Vol 1

The
BIOLOGY
0/
SEA TURTLES
Marine Science Series
The CRC Marine Science Series is dedicated to providing state-of-the-
art coverage of important topics in marine biology, marine chemistry, marine
geology, and physical oceanography. The series includes volumes that focus
on the synthesis of recent advances in marine science.
CRC MARINE SCIENCE SERIES
Series Editor
Michael J. Kennish^ Ph.D.
Published T itles
Artificial Reef Evaluation with Application to Natural Marine Habitats,
William Seaman, Jr.
Chemical Oceanography, Second Edition, Frank J. Millero
Coastal Ecosystem Processes, Daniel M. Along!
Ecology o f Estuaries: Anthropogenic Effects, Michael J. Kennish
Ecology o f Marine Bivalves: An Ecosystem Approach, Richard F. Dame
Ecology o f Marine Invertebrate Larvae, Larry MeEdward
Ecology o f Seashores, George A. Knox
Environmental Oceanography, Second Edition, Tom Beer
Estuary Restoration and Maintenance: The National Estuary Program,
Michael J. Kennish
Eutrophication Processes in Coastal Systems: Origin and Succession
o f Plankton Blooms and Effects on Secondary Production in
Gulf Coast Estuaries, Robert J. Livingston
Handbook o f Marine Mineral Deposits, David S. Cronan
Handbook fo r Restoring Tidal Wetlands, Joy B. Zedler
Intertidal Deposits: River Mouths, Tidal Flats, and Coastal Lagoons,
Doeke Eisma
Marine Chemical Ecology, James B. McClintock and Bill J. Baker
Morphodynamics o f Inner Continen tal Shelves, L. Donelson Wright
Ocean Pollution: Effects on Living Resources and Humans, Carl J. Sindermann
Physical Oceanographic Processes o f the Great Barrier Reef, Eric Wolanski
The Physiology o f Fishes, Second Edition, David H. Evans
Pollution Impacts on Marine Biotic Communities, Michael J. Kennish
Practical Handbook o f Estuarine and Marine Pollution, Michael J. Kennish
Practical Handbook o f Marine Science, Third Edition, Michael J. Kennish
Seagrasses: Monitoring, Ecology, Physiology, and Management,
Stephen A. Bortone
Trophic Organization in Coastal Systems, Robert J. Livingston
The
BIOLOGY
of
SEA TURTLES
Edited by
Peter L. Lutz
John A. Musick
CRC PR E SS
Boca Raton London New York Washington, D.C.
Library of Congress Cataloging-in-Publication Data
The biology of sea turtles / edited by Peter L. Lutz and John A.

Musick.
p. cm. — (CRC marine science series)
Includes bibliographical references (p. ) and index.
ISBN 0-8493-8422-2
1. Sea turtles. I. Lutz, Peter L. II. Musick, John A.
III. Series; Marine science series.
QL666.C536B56 1996
597.92—dc20 96-36432
CIP
This book contains information obtained from authentic and highly regarded sources. Reprinted material
is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable
efforts have been made to publish reliable data and information, but the author and the publisher cannot
assume responsibility for the validity of all materials or for the consequences of their use.
Apart from any fair dealing for the purpose of research or private study, or criticism or review, as permitted
under the UK Copyright Designs and Patents Act, 1988, this publication may not be reproduced, stored
or transmitted, in any form or by any means, electronic or mechanical, including photocopying, micro
filming, and recording, or by any information storage or retrieval system, without the prior permission
in writing of the publishers, or in the case of reprographic reproduction only in accordance with the
terms of the licenses issued by the Copyright Licensing Agency in the UK, or in accordance with the
terms of the license issued by the appropriate Reproduction Rights Organization outside the UK.
All rights reserved. Authorization to photocopy items for internal or personal use, or the personal or
internal use of specific clients, may be granted by CRC Press LLC, provided that $.50 per page
photocopied is paid directly to Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923
USA. The fee code for users of the Transactional Reporting Service is ISBN 0-8493-8422-
2/97/$0.00+$.50. The fee is subject to change without notice. For organizations that have been granted
a photocopy license by the CCC, a separate system of payment has been arranged.
The consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for
creating new works, or for resale. Specific permission must be obtained in writing from CRC Press LLC
for such copying.
Direct all inquiries to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation, without intent to infringe.
Visit the CRC Press Web site at www.crcpress.com
© 1997 CRC Press

No claim to original government works
International Standard Book Number 0-8493-8422-2
Library of Congress Card Number 96-36432
Printed in the United States of America 6 7 8 9 0
Printed on acid-free paper
Preface
The impetus for this book comes from the explosion in interest and involvement in
sea turtle biology and conservation. The interest in sea turtles is easy to understand.
The turtles belong to the most ancient line of living reptiles, first appearing more
than 200 million years ago in the late Triassic. When turtles first entered the sea is
not known, probably in the early Mesozoic, and for the next 100 million years,
during the rise and reign of the dinosaurs, the sea turtles shared the ocean with a
rich diversity of other air breathing reptiles, including the ichthyosaurs and plesio
saurs. But while the end of the Cretaceous witnessed the extinction of the dominant
large reptiles, the sea turtles continued to flourish up until very recent times when
their numbers have, through human interference, drastically declined. In order to
appreciate the causes of this collapse it must help to understand the selective advan
tages that have allowed sea turtles to survive so tenaciously, and, more particularly,
how they managed to hold their own against the comparatively recent invasions of
the sea by mammals. Up until the past 20 years or so most information on sea turtles
had been confined by the accessible, the nesting female and her eggs. But the use
of new technologies in data gathering — including mitochondrial DNA analyses,
remote sensing, and more sophisticated physiological monitoring techniques — has
resulted in a substantially greater understanding of sea turtle biology at all stages
of their life history. Major advances have been made in such broad areas as popu
lation genetics and phylogeny, sensory biology, migration and orientation, hatchling
behavior, age and growth, reproduction and endocrinology, sex determination, diving
physiology, and osmoregulation.
The results of these diverse studies are widely scattered in the literature which
this book brings together in a comprehensive summary; no similar reference has
previously been available. However, it is not a mere compilation, its focus is on how
sea turtles operate in, are adapted to, and are dependent upon their marine environ
ment. Considering the endangered status of these fascinating giant marine reptiles,
the book also deals with the threats to survival presented by manmade changes to
the ocean and coastal zones.
Peter L. Lutz
John A. Musîck
Acknowledgment
The idea for this book was bom out of many late nights of discussions with friends
and colleagues at annual meetings of the Sea Turtle Symposium. This convocation
has grown since 1981 from a small group of about 60 scientists and conservationists
to more than 700 workers drawn from 30 nations each year. We are extremely grateful
for those who contributed chapters, as well as those who freely gave of their precious
time as reviewers. This book is dedicated to the collegial spirit of scholarship and
camaraderie embodied in the Annual Sea Turtle Symposium.
The Editors
Peter L. Lutz, Ph.D., holds the McGinty Eminent Scholar Chair in Marine Biology
at Florida Atlantic University. Dr. Lutz received both his B.Sc. and Ph.D. from
Glasgow University, Scotland. After finishing his Ph.D. in 1970 he became a research
associate with Dr. Knut Schmidt-Nielsen at Duke University, with whom he worked
on avian respiration. In 1972 he became a Lecturer at the University of Bath,
England, and in 1974 an Associate Professor in Marine Biology at the Rosenstile
School of Marine and Atmospheric Science, University of Miami. In 1982 he was
promoted to Professor and in 1983 became Chairman of Marine Biology and Fish
eries, a post he held until he took up his present position in 1991.
As a comparative physiologist, he has worked on the physiology of a wide
variety of organisms, from liver flukes to duck-billed platypuses. His current interests
center on survival strategies of anoxia-tolerant brains and on the physiology of sea
turtles. Dr. Lutz is a member of the Society for Neuroscience, the American Phys
iological Society, and the Society for Experimental Biology. He is Series Editor for
the Marine Science Series published by CRC Press. He was a Governing Council
Member, Bahamas National Trust, and is a Fellow of the Explorers Club and member
of the Species Survival Commission of the lUCN. He has authored more than 150
research papers and 2 books.
John A. (Jack) Mustek, Ph.D., received his B.A. in Biology from Rutgers University
in 1962 and his M.A. and Ph.D. from Harvard University in 1964 and 1969, respec
tively. He has been on the faculty at the Virginia Institute of Marine Science, College
of William and Mary since 1967 and has successfully mentored 28 M.A. and 30
Ph.D. students. His research has covered many aspects of vertebrate ecology includ
ing community structure in deep sea fishes, shark population dynamics, and sea
turtle ecology. In 1985, he was elected a Fellow by the American Association for
the Advancement of Science. He has served as President of both the American
Elasmobranch Society and the Annual Sea Turtle Symposium and has also served
on numerous national and international advisory groups for conservation and man
agement, most recently on the International Union for the Conservation of Nature
working groups for sharks and sea turtles. He has authored or contributed to more
than 100 research papers and 3 books.
Contributors
Ralph A. Ackerman, Ph.D. Catherine M.F. Lohmann, Ph.D.
Department of Zoolology and Genetics Department of Biology
Iowa State University University of North Carolina
Ames, Iowa Chapel Hill, North Carolina
Karen A. Bjorndal, Ph.D. Kenneth J. Lohmann, Ph.D.

Archie Carr Center for Sea Turtle Department of Biology
Research University of North Carolina
and Chapel Hill, North Carolina
Department of Zoology
University of Florida Molly E. Lutcavage, Ph.D.
Gainesville, Florida Edgerton Research Center
New England Aquarium
Brian W. Bowen, Ph.D. Boston, Massachusetts
Archie Carr Center for Sea Turtle
Research Peter L. Lutz, Ph.D.
and Department of Biological Sciences
Department of Zoology Florida Atlantic University
University of Florida Boca Raton, Florida
Gainesville, Florida
Jeffrey D. Miller, Ph.D.
M.Y. Chaloupka, Ph.D. Queensland Department of Environment
Queensland Department of Environment Townsville, Queensland, Australia
and Heritage
Townsville, Queensland, Australia John A. Musick, Ph.D.
School of Marine Science
Robert H. George, D.V.M. Virginia Institute of Marine Science
Aquatic Animal Medicine Consultants Gloucester Point, Virginia
Gloucester, Virginia
Michael P. O’Connor, Ph.D.
Stephen A. Karl, Ph.D. Department of Bioscience and
Biology Department Biotechnology
University of South Florida Drexel University
Tampa, Florida Philadelphia, Pennsylvania
Colin J. Limpus, Ph.D. David W. Owens, Ph.D.

Queensland Department of Environment Department of Biology
and Heritage Texas A & M University
Brisbane, Australia College Station, Texas
Frank V. Paladino, Ph.D, Michael Salmon, Ph.D.
Department of Biology Department of Biological Sciences
Purdue University Florida Atlantic University
Fort Wayne, Indiana Boca Raton, Florida
Pamela T. Plotkin, Ph.D. James R. Spotila, Ph.D.

Department of Bioscience and Department of Bioscience and
Biotechnology Biotechnology
Drexel University Drexel University
Philadelphia, Pennsylvania Philadelphia, Pennsylvania
Blair E. Witherington, Ph.D.

Peter C.H. Pritchard, Ph.D.
Florida Marine Research Institute
Florida Audubon Society
Department of Environmental Protection
Winter Park, Florida
Tequesta, Florida
Jeanette Wyneken, Ph.D.

Department of Biological Sciences
Florida Atlantic University
Boca Raton. Florida
Table of Contents
Chapter 1
Evolution, Phylogeny, and Current Status............................................................. 1
Peter C.H. Pritchard
Chapter 2
Population Genetics, Phylogeography, and Molecular Evolution......................... 29
Brian W. Bowen and Stephen A. Karl
Chapter 3
Reproduction in Sea Turtles................................................................................... 51
Jejfrey D. Miller
Chapter 4
The Nest Environment and the Embryonic Development of Sea Turtles............. 83
Ralph A. Ackerman
Chapter 5
Orientation, Navigation, and Natal Beach Homing in Sea Turtles..................... 107
Kenneth J. Lohmann, Blair E. Witherington, Catherine M.F. Lohmann,
and Michael Salmon
Chapter 6
Habitat Utilization and Migration in Juvenile Sea Turtles.................................. 137
John A. Musick and Colin J. Limpus
Chapter 7
Sea Turtle Locomotion: Mechanics, Behavior, and Energetics........................... 165
Jeanette Wyneken
Chapter 8
Foraging Ecology and Nutrition of Sea Turtles................................................... 199
Karen A. Bjorndal
Chapter 9
Age, Growth, and Population Dynamics............................................................. 233
M.Y. Chaloupka and John A. Musick
Chapter 10
Diving Physiology................................................................................................. 277
Molly E. Lutcavage and Peter L. Lutz
Chapter 11
Thermal Biology.....................................................................................................297
James R. Spotila, Michael P. O'Connor, and Frank V Paladino
Chapter 12
Hormones in the Life History of Sea Turtles...................................................... 315
David W. Owens
Chapter 13
Salt, Water, and pH Balance in Sea Turtles......................................................... 343
Peter L, Lutz
Chapter 14
Health Problems and Diseases of SeaTurtles...................................................... 363
Robert H. George
Chapter 15
Human Impacts on Sea Turtle Survival................................................................ 387
Molly E. Lutcavage, Pamela Plotkin, Blair Witherington,
and Peter L. Lutz
Index.......................................................................................................................411
1 Evolution, Phylogeny, and
Current Status
Peter C. H. Pritchard
CO NTENTS
1.1 Introduction....................................................................................... 2
1.2 Marine Turtles................................................................................................... 3
1.3 Historical Classifications of Marine Turtles.....................................................5
1.4 Cladistic Placement of Marine Turtle Groups.................................................7
1.5 Marine Turtle Families..................................................................................... 7
1.6 Familial Definitions.......................................................................................... 8
1.6.1 Cheloniidae........................................................................................... 9
1.6.2 Dermochelyidae.................................................................................... 9
1.6.3 Toxochelyidae............................................................................... 9
1.6.4 Protostegidae........................................................................................ 10
1.7 Dermochelyid Diversity................................................................................... 10
1.8 Cheloniid Diversity...........................................................................................11
1.9 Generic Definitions of Living Cheloniids....................................................... 11
1.9.1 Cheionia...............................................................................................12
1.9.2 Eretmochelys....................................................................................... 12
1.9.3 Lepidochelys........................................................................................ 12
1.9.4 Caretta................................................................................................. 12
1.9.5 Natator.................................................................................................13
1.10 Affinities of Recent Cheloniid Species: Current Questions......................... 13
1.11 Survival Status of Living Cheloniids............................................................ 16
1.11.1 Legal and Formal Status..................................................................... 16
1.11.2 Biological Status................................................................................. 16
1.11.2.1 The Green Turtle, Cheionia mydas..................................... 17
1.11.2.2 The Black Turtle, Cheionia agassizii.................................. 18
1.11.2.3 The Flatback, Natator depressus......................................... 18
1.11.2.4 The Loggerhead, Caretta caretta........................................ 19
1.11.2.5 The Hawksbill, Eretmochelys imbricata.............................20
1.11.2.6 The Olive Ridley, Lepidochelys olivácea............................22
1.11.2.7 The Kemp’s Ridley, Lepidochelys kempi............................22
1.11.2.8 The Leatherback, Dermochelys coriacea............................23
References............................................ ...................................................................24
0-8493-8422-2/97/$0.00+$.50
© 1997 by CRC Press, Inc
The Biology of Sea Turtles
1.1 INTRODUCTION
Turtles, being reptiles and tied to the terrestrial environment for oviposition (with
the single partial exception of the chelid Chelodina nigosa [Kennett, 19931), are
generally assumed to have had terrestrial origins. This assumption is probably
correct. Additional evidence is provided by the earliest fossil turtles of the genus
Proganochelys and its relatives that had already achieved a broad distribution in the
northern continents (then still united as “Laurasia”) by the Triassic, and that appear
from their limb structure to have been terrestrial, or, at most, marshdiving forms
(Gaffney, 1990).
Authors such as Römer (1945) and Swinton (1958) assumed that the group of
early reptiles long known as the Cotylosauria, characterized inter alia by the absence
of true temporal fossae, was ancestral to the turtles. Yet intermediate stages were
unknown in the fossil record, the alleged “missing link,” Eunotosaurus Seeley 1892,
from the Permian of South Africa having now been reinterpreted, and Swinton,
although observing that “the Chelonian skull has many features in common with
that of Diadectid Cotylosaurs,” cautiously concluded that “a direct relationship is
not likely.”
Gaffney (1975) considered the turtles to be the sister group of the Synapsida
plus the Diapsida. But he later revised this opinion (Gaffney and Meylan, 1988),
and argued that the turtles were the sister group to the Diapsida alone. Subsequently,
Lee (1993) conducted a cladistic analysis of the Chelonii and various groups of
primitive amniotes and concluded, on the basis of 16 shared derived features, that
the pareiasaurs, a group of anapsid, large, terrestrial reptiles with short, heavy bodies,
and with large osteoderms possibly anticipatory of the chelonian shell, were evolu-
tionarily close to the earliest turtles. The pareiasaurs were, for a time, cosmopolitan
in distribution, but although they flourished in the Late Permian, they were not
persistent. Lee considered the evidence linking the chelonians to other traditionally
cotylosaurian groups such as the captorhinids and the procolophonoids to be weak.
In view of the terrestrial adaptations of the earliest turtles, it is noteworthy that
the subsequent evolution of the Order Chelonii has seen a remarkable proliferation
of aquatic taxa, whereas terrestriality has been retained (or reachieved?), amongst
modern forms primarily by the single family Testudinidae (with about 40 living
species). Some isolated fossil genera (e.g., Meiolania, Zangerlia — see Gaffney
[1983] and and Mlynarski [1972] for discussion) also appeared to have been terres
trial, as are a few living emydids and batagurids. Yet eleven of the twelve living
families are basically aquatic, although in some cases families that are considered
exclusively aquatic today, such as the Pelomedusidae and the Dermatemyidae, once
had significant terrestrial representation (Wood, 1985; Mlynarski, 1972). The great
majority of the extinct families were entirely aquatic also, as far as is known.
Moreover, even specialized aquatic forms, such as the Trionychidae, appeared very
early in the fossil record (Hutchison, 1982). Furthermore, as far back as the early
Jurassic, a newly described turtle, Kayentachelys, from the Kayenta fonnation of
eastern Arizona, considered to be the sister taxon for all other cryptodires, had a
carapace with all of the morphological features of modern, fully aquatic species
(Gaffney et al., 1987).
Evolution, Phylogeny, and Current Status 3
The successful penetration of aquatic niches by both early and modern turtles
was probably made possible by a remarkable example of preadaptation. Other living
aquatic reptiles, including the sea snakes and other snake species adapted to various
degrees for freshwater environments, marine iguanas, crocodilians, etc., as well as
the extinct ichthyosaurs, swim (or swam) by means of body and tail undulations not
dissimilar to those of typical fishes. Turtles, on the other hand, lost the capacity for
this form of propulsion when they developed the shortened, rigid body form and
corselet that has characterized the group since the Triassic. This body form offered
armored resistance to attack by predators, but the tradeoff was reduced speed and
agility, obliging those terrestrial chelonian species surviving in a world with increas
ingly sophisticated predators to adopt specialized (fossorial, cryptic, insular) life
styles (Pritchard, 1979).
Nonetheless, while the encarapaced, terrestrial body form with columnar, walk
ing-type limbs precluded aquatic locomotion by serpentine undulation, it was
remarkably preadapted for very different modes of propulsion, namely swimming
by means of alternating thrusts of limbs with webbed digits, or simultaneous strokes
with powerful paddlelike forelimbs, and these occurred repeatedly.
The shell required only modest modifications in the transition to aquatic life,
these generally taking the form of an overall lower, more streamlined profile, with
sharp, tapered edges to the carapace margins, and expansion of the plastral lobes,
thus facilitating lateral (swimming) rather than ventral (walking) limb movements
(Gaffney et al., 1987).
Only the limbs required profound modifications, and these were indeed forth
coming, the precise form of the newly reshaped limbs being governed by the degree
of aquatic specialization appropriate to the new life-style. In the vast majority of
aquatic chelonians, the limbs represent a compromise between the needs of swim
ming and of walking. In most swimming fonns the hind limbs are longer, more
powerful, and have more extensive webbing than the forelimbs, and they generate
the principal propulsive force, although their narrow specialization for this function
alone is constrained by the need in all chelonian species to utilize the hind limbs
for nest construction.
But in some of the most aquatic chelonians, the forelimbs provide the principal
propulsive force. A trend in this direction is evident in the soft-shelled turtles
(Trionychidae), in which the forelimbs have become substantially modified into
swimming paddles, and the foreclaws (and hindclaws) are reduced to three on each
limb. In these turtles, propulsive force is generated to a comparable degree by
forelimbs and hindlimbs. Forelimb specialization for swimming is carried further in
the family Carettochelyidae (now monotypic, but with an elaborate fossil record and
forebears widespread in northern continents [Mlynarski, 1976]), in which the fore
limbs are paddlelike and two-clawed, but still flexible.
1.2 MARINE TURTLES

Modification of the forelimbs for swimming reached its apogee in the modern marine
turtles of the families Cheloniidae and Dermochelyidae, in which the paddlelike
forelimbs achieve a substantial degree of rigidity by having elongate phalanges
enmeshed in a continuous matrix of tough connective tissue. This not only makes
independent mobility of individual digits impossible, but flexion of the blade of the
paddle as a whole is tightly constrained by the combination of the fibrous binding
of the digits and nonalignment of the corresponding joints between the individual
phalanges of adjacent digits (Zangerl, 1980). In such forms, the forelimb claws are
essentially rudimentary, reduced to one, or two at most. But the claw on the first
digit has a specialized secondary function in adult male cheloniids, in which it is
enlarged and hooklike, and is used for clasping the anterolateral marginal area of
the female during copulation.
Such limbs are poorly adapted for terrestrial locomotion in that the proximal
parts of the limbs bear almost the entire weight of the animal, but ability to locomote
on land cannot be compromised beyond a certain point because of the inescapable
marine chelonian committment to terrestrial oviposition. On the other hand, the
extraordinary demands for superior swimming ability and endurance made by a
marine life incorporating transoceanic migrations ensure that adaptations for aquatic
locomotion are paramount and uncompromised, whereas terrestrial locomotion can
be allowed to become slow and labored, as long as it is not entirely precluded.
Turtles are an important component of marine ecosystems — primarily tropical
and to a lesser degree subtropical ones, but with one species (Dermochelys coriacea)
showing remarkable adaptations for survival and function in very cold water (Frair
et ak, 1972; Greer et al., 1973). Today, however, only seven or eight species of
marine turtles survive, the majority distributed unevenly through all three tropical
oceans, but with three having relatively restricted distributions (the flatback, Natator
depressus, in northern Australia; Kemp’s ridley, Lepidochelys kempi, in the Gulf of
Mexico and North Atlantic; and the black turtle, Chelonia agassizH, in the eastern
Pacific).
Despite this paucity of species, the living marine turtles are not a relictual group.
They have great economic value as well as extreme vulnerability to mankind, at
least while nesting, and their inclusion on most lists of threatened or endangered
species is a reflection primarily of past overexploitation and current need for better
management rather than to inherently poor adaptation to post-Pleistocene conditions.
The documented great diversity of sea turtle taxa in the past probably derives
from several causes, including on the one hand the relatively good chances of
fossilization of these heavy-boned animals, and on the other hand the disappearance
of entire isolated oceans (such as the Niobrara Sea), once populated by diverse and
remarkable sea turtle species. Moreover, the overall evolutionary history of the group
is paralleled on a mega-scale by the history of life itself (Gould, 1980) or on a
human-economy scale by the reduction in the number of automobile manufacturers
in the U.S. during the last century even as cars have become more, rather than less
popular (Yates, 1996). In these examples, an early blossoming forth of numerous
fundamentally “experimental” types (Pritchard and Trebbau [1984] list 27 entirely
extinct genera in the Cheloniidae alone) became winnowed down, probably through
elimination of overspecialized types with poor response to changing conditions, to
a much-reduced diversity of modem survivors, characterized to a considerable degree
by being less rather than more specialized than their forebears (Zangerl, 1980).
Evolution, Phylogeny, and Current Status
1.3 HISTORICAL CLASSIFICATIONS OF MARINE

TURTLES
Linnaeus (1758) included all of the turtles and tortoises that he recognized in the
single genus Testudo, a taxon now restricted to a small group of circum-MediteiTa-
nean terrestrial species of the family Testudinidae. Subsequent early taxonomists in
this pre-Darwinian period failed to identify trenchant carapacial characters to justify
recognition of major divisions of chelonians, and this is understandable in that the
carapace superficially presents extraordinary conservatism. For example, the vast
majority of living and extinct chelonian taxa show the same number of vertebral
and costal scutes (five, and four pairs, respectively), as well as a similar suite of six
pairs of plastral scutes, the commonest variation being in the presence or absence
of an unpaired anterior scute, the intergular.
Turtles and tortoises, as a whole, clearly present a “natural group” (to use the
language of the pre-Darwinians), or a “monophyletic group” or “clade” (to use a
contemporary evolutionary concept), although different opinions have been
expressed on their “affinities” over the last two centuries. (MacLeay, in Jenyns
[1835], perceived or fancied a relationship between the chelonians and the cepha-
lopods, although Jenyns allowed that “the hiatus occurring between is very consid
erable”; and I once saw armadillo specimens stored in the same cabinet as the Bell
collection of chelonians in the attic of the Oxford University Museum, although
there was no one to interpret the implications of this juxtaposition.) Perhaps the
most remarkable “pseudoturtle” that has been described was the Triassic placodont
sauropterygian Placochelys — even its scientific name is turtlelike. This marine
reptile had an oval carapace studded with conical osteoderms, cheloniid-style flip
pers, a head with a narrowed, toothless “beak” (although flattened, shell-crushing
teeth were present on the maxillae and palatines), and reduction of the presacral
vertebrae to about 22 (18 in true turtles). A reconstruction of Placochelys was even
featured, as a sea turtle, in a Hungarian postage stamp in 1969. But the presence of
a single pair of well-developed temporal fenestra high up on each side of the skull,
and many fundamental structural differences in the carapace, make clear that parallel
evolution rather than relationship is the explanation (Römer, 1945, 1956).
The turtles have long been recognized as an order of the class Reptilia, and the
most commonly utilized names for that order include the pre-Linnaean Testudinata
(Klein, 1751), first adopted post-Linnaeus by Oppel (1811); Testudines (Wagler,
1830); and Chelonii (Brongniart, 1800). Bour and Dubois (1985) presented a well-
argued case for utilization of Chelonii as the valid ordinal name today, although this
has not had universal acceptance. Other names that had little or no subsequent use
include Fornicata (Haworth, 1825); Sterrochrotes (Ritgen, 1828); and Tylopoda
(Meyer, 1849).
The first multigeneric system of classification proposed for the Testudines was
that of Brongniart (1805), who considered habitat to be the key character, and
recognized Testudo for the terrestrial species, Emys for the freshwater ones, and
Chelonia for the marine ones. Oppel (1811) presented a dichotomous division of
the turtles at the subordinal level, reeognizing the Chelonii with oarlike forelimbs
and the Amydae with distinct digits. The same division with different names was
used by Merrem (1820) and Bell (1828), who called the marine species with oarlike
limbs Pinnata, and the species with distinct digits Digitata.
Further refinements followed. It was soon recognized that the limbs of tortoises
were as distinct from those of freshwater species as were the limbs of the marine
forms. Moreover, the leatherback turtle and also the freshwater softshells presented
radical distinctions from the typical chelonian body form. This led to Gray’s (1825)
recognition of five families, the Cheloniadae, the Sphargidae, the Trionicidae, the
Emydidae, and the Testudinidae.
Agassiz (1857) reviewed subsequent early nineteenth century classifications,
remarking sagely that the schema of Brongniart, Ritgen, Wagler, Duméril and
Bibron, and Prince Canino Bonaparte, although utilizing totally different names, had
a great deal in common. It was Wagler (1830), followed by Duméril and Bibron
(1836) who hit upon the character that is considered to be of the most fundamental
importance today, namely the plane of retraction of the neck. This feature not only
reflects in the articulations of the cervical vertebrae, but also correlates with impor
tant skull features (such as the trochlear mechanism for the jaw retractor muscles),
and shell characters (the pelvis fused to both carapace and plastron in the pleurodires,
and the presence of an intergular scute in the pleurodires, but only rarely in the
cryptodires). Many pleurodire groups also show remarkable parallelisms with cryp
todires, for example in the loss of the mesoplastra, as well as in the characteristic
carapacial scute mosaic.
Agassiz also commented that many of the classifications that he had reviewed
assumed (implicitly or explicitly) that the turtle species with oar- or paddlelike limbs
were “inferior” to those with separate digits, which in turn were “inferior” to those
with clublike walking feet. This evaluation of his predecessors’ beliefs was correct,
and indeed the assumption contains some elements of truth if we substitute the word
“primitive” for “inferior.” For example, the marine turtles do show the complete
(nonemarginate) skull roof considered characteristic of early turtles, and the char
acteristic elephantine foot of the testudinids is indeed probably a derived rather than
a primitive feature. But it should not be forgotten that both Agassiz and the ante
cedents whom he quoted were nonevolutionists, and thus concepts of primitive vs.
advanced features, or of “natural groups,” had at most a theological basis rather than
one based upon actual closeness of phylogenetic relationship.
Subsequent classifications, of living turtles at least, were based upon the primary
division of turtles and tortoises into Pleurodires and Cryptodires, although the proper
alignment of the leatherback and the softshells remained a challenge. The many
morphological peculiarities of the former (summarized by Gervais, 1872) led Cope
(1872) to declare this taxon the “sister group” (to use cladistic terminology) for all
other turtles. He named it “Athecoidea” (or, in family-level nomenclature, “Athe
cae”), meaning “without a shell.” This allocation was followed by Dollo (1886),
Boulenger (1889), Lydekker (1889), and various others, the latest author of note to
use it being Carr (1952). In these classifications, all other chelonians were identified
as Thecophora (“shell-bearing ones”).
Furthermore, the Thecophora were trichotomously divided into the superfamilies
Cryptodira, Pleurodira, and Trionychoidea (Boulenger, 1889). Such elevation of
status of the taxa with scuteless shells is understandable, in that it makes a mockery
of classifications founded upon details of scutation when some turtles have no scutes
at all. But, despite their extremely divergent morphology, Dermochelys and its fossil
relatives are now considered to be related to the marine turtles of the family Che-
loniidae (Gaffney and Meylan, 1988).
Yet Cope’s early (1872) designation of a taxon (Athecoidea) linking the pro-
tostegids and the dermochelyids, while shaken by Zangerl’s subsequent clarification
that protostegids (in contrast to demiochelyids) had normal chelonian scutes rather
than leathery skin, achieved some vindication with Gaffney and Meylan’s (1988)
proposal of an “Epifamily” Dermochelyoidae, with the same content. Moreover,
even Dollo, while recognizing the Athecae on the basis of its extreme morphological
modifications, considered it to be derived from the Cheloniidae (Dollo, 1886).
Dollo’s conclusion may be factually correct, although the divergence was so far back
that it is arguable whether or not the common ancestral stock was truly cheloniid.
But his nomenclatural inteipretation of its implications, although permissible by
classical taxonomists, is in conflict with the rules of Cladism, which seeks to establish
a nomenclature based purely upon temporal order of divergence rather than upon
degree or extent of divergence.
Today the trionychids also are recognized only as a family, related to the caret-
tochelyids, and incorporated within the Cryptodira (Meylan, 1987).
Various earlier, intermediate, or conflicting placements of the Dermochelyidae
are discussed by Pritchard and Trebbau (1984).
1.4 CLADISTIC PLACEMENT OF MARINE TURTLE

GROUPS
Gaffney and Meylan (1988) offer the following cladogram (Figure 1.1) for the marine
turtles with well-developed flippers (i.e., excluding the Thalassemyidae).
While I have many misgivings about the cladistic process as a whole (Pritchard,
1994), this cladogram has merit in that it recognizes areas of uncertainty without
attempting arbitrary resolution, including the familial placement of Notochelone,
Allopleuron, and Erquelinnesia, and the unresolved trichotomies for the osteopygid
genera as well as the Notochelone complex. It is also clearly preliminary in that, of
the 31 cheloniid or possibly cheloniid genera mentioned by Pritchard and Trebbau
(1984), only nine are mentioned. There have also been recent discoveries clarifying
the relationships between the extant cheloniid species (Bowen et al., 1993) that are
at variance with the cladogram.
1.5 MARINE TURTLE FAMILIES

Marine adaptation has been undertaken more than once in the history of the Chelonii.
The earliest marine turtles appeared in the Jurassic, with a marine species of the
family Pleurosternidae, Desmemys bertelsmanni, and an entirely marine family, the
Thalassemyidae also appearing in the Jurassic, possibly derived from the primitive,
probably freshwater Plesiochelyidae. The Thalassemyidae showed many carapacial
FIGURE 1.1 Cladogram “D”, Chelonidea.
parallels (or anticipations) with later marine turtle families, but the limbs remained
unspecialized for marine life (Hay, 1908), and Gaffney and Meylan (1988) recently
found little or nothing to separate Thalassemys itself from the plesiochelyids. Cer
tainly, Hay’s concept of the Thalassemyidae was a polyphyletic one, with his family
definition following rather than preceding the allocation of a mixed bag of genera
from at least three distinct families.
Some extinct Pelomedusid genera (e.g., Taphrosphys [Wood, 1974]; possibly
Stupendemys [Wood, 1976]) may also have been marine.
By the Cretaceous, four marine turtle families — the Toxochelyidae, Protoste-
gidae, Cheloniidae, and Dermochelyidae —were all established, the last two surviv
ing to the present. Whether or not these all represent independent invasions of the
marine environment is unclear, but it is probable (as Dollo observed) that the
dermochelyids represent an extraordinarily divergent offshoot of the cheloniids,
whereas some of the toxochelyids bear such a strong superficial resemblance to
certain cheloniids that the possibility of close relationship must be seriously enter
tained.
1.6 FAMILIAL DEFINITIONS

Detailed definitions are possible for the living marine turtle families (Pritchard and
Trebbau, 1984), although the proviso has to be stated that the applicability of soft-
tissue characters, established from the living taxa, to the fossil representatives is
obviously conjectural. Gaffney and Meylan (1988) give diagnoses for these families
Evolution, Phylogeny, and Current Status
based upon purely osteological features. Definitions for the entirely extinct marine
families are necessarily less complete.
1.6.1 C heloniidae
A family of turtles characterized by an extensively roofed skull with well-developed

rhamphothecae; secondary palate present; incompletely retractile or nonretractile
head; extremities in the form of nonretractile flippers covered with numerous small
scales; the forelimbs having highly elongate digits firmly bound together by con
nective tissue, the claws being reduced to one or two on each limb, and the radius
and ulna immobilized against independent movement by juxtaposed rugose surfaces;
shell covered with horny scutes, variable in number, but commonly including five
vertebráis and six pairs of plastral scutes, together with an unbroken series of three
or four pairs of inframarginals, multiple axillary scutes, and usually both an interguiar
and an interanal scute; the plastron often with persistent fontanelles, one in the
middle and others in the entoplastral and xiphiplastral regions; hyoplastra and hypo-
plastra not suturally connected mesially, but each with a series of coarse spikes that
interdigitate mesially; plastron not crucifonn, and posterior plastral lobe relatively
long and wide.
1.6.2 D ermochelyidae
A family of turtles characterized by extreme reduction of the bones of the carapace
and plastron (with the neural and peripheral bones of the carapace, and entoplastron
of the plastron, lacking; the pleurais reduced to endochondral ribs, separated by
wide fenestrae; and the plastral bones reduced to narrow splints, forming a ring of
bones surrounding a great fontanelle); development of a neomorphic epithecal shell
layer consisting of a mosaic of thousands of small polygonal bones; claws and shell
scutes lacking (scales only present in the first few weeks of life); skull without nasal
bones; jaw surfaces covered with keratin, but lacking differentiated rhamphothecae;
parasphenoid overlain by pterygoids; prefrontals in contact dorsally, with descending
processes that are moderately separated; unridged tomial surfaces; a generally neo-
tenic and oil-saturated skeleton; extensive areas of vascularized cartilage in the
vertebrae, limb girdles, and limb bones; very large body size; and marine habitat.
1.6.3 T oxochelyidae
The toxochelyids were a group of small to medium-sized, broad-shelled (circular
to cordiform) marine turtles that diverged from the early cheloniid stock in the early
Cretaceous and persisted in the proto-Atlantic and Niobrara seas until the late
Eocene. Various toxochelyid genera display progressive development of a secondary
palate, from Toxochelys with a primary palate only, through Ctenochelys with
incipient undershelving, a more advanced condition in Osteopygis, and extreme
development of the secondary palate in Erquelinnesia. The carapace ranged from
fully closed (i.e., without intercostal fontenalles) in Osteopygis to a highly reduced
condition with persistent, large fontanelles in Erquelinnesia. Features characteristic
of the family include the presence of small nasal bones and the well-developed
10 The Biology of Sea Turtles
peripherals (often reduced to narrow bars in cheloniids). The plastron, primitively

cruciform, was larger and with much wider bridges in later representatives. In one
subfamily (the Lophochelyinae), the neural ridge was highly developed and serrated
or tuberculate in lateral aspect. The toxochelyids differed from the thalassemyids
not only in the cruciform plastron, but also in having narrow vertebral scutes (much
narrower than the pleurais); unreduced neurals; and two suprapygals. In many
features of the skull they appeared intermediate between cheloniids and chelydrids
(i.e., snapping turtles). The toxochelyids are discussed by Hay (1908) and Zangerl
(1953, 1980).
1.6.4 Protostegidae
The protostegids were a group of large to gigantic turtles whose more primitive
members (e.g., Rhinochelys) have been found in Cretaceous deposits of France and
England, but which reached its maximal development (with the genera Protostega
and Archelon, known from Upper Cretaceous deposits of Texas, South Dakota,
Arkansas, Alabama, Colorado, and Kansas) in the Niobrara Sea. They showed certain
parallels with the dermochelyids, one of these being the possession of vascularized
chondroepiphyses (Rhodin, 1985), a feature present in a number of fossil turtle
genera (Psephophorus, Eosphargis, Archelon, Pneumatoarthrus, Corsochelys) as
well as the living Dermochelys. However, vascular chondroepiphyses, a condition
readily evident in the ends of the limb bones even of purely fossil material, may be
a parallelism connected with very large adult size (Pneumatoarthrus was originally
described as a dinosaur! [Cope, 1870; Baird, 1978]), and rapid growth potential
rather than an indicator of relationship.
The protostegids were characterized by large skulls (up to 100 cm in length in
the case of Arche Ion) with well-developed, sometimes strongly hooked beaks; no
secondary palate; reduced xiphiplastra and reduced or absent epiplastra; considerable
reduction of the carapacial armor, in some cases with intercostal fontanelles reaching
close to the vertebral column even in adults; a tendency towards keeled or tuberculate
middorsal keels (sometimes with carination present only on alternating neural
bones); and a slight to moderate degree of temporal emargination.
1.7 DERMOCHELYID DIVERSITY

The single extant dermochelyid, D. coriacea, shows very modest geographic vari
ation, and it is probable that no subspecies exist (but see Brongersma [in press] for
a discussion of potentially available names for different populations). Dermochelyids
appear not to fossilize well or easily, and the majority of the rather few fossil genera
reviewed by Pritchard and Trebbau (1984) are known from fragmentary material
only. For example, Cosmochelys (Eocene of Nigeria) is based upon fragments of
the epithecal carapace, and is not even considered dermochelyid by some authorities,
although it is so accepted by de Broin and Pironon (1980). Pseudosphargis (Late
Oligocène of Germany) is known from a cranial fragment only, and most species
of the polytypic Psephophorus (Eocene-Pliocene, Europe, North Africa, North
America) are known from shell fragments only, although P. rupeliensis is known
from several mostly complete specimens. In this genus, the shell is characterized by
thickened, strongly keeled epithecal bones that form a complete layer on the plastron
as well as the carapace (in contrast to Dermochelys).
In the genus Eosphargis (Eocene, Europe), a well-preserved skull shows typi
cally dermochelyid foim, but the mosaic epithecal layer was not yet developed
(Nielsen, 1959).
1.8 CHELONIID DIVERSITY

Pritchard and Trebbau (1984) inventoried 31 cheloniid genera (only four with living
representatives — Natator was resurrected later). However, many of these were
known from fragmentary material only — some only from a mandible, or a humerus
— and, despite the abundance of named forms, we remain a great distance from a
full comprehension of the phylogeny and morphological diversity of the family
Cheloniidae.
Probably the best brief summary is that of Zangerl (1980). He characterized
cheloniid evolution as a persistent theme of specialized forms that evolved, but failed
to persist, whilst a generalized stem group, adapted for near-shore rather than pelagic
waters, survived and progressively generated subsequent short-lived, relatively
pelagic, specialized offshoots. Thus, in the Cretaceous, named offshoots included,
on the one hand, Glyptochelone suyckerbuyki, and on the other hand a progression
of forms (Catapleura arkansaw; Desmatochelys lowi; Corsochelys haliniches; and
Allopleuwn hoffmarmi) that showed increasing pelagic specialization. Zangerl con
sidered pelagic adaptation to be manifest in elongate, persistently fontanelled cara
paces (as opposed to broad, fully ossified ones), and the plastron also with persistent
median and lateral fontanelles. Other specializations include the development of a
secondary palate and the conversion of the hind limbs into relatively stiff “rudders,”
rather than active propulsive members. But the situation is complex; many forms
show a mosaic of primitive and advanced characters, and one extant species (L.
oliváceo) has a fundamentally primitive shell, with broad, short outline and virtually
complete ossification, yet is so pelagic that the juveniles are almost never found,
despite this being the most abundant surviving marine turtle species.
Eocene cheloniids include Puppigerus camperi; Argillochelys subscriptata; and
Eochelone brabantica, only the last being considered to show pelagic specialization.
In the Oligocène, named taxa include Procolpochelys sp.; Oligochelone rupeliensis;
and ''Chelonia’' gwinneri. Miocene forms include P. grandaeva, P. melii, and Syl-
lomiis aegyptiacus. The only well-documented Pliocene species is Chelonia sismon-
dai.
1.9 GENERIC DEFINITIONS OF LIVING CHELONIIDS

It is easier to write a generic diagnosis for a large, polytypic genus than for a
monotypic one. For the former, a list of common features of the various recognized
species will suffice. However, for the latter, it is necessary to separate those char
acteristics that merely characterize the single known species from “true” generic
characters, that would permit correct diagnosis and allocation of a new potential
congener, fossil or living, that was unknown when the diagnosis was drawn up.
The diagnoses of Carr (1942) for four of the known cheloniid genera meet these
criteria, and are quoted below with very minor modifications. On the other hand,
the two recently presented and much more detailed diagnoses for Natator (Zangerl
et al., 1988; Limpus et al., 1988) are indistinguishable from diagnoses of the single
species N. depressus, in that they combine both fundamental (generic-level) and
species-specific data.
1.9.1 C helonia
One pair of prefrontal scales; homy cutting edge of lower jaw coarsely dentate, that
of upper Jaw strongly ribbed vertically; bony alveolar surface of upper jaw with a
low, but regularly raised auxiliary ridge behind the anterior ridge which is very
strong and terminates anteriorly in a pointed eminence at the posterolateral corner
of the premaxillary pit; costal scutes 4 pairs, well cornified, juxtaposed.
1.9.2 Eretmochelys
Two pairs of prefrontal scales; nuchal scute not in contact with first costáis; snout
elongate, narrow, the mandibular symphysis deeply excavated and not terminally
toothed; bony alveolar surface of upper jaw with a sharp-crested ridge; dorsal scutes
usually 4 pairs, usually thick and conspicuously imbricated; peripheral bones usually
11 (after Carr, 1942).
1.9.3 L epidochelys
Maxillaries not in contact, separated by vomer; frontal bone usually entering rim of
orbit; pterygoids markedly broadened anteriorly, the ectopterygoid processes strong;
fontanelles in choanal chamber near opening, not hidden by alveolar surface in
ventral aspect; external openings of orbits not concealed by overlying bones in
ventral aspect; descending processes of prefrontals not reaching palatines; lower jaw
with a more or less sharp and strong median elevation at the posterior border of the
bony alveolar surface, which may or may not extend forward as an elevated ridge;
four enlarged inframarginal scutes on the bridge; costal scutes thin, juxtaposed, 5
to 9 pairs (often asymmetrical); neural bones 11 to 15.
1.9.4 C aretta
Maxillaries in contact, not separated by vomer; pterygoids not, or but slightly,
broadened anteriorly, the ectopterygoid processes vestigial or lacking; fontanelles
far forward in choanal chamber, completely concealed by alveolar surface in ventral
aspect; external openings of orbits concealed by overlying bones in ventral aspect
(in mature specimens); descending processes of prefrontals connected with palatines;
bony alveolar surface of lower jaw smooth at posteromedian border; symphysis of
mandibles without longitudinal ridge; bridge with three enlarged inframarginal
scutes; costal scutes well comified, juxtaposed, usually five pairs; neural bones
usually 7 or 8.
1.9.5 N atator
Maxillaries widely separated by vomer; pterygoids broadened anteriorly and with
very strong ectopterygoid processes; fontanelles far forward in choanal chamber,
completely concealed by alveolar surface in ventral aspect; external openings of
orbits not concealed by overlying bones in ventral aspect; symphysis of mandible
with a strong median ridge terminating posteriorly in an elevated prominence;
external openings of internal carotid canals very large; head scalation characterized
by a pair of preoculars lying between maxillary rhamphotheca and prefrontals, post-
parietals variable and often asymmetrical, generally including an undivided median
element; plastral fontanelles closing at sides, but remaining open medially at matu
rity; four pairs of inframarginal scutes; carapacial scutes very thin, poorly kerati
nized, and with sulci tending to disappear with age; costáis four pairs. Neural bones
about twelve, with frequent transverse division of “standard” elements, and 2 or 3
suprapygals.
From the above diagnoses, the confusing “mosaic” of characters of Natator is
evident. The skull is so similar to that of L. kempi that some of the older skulls in
Australian museums have been identified and catalogued as that species.
1.10 AFFINITIES OF RECENT CHELONIID SPECIES:

CURRENT QUESTIONS
With only six or seven extant cheloniid species, the relationships between these
species have been surprisingly conjectural until very recent times. The key questions
may be phrased as follows:
1. What subfamilies (or “Tribes”) should be recognized, and what is the

content of each of the recognized subgroups?
2. What is the correct placement of the flatback turtle?
3. What is the relationship between the two forms of ridleys?
4. What is the status of the “black turtle” of the eastern Pacific?
Deraniyagala (1939) recognized a superfamily, the Chelonioidae, comprising

two families, the Cheloniidae (Chelonia and Eretmochelys), and the Carettidae
{Carena and Lepidochelys). The latter was distinguished by the greater number of
certain shell elements, specifically the peripheral bones and the costal scutes.
(Deraniyagala considered that four pairs of costáis and eleven pairs of peripheral
bones constituted a phylogenetieally reduced count, whereas today the proliferation
of elements beyond these numbers is considered to be the derived condition.)
Carr (1942) disagreed shaiply with Deraniyagala, in a paper that uneannily
anticipated the cladistic process. Using 25 characters (some not independent, and
unanalyzed in terms of primitive vs. derived), he found only four that justified placing
Eretmochelys with Chelonia, and 15 justifying placement of Eretmochelys with

Caretta and Lepidochelys. He concluded that the Cheloniidae should include only
Chelonia, the other three genera being placed in the Carettidae. Deraniyagala (1953)
partially agreed, or at least was willing to meet Carr half way, and recognized the
subfamilies Cheloniinae, Carettinae, and Eretmochelinae, all within the family Che
loniidae.
Subsequent authors were divided as to whom to follow. Frair (1979) and Prit
chard and Trebbau (1984) favored Carr (1942); Zangerl and Turnbull (1953), Mly-
narski (1969), and Smith and Smith (1980) followed Deraniyagala (1939), although
Smith and Smith overstated their case by claiming that it was “universally recog
nized.” Nobody followed Deraniyagala’s 1953 classification, although it was not
totally without merit.
Using genetic techniques unavailable to earlier workers, Bowen et al. (1993)
found that all phylogenetic analyses of mtDNA data supported placement of the
hawksbill with the Carettinae, suggesting that the unusual spongivorous diet of the
hawksbill probably evolved from a carnivorous rather than an herbivorous ancestral
condition.
The flatback turtle was originally described as Chelonia depressa by Carman,
1880. The type series was composite (including a specimen of C. mydas), but the
lectotype (MCZ 4473) was recognized by Carr (1942) as “an extraordinary speci
men,” with characters reminiscent of both Chelonia and Lepidochelys. Baur (1890)
thought that depressa belonged in its own genus, and McCullough (1908) agreed,
naming the new genus Natator, in combination with a new specific name, tessellatus.
Fry (1913) also agreed that the flatback was a distinctive species, and published a
detailed description. However, all of these works were generally ignored until Wil
liams et al. (1967) formally resurrected C. depressa as a valid cheloniid species.
Subsequent morphological analyses (Zangerl et al., 1988) and both electro
phoretic and osteological comparisons (Limpus et al., 1988) demonstrated conclu
sively that the flatback should be recognized as a distinct genus, Natator, with closer
affiliations with the Carettini than with the Chelonini. Limpus et al. did not examine
L. kempi or C. agassizi, but they presented a phylogeny of the remaining species,
based upon electrophoretic data, that placed Natator and L. olivácea close together,
Caretta and Eretmochelys in a separate branch and more distantly related, and
Chelonia mydas as the sister group to the other four taxa. However, Bowen et al.
(1993), using mtDNA analysis, found that Natator had comparably wide genetic
separations from both the Chelonini and the Carettini. The only conclusion one can
draw from this is that the flatback is the only extant representative of a new tribe,
the Natatorini.
Incidentally, in that the gender of Natator is not self-evident, it must be construed
as masculine because its describer (McCulloch, 1908) used it in combination with
a masculine-form epithet {‘‘tessellatus"'). The correct name for the flatback turtle is
thus N. depressus.
When examined side by side, the close relationship between the two species of
ridley (L. olivácea and L. kempi) is obvious. Indeed, there is no external difference
between a normal juvenile L. kempi and a specimen of L. olivácea with only five
pairs of costal scutes, a minority condition, but not rare in some areas (Pritchard,
1969). However, earlier workers lacked the material to make these comparisons, and
the two were placed in separate genera {Colpochelys and Lepidochelys) for a long
time. Carr (1942) had to report that prevailing opinion even then insisted either that
the two ridleys and the loggerhead should each be in separate genera, or that all
three forms should be united within Caretta. Yet, he added, it “seems evident that
kempii should be associated with olivácea in the genus Lepidochelys Fitzinger,” an
allocation presaged by Baur (1890) and followed by most subsequent authors (see
Pritchard, 1969, 1989 for details).
But there were holdouts, the most prestigious being Loveridge and Williams
(1957) and Mertens and Wermuth (1955), who considered kempi to be merely a
subspecies of L. olivácea. Once again, Bowen et al. (1993) offered a mitochondrial
DNA analysis, in this case demonstrating that, although the two forms were sister
taxa, they showed genetic distinction at the species level.
The “black turtle” was described as a full species, Chelonia agassizii, by Bocourt
(1868). C. mydas caninegra Caldwell 1962, appears to be a synonym, although the
assumption that this forni from the Gulf of California is identical to agassizii from
Pacific Guatemala has not been proven (but is probable). The pros and cons of
species vs. subspecies status for agassizii have been discussed by Mrosovsky (1983)
and Pritchard (1983). I have justified my own position (full species status) on
narrowly technical grounds (i.e., the nomenclature of the original describer, Bocourt,
should stand until formally changed), and on biological ones, in that agassizii and
mydas-Y\kç^ forms are sympatric in several places in the Pacific, including the Gal
apagos islands and New Guinea; the degree of differentiation in size, shape, and
color is more extreme than that found in any other Chelonia population; the dark
plastral coloration of agassizii is not environmentally derived; and there may be
purely physical reasons why a 40-kg adult male agassizii may be unable to mate
successfully with a 200-kg adult female mydas, even if the two do come in contact.
Subsequent contributions on the subject include that of Kamezaki and Matsui
(1995), who analyzed geographic variation in skull morphology of six Chelonia
populations (two from each tropical ocean), and concluded that the Galapagos
sample (i.e., agassizii) was completely separated from the other samples by a
canonical discriminant analysis, and that this result indicated distinctness of the
eastern Pacific population. On the other hand, Bowen et al. (1993) found that eastern
Pacific turtles represented but a small subset of lineage diversity within the broader
and deeper mtDNA gene tree for the globally distributed green turtle.
A conclusive decision on this issue thus remains elusive. Bowen et al. found
that nDNA data indicated very close relationship between the Galapagos and the
Pacific Mexican populations of "'agassiziC, but both lacked close affinities with
south-central and western Pacific populations, which argues in favor of species-
level status for agassizii. On the other hand, a phylogeny derived from nDNA links
the eastern Pacific populations with, inter alia, those of Ascension Island and the
Atlantic Coast of Africa (Parham and Zug, 1996)! This discordance of nDNA with
both morphology and geographic probability would, to some, substantially lower
the credibility of nDNA analysis as a technique for elucidating phylogenetic
relationships.
Systematists have been divided in recent years both in whether to recognize

subspecies at all, and, if the subspecies concept is to be rejected, whether to “pro
mote” previously recognized subspecies to full species status, or to “demote” them
into oblivion. Widespread, genetically complex species such as Chelonia mydas
present especial challenges when one tries to force them into rigid nomenclatorial
systems, and some of the implications of recognizing agassizii at all (including the
obligation to give nomenclatorial status to many other populations of Chelonia
mydas) are discussed by Parham and Zug (1996).
1.11 SURVIVAL STATUS OF LIVING CHELONIIDS

1.11.1 Legal and Formal S tatus
The U.S. Department of the Interior, under the authority of the Endangered Species
Act, currently lists C. mydas (endangered in Florida and Pacific Mexico) and Caretta
caretta as Threatened Species. E. imhricata, L. olivácea, L. kempi, and Dermochelys
coriácea are listed as endangered. N. depressus, the most restricted and least-known
of all living cheloniids, is unlisted.
The Convention on International Trade in Endangered Species of Flora and
Fauna (CITES) lists all marine turtles on its Appendix I (i.e., prohibited from
international trade from or to signatory countries).
The World Conservation Union (lUCN) has listed C. caretta as Vulnerable and
all other sea turtle species (except A. depressus) as Endangered. However, the lUCN,
in close cooperation with the Secretariat and Parties to CITES, has now adopted a
set of complex numerical and ostensibly objective criteria by which the status
category of a species should be deduced. While these criteria are expected to apply
only to species proposed for the CITES appendices in the future, they will be
incorporated across the board by lUCN. The criteria incorporate considerations of
actual global population numbers, fragmentation of habitat and populations, and
demonstrable population trends. For the great majority of species, the necessary data
are unlikely to be currently available. Among the sea turtles, even the “critically
endangered” L. kempi may only qualify as “conservation dependent,” in view of its
modest population recovery in recent years.
1.11.2 B iologïcal S tatus
The actual, biological status of marine turtles is a topic of considerable complexity.

Unlike, say, endemic species of fauna or flora on small, remote oceanic islands,
where a single ecological perturbation may precipitate a chain of events leading to
multiple extinctions, rather few marine species are faced with biological extinction,
and typically those that have been exterminated, such as Stellers’s Sea Cow or the
Caribbean monk seal, were large, intrinsically vulnerable species with very small
natural ranges.
Rather, marine species are faced with problems of overharvest, or even “popu
lation collapse” and “commercial extinction” at some point short of biological
extinction. In other words, the “demostat” for such species may be drastically shifted
to a lower position once anthropogenic stresses (directed or incidental) become

manifest and unremitting, but, typically, the new setting of the demostat is not at
the zero point. Nonetheless, it remains to be seen whether these conceptual gener
alizations about marine species are applicable to marine turtles, in that the unavoid
able terrestrial ovipositional excursions of all marine turtle species constitute a
special vulnerability to which fully marine species are not subject.
The biological status of the eight (or seven) surviving marine turtle species are
summarized below.
1.11.2.1 The Green Turtle, Chelonia mydas

This is a circumglobal species, most of whose important nesting and feeding grounds
lie within the tropics. It has major nesting colonies on mainland shores (such as
northwestern Costa Rica, or the coast of eastern Surinam), on barrier reef islands
(Queensland, Australia; d’Entrecasteaux Reef, New Caledonia), and on remote oce
anic islands (e.g.. Ascension Island, Atol das Rocas). In many places it has long
been harvested for meat and eggs. Demand for international commerce is now an
insignificant factor, but has been replaced by increasing demand for subsistence and
local markets by indigenous people, whose population increase has often not been
matched by an increase in real wealth or political opportunity.
St",
'■ iT ^ u ‘
FIGURE 1.2 The green turtle, Chelonia mydas.
Thus, it remains to be seen whether the Tortuguero, Costa Rica, nesting colony
of the green turtle, although still large and the most important by far in the Caribbean,
can indefinitely survive heavy subsistence take by Costa Ricans from the city of
Limon as well as by Miskito people from Caribbean Nicaragua, San Bias people
from Panama, and Guajiros from northeastern Colombia and northwestern Venezu
ela. Similar uncertainties are faced by the large Australian colonies, protected whilst
in Australian territory and waters, but heavily harvested in various islands of Indo
nesia (Limpus, 1994, 1995).
Here and there, green turtle colonies seem to be on the increase. In the last
decade, not only has regular green turtle nesting been observed where none was
seen before at Rancho Nuevo, Tamaulipas, Mexico, but also, on the mid-Atlantic
coast of Florida, U.S.A., green turtle nesting is now commonplace on beaches where
almost nothing but loggerheads nested two decades ago. These increases on well-
patrolled beaches seem to be real, and with good protection now offered to many
nesting colonies, including those of Atol das Rocas, Ascension Island, and Trindade,
the green turtle does not appear to be faced with imminent extinction.
Parenthetically, it is interesting to note that proponents of turtle farming and
ranching have observed that, although the green turtle has been sought by interna
tional gourmets for centuries, it is no more endangered than other sea turtle species,
and thus commercial utilization could not have been a significant factor in any
decline that the species may have experienced (Fosdick and Fosdick, 1994).
1.11.2.2 The Black Turtle, Chelonia agassizii
This species (possibly only a subspecies of C. mydas) is confined to the eastern

Pacific. The species is protected in the Galapagos Islands and is nominally protected
in Mexico also, where the important nesting grounds in Maruata Bay are patrolled
by teams from the Universidad de Michoacan. Nonetheless, individuals from both
the mainland and Galapagos nesting grounds are caught in uncontrolled numbers in
Peruvian waters, and are also subject to illegal harvest on the Mexican and Central
American Pacific coasts, including the Gulf of California. Furthermore, marine
conservation efforts in the Galapagos Islands have been subjected to severe chal
lenges by settlers and fishermen in the last two years (Pritchard, 1996).
Thus, although the numbers are probably lacking to establish an unequivocal
status categorization following the new numerical criteria, the outlook for C. agas
sizii is, in the long run, uncertain. Much hinges on the fundamental question of
whether or not total protection of nesting turtles, their eggs, and their nesting beach
can offset uncontrolled harvest on the feeding grounds hundreds or thousands of
kilometers away. Answers to this key question are not yet available.
1.11.2.3 The Flatback, Natator depressus
This species is unlisted by the authorities of USDI and lUCN because it is essentially
confined to the waters of a single nation (Australia), where it is protected by law
except from aboriginal harvest, and where most of the nesting beaches are remote
and undeveloped (with the exception of the easily accessible Mon Repos Beach in
Queensland). Moreover, Australian native people in general prefer to consume the
more abundant and more succulent C. mydas. The outlook for Natator would thus
appear to be good, but the vulnerability of individuals to incidental capture in prawn
trawls, and the overall relatively restricted range lending panspecific vulnerability
from single catastrophic environmental events, leave no grounds for complacency.
FIGUKIC 1.3 "Hie black tuide, Chekmia agassizM.
FIGURE 1.4 The flatback, Natator depressus.
1.11.2.4 The Loggerhead, Caretta caretta

The loggerhead is little sought for its flesh, and although the eggs are gathered in
some parts of the world, direct take for human consumption is not a major factor
in its survival prospects. Rather, this species has an “antitropical” distribution that
not only fragments its overall range into well-separated enclaves in the northern and
southwestern Indian Ocean, eastern Australia, Japan, southeastern U.S., the Medi
terranean, and southern Brazil, but it also brings the species into contact with
industrial and development stresses ranging from massive incidental capture in
Atlantic shrimp trawls to resort and recreational development of nesting beaches.
At present, it appears that nesting populations are declining as a result of incidental
catch in both southern Queensland, Australia (Limpus, 1994) and in the U.S. north
of Cape Canaveral, but on the other hand the larger populations in Florida south of
Cape Canaveral and also the relatively small population in Natal, South Africa, are
increasing (National Research Council, 1990).
FIGURE 1.5 The loggerhead. Caretta caretta.
The small or medium-size populations in the Mediterranean, nesting primarily

in Zakynthos and other islands of Greece and the Turkish coast, and with important
feeding grounds off the coast of Tunisia, are threatened by a variety of factors. These
include intensive touristic development of many of the best nesting beaches. On the
other hand, recent information (Venizelos, 1996) identifies by far the most important
loggerhead nesting beach in the entire Mediterranean in eastern Libya, with an
estimated 9000 nests along 1250 km of coast, and with no foreseeable prospects for
touristic development.
1.11.2.5 The Hawksbill, Eretmochelys imbricata
Intense national and international trade in hawksbill shell products and, in more
recent decades, entire stuffed hawksbill turtles, has led to widespread concerns that
the species is being seriously overexploited. Moreover, although this may be the
most commonly seen species in certain tropical reef habitats in the Caribbean Islands
or tropical Australia, such populations generally consist primarily of subadults, and
very few nesting “colonies” (as opposed to isolated nesting females) are known.
Certain!; it is fundamentally less abundant than the green turtle, but downward
The green turtle
t i* Chelonia m ydas
"'J' '.'T i
The black turtle

Chelonia a g a ssizii
The flatback
Natator depressus.
The loggerhead
Caretta caretta
The hawksbill
Eretm ochelys im bricata
Olive ridley
Lepidochelys oUvacea
Kemp’s ridley
Lepidochelys kern p i
The leatherback
D erm ocheiys coriacea
trends are often difficult to determine from available data, and some of the more
dire warnings about imminent extinction of the species have stemmed simply from
the contrast between the very substantial global harvest for shell and other products,
and the modest size of the known nesting populations. In a global analysis of the
status of the hawksbill, Groombridge and Luxmoore (1989) could only classify it
as “indeterminate.”
FIGURE 1.6 The hawkshill, Eretmochelys imbricata.
For many decades, Japan has constituted the largest importer of raw tortoiseshell.
However, Japan has now withdrawn its reservation to this species on the CITES
convention, where it is listed as Appendix I (banned in international commerce),
and has phased out importation of hawksbill products. On the other hand, Cuba —
a major habitat for the hawksbill and supplier of world tortoiseshell markets — has
recently expressed the opinion that its domestic hawksbills not only constitute a
nonmigratory stock, but also have a sufficiently healthy population to justify down
grading to CITES Appendix II. The intention is to develop hawksbill ranches in
Cuba to supply the Japanese demand. The implications of this move are ambiguous,
but the conditions for approved commercial ranching of Appendix II species agreed
by the Parties to CITES in November 1994 are sufficiently demanding that a legal
Cuban turtle ranching program may well be linked to much better protection of
Cuban, and regional, hawksbill stocks than has hitherto been available.
In a review of worldwide population trends of the hawksbill, Limpus (1995)
could only find one example of an increasing hawksbill population — that of the
Sabah Turtle Islands (Malaysia), where a more than tenfold increase has been logged
since 1969 (Trono, 1994).
1.11.2.6 The Olive Ridley^ Lepidochelys oUvacea

The olive ridley, although having relatively localized nesting, remains the most
numerous species of sea turtle in the world as a result of the continued existence of
a few sites of enormously aggregated nesting — two in Pacific Costa Rica, one in
Pacific Mexico, and two or three in northeastern India, with some more minor sites
in Nicaragua and scattered nesting along certain other tropical mainland shores.
Whether or not the existence of such numbers of turtles on these few nesting beaches
is reason to believe that no problems exist is debatable. In all cases except for the
limited egg harvest program at Playa Ostional, Costa Rica, these “arribada” beaches
are nominally protected, although incidental take by trawlers is significant in both
Costa Rica and India, and the Indian sites may also be threatened by fishery devel
opment plans along the Orissa coast.
FIGURE 1.7 The olive ridley, Lepidochelys olivacea.
There is some evidence that major “arribadas” are not permanent phenomena,
but rather they may wax and wane over cycles of at least several decades. Several
Mexican arribadas have disappeared in recent decades, and numbers are dropping
rapidly at the protected Nancite site in Costa Rica. Furthermore, the recent discovery
of a new site in Orissa, India, may correspond to the collapse of one of the two
previously documented arribadas along this coast. A former minor arribada in Suri
nam (the only one in the Atlantic) has now been decimated, presumably by a
combination of heavy egg collection for several decades and ongoing trawler mor
tality.
1.11.2,7 The Kemp's Ridley, Lepidochelys kempi
A 1947 film exists of an arribada of L. kempi estimated at 40,000 turtles at Rancho

Nuevo, Tamaulipas, Mexico. Apart from a few isolated cases, this was and is the
only known nesting site for this species. However, no subsequent data were available
from Rancho Nuevo until 1965, by which time the biggest arribada numbered less
than 5000 turtles, and despite beach protection this number continued to drop for
the next 20 years, by which time total nestings for the season numbered in the
hundreds only. Such nonresponse to protection is a classic indication of a population
for which recruitment has been nullified by near-total harvest of eggs, in this case
exacerbated by the intensity of shrimp trawling in the Gulf of Mexico (the primary
juvenile and subadult habitat and the only habitat of adult L. kempi) and, during at
least one episode in the late 1960s, beach slaughter of adults also (P. Burchfield,
personal communication). The nesting population reached a low point in the mid-
1980s, and has subsequently increased modestly, but steadily, possibly as a result
of the “head-starting” program initiated in 1978, about 10 years (thought to corre
spond to the approximate maturation time of the species) after the commencement
of the head-starting program. The species remains the rarest sea turtle in the world,
but the current apparent annual population increase may identify it more as “con
servation dependent” than actually “endangered.”
i |l ® i
FIGURE 1.8 The Kemp’s ridley, Lepidochelys kempi.
1.11.2.8 The Leatherback, Dermochelys conacea
None of the major nesting grounds of this species were discovered before the 1950s,
and many of them only in the 1960s or 1970s, and thus it is impossible to compare
contemporary population estimates with those of earlier in the century. Pritchard
(1982) noted recent discoveries of major nesting grounds in Mexico, and summarized
preceding discoveries that led to progressively increasing world population esti
mates.
The leatherback does not feature in international commerce, and its juvenile
stages (indeed, all stages between hatchling and adult) remain so cryptic that it is
FIGURE 1.9 The leatherback, Dermochelys coriacea.
unlikely that humans have any effect upon them. On the other hand, subsistence
take of eggs, and sometimes of nesting adults also, has been intense, especially in
the Eastern Pacific and Guyana, and while Asiatic nesting colonies (such as that in
Terengganu, Malaysia) are generally exploited for eggs rather than meat, this can
be equally devastating.
At present, the Atlantic colonies (especially in Trinidad, Suriname, and French
Guiana) appear to be reasonably secure and even increasing, as is the small nesting
colony in Natal, South Africa, and adjacent Mozambique. On the other hand, the
Terengganu colony has collapsed in recent years, and serious declines have been
documented in Pacific Mexico and Costa Rica, a result of the combination of beach
slaughter, egg collection, and serious incidental captures by fishing gear in the open
sea. Spotila et al. (in press) speculate that the species may be threatened with actual
extinction within the forseeable future, although Pritchard (in press) disagreed.
REFERENCES
Agassiz, J. L. R., Contributions to the natural history of the United States, Voi. 1, Little,
Brown, Boston, 1857, 452 p.
Baird, D., Pneumatoarthrus Cope, 1870, not a dinosaur but a sea turtle, Proc. Acad. Natl.
Sci. Philadelphia, 129, 71-81, 1878.
Baur, G., The genera of the Cheloniidae, Am. Nat., 24, 486-487, 1890.
Bell, T, Characters of the order, families, and genera of the Testudinata, J. Zool, 3, 513-516,
1828.
Bocourt, M.-M., Description des quelques chéloniens nouveau appartenant à la faune mex
icaine. A/tn. Sci. Nat., ser. 5. Zool.. 10. 121-122. 1868.
Boulenger, G. A., Catalogue of the Chelonians, Rhynchocephalians, and Crocodiles in the

British Museum (Natural History), British Museum (Natural History), London, 1889,
311 p.
Bour, R. and Dubois, A., Nomenclature ordinale et familiale des tortues (Reptilia), in de
Broin, F. and Jimenez-Fuentes, E., Eàs., Studia Palaeocheloniologica 1, Ediciones Univ.
de Salamanca, Studia Geologica Salmanticensia, Vol. Especial 1 [1984]: 77-86, 1985.
Bowen, B. W., Nelson, W. S., and Avise, J. C., A molecular phylogeny for marine turtles:
trait mapping, rate assessment, and conservation relevance,Fwc. Acad. Nati. Sci. U.S.A.,
90, 5574-5577, 1993.
Brongersma, L. D., On the availability of the namQ Dermochelys schlegelii (Gannan, 1884),
as a species or subspecies of the leatherback turtle, Chelonian Conservation and Biology,
in press.
Brongniart, A., Essai d’une classification naturelle des reptiles,5w//. Sci. Soc. Phil., 35, 81-82;
36, 89-91, 1800.
Brongniart, A., Essai d’une classification naturelle des reptiles, Badouin, Paris, 1805, 53 p.
CaiT, A. F., Notes on sea turtles, Proc. N. Engl. Zool. Club, 21, 1-16, 1942.
CaiT, A. F., Handbook of turtles, Comstock Assoc., Cornell University Press, Ithaca, NY,
1952,542 p.
Cope, E. D., Synopsis of the extinct Batrachia and Reptilia of North America, Appendix to
the Crocodilia, Trans. Am. Philos. Soc. (N.S.), 14(1), 83-85, 1870.
Cope, E. D., On the homologies of some of the cranial bones of the Reptilia and on the
systematic aiTangement of the class, Proc. Am. Assoc. Adv. Sci., 19, 194-246, 1871.
de Broin, F. and Pironon, B., Découverte d’une tortue dermochelyidée dans le Miocène d’Italie
centro-meridionale (Matese oriental). Province de Benevento, Riv. Ital. Palaeontol.
Stratigr, 86, 589-604, 1980.
Deraniyagala, P. E. P, The tetrapod reptiles of Ceylon, Colombo Museum, Ceylon, 1939,
412 p.
Deraniyagala, P. E. P., A colored atlas of some vertebrates from Ceylon, Vol. 2, Tetrapod
Reptilia, 1953, 101 p.
Dolio, L., Premierò note sur les chéloniens du Landenien (Eocene inferieur) de la Belgique,
Bull. Mus. R. Hist. Nat. Belg., 4, 129, 1886.
Dumeril, A. M. C. and Bibron, G., Eipétologie generale ou histoire naturelle complète des
reptiles, vol. 3, Librairie Encyclopédique de Roret, Paris, 1836, 680 p.
Fosdick, P. and Fosdick, S., Last Chance Lost?, Irving Naylor, York, Pennsylvania, 1-338,
1994.
Frair, W., Taxonomie relations among sea turtles elucidated by serological itsi^,Herpetolog-
ica, 35, 239-244, 1979.
Frair, W., Ackerman, R. G., and Mrosovsky, N., Body temperature of Dermochelys coriacea:
warm turtle from cold water. Science, i l l , 791-793, 1972.
Fry, D. B., On the status of Cheionia depressa, Carman. Ree. Aust. Mus., 10, 159-185, 1913.
Gaffney, E., A phylogeny and classification of the higher categories of turtles, i9w//. Am. Mus.
Nat. Hist., 155, 387-436, 1975.
Gaffney, E., The cranial morphology of the extinct homed ixxriXt, Meiolania platyceps, from
the Pleistocene of Lord Howe Island, Australia, ilw//. Am. Mus. Nat. Hist., 175, 361-480,
1983.
Gaffney, E. S., The comparative osteology of the Triassic turtle Proganochelys, Bull. Am.
Mus. Nat. Hist., 194, 1-263, 1990.
Gaffney, E. S., Hutchinson, J. H., Jenkins, F. A., and Meeker, L., Modern turtle origins: the
oldest known cryptodire. Science (AAAS), 237, 289-291, 1987.
Gaffney, E. S. and Meylan, P. A., A phylogeny of turtles, in Benton, M. J., Eá.,The Phylogeny
and Classification o f Tetrapods, Voi. 1, Amphibia, Reptilia, Birds, The Systematics
Association, Clarendon Press, Oxford, Spec. Voi. 35A, 157-219, 1988.
Garman, S., On certain species of the Chelonioidae,ßw//. Mus. Comp. Zool., 6 , 123-126, 1880.
Gervais, R, Nouv. Arch. Mus. Hist. Nat. (Paris), 8, 1872.
Gould, S. J., An early start, in The Panda’s Thumb. More Reflections in Natural History,
Norton, New York, 1980, 217-226.
Gray, J. E., A synopsis of the genera of reptiles and amphibia with a description of some new
species, Ann. Philos., ser. 2, 10, 193-217 (reprint 1966, Soc. Study Amphib. Reptiles).
Greer, A. E., Lazell, J. D., and Wright, R. M., Anatomical evidence for a counter-cuiTent heat
exchanger in the leatherback turtle (Dermochelys coriacea), Nature, 244, 181, 1973.
Groombridge, B. and Luxmoore, R., The green turtle and hawkshill (Reptilia: Cheloniidae):
world status, exploitation and trade, CITES, UNEP, 1989, 601 p.
Haworth, A. H., A binary arrangement of the Class Amphihm, Philos. Mag. (London), 65,
372-373, 1825.
Hay, O. P., The fossil turtles of North America, Carnegie Inst. Washington PubL, 75, 1-568,
1908.
Hutchison, J. H., Turtle, crocodilian and champsosaur diversity changes in the Cenozoic of
the North-Central region of western United States, Palaeogeogr. Palaeoclimatol. Palae-
oecol, 37, 149-164, 1982.
Jenyns, L., Report on the recent progress and present state of zoology. Report ... British
Association, 1834, 143-251, 1835.
Kamezaki, N. and Matsui, M., Geographic variation in skull morphlogy of the green turtle,
chelonia mydas, with a taxonomic discussion,/. HerpetoL, 29(1), 51-60, 1995.
Kennett, R., Of droughts and flooding rains and a turtle that nests underwater. Second world
Congress of Herpetology, Abstracts, 1993, 138-139.
Klein, Quadrup. Disp. Lipsiae 1751. (not seen; cited by Agassiz, 1857).
Lee, M. S. Y, The origin of the turtle body plan: bridging a famous morphological gap.
Science, 261, 1716-1720, 1993.
Limpus, C. J., Marine conservation. Marine turtles: ancient mariners in distress. Air Sea
Rescued. (Australia), 12(2), 99-100, 103, 105, 106-113, 1994.
Limpus, C. J., Global overview of the status of marine turtles: a 1995 viewpoint, in K.
Bjorndal, Ed., Biology and Conservation o f Sea Turtles, rev. ed., Smithsonian Institution
Press, Washington, D.C., 1995, 605-609.
Limpus, C. J., Gyuris, E., and Miller, J. D., Reassessment of the taxonomic status of the sea
turtle genus NATATOR McCulloch, 1908, with a redescription of the genus and species,
Trans. R. Soc. South Aust., 112(1), 1-9, 1988.
Linnaeus, C., Systema naturae per regna triae naturae, secundum classes, ordines, genera,
species, cum characteribus, differentiis, synonymis, locis, Ed. 10, Tomus 1. L. Salvii,
Stockholm, 1758, 823 p.
Loveridge, A. and Williams, E. E., Revision of the African tortoises and turtles of the suborder
Cryptodira, Bull. Mus. Comp. Zool., Harv. Univ., 115, 163-557, 1957.
Lydekker, R., Catalogue of the Fossil Reptilia in the British Museum (Natural History), Part
3, Chelonia, British Museum (Natural History), London, 1889, 239 p.
McCulloch, A. R., A new genus and species of turtle from North Australia,/?cc. Austr. Mus.,
1, 126-128, 1908.
Merrem, B., Tentamen Systematis Amphibiorum, Marburg, Germany, 1820, 191 p.
Mertens, R. and Wermuth, H., Die rezenten Schildkröten, Krokodile und Brückenechsen,
Zool. Jahrb. (Syst.), 83, 5, 323-440, 1955.
Meyer, F. I. C., System des Thierreichs, etc., Verh. Naturforsh Ver. Rheinland, 1949.
Meylan, R, The phylogenetic relationships of soft-shelled turtles (family Trionychidae),5w//.
Am. Mus. Nat. Hist., 186(1), 1-101, 1987.
Mlynarski, M., Fossile Schildkröten, A. Ziemsen, Wittenber, 1969, 128 p.
Mlynarski, M., Results of the Polish-Mongolian Palaeontological Expeditions — Part IV.

Zangerlia testudinimorpha n. gen. n. sp. a primitive land tortoise from the Upper Cre
taceous of Mongolia, Palaeontol. Pol., 27, 85-92, 1972.
Mlynarski, M., Handbuch der Palaoherpetologie, Part 7, Testudines, Gustav Fisher, Stuttgart,
1976, 130 p.
Mrosovsky, N., Conserving Sea Turtles, British Heipetological Society, London, 1983, 176 p.
National Research Council, Decline o f the Sea Turtles. Causes and Prevention, National
Academy Press, Washington, D.C., 1990, 259 p.
Nielsen, E., Eocene turtles from Denmark, Mcd<7c/. Danskl. Geol. Foren., 14, 96-114, 1959.
Oppel, M., De Ordnungen, Familien und Gattungen der Reptilien, als Prodrom einer Naturge
schichte derselben, J. Lindauer, München, 1811, 87 p.
Parham, J. F. and Zug, G. R., Chelonia agassizii — valid or not?, Mar. turtle Newsl., 72, 2-5,
1996.
Pritchard, P. C. H., Studies of the Systematics and Reproductive Cycles of the G qüus Lepi-
dochelys. Doctoral dissertation. University of Florida, Gainesville, 1969, 197 p.
Pritchard, P. C. H., Encyclopedia of Turtles, T. F. H. Publications, Jersey City, NJ, 1979, 895 p.
Pritchard, P. C. H., Nesting of the leatherback iurilc, Dermochelys coriácea, in Pacific Mexico,
and a new estimate of the world population of the species. Copda, 1982, 741-747, 1982.
Pritchard, P. C. H., Conserving sea turtles, by N. Mrosovsky (book review), Copc/d, 1983,
1108-1111, 1983.
Pritchard, P. C. H., Evolutionary relationships, osteology, morphology, and zoogeography of
Kemp’s ridley sea turtles, 157-164, in Caillouet, C. W. and Landry, A. M., Eds., Proc.
1st hit. Symp. on Kemp’s ridley sea turtle biology. Conservation and Management,
Galveston.
Pritchard, P. C. H., Cladism: the great delusion,//crp. Rev., 25(3), 103-110, 1994.
Pritchard, P. C. H., The Galapagos tortoises. Nomenclatural and Survival Status, Monogr. No.
1, Chelonan Research Foundation, 1996, 85 p.
Pritchard, P. C. H. and Trebbau, R, The turtles of Venezuela, Soc. Study Amphib. Rept. 1984,
404 p.
Rhodin, A. G. J., Comparative chondro-osseous development and growth of marine turtles,
Copeia, 1985(3), 752-771, 1985.
Ritgen, F. A., Versuch einer natürlichen Eintheilung der Amphibien, Aova Acr Nat. Cur, vol.
14, 1828.
Römer, A. S., Vertebrate Palaeontology, 2nd ed.. University of Chicago Press, Chicago, 1945,
687 p.
Römer, A. S., Osteology o f the Reptiles, University of Chicago Press, Chicago, 1956, 772 p.
Smith, H. M. and Smith, R. B., Synopsis of the Herpetofauna o f Mexico. VI. Guide to Mexican
Turtles, J. Johnson, North Bennington, VT, 1980, 1044 p.
Spotila, J. R., Dunham, A. E., Leslie, A. J., Steyermark, A. C., Plotkin, P. T, and Paladión,
F. V., Population decline of the leatherback turtle, Dermochelys coriácea: are leather
backs going extinct?, Che Ionian Conservation and Biology, Lunenburg, in press.
Swinton, W. E., Fossil Amphibians and Reptiles, British Museum (Natural History), British
Museum (Natural History), London, 1958, 118 p.
Trono, R. B., The Philippine-Sabah Turtle Islands: a critical management area for sea turtles
of the ASEAN region, in Proc. 1st ASEAN Symposium-Workshop on Marine Turtle
Conservation, Manila, Philippines, Manila: World Wildlife Fund, 167-180, 1993.
Venizelos, L., Mediterranean loggerheads are thriving in Libya!, Marmc Turtle Newsl., No.
72, 2, 1996.
Wagler, J. G., Natürliches System der Amphibien, Cotta, Munich, 1830, 354 p.
Williams, E. E., Grandison, A. G. C., and Carr, A. F., Chelonia depressa Garman re-investi
gated, Breviora, Mus. Comp. Zool. Harv. Univ., 271, 1-15, 1967.
Wood, R. C., Evolution of the pelomedusid turtles, Stud. Geol. Salmanticensia, Vol. Especial
1 (1984), 269-282, 1985.
Yates, B., An American love affair. 100 years of the automobile in America, Lz/e Magazine,
Winter 1996, 11-116.
Zangerl, R., The vertebrate fauna of the Selma Formation of Alabama. III. The turtles of the
family Protostegidae, Fieldiana G eol Mem., 3, 59-133, 1953.
Zangerl, R., Patterns of phylogenetic differentiation in the toxochelyid and cheloniid sea
turtles. Am. Zool., 20, 585-596, 1980.
Zangerl, R., Hendrickson, L. P, and Hendrickson, J. R., A redescription of the Australian
Flatback S&di'T\xn{Q,Natator depressus, Bishop Mus. Bull. Zool., Honolulu, 1,1-69,1988.
Zangerl, R. and Turnbull, W. D., Procolpochelys grandaeva (Leidy), an early carettine sea
turtle. Fieldiana Zool., 37. 345-382. 1953.
2 Population Genetics,
Phylogeography, and
Molecular Evolution
Brian W. Bowen and Stephen A. Karl
CONTENTS
2.1 Introduction...................................................................................................... 30
2.1.1 Technical Considerations.................................................................... 30
2.2 Molecular Systematics and Evolution............................................................ 31
2.2.1 D erm ochelys coriácea: an Ancient Lineage?................................. 32
2.2.2 E retm ochelys im bricata: the Origin of an Unusual
Dietary Habit....................................................................................... 33
2.2.3 N atator depressus: a Recent Taxonomic Reassignment....................33
2.2.4 L epidochelys kem pii and L. olivácea: Biogeography and
Conservation....................................................................................... 34
2.2.5 C helonia m ydas and C. agassizi: When is a Green Turtle not
a Green Turtle?................................................................................... 35
2.2.6 Marine Turtle Hybrids: the Oldest Bastards Known
to Science............................................................................................ 36
2.3 Green Turtle Population Structure.................................................................. 38
2.3.1 Maternal Genetic Markers: a Test of Natal Homing.........................38
2.3.2 Nuclear vs. Mitochondrial DNA: Contrasting Population
Genetic Structures within an Organismal Lineage............................38
2.4 Loggerhead Turtle Population Structure........................................................40
2.4.1 Regional Population Structure: the Southeast U.S.............................40
2.4.2 Feeding Ground Composition............................................................40
2.5 Hawksbill Turtle Population Structure...........................................................43
2.5.1 Population Genetic Structure: the Geographic
Scale of Nesting Populations..............................................................43
2.5.2 Genetic Markers and International Conservation..............................44
2.6 Summary......................................................................................................... 45
References................................................................................................................46
0-8493-8422-2/97/$0.00+$ 50
© 1997 by CRC Press, Inc 29
2.1 INTRODUCTION
The age of biotechnology is upon us, and even the most venerable branches of
natural history have been sculpted by the advancing tide. Although the application
of genetic analyses to marine turtles is a relatively recent undertaking, research
efforts have already proven informative. The success of genetic studies is attributable
in part to the paucity of information on marine turtle biology prior to the last decade.
In a group of little-known animals, relatively simple genetic observations (such as
a comparison of DNA sequences in putative sister species) may constitute substantial
scientific advances.
Molecular genetic assays have special attributes which are particularly well
suited for examining cryptic components of marine turtle life history and evolution.
The DNA in each cell can provide genetic markers to identify individuals, popula
tions, species, and higher taxa. In addition, DNA sequence comparisons can illumi
nate the evolutionary history of sea turtles, as well as aspects of reproductive behavior
and ecology. For example, do female sea turtles return to nest on their natal beaches?
A maternal genetic marker such as mitochondrial DNA (mtDNA) is appropriate for
tracing the dispersal of female lineages.^ Do males return to their natal rookeries,
or do they mate with females from other nesting beaches? Analysis of nuclear DNA
(nDNA) fragments can reveal the level of gene flow mediated by males. Does more
than one father contribute to a clutch? Microsatellite and minisatellite DNA analyses,
popularly known as DNA fingerprinting, can resolve the paternal contribution to a
cluster of eggs. How distinct is the nominal black turtle, Che Ionia agassizi, from
the green turtle, C. mydasl Because DNA sequences diverge at a more or less
predictable pace over evolutionary time scales, they are useful for reconstructing
phylogenetic relationships, especially for taxonomic groups which are difficult to
distinguish morphologically. The slow pace of morphological evolution in marine
turtles makes the application of molecular systematics especially useful for these
animals.
Molecular studies of marine turtle systematics have advanced beyond prelimi
nary stages, and now include nDNA as well as mtDNA phylogenies. Studies of
several of these species are currently underway, and many of the gaps in our
knowledge will be bridged in the next decade as research programs on five continents
have begun to apply molecular genetic assays to facets of marine turtle biology.
2.1.1 Technical Considerations

Through the late 1980s, direct DNA sequencing was relatively difficult and generally
beyond the resources available for investigations of natural history. A simple and
robust alternative is to survey gene sequences with DNA cutting enzymes, restriction
endonucleases, which cleave the DNA at specific 4-, 5-, or 6-base sequences.^ Much
of the early molecular work on sea turtles and other vertebrates was accomplished
by analyzing restriction fragment length polymorphisms (RFLPs).
Another early challenge to molecular genetic analysis was in comparing the
same (homologous) piece of DNA from each sampled individual. The nuclear
Population Genetics, Phylogeography, and Molecular Evolution 31
genome consists of tens of thousands of genes, and identifying the same gene from
scores of individuals was an imposing technical challenge. Initially, this problem
was circumvented by working with the extranuclear mitochondrial genome. The
mtDNA of vertebrates is about 16,000 to 17,000 nucleotides, a convenient size to
assay with restriction endonucleases. mtDNA also is maternally inherited in verte
brates, which allows the elucidation of gender-specific behaviors (such as female
nest site fidelity).
One of the most powerful and exciting additions to the molecular genetic
toolbox in recent years is the polymerase chain reaction (PCR).^-^ This process
allows researchers to synthesize millions of copies of a particular segment of DNA
from only a few starting molecules. Pieces of either the mitochondrial or nuclear
genome can now be efficiently amplified and characterized. The PCR process can
be applied to minute pieces of DNA-containing tissue such as a hair follicle, a toe
clip, or a single drop of blood. Furthermore, PCR techniques do not require fresh
tissue; samples can be preserved in ethanol, detergent-based preservative, or a
saturated salt (NaCl) buffer.^’^ With this approach, large numbers of individuals
from natural populations can be sampled relatively easily and without causing them
harm.
One of the fastest evolving segments of the mtDNA molecule is the control
region (sometimes referred to as the D-loop). This region is not a gene coding for
a protein, but rather the site of the origin for replication of the mtDNA molecule.
The relatively rapid rate of mutation allows fine-scale resolution of populations, and
for this reason control region DNA sequencing has become the method of choice
for population definition with mtDNA.^
Fine-scale population discrimination can also be accomplished with a relatively
new class of hypervariable nDNA m arkers.^T hese nDNA regions, often termed
microsatellite loci, generally consist of a region of repeated pairs, triplets, or qua
druplets of DNA nucleotides (such as GTGTGTGT... or AACAACAAC...). Undoubt
edly, this approach will prove valuable for the assessment of fine-scale population
structure and male-mediated gene flow in marine turtles.
2.2 MOLECULAR SYSTEMATICS AND EVOLUTION

To address issues of phylogenetic affiliation, researchers have collected biochemical
data,^^ immunological data,^"^ protein electrophoretic data,^-^ mtDNA sequences,
and nDNA sequences. A summary of evolutionary trees based on molecular genetic
data sets shows broad areas of congruence (Figure 2.1).
Perhaps the most fertile grounds for molecular systematic investigations are at
the subgeneric levels, where fine-scale evolutionary differentiation in marine turtles
is difficult to interpret against a background of ecotypic variation. Environmentally
induced variation is generally of unknown magnitude, such that true evolutionary
differences are difficult to distinguish from morphological plasticity. Molecular
genetic data provide a new and independent line of evidence to help resolve these
systematic controversies.
mtDNA PHYLOGENY
nDNA PHYLOGENY
*C. CARETTA
L. KEMPIl
L. OLIVACEA
>E. IMBRICATA
N. DEPRESSUS
C. MYDAS
D. CORIACEA
C. SERPENTINA
FIGURE 2.1 Phylogenetic relationships among marine turtles based on mtDNA'^ and
anonymous nDNA sequences.'^ Phylogenetic relationships were inferred by parsimony criteria
(PAUP, version 3.0.5)^'' and statistical support was evaluated by bootstrap resampling. Boot
strap values expressed as percentages are indicated at the nodes.
2.2.1 D e r m o c h e l y s c o r ia c e a i an A ncient L ineage?

The evolutionary distinction of the leatherback turtle, D. coriacea, from other sea
turtles has evoked particularly diverse opinions. The leatherback is distinguished
from other marine turtles by unusual skeletal features, partial endothermy, and a
highly modified external morphology. Based on this distinct morphology, Cope^^
suggested that D. coriacea was the sister taxon to all other extant turtles, and erected
Population Genetics, Phyiogeography, and Molecular Evolution 33
a suborder (Athecae) to distinguish the shell-less leatherback turtle. This designation

has been advocated intermittently through the 20th c e n tu r y .^ O th e r researchers
have maintained that differences between the leatherback and other marine turtles
are not fundamental ones, and these researchers prefer to distinguish D erm ochelys
at the family, subfamily, or genus level (reviewed in Reference 25; see also Refer
ences 13, 14, 26). Inherent in this debate is the question of whether or not marine
turtles are monophyletic: did the adaptation to the marine environment occur more
than once during the evolutionary history of turtles? A polyphyletic origin for marine
turtles would imply that members of the Testudines invaded the marine realm at
least twice.
Molecular phylogenies reveal a deep partition between D. coriacea and the other
marine turtles (Figure 2.1). The comparison of Dermochelyidae and Cheloniidae to
a cryptodiran outgroup supports the monophyly of marine turtles (relative to these
outgroups) with both nDNA and mtDNA a sse ssm e n ts.W h ile additional out
groups and nDNA sequence data are desirable to test this conclusion, available
evidence indicates that the leatherback turtle represents an early offshoot of the
lineage that gave rise to the other extant marine turtles. The adaptation to the marine
environment probably arose just once during the evolutionary history of extant
turtles. Hence the suborder designation Athecae is not supported by DNA sequence
analysis.
2.2.2 Eretmochelys imbricata: the Origin of an Unusual

Dietary Habit
Another debate centers on evolutionary relationships of the spongivorous hawksbill

turtle, E. im bricata. One systematic school assigns this species to the subfamily
Chelonini, typihed by the herbivorous green turtle, C. m ydas, while a second school
affiliates this species with the subfamily Carettini, typified by the carnivorous log
gerhead turtle, Caretta caretta.^^’^^ Since the capacity to exploit sponges as a food
source is rare among vertebrates,^^ it is of interest to frame this systematic issue in
terms of evolutionary ecology: did the dietary habit of the hawksbill turtle arise from
a herbivorous or a carnivorous ancestor?
Genetic analyses are concordant in indicating that the hawksbill turtle is a
member of the tribe C a r e t t i n i . W i t h i n this tribe, the nDNA data indicate a near
trichotomy for the genera E retm ochelys, Caretta, and Lepidochelys, and mtDNA
sequences support a C arettajLepidochelys grouping.W hile the relationships among
the three Carettini genera can be clarified by additional nDNA sequence compari
sons, the subfamily affiliation of E retm ochelys is unequivocal. The strength of the
subfamily affiliations permits the inference that since all extant members of Carettini
are carnivorous, the dietary habit of the hawksbill arose from a carnivorous ancestor.
2.2.3 Natator depressus: a Recent Taxonomic Reassignment
Until 1988 the flatback turtle was labeled C helonia depressa and was considered a
close relative of the green turtle, C. m ydas. In a taxonomic reassessment based on
morphology and biochemical evidence, Limpus et resurrected the genus N atator
for the flatback turtle. Zangerl et al.^‘ went further in invoking a third tribe within
Cheloniidae, Natatorini, for this species.
The mtDNA sequence data support the genus-level distinction of the flatback
turtle (N. depressus), and are not consistent with the earlier assignment to the genus
Chelonia}^'^'^ An expanded mtDNA data set indicates that flatback lineage is more
closely related to the Carettini than the Chelonini.^^ The nDNA data do not fully
resolve the phylogenetic position of N. depressus. Sequence data from additional
nuclear loci would be desirable to clarify this situation.
2.2.4 Lepidochelys kempii and L, olivácea: Biogeography

and Conservation
Over the last century, the Kemp’s and olive ridleys have been alternately described
as subspecies, species in the genus L epidochelys, or members of the loggerhead
genus Carettad^^^^ The two ridley forms are readily distinguished from each other
by coloration, but no external morphological character will unambiguously distin
guish these sister taxa. Hence some question exists regarding the evolutionary
distinctiveness of the Kemp’s ridley.
In the first thorough examination of ridley evolution, Pritchard^^ suggested that
progenitors of the Kemp’s and olive ridleys were initially isolated by the Central
American land bridge, now known to be 3 million to 4 million years old.^Minder
this scenario, the olive ridley may have evolved in the Indian-Pacific Ocean and only
recently invaded the Atlantic Ocean via South Africa. Pritchard^^ concluded that the
two forms merit species-level distinction, based on morphology and osteology.
However, this taxonomic decision required careful deliberation because of the overall
moiphological conservatism of the genus.
Both mtDNA and nDNA sequence data support the evolutionary distinctiveness
of the Kemp’s ridley. The levels of divergence between L. kem pii and L. olivácea
are consistent with the species-level assignments recommended by Pritchard:^^
mtDNA divergence between ridley forms is greater than the divergence observed
within other species of marine turtle^^’^"^ (but see Reference 17). These genetic data
also provisionally support the biogeographic scenario proposed by Pritchard.^^ Olive
ridleys in the Atlantic and Pacific oceans are closely related in terms of mtDNA
sequences and are divergent from Kemp’s ridley at approximately the level expected
for 3 million to 4 million years of isolation.
The DNA sequence data alone cannot indicate whether or not L. kem pii and L.
olivácea are good species. Nonetheless, the results of both the mtDNA and nDNA
studies indicate that these two groups are genetically distinct at a level which
indicates a long evolutionary separation. This genetic distinctiveness reinforces
Pritchard’s^^ conclusion that the two ridley forms are unique evolutionary units
meriting species-level designations. Accordingly, genetic data bolster the rationale
for intensive conservation efforts for L. kem pii.
2.2.5 Chelonia mydas and C. agassizh When Is a Green

Turtle not a Green Turtle?
Subspecific status has been proposed for several regional forms of the green turtle,
including populations in the Caribbean (C. m. viridis), South Atlantic (C. m. mydas),
Indo-West Pacific (C. m. japónica), Gulf of California (C. m. carrinegra), and East
Pacific (C. m. agassizi)?^'^^ Only the east Pacific form is widely accepted as a distinct
taxonomic entity, and this form is accorded full species status as C. agassizi, the
black turtle. However, the scientific evidence in support of this species designation
is s p a r s e . I n the most complete morphological analysis conducted to date,
Kamezaki and Matsui^^ examined skull characters in seven Chelonia populations,
concluding that the subtle differentiation of East Pacific specimens was consistent
with a subspecific designation, C. m, agassizi.
To resolve the global phylogeography of Chelonia, Bowen et al.^^ examined
mtDNA in 226 specimens from 15 locations across the tropical range of Chelonia,
and Karl et al.^^ surveyed the same populations with nDNA assays. A striking feature
of the mtDNA phytogeny is a grouping of observed haplotypes into two assemblages
that correspond precisely to major oceanic basins: (a) the Atlantic Ocean and Med
iterranean Sea; and (b) the Indian and Pacific Oceans (Figure 2.2a). This partition
is consistent with the biogeographic boundaries of green turtle distribution: the
southern extensions of Africa and South America are effective barriers to dispersal
of tropical fauna between Atlantic and Indian-Pacific r e gi ons The mtDNA analysis
does not indicate that the black turtle represents a unique lineage relative to other
green turtle populations and hence does not support a taxonomic distinction of East
Pacific Chelonia populations.
Because nDNA lineages appear to evolve more slowly than mtDNA lineages
(and therefore provide less resolution for shallow evolutionary separations), a robust
phylogeographic analysis was not possible with RFLP analysis of nDNA loci.^^
Nonetheless, nDNA RFLP data provide more support for population subdivision
elsewhere within C. mydas than for the specific status of C. a g a s s iz i.Subsequent
sequence analysis of nDNA reinforces the lack of genetic distinctiveness of C.
agassizi (Figure 2.2b).'^^
While the debate over C. agassizi has yet to be resolved, the recent evidence
seems to have moved the question from the species to the subspecies arena. The
mtDNA data support an Atlantic-Mediterranean vs. Indian-Pacific grouping, while
the morphological data^^ support an East Pacific subspecies. At present we are aware
of no strong scientific evidence for an East Pacific Chelonia species."^^
The mydas!agassizi debate has rekindled in recent years in part as a result of
dire trends in the East Pacific populations. In considering the conservation implica
tions of taxonomic status, one point bears consideration: independent of taxonomic
assignment, the distinctiveness of the East Pacific populations has been demonstrated
in terms of biogeographic isolation,"^® skull morphology,^^ and population-level
genetic partitions (mtDNA and nDNA genotype frequency shifts), and this is suffi
cient to warrant conservation efforts. However, it would be a mistake to support
species designations as a political tool to enhance conservation efforts.
POLYNESIA (n=2)
AUSTRALIA 100%
JAPAN 5%
JAPAN 70%
JAPAN 25%
HAWAII 73%
HAWAII 27%
GALAPAGOS 100%
PAC. MEXICO 100%
OMAN 100%
(POLYNESIA (n=1)
VENEZUELA 12%
COSTARICA 100%
FLORIDA 88%
FLORIDA 12%
( CYPRUS 100%
VENEZUELA 88%
SURINAM 100%
ASCENSION 3%
ASCENSION 97%
BRAZIL 94%
BRAZIL e%
WEST AFRICA 100%
r
0.8
”T
0.6
"T
0.4 0.2
n
0.0
SEQUENCE DIVERGENCE (%)
FIGURE 2.2a Phylogenetic tree summarizing the relationships among fourteen haplotypes
observed in a mtDNA RFLP survey of green turtle nesting colonies (n = 226)?*^ The geographic
location(s) and the frequencies of each haplotype at these locations are indicated to the right.
Haplotype frequencies for Polynesia were omitted due to small sample size (n = 3).
2.2.6 Marine Turtle Hybrids: the Oldest Bastards Known to

Science
Molecular genetic techniques have proven useful in the documentation of several

cases of natural hybridization of marine turtles."^^ One case of hybridization was
discovered serendipitously during global population genetic surveys of green and
loggerhead turtlesT^’"^'^ Four hatchling clutch-mates from Brazil, originally thought
to be green turtles, unexpectedly displayed loggerhead mtDNA restriction profiles
and cytochrome b gene sequences. All four individuals were screened at several
nDNA loci and revealed DNA patterns characteristic of both green and loggerhead
FIGURE 2.2b Phytogeny summarizing the relationships among 18 iiDNA genotypes from
across the range of Chelonia (1446 nucleotides from loci CM-12, CM-14, and CM-45). This
tree represents a consensus of 171 equally parsimonious arrangements, with the frequency
of each node in these trees indicated as a percentage. No clear distinction is observed between
Atlantic-Mediterranean and Indian-Pacific lineages. However, both nDNA and mtDNA trees
contradict arguments for the evolutionary or taxonomic distinction of the black turtle.
turtles. The combination of the nDNA and mtDNA evidence indicates that these
hatchlings were first-generation hybrids between a loggerhead female and a green
turtle male. Three other occurrences of natural hybridization also have been docu
mented with molecular genetic techniques — a hybrid between a loggerhead female
and a hawksbill male was found in Florida; a Kemp’s ridley female/loggerhead male
hybrid discovered in Chesapeake Bay during the summer of 1992; and a possible
second (or later)-generation hybrid of a female green turtle and male hawksbill
originally collected in Surinam in 1977 and maintained at Cayman Turtle Farm."^^
The existence of these hybrids is remarkable because the tribes Carettini and
Chelonini reflect an ancient division within the family Cheloniidae, dating to perhaps
50+ mya."^^ These crosses may be the oldest vertebrate lineages known to hybridize
in nature. Predictions about the impact of hybridization on natural populations are
difficult to make. It is therefore imperative that wildlife managers consider the
potential for hybridization before manipulating the natural histories of endangered
species.
2.3 GREEN TURTLE POPULATION STRUCTURE

2.3.1 Maternal G enetic Markers: a T est of N atal H oming
The strong nesting site fidelity observed in marine turtles prompted Carr^^ and others
to propose that the nesting site of a female green turtle is also her natal rookery.
HendricksoiT^ proposed an alternative explanation for female nest site fidelity,
labeled the “social facilitation hypothesis.” Under the social facilitation hypothesis,
first-time nesting females follow experienced breeders from the feeding grounds to
a nesting beach, and use this site for all subsequent nesting."^^
“Natal homing” and “social facilitation” hypotheses have proven difficult to test
directly, as no tag applied to a hatchling has been successfully recovered from an
adult."^^ Natal site philopatry, however, generates a testable prediction about the
genetic structuring of populations. If females return faithfully to their rookery of
origin, then each nesting population should be effectively isolated in terms of female-
transmitted traits (such as mtDNA). In contrast, social facilitation would allow high
rates of female-mediated gene flow between rookeries that share feeding grounds.
Tagging data have demonstrated that Brazilian feeding pastures are shared by
turtles from at least two nesting colonies, located in Surinam and Ascension
Island.^^ '’®This co-occupation of feeding habitat by two major nesting colonies is
an excellent basis for genetic tests of the natal homing and social facilitation hypoth
eses.
Analysis of mtDNA RFLP data and control region sequences demonstrate that
individuals from the Surinam rookery (n = 15) possess a haplotype at 100% fre
quency that is not observed at Ascension Island (n = 35, Figure 2.3). Despite
extensive overlap on feeding grounds, there are fixed genetic differences (e.g., no
sharing of haplotypes) between Surinam and Ascension s a m p l e s . T h e s e genetic
data constitute strong evidence of a barrier to female dispersal between nesting
populations, and thus are consistent with expectations of the natal homing hypoth
esis.
A second case involves the two major nesting locations on the Great Barrier
Reef. Heron Island represents the primary nesting area in the southern Great Barrier
Reef, and Raine Island is the largest nesting colony in the northern Great Barrier
Reef. Tagging data demonstrate that these populations overlap extensively along the
margins of Australia and the Coral Sea.^^ Despite this overlap, Raine Island (n =
15) and Heron Island (n = 15) are characterized by a nearly fixed difference in the
distribution of mtDNA haplotypes,^ again supporting natal homing in this species.
2.3.2 N uclear vs. M itochondrial DNA: C ontrasting

Population G enetic Structures within an
O rganismal L ineage
The analysis of multiply independent nDNA markers has helped to address several
of the issues concerning gender-specific gene flow and potential biases encountered
by single locus a s s a ys . I n a companion to the global mtDNA survey described
above, Karl et al.^^ used the same samples to investigate the degree of nDNA genetic
MATAPICA,
SURINAM
FIGURE 2.3 Green turtle nesting colonies at Matapica, Surinam and Ascension Island,
U.K. and distribution of tag recoveries along the coast of South America. Dark shading
indicates the area of overlap in feeding habitat along the coast of BraziL^^’-'’^The distribution
of mtDNA haplotypes (“C”, “D”, and “E”) are indicated for each rookery. Despite extensive
overlap on feeding grounds, no mtDNA haplotypes are shared between samples from these
two nesting colonies, consistent with a natal homing hypothesis for the migrations of nesting
females.^^’^^
isolation among nesting beaches. Five independent nDNA loci were screened for
RFLPs and the results were compared to the mtDNA analysis. Overall, there was a
low level of genetic variation observed in green turtles, consistent with previous
surveys using protein electrophoresis.“’^’^^ An analysis of nDNA interpopulation
genetic differentiation revealed that, although not as pronounced as the mtDNA data,
green turtle populations on a global scale are genetically isolated from each other.
The overall estimate of intraocean gene flow based on nuclear loci is higher than
that observed for mtDNA, but still indicative of restricted gene flow between nesting
colonies.
One of the primary goals of this study was to compare genetic results from the
strictly maternally inherited mtDNA with those of the biparentally inherited nDNA,
allowing the disentanglement of maternal from paternal influences on population
genetic structuring. The Ascension Island and Surinam nesting populations provide
an appropriate arena for such comparisons. The estimates of mtDNA (i.e., female)
gene flow between these populations was zero, because no mtDNA haplotype was
shared between populations. When assayed for genetic differences at the five nDNA
loci, the two populations were indistinguishable from each other. Estimates of
nuclear gene flow between the two populations were quite high, indicating that there
is current, or recent historical, gene flow between these two populations. This nDNA
gene flow must be mediated by males. Comparisons of gene flow based on these
nDNA and mtDNA data sets must be made with caution due to the low level of
variation detected at assayed nDNA loci. Nonetheless, the conclusion of occasional
male-mediated gene flow between nesting colonies warrants further investigation.
FitzSimmons et al.^^ have developed a suite of highly polymorphic nDNA
markers that may be capable of resolving many of the remaining questions concern
ing male-mediated gene flow and population structure in marine turtles.
2.4 LOGGERHEAD TURTLE POPULATION

STRUCTURE
2.4.1 R egional Population S tructure: the Southeast U.S.
Many of the questions posed for green turtles apply to loggerhead turtles as well.
Do loggerhead females return to nest on their natal beach? Are nesting populations
genetically distinct?
The nesting beaches of the southeastern U.S. are appropriate for examining
regional population structure. This area hosts one of the largest nesting aggregates
in the world, covering thousands of kilometers of coastline from Virginia to Texas.
However, nesting effort is not evenly distributed, nor is nesting continuous along
these coasts. In particular, southern Florida hosts the vast majority of nesting efforts
(perhaps 90%),^^ and this area is separated from other nesting areas (Georgia on the
Atlantic coast and the Florida panhandle in the Gulf of Mexico) by several hundred
kilometers of coastline in which nesting density is low.
To elucidate regional populations structure within this area, Bowen et al.'’^
surveyed nesting areas in South Carolina, Georgia, and Florida. RFLP analysis of
mtDNA revealed two haplotypes which accounted for 96% of nest samples. Despite
this low level of diversity, the distribution of these haplotypes has a strong geographic
component. A haplotype observed in 66% of southern Florida samples (n = 29), was
completely absent from the Georgia (n = 44) and South Carolina (n = 19) collections
(Figure 2.4). In contrast, proximal nesting sites (Georgia and South Carolina or
southeast Florida and southwest Florida) were not significantly different in haplotype
composition. Hence, nesting loggerhead of the southeastern U.S. comprise at least
two populations corresponding to southern Florida and Georgia/South Carolina
groupings. Analyses of mtDNA control region sequences and microsatellite loci are
likely to illuminate additional aspects of population structure. Nonetheless, the
current genetic definition of nesting populations in the southeastern U.S. is consistent
with the natal homing hypothesis for reproductive migrations.
2.4.2 Feeding G round C omposition

In many cases, marine turtle researchers have identified nesting areas and regional
feeding habitats, but do not know which nesting aggregate uses which feeding area.
FIGURE 2.4 Distribution of mtDNA haplotypes on nesting beaches of the southeastern

U.S. No significant haplotype frequency shifts are obsei*ved between Georgia and South
Carolina or between southeast and southwest Florida. However, haplotype “D”, observed at
66% frequency in southern Florida samples, is absent in Georgia and South Carolina samples,
indicating a restriction on contemporary gene flow between southern Florida and Geor-
gia/South Carolina nesting populations.^‘^
Mark/recapture studies have revealed the presence of cohorts from specific nesting
populations on feeding g r o u n d s , b u t the relative contribution of each regional
rookery to a feeding area remains unknown in most cases (but see Limpus et al.).^^
Satellite tracking has solved the problem of following individual turtles during
oceanic movements, and has revealed fascinating aspects of behavior and ecology.^®
However, the limited sample sizes currently available in satellite tracking studies do
not allow appraisals of population-level phenomena such as feeding ground com
position.
The unknown link between nesting populations and feeding aggregates
stands as one of the primary gaps in the scientific understanding of marine turtle
life history. Fortunately, the presence of rookery-specific genetic markers (and
significant haplotype frequency shifts between rookeries) affords an opportunity

to bridge this gap.^‘ Mixed stock assessment typically uses a maximum likelihood
algorithm to determine the relative contribution from several source populations
which best explains the genotype frequencies observed in the presumably
“mixed” aggregate.
Only a few genetic assessments of feeding grounds exist to date.^"^ To determine
the origin of turtles feeding in the Charleston Harbor entrance channel, Sears et
al.^'^ compared mtDNA haplotypes from feeding turtles to the haplotype frequen
cies at the Florida and Georgia/South Carolina nesting populations. If turtles from
these nesting areas were recruiting to the feeding grounds in numbers proportional
to their population size, then Florida turtles should make up the majority (approx
imately 90%) of the Charleston Harbor feeding population. Contrary to this expec
tation, results of the mixed stock assessment indicate that approximately 50% of
the turtles feeding in the harbor are from the Georgia/South Carolina nesting
population and approximately 50% are from the Florida nesting beaches. Thus,
the smaller Georgia/South Carolina nesting population is recruiting preferentially
to the Charleston feeding area, and therefore is especially vulnerable to distur
bances to habitats along this coast. Research now in progress should determine
whether or not this conclusion applies to other feeding habitats on the Atlantic
coast of North America.
A second conservation concern involves turtles captured incidentally in marine
fisheries. Loggerhead turtles are vulnerable to driftnets and longlines in the Pacific
Ocean, and in this case conservation concerns are intertwined with a remarkable
facet of loggerhead natural history. In the last three decades researchers have
documented a feeding aggregate of Juvenile loggerheads off the coast of Baja
Califomia.^^’’^'^ This population poses a prominent enigma because the nearest log
gerhead nesting beaches are in Japan and Australia, over 10,000 km distant. Where
do the Baja California loggerheads come from? Based on the recovery of a single
tagged turtle in the eastern Pacific, Uchida and Teruya^^ suggested that these turtles
may be coming from Japan. This hypothesis was not widely accepted because a
trans-Pacific migration would greatly exceed the known geographic scale of marine
turtle migrations.^^ The prevailing explanation for the presence of these juveniles
in Baja California involved an undiscovered rookery in the eastern Paci fi c. How
ever, the coastlines of Central and South America have been extensively surveyed
over the last two decades, and no evidence of a loggerhead nesting has been
forthcoming.^^
When samples from North Pacific driftnet mortalities (n = 34) and the Baja
California feeding aggregate (n = 26) were compared to the nesting beach haplotypes,
95% of these individuals matched the two haplotypes observed at Japanese nesting
beaches (Figure 2.5).^^ The remaining 5% may be derived from Australian nesting
beaches, although additional sampling would be necessary to confirm this. Japanese
loggerhead turtles apparently traverse the entire Pacific Ocean, about one third of
the planet, during juvenile migrations.
FIGURE 2.5 Distribution of loggerhead mtDNA haplotypes in Pacific nesting areas, North
Pacific driftnet fisheries, and Baja California feeding grounds?^ These data indicate that
juvenile loggerhead turtles from the nesting beaches in Japan traverse the North Pacific Ocean
in the course of developmental migrations. The westerly Northern Equatorial Current may
facilitate a return migration to adult habitats in the western Pacific. Juvenile turtles have been
recovered from New Zealand, Chile, and Colom bia,învoking the possibility that some of
the turtles from the nesting colony in Queensland, Australia undertake a similar transoceanic
migration.
2.5 HAWKSBILL TURTLE POPULATION STRUCTURE

2.5.1 Population G enetic Structure: the G eographic S cale
OF N esting Populations
The hawksbill turtle is considered to be a solitary nester in some locations, but at
least part of this solitary behavior may be attributed to the vastly reduced numbers
of this turtle relative to historical levels. Do female hawksbills return to their natal
beach? Broderick et al.'^^ found nearly fixed differences in the distribution of mtDNA
(control region) haplotypes between nesting areas in northeast and northwest Aus
tralia, but pairs of nesting beaches within each region were not distinct in terms of
haplotype frequencies. Bass et al.^"^ documented significant haplotype frequency
shifts between seven West Atlantic rookeries, demonstrating strong population struc
ture within the Caribbean region. These data are consistent with expectations of
natal homing. As is the case for green and loggerhead turtles, each hawksbill nesting
population constitutes a distinct demographic entity.
Both Broderick et al.^^ and Bass et al.^"^ report that in some cases nesting
aggregates separated by a few hundred kilometers are not distinct in terms of mtDNA
haplotypes. Lack of differentiation may be attributed in part to historical factors
such as recent colonization events. Movement between adjacent nesting habitats also
may play a role in preventing population subdivisions on a microgeographic scale.

The combined evidence from tagging data (occasional nest relocations) and mtDNA
data (sharing of haplotypes between proximal rookeries) indicate that nesting pop
ulations may encompass several regional nesting sites.
2.5.2 G enetic Markers and I nternational C onservation

Genetic markers can play a role in identifying the origin of marine turtle feeding
populations, but how can these data influence conservation strategies? The hawksbill
turtle is widely sought for the beautiful translucent scales, which may be worked
into jewelry, sculpture, and a variety of vanity items. As a result, many populations
were greatly reduced or hunted to extinction. Meylan^^ estimated that no more than
10.000 nesting females remain in the Caribbean. At the same time, annual harvest
in the Caribbean (including males, juveniles, and females) may have exceeded
10.000 turtles.^^ In view of the alarming depletion of this species, international trade
has been prohibited under CITES convention. The Bekko Corporation of Japan
continued to import hawksbill shell until December 1992, at which time the gov
ernment of Japan instituted a moratorium. At the time of this writing a local trade
continues in many countries, and there is continuing interest in reopening the lucra
tive international trade.
In response to the demand for hawksbill shell, the government of Cuba
announced in 1992 an intention to resume the commercial harvest of hawksbill
turtles on the reefs within Cuban territorial waters.'^^ The harvest on foraging habitat
is based on a fishery model which assumes that hawksbill turtles are nonmigra-
toryT^’^9 important point: if only turtles from Cuban nesting beaches are
taken, then no other countries are affected by this harvest. Marine turtle biologists
were skeptical of this assumption, but little information existed on the composition
of feeding grounds. Indeed, early reports suggested that hawksbill turtles are less
migratory than other marine turtles (see Reference 80), consistent with the fishery
model proposed by Doi et al.^^
To test the assumptions of this fishery model, members of the Archie Carr Center
for Sea Turtle Research (University of Florida) with Anna Bass (Louisiana State
University) applied a genetic test to the fishery model. In the simplest version of
this test, mtDNA from individuals from a nesting population at Mona Island, Puerto
Rico, was compared to samples from an adjacent juvenile feeding aggregate moni
tored by C. Diez (University of Central Florida) and R. van Dam (Scipps Institute
of Oceanography). If hawksbill turtles are nonmigratory, as assumed in the fishery
model, then the Mona Island nesting population and feeding aggregate should be
similar in terms of mtDNA haplotype composition. Contrary to predictions of the
fishery model, feeding ground samples were significantly different from the nest
samples.
Mixed stock assessment with the program UCON^^ indicated that the Mona
Island feeding area contains turtles from throughout the Caribbean region. Notably,
the surveyed nesting colony in Bahia, Brazil, did not contribute to the Mona Island
feeding aggregate at detectable levels. Bowen et al.^’ conclude that turtles recruit to
feeding grounds on a geographic scale of several hundred kilometers, but not over
the scale of 7000 km which separates Bahia, Brazil and Mona Island, Puerto Rico.
Therefore, a harvest of hawksbill turtles on feeding grounds within the territorial
waters of one nation will affect nesting populations within the territorial boundaries
of other nations.
2.6 SUMMARY
In the near future, relatively complete genetic inventories of nesting populations for
each species are likely to be completed. These inventories represent a milestone in
the conservation genetics of marine turtles. The (foreseeable) advantages of range
wide surveys are threefold:
a. Forensic identification of marine turtles or commercial by-products. DNA

from meat or shell can be used to identify the species and geographic
origin of marine turtle products in the marketplace. In a recent test case,
a green turtle recovered from a cargo vessel in California was determined
to be from a nesting colony in the southern or eastern Atlantic.
b. Definition of management units for this group of threatened and endan
gered reptiles. Analyses of maternally inherited genetic markers (mtDNA)
and biparentally inherited markers (nDNA) yield an especially rich body
of information about population structure and sex-specific dispersal.
c. Identification of feeding ground cohorts. Genetic markers and mixed stock
assessments can identify the source of turtles in feeding areas which are
vulnerable to human disturbance.
In addition to these benefits, molecular data are likely to illuminate other aspects
of natural history (such as paternity) which will have conservation applications
currently unforeseen by wildlife managers and biologists. In view of these ongoing
conservation benefits, it is regrettable that scientific permit restrictions continue to
play a role in defining the pace of conservation-oriented research.
The genetic data available to date prompt the following conclusions:
* Phylogenetic analyses of DNA sequences indicate that the leatherback

turtle is an early offshoot of the lineage which gave rise to Cheloniid sea
turtles. Thus the adaptation to marine habitats probably arose just once
during the evolutionary history of extant turtles. Sequence comparisons
support affiliation of the hawksbill turtle with the tribe Carettini, and
support the evolutionary distinction of the Kemp’s and olive ridley.
Sequence comparisons do not support the evolutionary distinction of the
East Pacific green turtle, Chelonia agassizi.
* Genetic data confirm the existence of hybrids among the deepest evolu
tionary lineages within Cheloniidae. These marine turtle taxa are possibly
the oldest species known to hybridize in nature.
* In green, loggerhead, and hawksbill turtles (the Cheloniid species sur
veyed to date), mtDNA data demonstrate strong population genetic struc
turing among regional nesting populations. These data are consistent with
natal homing and reinforce the concept that marine turtle nesting colonies
constitute fundamental units of wildlife management.
The nDNA data indicate that significant levels of gene flow occur between
some regional nesting populations and that this gene flow is mediated
primarily by males. More sensitive nDNA (microsatellite) analyses are
cuiTently being applied to elucidate genetic structure over finer geographic
scales. Sequence data from nDNA loci has generally been concordant
with the phylogenetic inferences based on mtDNA sequences.
Rookery-specific mtDNA polymorphisms can be used to resolve the con
tribution of nesting populations to a particular feeding ground. Genetic
“tags” demonstrate that loggerhead turtles cross the Pacific Ocean in the
course of developmental migrations. Mixed stock assessments indicate
that loggerhead and hawksbill feeding grounds typically contain turtles
from several nesting populations. These data are concordant with tagging
studies in demonstrating the international nature of marine turtle conser
vation.
REFERENCES
1. Avise, J. C., Mitochondrial DNA polymorphism and a genetics-demography connec
tion of conservation relevance, Conserw B iol, 9, 686, 1995.
2. Dowling, T. E., Mortiz, C., and Palmer, J. D., Nucleic acids II: restriction site analysis,
in Molecular Systematics, Hillis, D. M. and Moritz, C., Eds., Sinauer Associates,
Sunderland, 1990, 250.
3. Mullis, K., Faloona, F., Scharf, S., Saiki, R., Horn, G., and Erlich, H., Specific
enzymatic amplification of DNA in vitro: the polymerase chain reaction, Co/d Spring
Harbor Symp. Quant. Biol., 51, 263, 1986.
4. Saiki, R. K., Gelfand, D. H., Stoffel, S., Scharf, S. J., Higuchi, R., Horn, G. T., Mullis,
K. B., and Erlich, H. A., Primer-directed enzymatic amplification of DNA with a
thermostable DNA polymerase. Science, 239, 487, 1988.
5. Innés, M. A., Gelfand, D. H., and Sninsky, J. J., PCR Strategies, Academic Press,
San Diego, 1995.
6. Amos, B. and Hoelzel, A. R., Long-term preservation of whale skin for DNA analysis,
Rep. Int. Whal. Comm., Special Issue, 13, 99, 1991.
7. White, P. S. and Densmore, L. D., Mitochondrial DNA isolation, mMolecular Genetic
Analysis o f Populations: a Practical Approach, Hoelzel, A. R., Ed., IRL Press at
Oxford University Press, New York, 1992, 29.
8. Norman, J. A., Moritz, C., and Limpus, C. J., Mitochondrial DNA control region
polymorphisms: genetic markers for ecological studies of marine turtles. Mo/. Ecol.,
3, 363, 1994.
9. Allard, M. W., Miyamoto, M. M., Bjorndal, K. A., Bolten, A. B., and Bowen, B. W.,
Support for natal homing in green turtles from mitochondrial DNA sequences,
Copeia, 1994, 34, 1994.
10. Lahanas, P. N., Miyamoto, M. M., Bjorndal, K. A., and Bolten, A. B., Molecular
evolution and population genetics of Greater Caribbean green turtles (CheIonia
mydas) as inferred from mitochondrial DNA control region sequences, Genetica, 94,
57. 1994.
11. Tautz, D., Hypervariability of simple sequences as a general source for polymorphic
DNA markers, Nucleic Acids Res., 17, 6436, 1989.
12. Charlesworth, B., Sniegowski, P., and Stephan, W., The evolutionary dynamics of
repetitive DNA in eukaryotes. Nature, 371, 215, 1994.
13. Chen, B. Y., Mao, S. H., and Ling, Y. H., Evolutionary relationships of turtles
suggested by immunological cross-reactivity of albumins. Comp. Biochem. Physiol.,
66B, 421, 1980.
14. Frair, W., Taxonomic relations among sea turtles elucidated with serological tests,
Herpetologica, 35, 239, 1979.
15. Frair, W., Serum electrophoresis and sea turtle classification, Comp. Biochem. Phys
iol, 72B, 1, 1982.
16. Bowen, B. W., Nelson, W. S., and Avise, J. C., A molecular phylogeny for marine
turtles: trait mapping, rate assessment, and conservation relevance, Froc. Natl. Acad.
Sci. U.S.A., 90, 5574, 1993.
17. Dutton, P. H., Davis, S. K., Guerra, T., and Owens, D., Molecular phylogeny for
marine turtles based on sequences of the ND4-leucine tRNA and control regions of
mitochondrial DNA, Mol. Phyl. EvoL, 5, 511, 1996.
18. Karl, S. A., Sweger, D., and Severance, E., unpublished data, 1996.
19. Frair, W., Ackman, R. G., and Mrosovsky, N., Body temperature of Dermochelys
coriácea: warm turtle from cold water. Science, 111, 791, 1972.
20 . Pritchard, P. C. H. and Trebbau, P., The Turtles of Venezuela, mSociety fo r the Study
o f Amphibians and Reptiles. Fundación de Internados Rurales, Contrib. Herpetol.,
2, Caracas, Venezuela, 1984.
21. Rhodin, A. G. J., Comparative chondro-osseous development and growth of marine
turtles. Copela, 1985, 752, 1985.
22 . Cope, E. D., On the homologies of some of the cranial bones of the Reptilia, and on
the systematic arrangement of this class, Proc. Am. Assoc. Adv. Sci., 19, 194, 1871.
23. Hay, O. P., Further consideration of the shell of Chelys and of the constitution of the
armor of turtles in general, Proc. U.S. Natl. Mus., 73, 1, 1928.
24. Carr, A. F, Handbook o f Turtles, the Turtles o f the United States, Canada, and Baja
California, Cornell University Press, Ithaca, NY, 1952.
25. Pritchard, P. C. H., Encyclopedia o f Turtles, T. F. H. Publications, Neptune, NJ, 1979.
26. Zug, G. R., The penial morphology and the relationships of cryptodiran turtles,Occa^.
Pap. Mus. ZooL, Univ. Mich., 647, 1, 1966.
27. Carr, A. F, Notes on sea turtles, Proc. N. Engl. Zool. Club, 21, 1, 1942.
28. Hendrickson, J. R., Ecological strategies of sea turtles. Am. Z ool, 20, 597, 1980.
29. Meylan, A. B., Spongivory in hawksbill turtles: a diet of glass. Science, 239, 393,
1988.
30. Limpus, C. J., Gyuris, E., and Miller, J. D., Reassessment of the taxanomic status of
the sea turtle genus Natator McCulloch, 1908, with redescription of the genus and
species, Trans. R. Soc. S. Aust., 112, 1, 1988.
31. Zangerl, R., Hendrickson, L. P., and Hendrickson, J. R., A redescription of the
Australian flatback sea turtle, Natator depressus. Bishop Mus. Bull. Zool, 1,1, 1988.
32. Pritchard, P. C. H., Studies of the Systematics and Reproductive Cycles of the Genus
Lepidochelys, Dissertation thesis, University of Florida, 1969.
33. Keigwin, L. D., Isotopic paleooceanography of the Caribbean and East Pacific: role
of Panama uplift in Late Neocene time. Science, 111, 350, 1982.
34. Bowen, B. W., Meylan, A. B., and Avise, J. C., Evolutionary distinctiveness of the
endangered Kemp’s ridley sea íuxÚq, Nature, 352, 709, 1991.
35. Can', A. F, The Ascension Island green turtle colony, Copela, 1975, 547, 1975.
36. Mrosovsky, N., Conserving Sea Turtles, British Heipetological Society, London,
1983.
37. Kamezaki, N. and Matsiii, M., Geographic variation in the skull moiphology of the
green turtle, Chelonia mydas, with a taxonomic discussion,/. HerpetoL, 29, 51, 1995.
38. Bowen, B. W., Meylan, A. B., PeiTan-Ross, J., Limpus, C. J., Balazs, G. H., and
Avise, J. C., Global population structure and natural history of the green turtle
{Chelonia mydas) in terms of matriarchal phylogeny. Evolution, 46, 865, 1992.
39. Karl, S. A., Bowen, B. W., and Avise, J. C., Global population structure and male-
mediated gene flow in the green turtle {Chelonia mydas): RFLP analyses of anony
mous nuclear DNA regions, Genetics, 131, 163, 1992.
40. Briggs, J. C., Marine Zoogeography, McGraw-Hill, New York, 1974.
41. Karl, S. A. and Bowen, B. W., unpublished data, 1996.
42. Zug, G. R., Chelonia agassizii — valid or not?. Mar. Turtle News, 72, 2, 1996.
43. Karl, S. A., Bowen, B. W., and Avise, J. C., Hybridization among the ancient mariners:
characterization of marine turtle hybrids with molecular genetic markers,/. Hered.,
86, 262, 1995.
44. Bowen, B. W., Kamezaki, N., Limpus, C. J., Hughes, G. H., Meylan, A. B., and
Avise, J. C., Global phylogeography of the loggerhead turtle {Caretta caretta) as
indicated by mitochondrial DNA haplotypes, 48, 1820, 1994.
45. Avise, J. C., Bowen, B. W., Lamb, T., Meylan, A. B., and Bermingham, E., Mito
chondrial DNA evolution at a turtle’s pace: evidence for low genetic variability and
reduced mieroevolutionary rate in the Testudines, Mol. Biol. EvoL, 9, 457, 1992.
46. Carr, A. F., So Excellent a Eishe: a Natural History o f Sea Turtles, Scribner, New
York, 1967.
47. Hendriekson, J. R., The green sea turtle, Chelonia mydas (Linn.) in Malaya and
Sarawak, Proc. Zool. Soc. London, 130, 455, 1958.
48. Owens, D. W., Grassman, M. A., and Hendrickson, J. R., The imprinting hypothesis
and sea turtle reproduction, Herpetologica, 38, 124, 1982.
49. Carr, A. F, Rips, FADS, and little Xoggtvhc^diàs, BioScience, 36, 92, 1986.
50. Pritchard, P. C. H., Post-nesting movements of marine turtles (Cheloniidae and Der-
mochelyidae) tagged in the Guianas, Copeia, 1976, 749, 1976.
51. Encalada, S. E., Lahanas, P. N., Bjomdal, K. A., Bolten, A. B., Miyamoto, M. M.,
and Bowen, B. W., Phylogeography and population structure of the green turtle
{Chelonia mydas) in the Atlantic Ocean and Mediterranean Sea: a mitochondrial
DNA control region sequence assessment. Mol. EcoL, 5, 473, 1996.
52. Meylan, A. B., Bowen, B. W., and Avise, J. C., A genetic test of the natal homing
vs. social facilitation models for green turtle migration, Science, 248, 724, 1990.
53. Limpus, C. J., Miller, J. D., Parmenter, C. J., Reimer, D., McLachland, N., and Webb,
R., Migration of green {Chelonia mydas) and loggerhead {Caretta caretta) turtles to
and from eastern Australian rookeries, Wildl. Res., 19, 347, 1992.
54. Karl, S. A. and Avise, J. C., PCR-based assays of Mendellian polymorphisms from
anonymous single copy nuclear DNA: techniques and applications for population
genetics. Mol. Biol. EvoL, 10, 342, 1993.
55. Smith, M. H., Hillstad, H. O., Manlove, M. N., Straney, D. O., and Dean, J. M.,
Management implications of genetic variability in loggerhead and green sea turtles,
in Proc. 13th Int. Cong. Game Biol, 1978, 302.
56. Bonhomme, R, Salvidio, S., LeBeau, A., and Pasteur, G., Comparison genetique des
tortues vertes {Chelonia mydas) des Oceans Atlantique, Indien et Pacifique: une
illustration apparente de la théorie mullerienne classique de la structure genetique
des populations?, Genetica, 74, 89, 1987.
57. FitzSimmons, N. N., Moritz, C., and Moore, S. S., Conservation and dynamics of
microsatellite loci over 300 million years of marine turtle evolution, Mr?/. Biol. Evol.,
12, 432, 1995.
58. Murphy, T. M. and Hopkins-Murphy, S. R., Sea turtle and shrimp fishing interactions:
a summary and critique of relevant information. Center for Marine Conservation,
Washington, D.C., 1989.
59. Bowen, B. W., Avise, J. C., Richardson, J. I., Meylan, A. B., Margaritoulis, D., and
Hopkins-Murphy, S. R., Population structure of loggerhead turtles (Caretta caretta)
in the northwestern Atlantic Ocean and Mediterranean Sea, Conserv. Biol., 7, 834,
1993.
60. Plotkin, P. T., Byles, R. A., Rostal, D. C., and Owens, D. W., Independent vs. socially
facilitated oceanic migrations of the olive ridley, Lepidochelys olivácea, Mar. Biol.,
122, 137, 1995.
61. Broderick, D., Population Genetics of Hawksbill Turtles, B.Sc. thesis. University of
Queensland, St. Lucia, 1992.
62. Pella, J. J. and Milner, G. B., Use of genetic markers in stock composition analysis,
in Population Genetics and Fishery Management, Ryman, N. and Utter, R, Eds.,
University of Washington Press, Seattle, WA, 1987, 247.
63. Xu, S., Kobak, C. J., and Smouse, P. E., Constrained least squares estimation of
mixed population stock composition from mtDNA haplotype frequency data, Can.
J. Fish. Aquat. Sci., 51, 417, 1994.
64. Bowen, B. W., Tracking marine turtles with genetic markers, BioScience, 45, 528,
1995.
65. Sears, C. J., Bowen, B. W., Chapman, R. W., Galloway, S. B., Hopkins-Murphy,
S. R., and Woodley, C. M., Demographic composition of the juvenile loggerhead
sea turtle (Caretta caretta) feeding population off Charleston, South Carolina:
evidence from mitochondrial DNA markers, Mar. Biol., 123, 869, 1995.
66. Bartlett, G., Juvenile Caretta off the Pacific coast of Baja California, Not. Caguamas,
2, 1, 1989.
67. Ramirez Cruz, J. C., Pena Ramirez, I., and Villanueva Flores, D., Distribution y
abundancia de la tortuga perica. Caretta caretta Linnaeus (1758), en la costa occi
dental de Baja California Sur, Mexico Archelon, 1, 1, 1991. (Available from BIT-
MAR: Banco de Infonnation sobre Tortugas Marinas, Instituto de Ciencias del Mar
y Limnologia, Apartado Postal 811, Mazatlan, Sinaloa, Mexico 82000).
68. Uchida, S. and Teruya, H., Transpacific migration of a tagged loggerhead. Caretta
caretta, and tag-return results of loggerheads released from Okinawa, Japan, in
International Symposium on Sea Turtles 1988 in Japan, Uchida, I., Ed., Hemeji City
Aquarium, Hemeji City, Japan, 1991, 171.
69. Meylan, A. B., Sea turtle migration — evidence from tag returns, in Biology and
Conservation of Sea Turtles, Bjorndal, K. A., Ed., Smithsonian Institution Press,
Washington, D.C., 1982, 91.
70. Frazier, J., Misidentification of sea turtles in the eastern Pacific: Caretta caretta and
Lepidochelys olivácea, J. Herpetol., 19, 1, 1985.
71. Amorocho, D. F, Rubio, H., and Diaz, W., Observaciones sobre el estado actual de
las tortugas marinas en el Pacifico Columbiano, in Contribución al Conocimiento de
las Tortugas Marinas de Colombia, Mahencha, J. V. R. and Paez, H. S., Eds., Inst.
Nacional de los Recursos Naturales Renovables y del Ambiente, Bogota, 1992.
72. Bowen, B. W., Abreu-Grobois, F. A., Balazs, G. H., Kamezaki, N., Limpus, C. J.,
and Ferì, R. J., Trans-Pacific migrations of the loggerhead sea turtle demonstrated
with mitochondrial DNA markers. Proc. Natl. Acad. Sci. U.S.A., 92, 3731, 1995.
73. Broderick, D., Moritz, C., Miller, J. D., Guinea, M., Prince, R. J., and Limpus, C. J.,
Genetic studies of the hawksbill turtle: evidence for multiple stocks and mixed feeding
grounds in Australian waters, Pac. Conserv. Biol., 1, 123, 1994.
74. Bass, A., Good, D. A., Bjorndal, K. A., Richardson, J. I., Hillis, Z. M., Horrocks,
J. A., and Bowen, B. W., Testing models of population structure and female migra
tory behavior in the Caribbean hawksbill turtle, Eretmochelys imhricata, with
mtDNA control region sequences. Mo/. EcoL, 5, 321, 1996.
75. Meylan, A. B., Status report of the hawksbill turtle, in Proceedings o f the Second
Western Atlantic Turtle Symposium, Ogren, L., Berry, F., Bjorndal, K., Kumpf, H.,
Mast, R., Medina, G., Reichart, H., and Witham, R., Eds., NOAA Tech. Memo.
NMFS-SEFC-226, National Technical Information Service, Springfield, IL, 1989,
101.
76. Canin, J., international trade in sea turtle products, mProc. 9th Annual Workshop on
Sea Turtle Biology and Conservation, Eckert, A., Eckert, K. L., and Richardson,
T H., Compilers, NOAA Tech. Memo. NMFS-SEFC-232S, National Technical
Information Service, Springfield, IL, 1989, 27.
77. Ottenwalder, J. A. and Ross, J. P., The Cuban sea turtle fishery: description and needs
for management, in Proc. 11th Annual Workshop on Sea Turtle Biology and Conser
vation, Salmon, M. and Wyneken, J., Compilers, NOAA Tech. Memo. NMFS-SEFC-
302, National Technical Information Service, Springfield, VA, 1992, 90.
78. Doi, T, Marquez, R., Kimoto, H., and Azeno, N., Diagnosis and conservation of
hawksbill turtle population in the Cuban archipelago. The Bekko Industry, Japan,
unpublished technical report, 1992.
79. Heppell, S. S. and Crowder, L. B., Analysis of a fisheries model for harvest of
hawksbill sea turtles, Conserv. Biol., 10, 874, 1996.
80. Carr, A. F, Hirth, H., and Ogren, L., The ecology and migration of sea turtles. 6.
The hawksbill turtle in the Caribbean Sea, Am. Mus. Novit., 2248, 1, 1966.
81. Bowen, B. W., Bass, A. L., Garcia-Rodrigues, A., Diez, C. E., van Dam, R., Bolten,
A., Bjorndal, K. A., Miyamoto, M. M., and Ferl, R. J., Origin of hawksbill turtles
{Eretmochelys imhricata) in a Caribbean feeding area as indicated by mitochondrial
DNA seuqence analysis, Ecol. Appl., 6, 566, 1996.
82. Masuda, M., Nelson, S., and Pella, J., User manual for GIRLSEM, GIRLSYM and
CONSQRT, USA-DOC-NOAA-NMFS, Programs and user manual available from
US-Canada Salmon Program, Juneau, AK, 1991.
83. Encalada, S. E., Eckert, S. A., and Bowen, B. W., Forensic applications of mitochon
drial DNA markers: origin of a confiscated green turtle. Mar. Turtle Newsl, 66, 1,
1994.
84. Swofford, D. L., PAUP: phylogeny analysis using parsimony, version 3.0.5, Illinois
Natural History Survey, Champaign, IL, 1991.
85. Limpus, C., personal communication, 1995.
Q Reproduction In Sea
Turtles
Jeffrey D. Miller
CO NTENTS
3.1 Introduction..................................................................................................... 52
3.2 Adults.............................................................................................................. 52
3.2.1 Distribution......................................................................................... 52
3.2.2 Size and Maturity................................................................................ 54
3.2.3 Reproductive Behavior....................................................................... 54
3.2.3.1 Philopatry.............................................................................55
3.2.3.2 Nest Site Fidelity.................................................................55
3.2.3.3 Periodicity........................................................................... 55
3.2.3.4 Courtship and Mating................................................... 57
3.2.3.5 Nesting Activity...................................................................58
3.2.3.6 Nesting Process....................................................................60
3.2.4 Reproductive Potential........................................................................ 63
3.2.4.1 Shape and Size of Eggs.......................................................64
3.2.4.2 Number of Eggs and Clutches............................................64
3.3 Embryos.......................................................................................................... 66
3.3.1 Fertility and Infertility........................................................................ 66
3.3.2 Development....................................................................................... 66
3.3.3 Nest Environment............................................................................... 67
3.4 Hatchlings....................................................................................................... 68
3.4.1 Hatching Process................................................................................. 68
3.4.2 Hatchling Emergence.......................................................................... 68
3.4.3 Hatching Success................................................................................ 69
3.5 Conservation................................................................................................... 69
3.5.1 Nesting Beaches.................................................................................. 69
3.5.2 Greenhouse Effect...............................................................................70
References................................................................................................................71
()-8493-8422-2/97/$0 00+$.50
3,1 INTRODUCTION
Reproduction in marine turtles (Cheloniidae, Dermochelyidae) occurs within three
general constraints.^ First, nesting must occur during conditions which “are condu
cive to adult activity”.^ Second, nesting must occur during conditions which facilitate
embryonic development and survival. Third, hatchlings must emerge into conditions
that “are conducive to their survival”.^ Within these constraints, sea turtles share a
number of general reproductive characteristics. All species exhibit (1) iteroperous
reproduction^’^ with the possible exception of the Kemp’s ridley,^ (2) stereotyped
nesting behavior,"^ (3) laying of relatively large numbers of eggs several times during
the reproductive period,^’'’ and (4) relatively strong attachment to a particular location
for nesting,^’^ but inter- and intraspecific variations exists.^’^^ Together these con
straints and characteristics form the context of an ecologically successful reproduc
tive strategy.
Because most ecological studies on sea turtles have been conducted on the beach
during nesting, most deal with some aspect of reproduction. As a result the literature
concerning the reproductive biology of sea turtles is immense, albeit uneven. The
myriad research results concerning mating, nesting, behavior, reproduction, distri
bution, and population dynamics have been published in books,^'^"^ symposia,
r e v i e w j o u r n a l a rt i c l e s, s p e c ie s accounts,^^-^^ and collected references.^^’^^
In the current review, no attempt has been made to cite every paper that deals with
the topic; preferentially, reference has been made to the papers that provide clear,
concise information and that provide linkages into other relevant sources.
3.2 ADULTS
The seven species of sea turtles that reproduce on beaches around the world, Der~
m ochelys coriácea (leatherback turtle), C helonia m ydas (green turtle), Caretta
caretta (loggerhead turtle), E retm ochelys im hricata (hawksbill turtle), L epidochelys
olivácea (olive ridley), L. kem pi (Kemp’s ridley), and N a ta to r depressus (flatback
turtle), share a common life cycle with only minor variation^’^’^^ (Figure 3.1). All
species migrate, at least short distances, from foraging areas to mating areas, then
the males return to the foraging areas and the females move to the nesting areas.^^’^^
After a reproductive period of several months, the females return to their foraging
areas and begin to prepare for the next reproductive period, a few to several years
in the future.2^’“^
3.2.1 D istrìbution
As a consequence of their food and habitat requirements, adult sea turtles are
unevenly distributed throughout the circumglobal tropical and subtropical
While most species are widely distributed, L. kem pi is restricted
to the Gulf of Mexico'^^ and N. depressus is endemic to the Australian continental
shelf.Although foraging occurs over a wide range, including open ocean areas,
the primary foraging areas for marine turtles are located within the North and South
20°C isotherms of average sea surface temperature and on the relatively shallow
Reproduction In Sea Turtles 53
FIGURE 3.1 Generalized life cycle of sea turtles; individual species vary in the duration of phases. Dermochelys coriacea and at least some populations
of Lepidochelys olivacea remain palagic foragers throughout their lives. (Redrawn from Lanyon, J., Limpus, C. J., and Marsh, H., in Biology ofSeagrasses,
Sc*
SH<-°
ol
<Xg>
5(31
< £ <D
P S ffif
gT§
ff/ ra®. . Ta)O

cr w 2-oE .E15O)
3 c0 I/
!0) Eo
®c
Eg If
Larkum, A. W. D., McComb, A. J., and Sheperd, S. A., Eds., Elsevier, New York, 1989, 610.)
3 CD
_ -O F
is I <
1§
2
D) c-2
e Jo
TCO00>>
CD CD
2 OJ
CD ,
CO c1<
CO *-
CO
V) ®
2 Li
a C
0) CD
CO E e2 f^r’^to
a«r "JD
E<D D)
2
^ h
|r g
tl o
gsss^ CLi/ «/j •“to
CD
»g 5< sE, UJCD -*CCDD
05
2
LULU 001
HZ
<0-2co -C
ôr y-BC xS
5- jo2 B= * ••coE 5?c >w
<0C w Iff 11 *
2c Bto 10' < 23 «Q)
z^ J
A
"2f V CD .b= O
e : w oC <2c °\ i
S is q. co -g
° 0) O
<O z ||
O CEuS /
r
> >0) <^ c0 UjLL
z O
W:
CD CD -Q
K Q
lu
O G Q .i 0-LU co
^ LLtfl
3o>
DC o
£/)£
continental shelf a r e a s . T h e leatherback turtle, D. coriacea, is an excep

tion; it ranges far into cold temperate waters^^^ and makes very deep dives after
food.^^ Local distribution is also uneven, with each species occupying different,
occasionally overlapping habitats: C/z. m ydas inhabits shallow, near shore, and reef
areas of abundant seagrass and algae;^^ Ca. caretta is regularly found on sandy reefs
and in shallow bays with abundant crabs and molluscs;'’^ L. kem pi occurs in shallow
bays with sandy and/or muddy bottoms with abundant crabs and molluscs;'’^ L.
olivácea and N. depressus are associated with relatively deeper soft-bottomed hab
itats inhabited by crabs and other c r u s t a c e a n s E. im hricata most frequently
occurs in hard-bottomed and reef habitats containing sponges.^’*
Regardless of their foraging range, the major nesting areas for most species are
located in the tropical and subtropical regions.^® The major nesting sites of D.
coriacea, Ch. m ydas, N. depressus, E. im hricata, L. kem pi, and L. olivácea are
located on islands and mainland beaches between the Tropic of Capricorn and the
Tropic of Cancer.^^ -^^’''^’^^’^^’^^ L. kem pi nesting is restricted to a narrow band of
beaches in the western Gulf of Mexico.^^ The exception to the tropical nesting pattern
is the loggerhead turtle.^^ The major portion of Ca. caretta nesting occurs in warm
temperate and subtropical areas, with the exception of Masirah Island, Oman;'’^’^*
only minor, scattered nesting occurs in tropical areas.
3.2.2 S ize and Maturity

Marine turtles do not begin to breed at a uniform or minimum size.^^’^^ Using
laparoscopy, Limpus^^ has followed the maturation (changes in oviduct and ovaries,
testis and epididymis) of Ca. caretta in foraging areas of eastern Australia. Based
on repeated observations of gonads^"^ and growth rates of free-ranging turtles,the
passage through puberty may require 10 years with individual turtles starting and
finishing at different sizes.The results obtained by using laparoscopy indicate that
hawksbill, green, and loggerhead turtles typically begin breeding at just less than
the average breeding size for the population;^"^’^^’^^ some individuals may be 10 or
more centimeters longer than the minimum breeding size and still be immature or
just starting into puberty,'^“’^^ presumably, other species show similar patterns of
maturation. This means that size is not a reliable indicator of maturity or breeding
status.^2’^^
3.2.3 R eproductive B ehavior

The major aspects of reproduction are very similar among the seven species of sea
turtles as a result of similarities in both moiphological and ecological constraints
(very similar body shape and the requirement of laying eggs on land)."^ The places
and conditions under which sea turtles nest are also very similar; they must have
access to beaches with deep, loose sand that are above high tide"^ and they nest
predominantly during the warmer months, with some places hosting year-round
nesting.^ Within this context, similarity in the behavioral component of nesting is
to be expected.
3.2.3.1 Philopatry
Carr^^ made the distinction between regional discrimination (philopatry) and fine-
scale homing to beach (site fidelity) within the region for sea turtles. It is now well
established that marine turtles migrate between their foraging areas and their nesting
areas with a high degree of accuracy.^’^^ Tag-return data have established that
although some turtles migrate greater than 2600 km, most travel less than 1000
The migrations of the green turtle across international borders are well
documented;^^’^^ in contrast, N. depressus does not migrate beyond the continental
shelf of Australia. All other species migrate to varying degrees.^ 56 ,57,61,63,68
Although they may not necessarily return to the beach of their birth, genetic studies
have demonstrated that breeding sea turtles return to the region of their birth.^^'^^
3.2.3.2 Nest Site Fidelity
Once it has returned to the region of its birth and selected a nesting beach, a turtle
will tend to renest in relatively close proximity (0 to 5 km) during subsequent nesting
attempts within that nesting season, although a small percentage will utilize more
distant nesting sites in the general or within a few hundred kilome-
ters.88.89 Limpus et al.^^ reported that N. depressus returned to the nesting beach with
a high degree of accuracy regardless of whether they were returning after an “unsuc
cessful nesting attempt,... a successful nesting,... [or] in a later breeding season.”^^
Bjomdal et al.^^ reported the distance between intraseasonal renesting attempts of
38 loggerheads ranged up to 290 km; other records of intraseasonal nesting move
ments suggest that loggerhead turtles are capable of moving long distances, but the
proportion of individuals doing so is low.^' Green turtles show a high degree of nest
site fidelity. Bose and Le Gall^^ reported that most renesting attempts were within
200 m of the previous attempt, with a range of about 600 m over the 3-month nesting
season.
The ecological consequences of this behavior require elucidation, but it is clear
that the turtles are spreading their reproductive effort through time which reduces
the impact of an unpredictable environment on hatchling production.'^^ By laying
several clutches of eggs at approximately 2-week intervals, the likelihood of all eggs
being lost is greatly reduced.
3.2.3.3 Periodicity
In general, female sea turtles do not reproduce every
possible exception of L. kempi?'^^ However, males of at least some species (e.g.,
Ca. caretta) may breed every year"^'^’'^^’ or every two years. There is no reason to
assume that males and females use the same time scale for reproduction. The duration
of the period between reproductive seasons is defined as the remigration interval.^*
The mean remigration interval reported for female sea turtles varies among the
species (Table 3.1); the range in remigration interval is from 1 to 9 or more years,
depending on the s p e c i e s . 88,91 actual remigration interval for a pop
ulation may be longer than is currently understood; the discrepancy may result from
TABLE 3.1
Reproductive Characteristics of Marine Turtles; Reproducing Females
Clutch
Dominant Carapace frequency Renesfmg Remigration
period of length (clutches/ interval (days) interval
Species nesting (cm) season) (mean, range) (years)
Dermochelys Night 148.7 (1.7) 11 6.17 (0.47) 4 9.5, 9-10 2.28 (0.14) 5
coriacea
Cheionia Night 99.1 (1.9) 22 2.93 (0.28) 9 12, 10-17 2.86 (0.23) 9
mydas
Natator Night and Day 90.7 (0.9) 6 2.84 (—) 1 16, 2-23 2.65 (—) 1
depressus
Lepidochelys Day 64.6 (—) Í 1.80 (—) 1 20-28* 1.50 ( - ) 1
kempi
Lepidochelys Night 66.0(1.1)8 2.21 (0.79) 2 17,22, 30,45* 1.70 (0.30)2
olivácea
Eretmochelys Night and Day 78.6 (1.7) 15 2.74 (0.22) 5 14.5, 11-28 2.90 (0.11) 5
imb ficaia
Caretta caretta Night 87.0 (1.6) 18 3.49 (0.20) 4 14, 13-17 2.59 (0.15) 5
* Interval between arribadas.

