Prieto-Marquez 2008 - PHD Dissertation

THE FLORIDA STATE UNIVERSITY
COLLEGE OF ARTS AND SCIENCES
PHYLOGENY AND HISTORICAL BIOGEOGRAPHY OF
HADROSAURID DINOSAURS
By
ALBERT PRIETO-MÁRQUEZ
A Dissertation submitted to the

Department of Biological Science
in partial fulfillment of the
requirements for the degree of
Doctor of Philosophy
Degree Awarded:
Summer Semester, 2008
Copyright © 2008
Albert Prieto-Márquez
All Rights Reserved
The members of the Committee approve the dissertation of Albert Prieto-Marquez
defended on June 30, 2008.
_____________________________
Gregory M. Erickson
Professor Directing Dissertation
_____________________________
William C. Parker
Outside Committee Member
_____________________________
Scott J. Steppan
Committee Member
_____________________________
David L. Swofford
Committee Member
_____________________________
Fredrik Ronquist
Committee Member
Approved:
______________________________________
P. Bryant Chase, Chair, Department of Biological Science
The Office of Graduate Studies has verified and approved the above named committee
members.
ii
To mom, dad, my sister... and little cherry blossom Noa
A mamá, papá, mi hermana... y la pequeña Noa
iii
ACKNOWLEDGEMENTS
In the beginning, there was a typewriter, a kid’s dream and the wind flowing from the
open window, in the late night, merging with a soundscape in suburbial Barcelona, Spain.
Eighteen years later, that dream became realized into a four-year journey throughout 12 countries
from four continents. Data were collected, including 78,000 images documenting thousands of
hadrosaur specimens from 47 institutions. All this, to elucidate how one of the most successful
groups of herbivorous dinosaurs came to be and evolve up to its demise 65 million years ago.
But I could not have done it alone.
First and foremost, I would like to thank my supervisor, Gregory M. Erickson, for being
the most caring and excellent mentor I could have hoped for and a good friend. Not only was his
commitment and support to my education unflagging, but also he has been a constant inspiration
for hard work and scientific rigor. I had also the privilege to be the advisee of committee
members William C. Parker, Scott J. Steppan, Fredrik Ronquist and David L. Swofford. The use
of the R statistical package and the clustering methods in this project was possible thanks to
William C. Parker, who generously spent countless hours making all these mathematical
methodologies easy for me to understand and apply to my research. The knowledge on
phylogenetic methods that Scott J. Steppan, Fredrik Ronquist and David L. Swofford shared with
me was truly invaluable and had a positively strong influence in the outcome of this research.
This project would have never been possible without the generosity of Terry and Mary
Kohler, who provided the necessary funds for the research equipment and the extensive traveling
through a Charlotte and Walter Kohler Charitable Trust grant. I am also indebted to Florida State
University and the Department of Biological Science for covering my expensive tuition of non-
resident international student and supporting my living expenses during almost six years of
graduate studies. I have also received generous support from a grant from The Field Museum for
research in its collections, a Margaret Y. Menzel award from the Department of Biological
Science at Florida State University, and the National Science Foundation: EAR 0207744 and
DBI 0446224 grants presented to Gregory M. Erickson.
I am grateful to many people for access to specimens in their care: Mark Norell and Carl
Mehling (American Museum of Natural History, New York City, USA), Ted Daeschler
(Academy of Natural Sciences of Philadelphia, Pennsylvania, USA), Daniel Brinkman (Yale
iv
Peabody Museum, New Haven, Connecticut, USA), Peter Makovicky (The Field Museum,
Chicago, Illinois, USA), John R. Horner (Museum of the Rockies, Bozeman, Montana, USA),
David B. Weishampel (Center for Functional Anatomy and Evolution, Johns Hopkins
University, Baltimore, USA), Michael K. Brett-Surman (National Museum of Natural History,
Smithsonian Institution, Washington D. C., USA), Pacal Godefroit (Institut Royal des Sciences
Naturelles de Belgique, Brusels and Musée de l’Iguanodon, Bernissart, Belgium), Sandra
Chapman and Paul Barrett (The British Museum, London, UK), John Peel and Solweg Stuenes
(Museum of Evolution, Uppsala, Sweden), Gerdard Plodowski, Olaf Vogel and Claudia
Weißbrod (Naturmuseum Senckenberg, Frankfurt, Germany), Bernard Battail and Philip Taquet
(Museum National d’Histoire Naturelle, Paris, France), Tim Rowe, Jonathan R. Wagner, Lyndon
Murray, Pamela Owen, Wann Langston (Pickle Vertebrate Paleontology Laboratory at the
University of Texas at Austin and Texas Memorial Museum, Austin, Texas, USA), Louis Jacobs,
Bonnie Jacobs, Dale Wrinkler and Diana Vineyard (Shuler Museum of Paleontology, Southern
Methodist University, Dallas, Texas, USA), Andy Heckert, Spencer Lucas, Kevin M. Madalena,
Larry Rinehart and Michael Pierce (New Mexico Museum of Natural History and Science,
Albuquerque, New Mexico, USA), Luis Chiappe and Bruce McLeod (Los Angeles County
Museum of Natural History, Los Angeles, California, USA), Mark Goodwin and Patricia
Holroyd (University of California Museum of Paleontology, Berkeley, California, USA), Kieran
Shepherd, Margaret Feuerback and Robin Culbertson (Canadian Museum of Nature, Ottawa,
Canada), Kenneth Carpenter (Denver Museum of Natural History, Denver, Colorado, USA),
Tony Culver (University of Colorado Museum of Natural History, Boulder, Colorado), Kevin
Seymour, Ian Morrison, David C. Evans, Robert R. Reisz, Brian Iwama (Royal Ontario Museum,
Toronto, Ontario, Canada), Cristiano Dal Sasso and Simone Maganuco (Museo Civico di Storia
Naturale, Milan, Italy), Dan Grigorescu and Zoltan Csiki (Universitatea din Bucureşti,
Bucharest, Romania), Halska Osmolska, Karol Sabath (Instytut Paleobiologii, Polska Akademia
Nauk, Warsaw, Poland), Jolanta Kobalynski, Muzeum Ewolucji, Warsaw, Poland), Xu Xing, Liu
Liping, Liu Zhong Yun, Zhao Qin, Zheng Fang and Jia Cheng Kai (Institute of Vertebrate
Paleontology and Paleoanthropology, Beijing, China), Zhang Zhijun (Geological Museum of
China, Beijing, China), (the director of the China Dinosaur Park, Changzhou, Jiangsu Province,
China), Jin Li-Young, Chen Jun and Li Tao (Museum of Natural History of the Changchun
University of Science and Technology, Changchun, Jilin Province, China), Terry A. Gates,
v
Lindsay Zanno, Erick Lund and Mike Getty (Utah Museum of Natural History, Salt Lake City,
Utah, USA), María del Carmen Perrillat, Claudia Inés Serrano Brañas, Marisol Montellano and
Erandi García (Instituto de Geología de la Universidad Autónoma de México, Mexico D. F.),
René Hernández Rivera (Museo de Geología de la Universidad Autónoma de México), José
Manuel Padilla Gutiérrez, María del Rosario Gómez Núñez and Carlos René Delgado (Museo
del Desierto, Saltillo, Cohauila, México), Régulo Zapata, Raúl Dante and Ruth Zúñiga
(Paleontólogos Aficionados de Sabinas, A. C., Sabinas, Coahuila, México), Belinda Espinosa
and Claudio Arturo de León (Benemérita Escuela Normal Superior de Coahuila, Saltillo,
Coahuila, México), Matthew Lamanna and Amy Henrici (Carnegie Museum of Natural History,
Pittsburg, Pennsylvania, USA), James Lamb (McWane Science Center, Birmingham, Alabama,
USA), James Gardner (Royal Tyrrell Museum of Paleontology, Drumheller, Alberta, Canada),
Philip J. Currie (University of Alberta Laboratory for Vertebrate Paleontology, Edmonton,
Alberta, Canada), Alejandro Kramarz, Guillermo Salinas and Julia Desojo (Museo Argentino de
Ciencias Naturales Bernardino Rivadavia, Buenos Aires, Argentina), Carlos Muñoz (Museo
Provincial Carlos Ameghino, Cipoletti, Argentina), Rodrigo Gaete (Museu de la Conca Dellà,
Isona, Catalunya, Spain) and Angel Galobart (Institut Català de Paleontologia, Sabadell,
Barcelona).
Terry A. Gates provided digital images of Saurolophus angustirostris, Maiasaura
peeblesorum and a new hadrosaurid form the Wahweap Formation under study by him and John
R. Horner. Gregory M. Erickson provided video footage and digital images of S. angustirostris.
David C. Evans provided numerous additional images of lambeosaurine and saurolophine
hadrosaurids from Canada, Kazakhstan and Mongolia. David B. Weishampel provided
photographs of Edmontosaurus annectens and other hadrosaur taxa. James I. Kirlkland provided
digital images of disarticulated facial elements of the type specimen of Kritosaurus navajovius.
I thank Shantanu Joshi and Anuj Srivastava for letting me implement the Analysis of
Planar Shapes Using Geodesic Paths to the morphometric study of hadrosaur bone shapes.
Shantanu Joshi devoted numerous hours to collaborate with me in the search for patterns of
variation of hadrosaur skeletal elements. This was an important stage in my dissertation research.
I also thank Jim Wilgenbusch for helping me understand the command-based version of PAUP.
Paul van der Mark helped me to use the High Performance Computing system of FSU to run
computationally intensive analyses that otherwise, I would have not been able to conduct in the
vi
available time I had. Judy Bowers has been a key reference and a most efficient help in various
administrative issues during my graduate studies. Ann Thistle provided additional assistance on
the use of Microsoft Word software. Austin Mast and Trisha Spears allowed me to use their
Apple computers for running applications requiring Classic OS9. Katja Seltmann provided
assistant for documenting hadrosaur images in morphbank. Diego Pol helped me in
implementing and better understanding the Manhattan Stratigraphic Measure.
I will never forget the hospitality of all those who offered me a place to stay in their
homes: Peter Makovicky (Chicago, USA), Tomasz and Isa Krupa (Warsaw, Poland), Louis and
Bonnie Jacobs (Dallas, USA), Zoltan Csiki (Bucharest, Romania), René Hernández (México D.
F.), Rodrigo Gaete (Isona, Spain), Philip J. Currie (Edmonton, Canada) and María del Rosario
Gómez (Saltillo, México). Likewise, I very much appreciate the two-bedroom apartment that Jin
Li-Young offered me during my visit to Jilin University, Changchun, China. I will always
remember the kindness and guidance in Warsaw of Karol Sabath; the lunches and late afternoon
ice creams with Bernard Battail under the charm of Paris streets; the “asados” and guidance
through Buenos Aires provided by Guillermo Salinas; and the stimulating and insightful
conversations with colleagues Jonathan R. Wagner, David Evans, Terry A. Gates, Merrilee
Guenther, Jason Head and Pascal Godefroit.
Traveling to Utah and Buenos Aires would have not been possible without the help of my
friend Marian Marzin, who found for me affordable airfares and arranged other logistic issues of
those trips. Friends Erika Maticorena, Maria José Cordero, Álvaro Torres Calderón Cisneros
Campoverde, Vida Volkert and Tawainga Katsvairo, who were always there to listen when my
strength faded away during what seemed an interminable struggle.
Finally, but not lastly, I want to thank my parents. My mom, Gracia Márquez Ferrero,
gave me her ball-busting endurance and fighting spirit, without which I had not been able to
endure all those dozens of nights sleeping in airports from all over the world, among other
calamities. My dad, Jaime Prieto Bertrán, gave me his dedicated and methodic approach to work,
without which I would have not been able to complete the analysis, writing and illustration of
this dissertation.
Now, like then, the journey continues...
vii
TABLE OF CONTENTS
List of Tables .................................................................................................................. x

List of Figures............................................................................................................... xx
Abbreviations............................................................................................................... lxx
Abstract.....................................................................................................................lxxiii
INTRODUCTION .......................................................................................................... 1
1. HISTORICAL REVIEW OF STUDIES ON HADROSAURID SYSTEMATICS........ 5
First Discoveries ......................................................................................................... 5

The Pre-Cladistic Period ............................................................................................. 6
The Cladistic Period.................................................................................................. 16
2. PHYLOGENY OF HADROSAURID DINOSAURS ................................................ 43
Abstract .................................................................................................................... 44
Introduction .............................................................................................................. 44
Summary of Previous Studies on Hadrosaurid Systematics ....................................... 46
Materials and Methods.............................................................................................. 53
Results ...................................................................................................................... 66
Discussion ................................................................................................................ 86
Conclusions .............................................................................................................. 91
3. HISTORICAL BIOGEOGRAPHY OF HADROSAURID DINOSAURS................ 122
Abstract .................................................................................................................. 122

Introduction ............................................................................................................ 123
Review of Previous Studies on Hadrosaurid Biogeography ..................................... 124
Methods for the Inference of Ancestral Areas ......................................................... 129
Hadrosaurid Ancestral Distributions ....................................................................... 130
Global Biogeographic History of Hadrosaurids....................................................... 133
Discussion .............................................................................................................. 138
Conclusions ............................................................................................................ 143
4. OSTEOLOGY AND SYSTEMATICS OF THE HADROSAURID DINOSAUR

SECERNOSAURUS KOERNERI (= KRITOSAURUS AUSTRALIS) FROM THE LATE
CRETACEOUS OF ARGENTINA ............................................................................. 153
Abstract .................................................................................................................. 153

Introduction ............................................................................................................ 154
Materials and Methods............................................................................................ 157
viii
Revised Taxonomy of Secernosaurus koerneri and Kritosaurus australis ............... 158
Osteological Description of Secernosaurus koerneri ............................................... 160
Phylogenetic Position of Secernosaurus koerneri.................................................... 169
Historical Biogeography of Secernosaurus koerneri................................................ 171
Discussion .............................................................................................................. 172
Conclusions ............................................................................................................ 175
5. PARARHABDODON ISONENSIS AND TSINTAOSAURUS SPINORHINUS: A NEW

CLADE OF LAMBEOSAURINE HADROSAURIDS FROM EURASIA .................. 195
Abstract .................................................................................................................. 195

Introduction ............................................................................................................ 195
Materials and Methods............................................................................................ 198
Results.................................................................................................................... 200
Discussion .............................................................................................................. 203
Conclusions ............................................................................................................ 204
CONCLUSION........................................................................................................... 207
APPENDIX A. TAXON-CHARACTER STATE MATRIX........................................ 210
APPENDIX B. DENTAL CHARACTERS ................................................................. 255
APPENDIX C. MANDIBULAR CHARACTERS....................................................... 309
APPENDIX D. FACIAL CHARACTERS................................................................... 388
APPENDIX E. NEUROCRANIAL CHARACTERS................................................... 536
APPENDIX F. REGIONAL CRANIAL CHARACTERS ........................................... 576
APPENDIX G. VERTEBRAL AND OTHER AXIAL CHARACTERS...................... 623
APPENDIX H. PECTORAL AND FORELIMB CHARACTERS............................... 638
APPENDIX I. PELVIC CHARACTERS..................................................................... 697
APPENDIX J. PELVIC CHARACTERS .................................................................... 797
APPENDIX K. SYNAPOMORPHIES AND CLADE DEFINITIONS........................ 813
REFERENCES ........................................................................................................... 834
BIOGRAPHICAL SKETCH....................................................................................... 858
ix
LIST OF TABLES
Table 2.1. List of the outgroup taxa used in the phylogenetic analyses of hadrosaurid
dinosaurs, showing the genera and species considered in the present study. In the column of
the original names, when two citations are present the first one refers to the specific name
and the second to the generic name.............................................................................. 110
Table 2.2. List of the ingroup hadrosaurid taxa used in the phylogenetic analyses, showing
the hadrosaurid genera and species considered in the present study. In the column of the
original names, when two citations are present the first one refers to the specific name and
the second to the generic name .................................................................................... 112
Table 2.3. Explanation of the identifiers used to characterize the various models of clusters
that can be used to classify multidimensional data, such as those derived from the analyses of
planar shapes using geodesic paths (GDA; see Fig. 2.1B). Table modified from Fraley and
Raftery (2003)............................................................................................................. 118
Table 2.4. Tree-island profile of the heuristic search in the unweighted parsimony analysis of
hadrosaurid dinosaurs, for the data set excluding Nipponosaurus sachalinensis and
Barsboldia sicinskii..................................................................................................... 119
Table 2.5. Tree-island profile of the heuristic search from the unweighted parsimony analysis
of hadrosaurid dinosaurs, for the data set including Nipponosaurus sachalinensis and
Barsboldia sicinskii..................................................................................................... 120
Table 3.1. Coding of the Sankoff age character used in the calculation of the Manhattan
Stratigraphic Measure ................................................................................................. 150
Table 3.2. Sankoff step matrix of the age character used in the calculation of the Manhattan
Stratigraphic Measure (MSM*) ................................................................................... 152
Table A.1. Character-taxon matrix of the outgroup iguanodontoidean taxa used in the
phylogenetic analysis of the Hadrosauridae. Inapplicable characters were coded as “-“ and
missing data as “?”. Polymorphic characters were coded including in the same cell all states
present in a particular taxon separated by “&”. Ambiguity in the presence of character states
for a given taxa was indicated by “/” ........................................................................... 210
Table A.2. Character-taxon matrix of the non-lambeosaurine hadrosaurid taxa used in the
phylogenetic analysis of the Hadrosauridae. Inapplicable characters were coded as “-“ and
missing data as “?”. Polymorphic characters were coded including in the same cell all states
present in a particular taxon separated by “&”. Ambiguity in the presence of character states
for a given taxa was indicated by “/” ........................................................................... 225
x
Table A.3. Character-taxon matrix of the lambeosaurine taxa used in the phylogenetic
analysis of the Hadrosauridae. Inapplicable characters were coded as “-“ and missing data as
“?”. Polymorphic characters were coded including in the same cell all states present in a
particular taxon separated by “&”. Ambiguity in the presence of character states for a given
taxa was indicated by “/”............................................................................................. 240
Table B.1. Character 1. Values of the number (N) of alveolar positions of the dentaries used
to investigate the distribution and patterns of variation of this character, with the
corresponding character state coding derived from the K-Means clustering ................. 261
Table B.2. Character 2. Values of the number of teeth per cm of dental battery (tooth density;
D) for the dentaries used to investigate the distribution and pattern of variation of this
character. Included are the character state codings derived from K-Means clustering .. 267
Table B.3. Character 3. Values of the number of teeth per alveolus (N) for various
hadrosaurid dentaries, with the corresponding character state coding derived from K-Means
clustering analysis ....................................................................................................... 272
Table B.4. Character 5. Values of the height/width ratio (H/W) of tooth crowns for the
dentaries used to investigate the distribution and pattern of variation of this character, with
the corresponding character state coding derived from clustering using K-Means........ 274
Table B.5. Character 6. Values of the number of ridges (N) in the tooth crowns of several
hadrosaurid dentaries, with the character coding derived from K-Means clustering. A “0.5”
added to an integer value indicates the presence of a fainter secondary ridge............... 279
Table B.6. Character 9. Values of the crown-root angle (A) of the dentary teeth used to
investigate the distribution and patterns of variation of this character, with the corresponding
character state coding derived from K-Means clustering.............................................. 284
Table B.7. Character 7. Number of denticles per cm of crown margin (N) in the dentary teeth
of a sample of hadrosaurid dinosaurs. Ontogenetic stages are approximate based on the
relative size of the exemplars ...................................................................................... 287
Table B.8. Character 16. Values of the number of alveolar positions (N) of the maxillae used
to investigate the distribution and patterns of variation of this character, with the
corresponding character state coding derived from K-Means clustering....................... 292
Table B.9. Character 17. Values of the number of teeth per cm of maxillary dental battery
(tooth density, D) for the maxillae used to investigate the distribution and pattern of variation
of this character. Included is the corresponding character state coding derived from K-Means
Table B.10. Character 18. Comparison of the number of dentary and maxillary alveolar
positions in various hadrosaurid specimens for which both elements are preserved...... 310
xi
Table C.1. Character 24. Values of the ratio (R) between the mediolateral and rostrocaudal
length of the predentary, with the corresponding character state coding derived from K-
Means clustering analysis............................................................................................ 320
Table C.2. Values of the ratio (R) between the depth of the predentary rostral face and the
length of the lateral process, with the corresponding character state coding derived from K-
Table C.3. Character 26. Angle (A) between the predentary rostral face and the lateral
process, with the corresponding character state coding derived from K-Means clustering
analysis ....................................................................................................................... 324
Table C.4. Character 37. Ratio (R) between the length of the proximal slope of the dentary
and that of the dental battery for the specimens used to investigate the distribution and
pattern of variation of this character. Included are also the corresponding character state
coding derived from K-Means cluster analysis ............................................................ 335
Table C.5. Character 38. Angle of deflection (A) of the proximal slope in the dentaries used
to investigate the distribution and pattern of variation in this character. Included are also the
corresponding character coding derived from K-Means clustering analysis ................. 338
Table C.6. Character 39. Angle of deflection (A) of the proximal slope of the predentary joint
surface for the specimens used to investigate the distribution and pattern of variation of this
character. Included is also the corresponding character coding derived from K-Means
Table C.7. Character 40. Angle of deflection (A) of the ventral margin of the rostral region of
the dentary for the specimens used to investigate the distribution and pattern of variation of
this character. Included is the corresponding character coding derived from K-Means
Table C.8. Character 41. Relative position of the inflexion point where the ventral margin of
the dentary originates (ratio R; Fig. C.20). The values are given for the specimens used to
investigate the distribution and pattern of variation of this character. Included is also the
corresponding character coding derived from K-Means clustering analysis ................. 351
Table C.9. Character 42. Values for the ratio (R) of the lingual projection of the dentaries
used to investigate the distribution and pattern of variation of this character. Included is also
the corresponding character coding derived from K-Means clustering analysis............ 356
Table C.10. Character 43. Values of the angle of orientation of the symphysis (S) of the
dentaries used to investigate the distribution and pattern of variation of this character.
Included are also the corresponding character coding derived from K-Means clustering....
.................................................................................................................................... 360
xii
Table C.11. Character 47. Angle of rostral inclination (A) of the coronoid process of the
Included is also the corresponding character coding derived from K-Means clustering ......
.................................................................................................................................... 366
Table C.12. Character 51. Angle (A) between the lateral expansion of the base of the
coronoid process and the lateral side of the dentary for the specimens used to investigate the
distribution and pattern of variation of this character. Included is also the corresponding
character coding derived from K-Means clustering analysis ........................................ 371
Table C.13. Character 61. Angle (A) between the proximal medial side of the caudal process
and the proximal medial side of the surangular in the specimens used to investigate the
distribution and patterns of variation of this character. Included is also the corresponding
character coding derived from K-Means clustering...................................................... 383
Table D.1. Character 65. Ratio (R) between the maximum width of the premaxillary oral
margin and the minimum width of the skull at the post-oral constriction for a sample of
iguanodontoidean ornithopods. Included is the corresponding character coding derived from
K-Means clustering analysis........................................................................................ 409
Table D.2. Character 81. Length/width ratio of the external nares in a sample of
lambeosaurine hadrosaurs. Included is the corresponding character coding derived from K-
Table D.3. Character 95. Angle (A) between the dorsal margin of the maxillary rostroventral
region and the rostral segment of the maxillary tooth row for a sample of iguanodontideans.
Included is the corresponding character coding derived from K-Means clustering analysis
.................................................................................................................................... 447
Table D.4. Character 98. Position of the rostrolateral region of the maxilla (expressed as the
ratio (R) between the distance from its dorsal margin to the rostral end of the maxilla and the
rostrocaudal distance of the element) for a sample of iguanodontoideans. Included is the
Table D.5. Character 102. Number of foramina rostral and ventral to the jugal articulation
(excluding the large dorsal or rostrodorsal foramen) in a sample of iguanodontoidean
ornithopods. Abbreviations: lt = left; rt = right ........................................................... 464
Table D.6. Character 105. Ratio (R) between the length of the ectopterygoid shelf and the
length of the alveolar margin of the maxilla for a sample of iguanodontoidean ornithopods.
Included is the corresponding character coding derived from K-Means clustering analysis
.................................................................................................................................... 471
xiii
Table D.7. Character 106. Angle (A) between the ectopterygoid shelf and the caudal segment
of the tooth row of the maxilla for a sample of iguanodontoidean ornithopods. Included is the
Table D.8. Character 118. Ratio (R) between the dorsoventral depth of the caudoventral
flange and the minimum depth of the caudal constriction for a sample of iguanodontoidean
ornithopods. Included is the corresponding character state coding derived from K-Means
Table D.9. Character 121. Ratio (R) between the minimum depth of the caudal and rostral
constrictions for the jugals sampled to study the variation of this character, with the
Table D.10. Character 124. Angle (A) of curvature of the caudal margin of the quadrate of a
sample of iguanodontoidean ornithopods. Included is also the corresponding character state
coding derived from K-Means clustering analysis ....................................................... 501
Table D.11. Character 125. Ratio (R) between the distance from the mid-length of the notch
to the quadrate head and the total length of the element, for the quadrates sampled to study
the variation of this character Included is the corresponding character state coding derived
from K-Means clustering analysis ............................................................................... 504
Table D.12. Character 126. Angle (A) formed by the rostral margin of the dorsal half of the
quadratojugal notch with the caudal margin of the quadrate for a sample of iguanodontoidean
clustering .................................................................................................................... 508
Table D.13. Character 129. Ratio (R) between the rostrocaudal width of the lateral condyle
and the mediolateral width of the ventral end of the quadrate for a sample of
iguanodontoidean specimens. Included is also the character state coding derived from K-
Means clustering ......................................................................................................... 513
Table D.14. Character 142. Ratio (R) between prequadratic process length and the quadrate
cotylus with of the squamosal, with the corresponding character state coding derived from K-
Table E.1. Character 155. Ratio (R) between the length and the width of the parietal for the
specimens sampled to study the variation of this character, with the corresponding character
state coding derived from K-Means cluster analysis .................................................... 552
Table E.2. Character 162. Angle (A) between the ventral margins of the pterygoid processes
for the braincases sampled to study the variation of this character, with the corresponding
character state coding derived from K-Means clustering analysis ................................ 560
Table E.3. Character 166. Ratio (R) between the minimum mediolateral width of the
basisphenoid constriction and the maximum width of this element across the spheno-occipital
xiv
tubercles, for the braincases sampled to study the variation of this character. Included is the
Table F.1. Character 177. Angle (A) between the long axis of the external naris and the
maxillary tooth row for a sample of iguanodontoidean ornithopods. Included is the
Table F.2. Character 181. Ratio (R) between the length of the narial foramen and the distance
between the rostroventral corner of the premaxilla and the rostroventral margin of the
prefrontal, for the specimens sampled to study the variation of this character. Included is the
Table F.3. Character 203. Ratio (R) between the basal length of the skull (from the caudal
margin of the quadrate to the rostral end of the oral margin of the premaxilla) and its height
(from the ventral margin of the quadrate to the dorsal border of the squamosal) for a sample
of iguanodontoidean taxa. Included is the corresponding character state coding derived from
Table F.4. Character 204. Ratio (R) between the maximum mediolateral width of the skull
across the postorbitals and the mediolateral width across the quadrate cotylae, for the crania
sampled to study the variation of this character. Included is the corresponding character state
coding derived from K-Means clustering..................................................................... 619
Table G.1. Character 209. Ratio (R) between the height of the neural spine and the height of
the centrum of the tallest vertebra from the posterior dorsals or sacral vertebrae from a
sample of iguanodontoidean ornithopods. Included is the corresponding character state
Table G.2. Presence/absence of haemal sulcus and longitudinal ridge on the ventral surface
of the sacral vertebrae in various iguanodontoidean taxa. When a sulcus is present, it is found
along the caudal region of the sacrum (exceptions are a subadult Saurolophus angustirostris,
ZPAL MgD-I 159, where the feature is found cranially, and Charonosaurus jiayinensis, that
showed it running along the whole length of the sacrum)............................................. 632
Table G.3. Character 211. Minumum number (N) of co-ossified vertebrae (sacrodorsal and
sacrocaudal contributions included) found in the sacral region of various iguanodontoidean
ornithopods ................................................................................................................. 634
Table H.1. Ratio (R) between the length of the lateral margins of the scapular and glenoid
facets of the coracoid for a sample of iguanodontoidean ornithopods. Included is also the
corresponding character state coding derived from K-means clustering analysis.......... 647
Table H.2. Character 216. Angle (A) between the lateral margins of the scapular and glenoid
facets of the coracoid for a sample of iguanodontoidean ornithopods. Included is the
xv
Table H.3. Character 218. Ratio (R) between the dorsoventral length and the breadth of the
ventral process of the coracoid for a sample of iguanodontoidean ornithopods. Included is the
Table H.4. Character 221. Ratio (R) between the craniocaudal length and the dorsoventral
depth of the cranial end of the scapula, for a sample of iguanodontoidean ornithopods.
Included is the corresponding character state coding derived from K-Means clustering
analysis ....................................................................................................................... 656
Table H.5. Character 223. Ratio (R) between the maximum dorsoventral width of the
scapular blade and the proximal region of the scapula for a sample of iguanodontoidean
ornithopods. Included is the corresponding character state coding K-Means clustering......
.................................................................................................................................... 660
Table H.6. Character 224. Ratio (R) between the dorsoventral depth of the proximal
constriction and the dorsoventral depth of the cranial end of the scapula, for a sample of
iguanodontoidean ornithopods. Included is the corresponding character state coding derived
from K-Means clustering............................................................................................. 663
Table H.7. Character 227. Ratio (R) between the maximum craniocaudal length of the
craniodorsal region of the scapula (cranial to the acromion process) and the dorsoventral
height from the cranial end of the acromion process to the ventral apex of the glenoidal facet,
for a sample of iguanodontoidean scapulae. Included is the corresponding character state
coding derived from K-Means clustering..................................................................... 671
Table H.8. Character 229. Ratio (R) between the proximodistal length of the deltopectoral
crest and the proximodistal length of the humerus for a sample of iguanodontoidean
clustering .................................................................................................................... 674
Table H.10. Character 231. Angle (A) between the cranial margin of the deltopectoral crest
and the proximalmost margin of the cranial curvature of the humeral shaft, for the specimens
sampled to study the variation of this character Included is the corresponding character state
Table H.11. Character 232. Ratio (R) between the total length and the width of the lateral
surface of the proximal end of the humerus, for the specimens sampled to study the variation
of this character. Included is the corresponding character state coding derived from K-Means
width (at mid-length) of the ulna, for the exemplars used to investigate the distribution and
pattern of variation of this character. Included is the corresponding character state coding
derived from K-Means clustering ................................................................................ 683
xvi
Table H.13. Character 234. Ratio (R) between the length of the ulna and the length of the
humerus for a sample of iguanodontoidean ornithopods. Included is the corresponding
Table H.14. Character 237. Ratio (R) between the total length and the width, at mid-shaft, of
metacarpal III for the specimens sampled to study the variation of this character. Included is
the corresponding character state coding derived from K-Means clustering ................. 690
Table I.1. Character 243. Angle (A) of ventral deflection of the preacetabular process of the
ilium for a sample of iguanodontoidean ornithopods. Included is also the corresponding
Table I.2. Character 244. Ratio (R) between the craniocaudal length of the preacetabular
process and the craniocaudal length of the central plate of the lium for a sample of
iguanodontoidean ornithopods. Included is the corresponding character state coding derived
from K-Means clustering analysis ............................................................................... 712
Table I.3. List of specimens used to conduct a GDA analysis of the geometry of the lateral
profile of the preacetabular process of the ilium. The numbers correspond to the specimens
plotted in Fig. I.3......................................................................................................... 715
Table I.4. Character 245. Ratio (R) between the dorsoventral depth of the proximal region of
the preacetabular process and the dorsoventral depth of the proximal region central plate of
the ilum for a sample of iguanodontoidean ornithopods. Included is the corresponding
character state coding derived from K-Means cluster analysis ..................................... 718
Table I.5. Character 246. Ratio (R) between the dorsoventral depth and the craniocaudal
length of the central iliac plate for a sample of iguanodontoidean ornithopods. Included is the
Table I.6. Character 249. Ratio (R) between the craniocaudal width of the dorsal region of
the supraacetabular process and the craniocaudal length of the central iliac plate for a sample
of iguanodontoidean ornithopods. Included is the character state coding derived from K-
profile of the pubic peduncle of the ilium .................................................................... 733
profile of the ischial peduncle of the ilium................................................................... 738
Table I.9. Character 255. Ratio (R) between the craniocaudal length of the postacetabular
process and the craniocaudal length of the iliac central plate for a sample of
iguanodontoidean ornithopods. Included is the character state coding derived from K-Means
xvii
profile of the postacetabular process of the ilium......................................................... 748
profile of the prepubic process of the pubis ................................................................. 757
Table I.12. Character 271. Length/width ratio (R) of the ischial peduncle of the pubis for a
sample of iguanodontoidean ornithopods. Included is the character state coding derived from
Table I.13. Character 274. Ratio (R) between the craniocaudal length of the pubis (distance
from the acetabular margin to the distal margin of the prepubic process) and the distance
from the dorsal margin of the iliac peduncle to the ventral margin of the proximal postpubic
shaft, for a sample of iguanodontoidean ornithopods. Included is the corresponding character
state coding derived from K-Means clustering analysis ............................................... 770
Table 14. List of specimens used to conduct a GDA analysis of the geometry of the iliac
peduncle of the ischium............................................................................................... 773
Table I.15. Character 276. Orientation (angle A) of the distal articular surface of the iliac
peduncle of the ischium for a sample of iguanodontoidean ornithopods. Included is the
Table I.16. Character 277. Ratio (R) between the proximodistal length and the craniocaudal
width of the distal margin of the iliac peduncle of the ischium for a sample of
iguanodontoidean taxa. Included is the corresponding character state coding derived from K-
means clustering analysis ............................................................................................ 780
Table I.17. Character 279. Orientation of the pubic peduncle of the ischium, measured as the
angle (A) between the ischial shaft and the axis perpendicular to the articular margin of the
pubic peduncle, for a sample of iguanodontoidean taxa. Included is also the corresponding
character state coding derived from K-means clustering analysis................................. 784
Table I.18. Character 282. Ratio (R) between the dorsoventeal thickness of the ischial shaft
and its length for a sample of iguanodontoidean taxa. Included is also the corresponding
character state coding derived from K-means clustering analysis................................. 788
profile of the distal expansion the ischium................................................................... 790
Table I.20. Ratio (R) between the length of the distal “foot” and the craniocaudal length of
the shaft of the ischium for a sample of iguanodontoidean ornithopods. Included is also the
xviii
Table J.1. Character 294. Ratio (R) between the proximodistal length and the mediolateral
width at midshaft of metatarsal III, for the specimens sampled to study the variation of this
character. Included is the character state coding derived from K-Means clustering analysis
.................................................................................................................................... 805
xix
LIST OF FIGURES
Figure 1.1. The first remains of hadrosaurid dinosaurs, described by Joseph Leidy in 1856. A,
dentary crown of Trachodon mirabilis (nomen dubium; ANSP 9260). B, worn fragments of
hadrosaurid teeth and other unidentified parts, collected and described with the type of T.
mirabilis........................................................................................................................ 28
Figure 1.2. The first partial skeleton of a dinosaur found outside Europe; described by Joseph
Leidy in 1858 and named Hadrosaurus foulkii. A, dentary tooth crowns (ANSP 9201) in
lingual view. B, composite of the fragmentary maxilla (ANSP 9203-4) and partial postcranial
skeleton (ANSP 10005), lateral view. C, dentary crowns (ANSP 9201) in mesial or distal
views. Abbreviations: cdl = caudal vertebrae; cor = coracoid; drs = dorsal vertebrae; fm =
femur; hm = humerus; il = ilium; is = ischium; mx = maxilla; mttIV = metatarsal IV;
ph(pb)III1 = pedal phalanx III1; pb = pubis; rd = radius; ul = ulna ulna......................... 29
Figure 1.3. Phylogenetic hypothesis of the relationships among hadrosaurid dinosaurs by Lull
and Wright (1942). Redrawn after the latter................................................................... 30
Figure 1.4. Phylogenetic hypotheses of the evolutionary relationships among hadrosaurid

dinosaurs by A, Ostrom (1961) and B, Morris (1973). Redrawn after Ostrom (1961) and
Morris (1973), respectively ........................................................................................... 31
Figure 1.5. Phylogenetic hypotheses of the evolutionary relationships among hadrosaurid

dinosaurs by A, Hopson (1975) and B, Brett-Surman (1979). Redrawn after Hopson (1975)
and Brett-Surman (1979), respectively .......................................................................... 32
Figure 1.6. First cladograms showing the phylogenetic relationships among hadrosaurid
genera or species by A, Weishampel and Horner (1990) and B, Weishampel et al. (1993).
Redrawn after Weishampel and Horner (1990) and Weishampel et al. (1993), respectively
...........................................................................................................................................33
Figure 1.7. Phylogenetic hypotheses of hadrosaurid relationships by John R. Horner. A,

diphyletic origin of hadrosaurids (Horner, 1990). B, cladogram of Horner (1992). Redrawn
after Horner (1990; 1992), respectively ......................................................................... 34
Figure 1.8. Phylogenetic hypotheses of hadrosuaroid and hadrosaurid relationships by A,

Head (1998), B, Head (2001) and C, Kirkland (1998). Redrawn after Head (1998; 2001) and
Kirkland (1998), respectively ........................................................................................ 35
Figure 1.9. Phylogenetic hypotheses of hadrosuaroid and hadrosaurid relationships by

A, Godefroit et al. (1998) and B, Casanovas et al. (1999). Redrawn after Godefroit et al.
(1998) and Casanovas et al. (1999), respectively ........................................................... 36
xx
Figure 1.10. Phylogenetic hypotheses of hadrosaurid relationships by A, Hu et al. (2001) and
B, Wagner (2001). Redrawn after Hu et al. (2001) and Wagner (2001), respectively ..... 37
Figure 1.11. Phylogenetic hypotheses of hadrosaur relationships by A, Norman (2002) and B,

Hai-lu You et al. (2003). Redrawn after Norman (2002) and Hai-lu You et al. (2003),
respectively ................................................................................................................... 38
Figure 1.12. Hadrosaurid phylogenies by A, Suzuki et al. (2004) and B, Horner et al. (2004).
Redrawn after Suzuki et al. (2004) and Horner et al. (2004), respectively...................... 39
Figure 1.13. Phylogenetic hypotheses of hadrosaurid relationships by A, Prieto-Marquez

(2005), B, Prieto-Marquez et al. (2006a) and C, Prieto-Marquez et al. (2006b).
Abbreviations: H = Hadrosaurinae; L = Lambeosaurinae. Redrawn after Prieto-Marquez
(2005) and Prieto-Marquez et al. (2006a, b), respectively .............................................. 40
Figure 1.14. Hadrosaurid phylogenies by A, Gates and Sampson (2007) and B, Godefroit et
al. (2008). Redrawn after Gates and Sampson (2007) and Godefroit et al. (2008), respectively
...................................................................................................................................... 41
Figure 1.15. Lambeosaurine phylogenies by A, Evans and Reisz (2007) and B, Gates et al.
(2007). Redrawn after Evans and Reisz (2007) and Gates et al. (2007), respectively...... 42
Figure 2.1. Dentaries of two specimens of Edmontosaurus in lateral view, showing evidence
of postdepositional dorsoventral compresion in the form of bending lines (arrows). A, left
dentary of E. regalis, CMN 2289. B, right dentary of CMN 2289. C, right dentary of E.
annectens, cast of AMNH 5730 (= Anatotitan copei)..................................................... 92
Figure 2.2. Signs of postdepositional dorsoventral compression (arrows) in a specimen of

Edmontosaurus annectens, cast of AMNH 5730 (type of Anatotitan copei), lateral view.
A, dorsal region of the left quadrate. B, left postorbital.................................................. 93
Figure 2.3. Right prefrontal of Prosaurolophus maximus (MOR 454-6-24-6-2), part of the
holotype of P. blackfeetensis, showing the presence of a concave surface (arrows)
dorsomedial to the orbital margin, in A, dorsal and. B, lateral views.............................. 94
Figure 2.4. BIC plots resulting from clustering model analyses in Mclust. A, BIC plot two
clustering models of one dimensional dataset. B, BIC plot for various clustering models of
NMMDS derived from a GDA dissimilariy matrix. Identifiers of clustering models: EII =
spherical, equal volume and shape; VII = sperical, variable volume, equal shape; EEI =
diagonal, equal volume and shape; VEI = diagonal, variable volume, equal shape; EVI =
diagonal, equal volume, variable shape; VVI = diagonal, variable volume and shape; EEE =
ellipsoidal, equal volume, shape and orientation; EEV = ellipsoidal, equal volume and shape,
variable orientation; VEV = ellipsoidal, variable volume, equal shape; VVV = ellipsoidal,
variable volume and shape (Fraley and Raftery, 2003.................................................... 95
xxi
Figure 2.5. Strict consensus of the 41 most parsimonious trees resulting from the unweighted
parsimony nalysis of 49 hadrosaur taxa, excluding Barsboldia sicinskii and Nipponosaurus
sachalinensis. At each node, the pair of numbers above a branch represents, from left to
right, a decay index and a bootstrap proportion. Bootstrap proportions lower than 20 are
indicted by a “-”. Only taxa with complete skulls are illustrated with photographs ........ 96
Figure 2.6. Strict consensus of the 1422 most parsimonious trees resulting from the
unweighted parsimony analysis of 51 hadrosaur taxa, showing the position of Barsboldia
sicinskii and Nipponosaurus sachalinensis. At each node, the pair of numbers above a branch
represents, from left to right, a decay index and a bootstrap proportion. Bootstrap proportions
lower than 20 are indicted by a “-” ................................................................................ 97
Figure 2.7. Single most parsimonious tree resulting from the weighted parsimony analysis of
49 hadrosaur taxa, excluding Barsboldia sicinskii and Nipponosaurus sachalinensis. At each
node, the pair of numbers above a branch represents, from left to right, a decay index and a
bootstrap proportion. Bootstrap proportions lower than 20 are indicted by a “-”. Only taxa
with complete skulls are illustrated with photographs.................................................... 98
Figure 2.8. Single most parsimonious cladogram derived from the weighted parsimony
analysis, showing the names of the inferred clades used in this study. The numbers at the
nodes are tags used to refer to the clades defined in Appendix K. Likewise, names between
quotation marks are informal names created to facilitate the reference to specific clades ...
...................................................................................................................................... 99
Figure 2.9. Strict consensus of the 11 most parsimonious trees resulting from the weighted
parsimony analysis of 51 hadrosaur taxa, showing the position of Barsboldia sicinskii and
Nipponosaurus sachalinensis. At each node, the pair of numbers above a branch represents,
from left to right, a decay index and a bootstrap proportion. Bootstrap proportions lower than
20 are indicted by a “-”................................................................................................ 100
Figure 2.10. Consensus of the Bayesian analysis with equal rates of character change
showing the phylogenetic relationships of 49 hadrosaurs, excluding Barsboldia sicinskii and
Nipponosaurus sachalinensis. At each node, the number indicates a posterior probability.
Only the names of taxa with complete skulls are accompanied by a photograph .......... 101
Figure 2.11. Consensus tree resulting from the Bayesian analysis with equal rates of
character change showing the phylogenetic relationships of 51 hadrosaurs and the position of
Barsboldia sicinskii and Nipponosaurus sachalinensis. At each node, the number indicates a
posterior probability.................................................................................................... 102
Figure 2.12. Consensus tree of the Bayesian analysis with variable rates of character change,
showing the phylogenetic relationships of 49 hadrosaur taxa, excluding Barsboldia sicinskii
and Nipponosaurus sachalinensis. At each node, the number indicates a posterior probability.
Only the names of taxa with complete skulls are accompanied by a photograph .......... 103
xxii
Figure 2.13. Consensus tree derived from the Bayesian analysis with variable rates of
character change showing the phylogenetic relationships of 51 hadrosaur taxa, with the
position of Barsboldia sicinskii and Nipponosaurus sachalinensis. The numbers at the nodes
represent posterior probabilities................................................................................... 104
Figure 2.14. Ancestral state reconstructions of various cranial characters. A, expansion of the
premaxilla. B, caudal extension of the circumnarial fossa (rectangle) and presence of crest
(arrow). C, 31 or more tooth positions in the dentary. D, three teeth labiolingually exposed on
the dentary occlusal plane. E, ventral offset of the predentary-premaxilla oral contact. F,
elevation of the rostrodorsal region of the maxilla. G, more than 42 teeth in the dentary. H,
reversal to 31-41 dentary teeth. I, deep and invaginated caudal region of the circumnarial
fossa............................................................................................................................ 105
Figure 2.15. Ancestral state reconstructions of various pectoral and forelimb characters. A,
primitive condition in non-hadrosaurid iguanodontoidean of the dorsally oriented acromion
process (rectangle) of the scapula. B, horizontal acromion process (rectangle).
C, elongation, lateroventral projection and pronounced angulation of the deltopectoral crest
of the humerus. D, curved, rostrodorsally oriented acromion process (rectangle) of the
scapula in lambeosaurine hadrosaurids ........................................................................ 106
Figure 2.16. Ancestral state reconstructions of iliac characters. A, brevis shelf in non-
hadrosaurid iguanodontoideans (ventral view). B, absence of brevis shelf (ventral view). C
and D, brevis shelf resulting from medioventral rotation of the postacetabular process
(ventral view). E, ischial peduncle of the ilium composed of two protrusions. F, rostral shift
of the apex of the supraacetabular process relative to the caudal protrusion of the ischial
peduncle. G, craniocaudal shortening and lateroventral projection of the supraacetabular
process ........................................................................................................................ 107
Figure 2.17. Ancestral state reconstructions of lateral profile of the blade of the prepubic
process of the pubis. A, primitive condition in non-hadrosaurid iguanodontoideans. B,
morphology of the Edmontosaurus prepubic process. C, morphology of the “maiasaur”
prepubic process. D, saurolophine ancestral condition. E, lambeosaurine condition. F,
geometry of he prepubic process in Corythosaurus and Lambeosaurus derived from the
ancestral lambeosaurine condition by elongation or dorsoventral narrowing of the proximal
constriction ................................................................................................................. 108
Figure 2.18. Ancestral state reconstructions of ischial characters. A, cranioventral orientation

of the pubic peduncle in non-hadrosaurid iguanodontoideans. B, slight curvature of the iliac
peduncle in non-hadrosaurid iguanodontoideans. C, ischial “foot” in non-hadrosaurid
iguanodontoideans. D, horizontal orientation of the pubic peduncle in saurolophines. E, slight
cranioventral orientation of the pubic peduncle. F, absence of curvature in the iliac peduncle.
G, abence of ischial “foot”. H, caudodorsally extended and recurved iliac peduncle of
lambeosaurines. I, ischial “foot” of lambeosaurine hadrosaurids.................................. 109
xxiii
Figure 3.1. Time-calibrated phylogram based on the single most parsimonious tree derived
from weighted parsimony analysis of hadrosaurid relationships (Chapter 2, Fig. 2.5),
showing the Fitch parsimony reconstruction of ancestral areas. Grey lines indicate ambiguous
reconstructions. Geochronological ages from Gradstein et al. (2004)........................... 144
Figure 3.2. Time-calibrated phylogram based on the single most parsimonious tree derived
from weighted parsimony analysis of hadrosaurid relationships (Chapter 2, Fig. 2.5),
showing the DIVA reconstruction of ancestral areas. Grey lines indicate ambiguous
reconstructions. Black lines indicate dispersals leading to widespread ancestors and
subsequent vicariance Abbreviations: A = Asia; E = Europe; M = North America; v =
vicariance. Geochronological ages from Gradstein et al. (2004) .................................. 145
Figure 3.3. Paleogeographic reconstruction of continental coastlines during the Coniacian.

The arrows show the main dispersal routes that might have followed hadrosaur ornithopods.
Scisors indicate that vicariance may have ocurred upon dispersal of the most recent common
ancestor of a particular clade. Continental paleocoastlines redrawn after Smith et al. (1994),
Zharkov et al. (1998) and Blakey (2001) ..................................................................... 146
Figure 3.4. Paleogeographic reconstruction of continental coastlines during the Santonian.

The arrows show the main dispersal routes that might have followed primitive hadrosaurids.
Scisors indicate that vicariance may have ocurred upon dispersal of the most recent common
ancestor of a particular clade. Continental paleocoastlines redrawn after Smith et al. (1994),
Figure 3.5. Paleogeographic reconstruction of continental coastlines during the Campanian.

The arrows show the main dispersal routes that might have followed saurolophine and
lambeosaurine hadrosaurids. Scisors indicate that vicariance may have ocurred upon
dispersal of the most recent common ancestor of a particular clade. Continental
paleocoastlines redrawn after Hay et al. 1999), Blakey (2001), Pough et al. (2003) and
Hedges (2006)............................................................................................................. 148
Figure 3.6. Paleogeographic reconstruction of continental coastlines during the

Maastrichtian. The arrows show the main dispersal routes that might have followed
hadrosaurids. Scisors indicate that vicariance may have ocurred upon dispersal of the most
recent common ancestor of a particular clade. Continental paleocoastlines redrawn after
Figure 4.1. Geographical location of the most complete and abundant remains of hadrosaurid
dinosaurs found in South America. The map shows two provinces of southern Argentina, Río
Negro and Chubut. Redrawn after Bonaparte et al. (1984) and Salgado et al. (2007) ... 176
Figure 4.2. Mandibular elements of Secernosaurus koerneri. A and B, predentary, MACN-

RN 142, in rostral and caudal views, respectively. C, dentary, MACN-RN 142B, in medial
view. D, detail of tooth crowns, MACN-RN 142. E and F, dentary rostral fragment, MACN-
RN 142, in medial and rostral views. G, fragment of dentary dental battery, MACN
xxiv
(uncatalogued), in medial view. H and I, surangular, MACN-RN 991, in dorsal and lateral
views, respectively ...................................................................................................... 177
Figure 4.3. Facial elements of Secernosaurus koerneri. A, B and C, right maxilla, MACN-
RN 142, in lateral, medial and ventral views, respectively. D and E, rostral fragment of right
jugal, MACN (uncatalogued), in medial and dorsal views, respectively. F, prefrontal, MACN
(uncatalogued), in lateral view. G, squamosal fragment MACN-RN 991, in lateral view ...
.................................................................................................................................... 178
Figure 4.4. Skull roof and braincase of Secernosaurus koerneri, MACN-RN 142, in dorsal
view ............................................................................................................................ 179
Figure 4.5. Skull roof and braincase of Secernosaurus koerneri, MACN-RN 142. A, rostral,
B, lateral and C, caudal views...................................................................................... 180
Figure 4.6. Skull roof and braincase of Secernosaurus koerneri, MACN-RN 142, in ventral
view ............................................................................................................................ 181
Figure 4.7. Secernosaurus koerneri, fragment of basioccipital and basisphenoid, holotype

FMNH P13423, in A. dorsal, B, caudal, C, lateral and D, ventral views ...................... 182
Figure 4.8. Neurocranial elements of Secernosaurus koerneri. A and B, frontal and parietal,
MACN-RN 142, in dorsal and ventral views, respectively. C, right frontal, MACN
(uncatalogued), in ventral view. D, braincase fragment, MACN-RN 143, in left lateral view.
E and F, supraoccipital, MACN-RN 144, in right lateral and caudodorsal views, respecively
.................................................................................................................................... 183
Figure 4.9. Axial elements of Secernosaurus koerneri. A and B, cervical vertebra, MACN-
RN 826, in lateral and dorsal views, respectively. C and D, dorsal vertebra, MACN-RN 826,
in cranial and lateral views. E, partial sacrum, MACN-RN 2, in lateral view. F and G, caudal
vertebra, MACN-RN 2, in lateral and caudal views. H, sternal plate, MACN-RN 826, in
ventrolateral view........................................................................................................ 184
Figure 4.10. Axial elements of the holotype of Secernosaurus koerneri, FMNH P13423. A
and B, cervical centrum in right lateral and ventral views, respectively. C and D, dorsal
neural spine in caudal and lateral views, respectively. E and F, cranial dorsal vertebra in
caudal and lateral views, respectively. G and H, dorsal centrum in cranial and ventral views,
respectively. I, J and K, cranial dorsal vertebra in right lateral, cranial and dorsal views,
respectively. L and M, caudal centrum in cranial and right lateral views, respectively . 185
Figure 4.11. Pectoral and forelimb elements of Secernosaurus koerneri. A, left scapula,
MACN-RN 142, in lateral view. B, right scapula, MACN-RN 146, in lateral view. C,
coracoid, MACN-RN 2, in lateral view. D, juvenile humerus, MACN-RN 980, in
craniomedial view. E, left ulna, MACN-RN 145, in lateral view. F, left radius, MACN-RN
145, in dorsal view ...................................................................................................... 186
xxv
Figure 4.12. Pectoral and forelimb elements of the holotype specimen of Secernosaurus
koerneri, FMNH P13423. A, left scapula in lateral view. B and C, metacarpal III in dorsal
and proximal views, respectively. D and E, distal end of a right humerus in caudal and lateral
views, respectively. F and G, proximal region of a left ulna in cauda and medial views,
respectively ................................................................................................................. 187
Figure 4.13. Right ilium of the juvenile holotype specimen of Secernosaurus koerneri,
FMNH P13423, in A, lateral, B, dorsal, C, caudal and D, medial views...................... 188
Figure 4.14. Detail views of the holotype ilium of Secernosaurus koerneri, FMNH P13423,
showing the areas with signs of postdepositional deformation (arrows). A, postacetabular
process in lateral view. B, postcetabular process in caudoventral view. C, supraacetabular
process in lateroventral view ....................................................................................... 189
Figure 4.15. Pelvic elements of a juvenile specimen of Secernosaurus koerneri, holotype

FMNH P13423. A, left ilium in lateral view. B and C, right pubis in lateral and ventral views,
respectively. D, lateral view of the shaft of the ischium ............................................... 190
Figure 4.16. Pelvic elements of Secernosaurus koerneri. A, left ilium, MACN-RN 2, in

lateral view. B, right pubis, MACN-RN 2, in lateral view. C, proximal half of right ischium,
MACN-RN 2, in lateral view. D, left pubis, MACN-RN 2, in lateral view................... 191
Figure 4.17. Presence and absence of brevis shelf in the postacetabular process of four
species of hadrosaurs; images shown in ventral view. A, Bactrosaurus johnsoni SBDE 95E.
B, Brachylophosaurus canadensis, MOR 1071-8-2-98-469. C, Secernosaurus koerneri,
MACN-RN 2. D, Hypacrosaurus altispinus, AMNH 5204. Abbreviations; BS = brevis shelf
.................................................................................................................................... 192
Figure 4.18. Absence and presence (rectangles) of the dorsal ridge of the postacetabular
processes of the ilium in three species of hadrosaurids. A, Hypacrosaurus altispinus, AMNH
5204. B, Secernosaurus koerneri, MACN-RN 2. C, Big Bend UTEP hadrosaurid ....... 193
Figure 4.19. Hindlimb elements of Secernosaurus koerneri. A and B, left femur, MACN-RN
2, in cranial and medial views, respectively. C and D, tibia, MACN-RN 997, in lateral and
caudal views, respectively. E and G, left metatarsal II, MACN-RN 145, in medial and dorsal
views, respectively. F and H, left metatarsal III, MACN-RN 145, in medial and dorsal views,
respectively ................................................................................................................. 194
Figure 5.1. “Tsintaosaur” dentaries. A, Pararhabdodon isonensis, IPS-SRA 27, in dorsal and
B, rostral view. C, Tsintaosaurus spinorhinus, IVPP V723 in dorsal and D, rostral view...
.................................................................................................................................... 205
Figure 5.2. Lateral views of the jugal articular surfaces in the maxillae of various hadrosaur
species. A, Bactrosaurus johnsoni (AMNH 6553). B, cf. Telmatosaurus transsylvanicus
xxvi
(FGGUB R1010). C, Corythosaurus intermedius (CMN 8676); D, Brachylophosaurus
canadensis (MOR 1071-8-13-98-554). E, Pararhabdodon isonensis (IPS SRA 22); F,
Tsintaosaurus spinorhinus (IVPP V725) ..................................................................... 206
Figure B.1. Character 1. Boxplot showing the distribution of the number of dentary alveolar
positions in a sample of iguanodontoidean taxa. The inserted smaller graph shows the
Bayesian information criterion (BIC) for the various numbers of clusters in which the data
can be classified .......................................................................................................... 263
Figure B.2. Character 1. Scatterplots of the number of alveolar positions in the dentary
relative to the dental battery length for three species of hadrosaur ornithopods............ 264
relative to the dental battery length for three species of hadrosaurids........................... 265
Figure B.4. Character 1. Scatterplots of the number of alveolar positions in the dentary versus
the dental battery length for four hadrosaurid taxa....................................................... 266
Figure B.5. Character 2. Boxplot showing the distribution of the dentary tooth density in a
sample of iguanodontoidean taxa. The inserted smaller graph shows the Bayesian
information criterion (BIC) for various numbers of clusters in which these data can be
classified ..................................................................................................................... 269
Figure B.6. Character 2. Scatterplots of the dentary tooth density (teeth per cm of dental
battery) relative to the dental battery length for three species of hadrosaurids.............. 270
Figure B.7. Character 2. Scatterplots of the dentary tooth density (teeth per cm of dental
battery) versus the dental battery length for four species of hadrosaurids..................... 271
Figure B.8. Character 3. Boxplot showing the distribution of the number of teeth per alveolus
in dentary of a sample of iguanodontoidean taxa. The inserted smaller graph shows the
Bayesian information criterion (BIC) for various numbers of clusters in which these data can
be classified. The upper photograph shows four teeth per alveolus in Gryposaurus latidens
(AMNH 5465)............................................................................................................. 274
Figure B.9. Character 4. A, boxplot showing the distribution of the number of teeth exposed
in the entary occlusal plane of a sample of iguanodontoidean taxa. Teeth counted at mid-
length of the dental battery. B, occlusal plane of the left dentary of cf. Lambeosaurus sp.
(USNM 10309), where arrows point to individual teeth............................................... 275
Figure B.10. Character 5. Boxplot showing the distribution of the height/width ratio of
dentary tooth crowns in a sample of iguanodontoidean taxa. The inserted smaller graph
shows the Bayesian information criterion (BIC) for various numbers of clusters in which
these data can be classified. The upper photograph shows how this ratio was measured,
exemplified in the dental battery of Edmontosaurus annectens (MOR 003) ................. 278
xxvii
Figure B.11. Character 6. Dentary teeth showing the number and arrangement of ridges
(arrows) on their lingual enameled sides. A, Iguanodon sp. (BMNH R1831). B, Bactrosaurus
johnsoni (AMNH 6553). C, Lambeosaurus lambei (ROM 794). D, Parasaurolophus
cyrtocristatus (FMNH P27393). E, Brachylophosaurus canadensis (FMNH 862). F,
Edmontosaurus annectens (USNM 4808).................................................................... 281
Figure B.12. Character 7. Dentary teeth showing the position of the primary ridge (arrows)
on their lingual enameled sides. A, Hypacrosaurus stebingeri (MOR 549). B, Protohadros
byrdi (SMU 74582). C, Edmontosaurus regalis (CMN 2289). D, Prosaurolophus maximus
(CMN 8894)................................................................................................................ 282
Figure B.13. Character 8. Dentary teeth showing the absence or present of sinuousity
(arrows)in the primary ridge on their lingual enameled sides. A, Saurolophus osborni
(AMNH 5221). B, Brachylophosaurus canadensis (FMNH 862). C, Edmontosaurus anectens
(BMNH R4862). D, cf. Lambeosaurus sp. (USNM 10309) ......................................... 283
Figure B.14. Character 9. Boxplot showing the distribution of the angle between the root and
the crown of the dentary teeth in a sample of iguanodontoidean taxa. The inserted smaller
graph shows the Bayesian information criterion (BIC) for various numbers of clusters in
which the data can be classified. The two photographs show how this angle was measured,
exemplified in A, Brachylophosaurus canadensis (MOR 1071) and B, Iguanodon sp. (IPS
915) ............................................................................................................................ 285
Figure B.15. Character 10. Dentary teeth showing marginal denticles (inside the white
rectangles) on their lingual enameled sides. A, Iguanodon sp. (BMNH R1831). B,
Lambeosaurinae from Dinosaur Provincial Park (?Lambeosaurus; UALVP 11734). C,
Edmontosaurus regalis (CMN 2289)........................................................................... 286
Figure B.16. Character 11. Structure of the marginal denticles of dentary teeth. A,
Lambeosaurinae indeterminate from Dinosaur Provincial Park; the arrows point to two
indentations on a denticle. B, Lophorhothon atopus (AUMP 2295). C, Edmontosaurus cf.
regalis (unrecorded catalog number) from Roth Mason Quarry (South Dakota)........... 288
Figure B.17. Character 12. Comparison of the quantity and/or size of the denticles (white
rectangles) between the rostral and caudal margins of dentary tooth crowns. A, Bactrosaurus
johnsoni (AMNH 6581). B, Hadrosaurus foulkii (nomen dubium, ANSP 9201). C,
Corythosaurus casuarius (ROM 870). D, Lambeosaurinae indeterminate from Dinosaur
Provincial Park (Alberta, Canada) ............................................................................... 289
Figure B.18. Character 13. Mediolateral thickness of the dentary alveolar sulci (arrows).
A, Iguanodon sp. (BMNH R754). B, Brachylophosaurus canadensis (MOR 10717-25-98-
405) ............................................................................................................................ 290
xxviii
Figure B.19. Character 14. Shape of the dentary alveolar sulci (arrows). A, Iguanodon sp.
(BMNH R5764). B, Brachylophosaurus canadensis (MOR 1071-8-15-98-574) .......... 294
Figure B.20. Character 16. Boxplot showing the distribution of the number of maxillary
alveolar positions in a sample of iguanodontoidean taxa. The inserted smaller graph shows
the Bayesian information criterion (BIC) for various numbers of clusters in which the data
can be classified .......................................................................................................... 294
Figure B.21. Character 16. Scatterplots of the number of alveolar positions in the maxilla
relative to the dental battery length for two species of hadrosaur ornithopods.............. 295
relative to the dental battery length for three species of hadrosaurid ornithopods......... 296
Figure B.23. Character 17. Boxplot showing the distribution of the maxillary tooth density in
a sample of iguanodontoidean taxa. The inserted smaller graph shows the Bayesian
information criterion (BIC) for various numbers of clusters in which these data can be
classified ..................................................................................................................... 299
Figure B.24. Character 17. Scatterplots of the maxillary tooth density (teeth per cm of dental
battery) relative to the dental battery length for three species of hadrosaurids .............. 300
Figure B.25. Character 18. Scatterplot of the number of dentary teeth and that of the
maxillary teeth a sample of iguanodontoidean specimens ............................................ 302
Figure B.26. Character 19. Maxillary occlusal planes of various iguanodontoidean

orntihopds, showing the number of funcitonal teeth (arrows) exposed per alveolus. A,
Iguanodon atherfieldensis (BMNH R11521). B, cf. Telmatosaurus (BMNH R4911). C,
Corythosaurus intermedius (CMN 8676)..................................................................... 303
Figure B.27. Character 20. Maxillary tooth crowns showing the primary and, when present,
secondary and auxilliary ridges on their buccal enameled sides. A, Iguanodon atherfieldensis
(BMNH 11521). B, Iguanodon sp. (BMNH R454). C, Brachylophosaurus canadensis (MOR
1071-8-15-98-573). D, Lambeosaurus lambei (CMN 351) .......................................... 304
Figure B.28. Character 21. Maxillary tooth crowns showing the position of the primary ridge
(arrows) on their buccal enameled sides. A, Bactrosaurus johnsoni (juvenile specimen,
AMNH 6390). B, Bactrosaurus johnsoni (cast of SBDE 95/E5). C, Amurosaurus riabinini
(cast of AEHM 1/12)................................................................................................... 305
(arrows) on their buccal enameled sides. A, Equijubus normani (IVPP V12534), with
asymmetrically positioned primary ridge. B, Gilmoreosaurus mongoliensis (AMNH 6551),
with symmetrically positioned primary ridge............................................................... 306
xxix
Figure B.30. Character 22. Maxillary tooth crowns showing the absence (A) or presence (B)
of sinuosity in the primary ridge (arrows). A, Edmontosaurus regalis (CMN 2289). B,
Brachylophosaurus canadensis (CMN 8893) .............................................................. 307
Figure B.31. Character 23. Maxillary tooth crowns showing the size and shape of the
marginal denticles (arrows).A, Telmatosaurus transsylvanicus (BMNH R3388). B,
Gilmoreosaurus mongoliensis (AMNH 6551). C, Edmontosaurus regalis (CMN 2289)....
.................................................................................................................................... 308
Figure C.1. Character 24. Boxplot showing the distribution of the length/width ratio of the
predentary in a sample of iguanodontoidean ornithopods. The inserted smaller graph shows
can be classified. The photograph shows how the length and width of the predentary were
measured, exemplified in Protohadros byrdi (SMU 74582)......................................... 321
Figure C.2. Character 25. A, boxplot showing the distribution of the depth/length ratio of the
predentary in a sample of iguanodontoidean taxa. B, Bayesian information criterion (BIC) for
various numbersof clusters in which the data can be classified. C, Predentary of Gryposaurus
notabilis (CMN 2278) showing how the depth and length of this bone were measured......
.................................................................................................................................... 323
Figure C.3. Character 26. Boxplot showing the distribution of the angle between the rostral
surface of the predentary and the frontal plane of the skull in a sample of iguanodontoidean
ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC)
for various numbers of clusters in which the data can be classified. The photograph shows
how this angle (A) was measured, exemplified in Prosaurolophus maximus (MOR 447-7-27-
7-5) ............................................................................................................................. 326
Figure C.4. Characters 27 and 28. Predentary denticles of several iguanodontoidean

ornothopods. A, Ouranosaurus nigeriensis (cast of GDF 300). B, Brachylophosaurus
canadensis (MOR 1071-7-28-98-299). C, Corythosaurus intermedius (ROM 777). D, B.
canadensis (CMN 8893). E, Bactrosaurus johnsoni (AMNH 6372, juvenile). F,
Edmontosaurus sp. (UCMP unrecorded no., juvenile). G, Equijubus normani (IVPP 12534).
H, Edmontosaurus sp. (UCMP Fld 4.5-83-90, juvenile). I, Protohadros byrdi (SMU 74582).
J, Edmontosaurus sp. (UCMP FldAK 83-V-292, juvenile). K, cf. Corythosaurus ap. (USNM
11893, juvenile). L, Hypacrosaurus altispinus (CMN 8674). M, H. altispinus (CMN 8501).
N, Gryposaurus notabilis (MSNM V345). O, Gryposaurus notabilis (MOR 553S-7-25-8-78).
P, H. stebingeri (MOR 553S). Q, Prosaurolophus maximus (MOR 447-7-27-7-5). R, H.
stebingeri (MOR 553S-7-27-2-9). S, cf. Lambeosaurus sp. (USNM 10309). T, Iguanodon sp.
(BMNH R105). U, Maiasaura peeblesorum (YPM-PU 22405). V, Tsintaosaurus spinorhinus
(IVPP V723). W, Olorotitan ararhensis (cast of AEHM 2/845). X, C. intermedius (CMN
8676). Y, Lambeosaurus lambei (TMP 2005-00-28) ................................................... 327
xxx
Figure C.5. Characters 29-32. Dorsal views of several iguanodontoidean predentaries.
A, Iguanodon sp. (BMNH R105). B, I. atherfieldensis (BMNH R11521, juvenile). C,
Bactrosaurus johnsoni (AMNH 6372, juvenile). D. Gilmoreosaurus mongoliensis (AMNH
6369, juvenile). E. Protohadros byrdi (SMU 74582). F, cf. Lambeosaurus sp. (USNM
10309). G, Ouranosaurus nigeriensis (cast of GDF 300). H, Maiasaura peeblesorum (YPM-
PU 22405). I, Brachylophosaurus canadensis (MOR 1071-7-28-98-299) .................... 328
Figure C.6. Characters 29-32. Dorsal views of several hadrosaurid predentaries. A, B, and C,
Edmontosaurus sp. (UCMP unrcorded no, UCMP Fld 4.5-83-90, and UCMP FldAK 83-V-
292, respectively; juveniles). D, Gryposaurus monumentensis (RAM 6797). E and F,
Prosaurolophus maximus (MOR 447-7-27-7-5 and MOR 553S-7-12-8-5, respectively). G
and H, Gryposaurus cf. notabilis (MOR 553S-7-3-8-87 and 553S-7-25-8-78). I,
Hadrosauridae indeterminate (UCMP V82223) ........................................................... 329
Figure C.7. Characters 29-32. Lambeosaurine predentaries in dorsal view (except I, ventral
view). A and B, Hypacrosaurus stebingeri (MOR 553S). C, H. stebingeri (MOR 553S-7-27-
2-89). D and E, cf. Corythosaurus sp. (USNM 11893 and 16600, respectively; juveniles). F,
Tsintaosaurus spinorhinus (IVPP V725). G and H, H. altispinus (CMN 8674 and 8501,
respectively). I, Olorotitan ararhensis (cast of AEHM 2/845) ..................................... 330
Figure C.8. Character 33. Lateral views of two hadrosaurid predentaries. A,

Brachylophosaurus canadensis (MOR 1071-7-28-98-299), with an arrow pointing to the
lingual margin as the more caudally extended feature in the caudal end of the lateral process.
B, Corythosaurus intermedius
Figure C.8. Character 33. Lateral views of two hadrosaurid predentaries. A,