Note: Values are means of means of populations (standard deviation), number of populations included.
No account has been made of the number of samples from each population.
Based on data from Hirth,^ Dodd,^^ and Van Buskirk and Crowder.“^
tag loss,^^ incomplete survey coverage,or extremely long periods of quiescence.^^

Certainly, individuals in a population may have very long periods between repro
ductive episodes.
Supporting the observed reproductive periodicity is a poorly documented cycle
of energy accumulation, deposition, reorganization, and utilization. In general, while
in their foraging areas, sea turtles accumulate the energy reserves required to support
vitellogenesis over a variable period of several years depending, in part, on the
quality and quantity of food available.W hen a set of exogenous (e.g., photoperiod)
and endogenous (e.g., hormone levels and/or fat levels) factors interact, the conse
quence is r e p r o d u c t i o n . T h e timing of reproduction in marine turtles follows
behind periods of ample food during which the turtles accumulate fat reserves (one
to several years), complete vitellogenesis (10 to 12 months) and migrate to the
breeding and nesting areas (a few days to a few months). In the herbivorous green
turtle, the lag period varies; it may be as short as 2 years following an El Nino-
Southern Oscillation (ENSO) event^^^ or several years during anti-ENSO periods. It
is postulated that the ENSO weather events are associated with warming of sea
surface temperatures and increased growth of seagrass and algae, which has a
positive impact on the energy cycle and, hence, reproductive cycle of the green
tu rtl e. O the r species are carnivorous or omnivorous and their reproduction does
not appear to be as closely associated with fluctuations in major weather pattems.^^
Given the physiological demands of reproduction, iteroperous reproductive patterns

displayed by most species appear to be normal; the annual reproduction practiced
by a proportion of L. kempP^ requires detailed physiological study.
3.2.3.4 Courtship and Mating
There are only a few, mostly anecdotal, descriptions of courtship behavior among sea
turtles in the literature. Ehrhart^®"^ tabulated 15 studies which provide the foundation
for the current understanding of courtship and mating (Ch. E.
imbricatap^ Ca. carettaP^^'^^^ L. kempip^^ the most detailed descriptions concern
green turtles.
The range and significance of precopulatory behaviors cannot be determined
without more detailed underwater studies. The range of species-specific signals that
denote receptiveness (possibly: head bobbing, position in the water column, head
to head bumps, nuzzling, biting, movement of flippers) remain undefined. Precop
ulatory behaviors may have greater significance to females than to males; during
the mating period male green turtles attempt to court and mount objects that are
approximately the right size and general shape (e.g., skindivers, rough wooden
decoys, etc.).^^"^ Mounted pairs are most often seen at the surface of the water, but
copulation may occur anywhere in the water column."^^’^^’^^’^®^
Certainly, courtship and mating are not gentle processes.A female receives
bites (nips) to her flippers, neck and head that leave open sores requiring several
weeks to heal. She receives damage to her shell from the claws of the males as well.
Throughout the subsequent nesting season, the progressive healing of these injuries
provides an indication that mating took place only during the prenesting season.
Males are not excluded from receiving damage while several males are courting a
female. As one male positions himself to mount or becomes mounted on the female,
attendant males often bite the trailing edges of his flippers and/or the dorsal portion
of his tail. This biting may or may not unseat the mounted male at the time, but
certainly reduces his fitness to attempt mating with another female by causing severe
damage to the flippers and the tail. The more severe the injuries, the greater the
chance that the mounted male will be removed from the breeding pool, at least for
a period.
Three studies that were conducted on captive animals^®’ provide details on
aspects that are difficult to obtain in wild situations (e.g., duration of coupling and
resulting fertilization rates, period of receptivity, period of courtship activity).
Although a pair may remain coupled for hours, the actual transfer of sperm may not
require the entire period.
The male mounts the female quickly, usually at the surface with a lot of splash
ing, and hooks onto her carapace using the enlarged claws on his front flippers and
the large claws on his hind flippers to hold himself in place. The male curls his long
tail to bring their cloacae into contact. His penis is erected into her cloaca. The shape
of the penis with a bifurcation of the sperm duct at the tip allows for the transfer of
semen into each oviduct without passing through the environment of the female’s
cloaca. In captive situations the pair may remain coupled for 10 or more hours;
Wood and Wood^^ reported an increase in fertility that correlated with the duration
of the coupling period.
In the wild, after coming into reproductive readiness, females and males
migrate^’^^"^’^'^ from their foraging areas to copulate at courtship/mating areas^^ in
the region of the nesting area. Males show fidelity to particular courtship areas in
successive migrations.Females are in reproductive readiness for about 7 to 10 d;
males appear to be sexually active for about one month.^^’^^'^'’ Genetic studies have
demonstrated that males mate with several females and females mate with several
m a l e s . A t the end of the mating period the males leave, presumably returning
to their foraging areas.^^ Female green turtles^^ (and presumably females of most
other species) disperse from the mating areas to nesting sites in the region to which
they show fidelity.
Some confusion concerning courtship and malting crept into the literature
because some of the early observations suggested that mating for fertilization of
subsequent clutches occurred immediately following oviposition and that females
retained sperm for several years between reproductive Even
tually these ideas were challenged by Frazier,"^' who raised the following points: (1)
lack of evidence that females without claw marks on their carapace (i.e., not mated)
laid fertile eggs; (2) the nonadaptive requirement that virgin females migrate to mate,
but not lay eggs; (3) the nonadaptive behavior required of males to migrate before
mating; and (4) the mechanism to ensure sperm viability following long-term stor
age."^^ It is now generally accepted that mating occurs in the month or two just
preceding the first ovipositional cycle of the season.^^ Sperm have been found at the
junction of the magnum and the aglandular zones in the upper reaches of the
oviduct; presumably, the sperm in this area fertilize the ova as they enter the
magnum, which secrets albumen. Females do not store sperm between reproductive
seasons.
Absolutely nothing is known about the impact of endogenous factors, such as
ageing, on reproduction. For example, older males may have a longer period of
spermatogenesis than younger males. Similarly, older females may nest earlier in
the season and may lay more clutches. The testing of these ideas requires a known
age population. As individuals in long-term studies pass through puberty, join the
breeding portion of the population, and age, it may be possible to address these
issues.
3.2.3.5 Nesting Activity
Although there have been several attempts to define the qualities of a nesting beach,
it is not obvious why a population of sea turtles or an individual sea turtle uses a
particular beach.Mortimer^^^ identified general requirements for nesting beaches:
(1) the beach must be accessible from the sea, (2) the beach must be high enough
to prevent inundation of the eggs by tides or the underlying water table, (3) the
substrate must facilitate gas difussion, and (4) the substrate must be moist and fine
enough to prevent collapse of the egg chamber during construction. Mortimer^^®
examined physical and chemical characteristics of beach sand from 50 nesting areas
from around the world and found that variability was high. She concluded that
‘factors other than physiognomy of sand on nesting beaches may be as important,

or more important, in nest site s e l e c t i o n T h e result of beach selection and nest
site selection is that the eggs incubate in a low-salinity, high-humidity, well-
ventilated substrate which is not inundated during developmentd^’^^^’*^^ Given the
level of investigation concerning the nesting process, it seems odd that no one has
been able to define the process by which the turtle (any species) selects its nesting
beach or the site for the nest on the beach.
The gait used by sea turtles to move on land fits one of two patterns: simultaneous
or alternating; as a result, the tracks left in the sand are either symmetrical or
asymmetrical.A turtle using the simultaneous gait moves all four fiippers forward
at the same time, then using the flippers, moves its body forward; at the end of the
forward stroke the body rests on the substrate for at least a moment while all four
flippers are moved forward again to repeat the process. The resulting track is
symmetrical with the marks made by the front flippers forming diagonal grooves
on the opposite sides of the track and the tracks of the hind flippers forming paired
mounds on either side of the center of the track. The center of the track is the mark
left by the plastron in which the tail may occasionally drag. The simultaneous gait
is used by Chelonia, Natator, and D erm ochelys. A turtle using the alternating gait
moves one of its front flippers forward at the same time as it moves the hind flipper
on the opposite side forward; only two flippers move at a time. The resulting track
is asymmetrical with the marks made by the front flippers obviously offset. The
alternating gait is used by L epidochelys, E retm ochelys, and Caretta. The style of
the gait and the width of the track provide a clear indication of the species that has
attempted to nest (track width: Chelonia -100 cm, N atator -90 cm, D erm ochelys
-150+cm, L epidochelys -80 cm, E retm ochelys -75+cm, Caretta -90+cm).^2^
In order to avoid lethal temperatures during the day, most sea turtles nest at
night, although a small number of every species occasionally nest during daylight;
the exceptions to the rule are (1) L. kem pi, which nest predominantly during the
day,^^ (2) N. depressus, which regularly nest during the day;^*"^ and (3) L. olivácea,
when nesting in a r r i b a d a s . Presumably, these turtles are able to nest during
the day when temperatures and radiation from the sun are high because (1) they
avoid nesting during the hottest part of the year;^^^ (2) they avoid nesting during the
hottest part of the day by nesting in the late afternoon; ^ (3) they nest rather quickly;^
(4) they are lighter in color than other turtles and, presumably, absorb radiation at
a lower rate; and (5) they are able to dissipate excess heat to the environment via
convection, evaporation, and conduction.By timing nesting to avoid the hottest
time of the year, the turtles may actually place their clutches into beach temperatures
that support embryogensis, rather than being too hot.^^^ The thermal biology of
marine turtles in terrestrial (and aquatic) habitats requires much more investigation
before its importance to their survival can be understood.
The amount of time required to complete the nesting process varies among the
species:^ L. olivácea (-1 hr), L. kem pi (-1 hr), E. im bricata (-1 to 1.5 hr), and N,
depressus (-1 to 1.5 hr), Ca. caretta (-1 to 2 hr), D, coriácea (-1.5 hr), and Ch.
m ydas (-2 to 3 hr). In extreme cases, green turtles may take 7 or more hours to nest
on a dry, sandy beach; whereas hawksbill turtles may be out of the water for only
45 min to lay a clutch of eggs in a moist, vegetated dune area. Some of the variation
results from the habitat in which the turtles typically nest and some of the variation
is derived from the size of the turtle (among the cheloniid turtles, smaller species
nest faster).
3.2.3.6 Nesting Process
The general nesting process (beaching, digging, laying, filling-in, returning) is

essentially the same in all species of sea t u r t l e s , a l t h o u g h some differences
do occur (Table 3.2). The process has been subdivided in to 11, 9, and 7
Regardless of the defined steps, the general pattern contains (a)
emerging from the surf, (b) ascending the beach, (c) excavating the body pit, (d)
digging the egg chamber, (e) oviposition, (f) filling in the egg chamber, (g) filling
the body pit, and (h) returning to the sea. The nesting behavior of sea turtles is
constrained by their anatomy and by the environment in which the nest is con
structed."^ Most of the differences in nesting behavior may be attributed to the size
of the nesting turtle, which influences its speed and mobility on the beach and the
depth of the nest.
TABLE 3.2
Stages of Sea Turtle Nesting Behavior
Activity Species Differences
Stranding, Testing of sand, and Similar in all species; D. coriácea may be more direct than other
emergence from wave wash species; L. olivácea and L. kempi in arribadas
Ascent of the beach Similar in all species; usually fairly direct uphill unless debris causes
deviation
Selecting nest site Similar in all species; may be related to change in temperature
Clearing nest site All species actively clear area, possibly in response to litter and
vegetation at site
Excavating body pit Similar in all species; Ch. mydas digs proportionally deep body pit
Excavating nest hole Similar in all species; achieved by alternating use of hind flippers
Oviposition Similar in all species; hind flipper positioned differently
Filling and packing nest E. imbricata ladles sand over eggs; others scrape sand into hole; N.
chamber depressus, L. olivácea, m á L . kempi rock body forcefully to compact
sand; D. coriácea pivots with weight on rear body; other species use
hind legs to “knead” sand compact
Filling body pit Similar in all species; sand thrown backwards by front flippers
Course back to sea Similar in all species; move down slope
Reentering wave wash Similar in all species; typically a pause at edge of waves
Based on CaiT and Ogren,^^^ H e n d ric k s o n a n d Bustard, Greenham, and Limpus.-
Pritchard and Trebbau provide lucid descriptions of the nesting process of D.

coriácea, Ch. mydas, Ca. caretta, E. imbricata, and L. olivácea. The only rigorously
descriptive study of marine turtle nesting behavior is by Hailman and Elowson;^^^
focusing on the loggerhead, they determined that the ten basic phases of the nesting
process involve about 50 separate action patterns. These action patterns are thought
to be modified locomotor patterns derived from the gait. This level of analysis should
be applied to the other species to define similarities and differences among them.
Although the physical characteristics (slope, aspect, size of particles) of the
beach influence the nesting behavior, as well as the microenvironment in which the
eggs develop, in general the nesting sites used by sea turtles share several charac
teristics: exposed marine beaches, relatively loose sand, free of buried debris, and
above high tide level.
At the moment when the turtle is first supported by the sand it may be washed
over by several waves before it starts up the beach; while the turtle is on the wet,
hard, sloping part of the beach she continues mostly uphill. As the turtle proceeds,
it pauses after a few meters to rest, breathe, and, possibly, observe its surroundings.^^
During this period solitary and low-density nesting marine turtles are easily deterred
from continuing up the beach by activity on the beach; during arribada nesting by
L. olivácea and other high-density nesting situations (e.g., several thousand green
turtles attempting to nest at Raine Island, Australia; personal observation) turtles are
less easily disturbed. Debris and other obstacles may cause the female to change
direction or even to abandon the effort; she may return to the water for no obvious
reason. The tracks of aborted nesting attempts are often and erroneously called “false
crawls”;’“"^ the use of this term may be acceptable in the vernacular, but it is
misleading and implies some intent on the part of the turtle. The crawl is real; the
nesting attempt was aborted for some reason, albeit unknown. The occurrence of
aborted nesting attempts varies among species (low in N. depressus,^^ high in Ca.
caretta^^ and between nesting beaches (e.g., high in southeastern U.S., low in
Queensland, Australia).W hatever the reason(s) for aborting the nesting attempt,
the turtle usually returns to nest the same night or the following night (e.g., Ca.
caretta, 1.08 d;"^^ N. depressus, 1.17 d)^^ and most return to the same beach (e.g.,
Ca. caretta, N. depressus, 95%).^^
Once the turtle has crossed the temperature boundary between the cooler wet
sand of the beach slope and entered the warmer, drier, softer, flatter portion of the
b e a c h s h e may change direction several times before selecting a site at which to
construct her nest.^^^ All sea turtles prepare the nest site before digging the egg
chamber to varying degrees. The surface debris is removed by either simultaneous
or alternating sweeps of the front flippers. As the area in front of the turtle is cleared
the turtle “swims” forward onto the cleared substrate. The hind flippers may be used
to clear loose, dry substrate from immediately behind the turtle. The depth of the
body pit is related to the dryness of the surface sand and to the size of the turtle.
Very dry conditions require that a turtle digs deeper to locate the level of firmer
sand that will hold the shape of the egg chamber. Larger species (e.g., green turtles)
excavate relatively deep body pits; whereas smaller species (e.g., hawksbill turtles)
dig shallower body pits. After the turtle has moved forward into the “body pit”, the
front flippers cease action and are usually positioned near the carapace.
Sea turtles use their hind flippers to excavate a flask-shaped chamber that has a
narrow neck and a wider bottom"^^ unless the dryness and particle size of the substrate
causes the shape to a l t e r . A t first, the nesting sea turtle scrapes and flicks sand
away from the area where the chamber will be; then it begins to scoop sand from
under its posterior using the ventral portion of each hind flipper altematively.^^’^^^
The cup full of sand is placed to the side of the chamber area and the flipper is
quickly placed on top of it.^^ The other flipper is then used to remove a cup full of
sand; when this flipper is placed to the side of the chamber, the first flipper throws
the loose sand that has fallen on top of it forward and reaches into the nest chamber
for more sand.^^ This process ensures that during the filling-in process, the sand
placed on top of the eggs is moist sand that originated from the chamber as it was dug.
The actual shape of the chamber is constrained primarily by two anatomical
factors: the length of the hind limb and the size of the flipper (hand) portion. Because
hawksbill turtles have relatively short hind limbs and relatively small hind flipper
areas, they dig a comparatively shallow nest chamber that has a more rounded bottom
portion.^^ In contrast, green turtles have relatively long hind limbs with relatively
large hind flippers; as a result, the egg chamber dug by a green turtle is deeper and
extends more forward than that dug by a hawksbill turtle. Regardless of the size of
the hind flippers, by raising on its front flippers, a turtle is able to extend its reach
down and forward in the nest chamber.^^ The shape of the chamber is also effected
by the moisture content and particle size of the substrate as well as the presence of
buried debris. The nest is not inspected visually during the entire process; the
stereotyped alternating digging stops when the turtle cannot remove any more sand
while attempting to scrape the walls and bottom of the chamber. The digging
behavior is stereotyped so strongly that a turtle missing one hind flipper will move
the stump in sequence with the flipper actually removing sand from the chamber.
This one-legged style concludes with a lopsided chamber, the walls of which cannot
be enlarged by the one flipper, but that is not large enough to hold the entire clutch
of eggs. In these cases the eggs fill the neck of the chamber and may spill out the
top where they are mixed with sand during the filling process.
During oviposition the hind flippers of Dermochelys, Chelonia, and Natator are
positioned to cover the nest chamber; one flipper usually hangs inside the chamber.
In this position, one flipper may help keep sand from falling into the chamber while
the other may break the fall of the eggs into the chamber. In Carena, Eretmochelys,
and Lepidochelys both hind flippers are extended outwards on the sand behind the
turtle so that the opening to the nest is not covered by the flippers, but is covered
by the posterior margin of the carapace;^^ during oviposition the outer edges of the
flippers curl as eggs are released.
The eggs are laid singly or in groups of two or three, sometimes four. As the
cloaca contracts, the tip of the tail points downward and slightly forward, then relaxes
backward as the eggs drop. Results of a dissection of a green turtle that died during
oviposition at Raine Island, Australia indicate that eggs were laid from both oviducts
together, rather than from one and then the other, because 22 and 27 eggs remained
in the oviducts while 47 eggs were in the nest chamber at the time of death (Miller,
personal observation).
During egg laying, all species of sea turtles are relatively tolerant of a modest
level of external disturbance. The level of indifference to disturbance varies among
individuals; some will be completely disturbed by being tagged while others will
bite the sand in front of them; still others will show no response at all. Part of this
variation is a function of the number of eggs already laid. As a general rule, the
level of tolerance increases as the turtle lays more eggs. The level of tolerance to
disturbance remains high during the filling of the egg chamber and the covering
phases of the nesting cycle.
Once oviposition is complete the nest chamber is filled in, usually by scraping
moist sand from near the opening into the hole with the hind flippers, until the sand
above the chamber is higher than the floor of the body pit.^^^ As the neck of the
chamber fills, the sand is compacted using the hand and leg portions of the hind
limb. In the smaller species the compaction may appear quite vigorous.
After the hole is filled and compacted, the front flippers begin to throw sand
backwards over the nest site; during the process the turtle moves forward and
continues to throw sand backwards. At the end of the process the nest site may be
one or more meters behind the turtle and the remnant of the body pit is not near the
eggs.129 behavior was once interpreted to be camouflaging the nest,^'" presumably
from predators; while this interpretation may be in part true in some cases, most
predators can readily locate a nest within a few days of oviposition or near emer
gence. The primary function of this behavior is to reestablish the environment of
the beach around the nest; the sand thrown on top of the nest provides insulation in
terms of both temperature and moisture for the eggs.
3.2.4 R eproductive Potential

All species of sea turtles lay several clutches of eggs during a nesting season^’^ (Table
3.1) . Egg laying may occur year-round, with seasonal peak and trough periods, at
nesting sites closer to the equator, or may be distinctly seasonal, with at least several
months during which no turtles nest, at more temperate The
number of clutches laid during a reproductive season varies within and between
species.“ Determining the number of times a turtle nests during a reproductive season
is important, particularly if such data are averaged and used in calculations to
estimate the number of female turtles in the population. It is also important to
know the number of clutches laid per season in order to estimate the annual fecundity.
Unfortunately, the number of clutches laid is often derived from incomplete coverage
of the nesting season or nesting area; the apparent number of clutches laid may be
reduced by the loss of individuals from the nesting group.^’^^
Although all species of sea turtles average more than one clutch of eggs in a
reproductive season, the intervals between successive nesting attempts varies (Table
3.1) . The usual internesting period for D. coriacea is 9 or 10 for Ca. caretta
the internesting period varies from 12 to 16 The internesting period for Ch.
m ydas ranges from 10 to 14 N. depressus renests at intervals between
13 and 18 d.^^’ E. im hricata exhibits a renesting interval of between 13 and 15 d.^^
The two species of L epidochelys exhibit the longest internesting periods, 20 to 28
d for k e m p p ^ and 17 to 30 d for o liv a c e a P Intemesting periods of less than 6 d
are physiologically impossible for sea turtles (see below); when extremely short
intemesting intervals occur they represent the second attempt at laying a clutch
follow ing a divSturbance, and not a new clutch.^^ Because sea turtles may be deterred
from nesting by a variety o f subtle events, Limpus'^^ defined an internesting interval
as the interval between returning to the sea after having laid a clutch o f eggs and
when the turtle next emerges to lay, whether or not that attempt is successful.
3.2.4.1 Shape and Size of Eggs
All marine turtles lay white, spherical cleidoic eggs with flexible calcareous shells.^^
The size (diameter, shape) of eggs laid in each clutch varies slightly between clutches
laid by one female as well as considerably within and between species.^ D. coriácea
lays the largest eggs of all marine turtles (5.3 cm, 90 g), but N. depressus produces
eggs that are nearly the same size (5.1 cm, 80 g). Ch. mydas lays medium large eggs
(4.5 cm, 48 g), followed by Ca. caretta eggs (4.0 cm, 36 g); E. imbricata lays small
eggs (3.8 cm, 28 g), L. olivácea and L. kempi lay eggs of similar size (3.9 cm, 35
g; 3.8 cm, 30 g, respectively) (Table 3.2). At the level of species, large sea turtles
lay large eggs and small ones produce small eggs,^
In an examination of correlations among several life history traits at the species
level. Van Buskirk and Crowder^ used the means obtained for populations as reported
in the literature; they acknowledged that the data were uneven and contained errors.
Their analysis supports several generalities concerning sea turtle biology. They
reported that body size is correlated with “several reproductive traits including egg
size and overall reproductive effort”.'’ A general “trade-off between clutch size and
egg size was confirmed among the species” after adjusting for body size.^ However,
in their discussion of the unique pattern of a small clutch (mean = 52.8 g) and large
egg size (51.5 mm) found in N. depressus, they did not consider that Natator evolved
in the context of a wide range of carnivorous marsupials and varanid lizards that
occupied predatory niches in Australia similar to those occupied by placental car
nivores elsewhere.^^’*'*^ As would be expected, some traits (e.g., egg survivorship,
remigration interval) were not related to the body size of the female.^’Although their
analysis provides a number of insights into life history variation among the species
of marine turtles, until within and between regional variation and interpopulation
variation in life history traits for each species are better documented, including both
morphometric and genetic data, their analysis at the species level may be premature.
One factor that has not been adequately handled in assessment of fecundity in
sea turtles is the duration of the reproductive life of the species. Fecundity should
be calculated as the product of the clutch count, the number of clutches per breeding
season, and the number of breeding seasons in the life of the turtlehowever, the
number of breeding seasons in the life of a sea turtle is not known for any species.
Until reproductive life expectancy is determined, detailed comparisons of fecundity
among species is not Justified.^^
3.2.4.2 Number of Eggs and Clutches
The number and size of eggs, and the number of clutches laid (Table 3.3) represent
the result of an adaptive compromise for survival.^^ The number of eggs in a clutch
should be the result of the interaction among optimum egg size for hatchling/post-
TABLE 3.3
Reproductive Characteristics of Marine Turtles: Nesting Period and Sizes of
Eggs and Hatchlings
Hatchling
Clutch count Egg weight Egg diameter Egg volume weight
Species {# of eggs) (g) (mm) (cc) (g)
Derm ochelys 81.5 (3.6) 12 75.9 (4.2) 4 53.4 (0.5) 9 79.7 (2.4) 9 44.4 (4.16) 5
coriacea
Cheionia 112.8 (3.7) 24 46.1 (1.6) 10 44.9 (0.7) 17 45.8 (1.2) 17 24.6 (0.91) 11
mydas
N atator 52.8 (0.9) 6 51.4 (0.4) 3 51.5 (0.3) 6 70.8 (1.1) 6 39.3 (2.42)3
depressus
Lepidochelys 110.0 (—) 1 30.0 (—) 1 38.9 (—) 1 30.8 (—) 1 17.3 (—) 1
kempi
Lepidochelys 109.9 (1.8) 11 35.7 (—) 1 39.3 (0.4) 6 31.8 (1.1) 6 17 ( - ) 1
olivacea
Eretmochelys 130.0 (6.8) 17 26.6 (0.9) 5 37.8 (0.5) 1 28.7 (1.3) 11 14.8 (0.61)5
imhricata
Caretta 112.4 (2.2) 19 32.7 (2.8) 7 40.9 (0.4) 14 36.2 (1.1) 14 19.9 (0.68) 7
caretta
Values are means of means of populations (standard deviation), number of populations included.
Based on data from Hirth,^ M iller,an d Van Buskirk and Crowder.^
hatchling survival and dispersal, the selective advantage of the number of hatchlings
produced, and the physical and ecological limitations of the female. The number of
clutches laid during a reproductive period is the result of selection for the advantage
of separating groups of eggs in time and space in the context of the capacity of the
female to produce the eggs within the risks such as exposure to predation operating
within other limitations such as the length of the nesting season and the energy
required for reproduction.^^ Moll^^^ proposed two general patterns of reproduction
in turtles. The sea turtles fit pattern I: ‘Targe clutches of relatively small eggs; multiple
clutches produced during a ... nesting season; communal nesting in ... ancestral
nesting areas; careful construction of covered nest”.^° Within this paradigm, selection
favors fecundity; however, when approaching the extreme case, density dependent
factors such as predation (including by man) may intensify to cause local and
sometimes large-scale extinctions. The density of turtles attempting to nest on a
beach may have a direct impact on the number of eggs destroyed by subsequent
nesting attempts. Bustard and Tognetti^"^® reported on computer simulations of the
impact of various densities of nesting green turtles on egg destruction; they reported
that as density increases, so does the likelihood of clutches being destroyed. On the
nesting beach, density should be interpreted as both the number of turtles attempting
to nest and the number of clutches already deposited into the beach. However, the
importance of egg destruction should be reconsidered in the context of other factors
(e.g., bacterial infection, losses caused by inundation and erosion, etc.) that impact
on the production of an adequate level of hatchlings to sustain the population through

time.
3.3 EMBRYOS
In order to provide for the next generation, adult sea turtles must lay fertile eggs at
the right stage of development into an environment that facilitates development and
maximises hatchling emergence.^
3.3.1 Fertility and I nfertility

The fertility, as shown by hatching success and opening unhatched eggs, of marine
turtle eggs is high, typically greater than The actual fertility of the
eggs probably exceeds 95%, but the difficulty in separating intraoviducal death
and/or early embryonic death (before the formation of blood isles) from infertility
prevents a full assessment.^^
3.3.2 D evelopment
When a sea turtle arrives at the nesting beach she is carrying a full complement of
follicles to supply yolks for all the eggs she will lay during the season; the diameter
of the follicles will increase only slightly as the result of the absorption of water in
the period before ovulation.
Within the body of the female, a short period after the follicle has been ovulated
from the ovary, it enters the infundibulum of the oviduct; the follicle then passes
through the aglandular region into the magnum, where it is fertilized by sperm.
Because the sperm from all successful males are stored and mixed in the upper
portion of the oviduct, the follicles ovulated to become the yolks of the eggs in a
clutch are fertilized by sperm from several different males.Development begins
immediately following fertilization. The fertilized embryonic area on its yolk con
tinues into the oviduct where albumen is secreted from special cells in the lining of
the anterior glandular region of the oviduct. Once the yolk has been coated with
albumen, the inner shell membrane begins to foim, secreted from special cells in
the shell-forming segment of the oviduct.^^’^*'^’^"^^Following the formation of the shell
membrane, aragonite crystals begin to form the outer portion of the s h e l l . T h e
shell is not fully formed until at least the seventh day following ovulation.^'^
Although cleavage begins within hours of fertilization, development does not
advance beyond middle gastrulation while the egg is within the oviduct.^^ Eggs are
laid at middle gastrulation, before the chordomesodermal canal breaks through into
the subgerminai cavity. However, once oviposited, development resumes in a few
hours (4 to 8, depending on temperature); movement of the eggs after this time
results in death.
The first external sign that development is progressing is the development of a
white spot on the uppermost part of the egg;^^’^"^^ the embryo is situated just beneath
the shell. As the white spot becomes progressively bigger to eventually encompass
the entire shell, the embryonic membranes (vitelline, amnion, allantois, chorion)
develop.^^^ The embryological development of marine turtle embryos of all species

is similar in morphological detail and sequence up to stage 22, after which generic
and species-specific characteristics become increasingly evident.^^
When considering eggs it should be remembered that the energy and chemical
components contained in the follicles were derived from food eaten in the foraging
areas; the results of any chemical analysis of the eggs^"^^ reflects the situation in the
foraging area (not the nesting area). Because sea turtles using a nesting beach have
migrated from several foraging areas, unless the foraging area is known, analysis of
eggs (or other tissues) cannot be tied to any specific place. Because chemical analysis
tends to be rather expensive, the number of samples tends to be small, further
reducing the usefulness of the data except as an indicator. However, if the foraging
area is known, the data become extremely useful in interpreting situations in the
foraging area and the potential impact on reproductive success.
3.3.3 N est Environment

The conditions within the nest environment must be within the limits of embryonic
tolerance in three areas: gas exchange, moisture, and temperature.^^’^^^’^22,123,150
limits of embryonic tolerance are only partially defined for gas exchange. The
impact of different levels of moisture on the embryos is more clear. Mortality is
higher in drier conditions, suggesting that eggs of marine turtles are very sensitive
to desiccation;înundation for extended periods of time (hours) also increases
mortality. By comparison, the impact of temperature is well defined. Eggs that
incubate at temperatures lower than 23°C for the last third of incubation seldom
hatch; eggs held at temperatures greater than 33 °C for extended periods do not
hatch. Temperature has an inverse impact on the duration of incubation;^^ at 23 to
25°C, Ch. mydas eggs incubated for 94 d^^® and at 32°C eggs incubated for 49 d.^^^
In the natal beach the eggs incubate at temperatures between 24 and 33°C, although
an occasional short period may occur outside this Within the range
of 26 to 32°C, a change of EC adds or subtracts about 5 d to the incubation period.^^
The sexual differentiation of sea turtle embryos is determined by temperature
{Ca. Ch. L. olivacea;^^^ D. coriacea;^^^ E. imbricata.^^^
Although the pivotal temperature varies among the species and between populations
within a species,the generality that cooler temperatures produce males and warmer
temperatures produce females holds for all species. However, the mean daily tem
perature in natural nests is actually a poor predictor of the hatchling sex ratio when
nest temperatures fluctuate.The sex of Ca. caretta hatchlings “depends on the
daily proportion of embryonic development that occurs above the threshold temper
ature’ presumably other species also respond to the direct impact of temperature
rather than to the overall developmental rate to detemiine the sex of the hatchlings.
In nomial conditions the eggs absorb water soon after being oviposited and
become turgid. Usually, eggs increase in weight slightly during the term of incuba
tion; but it is more important that the eggs do not lose more than 40% of their initial
mass if they are to survive to hatching. Salt and organic material as well as the type
and microshape of the particles modify the impact of the hydric conditions on the
change in weight of eggs and hatchlings. Work on other species of turtlesuggests
that the hydric conditions will have a significant impact on the hatching success as
well as the mass and general vigor of the hatchlings.
The developing embryo usually experiences adequate gas exchange,‘22,162
although the gas diffusion may be affected by the particle size and water content of
the sand.‘22^162-166 inadequate gas exchange slows growth and increases embryonic
mortality.
3.4 HATCHLINGS
The eggs of marine turtles hatch after 6 to 13 weeks of incubation, depending on
the temperature.29 Hatchling sea turtles usually weigh slightly less than 50% (41 to
62%2) of their ovipositional egg weight (Table 3.2); large eggs produce large
hatchlings.^
3.4.1 HATCHrNG Process

Within the nest chamber, individual embryos use their caruncle to cut through the
amniotic and chorio-allantoic membranes as the hatching process begins. Once the
eggshell has been ruptured and the embryo begins to tear the shell, the fluid (allantoic,
albumen, amnionic) spills out and drains away. Because the egg has been turgid
throughout incubation and has occupied a volume, the loss of fluid creates a space
which fills with air. As a result of its struggles, the curled hatchling wiggles out of
the eggshell and begins to flatten which internalizes the remaining yolk. Within the
space created by one or a few eggs, hatchlings have diffieulty in flattening and
subsequently digging toward the surface. However, when a clutch hatches the total
volume created assists the hatchlings by allowing them to crawl past each other. The
discarded eggshells are shuffled downward and the space within the chamber,
although not physically larger, is more conducive to the straightening of the
hatchlings.
The mutual stimulation by the hatchlings was described as “social facilitation”
by Carr and Hirth. ‘^’2 Social facilitation describes the upward digging activity of the
hatchlings that is stimulated by the activity of other hatchlings, usually from the
bottom of the chamber. As the hatchlings dig upwards, the sand above them is
scratched away and shifts down through the mass of wiggling hatchlings to the
bottom of the chamber. In this way the hatchlings dig their chamber to the surface
of the sand. When the surface sand is damp, the remains of the chamber can be seen
as a cavity on the beach.
3.4.2 H atchling Emergence

Hatchling sea turtles usually emerge onto the surface of the beach during the early
evening. The nocturnal emergence of hatchlings is controlled by the gradient in the
sand temperature experienced as the hatchlings dig toward the s u r f a c e . ^ s
hatchlings dig toward the surface during the day, they reach warmer sand; while the
sand above them is warmer than the sand they are in, digging is inhibited until the
temperature gradient r e v e r s e s . At sunset the temperature of the sand drops rapidly
because the energy input from the sun is removed and the gradient of latent heat
contained in the sand reverses. After an hour or more the temperature of the sand
at the level of the hatchlings has dropped and the hatchlings begin to dig toward the
surface, arriving after dark.
The gas exchange at the level of the hatchlings may also act as a controlling
factor on digging activity. The digging activity of the massed hatchlings may require
more oxygen (or produce more carbon dioxide) than can readily diffuse through the
sand. In this case the hatchlings may become quiescent until their anaerobic respir
ation debt is repaid. The behavior of hatchlings near the surface may not be regulated
by oxygen as much as by temperature. Clutches of hatchlings emerge through the
final few centimeters of sand on cool, cloudy, and/or rainy days.^^®
Hatchlings face two problems if they emerge during the day: lethal temperatures
and predators. The temperature of the sand during the day may exceed 45°C at some
nesting beaches; certainly if the hatchlings have to cross a wide beach under direct
sun, many will die before reaching the water or receive burns from the heat of the
sand being crossed and die from the burns later or be unable to swim properly and
be consumed before they reach deep water. Further, the hatchlings are much more
visible to a wide range of predators during the day. Day or night and at different
tide levels, hatchlings face a large number of aquatic predators that reduce the
number of hatchlings reaching the relative safety of offshore water.
3.4.3 H atchïng S uccess

Clutches of sea turtle eggs typically have high hatching success (80% or more)
unless external factors (e.g., predation, environmental change, microbial infection,
etc.) intervene.^ After 55 or more days of incubation, it is difficult to determine if
an egg contained a fertilized embryo that died early (possibly within the oviduct or
during the first few days of incubation) or not. Fortunately, the contents of unhatched
eggs can often be categorized by the presence of blood (indicating early embryonic
death) or by a recognizable embryo at some stage of development.^'^
3.5 CONSERVATION
The general policy for the conservation of marine turtles was outlined in Bjorndal;^
variations and refinements of these ideas have been presented by Limpus.^^^ Recently
the lUCN Marine Turtle Specialist Group produced a global strategy that addresses
research, management, and conservation issues. These policy statements provide a
framework for global, regional, and local conservation efforts. BjorndaF also out
lined what was (and is) being done so that others may copy successful conservation
programs; together, these documents outline the data required to better define the
biological requirements of the turtles.
3.5.1 N esting B eaches

The focus of conservation protection of reproducing turtles divides into two general
categories: protection of the animals and protection of their habitat. It is rather
obvious that the important nesting areas, nesting activities, as well as the eggs and
hatchlings should be protected.^ However, achieving this may take different forms
depending on the assessment of the local problem(s); in one area minimizing
lighting^’ near an important nesting beach to reduce disorientation will solve a
specific problem, while in another area, limitation of human access to an important
nesting area may alleviate several problems.^’
Management of the nesting site must include management of embryos via their
eggs. Probably the best conservation action on the nesting beach is to leave the eggs
in situ and to reduce the impact of predators and poachers. However, in certain
circumstances, the shifting of eggs to protect them from inundation, predators,
poachers, etc., is more effective. If practical, eggs should be shifted to nearby sites
out of harm’s way in preference to shifting them to hatcheries. Whenever it is
considered necessary to shift eggs to safer areas of the beach or to hatcheries, they
should be moved and reburied within 4 h of being laid or allowed to remain in situ
for at least 25 d to reduce the impact of movement-induced m o r t a l i t y . T h e new
location for the eggs must provide adequate moisture, temperature, and gas exchange
to support the developing embryos. It is almost self-evident that hatchlings should
be released as soon as possible after emerging from the sand.
The second area of focus is habitat protection. This may be achieved through
the creation of special, limited-access areas such as national parks or other protected
areas. Protection of the nesting beaches is an obvious priority; foraging areas, too,
should receive high priority because it is from the foraging areas that the energy for
yolk deposition (and hence embryonic, hatchling, and post-hatchling survival) and
migration is obtained. However, other habitats used by the turtles also require
protection, including mating areas, internesting areas, migration routes, and hiber-
nacula.^ Practical measures for the conservation of sea turtles are discussed by
Pritchard et al.;^^"^ whatever techniques are employed, they must be based on a
knowledge of the needs of the turtles and an assessment of the problem. Further,
whatever remedial action is applied must be documented so that the success or failure
of the effort can be evaluated. Because sea turtles are long-lived animals, the results
of conservation management actions will not necessarily be obvious, particularly in
the short term.
3.5.2 G reenhouse Effect

One of the most challenging issues facing the conservation management of marine
turtles will be the long-term effect of the rise of the sea level as part of the greenhouse
effect. Given the reproductive characteristics of marine turtles (particularly natal
beach homing, nest site fidelity, temperature-dependant sex determination, decades
to sexual maturity, long breeding life), Mrosovsky et al.,^^^' Davenport,^^^ and
Limpus^^^ have speculated on the impact of global wanning on sea turtles. Several
points arise from their separate evaluations. The predicted increased atmospheric
temperatures will result in increased sand temperatures which in turn will affect the
sex ratio of hatchlings, with more females being produced. How will the population
respond? Will the pivotal temperature, which varies between populations, shift to
accommodate the shift in sex ratio? Certainly, populations do have different pivotal
temperatures with higher values recorded at nesting sites closer to the equator. Along
with sea level rise there will be increased stomiiness (frequency and intensity) which
will reduce the size of islands, particularly low sandy ones, and alter the shape of
mainland beaches.Questions arise concerning the behavioral response of the
turtles: will they shift nesting sites? Genetic studies suggest that they will not en
mass, but that some will. Will the turtles simply continue to nest until the recruitment
fails to support the population? With increased storminess, will the associated cool
ing pulses of temperature at egg depth compensate for the general temperature rise
and thereby help to maintain the required sex ratio? It is predicted that ocean currents
may alter; if this happens to a current that is supporting the dispersal of hatchlings
from a nesting site, will the hatchling survivorship be increased or decreased or
unchanged? Given the known impact of the ENSO on the number of green turtles
nesting*^^ and that the impact of the greenhouse changes (more toward the anti-
ENSO), will green turtles nest less frequently? In the face of complex responses to
long-term changes, more data are essential before management actions are imple
mented. The short-term impact of low-temperature pulses associated with humcanes
and cyclones needs to be examined with reference to the experimental work on sex
determination.^^® The behavioral response of nesting turtles to alterations in the beach
and off-shore environment resulting from tropical storms should be monitored to
give a basis for prediction of the changes that might result from greenhouse changes.
A great deal is known about the reproductive biology of marine turtles, but there
is much more research to be conducted if man is to reconcile his behavior with the
nonnegotiable biological constraints of marine turtles. Their need for unpolluted
foraging areas from which to gather the energy necessary for growth and reproduc
tion must be respected. Their need for unencumbered nesting areas that they have
selected, on which they may lay their eggs, must be guaranteed.
REFERENCES
1. Georges, A., Limpus, C. J,, and Parmenter, C. J., Natural history of the Chelonia, in
Fauna of Australia, Vol. 2A, Amphibia & Reptilia, Glasby, C. J., Ross, G. J. B., and
Beesley, P. L., Eds., Australian Government Publishing Service, CanbeiTa, 1993, 120.
2. Hirth, H. F., Some aspects of the nesting behavior and reproductive biology of sea
turtles, ^m. Zool, 20, 507, 1980.
3. NRC (National Research Council), The Decline o f the Sea Turtles, National Academy
of Science Press, Washington, D.C., 1990, 259.
4. Hendrickson, J. R., Nesting behavior of sea turtles with emphasis on physical and
behavior detemiinants of nesting success or failure, in Biology and Conservation o f
Sea Turtles, Bjorndal, K., Ed., Smithsonian Institution Press, Washington, D.C., 1982,
53.
5. Van Buskirk, J. and Crowder, L. B., Life-history variation in marine i\xvi\QS,Copeia,
1994, 66, 1994.
6. Bjorndal, K. A., Carr, A., Meylan, A. B., and Mortimer, J. A., Reproductive biology
of the hawksbill Eretmochelys imhricata at Tortuguero, Costa Rica, with notes on
the ecology of the species in the Caribbean, B/o/. Conserv., 34, 353, 1985.
7. Limpus, C. J., Miller, J. D., Parmenter, C. J., Reimer, D., McLachlan, N., and Webb,
R., Migration of green (Che Ionia my das) and loggerhead (Caretta caretta) turtles to
8. Bjorndal, K. A., Biology and Conservation o f Sea Turtles, Smithsonian Institution
Press, Washington, D.C., 1982.
9. Bustard, H. R., Sea Turtles: Natural History and Conservation, Collins, Sydney, 1972.
10. CaiT, A. F., Handbook of Turtles. The Turtles o f the United States, Canada, and Baja
California, Cornell University Press, Ithaca, NY, 1952.
11. Groombridge, B. and Luxmoore, R., The Green Turtle and Hawksbill (Reptilia:
Cheloniidae): World Status, Exploitation and Trade, Secretariat of the Convention
on International Trade of Endangered Species, Cambridge, 1989, 601.
12. Rebel, T. R, Sea Turtles and the Turtle Industry o f the West Indies, Florida, and the
Gulf o f Mexico, University of Miami Press, Coral Gables, FL, 1974.
13. Pritchard, P. C. H. and Trebbau, P, The Turtles o f Venezuela, Society for the Study
of Amphibians and Reptiles, Contrib. Herpetol. 2, Caracas, Venezuela, 1984, 403.
14. Pritchard, P. C. H., Encyclopedia o f Turtles, T.F.H. Publications, Neptune, NJ, 1979.
15. Anon., Symposium; behavioral and reproductive biology of sea turtles. Am. ZooL,
20, 485, 1980.
16. Bacon, R, Berry, F., Bjorndal, K., Hirth, H., Ogren, L., and Weber, M., Eds., Pro-
ceedings o f the Western Atlantic Turtle Symposium, Vol. 1, 306 pp., Vol. 2, 318 pp.,
Vol. 3, University of Miami Press, Miami, FL, 1984, 514.
17. Bjorndal, K. A., Bolten, A. B., Johnson, D, A., and Eliazar, R, CompWtxs,Proceedings
o f the 14th Annual Workshop on Sea Turtle Biology and Conservation, NOAA Tech.
Mem. NMFS-SEFSC-351, Miami, FL, 1994, 323.
18. Eckert, S. A., Eckert, K. L., and Richardson, T. H., Compilers, o f the
Ninth Annual Workshop on Sea Turtle Biology and Conservation, NOAA Tech.
Memo. NMFS-SEFSC-232, Miami, FL, 1994.
19. James, R., Compiler, Proceedings o f the Australian Marine Turtle Conservation
Workshop, Queensland Department of Environment and Heritage and Aust. Nature
Conservation Agency, CanbeiTa, Australia, 1994, 208.
20. Ogren, L., Berry, F., Bjorndal, K., Krumpf, H., Mast, R., Medina, G., Reichart, H.,
and Witham, R., Proceeding of the Second Western Atlantic Turtle S y m p o s i u m , A A
Tech. Memo. NMFS-SEFC-226, Panama City, FL, 1989.
21. Richardson, J. I. and Richardson, T. H., Compilers, Proceedings o f the Twelfth Annual
Workshop on Sea Turtle Biology and Conservation, NOAA Tech. Memo. NMFS-
SEFSC-361, Miami, FL, 1995, 274.
22. Richardson, T. H., Richardson, J. L, and Donnelly, M., Compilers, Proceedings o f
the Tenth Annual Workshop on Sea Turtle Biology and Conservation, NOAA Tech.
Memo. NMFS-SEFSC-278, Miami, FL, 1990, 286.
23. Schroeder, B., Compiler, Proceedings of the Eighth Annual Workshop on Sea Turtle
Biology and Conservation, NOAA Tech. Memo. NMFS-SEFSC-214, Miami, FL,
1988, 135.
24. Schroeder, B. and Witherington, B., Compilers, Proceedings o f the Thirteenth Annual
Workshop on Sea Turtle Biology and Conservation, NOAA Tech. Memo. NMFS-
SEFSC-341, Miami, FL, 1994, 281.
25. Bustard, H. R., Turtles of coral reefs and coral islands, in Biology and Geology o f
Coral Reefs, Vol. 3 Part 2 Biol., Jones, O. A. and Endean, E., Eds., Academic Press,
New York. 1976. 343.
26. Bustard, H. R., Population dynamics of sea turtles, in Turtles: Perspectives and
Research, Harless, M. and Morlock, H., Eds., John Wiley & Sons, New York, 1979,
523.
27. Limpus, C. J., Family Demiochelyidae, in Fauna o f Australia, Vol. 2A, Amphibia &
Reptilia, Glasby, C. J., Ross, G. J. B., and Beesley, P. L., Eds., Australian Government
Publishing Service, Canberra, Australia, 1993, 139.
28. Limpus, C. J. and Miller, J. D., Family Cheloniidae, in Fauna o f Australia, Vol. 2A,
Amphibia & Reptilia, Glasby, C. J., Ross, G. J. B., and Beesley, P. L., Eds., Australian
Government Publishing Service, Canberra, Australia, 1993, 113.
29. Miller, J. D., Embryology of marine turtles, in Biology o f the Reptilia, Vol. 14A,
Gans, C., Biliett, F. and Maderson, P. F. A., Eds., Wiley-Interscience, New York, 1985,
269.
30. Moll, E.O., Reproductive cycles and adaptations, in Turtles: Perspectives and
305.
31. Mrosovsky, N., Thermal biology of sea turtles. Am. ZooL, 20, 531, 1980.
32. Parkes, A. S., Reproduction in the green sea turtle, Che Ionia my das, Symp. ZooL Soc.
London, No. 46, 253, 1981.
33. Wilbur, H. M. and Morin, P. J., Life history evolution in turtles, in Biology o f the
Reptilia, Vol. 16, Ecology B, Gans, C. and Huey, R., Eds., Alan R. Liss, New York,
1988, 387.
34. Zwinenberg, A. J., The leatherback (Dermochelys coriácea), one of the largest living
reptiles. Bull. Md. Herpetol. Soc., 10, 42, 1974.
35. Zwinenberg, A. J., The olive ridley, Lepidochelys olivácea (Eschscholtz, 1829):
probably the most numerous marine turtle today. Bull. Md. Herpetol. Soc., 12, 75,
1976.
36. Zwinenberg, A. J., Kemp’s ridley, Lepidochelys kempii (Garman, 1880), undoubtedly
the most endangered marine turtle today (with notes on the current status of Lepido
chelys olivácea). Bull. Md. Herpetol. Soc., 13, 170, 1977.
37. Bjorndal, K. A., Demography of the breeding population of the green invÚQ,Chelonia
mydas, at Tortuguero, Costa Rica, Copeia, 1980, 525, 1980.
38. Bustard, H. R., Greenham, P, and Limpus, C. J., Nesting behavior of loggerhead and
flatback turtles in Queensland, Australia, Proc. K. Ned. Akad. Wet. Ser. C, 78, 111,
1975.
39. Carr, A., Hirth, H., and Ogren, L., Ecology and migrations of sea turtles. 6. The
hawksbill turtle in the Caribbean sea. Am. Mus. Novit., 2248, 1, 1966.
40. Carr, A., Carr, M. H., and Meylan, A. B., The ecology and migrations of sea turtles.
7. The west Caribbean green turtle colony. Bull. Am. Mus. Nat. Hist., 162, 1, 1978.
41. Frazier, J., Observations on sea turtles at Aldabra, Philos. Trans. R. Soc. London Ser.
B, 260, 373, 1971.
42. Hendrickson, J. R., The green sea turtle, Chelonia mydas (Linn.) in Malaya and
43. Hendrickson, J. R., The ecological strategies of sea turtles. Am. Zool., 20, 597, 1980.
44. Hughes, G. R., The sea turtles of South-East Africa. I. Status, morphology and
distributions, S. Afr. Assoc. Mar. Biol. Res. Oceanogr. Res. Inst. Invest. Rep., No. 35,
1974.
45. Hughes, G., The sea turtles of South-East Africa. II. The biology of the Tongaland
loggerhead turtle, Caretta caretta L., with comments on the leatherback turtle, Der-
mochelys coriácea L., and the green turtle, Chelonia my das L., in the study region,
S. Afr Assoc. Mar. Biol. Res. Oceanogr. Res. Inst. Invest. Rep., No. 35, 1974.
46. Limpus, C. J., The flatback turtle, Chelonia depressa Garman, in southeast Queen
sland, Australia, Herpetologica, 27, 431, 1971.
47. Limpus, C. J. and McLachlan, N. C., Observations on the leatherback turtle,Dermo-
chelys coriácea (L.), in Australia, Wildl. Res., 6, 105, 1979.
48. Pritchard, P. C. H., Sea turtles of the Guianas, ^ m//. FI. State Mus. Biol. Ser., 13, 85,
1969.
49. Schulz, J. R, Sea turtles nesting in Surinam, Z6>6>/. Verh. Rijksmus. Nat. Hist. Leiden,
No. 143, 1975.
50. Simon, M. H., Ulrich, G. F., and Parkes, A. S,, The green sea turtle {Chelonia mydas):
mating, nesting and hatching on a farm, 7. Zool. (London), 111, 411, 1975.
51. Wood, J. R. and Wood, F. E., Reproductive biology of captive green sea turtles
Chelonia mydas. Am. Zool., 20, 499, 1980.
52. Dodd, C. K., Jr., Synopsis of the biological data on the loggerhead sea IwriX^Caretta
caretta (Linnaeus 1758), U.S. Fish Wildl. Serv. Biol. Rep., 88, 110, 1988.
53. Hiith H. F, Synopsis of the Biological Data on the Green Turtle, Chelonia mydas
(Linnaeus) 1758, FAO Fish. Synop., 1971, 85.
54. Hirth, H. F., Chelonia Brongniart, green sea turtles. Catalogue o f American Amphib
ians & Reptiles, Society for the Study of Amphibians and Reptiles, Athens, OH,
248.1, 1980.
55. Hirth, H. F., Chelonia mydas (Linnaeus) green turtles, Catalogue o f American
Amphibians & Reptiles, Society for the Study of Amphibians and Reptiles, Athens,
OH, 249, 1, 1980.
56. Marquez, M. R., Synopsis o f Biological Data on the Kemps Ridley Turtle, Lepido-
chelys kempi (Garman, 1880), NOAA Tech. Memo. NMFS-SEFSC-343, Miami, FL,
1994, 91.
57. Pritchard, P. C. H., The Leatherback or Leathery Turtle, Dermochelys coriácea,
Monogr. 1, Int. Union Conserv. Natur. Res., Morges, Switzerland, 1971, 1.
58. Pritchard, P. C. H., Dermochelys coriácea, Catalogue o f American Amphibians &
Reptiles, Society for the Study of Amphibians and Reptiles, Athens, OH, 238.1, 1980.
59. Pritchard, P. C. H. and Marquez, R. M., Kemp’s Ridley Turtle or Atlantic Ridley
Lepidochelys olivácea, Monogr. 2, Int. Union Conserv. Natur. Res., 1973, 1.
60. Wilson, R. V. and Zug, G. R., Lepidochelys kempii (Gannan) Kemp’s Ridley sea
turtle, Tortuga Lora, Catalogue o f American Amphibians & Reptiles, Society for the
Study of Amphibians and Reptiles, Caracas, Venezuela, 509.1, 1991.
61. Witzell, W. N., Synopsis of biological data on the hawksbill turtle, Eretmochelys
imbricata (Linnaeus, 1766), FOA Fish. Synop., 137, 1983, 78.
62. Zug, G. R. and Ernst, C. H., Lepidochelys Fitzinger, Ridley sea turtles. Catalogue of
American Amphibians & Reptiles, Society for the Study of Amphibians and Reptiles,
Caracas, Venezuela, 587.1, 1994.
63. Marquez, M. R., Villanueva, O. A., and Penaflores, S. C., Sinopsis de Datos Biológicos
Sobre la Tortuga Goiñndi Lepidochelys olivácea (Eschscholtz, 1829), Sinopsis sobre
la Pesca (INP/S2, SAT), INP,1976, 2:61p.
64. Dodd, C. K., Caretta Rafinesque, loggerhead sea turtles, Catalogue o f American
Amphibians and Reptiles. Society for the Study of Amphibians and Reptiles, Caracas,
Venezuela. 482.1. 1990.
65. Dodd, C. K., Carena caretta (Linnaeus) loggerhead sea turtle, Catalogue o f American
Amphibians and Reptiles, Society for the Study of Amphibians and Reptiles, Caracas,
Venezuela, 483.1, 1990.
66 . Bonnet, B., Introduction Bibliographique a la Physiologie de la Tortue Verte, Che-
lonia mydas (L.)., Collection Travax et Documents, 4 (UER Sciences), Universitaire
de la Reunion, Université Française de l’Ocean Indien, 1980, 31 pp.
67. Gilboa, I. and Dowling, H. G., A bibliography on the green turtle, Chelonia mydas,
Herp. Info. Search Syst. Am. Mus. Nat. Hist. Publ. Herpt., 1,1, 1973.
68. Limpus, C. J., Gyuris, F., and Miller, J. D., Reassessment of the taxonomie status of
the sea turtle genus Natator McCulloch 1908, with a redescription of the genus and
species, Trans. R. Soc. S. Aust., 112, 1, 1988.
69. Fckert, S. A., Fckert, K. L., Ponganis, R, and Kooyman, G. L., Diving and foraging
behavior of leatherback sea turtles (Dermochelys coriacea). Can. J. ZooL, 67, 2834,
1989.
70. Sternberg, J., The Worldwide Distribution o f Sea Turtle Nesting Beaches, Center for
Environmental Education, Washington, D.C., 1981.
71. Gasperetti, J., Stimson, A. F., Miller, J. D., Ross, J. R, and Gasperetti, P. R., Turtles
of Arabia, Fauna o f Arabia, 13, 170, 1993.
72. Limpus, C. J., Couper P. J., and Read, M. A., The green turtle, Chelonia mydas, in
Queensland: population structure in a warm temperate feeding area, Mem. Queensl
Mus., 35, 139, 1994.
73. Limpus, C. J., Couper, P. J., and Read, M. A., The loggerhead turtle, Caretta caretta,
in Queensland: population structure in a warm temperate feeding area, Mem. Queensl
Mus., 37, 195, 1994.
74. Limpus, C. J., Puberty and first breeding in Caretta caretta, in Proceedings o f the
10th Annual Workshop on Sea Turtle Biology and Conservation,NOAA Tech. Memo.
NMFS-SEFSC-278, 1990, 81.
75. Limpus, C. J. and Walter, D. G., The growth of immature green turtles {Chelonia
mydas) under natural cou(Xit\om, Her pet ologica, 36, 162, 1980.
76. Limpus, C. J., The hawksbill turtle, Eretmochelys imbricata, in Queensland: popu
lation structure within a southern Great Barrier Reef feeding ground, Wildl. Res., 19,
489, 1992.
77. Limpus, C. J. and Reed, P, The green turtle, Chelonia mydas, in Queensland: pop
ulation structure in a coral reef feeding ground, in Biology o f Australasian Frogs and
Reptiles, Grigg, G., Shine, R., and Ehmann, H., Eds., Surrey Beatty and Sons, Sydney,
1985, 47.
78. Carr, A., The Ascension Island green turtle colony, Copeia, 1975, 547, 1975
79. Meyian, A. B., Sea turtle migration — Evidence from tag returns, in Biology and
Conservation of Sea Turtles, Bjorndal, K., Ed., Smithsonian Institution Press, Wash
ington, D.C., 1982, 91.
80. Bowen, B. W., Meyian, A. B., Ross, R, Limpus, C. J., Balazs, G. H., and Avise,
J. C., Global population structure and natural history of the green turtle (Chelonia
mydas) in terms of matriarchal phylogeny. Evolution, 46, 865, 1992.
81 . Bowen, B. W., Meyian, A. B., and Avise, J. C., An odyssey of the green turtle:
Ascension Island revisited, Proc. Natl. Acad. Sci., 86, 573, 1989.
82. Nonnan, J., Moritz, C., Limpus, C., and Prince, R., Population genetics as a tool for
managing marine turtle populations, in Proceedings o f the Australian Marine Turtle
Conservation Workshop, James, R., Compiler, Queensland Department of Environ
ment and Heritage and Australian Nature Conservation Agency, Canberra, Australia,
1994, 92.
83. Broderick, D., Moritz, C., Miller, J. D., Guinea, M., Prince, R. I. T., and Limpiis,
C. J., Genetic studies of the hawksbill turtle, Eretmochelys imhircata: evidence for
multiple stocks in Australian waters, Pac. Conserv. Biol., 1, 121, 1994.
84. Bowen, B., Avise, J. C., Richardson, J. L, Meylan, A. B., Margaritoulis, D., and
Hopkins-Murphy, S. R., Population structure of loggerhead turtles, Caretta caretta,
in northwestern Atlantic Ocean and Mediterranean Conserv. Biol., 1, 834, 1993.
85. Bowen, B. W., Kamezaki, N., Limpiis, C. J., Hughes, G. R., Meylan, A. B., and Avise,
J. C., Global phylogeography of the loggerhead turtle {Caretta caretta) as indicated
by mitochrondrial DNA haplotypes. Evolution, 48, 1820, 1994.
86 Bowen, B. W., Abreu-Grobois, F. A., Balazs, G. H., Kamezaki, N., Limpus, C. J.,
and Ferl, R. J., Trans-Pacific migrations of the loggerhead turtle (Caretta caretta)
demonstrated with mitochrondrial DNA markers, Proc. Natl. Acad. Sci., 92, 3734,
1995.
87. Gyruis, E. and Limpus, C. J., The loggerhead turtle, Caretta caretta, in Queensland:
population breeding structure, A Wildl. Res., 15, 197, 1988.
88 . Limpus, C. J., Fleay, A., and Baker, V., The flatback turtle, Chelonia depressa, in
Queensland: reproductive periodicity, philopatry and recruitment, Awvr. Wildl. Res.,
11, 579, 1984.
89. Limpus, C. J., Fleay, A., and Guinea, M., Sea turtles of the Capricorn Section, Great
Barrier Reef, in The Capricorn Section o f the Great Barrier Reef: Past, Present and
Future, Ward, W. T. and Saenger, R, Eds., Royal Society of Queensland and the
Australian Coral Reef Society, Brisbane, Australia, 1984, 61.
90. Bjomdal, K. A., Meylan, A. B., and Turner, B. J., Sea turtles nesting at Melbourne
Beach, Florida. I. Size, growth and reproductive biology,P/r>/. Conserv., 26, 65, 1983.
91. Limpus, C. J., A study of the Loggerhead Sea Turtle, Caretta caretta, in Eastern
Australia, PhD dissertation. University of Queensland, St Lucia, Australia, 1985.
92. Bose, P. and Le Gall, J. Y., Attachement spatial des tortues vertes Chelonia my das
aux plages de Tile de Tromelin (Ocean Indien), Oceanologica ACTA 1986, 9, 489,
1986.
93. Eckert, K. L., Environmental unpredictability and leatherback sea turtle (Permochelys
coriacea) nest loss, Herpetologica, 43, 315, 1987.
94. Wibbels, T, Owens, D. W., Limpus, C. J., Reed, P. C., and Amoss, M. S., Jr., Seasonal
changes in serum gonadal steroids associated with migration, mating and nesting in
the loggerhead sea turtle (Caretta caretta), Gen. Comp. Endocrinol, 79, 154, 1990.
95. Limpus, C. J., The green turtle, Chelonia mydas, in Queensland: breeding males in
the southern Great Barrier Reef, Wildl. Res., 20, 513, 1993.
96. Limpus, C. J., Estimation of tag loss in marine turtle research, Wildl. Res., 19, 457,
1992.
97. Hughes, G. R., Nesting cycles in sea turtles — typical or atypical, in Biology and
Conservation o f Sea Turtles, Bjorndal, K., Ed., Smithsonian Institution Press, Wash
ington, D.C., 1982, p 81.
98. Limpus, C. J., Eggler, R, and Miller, J. D., Long interval remigration in eastern
Australian Chelonia, in Proceedings o f the Thirteenth Annual Workshop on Sea Turtle
Biology and Conservation, Schroeder, B. and Witherington, B., Compilers, NOAA
Tech. Memo. NMFS-SEFSC-341, Miami, FL, 1994, 85.
99 Owens, D. W., The comparative reproductive physiology of sea turtles. Am. ZooL,
20, 549, 1980.
100. Licht, R, Evolutionary and functional aspects of pituitary gonadotropins in the green
turtles, Chelonia mydas. Am. ZooL, 20, 565, 1980.
101. Wibbels, T., Owens, D. W., Licht, P. Limpus, C. J., Reed, P. C., and Amoss, M. S.,
Jr., Serum gonadotropins and gonadal steroids associated with ovulation and egg
production in sea turtles, Gen. Comp. Endocrinol., 87, 71, 1992.
102. Licht, R, Owens, D., Cliffton, K., and Penaflores, C., Changes in LH and progesterone
associated with nesting cycle and ovulation in the olive ridley sea i\xn\Q,Lepidochelys
olivácea, Gen. Comp. Endocrinol., 48, 247, 1982.
103. Limpus, C. J. and Nicholls, N., The southern oscillation regulates the annual numbers
of green turtles {Chelonia mydas) breeding around northern Australia, Aust. Wildl.
Res., 15, 157, 1988.
104. Ehrhart, L. M., A review of sea turtle reproduction, R/o/ogy and Conservation o f Sea
Turtles, Bjorndal, K. A., Ed., Smithsonian Institution Press, Washington, D.C., 1982,
29.
105. Carr, A. F. and Giovannoli, L., The ecology and migrations of sea turtles. 2. Results
of field work in Costa Rica, 1955, Am. Mus. Novit., 1835, 1, 1957.
106. Booth, J. and Peters, J., Behavioral studies on the green turtle (Chelonia mydas) in
the sea, Anim. Behav., 20, 808, 1972.
107. Harrison, T., The edible turtle (Chelonia mydas) in Borneo. 2. Copulation, Sarawak
Mus. J., 6, 126, 1954.
108. Ulrich, G. F. and Owens, D. W., Preliminary note on reproduction of Chelonia mydas
under farm conditions, Proc. World Mariculture Soc\, 5, 205, 1974.
109. Witham, R., Breeding of a pair of pen-reared green turtles, Q. J. El. Acad. Sci., 33,
288, 1970.
110. Caldwell, D. K., The loggerhead turtles of Cape Romain, South Carolina (abridged
and annotated manuscript of W. P. Baldwin, Jr. and J. P. Loftin, Jr.), Bull. Fla. State
Mus. Biol. Ser., 4, 319, 1959.
111. Caldwell, D. K., Carr, A. F, and Ogren, L., Nesting and migration of the Atlantic
loggerhead turtle, Bull. Fla. State Mus. Biol. Ser., 4, 295, 1959.
112. Wood, F. G., Mating behavior of captive loggerhead turtles, Copc/úf, 1953, 184, 1953.
113. Chavez, H., Marcado y recaptura de individuos de tortuga lom, Inst. Nac. ínv. Biol.
Fres. Mexico, 19, 1, 1968.
114. Limpus, C. J., Pannenter, C. J., Baker, V., and Fleay, A., The flatback inxÚQChelonia
depressa, in Queensland: post-nesting migration and feeding ground distribution,
Aust. Wildl. Res., 10, 557, 1983.
115. Parmenter, C. J., Reproductive migration in the hawksbill turtle (Eretmochelys irnbri-
cata), Copeia, 1982, 271, 1983.
116. Harry, J. L. and Briscoe, D. A., Multiple paternity in the loggerhead turtle (Caretta
caretta), J. Genet., 19, 96, 1988.
117. Carr, A. F. and Ogren, L., The ecology and migrations of sea turtles. 4. The green
turtle in the Caribbean Sea, Bull. Am. Mus. Nat. Hist., 121, 1, 1960.
118. Carr, A. F. and Hirth, H., The ecology and migrations of sea turtles. 5. Comparative
features of isolated green turtle colonies, Am. Mus. Novit., 2091, 1, 1962.
119. Solomon, S. E. and Baird, T., Aspects of the biology of Chelonia mydas L., Oceanogr.
Mar. Biol. Ann. Rev., 17, 347, 1979.
120. Mortimer, J. A., The influence of beach sand characteristics on the nesting behavior
and clutch survival of green turtles (Chelonia mydas), Copeia, 1990, 802, 1990.
121 Mortimer, J. A., Factors influencing beach selection by nesting sea turtles, mBiology
and Conservation o f Sea Turtles, Bjorndal, K., Ed., Smithsonian Institution Press,
122. Ackerman, R. A., Physiological and ecological aspects of gas exchange by sea turtle
eggs, Am. ZooL, 20, 575, 1980.
123. Maloney, J. E., Darian-Smith, C., Takahashi, Y., and Limpus, C. J., The environment
for development of the embryonic loggerhead turtle (Caretta caretta) in Queensland,
Copeia, 1990, 378, 1990.
124. Pritchard, P. C. H., Bacon, R, Berry, A., Carr, A., Fletemeyer, J., Gallagher, R.,
Hopkins, S., Lankford, R., Marquez, M., Ogren, L., Pringle, W., Reichart, H., and
Witham, R., Manual o f Sea Turtle Research and Conservation Techniques, 2nd ed.,
Bjorndal, K. and Balazs, G., Eds., Center for Environmental Education, Washington,
D.C., 1983.
125. Spotila, J. R. and Standora, E. A., Environmental constraints on the thermal energetics
of sea turtles, Copeia, 1985, 694, 1985.
126. Marquez, M. R., Sea Turtles o f the World. An Annotated and Illustrated Catalogue
o f Sea Turtle Species Known to Date, FAO species catalogue, Vol. 11, FAO Fish.
Synop., Rome, Italy, 125, 1, 1990.
127. Guinea, M., Nesting seasonality of the flatback sea íuvíIq,Natator depressus (Garman)
at Fog Bay, Northern Territory, in Proc. Australian Marine Turtle Conservation
Workshop, James, R., Compiler, Queensland Department of Environment and Heri
tage and Australian Nature Conservation Agency, Canberra, Australia, 1994, 50.
128. Bustard, H. R. and Greenham, P, Nesting behavior of the green sea turtle on a Great
Barrier Reef island, Herpetologica, 25, 93, 1969.
129. Kaufman, R., Zur brutbiolooguie der meeresschildkrote, Caretta caretta caretta L.,
Mitt. Inst. Colombo-Aleman Invest, d e n t., 2, 46, 1968.
130. Tufts, C. E., Report on the Buritaca Marine Turtle Nesting Reserve with Emphasis
on Biological Data from ‘Operación Tortuga 1972’ and Recommendations fo r the
Future, Mimeographed report, Inderena, Bogota, Colombia, 1972.
131. Hailman, J. P. and Elowson, A. M., Ethogram of the nesting female loggerhead
{Caretta caretta), Herpetologica, 48, 1, 1992.
132. Stoneburner, D. L. and Richardson, J. I., Observations on the role of temperature in
loggerhead turtle nest site selection, Copeia, 1981, 238, 1981.
133. Bustard, H. R. and Greenham, P, Physical and chemical factors affecting hatching
in the green sea turtle, Chelonia mydas (L.), Ecology, 49, 269, 1968.
134. Hopkins, S. R., Murphy, T. M., Jr., Stansell, K. B., and Wilkinson, P. M., Biotic and
abiotic factors affecting nest mortality in the Atlantic loggerhead turtle, Annu.
Conf. Southeast. Assoc. Fish Wildl. Agencies, 32, 213, 1979.
135. Eherenfeld, D., Behavior associated with nesting, in Turtles: Perspectives and
417.
136. Hughes, G. R. and Mentis, A., Further studies on marine turtles in Tongaland, 2,
Lammergeyer, 3, 55, 1967.
137. Caldwell, D., Comments on the nesting behavior of Atlantic loggerhead sea turtles,
based primarily on tagging returns, Q. J. Fla. Acad. Sci., 25, 287, 1962.
138. Chavez, H., Tagging and recapture of the lora iuxile^ Lepdioclelys kempi, Int. Turtle
Tortoise J., 3, 14, 32, 1969.
139. Archer, M., Carnivorous Marsupials, Vol. 2, Royal Zoological Society of NSW,
Mosman, New South Wales, Australia.
140. Bustard, H. R. and Tognetti, K. P, Green sea turtles: a discrete simulation of density-
dependent population xe^gxxXuúon, Science, 163, 939, 1969.
141. Hughes, G. R., Bass, A., and Mentis, M., Further studies on marine turtles in Ton
galand, Lammergeyer, 1, 1, 1967.
142. Hughes, G. R., Further studies on marine turtles in Tongaland, 3 ,Lammergeyer, 12,
7. 1970.
143. Blanck, C. E. and Sawyer, R. H., Hatchery practices in relation to early embryology
of the loggerhead sea turtle, Caretta caretta (Linné), /. Exp. Mar. Biol. EcoL, 49,
163, 1981.
144. Aitken, R. N. and Solomon, S. E., Observations on the ultrastructure of the oviduct
of the Costa Rican green turtle {Cheionia mydas L.), J. Exp. Mar. Biol. EcoL, 21,
75, 1976.
145. Solomon, S. E. and Baird, T., Studies on the eggshell (oviductal and oviposited) of
Cheionia mydas L., J. Exp. Mar. Biol. EcoL, 22, 145, 1976.
146. Limpus, C. J., Baker, V., and Miller, J. D., Movement induced mortality of loggerhead
eggs, Herpetologica, 35, 335, 1979.
147. Parmenter, C. J., Incubation of the eggs of the green sea turtle. Cheionia mydas, in
Torres Strait, Australia: the effect of movement on hatchability, Aw.s’/. Wildl. Res., 7,
487, 1980.
148. Chan, E., White spot development, incubation and hatching success of leatherback
turtle {Dermochelys coriacea) eggs from Rantau Abang, Malaysia, Copeia, 1989, 42,
1989.
149. Sakai, H., Ichihashi, H., Suganuma, H., and Tatsukawa, R., Heavy metal monitoring
in sea turtles using eggs, Mar. Poll. Bull., 30, 347, 1995.
150. Packard, G. C. and Packard, M. J., The physiological ecology of reptilian eggs and
embryos, in Biology o f the Reptilia, Voi. 16, Ecology B, Gans, C. and Huey, R.,
Eds., Alan R. Liss, New York, 1988.
151. Ackerman, R. A. and Prange, H. D., Oxygen diffusion across a sea imiÌQ Cheionia
mydas eggshell. Comp. Biochem. Physiol, 43A, 905, 1972.
152. Ewert, M., The embryo and its egg: development and natural history, in Turtles:
Perspectives and Research, Harless, M. and. Morlock, H., Eds., John Wiley & Sons,
New York, 1979, 333.
153. Yntema, C. and Mrosovsky, N., Incubation temperature and sex ratio in hatchling
loggerhead turtles: a preliminary report. Mar. Turtle NewsL, 11,9, 1979.
154. Mrosovsky, N. and Yntema, C. L., Temperature dependence of sexual differentiation
in sea turtles: implications for conservation practices, 5/a/. Conserv., 18, 271, 1980.
155. Limpus, C. J., Reed, P, and Miller, J. D., Temperature dependent sex determination
in Queensland sea turtles: intraspecific variation in Caretta caretta, in Biology o f
Australasian Erogs and Reptiles, Grigg, G., Shine, R., and Ehmann, H., Eds., Surrey
Beatty and Sons, Sydney, Australia, 1985, 305.
156. Miller, J. D. and Limpus, C. J., Incubation period and sexual differentiation in the
green turtle Cheionia mydas L., in Proceedings o f the Melbourne Herpetological
Symposium, Banks, C. B. and Martin, A. A., Eds., Zoological Board of Victoria,
Melbourne, Victoria Australia, 1981, 66.
157. Morreale, S. J., Ruiz, G. J., Spotila, J. R., and Standola, E. A., Temperature dependent
sex determination: current practices threaten conservation of sea turtles, 216,
1245, 1982.
158. Ruiz, G., Standora, E., Spotilia, J., Morreales, S., Camhim, M., and Ehrenfeld, D.,
Artificai incubation of sea turtle eggs affects sex ratios of hatchlings. Abstract, ./o/ni
Ann. Meet. SSAR and Herpetol. League, 68, 1981.
159. Rimblot, R, Fretey, J., Mrosovsky, N., Lescure, J., and Pieau, C., Sexual differenti
ation as a function of the incubation temperature of eggs in the sea imiÌQ Dermochelys
coriacea (Vandelli, 1161) Amphibia-Reptilia 6, 83, 1985.
160. Mrosovsky, N., Bass, A., Corliss, L. A., Richardson, J. L, and Richardson, T. H.,
Pivotal and beach temperatures for hawksbill turtles nesting in Antigua, Caa. J. ZooL,
70, 1920. 1992.
161. Georges, A,, Limpus, C. J., and Stoutjesdijk, R., Hatchling sex in the marine turtle
Caretta caretta is detemiined by proportion of development at a temperature, not
daily duration of exposure, J. Exp. ZooL, 270, 432, 1994.
162. Prange, H. D. and Ackerman, R. A., Oxygen consumption and mechanism of gas
exchange of green turtle (Chelonia mydas) eggs and hatchlings, Copeia, 1974, 758,
1974.
163. Raotzkie, R., Mortality of loggerhead turtle eggs from excessive rainfall, Ecology,
40, 303, 1959.
164. Ackerman, R. A., Oxygen consumption by sea turtle {Chelonia mydas) eggs during
development. Physiol. ZooL, 54, 316, 1981.
165. Ackerman, R. A., Growth and gas exchange of embryonic sea turtles (Chelonia,
Caretta), Copeia, 1981, 757, 1981.
166. Kraemer, J. E. and Bell, R., Rain-induced mortality of eggs and hatchlings of log
gerhead sea turtles (Caretta caretta) on the Georgia coast, Herpetologica, 36, 72,
1980.
167. Carr, A. F. and Hirth, H., Social facilitation in green turtle siblings, Anim. Behav., 9,
68, 1961.
168. Bustard, H. R., Mechanism of nocturnal emergence from the nest in green turtle
hatchlings, Nature (London), 214, 317, 1967.
169. Mrosovsky, N., Nocturnal emergence of hatchling sea turtles: control by thermal
inhibition of diCiiviiy, Nature, 220, 1338, 1968.
170. Gyuris, E., Factors that control the emergence of green turtle hatchlings from the
nest, Wildl. Res., 20, 345, 1993.
171. Gyuris, E., The rate of predation by fishes on hatchlings of the green turtle, Coral
Reef, 13, 137, 1994.
172. Limpus, C. J., Management oriented research requirements, in Proc. Australian
Marine Turtle Conservation Workshop, James, R., Compiler, Queensland Department
of Environment and Heritage and Australian Nature Conservation Agency, Canberra,
Australia, 1994, 198.
173. MTSG, A global strategy for the conservation of marine turtles. International Union
for Conservation of Nature and Natural Resources: Marine Turtle Specialist Group,
24, 1995.
174. Salmon, M. and Witherington, B. E., Artificial lighting and seafinding by loggerhead
hatchlings: evidence for lunar modulation, Copeia, 1995, 931, 1995.
175. Limpus, C. J. and Fleay, A., Management and turtles, James Cook University Press,
Townsville, Queensland, Australia, 1983, 535.
176. Mrosovsky, N., Dutton, R, and Whitmore, C., Sex ratio of two species of sea turtles
nesting in Suriname, Can. J. ZooL, 62, 2227, 1984.
177. Davenport, J., Sea turtles and the greenhouse effect, Br. Herpetol. Soc. Bull., 29, 11,
1989.
178. Limpus, C. J., A marine resource case study: climate change and sea level rise —
probable impacts on marine turtles, in Climate Change and Sea Level Rise in the
South Pacific Region, Proceedings o f the 2nd SPREP Meeting, Hay, J. and Kaluwin,
C., Eds., Apia, Western Samoa, 1993, 157.
179. Main, A. R., Climate change and its impact on nature conservation in Australia, in
Greenhouse: Planning fo r Climate Change, Pearman, G. I., Ed., C.S.I.R.O., Mel
bourne. Victoria. Australia. 1989.
180. Georges, A., The influence of fluctuating temperature on hatchling sex ratios — a
model and proposed test using Caretta caretta, in Pfvc. Australian Marine Turtle
Conservation Workshop, James, R., Compiler, Queensland Department of Environ
ment and Heritage and Australian Nature Conservation Agency, 1994, 156.
181. Lanyon, J., Limpus, C. J., and Marsh, H. Dugongs and turtles; grazers in the seagrass
system, in Biology of Seagrasses, Larkum, A. W. D., McComb, A. J., and Shepherd,
S. A. Eds.. Elsevier. New York. 1989. 610.
4 The Nest Environment
and the Embryonic
Development of Sea
Turtles
Ralph A. Ackerman
CO NTENTS
4.1 Introduction..................................................................................................... 83
4.2 Embryonic Development................................................................................ 84
4.3 Temperature..................................................................................................... 85
4.4 Water......................................................................................................... 89
4.5 Gas Exchange................................................................................................. 97
4.6 Altered Beaches.............................................................................................. 99
Acknowledgments..................................................................................................103
References..............................................................................................................103
4.1 INTRODUCTION
Sea turtle eggs are deposited in a large clutch in a nest excavated by the female in
sandy marine beaches. Incubation requires several months during which time the
sea turtle embryo grows from a few cells to a fully formed organism capable of
independent existence. During this process energy stored in the egg by the female
is transformed into embryonic tissue. The developing embryo is coupled to the
nesting beach through the exchange of the O2 , CO2 , H2 O, and heat that is either
required or produced by the energy transformation process. The nesting beach is
the incubator for sea turtle embryonic development and operates by producing a
climate space that is appropriate for embryonic development. Deposition of the egg
clutch depends on the ability of the female to select and excavate a suitable nest
chamber. Successful incubation of the eggs depends on the presence of suitable
conditions in the beach sand. Among these conditions are temperature, humidity or
water potential, salinity, and levels of respiratory gases. A microclimate suitable for
incubation is generated by interaction among the physical characteristics of the
0-8493-8422-2/97/$0 00+$.50
material composing the beach, the physical structure of the beach, local climate,
and the eggs in the clutch. The microclimate is dynamic and changes with the state
of biological activity in the clutch and in the beach. Unfortunately, we know very
little about the microclimate of sea turtle eggs during incubation. We know even
less about how the physical characteristics of the beach influence the microclimate.
Our understanding of sea turtle embryonic adaptations for development in an
egg clutch buried in a nesting beach is not extensive. There is little information
available to describe how variation in beach characteristics, hydrology, topography,
or local climates influence the environment around eggs. The coupling between eggs
and beach is poorly described. No detailed climate space for a reptile egg of any
species has been described and, as a result, egg, embryonic, and developmental
adaptation are not well understood. This extensive lack of infomiation and insight
will not be remedied in this chapter. My focus will be on the environment for
incubation and on conceptual ways of examining and describing it while pointing
out remaining issues. Unfortunately, the lack of hard information on sea turtle
embryonic development and its environment limits the extent to which conceptual
synthesis is possible.
4.2 EMBRYONIC DEVELOPMENT

During the course of incubation, the embryo grows inside the egg from a few cells
at the beginning to a self-sufficient organism at hatching some 50 to 80 d later.
Throughout this process, the egg exchanges heat, H2 O, O2 , and CO2 with other eggs
in the clutch and with the beach surrounding the clutch. The developing embryo is
involved in this exchange, because it produces heat and CO2 and consumes O2 . The
embryo may also produce or require H2O. The osmotic environment of the egg must
be influenced to some extent by the exchange of H2O between the egg and the
environment. Because the extraembryonic parts of the egg are interposed between
the embryo and the environment, these structures have the potential to act as an
agent buffering or even regulating the embryonic milieu. Moreover, the embryo
clearly must be developing regulatory processes as it develops so that it is self-
sufficient at hatching. These regulatory processes can be expected to mature as
development progresses. Thus there is potential for a complex and shifting interac
tion between the egg and the embryo developing inside. Unfortunately, our knowl
edge of developmental and regulatory physiology inside the egg is rudimentary for
all reptile eggs and especially so for sea turtle eggs and embryos.
Embryonic growth has been described for a number of species. Miller^ reviewed
the embryology of sea turtles and presented a stage-by-stage description. He reports
that many measures of embryonic growth can be related to developmental time
(incubation period) using 2° and 3° polynomials. Ackerman^ described the change
in embryonic wet mass with time using a logistic equation. In this case, the growth
rate is described by a 2° polynomial. How embryonic growth may be influenced by
the environment or by exchange with the environment is unknown; however, embry
onic growth is unlikely to be independent of the environment.
Embryonic growth is a metabolic process. Fuel, stored in the egg yolk by the
female, is moved from the yolk to cells where it is transformed into energy and then
The Nest Environment and the Embryonic Development of Sea Turtles 85
used for differentiation and growth. Ackerman^ measured the O2 uptake of Caretta
and Chelonia eggs during development. The pattern of change in O2 uptake with
time was sigmoidal, peaking at about the time that hatching occurred and could be
described with a logistic curve. Vleck and Hoyt^ reported data for Dermochelys and
summarize the available data for reptile eggs in general. Peak O2 uptakes are a
function of initial egg size and, presumably, hatchling mass and increase with about
the 0.8 power of initial egg mass. Peak O2 uptakes for sea turtles are in the range
of 90 to 250 ml O2 day'k Total O2 cost of incubation is in the range of 38 to 90
kJ (after converting to energy units) and increases with about the 0.8 power of initial
egg mass. Note that the O2 cost is an estimate of the energy lost due to metabolic
inefficiency during development and is not an estimate of the total cost.
4.3 TEMPERATURE
Reptile embryos develop successfully over a range of temperatures.^'^ Typically,
incubation period (I, days) decreases as incubation temperatures (T, °C) increases.
The reciprocal of I can be used as a developmental rate (R^, days“^-^ Ackemian^
discusses how the relationship between I and T can be described. One descriptor is
the range of T over which development occurs with the end points being described
as upper and lower lethal temperatures. Another is the position of the curve relating
I or R^ to T (the response curve) on the T scale. A third is the slope of the response
curve of I or R^ to change in T. Sea turtle eggs are typically incubated at constant
temperature in the laboratory where I can be determined with the greatest precision.
Eggs in nature do not experience constant temperatures. While diurnal temperature
changes in the beach are damped by the depth of the nest, there is still likely to be
some diurnal change, and furthermore, nest temperatures may change systematically
over the course of incubation (see discussion below). Because there are no experi
mental or empirical observations, it is not clear whether or not the average of a
changing temperature produces the same effects as a constant temperature of the
same value. However, because temperature effects are typically nonlinear,^ we might
expect differences.
The thermal tolerance range (TTR) for development of sea turtle embryos
incubated at constant temperature appears to fall between about 25 to 27°C and 33
to 35°C and is around 10°C wide. Data for I and T for two sea turtle species are
shown in Figure 4.1. The data appear nearly linear as do the summarized data of
Miller. ‘ However, temperature effects on growth rate are typically nonlinear,^ and
nonlinear curves yield better fits to the data than do linear ones. The Arrhenius
equation is often used to describe temperature influences on rate functions^ and
takes the form:
R.
where R^ is a rate (days the pre-exponential factor (B, days'O. represents the
position of the curve on the T scale, R (kJ • mok^ • K~^) is the gas constant, T (°K)
is the incubator temperature, and E (kJ • mol~0 is the free energy. The quantity E
FIGURE 4.1 The influence of incubation temperature on incubation time. Open circles are
for the hawksbill turtle, Eretmochelys imhricata. The closed circles are for the loggerhead
turtle, Caretta caretta. Data are taken from the literature cited in the text.
can be used to describe the shape or sensitivity of the response of (or I) to change
in T. Q,o can also be used to define the shape or sensitivity,
_ |(T + 10) (rt-(t+io))

Qin =
where T is again in °K. Note that is temperature dependent, whereas E is single

valued. Ackerman^ reported an E of 38 kJ • mol ^ equivalent to a Q jq of 1.6. A
similar analysis for the Eretmochelys data shown in Figure 4.1 yielded an E of 6 6
kJ • moh^ equivalent to a QioOf 2.4 between 25 and 35°C. The values for the species
are probably not different,^ indicating that changes in temperature produce similar
changes in I. However, the position of the curves appear to be different, with the
Eretmochelys data displaced below the Caretta data. An explanation for this remains
to be found, but the difference may be related to beach temperature.
The influence of temperature on sea turtle embryonic sex determination is very
interesting and very puzzling. Temperature-dependent sex determination (TSD, also
called environmental sex determination, or ESD) among sea turtles operates to
produce female hatchlings at warm temperatures and males at cool. Mrosovsky^ has
recently reviewed TSD among sea turtles and Ewert et al.‘^for all turtles. The sensitive
period for sex determination appears to occur around the middle third of incuba
t i o n . T h e threshold temperature (or pivotal temperature, Tp^^) for the transition
from production of one sex to the other (i.e., a sex ratio of 1:1) is around 28 to 30°C
for all the species for which there are data. The amount of variation within species
is not well defined. Figure 4.2 illustrates the influence of constant incubation tem
perature on hatchling sex ratio for 4 species of sea turtle. The data are taken from
Paukstis and Janzen’^ and Mrosovsky et al.^^ A cumulative normal distribution
(represented by the line) is fitted to the data. This distribution is often used to describe
dose-response relationships.The estimated Tp^^ is 29.0°C with a standard deviation
of 1.77°C. Table 4.1 summarizes Tpj^ estimated by linear regression on data for the
individual species as well as on the pooled data. There is considerable scatter in the
data both within and among the species, but the Tp^^ values are quite similar among
the species. The data sets are not of sufficient quality to determine whether or not
Tpj^ varies within species. Typically, because of the endangered and threatened status
of sea turtles, only small samples with few replicates are collected. If we treat sex
determination as a binomially distributed random variable (i.e., either male or
female). Figure 4.3 illustrates the random variation which we might expect from
samples of size 10, replicated 10 ten times. Here, 10 replicates were drawn for 5
different, evenly spaced probabilities ranging from 0.1 to 0.9 with Tpi^ (p = 0.5) set
to 29°C. The situation is very much improved for larger sample sizes on the order
of 50 eggs. Much of the variation that is seen in sea turtle TSD data is likely to be
random. Tpj^ is of interest because it ought to be useful for predicting the beach
temperature required to produce a sex ratio of 50%. This assumes that the Tpj^
produced by constant incubator conditions can be extrapolated to the beach where
temperature is not constant. Mrosovsky^ reports that beach temperatures in some
locations are around the levels of Tp^^, but in other locations, such as the Atlantic
coast of Florida, average beach temperatures appear to be in excess of Tpj^ during
incubation with an excess production of female hatchlings as a result. The life history
consequences of imbalances in sex ratios are poorly understood.^ One problem, as
pointed out by Mrosovsky,^ is that we know little about the long-term variation of
beach temperature. This is of particular importance for long-lived animals such as
sea turtles. Unfortunately, there are few long-term temperature data available for
beaches. I have found a 12-year record for soil temperature at 10 cm near Gainesville,
FL. Figure 4.4 illustrates average July soil temperatures. The data for June and
August are similar. Although temperatures in beaches at greater depth are likely to
be somewhat cooler, we can see that the average temperature near Gainesville is
typically well in excess of Tpj^. This is consistent with the trend for nesting beach
temperature noted by Mrosovsky^ from a much smaller data base: beach tempera
tures at appropriate depths seem to be in excess of Tpj^ and ought to produce more
female than male hatchlings. It is possible if not likely that the Florida beaches have
been warm (in excess of Tpj^) for some time, perhaps in excess of a decade. To
understand TSD in this system we need to know what the normal beach temperature
is and how it varies, at least over some period of time in excess of the life span of
the sea turtles nesting on the beaches. This is a problem, but could potentially be
approached by correlating beach temperature with the temperature at some known
(and, hopefully, close) location.
Because temperature has such a pervasive influence on the embryonic develop
ment of sea turtles, it is important to understand the determinants of clutch temper
ature. The most important determinant is, of course, beach temperature. The second
important determinant is the exchange of heat between the clutch and the beach.
FIGURE 4.2 The influence of constant incubation temperature on hatchling sex ratio of
sea turtles. Loggerhead turtles (open circle), ridley turtles (diamond), green turtles (plus), and
hawksbill turtles (square) are shown. Data are taken from the literature as described in the
text. The line represents a cumulative noiTnal distribution fitted to the data as described in
the text.
TABLE 4.1
Pivotal Temperatures (°C) of Several Sea Turtle Species
Pivotal
Species temperature 95% Cl R2
CheIonia mydas 28.26 1.06 0.52

Caretta caretta 28.74 0.36 0.58
Lepidochelys olivácea 29.13 1.14 0.68
Eretmochelys imbricata 29.32 0.23 0.81
Grouped data 28.74 0.31 0.54
Temperatures were determined using inverse solution of a linear regies:
equation for sex ratio on temperature at a sex ratio of 50%.
There are extensive records of beach temperature, especially in regard to hatchling

sex d e t e r m i n a t i o n , b u t they tend not to be very long term or extensive. Beach
temperatures are typically in the range of 26 to 35°C. In some cases beach temper
ature is relatively constant over the period of incubation, and in others there may
be substantial change in beach temperature over incubation. Data for 7 beaches
along the Atlantic coast of Florida are shown in Figure 4.5. The data were measured
at a depth of 50 cm over a 24-h period, and, because there is little change in
FIGURE 4.3 The number of females resulting from samples from a binomial distribution.
Clutch size was 10 eggs and each sample for 5 probabilities between p = 0.1 and p = 0.9 was
run ten times. Only different results are shown. The line is a linear regression through the
data. At p = 0.5, Tpj^ was 29.0°C.
temperature at this depth, can be used as estimates of average monthly temperature.^^

Patterns of change in beach temperature surely reflect the physical characteristics
of the beach as well as the local boundary layer climate. As a result we can expect
considerable variation among beaches. Even distances of a few miles may be suffi
cient to influence patterns of beach temperature (Figure 4.5). Beach temperature is
determined by the exchange of thermal energy at the beach surface with the local
boundary climate and by the transmission of heat within the beach. Both the heat
balance at the surface of the soil and heat transmission within the soil column are
will u n d e r s t o o d , b u t this knowledge has never been applied to the analysis of
beach temperature. Models such as those of Horton et al.^^'^^ could be applied to
the thermal modeling of the sea turtle nesting beach with the intent of developing
more predictive relationships.
4.4 WATER
Sea turtle eggs contain a supply of water when they are oviposited. Water is stored
in the albumen at the beginning of incubation and is supplemented by water gained
through the transformation of yolk.^^ Because there are a number of anatomically
and physiologically different compartments inside the eggs, including the embryo
and extraembryonic membranes, the yolk sac, the chorioallantoic sac, and the
FIGURE 4.4 The average daily temperature at a soil depth of 10 cm in July at Gainesville,
FL. The solid line represents pivotal temperature for sea turtles. The asterisks represent average
daily temperatures for July 1995,
amniotic sac, water and osmotic balance may be complex. Vleck^^ has reviewed the
water economy and osmotic regulation of a number of reptile species, but we know
very little above the osmotic events and processes occurring inside the sea turtle
egg at any stage of embryonic development. Egg and embryonic osmotic balance
are especially interesting because the egg must be open to the exchange of water
vapor with the environment in order to exchange respiratory gases. The sea turtle
egg is coupled to the environment through the exchange of water vapor and possibly
through the exchange of liquid water.^^'^^ We know that for some species of reptile,
embryonic growth is influenced by egg water exchange^“^’^^ while for other species,
embryonic growth is independent of egg water exchange over some range of
exchange rates. Some reptile eggs must take up water in order to produce a viable
hatchling.A ckerm an et al.^^ suggested that water exchange was likely to be
relatively less important for large eggs such as sea turtles than for smaller reptile
eggs. We have recently found^^ that the embryonic growth of Caretta caretta is
independent of egg water exchange over a range of total egg water exchanges of
around -10 to +30% of initial egg mass. There also appeared to be little effect on
hatchability over this range, which is consistent with the observations of Bustard.*^
It is clear that hatching success can be correlated with some physical character
istics of the sand environment.-^^ Mortimer^^ found that hatching success of Ascension
FIGURE 4.5 Average monthly soil temperatures at a 50-cm depth in 8 beaches along the
Atlantic coast of Florida between Ft. Pierce Inlet and Boca Raton.
Island Cheionia mydas was negatively correlated with sand grain size. However,
there appears to be very little data on the water exchange of sea turtle eggs during
incubation. It seems to be generally accepted that eggs take up water during early
incubation and become more t urgid,but the quantity of water exchanged is unclear.
There is at least one reporU^ indicating that eggs may lose water near the end of
incubation, producing a reduction in clutch volume. There is very little available on
what happens in between. However, because the water in the sand surrounding the
clutch is likely to have a water potential on the order of -5 to -50 kPa and because
the water potential inside the egg is on the order of -900 kPa^^ “^^ there is a net
tendency of water to move into the egg during incubation. An added complication
for the understanding of egg water exchange is that sea turtle eggs are incubated in
large clutches rather than as single eggs. Eggs in the interior of the nest contact each
other and the spaces between because the interior of the clutch remains free of sand.
As the eggs take up water and swell, these spaces will shrink and disappear. Eggs
at the periphery of the nest contact sand over some fraction of their surface as well
as other eggs, a rather different physical situation. Moreover, clutches heat up by
several degrees during incubation with peak temperature occurring at the center of
the clutch. Temperature increases of this magnitude will have a significant impact
on egg water exchange.^
The general concepts of egg water exchange can be examined using a steady-
state model for the water exchange of a single egg in contact with soil:
M H.O L -A •(PeEH2O ■ ^ S ÌÌ2 0 (4.1)
where M hzo (liigM) is the rate of water exchange, Lp (mg day~^ kPa” 0 is the
permeability of the eggshell and shell membranes, A (cm^) is the area of the eggshell
and Peh20 (kPa) and Psh20 (kPa) are the vapor pressures of the egg and sand,
respectively. The product, Lp A, is often described as the water vapor conductance
of the eggshell:
G hjO = Lp ■a (4.2)
where G h20 = irig day~^ kPa~k Egg water exchange as modeled by Equation 4.1 is
driven by a vapor pressure (vp) difference between the egg and the surrounding sand
which describes the direction of the exchange, while the magnitude of the water
exchange is given by the product of the vp difference and the conductance. It is
important to note^^ that the permeability represents the sum, in series, of the eggshell
permeability (Lp®) and the soil permeability (Lp"):
1/Lp = 1/Lp® + 1/Lp" (4.3)
However, Lp" for moist sand is orders of magnitude greater than Lp®. As a result,
L ®becomes limiting^^ and Equation 4.3 reduces to:
1/Lp = 1/Lp® (4.4)
Thus Equation 4.2 should describe the only important conductance for water vapor
exchange of the sea turtle egg. The G h20 of only Caretta caretta appears to have
been measured (2175 mg • day~^ • kPa This value was close to that expected
for other flexible-shelled reptile eggs and about 3 orders of magnitude larger than
for a similarly sized bird egg.'^^ '^^ The magnitude of the Caretta G h20 means that
the eggs should lose water at a high rate in an environment only slightly drier than
the egg i tself. One implication seems to be that the environment of sea turtle eggs
must be very humid. In any event, the magnitude of the vp difference between the
inside and outside of the egg takes on much greater significance for the sea turtle
than it does for a bird egg. To my knowledge, no one has reported values for the vp
(Peiho) of sea turtle eggs. We can assume that the egg vp is equivalent to typical
vertebrate tissue vp. Data for other reptile eggs indicate that this assumption is not
unreasonable.^^’^^ It is possible, however, that Peh20 changes with incubation time
due to osmotic processes occurring inside the egg.^^ "^^Moreover, Peh20 is a function
of egg temperature and will change with time as the temperature of the egg changes
during development.^’^^
Sand vp (Psii20 ) at the other end of the vp difference driving egg water exchange
is a function of several variables. Psh20 is determined chiefly by the relationship
between the water content and water potential ('F) of the sand and is a function of
sand water content. Dry sand is comprised of solid sand grains and gas. When water
is present, gas is displaced until the water completely fills the gas space. The sand
is then considered to be saturated with water. At saturation, the hydraulic pressure
in the water is zero with reference to atmospheric pressure. As the water content of
the soil decreases from saturation, a negative pressure occurs in the soil water. This
pressure is often described as a water potential ('F, the potential energy per unit
volume, J • m"^) whose dimensions are equivalent to those of pressure (kPa) and
can be generated in several ways.^'^'^'^ ( 1 ) T' is a matric potential if it is due to the
capillary effects of surface tension in soil pores or to the binding of water to the
surface of soil particles.^^ This is a negative potential. As the soil dries, becomes
increasingly negative. A characteristic curve describes graphically the relationship
between sand water content and 'F of water. (2) An osmotic potential may be present
in the water if solutes such as salt are present. This is also a negative potential that
is proportional to the concentration of solute present. (3) If the liquid water in the
soil is connected hydraulically to a saturated water source (the water table), as is
likely to be the case for a nesting beach, then a gravitational potential is also present.
This potential is positive and is equivalent to 10 kPa for each meter above the water
table. The total potential in sand water is thus the sum of the matric (-), osmotic
(-), and gravitational (+) potentials. In terms of water transport in the sand, the
matric potentials tends to hold water in while the gravitational potential tends to
cause water to drain out. The vp of water in the soil gas phase is generally assumed
to be in equilibrium with 'F. Relative humidity (H,., %) can be related to water
potential (T^, 1 kPa = 1 J • kg"0 by:
= (R . T / MJ • ln(H,) (4.5)
where the gas constant R is 8.31 J • mok^ • the molar mass of water, M^, is
0.018 kg • mok’ and T is temperature (°K). Characteristic curves for sand from a
number of nesting beaches are shown in Figure 4.6. Only a limited range for W is
shown because totally dry sand is assumed to have a water potential of - 1 ,0 0 0 , 0 0 0
kPa.'^'’ Sand loses cohesiveness at 'F greater than -50 to -100 kPa and can no longer
support nest excavation. The horizontal section of the characteristic curve represents
the presence of bulk water in the sand. The vertical section of curve represents water
bound to the surface of the sand particles. Bound water remains after the bulk water
has drained from the soil."^^ When bulk water has been removed, leaving only bound
water, sand loses its cohesiveness and the individual particles no longer stick
together. The sand appears to be dry. When the external pressure exceeds the capillary
pressure holding water in the soil, the soil pores empty. When bulk water is present
in the soil, 'F is only weakly dependent on soil water content. This is pronounced
for sand (Figure 4.6), where water potential is still above -10 kPa even though most
of the water has been lost. I have data for sands from 16 different nesting beaches.
All release their bulk liquid water at matric water potentials less negative (lower)
than -10 to -15 kPa. In sands, the pores are relatively large and of similar size so
that emptying (or filling) occurs abruptly as matric forces are exceeded by drying
(or wetting) forces, but the characteristic curves are dependent to some extent on
sand grain size. More coarsely textured sands will hold water less tightly than more
finely textured sands because the pores are larger in the coarser sands. This means
that coarsely grained sands will empty at lower matric water potentials than more
finely grained sands.
cd
CL
c
CD
■*—*
o
a.
o3
cd
cd
Water Content (g g-^)

FIGURE 4.6 Characteristic curves for sand taken from a number of beaches along the
Atlantic coast of Florida. Details as described in the text.
The vp and present in the beach sand depends fundamentally on the quantity
of water in the sand and its distribution in the beach. The distribution of water in
the beach is a result of the balance between water entering and leaving a layer of
sand at any depth in the sand depth profile. If the distribution is stable it is because
on the average, the same quantity of water enters a layer as leaves it. Sea turtle
nesting beaches are characterized by what appear to be relatively stable distributions
of beach water that are far from equilibrium and dry out only slowly in the absence
of rain.^^ The elements of a water balance for a beach are presented by Ackerman.^^
Nesting beaches are likely to have a source of liquid water (a water table) present
not far from the surface (1-5 meters). The water table is close enough to the surface
to generate a gravitational potential in the beach. Water will drain downward toward
the water table, but slowly because the 'T difference driving water movement is
small (on the order of a few kPa). The chief source of water in the beach is rain.
Water injected by rain enters the beach and drains down its gravitational field until
it reaches equilibrium with the capillary forces (the matric water potential). Thus at
1 m above the water table the equilibrium matric potential would be -10 kPa,
offsetting the 10 kPa gravitational potential and producing a total potential of 0 kPa.
Note that sand with a NKof ~10 kPa is not dry and still retains 1 to 2% water. The
equilibrium condition is approached slowly because the potential differences are
usually small, as demonstrated by the Caretta nesting beaches in Israel where no
rainfall occurs during the nesting season. When drying at the surface occurs after
a rainfall, water is lost from the soil to the atmosphere.^"^’^^"^Âs water is lost at the
surface, it is drawn up from the underlying sand and the water potential of the sand
falls below its equilibrium point. Water is, in turn, pulled up from the free water
surface further down in the sand column. Water transport in the soil column produces
a characteristic profile in which, after a period of drying, a layer of dry sand comes
to overlie a layer of wetter sand. This distribution is familiar to anyone who has
worked on a sea turtle nesting beach where a layer of dry sand always overlies
moister sand underneath (in the absence of rain). The top, dry soil layer limits water
loss to the atmosphere from the wetter layer. Near equilibrium, liquid water may be
drawn up at a rate sufficient to balance the water lost across the suiface.^'^’^^'^^ The
evaporation from the surface is self-limiting. This is because the rate of water vapor
transport decreases as the length of the diffusion path increases. It is also because
the density of vapor in the pathway decreases substantially after short periods of
time.^'^’^^ The thickness of the dry layer will depend, among other things, on the time
since the infiltration of surface water, on the sand structure, and on the depth of the
water table. Given sufficient time after surface infiltration, an equilibrium distribution
may be reached, but this is likely to require weeks to months.^^ Thus in Israel, with
no rain during the nesting season, it appears to take in excess of 3 months to approach
equilibrium.^^ deeper the liquid water surface, the thicker the dry layer is likely
to be. It is also likely to be thicker in coarser sand than in finer sands because coarse
sands hold water less tightly than finer sands. This phenomenon may account for
the observation by Mortimer^^ that hatching success was lower in more coarsely
grained Ascension Island beaches. These beaches appeared to be drier than the more
finely grained beaches. Most likely, drying fronts had penetrated much more deeply
in the coarser beaches, resulting in a much drier environment for the eggs, perhaps
dry enough to produce an increase in egg mortality.
Reports of water content profiles in sea turtle nesting beaches are not extensive.
Ackerman^^ presented a profile for an Israeli beach taken near Caretta nests. Figure
4.7 presents the average water content in five beaches along the Atlantic coast of
Florida.'^^ The profiles appear as described above with a dry layer at the surface, an
intennediate layer with increased water content, and wet layer at the intersection of
the soil column with the water table. The moist intermediate layer is characterized
by an essentially constant water content throughout its depth. The water content was
around 4 to 6 % (by mass) and was far from the saturation values of around 35 to
45%. It is in this layer that sea turtle eggs are placed. The most important difference
between the Florida and Israeli beaches is in the frequency of rainfall. In Florida,
rain often occurs daily and weekly during the nesting season. The result is that the
water content of Florida beaches is higher throughout the season, while the water
content in Israel starts out high (but lower than in Florida) and decreases during the
nesting season as gravitational drainage occurs. Moreover, a drying front forms in
both locations, but is much deeper in the Israeli beach (15 to 20 cm) than in the
Florida beaches (5 to 10 cm). In both locations, the dry sand at the surface has a ¥
FIGURE 4.7 Water content profile for 5 beaches along the Atlantic coast of Florida. The
beach surface is at 0-cm depth. Water table is typically in excess of 1 m.
of around -100,000 kPa or more while the matric T in the intermediate layer is
around -5 to -10 kPa. The drying front where the transition from dry sand to moist
sand occurs is around 5 cm wide and has a ^ gradient on the order of 10,000 to
20,000 kPa • cm~h The presence of salt in the water will increase T somewhat.
Mortimer^^ and Rimkus et al."^^ report low levels of salt in beach water producing
osmotic potentials on the order of -10 to -60 kPa. These represent very low salt
concentrations because seawater has a water potential of around -2000 kPa. Rainfall
is likely to flush salt from beach water rapidly. Rimkus et al.,"^^ studying the effect
of beach renourishment on beach water, found that beaches renourished with sand
in a seawater slurry lost all their salt in two weeks or less. Average total (matric
plus osmotic potential) is shown for some Florida beaches in Figure 4.8. Total is
likely to be higher in locations such as Israel, where there is much less rainfall.^^^
It appears that the W and therefore the water vp in the sand around the sea turtle
clutch is relatively constant throughout the typical incubation period. If we assume
a sand temperature of 30°C and take a typical 'F to be around -30 kPa, then the
water vp in the sand is around 4.232 kPa. If the osmotic concentration of the egg
is typical of vertebrate tissue, then the water vp of the egg is around 4.207 kPa. The
water vp difference across the eggshell is around 0.025 kPa. Substituting this value
and the value for G h20 (2175 mg • day~^ • kPa~0 into Equation 1, we find that a
typical egg would take up around 54 mg day“^ of water vapor. Over 60 d, this is
around 3.3 g or about 6.5% of the mass of a 50-g Caretta egg. This water uptake
should be considered the maximum quantity possible because the eggs are incubated
in clutches rather than singly and because the clutch heats up. Eggs at the periphery
FIGURE 4.8 Total water potential (matric plus osmotic) as a function of depth for beaches
along the Atlantic coast of Florida. Data are averaged for 7 beaches. The increased variation
at depths greater than 80 cm is due to proximity to the water table which had a salinity typical
of seawater.
of the clutch will be closer to this value than eggs at the center of the clutch. Indeed,
eggs toward the center will heat up sufficiently to reverse the direction of water
vapor transport. An increase in egg temperature of 0.1 °C is sufficient to bring egg
water vp to the same level as sand water vp. Pearson"^^ analyzed the effect of clutch
position and heating on egg water exchange. His model indicated that eggs at the
periphery of the nest should experience a net water uptake and eggs at the center
of the nest a net water loss during incubation. Water loss for the central eggs occurred
near the end of incubation. The observation^'^ of a decrease in egg volume near the
end of incubation is consistent with the Pearson"^^ analysis.
4.5 GAS EXCHANGE

Sea turtle egg clutches are deposited in a medium, the nesting beach, which limits
respiratory exchange because O2 and CO2 are free to move only through the gas-
filled fraction of soil. This fraction is usually much less than half of the volume
present. Gas exchange in the clutch is limited for the same reason. Most of the clutch
is composed of the eggs themselves, and respiratory gases can only move through
the gas spaces among them. Respiratory gas exchange in these two circumstances
can occur by convection or diffusion. Convection in the beach can be induced by
temperature variation, changes in atmospheric pressure, or displacement of soil air
by water table movement. The magnitude of convection due to these sources is
thought to be very small."^^ Prange and Ackenuan"^^ found little or no effect of water
table movement (due to tidal pumping) on the gas exchange of Chelonia mydas
clutches, but Maloney et al.^^ speculated that tidal pumping may be important.
Diffusion of respiratory gases is likely to be the principle means of gas exchange
of sea turtle eggs. Net diffusion of any gas can only occur when there is a concen
tration difference for the gas. This means that, in order to supply the O2 or get rid
of the CO2 associated with developmental metabolism, concentration differences
must set up between the center and edge of the egg clutch and between the edge of
the egg clutch and the surrounding sand. As egg and clutch metabolism change with
time during incubation, so too should the concentration differences which occur to
supply the respiratory needs of the clutch. Ackerman'^^ measured P02 and Pco2
the sand surrounding the clutch and inside the clutch of C. mydas and Caretta caretta
nests in Costa Rica and Florida. Pq, decreased in the center of the clutch from near
atmospheric levels at the beginning of incubation to around 12 to 14 kPa near the
end of incubation, while Pco, increased to around 4 to 6 kPa near the end of
incubation. Maloney et al.^'^ reported similar patterns of change, but smaller magni
tudes in Caretta nests in Queensland. Ackerman"^^ used a one-dimensional, spherical
diffusion model to quantitatively describe gas change by sea turtle clutches, treating
the medium (the beach) as if it were infinite. Among other things, this implies that
clutch gas exchange is independent of the beach surface. However, Maloney et al.^'^
concluded that the model used by Ackerman"^'^ was not adequate to describe clutch
gas exchange in their location. In the situation described by Maloney et al.,^‘^ the
surface may well have influenced clutch gas exchange. This would mean that gas
exchange through the top of the nest would have a disproportionate influence on
nest gases. The modeP^ can be modified easily to deal with this by adding dimen
sions. An explanation for the observation of Maloney et al.‘^ may lie in the relation
ship between the gas diffusivity of the sand, its water content, and the distribution
of water in the sand above the clutch (as described earlier). The gas diffusivity of
sand (D ) can be approximated by the relationship
(4.6)
where is the gas diffusivity in air and (j)g is the fraction of the sand that is air
(taking into account the water present). More accurate estimators are available if
required.^"^ As described above, a dry layer of considerable thickness is usually found
at the beach surface. This layer will have a greater than the moist underlying
layers where the water content is on the order of 2 to 10%. Egg clutches incubating
in drier climate will tend to be closer to the dry layer (i.e., the dry layer will penetrate
into the beach more deeply) than egg clutches incubating in wetter climates. The
increased thickness of the dry layer should (at some point) increase the contribution
of gas exchange through the top of the nest to overall nest gas exchange. For locations
where extensive surface drying may occur, such as Israel or Queensland, a radial
(one-dimensional) gas exchange model might not account well for clutch gas
exchange, and a two-dimensional model might be more suitable. It is also worth
considering the influence of nest crowding on clutch gas exchange. The model used
by Ackerman"^'^ assumed that each nest in the beach was independent of all other
nests; that is, that nests were not “breathing” each other’s gases. When nests are
crowded together, as, for example, may occur in hatchery situations, the gas exchange
of nests will not be independent of each other. It is not clear what crowding means
quantitatively, but if nests are within a meter or less of each other then it is likely
that the influence of one nest on another must be taken into account. As crowding
increases, then clutch gas exchange through the top or bottom of the nest becomes
progressively more important than gas exchange through the sides, and a two-
dimensional planar model may become more applicable.
Clutch gas is important because it establishes the background concentrations of
respiratory gases in which the individual egg must exchange the gases necessary to
support the metabolic activity of the developing embryo. Gas exchange between the
clutch gases and blood in the chorioallantoic capillaries of the egg occurs across the
eggshell and associated eggshell membranes. This gas exchange process has been
reviewed by Deeming and Thompson^^ for reptile eggs, but has not been extensively
studied. Ackerman and Prange^^ reported the O2 diffusion coefficient for Chelonia
mydas eggshell and outer shell membrane to be around 5 x 10"^ cm^ s^^ cm“^kPa ^
a magnitude about twice that of a similarly sized chicken egg. Because the O2 uptake
of the sea turtle egg is only about 2 0 % that of the chicken egg, the O2 difference
across the eggshell and membrane must be much smaller than for the chicken egg.
An interesting aspect of this is that the diffusivity of the sea turtle shell must be a
function of its water content. When the egg is oviposited, the shell is flooded with
fluid. This fluid, if allowed to remain in the shell, would present such a large
resistance to O2 transport (given the shell thickness) that the embryo would run out
of O2 well before the end of incubation. Therefore, the shell must be dried out in
some may; furthermore, as water is taken up by the egg during incubation, that water
cannot be allowed to accumulate in the eggshell. The relationship between egg gas
exchange and egg water exchange is unexplored.
4.6 ALTERED BEACHES

Among other problems faced by sea turtles is the disappearance of nesting beaches,
due either to submergence (sea level changes) or to erosional processes. Recent
projections based on estimates of climate warming indicate that ocean level may
rise by about 1 ft by 2 1 0 0 (and more thereafter, if nothing changes), submerging
many beaches. Beaches are dynamic geological structures with finite life spans,
moving in time and space; moreover, sea level has changed many times over the
geological life span of sea turtle species. The problem, however, is that the predicted
sea level change may occur quite rapidly. This will accelerate beach dynamics. An
added complication is that beaches are valuable economic entities. As a consequence
there is strong political interest in replacing or maintaining beaches when they begin
to disappear. The state of Florida in the U.S., a major nesting site for loggerhead
turtles {Caretta caretta) as well as for leatherback {Dermochelys coraceia) and green
{Chelonia mydas) turtles, suffers from major erosion of its primary sea turtle nesting
beaches. Around 40% of the Atlantic coast beaches are experiencing erosion and
33% are eroding at a critical rate.^^ Because of the economic value of beaches to
Florida, the eroded beaches are usually restored to something resembling a natural
condition using artificial renourishment techniques. Beach renourishment often
involves locating a suitable source of sand offshore of the eroded beach, pumping
the sand in a seawater slurry onto the eroded beach to supplement the natural sand
present, and shaping the beach platform mechanically. Because renourishment has
no influence on the nearshore processes which eroded the beach in the first instance
(or on sea level increases), it generally has to be repeated at intervals. It is clear that
the process of renourishment produces a beach that is different in some ways from
natural beaches. In particular, renourished beaches appear to be harder than natural
beaches and a vertical, erosional scarp often forms in the surface zone of the beach.
It has been argued that this is due to compaction of the new beach m a t e r i a l . T h e
continued use by sea turtles of the renourished beaches for egg incubation has
generated concern that beach renourishment might adversely influence sea turtle
reproduction. This is of some concern because, as we have discussed above, it is
known that the incubation climate critically influences embryonic development in
sea turtles and other reptiles. Moreover, beach hardening may well impact nest
excavation and structure as well as emergence from the nest of hatchlings. However,
there is little information available on how renourishment might change the climate
for development in the beach.
Rimkus et al."^^ examined the hydric properties of eight renourished and seven
natural beaches along the Atlantic coast of Florida in order to provide adequate
replication for analyzing treatment effects. Although the distribution of water in both
types of beach were similar, renourished beaches appear to retain more water than
natural beaches. The field capacity^^’"^^ of renourished beaches was greater than that
of natural beaches. The difference in water retention (i.e., increased water content)
is consistent with a difference in the characteristic curves of renourished and natural
sand. Renourished sand appears to hold slightly (but significantly) more water at
the occuiTing in the sand. The difference in water retention may be attributable
to differences in the sorting of sand composing the beach. The natural process of
beach formation results in extensive sorting of the sand both by layers and within
layers.^^ The orientation of the layer is determined by the wave wash on the beach
and is typically inclined. Renourished beaches are unlikely to be sorted in the same
way because they are not constructed in the same way.
Renourished beaches had a lower bulk density and were less compact than
natural beaches."^® This finding was not consistent with earlier work where the reverse
was reported.^^"^^ However, the earlier findings were not based on measurements of
bulk density; rather, increased compaction was inferred from measurements of beach
shear resistance using cone penetrometers.^® Increases in shear resistance can be due
to increases in compaction, but they may also be due to other factors such as sand
particle shape and surface characteristics and to interactions (e.g., chemical) between
particles.®^ Scarp formation can be attributed to an increase in the shear resistance®^
rather than to beach compaction and thus the presence of scarps does not confimi
that a beach is compacted. Another explanation needs to be sought. A measure of
beach shear resistance can also be described as a measure of beach hardening or
strength. Shear resistance describes the ability of a soil, in this case, beach sand, to
resist sliding along internal surfaces. In addition to the external force operating on
the object, the object exerts a normal force on the surface over which it is tending
to move. For a beach, the nonnal force would be the weight of sand (overburden)
pressing down at the point of movement. The force required to just cause the surfaces
to slide along each other (the point at which the material fails) is called the shearing
stress (Tf) and is related to the angle of stress of the normal force (a) and the friction
angle ((|)) by the equation:
Tf = o tan (|) (4.7)
For soils, 0 is defined as the angle of internal friction. This angle represents the sum
of the sliding friction ((j)^) and the effect of interlocking by the particles (9):
+ e (4.8)
The contributions of and 0 can be separated experimentally. Sliding friction is

characteristic of a particular soil and has to do with the nature of the surfaces which
are in contact. Interlocking describes the tendency for particles to move out of each
other’s path as shear (failure) occurs and is strongly influenced by the distribution
of particle sizes present (gradation) and density (compaction). Equations 4.7 and
4.8 demonstrate how beach shear resistance (x^) could change when only (|)g increases
with no change in compaction or if compaction changes with no change in the type
of particles in the beach ((|)g). Thus, the measurement of increased beach shear
resistance does not necessarily establish that the beach is also compacted. Measure
ment of increased shear resistance in renourished beaches would indicate beach
compaction, only when the particles on the renourished beach were identical to those
on a reference beach. The relationship between shearing strength (x^) and the normal
force (a) is describe by the Coulombe equation:
c + a tan (|) (4.9)
where an additional variable, cohesion (c), has been added to Equation 4.7. This
relationship can be deteimined in the laboratory. However, another factor must also
be considered. The new material used to renourish beaches may interact chemically
through organic molecules bound to the surface. This could also result in increased
shear resistance and should be evaluated. The scarps observed on renourished
beaches in the tidal wash zone of the beach can be of considerable height and can
be dangerous to humans, while limiting sea turtle access to the beach. Increased
shear strength acts to increase the stability of slopes^^ with the result that scarps
should be more stable on renourished beaches which exhibit increased shear resis
tance. Vertical cuts, such as scarps on beaches, are common in gullying or erosional
situations. The height of the scarp (HJ can be calculated from the angle of internal
friction ((|)) and the cohesion (c). To a first-order approximation, the greater the (|)
the greater the height (HJ of the scarp that can develop:
H, = 2 c /y tan (45 + 0/2) (4.10)
where y is the unit weight of sand. Thus the magnitude of scarp formation on a
renourished beach could serve as an index to changes in shear resistance of the
beach if compaction were measured in some independent way.
The influence of beach renourishment on sea turtle reproduction or egg incuba

tion has not been studied extensively. However, because the hydric, thermal, and
respiratory properties of soil are a function of soil wetness, it might be expected
that the microclimate of renourished beaches is different from that of natural
beaches."^® Whether or not and to what extent this difference might impact egg
incubation can only be conjectured at this point and must ultimately be assessed
empirically. Egg water exchange is driven by the difference in SKbetween the interior
of the egg and the beach.^^ Observed differences'^® indicate that it is likely that the
water potentials of renourished and natural beaches are so similar that the hydric
climate in renourished and natural beaches are identical. The matric W are in the
range of -5 to -15 kPa for both types of beach.'^® If the total of the beach is the
sum of the due to salt and the matric 'F, then the value is around -20 to -40 kPa
for both renourished and natural beaches. Because the driving force for water
exchange between eggs and beach is likely to be around 800 kPa,^^^ small differences
between the two beach types are unlikely to be quantitatively important. However,
the increased wetness is likely to alter the thermal and respiratory exchange of the
eggs in the clutch. The differences in observed water contend® between renourished
and natural beaches could produce a substantial increase in the thermal
conductivity^'^’®^ of renourished relative to natural beaches. This difference is likely
to be important in two different contexts. First, sea turtle nests increase in temperature
above beach temperature as incubation progresses,^^ with the center of the nest
increasing in temperature to a greater extent than the periphery. The increase in
temperature is a function of the degree to which heat is conducted away from the
nest. Thus, if beach thermal conductivity increases, we would expect to find that the
temperature in nests in renourished beaches would increase substantially less than
in natural nests. The difference should be sufficient to influence both the duration
of incubation and the water exchange of the egg^’^^ in clutches in renourished
beaches. The second context involves the temperature of the beach rather than the
temperature of the clutch. Beach temperature will be strongly dependent on the
movement of heat inside the beach and on heat exchange at the beach surface. These
processes are well understood for soils in general,-'^ but, to my knowledge, have not
been applied to nesting beaches. Renourished beaches should have greater thermal
conductivity than natural beaches as a consequence of their increased in water
content. In addition, because water stores large quantities of heat (relatively) an
increase in beach water content will result in an increase in the heat capacity®^ of
the beach. The differences in these thermal properties are likely to cause differences
in the beach temperature of renourished and natural beaches on a daily and seasonal
basis. The magnitude of such differences cannot be easily or reliably predicted.
However, it is well known that incubation temperature influences important biolog
ical variables such as hatchling sex ratio, where even small differences in temperature
at the right time and temperature can have important effects on sex determination.
Sand water content also influences the gas permeability of sand.^'^ '^^ It appears likely
that observed water content differences'^® could result in a substantial decrease in
gas permeability of renourished beaches. It can be estimated'^® that the gas exchange
of nests in renourished beaches would be impeded to some extent. This is likely to
result in lower concentrations of oxygen and higher concentrations of carbon dioxide
in the center of the clutch. The biological relevance of this potential effect remains
to be investigated, but it has been shown experimentally that sea turtle eggs are
sensitive to altered respiratory gas exchange.^’^"^
ACKNOWLEDGMENTS
This work has been supported by grants from the Israel-U.S. Binational Science
Foundation, the National Science Foundation, and the Florida Department of Natural
Resources. Amos Ar, Razi Dmi’el, Robert Horton, Richard Seagrave, and William
Bardsley have contributed greatly to my work and my education.
REFERENCES
1. Miller, J. D., Embryology of marine turtles, in Cans, C., Billett, F., and Maderson,
R F. A., Eds., B io lo g y o f th e R e p tilia , D e v e lo p m e n t A, Vol. 14, John Wiley & Sons,
New York, 1985, 269-328.
2. Ackerman, R. A., Growth and gas exchange of embryonic sea turtles (C h elo n ia ,
C a re tta ), C o p e ia , 1981, 757-765, 1981.
3. Ackerman, R. A., Oxygen consumption by sea turtle eggs during development,
P h y s io l. Z o o L , 54, 316-324, 1980.
4. Vleck, C. M. and Hoyt, D. F, Metabolism and energetics of reptilian and avian
embryos, in E g g In c u b a tio n : Its E ffe c ts on E m b r y o n ic D e v e lo p m e n t in B ir d s a n d
R e p tile s, Deeming, D. C. and Ferguson, M. W. J., Eds., Cambridge University Press,
New York, 1991, 285-306.
5. Deeming, D. C. and Ferguson, M. W. J., Physiological effects of incubation temper
ature on embryonic development in reptiles and birds, in Deeming, D. C. and Fer
guson, M. W. J., Eds., E g g In c u b a tio n : Its E ffe c ts o n E m b iy o n ic D e v e lo p m e n t in
B ir d s a n d R e p tile s , Cambridge University Press, New York, 1991, 147-172.
6. Ewert, M. A,, Embryology of turtles, in Cans, C., Billet, F, and Maderson, P. F. A.,
Eds., B io lo g y o f th e R e p tilia , Vol. 14, D e v e lo p m e n t A, John Wiley & Sons, New York,
1985, 75-268.
7. Ackerman, R. A., Temperature, time, and reptile egg water exchange, Isr. J. Z o o L ,
40, 293-306, 1994.
8. Mrosovsky, N., Sex ratios of sea turtles,/. E xp . Z o o L , 270, 16-27, 1994.
9. Ewert, M. A., Jackson, D. R., and Nelson, C. E., Patterns of temperature-dependent
sex determination in turtles,/. E xp . Z o o L , 270, 3-15, 1994.
10. Yntema, C. L. and Mrosovsky, N., Critical periods and pivotal temperatures for sexual
differentiation in loggerhead sea turtles. C a n . J . Z o o L , 60, 1012-1016, 1982.
11. Yntema, C. L. and Mrosovsky, N., Sexual differentiation in hatchling loggerheads
{C a re tta c a re tta ) incubated at different controlled temperatures, H e r p e to lo g ic a , 36,
33-36, 1980.
12. Paukstis, G. L. and Janzen, F. J., Sex detemiination in reptiles: summary of effects
of constant temperature of incubation on sex ratios of offspring, S m ith s o n . H e r p e to l.
I n f S erv., 83, 1-28, 1990.
13. Mrosovsky, N., Bass, A., Corliss, L. A,, Richardson, J. I., and Richardson, T. H.,
Pivotal and beach temperatures for hawksbill turtles nesting in Antigua, Ca/i. /. Z o o L ,
70, 1920-1925. 1992.
14. Ashton, W. D., T h e L o g it T r a n sfo rm a tio n , Griffith, London, 1972, 88 pp.
15. Bustard, H. R., S e a T u r tle s, Taplinger Publishing, New York, 1972, 220 pp.
16. Horikoshi, K., Egg Survivorship and Primary Sex Ratio of Green Turtles, C h e lo n ia
m y d a s, at Tortuguero, Costa Rica, Ph.D. thesis. University of Florida, Gainesville,
FL, 1992.
17 Johannes, R. E. and Rimmer, D. W., Some distinguishing characteristics of nesting
beaches of the green turtle C h e lo n ia m y d a s on North west Cape Peninsula, Western
Australia, M ar. B io l., 83, 149-154, 1984.
18 Standora, E. A. and Spotila, J. R., Temperature dependent sex determination in sea
turtles, C o p e ia , 1985, 711-722, 1985.
19. Maloney, J. E., Darian-Smith, C., Takahashi, Y., and Limpus, C. J., The environment
for development of the embryonic loggerhead turtle {C aretta c a re tta ) in Queensland,
C o p e ia , 1990, 378-387, 1990.
20 . Silberstein, D., The physical condition prevailing in nests of the loggerhead sea turtle
C a re tta c a re tta , MS. thesis, Tel Aviv University, Tel Aviv, Israel, 1988.
21. Mrosovsky, N. and Provancha, J., Sex ratios of loggerhead sea turtles hatching on a
Florida beach. C a n . J . Z o o L , 67, 2533-2539, 1989.
22 . Maxwell, J. A., Motara, M. A., and Frank, G. H., A micro-environmental study of
the effect of temperature on the sex ratio of the loggerhead turtle, C a re tta c a re tta ,
from Tongaland, Natal, 5'. A fr. T ydskr. D ie r k ., 23, 342-350, 1988.
23. Soil Survey Staff, Soil Taxonomy, Vol. 436, Soil Conservation Service, U.S. Depart
ment of Agriculture, Washington, D.C., 1975, 109 pp.
24. Campbell, G. S., S o il P h y s ic s w ith B a s ic , Elsevier Scientific, New York, 1985,155 pp.
25. Hillel, D., A p p lic a tio n s o f S o il P h y s ic s . Academic Press, New York, 1985, 385 pp.
26. Horton, R., Canopy shading effects on soil heat and water ^ o w ,S o i l S c i. S o c . A m e r ic a
J ., 53, 669-679.
27. Horton, R. and Chung, S. O., Soil heat flow, in M o d e lin g P la n t a n d S o il S y s te m s ,
Agronomy Monograph no. 31, Hanks, J. and Ritchie, Eds., ASA-CSSA-SSSA, Mad
ison, WI, 1991, 397-438.
28. Horton, R., Aguirre-Luna, O., and Luna, P. J., Observed and predicted two-dimen
sional soil temperature distributions under a row crop. S o il S c i. S o c . A m . J ., 48,
1147-1152, 1984.
29 Ar, A. and Rahn, H., Water in the avian egg: overall budget for incubation,Aw. Z o o L ,
20, 373-384, 1980.
30, Vleck, D., Water economy and solute regulation of reptilian and avian embryos, in
E g g In c u b a tio n : Its E ffe c ts o n E m b r y o n ic D e v e lo p m e n t in B ir d s a n d R e p tile s , Deem
ing, D. C. and Ferguson, M. W. J., Eds., Cambridge University Press, New York,
1991, 245-260.
31 Ackerman, R. A., Seagrave, R. C., Dmi’el, R., and Ar, A., Water and heat exchange
between parchment-shelled reptile eggs and their surroundings, C o p e ia , 1985,
703-711, 1985.
32. Ackerman, R. A., Physical factors affecting the water exchange of buried reptile eggs,
in E g g I n c u b a tio n : Its E ffe c ts o n E m b r y o n ic D e v e lo p m e n t in B ir d s a n d R e p tile s ,
Deeming, D. C. and Ferguson, M. W. J., Eds., Cambridge University Press, New
York, 1991, 193-212.
33. Kam, Y. C. and Ackerman, R. A., The effect of incubation media on the water
exchange of snapping turtle (C h e ly d ra s e r p e n tin a ) eggs and hatchlings, J. C o m p .
P h y s io l. B , 160, 317-324, 1990.
34. Packard, G. C. and Packard M. J., Coupling of physiology of embryonic turtles to
the hydric environment, in R e s p ir a tio n a n d M e ta b o lis m o f E m b r y o n ic V erte b ra tes,
Seymour, R. S., Ed., Dr. W. Junk, The Netherlands, 1984, 99-119.
35. Packard, G. C. and Packard, M. J., The physiological ecology of reptilian eggs and
embryos, in B io lo g y o f th e R e p tilia , Vol. 16, Gans, C. and Huey, R. B., Eds., Alan
R. Liss, New York, 1988.
36. Tracy, C. R., Biophysical modeling in reptilian physiology and ecology, in B io lo g y
o f th e R e p tilia , Vol. 12, Gans, C. and Huey, R., Eds., Academic Press, New York,
1982, 275-321.
37. Bustard, H. R. and Greenham, R, Physical and chemical factors affecting hatching
in the green turtle, C h e lo n ia m y d a s (L.), E c o lo g y , 49, 269-276, 1968.
38. Mortimer, J. A., The influence of beach sand characteristics on the nesting behavior
and clutch survival of green turtles {C h e lo n ia m y d a s ), C o p e ia , 1990, 802-817, 1980.
39. Kraemer, J. E. and Richardson J. I., Volumetric reduction in nest contents of logger-
head sea turtles {C a retta c a re tta ) (Reptilia, Testudines, Cheloniidae) on the Georgia
coast,/. H e rp e to L , 13, 255-260, 1979.
40. Rimkus, T. A., Horton, R., and Ackemian, R. A., The effect of beach renourishment
on the hydric structure and climate of sea turtle nesting beaches along the Atlantic
coast of Florida, R e s to r a tio n E co L , in press.
41. Ackerman, R. A., Dmi’el, R., and Ar, A., Water and heat exchange characteristics of
some parchment-shelled reptile e g g s, P h y s io l. Z o o L , 58, 129-137, 1984.
42. Deeming, D. C. and Thompson, M. B., Gas exchange across the reptilian eggshell,
in E g g In c u b a tio n : Its E ffe c ts o n E m b iy o n ic D e v e lo p m e n t in B ir d s a n d R e p tile s ,
Deeming, D. C. and Ferguson, M. W. J., Eds., Cambridge University Press, New
York, 1991, 277-284.
43. Muth, A,, Water relations of desert iguana (D ip so s a u ru s d o r s a lis ) eggs. P h y s io l. Z o o L ,
54, 441-451, 1981.
44. Corey, A. T. and Klute, A., Application of the potential concept to soil water equi
librium and transport, S o il S c i. S o c. A m . J ., 49, 3-11, 1985.
45. Koorevaar, R, Menelik, G., and Dirksen, C., E le m e n ts o f S o il P h y s ic s , Elsevier, New
York, 1983, 228 pp.
46. Hillel, D., F u n d a m e n ta ls o f S o il P h y s ic s, Academic Press, New York, 1980, 1-413.
47. Pearson, S. D., Thermodynamics of Embryogenesis, Ph.D. thesis, Iowa State Uni
versity, Ames, IA, 1987, 128 pp.
48. Prange, H. D. and Ackerman, R. A., Oxygen consumption and mechanism of gas
exchange of the green turtle {C h e lo n ia m y d a s) egg and hatchlings, C o p e ia , 1974,
758-763, 1974.
49. Ackerman, R. A., The respiratory gas exchange of sea turtle nests {C h elo n ia , C a re tta ),
R esp ir. P h y s io l., 31, 19-38, 1977.
50. Ackerman, R. A. and Prange, H. D., Oxygen diffusion across a sea turtle ( p h e lo n ia
m y d a s) egg shell. C o m p . B io c h e m . P h y s io l., 43A, 905-909, 1972.
51. Clark, R. R., Beach conditions on Florida; a statewide inventory and identification
of the beach erosion problem areas in Florida, B e a c h e s S h o r e s Tech. D e s. M e m o .,
89-1, 1-167, 1989.
52. Ehrhart, L. M. and Raymond, P. W., The Effects of Beach Restoration on Marine
Turtle Nesting in South Brevard County, Florida, Technical Report, U.S. Army Engi
neering District, Jacksonville, FL, 1983.
53. Nelson, D. A. and Dickerson, D. D., Effects of beach nourishment on sea turtles,
Proceedings of 1989 Sea Turtle Conference, Jeckyll Island, GA, 125.
54 . Nelson, D. A. and Dickerson, D. D., Response of Nesting Sea Turtles to Tilling of
Compacted Beaches, Jupiter Island, Florida, Technical Report, U.S. Army Engineer
Waterways Experiment Station, Vicksburg, MS, 1988.
55. Nelson, D. A. and Dickerson, D. D., Effects of beach renourishment on sea turtles,
Proc. Beach Technology Conference 17, Florida Department of Beach and Shores,
Tallahassee, FL, 1988.
56. Nelson, D. A. and Mayes, C. H., St. Lucie Inlet Dredged Material Disposal Effects
on the Firmness of Sand Used by Nesting Turtles, Report 1-24, U.S. Army Engineer
Waterways Experiment Station, Vicksburg, MS, 1986.
57. Nelson, D. A., Mauck, K., and Fletemeyer, J., Physical Effects of Beach Nourishment
on Sea Turtle Nesting, Delray Beach, Florida, Technical Report EL-87-15:1-56, U.S.
Army Engineer Waterways Experiment Station, Vicksburg, MS, 1987.
58. Raymond, P. W., The Effects of Beach Restoration on Marine Turtle Nesting in South
Brevard County, Florida, MS thesis. University of Central Florida, Orlando, FL, 1984.
59. Carter, R. W. G., Coastal Environments, Academic Press, New York, 1988, 617 pp.
60. Nelson, D. A., Measurement of Sea Turtle Nesting Beach Compaction with a Cone
Penetrometer, Technical Report 1-2, U.S. Army Engineer Waterways Experiment
Station, Vicksburg, MS, 1989.
61. Spangler, M. G. and Handy, R. L., Soil Engineering, 4th ed.. Harper & Row, New
York, 1982.
62. Ackerman, R. A. and Seagrave, R. C., Modeling heat and mass exchange of buried
avian eggs, J. Exp. ZooL, 1, 87-97, 1987.
63. Jackson, R. D. and Taylor, S. A., Heat capacity and specific heat, in Methods o f Soil
Analysis, Klute, A., Ed., Soil Science Society of America, Madison, WI, 1986,
954-956.
eggs, Am. ZooL, 20, 575-583, 1980.
65. Rimkus, T. A. and Ackerman, R. A., unpublished data.
C Orientation, Navigation,
and Natal Beach Homing
in Sea Turtles
Kenneth j, Lohmann, Blair E, Witherington,
Catherine M. F. Lohmann, and Michael Salmon
CONTENTS
5.1 Introduction................................................................................................... 108

5.2 Hatchling Emergence and Sea-Finding........................................................ 109
5.2.1 Emergence from the Nest................................................................. 109
5.2.2 Sea-Finding Under Natural Conditions............................................ 110
5.2.2.1 Dependence upon Vision.................................................... 110
5.2.2.2 Light Intensity and Wavelength.......................................... 111
5.2.2.3 Light Direction.................................................................... 114
5.2.2.4 Shape and Form Vision....................................................... 115
5.2.2.5 Light Movement, Periodicity, and Polarization................. 116
5.2.3 Artificial Lighting and Misorientation............................................. 116
5.2.3.1 Characteristics of Artificial Light Fields............................ 116
5.2.3.2 Behavioral Responses......................................................... 116
5.2.3.3 Effects of Moonlight........................................................... 117
5.2.4 Summary of Sea-Finding................................................................. 117
5.3 Hatchling Orientation in the Ocean............................................................. 117
5.3.1 Orientation to Oceanic Waves.......................................................... 117
5.3.1.1 Detecting Waves While Under Water................................. 119
5.3.1.2 Orientation Within and Beyond the Wave
Refraction Zone................................................................. 120
5.3.2 Magnetic Compass Orientation........................................................ 120
5.3.2.1 Acquisition of a Magnetic Directional Preference............ 121
5.3.3 Summary of Orientation Mechanisms During
the Offshore Migration..................................................................... 123
5.4 Oriented Movements in the Open Ocean..................................................... 124
5.4.1 Detection of Magnetic Inclination Angle......................................... 124
5.4.2 Detection of Magnetic Field Intensity............................................. 125
5.5 Natal Beach Homing and Adult Navigation................................................ 125
0-8493-8422-2/97/$0 00+$.50
© 1997 by CRC Press, Inc. 107
5.5.1 The Chemical Imprinting Hypothesis.............................................. 126

5.5.1.1 Chemosensory Cues in Long-Distance Navigation......... 127
5.5.1.2 Chemical Cues in NatalBeach Recognition.......................127
5.5.2 Magnetic Map Hypothesis................................................................ 128
5.6 Concluding Remarks..................................................................................... 130
References................... 130
5.1 INTRODUCTION
The lives of sea turtles consist of a continuous series of migrations. As hatchlings,
the turtles swim from their natal beaches into the open sea,^ "*often taking refuge in
circular current systems (gyres) that serve as moving, open-ocean nursery grounds."^
The juveniles and subadults of many populations reside in coastal feeding areas that
are located hundreds or thousands of kilometers from the beaches on which the
turtles hatched.“’'^ Adult females, however, periodically leave their feeding sites and
migrate to their natal beaches to nest; after nesting, many return to their own specific
feeding areas.^ The itinerant lifestyle that characterizes sea turtles is thus inextricably
linked to an ability to orient and navigate accurately across large expanses of
seemingly featureless ocean.
In some sea turtle populations, migratory performance reaches extremes. The
total distances certain green turtles (Che Ionia my das) and loggerheads (Carena
carena) traverse over the span of their lifetimes exceed tens of thousands of kilo
meters — several times the diameter of the home ocean basin of the turtle.^ Adult
migrations between feeding and nesting habitats can require continuous swimming
for periods of several weeks.^’^^ In addition, the satellite tracks of migrating turtles
often lead straight across the open ocean and directly to the de s t i na t i on. The s e
position plots leave little doubt that adult turtles know where they are going.
The migrations of hatchling turtles are no less impressive. Within seconds of
emerging from their underground nests at night, hatchlings begin to crawl directly
toward an ocean that cannot be seen from the emergence site. The young turtles
enter the sea and immediately establish offshore headings, which are maintained
long after the hatchlings swim beyond sight of land.^’^’^^ Thus, even as hatchlings,
sea turtles are finely tuned migratory machines equipped with an array of mecha
nisms for guiding themselves across large expanses of open ocean.
For logistical reasons, most studies on orientation mechanisms in sea turtles
have focused on hatchlings rather than adults. Adults, after all, are powerful animals
that often exceed one hundred kilograms in weight; they are also difficult to keep
in captivity, and migrate only intermittently. In contrast, hatchlings are small, sea
sonally abundant, easy to manipulate in laboratory and field studies, and strongly
motivated to migrate offshore. In recent years, considerable progress has been made
in characterizing the orientation cues that guide hatchling turtles during their offshore
migration.
In this chapter, we begin by summarizing the orientation mechanisms used by
hatchling turtles, first on land as they move from their nests to the sea, then in the
ocean as they migrate offshore. We then review recent experimental evidence that
Orientation, Navigation, and Natal Beach Homing in Sea Turtles 109
hatchlings can detect magnetic field inclination angle and magnetic field intensity,
two geomagnetic features that vary across the surface of the earth and may provide
sea turtles with information on their global position. We conclude by reviewing
hypotheses of adult navigation and possible mechanisms by which adults might find
and recognize their natal beaches after spending years in distant oceanic regions.
5.2 HATCHLING EMERGENCE AND SEA-FINDING

5.2,1 Emergence from the N est
Sea turtle clutches usually contain between 50 and 170 eggs^^ and develop in nests
buried beneath the sand of oceanic beaches. After the eggs hatch, the hatchlings
must ascend to the beach surface from the underground egg chamber. Although this
upward journey is rarely more than about 1.0 m in distance, it can nevertheless be
difficult, particularly when the sand above the eggs has become compacted or
overgrown with plant roots. The emergence of hatchlings onto the beach is thus the
culmination of a considerable amount of activity that occurs underground after the
turtles hatch.
The most detailed accounts of hatchling behavior within the nest come from
observations on loggerheads^^ and green t u r t l e s . I n these species, most turtles
within a clutch pip (i.e., cut through the eggshell) within a few hours of each other.
After pipping, hatchlings remain quiescent for approximately one d, often with their
heads and front flippers protruding from the egg. During this time yolk is resorbed
into the body cavity and the plastron straightens. In addition, an airspace begins to
form in the upper egg chamber as fluid drains from pipped eggs into the surrounding
sand. This drainage reduces the volume that the eggs occupy and creates sufficient
space in the nest for subsequent hatchling movements.
After the quiescent period, hatchlings extricate themselves from their eggshells
and begin exaggerated crawling m ovem ents.This activity is synchronized among
siblings and occurs episodically in bouts that last several hours. Activity is elicited
by contact with moving siblings and inhibited by high temperatures (as occur during
mi dda y) . Soc i a l l y facilitated upward movement of the hatchling mass occurs as
turtles near the top of the cavity scrape sand from the ceiling of the chamber, while
those below trample it into the floor as they struggle to crawl upward. Thus, the
chamber containing the hatchlings gradually ascends until at last the turtles break
through the surface of the sand.
Hatchlings emerge en masse from the nest 1 to 7 d after pipping ^^^2 imme
diately crawl seaward. Each emergence usually involves between 20 and 120
hatchlings.^^ Emergence marks the beginning of the “hatchling frenzy,” a period of
high activity during which hatchlings enter the sea and swim away from land. Turtles
that pip late or are less active take part in smaller emergence events on subsequent
evenings, or may be unable to escape from the nest at all.^^
Because lower sand temperatures that occur after sunset elicit an increase in
hatchling activity, most hatchling emergences occur at night.^^’^^ Nevertheless, some
early morning emergences^'^ and diurnal emergences during rain storms^'^ also occur.
Among loggerhead nests, emergence activity has a relatively normal temporal
distribution, with a peak near midnight^^ (Figure 5.1). Emergence at night allows
hatchlings to cross the beach at a time when risks of hyperthermia and predation
are relatively low.
LU
2 CO
LU tZ
0 LU
QC >
LU in
LU
TIME
FIGURE 5.1 The timing of 157 loggerhead emergence events from natural nests. An
emergence event was defined as the movement of 10 or more hatchlings from a nest to the
sea. (From Witherington, B. E., Bjorndal, K. A., and McCabe, C. M.,Copeia, 4, 1165, 1990.
With permission.)
5.2.2 S ea-F inding U nder N atural C onditions

5.2.2.1 Dependence upon Vision
After emerging, hatchlings promptly crawl toward the sea. This “sea-finding” behav
ior occurs reliably in daylight or at night and under a wide range of weather
conditions.
Sensory deprivation experiments have demonstrated that the cues hatchlings use
in sea-finding are primarily visual.^^'^^ Hatchlings with their eyes covered and those
screened from a seaward view either crawled in circles or oriented in random
directions on the beach. Although hatchlings may also respond to nonvisual cues
such as beach slope, such environmental features appear to exert little or no influence
on directional movement when visual cues are present.^^’^"^
5.2.2.2 Light Intensity and Wavelength

Identifying light as the principal source of directional infomiation that hatchlings
use during sea-finding only slightly narrows the list of potential cues. Light has
many properties (e.g., intensity, wavelength, directivity, and polarization), any or all
of which might influence hatchling orientation,
“Brightness” is used to describe light intensity and wavelength as a function of
hatchling spectral sensitivity.^^’^^’^^'^^ Hatchlings appear to have a strong tendency
to orient toward the brightest d ire c tio n .“Brightest-direction” orientation may lead
turtles toward the horizon with the greatest portion of visible sky. On natural,
undeveloped beaches, such a horizon is nearly always in the seaward direction.
Characteristics of the brightest-direction response have been investigated by
covering specific parts of hatchlings’ eyes and observing the effect on sea-finding
orientation.^^’^^’^^ Orientation under these conditions is consistent with a complex
phototropotaxis model which proposes that a hatchling turns to maximize the
strength of visual input to multiple comparators in the retina of each eye."'^' As a
consequence, the hatchling orients toward the brightest direction. Because a multiple
sensory an'ay is involved, hatchlings are not only able to recognize an imbalance in
lateral visual input, but are also able to identify the direction they must turn toward
to regain input balance.
The properties of light that influence a hatchling’s assessment of brightness
depend in part upon the differing spectral absorption of photopigments and colored
oil droplets within the retina."^®"^^ Some understanding of the spectral characteristics
of brightness perception has been gained through electroretinography (ERG). For
the green turtle, the only species for which such data have been obtained, the eye
is most sensitive in the violet to yellow (400- to 600-nm) region of the visible
spectrum."^^ Thus, relative to humans, green turtle spectral sensitivity is skewed
toward the short wavelength (blue) end of the spectrum. Because sea water selec
tively attenuates red light more than blue light, these results are not surprising.
Early studies suggested that hatchlings preferred the colors that ERG analyses
indicated they could see best. In initial experiments, hatchlings were exposed to
light that had passed through various broad-band (multiple-wavelength transmission)
filters.^' Although reactions to specific wavelengths could not be assessed, it was
clear that green turtles were more strongly attracted to blue light than to red.
In subsequent studies, narrow-band filters were used in two-choice light prefer
ence experiments, and behavioral responses to specific wavelengths were found to
differ among four species t e s t e d . G r e e n turtle, hawksbill (Eretmochelys imbri-
cata), and olive ridley (Lepidochelys oliváceo) hatchlings were strongly attracted to
light in the near-ultraviolet to yellow region of the spectrum (360 to 600 nm), but
were relatively indifferent to light in the yellow-orange to red region (630 to 700
nm) (Figure 5.2). In contrast, loggerhead hatchlings showed an aversion to light in
the green-yellow to yellow region of the spectrum (560 to 600 nm) (Figure 5.2).
Although this xanthophobia (yellow light aversion) is unique to loggerheads, green
turtles are less strongly attracted to yellow light than to light of other wavelengths."^^
►•LOGGERHEAD -»GREEN TURTLE OLIVE RIDLEY — HAWKSBILL
FIGURE 5.2 Orientation responses to colored light of hatchlings from each of four species
of sea turtles. Turtles were tested in a modified Y-maze that contained two windows, one of
which was illuminated with light of a specific wavelength. Responses were measured as the
proportion of hatchlings (of 30 tested) that chose the illuminated window over an alternative
darkened window. A single quantum radiance at approximately full-moon irradiance (1.58 x
10"^ photons/s/m7nm) was used for the colored source. Aversion indicates that hatchlings
preferred the dark window to the lighted window. Indifference is the statistically random
choice of windows according to a binomial probability test at a = 0.05. (From Witherington,
B. E., Sea-Finding Behavior and the Use of Photic Orientation Cues by Hatchling Sea Turtles,
Ph.D. dissertation. University of Florida, Gainesville, 1992.)
Thus, whereas ERG data indicate that green turtles are highly sensitive to yellow
light,this does not imply that they are strongly attracted to it.
Xanthophobia in loggerheads only occurs at relatively high light intensities and
is probably dependent upon photopic (cone) vision."^^ To assess responses to low-
intensity light, the threshold intensity required to elicit a statistically significant
orientation response in loggerhead hatchlings was determined for several different
wavelengths (Figure 5.3). At lower light levels, turtles oriented toward the light
source at all wavelengths studied, including yellow. Ultraviolet-green light required
the lowest intensity to elicit orientation and was thus the most attractive to the
>-
>
LU
CO
-J
<
OC
5
X
LU
CD
WAVELENGTH (nm)
FIGURE 5.3 Behavioral sensitivity of loggerhead hatchlings to low-intensity spectral light.