Brachylophosaurus canadensis (MOR 1071-7-28-98-299), with an arrow pointing to the
lingual margin as the more caudally extended feature in the caudal end of the lateral process.
B, Corythosaurus intermedius (CMN 8676), with an arrow pointing to the lateral margin as
the most caudally extended feature in the caudal end of the lateral process.................. 331
Figure C.9. Character 34. Dorsolingual views of iguanodontoidean predentaries; arrows

indicate the ridge of the median lingual process. A, Iguanodon sp. (AMNH R105). B,
Gilmoreosaurus mongoliensis (AMNH 6369, juvenile). C, Brachylophosaurus canadensis
(MOR 1071-7-28-98-299). D, Hyppacrosaurus stebingeri (MOR-553S-7-27-2-89) ..... 332
Figure C.10. Character 35. Ventral views of two hadrosaurid predentaries; arrows show the
rostral end of the ventral median process. A, Brachylophosaurus canadensis (MOR 794). B,
Prosaurolophus maximus (MOR 553S-7-12-8-5) ........................................................ 333
extent of the indentation of the split of the ventral process. A, Prosaurolophus maximus
(MOR 553S-7-12-8-5). B, Brachylophosaurus canadensis (MOR 794) ....................... 334
xxxi
Figure C.12. Character 37. Boxplot showing the distribution of the ratio between the
proximal slope of the dentary edentulous portion and the length of the dental battery in a
information criterion (BIC) for various numbers of clusters in which the data can be
classified. The photograph shows how the two variables of this raio were measured,
exemplified in Corythosaurus casuarius (CMN 8676)................................................. 337
Figure C.13. Character 38. Boxplot showing the distribution of the angle between the
proximal slope of the dentary edentulous portion and the frontal plane of the skull in a
informaiton criterion (BIC) for various numbers of clusters in which the data can be
exemplified in Lambeosaurus lambei (CMN 8633) ..................................................... 340
Figure C.14. Character 39. Boxplot showing the distribution of the angle formed by the slope
of the rostral region of the dentary that articulates with the predentaryand the frontal plane of
the skull. The inserted smaller graph shows the Bayesian information criterion (BIC) for
various numbers of clusters in which the data can be classified. The photograph shows how
the two variables of this raio were measured, exemplified in Brachylophosaurus canadensis
(MOR 1071)................................................................................................................ 343
Figure C.15. Character 39. Scatterplots of the angle formed by the slope of the predentary
margin of the dentary and the frontal skull plane relative to the dental battery length for three
species of hadrosaurid ornithopods.............................................................................. 344
margin of the dentary and the frontal skull plane relative to the dental battery length for the
hadrosaurid Edmontosaurus ........................................................................................ 345
Figure C.17. Character 40. Boxplot showing the distribution of the angle of ventral deflection
of the rostral edentulous region of the dentary for a sample of iguanodontoidean taxa. The
inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers
of clusters in which the data can be classified. The photograph shows how this angle was
measured, exemplified in a dentary of Amurosaurus riabinini (AEHM 1/12)............... 348
Figure C.18. Character 40. Scatterplots of the angle of ventral deflection of the rostroventral
margin of the dentary relative to the dental battery length for four species of hadrosaurid
ornithopods ................................................................................................................. 349
margin of the dentary relative to the dental battery length for four species of hadrosaurid
ornithopods ................................................................................................................. 350
xxxii
Figure C.20. Character 41. Boxplot showing the distribution of ratio of the starting position
of the ventral deflection of the dentary rostroventral margin. The inserted smaller graph
shows the Bayesian information criterion (BIC) for various numbers of clusters in which
the data can be classified. The photograph shows how the two variables of this raio were
measured, exemplified in Lambeosaurus lambei (CMN 351)....................................... 353
Figure C.21. Character 41. Scatterplots of the ratio of the stating point of the ventral
reflection relative to the dental battery length for four species of hadrosaurid ornithopods.
.................................................................................................................................... 354
Figure C.22. Character 41. Scatterplots of the ratio of the stating point of the ventral
reflection relative to the dental battery length for four species of hadrosaurid ornithopods.
.................................................................................................................................... 355
Figure C.23. Character 42. Boxplot showing the distribution of the degree of lingual
projection of the symphyseal end of the dentary for a sample of iguanodontoidean taxa. The
of clusters in which the data can be classified. The photograph shows how the two variables
for this character were measured, exemplified in an indeterminate lambeosaurine dentary (?
Lambeosaurus, UALVP 11734) .................................................................................. 358
Figure C.24. Character 42. Scatterplots of the degree of lingual projection of the symphyseal
end of the dentary relative to the dental battery length for two hadrosaurid taxa .......... 359
Figure C.25. Character 43. Boxplot showing the distribution of the orientation of the dentary
symphysis relative to the lateral wall of the element for a sample of iguanodontoidean taxa.
The inserted smaller graph shows the Bayesian information criterion (BIC) for various
numbers of clusters in which the data can be classified. The photograph shows how this
character was measured, exemplified in a dentary of Brachylophosaurus canadensis (MOR
1071-7-15-98-226) ...................................................................................................... 361
Figure C.26. Character 43. Scatterplots of the orientation of the dentary symphysis
relative to the dental battery length for three hadrosaurid taxa ..................................... 362
Figure C.27. Character 44. Rostral views of dentaries showing the degree of lingual curvature
of the symphyseal region; that area curves approximately 90º in A and B, and gently forming
a wide arch in C. A, Iguanodon sp. (BMNH 28660). B, Parasaurolophus tubicen (NMMNH
P-25100). C, Maiasaura peeblesorum (YPM-PU 22405)............................................. 363
Figure C.28. Character 35. Medial views of dentaries showing the morphology of the dorsal
edge of the rostral end of the dentary (i.e., absence or presence of a concave profile). A,
Edmontosaurus annectens (USNM 4807). B, Prosaurolophus maximus (MOR 447). C,
Brachylophosaurus canadensis (FMNH 862). D, Lambeosaurus lambei (CMN 8633) ......
.................................................................................................................................... 364
xxxiii
Figure C.29. Character 46. Lateral views of hadrosaur dentaries, showing the development
and location of a ventral bulge (white rectangles) in this bone. A, Equijubus normani (IVPP
V12534). B, Jinzhousaurus yangi (IVPP V12691). C, Gryposaurus latidens (AMNH 5465).
D, Prosaurolophus maximusi (MOR 447-WSQ86)...................................................... 365
Figure C.30. Character 47. Boxplot showing the distribution of the angle of inclination of the
dentary coronoid process for a sample of iguanodontoidean taxa. The inserted smaller graph
shows the Bayesian information criterion (BIC) for various numbers of clusters in which the
data can be classified. The photograph shows how the two variables for this character were
measured, exemplified in a dentary of Prosaurolophus maximus (CMN 8894) ............ 368
Figure C.31. Characters 48 and 49. Medial views of the coronoid process of four
iguanodontoidean taxa. A, Iguanodon sp. (BMNH R754).]B, Brachylophosaurus canadensis
(CMN 8893); the white rectangle indicates the apomorphic caudodorsally pointed margin. B,
Prosaurolophus maximus (MOR 553S-7-12-8-5). C, Edmontosaurus annectens (BMNH
R3656). D, Lambeosaurine from Dinosaur Provincial Park (?Lambeosaurus, UALVP 11734)
.................................................................................................................................... 369
Figure C.32. Character 50. Ridges and striations on the medial side of the coronoid process
of the dentary, indicated by the arrows and the white rectangle. A, Bactrosaurus johnsoni
(BMNH 6581, juvenile specimen). B, Bactrosaurus johnsoni (PIN2540-1). C,
Parasaurolophus tubicen (NMMNH P-25100)............................................................ 370
Figure C.33. Character 51. Boxplot showing the distribution of the angle between the lateral
surface and the caudoventral region of the lateral side of the dentary for a sample of
iguanodontoidean taxa. The inserted smaller graph shows the Bayesian information criterion
(BIC) for various numbers of clusters in which the data can be classified. The photograph
shows how this character was measured, exemplified in a dentary of Maiasaura peeblesorum
(YPM-PU 22405)........................................................................................................ 373
Figure C.34. Character 52. Dorsal views of two hadrosaur dentaries showing the orientation
of the long axis of the occlusal plane relative to the lateral surface of the element. A,
Protohadros byrdi (SMU 74582). B, Prosaurolophus maximus (MOR 447)................ 374
Figure C.35. Character 53. Dorsal views of two iguanodontoidean dentaries showing the
presence and absence of lingual arching of the occlusal plane. A, Ouranosaurus nigeriensis
(GDF 300). B, Telmatosaurus transsylvanicus (BMNH R3386) .................................. 375
Figure C.36. Character 54. Medial views of three iguanodontoidean dentaries, with white
arrows showing the position of the caudal end of the dental battery relative to that of the
coronoid process. A, Iguanoodon atherfiendensis (BMNH R11521). B, Bactrosaurus
johnsoni (SBDE 95E5-12). C, Prosaurolophus maximus (MOR447-WSQ86) ............ 376
Figure C.37. Character 55. Degree of separation (arrows) between the dentary tooth row and
the coronoid process. A, Iguanodon atherfieldensis (BMNH R11521), caudal view. B, I.
xxxiv
atherfieldensis (BMNH R11521), dorsal view. C, Amurosaurus riabinini (AEHM 1/12),
caudal view. D, A. riabinini (AEHM 1/12), dorsal view .............................................. 377
Figure C.38. Character 56. Participation of the surangular in the coronoid process of the
dentary and shape of the rostrodorsal process (white rectangles) of the former. All images
display lateral views. A, Iguanodon atherfieldensis (BMNH R11521). B, cf. Bactrosaurus sp.
(MAL 0329-002). C, Edmontosaurus annectens (BMNH R4862). D, Prosaurolophus
maximus (MOR 447-3-8-5-86) .................................................................................... 378
Figure C.39. Character 57. Presence and absence of surangular foramen (arrows). A,
Iguanodon bernissartensis (IRSNB 1731), lateral view. B, Jinzhousaurus yangi (IVPP
V12691), lateral view. C, Brachylophosaurus canadensis (MOR 794), lateral view. D,
Prosaurolophus maximus (TMP 84.1.1), lateroventral view ........................................ 379
Figure C.40. Character 58. Presence and absence of surangular accessory foramen (black
arrow). A, Ouranosaurus nigeriensis (cast of GDF 300), lateral view. B, Parasaurolophus
tubicen (NMMNH P-25100), in lateral, and slightly ventral, view .............................. 380
Figure C.41. Character 59. Orientation of the lateral quadrate lap (white rectangles)
and convex surface of the surangular. A, Equijubus normani (IVPP V12534), lateral view.
B, Brachylophosaurus canadensis (CMN 8893), lateral view...................................... 381
Figure C.42. Character 60. Lateral curvature of the caudal process of the surangular (white
rectangles); all images in dorsal view. A, Brachylophosaurus canadensis (MOR 1071). B,
Amurosaurus riabinini (AEHM 1/12). C, Edmontosaurus (MOR 601). D, Lambeosaurus
lambei (CMN 2869). E, Gryposaurus monumentensis (RAM 6797). F, Parasaurolophus
tubicen (NMMNH P-25100)........................................................................................ 382
Figure C.43. Character 61. Boxplot showing the distribution of the angle between the
proximal medial margin of the surangular and the proximal medial margin of the caudal
process for a sample of iguanodontoidean taxa. The inserted smaller graph shows the
Bayesian information criterion (BIC) for various numbers of clusters in which the data can be
classified. The photograph shows how this angle was measured, exemplified in a dentary of
Lambeosaurus lambei (CMN 2869)............................................................................. 384
Figure C.44. Character 62. Lateral exposure of the angular (white arrows). A, Equijubus
equijubus (IVPP V12534), lateral view. B, Brachylophosaurus canadensis (TMP 104.90.1),
lateral view ................................................................................................................. 385
Figure C.45. Character 63. Presence and absence of coronoid bone. A, Corythosaurus
intermedius (ROM 777), lateral view. B, Iguanodon bernissartensis (cast of IRSNB 1534),
lateral view ................................................................................................................. 386
xxxv
Figure C.46. Character 64. Presence and absence of prearticular bone. A, Brachylophosaurus
canadensis (CMN 8893), medial view. B, Iguanodon atherfieldensis (BMNH R11521),
medial view. Abbreviations: an = angular; dt = dentary; prar = prearticular; sa = surangular;
spl = splenial ............................................................................................................... 387
Figure D.1. Character 65. Boxplot showing the distribution of the ratio of the medioteral
expansion of the premaxilla in a sample of iguanodontoidean taxa. The inserted smaller
which the data can be classified. The photograph shows how the two variables of this ratio
were measured, exemplified in the premaxilla of Gryposaurus notabilis (CMN 2278.. 410
Figure D.2. Character 66. Position of the premaxillary oral margin relative to the occlusal
plane of the dentition. A, Equijubus normani (IVPP V12534), lateral view. B Edmontosaurus
regalis (CMN 2288), lateral view................................................................................ 411
Figure D.3. Character 67. Morphology of the oral margin (arrows) of the premaxilla. State 0:
moderately expanded, dorsoventrally thicker towards the parasagittal plane of the snout and
slightly deflected ventrally. A, Iguanodon bernissartensis (IRSNB 1731), lateral view. B,
Ouranosaurus nigeriensis (cast of GDF 300), lateral view. C, Iguanodon atherfieldensis
(BMNH R11521), lateral view. D, Telmatosaurus transsylvanicus (BMNH R3386), lateral
view ............................................................................................................................ 412
Figure D.4. Character 67. Morphology of the oral margin (arrows) of the premaxilla. State 1:
moderately expanded, becoming thinner towards the parasagittal plane of the snout, so that
the narial fossa merges with the rostromedial region of the premaxilla adjacent to the contact
between both premaxillae. A, Corythosaurus intermedius (CMN 8676), lateral view. B,
Hypacrosaurus altispinus (CMN 8501), lateral view. C, “Lambeosaurus” laticaudus (LACM
17715), lateral view. D, Lambeosaurus lambei (CMN 2869), lateral view ................... 413
Figure D.5. Character 67. Morphology of the oral margin of the premaxilla. State 2: oral
margin (arrows) folded caudodorsally, thin and recurved. A, Gryposaurus notabilis (ROM
873), lateral view. B, Prosaurolophus maximus (CMN 2277), lateral view. C, Saurolophus
osborni (AMNH 5220), lateral view............................................................................ 414
Figure D.6. Character 67. Morphology of the oral margin of the premaxilla. State 3: ventrally
deflected and dorsoventrally expanded margin (arrows), forming a thick “lip-like” border. A,
Brachylophosaurus canadensis (MOR 794), lateral view. B, Edmontosaurus regalis (CMN
2288), lateral view. C, Hypacrosaurus stebingeri (MOR 549), lateral view. D, Maiasaura
peeblesorum (OTM F138), lateral view. E, Parasaurolophus walkeri (ROM 768), lateral
view. F, Tsintaosaurus spinorhinus (IVPP V725), lateral view................................... 415
Figure D.7. Character 68. Morphology of the rostrolateral corner (arrows) of the premaxillary
oral margin. A, Equijubus normani (IVPP V12534), lateral view. B, Lambeosaurus lambei
(ROM 794), lateral view.............................................................................................. 416
xxxvi
Figure D.8. Character 69. Presence and absence of a double denticle layer (arrows) ventral to
premaxillary oral margin. A, Iguanodon atherfieldensis (BMNH R11521), ventral view. B,
Telmatosaurus transsylvanicus (BMNH R3386), ventral view. C, Bactrosaurus johnsoni
(AMNH 6501), lateral view. D, Gryposaurus latidens (AMNH 5465), lateral view..... 417
Figure D.9. Character 70. Presence (arrows) and absence of premaxillary foramen. A,
Iguanodon bernissartensis (IRSNB 1731), caudodorsolateral view. B, Lambeosaurus lambei
(CMN 8633), dorsolateral view. C, Brachylophosaurus canadensis (CMN 8893), dorsolateral
view. D, Edmontosaurus annectens (SM R4036), lateral view..................................... 418
Figure D.10. Character 71. Presence and absence of accessory premaxillary foramen.
A, Iguanodon bernissartensis (CMN 2278), lateral view. B, Edmontosaurus annectens
(CMN 8509), lateral view. Abbreviations: pmxf = premaxillary foramen; apmxf = accessory
premaxillary foramen .................................................................................................. 419
Figure D.11. Character 72. Presence (arrow) and absence of premaxillary accessory fossa. A,
Equijubus normani (IVPP V12534), lateral view. B, Gryposaurus sp. (MOR 553S-7-23-8-
57), lateral view. C, Maiasaura peeblesorum (OTM F138), lateral view...................... 420
Figure D.12. Character 73. Presence (arrow) and absence of an additional accessory fossa in
the premaxilla. A, Brachylophosaurus canadensis (TMP 90.104.1), dorsolateral view. B,
Edmontosaurus regalis (CM 26258), dorsolateral view ............................................... 421
Figure D.13. Character 74. Elongation (arrows) of the premaxillary caudodorsal process over
the orbit. A, Gryposaurus sp. (MOR553S-7-18-91-107), lateral view. B, Corythosaurus sp.
(CMN 11375), lateral view. C, Gryposaurus notabilis (ROM 873), lateral view. D,
Lambeosaurus lambei (CMN 2869, lateral view.......................................................... 422
Figure D.14. Character 75. Presence and absence of vertical groove (arrow) on the
caudoventral process of the premaxilla. A, Parasaurolophus walkeri (ROM 768), lateral
view B, Corythosaurus casuarius (AMNH 5338), lateral view.................................... 423
Figure D.15. Character 67. Elongation of the caudoventral process of the adult premaxilla;
caudal end of this process indicated by arrows. A, Brachylophosaurus canadensis (TMP
90.104.1), lateral view. B, Saurolophus osborni (AMNH 5220), lateral view. C,
Lambeosaurus lambei (CMN 8703), lateral view. Abbreviations: lc = lacrimal; ns = nasal;
pf = prefrontal; pmx = premaxilla................................................................................ 424
Figure D.16. Character 77. Morphology of the caudal region of the caudoventral process of
the adult premaxilla. State 0: mediolaterally compressed and triangular. A,
Brachylophosaurus canadensis (CMN 8893), lateral view. B, Parasaurolophus walkeri
(ROM 768). Abbreviations: cddrs = caudodorsal; cdvnt = caudoventral; ns = nasal; pmx =
premaxilla ................................................................................................................... 425
xxxvii
the adult premaxilla. State 1: dorsoventrally broad and directed caudally or caudally and
slightly dorsally. A, Hypacrosaurus stebingeri (MOR 549), lateral view. B, Corythosaurus
casuarius (ROM 871), lateral view. Abbreviations: cddrs = caudodorsal; cdvnt =
caudoventral; ns = nasal; pmx = premaxilla................................................................. 426
the adult premaxilla. State 2: triangular and dorsoventrally expanded, laterally convex lobe
directed rostrodorsally. A, Hypacrosaurus stebingeri (MOR 549), lateral view. B,
Corythosaurus casuarius (ROM 871), lateral view. Abbreviations: cddrs = caudodorsal;
cdvnt = caudoventral; ns = nasal; pmx = premaxilla .................................................... 427
Figure D.19. Character 78. Presence of an accessory rostroventral flange (black rectangle) in
the caudodorsal premaxillary process that overlaps the lateral surface of the nasal in the
rostral region of a supracranial crest. Exemplified here in a specimen of Lambeosaurus
lambei (TMP 81.37.1), lateral view ............................................................................. 428
Figure D.20. Character 79. Laterodorsal profile of the caudodorsal and caudoventral margins
of the external bony naris. A, Gryposaurus notabilis (ROM 873), lateral view; the white
rectangle surrounds the area regarded homologous to the narial opening displayed in B, C
and D. B, Tsintaosaurus spinorhinus (IVPP V725), lateral view. C, Hypacrosaurus
stebingeri (MOR 549), lateral view. D, Parasaurolophus walkeri (ROM 768), lateral view
.................................................................................................................................... 429
Figure D.21. Character 80. Relative thickness of the caudal region of the caudodorsal and
caudoventral processes (indicated by the double-headed arrows). A, Parasaurolophus
walkeri (ROM 768), lateral view. B, “Lambeosaurus” laticaudus (LACM 17715), lateral
view. C, Hypacrosaurus altispinus (ROM 702), lateral view. Abbreviations: cddrs =
caudodorsal; cdvnt = caudoventral; pmx = premaxilla ................................................. 430
Figure D.22. Character 81. Boxplot showing the distribution of the length/width ratio of the
external bony naris of the premaxilla in a sample of lambeosaurine hadrosaurids. The
of clusters in which the data can be classified. The photograph shows the two variables of the
ratio were measured in a premaxilla of Corythosaurus casuarius (ROM 1933............. 431
Figure D.23. Character 82. Presence and absence of a well-developed dorsolateral flange
(rectangles) of the caudoventral process of the premaxilla. A, Brachylophosaurus canadensis
(MOR 794), dorsolateral view. B, Gryposaurus notabilis (ROM 873), dorsolateral view.
Abbreviations: mx = maxilla; ns = nasal; pmx = premaxilla ........................................ 433
Figure D.24. Character 83. Location of the nasal bone and nasal cavity in the adult skull. A,
Gryposaurus notabilis (CMN 2278), lateral view. B, Tsintaosaurus spinorhinus (IVPP
V725), lateral view. C, Corythosaurus casuarius (ROM 871), lateral view. Abbreviation:
ns = nasal .................................................................................................................... 434
xxxviii
Figure D.25. Character 84. Presence and absence of curvature of the caudodorsal region of
the nasal. A, Edmontosaurus annectens (MOR 003), lateral view. B, Hypacrosaurus
altispinus (CMN 8501), lateral view. Abbreviation: ns = nasal .................................... 435
Figure D.26. Character 85. Morphology of the rostral end of the nasal at the contact with the
dorsal process of the premaxilla. A, Gryposaurus notabilis (ROM 873), lateral view. B,
Maiasaura peeblesorum (ROM 44770), lateral view. C, Prosaurolophus maximus (ROM
1928), lateral view. Abbreviations: ns = nasal; pmx = premaxilla................................ 436
Figure D.27. Character 85. Morphology of the rostral end of the nasal at the contact with the
dorsal process of the premaxilla. A, Lambeosaurus lambei (ROM 869), lateral view. B,
Corythosaurus (ROM 870), lateral view. Abbreviations: ns = nasal; pmx = premaxilla .....
.................................................................................................................................... 437
Figure D.28. Character 86. Morphology of the nasal contact with the caudodorsal region of
the caudoventral premaxillary process at the caudal margin of the narial foramen.A,
Equijubus normani (IVPP V725), lateral view. B, Brachylophosaurus canadensis (CMN
8893), lateral view. C, Edmontosaurus sp. (BMNH R8927), lateral view. D, Prosaurolophus
maximus (CMN 2277), lateral view. Abbeviations: ns = nasal; pmx = premaxilla........ 438
Figure D.29. Character 87. Location of the rostral end of the dorsal process of the nasal
relative to the rostral margin of the narial foramen (indicated by the separation between the
pairs of white arrows). A, Gryposaurus notabilis (ROM 873), lateral view. B,
Prosaurolophus maximus (ROM 1928), lateral view. Abbreviations: ns = nasal; pmx =
premaxilla ................................................................................................................... 439
Figure D.30. Character 88. Presence of a caudoventral region of the nasal ventrally recurved
and hook-shaped, with a rostral process that inserts under the caudoventral process of the
permaxilla Exemplified here in a specimen of Lambeosaurus lambei (CMN 8703), lateral
view. Abbreviations: ns = nasal; pmx = premaxilla...................................................... 440
Figure D.31. Character 89. Caudal end of the nasals forming a pair of finger-like process
(arrows) on top of the frontals and centered around the sagittal plane of the skull roof.
Exemplified here in A, a subadult specimen of Kritosaurus navajovius (NMMNH P-16106,
dorsal view) and B, Prosaurolophus maximus (CMN 2870). Abbreviations: f = frontal;
ns = nasal; pf = prefrontal; po = postorbital ................................................................. 441
Figure D.32. Character 90. Caudal end of the nasals forming a pair of small and short
processes (arrows) that insert between the frontals at the sagittal plane of the skull roof.
Exemplified here in a specimen of Gryposaurus notabilis (AMNH 5350), dorsal view.
Abbreviations: f = frontal; ns = nasal; pf = prefrontal; po = postorbital........................ 442
xxxix
Figure D.33. Character 91. Position of the summit (asterisk) of the nasal arch relative to the
caudodorsal margin of the narial foramen (rectangle). A, Gryposaurus latidens (AMNH
5465) lateral view. B, G. notabilis (CMN 2278), lateral view ...................................... 443
Figure D.34. Character 92. Rostrodorsal process that is medially offset from the body of the
maxilla and extends medial to the caudovenventral process of premaxilla, shown in A,
Brachylophosaurus canadensis (MOR-1071-7-6-98-79), lateral view; and B, medial view of
the same specimen. Rostral end of the maxilla formed by a ventrally sloping rostrodorsal
shelf that underlies the premaxilla, show in C, Corythosaurus intermedius (CMN 8676),
lateral view; and D, medial view of the same specimen ............................................... 444
Figure D.35. Character 93. Absence and presence of lateral exposure of the rostrodorsal
process of the maxilla (arrow). A, Saurolophus osbori (CMN 8796), lateral view. B,
Brachylophosaurus canadensis (TMP 90.104.1), lateral view...................................... 445
Figure D.36. Character 94. Presence and absence of a pendant rostroventral process of
the maxilla. A, Velafrons coahuilensis (CPC-59), lateral view. B, Lambeosaurus laticaudus
(LACM 17715), lateral view ....................................................................................... 446
Figure D.37. Character 95. Boxplot showing the distribution of the angle between the dorsal
margin of the rostroventral process or shelf of the maxilla and the rostral segment of the tooth
row. The inserted smaller graph shows the Bayesian information criterion (BIC) for various
numbers of clusters in which the data can be classified. The photograph shows how this angle
was measured, exemplified in a maxilla of Corythosaurus casuarius (ROM 1933....... 449
Figure D.38. Character 96. Lateral profile of the lateral surface of the rostrodorsal region
(rectangles) of the maxilla. A, Saurolophus osborni (AMNH 5221), lateral view. B,
Corythosaurus casuarius (ROM 1933), lateral view. C, Pararhabdodon isonensis (IPS-SRA
22), lateral view .......................................................................................................... 450
Figure D.39. Character 93. Elevation of the lateral surface of the rostrodorsal region of the
maxilla. A, Brachylophosaurus canadensis (MOR 1071-7-6-98-79), lateral view. B,
Gryposaurus sp. (MOR 553S-8-26-9-55), lateral view ................................................ 451
Figure D.40. Character 98. Boxplot showing the distribution of the position of the dorsal
process and the dorsal margin of the dorsolateral eminence of the maxilla for a sample of
iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information
criterion (BIC) for various numbers of clusters in which the data can be classified. The
photograph shows how the two variables of this ratio were measured, exemplified in a
maxilla of Gryposaurus sp. (MOR 553S-8-26-9-55).................................................... 454
Figure D.41. Character 99. Morphology of the dorsal process of the maxilla. A, Bactrosaurus
johnsoni (SBDE 95/E5), lateral view. B, Maiasaura peeblesorum (OTM F138), lateral view.
C, Prosaurolophus maximus (TMP 84.1.1), lateral view. D, Amurosaurus riabinini (AEHM
xl
1/12), lateral view. E, Hypacrosaurus altispinus (CMN 8674), lateral view. F, Tsintaosaurus
spinorhinus (IVPP V725), lateral view ........................................................................ 455
Figure D.42. Character 1000. Morphology of the jugal articulation surface of the maxilla.
State 0: jugal joint protruding lateral to the caudal third of the maxilla as a mediolaterally
compressed finger-like process (rectangle and arrow). Exemplified here in a specimen of
Iguanodon atherfieldensis (BMNH R5764) in A, lateral and B, dorsal view................ 456
State 1: joint located on a promontory (rectangles) dorsal and rostral to the ectopterygoid
shelf, with a caudolaterally directed corner (arrow). Exemplified here in a specimen of
Bactrosaurus johnsoni (BMNH 6553) in A, lateral and B, dorsal view. C, cf. Telmatosaurus
sp. (FGGUB R1010), lateral view ............................................................................... 457
State 2: subtriangular joint surface (rectangles) for the jugal that is more laterally than
dorsally-facing with a lateroventrally-directed pointed corner (arrow) located adjacent and
slightly dorsal to the the lateral ridge of the ectopterygoid shelf. Exemplified here in a
specimen of Brachylophosaurus canadensis (MOR 1071-8-13-98-554) in A, lateral and B,
dorsal view. C, Corythosaurus intermedius (CMN 8676), lateral view ........................ 458
State 3: dorsally elevated jugal joint (distance between the ventral margin of the jugal joint
and ectopterygoid shelf nearly equal to depth of the caudal segment of the maxilla), caudal
margin of the joint flush with the caudal margin of the rostrodorsal eminence of the lateral
side of the maxilla Exemplified here in a specimen of Tsintaosaurus spinorhinus (IVPP
V725) in A, lateral and B, dorsal view. C, Pararhabdodon isonensis (IPS SRA 22), lateral
view ............................................................................................................................ 459
Figure D.46. Character 101. Arrangement of maxillary foramina (arrows) ventral and rostral
to the jugal articulation, shown in lateral view. A, Bactrosaurus johnsoni (cast of PIN 2549-
1). B, Equijubus normani (IVPP V12534). C, Gilmoreosaurus mongoliensis (AMNH 6551).
D, Iguanodon atherfeldensis (BMNH R11521). E, Brachylophosaurus canadensis (CMN
8893). F, B. canadensis (MOR 1071-7-6-98-79).......................................................... 460
to the jugal articulation, shown in lateral view. A, Brachylophosaurus canadensis (MOR
1071-8-13-98-559). B, B. canadensis (MOR 1071-8-15-98-573). C, B. canadensis (MOR
794). D, B. canadensis (TMP 90.104.1). E, Lambeosaurus lambei (CMN 2869). F, L. lambei
(TMP 81.37.01)........................................................................................................... 461
to the jugal articulation, shown in lateral view. A, Edmontosaurus regalis (AMNH 5254). B,
E. regalis (CMN 2289). C, E. regalis (ROM 801). D, Saurolophus osborni (AMNH 5220).
E, S. osborni (AMNH 5221, left maxilla). F, S. osborni (AMNH 5221, right maxilla.. 462
xli
to the jugal articulation, shown in lateral view. A, Prosaurolophus maximus (CMN 2277). B,
Corythosaurus intermedius (CMN 8676). C, C. intermedius (ROM 776). D, C. intermedius
(ROM 777). E, C. casuarius (ROM 1933). F, C. casuarius (TMP 82.37.01)................ 463
Figure D.50. Character 102. Boxplot showing the distribution of the number of maxillary
foramina ventral and rostral to the jugal articulation for a sample of iguanodontoidean
ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for
various numbers of clusters in which the data can be classified. The photograph shows the
shape and location of these foramina in a maxilla of Brachylophosaurus canadensis (MOR
1071-8-13-98-554) ...................................................................................................... 466
Figure D.51. Character 103. Location of the large rostral maxillary foramen (arrows).
A, Equijubus normani (IVPP V725), lateral view. B, Jinzhousaurus yangi (IVPP V12691),
lateral view. B, Bactrosaurus johnsoni (SBDE 95/E5), lateral view............................. 467
A, Brachylophosaurus canadensis (MOR 1071-8-13-98-559), rostral view. B, lateral view of
the same specimen. C, Corythosaurus sp. (TMP 82.37.01), rostral view. D, lateral view of the
same specimen ............................................................................................................ 468
Figure D.53. Character 104. State 0: presence of maxilla-lacrimal contact. A, Gryposaurus

notabilis (CMN 2278), lateral view. B, Prosaurolophus maximus (ROM 1928), lateral view.
C, Parasaurolophus walkeri (ROM 768), lateral view. D, Saurolophus osborni (AMNH
5220), lateral view. Abbreviations: j = jugal; lc = lacrimal; mx = maxilla; ns = nasal; pf =
prefrontal; pmx = premaxilla....................................................................................... 469
Figure D.54. Character 104. State 1: absence of maxilla-lacrimal contact. A,

Brachylophosaurus canadensis (TMP 90.104.1), lateral view. B, Corythosaurus casuarius
(AMNH 5338), lateral view. C, Edmontosaurus annectens (SM R4036), lateral view. D,
Maiasaura peeblesorum (ROM 44770), lateral view. Abbreviations: j = jugal; lc = lacrimal;
mx = maxilla; ns = nasal; pmx = premaxilla ................................................................ 470
Figure D.55. Character 105. Box plot showing the distribution of the relative length of the
ectopterygoid shelf. The inserted smaller graph shows the Bayesian information criterion
(BIC) for various numbers of clusters in which the data can be classified. The photograph to
the lower right corner shows how the two variables of this raio were measured, exemplified
in a maxilla of Maiasaura peeblesorum (OTM F138).................................................. 473
Figure D.56. Character 106. Boxplot showing the distribution of the angle of inclination of
the ectopterygoid shelf. The inserted smaller graph shows the Bayesian information criterion
(BIC) for various numbers of clusters in which the data can be classified. The photograph
shows how this angle was measured, exemplified in a maxilla of Brachylophosaurus
canadensis (MOR1071-7-6-98-79).............................................................................. 476
xlii
Figure D.57. Character 107. Morphology of the lateral emargination (arrows) of the
ectopterygoid shelf. A, Iguanodon atherfieldensis (BMNH R5764), lateral view. B,
Protohadros byrdi (SMU 74582), lateral view. C, cf. Telmatosaurus transsylvanicus (BMNH
R4911), lateral view. D, Edmontosaurus regalis (CMN 2289), lateral view................. 477
Figure D.58. Character 108. Position of the arcuate row of special foramina on the medial
side of the dentary. A, Iguanodon atherfieldensis (BMNH R5764), lateral view. B,
Brachylophosaurus canadensis (MOR 1071-8-13-98-559), lateral view...................... 478
Figure D.59. Character 109. Morphology of the lacrimal.. A, Maiasaura peeblesorum (ROM
44770), lateral view. B, Lambeosaurus lambei (ROM 1218), lateral view. Abbreviations: lc =
lacrimal....................................................................................................................... 479
Figure D.60. Character 110. Absence (A) and presence (B) of a prominent convexity (arrow),
rostral to the jugal notch, in the ventral margin of the lacrimal. A, Edmontosaurus annectens
(SM R4036), lateral view. B, Gryposaurus monumentensis (RAM 6797), lateral view ......
.................................................................................................................................... 480
Figure D.61. Character 111. Morphology of the apex (arrow) of the rostral process of the
jugal A, cf. Bactrosaurus johnsoni (MAL 0329-002), lateral view. B, Maiasaura
peeblesorum
(ROM 44770), lateral view. C, Edmontosaurus annectens (SM R4036), lateral view. D,
Gryposaurus notabilis (ROM 873), lateral view .......................................................... 481
Figure D.62. Character 111. Morphology of the apex (arrow) of the rostral process of the
jugal. A, Corythosaurus intermedius (ROM 777), lateral view. B, Lambeosaurus lambei
(ROM 1218), lateral view. C, Parasaurolophus tubicn (NMMNH P-25100), lateral view. D,
Hypacrosaurus altispinus (ROM 789), lateral view..................................................... 482
Figure D.63. Character 112. Dorsoventral expansion of the caudodorsal margin of the rostral
process of the jugal (rectangles). A, Equijubus normani (IVPP V12534), lateral view.
B, Lambeosaurus lambei (CMN 2869), lateral view .................................................... 483
Figure D.64. Character 113. Morphology of the triangular caudoventral margin of the rostral
process of the jugal (rectangles). A, Prosaurolophus maximus (TMP 84.1.1), lateral view. B,
Brachylophosaurus canadensis (TMP 90.104.1), lateral view. C, Corythosaurus casuarius
(ROM 1933), lateral view............................................................................................ 484
Figure D.65. Character 114. Location of the caudoventral apex of the rostral process relative
to the caudodorsal articulation with the lacrimal (circles) (having as reference the
longitudinal axis of the rostral process oriented horizontally). A, Edmontosaurus sp. (BMNH
R8297). B, Gryposaurus notabilis (ROM 764), lateral view. C, Prosaurolophus maximus
(ROM 1928), lateral view. D, Brachylophosaurus canadensis (TMP 90.104.1), lateral view
.................................................................................................................................... 485
xliii
Figure D.66. Character 115. Orientation of the medial articular surface (rectangles) of the
rostral process of the jugal. A, Tanius sinensis (PMU R240), medial view. B, Edmontosaurus
sp. (MOR 601), medial view ....................................................................................... 486
Figure D.67. Character 116. Inclination of the bony rim that bounds caudally the medial
articulation surface of the rostral process of the jugal (this inclination is considered relative to
the rostrocaudal longer axis of the jugal). A, Maiasaura peeblesorum (OTM F138), medial
view. B, Prosaurolophus maximus (MOR 454-6-24-6-2), medial view........................ 487
Figure D.68. Character 117. Rostrocaudal width of the curvature of the ventral margin of the
jugal, between the caudoventral flange and the rostral process. A, Prosaurolophus maximus
(MOR 447-7-6-7-1), lateral view. B, Charonosaurus jiayinensis (CUST unrecorded number),
lateral view ................................................................................................................. 488
Figure D.69. Character 118. Boxplot showing the distribution of the ratio between the
dorsoventral depth of the flange and the minimum depth of the caudal constriction of the
jugal) . The inserted smaller graph shows the Bayesian information criterion (BIC) for
various numbers of clusters in which the data can be classified. The photograph how the two
variables of this raio were measured, exemplified in a jugal of Brachylophosaurus
canadensis (MOR1071-7-6-98-79).............................................................................. 492
Figure D.71. Character 119. Lateral outlines of the quadratojugal flange of the jugal. A,
Equijubus normani (IVPP V12534). B, Iguanodon atherfieldensis (BMNH R5764). C,
Protohadros byrdi (SMU 74582). D, Brachylophosaurus canadensis (MOR 794). E, B.
canadensis (MOR 1071-7-27-98-453). F, B. canadensis (MOR 1071-7-31-99-281-O). . G,
Maiasaura peeblesorum (ROM 44770). H, B. canadensis (CMN 8893). I, Bactrosaurus
johnsoni (AMNH6379). J, B. johnsoni(AMNH 6396). K, Amurosaurus riabinini (AEHM
1/12). L, Corythosaurus casuarius (AMNH 5240). M, C. casuarius (AMNH 5382). N, C.
intermedius (CMN 8704) Ñ, C. intermedius (ROM 777). O, C. intermedius (ROM 845). P,
C. casuarius (ROM 868). Q, Corythosaurus sp. (CMN 34845). R, Corythosaurus sp. (ROM
1947). S, H. altispinus (CMN 8501). T, H. stebingeri (AMNH 5461). U, Lambeosaurus
lambei (CMN 8703). V, L. lambei (AMNH 5373). W, L. lambei (FMNH PR380). X, L.
lambei (AMNH 5240).Y, L. lambei (CMN 8633). Z, Velafrons coahuilensis (CPC-59). AA,
Edmontosaurus regalis (AMNH 5254). BB, Edmontosaurus sp. (BMNH R8297). CC, E.
regalis (CM 26258. DD, E. regalis (CMN 2288). EE, E. regalis (FMNH P15004). FF,
Prosaurolophus maximus (MOR 447). GG, E. regalis (MOR 801). HH, E. annectens (CMN
8509). II, E. annectens (UCM 20892). JJ, Edmontosaurus sp. (AMNH 5046). KK,
Gryposaurus notabilis (TMP 80.22.1). LL, G. notabilis (MSNM V345). MM, P. maximus
(CMN 2870). NN, P. maximus (AMNH 5386). ÑÑ, P. maximus (ROM 787). OO, P.
maximus (TMM 41262-1). PP, P. maximus (ROM 1928). QQ, S. osborni (AMNH 5220)
.................................................................................................................................... 494
Figure D.72. Character 120. Morphology of the ventral margin (rectangles) located between
the caudoventral and quadratojugal flanges of the jugal. A, Hypacrosaurus altispinus (ROM
xliv
702), lateral view. B, Velafrons coahuilensis (CPC-59), lateral view. C, Brachylophosaurus
canadensis (MOR 1071-7-31-99-281O), lateral view. D, Gryposaurus notabilis (ROM 873),
lateral view ................................................................................................................. 495
Figure D.73. Character 121. Boxplot showing the distribution of the relative depth of the
caudal and rostral constrictions of the adult jugal in a sample of iguanodontoidean
the two variables of this raio were measured, exemplified in a jugal of Prosaurolophus
maximus (MOR 447-7-6-7-1) ...................................................................................... 498
Figure D.74. Character 122. Boxplot showing the distribution of the ratio between the
minimum depth of the caudal constriction and he distance between the point of maximum
curvature of the infratemporal margin and the caudal margin of the lacrimal process .The
of clusters in which the data can be classified. The photograph at the lower right corner
shows how the two variables of this raio were measured, exemplified in a jugal of
Edmontosaurus sp. (MOR 601) ................................................................................... 499
Figure D.75. Character 123. Relative width of the orbital and infratemporal margins of the
jugal. A, Velafrons coahuilensis (CPC-59), lateral view. B, Maiasaura peeblesorum (OTM
F138), lateral view. C, Prosaurolophus maximus (MOR 447-7-6-7-1), lateral view..... 500
Figure D.76. Character 124. Boxplot showing the distribution of the angle of curvature of the
long axis of the quadrate. The inserted smaller graph shows the Bayesian information
photograph shows how this angle was measured, exemplified in a quadrate of Iguanodon
atherfieldensis (BMNH 11521) ................................................................................... 503
Figure D.77. Character 125. Boxplot showing the distribution of the ratio of the position of
the quadratojugal notch of the quadrate. The inserted smaller graph shows the Bayesian
exemplified in a quadrate of Prosaurolophus maximus (MOR 454-6-24-6-2) .............. 507
Figure D.78. Character 126. Boxplot showing the distribution of the angle of inclination of
the dorsal margin of the quadrate notch of the quadrate. The inserted smaller graph shows the
exemplified in a quadrate of Prosaurolophus maximus (TMP 98.50.1)........................ 510
Figure D.79. Character 127. Morphology of the lateral profile of the quadratojugal notch of
the quadrate. A, Hypacrosaurus stebingeri (MOR 455), lateral view. B, Tsintaosaurus
spinorhinus (IVPP V725), lateral view. C, Edmontosaurus sp. (ROM 658), lateral view ...
.................................................................................................................................... 511
xlv
Figure D.80. Character 128. Development of the squamosal buttress on the caudal margin of
the dorsal end of the of the quadrate. A, Prosaurolophus maximus (CMN 2870), lateral view.
B, Gryposaurus notabilis (TMP 80.22.1), lateral view................................................. 512
Figure D.81. Character 129. Boxplot showing the distribution of the width/length ratio of the
distal surface of the quadrate. The inserted smaller graph shows the Bayesian information
photograph shows how the two variables of this raio were measured, exemplified in a
quadrate of Edmontosaurus sp. (ROM 658)................................................................. 514
Figure D.82. Character 129. Degree of ventral offset of the lateral condyle of the distal end of
the quadrate relative to the medial condyle (correlated to the width/length ratio of the distal
surface of the quadrate). A, Bactrosaurus johnsoni (SBDE 95/E5), lateral view. B,
Brachylophosaurus canadensis (MOR 1071-8-13-98-559D), lateral view ................... 515
Figure D.83. Character 130. Development of a prefrontal crest (arrows). A, Edmontosaurus

sp. (AMNH 427), lateral view. B, dorsal view of the same specimen. C, Jaxartosaurus
aralensis (PIN 1/50009), dorsolateral view. D, Corythosaurus casuarius (ROM 871), lateral
view ............................................................................................................................ 516
Figure D.84. Character 130. Presence of a well-developed prefrontal crest (arrows). A,

Hypacrosaurus altispinus (ROM 702), lateral view. B, dorsal view of the same specimen. C,
H. stebingeri (MOR 553S-7-27-2-93), dorsal view...................................................... 517
Figure D.85. Character 131. Lateral profile of the rostrodorsal margin (rectangles) of the
prefrontal. A, Lambeosaurus lambei (ROM 1218), lateral view. B, Kritosaurus navajovius
(NMMNH P-16106), lateral view................................................................................ 518
Figure D.86. Character 132. Mediolateral breadth of the exposed rostroventral region
(rectangles) of the prefrontal. A, Edmontosaurus annectens (SM R4036), lateral view. A,
Lambeosaurus lambei (CMN 8633), lateral view......................................................... 519
Figure D.87. Character 133. State 0: exclusion of the prefrontal from the circumnarial fossa.
A, Gryposaurus notabilis (ROM 873),lateral view. B, G. notabilis (AMNH 5350),
rostrodorsolateral view. Abbreviations: bcf = boundary of the circumnarial fossa; ns = nasal;
pf = prefrontal ............................................................................................................. 520
Figure D.88. Character 133. State 1: inclusion of the prefrontal in the circumnarial fossa,
exemplified in a skull roof of Maiasaura peeblesorum (OTM F138). A, rostrodorsolateral
view. B, right lateral view. Abbreviations: bcf = boundary of the circumnarial fossa; ns =
nasal; pf = prefrontal ................................................................................................... 521
Figure D.89. Character 134. Absence and presence of eversion (arrow) of the rostrodorsal
margin of the prefrontal. A, Gryposaurus notabilis (ROM 873), lateral view. A,
Brachylophosaurus canadensis (MOR 794), lateral view ............................................ 522
xlvi
Figure D.90. Character 135. Exposure of the prefrontal-nasal contact. A, Brachylophosaurus
canadensis (MOR 794), lateral view. B, Lambeosaurus lambei (CMN 2869), lateral view. C,
Corythosaurus casuarius (ROM 871), lateral view. Abbreviations: ns = nasal; pf =
prefrontal; pmx = premaxilla....................................................................................... 523
Figure D.91. Character 136. Absence and presence of a dorsal promontorium (arrow) on the
rostral process of the postorbital. A, Iguanodon atherfieldensis (BMNH R11521), lateral
view. B, Brachylophosaurus canadensis (TMP 90.104.1), lateral view. C, Gryposaurus
notabilis (RAM 6797), lateral view. D, Saurolophus osborni (AMNH 5220), lateral view.
.................................................................................................................................... 524
Figure D.92. Character 136.Presence of a dorsal promontorium (arrows) on the rostral

process of the postorbital. A, Charonosaurus jiayinensis (CUST JIV1223), lateral view. B,
Parasaurolophus cyrtocristatus (UCMP 143270), lateral view. C, Lambeosaurus lambei
(YPM 3222), lateral view. D, L. lambei (CMN 2869), lateral view.............................. 525
Figure D.93. Character 137. Rostrocaudal constriction of the dorsal region of the
infratemporal fenestra (rectangle) and length of the caudal process of the postorbital. A,
Corythosaurus casuarius (ROM 776), lateral view. B, Brachylophosaurus canadensis (cast
of MOR 794). C, Parasaurolophus tubicen (NMMNH P-25100), lateral view............. 526
Figure D.94. Character 138. Morphology of the central body of the postorbital (arrows and
rectangle). A, Bactrosaurus johnsoni (AMNH 6365),lateral view. B, Equijubus normani
(IVPP V12534), lateral view. C, Iguanodon bernissartensis (IRSNB 1731), lateral view.
D, Amurosaurus riabinini (AEHM 1/12), lateral view. E, Edmontosaurus regalis (CMN
2288), lateral view. F, rostral view of the same specimen ............................................ 527
Figure D.95. Character 139. Length of the jugal process (rectangles) of the postorbital.
A, Corythosaurus casuarius (AMNH 5338), lateral view. B, Brachylophosaurus canadensis
(TMP 90.104.1), lateral view....................................................................................... 528
Figure D.96. Character 140. Morphology of the caudal end of the caudal process of the
postorbital at its articulation with the squamosal A, Jinzhousaurus yangi (IVPP V12691),
dorsolateral view. B, Bactrosaurus johnsoni (PIN 2549-1), dorsolateral view.............. 529
Figure D.97. Character 141. Caudal extension of the caudal ramus of the postorbital
(rectangles) overlapping the laterodorsal surface of the squamosal A, Gryposaurus notabilis
(CMN 2278), lateral view. B, Brachylophosaurus canadensis (CMN 8893), lateral view. C,
Parasaurolophus cyrtocristatus (UCMP 143270), lateral view. Abbreviations: po =
postorbital; qc = quadrate cotylus; sq = squamosal ...................................................... 530
Figure D.98. Character 142. Boxplot showing the distribution of the ratio of the length of the
precotyloid process of the squamosal for a sample of iguanodontoidean ornithopods. The
xlvii
of this ratio were measured, exemplified in a squamosal of Parasaurolophus walkeri (ROM
768) ............................................................................................................................ 532
Figure D.99. Character 143. Dorsoventral expansion of the caudolateral surface (rectangles)
of the squamosal. A, Brachylophosaurus canadensis view (CMN 8893). B, Hypacrosaurus
stebingeri (MOR 455) ................................................................................................. 533
Figure D.100. Character 144. Presence (arrows) and absence (rectangle) of separation of the
squamosals at the occipital margin of the skull roof. A, Kritosaurus navajovius (NMMNHP-
16106), dorsal view. B, Tsintaosaurus spinorhinus (IVPP V725), dorsal view. C,
Brachylophosaurus canadensis (MOR 1071-7-7-98-86), dorsal view. D, Hypacrosaurus
altispinus (CMN 8501), dorsal view............................................................................ 534
Figure D.101. Character 145. Presence and absence of rostromedial indenture of the medial
ramus of the squamosal . A, Bactrosaurus johnsoni (AMNH 6365), dorsal view. B,
Maiasaura peeblesorum (OTM F138), dorsal view ..................................................... 535
Figure. E.1. Character 146. Absence and presence (rectangles) of bifurcating rostromedian
region of the frontals. All specimens in dorsal view. A, Tsintaosaurus spinorhinus (IVPP
V725). B, Maiasaura peeblesorum (YPM-PU 22405). C, Tanius sinensis (PMU R240). D,
Secernosaurus koerneri (MACN-RN 144). E, Hypacrosaurus altispinus (CMN 8675).
Abbreviations: f =frontal; ns = nasal; po = postorbital ................................................. 543
Figure E.2. Character 147. Absence and presence (rectangle) of frontal fontanelle. A,
Prosaurolophus maximus (CMN 2870). B, Lophorhothon atopus (FMNH P27383).
Abbreviations: f =frontal; ns = nasal............................................................................ 544
Figure E.3. Character 148. Presence of a frontal platform. A, Amurosaurus riabinini

(cast of AEHM 1/12). B, Brachylophosaurus canadensis (MOR 1071-7-13-99-87-I, subadult
specimen). Abbreviation: fp = frontal platform............................................................ 545
Figure E.4. Character 149. Presence of a frontal platform. A, Charonosaurus jiayinensis

(CUST JV-12512). B, Parasaurolophus tubicen (PMU R222). C, Saurolophus osborni
(AMNH 5221). Abbreviation: fp = frontal platform..................................................... 546
Figure E.5. Character 150. Absence and presence (arrows) of median cleft in the frontal
platform. A, Saurolophus osborni (AMNH 5221), dorsal view. B, Parasaurolophus tubicen
PMU R222), dorsal view. C, Hypacrosaurus altispinus (CMN 8675), dorsal view.
Abbreviation: fp = frontal platform ............................................................................. 547
Figure E.6. Character 151. Absence and degree of exposure (rectangles) of the frontal in the
dorsal orbital margin. A, Edmontosaurus sp. (YPM 618), dorsal view. B, Gryposaurus
notabilis (ROM 764), dorsal view. C, cf. Corythosaurus sp. (AMNH 5433), dorsal view.
Abbreviations: f = frontal; pf = prefrontal; po = postorbital ......................................... 548
xlviii
Figure E.7. Character 152. Absence and presence of frontal doming (rectangle). A,
Gryposaurus notabilis (AMNH 5350). B, Amurosaurus riabinini (cast of AEHM 1/12).
Abbreviation: f = frontal.............................................................................................. 549
Figure E.8. Character 153. Length/width proportions of the ectocranial surface (rectangles) of
the frontal. A, Tanius sinensis (PMU R222), dorsal view. B, cf. Corythosaurus sp. (AMNH
5433), dorsal view. C, Charonosaurus jiayinensis (CUST JV-12512), dorsal view.
Abbreviation: f = frontal.............................................................................................. 550
Figure E.9. Character 154. Absence and presence of annular ridge. A, Brachylophosaurus
canadensis (FMNH 862), ventral view. B, cf. Corythosaurus sp. (USNM 11893), ventral
view ............................................................................................................................ 551
Figure E.10. Character 155. Boxplot showing the distribution of the ratio of the length/width
ratio of the parietal for a sample of iguanodontoidean ornithopods. The inserted smaller
which the data can be classified. The photograph shows how the two variables of this raio
were measured, exemplified in the parietal of Tanius sinensis (PMU R240) ................ 553
Figure E.11. Character 156. Orientation of the parietal midline crest (rectangles). A,
Gryposaurus notabilis (AMNH 5350). B, Hypacrosaurus stebingeri (MOR553S-7-27-2-93)
.................................................................................................................................... 554
Figure E.12. Character 157. Morphology of the rostromedian process of the parietal
(rectangle) that forms a crenulated suture in between the caudomedian margin of the frontals.
A, Tanius sinensis (PMU R222), dorsal view. B, cf. Corythosaurus sp. (AMNH 5433), dorsal
view. C, Edmontosaurus sp. (AMNH 427), dorsal view .............................................. 555
Figure E.13 Character 158. Rostral extension (rectangles) of the sagittal crest along the
dorsal surface of the parietal. Specimens shown in dorsal view. A, Edmontosaurus regalis
(CMN 2289). B, Gryposaurus notabilis (AMNH 5350). C, Hypacrosaurus stebingeri (MOR
553S-7-27-2-93). D, Corythosaurus intermedius (ROM 776). E, Amurosaurus riabinini
(AEHM 1/12). F, Charonosaurus jiayinensis (CUST JV1251-57) ............................... 556
Figure E.14. Character 159. Participation of the basioccipital in the ventral margin of the
foramen magnum. All specimens shown in caudal view. A, Equijubus normani (IVPP
V12534). B, Gryposaurus notabilis (AMNH 5350). C, Hypacrosaurus altispinus (CMN
2246, juvenile specimen). D, Corythosaurus intermedius (CMN 8676). Abbreviations:
bo = basioccipital; ex = exoccipital.............................................................................. 557
Figure E.15. Character 160. Orientation of the articulation margin of the occipital condyle
relative to the convex caudoventral border of the basioccipital. The arrows point to a shallow
cleft dividing the articulation margin of the occipital condyle from the caudal margin of the
basioccipital. A, Ouranosaurus nigeriensis (cast of GDF 300), lateral view. B, Bactrosaurus
xlix
johnsoni (AMNH 6365), lateral view. C, Amurosaurus riabinini (cast of AEHM 1/12), lateral
view ............................................................................................................................ 558
Figure E.16. Character 161. Length of the basioccipital constriction (arrows). A,

Bactrosaurus johnsoni (AMNH 6365), ventral view. B, Ouranosaurus nigeriensis (cast of
GDF 300), ventral view. C, Gryposaurus notabilis (CMN 2289), ventral view. D,
Charonosaurus jiayinensis (CUST JV1251-57), ventral view...................................... 559
Figure E.17. Character 162. Boxplot showing the distribution of the orientation of the
basipterygoid process of the basisphenoid for a sample of iguanodontoidean ornithopods. The
of this raio were measured, exemplified in the parietal of Charonosaurus jiayinensis (CUST
JV1251-57) ................................................................................................................. 561
Figure E.18. Character 163. Developement of the alar process (rectangles) of the
basisphenoid. A, Hypacrosaurus stebingeri (MOR 455). B, Brachylophosaurus canadensis
(MOR 1071-7-7-98-86)............................................................................................... 562
Figure E.19. Character 164. Absence and presence (arrows) of a ventral transverse ridge
between the basipterygoid processes of the basisphenoid. All specimens shown in ventral
view. A, Amurosaurus riabinini (cast of AEHM 1/12). B, Hypacrosaurus altispinus (CMN
8675). C, Parasaurolophus tubicen (PMU R222). D, Edmontosaurus regalis (CMN 2289)
.................................................................................................................................... 563
Figure E.20. Character 165. Absence and presence (rectangles) of a median ventral process
between the basipterygoid processes of the basisphenoid. All specimens shown in caudal
view. A, Corythosaurus intermedius (ROM 776). B, Hypacrosaurus stebingeri (MOR 455).
C, H. altispinus (ROM 702). D, Ouranosaurus nigeriensis (cast of GDF 300)............ 564
Figure E.21. Character 166. Boxplot showing the distribution of the the ratio between the
minimum mediolateral width of the rostral constriction and the maximum width of the
basisphenoid across the spheno-occipital tubercles for a sample of iguanodontoidean
the two variables of this raio were measured, exemplified in the parietal of Edmontosaurus
sp. (AMNH 427) ......................................................................................................... 566
Figure E.22. Character 167. Absence and presence (rectangle) of complete lateral osseous
closure of the ophthalmic sulcus of the laterosphenoid. A, Amurosaurus riabinini (cast of
AEHM 1/12), lateral view. B, Hypacrosaurus altispinus (CMN 8675), dorsal view.
Abbreviations: V1 = ophtalmic sulcus; V3 = mandibular sulcus .................................. 567
l
Figure E.23. Character 168. Great reduction of the length of the postorbital process
(rectangle) of the laterosphenoid. Neurocranium of Edmontosaurus regalis (CMN 2289) in
rostroventral view ....................................................................................................... 568
Figure E.24. Character 169. Orientation of the caudal surface (arrows) of the supraoccipital.
A, Tanius sinensis (PMU R240), caudodorsolateral view. B, Brachylophosaurus canadensis
(TMP 90.104.1), caudodorsolateral view..................................................................... 569
Figure E.25. Character 170. Absence and presence of a lateroventral corner of the
supraoccipital that is deeply inset into the exoccipital. A, Bactrosaurus johnsoni (AMNH
6365), caudal view. B, Corythosaurus intermedius (CMN 8676), caudal view ............ 570
Figure E.26. Character 171. Caudal extension of the exoccipital-supraoccipital shelf above
the foramen magnum. A, Hypacrosaurus altispinus (CMN 8675), ventral view. B,
Brachylophosaurus canadensis (MOR 1071-7-7-98-86), ventral view. C, Edmontosaurus
regalis (CMN 2289), ventral view............................................................................... 571
Figure E.27. Character 172. Absence and presence (rectangle) of complete lateral osseous
closure of the ophthalmic sulcus of the laterosphenoid. A, Charonosaurus jiayinensis (CUST
JV1251-57), caudal view. B, Tsintaosaurus spinorhinus (IVPP V725), caudal view.
Abbreviations: pp = paroccipital process ..................................................................... 572
Figure E.28. Character 173. Elevation (rectangles) of the proximodorsal region of the
quadrate wing of the pterygoid. A, Edmontosaurus regalis (CMN 2289), mediorostral view.
B, Brachylophosaurus canadensis (MOR 1071-7-23-98-387), mediorostral view........ 573
Figure E.29. Character 174. Ventral extension of the lamina (rectangles) located ventral to
the central buttress of the pterygoid. A, Gryposaurus sp. (MOR 399), mediorostral view. B,
Edmontosaurus regalis (CMN 2289), mediorostral view ............................................. 574
Figure E.30. Character 175. Presence and absence of ectopterygoid –jugal contact, as
evidenced on the medial side of the jugal. A, Iguanodon atherfieldensis (BMNH R5764),
medial view. B, Bactrosaurus johnsoni (CMN 6373), medial view.............................. 575
Figure F.1. Character 176. Lateral profile of the dorsal margin of the rostrum (brackets)
above the narial foramen or the bony naris. A, Brachylophosaurus canadensis (TMP
90.104.1), lateral view. B, Edmontosaurus regalis (CM 26258), lateral view............... 583
Figure F.2. Character 177. Boxplot showing the distribution of the angle between the dorsal
margin of the rostrum and the maxillary tooth row for a sample of iguanodontoidean
this angle was measured, exemplified in a subadult skull of Edmontosaurus annectens (CMN
8509)........................................................................................................................... 586
li
Figure F.3. Character 178. Exposure (A) and closure (B) of the nasal passages. A,
Prosaurolophus maximus (CMN 2277), lateral view. B, Corythosaurus intermedius (ROM
776), lateral view......................................................................................................... 587
Figure F.4. Character 179. Geometry of the narial foramen. A, BIC plot for various clustering
models of the NMMDS derived from the GDA dissimilariy matrix. B, coordinate projection
of the NMMDS data showing chosen clustering into three states. Abbreviations of clustering
models: EII = spherical, equal volume and shape; VII = sperical, variable volume, equal
shape; EEI = diagonal, equal volume and shape; VEI = diagonal, variable volume, equal
shape; EVI = diagonal, equal volume, variable shape; VVI = diagonal, variable volume and
shape; EEE = ellipsoidal, equal volume, shape and orientation; EEV = ellipsoidal, equal
volume and shape, variable orientation; VEV = ellipsoidal, variable volume, equal shape;
VVV = ellipsoidal, variable volume and shape (Fraley and Raftery, 2003) .................. 588
Figure F.5. Character 179. Geometry of the narial foramen (rectangle). A, Brachylophosaurus
canadensis (CMN 8893), lateral view. B, Edmontosaurus annectens (UCM 20892), lateral
view. C, Prosaurolophus maximus (TMP 84.1.1), lateral view .................................... 589
Figure F.6. Character 180. Degree of closure of the nasal passage on the lateral crest surface
between the caudoventral process of te premaxilla and the nasal. Arrows point to fontanellae.
A, Corythosaurus intermedius maximus (ROM 776), lateral view. B, Hypacrosaurus
altispinus (CMN 8501), lateral view............................................................................ 590
Figure F.7. Character 181. Boxplot showing the distribution of the ratio between the length
of the narial foramen and the distance between the rostroventral corner of the premaxilla and
the rostroventral margin of the prefrontal for a sample of iguanodontoidean ornithopods. The
of this ratio were measured, exemplified in the skull of Gryposaurus notabilis (CMN 2278)
.................................................................................................................................... 592
Figure F.8. Character 185. Caudal extent (arrows) of the nasal passages. A,
Brachylophosaurus canadensis (CMN 8893), lateral view. B, Corythosaurus intermedius
(UALVP 13), lateral view. C, Parasaurolophus cyrtocristatus (UCMP 143270), lateral view.
Abbreviation: cnsp = caudal extent of the nasal passage .............................................. 593
Figure F.9. Character 186. Composition of the caudal margin (rectangles) of the external
naris. A, Equijubus normani (IVPP V12534), lateral view. B, Brachylophosaurus canadensis
(MOR 794), lateral view. C, Lambeosaurus lambei (YPM 3222), lateral view. Abbreviations:
ns = nasal; pmx = premaxilla....................................................................................... 594
Figure F.10. Character 187. Absence and presence of circumnarial fossa (rectangle).
A, Corythosaurus intermedius (ROM 845), lateral view. B, Edmontosaurus regalis (ROM
801), lateral view......................................................................................................... 595
lii
Figure F.11. Character 188. Caudal extent of the circumnarial fossa. The arrows indicate the
lateral and caudal margins of the circumnarial fossa A, Iguanodon bernissartensis (IRSNB
1731), lateral view. B, Edmontosaurus regalis (ROM 801), lateral view. C, Maiasaura
peeblesorum (TCMI 2001.89.2), laterodorsal view (courtesy of Terry A. Gates) ......... 596
medial, lateral and(or caudal margins of the circumnarial fossa A, Brachylophosaurus
canadensis (MOR 794), dorsolateral view. B, Saurolophus angustirostris (unrecorded
catalog number), lateral view (photo courtesy of David Evans) ................................... 597
Figure F.13. Character 189. Degree of excavation of the caudal region (delimited by the
arrows) of the circumnarial fossa. A, Gryposaurus notabilis (ROM 873), lateral view. B,
Edmontosaurus regalis (FMNH P15004), lateral view. C, Prosaurolophus maximus (TMP
84.1.1), lateral view..................................................................................................... 598
Figure F.14. Character 190. Absence and presence (rectangle) of supracranial crest.
A, Edmontosaurus annectens (AMNH 5730), lateral view. B, Gryposaurus notabilis (MSNM
V345), lateral view...................................................................................................... 599
Figure F.15. Character 191. Composition of the supracranial crest. A, Brachylophosaurus

canadensis (MOR 1071-7-7-98-86), lateral view. B, Maiasaura peeblesorum (OTM F138),
lateral view. C, Hypacrosaurus stebingeri (MOR 549), lateral view. Abbreviations: f =
frontal; ns = nasal; pmx = premaxilla .......................................................................... 600
Figure F.16. Character 192. Relative contribution of the premaxilla and nasal bones to the
formation of hollow supracranial crests. A, Corythosaurus casuarius (ROM 871), lateral and
slightly caudal view. B, Lambeosurus magnicristatus (CMN 8705), lateral view.
Abbreviations: ns = nasal; pmx = premaxilla............................................................... 601
Figure F.17. Character 193. General morphology of supracranial crests (rectangles).

A, Ouranosaurus nigeriensis (GDF 300; photo courtesy of David B. Weishampel), lateral
view. B, Gryposaurus notabilis (CMN 2278), lateral view. C, Brachylophosaurus canadensis
(CMN 8893), laterodorsal view ................................................................................... 602

A, Saurolophus osborni (AMNH 5220), lateral view. B, Maiasaura peeblesorum (OTM
F138), lateral view. C, Brachylophosaurus canadensis (CMN 8893)........................... 603

A, Prosaurolophus maximus (CMN 2277), lateral view. B, Kritosaurus navajovius (BYU
12950), lateral view..................................................................................................... 604
Figure F.20. Character 193. General morphology of supracranial crests (arrows). A,

Corythosaurus casuarius (ROM 1933), lateral view. B, Parasaurolophus tubicen (NMMNH
P-25100), lateral view ................................................................................................. 605
liii
Figure F.21. Character 194. Hollow crest-snout angle along the dorsal margin of the
premaxilla in lateral view (in adults). A, Parasaurolophus walkeri (ROM 768), lateral view.
B, Hypacrosaurus altispinus (CMN 8501), lateral view. C, Corythosaurus casuarius (TMP
84.121.1), lateral view. D, Lambeosaurus lambei (FMNH PR380), lateral view .......... 606
Figure F.22. Character 195. Caudal extension of the caudoventral corner of the nasal in a
dome-shaped hollow supracranial crest. A, Hypacrosaurus altispinus (CMN 8501), lateral
view. B, Corythosaurus intermedius (ROM 776), lateral view. C, C. casuarius (ROM 1933),
lateral view. D, H. stebingeri (MOR 549), lateral view................................................ 607
Figure F.23. Character 196. Presence of palpebral bone (arrows). A, Iguanodon

bernissartensis (IRSNB 1726), lateral view. B, I. atherfieldensis (IRSNB 1551), lateral view
.................................................................................................................................... 608
Figure F.24. Character 197. Length/width proportions of the orbit. A, Iguanodon

atherfieldensis (BMNH 11521), lateral view. B, Brachylophosaurus canadensis (MOR 794),
lateral view. Abbreviation: O = orbit ........................................................................... 609
Figure F.25. Character 198. Absence and presence of paraquadratic foramen (rectangle).
A, Maiasaura peeblesorum (ROM 44770), lateral view. B, Equijubus normani (IVPP
V12534), lateral view.................................................................................................. 610
Figure F.26. Character 199. Height and width of the infratemporal fenestra relative to
those of the orbital cavity. A, Gryposaurus notabilis (CMN 2278), lateral view. B,
Corythosaurus intermedius (CMN 8503), lateral view. Abbreviations: itf = infratemporal
fenestra; o = orbit ........................................................................................................ 611
Figure F.27. Character 200. Shape and rostrocaudal breadth of the dorsal margin of the
infratemporal fenestra relative to that of the dorsal orbital margin. State 0:subrectangular
lateral profile, with a dorsal infratemporal margin that is approximately as wide as the ventral
margin. This figure shows the outlines of a number of hadrosaurid taxa in right lateral view.
A, Corythosaurus intermedius (CMN 8503). B, C. intermedius (CMN 8704). C, C.
intermedius (ROM 845). D, C. intermedius (TMP 80.23.4). E, Corythosaurus sp. (ROM
870). F, C. casuarius (AMNH 5240). G, C. casuarius (AMNH 5338). H, C. casuarius (ROM
871). I, Gryposaurus monumentensis (RAM 6797). J, G. notabilis (CMN 2278). K, G.
notabilis (MSNM V345). L, G. notabilis (ROM 873). M, G. notabilis (TMP 80.22.1). N,
Hypacrosaurus altispinus (CMN 8501). O, H. altispinus (ROM 789). P, Kritosaurus
navajovius (AMNH 5799). Q, K. navajovius (NMMNH P-16106). R, Lambeosaurus lambei
(AMNH 5340, juvenile). S, L. lambei (CMN 8703). T, L. lambei (CMN 8633, subadult). U,
L. lambei (AMNH 5353). V, L. lambei (AMNH 5373). W, L. lambei (ROM 1218 ...... 612
infratemporal fenestra relative to that of the dorsal orbital margin. State 1: subtriangular, with
a dorsal infratemporal margin that is narrower than the ventral margin. This figure shows the
liv
outlines of a number of hadrosaurid taxa in right lateral view. A, Brachylophosaurus
canadensis (MOR 794). B, Edmontosaurus annectens (CMN 8399). C, E. annectens (CMN
8509, subadult). D, E. annectens (DMNH 1943). E, E. annectens (SM R4036). F, E. regalis
(AMNH 5254). G, E. regalis (CM 26258). H, E. regalis (CMN 2288). I, E. regalis (ROM
801). J, E. regalis (FMNH P15004). K, Parasaurolophus walkeri (ROM 768). L,
Prosaurolophus maximus (CMN 2870). M, P. maximus (TMP 84.1.1). N, Saurolophus
osborni (AMNH 5220)................................................................................................ 613
Figure F.29. Character 201. Location of the dorsal margin of the infratemporal fenestra
relative to the dorsal margin of the orbit . A, Brachylophosaurus canadensis (TMP 90.104.1),
lateral view. B, Gryposaurus notabilis (CMN 2278), lateral view. C, Hypacrosaurus
altispinus (CMN 8501), lateral view............................................................................ 614
Figure 30. Character 202. Shape of the supratemporal fenestra. Outlines drawn in dorsal
view; the lateral side of the skull is to the left of the drawings. A, Bactrosaurus johnsoni
(PIN 2549-1. B, Edmontosaurus annectens (AMNH 5730). C, E. annectens (ROM 867,
subadult). D, E. annectens (SM R4036). E, Gryposaurus notabilis (AMNH 5350). F, G.
notabilis (CMN 2278). G, G. notabilis (ROM 764). H, Kritosaurus navajovius (NMMNH P-
16106, subadult). I, Maiasaura peeblesorum (OTM F138). J, M. peeblesorum (YPM-PU
22405). K, Prosaurolophus maximus (CMN 2870). L, P. maximus (ROM 787). M, P.
maximus (TMP 84.1.1). N, Secernosaurus koerneri (MACN-RN 144). O, Aralosaurus
tuberiferus (PIN 2229). P, cf. Lambeosaurus (USNM 10309). Q, Hypacrosaurus altispinus
(CMN 8675). R, H. altispinus (ROM 702). S, H. stebingeri (MOR 549). T, Jaxartosaurus
aralensis (PIN 1/50009). U, Ouranosaurus nigeriensis (GDF 300). V, H. altispinus (IVPP
V725).......................................................................................................................... 615
Figure F.31. Character 203. Boxplot showing the distribution of the ratio between the height
and length of the skull for a sample of iguanodontoidean ornithopods. The inserted smaller
were measured, exemplified in the skull of Corythosaurus intermedius (CMN 8704) .. 618
Figure F.32. Character 204. Boxplot showing the distribution of the ratio between the
maximum transverse width of the cranium across the postorbitals and the width across the
quadrate cotylus of the squamosals for a sample of iguanodontoidean ornithopods. The
of this raio were measured, exemplified in a skull roof of Hypacrosaurus altispinus (CMN
8675)........................................................................................................................... 621
Figure F.33. Character 205. Shape of the occiput and orientation of the quadrates in caudal
view. A, Iguanodon bernissartensis (cast of IRSNB 1534), caudal view. B,
Brachylophosaurus canadensis (CMN 8893), caudal view .......................................... 622
lv
Figure G.1. Character 206. Morphology of the dorsal flange of the axis. The arrow indicates
an embayment that separates the dorsal falnge from the postzygapophyseal region of the axis.
A, Indeterminate hadrosaurine (USNM 8629), lateral view. B, Brachylophosaurus
canadensis (MOR 794), lateral view. Abbreviation: df = dorsal flange........................ 626
Figure G.2. Character 207. Development of the postzygapophyseal procesess of cranial

and middle cervical vertebrae. A, Gilmoreosaurus mongoliensis (AMNH 6551), dorsal view.
B, Brachylophosaurus canadensis (CMN 8893), dorsal view. Abbreviations: na = neural arch
length; po = postzygapophyseal process length............................................................ 627
Figure G.3. Character 209. Boxplot showing the distribution of the ratio between the height
of the neural spine and that of the centrum of the tallest posterior dorsal or sacral vertebrae
(in adults) for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows
can be classified. The photograph shows how the two variables of this raio were measured,
exemplified in a dorsal vertebra of Bactrosaurus johnsoni (AMNH 6553)................... 629
Figure G.4. Character 210. Absence (A) and presence (B) of slightly elongated neural spines
in the cranial dorsal vertebrae (rectangles), forming a “wither-like” region above the pectoral
girdle. A, Edmontosaurus annectens (AMNH 5730; photo courtesy of David B.
Weishampel), lateral view. B, Lambeosaurus lambei (TMP 82.38.1), lateral view....... 630
Figure G.5. Ventral view of the sacrum of Iguanodon bernissartensis (IRSNB 1723)
showing the co-occurrence of a ventral sulcus and ridge (arrows) in the same specimen.
Abbreviations: r = ridge; s = sulcus ............................................................................. 631
Figure G.6. Character 212. Chevron length relative to the length of the neural spines in the
caudal vertebrae of the proximal half of the tail. A, Parasaurolophus cyrtocristatus (FMNH
P27393), lateral view. B, Edmontosaurus annectens (AMNH 5730), lateral view.
Abbreviations: c = chevron length; n = neural spine length.......................................... 635
Figure G.7. Shape of the craniomedial plate and the caudoventral process of various
iguanodontoidean sternals. Outlines drawn in cranioventral view.. A, Lambeosaurinae
indeterminate (ANSP 16969). B, Bactrosaurus johnsoni (). C, Brachylophosaurus
canadensis (MOR 794). D, B. canadensis (MOR 1071). E, B. canadensis (UCMP 130139).
F, Corythosaurus sp. (CMN 36141). G, Edmontosaurus sp. (LACM 23504, subadult). H,
Lambeosaurinae indeterminate (USNM 11950). I, Corythosaurus sp. (ROM 1947). J, E.
annectens (YPM 2182). K, E. regalis (CMN 2289). L, Hadrosauridae indeterminate (USNM
2413). M, Iguanodon atherfieldensis (IRSNB 1551). N, I. bernissartensis (IRSNB 1534). O,
I. bernissartensis (IRSNB 1535). P, I. bernissartensis (IRSNB 1725). Q, Iguanodon sp.
(BMNH R2218). R, Lophorhothon atopus (AUMP 2295). S, Lambeosaurus lambei (AMNH
5340, juvenile). T, Olorotitan ararhensis (AEHM 2/845). U, Prosaurolphus maximus (ROM
787). V, Tsintaosaurus spinorhinus (IVPP V725)........................................................ 636
lvi
Figure G.8. Character 213. Length of the “handle-like” caudolateral process of the sternal
relative to that of the craniomedial plate (excluding the caudoventral process). A, Iguanodon
sp. (FMNH R2218), rostroventral view. B, Olorotitan ararhensis (AEHM 1/845),
rostroventral view ....................................................................................................... 637
Figure H.1. Character 214. Coracoid size relative to the length of the scapula. A, Iguanodon
bernissartensis (IRSNB 1534), lateral view. B, Gryposaurus latidens (AMNH 5465), medial
view. Abbreviations: cor = coracoid; scp = scapula ..................................................... 646
Figure H.2. Character 215. Boxplot showing the distribution of the ratio between the length
of the lateral margin of the facet for the scapular articulation and the length of the lateral
margin of the glenoid for a sample of iguanodontoidean ornithopods. The inserted smaller
were measured, exemplified in an indeterminate hadrosaurid coracoid (TMP 80.29.1) ......
.................................................................................................................................... 648
Figure H.3. Character 216. Boxplot showing the distribution of the angle between the lateral
margins of the scapular facet and the glenoid for a sample of iguanodontoidean ornithopods.
numbers of clusters in which the data can be classified. The photograph shows how the two
variables of this raio were measured, exemplified in an indeterminate hadrosaurid coracoid
(TMP 80.29.1)............................................................................................................. 650
Figure H.4. Character 217. Morphology of the craniomedial margin (arrows) of the coracoid.
Specimens shown in lateral view. A, Lophorhothon atopus (AUMP 2295). B, Iguanodon sp.
(BMNH R6471). C, I. bernissartensis (IRSNB 1534). D, “Lambeosaurus laticaudus“
(LACM 17715). E, Secernosaurus koerneri (MACN-RN 2). F, cf. Hypacrosaurus stebingeri
(MOR 355-8-28-5-12). G, Olorotitan ararhensis (AEHM 2/845). H, Lambeosaurine
indeterminate (USNM 11950) ..................................................................................... 651
Figure H.5. Character 218. Boxplot showing the distribution of the ratio between the
dorsoventral depth and the breadth of the ventral process of the coracoid for a sample of
photograph shows how the two variables of this raio were measured, exemplified in an
indeterminate hadrosaurid coracoid (TMP 92.36.983) ................................................. 653
Figure H.6. Character 219. Degre of curvature of the ventral process of the coracoid (arrows).
Specimens shown in lateral view. A, Iguanodon sp. (BMNH R118). B, Ouranosaurus
nigeriensis (cast of GDF 300). C, Brachylophosaurus canadensis (CMN 8893). D, B.
canadensis (UCMP 130139). E, Edmontosaurus sp. (MOR 601). F, cf. Hypacrosaurus sp.
(MOR 609-89-177) ..................................................................................................... 654
lvii
Figure H.7. Character 220. Lateral profile of the dorsal margin of the scapula. A,
Bactrosaurus johnsoni (AMNH 6553), lateral view. B, Brachylophosaurus canadensis
(MOR 1071-7-18-98-298), lateral view)...................................................................... 655
dorsoventral depth of the proximal region of the scapula and its craniocaudal length for a
sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian
exemplified in a scapula of Secernosaurus koerneri (MACN-RN 142)........................ 658
Figure H.9. Character 222. Absence and presence of a convexity along the ventral margin of
the distal blade of the scapula. A, Secernosaurus koerneri (MACN-RN 142), lateral view.
B, Corythosaurus intermedius (ROM 845), lateral view .............................................. 659
dorsoventral depth of the distal end of the scapular blade and the depth of the proximal
region of the scapula for a sample of iguanodontoidean ornithopods. The inserted smaller
were measured, exemplified in a scapula of Lambeosaurus lambei (ROM 1218)......... 662
dorsoventral width of the proximal constriction and the dorsoventral depth of the cranial end
of the scapula for a sample of iguanodontoidean ornithopods. The inserted smaller graph
data can be classified. The photograph shows how the two variables of this raio were
measured, exemplified in a scapula of Gryposaurus latidens (AMNH 5465) ............... 665
Figure H.12. Characters 225 and 226. Morphology and orientation of the acromion process
(arrows) of the scapula. Specimens shown in lateral view. A, Iguanodon atherfieldensis
(IRSNB 1551). B, Parasaurolophus walkeri (ROM 768). C, Tanius sinensis (PMU R241). D,
Bactrosaurus johnsoni (AMNH 6553). E, Corythosaurus intermedius (CMN 8704). F,
Charonosaurus jiayinensis (CUST JII1398). G, Lambeosaurinae (? Hypacrosaurus
stebingeri, TMP 87.79.220; juvenile specimen). H, Maiasaura peeblesorum (ROM 44770).
I, Brachylophosaurus canadensis (UCMP 130139). J, Edmontosaurus annectens (SM
R4036). Exemplars A through G document state 0 of character 225; exemplars H through J
document state 1 of character 225. Specimens A through C document state 0 of character
225; specimens D through G document state 1 of character 226 .................................. 666
distance from the coracoid joint and the cranial end of the acromion process and the height
between this and the ventral apex of the glenoidal facet for a sample of iguanodontoidean
lviii
the two variables of this raio were measured, exemplified in a scapula of Secernosaurus
koerneri (MACN-RN 142) .......................................................................................... 669
Figure H.14. Character 228. Development of the deltoid ridge (arrows) of the scapula.
A, Hypacrosaurus stebingeri (MOR 549), lateral view. B, Secernosaurus koerneri (MACN-
RN 146), lateral view .................................................................................................. 670
proximodistal length of the deltopectoral crest and the proximodistal length of the humerus
for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the
exemplified in a humerus of Gryposaurus latidens (AMNH 5465) .............................. 673
Figure H.16. Character 230. Boxplot showing the distribution of the ratio between the width
of the humerus across the distal fourth of the deltopectoral crest and the width of the distal
shaft at the point of maximum curvature for a sample of iguanodontoidean ornithopods. The
of this raio were measured, exemplified in a humerus of Brachylophosaurus canadensis
(CMN 8893)................................................................................................................ 676
Figure H.17. Character 231. Boxplot showing the distribution of the angle between the
cranial margin of the crest and the proximalmost margin of the cranial curvature of the
humeral shaft for a sample of iguanodontoidean ornithopods. The inserted smaller graph
data can be classified. The photograph shows how this angle was measured, exemplified in a
humerus of Tanius sinensis (PMU R235) .................................................................... 679
Figure H.18. Character 232. Boxplot showing the distribution of the ratio between the total
length of the humerus and the width of the lateral surface of the proximal end of the element
classified. The photograph shows how this angle was measured, exemplified in a humerus of
Tsintaosaurus spinorhinus (IVPP V725) ..................................................................... 682
Figure H.19. Character 233. Boxplot showing the distribution of the length of the ulna
relative to its dorsoventral thickness (measured at mid-shaft) for a sample of
photograph shows how the two variable of this ratio were measured, exemplified in a ulna of
Parasaurolopus walkeri (ROM 768) ........................................................................... 685
lix
of the ulna and the length of the humerus for a sample of iguanodontoidean ornithopods. The
of clusters in which the data can be classified. The photograph shows how the two variable of
this ratio were measured, exemplified in a forelimb of Brachylophosaurus canadensis (MOR
794) ............................................................................................................................ 687
Figure H.21. Character 235. Composition of the carpus (rectangle and arrows). A, Iguanodon
bernissartensis (IRSNB 1535), dorsal view. B, Brachylophosaurus canadensis (MOR 794),
proximolateral view .................................................................................................... 688
Figure H.22. Character 236. Absence and presence of manual digit I. A, Iguanodon
bernissartensis (IRSNB 1535), dorsal view. B, Hadrosauridae indeterminate (TMP 84.36.29)
.................................................................................................................................... 689
and the width at mid-shaft of metacarpal III for a sample of iguanodontoidean ornithopods.
numbers of clusters in which the data can be classified. The photograph shows how the two
variable of this ratio were measured, exemplified in a metacarpal III of Tsintaosaurus
spinorhinus (IVPP V725) ............................................................................................ 691
Figure H.24. CHaracter 238. Proximodistal elongation of metacarpal V. A, Bactrosaurus

johnsoni (SBDE 95E). B, Maiasaura peeblesorum (ROM 44770) .............................. 692
Figure H.25. Character 239. Expansion of the proximal surface of metatarsal V. A,

Iguanodon bernissartensis (IRSNB 1535). B, Parasaurolophus walkeri (ROM 768) .. 693
Figure H.26. CHaracter 240. Length/width proportions of manual phalanx III1. A,