Behavioral sensitivity is represented as the inverse of light source radiance required to elicit
directed orientation in groups of individually released hatchlings (n = 30 per group). Thus,
a high behavioral sensitivity at a given wavelength implies that significant orientation toward
the light source was elicited by low light levels. The ordinate is a log scale of the units
(photons/s/m7nm/sr)“h (From Witherington, B. E., Sea-Finding Behavior and the Use of
Photic Orientation Cues by Hatchling Sea Turtles, Ph.D. dissertation. University of Florida,
Gainesville, 1992.)
hatchlings. The threshold intensity for yellow light was about one order of magnitude
higher, and that for red light about five orders of magnitude higher, than the threshold
intensities for the shorter wavelengths.
Such changes in sensitivity and orientation behavior in response to yellow light
may have a proximate explanation based upon loss of color discrimination at low
light intensities. The adaptive significance of these responses in hatchlings may be
to factor out light from bright celestial sources near the horizon. A rising or setting
sun or moon is rich in yellow light, due to scattering effects of the atmosphere of
the earth. A bias against yellow would make these sources less attractive than they
otherwise would be.
According to this hypothesis, the rising sun should have a minimal influence
upon orientation during sea-finding. Indeed, the sea-finding orientation of logger-
head, green turtle, and hawksbill hatchlings appears to be altered little or not at all
when the sun is close to the h o r i z o n . M i n o r influences of the sun on
orientation may be a function not only of color, but also of the way hatchlings
spatially integrate their measurement of brightness (see below).
5.2.23 Light Direction
Sensitivity to directional light can be described by a specific “cone of acceptance”

which indicates how much of the world a hatchling measures at any one instant
(Figure 5.4). An acceptance cone encompasses the angular span over which a
brightness measurement is integrated, and may include only a limited part of a
turtle’s entire field of view. The height and breadth of a hatchling’s acceptance cone
critically influence its determination of brightest direction.
FIGURE 5.4 A hypothetical cone of acceptance for the assessment of brightest direction
by a sea turtle hatchling. The vertical component of the cone (V) is approximately 10 to 3CP
from the horizontal plane. The horizontal component of the cone (H) is approximately 180P.
Light sources within this cone of acceptance are integrated into a measure of brightness for
the direction D.
The horizontal component of the acceptance cone for green turtle,olive ridley,"^^
and loggerhead hatchlings^^ has been inferred from studies of hatchling orientation
in controlled light fields. In these studies, hatchlings were placed within a vertically
striped cylinder in which the direction perceived as brightest depended on the width
of the turtles’ acceptance cones. The results indicated that hatchlings of each species
measured the brightest direction with a wide (about 180°) horizontal acceptance
cone (Figure 5.4). The vertical component was determined by analyzing responses
to light sources positioned at various vertical angles relative to the hatchling. The
vertical component has been estimated to be “a few degrees” for green turtles and
olive ridleys"^^ and between 10° below and 30° above the horizon for loggerheads.
The vertically narrow acceptance cone for hatchlings means that light closest to
the horizon plays the greatest role in determining orientation direction. The integra
tion of brightness over a horizontally broad acceptance cone means that irradiance
(light reaching the hatchling) is a more important consideration than radiance (light
emanating from a source). Integration of light over a broad range may also be
important for mitigating the effects of celestial light source position on orientation
direction; a broad acceptance cone gives hatchlings a larger “sample”, which dimin
ishes the influence of any single point source of light.
5.2.2.4 Shape and Form Vision
Although orientation towards the brightest direction is an important part of hatchling

sea-finding, hatchlings also respond to shape cues. Loggerhead^"^ and green turtle"^^
hatchlings orient away from elevated silhouettes. On most beaches this tendency
directs hatchlings away from dunes and vegetation and toward the ocean. However,
this response may in part be an epiphenomenon of brightest-direction orientation.
Dune silhouettes often darken the portion of the horizon within a hatchling’s cone
of acceptance.
Separating effects of shape from effects of brightness is possible using cue-
conflict e x p e r i m e n t s . B o t h green turtle^® and loggerhead^^ hatchlings showed
a tendency to orient away from sets of alternating black and white stripes, and toward
an opposing “open horizon”, even when both horizons were equally bright. In
loggerheads, orientation away from the stripes persisted even when the striped
horizon was about twice the brightness of the open horizon"^^ (with the angular
radiance measured according to the loggerhead acceptance cone). When the bright
ness of the striped horizon was increased to three times that of the open horizon,
hatchlings failed to orient significantly toward either direction. When the striped
horizon was made five times brighter than the open horizon, hatchlings crawled
toward the striped horizon.
A different series of experiments has suggested that hatchling loggerheads have
preferences for certain shape patterns.^^ For example, hatchlings preferred horizons
with wide vertical stripes to horizons with narrow vertical stripes; they also preferred
either a vertically striped horizon or an open uncluttered horizon to an opposing
checkered horizon. These responses indicate that hatchlings can distinguish between
different shape patterns and may, under at least some conditions, use shape as an
orientation cue.
Whether or not hatchlings use shape cues to locate the sea apparently depends
on what cues are available. In one series of experiments, fomi vision was disrupted
by fitting loggerhead hatchlings with goggles of wax paper,^^ and turtles still oriented
seaward. These results suggested that shape cues were not essential for sea-finding.
However, subsequent experiments were performed at moonset with animals in a
cylindrical arena surrounded by either clear acrylic or wax paper. Under these
conditions, hatchlings with a clear view of their surroundings oriented seaward
whereas those with impaired form vision oriented in the general direction of the
setting moon (the brightest direction).^^ Thus, when both brightness and shape cues
were available, form vision cues apparently took precedence over cues indicating
the brightest direction; hatchlings moved toward the brightest direction, however,
when form vision was disrupted.
On beaches, brightest direction alone can be an unreliable sea-finding cue.
Measurements of brightness that have taken into account appropriate spectral and
acceptance cone considerations indicate that the seaward direction is often, but not
always, the brightest.^^-^^ For example, although the seaward direction is nearly
always the brightest during clear, moonless nights, the brightest direction is not
always seaward when a full moon rises or sets.^^ Nevertheless, hatchlings released
at moonset and moonrise orient seaward without difiiculty^^.
S.2.2.5 Light Movement, Periodicity, and Polarization
As waves approach shore and break, the pattern of light reflected toward the beach
changes rapidly. If hatchlings can distinguish between the shimmering, constantly
shifting patterns of light reflected from waves and the more static light regime of
land, then the movement of light could play a role in guiding turtles seaward. Such
a possibility, however, has not yet been investigated.
The periodicity or flashing pattern of light (again, such as might be observed
from reflections off waves) might also serve as an orientation cue. However, evidence
suggests that green turtles do not distinguish between constant and flashing light
sources except when the off time of the flashing source is very long.^^ In this latter
case, turtles actually prefer a constant light source over the flashing one, a preference
which suggests that flashing patterns are not important in orientation. Moreover, the
avoidance of the slowly flashing source suggests that turtles are merely measuring
relative brightness of the two light sources and are integrating their measure of
brightness over time.
Because light reflected from water is usually polarized, polarized light might
also indicate the seaward direction. In orientation experiments, however, loggerheads
did not respond differently to polarized and unpolarized light.^^
5.2.3 A rtificial L ighting and M isorientation

5.2.3.1 Characteristics of Artificial Light Fields
Celestial light originates from intense, but distant sources. Such light is scattered
by the atmosphere and reflected by the surface of the earth. Natural light fields are
therefore only moderately directed. Artificial light fields, in contrast, are produced
by less intense, but closer sources; thus, they can be highly directed with one
direction much brighter than all others. Because artificial light sources are not intense
enough to brighten the remainder of the observer’s world, artificial light fields are
overwhelmingly dominated by the light sources that generate them.
5.2.3.2 Behavioral Responses

Artificial light fields with high directivity often elicit ‘Tight trapping” responses in
animals.^^ As an orienting animal approaches an artificial source, the light field the
animal perceives becomes less uniform (i.e., more directed) until the source becomes
the only visible feature. The animal is essentially blinded to all else.'’^
Despite their high directivity, artificial light fields do not necessarily have blind
ing characteristics when perceived from a distance. For instance, hatchlings beneath
an artificial light source circle as if blinded, but hatchlings considerable distances
away from a source often crawl directly toward it. On occasion hatchlings may crawl
for hundreds of meters toward distant lighting.^^
Under natural conditions, hatchlings can use multiple cues to find the ocean.
Thus, if a single cue such as brightness does not precisely indicate the seaward
direction, it seldom overrides all other cues, and turtles are still able to locate the
sea. Artificial light sources alter conditions by creating a beach environment in which
extremes of brightness exist. To a hatchling on a dark beach, an artificial light source
may thus become a supernormal stimulus that unambiguously indicates the seaward
direction. At such high levels of stimulation, hatchlings may ignore shape cues and
other features of the beach, or perhaps not even perceive them.
5.2.33 Effects of Moonlight
The common misconception that hatchlings emerge only during the full moon
probably originated because hatchlings are most easily observed on bright, moonlit
nights. In fact, no evidence exists that hatchling emergences are correlated with lunar
phase. The date of emergence is determined by the date eggs were deposited in the
nest and by the length of the incubation period, which depends in part upon tem
perature. Although nesting cycles correlated with moon phase may occur in some
mass nesting (“arribada”) species (e.g., olive ridley)^"^ and in a few loggerhead
populations,^^ variation in incubation time and other factors typically results in
emergences occurring throughout the lunar month.
Moonlight does affect the degree of sea-finding disruption caused by artificial
lighting. Hatchling misorientation in areas with artificial lighting is most frequent
and severe on nights near the new moon.^^ Higher levels of ambient light, as occur
during other moon phases, diminish the directivity of artificial light fields on the
beach. Under such conditions, hatchlings may rely more heavily on shape and
silhouette cues to guide themselves seaward.
5.2.4 Summary of Sea-Finding

Hatchling sea turtles find the ocean using visual cues. Among the features that
influence sea-finding are the brightness and color of the ambient light, and the
presence or absence of shapes and silhouettes. Hatchlings attend primarily to visual
cues within a restricted cone of acceptance along the horizon; thus, higher sources
of light, such as the risen moon, often have relatively little effect on orientation.
Under natural conditions, hatchlings find the sea by crawling toward the brighter,
lower oceanic horizon and away from the elevated silhouettes of vegetation and
dunes that usually border the landward edge of the beach.
5,3 HATCHLING ORIENTATION IN THE OCEAN

5.3.1 O rientation to O ceanic W aves
As hatchlings complete their crawl across the beach and enter the surf, they are
lifted off the sand by incoming waves. As soon as their flippers no longer contact
substrate, hatchlings begin to swim vigorously; simultaneously, they dive.^^ These
responses position turtles in the wave undertow, which quickly transports hatchlings
a short distance (usually about 5 to 10 m) from the beach.
Almost immediately after surfacing, hatchlings establish a course toward the

open ocean^’^’*^ and maintain it as they swim away from shore. Turtles apparently
rely on wave cues to establish offshore headings.'’^ Hatchling loggerheads,^^ green
turtles,^^ and leatherbacks^^^ {Dermochelys coriacea) tethered in floating arenas
placed at various distances offshore all swam into approaching waves. At times when
no waves were present, hatchlings either adopted courses toward seemingly random
directions or swam in circles.
Waves usually approach beaches from almost directly offshore, but storms and
unusual weather conditions occasionally generate waves that temporarily move in
other directions. In one series of experiments, free-swimming (i.e., untethered) green
turtle hatchlings were released at sea either under common conditions when waves
moved shoreward or during unusual weather conditions when waves moved in
atypical directions.^^ The hatchlings swam into waves during all experiments, even
when doing so resulted in orientation back towards land (Figure 5.5).
Direction of Wave Approach
FIGURE 5.5 A summary of experiments in which green turtle hatchlings were released at
sea near the east coast of Florida, U.S.A., on days when waves moved in various directions.
Each data point represents the mean angle of orientation of one group of hatchlings plotted
with respect to the angle of wave approach at the time of release. Hatchlings in each
experiment swam into waves, even when doing so resulted in orientation toward directions
other than offshore. (From Lohmann, K. J. and Lohmann, C. M. F .,i. Exp. Biol., I l l , 1,
1992. With permission.)
Laboratory experiments have also confirmed that loggerhead, green turtle, and
leatherback hatchlings can use waves as a directional cue.^^’^^ When hatchlings were
tethered in a wave tank, they swam into waves when waves were present, but oriented
randomly in still water.
Thus, hatchlings maintain seaward orientation early in the offshore migration

by using wave propagation direction as an orientation cue. Waves and swells are
refracted as they enter shallow, coastal areas; at beaches, they move almost directly
towards shore.Sw im m ing into waves therefore results in movement away from
land and towards the open sea.
5.3.1.1 Detecting Waves While Under Water

The ability of hatchlings to detect wave direction while swimming under water at
night suggests that turtles do not detect waves visually. Experiments in a wave tank
have confirmed that visual cues are not necessary for wave orientation; hatchlings
oriented into waves even in the absence of visible light.^^’^^
One way in which turtles might detect wave direction in darkness is to sense
the orbital movements associated with wave propagation. Objects under water near
the surface of the ocean describe a circular pattern of movement as waves pass
above.^^ In addition, the accelerations that occur in typical wave orbits exceed the
threshold that the vertebrate inner ear can detect.^^’^"^ Turtles might therefore deter
mine wave direction by monitoring the sequence of accelerations they experience
in the water c o l u m n . F o r example, a hatchling facing into approaching waves
is accelerated upward, backward, downward, and then forward with each wave cycle,
whereas a turtle oriented in the direction of wave movement would be accelerated
upward, forward, downward, and then backward (Figure 5.6). The turtle would need
only to distinguish between these two sequences to differentiate orientation against
and with wave propagation direction.
Wave Direction
FIGURE 5.6 The motion of a hatchling turtle swimming with and against the direction of
wave propagation. See text for details. (From Lohmann, K. J., Swartz, A. W., and Lohmann,
C. M. R, J. Exp. Biol., 198, 1079, 1995. With permission.)
To determine whether or not hatchlings might detect wave direction in this way,
turtles were tested on a wave motion simulator that reproduced in air the circular
movements that a hatchling would normally encounter while swimming beneath
small oceanic w aves.Because the swimming behavior of a hatchling is activated

when its ventral surface fails to contact s u b s t r a t e t u r t l e s could be tested while
“swimming” in air on the simulator.^^ Hatchlings subjected to orbital movements
that simulated waves approaching from their right sides attempted to turn right,
whereas movements that simulated waves from the left elicited left-turning behavior.
Movements simulating waves from directly in front of the turtles elicited little turning
in either direction. Thus, hatchlings evidently can determine the propagation direc
tion of waves by monitoring orbital movements.
Additional experiments with the wave simulator revealed that loggerhead
hatchlings respond most strongly to “wave orbits” with amplitudes and periods
closely resembling those of typical waves that occurred near the natal beach in
Florida.^^ Thus hatchlings may be “tuned” to recognize specific wave parameters at
the time they enter the sea.
5.3.1.2 Orientation Within and Beyond the Wave Refraction

Zone
Hatchlings that have just entered the ocean appear to orient exclusively on the basis
of waves; no evidence presently exists for the involvement of other directional cues
at this early stage of the offshore migration. Because waves entering shallow water
refract until they approach the beach directly, orienting into waves leads turtles
seaward. In deeper water farther from land, however, waves no longer provide a
reliable indicator of offshore direction. Evidence suggests, however, that hatchlings
may orient into waves for only a short time after entering the sea. Hatchling log
gerheads tracked from a Florida beach into the open ocean swam into waves initially,
but continued on the same seaward headings after entering offshore areas in which
wave direction no longer coincided with their established courses.*"'^ The ability to
maintain courses seemingly independent of wave direction implies that, after they
have distanced themselves from land, hatchlings use one or more alternative sources
of directional information to guide their movements.
5.3.2 Magnetic C ompass O rientation

Several laboratory experiments have demonstrated that loggerhead^^^’^^ and
leatherback^^ hatchlings can orient to the magnetic field of the earth. Thus, one
possibility is that magnetic compass orientation supplants wave orientation as
hatchlings distance themselves from shore.
Evidence exists for two functionally different types of magnetic compasses in
animals.^^ Inclination compasses do not detect the polarity of the field (i.e., north
vs. south); instead, such compasses define “poleward” as the direction along the
surface of the earth in which the angle formed between the magnetic field vector
and the gravity vector is smallest.^"^ For an animal with an inclination compass,
inverting the vertical component of the field has the same behavioral effect as
reversing the horizontal component.^^ In contrast, animals with polarity compasses
determine north using the polarity of the horizontal field component, and inverting
the vertical component has no effect on orientation.
Loggerhead hatchlings reversed their direction of orientation when the vertical

component of the magnetic field was inverted, but not when the vertical and hori
zontal components were reversed together.^^ These data provide evidence that the
loggerhead compass is based on inclination rather than polarity and is functionally
similar to the magnetic compass of birds.
5.3.2.1 Acquisition of a Magnetic Directional Preference
If the magnetic compass functions in the offshore migration of hatchlings, then

turtles must inherit or acquire a magnetic directional preference that reliably leads
them seaward. During initial experiments with hatchling loggerheads^^>^‘ and
leatherbacks^“ from Florida, hatchlings were permitted to establish a course toward
a dim light in the east (their normal migratory direction) before they were tested in
darkness. These animals subsequently oriented east to northeast in the geomagnetic
field. To determine whether or not the initial course of the turtles influenced their
subsequent magnetic orientation, turtles in one experiment were exposed to light
from either magnetic east or west before being tested in the dark.^^ Hatchlings that
had been exposed to light in the east oriented eastward, whereas those that had been
exposed to light in the west swam approximately westward (Figure 5.7). Reversing
the magnetic field resulted in a corresponding shift in orientation, demonstrating
that the turtles were indeed orienting magnetically in the dark. Another group of
turtles tested in darkness without prior exposure to light cues was not significantly
oriented (Figure 5.7).
East Light West Light No Light
O'* O'*
90 **
FIGURE 5.7 Orientation of hatchlings swimming in complete darkness in the geomagnetic

field after previous exposure to a light in the east (left diagram), a light in the west (middle),
or no light (right). Each data point represents the mean angle of orientation for a single
hatchling; arrows represent the mean angle of the group for the two significant distributions.
Dotted lines represent 95% confidence intervals for the mean. Turtles exposed to the east
light subsequently swam approximately eastward, whereas hatchlings exposed to the west
light swam approximately west. Turtles that were not previously exposed to light were not
significantly oriented. (From Lohmann, K. J. and Lohmann, C. M. F.,7. Exp. Biol., 190, 1,
These results indicate that the position of light cues, or perhaps just the experi
ence of maintaining a course toward a specific direction, can influence subsequent
magnetic orientation behavior. Moreover, because hatchlings without prior light

exposure oriented randomly, the results suggest that turtles do not emerge from their
nests with a preferred magnetic compass bearing, but instead must acquire one.
Although light cues play a critical role in the orientation of hatchlings on the
beach, turtles that enter the ocean under natural conditions initially orient into waves
and appear to ignore visual cues. Thus, if light cues are nomially involved in setting
the preferred magnetic direction, the process may occur during the beach crawl. To
investigate whether hatchlings can set a magnetic course while crawling from their
nests to the ocean, turtles were placed into one end of a short (4.1-m) runway with
a dim light placed in the opposite end, and permitted to crawl toward the light.^^
When a hatchling finished the crawl, the light was extinguished and the turtle was
transferred in darkness to a water-filled orientation arena, where its orientation was
monitored.
Turtles that had crawled towards an east light subsequently swam eastward in
darkness, whereas hatchlings that had crawled west swam west. Reversing the
magnetic field around the swimming hatchlings resulted in a corresponding shift in
orientation, demonstrating that the turtles were orienting to the magnetic field of the
earth. Other hatchlings placed into the runway in complete darkness and permitted
to crawl with no light present were not significantly oriented. These results are
consistent with the hypothesis that turtles emerge from their nests without an estab
lished directional preference; they may, however, acquire one while crawling a short
distance across the beach.
Recent experiments have also demonstrated that hatchlings can establish a mag
netic directional preference on the basis of wave cues.^^ Loggerhead hatchlings that
had never crawled across a beach were tethered inside a wave tank and allowed to
swim into waves for 30 min. The waves were then terminated. Half of the turtles
were allowed to continue swimming in the local geomagnetic field, whereas the
other half were subjected to a field with a reversed vertical component, a treatment
that has the same effect on the turtle magnetic compass as reversing the horizontal
component^ (see above). Hatchlings that swam in the field of the earth continued
to swim in the direction from which waves had previously approached. Turtles tested
in the reversed field, however, swam in the opposite direction. A third group of
turtles that swam without previous exposure to waves was not significantly oriented.
These results provide additional evidence that turtles do not inherit a magnetic
preference for the offshore direction, but instead acquire one that is based on other
directional cues.
Under laboratory conditions, then, hatchlings can establish a magnetic direc
tional preference in at least three different ways: by swimming towards a light source,
by crawling towards a light source, and by swimming into waves. Taken together,
the results suggest that the experience of maintaining a course, either on land or in
water, may be sufficient to set the magnetic compass. Under natural conditions,
hatchlings crawl across the beach toward the brighter, lower horizon of the open
sea, then continue offshore by orienting into approaching waves. Thus, one possible
interpretation of the laboratory experiments is that the seaward course turtles initiate
while crossing the beach and swimming away from land is transferred to the mag
netic compass, so that hatchlings can continue on the same heading after swimming
beyond the wave refraction zone and entering the open ocean.
5.3.3 S ummary of O rientation M echanisms during the

O ffshore M igration
Hatchling loggerheads from the east coast of Florida appear to rely upon three
distinct types of orientation cues during their migration from the natal beach to the
Gulf Stream (Figure 5.8). These cues appear to be used sequentially at different
stages of the migration.
1. visual cues
Beach
FIGURE 5.8 Diagram summarizing the orientation cues hypothesized to guide hatchling
Florida loggerheads from their nests to the Gulf Stream current. The beach is to the left and
progressively deeper water is to the right. Lines represent oceanic waves moving toward the
beach; as they enter shallow coastal areas, waves refract until they approach the beach directly.
Visual cues guide hatchlings from their nests to the sea. Near shore, turtles swim into refracted
waves, which provide a reliable cue for swimming seaward. In deeper water, waves no longer
provide a consistent cue for offshore orientation, and hatchlings are hypothesized to transfer
their seaward courses to their magnetic compasses.
On the beach, hatchlings crawl seaward by orienting toward the brighter, lower,
oceanic horizon. Once in the ocean, turtles initially swim offshore by orienting into
waves, apparently without regard to visual cues. By the time hatchlings swim beyond
the wave refraction zone, they have probably transferred the initial seaward heading
to their magnetic compasses, enabling them to maintain offshore courses long after
swimming beyond sight of land. Because detailed studies on orientation mechanisms
have been carried out only with loggerhead hatchlings from the east coast of Florida,
whether or not the mechanisms outlined above are generally applicable to other
populations and species is not known.
5.4 ORIENTED MOVEMENTS IN THE OPEN OCEAN

The migration of hatchling Florida loggerheads from their natal beach to the open
ocean is just the first step in a much longer transoceanic journey. Young loggerheads
evidently remain in the North Atlantic gyre for several years, during which time
they cross to the eastern side of the Atlantic Ocean before returning to the south
eastern U.S. coast as subadults.'^'^^’^^ Analyses of mitochondrial DNA from nesting
loggerheads has provided evidence that adult females eventually return to nest on
or near the same beaches where they themselves emerged as hatchlings.^’^*^^
Hatchling and juvenile loggerheads in the open sea may benefit from oriented
movements that serve to keep them within oceanic regions favorable for growth and
development. For example, whereas the warm waters of the Gulf Stream provide a
suitable environment for young turtles, straying beyond the latitudinal extremes of
the North Atlantic gyre can be fatal. As the northern edge of the gyre approaches
Portugal, the east-flowing current divides. The northern branch continues past Great
Britain and the water temperature decreases rapidly. Loggerheads swept north in
this current soon die from the cold."^’^^ Similarly, turtles that venture south of the
gyre risk being swept into the South Atlantic Current system and carried far from
their normal range. An ability to recognize the latitudinal extremes of the gyre, and
to respond by orienting in an appropriate direction, might therefore have considerable
adaptive value.
5.4.1 D etection of Magnetic I nclination A ngle

Several features of the magnetic field of the earth vary in a predictable way across
the surface of the earth.^^ Thus, the ability to detect the geomagnetic field might
assist turtles in remaining within the North Atlantic gyre, if turtles can detect
parameters of the field that vary latitudinally.
One geomagnetic feature strongly correlated with latitude is field line inclina
tion.^^ At each position on the globe, magnetic field lines intersect the surface of the
earth at a specific angle ranging from 0° (parallel to the earth) at the geomagnetic
equator to 90° at the magnetic poles. Thus, an animal able to distinguish between
different inclination angles might be able to approximate its latitude.
To determine if loggerheads can distinguish between different magnetic incli
nation angles, hatchlings were tethered in an arena sun*ounded by a computerized
coil system that could be used to generate earth-strength fields with different incli
nations.^^ Hatchlings exposed to a field with an inclination angle found on the
northern boundary of the North Atlantic gyre swam south-southwest (Figure 5.9).
In contrast, hatchlings exposed to an inclination angle found near the southern
boundary of the gyre swam in a northeasterly direction (Figure 5.9). Turtles exposed
to inclination angles they do not normally encounter (i.e., from north or south of
the North Atlantic gyre), or to a field inclination found well within the northern and
southern extremes of the gyre, were not significantly oriented. These results dem
onstrate that loggerheads can distinguish between different magnetic inclination
angles and perhaps derive from them an approximation of latitude.
60 ° 30°
0° 0®
90" 90"
FIGURE 5.9 Orientation of hatchlings tested in magnetic fields of the same intensity, but
different inclinations. Conventions as in Figure 5.7. Turtles exposed to a 6CF inclination angle
(found near the northern edge of the North Atlantic gyre) were significantly oriented toward
the south-southwest, whereas those exposed to a 3(T inclination angle (found near the southern
border of the gyre) swam in a northeasterly direction. (From Lohmann, K. J. and Lohmann,
C. M. F., J. Exp. Biol., 194, 23, 1994. With pemiission.)
5.4.2 D etection of Magnetic F ield I ntensity

A second geomagnetic feature that varies across the surface of the earth is field
intensity (strength). Recent experiments have demonstrated that loggerhead
hatchlings can perceive differences in intensity that they encounter along their
migratory route.
In the same apparatus used in the inclination angle study,hatchlings were
exposed to one of two field intensities that they normally encounter during their first
months in the sea. In all trials, the inclination angle of the field was kept constant.
Turtles tested in a field of 52,000 nT (a field 10.6% stronger than the natal beach
field, and one that hatchlings first encounter near South and North Carolina, U.S.)
swam eastward. Those exposed to a 43,000-nT field (a field 8.5% weaker than the
natal beach field, and one first encountered on the eastern side of the Atlantic near
Portugal) swam westward.
These results demonstrate that hatchlings can distinguish between field intensi
ties that occur in different locations along their migratory route. Moreover, because
eastward orientation near South Carolina and westward orientation near the coast
of Portugal would both function to keep hatchlings within the confines of the gyre,
the results are consistent with the hypothesis that turtles can use field intensity to
assess their global position.^^
5.5 NATAL BEACH HOMING AND ADULT

NAVIGATION
The natal beach homing hypothesis proposes that adult turtles return to nest near
the same sites where they themselves emerged as hatchlings. Until recently, little
direct evidence existed to support this hypothesis. Genetic analyses, however, have
now confirmed that such return migrations do indeed occur in at least several sea
turtle populations.^’’^’^^^’^'^
An ability to return to a specific nesting beach from hundreds or thousands of

kilometers away, and after years in distant oceanic or coastal habitats, is common
among sea turtles. Green turtles that nest on tiny Ascension Island, for example,
regularly migrate between their nesting beach and Brazilian feeding grounds, a
straight-line distance of more than 2000 km.^’^‘ Most Kemp’s ridley turtles (L. kempi)
throughout the Atlantic, Caribbean, and Gulf of Mexico converge from hundreds or
thousands of kilometers away to nest along a single, isolated beach in Mexico.^’^^
And loggerheads that nest in Japan apparently traverse the entire Pacific Ocean to
Baja California before returning to their natal beaches to nest.^
Sea turtles can also pinpoint small, isolated sites that are not nesting beaches.
Loggerheads and green turtles that nest along the Great Barrier Reef, for example,
routinely return to specific, widely dispersed feeding grounds that are sometimes
located hundreds or thousands of kilometers away from their nesting sites.^ Although
the feeding grounds of different individuals are often separated by hundreds of
kilometers and may lie in nearly any direction from the nesting beach, adult turtles
show great fidelity to both their feeding and nesting areas, and migrate between
them at appropriate times. Thus, the picture that emerges is that no “typical” migra
tory pathway exists for the adults of a given population. Instead, turtles converge
on nesting beaches from widely dispersed feeding grounds, and each individual
migrates back to its own feeding site after nesting is completed. Such precise
targeting of specific destinations over immense distances is difficult to explain
without hypothesizing an ability to determine geographic position relative to the
g O a l.^ 2 ’^4’^5,86
Little is known about the mechanisms that underlie long-distance navigation in

adult turtles. Adults are large, powerful animals that make poor subjects for labora
tory experiments. Until recently, tracking adults in the ocean was an arduous, labor-
intensive undertaking that usually provided relatively little information.^^ A prom
ising development is the successful use of satellite tracking technology to reconstruct
the paths of turtles migrating over long d i s t a n c e s . S u c h studies may at last
permit experimental manipulations of adults migrating under natural conditions, and
allow for direct tests of navigational theories.
For now, what little is known about adult navigation has been inferred from
studies of hatchlings and juveniles, and from analyses of satellite tracking experi
ments. Here we briefly summarize major hypotheses of sea turtle navigation and
natal beach homing.
5.5.1 T he C hemical Imprinting H ypothesis

The discovery that salmon use olfactory cues to recognize the tributaries in which
they hatched^^’^^ led to speculation that turtles might recognize their natal beaches
by detecting chemical features unique to each nesting area. The chemical imprinting
hypothesis for sea turtles proposes that hatchling turtles imprint upon chemical cues
unique to their natal beach and use this information years later as adults to return
to that same beach for nesting and mating.^’^^ Relatively little evidence, however,
has yet been obtained to either support or refute this hypothesis. Conditioning
experiments have revealed that green turtles can indeed detect small concentrations
of chemicals in water,"^*’^^ but the role that chemical cues play in long-distance
navigation and natal beach recognition remains almost entirely unknown.
5.5.1.1 Chemosensory Cues in Long-Distance Navigation

The chemical imprinting hypothesis fueled speculation that adult turtles might locate
their natal beaches by homing in on the source of a particular odor plume from
considerable distances away. In the case of the Ascension Island migration, for
example, adult green turtles in Brazilian waters were hypothesized to detect a plume
that originated at the island and then follow it across more than 2000 km of ocean
to its source.'^^ The chemical stimulus was thought to be carried to Brazil by the
west-flowing Southern Equatorial Current.
As oceanic cuiTents and the migratory paths of turtles have become better
characterized, however, evidence against the use of chemical cues in long-distance
navigation has begun to accumulate. In the case of Ascension, oceanographic anal
yses have revealed that the island falls within the narrow South Equatorial Counter-
current, which flows eastward rather than westward during the season when the
turtles undertake their nesting m igrations.Thus, any odor plume originating at
Ascension would presumably flow toward Africa instead of Brazil, and could thus
not serve as a guide for turtles approaching the island from the west.
Similar considerations at other nesting sites also argue against the hypothesis
that turtles navigate to their natal beaches by following odor plumes over great
distances. Green turtles that nest at Tortuguero, Costa Rica, for example, converge
on their nesting beaches from feeding grounds that are both upcurrent and down-
current;^ the same appears to be true for Kemp’s ridleys that nest at Rancho Nuevo,
Mexico,'^^ loggerheads that nest in Florida,^^ and green turtles and loggerheads that
nest along the Great Barrier Reef.^
Recent satellite tracking experiments have also revealed that migrating adult
green turtles often follow essentially straight paths to target sites from hundreds of
kilometers away, even when moving perpendicularly to water c ur r e nt s . Ani ma l s
orienting in a chemical plume are usually unable to move directly to the source
without employing a search strategy that involves frequent course changes.
The consistency of the headings observed in satellite tracking experiments thus
suggests that turtles do not rely upon chemical cues during long-distance naviga
tion.^“'"^
5.5.1.2 Chemical Cues in Natal Beach Recognition
Even if chemical cues play no role in long-distance navigation, they may still enable
nesting turtles to recognize their natal beaches after other navigational mechanisms
have brought the turtles into close proximity. Some limited evidence exists to support
this hypothesis, but the results are not yet fully conclusive.
In one study,‘^'^ Kemp’s ridley turtle eggs from Rancho Nuevo, Mexico were
incubated in sand from Padre Island, TX. After emerging from their eggs, hatchlings
were permitted to crawl across the Padre Island beach and swim through the surf,
whereupon they were recaptured and held in captivity for four months. These juvenile
turtles were then tested in a water-filled arena which contained four compartments
that the turtles could enter. One compartment contained a solution made from Padre
Island sand and seawater, one contained a similar solution made from sand and sea
water from a different location (Galveston, TX), and two contained untreated sea
water. The time that turtles spent in each compartment after entering was monitored.
Results indicated that turtles spent significantly more time per entry in the compart
ment containing Padre Island water than they spent in any of the other compartments.
Although these results are suggestive, all turtles tested in this initial study had
been “imprinted” to Padre Island sand and water; no attempt was made to test the
responses of turtles that had been similarly “imprinted” to water from another
location. Thus, the possibility that the turtles simply preferred Padre Island water to
the alternatives for reasons unrelated to early experience could not be excluded. A
second experiment was therefore carried out to determine whether turtles
“imprinted” to either Padre Island or Rancho Nuevo preferred water from their
respective natal b e a c h e s . N o such preferences could be discerned, but poor health
of the turtles may have adversely affected their performance.
In a subsequent experiment, green turtle eggs were incubated in sand that was
scented with one of two chemicals (morpholine or 2-phenylethanol). After hatching,
each turtle was held for an additional three months in water containing the same
chemical to which the turtle had been exposed as an embryo. After two additional
months without exposure to either chemical, the turtles were tested in the monitoring
tank, which this time contained solutions of morpholine, 2~phenylethanol, and
untreated sea water. Turtles that had previously been exposed to morpholine preferred
morpholine to 2-phenylethanol, whereas the opposite was true for turtles that had
been exposed to the 2-phenylethanol. Interestingly, however, additional groups of
turtles that had been exposed to the chemicals only while in the nest (for about 2
months) or only after emerging from the nest (i.e., while living in the water of the
holding tank for 3 months) failed to manifest these preferences.
The results demonstrate that, during certain developmental periods, turtles can
acquire a preference for specific chemicals and retain it for at least two months. The
significance of this finding, however, is not yet clear. Only turtles exposed to the
chemicals for a total of about 5 months (approximately 2 months in the nest, followed
by 3 months in water) acquired the preference. Under natural conditions, hatchlings
migrate beyond the waters of their natal beach within a few hours after leaving their
nests.^’^’^^ Thus, if turtles require several months of exposure to natal beach water
after hatching (in addition to exposure as embryos) for chemical imprinting to occur,
it is difficult to envision how such a process could occur under natural conditions.
Nevertheless, the initial results are intriguing, and additional studies are clearly
needed.
5.5.2 Magnetic Map H ypothesis

The findings that adult turtles can swim directly to specific destinations from hun
dreds of kilometers away^^'^"^ and can return to nesting sites following forced
displacements^^ ™ply that turtles can determine their position relative to a goal.^"^’^^’^^
Animals with such an ability are said to have a map sense.Although the nature of
the sea turtle map remains unknown, a new hypothesis proposes that turtles determine
position using geomagnetic parameters.
That sea turtles can derive at least some positional information from features of
the magnetic field of the earth now seems c l e a r . A t a minimum, such information
appears likely to provide an estimate of latitude; in the case of Florida loggerheads,
it may assist young turtles in remaining within the borders of the North Atlantic
gyre.i>82 ability of turtles to detect both field inclination and intensity, however,
raises the possibility that adults use a bicoordinate magnetic map to pinpoint their
global position.
The migration of Ascension Island green turtles provides one intriguing example
of how such a bicoordinate magnetic map might permit navigation over a large
oceanic region. Isoclinics (lines of equal inclination angle) and isodynamics (lines
of equal field intensity) form a nonorthogonal grid between South America and
Africa^^ (Figure 5.10). Thus, all locations between feeding grounds in Brazil and
nesting beaches at Ascension Island are defined by unique combinations of inclina
tion and intensity. A migrating turtle using a bicoordinate magnetic map based on
these two parameters could therefore determine its position anywhere along its route,
provided it had learned the magnetic features of Ascension and the gradients of field
inclination and intensity in the South Atlantic.
FIGURE 5.10 Isoclinics (dotted lines) and isodynamics (solid lines) in the oceanic region
surrounding Ascension Island. Adjacent isoclinics differ by 2° and adjacent isodynamics by
1000 iiT. The two geomagnetic features form a nonorthogonal grid that might, in principle,
provide Ascension Island turtles with a bicoordinate position-finding system as they migrate
between Ascension and the Brazilian coast. (From Lohmann, K. J. and Lohmann, C. M. F.,
Nature, 380, 59, 1996, With permission.)
if a magnetic map is used at all by turtles, it might range from a system that
provides only a crude estimate of global position to a highly developed sensory
system capable of pinpointing a specific beach. The resolution would presumably
depend on both the uniformity of the magnetic gradients over a given oceanic region
and on the sensitivity of the turtles to the two parameters. Although nothing is yet
known about the resolution with which turtles can detect specific magnetic features,
it is interesting to note that bees can detect magnetic anomalies in which the field
intensity differs from that of the local field of the earth by as little as 26 A
sensitivity of approximately 10 to 30 nT has been hypothesized for b i r d s . I f
turtles have similar abilities to discriminate intensity and can also detect inclination
angle with great precision, a navigational system capable of pinpointing specific
islands and beaches is at least theoretically possible. Alternatively, however, a mag
netic map might function only to guide turtles into the general vicinity of a nesting
beach, close enough for other cues (for example, chemical or visual) to take over
and allow final localization of the nesting site.
Although such speculation is inviting, whether or not turtles do indeed use a
bicoordinate magnetic map sense during long-distance migrations remains to be
determined. At present, the open-ocean navigational abilities of sea turtles remain
an enduring mystery of behavioral biology.
5.6 CONCLUDING REMARKS

We are now at the threshold of understanding how young sea turtles can travel
reliably from nesting beaches to nursery areas and how they can maintain position
within oceanic regions favorable for development. For loggerhead hatchlings, and
perhaps other species as well, orientation during the offshore migration appears to
depend on the sequential use of visual, wave, and magnetic cues. After young turtles
reach their nursery habitat, the magnetic field of the earth may be used not only in
orientation, but also in position-finding, enabling young turtles to stay within the
boundaries of gyres or other favorable areas.
As yet we can only speculate on the mechanisms that adult turtles use to navigate.
Nevertheless, research on hatchling orientation has provided a first glimpse into the
sensory information that older turtles may have at their disposal. The discovery that
young turtles can detect both magnetic field inclination and intensity gives credence
to the hypothesis that adults can integrate such parameters into bicoordinate magnetic
maps that allow precise journeying between nesting beaches and foraging sites.
Perhaps such a geomagnetic map sense will ultimately prove to be the key to the
extraordinary and seemingly inexplicable mystery of how sea turtles navigate.
REFERENCES
1. Lohmann, K. J. and Lohmann, C. M. F., Orientation and open-sea navigation in sea
turtles, J. Exp. Biol., 199, 73, 1996.
2. Frick, J., Orientation and behaviour of hatchling green sea turtles (Chelonia mydas)
in the sQa.Anim. Behav.. 24. 849. 1976.
3. Salmon, M. and Wyneken, J., Orientation and swimming behavior of hatchling log
gerhead turtles (Caretta caretta L.) during their offshore migration, /. Exp. Mar. Biol.
Ecol., 109, 137, 1987.
Carr, A., Rips, FADS, and little loggerheads, B/eSdc/icc, 36, 92, 1986.
Bowen, B. W., Avise, J. C., Richardson, J. I., Meylan, A. B., Margaritoulis, D., and
Hopkins-Murphy, S. R., Population structure of loggerhead turtles (Caretta caretta)
in the northwestern Atlantic Ocean and Mediterranean Sea, Conserx. Biol., 7, 834,
1993.
6 . Bowen, B. W., Abreu-Grobois, F. A., Balazs, G. H., Kamezaki, N., Limpus, C. J.,
and Ferl, R. J., Trans-Pacific migrations of the loggerhead turtle (Caretta caretta)
demonstrated with mitochondrial DNA markers, Proc. Natl. Acad. Sci., 92, 3731,
1995.
7. Sears, C. J., Bowen, B. W., Chapman, R. W., Galloway, S. B., Hopkins-Murphy,
S. R., and Woodley, C. M., Demographic composition of the feeding population of
juvenile loggerhead sea turtles (Caretta caretta) off Charleston, South Carolina:
evidence from mitochondrial DNA markers. Mar. Biol., 123, 869, 1995.
Limpus, C. J., Miller, J. D., Parmenter, C. J., Reimer, D., McLachland, N., and Webb,
R., Migration of green (Chelonia my das) and loggerhead (Caretta caretta) turtles to
9. Carr, A., The Sea Turtle: So Excellent a Eishe, University of Texas Press, Austin, 1986.
10. Mortimer, J. A, and Portier, K. M., Reproductive homing and intemesting behavior
of the green turtle (Chelonia mydas) at Ascension Island, South Atlantic Ocean,
Copeia, 4, 962, 1989.
11. Carr, A., The Ascension Island green turtle colony, Copeia, 3, 547, 1975.
12. Papi, F, Liew, H. C., Luschi, P, and Chan, E. H., Long-range migratory travel of a
green turtle tracked by satellite: evidence for navigational ability in the open sea.
Mar. B iol, 122, 171, 1995.
13. Balazs, G. H., Craig, R, Winton, B. R., and Miya, R. K., Satellite telemetry of green
turtles nesting at French Frigate Shoals, Hawaii, and Rose Atoll, American Samoa,
NOAA Tech. Memo., NMFS-SEFSC-351, 184, 1994.
14. Papi, F. and Luschi, R, Pinpointing Msla Meta’: the case of sea turtles and albatrosses,
J. Exp. B iol, 199, 65, 1996.
15. Ireland, L. C., Frick, J. A., and Wingate, D. B., Nighttime orientation of hatchling
green turtles (Chelonia mydas) in open ocean, in Animal Migration, Navigation, and
Homing, Schmidt-Koenig, K. and Keeton, W. T., Eds., Springer-Verlag, New York,
1978, 420.
16. Ehrhart, L. M., A review of sea turtle reproduction, in Biology and Conservation o f
Sea Turtles, Bjomdal, K. A., Ed., Smithsonian Institution Press, Washington, D.C.,
1982, 29.
17. Demmer, R. J., The Hatching and Emergence of Loggerhead TnxiXt (Caretta caretta)
Hatchlings, Masters thesis. University of Central Florida, Orlando, 1981.
18. Bustard, H. R., Sea Turtles. Natural History and Conservation, Collins, London,
1972.
19. Carr, A. and Hirth, H., Social facilitation in green turtle siblings, An/m. Behav., 9,
68, 1961.
20. Carr, A. and Ogren, L., The ecology and migrations of sea turtles. 4. The green turtle
in the Caribbean Sea, Am. Mus. Nat. Hist. B ull, 121, 1, 1960.
21. Mrosovsky, N., Nocturnal emergence of sea turtles: control by thermal inhibition of
activity, 220, 1338, 1968.
22. Christens, E., Nest emergence lag in loggerhead sea turtles, /. Herpetol., 24, 400,
1990.
23. Witherington, B. E., unpublished data, 1992.
24. Witherington, B. E., Human and Natural Causes of Marine Turtle Clutch and
Hatchling Mortality and Their Relationship to Hatchling Production on an Important
Florida Nesting Beach, Master’s thesis. University of Central Florida, Orlando, 1986.
26. Bustard, H. R., Mechanism of nocturnal emergence from the nest in green turtle
hatchlings. Nature, 214, 317, 1967.
27. Neville, A., Webster, W. D., Gouveia, J. E, Hendricks, E. L., Hendricks, I., Marvin,
G., and Marvin, W. H., The effects of nest temperature on hatchling emergence in
the loggerhead sea turtle {Caretta caretta), NOAA Tech. Memo. NMFS-SEFC-214,
71, 1988.
28. Witherington, B. E., Bjomdal, K. A., and McCabe, C. M., Temporal pattern of
nocturnal emergence of loggerhead turtle hatchlings from natural nests, Copeia, 4,
1165, 1990.
29. Chavez, H. M., Contreras, G., and Hernandez, T. P. E., On the coast of Tamaulipas,
Part Two, Int. Turtle Tortoise Soc. J., 2, 27, 1968.
30. Mrosovsky, N. and Shettleworth, S. J., On the orientation circle of the leatherback
turtle, Dermochelys coriacea, Anim. Behav., 23, 568, 1975.
31. Mrosovsky, N. and Shettleworth, S. J., Wavelength preferences and brightness cues
in the water-finding behaviour of sea turtles. Behaviour, 32, 211, 1968.
32 Mrosovsky, N., The water-finding ability of sea turtles, 5m m Behav. EvoL, 5, 202,
1972.
33. van Rhijn, F. A., Optic orientation in hatchlings of the sea turtle, Chelonia mydas. I.
Brightness: not the only optic cue in sea-finding orientation, Mar. Behav. Physiol.,
6, 105, 1979.
34 Salmon, M., Wyneken, J., Fritz, E., and Lucas, M., Seafinding by hatchling sea turtles:
role of brightness, silhouette and beach slope as orientation cues. Behaviour, 122,
56, 1992.
35. Ehrenfeld, D. W. and Carr, A., The role of vision in the sea-finding orientation of the
green turtle {Chelonia mydas), Anim. Behav., 15, 25, 1967.
36. Mrosovsky, N. and Kingsmill, S. E, How turtles find the sea, Z. TierpsychoL, 67,
237, 1985.
37. Witherington, B. E., Sea-Finding Behavior and the Use of Photic Orientation Cues
by Hatchling Sea Turtles, Ph.D. dissertation. University of Florida, Gainesville, 1992.
38. Mrosovsky, N. and Shettleworth, S. J., Further studies on the sea-finding mechanism
in green turtle hutchlings. Behaviour, 51, 195, 1974.
39. Mrosovsky, N., Granda, A. M., and Hay, T., Seaward orientation of hatchling turtles:
turning systems in the optic tectum. Brain Behav. EvoL, 16, 203, 1979.
40. Granda, A. M. and Haden, K. W., Retinal oil globule counts and distribution in two
species of turtles: Pseudemys scripta elegans (Wied) and Chelonia mydas mydas
(Linnaeus), Vision Res., 1079, 79, 1970.
41. Liebman, P. A. and Granda, A. M., Microspectrophotometric measurements of visual
pigments in two species of turtle, Pseudemys scripta and Chelonia mydas, Vision
Res., 11, 105, 1971.
42 , Granda, A. M. and Dvorak, C. A., Vision in turtles, \n Handbook o f Sensory Physi
ology, Vol. VII/5, The Visual System in Vertebrates, Crescitelli, E, Ed., Springer-
Verlag, Berlin, 1977, 451.
43. Granda, A. M. and O’Shea, P. J., Spectral sensitivity of the green turtle (phelonia
mydas mydas) determined by electrical responses to heterochromatic light, Brain
Behav. EvoL, 5, 143, 1972.
44. Mrosovsky, N. and Carr, A., Preference for light of short wavelengths in hatchling
green sea turtles, Chelonia mydas, tested on their natural nesting beaches, Behaviour,
28, 217, 1967.
45. Witherington, B. E. and Bjorndal, K. A., Influences of wavelength and intensity on
hatchling sea turtle phototaxis: implications for sea-finding behavior, CopczV/, 4,1060,
1991.
46. Parker, G. H., The crawling of young loggerhead turtles toward the sea,/. Exp. ZooL,
36, 323, 1922.
47. Mrosovsky, N., The influence of the sun’s position and elevated cues on the orientation
of hatchling sea turtles, An/m. Behav., 18, 648, 1970.
48. Verheijen, F. J. and Wildschut, J. T., The photic orientaion of sea turtles during water
finding behaviour, Neth. J. Sea Res., 1, 53, 1973.
49. Limpus, C., Sea turtle ocean finding behaviour, S'cdirc/z, 2, 385, 1971.
50. van Rhijn, F. A. and van Gorkom, J. C., Optic orientation in hatchlings of the sea
turtle, Chelonia mydas. III. Sea-finding behaviour; the role of photic and visual
orientation in animals walking on the spot under laboratory conditions. Mar. Behav.
Physiol, 9, 211, 1983.
Witherington, B. E., Bjorndal, K. A., and Bolten, A. B., An evaluation of the use of
dune structure to reduce effects of artificial lighting on hatchling sea turtle sea-finding
and offshore orientation, unpublished final report to the U.S. Air Force, Cape Canav
eral Air Force Station, FL, 1994.
52. Mrosovsky, N., Effects of flashing lights on sea-finding behavior of green turtles,
Behav. B iol, 22, 85, 1978.
53. Verheijen, F. J., The mechanisms of the trapping effect of artificial light sources upon
animals, Arc/z. N eerl Zool, 13, 1, 1958.
54. Cornelius, S. E., The Sea Turtles o f Santa Rosa National Park, Fundación de Parques
Nacionales Costa Rica, Costa Rica, 1986.
55. Burney, C. M., Mattison, C., and Fisher, L., The relationship of loggerhead nesting
patterns and moon phase in Broward County, NOAA Tech. Memo. NMFS-SEFC-
278, 161, 1991.
hatchlings: evidence for lunar modulation, Copeia, 4, 931, 1995.
57. Salmon, M. and Lohmann, K. J., Orientation cues used by hatchling loggerhead sea
turtles (Caretta caretta) during their offshore migration. Ethology, 83, 215, 1989.
58. Lohmann, K. J., Salmon, M., and Wyneken, J., Functional autonomy of land and sea
orientation systems in sea turtle hatchlings, B ull, 179, 214, 1990.
59. Lohmann, K. J. and Lohmann, C. M. F, Orientation to oceanic waves by green turtle
hatchlings, J. Exp. B iol, 171, 1, 1992.
60, Lohmann, K. J., How sea turtles navigate, Am., 266, 100, 1992.
61. Wyneken, J., Salmon, M., and Lohmann, K. J., Orientation by hatchling loggerhead
sea turtles Caretta caretta L. in a wave tank, J. Exp. Mar. Biol. Ecol, 139, 43, 1990.
62. Bascom, W., Waves and Beaches, Anchor Press Doubleday, New York, 1980.
63. Denny, M. W., Biology and the Mechanics o f the Wave-Swept Environment, Princeton
University Press, Princeton, NJ, 1988.
64. Cook, P. H., Directional infonnation from surface swell: some possibilities, in Mech
anisms of Migration in Fishes, McLeave, J. D., Arnold, G. P., Dodson, J. J., and
Neill. W. H.. Eds., Plenum Press, New York, 1984, 79.
65. Lohmann, K. J., Swartz, A. W., and Lohmann, C. M. R, Perception of ocean wave
direction by sea turtles, 7. Exp. B iol, 198, 1079, 1995.
66. Carr, A., Notes on the behavioral ecology of sea turtles, mBiology and Conservation
o f Sea Turtles, Bjorndal, K. A., Ed., Smithsonian Institution Press, Washington, D.C.,
1981, 19.
67. CaiT, A., Orientation problems in the high seas travel and terrestrial movements of
marine turtles, in Bio-Telemetry, Slater, L. E., Ed., MacMillan, New York, 1963, 179.
68. Manning, E. L., Cate, H. S., and Lohmann, K. J., Discrimination of ocean wave
features by hatchling sea turtles, Mar. B iol, in press.
69. Witherington, B. E., Observations of hatchling loggerhead turtles during the first few
days of the lost year(s), NOAA Tech. Mem. NMFS-SEFC-361, 154, 1995.
70. Lohmann, K. J., Magnetic orientation by hatchling loggerhead sea turtles (Caretta
caretta), J. Exp. Biol., 155, 37, 1991.
71. Light, P., Salmon, M., and Lohmann, K. J., Geomagnetic orientation of loggerhead
sea turtles: evidence for an inclination compass,/. Exp. Biol., 182, I, 1993.
72. Lohmann, K. J. and Lohmann, C. M. F., A light-independent magnetic compass in
the leatherback sea turtle, Biol. Bull., 185, 149, 1993.
73. Wiltschko, W. and Wiltschko, R., Magnetic orientation in birds, inCurrent Ornithol
ogy, Vol. 5, Johnston, R. R, Ed., Plenum Press, New York, 1988, 67.
74. Wiltschko, W. and Wiltschko, R., Magnetic compass of European robins. Science,
176, 62, 1972.
75. Lohmann, K. J. and Lohmann, C. M. R, Acquisition of magnetic directional prefer
ence in hatchling loggerhead sea turtles,/. Exp. Biol., 190, 1, 1994.
76. Lohmann, K. J., Lohmann, C. M. R, and Callaway, J. C., Evidence that beach crawl
direction sets the magnetic compass in loggerhead hatchlings, NOAA Tech. Memo.
NMFS-SEFC-361, 71, 1995.
77. Goff, M. D., Salmon, M., and Lohmann, K. J., The magnetic compass of loggerhead
sea turtle hatchlings: calibration by surface waves, NOAA Tech. Mem. NMFS-
SEFSC, in press.
78. Carr, A., New perspectives on the pelagic stage of sea turtle development, Conserv.
B iol, 1, 103, 1987.
79. Bolten, A. B., Martins, Ft. R., Bjorndal, K. A., and Gordon, J., Size distribution of
pelagic-stage loggerhead sea turtles (Caretta caretta) in the waters around the Azores
and Madeira, Arquipelago Life Mar. Sci., 11 A, 49, 1994.
80. Bowen, B. W., Kamezaki, N., Limpus, C. J., Hughes, G. R., Meylan, A. B., and Avise,
J. C., Global phylogeography of the loggerhead turtle (Caretta caretta) as indicated
by mitochondrial DNA haplotypes. Evolution, 48, 1820, 1994.
81. Skiles, D. D., The geomagnetic field: its nature, history, and biological relevance, in
Magnetite Biomineralization and Magnetoreception in Organisms, Kirschvink, J. L.,
Jones, D. S., and MacFadden, B. J., Eds., Plenum Press, New York, 1985, 43.
82. Lohmann, K. J. and Lohmann, C. M. R, Detection of magnetic inclination angle by
sea turtles: a possible mechanism for determining latitude,/. Exp. Biol, 194, 23,1994.
83. Lohmann, K. J. and Lohmann, C. M. R, Detection of magnetic field intensity by sea
turtles, Nature, 380, 59, 1996.
versus social facilitation models for green turtle migration, Science, 248, 724, 1990.
85. Gould, J. L., Are animal maps magnetic?, in Magnetite Biomineralization and Mag
netoreception in Organisms, Kirschvink, J. L., Jones, D. S., and MacFadden, B. J.,
Eds.. Plenum Press. New York, 1985, 257.
86 . Salmon, M. and Wyneken, J., Orientation by hatchling sea turtles: mechanisms and
implications, Herp. Nat. Hist., 2, 13, 1994.
87. Luschi, R, Papi, R, Liew, H. C., Chan, E. H., and Bonadonna, R, Long-distance
migration and homing after displacement in the green turtle (Chelonia mydas): a
satellite tracking study,/. Comp. Physiol. A., 178, 447, 1996.
Wisby, W. J. and Hasler, A. D., The effect of olfactory occlusion on migrating silver
salmon (O. kisutch), J. Fish. Res. Bd. Can., 11, 472, 1954.
89. Hasler, A. D. and Scholz, A. T., Olfactory Imprinting in Homing Salmon, Springer-
Verlag, Berlin, 1983.
90. Grassman, M., Chemosensory orientation behavior in juvenile sea turtles, Brain
Behav. Evol, 41, 224, 1993.
91. Mantón, M., Karr, A., and Ehrenfeld, W., Chemoreception in the migratory sea turtle,
Chelonia mydas, Biol. Bull., 143, 184, 1972.
92. Mantón, M. L., Karr, A., and Ehrenfeld, D. W., An operant method for the study of
chemoreception in the green turtle, Chelonia mydas. Brain Behav. Evol., 5, 188,1972.
93. Koch, A. L., CaiT, A., and Ehrenfeld, D. W., The problem of open-sea navigation:
the migration of the green turtle to Ascension Island,/. Theoret. Biol., 22, 163, 1969.
94. Brown, C. W., The significance of the south Atlantic equatorial countercurrent to the
ecology of the green turtle breeding population of Ascension Island,/. HerpetoL, 24,
81, 1990.
95. Carr, A., Panspecific reproductive convergence in Lepidochelys kempi, Ergeb. Biol.,
26, 298, 1963.
96. Meylan, A. B., Bjorndal, K. A., and Turner, B. J., Sea turtles nesting at Melbourne
Beach, Rlorida. II. Post-nesting movements of Caretta caretta, Biol. Conserv., 26,
79, 1983.
97. Baker, T. C., Chemical control of behavior, in Comprehensive Insect Physiology,
Biochemistry, and Pharmacology, Vol. 9, Behaviour, Kerkut, G. A. and Gilbert,
L. I., Eds., Pergamon Press, Oxford, 1985, 621.
98. Dusenbery, D. B., Sensory Ecology, W.H. Rreeman, New York, 1992.
99. Grassman, M. A., Owens, D. W., McVey, J. P, and Marquez, R. M., Olfactory-based
orientation in artificially imprinted sea Science, 224, 83, 1984.
100. Grassman, M. A. and Owens, D. W., A further evaluation of imprinting in Kemp’s
ridley sea turtle, in Proc. 1st hit. Symp. on Kemp’s Ridley Sea Turtle Biology, Con
servation and Management, Caillouet, C. W., Jr. and Landry, A. M., Jr., Eds., Texas
A&M University Sea Grant College Program, TAMU-SG-89-105, 1989, 90.
101. Grassman, M. and Owens, D., Chemosensory imprinting in juvenile green sea turtles,
Chelonia mydas, Anim. Behav., 35, 929, 1987.
102. Walker, M. M. and Bitterman, M. E., Honeybees can be trained to respond to very
small changes in geomagnetic field intensity,/. Exp. Biol., 145, 489, 1989.
103. Gould, J. L., The map sense of pigeons. Nature, 296, 205, 1982.
6 Migration
Habitat Utilization and
in Juvenile Sea
Turtles
John A. Musick and Colin J. Limpus
CO N TEN T
6.1 Introduction.....................................................................................................137
6.2 Methods......................................................................................................... 139
6.3 Early Juvenile Nursery Habitats................................................................... 140
6.3.1 Loggerhead, Caretta caretta..............................................................140
6.3.2 Kemp’s Ridley, Lepidochelys kempi................................................. 142
6.3.3 Olive Ridley, Lepidochelys olivácea................................................ 144
6.3.4 FlaXback Turtle, Natato?^ depressus .................................................144
6.3.5 Green Turtle, Chelonia m ydas......................................................... 144
6.3.6 Hawksbill, Eretmochelys imbricata.................................................. 145
6.3.7 Leatherback, Dermochelys coriácea................................................ 145
6.4 Later Juvenile DevelopmentalHabitats..........................................................146
6.4.1 Loggerhead........................................................................................ 146
6.4.2 Kemp’s Ridley.................................................................................. 148
6.4.3 Olive Ridley...................................................................................... 150
6.4.4 Flatback Turtle.................................................................................. 150
6.4.5 Green Turtle...................................................................................... 150
6.4.6 Hawksbill.......................................................................................... 152
6.4.7 Leatherback....................................................................................... 152
6.5 Conclusions................................................................................................... 153
Acknowledgment.................................................................................................. 154
References..............................................................................................................155
6.1 INTRODUCTION
Sea turtles are basically creatures that spend their entire lives in marine or estuarine
habitats. Their only remaining reptilian ties to terrestrial habitats are for nesting and
restricted cases of basking. Consequently, physiological, anatomical, and behavioral
adaptations have evolved largely in response to selection in the aquatic environment.
0-8493-8422-2/97/$0.00+$.50
and sea turtles share many common elements with larger fishes and cetaceans in
their habitat utilization and migrations. A generalized habitat model may be con
structed for sea turtles based on ontogenetic stages (Figure 6.1):
1. Early juvenile nursery habitat (usually pelagic and oceanic).

2. Later juvenile developmental habitat (usually demersal and neritic).
3. Adult foraging habitat.
4. Adult inter-nesting and/or breeding habitat.
FIGURE 6.1 Conceptual model of ontogenetic habitat stages in sea turtles.
All sea turtles move immediately to the sea after hatching, usually after dark,
and swim actively offshore. Most then undertake a mostly passive, denatant (sensu
Jones)^ migration drifting pelagically in oceanic gyre systems. Subsequently, after
a period of years, these now larger and older juveniles actively recruit to demersal
neritic developmental habitats in the tropical and temperate zones. Demersal juve
niles in some temperate zone populations make seasonal migrations to foraging areas
at higher latitudes in summer and lower latitudes in winter (see below) while those
in tropical areas are more localized in their movements. When approaching maturity,
pubescent turtles move into adult foraging habitats. In some populations adult
habitats are geographically distinct from juvenile developmental habitats;^ "în others
they may overlap or coincide.'’’^ Upon maturity as the nesting season approaches
adults make a contranatant migration toward the nesting beaches. Most mating occurs
Habitat Utilization and Migration in Juvenile Sea Turtles 139
at poorly defined courtship areas that are close to the nesting beaches relative to the
distant foraging areas. After mating the females move to their respective nesting
beaches.^’^ Courtship areas may be directly off the nesting beaches,^ or remote from
the beaches,depending on the population. During the nesting season, females
usually become resident in the internesting habitat in the vicinity of the nesting
beach. The focus of the present paper is habitat utilization and migration of juvenile
sea turtles and nursery and developmental habitats.
6.2 METHODS
Although some authors have attempted to define objective size catagories of sea
turtles by ontogenetic stages,temiinology for juvenile life history stages has been
varied and often imprecise. The words hatchling and neonate have been widely used
and clearly refer to animals that have quite recently hatched. The term post-hatchling
obviously refers to individuals that are larger and older than a hatchling, but the size
(or age) at which a hatchling becomes a post-hatchling varies by author and is ill
defined. Even less well defined is the size or age when a post-hatchling becomes a
juvenile or an “immature” or a “yearling”, and when this stage becomes the subadult
stage. Many authors have used the teim subadult to refer to all juvenile turtles that
have recruited to demersal, neritic habitats. This term is a misnomer for animals that
may not mature for 15 to 45 years. Critical ontogenetic habitat shifts such as that
from the pelagic nursery to the demersal developmental habitats occur in different
species or even populations at different ages and sizes. For instance, Eretmochelys
imbricata and Lepidochelys kempi usually make this shift at a smaller size (and
younger age) than Chelonia mydas, which in turn makes this shift at a smaller size
(and younger age), than does Caretta caretta (see below). Even within C. caretta,
western North Atlantic populations make the shift from the juvenile pelagic habitat
to demersal habitat at a smaller size (and younger age) than do western South Pacific
populations (see below). The adult stage would seem to be easily defined (i.e.,
minimum size of nesting females and clearly mature males with large dimorphic
tails); however, large numbers of Chelonia mydas, Caretta caretta, and E. imbricata
that are larger than the minimum breeding size of the nesting females recorded on
the beaches are still sexually immature and hence incapable of br ee di ng. Se xual l y
immature turtles that are within the size of breeding turtles also occur with L. kempii
and Dermochelys coriacea}"^ The average female commences to breed at a size only
slightly smaller than the average size of the entire nesting population.C onse
quently, the size and age range for transition from the juvenile to the adult stage is
usually broad within most populations. In order to avoid further confusion concern
ing ontogenetic stages, we have restricted our use to the terms hatchling (or neonate),
juvenile, and adult. Hatchlings refer to sea turtles that have quite recently hatched
and are still on the nesting beach or at sea only until they commence to feed (i.e.,
while they are dependent on the internalized yolk sac).^^ The at-sea period for the
hatchling is short (days), so that size is still essentially similar to that at hatching.
Juveniles include turtles that have commenced feeding, but have not attained sexual
maturity. The term adult will be used to describe all turtles that have attained adult
size, i.e., that are larger than minimum breeding size. Adult-sized turtles within
feeding areas will include a mixture of sexually immature and mature animals, unless
maturity is assessed via direct observation of the gonads and associated ducts or
from a past breeding history. We will reference other terms when used by other
authors. The size of sea turtles is usually given as curved carapace length (ccl, length
measured with tape over the curvature of the back) or straight carapace length (scl,
straight line length of the carapace measured with calipers). When the type of
measurement has not been designated in the literature we have referenced it as
undesignated length (ul).
6.3 EARLY JUVENILE NURSERY HABITATS

After emerging from the nest, neonate sea turtles crawl down the beach, swim into
the waves, and disappear offshore, not to reappear in the neritic habitat until they
have grown to be much larger juveniles. This largely unknown stage of sea turtle
life history became known as the “lost year”. Carr^^ noted that the lost year puzzle
was the “most substantial of all obstacles to understanding the ecology of sea turtles”,
but he also suggested that the lost year was spent in pelagic, oceanic habitats.
Evidence for that hypothesis has been slowly, but steadily accumulating.
6.3.1 L oggerhead, C aretta caretta
The early juvenile nursery habitat of the loggerhead has been best documented of
all sea turtle species. The occurrence of small loggerheads far at sea often in close
association with drifting Sargassum within the North Atlantic Ocean has been known
for several y e a r s . S e v e r a l authors have noted the potential importance of oceanic
convergence zones and major gyre systems in the distribution and local concentration
of juvenile loggerheads.^^ Recent studies have provided strong support for these
hypotheses. Witherington^^ has documented substantial numbers of neonate logger-
heads along the western edge of the Gulf Stream in convergence zones actively
feeding among Sargassum. Similarly, Richardson and McGilivary^^ recorded post
hatchling loggerheads actively feeding (on aeolian transported insects) in Sargassum.
Behavioral studies in the laboratory^^’^^ have shown that neonate loggerheads are
attracted to floating seaweed and hide in the weed motionless for long periods of
time. (The hatchlings’ basic brown and tan coloration renders them cryptic in the
Sargassum habitat.) Carr et al.^^ hypothesized that neonate loggerheads from the
southeastern U.S. swam offshore and entered the Florida current (subsequently the
Gulf Stream) where they became entrained and were passively transported across
the North Atlantic to the eastern Atlantic. Then they drifted south in the North
Atlantic gyre past the Azores and the Canary Islands and subsequently returned with
the North Equatorial CuiTent back into the western Atlantic (Figure 6.2). He based
this hypothesis on the small size classes (<40 cm scl) of loggerheads recorded
stranded in the eastern Atlantic.^^ (These “lost year” size classes are rare in the
western Atlantic). Witham^^ had estimated that a young sea turtle could make this
trip in about a year, but other evidence suggests it lasts longer. Several recent pa-
pers3o-33 shown that young loggerheads successfully make trans-Atlantic cross-
n:
O
g-)
DJ
D0
01
3Q .
FIGURE 6.2 North Atlantic gyre showing potential oceanic transport routes for pelagic juvenile sea turtles. (After Carr, 1987.)
ings. Juvenile loggerheads have been reported to be common and often captured
incidentally in longline fisheries around the Azores and Madeira.^^-^'^'^^ Lohmann et
al."^^ (see Chapter 5) showed that juvenile loggerheads could use the magnetic field
of the earth for navigation, and could make midcourse connections to maintain their
position in the North Atlantic gyre. Studies near the Azores suggest that juvenile
loggerheads appear to spend a much longer time in the oceanic realm than previously
supposed, the “lost year” more likely representing a “lost decade”.^^ This assertion
is supported by skeletochronological studies of pelagic juvenile loggerheads captured
incidentally in oceanic driftnets in the North Pacific.^^ These turtles were 2 to 8
years old and probably originated from nesting beaches in Japan. In addition, a large
pelagic feeding aggregation of juvenile loggerheads (20 to 30 cm ul) has been
documented off Baja California.Genetic studies have shown that these animals
originated in Japanese and Australian rookeries.^^ Within the Indian Ocean, Hughes"^^
has successfully plotted the dispersal of “tagged” juvenile C. caretta from the South
African (Natal) nesting beaches by the Agulhas Current into the Indian Ocean gyre.
Thus, much evidence has accumulated that loggerhead turtle populations nesting in
the northwest Atlantic, West Indian, and West Pacific oceans utilize oceanic, pelagic
nursery grounds, and migrate with the predominant ocean gyres from west to east
for several years before returning to their western neritic foraging and nesting
habitats. The duration of the pelagic stage is highly variable and probably ranges
from three to at least ten years or longer, depending on the individual and ocean
basin. Loggerheads as small as 25 to 35 cm (ccl) have been recorded stranded in
the Chesapeake Bay (U.S.) developmental habitat.^ These animals were probably
about two to four years of age."^^ In addition, Ke i na t hus i ng satellite telemetry,
found that head-started loggerheads released at two years of age off Virginia still
exhibited the same epipelagic surfacing behavior as neonates and migrated east to
the Gulf Stream. However, a head-started loggerhead released at three years of age
became demersal and followed the same autumn migration pattern to the south as
that followed by wild demersal juveniles from Chesapeake Bay. Thus, the ontoge
netic behavioral shift from epipelagic to demersal had taken place in this individual.
6.3.2 Kemp' s R idley, L epid ochelys kempi
Neonate Kemp’s ridleys disappear into the sea and do not reappear in the neritic
zone until they have reached a length of about 20 to 25 cm (scl)'^'^ or at about an
age of two years."^^ As their somber coloration would predict, juvenile Kemp’s ridleys
have also been reported in floating mats of vegetation.^^ Kemp’s ridleys only nest
on one restricted beach area (Rancho Nuevo) in Tamaulipas, Mexico. Thus, pelagic,
oceanic transport of neonates must be initially controlled by the hydrography in the
western Gulf of Mexico.'^^'^^ After initial entrainment in the anticyclonic Mexican
Current off the nesting beach, pelagic L. kempi should be swept into the northern
Gulf of Mexico, then eastward. Some juveniles are probably retained in the northern
Gulf until they migrate inshore and become demersal. Others may be swept south
into the Loop current, and then into the Florida Current and north into the Gulf
Stream"^^ (Figure 6.3). Most of these have grown sufficiently large to migrate inshore
and to adopt a demersal life-style by the time they reach the latitude of Long Island,
100*" 98° zcu
cr
30°N N
OJ
o'
o
D
Cl
<
O
Zi
FIGURE 6.3 Major hydrographic features in the Gulf of Mexico that may effect transport of pelagic juvenile Kemp’s ridley’s. (After Collard 1990.) w
NY, or southern New E n g l a n d . O t h e r s are carried by the Gulf Stream and North
Atlantic gyre into the eastern Atlantic^"^’^"^ (see review in Marquéz).^^ The fate of
these pelagic juveniles expatriated from the Gulf of Mexico has been conjectural,
some authors suggesting that they are lost to the breeding population.^^ "^^Conversely,
there is no reason to assume that Kemp’s ridleys do not have navigational abilities
equal to those of loggerheads and, therefore, probably are able to find their way
back to the Gulf of Mexico. Mounting evidence on the behavior of demersal juveniles
in neritic developmental habitats strongly supports the latter hypothesis (see below).
In addition, Meylan^^ reported that from 1980 to 1985, 29% of stranded ridleys
(mostly juveniles) were reported from the Atlantic coast of the U.S.; the remainder
from the Gulf of Mexico. If such a large proportion of the population regularly uses
Atlantic coast developmental habitats, it is likely that they can return to the Gulf of
,^léxico to breed.
6.3.3 O live R idley, L epid ochelys ouvacea
Neonate olive ridleys emerge from the nest at night and rapidly crawl across the
beach to enter the surf.'^^’^^ Nothing is known of the early juvenile nursery area,
although it most probably is pelagic and oceanic. Even the adults of some populations
appear to use pelagic, oceanic feeding areas when not breeding or nesting.
Pitman^® reported that 26 of 247 olive ridleys sighted at sea by shipboard observers
in the eastern Pacific were associated with floating Sargas sum. Unfortunately, he
did not note the size of these animals.
6.3.4 Flatback T urtle, N a t a t o r depressus
Flatback turtles emerge from the nest and rapidly enter the sea at night.^^ Evidently
they lack an oceanic stage^^’^^ and use neritic nursery areas while still feeding near
the top of the water column. Hatchling flatbacks are larger (5.7 to 6.2 cm scl) than
most other cheloniids^"^ and thus may have higher survivorship in the neritic habitat.
6.3.5 G reen T urtle, C h e lo n ia m ydas
Hatchling green turtles enter the sea and continue to swim actively offshore for at
least 24 Thereafter, the hatchlings apparently rest at night, but continue to
swim actively offshore during the day.^^’^^’^^ Carr^^ documented several pelagic,
oceanic records of Chelonia neonates or post-hatchlings (<20 cm scl) both in the
Atlantic and Pacific.^^ Many of these were in the vicinity of Sargassum driftlines.^^
However, the occurrence of juvenile Chelonia in or near Sargassum might be for
tuitous because both the animals and the weed could be brought together passively
in convergence z one s . The strong counter-coloration of neonate green sea turtles
suggests they are open-water animals, and experiments in the l a bor a t or ys howe d
that hatchling Chelonia avoided floating weed and spent a greater amount of time
swimming in open water than did Caretta or Eretmochelys. Carr and Meylan^^ have
speculated on the most probable routes of transport for early juvenile Chelonia
hatched at Toituguero, Costa Rica, based on major oceanographic circulation patterns
in the Caribbean. Witham^^ provided a model for the transport of neonate green
turtles from the east coast of Florida in the Gulf Stream, around the North Atlantic
Basin in the North Atlantic gyre, with return to the Caribbean and Florida in the
North Equatorial Current (the same model presented for loggerheads above) (Figure
6.2). He also provided tagging records of yearling green turtles released in Florida
and recovered in the Azores and Madeira. Oceanic records of juvenile Chelonia are
not as common as those for Caretta. In addition, the recruitment of Chelonia to
neritic developmental habitats in general occurs at smaller sizes (=30 to 40 cm
cci) 13,71-74 those for Caretta (=50 to 70 cm ccl).^’^ Therefore, Chelonia must
spend a shorter time in the oceanic nursery and recruits to demersal developmental
habitats at a younger age than does Caretta, or it has a much slower pelagic growth
rate.
6.3.6 H awksbill, E r e t m o c h e l y s ¡m b r ic a t a
Upon hatching, neonate hawksbill behavior is much like that reported for other sea
turtles with emergence at night and immediate movement to the sea.^^ Carr^^ provided
several pelagic records of early juvenile hawksbills (5 to 21 cm scl) many of which
were found in association with Sargassum. Other recent records have been provided
by Redfoot et al.,^^ Limpus et al.,^^ and Parker.^^ Unfortunately, other than the one
original record Parker provided (=23 cm ccl), most of the others he cited were based
on beach strandings, and many of the animals might have been neritic and demersal
prior to death. At least one of his records^^ was based on a benthic hawksbill captured
in clam tongs inside Chesapeake Bay. In the laboratory, neonate hawksbills were
attracted to floating weed and used it for cover where they remained motionless for
long peri ods. Appar ent l y at least some hatchlings may remain on reefs close to
their natal beaches.^^’^®However, Boulon^^ reported that juvenile hawksbills recruit
to the demersal coral reef habitat in the Virgin Islands at 20 to 25 cm (scl). This
range complements the largest verified pelagic records (21 cm scl, 23 cm ccl) noted
above. Thus hawksbills in the Atlantic recruit to the neritic developmental habitat
at a smaller size than either the loggerhead or green turtle, probably at an age of 1
to 3 years. Age and growth have not been determined in pelagic hawksbills, and this
tentative age estimate is based on growth rates in demersal juvenile hawksbills,^^
and pelagic juvenile loggerheads^^ and juvenile Kemp’s ridleys."^^ In the Indo-Pacific
region hawksbills recruit to inhabit coral reefs at a larger size (usually >35 cm [ccl])^
and presumably at a greater age.
6.3.7 L eatherback, D e r m o c h e l y s c o r ia c e a
Hatchling leatherbacks move immediately to the sea and swim actively offshore.
Leatherbacks are pelagic even as adults, and it is not surprising that neonates are
more active than other species of sea turtles, swimming offshore both day and night
for at least six d after entering the sea.^^ The fate of pelagic juvenile leatherbacks
after leaving the nesting beach is one of the great mysteries of sea turtle biology.
Other than neonates, there are very few records of leatherbacks <110 cm
If Zug and Parham’s^^ age estimates are coiTect, and leatherbacks grow rapidly, then
juveniles virtually disappear for four years. Lutcavage and Lutz^^ noted that small
leatherbacks need to consume gelatinous prey (the low-energy food upon which they
specialize) equal to their biomass every day in order to support routine metabolism.
Their food requirements must even be higher to promote rapid growth. In the oceanic
realm, production necessary to support high biomass of gelatinous (and other)
animals is found primarily in areas of major upwelling such as those along the
eastern sides of the major ocean basins or perhaps in the Equatorial Convergence
Zones. These areas may provide nursery grounds for leatherbacks.
6.4 LA TER JU V E N ILE D E V E LO P M E N T A L H A B IT A T S

6.4.1 L oggerhead
In the western Atlantic, juveniles recruit from oceanic pelagic to neritic demersal
habitats when as small as 25 to 30 cm ccP (Figure 6.4), but most are >50 cm (ccl)^
at an age of about 7 to 10 years."^^ Limpus et al.^ reported that loggerheads in Australia
do not recruit to demersal habitats until 70 cm (ccl) or greater (Figure 6.5). Larger
size may be in keeping with the much more expansive gyre system in the Pacific
(thus requiring more time for pelagic juveniles to make the transit), or to slower
growth in pelagic juveniles in the Pacific or both. This is supported by recaptures
of large juvenile loggerhead turtles which had originally been “tagged” at south
Queensland nesting beaches as hatchlings and which had recently recruited to res
idency in shallow inshore habitats in eastern Australia at 15 years of age or older.^’^"^
Loggerheads are anti tropical in distribution, and nest on subtropical and warm-
temperate beaches.^^ In the western Atlantic some demersal juvenile loggerheads
make strong seasonal foraging migrations into temperate latitudes, occurring com
monly as far north as Long Island, NY.^^ Chesapeake Bay is a major seasonal
developmental habitat in summer^ with 5,000 to 10,000 loggerheads present each
s u mm e r . A r o u n d 95% of the loggerheads that visit Chesapeake Bay are juveniles
(Figure 6.4). They usually enter the bay in late May or early June when water
temperatures rise to 16 to 18°C, and depart from late September to early Novem-
ber.^’^“^ Juveniles become resident for the summer along channel edges (5-13 m) and
forage back and forth along the bottom, passively with the tide within a home range
of 10 to 80 km^ with preferred ranges (within the home range) of about 5 to 15
km^.^^ These juveniles show strong foraging site fidelity. Animals displaced >100
km have returned to within a few kilometers of their point of origin within a few
weeks.^^’^^' Of 121 loggerheads tagged in Virginia, 48 were recaptured there in
subsequent (but not necessarily consecutive) seasons.Some were recaptured up to
four seasons, showing strong foraging ground fidelity between seasons. Bowen et
aT^^ showed that there were two nesting populations of loggerheads in the western
North Atlantic; one in Florida, and another north of there in Georgia and South
Carolina. Norrgard^^ found that juvenile loggerheads from Chesapeake Bay were
derived at about a 50/50 ratio from both nesting populations. Sears^^’^^ found similar
results for juvenile loggerheads from Charleston, SC. These data suggest that the
Georgia-South Carolina turtles selectively use more northerly summer developmen
tal habitats than do Florida turtles. The Florida nesting population is about nine
times larger, and the juvenile ratios should have been heavily skewed toward the
Sf>
im
u.
cr> 0T\ cT> OS CH Ch

(T
—» HI C
^T> CJ^ CT* CT» C?^
LO SD r - CO
OS OS OS OS
OS O «r- < \l
oc\j oro o o o
’sT l/> \D O
OS OI Ósl
O C
length (cm )
FIGURE 6.4 Length frequency distribution (ccl) for loggerheads stranded in Virginia from
1974 to 1995.
50 j
I male
40 - ¡Female
30
g
20
I
10
0 "I ^ 1 r I..T" "1 ^ r "'1 f r" r "I 1

0 10 20 30 40 50 60 70 80 90 100 110 120
CURVED CARAPACE LENGTH (cm)
FIGURE 6.5 Length frequency distribution (ccl) for loggerheads collected in foraging
habitats in Australia.
Florida population if there were a random distribution of juveniles from both

groups.Aerial surveys and satellite tracking studies have shown that when juvenile
loggerheads migrate from Chesapeake Bay in autumn, they travel relatively close
to the coast (<20 km) and move south, rounding Cape Hatteras around Decern-
ber.^3,100,101 xhey are joined in the fall by substantial members of juvenile loggerheads
from the sounds of North Carolina. By December most loggerheads have
migrated south of Oregon Inlet, NC,^®‘ and by January most are south of Cape
Hatteras. Most of the turtles that remain off North Carolina in January and
February are found at the edge of the Gulf Stream. Keinath'^^ found that the juvenile
loggerheads he tracked from Virginia wintered in two basically distinct areas: one
group went down the coast, stayed inshore (hesitating during warm spells at various
major inlets), and wintered inshore off southern Georgia and Florida (one traveled
as far as the Florida Keys). The other group followed the same inshore route south
of Cape Flatteras to about Cape Lookout, NC, then moved offshore to winter on
reefs along the shelf edge on the western side of the Gulf Stream. One might
hypothesize that the group wintering off North Carolina was derived from Geor-
gia/South Carolina rookeries, and those that wintered off Florida were from Florida
rookeries. Only further genetic research will tell.
Juvenile loggerheads of about the same size distribution (50 to 80 cm scl) are
common during the summer in estuaries as far south as Mosquito and Indian
Lagoons, Florida^®"^’^^^ and in the Gulf of Mexico. In South Carolina and Georgia
they exhibit seasonal emigration from estuaries into the ocean in the autumn, with
immigration occurring in the s p r i n g . I n Mosquito Lagoon, Florida, juvenile
Carena occur year-round, but are captured in smallest numbers during February and
March (the time of lowest water tem perature).The reason for this may be because
the turtles emigrate out into the ocean, or rather they may be inactive, actually
brumating in the mud during this period. Juvenile loggerheads commonly brumate
in winter by digging head first into the mud in the Canaveral Ship Channel (in the
ocean not far from Mosquito L a g o o n ) . W e have never observed any evidence
of brumation in Virginia, nor have Epperly et al.^^^ in North Carolina where winter
water temperatures fall below the lethal lower limit for loggerheads (5 to 6.5°C).^
Henwood^^^ noted that juvenile loggerheads reached their peak abundance in winter
in Canaveral Ship Channel, and that some turtles tagged there in winter were
recaptured to the north in summer. In contrast, to the well-defined seasonal foraging
migrations exhibited by juvenile loggerheads along the Atlantic coast of the U.S.,
both large (>70 cm ccl) juvenile and adult Caretta were resident all year in the coral
reefs of the southern Great Barrier Reef and in Moreton Bay, Australia in the
southwestern Pacific.^’^^"^ The most southerly of these sites, Moreton Bay, has an
annual water temperature range of 16 to 28°C, much more moderate than those
found in estuarine and coastal waters north of Florida in the U.S. After recruitment
into the eastern Australian demersal habitats, these juvenile loggerheads show strong
forage site fidelity to relatively small areas, mostly remaining within a few square
kilometers for the next 8 to 20 years of growth to sexual maturity. Adult
loggerheads, after completing their first breeding migration to distant nesting
beaches, return with high fidelity to the same juvenile foraging areas in which each
had completed its juvenile life.^^"* They also return to the same foraging sites fol
lowing subsequent breeding migrations. It is strongly indicated that individual
loggerheads in the southwestern Pacific occupy very localized home ranges for the
decades that span their later juvenile and entire adult lives.
6.4.2 K emp' s R idley

Initial recruitment of juveniles from pelagic oceanic to demersal neritic habitats
takes place at a size of 20 to 25 cm (scl) in the northern Gulf of Mexico and in New
England and New York waters.“^“^ Contrary to the assumption of Carr^^ and Carr et
the juveniles carried by the Gulf Stream into the Atlantic probably make an
active rather than a passive migration west from the edge of the continental shelf to
find suitable shallow developmental habitats such as Long Island Sound. Although
Gulf Stream gyres may impinge upon the edge of the continental shelf, they rarely
cross the shelf to near coastal habitats.În the Gulf of Mexico, shallow coastal
habitats serve as foraging areas for Kemp’s ridley throughout the year, although
there is evidence for seasonal offshore movements in response to low water temper
atures in winter."^^'^^^ OccuiTence of L. kempi in shallow water off northwest Florida
is seasonal, with no turtles captured in the coldest winter months. In New York
and New England, juveniles must emigrate coastally to the south or face cold
stunning and death in the winter.^^’^^’^^^
Chesapeake Bay serves as an important developmental habitat for juvenile L.
kempiP'^"^ The summer population size has been estimated to be 211 to 1083 (a
minimum estimate) with a mean size of about 40 cm (ccl)^^^ (Figure 6.6). Juvenile
L. kempi preferred shallower summer habitats in Long Island Sound (<8
and Chesapeake Bay (<5 m)^^ than were used by juvenile loggerheads,^^ and often
foraged in submerged aquatic grass beds for Callinectes sapidus and other crabs.^
The seasonal migration pattern is similar to that of the loggerhead, with immigration
in May and June and emigration from September to November. Of 57 Kemp’s ridleys
tagged in Chesapeake Bay, 2 were recaptured there in subsequent summer s Thus ,
site fidelity is possible for summer foraging locations. Upon leaving Chesapeake
Bay in autumn, juvenile L. kempi migrate down the coast, passing Cape Hatteras in
December and January. These larger juveniles are joined there by juveniles of the
same size from the North Carolina sounds and smaller juveniles from New York and
New England to form one of the densest concentrations of Kemp’s ridleys outside
the Gulf of Mexi co. Kei nat h*^^ used satellite transmitters to track three juvenile
Kemp’s ridleys released in Virginia in October. They followed the same inshore
route noted for loggerheads above, and two individuals migrated as far as Florida
by mid-January. (The transmitter failed on the third turtle in November when the
animal was approaching Cape Fear, NC). Henwood and Ogren^^^ noted that juvenile
Kemp’s ridleys reached their maximum abundance off Cape Canaveral, FL in Jan
uary to March, and that their tagging data suggested a northward movement in
summer and a southward movement in the winter. These data complement and
support the pattern of seasonal migration noted to the north.
Ogren"^^ reported that the entire northern coast of the Gulf of Mexico from Port
Aransas, TX to Cedar Key, FL served as demersal developmental habitat for juvenile
Kemp’s ridleys. Areas where significant concentrations of juvenile ridleys occurred
included: the Sabine Pass area of Texas; Caillou Bay, Terrabonne Parish, LA; and
Big Gulley east of Mobile Bay, AL. The smallest juveniles (20 to 25 cm scl) were
found in two areas: one off western Louisiana and eastern Texas, and the other off
northwest Florida.^'^’^^'^ Most juvenile L. kempi in the latter area were captured at
depths <6.1 m, and all ridleys <25 cm (scl) were from <1 m. Juvenile Kemp’s ridleys,
then, recruit to shallow estuarine and coastal habitats in the western Gulf of Mexico
and off New England and New York in summer. In winter they migrate south and/or
offshore to avoid cold temperatures, but may remain resident in some areas of the
45 T
iZ 15 -
length (cm)
FIGURE 6.6 Length frequency distribution (cel) for Kemp’s ridley stranded in Virginia
from 1979 to 1995.
Gulf of Mexico that are sufficiently warm. It is possible that some individuals may
brumate at temperatures
6.4.3 O líve R idley

Very little is known about the developmental habitats of older juvenile olive ridleys.^’^
Adults may use coastaF^425 qj. oceanic foraging h a b i t a t s a n d juveniles may be
able to use either as developmental habitats, depending on food availability. In
northern Australian continental shelf waters, large immature and adult-sized olive
ridleys (52.0 to 72.0 cm ccl) are captured all year-round in prawn trawls over soft-
bottomed habitats. ^27
6.4.4 Flatback T urtle

This species is endemic to the Australian continental shelf. Demersal juveniles have
been reported frequently from the prawn trawling grounds inshore from the Great
Barrier Reef and across northern Australia. The species appears to avoid reef habitats,
but is common in turbid, shallow, inshore water throughout northern Australia north
of 2 5 °.^ "^ 4 2 8 Adult flatbacks appear to live sympatricly with the larger juveniles.
6.4.5 G reen T urtle

Juvenile green turtles recruit to demersal developmental habitats at about 30 to 40
cm (ccl) or larger^^ '^^’^^’^'^ (Figure 6.7). In the western Atlantic the summer develop
mental habitat encompasses estuarine waters as far north as Long Island Sound,
Chesapeake Bay, and the North Carolina Sounds south throughout the tropics. Green
turtles north of Florida must migrate south in autumn or face the risk of cold
stunning.'^^’^^^ ‘^^429 Guseman and Ehrhart^^^^ noted that the smallest juvenile green
turtles (26.8 to 45.3 cm scl) to appear in the demersal developmental habitat off the
east coast of Florida occuiTed on sabellarid polychaete reefs, and that as juveniles
40
Hi adult
»pubescent
30 Oprepubescent
è
zUJ
3 20
o
LU
DC
Ll
10
0 10 20 30 40 50 60 70 80 90
IL
100 110 120 130
CURVED CARAPACE LENGTH (cm)
FIGURE 6.7 Length frequency distribution (ccl) for green turtles captured in Shoalwater-
Bay, Australia.
became larger (28.8 to 66.8 cm scl) they moved into the Indian River Lagoon seagrass
beds to forage. Henwood and Ogren^^^ also noted that juvenile green turtles off Cape
Canaveral were smaller than those in the Mosquito Lagoon estuary.H enw ood and
Ogren^^^ suggested that these small coastal green turtles might be in early develop
mental stages prior to the shift to herbivory and recruitment to shallow inshore
feeding pastures. Mendonca and Ehrhart^®"^ showed that juvenile green turtles
recruited to the Mosquito Lagoon (Florida) estuarine developmental habitat at 30 to
40 cm scl and that some individuals became resident, showing foraging site fidelity.
Many turtles remained in the Lagoon year-round until they approached maturity.
They also noted that the catch rates of juvenile Chelonia were highest in the summer,
but that the winter decline might be due in part to turtle inactivity (brumation?).
Severe cold-stunning events in winter in 1977 and 1978 showed that green turtles
were still common in the Lagoon. Felger et al.^^^ reported Chelonia to burrow in
the mud (brumate) in winter in the Gulf of California. Mendonca*^^ noted that
juvenile green turtles in Mosquito Lagoon wandered as far as 5 to 10 km a day
when the water was cooler (11 to 18°C) but established home ranges of 1.2 to 4.1
km when temperatures were above 25°C. Bjorndal and Bolton^^^ reported that tidal
embayments were important developmental habitats for juvenile green turtles in the
Bahamas. Their tagging studies showed that immigration occurred to other areas of
the Bahamas, Colombia, Cuba, Dominican Republic, Haiti, Nicaragua, Panama, and
Venezuela.
Balasz^"^ and G r e e n n o t e d that juvenile green turtles in the Pacific remained
on reef habitats and consumed macroalgae. Balasz^"^ also reported that almost all
juvenile green turtles he tagged in the Hawaiian Islands were recaptured in the same
resident areas. Similarly, all the juvenile green turtles tagged by G r e e n i n the
Galapagos were recaptured within the archipelago. In the southwest Pacific juvenile
greens occupy the same wide range of habitats as the adults, being common on coral
reefs where algae is the dominant food^ and in inshore embayments where the main
food items are seagrasses.^^ These juvenile greens have displayed strong feeding site
fidelity within their feeding areas (tens of square kilometers) over many yearsT^’^^^^
Some juvenile greens have been under observation in the same localized feeding
areas in the southern Great Barrier Reef for two decadesd"^ Juvenile green turtles
can be resident for years in tropical or subtropical developmental habitats with stable
or moderately stable seasonal temperatures. They may utilize continental foraging
areas in temperate latitudes (to about 40°N) in summer, but must return to subtropical
latitudes in winter to avoid cold stunning. Brumation may occur during colder
periods in subtropical estuaries.
6.4.6 H awksbill
Hawksbills are restricted to the tropics more than any other sea turtle. Juveniles
recruit to developmental habitats at a very small size (20 to 25 cm scl) or even as
hatchlings (see above). The developmental habitat includes shallow (<20 m) coral
reefs and mangrove estuaries^^ where there are abundant sponges, the principal food
of this s p e c i e s . Th e developmental habitat does not seem to differ from that of
the adult, and juveniles and adults are taken together in the same foraging areas.'^’^^'^
Boulon^^^ tagged juvenile hawksbills in the Virgin Islands and found that many
migrated to areas outside the Virgin Islands. (Nesting hawksbills from the Virgin
Islands have also been found to be mostly mi g r a t or y) . I n contrast, in eastern
Australia, juvenile hawksbills >35 cm (ccl) have displayed a high fidelity to localized
feeding areas in coral reef habitats with some individuals being associated with the
same site for over a decade."^ In addition, juvenile hawksbills less than 30 cm ccl
are uncommon in the Great Barrier Reef waters, and none have been recorded in
demersal developmental habitats.^^ Very little research has been conducted on behav
ior and habitat utilization in juvenile hawksbills.
6.4.7 L eatherback
Although leatherbacks are pelagic, they recruit seasonally to temperate and boreal
coastal habitats to feed on concentrations of jellyfish.^ In the western Atlantic,
juvenile Dermochelys appear in these habitats at about 110 to 120 cm (ccl), based
on stranding data^^’^"^®(Figure 6.8). If we accept the size at maturity (based on the
nesting population in St. Croix) as 137 cm (ccl) ( the size of the smallest nesting
female) or as 157 cm (ccl), (the approximate average size of nesting females),
then a substantial percentage of leatherbacks stranding along the U.S. east coast
must be large juveniles. Aerial surveys have shown large numbers of leatherbacks
off the northeast coast of Florida in February,southern Virginia in late April,
and the Gulf of Maine in summer.Leatherback standings from 1980 to 1991 in
Florida occurred mostly between October and April.^^ Sightings of leatherbacks off
Massachusetts peak in August, with no live sightings before June or after October.
Although several tagging and tracking studies have been done on adult females that
show long-distance oceanic migration to higher-latitude coastal feeding habitats,
virtually no work of this kind has been done on juveniles. Paladino et al.^“^^ pointed
out that leatherbacks were inertial endotherms, but that the high surface area-to-
9O
4im
»
11
1 ^ ^ 1
-------- --------- r
1
T--------
1
1
1
T--------T--------- — —
CTk os OS OS as CJS CiS
CIS OS OS OS OS os OS CJS
CM ro in
T-
vD OO OS o
CM
•t—
CM
CM
CM
o o o o o
in o o o
00 o o o o
^ CM ro MD r-
T-.
OS o ■»—- CM
CM CM CM
Length (cm )
FIGURE 6.8 Length frequency distribution (ccl) for leatherbacks stranded in Virginia from
1979 to 1995.
mass ratio in juveniles precluded body temperatures significantly higher than ambi
ent. The size range at which leatherbacks become efficient endotherms is probably
the same as that when they first appear in coastal, boreal waters (110 to 120 cm ccl).
The leatherback shares many attributes with the endotheimic scombroid hshes
(tunas). Both reproduce in the tropics. Both have been able to expand their effective
foraging ranges into higher latitudes and greater depths (into the thermocline), well
beyond the normal range of their nonendothermic relatives.
6.5 CONCLUSIONS
A comparison of the ontogenetic patterns of habitat utilization and migration among
the seven species of sea turtles leads to the following hypotheses: natural selection
will favor those nesting locations with the highest survivorship during the vulnerable
early juvenile stage when predation is highest. Given all sea turtles nest in tropical,
subtropical, or warni temperate latitudes, the most important determinant of nesting
beach locations probably is survivorship of hatchlings and small juveniles in the
nursery habitats.
In sea turtles as in fishes, the numbers and diversity of potential predators and
resulting mortality rates are inversely proportional to the size of the juvenile. The
advantage of pelagic, oceanic nurseries is the low density of predatory fishes and
sea birds there dictated by low primary production.‘^^’^‘^'^ (Productive-upwelling areas
are exceptions.) Based on predator mouth size alone, few potential avian or fish
predators, except s h a r k s , a r e capable of engulfing hard-bodied epipelagic prey
(such as sea turtles) >20 to 30 cm in diameter. Marine crocodiles, which can also
be significant predators of most size ranges of marine turtles, are largely restricted
to tropical inshore margins of marine turtle foraging habitats. Marine mammals
appear to be rarely predators of marine turtles. Hawksbills and Kemp’s ridleys recruit
from pelagic, oceanic nurseries to neritic, demersal developmental habitats at about
20 to 25 cm (scl); green turtles do so at about 30 cm. In order to avoid predation
from sharks and large teleosts, the smallest demersal recruits tend to use structured
habitats such as reefs that provide cover (hawksbills and green turtles) or very
shallow habitats where larger predators cannot go (Kemp’s ridleys). Juvenile log
gerheads stay in the pelagic, oceanic nursery much longer and to a much larger size.
Apparently, so do some populations of olive ridleys, which may use oceanic feeding
grounds as adults (mostly in up welling areas with abundant crustacean prey). Log
gerheads are opportunistic carnivores utilizing a wide variety of prey, including
molluscs, crustaceans, horseshoe crabs, and gelatinous m acroplankton.They seem
to be well adapted for long periods of opportunistic foraging (mostly on gelatinous
prey) in the oceanic realm, although they have relatively slow growth rates there.^^
Older juvenile and adult green, hawksbill, and Kemp’s ridley turtles have specialized
diets (herbivory, spongivory, and cancrivory, respectively), and might be compelled
by dietary energetic constraints to seek out more productive demersal developmental
habitats at smaller sizes than loggerheads. In addition, Kemp’s ridley, the smallest
species of sea turtle, seems to have a shorter life cycle and matures at an earlier age
than other species of the Cheloniidae."^^
Most species that primarily utilize pelagic, oceanic nurseries and have relatively
long (>3-year) oceanic stages tend to have high-density nesting localities located
adjacent to major oceanic currents that can maintain small juveniles passively and
safely for sufficient time for them to grow large enough to lower predation risk and
to have the mobility to actively swim to developmental habitats. Loggerheads, green
turtles, and both Kemp’s and olive ridleys seem to fit this pattern. Nesting is mostly
restricted to certain sites used (or historically used) by large numbers of nesters.
Small scattered nesting sites along the fringes or outside of these primary sites are
probably constrained by very low survivorship of the pelagic juveniles caused by
poor access to the passive hydrographic pathways that lead to the oceanic nurseries.
Hawksbills do not fit this pattern well. Although some concentrated nesting areas
exist^^® or existed historically, solitary nesting is widespread and frequent mostly on
islands.Hawksbills may have the shortest pelagic stage of all sea turtles (except
the flatback), or hatchlings may recruit directly to demersal developmental habitats
on coral reefs or mangrove flats. Also, they tend to occur in archipelagos where
local or regional hydrographic features may be more important than major oceanic
gyres during the short passive transport stage of the juveniles. Leatherbacks are
pelagic during all stages of their life, and utilize oceanic habitats more than any
other sea turtle.^^’^^* Although there are some high-density nesting a r e a s , m u c h
scattered nesting occurs, and this species may have lower re nesting fidelity than
other sea turtles. This apparent enigma may be explained by the larger size and
faster growth rates and superior swimming ability of neonate leatherbacks. These
attributes may render juvenile leatherbacks much less dependent on passive transport
to the oceanic nurseries. Young leatherbacks undoubtedly use major oceanic current
systems for transport, but may be more capable of actively reaching those passive
pathways.
ACKNOWLEDGMENT
Virginia Institute of Marine Sciences contribution # 2026.
REFERENCES
1. Jones, F. R., Fish Migration, Edward Arnold, London, 1968.
2. Musick, J. A., The Sea Turtles o f Virginia, 2nd rev. ed., Virginia Sea Grant Program,
Virginia Institute of Marine Science, Gloucester Point, VA, 1988.
3. Lutcavage, M. and Musick, J. A., Aspects of the biology of sea turtles in Virginia,
Copeia, 2, 449, 1985.
4. Limpus, C. J., The hawksbill turtle, Eretmochelys imhricata, in Queensland: popu
lation structure within a southern Great Barrier Reef feeding ground, Wild. Res., 19,
489, 1992.
5. Limpus, C. J. and Reed, P. C., The green turtle, Chelonia my das, in Queensland: a
preliminary description of the population structure in a coral reef feeding ground, in
Biology of Australiasian Frogs and Reptiles, Griggs, G., Shine, R., and Ehmann, H.,
Eds., Surrey Beatty and Sons, Sydney, 1985, 47.
6. Limpus, C. J., Couper, P. J., and Reed, M. A., The loggerhead Caretta caretta,
in Queensland: population structure in a warm temperate feeding dx^di,Mem. Queensl.
Mus., 37(1), 195, 1994.
Limpus, C. J., The green turtle, Chelonia mydas, in Queensland: breeding males,
Wild. Res., 20, 513, 1993.
Meylan, A. B. and Meylan, P. A., Description of a mighty fleet of green turtles, in
Proc. 13th Ann. Symp. Sea Turtle Biology and Conservation, Schroeder, B. A. and
Witherington, H. E., Compilers, NOAA Tech. Memo. NMFS-SEFSC-341, 107,
1994.
9. Balazs, G. H., Recovery records of adult green turtles observed or originally tagged
at French Frigate Shoals, northwestern Hawaiian Islands, NOAA Tech. Memo.
NMFS-SWFC-36, 42, 1983.
10. Pitman, R. L., Pelagic distribution and biology of sea turtles in the eastern tropical
Pacific, in Proc. Tenth Annu. Workshop on Sea Turtle Biology and Conservation,
Richardson, T. H., Richardson, J. I., and Donnely, M., Compilers, NOAA Tech.
Memo. NMFS-SEFC-278, 143, 1990.
11. Meylan, A., Behavioral ecology of the west Carribean green turtle (Chelonia mydas)
in the internesting habitat, in Biology and Conservation o f Sea Turtles, Bjorndal, D.,
Ed., Smithsonian Institation Press, Washington D.C., 1982, 67.
12. Dodd, D. C., Synopsis of the biological data on the loggerhead sea turtle, Caretta
caretta (Linnaeus, 1758), FAO Synopsis NMFS-149, USFWS Biological Report
88(14), n o , 1988.
13 Limpus, C. J., Couper, P. J., and Reed, M. A., The green turtle Chelonia mydas in
Queensland: Population structure in a warm temperate feeding area, Mem. Queensl.
Mus., 35(1), 139, 1994.
14. Limpus, C. J., unpublished data, 1966.
15. Limpus, C. J., Puberty and first breeding in Caretta caretta, NOAA Tech. Memo.
NMFS-SEFC-278, 81-3, 1991.
16. Miller, J. D., Embryology and marine turtles, in Biology o f the Reptilia, Gans, C.,
Billett, R, and Maderson, P. F. A., Eds., (14), 269, 1985.
17. Carr, A., So Excellent a Fish, Natural History Press, Garden City, NY, 1967, 249.
18. Carr, A., Notes on the behavioral ecology of sea turtles, in Biology and Conservation
o f Sea Turtles, Bjomdal, K. A., Ed., Smithsonian Institution Press, Washington, D.C.,
1982, 19.
19. Fletmeyer, J. R., Underwater tracking evidence that neonate loggerhead sea turtles
seek shelter in drifting Sargassum, Copeia, 1, 148, 1978.
20 . Smith, W. G., A neonate Atlantic loggerhead turtle, Caretta caretta, captured at sea,
Copeia, 4, 880, 1968.
21 Carr, A., Heath, H., and Ogren, L., The ecology and migrations of sea turtles: the
hawksbill turtle in the Caribbean Sea, Am. Mus. Novit., 2248, 1, 1966.
22 . Carr, A., New perspectives on the pelagic stage of sea turtle development, Conserv.
Biol., 1(2), 103, 1987.
23 Witham, R. M., The ‘Tost years” question in young sea turtles. Am. ZooL, 20, 525,
1980.
24. Witham, R., On the ecology of young sea turtles, E/. Sd., 54, 179, 1991.
25. Witherington, B. E., Some “lost year” turtles found, in Proc. 13th Annu. Symp. Sea
Turtle Biology and Conservation, Schroeder, B. A. and Witherington, H. E., Com
pilers, NOAA Tech. Memo. NMFS-SEFSC-341, 192, 1994.
26. Richardson, J. I. and McGillivary, R., Post-hatching loggerhead turtles eat insects in
Sargassum community. Mar. Turtle. Newsl., 55, 2, 1991.
27. Mellgren, J. R., Morra, M., Bushong, M., Harkins, S., and Krumke, V., Habitat
selection in three species of captive sea turtle hatchlings, in Proc. 14th Annu. Symp.
Sea Turtle Biology and Conservation, Bjorndal, K. A., Bolton, A. B., Johnson, P. A.,
and Eliazar, P. J., Compilers, NOAA Tech. Memo. NMFS-SEFSC-351, 259, 1994.
28. Mellgren, R. G., and Mann, M. A., Comparative behavior of hatchling sea turtles, in
Proc. 15th Annu. Symp. Sea Turtle Biology and Conservation, Keinath, J. A., Barnard,
D. E., Musick, J. A., and Bell, B. A., Compilers, NOAA Tech. Memo. NMFS-SEFSC-
387, 202, 1996.
29. Brongersma, L. D., European Atlantic turtles. Zoo/. Berhand. Leiden, 121,318, 1972.
30. Eckert, S. A. and Martins, H. R., Transatlantic travel by a juvenile loggerhead turtle.
Mar. Turtle Newsl., 45, 15, 1989.
31. Manzella, S. A. and Fontaine, C. T., Loggerhead sea turtle travels from Padre Island,
Texas to the mouth of the Adriatic Sea, Mar. Turtle. Newsl., 42, 7, 1988.
32. Bolten, A. B., Santana, C., and Bjorndal, K. A., Transatlantic crossing by a loggerhead
turtle, Mar. Turtle Newsl., 59, 7, 1992.
33. Bolton, A. B., Martins, H. R., Watali, M. L., Thomas, J. C., and Marcovaldi, M. A.,
Loggerhead released in Brazil recaptured in Azores, Mar. Turtle. Newsl, 48, 24, 1990.
34. Brongersma, L. D., Marine turtles of the eastern Atlantic ocean, in Biology and
Conservation o f Sea Turtles, Bjomdal, K. A., Ed., Smithsonian Institution Press,
35. Bolten, A. B., Martins, H. R., Bjorndal, K. A., and Gordon, J., Size distribution of
pelagic-stage loggerhead sea turtles {Caretta caretta) in the waters around the Azores
and Mdidmdi, Arquipelagic, 11 A, 49, 1993.
36. Bjomdal, K. A., Bolten, A. B., Gordon, J., and Caminas, J. A., Caretta caretta
(loggerhead) growth and pelagic movement, Herptol. Rev., 25, 23, 1994.
37 . Lohmann, K. J., Witherington, B. E., Lohmann, C. M., and Salmon, M., Orientation,
navigation and natal beach homing in sea turtles, in Biology o f Sea Turtles, Lutz, P.
and Musick. J. A.. Eds.. CRC Press. Boca Raton. FL, 1996, 109.
38. Bolton, A. B. and Balazs, G. H., Biology of the early pelagic stage — the “lost year”,
in Biology and Conservation o f Sea Turtles, Bjorndal, K. A., Ed., Smithsonian
Institution Press, Washington, D.C., 1995, 579.
39. Zug, G. R., Balzs, G. H., and Wetherall, J. A., Growth in juvenile loggerhead sea
turtles (Caretta caretta) in the north Pacific pelagic habitat, Copeia, 484, 1995.
40. Bowen, B. W., Abreu-Grobois, F. L., Balazs, G. H., Kamezaki, N., Limpus, C. J.,
and Perl, R. J., Trans-Pacific migrations of the loggerhead turtle (Caretta caretta)
demonstrated with mitochondrial DNA markers, Proc. Natl. Acad. Sci., 92, 3731,
1995.
41. Hughes, G. R., The sea turtles of southeast Africa. I. Investigational Report No. 36,
Oceanographic Research Institute, Durban, 1974.
42. Klinger, R. C. and Musick, J. A., Age and growth of loggerhead turtles (Caretta
caretta) from Chesapeake Bay, Copeia, 1995(1), 204, 1995.
43. Keinath, J. A., Movements and Behavior of Wild and Head-Started Sea Turtles, Ph.D.
dissertation, College of William and Mary, Williamsburg, VA, 1993.
44. Ogren, L. H., Distribution of juvenile and subadult Kemp’s ridley turtles: preliminary
results from the 1980-1987 suiweys, in Proc. 1st Int. Symp. Kemp’s Ridley Sea Turtle
Biology, Conservation and Management, Calliouet, C. W., and Landry, A. M., Eds.,
Texas A & M University Sea Grant College Program Spec. Publ. -89(105), 116, 1989.
45. Zug, G. R., Kalb, H., and Luzan, S. L., Age and growth in wild Kemp’s ridley sea
turtles (Lepidochelys kempi) from skeletochronological data., Biol. Consent., in press.
46. Manzella, S. A., Williams, J. A., Schroeder, B., and Teas, W., Juvenile head-started
Kemp’s ridleys found in floating grass mats. Mar. Turtle Newsl., 52, 5, 1991.
47. Collard, S. B., Guest editorial: speculation on the distribution of oceanic-stage sea
turtles, with emphasis on Kemp’s ridley in the Gulf of Mexico, Mnr. Turtle Nentsi,
48, 6, 1990.
48. Collard, S. B., The influence of oceanographic features in post-hatching sea turtle
distribution and dispersion in the pelagic environment, in Proc. 10th Annu. Workshop
on Sea Turtle Biology and Conservation, Richardson, T. H., Richardson, J. I., and
Donnely, M., Compilers, NOAA Tech. Memo. NMFS-SEFC-278, 111, 1990b.
49. Collard, S. B. and Ogren, L. H., Dispersal scenarios for pelagic post-hatchling sea
turtles, Bull. Mar. Sci., 47, 233, 1990.
50. Lazell, J. D., This Broken Archipelago: Cape Cod and the Islands, Amphibians and
Reptiles, Demeter Press, New York, 1976, 260.
51. Lazell, J. D., New England waters: critical habitat for marine turtles, Copeia,
290-295, 1980.
52. Burke, V. J. and Standora, E., Factors affecting stranding of cold-stunned juvenile
Kemp’s ridley and loggerhead sea turtles in Long Island, New York, Copeia, 1136,
1991.
53. Morreale, S. L., Meylan, A., Sadove, S., and Standora, E., Annual occurrence and
winter mortality of marine turtles in New York waters,/. Herpetol., 26, 301, 1992.
54. Bolten, A. B. and Martins, H. R., Kemp’s ridley captured in the Azores, Mar. Turtle
Newsl, 48, 23, 1990.
55. Marquéz, R., Synopsis of biological data on the Kemp’s ridley turtle Lepidochelys
kempi (Garman, 1880), NOAA Tech. Memo. NMFS-SEFSC-343, 91, 1994.
56. Meylan, A. B., Riddle of the ridleys, Hist., 95(11), 90, 1986.
57. Cornelius, S., The Sea Turtles of Santa Rosa National Park, Fondacion de Parcpies
Nacionales. Costa Rica. 65. 1986.
58. Reichart, H. A., Synopsis of biological data on the Olive ridley sea imiltLepidochelys
olivácea (Eschsholtz, 1829) in the western Atlantic, NOAA Tech Memo. NMFS-
SEFSC-336, 78, 4, 1993.
59. Plotkin, R T., Reproductive Behavior of the Olive Ridley Turtle (Lepidochelys oli
vácea) in the Eastern Pacific Ocean, Ph.D. dissertation, Texas A & M University,
College Station, TX, 1994.
60. Plotkin, R, Byles, R. A., and Owens, D. M., Ups and downs in the life of the Olive
ridley turtle: breeding versus post-reproductive diving behavior, in Proc. 15th A m u .
Sea Turtle Symposium, Keinath, J. A., Barnard, D. E., Musick, J. A., and Bell, B. A.,
Eds., NOAA Tech. Memo. NMFS-SEFC, 1996.
61. Limpus, C. J., The Flatback turtle, Chelonia depressa (Carman) in southwest Queen
sland, Australia, Herpetologica, 27, 434, 1971.
62. Walker, T. A. and Parmenter, C. J., Absence of a pelagic phase in the life cycle of
the fiatback turtle, Natator depressa (Carman), J. Biogeogr., 17, 275, 1990.
63. Limpus, C. J., Walker, T. A., and West, J., Posthatchling specimens and records from
the Australian region, in Proc. Marine Turtle Conservation Workshop, James, R.,
Compiler, Australian National Parks and Wildlife Service, Canberra, 1994, 86.
64. Limpus, C. J., Parmenter, C. J., Baker, V., and Fleay, A., The fiatback XmÚQ,Chelonia
Aust. Wildl. Res., 10, 557, 1983.
65. Frick, J., Orientation and behavior of hatchling green turtles (Chelonia my das) in the
sea, Anim. Behav., 24, 849, 1976.
66. Ireland, L. C., Frick, J. A., and Wingate, D. B., Nighttime orientation of hatchling
green turtles (Chelonia mydas) in the open ocean, in Animal Migration, Navigation,
and Homing, Schmidt-Koenig, K., and Keeton, W. J., Eds., Springer-Verlag, New
York, 1978, 420.
67. Wyneken, J. and Salmon, M., Frenzy and post-frenzy swimming activity in logger-
head, green and leatherback hatchling sea turtles, Copeia, 478, 1992.
68. Wyneken, J., Sea turtle locomotion: mechanisms, behavior and energetics, in Biology
o f Sea Turtles, Lutz, P. and Musick, J. A., Eds., CRC Press, Boca Raton, FL, 1996,
chap. 7.
69. Caldwell, D. K., Hatchling green sea turtles, Chelonia mydas at sea in the northeastern
Pacific Ocean, Bull. So. Calif. Acad. Sci., 68, 113, 1969.
70. Carr, A. and Meylan, A. B., Evidence of passive migration of green turtle hatchlings
in Sargassum, Copeia, 2, 366, 1980.
71. Bjomdal, K. A. and Bolton, A. B., Growth rates of immature green ixxvilts, Chelonia
mydas, on feeding grounds in the southern Bahamas, Copeia, 555, 1988.
72. Bjorndal, K. A. and Bolton, A. B., Comparison of length-frequency analyses for
estimation of growth parameter for a population of green turtles, Herpetologica,
51(2), 160, 1995
73. Keinath, J. A. and Musick, J. A., Atlantic green turtle Chelonia mydas (Linnaeus),
in Virginia's Endangered Species, Terwilliger, K., Coord., McDonald and Woodward,
Blacksburg, VA, 1991, 448.
74. Balazs, G. H., Growth rates of immature green turtles in the Hawaiian Archipelago,
in Biology and Conservation o f Sea Turtles, Bjomdal, K. A., Ed., Smithsonian
75. Carr, A., Heath, H., and Ogren, L., The ecology and migrations of sea turtles: the
hawksbill turtle in the Caribbean Sea, Am. Mus. Novit., 2248, 1, 1966.
76. Redfoot, W. E., Ehrhart, L. M., and Raymond, P. W., A juvenile Atlantic hawksbill
turtle Eretmochelys imbricata from Brevard County, Florida, FL Sci., 48, 193, 1985.
77. Parker, L. G., Encounter with a juvenile hawksbill turtle offshore Sapelo Island,
Georgia, Mar. Turtle NewsL, 71, 119, 1995.
78. Keinath, J. A., Musick, J. A., and Swingle, W. M., First verified record of the hawksbill
sea turtle (Eretmochelys imhricata) in Virginia waters, Catesbeina, 11(2), 37, 1991.
79. Witzell, W. N., Synopsis of the biological data on the Hawksbill imi\& Eretmochelys
imbricata (Linnaeus, 1766), E40 Eish. Synop., 137, 78, 1983.
80. Witzell, W. and Banner, A., The hawksbill turtle, Eretmochelys imbricata, in western
Samoa, Bull. Mar. Sci., 30(3), 571, 1980.
81. Boulon, R. H., Growth rates of wild juvenile hawksbill turtles Eretmochelys imbricata
in St. Thomas, United States, Virgin islands, Copeia, 811, 1994.
82. Carr, A. and Ogren, L., The ecology and migrations of sea turtles. 3: Dermochelys
in Costa Rica, Am. Mus. Novit., 1958, 1, 1959.
83. Caldwell, D. K., On the status of the Atlantic leatherback turtle,Dermochelys coriacea
coriacea as a visitant to Florida mating beaches, with natural history notes, Q. /. Ela.
Acad. Sci., 21(3), 285-291, 1959.
84. Brongersma, L. D., Miscellaneous notes on turtles. III,i^. Ned. Akad. Wet. Amsterdam
Ser. C, 73(4), 323, 1970.
85. National Marine Fisheries Service and U.S. Fish and Wildlife Service, Recovery Plan
for Leatherback Turtles in the U.S., Caribbean, Atlantic and Gulf of Mexico, National
Marine Fisheries Service, Washington, D.C., 65, 1992.
86 . Grant, G. S., Juvenile leatherback turtle caught by longline fishing in American
Samoa, Mar. Turtle Newsl., 66, 5, 1994.
87. Horrocks, J. A., Leatherbacks in Barbados, Mar. Turtle Newsl., 41, 7, 1987.
88 . Zug, G. R. and Parham, J. R, Age and growth in leatherback sea turtles,Dermochelys
coriacea (Reptilia, Dermochelyidae), Biol. Bull., in press.
89. Lutcavage, M., and Lutz, P. L., Metabolic rate and food energy requirements of the
leatherback sea turtle, Dermochelys coriacea, Copeia, 1986, 796, 1986.
90. Koblentz-Mishke, O. J., Volkovinsky, V. V., and Kaloanova, J. G., Plankton primary
production of the world oceans, in Scientific Exploration o f the South Pacific, Wooster,
W. S., Ed,, National Academy of Sciences, Washington, D.C., 1970, 183.
91. Ernst, C. H., Lovich, J. E., and Barbour, R. W., Turtles o f the United States and
Canada, Smithsonian Institution Press, Washington, D.C., 1944.
92. Shoop C. R. and Kenney, R. D., Seasonal distribution and abundances of loggerhead
and leatherback sea turtles in waters of the northeastern United States, Herpetol.
Monogr., 6, 43, 1992.
93. Byles, R. A., Behavior and Ecology of Sea Turtles from Chesapeake Bay, VA, Ph.D.
dissertation. College of William and Mary, Williamsburg, VA, 1988.
94. Keinath, J. A., Musick, J. A., and Byles, R. A., Aspects of the biology of Virginia’s
sea turtles: 1979-1986, Va. J. Sci., 38(4), 329, 1987.
95. Musick, J. A. and Keinath, J. A., unpublished, 1988.
96. Bowen, B. W., Richardson, J. I., Meylan, A. B., Margaritoulis, D., Hopkins-Murphy,
S. R., and Avise, J. C., Population structure of loggerhead turtles {Caretta caretta)
in the west Atlantic ocean and Mediterranean Sea, Conserv. Biol., 1, 834-844, 1993.
97. Norrgard, J., DeteiTnination of Stock Composition and Natal Origin of a Juvenile
Loggerhead Turtle Population (Caretta caretta) in Chesapeake Bay Using Mitochon
drial DNA Analysis, Master’s thesis. College of William and Mary, School of Marine
Science, Williamsburg, VA, 1995.
98 . Sears, C. J., Genetic Structure of a Local Loggerhead Sea Turtle Population Based
on Mitochodrial DNA Analysis, Masters thesis. University of Charleston, SC, 1993.
99. Sears, C. J., Preliminary genetic analysis of the population structure of Georgia
loggerhead sea turtles, in Proc. 14th Annu. Symp. Sea Turtle Biology and Conserva
tion, Bjorndal, K., Bolton, A. B., Johnson, D. A., and Elizar, P. J., Compilers, NOAA
Tech. Memo. NMFS-SEFSC-351, 135, 1994.
100. Keinath, J. A., Musick, J. A., and Barnard, D. E., Abundance and distribution of sea
turtles off North Carolina, Mineral Management Service, OCS Study, Virginia Insti
tute of Marine Science, Gloucester Point, VA, MMS 95-0024, 1, 1996.
101. Musick, J. A., Barnard, D., and Keinath, J. A., Aerial estimates of seasonal distribution
and abundance of sea turtles near the Cape Hatteras faunal barrier, in Proc. 13th
Annu. Symp. Sea Turtle Biology and Conservation, Schroeder, B. A. and Witherington,
H. E., Compilers, NOAA Tech. Memo. NMFS-SEFSC-341, 121, 1994.
102. Epperly, S. P., Braun, J., and Chester, A. J., Aerial surveys for sea turtles in North
Carolina inshore waters, Fish. Bull., 93, 254, 1995.
103. Epperly, S. R, Braun, J., Chester, A. J., Cross, F. A., Merriner, J. W., and Tester,
P. A., Winter distribution of sea turtles in the vicinity of Cape Hatteras and their
interaction with the summer flounder trawl f i s h e r y , Mar. Sci., 56(2), 547, 1995.
104. Mendonca, M. T. and Ehrhart, L. M., Activity, population size and structure of
immature Chelonia my das and C arena caretta in Mosquito Lagoon, Florida, Copeia,
I. 161, 1982.
105. Ehrhart, L. M., Marine turtles of the Indian River Lagoon System, F/. Sci., 46, 337,
1983.
106. LeBuff, C. R., The loggerhead turtle in the eastern G ulf o f Mexico, Caretta, Research
Inc., Sanibel, Florida, 216, 1990.
107. Hildebrand, H. H., A historical review of the status of sea turtle populations in the
western Gulf of Mexico, in Biology and Conservation o f Sea Turtles, Bjorndal, K.,
Ed., Smithsonian Institution Press, Washington, D.C., 447, 1982.
108. Keinath, J. A., Barnard, D. E., and Musick, J. A., Behavior of loggerhead sea turtles
in Savannah, Georgia, and Charleston, South Carolina shipping channels. Final report
to the U.S. Army Coips of Engineers, Vicksburg, MI, Contract Report FACW39-93-
C-0016, 75 pp. + appendices, 1995.
109. Carr, A., Ogren, L., and McVea, C., Apparent hibernation by the Atlantic loggerhead
turtle Caretta caretta off Cape Canaveral, Florida, Biol. Conserv., 19, 7~14, 1980.
110. Ogren, L. and McVea, C., Apparent hibernation by sea turtles in North American
waters, in Biology and Conservation of Sea Turtles, Bjorndal, K., Ed., Smithsonian
Institution Press, Washington, D.C., 127, 1982, 127.
111. Schwartz, F. J., Behaviorial and tolerance responses to cold water temperatures by
three species of sea turtles (Reptilia, Cheloniidae) in North Carolina, Fla. Mar. Rec.
Publ., 13, 16-18, 1978.
112. Lutz, P. L., Bergy, A., and Bergy, M., Effects of temperature on gas exchange and
acid-base balance in the sea turtle Caretta caretta at rest and during routine activity,
J. Exp. B iol, 144, 155-169, 1989.
113. Henwood, T. A., Movements and seasonal changes in loggerhead turtle, Caretta
caretta, aggregations in the vicinity of Cape Canaveral, Florida, Biol. Conserv., 40,
191, 1987.
114. Limpus, C. J., Fleay, A., and Guinea, M., Sea turtles of the Capricornia Section,
Barrier Reef, in The Capricornia Section o f the Great Barrier Reef: Past, Present
and Future, Ward, W. T. and Saeger, R, Eds., Royal Society of Queensland and
Australian Coral Reef Society, Brisbane, 61, 1984.
115. Limpus, C. J., The loggerhead turtle, Caretta caretta, in Queensland: feeding ground
selection following her first breeding season, NOAA Tech. Memo. NMFS-SEFC-
351, 78, 1994.
116. Limpus, C. J., The hawksbill turtle, Eretmochelys imbricata, in Queensland: popu
489, 1992.
117. Beardsly, R. C., Boycourt, W. C., and Hanson, D. Z., Physical oceanography of the
Middle Atlantic Bight, Am. Soc. LimnoL Oceanogr. Spec. Symp., 2, 20, 1976.
118. Rudloe, A., Rudloe, J., and Ogren, L., Populations of Atlantic ridley sea turtles
{Lepidochelys kempi) in Apalachee Bay, Florida, costal waters, in Proc. 9th Annu.
Workshop on Sea Turtle Conservation and Biology, Eckert, S. A., Eckert, K. L., and
Richardson, T. H., Eds., NOAA Tech. Mem. NMFS-SEFC-232, 151, 1989.
119. Schmid, J. R. and Ogren, L.H., Subadult Kemp’s ridley sea turtles in the southeastern
U.S.: results of long-tenn tagging studies, inProc. 11th Annu. Workshop on Sea Turtle
Biology and Conservation, Salmon, M. and Wyneken, J., Compilers, NOAA Tech.
Memo. NMFS-SEFSC 302, 14, 1992.
120. Morreale, S. J. and Standora, E. A., Occurrence, movement and behavior of the
Kemp’s ridley and other sea turtles in New York waters,/. HerpetoL, 26, 301, 1992.
121. Keinath, J. A., Barnard, D. E., Musick, J. A., and Bell, B. A., Kemp’s ridley sea
turtles from Virginia waters, in Proc. 14th Annu. Symp. Sea Turtle Biology and
Conservation, Bjorndal, K.A., Bolton, A. B., Johnson, P. A., and Eliazar, P. J.,
Compilers, NOAA Tech. Memo. NMFS-SEFSC-351, 70, 1994.
122. Standora, E. A., Morreale, S. J., Thompson, R. D., and Burke, V. S., Telemetric
monitoring of diving behavior and movements of juvenile Kemp’s ridleys, in Proc.
10th Annu. Workshop on Sea Turtle Biology and Conservation, Richardson, T. H.,
Richardson, J. I., and Donnely, M., Compilers, NOAA Tech. Memo. NMFS-SEFC-
278, 133, 1990.
123. Henwood, T. A. and Ogren, L. H., Distribution ann migration of immature Kemp’s
ridley turtles {Lepidochelys kempi) and green turtles {Chelonia mydas) off Florida,
Georgia and South Carolim, Northeast Gulf Sci., 9(2), 153, 1987.
124. Rudloe, A., Rudloe, J., and Ogren, L., Ocurrence of immature Kemp’s ridley turtles,
Lepidochelys kempi, in coastal waters of northwest ¥\oúáa. Northeast G ulf Sci., 12(1),
49, 1992.
125. Pritchard, P. C. H., Post-nesting movement of marine turtles (Cheloniidae and Der-
rnochelidae) tagged in the Guianas, Copeia, 1976, 749, 1976.
126. By les, R. and Plotkin, P, Comparison of the migratory behavior of the congeneric
sea turtles Lepidochelys olivácea and L. kempi, in Proc. 13th Annu. Symp. Sea Turtle
Biology and Conservation, Schroeder, B. A. and Witherington, H. E., Compilers,
NOAA Tech. Memo. NMFS-SEFSC-341, 39, 1994.
127. Harris, A., Species review: olive ridleys. in Proc. Marine Turtle Conservation Work
shop, James, R., Comp., Australian National Parks and Wildlife Service, Canberra,
1994, 58.
128. Poiner, I. R. and Hams, A. N. M., The incidental capture and mortality of sea turtles
in Australia’s northern prawn fishery, in Proc. Marine Turtle Conservation Workshop,
James, R., Comp., Australian National Parks and Wildlife Service, Canberra, 1994,
115.
129. Epperly, S. A., Braun, J., and Veishlow, A., Sea turtles in North Carolina waters,
Conserv. Biol., 2, 384, 1994.
130. Guseman, J. L. and Ehrhart, L. M., Green turtles on sabellariid worm reefs: initial
results from studies on the Florida Atlantic coast, in Proc. 10th Annu. Workshop on
Sea Turtle Biology and Conservation, Richardson, T. H., Richardson, J. L, and
Donnely, M., Compilers, NOAA Tech. Memo. NMFS-SEFC-278, 125, 1990.
131. Felger, R. S., Clifton, S. K., and Regal, R S., Winter dormancy in sea turtles:
independent discovery and exploitation in the Gulf of California by two local cultures.
Science, 191, 283, 1976.
132. Mendonca, M. T., Movements and feeding ecology of immature green turtles (C/?c-
lonia my das) in a Florida lagoon, Copeia, 4, 1013, 1983.
133. Bjorndal, K. A. and Bolton, A. B., Developmental migrations of juvenile green turtles
in the Bahamas, in Proc. 15th Annu. Symp. Sea Turtle Biology and Conservation,
Keinath, J. A., Barnard, D. E., Musick, J. A., and Bell, B. A., Compilers, NOAA
Tech. Memo. NMFS-SEFSC-387, 38, 1996.
134. Green, D., Growth rates of wild immature green turtles in the Galapagos Islands,
Ecuador,/. Herpetol, 27(3), 338, 1993.
135. Limpus, C. J. and Walter, D. G., The growth of immature green turtles (Chelonia
mydas) under natural conditions, Herpetologica, 36, 162, 1980.
136. Meylan, A. B., Spongivory in hawksbill turtles: a diet of g\‘a.s^. Science, 239, 393,
1988.
137. Broderick, D., Moritz, C., Miller, J. D., Guinea, M., Prince, R., and Limpus, C. J.,
Genetic studies of the Hawksbill ixxnlQ Eretmochelys imhricata: evidence for multiple
stocks in Australian waters, Pac. Conserv. Biol., 4, 124, 1994.
138. Boulon, R. H., Virgin Island turtle tag recoveries outside the U.S. Virgin Islands, in
Proc. 9th Annu. Workshop on Sea Turtle Conservation and Biology, Eckert, S. A.,
Eckert, K. L., and Richardson, T. H., Compilers, NOAA Tech. Memo. NMFS-SEFC-
232, 215, 1989.
139. Hillis, Z. M., The hawksbill turtles of Buck Island Reef National Monument: a shared
resource of the Caribean, Proc. 14th Annu. Symp. Sea Turtle Biology and Conserva
tion, Bjorndal, K.A., Bolton, A. B., Johnson, R A., and Eliazar, P. J., Compilers,
NOAA Tech. Memo. NMFS-SEFSC-351, 59, 1994.
140. Barnard, D. E., Keinath, J. A., and Musick, J. A., Distribution of ridley, green and
leatherback turtles in Chesapeake Bay and adjacent waters, mProc. 9th Annu. Work
shop on Sea Turtle Conservation and Biology, Eckert, S. A., Eckert, K. L., and
Richardson, T. H., Compilers, NOAA Tech. Memo. NMFS-SEFC-232, 201, 1989.
141. Boulon, R., Leatherback turtle (Dermochelys coriacea) nesting biolgy Sandy Point,
St. Croix: 1981-1990, in Proc. 11th Annu. Workshop on Sea Turtle Biology and
Conservation, Salmon, M. and Wynhken, J., Compilers, NOAA Tech. Memo. NMFS-
SEFC 302, 14, 1992.
142. Knowlton, A. R. and Weigle, B., A note on the distribution of leatherback turtles
(Dermochelys coriacea) along the Florida Coast in February 1988, inProc. 9th Annu.
Workshop on Sea Turtle Conservation and Biology, Eckert, S. A., Eckert, K. L., and
Richardson, T. H., Compilers, NOAA Tech. Memo. NMFS-SEFC-232, 83, 1989.
143. Musick, J. A., unpublished, 1989.
144. Prescott, R. L., Leatherbacks in Cape Cod Bay, Massachusetts 1977-1987, in Proc.
8th Annu. Workshop on Sea Turtle Conservation and Biology, NOAA Tech. Memo.
NMFS-SCFC-214, 83, 1988.
145. Paladino, F. V, O’Connor, M. R, and Spotila, J. R., Metabolism of leatherback turtles,
gigantothermy, and thermoregulation of dinosaurs, Vaiwrc, 344, 959, 1990.
146. Block, D. A., Finnerty, J. R., Stewart, A. F. R., and Kidd, J., Evolution of endothermy
in fish: Mapping physiological traits on a molecular ^hylogtny, Science, 260, 210,
1993.
147. Ryther, J. H., Photosynthesis and fish production in the sea, Science, 166, 72, 1969.
148. Stancyk, S. E., Non-human predators of sea turtles and their control, in Biology and
Conservatioin o f Sea Turtles, Bjorndal, K., Ed., Smithsonian Institution Press, Wash
ington, D.C., 139, 1982.
149. Bjorndal, K. A., Foraging, ecology and nutrition in sea turtles, in Biology o f Sea
Turtles, Lutz, P. and Musick, J. A., Eds. CRC Press, chap. 8.
150. Loop, K., Miller, J. D., and Limpus, C. J., Hawksbill turtle nesting on Milman Island,
Great Barrier Reef, Australia, Wild. Res., 22, 241, 1995.
151. Pritchard, P. C. H., Nesting of the leatherback turtle Dermochelys coriacea in Pacific
Mexico, with a new estimate of the world population status, Copeia, 3, 741, 1982.
152. Sternberg, J., The Worldwide Distribution of Sea Turtle Nesting Beaches, Center for
Environmental Conservation, Washington, D.C., 1981.
153. Pritchard, P. C. H., International migrations of South American sea turtles Chloniidae
and Dermochelidae, Am. Behav., 21, 18, 1973.
154. Keinath, J. A. and Musick, J. A., Movements and diving behavior of a leatherback
sea turtle, Dermochelys coriacea, Copeia, 4, 1010, 1993.
V Sea Turtle Locomotion:
Mechanisms, Behavior,
and Energetics
Jeanette Wyneken
CO NTENTS
7.1 Overview........................................................................................................ 166

1 2 Tenestrial Locomotion................................................................................... 167
7.2.1 Locomotion by Hatchlings — Quadrupedal Patterns...................... 168
1 2 .1.1 The Cheloniid Pattern........................................................ 168
7.2.1.2 The Dermochelid Pattern................................................... 168
7.2.2 Adult and Subadult Locomotion on Land........................................ 169
7.2.2.1 Chelonia.............................................................................169
1.2.2.2 Caretta, Eretmochelys, and Lepidochelys......................... 170
7.2.2.3 Natator............................................................................... 170
7.2.2.4 Dermochelys....................................................................... 170
7.3 Aquatic Locomotion...................................................................................... 171
7.3.1 Overview of Locomotor Adaptations............................................... 171
7.3.1.1 Flippers and the Supporting Skeleton............................... 172
7.3.1.2 Body Form......................................................................... 172
7.3.2 Thrust Production and Steering........................................................ 174
7.3.2.1 Lift-Based Locomotion...................................................... 174
1.2.2.2 Drag-Based Locomotion.................................................... 174
7.3.2.3 Mechanisms of Thrust Production during
Powerstroking.................................................................... 176
7.3.2.4 Steering.............................................................................. 180
7.3.3 Species-Specific Patterns of Swimming........................................... 180
7.4 Migratory Behavior of Hatchlings andPelagic LifeStages..........................181
7.4.1 Swimming Characteristics during the Frenzy and
Postfrenzy Periods............................................................................ 181
7.4.2 Swimming Characteristics of Pelagic Stages................................... 185
7.5 Migratory Swimming Behavior ofSubadultsand Adults...............................186
7.5.1 Swimming by Cheloniids................................................................. 186
7.5.2 Swimming by Dermochelys coriácea.............................................. 188
0-8493-8422-2/97/$0.00+$ 50
7.6 Energetic Adaptations for Migratory Swimming......................................... 189

7.6.1 Aerobic Adaptations and Behavioral Correlates.............................. 189
7.6.1.1 Hatchlings during Frenzy and Postfrenzy Swimming..... 189
7.6.1.2 Juveniles, Subadults, and Adults....................................... 190
7.6.2 Anaerobic Adaptations and Migratory Swimming.......................... 192
7.7 Conclusions....................................................................................................193
References...............................................................................................................194
7.1 O V E R V IE W
Marine turtles depart functionally and moiphologically from other chelonian lin
eages. Their form of locomotion and of their locomotor adaptations, perhaps more
than any other features, make sea turtles unique from other turtles. Very large adult
size, hypertrophied phalanges, distinctive locomotor patterns, relatively high activity
levels, and their nearly totally aquatic life-style are all deviations from the historical
emphasis of this morphologically conservative vertebrate group. Consequently, sea
turtles have attracted the attention of many authors who described, listed, or inves
tigated the adaptations that sea turtles possess for their marine life-style.
In general, the locomotion of turtles has unique features when compared with
that of other vertebrates. Because of the rigid carapace, all propulsion must come
from the limbs. While all terrestrial turtles and freshwater turtles (with the exception
of Carettochelys) utilize diagonally alternate limb movements to swim or crawl,^'^
marine turtles use both forelimbs, simultaneously swept through the water during
swimming.^’^®Though this swimming motion is technically described as an asym
metrical gait (because the forelimbs move simultaneously as a pair rather than a
forelimb and hindlimb acting as a pair; see Figure 7.1), the term is counterintuitive
when describing sea turtle locomotion. So for this discussion I will abandon typical
“gait jargon” and instead use terminology prevalent in the behavioral literature. Thus,
the most common locomotor pattern used by sea turtles is the “powerstroke,”^* ’^ in
which the forelimbs are swept through the water simultaneously, like flapping wings.
When covering distances beyond several body lengths, all sea turtles utilize the
powerstroke to produce thrust. During their post-hatching ontogeny, cheloniid sea
turtles employ a “dogpaddling” stroke (symmetrical gait) in which diagonally oppo
site limbs move together to produce thrust.^®’^^ Similarly, following offshore migra
tion, hatchlings and post-hatchlings primarily rely upon the a “rear flipper kick”
while they reside in the pelagic environment. In contrast, dermochelids
utilize the powerstroke almost exclusively and do not undergo similar ontogenetic
changes in locomotion.
Three major themes dominate any consideration of sea turtle locomotion. First,
cheloniids undergo ontogenetic changes in their predominant form of thrust produc
tion. These changes are closely linked to behavioral and ecological shifts. Under
most circumstances, the forelimbs have primary responsibility for thrust production.
Yet, all 4 limbs work to generate thrust during some phases of life in young sea
turtles. Second, during terrestrial locomotion, all four limbs are used to generate
Sea Turtle Locomotion: Mechanisms, Behavior, and Energetics 167
Symmetrical Gait
Diagonally opposite limbs move together
Asymmetrical Gait
Contralateral limbs move together
FIGURE 7.1 Diagrammatic representation of turtle gaits. Gaits are defined as symmetrical
or asymmetrical by both the duration of foot contact with the ground (or other foot action,
as in swimming) and by the phase relationship between the fore- and hindfeet.'^ These
designations are cumbersome when applied to sea turtle locomotion. Symmetrical gaits (upper
diagram) are typical of the locomotion by most nonmarine turtles as well as marine turtle
dogpaddling and crawling by Caretta, Eretmochelys, Lepidochelys, and hatchling Chelonia
and Natator. The powerstroke is an asymmetrical gait. Asymmetrical gaits (lower diagram)
are also used by adult Chelonia and Dermochelys during crawling.
thrust in all turtle species. However, the details of sea turtle terrestrial locomotion
make it distinct. Third, migratory behavior dominates sea turtle life. Morphological,
behavioral, and physiological adaptations are clearly related to efficient, long-dis
tance movement through the water (particularly during migration).
7.2 TER R ESTR IA L L O C O M O T IO N

The terrestrial locomotion of sea turtles, particularly adults, is often described as
clumsy^ or labored. These animals spend virtually their entire lives in water with
their weight supported by this medium, but when on land, they no longer have that
support.
Most hatchlings (with the exception of Dermochelys ) crawl by synchronous

movements of diagonally opposite limbs (i.e., left front-right rear). As the turtles
increase in mass, their locomotion becomes taxonomically dichotomous. Caretta,
Eretmochelys, and Lepidochelys crawl using a gait similar to that used as hatchlings.
Chelonia, Dermochelys, and, to a large extent, Natator protract, then retract, all four
limbs simultaneously while crawling (a “crutching” movement). Adult Natator, a
generalist in its locomotion, may use both gaits under different circumstances.^^
7.2.1 L ocomotion by H atchlings — Q uadrupedal Patterns

7.2.1.1 The Cheloniid Pattern
Cheloniid hatchlings crawl by protracting and retracting diagonally opposite limbs

(Figure 7.2). Thrust is produced when the palmar surface of the manus, the anatom
ical anterior edges of the manus and crus (front flipper), and the plantar surfaces of
the hind flippers, transfer force to the ground.
FIGURE 7.2 Crawling by a Caretta hatchling is typical of the locomotion of all other
cheloniid hatchlings. Diagonally opposite limbs move together. Tracings were made from
ciné recordings filmed at 32 fps. Elapsed time (t) is given in seconds to the left of each tracing.
7.2.1.2 The Dermochelid Pattern

Hatchling D. coriacea typically use a “swing and stance” limb c y c l e . T h e fore
limbs are simultaneously protracted, then repositioned once they touch the ground;
the body is then lifted as the limbs sweep back (Figure 7.3) to propel the body
forward. The hindlimbs move in unison with the forelimbs so that all four flippers
are protracted, then retracted, pushing back together on the sand.^^ Hindlimbs are
protracted, planted on the ground, and then retracted, pushing backward. Toward
FIGURE 7.3 Crawling by a Dermochelys hatchling using a “swing-and-stance” pattern of

movement. Unlike locomotion by cheloniid hatchlings, contralateral limbs move together.
Successive events are labeled 1 to 6 to the left of each drawing. (Redrawn from Renous, S.,
Mesogee, 48, 79-84, 1988. With permission.)
the end of retraction, the hind flippers are abducted and dragged as the forelimbs
complete their retraction. Unlike other turtles, the hatchling leatherback gait
includes a period when all four flippers are in the air simultaneously while the
ventrum supports the body. When D. coriacea is placed on a smooth surface it can
switch to use diagonally opposite limbs in an alternating pattern. This gait was
observed only under experimental conditions.
7.2.2 A dult a nd S ubadult Locom otion on La n d
Sea turtle crawling differs kinematically from mechanisms used by other turtle
species. The “crutching” movements seen in adult C. mydas, N. depressa, and D,
coriacea are unique within the Testudines. Adults of other cheloniid species employ
quadrupedal gaits in which diagonally opposite feet move as a pair. As marine turtles
approach adulthood and their mass increases, they support themselves on the carpus
and anterior edge of the manus rather than on the palmar surface.
7.2.2.1 Chelonia
Adult female green turtles typically come ashore only to nest, while males remain
permanently aquatic. However, at isolated locals, such as the remote northwestern
Hawaiian Islands (U.S.), green turtles of both sexes (adults as well as immature
turtles) crawl ashore to bask on oceanic beaches.^® Along coastal North Carolina,
juvenile C. mydas and L. kempi have been observed crawling on land,^°' presumably
to bask.
Adult turtles crawl by using simultaneous protraction, then retraction, of the

forelimbs aided by pushing of the rear limbs. Typically, the fore and hind flippers
are protracted and retracted in unison. However following nesting, the rear flippers
are occasionally used in an alternating fashion while the forelimbs continue to be
moved together. During more rapid land-based locomotion green turtles may use an
alternating limb pattem.^’*^^’^^^^ Under most conditions, this “crutching” motion lifts
the anterior carapace and plastron while the posterior part of the body is dragged
or lifted only slightly.
7.2.2.2 Caretta, Eretmochelys^ and Lepidochelys
All adult and subadult Caretta, Eretmochelys, and Lepidochelys move diagonal pairs
of limbs to produce thrust. This pattern is often described as a “typical” quadrupedal
or tetrapod gait, though it differs by lacking any axial undulations and both the
manus and the crus support the weight of the body.
7.2.2.3 Natator
Adult flatback turtles appear to rely on both methods of terrestrial locomotion,
depending upon circumstances. Bustard^ reports that this species uses simultaneous
movements of all 4 limbs when ascending the beach, but switches to thrust with
diagonal limbs moving together when crawling up a dune face.
7.2.2.4 Dermochelys
Only adult female leatherback turtles come ashore and only when ready to nest.
Detailed studies of terrestrial locomotion rely upon direct observation and films of
nesting emergences, and upon films of a juvenile raised in captivity.'^ Adult and
juvenile leatherback turtles use crutching movements of the forelimbs, with simul
taneous pushing by the hind limbs. This “swing and stance” gait of the adults and
juveniles is very similar to the locomotor pattern used by hatchlings (Figure 7.3).
However, leatherback turtles are massive and do not lift the body far, if at all, from
the sand.^^ The adult gait involves support of the weight of the animal on the plastron
while all 4 flippers are protracted in the air; the turtle then moves forward in a series
of lunges (Figure 7.4). A single captive-raised juvenile, crawling under experimental
conditions, occasionally used its hind limbs in an alternating fashion for a few steps.
In nature, juveniles are not known to crawl ashore.
Nesting leatherback turtles leave a sinusoidal track when crawling ashore.
Pritchard^^ suggests that twisting in the track is due to the efforts of the animal to
minimize the angle it has to climb “uphill” to nest. Another possible explanation is
that since the rear flippers are occasionally used alternately, the direction of push
(the thrust vector) from the rear could torque the body of the animal around as it
moves forward.
FIGURE 7.4 Crawling by an adult Dermochelys showing the “swing-and-stance” pattern

of movement. Adult crawling differs from that of a hatchling in that the weight of the body
is supported by the plastron while all four limbs are protracted (drawings 1 to 3). Successive
events (0.055 s apart) are labeled 1 to 7 to the left of each drawing. (Redrawn from Renous,
S., Mesogee, 48, 79-84, 1988. With pemiission.)
7.3 A Q U A T IC L O C O M O T IO N
7.3.1 O verview of Locom otor A daptations
Structural and adaptive diversity exist among marine turtle species along with con
siderable conservatism. Specifically, the form of the body and limbs is modified
from the basic turtle body design. Marine turtle body form is characterized by
reduction in the axial skeletal elements and increased fontanels in the carapace and
plast ron, a heavy head with a nonretractable neck,^^’^"^ and minimal cervical, axial,
and inguinal pouches. The pectoral limbs are modified into wing-like fore flippers
and the hind flippers have become paddle-like or rudder-like. Functionally, sea turtles
have exploited a mechanical system for propulsion that is not utilized by other turtle
groups. For most of their swimming activities, cheloniid and dermochelid sea turtles
use a largely lift-based mechanism of producing t h r u s t . T h e flippers
serve as wings (or propellers) as well as paddles. Two drag-based mechanisms,
“dogpaddling” and “rear flipper kicking,” are also in the repertoire of hatchling and
juvenile cheloniid locomotor patterns. However, these are reserved for activities
other than covering long distances quickly and efficiently.
The body fonri of sea turtles is highly streamlined when compared with semi-
aquatic and freshwater turtles.Streamlining contributes to efficiency while swim
ming by minimizing drag (the resistance to moving through the water).
7.3.1.1 Flippers and the Supporting Skeleton
Elongate flippers are typieal of secondarily aquatic vertebrates that propel themselves
with paired appendages (e.g., seals, penguins, marine turtles, and plesiosaurs).
The modification of the pectoral appendage from a limb with a typieal foot (Figure
7.5A) to an elongated, semirigid flipper Figure 7.5B) is, perhaps, the most striking
specialization of sea turtles. The radius and ulna are functionally fused by thick
connective tissue. The digits are elongated and made stiff by layers of fibrous
connective tissue overlaid by a scaled skin. The procoracoid of the pectoral apparatus,
located deep to the pectoralis and supracoracoideus muscles, is greatly expanded to
accommodate the proportionately greater muscle mass which retracts, abducts, and
rotates the flipper (Figure 7.5C, D).
The rear flippers are neither elongated nor particularly stiff. The interdigital
musele and connective tissue, as well as the overlying connective tissue, make the
hind flipper appear to be a paddle, though it functions both as a paddle in thrust
productions^ and as a rudder or elevator in steering.^^
Walker^^^ made qualitative comparisons between the limb skeletal components
of marine and aquatic turtles. The musculoskeletal system has been described in
detail by several investigators.^^’^s-34 orientation of the skeletal elements in the
fore- and hindlimbs of a crawling leatherback turtle was described by extrapolation
from radiographs of posed preserved limbs.
7.3.1.2 Body Form

Compared with nonpelagic turtles, the marine turtle shell is elongated along the
antero-posterior axis and streamlined. The axial skeleton is modified as described
above and the body has minimal cervical, axial, and inguinal pouches, undoubtedly
associated with minimizing drag.^^'^"^ All are features that have been associated with
increased emphasis of the pelagic life-style and well-developed migratory behavior.
Aleyev^^ described hydrodynamic features of the carapace in subadult sea turtles.
However, because he analyzed the shells alone (after removal of flippers and head),
only a limited assessment of streamlining was possible. Wyneken^^^ and Wyneken
and LaBarbera^*^ examined drag-reducing features of the body using dead hatchlings
without front flippers (the thrust producers). They showed that hatchling Chelonia
mydas, Caretta caretta, and D. coriacea accomplish drag reduction by streamlining
of body form, relatively low skin friction, and possibly through adjusting body
attitude while swimming. Logan and Morreale^^^ measured juvenile C. caretta, Che
lonia mydas, and L. kempi and calculated drag based upon Reynolds number. They
determined that drag was low in these animals and that loading of the carapace with
epibionts (common in juveniles) increased drag as well as the energy expended on
long migrations.
Radius Humerus BR INF

TR-H
FIGURE 7.5 The musculoskeletal system of a representative freshwater turtle (Trachemys

scripta) and marine turtle (Carena caretta). (A) The forelimb skeleton of T. scripta has
relatively short digits and long antebrachium whose elements (radius and ulna) are not
functionally fused. (B) The forelimb skeleton of C. caretta has relatively long digits and a
short antebrachium whose elements are functionally fused. (C and D) The superficial muscles
of the forelimbs. The muscles involved in forelimb retraction and adduction (e.g., biceps
superficialis [BS-L], pectoralis [PECT], and the posterior part of the supracoracoideus [not
shown]) are relatively small in T. scripta (C) when compared with those of C. caretta. The
protractors and abductors: deltoidius (DL) and the anterior part of the supracoracoideus (SUP
COR-A) are also hypertrophied. Other muscles shown act to flex, extend, and rotate the
antebrachium and manus. The black scale bar = 3 cm. Other abbreviations: ACL: acromio
coracoid ligament, ACR: acromion process; BR INF: brachialis inferior; BS-L: biceps super
ficialis — lateral part; TR-H triceps brachii — humeral head; TR-S triceps brachii — scapular
head. (Modified from Wyneken, J., Comparative and Functional Considerations of Locomo
tion in Turtles, Doctoral dissertation. University of Illinois, Champaign-Urbana, Illinois, 1988,
151 pp.)
7.3.2 T hrust P roduction and S teering
Sea turtles swim vigorously underwater using their forelimbs to provide thrust. The
synchronous sweeping of the flippers causes the axial body to rock so that it cycles
through positive and negative angles of attack, (but does not reach the point of stall).
During normal swimming all species avoid stall.
Sea turtle hat chl ingsand juveniles^^ differ from other marine migrants by
operating at moderate Reynolds numbers (Re = 1.1 to 4.6 x 10"^).^^ The Re for
hatchlings and juveniles are similar to those of small salmonids'^® but lower than
those recorded for many other aquatic migrants (e.g., tuna,^^ 3 x 10^; salmonid fish,"^®
2.5 X 10^ to 5.2 X 10^; penguins,^^’'^^ 1.26 to 1.76 x 10^; dolphins:'^^ 6.5 x 10^;
p o rp o is e s ,1.3 x 10^; whales,^^ 3 x 10^; and sea lions,'^'^ 2.03 to 2.87 x 10^). All
else being equal, similar Re imply similar flow conditions (and comparable relation
ships of inertial and viscous forces).
7.3.2.1 Lift-Based Locomotion

Lift-based mechanisms of thrust production are those in which the locomotor appa
ratus acts as a wing to generate lift forces. The orientation of the wing section with
respect to flow defines its action. At low angles of attack it produces lift which may
serve as thrust; at higher angles it may stall, countering propulsion by generating
high drag. Because lift forces are produced perpendicular to local flow and regardless
of the orientation of the wing with respect to gravity, they can serve as thrust. Sea
turtles use their foreflippers like wings, then paddles, during large portions of the
powerstroke (Figure 7.6), with lift arising from its convex surfaces. Net lift vectors
are oriented in such a way that they serve in propulsion (Figure 7.7).
7.3.2.2 Drag-Based Locomotion

Vertebrates that generate thrust by rowing or paddling use drag-based forms of
l o c o mo t i o n . T h e limb acts as a paddle with its blade perpendicular to the line of
movement through the water (Figure 7.8). For example, when a freshwater turtle
swims, the retracting limbs encounter resistance (drag) as they push back against
the water (Figure 7.9). The body moves forward relative to the paddle. The blade
is feathered during the recovery stroke (protraction) to minimize fonnation of coun
terproductive drag forces.
Drag produces thrust during (a) rear-flipper kicking, (b) part of the limb-retrac
tion phase of the powerstroke, and (c) much, if not all of the dogpaddle. During
their first few days in the water (the swimming frenzy period when hatchlings swim
hyperactively to distance themselves from s h o r e ) , t h e dogpaddle (Figure 7.10)
is used in alternation with the mostly lift-based powerstroke. When hatchlings are
breathing they use the dogpaddle stroke to elevate the anterior body and continue
making forward progress with the head extended above the surface of the water.
Following the swimming frenzy period, hatchling cheloniid sea turtles switch the
dominant mode of locomotion from the powerstroke to the drag-based rear flipper
kick and the d o g p a d d l e . T h e rear flipper kick consists of paddling or rowing
2 12
3
13
14
15
16
9 19
10 20
I
FIGURE 7.6 Powerstroking by a Chelonia mydas hatchling. Dark vertical bars to the left
of the tracings indicate where the forelimbs are undergoing simultaneous protraction. Open
vertical bars to the left of the tracings indicate that the forelimbs are undergoing simultaneous
retraction. The swimming turtle was filmed at 58 fps. Each tracing comes from consecutive
frames that are 0.017 s apart. Lift-based propulsion, at the end of protraction and start of
retraction, creates the greatest speeds. (Redrawn from Wyneken, J., Comparative and Func
tional Considerations of Locomotion in Turtles, Doctoral dissertation, University of Illinois,
Champaign-Urbana, Illinois, 1988, 151 pp.)
involving the hind flippers alone (Figure 7.11); the foreflippers (which play no role
in thrust production) are flexed and folded back over the carapace.
Lift i Lift 4- Drag

I
I
I
FLOW
FIGURE i n Diagrammatic explanation of lift forces that serve as forward thrust. Lift is
the force that results from pressure differences over a hydrofoil (such as a flipper). Lift is
perpendicular to the direction fluid moves over the foil and is perpendicular to drag (upper
diagram). A foil that is properly oriented relative to the flow can generate forward thrust
(lower two diagrams).
7.3.2.3 Mechanisms of Thrust Production during

Powerstroking
The predominant way in which sea turtles swim is by sweeping the foreflippers
roughly up and down as a pair (Figure 7.7). This motion, termed powerstroking in
most contemporary l i t e r a t u r e , h a s been assigned many other terms, some
more cumbersome than others (e.g., Parker: ‘‘submarine flight”; Carr and Ogren:^^
“the breast stroke”; Robinson:^® “subaqueous flight”; Davenport et al.:^^ “synchro
nous flapping of the foreflippers”). The powerstroke is a highly efficient gait that
produces propulsion during all phases of the stroke (upward-outward, then down-
ward-inward). During powerstroking, the hindlimbs usually do not contribute
thrust. Occasionally, some hatchlings coordinate slight simultaneous retraction of
r " Effective
irectionof
Direction of
c
W ÊÊÊÊÊKÊÊÊÊÊÊÊÊÊÊÊÊtÊÊÎ^fo^
Paddle Movement Water Flow
B
Effective Direction
^Direction of
of
Paddle Movement
Water Flow
FIGURE 7.8 Diagrammatic explanation of drag-based locomotion and drag forces that
serve as thrust. Drag-based propulsion involves orienting a paddle so that the blade is per
pendicular to the direction of paddle movement. (A) The blade of the paddle or flipper
encounters more resistance to moving through the water (drag) than the body does. (B) The
paddle blade or flipper during retraction in which it is feathered to minimize surface area.
(C) Propulsion (P) is generated on retraction. Protraction of the limb represents the recovery
(R) stroke.
the hindflippers with the downward-inward movement of the foreflippers, and

undoubtedly gain some thrust from this movement.
Walker^^ and Mordvinov"^^ both analyzed the mechanics of the powerstroke, but
both made fundamental errors in their identification of the appropriate mechanical
model by confusing wings (a lift-based propulsion) with paddles (drag-based pro
pulsion). Davenport et al.^^ identified the powerstroke as a lift-based mechanism of
thrust production. They measured forces from powerstroking juvenile green turtles
that were rigidly tethered in place. Their data showed forward thrust during all of
retraction (the “downstroke”) plus all but the end of protraction (when backwards
thrust was recorded). Wyneken^® used kinematic analyses of free-swimming Caretta
FIGURE 13 Drag-based propulsion by a swimming semiaquatic turtle was filmed as the

turtle used a symmetrical gait. Diagonally opposite limbs were protracted, then retracted
together, and act as paddles. The right front and left rear limbs of the turtle act as a couple
and start retraction together, producing forward thrust. The distal elements of the limbs are
then flexed to minimize surface area as they are brought forward. Time elapsed (in seconds)
is given to the left of each tracing.
and Chelonia hatchlings and found that the powerstroke consisted of both lift- and
drag-based components in series. Her data showed little forward thrust during the
first part of protraction, but at the end of protraction forward thrust was produced
as the flipper blade rotated through anteromedial extension. Forward thrust from lift
was also produced during the first half of retraction (when the blade twists and starts
its ventral sweep). During the second half of retraction, thrust was produced by the
flipper acting like a paddle (drag-based propulsion). Wyneken’s^®kinematic analyses
suggest that no thrust is produced during early protraction and the animal is carried
forward by momentum.
Sea Turtle Locomotion; Mechanisms, Behavior, and Energetics 179
t= 0 t = 0.34
t = 0.7 t = 0.45
t = 0.14 t = 0.62
t = 0.21 t = 0.67
t = 0.28
t = 0.74
FIGURE 7.10 Dogpaddling by a hatchling green turtle. Drag-based locomotion involves

all 4 limbs acting as oars or paddles. The flipper blade is swept ventrally and posteriorly, then
protracted with the anterior edge oriented forward, minimizing surface area. The turtle was
filmed at 32 fps. Elapsed time (in seconds) is shown for the sequence which begins at the
upper left and ends at the lower right.
FIGURE 7.11 Rear flipper kicking by a hatchling loggerhead sea turtle. Tracings show
synchronous protraction and retraction of the hind flippers. The hind limbs acts as paddles
and thrust production is drag-based. The turtle was filmed at 32 fps. Each tracing was made
at intervals of five frames.
In summary, Davenport et al.^^ and Wyneken^^^ came to slightly different con

clusions about the mechanical nature of the powerstroke. However, both the tech
niques as well as the ontogenetic stages studied differed substantially; thus the
differences are not surprising. Davenport et al.^^ studied juvenile C. mydas that were
rigidly tethered for force measurements; Wyneken*^ examined free-swimming
hatchling Caretta caretta and Chelonia mydas). The matter will only be resolved
when similar methods are used to study each life history stage.
The key feature of the powerstroke is that thrust is produced by all or virtually
all phases of forelimb movement. Prange^^ measured an average sustained thrust of
91 N (range: 66.7 to 364.8 N) from 10 adult green turtles as they swam away from
their nesting beach. Tucker et al."^^ measured the pull of two adult loggerheads as
they swam away from their nesting beach. Using their data, I calculated a thrust
estimate of 89 to 200 N. Davenport et al.^^ measured 0.43 to 1.00 N of thrust from
captive juvenile green turtles (523 to 1130 g) tethered to a force transducer via a
stiff element. Both the kinematic analyses and empirical measurements show that
the powerstroke can serve to generate stronger thrust (and faster swimming speeds)
than drag-based paddling. The powerstroke is a very efficient mode of locomotion
and very unlike the swimming gait of any other turtle species.
7.3.2.4 Steering
Turning can be accomplished by either forelimb movements alone, or by forelimb

movements combined with the rudder-like action by the hind limbs. Asynchrony in
the right and left forelimb movements was observed during turning by leatherbacks.^^
The duration of limb “flapping” movements becomes shorter on the side toward
which the turtle was turning. Hatchling Caretta use a combination of forelimb
propulsion and hindlimb rudder action in order to turn.^'^ '’®
7.3.3 S pecies-S pecific P atterns of S wimming
The form of the powerstroke is superficially similar in cheloniids and dermochelid

sea turtles, but differs in the details of the limb c y c l e . I n cheloniids {Caretta
and Chelonia) the powerstroke involves simultaneous protraction (anterior sweep)
of the forelimbs during which they move from a ventromedial position through a
laterally extended arc, then anteriorly. The flipper blades end in an anterodorsal
position (the “upstroke”). The forelimbs then rotate so that the preaxial edge faces
ventrally, before retraction begins. During retraction, the forelimbs sweep ventrally
through an anterodorsal to posteroventral arc, producing a “downstroke”^^^-'’ (Figure
7.6).
Deraniyagala^ described swimming by hatchling Dermochelys as composed of
two strokes: one in which the forelimbs are moved simultaneously while the hind-
limbs are spread “horizontally” and another, not mentioned by other investigators,
described as follows: “... the turtlet brings the inner edges of the hind limbs together,
then moves both up and down simultaneously in an undulating movement resembling
the action of the flukes of a cetacean. This shoots the animal forward at a more rapid
rate ....”. I have found no other accounts of this second pattern. Hence, it is probably
used infrequently. In hatchling and juvenile D. coriacea, the “axis” of the power-
stroke path differs from that of cheloniids and the route that the flipper follows is
opposite in direction. The flippers move upward and posteriorly, then downward
and anteriorly during normal powerstroking (Figure 7.12). It is difficult to divide the
leatherback powerstroke into neat protraction and retraction events because much
of the movement is upward and downward, with less anterior-posterior movement
than is seen in cheloniids. When swimming vigorously, the body tilts head down.
The flippers move posteriorly and downward, then anteriorly and upward relative
to the swimming path (or axis).^^
An overview comparing all locomotor characteristics of each marine turtle
species is provided in Table 7.1. Species-specific patterns are separated into temestrial
and aquatic locomotion and by ontogenetic categories in this summary. Because
little data exist for juveniles, those categories are omitted for simplicity.
7.4 MIGRATORY BEHAVIOR OF H ATCH LIN G S AN D

PELAGIC LIFE STAGES
Hatchling sea turtles undergo remarkable long-distance migrations in which they
often swim dozens of kilometers to reach offshore nursery areas.Hatchling migra
tion is accomplished during a swimming frenzy and postfrenzy period.
Several studies showed that hatchling sea turtles characteristically undergo a
period of hyperactive swimming activity upon entering the ocean for the first
time7 ’^^’^2 Wyneken and Salmon^^ studied activity patterns during the swimming
frenzy and postfrenzy by tethering hatchlings to activity sensors. The hatchlings
were free to swim in any direction, surface, or dive, but could not encounter tank
walls or tank bottom (a situation similar to what they would experience in the open
ocean). The study showed that hatchling loggerhead, leatherback, and green sea
turtles swam continuously through their first night and day; continuous swimming
typically stopped (ending the frenzy period) at some point during their second night
in the water. The swimming frenzy averaged 24 h in these species. Hatchling activity
declined by their second night in the water. By the third day in the water, loggerhead
and green turtle hatchlings swam only during the diurnal period. This pattern of
activity defines routine or postfrenzy behavior. Postfrenzy swimming by leatherback
hatchlings differed from that of cheloniids. Typically, they were active for the diurnal
period plus 30 to 40% of the nocturnal period. Thus hatchling sea turtles, unlike all
other reptiles, can perform vigorous locomotion for sustained periods during the
frenzy period. This continuous locomotion is associated with migratory movements
and reaches levels not seen in other reptiles.
7.4.1 S wimming C haracteristics during the Frenzy and
P ostfrenzy P eriods
The swimming behavior of hatchling Caretta, Cheionia, and Dermochelys was

documented by following free-swimming hatchlings moving offshore^^ and by mon
itoring tethered hatchlings in the laboratory."^^’^^ During the early part of the frenzy,
31
L 45
53
FIGURE 7.12 Powerstroking by a juvenile leatherback turtle. The “axis” of the leatherback
powerstroke path differs from that of cheloniids, and the route that the flipper follows is
opposite in direction so flippers move upward and posteriorly, then downward and anteriorly.
Numbers indicate the number of elapsed frames. From Renous, S. and Dels, V., J. Zool.
London, 230, 357-378, 1993. With permission.)
hatchling loggerhead and green turtles interrupt powerstroking with dogpaddling

when they surfaced to breathe. Leatherback hatchlings use powerstroking alone
(including when they surface to breathe) and were never observed to dogpaddle or
rear flipper kick. D. coriacea hatchlings showed only a slight decrease in stroke
amplitude over time; the flipper tips regularly extend above the water during the
frenzy, but ceased to break the surface of the water during postfrenzy swimming.'^^ ^^®
Witherington^'^ described the swimming behavior of hatchling Caretta (n = 15) which
TABLE 7.1
Summary of Locomotor Patterns
Hatchling
terrestrial Hatchling Adult terrestrial Adult
Species locomotion swimming locomotion swimming
Dermochelys Simultaneous Simultaneous beating of Simultaneous Simultaneous

coriacea movement of fore foreflippers; little or no movement of fore beating of
and hind flippers; thrust from hind flippers and hind flippers foreflippers; no
occasional thrust from hind
alternating steps flippers
Cheionia Diagonally Simultaneous beating of Simultaneous Simultaneous
mydas opposite limbs foreflippers; dogpaddling movement of fore beating of
moved together (diagonally opposite limbs and hind flippers foreflippers; no
moved together); rear thrust from hind
flipper kicking flippers
(simultaneous beating of
hind flippers)
Caretta Diagonally Simultaneous beating of Foreflippers Simultaneous
caretta opposite limbs foreflippers; dogpaddling moved beating of
moved together (diagonally opposite limbs alternately; hind foreflippers; no
moved together); rear flippers moved thrust from hind
flipper kicking together or flippers
(simultaneous beating of alternately
hind flippers)
Eretmochelys Diagonally Simultaneous beating of Foreflippers Simultaneous
imbricata opposite limbs foreflippers; dogpaddling moved beating of
hind flippers)
Lepidochelys Diagonally Simultaneous beating of Foreflippers Simultaneous
kempi opposite limbs foreflippers; dogpaddling moved beating of
hind flippers)
Lepidochelys Diagonally Simultaneous beating of Foreflippers Simultaneous
olivácea opposite limbs foreflippers; dogpaddling moved beating of
hind flippers)
TABLE 7.1 (continued)

Summary of Locomotor Patterns
Hatchling
terrestrial Hatchling Adult terrestrial Adult
Species locomotion swimming locomotion swimming
Natator Diagonally Simultaneous beating of Simultaneous Simultaneous

depressa opposite limbs foreflippers; dogpaddling movement of fore beating of
moved together (diagonally opposite limbs and hind flippers; foreflippers; no
moved together); rear or foreflipper thrust from hind
flipper kicking moved flippers
(simultaneous beating of alternately; hind
hind flippers) flippers moved
together or
alternately
he tracked as they swam offshore for 5 to 81 h. His descriptions of swimming in
the field very closely matched those observed in the laboratory.^
No published data are available on the swimming frenzy of Lepidochelys, Eret-
mochelys, or Natator. It has been hypothesized that Natator might have an abbre
viated frenzy since it apparently remains in continental shelf waters rather than
fleeing to the high seas.^^’^^
Free-swimming hatchlings (Caretta, Chelonia, and Dermochelys; n = 10 of each
species) that were followed offshore by boat or by snorkellers^^"^^’^^’^^^ typically
swam at or near the surface of the water. None of these turtles pulled floats (often
used in tracking studies). The cheloniid hatchlings occasionally dove and swam at
depths of 3 to 4 m, but, more typically, they swam within the upper 1 m of water.
Dermochelys hatchlings were more likely to dive (to depths of 4 to 18 m) while
swimming offshore but almost half the animals swam at the surface of the water.
Average swimming speeds for hatchlings during their frenzy (after correcting for
local water currents) were Dermochelys, 0.91 km/h; Caretta, 1.28 km/h; and Che
lonia, 1.57 km/h. These speeds for green turtle and loggerheads were similar to
those reported earlier for free-swimming cheloniids {Chelonia, 1.56 km/h;^^ Caretta,
1.26 km/h. ^2) They are also similar to the top speed found by Wither!ngton"^^ for
loggerhead hatchlings towing small floats (1.31 km/h), but greater than the typical
speeds (0.83 to 0.89 km/h) he observed at different sites, some with natural skylines
and some with heavy levels of photopollution. In a subsequent study, 15 hatchlings
towing floats attained higher speeds^"^ (mean = 3.76 km/h, S.E. = 0.49). However, it
is unclear if these data were corrected for the effects of currents.
In the three species examined, powerstroke rates were often highest at the start
of the frenzy (Table 7.2). By the end of 24 h, powerstroke rates decreased (on
average) and became quite variable. Based on 95% confidence limits, the foreflipper
stroke rates decrease significantly between the frenzy and postfrenzy in green turtles
and leatherbacks, but not in loggerheads (Table 7.2). Wyneken^^ showed that pow
erstroke bout durations varied with species. Generally, they are shorter at the start
of the frenzy in Caretta and Dermochelys, and shorter after the frenzy in Chelonia.
TABLE 7.2
Oxygen Consumption and Swimming Parameters for Hatchling Chelonia,
Caretta^ a n d Dermochelys*
Power-
VO2 Power- stroke Stroke
Migratory ml • g-’ • strokes bout Breath rate
Species phase min-^ per bout duration bout duration (strokes/s)
Cheionia Frenzy 0.021 ±0.001 18.4 ±2.3 8.4 ± 1.0 1.5 ±0.2 2.2 ± 0.4
Postfrenzy 0.013 ±0.001 13.5 ± 1.5 7.5 ± 0.9 2.8 ± 0.2 1.9 ±0.1
Resting 0.005 ± 0.0003
Caretta Frenzy 0.016 ± 0.001 21.8 ±2.5 10.6 ± 1.1 1.6 ±0.2 2.1 ±0.1
Postfrenzy 0.017 ±0.002 24.7 ± 3.0 11.6 ±2.0 2.2 ± 0.3 2.0 ±0.1
Resting 0.009 ± 0.0003
Derm ochelys Frenzy 0.010 ± 0.001 27.2 ± 3.0 19.0 ± 1.8 1.9 ± .0.3 1.4 ± 0.1
Postfrenzy 0.008 ± 0.001 24.7 ± 3.4 21.2 ± 3.4 2.4 ± 0.3 1.2 ±0.1
Resting 0.004 ± 0.0002
* All were active during the frenzy and postfrenzy, but inactive during resting measurements.
Means ±1 S.E. are based upon data collected every 5 min for 1 h. Individual turtles (n = 10 of each
species) that had not been fed were monitored separately as they swam in seawater or rested in a black
container. Both water and air temperatures were 24°C.
Surface times (during which hatchlings breathe) increase in duration following the
swimming frenzy (Table 7.2).
The combination of slow swimming speeds, slow flipper movements, and long
powerstroke bout durations in leatherbacks suggests a “marathon” strategy for the
migration (slow, steady, economical movements as opposed to rapid bursts of move
ment punctuated with many breathing interruptions). This contrasts sharply with the
“sprinter” strategy of hatchling green turtles: fast swimming speeds, quick flipper
movements, and short-duration powerstroke bouts.
7.4.2 S wimming C haracteristics of P elagic S tages
What little infomiation is available on the swimming behavior of pelagic-stage turtles

comes from rare observations in the field and from the study of turtles raised in
captivity. Deraniyagala* noted that leatherback hatchlings swam offshore at a depth
of 30 cm. He raised several leatherback hatchlings in captivity for up to 662 d, and
observed that they swam persistently without ever recognizing the tank sides as a
barrier.
The swimming behavior of juvenile green turtles was described in the labora-
tory.15,25 Juveniles (approximately 500 g) swam using the powerstroke or dogpaddle.
Green turtles greater than 500 g also were able to swim backward by kicking with
alternate hind flippers while extending the foreflippers in a protracted position
adjacent to the head. Juvenile green turtles did not use the synchronous rear flipper
kick.
Witherington'"^ found post-hatchling loggerheads (43 to 59 mm standard straight-
line carapace length) in the open ocean to be relatively inactive. Most swimming
was accomplished using the rear flipper kick (n = 10 turtles). Only one turtle was
observed using the powerstroke. Davenport and Clough described the swimming
behavior of juvenile loggerheads (400 to 600 g) raised in the laboratory. They found
that these animals spent their time at the surface. More than 95% of their surface
swimming was carried out using the rear flipper kick. The powerstroke was rarely
used, and then only during escape or diving attempts. Since young loggerheads are
positively buoyant, they have difficulty diving until they reach larger sizes.
Ecologically juvenile cheloniids are found in the epipelagic zone (far from bottom
structures and food in many cases).
The swimming behavior of captive juvenile Pacific ridley turtles (L. oliváceo)
was described by Davenport and Pearson. These turtles, like loggerheads, are
positively buoyant and spend their time at the surface of the water. The majority of
their swimming involved synchronous or asynchronous rear flipper kicking. Unlike
loggerheads, the olive ridley turtles did not carry their foreflippers flexed over the
carapace, but instead held them laterally, edge on to water flow. During escape the
turtles dogpaddled. The powerstroke was used rarely (only when the turtles tried to
vigorously escape handling).
7.5 MIGRATORY LOCOMOTOR BEHAVIOR OF

SUBADULTS AND ADULTS
Information on the swimming behavior of adult and subadult turtles during migration
is fragmentary. Undoubtedly, this gap is due to the logistical problems associated
with monitoring turtles over large distances and, sometimes, at great depths. Addi
tionally, once in the ocean the density of animals becomes relatively low; chance
encounters and observations on behavior are less frequent. The migration routes of
some populations of C. mydas (adults), Caretta caretta (juveniles and adults), L.
kempi (juveniles and adults), and D. coriacea (adults) and estimated swimming
speeds have been described (see below). Relatively few reports make mention of
the locomotor patterns that the turtles used.
7.5.1 S wimming by C heloniids
I have observed juvenile green turtles swimming along reefs in Hawaii and Florida,
and subadult loggerheads in Florida. All relied on the powerstroke for routine
swimming. No other flipper movement patterns were associated with locomotion in
these free-swimming animals. Captive C. mydas occasionally attempt to swim back
wards to escape a barrier. The mechanism is the same as described above. Captive
Chelonia mydas and Caretta caretta juveniles have been observed to push off the
bottom or move along the bottom using a modified dogpaddle stroke (diagonal limbs
moved together to push along the substratum).
Data collected in Hawaii and Florida indicate that juvenile Chelonia mydas may
live relatively sedentary lives once they recruit to foraging and resting areas from
the open o c e a n . A d u l t green turtles from both Atlantic and Pacific populations
are known to migrate long distances. Migrations from feeding areas to nesting
beaches at French Frigate Shoals, part of the Hawaiian Islands, can exceed 100 km.
Balazs^^ estimates that these turtles cover 23 km/d. Mature green turtles marked at
Tortuguero, Costa Rica; Ascension Island, South Atlantic Ocean; and Southern
Yemen were subsequently recaptured after long-distance migrations. These mark-
and-recapture studies gave typical swimming speeds of 23.5 km/d, 21.6 km/d, and
22.4 km/d, respectively (summarized by Hirth).^"^ Prange^^ measured swimming
speeds averaging 1.4 to 3.6 km/h by adult green turtles (n = 10) as they swam away
from their nesting beach at Tortuguero, Costa Rica. Oliver^^ reported slightly higher
cruising speeds of 1.4 to 2.2 km/h by a captive adult C. mydas. Both Prange’s and
Oliver’s measurements of swimming speeds are considerably higher than those
calculated from mark-and-recapture data. In both cases they may have observed
sprinting rather than routine swimming. The highest published average speeds (53
to 65 km/d) are based upon mark-and-recapture of females after leaving the nesting
beach (Shultz,^^ summarized by Meylan).^^ However, since the routes taken by free-
swimming adults are not known (nor are the effects of cunênts), the field estimates
may be too low or too high. Papi et al.^*^ tracked the post-nesting migration of an
adult green turtle by using satellite telemetry. Though the transmitter attached to the
shell increased the drag the turtle experienced, that increase is probably minimal
because the size of the turtle was quite large and the transmitter package was
relatively small and low in profile. The swimming speeds obtained (1.57 to 3.73
km/h = 37.7 to 89.5 km/d) were similar to those obtained from tag-return data, but
these estimates too do not account for displacements due to cuirents.
CarC and Parrish^'^ described swimming by captive adult loggerhead turtles.
Both indicated that the predominant method of thrust production was by the pow-
erstroke. Parish^^ noted that loggerheads typically completed 10 to 30 powerstrokes
per minute. Most published swimming speeds for this species are based upon
recaptures of marked turtles after a migration. These speeds ranged from 28.1 to
40.2 km/d.^°'^2 Tucker et al.^'^ obtained estimated swimming speeds from adult
female Caretta towing satellite transmitters. During their internesting movements
they averaged 0.45 km/h (range = 0.02 to 3.01 km/h). All free-swimming subadult
and adult Caretta (n = approximately 60) as well as juvenile and adult Chelonia (n
= approximately 84) that I observed swimming relied on the powerstroke. No
dogpaddling or rear flipper kicking has been observed in these sublittoral animals.
I observed that most swimming E. imbricata (juveniles and adults) in the waters
off South Florida used what appeared to be a typical cheloniid powerstroke. When
feeding, they occasionally crawl across the reef using diagonally opposite limbs. No
kinematic studies of hawksbill turtle swimming have been done and no quantitative
measurements exist of the thrust they generate from pushing against the water, the
hard bottom, or both.
Mark-and-recapture studies of E. imbricata suggest that juveniles are relatively
sedentary, though some long-distance migrations have been documented from
Indo-Pacific populations (summarized by Witzell).^^ One reporP"^ estimated the

swimming speed of a hawksbill turtle at 17.8 km/d (somewhat slower than those
speeds listed for green turtles and loggerheads). However, like the data given
above, these are also rough estimates, since routes were not known and the effects
of currents were not documented.
Tag returns and tracking provide most of the swimming speed data for ridley
turtles (L. kempi and L. oliváceo). However, in one study, captive-raised juvenile
Kemp’s ridley turtles were able to hold sustained speeds in a flow tank of 1.0 to 2.6
m/min after an exercise regime, while those not given exercise had less stamina and
had difficulty maintaining higher swimming speeds.^^ Mark-and-recapture of two
free-swimming Kemp’s ridley turtles provided speed estimates of 24 and 29.5 km/d^^
(summarized by Meylan).^^ Similar studies of olive ridley turtles suggest migratory
swimming speeds of 28.4 to 36.7 km/d^^ and 28.1 to 82.6 km/d^^ (summarized by
Meylan).^^ The very high estimated minimum swimming speeds require that the
turtle sustained swimming speeds of approximately a body length per second (0.94
m/s = 82.6 km/d). While such speeds are clearly possible in bursts, they are probably
hydrodynamically unreasonable as cruising speeds (Block et al.)^^ and probably
include considerable displacement from currents.
Flatback turtles (N. depressa) undergo long-distance migrations that are confined
to Australian continental shelf waters.^^’^^ Mark and recapture of 8 free-swimming
adult females provided data documenting minimum migratory distances of roughly
200 to 1300 km. Using the published recapture interval and minimum distance
covered (and after eliminating animals that died or were sick), it appears that the
flatback might take a more leisurely pace than other cheloniid species (1.5 to 8.8
km/d). However, as with other mark-and-recapture data, neither currents nor actual
routes were accounted for by such calculations.
7.5.2 S wimming by D erm o ch elys C oriácea
The leatherback turtle is known for the longest migrations (both horizontally and
vertically). Mark-and-recapture studies of the adult turtles show that this species
regularly undergoes long-distance migrations (the maximum recorded migration
distance approaches 6000 km).'^^’^^ Reports in the range of 500- to 1000-km migra
tions are not uncommon. An estimated travel speed of 40 km/d was logged for a
turtle carrying a satellite transmitter after nesting at French Guiana.^^ A female
leatherback towing a satellite transmitter was tracked during its intemesting move
ments.^^ It traveled a minimum distance of 515 km, with a mean speed of 1.2 km/h
(= 28.8 km/d). As with other estimated speeds, no account of the effects of cuirent
was possible. Leatherback turtles are also noted for spectacular diving capabilities
which necessarily involve active swimming. Eckert et al.^"^ documented a maximum
dive to 1300 m by one mature female, and many other dives to less great depths
(around 200 m). Limpus^^ reported that a leatherback took an octopus bait in 50 m
of water.
7.6 E N E R G E T IC A D A P T A T IO N S F O R M IG R A T O R Y
SW IM M IN G
Reptiles, generally, have low stamina and typically are capable of only short bursts
of high activityMarine turtles are obvious exceptions, as most swim for long
distances^^ and, in the case of hatchlings, for long durations without stopping."^"'
7.6.1 A erobic A daptations and B ehavioral C orrelates
7.6.1.1 Hatchlings during Frenzy and Postfrenzy Swimming
Wyneken'"^ measured mass-specific oxygen consumption (VO2 ) as hatchlings (C.

mydas, Caretta caretta, and D. coriacea, n = 10 of each species) swam on a long
tether inside an open-flow respirometer. As in experiments documenting activity
levels,"^^ hatchlings were free to swim in any direction, use any gait, but could not
contact the walls, top, or bottom of the respirometer. In all three species examined,
frenzy and postfrenzy (routine) swimming resulted in significantly higher rates of
O2 consumption than during rest. This indicates that hatchlings rely on aerobic
metabolism during at least part of their migratory swimming.
Caretta, Chelonia, and Dermochelys differ in aerobic scope (Figure 7.13); Che-
lonia had the broadest aerobic scope while Dermochelys had the narrowest. Caretta
was somewhat intermediate; low (resting) VO2 levels were greater than those of
Chelonia and Dermochelys while highest (frenzy swimming) VO2 levels were less
than those of Chelonia and Dermochelys, In Chelonia and Dermochelys, mean VO2
was highest at the start of the frenzy, then dropped so that after 24 h (during
postfrenzy swimming), O2 was consumed at a lower rate. In Caretta, VO2 was similar
during frenzy and postfrenzy swimming (Table 7.2, Figure 7.14). Both VO2 (Table
7.2) and cost of transport differ interspecifically and intraspecifically, between frenzy
and postfrenzy swimming activity in hatchlings. The VO2 measured for leatherback
hatchlings during the postfrenzy period was similar to that reported by Lutcavage
and Lutz.^^ Though their hatchlings were not tethered, they swam continuously
throughout the sampling period. The VO2 levels they recorded for swimming log
gerhead hatchlings (clearly postfrenzy) were much lower (0.0035 ml O2 • g ‘ • min
than those I found for postfrenzy hatchlings (0.017 ml O2 • g“^ • mimO- Since similar
temperatures were used in both studies, this difference may be a consequence of the
sporadic swimming they observed (the hatchlings I studied swam most of the time).
Generally, the highest rates of oxygen consumption are associated with the
fastest powerstroke rates observed early in the frenzy (the first 9 to 12 h). The stroke
rates gradually slow during the second half of the frenzy, but conelation between
activity level and oxygen consumption was not strong.^^ During both the frenzy and
postfrenzy periods, leatherback hatchlings consumed the least oxygen. During post
frenzy swimming, green turtles had lower VO2 values than did loggerheads (but
greater values than leatherbacks). Both leatherbacks and green turtle hatchlings
consumed more O2 during frenzy swimming than they did during postfrenzy (rou
tine) swimming. Loggerheads did not differ in VO2 between the frenzy and post
frenzy (Figure 7.14). When resting, loggerhead hatchlings consumed O2 at signifi
cantly greater rates than leatherbacks and green turtles (Table 7.2).
Aerobic Scope
c 0.028
1# Chelonia m ydas
w 0.024
1
■O
4> 0.020
E
3(ft
c
o
0-016 D erm ochelys
Q coriacea
C
0U)) 0.012
>
X
«
o 0.008
S 0.004
>O 0.000
FIGURE 7.13 Aerobic scope of three species of sea turtles. VO2 was measured for a 1.5-
h period (after a 30-min acclimation period) while the animals (i) swam in an open flow
respirometer or (ii) were rested quietly in a dark container. In all species maximum VQ was
measured during the swimming frenzy while hatchlings were powerstroking. Minimum VQ
was from resting measurements.
7.6.1.2 Juveniles, Subadults, and Adults
Prange^^ monitored O2 consumption by juvenile green turtles (n = 5) during rest and

swimming in a flow tank. Resting VO2 was 0.07 1 O2 • kg~^ • h ^ a value similar to
an estimate calculated by Mrosovsky and Pritchard^^ for resting adult green turtles
(0.069 1 O2 • kg“^ • Swimming juvenile green turtles increased their oxygen
consumption with swimming speed (range -^0.08 to 0.27 1 O2 • kg~^ • h~‘)- Prange
and Jackson^^ studied a juvenile green turtle. They noted that during swimming
(presumably powerstroking), oxygen consumption increased tenfold over resting
metabolic rates, indicating a heavy reliance on aerobic metabolism to fuel routine
swimming. Butler et al.^^ measured V02in juvenile Chelonia (n = 5) during rest and
during steady swimming (0.4, 0.5, and 0.6 m • s~*)- They found that mean resting
VO2 was approximately 1.98 ml • kg~^ • min k VO2 during swimming was signifi
cantly greater than during resting and increased with swimming speed (from 3.77
to 5.60 ml • kg'^ * min~0- Since O2 consumption is an indirect measure of energy
consumption, these data suggest that energy demands during migration may be
among the biologically important factors that determine how quickly a sea turtle
accomplishes a migration.
In another set of studies, Jackson and Prange^^ examined nesting Chelonia and
noted that resting VO2 decreased with increasing body mass while maximal (active)
VO2 remained unchanged. These observations suggest that larger animals are capable
of greater relative increases in activity than smaller animals. This observation may
be important in understanding hatchling migration and dispersal. Hatchlings migrate
offshore without heading toward a particular discreet goal (rather, they just need to
FIGURE 7.14 VO2 during frenzy swimming, postfrenzy swimming, and while resting. Each
reading is the mean ± SE for 10 animals. Mass specific O2 consumption was greatest during
frenzy swimming in Chelonia and Dermochelys, but not in Caretta. Swimming was aerobi
cally more expensive than resting, showing that aerobic metabolism is an important pathway
during hatchling migratory swimming.
locate deep water and favorable currents). Once hatchlings distance themselves from
land they become less active swimmers. Their small size and relatively low thrust
prevent them from countering the effects of major currents. Once offshore they might
be viewed as swimming drift bottles, largely at the mercy of currents. In contrast,
adults have well-documented long-distance migrations between specific sites (e.g.,
feeding and breeding grounds). Their migrations sometimes involve swimming
against currents, clearly an energetically more expensive proposition. One might
cautiously suggest that adults are free to accommodate a much greater increase in
activity with less cost.
Jackson and Prange^^ noted that the breathing pattern of turtles on land (rest
with ventilation punctuating bouts of exercise) was different from that of swimming
turtles (single breaths). Hence, VO2 levels from nesting turtles may not be good
approximations of VO2 for swimming animals.
No data are available describing oxygen consumption by swimming Eretmo-
chelys, Lepidochelys, or Natator.
Aerobic measurements from adult Dermochelys are limited to data from adult
females during nesting, during post-nesting crawling, or restrained turtles (captured
just after nesting)."^^'^^ There are no data available for swimming animals. If breathing
patterns differ substantially between swimming vs. nesting or crawling (as described
for Chelonia), then aerobic measurements from nesting females may not be good
predictors of VO2 for swimming leatherbacks.
7.6.2 A naerobic A daptations and M igratory Swimming

Anaerobic metabolism of hatchling Caretta and Chelonia was measured during
hatching, nest emergence, sea-finding, and offshore swimming by DiaP*-' and Baldwin
et al.^^ The two studies used different assays. DiaP^ measured whole body lactate;
Baldwin et al.^^ measured lactate in blood samples. The two studies reached quali
tatively similar conclusions about the relationships between anaerobic metabolism
and natural activity for hatchlings while on the beach. However, they reached
different conclusions about the role of anaerobic glycolysis during the first few hours
that hatchlings are in the water. DiaT^^ measured whole body lactate levels from
hatchlings that swam in the laboratory. He found they were low. When Baldwin et
al.^^ measured blood lactate from hatchlings that swam away from shore, levels were
high. To resolve this discrepancy, Baldwin et al,^^ conducted an ingenious set of
experiments. They measured lactate in repeated blood samples from the same
hatchlings that were resting, then swimming, or resting, then crawling. Their results
showed that hatchlings (1) accumulate high levels of lactate while crawling, but that
(2) those hatchlings which swam in an aquarium (in the laboratory) had low levels
of blood lactate. Those hatchlings that swam in open water had high levels. They
also observed that hatchlings swimming in aquaria altered their swimming behavior
and held their heads elevated above water much more often than free-swimming
turtles. There was no rapid “washout” of muscle lactate from field-tested animals
prior to swimming under field conditions. Hence, the low levels of lactate in swim
ming turtles that DiaP^ observed were probably due to a laboratory artifact (altered
swimming and breathing behavior). Baldwin et al.'^^ hypothesized that anaerobic
metabolism supplies a significant part of the energy for the offshore swim of
hatchlings. Hence, anaerobic glycolysis is clearly important to hatchlings during
most early phases of activity, both during digging to the surface of the nest and
during (at least) the early part of the swimming frenzy.
7.7 C O N C L U S IO N S
Sea turtles are the migratory specialists among the testudines. They possess a suite
of morphological, behavioral, and physiological characteristics that are adaptations
for sustained swimming. Unlike virtually all other turtle species, marine turtles
emphasize the mechanically efficient powerstroke as their primary swimming gait.
However, these migratory giants are forever tied to land for reproduction. Hence,
their morphological and mechanical traits must always reflect a compromise between
efficiency for the aquatic environment and effective for movement on land. Though
they spend a fraction of their lives on land, that portion of time is critical to
perpetuation of their genes, and hence is a strong selective pressure. Marine turtle
adaptations to the ocean environment are also constrained in another way. All of
their structural adaptations are built from and superimposed upon, a basic turtle body
plan (the armored tank, or box). Thus, they are also constrained by their ancestry.
During ontogeny, the ecology of the different life history stages changes as does
expression or emphasis of their adaptations for a migratory and marine life. This is
particularly evident in the ontogenetic shifts in locomotor behavior and mechanisms.
For example, the powerstroke is initially the primary gait of cheloniid hatchlings as
they distance themselves from land. It is largely abandoned in the pelagic (by some
species) where locomotion serves more for maneuvering during food gathering;
crypsis and/or carapace morphology provide protection. In the pelagic, dogpaddling
or rear flipper kicking predominate. The powerstroke returns as the primary gait
used by sublittoral (juvenile or subadult) life stages and adults. Activity patterns
(frenzy vs. routine swimming) change during ontogeny and correspond with eco
logical shifts and the associated migrations. The energetic costs of migratory swim
ming (and perhaps the energy pathways utilized: anaerobic and aerobic) not only
change with ontogenetic shifts in swimming behavior, but also may scale with body
size.
The cheloniid species are most similar in morphological, kinematic, and behav
ioral features. However, energetic adaptations may be more closely allied with the
ontogeny and ecology of a species than it is to phylogeny. Typically, cheloniids
switch swimming behavior and habitat several times in their life. For example,
Caretta caretta and L. kempp^ migrate away from predator-rich nearshore waters to
offshore nursery areas. Initially, they rely on a mechanically efficient (but for the
tiny hatchlings, energetically expensive) powerstroke and high levels of activity,
then, once in the pelagic environment their activity drops and their predominant
mode of propulsion changes. They remain pelagic animals for just a portion of their
lives, during which they grow considerably. They return to coastal waters as juveniles
and subadults. Their migrations along the coastlines are again executed using the
efficient powerstroke.
Dermochelys hatchlings are pelagic migrants and remain so throughout life. This
turtle swims relatively slowly using the mechanically efficient powerstroke for very
long periods of time during routine swimming,and has low aerobic costs. Covering
long distances efficiently is critical. Its powerstroke, while different in detail from
that of cheloniids, is still a mechanically efficient gait. This gait is not abandoned
during ontogeny and persists largely unchanged.
REFEREN CES
1. Deraniyagala, P. E. P., The tetrapod reptiles of Ceylon, Ceylon J. Sci., (Colombo

Museum Natural History Series), 1, 1-242, 1939.
Sarawak, Proc. Zool. Soc. Lond., Vol. 130, No. 4, 455-535, 1958.
3. Bustard, H. R., Sea Turtles: Their Natural History and Conservation, Collins, Lon
don, 1972.
4. Carr, A. F., The Handbook o f Turtles — Turtles o f the United States, Canada, and
Baja California, Cornell University Press, Ithaca, NY, 1952.
5. Mrosovsky, N., Conserving Sea Turtles, The British Heipetological Society, London,
1983.
6. Pritchard, P. C. H. and Trebeau, E., The Turtles o f Venezuela, S.S.A.R. Publications,
Cushing-Malloy, Ann Arbor, MI, 1982.
7. Renous, S., The ontogeny of locomotion in marine turtles, Sci. HerpetoL, 1995,
112-119, 1995.
8. Walker, W. F, Jr., Locomotion, in Turtles: Perspective and Research, Harless, M.
and Morlock, M., Eds., John Wiley & Sons, New York, 1979, 435-454.
9. Hildebrand, M., Walking and running, in Functional Vertebrate Morphology, Hilde
brand, M., Bramble, D. M., Liem, K. F, and Wake, D. B., Eds., The Belknap Press
of Harvard University, Cambridge, MA, Chapter 3, 1985, 38-57.
10. Wyneken, J., Comparative and Functional Considerations of Locomotion in Turtles,
Doctoral dissertation. University of Illinois, 1988, Champaign-Urbana, IL, 151 pp.
11. Carr, A. F. and Ogren, L., The ecology and migrations of sea turtles. 4. The green
turtle in the Caribbean Sea, Bull. Am. Mus. Nat. Hist., 121(1), 6 ^ 8 , 1960.
12. Salmon, M. and Wyneken, J., Orientation during the swimming frenzy period in
loggerhead sea turtles,/. Exp. Mar. Biol. EcoL, 109, 137-153, 1987.
13. Parker, G. H., The instinctive locomotor reactions of the loggerhead turtle in relation
to its senses. Comp. Psychol., II, 425-429, 1922.
14. Davenport, J. and Clough, W., Swimming and diving in young loggerhead sea turtles
(Caretta caretta L.), Copela, 1, 53-57, 1986.
15. Davenport, J. and Pearson, G. A., Observations on the swimming of the Pacific ridley,
Lepidochelys olivácea (Eschscholtz, 1829) comparisons with other sea turtles, Her-
petol. J., 4, 60-63, 1994.
16. Carr, A. F and Ogren, L., The ecology and migrations of sea turtles. 3. Dermochelys
in Costa Rica, Am. Mus. Novit., 1958, 1-29, 1959.
17. Renous, S. and Bels, V., Comparison between aquatic and terrestrial locomotion of
the leatherback sea turtle {Dermochelys coriáceo), J. Zool. London, 230, 357-378,
1993.
18. Renous, S., Retentissement de Tadaption a la vie palagique de la tortue luth (f>er-
mochelys coriáceo) sur sa locomotion terrestre, Mesogee, 48, 79-84, 1988.
19. Davenport, J., Locomotion in hatchling leatherback inrÜQS Dermochelys coriacea, J.

ZooL London, 212, 85-101, 1987.
20. Whittow, G. C. and Balazs, G. H., Basking behavior of the Hawaiian green turtle
{Chelonia mydas), Pac. Sei., 36(2), 129-139, 1982.
21. Pritchard, P. C. H., Encyclopedia o f Turtles, T. F. H. Publications, Neptune, NJ, 1979,
895 pp.
22. Zangerl, R., Patterns of phylogenetic differentiation in the toxochelid and cheloniid
sea turtles. Am. ZooL, 20, 585-596, 1980.
23. Walker, W. F., Jr., The locomotor apparatus of the testudines, in The Biology o f the
Reptilia, Gans, C. and Parsons, T. S., Eds., Academic Press, New York, 1973, 1-100.
24. Zangerl, R., The vertebrate fauna of the Selma formation of Alabama. III. The turtles
of the family Protostegidae. IV. The turtles of the family Toxochelidae, Fieldiana
GeoL Mem., 3, 57-297, 1953.
25. Davenport, J., Munks, S. A., and Oxford, P. J., A comparison of the swimming of
marine and freshwater turtles, Proc. R. Soc. London, 220, 447-^75, 1984.
26. Munks, S., The swimming mechanism of a marine turtle — Chelonia my das, Testudo.
J. Br. Chelonia Group, 2(3), 33-55, 1985.
27. Wyneken, J., Functional innovations in swimming analyses of sea turtle locomotor
patterns. Am. ZooL, 28, (13A), 1988.
28. Blake, R. J., Mechanics of drag based mechanisms of propulsion in aquatic verte
brates, Symp. ZooL Soc. London, 48, 29-52, 1981.
29. Clark, B. D. and Bemis, W., Kinematics of swimming of penguins at the Detroit Zoo,
/. ZooL London, 188, 411-428, 1979.
30. Robinson, J. A., The locomotion of plesiosaurs, Vene.? Jahrb. GeoL Paläontol. Ahh.,
149(3), 286-332, 1975.
31. Fürbringer, M., Zur vergleichen anatomie der schultermuskelen, Jena Zschr., 8,
175-280, 1894.
32. Hoffman, C. K., Reptilien. I. Schildkröten, in Klassen und Ordnungen des Their-
Reichs, Bronn, H. G., Ed., 6(3), 1-^42, 1890.
33. Burne, B. A., Notes on the musculature and visceral anatomy of the leathery turtle
(Dermochelys coriacea), Proc. ZooL Soc. London, 291-324, 1905.
34. Sieglbauer, F, Zur anatomie der Schildkröten extremität, Ac//. Anat. Physiol. Anat.
Abt., 183-280, 1909.
35. Aleyev, Y. G., Nekton, Dr. W. Junk, The Hague, 1977.
36. Wyneken, J. and LaBarbera, M., Locomotor strategies of hatchling sea turtles. Am.
ZooL, 27(4), 63A, 1987.
37. Logan, P. and Morreale, S. J., Hydrodynamic drag characteristics of juvenileL. kempi,
C. mydas, and C. caretta, in Proc. 13th Annu. Conf. Sea Turtle Biology and Conser
vation, Schroeder, B. A. and Witherington, B. E., Compilers, NOAA Tech. Memo.
NMES-SEESC-341, Miami, FL, 248, 1994.
38. Prange, H. D., Energetics of swimming of a sea turtle,/. Exp. BioL, 64, 1-12, 1976.
39. Vogel, S., Life in Moving Fluids, Princeton University Press, Princeton, NJ, 1983,
352 pp.
40. Webb, P. W., Hydrodynamics and energetics of propulsion, ßw//. Fish. Res. B. Canada,
190 (1975), 1-158, 1975.
4L Hui, C. A., Penguin swimming. I. Hydrodynamics, P/iys/o/. ZooL, 61(4), 333-343,
1988.
42. Newman, J. N. and Wu, T. Y , Hydromechanical aspects of fish swimming, in Swim
ming and Flying in Nature., Vol. 2., Wu, T. Y.-T., Brokaw, C. J., and Brennen, C.,
Eds.. Plenum Press. New York, 1975, 615-634.
43. Lang, T. G. and Norris, K. R, Swimming speed of the Pacific bottlenose porpoise,
Science, 151, 588-590, 1966.
44. Feldkamp, S. D., Swimming in the California sea lion: morphometries, drag and
energetics,/. Exp. Biol., 131, 117-135, 1987.
45. Wyneken, J. and Salmon, M., Frenzy and postfrenzy swimming activity in loggerhead,
leatherback, and green sea turtles, Copeia, 1992(2), 478^ 84, 1992.
46. Witherington, B. E., Orientation of hatchling loggerhead turtles at sea off artificially
lighted and dark beaches,/. Exp. Mar. Biol. Ecol, 149, 1-11, 1991.
47. Walker, W. R, Jr., Swimming in sea turtles of the family Cheloniidae, Copeia, (2),
229-233, 1971.
48. Mordvinov, Y. E., Morphological features of the locomotor organs and body of the
loggerhead and pond turtle, Vestn. Zool., 3, 58-62, 1986.
49. Tucker, A. D., FitzSimmons, N. N., and Limpus, C. J., Conservation implications of
internesting habitat use by loggerhead turtles {Caretta caretta) in Woongarra Marine
Park, Queensland, Australia, Pac. Conserv. Biol., 2, \5 1 -\6 6 , 1996.
50. Lohmann, K. J., Swartz, A., and Lohmann, C. M. R, Perception of ocean wave
direction by sea turtles,/. Exp. Biol., 198, 1079-1085, 1995.
51. Carr, A., New perspectives on the pelagic stage of sea turtle development, NOAA
Tech. Memo. NMFS-SEFC-190, Miami, FL, 1-36, 1986.
52. Wyneken, J., Goff, M., and Glenn, L., The trials and tribulations of swimming in the
near-shore environment, in Proc. 14th Annu. Symp. Sea Turtle Biology and Conser
vation, Bjomdal, K. A., Bolton, A. B., Johnson, D. A., and Eliazar, P. J., Compilers,
NOAA Tech. Memo. NMFS-SEFC-351, Miami, FL, 169-171, 1994.
53. Wyneken, J. and Salmon, M., Swimming the offshore marathon: analyses of early
swimming activity in loggerhead, green turtle and leatherback hatchlings, in Proc.
10th Annu. Workshop on Sea Turtle Biology and Conservation, Richardson, T. H.,
Richardson, J. L, and Donnelly, M., Compilers, NOAA Tech. Memo. NMFS-SEFC-
278, Miami, FL, 109, 1990.
54. Witherington, B. E., Observations of hatchling loggerhead turtles during the first few
days of the lost year(s), in Proc. 12th Annu. Workshop on Sea Turtle Biology and
Conservation, Richardson, J. I. and Richardson, T. H., Compilers, NOAA Tech.
Memo. NMFS-SEFSC-361, Miami, FL, 154-157, 1995.
the flatback turtle, Natator depressa (Garnian), /. Biogeog., 17, 275-278, 1990.
56. Yeomans, S. R. and Limpus, C. J., Behavior of the flatback sea turtle (Natator
depressus) hatchlings, in Proc. 14th Annu. Symp. Sea Turtle Biology and Conserva
tion, Bjorndal, K. A., Bolton, A. B., Johnson, D. A., and Eliazar, P. J., Compilers,
NOAA Tech. Memo. NMFS-SEFC-351, Miami, FL, 172, 1994.
57. Frick, J., Orientation and behaviour of hatchling green turtles {Chelonia mydas) in
the ^Q2L,Anim. Behav., 24, 849-857, 1976.
58. Wyneken, J., Comparison of oxygen utilization by hatchling loggerheads, greens, and
leatherbacks during the swimming frenzy: sprinting vs. marathon strategies revisited,
in Proc. 11th Annu. Workshop on Sea Turtle Biology and Conservation, NOAA Tech.
Memo. NMFS-SEFC.-232, Miami, FL, 131-132, 1991.
59. Witherington, B. E., Some lost year turtles found, in Proc. 14th Annu. Symp. Sea
Turtle Biology and Conservation, Bjorndal, K. A., Bolton, A. B., Johnson, D. A., and
Eliazar, P. J., Compilers, NOAA Tech. Memo. NMFS-SEFC-351, Miami, FL,
194-197, 1994.
60. Milsom, W. K., Development of buoyancy control in juvenile Atlantic loggerhead
turtles, Caretta caretta, Copeia,, 1975, 758-762, 1975.
61. Balazs, G. H., Synopsis of the biological data on the green turtle in the Hawaiian
Islands, NOAA Tech. Memo. NMFS, NOAA-TM-NMFS-SWFC-7, Honolulu, HI,
1980, 141 pp.
62. Mendonca, M. T., Movements and feeding ecology of immature green turtles (phe-
lonia my das) in a Florida lagoon, Copeia, 4, 1013-1023, 1983.
63. Brill, R. W., Balazs, G. H., Holland, K. N., Chang, R. K. C., Sullivan, S., and George,
J. C., Daily movements, habitat use, and submergence intervals of normal and tumor
bearing juvenile green turtles (Chelonia mydas L.) within a foraging area in the
Hawaiian Islands,./. Exp. Mar. Biol. EcoL, 185, 203-218, 1995.
64. Hirth, H. F, Synopsis of biological data on the green inxiXt Chelonia mydas (Linnaeus)
1758, FAO Eish. Synop., 85, 1971.
65. Oliver, J. A., The Natural History o f North American Amphibians and Reptiles, Van
Nostrand, New York, 1955, 359 pp.
66. Shultz, J. R, Sea turtles nesting in Suriname, Zo6>/. Verh. Rijksmus. Nat. Hist. Leiden,
143, 1-44, 1975.
67. Meylan, A., Sea turtle migration — evidence from tag returns, in Biology and
Conservation o f Sea Turtles, Revised ed., Bjorndal, K. A., Ed., Smithsonian Institution
Press, Washington, D.C., 1995, 91-100.
68. Papi, F, Liew, H. C., Luschi, R, and Chan, E. H., Long-range migratory travel of a
green turtle tracked by satellite: evidence for navigational ability in the open sea.
Mar Biol., 122, 171-175, 1995.
69. Parrish, F. K., Miscellaneous observations on the swimming behavior of captive sea
turtles. Bull. Mar. Sci. GulfCaribb., 8, 348-55, 1958.
70. Bustard, H. R. and Limpus, C., The first international recapture of an Australian
tagged loggerhead turtle, Herpetologica, 26, 358-359, 1970.
71. Hughes, D. A., The sea turtles of South-East Africa. 2. The biology of the Tongaland
loggerhead turtle Caretta caretta L. with comments on the leatherback turtle Dermo-
chelys coriacea L. and the green turtle Chelonia mydas in the study region. Investi
gational Report o f the Oceanographic Research Institute, Durban, South Africa, 36,
1-96, 1974.
72. Bell, R. and Richardson, J. I., An analysis of tag recoveries from loggerhead sea
turtles {Caretta caretta) nesting on Little Cumberland Island, Georgia, jp/onufa Marine
Research Publications, 33, 20-24, 1978.
73. Witzell, W. N., Synopsis of the biological data on the hawksbill turtle {Eretmochelys
imbricata Linnnaeus, 1766), FAO Fish. Synop., 137, 1-78, 1983.
74. de Silva, G. S., The status of sea turtle populations in east Malaysia and the South
China Sea, in Biology and Conservation o f Sea Turtles, Revised ed., Bjorndal, K. A.,
Ed., Smithsonian Institution Press, Washington, D.C., 1995, 327-337.
75. Stabenau, E. K., Landry, A. M., Jr., and Callouett, C. W., Jr., Swimming performance
of captive-reared Kemp’s ridley sea ixxn\t^,Lepidochelys kempi Garman,/ . Exp. Mar.
Biol. EcoL, 161(2), 213-222, 1992.
76. Chavez, H., Marcado y recaptura de individuos de tortuga lora (Lepidochelys kempi
Garman), Inst. Nac. Invest. Biol. Pesqueras, 19, 1-28, 1968.
77. Pritchard, P. C. H., International migrations of South American sea turtles (Cheloni-
idae and Dermocheliidae), Amm. Behav., 21, 18-27, 1973.
78. Block, B. A., Booth, D., and Carey, F. G., Direct measurement of swimming speeds
and depth of blue marlin, J. Exp. Biol., 166, 267-284, 1992.
79. Bustard, H. R. and Limpus, C., Observations on the flatback turtleC/ze/on/a depressa
Garman, Herpetologica, 25, 1, 29-34, 1969.
80. Limpus, C. J., Parmenter, C. J., Baker, V., and Fleay, A., The flatback inn\Q,Chelonia
Aust. Wild!. Res., 10, 557-561, 1983.
81. Eckert, K. L. and Eckert, S. A., Pre-reproductive movements of leatherback sea turtles
{Dermochelys coriacea) nesting in the Caribbean, Copeia, (2), 400-406, 1988.
82. Duron-DeFrenne, M., Premier suivi par satellite en Atlantique d’une tortue luth
Dermochelys coriacea, C.R. Acad. Sci. Paris, 304, Ser. Ill (15), 399^ 02, 1987.
turilQ, Dermochelys coriacea, Copeia, (4), 1010-1017, 1993.
84. Eckert, S. A., Eckert, K. A., Ponganis, P, and Kooyman, G. L., Diving and foraging
behavior of leatherback sea turtles Dermochelys coriacea. Can. J. Zool., 67,
2834-2840, 1989.
85. Limpus, C. J., A benthic feeding record from neritic water for the leathery turtle
Dermochelys coriacea, Copeia, 1984, 552-553, 1984.
86. Bennett, A. R, The energetics of reptilian activity, in Biology o f the Reptilia, Gans,
C. and Pough, F. H., Eds., Academic Press, New York, Vol. 13, 151-199, 1982.
87. Can', A. R, Carr, M. H., and Meylan, A. B., The ecology and migration of sea turtles.
7. The west Caribbean green turtle colony. Bull. Am. Mus. Nat. Hist., 162, 1-46, 1978.
88. Lutcavage, M. and Lutz, P. L., Metabolic rate and food energy requirements of the
leatherback sea turtle, Dermochelys coriacea, Copeia, (3), 796-798, 1986.
89. Mrosovsky, N. and Pritchard, P. C. H., Body temperature of Dermochelys coriacea
and other sea turtles, Copeia, 1971, 624-631, 1971.
90. Prange, H. D. and Jackson, D. C., Ventilation, gas exchange, and metabolic scaling
of a sea turtle, Respir. Physiol., 27, 369-377, 1976.
91. Butler, P. J., Milsom, W. K., and Woakes, A. J., Respiratory, cariovascular and
metabolic adjustments during steady state swimming in the green turtle, Chelonia
my das, J. Comp. Physiol., 154, 154-174, 1984.
92. Jackson, D. C. and Prange, H. D., Ventilation and gas exchange during rest and
exercise in adult green sea turtles,/. Comp. Physiol, 134, 315-319, 1979.
93. Lutcavage, M. E., Bushnell, P. G., and Jones, D. R., Oxygen transport in the leath
erback sea inxilc Dermochelys coriacea. Physiol. Zool, 63(5), 1012-1024, 1990.
94. Lutcavage, M. E., Bushnell, P. G., and Jones, D. R., Oxygen stores and aerobic
metabolism in the leatherback sea turtle, Can. J. Zool, 70, 348-351, 1992.
95. Paladino, R V, O’Connor, M. P, and Spotila, J. R., Respiratory physiology of laying,
active and resting leatherback turtles on the beach in Costa Rica, in Proc. 11th Annu.
Conf. Sea Turtle Biology and Conservation, Salmon M. and Wyneken, J., Compilers,
NOAA Tech. Memo. NMFS-SEFC-232, 93, 1992.
96. Dial, B. E., Energetics and performance during nest emergence and the hatchling
frenzy in loggerhead sea turtles (Caretta caretta), Herpetologica, 43(3), 307-315,
1987.
97. Baldwin, J., Gyuris, E., Mortimer, K., and Patak, A., Anaerobic metabolism during
dispersal of green and loggerhead turtle hatchlings, Crw?/?. Biochem. Physiol, 94A(4),
663-665, 1989.
98. Meylan, A. B., Riddle of the Ridleys, Nat. Hist., 95(1), 90, 91-102, 1986.
99. Witherington, B. E., unpublished observations.
100. Wyneken, J., unpublished observations,
101. Epperly, S., personal communication.
102. Balazs, G. H., unpublished observations.
g Foraging Ecology and
Nutrition of Sea Turtles
Karen A. Bjornclal
CO NTENTS
8.1 Introduction................................................................................................... 199

8.2 Foraging Ecology.......................................................................................... 200
8.2.1 Green Turtle, C helonia my d a s ..........................................................200
8.2.2 Loggerhead, Caretta ca retta .............................................................202
8.2.3 Hawksbill, E retm ochelys im bricata ..................................................204
8.2.4 Kemp’s Ridley, L epidochelys ke m p i .................................................206
8.2.5 Olive Ridley, L epidochelys o liv á c e a ................................................207
8.2.6 Flatback, N atator d ep re ssiis .............................................................208
8.2.7 Leatherback, D erm ochelys c o r iá c e a ................................................209
8.3 Diet Selection................................................................................................ 210
8.4 Role of Sea Turtles in Structuring Foraging HabitatCommunities............. 212
8.5 Digestive Processing and Nutrition..............................................................213
8.5.1 Intake................................................................................................. 213
8.5.2 Digestion........................................................................................... 214
8.5.3 Passage of Digesta............................................................................ 218
8.6 Anthropogenic Effects on Foraging Ecologyand Nutrition......................... 218
8.6.1 Marine Debris................................................................................... 218
8.6.2 Commercial Fisheries....................................................................... 219
8.6.3 Degradation of Foraging Habitats....................................................220
8.7 Conclusions................................................................................................... 221
8.8 Acknowledgments......................................................................................... 222
References............................................................................................................. 222
8.1. IN T R O D U C T IO N
The closest interaction of an organism with its environment is the ingestion of a
subset of that environment and the subsequent alteration and absorption of that subset
as it passes through the digestive tract of the organism. The absorbed nutrients fuel
the productivity — both growth and reproduction — of the organism. The pivotal
0-8493-8422-2/97/$0.00+$ 50
role that nutrition plays in the productivity of individuals and populations — and
thus to the conservation of species — has often been overlooked.
Differences in diet, either in quality or quantity, are believed to cause the great
differences in mean growth rates of green turtles from different foraging areas in
the Hawaiian Archipelago.^ An herbivorous diet has important consequences for the
life history parameters and survival outlook of green turtles,and green turtles have
significant effects on the nutrient cycling and community structure of their seagrass
foraging habitats.A significant correlation between indices of the El Niño Southern
Oscillation and the numbers of green turtles nesting approximately two years later
at Heron Island and Raine Island is believed to have a nutritional basis, but the
mechanism is not known.^'-^ No major changes in seagrass abundance in the foraging
areas of these green turtles have been reported, so that if the cause is nutritional,
variations in quality rather than quantity of seagrasses may be responsible.^
Only by understanding the quantitative aspects of diet selection, digestive pro
cessing, and nutrition can we hope to understand the role of sea turtles in marine
ecosystems and to elucidate how nutrition acts as regulating mechanisms in the
productivity of sea turtles. As we make progress in descriptions of population
structure and in development of population models, an understanding of how nutri
tion regulates productivity will allow us to move to the next level and to address
how we can enhance or modify productivity to improve the survival outlook of these
endangered species.
8.2 FORAGING ECOLOGY

Tremendous gaps remain in our understanding of the foraging ecology of sea turtles.
For some species, little progress in the elucidation of diet and foraging habitats has
been made since reviews summarized the state of our knowledge more than 10 years
ago.^’^
Results from some of the studies reviewed here must be interpreted with care.
Small sample sizes, possible misidentification of species, evaluation of reproductive
animals, and analysis of feces may contribute to misleading conclusions. Because
the foraging habitats of most populations of sea turtles are quite different from the
habitats off their nesting beaches, food items ingested by turtles during reproductive
migrations or nesting seasons may not reflect their usual diet on the foraging grounds.
Unfortunately, the reproductive status of the turtles is often not reported. Analyzing
diet based on feces can result in the omission or underestimation of more digestible
foods. Stomach lavage offers a better option for quantitative analysis of diet.^®’^^
8.2.1 G reen T urtlE; C h e lo n ia m ydas
Young green turtles are believed to occupy open ocean pelagic habitats, perhaps in
association with sargassum rafts in some areas, after leaving the nesting beach.
It is assumed that they are omnivorous with a strong tendency to carnivory during
this life stage.^ Records of diet from small pelagic green turtles are very few. One
hatchling, swimming away from a beach in Bermuda, dived down 1 m and ate a
Foraging Ecology and Nutrition of Sea Turtles 201
small ctenophored"^ The stomach of a green turtle (25.7 cm straight carapace length)
removed from the stomach of a shark caught 107 km off Durban, South Africa was
filled with the pelagic snail Janthina janthina}^ Plasma color provides additional
evidence of a carnivorous diet in pelagic-stage green turtles. When green turtles first
arrive on benthic foraging areas in the Bahamas, their plasma is unpigmented; after
they have begun feeding on seagrasses, their plasma has yellow pigmentation,^^ a
result of plant pigments.
Green turtles leave pelagic habitats and enter benthic foraging areas at a size of
20 to 25 cm carapace length in the western Atlanticand at 35 cm in Hawaii and
Australia. At that time they shift to an herbivorous diet, and, as herbivores, occupy
a feeding niche unique among sea turtles. Green turtles feed primarily on seagrasses
and algae, although they also consume animal matter, particularly jellyfish, salps,
and sponges. The early literature on feeding habits was reviewed by Mortimer.^^’^^
Few reports are available on the diet of green turtles in the Indian Ocean and
the Arabian Sea. Green turtles in the Gulf of Mannar and Palk Bay, India feed
primarily on the seagrasses Halophila ovalis, Thalassia sp., and the alga Gelidiella
acerosa; other seagrasses and algae were taken in smaller amounts as were squid
eggs and sponges.In South Yemen, the seagrasses Posidonia oceánica, Halodule
uninervis, Syringodium isoetifolium, and Cymodocea serrulata are the main diet
species for green t u r t l e s . G r e e n turtles in Oman feed on the seagrasses Halodule
uninervis, Halophila ovalis, Halophila ovata, and the algae Chaetomorpha aerea
and Sargassum illicifolium?^ Green turtles in the Seychelles Islands and in the
Comoro Archipelago feed primarily on Thalassodendron ciliatum and ingest smaller
amounts of other seagrasses and algae.
More data are available for green turtles on the Australian shelf. For green turtles
from Moreton Bay, Queensland, Australia, the mouth contents, in order of frequency
of occurrence, were the seagrasses Halophila ovalis, Halodule uninervis, Zostera
capricorni, Halophila spinulosa, and the alga Hypnea cervicornis; green turtles in
this area also feed on the jellyfish Catostylus mosaicus?^ In 44 green turtle stomachs
from Torres Strait, red algae were dominant (42 stomachs with >5% dry mass),
brown and green algae were equivalent (14 and 13 stomachs with >5% dry mass,
respectively), and seagrasses were present in 8 stomachs at >5% dry mass.^^ Based
on stomach contents from five green turtles feeding near Sir Edward Pellew Island,
Australia, the major diet species is the seagrass Halodule pinifolia with lesser
amounts of H. uninervis and Halophila spinulosa?^
Unfortunately, only preliminary results are available from the most quantitative
study to date on feeding habits and diet selection in green turtles.^® Based on stomach
lavage of 518 green turtles feeding on the reefs surrounding Heron Island, Australia,
the diet was composed of 38 species of red algae, 21 species of green algae, and 10
species of brown algae. Gelatinous animal tissue — Physalia and mollusc egg cases
— were also consumed.
Records of the diet of green turtles are available from a number of locations in
the rest of the Pacific. Algae comprise the diet of green turtles in the Ogasawara
Islands, J a p a n ; i n Tokelau, South Central Pacific;^^ and at Johnston Atoll.^^ The
diet of green turtles in the Hawaiian Archipelago includes 56 species of algae (of
which 9 species are the major diet components), 1 species of seagrass, and 9 types
of i n v e r t e b r a t e s F i v e species of algae intentionally introduced into Hawaii for
aquaculture projects are consumed by green turtles.
Green turtles along the East Pacific coast may have more carnivorous diets than
those in other regions. Algae were found in all stomachs from 20 green turtles
captured in Peru, jellyfish in 12 stomachs, fish in 12 stomachs, and molluscs in 10
stomachs.Analysis of stomach contents from 39 green turtles (carapace lengths
52 to 89 cm) captured near Pisco, Peru, revealed the following percent occurrences:
molluscs (primarily Mytilus, Nassarius, and Semele), 64%; algae (mainly Gigartina
and Rhodymenia), 51%; polychaetes, 49%; jellyfish and commensal amphipods
{Hyperia medusarum), 31%; fish and fish eggs, 23%; salt grass {Distichlis), 18%;
and crustaceans, 13%.^^ The stomach of a large (71-cm carapace length) green turtle
captured in a pelagic habitat 40 nautical miles off the coast of Manta, Ecuador
contained fish eggs along with a small amount of sargassum to which the eggs had
been attached."^® The digestive tracts of 7 green turtles captured off the coasts of
Colima and Jalisco, Mexico contained green and red algae (4 turtles) and a great
diversity of invertebrates and chordates (all turtles)."^‘
In the Caribbean, the seagrass Thalassia testudinum is the primary diet species
for the green turtle.^ '^^ Thalassia comprised 87% of the dry mass of contents from
202 stomachs from green turtles captured on foraging grounds off the Caribbean
coast of Nicaragua; other seagrasses (Syringodium filiforme, Halodule wrightii)
made up an additional 5%."^^ On the southwestern Cuban Shelf, 14 green turtles had
fed primarily on the seagrass T. testudinum and the sponge Chondrilla nuculad"^
Along the coast of Brazil, algae are the primary diet.'^^ '^^ Green turtles in a lagoon
in Florida, fed primarily on the seagrasses S, filiforme and H. wrightii with lesser
amounts of the seagrass Halophila engelmanni and red and green algae.Small
green turtles in the waters of Long Island, New York fed on the seagrass Z. marina
and on 5 species of algae — 3 green and 2 brown.
8 .2 .2 Loggerhead , C aretta caretta
For the first few years of life, loggerheads forage in open ocean pelagic habitats,
often — at least in the Atlantic — in association with sargassum. ^‘ Loggerheads
begin to move into shallower waters where they forage over a variety of benthic
hard- and soft-bottom habitats at a size of about 40 to 50 cm carapace length in the
Atlantic^® and at 70 to 80 cm on the Australian continental shelf and southwestern
Pacific basin.'’^ One or more large loggerheads were found associated with two
submarine geothermal springs on the West Florida Shelf, Gulf of Mexico.^^ Benthic
epifauna and nekton were considerably enriched in the area around the springs.
Diet items from post-hatchling loggerheads captured near the nesting beach have
been recorded off the coasts of South Africa and the southeastern U.S. Stomach
contents from 37 loggerhead hatchlings stranded on Cape Agulhas, South Africa,
yielded Physalia physalis, fragments of algae, bark, feathers, and fine grit that may
have been pieces of Janthina shells.Stomachs of five loggerhead hatchlings washed
ashore in Florida, in masses of sargassum during a hurricane contained sargassum
floats and leaf parts, two snails (Litiopa melanostoma) that associate with sargassum,
the pelagic snail Diacria trispinosa, and pieces of crustacean appendages.^^ Two
dead loggerhead hatchlings, both with small yolk sacs, were found at the west wall
of the Gulf Stream 93 km east of Florida, and their guts contained three categories
of food: terrestrial insects, marine animals, and marine plants.The insects included
ants (winged, sexual forms), a fly, an aphid, leafhoppers, planthoppers, beetles, and
unidentified thorax fragments. The authors suggested that aerial insect plankton
settled on the ocean surface and were accumulated in convergence zones. Marine
animals that had been ingested were hydrozoan colonies, goose and acorn barnacles,
amphipods, crab zoea, shrimp eye capsules, planktonic fish eggs, and unidentified
muscle and chitinous fragments. Marine plants ingested were sargassum leaf and
bladder fragments and unidentified algal fragments. Major diet items flushed from
the stomachs of 42 post-hatchlings off the coast of Florida included gelatinous
animals (predominately medusae and ctenophores), crustaceans (primarily larval
shrimp and crabs), hydrozoans, insects, gastropods, sargassum, and an actinid anem-
one.^^
The diet of pelagic juvenile loggerheads has been recorded in several studies.
Brongersma^^' reviewed the early literature on feeding habits of pelagic loggerheads
in the Atlantic. The stomachs from loggerheads taken near the Azores and Madeira
contained salps, jellyfish, amphipods Hyperia medusarum that associate with medu
sae, pteropods Hyalaea tridentata, the crab Nautilograpsus {= Planes) minutus,
bunches of the barnacle Lepas anatifera, many sygnathid fish Entelurus aequoreus,
squid, and the pelagic snail Janthina. Brongersma^^ expressed doubt that loggerheads
could catch fish, but suggested that the slow speed of the sygnathid fish and the
mass mortalities of these fish that have been observed in the area could account for
the presence of these fish in the loggerheads. A careful analysis of the gut contents
of five juvenile loggerheads from the waters around Madeira, Selvagens Islands, and
the Azores revealed that the major diet components were pelagic coelenterates
(primarily siphonophores and to a lesser extent scyphomedusae and hydromedusae),
salps {Pyrosoma atlanticum), gastropods {Janthina spp. and Pterotrachea spp.),
barnacles {Lepas spp.), and the isopod Idotea metallicaJ'^ The jellyfish Pelagia
noctiluca is an important prey species for North Atlantic loggerheads."^^ Food items
from six pelagic-stage loggerheads (5.2 to 30.0 cm straight carapace length) that
stranded dead along the south Texas coast included sargassum, jellyfish, Janthina,
Litiopa melanostoma, and decapod and stomatopod larvae.^^ Pelagic loggerheads in
the North Pacific feed primarily on Velella velella and Janthina sp.^'^
Dodd^'^ presented a thorough species list of diet items found in the digestive
tracts of all life stages of loggerheads. Loggerheads feeding in benthic environments
ingest a wide range of invertebrates. Some of these prey items would seem to yield
relatively low nutrient gain, such as horseshoe crabs, Limulus polyphemus, which
is the most common prey item of loggerheads in Virginia.^^ Post-pelagic-stage
loggerheads feed throughout the water column, capturing jellyfish and salps at the
surface and middle depths, but they apparently concentrate their foraging efforts on
the bottom. Loggerheads in Moreton Bay, Australia create broad depressions in the
substrate with sweeping movements of the front flippers, thus exposing burrowing
bivalves, such as Anadara trapezia, Pinna bicolor, and Solen grandis, which the
loggerheads then ingest.^* Loggerheads in this area also feed on the anemone Sti-
chodactyla haddoni and the crab Portuniis pelagicus.
Since the review by Dodd,‘’^ two quantitative analyses of loggerhead feeding
habits have been published. Feces were collected from 25 loggerheads captured in
the waters around Long Island, New York.^^ Crabs (Libinia emarginata. Cancer
irroratus, Pagurus pollicaris, and Ovalipes ocellatus) were present in the feces from
90% of the turtles, molluscs {Mytilus edulis and Busycon spp.) in 40% of the turtles,
and algae {Sargassum natans, Ulva sp., and Fucus spp.) in 20% of the turtles. As
noted by the authors, prey that undergo extensive digestion, such as jellyfish, will
be missed in a study based on feces. The stomach contents of another loggerhead
from this region yielded 18 seahorses {Hippocampus erectus), invertebrates, and
marine plants.
Digestive tract contents were analyzed from 82 loggerheads that stranded dead
in south Texas over a three-year period.^"^ The sea pen (Virgularia presbytes) was
the most common prey with a frequency of occurrence of 56% and comprising 59%
of the total prey dry weight. Crabs were the second most important prey item with
a frequency of occurrence of 88% and making up 29% of the total prey dry weight.
Nine species of crabs were identified; walking, or benthic, crabs were most impor
tant, while swimming (portunid) crabs were less commonly consumed. Molluscs,
tube worms, barnacles, fish, vegetation, sea pansies, whip corals, sea anemones,
mantis shrimp, penaeid shrimp, and Physalia physalis were also identified in the
gut contents. Significant dietary shifts occurred among seasons. Loggerheads fed
mainly on sea pens in the spring, and crabs were the major diet component in summer
and fall when crabs increase in abundance in the region.
8.2.3 H awksbill, E retm o ch elys im b r ic a t a
Post-hatchling hawksbills apparently are pelagic, often living in close association

with floating rafts of SargassumJ^'^^^^^ As with all species of sea turtles, data on the
diet of the pelagic stage of hawksbills are limited. Much of the material in the
digestive tracts of four hawksbills (14.0 to 21.3 cm straight carapace length) that
stranded dead on Florida beaches could not be identified, but Sargassum spp. (either
S. fliiitans or S. natans) and unidentified animal matter were found in large propor
tions in at least one of the animals.Small quantities of the following food items
were also found: Syringodium filiforme, Microdictyon sp., woody plant remains,
shell fragments of goose barnacles, eggs of pelagic fish, tunicate, crab chela, and
unidentified algae.
Hawksbills leave the pelagic and begin foraging in benthic habitats at a minimum
straight carapace length of 20 to 25 cm in the Caribbean^^ and at a minimum curved
carapace length of 35 cm in Australia.^^ Hawksbills forage most commonly over
coral reefs and rock outcroppings, but also feed over seagrass pastures in mangrove-
fringed bays.^^’^^’^®
In a thorough study of gut contents from post-pelagic hawksbills, Meylan^^
deposed the reigning theory that hawksbills were indiscriminate omnivores and
demonstrated that they specialize on a diet of sponges, at least in the Caribbean.
Sponges comprised 95.3% of the dry mass of total contents from the digestive tracts
of 61 hawksbills from throughout the Caribbean.^^ Hawksbills feed very selectively:
98.9% of the dry mass of all identified sponges from hawksbills were from 3 of the
13 orders of demosponges (Astrophorida, Hadromerida and Spirophorida). Hawks
bills avoid sponges with spongin, but silica spicules apparently are not a feeding
deterrent. The 10 highest-ranked prey sponges (in descending order) were Chondrilla
nucula, Ancorina sp., Geodia sp., Placospongia sp., Suberites sp., Myriastra sp.,
Ecionemia sp., Chondrosia sp., Aaptos sp., and Tethya cf. actinia.
Studies conducted in Cuban waters report similar results. In 8 mature hawksbills
taken during the breeding season in the Gulf of Batabano, all had sponges in their
stomachs (all contained Chondrilla nucula, one had consumed T. diploderma, and
two had unidentified sponge species)."^"^ In a larger study of 73 hawksbills, sponges
comprised over 90% of the diet of all hawksbills between 50 and 80 cm.^^ Nine
sponge species were identified; the 5 most common were C. nucula, Chondrosia
collectrix, G. gibberosa, Erylus ministrongylus, and T. aurantia.
At Mona Island, Puerto Rico, demosponges have been found to comprise essen
tially the entire diet of subadult hawksbills.^^ Another study reported the gut contents
of six hawksbills from Puerto Rico: one adult had 95% of the sponge Chondrilla
niicula, one adult had 90% of the sponge G. neptuni, one adult had ingested only
the sea cucumber Holothuria cubana, the stomachs of two juveniles contained only
the sponge C. nucula, and one juvenile had fed on four species of demosponges.^^
The stomachs from an additional eleven hawksbills from Puerto Rico were full of
demosponges, with C. nucida the most common species.^"^ Other accounts of con
sumption of sponges by hawksbills in the Atlantic system are reviewed by Mey-
lan.66-67
Invertebrates other than sponges have been reported as major diet components
in the Atlantic, and marine plants have been noted in gut contents. Other than
sponges, tunicates were the major diet component of 20 hawksbills captured off
Tortuguero, Costa Rica.^'^ The Costa Rican hawksbills had also ingested bryozoans,
coelenterates, molluscs, and marine plants. The sea m tm ont Anemonia sulcata made
up the bulk of the gut contents of a hawksbill from Selvagem Pequeña (north of the
Canary Islands), which also included other coelenterates, sponges, oceanic squid,
gastropods, a sea urchin, a spider crab {Inachus sp.), and algae.^^ Several marine
plants were recorded in the digestive tracts of Caribbean hawksbills
Early diet records for hawksbills in the Pacific and Indian oceans suggested a
more omnivorous diet; large quantities of vegetation such as Sargassum, Rhizophora,
Cymodocea, and unidentified algae were reported from stomach c o n t e n t s . T w o
hawksbills from the Philippines had seagrass, sponges, and algae (Eucheuma and
Codium) in their stomachs.^^ WitzelP® summarizes other accounts of hawksbills
feeding on a wide range of diet items in the Pacific.
More recent studies, however, report that sponges form the preponderance of
the diet in the Pacific and Indian oceans. Hawksbills feed on sponges and soft corals
in the Seychelles,and sponges were dominant in the digestive tract from 35
hawksbills from Cosmoledo, Seychelles (J. Mortimer, personal communication in
Meylan).^*" An immature hawksbill from Mohéli, Comores had consumed four types
of demosponges.^^ The stomachs of two subadult hawksbills from Masirah Island,
Oman contained only sponges.A mature hawksbill from Oahu, Hawaii had at least
three species of sponge in its digestive tract.^^ Sponges were the major food in two
hawksbill stomachs from South Africa and one from Pacific Panama.^'^ All reports
from the Pacific and Indian oceans are based on only a few turtles (other than the
personal communication of J. Mortimer); larger studies are needed to establish the
diet of hawksbills in the Pacific.
Hawksbills may undergo a period of omnivorous feeding in benthic habitats
before they adopt the specialized spongivory of larger juveniles and adults. Data
from three studies support this theory. First, two of the smallest individuals among
38 hawksbills studied by Meylan,^^ had significant quantities of nonsponge material
in their digestive tracts: a 23-cm hawksbill and a 26-cm hawksbill had consumed
invertebrates other than sponges, fish, and substrate (58 and 22% by dry weight,
respectively) in addition to sponges (42 and 78%, respectively). Three other hawks
bills in this size class had ingested 95 to 100% sponges. Second, feces collected
from a 33-cm hawksbill captured in the Miskito Cays, Nicaragua contained the red
alga Coelothrix irregularis (70% of the feces volume), tubes from two species of
polychaetes, sponge spicules, a hydroid, and pieces of a snail shell and a pelecypod
shell.Third, in Cuba, stomach contents were analyzed from 73 hawksbills ranging
in size from 30 to 90 cm carapace length.^^ Although sponges represented over 90%
of the diet for individuals larger than 50 cm, the stomachs from the two smallest
individuals had significant quantities of other foods. The red alga Gracilaria sp.
composed 84% (by weight) of the stomach contents from a 30-cm individual, and
the stomach from a 40-cm hawksbill contained 33% of an unidentified ascidean. In
both turtles, the rest of the stomach contents were sponges. Although the data in
support of a transitional benthic diet are limited, such a diet would support the turtles
as it adapts to the challenges — both mechanical (sharp silica spicules) and chemical
(toxic compounds) — of feeding on a sponge diet.
Purposeful ingestion of the coralline substrate^^ and substantial quantities of the
calcareous algae Halimeda incrassata^^ by gravid female hawksbills has been noted.
The authors suggest gravid females may consume these items as a source of calcium
for eggshell production.
8 . 2 .4 Kemp' s R idley, L epid ochelys kempi
Kemp’s ridleys apparently forage in surface waters during an early pelagic stage
that lasts until the turtles reach approximately 20 cm carapace l e n g t h . 84
size they move into relatively shallow water (less than 50 m) benthic foraging areas
with unconsolidated substrates.They will continue to forage in these areas, some
times moving long distances between feeding grounds.
The only report of diet of small, pelagic-stage Kemp’s ridleys is from two
individuals of less than 10 cm carapace length that stranded dead on south Texas
beaches.These turtles had consumed Sargassum sp., pelagic molluscs {Recluzia
rollandiana, Cavolina longirostris, and Litiopa melanostoma) that are often associ
ated with sargassum, and crabs that could not be identified.
Post-pelagic Kemp’s ridleys feed primarily on crabs. Crabs comprised the major
ity of the diet in 101 Kemp’s ridleys that stranded dead on Texas beaches with a
percent occurrence of 1171% and percent of total dry mass of digestive tract contents
of 93.6%.^^ Other components were molluscs (62.4% occurrence and 2.2% dry
mass), fish (25.7 and 0.4%), vegetation (61.4 and 0.3%), and shrimp (8.9 and 3.2%).
The turtles consumed large quantities of both portunid crabs (Arenaeus cribrarius,
Callinectes sapidus, C. similus, Portunus gibbessii, and Ovalipes floridanus) and
other crabs (Hepatus epheliticus, Libinia sp., and Persephona mediterranea). This
observation contradicts the earlier idea that Kemp’s ridleys specialized on portunid
crabs; the distribution of Kemp’s ridleys may relate to the distribution of all major
crab species consumed rather than to only the distribution of portunids.^^ The small
quantities of vegetation probably were ingested incidental to other items in post-
pelagic turtles, and the small amounts of fish and shrimp ingested were probably
the result of scavenging on dead bycatch from shrimping vessels or discarded bait.^^
The presence of Nassarius spp. — molluscs that scavenge on dead tissue — in 15%
of those turtles that had consumed shrimp and in 50% of those turtles that had
consumed fish support the latter conclusion.
Kemp’s ridleys in coastal waters of New York State also feed primarily on
c r a b s . S p i d e r crabs (L. emarginatus) were by far the most commonly consumed,
followed by rock crabs {Cancer irroratus) and lady crabs {O. ocellatus). Molluscs
(Argopectin irradians, Mytilus edulis, and N. trivitattus) were consumed in smaller
quantities; small amounts of vegetation {Fucus sp., S. natans, Ulva sp., and Z.
marina) also were ingested. Seahorses {Hippocampus erectus) — a fish that relies
on crypsis rather than flight to avoid predation — comprised a substantial volume
of the gut contents in two Kemp’s ridleys from Long Island, New York.^^ Marquez-
lists other accounts of the feeding habits of Kemp’s ridleys.
8 .2 .5 O live R idley, L epid ochelys o livácea
The habitat of juvenile olive ridleys is not known; few sightings of size classes
between hatchlings and adults have been recorded.Adult olive ridleys apparently
utilize a wide range of foraging habitats. They feed in deep water, as indicated by
capture in bottom trawls at depths of 80 to 110 m.^^ They are common in pelagic
h a b i t a t s , a n d they feed in relatively shallow benthic waters, sometimes near major
estuaries.^'^’^^
The largest study of the diet of olive ridleys was conducted on 139 mature turtles
(115 females and 24 males) that were captured off the Escobilla nesting beach in
Oaxaca, Mexico.^^ Salps (all in the genus Metcalfina, possibly the species M. hex
ágono) and fish composed the largest percentage of the total volume of stomach
contents (45 and 42%, respectively), but occurred in only 14 and 5% of the turtles,
respectively. The other constituents were molluscs (5% volume and 66% frequency
of occurrence), crustaceans (4 and 48%), algae (2 and 51%), bryozoans (0.2% and
13%), and fish eggs, sipunculids, and ascidians (combined 0.1 and 11%). Percent
volume and occurrence varied greatly among months and between sexes. Given the
large sample size, the high volume percentages for some of the constituents with
low percent occurrence (i.e., fish were found in only 5% of the turtles, but accounted
for 42% of total stomach content volume) indicates that many of the stomachs must
have contained only a small volume of contents. Because these turtles were in a
reproductive season, their stomach contents may not reflect the diet of olive ridleys
on their foraging grounds.
There are other reports for olive ridleys from Mexico. Jellyflsh and crabs were
the major components of the stomach contents of 20 adult females captured off the
state of GueiTero, Mexico.^^ Five olive ridleys, presumably captured in Mexican
waters, contained only Pleuroncodes planipes,^^ The stomachs of six olive ridleys
captured off the coasts of Colima and Jalisco, Mexico contained primarily benthic
invertebrates: crabs, pelecypods, and gastropods."^^ One turtle had consumed a scy-
phozoan medusa {Pelagia sp.).
Fritts^^ reviewed the notes of J.R. Slevin (1905 to 1906); Slevin reported that
fish eggs filled the digestive tracts of two female olive ridleys captured near the
Galápagos Islands, and small crabs filled the stomachs and intestines of two male
olive ridleys captured south of Cocos Island. The stomachs of two olive ridleys
captured about 40 nautical miles off Ecuador contained clumps of medusae."^^ Crabs,
jellyfish, and tunicates have been recorded in the stomachs of olive ridleys from the
eastern Pacific (unpublished reports cited in Mortimer).^ A mature female of 61-cm
carapace length found in New Zealand waters had been feeding extensively on
pelagic tunicates Pyrosoma and Salpa.^"^
The diets of olive ridleys in the Indian and Atlantic oceans are poorly known.
Deraniyagala^^ reported algae as the main diet component in olive ridleys from Sri
Lanka, although young pearl oysters and the sea urchin Clypeaster humilis were
also ingested. Stomachs from adult olive ridleys (number not stated) captured in
India were full of algae.^'’ The digestive tract from a female with about 200 ovarian
eggs caught at a 20- to 24-m depth off Surinam contained two small catfish, ten
snail shells, three small crab carapaces, and about 2 1 of what was believed to be
partially digested jellyfish.^^^
8 .2 .6 Flatback, N a t a t o r depressus
The flatback may be the only species of sea turtle that does not have an early pelagic
stage as surmised from the presence of remains of post-hatchlings (11- to 21-cm
carapace length) at island feeding stations of the white-bellied sea eagle, Haliaeetus
leucogaster!^'^’^^ The early post-hatchling stage is apparently spent within tens or
hundreds of kilometers of their natal beaches over the Australian continental shelf
where they inhabit shallow, turbid, weakly flushed, coastal waters protected from
ocean s w e l l s . P r i m a r y feeding habitat of larger flatbacks appears to be turbid,
shallow inshore waters off northeastern Australia and in the Gulf of Carpentaria,
most commonly in areas 5 to 20 m deep.^^ '^'^
The diet of the flatback is poorly documented. Stomach contents from two small
stranded flatbacks (14- and 22-cm carapace lengths) contained both planktonic
organisms (snails Janthina, siphonophores Porpina) and benthic organisms (corals,
molluscs, bryozoans), all of which could have been consumed in inshore waters.
Larger subadults and adults feed on jellyfish and on soft-bodied benthic invertebrates,
such as sea pens and soft corals, in habitats with unconsolidated substrates.
8 . 2 .7 Leatherback, D erm o ch elys co riácea
The leatherback is believed to be the most pelagic of all sea turtles, spending much
time in the open oceanHowever, leatherbacks will forage close to shore and over
continental s h e l v e s a n d have been reported feeding in water of less than 4-m
depthd^^"^ The distribution of foraging leatherbacks appears to be largely dependent
upon the distribution of the jellyfish, salps, and other gelatinous organisms upon
which they feed^^^'^®^ and which often accumulate near convergent zones or water-
mass boundariesd^^ Leatherbacks feed throughout the water column, from the
s u r f a c e t o great depths. A leatherback fed on octopus used for fish bait at a
depth of 50 m in western Australia.^The presence of a deep-water siphonophore
(Apolemia uvarid) in the stomach of a Mediterranean leatherback suggests leather
backs may feed at great depths; however, this siphonophore may be carried to the
surface by upwelling water.^^ Internesting leatherbacks off St. Croix may feed on
siphonophores, salps, and medusae within the deep scattering layer at n i g h t ; t h e
mean nocturnal dive depths for six leatherbacks ranged from 31.8 to 83.9 m.
Because leatherbacks apparently inhabit the pelagic zone throughout their lives
and because the ability to capture and consume the gelatinous prey species is not
size dependent, there would seem to be no reason for a diet shift between size
classes, as is apparently the case in other sea turtle species. A 15.6-cm leatherback
captured 200 to 250 nautical miles southwest of Acapulco, Mexico had jellyfish in
its digestive tract.
Bleakney^**^ and Brongersma*^^ established that the leatherback feeds primarily
on scyphomedusae, pelagic tunicates, and their commensals, parasites and prey. The
five leatherbacks examined by Bleakney*^^ had consumed the jellyfish Cyanea capil-
lata arctica and its amphipod commensal Hyperia medusarum. Diet records since
that time have all supported the theory that leatherbacks specialize on gelatinous
organisms. The stomach of a leatherback captured near Malta contained three species
of pelagic coelenterates: the siphonophore A. uvaria, an unidentified siphonophore,
and an unidentified scyphozoan.^^ The gut contents of six leatherbacks from southern
England and the North Sea contained almost entirely scyphozoan medusae of the
species C. capillata, C. lamarckii, Rhizostoma octopus, Aurelia aurita, Pelagia
noctiluca, and Chrysaora hysoscella; the leptomedusa Ae^w6>rea sp.; portions of the
hydroid Obelia dichotoma; many amphipods (scyphozoan commensals) H. galba;
and fragments of a fish, a crab, and algae. Leatherbacks off the coast of France
feed almost entirely on R. pulmo}'^'^ Libinia spinosa, a small crab that is a jellyfish
commensal, was reported from a leatherback captured off the coast of Uruguay.
There are a number of records of leatherbacks feeding on Stomolophus meleagris
off the Atlantic coast of the U.S. and in the Gulf of Mexico. A leatherback was
observed feeding on Aurelia sp. off the coast of Washington, U.S.^^^ Leatherbacks
in Australia are reported to feed on the jellyfish Catostylus mosaicus}^^ Davenport
and Balazs*^'^ compiled records of leatherbacks consuming pyrosomas — pelagic
colonies of tunicate zooids — from the Pacific and Mediterranean.
8.3 D IE T S E L E C T IO N
The development and duration of diet preferences in sea turtles are not understood.
Studies of early diet preferences in loggerhead hatchlings revealed that hatchlings
do forni preferences based on early feeding experience, but that these preferences
are reversible.
The extent to which the diet of green turtles is detemined by selective feeding
or by the relative abundance of different diet species has been addressed in several
studies. Green turtles in the Caribbean feed selectively by recropping plots of
Thalassia testudinum that they have previously g r aze d. By grazing on the short
leaves in the grazed plots rather than feeding randomly on Thalassia blades, they
ingest a diet that is higher in nitrogen and lower in lignin, a plant constituent that
is indigestible and lowers the digestibility of other structural carbohydrates in the
plants.
Fen'eira'^'’ attributed the high frequency of red algae in green turtle stomachs to
the greater abundance of red algae in the feeding areas. Balazs^^ suggested that both
relative abundance and feeding selectivity determine the feeding habits of Hawaiian
green turtles. Three small green turtles caught off Necker Island, Hawaii, feeding
over dense stands of Asparagopsis sp., Caulerpa sp., Laurencia sp., Sargassum sp.,
and Turbinaria sp., had only Caulerpa in their stomachs,suggesting they were
selecting that alga. In a series of foraging areas in Hawaii, the relative abundance
of available algae was quite different from the relative abundance of algae in stom
achs of green turtles.^"^ Sazima and Sazima"^^ reported that green turtles visually scan
the potential foraging area while slowly moving among algal patches. They suggest
that this scanning behavior and the differences between the algae available and the
algae consumed indicate that the turtles are feeding selectively. Ross^^ also attributed
the differential abundance of algal species in the habitat and in the diet of green
turtles in Oman to selective feeding. Garnett et al.^^ suggested that diet selection is
based on relative availability of diet species, although some selectivity for soft red
algae was detected in their study of stomach contents from 44 green turtles in Torres
Strait, Australia. In a preliminary report for a very thorough study of diet selection
in 518 green turtles that had ingested 69 species of algae near Heron Island, Australia,
Forbes^® concluded that green turtles did not ingest a diet based on relative abun
dances of the algae in the foraging habitat, but rather fed selectively.
BjomdaF '^^ hypothesized that microbial communities in the digestive tracts of
green turtles may affect diet selection, as suggested by three lines of evidence. First,
as reviewed by Mortimer,^ in many areas where both seagrasses and algae are present,
green turtles feed on either algae or on seagrasses, not on a mixture. Second, in
turtles (and dugongs) that feed primarily on seagrasses, algae appear undigested in
the feces in contrast to the very digested appearance of the seagrass (references in
Bjorndal)."^^ Conversely, in green turtles that feed primarily on algae, blades of
Thalassia in the posterior colon appear undigested, again in contrast to the very
digested appearance of algae surrounding them.^^^ Third, structural carbohydrates
in seagrasses are quite different from those in marine algae; structural carbohydrates
also vary considerably among algae. The microbial communities in the digestive
tracts of green turtles feeding on seagrasses would differ from those in green turtles
feeding on algae. BjorndaP suggested that these differences in microbial populations
could affect diet selection because turtles with gut microbes adapted to algae would
digest seagrasses less efficiently, and vice versa. Although gut microbial communities
adapt to long-term diet shifts by varying the number and relative abundance of
microbial species/^^ turtles would digest food less efficiently if they made succes
sive, short-term diet shifts. There is a parallel situation in Orkney sheep that feed
on algae on North Ronaldsay Island and that are occasionally shifted to pastures of
angiosperms.^^"^'^^^
Specificity of the gut microbial communities may be one component in the
optimal foraging strategy of the green turtle, but it will not overwhelm all others.
The extent of the restriction on diet selection in green turtles that BjorndaP hypoth
esized has been overemphasized by researchers working with sea turtles (e.g., Gar
nett et al.).^^ Green turtles, like all organisms, select a diet that will yield the most
nutrition for the least investment in search and handling costs. When vast pastures
of seagrass or algae are available, the optimal forage for green turtles may be that
to which its gut microbes are adapted (either all seagrass or all algae). However,
where food is limited or where food types are more dispersed, the greater search
and handling costs of seeking either an all-algae or all-seagrass diet may be greater
than the energy gain from more efficient digestion. In this case, the turtle would
ingest a mixed diet.
Animals consistently ingesting a mixed diet would almost certainly develop a
microbial community capable of degrading the various complex carbohydrates. In
some areas of Australia, green turtles ingest both seagrasses and algae and, in the
feces, both components have the appearance of being equally digested.H ow ever,
the microbial populations in these turtles would have to adapt continually to any
changes in the proportions of seagrass to algae and in the proportions of the various
algae. Even if green turtles ingesting a mixed diet have lower digestive efficiencies,
their nutrient gain may well be maximized by the ability to ingest a greater quantity
of the mixed diet more rapidly.
In an assessment of selective feeding in Kemp’s ridleys,^^^ relative abundance
of four species of crabs (two swimming crabs [Ovalipes ocellatus and Callinectes
sapidus] and two walking crabs [Libinia emarginata and Cancer irroratus]) in the
foraging areas of Kemp’s ridleys in New York waters were compared with the
abundance of those crabs in the digestive tracts of Kemp’s ridleys. The turtles
ingested more walking crabs and fewer swimming crabs than would be expected
from the abundance of the crabs in the environment, indicating that the small ridleys
in New York waters feed on slower-moving prey perhaps because the turtles have
just moved out of the pelagic habitat and are inexperienced foragers in benthic
habitats. By the time these turtles have moved south into Chesapeake Bay, Virginia
they are more adept predators and are better able to capture the swimming crab
Callinectes sapidus, which is their primary prey species in that region.
8.4 ROLE OF SEA TURTLES IN STRUCTURING

FORAGING HABITAT COMMUNITIES
Sea turtles can have major effects on nutrient cycling and community structure in
their foraging habitats. Under natural conditions, the high population densities that
sea turtles can attain make them major predators and grazers in their ecosystems.
Seagrass ecosystems, found in relatively shallow waters worldwide, are among the
most productive in the world.^^9 Grazing by green turtles has significant effects on
the structure and nutrient cycling in these systems. Green turtles establish and
maintain grazing plots in pastures of the seagrass T. testudinum by continually
recropping areas that they have earlier grazed and thus ingest a diet higher in protein
and lower in lignin."^^ These plots vary in size from 10 to 100 m^ and can be
maintained for over two y e a rs .G re e n turtles increase the speed of nutrient recy
cling in Thalassia beds by shortening the time required for normal decomposition
of Thalassia blades, perhaps as much as from eight weeks to a few days."^ Thalassia
blades that are consumed by green turtles and deposited in feces have greatly reduced
particle size and lower ( = enriched) C to N ratio compared to blades that die and
decompose in situd Continual cropping by green turtles can also stress Thalassia
plants, resulting in lower blade productivity and reduced leaf width, rhizome diam
eter, and leaf-shoot d e n s i t y . T h e stress may be caused by reduced nutrient
availability, which may result from decreased trapping of detritus within the seagrass
bed by blades that have been grazed to a few centimeters in height.'’ Thus, grazing
by green turtles can have both positive and negative effects on seagrass nutrition.
In addition to effects on nutrient cycling and plant moiphology and productivity,
grazing by green turtles can enhance invasion by early successional species and
change faunal densities and predator-prey relationships.^ Although never quantified,
grazing on algae by green turtles on coral reefs must have significant effects on
percent cover by algae in these ecosystems.
By preying upon sponges in coral reef habitats, hawksbills may affect diversity,
biomass, and succession in complex reef communities.^^ Sponges are major contrib
utors to reef biomass and compete with other reef organisms for space. Predation
by hawksbills, at natural population densities, could be a significant factor in this
competition. On a submerged reef in Puerto Rico, the haplosclerid sponge Niphates
digitalis was the species most commonly grazed by hawksbills of 26 sponge species
found in the area. Between 67 and 86% (n = 26) of all specimens of N. digitalis
had been grazed by hawksbills.^^ Hawksbills also make sponge tissue accessible to
other predators by biting through the tough outer covering of sponges, exposing the
soft inner sponge tissues to fish.^^
Loggerheads, as major predators on invertebrates, may affect community struc
ture in benthic habitats. Predation by loggerheads may be a major mortality factor
for saucer scallops (Amusium japonicum halloti) around Bundaberg, Australia^^"^ and
for large juvenile and adult queen conch {Stromhus gigas) in the Bahamas.*^'’
Studies are needed to explore the extent of competition among species of sea
turtles and between sea turtles and other species. In general, the different food habits
and foraging habitats of sea turtle species minimize competition for food resources
among sea t u r t l e s , b u t food resources are shared in some areas. Considerable
diet overlap was reported for juvenile Kemp’s ridleys and loggerheads in the waters
around Long Island, New York,^^ whereas larger size classes of the two species were
reported to exhibit habitat and diet partitioning in Virginia w a te rs .In one of the
few studies of diet overlap between sea turtles and other species, Sazima and
Sazima"^^ report that large herbivorous fishes {Acanthurus, Kyphosus, Sparisoma)
select different algae species than do green turtles in the same area, thus avoiding
competition.
All sea turtles serve as nutrient exporters from their foraging habitats. Some
species export nutrients on a daily basis by moving out of foraging areas to resting
habitats and defecating there. All species export nutrients on an annual basis when
mature individuals leave foraging habitats to migrate to nesting beaches where they
deposit substantial nutrient packets — in the form of eggs — that have been produced
from nutrients gleaned and stored on the foraging grounds.
8.5 DIGESTIVE PROCESSING AND NUTRITION

Only a few studies of nutrition and digestion in sea turtles have been undertaken.
Most studies have been conducted on the herbivorous green turtle; these studies have
limited application to the other, carnivorous, species of sea turtles. A knowledge of
quantitative nutrient requirements would help us understand how nutrition regulates
productivity, but there have been few such studies. Requirements for seven amino
acids have been determined for hatchling green t u r t l e s , a n d a requirement for
vitamin A in hatchling green turtles has been suggested.
8.5.1 I ntake
Intake — or the rate of food ingestion — has been measured in free-ranging green
turtles feeding on the seagrass T. testudinum in the Greater Caribbean in two stud-
ies.42,141 different approaches were used to estimate intake: BjomdaL^ used total
daily feces production and percentages of an indigestible marker in feces and in
Thalassia, and Williams used estimates of mean dive time, mean number of bites
taken per dive, amount of leaf shoot removed per bite, and mean number of hours
spent feeding per day. The two techniques yielded similar values for intake (Table
8 . 1).
Most physiological rates in poikilotherms are affected by temperature, and both
intake and passage rate in herbivorous reptiles are positively related to temperature.
Intake of trout pellets in post-hatchling green turtles increased significantly between
18 and 33°C: Qjo 18 to 23°C = 9.64, Qô 23 to 28°C = 2.82, and Qio 28 to 33°C =
1 .97.^43
Diet quality also has a substantial effect on intake. Green turtles of similar size
feeding on a high-quality pelleted diet^"^"^ had intakes three to four times greater than
those for a diet of Thalassia (Table 8.1). These differences can have significant
effects on nutrient gain. Combining these intake levels with the higher nitrogen
content of the pelleted diet and the greater digestibility of nitrogen (see Tables 8.2
and 8.3), green turtles on the pelleted diet consume at least 12 times more nitrogen
on a daily basis than do green turtles feeding on Thalassia.
TABLE 8.1
Intakes (g dry mass day~* turtle *) of Green
Turtles Feeding on Thalassia testudinum
Turtle size mean
(range) (kg) Intake Ref.
8 (7-9) 24 42
30 (28-32) 82 42
48 (46-50) 177 42
66 (64-68) 218 42
26 (4-60) 127 141
In the western Atlantic, green turtles typically have two feeding bouts each day
— one in the morning and one in the afternoon.'*^’'^^’*^'^However, in an area of reduced
food availability, Williams*"** found that green turtles fed continually over seagrass
beds during the nine h of daylight. In Hawaii, green turtles often feed at night; the
tidal cycle is more important in determining foraging times than is the diurnal
cycle. *^’^"*
The low nutritive value of jellyfish, as it pertains to the nutrition of leatherbacks,
has been repeatedly pointed oiit.**^ **^ The nutritive value of pyrosomas (gelatinous
tunicates) is apparently similar to that of jellyfish.**^ Large numbers of these gelat
inous organisms would have to be consumed to meet nutritional needs of a 900-kg
organism that is capable of maintaining body temperatures well above ambient. **^^’*"*'^
Duron*"*^’ estimated that an adult leatherback would eat about 50 large R. pulmo each
day (equivalent to approximately 200 1). A leatherback was seen to ingest 50 to 80
Stomolophus meleagris during a feeding session of unstated length off Myrtle Beach,
South Carolina.***"* The commensals that are ingested with the gelatinous organisms
or prey contained in their stomachs may be an important source of nutrition.**^
Six olive ridley hatchlings fed ad libitum on clam (Meretrix casta) tissue had
daily mean intake of 16.4 mg (SD 1.6 mg) dry mass per gram live body mass.*"*^
This dry mass intake was equivalent to a daily intake of 74.5 calories and 8.9 mg
of protein per gram live body mass.
8.5.2 D igestion
The extent to which food is digested — or digestibility — has been measured for
a number of diet components in wild green turtles feeding on natural diets of T.
testudinum‘^'‘^^'^'^'^ and the sponge Chondrilla nucula^^^ and for artificial diets in
captivity*"*"*’*^**’*'** (Table 8.2). The values for digestibilities from the study by
Bjorndal"*^ in Table 8.2 represent means of values measured in 12 consecutive months
in the southern Bahamas. There was no seasonal effect on nutrient composition of
Thalassia (Table 8.3) or on digestibility of Thalassia)'^ Among the four size classes
of green turtles, body size had a significant effect on digestion of organic matter,
energy, hemicellulose, and nitrogen, but not on digestion of cellulose.
TABLE 8.2
Digestibilities Measured in Green Turtles on Several Diets: Seagrass
(Thalassia testudinum), the Sponge (Chondnlla nucula), and Pelleted
Diets
Digestibility Turtle size No. of
Constituent Diet (%) (*<g) turtles Ref.
Dry matter Sponge 49 8 3 149

53 48 3 149
51 66 3 149
Pellet A 83-84 4-6 15 144
84-86 22-25 13 144
Organic matter Seagrass 45 8 3 42
58 30 3 42
67 48 3 42
65 66 3 42
77 50 1 148
65 82 1 148
Sponge 41 8 3 149
46 48 3 149
44 66 3 149
Energy Seagrass 34 8 3 42
50 30 3 42
62 48 3 42
58 66 3 42
69 50 1 148
64 82 1 148
Sponge 40 8 3 149
43 48 3 149
43 66 3 149
Pellet B 75 0.5-0.7 2 150
Pellet C 73 0.6-0.9 4 150
Carbon Seagrass 75 50 1 4
63 82 1 4
Cellulose Seagrass 85 8 3 42
85 30 3 42
89 48 3 42
86 66 3 42
94 50 1 148
77 82 1 148
Hemicellulose Seagrass 53 8 3 42
62 30 3 42
70 48 3 42
75 66 3 42
94 50 1 148
78 82 1 148
Total nitrogen Seagrass 15 8 3 42

Digestibilities Measured in Green Turtles on Several Diets: Seagrass
(Thalassia testudlnum)^ the Sponge (Chondnlla nucula)^ and Pelleted
Diets
Digestibility Turtle size No. of
Constituent Diet (%) H) turtles Ref.
39 30 3 42
45 48 3 42
54 66 3 42
Sponge 52 8 3 149
55 48 3 149
53 66 3 149
Pellet A 82-88 4-6 15 144
86-89 22-25 13 144
Pellet B 84 0.5-0.7 2 150
Pellet C 82 0.6-0.9 4 150
Organic nitrogen Seagrass 44 50 1 4
25 82 1 4
Experimental turtles in BjorndaE^ were the same as those in Bjomdal;^'^^^ turtles in BjomdaT^^
were the same as those in Thayer et alE
TABLE 8.3
Composition of Diets for which Digestibilities Are Presented in Table 8.2
Diet OM Energy Cellulose HC Lignin Nitrogen Si Ref.
Seagrass 74 14 33 7 3.4 2.7 — 42

Sponge 68 16 0 0 0 8.1 1.8 149
Pellet A — _ _ — 5.6-7.8 _ _ 144
Pellet B _ _ — _ _ — 7.0 — 150
Pellet C — — — — — 8.2 — 150
All constituents are expressed as percent of dry matter except energy is kJ/g dry matter. OM is organic
matter; HC is hemicellulose; Si is silica.
The low nitrogen digestibility values for green turtles feeding on Thalassia may
result from a shift of protein digestion from the small intestine to the large intestine,
where absorption of end products is much less efficient.^ The extent to which cell
contents (including protein) are accessible to digestive enzymes in the small intestine
— before cell walls have been degraded — is not clear. The low nitrogen
digestibilities may be responsible for the low growth rates of sea turtles feeding on
Thalassia?
Herbivorous reptiles rely on a microbial fermentation in the large intestine to

degrade plant cell w a lls ;th e only exception identified to date is the presence of
a femnentation in the small intestine of the freshwater turtle Pseudemys nelsoni}^"^
These fermentations generate end products that are critical to the nutrient balance
of the host reptiles. The primary end products are volatile fatty acids (VFA), an
important energy source in herbivorous re p tile s ,th a t supplied up to 100% of the
digestible energy intake in P, nelsoni on a foliage diet.^^"^
Green turtles harbor a microbial feiTnentation in their capacious large intes
tine. The cell walls (cellulose, hemicellulose) of Thalassia are digested to a
great extent in this fermentation (Table 8.2). The VFA and lactate produced in the
cecum of a green turtle feeding on Thalassia provided about 15% of its estimated
daily energy budget. However, the total energy contribution of the fermentation
is much greater than 15% because the microbial fermentation continues throughout
the large intestine, which has a much greater volume than the cecum.
The site of fermentation and relative concentrations of VFA along the gut were
similar between green turtles feeding on Thalassia and those feeding on an algae
diet, but the relative amounts of individual VFAs in the region of active fermentation
were different. In green turtles feeding on an algae diet, the relative proportions
were acetate > propionate > butyrate > valerate; for Thalassia-ÎQQàmg green turtles,
the proportions were acetate > butyrate > propionate. This difference may result
from differences in substrate composition and/or differences in species composition
of the bacterial or protozoal communities.
Although sponges are a primary food source for hawksbills and a minor food
source for loggerheads and green turtles, the only measure of digestibility of sponges
has been for the chicken-liver sponge, C. nucula, in the green t u r t l e . C. nucula,
perhaps because of its low concentration of silica s p i c u l e s , i s one of the primary
species ingested by hawksbills^^ and is also consumed by loggerheads"^"^ and green
tu rtle s.In the study, green turtles were feeding primarily on the seagrass T testudi-
num, but also ingesting sponges. The digestibilities of C. nucula (Table 8.2) were
low relative to values in reptiles feeding on animal tissue which typically exceed
80%.^^^ These low digestibility values may result from the high concentration of
collagen fibrils in C. nucula, which have an unknown, but possibly low, digestibil
ity.^ The low digestibilities may also be a result of a negative associative effect
between the ingested Thalassia and the sponges. Associative effects are any inter
action among diet items that cause a diet component to be digested to a different
extent than it would have been if the component had been ingested alone. A positive
associative effect between duckweed and insect larvae in the turtle Trachemys scripta
has been described. The much greater bulk of Thalassia in the digestive tract may
have blocked sponges from contact with proteolytic enzymes. To test for an asso
ciative effect, digestibilities for the sponge would have to be measured in green
turtles feeding only on C. nucula and compared with digestibilities for sponges in
the mixed diet.
Green turtles attain much higher digestibilities when feeding on pelleted diets
in captivity (Table 8.2). These higher digestibilities result from three primary factors.
First, the pelleted diets have higher concentrations of very digestible nutrients (nitro
gen, soluble carbohydrates) and lower concentrations of less digestible constituents
(structural carbohydrates) than do seagrasses and algae. Second, because of these
shifts in nutrient concentrations, the pelleted diets are digested to a greater extent
by endogenous enzymes in the small intestine, whereas algae and seagrasses are
digested to a greater extent by microorganisms in the large intestine.^’^^^ Digestion
in the small intestine results in more efficient and complete nutrient absorption.
Third, particle size of the pelleted diets is very fine, compared with the large particle
size that is ingested by green turtles feeding on seagrasses and algae. Because turtles
cannot chew their food, particle size of ingested seagrass and algae are typically
several centimeters square. Particle size of digesta has a significant negative effect
on rates of fermentation and digestion in herbivorous reptiles and mammals.
8.5.3 Passage of D igesta

Passage rates of digesta through the digestive tract are affected by feeding rate^^^
and tem p eratu re.In captive-raised loggerheads of 1- to 2-kg body mass on a diet
of trout pellets, total gut clearance time decreased with increasing temperature
between 20 and 30°C (Q^q = 1*6). The effect was greater between 20 and 25°C than
between 25 and 30°C.^^^
Digesta apparently passes through the different sections of the gut at different
rates. In post-hatchling green turtles (30 to 150 g) fed a diet of trout pellets mixed
with barium, most of the transit time was spent in the esophagus, stomach, and small
intestine; in yearling green turtles (0.5 to 1 kg), the same diet was retained for the
longest period in the large in te s tin e .In yearlings, the digesta in the hindgut moved
to and fro, mixing digesta from different m e a ls .T h is mixing is probably the result
of antiperistalsis — peristaltic movements that originate near the coprodeum and
move along the large intestine toward the ileocolic valve. Antiperistalsis can have
significant ramifications for reptile nutrition by delaying movement of digesta in the
large intestine to allow for more complete fermentation and absorption and by
moving urinary nitrogen from the cloaca to the proximal hindgut where it provides
an important source of nitrogen. Such mixing of digesta also challenges current
models of digestive processing in reptiles (e.g., Meienberger et al.)^^^ that assume
that movement of digesta in reptiles is similar to that in a plug-flow reactor.
8.6 ANTHROPOGENIC EFFECTS ON FORAGING

ECOLOGY AND NUTRITION
8.6.1 Marine D ebris
Balazs'^^’ provided a thorough review of accounts of marine debris ingestion by sea
turtles and documented the seriousness of this problem. Carr^^^ then focused attention
on the threat to young pelagic-stage sea turtles that inhabit convergence zones in
which floating debris is also concentrated. Since these publications appeared, a
number of studies have reported high incidence of debris ingestion in sea turtles
(references in Bjomdal et al.)d^^ In particular, the concerns that were raised by

Carr^^^ about small, pelagic turtles have been confirmed. For example, in 50 post
hatchling loggerheads captured off the coast of Florida, plastics and synthetic fibers
were found in 32% of the turtles, tar was flushed from the stomachs of 35% of the
turtles, and over 50% of the turtles had tar on their jaws.^^
The amount of debris ingested is usually s m a l l b u t it is important to put
these apparently small percentages in perspective. In a study of 43 green turtles that
stranded dead in Florida, 24 turtles had ingested debris.Ingestion of debris was
clearly responsible for the death of at least two turtles, by obstructing the gut. In
mass and volume, the debris in the two turtles were 2.2 and 6.5 g and 3.0 and 4.5
ml, respectively. These values fall within the ranges of mass (0.01 to 7.0 g) and
volume (0.01 to 5.0 ml) of debris found in the other 22 turtles in which the debris
appeared to be passing harmlessly through the gut. Two conclusions can be drawn
from these values. First, small amounts of debris can kill a sea turtle, and, second,
the predictability of such mortality may well be low. A given piece of debris could
pass through the gut of a turtle many times without becoming lodged in the gut, but
during one transit, the debris could become oriented in such a way as to block the
gut and result in the death of the animal.
Studies of debris ingestion need to move beyond documenting occurrence and
begin to assess effects on populations — both mortality and sublethal effects.
Mortality resulting from ingestion of debris is extremely difficulty to estimate.
Sublethal effects are even more difficult to estimate, are probably more common,
and may well be more deleterious to sea turtle populations than direct mortality by
decreasing the productivity (growth and reproduction) of sea turtles.
Mechanisms for sublethal effects include absorption of toxins from the debris,
mechanical abrasion or blockage of absorptive surfaces in the digestive tract, and
displacement of nutritious food with debris so that nutrient gain is decreased (=
nutrient dilution). Absoiption of petroleum hydrocarbons was evaluated in sea turtles
found dead following a major oil spill in T e x a s.T h e potential problems of absorp
tion of plasticizers in sea turtles have been discussed. Small pieces of latex and
plastic sheeting can be retained in the digestive tract of sea turtles for up to four
months, and the latex appeared to have deteriorated during that time.‘^‘ Effects on
gut function as a result of plastic and latex ingestion include decline in blood glucose
levels, indicating possible interference with absorption of nutrients or metabolism,
interference in gut lipid metabolism, and gas accumulation in the large intestine,
resulting in loss of buoyancy control. Effect of nutrient dilution on intake is being
evaluated in loggerhead hatchlings.
8.6.2 C ommercial F isheries

Commercial fisheries have been recognized for many years as a serious source of
sea turtle mortality as a result of incidental capture. Degradation of foraging habitat
as a result of fishing activities has received less attention. Destructive fishing prac
tices include the use of dynamite or bleach in coral reef areas and the use of bottom
trawls in benthic communities. Most of the trawlable shelves of the world are
impacted by trawling, which results in massive changes in community structure.
Habitat degradation resulting from these practices decreases the quantity and quality
of foraging habitats for sea turtles.
A concern that has not yet been addressed is that of direct competition for food
between humans and sea turtles. Recent estimates^^^ indicate that commercial fish
eries in continental shelf systems have a high primary productivity requirement of
24 to 35%, which may not be sustainable and may threaten the biodiversity of these
systems. In addition, a higher primary productivity requirement would starve top
predators such as marine mammals and b i r d s . A s commercial fisheries become
increasingly efficient, and as more commercial species of fish and invertebrates
approach commercial — if not biological — extinction, sea turtles that feed on these
commercial species may suffer reduced feeding rates because of limited food
resources. Those sea turtle species that do not feed on commercial species may find
that availability of their dietary species have been negatively affected through
bycatch in commercial fisheries or as a result of the disruption in marine food webs
caused by the dramatic population/species shifts and habitat degradation resulting
from commercial fisheries. The low primary production requirement of ectothermic
sea turtles relative to endotherms may provide some protection from competition
with humans.
The sublethal effects of food limitation resulting from such competition will be
difficult to discern, but may reduce significantly the productivity of sea turtle pop
ulations by lowering growth rates, delaying onset of sexual maturity, and reducing
reproductive output. To monitor these sublethal effects, baseline data on rates of
growth and intakes on natural diets are needed for many populations of each sea
turtle species. These populations should represent a broad geographic range and all
habitat types within each species. Only with this information can the amount of food
resources necessary to sustain populations of sea turtles be determined.
If a population limited by food availability was released from that food limitation,
it has been suggested that compensatory gain may act to increase productivity and
hence population recovery.Compensatory gain is realized when an organism that
has had poor nutrition — often as a result of limited intake or poor quality food —
is provided better nutrition. For a period of time, the previously undernourished
individual will have a lower feed-to-gain ratio than an individual that has been
maintained on the better diet throughout its life. Compensatory gain is well known
in mammals, but has not been studied in reptiles.
8.6,3 D egradation of Foraging H abitats

In addition to degradation of foraging habitats by commercial fisheries, other human
activities degrade sea turtle foraging habitats. Scars in seagrass beds from anchors
or propellers can seriously reduce the standing crop and productivity of seagrasses.
Anchoring by 15 to 50 boats each night in two bays in St. John, U.S. Virgin Islands
caused the loss of 6.5 m^ of seagrass (Thalassia) pastures each day.^"^^ Recovery of
the seagrasses within the anchor scars was slow; after 7 months there was little
regrowth of seagrasses in the scars.
Deposition of silt on coral reefs, rocky bottom habitats, and seagrass beds —
most commonly as a result of inappropriate land management practices — decreases
the amount of foraging habitat available to sea turtles. Oil spills are a constant threat
to sea turtle foraging habitats. The Kemp’s ridley is particularly vulnerable because
the two major feeding areas of adult ridleys are in areas of intense development for
offshore oil production in the Gulf of Mexico.
In a different twist, construction of offshore oil platforms, particularly in the
Gulf of Mexico, creates hard-bottom communities that are favored by foraging
loggerheads. These structures — which currently number approximately 3800 and
provide the equivalent of about 5000 km^ of hard bottom, increasing the amount of
that habitat in the Gulf of Mexico by 27% — may remain in place for decades and
acquire a complex community of invertebrates.W hen the oil platforms are
removed (currently at a rate of about 100 per year), as required by law, most are
removed with explosives that destroy the habitat and can kill turtles in the area.^^^’^^^
Much research is needed to elucidate the relationships between the foraging ecology
of sea turtles and their role in marine ecosystems and between the nutrition of sea
turtles and their productivity. Surprising gaps remain in our knowledge of feeding
habits. Some of these gaps can be generalized to ocean basins, species, and life
stages. There is a paucity of data for all species from the Indian Ocean, for the early
life stages of all species, and, among species, the diet of the flatback and the olive
ridley are the least known.
The study of nutrition in sea turtles is in its infancy. Studies are needed on all
aspects of intake, digestion, and passage of digesta and on quantitative nutritional
requirements. To the extent possible, these studies should involve natural diets and
turtles that have not been raised in captivity.
Although understanding diet selection is critical for assessing habitat quality —
an undertaking that is ever more critical as wildlife managers are expected to make
difficult decisions on which habitats to protect and which to abandon — such studies
are still rare. Quantitative studies are needed that address diet selectivity in sea turtles
by relating relative abundance of potential food items in the environment to the food
items that are ingested.
We have failed to evaluate adequately the role of sea turtles in their ecosystems.
Such studies are essential not only to improve our understanding of how marine
ecosystems function, but also to assess what is lost at the ecosystem level when sea
turtle populations become severely reduced or extinct.
Finally, the negative effects of human activities on the foraging habitats of sea
turtles and their diet quality and availability must be quantified and controlled. Such
quantification will be difficult because many of the negative effects are sublethal or
are indirect results of changes in the delicate balance of marine ecosystems, and
thus difficult to trace.
8.8 A CKN O W LED G M EN TS