Parasaurolophus walkeri (ROM 768). B, Corythosaurus intermedius (ROM 845). C,
Maiasaura peeblesorum (ROM 44770) ....................................................................... 694
Figure H.27. Character 241. Shape of manual ungual II. A, Iguanodon bernissartensis
(IRSNB 1534). B, Brachylophosaurus canadensis (MOR 794) ................................... 695
Figure H.28. Character 242. Proximodistal length of manual palanx II1 relative to that of II2.
A, Iguanodon atherfieldensis (IRSNB 1551). B, Lambeosaurus lambei (CMN 8703).. 696
Figure I.1. Character 243. Boxplot showing the distribution of the angle of ventral deflection
of the preacetabular process of the ilium for a sample of iguanodontoidean ornithopods. The
of clusters in which the data can be classified. The photograph shows how this angle was
measured, exemplified in the ilium of Barsboldia sicinskii (ZPAL MgD-I 110)........... 711
lx
Figure I.2. Character 244. Boxplot showing the distribution of the ratio between the length of
the preacetabular process and that of the central plate of the ilium for a sample of
iguanodontoidean ornithopods. The ilium of Ouranosaurus nigeriensis (GDF 300) is
excluded from the the cluster analysis because it is very autapomorphic. The inserted smaller
which the data can be classified. The photograph shows how the two variable of this ratio
were measured, exemplified in an ilium of Brachylophosaurus canadensis (MOR 794) ....
.................................................................................................................................... 714
Figure I.3. Bivariate plot between two axes of the NMMDS analysis of a GDA dissimilarity
matrix of the geometry of the preacetabular process of the ilium. Each dot is surrounded by
the outline of the preacetabular process of the specimen that it represents. Colors indicate the
following groups: green = non-hadrosaurid Iguanodontoidea; blue = non-lambeosaurine
Hadrosauridae; red = Lambeosaurinae; yellow = Hadrosauridae indeterminate. The inserted
smaller plot shows the Bayesian Information Criterion (BIC) for various numbers of clusters
in which the data may be classified. Cluster model identifiers in the BIC plot are as in Fig.
F.4............................................................................................................................... 717
Figure I.4. Character 245. Boxplot showing the distribution of the a ratio between the depth
of the proximal region of the preacetabular process and the dorsoventral distance between the
pubic peduncle and the dorsal margin of the ilium for a sample of iguanodontoidean
ornithopods.The inserted smaller graph shows the Bayesian information criterion (BIC) for
the two variable of this ratio were measured, exemplified in an ilium of Prosaurolophus
maximus (ROM 787)................................................................................................... 720
Figure I.5. Character 246. Boxplot showing the distribution of the ratio between the depth of
the central plate of the ilium and the distance between the pubic peduncle and the
caudodorsal prominence of the ischial peduncle, for a sample of iguanodontoidean
the two variable of this ratio were measured, exemplified in the ilium of Secernosaurus
koerneri (MACN-RN 2) .............................................................................................. 723
Figure I.6. Character 247. Position of the apex (X) of the supraacetabular process relative to
that of the caudal tuberosity (white solid circle) of the ischial peduncle. A, Gilmoreosaurus
mongoliensis (AMNH 6551), lateral view. B, Brachylophosaurus canadensis (MOR 794),
lateral view ................................................................................................................. 724
Figure I.7. Character 248. Lateroventral expansion of the supraacetabular process of the
ilium. A, Ouranosaurus nigeriensis (GDF 300), lateral view. B, Bactrosaurus johnsoni
(SBDE 95E), lateral view. C, Prosaurolophus maximus (ROM 787), lateral view. D,
Parasaurolophus cyrtocristatus (FMNH P27393), lateral view. Abbreviations: S =
supraacetabular process ............................................................................................... 725
lxi
Figure I.8. Character 249. Boxplot showing the distribution of the ratio between the breadth
of the supraacetabular process along its dorsal region and the craniocaudal length of the
central iliac plate from the caudal ischial peduncle to the pubic peduncle, for a sample of
photograph shows how the two variable of this ratio were measured, exemplified in a right
ilium of Prosaurolophus maximus (MOR 787)............................................................ 728
Figure I.9. Character 250. Asymmetry and symmetry of the lateroventral margin of the
supraacetabular process of the ilium. A, Maiasaura peeblesorum (ROM 44770), lateral view.
B, Tanius sinensis (PMU R242), lateral view. B, Gryposaurus latidens (AMNH 5465),
lateral view. C, Hypacrosaurus stebingeri (MOR 549), lateral view ............................ 729
Figure I.10. Character 251. Morphology of the lateroventral margin of the supraacetabular
process of the ilium. A, Iguanodon bernissartensis (IRSNB 1551), lateral view. B,
Bactrosaurus johnsoni (SBDE 95E), lateral view. C, Prosaurolophus maximus (TMP 84.1.1),
lateral view. D, Brachylophosaurus canadensis (MOR 794), lateral view; arrows point to the
two poorly defined lobes of the process....................................................................... 730
Figure I.11. Character 252. Presence of a ridge (arrows) uniting the caudodorsal margin of
the supraacetabular process and the proximodorsal margin of the postacetabular process of
the ilium. A, Gilmoreosaurus mongoliensis (AMNH 6551), lateral view. B, “Hypacrosaurus”
stebingeri (MOR 549), lateral view. C, Secernosaurus koerneri (MACN-RN 2), lateral view.
D, cf. Parasaurolophus sp. (TMP 92.53.21), lateral view ............................................ 731
Figure I.12. Character 252. Absence of demarcated margin between the caudal margin of the
supraacetabular process and the proximodorsal margin of the postacetabular process. A,
Brachylophosaurus canadensis (MOR 794), lateral view. B, Gryposaurus notabilis (ROM
764), lateral view. C, Saurolophus angustirostris (unrecorded number), lateral view. D,
Hadrosauridae indeterminate (YPM-PU 16970; ?Gryposaurus), lateral view............... 732
Figure I.13. Bivariate plot of two axes of the NMMDS analysis of a GDA dissimilarity
matrix of the geometry of the pubic peduncle of the ilium of 52 iguanodontoidean specimens.
Each dot is surrounded by the outline (rotated 90º) of the pubic peduncle of the specimen that
it represents. Colors indicate the following groups: green = non-hadrosaurid
Iguanodontoidea; blue = non-lambeosaurine Hadrosauridae; red = Lambeosaurinae; yellow =
Hadrosauridae indeterminate. The inserted smaller plot shows the Bayesian Information
Criterion (BIC) for various numbers of clusters in which the data may be classified.
Abbreviations in the BIC plot are as in Fig. F.4........................................................... 736
Figure I.14. Character 253. Morphology of the pubic peduncle of the ilium. A, Iguanodon sp.
(BMNH R9290), lateral view. B, Brachylophosaurus canadensis (MOR 794), lateral view
.................................................................................................................................... 737
lxii
Figure I.15. 3D plot of three axes of the NMMDS analysis of a GDA dissimilarity matrix of
the geometry of the ischial peduncle of the ilium in 53 iguanodontoidean specimens. Each
dot is surrounded by the outline of the preacetabular process of the specimen that it
repersents.Colors indicate the following groups: green = non-hadrosaurid Iguanodontoidea;
blue = non-lambeosaurine Hadrosauridae; red = Lambeosaurinae; yellow = Hadrosauridae
indeterminate. The inserted smaller plot shows the Bayesian Information Criterion (BIC) for
various numbers of clusters in which the data may be classified. Abbreviations in the BIC
plot are as in Fig. F.4................................................................................................... 740
Figure I.16. Character 254. Morphology of the ischial peduncle (arrows) of the ilium. A,
Iguanodon sp. (BMNH R113), lateral view. B, Iguanodon sp. (BMNH R114), lateral view.
C, Tanius sinensis (PMU R242), lateral view. D, Bactrosaurus johnsoni (SBDE 95E/25),
lateral view. E, Gilmoreosaurus mongoliensis (AMNH 6551), lateral view. F,
Lambeosaurinae indeterminate (USNM 11950; ?Corythosaurus sp.), lateral view....... 741
Figure I.17. Character 255. Boxplot showing the distribution of the ratio between the length
of the postacetabular process and that of the central plate of the ilium, for a sample of
photograph shows how the two variable of this ratio were measured, exemplified in an ilium
of Brachylophosaurus canadensis (MOR 794) ............................................................ 744
Figure I.18. Character 256. Presence and absence of brevis shelf (arrow and rectangle) at the
base of the iliac postacetabular process. A, Bactrosaurus johnsoni (SBDE 95E), medial view.
B, ventral view of SBDE 95E. C, Brachylophosaurus canadensis (MOR 1071-8-2-98-469),
medial view. D, ventral view of MOR 1071-8-2-98-469.............................................. 745
Figure I.19. Character 257. Medial ridge (arrows) of the postacetabular process and
orientation of the brevis shelf (when present). The rectangles highlight the caudal region of
the postacetabular process in dorsal view. A, Hypacrosaurus stebingeri (MOR 549), medial
view. B and C, Gilmoreosaurus mongoliensis (AMNH 6551), medial and dorsal views,
respectively. D and E, Hypacrosaurus stebingeri (AMNH 5204), medial and dorsal views,
respectively. F and G, Secernosaurus koerneri (MACN-RN 2), medial and dorsal views,
respectively ................................................................................................................. 746
Figure I.20. Character 258. Absence and presence of a median ridge on the laterodorsal
surface of the postacetabular process of the ilium. A, Hypacrosaurus altispinus (AMNH
5204), caudal view. B, Secernosaurus koerneri (MACN-RN 2), caudal view .............. 747
Figure I.21. Character 259. Bivariate plot between two axes of the NMMDS analysis of a
GDA dissimilarity matrix of the geometry of the postacetabular process of the ilium. Each
represents. Colors indicate the following groups: green = non-hadrosaurid Iguanodontoidea;
lxiii
Figure I.22. Character 259. Lateral profile of the postacetabular process of the ilium.
A, Iguanodon atherfieldensis (IRSNB 1551), lateral view. B, Lambeosaurus lambei (ROM
1218)........................................................................................................................... 751
Figure I.23. Character 260. Relative orientation of the dorsal and ventral margins of the
lateral or laterodorsal surface of the postacetabular process of the ilium. A, Hadrosauridae
indeterminate (UTEP P37.7.222), lateral view. B, Tsintaosaurus spinorhinus (IVPP V725)
.................................................................................................................................... 752
Figure I.24. Character 261. Position of the medial sacral ridge (arrows) within the medial
surface of the central iliac plate. A, Iguanodon sp. (BMNH R113), medial view. B,
Bactrosaurus johnsoni (cast of SBDE 95E), medial view ............................................ 753
Figure I.25. Character 262. Orientation of the medial sacral ridge (arrows) of the ilium. A,
Ouranosaurus nigeriensis (GDF 300), medial view. B, Tanius sinensis (PMU R242), medial
view. C, Barsboldia sicinskii (ZPAL MgD-I 110), medial view. D, Brachylophosaurus
canadensis (MOR 1071-8-2-98-469), medial view ...................................................... 754
Figure I.26. Character 263. Lateral profile of the dorsal margin of the ilium above the
supraacetabular process. The arrow indicates the presence of a concave dorsal margin. A,
Iguanodon bernissartensis (IRSNB 1534), lateral view. B, Charonosaurus jiayinensis
(CUST JIII430) ........................................................................................................... 755
Figure I.27. Character 264. Orientation of the dorsal margin of the prepubic process of the
pubis. A, Ouranosaurus nigeriensis (GDF 300), lateral view. B, Prosaurolophus maximus
(MOR 454) ................................................................................................................. 756
the geometry of the prepubic process of the pubis for 36 iguanodontoidean specimens. Each
indeterminate. Pubic profiles for each point are facing rostrolaterally towards the reader. The
inserted smaller plot shows the Bayesian Information Criterion (BIC) for various numbers of
clusters in which the data may be classified. Abbreviations in the BIC plot are as in Fig. F.4
.................................................................................................................................... 758
Figure I.29. Character 265. Lateral profile of the prepubic process of the pubis. A, Iguanodon
sp. (BMNH R2194), lateral view. B, Lambeosaurus laticaudus (LACM 26874).......... 759
lxiv
Figure I.30. Character 265. Geometry of the distal blade of the prepubic process of the pubis.
A, Brachylophosaurus canadensis (MOR 1071- 8-20-98-587), lateral view. B,
Edmontosaurus regalis (CMN 2289), lateral view. B, Gryposaurus notabilis (ROM 764),
lateral view ................................................................................................................. 760
Figure I.31. Character 266. Depth of the dorsoventral expansion of the distal region of the
prepubic process relative to the width of the acetabular margin of the pubis. A,
Secernosaurus koerneri (MACN-RN 2), right medial view. B, cf. Corythosaurus sp.
(AMNH 3971)............................................................................................................. 761
Figure I.32. Character 267. Length of the distal blade (d) of the prepubic process relatve to
that of the prepubic constriction (n). A, Lambeosaurinae (?Corythosaurus sp.; GDF 300),
lateral view. B, Secernosaurus koerneri (MACN-RN 2), lateral view. C, Bactrosaurus
johnsoni (SBDE 95E5/43), lateral view....................................................................... 762
Figure I.33. Character 268. relative position of maximum concavity of the dorsal (asterisk)
and ventral (circle) margins of the prepubic process of the pubis. A, Ouranosaurus
nigeriensis (GDF 300), lateral view. B, Gryposaurus latidens (AMNH 5465) ............. 763
Figure I.34. Character 269. Morphology of the acetabular margin of the pubis. A,
Gilmoreosaurus mongoliensis (AMNH 6551), lateral view. The rectangle surrounds a rudge
that connects the iliac peduncle with the proximal region of the ischial peduncle. B, cf.
Tsintaosaurus spinorhinus (IVPP K149) ..................................................................... 764
Figure I.35. Character 270. Presence and absence of an obturator gutter or foramen ventral to
the ischial peduncle of the pubis. A, Iguanodon sp. (BMNH R28504), lateral view. B,
Edmontosaurus sp. (LACM 7241). Abbreviations: ip = ischial peduncle; og = obturator
gutter........................................................................................................................... 765
Figure I.36. Character 271. Boxplot showing the distribution of the length/width ratio of the
ischial peduncle of the pubis for a sample of iguanodontoidean ornithopods. The inserted
smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters
in which the data can be classified. The photograph shows how the two variable of this ratio
were measured, exemplified in a pubis of an indeterminate hadrosaurid (TMP 1966.27.24)
.................................................................................................................................... 767
Figure I.37. Character 272. Lateroventral protuberance (arrow) on the proximal region of the
ischial peduncle of the pubis A, Prosaurolophus maximus (MOR 454), lateral view. B,
Hypacrosaurus stebingeri (MOR 549) ........................................................................ 768
Figure I.38. Character 273. Depth/width proportions of the iliac peduncle (rectangles) of the
pubis. A, Iguanodon sp. (BMNH R2194), lateral view. B, Corythosaurus intermedius (ROM
845) ............................................................................................................................ 769
lxv
Figure I.39. Character 274. Boxplot showing the distribution of the ratio between the distance
from the acetabular margin to the distal border of the prepubic process and the distance from
the dorsal margin of the iliac peduncle to the ventral border of the proximal postpubic shaft,
classified. The photograph shows how the two variable of this ratio were measured,
exemplified in a pubis of Bactrosaurus johnsoni (SBDE 95E5/43).............................. 772
Figure I.40. Character 275. Morphology of the distal region of the iliac peduncle of the
ischium. A, BIC plot for various clustering models of the NMMDS derived from the GDA
dissimilariy matrix. B, coordinate projection of the NMMDS data showing chosen clustering
into three states. Abbreviations of clustering models. Abbreviations for the BIC plot are as in
Fig. F.4 ....................................................................................................................... 775
Figure I.41. Character 275. Degree of development of a curved distal margin (rectangles) of
the iliac peduncle of the ischium. A, Brachylophosaurus canadensis (MOR 794), lateral
view. B, Iguanodon atherfieldensis (BMNH 11521), lateral view. C, B, Tsintaosaurus
spinorhinus (IVPP V725), lateral view ........................................................................ 776
Figure I.42. Character 276. Boxplot showing the distribution of the angle between the distal
articular surface of the iliac peduncle of the ischium and the acentabular margin of the
peduncle, for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows
can be classified. The photograph shows how this angle was measured, exemplified in an
ischium of Lambeosaurus lambei (ROM 1218) ........................................................... 779
and the width of the iliac peduncle of the ischium for a sample of iguanodontoidean
the two variable of this ratio were measured, exemplified in an inchium of Charonosaurus
jiayinensis (CUST JV679)........................................................................................... 782
Figure I.44. Character 278. Relative orientation of the acetabular and caudodorsal margins of
the iliac peduncle of the ischium A, Iguanodon sp. (BMNH R51623), lateral view. B,
Brachylophosaurus canadensis ( MOR 454), lateral view ........................................... 783
Figure I.45. Character 279. Orientation of the craniocaudal axis of the pubic peduncle
(perpendicular to its articular margin) relative to the ischial shaft for a sample of
photograph shows how the two variable of this ratio were measured, exemplified in an
ischium of Corythosaurus intermedius (ROM 845) ..................................................... 785
lxvi
Figure I.46. Character 280. Length/width proportions of the pubic peduncle of the ishium.
A, Iguanodon atherfieldensis (BMNH R11521), lateral view. B, Tsintaosaurus spinorhinus
(IVPP V725). C, Secernosaurus koerneri (MACN-RN 2), lateral view........................ 786
Figure I.47. Character 281. Position of the dorsal acetabular margin of the pubic peduncle
relative to the dorsal margin of the ischial shaft. A, Velafrons coahuilensis (CPC-59), lateral
view. B, Brachylophosaurus canadensis (MOR 794), lateral view............................... 787
Figure I.48. Character 282. Boxplot showing the distribution of the ratio between the
dorsoventral width of the midshft of the ischium and the length of the ischial shaft for a
sample of iguanodontoidean ornithopods. The proximal end of the shaft is considered to be
located where the dorsal profile of the ischium begins to rise to form the base of the iliac
peduncle.The inserted smaller graph shows the Bayesian information criterion (BIC) for
the two variables of this ratio were measured, exemplified in an ishium of Parasaurolophus
cyrtocristatus (FMNH P27393) ................................................................................... 789
the geometry of the prepubic process of the pubis for 36 iguanodontoidean specimens. Each
Figure I.50. Character 283. Morphology of the distal end of the ischium. A, Shantungosaurus
giganteus (GMV 1780), lateral view. B, Hypacrosaurus altispinus (CMN 8501), lateral view
.................................................................................................................................... 792
of the distal expansion of the ischium and that of the ischial shaft for a sample of
photograph shows how the two variables of this ratio were measured, exemplified in a
ischium of Lambeosaurus laticaudus (LACM 17715) ................................................. 794
Figure I.52. Character 285. Morphology of the cranial margin of the ventral expansion of the
distal ischial shaft. A, Iguanodon sp. (BMNH R9290), lateral view. B, Parasaurolophus
cyrtocristatus (FMNH P27393), lareral view............................................................... 795
Figure I.53. Character 286. Orientation of the long axis of the distal expansion of ths ischium
relative to the ischial shaft. A, Lambeosaurus lambei (ROM 1218), lateral view. B,
Lambeosurinae (?Corythosaurus sp.; USNM 15581), lateral view............................... 796
lxvii
Figure J.1. Character 287. Degree of curvature of the distal half of the femoral shaft.
A, Iguanodon sp. (BMNH R2734), laterocaudal view. B, Hadrosaurus foulkii (ANSP
10006), lateral view..................................................................................................... 800
Figure J.2. Character 288. Lateral profile of the caudoventral margin (rectangles) of the
fourth trochanter of the femur. A, Iguanodon bernissartensis (IRSNB 1731), lateral view. B,
Bactrosaurus johnsoni (AMNH 6553), lateral view..................................................... 801
Figure J.3. Character 289. Distal extension of the cnemial crest (rectangles) of the tibia.
A, Iguanodon bernissartensis (unrecorded catalog number), medial view. B,
Parasaurolophus cyrtocristatus (FMNH P27393), lateral view ................................... 802
Figure J.4. Character 290. Expansion of the distal end (rectangles) of the fibula. A,
Bactrosaurus johnsoni (AMNH 6553), lateral view. B, Tsintaosaurus spinorhinus (IVPP
V725), lateral view...................................................................................................... 803
Figure J.5. Character 291. Development of the medial platform (rectangles) of the astragalus.
Specimens displayed in cranial view. A, Iguanodon bernissartensis (IRSNB 1534). B,
Edmontosaurus annectens (CMN 8399). C, Brachylophosaurus canadensis (MOR 794).
D, Parasaurolophus cyrtocristatus (FMNH P27393). E, Lambeosaurus lambei (ROM 1218).
F, Tsintaosaurus spinorhinus (IVPP V725). Abbreviations: as = astragalus; cc = calcaneum;
fb = fubula; tb = tibia .................................................................................................. 804
Figure J.6. Character 294 Boxplot showing the distribution of the length/width ratio of
metatarsal III for a sample of iguanodontoidean ornithopods. The inserted smaller graph
data can be classified. The photograph shows how the two variables of this ratio were
measured, exemplified in a third metatarsal of Charonosaurus jiayinensis (CUST JV1236)
.................................................................................................................................... 807
Figure J.7. Character 295. Length/width proportions of pedal phalanx II2 (rectangles). A,
Iguanodon bernissartensis (IRSNB 1723), dorsal view. B, Corythosaurus intermedius (ROM
845) dorsal view. C, Brachylophosaurus canadensis (MOR 1071), dorsal view .......... 808
Figure J.8. Character 296. Length/width proportions of pedal phalanges III2 and III3.
Specimens displayed in dorsal view. A, Shantungosaurus giganteus (GMV 1780). B,
Edmontosaurus regalis (CMN 2289). C, Hypacrosaurus altispinus (CMN 8501). D,
Saurolophus angustirostris (ZPAL MgD-I 157, subadult specimen ............................. 809
Figure J.9. Character 297. Length/width proportions of pedal phalanges IV2-IV4. A,

Brachylophosaurus canadensis (MOR 1071), dorsal view. B, Edmontosaurus anenctens (SM
R3046), dorsal view. C, Gryposaurus notabilis (ROM 764), dorsal view. D,
Shantungosaurus giganteus (GMV 1780), dorsal view ............................................... 810
lxviii
Figure J.10. Character 298. Morphology of the pedal ungual phalanges. A, Gilmoreosaurus
mongoliensis (AMNH 6551), dorsal view. B, Brachylophosaurus canadensis (MOR 1071-7-
31-99-273), dorsal view .............................................................................................. 811
Figure J.11. Character 299. Absence and presence of plantar ridge (arrow) in pedal ungual
phalanges. A, Velafrons coahuilensis (SEPCCP 1-4061), cranial view. B,
Brachylophosaurus canadensis (MOR 1071-1071-7-31-99-273), cranial view ............ 812
lxix
LIST OF ABBREVIATIONS
Institutional Abbreviations
AEHM1 Amur Natural History Museum, Blagoveschensk, Russia

AMNH American Museum of Natural History, New York City, New York, USA
ANSP Academy of Natural Sciences of Philadelphia, Pennsylvania, USA
AUMP University of Alabama Museum of Paleontology, Auburn, Alabama, USA
BMNH The British Museum, London, UK
BYU Brigham Young University Museum, Provo, Utah, USA
CM Carnegie Museum of Natural History, Pittsburgh, Pennsylvania, USA
CMN Canadian Museum of Nature, Ottawa, Ontario, Canada
CUST Changchun University of Science and Technology, Changchun, Jilin, China
DMNH Denver Museum of Natural History, Denver, Colorado, USA
FGGUB Facultatea de Geologie si Geofizica, Universitatea Bucuresti, Bucharest, Romania
FMNH The Field Museum, Chicago, Illinois, USA
GDF Gadofaoua; specimen casts accessed at the MNHN, Paris, France
GMH1 Geological Museum of Heilongjang Province, Harbin, Heilongjang, China
GMV National Geological Museum of China, Beijing, China
IPS Museu de l’Institut Català de Paleontologia, Sabadell, Barcelona, Spain
IRSNB Institut Royal des Sciences Naturelles de Belgique, Brusels, Belgium
IVPP Institute of Vertebrate Paleontology and Paleoanthropology, Beijing, China
LACM Natural History Museum of Los Angeles County, Los Angeles, California, USA
MACN Museo Argentino de Ciencias Naturales Bernardino Rivadavia, Buenos Aires,
Argentina
MAFI1 Magyar Állami Foldtani Intézet, Budapest, Hungary
MAL Museum of Ancient Life, Lehi, Utah, USA
MCD Museu de la Conca Dellà, Isona, Lleida, Spain
MNHN Muséum National d’Histoire Naturelle, Paris, France
MOR Museum of the Rockies, Bozeman, Montana, USA
MPC Mongolian Paleontological Center, Academy of Sciences, Ulan Bataar, Mongolia
MPCA Museo Provincial Carlos Ameghino, Cipoletti, Argentina
MSNM Museo Civico di Storia Naturale di Milano, Milano, Italy
OTM Old Trail Museum, Chateau, Montana, USA
PIN1 Palaeontologiceski Institut, Akademii Nauk, Moscow, Russia.
PMU Museum of Evolution, Uppsala University, Uppsala, Sweden
NMMNH New Mexico Museum of Natural History and Science, Albuquerque, New
Mexico, USA
RAM Raymond M. Alf Museum, Claremont, California, USA
ROM Royal Ontario Museum, Toronto, Ontario, Canada
SBDE Sino-Belgian Dinosaur Expedition
SM Naturmuseum Senckenberg, Frankfurt, Germany
SMU Shuler Museum of Paleontology, Southern Methodist University, Dallas, Texas,
USA
lxx
TCMI The Children’s Museum of Indianapolis, Indiana, USA
TMM Texas Memorial Museum, Austin, Texas, USA
TMP Royal Tyrrell Museum of Paleontology, Drumheller, Alberta, Canada
UALVP University of Alberta Laboratory for Vertebrate Paleontology, Edmonton,
Alberta, Canada
UCM University of Colorado Museum, Boulder, Colorado, USA
UCMP University of California Museum of Paleontology, Berkeley, California, USA
USNM National Museum of Natural History, Washington D. C., USA
UTEP Centennial Museum at the University of Texas at El Paso, Texas, USA
YPM Yale Peabody Museum of Paleontology, New Haven, Connecticut, USA
ZPAL Instytut Paleobiologii, Polska Akademia Nauk, Warsaw, Poland
1
Specimens from these institutions were not examined by the author and had to be scored from
the literature or from digital images.
Taxonomic Abbreviations
Am Amurosaurus riabinini
Ar Aralosaurus tuberiferus
Ba Bactrosaurus johnsoni
Bb Barsboldia sicinskii
Br Brachylophosaurus canadensis
Ch Charonosaurus jiayinensis
Cl Claosaurus agilis
Co Corythosaurus sp.
Coi Corythosaurus intermedius
Cos Corythosaurus casuarius
Ed Edmontosaurus sp.
Eda Edmontosaurus annectens
Edr Edmontosaurus regalis
Eo Eolambia caroljonesa
Eq Equijubus normani
Gi Gilmoreosaurus mongoliensis
Gr Gryposaurus sp.
Grl Gryposaurus latidens
Grm Gryposaurus monumentensis
Grn Gryposaurus notabilis
Hd Hadrosaurus foulkii
Hy Hypacrosaurus altispinus
Hys Hypacrosaurus stebingeri
Ig Iguanodon sp.
Iga Iguanodon atherfieldensis
lxxi
Igb Iguanodon bernissartensis
Jx Jaxartosaurus aralensis
Kb Kerberosaurus manakini
Kr Kritosaurus navajovius
lc Lambeosaurus laticaudus
Lh Lophorhothon atopus
Lm Lambeosaurus lambei
Lmm Lambeosaurus magnicristatus
Ma Maiasaura peeblesorum
Ol Olorotitan ararhensis
Ou Ouranosaurus nigeriensis
Pa Parasaurolophus sp.
Pac Parasaurolophus walkeri
Pat Parasaurolophus tubicen
Paw Parasaurolophus walkeri
Pb Probactrosaurus spp.
Ph Pararhabdodon isonensis
Pr Prosaurolophus maximus
Pt Protohadros byrdi
Saa Saurolophus angustirostris
sal Salitral Moreno OTU
Sao Saurolophus osborni
Sh Shantungosaurus giganteus
Sy Sahaliyania elunchunorum
Se Secernosaurus koerneri
Ta Tanius sinensis
Te Telmatosaurus transsylvanicus
Ts Tsintaosaurus spinorhinus
Ve Velafrons coahuilensis
Wu Wulagasaurus dongi
x Hadrosauridae indeterminate
lxxii
ABSTRACT
Hadrosaurids were the most diverse and abundant dinosaurs at the end of the Cretaceous.
However, their phylogeny is incompletely known and the relationships of many taxa, particularly
European and South American, remains unresolved. Questions remain regarding the timing of
their origin and which attributes might have allowed these animals to diversify and colonize
nearly all continents by the late Campanian. Likewise, the center of origin and subsequent
biogeographical history of hadrosaurids has remained contentious.
With the goal of elucidating the evolutionary history of hadrosaurids, I present the most
comprehensive and resolved phylogeny of these animals ever estimated, using a complete
taxonomic sampling at specific level. I also report the results of a biogeographic analysis seeking
to establish the ancestral area of Hadrosauridae, testing whether they originated in Asia, North or
South America. In addition I used the same method to track their diversification in relation to
intercontinental connections throughout their tenure.
Parsimony and Bayesian methods were implemented to elucidate the phylogenetic
relationships of all hadrosaurid species. These included specimens of 41 hadrosaurids from
Europe, Asia and North and South America. Outgroup taxa included tweleve iguanodontoidean
species from Europe, North America and Asia. New characters were defined and old ones revised
on the basis of new data collected from first hand examination of specimens. Traditional and
geometric morphometrics were applied to discover patterns of variation containing phylogenetic
information. Continuous geometries were studied using the Analysis of Planar Shapes Using
Geodesic Paths, a novel landmark-free method that considers the continuous non-linear geometry
of the bones. In this way, a total of 299 phylogenetically informative characters (205 cranial and
94 postcranial) were defined and documented, the most extensive character data set ever
constructed for hadrosaurid dinosaurs.
In general, parsimony and Bayesian analyses confirmed the dichotomic evolution of
hadrosaurids into Saurolophinae and the hollow-crested Lambeosaurinae. Saurolophines consisted
of “saurolophs” and “kritosaurs”. Lambeosaurines consisted of a succession of Eurasian
outgroups to two major clades, “parasaurolophs” and “amurosaurs” (“hypacrosaurs” and
“corythosaurs”). Hadrosauridae was redefined as the clade stemming from the most recent
common ancestor of Hadrosaurus foulkii and Parasaurolophus walkeri. Its monophyly was
lxxiii
unambiguously supported by an iliac supraacetabular process that projects lateroventrally
between half and three quarters of the dorsoventral depth of the central plate of the ilium and a
craniocaudally short supraacetabular process that is less than 55% the length of the central plate
of the ilium. The closest outgroup taxa to Hadrosauridae lived in eastern North America. The
hadrosaurid radiation and the divergence of saurolophines from lambeosaurines occurred no later
than the Santonian and was coincident with the evolution of a suite of mandibular characters (i.e.,
increased number of tooth families, presence of three teeth forming the dentary occlusal plane,
ventral offset of the oral predentary-premaxilla contact). These characters may have been key
innovations. These results suggest that feeding adaptations might have played a central role in the
diversification of hadrosaurids. Circumnarial fossae and cranial crests evolved prior to the
hadrosaurid radiation. Both of these characters were reconstructed to have evolved at the same
time.
Ancestral areas were reconstructed on the phylogeny derived from the weighted
parsimony analysis of Hadrosauridae. Fitch parsimony and the Dispersal-Vicariance (DIVA)
method were implemented to reconstruct ancestral areas for all clades of Hadrosauria. The
results show that the genesis of Hadrosauridae occurred in eastern North America during the late
Santonian. Soon after their origin, hadrosaurids dispersed to Asia, the ancestral area for the
major Saurolophinae-Lambeosaurinae divergence. Lambeosaurines and saurolophines dispersed
to North America and underwent major cladogenesis during the late Campanian. Within
Saurolophinae, “maiasaurs” and “saurolophs” returned to Asia, while “kritosaurs” colonized
South America by the late Campanian. Within Lambeosaurinae, “tsintaosaurs” dispersed to
Europe no later than the late Campanian. The North American radiations of “hypacrosaurs”,
“parasaurolophs” and “corythosaurs” may have represented independent dispersal events from
the Asian continent or, alternatively, were part of a single dispersal in the late Campanian with
posterior occupations of Asia by a few species from those clades. Vicariant events may have
occurred following several of the inferred hadrosaurid dispersals. Thus, both vicariance and
dispersal may have been instrumental in shaping the recorded distribution of hadrosaurids.
Two additional studies were conducted on European and South American hadrosaurids.
Although well represented in Asia and North America, the presence of this animals in Europe
and South America is known only from rare and fragmentary remains that are poorly
documented and mostly unstudied. As a result, the impact of these animals on the phylogenetics
lxxiv
and biogeography of hadrosaurids as a whole are not known. Here, I provide a revised and
complete osteology of the type specimens and hypodigms for the only two taxa known from
South America, Secernosaurus koerneri and Kritosaurus australis. K. australis is regarded as a
junior synonym of S. koerneri, based on a combination of iliac and pubic characters unique to
these two taxa. Inclusion of S. koerneri within the genus Kritosaurus is not supported by the
phylogenetic analysis. S. koerneri is inferred to be a member of the “kritosaur” clade within
Saurolophinae, as the sister taxon to the Argentinean OTU from Salitral Moreno. Another
unnamed hadrosaurid, an OTU from Big Bend National Park, Texas, is positioned as the closest
outgroup to the South American clade. The results of this biogeographical analysis indicates that
the Secernosaurus clade originated in South America during the late Campanian after a dispersal
event (probably followed by vicariance) from southern North America before the end of that
geologic stage.
Finally, an regarding European hadrosaurids, the observation of previously unrecognized
characters in the maxilla and dentary of Tsintaosaurus spinorhinus (Campanian of China) and
Pararhabdodon isonensis (Maastrichtian of Spain) led to a revision of the taxonomy and
phylogenetic relationships of the latter. In particular, the extreme elongation of the symphyseal
region of the type and only material of Koutalisaurus kohlerorum (a dentary from the
Maastrichtian of Spain) was also observed in T. spinorhinus. This implied that K. kohlerorum is
indistinct from T. spinorhinus. This, in combination with the fact that P. isonensis and T.
spinorhinus share a maxilla with elevated jugal joint continuous with the ectopterygoid ridge (a
character not seen among Iguanodontoidea), led me to the conclusion that K. kohlerorum as a
junior synonym of P. isonensis. The incorporation of those new characters in Bayesian and
parsimony phylogenetic analyses of Hadrosauridae resulted in the inference that Pararhabdodon
isonensis and Tsintaosaurus spinorhinus form a clade of basal lambeosaurines—the
“tsintaosaurs”. Fitch parsimony and Dispersal-Vicariance reconstruction of ancestral areas on the
resulting phylogeny indicated that “tsintaosaurs” originated in eastern Asia during the middle or
late Campanian and that P. isonensis represents a dispersal event (followed by vicariance) to
Europe that occurred before the end of that geologic stage.
lxxv
INTRODUCTION
The Hadrosauridae is the most derived clade of ornithischian ornithopods, representing

one of the most successful dinosaurian radiations (Horner et al., 2004). Hadrosaurid remains
have been found in Upper Cretaceous strata of Europe, Asia, the Americas, and Antarctica
(Buffetaut and Tong-Buffetaut, 1993; Forster, 1997b; Case et al., 2000; Lund and Gates, 2006).
Their fossil record is very rich, probably the most complete among dinosaurs. It includes fully
articulated skeletons (Lull and Wright, 1942), thousands of disarticulated bones from
multindividual assemblages (Dodson, 1971; Varricchio and Horner, 1993), growth series
(Horner 2000), soft tissue impressions (Osborn, 1912; Murphy et al., 2007), eggs, nests and
neonates (Horner and Makela, 1979; Horner, 1982; Horner and Currie, 1994) and trackways
(Currie et al., 1991).
Hadrosaurids have received considerable attention due to their supracranial adornments.
In many species, the nasal or the premaxilla and nasal bones extended caudally and/or dorsally to
form either solid (Prieto-Marquez, 2005) or hollow supraorbital crests (Dodson, 1975). The
function of these crests was probably related to species recognition and other display behaviors
(Hopson, 1975; Evans, 2006). Hadrosaurids are also notable for having hundreds of closely
packed teeth (dental batteries) with masticatory capabilities (Weishampel, 1984; Norman and
Weishampel, 1985), mediolaterally expanded rostra (Morris, 1970) and hypertrophied nasal
passages (Ostrom, 1961; Weishampel, 1981). The earliest record of the Hadrosauridae sensu
stricto is from the Lower Campanian (Horner, 1992) and the largest taxonomic diversity of the
clade is recorded from the late Campanian to the early Maastrichtian (Gates, 2007).
Because of their abundance, widespread distribution, and excellent preservation,
hadrosaurids provide an excellent model for studying dinosaurian evolution and paleobiology.
For example, they can be used to elucidate how geological events affected the evolution and
distribution of terrestrial faunas, how their anatomy relates to biomechanical form, function and
ecology, and, owing to their large sample sizes, how anatomical variation relates to species
recognition, reproductive behavior, and population biology. These and many other questions can
be best approached in a phylogenetic context to provide an all-encompassing evolutionary
understanding for the group. However, a comprehensive, well-resolved, accurate and robust
1
phylogeny of these animals does not exist. Likewise, the origin and biogeographic history of
hadrosaurids remains uncertain.
Resolving hadrosaurid relationships at lower taxonomic levels is crucial if one is to
understand their origin, how they became so widespread, and they acquired their unique suite of
derived skeletal attributes. Previous phylogenetic studies have only included a limited number of
taxa and/or a limited number of characters. A cladistic analysis integrating all taxa from all
continents using modern methods of phylogenetic inference has yet to be conducted. This is
mainly because most of the material from Europe, Asia and South America remains poorly
studied and has yet to be compared with the much better known North American taxa. The vast
majority of studies have mostly incorporated hadrosaurids from North America and only very
distantly related outgroup taxa (Weishampel et al., 1993; Horner et al., 2004). Certainly, the
inclusion of hadrosaurid and closely related outgroup taxa from Asia (Godefroit et al., 1998;
Norman, 2002; Hai-Lu You et al., 2003), South America (Brett-Surman, 1979; Bonaparte et al.,
1984; Powell, 1987) and Europe (Casanovas et al., 1999; Lopez-Martinez et al., 2001; Prieto-
Márquez et al., 2006b) stands to augment our current view of the evolutionary history of these
animals. Paleontologists have been also excluded phylogenetically informative characters in their
phylogenies to minimize the amount of missing data when skeletal parts were missing in their
taxa of interest. Thus, it is common to find phylogenies based on only a few dozens (and often
even less) characters (e.g., Head, 1998; Casanovas et al., 1999; Godefroit et al., 2000; Hai-lu
You et al., 2003). Such practices can reduce phylogenetic accuracy (Wiens, 1998; 2003).
Additionally, characters used in phylogenetic studies are rarely informative enough to
resolve lower level relationships within the Hadrosauridae, especially those from the postcrania.
This is a consequence of our relatively superficial knowledge about the morphological variation
in the hadrosaurid skeleton. As a consequence, intraspecific individual and ontogenetic patterns
of variation have clouded phylogenetic signals. This problem impacts also the number and
diagnoses of recognized species. Only a few studies have been conducted that shed light into this
problem (Dodson, 1975; Chapman and Brett-Surman, 1990; Egi and Weishampel, 2002; Brett-
Surman and Wagner, 2007). However, none of them has been comprehensive enough to
encompass all skeletal areas and elements while at the same time being detailed enough to
include all possible sources of variation.
2
Finally, no quantitative analysis integrating phylogenetic and geographical data has been
undertaken to reveal the birthplace of hadrosaurids or explain how they achieved their known
distribution. Instead, previous biogeographical studies consisted of literal readings of recorded
distributions in the fossil record, typically leading to narratives of dispersal or vicariant scenarios
with centers of origin for the major clades (Brett-Surman, 1979; Weishampel and Weishampel,
1983; Head, 1998; 2001; Casanovas et al., 1999; Wagner, 2001; Godefroit et al., 1998; 2003;
2004b). Such an approach, while useful for building working hypotheses, is prone to error
because does not account for extinction or lack of record as explanations for the absence of taxa
in some areas. This critique can be extended to most studies of dinosaur biogeography (LeLoeuff
et al., 1992; Russell, 1993; Prieto-Marquez et al., 2000; Sereno, 1997; 1999a, b; Sereno et al.,
2004). Approaching biogeographical research with current methods of cladistic biogeography
(Morrone and Crisci, 1995; Lieberman, 2000; 2003) would provide testability and rigor to the
field. A few authors have recently implemented analytical methods for inferring biogeographical
histories for dinosaurs in a broad scale (Chapman, 1997; Upchurch et al., 2002; Holtz et al.,
2004) and crocodyliforms (Turner, 2004), but not yet specifically for hadrosaurids.
The cardinal goal in the present research is to resolve the phylogenetic relationships of
the Hadrosauridae to the species level. One of my novel contributions will be the inclusion of
newly found and poorly known forms from Europe and South America. The phylogeny will
serve as the underpinning for addressing a number of questions. When did the hadrosaurid
radiation take place? Which apomorphies characterize the hadrosaurid radiation and what is the
order of acquisition of these apomorphies along the iguanodontoidean lineage? Where is the
location of the ancestral area of the Hadrosauridae? What was the relative influence of
vicariance, dispersal and regional extinction in the geographical distribution of hadrosaurids?
The results of my cladistic analyses depend on morphological characters reflecting actual
phylogenetic signals and encompassing all the information that the available skeletal material
can provide. Thus, my first concern will be the analysis of patterns of variation in the
morphology of hadrosaurid skeletal elements to reassess, discover and define autapomorphic and
phylogenetically informative characters. These characters will form the basis for the subsequent
analyses of phylogenetic inference and reconstruction of historical biogeography.
3
The first chapter of this study provides an all encompassing review of previous research
on the systematics of hadrosaurid dinosaurs, spanning from the first scientific recognition of
these animals in the middle of the nineteenth century to the present day.
The second includes a description of the methodology that I implemented to define
phylogenetically informative characters and delimit character states, as well as that used for
phylogenetic inference. This chapter then provides phylogenetic hypotheses of the evolutionary
relationships of hadrosaurid dinosaurs and revises the taxonomy of the clade.
The third chapter is devoted to the reconstruction of ancestral areas for the clades inferred
in the phylogenies presented in the previous chapter. It aims to provide a hypothesis on the origin
and historical biogeography of hadrosaurid dinosaurs.
The final two chapters provide further detail on two hadrosaurid subclades inferred in the
phylogenetic reconstructions of chapter two. These subclades are of particular interest because
contain poorly studied taxa from continental areas where hadrosaurids have been relatively
unstudied and whose remains are scarce. Specifically, chapter three focuses on the systematics of
Secernosaurus koerneri, from the Campanian of Argentina. A complete description of the
available material is provided, along with an emended systematic paleontology that clarifies the
taxonomy and relationships of this hitherto enigmatic hadrosaurid. The final chapter provides
clarification of important diagnostic characters of a hitherto unrecognized clade of Eurasian
lambeosaurines composed of Tsintaosaurus spinorhinus (Campanian of China) and
Pararhabdodon isonensis (Maastrichtian of Spain).
The appendices that follow illustrate and document nearly all the 299 characters used for
the phylogenetic inference of hadrosaurid interrelationships. The numerical data for quantitative
characters are also provided. As in many other phylogenetic studies, the matrix of characters is
presented as an appendix.
I should point out that this study is not intended as a general monographic review of
hadrosaurid dinosaurs. Thus, only data and descriptive information relevant to apomorphies, with
a role in inferring phylogenetic relationships, will be specified.
I hope that this research will provide the needed backbone for future studies seeking a
more complete understanding of dinosaur paleobiology and evolution. Furthermore, and in a
broader scale, the fossil record of hadrosaurids offers an excellent model system for future
research about how evolution is influenced by long-term geological changes.
4
CHAPTER 1
HISTORICAL REVIEW OF STUDIES ON HADROSAURID SYSTEMATICS
First Discoveries
The first description of fossil remains referable to a hadrosaurid dinosaur was published
by Joseph Leidy in 1856. These remains consisted of fragments of several tooth crowns found by
Ferdinand Vandiveer Hayden in the badlands of the Judith River Formation of Montana
(Fig.1.1). Leidy (1856a) erected the new genus and species Trachodon mirabilis from some of
these dental remains (ANSP 9260; Fig. 1.1A) and considered his new taxon as closely related to
Iguanodon from the lower Cretaceous of England. Later in the same year, Leidy published on the
discovery of two caudal vertebrae (USNM 219 and 221) and a proximal pedal phalanx (USNM
220) from the Lance Formation of Nebraska, for which he erected the name Thespesius
occidentalis (Leidy, 1856b).
Two years after the description of these fragmentary remains, a naturalist from
Philadelphia, William Parker Foulke, learned of the existence of fossil bones in the Cretaceous
marls of the Woodbury Formation, in Haddonfield, New Jersey (Foulke, 1859; Weishampel and
White, 2003). Foulke contacted Joseph Leidy at the Academy of Natural Sciences in
Philadelphia and excavations took place, unveiling large appendicular and vertebral bones
belonging to a hadrosaur (Weishampel and Young, 1996; Warren, 1998). Leidy (1858, reprinted
in 1859) described the teeth, fragmentary maxilla, partial vertebrae, forelimb and hindlimb of
this hadrosaur, that he named Hadrosaurus foulkii (Fig. 1.2). This specimen was the first
dinosaur known from partially complete skeletal remains discovered in the Americas and the
first dinosaur skeleton to be mounted for exhibition (Ruschenberger and Tryon, 1879;
Anonymous, 1922). As in the case of Trachodon, Leidy (1858) noticed the resemblance of the
teeth and vertebrae with those of Iguanodon.
However, scientists would not gain a complete knowledge of the skeletal anatomy of
hadrosaurids until the end of the nineteenth century. The first nearly complete skeleton of a
hadrosaurid was found in 1882 by two fossil collectors of Cope, J. L. Worthman and R. S. Hill,
in strata from the Lance Formation of South Dakota, near Black Hills (Lund and Gates, 2006).
The specimen, AMNH 5730, was named Diclonius mirabilis by Cope (1883). This finding was
5
followed in 1891 by the collection of two skeletons (USNM 2414 and YPM 2182) from the
Lance Formation of Wyoming (Creisler, 2007). Marsh (1892) referred these skeletons to the
genus Claosaurus and erected the species C. annectens. In 1904, Barnum Brown collected
another complete skeleton from the same species (AMNH 5886), from an outcrop of the Lance
Formation in a region known as Crooked Creek, in Montana. These four specimens are referable
to Edmontosaurus annectens. The two skeletons AMNH 5730 and 5886 were mounted at the
American Museum of Natural History, in a quadropedal and bipedal posture, respectively. These
two specimens can still be seen in the Hall of Ornithischian Dinosaurs.
The Pre-Cladistic Period
The name Hadrosauridae was erected by Cope in 1869 in his extensive synopsis of fossil
amphibians and reptiles of North America. Cope (1869) diagnosed the Hadrosauridae on the
basis of “teeth in several rows forming a vertical pavement; metatarsals ?three” (p. 91). Thus,
Cope realized that these animals possessed a well-developed dental battery and that this trait
probably characterized the group. The family Hadrosauridae was composed of Hadrosaurus, the
only hadrosaurid genus recognized by Cope at the time. Within Hadrosaurus, he distinguished
two species: H. mirabilis (= Trachodon mirabilis Leidy, 1856) and H. foulkii (Leidy, 1858).
With the erection of the Hadrosauridae, Cope realized that these animals were anatomically
distinct from other members of the Ornithopoda, at the time thought to include Iguanodon and
the thyreophorans Secelidosaurus and Hylaeosaurus.
However, in1888 Lydekker ignored Cope’s Hadrosauridae and erected the
Trachodontidae, which included the genera Hadrosaurus, Orthomerus and Diclonius (=
Edmontosaurus). Furthermore, he regarded Hadrosaurus as a junior synonym of Trachodon.
Lydekker characterized the Trachodontidae as having “teeth arranged in vertical rows, and
articulating with one another so as to form a more or less complete pavement. In the one genus
(Diclonius) in which it is known the skull is elongated and much depressed, and the premaxillae
are extended a considerable distance of the large nares, thus producing a very long edentulous
portion” (Lydekker, 1888: p. 241). From this diagnosis it is clear that Lydekker recognized the
derived anatomy of the expanded oral region of the hadrosaurid skull. Otherwise, his diagnosis
of the Trachodontidae was very similar to that of Cope’s for the Hadrosauridae.
6
Similarly, Marsh (1890) used the name Trachodontidae for one of the two families that
he classified in the Ornithopoda. He diagnosed the Trachodontidae on the basis of “large size,
with teeth of the Hadrosaurus type, in many rows. Cervical vertebrae opithocoelian. Limb bones
hollow. Fore limbs small. Feet digitigrade.” (Marsh, 1890: p. 424).
The name Trachodontidae prevailed until the beginning of the twentieth century. Brown
(1914) presented an early classification of the seven genera of Trachodontidae known at that
time. He considered two subfamilies, Trachodontinae and Saurolophinae. Trachodontines were
diagnosed by the possession of an ischium with unexpanded distal end and crestless skulls.
Brown included Trachodon, Claosaurus, Kritosaurus and Hadrosaurus in this subfamily.
Conversely, Brown diagnosed saurolophines as having cranial crests and an expanded distal end
of the ischium (also informally known as ischial “boot” or “foot”). In this subfamily he included
Saurolophus, Hypacrosaurus and the then newly discovered Corythosaurus. Brown also realized
that the skull and, to a lesser degree, the pelvic girdle, contained the majority of the diagnostic
characters of these animals. Likewise, he became aware of the potential risk of confounding
individual and ontogenetic characters with those useful in delimiting taxa and inferring
evolutionary relationships.
In 1918, Lambe invalidated the name Trachodontidae and resurrected Hadrosauridae. He
preferred the latter name because it had precedence over Trachodontidae. Likewise, Lambe also
preferred to use Hadrosaurus as the type genus for the family instead of Trachodon because the
former was founded upon more abundant and better-preserved material. Furthermore, he
regarded the genus Trachodon as being of dubious validity because the type material lacked
sufficient diagnostic characters. Although Lambe agreed with Brown (1914) in grouping
hadrosaurids into two subfamilies, he also replaced the name Trachodontinae with
Hadrosaurinae. Whithin the latter subfamily, Lambe included Gryposaurus, Edmontosaurus,
Claosaurus and Diclonius (= Edmontosaurus annectens). The Saurolophinae was composed of
Stephanosaurus (= Lambeosaurus), Prosaurolophus, Corythosaurus, Saurolophus and
Cheneosaurus (= juvenile Hypacrosaurus altispinus).
Later, Lambe (1920) divided the Hadrosauridae into three subfamilies: Hadrosaurinae,
Saurolophinae and Stephanosaurinae. Hadrosaurines were characterized by having a flat
supraorbital region, unexpanded distal end of the ischium and lacrimal of moderate size, among
other characters. In this subfamily he included Gryposaurus, Kritosaurus, Edmontosaurus.
7
Diclonius (= E. annectens) and Claosaurus. Prosaurolopus and Saurolophus were grouped under
their own subfamily, the Saurolophinae. Finally, the Stephanosaurinae included the hollow-
crested forms Hypacrosaurus, Cheneosaurus, Stephanosaurus and Corythosaurus. Lambe
realized the large suite of characters shared between hadrosaurines and saurolophines. However,
he maintained the distinction of the latter subfamily because he thought that saurolophines had a
“footed” ischium, as did lambeosaurines, but not hadrosaurines.
In the same publication, Lambe noticed several derived characters that would later be
used to diagnose lambeosaurine hadrosaurids: a reduced lacrimal, relatively short skulls, external
nares completely formed by premaxillae, caudodorsal retraction and hypertrophy of the nasal
passages, the reduction of the frontoparietal region, and, of course, the hollow supracranial
crests. Furthermore, Lambe also noticed derived characters common to hadrosaurines and
saurolophines, such as the caudally extended circumnarial fossa (“elongated narial vacuity,
opening through the skull, enclosed above by the nasals and premaxillae”; Lambe, 1920: p. 69).
In 1923, Parks invalidated the genus Stephanosaurus Lambe, 1914. This had important
implications for the subfamiliar name of hollow crested hadrosaurids. Lambe (1902) had erected
the species Trachodon marginatus for a dentary, maxilla and several appendicular elements of a
hadrosaurid collected in the Campanian Belly River beds (Judith River Group) in the Red Deer
River region of Alberta, Canada. Later, he established the genus Stephanosaurus for that species
as S. marginatus and referred to this taxon two skulls (one of them being CMN 351) that were
found at a later time from the same formation and region (Lambe, 1914b). Parks (1923), in
agreement with Brown (1914), saw no prove of the referral of the two skulls to S. marginatus
and regarded them as belonging to a new genus and species of hadrosaurid, that he named
Lambeosaurus lambei. Because Parks considered Stephanosaurus has having been founded upon
undiagnostic material, he invalidated the name Stephanosaurinae and erected the new subfamilial
name Lambeosaurinae. This name has been maintained up to the present day to designate those
hadrosaurids with caudodorsally retracted nasal passages and hollow supracranial crests.
During the first decades of the twentieth century scientists were still more concerned with
the osteology and taxonomy of the increasing number of hadrosaurid specimens being collected
than in the origin and evolutionary history of the group. Occasionally, a few speculations were
offered on the evolutionary history of these animals based on personal judgement of osteological
similarities. For example, Gilmore (1924), in describing a complete skull (CMN 8501) of
8
Hypacrosaurus altispinus, commented on the possible covergent evolution of the receding lateral
profile of the rostrum in this species and Cheneosaurus tolmanensis (a juvenile H. altispinus,
CMN 2246). Parks (1922) regarded Parasaurolophus as a close relative of Saurolophus on the
basis of the shape of its crest, even though he also recognized that there were other attributes
shared between the former genus and lambeosaurines. The fact that Parks gave more importance
to the morphology of the cranial crests exemplifies how scientists were subjectively biased in
their criteria for establishing taxonomic units and hypotheses of evolutionary relationships. This
approach would be common until the application of cladistic methods in the middle eighties
(however, with a few exceptions; see below).
In Europe, Nopcsa (1928) maintained the name Trachodontidae and divided the family
into four subfamilies: Protrachodontinae, Trachodontinae, Saurolophinae and Lambeosaurinae.
The Protrachodontinae, that Nopcsa characterized as having a short premaxilla and short teeth,
included nomina dubia such as Orthomerus and Syngonosaurus. The Trachodontinae was
characterized as having “numerous teeth” and lack of cranial ornamentation; it included
Claosaurus, Edmontosaurus, Hadrosaurus, Kritosaurus, Thespesius (= Edmontosaurus
annectens) and Trachodon. The Saurolophinae was diagnosed as having “very numerous teeth”
and “males with median horn-like protuberance on skull” (Nopcsa, 1928: p. 185); it included
Parasaurolophus, Prosaurolophus and Saurolophus. Finally, the Lambeosaurinae was diagnosed
as possessing “very numerous teeth, males with helmet-like crest on skull” (p. 185); it included
Cheneosaurus, Corythosaurus, Lambeosaurus and Hypacrosaurus.
By the 1940’s, hadrosaurids were already known from abundant, complete and well-
preserved skeletal material collected from Campanian-lower Maastrichtian strata in North
America, most remarkably from the northern states of the United States and adjacent regions in
southern Canada, such as the Dinosaur Provincial Park. Classic genera such as Corythosaurus,
Lambeosaurus, Hypacrosaurus, Parasaurolophus, Gryposaurus, Prosaurolophus, Saurolophus
and Edmontosaurus became known from several exquisitely preserved complete skulls and, in
many cases, articulated postcranial skeletons. These materials allowed scientists to gain a more
accurate knowledge of the great taxonomic diversity and wide variety of forms of hadrosaurid
dinosaurs. Outside North America, new species (most of them apparently primitive) had also
been erected indicating the nearly cosmopolitan distribution of these animals during the late
Cretaceous. In Europe, Nopcsa (1900) had described and named Telmatosaurus transsylvanicus,
9
from the Maastrichtian of Rumania. In Asia, a few taxa had also been erected by that time. These
included: Mandschurosaurus amurensis (Riabinin, 1930; regarded as nomen dubium by Brett-
Surman [1979]) from near the banks of the Amur River, along the Russia-China border in
Heilongjang Province; Bactrosaurus johnsoni and Gilmoreosurus mongoliensis (Gilmore, 1933)
form the late Maastrichtian-early Campanian Iren Dabasu Formation of Inner Mongolia, China;
and Nipponosaurus sachalinensis from the Santonian-early Campanian Upper Yezo Group of the
Russian island of Sakhalin (Nagao, 1936).
In their classic monograph on the morphology and systematics of North American
hadrosaurids, Richard S. Lull and Nelda E. Wright (1942) presented a phylogenetic hypothesis of
the evolutionary history of these animals (Fig. 1.3). These authors supported the main split of
hadrosaurids into two major groups: the “flat-headed” or Hadrosaurinae and the “crested
hadrosaurs” or Lambeosaurinae. Hadrosaurines were divided into three independent lineages of
ancestor-descendant relationships. These authors did not specify any ancestors at the base of
each of the lineages. In the first lineage Gryposaurus evolved into Kritosaurus. In the second
lineage Edmontosaurus regalis gave rise to Anatosaurus edmontoni (= Edmontosaurus
annectens), that then evolved into a dichotomy integrated by A. annectens and A. copei (=
Edmontosaurus annectens). A third lineage showed Prosaurolophus as the ancestor of
Saurolophus. Within lambeosaurines, they considered four lineages. The first one consisted of
the relatively small and low-crested “cheneosaurs” (juveniles of other lambeosaurine genera such
as Hypacrosaurus [Dodson, 1975]). In the second lineage Corythosaurus evolved into H.
altispinus. A third lineage consisted solely of Lambeosaurus. Finally, the fourth lineage featured
Parasaurolophus walkeri as the ancestor of P. tubicen. All these lineages were based on the
cranial anatomy of the more complete hadrosaurid specimens known at the time. Furthermore,
Lull and Wright (1942) poined out that “kritosaurs” (Gryposaurus and Kritosaurus) and
“saurolophines” (Saurolophus and Prosaurolophus) were morphologically intermediate forms
between non-crested and the lambeosaurine crested forms. “Kritosaurs” were regarded as
intermediate for having a simple solid crest, yet lacking the “footed” ischium and narial structure
typical of lambeosaurines. Prosaurolophus and Saurolophus were regarded as intermediate for
having cranial crests that were “utterly different in shape and appearance from those of the
Lambeosaurinae” (Lull and Wright, 1942: p. 49).
10
Sternberg (1954) proposed the abandonment of the subfamily Saurolophinae and the
inclusion of Prosaurolophus and Saurolophus within the Hadrosaurinae. Until that time
Saurolophus had been classified outside hadrosaurines because it was thought to have a “footed”
ischium. Although the type skeleton of Saurolophus (AMNH 5220) did not preserve the distal
end of the ischium, the discovery of a disarticulated “footed” ischium (AMNH 5225) from the
same region and formation led Brown (1914) to assume that it belonged to the former genus.
Brown then regarded AMNH 5225 as the plesiotype of Saurolophus. Prosaurolophus, the type
of which lacked the postcranial skeleton, had been allied to Saurolophus because its skull
morphology was thought to be ancestral to the latter. However, Parks (1924) described a nearly
complete skeleton of Prosaurolophus (ROM 787) that clearly showed an “unfooted” ischium. As
Sternberg (1954) pointed out, it was unproven that AMNH 5225 could be referred to
Saurolophus (actually, this ischium is most likely referable to Hypacrosaurus altispinus; Prieto-
Marquez, unpublished data). Rather, Sternberg considered that the ischium of Saurolophus
would probably not be footed as in Prosaurolophus, provided that both genera were regarded as
closely related. For this reason, he abandoned the name Saurolophinae.
Furthermore, Sternberg (1954) found no substantial differences among lambeosaurine
taxa to separate them into subfamilies, as opposed to Lull and Wright (1942), who had erected
the Cheneosaurinae for the small forms Cheneosaurus (juvenile Hypacrosaurus) and
Procheneosaurus (= Tetragonosaurus, that consisted of juvenile Corythosaurus and
Lambeosaurus; Evans et al. [2005]; Evans [2007]). In Sterneberg’s view, the most fundamental
characters in lambeosaurines were the modifications of their premaxillae and nasals in relation to
the hypertrophied nasal passages. And since these modifications were also present in
“cheneosaurs”, there was no justification to make such distinction. He also believed that there
were more differences between Parasaurolophus and Corythosaurus than between the latter and
Cheneosaurus (= Hypacrosaurus).
Thus, Sternberg (1954) favored the dichotomic classification of hadrosaurids into
Hadrosaurinae and Lambeosaurinae. He recognized that postcranial characters such as the
relative shortening of the humerus relative to the length of the forearm, the number of co-ossified
sacral vertebrae and the length of dorsal neural spines could be variable within subfamilies.
Consequently, he gave more importance to the premaxillonasal complex as a potential source of
diagnostic and phylogenetically informative characters.
11
Young (1958) supported Sternberg’s invalidation of the subfamily Cheneosaurinae but
regarded the latter as primitive within the Lambeosaurinae because of the less derived
morphology of the premaxillonasal region. However, the study of Tsintaosaurus spinorhinus,
from the Campanian of the Wangshi Group the Chinese Province of Shandong, with its
rostrodorsally directed tubular nasal, led Young to reconsider the validity of the Saurolophinae.
Young did not think that forms with solid nasal crests such as Saurolophus and Tsintaosaurus
should be classified with “flat-headed” hadrosaurids under a single subfamily. Thus, within the
Saurolophinae he included those hadrosaurids similar to the Hadrosaurinae in the morphology of
the premaxilla but with “nasals specialized in various ways, thickened or forming boss or as a
tube-like apparatus directed either backwards or nearly vertical” (Young, 1958: p. 106). In this
subfamily Young included Tsintaosaurus, Saurolophus, Prosaurolophus and Kritosaurus (=
Gryposaurus, excluding K. navajovius). The other crested hadrosaurids where the premaxilla and
nasal extend caudal to the braincase were included in the Lambeosaurinae. In the Hadrosaurinae,
Young included only those forms with the nasal in “normal position and never specialized”
(Young, 1958: 106): Thespesius (= Edmontosaurus), Hadrosaurus, Anatosaurus (=
Edmontosaurus) and Mandchurosaurus (= Gilmoreosaurus, in part) from China. Finally, a
fourth family was considered, the Prohadrosaurinae (= Protrachodontinae of Nopcsa, 1928), to
accommodate those more primitive forms with relatively lesser tooth count and short premaxilla.
Around the same time, Friedrich von Huene (1956) erected the name Hadrosauria as a
superfamily that included five families: the Prohadrosauridae, Hadrosauridae, Saurolophidae,
Cheneosauridae and Lambeosauridae. In the Prohadrosauridae, von Huene included relatively
basal forms such as Telmatosaurus and Tanius. Within the Hadrosauridae, he included the
nomen dubium Ornithotarsus, along with “flat-headed” or “low-crested” taxa such as Thespesius
(= Edmontosaurus annectens), Anatosaurus (E. annectens), Edmontosaurus, Mandschurosaurus
(= Gilmoreosaurus, in part), Kritosaurus (= K. navajovius and Gryposaurus spp.) and
Brachylophosaurus. The Saurolophidae was composed of Saurolophus, Prosaurolophus and
Bactrosaurus. As in Lull and Wright (1942), the Cheneosauridae included the small (juvenile
lambeosaurine) forms Cheneosaurus and Procheneosaurus. Finally, von Huene’s
Lambeosaurinae was equivalent to the Lambeosaurinae of other authors in that it included those
hadrosaurs with large hollow crests, such as Lambeosaurus, Corythosaurus, Hypacrosaurus and
12
Parasaurolophus. However, von Huene also considered as “lambeosaurids” Trachodon (nomen
dubium), Jaxartosaurus and Nipponosaurus.
In a monograph on the cranial morphology of North American hadrosaurids, Ostrom
(1961) supported the re-establishment of the Saurolophinae as a third subfamily of hadrosaurids.
In this group he included Brachylophosaurus, Prosaurolophus and Saurolophus. Ostrom’s
distinguishing feature for these three genera was the possession of what he called a “pseudo-
narial” crest. He pointed out that this cranial crest was not part of the nasal passage, in contrast to
the “true” crests of lambeosaurines. He regarded the Saurolophinae as intermediate between the
non-crested Hadrosaurinae and the crested Lambeosaurinae. In the Hadrosaurinae, Ostrom
included the genera Anatosaurus (= Edmontosaurus), Edmontosaurus and Kritosaurus (=
Gryposaurus). Within the Lambeosaurinae, Ostrom recognized the following genera:
Cheneosaurus (= juvenile Hypacrosaurus), Corythosaurus, Hypacrosaurus, Lambeosaurus,
Parasaurolophus and Procheneosaurus (= juvenile Corythosaurus and Lambeosaurus).
Furthermore, he suggested that Claosaurus, referred to Hadrosauridae incertae sedis, might have
been the most primitive member of the Hadrosauridae (Fig. 1.4A). He based this idea on the
primitive morphology of the postcrania and dentition of this genus, as well as its relatively low
stratigraphic position in the Niobrara Chalk Formation (the oldest recorded hadrosaur known at
that time).
Morris (1973) recognized five independent lineages within hadrosaurids (Fig. 1.4B). In
the first one, a stock composed of Anatosaurus (= Edmontosaurus) annectens gave rise in the
Campanian to Gryposaurus notabilis and Kritosaurus navajovius and in the Maastrichtian to E.
regalis. In the second lineage Prosaurolophus appeared as the ancestor of Saurolophus. In the
third lineage Corythosaurus evolved into Hypacrosaurus. In the fourth lineage Lambeosaurus
evolved into “Lambeosaurus” laticaudus. Morris (1973) interpreted the deep tails of
Hypacrosaurus and “L.” laticaudus as an adaptation for living in aquatic habitats. Finally, the
fifth lineage consisted solely of Parasaurolophus. No ancestral taxon was hypothesized for
Parasaurolophus, the other lineages or hadrosaurids as a whole.
A few years later, Dodson (1975) published a seminal paper on the taxonomy, ontogeny
and sexual dimorphism of lambeosaurine dinosaurs from the Oldman Formation of southern
Aberta, Canada. This author measured 45 landmarks in 36 skulls from 12 species corresponding
to three lambeosaurine genera, Procheneosaurus, Corythosaurus and Lambeosaurus. The data
13
were then used to conduct a series of bivariate and multivariate statistics. Dodson concluded that
only three taxa were valid: Corythosaurus casuarius, Lambeosaurus lambei and L.
magnicristatus. Redundant taxa were found to represent subadult specimens and sexual morphs
of those three species.
In an often-cited paper on the functional morphology of the facial and supraorbital
structures of hadrosaurid crania, Hopson (1975) supported Sternberg’s division of the
Hadrosauridae into the two subfamilies Hadrosaurinae and Lambeosaurinae (Fig. 1.5A).
However, he divided the Hadrosaurinae into three informally named groups on the basis of their
facial anatomy. The first group was the “kritosaurs”, composed of Kritosaurus (= Gryposaurus,
excluding K. navajovius), Brachylophosaurus and Aralosaurus. The second group, or
“edmontosaurs”, was composed of Anatosaurus (= Edmontosaurus annectens) and
Edmontosaurus (E. regalis). The third group was the “saurolophines”, composed of
Prosaurolophus, Saurolophus, Tsintaosaurus and Lophorhothon. Hopson (1975) regarded the
premaxillonasal structure of “kritosaurs” as the ancestral condition from which the deeply
excavated narial region (circumnarial fossa) of other forms derived. He also considered the genus
Lophorhothon, described by Langston in 1960 from strata corresponding to the Selma Formation
of Alabama (southern U.S.), as intermediate between “kritosaurs” and “saurolophines”. Thus,
Hopson regarded as primitive the nasal arch and lightly incised circumnarial fossa of
Gryposaurus. He also maintained that the derived, deeply excavated circumnarial fossae of
“edmontosaurs” and “saurolophines” evolved independently in several hadrosaurid lineages.
Furthermore, he suggested that the enclosure of the nasal passages of lambeosaurines might have
been derived from the deep circumnarial structure of a hypothetical “edmontosaur”-like ancestor.
Specifically, he pointed out that the laterally exposed circumnarial fossa would have undergone
caudodorsal migration into the supraorbital region of the skull, while experiencing further
excavation and vaulting to be enclosed by bone and become the lateral diverticula inside the
hollow crests of lambeosaurines (Hopson, 1975: Fig. 9).
A year later, Brett-Surman (1975) concluded a Master’s thesis on the appendicular
skeleton of hadrosaurid dinosaurs, where he provided a comprehensive taxonomic revision and
phylogeny for these animals. This work represents one of the landmarks of hadrosaurid studies
because it included taxa from Europe, Asia and South America and showed that the postcrania of
hadrosaurids had a greater utility for diagnosing taxa and elucidating phylogenetic relationships
14
than previously recognized. Brett-Surman recognized five morphotypes of ilia, four morphotypes
of ischia and two morphotypes of tibiae, fibulae and tarsi.
Brett-Surman (1975) recognized only the Hadrosaurinae and Lambeosaurinae
subfamilies, merging the Saurolophinae with the former subfamily. Within hadrosaurines, Brett-
Surman recognized three lineages. The first one was the Edmontosaurus lineage, that also
included Shantungosaurus, Tanius, Hadrosaurus, Claosaurus, Telmatosaurus, Gilmoreosaurus
and Secernosaurus. The second lineage, or Kritosaurus group, also included Aralosaurus,
Lophorhothon and Brachylophosaurus. The third lineage was composed of Prosaurolophus and
Saurolophus. Within lambeosaurines, two lineages were recognized. The Corythosaurus lineage
also included Jaxartosaurus, Lambeosaurus and Hypacrosaurus. The second group, or
Parasaurolophus lineage, also included Bactrosaurus and Tsintaosaurus.
The phylogeny proposed by Brett-Surman was published in 1979 and represented the first
evolutionary hypothesis to encompass all genera known at the time from all continents (Fig.
1.5B). Furthermore, he specified Asian species as ancestors for the Hadrosauridae and its major
groups. Thus, Probactrosaurus was placed as the base of the tree as the ancestor of all
hadrosaurids. Among lambeosaurines, Brett-Surman proposed Jaxartosaurus as the ancestor of
the Corythosaurus lineage and Bactrosaurus as the ancestor of the Parasaurolophus lineage. On
the hadrosaurine tree, Gilmoreosaurus was regarded as the ancestral species for this subfamily.
Brett-Surman characterized the Hadrosaurinae on the basis of a relatively low sacral neural
spines with a length/width ratio less than 4.5, ventrally grooved sacral centra, long and gracile
limb elements, absence of distal ischial expansion, prepubic process with a long and thin
proximal constriction, mean length/height ratio of the ilium of 4.4, convoluted or slightly
expanded premaxilla, long and straight mandible and, in some taxa, presence of solid narial
crests. In contrast, the Lambeosaurinae was characterized by having relatively high sacral neural
spines with a length/width ratio greater than 4.5, ventrally ridged sacral centra, shorter and more
robust limb elements, expanded distal end of the ischium, prepubic process with a short and thick
proximal constriction, ilium with a mean length/height of 5.4, greatly expanded premaxilla,
shorter and anteriorly decurved mandible and hollow cranial crests.
Brett-Surman returned to studying hadrosaurids in 1989 in his doctoral dissertation with a
study of greater scope. He revised all cranial and postcranial characters previously used in
hadrosaurid systamatics and presented a more formal nomenclature and classification of the
15
same groups that he had previously recognized. Lineages were given the rank of tribe. Thus, the
Hadrosaurinae was composed of the tribes Edmontosaurini, Hadrosaurini and Saurolophini,
whereas the Lambeosaurinae was intergrated by the tribes Corythosaurini and Parasaurolophini.
Numerous characters were provided as diagnostic of the Hadrosauridae, Hadrosaurinae,
Lambeosaurinae and the five tribes. Brett-Surman suggested a series of evolutionary trends from
the Edmontosaurini to the Parasaurolophini. In the skull, he proposed that the general trend was
towards decreasing the length of the dentary, the size of the external nares and the circumnarial
fossae, while the nasal passages became enlarged. In the postcrania, Brett-Surman hypothesized
changes towards a more robust pelvic girdle, with deeper ilia and pubes and increased height of
dorsal and sacral neural spines. This study became one of the reference works on hadrosaurid
systematics during the end of the twentieth century.
The Cladistic Period
As indicated before, a drawback common to all phylogenetic hypotheses published before

the application of cladistic methods was the implicit subjective weighting of characters.
Scientists seemed to see characters as evolving in one direction, with no consideration for
reversals or homoplasy as having ocurred. Researchers used personal judgment to consider
particular characters as those from which groups were established and evolutionary trends
hypothesized. Character conflict was not an issue because researches subjectively decided which
traits dictated taxonomic diagnoses, systematic classifications and evolutionary hypotheses.
Nevertheless, with the advent of phylogenetic systematics (Hennig, 1966), researchers had a tool
to minimize subjectivity and conduct more explicit and operative analyses of evolutionary
relationships where all observed characters, from the most conspicuous to the most obscure and
apparently insignificant, could play a role in the formulation of phylogenetic hypotheses.
Norman (1984) conducted the first cladistic analysis to include hadrosaurids. Although he
focused on the Ornithischia as a whole, his phylogeny supported the monophyly of hadrosaurids
as the most derived clade within the Ornithopoda. Sereno (1986) produced another early
cladogram resolving the phylogenetic relationships of the major clades of ornithischian
dinosaurs. In this tree, he showed a monophyletic Hadrosauridae with Ouranosaurus, a very
unique form from the lower Cretaceous of Niger, as its sister taxon. Together, the latter two taxa
16
formed the clade, of superfamily rank, Hadrosauroidea. A more inclusive clade, of hyperfamily
rank and composed of the Hadrosauroidea and Iguanodon, was the Iguanodontoidea (Hay, 1902;
later Sereno [1997] would substitute this name for Hadrosauriformes to avoid having two taxa of
different ranks with the same suffix “–oidea”).
The first cladistic analysis to focus on the relationships within the Hadrosauridae was that
of Weishampel and Horner (1990) (Fig. 1.6A). These authors provided a concise but
comprehensive review of knowledge on the taxonomy, phylogeny and paleobiology of
hadrosaurids accumulated up to that time. Their phylogenetic tree showed a monophyletic
Hadrosauridae composed of two major clades, the Hadrosauridae and the Lambeosaurinae.
Within hadrosaurines, there were two clades. One of them was composed of Gryposaurus,
Aralosaurus, Maiasaura and Brachylophosaurus. The other clade included Saurolophus,
Prosaurolophus, Lophorhothon, Edmontosaurus and Shantungosaurus. The Lambeosaurine
included Parasaurolophus and a subclade composed of Corythosaurus, Hypacrosaurus and
Lambeosaurus. Bactrosaurus was tentatively placed within lambeosaurines. Telmatosaurus,
followed by Gilmoreosaurus, were regarded as the most primitive hadrosaurids and were placed
outside hadrosaurines and lambeosaurines. Hadrosaurid monophyly was based on the following
synapomorphies: three or more teeth per alveolus, elevation of the dorsal process of the maxilla
and reposition of the antorbital fenestra, loss of surangular and paraquadrate foramina, fusion of
the palpebral to the dorsal orbital margin, eight to ten sacaral vertebrae and reduction of the shaft
of the prepubic process, among other characters. Hadrosaurines were characterized by transverse
expansion of the premaxilla, circumnarial fossa and ventrally grooved sacral centra.
Lambeosaurines were diagnosed as having a supraorbitally positioned nasal cavity, jugal with
truncated maxillary articulation, dentary teeth with root-crown angle greater than 145º, elongated
neural spines, robust appendicular elements and prominently footed ischium. Although
composed mostly of North American taxa, this phylogeny became one of the reference studies
on hadrosaurid evolutionary history in the years that follow.
Horner (1985; 1990) proposed a diphyletic origin for hadrosaurids (Fig. 1.7A). He
redefined the Hadrosauridae to include only hadrosaurines and the Lambeosaurinae was elevated
to family rank and renamed as Lambeosauridae. He used the name Hadrosauria for the clade
composed of Hadrosauridae and Iguanodon and created Lambeosauria for the clade composed of
Lambeosauridae and Ouranosaurus. The Lambeosauria and Hadrosauria were then united as
17
sister taxa under the Iguanodontoidea. Horner’s argument in support of his diphyletic hypothesis
was that Iguanodon shared more characters with his Hadrosauridae than with “lambeosaurids”,
whereas Ouranosaurus shared more characters with the latter group than with remaining
hadrosaurids. According to Horner (1990), the most important synapomorphies of the
Hadrosauria were the rostrodorsal process of the maxilla, the extension of the rostrodorsal bar of
the nasal over the narial foramen, near contact or fusion of the supraorbitals to the postorbital,
frontal and prefrontal, the folding of the caudodorsal surface of the squamosal, crown-root angle
of dentary teeth less than 130º, grooved ventral sacral centra, scapular blade with parallel
borders, long and narrow prepubic process and a ischial shaft that narrows caudal to the obturator
process. On the other hand, synapomorphies of the Lambeosauria included a rostroventral
maxillary shelf, absence of premaxillary foramina, ventrally excavated nasals, dorsolateral
bosses on the supraoccipital for reception of the squamosals, tall neural spines and a short
prepubic process with a dorsoventrally expanded distal blade. However, Horner provided no
explicit mention of the methodology used to infer his phylogeny.
In a monograph on the cranial morphology of Prosaurolophus, Horner (1992) presented a
phylogeny of the Hadrosauridae (sensu Horner, 1990) (Fig. 1.7B). Four clades of subfamiliar
rank were considered: the Maiasaurinae (Brchylophosaurus and Maiasaura), the Gryposaurinae
(species of Gryposaurus), the Saurolophinae (Kritosaurus, Saurolophus and Prosaurolophus)
and the Edmontosaurinae (Edmontosaurus and Anatotitan copei [= E. annectens]). These clades
formed an unbalanced cladogram and were nested within progressively more exclusive clades,
evolving sequencially from the Maiasaurinae to the Edmontosaurinae. Again, there was no
mention of the use of a method phylogenetic inference. Synapomorphies were provided for each
clade but character optimization appears to have been conducted by personal judgement. Thus,
characters were implicitly weighted (e.g., Kritosaurus was regarded as primitive within the
Saurolophinae due to the less developed nature of its crest, undermining the potential role of
other postcranial and cranial characters in the evolution of these animals).
Horner et al. (1992) proposed a series of anagenetic evolutionary changes within the
lineages of several dinosaur clades that lived during the Campanian-Maastrichtian of western
North America at the east of the Rocky Mountain front. In the case of hadrosaurids, these authors
hypothesized that Hypacrosaurus stebingeri represented a transitional metaspecies between a
taxon closely related to Lambeosaurus from the Campanian Judith River Formation and H.
18
altispinus from the lower Maastrichtian Horseshow Canyon Formation. Horner and colleagues
saw no autapomorphies in H. stebingeri but rather a combination of characters present in
Lambeosaurus and H. altispinus.
One of the most cited phylogenetic analysis of hadrosaurids is that of Weishampel et al.
(1993) (Fig. 1.6B). This was the first published phylogeny of these animals that included a
matrix and a list of the characters used in the phylogenetic inference. These authors provided
further support to the monophyly of the Hadrosauridae and its two major subclades, the
Hadrosaurinae and Lambeosaurinae. Consequently, the monophyly of these three clades became
firmly established and widely accepted. Weishampel and colleagues erected a new name, the
Euhadrosauria, for the clade composed of the Hadrosaurinae and the Lambeosaurinae. Their
Hadrosauridae was a more inclusive clade composed of the European genus Telmatosaurus and
the Euhadrosauria. Hadrosaurid relationships were as in Weishampel and Horner (1990).
Synapomorphies of the Hadrosauridae included the elevation of the dorsal process of the maxilla,
accompanied by migration of the antorbital fenestra onto the dorsal maxillary surface, absence of
paraquadratic and surangular foramina, medially located surangular, narrow maxillary teeth,
three or more dentary alveolar positions and reduction of the dorsal margin of the scapular blade.
Euhadrosaurian synpomorhies included the absence of denticulate premaxillary oral margin,
quadrate with narrow mandibular condyle, narrow dentary teeth, large single carina on dentary
teeth and angular deltopectoral crest.
In a brief abstract, Forster and Sereno (1994) confirmed the monophyly of the
Hadrosauridae and its two major subclades, the Hadrosaurinae and Lambeosaurinae. In their
work, the Hadrosaurinae contained two clades. The first one was composed of Maiasaura,
Edmontosaurus and Anatotitan (= E. annectens), characterized by broad snouts. Remaining
hadrosaurines (Gryposaurus, Kritosaurus, Lophorhothon, Saurolophus and Prosaurolophus)
formed a clade characterized by narial fossae that join along the dorsal margin of the internarial
bar. In a later abstract, Forster (1997a) restricted the name Hadrosauridae to include only the
Hadrosaurinae and the Lambeosaurinae. All taxa lying outside this clade were referred to the
clade Iguanodontia.
During the end of the twentieth and beginning of the twenty-first centuries phylogenetic
analyses became customary in systematic studies. Hadrosaurid phylogenies proliferated in the
19
literature with the revision, erection and description of old and new taxa. All analyses agreed in
that the ancestry of hadrosaurids was to be found among basal iguanodontoideans.
The analysis of Head (1998) included a few taxa less derived than hadrosaurines and
lambeosaurines from North America, Europe and Asia. In his phylogeny (Fig. 1.8A), the
Hadrosauridae was a more inclusive clade because, besides the Hadrosaurinae and the
Lambeosurinae, contained Bactrosaurus, Gilmoreosaurus, Telmatosaurus and the then newly
discovered Protohadros. The analyses of Weishampel et al.(1993) and Head (1998) share the
practice of arbitrarily giving special importance to a species by redefining the Hadrosauridae as
including that species, its most recent common ancestor and all remaining iguanodontoideans. In
the case of Weishampel et al (1993) that arbitrary “special species” was Telmatosurus
transsylvanicus; in Head (1998) it was Protohadros byrdi. In the latter study, Bactrosaurus was
found to be the sister taxon to the Lambeosaurinae and these two taxonomic units combined
formed a sister clade to the Hadrosaurinae. Hadrosaurid synapomorphies included the lateral
expansion of the rostrum, the elevation of the dorsal maxillary process and migration of the
antorbital fenestra to caudodorsal region of the maxilla, absence of paraquadratic foramen,
maxillary teeth symmetrical around a large median carina and “miniaturization” of maxillary
teeth. No further details were given regarding the relationships of hadrosaurine and lambeosurine
hadrosaurids.
Kirkland (1998) presented a phylogeny of hadrosaurids very similar to that of
Weishampel and Horner (1990). The only difference was the position of Eolambia, from the
Cenomanian Cedar Mountain Formation of Utah, western United States, as the first taxon to
branch off the main lambeosurine lineage (Fig. 1.8C). Head (2001) and Godefroit (2001) refuted
this hypothesis and provided a phylogeny where Eolambia occupied a more basal position
outside the Hadrosauridae (Fig. 1.8B).
Godefroit et al. (1998) agreed with Forster (1997a) in restricting the Hadrosauridae to the
much more stable clade composed of the Hadrosaurinae and Lambeosaurinae (Fig. 1.9A). In this
phylogeny, Bactrosaurus, Telmatosaurus and the Hadrosauridae formed the superfamily
Hadrosauroidea. This clade was diagnosed by the possession of more than one replacement tooth
per alveolus, nearly vertical articular surface of the occipital condyle, long and slender
basipterygoid processes of the basisphenoid that project ventral to the level of the occipital
condyle, cranial fontanellae present in subadult specimens, absence of surangular foramen,
20
miniaturization of maxillary teeth, single large carina on maxillary tooth crowns, ventrally
deflected preacetabular process of the ilium, developed supraacetabular process of the ilium,
fully open obturator foramen of the pubis and transverse widening of the proximal head of the
tibia extending to the diaphysis. The family Hadrosauridae was supported by the reduction of
premaxillary and predentary denticulations, narrow and sub-hemispherical mandibular condyle,
lack of indentation of the caudal border of the skull, medial rami of the squamosals nearly in
contact, mandibular diastema, rostrally inclined coronoid process of the dentary, dentary teeth
miniaturized and not recurved, dentary teeth with a large single carina, odontoid process free
from the axis and angular deltopectoral crest. Hadrosaurines were diagnosed by the possession of
reflected premaxillary oral margin, external nares enlarged up to 40% the basal skull length,
maxilla symmetrical in lateral view, circumnarial fossa extending into the nasal, crown-root
angle less than 130º, caudal sacral centra with ventral sulcus and ischial shaft tapering distally.
Finally, lambeosaurine hadrosaurids were diagnosed on the basis of the complete exclusion of
the frontal from the orbital margin, shortened parietal, lacrimal laterally “lapped” by the maxilla,
maxillary shelf, hollow supracranial crest, external naries entirely surrounded by the premaxilla,
rounded and truncated rostral process of the jugal, absence of premaxillary foramen and very tall
neural spines.
The analysis of Casanovas et al. (1999) was the first to include one of the only two
hadrosaurids named in Europe at the time, Pararhabdodon isonensis, from the Maastrichtian of
the south-central foredeep basin of northern Spain. These authors found P. isonensis to be the
most primitive lambeosaurine hadrosaurid (Fig. 1.9B). Characters uniting the Spaniard species to
the Lambeosaurinae included the presence of a maxillary shelf, truncated and jugal-maxilla
articulation and deltopectoral crest of the humerus projecting laterally and distally. Hadrosaurid
monophyly was supported by the elevation of the dorsal maxillary process, absence of
paraquadratic and surangular foramina, dental battery composed of at least 60 tooth families,
three of more dentary teeth per alveolus, a minimum of eight fused vertebrae in the sacrum and
reduced dorsal margin of the scapular blade. The Hadrosauridae was again composed of
Telmatosaurus and Euhadrosauria. The Hadrosaurinae was scored as a single taxonomic unit and
the Lambeosurinae was composed of P. isonensis, Parasaurolophus and the most derived
Corythosaurini.
21
In a monographic revision of the osteology and systematics of the Chinese species
Shantungosaurus giganteus, Hu et al. (2001) presented a rather unique topology of the
relationships among hadrosurids. These authors included a large number of taxa and showed a
monophyletic Hadrosauridae composed of the Lambeosaurinae and an unnamed clade divided
into the Saurolophinae and the Hadrosaurinae (Fig. 1.10A). Hadrosaurines included two main
subclades. One of them was composed of Eurasian forms such as Telmatosaurus,
Shantungosaurus and Tanius and North American taxa such as Claosaurus and Edmontosaurus.
The other hadrosaurine clade was composed of Brachylophosaurus, Maiasaura, Gryposaurus
and Aralosaurus. The Saurolophinae included Kritosaurus, Lophorhothon, Saurolophus,
Prosaurolophus and a clade composed of Tsintaosaurus and Jaxartosaurus. The
Lambeosaurinae included previously unheard relationships, such as Bactrosaurus being sister
taxon to Nipponosaurus and Parasaurolophus as sister taxon to Hypacrosaurus.
In the same year, another substantially different phylogeny was proposed by Wagner
(2001), where the Hadrosaurinae-Lambeosaurinae dichotomy was not recovered (Fig. 1.10B).
Instead, hadrosaurid evolution consisted of a succession of progressively more derived clades. A
newly defined Hadrosaurinae, restricted to include only Hadrosaurus and Kritosaurus (=
Gryposaurus spp. and K. navajovius), was the first clade to branch off the main hadrosaurid
stock. Next to evolve was Brachylophosaurus and the Saurolophidae. The latter clade was
composed of a newly defined Saurolophinae (Saurolophus and Edmontosaurus) and
lambeosaurines. Finally, the Lambeosaurinae included Hypacrosaurus, Lambeosurus,
Jaxartosaurus and Parasaurolophus as the most derived taxon. This phylogeny echoed Hopson’s
(1975) ideas of the evolution of the facial skeleton, in that there was an evolutionary trend
among hadrosaurids towards increasing the enclosure and dorsal migration of the nasal passages.
This changes ranged from the slightly incised circumnarial fossa of Kritosaurus to the
completely enclosed and convoluted structure of the nasal passage of Parasaurolophus.
However, Wagner’s phylogeny was not the outcome of a tree search following an optimality
criterion and, as in the case of Horner (1990; 1992), clades were apparently formed on the basis
of personal judgement of primary homology assessments.
Wagner redefined the Hadrosauria as “the most inclusive clade containing Hadrosaurus
foulkii Leidy 1858, but not Iguanodon bernissartensis Boulenger 1881 nor Ouranosaurus
nigeriensis Taquet 1976” (Wagner 2001: p. 114). The Hadrosauroidea was redefined as “the
22
most inclusive clade containing Hadrosaurus foulkii Leidy 1858, Bactrosaurus johnsoni Gilmore
1933, Gilmoreosaurus mongoliensis Gilmore 1933 and Telmatosaurus transsylvanicus Nopcsa
1900” (p. 117). The Hadrosauridae was redefined as “the most exclusive clade containing
Hadrosaurus foulkii Leidy 1858, Saurolophus orborni Brown 1912 and Lambeosaurus lambei
Parks 1923” (p. 123). Furthermore, Wagner synonymized several monotypic genera under a
single genus. For example, Prosaurolophus maximus, Maiasaura peeblesorum and
Corythosaurus casuarius were included under Saurolophus (as S. maximus), Brachylophosaurus
(as B. peeblesorum) and Hypacrosaurus (as H. casuarius), respectively.
Norman (2002) conducted a phylogenetic analysis where he focused on the relationships
of iguanodontoidean taxa outside the Hadrosauridae (Fig. 1.11A). In his phylogeny,
Probactrosaurus appeared as the sister taxon to the Hadrosauridae and Protohadros, Eolambia
and Altirhinus represented progressively less derived outgroup taxa. Telmatosaurus and
Bactrosaurus were hadrosaurids leading, in this order, to the Euhadrosauria (sensu Weishampel
et al. 1993).
The analysis of Hai-lu You et al. (2003) included several basal hadrosauroid genera from
Asia and also focused on the relationships of hadrosaurid outgroup taxa. In their topology (Fig.
1.11B), the Hadrosauroidea contained, from less to more derived, Equijubus, Probactrosaurus,
Bactrosaurus, Protohadros. Telmatosaurus and the Hadrosauridae. These authors emphasized a
step-wise pattern of evolution in the development of mobile maxillae and elaborate dental
batteries from basal hadrosauroids to the crown clade Hadrosauridae.
One of the more recent phylogenies to focus on the relationships of lambeosaurine
hadrosurids was that of Suzuki et al. (2004) (Fig. 1.12A). In their analysis, Parasaurolophus was
the first genus to split from the lambeosaurine lineage. Next in evolve was a polytomy composed
of Hypacrosaurus stebingeri and a clade formed by H. altispinus and Nipponosaurus. Finally,
another clade was recovered including Corythosaurus and Lambeosaurus as sister taxa.
Bactrosaurus was inferred to be the closest outgroup to the Lambeosaurinae. No hadrosaurine
taxon was included in this analysis.
Horner et al. (2004) updated the review on the Hadrosauridae presented in Weishampel
and Horner (1990) (Fig. 1.12B). In the new phylogeny, the Hadrosauridae was composed of
Telmatosaurus and the Euhadrosauria. The latter clade was composed of the Hadrosaurinae and
the Lambeosaurinae. Within the Hadrosaurinae, Lophorhothon was considered as the most basal
23
form. Two hadrosaurine clades were recognized. One of the clades was composed of
Brachylophosaurus, Maiasaura, Gryposaurus, Edmontosaurus and Prosaurolophus. The other
hadrosaurine clade consisted of a polytomy of Naashoibitosaurus (= Kritosaurus navajovius),
Saurolophus and “Kritosaurus” asutralis. On the other part of the tree, lambeosaurines included
an unbalanced tree composed of Tsintaosaurus, Parasaurolophus and a polytomy formed by
Lambeosaurus, Hypacrosaurus and Corythosaurus. Of interest was the incorporation of taxa
from all continents, including K. australis from Argentina. The monophyly of the Hadrosauridae
was supported by the presence of three or more replacement teeth per alveolus, dentary tooth row
terminating caudal to the coronoid process, caudalmost termination of the dentary well behind
the coronoid process, absence of surangular foramen, absence of ectopterygoid-jugal articulation,
absence or fusion of the supraorbital to the margin of the orbit, elevation of the
postzygapophyseal arches of cervical vertebrae well above the level of the neural arch, long and
dorsally arched postzygapophyseal processes, long cranioventral process of the coracoid that
extends well below the glenoid, dorsoventrally narrow proximal region of the scapula with
horizontally projecting acromion process, scapula with notched cranioventral corner and reduced
articular are for the coracoid and deep intercondylar groove of the femur, whose edges nearly or
completely meet cranially to form an extersor tunnel. The Hadrosaurinae was supported by a
single unambiguous synapomorphy, the presence of a caudal margin in the circumnarial fossa.
Horner and colleagues recognized the great degree of homoplasy present among the members of
that clade. The Lambeosaurinae was united by only four unambiguous synapomorphies:
maxillary rostrodorsal shelf, gooeve on the caudolateral process of the premaxilla that
communicates with the maxillary foramen on the maxilla-premaxilla contact surface and with the
lateral fontanelle of the crest, low maxillary apex and a parietal crest that is less than half the
length of the supratemporal fenestra.
In revising the cranial anatomy of Brachylophosaurus canadensis, Prieto-Marquez (2005)
confirmed the close relationship between this species and Maiasaura peeblesorum. His
phylogeny (Fig. 1.13A) focused on the interrelationships among hadrosaurine taxa and depicted
Edmontosaurus as the sister taxon to a clade composed of Gryposaurus, Prosaurolophus and
Saurolophus. A year later, Prieto-Marquez and colleagues (2006a) revised the phylogenetic
position of Hadrosaurus foulkii and supported the inclusion of this species within the
24
Hadrosauridae (Fig. 1.13B). The majority rule cladogram suggested the possibility that H. foukii
was a member of the Hadrosaurinae.
The phylogenetic analysis of Prieto-Marquez et al. (2006b) was one of the most
comprehensive ever conducted, in terms of the number of hadrosaurid taxa included and skeletal
representation as coded in the characters used (Fig. 1.13C). This phylogeny supported the clade
Hadrosauroidea as containing a number of species more derived than Iguanodon but more
plesiomorphic than hadrosaurids. The Spaniard genus Pararhabdodon was found to be the sister
taxon to the Hadrosauridae. The latter clade was restricted to include the Hadrosaurinae and
Lambeosaurinae. Within the Hadrosaurinae, Edmontosaurus appeared as the first lineage to
evolve. Two other clades were recognized: one composed of Brachylophosaurus and Maiasaura
and another integrated by Gryposaurus, Saurolophus and Prosaurolophus. Within
lambeosaurines, three lineages were inferred with polytomic relationships. One of them gave rise
to Parasaurolophus and Olorotitan. Another lineage was represented by Hypacrosaurus
altispinus. Finally, a third lineage included a succession of progressively more inclusive clades
that culminated in the sister taxa Lambeosaurus and Corythosaurus. The genus Hypacrosaurus
was found to be paraphyletic.
In recent years, the work of Godefroit and colleagues (2000; 2001; 2003; 2004a, b; 2008)
has had considerable impact in our knowledge of hadrosaurid taxonomic diversity, phylogeny
and biogeography. Several expeditions working in the Upper Maastrichtian of the Amur region
of far eastern Russia and adjacent regions of China led to the discovery of various new
lambeosaurine (Charonosaurus, Olorotitan, Shaliyania) and hadrosaurine (Kerberosaurus,
Wulagasaurus) genera. With the description of each new taxon, Godefroit and colleagues
inferred a new phylogeny resolving lambeosaurine and/or hadrosaurine relationships. The
topology of Godefroit et al. (2008) recovers the clades inferred in all previous collaborative
works by Godefroit and various colleagues. This phylogeny (Fig. 1.14B) is one of the most
comprehensive in terms of the number of included taxa, with a rather balanced number of genera
from North America and Asia. The Hadrosaurinae was composed of Wulagasaurus and a clade
integrated by two main subclades. The first subclade was composed of Maiasaura and
Brachylophosaurus. The second subclade included Gryposaurus, Kerberosaurus,
Prosaurolophus and Edmontosaurus. On the other side of the cladogram, lambeosaurines
included a succession of progressively more derived Asian genera: Aralosaurus, Tsintaosaurus,
25
Jaxartosaurus, Amurosaurus and a derived clade composed of two lineages. The first lineage
was integrated by Lambeosaurus and a polytomy of mostly dome-crested genera, Corythosaurus,
Hypacrosaurus and Olorotitan. The other derived clade of lambeosurines was composed of
Parasaurolophus, Charonosaurus and Sahaliyania.
A recent analysis of hadrosaurine relationships by Gates and Sampson (2007) showed
Lophorhothon as the first taxon to evolve within this group (Fig. 1.14A). The Hadrosaurinae was
composed of two major clades. One clade included Edmontosaurus, Saurolophus and
Prosaurolophus. The other clade consisted of Naashoibitosaurus (= Kritosaurus navajovius),
four species of Gryposaurus with unresolved relationships and the Brachylophosaurus-
Maiasaura clade.
Evans and Reisz (2007) presented a species-level phylogeny of lambeosaurine
hadrosaurids (Fig. 1.15A). Their topology showed a succession of Asian taxa (from Aralosaurus
tuberiferus to Amurosaurus riabinini) as outgroups to a specious dichotomic clade. The first
clade of this dichotomy was composed of Charonosaurus jiayinensis and three species of
Parasaurolophus. The other clade was composed of Lambeosaurus and un unresolved
trichotomy composed of Olorotitan, Corythosaurus and a monophyletic Hypacrosaurus. Their
phylogeny revealed that a greatly enlarged cranial crest evolved independently at least three
times within the Lambeosaurinae. This conclusion was reached after the comparison of the size
of the crest between sister taxa, where one of the species showed a relative enlargment of the
crest (e.g., L. magnicristatus relative to L. lambei and P. tubicen relative to P. cyrtocristatus).
Finally, the most recent phylogeny of lambeosaurine interrelationships was presented by
Gates et al. (2007) (Fig. 1.15B). This phylogeny was very similar to that of Evans and Reisz
(2007), which is hardly surprising since Gates and colleagues included almost the same taxa and
used the same character list presented in the latter study. The analysis of Gates and colleagues
differed from that in Evans and Reisz (2007) in presenting a polytomy composed of Olorotitan
ararhensis, Lambeosaurus spp., Corythosaurus casuarius, Hypacrosaurus spp. and Velafrons
coahuilensis, and uniting Probactrosaurus gobiensis and Telmatosaurus transsylvanicus under
their own clade.
The future of systematic studies of hadrosaurid dinosaurs will need a greater emphasis in
studies of European, South American and Asian taxa if one is to expect to achieve a more
accurate and global understanding of the evolution of these animals. Furthermore, the present
26
treatise is intended to serve as the starting point of a new trend whereby characters are
documented and the data and methods for their definition made explicit. This, coupled with the
implementation of the new developments in the theory and practice of phylogenetic inference,
should bring us nearer to the true phylogenetic history of this successful group of dinosaurs.
27
A
Figure 1.1. The first remains of hadrosaurid dinosaurs, described by Joseph Leidy in 1856. A, dentary
crown of Trachodon mirabilis (nomen dubium; ANSP 9260). B, worn fragments of hadrosaurid teeth and
other unidentified parts, collected and described with the type of T. mirabilis.
28
1 cm
A
C
Figure 1.2. The first partial skeleton of a dinosaur found outside Europe; described by Joseph Leidy in 1858 and named Hadrosaurus foulkii.
A, dentary tooth crowns (ANSP 9201) in lingual view . B, composite of the fragmentary maxilla (ANSP 9203-4) and partial postcranial skeleton
(ANSP 10005), lateral view. C, dentary crowns (ANSP 9201) in mesial or distal views. Abbreviations: cdl = caudal vertebrae; cor = coracoid;
drs = dorsal vertebrae; fm = femur; hm = humerus; il = ilium; is = ischium; mx = maxilla; mttIV = metatarsal IV; ph(pb)III1 = pedal phalanx III1;
pb = pubis; rd = radius; ul = ulna;
29
Figure 1.3. Phylogenetic hypothesis of the relationships among hadrosaurid dinosaurs by Lull and Wright (1942). Redrawn after the latter.
30
A
Figure 1.4. Phylogenetic hypotheses of the evolutionary relationships among hadrosaurid dinosaurs by
A, Ostrom (1961) and B, Morris (1973). Redrawn after Ostrom (1961) and Morris (1973), respectively.
31
A
Figure 1.5. Phylogenetic hypotheses of the evolutionary relationships among hadrosaurid dinosaurs
by A, Hopson (1975) and B, Brett-Surman (1979). Redrawn after Hopson (1975) and Brett-Surman (1979),
respectively.
32
Iguanodon bernissartensis
Ouranosaurus nigeriensis
Telmatosaurus transsylvanicus
Gryposaurus notabilis
Maiasaura peeblesorum
Brachylophosaurus canadensis
Edmontosaurus regalis
33
Anatotitan copei
Prosaurolophus maximus
HADROSAURINAE
Saurolophus osborni
HADROSAURIDAE
Parasaurolophus walkeri
EUHADROSAURIA
Hypacrosaurus altispinus
Weishampel and Horner (1990) and Weishampel et al. (1993), respectively.