I am grateful to Alan Bolten for sharing his insights that have contributed to the
ideas expressed in this chapter and for his constructive comments on the manuscript.
I thank Peter Eliazar for his help with the references in this chapter. The on-line sea
turtle bibliography maintained by the Archie Carr Center for Sea Turtle Research,
University of Florida, facilitated the literature review for this chapter.
R EFEREN CES
Balazs, G. H., Growth rates of immature green turtles in the Hawaiian Archipelago,
in Biology and Conservation of Sea Turtles, Bjomdal, K. A., Ed., Smithsonian
2. Bjomdal, K. A., The consequences of herbivory for the life history pattern of the
Caribbean green turtle, in Biology and Conservation o f Sea Turtles, Bjomdal, K. A.,
Ed., Smithsonian Institution Press, Washington, D.C., 1982, 111.
Bjomdal, K. A., Nutritional ecology of sea turtles, Copeia, 1985, 736, 1985.
Thayer, G. W., Engel, D. W., and Bjomdal, K. A., Evidence for short-circuiting of
the detritus cycle of seagrass beds by the green turtle, Chelonia mydas L., J. Exp.
Mar. B iol EcoL, 62, 173, 1982.
5. Thayer, G. W., Bjomdal, K. A., Ogden, J. C., Williams, S. L., and Zieman, J. C.,
Role of larger herbivores in seagrass communities, 7, 351, 1984.
6. Limpus, C. J. and Nicholls, N., The southern oscillation regulates the annual numbers
of green turtles {Chelonia mydas) breeding around northern Australia, Aust. J. Wildl.
Res., 15, 157, 1988.
7. Limpus, C. J. and Nicholls, N., Progress report on the study of the interaction of the
El Niño Southern Oscillation on annual Chelonia mydas numbers at the southern
Great Barrier Reef rookeries, in Proceedings of the Australian Marine Turtle Con
servation Workshop held at Sea World Nara Resort, Gold Coast, Queensland Depart
ment of Environment and Heritage, and Australian Nature Conservation Agency,
Queensland, Australia, 1994, 73.
Lanyon, J. M., Limpus, C. J., and Marsh, H., Dugongs and turtles; grazers in the
seagrass system, in Biology of Seagrasses, Larkum, A. W. D., McComb, A. J., and
Shepherd, S. A., Eds., Elsevier, Amsterdam, 1989, 610.
Mortimer, J. A., Feeding ecology of sea turtles, in Biology and Conservation o f Sea
Turtles, Bjomdal, K. A., Ed., Smithsonian Institution Press, Washington, D.C., 1982,
103.
10, Balazs, G. H., Field methods for sampling the dietary components of green turtles
Chelonia mydas, HerpetoL Rev., 11, 5, 1980,
11 . Forbes, G. A. and Limpus, C. J., A non-lethal method for retrieving stomach contents
from sea turtles, Wildl. Res., 20, 339, 1993,
12. Carr, A., New perspectives on the pelagic stage of sea turtle development, Conserv.
B iol, 1, 103, 1987.
13. Walker, T. A., Post-hatchling dispersal of sea turtles, in Proceedings of the Australian
Marine Turtle Conservation Workshop held at Sea World Nara Resort, Gold Coast,
Queensland Department of Environment and Heritage, and Australian Nature Con
servation Agency, Queensland, Australia, 1994, 79.
14. Frick, J., Orientation and behaviour of hatchling green turtles (phelonia mydas) in
the sea, Anim. Behav., 24, 849, 1976.
15. Hughes, G. R., The sea turtles of south-east Africa. II. The biology of the Tongaland
loggerhead turtle Caretta caretta L. with comments on the leatherback turtle Dermo-
chelys coriacea L. and the green turtle Chelonia mydas L. in the study region,
Oceanogr. Res. Institute (Durban) Invest. Rep., No. 36, 1974.
16. Bolten, A. B., and Bjorndal, K. A., Blood profiles for a wild population of green
turtles (Chelonia mydas) in the southern Bahamas: size-specific and sex-specific
relationships, J. Wildl. Dis., 28, 407, 1992.
17. Nakamura, K., Carotenoids in serum of Pacific green turtle, Chelonia mydas. Bull.
Jp. Soc. Sci. Fish., 46, 909, 1980.
18. Bjorndal, K. A., and Bolten, A. B., Growth rates of immature green imilQS, Chelonia
mydas, on feeding grounds in the southern Bahamas, Copeia, 1988, 555, 1988.
19. Balazs, G. H., Synopsis of biological data on the green turtle in the Hawaiian Islands,
NOAA Tech. Memo. NOAA-TM-NMFS-SWFC-7, Honolulu, Hawaii, 1980.
20. Limpus, C. J., Couper, P. J., and Read, M. A., The green turtle, Chelonia mydas, in
Queensland: population structure in a warm temperate feeding area, Mem. Queens
land Mus., 35, 139, 1994.
21. Mortimer, J. A., The feeding ecology of the West Caribbean green turtle (Chelonia
mydas) in Nicaragua, Biotropica, 13, 49, 1981.
22. Agastheesapillai, A. and Thiagarajan, R., Biology of the green iwxiXt Chelonia mydas
(Linnaeus) in the Gulf of Mannar and Palk Bay,/. Mar. Biol. Assoc. India, 21, 45,
1979.
23. Hirth, H. F. and Carr, A., The green turtle in the Gulf of Aden and the Seychelles
Islands, Verb. K. Ned. Akad. Wet. Afd. Natuurk. Reeks, 58, 1, 1970.
24. Hirth, H. F., Klikoff, L. G., and Harper, K. T., Sea grasses at Khor Umaira, People’s
Democratic Republic of Yemen with reference to their role in the diet of the green
turtle, Chelonia mydas. Fish. Bull., 71, 1093, 1973.
25. Ross, J. R, Biology of the green turtle, Chelonia mydas, on an Arabian feeding ground,
J. Herpetol, 19, 459, 1985.
26. Frazier, J., Marine turtles in the Seychelles and adjacent territories, in Bio geography
and Ecology o f the Seychelles Islands, Stoddart, D. R., Ed., Dr. W. Junk Publishers,
The Hague, Netherlands, 1984, 417.
27. Frazier, J., Marine Turtles in the Comoro Archipelago, Verhandelingen der Konin-
klijke Nederlandse Akademie van Wetenschappen, Afd. Natuurkunde, Tweede Reeks,
Deel 84, North-Holland, Amsterdam, 1985.
28. Garnett, S. T., Rice, I. R., and Scott, F. J., The diet of the green turtle, Chelonia
mydas (L.), in Torres Sirmi, Aust. Wildl. Res., 12, 103, 1985.
29. Limpus, C. J. and Reed, P. C., Green sea turtles stranded by Cyclone Kathy on the
south-western coast of the Gulf of Carpentaria, Wildl. Res., 12, 523, 1985.
30. Forbes, G. A., The diet of the green turtle in an algal-based coral reef community -
Heron Island, Australia, in Proc. 13th Annu. Symp. Sea Turtle Biology and Conser
vation, Schroeder, B. A. and Witherington, B. E., Compilers, NOAA Tech. Memo.
NMFS-SEFSC-341, Miami, FL, 1994, 57.
31. Kurata, Y, Yoneyama, S., Tsutsumi, K., Kimura, J., and Hosokawa, S., Experiments
to increase number of green turtles through the release of the young (VII. Aoumigame
zoshoku horyu shiken), Ogasawara Fishery Center, Tokyo Metropolitan Government,
Research Report 3:58, 1978.
32. Balazs, G. H., Sea turtles and their traditional usage in Tokelau,/Ire// Res. Bull., No.
279, 1, 1983.
33. Balazs, G. H., Status and ecology of marine turtles at Johnson Atoll, Are// Res. Bull.,
No. 285, 1, 1985.
34. Balazs, G. H., Forsyth, R. G., and Kam, A. K. H., Preliminary assessment of habitat
utilization by Hawaiian green turtles in their resident foraging pastures, NOAA Tech.
Memo. NOAA-TM-NMFS-SWFC-71, Honolulu, HI, 1987.
35. Balazs, G. H., Miya, R. K., and Finn, M. A., Aspects of green turtles in their feeding,
resting, and cleaning areas off Waikiki Beach, mProc. 13th Annu. Symp. Sea Turtle
Biology and Conservation, Schroeder, B. A. and Witherington, B. E., Compilers,
NOAA Tech. Memo. NMFS-SEFSC-341, Miami, FL, 1994, 15.
36. Russell, D. J. and Balazs, G. H., Colonization by the alien marine alga Hypnea
musciformis (Wulfen) J. Ag. (Rhodophyta: Gigartinales) in the Hawaiian Islands and
its utilization by the green turtle, Chelonia mydas L., Aquat. Bot., 47, 53, 1994.
37. Russell, D. J. and Balazs, G. H., Utilization of alien algal species by sea turtles in
Hawaii, in Proc. Conference and Workshop on Nonindigenous Estuarine & Marine
Organisms (NEMO), U.S. Department of Commerce, NOAA, Washington, D.C.,
1994, 93.
38. Paredes, R. R, Introducción al Estudio Biológico de Chelonia mydas agassizi en el
Perfil de Pisco, Masters thesis. Universidad Nacional Federico Villareal, Lima, Peru,
1969.
39. Hays Brown, C. and Brown, W. M., Status of sea turtles in the southeastern Pacific:
emphasis on Peru, m Biology and Conservation of Sea Turtles, Bjorndal, K. A., Ed.,
Smithsonian Institution Press, Washington, D.C., 1982, 235.
40. Fritts, T. H., Pelagic feeding habits of turtles in the eastern Pacific,Mar. Turtle NewsL,
No. 17, 4, 1981.
41. Casas-Andreu, G. and Gómez-Aguirre, S., Contribución al conocimiento de los hábi
tos alimenticios de Lepidochelys olivácea y Chelonia mydas agassizi (Reptilia, Che-
loniidae) en el Pacífico Mexicano, Bolm Inst. Oceanogr. Sao Paulo, 29, 87, 1980.
42. Bjorndal, K. A., Nutrition and grazing behavior of the green turtle, Chelonia mydas.
Mar. Biol., 56, 147, 1980.
43. Mortimer, J. A., Observations on the Feeding Ecology of the Green Turtle, Chelonia
mydas, in the Western Caribbean, Thesis, University of Florida, Gainesville, FL, 1976.
44. Acevedo, M., Gómez, O., and Berovides, V., Alimentación de tres especies de que-
lonios marinos en la plataforma suroccidental de Cuba,/?^v. Invest. Mar., 5, 29, 1984.
45. Ferreira, M. M., Sobre a alimenta^áo da aruaná, Chelonia mydas Linnaeus, ao longo
da costa do Estado do Ceará, Ar^. Est. Biol. Mar. Univ. Fed. Ceará, 8, 83, 1968.
46. Sazima, I. and Sazima, M., Aspectos de comportamento alimentar e dieta da tartaruga
marinha, Chelonia mydas, no litoral norte paulista, Bolm Inst. Oceanogr. Sao Paulo,
32, 199, 1983.
47. Mendonga, M. T, Movements and feeding ecology of immature green turtles (phe-
lonia mydas) in a Florida lagoon, Copeia, 1983, 1013, 1983.
48. Burke, V. J., Morreale, S. J., Logan, R, and Standora, E. A., Diet of green turtles
(Chelonia mydas) in the waters of Long Island, N.Y., in Proc. 11th Annu. Workshop
on Sea Turtle Biology and Conservation, Salmon, M. and Wyneken, J., Compilers,
NOAA Tech. Memo. NMFS-SEFSC-302, Miami, Florida, 1991, 140.
49. Bolten, A. B. and Balazs, G. H., Biology of the early pelagic stage — the ‘Tost year”,
in Biology and Conservation o f Sea Turtles, Rev. ed., Bjomdal, K. A., Ed., Smithso
nian Institution Press, Washington, D.C., 1995, 579.
50. Carr, A., Rips, FADS, and little loggerheads, B/oSc/cncé', 36, 92, 1986.
51. Limpus, C. J., Fleay, A., and Guinea, M., Sea turtles in the Capricornia Section, Great
Barrier Reef, in The Capricornia Section o f the Great Barrier Reef: Past, Present
and Future, Ward, W. T. and Saenger, R, Eds., Royal Society of Queensland and
Australian Coral Reef Society, Brisbane, 1984, 61.
52. Fanning, K. A., Betzer, P. R., Byrne, R. H., Breland, J. A., and Jolley, R. R., Char
acteristic of a submarine geothermal spring on the West Florida Shelf, FI. Sci., 42
(Suppl.), 21, 1979.
53. Carr, A. and Meylan, A. B., Evidence of passive migration of green turtle hatchlings
in sargassum, Copeia, 1980, 366, 1980.
54. Richardson, J. I. and McGillivary, R, Post-hatchling loggerhead turtles eat insects in
Sargassum community, Mar. Turtle Newsl., No. 55, 2, 1991.
55. Witherington, B. E., Flotsam, jetsam, post-hatchling loggerheads, and the advecting
surface smorgasbord, in Froc. 14th Annu. Symp. Sea Turtle Biology and Conservation,
Bjorndal, K. A., Bolten, A. B., Johnson, D. A., and Eliazar, P J., Compilers, NOAA
Tech. Memo. NMFS-SEFSC-351, Miami, Florida, 1994, 166.
56. Brongersma, L. D., European Atlantic turtles. Zoo/. Verb. (Leiden), No. 121, 1972.
57. Van Nierop, M. M. and den Hartog, J. C., A study on the gut contents of five juvenile
loggerhead turtles, Caretta caretta (Linnaeus) (Reptilia, Cheloniidae), from the south
eastern part of the North Atlantic Ocean, with emphasis on coelenterate identification,
Zool. Meded. Leiden, 59, 35, 1984.
58. Plotkin, P, Feeding ecology of the loggerhead sea turtle in the northwestern Gulf of
Mexico, in Proc. 9th Annu. Workshop on Sea Turtle Conservation and Biology, Eckert,
S. A., Eckert, K. L., and Richardson, T. H., Compilers, NOAA Tech. Memo. NMFS-
SEFC-232, Miami, Florida, 1989, 139.
59. Dodd, C. K., Jr., Synopsis of the biological data on the loggerhead sea turtle Careiia
caretta (Linnaeus 1758), U.S. Fish Wildl. Serv. Biol. Rep., 88(14), 1988.
60. Lutcavage, M. and Musick, J. A., Aspects of the biology of sea turtles in Virginia,
Copeia, 1985, 449, 1985.
61. Limpus, C. J., Couper, P. J., and Read, M. A., The loggerhead turtle, Cai'etta caretta,
in Queensland: population structure in a warm temperate feeding area,Mem. Queensl.
Mus., 37, 195, 1994.
62. Burke, V. J., Standora, E. A., and Morreale, S. J., Diet of juvenile Kemp’s ridley and
loggerhead sea turtles from Long Island, New York, Copeia, 1993, 1176, 1993.
63. Burke, V. J., Morreale, S. J., and Rhodin, A. G. J., Lepidochelys kempii (Kemp’s
ridley sea turtle) and Caretta caretta (loggerhead sea turtle). Diet, Herpetol. Rev., 24,
31, 1993.
64. Plotkin, P. T., Wicksten, M. K., and Amos, A. F, Feeding ecology of the loggerhead
sea turtle Caretta caretta in the northwestern Gulf of Mexico, Mar. Biol., 115, 1, 1993.
65. Hornell, J., The turtle fisheries of the Seychelles Islands, His Majesty’s Stationery
Office, London, 1927.
66. Meylan, A. B., Feeding Ecology of the Hawksbill Turtle (Eretmochelys imbricata):
Spongivory as a Feeding Niche in the Coral Reef Community, Dissertation, University
of Florida, Gainesville, FL, 1984.
67. Meylan, A., Spongivory in hawksbill turtles: a diet of gluss. Science, 239, 393, 1988.
68 . Limpus, C. J,, The hawksbill turtle, Eretmochelys imbricata, in Queensland: popu
489, 1992.
69 . Pritchard, P. C. H. and Trebbau, P, The Turtles of Venezuela. Contributions to
Herpetology, No. 2, Society for the Study of Amphibians and Reptiles, Oxford, Ohio,
1984.
70 . National Marine Fisheries Service and U.S. Fish and Wildlife Service, Recovery plan
for hawksbill turtles in the U.S. Caribbean Sea, Atlantic Ocean, and Gulf of Mexico,
National Marine Fisheries Service, St. Petersburg, FL, 1993.
71. Anderes Alvarez, B. L. and Uchida, I., Study of hawksbill turtle {Eretmochelys
imbricata) stomach content in Cuban waters, in Study of the Hawksbill Turtle in
Cuba (I), Ministry of Fishing Industry, Cuba, 1994, 27.
72. Diez, C. E. and van Dam, R., Foraging ecology of juvenile and subadult hawksbills
{Eretmochelys imbricata), Preliminary Research Report, 1992.
73. Vicente, V. P. and Carballeira, N. M., Studies on the feeding ecology of the hawksbill
turtle, Eretmochelys imbricata, in Puerto Rico, in Proc. 11th Annu. Workshop on Sea
Turtle Biology and Conservation, Salmon, M. and Wyneken, J,, Compilers, NOAA
74. Vicente, V. R, Spongivory in Caribbean hawksbill turtles, Eretmochelys imbricata:
Data from stranded specimens, in Proc. 13th Annu. Symp. Sea Turtle Biology and
Conservation, Schroeder, B. A. and Witherington, B. E., Compilers, NOAA Tech.
75. Carr, A. and Stancyk, S., Observations on the ecology and survival outlook of the
hawksbill turtle, Bio/. Conserv., 8, 161, 1975.
76. Den Hartog, J. C., Notes on the food of sea turtles: Ef'etmochelys imbricata (Linnaeus)
and Dermochelys coriácea (Linnaeus), Neth. J. Zool., 30, 595, 1980.
77. Carr, A., Handbook o f Turtles, Cornell University Press, Ithaca, NY, 1952.
78. Deraniyagala, P. E. R, A Colored Atlas o f Some Vertebrates from Ceylon. Vol. II,
Tetrapod Reptilia, Ceylon National Museums Publication, The Ceylon Government
Press, Colombo, Ceylon, 1953.
79. Alcala, A. C., Observations on the ecology of the Pacific hawksbill turtle in the central
Visayas, Philippines, Eish. Res. J. Philipp., 5, 42, 1980.
80. Witzell, W. N., Synopsis of biological data on the hawksbill turtle Eretmochelys
imbricata (Linnaeus, 1766), FAO Fish. Synop., No. 137, 1983.
81. Ross, J. R, Flawksbiil turtle Eretmochelys imbricata in the Sultanate of Oman, Biol.
Conserv., 19, 99, 1981.
82. Balazs, G. H., A hawksbill in Kaneohe Bay, Oahu, ‘Elepaio, 38, 128, 1978.
83. Bjomdal, K. A., Carr, A., Meylan, A. B., and Mortimer, J. A., Reproductive biology
of the hawksbill Eretmochelys imbricata at Tortuguero, Costa Rica, with notes on
the ecology of the species in the Caribbean, B/o/. Conserv., 34, 353, 1985.
84. Ogren, L. H., Distribution of juvenile and subadult Kemp’s ridley turtles; Preliminary
results from the 1984-1987 surveys, in Proc. 1st hit. Symp. Kemp's Ridley Sea Turtle
Biology, Conservation and Management, Caillouet, C. W., Jr. and Landry, A. M., Jr.,
Eds., Sea Grant College Program, Galveston, TX, 1989, 116.
85. Márquez-M., R., Synopsis of biological data on the Kemp’s ridley turtle,Lepidochelys
kempi (Garman, 1880), NOAA Tech. Memo. NMFS-SEFSC-343, Miami, FL, 1994.
86 . Shaver, D. J., Feeding ecology of wild and head-started Kemp’s ridley sea turtles in
south Texas waters, J. Herpetol., 25, 327, 1991.
87. Burke, V. J., Morreale, S. J., and Standora, E. A., Diet of the Kemp’s ridley sea turtle,
Lepidochelys kempii, in New York waters. Fish. Bull., 92, 26, 1994.
88. Hughes, G. R., The sea turtles of south-east Africa. I. Status, morphology and dis
tributions, Oceanogr. Res. Inst. Invest. Rep., No. 35, 1974.
89 . Pitman, R. L., Seabird associations with marine turtles in the eastern Pacific Ocean,
Colon. Waterbirds. 16. 194. 1993.
90. Plotkin, P. T., Bytes, R. A., and Owens, D. W., Migratory and reproductive behavior
of Lepidochelys olivácea in the eastern Pacific Ocean, inProc. 13th Annu. Symp. Sea
Turtle Biology and Conservation, Schroeder, B. A. and Witherington, B. E., Compil
ers, NOAA Tech. Memo. NMFS-SEFSC-341, Miami, FL, 1994, 138.
9 1. Márquez-M., R., Villanueva O., A., and Peñaflores S., C., Sinopsis de datos biológicos
sobre la tortuga golfina, Lepidochelys olivácea (Eschscholtz, 1829), INP Sinopsis
Sobre la Pesca No. 2, 1976.
92. Montenegro Silva, B. C., Bernal Gonzalez, N. G., and Martínez Guerrero, A., Estudio
del contenido estomacal de la tortuga marina Lepidochelys olivácea, en la costa de
Oaxaca, México, An. Inst. Cieñe. Mar Limnol. Univ. Nal. Antón. México, 13, 121,
1986.
93. Fritts, T. H., Marine turtles of the Galápagos Islands and adjacent areas of the eastern
Pacific on the basis of observations made by J. R. Slevin 1905-1906,/. HerpetoL,
15, 293, 1981.
94. Limpus, C. J. and Roper, P. A., On the Victorian coast — a Pacific ridley sea-turtle
Lepidochelys olivaceae (Eschscholtz), Victorian Nat., 94, 54, 1977.
95. Biswas, S., A report on the olive ridley, Lepidochelys olivácea (Eschscholtz)
[TestudinesrCheloniidae] of Bay of Bengal, Rcc. Zool. Surv. India, 19, 275, 1982.
96. Caldwell, D. K., Rathjen, W. F, and Hsu, B. C. C., Surinam ridleys at sea,7/î/. Turtle
Tortoise S o c . 3, 4, 1969.
the flatback turtle, Natator depressus (Garman), /. Biogeogr., 17, 275, 1990.
98. Walker, T. A„ Juvenile flatback turtles in proximity to coastal nesting islands in the
Great Barrier Reef Province, J. HerpetoL, 25, 246, 1991.
99. Limpus, C. J., Parmenter, C. J., Baker, V., and Fleay, A., The flatback inrÚQ,Chelonia
Aust. Wildl. Res., 10, 557, 1983.
100. Zangerl, R., Hendrickson, L. P., and Hendrickson, J. R., A redescription of the
Australian flatback sea turtle, Natator depressus. Bishop Museum Bulletin in Zoology
I, Bishop Museum Press, Honolulu, 1988.
101. Limpus, C. J., Gyuris, E., and Miller, J. D., Reassessment of the taxonomic status of
the sea turtle genus Natator McCulloch, 1908, with a redescription of the genus and
species, Trans. R. Soc. South Aust., 112, 1, 1988.
102. Lee, D. S. and Palmer, W. M., Records of leatherback i\xn\e^s,Dermochelys coriácea
(Linnaeus), and other marine turtles in North Carolina waters, Brimleyana, 5, 95,
1981.
103. Hoffman, W. and Fritts, T. H., Sea turtle distribution along the boundary of the Gulf
Stream Current off eastern Florida, Herpetologica, 38, 405, 1982.
104. Grant, G. S. and Ferrell, D., Leatherback turtle, Dermochelys coriácea (Reptilia:
Dermochelidae): notes on near-shore feeding behavior and association with cobia,
Brimleyana, 19, 77, 1993.
105. Leary, T. R., A schooling of leatherback ixxxile^s, Dermochelys coriácea coriácea, on
the Texas coast, Copeia, 1957, 232, 1957.
106. Lazell, J. D., Jr., New England waters: critical habitat for marine turtles, Copeia,
1980, 290, 1980.
107. Duron, M., Quero, J. C., and Duron, P, Présence dans les eaux côtières de France et
de Guyane fréquentées par Dermochelys coriácea L., de Remora remora L., et de
Rhizostoma pulmo L., Annal. Soc. Sci. Nat. Charente-Mar., 1, 147, 1983.
108. Collard, S. B., Leatherback turtles feeding near a watermass boundary in the eastern
Gulf of Mexico. Mar. Turtle Newsl., No. 50. 12. 1990.
109. Duron, M. and Duron, R, Des tortues luths dans le pertuis Charentais, Cow/r. Nat.,
69, 37, 1980.
no. Eisenberg, J. F. and Frazier, J., A leatherback turtle {Dermochelys coriacea) feeding
in the wild, J. Herpetol, 17, 81, 1983.
111. Limpus, C. J., A benthic feeding record from neritic waters for the leathery turtle
{Dermochelys coriacea), Copeia, 1984, 552, 1984.
112. Eckert, S. A., Eckert, K. L., Ponganis, P., and Kooyman, G. L., Diving and foraging
behavior of leatherback sea turtles {Dermochelys coriacea). Can. J. ZooL, 67, 2834,
1989.
113. Brongersma, L. D., Miscellaneous notes on turtles. III. Proc. K. Ned. Akad. Wet-
Amsterdam, Ser. C, 73, 323, 1970.
114. Bleakney, J. S., Reports of marine turtles from New England and eastern Canada,
Can. Field Nat., 79, 120, 1965.
115. Brongersma, L. D., Miscellaneous notes on turtles. IIA and llB.Proc. K. Ned. Akad.
Wet.-Amsterdam, Ser. C, 72, 76, 1969.
116. Den Hartog, J. C. and van Nierop, M. M., A study on the gut contents of six leathery
turtles Dermochelys coriacea (Linnaeus) (Reptilia:Testudines:Dermochelyidae) from
British waters and from the Netherlands, Zoo/. Verh. (Leiden), No. 209, 1984.
117. Frazier, J., Meneghel, M. D., and Achaval, E, A clarification on the feeding habits
of Dermochelys coriacea, J. Herpetol., 19, 159, 1985.
118. Limpus, C. J. and McLachlan, N., The conservation status of the leatherback turtle,
Dermochelys coriacea, in Australia, in Proceedings of the Australian Marine Turtle
Conseiwation Workshop held at Sea World Nara Resort, Gold Coast, Queensland
Department of Environment and Heritage, and Australian Nature Conservation
Agency, Queensland, Australia, 1994, 68.
119. Davenport, J. and Balazs, G. H., “Fiery bodies” — are pyrosomas an important
component of the diet of leatherback turtles?, Br. Herpetol. Soc. Bull., 37, 33, 1991.
120. Grassman, M. A. and Owens, D. W., Development and extinction of food preferences
in the loggerhead sea turtle, Caretta caretta, Copeia, 1982, 965, 1982.
121. Steele, C. W., Grassman, M. A., Owens, D. W., and Matis, J. H., Application of
decision theory in understanding food choice behavior of hatchling loggerhead sea
turtles and chemosensory imprinting in juvenile loggerhead sea turtles, Experientia,
45, 202, 1989.
122. Balazs, G. H., Ecological aspects of green turtles at Necker Island, Report Hawaii
Institute of Marine Biology, Kaneohe, HI, 1977.
123. Hungate, R. E., The Rumen and Its Microbes, Academic Press, New York, 1966.
124. Greenwood, Y., Hall, F. J., Orpin, C. G., and Paterson, I. W., Microbiology of seaweed
digestion in Orkney sheep,/. Physiol. London, 343, 121, 1983.
125. Greenwood, Y, Orpin, C. G., and Paterson, I. W., Digestibility of seaweeds in Orkney
sheep,/. Physiol. London, 343, 120, 1983.
126. Orpin, C. G., Greenwood, Y , Hall, F. J., and Paterson, I. W , The rumen microbiology
of seaweed digestion in Orkney sheep,/. Appl. BacterioL, 59, 585, 1985.
127. Morreale, S. J. and Standora, E. A., Habitat use and feeding activity of juvenile
Kemp’s ridleys in inshore waters of the northeastern U.S., in Proc. 11th Annu.
Workshop on Sea Turtle Biology and Conservation, Salmon, M. and Wyneken, J.,
Compilers, NOAA Tech. Memo. NMFS-SEFSC-302, Miami, FL, 1991, 75.
128. Keinath, J. A., Musick, J. A., and Byles, R. A., Aspects of the biology of Virginia’s
sea turtles: 1979-1986. F /. Sci.. 38. 329. 1987.
129. McRoy, C. P. and McMillan, C., Production ecology and physiology of seagrasses,
in Seagrass Ecosystems: a Scientific Perspective, McRoy, C. P. and Helfferich, C.,
Eds., Marcel Dekker, New York, 1977, 53.
130. Ogden, J. C., Robinson, L., Whitlock, K., Daganhardt, H., and Cebula, R., Diel
foraging patterns in juvenile green turtles (Chelonia mydas L.) in St. Croix, United
States Virgin Island, J. Exp. Mar. Biol. Ecol., 66, 199, 1983.
131. Greenway, M. The effects of cropping on the growth of Thalassia testudinum (König)
in fdiVñdiicdi, Aquaculture, 4, 199, 1974.
132. Tighe, S. A., Grazing of Seagrasses by the Green Turtle {Chelonia mydas) in St.
Croix, U.S.V.I., Thesis, Fairleigh Dickinson University, Rutherford, NJ, 1981.
133. Zieman, J. C., Iverson, R. L., and Ogden, J. C., Herbivory effects on Thalassia
testudinum leaf growth and nitrogen content, Mar. Ecol. Progr. Ser., 15, 151, 1984.
134. Dredge, M, C. L., Growth and mortality in an isolated bed of saucer scallops,
japonicum balloti (Bernardi), Queensl. J. Agrie. Anim. Sei., 42, 11, 1985.
135. Iversen, E. S., Jory, D. E., and Bannerot, S. R, Predation on queen conchs, Strombus
gigas, in the Bahamas, Bull. Mar. Sei., 39, 61, 1986.
136. Hughes, G., Is a sea turtle no more than an amioured stomach?. Bull. South Afr.
Assoc. Mar. Biol. Res., 11, 12, 1974.
137. Hendrickson, J. R., The ecological strategies of sea turtles, Am. Zool., 20, 597, 1980.
138. Wood, F. E. and Wood, J. R., Quantitative requirements of the hatchling green sea
turtle, Chelonia mydas, for valine, leucine, isoleucine and phenylalanine, J. Nutr.,
107, 1502, 1977.
139. Wood, J. R. and Wood, F. E., Quantitative requirements of the hatchling green sea
turtle for lysine, tryptophan, and methionine,/. Nutr., 107, 171, 1977.
140. Patterson, J. I., Vitamin A Requirement of the Hatchling Sea Turtle, Chelonia mydas.
Thesis, University of Arizona, Tucson, 1974.
141. Williams, S. L., Thalassia testudinum productivity and grazing by green turtles in a
highly disturbed seagrass bed. Mar. Biol., 98, 447, 1988.
142. Bjorndal, K. A., Fermentation in reptiles and amphibians, in Gastrointestinal Micro
biology, Vol. 1, Mackie, R. I. and White, B. A., Eds., Chapman and Hall, New York,
in press.
143. Davenport, J., Antipas, S., and Blake, E., Observations of gut function in young green
turtles Chelonia mydas L., Herpetol. J., 1, 336, 1989.
144. Wood, J. R. and Wood, F. E., Growth and digestibility for the green turtle (Çhelonia
mydas) fed diets containing varying protein \ovo\s. Aquaculture, 25, 269, 1981.
145. Lutcavage, M. and Lutz, P. L., Metabolic rate and food energy requirements of the
leatherback sea turtle, Dermochelys coriácea, Copeia, 1986, 796, 1986.
146. Duron, M., Contribution à l’Étude de la Biologie áo Dermochelys coriácea (Linné)
dans les Pertuis Charentais, Thèse P Université de Bordeaux 3me cycle, 1978.
147. Vijayakumaran, M., Rajagopalan, M., and Silas, E. G., Food intake and conversion
in hatchlings of olive ridley Lepidochelys olivácea fed animal and plant food, in Sea
Turtle Research and Conservation, Silas, E. G., Vijayakumaran, M., and Meenak-
shisundaram, P. T., Eds., CMFRI Bulletin No. 35, Cochin, India, 1984, 41.
148. Bjorndal, K. A., Cellulose digestion and volatile fatty acid production in the green
turtle, Chelonia mydas. Comp. Biochem. Physiol., 63A, 127, 1979.
149. Bjorndal, K. A., Digestibility of the sponge Chondrilla nucula in the green turtle,
Chelonia mydas. Bull. Mar. Sei., 47, 567, 1990.
150. Hadjichristophorou, M., and Grove, D. J., A study of appetite, digestion and growth
in juvenile green turtle {Chelonia mydas L.) fed on artificial diets. Aquaculture, 30,
191. 1983.
151. Davenport, J. and Scott, C. R., Individuality of growth, appetite, metabolic rate and
assimilation of nutrients in young green turtles (Chelonia my das L.), Herpetol. 3,
26, 1993.
152. Bjorndal, K. A., Bolten, A. B., and Moore, J. E., Digestive fermentation in herbivores:
effect of food particle size. Physiol. ZooL, 63, 710, 1990.
153. Bjorndal, K. A. and Bolten, A. B., Body size and digestive efficiency in a herbivorous
freshwater turtle: advantages of small bite size. Physiol. ZooL, 65, 1028, 1992.
154. Bjorndal, K. A. and Bolten, A. B., Digestive processing in a herbivorous freshwater
turtle: consequences of small intestine fermentation, P/2y.9/<9/. ZooL, 63, 1232, 1990.
155. Bjorndal, K. A., Suganuma, H., and Bolten, A. B., Digestive fermentation in green
turtles, Chelonia mydas, feeding on algae. Bull. Mar. Sci., 48, 166, 1991.
156. Zimmerman, L. C. and Tracy, C. R., Interactions between the environment and
ectothermy and herbivory in reptiles. Physiol. ZooL, 62, 374, 1989.
157. Meylan, A., The role of sponge collagens in the diet of the hawksbill turtle {Eretmo-
chelys imhricata), in Biology o f Invertebrate and Lower Vertebrate Collagens, Bairati,
A. and Garrone, R., Eds., Plenum Press, New York, 1985, 191.
158. Meylan, A., Nutritional characteristics of sponges in the diet of the hawksbill turtle,
Eretmochelys imhricata, in New Perspectives in Sponge Biology, Rutzler, K., Ed.,
Smithsonian Institution Press, Washington, D.C., 1990, 472.
159. Bjorndal, K. A., Diet mixing: nonadditive interactions of diet items in an omnivorous
freshwater turtle. Ecology, 72, 1234, 1991.
160. Bjorndal, K.Â. and Bolten, A. B., Digestive efficiencies in herbivorous and omniv
orous freshwater turtles on plant diets: do herbivores have a nutritional advantage?.
Physiol. ZooL, 66, 384, 1993.
161. Davenport, J. and Oxford, P. J., Feeding, gut dynamics, digestion and oxygen con
sumption in hatchling green turtles {Chelonia mydas L.), B r J. Herpetol., 6, 351,
1984.
162. Birse, R. F. and Davenport, J., A study of gut function in young loggerhead sea turtles,
Caretta caretta L. at various temperatures, Herpetol. J., 1, 170, 1987.
163. Meienberger, C., Wallis, I. R., and Nagy, K. A., Food intake rate and body mass
influence transit time and digestibility in the desert tortoise (Xerobates agassizii).
Physiol. ZooL, 66, 847, 1993.
164. Penry, D. L., and Jumars, P. A., Modeling animal guts as chemical reactors,Aw. Nat.,
129, 69, 1987.
165. Balazs, G. H., Impact of ocean debris on marine turtles: entanglement and ingestion,
in Proc. Workshop on the Fate and Impact o f Marine Debris, Shomura, R. S. and
Yoshida, H. O., Eds., NOAA Tech. Memo. NOAA-TM-NMFS-SWFC-54, Honolulu,
HI, 1985, 387.
166. Carr, A., Impact of nondegradable marine debris on the ecology and survival outlook
of sea turtles. M ar Pollut. Bull., 18, 352, 1987.
167. Bjorndal, K. A., Bolten, A. B., and Lagueux, C. J., Ingestion of marine debris by
juvenile sea turtles in coastal Florida habitats, Mar. Pollut. Bull., 28, 154, 1994.
168. National Research Council, Decline o f the Sea Turtles: Causes and Prevention,
National Academy Press, Washington, D.C., 1990.
169. Hall, R. J., Belisle, A. A., and Sileo, L., Residues of petroleum hydrocarbons in
tissues of sea turtles exposed to the Ixtoc I oil spill,/. Wildl. Dis., 19, 106, 1983.
170. Hutchinson, J. and Simmonds, M., Escalation of threats to marine turtles. Oryx, 26,
95. 1992.
171. Lutz, P. L., Studies on the ingestion of plastic and latex by sea turtles, in Free. 2nd
bit. Conf. Marine Debris, Shomura, R. S. and Godfrey, M. L., Eds., NOAA Tech.
Memo. NOAA-TM-NMFS-SWFSC-154, Honolulu, HI, 1990, 719.
172. Schulman, A. A. and Lutz, P., The effect of plastic ingestion on lipid metabolism in
the green sea turtle {Cheionia mydas), in Proc. 12th Anna. Workshop on Sea Turtle
Biology and Conservation, Richardson, J. I. and Richardson, T. H., Compilers, NOAA
173. Pauly, D. and Christensen, V., Primary production required to sustain global fisheries,
Nature, 374, 255, 1995.
174. Wetherall, J. A., Balazs, G. H., Tokunaga, R. A., and Yong, M. Y. Y, Bycatch of
marine turtles in North Pacific high-seas driftnet fisheries and impacts on the stocks,
North Pac. Comm. Bull., 53, 519, 1993.
175. U.S. Fish and Wildlife Service, and National Marine Fisheries Service, Recovery
plan for the Kemp’s ridley sea turtle {Lepidochelys kempii). National Marine Fisheries
SeiTice, St. Petersburg, FL, 1992.
176. National Research Council, An Assessment o f Techniques fo r Removing Offshore
Structures, National Academy Press, Washington, D.C., 1996.
177. Bjorndal, K. A., unpublished data.
178. Limpus, C., personal communication.
179. McCauley, S. J. and Bjorndal, K. A., unpublished data.
Q Age, Growth, and
Population Dynamics
M. Y. Chaloupka and John A. Musick
9.1 Introduction................................................................................................... 234

9.2 Methodological Background..........................................................................235
9.2.1 Sampling Issues......................................... .......................................235
9.2.1.1 Demographic Unit............................................................. 235
9.2.1.2 Demographic Sampling Design........................................ 235
9.2.2 Mark-Recapture Procedures.............................................................236
9.2.2.1 Marking Methods.............................................................. 237
9.2.2.2 Tagging Protocols............................................................. 237
9.2.2.3 Tag Loss Analysis............................................................. 238
9.2.3 Skeletochronology............................................................................. 238
9.2.3.1 Growth Mark Validation...................................................239
9.2.3.2 Layer Loss CoiTection Protocols...................................... 239
9.2.3.3 Length-at-Age Back Calculation......................................240
9.2.4 Growth............................................................................................... 241
9.2.4.1 Growth Measurement................................................... .....241
9.2.4.2 Age-Based Growth Functions........................................... 241
9.2.4.3 Mark-Recapture Analogues.............................................. 242
9.3. Growth Models............................................................................................ 246
9.3.1 Age-Based Growth Model................................................................ 247
9.3.1.1 Cross-Sectional Sampling................................................. 247
9.3.1.1.1 Known A ge...................................................... 247
9.3.1.1.2 Estimated Age (Skeletochronology)................248
9.3.1.2 Longitudinal Sampling........... ......................................... 249
9.3.1.3 Mixed Longitudinal Sampling.......................................... 250
9.3.1.3.1 KnownAge (Headstarted Stock)......................250
9.3.1.3.2 Known Age (Wild Stock).................................251
9.3.2 Size-Based Growth Model................................................................251
9.3.2.1 Cross-Sectional Sampling................................................. 251
9.3.2.1.1 Length Frequency............................................251
9.3.2.1.2 Parametric Growth Rate Model.......................253
9.3.2.2 Longitudinal Sampling..................................................... 253
9.3.2.3 Mixed Longitudinal Sampling.......................................... 253
9.3.2.3.1 Known Age (Headstarted Stock).....................253
0-8493-8422-2/97/$0 00+$.50
9.3.2.3.2Growth Interval.................................................... 254

9.3.2.33 Size-Based Growth Analogues........................254
9.3.2.3.4 Nonparametric Growth Rate Model................255
9.4. Comparative Life History Patterns.............................................................. 256
9.4.1 Comparative Life History Traits...................................................... 256
9.4.2 Comparative Demography................................................................ 259
9.5. Population Dynamics.................................................................................... 261
9.5.1 Deterministic Cohort Models........................................................... 262
9.5.1.1 Static Age-Structured Models.......................................... 262
9.5.1.2 Dynamic Age-Structured Models...................................... 263
9.5.1.3 Dynamic Stage-Structured Models.................................. 264
9.5.2 Stochastic Simulation Model............................................................ 267
9.5.2.1 Dynamical Systems.......................................................... 267
9.5.2.2 Monte Carlo Process........................................................ 268
9.6. Conclusion.................................................................................................... 268
Acknowledgment................................................................................................... 270
References........................................................................ 270
9.1 INTRODUCTION
Many sea turtle stocks are in decline worldwide due to excessive harvesting and
destruction of nesting and foraging habitats.^ Yet population abundance for many
stocks can display stochastic fluctuations with no apparent long-term trend. How
are we to know then whether a stock is really at risk and what risk management
strategies might be appropriate for protection of the stock? These are challenging
questions that cannot be resolved without a thorough understanding of sea turtle
growth and population dynamics. Age and size are related components of somatic
growth, which itself is a major determinant of population dynamics and macroevo-
lutionary change.
Therefore, we review in broad terms a wide range of studies on age, somatic
growth, and the population dynamics of sea turtle stocks. Firstly, we present an
overview of some of the major methodological issues relevant to understanding sea
turtle growth and population modeling. This section is essential to appreciate the
methodological themes apparent in the remaining sections. Many of the methods
used need careful scrutiny of the underlying statistical properties and assumptions.
Therefore, the main emphasis in this review is on methodological issues.
Then we review critically a compilation of sea turtle growth studies arranged in
a classification table providing a framework for grouping similar approaches. Next
we discuss the complementary approaches of comparative life history analysis and
comparative demography as a precursor to population dynamics. In the last section
we review critically all known studies on sea turtle population dynamics, also
arranged in a classification table. We conclude with a brief overview of the review
focusing on recurrent methodological issues.
Age, Growth, and Population Dynamics 235
9.2 METHODOLOGICAL BACKGROUND

9.2.1 S ampling Issues
Ecological studies are based on many presuppositions including two key sampling
design issues — (1) the stock is distinguishable over time and space and (2) the
demographic processes of that stock are time invariant. The first design issue is
reasonably well understood, but the second issue of implicit time invariant sampling
is very poorly understood in most ecological research, let alone in sea turtle studies.
9.2.1.1 Demographic Unit

The identification of discrete demographic units is fundamental to understanding
growth dynamics, population dynamics, and macroevolutionary change. Recent stud
ies on growth and dispersal in the epipelagic developmental habitat^’^^’^ as well as
in foraging areas^'^^ are helping to provide a more complete description of sea turtle
ontogeny and spatial dispersion patterns. So there is no apparent reason why eco
logical modeling of sea turtle stocks would take a restricted perspective of sea turtle
demography.
However, identifying a discrete demographic unit poses a major challenge. Sea
turtle breeding stocks are likely to comprise multiple rookeries within a region, while
foraging areas and developmental habitats comprise a mix of turtles from several
genetically distinct stocks.^>^^ But the intermingling of individuals from different
stocks in different habitats makes it difficult to distinguish individuals from each
stock and so makes it difficult to recognize the spatial dispersion pattern for each
stock. It is important therefore to be able to recognize individuals from the different
stocks in these separate habitats, the main reason being to link individuals with their
correct breeding, developmental, and foraging habitats so that the focus of a study
and conclusions are relevant to that stock.
It is now possible to distinguish discrete sea turtle demographic units across
breeding, foraging, and developmental habitats using a mark-recapture program
combining stock-specific genetic markers and physical individual tagging.^’^^
Assuming demographic independence for distinct stocks and the capacity to recog
nize each stock, it follows that the appropriate management and population modeling
focus should be a genetically distinct breeding stock and its associated developmental
habitats and foraging areas. For practical purposes the management focus might in
some cases be limited geographically because of trans-boundary jurisdictions. Nev
ertheless, all sea turtle growth and population studies should be referenced to a
genetically distinct stock enabling a better understanding of the full spatial dispersion
pattern of each stock.
9.2.1.2 Demographic Sampling Design
Estimates of demographic rates such as mortality, growth, and recruitment rates are
a key part of modeling population dynamics. However, demographic processes are
complex time-dependent functions comprising three confounded time effects (age,

year, cohort). In other words there is natural variability in demographic rates attrib
utable to age-specific, interannual, and cohort-specific factors. Year effect is consid
ered an environmental factor while the cohort effect is usually considered to be a
genetic factor.
Confounded time effects can result in biased estimates, so demographic analyses
need to uncouple the time effects by accounting for age, year, and cohort factors in,
say, a tabular model known as a lexis diagram)^ or a statistical model. Accounting
for time-dependent effects has rarely been done in sea turtle studies. It is not always
possible to uncouple the time effects in many studies. For instance, Tucker et al.,^"^
in one of the rare ecological studies to recognize the issue of confounded time effects,
proposed a method to account for interannual growth rate variation in freshwater
turtle species turtles. They concluded that failure to account for this time effect
would lead to incorrect growth rate estimation. Age was determined by independent
means, but the authors seem unaware that both age and year were confounded due
to the implicit cross-sectional sampling design. It is important to understand that
there are three time effects involved and, if it is not possible to uncouple all of them,
that you must be aware which factors are confounded.
Any form of time-dependent demographic sampling that involves cross-sectional
sampling (random sampling with replacement) of a turtle stock will confound the
main effects of age and cohort as well as the interactions. Longitudinal sampling
(repeated sampling on the same individuals) will confound age and year main effects
and interactions. Mixed longitudinal sampling (sampling with partial replacement)
will confound only the interactions if age is known or will confound year, cohort,
and interactions if age is unknown. All sea turtle growth studies are based on one
of these sampling schemes with major implications for interpretation. A discussion
of age, year, and cohort effects in relation to implicit sampling design is presented
by Berger,^^ while Fitzhugh^^ gives a classification of data types used in growth
studies.
9.2.2 M ark-R ecapture P rocedures
Metal tagging of flippers is the most common approach used for mark-recapture
programs designed for long-tenn monitoring of individual t u r t l e s . O v e r time
a detailed mark-recapture profile is compiled for each turtle comprising morpho
metric measurements at first and subsequent recaptures, time of first capture, time-
at-large since first capture, and in some cases determination of sex, maturity, and
reproductive status derived from visual examination of gonads using laparoscopy.^^’^^
These profiles provide the basis for estimating demographic parameters needed
for population modeling such as mortality r a t e s , r e c r u i t m e n t ra te s ,in te r
habitat dispersal rates,^^ annual remigration rates,^'^ and sex-, size-, and age-specific
growth rates.^^’^^‘^ The collection of demographic information on individual turtles
requires a reliable means of capture and marking to enable recognition at recapture.
Clearly, the reliability of the marking methods is a concern for long-term mark-
recapture programs.
9.2.2.1 Marking Methods

The methods of capture of individual turtles range from passive capture of cold-
stunned turtles or nesting females to active approaches used in shallow water such
as netting, beach jumping, and the turtle rodeo technique.^^ Deeper water capture is
usually done using SCUBA. Miller^^ provides a thorough descriptive overview of
the range of methods used to catch sea turtles in foraging grounds and the specific
data-oriented purpose of each method. There have been no studies of the demo
graphic sampling bias associated with any of these capture methods — all studies
presume that the sample is an unbiased representation of the population under study.
Following capture, a variety of tagging devices have been used to mark individual
sea turtles depending on cost, study purpose, and ontogenetic stage. Hatchlings or
juveniles have been marked with mutilation notches,^’^ injected with magnetically
coded wire tags,^^ autografts,^^ or plastic tags.^^ Immature and adult turtles are
usually marked with one or more flipper tags of various materials such as plastic,
m o n e l , i n c o n e l , ^ * ' ^ ’^^’^^’^^ or titanium.
Other forms of postnatal marking include passive integrated transponders (PIT)
tags, where metal tags were unable to be secured,^^ and radio or satellite biotelemetry
for tracking in real time of local movemenP^ '^®or long-distance dispersal."^^ Skeletal
elements have also been marked by injection of oxytetracycline (OTC) (see Section
9.2.3) to record a bone growth layer benchmark suitable for skeletochronological
aging.^^
9.2.2.2 Tagging Protocols
Most sea turtle monitoring programs use multiple tagging with subsequent replace
ment protocols, although a few have relied on single tagging such as for green turtles
nesting in Malaya and Saraw ak,green turtles nesting at Tromelin and Europa in
the Indian Ocean,^^ and for recaptured headstarted Kemp’s ridleys released in the
Gulf of Mexico.^^ While many long-term sea turtle monitoring programs have used
multiple tagging protocols, most of them have also used monel alloy tags, which
are known to corrode rapidly in seawater.'^^ '^'^ Clearly, studies that used single monel
tagging^^’^^’^'^’^^"^^ are most at risk, although Caillouet et al.^^ have also used single
inconel tags, and Limpus et al.,^’^ Limpus,^^ and Frazer et aU^^ have long since moved
to multiple titanium or inconel tagging. Single tagging, even if it is an inconel tag'^^
or a titanium tag,^® is not a prudent practice. Plastic tags are known to deteriorate
rapidly in seawater, probably resulting in low tag retention rates.
Both inconel and titanium tags have outstanding long-term durability in seawa
ter. 18,44 However, both monel and inconel are alloys containing nickel, which is a
known antifouling agent. Titanium, on the other hand, has no antifouling properties,
although it also has no known toxic properties. Limpus'^'^ has noted that titanium
tags on recaptured sea turtles in the Great Baixier Reef are often covered with
barnacles, which he attributes to potential tag loss. Wood and Wood^^ suggested an
association between titanium tags and fibropapillomas on recaptured headstarted
green turtles released from the Cayman Islands. Studies using both inconel and
titanium tags^’^’^®conducted in accordance with a properly designed experimental
protocol could help to resolve these issues. Clearly, more research on tag design,
material technology, and tag retention rates is needed, but it might well be that
inconel and perhaps titanium are the preferred materials for long-term monitoring
of sea turtle stocks.
9.2.2.3 Tag Loss Analysis
Sea turtle tag loss is a complex function of time-dependent competing risks"^^’"^"^such

as (1) incorrect application of tag, (2) loss due to mechanical stress at nesting or
from courtship biting, (3) tissue necrosis, (4) tag corrosion, or (5) censored record
due to death and unrecovered tag. These risk factors suggest why multiple tagging
with replacement is a prudent practice. The first two factors reflect an early failure
period; tissue necrosis reflects variable, but probably relatively early failure, while
corrosion and censoring reflect a later failure period. The underlying hazard function
is nonmonotonic or multimodal where tag loss is high initially due to early risk
factors, followed by a period of decreasing risk, and then followed by a period of
increasing risk due to corrosion and observation censoring as both the tag and the
turtle age.
This hazard function explains the finding in Henwood"^^ that his analysis pre
dicted far higher loss rates than would seem consistent with recorded renesting rates
of tagged females. The linear regression model used by Henwood"^^ assumes a
constant hazard function, no competing risks, and no censoring processes and is not
suitable for analysis of tag retention.
Limpus,"^"^ in a detailed analysis of tag retention risk, used a logit regression
approach, which assumes a monotone increasing hazard. He assumed no censoring
processes and the implicit hazard function would be inappropriate if an early failure
period occurs. Censored regression models with risk-specific nonmonotone hazard
functions'^^ might be worth considering in future analyses. Nonetheless, Limpus"^"^
provided a major contribution to tag loss analysis, supporting confidence in results
obtained from his long-term tagging studies of Great Barrier Reef loggerhead,
hawsbill, and green turtle stocks.
9.2.3 Skeletochronology
Periodic marks or growth checks on hard parts such as scales or bones have been
used to estimate vertebrate age for almost 100 years.^* The procedure assumes
amongst other things that (1) a constant proportional relationship exists between
hard part growth and somatic growth and (2) that discernible periodic growth rings
or annuli are deposited at regular constant time intervals. Zug et al.^^ reviewed the
history of the use or attempted use of bony rings to estimate age (skeletochronology)
in sea turtles. The future of skeletochronology for estimating sea turtle ages rests
with resolving three issues — (1) validation of universal annual deposition of growth
layers, (2) development of a robust method to correct age estimates for growth layer
loss, and (3) validation of constant proportional allometry between bone growth and
somatic growth to enable application of length-at-age back calculation methods.
9.2.3.1 Growth Mark Validation
Zug et al.^^ used the humeral bones to estimate age and growth for northwestern
Atlantic loggerheads. They assumed that growth zones apparent following histolog
ical processing of the bone were deposited on an annual basis, although others^^ had
mixed success previously in validating the annual nature of bony rings in sea turtles.
Klinger and Musick"^^ have validated the annual nature of the humeral growth
layers in loggerheads using OTC injection. Several turtles were subsequently
stranded dead or were recaptured live by fishermen after being at large for up to
four years. Humeri were removed from dead animals and processed according to
Zug et al.^^ The number of growth rings deposited beyond the fluorescent OTC
benchmark were then counted. Live recaptured turtles were anaesthetized and
humeral bone biopsy samples were taken and histologically processed as in whole
humeral sections.^"^ The results of this study showed that one growth layer was
deposited for each year a turtle was at large for this stock after OTC marking.
Annular growth zones in the humeri of a wide size range of western North
Atlantic loggerheads have now been validated. In addition, growth zones have been
validated in a small sample of Lepidochelys kempi?^^ It remains to be seen whether
such zones are formed on an annual basis over the entire size range of L. kempi, or
for small pelagic loggerheads.
Frazier^^ also used the OTC method to attempt to validate the nature of growth
zones in a captive group of green turtles, but had poor results. Zug et al.^^ suggested
that this was likely because the growth rates were accelerated in captivity and so
that the normal pattern of bone growth might have been disrupted. This is an
important point because supposedly aseasonal environments such as the tropics may
not induce strong seasonal pulses in growth commonly found for more warm tem
perate species such as loggerheads or Kemp’s ridley — the two species subject to
most aging studies. Annular layers have yet to be validated for sea turtle species
with a tropical distribution pattern such as greens and hawksbills.
9.2.3.2 Layer Loss Correction Protocols
Growth layer resorption in skeletal elements has been found to occur in post-pelagic
growth phases of loggerheads, Kemp’s ridleys, and leatherbacks.'^^’^^’^^ ^’^ Zug et al.^
have also shown that growth layer resoiption also occurs in pelagic loggerheads less
than 20 cm curved carapace length (ccl) or about 2 years old caught in the North
Pacific high seas driftnet fishery. Therefore, it seems likely that growth layer resorp
tion occurs in all postnatal growth phases of sea turtles.
Zug et al.^^ proposed originally a simple absolute growth rate averaging method
to compensate for growth layer loss when determining age estimates for post-pelagic
loggerheads. However, this method may overestimate the number of layers because
growth may be faster and growth layers may be wider in younger, fast-growing
turtles than in older, slower-growing turtles.O ther adjustment methods to correct
for growth layer loss have since been developed and are now more widely applied
— (1) correction factor p r o t o c o l , (2) regression growth protocol,^'^^ and (3)
the ranking protocol.^ '’^-'^^ '’^ The correction factor and regression growth protocols
are based on several assumptions related to constant relative growth rates predicated
on a von Bertalanffy growth function and in some cases application of the Fraser-
Lee back calculation function The use of the Fraser-Lee BCF is also
discussed in the next section concerning length-at-age estimation. The Zug ranking
protocol'’'îs simpler and based on fewer assumptions. Nonetheless, all three methods
require further validation and development to warrant confidence in the age estimates
derived using these adjustment methods.
9.2.3.3 Length-at-Age Back Calculation
The standard Fraser-Lee back calculation method was used to derive loggerhead and
Kemp’s ridley length-at-age records.^^’"^^’^^’^^ The specific reason for back calculation
other than for growth layer loss adjustment was unclear, but is presumed to be for
increasing the sample size for fitting an age-size growth function. The sample could
also include age-length estimates for lengths not observed in the study. Length back
calculation derives multiple age-length measurements for each individual, depending
on the ontogenetic stage reached by the individual at time of biopsy or death. Back
calculation in theory, then, represents mixed longitudinal sampling, which is desir
able. However, it has been common practice to use a standard Fraser-Lee BCF
providing only a length estimate at the earliest discernible growth mark.^^’^^’^^’^^ This
practice yields only mixed cross-sectional data at the expense of discarding growth
information.
Zug et al.,^^ in an analysis of Kemp’s ridley growth, also used Fraser-Lee-derived
back calculated lengths for fitting the Fabens size-based analogue of the von Ber
talanffy growth function. The Fabens analogue is not appropriate for determining
an age-length function^® (see Section 9.2.4.3). Moreover, the parameter form of the
von Bertalanffy growth function fitted to all these age-length data sets is known to
have poor statistical p ro p e rtie s .T h e s e issues are discussed in more detail in
Section 9.2.4.
There appears to be much confusion in the sea turtle skeletochronological liter
ature regarding the correct approach for using proportional BCFs. The main problems
seem to be due to a misunderstanding of the following: (1) implicit proportionality
hypothesis, (2) correct form of the standard Fraser-Lee equation, (3) potential of the
general Fraser-Lee equation to derive multiple length-at-age measurements per indi
vidual, and (4) correct linear regression to use, depending on the particular BCF used.
Assuming a scale proportional hypothesis, a regression of the data set comprising
the humerus radial length (y variable) and straight carapace length (x variable)
recorded from each individual at capture should be applied.^^ Ricker^“^ argues that
the order of (y) and (x) is irrelevant so long as a symmetrical linear regression model
such as the reduced major axis model is used to derive the parameter estimates. If
the fitted functional relationship is linear, then it is possible using the regression
parameter estimates to derive back calculated length estimates for a specific indi
vidual at all previous ages indicated by the growth checks using an appropriate BCF,
such as the general Fraser-Lee equation.^^ This generates a set of length-at-age data
with variable number of length-age records per individual for fitting an age-size
growth function. There is, of course, no necessity to use a regression model in the
first place, since biologically determined values where available could just as validly
be used in any proportional Irrespective of method, current BCFs do not take
into account the time-dependent nature of the relationship between skeletal and
somatic growth rates. Future sea turtle aging models using skeletochronology and
back calculation will need to evaluate the assumptions of proportionality and time-
invariant growth rates.
9.2.4 G rowth
9.2.4.1 Growth Measurement
Somatic growth for a mark-recaptured turtle is described in terms of length or weight

with length measured as either straight carapace length (scl) or curved carapace
length (ccl). There seems little basis for preferring scl over ccl measurements except
perhaps for leatherbacks, where ccl is more appropriate (see Zug and Parham).^^
Somatic growth is a dynamic process and can be approximated in terms of either
absolute growth rates or specific growth rates.A bsolute growth rates are described
as change in size measured as length or weight recorded over a time interval of
variable duration. Specific growth rates are described as the absolute growth rate
per unit size estimated simply as the logarithmic absolute growth rate.
Strictly speaking, these are only really rates if the time interval is a constant
unit interval. In sea turtle studies annual growth is usually measured as the average
rate of size gain recorded over the time interval between first capture and recapture,
which is rarely, if ever, one year. This presumes that the absolute growth rate is
locally linear or constant across the time interval. For small time intervals this might
not be a serious issue. But as the interval between recaptures increases, then it might
well be important, since growth is unlikely to be linear, causing major growth rate
measurement error. Chaloupka and Limpus^^ and Limpus and Chaloupka^^"^ found
that the recapture interval measured in years was not a significant factor affecting
absolute growth rates in their study of the southern Great Barrier Reef (sGBR) green
turtle stock. This was a limited test of this problem and needs to be tested for each
specific study before mark-recapture interval could be discounted.
Most studies of sea turtle growth have used absolute growth rates, but Bjomdal
and BolteiT^ used both absolute and specific growth rates in their study of immature
green turtles resident in the Bahamas. For comparative growth analysis, specific
growth rates have some advantages based on additive statistical properties and per
unit measurement. On the other hand, many studies are concerned with estimating
a size-based growth function rather than summarizing size-specific growth in a single
numeric measure, and so absolute growth may be appropriate.
9.2.4.2 Age-Based Growth Functions
The main purpose of growth studies when age is known has been to fit a parametric
growth function useful for estimating important demographic processes such as
maturation rate. The von Bertalanffy growth function has been widely used in sea
turtle growth studies (see Section 9.3). The age-based form of this function has no
inflexion representing a monotone declining growth rate as the turtle ages through
the postnatal development phase, irrespective of whether absolute or specific growth
is used. Interestingly, of all the studies on sea turtle growth purporting von Berta-
lanffy growth, only two studies^^’^^ showed a monotone declining size-specific
growth rate, and then only for part of the postnatal development phase.
The von Bertalanffy growth function has very poor statistical properties resulting
in misleading and biased growth parameter estimates.^’ The same problems apply
to the generalized von Bertalanffy or Chapman-Richards function^^ used in a known-
age growth study of West Atlantic immature green turtles (see Table 9.1). An even
more generalized model known as the Richards function has been shown to have
the worst statistical properties of any known function.T his function, also known
as the generalized logistic or Richards-Nelder function, has not yet been used in sea
turtle growth studies.
Ratkowsky^^ has shown that other common parametric age-based growth func
tions used in sea turtle growth studies such as the logistic^^ and Gompertz^'^ functions
have excellent statistical properties. The sex-specific Weibull functions used by
Chaloupka and Limpus^^ also have good statistical p ro p erties.O f course, this
applies so long as the function is fitted to data spanning the whole postnatal devel
opment phase.
Ratkowsky^^ has also shown that the statistical properties of many nonlinear
parametric growth functions can be improved significantly by using expected-value
parameters. This involves replacing some function parameters with known or fitted
response variable values that correspond with, say, the lowest and highest known
values of the explanatory variable. Say for instance that the response variable was
ccl with known age (years) as the explanatory variable, then initial estimates of
suitable expected-values can be read from a nonparametric regression smooth^'^ of
the bivariate data plot. The initial estimates would replace unknown parameters in
a reparameterized form of the function being used. Recall that this concept of using
biologically relevant values was also suggested for length-at-age back calculation
procedures (see Section 9.2.3.3).
Schnute and Fournier^® and Ratkowsky^* provide expected-value parameter
forms of the von Bertalanffy function. All reparameterizations of the von Bertalanffy
growth function have excellent statistical properties.^^ A point implicit in the
expected-value approach is that growth functions are only fit to the range of observed
data. Extrapolating any function beyond the empirical range is fraught with diffi
culties. This principle applies irrespective of the growth function used, but has not
been well adhered to in sea turtle growth studies (see Section 9.3).
9.2.4.3 Mark-Recapture Analogues

It has been common practice to use mark-recapture or size-based analogues of the
parametric growth functions when age is unknown, which is invariably the case with
sea turtles. Mark-recapture analogues used are the absolute Fabens^^ and specific
Kaufman^^ growth analogues of the von Bertalanffy growth function and the absolute
growth analogue of the logistic growth fu n ctio n .It is a simple task to derive these
types of analogues for many other parametric growth functions.
e>
ra
fD
TABLE 9.1
n
Classification Summary of Sea Turtle Growth-at-Age and Growth-at-Size Literature o
Examples
Growth
Focus Sampling Method Ref. (year;no.) Species Phase Location Metric function 13
Q_
“U
Age Cross- Known age Ackerman (1981; 80) Green/loggerhead Embryonic Western Atlantic Weight Logistic o3
TC
sectional
Estimated age Zug and Parham (1996;57) Leatherback Hatchling-adult Pacific/Florida ccl von Bertalanffy O
13
(skeletochronology) Klinger and Musick (1995;36)^ Loggerhead Immature Chesapeake Bay sci von Bertalanffy
a
Zug et al. (1995;6) Loggerhead Pelagic North Pacific scl/ccl von Bertalanffy^
Parham and Zug (1997;58)^ Loggerhead Subadult-adult Georgia ccl von Bertalanffy a;
3
Zug et al. (1996;56)‘ Ridley (kemp) Juvenile-adult U.S. east/Gulf scl/ccl von Bertalanffy/ n
coast Fabens
Longitudinal Known age captive- Witham and Futch (1977;83) Green/loggerhead Yearling Florida Weight/scl Monotonie?
reared Witzell (1980;87) Hawksbill Immature Western Samoa sci von Bertalanffy^
Uchida (1967;90) Loggerhead Hatchling-immature Japan Weight/scl Logistic
Frazer and Schwartz (1984;67) Loggerhead Hatchling-immature North Carolina sci Logistic
Marquez (1972;88) Ridley (kemp) Hatchling-immature Gulf of Mexico sci von Bertalanffy
Caillouet et al. (1986;68) Ridley (kemp) Hatchling-immature Gulf of Mexico Weight Gompertz
Phasuk & Rongmuangsart Ridley (olive) Hatchling-immature Phuket Weight/scl Nonmonotonic
(1973;84)
Mohanty-Hejmadi (1992;89) Ridley (olive) Hatchling-immature Orissa Weight/scl Nonmonotonic
Mixed Known age recapture Ehrhardt and Witham Green Hatchling-immature Florida Weight Chapman-
(headstarted) (1992;91)2 Richards
Caillouet et al. (1995;37)‘ Ridley (kemp) Yearling-immature Gulf of Mexico sci von Bertalanffy
Known age recapture Limpus et al. (1994;8)- Green Hatchling-immature Moreton Bay ccl
(wild) Limpus et al. (1994;9)^ Loggerhead Hatchling-immature Moreton Bay ccl
KÎ
e>
0ra
O
ification Summary of Sea Turtle Growth-at-Age and Growth-at-Size Literature o
Examples
Growth ÎD
Sampling Method Ref. (year;no.) Species Phase Location Metric function D
Q_
-g
Size Cross- Length frequency Le Toquin et al. (1980;92) Green Immature Mauritania sci von Bertalanffy
O
■D
sectional modes Bjomdal et al. (1995;93) Green Immature Bahamas sci von Bertalanffy E.
Parametric growth Bolten et al. (1992;66) Green Immature Bahamas sci Monotonie Ö'
13
rate modeling
o
Longitudinal Known age captive Swingle et al. (1993;96) Loggerhead Hatchling-immature Virginia Weight/scl Monotonie 13
reared o>
Mixed Known age recapture Kowarsky and Capelle Green Yearling Torres Strait cd
(headstarted) (1979;45)2
Wood and Wood (1993;30)^ Green Hatchling-immature Cayman Islands Weight/ccl
Bjomdal and Bolten (1988;47) Loggerhead Immature Bahamas sci Monotonie
Growth Limpus and Walter (1980;46) Green Immature-adult Great Barrier cd Nonmonotonic
interval/growth Reef
record Balazs et al. (1994; 18) Green Immature Hawaii sci Nonmonotonic
Bjomdal and Bolten (1988;33) Green Immature Bahamas sci Monotonie
Collazo et al. (1992;35)^ Green Immature Puerto Rico sci Nonmonotonic
Green (1993;32)i Green Immature Galapagos sci Monotonie
Islands
Mendonca (1981 ;97) Green/loggerhead Immature Eastern Florida sci Monotonie
Limpus (1992;20) Hawksbill Immature-adult Great Barrier ccl Nonmonotonic
Reef
Boulon (1994;34)i Hawksbill Immature U.S. Virgin sci Monotonie
Islands
KJ
4ì»
>
era
0)
O
O
g.
OJ
Z3
Q_
~0
Bjomdaletal. (!994;31)2 Loggerhead Pelagic (incomplete) Northeast scl/ccl Nonmonotonic ■O
O
c
Atlantic
Mark-recapture Bjomdal and Bolten (1988;33) Green Immature Bahamas scl Kaufman^ O
13
growth analogue Boulon and Frazer (1990;98)^ Green Immature U.S. Virgin scl Fabens O
Islands U
eu
Frazer and Ehrhart (1985;81)^ Green/loggerhead Immature Eastern Florida scl Fabens/Schoener 3
Frazer et al. (1994;49)' Loggerhead Subadult-adult Great Barrier ccl Fabens/Schoener
Reef
Bjomdal et al. (1995;93) Green Immature Bahamas scl Fabens
Nonparametric Chaloupka & Limpus Green Immature-adult Great Barrier ccl Nonmonotonic
growth rate (1996;28)^ Reef
modeling Chaloupka & Limpus Hawksbill Immature-adult Great Barrier ccl Nonmonotonic
(1996;91a) Reef
^ Explicit mixed cross-sectional; ^ Implicit mixed longitudinal, but mixed cross-sectional due to insufficient recaptures; ^ Reanalyzed here; von Bertalanffy function fitted
to left-censored data; ^ Specific growth rate metric; ^ See also Limpus and Chaloupka.^^^
to
Ü1
These analogues have been used for two main purposes — (1) to derive growth
parameter estimates from absolute or specific growth data sets provided from mark-
recapture studies and (2) to substitute these parameter estimates into an analogous
age-based growth function to generate a size-at-age function. There are two major
problems with both practices. Firstly, the poor statistical behavior of the von Ber-
talanffy function has also been shown for the Fabens a n a l o g u e , a n d using similar
arguments it follows that the logistic analogue^^ would have poor properties.
Francis^^ gives an expected-value reparameterization of the Fabens analogue while
Francis^'’ gives an analogue of Schnute’s general growth model,^^' which can also be
used in expected-value form. Bear in mind that these models provide growth-at-size
functions, not growth-at-age functions. Secondly, the widely accepted practice in
sea turtle studies of parameter substitution (see Section 9.3.2.3.3) is inappropriate
because size-based growth parameters are not equivalent to age-based growth param
eters — they are analogous, but not equivalent as clearly shown by Francis.^^ Put
simply, replacing the growth parameters in, say, the von Bertalanffy age-based
function with parameter estimates derived from either the Fabens^ ^ or Kaufman^^
size-based analogues is invalid. The same applies to use of all other analogues. The
only apparent way around this problem when age is unknown is to use numerical
integration to derived separate growth-at-size and growth-at-time-at-large functions.
This could also be done using a Francis analogue,^^ but, as discussed below, ana
logues cannot take into account the confounded time-dependent nature of growth
data unlike the statistical modeling and numerical integration approach of Chaloupka
and Limpus^^ and Limpus and Chaloupka.
Finally, a point relevant to statistical fitting of age- and size-based growth
functions concerns goodness of fit. It is common practice in sea turtle growth studies
using nonlinear regression to use R-square as a numerical goodness of fit measure.
This is a fallacious practice (Ratkowsky^'^ and references therein). Goodness of fit
for nonlinear regression is a matter of judgment based on residual variance, exam
ination of residuals, and assessment of underlying assumptions.
93 G R O W T H M O D E LS
A wide selection of sea turtle growth studies have been arranged in Table 9.1 into
one of twelve groups according to the following factors: (1) whether the main focus
was on age or size, (2) the implicit sampling design (cross sectional, longitudinal,
mixed longitudinal), and (3) the specific analytical method used. Table 9.1 includes
the citation, species, ontogenetic range, location, growth metric, and underlying
explicit or implicit growth behavior. We discuss the groups below, but first we raise
four major issues relevant to a critical review of all groups.
Firstly, sea turtle development undoubtedly comprises separate growth compart
ments representing distinct ontogenetic phases — Ricker^'“’ refers to ontogenetic
phases as growth stanzas. Just how many phases is unknown but at the very least
there are distinct prenatal and postnatal developmental phases. The number of phases
within the postnatal period will depend on the ontogenetic habit of each species.
For instance, the flatback {Natator depressus) has no epipelagic development phase,
but remains in a neritic habitat.^^ Notwithstanding individual growth variability.
distinct developmental phases are most likely to have different somatic growth
characteristics.^® However, it has been a very common practice in sea turtle growth
studies to fit a monophasic growth function to a limited ontogenetic range assuming
that the chosen parametric function reflected the entire (rarely observed) postnatal
development period. There is simply no substantive basis for such a practice. Most
of the growth studies shown in Table 9.1 that fitted growth functions have done this,
although some studies such as those of Bjorndal and Bolten^^ have long recognized
this as an inappropriate practice.
Secondly, another widespread practice (Table 9.1) has been to discard multiple
growth measurements per individual, thereby turning implicit mixed longitudinal
sampling into explicit mixed cross-sectional sampling (see Section 9.2.1.2; Table
9.1 footnotes; also Fitzhugh^^ for a discussion of data types). The usual approach
has been to use only data from first capture and last recapture and discard recapture
events (if any) in between. The same approach is prevalent in skeletochronological
work where growth marks between the earliest and the last are disregarded in growth
measurement for growth layer adjustment procedures. There is no apparent basis
for this practice other than perhaps a misunderstanding that repeated observations
on some individuals and not others in a data set is invalid. It is not, and in fact mixed
longitudinal sampling (sampling with partial replacement) has many desirable
properties — so why throw away useful growth data? Thirdly, a recurrent
although not widespread practice that is inappropriate in sea turtle growth studies
is the discarding of zero or negative growth measurements prior to analysis. We can
find no statistical basis for such a practice. Lastly, it must be borne in mind that
many studies shown in Table 1 were based on mark-recapture programs using
unreliable tag types and tagging protocols (see Sections 9.2.2.2 and 9.2.2.3 above).
9.3.1 A ge-B ased G rowth M odel
The 18 studies classified as age-focused in Table 9.1 used either known-age turtles
or estimated age using skeletochronology (see Section 9.2.3). All three sampling
designs (cross-sectional, longitudinal, and mixed longitudinal) were used, reflecting
the diversity of study settings and research purposes. If skeletochronology is coupled
with back calculation (see Section 9.2.3.3), then it is mixed longitudinal sampling,
else it is cross-sectional sampling, which confounds age and cohort effects.
9.3.1.1 Cross-Sectional Sampling
9.3.1.1.1 Known age

Ackerman^^^ studied embryonic growth in green turtles and loggerheads under a
quasi-experimental field situation. Eggs were relocated from several West Atlantic
beaches and incubated under beach conditions at the experimental sites. Eggs were
sampled sacrificially over time from the relocated clutches to determine embryo
growth (wet weight) for each stock and species. Sampling was cross-sectional,
assuming eggs were randomly sampled at each time both within and across the
clutches; otherwise there was additional uncontrolled error due to intraclutch corre
lation. Logistic growth functions fitted the species-specific weight-at-age growth
trajectories, suggesting a nonmonotone age- or size-specific growth rate function.

Growth of loggerhead and green turtle embryos slows prior to hatching, which is a
characteristic growth behavior known for producing precocial offspring (hatchlings)
capable of surviving without parental care.^°
9 J .L 1 .2 Estimated age (skeletochronology)
The skeletochronological techniques developed by Zug and colleagues^’^^’^^ '’'^ and
Klinger and Musick^^ "^^for indirect age estimation reflects an important development
in sea turtle growth studies (see Section 9.2.3 for details). The size-at-age growth
functions generated by Klinger and Musick^^ and Parham and Zug^^ for warm
temperate loggerhead stocks support age at maturity estimates derived from wild
stock mark-recapture s t u d i e s . B a s e d on skeletochronological and mark-recap
ture studies, mean age at sexual maturity for wild stock loggerheads seems to range
between 25 to 35 years of age, depending on the stock. The size-at-age growth
function generated by Zug et al.^^ for the Kemp’s ridley suggests later age at maturity
than estimates derived from captive reared^^ or headstarted mark-recapture studies.
Zug et al.^^ suggest that mean age at sexual maturity for wild Kemp’s ridleys ranges
between 11 and 16 years of age — a reanalysis of these data has suggested that 15
to 20 years of age may be a better interval estimate of mean age at sexual maturity.
Zug and Parham^^ have recently derived leatherback age estimates using skeleto
chronological analysis of sclerotic ossicles. These are the first substantive growth
records for wild stock leatherbacks. Bearing in mind the very small sample size,
Zug and Parham^^ suggest that mean age at sexual maturity in wild stock leatherbacks
is around 13 to 14 years of age.
A problem with most of these studies relates to the von Bertalanffy growth
function used.^^’^^"* The growth function was fitted in all cases to a limited age-size
range, yet was used to represent the entire postnatal development period. This is not
appropriate, especially if postnatal development is polyphasic.^'"’'’^^ Whether or not
the von Bertalanffy growth function is appropriate for sea turtle development is
questionable, and just about any function could fit such a narrow range of data —
this may explain why several parametric functions are often found to provide a
similar fit to the data. The von Bertalanffy function also has very poor statistical
properties (see Section 9.2.4.2).
Moreover, the von Bertalanffy function implies monotonie age- or size-specific
growth rates for which no empirical support was provided. Zug et al.^ used skele
tochronology to estimate age in loggerhead juveniles caught in the North Pacific
high seas driftnet fishery. A von Bertalanffy growth function was also used to
determine size-at-age and age-specific growth rates. Chaloupka^"* has reanalyzed
these data and found that growth was polyphasic and best fit using a nonlinear
regression comprising the summation of two logistic functions reflecting growth
spurts in the first year or so, and then again around 5 years of age — simply plotting
the data in Zug et al.^ will reveal that growth was polyphasic and would be poorly
fit by a von Bertalanffy function (see also growth records for pelagic loggerheads
in Bjomdal et al.).^^ Irrespective of the analysis and small sample size, this study^
makes a significant contribution to understanding the epipelagic phase for logger
heads (see also Bowen et al.^) and suggests that the pelagic phase for some stocks
may well span 10 or more years.
Growth mark validation, growth layer loss correction, and length back calcula
tion are major problems for all skeletochronological studies, and growth findings
need to be considered cautiously (see Section 9.2.3). , Zug et al.^^ employed an
intriguing form of back calculation coupled with the Fabens size-based analogue^^
of the von Bertalanffy growth function to generate an apparently age-free estimate
of age-based growth. Substitution of size-based growth parameters, say, from the
Fabens function into the conventional age-based von Bertalanffy function is invalid
(see Section 9.2.4.3).
Klinger and Musick,^^ Zug et al.,^^ and Parham and Zug^^ found significant
growth variability, which is attributable to natural variability in growth rates due to
age, year, and cohort effects. However, they also used growth record discarding so
sampling was explicit mixed cross-sectional confounding age and cohort effects.
The irony is that using back calculation properly (see Section 9.2.3.3) they would
have had mixed longitudinal sampling rather than cross sectional. Skeletochrono
logical age estimation coupled with length back calculation provides the best oppor
tunity for undertaking age, year, and cohort analysis of growth rates (see Sections
9.2.1.2, and 9.3.2.3.4), which could lead to new insights into sea turtle growth
dynamics.
9.3.1.2 Longitudinal Sampling
Many sea turtle growth studies have been based on known-age, captive-reared turtles
(Table 9.1). Such studies involve longitudinal sampling, which confounds age and
year effects on growth rates. This is not a major issue if the year effect is irrelevant,
say, in a controlled experiment to study dietary effects on g r o w t h . D u e to space
constraints we will not consider the large number of these sorts of captivity-based
growth experiments, but foreshadow that a critical review is wan'anted. A number
of compilations of captive-reared growth information are a v a i l a b l e . T h e main
insight is that all species develop much faster under various captive conditions than
is apparent in wild stocks. It is also apparent that growth variability of a study cohort
increases with age even under captive conditions, suggesting significant individual
variability in age to sexual maturity.
Several studies^^’^^’^^ fitted parametric growth functions to estimate growth rates
and mean age to maturity. Marquez^^ fitted a von Bertalanffy function to Kemp’s
ridley growth data, implying monotone decreasing size-specific growth rates. On
the other hand, Caillouet et al.^^ used a Gompertz model for Kemp’s ridley, implying
nonmonotone size-specific growth rates. The Gompertz function has excellent sta
tistical properties and the von Bertalanffy function has poor properties.^^ Recall that
Zug et al.^^ also fitted a von Bertalanffy function to estimated age at size data for
wild stock Kemp’s ridleys. All three studies^^’^^’^^ fitted functions to incomplete
ontogenetic ranges. Whether or not this species has nonmonotone size- or age-
specific growth is unresolved although Chaloupka and Zug^^^ have proposed a new
polyphasic age-specific growth model for the Kemp’s ridley based on skeletochro-
nological data derived from a cross-sectional sampling design. Moreover, Mar-

quéz’s^^ data included a mix of captive and mark-recapture growth records.
While longitudinal designs confound age and year effects, cohort effects are not
necessarily affected. For instance, Caillouet et al.^^ found strong cohort (year class)
effects in a study of Kemp’s ridley growth in captivity. It is noteworthy that the
mean growth function for an initial study cohort of about 20 olive ridleys^"^ shows
seasonal growth. Neither Phasuk and Rongmuangsart^"^ nor Mohanty-Hejmadi^^ fit
ted size-at-age functions to their olive ridley growth data, but it is clear from graphical
presentations in both studies that age-specific growth rates were nonmonotonic.
Uchida^^ and Frazer and Schwartz^^ fitted the logistic function to the growth
profiles of individual loggerheads, suggesting nonmonotone size-specific growth.
The logistic function in most parameter forms has excellent statistical properties.^^
Both studies fitted functions to incomplete ontogenetic ranges. The two loggerheads
reared under controlled captive conditions^^^ grew two to three times faster than the
two loggerheads reared under seminatural captive conditions.The difference was
attributed to experimental settings related to temperature and food treatm ents.The
difference in experimental setting reinforces the confounding of age and year effects
inherent in longitudinal studies. Under controlled experimental conditions'^^ the
confounded year effect will be masked. But in seminatural experiments spanning
nearly 14 years, like those of Frazer and Schwartz,^^ then the confounding of age
and year effects will not be masked if year effects occur. This might be the case in
Frazer and Schwartz^^ where the individual profiles display irregular monophasic
logistic growth behavior more suggestive of polyphasic logistic growth.
We caution against using captive-reared growth information for drawing con
clusions about growth dynamics of wild stock sea turtles. Longitudinal growth
studies confound age and year effects, irrespective of whether controlled experimen
tal or seminatural rearing conditions were used. Moreover, some captive study
designs are questionable variations of conventional sampling design. Of note is
Witzell,^^ who fitted a von Bertalanffy function to hawksbill size-at-age data and
inexplicably censored all data prior to 53 weeks of age. The sampling design was
also extremely complex, starting as cross sectional for the first 18 weeks, then at 32
weeks it became longitudinal when six selected turtles were followed to 84 weeks,
and then two selected turtles were followed until study end at 128 weeks.
9.3.1.3 Mixed Longitudinal Sampling
93.1.3,1 Known age (headstarted stock)

Although assigned to a mixed longitudinal group, both studies^^’^* used explicit
mixed cross-sectional sampling. Caillouet et al."^^ used growth record discarding,
while Ehrhardt and Witham^^ had insufficient recaptures per individual at the time
of the study. Therefore, age and cohort effects are confounded in both studies.
Caillouet et al.^^ fitted the von Bertalanffy function to Kemp’s ridley data, implying
monotonie size- or age-specific growth rate functions (cf. Caillouet et al.,^^ implying
nonmonotonic growth rates in captivity). Ehrhardt and Witham^* fitted the general
ized von Bertalanffy function to green turtle data, implying nonmonotonic size- or
age-specific growth rate functions. Both the standard and generalized von Bertalanffy
functions have poor statistical properties'^ (see Section 9.2.4.2) and, moreover, both
functions were fitted to limited data ranges.
Ehrhardt and Witham^^ made no mention of tagging nor provided standard errors
for fitted parameters, making it difficult to assess data reliability or model adequacy.
Irrespective of these concerns, the empirical growth pattern for the ontogenentic
range observed in their study was asymmetric, implying nonmonotone size- and
age-specific growth rate functions. Ehrhardt and Witham^* provided a time derivative
of the estimated size-at-age function to derive the age-specific growth rate function.
The age-specific growth rate function for this headstarted green turtle stock displayed
increasing growth from hatching to about 10 years of age, then slowly decreasing
to sexual maturity. Ehrhardt and Witham^* used the Chapman-Richards (generalized
von Bertalanffy) growth function, which is asymmetric about the inflexion rather
than skew-symmetric, such as the logistic. If an asymmetric function is needed, then
the Gompertz or an expected-value reparameterization of the Chapman-Richards
function^^ might be appropriate. The Weibull-type growth function used by Cha-
loupka and Limpus^^ (see Section 9.3.2.3.4) is also a flexible asymmetric parametric
function with good statistical properties.
The value of headstarted growth studies is questionable, since growth rates were
accelerated at a critical ontogenetic phase with unknown consequences, an issue
acknowledged by Caillouet et al.^^ Nonetheless, headstarted studies are more infor
mative than captive-reared studies and cannot be discounted until a wide range of
reliable wild stock studies have been earned out — which is not the case at present.
9.3.1.3.2 Known age (wild stock)
Since 1976 around 130,000 loggerhead and 109,000 green turtle hatchlings have
been marked with a mutilation tagging protocol in the sGBR r egi on. One immature
female loggerhead (15.2 years, 75.6 cm ccl), one immature female green turtle (8.75
years, 54.4 cm ccl), and one immature green turtle male (11.25 years, 68 cm ccl)
have now been recovered in a warm temperate foraging area well south of the sGBR
rookeries.^ If tag recognition is reliable, these data are invaluable, but insufficient
at this stage to be of major use for modeling growth dynamics. Other long-term
tagging and monitoring programs will undoubtedly also begin soon to provide known
age wild stock recoveries.
9.3.2 S ize-B ased G rowth M odel
The 21 studies classified as size-focused in Table 9.1 used either known age turtles
(captive-reared, headstarted) or turtles of unknown age. In most cases the purpose
of the study was to investigate growth rates, although many also had the explicit
intention of determining a size-at-age function from length data. All three sampling
designs (cross-sectional, longitudinal, mixed longitudinal) were used, reflecting the
diversity of study settings and research purposes. Only three studies (Bolten et al.,^^
Chaloupka and Limpus,^^ and Chaloupka and Limpus have taken a statistical
approach to better understand growth dynamics. Two studies (Chaloupka and
Linipus^^ and Chaloupka and Limpus'^)have attempted to take into account the
confounded time effects of time-sampled growth information.
252 The Bloiogy of Sea Turtles
9.3.2.1 Cross-Sectional Sampling
9.3.2. h i Length frequency

Although the main focus of the following studies was to estimate an age-size function
rather than growth rates, they are nonetheless size-based models. The sampling is
cross-sectional, so age and cohort time effects are confounded. Length frequency
analysis assumes that a length frequency distribution comprises a series of distinct
modes or pseudo-cohorts corresponding to specific age classes. The lengths in each
age class are assumed to have a parametric distribution around their mean length
(usually Gaussian) and be of equal variance. The mean lengths at age are also
assumed to lie along a parametric growth function — usually assumed to be a von
Bertalanffy growth function.
Le Toquin et al.^^ used single-sample length frequency analysis to estimate size-
at-age functions for an Indian Ocean immature green turtle stock. The sample was
too small, spanned an inadequate ontogenetic range, and the technique for mode
detection was inappropriate. Single-sample length frequency analysis is discussed
in Ricker.^^ While not assigned to this group, it is worth mentioning that Frazer^^
used the Fabens analogue^^ to convert a single-sample length frequency distribution
to derive an age frequency distribution (see Section 9.3.2.3.3).
Bjorndal et al.^^ used sequential sample length-frequency analysis^"^ to estimate
growth and age-size functions for an immature green turtle stock resident in the
southern Bahamas. They compared the estimated growth parameters with growth
interval and Fabens analogue growth parameters derived from a mark-recapture
program. Bjorndal et al.^^^ concluded from their length frequency analysis that it
takes the immature turtles at this site about 12 years to grow from recruitment (about
25 cm scl) to 70 cm scl when they disperse to other habitats.
The Bjorndal et al.^^ comparative study presents an interesting dilemma, since
the length frequency analysis was based on cross sectional sampling (age and cohort
confounded), whereas the growth interval and the Fabens function were based on
mixed longitudinal sampling of unknown age (year and cohort confounded). Based
on statistical sampling considerations alone it seems that the approaches are incom
parable. Moreover, without an independent and corroborative source of age, only a
weak form of inference was logically possible. Also, the study assumed that growth
for this stock was represented by a von Bertalanffy function, implying monotonie
size- and age-specific growth rates for the entire postnatal development phase. There
was no empirical basis to this assumption.
Sequential length frequency analysis using, say, MULTIFAN,^"^ is not just a
growth function; it is in fact a sophisticated modeling system. Therefore, it should
come as no surprise that it is one of the most assumption-laden parametric approaches
to growth modeling. Modeling length frequencies or age frequencies, for that matter,
without auxiliary biological information is a notoriously difficult task. Therefore,
length frequency methods need constraints imposed on the von Bertalanffy growth
function parameters; otherwise the model is not estimable.
Imposing biologically reasonable constraints is not a simple task and is quite
subjective.^"^ Yet the main advantage claimed for length frequency analysis is that it
does not need expensive mark-recapture programs. It is ironic then that the approach
depends on inclusion of demographic information to provide the model constraints

that can only be obtained by a mark-recapture program in the first place. Other more
recent length frequency modeling approaches based on artificial intelligence
algorithms'^ depend even more on incorporating reasonable biological constraints.
We reserve our judgment on the utility of the length frequency approach, irrespective
of the modeling approach used.
9.3.2.1.2 Parametric growth rate model
Bolten et al.^^ were the first to adopt a statistical modeling approach to analyzing
growth rate infomiation for sea turtles. They derived size-specific absolute growth
rates for 88 immature green turtles in a developmental habitat in the Bahamas. The
size-specific growth function was monotone decreasing. Sex was determined for
each turtle using radioimmunoassay. Using a linear regression model, Bolten et al.^^
found no sex-specific difference in the size-specific growth rates for this stock, at
least not for the range of sizes in the study (about 32 to 60 cm scl). The sampling
design was cross-sectional, confounding age and cohort time effects. The year and
cohort effects were also confounded because age was unknown. Age was unknown,
so year and cohort are also confounded. Since growth was the focus, not age, then
age/cohort confounding is not important. On the other hand, year and cohort effects
are important, especially since the site is a developmental habitat, which would be
subject to highly variable annual recruitment and growth (see Bjomdal et al.).^^
9.3.2.2 Longitudinal Sampling
Swingle et aU^^' provide weekly growth records for twelve conspecific captive-reared
loggerheads monitored for two years at several different sites. These data reflect
longitudinal sampling, hence age and year effects are confounded, making conclu
sions about wild stocks questionable. Age was known, but the focus was growth
rates rather than age. The weekly growth profiles enable investigation of potential
seasonal growth and perhaps polyphasic growth similar to the study by Phasuk and
Rongmuangsart^'^ on captive-reared olive ridleys discussed above (see Section
9.3.1.2).
9.3.2.3 Mixed Longitudinal Sampling
9.3.2.3.1 Known age (headstarted stock)

Age was known, but the authors^^'^^ '^^ focused on growth rather than age. Sample
sizes were also extremely small for two studies — 12 green and 2 hawksbill
recoveries by Kowarsky and Capelle"^^ and five recoveries from three loggerheads."^^
Monotonie size-specific growth was inferred by Bjorndal and Bolten"^^ for a logger-
head stock in the Bahamas, but there was no empirical basis for this view.
Since 1980 around 27,000 green turtle hatchlings and yearlings have been
marked with autografts and mutilation tags for hatchlings and single plastic or
titanium tags for yearlings, and released around the Cayman Islands.^^^ A number of
recaptures have occurred, providing information on headstarted growth rates, local
dispersal, and incidence of fibropapillomas. As already mentioned, the value of
headstarted growth studies is questionable.
9 3 .2 3 .2 Growth interval
Besides the captive-reared growth studies, this group reflects the bulk of sea turtle
growth studies. This is not surprising given that many studies were motivated by
the apparent disparity between growth information obtained from captive-reared
turtles and data obtained from wild stock observations.
Sampling design was mixed longitudinal (but age unknown) except for Green,
B o u l o n , and Collazo et al.,^^ where it was explicit mixed cross sectional due to
growth record discarding, thereby confounding age and cohort effects. As already
mentioned, growth record discarding is an unnecessary practice. For all studies the
year and cohort effects are confounded because age was unknown. Most studies
used multiple monel tagging protocols (see Section 9.2.2.2) except for Balazs et
al.^^ (inconel tags), Limpus^® (titanium tags), and Bjorndal et al.^^ (plastic tags).
Most studies only covered a limited ontogenetic range and provided small sample
sizes.
The basic method is the compilation of mark-recapture growth records over
time. Then mean annual absolute growth rates are calculated and grouped into 5-
or 10-cm-size increments referenced either to mean size or initial size at first capture
to indicate size-specific growth — the appropriate size reference metric being mean
size rather than initial size.^^>^^^ The approach is simple and has been the basis for
useful comparative analyses across a broad geographic r a n g e . S e v e r a l
studies^^’^^’^^’^-'’''^^ found nonmonotonic size-specific growth patterns for green turtles,
loggerheads, and hawksbills from several different geographic locations. Nonmono
tonic size-specific growth implies asymmetric (e.g., Gompertz type) or skew-sym
metric (e.g., logistic type) size-at-age functions. Three studies^^’^"^’^^ (but see also
Bjorndal et al.^^ and Boulon and Frazer^^ found monotonie size-specific growth for
green turtles, loggerheads, and hawksbills from western Atlantic-Caribbean loca
tions. Monotonie size-specific growth implies a noninflexion asymptotic size-at-age
function (e.g., von Bertalanffy type).
The growth interval approach is useful, but limited because the grouping arrange
ment precludes application of individual covariate information essential for any
further statistical modeling (cf. Limpus^^ and Chaloupka and Limpus^^% discussed
in Section 9.3.2.3.4 below).
9.3.2.3.3 Size-based growth analogues
Bjorndal and Bolten^^ used the Kaufman^^ size-based analogue of the von Bertalanffy
growth function to study specific growth in an immature green turtle stock in the
Bahamas. The main focus of this study was growth, not age. The sampling design
was mixed longitudinal, but as age was unknown, year and cohort effects are
confounded. They used the Kaufman^^ approach to determine if the size-specific
growth data implied either logistic, von Bertalanffy, or Gompertz type size-at-age
growth — growth was specific, not absolute growth. The Kaufman approach pre
cluded the need to specify unknown growth parameters such as mean adult size.
They found that von Bertalanffy growth behavior best reflected the size-specific
growth pattern for the recorded size range. The empirical size-specific growth func
tion was monotone decreasing for the recorded size range (25 to 70 cm scl). But
the Kaufman analogue is inappropriate for two reasons — (1) it isn’t applicable for
estimating von Bertalanffy parameters anyway (see Kaufman)^^ and (2) replacing
parameters in age-based functions with parameters from size-based analogues is
invalid (see Section 9.2.43), although this latter use was intentionally avoided by
Bjorndal and Bolton.
The Bjorndal and Bolten^^ study has been extended by Bjorndal et al.^^ in a
comparison of sequential length frequency analysis, growth interval method, and
Fabens analogue of the von Bertalanffy growth function (see also Section 9.3.2.1.1).
This time the main focus was age, but the method was still size based. They used
the Fabens analogue to derive a size-at-age function fitted only to the observed size
range. The Fabens size based analogue has known poor statistical properties'"^ and,
moreover, using any size-based analogue to provide age-based parameter estimates
for the von Bertalanffy growth function is invalid (Francis^^ and references therein;
see Section 9.2.4.3). Nonetheless, it appears the empirical size-specific growth
function for this 25 to 70 cm scl green turtle stock^^*^^ does display monotone
decreasing size-specific growth rates unlike, say, for green turtle stocks in the sGBR
region^^’2^‘^ (see Section 9.3.2.3.4).
The main focus of Frazer et al.,"^^ Frazer and Ehrhart,^^ and Boulon and Frazer^^
and was on deriving size-at-age functions for green turtles and loggerheads from
mark-recapture length data. In fact, the growth function derived by Frazer et al."^^
was used to form the age-structure of a loggerhead population modeF"^ (see Section
9.5.2.2). Recall also that Zug et al.^^ used the Fabens analogue to derive a size-at-
age function for age-estimated Kemp’s ridley in a mixed cross-sectional study (see
Section 9.3.1.1.2).
The sampling design for these four studies should be mixed longitudinal, but
because of growth record discarding it was explicit mixed cross-sectional with age
and cohort effects confounded. All four studies fitted von Bertalanffy type growth
functions to a limited ontogenetic range, assuming that the function reflected the
entire unobserved postnatal development period. This is simply not appropriate. All
four studies used the Fabens size-based analogue approach to derive age-based
growth parameters to substitute into the von Bertalanffy age-based growth function.
As already mentioned this approach is invalid (Francis^^ and references therein; see
Section 9.2.4.3). Frazer and Ehrhart’s study^^ was extended by Frazer^^ using the
Fabens analogue to back transform a length frequency distribution for stranded
loggerheads to derive an age frequency distribution (catch curve) for estimating age-
specific survival. The same issues remained unresolved. We discussed elsewhere
(see Section 9.2.4.3) suitable approaches to size-based growth studies and size-based
analogues.
All seven studies discussed concluded that somatic growth
was best expressed as a von Bertalanffy growth function, implying monotonie size-
and age-specific growth rate patterns. Excluding Bjorndal and Bolten,^'^ the remain
ing six studies"^^’^^’^^’^^’^^'^^ concluded that the von Bertalanffy function reflected
growth for the entire postnatal development phase, which was highly questionable
since no study had sufficient data range, and these six studies used inappropriate
size-based analogues to derive such a conclusion.
93.2.3A Nonparametric growth rate models

Chaloupka and Limpus^^ developed size-, sex-, and age-specific somatic growth rate
models for the sGBR green turtle stock using a large data set derived from a long
term mark-recapture program^^ — see also Limpus and Chaloupka.^^"' Tagging used
the multiple titanium tagging protocol and the sampling design was mixed longitu
dinal (>1 measurements per individual) with data spanning the entire post-pelagic
developmental phase.
They modeled absolute growth rate as a function of informative covariates such
as size (ccl), sex, and recapture interval using nonparametric regression modeling
known as generalized additive modeling.They found nonmonotone size-specific
growth rates that increased from recruitment size (>35 cm) to maximum growth rate
around 60 to 65 cm ccl and then decreasing slowly to about 100 cm ccl. Female
growth was significantly different from male growth, diverging at about 60 cm ccl.
They found significant annual variation in female growth rates.
Using numerical integration of B-splines they derived size to time-at-large
growth functions for each sex from the size-specific growth functions. Their empir
ical solution models suggest that both females and males in the sGBR green turtle
stock must be >30 to 35 years-at-large before reaching sexual maturity. Time-specific
growth functions for each sex were also provided, suggesting maximum growth for
females occurs at 10 years-at-large (60 cm ccl) and at 12 years-at-large (63 cm ccl)
for males. The size to time-at-large functions were modeled using a system of
nonlinear equations to provide analytical solutions to the mean size to time-at-large
growth curve for each sex. These Weibull-type functions have good statistical
properties'^ and fitted the numerically integrated growth data well.
A generalized additive modeling approach was also used by Chaloupka and
Limpus^^"* to analyze absolute growth rates for a sGBR hawksbill stock.^^^ Sampling
was mixed longitudinal (>1 measurements per individual) with data spanning the
entire post-pelagic developmental phase. They found a nonmonotone size-specific
growth function and distinct sex-dependent growth rates.
The nonparametric statistical modeling approach developed by Chaloupka and
Limpus^^ has many useful features including (1) the data determine the underlying
growth function rather than using a parametric function that bears little resemblance
to the data, and (2) the growth data are conditioned on informative covariates to
support a better understanding of the growth process itself. There is a price, however
— it is more complex. Moreover, year and cohort effects were confounded in both
this study, Limpus and Chaloupka.^^"* and Chaloupka and Limpus^^'' because age was
unknown despite the sampling designs being mixed longitudinal.
9.4 C O M P A R A TIV E LIFE H IS T O R Y PATTERN S

9.4.1 C omparative Life H istory T raits
There are three common schemes for interpreting life history strategies: (1) deter
ministic r-K theory, (2) stochastic bet-hedging theory, or (3) age-specific reproduc
tion axioms focusing on residual reproductive value (see S t e a ms ) . T h e schemes
are all special cases of a more general demographic classification of environmental
setting or habitat. The study of growth dynamics focuses on demographic implica

tions of ontogeny, while life history theories focus on demographic implications of
phytogeny. The demographic implications of the interaction between growth, life
history, and environment forms the basis for population dynamics and macroevolu-
tionary change.
The r-K theory assumes density independence and environmental invariance and
has been co-opted to support various linear deterministic growth, population, and
macroevolutionary theories. However, variability in timing of maturation and
growth^^^ and environmental stochasticity^®^ seem more likely to be the keys to
unlocking complex life history patterns. Nevertheless, the most common framework
apparent in sea turtle research is the r-K theory. For instance, r-K theory is implicit
in all deterministic stable and stationary age-type models shown in Table 9.2. In
fact, it seems explicit in the life table s c h e d u l e s a n d the matrix projection
models^"^’^^^^'^^^ where the presumption is for a stable environment focusing on adult
selective pressures (see Steams). Interestingly, there are few tests^^^'' of life history
theories using sea turtle growth, population, or macroevolutionary patterns.
Two approaches that could be adopted for such a study are: (1) a prospective
approach based on stochastic simulations (e.g., see Section 9.5.2.1; Chaloupka and
Limpus)^^^ to test demographic forecasts of competing life history theories on the
temporal behavior of a hypothetical population, and (2) a retrospective approach
based on comparative analysis of life history patterns recorded from various envi
ronmental settings. The two approaches are not necessarily mutually exclusive. The
prospective approach has not been used in sea turtle research, although a limited
comparative test of bet-hedging theory was undertaken recently by Cunnington and
B r o o k s u s i n g detemiinistic matrix projection models (see Section 9.5.1.3). Mean
while, important qualitative progress has been made using the second approach of
stochastic comparative analysis of static life history traits.
Van Buskirk and Crowder^ produced a compendium of selected life history
traits for 96 stocks of the seven extant sea turtle species, enabling analysis of both
intergeneric and intraspecific covariation. They used these data for an analysis of
intergeneric life history covariation based on a combination of correlation and linear
regression-derived residuals to control for size. Given the nonorthogonal hierarchical
nature of such data sets, it might be worthwhile in future to consider the use of
classification t r e e s o r visualization tools such as coplots^^ for exploring the rela
tionship between size, phylogeny, and life history patterns. Despite the limited and
incomplete data set and limited methodology, they nonetheless were able to draw
several broad qualitative conclusions. Perhaps the most important conclusion disre
garding adult body size was that the seven extant species of sea turtles could be
readily classified into two distinct groups based on mean fecundity and neonate
characteristics — group 1 comprised leatherbacks and flatbacks with smaller clutch
size, larger egg volume, and larger hatchling size, while group 2 comprised greens,
loggerheads, hawksbills, and both ridleys with larger clutch size, smaller egg volume,
and smaller hatchling size. It may be no coincidence that both group 1 species
(leatherbacks, flatbacks) also have similar adult gait, egg-laying posture, and adult
carapacial skin.^*^
K5
05
TABLE 9.2
Classification Summary of Sea Turtle Population Dynamics Literature
Examples
/-state Mode Time Focus Ref. (year; no.) Species Modeling approach Location Purpose
Distribution Deterministic Static Age Marquez et al. (1982a; 124) olive Ridley Sequential population analysis Eastern PacificDemographicanalysis
Marquez et al. (1982b; 125) Kemp’s Ridley Sequential population analysis Gulf of MexicoDemographicanalysis
Richardson & Richardson Loggerhead Life table simulation Georgia, U.S. Demographic analysis
(1982; 126)
Frazer (1983, 1984, Loggerhead Life table schedules Georgia, U.S. Demographic analysis
1986; 103-105)
Dynamic Age Doi et al. (1992; 127) Hawksbill General dynamic pool Cuba Policy evaluation
Crowder et al. (1994; 108) Loggerhead BLL matrix projection Southeastern U.S. Policy evaluation
______________ Heppell et al. (1996;24) Loggerhead BLL matrix projection Great Barrier Reef Policy evaluation
Stage Crouse et al. (1987; 106) Loggerhead Lefkovitch matrix projection Southeastern U.S. Policy evaluation
Laurent et al. (1992; 107) Loggerhead Lefkovitch matrix projection Mediterranean Demographic analysis
Crowder et al. (1994; 108) Loggerhead Lefkovitch matrix projection Southeastern U.S. Policy evaluation
Somers (1994; 109) Loggerhead Lefkovitch matrix projection Great Barrier Reef Policy evaluation
Heppell et al. (1996;24) Loggerhead Lefkovitch matrix projection Great Barrier Reef Demographic analysis
Stochastic Dynamic Hybrid Chaloupka & Limpus Green Nonlinear system simulation Great Barrier Reef Policy design/evaluation
IT
(1996;110) CD
CO
Chaloupka & Limpus Loggerhead Nonlinear system simulation Great Barrier Reef Policy evaluation
(1996; 123a) o
Configuration Stochastic Static Age Bustard and Tognetti Green Monte Carlo simulation Great Barrier Reef Demographic analysis
(1969; 137) Ln
CD
Oi
While not stated in van Buskirk and Crowder,^these two groups represent
distinct ontogenetic patterns. Group 1 species live in one general habitat irrespective
of ontogeny — flatbacks remain neritic with no pelagic phased* while leatherbacks
remain pelagic except, of course, when nesting. On the other hand, group 2 species
have a distinct postnatal pelagic phase proceeded years later by recruitment to a
neritic immature adult phase. The transition from pelagic to neritic habitat represents
a major ontogenetic shift and is one of the many reasons why a single growth function
is inappropriate for representing postnatal growth of at least group 2 species, where
growth is clearly polyphasic^^’^^^ (see Section 9.3.1.1.2).
Another possibility regarding these two ontogenetic groups is that they might
well reflect two distinct phyletic groups based on paedomorphic heterochrony (see
Alberch et al.).*®^ Rhodin et al.^^"^ consider that the leatherback skeletal morphology
reflects neoteny, which is pedomorphism due to somatic retardation.N eoteny
results from an ontogeny with a longer exposure to high juvenile growth rates,
leading to a morphologically retarded adult. It is interesting then that leatherbacks
appear to have by far the highest juvenile growth rates of all sea turtle species,
reaching sexual maturity on average in around 13 to 14 years.W hether or not
flatbacks are a macroevolutionary product of neotenic heterochrony like the leath
erback is not known, although flatbacks do retain a number of juvenile features in
adulthood, such as carnivory and poorly keratinized skin and scales.^^’^^^ If they are
neotenic, then we would expect flatbacks to also have higher growth rates, which is
feasible given that they are restricted to a continental shelf habitat with feeding
grounds in shallow, turbid, inshore w a t e r s . I f higher growth rates occur, then
perhaps earlier maturation rates offset the impact of lower flatback fecundity^ on
flatback population dynamics — else lower fecundity could be offset by higher
hatchling and juvenile survival rates, but we just do not know. And therein lies a
consequence of the complex nature of sea turtle life history, growth, and population
dynamics — a litany of unresolved “if-then-else-but” statements.
There are insufficient comparative studies of sea turtle ecology, reproduction,
embryology, morphology, and somatic growth to support further insight into life
history patterns. Comprehensive studies of growth dynamics and population dynam
ics across a range of environmental settings are needed to understand the complex
relationship between sea turtle ontogeny, phytogeny, and macroevolutionary change.
9.4,2 C omparative D emography
Most sea turtles have an overlapping iteroparous life cycle — long life expectancy
coupled with discrete multiple breeding seasons and overlapping generations. The
nesting season is usually in summer, but a northern Australian flatback stock (N.
depressus) has a winter nesting season.^Several stocks have also been shown to
nest year-round and so have a continuous iteroparous life cycle (e.g., Sarawak green
turtles,Crab Island flatbacks).^Such complexity of life cycles and geographic
variability provides a rich anea for comparative demographic studies.
Comparison of limited growth interval profiles for some stocks indicated inter
regional differences in size-specific growth^^^’^^’^^ and mean adult size^^^ for green
turtles and loggerheads. A difference in benthic phase recruitment size is also
apparent between the Caribbean and the sGBR regions for green turtles, loggerheads,
and hawksbills. In the Caribbean region, greens recruit at 20 to 25 cm scl, logger-
heads at 40 to 45 cm scl, and hawksbills at 15 to 25 cm sclT On the other hand, in
the sGBR region greens recruit at 35 to 40 cm ccl, loggerheads at 68 cm ccl, and
hawksbills at 40 cm ccl.^’^^'^^ Even accounting for tranformation of scl to ccl the
recruitment size differences between regions and between loggerheads and the other
species are considerable. Clearly, there are common and yet local or regional patterns
in sea turtle demography, suggesting that there is sufficient phenotypic variability
to warrant comparative demographic analyses to complement life history analyses.
Zug et al.^^ summarized captive and wild stock growth rates and estimated age
at sexual maturity from many sources, but these data are of limited value for
comparative purposes. Important static s u r v i v a l a n d fecundityschedules have
been derived for a loggerhead nesting stock (Little Cumberland Island, GA). Com
parison of these life schedules with similar schedules derived for other loggerhead
stocks would fomi the basis for a comparative demography, but there seem to be
few age-specific or size-specific survival or fecundity schedules available.
BjomdaE^ presented both cohort and relative age-specific survivorship data for
a green turtle stock nesting at Tortuguero, Costa Rica. Chaloupka^^^'* has used a
statistical model to account for the main effects of age, year, and cohort in a
reanalysis of BjorndaLs^^ data. This modeling approach based on a generalized linear
model with extra-Poisson sampling variation (see McCullagh and Nelder)^^^ provides
an age-specific hazard function conditioned on the time-dependent effects (age, year,
cohort). This time-conditioned hazard function would be suitable for inter-stock
comparison if other such hazard functions were avaliable.
Despite these studies there are no comparative analyses of intra- and interspecific
demography based on key attributes such as age- or size-specific mortality (hazard),
fecundity, somatic growth, or recruitment. It is certainly more difficult to compare
demographic functions than static life history traits based on, say, mean fecundity
or relative reproductive value. Nonetheless, important insights into demography and
population dynamics will arise when such analyses are undertaken. To encourage
such an analysis, which would need to be a collaborative research effort, we outline
a standard modeling procedure that might be suitable for comparative demographic
analysis based on somatic growth dynamics and mortality. Firstly, a statistical anal
ysis of age-specific growth if age known or else size-specific growth rates condi
tioned on time-dependent effects (year, cohort) is needed to estimate the growth
function. If the estimated growth function is size specific, then numerical integration
is needed to transform the function into age-specific form.^^’^^^ Then a suitable
parametric growth function could be fitted to the conditioned size-at-age data derived
from either the age-specific or size-specific forms. The Weibull-type nonlinear equa
tion is useful here for comparative puiposes, especially when used as a system of
nonlinear equations (see Chaloupka and Limpus).^^ The single equation fonn is^^
y = a -- P exp[-exp(-y)fl (9.1)
where: y = size (length, weight)

t = time (age or year-at-large)
a = asymptote
p = asymptote - y intercept
Y= growth coefficient
5 = dimensionless shape parameter
If the shape parameter 5 = 1 , then the growth function is an asymptotic regression
function indicating monotone decreasing size-specific growth rates — a von
Bertalanffy type function. The shape parameter enables the equation to fit either
skew-symmetric or asymmetric functions. This model is the only flexible shape
equation with known good statistical properties, and expected-value forms are
avai l abl e. It is easy to derive a size-based analogue of this function, but its
utility is questionable since it is our claim that all time-dependent demographic
data needs to account for the age, year, and cohort effects first.
The statistical estimation Equation (9.1) or a system of equations (9.1) will
include an error term, which can be autoregressive or moving average to account
for the nth order autocorrelation. In a system of equations the parameters can be
either unique to each equation or common across equations, as can the autocorre
lation parameters. The system of nonlinear equations is then best estimated using
nonlinear SUR.^^^
The Weibull-type model would then provide estimates of growth dynamics from
several stocks expressed in summary form as (1) mean mature adult size (a), (2)
growth coefficient (y), and (3) the intrinsic growth function shape parameter (5).
These three parameters coupled with estimated mean age at sexual maturity would
then provide a quantitative basis for inter-stock demographic comparison. These
four key parameters could be derived also as sex-specific functions using the same
process. The Weibull-type modeling procedure presented could also be applied to
survivorship data for inter-stock comparative analysis. The advantage of using one
reliable growth function such as Equation (9.1) for comparing all stocks (sex,
geographic, species) is that it would enable comparison of growth functions among
the stocks in terms of the same growth parameters and estimates of standard errors.
Such a form of comparative demography based on growth and survivorship
would complement well conventional life history analysis such as that of van Buskirk
and Crowder. ^^^The problem with just analysis of life history traits is that phenotypic
variability of traits such as reproductive output is only epigenetic and of low heri-
tability, whereas somatic growth rates, while environmentally influenced, have high
heritability (see Stearns).M oreover, growth dynamics is a fundamental aspect of
macroevolutionary change.Combining comparative demography and comparative
life history would provide better insights into population dynamics. Major challenges
exist in undertaking the modeling procedure related to lack of sufficient time-
structured growth and mortality data sets spanning the entire post-pelagic develop
mental phase.
9.5 P O P U LA T IO N D Y N A M IC S
A population is a set of individuals with a common life cycle, which is a sequence
of morphological states structured by age and/or developmental stages. Population
dynamics is about the analysis of temporal change in abundance and spatial structure
to explore how a population responds to natural and human-induced perturbations.

The abundance of individuals over time in each state is determined by five demo
graphic processes or rates (birth, death, growth, immigration, emigration) subject to
varying levels of environmental, demographic, and genetic stochasticity. The chal
lenge for population dynamics is to integrate existing knowledge about specific life
cycles with the demographic processes. Usually this involves mathematical modeling
of the spatiotemporal behavior of the states as some function of the demographic
rates — the rates may be time varying. Population dynamic models are useful for
helping to provide insight into the complex mechanisms affecting temporal and
spatial change in population abundance on a scale not amenable to experimental
control or observation.
wSea turtle demographic analysis and population modeling are very much in their
infancy. All known sea turtle population models have been anânged in Table 9.2
into one of five groups according to the following factors — (1) model type as either
/-state distribution or /-state configuration (see DeAngelis et al.),^^® (2) whether
model was deterministic or stochastic (see Getz and H a i g h t ) , (3) whether model
was static (steady state) or dynamic, and (4) whether the demographic structure was
age or stage classified (see Manl y, McDonal d and CaswelP^^). Table 9.2 includes
the citation, species, specific modeling approach, location, and model purpose. The
distinction between /-state distribution and configuration models was included to
place sea turtle models in a broader modeling context, since so-called individual-
based models (IBMs, /-state configuration) are emerging in other fields.
The distinction between static and dynamic is a bit arbitrary, since only the
difference equation models of Chaloupka and Limpus^^’^^^'*are really dynamic where
the demographic processes are time varying and involve feedback mechanisms.
However, many aut hor sconsi der models to be dynamic when a model has a
potential capacity for modeling so-called transient and time-dependent behavior.
This is a rather liberal definition of dynamic, but is adopted here for consistency.
Interestingly, all the deterministic models in Table 9.2 (sequential population, life
table, dynamic pool, matrix projections) are related to a generalized matrix modeT^^
which itself can be interpreted within a more general difference equation modeling
framework.
9.5.1 D eterministic C ohort M odels
All the deterministic models in Table 9.2^4 11)9 ,124-127 ^^sed on a deterministic
and linear theory of population growth where demographic rates are assumed to be
density independent and stationary (time invariant), resulting in a stable age distri
but i on. I n other words, demographic rates are assumed to be time independent,
thereby confounding age, year, and cohort effects, one implication being that indi
viduals are assumed to develop at the same rate and so reach sexual maturity at the
same age. These assumptions are expressed as knife-edge recruitment in the Marquez
et models.
9.5.1.1 Static Age-Structured Models
All models assigned here (Marquez et Richardson and Richardson,

F r a z e r ) a r e some form of age-structured life table model (a life table could of
course be size- or life-state structured).
Marquez et al.^24,125 developed population models for hypothetical populations
of olive and Kemp’s ridleys based on demographic information related to stocks
nesting along the Mexican coast. Both models are based on a variant of sequential
population analysis (SPA) common in fisheries stock assessment (see Megrey and
Wespestad).’^^ SPA is a detemiinistic method based on many assumptions, including
constant annual recruitment independent of the spawning stock and both a growth
function and constant hazard function that are age and year invariant. The demo
graphic information available at the time the model was developed was limited. They
determined female mortality rates from a mark-recapture program and assumed both
species reached sexual maturity between 7 to 8 years of age. Zug et al.^^^ suggested
that mean age at sexual maturity for wild Kemp’s ridley ranges between 11 to 16
years of age while Chaloupka and Zug'’^'* consider that 15 to 10 years of age might
be a better interval estimate. Marquez et used their SPA variant to derive
hindcast age-cohort stock estimates. SPA is a derivative of virtual population (VPA)
or cohort analysis. ^^8 same assumptions as VPA except that the
hazard function implicit in Marquez et comprises natural mortality, but no
fishing-related mortality. While not immediately obvious, SPA is also closely related
to age-structured matrix models.^“’
Richardson and Richardsondeveloped a simulation model for a hypothetical
cohort to reflect the loggerhead nesting stock dynamics at Little Cumberland Island,
GA. The model is best described as a life table with a stock-recruitment function
based on the product of relative age-specific fecundity and mortality schedules. It
incoiporated stationary 5- or 10-year cyclical patterns in fecundity to simulate
cyclical patterns in stock abundance reflective of the observed nesting stock at Little
Cumberland Island. F r a z e r d e r i v e d static life table schedules for the same
loggerhead stock, which were major refinements of the survival and fecundity
schedules used in Richardson and Richardson. The F r a z e r l i f e table schedules
were the first schedules produced for sea turtles and are still the only substantive
static life table models available. These age-specific survival and fecundity schedules
provide information useful for comparative demographic analysis. Both the SPA
models^^'^’*^^ and the static life table schedules^®^"^^^ are very useful for parameterizing
age-structured simulation models discussed in the following sections.
9.5.1.2 Dynamic Age-Structured Models
All models assigned here (Heppell et al.,“"^ Crowder et al.,^°^ Doi et al.)^^^ are some
form of deterministic age-structured matrix model. Doi et al.^^^ used a general form
of dynamic pool or so-called yield-per-recruit stock assessment m o d e l . T h e
dynamic pool model used by Doi et al.*^^ is also a form of generalized Bemardelli-
Lewis-Leslie (BLL) m a t r i x . T h e BLL model is a common mathematical descrip
tion of density-independent growth of an age-structured population observed at
discrete time i nt er val s . The BLL matrix models used by Crowder et al.^^^ and
Heppell et al.^"^ were complements of stage-structured loggerhead models, so we
leave these two studies until the next section.
Doi et al.^^^ developed an extended dynamic pool model to estimate optimal
harvesting yield given hypothetical levels of fishing intensity for a hawksbill fishery
based in Cuba. The extensions included an age-structured cohort, gradual adult
maturation, a piece-wise linear age-specific harvesting risk function and a piece-
wise linear Beverton-Holt-type stock-recruitment function. Fishing mortality and
natural mortality (hazard) were assumed to be continuous and separable functions.
The model makes other major assumptions — (1) growth is well described by a von
Bertalanffy function; (2) a year- and cohort-invariant von Bertalanffy age-length key
to back transform catch-at-length to catch-at-age, since age is unknown; (3) constant
time-invariant recruitment; (4) a constant time-invariant hazard function; and (5) a
constant time-invariant Baranov-type catch function. It is a sophisticated, but
assumption-laden model that is very common in fisheries stock assessment. It incor
porated SPA (see Section 9.5.1.1) to review transient behavior induced by changes
to model parameter inputs. Doi et al.^^^ concluded that the escapement size for
hawksbills in the Cuban fishery would need to be increased to 70 cm scl (12 years
old) to ensure the long-term viability of this stock.
The major concern with this model is that it was based on very little biological
data about the hawksbill stock itself. Doi et al.^^^ relied on captive-reared growth
information (see concerns discussed in Section 9.3.1.2 on captive studies) or theo
retical inputs such as a presumed maximum exposure of immature hawksbills to the
fishery no later than 7 years of age, complete sexual maturation by 14 years of age,
and negligible dispersal to or from the Cuban stock. There were no valid estimates
of age- or size-specific natural mortality, let alone a hazard function conditioned on
age, year, and cohort effects. Since stock abundance in foraging or nesting habitats
was either unknown or poorly known, the model had no empirical stock reference
behavior to assess the heuristic value of the model. Sullivan et al.^"*^ present a general
stochastic size-structured model perhaps more suitable than the extended dynamic
pool model for analysis of sea turtle stock assessment when age is unknown and
the analysis is based on the catch-at-length composition. However, any model will
be of limited value if there are no valid stock-specific demographic data.
9.5.1.3 Dynamic Stage-Structured Models
The models assigned here (Crouse et al.,^^*" Laurent et al.,^®^ Crowder et al.,^^^
Somers,*®^ Heppell et al.)^"^ are all deterministic stage-structured variants of a general
linear matrix model, which includes the BLL matrix model introduced in Section
9.5.1.2 as a special case.^î Pnear stage-structured matrix model (Lefkovitch
model) is widely used in ecological m o d e l i n g . N o n l i n e a r extensions*^^ and
linear stochastic analogues of the Lefkovitch model have been proposed.
The main purpose of all these models on loggerheads^"^’*®^'^^"^ was to estimate
stage-specific reproductive value and demographic parameter sensitivity, but increas
ingly they are being also used to simulate temporal stock abundance in response to
demographic perturbations.
Crouse et developed the first comprehensive stage-structured matrix model

for a sea turtle although similar projection models for green sea turtles pre-existedd^^^^
The model was based on life table schedules developed previously for the Little
Cumberland Island loggerhead stockd®^'^^^ Stage durations were estimated from
those schedules. The model comprised seven size-based stages assuming sexual
maturity at 22 years and a 54-year life expectancy. More recent age estimates"^^*^^
suggest that mean age to sexual maturity could range from 25 to 35 years of age or
older. Crouse et al.^®^ concluded from sensitivity analyses using elasticity measures
that reducing annual mortality of large juveniles (stage 3) was most important to
ensure long-term viability of the stock. Stage 3 individuals were most at risk to
drowning in the shrimp fishery coincident with the modeled stock. There was no
empirical stock reference behavior to assess the heuristic value of this model.
Crowder et al.^^^ extended the original Crouse et al.^®^ model by simplifying the
stage structure and increasing small juvenile (stage 2) mortality based on revised
data. Stage durations were derived from a von Bertalanffy growth function presented
in Frazer and Ehrhart.^* The purpose of the study was to evaluate the potential impact
of shrimp trawling on loggerhead mortality. Like Crouse et al.,^®^ they concluded
from a sensitivity analysis (elasticity) of the stage-structured model that reducing
annual mortality of large juveniles (stage 3) was essential for stock viability. Crowder
et al.^^^^ used the stage-structured model to forecast stage-specific loggerhead stock
recovery rates following introduction of turtle excluder devices (TEDs) in the shrimp
fishery operating in offshore waters. There was no empirical stock reference behavior
to assess the heuristic value of this model.
Crowder et al.'^^ then used a BLL matrix form of their model with 54 age classes
to simulate the expected time-dependent (transient) response in nesting female
numbers following TED introduction. The BLL model was needed, since the five
stage-class model assumed stable age within stages and would display anomalous
responses to time-varying perturbations, which was their intention by imposing a
step-function mortality change at specific age and time instances. Recall that the
deterministic stage-structured models in Table 9.2 assume indistinguishable individ
uals in a stage of which a fixed proportion will always progress to the next stage
even if no new individuals are entering the current stage due to, say, previous 100%
egg mortality. If this model was used to simulate temporal stock behavior, then
anomalous abundance patterns would occur since there is no individual develop
mental period within a stage — it is a design defect in some stage-structured models
where stages are defined on age criteria rather than developmental states. Unless
distributed delay maturation for each stage is incoiporated (see Section 9.5.2.1) all
the deterministic stage-structured models cannot in their current form cope with
time varying changes.
The BLL matrix approach was used by Crowder et al.^^^ to get around this
problem. But it then introduces other problems associated with assuming all demo
graphic responses are only age dependent and that all responses occur only at fixed
ages within a fixed 54-year life expectancy — quite a limiting set of assumptions
which was the rationale for using stage-structured models in the first place. There
was no empirical stock reference behavior to assess the heuristic value of this model.
Nonetheless, this is a particularly useful simulation because it provided for the first
time important qualitative insights into the long time delays inherent in loggerhead
population recovery following TED introduction. Crowder et al.^^^ used the Crowder
et stage-structured model to forecast population recovery rates using revised
estimates of TED-reduced mortality derived from their recent analysis^"** of logger-
head strandings.
Laurent et al.^^^ developed a stage-structure model for a Mediterranean logger-
head stock using six stages (4 size-based stages and 2 reproductive status stages).
Stage transition (maturation rate) was knife-edge and time-invariant as for similar
Laurent et al.^®^ concluded from a sensitivity analysis (elastic
ity) that reducing annual mortality of loggerheads larger than 70 cm scl (large
juveniles and adults) was essential for stock viability. The model was based on little
empirical demographic information relevant to the stock. There was no empirical
stock reference behavior to assess the heuristic value of this model.
Som ersdeveloped a stage-structured model for the sGBR loggerhead stock.
This model was similar to Crouse et al.,^^^ the main difference being a modified
size-based classification and mainly theoretical demographic parameters reflecting
the stock in a hypothetical reefal setting. Unlike Crouse et al.,'^^ Somersconcluded
that protection of eggs/hatchlings would have a major impact on long-tenn stock
viability. The reason for the difference in conclusion between the two studies was
due to higher egg/hatchling stage mortality rates used in Somers. There was no
empirical stock reference behavior to assess the heuristic value of this model.
Heppell et al.^"^ also developed a stage-structured model for the sGBR loggerhead
stock. This model was similar to Crouse et al.,‘^^ the main difference being a stage
classification for immature and adult loggerheads based on reproductive condition
determined from laparoscopy. The demographic parameters were relevant to a stock
in a hypothetical reefal setting and to a sGBR stock nesting at the adjacent mainland
rookery (Mon Repos). This model used stage-specific survival rates for immature
and mature loggerheads derived from Cormack-Jolly-Seber a n a l y s i s o f capture
profiles derived from a mark-recapture program.^ The survivorship analysis assumed
time-invariant mortality confounding year and cohort effects. Stage durations were
derived from the von Bertalanffy growth function presented in Frazer et al.^^^
Heppell et al.^"^ concluded from sensitivity analysis (elasticity) that reducing
annual mortality of loggerheads larger than about 70 cm ccl was essential for stock
viability. Like Crowder et al.,’^^^ Heppell et al.^“^ then used a BLL matrix form of
their model with 35 age classes to simulate time-dependent responses in adult female
abundance in reply to (1) levels of trawling-related mortality or (2) previous fox
predation of eggs at the mainland rookery. The two mortality sources were not
simulated simultaneously. There were no empirical stock reference behaviors to
assess the heuristic value of these models.
All the stage-structured models discussed in this section assumed time-invariant
and density-independent demographic rates. Therefore, age, year, and cohort effects
are confounded. Individuals in a stage were also assumed indistinguishable with
respect to growth development and time spent in that stage. In other words, all
individuals develop at the same rate and progress to the next stage with the same
maturation rate — similar to the constant knife-edge recruitment assumption of SPA
(Section 9.5.1.1). Many sea turtle growth have shown sig

nificant individual variability, suggesting that including variable development rates
in future models might be worthwhile.
Four of the stage-structured studies^"^’^®^'^®^ assigned to this section used sensi
tivity analyses based on elasticity measures. Elasticity does not recognize covariation
and assumes that each trait is independent of all other traits. Yet van Buskirk and
Crowder,în a review of life history traits, found correlation between reproductive
traits. Covariation in demographic rates might well explain why several studies found
stage-specific parameter insensitivity to fixed stage durations, yet the most sensitive
stages to the other parameters were those stages with the longest stage dura
t i o n s . I t is worth noting that Tom^^®^ showed quite some time ago that simple
matrix projection models are indeed highly sensitive to stage duration. Van
Thenderen^^^ has proposed the use of integrated elasticity measures that attempt to
account for correlation in traits such as survival and fecundity. Application of
integrated elasticity measures might well change some of the conclusions of sea
turtle studies based on deterministic linear stage-classified models.
9.5.2 S tochastic S imulation M odels
9.5.2.1 Dynamical Systems
Chaloupka and Limpus^^^ developed a simulation model of the population dynamics

for green turtles comprising the sGBR genetic stock. The model was a hybrid /-state
distribution type based on a stage-classified demography comprising age-based, and
reproductive status-based stages and it considered both foraging and rookery areas.
Age-based stage durations were derived from weibull-type growth functions pre
sented in Chaloupka and Limpus^^ (see also Equation 9.1). The model comprised a
dynamic system of stage-specific difference equations linked with demographic rates
reflecting nonlinear, feedback, time varying, distributed delay, and stochastic prop
erties. It included demographic stochasticity and also environmental stochasticity
due to the delayed impact of ENSO events on female breeding likelihood (see Limpus
and Nicholls).^ Mortality estimates were derived from both empirical and theoretical
sources. Growth and maturation were assumed to be heterogeneous processes, so
stage transitions were modeled as distributed-delay rates based on Erlangian func
tions (see Blythe et al.).^^"^
The model is part of a larger heuristic modeling system designed as a tool to
help understand the dynamic nature of sea turtle demography, harvesting behavior,
and impact of specific conservation policies. Chaloupka and Limpus^’®used this
simulation model to investigate the potential impact of egg harvesting on the long
term viability of the sGBR green turtle stock at all stages of the life cycle. While
the temporal behavior of most of the stages is unknown, their simulation model
outcomes reflected well the known annual nesting stock at Heron Island (sGBR)
and long-term patterns in variability of southeast Asian green turtle nesting stocks
with a known egg harvesting history.^ In other words, this model reflected well
various independent regional reference behaviors.
Sensitivity analysis was undertaken using parameter perturbation and fractional

factorial sampling design (see Steinhorst et al.).‘^^ They found that their system was
sensitive to changes in the proportion of females in the foraging areas preparing to
breed each year, which is itself a function of stochastic environmental factors and
post-breeding return time — a finding supporting the view of Miller^® for the
importance of conducting foraging area studies.
Chaloupka and Limpus^^^^ have designed a similar dynamical system model for
the sGBR loggerhead stock to support a risk-based evaluation of trawl fishery
impacts and for testing policy options to mitigate trawling risk. The model builds
on the deterministic model of Heppell et al.,^'^ but includes distributed-delay matu
ration, stochastic demographic parameters, and stochastic breeding likelihood. Their
simulation model outcomes reflected well an empirical reference behavior — the
annual Mon Repos (sGBR) nesting stock. They used this model to assess the relative
risks of trawling simultaneously with other known mortality sources on the long
term viability of this particular stock. The modeT^^'' complements empirical studies^^^
on trawl-related risks for sea turtles in Australian waters.
None of the models in Table 9.2 incorporate explicit spatial structure. The models
of Chaloupka and Limpus^^^’^^^‘^ can be extended to include explicit spatial structure
and multispecies systems, but the approach becomes awkward for modeling more
than a moderate number of subpopulations or interacting species.
9.5.2.2 Monte Carlo Process
Bustard and Tognetti*^^ designed probably the first sort of /-state configuration
modeT^^ to investigate sea turtle population dynamics. The model was a Monte Carlo
experiment (see Rubenstein)*^^ to simulate the probability of a green turtle nest being
destroyed as a function of increasing female nesting density. The model was not
well documented, but it seems that hypothetical individual turtles were tracked over
a 35-d nesting cycle as each turtle attempted to nest in a fixed beach area at 15-d
intervals. The model was based on the sGBR green turtle stock nesting on Heron
Island. They used the model to derive a linear function between nesting density and
percentage nest destruction where every 100 additional nesters resulted in an addi
tional 5% nest destruction — all eggs in the nest are destroyed, not a portion.
Bustard and Tognetti^'^^ then proposed that nest destruction was a major density-
dependent mechanism of population regulation. Their linear numerical response
function implies a dome-shaped or extreme Ricker-type^^ stock recruitment function.
There is little basis for adopting either their linear numerical response function nor
the implicit stock-recruitment function. There are no empirical records of extensive
nest destruction for any nesting stock for any species in the Great Barrier Reef
region. Even if such a numerical response function existed it would still be only one
of many factors affecting population dynamics.
9.6 C O N C L U S IO N
Substantial work has been carried out on growth dynamics of sea turtles, increasingly
from wild stocks. Table 9.1 presents many of these growth studies arranged to foster
easy selection of conspeciiic information classified by the underlying methodological

approach. For instance, if information for Kemp’s ridley is needed, then the appro
priate selection would be (1) age cross-sectional (Zug et al.,^^; Chaloupka and
Zug56a), Ç2) age longitudinal (Marquéz,^^ Caillouet et al.,^^) and (3) age-mixed
headstarted (Caillouet et al.).^^ For hawksbills the selection would be (1) age longi
tudinal (Witzel),^^ (2) size-mixed growth interval (Limpus,^® Bo u l o n ) , and (3) size-
mixed nonparametric growth modeling (Chaloupka and Limpus).‘^^'‘
Nonetheless, it is premature for a full-scale comparison of sea turtle growth
dynamics. Many studies relied on inappropriate growth functions, questionable
tagging protocols, and were fitted to limited ontogenetic ranges. Many studies fitted
age- or size-based forms of the von Bertalanffy growth function, yet there seems
little basis for justifying this function either on empirical or statistical grounds. A
necessary condition for a von Bertalanffy-type function is a size- or age-specific
growth rate function that is monotone decreasing over the entire postnatal devel
opment phase — this has not been shown for any sea turtle growth study. Few
studies accounted for time-dependent effects (age, year, cohort) within the con
straints of the sampling design (see Table 9.1), which influences the validity of age
estimation and growth parameter estimates. These sorts of issues will need to be
considered in designing future skeletochronological and somatic growth studies for
sea turtles.
Table 9.2 shows that many sea turtle numerical population models have been
developed over the last 25 years using a variety of related mathematical approaches.
All models have provided significant insight into the dynamics of sea turtle stocks
and many have been concerned with conservation policy design and evaluation. We
agree with Oreskes et a l . that the main value of numerical models is heuristic —
that is, to challenge mind sets and assumptions and encourage more focused empir
ical research. On this basis all the population models shown in Table 9.2 have been
successful.
Sea turtle population modeling, however, is very much in its infancy. The limited
data on size- and age-specific growth and mortality rates is a major limitation. So
too is the lack of information on distribution of stage transition rates for any sea
turtle stock. There are few reliable stock abundance estimates for either breeding
areas or foraging areas, and this was one of the reasons why many studies failed to
provide an empirical stock reference behavior to assess the heuristic value of the
model. Little is also known of the functional responses of sea turtle stocks to food
availability in the foraging areas and whether or not there are any time-dependent
or density-dependent effects. Less is known about post-recruitment dispersal rates
and spatial population structure. Even less is known about the pelagic developmental
habitat and post-pelagic recruitment rates. Estimating demographic rates and the
inherent variability in those rates represents a major challenge for sea turtle research
programs. Long-term mark-recapture programs coupled with reliable tagging and
innovative techniques such as skeletochronology, biotelemetry, and laparoscopy
should form the foundation of such a research program. Clearly, many such programs
are well in place and we can expect to start reaping the benefits over the next few
years.
ACKNOWLEDGMENT
Virginia Institute of Marine Sciences contribution # 2027.
REFERENCES
1. National Research Council, Decline of Sea Turtles: Causes and Prevention, National
Academy Press, Washington, D.C., 1990.
2. Bjorndal, K. A., Bolten, A. B., and Lagueux, C. J., Decline of the nesting population of
hawksbill turtles at Tortuguero, Costa Rica, Conserv. Biol., 7(4), 925, 1993.
3. Limpus, C. J., The worldwide status of marine turtle conservation, in Proc. 1st ASEAN
Symposium-Workshop on Marine Turtle Conservation, Nacu, A., Trono, R., Palma, J. A.,
Torres, D., and Agas, F., Eds.,. Manila, 1993, 43.
4. Horikoshi, K., Suganuma, H., Tachikawa, H., Sato, R, and Yamaguchi, M., Decline of
Ogasawara green turtle population in Japan, in Proc. 14th Annu. Symp. Sea Turtle Biology
and Conservation, Bjorndal, K. A., Bolten, A. B., Johnson, D. A., and Eliazar, P. J.,
Compilers, NOAA Tech. Memo. NMFS-SEFSC-351, Miami, 1994, 235.
5. Limpus, C. J. and Nicholls, N., Progress report on the study of the interaction of the El
Nino-Southem Oscillation on annual Chelonia mydas at the southern Great Barrier Reef
rookeries, in Proc. Marine Turtle Conservation Workshop, James, R., Compiler, Australian
National Parks and Wildlife Service, Canberra, 1994, 73.
6. Zug, G. R., Balazs, G. H., and Wetherall, J. A., Growth in juvenile loggerhead sea turtles
(Caretta caretta) in the North pacific pelagic habitat, Copeia, 1995(2), 484, 1995.
6a. Chaloupka, M. Y., Polyphasic growth in pelagic loggerhead sea turtles, Copeia, 1996,
submitted.
7. Bowen, B. W., Abreu-Grobois, F. A., Balazs, G. H., Kamezaki, N., Limpus, C. J., and
Perl, R. J., Trans-Pacific migrations of the loggerhead turtle (Caretta caretta) demonstrated
with mitochondrial DNA markers, Proc. Natl. Acad. Sci. U.S.A., 92, 3731, 1995.
Queensland: population structure in a warm temperate feeding area, Mem. Queensl. Mus.,
35(1), 139, 1994.
9. Limpus, C. J., Couper, P. J., and Read, M. A., The loggerhead turtle, Caretta caretta, in
Queensland; population structure in a warm temperate feeding area, Mem. Queensl. Mus.,
37(1), 195, 1994.
10. Miller, J. D., Describing sea turtle population structure in foraging areas: a review of
current methods, in Proc. Marine Turtle Conservation Workshop, James, R., Compiler,
Australian National Parks and Wildlife Service, Canberra, 1994, 118.
11. Broderick, D., Moritz, C., Miller, J. D., Guinea, M., Prince, R. I. T, and Limpus, C. J.,
Genetic studies of the hawksbill turtle Eretmochelys imbricata: evidence for multiple
stocks in Australian waters, Pac. Conserv. Biol, 1, 123, 1994.
12. Keyfitz, N., Applied Mathematical Demography, 2nd ed., Springer-Verlag, New York,
1985.
13. Brillinger, D. R., The natural variability of vital rates and associated statistics. Biometrics,
42, 693, 1986.
14. Tucker, J. K., Maher, R. J., and Theiling, C. H., Year-to-year variation in growth in the
red-eared turtle, Trachemys scripta elegans, Herpetologica, 51(3), 354, 1995.
15. Berger, M. P. R, A comparison of longitudinal, mixed longitudinal and cross sectional
designs, 7. Educ. Stat., 11(3), 171, 1986.
16. Fitzhugh, H. A., Analysis of growth curves and strategies for altering their shape, J. Anim.
Sci.. 42. 1036. 1975.
17. Hendrickson, J. R., The green sea turtle, Chelonia mydas (Linn.), in Malaya and Sarawa,
J. Zool. Soc. London, 130, 455, 1958.
18. Balazs, G. H., Dudley, W. C., Hallacher, L. E., Coney, J. R, Koga, S.K., Ecology and
cultural significance of sea turtles at Punalu’u, Hawaii, in Proc. 14th Annu. Symp. Sea
Turtle Biology and Conservation, Bjorndal, K. A., Bolten, A. B., Johnson, D. A., and
Eliazar, P. J., Compilers, NOAA Tech. Memo. NMFS-SEFSC-351, 1994, 10.
19. Le Gall, J.-Y., Bose, P, Chateau, D., and Taquet, M., Estimation du nombre de tortues
vertes femelles adultes Chelonia mydas par saison de ponte a Tromelin et Europa (Ocean
Indien) (1973-1985), Oceanogr. Trop., 21(1), 3, 1986.
20. Limpus, C. J., The hawksbill turtle, Eretmochelys imbricata, in Queensland: population
structure within a southern Great Barrier Reef feeding ground, Wildl. Res., 19(3), 489,
1992.
21 . Wood, J., Wood, K., Critchley, D., Wildt, E., and Bush, M., Laparoscopy of the green sea
turtle, Chelonia mydas, Br. J. HerpetoL, 6, 323, 1983.
22 . Limpus, C. J. and Read, P. C., The green turtle, Chelonia mydas, in Queensland: a
preliminary description of the population structure in a coral reef feeding ground, in
Biology of Australasian Frogs and Reptiles, Grigg, G., Shine, R., and Ehmann, H., Eds.,
Royal Society of New South Wales, Sydney, 1985, 47.
23. Bjorndal, K. A., Demography of the breeding population of the green turtle, Chelonia
mydas, at Tortuguero, Costa Rica, Copeia, 1980(3), 525, 1980.
24. Heppell, S. A., Limpus, C. J., Crouse, D. T., Frazer, N. B., and Crowder, L. B., Population
model analysis for the loggerhead sea turtle, Caretta caretta, in Queensland, Wildl. Res.,
23, 143, 1996.
25. Parmenter, C. J. and Limpus, C. J., Female recruitment, reproductive longevity and
inferred hatchling survivorship for the flatback turtle (Natator depressus) at a major eastern
Australian rookery, Copeia, 1995(2), 474, 1995.
26. Limpus, C. J., Miller, J. D., Parmenter, C. J., Reimer, D., McLachlan, N., and Webb, R.,
Migration of green {Chelonia mydas) and loggerhead (Caretta caretta) turtles to and from
eastern Australian rookeries, Wildl. Res., 19, 347, 1992.
27. Limpus, C. J., Eggler, P, and Miller, J. D., Long interval remigration in eastern Australia,
in Proc. 13th Annu. Symp. Sea Turtle Biology and Conservation, Schroeder, B. A., and
Witherington, B. E., Compilers, NOAA Tech. Memo. NMFS-SEFSC-341, 1994, 85.
28. Chaloupka, M. Y. and Limpus, C. J., Robust statistical modeling of Chelonia mydas growth
rates — southern Great Barrier Reef, in Proc. 15th Annu. Symp. Sea Turtle Biology and
Conservation, Musick, J. A., Compiler, NOAA Tech. Memo. NMFS-SEFSC-387, 1996.
28a. Limpus, C. J., Chaloupka, M. Y, Nonparametric regression modelling of green sea turtle
growth rates (southern Great Barrier Reef), Marine Ecology Progress Series, 1996, sub
mitted.
29. Schwartz, F. J., A long-tenn internal tag for sea turtles. Northeast GulfSci., 5, 87, 1981.
30. Wood, F. and Wood, J., Release and recapture of captive-reared green sea turtles, Chelonia
mydas, in the waters surrounding the Cayman Islands, HerpetoL 3, 84, 1993.
31. Bjorndal, K. A., Bolten, A. B., Gordon, J., and Caminas, J. A., Caretta caretta (logger-
head): growth and pelagic movement, HerpetoL Rev., 25(1), 23, 1994.
32. Green, D., Growth rates of wild immature green turtles in the Galapagos Islands, Ecuador,
J. HerpetoL, 27(3), 338, 1993.
33. Bjorndal, K. A. and Bolten, A. B., Growth rates of immature green turtles, Chelonia
mydas, on feeding grounds in the southern Bahamas, Copeia, 1988, 555, 1988.
34. Boulon, R. H., Growth rates of wild juvenile hawksbill turtles, Eretmochelys imbricata,
in St. Thomas, United States Virgin Islands, Copeia, 1994(3), 811, 1994.
35. Collazo, J. A., Boulon, R. H., and Tallevast, T. L., Abundance and growth patterns of
Chelonia mydas in Culebra, Puerto Rico, J. HerpetoL, 26, 293, 1992.
36. Klinger, R. C. and Musick, J. A., Age and growth of loggerhead turtles (Caretta caretta)
from Chesapeake Bay, Copeia, 1995(1), 204, 1995.
37. Caillouet, C. W., Fontaine, C. T., Manzella-Tiipak, S. A., and Williams, T. D., Growth of
headstarted Kemp’s ridley sea turtles (Lepidochelys kempii) following release, Chelonian
Conserv. BioL, 1(3), 231, 1995.
38. Parmenter, C. J., A preliminary evaluation of the performance of passive integrated
transponders and metal tags in a population study of the flatback sea turtle (Natator
depressus), Wildl. Res., 20(3), 375, 1993.
39. Tucker, A. D., Fitzsimmons, N. N., and Limpus, C. J., Conservation implications of
internesting habitat use by loggerhead turtles Caretta caretta in Wongarra Marine Park,
Queensland, Australia, Pac. Conserv. Biol., 2, 157, 1995.
40. Plotkin, P. T., Byles, R. A., Postal, D. C., and Owens, D. W., Independent versus socially
facilitated oceanic migrations of the olive ridley, Lepidochelys olivácea. Mar. Biol., 122,
137, 1995.
41. Papi, R, Liew, H. C., Luschi, R, and Chan, E. H., Long-range migratory travel of a green
turtle tracked by satellite: evidence for navigational ability in the open sea. Mar. Biol.,
122, 171, 1995.
42. Klinger, R. C. and Musick, J. A., Annular growth layers in juvenile loggerhead turtles
{Caretta caretta). Bull. Mar. Sci., 51(2), 224, 1992.
43. Henwood, T. A., Losses of monel flipper tags from loggerhead sea turtles, Caretta caretta,
J. Herpetol., 20(2), 276, 1986.
44. Limpus, C. J., Estimation of tag loss in marine turtle research, Wildl. Res., 19(3), 457,
1992.
45. Kowarsky, J, and Capelle, M., Returns of pond-reared juvenile green turtles tagged and
released in Torres strait. Northern Australia, Biol. Conserv., 15, 207, 1979.
46. Limpus, C. J. and Walter, D., The growth of immature green turtles {Chelonia mydas)
under natural conditions. Herpetologica, 36(2), 162, 1980.
47. Bjorndal, K. A. and Bolten, A. B., Growth rates of juvenile loggerheads, Caretta caretta,
in the southern Bahamas, J. Herpetol, 22, 480, 1988.
48. Le Gall, J-Y., Lebeau, A., and Kopp, J., Monitoring green turtles at Tromelin and Europa
(Indian Ocean), 1970-1984, Mar. Turtle NewsL, 29, 2, 1984.
49. Frazer, N. B., Limpus, C. J., and Greene, J. L., Growth and estimated age at maturity of
Queensland loggerheads, in Proc. 14th Annu. Symp. Sea Turtle Biology and Conservation,
Bjorndal, K. A., Bolten, A. B., Johnson, D. A., and Eliazar, P. J., Compilers, NOAA Tech.
Memo. NMFS-SEFSC-351, Miami, 1994, 42.
50. Lawless, J. R, Statistical Models and Methods for Lifetime Data, John Wiley & Sons,
New York, 1982.
51. Carlander, K. D., A history of scale age and growth studies of North American freshwater
fish, in Age and Growth of Fish, Summerfelt, R. C. and Hall, G. E., Eds., Iowa State
University Press, Ames, 1987, 3.
52. Zug, G. R., Wynn, A. H., and Ruckdeschel, C., Age determination of loggerhead sea
turtles, Caretta caretta, by incremental growth marks in the skeleton, Smithson. Contrib.
ZooL, 427, 1, 1986.
53. Frazier, J., Ageing studies of sea turtles, Report to National Marine Fisheries Service, NA
81-GA-A-C-00018, and U.S. Fish and Wildlife Service, NA 8LGF-A-184, Washington,
D.C., 1982.
54. Klinger, R. C., George, R. H., and Musick, J. A., Bone biopsy techniques for sea turtles,
Herpetol. Rev., in press.
54a. Klinger, R. C. and Musick, J. A., unpublished.
55. Frazier, J., Tetracycline as an in vivo label in bones of green turtles, Chelonia mydas (L.),
Herpetologica, 41(2), 228, 1985.
56. Zug, G. R., Kalb., H. J., and Luzar, S., J., Age and growth in wild Kemp’s ridley sea
turtles {Lepidochelys kempii) from skeletochronological data, Biol. Conserv., in press.
56a. Chaloupka, M. Y. and Zug, G. R., A polyphasic growth function for the endangered Kemp’s
ridley sea turtle, Fish. Bull, 1996, submitted.
57. Zug, G. R. and Parham, J. R, Age and growth in leatherback sea turtles, Dermochelys
coriacea (Reptilia, Dermochelyidae), Biol. Bull., submitted.
58. Parham, J. F. and Zug, G. R., Age and growth of loggerhead sea turtles (Caretta caretta)
of coastal Georgia: an assessment of skeletochronological age-estimates. Bull. Mar. Sci.,
60(1), 1997.
59. Zug, G. R., Age determination of long-lived reptiles; some techniques for sea turtles, Ann.
Sci. Nat. (ZooL, Paris), I3e Ser. 11, 219, 1990.
60. Francis, R. I. C. C., Maximum likelihood estimation of growth variability from tagging
data, N.Z. J. Mar. Freshwater Res., 22, 42, 1988.
61. Ratkowsky, D. A., Statistical properties of alternative parameterizations of the von Ber-
talanffy growth curve. Can. J. Fish. Aquat. Sci., 43, 742, 1986.
62. Ratkowsky, D. A., Handbook of Nonlinear Regression Models, Marcel Dekker, New York,
1990.
63. Francis, R. I. C. C., Back-calculation of fish length: a critical review, J. Fish Biol., 36,
883, 1990.
64. Ricker, W. E., Back-calculation of fish lengths based on proportionality between scale
and length increments. Can. J. Fish. Aquat. Sci., 49, 1018, 1992.
65. Ricker, W. E., Handbook of Computations for Biological Statistics of Fish Populations,
Bulletin of the Fisheries Research Board of Canada, 191, 1975.
66. Bolten, A. B., Bjorndal, K. A., Grumbles, J. S., and Owens, D. W., Sex ratio and sex-
specific growth rates in immature green turtles, Chelonia mydas, in the southern Bahamas,
Copeia, 1992, 1098, 1992.
67. Frazer, N. B. and Schwartz, F. J., Growth curves for captive loggerhead turtles, Caretta
caretta, in North Carolina, USA, Bull. Mar. Sci., 34, 485, 1984.
68 . Caillouet, C. W., Koi, D. B., Fontaine, C. T., Williams, T. D., Browning, W. J., and Harris,
R. M., Growth and survival of Kemp’s Ridley sea turtle, Lepidochelys kempi, in captivity,
NOAA Tech. Memo. NMFS-SEFC-186, Miami, 1986.
69. Cleveland, W. S., Visualizing Data, Hobart Press, New Jersey, 1993.
70. Schnute, J. and Fournier, D., A new approach to length-frequency analysis: growth struc
ture, Can. J. Fish. Aquat. Sci., 37, 337, 1980.
71. Fabens, A. J., Properties and fitting of the von Bertalanffy growth curve. Growth, 29, 265,
1965.
72. Kaufman, K. W., Fitting and using growth curves, Oecologia, 49, 293, 1981.
73. Schoener, T. W. and Schoener, A., Estimating and interpreting body-size growth in some
Anolis lizards, Copeia, 1978(3), 390, 1978.
74. Chien, Y.-H. and Condrey, R. E., Bias in estimating growth parameters using Faben’s
mark-recapture procedure, Asian Fish. Sci., 1, 65, 1987.
75. Francis, R. I. C. C., An alternative mark-recapture analogue of Schnute’s growth model.
Fish. Res., 23, 95, 1995.
76. Schnute, J., A versatile growth model with statistically stable parameters. Can. J. Fish.
Aquat. Sci., 38, 1128, 1981.
77. Walker, T. A., Juvenile flatback turtles in proximity to coastal nesting islands in the Great
Barrier Reef province, J. Herpetol, 25(2), 246, 1991.
78. Duncan G. J. and Kalton, G., Issues of design and analysis of surveys across time, Int.
Stat. Rev., 55(1), 97, 1987.
79. Skalski, J. R., A design for long-term status and trends monitoring, J. Environ. Manag.,
30, 139, 1990.
80. Ackerman, R. A., Growth and gas exchange of embryonic sea turtles {Chelonia, Caretta,
Copeia, 1981(4), 757, 1981.
81. Frazer, N. B. and Ehrhart, L. M., Preliminary growth models for green, Chelonia mydas,
and loggerhead, Caretta caretta, turtles in the wild, Copeia, 1985, 73, 1985.
82. Frazer, N. B., Survival of large juvenile loggerheads {Carena carena) in the wild, J.
HerpetoC 21, 232, 1987.
83. Witham, R., and Futch, C. R., Early growth and oceanic survival of pen-reared sea turtles,
84. Phasuk B., and Rongmuangsart, S., Growth studies on the ridley turtle, Lepidochelys
olivácea olivácea, in captivity and the effect of food preference on growth, Phuket Mar.
Biol. Cent. Res. Bull., 1,1, 1973.
85. Davenport, J. and Scott, C. R., Individual growth and allometry of young green turtles
{Chelonia mydas), Herpetol. J., 3, 19, 1993.
86. Godinez-Dominguez, E., Carretero-Montes, R. E., Silva-Batiz, F. de A., Ruiz, S., and
Aguilar, B., Crecimiento de neonatos de Chelonia agassizii (Testudines: Chelonidae) en
cautiverio, Rev. Biol. Trop., 41(2), 253, 1993.
87. Witzell, W. N., Growth of captive hawksbill turtles, Eretmochelys imbricata, in western
Samoa, Bull. Mar. Sci., 30, 909, 1980.
88. Marquéz M. R., Resultados preliminares sobre edad y crecimiento de la tortuga lora,
Lepidochelys kempi (Garman), Memorias IV Congreso Nacional de Oceanografía (Mex
ico), 419, 1972.
89. Mohanty-Hejmadi, P., A study of ecology, breeding patterns, development and karyotype
patterns of the olive ridley, Lepidochelys olivácea, of Gahirmatha, Orissa, Pranikee—J.
Zool. Soc. Orissa, 9, 1, 1992.
90. Uchida, I., On the growth of the loggerhead turtle. Carena carena, under rearing condition.
Bull. Jpn. Sci. Fish., 33(6), 497, 1967.
91. Ehrhardt, N. M. and Witham, R., Analysis of growth of the green sea turtle {Chelonia
mydas) in the western central Atlantic, Bull. Mar: Sci., 50(2), 275, 1992.
91a. Chaloupka; M. Y. and Limpus, C. J., Robust statistical modelling of hawsbill sea turtle
growth rates (southern Great Barrier Reef), Marine Ecology Piôgress Series, submitted.
92. Le Toquin, A., Galmel, E., and Trotignon, J., Morphologie, croissance individuelle et
dynamique des populations de la tortue verte {Chelonia mydas L.) au Banc D’Arguin
(Republique Islamique de Mauritanie), Terre Vie, 34, 271, 1980.
93. Bjomdal, K. A., Bolten, A. B., Coan, A. L., and Kleiber, P, Estimation of green turtle
{Chelonia mydas) growth rates from length-frequency analysis, Copeia, 1995(1), 71,1995.
94. Fournier, D. A., Sibert, J. R., Majkowski, J., and Hampton, J., MULTIFAN a likelihood-
based method for estimating growth parameters and age composition from multiple length
frequency data sets illustrated using data for southern bluefín tuna {Thunnus maccoyii).
Can. J. Fish. Aquat. Sci., 37, 301, 1990.
95. Golikov, S. Y, Sakuramoto, K., Kitahara, T., and Harada, Y, Length-frequency analysis
using the genetic algorithms. Fish. Sci., 61(1), 37, 1995.
96. Swingle, W. M., Warmolts, D. I., Keinath, J. A., and Musick, J. A., Exceptional growth
rates of captive loggerhead sea turtles. Carmena carena. Zoo Biol., 12(5), 491, 1993.
97. Mendonca, M. T., Comparative growth rates of wild immature Chelonia mydas and Caretta
caretta in Florida, J. Herpetol., 15(4), 447, 1981.
98. Boulon, R. H. and Frazer, N. B., Growth of wild juvenile Caribbean green turtles, Chelonia
mydas, J. Herpetol, 24, 441, 1990.
99. Hastie, T. J. and Tibshirani, R. J., Generalized additive models: some applications, J. Am.
Stat. Assoc., 82(398), 371, 1987.
100. Steams, S. C., The Evolution of Life Histories, Oxford University Press, Oxford, 1992.
101. Alberch, R, Gould, S. J., Oster, G. E, and Wake, D. B., Size and shape in ontogeny and
phylogeny. Paleobiology, 5(3), 296, 1979.
102. Tuljapurkar, S., Population Dynamics in Variable Envirvnments, Lecture Notes in Biom
athematics, Springer-Verlag, New York, 1990.
103. Frazer, N. B., Survivorship of adult female loggerhead sea turtles, Caretta caretta, nesting
on Little Cumberland Island, Georgia, USA, Herpetologica, 39(4), 436, 1983.
104. Frazer, N. B., A model for assessing mean age-specific fecundity in sea turtle populations,
Herpetologica, 40(3), 281, 1984.
105. Frazer, N. B., Survival from egg to adulthood in a declining population of loggerhead
turtles, Caretta caretta, Herpetologica, 42(1), 47, 1986.
106. Crouse, D. T., Crowder, L. B., and Caswell, H., A stage-based population model for
loggerhead sea turtles and implications for conservation. Ecology, 6 8 , 1412, 1987.
107. Laurent, L., Clobert, J., and Lescure, J., The demographic modeling of the Mediten'anean
loggerhead sea turtle population: first results, Rapp. Comm. Int. Mer Medit., 33, 5, 1992.
108. Crowder, L. B., Crouse, D. T., Heppell, S. S., and Martin, T. H., Predicting the impact of
turtle excluder devices on loggerhead sea turtle populations, Ecol. AppL, 4(3), 437, 1994.
109. Somers, I., Modelling loggerhead turtle populations, in Proc. Marine Turtle Conservation
Workshop, James, R., Compiler, Australian National Parks and Wildlife Seiwice, Canberra,
1994, 142.
109a. Cunningham, D.C. and Brooks, R. J., Bet-hedging theory and eigenelasticity: a compar
ison of the life histories of loggerhead sea turtles {Caretta caretta) and snapping turtles
{Chelydra serpentina). Can. J. ZooL, 74, 291, 1996.
110. Chaloupka, M. Y. and Limpus, C. J., Heuristic modeling of Chelonia mydas population
dynamics — southern Great Barrier Reef, in Proc. 15th Annu. Symp. Sea Turtle Biology
and Conservation, Keinath, J. A., Barnard, D. E., Mustek, J. A., and Bell, B. A., Compilers,
NOAA Tech. Memo. NMFS-SEFSC-387, 1996.
111. Van Buskirk, J. and Crowder, L. B., Life-history variation in marine turtles, Copeia,
1994(1), 6 6 , 1994.
112. Breiman, L., Friedman, J. H., Olshen, R. A., and Stone, C. J., Classification and Regression
Trees, Wadsworth International Group, Belmont, CA, 1984.
113. Limpus, C. J., Gyuris, E., and Miller, J. D., Reassessment of the taxonomic status of the
sea turtle genus Natator McCulloch 1908, with a redescription of the genus and species,
Trans. R. Soc. South Aust., 112(1), 1, 1988.
114. Rhodin, A. G. J., Ogden, J. A., and Conlogue, G. J., Chondro-osseous morphology of
Dermochelys coriacea: a marine reptile with mammalian skeletal features, Nature, 290,
244, 1981.
115. Limpus, C. J., Parmenter, C. J., Baker, V., and Eleay, A., The flatback turtle, Chelonia
depressa, in Queensland: post-nesting migration and feeding ground distribution, Aust.
Wildl. Res., 10, 557, 1983.
116. Guinea, M. L., (1994). A possible model to explain winter nesting by the flatback turtle,
Natator depressus at Fog Bay, Northern Territory, in Proc. Marine Turtle Conservation
Workshop, James, R., Compiler, Australian National Parks and Wildlife Service, Canberra,
1994, 154.
117. Limpus, C. J., Parmenter, C. J., Baker, V., and Fleay, A., The Crab Island sea turtle rookery
in the north-eastern Gulf of Carpentaria, Aust. Wildl. Res. 10, 173, 1983.
117a. Chaloupka, M. Y, Robust statistical modelling of Tortuguero green turtle survivorship
rates. Can. J. Eish. Aquat. Sci., to be submitted.
118. McCullagh, P. and Nelder, J. A., Generalized Linear Models, 2nd ed.. Monographs in
Statistics and Applied Probability 37, Chapman and Hall, London, 1989.
119. Judge, G. G., Griffiths, W. E., Hill, R. C., Lutkepohl, H., and Lee, T.-C., Theory and
Practice of Econometrics, 2nd ed., John Wiley & Sons, New York, 1985.
120. DeAngelis, D. L., Rose, K. A., Crowder, L. B., Marschall, E. A., and Lika, D., Fish cohort
dynamics: application of complementary modeling approaches. Am. Nat., 142(4), 604,
1993.
121. Getz, W. M., and Haight, R. G., Population Harvesting: Demographic Models of Fish,
Forests and Animal Resources, Monographs in Population Biology 27, Princeton Univer
sity Press, Princeton, 1989.
122. Manly, B. F. J., Stage-Structured Populations: Sampling, Analysis and Simulation, Chap
man and Hall, New York, 1990.
123. McDonald, D. B. and Caswell, H., Matrix methods for avian demography, in Current
Ornithology, Vol. 10, Power, D. M., Ed., Plenum Press, New York, 1993, 139.
123a. Chaloupka, M. Y and Limpus, C. J., Heuristic simulation modelling of Fish, bull., to be
submitted.
124. Marquez M. R., Penaflores S, C., Villanueva O, A., and Diaz F, J., A model for diagnosis
of populations of olive ridleys and green turtles of west Pacific tropical coasts, in Biology
and Conservation of Sea Turtles, Bjomdal, K. A., Ed, Smithsonian Institution Press,
125. Marquez M. R., Villanueva O, A., and Sanchez Perez, M., The population of the Kemp’s
ridley sea turtle in the Gulf of Mexico — Lepidochelys kempii, in Biology and Conser
vation of Sea Turtles, Bjomdal, K. A., Ed., Smithsonian Institution Press, Washington,
D.C., 1982, 159.
126. Richardson, J. I. and Richardson, T. H., An experimental population model for the log
gerhead sea turtle {Caretta carettd), in Biology and Conservation o f Sea Turtles, Bjomdal,
K. A., Ed., Smithsonian Institution Press, Washington, D.C., 1982, 165.
127. Doi, T., Marquez M. R., Kimoto, H., and Azeno, N., Diagnosis and Conservation of the
Hawksbill Turtle Population in the Cuban Archipelago, Technical Report for the Bekko
Industry, Japan, 1992.
128. Megrey, B. A. and Wespestad, V. G., A review of biological assumptions underlying fishery
assessment models, in Fishery Science and Management: Objectives and Limitations,
Wooster, W.S., Ed, Lecture Notes on Coastal and Estuarine Studies, 28, 31, 1988.
129. Mohn, R. K., Generalizations and recent usages of yield per recruit analysis. Can. Spec.
Publ. Fish. Aquat. Sci., 92, 318, 1986.
130. Sullivan, P. J., Lai, H.-L., and Galluci, V. R, A catch-at-length analysis that incorporates
a stochastic model of growth. Can. J. Fish. Aquat. Sci., Al, 184, 1990.
130a, Tom, J. S. C., Bounds on the dominant eigenvalue of a population projection models,
Math. BioscL, 57, 175, 1981.
131. Crowder, L. B., Hopkins-Murphy, S. R., and Royle, J. A., Effects of turtle excluder devices
(TEDs) on loggerhead sea turtle strandings with implications for conservation, Copeia,
1995(4), 773, 1995.
132. Lebreton, J.-D., Burnham, K. P., Clobert, J., and Anderson, D. R., Modeling survival and
testing biological hypotheses using marked animals: a unified approach with case studies,
Ecol. Monogr., 62(1), 67, 1992.
133. Van Thenderen, P. H., Life cycle trade-offs in matrix population models. Ecology, 76(8),
2482, 195.
134. Blythe, S. R, Nisbet, R. M., and Gurney, W. S. C., The dynamics of population models
with distributed maturation periods, Theor. Popul. Biol., 25, 289, 1984.
135. Steinhorst, R. K., Hunt, H. W., Innis, G. S., and Haydock, K. P, Sensitivity analyzes of
the ELM model, in Grassland Simulation Model, Innis, G. S., Ed., Springer-Verlag, New
York, 1978, 231.
136. Poiner, I. R. and Harris, A. N., The incidental capture, direct mortality and delayed
mortality of turtles in the Northern Prawn Fishery, Mar. Biol., in press.
137. Bustard, H. R. and Tognetti, K. R, Green sea turtles: a discrete simulation of density-
dependent population regulation. Science, 163, 939, 1969.
138. Rubenstein, R. Y, Simulation and the Monte Carlo Method, John Wiley & Sons, New
York, 1981.
139. Oreskes, N., Shrader-Frechette, K., and Belitz, K., Verification, validation, and confirma
tion of numerical models in the earth sciences. Science, 263, 641, 1994.
■j Q Diving Physiology
M olly E. Lutcavage and Peter L. Lutz
CO NTENTS
10.1 Introduction....................................................................... 277