Corythosaurus casuarius
Lambeosaurus lambei
LAMBEOSAURINAE
Figure 1.6. First cladograms showing the phylogenetic relationships among hadrosaurid genera
or species by A, Weishampel and Horner (1990) and B, Weishampel et al. (1993). Redrawn after
B
A
Maiasura peeblesorum
Brachylophosaurus goddwini
1992), respectively.
Iguanodon
Gryposaurus latidens
MAIASAURINAE
Gryposaurus incurvimanus
Hadrosauridae
Kritosaurus navajovius
HADROSAURIA
GRYPOSAURINAE
Saurolophus angustirostris
34
Saurolophus osborni
Ouranosaurus
Prosaurolophus blackfeetensis
HADROSAURIDAE
IGUANODONTOIDEA
SAUROLOPHINAE
Edmontosaurus annectens
Lambeosauridae
Edmontosaurus saskatchewanensis
LAMBEOSAURIA
Anatotitan copei
EDMONTOSAURINAE
Figure 1.7. Phylogenetic hypotheses of hadrosaurid relationships by John R. Horner. A, diphyletic

A
origin of hadrosaurids (Horner, 1990). B, cladogram of Horner (1992). Redrawn after Horner (1990;
Iguanodon
Iguanodon
Ouranosaurus
Ouranosaurus
Probactrosaurus
Probactrosaurus
Protohadros
Telmatosaurus
Gilmoreosaurus
Gilmoreosaurus
Telmatosaurus
Gryposaurus
Bactrosaurus
Maiasaura
Lambeosaurinae
Brachylophosaurus A
Hadrosaurinae
Edmontosaurus
HADROSAURIDAE
Iguanodon
35
Anatotitan
Ouranosaurus
Prosaurolophus
Altirhinus
Saurolophus
HADROSAURINAE
Probactrosaurus
Eolambia
Eolambia
HADROSAURIDAE
Bactrosaurus
Protohadros
EUHADROSAURIA
Parasaurolophus
Telmatosaurus
Corythosaurus Gilmoreosaurus
Hypacrosaurus Bactrosaurus
Lambeosaurus Lambeosaurus
LAMBEOSAURINAE
C
Hadrosaurinae
B
Figure 1.8. Phylogenetic hypotheses of hadrosuaroid and hadrosaurid relationships by A, Head (1998),
B, Head (2001) and C, Kirkland (1998). Redrawn after Head (1998; 2001) andKirkland (1998), respectively.
Lambeosaurinae
Hadrosaurinae
Telmatosaurus
Ouranosaurus
Camptosaurus
Bactrosaurus
Iguanodon
IGUANODONTIDAE
HADROSAURIDAE
HADROSAUROIDEA
ANKYLOPOLLEXIA
A
Parasaurolophus
Corythosaurini
Pararhabdodon
Hadrosaurinae
Telmatosaurus
Ouranosaurus
LAMBEOSAURINAE
EUHADROSAURIA
HADROSAURIDAE
Figure 1.9. Phylogenetic hypotheses of hadrosuaroid and hadrosaurid relationships by

A, Godefroit et al. (1998) and B, Casanovas et al. (1999). Redrawn after Godefroit et al. (1998)
and Casanovas et al. (1999), respectively.
36
Iguanodon Camptosaurus
Iguanodon
Lurdusaurus
Ouranosaurus
Altirhinus
Probactrosaurus
Probactrosaurus
Shantungosaurus
Eolambia
Tanius
Protohadros Mandschurosaurus
Nanyangosaurus Telmatosaurus
Bactrosaurus Anatosaurus
(2) = SAUROLOPHINAE
(1) = HADROSAURINAE
Telmatosaurus Edmontosaurus
Tanius Claosaurus
Aralosaurus
Pararhabdodon
Gryposaurus
Gilmoreosaurus
Maiasaura
Claosaurus
HADROSAURINAE
Brachylophosaurus
37
Lophorhothon
Kritosaurus
Secernosaurus Saurolophus
Kritosaurus Prosaurolophus
IGUANODONTIDAE
HADROSAURIA
Hadrosaurus Lophorhothon
(1)
Brachylophosaurus Tsintaosaurus
HADROSAURIDAE
Edmontosaurus Jaxartosaurus
Bactrosaurus
Saurolophus
(2)
HADROSAUROIDEA
Nipponosaurus
SAUROLOPHINAE
Hypacrosaurus
Corythosaurus
Lambeosaurus
EUHADROSAURIA
Parasaurolophus
Jaxartosaurus
Hypacrosaurus
Parasaurolophus
HADROSAURIDAE
Lambeosaurus
B, Wagner (2001). Redrawn after Hu et al. (2001) and Wagner (2001), respectively.
B
A
SAUROLOPHIDAE
LAMBEOSAURINAE
LAMBEOSAURINAE
Figure 1.10. Phylogenetic hypotheses of hadrosaurid relationships by A, Hu et al. (2001) and

Hypsilophodon Dryosaurus
Camptosaurus
Helvetica
Lurdusaurus
Camptosaurus
Mazongshan
Iguanodon Iguanodon b
Iguanodon a
Altirhinus
Ouranosaurus
Ouranosaurus
Altirhinus
Jinzhousaurus Eolambia
38
Protohadros
Equijubus
Probactrosaurus
Probactrosaurus
Telmatosaurus
Bactrosaurus Bactrosaurus
Corythosaurus
Protohadros
IGUANODONTIA
Parasaurolophus
Telmatosaurus
Saurolophus
IGUANODONTOIDEA
Corythosaurus Brachylophosaurus
Edmontosaurus
HADROSAUROIDEA
Prosaurolophus
HADROSAURIDAE
Figure 1.11. Phylogenetic hypotheses of hadrosuaroid relationships by A, Norman (2002) and

EUHADROSAURIA
B, Hai-lu You et al. (2003). Redrawn after Norman (2002) and Hai-lu You et al. (2003), respectively.
HADROSAURIDAE
Iguanodon
Camptosaurus
Probactrosaurus
Protohadros Iguanodon bernissartensis
Eolambia
Iguanodon atherfieldensis
Telmatosaurus
Ouranosaurus
Maiasaura
Brachylophosaurus Altirhinus
Edmontosaurus
Bactrosaurus
Gryposaurus
Prosaurolophus Parasaurolophus
39
Saurolophus
Hypacrosaurus stebingeri
Naashoibitosaurus
"Kritosaurus" australis
HADROSAURINAE
Lophorhothon Nipponosaurus
Tsintaosaurus
HADROSAURIDAE
Corythosaurus
EUHADROSAURIA
Parasaurolophus
Redrawn after Suzuki et al. (2004) and Horner et al. (2004), respectively.
Lambeosaurus Lambeosaurus magnicristatus
LAMBEOSAURINAE
Corythosaurus
Lambeosaurus lambei
Hypacrosaurus
B
A
Figure 1.12. Hadrosaurid phylogenies by A, Suzuki et al. (2004) and B, Horner et al. (2004).
LAMBEOSAURINAE
Iguanodon atherfieldensis Iguanodon
Iguanodon bernissartensis Protohadros
Probactrosaurus gobiensis
respectively.
Telmatosaurus
Maiasaura
Protohadros byrdi
Brachylophosaurus
Bactrosaurus johnsoni
H Gryposaurus
GIlmoreosaurus mongoliensis
Prosaurolophus
Telmatosaurus transsylvanivus
Tanius sinensis Saurolophus
Pararhabdodon isonensis Edmontosaurus

A
Brachylophosaurus canadensis Hypacrosaurus

40
Iguanodon
Gryposaurus spp.
Telmatosaurus
Saurolophus spp. Maiasaura
Prosaurolophus spp. Brachylophosaurus
Edmontosaurus spp. Gryposaurus

Kritosaurus
H
Lambeosaurus spp.
HADROSAUROIDEA
Saurolophus
HADROSAURINAE
Prosaurolophus
Amurosaurus riabinini Hadrosaurus
Hypacrosaurus stebingeri Edmontosaurus
HADROSAURIDAE
Hypacrosaurus altispinus Parasaurolophus
Lambeosaurus
HADROSAURIDAE
Parasaurolophus spp.
L
EUHADROSAURIA
Hypacrosaurus
Olorotitan ararhensis
Corythosaurus
Figure 1.13. Phylogenetic hypotheses of hadrosaurid relationships by A, Prieto-Marquez (2005),

C
L = Lambeosaurinae. Redrawn after Prieto-Marquez (2005) and Prieto-Marquez et al. (2006a, b),
B
LAMBEOSAURINAE
B, Prieto-Marquez et al. (2006a) and C, Prieto-Marquez et al. (2006b). Abbreviations: H = Hadrosaurinae;

Bactrosaurus Iguanodon
Wulagasaurus
Eolambia
Brachylophosaurus
Maiasaura Telmatosaurus
Gryposaurus Corythosaurus
Kerberosaurus
Lophorhothon
Saurolophus
Prosaurolophus Edmontosaurus spp.
Edmontosaurus
Prosaurolophus spp.
Anatotitan
HADROSAURINAE
41
Saurolophus spp.
Aralosaurus
Tsintaosaurus Naashoibitosaurus
Jaxartosaurus
Amurosaurus
Olorotitan Gryposaurus incurvimanus
HADROSAURIDAE
Hypacrosaurus
Corythosaurus
Gryposaurus monumentensis
Lambeosaurus
Sahaliyania Maiasaura
Redrawn after Gates and Sampson (2007) and Godefroit et al. (2008), respectively.
Charonosaurus
LAMBEOSAURINAE
Brachylophosaurus
Parasaurolophus
A
Figure 1.14. Hadrosaurid phylogenies by A, Gates and Sampson (2007) and B, Godefroit et al. (2008).
Probactrosaurus gobiensis Probactrosaurus gobiensis
Telmatosaurus transsylvanicus Telmatosaurus transsylvanicus
Gryposaurus incurvimanus Gryposaurus incurvimanus
Aralosaurus tuberiferus Aralosaurus tuberiferus
Tsintaosaurus spinorhinus Tsintaosaurus spinorhinus
Jaxartosaurus aralensis Jaxartosaurus aralensis
Amurosaurus riabinini Amurosaurus riabinini
Charonosaurus jiayinensis Charonosaurus jiayinensis
Parasurolophus tubicen Parasaurolophus cyrtocristatus
42
Parasaurolophus walkeri Parasaurolophus walkeri
Parasaurolophus cyrtocristatus Parasaurolophus tubicen
Olorotitan arharensis Nipponosaurus sachalinensis
Lambeosaurus magnicristatus Lambeosaurus lambei
Lambeosaurus lambei Lambeosaurus magnicristatus
Corythosaurus casuarius Olorotitan arharensis

LAMBEOSAURINAE
Hypacrosaurus altispinus Corythosaurus casuarius
Hypacrosaurus stebingeri Hypacrosaurus stebingeri
Redrawn after Evans and Reisz (2007) and Gates et al. (2007), respectively.
Velafrons coahuilensis Hypacrosaurus altispinus
B
A
Figure 1.15. Lambeosaurine phylogenies by A, Evans and Reisz (2007) and B, Gates et al. (2007).
CHAPTER 2
PHYLOGENY OF HADROSAURID DINOSAURS
Abstract
The Late Cretaceous hadrosaurids were the most derived clade of ornithopod dinosaurs.
Questions remain regarding the timing of their origin and which attributes might have allowed
these animals to diversify and colonize nearly all continents by the late Campanian. Parsimony
and Bayesian methods were implemented to elucidate the phylogenetic relationships of all
hadrosaurid species. These included specimens of 41 hadrosaurids from Europe, Asia and North
and South America. Outgroup taxa included tweleve iguanodontoidean species from Europe,
North America and Asia. New characters were defined and old ones revised on the basis of new
data collected from first hand examination of specimens. Traditional and geometric
morphometrics were applied to discover patterns of variation containing phylogenetic
information. Continuous geometries were studied using the Analysis of Planar Shapes Using
Geodesic Paths, a novel landmark-free method that considers the continuous non-linear geometry
of the bones. In this way, a total of 299 phylogenetically informative characters (205 cranial and
94 postcranial) were defined and documented, the most extensive character data set ever
constructed for hadrosaurid dinosaurs.
In general, parsimony and Bayesian analyses confirmed the dichotomic evolution of
hadrosaurids into Saurolophinae and the hollow-crested Lambeosaurinae. Saurolophines consisted
of “saurolophs” and “kritosaurs”. Lambeosaurines consisted of a succession of Eurasian
outgroups to two major clades, “parasaurolophs” and “amurosaurs” (“hypacrosaurs” and
“corythosaurs”). Hadrosauridae was redefined as the clade stemming from the most recent
common ancestor of Hadrosaurus foulkii and Parasaurolophus walkeri. Its monophyly was
unambiguously supported by an iliac supraacetabular process that projects lateroventrally
between half and three quarters of the dorsoventral depth of the central plate of the ilium and a
craniocaudally short supraacetabular process that is less than 55% the length of the central plate
of the ilium. The closest outgroup taxa to Hadrosauridae lived in eastern North America. The
hadrosaurid radiation and the divergence of saurolophines from lambeosaurines occurred no later
than the Santonian and was coincident with the evolution of a suite of mandibular characters (i.e.,
43
increased number of tooth families, presence of three teeth forming the dentary occlusal plane,
ventral offset of the oral predentary-premaxilla contact). These characters may have been key
innovations. These results suggest that feeding adaptations might have played a central role in the
diversification of hadrosaurids. Circumnarial fossae and cranial crests evolved prior to the
hadrosaurid radiation. Both of these characters were reconstructed to have evolved at the same
time.
Introduction
Hadrosauridae is one of the most derived and diverse clades of herbivorous dinosaurs
(Horner et al., 2004). These animals were very common during the Late Cretaceous of Europe,
Asia, the Americas and Antarctica (Forster, 1997; Case et al., 2000; Lund and Gates, 2006).
Their fossil record is the richest among non-avian dinosaurs. It includeds dozens of articulated
skeletons (Lull and Wright, 1942), thousands of disarticulated bones from multindividual
assemblages (Dodson, 1971; Varricchio and Horner, 1993), growth series (Horner et al., 2000),
soft tissue impressions (Osborn, 1912; Murphy et al., 2007), stomach contents, coprolites,
trackways (Currie et al., 1991) and eggs, nests and neonates (Horner, 1979; 1982; Horner and
Currie, 1994).
Hadrosaurids are characterized for having mediolaterally-expanded rostra (Morris, 1973),
hypertrophied nasal passages (Weishampel, 1981) and, in many species, conspicuous
supracranial crests (Lull & Wright, 1942; Ostrom, 1962). The function of some of these crests
might be related to species recognition and other display behaviors (Hopson, 1975; Evans, 2006).
The earliest record of Hadrosauridae is from the Santonian (Averianov and Nessov, 1995;
Averianov, 2007) and the largest taxonomic diversity of the clade spans from the late Campanian
to the early Maastrichtian (Gates, 2007).
Because of their abundance, widespread distribution, and excellent preservation,
hadrosaurids provide an excellent model for studying dinosaurian evolution and paleobiology.
For example, they can be used to elucidate how geological events affected the evolution and
distribution of terrestrial faunas, how their anatomy relates to biomechanical form, function and
ecology, and, owing to their large sample sizes, how anatomical variation relates to species
recognition, reproductive behavior, and population biology. These and many other questions can
44
be best approached in a phylogenetic context to provide an all-encompassing evolutionary
understanding for the group. However, a comprehensive and well-resolved phylogeny of
hadrosaurids has yet to be inferred.
An accurate and well-resolved phylogeny of hadrosurids is crucial if one is to understand
their origin, how they became so widespread, and how they acquired their unique suite of
skeletal attributes. However, a cladistic analysis with complete taxonomic sampling and using
modern methods of phylogenetic inference has yet to be conducted. Previous cladistic studies
have only included a limited number of taxa and/or a limited number of characters documenting
selected regions of the hadrosaurid anatomy owing to missing data (Head, 1998; Casanovas et
al., 1999; Godefroit et al., 2000; Hai-lu You et al., 2003). This has occurred inspite of theoretical
studies that have shown that the addition of phylogenetically informative characters generally
increses accuracy, even when there are numerous missing entries (Wiens, 1998; 2003). Other
studies have demonstrated that including a large number of taxa typically leads to an increase in
phylogenetic accuracy even for incompletely scored taxa (Wiens and Reeder, 1995; Hillis,
1996). In this regard, the inclusion of the hadrosaurid and outgroup hadrosaur species from Asia
(Godefroit et al., 1998; Norman, 2002; Hai-Lu You et al., 2003), South America (Brett-Surman,
1979; Bonaparte et al., 1984; Powell, 1987) and Europe (Casanovas et al., 1999; Lopez-Martinez
et al., 2001; Prieto-Márquez et al., 2006b) that have not been utilized could provide a more
accurate hypothesis of the evolutionary history of these animals.
Additionally, characters used in phylogenetic studies of hadrosaurids are rarely
informative enough to resolve lower level relationships. This is especially true for those from the
postcrania. This may be a consequence of our superficial understnding of intraspecific
(ontogenetic and individual) morphological skeletal variation. This problem may impact the
number and diagnoses of recognized species. Only a few studies have been conducted to shed
some light into this problem (Dodson, 1975; Chapman and Brett-Surman, 1990; Egi and
Weishampel, 1997; Brett-Surman and Wagner, 2007). However, none of them has been
comprehensive enough to encompass all skeletal areas and elements while at the same time being
detailed enough to include all possible sources of variation.
Finally, in virtually all the previous studies the data and methods used to discover and
define characters and character states were never explicit. Indeed, this is not exclusive to the
research in hadrosaurid systematics, but it is a common practice in vertebrate paleontology.
45
Thus, characters are very rarely documented and, at most, a list of characters and character states
is provided.
My cardinal goal in this study was to resolve the phylogenetic relationships of
Hadrosauridae at the specific level using complete taxonomic sampling and controlling for
ontogenetic and intraspecific variation. To achieve this, I conducted an exhaustive revision of the
patterns of variation in the crania and postcrania of hadrosaurids that contain phylogenetic
information. This resulted in the most extensive character data for a phylogenetic study of these
animals. The inferred phylogeny served as the underpinning to address a number of questions on
the evolutionary history of these animals. Specifically, these included the following. When did
the hadrosaurid radiation take place? Which apomorphies characterize the hadrosaurid radiation
and what is the order of acquisition of these apomorphies along the hadrosaur lineage? Is there
evidence for key innovations springbording the radiation of hadrosaurids?
Summary of Previous Studies on Hadrosaurid Systematics
The Pre-Cladistic Era

The first description of hadrosaurid remains consisted of tooth fragments collected in the
Judith River Formation of Montana (Trachodon mirabilis; Leidy, 1856). The family
Hadrosauridae was originally erected by Cope to classify the two species known at the time of
this group of ornithopods, Hadrosaurus foulkii and H. mirabilis (= Trachodon mirabilis) (Cope,
1869). Yet, scientists would not gain a complete knowledge of the skeletal anatomy of
hadrosaurids until the end of the nineteenth century, when the first complete articulated skeletons
were found (Cope, 1883; Lund and Gates, 2006). These first complete skeletons, corresponding
to the genus Edmontosaurus, came from late Maastrichtian formations of Dakota and Wyoming,
northern United States. In the next few decades, numerous complete skulls and skeletons were
collected in the western interior of the United States and southern Alberta, Canada (late
Campanian strata of the Dinosaur Provincial Park), including genera such as Kritosaurus
(Brown, 1910), Saurolophus (Brown, 1912), Hypacrosaurus (Brown, 1913), Corythosaurus
(Brown, 1914), Prosaurolophus (Brown, 1916), Gryposaurus (Parks, 1920), Parasaurolophus
(Parks, 1922) and Lambeosaurus (Gilmore, 1924).
46
Brown (1914) was the first to propose the split of hadrosaurids into the flat-headed forms
(Trachodontinae) and the crested forms (Saurolophinae). Lambe (1920) suggested that
hadrosaurids diversified into three main lineages, based on the morphology and the absence or
presence of cranial crests: Hadrosaurinae (= Trachodontinae or non-crested forms),
Saurolophinae (solid-crested forms) and the Stephanosaurinae (hollow-crested forms) (Lambe,
1920). Parks (1923) renamed the Stephanosaurinae as Lambeosaurinae, after invalidating the
name Stephanosaurus upon which the former subfamily had been based.
In their classic monograph on North American hadrosaurids, Lull and Wright (1942)
included Saurolophinae within Hadrosaurinae and supported the dichotomic classification of
hadrosaurids into Lambeosaurinae and Hadrosaurinae (Fig. 1.3, Chapter 1 of this dissertation).
These authors regarded “kritosaurs” (Gryposaurus and Kritosaurus) and “saurolophines”
(Saurolophus and Prosaurolophus) as morphologically intermediate between non-crested and
lambeosaurine hadrosaurids. The dichotomic classification of hadrosaurids was also supported
by Sternberg (1954), who, in addition, proposed the abandonment of Saurolophinae and the
inclusion of Prosaurolophus and Saurolophus within Hadrosaurinae.
Friedrich von Huene (1956) erected the name Hadrosauria to include hadrosaurids and all
those forms that were morphologically more similar to them than to Iguanodon. Hadrosauria
included the following families: Prohadrosauridae, Hadrosauridae, Saurolophidae,
Cheneosauridae and Lambeosauridae. In the Prohadrosauridae, von Huene included relatively
basal forms such as Telmatosaurus and Tanius. Within the Hadrosauridae, he included the
nomen dubium Ornithotarsus, along with “flat-headed” or “low-crested” taxa such as Thespesius
(= Edmontosaurus annectens), Anatosaurus (E. annectens), Edmontosaurus, Mandschurosaurus
(= Gilmoreosaurus, in part), Kritosaurus (= K. navajovius and Gryposaurus spp.) and
Brachylophosaurus. The Saurolophidae was composed of Saurolophus, Prosaurolophus and
Bactrosaurus. As in Lull and Wright (1942), the Cheneosauridae included the small (juvenile
lambeosaurine) forms Cheneosaurus and Procheneosaurus. Finally, von Huene’s
Lambeosaurinae was equivalent to the Lambeosaurinae of other authors in that it included those
hadrosaurs with large hollow crests, such as Lambeosaurus, Corythosaurus, Hypacrosaurus and
Parasaurolophus. However, von Huene also considered as “lambeosaurids” Trachodon (nomen
dubium), Jaxartosaurus and Nipponosaurus.
47
Young (1958) concluded that Saurolophinae was a valid taxonomic grouping, one that
included solid-crested forms, that, along with Hadrosaurinae, Lambeosaurinae and the primitive
Prohadrosaurinae, formed Hadrosauridae. Around the same time, Friedrich von Huene (1956)
followed a similar scheme of classification. He erected the name Hadrosauria as a superfamily
that included five families: the Prohadrosauridae, Hadrosauridae, Saurolophidae, Cheneosauridae
and Lambeosauridae.
Ostrom (1961) supported the re-establishment of Saurolophinae as a subfamily of
hadrosaurids possessing solid cranial crests that were excluded from the nasal passages, in
contrast to those of lambeosaurines that were hollow and contained the nassal passages (Fig.
1.4A, Chapter 1 of this dissertation). Ostrom regarded Saurolophinae as intermediate between
the non-crested Hadrosaurinae and the crested Lambeosaurinae. Furthermore, he suggested that
Claosaurus might have been the most primitive hadrosaurid because of the late Coniacian
stratigraphic position (thought to be the oldest occurrence of a hadrosaur at the time) and the
primitive morphology of the postcranial bones.
Hopson (1975) supported Sternberg’s division of Hadrosauridae into the two subfamilies
Hadrosaurinae and Lambeosaurinae (Fig. 1.5A, Chapter 1 of this dissertation). However, he
divided Hadrosaurinae into three informally named groups on the basis of their facial anatomy:
the “kritosaurs”, the “edmontosaurs” and the “saurolophines”, composed of Prosaurolophus,
Saurolophus, Tsintaosaurus and Lophorhothon. Hopson (1975) regarded the premaxillonasal
structure of “kritosaurs” as the ancestral condition from which the more excavated circumnarial
fossae of other forms derived. He also considered the genus Lophorhothon as intermediate
between “kritosaurs” and “saurolophines”.
The work of Brett-Surman (1976; 1979; 1989) represented one of the landmarks of
hadrosaurid systematics because for the first time taxa from Europe, Asia and South America
were included and his extensive revision of the taxonomy and postcranial anatomy of these
animals. That study also led to a greater appreciation of the postcrania as a source of diagnostic
and phylogenetically informative characters. In 1979, he presented the first evolutionary
hypothesis to encompass all genera known at the time from Europe, Asia and the Americas (Fig.
1.5B, Chapter 1 of this dissertation). Probactrosaurus was placed as the base of the tree as the
ancestor of all hadrosaurids. Among lambeosaurines, Jaxartosaurus was posited as the direct
ancestor of a Corythosaurus lineage and Bactrosaurus as the ancestor of a Parasaurolophus
48
lineage. Among hadrosaurines, Gilmoreosaurus was inferred to be the direct ancestor of this
subfamily.
The Advent of Phylogenetic Systematics
All the above classification schemes and phylogenetic hypotheses were characterized by
subjective weighting of characters. Researchers used personal judgment to consider particular
characters as those from which groups were established and evolutionary trends hypothesized.
This approach changed with the implementation of cladistic methods of phylogenetic inference.
Rsearchers for the first time had a tool to minimize subjectivity and conduct more explicit and
operative analyses of evolutionary relationships where all observed characters could play a role
in the formulation of phylogenetic hypotheses.
Norman (1984) and Sereno (1986) carried out the first cladistic analyses of iguanodontian
ornithopods, supporting the monophyly of Hadrosauridae and its divergence into Hadrosaurinae
and Lambeosaurinae. This dichotomic topology and the monophyly of these three clades figured
prominently in subsequent phylogenetic analyses (Weishampel and Horner, 1990; Weishampel
et al., 1993; Forster and Sereno, 1994; Forster, 1997a; Head 1998; 2001; Godefroit et al., 1998;
2008; Norman, 2002; Hai-lu You et al., 2003; Horner et al., 2004; Prieto-Marquez et al., 2006b;
Evans and Reisz, 2007; Gates et al., 2007).
The cladistic analysis of Weishampel and Horner (1990) was the first to focus on the
relationships within Hadrosauridae (Fig. 1.6A, Chapter 1 of this dissertation). Within
hadrosaurines, two clades were identifed. One of them was composed of Gryposaurus,
Aralosaurus, Maiasaura and Brachylophosaurus. The other clade included Saurolophus,
Prosaurolophus, Lophorhothon, Edmontosaurus and Shantungosaurus. Lambeosaurinae
included Parasaurolophus and a subclade composed of Corythosaurus, Hypacrosaurus and
Lambeosaurus. Bactrosaurus was tentatively placed within lambeosaurines. Telmatosaurus and
Gilmoreosaurus, were regarded as the most primitive hadrosaurids and were placed immediately
outside hadrosaurines and lambeosaurines.
Horner (1985; 1990) proposed a diphyletic origin for hadrosaurids. He redefined
Hadrosauridae to include only hadrosaurines (Fig. 1.7A, Chapter 1 of this dissertation).
Lambeosaurinae was elevated to family rank and renamed as Lambeosauridae. He used the name
Hadrosauria (Von Huene, 1956) for the clade including Hadrosauridae and Iguanodon and
erected Lambeosauria as the clade including Lambeosauridae and Ouranosaurus. Lambeosauria
49
and Hadrosauria were then united as sister taxa within Iguanodontoidea. Horner’s argument in
support of his diphyletic hypothesis was that Iguanodon shared more characters with his
definition of Hadrosauridae than with “lambeosaurids”, whereas Ouranosaurus shared more
characters with the latter group than with remaining hadrosaurids. His subsequent phylogeny of
Hadrosauridae (Horner, 1992) included an unbalanced tree composed of four clades, i.e., the
Maiasaurinae, the Gryposaurinae, Saurolophinae and the Edmontosaurinae, respectively (Fig.
1.7B, Chapter 1 of this dissertation).
One of the most cited phylogenetic analyses is that of Weishampel et al. (1993), the first
hadrosaurid phylogeny that included a matrix along with the list of characters. These authors
erected a new name, Euhadrosauria, for the clade composed of Hadrosaurinae and
Lambeosaurinae (Fig. 1.6B, Chapter 1 of this dissertation). Their Hadrosauridae was a more
inclusive clade composed of Telmatosaurus and the Euhadrosauria. Euhadrosaur relationships
were the same as those in the analysis of Weishampel and Horner (1990).
In a brief abstract, Forster and Sereno (1994) presented the results of a phylogeny where
Hadrosaurinae contained two clades. The first clade was composed of broad-snouted forms such
as Maiasaura and Edmontosaurus. Remaining hadrosaurines formed a clade characterized by
narial fossae joined along the dorsal margin of the internarial bar. In a later abstract, Forster
(1997a) restricted the name Hadrosauridae to include only Hadrosaurinae and Lambeosaurinae.
All taxa lying outside this clade were referred to the more inclusive Iguanodontia.
At the end of the twentieth and beginning of the twenty-first century cladistic analyses
became customary in systematic studies. Hadrosaurid phylogenies proliferated in the literature
with the revision, erection and description of new taxa. All analyses were in agreement that the
ancestry of hadrosaurids was to be found among basal iguanodontoideans.
Head (1998) proposed a phylogeny whereby primitive taxa such as Protohadros,
Gilmoreosaurus, Telmatosaurus and Bactrosaurus were outgroup genera to Hadrosauridae.
Bactrosaurus was sister taxon to Lambeosaurinae (Fig. 1.8A, Chapter 1 of this dissertation). In
the same year, Godefroit et al. (1998) (Fig. 1.9A, Chapter 1 of this dissertation) agreed with
Forster (1997a) in restricting Hadrosauridae to the more stable clade composed of Hadrosaurinae
and Lambeosaurinae.
Subsequently, Wagner (2001) and Hu (2001) proposed phylogenetic hypotheses that
were very different from those published up until that time. In the phylogeny of Hu et al. (2001)
50
non-lambeosaurine hadrosaurids formed an unnamed clade composed of Saurolophinae and
Hadrosaurinae (Fig. 1.10A, Chapter 1 of this dissertation). Hadrosaurines included two main
subclades. One of them included primitive genera such as Telmatosaurus, Tanius and
Claosaurus and Edmontosaurus. The other hadrosaurine clade was composed of forms such as
Maiasaura, Gryposaurus and Aralosaurus. Saurolophinae included Kritosaurus, Lophorhothon,
Saurolophus, Prosaurolophus and a clade composed of Tsintaosaurus and Jaxartosaurus.
Lambeosaurinae included Bactrosaurus being sister taxon to Nipponosaurus and
Parasaurolophus as sister taxon to Hypacrosaurus. On the other hand, Wagner’s (2001)
phylogeny formed an unbalanced tree, with the name Hadrosaurinae restricted to include only
Hadrosaurus and Kritosaurus (= Gryposaurus spp. and K. navajovius) (Fig. 1.10B, Chapter 1 of
this dissertation). Brachylophosaurus was found to be the sister taxon to the Saurolophidae. The
latter clade included Saurolophus, Edmontosaurus and lambeosaurines.
In Norman’s (2002) analysis Probactrosaurus appeared as the sister taxon to
Hadrosauridae and Protohadros, Eolambia and Altirhinus represented progressively less derived
outgroup taxa (Fig. 1.11A, Chapter 1 of this dissertation). Telmatosaurus and Bactrosaurus were
hadrosaurids leading crownward towards the Euhadrosauria. Conversely, Hai-lu You et al.
(2003) included several basal hadrosaur genera from Asia and proposed that hadrosaurid
outgroups were, moving crownward toward Hadrosauridae, Equijubus, Probactrosaurus,
Bactrosaurus, Protohadros and Telmatosaurus (Fig. 1.11B, Chapter 1 of this dissertation).
In 2004, Horner et al. updated their phylogeny of Hadrosauridae (Fig. 1.12B, Chapter 1
of this dissertation). Within Hadrosaurinae, Lophorhothon now appeared as the most basal form.
Two hadrosaurine clades were recognized. One of the clades was composed of
Brachylophosaurus, Maiasaura, Gryposaurus, Edmontosaurus and Prosaurolophus. The other
hadrosaurine clade consisted of a polytomy of Naashoibitosaurus (= Kritosaurus navajovius),
Saurolophus and “Kritosaurus” australis. Lambeosaurines included an unbalanced tree
composed of Tsintaosaurus, Parasaurolophus and a polytomy formed by Lambeosaurus,
Hypacrosaurus and Corythosaurus.
The analysis of Suzuki et al. (2004) focused on Lambeosaurinae (Fig. 1.12A, Chapter 1
of this dissertation). Bactrosaurus was found to be the closest outgroup to this clade. In their
analysis, Parasaurolophus figured as outgroup to all other lambeosaurines. These included a
51
polytomy composed of Hypacrosaurus stebingeri, H. altispinus and Nipponosaurus and a clade
including Corythosaurus and Lambeosaurus.
Prieto-Marquez et al. (2006b) restricted Hadrosauridae to include Hadrosaurinae and
Lambeosaurinae (Fig. 1.13C, Chapter 1 of this dissertation). Edmontosaurus appeared as
outgroup to remaining hadrosaurines. Two other clades were recognized: one composed of
Brachylophosaurus and Maiasaura and another integrated by Gryposaurus, Saurolophus and
Prosaurolophus. Lambeosaurines were composed of three clades of unresolved relationships.
The first clade included Parasaurolophus and Olorotitan. Another lineage consisted of
Hypacrosaurus altispinus. Finally, a third clade included a succession of progressively more
inclusive clades that culminated in the sister taxa Lambeosaurus and Corythosaurus. The genus
Hypacrosaurus was found to be paraphyletic.
In recent years, the work of Godefroit et al. (2000; 2001; 2003; 2004a, b; 2008) in the
Upper Maastrichtian of the Amur region of far eastern Russia and adjacent regions of China has
had considerable impact in our knowledge of hadrosaurid taxonomic diversity, phylogeny and
biogeography. Various new lambeosaurine (Charonosaurus, Olorotitan, Sahaliyania) and
hadrosaurine (Kerberosaurus, Wulagasaurus) genera were described by Godefroit et al. (2000;
2003; 2004 a, b; 2008) and Bolotsky and Godefroit (2004). The phylogeny of Godefroit et al.
(2008) recovered the clades and topologies inferred in all previous collaborative works by
Godefroit and colleagues (Fig. 1.14B, Chapter 1 of this dissertation). Hadrosaurinae was
composed of Wulagasaurus and a monophyletic group including the Maiasaura-
Brachylophosaurus clade and another clade formed by Gryposaurus, Kerberosaurus,
Prosaurolophus and Edmontosaurus. On the other side of the cladogram, lambeosaurines were
composed of an unbalanced topology that included Aralosaurus, Tsintaosaurus, Jaxartosaurus,
Amurosaurus and a clade composed of two lineages. The first lineage included Lambeosaurus
and a polytomy formed by Corythosaurus, Hypacrosaurus and Olorotitan. The other derived
was composed of Parasaurolophus, Charonosaurus and Sahaliyania.
Gates and Sampson (2007) inferred the position of Lophorhothon as the most basal
hadrosaurine (Fig. 1.14A, Chapter 1 of this dissertation). Two subclades were inferred among
hadrosaurines. One clade included Edmontosaurus, Saurolophus and Prosaurolophus. The other
clade consisted of Naashoibitosaurus (= Kritosaurus navajovius) and a polytomy formed by four
species of Gryposaurus and the Brachylophosaurus-Maiasaura clade.
52
Finally, two similar phylogenies of lambeosaurine hadrosaurids were recently presented
by Evans and Reisz (2007) and Gates et al. (2007). The former authors showed a succession of
Asian taxa (from Aralosaurus tuberiferus to Amurosaurus riabinini) as outgroups to a specious
dichotomic clade (Fig. 1.15A, Chapter 1 of this dissertation). The first clade of this dichotomy
was composed of Charonosaurus and three species of Parasaurolophus. The other clade was
composed of Lambeosaurus and a trichotomy composed of Olorotitan, Corythosaurus and a
monophyletic Hypacrosaurus. Their phylogeny revealed that a greatly enlarged cranial crest
evolved independently at least three times within Lambeosaurinae. The topology of the analysis
of Gates et al. (2007) differed from that in Evans and Reisz (2007) in having a polytomy
composed of Olorotitan ararhensis, Lambeosaurus spp., Corythosaurus casuarius,
Hypacrosaurus spp. and Velafrons coahuilensis, and in having Probactrosaurus gobiensis and
Telmatosaurus transsylvanicus united under their own clade (Fig. 1.15B, Chapter 1 of this
dissertation).
Materials and Methods
Selection of Taxa and Exemplars