10.2 Natural Dive Patterns.................................................................................. 278
10.3 Lung Structure and Function...................................................................... 279
10.3.1 Ventilatory Patterns....................................................................... 279
10.3.2 Lung Structure.............................................................................. 280
10.3.3 Pressure and the Respiratory System...........................................281
10.3.4 Pulmonary Gas Exchange............................................................. 281
10.3.5 Pulmonary Perfusion.....................................................................282
10.3.6 Nonrespiratory Functions.............................................................282
10.4 Oxygen Transport....................................................................................... 283
10.4.1 Oxygen Consumption.................................................................. 283
10.4.2 Oxygen Store................................................................................ 283
10.4.3 Blood Gases During Breath-Hold Dives......................................284
10.4.4 Blood Oxygen Transport..............................................................284
10.5 Forced-Diving Responses........................................................................... 288
10.6 Anoxic Tolerance and Hibernation............................................................. 289
10.7 Conclusions................................................................................................. 290
References..............................................................................................................291
10.1 INTRODUCTION
Sea turtles are among the longest and deepest diving of the air-breathing vertebrates.
In the wild, most sea turtles spend as little as 3 to 6% of their time at the surface,
where energetic and predation costs may be high, and can thus be considered truly
subaquatic. The central features of their diving ability involve an efficient oxygen
transport system and an extraordinary tolerance of hypoxia which allows maximal
use of limited oxygen stores. Many of the physiological traits that support this breath-
hold mode of life, such as intermittent breathing patterns, adjustable metabolism,
and hypoxia tolerance, are common reptilian features and undoubtedly were present
in the land-dwelling reptilian ancestors of the sea turtle. However, sea turtles have
distinctive modifications in morphology and physiology that allow them to fully
exploit the marine environment in unique ways, and interesting parallels can be
made with the aquatic adaptations shown by marine mammals.^
0-8493-8422-2/97/$0.00+$.50
For efficient underwater locomotion the sea turtle body plan includes limbs
modified as hydrofoils, a reduction in carapace mass, shortened neck, and body
streamlining. They have nasal passages that close upon diving, orbital salt glands
for salt excretion (see Chapter 14), and some (leatherbacks) can conserve body heat
sufficiently to allow them to forage in cold temperate waters'^ (see Chapter 11).
However, the most important respiratory adaptations for successful breath-hold
diving are those that facilitate efficient and rapid gas exchange when the turtles are
on the surface, and maximize oxygen storage and tissue oxygen delivery while they
are submerged. The important metabolic adaptations that enhance diving prowess
include having an increased anaerobic scope and an enhanced tolerance of hypoxia.
10.2 N A T U R A L D IV E PATTERN S
Sea turtles show a wide range in diving depth and duration, although records for
some species such as the Australian flatback, Natator depressus, are lacking, and
we have no information on adult males of most species (Table 10.1). Deepest dive
depths have been recorded from adult female leatherback turtles, Dermochelys
coriácea (>1000 m),^^^ followed by the olive ridley, Lepidochelys olivácea (290 m,
sex unknown),^^ and adult female loggerheads, Caretta caretta (233 m).^ Although
hawksbills, Eretmochelys imhricata,^^ green, Chelonia mydas,^^ and Kemp’s ridley
turtles, L. kempi,^~^^ tend to remain in shallow water (from 20 to 50 m), a record of
110 m has been given for a green turtle.Hawksbill turtles make the longest routine
dives reported so far (56.1 min, sex unknown),followed by internesting female
olive ridleys (54.3 min),^^ who made longer dives than breeding or post-breeding
males (28.6 and 20.5 min, respectively).The longest reported voluntary dives in
sea turtles range from 2 to 5 h.^ In the olive ridley, diving and surface times of turtles
tracked for extended periods with satellite transmitters varied in relation to repro
duction and migration activities.^* Similarly, satellite-tagged juvenile loggerhead and
Kemp’s ridley turtles showed different mean dive depths and surface intervals,
depending on whether they were located in shallow coastal areas, (short surface
intervals) or in deeper, offshore areas (longer surface intervals).^ ***
In most cases, there seems to be little relation between maximum dive depth
and dive duration. For example, a dive to 211 m by a satellite-tracked loggerhead
off Japan required only 10 min,^ while dives by loggerheads in shallow areas of
Chesapeake Bay are considerably longer.**^^ Perhaps surprisingly, the largest and
deepest-diving sea turtle, D. coriácea, appears to have the shortest routine dives (4
to 11 min).*^-2® Possibly the price of having higher body temperatures and metabolic
rates is that oxygen stores are expended more quickly than in smaller, slower-paced
chelonid sea turtles (see Chapter 11).^*
Despite the brevity of their respiratory phases, sea turtles sometimes spend as
much as 19 to 26% of their time at the surface, engaged in surface basking, feeding,
orientation, and mating.^ * T h e wide variation evident both within and across
species reminds us that sea turtle diving and surface habitats are a reflection not
only of their size and physiological attributes, but of their ecology, environment,
and life history.
Diving Physiology 279
TABLE 10.1
Sea Turtle Dive Records from Published Sources
Dive depth (meters) Dive length (min) % Time
Species Maximum Routine Maximum Routine submerged
Loggerhead (Caretta caretta)
fpn-^’^’ 211-233 9-22 17-30
fpn^ 99 13.5-16.6 14.8-20.5
SubadulP'^ 9-22 19-30 80-94
Kemps ridley (Lepidochelys kempi)
SubadulL’^ <50 300 12.7-18.1 96
ipn>« 167 16.7
Pacific ridley (L. olivácea)
fpn“ 290“ .'2 54.3
mb“ 28.6
Green turtle (Cheionia mydas)
714 110
Subadult‘3 <20 66 9-23
Hawksbill (Eretmochelys imhricata)
inf, d 15 73.5 56.1
Leatherback (Dermochelys coriacea)
>1000 37 4-11
fpn“ 475 50-84 37.4 10-14.5
fpn*^ 2-11
Subadult^^^ 7.7 74-91
Note: Abbreviations: fpn, post-nesting female; inf, internesting female; mb, breeding male; d, daytime;
? not given.
* Misidentified as C. mydas in original report.
10.3 LUNG STRUCTURE AND FUNCTION

During the short time they are on the surface, active breath-hold divers must eliminate
all of the CO2 accumulated during the previous dive and take on sufficient oxygen
for the next dive. Lung adaptations to handle rates of gas exchange that are much
more rapid than those needed by terrestrial relatives should have a high selection value.
10.3.1 V entilatory Patterns

Sea turtles only require a few breaths lasting less than 2 to 3 sec to empty and refill
their lungs, even after being submerged for long p e r i o d s H o w e v e r , there is a
tendency for the number of breaths per breathing episode to increase with submer
gence time, regardless of activity.^^’^^
Increased ventilatory frequency is seen during nesting activity and when swim
ming at high speeds. For example, in leatherback turtles the breathing frequency
increased from 1.0 to 3.3 breaths between the egg deposition and sand
throwing p h a s e s , a n d the ventilatory frequency increased sevenfold over resting
values when juvenile greens swam at 0.4 m sec“în a swimming tunnel.^^ Breathing
hypoxic or hypercapnic-inspired gases caused a shift in breathing patterns in both
green and loggerhead turtles, going from intermittent single breaths to clusters of
multiple breaths, shortened breath holds, and eventually, nearly continuous breath-
i n g .2 6 ,29-32
Sea turtle tidal volumes are highly variable, and are strongly influenced by the
physical conditions the measurements are made under, i.e., whether the animal is
swimming in water, held on land, or placed in a supine position.^'^’^^ Values range
from 4 to 14 ml kg^^ in D. coriacea,^^ 33 to 49 ml kg"^ in Caretta and
24 to 187 ml kg~^ in Chelonia mydas, which represent 27% to over 80% of the
total lung volumes for these species. These tidal volumes are much greater than
those found in other reptiles (e.g., Trachemys scripta, 6.9 ml kg'^^^ Crocodylus
niloticus, 11.0 ml kg~0^^ and allow the sea turtle nearly complete exchange of lung
gases in a few respiratory bouts. Most marine mammals also have much greater tidal
volumes than their terrestrial relatives.
10.3.2 L ung Structure

Sea turtles have wedge-shaped lungs that lie under the carapace within the pleuro
peritoneal cavity and are firmly attached to the dorsal body wall along the vertebral
axis.23,32,39,40 lungs are spongy and multicameral like those of other highly
aerobic reptiles such as Varanus,^^ but early anatomical work identified unique
structures not found in reptilian lungs. These structures include strongly reinforced,
large-diameter airways, and an intrapulmonary bronchus subdividing into reinforced
secondary airways that branch from the primary bronchus and taper gradually in
diameter toward the peripheral margins of the lung.^^ More recent histologic studies
on loggerhead and green turtle lungs showed that cartilage occurs in all but the
central leaflet of respiratory bronchioles, and all airways contain smooth muscle in
close association with an elastic fiber matrix.'^^"^'^ The parenchyma is homogenous,
fibrous connective tissue and prominent myoelastic bundles occur proximal to the
alveoli
In the marine mammals similar structural features facilitate high ventilatory flow
rates by providing mechanical support against airway collapse from hydrostatic
pressure, and it is very probable that the same function is provided in sea turtles.
Maximum expiratory flow rates in sea turtles are indeed at least an order of magni
tude higher than those reported in other reptiles and are only slightly below the
range reported for marine mammab.^^’^"^For example, during spontaneous breathing
on land, expiratory flow in an adult Chelonia mydas reached 120 ml sec~^ kg~^ or
nearly 12 1 sec~^^^ Swimming juvenile green and loggerhead turtles resting on land
had peak flow rates of 57 and 16 ml see~^ kg“^ r es pect i vel y, much higher than
those reported in the semiaquatic turtle T. scripta (0.23 ml sec^^ kg~0-^^
Unlike mammals, sea turtles lack a diaphragm, and pelvic, gular, and pectoral
muscles are recruited to ventilate the lung."^^ In consequence, both inspiration and
expiration are active, and the respiratory muscles perform work, improving the
respiratory pump capacity beyond that of the passive elastic properties of the lung
itself. The ventilation costs associated with large tidal volumes are further offset by
the compliant lung and potential energy stored in the elastic recoil properties of lung
tissue.^'^’'^^
10.3.3 Pressure and the R espiratory System

Despite reinforcement with connective and myoelastic tissue, the respiratory system
compliance of the sea turtle (C^ = 11 ml cm H20~^ kg~^ Caretta carettay^ closely
reflect the pressure-volume characteristics of the body wall, and is only slightly less
than reptiles having simple, less structurally modified lungs (e.g., Gecko 16 ml cm
H20“^ In deep divers, compliant chest walls prevent “thoracic squeeze”
during lung collapse, a condition that could impede central circulation."^^ Complete
lung collapse is believed to occur in deep-diving sea turtles at pressures equivalent
to depths ranging from 80 to 160 m.^"^ Also in sea turtles and other deep-diving
mammals, the ratio of nonrespiratory airway volume to residual lung volume is
large, so that at depth a considerable volume of gas will remain only in the reinforced
a i r w a y . T h e confinement of lung gas to nonrespiratory areas during deep dives
will help prevent nitrogen supersaturation of tissues, commonly known as the
“b ends” .24.37,5o
However, there is some evidence that sea turtles are not completely protected
against the bends, at least under accelerated decompression produced in some lab
oratory studies. Gas emboli have been seen in the capillaries of green sea turtles
that died after rapid ascent from pressure chamber dives of over 14.5 atms.^"^ Indeed,
it has been suggested that bends-related damage accounts for an apparent avascular
necrosis seen in the long bones of extinct sea turtles and mososaurs.^^’^’^ However,
similar sequelae have not been identified in bones taken from extant sea turtles.^^
Given its relatively small and collapsible lung, there seems to be no compelling
evidence that the sea turtle is normally at risk of decompression sickness. A further
protection may be provided by vascular modifications. All sea turtles examined so
far have a thickened muscular area (bulbous arteriosis) of the pulmonary artery that
may constrict during deep diving and invoke a right to left shunt.^^ A diving-related
increase in pulmonary vascular r e s i s t a n c e a n d a shift of circulation away from
the pulmonary circuit may offer additional protection against the bends.
10.3.4 Pulmonary G as Exchange

Sea turtles have the highest rates of oxygen consumption and the greatest aerobic
scopes of any reptile. For example, green and leatherback sea turtles can increase
resting oxygen uptake rate by 8- to 15-fold and attain maximal rates comparable to
the resting rates of most m a m m a l s . T h i s aerobic scope is facilitated by the
extensively subdivided respiratory surface. Sea turtle lungs provide a comparatively
much greater area for gas exchange, and lower resistance to gas transfer, than seen
in most reptiles. For example, the diffusion capacity of the loggerhead lung (0.11
ml min~^ kg ' Torr' at 25°C) is about twice that of nonvaranid reptiles and almost
25% of values for the resting mammal."'^ A high pulmonary diffusion capacity would
be of advantage during prolonged submergence, when sea turtles deplete lung,

arterial, and venous O2 stores. A low-resistance lung would also support the high
metabolic rates seen in maximally exercising sea turtles by maintaining high satu
ration levels in arterial blood.^^’^^"^^
10.3.5 Pulmonary Perfusion

Blood flow patterns can change dramatically during diving. In juvenile greens during
ventilatory bouts swimming at speeds of 4 m secpul monar y blood flow increased
and heart rate nearly doubled.^^ "^^The increase in cardiac output was mainly brought
about through an elevated heart rate with only minor changes in cardiac stroke flow.^^
In green sea turtles, vascular resistance falls during ventilation and exercise and
increases during diving.^^'^'^’^'^
Intracardiac shunting has been seen in force-dived freshwater turtles and in the
green sea turtle.^^ '^^’^'^’^^ But intracardiac shunts appear to be of minor significance
during short aerobic dives in green and loggerhead sea turtles.^^'^^’^'^ However, right
to left and left to right shunts, which would result in a substantial redistribution of
blood away from or to the pulmonary circuit, could account for the observed wide
variations in the depletion of lung and arterial PO2 (see below). There is also evidence
of pulsatile flow to the lungs during resting nonventilatory periods in Chelonia
mydas^^ and in fresh water turtles,believed to be caused by vagally mediated
vasoconstriction in the pulmonary artery.
Pulmonary perfusion has been measured in loggerhead and green sea turtles.
Using gas dilution techniques (25°C) the values for Caretta caretta (86 ml mim^
kg-i)43 significantly greater than those found for Chelonia mydas (24 ml min~’
kg-i) 54 latter being in agreement with measurements using flow probes in juvenile
green turtles (26.5 ml min~^ kg"^).^^ This suggests that loggerheads may have a higher
blood convection requirement and therefore a small arterial-venous O2 content dif
ference as compared to that of C. mydas^^
10.3.6 N onrespiratory F unctions

Many aquatic reptiles appear to regulate the volume of air in their lungs to adjust
buoyancy while diving. In shallow-diving loggerhead turtles the breath hold lung
volume is close to the value predicted for neutral buoyancy^^’^^ and a fine scale
control of buoyancy may be achieved by shifting air between pulmonary compart
ments.^^ Although the pulmonary musculature of loggerhead embryos is fully present
at hatching,like other turtles, initially they can only dive to a few meters, and
appear to require several months to develop full buoyancy control.^^’^^’^^ Diving depth
and breath-hold duration increase with the size of the lung oxygen store (which
increases with biomass), the maturation of the oxygen transport system, and struc
tural development of the lung.
It is important to note that in addition to its roles in gas exchange and buoyancy
regulation, the sea turtle lung may serve other purposes. Because the lung lies directly
under the carapace, it has been suggested that during surface basking blood circulated
through the lung may be used to transport radiant heat from the sun to other regions
of the body.^^ Nesting leatherback turtles sometimes produce quite audible growling
sounds, but the role of the respiratory system in vocalization is not well documented
in the sea turtles.
10.4 O X YG EN TRA N SPO RT
10.4.1 O xygen C onsumption

The dive duration of sea turtles is a function not only of the total oxygen store, but
also of the metabolic rate during the dive, and the latter is dependent on size, activity,
temperature, and hormonal and dietary status. Wide differences in oxygen consump
tion (VO2 ) are found according to activity levels. Because of difficulties associated
with sampling, metabolic rates are hard to measure on freely diving sea turtles, and
most VO2 rates during diving are interpolated from land or laboratory studies. For
example, in leatherback turtles VO2 increased 15-fold between egg-laying and
sand-throwing activity (VO2 = 0.25 ml mim^ kg^^ egg laying,^^ 3.7 ml min'^ kg~^
sand-throwing).A threefold increase has been recorded in juvenile green turtles
going from rest (1.98 ml min~^ kg'0 to a swimming speed of 0.6 m sec^^ (5.6 ml
miir^ kg 1)"^^ and a threefold increase has also been seen in loggerheads from rest
(1.0 ml min^^ kg"^ 25°C) to moderate swimming activity.^^ In other studies VO2
rates for C. mydas range from 1.1 to 2.4 ml mim^ kg"^ (rest)*^^ to 4.5 ml mim^ kg~^
(sand-throwing).^"^’^2 Leatherback hatchlings swimming continuously had VO2 rates
(4.7 ml mim^ kg“^ at 24°C)^^ comparable to and even slightly higher than maximal
rates recorded in hatchling green turtles.^"^
In the sea turtle, increases in oxygen uptake are brought about through increases
in the respiratory frequency, tidal volume, and the amount of oxygen extracted from
lung air. The efficiency of oxygen extracted from lung air is estimated by measuring
the air convection requirement (ACR), the ratio of lung ventilation to oxygen uptake
(Vg / VO2 ). Although ACR values range widely in sea turtles (e.g., loggerhead 16
to 120,^^ green turtles 17.3 to 42.2,'^^’^'^ nesting leatherbacks 37.1),^^’^^ there appears
to be a general tendency for the ACR to decline with increased activity and with a
decrease in temperature.^^’^^'^^ Changes in respiratory frequency account for most
increases in Vg, yet tidal volumes also vary somewhat with activity.^'^’^^’^^’^'^
10.4.2 O xygen Store

It has been suggested that diving birds and mammals typically store oxygen in the
blood and tissues, while amphibians and reptiles use the lungs as the major oxygen
store.^^’^^’^'^ However, consideration of the functional aspects of breath-hold diving
indicates that diving strategies may rest on adaptational rather than phylogenetic
grounds, centering around the different demands of shallow vs. deep diving.^“’’^^
Shallow divers (coastal, estuarine, and freshwater inhabitants) typically inhale
before a dive and depend upon the lung as a major oxygen store. This set of animals
includes most aquatic turt les,the duckbill platypus,the sea otter,^^ manatees,^^
and dolphins.By contrast, the more oceanic species (e.g., the large cetaceans, some
of the pinnipeds, and the leatherback sea turtle) who dive deeply rely more upon
blood and tissue stores for oxygen and accordingly have a different set of adaptations.
For example, in the loggerhead sea turtle some 72% of stored oxygen is carried in
the lungs and tissue oxygen stores are of minor importance.^^ On the other hand,
the deep-diving leatherback turtles have a distinctly different oxygen store strategy
compared to chelonid sea turtles.^^ ^? hematocrit, hemobglobin, and myoglobin
concentrations of the leatherback are among the highest recorded for reptiles, and
approach levels found in diving mammals. The blood and tissue oxygen store of the
leatherback (15.2 ml kg“0 is larger than that of the lung (12.2 ml k g w h e r e a s in
other sea turtles the lung store is larger by at least a factor of two (Figure 10.1).^^’^^’^^
Female leatherback turtles routinely dive to depths that would result in complete
lung collapse, whereas in shallow-diving turtles the lung would remain partially
inflated throughout the dive.^^*^^ However, despite differences in distribution, total
oxygen stores calculated for loggerhead turtles (2 2 . 2 ml O2 kg"^ and the leather
back (27.4 ml O2 kg~0^^ are surprisingly similar. Based on oxygen consumption
measurements and estimates of total oxygen stores, aerobic dive limits of 33 and
70 min have been calculated for a 20-kg loggerhead^^ and the adult leatherback,^^
respectively, encompassing the ranges of dive duration seen in nature.
10.4.3 B lood G ases D uring Breath-H old D ives

The blood gas transport systems of terrestrial vertebrates normally function within
narrow ranges of alveolar PO2 , PCO2 , and blood pH. In contrast, one would expect
that during a dive the sea turtle blood must be able to operate effectively in the face
of intensified hypoxia and hypercapnia.
Unfortunately, there are few blood gas measurements in voluntarily diving sea
turtles. In subadult loggerhead turtles, studied under laboratory conditions, a steady
decline in arterial PO2 was seen during dives, starting from initial values of 112.4
Torr PO2 and falling to 3.9 Ton* after 25 min of breath-hold diving.^^ Interestingly,
the concomitant increases in PCO2 were relatively small, typically <10 Torr for 20-
min dives, suggesting that an efficient C0 2 /bicarbonate and ionic exchange system
was operating. Arterial pH, in consequence, did not decline by more than 0.03 to
0.1 pH units, even during long dives, and sometimes even increased slightly.^^ Such
dives are probably fully aerobic, since blood lactate values obtained from comparably
sized loggerheads under similar conditions were less than 0.2 to 0.4 mM?^ Indeed,
most voluntary dives are terminated at or before the arterial PO2 reaches about 2 0
Torr in freshwater turtles,^^’^^’^®crocodiles,^^ sea snakes,^ and diving b i r d s . On the
other hand, in a 15-min dive by a tethered adult green turtle, arterial PO2 fell to
below 30 Torr and arterial saturation declined from 90 to 45%, there was a substantial
decrease in blood pH, and a tenfold increase in blood lactate concentration. In
the stress of trying to escape, oxygen stores had been used up rapidly and anaerobic
glycolysis had been activated.
10.4.4 Blood O xygen T ransport

For sea turtles (and also marine mammals) the respiratory properties of blood may
depend upon whether oxygen is primarily stored in the tissues or in the lung during
a
<’
era
~u
'<
eora
FIGURE 10.1 Oxygen stores in diving reptiles are divided between the lung, blood, and tissues.
OD
the dive, as the problems of blood oxygen transport are quite different in either
For example, marine mammals and the leatherback sea turtle that store
oxygen in the blood often have comparatively high hematocrits, which increase the
blood oxygen-carrying capacities (Table 10.2). The deep-diving leatherback, for
example, has the greatest hematocrit values of any reptile (32 to 38%).^^ But elevated
hematocrits cause increased blood viscosity, which can make lung to tissue oxygen
transport energetically much more expensive.^ For both adult and hatchling green
turtles. Wells and Baldwin^^ found an exponential increase in blood viscosity at
hematocrits over 30%, and calculated that the optimal hematocrit for this species is
around 30%.
TABLE 10.2
Blood Oxygen Affinity and Transport Properties of Diving Reptiles and
Mammals
Leather- Logger- Green Killer- Weddell
back^^ head^^ turtle^^ Crocodile^'^ whale^^ seap2
Maximum dive >1000 233 110 <30 260 600

depth (m)
Hematocrit (%) 39 29 30 28 44 58
Hemoglobin 15.6 9.8 8.8 8.7 16.0 17-22
(g dl- ')
Myoglobin 4.9 2.9” 44.6”
(mg g'O
O2 carrying 21 7.5-11.9 12.4 23.7 31.6
capacity (vol%) 8.8
P50 (Ton-) 40 47 29 22 31 29 ‘
18.253
pH at P50 7.52 7.45 7.45 7.5 7.4 7.4
7.6”
Bohr effect -0.34 -0.34 -0.30 -0.43 -0.74 -0.61
-0.59”
Hill number 2.7 2.8 2.7 2.6
2 .8“
P50 is the PO2 at 50% saturation. Hill number is the slope of the line describing the relationship
N o te :
between blood saturation log (S/100 - S) and partial pressure of oxygen, log PO2.
By contrast, for animals that use the lung as an oxygen store, the blood must
continue to pick up oxygen as the dive progresses, in face of declining lung PO2
and blood pH. Here, a large Bohr effect would be disadvantageous for oxygen
binding, since the lung PO2 is falling concurrently with pH. Loggerhead and green
sea turtles not only have Bohr values at the low end of the reptilian range, but, more
significantly, the Bohr effects decline substantially at low saturation levels, i.e., at
the operating region towards the end of a dive when the lung pH and blood PO2 are
lowest.^'^ The shapes of the oxygen binding curves in both species are also favorable
for oxygen extraction from the lungs during a dive, in that they are steep rather than
sigmoidal at low saturation levels (Hill n coefficient approaches 1).33,65 indeed, the
kinetics of the oxygen-hemoglobin interaction in the green sea turtle would appear
to favor oxygen uptake rather than oxygen release.^^ These adaptations allow oxygen
to be stripped from the lung in the sea turtle to almost below detectable limits.^^’^'^’^^
Interestingly, the variation in blood oxygen affinities between different species
is very wide (Figure 10.2). The highest affinity is seen by the green turtle (Pso= 27
Torr PO2 at pH 7.4),^^ the leatherback turtle has a P5Qof 40 Toit PO2 at PCO2 of
4 . 8 % , and the loggerhead turtle has the lowest affinity reported (PO2 = 49 Torr at
pH 7 .4 ).33,65,79 adaptive significance of such a wide range on oxygen affinities
is not at all clear, since all animals including the leatherback have basically the same
mode of living, i.e., breath-hold diving. By contrast to greens and loggerheads, the
slope of the oxygen dissociation curve of the leatherback at low saturation is shal-
low.27
FIGURE 10.2 Comparison of blood oxygen binding curves in the (b) leatherback (25°C,
pH = 7.52, 4.8% € 0 2 ),^* and in the green (a) and loggerhead (c) sea turtles (25"^C, pH = 7.45,
PCO2 = 37 Torr).^^ The dotted lines indicate position of P50.
Green and loggerhead hemoglobins appear to be adapted to a role in O2 delivery

in that they have an oxygen binding site that remains strained under all physiological
conditions.^^’^^ The dominance of the strained “T state” acts to preserve the hemo
globins of the sea turtle from uncontrolled stripping of oxygen, allowing modulation
to be dictated by the partial pressure of O2 at the tissue level.
It appears that sea turtle hemoglobins are affected less by changes in temperature
than other reptilian species. In one study, purified hemoglobin from C. mydas showed
an increase in O2 binding with increased temperature, and adult loggerhead hemo

globin, plus cofactors, displayed only a slight decrease of O2 affinity with temper
ature, suggesting that in sea turtles, hemoglobin-mediated oxygen delivery is inde
pendent of temperature.^^’^^’^® However, the functional significance of such
temperature independence is not clear, since the loggerhead is unlikely to experience
marked gradients in body temperatures, and the effect may relate more to the
interdependency of factors that influence hemoglobin oxygen affinity.^^^
Interestingly, while the oxygen affinity of the green turtle hatchling is similar
to that of the adult, the oxygen affinity of pupified (or stripped) hatchling hemoglobin
is much greater than that of the adult.^^’-^^ The reason appears to be that while the
erythrocyte organic phosphates act to decrease the oxygen affinity of the hatchling
embryonic hemoglobin (HbE), they have little effect on the affinity of adult hemo
globin (HbA). Hematocrit and hemoglobin concentrations in Kemp’s ridley, green,
loggerhead, and olive ridley turtles appeared to increase with age.^^ In L. kempi,
there was a substantial increase in these parameters by 7 months, but adult levels
were not reached until 16 months.
10.5 F O R C E D -D IV IN G R ESPO N SES

While most voluntary dives appear to be aerobic, showing little if any increases in
blood lactate and only minor changes in acid-base status, the story is quite different
in forcibly submerged turtles, where oxygen stores are rapidly consumed, anaerobic
glycolysis is activated, and acid-base balance is disturbed, sometimes to lethal levels.
It is now recognized that the “classic” dive response, which was identified in
early studies on enforced submergence in marine mammals and r e p t i l e s , i s
essentially an emergency reaction to special circumstances and rarely occurs during
routine dives.^^’^^ In the “classic” dive response, underwater endurance is maximized
through circulatory adjustments (a rapid onset of bradycardia and severe peripheral
ischemia) which spare oxygen and aerobic substrates for the brain and heart, while
other tissues become anaerobic. The cost, however, is that the animal incurs an
oxygen debt, which must be repaid when breathing is resumed.
In forcibly submerged adult green turtles the heart rate declined sharply from
20 to 30 b mim^ to less than 1 beat min"^;^'^ after 90 min the lung oxygen was
virtually zero, and no oxygen was measurable in the blood after 5 \\P In these
experiments no increase in blood lactate was seen until the first 30 to 60 min of
enforced submergence, when blood lactate reached 90 to 100 mg mkk^^ Recovery
to predive levels required 15 h or more. A somewhat different pattern was seen in
forcibly submerged loggerhead turtles. In these animals, blood oxygen was depleted
to negligible levels in less than 30 min, and venous pH fell from 7.5 to 6.9 within
90 min.^^ Blood lactate increased from 9.9 mM in 30 min to 18 mM in 90 min forced
dive. Following 30- to 90-min dives, loggerhead and green turtles hyperventilated
for up to 30 min, yet remained acidotic for hours.^^ The increase in blood lactate
occurring during prolonged forced dives in juvenile loggerheads appeared to be
slightly more severe than that reported in adult green turtles. In the latter case, lactate
concentration remained low until rapidly released as a spike upon initiation of
breathing.
It is most likely that the rapidity and extent of the internal changes that occur
during forced or emergency dives are functions of the intensity of underwater
struggling activity as well as the length of submergence. For example, oxygen stores
were depleted within 15 min in tethered green sea turtles diving to escape.^"^ Under
such circumstances survival could be compromised if the animal remains trapped
underwater, caught up in gear such as a shrimp trawl. Indeed, accidental drowning
in shrimp trawls, driftnets, longlines, and other fishing gear has been identified as
the most important source of turtle mortality in the U.S.*^^ In a field study examining
the effects of shrimp trawl tow times on sea turtle survival, there was a strong positive
correlation between tow time and sea turtle deaths.^^ There is also evidence that
trawl capture can cause a rapid and severe disturbance in acid-base balance. Juvenile
Kemp’s ridley turtles subjected to experimental trawls lasting a maximum of only
7.3 min (27°C) showed a marked metabolic acidosis (a sixfold lactate increase from
pretrawl conditions) and increased breathing frequency nearly tenfold upon emer
gence.^^ Loggerhead turtles captured in shrimp trawls with tow times of less than
30 min showed severe metabolic acidosis with blood lactate values ranging from
8.8 to 16.2 Lactate recovery times were long, taking as much as 20 h in
loggerheads and Kemp’s ridleys, indicating that turtles are probably more susceptible
to lethal metabolic acidosis if they experience multiple captures, because they would
not have had time to process lactacid loads.^^’^^’^^
Additional factors such as size, activity, water temperature, and interspecific
differences also bear directly on metabolic rates and aerobic dive limits and will
also influence trawl endurance times.^'’ For example, larger sea turtles are capable
of longer voluntary dives than small turtles. Juvenile sea turtles therefore may be
more vulnerable to the stress of enforced submergence than adults, especially during
the warmer months when routine metabolic rates are higher.^^ Disease factors and
hormonal status may also play a role in anoxic survival in forced submergence. For
instance, green turtles afflicted with fibropapillomas or spirorchidaisis may be espe
cially vulnerable to trawl stress. Hematocrits as low as 25% of normal values occur
in afflicted turtles, which would cause a reduction in the blood oxygen transport
capacity. In freshwater turtles, thyroid hormones appear to have a role in setting
metabolic rate, suggesting that hormones may also influence the metabolic status of
a sea turtle.^^
10.6 A N O X IC T O L E R A N C E A N D H IB E R N A T IO N
Characteristically, the vertebrate brain has an absolute dependence on oxygen and
dies with a few minutes of its absence. This is true of the marine mammal, and
maintaining an adequate supply of oxygen to the brain is probably the ultimate
determinant of dive endurance. However, the brains of a few species of freshwater
turtle, and at least the loggerhead sea turtle, have the extraordinary ability to survive
many hours of anoxia, indicating that these species can remain underwater without
breathing for greatly extended periods.
The mechanisms that protect their brains from anoxic failure have been the
subject of several recent r e v i e w s . I n essence, the anoxic turtle brain is able to
maintain ATP levels and ionic homeostasis by severely reducing its metabolic
demands to a level that can be fully met by anaerobic glycolysis. Factors involved
include an increase in the concentration and release of inhibitory neurotransmitters
such as gamma aminobutyric acid and the cofactors such as adenosine and gamma
butyric acid, and a reduction in the concentration of excitatory neurotransmitters
such as glutamate.^^’^"^
The ability to tolerate complete anoxia may allow some freshwater turtle species
such as Chrysemys pietà to hibernate in frozen-over anoxic freshwater pools.'^^ For
sea turtles its use is not so clear. There are a few reports of possible hibernation in
sea turtles. Felger et al.^^ described an assemblage of overwintering green turtles,
Chelonia agassizi, in the Gulf of California, partially buried in the sandy bottom
and believed to have been resident from 1 to 3 months (presumably without eating
or breathing) at water temperatures below 15°C.^^ Torpid sea turtles have also been
recorded off central Florida and coastal Georgia. In the winter of 1977 to 1978 large
numbers of loggerheads were caught by trawling in the Cape Canaveral ship chan
nel These animals were stained black by the anoxic sediments, suggesting that
they had been lodged in the channel mud for a long period. Laboratory studies also
indicate a dramatic reduction in activity in seawater temperatures below 11 to
15"^C.^^’^^ However, it has not been established that cold ocean waters actually cause
sea turtles to become apneic and completely inactive for days. Indeed, there is
evidence from satellite tracking studies that migrating loggerhead and Kemp’s ridleys
remain active below 10°C and as low as 6°C.^ Loggerhead turtles subjected to
seasonal low water temperatures off Tunisia reduce activity, but continue to forage.^^
Apart from incidences of cold stunning of young juveniles in New England waters,
it is likely that as winter approaches most sea turtles living in temperate waters start
to migrate towards warmer tropical waters and do not experience severely cold
temperatures.
From recent studies we have started to gain an appreciation as to how sea turtles
and other divers manage breath-hold dives. This involves fairly radical adaptations
in the lung, blood, and tissue compared to other reptiles. Interesting parallels between
marine mammals and sea turtles in respiratory system structure and in deep diving
vs. shallow diving strategy provide fascinating examples of convergent evolution.
Yet much remains to be learned about the integration of physiological and cognitive
responses to diving. For example, heart rate, pulmonary, and peripheral circulation
may change abruptly in anticipation of surfacing or diving, and also change with
activity during a dive, suggesting that the central nervous system coordinates and
modulates the dive r e s p o n s e . T h e details of these responses in free-diving
animals remain elusive.
We most especially need to know the characteristics and scope of the physio
logical responses of the sea turtle in the field. Huge gaps in our knowledge include
how natural respiratory and cardiovascular patterns and alterations in blood gases
relate to underwater activity, and how these variables change in relation to diurnal
activity and season. Very recently, newly developed physiological sensors and
miniaturized computers that record heart rate, swim velocity, and blood flow infor
mation on free-ranging pinnipeds and diving birds have been deployed on sea turtles.
Satellite telemetry has also opened new avenues of research on the pelagic behavior
of sea turtles, but how the different sea turtle species and their diving habits are
affected by environmental changes such as cold snaps is poorly documented. Finally,
it is a goal of sea turtle conservation efforts worldwide to reduce sea turtle deaths
in fishing gear, identified as the greatest single threat to their survival (see Chapter
13). We must then fully understand how enforced submergence produces lethal stress
in animals better known for their anoxic tolerance and durability.
REFERENCES
1. Seymour, R. S., Physiological adaptations to aquatic life, 'mBiology o f the Reptilia,
Vol. 13, Gans, C., Ed., Academic Press, New York, 1982, chap. 1.
2. Glass, M. L. and Wood, S. C., Gas exchange and control of breathing in reptiles,
Physiol. Rev., 63, 232, 1983.
3. Lutz, P. L. and Hochachka, P, Hypoxic defense mechanisms: a comparison between
diving reptiles and mammals, in Cellular Defense Strategies against Hypoxia, Hoch
achka, P. W., Lutz, P. L., Sick, T., Rosenthal, M., and van den Thilart, G., Eds., CRC
Press, Boca Raton, FL, 1993, 459.
4. Greer, A. E., Lazell, J. D., and Wright, R. M., Anatomical evidence for a counter-
current heat exchanger in the leatherback turtle (Dermochelys coriáceo), Nature, 244,
181, 1973.
5. Sakamoto, W., Uchida, I., Naito, Y., Kureha, K., Tujimura, M., and Sato, K., Deep
diving behavior of the loggerhead turtle near the frontal zone, Nippon Suisan Gak-
kaishi, 56, 1435, 1990.
6 . Sakamoto, W., Naito, Y, Uchida, I., and Kureha, K., Circadian rhythm on diving
motion of the loggerhead turtle Caretta caretta during intemesting and its fluctuations
induced by the oceanic environmental events, Nippon Suisan Gakkaishi, 56, 263,
1990.
7. Soma, M., Radio biotelemetry system applied to migratory study of turtle,/. Fac.
Mar. Sci. Technol. Tokai Univ., 21, 47, 1985.
8 . Byles, R. A., Behavior and Ecology of Sea Turtles from Chesapeake Bay, Virginia,
Ph.D. dissertation. College of William and Mary, Williamsburg, VA, 1988.
9. Keinath, J. A., Movements and Behavior of Wild and Head-Started Sea Turtles, Ph.D.
dissertation. College of William & Mary, Williamsburg, VA, 1993.
10. Mendonca, M. T. and Pritchard, P. C. H., Offshore movements of post-nesting Kemp’s
ridley sea turtles (Lepidochelys kempi), Herpetologica, 42, 373, 1986.
11. Plotkin, P. T, The Migratory and Reproductive Behavior of the Olive Ridley, Lcp/-
dochelys oliváceo (Eschscholtz, 1829) in the Eastern Paciflc Ocean, Ph.D. disserta
tion, Texas A&M University, College Station, 1994.
13. Brill, R. W., Balazs, G. H., Holland, K. N., Chang, R. K. C., Sullivan, S., and George,
J., Daily movements, habitat use, and submergence intervals of normal and tumor
bearing juvenile green turtles (Chelonia mydas L.) within a foraging area in the
Hawaiian islands,/. Exp. Mar. Biol. EcoL, 185, 203, 1995.
14. Berkson, H., Physiological adjustments to deep diving in the Pacific Green turtle
{Chelonia mydas agassizii). Comp. Biochem. Physiol., 21, 507, 1967.
15. Starbird, C. H. and Hillis, Z., Telemetry studies of the internesting movements and
behavior of hawksbill sea turtles {Eretmochelys imhricata) around Buck Island Reef
National Monument, St. Crois, U.S. Virgin Islands, in Proc. 12th Ann. Workshop on
Sea Turtle Biology and Conservation, NOAA Tech. Memo. NMFS-SEFSC-361,
Miami, Florida, 134, 1995.
16. Eckert, S. A., Eckert, K. L., Ponganis, P., and Kooyman, G. L., Diving and foraging
behavior of leatherback sea turtles, (Dermochelys coriacea). Can. J. Zool., 61, 2834,
1989.
17. Eckert, S. A., Nellis, D. W., Eckert, K. L., and Kooyman, G. L., Diving patterns of
two leatherback sea turtles (Dermochelys coriacea) during intemesting intervals at
Sandy Point, St. Croix, U.S. Virgin Herpetologica, 42, 381, 1986.
im ilt, Dermochelys coriacea, Copeia, 1993, 1010, 1993.
19. Standora, E. A., Spotila, J. R., Keinath, J. A., and Shoop, C. R., Body temperatures,
diving cycles, and movement of a subadult leatherback turtle, Dermochelys coriacea,
20. Duron, M. D., Contribution a FEtude de la Biologie d^Dermochelys coriacea (Linne)
dans les Pertuis Crentais, Ph.D. dissertation. University of Bordeaux, France, 1978.
metabolism in the leatherback sea turtle. Can. J. ZooL, 70, 348, 1992.
22. Berkson, H., Physiological adjustments to prolonged diving in the Pacific green turtle
(Chelonia mydas agassizii). Comp. Biochem. Physiol., 18, 101, 1966.
23. Tenney, S. M., Bartlett, D., Färber, J. R, and Remmers, J. E., Mechanics of the
respiratory cycle in the green turtle (Chelonia mydas), Respir. Physiol., 22, 361,1974.
24. Lutcavage, M. E., Lutz, P. L., and Baier, H., Respiratory mechanics in the loggerhead
sea turtle, Caretta caretta, Respir. Physiol., 76, 13, 1989.
25. Lutcavage, M. E. and Lutz, P. L., Voluntary diving metabolism and ventilation in the
loggerhead sea turtle, J. Exp. Mar. Biol. EcoL, 147, 287, 1991.
26. West, N. H., Butler, P. J., and Bevan, R. M., Pulmonary blood flow at rest and during
swimming in the green turtle, Chelonia mydas. Physiol. ZooL, 65, 287, 1992.
27. Lutcavage, M. E., Bushnell, P. G., and Jones, D. R., Oxygen transport in the leath
erback sea turtle Dermochelys coriacea. Physiol. ZooL, 63, 1012, 1990.
28. Paladino, R, O’Connor, M. R, and Spotila, J. R., Metabolism of leatherback turtles,
gigantothermy, and thermoregulation of dinosaurs. Nature, 344, 859, 1990.
29. Jackson, D. C., Kraus, D. R., and Rrange, H. D., Ventilatory response to inspired CQ
in the sea turtle: effects of body size and t e m p e r a t u r e , Physiol., 38, 71, 1979.
30. Kraus, D. R. and Jackson, D. C., Temperature effects on ventilation and acid-base
balance of the green turtle. Am, J. Physiol., 239, R254, 1980.
31. Lutz, P. L., Bergey, A., and Bergey, M., Effects of temperature on gas exchange and
J. Exp. Biol., 144, 155-169, 1989.
32. Jackson, D. C., Respiration and respiratory control in the green turtle, Chelonia
mydas, Copeia, 1985, 664, 1985.
33. Lutz, P. L. and Bentley, T. B., Respiratory physiology of diving in the sea turtle,
Copeia, 1985, 671, 1985.
34. Jackson, D. C, and Prange, H. D., Ventilation and gas exchange during rest and
exercise in adult green sea turtles,/. Comp. Physiol., 134, 315, 1979.
35. Vitalis, T. Z. and Milsom, W. K., Pulmonary mechanics and the work of breathing
in the semi-aquatic turtle, Pseudemys scripta, J. Exp. Biol., 125, 137, 1986.
36. Glass, M. L. and Johansen, K., Periodic breathing in the crocodile, Crocodylus
niloticus: consequences for the gas exchange ratio and control of breathing, J. Exp.
Zoo/., 208, 319, 1979.
37. Kooyman, G. L., Respiratory adaptations in marine mammals, Am. Zoo/., 13, 457,
1973.
38. Kooyman, G. L., Castellini, M. A., and Davis, R. W., Physiology of diving in marine
mammals, Amiw. Rev. Physiol., 43, 343, 1981.
39. Jacobs, W., Die lunge der seeschildkrote Caretta caretta (L.) als schwebeorgan, Z.
Vgl. Physiol., 24, 1, 1939.
40. Milani, A., Beitrage zur Kenntnis der Reptilienlunge. II. Zoo/. Jahrb. Abt. Anat.
Ontog. Tiere, 10, 93, 1897.
41. Perry, S. F. and Duncker, H. R., Lung architecture, volume and static mechanics in
five species of lizards, Respir. Physiol., 34, 61, 1978.
42. Solomon, S. E. and Purton, M., The respiratory epithelium of the lung in the green
turtle (Chelonia mydas L.), J. Anat., 139, 353, 1984.
43. Lutcavage, M., Lutz, P. L., and Baier, H., Gas exchange in the loggerhead sea turtle,
Caretta caretta, J. Exp. Biol., 131, 365, 1987.
44. Perry, S. F., Darian-Smith, C., Alston, J., Limpus, C. J., and Maloney, J. E., Histo
logical structure of the lungs of the loggerhead turtle, Caretta caretta, before and
after hatching. Copela, 1989, 1000, 1989.
45. Leith, D., Lowe, R., and Gillespie, J., Mechanics of baleen whale lungs, Foi/. Proc.
Fed. Am. Soc. Exp. Biol., 31, 335, 1972.
46. Leith, D. E., Comparative mammalian respiratory mochmics. Physiologist, 19, 485,
1976.
47. Butler, P. J., Milsom, W. K., and Woakes, A. J., Respiratory cardiovascular and
metabolic adjustments during steady state swimming in the green turtle, Chelonia
mydas, J. Comp. Physiol. B, 154, 167, 1984.
48. Gans, C. and Hughes, G. M., The mechanisms of lung ventilation in the tortoise
Testudo graeca Linne, J. Exp. Biol., 47, 1, 1967.
49. Milsom, W. K., The interrelationship between pulmonary mechanics and spontaneous
breathing pattern in the Tokay lizard, Gekko gecko, J. Exp. Biol., 113, 203, 1984.
50. Kooyman, G. L., Pressure and the diver. Can. J. Zook, 6 6 , 84, 1988.
51. Rothschild, B. and Martin, L. D., Avascular necrosis: occun'ence in diving Cretaceous
mosasaurs. Science, 236, 75, 1987.
52. Rothschild, B. M., Decompression syndrome in turtles: implications of the fossil
record pathology, avascular necrosis, A/îaz. Carnegie Inst., 59, 287, 1990.
53. Sapsford, C. W., Anatomical evidence for intracardiac blood shunting in marine
turtles, Zool. Afr., 13, 57, 1978.
54. Wood, S. J., Gatz, R. N., and Glass, M. L., Oxygen transport in the green sea turtle,
J. Comp. Physiol, 154, 275, 1984.
55. Burggren, W. W. and Shelton, G., Gas exchange and transport during intermittent
breathing in chelonian reptiles,/. Exp. B iol, 82, 75, 1979.
56. Burggren, W., Smits, A., and Evans, B., Arterial O2 homeostasis during diving in the
turtle Chelodina longicollis. Physiol. Zool, 62, 6 6 8 , 1989.
57. Milsom, W. K., Development of buoyancy control in juvenile Atlantic loggerhead
turtles, Caretta caretta, Copeia, 1975, 758, 1975.
58. Milsom, W. K. C. and Johansen, K., The effect of buoyancy induced lung volume
changes on respiratory frequency in a chelonian (Caretta caretta), J. Comp. Physiol,
98. 157. 1975.
59. Salmon, M. and Wyneken, J., Orientation and swimming behavior of hatchling log
gerhead sea turtles (Caretta caretta L.) during their offshore migration, J. Exp. Mar.
Biol. EcoL, 109, 137, 1987.
60. Davenport, J. and Clough, W., Swimming and diving in young loggerhead sea turtles
{Caretta caretta L.), Copeia, 1986, 53, 1986.
61. Sapsford, C. W. and Van Der Riet, M., Uptake of solar radiation by the sea turtle,
Caretta caretta, during voluntary surface basking. Comp. Biochem. Physiol., 63A,
471, 1979.
62. Prange, H. D. and Jackson, D. C., Ventilation, gas exchange and metabolic scaling
of a sea turtle, Respir Physiol., 27, 369, 1976.
63. Lutcavage, M. and Lutz, P. L., Metabolic rate and feeding requirements of the
leatherback tmtlQ, Dermochelys coriacea, Copeia, 1976, 796, 1986.
64. Prange, H. D. and Ackerman, R. A., Oxygen consumption and mechanisms of gas
exchange of green turtle {Chelonia mydas) eggs and hatchlings, Copeia, 1974, 758,
1974.
65. Lapennas, G. N. and Lutz, P. L., Oxygen affinity of sea turtle blood,Respir Physiol,
48, 59, 1982.
65. Butler, P. J. and Jones, D. R., The comparative physiology of diving in vertebrates,
in Advances in Comparative Physiology and Biochemistry, Vol. 8 , Academic Press,
New York, 1982.
6 6 . Lutz, P. L., Oxygen transport in vertebrate blood. Challenges, in A Companion to
Animal Physiology, Johansen, K. and Taylor, C. R., Eds., Cambridge University Press,
New York,' 1982.
67. Snyder, G. K., Respiratory adaptations in diving mammals, Physiol, 54, 269,
1983.
6 8 . Evans, B. K., Jones, D. R., Baldwin, J., and Gabbot, G. R. J., Diving ability of the
platypus, Aust. J. Z ool, 42, 17, 1994.
69. Gallivan, G. J. and Best, R. C., Metabolism and respiration of the Amazonian manatee
{Trichechus inunguis). Physiol Zool, 53, 245, 1980.
70. Burggren, W., Cardiovascular responses to diving and their relation to lung and blood
oxygen stores in vertebrates. Can. J. Zool, 66, 20, 1988.
71. Wright, J. C., Energy metabolism during unrestrained submergence in the saltwater
crocodile Crocodylus porosus. Physiol. Zool, 60, 515, 1987.
72. Lenfant, C., Physiological properties of blood of marine mammals, in The Biology
o f Marine Mammals, Andersen, H. T, Ed., Academic Press, New York, 1969, 95.
73. McCutcheon, E. H., The respiratory mechanism in turtles, Physiol Z ool, 16, 255,
1943.
74. Grigg, G. C. and Caimcross, M., Respiratory properties of the blood oiCrocodylus
porosus, Respir. Physiol, 41, 367, 1980.
75. Castellini, M. A. and Somero, G. N., Buffering capacity of vertebrate muscle: cor
relations with potentials for anaerobic function,/. Comp. Physiol, 143, 191, 1981.
76. Wells, R. M. G. and Baldwin, J., Oxygen transport in marine green turtle (phelonia
mydas) hatchlings: blood viscosity and control of hemoglobin oxygen-affinity,/. Exp.
B iol, 103, 154, 275, 1994.
77. Eriedman, J. M., Simon, S. R., and Scott, T. W., Structure and function in sea turtle
hemoglobins, Copeia, 1985, 679, 1985.
78. Giardina, B., Galtieri, A., Lania, A., Ascenzi, R, Desideri, A., Cerroni, L., and Condo,
S. G., Reduced sensitivity of O2 transport to allosteric effectors and temperature in
loggerhead sea turtle hemoglobin: functional and spectroscopic sixxdy, Biochim. Bio-
phys. Acta, 1159, 129, 1992.
79. Tsaaks, R. E., Harkness, D. R., and Witham, R R., Relationship between the major
phosphorylated metabolic intermediates and oxygen affinity of whole blood in the
loggerhead {Caretta caretta) and the green sea turtle {Chelonia mydas) during devel
opment, Dev. Biol., 62, 344, 1978.
80. Lutz, P. L. and Storey, K., Strategies for dealing with variations in gas tensions-
vertebrates and invertebrates. Handbook o f Comparative Physiology, Am. Physiol
Soc., in press.
81. Lutz, P. L. and Lapennas, G. N., Effects of pH, CO2 and organic phosphates on
oxygen affinity of sea turtle h e m o g l o b i n s , Physiol., 48, 75, 1982.
82. Davis, B. J., Developmental changes in the blood oxygen transport system of Kemp’s
ridley sea turtle, Lepidochelys kempi. Can. J. ZooL, 69, 2660, 1991.
83. Irving, L., Scholander, P. R, and Grinnell, S. W., Significance of the heart rate to the
diving ability of seals,/. Cell.Comp. Physiol., 18, 283, 1941.
84. Scholander, P. R, Experimental investigations on the respiratory function in diving
mammals and birds, Hvalradets Skr., 22, 1, 1940.
85. Bert, R, Leçons sur la Physiologie Comparée de la Respiration, Bailliere, Paris, 1870.
8 6 . National Research Council, Decline o f the Sea Turtles: Causes and Prevention,
National Academy Press, Washington, D.C., 1985, 355 pp.
87. Henwood, T. A. and Stuntz, W. E., Analysis of sea turtle captures and mortalities
during commercial shrimp trawling. Fish. Bull., 85, 813, 1987.
8 8 . Stabenau, E. K., Heming, T. A., and Mitchell, J. R, Respiratory, acid-base and ionic
status of Kemp’s ridley sea turtles {Lepidochelys kempi) subjected to trawling. Comp.
Biochem. Physiol, 99A, 107, 1991.
89. Lutz, P. L. and Dunbar-Cooper, A., Variations in the blood chemistry of the loggerhead
sea turtle, Caretta caretta. Fish. Bull, 85, 37-43, 1987.
90. Keiver, K. M. and Hochachka, P. W., Catecholamine stimulation of hepatic glyco-
genolysis during anoxia in the turtle Chrysemys pietà. Am. J. Physiol., 30, R1441,
1991.
91. Lutz, P. L., LaManna, J. C., Adams, M. R., and Rosenthal, M., Cerebral resistance
to anoxia in the marine turtles, Physiol, 41, 241, 1980.
92. Lutz, P. L., Anoxic defense mechanisms in the vertebrate hvCm, Annu. Rev. Physiol,
54, 601, 1992.
93. Lutz, P. L. and Nilsson, G. E., The Brain without Oxygen - Cases o f Failure,
Mechanisms for Survival, Landis Press, Austin, TX, 1994.
94. Nilsson, G. E., Lutz, P. L., and Jackson, T. L., Neurotransmitters and anoxic survival
in the brain: a comparison between anoxia tolerant and anoxia intolerant vertebrates,
Physiol. Zool, 64, 638, 1991.
95. Ultsch, G. R., Ecology and physiology of hibernation and overwintering among
freshwater fishes, turtles, and snakes, Rev., 435, 1989.
96. Pelger, R. S., Cliffton, K., and Regal, P. J., Winter dormancy in sea turtles: independent
discovery and exploitation in the Gulf of California by two local culiurts. Science,
191, 283, 1976.
97. Carr, A., Ogren, L., and MeVea, C., Apparent hibernation by the Atlantic loggerhead
turtle Caretta caretta off Cape Canaveral, Rlorida, B/o/. Conserv., 19, 7, 1980-81.
98. Moon, D. Y., The Responses of Sea Turtles to Temperature Changes: Behavior,
Metabolism, and Thyroid Hormones, Ph.D. dissertation, Texas A&M University,
College Station, 1992.
99. Laurent, L. and Lescure, J., L’Hivernage des tortues caouannes Caretta caretta (L.)
dans le sud tunisien, Rev. Ecol. {Terre Vie), 49, 63, 1994.
100. Lazell, J. D., New England waters: critical habitat for marine turtles, Cope/«, 1980,
449, 1980.
101. Lacombe, A. M. A. and Jones, D. R., Role of adrenal catecholamines during forced
submergence in ducks. Am. /. Physiol., 30, R1346, 1991.
1 "I Thermal Biology
James R. Spotila, Michael P. O'Connor,
and Frank V. Paladino
CO NTENTS
11.1 Introduction.................................................................................................. 297

11.2 Biophysical Constraints.............................................................................. 298
11.3 Sea Turtles on Land.................................................................................... 303
11.4 Basking........................................................................................................ 304
11.5 Sea Turtles in Water................................................................................... 306
11.6 Metabolism, Body Size, and Thermoregulation.........................................309
11.7 Conclusions.................................................................................................. 311
References............................................................................................................. 312
11.1 IN T R O D U C T IO N
The biology of sea turtles is intimately tied to the thermal constraints of their
environment. From the time a hatchling enters the water, a sea turtle lives in an
environment that is thermally stable in any given place, but that can vary in tem
perature through time, with depth, and with geographic location. In general, it is
only the female sea turtle that comes onto land and only when she is nesting.
Nevertheless, temperature plays an important role in the biology of the nesting
process as well. Sea turtles are unusual among reptiles because their large body size
allows adults to use insulation and blood flow to alter or control body temperature
(Tb). Therefore, there is a large difference in the ability of hatchlings, juveniles, and
adults to thermoregulate actively.
The environmental constraints on the thermal energetics of sea turtles were
reviewed by Spotila and Standora,‘ and recent advances were reviewed by Spotila.^
Considerable progress has been made in the last 10 years in determining the ther
moregulatory mechanisms of sea turtles. This has been accomplished by laboratory
studies and by the application of sophisticated physiological techniques to field
conditions. In this chapter we will review the basic biophysical constraints on the
thermal biology of sea turtles, the themioregulation of sea turtles on land and in the
water, and the role of body size and metabolism in thermoregulation. For a review
of early research on the thermal biology of sea turtles see Mrosovsky.^
0-8493-8422-2/97/$0.00+$ 50
© 1997 by CRC Press, Inc, 297
Conduction-convection
FIGURE 11.1 Heat exchange between sea turtles and their environment. On land, heat
transfer occurs via radiation, convection (wind), evaporation, and conduction. In water, heat
transfer occurs by conduction-convection. (From Spotila, J. R. and Standora, E. A.,Copeia,
1985, 694, 1985. With permission.)
11.2 BIOPHYSICAL CONSTRAINTS

Energy exchange controls the of sea turtles. Chemical energy from food drives
metabolism, which in turn provides an internal source of heat. However, in all but
the largest animals, energy exchange with the environment is the main source of
heat for a sea turtle whether it is on land or in the water. On land, heat energy
exchange is by radiation, convection, evaporation, and conduction."^ In water, thermal
radiation is absorbed near the surface and heat exchange is primarily by conduction
and convection^’^ (Figure 11.1). In his classic book. Energy Exchange in the Bio
sphere, Gates^ explained the energy environment in which we live and detailed heat
transfer by radiation and convection. His basic message was that
HEAT IN = HEAT OUT + HEAT STORED
By using simple steady-state or equilibrium equations, he indicated how to calculate

the energy balance of an organism. The heat storage term is ignored in steady-state
analysis. Following the convention of Gates,^ this steady-state energy budget equa
tion is given as:
Q abs + M - R + C + E + G ( 11 . 1)
Thermal Biology 299
where:
= radiation absorbed by the surface of the animal from the sun (solar radiation)
and the surroundings (thermal radiation) (W m“^)
M = metabolic heat production (W m“^)
R = thermal radiation emitted by the surface of the animal (W m“^)
C = heat energy lost by convection (W m'^)
E = heat energy lost by evaporation of water or heat energy gained by condensation
of water (W m~^)
G = heat energy lost or gained by conduction through direct physical contact of
the animal with soil, water, or substrate (W
Using this equation we can predict the of a plant or animal if we know its heat
transfer properties and the thermal energy environment to which it is exposed. Details
of the mechanisms of heat transfer in animals are presented by Spotila et al.^ and
do not need to be repeated here.
Porter and Gates'^ apply this approach to animals to produce climate space
diagrams. These diagrams give thermal limits for birds, mammals, and reptiles.
While useful for small animals, climate space diagrams are less useful for large
reptiles because large crocodilians and sea turtles have high internal heat storage
capacities and therefore large thermal inertia.^’^^Therefore, climate space diagrams
indicate long-term or average limits on the thermal biology of these large animals.
Of more interest for sea turtles are transient energy budgets. The reason for this is
that we need to predict body temperature of a sea turtle through time in order to
understand its thermal biology.
Equilibrium and transient energy budget equations make different assumptions,
are solved by different methods, and provide answers to different questions. Equi
librium equations provide information about the thermal environment of the animal.
The Tb computed from Equation 11.1 is the T^ that a sea turtle would come to if
the turtle stayed in one place for a long time and neither the environment nor the
turtle changed. Since most sea turtles have large thermal inertia and, thus, seldom
equilibrate with environmental conditions, their T^, tends to average out the changing
environmental temperature over several hours or days. As the thermal environment
changes, the turtle T^ is also changing (Figure 11.2). It approaches the environmental
temperature with a lag due to its large heat capacity. Environmental temperature
reaches a peak and starts to drop while the body core of the sea turtle is still warming
up. Body temperature reaches a peak while environmental temperature is dropping,
and then follows it down. In essence, the T^ of the turtle is chasing the effective
temperature of the environment. Even so, the turtle should respond behaviorally to
a changing thermal environment because it senses changes in skin temperature. The
turtle has to anticipate changes in environmental heat load before they occur to avoid
overheating due to its own internal heat transfer lag. A good example of this problem
is reported by Colbert et al.,^* who found that when they removed an alligator from
the sun as its T^ approached the lethal value, T^ continued to rise for some time,
and the alligator died of heat stress despite being in the shade for hours. Thus, we
Time
FIGURE 11.2 An idealized relationship of body temperature (T^,) change to environmental
temperature change in a variable environment. Environmental temperature is the operative
environmental temperature (see text). In a large animal such as a sea turtle, the core
never comes to equilibrium with the % if that temperature fluctuates rapidly. The is
moderated or dampened and changes more slowly and less dramatically than the It appears
as if the chases, but never catches the T^. Thus, a large sea turtle has a more constant
temperature than its environment because of its thermal inertia. (From Spotila, J. R., O’Con
nor, M. R, Dodson, R, and Raladino, F. W.,Mod. GeoL, 16, 203, 1991. With permission.)
need to have a means to study the instantaneous response of sea turtles to their
thermal environment and we need to have a means to study transient heat transfer,
because this determines at any given time.
We can study the equilibrium and instantaneous energy exchange of a sea turtle
by making use of the concept of operative environmental temperature (TJ as defined
by Bakken and Gates,Bakken,*^ and Bakken et al.^"^ The Tg is a temperature that
indicates the heat load that the environment places on the surface of an animal. It
averages the effect of solar radiation, heat radiation from the suiToundings, and
convection into a single temperature value. This is the temperature of a darkened
environmental chamber with black walls that would provide the same thermal envi
ronment (heat load) as that experienced by the animal under the conditions in which
the Tg is being measured. The Tg can be computed from Equation 11.1 or can be
measured with an operative temperature model that has the same size, shape, and
radiative properties as the real animal in question. Thus, Standora et al.^^ used a
large hollow copper model of a sea turtle to measure the Tg of green turtles. Cheionia
mydas, on the beach at Tortuguero, Costa Rica. The use of such a model would
allow a detailed study of phenomena such as the basking of green turtles at French
Frigate Shoals in the North Pacific. Caution must be applied with this technique,
however, because it is not a “cure all” for all biophysical studies of animals. Many
poorly conceived and executed studies have produced unclear data because the
investigators did not take the time to understand the intricacies of the Tg concept.
Thermal Biology 301
Bakken^^ provided a summary and critical review of the uses and measurements of
Tg. We do not repeat that analysis here. Potential users of this technique would be
well advised to consult this and other papers by Bakken to develop clearly their
question and methodology before carrying out a study of sea turtles.
Most importantly, before using a model one must understand that such models
(particularly for large animals) have several limitations. (1) Models must have the
same heat exchange properties (radiation, convection) as the real animal. It is not
sufficient to take a piece of copper pipe or a large sphere and paint it some color
that appears to look like a sea turtle, and then assume that the model represents the
turtle. The visual color may have little relevance to the absorptivity of the surface
to visible and near-infrared solar radiation, because the near infrared is a large
component of the total absorbed radiation, and our eyes cannot see it. (2) The T^
models do not account for metabolism and do not model evaporation effects very
well, unless you cover the model with a wet surface, but that would change the
radiative properties as well. Of course, sea turtles do not have wet skins, so that is
less of a problem than for amphibians. (3) Models must be assessed in the same
thermal environment as that experienced by the animal. For example, a sea turtle
model needs to be in the same position as the sea turtle on the beach. If it is a green
turtle model for a nesting turtle, then it needs to be in a body pit. (4) A model of a
large animal such as a sea turtle may have problems with gradients among its parts.
That is, the top surface (carapace) will be hotter than the true T^ of the animal, while
one of the limbs will be another temperature, etc. The gradients in Tg may be real
or may be an artifact of the model. It is possible that the Tg of the carapace is different
than the real Tg of the plastron or front limb because those parts of the turtle
experience different heat loads, or those differences may be due to a problem of air
circulation in the model. That would have to be determined by more complicated
modeling and computational efforts. In using a Tg model of an animal, we are looking
for the average Tg of the whole animal. Therefore, the true Tg will be best estimated
by the internal air temperature of the model. If the regional differences in Tg are
very troublesome, it may be necessary to put a small fan into the model to assure
good air circulation. (5) By design, a Tg model is not supposed to measure T^. It is
designed to measure the heat load on the surface of the animal. That is why we do
not fill it with water or some other substance. We do not want it to respond with
the same time constant as the real animal. We are measuring the heat exchange at
the surface, not the T^ of the animal.
Transient energy budget equations predict how the T^ of an animal will change
with time. If the animal is already at its equilibrium T^ and the environment is stable,
no change in T^ is predicted. Thus, the equilibrium energy budget equation is a
special case of the transient energy budget equation with the heat storage term = 0.
The transient model is more generalized. While the equilibrium equation can be
solved as an algebraic equation, the transient model involves the solution of a
differential equation. There are standard solutions for transient energy budget equa
tions,^’ and transient models have proven useful for consideration of heat transfer
in large reptiles, including sea turtles^^^ and d i n o s a u r s . T h e basic conclusion of
these studies is that as body size of an animal increases, fluctuations in core T^, are
reduced on hourly to daily time scales. For example, calculations by Spotila et al.^^
indicate that the of the dinosaur Compsognathus (1.96 kg) would vary with
fluctuations in Tg of a few hours, like a large green iguana, while Tj, of Deinonychus
(75 kg) would vary with fluctuations on the order of less than a day, like an olive
ridley turtle (Lepidochelys oliváceo) on land, and Ti, of Tenontosaurus (624 kg)
would vary with % changes on the order of a week, like a leatherback turtle
{Dermochelys coriáceo) on land (Figure 11.3). Of course, sea turtles in water would
be exposed to more rapid heat transfer at the surface of the skin (see below). Thus,
differences in body size alone, both within and between species, result in different
thermoregulatory problems for sea turtles. Sea turtles can only thermoregulate within
the constraints placed upon them by the interaction of their physiological, behavioral,
and physical characteristics and the physical characteristics of their environment.
Within those constraints biological interactions, for example, predation by sharks,
will further limit the ability of sea turtles to thermoregulate.
FIGURE 11.3 Predicted responses of dinosaur body temperatures (T^) to variations in the
thermal environment. Fractional amplitudes predicted for various-sized dinosaurs when
exposed to fluctuations in the thermal environment with periods ranging from 1 min to l(f
min (approximately two years). Fractional amplitudes are changes in \ expressed as a fraction
of the total difference in between one equilibrium and another. Thus, a fractional
amplitude of 0.5 indicates that T^ would change half the way from one equilibrium temperature
to another. Different symbols and lines represent different masses. Multiple symbols for a
particular mass and period represent the effect of varying cardiac output from 20% of the
predicted resting flow rate to 10 times the resting rate. Values are presented for 0.2, 0.5, 1.0,
2.0, 5.0, and 10.0 times resting flow rates. Lines go through values for 1.0 times resting flow
rate. In all cases, lower cardiac outputs result in lower predicted fractional amplitudes,
indicating that core T^ would change less in response to changing environmental temperature
with lower rates of blood flow. Increases in body size and decreases in blood flow isolate
core Tb from environmental temperature. (From Spotila, J. R., O’Connor, M. R, Dodson, R,
and Paladino, F. V., Mod. GeoL, 16, 203, 1991. With permission.)
Thermal Biology 303
11.3 SEA TURTLES ON LAND

Most sea turtles nest at night because exposure to the hot sun during the day would
lead to lethal heat gain. Large sea turtles in tropical water are relatively warm and
crawl onto the beach with of 29 to 32°C. They are already close to their upper
limit for thermal tolerance (which we assume is near 40°C, although it has not been
measured). A rise of a few degrees in Ti, would place them in danger of overheating.
Green turtles walking on the beach at Tortuguero, Costa Rica during midmorning
experienced a rapid rise in temperatures of the plastron and carapace, while (near
liver and pectoral muscles) rose at a slower rate.^^ Using multichannel telemetry we
found that, in full sun, plastron temperature of a 117-kg turtle rose from 36.5 to
42.9°C, carapace temperature rose from 34.5 to 53.8°C, and deep body temperature
rose from 32.1 to 35.UC in 20 min. The gradient across the thickness of the carapace
was 17.1°C and across the thickness of the plastron was 5.5°C. The T^ measured
for a model green turtle on the beach was 44°C. At this point we allowed the turtle
to enter the surf (28.5°C), and external temperatures dropped. After 17 min the
carapace surface dropped to 40.6°C while the inner surface rose from 36.7 to 36.9°C.
External plastron temperature dropped to 38.8°C while inner surface temperature
rose from 37.4 to 37.6°C. Body temperature continued to rise (0.7°C) and reached
35.8°C. Heat continued to move into the turtle despite the fact that its shell was
cooling in the water. This was due to internal heat lag similar to the phenomenon
encountered by Colbert et al.^^ with the alligator discussed above.
Under overcast sky, T^ of two green turtles walking on the beach rose 2.6 and
2.8°C h”^ Heating occurred primarily through the carapace, while the plastron
remained cooler than the deep body and acted as a heat sink. Under clear sky at
night, Tb of green turtles walking on the beach dropped slowly (0.3°C h“0 over 2
h. This was due to theimal inertia, which was also indicated in the thermal time
constants for green turtles. A time constant is the time it takes for the temperature
of an object or animal to change about 67% of the way to equilibrium when the
object or animal is transferred from one constant T^ to another. Time constants
of three turtles ranged from 420 min for a 54-kg turtle to 690 min for a 104-kg turtle
(Table 11.1).
These thermal characteristics of green turtles combine to prevent them from
nesting during the day. Spotila and Standora^ computed that a green turtle nesting
in the full sun (air temperature, T^ = 28°C, radiation absorbed of 864 W m"^ on the
upper surface and 432 W nr^ averaged over the entire surface, heat gain from sand
at 40°C of 212 W averaged over the entire surface) would warm up 3 to 6°C
due to solar heating and 2 to 4°C due to elevated metabolism during nesting. Thus,
it would reach 36 to 41°C, suffer heat stress, and probably have to return to the
water before completing nesting. In addition, its superficial tissues would be much
hotter and would suffer heat damage. Only small sea turtles like L. olivácea can
regularly nest during the day. Intuitively, we might think that such a turtle would
heat up faster than a green turtle because of their smaller surface-to-volume ratio.
However, they do not do so because of the effect of convection. Olive ridleys nest
primarily on windy and cloudy days when there is less heating due to solar radiation.
In addition, their smaller adult body size results in more effective convective cooling.
TABLE 11.1
Time Constants for Cooling Experiments in Green
Turtles in an Environmental Chamber Aboard the Alpha
Helix at Cayos Miskito, Nicaragua in August, 1978
Temperatures (°C)
Turtle mass (kg) Ta Te Tb AT X (min)
54 16.4 16.4 17.2 0.8 420

61 16.4 16.4 17.1 0.7 460
104 16.4 16.4 17.9 1.5 690
N o te :Tg is mean air temperature; is mean operative environmental tem

perature as measured with a hollow copper turtle model painted to have the
same absorptivity as a green turtle; Tj, is final equilibrium body temperature
of turtle; AT is Tj, • T^, and x is the time constant computed as the time
required for 63.2% of the temperature change between the initial body
temperature and final body temperature at equilibrium. (Data from Standora,
E. A., Spotila, J. R., and Foley, R. E., J . T h erm . B io l., 1 , 159, 1982. With
permission.)
The differential effects of body size and wind speed on environmental heat load are
apparent in the for different sized turtles. A 50-kg turtle with little wind would
have a of 58.3°C, while an increase of wind to 5 m s~* would lower to 30.0°C.
At the same wind speeds a 200-kg turtle would be exposed to of 58.7 and 36.0°C.
For an explanation of the relationship between body size, wind speed, and convection
coefficient in sea turtles see Spotila and Standora.^
11.4 BASKING
Sea turtles bask on land and in the water. Terrestrial basking by green turtles occurs
in the northwestern Hawaiian I s l a n d s , i n Australia,^^ and historically in the Gal
apagos Islands and Mexico.^^ Whittow and Balazs^^ completed the most comprehen
sive study of this phenomenon at French Frigate Shoals in the northwestern Hawaiian
Islands. Here turtles come ashore and bask on the side of islands facing the outer
reef and the prevailing northeast trade winds. Sand temperatures and black globe
temperatures (an approximation of were lower on beaches used for basking
than on beaches on the opposite side of the island that were not used by basking
turtles (Figure 11.4). If we use black globe temperature as a measure of Tg, then it
is apparent that beaches used for basking are cooler than nonbasking beaches.
Perhaps turtles would overheat if forced to stay out on the hotter beaches.
Our calculations of heat stress in green turtles (above) suggest that T^ of 40°C
or higher would result in heat stress for this turtle. Carapace temperatures were as
high as 40.0 to 42.8°C in basking turtles. By flipping sand onto their flippers and
carapace, turtles lowered surface temperatures by as much as 10°C. This would
Thermal Biology 305
Time (h)
FIGURE 11.4 Environmental conditions experienced by green turtles basking on French
Frigate Shoals in the Pacific Ocean.^^ Upper lines (triangles) indicate temperatures measured
with a black globe thermometer and lower lines (circles) indicate temperatures of the sand.
Temperatures on the basking beach (solid symbols) are lower than temperatures of the non
basking beach (open symbols). The black globe temperature, which approximates T^, is much
higher than sand temperature on both beaches. (From Spotila, J. R. and Standora, E. A.,
Copeia, 1985, 694, 1985. With permission.)
reduce heat load on the peripheral tissues and allow the turtle to warm up deep
without overheating the peripheral tissues. Basking on land was most prevalent
among female turtles and appeared to be fostered by cool ocean temperatures
(26.3°C) and a combination of white sand beaches with steady wind and moderate
solar intensity. It is possible that females may be able to accelerate development of
eggs by being at a higher T^. Some authors have also suggested that these turtles
are not thermoregulating because they do not use the hottest beach. Instead they
may be avoiding predators such as sharks and avoiding mating encounters with male
turtles. Data available to date are consistent with a themioregulatory interpretation
of this basking behavior. Additional studies are needed to clarify the driving force
behind this behavior. These should include detailed studies of basking behavior of
individual turtles and accurate measurements of the T^ to which these turtles are
being exposed.
11.5 SEA T U R T LE S IN W ATER

Water places much tighter constraints on the thermal biology of sea turtles than does
the environment on land. Water supports convection coefficients that are 100 times
greater than in air^ because it has a high heat capacity and high thennal conductivity.
This allows water to act as a temperature stabilizer. It tends to keep animals at the
same temperature as their environment. Liquid water is highly transparent to visible
radiation, but absorbs in the ultraviolet and infrared. Thus, a turtle under water can
absorb some solar radiation, but cannot reradiate heat to its sun'oundings. Heat
transfer is very rapid via conduction and convection. The rapid heat transfer between
the turtle and its environment strongly limits the warming effect of metabolism. In
general, Ti, of inactive loggerhead (Caretta caretta), green, and olive ridley turtles
are within 1 to of water temperature (T^) (reviewed in Reference 1). Heat
transfer occurs through the soft skin of the neck and proximal area of the flippers,
followed by the plastron, carapace, and scaled epidermis of the rest of the flippers.
Leatherbacks are typically 30.5 to 32.0°C when they come ashore to nest.^^’^'^ Thus,
in tropical waters, they have T^, similar to those of green turtles. In subarctic waters,
they are much warmer than T^ due to their large body size and thennoregulatory
capabilities (see below).
Both green turtles and loggerhead turtles maintain T^, slightly higher than T^
while swimming and resting in the water. Standora et al.^'^ used multichannel sonic
telemetry to determine that two green turtles had elevated body temperatures while
swimming off Tortuguero, Costa Rica. One turtle (110 kg) had a T^ of 0.7 to 1.6°C
above T^ when alternately resting and swimming slowly a few hundred meters from
shore (Figure 11.5). The other (121 kg) had a T^ of 32.8 to 37.1°C while swimming
rapidly in water of 29.1°C (Figure 11.6). They concluded that the green turtle was
endothermic and could raise T^, through its swimming activity. This was supported
by the results of Prange and J a c k s o n , w h o demonstrated that green turtles have
a highly aerobic metabolism and raise their metabolism 10 times above standard
metabolism when active. Warm pectoral muscles probably increase the swimming
ability of this turtle and may facilitate long-distance migrations.
In a remarkable series of studies, Naito et al.^^ and Sakamoto et al.^^’^^ used
satellite telemetry and microminiature data recorders to track the movements of a
loggerhead turtle during its internesting period and to measure its dive depths, T^,
and T^. They found that stomach temperatures of the turtle were up to 2 to 3"^C
above T^ during a 20-d period while she swam and dove to a maximum depth of
233 m during her intemesting m o v e m e n t s . I n one instance, during a period of
127 min of active swimming the turtle raised its stomach temperature from 24.5 to
25.0°C. Since the accuracy of the thenuistor was +/- 0.15°C,^^ this rise was just
high enough to indicate that the turtle actually did warm up from metabolic activity.
In a more extensive study, Sato et al.^^ used time-temperature recorders, time-
depth recorders, and time-light intensity recorders to determine the relationship
between T^ and T^ and to determine if elevated T^ were caused by basking. Sapsford
and van der RieF^ had reported that a 42-kg loggerhead raised its T^ 3.8°C above
T,y by basking in sunlight while keeping a substantial portion of its carapace exposed
above the water surface. Sato et al.^^ found that 8 loggerheads had a mean T^ of
Thermal Biology 307
FIGURE 11.5 Temperatures of a 110-kg adult female green turtle swimming in the Car
ibbean Sea at Tortuguero, Costa Rica. A multichannel sonic transmitter with remotely posi
tioned thermistors transmitted temperatures from the deep body (triangles), inside the plastron
(squares), carapace surface (open circles), and water (crosses). This female was released from
land at 1100 h, remained a few hundred meters off shore and was relatively inactive.^^ (From
Spotila, J. R. and Standora, E. A., Copeia, 1985, 694, 1985. With pennission.)
0.9°C above and that this difference increased with increasing body mass to a
maximum of 1.7°C (Figure 11.7). These turtles did not so much regulate at a constant
T^, as maintain a relatively constant temperature difference between T^ and T^. They
spent most of their time within 30 m of the surface, and the light intensity data
suggested that they were not basking. Body temperatures were higher than T^ both
on cloudy days and at night. It appeared that the relative constancy of in these
turtles was due in part to thermal inertia as would be expected from biophysical
considerations. A 70-kg loggerhead would have a time constant of perhaps an hour
in water (see above and References 1, 20, 22, and 23) and a resultant overall thermal
lag of a few hours. Thus, variation in T^ would be dampened out and it would remain
somewhat above T^ with less variability as the turtle swam from a water mass at
one temperature into a water mass at another temperature. Thus, as cautioned by
Neill and Stevens,w hile these data indicate that loggerheads enjoy considerable
thermal inertia and have some endothermic capabilities, they do not conclusively
demonstrate that these turtles regulate T^, by physiological means.
Despite the thermoregulatory adaptations of green turtles and loggerheads, nei
ther species functions well at lower T^. This makes them susceptible to cold stunning,
a winter phenomenon in temperate and subtropical waters in which the T^ drops
CA0f0ifta mydos 2 8 Aug. r t
38.0“
U
34.0-
m
V*
3
O
w
CL
£ 30.0-
<0
H
28.0-
--- 1--------- j----------j---- —r

10*00 14*00 IS'OO
Tim e (h)
FIGURE 11.6 Temperatures of a 121-kg adult female green turtle swimming in the
Caribbean Sea near Tortuguero, Costa Rica. A multichannel sonic transmitter with remotely
positioned thermistors transmitted temperatures from the deep body (triangles), inside the
plastron (solid squares), plastron surface (open squares), inside the carapace (solid circles),
carapace surface (open circles), and water (crosses). Arrow indicates when turtle entered
the water (29°C). This turtle was continuously active and pulled a large polyurethane float
and long length (30 m) of manilla rope (diameter 1 cm).^^ (From Spotila, J. R. and Standora,
E. A., Copeia, 1985, 694, 1985. With permission.)
quickly (in a few days) and turtles become in cap acitated .K em p ’s ridley, L. kempi,
turtles are similarly affected. Cold stunning occurs when T^ drops below 8°C before
sea turtles can swim away from an area during the sudden onset of a cold front. The
turtles lose their ability to swim and dive, lose control of buoyancy, and float to the
surface. If not rescued and resuscitated by humans, most of these turtles die. The
physiological basis for this response is unknown, but probably involves disruption
of metabolic pathways and disruption of ion gradients across membranes
There remains a need for studies to determine the low-temperature tolerance of
sea turtles, their high-temperature tolerance, and their preferred temperature. These
studies can be most easily done on hatchlings, but are also needed for juveniles,
subadults, and adults. It may be impractical to study large individuals and difficult
to obtain permits to study the thermal tolerances of a sea turtle like the Kemp’s
ridley, which is highly endangered. However, carefully designed and nonlethal
Thermal Biology 309
FIGURE 11.7 The relationship of mean difference between body temperature and water
temperature to body mass of loggerhead turtles swimming in the Pacific Ocean off Japan
during their intemesting periods. Mean values are presented for each turtle. The linear
regression is significant at p = 0.05. (Redrawn from Sato, K., Sakamato, W., Matsuzawa, Y.,
Tanaka, H., Minamikawa, S., and Naito, Y, Mar. Biol., 123, 197, 1995. With permission.)
studies of critical thermal maximum and critical thermal minimum (CTM^^ and
will elucidate thermal limits, and preferred-temperature studies will clarify
the role of temperature in behavior and physiology of these species. These studies
can be done without harming individual turtles and should be done so that we can
understand the thermal biology of these species and avoid loss of sea turtles to
thermal events such as cold stunning in the future.
11.6 M ETA BO LISM , B O D Y SIZE , A N D

T H E R M O R E G U LA T IO N
Are any sea turtles warm blooded? That is, do they have a high mammalian-like
metabolic rate? We discussed that question in our previous review^ and concluded
that leatherbacks should be capable of maintaining large temperature differences
between their body core and the surrounding water even with a low metabolic rate.
Studies since 1985 have supported that hypothesis. Leatherbacks routinely occur in
northern waters off Newfoundland where they occupy water with temperatures from
0 to 15°C (mean = 12.6°C).'^'^ Frair et al. recorded a T^ of 25.5°C for a captive
leatherback in 7.5°C seawater. This suggests that leatherbacks can thermoregulate
in cold water, although the effect of thermal inertia in this particular turtle is
unknown.
We measured the metabolic rates of six adult female leatherbacks while they
were resting, walking, and covering their nests by analyzing expired air with gas
analyzers (Figure 11.8).^^ Metabolic rates for leatherbacks at rest were above those
predicted by allometric relationships for green turtles and other reptiles scaled to
leatherback size (0.387 W kg“^ 1.15 ml O2 kg~‘ min ^^ but half the values predicted
for mammals of this size. Lutcavage et al."^^ reported similar metabolic rates (1.1 ml
O2 kg”*min”*) for leatherbacks using another method (difference in gas concentration
during rebreathing). It is interesting to note that hatchling leatherbacks also have a
metabolic rate (0.286 1 kg”*h”*= 4.77 ml kg”*min *) three times that of green turtle
and loggerhead hatchlings."*^ Leatherback hatchlings swam continuously in the
respirometer and maintained a high level of activity for hours. They may have a
higher routine (noimal) metabolic rate than the other species, as do the adults.
2.0
1.5-
no
CO
©
= 3
1.0
^1
CO ^
Vi
m
CO
0.5 +
■f
0 .0 - — i
200 250 300 350 400 450
Mass (kg)
FIGURE 11.8 Mass specific metabolic rates for nesting leatherback turtles at Tortuguero,
Costa Rica. We measured metabolic rate for turtles at rest (crosses), while they were covering
their nests (solid squares), and while they were walking on the beach (open circles). We
collected expired air in large Douglas bags and measured the Oj and CO2 content of that air
with a Beckman C-2 gas analyzer and Scholander 0.5-ml gas analyzer. The lowest values of
active metabolism were for turtles when they were covering their nests with their hind flippers
after laying their eggs. The highest values were for turtles vigorously throwing sand with
hind and foreflippers, crawling, and rotating the body from side to side while covering the
nest with sand. (From Paladino, F. V., O’Connor, M. R, and Spotila, J. 344, 858,
We*** assessed the thermoregulatory capabilities of leatherbacks by mathemati

cally analyzing heat exchange within the turtle, and between the turtle and the
environment. As body size increases, resting metabolism can maintain increasingly
larger core-skin temperature differences. Given their large adipose tissue layer (6 to
7 cm)"*"* and their large body size, leatherbacks can use changes in blood flow to the
skin and periphery to regulate body temperature such that they maintain warm
Thermal Biology 311
temperatures in the North Atlantic and avoid overheating in warni tropical waters.
When leatherbacks are nesting it is typical to observe changes in color of their soft
skin. When they crawl out on the beach their skin is pale. After nesting their skin
is bright pink. We measured blood flow in the skin (unpublished data) and found
that blood flow when the skin appears pink is more than 10 times higher than when
the skin is pale. This increased blood flow helps to cool the turtles by bypassing the
insulation and bringing heat to the surface where it can be transferred to the cool
night air. This strategy would be ineffective during the day when T^ is much higher
(see above). Thus, while the leatherback has a large enough body size to dampen
the effects of solar heating if it nested during the day, the high T^ at its surface would
make thermoregulatory strategy of the leatherback of changing blood flow to the
surface to regulate T^ ineffective. Thus, while it could avoid overheating from
external heat load due to its themial inertia, it could not get rid of the excess heat
generated within its body during the nesting process. Here again, biophysical con
straints would combine with characteristics of the anatomy and physiology of the
sea turtle to detemiine its behavior.
11.7 CONCLUSIONS
The thermal biology of sea turtles is determined by the interaction of the biophysical
constraints imposed by the environment with the anatomical, physiological, and
behavioral characteristics of the animals. We can obtain an understanding of the
thermal constraints on these turtles by considering the limitations imposed by heat
energy exchange with the environment. Judicious use of biophysical models can
help to pose useful questions and to answer them. On land, sea turtles are affected
by rapid heat gain during the day, and leatherbacks are also affected by the inability
to dump metabolic heat during nesting activity. Heat gain and the inability to lose
heat prevent large species of sea turtles from nesting during the heat of the day. In
water, heat gain and loss are very rapid and T^ tend to remain close to T^. Larger
sea turtles such as loggerheads and green turtles can maintain a small (1 to 2°C)
temperature difference between the body core and the water. Green turtles when
swimming vigorously can be as much as 7°C above T^. This endothermic capacity
is even greater in leatherback turtles that can be active in cold northern waters that
have temperatures between 0 and 15°C. They use a combination of large body size,
thick insulation, an elevated metabolism, and changes in blood flow to remain
wanner than the water in cold oceans and to avoid overheating in tropical waters
and on the nesting beach at night. Their strategy of changing blood flow to dump
heat on the nesting beach is ineffective during the heat of the day, because the T^
of their skin is too high to allow heat loss by convection. As we stated in 1985,
additional experiments are still needed to elucidate the thermal biology of sea turtles.
These include such simple measurements as determination of the preferred T^ and
CTMs of these animals, as well as more sophisticated measurements of the thermal
biology of sea turtles free ranging in the ocean. With the increased availability of
sophisticated sensors, recorders, and telemetry, such measurements can now be a
reality.
R EFEREN CES
Spotila, J. R. and Standora, E. A., Environmental constraints on the thermal energetics

of sea turtles, Copeia, 1985, 694, 1985.
2. Spotila, J. R., Metabolism, physiology, and thermoregulation, in Biology and Con
servation o f Sea Turtles, Rev. ed., Bjomdal, K. A., Ed., Smithsonian Institution Press,
Mrosovsky, N., Thermal biology of sea turtles. Am. ZooL, 20, 531, 1980.
Porter, W. P. and Gates, D. M., Thermodynamic equilibria of animals with environ
ment, EcoL Monogr., 39, 245, 1969.
Erskine, D. J. and Spotila, J. R., Heat-energy-budget analysis and heat transfer in the
largemouth blackbass (Micropterus salmoides), Physiol. ZooL, 50, 157, 1977.
Crawford, K. M., Spotila, J. R., and Standora, E. A., Operative environmental tem
peratures and basking behavior of the turtle, Pseudemys scripta. Ecology, 64, 989,
1983.
Gates, D. M., Energy Exchange in the Biosphere, Harper and Row, New York, 1962.
Gates, D. M., Biophysical Ecology, Springer-Verlag, New York, 1980.
Spotila, J. R., O’Connor, M. R, and Bakken, G. S., Biophysics of heat and mass
transfer, in Environmental Physiology o f the Amphibians, Feder, M. E. and Burggren,
W. W., Eds., University of Chicago Press, Chicago, 1992, 59.
10. Spotila, J. R., Soule, O. H., and Gates, D. M., The biophysical ecology of the alligator:
heat energy budgets and climate spaces, Eco/ogy, 53, 1094, 1972.
11. Colbert, E. H., Cowles, R. B., and Bogert, C. M., Temperature tolerances in the
American alligator and their bearing on the habits, evolution and extinction of the
dinosaurs. Bull. Am. Mus. Nat. Hist., 8 6 , 327, 1946.
12. Bakken, G. S., and Gates, D. M., Heat-transfer analysis of animals: some implications
for field ecology, physiology, and evolution, in Perspectives in Biophysical Ecology,
Gates, D. M. and Schmerl, R., Eds., Springer-Verlag, New York, 1975, 255.
13. Bakken, G. S., A two-dimensional operative-temperature model for thermal energy
management by animals, J. Thermal Biol., 6 , 23, 1981.
14. Bakken, G. S., Santee, W. R., and Erskine, D. E., Operative and standard operative
temperatures: tools for thermal energetics studies. Am. ZooL, 25, 933, 1985.
15. Standora, E. A., Spotila, J. R., and Foley, R. E., Regional endothermy in the sea turtle,
Chelonia mydas, J. Thermal Biol., 1, 159, 1982.
16. Whittow, G. C. and Balazs, G. H., Basking behavior of the Hawaiian green turtle
{Chelonia mydas), Pac. Sci., 36, 129, 1982.
17. Bakken, G. S., Measurement and application of operative and standard operative
temperatures in ecology. Am. ZooL, 32, 194, 1992.
18. White, F. M., Heat Transfer, Addison-Wesley, Reading, MA, 1984, chap. 4.
19. Paladino, F. V, O’Connor, M. R, and Spotila, J. R., Metabolism of leatherback turtles,
gigantothermy, and thermoregulation of dinosaurs. Nature, 344, 858, 1990.
20. Spotila, J. R., Lommen, P. W., Bakken, G. S., and Gates, D. M., A mathematical
model for body temperatures of large reptiles: implications for dinosaur ecology,Am.
Nat., 107, 391, 1973.
21. Dunham, A. E., Overall, K. L., Porter, W. R, and Forster, C. A., Implications of
ecological energetics and biophysical and developmental constraints for life-history
variation in dinosaurs, in Paleobiology o f the Dinosaurs, Farlow, J. O., Ed., Geolog
ical Society of America Special Paper 238, Boulder, CO, 1989, 1.
22. Spotila, J. R., O’Connor, M. R, Dodson, R, and Raladino, F. V, Hot and cold running
dinosaurs: body size, metabolism and migration, Mo<7. Geol., 16, 203, 1991.
Thermal Biology 313
23. Bakken, G. S., An improved method for determining thermal conductance and equi
librium body temperature with cooling curve experiments,/. Therm. Biol., 1, 169,
1976.
24. Balazs, G. H., Green turtle migrations in the Hawaiian archipelago,^//)/. Conserv.,
9, 125, 1976.
25. Balazs, G. H., Synopsis of biological data on the green turtle in the Hawaiian Islands,
NOAA Tech. Memo. NMFS, NOAA-TM-NMFS-SWFC-7, Honolulu, Hawaii, 1980.
26. Balazs, G. H. and Ross, E., Observations on the basking habit in the captive juvenile
Pacific green turtle, Copeia, 1974, 542, 1974.
27. Bustard, R., Sea Turtles, Natural History and Conservation, Taplinger Publishing,
New York, 1973.
28. Fritts, T. H., Marine turtles of the Galapagos Islands and adjacent areas of the eastern
Pacific on the basis of observations made by J. R. Selvin 1905-1906,/. HerpetoL,
15, 293, 1981.
29. Paladino, F. V., Spotila, J. R., O’Connor, M. R, and Gatten, R. E., Jr., Respiratory
physiology of leatherback turtles {Dermochelys coriacea) while resting, laying eggs,
and active on land, Chelonian Conservation and Biology, in press.
30. Prange, H. D., Energetics of swimming of a sea turtle,/. Exp. Biol., 64, 1, 1976.
31. Prange, H. D. and Jackson, D. C., Ventilation, gas exchange and metabolic scaling
of a sea turtle, Respir. Physiol., 27, 369, 1976.
32. Jackson, D. C. and Prange, H. D., Ventilation and gas exchange during rest and
exercise in adult green sea turtles,/. Comp. Physiol., 134, 315, 1979.
33. Naito, Y., Sakamoto, W., Uchida, L, Kureha, K., and Ebisawa, T., Estimation of
migration route of the loggerhead turtle Caretta caretta around the nesting ground,
Nippon Suisan Gakkaishi, 56, 255, 1990.
34. Sakamoto, W., Uchida, I., Naito, Y, Kureha, K., Tujimura, M., and Sato, K., Deep
diving behavior of the loggerhead turtle near the frontal zone, Nippon Suisan Gak
kaishi, 56, 1435, 1990.
35. Sakamoto, W., Naito, Y, and Uchida, I., Measurement of the migration route of the
loggerhead turtle by Argos and a biorecorder Biotelemetry, XI, 112, 1990.
36. Sakamoto, W. and Yuzuriha, H., Tracking free-swimming marine animals, S£?a Tech-
noL, 34, 25, 1993.
37. Sato, K., Sakamoto, W., Matsuzawa, Y, Tanaka, H., Minamikawa, S., and Naito, Y,
Body temperature independence of solar radiation in free-ranging loggerhead turtles,
Caretta caretta, during intemesting periods. Mar. B iol, 123, 197, 1995.
38 Sapsford, C. W. and van der Riet, M., Uptake of solar radiation by the sea turtle,
Caretta caretta, during voluntary surface basking. Comp. Biochem. Physiol., 63A,
471, 1979.
39. Neill, W. H. and Stevens, E. D., Thennal inertia versus thermoregulation in “warm”
turtles and tunas, Science, 184, 1008, 1974.
40 Witherington, B. E. and Ehrhart, L. M., Hypothermic stunning and mortality of marine
turtles in the Indian River Lagoon system, Florida, Copeia, 1989, 696, 1989.
41. Morreale, S. J., Meylan, A. B., Sadove, S. S., and Standora, E. A., Annual occurrence
and winter mortality of marine turtles in New York waters,/. HerpetoL, 26, 301,1992.
42, Hochachka, P. W. and Dunn, J. F, Protecting cells and tissues against hypoxia and
hypothermia, in Hypoxia and Cold, Sutton, J. R., Houston, C. S., and Coates, G.,
Eds., Praeger, New York, 1987, 57.
43 . Hutchison, V. H. and Dupre, Thermoregulation, in Environmental Physiology o f the
Amphibians, Feder, M. E. and Burggren, W. W., Eds., University of Chicago Press,
Chicago, 1992, 206.
44. Goff, G. P. and Lien, J., Atlantic leatherback iuxiX^^.Dermochelys coriacea, in cold
water off Newfoundland and Labrador, Can. Field Nat., 102, 1, 1988.
45. Frair, W., Ackman, R. G., and Mrosovsky, N., Body temperature of Derrnochelys
coriacea: warm turtle from cold water, Science, 177, 791, 1972.
metabolism in the leatherback sea turtle, Can. J. Zool., 70, 348, 1992.
47. Lutcavage, M. E. and Lutz, P. L., Metabolic rate and food energy requirements of
the leatherback sea turtle, Derrnochelys coriacea, Copeia, 1986, 796, 1986.
1 2 Hormones in the Life
History of Sea Turtles
David W. Owens
CO NTENTS
12.1 Introduction................................................................................................. 315

12.2 Eggs and Hatchlings.................................................................................. 316
12.2.1 Sex Determination........................................................................ 316
12.2.2 Corticosterone and the Imprinting Hypothesis............................317
12.3 Juveniles and Subadults.............................................................................. 317
12.3.1 Testosterone and the Sexing Technique.......................................319
12.3.2 Testosterone and Secondary Sex Characteristics.........................321
12.4 Endocrine Patterns During Reproduction...................................................322
12.4.1 Fem ales........................................................................................ 324
12.4.1.1 Off Season.................................................................... 324
12.4.1.2 Migration and Courtship..............................................324
12.4.1.3 Nesting.......................................................................... 326
12.4.2 Males............................................................................................. 329
12.4.2.1 Off Season and Annual Cycles............. 329
12.4.2.2 Migration and Courtship..............................................330
12.5 Pituitary Hormone Studies.......................................................................... 331
12.6 Metabolism and Stress................................................................................ 332
12.6.1 Thyroid Metabolism..................................................................... 332
12.6.2 Corticosterone, Stress, and Activity.............................................333
12.7 Conclusion and Summary........................................................................... 334
12.7.1 Summary of Practical Observations.............................................335
12.7.2 Summary of Interesting Observations andQuestions................... 336
Acknowledgments................................................................................................. 336
References............................................................................................................. 337
The presence of a chapter on hormones in a sea turtle book may surprise some
readers. As it turns out, the initiation of sea turtle farming on Grand Cayman Island
around 1969 was the initial stimulus for a long series of endocrine studies on sea
0-8493-8422-2/97/$0.00+$.50
turtles. Paul Licht of the University of California at Berkeley and Harold Papkoff
at the University of California at San Francisco were looking for a reptilian model
to be used in studies of the vertebrate pituitary gland. The project utilized hundreds
of fresh pituitaries from green sea turtles {Chelonia mydas) for purification of their
peptide hormones, which were then compared with many other species for structural
and biological properties.Subsequently, in an effort to improve captive breeding
potential, the Cayman turtle farm also supported many on-site endocrinological and
behavioral studies of green turtles^'^ and Kemp’s ridleys (Lepidochelys kempi)d'^
Because of this extensive record of endocrine study on sea turtles (see reviews by
Owens and Morris^ and Licht^^) they are generally the best known model for the
reptilian endocrine system. Indeed, because of the development and use of radioim
munoassays derived from sea turtle pituitary hormones,^^ we have also learned a
great deal about many other chelonian species.
While the focus of earlier reviews"^-^^ has been on the research done in the
laboratory and with captive populations, this summary will concentrate on the work
of the past decade which has been directed toward studies of natural populations as
well as species other than the green sea turtle. Sea turtles have been rather easy to
study during the brief period when the females emerge to nest or while the hatchlings
are developing and hatching. On the other hand, the remaining, completely marine
life stages constitute more than 99% of a sea turtle’s very long life. At these times
sea turtles are usually difficult to observe for any extended period. In addition, when
they are captured from the sea, it is often problematical to determine their sex, age,
and behavior/reproductive condition. For these reasons we have previously espoused
the use of endocrine evaluations, laparoscopy,^^ and more recently, ultrasound
imaging‘s to diagnose the specific life stage of the individual turtle. For example,
this type of diagnosis has proven very useful for choosing animals at specific times
in their reproductive cycle for use in satellite transmitter tracking studies.
12.2 EGGS AND HATCHLINGS

12.2.1 S ex D etermination
In sea turtles, all of which are thought to show temperature-dependent sex determi
nation (TSD), the developing gonad itself does not appear to give the signal for its
own differentiation into a testis or ovary. Merchant-Larios and Villalpando found
that if they incubated olive ridley embryos at lower, male-producing temperatures,
then removed the gonad from this embryo, it continued to develop as a male (testis)
even when the isolated gonad was warmed to temperatures which would normally
produce an ovary. The same thing happened in a temperature shift in the opposite
direction, with the ovary continuing to develop even when removed and cooled to
temperatures which would normally produce males. Thus, these authors suggest that
an externally derived factor or factors moves into the developing gonadal tissue
during the sensitive period to induce the normal differentiation. Desvages et al.^^
have proposed that the P-450 arom gene may be the key transcription step in that
it would somehow be thermally sensitive and produce the critical enzyme aromatase,
which in turn invades the gonad and converts available androgen steroid substrate
Hormones in the Life History of Sea Turtles 317
to estrogens. Using leatherback eggs, these researchers found higher levels of aro
matase at 30.5°C than at 27°C and only observed estrogen production at the higher
temperature. Their model basically states that if the higher female-producing tem
peratures do not occur in an individual, there will be no aromatase, no estrogens,
and the embryo will remain a male.
Additional support for the aromatase-estrogen theory for female differentiation
in sea turtles is from the work of Crain et al."° and Gross et al.,^^ who measured
steroids in the chorioallantoic/amniotic fluid of hatchling loggerheads. They found
the ratio of estrogen to testosterone is higher in hatchlings raised at warmer tem
peratures. They also proposed the use of this ratio as a noninvasive sexing technique.
In the search for the primary genetic mechanism for TSD, heat shock proteins have
also been suggested for the transduction of environmental temperature.
12.2.2 C orticosterone and the Imprinting H ypothesis

Birds appear to be able to imprint optimally when their corticosterone levels are
low.^"^ The working hypothesis states that the hypothalamus-adrenal axis is sup
pressed during this “critical period”. Morris^^ first looked at the corticosterone (B)
levels in hatchling sea turtles to determine how various stressful events in their early
life might be related to the hypothetical critical period for imprinting suspected for
sea turtles. Her data showed a dramatic peak in corticosterone, and presumably
stress, associated with piping and escaping the egg, as well as a lesser peak as the
hatchling first enters the water. In addition, relatively low corticosterone was seen
just prior to emergence from the nest and after several days in the water. Morris^^
felt that these low corticosterone episodes would be the most likely times for
olfactory or other types of imprinting to be occurring in the hatchlings. This result
was distinctive enough that Schwantes^^ felt it should be repeated, with some addi
tional sampling times. She did this several years later in the same species (Figure
12.1). She confirmed the results of Moitìs^^ and, in an additional group of turtles
sampled seven d prior to hatching, noted that there were very low corticosterone
levels near the end of embryonic development prior to the corticosterone surge at
hatching. Thus, if corticosterone is negatively correlated with the critical period, the
most likely times for imprinting would be just prior to hatch and at 4 to 5 d of age
as the turtle enters the water. With good evidence for olfactory imprinting capabilities
in sea tu rtle s,a n d increasing molecular genetics data supporting the imprinting
hypothesis,^^ it is important for conservationists to consider the needs of hatchlings
during the presumed imprinting times in the first few weeks of life of a turtle.
12.3 JU V E N ILE S A N D S U B A D U L T S
In the transitions which occur from hatchling to adult life, sea turtles grow in size
at up to four orders of magnitude from their hatching weight and spend an estimated
10 to 50 years in the process, depending on the species and the quality of the foraging
habitat. Indeed, sea turtles change so much in size that they require different habitats
and have different niche characteristics, depending on their size. Sea turtle scientists
have come to describe these different situations as developmental habitats.
FIGURE 12.1 Corticosterone in loggerhead hatchlings shows two pronounced peaks. The
first occurs at hatching and the second when the turtle first enters the water at four to five d
of age. It is at the low points that imprinting may be occurring. Different letters above bars
represent significant differences in means for a time using repeated measures ANOVA. (From
Schwantes, N., M.S. thesis, Texas A&M University, College Station, 1986. With permission.)
Although the supporting data are circumstantial at this time, we propose that dis
tinctive endocrine changes accompany and potentially regulate some of these devel
opmental habitat changes. Initially, as discussed above, B surges during the transi
tions from egg to hatchling as well as subsequently as the hatchling emerges from
the nest and migrates down to the water (Figure 12.1). During the pelagic phase we
know nothing about sea turtle homiones; however, post-pelagic immatures show
measurable testosterone with males having a seasonality and apparently temperature
dependence correlated to their testosterone levels (Figure 12.2). Is it possible that
this spring and summer surge in testosterone could be inducing turtles to move to
better or different foraging grounds? Indeed, seasonally speaking, when do sea turtles
make habitat changes? One certain time is in the cooling months when many tracking
studies have shown that turtles will leave for warmer areas.
According to Limpus,^^ sea turtles have a rather lengthy subadult period. In this
discussion we are equating subadult with the period of pubertal changes. It is during
this time that we first see the transition away from a temperature-related (controlled?)
androgen cycle to what we believe is a photoperiod-related (controlled?) cycle.
Figure 12.3 shows the dramatic early season peak in testosterone well prior to a
modest peak shown in immatures (Figure 12.2). We further propose that the initiation
of this photoperiod-related testosterone surge in subadults is also the time for the
I
FIGURE 12.2 Circulating testosterone levels in immature male Kemp’s ridleys at Sea
Arama, Galveston showing clear temperature influence at this age. (From Morris, Y., M.S.
thesis, Texas A&M University, College Station, 1982. With permission.)
transition to the final developmental foraging habitat where they will remain and
from which they will initiate their reproductive migrations. Sea turtles do have a
well-developed photosensitive pineal gland which produces the hormone melatonin,
which has been shown to regulate reproductive cycles in many vertebrate species.
We have previously reviewed this literature.
12,3.1 T estosterone a nd the S exing T echnique
The sex of immature sea turtles is difficult to determine with external evaluations.
For several years we have been using the higher testosterone levels seen in immature
males as a sex-distinguishing technique for generating sex ratios. Because of TSD
in sea turtles, it has been important to generate sex ratios for use in population
modeling as well as to attempt to explain in evolutionary terms the biological
importance of TSD systems generally. Table 12.1 summarizes some of the published
Time of year
FIGURE 12.3 Circulating testosterone from 80 adult loggerheads netted from Cape Canav
eral Channel between January and August. Note apparent separation into reproductively active
and inactive subsets. (From Wibbels, T., Owens, D. W., Morris, Y. A., and Amoss, M., in
Ecology o f East Elorida Sea Turtles, Witzell, Ed., U.S. Department of Commerce, NOAA
Tech. Rep., NMFS-53, Miami, FL, 1987. With permission.)
observations to date where testosterone has been used successfully to sex sea turtles.
It has been our practice when possible to use laparoscopic examination to indepen
dently verify the hormone technique. In most cases to date the laparoscopy has
proven the effectiveness of the testosterone-based system (Table 12.1). In two exper
iments, one by Schroeder and Owens^^ and a second by Coyne and Landry,
however, the testosterone technique produced ambiguous results. In the first case,
cold stunned greens, rescued from Mosquito Lagoon, FI, were transported to Orlando
where they were maintained for six weeks in tanks at 25 to 26°C. The testosterone
levels derived from blood samples taken after the six weeks of recovery did not
show the normal bimodal set of peaks, with females having a low peak at 0 to 20
pg/ml and males a higher peak above 30 pg/m\P Because the authors were able to
determine the sex of 66 other turtles which had died in the same cold-stunning event,
a good approximation of the expected sex ratio was available (1.77F:1.00M). Since
the hormone levels did not conform to this ratio, these authors felt the winter warming
of the surviving turtles may have caused an unusual increase in testosterone secretion
which smoothed out the expected bimodal curve, rendering the sexing technique
useless. Captive stress may also have been involved, since Coyne and Landry’s
unpublished work^"^ on Kemp’s ridleys indicates that the initial sample did predict
sex well, but that subsequent samples, taken over the next several hours, showed
increased variability in testosterone levels. Since captive stress is well known to
cause alterations in testosterone and since testosterone is temperature sensitive^“^ in
sea turtles, it is strongly recommended that all blood samples be taken as soon as
possible (within 15 min) after capture of the turtles. Since other blood-transported
TABLE 12.1
Sex Ratios of Immature Sea Turtle Populations Based on Testosterone Levels
Species #/Sex T pg/ml Ratio F:M laparoscopy Research site
C a r e tta c a r e tta 99F 5-30.1 1.74:1.00 22 Cape Canaveral, FL'^^

57M 76.3-560
lOU 30.1-76.3
45F 0-30 3.21:1.00 No Hutchinson Is., FL^^
14M 40-up
2U 30-40
14F 0-30 1.40:1.00 No Indian River, FL^^
lOM 40-up
OU 30-^0
IIF 0-30 1.96:1.00 No Chesapeake Bay, VA^^
6M 40-up
4U 30-40
21F 4.1-24 1.00:3.14 All Heron Is., Australia^^
66M 25.2-180
163F 0-30 2.67:1.00 No Hutchinson Is. FL^^
61M 40-up
148F 0-30 2.10:1.00 No Hutchinson Is., FH^^
70M 40-up
5U 30-40
C h e io n ia m y d a s 133F not given 2.00:1.00 All Heron Is., Australia^^
67M
65F 0-10 1.40:1.00 No Bahamas^^'^
46M 20-up
9U 10-20
32F 0-20 0.96:1.00 No Hawaipo
31M 30-up
3U 20-30
E r e tm o c h e ly s 21F 0-10 4.20:1.00 All Heron Is., Australia^*^
im b r ic a ta 5M 20-up
41F 0-16.1 1.00:1.41 14 Mona Is., Puerto
58M 30-up Rico"^*
lU 16.1-30
factors may also be impacted by stress, this recommendation is relavent to all types
of physiological work on turtles.
12.3.2 T estosterone a nd S econdary S ex C haracteristics
As mentioned above, juvenile and subadult males begin to secrete more and more
testosterone as they age. Morris^^ observed that there is a positive correlation with
ambient temperature and testosterone in these young animals. Early testosterone
injection studies at the Cayman Island Turtle Farm showed that both the tail and
enclosed penis elongate predictably when this hormone or gonadotropin is
administered.^ In addition, testosterone and gonadotropin stimulate courtship and

mating behavior in immature animals. Two other male secondary sexual character
istics, the curved front claws which are used to grip the females when mating and
the softened plastrons,are probably under the control of testosterone. In Figure
12.4 we show the curving claw, elongating tail, and softening central plastron area
of a 99-cm curved carapace length (ccl) subadult male loggerhead. For comparison,
a mature adult Kemp’s ridley is also shown with his well-developed tail and soft
plastron. The anatomical function of the softened male plastron has not been inves
tigated. We have noticed in the field with olive ridleys that the hard, domed carapace
of a mounted female fits securely into the softened medial plastron area of the mating
male; thus it may be serving to increase gripping capability. Since the softened
plastron of the male is also well vascularized (it often turns pink in captured olive
ridley males), it may also serve a sensory function in mating. At this time we have
also been unable to evaluate the potential for annual cyclic changes in the softening
of the plastron as the annual androgen cycle fluctuates."^^
Females do not appear to have any secondary sexual characteristics. There is
one type of mutilation scaring, however, which is typical of the adult females of
some populations of hard-shelled sea turtles. When males mount and cling to the
carapace of the female during courtship and mating, they often dig into the anterior
lateral edges of the carapace of the female with their flipper claws. In the short term,
a scab is apparent on each side of the female, while over the years a shallow
depression or even a pronounced notch can be produced. We suspect that the scab
could be an indicator of time since last mating.
12.4 E N D O C R IN E PATTERN S D U R IN G
R E P R O D U C T IO N
Ecological modelers of sea turtle populations require accurate numbers on sex ratios,
percent of animals reproductively active each year, and age to and duration of
reproductive life history. For several years we have been attempting to use hormone
levels as an indirect way to determine the reproductive and behavioral status of the
individual sea turtle. There are many reasons why this would be a useful
capability in these species. In the first place, female and possibly some male sea
turtles are not always reproductively active each year. While we have found that the
ridley species often reproduce in sequential years, they do also regularly skip years.
The larger sea turtle species rarely show repeat nesting in the next season, with the
exception of captive maintained turtles.Thus, when we see a turtle in the water it
is not usually clear what its reproductive state may be. Specifically, time of year or
the location where the turtle is captured does not always predict reproductive con
dition as it might in many species. Particularly in areas where both feeding and
reproductive populations coexist in the same habitat (Heron Island, Australia being
a good example), reproductively mixed populations can be found. On the other hand,
all of the turtles at a place like Ascension Island or Tortuguero, Costa Rica are
thought to be reproductively active since there are limited foraging grounds near
these nesting beaches.
FIGURE 12.4 (a, b, c) Photos of a 99-cm cci subadult loggerhead male: (a) shows the front
flipper claw is starting to curve; (b) the tail is clearly elongating, but will get much longer;
(c) the midline of the plastron is softening and has no keratin (a black line marks the soft
area), (d) An adult Kemp’s ridley is shown with fully dekeratinized plastron and a fully
developed tail.
Another reason why the hormone indicators can prove useful in adults is that
there are no clear anatomical features which relate directly to reproductive condition.
As far as we know there are no seasonal or reproductive color changes, or modifi
cations in secondary sex characteristics. There is, however, one possible exception
which needs to be examined more closely. This is the softening of the plastron of
adult males, as discussed above."^^
An additional reason for using hormone indicators of reproductive status is the
confusion that commonly occurs between very large subadults and adults. As
Limpus^^ has noted, one can locate some very large individuals which are still not
reproductively active, and indeed they may be several years from adulthood. In some
cases it is even possible to misidentify a large subadult male as a female due to his
lack of secondary sexual characteristics. Limpus^^ was able to describe the occur
rence of this situation by employing laparoscopic surgery.
A final consideration is that laparoscopic surgery, followed by biopsies where
appropriate (in males), is certainly an effective way to determine reproductive con
dition of both males and females. However, this is an involved technique which
requires specialized training, considerable field equipment, and some risk to the
turtle. Eventually it is hoped that a blood sample, which is quickly and easily
obtained for hormone analysis, will suffice for a full determination of the individuaTs
reproductive condition.
12.4.1 Females
The endocrinology of reproductive adult females is better studied than other sea
turtle life stages because it was the emphasis of the early work at the Cayman turtle
Farm as well as subsequent field projects with olive ridleys, greens, and loggerheads.
Crawling, digging, and nesting females are also very obvious, accessible, and easily
bled for hormone analyses.
12.4.1.1 Off Season
At Heron Island, Wibbels et al."^^ used the rodeo technique to collect from 4 to 13
reproductively inactive adult females per month between July and November. Several
of these turtles were known to have nested 1 or 2 years previously, they all had
atrophic or early growth ovaries upon laparoscopic examination, and their hormone
levels showed no significant seasonal changes. On the other hand, the ovaries of the
10 to 20% of the captured females which were reproductively active at this time
showed dramatic follicular growth with hundreds of approximately 1.5-cm-diameter
follicles and significant increases in circulating hormones.
12.4.1.2 Migration and Courtship
When we compared hormone levels of reproductively active vs. inactive loggerhead

females at Heron Island in the winter (July/August), there were no significant
differences; however, in the spring both estrogen and testosterone were much higher
in the active group, even while all the turtles were still feeding and remained on the
foraging grounds. In other words, the initial increases in steroid levels seen in the
reproductive subset did not cause them to depart immediately on their migration.
Thus, it appears that the general increase in circulating steroids we observed in the
spring is not the stimulus for migration. On the other hand, just prior to the departure
of the reproductively active loggerheads from the Heron Island feeding ground, we
did observe a dramatic and consistent drop in circulating estrogen as testosterone
went to its highest level up to that time in the year (Figures 12.5 and 12.6). At the
Cayman turtle farm during the mating season (they were obviously not able to
migrate) this same dip in estrogen and surge in testosterone was observed 8 years
earlier in green turtles.^ A particular challenge with green turtles is that estrogen
seems to circulate at lower levels than it does in other sea turtles.Regardless of
this potential species-specific difference, it is clear the circulating testosterone surges
while estrogen drops off at the time of migration in the reproductively active female.
o
02
W
H
cn
o
H
(n
FIGURE 12.5 Circulating hormone levels for a single turtle indicate a sudden increase in
estradiol followed by a shaip drop just as testosterone makes a pronounced surge. As the
testosterone peaked the turtle migrated away from the feeding ground. Also note the relatively
low estradiol compared to testosterone levels. (From Wibbels, T. et al., Owens, D. W., Limpus,
C., Reed, P., and Amoss, M., Gen. Comp. Endocrinol, 79, 154, 1990. With permission.)
At the Cayman turtle fami, greens and Kemp’s ridleys followed the chronology
of mating as depicted in Figure 12.6 with a discrete receptivity period about 28 d
prior to the first nest and 14 d prior to the first ovulation. This same pattern has been
observed for greens and loggerheads in Australia.^^ '^^ Female black turtles (C. agaS'
sizi) may not conform to this chronology based on observations by Alvarado and
Figueroa,"^^ Owens et al.,'^'^ and Rostal et al.“^^ Similarly, the chronological sequence
of Migration — Mating — Ovulation — Nesting may not be a constant for olive
ridleys either, since we have seen several individuals at Nancite, Costa Rica which
did not appear to conform to this pattern. For some time we have been testing the
hypothesis that testosterone levels in females receptive to mating will be at their
season high. While this idea seemed to work at the Cayman turtle farm, and was
1 2 3 4
REPRODUCTIVE SEASON (MONTHS)
FIGURE 12.6 A general model for endocrine control and correlations with reproductive
behavior in male (top) and female (bottom) sea turtles. The dashed boxes indicate behavioral
components which are poorly understood. In this example the female ovulated and nested
three times. The hormone curves shown have been placed on a common scale for relative
comparisons of the timing of changes. The maximum typical level in picograms or nanograms
of circulating homione is indicated for each curve.
generally supported by the loggerhead data from Heron Island and the olive ridley
study at Nancite (Figure 12.6), more work needs to be done on this interesting
question. The primary difficulty has been in confirming true mating events as
opposed to mounted pairs which were not actually copulating. This later situation
appears to be much more common than we anticipated (personal observations). We
suggest that if the hypothesized testosterone peak for mating receptivity holds up,
it could prove useful in understanding much about sea turtle mating systems. For
example, the group of mating C. agassizi observed by Rostal et al."^^ in Michoacan
did not have significantly elevated testosterone compared to nesters. Were they truly
mating? It is possible that the females were not in their receptive period, but were
simply caught off guard by the very aggressive males in this species.
12.4.1.3 Nesting
The physiological and behavioral events of nesting in sea turtles are quite dynamic
and surprisingly fast considering the size and reptilian ancestry of the animal. In
Figure 12.6 we have summarized several studies^’^^'^^'^^^’'^^ to track many of the
endocrine aspects of the prenesting and nesting process. Licht and associates^' first
developed and applied homologous radioimmunoassays for protein hormones in a

reptile using the green sea turtle. This information has been reviewed previously.
An LH and progesterone surge occurs at the first ovulation 10 to 14 d prior to the
first clutch and within 48 h after nesting for each successive clutch. This same, very
dynamic set of hormone peaks has now been observed for loggerheads, greens, and'
both ridley species.
When the original pituitary hormone purifications were done for structure studies
and to construct the radioimmunoassays, a protein contaminant of about the same
size as the FSH molecule was inadvertently included with the material injected into
the rabbits."^^ When the rabbits made antibodies to the FSH, it turned out that they
also made more abundant antibodies to the contaminant protein, which, unfortu
nately, dominated the assay. Interestingly, this contaminant showed a peak level in
the blood just at the time of nesting, and only very briefly then. Based on molecular
weight and other characteristics, Licht and colleagues^^ thought the molecule could
be a neurophysin-like protein which is known to be secreted with the octapeptide
hormones (possibly as protective carriers) of mammals. Figler^^ became interested
in one of these small hormones, arginine vasotocin (AVT), for his thesis work
because he thought it might be secreted by the impressive pineal gland of the sea
turtle.^“ Figler^^ found that the pineal had very little AVT, but at the same time
observed a pronounced peak of AVT during nesting. Collaborating with Licht, he
found that for loggerheads and olive ridleys, both AVT and neurophysin (formerly
thought to be FSH) reach their peak in the circulation just as the first egg is dropped
from the cloaca.^^ Thus it is thought that this reptilian neurophysin is secreted
together with AVT from the neurohypophysis and possibly serves to protect the AVT
during the early phase of its passage from the brain to the oviduct. In the bloodstream
the two molecules appear to separate. Since the neurophysin is a larger protein
molecule, it seems to take longer to be catabolized in the blood, thus explaining its
persistence in the blood even as the turtles reenter the water (Figure 12.7). AVT was
also shown by Figler^“^ to increase contractility of the oviduct of the olive ridley.
Guillette and associates^'’ have also studied the endocrine control of the nesting
process in loggerheads. Like Figler, they carefully monitored behavior and the
nesting process to obtain a very precise set of samples taken at close intervals. They
showed the dynamic nature of prostaglandins F (PGF) and E2 (PGE2) (Figure 12.8)
and hypothesized that AVT may have induced the synthesis of PGF in the oviduct.
PGF, like AVT, would have roles in oviduct contraction as has been shown in several
other species. Guillette et al.^^ also hypothesized that PGE2 would have a role in
vaginal/cervical relaxation. These several small hormones no doubt work in concert
to facilitate contractility and rapid egg transport out of the oviduct into the nest.
Testosterone is also of interest and potential use in the nesting female. Wibbels
et al.^^ noticed that with each successive clutch, the loggerhead female shows a
distinct stepwise drop in her circulating testosterone (Figure 12.6). Since the source
of this testosterone is most likely to be the mature follicles in the ovary, we suspect
that each drop is due to the sequential removal of follicles from the ovary with each
ovulation. When all the follicles have ovulated, the testosterone reaches basal levels.
There are some fluctuations in the stepwise drops with species which produce greater
numbers of clutches; however, in the ridley species, testosterone levels appear to
FIGURE 12.7 Circulating levels of neurophysin and arginine vasotocin during stages of
nesting in the olive ridley sea turtle. Letters above means indicate significant differences as
explained in Figure 12.1. (From Figler, R. A., Mackenzie, D. S., Owens, D. W., Licht, R, and
Amoss, M. S., Gen. and Comp. Endocrinol, 73, 223, 1989. With permission.)
predict nest number quite precisely. Rostal et al.^^ were even able to use testosterone
levels to accurately predict which clutch and the total number of clutches laid by a
given nesting female.
Caret fa caretta
FIGURE 12.8 Circulating levels of several hormones during nesting stages in the logger-
head sea turtle. Nesting stages are E, emerging; BP, body pit construction; N, nest chamber
construction; EO, early oviposition; LO, late oviposition; C, covering body pit; R, returning
to the ocean. *Data taken from Reference 55. (From Guillette L. J., Jr., Bjorndal, K. A.,
Bolten, A., Gross, T., Palmer, B., Witherington, B., and Matter, J., Gen. Comp. Endocrinol,
82, 121, 1991. With permission.)
12.4.2 M ales
The male sea turtles are less well studied than females because they do not come
ashore. However, since their energy investment in gamete production is also less
complicated than females, we suspect that they have a simpler hormone-control
system.
12.4.2.1 Off Season and Annual Cycles
Wibbels et al.^^ were able to sample 80 adult male loggerheads at the Cape Canaveral
Ship Channel between January and August (Figure 12.3). This unique man-made
channel is distinctive in that large numbers of adult, reproductively active males

arrive there first in the early spring, followed in the next few months by an arrival
of adult females. Since mating is commonly seen by boaters in the area, we suspect
that the coastal region near the channel is a courtship and mating ground. In the
summer, many of the females that arrive have been shown to nest in the important
beaches to the north and south of the channel, including the Archie Carr National
Wildlife Refuge. Among the males sampled, a subset of 11 of the 80 had testosterone
levels an order of magnitude lower than the others, who were clearly showing a
dramatic annual cycle. Since this channel appears to draw in both migratory repro
ductive adults"^^ and other age classes of loggerheads at most times of the year, one
interpretation of the data is that approximately 14% (11 of 80) of the adult males
of this population were not reproductively active during this sampling period.^^ This
result contrasts with another study of loggerheads at Heron Island in Australia, where
every male caught on the feeding ground appeared to be cycling and maintaining
elevated testosterone levels appropriate for the season."^^ We do not know if all of
these males actually did migrate away from Heron Island. One additional interpre
tation of the Cape Canaveral data is that the 14% of the males which were low in
testosterone had been caught repeatedly in fishing trawls along the intensly fished
coast and were suffering stress-induced reproductive failure. Reproductively active
adult loggerheads are regularly caught along this coastline.^^
Wibbels et al."^^ described the spermatogenic cycle of sea turtles (Table 12.2) as
prenuptial, with sperm formation and highest testosterone levels in the winter and
early spring and dropping testosterone levels as the mating season progresses. In
captivity, male green turtles showed a similar annual testosterone cycle.^ The pre
nuptial spermatogenic cycle may be characteristic of all sea turtles, since the same
pattern was reported by Engstrom^^ in green turtles. In addition, our limited sampling
of olive ridleys in Mexico^^ and Costa Rica supports this same pattern. In their green
turtle studies, Licht et al.^^ also noted that there was no daily cycle in male test
osterone, which is important to know, since single samples taken at various times
in the day might be suspect as being nonrepresentative, if indeed there were daily
cycles of the hormones. Interestingly, a few of these captive green males also showed
very low testosterone for the whole year, indicating that they too (like many females)
were taking the year off in terms of reproduction.
12.4.2.2 Migration and Courtship
From the work at Heron Island"^^ and our experiences at Cape Canaveral as discussed
above , t he impression we have is that loggerhead males begin their reproductive
migrations before the females. Supporting this idea are the testosterone data for both
of these populations which show a seasonal peak for males that is earlier than in
the females. We also noticed that some males left the Heron Island foraging area
well before the females departed on their reproductive migrations, although some
males did not seem to leave at all. Several years ago we administered testosterone
to immature greens (30 to 50 kg), which resulted in courtship and mating attempts
in animals which had not yet even initiated spermiation.'’ At this time, however, the
TABLE 12.2
Spermatogenic Stages Observed in Loggerhead Sea Turtles Captured on Heron
Island Reef
Month and number of turtles observed
stage Seminiferous tubules j F M
M A M J J A S O N D
1 Involuted with only spermatogonia; 5 5

spermatozoa may be present in
lumen
2 Primary spermatocytes present; 1 3
spermatogonia becoming abundant
3 Secondary spermatocytes and early 2
spermatids abundant
4 Transforming spermatids and some 6
spermatozoa
5 Spermatids and spermatozoa 4 4 1 2
abundant
6 Maximal spermiogenesis 7 4 3
7 Spermatozoa abundant; spermatids 1
and spermatocytes reduced
8 Few spermatozoa; spermatids and 3
spermatocytes may be absent;
spermatozoa may be abundant in
lumen
Adapted from Wibbels, T., Owens, D. W., Limpus, C., Reed, R, and Amoss, M., G e n . C o m p . E n d o c r in o l,
79, 154, 1990. With permission.
data on linking testosterone to migration behavior are primarily correlations, since

direct experimentation has not been attempted in sea turtles.
12.5 P IT U IT A R Y H O R M O N E S T U D IE S
As mentioned above, the pituitary hormones of the green sea turtle have been under
study for nearly 25 years. This aspect of sea turtle endocrinology has been of greatest
interest to the comparative/evolutionary zoologist. For this reason we will only
highlight some of the recent advances in this area.
The structure and function of the pituitary protein hormones have been of special
interest to the comparative endocrine community due to the sea turtle’s (any turtles)
clear link to the stem reptiles. Although perhaps an overly simplistic comparison,
they are seen as living relatives of the oldest amniotic vertebrates. Among turtles in
general, sea turtles happened to be chosen because large numbers of their large
pituitaries were available for study. Licht and Papkoff^ ^ reviewed their physiology
and structure studies of the three distinct glycoprotein hormones FSH, LH, and TSH,
showing LH to be the smallest, followed by TSH then FSH.^^ This is of interest
because, for a long time, we thought that FSH and LH might be a single molecule
with a somehow more “primitive” function. In testing different hormone preparations
in different vertebrate species, they concluded that “The relative importance of FSH
and LH may vary among turtles, and the sea turtle may be more like that of mammals
and birds’’.^^ Thus sea turtles may not even conform to a standard turtle model.
Sea turtle prolactin (PRL) and growth hormone (GH) have also been of particular
interest to the endocrinologist. As far as we can tell they were the first large proteins
ever completely sequenced from a r e p t i l e . P R L is slightly larger at 198 amino
acid residues,^^ while GH weighs in at 190.^^ The predicted somatotropic role for
GH has been strongly supported, along with a potential role in final gamete matu
ration.^ Evolutionarily speaking, the structure of the sea turtle GH molecule is
reassuring since it maps out closer to the birds than anything else (Figure 12.9).^^
PRL has proven particularly enigmatic in turtles. On one hand we know the
detailed structure of the molecule, but on the other hand we do not know what it
does in turtles.^ This is particularly surprising since PRL is well known to have roles
in migration, osmoregulation, and reproduction in vertebrates such as birds.
ó
FIGURE 12.9 Molecular phylogenetic tree of GHs on the basis of sequence comparisons.
The “Unit Evolutionary Period” figures for the tetrapod and teleost trunks are estimated to
be 9.6 and 6.3. The numbers of observed amino acid changes per 100 links are shown on the
tree. (From Yasuda, A., Yamaguchi, K., Papkoff, H., Yokoo, Y, and Kawauchi, H.,Gen. Comp.
EndocrinoL, 73, 242, 1989. With permission.)
12.6 M ETA B O LISM A N D STRESS
12.6.1 T hyroid M etabolism