The present study included 12 outgroup and 41 ingroup (hadrosaurid) taxa. Outgroup
taxa consisted of 2 non-hadrosaur iguanodontoideans and 10 non-hadrosaurid hadrosaurs (Table
2.1). Ingroup taxa comprised all valid species of hadrosaurids (Table 2.2). Species of Iguanodon
and Ouranosaurus nigeriensis, that are iguanodontoideans distantly related to hadrosaurids, were
chosen to root the tree by outgroup comparison (Wiley, 1981; Maddison et al., 1984). The
majority of taxa included in this analysis were scored from direct examination of actual fossil
material. The following species were not accessible to the author for logistic reasons and had to
be scored using anatomical descriptions from the literature: Probactrosaurus gobiensis (Norman,
2002), Eolambia caroljonesa (Kirlkland, 1998; Head, 2001), Wulagasaurus dongi (Godefroit et
al., 2008), Sahaliyania elunchunorum (Godefroit et al., 2008), Kerberosaurus manakini
(Bolotsky and Godefroit, 2004), Aralosaurus tuberiferus (Godefroit et al., 2004b) and some
skeletal elements of Shantungosaurus giganteus (Hu et al., 2001). Jaxartosaurus aralensis and
most of the adult material of Saurolophus angustirostris were exclusively scored from digital
photographs and video footage.
53
Taxonomic revisions. The taxonomy of a number of hadrosaur taxa has been historically
contentous. Here I provide a brief dicussion of my position regarding the taxonomic status of
these taxa. This is summarized in Tables 2.1 and 2.2.
I followed Godefroit et al. (1998) in the material referred to Bactrosaurus johnsoni and
Gilmoreosaurus mongoliensis. In particular, the bones from the AMNH quarries (field numbers)
149 and 145, AMNH 6551 and the predentary AMNH 6369 were used to score G. mongoliensis.
The material regarded here as belonging to B. johnsoni included all the original bones
documented by Gilmore (1933) from AMNH quarry 141, as well as the cranial remains of PIN
2549-1 and the more recently discovered cranial and postcranial bones of the SBDE 95/E5
collection (Godefroit et al., 1998).
Dalla Vechia (2006; 2007) pointed out that all the material that has been referred to T.
transsylvanicus (Nopcsa, 1900; 1928; Weishampel et al., 1993; Therrien, 2005) was collected
from four different formations in the Hateg Basin, the fossiliferous horizons of which span a
combined sedimentary sequence of at least 800 m. No data was recorded on the specific
stratigraphic provenance of the fossils. Thus, material referable to this species was here restricted
to the type skull BMNH 3386 and other material that was regarded by Nopcsa (1900) as
belonging to that same specimen: BMNH 3387, 3388 and 3841. In addition, two maxillae,
FGUUB 1010 and BMNH R4911, were tentatively referred here to T. transsylvanicus for being
morphologically indistinct from that in BMNH 3386.
Horner (1992) and Williamson (2000) summarized the evidence supporting the
synonymy of Anasazisaurus horneri (Hunt and Lucas, 1993; Lucas et al., 2006) and
Naashoibitosaurus ostromi (Hunt and Lucas, 1993) with Kritosaurus navajovius. In agreement
with Williamson (2000), the specimens BYU 12950 and NMMNH P-16106 were referred to K.
navajovius.
I also concurred with Wagner (2001) in that Gryposaurus incurvimanus is a junior
synonym of G. notabilis. Parks (1920) originally distinguished G. incurvimanus on the basis of
its deep and narrow skull, the presence of a “probable” median protuberance, the “peculiarities”
of the integument in ROM 764 with dermal callosities along the median line of the vertebral
column, two wedge-shaped manual phalanges, four phalanges in the 5th digit and “peculiarities”
of mandibular teeth. None of these characters allows distinction of G. incurvimanus (ROM 764
and TMP 80.22.1) from other hadrosaurid taxa. Additional characters provided by Parks (1920)
54
to distinguish G. incurvimanus from G. notabilis are not useful in this regard either. Thus, the
shape of the orbit is variable within G. notabilis and is not diagnostic to generic or specific level.
The more rostral position of the nasal protuberance in ROM 764 and TMP 80.22.1 is probably a
subadult condition, as is the lack of emargination on the rostral margin of the orbit (probably
related to the immature development of the nasal arch). That the G. incurvimanus specimens
represent subadult individuals of G. notabilis is supported by the smaller size of the skulls of G.
incurvimanus (58-60 cm in length, compared to the approximately 80 cm in length of the G.
notabilis skulls). The greater separation between the jugal and the quadrate varies among
hadrosaurid specimens depending on the circumstances of preservation and articulation. Finally,
the U or V-shape of the caudodorsal margin of the external naris is too variable among
specimens of G. notabilis to have diagnostic value: it is U-shaped in ROM 873 but more V-
shaped in CMN 2278 and MSNM V345.
More recently, Gates and Sampson (2007) have regarded Gryposaurus incurvimanus as a
distinct species and provided a revised diagnosis. New characters provided by these authors as
diagnostic for this species were: dorsal premaxillary process that is more concave caudally than
in other species of the genus; jugal with a small spur on the caudal margin of the caudoventral
flange; and slightly excavated ventral surface of the nasal hump. The degree of excavation of the
caudal region of the dorsal process of the premaxilla could only be observed in TMP 80.22.1 for
G. incurvimaus, since it was not preserved (and, unfortunately, entirely reconstructed) in ROM
764. Comparison of TMP 80.22.1 with G. notabilis skulls (e.g., ROM 873 and CMN 2278)
showed no differences in the relief of the caudal region of the premaxillary dorsal process among
these specimens. The slight excavation of the ventral surface of the nasal hump or arch was
present in all G. notabilis specimens as well as in TMP 80.22.1 and corresponded to the lightly
incised caudodorsal region of the circumnarial fossa. Examination of TMP 80.22.1 and ROM
764 revealed no spur on the caudal margin of the caudoventral flange that might allow
distinction between the jugals of these skulls and those of the G. notabilis specimens. Immature
characters of G. notabilis that Gates and Sampson (2007) considered as a unique combination to
diagnose G. incurvimanus included: gracility, smaller premaxillary lip, premaxillae angled more
steeply and rostrocaudally shortened and deeper skull. Finally, the infratemporal fenestra being
larger than the orbit was a character present also in G. notabilis and the straight ventral margin of
the maxilla was too variable intraspecifically to be of diagnostic value.
55
None of the characters used by Chapman and Brett-Surman (1990) to diagnose A. copei
were unique or allowed distinction of this taxon from other hadrosauids. Likewise, examination
of the skull of AMNH 5730 revealed several areas with signs of postdepositional dorsoventral
compression. For example, each dentary had a longitudinal ridge that protruding laterally,
probably produced by bending of the lateral surface of the bone (the same deformation ridge was
present in the right [but not the left] dentary of the paratype specimen of E. regalis, CMN 2289)
(Fig. 2.1); the dorsal region of the left quadrate had a transverse fissure (Fig. 2.2A); and the left
postorbital “inflation” of its central body appeared crushed inwards (medioventrally) (Fig. 2.2B),
with a nearly vertical indented line on the lateral bone surface. Aside from the very shallow
skull, AMNH 5730 and 5886 were indistinguishable from any specimen of E. annectens.
Therefore, A. copei was regarded as a junior synonym of E. annectens in the present study.
I also concluded that Edmontosaurus saskatchewanensis (CMN 8509; Sternberg, 1926)
was founded on subadult material of Edmontosaurus annectens. Differences in skull morphology
between CMN 8509 and other specimens of E. annectens and E. regalis, such as the poorly
developed lateral inflation of the central body of the postorbital and the relatively gracile jugal,
could be attributed to the immature development of the specimen. Likewise, even at its relatively
smaller size, CMN 8509 already had a preorbital region that was nearly 1.6 times the length of
the quadrate, indicating that later in ontogeny the skull of this animal would probably reach the
more elongated proportions of E. annectens. Therefore, E. saskatchewanensis was regarded as a
junior synonym of E. annectens.
Horner (1992) described Prosaurolophus blackfeetensis from the Upper Two Medicine
Formation of Montana. This species was diagnosed by the exclusion of the prefrontal and
premaxilla from the lateral concavity of the nasal, as well as by the greater caudal extension of
the crest relative to that of P. maximus. However, the dorsal surface of the prefrontal in P.
blackfeetensis (e.g., MOR 454-6-24-6-2) was actually concave medial to the orbital rim (Fig.
2.3), as in P. maximus. As for the length of the crest, this was variable within the genus (e.g.,
longer in ROM 787 than in the larger TMP 84.1.1) and actually not longer in MOR 454 relative
to that of other specimens within the genus. Thus, I agreed with Wagner (2001) in regarding P.
blackfeetensis as a junior synonym of P. maximus.
Secernosaurus koerneri (Brett-Surman, 1979) was regarded as a senior synonym of
Kritosaurus australis (Bonaparte et al., 1984). The type material of Secernosaurus koerneri
56
(FMNH P13423) was indistinct from the type and other referred material of Kritosaurus
australis (Bonaparte et al., 1984). The morphology of the pubis of FMNH P13423 was nearly
identical to that of K. australis (MACN-RN 2), particularly the geometry of prebupic process,
the length and width of the proximal constriction and the length of the ischial peduncle.
Likewise, the right ilium of FMNH P13423 (the left one is distorted) shared with those of the
MACN specimens a shallow proximal margin of the preacetabular process, craniocaudally
extensive supraacetabular process that did not project ventrally more than half the depth of the
iliac central plate and a morphologically unique brevis shelf.
The hadrosaurid material from the Salitral Moreno bonebed (Powell, 1987), from Rio
Negro, Argentina, was also included in my analysis as a single operational taxonomic unit
(OTU) to further document the diversity of these animals in South America. This material is
currently being described by a team of paleontologists from the Centro Paleontologico Lago Los
Barreales lead by Jorge Calvo.
Two other unnamed hadrosaurids were also included as OTU’s to further represent the
diversity of hadrosaurids in southern North America. The first OTU was a set of skeletal
elements from the UTEP collection at the University of Texas in Austin, documented by Davis
(1983) and regarded by Wagner (2001) as a lambeosaurine hadrosaurid. The ilium (UTEP
P37.7.222) was particularly interesting because it was the only hadrosaurid specimen that I
observed that shared a unique morphological type of brevis shelf with the ilia of Secernosaurus
koerneri and those collected from Salitral Moreno. The second OTU was a recently described
large skeleton collected near the Mexican town of Sabinas, Coahuila, referred to Kritosaurus sp.
by Kirkland et al. (2006).
Character Definition and Character State Delimitation
A pivotal stage in the estimation of hadrosaurid interrelationships was the definition of
morphological characters. This required the reassessment of characters used in previous analyses
by other authors (some of which needed to be redefined in face of the new data) as well as the
discovery of new ones based on the analysis of patterns of variation in the crania and postcrania
of hadrosaurids. I followed three approaches for discovering and evaluating characters from
skeletal morphological variation: qualitative comparative anatomy of specimens, traditional
morphometrics, and geometric morphometrics.
57
Qualitative osteological comparisons and traditional morphometrics. Primary homology
was assessed on thousands of cranial and postcranial bones through direct visual comparisons.
Implicit in these comparisons was the application of the homology criteria of conjunction and
similarity (topographic, structural and compositional correspondence) (Patterson, 1982; Hawkins
et al., 1997; Hawkins, 2000). In addition, linear and angular measurements were taken from the
actual specimens and from high-resolution digital images of the bones taken by myself. In the
later case, measurements were obtained using the Image J program (Abramoff et al., 2004).
Linear measurements were used to calculate ratios that served to describe quantitatively a
particular anatomical character.
Geometric morphometrics: analysis of planar shapes using geodesic paths. The analysis of
planar shapes using geodesic paths is a relatively new morphometric method developed by
Klassen et al. (2004). It has also been referred as Geodesic Distance Analysis (GDA) by Prieto-
Marquez et al. (2007) in its first implementeation to research within the biological sciences.
Prieto-Marquez et al. (2007) used GDA to quantify differences in the geometry of the pelvic
canal of iguanids and crocodilian reptiles.
In GDA dissimilarities between shapes are quantified using geodesic distances within a
Riemannian (i.e., non-Euclidean) shape space. The term “geodesic” refers to the shortest path
between two points in a curved space. GDA considers the shapes of planar continuous curves in
R2. Shapes are invariant to rigid motions (translation and rotation) and uniform scaling. Closed
curves are also invariant to the placement of origin (or starting point) of the parameterization.
The objects of study are represented by the continuous curves of their boundaries. The arc length
of these contours is used as a parameter to model the shapes. The angle between the tangent
vector to the curve and the positive x-axis is defined as a function of the arc length (Srivastava et
al., 2005). In this way, shapes are compared and represented using their angle functions. Then,
geodesic paths (or geodesics) are constructed between them. A geodesic between two shapes is
the path that minimizes the difference between their angle functions. I implemented an updated
version of the method, called “elastic” GDA (Mio et al., 2007). While the original approach
described in Klassen et al. (2004) modeled shapes by “bending-only”, in the approach of Mio et
al. (2007) curves along the geodesic paths are transformed by bending, stretching or compressing
segments non-uniformly along their lengths. Thus, elastic GDA is a more appropiate method to
study local changes among shapes, quantifying only those differences that are found in smaller,
58
more concrete regions of the curves. Likewise, elastic GDA allows modeling of open curves as
well as as closed ones.
The benefit of GDA is that it allows for the quantification of continuous curves that are
very difficult or impossible to discretize using landmarks. Thus, the method was applied in
searching for and defining patterns of variation on skeletal elements where the identification of
homologous points is ambiguous. An example is the lateral profile of the prepubic process of the
pubis, where it cannot be assumed that a given point along its margin is homologous to that of
another exemplar. Landmark-based methods (Bookstein et al., 1985; Dryden and Mardia, 1998;
Zelditch et al., 2000) are limited in scope because they depend heavily on the chosen
homologous points. Several other methods of shape analysis represent geometries by parametric
splines using Fourier Descriptors (Persoon and Fu, 1986). However, a drawback of using splines
is that Fourier Descriptors ignore the non-linear geometry of the shape space of closed curves.
While such representations can be used to classify shapes, it is not possible to perform statistical
analyses that are intrinsic to the shape space. This is because spline control points only represent
a finite number of coefficients and the continuity of the boundaries that connect those
coefficients are ignored; the space of all spline functions becomes a finite dimensional shape
space, unlike the infinite space of shapes of continuous curves. Another statistical approach,
called Active Shape Models (Cootes et al., 1995), uses principal component analysis of
landmarks to model shape variability. Despite its simplicity and efficacy, this approach is rather
limited because it ignores the non-linear geometry of shape space where the geometries of study
are contained.
In practice, the continuous curves for input into the GDA analysis were obtained by
tracing the contour of fossil bones with a Wacom Graphire graphic tablet (Wacom Co., Ltd,
Saltama, Japan) in Illustrator CS2 (Adobe Systems Inc., San Jose, USA). Each drawing of a
curve was saved as a vector file and imported into Matlab version 6.5 (The MathWorks, Natick,
USA), the software used to run the GDA algorithms. The GDA output consisted of a matrix of
pair wise geodesic distances among all the curves of a collection of bone shapes. Because the
application of the method of cluster analysis implemented in this study required that the data had
a finite number of dimensions, the next step consisted on reducing the number of dimensions of
the GDA dissimilarity matrix. This was accomplished via non-metric multidimensional scaling
(or NMMDS; Young, 1985), implemented in the R statistical package package (Ihaka and
59
Gentleman, 1996). NMMDS transforms an infinite, non-euclidean space into a coordinate
Euclidean space with a limited number of dimensions. The method uses Euclidean distance to
model dissimilarity by means of a monotonic transformation (Young, 1985).
Discretization of continuous data derived from morphometric analyses. While
morphometric approaches were used to discover and evaluate patterns of variation containing
phylogenetic information, other methods were required to transform those patterns into discrete
character states. A model-based clustering method was applied to these quantitative data in order
to minimize the subjectivity in distinguishing clusters and deciding the number character states
to be considered. I implemented a technique developed by Fraley and Raftery (2002) integrated
in the R statistical package as a program called Mclust (Fraley and Raftery, 2003; 2006). It
provides a criterion that aids in deciding how many clusters can be distinguished from a
distribution of data points. This method combines hierarchical clustering, the Expectation-
Maximization algorithm (McLachlan and Krishnan, 1997) and a Bayesian-Information Criterion
(or BIC; Fraley et al., 2005).
The result of this type of analysis is summarized in a bivariate plot. For example, Fig
2.4A shows a graphical output from Mclust resulting from cluster analysis of the angle between
the acetabular and articular margins of the iliac peduncle of the ischium (character 276;
Appendix I, Fig. I.35). The x-axis represents the number of clusters into which the data may be
classified and the y-axis shows the values of the Bayesian Information Criterion (BIC). The BIC
is the value of the maximized log-likelihood with a cost on the number of parameters of each
model (Fraley and Raftery, 2003). Each one of the two lines in the plot represents a clustering
model. Because the data used in the analysis contains only one dimension, there are only two
possible models (hence, only two lines are shown in the plot). These models contain ellipsoidal
clusters that vary according to the value of the BIC. Likewise, the BIC varies for each one of the
number of clusters considered for each model. These two models differ depending on whether
the variance is equal or variable (abbreviated as “E” and “V”, respectively, as shown in the lower
right corner of the graph). The method favors selecting the number of clusters that correspond to
the greater (i.e., less negative) BIC values. In the case of Fig. 2.4A, the method favored grouping
the data into two clusters of equal variance because that model showed the greatest (less
negative) BIC value. Three clusters of equal variance would be the next optimal grouping of the
data and so forth.
60
In multidimensional data sets such as those derived from the analysis of planar shapes
using geodesic paths, the method considers also the geometric characteristics of various
clustering models, like orientation, volume, or shape of the clusters. An example is shown in Fig.
2.4B, consisting on a model clustering analysis of the shape of the postacetabular process of the
ilium (character 259; Appendix I, Figs. I.21 and I.22). A total of eight colored lines are present in
that bivariate plot, each one representing a clustering model characterized by a specific shape,
volume and distribution of the clusters. Table 2.3 explains the meaning of the identifiers
appearing in the lower right corner of the plot. In the case of Fig. 2.4B, for character 259, the
method favored clustering the data into two groups of equal shape, equal volume and with data
points distributed diagonally (abbreviated EEI).
Boxplot distributions of the raw data were also considered in deciding the number of
clusters in which each quantitative character was divided. These graphs provided a visual
representation of the variation of a ratio or an angle among the sampled taxa, as well as an
insight into the variance within each species. The information provided by the BIC and boxplots
was combined to help decide how many character states would be considered for a given
character. In a few cases, the data would be clearly separated in the boxplot in a determinate
number of clusters that was not supported by the BIC plot. In such occasions, I favored the
pattern displayed in the boxplot in my decision of establishing the number of states for the
characters.
Finally, K-Means clustering analysis was used to assign each specimen of the data to one
of more of the character states resulting from the analyses described above.
Character State Coding. A total of 299 characters (205 cranial and 94 postcranial) were
included in this study. These characters were illustrated and documented in the appendices
(Appendix B through J).
The absence/presence coding of Pleijel (1995; equivalent to the nominal variable coding
of Pimentel and Riggins, 1987 and to the reductive coding of Wilkinson, 1995) was applied to
qualitative attributes that were either absent or present (with no further variation) among the taxa
of study. One example of such characters is the surangular foramen, present ancestrally in
Iguanodon and some hadrosaurs but lacking in hadrosaurids (character 58; Appendix C, Fig.
C.40).
61
The conventional coding of Hawkins et al. (1997) was applied when character states
consisted on alternate forms of a particular character. This form of coding was appropriate in
many qualitative characters such as the position of the caudal end of the tooth row relative to the
coronoid process. This character was composed of three states: tooth row located caudal to, flush
with or rostral to the coronoid process (character 54; Appendix C, Fig. C.36).
Another coding form used in this study was the composite coding method of Wilkinson
(1995). This method consists on including, under a single character, more than one attribute that
otherwise would be coded separately. This form of coding was applied to include single
character attributes that were not independent. An example is character 59, the orientation of the
convex side of the lateral flange and the lateroventral surface of the main body of the surangular
(Appendix C, Fig. C.41). An alternative to composite coding for this character would be to
construct a conventional character for the orientation of the lateral flange and another for the
orientation of the lateroventral surface if these were found to vary independently.
Numerous quantitative characters were defined using ratios from linear measurements.
Those characters were best discretized following the ratio coding of Hawkins (2000) because it is
the relationship between each of the two particular variables of the ratio that was
phylogenetically informative. For example, character 203, the total length of the skull relative to
its height. This character contained two states: length/height ratio of 2 or less or greater than 2
(Appendix F, Fig. F.31).
Finally, a subset of the characters was coded as logically related (Hawkins, 2000). This
form of coding was appropriate when two or more states provided different descriptors of a
single variable. An example of this form of coding is found in character 211, the number of
sacral vertebrae. Two states were recognized: sacrum composed of up to seven vertebrae and
composed of eight or more vertebrae (Appendix G, Table G.3).
Polymorphic characters were coded by scoring all states found in a given taxon in the
same cell of the character matrix (i.e., using the notation “0 &1 & 2” when states 0 through 2 are
present in the same taxon). Because fossil remains are often incompletely preserved, it is not
uncommon to encounter uncertainty in scoring characters. The shape of the quadratojugal notch
of the quadrate (character 127) will illustrate this case. This character includes three states
(Appendix D, Fig. D.79). In Kerberosaurus manakini, the notch is partially preserved so that,
while it is possible to rule out the presence of states “0”, it is uncertain which of the other two
62
states was present in this species. In those instances I scored in the same cell of the data matrix
the state codes that were possible for the taxon (in the example of K. manakini, using the
notation “1/2”).
Inapplicable characters were coded as “-”. This coding was chosen because various
studies have concluded that this is the less problematic way of dealing with inapplicable
characters (Strong and Lipscomb, 1999; Seitz et al., 2000).
All characters were equally weighted. The only ordered characters were the number of
dentary and maxillary alveoli (characters 1 and 16, respectively; Appendix B, Figs. B.1-B.4 and
B.20-B.22) and the maximum number of teeth exposed on the dentary and maxillary occlusal
planes (characters 4 and 19, respectively; Appendix B, Figs. B.9 and B.26). Justification for
ordering these four characters came from observing sequences of changes among characters that
were consistent throughout ontogeny and throughout the stratigraphic succession of the fossil
record of iguanodontoideans. For example, in the case of character 1, there is a positve
correlation between the number of tooth families and specimen size (Appendix B, Figs. B.2-B.4).
At the same time, oldest and more basal taxa tend to show less of teeth in their dentaries than
stratigraphically younger and relatively more derived taxa. This same pattern was observed for
characters 4 (number of functional teeth in the dentary occlusal plane; Appendix B, Fiog. B.9),
16 (number of maxillary tooth families; Appendix B, Figs. B.20-B.22) and 19 (number of
functional teeth in the maxillary occlusal plane; Appendix B, Fiog. B.26). Although many other
cranial and postcranial elements are known in subadult and adult specimens, in most cases these
were recorded only in a few taxa or in a few exemplars within a species. For example, although
numerous complete skulls document a growth series in the lambeosaurine genera Corythosaurus
and Lambeosaurus, similar cranial ontogentic series for other taxa are wanting. Likewise, in spite
of as the completely known cranial anatomy of Corythosaurus and Lambeosaurus throughout
development, ontogenetic data for for the postcrania are lacking
Methods of Phylogenetic Inference
A total of eight analyses were conducted (four analytical approaches on two data sets).
The analyses included weighted and unweighted parsimony and Bayesian with equal and
variable rates of character change. One of the character matrices had Barsboldia sicinskii and
Nipponosaurus sachalinensis pruned from the data, while the other included these two species.
B. sicinskii is an enigmatic hadrosaurid from the Maastrichtian Nemegt Formation of Mongolia
63
(Maryanska and Osmolska, 1981) known only from a single largely incomplete (a metatarsal, the
left ilium and sacral and several caudal vertebrae) exemplar. Thus, this is a “wildcard” taxon
(Wilkinson, 2003) that was included in this study for the sake of completeness with the intention
of shedding light into its phylogenetic relationship to other hadrosaurids. Nipponosaurus
sachalinensis is known from a small juvenile specimen collected from the late Santonian-early
Campanian Upper Yezo Group in Sakhalin, Russia (Nagao, 1936). Most of the diagnostic and
phylogenetically informative bones of the skull are missing. Among the few preserved bones
(mostly mandibular and dental elements), most are ontogenetically variable in hadrosaurids.
Maximum parsimony analyses. The search for the optimal tree(s) using maximum
parsimony was carried out in PAUP version 4.0b10 (Swofford, 2002). Two methodologies were
implemented. In the first one, parsimony was applied with all characters having the same weight.
For the second approach, I implemented the weighted parsimony method of Goloboff (1993). In
this method, the “best fits” (i.e., high consistency indices) of characters are used to calculate
weights simultaneously as the tree is built. In this way, characters exhibiting a greater degree of
homoplasy are downweighted relative to those with greater consistency indices. The rationale for
this procedure is that those characters that exhibit less homoplasy are more reliable indictors of
phylogenetic relationships (Goloboff, 1993). The Goloboff k fit parameter (constant of concvity
of the homoplasy fitting function) was set to 3. As the value of k increases, the function weights
less drastically (Goloboff, 1993). I chose a value of 3 as a conservative compromise between the
default of 2 in PAUP version 4.0b10 (Swofford, 2002) and higher values.
I performed a heuristic search using a random additional sequence of 10,000 replicates
for the unweighted parsimony analysis of the pruned data set. For the unweighted parsimony
analysis of the complete data set and the weighted parsimony analyses I used random additional
sequences of 5,000 replicates. In all the searches, I utilized the option of branch-swapping by
tree-bisection-reconnection in all parsimony analyses (TBR) to minimize algorithm “greediness”
and maximize the thoroughness of the search (Swofford et al., 1996a).
Bremer support (Bremer, 1988) was assessed by computing decay indices (Donoghue et
al., 1992) with MacClade version 4.0 (Maddison and Maddison, 2003) and PAUP. Bootstrap
proportions (Felsenstein, 1985) were computed in PAUP. Two bootstrap settings were used. For
the unweighted parsimony analyses, the bootstrap was set to 10,000 replicates using heuristic
searches, where each search was performed using random additional sequences with branch-
64
swapping by TBR and 1,000 replicates. For the weighted parsimony analyses, the bootstrap was
set to 5,000 replicates using heuristic searches, where each search was conducted using random
additional sequences with branch-swapping by SPR (Subtree Pruning and Regrafting) and 25
replicates. Reconstruction of ancestral states was conducted using parsimony in MacClade
version 4.0.
Bayesian analyses. There have been relatively few studies exploring the application of
Bayesian analyses to morphological data (Lewis, 2001; Nylander et al., 2004; Lee, 2005; Wiens
et al., 2005). Furthermore , in only a few cases Bayesian methods have been used to reconstruct
the phylogeny of fossil groups (Snively et al., 2004; Muller and Reisz, 2006). Notably these
included one on hadrosaurids (Evans, 2007b). Here, I used the maximum likelihood model of
Lewis (2001) (Markov k or Mk) for discrete morphological data, implemented in MrBayes
version 3.1.2 (Huelsenbeck and Ronquist, 2001). Two different models were applied. In a first
model, I assumed equal rates of character change. In the second model I incorporated the gamma
parameter to allow for variable rates of character change, as in Nylander et al. (2004). Priors
other than the default for the models were not specified for all the analyses.
For the pruned data set (51 taxa), the analysis without gamma used six chains and ran for
10,000,000 generations, with samples being taken every 100 generations. Stationarity was
achieved with a standard deviation of split frequencies of 0.005. Trees generated during the first
100,000 generations were discarded as “burn-in”. This was assessed by visual examination of
plots generated with the “sump” command of MrBayes. The analysis with gamma used four
chains and 4,000,000 generations sampled every 100 generations, reaching stationarity with a
0.008 standard deviation of split frequencies. Trees generated during the first 100,000
generations were also discarded as “burn-in”.
For the complete data set (53 taxa), the analysis without gamma used six chains and ran
for 10,000,000 generations sampled every 100 generations. Stationarity was achieved with a
standard deviation of split frequencies of 0.005. Trees generated during the first 100,000
generations were discarded as “burn-in”. This was assessed by visual examination of plots
generated with the “sump” command of MrBayes. The analysis with gamma used four chains
and 4,000,000 generations sampled every 100 generations, reaching stationarity with a 0.009
standard deviation of split frequencies. Trees generated during the first 100,000 generations were
also discarded as “burn-in”.
65
Results
The unweighted parsimony analysis of the pruned data set resulted in 41 equally most
parsimonious trees of 971 steps each. The strict consensus tree is shown in Fig. 2.5. The tree-
island profile of the heuristic search was characterized by a large island containing 15 most
parsimonious trees and 26 smaller islands containing one tree each. The large island was hit 3476
times (Table 2.4). The weighted parsimony analysis of the pruned data set resulted in one single
most parsimonious tree with a Goloboff fit of 222.19 (Fig. 2.7). The single island containing this
optimal tree was hit 619 times during the search.
In both of these parsimony analyses, the consistency index had a value of 0.44 and the
retention index a value of 0.76. Such a low consistency index reflected the large amount of
homoplasy present in these data. Likewise, in both parsimony analyses most clades were weakly
supported, as shown by the very low bootstrap proportions and decay indices (Figs. 2.5 and 2.7).
The groupings with greater bootstrap proportions were found among the most inclusive clades
outside hadrosaurids and also among those clades formed by multispecific genera. A similar
pattern of support was shown by the posterior probabilities in the Bayesian consensus trees,
although in general all the clades showed greater support than in the parsimony trees. The most
inclusive clades were very well supported, as well as the two major subclades with hadrosaurids
and the clades composed of multispecific genera.
Adding Nipponosaurus sachalinensis and Barsboldia sicinskii to the weighted and
unweighted parsimony analyses resulted in considerable loss of resolution within saurolophines
(Fig. 2.6). The unweighted parsimony analysis resulted in 1422 most parsimonious trees with a
tree length of 979 each. 1372 of these MPTs were found in a large island, hit 1639 times during
the analysis; remaining MPTs were formed single tree-islands that were hit only once during the
search (Table 2.5). The weighted parsimony analysis resulted in 11 most parsimonious trees with
an optimal score of 221.46. These 11 most parsimonious trees were found in a single island that
was hit 945 times during the search. The consistency and retention indices in these two
parsimony analyses were as in the analyses of the pruned data set, namely 0.44 and 0.76,
respectively. Support measures such as bootstrap proportions and decay indices in the parsimony
trees, as well as the posterior probabilities in the Bayesian trees, remained low and showed the
66
same distribution of poorly and better supported clades as in the analyses excluding B. sicinskii
and N. sachalinensis (Figs. 2.6 and 2.9).
Summary of Hadrosaurid Relationships
The topologies resulting from all the analyses an unbalanced topology of outgroup taxa to
Hadrosauridae (Figs. 2.5-2.13). The clades stemming from the most distant outgroup taxa to
hadrosaurids were moderately well supported in the parsimony analyses, with bootstrap
proportions approaching 0.8, and very well supported in the Bayesian phylogenies, with
posterior probabilities of 1 or nearly 1.
Informal names were created in order to facilitate reference to particular clades. Figure
2.8 shows the single most parsimonious cladogram derived from the weighted parsimony
analysis with these informal names written between quotation marks. In the following sections,
characters will be referred to by a number followed by a “-” and the character state number.
These numbers can be used to find the definition and documentation of the character in
Appendices B through J. Clade definitions (Appendix K) followed the requirements for
establishing clade names stated in Article 9 of the PhyloCode (Cantino and de Queiroz, 2007).
In contrast to previous studies implementing parsimony and Bayesian methods to fossil
data (Muller and Reisz, 2006; Evans, 2007b), important differences were found among the
inferred phylogenies. This was particularly evident when comparing the tree from the Bayesian
analysis with gamma and those from the other three analyses. In the phylogenies resulting from
the weighted parsimony and Bayesian analysis without gamma (Figs. 2.5-2.11), hadrosaurids
were composed of two major clades, Saurolophinae and Lambeosaurinae. This general pattern is
congruent with the most accepted cladistic hypothesis of the relationships of these animals
derived from numerous studies over the last two decades (Sereno, 1986; Weishampel and
Horner, 1990; Forster, 1997b; Horner et al., 2004). However, the unweighted parsimony analysis
produced a phylogeny where “maiasaurs” appeared as outgroup to the Euhadrosauria, the clade
formed by saurolophines and lambeosaurines (Figs. 2.5 and 2.6). Support for the exclusion of
“maiasaurs” from euhadrosaurians was relatively low, with a bootstrap proportion of 0.45 and a
decay index of 3. The topology inferred in the Bayesian analysis with gamma (Figs. 2.12 and
2.13) showed a result even more incongruent with the “classic” dichotomic evolution of
hadrosaurids (Lull and Wright, 1942; Weishampel and Horner, 1990; Horner et al., 2004;
Godefroit et al., 2008). In that analysis, hadrosaurids formed a polytomy composed of
67
“maiasaurs”, Gryposaurus, Kritosaurus, the Secernosaurus clade and an unbalanced tree leading
to lambeosaurines. A remarkable and unexpected result was that Edmontosaurus and
Shantungosaurus figured as successive outgroup taxa to Lambeosaurinae. This topology echoes
the phylogenetic inferences of Hopson (1975) and Wagner (2001) positing that the evolution of
the enclosed nasal passages in lambeosaurines derived from a common ancestor of the latter and
Edmontosaur-like saurolophines.
In all of my phylogenies the most basal members of Lambeosaurinae were species from
Eurasia, such as Jaxartosaurus aralensis, Tsintaosaurus spinorhinus and Pararhabdodon
isonensis (Figs. 2.5-2.13). Aralosaurus tuberiferus was inferred to be the most basal
lambeosaurine in all but the Bayesian analysis without gamma, where this taxon appeared as
outgroup to the Euhadrosauria. The following more exclusive clades were present in all
phylogenies: the Sahaliyania-Amurosaurus clade, the Charonosaurus-Parasaurolophus clade
(“parasaurolophs”), the Pararhabdodon-Tsintaosaurus clade (“tsintaosaurs”) and the
Corythosaurus-Lambeosaurus clade (“corythosaurs”, with or without Hypacrosaurus
stebingeri). Only the pasimony analyses were able to resolve the relationships among
Hypacrosaurus altispinus, Lambeosaurus laticaudus and Velafrons coahuilensis. Likewise, both
parsimony analyses agreed in having “corythosaurs” united with “hypacrosaurs” to form the
“amurosaur” clade. Likewise “amurosaurs” were inferred to form the sister clade to
“parasaurolophs”. However, these clades were very weakly supported, with the exception of the
most exclusive groupings of species into genera. The low support for these clades and the
polytomic relationships shown in the Bayesian analyses (Figs. 2.10-2.13) may be due to the short
divergence times of these species.
The monophyly of Saurolophinae was supported in all the phylogenies with the exception
of that derived from the Bayesian analysis with gamma. This clade is equivalent to
Hadrosaurinae of previous authors (Lull and Wright, 1942; Sereno, 1986; Weishampel and
Horner, 1990; Weishampel et al., 1993; Godefroit et al., 1998; 2008; Head, 1998; Norman,
2002; Horner et al., 2004; Gates et al., 2007; Evans and Reisz, 2007). The name Saurolophinae
(Brown, 1914) will be used here instead of Hadrosaurinae. This is because Hadrosaurus, the
genotype of Hadrosaurinae, did not form part this clade.
In the analyses that did recover a monophyletic and well-resolved Saurolophinae,
Shantungosaurus was inferred to be the outgroup taxon to all other saurolophines (Figs. 2.5, 2.7,
68
2.10 and 2.11). Likewise, two clades were always present: the “kritosaur” clade (composed of
Kritosaurus, Gryposaurus, Secernosaurus and the Salitral Moreno and Big Bend UTEP OTU’s)
and the Prosaurolophus-Saurolophus clade. Differences among the phylogenies showing these
clades consisted on the position of Edmontosaurus and “maisaurs”. Thus, in the unweighted
parsimony tree, “maiasaurs” were excluded from Euhadrosauria (Fig. 2.5) and Edmontosaurus
was the outgroup genus to a clade composed of “kritosaurs” and Prosaurolophus-Saurolophus.
In the weighted parsimony tree, “maiasaurs” were included in Saurolophinae as the sister clade
to “kritosaurs”, while Edmontosaurus became the sister genus to the Prosaurolophus-
Saurolophus clade (Fig. 2.7).
The Bayesian phylogenies with no gamma (Figs. 2.10 and 2.11) shared topological
features in both parsimony analyses. Like the weighted parsimony tree, the Bayesian analysis
with no Gamma inferred “maiasaurs” as sister clade to “kritosaurs”. In the unweighted
parsimony tree the Bayesian phylogeny placed Edmontosaurus as an outgroup to all other
saurolophines except Shantungosaurus.
Finally, the consensus tree derived from the Bayesian analysis with gamma contained a
paraphyletic Saurolophinae (Fig. 2.12). The Secernosaurus-Salitral Moreno clade, Gryposaurus,
the Big Bend OTU, Kritosaurus and a large clade with remaining euhadrosaurians, including
lambeosaurines, formed a well-supported polytomy. The Prosaurolophus-Saurolophus clade, the
Sabinas OTU, Kerberosaurus and Edmontosaurus were part of an unbalanced tree leading to
lambeosaurines. However, the placement of Edmontosaurus and Shantungosaurus in that clade
as ancestors of lambeosaurines was weakly supported by relattively low posterior probabilities.
The Interrelationships of Non-Hadrosaurid Hadrosaurs
The most inclusive clade considered in this study was Hadrosauria (von Huene, 1956). In
this study, the concept of Hadrosauria was as in the original work of von Huene (1956) in that it
included all those iguanodontians morphologically and phylogenetically closer to hadrosaurids
than to Iguanodon. Hadrosauria was defined here as the most inclusive clade containing
Equijubus normani, but not Iguanodon spp. and Ouranosaurus nigeriensis. This clade was
supported by the following unambiguous synapomorphies: rostral end of the rostrodorsal process
of the nasal reaching the rostral margin of the narial foramen (character 87-1; reversed in
Kritosaurus navajovius, Gryposaurus, Secernosaurus koerneri and the Salitral Moreno and Big
Bend UTEP OTU’s); centered dorsolateral promontory of the maxilla, base of dorsal process
69
positioned slightly caudal to the mid-length of the element (character 98-1); articulation margin
of the occipital condyle facing caudally and divided from the caudal border of the basioccipital
by a shallow cleft (character 160-1); palpebral bone absent or fused to the orbital margin
(character 196-0).
Outgroup hadrosaur taxa to Hadrosauridae formed an unbalanced tree of Asian and North
American species (Figs. 2.5-2.13). The unweighted parsimony and Bayesian analysis without
gamma joined Eolambia caroljonesa and Protohadros byrdi in a clade (Figs. 2.5 and 2.10),
supported by the following unambiguous synapomorphies: origination of the deflection of the
ventral margin of the dentary located rostrally, between 66% and 78% of the length of the dental
battery (character 41-2; convergent in Ouranosaurus nigeriensis, Gryposaurus latidens,
Shantungosaurus giganteus, Parasaurolophus walkeri, P. tubicen, Tsintaosaurus spinorhinus,
Pararhandodon isonensis, Corythosaurus spp. and “Hypacrosaurus” stebingeri); moderately
expanded jugal flange, being more than 36% and up to 55% the depth of the caudal constriction
of the jugal (character 118-1; convergent in Iguanodon spp., G. notabilis, “hypacrosaurs” and
Tsintaosaurus spinorhinus); and jugal with infratemporal margin wider than that of the orbit
(character 123-2; convergent in Gryposaurus notabilis, G. monumentensis and Prosaurolophus
maximus). In contrast, the weighted parsimony and Bayesian analysis with gamma supported the
divergence of E. caroljonesa earlier than P. byrdi from the lineage leading to hadrosaurids.
Closer to Hadrosauridae we find a sequence of Eurasian species, Bactrosaurus johnsoni,
Gilmoreosaurus mongoiliensis and Telmatosaurus transsylvanicus, in this order. This result,
present in of all the phylogenies, is in conflict with previous hypotheses that placed T.
transsylvanicus as the first (Weishampel and Horner, 1990; Weishampel et al., 1993; Kirkland,
1998; Horner et al., 2004; Suzuki et al., 2004; Prieto-Marquez et al., 2006a) or among the first
(Head, 1998; Hu et al., 2001) hadrosaurid lineage. These findings are however more congruent
with the hypothesis of Wagner (2001), who placed T. transsylvanicus as a more distant outgroup
to hadrosaurids (Chapter 1 of this dissertation, Fig. 1.10B). In the present analysis, the position
of T. transsylvanicus relative to remaining hadrosaurs was supported by the following
unambiguous synapomorphies: at least three teeth per alveolus arranged dorsoventrally at mid
length of the dental battery (character 2-1); 1.25 or more teeth per cm of maxillary dental battery
(character 17-1; reversed in Gryposaurus, Saurolophus angustirostris and Charonosaurus
jiayinensis); medial side of the coronoid process ornamented with only fine striations (character
70
50-0); absence of lingual bowing of the occlusal plane, long axis of the occlusal plane parallel to
the lateral side of the dentary (character 53-1; convergent in Protohadros byrdi); ectopterygoid
shelf more than 35% the length of the alveolar margin of the maxilla (character 105-2; reversed
in Kerberosaurus manakini and Sahaliyania elunchunorum); subtriangular ventral surface of the
quadrate, lateral condyle rostrocaudally expanded and much larger than the medial condyle, the
lateral condyle being well offset ventrally relative to the ventral surface of the medial condyle
(character 129-1); and relatively high and long postzygapophyseal processes of cervical
vertebrae that are three times or more longer than the width of the neural arch (character 207-1;
reversed in Shantungosaurus giganteus).
All the analyses, with exception of the unweighted parsimony tree, showed that the
closest outgroup taxa to Hadrosauridae were two North American species, Lophorhothon atopus
and Claosaurus agilis, respectively (Figs. 2.7-2.13). The relationships among Telmatosaurus
transsylvanicus, Lophorhothon atopus and Claosaurus agilis were unresolved in the unweighted
parsimony analysis.
Hadrosauridae
This study was congruent with the original concept of Hadrosauridae of Cope (1869),
where Hadrosaurus foulkii is the type genus and species of the family. Thus, Hadrosauridae was
here defined as the clade stemming from the most recent common ancestor of Hadrosaurus
foulkii and Parasaurolophus walkeri. Hadrosauridae was supported by two unambiguous
synapomorphies: iliac supraacetabular process that projects lateroventrally between half and
three quarters of the dorsoventral depth of the central plate of the iilium (character 248-2;
convergent in Lophorhothon atopus); and craniocaudally short supraacetabular process, less than
55% the length of the central plate of the ilium (character 249-3; reversed in saurolophines,
except Edmontosaurus and Shantungosaurus giganteus).
Because the cranium of H. foulkii remains unknown (aside from a handful of teeth and a
a pair of poorly preserved maxillary fragments), numerous additional synapomorphies that might
also be diagnostic for hadrosaurids could only be ambiguously reconstructed. When the
accelerated transformation option (ACCTRAN) was applied to the ancestral state reconstructions
the following characters became synapomorphic for hadrosaurids: dentary with more than 30
tooth positions; loss of all but the primary ridge of dentary tooth crowns (reversed in Kritosaurus
navajovius, Gryposaurus notabilis, Olorotitan ararhensis, “amurosaurs” and “parasaurolophs”
71
except Charonosaurus jiayinensis); maxilla with more than 32 tooth positions (convergent in
Protohadros byrdi); maxilla with two functional teeth throughout most of the length of the dental
battery (these gradually changing to one near the rostral and caudal ends of the bone); dentary
symphysis forming an angle up to 15º with the lateral side of the rostral half of the element
(convergent in Ouranosaurus nigeriensis and P. byrdi); symphyseal region of the dentary gently
curved lingually forming a wide arch, with the labial convex side facing labioventrally (reversed
in Olorotitan ararhensis and “parasaurolophs”); dentary coronoid process rostrally inclined with
an angle between 69º and 82º; well developed expansion of both the caudal and, especially, the
rostral margins of the coronoid process; well developed expansion of the lateral side of the
dentary ventral to the coronoid process with an angle up to 165º between the lateral surface of
the dentary and that of the region caudoventral to the coronoid process; long axis of the dentary
occlusal plane parallel to the lateral side of the bone; caudal end of the dental battery located
caudal to the caudal margin of the coronoid process; ventrally-facing convex side of the lateral
flange and lateroventral surface of the the surangular; strong ventral deflection of the
premaxillary oral margin (so that, approximately, the dorsoventral distance between the occlusal
plane and the level of the premaxillary oral margin is equal to or larger than the mean depth of
the dentary) (convergent in Eolambia caroljonesa and Protohadros byrdi); subtriangular joint
surface for the jugal that is more laterally than dorsally-facing, with a lateroventrally-directed
pointed corner that is located adjacent and slightly dorsal to the proximal end of the lateral ridge
of the ectopterygoid shelf; maxillary foramina forming either a row or cluster that is oriented
rostrodorsally, ventral to the jugal articulation (reversed in Aralosaurus tuberiferus, Sahaliyania
elunchunorum and Amurosaurus riabinini); large rostral maxillary foramen opening on the
rostrolateral body of the maxilla, within the dorsal half of the rostrodorsal margin of the element
and exposed in lateral view; maxilla-lacrimal contact largely covered externally by the
premaxilla-jugal contact (reversed in saurolophines except Shantungosaurus riabinini and
Edmontosaurus spp. and in “parasaurolophs”); dorsoventrally deep caudodorsal margin of the of
the rostral process of the jugal (60-90% as deep as the rostral jugal constriction), being dorsally
or slightly recurved caudodorsally, forming the rostroventral corner of the orbital rim;
caudoventral apex of the rostral process of the jugal located ventral to the caudal margin of the
lacrimal process (reversed in Saurolophinae); medial articular surface of the rostral process of
the jugal facing medially and bounded only caudally by a rim of bone; mediolaterally spread out
72
basipterygoid processes of the basisphenoid forming a ventral angle of 100º or greater
(convergent in Iguanodon spp., O. nigeriensis, Tanius sinensis and reversed in Kerberosaurus
manakini and “amurosaurs”, “parasaurolophs” and Olorotitan ararhensis); recurved ventral
process of the coracoid so that the process is caudoventrally directed (reversed in
Shantungosaurus giganteus and some specimens of Brachylophosaurus canandesis); curved and
dorsally convex dorsal margin of the scapula, curvature starting at the level of the dorsal margin
of the pseudoacromion process and being most pronounced over the dorsoventral constriction of
the bone (convergent in O. nigeriensis, Tanius sinensis and reversed in Kerberosaurus manakini
and the Sabinas OTU); relatively long scapula, length/width ratio greater than 4 (reversed in
Shantungosaurus giganteus and “kritosaurs”); distal end of the scapular blade being as deep as or
deeper than the proximal region of the scapula (convergent in Bactrosaurus johnsoni and
reversed in Shantungosaurus giganteus and “kritosaurs”); and proximal constriction of the
scapula more than 60% the width of the cranial end of the bone (reversed in the UTEP OTU, K.
manakini, the Sabinas OTU, Pararhabdodon isonensis, Tsintaosaurus spinorhinus and
“amurosaurs” except Corythosaurus spp.).
The following ambiguous synapomorphies supported Hadrosauridae when the delayed
transformation option (DELTRAN) was applied to the ancestral state reconstructions: primary
ridge on dentary teeth located at the center of the crown for most teeth, with some teeth within
the same dentition displaying a slight caudal offset of the primary ridge (convergent in
“Claosaurus agilis” and Lophorhothon atopus.); ventral apex of the iliac supraacetabular
process located craniodorsally relative to the caudoventral margin of the lateral ridge of caudal
protuberance of the ischial peduncle convergent in “Claosaurus agilis”); supraacetabular
process that projects lateroventrally between half and three quarters of the dorsoventral depth of
the ilium (convergent in L. atopus and reversed Velafrons coahuilensis and Sahaliyania
elunchunorum); and distinctly depressed dorsal margin of the ilium over the supraacetabular
process and dorsally bowed over the proximal region of the preacetabular process (convergent in
L. atopus).
Euhadrosauria and the Hadrosaurid Dichotomy
The clade Euhadrosauria was originally defined by Weishampel et al. (1993) to include
all hadrosaurids except Telmatosaurus transsylvanicus. Euhadrosauria (= Saurolophinae plus
Lambeosaurinae). Here, Euhadrosauria was used also to emcompass the Saurolophinae-
73
Lambeosurinae. Formally, it was here defined as the clade stemming from the most recent
common ancestor of Edmontosaurus annectens and Parasaurolophus walkeri. The
Euhadrosauria was supported by the following unambiguous synapomorphies: dentary denticles
formed by two, three or more indentations in their apical margins (character 11-0); long
deltopectoral crest that is more than 55% the total length of the humerus (character 229-2;
reversed in “maiasaurs”, “kritosaurs”, except the Salitral Moreno OTU, Prosaurolophus
maximus and Charonosaurus jiayinensis); narrow and subsquared laterodistal margin of the
deltopectoral crest of the humerus with an angle less than 115º (character 231-2; reversed in
Brachylophosaurus and Maiasaura); and caudodorsally oriented dorsal margin of the
postacetabular process of the ilium (character 260-1; reversed in the Big Bend UTEP
hadrosaurid).
The following synapomorphies also supported the Euhadrosauria, although they were
ambiguous because the condition of these characters is unknown in the most immediate outgroup
taxa, Claosaurus, Lophorhothon and Hadrosaurus: more than 30 alveolar positions in the
dentary (with DELTRAN, character 1-1; increasing to more than 41 in saurolophines, with a
reversal back to more than 30 and less than 41 in “wulagasaurs”); more than 33 alveolar
positions in the maxilla (with DELTRAN, character 16-1; increasing to more than 45 in
saurolophines, with a reversal back to more than 33 and less than 44 in “maiasaurs”,
Gryposaurus spp. and the Secernosaurus clade); three functional teeth on the occlusal plane of
the dentary throughout most of the dental battery length (with DELTRAN, character 4.2); two
functional teeth on the occlusal plane of the dentary throughout most of the dental battery length
(with DELTRAN, character 19-2); increase in the ventral offset of the premaxillary-predentary
oral margin relative to the occlusal plane so that the dorsoventral distance between the occlusal
plane and the level of the premaxillary oral margin is equal to, or larger, than the mean depth of
the dentary (with DELTRAN, character 66-1); elevation of the rostrodorsal region of the maxilla
(with DELTRAN, character 97-1; reversed in Kerberosaurus manakini and “maiasaurs” except
Wulagasaurus dongi); and lateroventrally expanded deltopectoral crest, so that the ratio between
its width and that of the humeral shaft is 1.65 or more (character 230-1; codition unknown in
Aralosaurus and Jaxartosarus, reconstructed as present using ACCTRAN; convergent in Tanius,
Bactrosaurus and Gilmoreosaurus).
74
Saurolophinae and the Revised Hadrosaurinae
As explained before, historically one of the two major subclades of hadrosaurids,
Hadrosaurinae, included all those forms without or with solid crests and hypertrophied nasal
passages with more or less developed circumnarial fossae exposed on the lateral side of the skull
(Lull and Wright, 1942; Weishampel and Horner, 1990; Horner et al., 2004). The clade was
recovered in most trees derived from the present analyses. However, in all the phylogenies
inferred in this study the type of Hadrosaurinae, Hadrosaurus foulkii, was excluded from the
clade. Article 29.1 of the International Code of Zoological Nomenclature states that family-group
names must contain the stem of the name of type genus. Thus, if Hadrosaurus foulkii is not a
member of the clade Hadrosaurinae, the name of this monophyletic group cannot contain the
stem “hadrosaur”. Therefore, the name Hadrosaurinae may only be restricted to the lineage
formed by H. foulkii. Instead, this major clade of hadrosaurids was here given the name
Saurolophinae, originally coined by Brown (1914) (see Chapter 1 of this dissertation).
Saurolophinae was here defined as the most inclusive clade containing Edmontosaurus
annectens but not Lambeosaurus lambei. (Fig. 2.7). I selected Saurolophus osborni as the type
genus and species of this newly defined Saurolophinae. Henceforth, in the following discussions
the sister clade to Lambeosaurinae shall be known as Saurolophinae.
Unambiguous synapomorpies supporting Saurolophinae included: more than 42 alveolar
positions in the dentary (character 1-2; reversed in “maiasaurs” and “kritosaurs”); five or more
teeth per alveoli arranged dorsoventrally at mid length of the dentary dental battery (character 3-
3; reversed in “maiasaurs” and “kritosaurs”, except Gryposaurus); dentary marginal denticles
absent or very reduced to small papillae along the apical half of the dorsal half of the crown
(character 10-2; convergent in Bactrosaurus johnsoni, Charonosaurus jiayinensis and
“hypacrosaurs”); 45 or more tooth positions in the maxillary dental battery (character 16-2;
reversed “maiasaurs” and “kritosaurs”, except Kritosaurus navajovius, and convergent in
Hypacrosaurus stebingeri and Olorotitan ararhensis); wide arcuate and symmetrical
quadratojugal notch of the quadrate, the ventral half of the notch being rostroventrally-directed,
nearly as straight as its dorsal half (character 127-2; reversed in “saurolophs”); relatively long
parietal, length/width ratio greater than 2.35 (character 155-2; reversed in the Wahweap
hadrosaurid and Kritosaurus navajovius); relatively long exoccipital-supraoccipital shelf above
the foramen magnum that is substantially longer (often twice or more) than the diameter of the
75
foramen magnum (character 171-2; reversed in “maiasaurs”); caudodorsal margin of the iliac
supraacetabular process poorly defined and discontinuous with the dorsal margin of the proximal
region of the postacetabular process, lacking a well demarcated caudodorsal ridge (character
252-1; reversed in Gryposaurus latidens, Secernosaurus koerneri and the Salitral Moreno OTU);
long iliac peduncle of the ischium, with a length/width ratio greater than 2 (character 277-2;
convergent in Lophorhothon atopus and reversed in Brachylophosaurus canadensis, Maiasaura
peeblesorum and Gryposaurus notabilis); pubic peduncle of the ischium oriented parallel to the
ischial shaft (character 279-2; reversed in Saurolophus angustirostris); and dorsal acetabular
margin of the pubic peduncle of the ischium located above the level of the dorsal margin of the
ischial shaft (character 281-1; reversed in the Salitral Moreno OTU and S. angustirostris).
The position of “maiasaurs”. The clade including Brachylophosaurus canadensis and
Maiasaura peeblesorum has always been considered a member of Saurolophinae (Horner and
Weishampel 1990; Weishampel et al., 1993; Horner, 1992; Horner et al., 2004; Prieto-Marquez,
2005; Gates and Sampson, 2007; Godefroit et al., 2008). However, the consensus tree derived
from the unweighted parsimony analysis showed “maiasaurs” excluded not only from
Saurolophinae, but also from the Euhadrosauria as the first lineage to evolve after Hadrosaurus
foulkii (Figs. 2.5). Two unambiguous synapomorphies supporting this position of “maiasaurs”:
long ulna, up to 20% longer than the humerus (character 234-1) and caudodorsally oriented
dorsal margin of the postacetabular process of the ilium, rising dorsally relative to the acetabular
margin (character 260-1). In contrast, the Bayesian analyses with equal rates (Fig. 2.10) and the
weighted parsimony tree (Fig. 2.7) placed “maiasaurs” within Saurolophinae. The Bayesian
analysis with variable rates did not resolve the position of “maiasaurs”, positioning the clade in a
polytomic relationship with Gryposaurus, “saurolophs” and other clades, including
lambeosaurines (Fig. 2.12). Unambiguous synapomorphies supporting the position of
“maiasaurs” within saurolophines as a sister clade to “kritosaurs” included: dentary dental
battery with 31 to 42 alveolar positions (character 1-1; convergent in Lambeosaurinae) and at
least four teeth per alveoli arranged dorsoventrally at mid-length of the dentary dental battery
(character 3-2; convergent in Corythosaurus and Parasaurolophus).
The position of the Prosaurolophus-Saurolophus clade. The relationships
Prosaurolophus and Saurolophus within saurolophines represented one of the most notable cases
of character conflict among hadrosaurids in this study. On one hand, the facial anatomy of
76
Prosaurolophus and Saurolophus shares a number of characters with Edmontosaurus, such as
the elongated mandible with more than 50 alveolar positions per tooth row and the deeply
excavated caudal region of the circumnarial fossa. On the other hand, several other characters are
shared between the Prosaurolophus-Saurolophus clade and Gryposaurus, such as a thin and
reflected premaxillary oral margin and a prepubic blade with a squared dorsal margin. Thus,
previous analyses have shown that the Prosaurolophus-Saurolophus clade was either closely
related to Edmontosaurus (Horner and Weishampel, 1990; Weishampel et al., 1993; Kirkland,
1998; Gates and Sampson, 2007; Godefroit et al., 2008) or to Gryposaurus (Prieto-Marquez,
2005; Prieto-Marquez et al., 2006b).
In both Bayesian analyses, the position of Prosaurolophus and Saurolophus could not be
resolved (Figs. 2.10-2.13). However, in the weighted parsimony analysis (Fig. 2.7),
Prosaurolophus and Saurolophus were unambiguously related to Edmontosaurus, under the
“sauroloph” clade, on the basis of the following synapomorphies: ventral margin of the dentary
with a wide and prominent bulge rostral to the coronoid process (character 46-3); rostral end of
the nasal forming a long and subrectangular process with slightly rounded corners at the contact
with the dorsal process of the premaxilla (character 85-2); rostrocaudally constricted dorsal
region of the infratemporal fenestra caused by a nearly straight and oblique caudoventral margin
of the caudodorsal region of the postorbital (dorsal region of infratemporal fenestra typically
subtriangular) (character 137-1); caudal region of the circumnarial fossa deeply incised and
invaginated (character 189-1); short neural spines in caudal dorsal and sacral vertebrae, being up
to 2.11 times longer than the depth of their centra (character 209-0; convergent in Gryposaurus
and the Big Bend UTEP hadrosaurid); distal region of the scapular blade that is as deep as or
deeper than the proximal region of the scapula (character 223-1; convergent in Bactrosaurus
johnsoni, Brachylophosaurus canandensis, Maiasaura peeblesorum and lambeosaurines); an
absent or very faintly developed lateroventral protuberance of the proximal region of the ischial
peduncle of the pubis (character 272-0; convergent in Iguanodon, Ouranosaurus nigeriensis,
“kritosaurs”, Sahaliyania elunchunorum and Charonosaurus jiayinensis).
A contrasting hypothesis resulted from the unweighted parsimony analysis (Fig. 2.5).
Prosaurolophus and Saurolophus were found to be allied to the clade containing Gryposaurus,
but not to Edmontosaurus. Unambiguous synapomorphies in support of this relationship
included: proximal edentulous slope of the dentary as wide as 20 to 31% of the length of the
77
dental battery (character 37-1; convergent in Protohadros byrdi and all lambeosaurines except
Aralosaurus and Jaxartosaurus); oral margin of the premaxilla folded caudodorsally into a thin
recurved border (character 67-2); rostrodorsal process of the maxilla not exposed through the
narial foramen (character 93-0); relatively wide and shallow embayment along the ventral
margin of the jugal (character 117-0); long craniodorsal region of the scapula, reaching a length
of 45% of the depth of the proximal region of the scapula (character 227-1); rectangular dorsal
margin of the prepubic proces of the pubis (character 265-4); proximal constriction and distal
expansion of the prepubic process of the pubis having approximately the same length (character
267-1; convergent in Ouranosaurus nigeriensis, Lophorhothon atopus, Hadrosaurus foulkii and
“maiasaurs”); long ischial peduncle of the pubis, length/width ratio of 3 or more (character 271-
2); and disc-shaped pedal phalanges III2 and III3 more than three times wider than they are
proximodistally long (character 296-1; convergent in and all lambeosaurines except Aralosaurus
and Jaxartosaurus).
The position of Barsboldia sicinskii. Barsboldia sicinskii from the Nemegt Formation of
Mongolia has long been considered a lambeosaurine (Maryanska and Osmolska, 1981;
Weishampel and Horner, 1990; Horner et al., 2004) based on the great elongation of the neural
spines of the sacral and anterior caudal vertebrae (Maryanska and Osmolska, 1981). However, in
all the phylogenies of this study this species was inferred to be a member of Saurolophinae.
While the precise position of B. sicinskii within the Saurolophine was unresolved in the strict
consensus of both parsimony analyses (Figs. 2.6 and 2.9), in both Bayesian analyses this species
appeared as an outgroup taxon to all other saurolophines (Figs. 2.11 and 2.13) (with exception of
Shantungosaurus giganteus). A single unambiguous synapomorphy united Barsboldia sicinskii
to saurolophines: a poorly defined caudodorsal margin of the ilium that is discontinuous with the
dorsal margin of the proximal region of the postacetabular process due to the lack of a well
demarcated caudodorsal ridge.
Lambeosaurinae
This clade of “hollow-crested” hadrosaurids was defined here as the most inclusive clade
containing Lambeosaurus lambei but not Edmontosaurus annectens. Lambeosaurine monophyly
was supported by the following unambiguous synapomorphies: absence of a maxillary
rostrodorsal process, the rostral end of the maxilla forming a ventrally sloping rostrodorsal shelf
that underlies the premaxilla (character 92-1); dorsal maxillary process dorsoventrally taller than
78
it is wide with a peaked and caudally inclined apex (character 99-1); ventral margin of the jugal
with a relatively narrow and deep embayment (character 117-1; convergent in “maiasaurs”
except Wulagasaurus dongi and in Gryposaurus monumentensis and Edmontosaurus) and oval
supratemporal fenestra that is wider mediolaterally than rostrocaudally (character 202-1).
The partially preserved anatomy of Aralosaurus tuberiferus caused the following
characters to be ambiguous synapomorphies; these characters also supported Lambeosaurinae
when ACCTRAN was applied to the ancestral state reconstructions: ratio between the length of
the proximal edentulous slope of the dentary and that of the dental battery ranging from 0.20 to
0.31 (convergent in P. byrdi, Wulagasaurus dongi, Kerberosaurus manakini, the Sabinas OTU
and P. maximus); elongate caudodorsal processes that extends caudally to meet the caudoventral
process and forming the caudal margin of the external nares; large maxillary foramen located on
the dorsal surface of the maxilla along the maxilla-premaxilla contact that is not exposed
laterally; lacrimal short rostrocaudally with a relatively shorter and thinner rostral process;
caudoventral margin of the rostral process of the jugal being at least twice as deep as it is wide
that is sharply pointed and often recurved caudally; quadratojugal notch located near the mid-
length of the quadrate with a ratio between the distance from the mid-length of the notch to the
qudadrate head and the dorsoventral length of the element less than 0.6 (convergent in
Iguanodon spp., Ouranosaurus nigeriensis, Equijubus normani, Eolambia carljonesa and
Maiasaura peeblesorum); frontal completely excluded from the orbital margin by an extensive
articulation between the prefrontal and postorbital (convergent in Saurolophus spp.); sharp
annular ridge on the ventral surface of the frontal that defines the rostral extent of the cerebral
fossa; sagittal crest of the parietal that deepens caudally and is strongly inclined ventrally
(convergent in S. osborni); nasal passage nearly or completely enclosed by bone and formation
of internal cavities and passages such as lateral diverticula and a common median chamber;
caudal margin of the external naris formed entirely by the premaxilla; absence of circumnarial
fossa exposed on the lateral side of the skull; supracranial crest primarily composed of the nasals
and premaxillae; curved pseudoacromion process of the scapula, so that the cranial region is
dorsally or craniodorsally directed (convergent in iguanodontoidean outgroup taxa to all
hadrosaurs stemming from the most recent common ancestor of Telmatosaurus and
Parasaurolophus); ulna 20% longer than the humerus (reversed in Charonosaurus jiayinensis
and Parasaurolophus spp. and convergent in Brachylophosaurus canadensis); well-developed
79
recurvature of the caudodorsal corner of the iliac peduncle of the ischium so that the peduncle
appears “thumb-like” in lateral and medial profiles; distal end of the ischial shaft ventrally
expanded forming a large “foot-like” process (convergent in iguanodontoidean outgroup taxa to
all hadrosaurs stemming from the most recent common ancestor of Telmatosaurus and
Parasaurolophus); club-shaped distal end of the fibula with a concave cranial margin and a
prominent cranial expansion (reversed in “hypacrosaurs”); and short medial platform of the
astragalus, tapering laterally to underlie only part of the medial malleolus of the tibia.
The position of Aralosaurus tuberiferus. Aralosaurus tuberiferus was once considered a
saurolophine hadrosaurid closely related to Gryposaurus (Rozhdestvensky, 1968; Weishampel
and Horner, 1990; Norman and Sues, 2002), Aralosaurus tuberiferus has recently been revised
and included within the Lambeosurinae (Godefroit et al., 2004b). Horner (1992) first noted the
presence of a maxillary shelf in A. tuberiferus, one of the most obvious and undisputed
lambeosaurine synapomorphies. In addition to this character, Godefroit et al. (2004b) added the
shortened parietal, high squamosal and the possible presence of a hollow nasal crest rostral to the
orbits in making the same referal.
Both parsimony analyses, as well as the Bayesian analysis with gamma, confirmed
Horner’s and Godefroit and colleagues’ hypotheses that Aralosaurus tuberiferus is a
lambeosaurine (Figs. 2.5-2.9 and 2.12-2.13). Likewise, the phylogenetic position inferred by
Godefroit et al. (2004b), Gates et al. (2007) and Evans and Resiz (2007) that placed A.
tuberiferus as outgroup to all the other members of the clade was also supported. This result was
unambiguously supported by the following synapomorphies: absence of maxillary rostrodorsal
process; the rostral end of the maxilla formsa ventrally sloping rostrodorsal shelf that underlies
the premaxilla (character 92-1); dorsal maxillary process dorsoventrally taller than wide with a
peaked and caudally inclined apex (character 99-1); ventral margin of the jugal with a relatively
narrow and deep embayment (character 117-1; convergent in “maiasaurs” except Wulagasaurus
dongi and in Gryposaurus monumentensis and Edmontosaurus); and oval supratemporal
fenestra, wider mediolaterally than rostrocaudally (character 202-1).
An unexpected result was the placement of Aralosaurus tuberiferus as the sister taxon to
Euhadrosauria in the consensus tree derived from the Bayesian analysis without gamma (Fig.
2.10). In the tree derived from that analysis, four unambiguous synapomorphies united A.
tuberiferus and the Euhadrosauria. Thes eincluded 33 to 44 tooth positions in the maxilla
80
(character 16-1); elevation of the rostrodorsal region of the maxilla (character 97-1); wedge-
shaped and pointed rostral apex of the rostral process of the maxilla that is positioned within the
dorsal half of the rostral process of the jugal (character 111-1); deep caudodorsal margin of the
rostral process of the jugal (about 60-90% as deep as the rostral jugal constriction), that is
dorsally or slightly recurved caudodorsally and forms the rostroventral corner of the orbital rim
(character 112-1).
The position of Olorotitan ararhensis. Olorotitan ararhensis is a species of
lambeosaurine described by Godefroit et al. (2003) from material collected in Maastrichtian
strata near Kundur, in the Amur region of Far eastern Russia. Previous phylogenies have shown
O. ararhensis as most closely allied to “corythosaurs” (Godefroit et al., 2003; Evans and Reisz,
2007; Gates et al., 2007; Godefroit et al., 2008). This hypothesis was not supported in any of the
present analyses. Nevertheless, the position of this species varied in each of the inferred
phylogenies.
The Bayesian analysis without gamma placed O. ararhensis as outgroup to
“parasaurolophs” (Fig. 2.10). This hypothesis was supported by the following unambiguous
synapomorphies: lacriform external naris with a caudal constriction that occurs abruptly and is
primarily composed of either a lateroventral expansion of the caudodorsal process or a dorsal
expansion of the caudoventral process of the premaxilla (character 79-2; convergent in
Hypacrosaurus altispinus); primary ridge of dentary teeth straight in some crowns but sinuous in
others (character 8-1; convergent in “tsintaosaurus”, Shantungosaurus giganteus and outgroup
taxa to all hadrosaurs stemming from the most recent common ancestor of Gilmoreosaurus and
Parasaurolophus); and fontanellae on lateral crests surface completely closed in adults
(character 180-1; convergent in H. altispinus). On the other hand, the Bayesian analyses with
gamma placed O. ararhensis in a polytomy with “parasaurolophs” and “amurosaurs” (Fig. 2.12).
Both parsimony analyses shared the removal of Olorotitan ararhensis from its close relationship
with “parasaurolophs”, yet in different positions within Lambeosaurinae. The tree derived from
the unweighted parsimony analysis showed O. ararhensis as the sister taxon to “amurosaurs”
(Fig. 2.5). Unambiguous synapomorphies supporting this position included: dentary proximal-
most edentulous slope forming an angle of 150º or greater with the frontal plane of the skull
(chracter 38-1; convergent in Ouranosaurus nigeriensis, Bactrosaurus johnsoni,
Probactrosaurus spp., Brachylophosaurus canadensis, saurolophines except “kritosaurs” and
81
Parasaurolophus tubicen); equally wide caudal regions of the caudodorsal and caudoventral
processes of the premaxilla (character 80-1); and subrectangular dorsal margin of the
infratemporal fenestra that is approximately as wide as the ventral margin (character 200-0; a
reversal in the context of Lambeosaurinae). In contrast, the weighted parsimony analysis placed
O. ararhensis as the sister taxon to”amurosaurs” and “parasaurolophs” (Fig. 2.7). This position
was supported by the following unambiguous synapomorphies: dentary tooth crowns with a
primary ridge accompanied by one or two faint and shorter ridges (character 6-2; convergent in
Talmatosaurus transsylvanicus, Lophorhothon atopus, Kritosaurus navajovius and Gryposaurus
notabilis); ratio between the dorsoventral depth of the predentary rostral face (excluding the
median ventral process) and the length of the lateral process of 0.38 or less (character 25-1;
convergent in Protohadros byrdi); predentary rostral surface forming an angle of 40º or less with
the dorsal margin of the lateral process (character 26-3; convergent in Saurolophus osborni and
reversed in Velafrons coahuilensis); laterally curved caudal process of the surangular (character
60-1; convergent in Ouranosaurus nigeriensis and Bactrosaurus johnsoni); medial margin of the
proximal region of the surangular forming an angle up to 148º with the medial margin of the
proximal region of the caudal process (character 61-1; reversed in Hypacrosaurus stebingeri);
rostrolateral corner of the oral margin of the premaxilla triangular, dorsoventrally expanded,
ventrally-deflected and pointed (character 68-1; convergent in Eolambia caroljonesa, B. johnsoni
and “maiasaurs”, except Wulagasaurus dongi); lacriform and relatively short external
premaxillary naris, caudal constriction occurring abruptly and primarily composed of either a
lateroventral expansion of the caudodorsal process or a dorsal expansion of the caudoventral
process of the premaxilla (character 79-2; reversed in “amurosaurs”, except H. altispinus);
triangular and rostrocaudally compressed dorsolateral region of the rostral half of the maxilla
(character 96-1); dorsomedial margin of the prefrontal developed into a caudodorsally-oriented
crest that extends caudally over the dorsal surface of the frontal and above the prefrontal-
postorbital articulation in adults (character 130-2); and well-developed ventral convexity along
the distal half of the scaular blade (character 222-1; convergent in Prosaurolophus maximus and
reversed in Parasaurolophus cyrtocristatus).
The paraphyly of Hypacrosaurus. This study did not support the inclusion of H.
stebingeri within the genus Hypacrosaurus. Horner et al. (1992) originally regarded H.
stebingeri, from the late Campanian of the Two Medicine Fm. of Montana, USA, as having
82
evolved anagenetically from a taxon closely related to Lambeosaurus spp. These authors pointed
out that H. stebingeri had a cranial crest mostly composed of premaxillae like Lambeosaurus and
reduced narial opening like Hypacrosaurus altispinus. From this evidence they deduced that H.
stebingeri was a metaspecies, a taxon that had evolved through anagenesis from a species of
Lambeosaurus. However, my examination of the type skull of H. stebingeri, MOR 549, showed
that the narial opening is very elongated, not restricted, as in species of Lambeosaurus and
Corythosaurus. Furthermore, although the premaxilla forms a large part of the crest in H.
stebingeri, its contribution in no way appraoches the condition in Lambeosaurus. Neverthteless,
H. stebingeri also have greatly elongated dorsal vertebraral neural spines as in H. altispinus. The
predentary of H. stebingeri contains an autapomorphy: the mediolateral constriction of the rostral
region of the ventral median process of the predentary. This means that H. stebingeri is not a
metataxon (Horner et al., 1992; Horner and Currie, 1994) but a distinct species diagnosable by
this predentary autapomorphy.
Two recent phylogenetic analyses inferred the position of Hypacrosaurus stebingeri as
the sister taxon to H. altispinus (Evans and Reisz, 2007; Gates et al., 2007), whereas two others
could not resolve its position beyond its being a Lambeosaurine (Suzuki et al., 2004; Prieto-
Marquez et al., 2006b). In the present study, both parsimony analyses placed H. stebingeri
within the “corythosaur” clade and as sister taxon to Corythosaurus (Figs. 2.5 and 2.7).
Unambiguous synpomorphies supporing the inclusion of H. stebingeri within the “corythosaur”
clade included: mesial margin of dentary tooth crowns with larger and either more or less
denticles than the distal one, crowns imbricated so that the mesial margin overlaps the distal
margin of the adjacent crown (character 12-1, convergent in Telmatosaurus transsylvanicus);
ratio between the maximum mediolateral width of the predentary and the maximum rostrocaudal
length of the lateral process less than 1.2 (character 24-0; convergent in non-lambeosaurine
iguanodontoideans, except Ouranosaurus nigeriensis, Brachylophosaurus canadensis,
Maiasaura peeblesorum and Edmontosaurus); length/width ratio of the external bony naris of the
premaxilla greater than 2.95 (character 81-2); and nasal vestibule folded into an S-loop in the
enclosed premaxillary passages rostral to the dorsal process of the maxilla (character 182-1).
The following unambiguous synapomorphies support H. stebingeri as the sister taxon to
Corythosaurus: caudal region of the caudoventral process of the adult premaxilla that is
dorsoventrally broad and directed caudally (and slightly dorsally) (character 77-1); absence of
83
the short median ventral process located between the basipterygoid processes of the basisphenoid
(character 165-0; convergent in Parasaurolophus cyrtocristatus); and relatively short and stocky
humerus with a length/width ratio less than 4.25 (character 232-0; convergent in Tanius sinensis,
Saurolophus angustirostris, Olorotitan ararhensis, Hypacrosaurus altispinus and
Parasaurolophus).
The inclusion of H. stebingeri within “corythosaurs” was also supported in the Bayesian
phylogeny with gamma. Nevertheless, it was placed as an outgroup to a clade formed by
Lambeosaurus and Corythosaurus (Fig. 2.12). The relationships of H. stebingeri could not be
resolved in the Bayesian analysis without gamma and this species formed a polytomy with other
lambeosaurine taxa (Fig. 2.10).
The position of Nipponosaurus sachalinensis. Previous analysis have hypothesized a
close relationship between N. sachalinensis and Hypacrosaurus altispinus (Suzuki et al., 2004)
or positioned the former species as the sister taxon to a clade formed by Corythosaurus,
Lambeosaurus and Hypacrosaurus (Evans and Reisz, 2007). In the present study the two
Bayesian analyses did not support the position of Nipponosaurus sachalinensis within
”hypacrosaurs”. Instead it was part of a polytomy awith “parasaurolophs”, “corythosaurs” and
other lambeosaurines except Aralosaurus, Jaxartosaurus and “tsinatosaurus” (Fig. 2.10 and
2.12). The position of N. sachalinensis was supported by the following ambiguous
synapomorphies (the condition of which is unknown in Aralosaurus, Jaxartosaurus and
“tsintaosaurs”): elongate caudodorsal processes that extends caudally to meet the caudoventral
process and forming the caudal margin of the functional external nares (character 74-1); nasal
bone retracted caudal to the rostrum and occupying a supracranial position in the skull and with
the ventral region of the nasal meeting the prefrontal caudal to the rostral margin of the orbit
resulting in a convoluted narial passage and hollow crest that extend supraorbitally (character 83-
2); and nasal passage nearly or completely enclosed by bone (character 178-1).
In contrast, the unweighted parsimony analysis supported the inclusion of Nipponosaurus
sachalinensis in a polytomy with Hypacrosaurus altispinus, Lambeosaurus laticaudus and
Velafrons coahuilensis (Fig. 2.6). Support for this positioning included the presence of a well-
defined ridge forming the medial margin of a medioventrally-facing shelf, with a postacetabular
process that is progressively expanded mediolaterally towards the caudal end (character 257-1).
An alternative hypothesis was found in weighted parsimony analysis. The consensus tree (Fig.
84
2.9) pointed to N. suchalinensis as an outgroup to a clade comprising most lambeosaurines,
composed of Olorotitan arahensis, “amurosaurs” and “parasaurolophs”. This result was
supported by the following unambiguous synapomorphoes: dentary tooth crowns with a primary
ridge and one or two faint and shorter ridges (character 6-2); equally wide caudal regions of the
caudodorsal and caudoventral processes (character 80-1); and nasals that are rotated and folded
caudodorsally (character 84-1).
Lambeosaurus laticaudus. The lambeosaurine Lambeosaurus laticaudus was described
by Morris (1981) from partial cranial and postcranial elements collected in Campanian deposits
of El Gallo Formation from Baja California, Mexico. None of the phylogenetic analyses
conducted in the present study supported the inclusion of L. laticaudus in the genus
Lambeosaurus of Morris (1981). The Bayesian analysis with no gamma could not resolve the
position of L. laticaudus within a rather inclusive clade that excluded “tsintaosaurs” and
Jaxartosaurus aralensis. The Bayesian analysis with gamma positioned L. laticaudus within
“amurosaurs” in a polytomy with the Sahaliyania-Amurosaurus clade, Corythosaurus spp.,
Lambeosaurus spp., Hypacrosaurus altispinus, H. stebingeri and Velafrons coahuilensis.
Only the parsimony analyses provided a resolved phylogenetic position for
Lambeosaurus laticaudus, namely as the sister taxon to Hypacrosaurus altispinus within the
“hypacrosaur” clade (Figs. 2.5 and 2.7). This relationship was recovered in both of the
parsimony analyses. Unambiguous synapomorphies supporting the position of L. laticaudus
within “hypacrosaurs” included: dentary denticles absent or very reduced to small papillae along
the apical half of the dorsal half of the crown (character 10-2; convergent in Charonosaurus
jiayinensis, all saurolophines except Gryposaurus latidens and the Wahweap hadrosaurid and
Bactrosaurus johnsoni); and a subtriangular distal end of the fibula with a straight or slightly
concave cranial margin (character 290-0; convergent in all other iguanodontoideans, except
Tanius sinensis). Unambiguous synapomorphies shared among L. laticaudus, Velafrons
coahuilensis and H. altispinus included: presence of brevis shelf at the base of the iliac
postacetabular process (character 256-1, condition inferred through parsimony but not preserved
in specimens of L. laticaudus; convergent in Tsintaosaurus spinorhinus, Secernosaurus koerneri,
the Salitral Moreno and UTEP OTU’s and iguanodontoidean outgroup taxa to all hadrosaurs
stemming from the most recent common ancestor of Telmatosaurus and Parasaurolophus); and
well-defined ridge forming the medial margin of a medioventrally-facing shelf in the
85
postacetabular process that is progressively expanded mediolaterally towards the caudal end
(character 257-2). Likewise, L. laticaudus and H. altipinus were unambiguously united by
having a thick ischial shaft, with a breadth greater than 7.5% of its length (character 282-2;
convergent in Bactrosaurus johnsoni, Secernosaurus koerneri, the Salitral Moreno OTU,
Corythosaurus, Lambeosaurus magnicristatus and Parasaurolophus cyrtocristatus).
Discussion
From all the phylogenies inferred in this study, I consider the hypothesis from the
weighted parsimony analysis as most likely to reflect the true phylogeny of hadrosaurids. This is
because it required the least number of changes to explain the origin of a complex of characters
in the hadrosaur skull. These characters are in the premaxillo-nasal complex of the facial
skeleton of hadrosaurs. Specifically, they consist of the lost of the rostrodorsal process of the
maxilla (associated to a caudodorsal rotation of the face), the elevation and reconfiguration of the
skull roof to form crest-like structures and the extension and development of the circumnarial
fossa caudal to the narial foramen. Cranial crests and circumnarial fossae dramatically changed
the anatomy of the skull in hadrosaur lineage. Indeed, the most distinct morphological changes in
the skull between hadrosaurs and remaining members of the Ornithopoda consist of the
development of these cranial crests and circumnarial structures (dental batteries and feeding-
related attributes).
In the weighted parsimony analysis, both cranial crests and caudal extension of
circumnarial fossae were reconstructed to have a single origin (coincident with the evolution of
the most recent common ancestor of Lophorothon and Parasaurolophus) (Fig. 2.14B).
Conversely, the alternative topology provided by the unweighted parsimony analysis differed
from all others in placing “maiasaurs” as outgroup to euhadrosaurs. The ramifications of this are
that cranial crests require five independent derivations (in Lophorhothon, “maiasaurs”, the clade
including “kritosaurs”, Prosaurolophus and Saurolophus, and in lambeosaurines) to be
explained. In my opinion this seems less likely than a single origination event for these crests.
Similarly, the extreme development of circumnarial fossae present in Edmontosaurus,
Prosaurolophus and Saurolophus is also reconstructed as having evolved once in the weighted
parsimony phylogeny because those three genera are united in their own clade (Fig. 2.14I). This
86
condition would require two independent origins in the unweighted parsimony phylogeny, since
in that hypothesis Edmontosaurus and Prosaurolophus-Saurolophus do not form a clade.
I also consider more likely that Aralosaurus tuberiferus is a member of Lambeosaurinae,
as shown in all but the phylogeny derived from the Bayesian analysis without gamma. A.
tuberiferus has a maxilla that lacks a rostrodorsal process but possesses a premaxillary shelf, a
character present in all lambeosaurines. In this clade, the presence of this shelf supports the
premaxilla and is always associated to a caudodorsal rotation of the facial premaxillonasal
structure of the skull (Wagner, 2001). Acceptance of the phylogeny inferred in the Bayesian
analysis with no gamma would require the independent origin of these attributes in Aralosaurus,
its subsequent loss ancestrally in Euhadrosauria and a second, independent origin in
lambeosaurines. It is more conservative and parsimonious to choose a hypothesis where these
typically lambeosaurine characters evolved only once in the Euhadrosauria.
Therefore, the following discussion of character evolution in hadrosaurids is based on the
phylogeny inferred using weighted parsimony. Appendix K contains a complete list of all the
clades inferred in that analysis, including their definitions, recorded temporal ranges and
unambiguous synapomorphies in support of each clade.
Character Evolution in Hadrosaurid Dinosaurs
The following discussion is concerned with the origin and evolution of the more
distinctive characters of hadrosaurids. The evolutionary history of these characters was derived
from ancestral state reconstructions using parsimony on the clades inferred in the weighted
parsimony phylogeny (Fig. 2.7).
Cranium. One of the most distinctive attributes of the hadrosaur skull is the
mediolaterally expanded rostrum from which the “duck bill dinosaur” homonym derives.
Anatomically, it is the expansion of the premaxillary bone that contributes to the bill (Fig.
2.14A). This analysis revealed that substantial expansion of the premaxilla first occurred in the
most recent common ancestor of Protohadros byrdi and Parasaurolophus walkeri, where the
premaxillary oral margin became between 1.65 and twice as broad as the postoral constriction
(character 65; Appendix D, Fig. D.1). This condition was retained in all hadrosaurs. In
“saurolophs” and “wulagasaurs” the premaxilla underwent further mediolateral expansion
relative to the ancestral condition (more than twice as broad as the postoral constriction).
87
Reversals to the ancestral condition occurred in Gryposaurus notabilis, G. monumentensis and
Saurolophus spp.
Cranial crests and the caudal extension of the circumnarial fossa were reconstructed to
appear for the first time in the most recent common ancestor of Lophorhothon atopus and
Parasaurolophus walkeri, no later than the late Santonian (Fig. 2.14B). This co-occurrence of
these two characters supports Hopson (1975) and Wagner (2001; 2004), who sugegsted that, at
least ancestrally, crests evolved as extensions of the skull roof for providing additional surface
for the caudal expansion of the circumnarial fossa. A more extreme excavation of the
circumnarial fossa occurred later in “saurolophs” (Fig. 2.14I).
The origin of Euhadrosauria correlates with the evolution, no later than the middle-late
Santonian, of several characters related to feeding. The number of alveolar positions in the
dentary and maxilla increased to more than 30 and more than 33, respectively (Fig. 2.14C). This
condition was retained in lambeosaurines. In saurolophines the dentary and maxillary dental
batteries acquired more than 42 and 45 tooth families, respectively (Fig. 2.14G). However, the
number of dentary and maxillary teeth in “maiasaurs” and “kritosaurs” became reversed to the
ancestral euhadrosaur condition (Fig. 2.14H). Associated with the increase of the number of
tooth families in the maxilla and dentary is an increase in the width of the occlusal plane. This
was achieved by bringing three (rarely four) teeth, arranged labiolingually into functional
positions (Fig. 2.14D). Concurrent with these dental changes were modifications of the jaws so
that in the plane of occlusion relative to the quadrate artculuar joint (jaw joint). This is reflected
in my documentation of the oral contact between the premaxilla and predentary descending well
below the occlusal plane in euhadrosaurians so that the dorsoventral distance between the
occlusal plane and the level of the premaxillary oral margin is equal to, or larger, than the mean
depth of the dentary (Fig. 2.14E).
The fact that all these characters appeared for the first time in the most recent common
ancestor of all euhadrosaurs opens up the possibility that they might have been key innovations
(Hunter and Jernvall, 1995; Bond and Opell, 1998) that helped spring the remarkable radiation of
hadrosaurids into the diverse and successful saurolophine and lambeosaurine clades.
At the same time, the rostrodorsal region of the maxilla became elevated relative to the
depth observed in outgroup taxa to Euhadrosauria (Fig. 2.14F). Previous authors have correlated
the elevation of the maxillary rostrodorsal region to the migration of the antorbital fenestra
88
dorsally near the premaxillary articulation surface (Weishampel et al., 1993; Godefroit et al.,
2001). The occurence of this character at the origin of Euhadrosauria might also have been
instrumental in the hadrosaurid radiation. A reversal of this condition to a relatively low
rostrodorsal region of the maxilla occurred in “maiasaurs” (except Wulagasaurus),
Kerberosaurus and the lambeosaurine Sahaliyania.
Postcranium. Several modifications to the humeral anatomy evolved in the most recent
common ancestor of euhadrosaurians (Fig. 2.15C). Specifically, ancestral state reconstructions of
characters of the deltopectoral crest indicate that it became elongated to greater than than 55%
the total length of the humerus (reversed in “maiasaurs” and “kritosaurs”, except the Salitral
Moreno OTU). The laterodistal margin evolved to be more angulose and subsquared, with
margins forming an angle less than 115º (reversed in Brachylophosaurus and Maiasaura).
Possibly allied with these changes is an increase in the projection of the deltopectoral crest so
that the ratio between its width and that of the humeral shaft was 1.65 or more (convergent in
hadrosaur outgroup taxa Tanius, Bactrosaurus and Gilmoreosaurus). These apomorphies
originated in the most recent common ancestor of euhadrosaurs. The deltopectoral crest further
expanded in lambeosaurines, so that its width doubled that of the humeral shaft.
Primitively, the acromion process of the scapula in iguanodontoidean outgroups to
hadrosaurids is dorsally projected (Fig. 2.15A). In lambeosaurines, this process is also dorsally
oriented but with a more rostral component (Fig. 2.15D). Ancestral state reconstruction for this
character suggests that the acromion process changed from dorsally oriented to horizontal in the
most recent common ancestor of Lophorhothon and Parasaurolophus (Fig. 2.15B) during the
late Coniacian-early Santonian and that the lambeosaurine condition was independently derived.
In the pelvic girdle, the iliac characters commonly observed in hadrosaurids appear to have
evolved before the euhadrosaur radiation. In the ilium, the change from an ischal peduncle
composed of a single knob to one composed of two protuberances (Godefroit et al., 2001)
occurred in the most recent common ancestor of Gilmoreosaurus and Parasaurolophus during
the late Coniacian-early Santonian (Fig. 2.16E). The subsequent cranial shift of the apex of the
supraacetabular process relative to the position of the caudal protuberance of the ischial peduncle
evolved no later than the late Coniacian in the most recent common ancestor of Claosaurus agilis
and Parasaurolophus (Fig. 2.16F). The reduction of the craniocaudal width and the lateroventral
expansion of the supraacetabular process were acquired in the most recent common ancestor of
89
Hadrosauridae no later than the late Santonian (Fig. 2.16G). Finally, the brevis shelf of the
postacetabular process was lost no later than in the most recent common ancestor of
Lophorhothon and Parasaurolophus during the late Coniacian-early Santonian (Fig. 2.16B).
However, the postacetabular process of “hypacrosaurs” and the Secernosaurus clade shows a
brevis shelf that is posited to have evolved independently in each of these two clades (Fig. 2.16C
and D). The independent origin of the brevis shelf in these clades appears to have involved the
same mechanism, that being a medioventral twisting of the postacetabular process. This contrasts
with being acquired through mediolateral thickening of the postacetabular process, as is the case
in primitive iguanodontoideans.
The geometry of the blade of the prepubic process of the pubis is one of the most
dignostic characters of the hadrosaurid postcrania. Taxa such as Gryposaurus spp.,
Prosaurolophus maximus, Saurolophus spp. and the Sabinas hadrosaurid display a characteristic
prepubic blade where the dorsal margin forms a rectangular edge and distally becomes parallel to
the ventral margin (Appendix I, Fig. I.30). This condition was found to be synapomorphic for all
saurolophines except Shantungosaurus (Fig. 2.17D). The smoother and rounded dorsal margins
of the prepubic blades of Edmontosaurus (Fig. 2.17B) and “maiasaurs” (Fig. 2.17C) represent
derived conditions from the squared ancestral geometry of the other saurolophines. On the other
side of the euhadrosaur tree, the “fist-like” prepubic blade of lambeosaurines (Fig. 2.17E)
represents a modification of the primitive geometry present in outgroup iguanodontians, where
the blade is more dorsally expanded and the ventral concavity is shifted caudally (Fig. 2.17A).
Lambeosaurines and primitive iguanodontoideans share three characters of the ischium:
caudally curved caudodistal margin of the iliac peduncle cranioventrally oriented pubic peduncle
and the presence of a ventral expansion in the distal shaft (“ischial foot”) (Figs. 2.18A-C).
However, the cranioventral orientation of the pubic peduncle is much more exaggerated in
primitive iguanodontoideans (Fig. 2.18A), while the curvature of the iliac peduncle is much more
developed (“thumb-like” lateral profile) in lambeosaurines (Fig. 2.18H). Ancestral state
reconstructions for these characters indicated that caudal curvature of the iliac peduncle and the
“ischial foot” evolved independently in lambeosaurines (Figs. 2.18H and I). This occurred after
the latter became lost and no later than in the most recent common ancestor of Lophorhothon and
Hadrosauridae (Fig. 2.18G) during the late Coniacian-early Santonian. The pubic peduncle
became slightly cranioventrally oriented (i.e., more rostrally than ventrally) in the most recent
90
common ancestor of Bactrosaurus and Parasaurolophus (Fig. 2.18E) during the late Coniacian.
This condition was retained in the most recent common ancestor of euhadrosaurs and in
lambeosaurines. Notably, Saurolophinae evolved a derived condition of this character whereby
the pubic peduncle was oriented horizontally, parallel to the long axis of the ischial shaft (Fig.
2.18F).
Conclusions
Parsimony and Bayesian analyses were conducted on a complete taxonomic sample of

hadrosaurid species. The closest outgroup taxa to Hadrosauridae were found in eastern North
America. Hadrosauridae was redefined as the clade stemming from the most recent common
ancestor of Hadrosaurus foulkii and Parasaurolophus walkeri. In general these analyses
confirmed the evolution of hadrosaurids into Saurolophinae (Hadrosaurinae of most previous
authors) and Lambeosaurinae. Saurolophines consist of “saurolophs” and “kritosaurs”.
Lambeosaurines consist of a succession of Eurasian outgroups to two major clades,
“parasaurolophs” and “amurosaurs” (“hypacrosaurs” and “corythosaurs”).
The hadrosaurid radiation and the divergence of saurolophines from lambeosaurines
occurred no later than the Santonian and was coincident with the evolution of a suite of
mandibular characters (i.e., increased number of alveolar positions, presence of three teeth
forming the dentary occlusal plane, ventral offset of the oral predentary-premaxilla contact).
These results suggest that feeding adaptations might have played a central role in the
diversification of hadrosaurids.
Finally, two of the most distinctive and unique attributes of hadrosaurids, circumnarial
fossae and cranial crests, evolved prior to the major radiation of these animals. Likewise, the fact
that both characters were reconstructed to have evolved at the same time supports the possibility
that cranial crests evolved to provide additional space in the skull to allow for the caudal
extension of circumnarial fossae.
91
A
Figure 2.1. Dentaries of two specimens of Edmontosaurus in lateral view, showing evidence of
postdepositional dorsoventral compresion in the form of bending lines (arrows). A, left dentary of
E. regalis, CMN 2289. B, right dentary of CMN 2289. C, right dentary of E. annectens, cast of
AMNH 5730 (= Anatotitan copei).
92
A
Figure 2.2. Signs of postdepositional dorsoventral compression (arrows) in a specimen of

Edmontosaurus annectens, cast of AMNH 5730 (type of Anatotitan copei), lateral view.
A, dorsal region of the left quadrate. B, left postorbital.
93
A
Figure 2.3. Right prefrontal of Prosaurolophus maximus (MOR 454-6-24-6-2), part of the holotype
of P. blackfeetensis, showing the presence of a concave surface (arrows) dorsomedial to the orbital
margin, in A, dorsal and. B, lateral views.
94
−310
−320
BIC
−330
−340
−350
E V
2 4 6 8 A
Number of Clusters
● ●
●
900
● ●
●
●
●
800
●
700
BIC
600
500
EII VVI
VII EEE
400
● EEI EEV
VEI VEV
EVI VVV
2 4 6 8 B
Number of Clusters
Figure 2.4. BIC plots resulting from clustering model analyses in Mclust. A, BIC plot two clustering models
of one dimensional dataset. B, BIC plot for various clustering models of NMMDS derived from a GDA
dissimilariy matrix. Identifiers of clustering models: EII = spherical, equal volume and shape; VII = sperical, variable
volume, equal shape; EEI = diagonal, equal volume and shape; VEI = diagonal, variable volume, equal shape;
EVI = diagonal, equal volume, variable shape; VVI = diagonal, variable volume and shape; EEE = ellipsoidal, equal
volume, shape and orientation; EEV = ellipsoidal, equal volume and shape, variable orientation; VEV = ellipsoidal,
variable volume, equal shape; VVV = ellipsoidal, variable volume and shape (Fraley and Raftery, 2003).
95
1/- Hypacrosaurus altispinus
1/- Lambeosaurus laticaudus
1/- Velafrons coahuilensis
1/28 Sahaliyania elunchunorum
2/- Amurosaurus riabinini
2/81 Corythosaurus casuarius
1/- Corythosaurus intermedius
1/- 1/- Hypacrosaurus stebingeri
3/89 Lambeosaurus lambei
Lambeosaurus magnicristatus
3/38 Olorotitan ararhensis
1/77 Parasaurolophus walkeri
3/73 Parasaurolophus tubicen
2/49 4/57 Parasaurolophus cyrtocristatus
Charonosaurus jiayinensis
2/55
2/61 Pararhabdodon isonensis
LAMBEOSAURINAE
3/47 Tsintaosaurus spinorhinus
Jaxartosaurus aralensis
Aralosaurus tuberiferus
1/75 Gryposaurus notabilis
3/59 Gryposaurus monumentensis
2/- Gryposaurus latidens

4/93 Secernosaurus koerneri
3/- 2/20 Salitral Moreno OTU
EUHADROSAURIA 3/-
1/- UTEP Big Bend OTU
Wulagasaurus dongi
2/-
6/99 Saurolophus osborni
3/57
2/- 1/- Prosaurolophus maximus
3/45
Kerberosaurus manakini
SAUROLOPHINAE Sabinas OTU
5/90 Edmontosaurus annectens
2/-
HADROSAURIDAE 1/37
Shantungosaurus giganteus
2/75 Brachylophosaurus canadensis
3/88 Maiasaura peeblesorum
1/53 Wahweap OTU
Hadrosaurus foulkii HADROSAURINAE
1/22 Telmatosaurus transsylvanicus
Lophorhothon atopus
1/27 Claosaurus agilis
2/50 Gilmoreosaurus mongoliensis
5/84
Tanius sinensis
3/70 1/37 Eolambia caroljonesa
HADROSAURIA Protohadros byrdi
4/71 Probactrosaurus spp.
Equijubus normani
Iguanodon spp.
Ouranosaurus nigeriensis 10 changes
Figure 2.5. Strict consensus of the 41 most parsimonious trees resulting from the unweighted parsimony
analysis of 49 hadrosaur taxa, excluding Barsboldia sicinskii and Nipponosaurus sachalinensis. At
each node, the pair of numbers above a branch represents, from left to right, a decay index and a bootstrap
proportion. Bootstrap proportions lower than 20 are indicted by a “-”. Only taxa with complete skulls
are illustrated with photographs.
96
1/- Nipponosaurus sachalinensis
Lambeosaurus laticaudus
1/-
Velafrons coahuilensis
1/27 Sahaliyania elunchunorum
2/- Amurosaurus riabinini
2/81 Corythosaurus casuarius
1/- Corythosaurus intermedius
1/- Hypacrosaurus stebingeri
1/- Lambeosaurus lambei
3/88 Lambeosaurus magnicristatus
2/20 Olorotitan ararhensis
1/75 Parasaurolophus walkeri
3/69 Parasaurolophus tubicen
2/30 4/54
Parasaurolophus cyrtocristatus
2/36
2/61 Pararhabdodon isonensis
LAMBEOSAURINAE 1/32 Tsintaosaurus spinorhinus
1/73 Gryposaurus notabilis
2/59 Gryposaurus monumentensis
EUHADROSAURIA 2/- 4/93 Secernosaurus koerneri
Salitral Moreno OTU
2/20
UTEP OTU
1/96 Saurolophus osborni
1/55 Saurolophus angustirostris
SAUROLOPHINAE 1/-
Wulagasaurus dongi
3/44 Edmontosaurus annectens
HADROSAURIDAE 1/52 Sabinas OTU
Barsboldia sicinskii
2/74 Brachylophosaurus canadensis
3/87 Maiasaura peeblesorum
2/59
Wahweap OTU
1/23 Telmatosaurus transsylvanicus
Lophorhothon atopus
1/29 Claosaurus agilis
2/40 Gilmoreosaurus mongoliensis
5/84
Tanius sinensis
3/70 Eolambia caroljonesa
HADROSAURIA 1/34 Protohadros byrdi
4/71 Probactrosaurus spp
Equijubus normani
Iguanodon spp.
Figure 2.6. Strict consensus of the 1422 most parsimonious trees resulting from the unweighted parsimony
analysis of 51 hadrosaur taxa, showing the position of Barsboldia sicinskii and Nipponosaurus
sachalinensis. At each node, the pair of numbers above a branch represents, from left to right, a decay index
and a bootstrap proportion. Bootstrap proportions lower than 20 are indicted by a “-”.
97
.04/-
.05/-
.05/- Velafrons coahuilensis
Sahaliyania elunchunorum
.05/39 Amurosaurus riabinini
.05/25
.12/79 Corythosaurus casuarius
.07/47 Corythosaurus intermedius
.05/25 .07/38 Lambeosaurus lambei
.19/95 Lambeosaurus magnicristatus
.11/84 Parasaurolophus walkeri
.30/88 .19/77 Parasaurolophus tubicen
.42/89 Charonosaurus jiayinensis
.22/73
.25/75 .29/83 Pararhabdodon isonensis
LAMBEOSAURINAE .19/74 Tsintaosaurus spinorhinus
Wahweap OTU
.19/95
.17Brachylophosaurus canadensis
.02/27 84
Wulagasaurus dongi
.29/80 Gryposaurus notabilis
.02/- .29/90 Gryposaurus monumentensis
EUHADROSAURIA .07/68 .05/-
.11/86 Secernosaurus koerneri
.03/- Salitral Moreno OTU
.07/20
.05/39 UTEP Big Bend OTU
.02/- Sabinas OTU
HADROSAURIDAE .08/53 .13/60
.87/100 Saurolophus osborni
.02/- Saurolophus angustirostris
.03/- Edmontosaurus annectens
SAUROLOPHINAE .16/94
.03/42 .03/22 Kerberosaurus manakini
.19/62
.29/73 Claosaurus agilis
.07/30 Lophorhothon atopus
.19/40
Gilmoreosaurus mongoliensis
.11/48
.10/79
Tanius sinensis
.26/68
Protohadros byrdi
.20/80
HADROSAURIA Eolambia caroljonesa
Probactrosaurus spp.
.10/54
Equijubus normani
Iguanodon spp.
Figure 2.7. Single most parsimonious tree resulting from the weighted parsimony analysis of 49 hadrosaur
taxa, excluding Barsboldia sicinskii and Nipponosaurus sachalinensis. At each node, the pair of numbers
above a branch represents, from left to right, a decay index and a bootstrap proportion. Bootstrap proportions
lower than 20 are indicted by a “-”. Only taxa with complete skulls are illustrated with photographs.
98
Iguanodon spp.
1 Equijubus normani
2 Protohadros byrdi
3 Tanius sinensis
4 Bactrosaurus johnsoni
5 Gilmoreosaurus mongoliensis
6 Telmatosaurus transsylvanicus
7 Lophorhothon atopus
8 Claosaurus agilis
9 Shantungosaurus giganteus
SAUROLOPHINAE 13 Kerberosaurus manakini
10 17 Edmontosaurus annectens
16 Edmontosaurus regalis
Prosaurolophus maximus “saurolophs”
14
Saurolophus osborni
HADROSAURIDAE 11 18
19 Saurolophus angustirostris
15 Sabinas OTU
22 Wulagasaurus dongi
Wahweap OTU
“maiasaurs”
20 Brachylophosaurus canadensis
23
24 Maiasaura peeblesorum
EUHADROSAURIA 12 21 Kritosaurus navajovius “wulagasaurs”
27 Gryposaurus latidens
25 “kritosaurs”
28 Gryposaurus notabilis
26 UTEP OTU
29 Salitral Moreno OTU
30 Secernosaurus koerneri
LAMBEOSAURINAE 34 Pararhabdodon isonensis
31 Tsintaosaurus spinorhinus
“tsintaosaurs”
32 Olorotitan ararhensis
43 Sahaliyania elunchunorum
33 42 Amurosaurus riabinini
Hypacrosaurus altispinus “hypacrosaurs”
35 37 44 Velafrons coahuilensis
45 “amurosaurs”
Lambeosaurus lambei
41 Lambeosaurus magnicristatus
38 Hypacrosaurus stebingeri “corythosaurs”
36 Corythosaurus casuarius
39
40 Corythosaurus intermedius
46 Parasaurolophus tubicen
“parasaurolophs”
47
48 Parasaurolophus cyrtocristatus
Figure 2.8. Single most parsimonious cladogram derived from the weighted parsimony analysis,
showing the names of the inferred clades used in this study. The numbers at the nodes are tags used
to refer to the clades defined in Appendix K. Likewise, names between quotation marks are informal
names created to facilitate the reference to specific clades.
99
.03/- Hypacrosaurus altispinus
.12/- Lambeosaurus laticaudus
.07/- Velafrons coahuilensis
.10/33 Amurosaurus riabinini
.05/23
.13/77 Corythosaurus casuarius
.05/49 Corythosaurus intermedius
.18/21 .15/34 Lambeosaurus lambei
.10/95 Lambeosaurus magnicristatus
.10/84 Parasaurolophus walkeri
.19/- .15/73 Parasaurolophus tubicen
.06/- .22/85
.19/49
Nipponosaurus sachalinensis
.12/54
.19/88
Pararhabdodon isonensis
LAMBEOSAURINAE .20/50 Tsintaosaurus spinorhinus
Wahweap OTU
.10/93
.12 Brachylophosaurus canadensis
.09/25 83
Wulagasaurus dongi
.02/81 Gryposaurus notabilis
.12/- .12/87 Gryposaurus monumentensis
EUHADROSAURIA .19/71 .28/-
.18/87 Secernosaurus koerneri
.25/- Salitral Moreno OTU
.05/-
.24/42 UTEP Big Bend OTU
Sabinas OTU
.40/97 Saurolophus osborni
HADROSAURIDAE .23/52 .29/28 .16/57 Saurolophus angustirostris
SAUROLOPHINAE
.25/44 Kerberosaurus manakini
.11/61
.29/70 Claosaurus agilis
.02/32 Lophorhothon atopus
.11/37
.11/46
.10/79
Tanius sinensis
.15/66
Protohadros byrdi
.18/79
.16/55
Equijubus normani
Iguanodon spp.
Figure 2.9. Strict consensus of the 11 most parsimonious trees resulting from the weighted parsimony
analysis of 51 hadrosaur taxa, showing the position of Barsboldia sicinskii and Nipponosaurus
sachalinensis. At each node, the pair of numbers above a branch represents, from left to right, a decay
index and a bootstrap proportion. Bootstrap proportions lower than 20 are indicted by a “-”.
100
Iguanodon spp.
Equijubus normani
0.99
HADROSAURIA 0.82 Eolambia caroljonesa
1 Protohadros byrdi
Tanius sinensis
1
0.99 Gilmoreosaurus mongoliensis
0.68 Telmatosaurus transsylvanicus
0.73 Lophorhothon atopus
0.99 Claosaurus agilis
0.96
0.95 Shantungosaurus giganteus
0.98 Edmontosaurus annectens
HADROSAURIDAE 0.87 Wulagasaurus dongi
0.91
0.87
SAUROLOPHINAE 1 Maiasaura peeblesorum
0.73 Wahweap OTU

0.75 Kritosaurus navajovius
0.84 Gryposaurus notabilis
0.98
0.79
1 Secernosaurus koerneri
0.80 Salitral Moreno OTU
UTEP Big Bend OTU
EUHADROASAURIA 0.62 Sabinas OTU
0.79
1
1 Saurolophus osborni
0.95 Tsintaosaurus spinorhinus
0.75 Sahaliyania elunchunorum
LAMBEOSAURINAE 0.94 Amurosaurus riabinini
0.81 Parasaurolophus walkeri
0.98
Parasaurolophus tubicen
0.83 0.82 1 Parasaurolophus cyrtocristatus
0.81 Corythosaurus casuarius
Corythosaurus intermedius
0.88
0.99
1 Lambeosaurus lambei
0.3
Figure 2.10. Consensus of the Bayesian analysis with equal rates of character change showing the
phylogenetic relationships of 49 hadrosaurs, excluding Barsboldia sicinskii and Nipponosaurus
sachalinensis. At each node, the number indicates a posterior probability. Only the names of taxa with
complete skulls are accompanied by a photograph.
101
Iguanodon spp.
Equijubus normani
.99
HADROSAURIA 0.82 Eolambia caroljonesa
1 Protohadros byrdi
1 Tanius sinensis
0.99
0.68
0.96 Hadrosaurus foulkii HADROSAURINAE
0.95 Jaxartosaurus aralensis
Tsintaosaurus spinorhinus
HADROSAURIDAE 0.84 0.97 Pararhabdodon isonensis
0.88 Hypacrosaurus stebingeri
LAMBEOSAURINAE Hypacrosaurus altispinus
0.73 Lambeosaurus laticaudus
0.78 Velafrons coahuilensis
0.87 Nipponosaurus sachalinensis
0.60
Amurosaurus riabinini
0.88
EUHADROASAURIA 0.58 Lambeosaurus magnicristatus
1 Lambeosaurus lambei
0.79
0.99 1
0.81 Parasaurolophus tubicen
SAUROLOPHINAE 0.66
0.58 Sabinas OTU
1
1 Saurolophus osborni
0.70 Saurolophus angustirostris
Wulagasaurus dongi
0.84
0.99
Wahweap OTU
0.69 Kritosaurus navajovius
UTEP Big Bend OTU
0.99 Secernosaurus koerneri
0.4 Gryposaurus latidens
0.98
Figure 2.11. Consensus tree resulting from the Bayesian analysis with equal rates of character change
showing the phylogenetic relationships of 51 hadrosaurs and the position of Barsboldia sicinskii and
Nipponosaurus sachalinensis. At each node, the number indicates a posterior probability.
102
Iguanodon spp.
Equijubus normani
1
1
Protohadros byrdi
1
Tanius sinensis
0.51
0.98
0.98
UTEP Big Bend OTU
Secernosaurus koerneri
0.89 0.83
Salitral Moreno OTU
0.83 0.99 Gryposaurus latidens
HADROSAURIDAE Gryposaurus monumentensis
Sabinas OTU
0.97
0.97 1 Saurolophus osborni
EUHADROSAURIA 0.62 Saurolophus angustirostris
0.66
0.93 0.85 Tsintaosaurus spinorhinus
0.98 Olorotitan ararhensis
0.85 0.99
1 0.98 Parasaurolophus cyrtocristatus
0.68 Corythosaurus intermedius
0.62
Lambeosaurus lambei
0.71
Wulagasaurus dongi
0.81 Brachylophosaurus canadensis
0.4 0.77 1 Maiasaura peeblesorum
Wahweap OTU
Figure 2.12. Consensus tree of the Bayesian analysis with variable rates of character change, showing
the phylogenetic relationships of 49 hadrosaur taxa, excluding Barsboldia sicinskii and Nipponosaurus
sachalinensis. At each node, the number indicates a posterior probability. Only the names of taxa with
complete skulls are accompanied by a photograph.
103
Iguanodon spp.
Equijubus normani
.99
1
Protohadros byrdi
1 Tanius sinensis
0.52 Bactrosaurus johnsoni
0.98 Gilmoreosaurus mongoliensis
Lophorhothon atopus
0.98
UTEP Big Bend OTU
0.84
0.82 0.99 Gryposaurus latidens
HADROSAURIDAE Gryposaurus monumentensis
Sabinas OTU
0.98 1 Saurolophus osborni
0.98
0.68 Saurolophus angustirostris
EUHADROSAURIA Kerberosaurus manakini
0.72 Shantungosaurus giganteus
0.54 Aralosaurus tuberiferus
0.70 0.89 Tsintaosaurus spinorhinus
0.88
Nipponosaurus sachalinensis
0.73 Charonosaurus jiayinensis
0.98
0.98 Parasaurolophus tubicen
0.87
0.69 Corythosaurus intermedius
Lambeosaurus lambei
0.74
0.53
Wulagasaurus dongi
0.4
0.80 Brachylophosaurus canadensis
0.79 1 Maiasaura peeblesorum
Wahweap OTU
Figure 2.13. Consensus tree derived from the Bayesian analysis with variable rates of character change
showing the phylogenetic relationships of 51 hadrosaur taxa, with the position of Barsboldia sicinskii
and Nipponosaurus sachalinensis. The numbers at the nodes represent posterior probabilities.
104
Iguanodon spp.
Equijubus normani
Eolambia caroljonesa
HADROSAURIA
Protohadros byrdi
Tanius sinensis
A
Lophorhothon atopus
Claosaurus agilis
Hadrosaurus foulkii
SAUROLOPHINAE
Saurolophus osborni
“saurolophs”
G
I
Sabinas OTU
Wulagasaurus dongi
Wahweap OTU
HADROSAURIDAE
“maiasaurs”
B
105
“wulagasaurs”
UTEP OTU
“kritosaurs”
Salitral Moreno OTU

H
EUHADROSAURIA
C
LAMBEOSAURINAE
“tsintaosaurs” “hypacrosaurs”
Lambeosaurus lambei
“amurosaurs”
42 teeth in the dentary. H, reversal to 31-41 dentary teeth. I, deep and invaginated caudal region of the circumnarial fossa.
E
Parasaur olophus cyrtocristatus
“corythosaurs” “parasaurolophs”
occlusal plane. E, ventral offset of the predentary-premaxilla oral contact. F, elevation of the rostrodorsal region of the maxilla. G, more than
fossa (rectangle) and presence of crest (arrow). C, 31 or more tooth positions in the dentary. D, three teeth labiolingually exposed on the dentary
Figure 2.14. Ancestral state reconstructions of various cranial characters. A, expansion of the premaxilla. B, caudal extension of the circumnarial
Iguanodon spp.
Equijubus normani
Protohadros byrdi
Tanius sinensis
A
HADROSAURIA
Lophorhothon atopus
Claosaurus agilis
Hadrosaurus foulkii
SAUROLOPHINAE
Saurolophus osborni
“saurolophs”
B
Sabinas OTU
Wulagasaurus dongi
Wahweap OTU
HADROSAURIDAE
“maiasaurs”
106
C
UTEP OTU
“kritosaurs”
Salitral Moreno OTU

EUHADROSAURIA
oriented acromion process (rectangle) of the scapula in lambeosaurine hadrosaurids.

LAMBEOSAURINAE
Lambeosaurus lambei
D
Figure 2.15. Ancestral state reconstructions of various pectoral and forelimb characters. A, primitive condition in non-hadrosaurid
iguanodontoidean of the dorsally oriented acromion process (rectangle) of the scapula. B, horizontal acromion process (rectangle).

C, elongation, lateroventral projection and pronounced angulation of the deltopectoral crest of the humerus. D, curved, rostrodorsally
Iguanodon spp.
Equijubus normani
HADROSAURIA
Protohadros byrdi
Tanius sinensis
A
Lophorhothon atopus
Claosaurus agilis
Hadrosaurus foulkii
SAUROLOPHINAE
E
Saurolophus osborni
“saurolophs”
B
Sabinas OTU
Wulagasaurus dongi
Wahweap OTU
“maiasaurs”
107
UTEP OTU
“kritosaurs”
EUHADROSAURIA
Salitral Moreno OTU

F
C
LAMBEOSAURINAE
D
HADROSAURIDAE
the ischial peduncle. G, craniocaudal shortening and lateroventral projection of the supraacetabular process.
Lambeosaurus lambei
G
Figure 2.16. Ancestral state reconstructions of iliac characters. A, brevis shelf in non-hadrosaurid iguanodontoideans (ventral view). B, absence
of brevis shelf (ventral view). C and D, brevis shelf resulting from medioventral rotation of the postacetabular process (ventral view). E, ischial
peduncle of the ilium composed of two protrusions. F, rostral shift of the apex of the supraacetabular process relative to the caudal protrusion of
Iguanodon spp.
Equijubus normani
Protohadros byrdi
A
Tanius sinensis
HADROSAURIA
Lophorhothon atopus
Claosaurus agilis
Hadrosaurus foulkii
SAUROLOPHINAE
Saurolophus osborni
B
Sabinas OTU
Wulagasaurus dongi
Wahweap OTU
D
HADROSAURIDAE Brachylophosaurus canadensis
“maiasaurs”
108
C
UTEP OTU
“kritosaurs”
EUHADROSAURIA
Salitral Moreno OTU

LAMBEOSAURINAE
Lambeosaurus lambei
E
D, saurolophine ancestral condition. E, lambeosaurine condition. F, geometry of he prepubic process in Corythosaurus and
F
Lambeosaurus derived from the ancestral lambeosaurine condition by elongation or dorsoventral narrowing of the proximal constriction.

Figure 2.17. Ancestral state reconstructions of lateral profile of the blade of the prepubic process of the pubis. A, primitive condition in non-
hadrosaurid iguanodontoideans. B, morphology of the Edmontosaurus prepubic process. C, morphology of the “maiasaur” prepubic process.
Iguanodon spp.
Equijubus normani
A
Protohadros byrdi
HADROSAURIA
C
Tanius sinensis
B
Lophorhothon atopus
Claosaurus agilis
Hadrosaurus foulkii
E
Saurolophus osborni
“saurolophs”
Sabinas OTU
Wulagasaurus dongi
F
Wahweap OTU
SAUROLOPHINAE
“maiasaurs”
HADROSAURIDAE
D
109
UTEP OTU
“kritosaurs”
Salitral Moreno OTU

EUHADROSAURIA
LAMBEOSAURINAE
Lambeosaurus lambei
H
I
B, slight curvature of the iliac peduncle in non-hadrosaurid iguanodontoideans. C, ischial “foot” in non-hadrosaurid iguanodontoideans. D, horizontal
G, abence of ischial “foot”. H, caudodorsally extended and recurved iliac peduncle of lambeosaurines. I, ischial “foot” of lambeosaurine hadrosaurids.
orientation of the pubic peduncle in saurolophines. E, slight cranioventral orientation of the pubic peduncle. F, absence of curvature in the iliac peduncle.
Figure 2.18. Ancestral state reconstructions of ischial characters. A, cranioventral orientation of the pubic peduncle in non-hadrosaurid iguanodontoideans.
Table 2.1. List of the outgroup taxa used in the phylogenetic analyses of hadrosaurid dinosaurs, showing the genera and species
considered in the present study. In the column of the original names, when two citations are present the first one refers to the specific
name and the second to the generic name.
Outgroup taxa Outgroup taxa Occurrence Age

(original name) (this study)
Bactrosaurus johnsoni Gilmore, Bactrosaurus johnsoni Iren Dabasu Fm., Inner late Campanian-?early
1933 Mongolia, China Maastrichtian (Van
Itterbeeck et al., 2005)
Claosaurus agilis Marsh, 1872; Claosaurus agilis Niobrara Chalk Fm. (Smoky Hill late Coniacian (Everhart and
1890 Chalk Member), Kansas, USA Ewell, 2006)
Eolambia caroljonesa Kirkland, Eolambia caroljonesa Cedar Mountain Fm., Utah, late Albian-Early
1998 USA Cenomanian (Gates and
Lund, 2006)
Equijubus normani Hai-lu You et Equijubus normani Xinnminbao Group (middle late Barremian-Aptian (Tang
al., 2003 Grey Unit), Gongpoquan Basin, et al., 2001)
Mazongshan region, Gansu
Province, China
Gilmoreosaurus mongoliensis Gilmoreosaurus mongoliensis Iren Dabasu Fm., Inner late Campanian-?early
Gilmore, 1933; Brett-Surman, 1979 Mongolia, China Maastrichtian (Van
Itterbeeck et al., 2005)
Iguanodon Mantell, 1825 Iguanodon Numerous formations Valanginian-Albian
throughout western Europe, (Norman, 2004)
Mongolia and USA
Lophorhothon atopus Langston, Lophorhothon atopus Mooresville Chalk Fm. (lower Santonian-Campanian
1960 unnamed member), Alabama, boundary (Lamb, pers.
USA comm.)
Ouranosaurus nigeriensis Taquet, Ouranosaurus nigeriensis Upper Elrhaz Fm., Niger late Aptian (Paul, 2008)
1976
110
Table 2.1—continued.
Outgroup taxa Outgroup taxa Occurrence Age

Probactrosaurus Rozhdestvensky, 1966 Probactrosaurus Dashiguo Fm., Mongolia Barremian-Albian (Norman,
2002)
Protohadros byrdi Head, 1998 Protohadros byrdi Woodbine Fm., Texas, USA middle Cenomanian (Head,
1998)
Tanius sinensis Wiman, 1929 Tanius sinensis Jiangjun Fm. (Hu et al., 2001), early Campanian (Hong and
Wangshi Group, Shandong Miyata, 1999)
Province, China
Telmatosaurus transsylvanicus Nopcsa, Telmatosaurus transsylvanicus ?Sampetru Fm., near Sampetru early Maastrichtian (Dalla
1900; 1903 (= Szentpeterfalva), Sibisel River Vecchia, pers. comm.)
valley, Romania
111
Table 2.2. List of the ingroup hadrosaurid taxa used in the phylogenetic analyses, showing the hadrosaurid genera and species
considered in the present study. In the column of the original names, when two citations are present the first one refers to the specific
name and the second to the generic name.
Ingroup taxa Ingroup taxa Occurrence Age

Amurosaurus riabinini Bolotsky Amurosaurus riabinini Udurchukan Fm., Amur middle-late Maastrichtian
and Kurzanov, 1991 Region, Russia (Godefroit et al., 2004a)
Anasazisaurus horneri Hunt and Kritosaurus navajovius Lower Kirtland Fm. late Campanian (Gates and Lund,
Lucas, 1993 (Farmington Member), 2006)
Kimbeto Arroyo, New
Mexico, USA
Anatotitan copei Chapman and Edmontosaurus annectens Hell Creek Fm., Montana late Maastrichtian (Horner et al.,
Brett-Surman, 1990 and South Dakota, USA; 2004)
Lance Fm., Wyoming,
USA
Aralosaurus tuberiferus Aralosaurus tuberiferus Bostobe (Bostobynskaya) late Santonian-early Campanian
Rozhdestvensky, 1968 Fm., Shak Shak locality, (Kordikova et al., 2001;
northeastern Aral Sea Averianov, 2007)
region, southwestern
Kazakhstan
Barsboldia sicinskii Maryanska Barsboldia sicinskii Nemegt Fm., northern middle Maastrichtian (Wilson,
and Osmolska, 1981 Sayr, Nemegt, Mongolia 2005)
Brachylophosaurus canadensis Brachylophosaurus canadensis Judith River Fm., late Campanian (LaRock, 2001)
Sternberg, 1953 Montana, USA; Oldman
Fm., Alberta, Canada
Brachylophosaurus goodwini Brachylophosaurus canadensis Judith River Fm., Hill late Campanian (LaRock, 2001)
Horner, 1988 County, Montana, USA
112

Charonosaurus jiayinensis Charonosaurus jiayinensis Yuliangze Fm., near late Maastrichtian (Godefroit et
Godefroit et al., 2000 Jiayin, Heilongjang al., 2000)
Province, northeastern
China
Corythosaurus casuarius Brown, Corythosaurus casuarius Dinosaur Park Fm. (lower late Campanian (Evans, 2007)
1914 third), Alberta, Canada
Corythosaurus intermedius Parks, Corythosaurus intermedius Dinosaur Park Fm. late Campanian (Evans, 2007)
1923 (middle third), Alberta,
Canada
Edmontosaurus annectens Marsh, Edmontosaurus annectens Hell Creek Fm., Montana late Maastrichtian (Horner et al.,
1892 and South Dakota, USA; 2004)
Lance Fm., South Dakota
and Wyoming, USA;
Laramie Fm., Colorado,
USA; Scollard Fm.,
Alberta, Canada
Edmontosaurus saskatchewanensis Edmontosaurus annectens Frenchman Fm., late Maastrichtian (Horner et al.,
Sternberg, 1926 Saskatchewan, Canada 2004)
Edmontosaurus regalis Lambe, Edmontosaurus regalis Hell Creek Fm., Montana, early-late Maastrichtian (Horner
1917 North and South Dakota, et al., 2004)
USA; Lance Fm.,
Wyoming, USA; Laramie
Fm., Colorado, USA;
Scollard Fm., St. Mary
River Fm. and Horeshoe
Canyon Fm., Canada
113

Gryposaurus notabilis Lambe, Gryposaurus notabilis Dinosaur Park Fm. (lower late Campanian (Evans, 2007)
1914a third), Alberta, Canada
Gryposaurus incurvimanus Parks, Gryposaurus notabilis Dinosaur Park Fm. (lower late Campanian (Gates, 2007)
Gryposaurus latidens Horner, Gryposaurus latidens Two Medicine Fm. (lower late Santonian-early Campanian
1992 section), near Two (Horner, 1992)
Medicine River, Pondera
County, Montana, USA
Gryposaurus monumentensis Gates Gryposaurus monumentensis Kaiparowits Fm., southern late Campanian (Gates and
and Sampson, 2007 Utah, USA Sampson, 2007)
Hadrosaurus foulkii Leidy, 1858 Hadrosaurus foulkii Woodbury Fm., Campanian (Prieto-Marquez et
Haddonfield, New Jersey, al., 2006a)
USA
Hypacrosaurus altispinus Brown, Hypacrosaurus altispinus Horseshoe Canyon Fm., early Maastrichtian (Evans, 2007)
1912 Alberta, Canada
Hypacrosaurus stebingeri Horner Hypacrosaurus stebingeri Two Medicine Fm. (upper middle-late Campanian (Horner
and Currie, 1994 section), Montana, USA; et al., 2004)
Oldman Fm., Alberta,
Canada
Jaxartosaurus aralensis Riabinin, Jaxartosaurus aralensis Syuksyuk Fm., Santonian (Averianov and
1939 Kazakhstan Nessov, 1995)
Kerberosaurus manakini Bolotsky Kerberosaurus manakini Tsagayan Fm. (upper ?late Maastrichtian (Bolotsky and
and Godefroit, 2004 section), Nagornaia street Godefroit, 2004)
in Blagoveschensk city,
Amur Region, Russia
114

Koutalisaurus kohlerorum Prieto- Pararhabdodon isonensis Tremp Fm., Catalunya, late Maastrichtian (Prieto-
Marquez et al., 2006b northeastern Spain Marquez et al., 2006b)
Kritosaurus navajovius Brown, Kritosaurus navajovius Kirtland Fm. (De-Na-Zin late Campanian (Sullivan, 1999)
1910 Member), San Juan Basin,
New Mexico, USA
Kritosaurus australis Bonaparte et Secernosaurus koerneri Los Alamitos Fm., Rio late Campanian-early
al., 1984 Negro Province, Maastrichtian (Apesteguia, 2005)
northeastern Patagonia,
Argentina
Kritosaurus sp. (referral by Sabinas OTU Olmos Fm. (lower latest Campanian (Krikland et al.,
Kirkland et al., 2006) section), near Sabinas, 2006)
Coahuila, Mexico
Lambeosaurus lambei Parks, 1923 Lambeosaurus lambei Dinosaur Park Fm. late Campanian (Evans, 2007)
(middle-upper sections),
Alberta, Canada
Lambeosaurus magnicristatus Lambeosaurus magnicristatus Dinosaur Park Fm. (upper late Campanian (Evans, 2007)
Sternberg, 1935 section), Alberta, Canada
Lambeosaurus laticaudus Morris, Lambeosaurus laticaudus El Gallo Fm., north of Campanian (Horner et al., 2004)
1981 Arroyo del Rosario, El
Rosario, northern Baja
California, Mexico
Maiasaura peeblesorum Horner Maiasaura peeblesorum Two Medicine Fm., middle-late Campanian (Horner
and Makela, 1979 Montana, USA et al., 2004)
Naashoibitosaurus normani Lucas Kritosaurus navajovius Kirtland Fm. (Naashoibito late Campanian (Gates and Lund,
and Hunt, 1993 Member), New Mexico, 2006)
USA
115

Nipponosaurus sachalinensis Nipponosaurus sachalinensis Upper Yezo Group, late Santonian-early Campanian
Nagao, 1936 Sinegorsk, southern (Suzuki et al., 2004)
Sakhaklin, Russia
Olorotitan ararhensis Godefroit et Olorotitan ararhensis Kundur, Amur Region, middle-late Maastrichtian
al., 2003 Far Eastern Russia (Godefroit et al., 2003)
Pararhabdodon isonensis Pararhabdodon isonensis Tremp Fm., Catalunya, late Maastrichtian (Prieto-
Casanovas-Caldellas et al., 1993 northeastern Spain Marquez et al., 2006b)
Parasaurolophus walkeri Parks, Parasaurolophus walkeri Dinosaur Park Fm. (lower late Campanian (Evans, 2007)
Parasaurolophus tubicen Wiman, Parasaurolophus tubicen Kirtland Fm. (lower late Campanian (Horner et al.,
1931 section), New Mexico, 2004)
USA
Parasaurolophus cyrtocristatus Parasaurolophus cyrtocristatus Kaiparowits Fm., Utah, late Campanian (Horner et al.,
Ostrom, 1961 USA; Fruitland Fm., New 2004)
Mexico, USA
Prosaurolophus maximus Brown, Prosaurolophus maximus Dinosaur Park Fm. (upper late Campanian (Evans, 2007)
1916 half), Alberta, Canada
Prosaurolophus blackfeetensis Prosaurolophus maximus Two Medicine Fm. (upper late Campanian (Horner, 1992)
Horner, 1992 section), Landslide Butte,
Glacier County, Montana,
USA
Sahaliyania elunchunorum Sahaliyania elunchunorum Yuliangze Fm., near ?late Maastrichtian (Godefroit et
Godefroit et al., 2008 Wulaga, Heilongjang al., 2008)
Province, northeastern
China
116

Saurolophus osborni Brown, 1913 Saurolophus osborni Horseshoe Canyon Fm., early Maastrichtian (Horner et al.,
Alberta, Canada 2004)
Saurolophus angustirostris Saurolophus angustirostris Nemegt Fm. (white beds middle Maastrichtian (Wilson, 2005)
Rozhdestvensky, 1952 of Hermiin Tsav), N. Sayr,
Nemegt, Mongolia
Secernosaurus koerneri Brett-Surman, Secernosaurus koerneri Upper Bajo Barreal Fm., late Campanian-early Maastrichtian
1979 Lago Colhue Huapi, Rio (Salinas, pers. comm.)
Negro, Argentina
Shantungosaurus giganteus Hu, 1972 Shantungosaurus giganteus Xingezhuang Fm. (Hu et early Campanian (Hong and Miyata,
al., 2001), Wangshi 1999)
Group, Shandong, China
Tsintaosaurus spinorhinus Young, 1958 Tsintaosaurus spinorhinus Jingangkou Fm. (Hu et al., early Campanian (Hong and Miyata,
2001), Wangshi Group, 1999)
Shandong Province, China
Velafrons coahuilensis Gates et al., Velafrons coahuilensis Cerro del Pueblo Fm., late Campanian (Gates et al., 2007)
2007 Coahuila, Mexico
Wulagasaurus dongi Godefroit et al., Wulagasaurus dongi Yuliangze Fm., near ?late Maastrichtian (Godefroit et al.,
2008 Wulaga, Heilongjang 2008)
Province, NE China
Unnamed hadrosaurid from Big Bend Big Bend UTEP OTU Aguja Fm. (upper shale Campanian-?early Maastrichtian
National Park (Davies, 1983) member), Texas, USA (Wagner, 2001)
Unnamed hadrosaurid from the Wahweap OTU Wahweap Fm (upper late Campanian (Gates, 2007)
Wahweap Fm. (Gates, pers. comm.) section)
Unnamed hadrosaurid from Salitral Salitral Moreno OTU Allen Fm., Rio Negro late Campanian-early Maastrichtian
Moreno (Powell, 1987) Province, Argentina (Salgado et al., 2007)
117
Table 2.3. Explanation of the identifiers used to characterize the various models of
clusters that can be used to classify multidimensional data, such as those derived from the
analyses of planar shapes using geodesic paths (GDA; see Fig. 2.1B). Table modified
from Fraley and Raftery (2003).
Identifier Cluster Cluster Cluster Cluster

Distribution Volume Shape Orientation
EII Spherical Equal Equal Not applicable
VII Spherical Variable Equal Not applicable
EEI Diagonal Equal Equal Coordinate axes
VEI Diagonal Variable Equal Coordinate axes
EVI Diagonal Equal Variable Coordinate axes
VVI Diagonal Variable Variable Coordinate axes
EEE Ellipsoidal Equal Equal Equal
VVV Ellipsoidal Variable Variable Variable
EEV Ellipsoidal Equal Equal Variable
VEV Ellipsoidal Variable Equal Variable
118
Table 2.4. Tree-island profile of the heuristic search in the unweighted parsimony
analysis of hadrosaurid dinosaurs, for the data set excluding Nipponosaurus sachalinensis
and Barsboldia sicinskii.
First Island Size First Tree Last Tree Score First Replicate Times Hit
1 15 1 15 971 1 3476
2 1 16 16 971 6 1
3 1 17 17 971 10 1
4 1 18 18 971 28 1
5 1 19 19 971 61 1
6 1 20 20 971 63 1
7 1 21 21 971 72 1
8 1 22 22 971 75 1
9 1 23 23 971 84 1
10 1 24 24 971 139 1
11 1 25 25 971 157 1
12 1 26 26 971 166 1
13 1 27 27 971 206 1
14 1 28 28 971 208 1
15 1 29 29 971 213 1
16 1 30 30 971 224 1
17 1 31 31 971 232 1
18 1 32 32 971 269 1
19 1 33 33 971 277 1
20 1 34 34 971 278 1
21 1 35 35 971 282 1
22 1 36 36 971 302 1
23 1 37 37 971 310 1
24 1 38 38 971 342 1
25 1 39 39 971 381 1
26 1 40 40 971 382 1
27 1 41 41 971 389 228
119
Table 2.5. Tree-island profile of the heuristic search from the unweighted parsimony
analysis of hadrosaurid dinosaurs, for the data set including Nipponosaurus sachalinensis
and Barsboldia sicinskii.
1 1372 1 1372 979 1 1639
2 1 1373 1373 979 3 1
3 1 1374 1374 979 35 1
4 1 1375 1375 979 55 1
5 1 1376 1376 979 84 1
6 1 1377 1377 979 142 1
7 1 1378 1378 979 203 1
8 1 1379 1379 979 207 1
9 1 1380 1380 979 236 1
10 1 1381 1381 979 258 1
11 1 1382 1382 979 293 1
12 1 1383 1383 979 310 1
13 1 1384 1384 979 361 1
14 1 1385 1385 979 381 1
15 1 1386 1386 979 394 1
16 1 1387 1387 979 426 1
17 1 1388 1388 979 517 1
18 1 1389 1389 979 523 1
19 1 1390 1390 979 571 1
20 1 1391 1391 979 573 1
21 1 1392 1392 979 591 1
22 1 1393 1393 979 594 1
23 1 1394 1394 979 902 1
24 1 1395 1395 979 1398 1
25 1 1396 1396 979 1404 1
26 1 1397 1397 979 1418 1
27 1 1398 1398 979 2172 1
28 1 1399 1399 979 2214 1
29 1 1400 1400 979 2401 1
30 1 1401 1401 979 2426 1
31 1 1402 1402 979 2441 1
32 1 1403 1403 979 2453 1
33 1 1404 1404 979 2609 1
34 1 1405 1405 979 2610 1
35 1 1406 1406 979 2724 1
36 1 1407 1407 979 2766 1
37 1 1408 1408 979 2980 1
38 1 1409 1409 979 3190 1
120
39 1 1410 1410 979 3260 1
40 1 1411 1411 979 3266 1
41 1 1412 1412 979 3449 1
42 1 1413 1413 979 3486 1
43 1 1414 1414 979 3663 1
44 1 1415 1415 979 3949 1
45 1 1416 1416 979 4030 1
46 1 1417 1417 979 4239 1
47 1 1418 1418 979 4244 1
48 1 1419 1419 979 4326 1
49 1 1420 1420 979 4402 1
50 1 1421 1421 979 4596 1
51 1 1422 1422 979 4649 1
121
CHAPTER 3
HISTORICAL BIOGEOGRAPHY OF HADROSAURID DINOSAURS
Abstract
Hadrosaurids were the most diverse and abundant herbivorous dinosaurs during the late
Cretaceous of Europe, Asia, the Americas and Antarctica. However, their center of origin and
subsequent biogeographical history has remained contentious. Previous analyses have omitted
entire continental areas and have not employed modern quantitative methods of historical
biogeography. Here, I report results of a biogeographic analysis seeking to establish the ancestral
area of Hadrosauridae, testing whether they originated in Asia, North or South America. In
addition I used the same method to track their diversification in relation to intercontinental
connections throughout their tenure.
Ancestral areas were reconstructed on a weighted parsimony phylogeny derived from the
latest and most comprehensive phylogenetic analysis of Hadrosauridae (Chapter 2, this
dissertation). Fitch parsimony and the Dispersal-Vicariance (DIVA) method were implemented
to reconstruct ancestral areas for all clades of Hadrosauria.
The results show that the genesis of Hadrosauridae occurred in eastern North America
during the late Santonian. Soon after their origin, hadrosaurids dispersed to Asia, the ancestral
area for the major Saurolophinae-Lambeosaurinae divergence. Lambeosaurines and
saurolophines dispersed to North America and underwent major cladogenesis during the late
Campanian. Within Saurolophinae, “maiasaurs” and “saurolophs” returned to Asia, while
“kritosaurs” colonized South America by the late Campanian. Within Lambeosaurinae,
“tsintaosaurs” dispersed to Europe no later than the late Campanian. The North American
radiations of “hypacrosaurs”, “parasaurolophs” and “corythosaurs” may have represented
independent dispersal events from the Asian continent or, alternatively, were part of a single
dispersal in the late Campanian with posterior occupations of Asia by a few species from those
clades. Vicariant events may have occurred following several of the inferred hadrosaurid
dispersals. Thus, both vicariance and dispersal may have been instrumental in shaping the
recorded distribution of hadrosaurids.
122
Introduction
Hadrosaurids were the most common and diverse herbivores from the Late Cretaceous of
Europe, Asia, the Americas and Antarctica (Weishampel, 1990; Forster, 1997; Case et al., 2000).
Their fossil record is the most complete among dinosaurs and includes fully articulated skeletons
(Lull and Wright, 1942), thousands of disarticulated bones from multi-individual assemblages
(Dodson, 1971; Varricchio and Horner, 1993), growth series (Horner et al., 2000), soft tissue
impressions (Osborn, 1912; Murphy et al., 2007), eggs, nests and neonates (Horner, 1979; 1982;
Horner and Currie, 1994), and trackways (Currie et al., 1991).
Despite the abundance and excellent preservation of hadrosaurid anatomical data, our
understanding of hadrosaurid biogeography is poor. This is because previous reconstructions of
hadrosaurid biogeography have been non-global, based on phylogenies that used a limited
taxonomic sampling. In addition, no study has attempted to elucidate the birthplace and
biogeographic history of these animals using modern quantitative methods of historical
biogeography and a phylogeny based on a complete taxonomic sampling of the clade. Instead,
previous reconstructions of hadrosaurid biogeography consisted on literal readings of recorded
distributions in the fossil record, typically leading to narratives of dispersal or vicariant scenarios
with centers of origin for the major clades (Brett-Surman, 1979; Weishampel and Weishampel,
1983; Head, 1998; 2001; Casanovas et al., 1999; Wagner, 2001; Godefroit et al., 1998; 2003;
2004b). Such approaches, while useful for establishing working hypotheses, lack the rigor and
testability provided by current methods of inference of historical biogeography (Morrone and
Crisci, 1995; Lieberman, 2000; 2003). Likewise, these approaches suffered from the same
shortcoming than pre-cladistic phylogenetic hypothesis, i.e., that ancestral state reconstruction
were based on opinion rather than using a explicit and testable methodology. Although few
authors have recently implemented analytical methods for inferring biogeographical histories for
major dinosaurian clades (Chapman, 1997; Upchurch et al., 2002; Holtz et al., 2004), they have
not specifically addressed hadrosaurids.
Traditionally, Asia has been considered the most likely area of origin for hadrosaurids
(Wiman, 1929; Rozhdestvensky, 1967). This is because primitive taxa preceding the radiation
were first discovered in China and Mongolia (Gilmoreosaurus mongoliensis, Gilmore, 1933;
Bactrosaurus johnsoni, Godefroit et al., 1998). However, the discovery of primitive hadrosaurs
123
in Asia (Probactrosaurus gobiensis, Norman, 2002; Equijubus normani, Hai-lu You et al.,
2003), North America (Eolambia caroljonesa, Kirkland, 1998; Protohadros byrdi, Head, 1998),
and Europe (Telmatosaurus transsylvanicus, Weishampel et al., 1993; Pararhabdodon isonensis,
Prieto-Marquez et al., 2006b), spanning a broad stratigraphic range, has complicated this issue
by offering alternative possible ancestral areas. These alternative scenarios include their genesis
in Europe (Weishampel et al., 1993; Company et al., 1998; Casanovas et al., 1999), North
America (Kirkland, 1998; Head, 1998; 2001; Horner et al., 2004), and South America (Brett-
Surman, 1979; Powell, 1987). Biogeographical reconstructions have involved dispersal and
vicariance as the primary factors affecting hadrosaurid distribution (Brett-Surman, 1979; 1989;
Weishampel and Weishampel, 1983; Russell, 1993; Head, 1998; Casanovas et al., 1999;
Godefroit et al., 2003; 2004a).
My goals in this study were two-fold. First to test the previous hypotheses regarding the
center of origin of hadrosaurids, specifically whether they originated in Asia, North or South
America. Second, establish how hadrosaurids achieved their nearly cosmopolitan distribution.
Unlike previous studies, my analysis was the first one to use a phylogeny based on a complete
taxonomic sampling of hadrosaurids and explicit quantitative methods for the inference of
ancestral areas.
Review of Previous Studies on Hadrosaurid Biogeography
The Pre-Cladistic Era

The first ideas about hadrosaurid biogeography were published by Lull and Wright
(1942) in the 1940s, nearly a century after their initial discovery. These authors suggested that
the “flat-headed” (saurolophine) hadrosaurids of western North America might have dispersed
from the eastern side of the continent, following a major regression of the western Interior Sea
during the late Cretaceous. Whether they viewed the east as the ancestral area for the
suarolophines is unclear and explanations for forms on other continents known at the time were
not considered.
Morris (1973) addressed the biogeography of a number of North American hadrosaurids
during the Campanian and Maastrichtian. He proposed that Lambeosaurus, Prosaurolophus and
Saurolophus had dispersed from southern Canada to Baja California along the Pacific coast.
124
Morris also suggested that Edmontosaurus and Gryposaurus had dispersed from southern
Alberta to as far south as New Mexico, along a routed that extended east to the Rockies. In
contrast, Corythosaurus and Hypacrosaurus would have been restricted to the northern Rocky
Mountain region of North America.
Brett-Surman (1975; 1979) presented the first global hypothesis of hadrosaurid
biogeography. He regarded Asia as the most likely region that harbored the ancestral area of
hadrosaurids during pre-Santonian times. This was because various outgroup taxa to
hadrosaurids, such as Probactrosaurus, Bactrosaurus and Gilmoreosaurus, had been found in
Asia. He then considered possible dispersal routes to Europe and North America through the
Turgai Strait. From Europe, he hypothesized connections to Africa, via the Iberian Peninsula,
and to South America through the Amazon basin (Brett-Surman, 1976). An alternative route to
South America from Europe, later suggested by Brett-Surman (1979), was via the east coast of
North America across the Caribbean arc by means of island hopping. Dispersals to North
America from Asia would have been possible through the Bering Strait or through Europe.
However, Brett-Surman (1979) also entertained the alternative hypothesis that
hadrosaurids might have originated in South America. He suggested this possibility after the
study of Secernosaurus koerneri, a taxon from the late Campanian-early Maastrichtian of
southern Argentina (see Chapter 4 of this dissertation). He regarded it as a primitive hadrosaurid
that had survived into the Maastrichtian of Argentina. From South America, dispersals to
Laurasia would have occurred via Africa and the Iberian Peninsula or, alternatively, through
North America via Panama or the Caribbean arc. Although Brett-Surman considered the
possibility that vicariance might have also been involved in the distribution of hadrosaurids in
Laurasia and Gondwana, he tentatively preferred a dispersal model because of the absence of a
hadrosaurid fossil record in Oceania and Africa. For saurolophines, Brett-Surman (1989) posited
an Asian or North American origin. He based this hypothesis on the presence in those continents
of taxa that he regarded as relatively primitive within the clade, such as Hadrosaurus,
Brachylophosaurus, Kritosaurus in North America and Aralosaurus in Asia.
Casamiquela (1964), Weishampel and Weishampel (1983) and Bonaparte et al. (1984)
explained the presence of South American hadrosaurids (Kritosaurus australis) as dispersals
from North America. Weishampel and Weishampel (1983) found support for this hypothesis in
the finding of a femur with hadrosaur affinities in the Cenomanian of Honduras (Horne, 1994).
125
Weishampel and Weishampel (1983) pointed out that the Nicoya Complex, a volcanic chain of
Upper Cretaceous age that extended from Central America to Ecuador, might have served as a
land connection between the two Americas. Because these authors regarded the Argentinian
species Kritosaurus australis as a primitive hadrosaurid, they proposed that the dispersal took
place early in the evolutionary history of the clade. Likewise, Bonaparte et al. (1984) posited that
these dispersals were part of a bidirectional faunal exchange between North and South America,
including many other groups of vertebrates such as titanosaur sauropods, snakes, birds and
mammals. This faunal exchange would have occurred during the Campanian-early
Maastrichtian. In their view, the dispersals most probably took place through Central America,
rather than through the Antilles, because the North American vertebrate fauna most closely
related to that in South America had been found in the western region of Central America.
The Cladistic Era
Soon after cladistic analyses were applied to the reconstruction of hadrosaur phylogenetic
relationships, paleontologists began to compare the topology of their cladograms with the
geographical and stratigraphical distributions of these animals to infer biogreographic patterns of
dispersal and vicariance. This resulted in more elaborate and testable biogographical hypotheses.
to explain the widespread distribution of hadrosaurids during the Campanian and Maastrichtian.
Specifically, Milner and Norman (1984) proposed that hadrosaurids originated in eastern
Asia during the Bajocian from iguanodontian ancestors through vicariance. Support for this
hypothesis came from the finding of the most primitive hadrosaur outgroups to hadrosaurids in
central Asia. These authors further suggested that the splitting of hadrosaurids into saurolophines
and lambeosaurines took place in Asia through vicariance. This was followed by dispersals
during the later stages of the Cretaceous. Thus, the occurrence in the latest cretaceous of
Saurolophus and lambeosaurines in eastern Asia and western North America was explained by
dispersal events eastwards from Asia through the Bering Strait. Later, Russell (1993) would
support the existence of a faunal exchange through the Bering corridor not only for hadrosaurids,
but also for many other dinosaur clades.
In one of the few studies of the European hadrosaur fossil record, Weishampel et al.
(1991; 1993) regarded Telmatosaurus transsylvanicus, from the Maastrichtian of Romania, as a
primitive hadrosaurid that lived in Europe as a relict species. This species would have remained
isolated in the large archipelago that formed southern Europe during most of the late Cretaceous.
126
Head (1998; 2001) also supported the survival of primitive hadrosaurids in South
America (Secernosaurus koerneri) and Europe (Pararhabdodon isonensis) into the latest
Cretaceous. In addition, he proposed that hadrosaurids might have originated in North America
during the Cenomanian (Head, 1998). From there, they would have dispersed to Asia by the
Turonian and to South America during the Campanian-Maastrichtian. He suggested also that
short-term regressions of the western Interior Seaway in North America during the Albian and
Cenomanian might have opened areas for potential dispersals from the east to the west in North
America and to Asia through the Bering Strait. He contended that this origin was as likely as the
Asian origin that was the paradigm at the time. In support of his North American origination
hypothesis he alluded to the presence of iguanodontians in the early Cretaceous of this continent
and the finding of what he considered the most basal hadrosaurid known at the time,
Protohadros byrdi from the Cenomanian of Texas. He favored a dispersal model rather than
vicariance because no conclusive hadrosaurid remains had been found in strata old enough to
allow congruence between the sequence of continental break-up and the hadrosaurid distribution.
However, he recognized the ambiguity in deciding a North American or Asian origin for
hadrosaurids. This is because while the oldest and most primitive hadrosaurid (Protohadros) was
found in North America, the closest outgroup taxa to Hadrosauridae were recorded from Asia.
Casanovas et al. (1999) regarded Pararhabdodon isonensis, from the Maastrichtian of
northern Spain, as the most primitive lambeosaurine. These authors posited a dispersal of
lambeosaurines to Europe and the subsequent onset of geographical barriers that isolated the
western part of the continent during a transgressive period spanning from the Albian to the
Maastrichtian.
With the re-evaluation of Eolambia, from the Cenomanian Cedar Mountain Formation of
Utah, as a hadrosaur closely related to the Mongolian genus Probactrosaurus, Head (2001)
expanded the North American record of these animals. He considered two biogeographical
scenarios. In the first, Eolambia was either a basal relative or the sister taxon to
Probactrosaurus. This implied multiple dispersal events from Asia to North America from the
early to late Cretaceous. In a second scenario, Eolambia was placed crownwards relative to
Probactrosaurus. This would only require a single dispersal episode from Asia to North America
to explain the presence of the Eolambia in the latter continent.
127
The subsequent discovery of another non-hadrosaurid hadrosaur, Equijubus normani,
from the late Barremian-Aptian of China, provided Hai-lu You et al. (2003) new evidence in
support of the Asian origin hypothesis. In contrast, Horner et al. (2004) supported a North
American origin for hadrosaurids based on the geographical distribution of the closest outgroup
taxa to their Hadrosauridae, Protohadros and Eolambia. Based on their hadrosaurid phylogeny,
these authors explained the presence of Telmatosaurus in Europe as a dispersal event in the mid-
Cretaceous. Horner et al. (2004) also thought that the divergence of saurolophines and
lambeosaurines took place in North America in the mid-Cretaceous. Within saurolophines, these
authors posited a dispersal event into Asia no later than the early Maastrichtian (Saurolophus
angustirostris) and another dispersal event into South America no later than the Campanian
(Kritosaurus australis). Within lambeosaurines, they hypothesized a dispersal to Asia no later
than the Campanian. This served to account for the presence in China of Tsintaosaurus
spinorhinus.
Godefroit et al. (2001) and Bolotsky and Godefroit (2004) proposed that several lineages
of Asian hadrosaurids dispersed to North America no later than the Campanian via the Bering
Strait. Among saurolophines, Kerberosaurus manakini, from the late Maastrichtian of eastern
Russia, would have occupied North America before the cladogenesis of the Prosaurolophus-
Saurolophus clade. Within lambeosaurines, Charonosaurus jiayinensis, the sister taxon to
Parasaurolophus spp., would have also dispersed to North America before the evolution of the
latter.
Gates and Evans (2005) and Gates (2007) reported finer scale patterns in the distribution
of hadrosaurids during the Campanian of western North America. They found that, at a given
time for a particular region, at least one saurolophine and one lambeosaurine lived
contemporaneously (e.g., Prosaurolophus maximus-Hypacrosurus stebingeri in the Upper Two
Medicine Formation of Montana and Kritosaurus navajovius-Parasaurolophus tubicen in the
Kirtland Formation of New Mexico). The number of contemporaneous taxa however was greater
in the fauna from Dinosaur Park Formation in southern Canada. They showed that there was
considerable variation in the size of the geographical ranges for hadrosaurid genera.
Edmontosaurus was found from southern Alberta to Alaska and Gryposaurus and
Parasaurolophus were recorded from southern Alberta to the Southwest of the United States.
128
Other taxa such as Maiasaura, Corythosaurus and Lambeosaurus were restricted to smaller
areas, such as the Western Interior of the United States and southern Canada, respectively.
Gates (2007) noted that in general saurolophines seemed to have larger geographical
ranges than lambeosaurines. He identified the highest diversity of hadrosaurids in the Western
Interior Basin of North America at approximately 75 million years ago, during the late
Campanian, with 8 species recorded. During that time frame, he revealed the existence of faunal
turnover within the Prosaurolophus-Edmontosaurus and the Brachylophosaurus-Gryposaurus
clades. Likewise, Gates (2007) supported a connection between North America and Asia no later
than the early Maastrichtian (age of the genus Saurolophus, found in both continents). The lack
of pre-Maastrichtian derived hadrosaurids in Asia, combined with the record in this continent of
Maastrichtian taxa closely allied to North American ones, led him to posit a direction of dispersal
of hadrosaurids from North America to Asia.
Most recently, Godefroit et al. (2008) supported an Asian origin for both saurolophines
and lambeosaurines after inferring a phylogeny where the most basal members of each clade
lived in Asia. From Asia, lambeosaurines and saurolophines were postulated to have dispersed to
North America no later than the Campanian. This would have occurred through the Bering
Straight, that in their view remained opened from the Albian-Aptian until the end of the
Cretaceous.
Methods for the Inference of Ancestral Areas
Ancestral areas were reconstructed on the phylogeny derived from the weighted
parsimony analysis performed in Chapter 2 (Fig. 2.5). It has been shown that different methods
for reconstructing ancestral clade distributions may offer different results using the same data
(Morrone and Carpenter, 1995; Lieberman, 2000). Here, I compared the results of two widely
used pattern-based and event-based methods (Crisci et al., 2003). In pattern-based methods,
divergent patterns in the distribution of taxa are derived by regarding vicariance as the null
hypothesis and treat the presence of dispersal and extinction as homoplasy in the data (van Veller
et al., 2000). On the other hand, event-based methods integrate phylogenetic information with
explicit models of the processes (such as dispersal, vicariance or extinction) involved in the
129
distribution of taxa; these different processes are assigned a cost or benefit value in the analysis
(Ronquist, 1996; Sanmartin and Ronquist, 2004).
In the first approach I conducted ancestral state reconstructions of the areas occupied by
hadrosaurid and non-hadrosaurid iguanodontoideans using Fitch (unordered) optimization. This
method was implemented using MacClade version 4.0. The algorithm for reconstructing
ancestral areas is explained in Maddison and Maddison (2003) and is based on Fitch (1971).
In the second approach I implemented the Dispersal-Vicariance Analysis (or DIVA;
Ronquist, 1997). DIVA does not make a priori assumptions about speciation and biogeographic
processes and uses a model in which vicariant events, sympatric speciation, dispersals, and
regional extinctions may be given different costs using a tridimentional cost matrix. The method
accounts for reticulate distributional patterns of species and was implemented using the program
DIVA 1.1 (Ronquist, 1996).
The iguanodontoidean (hadrosaurids and outgroup) taxa in this analysis are distributed
among four large areas: Europe, Asia, North and South America.
Hadrosaurid Ancestral Distributions
Hadrosaur Outgroup Taxa

Both parsimony and DIVA reconstructed the origin of the Hadrosauria in Asia no later
than the early Aptian (Figs. 3.1 and 3.2). Subsequently, during the Cenomanian, members of
Hadrosauria are found in North America (Eolambia caroljonesa in Utah and Protohadros byrdi
in Texas). On one hand, DIVA ambiguously reconstructed the origin of the lineages leading to E.
caroljonesa and P. byrdi in Asia or, alternatively, in both Asia and North America (Fig. 3.2). On
the other hand, when DELTRAN was applied to the parsimony reconstruction, both clades were
inferred to have originated in Asia. When ACCTRAN was applied the clades were estimated to
have originated in North America (Fig. 3.1).
The next three clades that evolved within the Hadrosauria originated in Asia during the
late Coniacian-early Santonian and are represented by forms discovered in Campanian to early
Maastrichtian formations from China (Tanius sinensis, Bactrosaurus johnsoni and
Gilmoreosaurus mongoliensis) (Fig. 3.1).
130
Only ambiguous reconstructions were provided by parsimony (Fig. 3.1) and DIVA (Fig.
3.2) for the area occupied by the most recent common ancestor of Telmatosaurus transsylvanicus
(from the Maastrichtian of Romania) and the remaining hadrosaurs. Parsimony optimization with
DELTRAN reconstructed the origin of this clade in Asia, while the ACCTRAN option did not
shed any light in this respect. On the other hand, DIVA inferred that this ancestor was
widespread in Europe and Asia.
The next two lineages of the hadrosaur phylogeny stemmed from ancestors that lived in
the eastern half of North America according to the unambiguous reconstructions of both
parsimony and DIVA. The first of these clades originated in middle Coniacian and contains the
lineage leading to Lophorhothon atopus, from the late Santonian Mooreville Formation of
Alabama (Fig. 3.1). The second clade originated also in eastern North America around the same
time (middle-late Coniacian) and is represented by Claosaurus agilis from late Coniacian
(Everhart and Ewell, 2006) strata from the Niobrara Chalk Formation of Kansas (Fig. 3.2).
Hadrosauridae and the Saurolophine-Lambeosaurine Divergence
The Fitch parsimony analysis indicated that the ancestral area of Hadrosauridae was
located in eastern North America (Fig. 3.1), while the DIVA results showed a reconstruction
where the most recent common ancestor of Hadrosaurus and remaining hadrosaurids would have
had a widespread distribution in Asia and North America (Fig. 3.2).
However, the ancestral area for the major radiation, Euhadrosauria, as well as the split
between Saurolophinae and Lambeosaurinae, was unambiguously inferred to have occurred in
Asia by both methods. Both saurolophines and lambeosaurines were ancestrally distributed in
Asia according to the unambiguous results reconstructed by both parsimony and DIVA.
Saurolophine Ancestral Areas
The two most basal saurolophines, Shantungosaurus giganteus and Kerberosaurus
manakini, have an Asian distribution. The ancestral area of the clade stemming from the most
recent common ancestor of S. giganteus and Gryposaurus notabilis was reconstructed in Asia
after both the parsimony and DIVA analyses. The next most inclusive clade of saurolophines
stemmed from the most recent common ancestor of K. manakini and G. notabilis. The parsimony
results unambiguously estimate the ancestral area of the latter clade in Asia (Fig. 3.1), while the
DIVA results place the distribution of this ancestor in Asia and North America (Fig. 3.2).
131
The remaining saurolophine clades (except Saurolophus and the Secernosaurus clade)
were all inferred to have North American ancestral distributions by the DIVA and parsimony
analyses. DIVA estimated a widespread distribution in North America and Asia for the most
recent common ancestor of Saurolophus (Fig. 3.2). In contrast, the parsimony results point to
North America as the ancestral area for this genus (Fig. 3.1). On the other side of the
saurolophine tree, the parsimony results reconstruct the origin of the Secernosaurus clade in
North America Fig. 3.1). In contrast, the DIVA results infer a widespread distribution in both
Americas for the most recent common ancestor of this monophyletic group (Fig. 3.2).
Lambeosaurine Ancestral Areas
The three most inclusive lambeosaurine clades (i.e., those stemming from the most recent
common ancestors of Jaxartosaurus aralensis, Tsintaosaurus spinorhinus and Olorotitan
ararhensis) were unambiguously reconstructed as having originated in Asia in both the
parsimony and DIVA results.
However, the ancestral area for the most recent common ancestor of the clade composed
of “parasaurolophs” and “amurosaurs” could not be unambiguously reconstructed with either
parsimony or DIVA. Using ACCTRAN optimizamation, parsimony estimated the ancestral area
as being in North America. Asia was the area estimated when using DELTRAN. In contrast,
DIVA provided two possible ancestral distributions, Asia and a widespread distribution in Asia
and North America.
Similarly, the parsimony analysis suggests the presence of the most recent common
ancestor of “amurosaurs” in North America when using ACCTRAN and in Asia when using
DELTRAN (Fig. 3.1). On the other hand, the DIVA results estimate that the most recent
common ancestor of this clade was found in North America or widespread between this
continent and Asia (Fig. 3.2).
Within “amurosaurs”, “hypacrosaurs” include two clades, one originating in Asia
(Amurosaurus riabinini, Sahaliyania elunchunorum) and the other in North America
(Hypacrosaurus altispinus, Lambeosaurus laticaudus, Velafrons coahuilensis). However, the
ancestral area of “hypacrosaurs” could only be unambiguously reconstructed in the DIVA
analysis where the most recent common ancestor of the clade was inferred to have been
widespread in Asia and North America (Fig. 3.2). Using parsimony, the ancestral area of this
clade was inferred to be in Asia with DELTRAN and North America with ACCTRAN (Fig. 3.1).
132
On the other side of the amurosaur tree, “corythosaurs”, those lambeosaurines forming the sister
clade to “hypacrosaurs”, were unambiguously reconstructed to have originated in North America
by both the parsimony and DIVA analyses.
Finally, the birthplace of the sister clade to “amurosaurs”, “parasaurolophs”, could not be
unambiguously inferred. On one hand, the parsimony analysis reconstructed the ancestral area of
“parasaurolophs” in Asia with DELTRAN and North America with ACCTRAN (Fig. 3.1). On
the other hand, the DIVA analysis estimated the ancestral distribution of the clade as widespread
in Asia and North America (Fig. 3.2).
Global Biogeographic History of Hadrosaurids
This section provides a hypothesis of how hadrosaurids achieved their nearly

cosmopolitan distribution as they evolved and diversified to become the dominant ornithischian
dinosaurs during the late Cretaceous. It combines the information derived from the
reconstruction of ancestral areas, shown above, with reconstructions of the late Cretaceous
paleogeography of the continental landmasses culled from the literature.
Origin of Hadrosauridae
During the late Coniacian and early Santonian, several hadrosaur taxa ancestral to
Hadrosauridae lived in Eurasia. These include Bactrosaurus johnsoni, Gilmoreosaurus
mongoliensis, and Tanius sinensis in eastern Asia and Telmatosaurus transsylvanicus in eastern
Europe. Sometime during the late Coniacian, hadrosaurs dispersed to southeastern North
America, where Lophorhothon atopus and Claosaurus agilis, the closest outgroups to
hadrosaurids are found (Fig. 3.3). Given the presence in Europe of the closest outgroup (T.
transsylvanicus) to the clade stemming from L. atopus and remaining hadrosaurs, it is likely that
these animals dispersed to North America by “island hopping” through the mosaic of islands that
formed the European continent at that time (Haq et al., 1987; Dalla Vecchia, 2006).
Alternatively, Asian hadrosaurs might have also dispersed to North America via southern
Greenland and the Labrador Peninsula or nearby regions (Fig. 3.3). Paleogeographical
reconstructions of the coastlines during the Coniacian (Zharkov et al., 1998) show northern
Europe, Greenland and Northeastern North America in close proximity, adding geographical
credence to makes this hypothesis. In this way, the most recent common ancestor of hadrosaurids
133
may have become widespread throughout Asia, Europe and southeastern North America.
Subsequently, the first members of Hadrosauridae would have evolved in eastern North America
through vicariance, perhaps caused by the discontinuation of land connections among Eurasian
and American landmasses. This hypothesis is supported by the DIVA analysis and the presence
in eastern North America of Hadrosaurus foulkii, the most basal hadrosaurid, and the closest
hadrosaurid outgroups.
Primitive hadrosaurids would have then dispersed to Asia no later than the middle or late
Santonian. A possible dispersal route from eastern North America may have been through
Northeastern Canada, Greenland and northern Eurasia (Fig. 3.4). Smith et al. (1994) and
Zharkov et al. (1998) showed that during the Santonian land connections existed between the
northeastern United States, Canada and Greenland. A series of smaller landmasses also existed
between North America and Asia (Zharkov et al., 1998) that may have formed an alternative
stepping stone route for hadrosaurids. Dispersals from North America to Asia are congruent with
the presence of the most basal euhadrosaurs in eastern (Shantungosaurus, Kerberosaurus) and
Central (Aralosaurus, Jaxartosaurus) Asia. An extensive seaway separated western from eastern
North America, with no land connections during the Santonian between the northern, western
and eastern regions of North America (Blakey, 2001). Thus, it is more likely that hadrosaurids
dispersed to Asia eastwards, than westwards through Beringia.
No later than the middle or late Santonian and soon after their initial occupation of Asia,
hadrosaurids experienced their main divergence into saurolophines and lambeosaurines.
Saurolophine Biogeography
Saurolophines dispersed to western North America during the late Santonian or early
Campanian. The shortest route for this dispersal would have been across the Bering land bridge,
which may have existed during that time (Zharkov et al., 1998; Averianov and Archibald, 2003)
(Fig. 3.4). This scenario is congruent with the eastern Asian distribution of the most basal
saurolophines (Kerberosaurus, Shantungosaurus) and that of the more derived North American
lineages, represented mostly by forms from western North America (Edmontosaurus,
Prosaurolophus, Saurolophus, Kritosaurus, Gryposaurus, Maiasaura, Brachylophosaurus).
After the invasion of western North America, saurolophines experienced extensive cladogenesis
in this continent during the Campanian. Notably, all but four taxa have been recorded in North
America. These four taxa appear to have dispersed from North America. Two of them are
134
posited to have returned to Asia and the other to South America (possibly continuing into
Antarctica).
One of these saurolophine dispersals to Asia occurred within “maiasaurs” no later than
the late Campanian. This explains the presence of Wulagasaurus dongi in the far region of
eastern Russia, The remaining “maiasaurs”, as well as the most recent common ancestor of the
clade, evolved in western North America. No later than the late Campanian an outgroup lineage
to remaining “maiasaurs” occupied eastern Asia (Fig. 3.5). The most likely, shortest route from
western North America to eastern Asia would have been via the Bering land bridge (Averianov
and Archibald, 2003). According to the DIVA and parsimony reconstructions of ancestral areas
on the phylogeny, no vicariant event would have been involved in the biogeographic history of
known “maiasaurs”.
Another Asian dispersal accounts for the presence of Saurolophus angustirostris in
Central Asia by the late Campanian-early Maastrichtian. The most recent common ancestor of S.
angustirrostris and its American sister species, S. osborni, probably lived in western North
America, as inferred in the parsimony and DIVA analyses. A slightly different scenario is
suggested by the ancestral area reconstruction using the latter method. According to this the most
recent common ancestor of Saurolophus spp. became widespread in Asia and North America.
After the Asian dispersal, S. angustirrostris would have diverged from S. osborni through
vicariance. The most obvious dispersal route would have been via Beringia (Russell, 1993) (Fig.
3.6).
Finally, a third dispersal event to South America within “kritosaurs” explains the
presence of the only hadrosaurids recorded so far from this continent, Secernosaurus koerneri
and the Salitral Moreno OTU. Both are recorded from the late Campanian-early Maastrichtian of
Patagonia, Argentina. This dispersal, would have occurred no later than the late Campanian as
was previously posited by Weishampel and Weishampel (1983) and Bonaparte et al. (1984).
Kritosaurus navajovius and Gryposaurus spp., outgroup taxa to the Secernosaurus clade
(composed of the two South American taxa and the Big Bend OTU), lived in western North
America during the early Campanian-early Maastrichtian. According to the parsimony analysis,
the ancestral area of the Secernosaurus clade was probably located in North America. However,
the DIVA analysis inferred a widespread distribution for the ancestor of this clade in North and
South America after which time the two Argentinian taxa would have diverged during a
135
vicariance event. Both of these scenarios are congruent with the distribution of the outgroup
taxon to the S. koerneri-Salitral Moreno clade. The outgroup taxon to the South American clade
is an unnamed hadrosaurid from the late Campanian of Big Bend National Park, Texas. The
presence of this taxon in southern North America provides an intermediate stage in the
hypothetical dispersal of “kritosaurs” from western North America to South America during the
late Campanian.
The existence of a land connection between both American continents during the late
Cretaceous has recently been supported by several studies (Pough et al., 2004). In particular,
subduction of the southern part of the North American plate under the eastward-moving
Caribbean plate may have given rise to a Proto-Antillean volcanic arc (Duellman, 2001; Pindell
and Kennan, 2002; Hedges, 2006). This chain of islands would have been located between North
and South America in the position currently occupied by Central America, thereby providing a
dispersal route for hadrosaurids and other vertebrates during at least the Late Campanian
(Hedges, 2006). The Antilles are believed to have subsequently moved eastward during the
migration of the Caribbean plate (Duellman, 2001). By the Maastrichtian, this connection would
have been interrupted (Pindell and Kennan, 2002). This scenario is congruent with vicariance
being involved in the evolution of the Secernosaurus-Salitral Moreno clade in South America. A
similar vicariance scenario, whereby the Caribbean plate carried ancient Mesozoic biota
eastwards, has previously been posited to explain the evolution of various endemic terrestrial
vertebrates in the Antilles (Hedges, 2006).
Finally, Case et al. (2000) reported an isolated hadrosaurid tooth from the late
Maastrichtian Lopez de Bertodano Formtion, in Vega Island, Antarctica. I concur in these
authors in regarding this tooth as belonging to a member of Saurolophinae. My reasons for this
referral are the absence of all but a median carina and the lack of, or very small size of marginal
denticles. Hadrosaurids may have dispersed into Antarctica through an isthmus that connected
this continent with South America during, at least, the late Campanian and early Maastrichtian
(Zharkov et al., 1998; Hay et al., 1999) (Fig. 3.5).
Lambeosaurine Biogeography
Lambeosaurine hadrosaurids originated in Asia during the middle or late Santonian. This
is in agreement with the biogeographic hypothesis of Godefroit et al. (2008). However, while
136
saurolophines had already occupied North America by the late Santonian, lambeosaurines
remained restricted to Asia until the Campanian (Figs. 3.1 and 3.2).
Early in their evolution (i.e., sometime in the middle Campanian) “tsintaosaur”
lambeosaurines dispersed to Europe. This dispersal explains the presence of Pararhabdodon
isonensis in the late Maastrichtian of the Iberian Peninsula. The most recent common ancestor of
the clade lived in Asia during Campanian times. Although no continuous land connection
between Asia and Europe has been reconstructed for the late Cretaceous, “tsintaosaurs” may
have used the mosaic of small landmasses that then formed Europe (Blakey, 2001) as stepping-
stones to reach the western region of this continent. This occurred by the Maastrichtian. The
DIVA reconstruction of the ancestral area of the clade inferred a widespread ancestor across
Eurasia and subsequent cladogenesis between P. isonensis and Tsintaosaurus spinorhinus
through vicariance. This hypothesis supports similar conclusions reached by Casanovas et al.
(1999). The mosaic paleogeographic configuration of Europe during that time supports the
possibility of vicariant events operating due to the intermittent nature of the land connections
owing to relatively high eustatic sea levels (Dalla Vecchia, 2006).
No later than the late Campanian, lambeosaurines occupied western North
America,where they underwent extensive cladogenesis. This lambeosaurine radiation gave rise to
the “amurosaur” (“hypacrosaurs” + “corythosaurs”) and “parasauroloph” clades. However, the
ancestral areas for the most recent common ancestor of “amurosaurs” and the clade composed of
the latter and “parasaurolophs” could not be unambiguously reconstructed (Fig. 3.5; double-
headed arrows). In particular, it is ambiguous whether all North American “amurosaurs” and
“parasaurolophs” radiated after a single dispersal event to this continent or whether
“hypacrosaurs”, “corythosaurs” and “parasurolophs” evolved after three independent dispersal
events from Asia (Figs. 3.1 and 3.2).
Within North American “hypacrosaurs”, it is likely that their most recent common
ancestor originated in the southern region of the continent. Evidence for this is the time
calibrated phylogram for the subclade showing a northward shift in the distribution of its
constituent species from the late Campanian to the early Maastrichtian (Fig. 3.1). Specifically,
Velafrons coahuilensis, the outgroup to the Lambeosaurus laticaudus-Hypacrosaurus altispinus
clade, lived in Mexico during the Campanian. L. laticaudus also lived in the late Campanian of
Mexico. The occurrence of its sister taxon H. altispinus in the early Maastrichtian of northern
137
North America could have been the result of a northward dispersal. “Corythosaurs” remained
restricted to western North America (Fig. 3.2).
If the most recent common ancestor of “parasaurolophs” lived in North America,
Charonosaurus jiayinensis from the Maastrichtian of China may represent an Asian dispersal of
the clade. Alternatively, the results of the DIVA analysis suggest that it may have evolved by
vicariance after widespread distribution throughout America and Asia by the “parasauroloph”
ancestor. In contrast, if the most recent common ancestor of “parasaurolophs” lived in Asia, the
genus Parasaurolophus would have evolved after dispersal from Asia to North America by the
ancestor of the clade.
Discussion
The validity of the inferred time of origin and the timing and sequence of the subsequent
historical biogeographic events hypothesized in this study depends (besides on the accuracy of
the phylogenetic hypothesis) on the accuracy of the stratigraphic ages of the taxa that were
analyzed. In general, there are very few stratigraphic sequences that have yielded hadrosaur
material whose age is known with relatively good precision.
Two exceptions where aging has been possible with a relatively good degree of accuracy
and precision are the late Campanian deposits of the Dinosaur Provincial Park, in southern
Canada (that yielded a rich fossil record of saurolophines and lambeosaurines) and the late
Maastrichtian strata from the South-Central Foredeep Pyrenean Basin of Northeastern Spain
(that yielded one of the only two named hadrosaurs in Europe, the lambeosaurine
Pararhabdodon isonensis).
Unfortunately, the age of several hadrosaur species that occupy basal positions in the
phylogeny has been contentious. The age of these basal species is important for unraveling the
age of first appearance of particular clades, within which those species are outgroup to all the
other taxa. This is exemplified by the uncertainly that has surrounded the age of the Iren Dabasu
Formation in Inner Mongolia, southern China. This formation yielded two outgroup taxa to
hadrosaurids, Bactrosaurus johnsoni and Gilmoreosaurus mongoliensis. The age of the Iren
Dabasu Formation has been estimated as lower Cretaceous (Berkey and Morris, 1927),
Cenomanian (Rozhdestvensky, 1966; 1977), pre-Turonian (Weishampel and Horner, 1986), pre-
138
Santonian (Brett-Surman, 1979), early late Creatceous (Godefroit et al., 1998), early Campanian
(Jerzykiewicz and Russell, 1991; Currie and Eberth, 1993) and Maastrichtian (Chen, 1983; Liu
and Wu, 1990). More recently, Van Itterbeeck et al. (2005) re-studied the stratigraphy and
sedimentology of the Iren Dabasu Formation and concluded that it is most probably latest
Campanian-early Maastrichtian in age. They based this conclusion on the age provided by
microfossils, particularly four species of charophytes and eight species of ostracods.
The Wangshi Group, in the Chinese Province of Shandong, yielded a primitive hadrosaur
fauna, including Tanius sinensis (Wiman, 1929), the most basal saurolophine Shantungosaurus
giganteus (Hu, 1972) and the basal lambeosaurine Tsintaosaurus spinorhinus (Young, 1958).
Estimations of the age of the Wangshi Group vary from Coniacian to the Maastrichtian (Zhang et
al., 2003; Horner et al., 2004) or simply to the Campanian (Buffetaut, 1995; Godefroit et al.,
2003; Bolotsky and Godefroit, 2004). However, Zhang et al. (2008) recently reported that basalt
floods in the Wangshi Group date to 74 million years old, placing these strata in the middle
Campanian. Additionally, Hong and Miyata (1999) dated the Wangshi Group as being between
82.4 and 81.8 million years old on the basis of fission track zircon ages and the fossil gastropod
Campeloma liui.
The age of Aralosaurus tuberiferus is of great importance for estimating the time of
origin of Lambeosurinae because this species is the outgroup taxon to all other members of the
clade. A. tuberiferus was collected from the Shakh Shakh locality of the Bostobe Formation, in
the Northeastern Aral Sea Region, Kazakhstan (Godefroit et al., 2004b). The Bostobe Formation
has also been known as Beleuta (Hwang et al., 2002) or Bostobynskaya (Dyke and Malakhov,
2004) Formation. The age of this formation has received different estimates, from Turonian
(Nessov and Khissarova, 1988), Turonian-Santonian (Godefroit et al., 2004b), Santonian
(Averianov, 2007), Santonian-Campanian (Kordikova et al., 2001), and Campanian (Nessov,
1995).
In addition to the uncertainty in the age estimates of hadrosaurs, the completeness of
theirfossil record and the stratigraphic fit of the phylogeny utilized in this analysis can affect the
biogeographic hypotheses herein presented. The time-calibrated phylograms (Fig. 3.1 and 3.2)
showed a substantial amount of time represented by ghost lineages. In order to show
quantitatively the extent of this apparent lack of representation in the evolutionary history of
Hadrosauria, I estimated the completeness of the fossil record of these animals. This was
139
accomplished using the analytical method of Foote and Raup (1996). Unlike other approaches
(Norell and Novacek, 1992; Benton and Storrs, 1994; Smith and Littlewood, 1994) that provide
an assessment of the proportion of preserved taxa in the fossil record based on a phylogenetic
hypothesis, Foote and Raup’s method is based only on the duration of the taxa of interest and
does not necessitate a phylogeny. This method was preferred because it does not incorporate
potential errors in the accuracy of the phylogeny.
In their approach, Foote and Raup (1996) estimated the proportion of taxa preserved in
the fossil record of a given group by combining estimates of the preservation probability and
estimates of the original durations of its constituent taxa. Specifically, the completeness of the
fossil record was computed using the following expression:
PP = ∑∞T=1 P1(T) · h(T),
where PP was the proportion of preserved fossil record, P1(T) the probability that a taxon was
preserved at least once given an original duration of T stratigraphic units and h(T) the probability
that the original durations were equal to T. P1(T) was given by the following expression:
P1(T) = 1 – (1 – R) T ,
where R was the probability that a given species was preserved at least once in one stratigraphic
unit and T the number of stratigraphic units where that species was recorded. R was calculated
using the following equation:
R = f(2)2 / [f(1) · f(3)],
where f(1), f(2) and f(3) were the frequencies of hadrosaur species preserved in one, two and
three stratigraphic units, respectively. For the calculation of f(t) (the frequencies of hadrosaur
species with t preserved stratigraphic units), each geologic stage was divided into three parts
(i.e., units: early, middle and late). The next step was to compile the number of stratigraphic
units recorded for every hadrosaur species (Table 3.1). Thus, for example, a hadrosaur recorded
from the late Santonian to the early Campanian would be assigned a value of two stratigraphic
units and a species recorded from the Santonian would be assigned a value of three units.
Finally, the probability that the original duration of a taxon was equal to T stratigraphic
units was calculated using the following expression:
h1(T) = e –q(T – 1) – e –qT.
140
The calculation of this probability assumed an exponential distribution of the original
durations with constant extinction rate (Alba et al., 2001). This constant, designated as q in the
formula above, was obtained from the natural logarithm of the slope of the range of frequencies
f(t).
In the case of hadrosaurs, the probability that a species was recorded at least once in a
stratigraphic unit was 0.61. The completeness of the hadrosaur fossil record was 0.69. Two
limitations inherent to the hadrosaur data in Table 3.1 should be kept in mind as potential sources
of error for this estimate. First, and as explained above, hadrosaur ages are not precise point
estimates but consist on time intervals subject to error and uncertainty. Second, the slope q could
only be approximated as the difference between the natural logarithms of f(2) and f(3). This is
because the small (relative to other taxonomic groups such as Neogene mammals) number of
hadrosaur species did not include any taxa with longer durations than three stratigraphic units.
With this mind, the hadrosaur fossil record appeared to be reasonably complete, a result hardly
surprising given the fact that this clade is one of the best represented within non-avian dinosaurs.
Thus, potential inaccuracies in the inference of the biogeographic history of hadrosaurs are less
likely to be due to the incompleteness of their fossil record.
I also assessed the fit between the stratigraphic ages of hadrosaurid taxa and the order of
branching events of the phylogeny considered in this study. This was accomplished by
implementing the Manhattan Stratigraphic Measure (Siddall, 1998) as modified by Pol and
Norell (2001; MSM*). The MSM* has the advantage of not being affected by tree shape and
being much more sensitive to the age differences among taxa than other measures such the SRC
(Norell and Novacek, 1992) and SCI (Huelsenbeck, 1994). Conceptually, the MSM* is
analogous to the consistency index (Kluge and Farris, 1969). Calculation of the MSM* is based
on the optimization of an age character using a symmetrical Sankoff step matrix of absolute age
differences (Table 3.1 and 3.2). The MSM* for Hadrosauria resulted in a value of 0.13, with a P-
value of 0.08; for Hadrosauridae, MSM* resulted in a value of 0.11 and a P-value of 0.01. These
measures indicate a poor degree of congruence between the sequence of appearance of
hadrosaurs in the stratigraphical record and the branching sequence in the phylogeny. However,
at least for hadrosaurids, there is a significant fit between the stratigraphy and the phylogeny,
even if this is poor. This low degree of congruence between the stratigraphic succession of taxa
and the phylogeny contrasts with the relatively good completeness of the hadrosaur fossil record.
141
Norell and Novacek (1992) found a similar discrepancy between the stratigraphic record and the
phylogeny of hadrosaurs. Likewise, these authors also noted other cases in which a clade had a
well-represented fossil record but a poor stratigraphic fit with the phylogeny, such as primates.
Norell and Novacek (1992) explained this incongruent pattern as caused by the early branching
of a few taxa in the phylogeny that were recorded later in the stratigraphic succession. In the case
of hadrosaurs, there are a number of taxa that follow the same pattern, notably various outgroup
species to several clades (e.g., Wulagasaurus dongi for “maisaurs”, Telmatosaurus
transsylvanicus within hadrosaurs, Kerberosaurus manakini within saurolophines, Kritosaurus
navajovius for “kritosaurs” and Charonosaurus jiayinensis for “parasaurolophs”; Fig. 3.1).
Although this study inferred the birthplace of Hadrosauridae in North America, the major
evolutionary event of the radiation of Euhadrosauria was inferred to have occurred in Asia. In a
broader context, this hypothesis is congruent with the dinosaurian pattern of origination indicated
by Russell (1993). Russell emphasized the importance of Asia as the ancestral area for numerous
groups of non-avian dinosaurs, particularly those with well-represented distributions in North
America. More recently, an Asian origin has been posited for a number of dinosaurian
contemporaries of hadrosaurids, such as dromaeosaurids (Norell and Makovicky, 2004),
troodontids (Makovicky and Norell, 2004), therizinosauroids (Clark et al., 2004),
pachycephalosaurs (Maryanska et al., 2004) and neoceratopsians (Hai-lu You and Dodson,
2004). Similarly to the saurolophine and lambeosaurine pattern, dispersals between North
America and Asia via the Bering land bridge have also been proposed for other late Cretaceous
dinosaurian clades (Russell, 1993). For example, troodontids appear to have dispersed back to
North America no later than the Campanian (Makovicky and Norell, 2004). Ornithomimids and
therizonosauroids require dispersal events to North America to account for their biogeography
(Makovicky and Norell, 2004; Clark et al., 2004). Pachycephalosaurs, like hadrosaurids, showed
a bidirectional dispersal pattern between North America and Asia (Maryanska et al., 2004). The
mammalian record is weakly in accord with the hadrosaurid pattern since faunal exchanges
between North America and Asia of late Cretaceous mammals appear to have been uncommon.
Only two cases are known, one before the Cenomanian and another in the middle Campanian
(Weil, 2004).
“Kritosaur” saurolophines may have been accompanied by other groups of vertebrates in
their dispersal to South America. Didelphid marsupials and several groups of placental mammals
142
with North American affinities are recorded in the late Cretaceous of South America (Bonaparte
and Rougier, 1987; Cox, 2000).
Conclusions
The present study supports North America as the center of origin for Hadrosauridae.
Their genesis for this event is traced to the middle-late Santonian. However, the major
hadrosaurid radiation, signaled in by the split between Saurolophinae and Lambeosaurinae,
occurred in Asia no later than the end of the Santonian. Saurolophines dispersed back to North
America during the late Santonian, where they showed considerable radiation in the Campanian.
Two dispersal events during that period are inferred to have occurred. First to southeastern and
Central Asia and a second one to South America and perhaps Antarctica. From Asia,
lambeosaurines dispersed to Europe and North America no later than the late Campanian.
However, it is uncertain at which point in the lambeosaurine phylogenythese animals dispersed
to North America, and whether a single dispersal event to the American continent with
subsequent occupations of Asia or independent dispersals to North America in the
“parasauroloph” and “amurosaur” clades.
Vicariance may have occurred follwoing several of the inferred hadrosaurid dispersals.
Specifically, the Santonian saurolophine dispersal to Asia, the late Campanian dispersal to South
America, the lambeosaurine dispersal to Europe and subsequent Campanian exchanges between
North America and Asia may have been followed by vicariant events due to the temporary
interruption of geographic connections and/or other yet unrecognized factors.
The poor but significant congruence between the hadrosaurid stratigraphic record and
their phylogenetic relationships, as well as their relatively complete fossil record, indicate that
the historical biogeogeography reconstructed in this study is based on a sample of species that
represents relatively well the diversity of hadrosaurs that once existed.
143
TURONIAN SANTONIAN
APTIAN ALBIAN CENOMANIAN CONIACIAN CAMPANIAN MAASTRICHTIAN
125.0 112.0 99.6 93.5 89.3 85.8 83.5 70.6 65.5
Iguanodon spp.
Equijubus normani
HADROSAURIA
Protohadros byrdi
Tanius sinensis
Lophorhothon atopus
Claosaurus agilis
Hadrosaurus foulkii
SAUROLOPHINAE
HADROSAURIDAE Saurolophus osborni
Sabinas OTU
Wulagasaurus dongi
Wahweap OTU
EUHADROSAURIA
Salitral Moreno OTU
UTEP Big Bend OTU
LAMBEOSAURINAE Tsintaosaurus spinorhinus
Lambeosaurus lambei
Figure 3.1. Time-calibrated phylogram based on the single most parsimonious tree derived from weighted
parsimony analysis of hadrosaurid relationships (Chapter 2, Fig. 2.5), showing the Fitch parsimony
reconstruction of ancestral areas. Grey lines indicate ambiguous reconstructions. Geochronological ages
from Gradstein et al. (2004).
144
TURONIAN SANTONIAN
APTIAN ALBIAN CENOMANIAN CONIACIAN CAMPANIAN MAASTRICHTIAN
125.0 112.0 99.6 93.5 89.3 85.8 83.5 70.6 65.5
Iguanodon spp.
Equijubus normani
(A)(AM)
HADROSAURIA
(M)(AM) Protohadros byrdi
v Tanius sinensis
(AE)(AM)(AEM)
Lophorhothon atopus
v Claosaurus agilis
Hadrosaurus foulkii
SAUROLOPHINAE
v
v Prosaurolophus maximus
HADROSAURIDAE v Saurolophus osborni
Sabinas OTU
Wulagasaurus dongi
Wahweap OTU
EUHADROSAURIA
v Salitral Moreno OTU
UTEP Big Bend OTU
LAMBEOSAURINAE v Tsintaosaurus spinorhinus
v Sahaliyania elunchunorum
(M)(AM) Velafrons coahuilensis
(A)(AM) Lambeosaurus lambei
v Parasaurolophus cyrtocristatus
Figure 3.2. Time-calibrated phylogram based on the single most parsimonious tree derived from weighted
parsimony analysis of hadrosaurid relationships (Chapter 2, Fig. 2.5), showing the DIVA reconstruction of
ancestral areas. Grey lines indicate ambiguous reconstructions. Black lines indicate dispersals leading to
widespread ancestors and subsequent vicariance. Abbreviations: A = Asia; E = Europe; M = North America;
v = vicariance. Geochronological ages from Gradstein et al. (2004).
145
Figure 3.3. Paleogeographic reconstruction of continental coastlines during the Coniacian. The arrows show the main dispersal routes
that might have followed hadrosaur ornithopods. Scisors indicate that vicariance may have ocurred upon dispersal of the most recent
common ancestor of a particular clade. Continental paleocoastlines redrawn after Smith et al. (1994), Zharkov et al. (1998) and
Blakey (2001).
146
Figure 3.4. Paleogeographic reconstruction of continental coastlines during the Santonian. The arrows show the main dispersal routes that
might have followed primitive hadrosaurids. Scisors indicate that vicariance may have ocurred upon dispersal of the most recent common
ancestor of a particular clade. Continental paleocoastlines redrawn after Smith et al. (1994), Zharkov et al. (1998) and Blakey (2001).
147
Figure 3.5. Paleogeographic reconstruction of continental coastlines during the Campanian. The arrows show the main dispersal routes that
might have followed saurolophine and lambeosaurine hadrosaurids. Scisors indicate that vicariance may have ocurred upon dispersal of the
most recent common ancestor of a particular clade. Continental paleocoastlines redrawn after Hay et al. (1999), Blakey (2001), Pough et al.
(2003) and Hedges (2006).
148
Figure 3.6. Paleogeographic reconstruction of continental coastlines during the Maastrichtian. The arrows show the main dispersal routes
that might have followed hadrosaurids. Scisors indicate that vicariance may have ocurred upon dispersal of the most recent common
ancestor of a particular clade. Continental paleocoastlines redrawn after Zharkov et al. (1998) and Blakey (2001).
149
Table 3.1. Coding of the Sankoff age character used in the calculation of the Manhattan
Stratigraphic Measure.
Taxa Age State Geochronological Stage

(million Code
years)
Edmontosaurus annectens 65 a late Maastrichtian
Edmontosaurus regalis 65 a late Maastrichtian
Kerberosaurus manakini 65 a late Maastrichtian
Wulagasaurus dongi 65 a late Maastrichtian
Pararhabdodon isonensis 65 a late Maastrichtian
Charonosaurus jiayinensis 65 a late Maastrichtian
Olorotitan ararhensis 65 a late Maastrichtian
Amurosaurus riabinini 65 a late Maastrichtian
Sahaliyania elunchunorum 65 a late Maastrichtian
Telmatosaurus transsylvanicus 68 b early Maastrichtian
Saurolophus osborni 68 b early Maastrichtian
Saurolophus angustirostris 68 b early Maastrichtian
Hypacrosaurus altispinus 68 b early Maastrichtian
Bactrosaurus johnsoni 70 c late Campanian-early Maastrichtian
Gilmoreosaurus mongoliensis 70 c late Campanian-early Maastrichtian
Kritosaurus navajovius 70 c late Campanian-early Maastrichtian
Salitral Moreno OTU 70 c late Campanian-early Maastrichtian
Secernosaurus koerneri 70 c late Campanian-early Maastrichtian
Hadrosaurus foulkii 74 d late Campanian
Wahweap OTU 74 d late Campanian
Maiasaura peeblesorum 74 d late Campanian
Brachylophosaurus canadensis 74 d late Campanian
Sabinas OTU 74 d late Campanian
Prosurolophus maximus 74 d late Campanian
Big Bend UTEP OTU 74 d late Campanian
Gryposaurus notabilis 74 d late Campanian
Gryposaurus monumentensis 74 d late Campanian
Parasaurolophus walkeri 74 d late Campanian
Parasaurolophus tubicen 74 d late Campanian
Parasaurolophus cyrtocristatus 74 d late Campanian
Lambeosaurus magnicristatus 74 d late Campanian
Lambeosaurus lambei 74 d late Campanian
Hypacrosaurus stebingeri 74 d late Campanian
Corythosaurus casuarius 74 d late Campanian
Corythosaurus intermedius 74 d late Campanian
Velafrons coahuilensis 74 d late Campanian
Lambeosaurus laticaudus 74 d late Campanian
150
Taxa Age State Geochronological Stage

(million Code
years)
Tanius sinensis 80 e early Campanian
Shantungosaurus giganteus 80 e early Campanian
Gryposaurus latidens 80 e early Campanian
Tsintaosaurus spinorhinus 80 e early Campanian
Lophorhothon atopus 83 f late Santonian-early Campanian
Aralosaurus tuberiferus 83 f late Santonian-early Campanian
Jaxartosaurus aralensis 84 g Santonian
Claosaurus agilis 86 h late Coniacian
Protohadros byrdi 95 i middle Cenomanian
Eolambia caroljonesa 98 j early Cenomanian
Probactrosaurus gobiensis 110 k Albian
Equijubus normani 125 l early Aptian
151
Table 3.2. Sankoff step matrix of the age character used in the calculation of the
Manhattan Stratigraphic Measure (MSM*).
a b c d e f g h i j k l
a 0 3 5 9 15 18 19 21 30 33 45 60
b i 0 2 6 12 15 16 18 27 30 42 57
c i i 0 4 10 13 14 16 25 28 40 55
d i i i 0 6 9 10 12 21 24 36 51
e i i i i 0 3 4 6 15 18 30 45
f i i i i i 0 1 3 12 15 27 42
g i i i i i i 0 2 11 14 26 41
h i i i i i i i 0 9 12 24 39
i i i i i i i i i 0 3 15 30
j i i i i i i i i i 0 12 27
k i i i i i i i i i i 0 15
l i i i i i i i i i i i 0
152
CHAPTER 4
OSTEOLOGY AND SYSTEMATICS OF THE HADROSAURID DINOSAUR
SECERNOSAURUS KOERNERI (= KRITOSAURUS AUSTRALIS) FROM THE LATE
CRETACEOUS OF ARGENTINA
Abstract
Hadrosaurids form the most diverse and derived clade of ornithopod dinosaurs. Although
well represented in Asia and North America, the presence of this clade in South America is
known only from rare and fragmentary remains that are poorly documented and mostly
unstudied. As a result, the impact of these animals on the phylogenetics and biogeography of
hadrosaurids as a whole are not known. Here, I provide a revised and complete osteology of the
type specimens and hypodigms for the only two taxa known from South America, Secernosaurus
koerneri and Kritosaurus australis,. Likewise, I infer the phylogenetic position and historical
biogeography of South American hadrosaurids using a nearly complete taxonomic sampling of
hadrosaurid species. Parsimony and Bayesian methods were used to infer phylogenetic
relationships, while Fitch parsimony and Dispersal-Vicariance analyses were implemented to
reconstruct ancestral areas.
Kritosaurus australis is regarded as a junior synonym of Secernosaurus koerneri, based
on a combination of iliac and pubic characters unique to these two taxa. Inclusion of S. koerneri
within the genus Kritosaurus is not supported by the phylogenetic analysis. S. koerneri is
inferred to be a member of the “kritosaur” clade within Saurolophinae, as the sister taxon to the
Argentinean OTU from Salitral Moreno. Another unnamed hadrosaurid, an OTU from Big Bend
National Park, Texas, is positioned as the closest outgroup to the South American clade. The
results of this biogeographical analysis indicates that the Secernosaurus clade originated in South
America during the late Campanian after a dispersal event (probably followed by vicariance)
from southern North America before the end of that geologic stage.
153
Introduction
Hadrosauridae is a diverse clade of herbivorous dinosaurs from the Late Cretaceous of