In sea turtles both thyroxin T4 and triiodothyronine T3 are measurable by radioim
munoassay, with T4 circulating at much higher levels than does T3 .^'^ Interestingly,
Moon^"^ detected T3 only in the captive animals. He also noted that captive sea turtles
typically have higher levels of T4 (10 ng/ml in greens) than wild turtles (2 to 3 ng/ml
in both greens and loggerheads) and hypothesized that this may be due to the captive
conditions of constant temperature and high protein diet. We have noted over the
years that many captive sea turtles are overfed and obese.
In captive greens at the Cayman turtle farm, Licht et al.^^ showed no variation
in adult green turtle levels of T4 over three annual cycles. At a Galveston, TX
aquarium, however. Moon et al.^"^ found a significant annual cycle for female Kemp’s
ridleys, with a peak in the winter and spring correlating with follicular development.
Rostal^ later duplicated this result with female Kemp’s ridleys at the Cayman farm.
In the only attempt at this sort of annual thyroid study in wild turtles, Wibbels et
al.^^ thought he also detected an annual cycle in females at Heron Island. Adult
males and immatures in all situations except the one described below (hibernation?)
showed no annual cycle for T4 .
Winter dormancy (hibernation?) has been described in populations of black
turtles (C. agassizi) in the Gulf of California^^ and loggerheads (Caretta carettd) in
Cape Canaveral Channel, FL.^^ Since the thyroid axis is often involved in hibernation
and metabolic adaptations in other animals, we hypothesized an involvement of T4
in these sea turtles. Haynes and MacKenzie^^ found a temperature dependency of
thyroid hormone binding in plasma using both in vitro and in vivo techniques in
greens, loggerheads, and Kemp’s ridleys. Similarly, when Moon^^ subjected captive
immature Kemp’s ridleys and greens to gradual cooling in order to simulate hiber
nation conditions, the turtles showed greatly reduced circulating T3 and T4 levels
compared to controls. At the same time, the turtles being cooled showed reduced
feeding and activity levels and appeared to be exhibiting the dormancy conditions
as described for wild turtles. The thyroid axis clearly changed under these conditions;
however, no direct cause/effect relationship with hibernation was verified by these
protocols.
1 2 .6 . 2 C orticosterone, S tress, an d A ctivity
In 20-d-old hatchling loggerheads, Moiris^^ first showed that the hormone corticos
terone could be significantly elevated when the animals were under stress. She
exposed the turtles to 15 min of repeated dropping onto a sand substrate from a
height of only 3 cm. Hatchlings which were three d old showed a similar, but
nonsignificant response to the same stress, possibly because their background levels
of corticosterone were naturally three times higher than the older hatchlings. In a
study of captive juvenile loggerheads, Schwantes^^ found a distinct diurnal pattern
of corticosterone secretion, with highest levels occurring in the early morning. This
peak probably correlates with the highest activity period of the day for the species.
Juvenile loggerheads stressed by neF^ or trawl capture^^’^^ also had significantly
elevated corticosterone. Studying adult olive ridleys, Schwantes^^ found that mating
males which had been hand captured in the ocean and males awaiting slaughter in
a severely overcrowded holding tank showed significantly higher corticosterone than
females under the same conditions. It was suspected that this sex difference was
related to the reproductive condition of these turtles, with females somehow less
responsive to stress.Valverde et have recently initiated an extensive series of

experiments to study stress in reproductively active sea turtles. Preliminary results
reported in three abstracts^^'^^ suggest that nesting female olive ridleys do show a
modest stress response (elevated corticosterone) when nesting, but that a significant
number of individuals do not show added stress indications (further elevated corti
costerone) when flipped on their backs after nesting. In contrast, males captured
while mating always showed a significant increase in corticosterone when placed
on their carapaces.However, post-nesting females stressed by restraint in a shaded
corral on the beach did not show the expected peak in progesterone which would
have indicated ovulation of the next clutch.^^ Unrestrained post-nesters captured in
the water the day after nesting did show the peak in progesterone and had ovulated,
based on ultrasound of the oviducts. Therefore, the ridleys which were held on the
beach demonstrated a stress response which impacted their reproductive patterns.
In surprising contrast to these experiments, Licht et al.^^ flipped nesting greens at
Aves Island and found that they showed the predicted progesterone surge anyway,
on the following morning, despite being upside down! Since we have never seen
the progesterone surge without ovulation, we think these greens may have ovulated
despite the stress. Post-nesting loggerheads which were hauled upside down by
wheel barrow for several hundred meters and held in a crowded shallow tank for 3
d also ovulated (verified by laparoscopy) and showed the progesterone peak."^^
Further evidence of this ovulatory sensitivity in olive ridleys is found in an experi
ment conducted in Oaxaca, Mexico where only one of 12 turtles ovulated in small
holding tanks compared to 50% of a control sample taken at sea.^^ These results
suggest species-specific differences among the sea turtles, with the olive ridleys
potentially having an ovulation reflex which is more sensitive to inhibition by stress.
12.7 C O N C L U S IO N A N D SU M M A RY
Sea turtles have proven very useful in hormone studies for two important reasons.
First, they are relatively large animals who can therefore give blood easily and
repeatedly without suffering physiologically.^^ At Heron Island, for example, we
hand captured (turtle rodeo) and bled (15 ml) some females up to nine times between
July and December. Recaptures of these individuals indicated that all of these females
made normal migrations, had typically high fertility in their clutches, and produced
normal numbers of clutches that season. A second reason for their value in endo
crinology has been discussed above in that their commercial use provided pituitaries
for biochemical purification of protein hormones.
In addition to being useful models of adaptation, a solid reason for choosing
sea turtles for these types of studies is that they deserve and need the attention. Since
they fall into the category which one of my colleagues calls the ‘‘Charismatic mega
vertebrates”, they are inherently interesting. The good news is that we can take
advantage of this popularity at the same time as we look for practical ways to solve
conservation problems which man has created.
Unfortunately, because of the amazing diversity of solutions to problems that
animals have arrived at through time, it is definitely not easy to speculate on how
Hormones ¡n the Life History of Sea Turtles 335
or with what particular molecule a given species may have taken advantage of (in
an evolutionary sense) to solve adaptation problems. Sea turtles, for example, can
be said to be secondarily marine, in that their suspected ancestors were terrestrial
and aquatic. As a consequence, they appear to have secondarily arrived at or evolved
solutions to problems encountered by such behaviors as living in salt water, initiating
long migrations, open-sea navigation, using prenuptial gametogenesis, and so on.
As a consequence of the above behaviors, many of these adaptations in sea turtles
may be unique to their marine lineage. This possibility further argues for continued
basic research on the physiology and behavior of these unique and often endangered
species.
12.7.1 S ummary of P ractical O bservations
1. Yearling and older immature male sea turtles of all species tested maintain
higher levels of circulating testosterone than females. This difference is
useful in detennining sex ratios for these species since secondary sexual
characteristics do not appear for several years and since sea turtles have
TSD.35
2. At hatching, the chorioallantoic/amniotic fluid contains steroids from the
embryo. The ratio of estrogen to testosterone is greater in the fluids from
hatching females than from males. These data have been used in a non-
invasive technique to generate a sex ratio from freshly hatched
clutches.^^’^^
3. Where direct observation is limited, estimating the number of clutches
produced by a female in a season has been difficult. The level of test
osterone predicts clutch number in some species; thus it may prove useful
in quantifying clutch numbers.
4. In satellite tracking studies, testosterone levels in the female and the male
have proven useful in predicting migratory, nesting, and mating behavior.
5. Corticosterone correlates to level of stress at several stages in the lives of
sea turtles. Using corticosterone in wild, injured, or captive sea turtles
may eventually prove useful in evaluating survival prospects and general
health.^^’^^^
6 . A sudden peak in progesterone predicts the time of ovulation in sea
turtles.^’^
7. AVT is important in oviducal contraction at the time of n e s tin g .^ In
clinical situations it is often possible to force oviposition by injections of
AVT or oxytocin.
8 . By using a composite of circulating estrogen, testosterone, and progest
erone, along with capture data, it is possible to determine the reproductive

condition of an individual female.^’
9. To avoid the possibility of stress or temperature-altered levels of hor
mones, blood samples should be taken from turtles as soon as they are
captured in the wild.^'^
12.7.2 S ummary of Interesting O bservations a n d Q uestions
1. Birds appear to be a very useful model for marine turtle adaptation. Levels
of corticosterone in hatchlings show peaks and dips at times appropriate
for imprinting style learning to occur.
2 . Steroids play a pivotal role in sex determination in embryos. It appears,
however, that an extrinsic factor may be the initial switch to one sex or
the o t h e r . M u c h more work needs to be done in this area.
3. Environmental temperature plays a role in modifying the circulating levels
of steroids in immature sea turtles. For example, testosterone levels are
significantly higher in warmer water.^^
4. Environmental temperature does not correlate with circulating levels of
steroids in adults. On the contrary and hypothetically, photoperiod may
provide the primary seasonal cues for reproduction in these prenuptial
species.
Capture stress can alter circulating testosterone levels which can in turn
be correlated with higher corticosterone levels, and, in the case of nesting
females, altered progesterone levels.
6 . Secondary sexual characteristics are induced by elevated levels of test
osterone.^ In the case of the softening of the plastron, this has not been
verified."^^
7. There are clearly two gonadotropins in sea turtles (FSH and LH). Circu
lating LH dynamics are consistent with a role in ovulation and progest
erone production. FSH also peaks at the time of ovulation, but has not
been well studied and no distinct role is evident.
8 . Sea turtles produce PRL in their pituitary. At this time we have no idea
what PRL does in sea turtles.^
Thyroid honnones (T3 , T4 ) remain fairly constant in immature sea turtles
under “normal” temperature fluctuations. In the lab, turtles exposed to
very cold conditions (down to IHC) demonstrated a significant drop in
T3 and T4 and appeared to approximate a state of hibernation.^’^^
10. The timing of migration correlates closely in both sexes with the peak
testosterone level during the reproductive season.
11. Even though testosterone is elevated during mating receptivity in the
female, it is still not clear if testosterone defines and controls the mating
receptivity in sea turtles.'^^
12. Estrogen circulates at unusually low levels in greens and at low levels in
all sea turtles. A full appreciation for the seasonal dynamics and gona
dotropin controls over estrogen is needed.
A CK N O W LED G M EN TS
A special thanks to my former and present students whose sweat and even blood
have gone into the gathering of these data. Dr. Janice Grumbles and Rhonda Patterson
also put a huge amount of work into many assays which are reviewed here. Drs.
Duncan MacKenzie and Roldan Valverde generously critiqued portions of the manu
script. The research from our lab has been supported by grants from the National
Science Foundation (BNS-8819940 and IBN-9124014) and the Texas A&M Uni
versity Sea Grant College Program, supported by the NOAA office of Sea Grant,
Department of Commerce, under grant number NA16RG0457-01.
REFERENCES
1. Licht, P., Farmer, S. W., and Papkoff, H., Further studies on the chemical nature of
reptilian gonadotropins: FSH and LH in the american alligator and green sea turtle,
B iol Reprod., 14, 222, 1976.
2. Licht, R, MacKenzie, D. S., Papkoff, H., and Farmer, S. W., Immunological studies
with the gonadotropins and their subunits from the green sea turtle, Cheionia my das,
Gen. Comp. Endocrinol., 33, 231, 1977.
3. Owens, D. W., A preliminary experiment on the reproductive endocrinology of the
green sea turtle (Cheionia mydas), Proc. World Maricult. Soc., 5, 215, 1974.
4. Ulrich, G. and Owens, D. W., Preliminary observations on the reproduction of Che-
lonia mydas under fami conditions, Proc. World Maricult. Soc., 5, 205, 1974.
5. Owens, D. W., The Endocrine Control of Reproduction and Growth in the Green Sea
Turtle Cheionia mydas, Ph.D. dissertation. University of Arizona, Tucson, 1976, 108
pp.
6 . Licht, P, Wood, J., Owens, D. W., and Wood, F., Serum gonadotropins and steroids
associated with breeding activities in the green sea turtle Cheionia mydas. I. Captive
animals, Gen. Comp. Endocrinol., 39, 274, 1979.
7. Rostal, D. C., Owens, D. W., Louis, E. E., and Amoss, M. S., The reproductive biology
of captive Kemp’s ridleys (Lepidochelys kempi): pilot studies, in Proc. 8th Annu.
Workshop on Sea Turtle Conservation and Biology, Schroeder, B. A., Compiler,
NOAA Tech. Memo. NMFS-SEFC-214, Miami, FL, 91, 1988.
8 . Rostal, D. C., The Reproductive Behavior and Physiology of the Kemp’s Ridley Sea
Turtle, Lepidochelys kempi (Garman, 1880), Ph.D. dissertation, Texas A&M Univer
sity, College Station, 1991, 154 pp.
9. Owens, D.W. and Morris, Y. A., The comparative endocrinology of sea turtles,C6>p^/<3 ,
3, 723, 1985.
10. Licht, P, Evolutionary divergence in the structure and function of pituitary gonadot
ropins of tetrapod vertebrates, Aa?2. Zool, 23, 673, 1983.
11. Licht, P. and Papkoff, H., Réévaluation of the relative activities of the pituitary
glycoprotein hormones (follicle stimulating hormone, luteinizing hormone, and thy
roid stimulating hormone) from the green sea turtle, Chelonia mydas, Gen. Comp.
Endocrinol., 58, 443, 1985.
12. Licht, P, Endocrine patterns in the reproductive cycle of Herpetologica, 38,
51, 1982.
13. Licht, P, Breitenbach, G., and Congdon, J., Seasonal cycles in testicular activity,
gonadotropin, and thyroxine in the painted turtle, Chrysemys pietà, under natural
conditions, Gen. Comp. Endocrinol, 59, 130, 1985.
14. Mendonca, M. and Licht, P, Seasonal cycles in gonadal activity and plasma gona
dotropin in the musk turtle, Sternotherus odoratus, Gen. Comp. Endocrinol, 62, 459,
1986.
15. Owens, D. W., The role of reproductive physiology in the conservation of sea turtles,
inBiology and Conservation o f Sea Turtles, Bjorndal, K., Ed., Smithsonian Institution
Press, Washington, D.C., 1996.
16. Rostal, D. C., Robeck, T., Owens, D. W., and Kraemer, D. C., Ultrasonic imaging of
ovaries and eggs in Kemp’s ridley sea turtles (Lepidochelys kempi), J. Zoo WildL
Med., 21, 27, 1990.
17. Plotkin, P. T., Byles, R., Rostal, D. C., and Owens, D. W., Independent versus socially
facilitated oceanic migrations of the olive ridley Lepidochelys olivácea. Mar. Biol.,
122, 137, 1995.
18. Merchant-Larios, H. and Villalpando, I., Effect of temperature on gonadal sex dif
ferentiation in the sea turtle Lepidochelys olivácea: an organ culture study, J. Exp.
ZooL, 254, 327, 1990.
19. Desvages, G., Girondot, M., and Pieau, C., Sensitive stages for the effects of tem
perature on gonadal activity in embryos of the marine turtle Dermochelys coriácea,
Gen. Comp. Endocrinol, 92, 54, 1993.
20. Crain, D.A., Gross, T. S., Bolton, A., Bjorndal, K., Carthy, R., and Guillette, L., Jr.,
Development of a non-invasive sexing technique for hatchling loggerhead turtles
(Caretta caretta), in Proc. 14th Annu. Symp. Sea Turtle Biology and Conservation,
Bjorndal, K. A., Bolton, A. B., Johnson, D. A., and Elizar, P. J., Compilers, NOAA
21. Gross, T., Crain, D., Bjorndal, K., Bolten, A., and Carthy, R., Identification of sex in
hatchling loggerhead turtles {Caretta caretta) by analysis of steroid concentrations
in chorioallantoic/amniotic fluid, Gen. Comp. Endocrinol, 99, 204, 1995.
22. Harry, J. L., Williams, K. L., and Brisco, D. A., Sex determination in loggerhead
turtles: differential expression of two hnRNP pxoXCim, Development, 109, 305, 1990.
23. Harry, J. L., Brisco, D. A., and Williams, K. L., Putting the heat on sex determination,
Genética, 87, 1, 1992.
24. Martin, J. T., Imprinting behavior: pituitary-adrenocortical modulation of the
approach response, Science, 200, 565, 1978.
25. Morris, Y. A., Steroid Dynamics in Immature Sea Turtles, M.S. thesis, Texas A&M
University, College Station, 1982, 92 pp.
26 Schwantes, N., Aspects of Corticosterone Levels in Two Species of Sea Turtles
{Caretta caretta) and {Lepidochelys olivácea), M.S. thesis, Texas A&M University,
College Station, 1986, 60.
27 Owens, D. W., Comuzzie, D. C., and Grassman, M. A., Chemoreception in the homing
and orientation behavior of amphibians and reptiles, with special reference to sea
turtles, in Chemical Signals in Vertebrates, Vol. IV, Ecology, Evolution and Compar
ative Biology, Duvall, D., Muller-Schwarze, D., and Silverstein, R. M., Eds., Plenum
Press, New York, 1986, 341.
versus social facilitation models for the green turtle migration. Science, 248, 724,
1990.
29, Carr, A., Rips, FADS, and little loggerheads, 36, 92, 1986.
30, Renaud, L. M. and Carpenter, J. A., Movements and submergence patterns of log
gerhead turtles {Caretta caretta) in the Gulf of Mexico determined through satellite
telemetry, Bull. Mar. Sei., 55, 1, 1994.
31 Limpus, C. J., A Study of the Loggerhead Sea Turtle, Caretta caretta in Eastern
Australia, Ph.D. dissertation. University of Queensland, Lucia, Australia, 1985, 484
pp.
32. Owens, D. W. and Gem, W. A.,The pineal gland and melatonin in sea turtles, in
Current Trends in Comparative Endocrinology, Lofts, B. and Holmes, W. N., Eds.,
Hong Kong University Press, Hong Kong, 1985, 645.
33. Schroeder, B. and Owens, D., Sex ratio of immature green turtles in an east central
Florida developmental habitat, in 1993 Proc. 13th Annu. Symp. Sea Turtle Biology
and Conservation, Schroeder, B. A and Witherington, B. E., Compilers, NOAA Tech.
34. Coyne, M. and Landry, A., unpublished data, 1995.
35. Wibbels, T., Owens, D. W., Morris, Y. A., and Amoss, M., Sexing techniques and sex
ratios for immature loggerhead sea turtles captured along the Atlantic coast of the
U.S, in Ecology o f East Florida Sea Turtles, Witzel, W., Ed., U.S. Department of
Commerce, NOAA Tech. Rep., NMFS-53, Miami, FL, 1987, 65.
36. Wibbels, T., Owens, D. W., Limpus, C. J., Martin, R. E., and Amoss, M. S., A sea
turtle sex ratio. Am. ZooL, 27, 21A, 1987.
37. Wibbels, T., Martin, R. E., Owens, D. W., and Amoss, M. S., Jr., Female-biased sex
ratio of immature sea turtles inhabiting the Atlantic coastal waters of Florida, Can.
J. ZooL, 69, 2973, 1991.
38. Wibbels, T., Owens, D. W., Limpus, C, J., and Amoss, M. S., Field testing of a sexing
technique for immature sea turtles, in Proc. 2nd Western Atlantic Turtle Symposium,
NOAA Tech. Memo. NMFS-SEFC-226, Panama City, FL, 1989, 349.
39. Bolten, A. B., Bjomdal, K. A., Grumbles, J., and Owens, D. W., Sex ratio and sex-
specific growth rates in immature green sea turtles, Chelonia mydas, in the southern
Bahamas, Copeia, (4), 1098, 1992.
40. Wibbels, T., Balazs, G., Owens, D. W., and Amoss, M. S., Sex ratio of immature
green turtles inhabiting the Hawaiian archipelago,/. HerpetoL, 27, 327, 1993.
41. Diez, E. C. and van Dam, R., Foraging ecology and population dynamics of the
hawksbill {Eretmochelys imhricata) at Mona Island, Puerto Rico; summary report
for 1992-1994, Mona Island Hawksbill Research, Santa Práxedes 1632, Urb. Sagrado
Corazón, San Juan, Puerto Rico.
42. Wibbels, T., Owens, D. W., and Rostal, D. R., Soft piastra of adult male sea turtles;
an apparent secondary sexual characteristic,//crpcio/. Rev., 22, 47, 1991.
43. Wibbels, T., Owens, D. W., Limpus, C., Reed, P, and Amoss, M., Seasonal changes
in gonadal steroid concentrations associated with migration, mating, and nesting in
loggerhead sea turtles, Gen. Comp. Endocrinol., 79, 154, 1990.
44. Owens, D. W., Wibbels, T, Comuzzie, D., Rostal, D., and Limpus C. J., Sea turtle
reproductive chronology; the model and the questions, in Proc. 9th Annu. Workshop
Sea Turtle Conservation and Biology, Eckert, S., Eckert, K., and Richardson, T,
Compilers, NOAA Tech. Memo. NMFS-SEFC-232, Athens, GA, 135, 1989.
45. Wood, J. R. and Wood, F. E., Reproductive biology of captive green sea turtles
Chelonia mydas. Am. ZooL, 20, 499, 1980.
46. Wibbels, T, Owens, D., Licht, R, Limpus, C., Reed, R, and Amoss, M., Jr., Serum
gonadotropins and gonadal steroids associated with ovulation and egg production in
sea turtles, Gen. Comp. Endocrinol., 87, 71, 1992.
47 . Alvarado, J. and Figueroa, A., Breeding dynamics in the black turtle ((Chelonia
agassizi) in Michoacan, Mexico, in Proc. 9th Annu. Workshop Sea Turtle Conserva
tion and Biology, Eckert, S., Eckert, K., and Richardson, T, Compilers, NOAA Tech.
Memo. NMFS-SEFC-232. Athens. GA, 5, 1989.
48. Rostal, D., Alvarado, J., Grumbles, J., and Owens, D., Observations on the reproduc
tive biology of the black turtle, Chelonia agassizi, at Playa Colola, Michoacan,
Mexico, in Proc. 10th A m u. Workshop Sea Turtle Conservation and Biology, Rich
ardson, T., Richardson, J., and Donnelly, M., Compilers, NOAA Tech. Memo. NMFS-
SEFC-278, Miami, FL, 255, 1991.
49. Licht, P. and Papkoff, H., Réévaluation of the relative activities of the pituitary
glycoprotein hormones (follicle-stimulating hormone, luteinizing hormone, and thy
rotropin) from the green sea turtle, Chelonia mydas, Gen. Comp. Endocrinol., 58,
443, 1985.
50. Licht, R, Pickering, B., Papkoff, H., Pearson, A., and Bonna-Gallo, A., Presence of
a neurophysin-like precursor in the green turtle {Chelonia mydas), J. Endocrinol,
103, 97, 1984.
51. Figler, R., Changes in Plasma Concentration of Arginine Vasotocin During Oviposi-
tion in Sea Turtles, M.S. thesis, Texas A&M University, College Station, 1987, 106 pp.
52. Owens, D. W. and Ralph, C. L., The pineal-paraphyseal complex in sea turtles, I.
Light microscopic description,/. M orphol, 158, 169, 1978.
53. Figler, R. A., MacKenzie, D. S., Owens, D. W., Licht, R, and Amoss, M. S., Increased
levels of arginine vasotocin and neurophysin during nesting in sea turtles, Gc/î . Comp.
Endocrinol, 73, 223, 1989.
54. Figler, R. A., Unpublished data, 1984.
55. Guillette, L. J., Jr., Bjomdal, K. A., Bolten, A., Gross, T, Palmer, B., Witherington,
B., and Matter, J., Plasma estradiol-17B, progesterone, prostaglandin F, and prostag
landin E2 concentrations during natural oviposition in the loggerhead turtle (Caretta
caretta), Gen. Comp. Endocrinol, 82, 121, 1991.
56. Rostal, D., Grumbles, J., and Owens, D., Physiological evidence of higher fecundity
in wild Kemp’s ridleys: implications to population estimates, in 1991 Proc. 11th Annu.
Workshop Sea Turtle Biology and Conservation, Salmon, M. and Wyneken, J., Com
pilers, NOAA Tech. Memo. NMFS-SEFC 302, Miami, FL, 180, 1992.
57. Wibbels, T , Owens, D. W., and Amoss, M., Seasonal changes in serum testosterone
titers of loggerhead sea turtles captured along the Atlantic Coast of the U.S, mEcology
of East Elorida Sea Turtles, Witzel, W., Ed., U.S. Department of Commerce, NOAA
Tech. Rep. NMFS-53, Miami, FL, 1987, 59 pp.
58. Engstrom, T, Observations on the testicular cycle of the green iuvÛQ,Chelonia mydas,
in the Caribbean, in Proc. 14th Annu. Symp. Sea Turtle Biology and Conservation,
Bjomdal, K. A., Bolton, A., Johnson, D., and Elizar, P. J., Compilers, NOAA Tech.
59. Owens, D. W., The comparative reproductive physiology of sea turtles. Am. Zool,
20, 546, 1980.
60. Licht, R, Wood, J., and Wood, F., Annual and diurnal cycles in plasma testosterone
and thyroxin in the male green sea turtle Chelonia mydas, Gen. Comp. Endocrinol,
57, 335, 1985.
61. Yasuda, A., Kawauchi, H., and Papkoff, H., The complete amino acid sequence of
prolactin from the sea turtle {Chelonia mydas), Gen. Comp. Endocrinol, 80, 363,
1990.
62. Yasuda, A., Yamaguchi, K., Papkoff, H., Yokoo, Y, and Kawauchi, H., The complete
amino acid sequence of growth hormone from the sea turtle {Chelonia mydas), Gen.
Comp. Endocrinol, 73, 242, 1989.
63. Moon, D.-Y, The Responses of Sea Turtles to Temperature Changes; Behavior,
Metabolism, and Thyroid Hormones, Ph.D. dissertation, Texas A&M University,
College Station, 1992, 195 pp.
64. Moon, D.-Y., Rostal, D., MacKenzie, D., and Owens, D., Serum thyroid hormones
in sea turtles, in Proc. 10th Annu. Workshop Sea Turtle Conseiâtion and Biology,
Richardson, T., Richardson, J., and Donnelly, M., Compilers, NOAA Tech. Memo.
NMFS-SEFC-278, Athens, GA, 253, 1991.
65. Wibbels, T., MacKenzie, D. S., Owens, D. W., Amoss, M. S., and Limpus, C. J.,
Aspects of thyroid physiology in loggerhead sea turtles, Caretta caretta, Am. ZooL,
26(4), 564, 1986.
6 6 . Felger, R. S., Cliffton, K., and Regal, R J., Winter dormancy in sea turtles: independent
discovery and exploitation in the Gulf of California by two local cultures. Science,
191, 283, 1976.
67. Carr, A., Ogren, L., and Me Vea, C., Apparent hibernation by the Atlantic loggerhead
turtle Caretta caretta off Cape Canaveral, Florida, Biol. Conservation, 19, 7, 1981.
6 8 . Haynes, S. and MacKenzie, D., Thyroid hormone binding to serum proteins in sea
turtles, Am. ZooL, 30, 124A, 1990.
69. Valverde, R. A., Owens, D. W., MacKenzie, D. S., and Amoss, M. S., Corticosteroids
during nesting in the olive ridley {Lepidochelys olivácea) sea turtle at Playa Nancite,
Costa Rica, in 1993 Proc. 13th Annu. Symp. Sea Turtle Biology and Conservation,
Schroeder, B. A. and Witherington, B. E., Compilers, NOAA Tech. Memo. NMFS-
SEFSC-341, Miami, FL, 183, 1994.
70. Valverde, R. A., MacKenzie, D. S., Owens, D. W., and Amoss, M. S., Adrenal
responsiveness during nesting of the olive ridley sea turtle (Lepidochelys olivácea).
Am. ZooL, 32, 21A, 1992
71. Valverde, R. A., Owens, D. W., MacKenzie, D. S., and Patterson, R., Hormone levels
and ovulation correlates in the olive ridley sea turtle, in Proc. 14th Annu. Symp. Sea
Turtle Biology and Conservation, Bjomdal, K. A., Bolton, A., Johnson, D., and Elizar,
P. J., Compilers, NOAA Tech. Memo. NMFS-SEFSC-351, Miami, FL, 156, 1994.
72. Licht, P, Rainey, W., and Cliffton, K., Serum gonadotropins and steroids associated
with breeding activities in the green sea turtle, Chelonia my das. II. Mating and nesting
in natural populations, Gen. Comp. Endocrinol., 40, 116, 1980.
73. Licht, R, Owens, D. W., Clifton, K., and Penaflores, C., Changes in LH and proges
terone associated with the nesting cycle and ovulation in the olive ridley sea turtle
Lepidochelys olivácea, Gen. Comp. Endocrinol., 48, 247, 1982.
'j Q Salt, Water, and pH
Balance in the Sea Turtle
Peter L. Lutz
CO N TEN TS
13.1 Introduction................................................................................................. 343

13.2 Salt and Water Exchange............................................................................ 344
13.2.1 Water Loss.................................................................................... 344
13.2.2 Salt Loss........................................................................................ 345
13.2.3 Salt G ain.............. 345
13.3 Osmotic and IonicRegulation..................................................................... 347
13.3.1 Salt Gland..................................................................................... 347
13.3.1.1 Size............................................................................... 347
13.3.1.2 Structure....................................................................... 347
13.3.1.3 Concentration and Content..........................................348
13.3.1.4 Llow Rate and Salt Efflux...........................................350
13.3.1.5 Mechanismsfor Tear Eormation.................................. 351
13.3.1.6 Mechanismsfor Salt Gland Activation........................ 352
13.3.2 Kidney........................................................................................... 352
13.4 Acid-Base Balance...................................................................................... 353
13.4.1 Diving and pH............................................................................... 354
13.4.2 Temperature and p H ..................................................................... 355
13.4.3 Regulation of p H .......................................................................... 356
13.5 Concluding Remarks................................................................................... 357
References..............................................................................................................357
13.1 INTRODUCTION
Sea turtles live in a medium that is almost three times more concentrated than their
body fluids, some ions having even greater plasma/sea water gradients (i.e., 23-fold
for Mg^+, Table 13.1). Like most marine vertebrates, therefore, sea turtles must
continuously combat the problems of water loss and salt gain. How they maintain
internal osmotic and ionic homeostasis and how they regulate the related acid-base
balance is the subject of this chapter.
0-8493-8422-2/97/$0.00+$ 50
TABLE 13.1
Salt Concentrations (mm H ) of Sea Turtle Plasma and Seawater, and the
Ratios of Average Plasma Concentrations to Those of Seawater
Osmotic
Na K Ca Mg Br Cl pressure Ref.
Seawater 470 10 10.2 54 0.8 548 949 79

Plasma
C h e lo n ia m y d a s 158 1.5 370 3
C a r e tta c a r e tta 145 3.8 1.5 2.1 107 321 1
140 3.2 1.3 1.2 0.3 110 320 35
Plasma/seawater ratios 3.2 3.5 13 32.7 2.7 5.1 2.8
For sea turtles the major potential routes for water loss are the respiratory tract,
the kidney, the salt gland, the gut (feces), and the integument. The major routes for
salt gain are the integument and gut (ingestion). In essence, these pathways are
combated through: (1) structural devices that act to mechanically reduce water and
salt flux and (2) physiological mechanisms that work to actively extrude gained ions,
obtain free water, and regulate the internal ionic millieu.
Although sea turtles appear to maintain tight control over the osmotic and ionic
composition of their plasma while they are at sea,^’^ it should be pointed out that
under special circumstances they can withstand substantial internal changes. Holmes
and McBean^ found that juvenile green sea turtles {Chelonia mydas) tolerated a fall
in plasma sodium from 157.8 mmol to 130.2 mmol 1~^ during 2 months in fresh
water, and Staub"^ reported a fall in plasma osmolality from 270 to 207 mOsm in
green sea turtles after about 14 weeks in fresh/brackish water (l.l ppt). A similar
robustness with regards to changes in internal salt/water balance is also suggested
from the old observations that captured green sea turtles can survive up to two
months held on ship deck, provided they were kept in the shade and occasionally
wetted with seawater:^ an ability that made them an invaluable source of freshmeat
for the early European mariners in the Caribbean.
13.2 SALT AND WATER EXCHANGE

13.2.1 W ater Lo ss
While respiration can be a major source of water loss for terrestrial air-breathing
vertebrates,^ this route is probably minimal for sea turtles since the relative humidity
of the ocean surface air that they inspire will be about 100% and the expiratory gas
will have a similar water content.^ On those occasions when the turtle has an elevated
body temperature (see Chapter 11) some respiratory water loss will occur, but this
will be mitigated by the high efficiency of the sea turtle lung in extracting oxygen,^
since the greater the amount of oxygen uptake per breath, the less the ratio of
respiratory water loss/oxygen consumption.
Salt, Water, and pH Balance in the Sea Turtle 345
For an aquatic animal the rate of water loss across the integument will be a
function of its permeabilty and the transcutaneous osmotic gradient. The reptilian
epidermis differs from that of the mammal in having two types of keratinous protein,
a-keratin (hair-like) and (i-keratin (feather-like). In sea turtles the neck and leg have
a-type keratin while the shell scutes have p-type.^ Although the thickness of keratin
will be a factor in impeding cutaneous water loss by increasing the diffusion path
length, the primary barrier to water diffusion is provided by the lipid embedded in
the keratinous framework.^ In these respects the structure of the marine turtle skin
and shell indicates a high resistance to water diffusion.^’^^
Water loss from the passage of feces or urine has not been quantified, but the
estimates of very low rates for cloacal salt loss (see below) indicate that these routes
are likewise minor. By contrast, the salt gland provides a net gain of water (see
below).
13.2.2 Salt L oss

Na+ efflux estimates for sea turtles are in the range of 130 to 300 pmol 100 g~^
}^-i 3,10,11 |3 elow the values for marine teleosts (about 1000 pmol 100 g"^
h-i) 12 5 m jg same order of magnitude as marine elasmobranchs (50 to 100
jiimol 100 g“^ h“0,^^ indicating that like marine elasmobranchs, sea turtles have a
low epithelial permeability for Na+.^^ The bulk of this salt loss is from the head
region, presumably via the salt glands (60% in the loggerhead,90% in the green,
with cloacal loss accounting for less than 5% of the total).^’^^^’^^ From cloacal salt
loss values, Kooistra and Evanscalculate a cloacal flow rate of 66 \x\ 100 g"^ h~^
in hatchling green sea turtles. The carapacal scutes of C.mydas have permeabilities
for Na, K, and Br that are two to four orders of magnitude less than that for water.^
13.2.3 Salt G ain

Ingestion is probably the major source for salt loading. The diet of marine turtles
consists mainly of marine invertebrates or, in the case of C. mydas, various marine
grasses and algae. Since marine invertebrates and plants have as much as three times
the salt concentration as sea turtle body fluids,^ the salt burden from feeding could
be considerable. An additional, and perhaps even more important, intake will come
from seawater swallowed along with the food. Mechanisms to reduce the salt load
from this “incidental drinking” could be of major significance.^^
The esophagus provides the primary defense against “incidental drinking”. One
of the most peculiar features of sea turtles is the presence of long, closely packed
conical papillae throughout the length of the esophagus (Figure 13.1). These con
stitute a unique feature of the alimentary system of sea turtles, setting them anatom
ically apart from “general reptiles”. The papillae are typically stiff projections with
pointed tips, oriented posteriorly towards the stomach. The sea turtle alimentary
tract also possesses a powerful sphincter which divides the stomach from the esoph
agus (green and loggerhead sea turtles,leatherback).^^
It is unlikely that the papillae function in nutrient or fluid transfer, as they are
covered with keratinized epithelia with an extensive layer of connective tissue
FIGURE 13.1 Esophageal papillae in leatherback sea turtle with entrapped jellyfish.
underneath and have a thick cornified layer on the s p i n e . B u t they probably play
a role in reducing seawater intake. X-ray studies on green sea turtles using barium
solutions indicate that ingested food and seawater are initially prevented from passing
from the esophagus into the stomach by the closing of the sphincter. The contraction
of the esophagus then expels seawater out of the mouth or nares while the papillae
act to trap the food.^^ After this action the sphincter relaxes, allowing the passage
of the food bolus to the stomach. Significantly, hawksbills, loggerheads, and green
sea turtles have all been observed to forcefully eject seawater out of their nostrils
while feeding.
Although minimizing the osmotic burden from “incidental drinking” may be
important for adult sea turtles, there is evidence that hatchlings require some degree
of seawater intake. Bennett et al.*^ found that hatchling loggerhead turtles (Caretta
caretta), who can lose up to 12% of their body weight during emergence from the
nest, drink about 4 ml seawater 100 g'^ day"^ when placed in seawater and regain
their initial weight within about 10 to 15 d even without feeding. However, if not
allowed to drink they rapidly lose weight through dehydration. Marshall and
Cooper^® also found that hatchling Chelonia mydas in seawater gain weight before
they start to feed and calculate a similar drinking rate (2.4 to 8.5 ml seawater 100
g-i day~0- Drinking has also been recorded in juvenile green sea turtles. Using phenol
red as a marker. Holmes and McBean^ found that unfed juvenile green sea turtles
ingested 1.33 ml seawater 100 day^k However, it has been argued that sea turtles
normally avoid drinking seawater.
13.3 OSMOTIC AND IONIC REGULATION

13.3.1 S alt G land
With the apparent exception of the marine mammals, the kidneys of marine verte
brates are insufficient to handle their salt influx and all have extrarenal mechanisms
for excreting salt. As a group, reptiles are unique in the variety of organs used as
salt glands which include modified salivary glands (sea snakes), lingual glands
(crocodiles), and nasal glands (lizards).In sea turtles, salt excretion is performed
by highly modified lachrymal glands, whose function was first discovered and
investigated by Schmidt-Nielsen and Fänge.^"^
13.3.1.1 Size
The sea turtle salt gland is one of the more conspicious organs in the skull (Figure
13.2) whose importance perhaps can be gauged from the fact that in the leatherback
the salt gland mass is almost twice that of the brain^^ (alternatively, of course, this
may indicate the relative unimportance of the brain for these animals). Interestingly,
hatchlings have proportionately much larger salt glands than adults — i.e., for
hatchling green sea turtles (53 g) the salt gland is 0.3% body weight and for 100-
to 200-g leatherbacks the salt gland is about 0.4% of body w eight,w hereas for
large Caretta carreta (130 kg), the salt gland is only 0.046% body w eig h t and for
40-kg green sea turtles the salt gland is 0.045% body weight.^ In the smaller ridley’s
turtle (Lepidochelys oliváceo) weighing 9.1 kg, the salt gland was 0.067% body
weight.^ This large difference in relative salt gland weights between hatchlings and
adults may be related to the necessity for hatchlings to drink seawater when they
first enter the sea (see above). By comparison the relative salt gland weight for a
wide range of sizes of sea bird species is about 0.08% body weight.^^
13.3.1.2 Structure
The gross, histological, and ultrastructural morphology of the sea turtle salt gland
has been well studied (e.g., Schmidt-Nielsen and F ä n g e ,Ellis and Abel,^^ Abel and
E llis,an d Marshall and Saddlier).^^
Briefly, sea turtle salt glands are modified lachrymal glands, each gland consist
ing of about one hundred lobules separated by blood vessels and connective tissue.^"^
The lobules contain many blind-ending secretory tubules which drain into a central
canal within each lobule, and the central canals combine to form secondary ducts
and a sac-like main duct which opens into the posterior canthus.^^’^^ The central
canal and secretory tubules have a rich blood supply^^ and there is extensive vascu
larization of the duct.^^ The central canals are surrounded by broad sheaths of
connective tissue with a rich network of cholinesterase staining nerve fibers around
each tubule; a nerve network reactive to monoamine oxidase is also found in the
perilobular connective tissu e.T h e tubule is composed of three types of cell: small
basophilic cells in the tubule terminal segments, principle secretory cells separated
FIGURE 13.2 MRI of head of leatherback sea turtle showing salt gland ducts converging
at the upper posterior region of the eye socket (acknowledgements W. Coles and A. Schulman).
by broad intercellular spaces, and small oval basal cells in the body of the tubule.
The principle secretory cells are characterized by having large quantities of mito
chondria and have convoluted lateral membranes that interlock with similar processes
in adjacent c e l l s . T h e surface area of the principle cells that is exposed to the
extracellular fluid is, in consequence, extensive.The intercellular channels of the
principle cells of sea turtles are filled with a mucopolysaccaride, a feature not seen
in bird salt glands,^”ând the main duct of the turtle salt gland is lined with mucocytes
which secrete a thick luminal layer of mucus lining the duct lumen.^^ Marshall and
Saddlier^^ suggest that the general organization of the sea turtle salt gland duct is
structurally more similar to that of the main duct of the elasmobranch rectal gland
than the ducts of the avian nasal gland, possibly indicating a functional similarity
between the elasmobranch rectal gland and the sea turtle salt gland.
13.3.1.3 Concentration and Content
When the salt gland is not active the loggerhead sea turtle produces a scanty secretion
that is isosmotic with plasma, i.e., 300 to 400 mosmol kg"^ (Figure 13.3). However,
when the gland is stimulated, tear concentrations rise rapidly to a plateau of about
1680 to 2000 mosmol kg“^ average 1900 mosmol kg~* (Figure 13.3), showing that
the sea turtle salt gland discharges a fluid with a concentration more than six times
that of the blood and twice that of seawater (Table 13.2). This all or nothing plateau
o
£
FIGURE 13.3 Changes in flow rate and concentration of salt gland fluid in the green sea
turtle following salt gland activation (time zero). (From Nicholson, S. E. and Lutz, P. L .,/.
Exp. Biol., 144, 171-184, 1989. With permission.)
is maintained for the duration of the response period, after which it rapidly declines.
The concentrating power of the sea turtle salt gland is greater than most marine
birds achieve^^ and twice that of the shark rectal gland.^^ In practical ternis the result
is that the osmotic load incurred by a sea turtle ingesting 1 1 of seawater would be
unburdened by the secretion of about 500 ml tears, leaving a net gain of 500 ml free
water.
Chloride and sodium are the predominant ionic components of sea turtle tears,
but, unlike the salt gland fluid of the bird, which is almost pure NaCl,^^ green turtle
tears have substantial amounts of magnesium (36 to 54 mmol 1~0» calcium (20 to
36 mmol LO» potassium (12.5 to 30 mmol LO? bicarbonate (5.5 mmol ), and even
bromine (2.2 to 6.24 mmol LO (Table 13.2).^^'^^ Avian salt gland fluid has, by
contrast, less than 1 mmol magnesium and calcium combined.Turtle salt gland
fluid also contains appreciable amounts of urea (8.9 mmol LO at concentrations
similar to those found in blood. But interestingly, the urea levels remain constant as
osmotic pressure increases, suggesting that this osmolyte is in passive equilibrium
across the salt gland duct and is not being actively added to or removed from the
salt gland fluid.^^ Glucose concentrations are much less than blood (0.5 mmol
salt gland fluid, plasma 8.0 mmol 10. indicating that glucose is either not entering
the salt gland fluid or is actively removed from it. Protein is virtually absent (less
than 0.01 mg mkO-^^
Comparing the ionic composition of sea turtle salt gland fluid with that of plasma,
the greatest gradient is shown by Mg^+ (about x 18), greatly exceeding the sixfold
difference in osmotic pressure. To some degree, this may reflect the ionic challenge
presented by seawater imbibation, e.g., Mg^+ has the highest seawater/plasma gra
dient of all of the common ions (x 24, Table 13.1). Interestingly, the relative
concentrations of the major ions, Naând CL, in sea turtle salt gland fluid are much
closer to those for seawater (NaVCL = 0.89, green sea turtle tears; Na+/CL = 0.86
TABLE 13.2
Salt Gland Fluid and Urine Composition (mmol H ) of Various Sea Turtles
Osmotic
pressure
Na K Ca Mg Br Cl NH 3 Urea (mosm b^) Ref.
Salt gland fluid

Chelonia 867 29.9 20.4* 35.8 2.18* 914 8.9 1900 33,*34
mydas 685 21 32
827 17 983 1854 29
Caretta 878 31 992 32
caretta
Urine
Chelonia 11.5 8.1 3
mydas 22 34 3.4 18.3 9 2 320 39
0.14 12.9 49
Lepidochelys 20.5 8.6 3
oliva
Dermochelys 76.4 360 16
coriacea
* From Hudson and Lutz.^'*
seawater) than plasma (NaVCh = 1.30, plasma green sea turtle).It would appear
that the ionic composition of the sea turtle salt gland fluid reflects more the ionic
burden from ingested seawater rather than the composition of the fluid from which
it is derived (ecf/plasma). This indicates that not only is the salt gland important in
water economy, but it also is concerned in internal ionic regulation.
Nicholson and Lutz^^ found that the composition of the salt gland fluid was
fairly constant throughout the secretory period, and stimulating the salt gland by
loading the turtles with hypertonic NaCl solution or hypertonic sucrose solution also
had little effect on the ionic composition of the tears.This suggests that once
activated, the sea turtle salt gland produces a fluid of fixed composition. Some lizards,
on the other hand, can vary the composition of their salt gland secretions; in
particular, considerable changes in the NaVK+ ratios can be produced by altering
the size and nature of the salt load.^^ By contrast. Holmes and McBean"* found that
the Na+/K^ ratio of the salt gland fluid for green sea turtles did not change over time.
This inflexibility in the composition of sea turtle salt gland fluid may reflect that the
marine life of sea turtles presents them with more uniform salt “insult” compared
to the terrestrial lizards.
13.3.1.4 Flow Rate and Salt Efflux

In the loggerhead, tear production typically involved a rapid increase in flow rate to
a peak, followed by a decline to a lower level, around which output fluctuated until
it gradually dropped off to zero (Figure 13.3). The initial increase in flow coincided
with the rise in tear concentration, and the decline in concentration at the end of the
secretory bout corresponded to the terminal fall in flow. A similar flow pattern has
also been reported in birds. For example, McFarland^^ observed that in salt-loaded
sea gulls the salt gland secretion rate rapidly peaked and then dropped to a lower
level where it leveled off and eventually ceased. On the other hand, leatherback sea
turtles (Dermochelys coriacea) appear to cry continuously,^^ and hatchling green
sea turtles maintain high secretion rates throughout the secretory period.^^ It is most
likely that flow patterns relate to the nature of the salt loads naturally encountered,
the hatchlings having particular problems in maintaining water balance (see above)
and the soft-skinned leatherbacks, which feed on jellyfish, having a higher salt influx
than other turtle species. In the loggerhead the duration of salt gland activity was
affected by the salt load.^^ With salt loads of 2.5 to 10 mmol ion kg body mass”^
the salt gland was active for 3 to 8 h; salt loads of 20 mmol ion kg body mass“^
produced responses lasting 15 to 18 h.
Several authors report similar cephalic/salt gland Na“^ efflux rates for the green
turtle (119 p.mol 100 g^^ h“^ Holmes and McBean;^ 118 pmol 100 g~^ h~^ Kooistra
and Evans;^ *137.8 jiimol 100 g"^ h \ Nicholson and Lutz).^^ These values are in the
same range as other reptiles with salt glands 1 to 255 pmol Na+ 100 g'^ h"k^'^
However, on a short-term basis, very much higher maximal rates (10,700 p.mol 100
g 1 h'l) were obtained by small (28-g) hatchling greens.^*^ This extraordinary high
secretion capacity in the hatchling may be related to its need to drink seawater when
it first enters the sea.*^
Interestingly, the left and right eyes of individual turtles often show marked
differences in tear flow and concentration. This has been reported for subadult green
turtles,hatchling green turtles,^^ and leatherback turtles.^^ Unilateral differences
in flow from the nasal salt glands have also been reported in the desert lizard
Uwmastyx acanthinurus,'^^ so the phenomenon is not limited to chelonians. No
comparisons of the salt gland secretions from the two eyes of any avian species have
been reported, presumably because avian nasal glands empty into the nostrils where
the tears mix across an orifice in the nasal septum.^^
13.3.1.5 Mechanisms for Tear Formation
Although the particular concentration mechanisms employed by chelonian salt

glands are unknown, the basic process of secretion of extrarenal tissues appears to
be remarkably similar throughout the vertebrates. These involve the establishment
of a transcellular membrane sodium gradient through the activity of basolateral Na-
K pump and the coupled cotransport of ions with sodium as it moves into the cell
down the Na+ gradient."^* Using X-ray microanalyses, MarshalP^ found that the
principle cells contained high intracellular Cl concentrations (81 mmol FO and that
during secretion the Na+ intracellular concentrations increased from 13 to 34 mmol
cell water. This situation is consistent with a secretion mechanism that involves
the coupled entry of chloride with sodium via a Na-2C1-K cotransporter in a manner
similar to that described for the rectal gland of the elasmobranchs.^^ However, other
factors may also be at work. For example, Ellis and AbeF^ suggest that the muco
polysaccharide filling the intercellular channels of principle cells, a feature peculiar
to sea turtles, may function as an ‘Ton exchange resin”.
Interestingly, the fluid in the lumina of the sea turtle secretory tubules is isosmotic
with b l o o d , a situation similar to that found in birds where the initial fluid at the
blind end of the salt gland tubules of birds is also isotonic with blood.'^^ Marshall
et al.^^ '^^have suggested that a standing osmotic gradient across the duct wall causes
the primary fluid to be concentrated by water absorption as it flows along the duct,
and that further ionic modifications in the fluid composition are brought about by
ion pumps. Osmotic extraction of water across a standing gradient would explain
the independence of concentration and flow rate seen in the sea turtle salt gland,
since once the luminal fluid came into equilibrium with the standing gradient no
further water would be taken up, and increasing the flow would merely displace the
position along the duct where equilibrium is achieved.^^
An interesting new hypothesis concerning the production of high osmotic con
centrations has recently been put forward by Schmidt-Nielsen.'^'^ The proposed mech
anism, which compasses the working of insect malphigian tubules to the mammalian
kidney, involves the production of hydrostatic pressures in the salt secreting tissues
through the action of peristaltic contractions in the muscle surrounding the organ.
The raised hydrostatic pressure acts to concentrate the luminal fluid by forcing water
across membrane pores that are selectively permeable to water. Further studies on
the highly effective sea turtle salt glands are clearly called for.
On the assumption that concentration of the expressed salt gland fluid is mainly
the result of an osmotic transfer of water across the duct wall, Nicholson and Lutz^^
calculate a maximum secretory rate 9.41 ml kg~* h~^ into the secretory tubules and
that 83% of the water from this initial isosmotic fluid is absorbed as the fluid passes
down the ducts. On a mass salt gland basis, this translates into a maximal initial
secretory rate of 0.32 ml g salt gland~^ mim^ comparable to the glomerular filtration
rate for humans (0.5 ml g kidney“^ min O-"^^
13.3.1.6 Mechanisms for Salt Gland Activation
While little is known of the mechanisms that control sea turtle salt gland activity,
there appear to be some similarities with the bird salt gland. In birds the primary
control for activation is neuronal acting via the cholinergic nerves, with hormonal
influence secondary.In sea turtles, cholinergic nerves also appear to be involved
in salt gland activation.^'^ The intracellular activation of salt gland secretion in
Malachlemys (the only turtle species to date investigated) is also similar to that of
birds in that it is mediated by changes in cyclic AMP and involves an increase in
intracellular calcium.'^^ Although there is no information on intracellular mechanisms
in sea turtle salt glands, it would be surprising if they were very different. The recent
finding that a melanocyte stimulating hormone in birds is targeted to directly activate
their salt glands'^^ is certainly worth following up in turtles.
13.3.2 K idney
The primary function of the kidney is to eliminate waste products and regulate the
electrolyte composition of the blood. However, in reptiles the cloaca, colon, and
urinary bladder are also important organs which modify the urine produced by the
kidney.
A detailed treatment of the morphology of the sea turtle kidney is given by
Solomon.'^^ essence, sea turtle kidneys (Chelonia mydas) are flattened lobed bodies
closely applied to the posterior wall of the pleuroperitoneal cavity. The nephron can
be divided into several distinct segments: a glomerulus, similar in structure to that
of the mammal, a proximal tubule, an intermediate segment, distal convoluted tubule,
and collecting tubules. This complex nephron would appear to be fully equipped to
perform the functions of filtration, absoiption, and secretion. Unlike snakes or
crocodiles, turtles have urinary bladders.^ We have no information on the operation
of the bladder in sea turtles, but in fresh water turtles the bladder is involved in the
active transport of both Na+ and CU, the secretion of H+, and the reabsorption of
HC0 3 ~.2^’48 jg probable that the sea turtle bladder possesses similar ion transport
mechanisms.
The composition of urine from several sea turtle species is shown in Table 13.2.
Na^ and CU are the major ionic constituents, but are present at much lower concen
trations than the blood (Table 13.1). Only urine K+ values are substantially higher
than the blood. The greatest variations are seen in the ammonia and urea measure
ments. BJorndal"^^ found little ammonia and a substantial amount of urea in the urine
of C. mydas. By contrast, Prange and Greenwald^^ found substantial ammonia, but
little urea, and KhaliP® only found urea in the urine of one out of four C. mydas
tested (8 . 8 mmol/1). It is probable that the urinary concentrations of these two
constituents vary widely according the nutritional and water balance status of indi
viduals. In the fresh water Trachemys scripta, for example, ammonia can account
for from 4 to 44% of total urinary nitrogen and urea from 45 to 95% of total urinary
nitrogen.^ ^
Under normal circumstances the osmotic concentration of urine is isosmotic to
that of the body fluids. For example, osmolal urine/plasma ratios of 0 .9 5 ^ 2 ^nd 1.0^^
have been reported for C. mydas and 0.99 for the leatherback.*^ However, the urine
may become more concentrated than plasma after prolonged dehydration (for C.
mydas 1 0 to 2 0 d out of water the average osmolal urine/plasma ratio = 1.15).29
Interestingly, the sum of the measured constituents is very much less than the osmotic
concentration (less than one third),^^ indicating that there are substantial amounts
of dissolved constituents yet to be discovered. The glomerular filtration rate (GRF)
of the C. mydas kidney is 14.3 ml kg' *h~*,^2 y^j-y much higher than the banded sea
snake (Laticauda colubrina, GFR = 0.5 ml kg *h *)^2 ^nd three times that for T.
scripta in fresh water (GFR = 4.73 ml kg * h~*).^2 Qu a daily basis the green sea
turtle kidney filters about 340 ml fluid per kilogram which, compared to 2 to 6 ml
kg *day *urine production (calculated from Prange and Greenwald),^^ indicates that
greater than 98% of the filtrate is reabsorbed prior to voiding.
13.4 A C ID -B A S E B A LA N C E
In biological solutions [HH and [HCO3“] can be considered as dependent variables
whose concentrations are determined by the independent variables: the difference
in charge between the sum of the strong ion electrolytes, the PCO2 , and the total
weak acid present.'’^ This means that the ionic composition and acid-base balance
of the blood are functionaly linked, and according to Stewart,changes in [H+] are
only brought about by alterations in one or more of these independent variables.
More importantly, solutions separated by biological membranes can interact in acid-
base terms only by processes that affect the values of the independent variables. The
pH of blood plasma is essentially determined by a C0 2 ~HC0 3 ~ buffer system that
can be described by the Henderson-Hasselbach equation:
pH = pKa + log([HC0 3 ]/aC 0 2 • PCO2
where aC 0 9 is the solubilty coefficient relating [CO2 ] and PCO2 , and pKa is the
apparent pK of this reaction.
Since changes in pH have such wide metabolic effects, disturbances in acid-
base balance are generally resisted.
13.4.1 D iving and pH

For diving sea turtles the problem of maintaining pH is particularly acute since they
have a greater commitment to breath-hold diving than any other reptile. Loggerhead
sea turtles, for example, routinely spending 97% of their time submerged.^^ During
a long apneic dive respiratory CO2 will build up, and when O2 stores become
insufficient to meet metabolic needs, anaerobic metabolism will result in the accu
mulation of lactic acid, and the increasing lactate would produce further molecular
CO2 through the acid titration of bicarbonate.
Unfortunately, there are few studies on free-diving turtles. In loggerheads, qui
escent voluntary dives of 40 min produced a fall in arterial PO2 from 107.2 to 23
Torr, but only minor changes in PCO2 and acid-base balance were seen, and there
was no indication that anaerobic glycolysis had been activated during this long dive.'^^
By contrast. Wood et al.^^ found that as little as 15 min of vigorous underwater
swimming by green sea turtles produced a fall in blood pH from 7.66 to 7.21 and
a substantial increase in blood lactate (from 0.6 to 10.4 mmol 10- The importance
of underwater activity in influencing changes in blood acid-base balance is demon
strated from forced-dive studies. For example, in the loggerhead, blood lactate
increased to 10 mmol after 30 min forced dive, and at 90 min forced submergance
the venous PO2 approached zero, pH had fallen to to 6.7, and PCO2 had risen to
120 Torr.^^ A similar depletion of blood O2 has been seen in the green sea turtle
after 1 0 0 min of forced dive,^^ but in this case blood lactate did not increase until
after air respiration was resumed. However, Hslatala et al.^^^ found blood lactate
levels greater than 35 mmol after a similar period (2 h) of forced submergence
in green sea turtles. As a practical consideration it might be expected that turtles
trapped and struggling in shrimp trawls undergo stress-related changes in acid-base
balance. And indeed, blood lactate levels can reach as high as 16.2 mmol in
loggerheads after 30 min in trawls^ and for Lepidochelys kempi as little as 7.3 min
trapped in commercial shrimp trawls produced a fall in blood pH from 7.40 to 7.04
and an increase in lactate to as high as 10.2 mmol Such disturbances in acid-

base balance could compromise viability, especially if the same turtles are recaptured
in shrimp trawls before they can fully recover, which may take many hours.*
Substantial (but sustainable) changes in acid-base balance might also be expected
in underwater hibernating turtles.This is certainly the case for the fresh water
turtle Chrysemys pietà, which survives at least 3 months submerged in cold (3°C)
anoxic water, and shows dramatic changes in plasma composition, particularly
calcium and magnesium, with lactate concentrations rising to as much as 2 0 0 mmol
|-i 63 information if such changes occur in sea turtles, but there are
reports of underwater aggregations of dormant sea turtles overwintering in Baja
California {Cheionia mydasY"^ and Cape Canaveral, FL (Caretta carreta).^^
In the fresh water turtle (T. scripta) effective intracellular pH regulatory mech
anisms protect the intracellular pH of the major internal organs (liver, heart, skeletal
muscle, brain) from the large decreases in blood pH that occur on anoxia.^*^ Again
we have no information on these processes in sea turtles, but it is likely that they
occur.
13.4.2 T emperature and pH

Temperature has profound effects on the acid-base balance of reptiles. In many
reptiles increasing body temperature is associated with decreasing blood pH such
that the relationship between arterial blood pH and body temperature (ApH/AT) is
similar to that for neutral water (from 10 to 30°C the value of ApH/AT for neutral
water ranges from -0.018 to -0.016)^^ with the result that the relative alkalinity of
the blood remains constant.^*^ It has been suggested that this shift maintains a constant
net charge state in blood proteins.While there are a number of exceptions to this
rule (for example, the varanid lizards exhibit an almost constant arterial blood pH
over a wide range of body temperatures),^*^* sea turtles appear to be conformers. For
example, for immature green turtles the ApH/AT ratio for blood (-0.0134) is similar
to that for neutral water over the body temperature range 25 to 35°C,^* and for
loggerhead venous blood the ApH/AT ratio is -0.017 over the range 15 to 30°C
(Figure B.da)."^^
However, the mechanisms that produce constant relative alkalinity may be dis
turbed when sea turtles are stressed outside their normal temperature r a n g e s .^ In
green sea turtles, the relationship between blood pH and temperature breaks down
when temperatures fall from 25 to 15°C,^* and 15°C may be below the temperature
range where green turtles are normally active.^^ By contrast, in the loggerhead sea
turtle the ApH/AT relationship is maintained to as low as 15°C.^^ But between 15
to 19°C there is a sudden break, with a sharp rise in blood pH and an increase in
blood lactate (Figure 13.4a).^^ The loggerhead has a much wider geographical range
than the green, being routinely found from the tropics to Virginia, U.S., and therefore
would normally experience colder body temperatures. Temperatures around 12°C
may be the stunning temperature in this species, causing the animals to lose buoyancy
control and “float”.^^
Temperature (®C)
FIGURE 13.4 The effect of temperature on blood pH in loggerhead sea turtles (a^^ and
on the air convection requirement in the green turtle (b)7*
13.4.3 Regulation of p H
In many reptiles, and particularly turtles, the temperature-related blood pH changes
are achieved through altering the efficiency of lung gas exchange, whereby the
increase in oxygen consumption (VO2 ) caused by a rise in body temperature is not
matched by an equivalent increase in lung ventilation (VE).^"^ This results in an
increase in blood PCO2 with body temperature and a fall in the / VO2 ratio
(called the air convection requirement, ACR). If the blood CO2 content and therefore
blood [HCO3 ], remain constant, blood pH is controlled by the change in PC0 2 .^^
Indeed, increasing body temperature results in an increase in blood PCO2 and a
decrease in ACR (Figure 13.4b) in both the green^^ and loggerhead.^^ But the
situation with respect to temperature-related changes in [HCO3 ] is somewhat con
fused by the fact that different authors have used different functions in calculating
the value of the apparent pK for the Henderson-Hasselbach equation (see above).
One of the most widely used set of values^^ only accounts for the effect of temper
ature on apparent pK and makes no correction for the effect of pH. The more accurate
functions given by Severinghaus^^ which take into account both pH and temperature,
have been derived for the mammalian situation. Heisler^^ has developed complex
equations for deriving pKa values in ectotherms that have been successfully applied
to blood plasma of the Kemp’s ridley sea turtle.^^ Applying the Severinghaus^^
functions to the data for the green turtle of Kraus and Jackson^’ shows a decline in
bicarbonate with temperature, and a similar result has been calculated for the log-
gerhead.^2 indicates that in sea turtles the temperature-related adjustments of
blood pH are not simply a matter of respiratory adjustments of blood PCO2 , but are
also managed at the tissue levels. More direct evidence of the involvement of
nonrespiratory mechanisms in temperature-related pH regulation is seen in the log
gerhead, where the fall in blood pH produced by increasing temperature is associated
with an increase in plasma potassium.^^ An identical association between plasma K+
and body temperature has been found in a field study of a large number of “wild”
loggerheads.^ A similar relationship between blood pH and K+ concentration is found
in mammals and is thought to be the result of intracellular buffering of blood pH
whereby extracellular H+ ions are taken up by the cell in exchange for K+ ions."^^
13.5 C O N C L U D IN G REM ARKS

Understanding the mechanisms by which sea turtles maintain internal osmotic and
ionic homeostasis and regulate the related acid-base balance is not only of interest
for comparative physiology, but it is important for an appreciation of the intimate
relationship these species have with their environment. We have discussed how
physiological attributes can be intimately linked to species and life history differ
ences. For instance, the fact that in the leather-back sea turtle the salt gland is
continuously active may be related to their jellyfish diet producing a higher salt
burden than that experienced by other turtle species. The comparatively large and
active salt glands of hatchlings compared to adults is probably related to their
obligatory drinking of seawater when they first enter the sea. Species differences in
the temperature-related failure of acid-base regulation may be one of the physiolog
ical factors that determine the geographical range of a species. And finally, there is
evidence that disruption in acid-base balance is the major contributing cause of
trawl-capture mortality. But our understanding of these processes is only sketchy;
little is known of the mechanisms that control sea turtle salt gland activity, even less
of kidney function. Clearly, much work on sea turtle osmotic and acid-base physi
ology remains to be done, and, considering the vulnerability of these last surviving
of the truly marine reptiles, the need is urgent.
REFEREN CES
1. Lutz, R L. and Dunbar-Cooper, A., Variations in the blood chemistry of the loggerhead
sea turtle Caretta caretta, Fish. Bull., 1985, 37-43, 1987.
2. Bolton, A. B. and Bjorndal, K. A., Blood profiles for a wild population of green
turtles Chelonia mydas in the Southern Bahamas: size-specific and sex-specific rela
tionships, J. Wildl. Dis., 28, 407-413, 1992.
3. Holmes, W. N. and McBean, R. L., Some aspects of electrolyte excretion in the green
turtle, Chelonia mydas mydas, J. Exp. Biol., 41, 81-90, 1964.
4. Staub, F. J., The Effects of Environmental Osmotic Variation on Serum Electrolyte
Levels and on the Histology and Histochemistry of the Salt-Secreting (Lachrymal)
Gland of the Green Sea Turtle, Chelonia mydas, Ph.D. thesis, Lfniversity of Rhode
Island, Kingston, 1975.
Carr, A. E, The Passing of the fleet, A./.ß.S. Bull., 4, 17-19, 1954.
Bentley, P. J., Osmoregulation, in Biology o f the Reptilia, Vol. 5, Gans, C. and Dawson,
W. R., Eds., Academic Press, New York, 1976, 365-412.
Lutcavage, M. E., Lutz, P. L., and Baier, H., Gas exchange in the logger head sea
turtle Caretta caretta, J. Exp. Biol., 131, 365-372, 1987.
Stokes, G. D., Permeability Characteristics and Possible Pore Structure of Reptilian
and Artificial Keratin Membranes, Ph.D. thesis, Pennsylvania State University, Uni
versity Park, 1981.
9. Lillywhite, H. B. and Maderson, P. F. A., Skin structure and permeability, mBiology
o f the Reptilia, Vol. 12, Physiology, Gans, C. and Pough, F. H., Eds., Academic Press,
New York, 1982, 397-442.
10. Evans, D. H., The sodium balance of the euryhaline marine logger head turtle,
Carretta carretta, J. Comp. Physiol., 83, 179-185, 1973.
11. Kooistra, T. A. and Evans, D. H., Sodium balance in the green Xnxút,Chelonia mydas,
in seawater and freshwater,/. Comp. Physiol., 107, 229-240, 1976.
12. Evans, D. H., Osmotic and ionic regulation, in The Physiology o f Eishes, Evans,
D. H., Ed., CRC Press, Boca Raton FL, 1993, 315-342.
13. Mazzotti, F. J. and Dunson, W. A., Osmoregulation in crocodilians, Am. ZooL, 29,
903-920, 1989.
14. Bjomdal, K. A., Nutritional ecology of sea turtles, Copem, 1985(3), 736-751, 1985.
15. Thompson, D. H., A comparative Study of the Anatomy and Histology of the Oral
Cavity and Alimentary Canal of Two Sea Turtles, M.S. thesis, James Cook University
of North Queensland, Australia, 1980.
16. Lutz, P. L., Unpublished.
17. Yoshie, S. and Honma, Y, Light and scanning electron microscopic studies on the
esophageal spines in the Pacific Ridley turtle, Lepidochelys olivácea, Arch. Histol.
Jap., 38(4), 339-347, 1976.
18. Guibe, J., L’Appareil digestif, in Trade de Zoologie, Tome XIV, Grasse, P. P, Ed.,
Libraires de TAcademie de Medicine, Paris, 1970, 521-548.
19. Bennett, J. M., Taplin, L. E., and Grigg, G. C., Sea water drinking as a homeostatic
response to dehydration in hatchling logger head turtles Caretta caretta. Comp.
Biochem. Physiol., 83A, 507-513, 1986.
20. Marshall, A. T. and Cooper, P. D., Secretory capacity of the lachrymal salt gland of
hatchling sea turtles, Chelonia mydas, J. Comp. Physiol., 157B, 821-827, 1988.
21 . Dunson, W. A., Control mechanisms in reptiles, in Mechanisms o f Osmoregulation
in Animals: Regulation o f Cell Volume, Gilles, R., Ed., Wiley Interscience, New York,
1979, 273-322.
22. Minnich, J. E., The use of water, in Biology o f the Reptilia, Vol. 12, Gans, C. and
Pough, F. H., Eds., Academic Press, New York, 1982, 325-396.
23. Dunson, W. A., Salt glands in reptiles, in Biology of the Reptilia. Vol. 5, Gans, C.
and Dawson, W. R., Eds., Academic Press, New York, 1976, 329-353.
24. Schmidt-Nielsen, K. and Fänge, R., Salt glands in marine rcpiilQS, Nature (London),
183. 783-785. 1958.
25. Hudson, H. D. and Lutz, P. L., Salt gland function in the leather back sea turtle
Dermochelys coriacea, Copeia, 1986(1), 247-249, 1986.
26. Peaker, M. and Linzell, J. L., Salt Glands in Birds and Reptiles, Cambridge University
Press, New York, 1975, 307 pp.
27 Ellis, R. A. and Abel, J. H., Intercellular channels in the salt-secreting glands of
marine turtles, Sdcncc, 144, 1340-1342, 1964.
28 Abel, J. H. and Ellis, R. A., Histochemical and electron microscope observations on
the salt secreting lachrymal glands of marine turtles, Am. J. Anat., 118, 337-358,
1966.
29. Marshall, A. T. and Saddlier, S. R., The duct system of the lachrymal gland of the
green sea turtle, Chelonia mydas. Cell Tissue Res., 257, 399-404, 1989.
30. Marshall, A. T., Intracellular and luminal ion concentrations in sea turtle salt glands
determined by X-ray microanalysis,./. Comp. Physiol., 159, 609-616, 1989.
31. Epstein, F. H., Stoff, J. S., and Silva, R, Chloride secretion by shark rectal gland,
in Chloride Transport Coupling in Biological Membranes and Epithelia, Gerenscer,
G. A., Ed., Elsevier Press, Amsterdam, 1984, 374-378.
32. Schmidt-Nielsen, K., The salt secreting gland of marine birds. Circulation, 21,
955-967, 1960.
33. Nicholson, S. E. and Lutz, P. L., Salt gland function in the green seaturtle,/. Exp.
Biol., 144, 171-184, 1989.
34. Hudson, H. D. and Lutz, P. L., Salt gland function in the Atlantic green turtle.
Physiologist, 28, 343, 1985.
35. Lutz, P. L., Lutcavage, M., and Hudson, D., Physiological effects of oil on sea turtles,
in Study on the Effects of Oil on Marine Turtles, Mineral Management Service
Contract Number 14-12-0001-30063, 1986, 91-127.
36. Shuttleworth, T. J., Thompson, J. L., and Dantzler, W. H., Potassium secretion by
nasal salt glands of desert lizard Sauromalus obesus, Am. J. Physiol., 22, R83-R90,
1987.
37. McFarland, L. D., Minimal salt load required to induce secretion from the nasal
glands of sea gulls, Nature, (London), 204, 1202-1203, 1964.
38. Dantzler, W. H. and Holmes, W. N., Water and mineral metabolism in reptilia, in
Chemical Zoology, Vol. II, Florkin, M. and Scheer, B. T, Eds., Academic Press,
New York, 1974, 277-336.
39. Prange, H. D. and Greenwald, L., Effects of dehydration on the urine concentration
and salt gland secretion of the green sea turtle. Comp. Biochem. Physiol., 6 6 A,
133-136, 1980.
40. van Lennep, E. W. and Komnick, H., Fine structure of the nasal salt gland in the
desert lizard Uromastyx acanthinurus, Cytobiologie, 2, 47-67, 1970.
4L Shuttleworth, T. J. and Thompson, J. L., Secretory activity insalt glands of birds and
turtles: stimulation via cyclic AMP, Am. J. Physiol, 21, R428-R432, 1987.
42. Marshall, A. T., Hyatt, A. D., Phillips, J. G., and Condron, R. G., Isosmotic secretion
in the avian nasal salt gland: X-ray microanalysis of luminal and intracellular ion
distributions,/. Comp. Physiol, 156B, 213-227, 1985.
43. Ellis, R. A., Goertemiller, C. C., Jr., and Stetson, D. L., Significance of Teaky’ cell
junctions in avian salt gland, Nature, (London), 268, 555-556, 1977.
44. Schmidt-Nielsen, B., The renal concentrating mechanism in insects and mammals: a
new hypothesis involving hydrostatic pressures. Am. J. Physiol, 37, R1087-R1100,
1995.
45. Rose, B. D., Clinical Physiology o f Acid-Base and Electrolyte Disorders, McGraw-
Hill. New York. 1977.
46. Ituniza, F. C., Venosa, R. A., Pujol, M. G., and Quintas, N. B., Alpha-melanocyte
stimulating hormone stimulates sodium secretion in the salt gland of the duck, I.
Comp. Endocrinol., 87, 369-374, 1992.
47. Solomon, S. E., The moiphology of the kidney of the green turtle Cheionia mydas,
J. Anat., 140, 355-369, 1985.
48. Stetson, D. L., Turtle urinary bladder: regulation of ion transport by dynamic changes
in plasma membrane area, Am. J. Physiol, 257, R973-R981, 1989.
49. Bjomdal, K. A., Urine concentrations of ammonia, urea and uric acid in the green
turtle, Cheionia mydas. Comp. Biochem. Physiol, 63A, 509-510, 1979.
50. Khalil, F., Excretion in reptiles: non-protein nitrogen constituents of the urine of the
sea turtle, Cheionia mydas, J. B iol Chem., 171, 611-616, 1947.
51. Dantzler, W. H., Renal function (with special emphasis on nitrogen excretion), in
Biology of the Reptilia, Voi. 5, Cans, C. and Dawson, W. R., Eds., Academic Press,
New York, 1976, 447-503.
52. Schmidt-Nielsen, B. and Davies, L. E., Fluid transport and tubular intercellular spaces
in reptilian kidneys. Science, 159, 1105-1108, 1968.
53. Dantzler, W. H. and Schmidt-Nielsen, B., Excretion in fresh water im ilt Pseudemys
scripta and desert tortoise Gopherus agassizii. Am. J. Physiol, 210, 198-210, 1966.
54. Stewart, P. A., Independent and dependent variables of acid-base control, Respir.
Physiol, 33, 9-26, 1978.
55. Busa, W. B. and Nuccitelli, R., Metabolic regulation via intracellular pH, Am. J.
Physiol, 246, R409-R438, 1984.
56. Lutz, P. L. and Bentley, T. B., Adaptations for diving in the sea turtle, Copcm, 1985,
671-697, 1985.
loggerhead sea turtle, J. Exp. Mar. B iol Ecol, 147, 287, 1991.
58. Wood, S. C., Gatz, R. N., and Glass, M. L., Oxygen transport in the green sea turtle,
/. Comp. Physiol, 154, 275-280, 1984.
59. Berkson, H., Physiological adjustments to prolonged diving in the Pacific Green turtle
Cheionia mydas agassizii. Comp. Biochem. Physiol, 18, 101, 1966.
60 Hslatala, M. P., Standaert, T. A., and Pierson, D. J., Hochachka, P W., Owen, T. G.,
Allen, J. R, and Whittow, G. C., Multiple end products of anaerobiosis in diving
vertebrates. Comp. Biochem. Physiol, 50B, 17, 1975.
61 Stabenau, E. K., Hemming, T. A., and Mitchell, J. F., Respiratory, acid-base and ionic
status of Kemp’s ridley sea turtles Lepidochelys kempi subjected to trawling. Comp.
Biochem. Physiol, 99A, 107-111, 1991.
62. Lutz, P. L., Interaction between hypometabolism and acid-base balance in vertebrates.
Can. J. Z ool, 67, 3018-3028, 1989.
63. Ultsch, G. R. and Jackson, D. C., Long-term submergence at 3°C of the turtle,
Chrysemys pietà bellii, in normoxic and severe lyhypoxic water. I. Survival, gas
exchange and acid-base status,/. Exp. B iol, 96, 11-28, 1982.
64. Felger, R. S., Cliffton, K., and Regal, P. J., Winter dormancy in sea turtles: independent
discovery and exploitation in the Gulf of California by two local cultures, Science,
191, 283-284, 1976.
65. Carr, A., Ogren, L., and MeVea, C., Apparent hibernation by the Atlantic logger head
turtle Caretta caretta off Cape Canaveral, Florida, B io l Conserv., 19, 7, 1980-81.
66 Wasser, J. S., Warburton, S. J., and Jackson, D. C., Extracellular and intracellular
acid-base effects of submergence anoxia and nitrogen breathing in turtles, Respir.
Physiol, 83, 293-252, 1991.
67. Truchot, J. R, L’Equilibre acido-basique extracellulaire et sa regulation dans les divers

groupes animaux, 7. Physiol, 11, 529-580, 1981.
6 8 . Howell, B. J. and Rahn, H., Regulation of acid-base balance in reptiles, 'mBiology
of the Reptilia. Vol. 5, Cans, C. and Dawson, W. R., Eds., Academic Press, New York,
1976, 335-363.
69. Reeves, R. B., Temperature-induced changes in blood acid-base status: pH and PCQ
in a binary buffer, J. AppL Physiol, 40, 752-761, 1976.
70. Wood, S. C., Glass, M. L., and Johansen, K., Effects of temperature on respiratory
and acid-base balance in a monitor lizard,/. Comp. Physiol, 16, 287-296, 1977.
71. Kraus, D. R. and Jackson, D. C., Temperature effects on ventilation and acid-base
balance of the green turtle, Am. /. Physiol, 239, R254-R258, 1980.
72. Lutz, P. L., Bergey, A., and Bergey, M., Effects of temperature on gas exchange and
/. Exp. B iol, 144, 155-169, 1989.
73. Jackson, D. C., Respiration and respiratory control in the green turtle, Che Ionia
mydas, Copeia, 1985, 664-671, 1985.
74. Shelton, G., Jones, D. R., and Milsom, W. K., Control of breathing in ectothermic
vertebrates, in Handbook o f Physiology, Section 3, The Respiratory System, Fishman,
A. R, Ed., American Physiological Society, Bethesda, MD, 1986, 857-909.
75. Jackson, D. C., Strategies of blood acid-base control in ectothermine vertebrates, in
A Companion to Animal Physiology, Taylor, C. R., Johansen, K., and Bolis, L,, Eds.,
Cambridge University Press, Cambridge, 1982, 73-90.
76. Severinghaus, J. W., Blood gas concentration, in Handbook o f Physiology, Respira
tion, American Physiological Society, Washington, D.C., 1965, 1475-1487.
77. Heisler, N., Buffering and transmembrane ion transfer processes, in Acid-Base Reg
ulation in Animals, Heisler, N., Ed., Elsevier, Amsterdam, 1986, 3-47.
78. Stabenau, E. K. and Heming, T. A,, Detennination of the constants of the Henderson-
Hasselbach equation, a CO2 and pKa in sea turtle plasma, J. Exp. Biol, 180, 311-314,
1993.
79. Kennish, M. J., Practical Handbook of Marine Science, 2nd ed., CRC Marine Science
Series. CRC Press. Boca Raton. FL. 1994.
1 /I Health Problems and
Diseases of Sea Turtles
Robert H, George
CO N TEN TS
14.1 Introduction........................................ 364

14.2 Nutritional Problems and Requirements....... ............................................ 365
14.2.1 Malnutrition.......................................... 365
14.2.2 Metabolic Bone Disease............................................................... 365
14.2.3 Iron Deficiency.............................................................................366
14.3 Bacterial Diseases....................................................................................... 366
14.3.1 Ulcerative Stomatitis — Obstructive Rhinitis — Pneumonia....367
14.3.2 Mycobacterial Infections.............................................................. 367
14.3.3 Dermal Bacterial Infections..........................................................368
14.3.4 Bacterial Encephalitis................................................................... 368
14.3.5 Chlamydial Infections................................................................... 369
Ì4À Mycotic Diseases........................................................................................ 369
14.4.1 Dermal or Superficial Mycoses....................................................369
14.4.2 Systemic or Deep Mycoses.......................................................... 370
14.5 Viral Diseases............................................................................................. 370
14.5.1 Heipesvirus Respiratory Disease..................................................370
14.5.2 Gray-Patch Disease....................................................................... 371
14.6 Fibropapillomatosis..................................................................................... 371
14.6.1 History...................................................... 371
14.6.2 Epidemiology................................................................................ 372
14.6.3 Pathology....................................................................................... 372
14.6.4 Histology....................................................................................... 372
14.6.5 Etiology......................................................................................... 373
14.6.6 Detection....................................................................................... 374
14.6.7 Treatment...................................................................................... 374
14.7 Parasitic Conditions.................................................................................... 375
14.7.1 Ectoparasites................................................................................. 375
14.7.1.1 Leeches......................................................................... 375
14.7.1.2 Barnacles...................................................................... 375
14.7.2 Endoparasites................................................................................ 375
14.7.2.1 Protozoans..................................................................... 375
0-8493-8422-2/97/$0.00+$ 50
14.7.2.2 Helminths..................................................................... 376

14.8 Environmental Health Problems................................................................. 377
14.8.1 Naturally Occurring Problems...................................................... 377
14.8.1.1 Traumatic Injuries........................................................ 377
14.8.1.2 Hypothermic Stunning................................................. 377
14.8.2 Anthropogenic Problems.............................................................. 378
14.8.2.1 Entanglement........................................ 378
14.8.2.2 Ingestion of Anthropogenic Debris.............................378
14.8.2.3 Chemical Pollutants..................................................... 379
14.9 Biomedical Techniques............................................................................... 379
14.9.1 Hematology and Plasma............................................................... 379
14.9.2 Anesthesia........ .............................................................................380
14.9.3 Resuscitation................................................................................. 380
References.............................................................................................................. 382
Sea turtles are affected by a variety of health problems. Some of these problems are
naturally occurring processes, such as parasitism, and are observed in both wild and
captive animals. Other conditions, such as certain nutritional deficiencies, are essen
tially a result of prolonged captivity. The term captive animal will be used throughout
this chapter to indicate animals held in confinement, including those in oceanariums,
commercial farms, research institutions, and head-start programs.
The environment sea turtles occupy predisposes them to some of the disease
processes with which they must contend. For example, how a turtle deals with
exposure to an infectious organism and its response to physical trauma is greatly
affected by environmental stress. While little is known about the sea turtle immune
system, it can be assumed that, like other animals, they suffer a reduction in ability
to fight disease when stressed. In stressed sea turtles, the adrenal glands release
corticosterone^’2 which may reduce the humoral and/or cell-mediated defense mech
anisms of the turtle, thus inhibiting the ability of the immune system to respond to
infectious agents. Stresses can be environmental (salinity, pollution, temperature,
etc.), nutritional, or physical (trauma). Low ambient temperature has been shown to
reduce immunoglobulin production^ and the competence of the immune response in
reptiles.
This chapter describes some of the health and disease problems commonly
observed in both wild and captive sea turtles. Only those problems with high
morbidity and mortality rates or those syndromes which are commonly noted are
included. The health problems discussed here are grouped according to etiology. In
cases where the cause of a disease process is unproven or there are multiple factors
involved in the pathogenesis, the disease is described in the section concerning the
primary or most probable agent.
Health Problems and Diseases of Sea Turtles 365
14,2 N U T R IT IO N A L PR O BLEM S A N D R EQ U IR EM EN TS
Captive sea turtles have been fed a variety of diets and held under a variety of
conditions. They have been successfully raised, maintained, and propagated on
commercial pelleted turtle food, modified pelleted trout ration, and balanced gelatin
diets. Most prepared diets range from 25 to 35% protein, but diets up to 45% protein
have been used."^ Kemp’s ridleys^ (Lepidochelys kempii) in government-sponsored
head-start programs have been maintained by feeding them a commercial diet at the
rate of 5% body weight daily (BWD) to hatchlings and 1.5% BWD to yearlings. At
the Cayman Turtle Farm^ Grand Cayman, B.W.I., yearling green turtles (Chelonia
mydas) have been successfully fed a similar diet at the rate of 1.2% BWD, and 44-
month-old animals were fed 0.8% BWD.
14.2.1 M alnutrition
Wild sea turtles that are able to successfully fulfill their nutritional needs do not
appear to exhibit signs of nutrient deficiencies. Free-ranging animals found in a
weak and emaciated condition generally have a physical or medical problem that
inhibits their ability to procure food. Such problems can include damage to flippers,
mouth parts, or the eyes. Sea turtles suffering from a variety of chronic diseases
and/or carrying a heavy parasite burden may cease to feed for long periods of time
and waste away. Cachectic myopathy^ is the general wasting exhibited by nonfeeding
and/or heavily parasitized animals. These individuals are characterized by concave
plastrons, sunken eyes, muscular atrophy, and reduced weight. Analysis of blood
samples will often show anemia, hypoproteinemia, and hypoglycemia. Necropsy
reveals reduced muscle mass and little fat. In extreme cases, serous atrophy of fat
can be observed.
Cachectic myopathy can be observed in captive turtles as well as wild ones. Any
disease process causing cessation of feeding will eventually produce a malnourished
sea turtle. For example, hatchlings fed peneid shrimp^ and larger turtles fed inap
propriate vegetation^ have been noted to develop gastrointestinal (GI) obstructions.
Obstructed turtles will stop feeding, waste away, and die unless the blockage is
relieved.
14.2.2 M etabolic B one D isease
Because humans control the dietary and environmental parameters of turtles in

captivity, there are several nutritional problems that have been documented in captive
animals. Metabolic bone disease due to improper calcium and phosphorous levels
has been documented in a number of animals.^’^®Diets low in Ca and high in P will
stimulate the parathyroid glands to release parathyroid hormone (PTH), which helps
to maintain serum Ca levels by mobilizing Ca from the bony matrix. When logger-
head (Caretta caretta) hatchlings at the North Carolina Aquarium at Pine Knoll were
inadvertently fed diets consistently low in calcium and high in phosphorous (prima
rily freeze-dried krill), the turtles became hypocalcémie^^ (median Ca of 2 . 6 mg/dl;
normal is 6.4-9 mg/dl) and hyperphosphatémie (median P of 14.2 mg/dl; nonnal is

4.6 to 7.4 mg/dl). The resulting hyperparathyroid condition induced demineralization
of bones and resulted in pathological fractures. Calcium and phosphorous levels
recorded^^ for juvenile loggerheads in the Chesapeake Bay indicate a Ca to P ratio
of less than 1:1 may be normal for some life stages of wild sea turtles. Although
the optimum Ca to P ratio for most sea turtles has not yet been determined, rapidly
growing captive animals continually fed diets below the generally accepted 1 : 1 to
2:1 ratio are prone to develop secondary nutritional hyperparathyroidism. A gelatin-
based diet has been developed to meet the Ca and P needs of young sea turtles.’^ It
is also essential to provide captive turtles adequate ultraviolet radiation and dietary
vitamin D3 , which are both necessary for Ca uptake from the intestine. The exact
UV wavelength required to assure optimum Ca absorption is unknown, but the use
of lights commercially available for the culture of reptiles should provide sufficient
UV radiation if used without plastic diffusers.
14.2.3 Iron D eficiency
Juvenile captive turtles regularly fed diets consisting solely of squid and fish have
developed life-threatening anemias.^ Kemp’s ridleys and loggerheads have been
presented for care because they were found listless, floating, and hyperventilating
due to poor oxygen-carrying capacity. On examination, packed cell volumes (PCVs)
varyied from 4 to 10%. This contrasts with normal PCVs for juvenile turtles in the
Chesapeake Bay, which range from 24 to 34%. All of the anemic turtles examined
were emaciated and showed signs of malnutrition. The animals responded well to
dietary adjustment and iron supplementation, after which they exhibited sequential
rises in their PCVs, weight gains, and increased activity.
14.3 B A C T E R IA L D ISEA SES

While bacterial infections occur in free-roaming sea turtles, the incidence is relatively
rare. The combination of a tough integument and competent immune system mini
mizes the opportunities for bacteria to gain entrance and colonize tissues of sea
turtles. Traumatic injury to dermal tissues and aspiration of seawater are the primary
routes by which bacteria enter the body of a turtle, and these entry points can lead
to abcesses and aspiration pneumonias, respectively. Once an infection occurs,
bacteria can enter the bloodstream and be disseminated throughout the body. Such
a bacteremia may result in multifocal abcesses or lethal septicemia.
By contrast, bacterial infections are common in captive sea turtles. Large pop
ulations of bacteria, which may accumulate in closed recirculating water systems or
crowded open flow-through systems, provide a ready source of pathogens. Turtles
confined together often traumatize each other by aggressive biting or damage their
integument on tank sides, barriers, or scenery. Such dermal injuries allow the patho
genic bacteria to enter tissues, and possibly the bloodstream, often resulting in
clinical disease. Turtles that aspirate water from a system containing poor-quality
water are more apt to inhale pathogenic organisms and develop pneumonia than
wild animals or captive animals kept in good-quality water. In addition, sick sea
turtles in close proximity to healthy turtles can “seed” the water with bacteria,
therefore increasing the likelihood of spread of disease. Any infectious agent, be it
bacterial, viral, fungal, or parasitic, can be easily transmitted in a closed system.
A variety of bacteria have been cultured from diseased sea turtles. The most
commonly cultured pathogens are Aeromonas hydrophila, Vibrio alginolyticus,
Escherichia coli, Citrobacter sp., Enterobacter sp., Proteus sp.. Pseudomonas ^p..
Salmonella sp., Mycobacterium sp., Edwardsiella sp., Arizona sp., and Elavobacte-
rium sp. Quite often, several species have been cultured from a single lesion. In
such cases it has not been clear whether only one type of bacteria was responsible
for the disease process or if the various bacteria functioned synergistically. Some of
the organisms cultured could have been contaminants and not related to the infection.
To further confuse the issue, some cultured lesions yielded fungal organisms in
addition to the bacteria.
14.3.1 U lcerative S tomatitis O bstructive R hinitis

P neumonia
A group of bacterial diseases known as ulcerative stomatitis (US), obstructive rhinitis

(OR), and bronchopneumonia (BP) occurs in captive sea turtles. The US-OR-BP
complex has been reported to cause high mortality rates in hatchling and juvenile
green and loggerhead turtles,^’ and the diseases can be present singly or in com
bination. The first sign of disease is often a plug of caseous material in one of the
nares or in the pharyngeal area of the oral cavity. Sick turtles cease to feed, become
listless, and float. If BP is present, there is often a loss of equilibrium or the ability
to maintain an evenly distributed neutral bouyancy, and the turtle floats with one
side below the horizontal. Untreated, the mortality rate can be as high as 70%.
Topical treatment, systemic antibiotics, and isolation of affected individuals can
greatly reduce the mortality rate.
Depending on the diseases involved, necropsy of turtles suffering from the US-
OR-BP complex can reveal a variety of lesions. Caseous material can be found in
the nares, oral cavity, trachea, and bronchi. The oral mucosa may be inflamed and
ulcerated. Oral lesions are most commonly located lateral to the tongue and on the
caudal aspect of the hard palate. If infected, the lungs may exhibit focal areas of
pneumonia or a generalized BP. The varieties of bacteria most commonly cultured
from affected turtles are Vibrio alginolyticus, Aeromonas hydrophila, and Elavobac-
terium sp.
14.3.2 M ycobacterial Infections
There are a few reports of mycobacterial infections occurring in captive turtles.^ '^’^^
In an outbreak of tuberculosis, acid-fast rods were found in histological preparations
of the lungs, livers, spleens, and kidneys from 6 of 120 hatchling green turtles
collected from the wild. Signs of infection appeared in the six animals over a two-
month period, and all of the infected animals died. M. avium was cultured from all
of these turtles, and the investigators considered the possibility of airborne trans
mission. In a separate case, M. marinum was cultured from the lungs of an infected
head-started Kemp’s ridley. M. marinum may be cultured from seawater, suggesting

that this infection may have been due to ingestion or aspiration of seawater,
14.3.3 D ermal B acterial Infections
Bacterial infections of the dermis are both common and highly visible in captive
sea turtles, and a variety of bacterial dermatitis syndromes are described in the
literature.^’^’^’^^'^^ Names such as focal errosive dermatitis (FED), septicemia ulcer
ative cutaneous disease (SCUD), and papillary dermatitis (PD) not only identify a
type of bacterial infection, but describe the predominant lesions produced by such
an infection. The most frequently cultured organisms are Aeromonas sp., V. algi-
nolyticus, Pseudamonas sp., Proteus sp., and Citrobacter sp. The same genus or
suite of microorganisms may be cultured from more than one of these diseases; it
is not understood why a single bacterial species, such as A. hydrophila, may cause
a variety of different disease syndromes.
Commonly described lesions involve ulceration and discoloration of the dermis
(FED, SCUD). The ulceration may be superficial or deep. Dermal infections with
deeply ulcerative lesions (SCUD) may provide a route for bacteria to enter the
bloodstream and lead to lethal septicemias. Other bacterial syndromes, such as PD,
are more superficial and proliferative in nature, and are characterized by the devel-
opement of papilla-like projections.
Response of dermal infeetions to antibiotic treatment varies greatly. The best
results have been obtained with a combination of topical and systemic antibiotics.
Efficacy of antibiotic treatment is greatly enhanced by an improvement in water
quality. In addition, the spread of disease is reduced by isolation of sick individuals
and frequent water changes.
14.3.4 B acterial Encephalitis
Corynebacterium sp. have been implicated in a multifocal bacterial encephalitis in

stranded juvenile loggerhead sea turtles. Turtles suffering from this form of enceph
alitis have been found in the Chesapeake Bay^^ and along the coast of Florida.
Though found floating or stranded, affected turtles had no apparent medical prob
lems. While dull and listless when undisturbed, the animals moved their flippers
spastically when handled. All of the affected individuals also showed a characteristic
hyperflexion of the neck when handled or otherwise disturbed.
Turtles with encephalitis had high white blood cell counts and elevated creatinine
kinase levels. On necropsy, multiple areas of hemorrhage and necrosis were visible
in the brain and meninges. Several of the turtles had caseous nodules between the
cerebellum and the brain stem.
Histological examination of affected brain tissue revealed the presence of a
granulomatous meningoencephalitis (GME). The granulomatous reaction was char
acterized by the presence of a large number of multinucleate giant cells. A Coryne-
bactium sp. was cultured from several of the necropsied turtles. Spirorchid fluke
eggs also were found in tissue sections from the turtles sampled. It is unclear whether
the fluke eggs played an active part in the development of a bacterial GME or were
a serendipitous finding.
14.3.5 C hlamydial Infections
Infections due to Clamydia psittaci are common in mammals and birds, but are rare
in reptiles. Only one occurrence of chlamydiosis in captive sea turtles has been
documented.^^ An epizootic disease outbreak at the Cayman Turtle Farm resulted in
the death of several hundred juvenile green turtles. Sick turtles were lethargic, weak,
and often found floating on the surface.
Lesions in necropsied animals were primarily confined to the heart and liver.
Several 1- to 10-mm gray patchy lesions were noted on the hearts. The livers were
enlarged and friable. Microscopic examination of the necropsy specimens revealed
necrosis of the myocardial and hepatic tissues. Chlamydial antigens were detected
in tissue samples from dead turtles, visualized with electron microscopy and cultured
in vitro.
14.4 M Y C O T IC D ISEA SES

There are few documented reports of mycotic (fungal) infections in sea turtles. Those
cases mentioned in the literature have involved both captive^’^’^’^^ and free-roaming^^
sea turtles. Because of their solitary habits, wild turtles seem far less vulnerable to
fungal pathogen exposure. However, due to the relatively crowded conditions often
necessary in mariculture and exhibition facilities, fungi can spread from turtle to
turtle, causing disease outbreaks with both high morbidity and mortality rates.
Mycotic infections are often divided into two groups depending on the locations
of the lesions. Infections of the dermis are termed superficial or dermal mycoses,
while infections of internal organs are referred to as systemic mycoses.
14.4.1 D ermal or S uperficial M ycoses
When fungal organisms colonize the dermal tissues of sea turtles, the infection can
manifest itself in varying ways. Aspergillus sp. have been found in lesions on the
flippers of hatchling loggerheads.^ These animals had focal, black, dry, gangrenous
areas on the leading edges and tips of the front and rear flippers. The areas underwent
auto-amputation and resolved when treated with topical iodine. Leong et al.^
described a variety of necrotic lesions on the head, neck, and shell of head-started
Kemp’s ridleys. The lesions appeared to be due to a number of different fungal
pathogens, and varied from mild to severe. Water quality appeared to be an important
factor in the incidence of disease in both the Aspergillus outbreak and the fungal
infections in captive ridleys. In both cases, improvement in water quality, brought
about by changing from static to free-flowing water systems, greatly reduced the
incidence of infection. The improvement in water quality also increased the efficacy
of treatment provided for infected turtles.
14.4.2 S ystemic or D eep M ycoses
Systemic mycotic infections of wild and captive turtles occur primarily in the lungs.
Yet while essentially a pulmonary disease, mycotic granulomas also can be found
in the liver and throughout the coelomic cavity. Infected animals are listless and
often float or swim with one side lower than the other. Leong et al.^ described various
systemic fungal disease processes in head-started Kemp’s ridleys, including pulmo
nary mycoses, mycotic osteomyelitis, and yolk sac mycoses. The mortality rate for
affected turtles was very high.
In a group of juvenile captive green turtles with systemic mycotic infections,
Jacobson et al.^^ found a number of animals had a consolidated left lung and an
emphysematous right lung, explaining why these sick animals floated in a nonhor
izontal position. On necropsy the lungs were found to contain multiple firm, round,
white nodules of varying diameters. Histological examination of the lesions revealed
caseous necrotic material and branching septate hyphae, surrounded by a granulo
matous inflammatory reaction.
Paecilomyces sp., Sporotrichium sp., Scolecobasidium sp. and Cladosporium
sp. organisms have been isolated from infected t u r t l e s . T h e s e fungi have been
incriminated as the agents responsible for a variety of systemic mycotic infections.
Reduced body temperature and subsequent impairment of the immune system
of a turtle may predispose the animal to a mycotic infection. The cases of fungal
pneumonia reported in the literature often involve animals exposed to cold temper
atures. Lewbart and Medway^^ reported a turtle with mycotic lung disease found
floating off the coast of Delaware in October (no water temperature available). In
the outbreak of mycotic pneumonia described by Jacobson et al.,^^ mortality peaked
during the month with the lowest water temperatures.
14.5 V IR A L D ISEA SES

Only two viral diseases are known to occur in sea turtles: herpesvirus respiratory
disease, and gray-patch disease (GPD). A third condition, green turtle fibropapilloma
disease, may have a viral etiology. This important syndrome, however, will be
discussed in a separate section.
14.5.1 H erpesvirus R espiratory D isease
A herpesvirus^^ was incriminated in an outbreak of respiratory disease involving 15-

to 20-month-old captive green sea turtles. Affected turtles had caseous necrotic
lesions in the mouth, caseous conjunctivitis, tracheitis, and BR These gross lesions
in turtles suffering from what is now termed lung, eye, and tracheal disease (LETD)
were similar to the lesions described by Glazebrook et al.^“^ for ulcerative stomatitis.
However, Glazebrook et al. were unable to isolate viral particles in turtles with US.
Intranuclear inclusion bodies were microscopically visible in tissue preparations
from turtles with LETD. Electron microscopy indicated the inclusion bodies con
sisted of viral particles, and although a variety of Gram-negative bacteria were
isolated from the respiratory tract lesions, a herpesvirus was thought to play a major
role in the disease process.
14.5.2 G ray-Patch D isease

GPD is a stress-induced disease that is caused by a herpesvirus and mainly affects
juvenile captive green sea turtles.Susceptibility to the disease varies with age.
Turtles 2 to 6 weeks of age seldom show skin lesions when innoculated with the
virus, but most of these turtles die quickly. The majority of juveniles in the 8 - to
15-week age group survive when exposed to the virus, and subsequently develop
the classic spreading gray patchy lesions. The stress of high water temperature helps
induce GPD and increases the severity of the lesions. Other stressors, such as
overcrowding and poor water quality, may also play a part in the pathogenesis of
GPD, but high temperature or rapid temperature changes are the most important
environmental stressors. For that reason, disease outbreaks are more common in the
summer months when water temperatures are higher. Turtles that survive GPD appear
to develop immunity and do not acquire the disease again.
Two types of skin lesions may appear on turtles with GPD. Some exhibit a
benign form of the disease in which numerous nonspreading papules or pustules
develop on the head and front flippers. Other turtles have rapidly spreading gray
lesions with raised edges, which represent areas of superficial maceration. These
lesions can cover large areas of the dermis, and in some cases may involve the entire
turtle. Many young turtles die from this more aggressive form of GPD. Microscop
ically, both types of lesions are characterized by hyperkeratosis and acanthosis of
the dermis. Intranuclear inclusion bodies are present in deimal cells, and herpes
like viral particles can be detected with electron microscopy. Transmission studies^'^
indicated the herpesvirus involved in GPD is a naturally occurring virus in wild
green turtles.
14.6 FIBROPAPILLOMATOSIS
Cutaneous fibropapillomatosis is a major epizootic disease affecting a variety of
species of sea turtles. The predominant lesions associated with this disease are
fibromas, cutaneous papillomas, and fibropapillomas. Because it has been reported
primarily in green turtles, this disease syndrome has been designated green turtle
fibropapillomatosis (GTFP). Even so, GTFP has also been documented in loggerhead
turtles and there are reports of it occurring in olive ridley turtles (L. olivácea),
hawksbills (Eretmochelys imbracata), and flatback turtles (Natator depressus)?^
14.6.1 H istory
Lesions consistent with GTFP were first described in green turtles from Florida
waters in 1938,^^ and in 1958 GTFP lesions were reported in Hawaiian green turtles.
Since then, GTFP disease has been identified around the world, in captive turtles as
well as wild animals.
14.6.2 Epidemiology
The incidence of turtles with GTFP has increased rapidly since the 1980s, but varies
between years and locations. In some populations no animals are affected, while
elsewhere up to 92% of individuals may exhibit GTFP lesions. Turtles frequenting
near-shore waters, areas adjacent to large human populations, and areas with low
water turnover, such as lagoons, have a higher incidence of GTFP than individuals
in deeper, more remote waters.
The lesions associated with GTFP seem to increase in size more rapidly in
summer,^^ but current statistics show that more diseased turtles strand in winter than
in summer. As the lesions grow and/or increase in number, they may cause the turtle
to become debilitated. External growths may impede the ability of the animal to
feed or navigate; internal growths can lead to pneumonia, liver disease, intestinal
obstruction, or kidney disease. While the disease slowly progresses in most turtles,
there are reports of spontaneous lesion regression occurring in some affected animals.
Body size and age class are associated with GTFP infection. Juvenile turtles
appear to be most affected: individuals in the 40- to 90-cm carapace length and 10-
to 30-kg weight class^° have the highest incidence of disease and the most extensive
lesions. Conversely, lesions in nesting adults are rare.
14.6.3 Pathology
As previously mentioned, the three primary lesions associated with GTFP are fibro
mas, cutaneous papillomas, and fibropapillomas. The number of lesions may vary
from single to many. They may be smooth or ulcerated, sessile or pedunculated
(Figure 14.1), small (0.1 cm) or large (30 cm), and cutaneous or systemic. Lesions
are commonly found in the conjunctiva, chin, neck (Figure 14.2), flippers, base of
the tail, and axillary and inguinal areas. The lesions are pigmented similarly to the
tissues from which they arise.
When present, visceral lesions occur in the lungs, liver, GI tract, and kidneys of
GTFP-affected turtles. This distribution of visceral lesions suggests a hematogenous
spread of the disease. Visceral growths are usually firm white nodules embedded in
the parenchyma of the affected organ. While the majority of lesions are well demar
cated, some growths may be characterized by an infiltrative margin.
14.6.4 H istology
Typically, the microscopic appearance of lesions associated with GTFP is one of
papillary epidermal hyperplasia and/or dermal hyperplasia.^The histological find
ings vary with the three gross lesion types. Papillomas are made up of hyperplastic
epidermis, fibromas consist of hyperplastic dermal tissues, and fibropapillomas are
characterized by a proliferation of both epidermal and dermal tissues. According to
Jacobson et al.,^^ the earliest detectable lesions are a ballooning degeneration of the
stratum basale and epidermal hyperplasia. Both the dermal cells and fibroblasts
present in these tissues are well differentiated and benign in appearance. The non-
malignant status of GTFP tumors is further supported by flow cytometric DNA
FIGURE 14.1 Pedunculated iibropapilloma lesion. (Photo courtesy of W. G. Teas.)
data.^"^ These data indicate cells from GTFP lesions have a normal cell division cycle
and are benign in nature.
The presence of trematode eggs, perivascular lymphocytic cuffing, and multi-
nucleate giant cells are also common histologic features of GTFP. These features
are often present in other conditions, and the significance of these findings in GTFD
is open to speculation.
14.6.5 Etiology
Based on serological evidence and transmission studies, a subcellular infectious

agenF'" has been implicated in the development of GTFP lesions. The gross appear
ance and microscopic structure of GTFD lesions suggests a papillomavirus as the
causative agent, but no evidence for the presence of such a pathogen has been found.
Work by Herbst et al.^^ and Jacobson et al.^^ points to a herpesvirus as the cause,
but in either case the etiological agent remains unproven.
A variety of environmental stressors have been implicated in the pathogenesis
of GTFP. The presence or absence of these stressors may effect the onset and course
of the disease. Some of these stressors have also been investigated as the possible
FIGURE 14.2 Large fibropapilloma lesions. (Photo courtesy of G. H. Balazs.)
primary etiology of GTFP. Bacterial infections, UV-B radiation, spirorchid blood-

fluke eggs, leeches, chemical contaminants, or a combination of these stressors may
weaken the immune system of the turtle and contribute to the onset of the disease.
There may even be a genetic factor that makes certain individuals more susceptible
to developing GTFP.^^
14.6.6 D etection
Early detection of GTFP through the use of serological tests may help in defining
and monitoring the level of disease in sea turtles. Currently, monoclonal antibody
tests are being developed that may lead to a method for identifying turtles with GTFP
prior to development of visible lesions.
14.6.7 T reatment
There is no effective treatment for turtles suffering from GTFP. Once the étiologie
agent is isolated and purified, specific and effective treatments as well as preventive
measures may be developed. Currently, treatment is aimed at reducing the tumor
burden which affects the ability of the turtle to function normally. Tumors may be
surgically extirpated from the eyes and mouth to allow the animal to see and feed.
Also, fibropapillomas have been removed using cryosurgical techniques instead of
excisional surgery. Cryosurgery may stimulate the immune system of the turtle and
cause regression of tumors besides those frozen.
14.7 PARASITIC CONDITIONS

A variety of parasitic infections,both external and internal, have been documented
in sea turtles. Some parasites, such as leeches, may occur in large numbers and
cause extensive damage. Over a period of time, such severe infestations can be very
debilitating. In contrast, other parasites such as nematodes have little pathological
impact on the host animal.
14.7.1 Ectoparasites
14.7.1.1 Leeches
Leeches can occur in small or large numbers. They are most often located on the
soft tissue surfaces of the axilla and inguinal areas, but can attach around the eyes,
mouth, and cloaca. Aggregations of yellow egg cases may be found on the carapace,
plastron, and flippers.
Two significant species of leeches are known to parasitize sea turtles. Ozobran-
chus branchiatus is found only on green turtles inhabiting tropical waters. O. margoi
occurs on most species of sea turtles, is found on turtles inhabiting cooler waters,
and has a more worldwide distribution than O. branchiatus.
Leeches use their oral sucker to obtain a blood meal from the host. Animals
carrying large numbers of leeches become anemic and have extensive areas of
macerated dermal tissue. In addition to their debilitating effects, leeches can also
act as vectors for various disease-producing organisms. There may be a relationship
between turtles with large leech infestations and the occurrence of the GTFP.
Since leeches can complete their life cyle on the host turtle, captive animals can
rapidly develop severe infestations. Currently, the most effective way to remove
leeches from an infested turtle is a 90-min fresh-water bath.^^
14.7.1.2 Barnacles
Several varieties of barnacle are found on sea turtles,and large numbers of these
epizoans can contribute to the overall stress on a sea turtle. Some varieties only
negatively affect the animal by increasing surface drag, while other varieties can
damage the underlying shell and skin and allow bacterial or fungal pathogens to enter.
14.7.2 Endoparasites
A variety of endoparasites can occur in sea turtles. While the adult organisms may
have a deleterious effect on host animals, the majority of damage is done by immature
stages.
14.7.2.1 Protozoans
Pathological conditions due to amebic and coccidian parasites has been described
in sea turtles. Entameba invadens, which is ubiquitous in freshwater turtles and
snakes, has been implicated in the death of both green and loggerhead turtles.^^ E,
invadens cannot live in salt water and has only been reported in captive turtles. The
amebae cause an enterohepatitis after being introduced to a turtle, most likely in
contaminated food. Animals with amebic enterohepatitis have high white blood cell
counts, high liver enzyme levels (SGOT/AST), and amebae in stool samples. If
detected early, the infection may be treated with metronidazole.
Outbreaks of coccidiosis among wild"^° and captive turtles have been ascribed
to Caryospora cheloniae. In 1991, an epizootic infection occurred among large
subadult green turtles in Moreton Bay, Australia. On necropsy, the turtles exhibited
severe ulcerative enteritis. Microscopic examination of intestinal scrapings revealed
large numbers of enlongated C. cheloniae oocysts. Both the intestinal lesions and
oocysts were similar to those occurring in a group of sick captive green turtles
described by Leibovitz et al."^^ Treatment with antibiotics and sulfonamides was of
no benefit to either wild or captive animals. Eimeria caretta has been described in
two loggerhead turtles.T here was no sign of intestinal disease in these animals,
but oocysts were found in routine stool samples.
14.7.2.2 Helminths
Sea turtles play host to a variety of helminths. While many genera of tremátodos
and nematodes are recorded in the literature, no cestodes are reported.
Sea turtles are hosts to a large variety of tremátodos. With only one exception,
these tremátodos are digenean. Sea turtles acquire such flukes by ingesting unknown
cercariae-rich intermediate hosts. The majority of adult flukes reside in the GI tract
and cause minor irritation and damage. Heavy infestations can produce clinical
disease and severely debilitate the host.
Spirorchid tremátodos are the most damaging parasites with which sea turtles
contend. Adult spirorchid flukes reside in the vascular system, whereby their eggs
are disseminated throughout the body. Spirorchidiasis is a chronic debilitating dis
ease of older juvenile loggerheads, and affects up to 30% of the Atlantic loggerhead
population."^^ Spirorchid flukes also occur in green turtles and hawksbill turtles, and
are found worldwide in both wild and captive individuals."^'^'^^
Adult flukes reside in the heart and major blood vessels, causing an epithelial
thickening of the vessel walls and vasculitis. The eggs become disseminated through
out the vascular system and result in a vasculitis and granulomatous reaction wher
ever they lodge. Common sites for lesions to occur are the brain, lungs, liver, kidney,
and intestines. Intestinal lesions are raised, linear, and discolored. These greenish-
black areas occur primarily in the hindgut, and when incised release a yellow-brown
gritty material made up of masses of eggs. Microscopically the lesions contain
multinucleated giant cells and exhibit perivascular lymphocytic cuffing.
The four most prominent genera of spirorchids inhabiting sea turtles are Hap-
lotrema, Learedius, Carettacola, and Neospirorchis. Their eggs are elongated and
often have hooked terminal processes. The negatively bouyant eggs may be found
in feces of infected turtles. These blood flukes may serve as disease vectors, and
their role in such diseases as GTFP is being investigated. To aid such investigations.
as well as to study the epidemiology of blood fluke infections, ELISA tests have
been developed"^^ to detect antibodies to spirorchid flukes.
There are several nematodes known to inhabit the intestinal tracts of sea turtles.
The majority of these wonns are ascarid-like and reside in the pyloroduodenal area
of the GI tract. Nematodes such as Sulcascaris sulcata^^ are cosmopolitan in distri
bution and can be found in several species of sea turtle. Nematodes cause a mild
hemorrhagic enteritis and, unless they occur in large numbers, are of little clinical
significance. The negatively bouyant eggs of S. sulcata and other nematodes are
passed in the feces of the turtle. The eggs may then be ingested by molluscs which
serve as intermediate hosts.
14.8 E N V IR O N M E N T A L H E A LT H PR O B LEM S
A variety of health problems occur in sea turtles which may be directly or indirectly
brought about by local environmental factors. Some of these detrimental factors
occur naturally, such as rapid water temperature changes. Other factors involve
human activities, such as littering the oceans with anthropogenic materials.
14.8.1 N aturally O ccurring Problems

14.8.1.1 Traumatic Injuries
Turtles often suffer bite wounds from predators such as birds, fish, and sharks.
Wounds can vary from minor cuts and abrasions, to the loss of an entire flipper or
other body part. Females are occasionally found with bite wounds inflicted by an
aggressive male during breeding, and also can sustain traumatic injuries while on
nesting beaches.
14.8.1.2 Hypothermic Stunning
Hypothermic or cold stunning occurs when a turtle is exposed to cold water for a
period of time. As the core body temperature of the turtle drops, it becomes unable
to function normally. When water temperatures drop to 10°C, turtles bob to the
surface and float ineffectually. If temperatures drop to 5 to 6 °C death may occur.
Cold-stunning events have occurred from the Gulf of Mexico to New England, as
well as in Western Europe. In-depth reviews of such cold-stunning events have been
reported by a number of authors.'^^ '^^
Inshore populations of turtles are most susceptible to cold stunning because the
water temperature can change rapidly in shallow waters. Turtles inhabiting shallow
lagoon systems are particularly at risk when the weather cools rapidly. Not only will
a lagoon system sustain a rapid temperature drop, but turtles may be unable to easily
escape the lagoon and flee the cooling waters. Smaller animals cold stun sooner
than larger ones. Recorded cold-stun events have included loggerheads, greens, and
Kemp’s ridleys.
The main goal in treating a hypothermic turtle is to gradually raise the core body
temperature of the animal by immersion in warm water. A variety of medications
have been suggested to enhance body functions and aggressively treat severely
stunned animals.
Salt glands normally provide a way for sea turtles to remove excess sodium,
potassium, chlorine, calcium, magnesium, and phosphorous from their systems.
Cold-stunned animals often have elevated values for these ions.^"^^’^^ Turtles that
have survived cold stunning often have decreased salt gland function for a short
period of time. Such animals may be maintained in fresh water until salt gland
function returns and plasma electrolytes approach normal levels.
14.8.2 A nthropogenic Problems

14.8.2.1 Entanglement
Sea turtles often become entangled in a variety of fishing gear, lines, cables, or
refuse. In the broadest sense, capture in a shrimp or flounder trawl can be considered
a form of entanglement. Either of two basic types of injuries may occur when a sea
turtle becomes trapped in a net or other man-made object.
First, a local injury may occur where the turtle is caught in the fishing gear. An
example would be a linear necrotic band around a flipper that is wrapped in a crabpot
line. The injury may be superficial or deep, and may even require the amputation
of the flipper.
The second type of injury occurs when a turtle is caught in a gill or trawl net
and is kept submerged for a prolonged period of time. It may or may not drown, as
some are able to hold their breath for lengthy periods. When rescued, some trapped
turtles may be in an anoxic state and comatose.^^ The casual observer may not notice
a very occasional heartbeat or even a shallow breath. An areflexic turtle removed
from a trawl net should be carefully checked for any signs of life. Even if no life
signs are noted, the animal should be placed in the shade and monitored for several
hours. With proper supportive care, some of these animals may be revived. The brain
has biochemical adaptations that allow sea turtles to withstand lengthy periods of
anoxia^^ during prolonged dives or hibernation. This normal response to anoxia may
therefore provide some level of protection to forcible submergence.
14.8.2.2 Ingestion of Anthropogenic Debris
Large numbers of stranded sea turtles are found to have man-made debris in their
digestive tracts. Turtles tend to ingest debris frequently, but in small amounts. Debris
of any size may be found throughout the GI tract. While all species of turtles have
been found to ingest foreign matter, Kemp’s ridleys seem to be less likely to do so
than other species.^^
Intestinal foreign bodies can harm the health of the turtle both directly and
indirectly. Plastics and other materials may block the intestinal tract, or cause local
necrosis and ulceration. Indirectly, ingested waste material may degrade the condi
tion of the turtle by interfering with lipid metabolism,increasing intestinal transit
time, or contributing to the accumulation of intestinal gas and uncontrollable floating.
Tar and crude oil may be ingested and act as a gastric foreign body, interfere
with intestinal function, and possibly cause serious toxicity problems (discussed
further in the following section). Petroleum products can also cause external fouling
and act as a fonn of entanglement. According to stranding records, green turtles and
hawksbills are more apt to be impacted by tar and oil than other species.
14.8.2.3 Chemical Pollutants

Turtles of all life stages have been tested for the presence of a variety of chemical
pollutants. Analysis of Atlantic loggerhead and green turtle eggs and subadults for
organochlorines^^“^^ indicated the presence of low levels of PCBs and DDE, a
metabolite of DDT. These levels were lower than values obtained for other coastal
reptiles. An adult male leatherback^^ (Dermochelys coriaced) that stranded on the
Welsh coast was found to have appreciable PCB levels, but these levels were orders
of magnitude lower than PCB levels in birds and marine mammals from the same
geographic area.
Not all sea turtle species accumulate tissue residues at the same rate. Logger-
heads consistently have higher levels of both PCBs and DDE than green turtles, and
it has been hypothesized that the variation is due to dietary differences. No corre
lation was found between organochlorine, PCB, organophosphate, or carbamate
levels and the presence of GTFP disease.
Exposure to tar and oil can cause a host of changes and problems in sea turtles.^®
In addition to decreasing mobility due to fouling, exposure of skin to crude oil
may cause necrosis and sloughing of tissue. Crude-oil exposure also affects red
and white blood cell counts, glucose metabolism, and salt gland function. Indeed,
the salt glands may cease to function for up to two weeks after oil exposure. Turtles
partition hydrocarbons from exogenous tar and oil in their tissues. The oil compo
nents are processed and either ejected from the body or sequestered in various
tissue compartments. Some of the petroleum-derived alkanes taken up by tissues
are selectively removed, while others are more resistant and remain in the body
for prolonged periods.^' Little is known about the cumulative effects of these
compounds.
14.9 BIOMEDICAL TECHNIQUES

A variety of procedures or tests, such as analysis of plasma chemistries, may be
employed to assess the health or physiological state of a sea turtle. Additional
techniques may be used to repair damaged shells, heal soft tissue injuries, resuscitate
comatose animals, or to help solve other health problems.
14.9.1 H ematology and Plasma

The first step in health assessment of sea turtles through examination of blood
components is acquiring the blood sample. Blood samples can be taken from the
dorsal cervical sin u s.T h e sinuses run lateral to and parallel to the muscular ridges
in the dorsal neck region, and become larger as they approach the carapace. This
location can be used for venapuncture on all species and sizes, from 35-g hatchlings
to 900-kg adults.
Proper handling of samples is imperative for obtaining meaningful results. When

possible, hematology slides should be made and fixed soon after the blood is drawn.
Plasma samples are preferred to serum samples because turtle blood clots incom
pletely. Due to this inconsistent clotting, there are often fibrin strands in serum
samples which may clog blood analyzer intake ports and result in erroneous values.
Blood should be stored in lithium or sodium heparin tubes and be immediately
refrigerated. The use of tubes with serum or plasma separators is helpful in isolating
the plasma from the cells. This is important because if the blood is left whole for a
period of hours, glucose and potassium values^"^ may be altered. Chemistry values
may vary with the type of autoanalyzer used or the individual test methodology.^^
This variability may make it difficult to compare values and results from different
laboratories and research programs.
Results of any plasma chemistry evaluation should be inteipreted only by com
paring them to normal values established by the same laboratory employing the
same techniques. Also, normal plasma chemistry values (Table 14.1) for any given
enzyme or element such as calcium may vary from species to species. Within species,
normal values may vary with the geographic location of the sampled animal, the
age of the individual, or its activity.
Research reports on the hematology of sea turtles^'’’^ând other reptiles^^ identify
a variety of cell types. These include red blood cells, thrombocytes, eosinophils,
T mphocytes (large and small varieties), heterophils, neutrophils, and monocytes.
Not all of the cell types have been documented to occur in each species of turtle.
For example, neutrophils are common in loggerhead blood, but have not been
recorded in samples from Kemp’s ridleys. White blood cell counts can be performed
using a variety of dilution and staining protocols in conjunction with a hemocytom-
eter counting chamber.
14.9.2 A nesthesia
Sea turtles can be anesthetized with injectable or inhalation anesthetics. Ketamine
hydrochloride ( 1 0 ml at 1 0 0 mg/ml) mixed with acepromazine maleate ( 1 ml at 1 0
mg/ml) can be used as a stock solution (ket/ace) and administered intravenously
(IV) at a rate of 25 mg/kg of body weight. This dosage was developed for 10-kg
turtles and should be allometrically scaled^^ for use in larger turtles. Isoflurane alone
or in combination with nitrous oxide can be used to induce and maintain general
anesthesia. The turtles can be masked down with the inhalant anesthetics and then
intubated, or they can be administered ket/ace IV and intubated.
Depth of anesthesia can be monitored by assessing muscle tone and ocular blink
reflexes. In addition, if instrumentation is available, the heart rate can be followed
with an ultrasonic doppler or an electrocardiograph.
14.9.3 Resuscitation
Turtles that have become comatose from anoxic submergence, cold stunning, or an
adverse reaction to anesthesia can be successfully resuscitated.^^ Comatose turtles
TABLE 14.1
Hematology and Blood Chemistry Values for Healthy, Wild, Juvenile Sea
Turtles
Lepidocheiys
Species Caretta caretta^ kempii^^ Cheionia mydas^^
Location Chesapeake Bay New York Bight Bahamas
Sample Plasma Serum Plasma
Number tested n = 50 tab n = 60 n = 100
Parameter Mean SD Mean SD Mean SD
Packed cell volume (%) 29 5 31.1 13.5 35.2 3.2‘

Glucose (mg/dl) 100 18 115 42 114 15
Urea nitrogen (mg/dl) 92 13 14 21 14 16
Uric acid (mg/dl) — — _ _ — 1.5 0.6
Sodium (meq/1) 157 4 153 21 172 5
Potassium (meq/1) 3.6 0.5 3.6 1.3 5.3 0.6
Chloride (meq/1) 112 17 115 14 113 5
Calcium (mg/dl) 7.7 1.3 7.4 0.1 9.1 2.1
Phosphorus (mg/dl) 5.9 1.3 6.8 1.4 6.7 1.2
AST2((U/1) 285 120 145 42 178 50
Aik phos^ (U/l) 53 25 89 57 43 16
Total protein (g/dl) 3 1.1 3.1 0.5 5.1 0.8
Albumin (g/dl) 1.3 1.1 1.3 0.2 1.5 0.2
LDH“' (U/l) 310 484 1299 638 135 61
Creatnine kinase (U/l) 1680 2043 4460 3074 — —
* N = 106.
2 Aspartate aminotransferase.
^ Alkaline phosphatase.
Lactic dehydrogenase.
maintained at their optimum temperature (25 to 30°C) will revive faster than turtles
with lower body temperatures.
The first step should be to establish an open airway. After placing a bite block
into the animal’s mouth, a cuffed endotracheal tube should be inserted through the
larynx and into the trachea. The lungs can then be inflated by using a positive
pressure resuscitation bag. Care should be taken not to overinflate the lungs. Rather,
they should be gradually inflated until the carapace starts to rise. The turtle should
be ventilated several times per minute until the animal shows signs of reviving.
As the turtle starts to revive its heart rate will increase and it may move its front
flippers. Shortly after voluntary movement begins, the turtle will generally breathe
on its own. at which time the endotracheal tube should be removed.
REFERENCES
1. Morris, Y. A. and Owens, D. W., Corticosterone and stress in sea turtles. Am. ZooL,
22(4), 956, 1982.
2. Schwantes, N. L., Aspects of Circulating Corticosterone in Sea Turtles, MS thesis,
Texas A&M University, College Station, 1986.
3. Zapata, A. G., Varas, A., and Torroba, M., Seasonal variations in the immune system
of lower vertebrates, Immunol. Today, 13, 142, 1992.
4. Wood, J. R. and Wood, F. E., Captive rearing and breeding Kemp’s ridley sea turtles
at Cayman Turtle Farm (1983) Ltd., in Proc. 1st Int. Symp. on Kemps Ridley Sea
Turtle Biology, Conservation, and Management, Caillouet, C. W., Jr. and Landry,
A. M., Jr., Eds., Texas A&M University, Sea Grant College, Galveston, TAMU-SG-
89A05, 1989, 237.
5. Leong, J. K., Smith, D. L., Revera, D. B., Clary, J. C., Lewis, D. FI., Scott, J. L., and
DiNuzzo, A. R., Flealth care and diseases of captive-reared Loggerhead and Kemp’s
ridley sea turtles, in Proc. 1st Int. Symp. on Kemp’s Ridley Sea Turtle Biology,
Consetâtion, and Management, Caillouet, C. W. and Landry, A. M., Eds., Texas
A&M Sea Grant, Galveston, TAMU-SG-89-105, 1989, 178.
6 . Wood, J. R. and Wood, F. E., Growth and digestibility for the green turtle (Chelonia
mydas) fed diets containing varying protein \Qvt\s, Aquaculture, 25, 269, 1981.
7. Glazebrook, J. S. and Campbell, R. S., A survey of the diseases of marine turtles in
northern Australia. II. Oceanarium-reared and wild turtles, Dis. Aquat. Org., 9, 97,
1990.
8 . George, R. H., unpublished data.
9. Glazebrook, J. S. and Campbell, R. S., A survey of the diseases of marine turtles in
northern Australia. I. Farmed turtles, ZJA. Aquat. Org., 9, 83, 1990.
10. Fowler, M. E., Metabolic bone disease, in Zoo and Wild An. Med., 2nd ed., W. B.
Saunders, Philadelphia, 1986, 70.
11. Stamper, A. and Lewbart, G. A., Personal communication, 1995.
12. Bellmund, S. A., Assessing Environmental Stress on the Loggerhead Sea Turtle
(Caretta caretta) in Virginia Waters, MS thesis. School of Marine Science, College
of William and Mary, Williamsburg, VA, 1988.
13. Choromanski, J. M., George, R. H., and Bellmund, S. A., Nutritional benefit of a
marine animal gelatin diet as measured by sea turtle blood chemistry values, in Proc.
Am. Assoc, of Zoos, Parks, and Aquariums, Portland, Oregon, 1987, 501.
14. Glazebrook, J. S., Campbell, R. S., and Thomas, A. T, Studies on an ulcerative
stomatitis-obstructive rhinitis-pneumonia disease complex in hatchling and juvenile
sea turtles, Chelonia mydas and Caretta caretta, Dis. Aquat. Org., 16, 133, 1993.
15. Kinne, O., Diseases o f Marine Animals, Vol. 4, Part 2, Biologische Anstalt Helgoland,
Hamburg, 1985, 543.
16. Wiles, M. and Rand, T. G., Integumental ulcerative disease in a loggerhead turtle
Caretta caretta at the Bermuda Aquarium: microbiology and histopathology, Dis.
Aquat. Org., 3, 85, 1987.
17. Frye, F L., Biomedical and Surgical Aspects o f Captive Reptile Husbandry,Veitrimry
Medicine Publishing Co., Edwardsville, KS, 1981, 146.
18. George, R. H., Wolke, R. E., and Keinath, J. A., A new reason why apparently healthy
sea turtles strand in warm weather, in Piôc. 12th Annu. Workshop on Sea Turtle
Conservation and Biology, Richardson, J. I. and Richardson, T. H., Compilers,
N.M.F.S. Tech. Memo., NOAA-TM-NMFS-SEFSC-361, Miami, FL, 41, 1995.
19. Walsh. M.. Personal communication, 1995.
20. Homer, B. L., Jacobson, E. R., Schumacher, J., and Scherba, G., Chlamydiosis in
mariculture-reared green sea turtles (Chelonia mydas), Vet. Pathol., 3 1 ,1 , 1994.
21. Jacobson, E. R., Gaskin, J. M., Shields, R. R, and White, R. H., Mycotic pneu
monia in mariculture-reared green sea turtles,/. Am. Vet. Med. Assoc., 175, 929,
1979.
22. Lewbart, G. A. and Medway, W., A case of mycotic lung disease in a wild caught
juvenile sea turtle, J. Small Exotic Anim. Med., 2(2), 58, 1993.
23. Jacobson, E. R., Gaskin, J. M., Roelke, M., Greiner, E. C., and Allen, J., Conjunc
tivitis, tracheitis, and pneumonia associated with herpesvirus infection in green sea
turtles, J. Am. Vet. Med. Assoc., 189(9), 1020, 1986.
24. Haines, H., A herpesvirus disease of green sea turtles in aquaculture,Mar. Fish. Rev.,
1294, 33, 1978.
25. Herbst, L. H., Fibropapillomatosis of marine turtles, Annu. Rev. Fish Dis., 4, 389,
1994.
26. Smith, G. M. and Coates, C. W., Fibro-epithelial growths of the skin in large marine
turtles Chelonia mydas, Zoologica, 23, 93, 1938.
27. Balazs, G. H., Current status of fibropapillomas in the Hawaiian green inxilQ,Che Ionia
mydas, in Research Plan for Marine Turtle Fibropapilloma, Balazs, G. H. and Pooley,
S. G., Eds., N.M.F.S. Tech. Memo. NOAA-TM-NMFS-SWFC-156, Honolulu, HI,
47, 1991.
28. Limpus, C. J. and Miller, J. D., The occurence of cutaneous fibropapillomas in marine
turtles in Queensland, in Proc. Australian Mar. Turtle Conservation Workshop, Rus
sell James, Compiler, Queensland Department of Environment and Heritage and
Australian Nature Conservation Agency, Brisbane, 1990, 186.
Queensland; population structure in a warm temperate feeding area, Mem. Queensl.
Mus., 35(1), 139, 1994.
30. Ehrhart, L. M., Fibropapillomas in green turtles of the Indian River lagoon, Florida:
distribution over time and area, 'm Research Plan fo r Marine Turtle Fibropapilloma,
Balazs, G. H. and Pooley, S. G., Eds., N.M.F.S. Tech. Memo. NOAA-TM-NMFS-
SWFC-156, Honolulu, HI, 59, 1991.
31. Harshbarger, J. C., Sea turtle fibropapilloma cases in the registry of tumors in lower
animals, in Research Plan fo r Marine Turtle Fibropapilloma, Balazs, G. H. and
Pooley, S. G., Eds., N.M.F.S. Tech. Memo. NOAA-TM-NMFS-SWFC-156, Hono
lulu, HI, 63, 1991.
32. Schlumberger, H. G. and Lucke, B., Tumors of fishes, amphibians, and reptiles,
Cancer Res., 8, 657, 1948.
33. Jacobson, E. R., Mansell, J. L., Sundberg, J. P, Hajjar, L., Reichmann, M. E., Ehrhart,
L. M., Walsh, M., and Murru, F, Cutaneous fibropapillomas of green turtles (Chelonia
mydas), J. Comp. Pathol., 101, 40, 1989.
34. Papadi, G. P, Balazs, G. H., and Jacobson, E. R., Flow cytometric DNA content
analysis of fibropapillomas in green turtles Chelonia mydas, Dis. Aquat. Org., 22,
13, 1995.
35. Herbst, L. H., Jacobson, E. R., Moretti, R., Brown, T., Sundberg, J. P, and Brown,
P. A., Experimental transmission of green turtle fibropapillomatosis using cell-free
tumor extracts, Dis. Aquatic Org., 22, 1, 1995.
36. Jacobson, E. R., Buergelt, C., Williams, B., and Harris, R. K., Herpesvirus in cuta
neous fibropapillomas of the green inxilt Chelonia mydas, Dis. Aquat. Org., 12(1-6),
2. 1991.
37. Morris, R. A. and Balazs, G. H., Experimental use of cryosurgery to treat fibropap-
illomas in the green turtle, Chelonia my das, in Proc. 13 th Annu. Workshop on Sea
Turtle Conservation and Biology, Schroeder, B. A. and Whitherington, B. E., Com
pilers., N.M.F.S. Tech. Memo. NOAA-TM-NMFS-SEFSC-341, Miami, FL, 111,
1994
38. Choy, B. K., Balazs, G. H., and Daily, M., A new therapy for marine turtles parasitized
by the piscicolid leech, Ozobranchus branchiatus, Herpetol. Rev., 20(4), 89, 1989.
39. George, R. H., Jones, W., and Musick, J. A., Enterohepatitis due to Entamoeba
invadans in captive Caretta caretta, in Proc. 10th Annu. Workshop on Sea Turtle
Consef-vation and Biology, Richardson, T. H., Richardson, J. I., and Donnelly, M.,
Compilers, N.M.F.S. Tech. Memo. NOAA-TM-NMFS-SEFSC-278, Miami, FL, 93,
1990
40. Gordon, A. N., Kelly, W. R., and Lester, J. G., Epizootic mortality of free-living green
turtles, Chelonia mydas, due to coccidiosis, /, Wildlife Dis., 29(3), 490, 1993.
41. Leibovitz, L., Rebell, G., and Boucher, G. C., Caryospora cheloniae SR N.: A
coccidial pathogen of mariculture-reared green sea turtles (Chelonia mydas), J. Wildl.
Dis., 14, 269, 1978.
42. Upton, S. J., Odell, D. K., and Walsh, M. T.,Eimeria caretta sp. nov. (Apicomplesa:
Eimeriidae) from the loggerhead sea turtle, Caretta caretta (Testudines), Can. J.
ZooL, 68, 1268, 1990.
43. Wolke, R. E., Brooks, D. R,, and George, A., Spirorchidiasis in loggerhead sea turtles
(Caretta caretta), J. Wildl. Dis., 18(2), 175, 1982.
44. Greiner, E. C., Forrester, D. J., and Jacobson, E. R., Helminths of mariculture-reared
green turtles (Chelonia mydas) from Grand Cayman, British West Indies, in Proc.
Helminthol. Soc. Wash., 47(1), 142, 1980.
45. Glazebrook, J. S., Campbell, R. S., and Blair, D., Studies on cardiovascular fluke
(Digenea: Spirorchiidae) infections in sea turtles from the Great Barrier Reef, Queen
sland, Australia,/. Comp. Pathol., 101, 231, 1989.
46. Graczyk, T. K., Aguirre, A. A., and Balazs, G. H., Detection by elisa of circulating
anti-blood fluke (Carettacola, Hapalotrema, Learedius) immunoglobulins in Hawai
ian green turtles (Chelonia mydas), J. Parasitol., 81(3), 416, 1995.
47. Witherington, B. E. and Ehrhart, L. M., Hypothermic stunning and mortality of marine
turtles in the Indian River lagoon system, Florida, Copeia, 3, 696, 1989.
48. Morreale, S. J., Meylan, A. B., Sadove, S. S., and Standora, E. E., Annual occurrence
and winter mortality of marine turtles in New York waters,/. Herpetol., 26, 301, 1993.
49. Nicolson, S. W. and Lutz, P. L., Salt gland function in the green sea imtlQ Chelonia
mydas, J. Exp. Biol.,, 144, 171, 1989.
50. Carminati, C. E., Gerle, E., Kiehn, L. L., and Pisciotta, R. P., Blood chemistry
camparison of healthy vs. hypothemiic juvenile Kemp’s ridley sea turtles (fepido-
chelys kempi), in Proc. 14th Annu. Workshop on Sea Turtle Conservation and Biology,
BJorndal, K. A., Bolton, A. B., and Johnson, D. A., Compilers, N.M.F.S. Tech. Memo.
NOAA-TM-NMFS-SEFSC-351, Miami, FL, 203, 1994.
51. Shoop, R. C., Ruckdeschel, C. A., and Wolke, R. E., The myth of the drowned turtle,
in Proc. lOh Annu. Workshop on Sea Turtle Conservation and Biology, Richardson,
T. H., Richardson, J. I., and Donnelly, M., Compilers, N.M.F.S. Tech Mem., NOAA-
TM-NMFS-SEFSC-278, Miami, FL, 93, 1990.
52. Lutz, P. L., LaManna, J. C., Adams, M. R., and Rosenthal, M., Cerebral resistance
to anoxia in the marine turtle, Respir. Physiol., 41, 241, 1980.
53. BJorndal, K. A., Bolten, A. B., and Lagueux, C. J., Ingestion of marine debris by
juvenile sea turtles in coastal Florida habitats. Mar. Pollut. Bull., 28(3), 154, 1994.
54. Schulman, A. A. and Lutz, P. L., The effect of plastic ingestion on lipid metabolism
in the green sea turtle {Chelonia mydas), in Proc. 12th Annu. Workshop on Sea Turtle
Conservation and Biology, Richardson J. I. and Richardson, T. H., Compilers,
N.M.ES. Tech. Memo. NOAA-TM-NMFS-SEFSC-361, Miami, FL, 122, 1995.
55. Clark, D. R. and Krynitsky, A. J., Organochlorine residues in eggs of loggerhead and
green sea turtles nesting at Merritt Island, Florida — July and August l916,Pestic.
Monit. /., 14(1), 7, 1980.
56. McKim, J. M. and Johnson, K. L., Polychlorinated biphenyls and p,p'-DDE in
loggerhead and green postyearling Atlantic sea turtles. Bull. Environ. Contam. Tox
icol., 31, 53, 1983.
57. Hillestad, H. O., Reinold, R. J., Stickney, R. R., Windom, H. L., and Jenkins, J. H.,
Pesticides, heavy metals and radionuclide uptake in loggerhead sea turles from South
Carolina and Georgia, Herpetol. Rev., 5(3), 75, 1974.
58. Davenport, J., Wrench, J., McEvoy, J., and Camacho-Ibar, V., Metal and PCB con
centrations in the “Harlech” leatherback, Mar. Turtle Newsl., 48, 1, 1990.
59. Aguirre, A. A., Balazs, G. H., Zimmennan, B., and Galey, R. D., Organic contami
nants and trace metals in the tissues of green turtles {Chelonia mydas) afflicted with
fibaropapillomas in the Hawaiian Islands, Mar. Pollut. Bull., 28(2), 109, 1994.
60. Lutcavage, M. E., Lutz, P. L., Bossart, G. D., and Hudson, D. M., Physiologic and
clinicopathologic effects of crude oil on loggerhead sea turtles, Arc/z. Environ. Con
tam. Toxicol., 28, 417, 1995.
tissues of sea turtles exposed to the Ixtoc I oil spill,/. Wildl. Dis., 19(2), 106, 1982.
62. Owens, D. W. and Ruiz, G. J., New methods of obtaining blood and cerebrospinal
fluid from marine turtles, Herpetologica, 36, 17, 1980.
63. Jacobson, E. R., Laboratory investigations, in Manual o f Reptiles, Beynon, P. H., Ed.,
British Small Animal Veterinary Assoc., Cheltenham, 1992, 49.
64. Bolten, A. B., Jacobson, E. R., and Bjorndal, K. A., Effects of anticoagulant and
autoanalyzer on blood biochemical values of loggerhead sea turtles ifdaretta caretta).
Am. J. Vet. Res., 12, 2224, 1992.
65. Wood, F. E. and Ebanks, G. K., Blood cytology and hematology of the green sea
turtle, Chelonia mydas, Herpetologica, 40(3), 331, 1984.
66. Cannon, M. S., The morphology and cytochemistry of the blood leukocytes of Kemp’s
ridley sea turtle {Lepidochelys kernpi). Can. J. ZooL, 70, 1336, 1992.
67. Frye, F. L., Hematology as applied to clinical reptile medicine,Reptile Care, an Atlas
o f Diseases and Treatments, Vol. 1, Frye, F. L., Ed., T.F.H. Publications, Neptune
City, NJ, 1991, chap. 7.
68. Pokras, M. A., Therapeutics, Manual o f Reptiles, Beynon, P. H., Ed., British, Small
Animal Veterinary Assoc., Cheltenham, 1992, 194.
69. Balazs, G. H., Resuscitation of a comatose green turtle, Herpetol. Rev., 17(4), 79,
1986.
70. Bolten, A. B. and Bjonidal, K., Blood profiles for a wild population of green turtles
{Chelonia mydas) in the southern Bahamas: size-specific and sex-specific relation
ships, J. Wildl. Dis., 28(3), 407, 1992.
'j C Human Impacts on Sea
Turtle Survival
Molly E. Lutcavage, Pamela Plotkin, Blair
Witherington, and Peter L. Lutz
CO NTENTS
15.1 Introduction................................................................................................. 388