Europe, Asia, the Americas, and Antarctica (Weishampel et al., 1990; Forster, 1997; Case et al.,
2000). In North America and Asia numerous hadrosaurid species have been erected based on
numerous, well-preserved remains, including complete articulated skeletons, multi-individual
aggregations, embryonic material and soft tissue impressions (Rozhdestvenskii, 1957; Dodson,
1971; Varricchio and Horner, 1993; Godefroit et al., 2004). In contrast, our knowledge of
hadrosaurid diversity and evolution in South America is comparatively poor with very few taxa
erected to date (Brett-Surman, 1979; Bonaparte et al., 1984).
Although hadrosaurids have been recorded in South America since early in the twentieth
century (Brett-Surman, 1975), most of their remains are too fragmentary to diagnose to generic
or specific levels (Casamiquela, 1964; 1980; Alonso, 1980; Alonso and Marquillas, 1986;
Bonaparte, 1996; Coria, 1999; Gonzalez Riga and Casadio, 2000). Consequently, of the
approximately 50 valid species of hadrosaurids, only two derive from South America. Nearly all
the discoveries of South American Hadrosauria have occurred in Argentina, with the exception
of a caudal vertebra from the Aptian-Albian of Brazil and two teeth from the Aptian of Uruguay
(Salinas et al., 2005). The most complete and best preserved materials belong to the type of
Secernosaurus koerneri (Brett-Surman, 1979), the type and hypodigm of Kritosaurus australis
(Bonaparte et al., 1984). (Note: an extensive collection from the late Campanian-early
Maastrichtian Allen Formation [Río Negro province, southern Argentina] Salitral Moreno
bonebed is also known but is curently undescribed [Powell, 1987]). The present study focuses on
the material of S. koerneri and K. australis.
The remains of Secernosaurus koerneri, the first hadrosaurid named from South
America, were collected in 1923 by an expedition from The Field Museum (Chicago, IL) led by
J. B. Abbott (Brett-Surman, 1975). The team recovered the partial skeleton of a juvenile
hadrosaurid from Upper Cretaceous strata near the head of the Chico River, east of Lake Colhué
Huapi (northeastern Chubut province, Argentina) (Fig. 4.1). This material remained unstudied
until the 1970’s at which time Brett-Surman provided a cursory diagnosis (1975; 1979). He
regarded it as the type of a new genus and species of hadrosaurid, S. koerneri. This diagnosis was
based on characters of the supraacetabular and postacetabular processes of the ilium (Brett-
154
Surman, 1979). Other materials were never described. Brett-Surman (1979) first considered S.
koerneri as a primitive form outside Euhadrosauria, although more closely related to
Saurolophinae than to Lambeosaurinae. Based on his interpretation of iliac primitive characters
in S. koerneri he entertained the possibility of a South American origin for hadrosaurids. This
stood in contradiction to the more widely accepted paradigm of an Asian origin. Later, Brett-
Surman (1989) regarded S. koerneri as Hadrosaurinae (Saurolophinae of the present study)
incertae sedis due to the lack of diagnostic cranial data needed for a more precise classification.
From 1982 to 1986, numerous cranial and postcranial elements corresponding to several
hadrosaurid specimens were collected in the late Campanian-early Maastrichtian Los Alamitos
Formation, in the Arroyo Verde region of southeastern Rio Negro province (Bonaparte et al.,
1984) near the border with Chubut province in southern Argentina (Fig. 4.1). Bonaparte et al.
(1984) referred these materials to the genus Kritosaurus on the basis of pelvic and dentary
characters and erected the new species K. australis. The presence of a species of Kritosaurus, a
genus otherwise known only from the western interior of the United States, in South America led
Bonaparte et al. (1984) and Bonaparte and Rougier (1987) to entertain the possibility that
hadrosaurids from North America had colonized Patagonia.
When comparing Kritosaurus australis with Secernosaurus koerneri, Bonaparte et al.
(1984) and Bonaparte (1996) found practically no meaningful differences between the pubis and
scapula of these taxa. However, these authors maintained their validity on the grounds that the
iliac postacetabular process of S. koerneri is more triangular in lateral view than in K. australis.
In addition, Bonaparte et al. (1984), Bonaparte (1996) and Wagner (2001) regarded the
morphology of the supraacetabular and postacetabular processes of the ilium of S. koerneri as
affected by post-mortem distortion. Wagner (2001) also found K. australis indistinct from S.
koerneri and referred the former to the genus Secernosaurus as S. australis. He maintained the
specific distinction between the Chubut and Rio Negro specimens because of their temporal and
geographical separation. More recently, Salinas et al. (2006) found K. australis to be closely
related to non-hadrosaurid hadrosaurs from Europe such as Telmatosaurus transsylvanicus and
not cogeneric with Kritosaurus. The authors based this conclusion on what they felt were
distinctive triangular predentary denticles and short dentary edentulous portion and the shape of
the scapular blade of K. australis.
155
Only Horner et al. (2004) have included Kritosaurus australis in a cladistic analysis of
hadrosaurid relationships. In their phylogeny, this species formed a polytomy with Saurolophus
and Naashoibitosaurus (= Kritosaurus) (Chapter 1, Fig. 1.12, this dissertation) within
Hadrosaurinae (Saurolophinae of this dissertation). That polytomy was supported by one
synapomorphy: a pointed and symmetrically triangular rostral process of the jugal. Horner and
colleagues considered Saurolophus to be more closely related to K. australis due to the presence
of low zygapophyseal peduncles on cervical vertebrae. Likewise, these authors found no
characters supporting the inclusion of K. australis within the genus Kritosaurus.
Because the aforementioned anatomical analyses are incomplete and the phylogenetic
context and biogeography of South American hadrosaurids have not been scrutinized in the light
of modern techniques or in a comprehensive analyses encompassing all known hadrosaurid taxa,
I chose to reanalyze these fossils and establish their true import. To achieve this I described and
documented the anatomy of the only two taxa known in South America, Secernosaurus koerneri
and Kritosaurus australis. I placed them in a phylogenetic context of all known hadrosaurids and
reconstructed their biogeographical history. In doing so, I addressed a number of questions
regarding the taxonomy and evolutionary history of South American hadrosaurids. Is the
material for S. koerneri diagnostic? Is K. australis cogeneric and conspecific with S. koerneri? Is
K. australis cogeneric with Kritosaurus (= Gryposaurus)? What is the phylogenetic position of
South American taxa within Hadrosauria and Hadrosauridae? Where did the most recent
common ancestor of South American hadrosaurids evolve? How and when did hadrosaurids
occupy South America?
Institutional Abbreviations—AMNH, American Museum of Natural History, New
York, New York, USA; DMNH, Denver Museum of Natural History, Denver, Colorado, USA;
FMNH, The Field Museum, Chicago, USA; MACN, Museo Argentino de Ciencias Naturales
Bernardino Rivadavia, Buenos Aires, Argentina; MOR, Museum of the Rockies, Bozeman,
Montana, USA; MSNM, Museo Civico di Storia Naturale di Milano, Milano, Italy; ROM, Royal
Ontario Museum, Toronto, Ontario, Canada; UTEP, Centennial Museum at the University of
Texas at El Paso, Texas, USA.
156
I examined the type and all material referable to Secernosaurus koerneri housed at the
FMNH. I also examined the holotype and hypodigm of Kritosaurus australis housed at the
MACN.
The phylogenetic position of Secernosaurus koerneri (= Kritosaurus australis) was
reassessed using a nearly complete taxonomic sampling of hadrosaurid dinosaurs. (Unless
otherwise indicated, all of the following anatomical comparisons between the aforementioned
materials and those of other taxa were based on direct examinations of specimens.) This analysis
included 12 outgroup and 39 ingroup (hadrosaurid) taxa. Outgroup taxa consisted of 2 non-
hadrosaur iguanodontoideans and 10 non-hadrosaurid hadrosaurs. Species of Iguanodon and
Ouranosaurus nigeriensis that are iguanodontoideans distantly related to hadrosaurids were used
to root the tree by outgroup comparison (Wiley, 1981; Maddison et al., 1984). The searches for
the optimal tree(s) using maximum parsimony were carried out in PAUP version 4.0b10
(Swofford, 2002). Two methodologies were implemented. In the first one, parsimony was
applied with all of the characters having the same weight. In the second I applied the weighted
parsimony of Goloboff (1993), with a k fit parameter of 3. I performed heuristic searches using a
random additional sequence of 10,000 replicates for the unweighted parsimony and 5,000
replicates for the weighted parsimony analysis. I used the option of branch-swapping by tree-
bisection-reconnection in both parsimony analyses (TBR) (Swofford et al., 1996a). Bremer
support (Bremer, 1988) was assessed by computing decay indices (Donoghue et al., 1992) with
MacClade version 4.0 (Maddison and Maddison, 2003) and PAUP. Bootstrap proportions
(Felsenstein, 1985) were computed in PAUP. Two bootstrap settings were used. For the
unweighted parsimony analyses the bootstrap was set to 10,000 replicates using heuristic
searches where each was performed using random additional sequences with branch-swapping
by TBR and 1,000 replicates. For the weighted parsimony analyses the bootstrap was set to 5,000
replicates using heuristic searches where each search was conducted using random additional
sequences with branch-swapping by SPR (Subtree Pruning and Regrafting) and 25 replicates. I
also conducted Bayesian analyses using the maximum likelihood model of Lewis (2001)
(Markov k or Mk) for discrete morphological data and implemented in MrBayes version 3.1.2
(Huelsenbeck and Ronquist, 2001). Two different models were applied. In a first model, I
157
assumed equal rates of change among characters. In the second model I incorporated the gamma
parameter to allow for variable rates of character change (Nylander et al., 2004). Priors other
than the default for the models were not specified. The analysis without gamma used six chains
and was run for 10,000,000 generations that were sampled every 100 generations. Stationarity
was achieved with a split frequency standard deviation of 0.005. Trees generated during the first
100,000 generations were discarded as “burn-in”. The analysis with gamma used four chains and
4,000,000 generations sampled every 100 generations, reaching stationarity with a split
frequency standard deviation of 0.008. Trees generated during the first 100,000 generations were
also discarded as “burn-in”. Ancestral states were reconstructed using parsimony in MacClade
version 4.0 (Maddison and Maddison, 2003).
Inference of the historical biogeography of Secernosaurus koerneri in the context of all
other Hadrosauria required the reconstruction of ancestral areas. These were reconstructed on a
weighted parsimony phylogeny derived from the latest and most comprehensive phylogenetic
analysis of Hadrosauridae (Chapter 2, this dissertation). Fitch parsimony (Fitch, 1971) and the
Dispersal-Vicariance (Ronquist, 1996) method were implemented to reconstruct ancestral areas
for all hadrosaur clades (Chapter 3, this dissertation). The parsimony analysis for reconstructing
ancestral areas was conducted in MacClade version 4.0 (Maddison and Maddison, 2003) and the
DIVA analysis was conducted in DIVA version 1.1 (Ronquist, 1997).
Revised Taxonomy of Secernosaurus koerneri and Kritosaurus australis
Dinosauria Owen, 1842

Ornithischia Seeley, 1888
Ornithopoda Marsh, 1881
Hadrosauria Von Huene, 1956
Hadrosauridae Cope,1869
Saurolophinae Brown, 1914
Secernosaurus Brett-Surman, 1979
S. koerneri Brett-Surman, 1979.
Synonymy. Kritosaurus australis Bonaparte et al., 1984; Secernosaurus australis
Wagner, 2001.
158
Holotype. FMNH PP13423: two dental battery fragments, partial basioccipital and
basisphenoid, two cervical centra, two neural arches of cranial dorsal vertebrae, one nearly
complete cranial dorsal vertebra, four centra (one with a neural arch) from middle dorsal
vertebrae, two centra and one fragment with postzygapophyses from caudal dorsal vertebrae, a
sacral rib, neural spine of caudal dorsal or sacral vertebra, five anterior and one posterior caudal
centra, three fragmentary ribs, nearly complete right scapula, distal end of the right humerus,
proximal ulnar fragment, right metacarpal III, nearly complete right ilium, partial left ilium, right
pubis, left and right ischial shafts, distal femoral condyle, and multiple small fragments of
unknown anatomical origin. FMNH PP13423 is probably a small subadult specimen, as noted by
Brett-Surman (1989) based on small body size, (TL ~4-5 m), unfused neural arches, and
disarticulated basisphenoid-basioccipital elements.
Referred material. MACN-RN 2 (originally the type of Kritosaurus australis): right
prefrontal, left postorbital, left and right sternals, right coracoid and scapula, left and right ilia,
left and right ischia lacking the distal half of each shaft, left and right pubes, left femur, proximal
right femur, a partial sacrum including four fused centra, several disarticulated sacral vertebrae
and ribs, and various middle caudal vertebrae. MACN-RN 142: fragmentary braincase with
parietal, frontal and postorbital, a right frontal, a fragment of predentary, right maxilla, rostral
regions of the dentaries, subadult left dentary, and a left scapula. MACN-RN 143: fragmentary
braincase including parts of the basioccipital, prootic and basisphenoid. MACN-RN 144: partial
skull roof and braincase including supraoccipital, parietal, frontals, postorbitals, and a fragment
of squamosal. MACN-RN 145: left radius and ulna, distal end of a femur, left metatarsals II and
III, and various middle dorsal vertebrae. MACN-RN 146: right scapula, left ilium lacking the
preacetabular process, fragmentary preacetabular process, proximal region of a left ischium, .
MACN-RN 826: right sternal, proximal half of a right scapula, left ilium lacking the pubic and
ischial peduncles, and various cervical and cranial and middle dorsal vertebrae. MACN-RN 987:
articulated series of distal caudal vertebrae. MACN-RN 990: pedal phalange III1. MACN-RN
991:fragment of left squamosal, fragment of right surangular. MACN-RN 997: right tibia.
MACN-RN 998: right dentary (heavily reconstructed with plaster). Based on the relative size
between these specimens and the mean size of larger hadrosaurid specimens from around the
world, it is likely that the Los Alamitos fossils represent individuals that are larger than the type
but are have not reached full adult size.
159
Locality and horizon. The holotype of Secernosaurus koerneri was most likely collected
from late Campanian-early Maastrichtian strata of the Upper Bajo Barreal Formation (see
discussion below). The label that accompanies the fossil material in the collections of FMNH
states that this material was found two miles to the east of the head of the Chico River. This
location is near the Lake Colhué Huapi and to the west of the Gulf of San Jorge (southeastern
Chubut province, Patagonia, southern Argentina) (Casal et al., 2007). The MACN specimens
(holotype and hypodigm of Kritosaurus australis) were collected from late Campanian-early
Maastrichtian strata from the middle section of the Los Alamitos Formation, near the Arroyo
Verde, southeastern Río Negro province, near the border with Chubut province (Patagonia,
southern Argentina) (Bonaparte et al., 1984).
Revised diagnosis. Hadrosaurid dinosaur that differs from all other hadrosaurids on the
basis of the following autapomorphies: V-shaped nasofrontal suture where the rostral margin of
each frontal forms a median triangular process and has a deep and concave recess to
accommodate the nasal bone; pubis with a prepubic process that differs from that of all other
hadrosauroids (except that of species of Gryposaurus and Prosaurolophus) in having an angular
dorsal margin at its maximum dorsoventral expansion; the prepupic process differs from that of
Prosaurolophus and Gryposaurus in having a ventral margin that forms a wider arch and having
a more gentle ventral deflection; ilium with the unique combination of these four characters:
dorsomedially twisted postacetabular process forming a ventromedially oriented brevis shelf;
median longitudinal ridge on the dorsal surface of the postacetabular process; postacetabular
process nearly as long as the iliac central plate and oriented caudodorsally; and craniocaudally
extensive supraacetabular process.
Osteological Description of Secernosaurus koerneri
Mandibular Complex of the Cranium

The predentary is represented by a rostral fragment (Fig. 4.2A-B). It is labiolingually
compressed and shows four nearly complete denticles. The rostral surface is flat. Medially, the
bone becomes slightly thicker labiolingually. Two additional denticles were broken at their bases
but their locations are seen medial to the other four denticles. The denticles are relatively tall,
slightly recurved lingually, labiolingually compressed, and are subrectangular. The denticles are
160
closely spaced with a separation of up to 25% basal length. The medial-most denticle appears
triangular and very tall, but both of these features result from erosion of their margins. The
lingual side of the predentary preserves part of the transverse lingual groove.
The most complete dentary, MACN-RN 142B (Fig. 4.2C), is missing the coronoid
process and most of the caudal region. A larger individual (MACN-RN 142; Fig. 4.2E-F) is
represented only by the symphyseal regions. (There is another dentary, MACN-RN 998, but it is
so heavily reconstructed that it is uncertain how much of the specimen is actual fossil bone and is
not utilized in the present description.) The proximal edentulous margin of the dentary is very
short in both specimens, with a length equivalent to 10-15% of the length of the dental battery.
The proximal edentulous margin forms an angle of approximately 140º from horizontal. The
ratio between the lingual projection of the symphyseal end of the dentary and the labiolingual
width of the dentary is approximately 2. The angle of deflection of the ventral margin of dentary
end is 20º relative to the horizontal. The symphyseal end gently curves lingually forming a wide
arch. In MACN-RN 142B there are ten visible alveoli. There would have been a minimum
number of 25 to 30 tooth positions based on an extrapolation of the length over which these
alveoli span to the the total length of the dentary. Each alveolus is narrow and vertically oriented.
There are at least four teeth per alveolus and these are arranged dorsoventrally. The height/width
ratio of dentary tooth crowns is approximately 2.8. There is a single ridge on the enameled side
of tooth crowns. This ridge is straight and occupies a median position in the crown. Marginal
denticles are small but are too poorly preserved to deduce their specific morphology. Both the
mesial and distal margins of the tooth have the same density and size of denticles.
The surangular is represented by a fragment containing the articulation surface for the
quadrate, part of the medial joint surface for the splenial and angular, and the proximal region of
the caudal process (Fig. 4.2H-I). The lateral flange that bounds rthe articulation surface for the
quadrate cotylus rostrolaterally is convex and expands laterally forming a wide semicircle. The
quadrate joint surface is concave and its medial wall faces laterodorsally. The ventral region of
the surangular is medially offset and bears the articular surface for the angular. This surface is
vertically inclined and dorsoventrally shallow. The ventral surface of the bone is rostrocaudally
convex and smooth.
161
Facial Complex of the Cranium
The maxilla is represented by a single exemplar. Its external surface is heavily eroded,
exposing the structure of the bone in several areas. It is subtriangular in lateral and medial views
and is mediolaterally compressed. The maxilla of Secernosaurus koerneri is characterized by a
deep rostrolateral region (Fig. 4.2A-B), a condition observed in species of Gryposaurus. The
lateral surface of the rostral end is eroded to the level of the dental battery and the dorsal
premaxillary surface of the rostroventral process is missing. The rostroventral region of the
maxilla is steeply inclined ventrally and forms an angle greater than 40º with the tooth row. The
dorsal process is not preserved. The rostrolateral promontory is located rostral to the mid-length
of the maxilla as in Gryposaurus. The lateroventral margin of the jugal articulation surface forms
a prominent ridge. This ridge is caudally continuous with the lateral emargination of the
ectopterygoid shelf, a condition observed in all Euhadrosauria. The ectopterygoid shelf
comprises approximately 40 to 45% of the length of the maxilla. In S. koerneri this shelf is
ventrally inclined 16 degrees relative to the horizontal long axis of the maxilla. An even more
inclined ectopterygoid shelf is found primitively in non-hadrosaurid hadrosaurs such as
Bactrosaurus johnsoni and Protohadros byrdi. The lateral ecopterygoid emargination in S.
koerneri is dorsoventrally thick and prominent. Adjacent and rostroventral to the jugal joint there
is a large ellipsoidal foramen. On the medial side of the maxilla, the row of alveolar foramina is
located within the dorsal half of the mid-dorsoventral depth of the element. The exact number of
alveolar positions in the dental battery is uncertain but a minimum number of 40 teeth can be
observed. The occlusal plane contains a maximum of two teeth arranged labiolingually (Fig.
4.2C), as in all other hadrosaurids.
The jugal is represented by a fragment that includes the orbital margin and the proximal
region of the rostral process (Fig. 4.2D-E). The apex of the rostral process of the jugal is missing.
The medial surface of the rostral process is bounded caudally by a sharp ridge. The orbital
margin forms a wide arch. The facet for the jugal in the maxilla is triangular in lateral profile.
This suggests that the rostral process of the jugal was also triangular and had a rostrally-pointed
apex.
The prefrontal is represented by a fragment that is missing most of the rostroventral
region (Fig. 4.3F). The lateral orbital margin is thicker caudally near the articulation with the
162
postorbital. Rostrally, the prefrontal curves rostroventrally so that the rostrodorsal margin of the
orbit was most probably subrectangular. The dorsal surface of the prefrontal is slightly concave.
The only known postorbital was found articulated to the partial skull roof MACN-RN
144. The postorbital of Secernosaurus koerneri has a broad central region (Fig. 4.5B). Its dorsal
surface is slightly convex, so that the dorsal margin is slightly arched in lateral profile. The
rostral and caudal borders of the orbital margin form an angle of 100º. The proximal region of
the jugal process is mediolaterally wide and forms an extensive concave surface that bounds the
orbital cavity caudally (Fig. 4.5A)
The squamosal is represented by a fragment that includes the quadrate cotylus and most
of the dorsal surface of the element. The caudal, rostral and medial processes are not preserved
(Fig. 4.3G). The dorsal surface of the squamosal is gently convex. The quadrate cotylus is deep
and rostrocaudally wide.
Neurocranial Complex of the Cranium
The frontal of Secernosaurus koerneri is unique among hadrosaurids with regard to the
nasal articulation surface. In S. koerneri, the rostral and median margin of the frontal contains a
pair of rostrolaterally-oriented processes (Figs. 4.4, 4.5A and 4.6). Together, these processes
form a V-shaped margin. Each process has a ventrally recessed surface that is rostrally offset
from the dorsal surface of the frontal. A bifid nasofrontal joint is also present in other
saurolophines such as Gryposaurus spp. and Edmontosaurus spp., and in hadrosaurid outgroups
such as Bactrosaurus johnsoni, Lophorhothon atopus and Tanius sinensis. However, the specific
morphology of this frontal region in S. koerneri, with its recessed surface and rostrally offset
apex, is autapomorphic for this species. Caudal to the nasofrontal articulation, the dorsal surface
of the frontal is nearly flat, and only become slightly concave laterally near the articulation with
the postorbital. Caudally and medially, the frontal rises slightly dorsally to meet the parietal. The
ectocranial surface of the frontal is relatively elongated rostrocaudally with a length/width ratio
greater than 0.8. On the ventral side of the frontal, the cerebral fossa is moderately deep and
bounded laterally by a thick border that contains the joint surfaces for the presphenoid and
orbitosphenoid (Fig. 4.8C). The annular ridge is low and gently rounded. The olfactory
depression is only partially preserved but it appears to have been as wide or wider mediolaterally
than the cerebral fossa (Fig. 4.8B-C).
163
The parietal is rostrocaudally elongated, with a length/width ratio greater than 2.35. This
condition is also found in other saurolophines. The parasagittal crest is nearly straight, with a
slight caudoventral inclination. The median process that articulates with the frontal is
rostrocaudally short and subtriangular.
The basioccipital (Figs. 4.5C, 4.6 and 4.7) is slightly rostrocaudally longer than it is
mediolaterally wide. The transversal suture with the basisphenoid is shown on the ventral side of
FMNH P13423 rostral to the spheno-occipital tubercles (Fig. 4.7D). The caudal two-fifths of the
basioccipital, which include the occipital condyle, is caudally shallow due to a ventral recess. It
is also distinctly narrower mediolaterally. Rostral to this area, the basioccipital is wider and
contains two eroded spheno-occipital tubercles. The long axis of each of these tubercles is
rostromedially directed. The spheno-occipital tubercles are separated by a median triangular
excavation that widens caudally (Fig. 4.7D). Laterally, the dorsal margin preserves the
articulation surface for the exoccipital and the ventral border of the vestibulocochlear foramen
(Fig. 4.7A-C).
Rostral to the basioccipital is the basisphenoid (Figs. 4.5C, 4.6 and 4.7). This bone is
incompletely preserved in all specimens. They lack the rostral region that supports the pterygoid
processes. Rostral to the sphenooccipital tubercles, the basisphenoid of S. koerneri lacks the
distinct mediolateral constriction that precedes the pterygoid processes in all other hadrosauroids.
Instead, the bone is mediolaterally wide and has a small and short knob-like lateral expansion at
mid-length of the rostrocaudal extension of the basisphenoid (Fig. 4.7D).
The postorbital process of the laterosphenoid, that projects laterally from the the
braincase, is thick and robust (Fig. 4.6). The basisphenoid process forms the anterior boundary of
the trigeminal foramen. The laterosphenoid does not enclose the ophthalmic sulcus that branches
out of the trigeminal foramen (Fig. 4.8D). The prootic and the postorbital processes form a
characteristic and extensively arched, concave and caudolaterally-facing surface. This is located
adjacent and caudal to the posterior limit of the orbital cavity.
Other Axial Elements
Vertebral elements are indistinct from those of all other hadrosaurids. The cervical centra
are opisthocoelous, mediolaterally broad and craniocaudally elongated (Figs. 4.9A-B and 4.10A-
B). Each parapophysis is large and oval, with its long axis directed craniocaudally.
164
Postzygapophyses are supported by large and massive postzygapophyseal processes (Fig. 4.9A-
B). These processes arch caudodorsally and diverge laterally from the neural arch.
The neural arches of the cranial dorsal vertebrae (Fig. 4.10E-F) are caudally tilted and the
bases of the neural spines are craniocaudally broad. Ventral and medial to the postzygapophyses
there are two large and deep excavations. The centra are craniocaudally longer than they are deep
and wide. The middle dorsals (Fig. 4.9C-D and 4.10G-H) have slightly opisthocoelous centra
with heart-shaped and equidimensional cranial and ventral surfaces. The lateral surfaces are
craniocaudally concave and converge ventrally into a keel-like structure. The transverse
processes are dorsolaterally and caudolaterally oriented. The centra of the more caudal dorsals
are craniocaudally narrower (Fig. 4.10G-H). Otherwise they are morphologically similar to those
of the middle dorsals.
Sacral vertebrae are represented by a partial sacrum (Fig. 4.9E) and disarticulated neural
arches with ribs and partially complete neural spines. The preserved portion of the sacrum
contains four fused centra. Each transverse process fuses ventrally with the iliac bar. This union
forms a thin wing of bone that stretches from the transverse processes ventrally and laterally to
the iliac blade. Sacral ribs are wing-like laminae that extend laterally between the transverse
processes and the centra.
The centra of the caudal vertebrae have cranial and caudal surfaces with hexagonal
contours (Fig. 4F-G). These centra are craniocaudally compressed and have concave lateral
surfaces. The postzygapophyses are small circular facets located on the caudal sides of the bases
of the neural spines. Transverse processes are narrow, rod-like, and horizontal. Neural spines are
mediolaterally thick and project caudodorsally. The neural canals are circular and relatively
small.
The sternal is formed by a paddle-like expanded craniomedial plate located at the end of
a long and elongated, “handle-like” caudolateral process (Fig. 4.9H). The medial side of the
sternal is slightly concave, particularly on the craniomedial plate. The caudolateral processt is
longer than the cranial plate. The process is mediolaterally compressed and ellipsoidal in cross
section. The caudoventral process is very prominent. In the more complete specimen (MACN-
RN 2) its length is greater than the width of the caudolateral process.
165
Pectoral Girdle
The coracoid (Fig. 4.11C) has a long and curved ventral process, with a length/width
ratio of 0.78. The humeral and scapular facets form an angle of 126º. The lateral margin of the
scapular facet is 17% longer than the lateral margin of the humeral facet. The craniomedial
margin of the coracoid is concave and ends medially into a prominent bicipital knob.
The acromion process of the scapula (Figs. 4.11A-B and 4.12A) is relatively large and
projects laterally. Dorsal to the acromion process, the surface of the bone merges rostrally into a
slightly concave surface, that connects to the coracoid facet. Caudal to the acromion process, the
lateral surface of the scapula has a pronounced and robust deltoid ridge. The narrowest region of
the scapula (“scapular neck”) amounts to 60% of the dorsoventral depth of the proximal end of
the element. The lateral surface of the proximal region of the scapular blade, caudal to deltoid
ridge, is dorsoventrally convex. The medial side is flat. The distal region of the scapular blade
reaches from 90 to 95% of the depth of the proximal region of the scapula.
Forelimb Elements
Forelimb elements in Secernosaurus koerneri are indistinct from those of other
hadrosaurids. The humerus is represented by a fragment of the right distal end in FMNH
PP13243 (Fig. 4.12D-E). The ulnar condyle is mediolaterally and craniocaudally broader than
the radial one. The condyles are separated by well-incised notches. More proximally the humerus
narrows into the shaft. The mediolateral diameter of the most proximal region preserved is about
60% as broad as the breadth of the bone across the distal condyles. A small isolated juvenile
humerus was also found in the Los Alamitos collection (Fig. 4.11D). In this exemplar the
deltopectoral crest is shorter than half the length of the humerus. The craniomedial region of the
crest is missing.
The lateral and medial flanges of the proximal region of the ulna (Figs. 4.11E and Fig.
4.12F-G) enclose a wide radial notch. The lateral flange is less prominent and its surface is
convex. Proximally and throughout most of its length, the shaft of the ulna has a triangular cross-
section. The olecranon process is subconical and prominent. The ulna thins progressively
distally, but it is slightly expanded at the distal end.
The radius (Fig. 4.11F) is subcylindrical in overall shape. It is as long as the ulna. The
proximal end is cup-like and has a flattened proximal surface. Distally the radius is also
expanded but more gradually and to a lesser than proximally.
166
The lateral margin of the proximal surface of metacarpal III (Fig. 4.12B-C) is slightly
longer and less subvertical than the medial margin. The dorsal surface is flat. The lateral and
medial sides contain longitudinal striations indicating where metacarpal II and IV would
articulate.
Pelvic Girdle
In the ilium (Figs. 4.13, 4.15A and 4.16A), the preacetabular process forms an angle of
145º with the plane containing the ventral margins of the pubic and ischial peduncles. The
proximal region of the preacetabular process is as deep as half the depth of the central plate of
the ilium. As in all hadrosaurids, the ventral apex of the supraacetabular process is located rostral
to the lateral ridge of the caudal protuberance of the ischial peduncle. The supraacetabular
process projects ventrally to a level that is between half and 75% of the depth of the iliac plate.
The process is craniocaudally broad, comprising 75% of the length of the central plate. The
supraacetabular process is asymmetrical with a caudally skewed V-shaped lateral profile. The
caudodorsal margin of the process forms a well-defined ridge that is continuous with the
proximodorsal margin of the postacetabular process. The pubic peduncle is triangular as in all
other Hadrosauridae. The ischial peduncle is formed by two protrusions. The caudal one ismore
dorsally located as in all other hadrosaurids. The postacetabular process is nearly as long as the
central plate. The most notable attribute of this process is that it is dorsomedially twisted and
forms a medioventral surface (Figs. 4.13, 4.17D and 4.18B) similar to the brevis shelf of various
hadrosaurid outgroup taxa, such as Bactrosaurus johnsoni and Tanius sinensis. Within
Hadrosauridae, this shelf is only seen in “hypacrosaur” lambeosaurines (e.g., Hypacrosaurus
altispinus and Velafrons coahuilensis), and in the Big Bend UTEP and the Salitral Moreno
hadrosaurids (Fig. 4.17). The laterodorsal surface of the postacetabular process of S. koerneri has
a median longitudinal ridge that is only shared by the Big Bend UTEP and the Salitral Moreno
hadrosaurids (4.18A and C).
The pubis of Secernosaurus koerneri is characterized by a long and rectangular prepubic
process (Figs. 4.15B-C and 4.16B and D). A similar morphology is found in Gryposaurus spp.,
Prosaurolophus maximus and Saurolophus spp. The process is ventrally deflected and not deeper
than the acetabular margin. The proximal constriction is as long as the distal expanded region of
the prepubic process. The point of maximum concavity of the dorsal margin of proximal
constriction is located approximately above that of the ventral margin. The ventral margin of the
167
prepubic process forms a very wide and open arch. The iliac peduncle is large and massive. As in
other hadrosaurids, it is tetrahedral in shape, with a concave caudolateral surface and slightly
convex cranial and medial sides. The ischial peduncle is very long as lender, with a length/width
ratio of nearly 4.5 in the Los Alamitos specimens (Fig. 4.16B and D) and 4.0 in the smaller type
exemplar (Fig. 4.15B). Such a long and slender ischial peduncle is shared by P. blackfeetensis
(e.g., TMP 84.1.1 and MOR 454; Appendix I, Table I.12).
In the ischium (Figs. 4.15D and 4.16C), the iliac peduncle is long, with a length/width
ratio of 2.3. The caudodorsal margin of the distal region of the iliac peduncle is slightly rounded.
Its distal articular facet forms an angle of 128º with the acetabular margin of the peduncle. The
ischial shaft is caudodorsally directed, forming an angle greater than 180º with the long axis of
the pubic peduncle.The ischial shaft is only represented in the type FMNH P13423 (Fig. 4.15D).
The ischial shaft is craniocaudally elongated and relatively thin. Proximally, the bone is
mediolaterally expanded and mediolaterally convex and is triangular in cross section. This
convexity is produced by a smooth dorsal ridge, which gradually disappears distally as the shaft
becomes mediolaterally compressed and rod-like. The medial side has longitudinal striations
indicating where the other ischial shaft joined. The preserved distal end is oval in cross section.
Hindlimb Elements
Like the forelimb and the axial skeleton, these elements in Secernosaurus koerneri are
indistinct from those of all other hadrosaurids. Accordingly, the femur is straight and columnar
(Fig. 4.19A-B). A lateral indentation separates the femoral head from the medial surface of the
greater trochanter that forms the lateral side of the proximal end of the femur. The fourth
trochanter comprises nearly 30% of the total length of the femur. The distal condyles are
mediolaterally compressed and are more caudally than cranially expanded. The medial condyle
is larger and mediolaterally thicker.
In the tibia (Fig. 4.19C-D), the cnemial crest extends distally into the cranial surface of
the proximal half of the element. The expanded proximal third of the tibia is cup-like in medial
outline and more than twice the diameter of the central shaft. The cnemial crest expands
anterolaterally forming the anterolateral side of the proximal end of the tibia. The two
caudomedial condyles are eroded away. The shaft of the tibia is slightly compressed
mediolaterally.
168
Metatarsal II (Fig. 4.19E and G) is mediolaterally-compressed. Distally, the element
becomes mediolaterally wider. This metatarsal is dorsoventrally expanded proximally and to a
lesser degree distally. The proximal surface is trapezoidal, while the distal face is strongly
compressed mediolaterally and dorsoventrally elongated, lenticular in outline. The distal surface
is cranially inclined. The ventral side forms a ridge along the proximal third of the metacarpal.
Metatarsal III (Fig. 4.19F and H) is the longer and more robust than metatarsal II. It is
dorsoventrally compressed and subrectangular at its mid-length cross-section. The proximal end
is mediolaterally expanded. The distal end is dorsoventrally expanded. The medial side is
concave proximally to articulate with the lateral, convex profile of metatarsal II. Proximally
there is a flange that protrudes ventrally. The proximal surface is crescentic and holds the convex
lateral side of the proximal surface of metatarsal II. The distal surface is subrectangular,
mediolaterally concave and dorsoventrally convex.
Phylogenetic Position of Secernosaurus koerneri
The parsimony and Bayesian analyses inferred that Secernosarus koerneri is a member of
the “kritosaur” clade within Saurolophinae. Unambiguous synapomorphies in support of the
inclusion of Secernosaurus koerneri within Saurolophinae are: dentary marginal denticles absent
or very reduced to small papillae along the apical half of the dorsal half of the crown (character
10-2); relatively long parietal, length/width ratio greater than 2.35 (character 155-2); relatively
long exoccipital-supraoccipital shelf above the foramen magnum, that is substantially longer
(often twice or more) than the diameter of the foramen magnum (character 171-2); long iliac
peduncle of the ischium with a length/width ratio greater than 2 (character 277-2); and pubic
peduncle of the ischium oriented parallel or nearly parallel to the ischial shaft (character 279-2).
Inclusion of S. koerneri within “kritosaurs” is supported by the following unambiguous
synapomorphies: rostral end of the rostrodorsal process of the nasal not reaching the rostral
margin of the narial foramen (character 87-0); rostroventral process dipping steeply ventrally and
forming an angle of 40º or greater with the tooth row: rostral region of the maxilla appearing
dorsally “inflated” and craniocaudally compressed (character 95-2); caudal end of the postorbital
caudal ramus extending to a point rostral to the quadrate cotylus without overlapping the latter
(character 141-0); and dorsal margin of the infratemporal fenestra located substantially more
169
dorsally than the dorsal margin of the orbit with the caudal region of the skull roof inclined
rostroventrally relative to the frontal plane (character 201-1).
Likewise, all the analyses recovered a clade composed of Secernosaurus koerneri and the
Salitral Moreno hadrosaurid (which was scored as an unnamed OTU) (see Chapter 2, Figs. 2.5
through 2.13). This is supported by the following unambiguous synapomorphies (characters and
character states numbers refer to the documentation shown in Appendices B through J): angle
between the lateral margins of the facet for scapular articulation and the glenoid greater than
115º (character 216-0; convergent in outgroup taxa to euhadrosaurs and “tsintaosaurs”,
Jaxartosaurus aralensis and Aralosaurus tuberiferus); asymmetrical iliac supraacetabular
process with a caudally skewed lateral profile (character 250-0; convergent in
Brachylophosaurus canadensis, Maiasaura peeblesorum, outgroup taxa to euhadrosaurs,
“tsintaosaurs”, A. tuberiferus and J. aralensis); caudodorsal margin of the lateroventral rim of the
supraacetabular process forming a well-defined ridge that is continuous with the dorsal margin of
the proximal region of the postacetabular process (character 252-0; convergent in Gryposaurus
latidens, Prosaurolophus maximus and all other iguanodontoideans, except remaining
saurolophines); and thick ischial shaft, thickness greater than 7.5% its length (character 282-2;
convergent in Bactrosaurus johnsoni, Hypacrosaurus altispinus, Lambeosaurus laticaudus,
Corythosaurus and Parasaurolophus cyrtocristatus).
In addition, the parsimony analyses inferred that the closest outgroup to the
Secernosaurus-Salitral Moreno OTU clade is the Big Bend UTEP OTU. This relationship is
supported by the following unambiguous synapomorphies: coracoid with slightly longer scapular
facet: ratio between the length of the lateral margin of the facet for the scapular articulation and
the length of the lateral margin of the glenoid greater than 1 and up to 1.30 (character 215-1;
convergent in Probactrosaurus, Lophorhothon atopus and the Sabinas OTU); brevis shelf at the
base of the postacetabular process of the ilium (character 256-1; convergent in the
iguanodontoidean outgroup taxa to clade 9, Tsintaosaurus spinorhinus and the “hypacrosaurs”);
well-developed medioventral ridge on the medial side of the postacetabular process that forms an
oblique and expanded flange forming the medial margin of an extensive brevis shelf that faces
more ventrally than medially (character 257-3); and presence of a craniocaudally-oriented
median ridge on the laterodorsal surface of the postacetabular process (character 258-1).
170
Historical Biogeography of Secernosaurus koerneri
The reconstruction of ancestral areas using Fitch parsimony and DIVA (Ronquist, 1996)
suggests that hadrosaurids occupied South America after a dispersal event from within
“kritosaurs”. This dispersal would have occurred no later than the late Campanian and is in
agreement with a similar conclusion reached by Weishampel and Weishampel (1983) and
Bonaparte et al. (1984). Kritosaurus navajovius and Gryposaurus spp., outgroup taxa to the
Secernosaurus clade (composed of the two South American taxa and the Big Bend OTU) lived
in western North America during the early Campanian-early Maastrichtian. According to the
parsimony analysis the ancestral area of the Secernosaurus clade was probably located in North
America. However, the DIVA analysis inferred a widespread distribution for the ancestor of this
clade in North and South America after which time the two Argentinean taxa would have
diverged after a vicariant event. Both of these scenarios are congruent with the distribution of the
outgroup taxon to the S. koerneri-Salitral Moreno clade in the late Campanian of Big Bend
National Park, Texas. The presence of this taxon in southern North America provides an
intermediate stage in the hypothetical dispersal of “kritosaurs” from western North America to
South America during the late Campanian.
The existence of a land connection between both American continents during the late
Cretaceous has recently been supported by several studies (Pough et al., 2004). In particular,
subduction of the southern part of the North American plate under the eastward-moving
Caribbean plate may have given rise to a Proto-Antillean volcanic arc (Duellman, 2001; Pindell
and Kennan, 2002; Hedges, 2006). This chain of islands would have been located between North
and South America in the position currently occupied by Central America, thereby providing a
dispersal route for hadrosaurids and other vertebrates during at least the Late Campanian
(Hedges, 2006). The Antilles are believed to have subsequently moved eastward during the
migration of the Caribbean plate (Duellman, 2001). By the Maastrichtian, this connection would
have been interrupted (Pindell and Kennan, 2002). This scenario is congruent with vicariance
being involved in the evolution of the Secernosaurus-Salitral Moreno clade in South America. A
similar vicariance scenario, whereby the Caribbean plate carried ancient Mesozoic biota
eastwards, has previously been posited to explain the evolution of various endemic terrestrial
vertebrates in the Antilles (Hedges, 2006).
171
Discussion
Provenance of the Holotype of Secernosaurus koerneri

In the original publication on Secernosaurus koerneri, Brett-Surman (1979) reported that
the type specimen of this hadrosaurid was collected from the San Jorge Formation, near the head
of the Chico River. However, Windhausen (1918) used the name San Jorge Formation to refer to
a marine sequence of strata of late Paleocene-early Eocene age. More recently, González Riga
and Casadio (2000) indicated that the only San Jorge Formation currently known in Argentina
does not contain fossils and is of early Paleozoic age. Thus, it seems unlikely that the type of S.
koerneri came from the San Jorge Formation in so far as it is currently defined.
Weishampel and Horner (1990) and Casal et al. (2007) indicated that Secernosaurus
koerneri came from the Bajo Barreal Formation, while Bonaparte (1996), Novas (1997) and
Case et al. (2000) pointed out that it came from the Laguna Palacios Formation. The late
Cretaceous strata of the San Jorge Basin, that comprises the region where S. koerneri was
originally found, is composed of the upper two formations of the Chubut Group, the Bajo Barreal
and the Laguna Palacios Formations. The Laguna Palacios Formation overlies the Bajo Barreal
Formation in the western region of the basin (Umazano et al., 2008). In the eastern side, the
Laguna Palacios Formation grades laterally into the Bajo Barreal Formation (Umazano et al.,
2008). Genise et al. (2002) pointed out that no dinosaur remains have been found in the Laguna
Palacios Formation and that those purportedly recovered from that unit were found in areas
where the latter is not exposed. That leaves the Upper Member of the Bajo Barreal Formation as
the most plausible provenance of the fossils. This view is supported by the finding of an isolated
hadrosaurid ilum in the Upper Member of the Bajo Barreal Formation, near the area where the S.
koerneri material was found (Luna et al. (2003). The age of this section of the Bajo Barreal
Formation has been estimated as Campanian-Maastrichtian (Casal et al., 2007).
Synonymy of Kritosaurus australis with Secernosaurus koerneri
Brett-Surman (1979) diagnosed Secernosaurus koerneri on the basis of the following
iliac characters: postacetabular process that is greatly deflected dorsomedially and elongate and
is unlike that of any hadrosaur or iguanodontian; preacetabular process deflected ventrally at an
angle greater than in Hadrosaurus (that he synonimized with Kritosaurus); and supraacetabular
172
process relatively smaller than in any hadrosaur of the same size. However, none of these
characters are autapomorphic for S. koerneri.
I agree with Bonaparte et al. (1984), Bonaparte (1996) and Wagner (2001) that the right
ilium of FMNH P13423 is distorted. At first glance, the ilium of S. koerneri appears to be unique
in having a long, tapering, and dorsoventrally compressed postacetabular process (Fig. 4.14).
However, the tapering shape of the postacetabular process is most likely the result of
deformation. Several signs of bone breakage support this possibility. The proximal half of the
lateral surface of the postacetabular process is collapsed inwards (medially) as if the process had
been subjected to dorsoventrally oriented compression (Fig. 4.14A). The distal half of the
process shows signs of medially and ventrally directed twisting. This is evidenced by a
longitudinal crushing line on the ventral surface that produces a concave ventral side (Fig.
4.14B). This deformation could have produced the tapering morphology. Unfortunately, the lack
of postacetabular process in the left ilium prevents an accurate estimation of the extent of the
deformation in this process. Still, the dorsomedial twisting and the craniocaudal elongation of the
postacetabular process are probably natural. However, I do not consider the elongation of the
postacetabular process as autapomorphic for S. koerneri. A very long postacetabular process
(i.e., more 90% the length of the central body of the ilium) is also found in Prosaurolophus
blackfeetensis, Saurolophus osborni and Gryposaurus spp. (character 255; Appendix I, Fig. I.17
of this dissertation). The dorsomedial twisting is probably an actual character (see below) that
has been accentuated by the post-depositional distortion of FMNH P13423.
The small size of the supraacetabular process is here attributed to the same post-
depositional deformation. In particular, the lateral surface of the central plate of the right ilium of
FMNH P13423 ventral to the supraacetabular process shows breakage marks sugegstive of the
process beeing latrodorsaly displaced (Fig. 4.14C, white arrows). The supraacetabular process in
the non-deformed left ilium of FMNH P13423 (Fig. 4.15A), although missing the lateral edge,
appears to have been as large and ventrally projected as in any member of Hadrosauridae. In the
absence of additional material in the Bajo Barreal collections, I shall assume that both the left
and right ilia of FMNH P13423 belong to the same individual and that the left supraacetabular
process preserves the actual condition in this species.
173
The ventral deflection of the preacetabular process is within the range of state one of
character 243 (angle less than 150º). This condition is present in all hadrosaurids and various
non-hadrosaurid hadrosaurs (Appendix I, Fig. I.1 of this dissertation).
Bonaparte et al. (1984) diagnosed Kritosaurus australis based on the “cranial features
observable in the dentition, pectoral and pelvic girdles that are coincident with those of the
different species of Kritosaurus, except for the more prominent denticles in the predentary and a
slight difference in the dorsal margin of the ilium, which is convex up to the middle of the iliac
plate and concave beyond that point, while in Kritosaurus spp. the inflexion point between the
convex and concave margins occurs more caudally” (p. 292). Characters coincident with
Kritosaurus (= Gryposaurus of this dissertation) were further explained (with little specification)
as follows: the curvature, cross section and length of the preacetabular process; the proportions
and morphology of the central plate of the ilium;the morphology and extension of the iliac
postacetabular process; the extension of the edentulous region of the dentary and the morphology
of the symphyseal region. However, the two characters that led Bonaparte et al. (1984) to erect a
new species for the Los Alamitos specimens were the cited prominence of the predentary
denticles and the inflextion point between the convexity and concavity of the dorsal margin of
the ilium being above the mid-length of the acetabulum. However, these two characters are not
autapomorphic of K. australis.
Predentary denticles being particularly developed and prominent in K. australis are not
unique to this species. For example, I have observed similarly large denticles in the specimen
MSNM V345 of Gryposaurus notabilis (Appendix C, Fig. C.4N). Likewise, predentary denticles
are usually eroded and incompletely preserved in the majority of hadrosaurid specimens,
especially in those including complete articulated skeletons (Appendix C, character 27). This
probably represents a preservational bias against finding large predentary denticles in other taxa.
As for the location of the inflexion point between the concave and convex profiles of the
dorsal margin of the ilium, it is also found above the mid-length of the acetabular margin in other
hadrosaurids. Examples are Brachylophosaurus canadensis (MOR 794), Hypacrosaurus
altispinus (AMNH 5204), Edmontosaurus annectens (DMNH 1943), Maiasaura peeblesorum
(ROM 44770) and Gryposaurus notabilis (ROM 764).
Various morphological attributes that are unique to and shared by Secernosaurus
koerneri and Kritosaurus australis led me to regard the latter as a junior synonym of the Chico
174
River species. The type and hypodigm of K. australis are indistinct from the type of S. koerneri.
The pubis of FMNH P13423 is morphologically identical to those of MACN-RN 2, including the
lateral profile of the distal prebupic process, the long proximal constriction and the long and
slender of the ischial peduncle. The geometry of the prepubic process is the only postcranial
character that can be diagnostic to generic and specific levels (e.g., Edmontosaurus annectens;
see character 265 in Appendix I, Figs. I.28-I.30, this dissertation).
The ilium of S. koerneri is also indistinct from those of K. australis. In both taxa, the
ilium has a craniocaudally extensive and asymmetrical supraacetabular process. The
postacetabular process is dorsomedially twisted and shows a brevis shelf on its ventromedial
surface. The dorsal surface of the process has a median longitudinal ridge. These three iliac
characters are only found in these two taxa (and in ilia recovered from the Salitral Moreno
bonebed that future studies may show to belong to S. koerneri). The only other hadrosaurid with
an ilium having a median ridge on the dorsal surface of the postacetabular process and brevis
shelf originating from dorsomedial twist of this process is the unnamed hadrosaurid from the
UTEP collection from Big Bend National Park, Texas (Davis, 1983). However, the latter taxon
has a craniocaudally shorter supraacetabular process and the postacetabular process is
horizontally oriented, not caudodorsally projected as in S. korneri.
Conclusions
A re-evaluation of the osteology of the types and referred materials of Kritosaurus

australis and Secernosaurus koerneri led me to conclude that the former is a junior synonym of
the latter. The results of a phylogenetic analysis of S. koerneri with a nearly complete taxonomic
sampling of Hadrosauridae did not support the inclusion of this species within the genus
Kritosaurus. However, S. koerneri was inferred to be a member of the “kritosaur” clade of
Saurolophinae as the sister taxon of the Argentinean OTU from Salitral Moreno. Another
unnamed hadrosaurid, an OTU from Big Bend National Park, Texas, was positioned as the
closest outgroup to the South American clade. The reconstruction of the ancestral area for the
South American hadrosaurid clade using Fitch parsimony and Dispersal-Vicariance analysis
supported the previous hypothesis that posited a dispersal from North to South America no later
than the late Campanian.
175
Neuquén Cipoletti
General Roca
SALITRAL MORENO Negro River

El Cuy (Allen Fm.)
RIO NEGRO
Arroyo Salado
San Carlos SECERNOSAURUS
de Bariloche (Los Alamitos Fm.)
Arroyo Verde
Cona Niyeu
CHUBUT
Chubut River
100 km
Chico River
Lake
Colhué Huapi
Lake Muster
Gulf of San Jorge
Sarmiento
Comodoro Rivadavia
SECERNOSAURUS
(Bajo Barreal Fm.)
Figure 4.1. Geographical location of the most complete and abundant remains of hadrosaurid
dinosaurs found in South America. The map shows two provinces of southern Argentina, Río Negro
and Chubut. Redrawn after Bonaparte et al. (1984) and Salgado et al. (2007).
176
Denticles
2.5 cm
Rostral surface
A B
5 cm
Alveolar sulci
Predentary joint
Meckelian groove
Alveoli and tooth crowns

C
2.5 cm
2.5 cm
E
Symphysis
5 cm
D F
Tooth crowns
Quadrate joint surface

H
2.5 cm
2.5 cm G I
Figure 4.2. Mandibular elements of Secernosaurus koerneri. A and B, predentary, MACN-RN 142, in rostral
and caudal views, respectively. C, dentary, MACN-RN 142B, in medial view. D, detail of tooth crowns,
MACN-RN 142. E and F, dentary rostral fragment, MACN-RN 142, in medial and rostral views. G, fragment
of dentary dental battery, MACN (uncatalogued), in medial view. H and I, surangular, MACN-RN 991,
in dorsal and lateral views,respectively.
177
Jugal joint
5 cm
Ectopterygoid shelf
A
Ectopterygoid ridge
Alveolar sulci
Alveolar foramina
B
5 cm
Ectopterygoid ridge
Dentition C
Orbital margin
Orbital margin
2.5 cm
Maxillary facet
Rostral
process
D F
2.5 cm
Maxillary facet E 2.5 cm

G
Quadrate
cotylus
Figure 4.3. Facial elements of Secernosaurus koerneri. A, B and C, right maxilla, MACN-RN 142,
in lateral, medial and ventral views, respectively. D and E, rostral fragment of right jugal, MACN
(uncatalogued), in medial and dorsal views, respectively. F, prefrontal, MACN (uncatalogued), in lateral
view. G, squamosal fragment MACN-RN 991, in lateral view.
178
Frontal (nasal joint)
Frontal
Postorbital
Parietal
Laterosphenoid
Supratemporal
fenestra
5 cm
Squamosal
Supraoccipital
Exoccipital
Figure 4.4. Skull roof and braincase of Secernosaurus koerneri, MACN-RN 142, in dorsal view.
179
Parietal
Frontal
Postorbital
Orbitosphenoid
Laterosphenoid
5 cm
A
Squamosal Opisthotic-Exoccipital
5 cm
Paroccipital
process
Postorbital
Frontal
Exoccipital
Paroccipital
process
Foramen magnum
5 cm
Basioccipital
Basisphenoid C
Figure 4.5. Skull roof and braincase of Secernosaurus koerneri, MACN-RN 142. A, rostral, B, lateral
and C, caudal views.
180
Frontal
Presphenoid
Orbitosphenoid
Postorbital
Laterosphenoid
Basisphenoid
5 cm
Basioccipital
Exoccipital
Figure 4.6. Skull roof and braincase of Secernosaurus koerneri, MACN-RN 142, in ventral view.
181
Figure 4.7. Secernosaurus koerneri, fragment of basioccipital and basisphenoid, holotype
FMNH P13423, in A. dorsal, B, caudal, C, lateral and D, ventral views.
182
Olfactory Olfactory
Frontal Frontal depression depression
Presphenoid
joint
surface Annular ridge
Orbitosphenoid
joint
Cerebral
surface
fossa
2.5 cm Cerebral
Parietal fossa
Parietal
A B 2.5 cm C
Prootic
Opisthotic-
Exoccipital Laterosphenoid
2.5 cm
V
V1
VII
V3
2.5 cm basisphenoid
D E F
Figure 4.8. Neurocranial elements of Secernosaurus koerneri. A and B, frontal and parietal, MACN-RN
142, in dorsal and ventral views, respectively. C, right frontal, MACN (uncatalogued), in ventral view.
D, braincase fragment, MACN-RN 143, in left lateral view. E and F, supraoccipital, MACN-RN 144, in
right lateral and caudodorsal views, respecively.
183
Postzygapophyseal Prezygapohysis
process Neural spine
Neural spine
5 cm
Transverse
5 cm process
5 cm Postzyga-
pophysis
A Postzygapophyseal B
process
Prezygpohysis
10 cm
C D
E
Centra
Proximal craniomedial
plate
5 cm
Prezygapophysis 5 cm
Caudolateral process
Neural spine (”handle”)
Caudoventral process
F G H
Figure 4.9. Axial elements of Secernosaurus koerneri. A and B, cervical vertebra, MACN-RN 826,
in lateral and dorsal views, respectively. C and D, dorsal vertebra, MACN-RN 826, in cranial and lateral
views. E, partial sacrum, MACN-RN 2, in lateral view. F and G, caudal vertebra, MACN-RN 2, in lateral
and caudal views. H, sternal plate, MACN-RN 826, in ventrolateral view.
184
Figure 4.10. Axial elements of the holotype of Secernosaurus koerneri, FMNH P13423. A and B, cervical
centrum in right lateral and ventral views, respectively. C and D, dorsal neural spine in caudal and lateral
views, respectively. E and F, cranial dorsal vertebra in caudal and lateral views, respectively. G and H, dorsal
centrum in cranial and ventral views, respectively. I, J and K, cranial dorsal vertebra in right lateral, cranial
and dorsal views, respectively. L and M, caudal centrum in cranial and right lateral views, respectively.
185
Acromion process
10 cm
Coracoid Deltoid ridge

facet
Scapular blade
A
Acromion process
Gleonoid Deltoid ridge
Coracoid
10 cm facet
Scapular facet
Bicipital sulcus
Bicipital
knob 5 cm B
Humeral Gleonoid
facet
Deltopectoral
5 cm crest
Ulnar condyle
(proximal region)
C D
Radial condyle
Ventral process
Olecranon process
10 cm E
10 cm F
Figure 4.11. Pectoral and forelimb elements of Secernosaurus koerneri. A, left scapula, MACN-RN 142,
in lateral view. B, right scapula, MACN-RN 146, in lateral view. C, coracoid, MACN-RN 2, in lateral
view. D, juvenile humerus, MACN-RN 980, in craniomedial view. E, left ulna, MACN-RN 145, in lateral
view. F, left radius, MACN-RN 145, in dorsal view.
186
Figure 4.12. Pectoral and forelimb elements of the holotype specimen of Secernosaurus koerneri,
FMNH P13423. A, left scapula in lateral view. B and C, metacarpal III in dorsal and proximal views,
respectively. D and E, distal end of a right humerus in caudal and lateral views, respectively. F and G,
proximal region of a left ulna in cauda and medial views, respectively.
187
Figure 4.13. Right ilium of the juvenile holotype specimen of Secernosaurus koerneri, FMNH
P13423, in A, lateral, B, dorsal, C, caudal and D, medial views.
188
Figure 4.14. Detail views of the holotype ilium of Secernosaurus koerneri, FMNH P13423, showing the
areas with signs of postdepositional deformation (arrows). A, postacetabular process in lateral view.
B, postcetabular process in caudoventral view. C, supraacetabular process in lateroventral view.
189
Figure 4.15. Pelvic elements of a juvenile specimen of Secernosaurus koerneri, holotype FMNH
P13423. A, left ilium in lateral view. B and C, right pubis in lateral and ventral views, respectively.
D, lateral view of the shaft of the ischium.
190
Supraacetabular
Postorbital process
process
Preacetabular process
10 cm
Ischial peduncle
Pubic peduncle
Acetabular margin A
Iliac peduncle Prepubic process
Acetabular margin
Ischial peduncle 10 cm
Iliac process
B
10 cm Acetabular margin
Pubic process
Obturator process C
Prepubic process
Iliac peduncle
Acetabular margin
10 cm
Ischial process
Figure 4.16. Pelvic elements of Secernosaurus koerneri. A, left ilium, MACN-RN 2, in lateral view.
B, right pubis, MACN-RN 2, in lateral view. C, proximal half of right ischium, MACN-RN 2, in lateral
view. D, left pubis, MACN-RN 2, in lateral view.
191
BS
5 cm
5 cm
5 cm
BS
5 cm
BS
Figure 4.17. Presence and absence of brevis shelf in the postacetabular process of four species of
hadrosaurs; images shown in ventral view. A, Bactrosaurus johnsoni, SBDE 95E. B, Brachylophosaurus
canadensis, MOR 1071-8-2-98-469. C, Secernosaurus koerneri, MACN-RN 2. D, Hypacrosaurus
altispinus, AMNH 5204. Abbreviations: BS = brevis shelf.
192
2.5 cm
5 cm
5 cm
A B C
Figure 4.18. Absence and presence (rectangles) of the dorsal ridge of the postacetabular processes of the ilium in three species of
hadrosaurids. A, Hypacrosaurus altispinus, AMNH 5204. B, Secernosaurus koerneri, MACN-RN 2. C, Big Bend UTEP hadrosaurid.
193
Femoral Lateral condyle
head
Greater
trochanter
Cnemial
10 cm crest
10 cm
Fourth
trochanter
Medial
condyle
Medial
External
condyle
malleolus
Lateral condyle
A B C D
5 cm
5 cm 5 cm
E
5 cm
F G H
Figure 4.19. Hindlimb elements of Secernosaurus koerneri. A and B, left femur, MACN-RN 2, in
cranial and medial views, respectively. C and D, tibia, MACN-RN 997, in lateral and caudal views,
respectively. E and G, left metatarsal II, MACN-RN 145, in medial and dorsal views, respectively.
F and H, left metatarsal III, MACN-RN 145, in medial and dorsal views, respectively.
194
CHAPTER 5
PARARHABDODON ISONENSIS AND TSINTAOSURUS SPINORHINUS: A NEW
CLADE OF LAMBEOSAURINE HADROSAURIDS FROM EURASIA
Abstract
Lambeosaurinae is one of the most diverse and derived clades of ornithopods. Although
well represented in Asia and North America, the presence of this clade in Europe has been
contentious. This is because their remains are rare and fragmentary, and its phylogenetic
implications poorly understood.
The observation of previously unrecognized characters in the maxilla and dentary of
Tsintaosaurus spinorhinus (Campanian of China) and Pararhabdodon isonensis (Maastrichtian
of Spain) led to a revision of the taxonomy and phylogenetic relationships of the latter. In
particular, the extreme elongation of the symphyseal region of the type and only material of
Koutalisaurus kohlerorum (a dentary from the Maastrichtian of Spain) was also observed in T.
spinorhinus. This implied that K. kohlerorum is indistinct from T. spinorhinus. This, in
combination with the fact that P. isonensis and T. spinorhinus share a maxilla with elevated jugal
joint continuous with the ectopterygoid ridge (a character not seen among Iguanodontoidea), led
me to the conclusion that K. kohlerorum as a junior synonym of P. isonensis.
The incorporation of those new characters in Bayesian and parsimony phylogenetic
analyses of Hadrosauridae resulted in the inference that Pararhabdodon isonensis and
Tsintaosaurus spinorhinus form a clade of basal lambeosaurines—the “tsintaosaurs”. Fitch
parsimony and Dispersal-Vicariance reconstruction of ancestral areas on the resulting phylogeny
indicated that “tsintaosaurs” originated in eastern Asia during the middle or late Campanian and
that P. isonensis represents a dispersal event (followed by vicariance) to Europe that occurred
before the end of that geologic stage.
Introduction
Lambeosaurine hadrosaurids comprise one of the most derived clades of ornithopod

dinosaurs (Horner et al., 2004). Lambeosaurines stand out among ornithopod dinosaurs in having
195
caudodorsally located nasal passages (Weishampel, 1981; Wagner, 2004) and pronounced
supracranial crests, the functions of which may have been for species recognition and sexual
display (Hopson, 1975; Evans, 2006). The fossil record of these large herbivores in North
America and Asia is among the best for non-avian dinosaurs and attests to their great diversity
(Lund and Gates, 2006). In contrast, European remains of Hadrosauria are quite rare (Lopez-
Martinez et al., 2001; Prieto-Marquez et al., 2006b) and have only allowed the erection of two
species, Telmatosaurus transsylvanicus (middle Maastrichtian Sinpetru and Densus Ciula
Formations of Romania; Weishampel et al., 1993) and Pararhabdodon isonensis (late
Maastrichtian of the Tremp Formation of Spain; Casanovas-Cladellas et al., 1993).
The phylogenetic relationships of Pararhabdodon isonensis have been contentious. It is
the only known European lambeosaurine hadrosaurid? (Casanovas et al., 1999) or another
outgroup to Hadrosauridae such as Telmatosaurus transsylvanicus (Head, 2001; Prieto-Marquez
et al., 2006b). Resolution of this question is important for a global understanding of the
phylogenetic relationships and historical biogeography of Lambeosurinae. If P. isonensis is not a
member of Lambeosaurinae, then it would stand that the lambeosaurine radiation (see Chapter 2,
this dissertation) did not spread into Europe and remained restricted to Asia and North America.
An isolated hadrosaurid ilium, collected from the Upper Maastrichtian Blasi 3 locality
from the Conques Formtion, near the town of Aren (Huesca province, northern Spain), was
referred to Lambeosaurinae by Cruzado-Caballero et al. (2005; 2007). This referral was based on
the arcuate and ventrally deflected preacetabular process, a supraacetabular process that projects
lateroventrally to mid-height of the central iliac plate, and the relatively deep proportions of the
ilium (Cruzado-Caballero et al., 2005). However, the character analysis conducted in this
dissertation (Chapter 2 and Appendices B through J) show that these attributes are not diagnostic
of Lambeosaurinae. The angle of ventral deflection of the preacetabular process is pronounced
(angle of 150º or less) in all hadrosaurids and various non-hadrosaurid hadrosaurs; it is low
(angle greater than 150º) in a few non-hadrosaurid iguanodontoideans such as Claosaurus agilis,
Probactrosaurus gobiensis, Equijubus normani, Ouranosaurus nigeriensis and some (but not all)
specimens of Iguanodon spp. (character 243; Appendix I, Table I.1 and Fig. I.1). A
supraacetabular process that is ventrally expanded to middle of the iliac central plate is a
hadrosaurid synapomorphy (Chapter 2 of this dissertation, Fig. 2.14). Finally, when Cruzado-
Caballero et al. (2005) calculated the length/depth ratio of the ilium, they also included the
196
lengths of the preacetabular and postaccetabular processes in their measurements. Thus, their
ratio is dependent upon the proportions of the latter processes and does not exclusively reflect
the depth of the ilium (which consists of that of its central plate). I calculated the length/depth
ratio restricting the measurements to the proportions of the central plate (character 246;
Appendix I, Table I.5 and Fig. I.5). I found that, although a relatively deep ilium (ratio greater
than 0.8) is indeed present in most lambeosaurines (Parasaurolophus cyrtocristatus being the
exception), it is not exclusive to this clade. A deep ilium is also present in the saurolophine
Brachylophosaurus canadensis and in all sampled non-hadrosaurid Iguanodontoidea. Thus, in
the absence of additional lambeosaurine characters, the depth of the ilium does not by itself
allow referral to Lambeosaurinae.
This study aims to unambiguously establish the phylogenetic relationships of
Pararhabdodon isonensis and clarify its taxonomy. In doing so, I ask whether lambeosaurine
hadrosaurids did in fact colonize Europe. If they did, I would further infer how these European
lambeosaurines relate to all other forms from Asia and North America, and how they came to
occupy the former continent.
Taxonomic History of Pararhabdodon isonensis
Originally, Pararhabdodon isonensis was referred to Iguanodontia incertae sedis by
Casanovas-Cladellas et al. (1993). Subsequently, Casanovas and colleagues (1993) regarded P.
isonensis as the most basal lambeosaurine. These authors based this conclusion on the presence
in P. isonensis of a sacrum with a haemal sulcus and a moderately developed distal expansion of
the ischium. However, haemal sulci are present in nearly all iguanodontoidean species and
within some taxa (e.g. Bactrosaurus johnsoni) are variable (Appendix G, Table G.2). The second
feature, a moderately expanded ischial expansion, is also found in non-hadrosaurid
iguanodontians (Appendix I, character 283).
In a recent revision of the osteology and phylogenetic position of P. isonensis, Prieto-
Marquez et al. (2006b) concluded that this species is not a lambeosaurine after all but a non-
hadrosaurid member of Hadrosauria. Their phylogenetic analysis reconstructed P. isonensis as
the sister taxon to the clade composed of Hadrosaurinae and Lambeosaurinae (Hadrosauridae
sensu Prieto-Marquez, 2006b; Euhadrosauria sensu Weishampel et al., 1993 and Prieto-Marquez,
Chapter 2 of this dissertation). This relationship was supported by three unambiguous
synapomorphies: a maxilla with three or four foramina, 35 to 40 alveolar positions in the
197
maxillary dental battery, and cervical vertebrae with strongly arched postzygapophyseal
processes.
Among the material referred to Pararhabdodon isonensis by Casanovas et al. (1999) is
an isolated dentary, IPS SRA 27 (Fig. 5.1A and B). This element was found in the Les Llaus
locality that is stratigraphically and geographically separated from the other localities that have
yielded P. isonensis material. This fact, added to the absence of dentaries from those localities,
prevented Prieto-Marquez et al. (2006b) from referring IPS SRA 27 to P. isonensis.
Additionally, they observed that IPS SRA 27 has what then appeared to be a unique character
nowhere found among Hadrosauridae or any other member of Iguanodontoidea. Specifically, it
has an edentulous region that is not only very elongated, and additionally shows extreme medial
projection. The projection is so severe that the distance between the symphysis and the lateral
wall of the element is three times its mediolateral breadth (Fig. 5.1). The presence of this dentary
character and its being found in isolation led to Prieto-Marquez et al. (2006b) erecting a new
genus and species, Koutalisaurus kohlerorum based exclusively on the dentary attributes.
I re-examined all the materials referred to Pararhabdodon isonensis and the holotype of
Koutalisaurus kohlerorum. These are housed at the Institut Català de Paleontologia (former
Institute of Paleontology “Miquel Crusafont”, Sabadell, Barcelona, Spain) and the Museu de la
Conca Dellà (Isona, Lleida, Spain).
The phylogenetic position of Pararhabdodon isonensis was reassessed using a nearly
complete taxonomic sampling of hadrosaurid dinosaurs. (Unless otherwise indicated, all of the
following anatomical comparisons between the aforementioned materials and those of other taxa
were based on direct examinations of the specimens.) This analysis included 12 outgroup and 39
ingroup (hadrosaurid) taxa. Outgroup taxa consisted of 2 non-hadrosaur iguanodontoideans and
10 non-hadrosaurid hadrosaurs. Species of Iguanodon and Ouranosaurus nigeriensis that are
iguanodontoideans distantly related to hadrosaurids were used to root the tree by outgroup
comparison (Wiley, 1981; Maddison et al., 1984). The searches for the optimal tree(s) using
maximum parsimony were carried out in PAUP version 4.0b10 (Swofford, 2002). Two
methodologies were implemented. In the first one, parsimony was applied with all of the
198
characters having the same weight. In the second I applied the weighted parsimony of Goloboff
(1993), with a k fit parameter of 3. I performed heuristic searches using a random additional
sequence of 10,000 replicates for the unweighted parsimony and 5,000 replicates for the
weighted parsimony analysis. I used the option of branch-swapping by tree-bisection-
reconnection in both parsimony analyses (TBR) (Swofford et al., 1996a). Bremer support
(Bremer, 1988) was assessed by computing decay indices (Donoghue et al., 1992) with
MacClade version 4.0 (Maddison and Maddison, 2003) and PAUP. Bootstrap proportions
(Felsenstein, 1985) were computed in PAUP. Two bootstrap settings were used. For the
unweighted parsimony analyses the bootstrap was set to 10,000 replicates using heuristic
searches where each was performed using random additional sequences with branch-swapping
by TBR and 1,000 replicates. For the weighted parsimony analyses the bootstrap was set to 5,000
replicates using heuristic searches where each search was conducted using random additional
sequences with branch-swapping by SPR (Subtree Pruning and Regrafting) and 25 replicates. I
also conducted Bayesian analyses using the maximum likelihood model of Lewis (2001)
(Markov k or Mk) for discrete morphological data and implemented in MrBayes version 3.1.2
(Huelsenbeck and Ronquist, 2001). Two different models were applied. In a first model, I
assumed equal rates of change among characters. In the second model I incorporated the gamma
parameter to allow for variable rates of character change (Nylander et al., 2004). Priors other
than the default for the models were not specified. The analysis without gamma used six chains
and was run for 10,000,000 generations that were sampled every 100 generations. Stationarity
was achieved with a split frequency standard deviation of 0.005. Trees generated during the first
100,000 generations were discarded as “burn-in”. The analysis with gamma used four chains and
4,000,000 generations sampled every 100 generations, reaching stationarity with a split
frequency standard deviation of 0.008. Trees generated during the first 100,000 generations were
also discarded as “burn-in”. Ancestral states were reconstructed using parsimony in MacClade
version 4.0 (Maddison and Maddison, 2003).
Inference of the historical biogeography of Pararhabdodon isonensis in the context of all
other Hadrosauria required the reconstruction of ancestral areas. These were reconstructed on a
weighted parsimony phylogeny derived from the latest and most comprehensive phylogenetic
analysis of Hadrosauridae (Chapter 2, this dissertation). Fitch parsimony (Fitch, 1971) and the
Dispersal-Vicariance (Ronquist, 1996) method were implemented to reconstruct ancestral areas
199
for all hadrosaur clades (Chapter 3, this dissertation). The parsimony analysis for reconstructing
ancestral areas was conducted in MacClade version 4.0 (Maddison and Maddison, 2003) and the
DIVA analysis was conducted in DIVA version 1.1 (Ronquist, 1997).
Results
Anatomical Observations
During my examinations of the collections of the Institute of Vertebrate Paleontology and
Paleoanthropology (IVPP), I came across some of the dentaries (IVPP V725 and V723) referred
by Young (1958) to Tsintaosaurus spinorhinus, from the Wangshi Group of Shandong, China
(Fig. 5.1C and D). Notably, these elements share the same edentulous morphology seen in IPS
SRA 27 (i.e., extensive elongation and medial projection of the symphyseal region). This
similarity implies (assuming they are homologous) that: 1) Koutalisaurus kohlerorum cannot be
distinguished from T. spinorhinus; 2) the holotype and only material of K. kohlerorum is non-
diagnostic; and 3) K. kohlerorum is nomen dubium.
Additional comparisons between the maxilla of Pararhabdodon isonensis (IPS SRA 22;
Fig. 5.2E) and Tsintaosaurus spinorhinus (IVPP V725; Fig. 5.2F) showed that these two species
share unique characters that, as in the case of the dentary, are not present in any other known
hadrosaur. These characters include: 1) a greatly elevated jugal joint (with a distance between the
ventral margin of this joint and the lateral ridge of the ectopterygoid shelf that is at least as great
as the dorsoventral depth of the lateral wall of the caudal region of the maxilla) and 2) a concave
lateral surface of the maxilla ventral to the jugal joint. Prieto-Marquez et al. (2006) considered
these two characters as autapomorphies of P. isonensis.
The overall pattern that emerged from my examinations is as follows. The hadrosaurid
bones from the Campanian Wangshi Group of Shandong, China include dentaries with long
medially projected edentulous regions and maxillae with highly positioned jugal joints. The
hadrosaurid fossils from the late Maastrichtian of northern Spain contains anatomically
comparable elements. How can this be reconciled? I believe that given the great temporal and
spatial distance between these two associations, it is safe to assume that the dentaries and
maxillae are referable to a single taxon in each case: Pararhabdodon isonensis in Spain and
200
Tsintaosaurus spinorhinus in China. Therefore, the holotype dentary of K. kohlerorum is
referrable to P. isonensis; K. kohlerorum becomes a junior synonym of P. isonensis.
These findings led to a reassessment of the taxonomy and phylogenetic relationships of
Pararhabdodon isonensis.
Revised Taxonomy of Pararhabdodon isonensis
Dinosauria Owen, 1842
Ornithischia Seeley, 1888
Ornithopoda Marsh, 1881
Hadrosauria Von Huene, 1956
Hadrosauridae Cope, 1896
Lambeosaurinae Parks, 1923
Pararhabdodon Casanovas-Cladellas, Santafé-Llopis, and Isidro-Llorens, 1993
P. isonensis Casanovas-Cladellas, Santafé-Llopis, and Pereda-Suberbiola, 1997
Holotype. IPS SRA 1: nearly complete middle-caudal cervical vertebra.
Paratype. IPS SRA 15: left humerus. IPS SRA 16: proximal fragment of left scapula. IPS
SRA 18: cranial or middle cervical vertebra.
Synonymy. Koutalisaurus kohlerorum (Prieto-Marquez et al. 2006b)
Referred specimens. MCD 4730: cranial cervical vertebrae. IPS SRA 25: middle cervical
vertebra. IPS SRA 18: caudal middle cervical vertebra. IPS SRA 13: centrum of dorsal
vertebra. MCD 4731: cranial dorsal vertebra. IPS 693-13, IPS SRA 12 and 20: caudal middle
dorsal vertebrae. IPS SRA 17: complete caudal vertebra. IPS SRA 22: nearly complete right
maxilla. IPS SRA 23: caudal half of left maxilla. IPS SRA 24: nearly complete sacrum. IPS 693-
12: fragment of proximal rib. IPS SRA 26: distal end of right ischium. IPS SRA 27: left dentary
missing the teeth and with partially preserved coronoid process; this dentary is the former
holotype and only known material of Koutalisaurus kohlerorum (Prieto-Marquez et al., 2006b).
Locality and age. Sant Romà d’Abella (near the town of Isona) and Les Llaus (near the
town of Abella de la Conca), Lleida province, northeastern Spain. Both localities belong to the
Tremp Formation, Upper Cretaceous (Upper Maastrichtian)
Revised diagnosis. Lambeosaurine hadrosaurid that differs from all other Hadrosauridae
and Hadrosauria in the possession of a maxilla with a jugal joint that is elevated above the level
of the lateral ridge of the ectopterygoid shelf and connected with this latter by a concave caudal
201
border. Pararhabdodon isonensis differs from Tsintaosaurus spinorhinus (the only other taxon
known to have this type of jugal joint) in having a much wider, nearly subrectangular
rostrodorsal region of the maxilla.
Revised Phylogenetic Position of Pararhabdodon isonensis
The parsimony and Bayesian analyses inferred a clade composed of Pararhabdodon
isonensis and Tsintaosaurus spinorhinus (see Chapter 2, Figs. 2.5 through 2.13). This clade, that
I informally refer to as “tsintaosaurs” for heuristic purposes diverged after Jaxartosaurus and
Aralosaurus. “Tsintaosaurs” are the outgroup clade to all remaining lambeosaurines. The clade is
supported by the following unambiguous synapomorphies: great medial projection of the
symphyseal region of the dentary with a ratio between the labiolingual extension of the rostral
end and the maximum labiolingual width of the dentary greater than 2.85 (character 42-2);
rostral edentulous region of the dentary for articulation with the predentary ranging from having
a very subtle concavity (almost straight) to straight or even displaying a subtle convexity
(character 45-0; convergent in Ouranosaurus nigeriensis, the Sabinas hadrosaurid and
“saurolophs”), and a dorsally elevated jugal joint characterized by a ventral margin of the jugal
joint and ectopterygoid shelf that is nearly equal to the depth of the caudal segment of the
maxilla and a caudal margin of the joint flush with the caudal margin of the rostrodorsal
promontory of the lateral side of the maxilla (character 100-3).
Inclusion of “tsintaosaurs” within Lambeosarinae was unambiguously supported by a
relatively short ectocranial surface of the frontal, with a length/width ratio greater than 0.4 and
up to 0.8 (character 153-1) and a sagittal crest of the parietal that extends along the entire length
of the element, fading away at the rostral region where the parietal is rostrocaudally shorter than
it is wide (character 158-1; reversed in Hypacrosaurus altispinus).
Parsimony ancestral state reconstruction of the morphology of the jugal joint showed that
the unique condition present in “tsintaosaurs” is derived from the same state present in
saurolophine and remaining lambeosaurine hadrosaurids. This means that the elevation of the
articulation surface in “tsintaosaurs” was not a retention of the plesiomorphic hadrosaur
condition, but rather is a modification of the morphology present in Hadrosauridae. Although
“tsintaosaurs” elevated the position of the jugal joint, they retained the connection of this surface
with the lateral emargination of the ectopterygoid shelf like hadrosaurids.
202
The reconstruction of ancestral areas using Fitch parsimony and DIVA (Ronquist, 1996)
indicated that “tsintaosaurs” originated in eastern Asia during the middle or late Campanian.
Pararhabdodon isonensis represents a dispersal event to Europe (see Chapter 3 of this
dissertation) that occurred no later than the middle-late Campanian. Although no continuous land
connection between Asia and Europe has been reconstructed for the late Cretaceous,
“tsintaosaurs” may have used the mosaic of small landmasses that then formed Europe (Blakey,
2001) as stepping-stones to reach the western region of this continent. The DIVA reconstruction
of the ancestral area of the clade inferred a widespread ancestor across Eurasia and subsequent
cladogenesis between P. isonensis and Tsintaosaurus spinorhinus through vicariance. This
hypothesis supports similar conclusions reached by Casanovas et al. (1999). The mosaic
paleogeographic configuration of Europe during that time supports the possibility of vicariant
events operating due to the intermittent nature of the land connections. The latter were caused by
relatively high eustatic sea levels (Dalla Vecchia, 2006).
Discussion
Prieto-Marquez et al. (2006) considered two different types of jugal joints among
hadrosaurs depending on the joint’s position relative to the lateral emargination of the
ectopterygoid shelf. In one character state, the jugal joint is elevated and detached from the
lateral ridge of the etopterygoid shelf. This condition is found in primitive forms such as
Bactrosaurus johnsoni, Gilmoreosaurus mongoliensis and Telmatosaurus transsylvanicus. In the
second state, the ventral margin of the jugal joint lies flush with the lateral margin of the
ectoptergoyd shelf. This condition is typically found in hadrosaurine and lambeosaurine
hadrosaurids. In the phylogenetic analyses of Prieto-Marquez et al. (2006) P. isonensis was
scored as having the condition present in primitive non-hadrosaurid hadrosaurs. This scoring was
critical for inferring the placement of P. isonensis outside of Euhadrosauria.
The data I derived from the examination and comparison of Tsintaosaurus spinorhinus
with other hadrosaurs led to a revision of this maxillary character. The jugal articulation surface
in Pararhabdodon isonensis and T. spinorhinus is quite distinct from both the condition in non-
hadrosaurid hadrosaurs and that of hadrosaurids. Specifically (Fig. 5.2A and B) the joint surface
is connected to the lateral margin of the ectopterygoid ridge through a concave caudal border
203
(Fig. 5.2E and F). In hadrosaurines (Fig. 5.2D) and other lambeosaurines (Fig. 5.2C) the jugal
articular surface lies well above the level of the lateral margin of the ectopterygoid shelf. Due to
these differences I defined this condition as a separate character state for P. isonensis and T.
spinorhinus.
This study provided support for the conclusions of Casanovas et al. (1999) that
Pararhabdodon isonensis is a basal lambeosaurinae and that its presence in Europe may have
been the result of vicariance following dispersal from Asia. However, I based similar
conclusions on different characters, as shown above, and on different phylogenetic relationships
among lambeosaurine hadrosaurids (see Chapter 1 of dissertation, Fig. 1.9B).
The establishment of the presence of Lambeosaurinae in Western Europe greatly expands
the distribution of this clade, as well as that of Hadrosauridae in general. The recognition of the
synapomorphice maxillary and dentary characters in ”tsintaosaurs” that are, not only unique
within Lambeosaurinae, but also within Iguanodontoidea. This indicates that the anatomical and
taxonomic diversity of hadrosaurids is more diverse than previously recognized.
Conclusions
New anatomical information has been presented showing that Koutalisaurus kohlerorum,
from the Maastrichtian of Lleida, northern Spain, is most probably the junior synonym of the
contemporary taxon Pararhabdodon isonensis from the same region. Dentary and maxillary
characters previously considered as autapomorphies of K. kohlerorum and P. isonensis,
respectively, are shown to be synapomorphies uniting the latter with Tsintaosaurus spinorhinus
from the Campanian of the Wangshi Group, Shandong Province, China.
This study confirmed the presence of Lambeosaurinae in Europe with the inference that
Pararhabdodon isonensis is a member of this clade. Tsintaosaurus spinorhinus and P. isonensis
were inferred to form a clade of basal lambeosaurines characterized by a maxilla with elevated
jugal joint (continuous with the ectopterygoid ridge) and extreme medial projection of the
symphyseal region of the dentary. This clade originated in Asia during the middle or late
Campanian and subsequently dispersed to Europe no later than the late Campanian. The lineage
leading to Pararhabdodon isonensis may have diverged through vicariance.
204
Figure 5.1. “Tsintaosaur” dentaries. A, Pararhabdodon isonensis, IPS-SRA 27, in dorsal and B, rostral view. C, Tsintaosaurus spinorhinus,
IVPP V723 in dorsal and D, rostral view.
205
Figure 5.2. Lateral views of the jugal articular surfaces in the maxillae of various hadrosaur species.
A, Bactrosaurus johnsoni (AMNH 6553). B, cf. Telmatosaurus transsylvanicus (FGGUB R1010).
C, Corythosaurus intermedius (CMN 8676); D, Brachylophosaurus canadensis (MOR 1071-8-13-98-554);
E, Pararhabdodon isonensis (IPS SRA 22); F, Tsintaosaurus spinorhinus (IVPP V725).
206
CONCLUSION
Parsimony and Bayesian analyses were conducted on a complete taxonomic sample of