15.2 Habitat Alteration and Loss........................................................................388
15.2.1 Nesting Beach Concerns...............................................................388
15.2.1.1 Beach Armoring...........................................................389
15.2.1.2 Beach Nourishment and Sand Mining........................ 389
15.2.1.3 Beach Cleaning and Beach Driving............................390
15.2.1.4 Human Presence on the Beach....................................390
15.2.1.5 Artificial Lighting........................................................ 391
15.2.2 Boat Strikes................................................................................... 392
15.2.3 Dredging, Explosive Platform Removal, andOther Activities....392
15.2.4 Depradation of Sea Turtle Nests..................................................392
15.3 Oil Pollution................................................................................................ 392
15.3.1 Oil Pollution Concerns.................................................................392
15.3.2 Laboratory Studies........................................................................ 394
15.3.3 Oil Spill Response Activities........................................................ 395
15.4 Other Pollution Sources andEntanglement................................................ 395
15.4.1 Plastics and Nonbiodegradable Debris.........................................396
15.4.2 Ingestion.........................................................................................396
15.4.3 Physiological Effects of Debris Ingestion....................................397
15.4.4 Entanglement................................................................................ 398
15.4.5 Types of Debris and Sourcesof Origin......................................... 398
15.5 Fishing and Incidental Capture.................................................................. 398
15.5.1 Shrimp Trawling and Turtle Excluder Devices............................399
15.5.2 Pelagic Fishing Gear.....................................................................400
15.5.3 Coastal Fishing Activities.............................................................401
15.5.4 Artisanal and Commercial Sea Turtle Fisheries..........................402
15.6 Conclusions........................................................................ 403
References............................................................................................................. 404
0-8493-8422-2/97/$0 00+$.50
As detailed in the introductory chapter by Pritchard, sea turtles have a long and
ancient history. Sea turtles arose in the early Mesozoic, and for the next 100 million
years, during the rise and reign of the dinosaurs, they shared the oceans with a rich
diversity of other air-breathing marine reptiles. By the end of the Cretaceous era,
the sea-going ichthyosaurs and plesiosaurs were extinct, but sea turtles tenaciously
survived and indeed flourished up until very recent times. But today their numbers
are drastically reduced to the point that all seven remaining sea turtle species are
considered either threatened or endangered on a worldwide basis. Undoubtedly,
human interference is the cause of this collapse. The challenges that sea turtles now
face from human activities impact every stage of their life cycle, from loss of nesting
beach and foraging habitats to mortalities on the high seas through intense pelagic
fishing practices. They are also harmed by the increasing loads of nonbiodegradable
waste and pollutants that the oceans and coastal zones now receive.
Clearly, a full consideration of all human activities that adversely impact sea
turtles and required conservation measures would be beyond the scope of a single
chapter. However, a review of the more compelling effects will give an indication
of the extent of the challenges sea turtles face and the efforts required to protect
them. A word of caution: by its very nature much of the data on this subject is only
found in the “grey literature” of government and conservation organization reports
and such like, and is therefore scientifically unreviewed. This is sometimes the case
in conservation biology, where threats to survival are first recognized and conser
vation remedies proposed by biologists familiar with the status of local populations.
15.2 H A B ITA T A LT E R A T IO N A N D LO SS
15.2.1 N esting B each C oncerns
The translocation of nests into hatcheries is probably the most common sea turtle
conservation technique undertaken when beach operations threaten sea turtles. These
efforts are usually practiced where serious threats to embryos or hatchlings are
thought to exist. High levels of egg poaching, predation, erosion, artificial lighting,
beach nourishment projects, and vehicular and foot traffic have been cited as justi
fication for nest translocation. However, there are potential problems inherent in
moving eggs and incubating them under artificial conditions, including increased
egg mortality and reductions in genetic and realized fitness.
Translocated sea turtle nests can have lower hatching rates than undisturbed
natural nests,although under certain conditions translocated nests may do better.^
Additional problems with translocation arise from changes in the incubation envi
ronment. Differences in temperature, hydric environment, and gas exchange affect
hatching rates and possibly the vigor and survivorship of hatchlings."^ During the
initial years of the Kemp’s ridley conservation program, low egg incubation tem
peratures resulted in the production of hatchlings that were primarily males.^ This
example shows that merely counting the number of mostly male hatchlings produced
does not reveal the success of a translocation program. Important assessments for
Human Impacts on Sea Turtle Survival 389
conservation include both number of hatchlings and their fitness — i.e., likelihood
of surviving, returning to an appropriate beach, and leaving offspring that themselves
survive.
15.2.1.1 Beach Armoring
Beach armoring consists of hardened structures meant to protect dune property from
erosion. This includes vertical or inclined concrete sea walls, wooden walls, rock
revetments, and sandbag/sandtube structures.^ From the perspective of a nesting
turtle, armoring is detrimental if it is rigid enough to impede access to the nesting
habitat. Even “soft” structures such as sand fences may pose threats similar to harder,
more pemianent fortifications. Structures meant to control longshore sand movement
such as groins and jetties also may present similar barriers to nesting turtles.
The greatest threat that armoring poses for sea turtles is to eliminate nesting
habitat. An example of how this can occur can be found in the case of the barrier
island nesting beach at Jupiter Island, FL. This partially developed island suffers a
net erosion of sand largely due to the sand starvation effects of an inlet Jetty. Whereas
both developed and undeveloped stretches of Jupiter Island are losing sand from the
beach, only the developed stretches have been armored. On the natural beach, dune
sand has replaced eroded beach, but on armored stretches, waves wash up to the
foot of walls and revetments, leaving no dry beach for nesting.
The problems caused by beach armoring will be greatest where coastal devel
opment is most widespread. This is especially the case in the U.S. and Mediterranean
where coastal development is older and dune structures have had more time to
become threatened by erosion. Elsewhere, the installment of armoring and its effects
on sea turtle nesting will become an important conservation issue as sea level rises,
beaches erode, and buildings are defended from the sea.
Whether sea turtles kept from nesting at armored beaches may forgo nesting
altogether or move to adjacent suitable nesting beaches is unknown. Although a
concentration of nesting effort at suitable nesting sites would allow reproduction to
continue, an increase in density-dependent mortality of eggs and hatchlings would
ensue.
15.2.1.2 Beach Nourishment and Sand Mining

Beach nourishment involves mechanically dumping or pumping sand onto an eroded
beach. Because this is an expensive undertaking, these activities are primarily con
ducted on the beaches of developed countries (especially the U.S. and Europe) where
the economic value of the beach makes such an effort worthwhile.
Beach nourishment is a preferred alternative to armoring. However, beach nour
ishment replaces rather than maintains the original nesting habitat, and the suitability
of this renourished habitat for sea turtle nesting will depend on the quality of fill
material and the methods used to deposit it. Some nourished beaches have a high
clay, silt, and shell content that makes them too compact for nest excavation. Often
on nourished beaches, wave action causes steep escarpments to form that prevent
turtles from reaching the upper beach. Properties of an artificially nourished substrate
that differ from a natural beach include sorting, moisture content, reflection, and
conduction. These properties affect the architecture of the egg chamber, incubation
temperature, gas exchange, and water uptake for a developing clutch and result in
diminished egg and hatchling survivorship.
In Florida, exhaustion of local sand sources for beach renourishment has stim
ulated the use of aragonite sand from the Bahamas. Incubation temperatures of
clutches in the white aragonite sand are 1 to 2°C lower than for natural Florida
silicate sands, a factor which may have an important effect on the hatchling sex
ratios (see chapter by Ackerman).^
There are economic incentives to nourish beaches during the summer nesting
season when costs are lower due to calmer seas. Renourishment projects during the
nesting season are already taking place in Florida, with an environmental cost to
sea turtles. Although large-scale translocations of eggs to hatcheries precede these
projects, an estimated 8% of the nests are missed by field workers and probably
destroyed during beach nourishment activities. Translocation itself can cause egg
mortality, and the process of nourishment involving lighting, sand spreading, and
pumping activities could disturb nesting females, and produce disorientation and
mortality in hatchlings.
In sand mining operations, large quantities of sand are dug from beaches to be
used as fill, in production of concrete, and for other construction activities. Sand
mining destroys the beach, eliminates nesting, and is recognized as a threat to some
sea turtle populations, primarily in the Mediterranean.^-^^^
15.2.1.3 Beach Cleaning and Beach Driving

Beach cleaning and beach driving involve vehicular traffic and its effects. Heavy
vehicles can crush developing eggs and pre-emergent hatchlings, and the tire ruts
resulting from traffic can “entrap” hatchlings, allowing them to become exhausted
or taken by p red ato rs.L o g g erh ead hatchlings can escape from a 3-cm-deep
footprint, but are unable to escape a tire rut of similar depth. Hatchlings may crawl
along ruts due to their tendency for orientation toward open areas. Vehicles operated
at night can disturb nesting females and crush emerging hatchlings crawling toward
the sea.
The principal deleterious impact of beach cleaning on sea turtles is to uncover
and destroy nests. Raking can also leave ruts or ridges that disrupt hatchling sea
finding behavior. Beach cleaning can be regulated so that the effect on sea turtle
nests is low.
15.2.1.4 Human Presence on the Beach
Although the effects of human foot traffic on sea turtle nests are largely unstudied,
these perceived impacts are commonly used as a rationale for translocating nests.
As with vehicular traffic, foot traffic has the potential to damage buried eggs and
harm pre-emergent hatchlings. Human visitation at night can cause turtles to abort
nesting attempts, although where visitation is controlled within guided “turtle watch”
groups, this disturbance is m i n i m a l . “Turtle watch” outings have an important
conservation and education potential and are becoming a common forni of eco-
tourism. However, even well-intentioned eco-tourism programs can lead to unex
pected conflicts. For example, in Costa Rica, former subsistence turtle hunters and
“hueveros”, or egg collectors, have been trained as turtle guides, and now derive
income by leading nesting beach tours. There is some indication that the strong
economic incentive to attract a large number of tourists to some important nesting
beaches may be at odds with the need to minimize disturbance to nesting females
and emergent hatchlings. Fortunately, these problems can be addressed through
educational programs and goal setting by the conservation and residential commu
nities alike.
15.2.1.5 Artificial Lighting
Artificial lighting on nesting beaches is detrimental to sea turtles because it disrupts

critical behaviors, including nest-site choice and the nocturnal sea-finding behavior
of both hatchlings and nesting females. There is both associative and direct exper
imental evidence showing that artificial lighting on beaches deters sea turtles from
nesting.Reduced nesting on lighted beaches appears to be the result of fewer
attempts at emerging onto the beach rather than the abandoning of attempts after
turtles emerge onto the beach. Nesting that does occur on or near lighted beaches
can produce high hatchling mortality. Hatchlings move toward artificial light sources
rather than the sea, and succumb to exhaustion, dehydration, and predation. 14,19-22
Hatchling mortality from artificial lighting is difficult to detect and is generally
underestimated. Effects of lighting vary with the lunar cycle and are greatest during
the new moon period.^Âs a general rule, any artificial light source that is visible
from the nesting beach may disorient sea turtles.
Many options exist for managing light near nesting beaches. Light sources can
be minimized in number and wattage, shielded, redirected, lowered, recessed, or
repositioned behind shielding objects so that light from the source does not reach
the beach. Installation of timers and motion-detector switches will ensure that light
ing is on only when needed. Interior lighting can be reduced by moving lamps away
from windows, drawing blinds after dark, and tinting windows.
Light sources that emit comparatively low levels of short-wavelength (blue and
green) light affect both hatchlings and nesting adults less than sources emitting
higher levels of short-wavelength light. For this reason, low-pressure sodium
vapor luminaires (a monochromatic yellow source) make good substitutes for more
disruptive lighting.^4Yellow-tinted incandescent “bug light” bulbs are another appro
priate substitute where lighting problems have been only partially solved by other
means.
In the U.S., Costa Rica, Greece, and Australia, there are conservation programs
in place to educate beach visitors and residents about the harmful effect of lighting
on sea turtles. In Florida and South Carolina, local power companies have distributed
educational materials to coastal consumers. Presently, most Florida communities
near nesting beaches have light-management ordinances designed to protect sea
turtles.
15.2.2 B oat S trikes
Sea turtle stranding data for the U.S. Gulf of Mexico and Atlantic coasts, Puerto
Rico, and the U.S. Virgin Islands show that between 1986 and 1993, about 9% of
living and dead stranded sea turtles had propeller or other boat strike injuries (n =
16,102).^^’^^ In Florida where coastal boating is popular, frequency of boat injuries
between 1991 and 1993 was 18% of 2,156 strandings. Countless numbers of log
gerheads and 5 to 50 Kemp’s ridley turtles are estimated to be killed by vessel traffic
per year in the U.S.^^ Although some of these strikes may be post-mortem, the data
show that vessel traffic is an important cause of sea turtle mortality.
15.2.3 D redging , Explosive P latform R emoval, and O ther

A ctivities
Dredge and fill activities occur in many of the nearshore seasonal habitats of sea
turtles worldwide. Operations range in scope from incidental propeller dredging by
recreational boats to large-scale dredging of channels and fill for land reclamation.
Sea turtles have been killed by hopper dredging in Florida and Georgia, but the
extent of mortality in other geographic regions is unknown.^^
Sea turtles associate with offshore oil and gas platforms, and can be hamied by
underwater explosions detonated to remove platforms no longer in service.^^ These
explosions can cause capillary damage, disorientation, and loss of motor control in
sea turtles. Although sea turtles far from explosion sites may suffer only disorien
tation, those near detonation sites will most likely sustain fatal injuries.^® Estimates
of sea turtles killed from explosive platform removal in the Gulf of Mexico range
from 50 to 500 per year.^^ Petroleum seismographic canons and military maneuvers
involving explosives also have the potential to harm sea turtles, but specific infor
mation on their impacts is lacking.
15.2.4 D epradation of S ea T urtle N ests
Companion animals and pests attracted to trash and food remains left by humans
can have a devastating impact on the success rates of sea turtle nesting beaches.
Nesting depredation by feral hogs, pigs, foxes, coyotes, coatis, mongooses, and dogs
can destroy up to 100% of nests.Municipalities have taken steps to eliminate or
reduce this disturbance with predator removal programs and the use of tamper-proof
trash receptacles.
15.3 O IL P O L L U T IO N
15.3.1 O il P ollution C oncerns
Sea turtle populations around the world coexist with oil and gas exploration, trans
portation, and development. The risk of this association is demonstrated in the
growing record of turtles harmed by oil in the Gulf of Mexico,^ ^ the Atlantic,^^ the
Caribbean Sea,^^ the Gulf of Iraq,^^ the M editerranean,and the Red Sea.-^^’ Indeed,
the extent of oil and tar pollution in ocean surface waters is enormous. The Sargasso
Sea alone entraps an estimated 70,000 metric tons of tar.^^
Sea turtles are exposed to the hamiful effects of oil pollution in many ways
(Figure 15.1). By continuously resurfacing through an oil slick to breathe, adults
would experience prolonged physical contact with the floating oil. Ocean convergent
zones and drift lines, where pelagic juveniles are found, are also regions where
floating tar accumulates. The sea turtle’s mode of respiration, a rapid inspiration of
surface layer air before diving, would introduce petroleum vapor into the lungs.
Eating contaminated food or tar balls brings petroleum materials into the intestine.
On nesting beaches, oil deposits could interfere with normal development of the
embryos in the egg clutch as well as present a lethal hazard to newly emerged
hatchlings. It is clear that all sea turtle species and life stages are vulnerable to the
harmful effects of oil, through direct contact or by fouling of their habitats.
ROUTE RESULT
Carcinogenesis — DEATH
SKIN
Parasitism, Disease ► Viability — PREDATION
Decreased INCREASES
Âerobic scope
LUNG decreased Foraging time
"reduced
Dive time
decreased GROWTH
REDUCED
Âssimilation
GUT decreased
, Organ ---- PERFORMANCE
Internal effects - Dysfunction DEPRESSED
Hormone ----------- ^ REPRODUCTION
Balance disturbed DECREASED
SENSE - Interference BEHAVIOR
ORGANS PERTURBED
EGGS - Abnormal
Development DEATH
FIGURE 15.1 Conceptual model of potential effects of prolonged oil exposure on sea
turtles. The primary impacts on organ systems, drawn from documented effects in laboratory
studies, may work synergistically to produce behavioral changes, interference with sensory
systems, reduced fitness, and eventual death. (From Lutz et al.^^)
There is direct evidence that sea turtles have been seriously harmed by oil spills.
A recent fuel oil spill oiled loggerhead nests and jeopardized hatchlings and adult
females in Tampa Bay, FL. In 1979, the Mexican Ixtoc blowout oil breached Rancho
Nuevo, the only known major nesting beach of the Kemp’s ridley, and oiled sea
turtles were subsequently seen in Texas coastal wa tersThere is evidence that in
1983, 180 hawksbill turtles were killed by oil spilled off the islands of Jana and
Karan, near Saudi Arabia."^^ Although the documentation of the numbers of sea turtles
killed by oil spilled during the Gulf of Iraq conflict is scanty, estimates range from
tens to hundreds.'^^
On a yearly basis about 1% of strandings identified by the U.S. sea turtle
stranding network are associated with oil, with higher rates in South Florida (3%
of strandings) and Texas beaches (3 to 6.3% of s t r a n d i n g s ) . F r o m 1980 to 1984,
37.5% of Dade County, FL sea turtle strandings were petroleum related, probably
due to the heavy surface and beach tar occurrences reported in South Florida."^^ Rates
may also be high in transport routes through the Caribbean and other areas. Oil
removed from stranded sea turtles in Florida and Texas has been identified primarily
as tanker washings from diverse sources."^^ Pelagic size-class loggerheads, hawks-
bills, and green turtles are known to strand, having been mired in oil.^^
In the western Atlantic and Gulf of Mexico offshore, sea turtle distribution has
been linked to zones of convergence.^^-^'^ Convergence fronts and Langmuir cells
concentrate zooplankton and other prey items, serving as dependable feeding zones
for sea turtles, marine birds, and other pelagic foragers. Unfortunately, marine debris
and oil also collect in these areas, which increases the probability of prolonged
contact with oil, tar, and harmful d e b r i s . I n convergence fronts east of Florida,
tar was found in the mouth, esophagus, or stomach of 65 out of 103 post-hatchling
loggerheads."^^ A loggerhead sea turtle sighted during a Gulf of Mexico aerial survey
surfaced repeatedly within a surface oil slick for over an hour."^^
Dramatic as it is, direct evidence of the effects of oil on sea turtles must grossly
underrepresent the problem. Oiling on the high seas and inaccessible beaches are
not likely to be documented. Secondly, only a small percentage of the animals killed
by oil would presumably strand, and of these, even fewer are likely to be found and
recorded.
15.3.2 Laboratory S tudies
Although lethal effects of oil pollution on sea turtles have been demonstrated, e.g.,
hatchlings apparently starved to death as a result of their beaks and esophagus being
blocked with tar balls, sublethal encounters are likely to be much more common,
and may have a greater deleterious impact on sea turtle populations. Laboratory
studies indicated that sea turtles will pursue and swallow tar balls indiscriminately
and that they appear to be unable to detect and avoid oil slicks."^^
Despite the high tolerance of sea turtles to physical damage, they are surprisingly
sensitive to oil. In laboratory studies on juvenile loggerhead turtles, almost all of
the major physiological systems were adversely affected by short exposures to
weathered South Louisiana crude oil.^^’^^ Significant changes in respiration, diving
patterns, energy metabolism, and blood chemistry occurred following oiling. A three-
to sixfold increase in white blood cell count suggested that a rapid immune response
was initiated in oiled turtles. Hematocrit and hemoglobin concentration decreased
slightly during oiling. These parameters are critical components of the oxygen trans
port system of the blood. Changes in metabolic indicators showed temporary inter
ference with digestion. The salt glands, paired organs located in the eye orbits (see
Chapter 13), failed to produce fluid several days after oil exposure, although function
returned within two weeks.^^’^^ Mechanical blockage of tear ducts or an inflammatory
response to the presence of oil may have inhibited salt gland function. Prolonged
salt gland failure could have serious if not fatal consequences, since it would interfere
with water balance and ion regulation. Turtles also experienced acute contact der
matitis, or disruption of the architecture of the skin, mucous membranes, and sensory
o r g a n s . A break in the skin barrier could act as a portal of entry for pathogenic
organisms, leading to infection and debilitation.
Petroleum hydrocarbons were not detected in skin or blood samples from log
gerhead sea turtles that were exposed to a weathered oil slick for no more than 96
However, there is some evidence of bioaccumulation in sea turtles exposed for
longer periods. After the Gulf of Iraq war, a stranded green sea turtle had 4050 and
310 mg of oil per kilogram of tissue in its liver and stomach, respectively."^^ Externally
the turtle did not appear to have been oiled, and until necropsy had been assumed
to have died of natural causes.
Kemp’s ridley sea turtle embryos from eggs incubated in sand contaminated in
situ by the Ixtoc blowout had survival rates similar to eggs incubated in unoiled
sand."^^ However, Ixtoc oil was highly weathered, and would have lacked most of its
volatile components. In contrast, the number of viable eggs and hatchlings surviving
to be released from nests incubated in the laboratory were significantly lower in
nests contaminated by unweathered crude oil than that of controls.^® It should be
emphasized that the laboratory studies only concerned the effects of weathered oil
— prolonged contact with freshly spilled oil would undoubtedly be considerably
more harmful. It stands to reason that the cumulative harmful effects of an oil spill
can be catastrophic.
15.3,3 O il S pill R esponse A ctivities
Oil-spill response activities have great potential to harm sea turtles if earned out
without regard to sea turtle behavior. Beach cleaning operations involve artificial
lighting and heavy machinery which may disturb the beach profile and disorient
nesting sea turtles and hatchlings. Because of the timing of critical events during
embryogenesis, sea turtle eggs can only be moved during specific periods of incu
bation, which limits relocation as a mitigation option.^®’^^
Booms used to contain oil may impede access, cause direct fouling or entangle
ment, and increase predation risks to hatchlings. Increased vehicular and vessel
traffic in shallow areas such as seagrass beds can adversely affect foraging habitat
and cause displacement from preferred areas. Oil spill contingency plans for areas
where sea turtles are found should have specific response procedures in place to
protect sea turtles.
15.4 O T H E R P O L L U T IO N S O U R C E S A N D
E N TA N G LE M EN T
Sea turtles can achieve life spans greater than 50 years, and have a potential to
bioaccumulate heavy metals and pesticides. In the coastal zone, chronic pollution
from industrial, agricultural wastes, and urban runoff is a threat to sea turtles.^^’^"^’^^’^^
Unfortunately, there is little information about this problem. In Chesapeake Bay,
DDT and its metabolites DDE and DDD were found in gonad, fat, liver, and muscle
tissues of loggerheads and Kemp’s ridleys.^"^ In some marine mammals, chronic
pollution from industrial or agricultural sources has been linked with immune sup
pression, raising a concern for sea turtles. Marine turtle fibropapilloma disease is
currently associated with a viral infection (see Chapter 14); however, the expression
of the disease may be mediated by a compromised immune system (Figure 15.2).^^
15.4.1 P lastics an d N onbiodegradable D ebris
In 1985, a published report documenting 79 cases of ingestion and 60 cases of

entanglement in marine debris by sea turtles worldwide focused attention on this
issue.Since then there has been an increased effort to quantify incidents of entan
glement and ingestion. Although there is no quantitative estimate of the worldwide
threat of marine debris to sea turtles, results from field and laboratory investigations
have played an important role in legislation banning the disposal of plastics at sea.^^’^^
FIGURE 15.2 A hawksbill turtle found entangled in plastic debris and an onion wrap.
Much of the marine debris that entangles sea turtles originates from boats, ships, fishing
vessels, and offshore oil and gas platforms. (Photo courtesy of Anthony Amos, University of
Texas; Marine Science Institute, Port Aransas, Texas.)
15.4.2 Ingestion
Reports of debris ingestion exist for almost all sea turtle species and life stages.
However, the dependence of pelagic juveniles upon convergence zones, where
floating debris concentrates, and the indiscriminate nature of their feeding strategy
leave pelagic turtles most susceptible to debris ingestion.^^’^"^ Adult and subadult
leatherback and loggerhead turtles utilize convergence fronts for travel routes.
Unfortunately, they too are at increased risk of ingesting and becoming entangled
in debris.
Marine debris is also prevalent in nearshore habitats, and there are numerous
reports documenting the occurrence of marine debris in the digestive tracts of post
hatchling and juvenile loggerheads,^^ green tuitles,^^ l ea th erb ac ks, K em p’s rid-
leys,"^"^ and hawksbill turtles.
Debris ingestion by sea turtles appears to be high in some regions and low in
others. For example, 32% of post-hatchling loggerheads captured off the east coast
of Florida had ingested plastics and synthetic fibers,and high frequencies of debris
ingestion has been reported for juvenile sea turtles in the northwestern Gulf of
Mexico^^’^^ and on the Gulf and Atlantic coasts of Florida.^"^ In contrast, low fre
quencies of debris ingestion were reported for Kemp’s ridleys, loggerheads, and
green turtles in the New York Bight.^^ Loggerhead and leatherback turtles apparently
ingest more debris than other species.The benthic foraging strategy of loggerheads
in nearshore habitats, where debris accumulation may be quite high, lends itself to
increased chance of encounter with debris. Leatherbacks are pelagic foragers that
prey on cnidarians and other gelatinous invertebrates, and it has been suggested that
they may not distinguish between their gelatinous prey and transparent plastic, the
type of debris most often present in their digestive tracts.
The list of materials found in turtle digestive tracks is extraordinary. Plastic bags,
sheeting, beads, pellets, line, rope, strapping, pieces from bottles, and hard pieces
of unknown origin are common items ingested by sea turtles.^^ '^'^ Sea turtles have
also ingested latex balloons, aluminum, paper, cardboard, styrofoam, rubber, string,
cigarette filters, wax, cellophane, fish hooks, charcoal, and g l a s s . I n d e e d , it is
possible that hungry sea turtles, particularly loggerheads, will swallow almost any
material of a suitable size and consistency and will continue to do so until satiated.
15.4.3 P hysiological Effects of D ebris I ngestion
Debris ingestion is not often directly implicated in turtle mortality, namely because
only small quantities of debris well mixed with natural food items are generally
seen in the digestive tract."^"^’^"^’^^ Debris in the digestive tract is commonly seen during
necropsies^^ and is defecated by live captured sea t u r t l e s . I s the ingestion of
debris, in particular plastic and latex, by sea turtles therefore any cause for concern?
Clearly, if sufficient material is swallowed to cause a complete stoppage of the gut,
death will result from starvation. This appears to be a relatively rare event in
comparison with the frequency with which smaller amounts are ingested.However,
the situation is much more serious if sea turtles are harmed by ingestion of plastics
and latex at quantities much smaller than those necessary for gut strangulation.
Controlled studies where turtles were presented with small amounts of debris
demonstrated that when hungry, sea turtles will actively consume plastic and latex
material.^®’^^ The sojourn of the ingested latex material in the gut ranged from a few
days to four months, indicating that some of the latex pieces were being held up in
the intestine. Although little change in blood chemistry was noted following inges
tion in these experiments, except for a fall in blood glucose levels, some turtles
became positively buoyant.^® '^^ In the wild, positive buoyancy would render afflicted
individuals more susceptible to collisions with boats, increase the chance of inci
dental capture in some commercial fisheries, and increase predation risk.^^ Clearly,
information is needed on the physiological effects of ingestion of larger quantities
of plastics by sea turtles and the levels at which adverse impacts occur.^^ It has also
been suggested that ingested plastic could result in PCB accumulation.^"^
1 5 . 4 .4 Entanglement
Entanglement in derelict fishing gear is a serious problem for sea turtles. Entangle
ment may occur when sea turtles accidentally swim into the derelict fishing gear or
when they are attracted to the accumulated epibionts encrusting old floating debris.
Leatherbacks have been reported entangled in active crab and lobster traps and even
whelk lines,^^’^^ and there are records of juvenile hawksbills trapped in plastic-woven
sacks.Entanglement can result in reduced mobility, making the turtle unable to
feed or flee from predators. If forced to remain at the surface they may become more
susceptible to collision with boats or incidental capture. Trailing debris may trap
sea turtles between rocks or ledges, resulting in death from drowning, and constric
tion of the neck and flippers can amputate limbs and consequently lead to death
from infection.
15.4.5 T ypes of D ebris and S ources of O rigin
Much of the marine debris that entangles sea turtles originates from boats and ships
and from offshore oil, gas, and chemical platforms. Some marine debris comes from
land, entering the oceans via rivers and beaches. Plastics are the most abundant type
of anthropogenic debris found on beaches and in the oceans.^^ Plastics are also the
most common type of debris sea turtles become entangled in and ingest.^^ As the
general use of plastics for industrial and commercial purposes increased over the
last 40 years, so have the amounts and types of plastic appearing in the oceans.
Fishing gear, and in particular monofilament line, is one of the most commonly
encountered anthropogenic debris that endangers sea turtles, and may account for
68% of all entanglement cases.^^’^^ In addition to material commonly perceived as
gear, plastic-fiber bags, sheeting, strapping, cloth, burlap bags, plastic 6-pack yokes,
steel cable, aluminum beach lounge chairs, and polypropylene rope are also impli
cated.
15.5 F IS H IN G A N D IN C ID E N T A L C A P T U R E
In a comprehensive review of sources of mortality in sea turtles conducted by the
National Academy of Science, incidental capture in shrimp trawls was said to account
for more deaths than all other sources of human activities combined.^^ Before the
implementation of protective measures, direct mortality of an estimated 5,000 to
50,000 loggerheads and 500 to 5,000 Kemp’s ridleys was believed to occur yearly
in U.S. waters as a result of drowning in shrimp trawls.Sea turtles have also been
killed when incidentally captured in purse seine, gill nets, and various types of
untended fishing gear. Lobster and crab pots and hook and line fishing also cause
entanglement, mutilation, and debilitation. The mortalities associated with these
fishing categories are estimated to be about 10% of that attributed to shrimp trawl-
ing.2’
Because sea turtles are exceptional breath-hold divers, there was initial skepti
cism about the estimated numbers of turtles caught and killed by inadvertent capture
in trawls. This was shortly cast aside by a series of field and laboratory studies
examining the impacts of enforced submergence. A study examining the relationship
of tow time and sea turtle mortality showed that mortality was strongly dependent
on trawling duration, with the proportion of dead or comatose sea turtles rising from
0% for the first 50 min of capture to 70% after 90 min capture.^"^ Loggerhead turtles
captured in shrimp trawls with tow times of less than 30 min showed severe acidosis,
and acid-base recovery times took as much as 20 h.^"^ In laboratory and field studies,
forcibly submerged loggerhead, green, and Kemp’s ridley sea turtles suffered sub
stantial acidosis within 30 min, and also required many hours to restore acid-base
b a l a n c e . T h e physiological disruptions associated with enforced submergence
(see Chapter 10) would be sufficient to cause a comatose state leading eventually
to death.^^ Sea turtles forcibly submerged in any type of restrictive fishing gear would
eventually suffer fatal consequences from prolonged anoxia and/or seawater infil
tration of the lung.
15.5.1 S hrimp T rawling and T urtle Excluder D evices
In 1978, the U.S. National Marine Fisheries Service and Sea Grant Program under
took development of trawl modifications (turtle excluder device, or TED) that would
allow captured sea turtles to escape the trawl through an escape hatch (Figures 15.3
and 15,4). By 1987, regulations required seasonal TED use in offshore shrimp
trawlers from North Carolina to Texas, although these were not finalized and fully
enforced until September 1989. Following years of unsuccessful voluntary compli
ance programs, legislation was passed in 1994 requiring year-round use of certified
TEDs by U.S. shrimpers from North Carolina to Texas. The use of TEDs in South
Carolina waters is believed to have reduced annual loggerhead strandings by 44%.^^
Repeated capture by trawlers in intensely fished areas may cause some sea turtle
mortality (despite the use of TEDs) in cases where sea turtles are unable to com
pensate for repeated metabolic stress.
Because of the high product value, shrimping intensity is high in coastal regions
worldwide.^'^ Given the highly migratory nature of sea turtles, protective measures
on the part of one country would shortly be undone by unrestricted fishing practices
of adjacent countries. In order to conserve the sea turtles, mitigation of incidental
catch must be addressed by all shrimp-fishing nations. In 1993, Mexico required
that offshore trawlers use TEDs in Atlantic and Caribbean fishing zones. The concept
and application of the TED has expanded, and these devices are now known as
“bycatch excluder devices” in recognition of their value in reducing undesired
bycatch of finfish by as much as 50 to 70%.^^ The U.S. National Marine Fisheries
FIGURE 15.3 A loggerhead sea turtle shown escaping from a trawl net via the turtle
excluder device (TED). (Photo courtesy of NOAA, National Marine Fisheries Service, Mis
sissippi Laboratories, Pascagoula Facility, MS.)
Service is currently providing bycatch and TED advisory services to fishing nations
of the Pacific Rim and Indian Ocean.
15.5.2 P elagic Fishing G ear
Sea turtles are also vulnerable to capture in pelagic longline, paired trawl, and gill
net fisheries, a source of increasing concern for both fisheries managers and the
industry. Incidental sea turtle take on swordfish and tuna fishing grounds had been
noted since the 1950s, yet more recent information suggests that interactions may
be higher than acknowledged. The pelagic swordfish fishery has shifted focus since
the late 1980s, encompassing wider fishing areas in the Atlantic and Caribbean, and
new gear types such as paired trawls. Estimated total take of sea turtles in the
Northwest Atlantic for 1989 to 1992 was 1218 turtles.From 1991 to 1993, NMFS
observers on 54 pelagic longline trips targeting swordfish, tuna, sharks, and mixed
species recorded 56 leatherbacks and 29 hardshelled or unidentified sea turtles.In
the Hawaiian longline tuna fishery,^^ leatherback captures were documented from
January to December 1992.^^
This pattern of high incidental catches may have been common in U.S. waters
since the opening of the fisheries, and appears to be selective for certain species.
WitzelE^ estimated that 79.6% of the observed take (57 sea turtles) of the Japanese
tuna longline fleet in the Gulf of Mexico and Atlantic (1978 to 1981) were leather
backs. Lethal encounters appear to be higher for chelonid sea turtles, usually found
impaled on the hook.^^’^"^’^^ Leatherbacks are usually found alive, but fouled in
longline gangions (drop lines) or buoy lines.^^’^'^Longlines are soaked after sunset,
FIGURE 15.4 A schematic showing an example of a recent turtle excluder device. A variety
of approved TED designs are available to maximize exclusion of sea turtles and bycatch under
different fishing conditions. (Schematic courtesy of NOAA, National Marine Fisheries Ser
vice, Mississippi Laboratories, Pascagoula Facility, MS.)
and chemical lightsticks are commonly used to lure target species to the bait.
Leatherbacks may be attracted to the lightsticks, or medusae fouled on longline
sections, whereas chelonid species such as loggerheads, greens, and Kemp’s and
Pacific ridleys appear to seek baited hooks. Even if successfully freed from longline
gear, sea turtles may suffer subsequent mortality from hook wounds, internal bleed
ing, or entanglement.
15.5.3 C oastal Fishing A ctivities
In recent years the documentation of sea turtles in coastal regions and their incidental
take by inshore fisheries has received considerable attention. Collectively, unattended
nets set in shallow waters and fisheries other than shrimping are the second largest
source of mortality to sea turtles.Coastal fishing practices including pompano gill
nets and trammel nets for shark and bluefish (Atlantic Florida), summer flounder
trawling (southern Virginia to North Carolina), shark and sturgeon nets (Georgia,
South Carolina), and pound nets and weirs (Atlantic coast) also have incidental takes
of sea turtles.^^’^^’^^ Sea turtle mortality associated with these fisheries varies in
response to the seasonal abundance of turtles and to the intensity and timing of
fishing effort. The short pulses of high mortalities associated with some of these
fisheries may disproportionately impact highly depleted species such as the Kemp’s
ridleys. Gear modification and short-term closures are being evaluated as mitigating
measures for some fisheries.
If we are to protect sea turtles from further population decline we clearly need
to encourage the development of fishing techniques that minimize or eliminate
incidental take. This task is difficult to achieve until we understand the nature of
sea turtle migration and foraging routes, particularly in offshore areas where migra
tion routes are poorly known. Incidental take records and anecdotal observations
from the fisheries document a notable abundance of sea turtles on shelf break (200-
m depth contour) or at edges of oceanic gyre systems such as the Gulf Stream
western boundary and Gulf of Mexico Loop Current.^'^’^'^’^^’^'^’^^’^^ New technologies
such as satellite tracking, remote sensing of the environment, and other techniques
may help quantify the types and degrees of interaction of sea turtles with human
activities on the high seas. Observer programs and better recordkeeping by fishermen
may also help quantify the impacts of fishing and offer solutions for mitigation
efforts.
15.5.4 A rtisanal and C ommercial S ea T urtle Fisheries
The worldwide decline of nearly all sea turtle populations can be attributed in part
to the direct take of turtles and their eggs by commercial and subsistence sea turtle
fisheries. Particularly in developing countries, the need for protein and income
generated by sea turtles and their by-products greatly exceeds any desire for their
preservation. Although all sea turtle life stages have been exploited, perhaps with
the exception of hatchlings, breeding adults and their eggs have been most vulner
able. Nesting females and the eggs they leave buried in the sand have provided easy
access to an otherwise difficult to reach resource. Mating areas and nearshore feeding
grounds, particularly reefs frequented by hawksbills, also provide relatively easy
access to large numbers of sea turtles.
In many cultures, sea turtle eggs are eaten not only for nutrition, but also because
of puiported aphrodisiac qualities. Green turtles are sought after primarily for their
meat and cartilage (calipee), as well as their shells and bodies, which are sometimes
stuffed and sold as curios for the tourist trade. Even bones are used to make
houseware items, decorative artwork, ornaments, jewelry, and fertilizer.Pacific
ridley skin is used to fashion boots, shoes, handbags, and other accessories, and oil
rendered from the carcass is used for everything from skin lotion additives to boat
caulking agents. Hawksbills are highly desired for their beautiful shells, which are
polished and hung on the wall as a decorative ornament, or the scutes from their
shells (“bekko”) used to craft jewelry, combs, and eyeglasses.
Since prohibited by the Endangered Species Act of 1973, commercial sea turtle
harvests no longer exist within continental U.S. waters, and were recently banned
in Mexico.^^ Sea turtle fisheries, nonetheless, continue elsewhere. Since the mid-
1970s, the primary importers of sea turtle shell and skin have been Japan, Hong
Kong, China, Singapore, Taiwan, and some European countries.^^^^^^’^^ The major
exporters have been Panama, Honduras, Jamaica, Haiti, Cuba, Cayman Islands,
Maldives, Fiji, Solomon Islands, Indonesia, Philippines, Thailand, Kenya, and Tan
zania.^®These countries generally harvest only turtles from their beaches and
territorial waters, and in many instances feel their activities are justified because
they exploit their own resources. But in reality, sea turtles are highly migratory
species that, on a geographic scale, are a regional resource to be managed and shared
among many countries.
Commercial exploitation has resulted in overharvesting and in declines or local
extinctions of the populations harvested. Unfortunately, the effects of a commercial
harvest are not immediately apparent and may give the false impression that con
tinued exploitation will not be detrimental to the population being harvested. This
is best illustrated by the Pacific sea turtle fishery in Mexico which harvested hundreds
of thousands of olive ridleys annually over several decades. This lead to the demise
of at least three local nesting populations and a precipitous decline of the species
in the eastern Pacific Ocean.
Legal and illegal subsistence sea turtle hunts and egg harvests occur worldwide,
but in contrast to commercial sea turtle fisheries, they supply local markets or tribes,
and in some instances the international tourist trade.^^ The level of subsistence take
and the corresponding threat to a particular population is difficult to determine
because the number of sea turtles and eggs taken generally is not documented or
reported. In some regions subsistence hunts are controlled by annual quotas and
appear to pose little threat to the resource.In other areas where quotas are nonex
istent, indigenous peoples have long ago exceeded the sustainable level of take and
have extirpated sea turtle populations.
There is ample evidence that human activity is seriously eroding once abundant sea
turtle populations. Much of the impact is unintentional, a consequence of the
increased exploitation of marine and coastal waters — the more intense fishing effort
results in an increase in the incidental catch of sea turtles and a higher risk of
entanglement in discarded gear. Debris and pollution in coastal areas take their toll
year after year. The inexorable spread of beach development eats away at natural
sea turtle nesting habitats and deters nesting females. Even well-intentioned eco-
tourism programs, which seek to educate visitors to turtle nesting beaches and offer
alternative employment for subsistence turtle hunters, have the potential for harm.
Poorly regulated foot traffic, noise, and lights create disturbances that may deter
nesting females and threaten hatchlings. There are also deliberate threats to sea turtle
survival, a consequence of both local economic pressures and international commer
cial interests which drive continued (even if illegal) fishing for juveniles and adults
and a massive taking of eggs.
In order to be biologically effective, conservation programs must be firmly based
on knowledge of how and to what extent humans are jeopardizing turtle survival.
In order to be accepted both politically and practically, conservation programs must
also recognize the economic forces behind the disturbing influences. These will be
different in different countries and even in different regions, so that implementable
solutions are/will be complex. Nevertheless, action is urgently need to halt the
decline and turn the situation around. Otherwise, the extinction of local populations
and even some species is inevitable.
R EFER EN C ES
1. Pritchard, P. C. H., Why nest relocation programs may be harmful to sea turtles, in
P ro c. 5 th A n n u a l N a t. C o n f. o n B e a c h P r e s e r v a tio n T e c h n o lo g y , Tait, L. S., Ed.,
Florida Shore and Beach Preservation Association, Tallahassee, FL, 408, 1992.
2. Pritchard, R, Bacon, R, Berry, R, Carr, A., Fletemeyer, J., Gallagher, R., Hopkins,
S., Lankfor, R., Marquez, R., Ogren, L., Rringle, W., Reichart, H., and Witham, R.,
M a n u a l o f S e a T u rtle R e s e a r c h a n d C o n s e r v a tio n T e c h n iq u e s, 2nd ed., Bjorndal, K.
and Balazs, G., Eds., Center for Environmental Education, Washington, D.C., 1983.
3. Wyneken, J., Burke, T. J., Slamon, M., and Pedersen, D. K., Egg failure in natural
and relocated sea turtle nests, J. H e r p e to L , 22, 88, 1988.
eggs, A m . Z o o l., 20, 575, 1980.
5. Shaver, D. J., Owens, D. W., Chaney, A. H., and Caillouet, C. W., Jr., Burchfield, R,
and Marquez, R., Styrofoam box and beach temperatures in relation to incubation
and sex ratios of Kemp’s ridley sea turtles, in P ro c . 8 th A n n u . W o rk sh o p o n S e a T u rtle
C o n s e r v a tio n a n d B io lo g y , Schroeder, B. A., Compiler, NOAA Tech. Memo. NMFS-
SEFC-214, Miami, FL, 103, 1988.
6. Pilkey, O. H., Jr., Sharma, D. C., Wanless, H. R., Doyle, L. H., Pilkey, O. H., Sr.,
Neal, W. J., and Gruver, B. L., L iv in g w ith th e E a s t F lo r id a S h o re , Duke University
Press, Durham, NC, 1984.
7. Shulman, A. A., Milton, S. A., and Lutz, R L., Aragonite sand as a nesting substrate
and its effect on C a re tta c a re tta nests, in P ro c. 1 4 th A n n u . S y m p . S e a T u rtle B io lo g y
a n d C o n se rv a tio n , Bjorndal, K. A., Bolten, A. B., Johnson, D. A., and Eliazar, R. J.,
Compilers, NOAA Tech. Memo. NMFS-SEFSC-351, Miami, FL, 134, 1994.
8. Whitmore, C. R, Mitigation of negative impact of tourism on nesting beaches and
turtles in Turkey, in P ro c . 1 3 th A n n u . S y m p . S e a T u rtle B io lo g y a n d C o n se rv a tio n ,
Schroeder, B. A. and Witherington, B. E., Compilers, NOAA Tech. Memo. NMFS-
SEFSC-341, Miami, FL, 190, 1994.
9. Mortimer, J. A., Reproductive Ecology of the Green Turtle, C h e lo n ia m y d a s, at
Ascension Island, Ph.D. dissertation. University of Florida, Gainesville, 1981.
10. Stella, I., The sea turtles in the Eastern Mediterranean in danger of extinction, in
B io lo g y a n d C o n s e r v a tio n o f S e a T u rtles, Bjorndal, K. A., Ed., Smithsonian Institution
Press, Washington, D.C., 417, 1982.
11. Hosier, R E., Kochhar, M., and Thayer, V., Off-road vehicle and pedestrian track
effects on the sea-approach of hatchling loggerhead ixxriXt^, E n v ir o n C o n se rv , 8 , 158,
1981.
12. Cox, J. H., Percival, H. R, and Colwell, S. V., Impact of vehicular traffic on beach
habitat and wildlife at Cape San Bias, Florida, Florida Cooperative Fish and Wildlife
Research Unit, U.S. Biological Survey Tech. Rep. 50, 1994.
13. Salmon, M., Wyneken, J., Fritz, E., and Lucas, M., Seafinding by hatchling sea turtles:
role of brightness, silhouette and beach slope as orientation cu&s. B e h a v io u r , 122, 56,
1992.
14. Mann, T. M., Impact of developed coastline on nesting and hatchling sea turtles in
Southeastern Florida, F la . M ar. R e s . P u b l., 33, 53, 1978.
15. Fangman, M. S. and Rittmaster, K. A., Effects of human beach usage on the temporal
distribution of loggerhead nesting activities, in P ro c. 1 3 th A n n u . S y m p . S e a T u r tle
B io lo g y a n d C o n s e r v a tio n , Schroeder, B. A. and Witherington, B. E., Compilers,
NOAA Tech. Memo. NMFS-SEFC-34L Miami, FL, 222, 1993.
16. Johnson, S. A., Bjorndal, K. A., and Bollen, A. B., Influence of organized turtle
watches on loggerhead nesting behavior and hatchling production in Florida, m P r o c .
1 4 th A n n u . S y m p . S e a T u r tle B io lo g y a n d C o n s e r v a tio n , Bjorndal, K. A., Bollen,
A. B., Johnson, D. A., and Eliazar, P. J., Compilers, NOAA Tech. Memo. NMFS-
SEFSC-351, Miami, FL, 64, 1995.
17. Stancyk, S. E., Non-human predators of sea turtles and their control, i n B io lo g y a n d
C o n s e r v a tio n o f S e a T u rtles, Bjorndal, K. A., Ed., Smithsonian Institution Press,
18. Witherington, B. E., Behavioral responses of nesting sea turtles to artificial lighting,
H e r p e to lo g ic a , 48, 31, 1992.
19. Peters, A. and Verhoeven, K. J. R, Impact of artificial lighting on the seaward
orientation of hatchling loggerhead turtles,/. H e r p e to L , 28, 112, 1994.
20. Philibosian, R., Disorientation of hawksbill turtle \\diic\\Ymgs,Eretniochelys imbricata,
by stadium lights, Copela, 1976, 824, 1976.
21. McFarlane, R. W., Disorientation of loggerhead hatchlings by artificial road lighting,
C o p e la , 1963, 153, 1963.
hatchlings: evidence for lunar modulation, C o p e la , 1995, 931, 1995.
23. Witherington, B. E. and Bjorndal, K. A., Influences of wavelength and intensity on
hatchling sea turtle phototaxis: implications for sea-finding behavior, Cope/«, 1991,
1060, 1991.
24. Witherington, B. E. and Bjorndal, K. A., Influences of artificial lighting on the seaward
orientation of hatchling loggerhead turtles {C aretta c a re tta ), B io l. C o n se rv ., 55, 139,
1991.
25. Schroeder, B. A., A n n u . R e p o r t o f the S e a T u rtle S tr a n d in g a n d S a lv a g e N e tw o r k
A tla n tic and G u l f C o a s ts o f th e U n ite d S ta te s J a n u a r y - D e c e m b e r 1986,
NOAA/NMFS, Coastal Resource Division contribution number CRD-87/88-12,
Miami, FL, 1987.
26. Teas, W. G., A n n u . R e p o r t o f th e S e a T u rtle S tr a n d in g a n d S a lv a g e N e tw o r k A tla n tic
a n d G u l f C o a s ts o f th e U n ite d S ta te s J a n u a r y - D e c e m b e r 1993, NOAA/NMFS,
Miami Laboratory contribution number MIA-94/95-12, Miami, FL, 1994.
27. National Research Council, D e c lin e o f th e S e a T u rtle s: C a u s e s a n d P re v e n tio n ,
National Academy Press, Washington, D.C., 1990.
28. Slay, C. K. and Richardson, J. I., King’s Bay, Georgia: dredging and turtles, in P r o c .
8 th A n n u . W o rk sh o p o n S e a T u rtle C o n s e r v a tio n a n d B io lo g y, NOAA Tech. Mem.
NOAA-TM-NMFS-SEFSC-214, Miami, FL, 109, 1988.
29. Klima, E. R, Gitschlag, G. R., and Renard, M. L., Impacts of the explosive removal
of offshore petroleum platforms on sea turtles and dolphins, Ma/: F ish . R ev., 50, 33,
1988.
30. Duronslet, M. J., Caillouet, C. W., Manzella, S., Indelicata, K. W., Fontaine, C. T.,
Revera, D. B., Williams, T., and Boss, D., The effects of an underwater explosion on
the sea turtles Lepiochelys kempi and Caretta caretta with observations of effects on
other marine organisms (trip report), U.S. Department of Commerce, National Marine
Fisheries Service, Southeast Fisheries Center, Galveston, TX, 1986.
tissues of sea turtles exposed to the Ixtoc I oil spill,/. W ildl. D is ., 19, 106, 1983.
32. Witham, P. R., 1978. Does a problem exist relative to small sea turtles and oil spills?,
in Proc. Conf. the Assessment of Ecological Impacts of Oil Spills, American Institute
of Biological Science, 1978.
33. Raloff, J., When sea turtles are awash in oil, Sci. News, 30, 358, 1986.
34. Hutchinson, J. and Simmonds, M., Escalation of threats to marine turtles, Oryx, 26,
95, 1992.
35. Gramentz, D., Involvement of loggerhead turtle with the plastic, metal, and hydro
carbon pollution in the central Mediterranean, Mar. Pollut. Bull., 19, 11, 1988.
36. Frazier, J. and Salas, S., The status of marine turtles in the Egyptian Red S q2ì,B ìoI.
Conserv., 30, 41, 1984.
37. Butler, J. N., Morris, B. F., and Sass, J., Pelagic Tar from Bermuda and the Sargasso
Sea, The Bermuda Biological Station for Research, Cambridge, MA, 1973.
38. Lutz, R L., Lutcavage, M., and Hudson, D., Physiological effects, in Final Report.
Study of the Effect of Oil on Marine Turtles, Vargo, S., Lutz, P. L., Odell, D. K., Van
VIeet, T, and Bossart, G., Eds., Minerals Management Service Contract Number 14-
12-0001-30063, Florida Inst, of Oceanography, St. Petersburg, FL, 1986.
39. Lutcavage, M., Lutz, R, Bossart, G., and Hudson, D., Physiologic and clinicopatho-
logic effects of crude oil on loggerhead sea turtles, Arc/z. Environ. Contam. Toxicol,
28, 417, 1995.
40. Fritts, T. H. and McGehee, M. A., Effects of Petroleum on the Development and
Survival of Marine Turtle Embryos, U.S. Fish and Wildlife Service, U.S. Department
of the Interior, Washington, D.C., Contract No. 14-16-0009-80-946, FWS/OBS-
81/37, 1981.
41. Anon., Oil Spill Intelligence Report, Iraq forces strike 26-well platform south of
Nowruz, Volume VI, 22 April, 1983.
42. Greenpeace, The Environmental Legacy o f the G ulf War, Greenpeace International,
Amsterdam, 1992.
43. Rabalais, S. C. and Rabalais, N. N., The occurrence of sea turtles on the south Texas
coast, Contrib. Mar. Sci., 23, 123, 1980.
44. Plotkin, P. and Amos, A. R, Effects of anthropogenic debris on sea turtles in the
northwestern Gulf of Mexico, in Proc. Second Int. Conf. on Marine Debris, Shomura,
R. S. and Godfrey, M. L., Eds., NOAA Tech. Memo. NMFS-SWFS-154, Honolulu,
HI, 1990.
45. Van VIeet, T. and Pauly, G. G., Characterization of oil residues scraped from stranded
sea turtles from the Gulf of Mexico, Cariò. J. Sci., 23, 77, 1987.
46. Witherington, B. E., Flotsam, jetsam, post-hatchling loggerheads, and the advecting
surface smorgasbord, in Proc. 14th Annu. Symp. Sea Turtle Biology and Conservation,
Bjomdal, K. A., Bolten, A. B., Johnson, D. A., and Eliazar, P. J., Compilers, NOAA
47. Loehefener, R. R., Hoggard, W., Roden, C. L., Mullin, K. D., and Rogers, C. M.,
Petroleum structures and the distribution of sea turtles, in Proc. Spring Ternary Gulf
of Mexico Studies Meeting, Minerals Management Service, U.S. Department of the
Interior, New Orleans, 1989, 31.
48. Odell, D. K. and MacMurray, C., Behavioral response to oil, in Final Report. Study
of the Effect of oil on Marine Turtles, Vargo, S., Lutz, P. L., Odell, D. K., Van VIeet,
T , and Bossart, G., Eds., Minerals Management Service Contract Number 14-12-
0001-30063, Florida Inst, of Oceanography, St. Petersburg, FL, 1986.
49. Bossart, G. D., Clinicopathological Effects, in Final Report. Study of the Effect of
Oil on Marine Turtles, Vargo, S., Lutz, P. L., Odell, D. K., Van VIeet, T , and Bossart,
G., Eds., Minerals Management Service Contract Number 14-12-0001-30063, Florida
Inst, of Oceanography, St. Petersburg, FL, 1986.
50. McGehee, M. A., Factors Affecting the Hatching Success of Loggerhead Sea Turtle
Eggs {C a retta c a re tta c a re tta ), Masters thesis, University of Central Florida, Orlando,
1979.
51. Blande, C. E. and Sawyer, R. H., 1981. Hatchery practices in relation to early
embryology of the loggerhead sea turtle. C a re tta c a r e tta (Linné), J. E x p . M ar. B io l.
E c o L , 49, 163, 1981.
52. Frazier, J. G., Marine turtles and problems in coastal management, i n C o a s t a l Z o n e
'80, Voi. Ill, Edge, B. L., Ed., Proc. 2nd Symposium on Coastal and Ocean Manage
ment, 2395, 1980.
53. Clark, D. R., Jr. and Krynitsy, A. J., DDE residues and artificial incubation of
loggerhead sea turtle eggs. B u ll. E n v iro n . C o n ta m . T o x ic o l, 34, 1985, 121.
54. Rybitski, M. J., Hale, R. C., and Musick, J. A., Distribution of organochlorine
pollutants in Atlantic sea turtles, C o p e ia , 1995, 372, 1995.
55. Herbst, L. H., Fibropapillomatosis of marine turtles, A n n u . R ev. F is h D is., 4, 389,
1994.
56. Balazs, G. H., Impact of ocean debris on marine turtles, in P ro c . W o rk sh o p o n the
F a te a n d I m p a c t o f M a r in e D e b ris, Shomura, R. S. and Godfrey, M. L., Eds., NOAA
Tech. Mem. NMFS-SWFC-54, 580, Honolulu, HI, 1985.
57. Shoop, C. R. and Kenney, R. D., Seasonal distributions and abundances of loggerhead
and leatherback sea turtles in waters of the Northeastern United States, H e rp e to l.
M o n o g r., 6, 43, 1992.
58. Epperly, S. R, Braun, J., and Chester, A. J., Aerial surveys for sea turtles in North
Carolina inshore waters. F is h . B u ll., 93, 254, 1995.
59. Brongersma, L. D., Notes upon some turtles from the Canary Islands and from
Madeira, P ro c. K . N e d . A k a d . Wet. S e n C, 128, 1968.
60. vanNierop, M. M. and denHartog, J. C., A study of the gut contents of five juvenile
loggerhead turtles. C a re tta c a re tta (Linnaeus) (Reptilia, Cheloniidae), from the south
eastern part of the North Atlantic Ocean, with emphasis on coelenterate identification,
Z o o l. M e d e d . (L e id e n ), 59, 35, 1984.
61. Hughes, G. R., T h e S e a T u rtle s o f S o u th - E a s t A fr ic a II, Investigational Report No.
36, Oceanographic Research Institute, Durban, South Africa, 1974.
62. Plotkin, P. and Amos, A. F, Entanglement in and ingestion of marine debris by sea
turtles stranded along the south Texas coast, in P ro c . 8 th A n n u . W o rk sh o p on S e a
T u rtle B io lo g y a n d C o n se rv a tio n , NOAA Tech. Memo. NMFS-SEFC-214, Miami,
FL, 1988.
63. Plotkin, P. T, Wicksten, M. K., and Amos, A. F, Feeding ecology of the loggerhead
sea turtle C a re tta c a re tta in the northwestern Gulf of Mexico, Mar. B io l., 115,1,1993.
64. Bjomdal, K. A., Bolten, A. B., and Lageux, C. J., Ingestion of marine debris by
juvenile sea turtles in coastal Florida habitats. Mar P o ll. B u ll., 28, 154, 1994.
65. Sadove, S. S. and Morreale, S. J., Marine mammal and sea turtle encounters with
marine debris in the New York Bight and the Northeast Atlantic, in P ro c . 2 n d In t.
C o n f. o n M a r in e D e b ris, Shomura, R. S. and Godfrey, M. L., Eds., NOAA Tech.
Memo. NMFS-SWFS-154, Honolulu, HI, 1990.
66. Mrosovsky, N., Plastic jellyfish. Mar T u rtle N e w s i. , 17, 5, 1981.
67. Shaver, D. J., Feeding ecology of wild and head-started Kemp’s ridley sea turtles in
south Texas waters, J . H e r p e t o l, 25, 327, 1993.
68 . Burke, V. J., Standora, E. A., and Morreale, S. J., Diet of juvenile Kemp’s ridley and
loggerhead sea turtles from Long Island, New York, C o p e ia , 1993, 1176, 1993.
69. Werner, S. A. and Landry, A. M., Feeding ecology of wild and headstarted Kemp’s
ridley sea turtles (L e p id o c h e ly s k e m p ii), in P ro c . 1 4 th A n n u . W o rk sh o p o n S e a T u rtle
B io lo g y a n d C o n se rv a tio n , NOAA Tech. Memo. NMFS-SEFSC-351, Miami, FL,
1994.
70. Lutz, R, Studies on the ingestion of plastic and latex by sea turtles, m P r o c . 2 n d In t.
C o n f. o n M a r in e D e b r is , Shomura, R. S. and Godfrey, M. L., Eds., NOAA Tech.
Memo. NMFS-SWFS-154, Honolulu, HI, 1990.
71. Lutz, P. L. and Alfaro-Schulman, A. A., T h e E ffe c ts o f C h ro n ic P la s tic In g e s tio n on
G ree n S e a T u rtle s, Report NOAASB21-WCH06134, U.S, Department of Commerce,
Miami, FL, 1991.
72. Morgan, P. J., Occurrence of leatherback turtles (Demnochelys coriacea) in the British
Isles in 1988 with reference to a record specimen, in P ro c . 9 th A n n u . W o rk sh o p o f
S e a T u rtle C o n s e r v a tio n a n d B io lo g y , Eckert, S. A., Eckert, K. L., and Richardson,
T. H., Eds., NOAA Tech. Memo. NMFS-SEFSC-232, Miami, FL, 1989.
73. O’Hara, K. J. and ludicello, S., P la s tic s in th e O c e a n : M o re th a n a L itte r P ro b lem ,
Center for Environmental Education, Washington, D.C., 1987, 128 pp.
74. Henwood, T. and Stuntz, W. E., Analysis of sea turtle captures and mortalities during
commercial shrimp trawling. F is h . B u ll., 85, 813, 1987.
75. Lutz, P. L. and Dunbar-Cooper, A., Variations in the blood chemistry of the loggerhead
sea turtle, C a re tta c a re tta . F ish . B u ll., 85, 37, 1987.
76. Lutz, P. L. and Bentley, T. B., Respiratory physiology of diving in the sea turtle,
C o p e ia , 1985, 671, 1985.
77. Stabenau, E. K., Heming, T. A., and Mitchell, J. F., Respiratory, acid-base and ionic
status of Kemp’s ridley sea turtles { L e p id o c h e ly s k e m p i) subjected to trawling, C o m p .
B io c h e m . P h y s io l., 9 9 A , 107, 1991.
loggerhead sea turtle, J. E xp . M ar. B io l. EcoL, 147, 287, 1991.
79. Crowder, L. B., Hopkins-Murphy, S. R., and Royle, J. A., Effect of turtle excluder
devices (TEDs) on loggerhead sea turtle strandings with implications for conserva
tion, C o p e ia , 1995, 773, 1995.
80. Weber, M., Crouse, D., Irvin, R., and ludicello, S., D e la y a n d D e n ia l, a P o litic a l
H is to r y o f S e a T u r tle s a n d S h r im p F is h in g , Center for Marine Conservation, Wash
ington, D.C., 1995.
81. Gerrior, P, Incidental take of sea turtles in the northeast. Longline Fishery-Sea Turtle
Interactions, in Pelagic Longline Fishery-Sea Turtle Interactions; Proceedings of a
Workshop, Williams, R, Anninos, R, Plotkin, R T , and Salvini, K. L., Compilers,
U. S. Department of Commerce, NOAA Tech. Memo. NMFS-OPR, 1996, Silver
Springs, MD, 73 p.
82. Balazs, G. H. and Pooley, S. G., Research Plan to Assess Marine Turtle Hooking
Mortality: Results of an Expert Workshop held in Honolulu, Hawaii, Balazs, G. H.
and Pooley, S. G., Eds., U.S. Department of Commerce, Administrative Report H-
93-18, Silver Springs, MD, 1993.
83. Witzell, W. N., The incidental capture of sea turtles in the Atlantic U.S. Fishery
Conservation Zone by the Japanese tuna longline fleet, 1978-81. Mar. F is h . R ev., 46,
56, 1984.
84. Witzell, W., The incidental capture of sea turtles by the U.S. pelagic longline fleet
in the Western Atlantic ocean, in P e la g ic L o n g lin e F is h e r y - S e a T u rtle I n te r a c tio n s :
P r o c e e d in g s o f a W o rk sh o p , Williams, R, Anninos, R, Plotkin, P. T , and Salvini,
K. L., Compilers, U.S. Department of Commerce, NOAA Tech. Memo. NMFS-
OPR, Silver Springs, MD, 1996, 73 pp.
85. Ogren, L. H., Sea turtle interactions with the longline fishery in the Gulf of Mexico,
in Research Plan to Assess Marine Turtle Hooking Mortality: Results of an Expert
Workshop Held in Honolulu, Hawaii, U.S. Department of Commerce, Administrative
Report H-93-18, Balazs, G. H. and Pooley, S. G., Eds., 1993, 164 pp.
86. Epperly, S. P, Braun, J., Chester, A. J., Cross, F. A., Merriner, J. V., and Tester,
P. A., Winter distribution of sea turtles in the vicinity of Cape Hatteras and their
interactions with the summer flounder trawl fishery, Bw//. M a r. S c i., 56, 547, 1995.
87. Lutcavage, M. E. and Musick, J. A., Aspects of the biology of sea turtles in Virginia,
C o p e ia , 1985, 449, 1985.
88. Hoffman, W. and Fritts, T., Sea turtle distribution along the boundary of the Gulf
Stream Current off eastern Florida, H e r p e to lo g ic a , 38, 405, 1982.
89. Thompson, N. B., P e la g ic d is tr ib u tio n o f tu r tle s in th e G u l f o f M e x ic o , NOAA Tech.
Memo. NMFS-SEFC-286, National Marine Fisheries Service, Miami, FL, 1991.
90. Mack, D. Duplaix, N., and Wells, S., Sea turtles, animals of divisible parts: Interna
tional trade in sea turtle products, in B io lo g y a n d C o n s e r v a tio n o f S e a T u rtle s,
Bjorndal, K. A., Ed., Smithsonian Institution Press, Washington, D.C., 1982, 545.
91. Canin, J., International trade in sea turtle products, P ro c . 9 th A n n u . W o rk sh o p o n S e a
T u rtle B io lo g y a n d C o n se rv a tio n , NOAA Tech. Memo. NMFS-SEFC-232, Miami,
FL, 1989.
92. Aridjis, H., Mexico proclaims total ban on harvest of turtles and eggs. M a r T u r tle
N e w sle tte r , 50, 1, 1990.
93. Donnelly, M., International trade in hawksbill sea turtle shell in the wider Caribbean,
in P ro c. 9 th A n n u . W o rk sh o p o n S e a T u rtle B io lo g y a n d C o n se rv a tio n , NOAA Tech.
Memo. NMFS-SEFC-232, Miami, FL, 1989.
94. Cornelius, S. E. and Robinson, D. C., Post-nesting movements of female olive ridley
turtles tagged in Costa Rica, V ida S ilv e stre N e o tr o p ic a l, 1, 12, 1986.
95. Ross, J. P. and Carr, D., Comprehensive management plan for Tortuguero, Costa Rica,
in P ro c. 9 th A n n u . W o rk sh o p o n S e a T u rtle B io lo g y a n d C o n se rv a tio n , NOAA Tech.
Memo. NMFS-SEFC-232. Miami. FL. 1989.
Index
Amnion, 66
Amniotic sac, 89-90
A a p to s , 205 A m u s iu m ja p o n ic u m b a llo ti, 212
Absolute growth rates, 241, 254, 256, 303 Amydae, 6
Absorptivity, 301 A n a d a r a tr a p e z ia , 204
Abundance, 261, 262 Anaerobic adaptations and migratory swimming,
A c a n th u r u s , 213 192-193, 288-290
Acepromazine maleate, 380 Anaerobic metabolism, 192, 288-290
Acid-base balance, 288, 353-357 Anapsid, 2
ACR (air convection requirement), 283, 356 A n c o r in a , 205
Adults, 139-140 Androgen, 316
homiones, 317-319 Anemia, 366, 375
migratory swimming, 186-188, 190-192 A n e n o m ia s u lc a ta , 205
reproduction, 52-65 Anesthesia, 380
A e q u o r e a , 209 Annuli, time intervals for deposit, 238-239
Aerial surveys, 152 Anoxic tolerance, 289-290, 378
Aerobic metabolism, 190, 284—288 Anthropogenic effects. S e e a ls o Pollution; Shrimp
Aerobic scope, 189, 281 trawls
A e r o m o n a s , 368 effects of fishing gear, 398, 399-401
A e r o m o n a s h y d r o p h ila , 367, 368 entanglement, 378, 396, 398
Affinities, of recent species, 13-16 on foraging ecology and nutrition, 218-221
Age ingestion of marine debris, 218-219, 378-379,
estimated, 248-249 396-398
known, 247-248, 250-251, 253 loss of habitat, 99-103, 388-392
reproduction and, 58 Antiperistalsis, 218
Age-based growth models, 241-242, 243-245, Antitropical distribution, 20
247-251 A p o le m ia u v a r ia , 209
Age-specific reproduction axioms, 256 Aquatic locomotion, 171-181, 183-184. S e e a ls o
Age-structured models Swimming
dynamic, 263-264 aerobic adaptations, 189-192, 284-288
static, 262-263 anaerobic adaptations, 192-193, 288-290
Air circulation, 301 Aragonite, 66
Air convection requirement (ACR), 283, 356 A rc h e lo n , 8, 10
Air temperature, 304 Archie Carr Center for Sea Turtle Research, 44
Alabama, L e p id o c h e ly s k e m p i, 149 A r e n a e u s c r ib r a r iu s , 207
Albumin, 381 A r g illo c h e ly s s u b s c r ip ta ta , 11
Algae. S e e a ls o S a r g a ss iim Arginine vasotocin (AVT), 327, 328, 335
as food resource, 140, 144, 145, 200-209, 210 A r g o p e c tin ir r a d ia n s , 207
v¿’. seagrass diet, 210-211 A r iz o n a , 367
Allantois, 66 Aromatase, 316
A llo p le u r o n , 7, 8 Arom gene, 316
A llo p le u r o n h o jfm a n n i, 11 Arrhenius equation, 85-86
304
A lp h a H e lix , Arribada beaches, 22, 59
Altered beaches, 99-103, 388-392 Artificial lighting, 116-117, 388, 391
Alternating gait, 59 Ascension Island, U.K.
Amebiasis, 375-376 C h e lo n ia a g a s s iz i, 15
Ammonia, 353 C h e lo n ia m y d a s , 17, 18, 38, 39, 126, 129
411
chemosensory cues in plume, 127 Behavior

co-occupation of feeding habitat, 38 to artificial light, 116-117, 388, 391
magnetic map in migration to, 129 nesting, 52, 60, 61, 62
phylogeny of nuclear DNA genotypes, 37 precopulatory, 57
A s p a r a g o p s is , 210 reproductive, 54
Aspartate aminotransferase, 381 upon emergence. S e e Frenzy period
A s p e r g illu s , 369 Bekko, 405
Associative effects, between T h a la s s ia and Bends, 281
sponges, 217 Bernardelli-Lewis-Leslie matrix (BLL), 263,
Asymmetrical gait, 166, 167 264
Athecae, 33 Bibron, 6
Athecoidea, 6, 7 Bicarbonate
Atlantic coast, D e r m o c h e ly s c o r ia c e a , 209 pH and, 354, 356, 357
Atlantic Mexico, phylogeny of nuclear DNA in salt gland fluid, 349
genotypes, 37 Biological status, 16-24
Atol das Rocas, 17, 18 Biophysical constraints, on thermal biology,
ATP levels, 289 298-302, 311
A u r e lia a u r ita , 209 Birds, salt glands, 348, 349
Australia, 19, 20, 21. S e e a ls o Great Barrier Birth rate, 261
Reef Biting, during mating, 57, 322, 377
C a r e tta c a r e tta , 42, 43, 212 Black turtle. S e e C h e Io n ia a g a s s iz i
C h e io n ia m y d a s , 210, 211 Bladder, 353
D e r m o c h e ly s c o r ia c e a , 209 BLL (Bernardelli-Lewis-Leslie matrix), 263, 264
E r e tm o c h e ly s im b r ic a ta , 152 Blood. S e e Plasma
Heron Island, 38, 200, 210, 267, 333 Blood flow, 282, 302, 311
N a ta to r d e p r e s s u s , 208, 259 Blood gases, 284
nesting locations, 38 Blood lactate, 192, 284, 288, 354, 355
Queensland, 17, 19, 20, 201 Blood oxygen affinity, 286, 287, 288
Raine Island, 61, 200 Blood oxygen transport, 284-288
turtle diet, 201, 203 Boat strikes, 377, 378, 392
Autografts, 237, 253 Body form, 171-172
Avascular necrosis, 281 Body pit, 301
AVT (arginine vasotoci), 327, 328, 335 Body size. S e e Size
Azores, 142, 203 Body temperature
regulation of, 297
size and, 301, 304, 309
B
vx. environmental temperature, 299-300,
Back calculation function (BCF), 240-241, 306
249 Bonaparte, Prince Canino, 6
Bacterial diseases, 366-369 Bones. S e e Skeletochronology; Skeleton
Bahamas, 151 Bony rings (annuli), 238-239
Baja California, C a r e tta c a r e tta , 42, 43, 126 Boreal coastal habitat, 152
Barnacles, as parasites, 375 Brazil
Basking. S e e Surface basking E r e tm o c h e ly s im b r ic a ta , 44-45
Bastards, 36-37 feeding grounds in, 126
Batagurids, 2 southern, 20
BCF (back calculation function), 240-241, 249 turtle diet, 202
Beach armoring, 389 Breast stroke. S e e Powerstroke
Beach cleaning, 390, 395 Breath-hold dives, 284
Beach crawl. S e e Hatchling emergence Breathing frequency, 279-280
Beach driving, 390 Brightness, 111, 114, 115-116, 117
Beaches. S e e Habitat; Nesting beaches Bromine, 344, 349, 350
Beach jumping, 237 Bronchopneumonia (BP), 367
Beach nourishment, 389-390 Brongiart, 6
Bees. 130 Brumation. 148, 149, 152
Index 413
Buoyancy hematology and blood chemistry values, 381

during a dive, 282 hibernation, possible, 333, 355
loss by cold stunning, 308, 355 honnones, 318, 321, 324-326, 327, 329-331
loss from ingestion of marine debris, 219, 398 incubation temperature V5. incubation time, 86
B u s y c o n , 204 later juvenile development, 140, 142
Bycatch excluder devices, 399 lung structure and function, 280, 281, 282
metabolism and stress, 333, 334
migration, 108, 140-142
molecular systematics, 31, 36-37
C . s e r p e n tin a , 32 natal homing, 55
C a lc a r ic h e ly s , 8 nest environment, 67, 99
Calcium nesting behavior, 60, 61, 62, 327, 329
metabolic bone disease and, 365-366 orientation
in plasma, 344, 381 to light, 114, 115, 117, 123, 130
in salt gland fluid, 349, 350 to magnetic field, 120-121, 122, 123, 130,
in seawater, 344 142
in urine, 350 to the ocean, 118, 123, 130
C a llin e c te s s a p id u s , 149, 207, 211 in the open ocean, 124, 125
C a llin e c te s s im ilu s , 207 oxygen transport system, 283, 284, 286,
C a n c e r ir r o r a tiis , 204, 207, 211 287-288
Canonical discriminant analysis, 15 pH regulation, 354, 355, 356, 357
Captive-reared turtles, 249, 250, 251 pivotal temperature, 88
Captivity, nutritional deficiency during, 364 plasma, salt concentrations of, 344
Captorhinids, 2 population structure, 40-43, 45-46
Capture reproduction, 52, 53, 248
incidental. S e e Shrimp trawls, incidental reproductive characteristics, 55, 56, 59
capture in reproductive potential, 15, 63
methodology, for study, 237 salt gland, 347, 350
Carapace, 171 spectral sensitivity. 111, 112, 113
harvest of, 402 suiwival status, 16, 19-20
thermal biology and the, 303, 304 terrestrial locomotion, 168, 170, 183
Carapace length, 56, 147, 149-150 thermal biology, 304, 306, 307, 309
length frequency analysis, 251-253 C a r e tta c o la , 376
straight v.s’. curved, 241, 259 Carettidae, 13, 14
Carbohydrates, 210-211, 218 Carettini, 14, 33, 37
Carbon dioxide Carettochelyidae, 3
accumulated, 279, 354 Carettochelyids, 6
blood pH and, 356, 357 C a r e tto c h e ly s , 166
Cardiac output, 302 Caribbean, 20
Caretta C h e lo n ia m y d a s , 210, 259
cladistic placement, 8 E r e tm o c h e ly s im b r ic a ta , 43, 205
classification, 13, 15, 34 turtle diet, 202, 205
gait, 59 Caribbean monk seal, 16
C a r e tta c a r e tta , 12-13 C a r y o s p o r a c h e lo n ia e , 376
aquatic locomotion, 172, 177, 180 C a ta p le u r a a r k a n sa w , 11
hatchlings, 181, 182, 183, 184, 185, 186, C a to s ty lu s m o s a ic u s , 201, 209
189, 190 C a u le r p a , 210
subadults and adults, 183, 186, 187, 191, C a v o lin a lo n g ir o s tr is , 206
192 Cayman Island Turtle Farm, 37, 315, 321, 324,
distribution, 20, 40-43, 54 333, 365
diving, 278, 279, 288, 289 Celestial light, 113, 115, 116, 117
embryonic development, 90, 92, 98 Central American land bridge, 34
foraging ecology, 202-204 Cephalopods, 5
habitat, 54, 140, 142 C h a e to m o r p h a a e r e a , 201
health problems, 365, 366, 368, 376 C h e lo d in o r u g o s a , 2
C h e lo n ia , 12 survival status, 16, 17-18

current classification, 13-16 teiTestrial locomotion, 168, 169, 183
distribution, 35 thermal biology, 304, 306, 307, 308, 311
gait, 59 C h e lo n ia m y d a s c a n in e g r a , 14, 15
historical classification, 5 C h e lo n ia s is m o n d a i, 11
Chelonia agassizi Chelonii, 5, 7
classification, 14, 15, 18, 45 Cheloniidae
distribution, 4, 18-19 cladistic placement, 8
hibernation, 290, 333 classification, 6, 13-16
hormones, 325, 326 description, 9
molecular systematics, 32, 35-36 diversity, 11
survival status, 13, 15 evolution, 3 ^ , 8, 11, 33, 37
Cheloniadae, 6 reproduction, 52
C h e lo n ia d e p r e s s a , 14, 33. S e e a ls o N a ta to r Cheloniids
d ep ressu s affinities of recent species, 13-16
11
“C helonia^' g w in n e r i, cladistic placement, 7, 8
Chelonia mydas generic definitions of extant, 11-13
aquatic locomotion, 172, 175, 178, 179, 180, locomotion pattern, 166, 168
183 skull features, 10
hatchlings, 181, 183, 184, 185, 189, 190 survival status of, 16-24
subadults and adults, 183, 186, 187, 190, swimming by, 186-188
191, 192 Cheloniinae, classification, 14
classification, 14, 15 Chelonini, 14, 33, 37
consumption of, 19 C h e lo n is , 8
courtship and mating, 57-58 C h e lo s p h a r g is , 8
diet selection, 211 Chelydrids, 10
distribution, 14, 15, 17-18, 35, 38, 39, 54 Chemical imprinting hypothesis, 126-128
diving, 278, 279, 288 Chemical pollutants, 379
early juvenile nursery habitat, 144-145 Chemosensory cues, 127-128
embryonic development, 97 Chesapeake Bay, 37, 142
foraging ecology, 200-202 C a r e tta c a r e tta , 146, 147
habitat, 54 C h e lo n ia m y d a s , 150
health problems, 369, 371, 375, 376 L e p id o c h e ly s k e m p i, 149, 211
hematology and blood chemistry values, 381 Chlamydial infections, 369
hibernation, possible, 355 Chloride
hoimones, 316, 325 in plasma, 344, 381
later juvenile development, 144-145 in salt gland fluid, 349, 350, 351
lung structure and function, 280, 282 in seawater, 344
metabolism and stress, 333 in urine, 350
migration, 108, 144-145 C h o n d r illa n u c u la , 202, 205, 214-217
molecular systematics, 32, 35-37 Chondroepiphyses, vascularized, 10
natal homing, 55 C h o n d r o s ia , 205
nest environment, 67, 99 C h o n d r o s ia c o lle c tr ix , 205
nesting behavior, 60, 62 Chorioallantoic/amniotic fluid, 317
orientation, 114, 115, 118 Chorioallantoic sac, 89, 99
oxygen transport system, 283, 286, 287, 288 Chorion, 66
pH regulation, 354, 355, 356 C h r y s a o r a h y s o s c e lla , 209
pivotal temperature, 88 C h r y s e m y s p ie tà , 290, 355
plasma, salt concentrations of, 344, 350 CITES, 16,21,44
population structure, 38-^0 C itr o b a c te r , 367, 368
reproductive characteristics, 52, 53, 56, 59 Cladism, 6, 7, 8
reproductive potential, 63, 64, 65 C la d o s p o r iu m , 370
salt gland, 347, 349, 350, 351 C la m y d ia p s itta c i, 369
sex ratios, 321 “Classic” dive response, 288
spectral sensitivity, 111, 112 Classification, historical, 5-7
Index 415
Claws, 322, 323 Convention on International Trade in Endangered

Climate space diagrams, 299 Species of Flora and Fauna
Cloaca, 57, 218 (CITES), 16, 21
Clutches Appendix II, 21
effect of egg position in, 96-97 E r e tm o c h e ly s im b r ic a ta , 44
frequency, 56 Convergence fronts, 146, 394, 397
per reproductive season, 63 Coral reefs, habitat degradation, 221
size, 64-65, 109 Coral Sea, nesting locations, 38
volume, 91 Correction factor protocol, 239, 240
C ly p e a s te r h u m ilis, 208 C o r s o c h e ly s , 10
Coastal development, 389-390 C o r s o c h e ly s h a lin ic h e s , 11
Coastal fishing activities, 401-402 Corticosterone, 317, 318, 333-334, 335
Coccidiosis, 376 C o r y n e b a c te r iu m , 368
C o d iu m , 205 C o s m o c h e ly s , 10
C o e lo th r ix ir r e g u la r is , 2 0 6 Costa Rica, 17, 24, 98
Cohort, sampling issues and, 236, 249, 253 Pacific Costa Rica, 22
Cold stunning, 149, 290, 307-308, 355, 377-378, Tortuguero, 127, 144, 205, 260
379 Cotylosauria, 2
Colombia, 18, 151 Coulombe equation, 101
Color Courtship, 57-58, 322, 325-326, 330-331
of skin, 311 Crab pots, 399
visual, 301 Crawling, 61, 166, 168, 169, 170-171, 310
C o lp o c h e ly s , 15 Creatnine kinase, 381
Comatose turtles, 378, 380-381 Cretaceous Period, 8, 10, 11
Commerce Critical thermal maximum (CTM^^), 309, 311
of carapace, 21, 402 Critical thermal minimum (CTMjft), 309, 311
C h e Ion ia m y d a s , 402 Crocodiles, 3, 153, 286, 299
D e r m o c h e ly s c o r ia c e a , not involved in, C r o c o d y lu s n ilo tic u s , 280
23 Cross-sectional sampling, 236, 246, 247
of eggs, 267, 402, 403 in age-based growth model, 247-249
E r e tm o c h e ly s im b r ic a ta , 44, 402 in size-based growth model, 251-253
forensic identification of meat or shell, 45 Crude oil. S e e Petroleum products
overharvest, 16, 403 “Crutching” movement, 168, 169, 170
of real estate, beach value, 99 Cryptodira, 6
of turtles, 402-403 Cryptodires, 6
Community structure C te n o c h e ly s , 8, 9
effect of habitat degradation on, 220-221 Cuba, 21, 151
of foraging habitat, 212-213 E r e tm o c h e ly s im b r ic a ta , 44, 205, 206, 264
Compaction, 100 turtle diet, 202, 205, 206
Comparative demography, 259-261 C y a n e a c a p illa ta a r c tic a , 209
Comparative life history patterns, 256-261 C y a n e a la m a r c k ii, 209
Compasses, internal, 119-120 Cyclic AMP, 352
Compensatory gain, 220 C y m o d o c e a , 205
Competition, 212, 213, 220 C y m o d o c e a s e r r u la ta , 201
Compliance, 281 Cytochrome B, 36
C o m p s o g n a th u s , 302
Condensation of water, 299
Conduction, 298 D
Cone of acceptance, 114 Death. S e e Mortality
Cones, 112 Death rates, 235, 236, 261
Conservation, 69-71, 388-389, 403 Decompression sickness, 281
genetic markers and, 44-45 Deep mycoses, 370
nest site management, 70 Dehydration, 346, 353
species designation and, 35 D e in o n y c h u s , 302
Convection, 97, 298, 299, 303, 306 Demographic sampling design, 235-236
Demographic unit, 235 D e s m e m y s b e r te ls m a n n i, 1

Demography, comparative, 259-261 Deterministic cohort models, 262-267
“Demostat,” 16-17 203
D ia c r ia tr is p in o s a ,
d’Entrecasteaux Reef, 17 Diadectid cotylosaurs, 2
Depradation, 392 Diapsida, 2
Dermal bacterial infections, 368 Diet, 200
Dermal mycotic diseases, 369 foraging ecology, 200-209, 218-221
Dermatemyidae, 2 pelleted, 215-218, 365
Demiochelids, locomotion pattern, 168-169 quality of, 213
Dermochelyidae selection of, 210-211
cladistic placement, 8 Differential equation, 301
classification, 6 Diffusion, 97-98
description, 9 Diffusion capacity, 281
evolution, 3 - 4 , 8, 33 Digesta, passage of, 218
reproduction, 52 Digestibility of food, 214-216
Dermochelyids, 6, 10-11 Digestive processes, 200, 213-218
Dermochelyoidea, 6, 8 Digitata, 6
D e r m o c h e ly s c o r ia c e a , 10 Digits, evolution of, 4
adaptations to cold water, 4 Dinosaurs, 301, 302
aquatic locomotion, 172, 180-181, 182-183 Direction of light, 114-115
hatchlings, 181, 182, 183, 184, 185, 189 Disease. S e e Health problems
postfrenzy swimming, uniqueness of, 181 Dispersal, 190
powerstroke, uniqueness of, 182-183 D is tic h its, 2 0 2
subadults and adults, 154, 184, 188, 191, Distribution. S e e a ls o u n d e r in d iv id u a l g e n e r a
194 antitropical, 20
cladistic placement, 8 phylogeography, 35
consumption of, 24 Diurnal changes
distribution, 23-24, 54 corticosterone secretion, 333
dive records, 278, 279 in temperature, 85
early juvenile nursery habitat, 145-146 Diversity, 4
evolution, 32-33 Diving
foraging ecology, 209 aerobic limits, 289
gait, 59 breath-hold, 284
historical classification, 6 “classic” dive response, 288
later juvenile development, 145-146 depth of, 188, 278, 282
metabolic rate, 310 effects of oil exposure, 394
molecular systematics, 32-33 forced, 288-289, 354, 399
morphology, 32 longest voluntary dive, 278
nest environment, 67, 99 natural patterns, 278-279
nesting behavior, 60, 62 percent time at surface, 277, 278
orientation to magnetic field, 120, 121 pH and, 354-355
orientation to the ocean, 118 physiology of, 277-291
oxygen transport system, 283, 284, 286, 287 D-loop, 31
reproduction, 52, 53 DNA sequence analysis, 30, 332. S e e a ls o
reproductive characteristics, 56, 59 Mitochondrial DNA analysis
reproductive potential, 63, 64, 65 microsatellite DNA analysis, 30, 31, 46
salt gland, 347, 350, 351, 357 minisatellite DNA analysis, 30
sexual differentiation, 317 nuclear, 30-37, 38-39, 40, 46
survival status, 16, 23-24 pace of divergence, 30
terrestrial locomotion, 168-169, 170-171, 183 Dogpaddling stroke, 166,171,174,179,183-184,
thermoregulation, 302, 309-311 187, 193
tidal volume, 280 Dominican Republic, 151
vascularized chondroepiphyses, 10 Drag, from transmitters, 187
D e s m a to c h e ly s , 8 Drag-based locomotion, 174-176, 177, 179
D e s m a to c h e ly s lo w i, 11 Drag forces, 172, 176, 177
Index 417
Dredging, 392 Emmigration rate, 261

Driftnets, 42 Emydidae, 6
Drinking Emydids, 2
hatchling emergence and, 346, 347, 351 E m ys, 5
incidental, 345 Encephalitis, 368-369
Drowning, 289, 378, 398-399 Endangered Species Act of 1973, 16, 402
Dynamic age-structured models, 263-264 Endocrine, 315, 322, 324-331
Dynamical systems, stochastic simulation models, Endothermic capacity, 311
267-268 Energetic adaptations to migratory swimming,
Dynamic pool, 263 189-192, 193
Dynamic stage-structured models, 264-267 Energy budget equation, 298-299, 301
Energy exchange, 298, 300, 301
England, D e r m o c h e ly s c o r ia c e a , 209
ENSO. S e e El Nino-Southern Oscillation
Eastern Pacific, 23, 202 E n ta m e b a in v a d e n s , 375-376
E c io n e m ia , 205 Entanglement, 42, 378, 396, 398
Ecotypic variation, 31 E n te lu r u s a e q u o r e u s , 203
Ecuador E n te ro b a c te r , 367
208
L e p id o c h e ly s o liv á c e a , Environment
turtle diet, 202 ecotypic variation, 31
Education nest, 67-68
on effects of artificial lighting, 391 salt water, ionic regulation and pH balance,
“turtle watch” outings, 390-391 343-357
E d w a r d s ie lla , 367 stochastic fluctuations of the, 257, 267
Egg chamber, 61, 62 temperature, 299-301
Egg laying, 62-63 Environmental chamber, 304
Eggs. S e e a ls o Embryos Environmental sex determination (ESD). S e e
age, 247-248 Temperature-dependent sexual
collectors of, 391 differentiation
diameter, 65 E o c h e lo n e , 8
effects of beach cleaning and beach driving, E o c h e lo n e b r a b a n tic a , 11
390 E o s p h a r g is , 8, 10, 11
effects of oil exposure, 393, 394 Epithecal bones, 11
incubation time, 85 Equatorial Convergence Zones, 146
number of, 52, 64-65 Equilibrium energy budget equation, 301
osmotic balance, 90, 96 Eretmochelinae, 14
in oviduct, 62 E r e tm o c h e ly s , 12
shape and size, 64 cladistic placement, 8
translocation of, 390 classification, 13-14
unhatched, 69 gait, 59
volume, 65 Eretmochelys imbricata
weight, 65 aquatic locomotion, 183, 184, 187-188
Eggshell, 66, 92, 99 distribution, 20-21, 44-45, 54
E im e r ia c a r e tta , 376 diving records, 278, 279
Elasticity measures (sensitivity), 264, 265, 266, early juvenile nursery habitat, 145
267 foraging ecology, 33, 204-206
El Nino-Southern Oscillation (ENSO) habitat, 54
female breeding likelihood and, 267 health problems, 371, 376
greenhouse effect and, 71 incubation temperature v.s\ incubation time,
numbers of nesting turtles and, 200 86
reproductive periodicity and, 56 later juvenile development, 145
Embryos, 66-68 migration, 44, 144
development, 86-87 molecular systematics, 32
nest environment and, 83-103 nest environment, 67
Emergence. S e e Hatchling emergence nesting behavior, 60, 61, 62
pivotal temperature, 88 Feeding ground composition, C a r e tta c a r e tta ,

population structure, 43-45, 46 40-43
reproduction, 52, 53 Females, endocrinology, 324
reproductive characteristics, 56, 59 Fertility, 58, 66
reproductive potential, 63, 64, 65 Fibropapillomatosis, 237, 253, 289, 371-374,
sex ratios, 321 396
spectral sensitivity of, 111, 112 Fidelity. S e e Site fidelity
survival status, 16, 20-21 Fisheries, 219-220
terrestrial locomotion, 168, 170, 183 coastal activities, 401
Er/iis m in is tr o n g y lu s , 205 pelagic, 400-401
Erosion, 99-100 sea turtle, 402-403
E r q u e lin n e s ia , 7, 8, 9 Fishing gear, 398, 399-401
E s c h e r ic h ia c o li, 367 Flatback turtle. S e e N a ta to r d e p r e s s u s
ESD. S e e Temperature-dependent sexual F la v o b a c te r iu m , 367
differentiation Flippers, 171, 172
Esophagus, 345, 346 in dogpaddling, 183-184
Estradiol, 325 in lift-based locomotion, 174
Estrogen, 336 in powerstroke, 175, 176
mating and, 325-326 tagging of. S e e Mark-and-recapture studies
migration and, 324—325 in terrestrial locomotion, 183-184
production, 317 in thermoregulation, 304
E u c h e u m a , 205 Florida, 18, 20
E u n o to s a u r u s , as “missing link,” 2 beach erosion in, 99
Evaporation, 298, 299, 301 C a r e tta c a r e tta , 40-43, 127, 146-147,
Evolution, 2-7 148
locomotor adaptations, 166, 171-173 C h e lo n ia m y d a s , 150-151
macroevolutionary change, 257, 261 chemosensory cues, 127
marine adaptations, 7 D e r m o c h e ly s c o r ia c e a , 152
molecular systematics and, 31-37 L e p id o c h e ly s k e m p i, 149
morphological adaptations to diving, 278, phylogeny of nuclear DNA genotypes,
290-291 37
pace of DNA divergence, 30, 35 turtle diet, 202, 203
Evolutionary ecology, 33 water content of beaches, 95-96
Exploitation. S e e Commerce Florida CuiTent, 142, 143
Explosions, underwater, 392 Flow rate
Eye, lung, eye, and tracheal disease (LETD), cardiac output, 302
370 expiratory, 280
of salt gland fluid, 350-351
Flukes. S e e Parasites
Focal errosive dermatitis (FED), 368
Fabens size-based analogue, 240-241, 249, 252, Follicles, 66, 67
255 Food limitation, 220
False crawls, 61 Food resources
Familial definitions, 8-10 algae, 140, 144, 145, 200-209, 210
Farms, sea turtle, 21,315. S e e a ls o Cayman Island amphipods, 203, 209
Turtle Farm anenomes, 203, 204, 205
Fat reserves, 56 anthropogenic effects on, 218-221
Feces, 200, 204, 206, 212, 263, 345 ascidians, 206, 207
Fecundity, 63, 259, 260 bivalves, 204, 208, 211
Feeding areas bryozoans, 208
adult foraging, 138 coelenterates, 203, 205
degradation of, 220-221 coral, 204, 208, 212
foraging ecology, 200-209, 218-221 crustaceans, 153,202,203,204,205,206-207,
migration to nesting beaches, 108, 126, 208
187 diet selection, 210-211
Index 419
egg composition and, 67 204, 207

F u cu s,
fish and fish eggs, 202,203,204,206,207,208, Fungal infections, 369-370
209
foraging ecology, 200-209, 218-221
gelatinous animals, 145-146, 201, 203, 209,
214 Gait, on land, 59, 166, 167
hydrozoans, 203 Galapagos Islands
insects, 203, 217 C h e lo n ia a g a s s iz i 'ànà m yd a s-W k Q Ío x m s, 15, 18
isopods, 203 L e p id o c h e ly s o liv á c e a , 208
jellyhsh, 152,201,203,204,208,209,214,351 Gas analyzers, 310
molluscs and mollusc egg cases, 201,202,204, Gas exchange, 70, 85, 97-99
206, 207, 208 embryo development and, 68, 84
octopus, 209 hatchling digging activity and, 69
pellets, 215-218, 365 oxygen uptake rates, 85
polychaetes, 202, 204 pulmonary, 279, 280-281
reproductive periodicity and, 56 in renourished beaches, 102-103
salps, 201, 203, 207, 209 Gastrointestinal tract
salt grass, 202 effects of petroleum exposure, 393
scyphozoan medusae, 209 gut microbes, 210, 211, 217
sea cucumber, 205 G e l i d i d l a a c e r o s a , 201
seagrasses, 200, 201, 202, 204, 210, 212 Gene flow, 38-39
digestibility of, 214—217 Genetic assays. S e e a ls o DNA sequence
seahorses, 204, 207 analysis
sea pens, 204, 208 evolutionary tree based on, 32, 36, 37
sea turtles as, 403 paucity of, in sea turtles, 30
sea urchins, 205, 208 population genetics, 29-46
siphonophores, 208, 209 Genetic markers, 235
sipunculids, 207 conservation and, 44-45
snails, 201, 202, 203 maternal, 38
sponges, 152, 201, 204-206, 212 nuclear, 31
digestibility of, 214—217 G e o d ia , 205
squid and squid eggs, 201, 203, 205 G e o d ia g ib b e r o s a , 205
tunicates, 208, 209 G e o d ia n e p tu n i, 205
Foraging ecology, 200-209, 218-221. S e e a ls o Georgia, C a r e tta c a r e tta , 40, 41, 42, 146,
Diet, selection of 263
Foraging habitat, community structure of, GFR (glomerular filtration rate), 353
212-213 GH (growth hormone), 332
Foraging migrations, 146, 148 G ig a r tin a , 202
Foraging site fidelity, 146, 148, 151 Gill nets, 399, 400, 401
Forced submergence, 288-289, 354, 399 Global position, 125, 129, 130
Forensic identification, 44 Global warming, 70-71
Form vision, 115 Glomerular filtration rate (GFR), 353
Fomicata, 5 Glucose, 219, 349, 381
Fractional amplitude, 302 Glycolysis, 192, 193, 288, 290
France, D e r m o c h e ly s c o r ia c e a , 209 G ly p to c h e lo n e s u y c k e r b u y k i, 11
Fraser-Lee back calculation function. S e e Back Gompertz function, 242, 249, 251, 254
calculation function Gonadotropin, 321, 322, 326
French Frigate Shoals, Hawaii, 187 Goodness-of-fit, 246
French Guiana, 24 G r a c ila r ia , 206
Frenzy period, 109, 174 Grand Cayman Island, 315. S e e a ls o Cayman
duration of, 181 Island Turtle Fann
oxygen consumption during, 189, 190, Granulomatous meningoencephalitis (GME),
191 368
swimming characteristics during, 181-185 Gray-patch disease, 371
FSH. 331-332, 336 Grazers, 212
Great Barrier Reef inter-nesting, 138

C a r e tta c a r e tta , 126, 127, 266, 267-268 later juvenile development, 144, 146-153
C h e lo n ia m y d a s , 126, 127, 152, 251, 255,259, loss of, 99-103, 220-221, 388-392
267, 268 protection of. S e e Conservation
chemosensory cues of, 127 Haiti, 151
N a ta to r d e p r e s s u s , 150 H a lia e e tu s le u c o g a s te r , 208
nesting locations, 38, 61, 126, 200 H a lim e d a in c r a s s a ta , 206
turtle diet, 210 H a lo d u le p in if o lia , 201
turtle thyroid study, 333 H a lo d u le u n in e rv is , 201
Greece, 20 H a lo d u le w r ig h tii, 202
Greenhouse effect, 70-71 H a lo p h ila e n g e lm a n n i, 202
Green turtle. S e e C h e lo n ia m y d a s H a lo p h ila o v a Us, 201
Green turtle fibropapillomatosis (GTFP), H a lo p h ila o v a ta , 201
371-373 H a lo p h ila s p in u lo s a , 201
Growth H a p lo tr e m a , 376
of captive-reared turtles, 249, 250 Hardening of beaches, 100
hard part V5. somatic, 238 Harvest of turtles. S e e Commerce
measurement of, 200, 236, 241, 247 Hatching success, 58, 66, 69
methods, 241-246 Hatchling emergence, 68-69, 109-110
modeling, 246-256 artificial lighting and misorientation, 116-117,
polyphasic, 253 388, 391
Growth-at-age studies, 246 behavior, 109
summary o f literature, 243-245 imprinting, 126-128, 317, 318, 336
Growth-at-size studies, 246 sea-finding under natural conditions, 110-116
summary of literature, 243-245 timing of, 109-110
Growth hormone (GH), 332 Hatchlings, 68-69, 139. S e e a ls o Frenzy period;
Growth interval, 254, 259 Postfrenzy period
Growth mark validation, 239 hormones, 317, 318
Growth stanzas. S e e Ontogenetic stages locomotion patterns, 168-169, 174,
Growth zones (annuli), 238-239 183-184
Guajiros, 18 metabolic rate, 310
Gulf of California, 18 migratory swimming, 181-186, 189-190,
Gulf of Carpentaria, 208 192-193
Gulf of Mexico, 23, 142, 149 salt glands, 347
C a r e tta c a r e tta , 202 seawater drinking by, 346, 347, 351
C h e lo n ia m y d a s , 151 suiwival of, and eggs per clutch, 64-65
D e r m o c h e ly s c o r ia c e a , 209 thermal biology, 297, 308
oil production in the, 221 weight, 65
Gulf Stream, 142, 144, 147, 148-149 Hawaiian Islands
Gut microbes, 210, 211, 217 C h e lo n ia m y d a s , 151, 187, 210
Guyana, 23 E r e tm o c h e ly s im b r ic a ta , 206
Gyres, 108, 124, 129, 130 French Frigate Shoals, 187
phylogeny of nuclear DNA genotypes, 37
turtle diet, 201-202, 206
H
Hawksbill turtle. S e e E r e tm o c h e ly s im b r ic a ta
Habitat, 54. S e e a ls o Nesting beaches Hazard function, 238
adult foraging, 138 Head, 171
boreal coastal, 152 “Head-starting” program, 23
breeding, 138 Health problems
coastal fishing activities, 401-402 bacterial diseases, 366-369
co-occupation of feeding, 38 biomedical techniques for, 379-381
early juvenile nursery, 140-146 from chemical pollutants, 379
ecotypic variation, 31 environmental, 377-379
foraging, 212-213 fibropapillomatosis, 237, 253, 289, 371-374,
hormones and changes in, 318-319 396
Index 421
immunosupression due to stress, 364, 370,374, stress and predisposition to disease, 364, 370,
396 374, 396
from ingestion of marine debris, 219, 220, temperature and the, 364, 370
378-379 Immunological data, 3 1
iron deficiency, 366 Imprinting
malnutrition, 365 chemical imprinting hypothesis, 126-128
metabolic bone disease, 365-366 corticosterone and, 317, 318, 336
mycotic diseases, 369-370 In a c h u s , 205
parasites, 364, 368-369, 374, 375-377 Incidental capture. S e e Fisheries; Shrimp
viral diseases, 237, 253, 289, 370-371 trawls
Heat energy exchange, 298, 301 Incidental drinking, 345-346
Heat gain, 311 Inclination compass, 119
Heat lag, 303, 307 Inconel, 237, 254
Heat load, 300, 301, 305 Incubation, 83, 84-85
Heat sink, 303 temperature during, 85-89
Heat storage, 298, 299 water storage in, 89
Heat stress, 299, 304 India, 22, 201
Heat transfer, 298, 299, 306 Indian Ocean, 20
Helminths, 376-377 Infections, 366-369
Hematocrit, 286 Infertility, 66
Hematology, 379-380 Ingestion
Hemoglobin, 287-288 of food, 213-214, 345
H e p a tiis e p h e litic u s , 207 of marine debris, 218-219, 378-379, 396-398
Heron Island, Australia. S e e a ls o Great Barrier Instantaneous energy exchange, 300
Reef Insulation, 311
nesting locations, 38, 200, 267 Intake, 213-214
thyroid study, 333 Interhabitat dispersal rates, 236
turtle diet, 210 Internesting period, 63
Herpesvirus, 370-371 Intracaidiac shunting, 282
Hibernation, 289-290, 333, 336, 355 Ionic regulation, 347-353
H ip p o c a m p u s e re c tu s , 204, 207 Iron deficiency, 366, 375
H o lo th w ia c u h a n a , 205 Islands and archipelagos, 154
Home range, 146, 148, 151 Isoclinics, 129
Homing. S e e Natal beach homing Isodynamics, 129
Hopper dredging, 392 Israel, 95
Hormones, 56, 289, 315-337, 334, 335 Iteroparous life cycle, 52, 259
Human impacts. S e e Anthropogenic effects lUCN (World Conservation Union), 16, 18
Humidity, 85
H y a la e a triclen ta ta , 203
Hybrids, 36-37, 45 j
Hydric conditions, 67-68 J a n th in a , 201, 202, 203, 208
H y p e r ia g a lb a , 209 Japan, 20
H y p e r ia m e d u s a ru m , 202, 203, 209 C a r e tta c a r e tta , 42, 43, 126, 142
Hyperthermic stunning. S e e Cold tortoiseshell imports to, 21
stunning turtle diet, 201
H y p n e a c e r v ic o r n is , 201 Jellyfish. S e e a ls o Food resources, jellyfish
nutritive value of, 214
salt load and, 351, 357
I Jurassic Period, 2
Ichthyosaurs, 3 Juveniles, 139
203
I d o te a m e ta llic a , early nursery habitat, 140-146
Iguanas, marine, 3 hormones, 317-329
Immigration rate, 261 later development habitat, 144, 146-153
Immune system orientation of, 124
effects of oil exposure, 394 thermal biology, 308
K metabolism and stress, 333

migration, 142-144
Kayentachelys molecular systematics, 32, 34
Kayenta fomiation nesting behavior, 60, 62
Kemp’s ridley. S e e L e p id o c h e ly s kem pi pH regulation, 354, 357
Keratin, 345 reproductive characteristics, 52, 53,55,56, 59,
Ketamine hydrochloride, 380 63
Kidneys, 352-353 reproductive potential, 63, 64, 65
K y p h o s u s , 213 survival status, 16, 22-23
terrestrial locomotion, 183
L e p id o c h e ly s o liv á c e o , 52, 53
aquatic locomotion, 183, 186, 188
Laboratory studies, effects of oil pollution, classification, 11, 14, 15
394-395 core body temperature, 302
Lactate, 192, 284, 288, 354, 355 distribution, 54
Lactic dehydrogenase, 381 diving records, 278, 279
Land foraging ecology, 207-208
terrestrial locomotion, 4, 167-171, 183, 184 habitat, 54, 144
terrestrial thermal biology, 59, 297-311 health problems, 371
Laparoscopy, 54, 236, 316, 320, 324 hormones, 326, 327, 328
Latex, 219, 398 later juvenile development, 144
L a tic a u d a c o lu b r in a , 353 metabolism and stress, 333, 334
Latitude, 124, 129 molecular systematics, 32, 34
“Laurasia,” 2 nest environment, 67
L a u r e n c ia , 210 nesting behavior, 60, 61, 62, 303, 327
Layer loss correction protocols, 239-240 orientation, 114, 117
Leatherback turtle. S e e D e r m o c h e ly s c o r ia c e a pivotal temperature, 88
Leeches, 375 reproductive characteristics, 56, 59
Length. S e e Carapace length reproductive potential, 63, 64, 65
Length-at-age back calculation function, 240-241 salt glands, 347, 350
Length frequency analysis, 251-253 survival status, 16, 22
L e p a s a n a tife r a , 203 tenestrial locomotion, 183
L e p id o c h e ly s , 12 LETD (lung, eye, and tracheal disease), 370
aquatic locomotion, 172, 183 Lexis diagram, 236
cladistic placement, 8 LH, 327, 331-332, 336
classification, 13, 14, 15 L ib in ia , 207
gait, 59 L ib in ia e m a r g in a to , 204, 207, 211
relationship between two extant species, 13, L ib in ia s p in o s a , 209
14-15 Libya, 20
terrestrial locomotion, 168, 169, 170, 183 Life expectancy, 154, 395
Lepidochelys kempi Life history traits, 256-259
aquatic locomotion, 183, 184, 186, 188, 192, Life table, 262
193 Lift-based locomotion, 171, 174, 175, 176
classification, 13, 14, 15 Lift forces, 176
diet selection, 211 Light
distribution, 4, 22-23, 52, 54 artificial, and misorientation, 116-117, 388,
diving, 278, 279, 289, 354-355 391,401
effects of oil production on, 221 brightness. 111, 114, 115-116, 117, 307
foraging ecology, 206-207 from celestial sources, 113, 116, 117
growth studies, 239 direction, 114-115
habitat, 54, 142-143 lunar, 115, 117
health problems, 365, 366, 369, 371 orientation to, 111-113, 114, 122, 123
hematology and blood chemistry values, 381 sea-finding and the, 113
hormones, 316, 319, 320, 323, 325 wavelength, 111-113, 306
later juvenile development, 142-143 Light cues, 121, 122, 123
Index 423
Lightsticks, 401 Marine mammals, blood oxygen transport, 286

Light trapping, 116 Mark-and-recapture studies
L im u lu s p o ly p h e m u s , 203 analogues, 242, 246
Linnaeus, 5 compilation of data, 187-188, 254
203, 206
L i tio p a m e la n o s to m a , methods, 236-238
Lizards, 351, 355 of migration patterns, 55
Lobster pots, 399 Markers. S e e Genetic markers
Locomotion. S e e a ls o Aquatic locomotion; Maiking methods, 237
Crawling; Swimming Maruata Bay, Mexico, 18
drag-based, 174-176, 177, 179 Massachusetts, D e r m o c h e ly s c o r iá c e a , 152
lift-based, 171, 174, 175, 176 Mass-specific oxygen consumption. S e e Oxygen
terrestrial, 4, 167-171, 183, 184 consumption
Locomotor adaptations, 166, 171-173 Matapica, 39
Loggerhead turtle. S e e C a r e tta c a r e tta Maternal genetic markers, 38
Logistic funtions, 242 Maternal lineages, 30
Longitudinal sampling, 236, 246 Mating, 57-58, 322, 325-326
in age-based growth model, 249-250 Matrix projection models, 257
in size-based growth model, 253 Maturation, 54, 70, 263
Longlines, 42, 400^01 Mediterranean, 20, 209
Lophochelyniae, 10 M e r e tr ix c a s ta , 214
Loss of tags, 238 Mesirah Island, Oman, 54
“Lost year,” 140, 142 Mesoplastra, 6
Louisiana, L e p id o c h e ly s k e m p i, 149 Metabolic bone disease, 365-366
Low-temperature tolerance, 308 Metabolic heat production, 299, 301, 303, 306
Luminal fluid, 352 Metabolic rate, 309, 310, 311
Lung, eye, and tracheal disease (LETD), 370 Metabolism
Lungs stress and, 332-334
collapse of, 281 thermoregulation and, 309-311
effects of petroleum exposure, 393 M e tc a lfin a h e x a g o n a , 207
structure and function, 279-283 Methodology
Lung ventilation, 356 anesthesia, 380
blood sample for hormones, 334, 335
capture, 237
M
growth, 241-246
M . te m m in c k ii, 32 hematology, 379-380
Madiera, 202, 203 mark-and-recapture studies, 236-238
Magnesium resuscitation, 380-381
in plasma, 344 sampling issues, 235-236
in salt gland fluid, 349, 350 skeletochronology, 238-241
in seawater, 344 Mexican Current, 142, 143
in urine, 350 Mexico, 23. S e e a ls o Pacific Mexico; Rancho
Magnetically coded tags, 237 Nuevo, Tamaulipas, Mexico
Magnetic compass orientation, 120-123 D e r m o c h e ly s c o r iá c e a , 209
Magnetic field intensity, 125, 129 L e p id o c h e ly s k e m p i, 126
Magnetic inclination angle, 124-125, 129 L e p id o c h e ly s o liv á c e a , 2 0 1 , 208, 334
Magnetic map hypothesis, 128-130 turtle diet, 202
Magnetic sensitivity, 130 Microbial communities, in digestive tract, 210,
M a la ch le rn y s, 352 211, 217
Malaysia, 21, 24 M ic r o d ic ty o n , 204
Male-mediated gene flow, 38-40 Microsatellite DNA analysis, 30, 31, 46
Malnutrition, 365 Migration, 108. S e e a ls o Mark-and-recapture
Malta, D e r m o c h e ly s c o r iá c e a , 209 studies; Natal beach homing
Management. S e e Conseiwation C a r e tta c a r e tta , 42, 43, 140-142
Map sense, 129, 130 C h e lo n ia m y d a s , 144-145
Marine debris, 218-219, 378-379, 394, 396-398 to couitship/mating areas, 58
distance, 188 Mosaic epithecal layer, 11

E r e tm o c h e ly s im h r ic a ta , 44, 144 Mozambique, 24
foraging, 146, 148 MULTIFAN, 252
hormones and, 324-325, 330 Musculoskeletal system, 172, 173
L e p id o c h y le s k e m p i, 142-144 Mutilation notches, in tag studies, 237, 253
post-nesting, 187 Mutilation scaring, from mating, 322, 377
thermal biology and, 306 Mycobacterial infections, 367-368
Migratory behavior. S e e Aquatic locomotion; M y c o b a c te r iu m , 367
Frenzy period; Postfrenzy M y c o b a c te r iu m a v iu m , 367
period; Swimming M y c o b a c te r iu m m a r in u m , 367, 368
Mining, sand, 389-390 Mycotic diseases, 369-370
Minisatellite DNA analysis, 30 M y r ia s tr a , 205
Miskito people, 17 M y tilu s , 202
Mitochondrial DNA analysis, 39, 40, 43, 44, 46 M y tilu s e d u lis , 204, 207
affinities and, 14, 15, 30, 31-37
control region (D-loop), 31
female-isolated traits, 38, 124 N
Mixed longitudinal sampling, 236, 246, 247 Nancite site, Costa Rica, 22
in age-based growth model, 250-251 Nasal salt glands, 351
in growth interval studies, 254 N a s s a r iu s , 202, 207
in size-based growth model, 253-256 N a s s a r in s tr iv ita ttu s , 207
Mixed stock assessment, 42, 44, 45 Natal, South Africa, 20, 24
Modeling, 269 Natal beach homing, 38, 52, 55, 107-130
biased estimates, 236 adult navigation and, 125-130
demographic unit and, 235 C a r e tta c a r e tta , 38, 40
environmental temperature, 300-301 chemical cues in recognition, 127-128
growth, 246-256 E r e tm o c h e ly s im h r ic a ta , 43
population dynamics, 262 N a ta to r , 12, 13
stochastic simulation, 267-268 classification, 14
Molecular systematics, evolution and, 31-37 fossil, 11
Mona Island, Puerto Rico gait, 59
E r e tm o c h e ly s im h r ic a ta , 44-45, 205 terrestrial locomotion, 168
turtle diet, 205 N a ta to r d e p r e s s u s , 12
Monel, 237 aquatic locomotion, 184, 188, 192
Monitoring programs, 237 classification, 13, 14
Monofilament line, 398 distribution, 4, 19, 52, 54
Mon Repos Beach, Australia, 19 diving records, 278, 279
Monte Carlo process, 268 foraging ecology, 208
Moonlight, 115, 117 habitat, 54, 144
Morpholine, 128 health problems, 371
Morphology later juvenile development, 144
adaptations to diving, 278 migration, 55
kidneys, 353 molecular systematics, 32, 33-34
salt glands, 347-348 natal homing, 55
Morphometric measurements, 236 nesting behavior, 61, 62
Mortality ontogenetic stages, 246
coastal fishing activities, 401 reproduction, 52, 53
cold stunning, 308 reproductive characteristics, 56, 59
forced dives, 289 reproductive potential, 63, 64, 65
incidental capture, 398 survival status, 16, 19
ingestion of marine debris, 219, 378-379 terrestrial locomotion, 169, 170, 184
in modeling, 263 Natatorini, 34
pelagic fishing gear, 400 N a ta to r te s s e lla tu s , 14
translocation, 390 N a litilo g r a p s u s ( = P la n e s ) m in u tu s, 203
Mortality rates, 235, 236, 261 Navigation
Index 425
chemical imprinting hypothesis, 126-128 C a r e tta c a r e tta , 147, 148

chemosensory cues in, 127 magnetic inclination angle, 125
magnetic map hypothesis, 128-130 North Equatorial CuiTent, 145
natal beach homing and, 125-130 N o to c h e lo n e , 7, 8
nDNA. S e e Nuclear DNA Nuclear DNA analysis, 15, 30-37
Neck, 171 gene flow and, 38-39, 40, 46
Nematodes, 376, 377 Nursery habitat, 140-146
Neonate, 139 Nutrients
N e o s p ir o r c h is , 376 cycling, 212
Neritic nursery areas, 144 dilution, 219
Neritic zone, 142 export, 213
Nest chamber, 61-62, 68, 329 requirements, 365
Nest environment, 67-68 Nutrition. S e e Diet; Digestive processes; Food
Nesting resources; Foraging ecology
colonies V5. females, 21 Nutritional problems, 365-366
dominant period of, 56, 59, 61, 303
hormones and, 326-329
process, 60-63
o
time required for, 59-60 O h e lia d ic h o to m a , 209
Nesting activity, 58-60 Obstructive rhinitis (OR)
Nesting beaches. S e e a ls o Habitat Oceanic habitat, 140, 144, 153
altered, 99-103, 389-390 Oil spill contingency plans, 395
arribada, 22, 59 Oil spills, 219, 221, 393-394, 395
conservation of, 69-70, 388-389 Olfactory cues, 126, 317
depredation by animals, 392 Oligocene Epoch, 11
human presence on, 390-391 O lig o c h e lo n e r u p e lie n s is , 11
interstitial water in, 89-97 Olive ridley. S e e L e p id o c h e ly s o liv á c e a
qualities of, 54, 58-59, 61 Oman, 54
requirements for, 70 C h e lo n ia m y d a s , 210
sand grain size, 91 phylogeny of nuclear DNA genotypes, 37
slope, 110 turtle diet, 201
temperature range of, 88-89, 90, 91 Ontogenetic stages, 138, 139
Nesting behavior, 52, 60, 61, 62 growth models and, 246-247, 250-251, 257,
Nesting populations, V5. feeding aggregates, 259
44-45 locomotion and, 166
Nest site fidelity, 55, 70. S e e a ls o Natal beach Operative environmental temperature, 299-300,
homing 304
Netting, 42, 237, 399, 400, 401 Optimal foraging strategy, 211
Neurophysin, 328 Orbital movements of waves, 119, 120
New Caledonia, 17 Organochlorines, 379
New Guinea, 15 Orientation, 108-109, 123
New York of flipper in lift-based locomotion, 174
C a r e tta c a r e tta , 213 to light, 111-113, 114, 122, 123
L e p id o c h e ly s k e m p i, 207, 211, 213 to magnetic field, 120-123
turtle diet, 204, 207 to the ocean, 117-123
Nicaragua, 17, 22, 151 in open ocean, 124-125
E re tm o c h e ly s im b r ic a ta , 206 to shape, 115-116
turtle diet, 202, 206 Orissa coast, India, 22
Night nesting. S e e Nesting, dominant period of Osmotic pressure, 344
Niobrara Sea, 4 Osmotic regulation, 347-353
N ip h a te s d ig ita lis , 212 Osteopygid, 7, 8
Nonlinear regression, 242, 247, 248 Osteopygidae, 8
Nonparametric growth rate models, 255-256 O s te o p y g is , 9
North Atlantic gyre, 124, 125, 129, 142, 145 OTC (Oxytetracycline), 237, 239
North Carolina O v a lip e s f lo r id a n u s , 207
O v a lip e s o c e lla tu s , 204, 207, 211 Pelagic habitat, 140, 153

Overheating, 311 Pelagic phase
Oviducts, 57, 66 duration of, 248
Oviposition, 58, 62-63, 66, 327, 335 “lost year,” 140, 142
Oxygen affinity, 286, 287, 288 Pelagic specialization, 11
Oxygen binding curves, 286-287 Pelleted diets, digestibility, 215-218
Oxygen consumption, 281, 283 Pelomedusidae, 2
during migratory swimming, 185, 189-192 Pelomedusids, 8
pH regulation through, 356 Penis, 57
Oxygen dissociation curve, 287 Peptide hormones, 316
Oxygen-hemoglobin interaction, 287, 288 Periodicity of reproduction, 55-57, 63, 326
Oxygen stores, 283-284, 285, 286 P e r s e p h o n a m e d ite r r a n e a , 207
Oxygen transport system, 282, 283-288 Peru, turtle diet, 202
Oxygen uptake, 85 Pesticides, 396
Oxytetracycline (OTC), 237, 239 Petroleum products, 392-395
O z o b r a n c h u s b r a n c h ia tu s , 375 exposure routes, 393
O z o b r a n c h u s m a r g o i, 375 ingestion of, 219, 378-379
laboratory studies on effects of, 394-395
salt gland function and, 379
P pH
Pacific Costa Rica, 22 diving and, 354-355
Pacific Mexico, 22, 24 regulation of, 356-357
C h e Ion ia a g a s s iz i, 15 temperature and, 355-356
phylogeny of nuclear DNA genotypes, 37 Phases of growth. S e e Ontogenetic stages
Pacific Ocean 2-Phenylethanol, 128
Eastern, 23 Philippines, 205
South, 201 Philopatry, 55
trans- migration, 42 Phosphorous, 365-366, 381
Pacific ridleys. S e e L e p id o c h e ly s o liv á c e a Photoperiod, 56
Paddles Photopic vision, 112
evolution of, 3-4 Phototropotaxis, 111
in locomotion. S e e Drag-based locomotion Phylogeny, 7-16, 45, 256, 332
Padre Island, Texas, 127-128 Phylogeography, C h e lo n ia , 35
P a e c ilo m y c e s , 370 P h y s a lia p h y s a lis , 202, 204
Paedomorphic heterochrony, 259 P in n a b ic o lo r , 204
P a g u n is p o llic a r is , 204 Pinnata, 6
Paired trawl, 400 Pipping, 109
Panama, 18, 151, 206 PIT (passive integrated transponders), 237
Papillae, esophageal, 345, 346 Pituitary hormone, 331-332, 334
Papillary dermatitis (PD), 368 Pivotal temperature, 86, 88, 90
Parameter substitution, 246 P la c o c h e ly s , 5
Parametric growth rate model, 253 P l a c o s p o n g ia , 205
Parasites, 364, 368-369, 374, 375-377 Plasma
Pareiasaurs, 2 chemistry of, 381
P-450 arom gene, 316 examination of, 379-380
Passage rates, 218 osmotic pressure, 344
Passive integrated transponder (PIT) tags, 237 pH of, 354
Paternal contribution, 30, 45 salt concentrations, 344
Pathogens, 367, 368, 369, 370, 371 Plastics, 219, 378, 396, 397
PCBs (polychlorinated biphenyls), 379 Plastic tags, 237, 253, 254
PCR (polymerase chain reaction), 31 Plastron, 10, 301
Pectoral limbs, 171 softened male, 322, 323
P e la g ia , 208 support by, during crawling, 170, 171
P e la g ia n o c tilu c a , 203, 209 thermal biology and the, 303
Pelagic fishing gear, 400-401 Platform removal, 392
Index 427
Playa Ostional, Costa Rica, 22 Predation, 110, 388

Plesiochelyidae, 7 by turtles, 212
Plesiochelyids, 8 of turtles, 153-154
Pleurodires, 5 Pressure, and the respiratoi7 system, 281
P le u r o n e c te s p la n ip e s , 208 PRL (prolactin), 332, 336
Pleurostemidae, 7 Procolophonoids, 2
P n e u m a to a r th r u s , 10 P r o c o lp o c h e ly s , 11
Pneumonia, 367 P r o c o lp o c h e ly s g r a n d a e v a , 11
Poaching, 388 P r o c o lp o c h e ly s m e lii, 11
Polarity compass, 119 Productivity, 200, 221
Polarized light, 116 P r o g a n o c h e ly s , 2
Pollution. S e e a ls o Entanglement; Fishing gear; Progesterone, 326, 327, 334, 335
Shrimp trawls Prolactin (PRL), 332, 336
chemical pollutants, 379 Prostaglandins, 327
marine debris, 218-219, 378-379, 394, Protein electrophoresis, 31, 39
396-398 Protein hormones, 331
petroleum hydrocarbons, 219, 378-379, P r o te u s , 367, 368
392-395 P r o to s te g a , 8, 10
Polychlorinated biphenyls (PCB), 379, 398 Protostegidae, 8, 10
Polymerase chain reaction (PCR), 31 Protostegids, 6
Polyphasic growth, 253 Protozoans, 375-376
Pompano gill nets, 401 Protraction, 170, 175, 177
Population dynamics, 261-268 P s e p h o p h o r u s , 8, 10
stochastic fluctuations, 234 P s e p h o p h o r u s r u p e lie n s is , 10-11
summary of literature, 258 P s e u d e m y s n e ls o n i, 217
Population genetics, 29-46 P s e u d o m o n a s , 367, 368
Population structure P s e u d o s p h a r g is , 10
C a r e n a c a r e n a , 40-43 “Pseudoturtle,” 5
C h e lo n ia m y d a s , 38-40 P te r o tr a c h e a , 203
E r e tm o c h e ly s im b r ic a ta , 4 3 ^ 5 Puerto Rico
Population survival programs, “head-starting” E r e tm o c h e ly s im b r ic a ta , 44-45, 205,
program, 23 212
P o r p in a , 208 turtle diet, 205
Portugal, magnetic inclination angle, 125 Pulmonary gas exchange, 279, 280-281
P o r tu n u s g ib b e s s ii, 207 Pulmonary perfusion, 282
P o r tu n u s p e la g ic u s , 204 P u p p ig e r u s , 8
P o s id o n ia o c e á n ic a , 201 P u p p ig e r u s c a m p e r i, 11
Postfrenzy period Purse seine, 399
oxygen consumption, 189, 191 Putative sister species, 30
swimming characteristics during, 181-185 P y r o s o m a , 208
Post-hatchling, 139 P y r o s o m a a tla n tic u m , 203
Potassium P y s a lia , 201
blood pH and, 357
in plasma, 344, 381
in salt gland fluid, 349, 350, 351
in seawater, 344 Q i^ 86
in urine, 350 Queensland, Australia, 17, 19, 20, 201
Pouches, 171
Pound nets, 401
Powerstroke, 166, 174, 175, 193
R
mechanisms of, 176-178, 180 Radiation
oxygen consumption during, 189, 192 solar, 298, 303, 306, 311
rates, 184-185, 187 thermal, 298, 299, 300, 303
uniqueness of D e r m o c h e ly s axis, 182-183 ultraviolet, 366, 374
Precopulatory behavior, 57 Radioimmunoassay, 253, 327, 332
Raine Island, Australia, 38, 61, 200. S e e a ls o R h in o c h e ly s , 10

Great Barrier Reef R h iz o s to m a o c to p u s ,209
Ranches, sea turtle, 21, 315. S e e a ls o Cayman R h iz o s to m a p u lm o , 209, 214
Island Turtle Farm R h o d o m e n ia , 202
Rancho Nuevo, Tamaulipas, Mexico, 18, R h ozoph ora, 205
22-23 Ridleys. S e e L e p id o c h e ly s k e m p i; L e p id o c h e ly s
chemosensory cues, 127, 128 o liv á c e a
L e p id o c h e ly s k e m p i, 127, 142 Ritgen, 6
Random sampling, with replacement, 236 r-K theory, 256, 257
Ranking protocol, 240 Rodeo technique, 237, 324, 334
Rear flipper kick, 166, 171, 174, 179, 186, 187, “Rudders,” conversion of hind limbs into, 11
193
R e c liiz ia r o lla n d ia n a , 206
Recruitment
knife-edge, 262, 265 Sabah Turtle Islands, Malaysia, 21
rates, 235, 236 Salinity, 85, 96, 344
size at, 259 S a lm o n e lla , 3 6 1
Regional discrimination, 55 S a lp a , 208
Regression growth protocol, 239, 240 Salt
Regression models, tag loss analysis, 238 gain, 344, 345-346
Remigration interval, 55, 56 loss, 345
Remigration rates, 236 and water exchange, 344-346
Renesting interval, 56 Salt gland fluid, 348-350
Renourishment of beaches, 99, 100, 102 mechanism of tear formation, 351-352
Reproduction, 51-71 secretory rate, 350-351
behavior during, 54 Salt glands, 345, 347-352, 357, 378
generalized life cycle, 53 cold stunning and, 379
stress and, 333-334 effects of oil exposure, 379, 394-395
Reproductive periodicity, 55-57, 63, 326 Salt load, 351
Reproductive potential, 63-64, 322 Sampling, methods, 235-236
Reproductive readiness, 58 San Bias people, 17
Resorption, of skeletal elements, 239-240 Sand mining, 389-390
Resource partitioning, 212 Sargassum
Respiration as food resource, 205, 210
adaptations in, 278, 290-291 as habitat, 140, 144, 145, 202-203
effects of oil exposure, 393, 394 S a r g a s s u m illic if o liu m , 201
mode of, 393 S a r g a s s u m n a ta n s , 204, 207
phases, 278 Satellite tracking, 41
water loss and, 344 of dives, 278, 291
Respiratory system of migration, 108, 126, 127, 187
herpesvirus respiratory disease, 370-371 of pelagic “lost year,” 142
lung structure and function, 279-283 tagging and, 237
pressure and the, 281 Sauropterygians, 5
Restriction fragment length polymorphisms Scales, 259
(RFLP), 30 Scarps, 100, 101
C a r e n a c a r e tta , 40 S c o le c o h a s id iiim , 370
C h e lo n ia , 35, 36, 38 SCUBA, 237
in mtDNA vs. nDNA analysis, 39 Sea-finding, 109-117
Resuscitation, 378, 380-381 Seagrass, 200, 201, 210-211
Retina, 111 Seals, 286
Retraction, 170, 175, 176, 177 Sea snakes, 3, 353, 376
Reynolds number, 172, 174 Seawater
RFLP (restriction fragment length incidental drinking of, 345-346
polymorphisms), 30 osmotic pressure, 344
R h e te c h e ly s , 8 salt concentrations, 344, 349
Index 429
Secondary sex characteristics, 321-322, 324, 335 heat transfer at the surface, 302
Secretory tubules, 347-348, 352 temperature, 299
Selvagens Islands, 203 water loss through, 345
S e m e le , 202 Slope, 110
Sensitivity analysis (elasticity), 264, 265, 266, Social facilitation, 38, 68
267 Sodium
Sensory deprivation experiments, 110 in plasma, 344, 381
Septicemia ulcerative cutaneous disease (SCUD), in salt gland fluid, 345, 349, 350, 351
368 in seawater, 344
Sequential population analysis (SPA), 262-263, in urine, 350
266 Soft-shelled turtles, 3, 6
Sexing technique, 319-321 Solar radiation, 298, 303, 306, 311
Sex ratios, 319-321, 335 S o le n g r a n á is , 204
Sex-specific growth functions, 242, 256 Somatic growth, 238, 241, 255
Sexual characteristics, secondaiy, 321-322, 324, South Africa, 20, 24, 142, 202, 205
335 South Atlantic Current, 124
Sexual differentiation, temperature-dependent, South Carolina
67, 70, 86-88, 102, C a r e tta c a r e tta , 40, 41, 42, 146, 147
316-317,319 D e r m o c h e ly s c o r iá c e a , 214
Sexual maturity, age of, 263 magnetic inclination angle, 125
Seychelles Islands, 201, 205 Southern Equatorial Cunent, 127
Shape parameter, 260-261 South Pacific, turtle diet, 201
Shape vision, 115 South Yemen, 201
Shark nets, 401 S p a r is o m a , 213
Sharks, 153, 154, 302, 305, 349 SPA (sequential population analysis), 262-263,
Shear resistance, 100-101 266
Shrimp trawls Spatial dispersion pattern, 235
incidental capture in, 20, 23, 219, 354-355, Specialization, pelagic, 11
378,398-400 Species diversity, evolutionary trends and, 4
turtle excluder devices on, 399-400 Spectral sensitivity, 111
Shunting, intracardiac, 282 Sperm
Simultaneous gait, 59 formation of, 330, 331
Site fidelity retention of, by females, 58
foraging sites, 146, 148, 151 Sperm duct, 57
nest sites, 55, 70 Sphargidae, 6
Size Spirorchidaisis, 289, 376
blood flow and, 302 S p o r o tr ic h iu m , 370
breeding and, 54, 139 Sri Lanka, L e p id o c h e ly s o liv á c e a , 208
habitat shifting and, 139, 142, 145, 148, Stage-structured models, dynamic, 264-267
152-153 Static age-stmctured models, 262-263
thermoregulation and, 301, 304, 309-311 Status, of survival, 16-24, 153
Size-at-age functions, 242, 246, 249, 251, 254 Steady-state equations, 298
Size-based growth model, 251-256 Steering, 180
Size-specific growth rate, 242, 243-245, 254-255 Stellar’s Sea Cow, 16
Skeletochronology, 238-241, 248-249 Steroochrotes, 5
Skeleton, 171, 172, 173 S tic h o d a c ty la h a d d o n i, 204
metabolic bone disease, 365-366 Stochastic bet-hedging theory, 256
resorption of, 239-240 Stochastic fluctuations
tagging of, 237 environmental, 257, 267
Skin, 259 population abundance, 234, 267
color, 311 simulation models, 267-268
dermal bacterial infections, 368 Stomach, temperature of, 306
dermal mycotic diseases, 369 Stomach lavage, 200
effects of oil exposure, 393, 395 S to m o lo p h u s m e le a g r is , 209, 214
gray-patch disease, 371 Strandings, 394
Streamlined body form, 172 Synchronous flapping of the foreflippers. See

Stress, 317, 321, 330 Powerstroke
corticosteroid secretion and, 333-334, 335 S y r in g o d iu m filifo r m e , 202, 204
metabolism and, 332-334 S y r in g o d iu m is o e tifo liu m , 201
susceptibility to disease and, 364, 370, 371, Systemic mycoses, 370
373-374
S tr o m b u s g ig a s , 212
S tu p e n d e m y s , 8
Sturgeon nets, 401 Tagging protocols, 237-238. S e e a ls o Mark-and-
Subadults, 139 recapture studies
hormones, 317-319 Tags
migration of, 124 loss analysis, 238
migratory swimming, 186-188, magnetically coded, 237
190-192 passive integrated transponders, 237
Subaqueous flight. S e e Powerstroke plastic, 237, 253, 254
S u b e r ite s , 205 titanium, 237, 253, 254, 256
Submarine flight. S e e Powerstroke types of, 237
Submergence of beaches, 99 Tail, 322, 323
S u lc a s c a r is s u lc a ta , 377 Tanker washing, 394
Summer flounder nets, 401 T a p h ro s p h y s, 8
Sun. S e e Light; Solar radiation Tar. S e e Petroleum products
Surface basking, 169, 282, 300, 304-305, Tears, 348-352
306-307 effects of oil exposure, 394-395
Surface times, 185 mechanism of formation, 351-352
Surface-to-volume ratio, 303 TED (turtle excluder devices), 265, 399-400
Surinam Temperature. S e e a ls o Body temperature
C h e lo n ia m y d a s , 17, 37, 38, 39 buoyancy control and, 308, 355
D e r m o c h e ly s c o r iá c e a , 24 effect on hemoglobin, 287
L e p id o c h e ly s o liv á c e a , 22 and the immune system, 364, 370
Survival incubation, embryo development and,
assumptions in analysis, 266 85-89
size-specific, 260 pH and, 355-356
status of turtles, 16-24, 153. S e e a ls o regulation of, 59, 297-311
Endangered Species Act of 1973 renourished beaches and, 102
Swimming. S e e a ls o Aquatic locomotion; for resuscitation, 381
Powerstroke; Thrust salt gland function and, 378
aerobic adaptations to, 189-192, of sand
284-288 basking and, 304, 305
anaerobic adaptations to, 192-193, hatchling emergence and, 68-69, 85,
288-290 109-110
aquatic locomotion, 171-181, 183-184 heat exchange with clutch, 87
backward, 185-186 of the stomach, 306
behavior, 120 tolerances for extreme, 308
D e r m o c h e ly s c o r iá c e a , 154, 183 water exchange and, 91
evolution of, 3-4 Temperature-dependent sexual differentiation, 67,
of hatchlings, 181-186 70, 86-88, 102,
oxygen consumption during, 283 316-317, 319
speeds, 184, 187, 188 T e n o n to s a u r u s , 302
of subadults and adults, 186-188 Terengganu, Malaysia, 24
thermal biology and, 306-309 Terrestrial locomotion, 4, 167-171, 183, 184
“Swing and stance” limb cycle, 168-169, 170, Terrestrial thermal biology, 303-304
171 Testosterone, 317, 318, 319-322, 335, 336
S y llo m u s a e g y p tia c u s , 11 mating, 325-326
Symmetrical gait, 166, 167 migration, 324-325
Synapsida, 2 nesting, 327-328
Index 431
sperm formation, 330, 331 chemosensory cues, 127

Testudinata, 5 205
E r e tm o c h e ly s im b r ic a ta ,
Testudines turtle diet, 205
historical classification, 5 Total absorbed radiation, 301
locomotion, 169 Tourism, 390-391, 402
molecular systematics, 33 Tow times, 399
Testudinidae Toxochelyidae, 8, 9-10
evolution, 2 T o x o c h e ly s , 8, 9
historical classification, 5, 6 Toxychelyids, 10
Testudinids, 5, 6 Trachea, lung, eye, and tracheal disease (LETD),
T estu d o , 5 370
T e th ya a u r a n tia , 205 Trachemys scripta
T eth ya cf. a c tin ia , 205 diet, 217
T eth ya d ip lo d e r m a , 205 musculoskeletal system, 173
Texas pH balance, 355
chemical cues study, 127-128 tidal volume, 280
L e p id o c h e ly s k e m p i, 149, 206-207 urine composition, 353
Thalassemyidae, 7-8 Trade. S e e Commerce
Thalassemyids, 10 Trade winds, 306
T h a la s s e m y s , 8 Traffic, effects of, 388, 390
T h a la s s ia , 201, 202, 220 Trammel nets, 401
T h a la s s ia te stu d in u m , 202, 210, 212, 213, Transient energy budget equation, 301
214-217 Trans-Pacific migration, 42
T h a la s s o d e n d r o n c ilia tu m , 201 Trauma, 377, 378, 396, 398
Thecophora, 6 Trawling. S e e Shrimp trawls
Thermal biology, 59, 297-311 Tremátodos, 376
Thermal inertia, 299, 303, 307, 309 Triassic Period, 3
Thermal lag, 303, 307 Triiodothyronine, 332-333
Thermal radiation, 298, 299, 300, 303 Trinidad, 18, 24
Thermal time constants, 303, 304 Trionicidae, 6
Thermal tolerance range (TTR), 85 Trionychidae, 2, 3
Thermal tolerances, 85, 308 Trionychids, 6
Thermoregulation, 297, 302, 307, Trionychoidea, 6
309-311 TSD. S e e Temperature-dependent sexual
Thoracic squeeze, 281 differentiation
Thrust, 187 TSH, 331
flippers during, 183-184 Tunisia, 20
production of, 174-180 Tur b in a r ia , 210
Thyroid metabolism, 332-333, 336 Turkey, 20
Thyroxin, 332-333 Turtle excluder devices (TED), 265,
Tidal volume, 280 399-400
Time constants, thermal, 303, 304, 307 Tylopods, 5
Time-depth recorders, 306
Time effects, in population dynamics modeling,
235-236, 251, 269 u
Time-light intensity recorders, 306 UCON program, 44
Time-temperature recorders, 306 Ulcerative stomatitis (US), 367
Tissue Ultraviolet radiation, 366, 374
adipose, 310 U lv a , 204, 207
oxygen stores in, 284, 285 United Kingdom. S e e Ascension Island;
peripheral, 305 England
Titanium tags, 237, 253, 254, 256 Universidad de Michoacan, 18
Torres Strait, Australia, 201, 210 Upwelling, 153, 209
Tortuguero, Costa Rica Urea, 349, 353, 381
C h e lo n ia m y d a s , 17, 127, 144, 260 Uric acid, 381
Urine effect of chemical pollutants in, 379

salt concentrations of, 350, 353 effect of cold. S e e Cold stunning
water loss through, 345 embryonic development and, 89-97
U r o m a s ty x a c a n th in u n is , 351 loss, 344-345
Uruguay, D e r m o c h e ly s c o r ia c e a , 209 quality of, and dermal infection, 369
U.S. Department of the Interior, 16, 18 in renourished beaches, 100
salt and, exchange, 344—346
thermal biology in, 303, 306-309
Wat^r content profiles, 95, 96
V a ran u s, 280 Wat^r potential
V elella v e le lla , 203 egg incubation and, 83, 91, 93, 94, 95, 96
Venapuncture, 379 in natural vs. renourished beaches, 102
Venezuela, 18, 151 Water table, 94, 97
Ventilatory patterns, 279-280 Wave cues, 122, 123
V ib rio a lg in o ly tic u s , 367, 368 Wave motion simulator, 119, 120
Viral diseases, 370-371. S e e a ls o Wave refraction zone, 120
Fibropapillomatosis Waves
Virgin Islands detection of, under water, 119-120
D e r m o c h e ly s c o r ia c e a , 152, 209 light reflecting from, 116
E r e tm o c h e ly s im h r ic a ta , 152 orientation to, 117-120, 122, 123
V irg u la ria p r e s b y te s , 204 parameters, 120
Vision. S e e a ls o Light propagation of, 119
color, 301 Weibull growth function, 242, 251,256,260,261,
photopic, M2 267
sea-finding and, 110 Weirs, 401
shape and form, 115-116 Whales, 286
Visual cues, 122, 123 Wind speed, 304
Vitamin Dg_ 366 Winter dormancy, 289-290, 333, 336
Vitelline, 66 World Conservation Union (lUCN), 16, 18
Vitellogenesis, 56
Vocalization, 283
Volatile fatty acids, 217
Voluntary dives, 278 Xanthophobia, 111, 112, 113
von Bertalanffy growth function, 240, 241, 242,
268-269
in age-based growth models, 248, 250, 264
inappropriate use of, 248, 250 Yearlings, 139
in size-based growth models, 252, 254, 255 Yield-per-recruit, 263
stage durations from, 265 Yolk sac, 89, 109
W
Wagler, 6 Zakynthos, Greece, 20
Warm-bloodedness, 309 201
Z o s te r a c a p r ic o n i,
Water Z o s te r a m a r in a , 202, 207

Vol 1

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Vol 1

Uploaded by

Copyright:

Available Formats

The

CRC MARINE SCIENCE SERIES

Michael J. Kennish^ Ph.D.

The biology of sea turtles / edited by Peter L. Lutz and John A.

Visit the CRC Press Web site at www.crcpress.com

© 1997 CRC Press

Karen A. Bjorndal, Ph.D. Kenneth J. Lohmann, Ph.D.

Colin J. Limpus, Ph.D. David W. Owens, Ph.D.

Pamela T. Plotkin, Ph.D. James R. Spotila, Ph.D.

Blair E. Witherington, Ph.D.

Jeanette Wyneken, Ph.D.

1.2 MARINE TURTLES

1.3 HISTORICAL CLASSIFICATIONS OF MARINE

1.4 CLADISTIC PLACEMENT OF MARINE TURTLE

1.5 MARINE TURTLE FAMILIES

FIGURE 1.1 Cladogram “D”, Chelonidea.

1.6 FAMILIAL DEFINITIONS

A family of turtles characterized by an extensively roofed skull with well-developed

peripherals (often reduced to narrow bars in cheloniids). The plastron, primitively

1.7 DERMOCHELYID DIVERSITY

1.8 CHELONIID DIVERSITY

1.9 GENERIC DEFINITIONS OF LIVING CHELONIIDS

1.10 AFFINITIES OF RECENT CHELONIID SPECIES:

1. What subfamilies (or “Tribes”) should be recognized, and what is the

Deraniyagala (1939) recognized a superfamily, the Chelonioidae, comprising

Eretmochelys with Chelonia, and 15 justifying placement of Eretmochelys with

Systematists have been divided in recent years both in whether to recognize

1.11 SURVIVAL STATUS OF LIVING CHELONIIDS

1.11.2 B iologïcal S tatus

The actual, biological status of marine turtles is a topic of considerable complexity.

to a lower position once anthropogenic stresses (directed or incidental) become

1.11.2.1 The Green Turtle, Chelonia mydas

FIGURE 1.2 The green turtle, Chelonia mydas.

1.11.2.2 The Black Turtle, Chelonia agassizii

This species (possibly only a subspecies of C. mydas) is confined to the eastern

1.11.2.3 The Flatback, Natator depressus

FIGUKIC 1.3 "Hie black tuide, Chekmia agassizM.

FIGURE 1.4 The flatback, Natator depressus.

1.11.2.4 The Loggerhead, Caretta caretta

FIGURE 1.5 The loggerhead. Caretta caretta.

The small or medium-size populations in the Mediterranean, nesting primarily

1.11.2.5 The Hawksbill, Eretmochelys imbricata

The black turtle

FIGURE 1.6 The hawkshill, Eretmochelys imbricata.

1.11.2.6 The Olive Ridley^ Lepidochelys oUvacea

FIGURE 1.7 The olive ridley, Lepidochelys olivacea.

1.11.2,7 The Kemp's Ridley, Lepidochelys kempi

A 1947 film exists of an arribada of L. kempi estimated at 40,000 turtles at Rancho

FIGURE 1.8 The Kemp’s ridley, Lepidochelys kempi.

1.11.2.8 The Leatherback, Dermochelys conacea

FIGURE 1.9 The leatherback, Dermochelys coriacea.

Boulenger, G. A., Catalogue of the Chelonians, Rhynchocephalians, and Crocodiles in the

Mlynarski, M., Results of the Polish-Mongolian Palaeontological Expeditions — Part IV.

2.1.1 Technical Considerations

2.2 MOLECULAR SYSTEMATICS AND EVOLUTION

2.2.1 D e r m o c h e l y s c o r ia c e a i an A ncient L ineage?

a suborder (Athecae) to distinguish the shell-less leatherback turtle. This designation

2.2.2 Eretmochelys imbricata: the Origin of an Unusual

Another debate centers on evolutionary relationships of the spongivorous hawksbill

2.2.3 Natator depressus: a Recent Taxonomic Reassignment

2.2.4 Lepidochelys kempii and L, olivácea: Biogeography

2.2.5 Chelonia mydas and C. agassizh When Is a Green

2.2.6 Marine Turtle Hybrids: the Oldest Bastards Known to

Molecular genetic techniques have proven useful in the documentation of several

continental shelf a r e a s . T h e leatherback turtle, D. coriacea, is an excep

The influence of beach renourishment on sea turtle reproduction or egg incuba