hadrosaurid species. The closest outgroup taxa to Hadrosauridae were found in eastern North
America. Hadrosauridae was redefined as the clade stemming from the most recent common
ancestor of Hadrosaurus foulkii and Parasaurolophus walkeri. In general these analyses
confirmed the evolution of hadrosaurids into Saurolophinae (Hadrosaurinae of most previous
authors) and Lambeosaurinae. Saurolophines consist of “saurolophs” and “kritosaurs”.
Lambeosaurines consist of a succession of Eurasian outgroups to two major clades,
“parasaurolophs” and “amurosaurs” (“hypacrosaurs” and “corythosaurs”).
The hadrosaurid radiation and the divergence of saurolophines from lambeosaurines
occurred no later than the Santonian and was coincident with the evolution of a suite of
mandibular characters (i.e., increased number of alveolar positions, presence of three teeth
forming the dentary occlusal plane, ventral offset of the oral predentary-premaxilla contact).
These results suggest that feeding adaptations might have played a central role in the
diversification of hadrosaurids. Two of the most distinctive and unique attributes of hadrosaurids,
circumnarial fossae and cranial crests, evolved prior to the major radiation of these animals.
Likewise, the fact that both characters were reconstructed to have evolved at the same time
supports the possibility that cranial crests evolved to provide additional space in the skull to allow
for the caudal extension of circumnarial fossae.
The present study supports North America as the center of origin for Hadrosauridae.
Their genesis for this event is traced to the middle-late Santonian. However, the major
hadrosaurid radiation , signaled in by the split between Saurolophinae and Lambeosaurinae,
occurred in Asia no later than the end of the Santonian. Saurolophines dispersed back to North
America during the late Santonian, where they showed considerable radiation in the Campanian.
Two dispersal events during that period are inferred to have occurred. First to southeastern and
Central Asia and a second one to South America and perhaps Antarctica. From Asia,
lambeosaurines dispersed to Europe and North America no later than the late Campanian.
However, it is uncertain at which point in the lambeosaurine phylogenythese animals dispersed
to North America, and whether a single dispersal event to the American continent with
subsequent occupations of Asia or independent dispersals to North America in the
207
“parasauroloph” and “amurosaur” clades. Vicariance may have occurred following several of the
inferred hadrosaurid dispersals. Specifically, the Santonian saurolophine dispersal to Asia, the
late Campanian dispersal to South America, the lambeosaurine dispersal to Europe and
subsequent Campanian exchanges between North America and Asia may have been followed by
vicariant events due to the temporary interruption of geographic connections and/or other yet
unrecognized factors.
A re-evaluation of the osteology of the types and referred materials of Kritosaurus
australis and Secernosaurus koerneri led me to conclude that the former is a junior synonym of
the latter. The results of a phylogenetic analysis of S. koerneri with a nearly complete taxonomic
sampling of Hadrosauridae did not support the inclusion of this species within the genus
Kritosaurus. However, S. koerneri was inferred to be a member of the “kritosaur” clade of
Saurolophinae as the sister taxon of the Argentinean OTU from Salitral Moreno. Another
unnamed hadrosaurid, an OTU from Big Bend National Park, Texas, was positioned as the
closest outgroup to the South American clade. The reconstruction of the ancestral area for the
South American hadrosaurid clade using Fitch parsimony and Dispersal-Vicariance analysis
supported the previous hypothesis that posited a dispersal from North to South America no later
than the late Campanian.
New anatomical information has been presented showing that Koutalisaurus kohlerorum,
from the Maastrichtian of Lleida, northern Spain, is most probably the junior synonym of the
contemporary taxon Pararhabdodon isonensis from the same region. Dentary and maxillary
characters previously considered as autapomorphies of K. kohlerorum and P. isonensis,
respectively, are shown to be synapomorphies uniting the latter with Tsintaosaurus spinorhinus
from the Campanian of the Wangshi Group, Shandong Province, China.
This study confirmed the presence of Lambeosaurinae in Europe with the inference that
Pararhabdodon isonensis is a member of this clade. Tsintaosaurus spinorhinus and P. isonensis
were inferred to form a clade of basal lambeosaurines characterized by a maxilla with elevated
jugal joint (continuous with the ectopterygoid ridge) and extreme medial projection of the
symphyseal region of the dentary. This clade originated in Asia during the middle or late
Campanian and subsequently dispersed to Europe no later than the late Campanian. The lineage
leading to Pararhabdodon isonensis may have diverged through vicariance.
208
The future of systematic studies of hadrosaurid dinosaurs will need a greater emphasis in
studies of European, South American and Asian taxa if one is to expect to achieve a more
accurate and global understanding of the evolution of these animals. Furthermore, the present
treatise is intended to serve as the starting point of a new trend whereby characters are
documented and the data and methods for their definition made explicit. This, coupled with the
implementation of the new developments in the theory and practice of phylogenetic inference,
should bring us nearer to the true phylogenetic history of this successful group of dinosaurs.
209
APPENDIX A
TAXON-CHARACTER STATE MATRIX
Table A.1. Character-taxon matrix of the outgroup iguanodontoidean taxa used in the phylogenetic analysis of the Hadrosauridae.
Inapplicable characters were coded as “-“ and missing data as “?”. Polymorphic characters were coded including in the same cell all
states present in a particular taxon separated by “&”. Ambiguity in the presence of character states for a given taxa was indicated by
“/”.
Taxon/Character 5 10 15 20
Bactrosaurus johnsoni 0 0&1 1 1 1&2 1 0 0 2 2 ? 0 1 1 1 0 0&1 1 1 1
Claosaurus agilis ? ? ? ? 1 3 1 0 ? 1 ? 0 ? ? ? ? ? ? ? 1
Eolambia caroljonesa 0 0 1 1 0 3 0 0 ? 1 0 0 ? 0 1 0 0 ? 0 ?
Equijubus normani 0 ? 1 ? 0 0 0 0 ? 0 0 0 ? ? ? 0 0 ? ? 0
Gilmoreosaurus mongoliensis 0 0 1 1 2 ? ? ? ? ? ? ? 1 1 ? 0 0 1 1 1
Iguanodon spp. 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Lophorhothon atopus ? ? ? ? 2 2 1 0 2 1 1 0 ? ? 1 0 ? ? ? 1
Ouranosaurus nigeriensis 0 0 1 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0
Probactrosaurus spp. 0 0 1 1 1 1 0 0 2 0 0 0 ? 1 1 0 0 ? ? 0
Protohadros byrdi 0 0 1 ? 0 1 0&1 0 1 1 1 0 1 1 1 1 0 1 ? ?
Tanius sinensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Telmatosaurus transsylvanicus 0 1 1 ? 1 2 0 0 ? 0 ? 1 1 1 1 0 1 ? 1 1
210
Table A.1—continued.
Bactrosaurus johnsoni 0&1 0 1 0 0 1 0 0 0 1 0 1 0 0 ? 0 0 1 2 1
Claosaurus agilis 1 ? 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Eolambia caroljonesa ? 0 1 ? ? ? ? ? ? ? ? ? ? ? ? ? 0 0 1 0
Equijubus normani 0 0 0 0 0 1 0 0 1 0 0 0 1 0 0 ? 0 0 1 0
Gilmoreosaurus mongoliensis 1 0 1 0 ? 0 ? ? 0 0 0 1 0 0 ? ? 0 ? ? 0
Iguanodon spp. 0 0 0 0 0 0&1 0 0 0 0 0 0 0 0 0 0 0 0&1 1 0&1
Lophorhothon atopus 1 0 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Ouranosaurus nigeriensis 0 0 ? 2 0 1 ? 0 0 0 2 2 ? 0 ? ? 0 1 0 0
Probactrosaurus spp. ? 0 ? 0 0 ? 0 0 0 0 0 1 1 0 ? ? 0 1 1 0
Protohadros byrdi 1 0 1 0 1 0 0 0 ? ? 1 4 1 1 0 0 1 0 1 2
Tanius sinensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Telmatosaurus transsylvanicus 1 0 0 ? ? ? ? ? ? ? ? ? ? ? ? ? 0 0 0 0
211
Bactrosaurus johnsoni 1 1 0 0&1 1 2 0 0&1 0 1 0 0 0 1 1 0 1 1 0 1
Claosaurus agilis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Eolambia caroljonesa 2 1 0 0&1 1 1 0 0 0 ? 0 0 0 1 ? ? 0 0 ? 0
Equijubus normani 1 ? 0 ? 1 0 0 0 0 ? 0 ? ? ? ? 0 0 1 0 ?
Gilmoreosaurus mongoliensis 1 0 0 0 1 ? ? ? ? 1 0 0 0 1 1 ? ? ? ? ?
Iguanodon spp. 0&1 0 0 0 1 0 0 0 0 1 0 0 0 0 0&1 0 0 0 0 0
Lophorhothon atopus ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Ouranosaurus nigeriensis 2 0 1 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 1
Probactrosaurus spp. 1 1 0 ? 1 2 0 0 0 1 0 0 0 1 ? ? 0 ? 0 ?
Protohadros byrdi 2 1 1 0 1 2 0 0 0 1 0 0 1 1 0 0 0 1 0 0
Tanius sinensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Telmatosaurus transsylvanicus 0 0 0 0 1 2 0 0 0 0 0 0 1 1 0 ? 1 1 0 0
212
Bactrosaurus johnsoni 0 1 0 0 1 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0
Eolambia caroljonesa 0 0 ? 0 ? ? 0 1 0 0 0 0 0 0 0 0 0 0 0 0
Equijubus normani ? 0 ? ? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Gilmoreosaurus mongoliensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Iguanodon spp. 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Lophorhothon atopus ? ? ? ? ? ? ? ? ? ? ? ? ? 0 0 0 0 0 ? ?
Ouranosaurus nigeriensis 0 0 1 1 0 ? 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Probactrosaurus spp. ? ? ? ? 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 0
Protohadros byrdi 0 0 ? ? 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Tanius sinensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Telmatosaurus transsylvanicus 0 1 0 0 ? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
213
Bactrosaurus johnsoni - 0 0 0 0 0 ? 0 0 0 - 0 ? 0 1 0 0 1 0 1
Claosaurus agilis - ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Eolambia caroljonesa - 0 0 0 ? ? ? 0 0 0 0 0 ? 0 ? 0 0 ? ? ?
Equijubus normani - 0 0 0 0 0 1 0 0 0 0 0 0 1 ? 0 0 1 ? ?
Gilmoreosaurus mongoliensis - ? 0 0 0 0 ? 0 ? ? ? 0 ? 0 1 0 0 2 0 1
Iguanodon spp. - 0 0 0 0 0 0 0 0 0 0 0 0 0&1 0 0 0 0 0 0
Lophorhothon atopus - ? 0 0 ? ? ? 0 0 0 - 0 ? 0 0 0 0 2 ? ?
Ouranosaurus nigeriensis - 0 0 0 0 0 0 0 1 0 0 ? 0 0 0 0 0 0 0 0
Probactrosaurus spp. - 0 0 0 0 ? ? 0 ? ? ? 0 0 0 ? ? ? ? ? 0
Protohadros byrdi - 0 0 0 ? 0 ? 0 ? ? ? 0 1 0 1 0 0 1 0 1
Tanius sinensis - ? ? ? ? ? ? 0 0 0 - ? ? ? ? ? ? ? ? ?
Telmatosaurus transsylvanicus - 0 0 0 ? ? ? 0 0 0 - 0 ? 0 1 0 0 2 0 1
214

Bactrosaurus johnsoni 0 0 0 0 1 0 1 1 0 0 0 0 0 0 0 - 0 2 1 0
Eolambia caroljonesa ? ? ? ? 1 0 ? ? ? ? ? 0 0 ? 0 - 0 1 ? 0
Equijubus normani 0 0 0 0 0 0 ? ? 0 0 0 0 0 0 0 - 0 0 0 0
Gilmoreosaurus mongoliensis 0 0 0 ? 1 1 1 1 0 0 ? 0 ? 0 0 - 0 ? ? 0
Iguanodon spp. 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 - 0 1 0 0
Lophorhothon atopus ? ? 0 ? ? ? ? ? 0 ? 0 0 ? 0 0 - ? ? ? ?
Ouranosaurus nigeriensis 0 1 0 0 0 3 2 0 0 0 ? 0 0 0 0 - 0 0 0 0
Probactrosaurus spp. ? ? ? 0 ? 0 0 ? ? ? 0 0 0 0 0 - 0 0 ? ?
Protohadros byrdi 0 1 0 ? 1 0 1 1 ? ? 0 0 0 0 0 - 0 1 0 0
Tanius sinensis ? ? ? ? ? ? ? ? ? ? ? 0 0 0 0 - 0 0 0 0
Telmatosaurus transsylvanicus 0 ? 0 0 2 1 1 1 ? ? 0 ? ? 0 0 - 0 ? ? ?
215

Bactrosaurus johnsoni 0 0 1 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 1
Eolambia caroljonesa 1 0 2 ? 0 0 0 ? ? ? ? ? ? ? 0 ? ? ? ? ?
Equijubus normani 1 0 1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 ?
Gilmoreosaurus mongoliensis ? 0 ? 0 1 1 1 1 0 0 ? ? ? ? 0 0 ? ? ? ?
Iguanodon spp. 0&1 0 0 0&1 0 0 0 1 0 0 0 0 0 0 0 0 0&1 0 1 1
Lophorhothon atopus ? ? ? 0 1 1 1 1 1 0 0 0 0 0 0 0 ? 0 ? ?
Probactrosaurus spp. ? 0 ? ? ? ? ? ? ? ? ? ? ? ? 0 0 0 ? ? 0
Protohadros byrdi 1 0 2 0 1 0 0 ? 0 0 ? 0 0 0 0 0 ? 1 ? ?
Tanius sinensis 0 0 1 0 1 ? ? 1 ? ? ? 0 ? ? 0 0 0 1 1 ?
Telmatosaurus transsylvanicus ? 0 0 0 1 ? ? 1 1 ? ? ? ? ? 0 0 ? ? ? ?
216

Iguanodon spp. 0 1 0 0 0 ? 0 0 0 - 0 0 0 ? 1 0 0 0 0 0
Eolambia caroljonesa ? ? ? ? ? ? ? 0 0 ? 0 ? 0 ? ? ? ? ? 1 1
Equijubus normani 0 1 0 ? ? ? 0 0 0 - 0 0 0 ? 1 0 ? 0 0 1
Ouranosaurus nigeriensis 1 ? 0 0 ? ? ? 0 0 - 0 0 0 ? 0 0 1 0 1 0
Probactrosaurus spp. 0 ? 0 0 0 ? ? 0 0 - 0 0 0 ? 1 0 0&1 0 ? ?
Protohadros byrdi ? ? ? ? ? ? ? 0 0 - 0 0 0 0 ? ? ? ? ? ?
Bactrosaurus johnsoni 1 1 0 0 0 1 1 0 0 - 0 0 0 0 1 0 0 0 1 1
Gilmoreosaurus mongoliensis ? ? ? ? ? ? ? ? ? ? 0 0 ? ? ? ? ? ? ? ?
Tanius sinensis 1 ? 0 ? 0 1 1 0 0 - 0 0 0 ? 1 0 0 0 1 1
Telmatosaurus transsylvanicus ? ? 0 0 ? 1 1 0 0 - 0 0 0 ? 1 0 ? ? 1 1
Lophorhothon atopus ? 2 0 0 1 1 1 0 0 - 0 0 0 0 1 ? ? 0 1 1
217
Character/Taxon 165 170 175 180

Bactrosaurus johnsoni 0 0 0 1 1 1 0 0 1 0 0 1 ? ? 1 ? 0 0 ? -
Eolambia caroljonesa ? 0 ? ? ? ? ? ? 0 0 0 ? ? ? 0 ? ? 0 ? -
Equijubus normani ? ? 0 ? ? ? 0 ? ? ? 0 ? ? ? ? 0 1 0 0 -
Gilmoreosaurus mongoliensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 0 ? -
Iguanodon spp. 0 ? 0 ? ? 1 0 0 0 0 0 0 0 ? 0 0 0&1 0 1 -
Lophorhothon atopus ? ? ? ? ? ? ? 0 ? ? 0 ? ? 0 0 ? ? 0 ? -
Ouranosaurus nigeriensis 0 1 0 1 1 1 0 0 0 0 0 1 0 0 0 1 0 0 1 -
Probactrosaurus spp. ? ? ? ? ? ? 0 ? ? ? 0 ? ? ? 0 ? ? 0 ? -
Protohadros byrdi ? ? ? ? ? ? ? ? ? ? ? ? ? ? 0 ? 1 0 1 -
Tanius sinensis 1 1 0 1 ? 1 0 0 0 0 0 1 0 0 1 ? ? ? ? -
Telmatosaurus transsylvanicus 1 0 ? ? ? 1 0 0 ? 1 ? ? ? ? 1 ? 1 0 1 -
218

Bactrosaurus johnsoni ? 0 - - 0 0 1 0 - 0 0 - - - - 1 0 1 0 0
Claosaurus agilis ? ? ? ? ? ? ? ? ? ? ? ? ? ? - ? ? ? ? ?
Eolambia caroljonesa ? 0 - - 0 ? 1 ? ? ? 0 - ? - - ? ? ? ? ?
Equijubus normani 0 0 - - 0 0 1 0 - 0 0 - - - - 0 0 1 0 0
Gilmoreosaurus mongoliensis ? ? - - 0 0 ? ? ? ? 0 - - - - ? ? ? ? ?
Iguanodon spp. 0 0 - - 0 0 1 0 - 0 0 - - - - 1 0 1 0&1 0
Lophorhothon atopus ? 0 - - 0 ? 1 1 0 1 0 - 5 - - 0 ? ? 1 0
Ouranosaurus nigeriensis 0 0 - - 0 0 1 0 - 1 0 - 0 - - 1 0 1 1 0
Probactrosaurus spp. ? 0 - - 0 ? 1 ? ? ? 0 - - - - ? 0 ? 1 0
Protohadros byrdi 0 0 - - 0 0 1 0 - 0 0 - - - - 0 ? ? ? 0
Tanius sinensis ? ? ? ? ? ? ? ? ? ? 0 - - - - 0 ? ? ? 0
Telmatosaurus transsylvanicus 0 0 - - 0 ? 1 ? ? ? 0 - - - - 0 ? 0 ? 0
219

Bactrosaurus johnsoni 0 0 ? 0 ? 0 0 ? 1 ? 0 ? 1 0 2 0 0 0 0 0
Claosaurus agilis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 0
Eolambia caroljonesa ? ? ? ? ? ? ? ? ? ? 0 ? ? ? ? ? ? ? ? ?
Equijubus normani 0 ? 0 0 ? 0 0 0 ? 0 0 ? ? ? ? ? ? ? ? ?
Gilmoreosaurus mongoliensis ? ? ? ? ? ? 0 ? ? ? ? ? ? ? 2 0 0 0 0 0
Iguanodon spp, 0 0 0 0 0 0 0 0 1 0 0&1 0 0 0 0 0 0 0 0 0
Lophorhothon atopus 0 0 ? 1 ? ? ? ? ? ? ? ? 0 0 1 0 0 0 0 0
Probactrosaurus spp, 0 0 0 1 ? ? ? ? ? ? 0 ? 0 0 1 0 0 0 0 0
Protohadros byrdi 0 ? 0 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Tanius sinensis 0 0 ? 1 ? 1 ? ? 2 ? ? ? ? ? ? ? ? ? ? 1
Telmatosaurus transsylvanicus 0 0 0 ? ? 0 1 ? ? ? ? ? ? ? ? ? ? ? ? ?
220

Bactrosaurus johnsoni 0 0 1 0 0 1 0 1 0 1 0&1 1 0 0 ? ? 1 0 1 ?
Claosaurus agilis 0 0 0 0 1 - 0 0 0 0 0 1 0 0 ? ? ? ? ? ?
Eolambia caroljonesa ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Equijubus normani ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Gilmoreosaurus mongoliensis ? 0 0 0 0 1 0 1 0 1 0 1 1 0/1 ? ? 1 ? ? ?
Iguanodon spp. 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1
Lophorhothon atopus 0 0 0 0 1 - 0 0 0 0 0 2 1 0 ? ? 1 0 0 ?
Probactrosaurus spp. 0 0 0 0 0 0 0 0 0 0 0 1 1 ? ? 0 1 ? ? 1
Protohadros byrdi ? ? ? ? ? ? ? ? 0 0 0 ? ? ? ? ? ? ? ? ?
Tanius sinensis 0 0 0 0 0 0 0 0 0 1 1 0 ? ? ? ? ? ? ? ?
Telmatosaurus transsylvanicus ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
221

Bactrosaurus johnsoni 1 ? 1 1 2 0 0 1 1 0 1 0 1 1 0&1 1 1 0 0 0
Claosaurus agilis ? ? 0 ? 1 1 1 1 1 ? ? 0 1 2 1 0 ? 0 1 0
Eolambia caroljonesa ? ? 1 1 ? ? ? 0 0 0 0 0 1 ? ? 1 ? 0 ? 0
Equijubus normani ? ? 0 1 0 0 0 0 0 0 0 0 0 0 0 ? ? 0 0 0
Gilmoreosaurus mongoliensis ? ? 1 1 2 1 0 1 1 0 1 0 1 2 0 1 1 0 1 0
Iguanodon spp. 0 0 0&1 1 0 0&1 0 0 0 0 0 0 0 0 0&1 1 1 0 0 0
Lophorhothon atopus 1 ? 1 1 2 ? 0 2 1 0 2 ? 1 ? ? 0 0 0 1 0
Probactrosaurus spp. 1 ? 0 ? 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0
Protohadros byrdi ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Tanius sinensis ? ? 1 1 2 0 0 0 1 0 1 0 1 1 1 1 1 0 1 0
222

Bactrosaurus johnsoni 1 0 0 0 ? 1 2 0 0 1 1 1 1 1 1 1 1 1 1 1
Claosaurus agilis ? ? 0 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Eolambia caroljonesa ? ? 0 ? ? ? ? ? ? ? ? ? ? ? 1 1 1 1 0 1
Equijubus normani ? ? 0 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Gilmoreosaurus mongoliensis 1 0 0&1 ? ? ? ? ? 0 ? ? ? 1 1 1 1 1 1 1 1
Iguanodon spp. 0 0 0 0 0 0 0 0 0 0 0 0 0 0&1 1 0 0 0 0 0
Lophorhothon atopus 1 1 1 1 ? ? 1 1 0 1 ? ? 1 1 0 1 2 1 ? ?
Probactrosaurus spp. 1 0 0 0 0 1 2 0 1 ? ? ? 1 1 1 0 1 0 0 1
Protohadros byrdi ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Tanius sinensis 0 0 0 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
223

Bactrosaurus johnsoni 0 2 1 0 0 1 1 1 1 0 0 ? 0 1 1 0 0 1 0
Claosaurus agilis ? ? ? ? ? ? ? ? ? ? ? ? 0 0 0 0 ? ? ?
Eolambia caroljonesa 0 1 1 0 1 0 ? ? ? ? ? ? ? ? ? ? ? ? ?
Equijubus normani ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Gilmoreosaurus mongoliensis 0 1 1 0 0 0 1 1 1 0 0 ? 0 1 1 0 0 0 0
Iguanodon spp. 0 1 1 0 0 0 0 0 0 0 0 0 1 0&1 0 0 0 0 0
Lophorhothon atopus ? ? 0 - - - 1 0 1 0 0 ? 0 1 1 0 0 0 0
Ouranosaurus nigeriensis 0 1 1 0 0 0 0 0 0 0 0 ? 0 1 1 0 0 0 0
Probactrosaurus spp. 0 1 1 0 0 0 0 0 ? ? ? ? ? 1 1 0 0 0 0
Protohadros byrdi ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Tanius sinensis ? ? ? ? ? ? 1 ? ? 1 ? ? ? ? ? ? ? ? ?
Telmatosaurus transsylvanicus ? ? ? ? ? ? ? ? ? 0 ? ? ? ? ? ? ? ? ?
224
Table A.2. Character-taxon matrix of the non-lambeosaurine hadrosaurid taxa used in the phylogenetic analysis of the Hadrosauridae.
Inapplicable characters were coded as “-“ and missing data as “?”. Polymorphic characters were coded including in the same cell all
states present in a particular taxon separated by “&”. Ambiguity in the presence of character states for a given taxa was indicated by
“/”.
Barsboldia sicinskii ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Brachylophosaurus canadensis 1 0&1 2 2 2 3 1 1 2 2 ? 0 1 1 1 1 1 1 2 1
Edmontosaurus annectens 2 1 3 2 1 3 1 0&1 2 2 2 0 1 1 1 2 1 1 2 1
Edmontosaurus regalis 2 1 3 2 2 3 1 0 2 2 2 0 1 1 1 2 1 1 2 1
Gryposaurus latidens 1 0 3 2 0 3 1 1 1 1 0 0 1 1 1 1 0 ? 2 1
Gryposaurus monumentensis 1 0 3 2 1 3 1 0 ? 2 ? 0 1 1 1 1 0 1 2 1
Gryposaurus notabilis 1 0&1 3 2 1 2 1 0 1 2 ? 0 1 1 1 1 0&1 1 2 1
Hadrosaurus foulkii ? ? ? ? 1 3 1 0 1 1 1 0 1 1 1 ? ? ? ? 1
Kerberosaurus manakini ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 2 1 ? 2 1
Kritosaurus navajovius 1 1 2 2 2 2 1 0 ? 2 ? 0 1 1 1 2 1 1 2 1
Maiasaura peeblesorum 1 1 2 2 2 3 1 1 2 2 ? 0 1 1 1 1 1 1 2 1
Prosaurolophus maximus 2 0&1 3 2 2 3 1 0 1 2 ? 0 1 1 1 2 1 1 2 1
Sabinas OTU 2 1 3 2 ? ? ? ? ? ? ? ? 1 1 ? 2 1 1 2 1
Salitral Moreno OTU 1 ? 2 2 2 3 1 0 2 2 ? 0 1 1 1 1 1 1 2 1
Saurolophus angustirostris 2 0 3 2 1 3 1 0 ? 2 ? 0 1 1 1 2 1 1 2 1
Saurolophus osborni 2 1 3 2 2 3 1 0 1 2 ? 0 1 1 1 2 1 1 2 1
Secernosaurus koerneri 1 ? 2 ? 2 3 1 0 ? 2 ? 0 1 1 1 1 1 1 ? 1
Shantungosaurus giganteus 2 1 3 2 1 3 1 1 ? 2 ? 0 1 1 1 2 1 ? ? 1
Wahweap OTU 1 1 2 2 2 3 1 1 2 1 ? 0 1 1 1 1 1 1 2 1
UTEP OTU ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Wulagasaurus dongi 1 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
225
Brachylophosaurus canadensis 1 0&1 1 2 0 2 1 1 1 1 2 2 0 1 1 1 3 1 2 0&1&2
Edmontosaurus annectens 1 ? 2 1 0&1 1 1 1 ? 1 1 4 0 1 1 1 2&3 1 0 0
Edmontosaurus regalis 1 0 2 1 0 1 1 1 ? 1 1 4 0 1 1 1 2&3 1 0 0
Gryposaurus latidens 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 2 1 2 2
Gryposaurus monumentensis 1 0 2 0 0 1 1 1 0 1 0 4 1 1 1 0 2 0 2 2
Gryposaurus notabilis 1 ? ? 0 0 2 0 0 0 1 0 4 1 1 1 0 1&2 0 2 1&2
Hadrosaurus foulkii 1 0 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Kerberosaurus manakini 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Kritosaurus navajovius 1 0 ? 0 0 1 ? ? 1 1 0 2 1 1 ? ? 0 0 1 2
Maiasaura peeblesorum 1 ? ? 2 0 2 1 1 1 1 2 2 0 1 1 1 2 0 2 1
Prosaurolophus maximus 1 0 ? 0 0 2&3 1 1 0 1 1 4 1 1 1 1 1 1 0 0&1
Sabinas OTU ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 1 ? 0 1
Salitral Moreno OTU 1 0 ? ? ? ? ? ? ? ? ? ? ? ? ? ? 0 0 2 1
Saurolophus angustirostris 1 ? ? 0 0 2 1 1 0 1 0 4 1 1 1 1 1&2 1 0 0
Saurolophus osborni 1 0 ? 0 0 3 1 1 0 1 0 4 1 1 1 ? 2 1 0 0
Secernosaurus koerneri 1 0 ? ? ? ? 1 1 ? ? ? ? ? ? ? ? 0 0 2 1
Shantungosaurus giganteus 1 ? ? ? 0 2 ? ? ? ? 0 ? ? ? ? ? 2 1 0 0
Wahweap OTU 1 1 1 ? ? ? ? ? ? ? ? ? ? ? ? ? 2 0 2 0&1
UTEP OTU ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Wulagasaurus dongi ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 1 1 1 0
226
Brachylophosaurus canadensis 0&1&2 1 1 1 1 2 1 1 1 0 1 1 1 2 1 1 1 1 1 0
Edmontosaurus annectens 0&1 0 1 1 0 3 1&2 1 0 0 1 1 1 2 1 1 1 1 1 0
Edmontosaurus regalis 0&1 0 1 1 0 3 1&2 1 0 0 1 1 1 2 1 1 1 1 1 0
Gryposaurus latidens 2 0 1 1 1 2 1 1 0 0 1 1 1 2 1 1 1 1 1 0
Gryposaurus monumentensis 1 0 1 1 1 2 1 ? 0 0 1 1 1 2 1 1 1 1 1 0
Gryposaurus notabilis 0&1 0 1 1 1 2 1 1 0 0 1 1 1 2 1 1 1 1 1 0
Hadrosaurus foulkii ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Kerberosaurus manakini ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Kritosaurus navajovius 0 0 1 1 1 2 1 1 0 0 1 1 1 2 1 1 1 1 1 0
Maiasaura peeblesorum 1 0 1 1 1 2 1 1 1 0 1 1 1 2 1 1 1 1 1 0
Prosaurolophus maximus 0 0 1 1 0 3 1&2 1 0 0 1 1 1 2 1 1 1 1 1 0
Sabinas OTU 0 0 1 1 0 2 1 1 0 0 1 1 1 2 1 ? ? ? ? ?
Salitral Moreno OTU 1 1 1 1 1 2 1 1 0 0 1 1 1 2 1 ? 1 ? 1 ?
Saurolophus angustirostris 0 0 1 1 0 3 1 1 0 0 1 1 1 2 1 1 1 1 1 0
Saurolophus osborni 0 0 1 1 0 3 1 1 0 0 1 1 1 2 1 1 1 1 1 0
Secernosaurus koerneri 1 1 1 1 1 2 ? 1 0 0 ? 1 1 2 ? ? 1 ? 1 ?
Shantungosaurus giganteus 2 ? 1 1 1 2 2 1 0 0 ? ? 1 2 1 ? ? ? ? ?
Wahweap OTU 0 0/1 1 1 1 2 1 1 ? 0 1 1 1 2 1 1 1 1 1 0
UTEP OTU ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Wulagasaurus dongi 1 0/1 ? ? 1 2 1 ? ? 0 1 ? 1 2 1 ? ? ? ? ?
227
Brachylophosaurus canadensis 0 1 0 0 2 1 3 1 1 1 0 1 0 0 0 0 0 0 0 0
Edmontosaurus annectens 0 1 0 0 2 1 3 0 1 1 1 1 1 0 0 0 0 0 0 0
Gryposaurus latidens 0 1 0 0 2 1 2 0 1 1 ? ? 0 0 0 0 0 0 0 0
Gryposaurus notabilis 0 1 0 0 1 1 2 0 1 1 0 1 0 0 0 0 0 0 0 0
Kerberosaurus manakini ? ? ? ? ? ? ? ? ? ? ? ? ? ? 0 ? ? ? ? ?
Kritosaurus navajovius ? 1 0 0 ? ? ? 0 ? ? ? ? ? 0 0 0 0 0 0 0
Prosaurolophus maximus 0 1 0 0 2 1 2 0 1 1 0 1 0 0 0 0 0 0 0 0
Sabinas OTU ? 1 0 0 ? ? ? ? ? ? ? ? ? ? 0 0 ? ? ? ?
Salitral Moreno OTU ? ? 0 0 ? ? ? ? 1 1 0 0 0 ? 0 ? ? ? ? ?
Secernosaurus koerneri ? 1 0 0 ? ? ? ? ? ? ? ? ? ? 0 ? ? ? ? ?
Shantungosaurus giganteus ? 1 0 0 ? ? ? 0 ? ? ? ? ? 0 0 0 0 0 ? ?
Wahweap OTU 0 1 0 0 2 ? 3 1 ? 1 0 1 0 0 0 0 0 0 0 0
UTEP OTU ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Wulagasaurus dongi ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
228
Brachylophosaurus canadensis - 0 0 0 1 1 1 0 0 0 - 0 1 0 1 0 0 2 0 2
Edmontosaurus annectens - 0 0 0 2 2 1 0 1 0 - 0 1 0 0 0 1 2 0 2
Edmontosaurus regalis - 0 0 0 2 2 1 0 1 0 - 0 1 0 0 0 1 2 0 2
Gryposaurus latidens - ? 0 0 0 1 0 0 ? ? 0 0 0 0 2 0 1 3 0 2
Gryposaurus monumentensis - 1 0 0 0 1 0 0 0 1 1 0 0 0 2 0 1 3 0 2
Gryposaurus notabilis - 1 0 0 0 1 0 0 0 1 1 0 0 0 2 0 1 3 0 2
Hadrosaurus foulkii - ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Kerberosaurus manakini - ? 0 0 0 ? ? 0 ? ? ? 0 ? 0 1 0 0 2 0 2
Kritosaurus navajovius - 0 0 0 0 1 0 0 1 0 - 0 0 0 2 0 1 2 0 2
Maiasaura peeblesorum - 0 0 0 1 1 1 0 0 0 - 0 1 0 1 0 0 2 0 2
Prosaurolophus maximus - 0 0 0 2 2 1 0 1 0 - 0 0 0 0&1 0 1 2 0 2
Sabinas OTU - ? ? ? ? ? ? 0 ? ? ? 0 ? 0 1 0 ? 2 0 2
Salitral Moreno OTU - ? 0 0 ? ? ? 0 0 0 ? 0 ? 0 2 0 1 3 0 2
Saurolophus angustirostris - 0 0 0 2 2 1 0 0 0 - 0 0 0 1 0 1 2 0 2
Saurolophus osborni - 0 0 0 2 2 1 0 0 0 - 0 0 0 1 0 1 2 0 2
Secernosaurus koerneri ? ? ? ? ? ? ? 0 0 0 - 0 ? 0 2 0 1 3 0 2
Shantungosaurus giganteus - ? 0 0 ? ? ? 0 ? ? ? 0 ? ? 0 0 1 2 0 2
Wahweap OTU - 0 0 0 ? ? ? 0 0 0 - 0 ? 0 1 0 0 2 0 2
UTEP OTU ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Wulagasaurus dongi - ? ? ? ? ? ? ? ? ? ? ? ? 0 ? 0 1 ? ? 2
229

Gryposaurus latidens 1 1 1 0 2 3 2 1 ? ? ? ? ? ? 1 ? ? ? ? ?
Gryposaurus monumentensis 1 1 1 0 2 3 2 1 0 1 1 1 1 0 1 ? 1 2 3 1
Hadrosaurus foulkii ? ? ? ? ? ? 2 ? ? ? ? ? ? ? ? ? ? ? ? ?
Kerberosaurus manakini 1 1 1 ? 1 ? 2 1 ? ? 1 1 0 0 1 1 0 0 3 0
Kritosaurus navajovius 1 1 1 ? 2 3 2 1 0 0 1 1 1 0 1 0 0 2 3 1
Prosaurolophus maximus 1 1 1 0 2 3 2 1 0 1 1 1 0 0 1 1 0 0&1 3 0
Sabinas OTU ? 1 1 ? 2 3 2 1 ? ? ? ? ? ? ? ? ? ? ? ?
Salitral Moreno OTU 1 1 1 ? 2 2 2 1 ? ? ? ? ? ? ? ? ? ? ? ?
Secernosaurus koerneri ? ? ? ? 2 1 2 1 ? ? ? ? ? ? ? ? ? ? ? ?
Shantungosaurus giganteus 1 1 1 ? 2 ? 2 1 ? ? ? ? ? ? ? ? ? ? ? ?
Wahweap OTU 1 1 1 ? 2 2 2 1 0 0 0 1 1 1 1 0 1 2 2 1
UTEP OTU ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Wulagasaurus dongi 1 1 ? ? 2 2 2 1 ? ? 0 1 ? 0 1 0 ? ? ? ?
230

Brachylophosaurus canadensis 1 0&1 1 0 1 1 2 1 1 0 0 0 0 1 0 0 0&1 0&1 1&2 1
Edmontosaurus annectens 2 1 1 0 1 1 2 0&1 1 0 1 0 0 0 0 0 1 2 0 1
Gryposaurus latidens ? ? ? ? ? ? ? ? ? 0 ? ? 0 ? 0 ? 0 ? ? ?
Kerberosaurus manakini 1 1 1 0 1 1 2/1 ? 1 0 1 0 0 0 0 ? ? ? ? ?
Kritosaurus navajovius 1 0 1 0 1 1 2 1 1 0 1 0 0 0 0 0 0 1 1 1
Maiasaura peeblesorum 1 0 1 0 0 1 2 1 1 0 0 0 1 1 0 0 0 0&1 1 1
Prosaurolophus maximus 0 1 2 0 1 1 1 0 1 0 1 0 1 0&1 0 0 1 1 1 1
Sabinosaur ? ? ? 0 1 1 2 1 1 0 ? ? ? ? ? 0 ? ? ? ?
Salitral Moreno OTU ? ? ? ? ? ? ? 1 ? ? ? ? ? ? ? ? ? ? ? ?
Saurolophus angustirostris 0 1 1&2 0 1 1 1 0 1 0 0 0 1 1 0 1 2 1 2 1
Saurolophus osborni 0 1 1&2 0 1 1 1 0 1 0 0 0 1 1 0 1 2 1 2 1
Secernosaurus koerneri ? ? ? ? ? ? ? ? ? 0 1 0 0 0 0 0 0 0 ? ?
Shantungosaurus giganteus ? ? ? 1 1 1 2 0 1 0 0 0 0 0 0 0 ? 1 ? 1
Wahweap OTU 1 0 1 ? ? ? ? ? ? 0 1 0 0 0 0 0 1 ? ? 1
UTEP OTU ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Wulagasaurus dongi ? ? 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
231

Edmontosaurus annectens 1 1 0 1 1 1 0 0 0 - 0 0 0 0 2 0 2 0 1 1
Edmontosaurus regalis 1 1 0 1 1 1 0 0 0 - 0 0 0 0 2 0 2 0 1 1
Gryposaurus latidens ? 2 0 ? ? ? ? 0 0 ? ? ? ? ? ? ? ? ? ? ?
Gryposaurus monumentensis 0 2 0 ? 1 1 0 0 0 - 1 0 0 ? 2 0 ? 0 1 1
Gryposaurus notabilis 0 2 0 1 1 1 0 0 0 - 1 0 0 0 2 0 0 0 1 1
Kerberosaurus manakini ? ? 0 ? 1 1 ? 0 0 - ? 0 0 0 2 0 ? 0 1 ?
Kritosaurus navajovius 0 2 0 0 1 ? 0 0 0 - 1 0 0 ? 1 0 1 0 1 1
Maiasaura peeblesorum 1 0 0 2 1 0 0 0 0 - 1 0 0 ? 1&2 0 1 0 1 1
Prosaurolophus maximus 1 2 0 1 1 ? 0 0 0 - 0 0 0 0 2 0 1 0 1 1
Sabinas OTU ? ? 0 ? 0 ? ? ? ? ? ? ? ? ? 2 0 ? 0 1 1
Salitral Moreno OTU ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Saurolophus angustirostris 1 2 1 2 1 0 0 0 1 1 2 0 0 ? 2 0 1 0 1 1
Saurolophus osborni 1 2 1 2 1 0 0 0 1 1 2 0 0 ? ? 1 ? 0 1 1
Secernosaurus koerneri ? ? 0 ? 1 1 0 0 0 - ? 0 0 0 2 0 1 0 1 1
Shantungosaurus giganteus 1 ? 0 2 1 1 ? 0 0 ? 0 0 0 ? 2 0 0 0 1 1
Wahweap OTU 1 1 0 2 1 0 0 0 0 - 1 0 0 0 1 0 0 0 1 1
UTEP OTU ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
232

Brachylophosaurus canadensis 1 1 1 1 1 2 0 0 1 1 1 1 1 0 1 0 1&2 0 0 -
Edmontosaurus annectens 1 1 0 1 1 1 0 1 1 1 2 1 0 1 1 1 0 0 0 -
Edmontosaurus regalis 1 1 0 1 1 1 0 1 1 1 2 1 0 1 1 1 0 0 0 -
Gryposaurus latidens ? ? ? ? ? ? ? ? ? ? ? ? ? ? 1 1 ? 0 ? -
Gryposaurus monumentensis ? 1 ? ? ? 1 0 0 1 1 ? 1 0 0 1 1 2 0 1 -
Gryposaurus notabilis 1 ? 0 1 1 1 0 0 1 1 2 1 0 0 1 1 2 0 1 -
Kerberosaurus manakini 1 0 ? ? ? ? 0 ? ? ? 2 ? ? ? 1 ? ? 0 ? -
Kritosaurus navajovius 1 1 0 1 1 1 0 0 1 1 2 1 ? 0 1 ? ? 0 ? -
Maiasaura peeblesorum 1 1 1 1 1 2 0 0 1 1 1 1 1 0 1 0 1 0 0 -
Prosaurolophus maximus 1 1 0 1 1 0 0 0 1 1 2 1 0 0 ? 1 1 0 2 -
Sabinas OTU 1 1 ? 1 1 0 1 ? 1 1 2 ? ? 0 ? ? ? 0 ? -
Salitral Moreno OTU ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 0 ? -
Saurolophus angustirostris 1 ? 0 ? ? 0 0 0 1 1 2 1 ? ? 1 1 1 0 2 -
Saurolophus osborni 1 ? 0 ? ? 0 0 0 1 1 2 1 0 0 1 1 1 0 2 -
Secernosaurus koerneri 1 1 0 1 ? 0 0 0 1 1 2 1 ? ? ? ? ? ? ? -
Shantungosaurus giganteus ? ? ? ? ? ? ? ? 1 1 2 1 ? ? ? ? 0/1 0 ? -
Wahweap OTU 1 1 1 1 1 2 0 ? 1 1 1 1 ? ? 1 ? ? 0 ? -
UTEP OTU ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
233

Brachylophosaurus canadensis 1 0 - - 0 1 1 3 0 1 0 - 2 - - 0 1 0 1 1
Edmontosaurus annectens 1 0 - - 0 1 1 1 1 0 0 - - - - 0 1 0 1 1
Edmontosaurus regalis 1 0 - - 0 1 1 1 1 0 0 - - - - 0 1 0 1 1
Gryposaurus latidens ? 0 - - 0 1 1 1 0 1 0 - 1 - - 0 ? 0 ? 0
Gryposaurus monumentensis 3 0 - - 0 1 1 1 0 1 0 - 1 - - 0 1 0 1 0
Gryposaurus notabilis 3 0 - - 0 1 1 1 0 1 0 - 1 - - 0 1 0 0 0
Hadrosaurus foulkii ? ? ? ? ? ? ? ? ? ? ? ? ? ? - ? ? ? ? ?
Kerberosaurus manakini ? 0 - - 0 ? 1 1 ? ? ? ? ? ? - 0 ? 0 ? ?
Kritosaurus navajovius ? 0 - - 0 1 1 2 0 1 0 - 5 - - 0 1 0 0&1 0
Maiasaura peeblesorum 0 0 - - 0 1 1 2 0 1 1 - 4 - - 0 1 0 1 1
Prosaurolophus maximus 2 0 - - 0 1 1 2 1 1 0 - 5 - - 0 1 0 1 1
Sabinas OTU ? 0 - - 0 ? ? ? ? ? 0 - - - - 0 ? 0 ? ?
Salitral Moreno OTU ? 0 - - 0 1 1 1 0 ? ? ? ? ? - ? ? ? ? ?
Saurolophus angustirostris 2 0 - - 0 1 1 3 1 1 0 - 3 - - 0 1 0 1 1
Saurolophus osborni 2 0 - - 0 1 1 3 1 1 0 - 3 - - 0 1 0 1 1
Secernosaurus koerneri ? 0 - - 0 ? ? ? ? ? ? ? ? ? - 0 ? ? ? ?
Shantungosaurus giganteus ? 0 - - 0 ? 1 ? ? 0 0 - - - - 0 ? ? ? ?
Wahweap OTU ? 0 - - 0 1 1 ? ? 0 0 - - - - 0 1 0 1 1
UTEP OTU ? ? ? ? ? ? ? ? ? ? ? ? ? ? - ? ? ? ? ?
Wulagasaurus dongi ? ? ? ? ? ? ? ? ? ? ? ? ? ? - ? ? ? ? ?
234

Barsboldia sicinskii ? ? ? ? ? ? ? ? 2 ? 1 ? ? ? ? ? ? ? ? ?
Brachylophosaurus canadensis 0 0 0&1 0 1 1 1 1 1 0 1 1 1 1 2 1 1 2 1 1
Gryposaurus latidens ? ? ? ? ? 0 1 ? 0 ? 1 1 1 1 2 1 1 1 1 1
Gryposaurus monumentensis 1 0 0 1 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Kritosaurus navajovius 1 0 0 1 1 ? 1 ? ? ? ? ? ? ? ? ? ? ? ? ?
Maiasaura peeblesorum 0 0 0 0 1 1 1 1 1 0 1 1 ? 1 2 1 1 ? 1 1
Prosaurolophus maximus 0 0 0&1 1 1 ? 1 1 0 0 1 1 1 1 2 1 1 2 1 1
Sabinas OTU ? ? ? ? ? ? 1 ? 1 ? 1 ? ? 1 1 1 ? 2 ? 0
Salitral Moreno OTU ? ? ? ? ? ? 1 ? 1 ? ? ? 1 1 1 0 1 1 1 1
Saurolophus osborni 0 ? 1 1 1 1 1 1 0 1 1 1 ? 1 2 1 1 2 1 1
Secernosaurus koerneri ? 0 ? 1 ? ? 1 ? 1 ? ? ? 1 1 1 0 1 1 1 1
Shantungosaurus giganteus 0 0 ? 1 ? ? 0 ? 1 1 1 0 1 1 2 1 1 1 0 1
UTEP OTU ? ? ? ? ? ? ? ? 0 ? 0 ? ? ? 1 1 1 1 1 1
Wahweap OTU 0 0 1 0 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Wulagasaurus dongi ? ? ? ? ? ? ? ? ? ? ? ? 1 ? ? ? ? ? ? 1
235

Brachylophosaurus canadensis 1 0&1 1 0&1 1 - 0 1 1 0 1 1 1 2 1 1 1 1 0 2
Edmontosaurus annectens 1 0 1 1 1 - 0 1 1&2 1 1 1 0&1 1 1 1 1 1 1 1
Edmontosaurus regalis 1 0 1 1 1 - 0 1 2 1 2 1 1 1 1 1 1 1 1 1
Gryposaurus latidens 0 0 0 1 1 - 1 1 1 0&1 1&2 2 0 1 ? ? 1 ? ? ?
Gryposaurus monumentensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Gryposaurus notabilis 0 0 0 1 1 - 1 1 1 1&2 2 1&2 1 1 1 1 1 1 1 1
Hadrosaurus foulkii ? ? ? ? ? ? ? ? 0 0 1 1 1 0 ? ? ? ? ? ?
Kritosaurus navajovius ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Maiasaura peeblesorum 1 0 1 1 1 - 0 1 1 0 1 1 1 1 1 1 1 ? 0 2
Prosaurolophus maximus 1 1 1 1 1 - 1 1 1 1 2 1&2 1 1 1 1 1 1 1 1
Sabinas OTU ? ? ? 0 1 - 1 1 ? ? ? ? ? ? ? ? ? ? ? ?
Salitral Moreno OTU 0 0 0 1 1 - 1 1 2 1 1 1 ? ? ? ? 1 ? ? ?
Saurolophus angustirostris 1 0 1 1 1 - 0 1 2 2 2 0 0 1 1 1 1 ? ? 1
Saurolophus osborni 1 0 1 1 1 - 0 1 ? 2 2 0/1 0 1 1 1 1 ? ? ?
Secernosaurus koerneri 0 0 0 1 1 - 1 1 ? ? ? ? 1 ? ? ? 1 ? ? ?
Shantungosaurus giganteus 0 0 0 1 1 - 0 0 2 1 2 1 1 1 ? ? 1 ? ? ?
Wahweap OTU ? ? ? ? ? ? ? ? ? ? ? ? 1 1 ? ? ? ? ? ?
UTEP OTU ? 0 ? 0 1 - 1 1 1 1 2 1 1 1/2 ? ? 1 ? ? ?
Wulagasaurus dongi 1 0 0 ? 1 - ? ? 1 2 2 1 ? ? ? ? ? ? ? ?
236

Barsboldia sicinskii ? ? 1 1 2 1 1 ? 3 ? ? 1 1 2 ? ? ? ? ? 1
Gryposaurus latidens ? ? 1 1 1 ? 1 2 1 1 2 0 1 2 2 0 0 0 1 1
Gryposaurus monumentensis ? ? ? ? ? ? ? ? ? ? ? ? ? 2 ? ? ? ? ? ?
Gryposaurus notabilis 1 1 1 1 1 1 1 2 1 1 2 1 1 2 2 ? ? ? 1 1
Hadrosaurus foulkii ? ? 1 ? 2 0 1 2 3 ? ? 0 1 2 ? ? ? ? 1 0
Kritosaurus navajovius ? ? ? ? ? ? ? ? ? ? ? ? ? 2 ? ? ? ? ? ?
Prosaurolophus maximus 1 1 1 1 2 1 1 2&3 2 1 2 0 1 2 2 0 0 0 1 1
Sabinas OTU ? ? ? ? ? 1 1 2 2 1 2 1 1 2 1 0 0 0 1 1
Salitral Moreno OTU ? ? 1 1 1 1 1 2 1 0 2 0 1 2 1 1 3 1 1 1
Secernosaurus koerneri ? ? 1 1 1 1 1 2 1 0 2 0 1 2 1 1 3 1 1 1
Shantungosaurus giganteus ? ? 1 1 2 1 1 2 3 1 2 1 1 2 1 0 0 0 1 1
Wahweap OTU ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
UTEP OTU 1 ? 1 1 2 0 1 2 1 1 2 1 1 2 1 1 3 1 1 0
237

Barsboldia sicinskii 1 1 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Gryposaurus latidens 1 1 1 1 4 1 1 1 1 1 ? 0 1 2 ? ? ? ? ? ?
Gryposaurus monumentensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Gryposaurus notabilis 1 ? 1 1 4 1 1 1 1 1 ? ? ? 2 0 1 1 1 2 2
Hadrosaurus foulkii 1 1 1 ? ? ? ? ? 1 1 ? ? ? ? 0 1 1 1 ? 0
Kritosaurus navajovius ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Prosaurolophus maximus 1 1 1 1 4 0 1 1 1 1 2 0 1 2 0 1 2 1 2 2
Sabinas OTU 1 1 1 1 4 0 1 1 1 1 ? 1 1 2 0 1 2 1 2 2
Salitral Moreno OTU 1 1 1 1 ? ? 1 1 1 1 ? ? 1 2 0 1 2 1 2 2
Saurolophus angustirostris 1 ? 1 1 4 1 1 1 1 1 ? ? 1 1 0 1 2 1 1 2
Saurolophus osborni 1 ? 1 1 4 1 1 1 1 1 ? ? 1 1 0 1 ? 1 ? ?
Secernosaurus koerneri 1 1 1 1 4 0 1 1 1 1 2 0 1 2 0 1 2 1 2 2
Shantungosaurus giganteus 1 1 1 1 3 ? 0 1 1 1 1 1 1 0 0 1 2 1 2 2
Wahweap OTU ? ? ? ? ? ? ? 1 1 1 0 1 1 ? ? ? ? ? ? ?
UTEP OTU 1 1 1 ? ? ? ? 1 1 1 ? 0 1 ? 0 0 0 1 2 ?
Wulagasaurus dongi ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 1 ? ?
238

Barsboldia sicinskii ? ? ? ? ? ? ? ? ? ? ? ? 0 0 ? ? ? ? ?
Brachylophosaurus canadensis 1 1 0 - - - 1 1 1 0 0 1 0 1 1 0 0 1 1
Edmontosaurus annectens 0&1 0 0 - - - 1 1 1 0 0 1 0 1 1 0 0 1 0
Edmontosaurus regalis 1 0 0 - - - 1 1 1 0 0 1 0 0 1 0 ? 1 0
Gryposaurus latidens ? ? 0 - - - ? ? ? 0 ? ? ? ? ? ? ? 1 0
Gryposaurus monumentensis ? ? ? - - - ? ? ? ? ? ? ? ? ? ? ? ? ?
Gryposaurus notabilis 1 0 0 - - - 1 1 1 0 0 1 0 0 1 1 1 1 0
Hadrosaurus foulkii 0 ? 0 - - - 1 1 1 ? ? ? 0 ? ? ? ? ? ?
Kerberosaurus manakini ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Kritosaurus navajovius ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Maiasaura peeblesorum 1 1 0 - - - 1 1 1 0 0 1 0 1 ? 0 0 1 1
Prosaurolophus maximus 1 1 0 - - - 1 1 1 0 0 1 0 1 1 1 1 1 0
Sabinas OTU 1 0 0 - - - 1 1 1 ? ? ? ? ? ? ? ? ? ?
Salitral Moreno OTU 0 2 0 - - - 1 1 1 0 0 ? 0 0 1 ? ? 1 0
Saurolophus angustirostris 0 1 0 - - - 1 1 1 ? 0 1 0 0 1 1 1 1 0
Saurolophus osborni ? 1 0 - - - 1 1 1 0 0 1 0 0 1 1 1 1 0
Secernosaurus koerneri 1 2 0 - - - 1 1 1 ? ? ? 0 0 ? ? ? ? ?
Shantungosaurus giganteus 1 1 0 - - - 1 1 1 0 0 ? 0 0 0 0 1 1 0
Wahweap OTU ? ? ? ? ? ? 1 1 ? ? ? ? ? ? ? ? ? ? ?
UTEP OTU ? 0 0 - - - 1 1 ? ? ? ? 0 1 ? ? ? ? 0
Wulagasaurus dongi ? 0 0 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
239
Table A.3. Character-taxon matrix of the lambeosaurine taxa used in the phylogenetic analysis of the Hadrosauridae. Inapplicable
characters were coded as “-“ and missing data as “?”. Polymorphic characters were coded including in the same cell all states present
in a particular taxon separated by “&”. Ambiguity in the presence of character states for a given taxa was indicated by “/”.
Amurosaurus riabinini 1 1 1 2 2 2 1 1 2 2 ? 0 1 1 1 1 1 1 2 1
Aralosaurus tuberiferus ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 1 1 ? ? 1
Charonosaurus jiayinensis 1 0 1 2 3 3 1 1 2 2 0 ? 1 1 1 ? ? ? ? ?
Corythosaurus casuarius 1 1 2 2 2 2 1 1 2 1 0 1 1 1 1 1 1 1 2 1
Corythosaurus intermedius 1 1 2 2 2 2 1 1 2 1 0 1 1 1 1 1 1 1 2 1
Hypacrosaurus altispinus 1 1 1 2 3 2 1 1 ? 2 ? 0 1 1 1 1 1 1 2 1
Hypacrosaurus stebingeri 1 1 1 2 3 2 0&1 0 1 1 0 1 1 1 1 2 1 1 2 1
Jaxartosaurus aralensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Lambeosaurus lambei 1 1 1&2 2 2&3 2 1 1 2 1 0 1 1 1 1 1&2 1 1 2 1
Lambeosaurus laticaudus 1 1 ? 2 2 2 1 1 2 ? ? ? 1 1 1 1 1 ? 2 1
Lambeosaurus magnicristatus 1 1 ? 2 ? 2 1 ? ? ? ? ? 1 1 1 1&2 1 1 2 1
Nipponosaurus sachalinensis ? ? 2 2 ? 2 1 1 ? 2 ? 0 ? ? ? ? ? 1 ? 1
Olorotitan ararhensis 1 1 ? 2 2 2 1 1 2 ? ? ? ? 1 1 2 1 1 2 1
Pararhabdodon isonensis 1 1 ? ? ? ? ? ? ? ? ? ? 1 1 1 1 1 ? ? ?
Parasaurolophus cyrtocristatus 1 1 2 2 2 2 1 1 2 1 0 0 ? 1 1 ? ? ? ? ?
Parasaurolophus tubicen 1 1 2 2 3 2 1 1 ? 1 0 0 ? 1 1 1 1 1 2 1
Parasaurolophus walkeri 1 1 2 2 ? ? ? ? ? ? ? ? ? 1 1 ? 1 ? ? 1
Sahaliyania elunchunorum 1 1 1 ? 2 2 1 1 ? 2 ? 0 1 1 1 ? ? ? ? ?
Tsintaosaurus spinorhinus 1 1 1 2 2&3 3 1 1 1 1 0 0 1 1 1 1 1 ? ? 1
Velaphrons coahuilensis 1 1 ? 2 2 2 1 ? ? ? ? ? 1 1 1 1 1 ? 2 1
240
Amurosaurus riabinini 1 0 1 ? ? ? ? ? ? ? ? ? ? ? ? ? 2 0 1 2
Aralosaurus tuberiferus 1 0 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Charonosaurus jiayinensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 0 0 ? 1
Corythosaurus casuarius 1 ? ? 0 1 3 1 2 1 1 1 3 1 1 1 0 1 1 1 1&2
Corythosaurus intermedius 1 ? ? 0 1 3 1 2 1 1 1 3 1 1 1 0 1 1 1 1&2
Hypacrosaurus altispinus 1 1 2 ? 1 3 1 2 2 1 1 3 1 1 1 ? 1 1 1 1
Hypacrosaurus stebingeri ? ? ? 0 1 3 1 2 1 1 1 3 1 1 1 0 1 0 1 2
Lambeosaurus lambei 1 1 1 0 1 2&3 1 1 0 1 1 3&4 1 1 1 0 1 1 1 1
Lambeosaurus laticaudus 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Lambeosaurus magnicristatus 1 1 1 ? 1 2&3 1 ? 0 1 1 3 1 1 1 0 1 1 1 1
Nipponosaurus sachalinensis 1 1 2 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Olorotitan ararhensis 1 ? ? 1 1 3 1 1 1 1 1 ? ? 1 ? 1 2 1 1 1
Pararhabdodon isonensis ? ? ? ? ? ? ? ? ? ? ? ? 0 ? ? ? 1 0 0 2
Parasaurolophus cyrtocristatus ? ? ? ? ? ? ? ? ? ? ? ? ? 1 ? ? ? ? ? ?
Parasaurolophus tubicen 1 1 1 ? ? ? ? ? ? ? ? ? ? 1 ? ? 1 1 2 0
Parasaurolophus walkeri 1 1 1 1 1 3 1 2 ? 1 1 4 1 1 1 0 1 0 2 0
Sahaliyania elunchunorum ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 2 0 2 2
Tsintaosaurus spinorhinus 1 ? ? 1 0 2 1 2 0 1 1 3 1 1 1 0 1 0 0 2
Velaphrons coahuilensis 1 ? ? 1 1 2 1 2 ? 1 1 3 ? 1 ? ? ? 1 2 1
241
Amurosaurus riabinini 2 1 1 1 1 2 1&2 1 0 0 1 1 1 2 1 1 1 1 1 1
Aralosaurus tuberiferus ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Charonosaurus jiayinensis 1 1 1 0 1 2 1 1 0 0 1 1 1 2 1 1 1 1 1 ?
Corythosaurus casuarius 0&1 1 1 0&1 1 2 1 1 0 0 1 1 1 2 1 1 1 1 1 1
Corythosaurus intermedius 0&1 1 1 0&1 1 2 1 1 0 0 1 1 1 2 1 1 1 1 1 1
Hypacrosaurus altispinus 1&2 1 1 1 1 2 ? 1 0 0 ? 1 1 2 1 1 1 1 1 ?
Hypacrosaurus stebingeri 2 0 1 1 1 2 1 1 0 0 1 1 1 2 1 1 1 1 1 1
Lambeosaurus lambei 1 0 1 0&1 1 2 1 1 0 0 1 1 1 2 1 1 1 1 1 1
Lambeosaurus laticaudus ? ? ? ? ? ? ? ? ? ? ? 1 1 2 ? ? ? ? ? ?
Lambeosaurus magnicristatus 1 ? ? 1 1 2 1 1 0 ? 1 1 1 2 1 1 1 1 1 ?
Nipponosaurus sachalinensis ? ? ? ? ? 2 1 1 0 0 1 ? ? 2 1 1 1 1 1 ?
Olorotitan ararhensis 1 1 1 0 1 2 1 1 0 0 1 1 1 2 1 1 1 1 1 1
Pararhabdodon isonensis 2 2 1 1 0 ? ? ? ? ? 1 1 1 2 1 ? ? ? ? ?
Parasaurolophus cyrtocristatus ? ? ? ? 1 2 ? ? ? 0 ? 1 1 2 1 1 1 1 1 ?
Parasaurolophus tubicen 2 1 1 0 1 2 1 1 0 0 1 1 1 2 1 1 1 1 1 1
Parasaurolophus walkeri 2 1 1 ? 1 2 1 ? ? 0 1 1 1 2 1 1 1 1 1 1
Sahaliyania elunchunorum 2 1 ? 1 0&1 2 1&2 1 0 0 1 1 1 2 1 ? ? ? ? ?
Tsintaosaurus spinorhinus 2 2 1 1 0 2 1&2 1 0 0 1 1 1 2 1 1 1 1 1 0
Velaphrons coahuilensis 1 1 1 1 1 2 1 1 0 0 1 1 1 2 1 ? ? ? ? ?
242
Amurosaurus riabinini 1 1 0 0 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Aralosaurus tuberiferus ? ? 0 ? ? ? ? ? ? ? ? ? ? ? ? 0 ? 0 ? ?
Charonosaurus jiayinensis ? 1 0 0 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Hypacrosaurus altispinus ? 1 0 0 1 1 1 1 1 0 0 0 0 1 1 2 2 0 2 2
Lambeosaurus lambei 1 1 0 0 1 1 1 1 1 0 0 0 0 1 1 2 2 1 1 1
Lambeosaurus laticaudus ? ? ? 0 1 1 1 0 1 0 0 0 0 1 ? ? ? ? 1 1
Lambeosaurus magnicristatus ? 1 0 0 1 1 1 1 1 0 0 0 0 1 1 2 2 1 1 1
Nipponosaurus sachalinensis ? 1 0 0 ? 1 ? ? ? ? ? ? ? 1 ? ? ? ? ? 1
Olorotitan ararhensis 1 1 0 0 1 1 3 1 1 0 0 0 0 1 ? 2 ? 0 2 1
Pararhabdodon isonensis ? ? ? 0 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Parasaurolophus cyrtocristatus ? 1 0 0 ? ? ? ? ? ? ? ? ? 1 0 0 0 0 ? ?
Parasaurolophus tubicen 1 1 0 0 ? ? ? ? ? ? ? ? ? 1 0 0 0 0 ? ?
Parasaurolophus walkeri 1 1 0 0 1 1 3 1 ? 0 0 0 0 1 0 0 0 0 2 0
Sahaliyania elunchunorum ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Tsintaosaurus spinorhinus 0 1 0 0 0 1 3 0 1 ? 0 1 0 ? ? 2 ? ? 0 0
Velaphrons coahuilensis ? 1 0 0 1 1 0 1 1 0 0 0 0 1 1 2 ? 0 1 1
243
Amurosaurus riabinini ? - 2 1 ? - - ? 0 0 - 1 - 0 0 1 1 2 1 2
Aralosaurus tuberiferus ? ? ? ? ? ? ? ? ? ? ? 1 - 0 0 0 1 2 1 2
Charonosaurus jiayinensis ? - 2 1 ? - - ? 0 0 - ? - 0 ? ? ? ? ? ?
Corythosaurus casuarius 2 - 2 1 4 - - 1 0 0 - 1 - 0 0 1 1 2 1 2
Corythosaurus intermedius 2 - 2 1 4 - - 1 0 0 - 1 - 0 0 1 1 2 1 2
Hypacrosaurus altispinus 0 - 2 1 4 - - 1 0 0 - 1 - 0 0 1 1 2 1 2
Hypacrosaurus stebingeri 2 - 2 1 4 - - 1 0 0 - 1 - 0 0 1 1 2 1 2
Jaxartosaurus aralensis ? ? ? ? ? ? - ? 0 0 - ? ? ? ? ? ? ? ? ?
Lambeosaurus lambei 2 - 2 1 3 - - 0 0 0 - 1 - 0 0 1 1 1&2 1 2
Lambeosaurus laticaudus 1 - ? ? ? - - ? ? 0 - 1 - 1 0 1 ? 2 1 2
Lambeosaurus magnicristatus 2 - 2 1 3 - - 0 0 0 - 1 - 0 0 1 1 2 1 2
Nipponosaurus sachalinensis ? - 2 1 ? - - ? ? 0 - 1 - ? ? ? ? ? ? 2
Olorotitan ararhensis 1 - 2 ? ? - - 0 ? 0 - 1 - 0 0 1 1 1 1 2
Pararhabdodon isonensis ? ? ? ? ? ? ? ? ? 0 - ? ? ? 0 0 1 1 ? 3
Parasaurolophus cyrtocristatus 0 - 2 ? ? - - 0 ? 0 - 1 - 0 ? ? ? ? 1 ?
Parasaurolophus tubicen 0 - 2 ? ? - - 0 0 0 - 1 - 0 1 1 1 1 1 2
Parasaurolophus walkeri 0 - 2 ? ? - - 0 0 0 - 1 - 0 1 1 1 1 1 2
Sahaliyania elunchunorum ? ? ? ? ? ? ? ? ? ? ? 1 - 0 0 1 0 1 ? 2
Tsintaosaurus spinorhinus - - 1 0 3 - - 0 0 0 - 1 - 1 0 0 1 1 1 3
Velaphrons coahuilensis 1 - 2 1 4 - - ? 0 0 - 1 - 0 0 1 1 2 1 2
244

Amurosaurus riabinini 0 1 2 ? 2 3 2 1 ? ? 2 1 ? 1 1 0 1 1 1 0
Aralosaurus tuberiferus 0 1 ? ? 2 0 2 1 ? ? 1 1 ? ? ? ? 1 ? ? 0
Charonosaurus jiayinensis ? ? ? ? ? ? ? ? ? ? 2 1 2 1 1 0 1 0 1 0
Corythosaurus casuarius 1 1 2 1 2 3 2 1 1 0 2 1 2 1 1 0 1 0&1 1 0
Corythosaurus intermedius 1 1 2 1 2 3 2 1 1 0 2 1 2 1 1 0 1 0&1 1 0
Hypacrosaurus altispinus 1 1 2 1 2 3 2 1 1 0 4 1 2 1 1 0 1 0&1 1 0
Lambeosaurus lambei 1 1 2 1 2 3 2 1 1 0 2 1 2 1 1 0 1 0&1 1 0
Lambeosaurus laticaudus 1 1 2 ? 2 3 2 1 ? ? 2 1 2 1 1 0 1 1 ? 0
Lambeosaurus magnicristatus 1 1 2 1 2 3 2 ? 1 0 2 1 2 1 1 0 1 0 1 0
Nipponosaurus sachalinensis ? ? ? ? ? ? ? 1 ? ? ? ? ? ? ? ? ? ? ? ?
Olorotitan ararhensis 1 1 2 1 2 3 2 1 ? ? 4 1 2 1 1 0 1 0 1 0
Pararhabdodon isonensis 1 1 ? ? 2 3 2 1 ? ? ? ? ? ? ? - ? ? ? ?
Parasaurolophus cyrtocristatus ? ? ? 0 ? ? ? ? 1 ? ? ? ? ? 1 ? ? ? ? ?
Parasaurolophus tubicen 1 1 2 0 2 3 2 1 1 ? 3 1 2 1 1 0 1 0 1 0
Parasaurolophus walkeri 1 1 ? 0 2 3 2 ? 1 ? 3 1 2 1 1 ? 1 0 1 0
Sahaliyania elunchunorum ? 1 2 ? 1 ? ? 0 ? ? 2 1 2 1 1 0 1 1 1 ?
Tsintaosaurus spinorhinus 1 1 ? ? 2 3 2 1 ? ? 4 1 2 ? 1 0 1 1 1 0
Velaphrons coahuilensis 1 1 2 1 2 1 2 1 ? ? 2 1 2 1 1 0 1 1 1 0
245

Amurosaurus riabinini 1 1 1 ? ? ? ? ? 1 2 0 1 0 0 ? 1 0 1 0 ?
Aralosaurus tuberiferus 1 ? 0 ? ? ? ? ? ? 0 0 0 0 0 ? 0 0 1 ? ?
Charonosaurus jiayinensis 1 1 0 ? ? ? ? ? ? 2 ? ? 0 ? ? 1 0 1 1 1
Hypacrosaurus altispinus 1 1 1 1 0 0 0 0 1 2 0 1 0 0 2 1 0 1 0 0
Jaxartosaurus aralensis ? ? ? ? ? ? ? ? ? 1 ? ? 0 ? ? 0 2 1 ? 1
Lambeosaurus laticaudus 1 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Lambeosaurus magnicristatus 1 1 1 1 0 0 0 0 1 2 0 1 0 0 1 1 0 1 0 1
Nipponosaurus sachalinensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Olorotitan ararhensis 1 1 1 1 ? ? ? 0 ? 2 0 ? 0 0 ? ? ? ? ? ?
Pararhabdodon isonensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Parasaurolophus cyrtocristatus ? ? ? ? ? ? ? ? ? 2 0 1 0 0 2 1 0 1 1 1
Parasaurolophus tubicen 2 1 0 1 ? 0 0 0 1 2 0 1 0 0 2 1 2 1 0 1
Parasaurolophus walkeri 2 1 0 1 0 0 0 0 1 2 0 1 0 0 2 1 2 1 0 1
Sahaliyania elunchunorum 1 1 0 1 1 0 0 ? 1 ? ? ? ? ? ? ? ? ? ? ?
Tsintaosaurus spinorhinus 0 1 1&2 1 0 1 1 0 1 0 ? 0 ? 0 0 0 ? ? ? ?
Velaphrons coahuilensis 1 1 0 1 1 0 0 0 1 2 ? ? 0 ? ? 1 0 ? ? 1
246

Amurosaurus riabinini ? 1 1 0/1 1 1 0 1 0 1 2 1 1 ? 1 1 0 2 1 1
Aralosaurus tuberiferus 1 ? 0 ? 1 ? ? 0 0 ? ? 0 0 ? 1 ? ? ? ? ?
Charonosaurus jiayinensis 2 ? 1 2 1 0 0 0 1 ? 2 1 2 1 1 1 1 2 1 1
Jaxartosaurus aralensis 1 2 0 ? 1 ? ? 1 0 1 2 1 0 ? 0 1 ? 0 1 1
Lambeosaurus laticaudus ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Lambeosaurus magnicristatus 1 1 1 2 1 ? 0 1 0 1 2 1 ? ? 0 1 ? 1 1 1
Nipponosaurus sachalinensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Olorotitan ararhensis ? ? ? ? 1 ? ? ? ? ? 2 ? 1 ? ? 1 ? ? 1 1
Parasaurolophus cyrtocristatus 2 2 1 2 1 0 0 0 1 0 2 1 2 ? 0 1 1 1 1 1
Parasaurolophus tubicen 1 2 1 2 1 0 0 0 1 0 2 1 2 1 0 1 1 1 1 1
Sahaliyania elunchunorum ? ? ? ? ? 1 ? 1 0 1 ? ? 1 ? 0 ? 1 2 1 1
Tsintaosaurus spinorhinus 1 ? 0 0 1 0 0 0 0 ? 2 1 1 ? 0 1 1 1 1 1
Velaphrons coahuilensis 1 ? ? ? 1 ? ? 1 0 1 ? ? 1 ? 0 1 1 1 ? ?
247

Amurosaurus riabinini 1 0 0 0 1 1 0 0 ? 1 0 0 ? ? 1 ? ? 1 - ?
Aralosaurus tuberiferus ? ? ? ? ? ? ? ? 1 1 ? ? ? ? ? ? ? ? ? ?
Charonosaurus jiayinensis 1 0 0 ? ? 1 0 0 ? 0 0 0 ? ? 1 ? ? 1 - ?
Corythosaurus casuarius 1 0 0 1 0 0 0 0 1 1 0 1 0 0 1 1 0&1 1 - 0
Corythosaurus intermedius 1 0 0 1 0 0 0 0 1 1 0 1 0 0 1 1 0&1 1 - 0
Hypacrosaurus altispinus 1 0 0 0 1 0 0&1 0 1 1 0 1 0 0 1 1 0&1 1 - 1
Hypacrosaurus stebingeri 1 0&1 0 1 0 1 0 0 1 1 0 1 ? ? 1 1 1 1 - 1
Jaxartosaurus aralensis 1 ? 0 ? ? ? 0 0 1 1 0 1 ? ? ? ? ? ? ? ?
Lambeosaurus lambei 1 0 0 1 1 0 1 0 1 1 0 1 0 0 1 1 1 1 - 0
Lambeosaurus laticaudus ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 1 ? 1 - ?
Lambeosaurus magnicristatus ? ? ? ? ? ? 1 ? 1 1 0 1 ? ? 1 1 1 1 - 0
Nipponosaurus sachalinensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? 1 ? ? 1 - ?
Olorotitan ararhensis ? ? ? ? ? ? ? ? ? ? ? ? 0 0 ? 1 1 1 - ?
Parasaurolophus cyrtocristatus 1 0 0 1 0 0 0 0 1 1 0 1 ? ? ? 1 2 1 - 1
Parasaurolophus tubicen 1 0 0 1 ? 0 0 0 1 1 0 1 ? ? 1 1 2 1 - 1
Parasaurolophus walkeri ? ? ? ? ? ? 0 ? 1 1 0 1 ? ? 1 1 2 1 - 1
Sahaliyania elunchunorum ? ? 0 ? ? ? 0 0 ? ? 0 ? ? ? ? ? ? ? ? ?
Tsintaosaurus spinorhinus 1 1 0 1 ? 0 ? 0 1 1 0 0 ? ? 1 1 ? ? ? ?
Velaphrons coahuilensis ? ? ? ? ? ? ? ? 1 1 0 1 ? ? 1 1 1 1 - ?
248

Amurosaurus riabinini - ? ? ? ? ? 0 - - 1 ? ? ? ? ? 0 1 0 ? ?
Aralosaurus tuberiferus ? ? ? ? ? ? ? ? ? 1 ? ? ? ? ? 0 ? 0 1 ?
Charonosaurus jiayinensis - ? ? ? ? ? 0 - - 1 ? ? ? ? ? 0 1 0 1 ?
Corythosaurus casuarius - 1 0 0 1 2 0 - - 1 2 0 6 2 1 0 1 0 1 0
Corythosaurus intermedius - 1 0 0 1 2 0 - - 1 2 0 6 2 0 0 1 0 1 0
Hypacrosaurus altispinus - 0 0 0 1 2 0 - - 1 2 0 6 1 0 0 1 0 1 0
Hypacrosaurus stebingeri - 1 0 0 1 2 0 - - 1 2 0 6 1 1 0 1 0 1 0
Jaxartosaurus aralensis ? ? ? ? ? ? ? ? ? 1 ? ? ? ? ? 0 ? ? 1 ?
Lambeosaurus lambei - 1 1 0 1 2 0 - - 1 2 1 6 3 0 0 1 0 1 0
Lambeosaurus laticaudus - ? ? ? ? 2 0 - - ? ? ? ? ? ? 0 ? ? ? ?
Lambeosaurus magnicristatus - ? ? 0 1 2 0 - - 1 2 1 6 3 1 0 1 0 1 0
Nipponosaurus sachalinensis - ? ? ? ? 2 0 - - 1 2 ? ? ? ? 0 ? ? ? ?
Olorotitan ararhensis - ? ? ? 1 2 0 - - 1 2 0 ? 1 ? 0 1 0 1 0
Parasaurolophus cyrtocristatus - 0 2 1 2 2 0 - - 1 2 1 7 0 - 0 1 0 1 ?
Parasaurolophus tubicen - 0 2 1 2 2 0 - - 1 2 1 7 0 - 0 1 0 1 ?
Parasaurolophus walkeri - 0 2 1 2 2 0 - - 1 2 1 7 0 - 0 1 0 1 1
Sahaliyania elunchunorum ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Tsintaosaurus spinorhinus ? ? ? ? ? ? 0 - - 1 2 ? ? ? ? 0 1 0 1 ?
Velaphrons coahuilensis - ? ? ? ? 2 0 - - 1 2 ? 6 ? ? 0 1 ? 1 0
249

Amurosaurus riabinini ? 1 0 1 ? ? 1 ? ? ? ? ? 0 1 2 1 1 1 1 1
Aralosaurus tuberiferus 0 1 0 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Charonosaurus jiayinensis ? 1 0 1 ? ? 1 ? ? ? 1 ? ? ? ? ? ? ? ? 1
Corythosaurus intermedius 2 1 0 1 1 0 1 1 1 1 1 ? 1 1 2 1 1 1 1 1
Hypacrosaurus stebingeri 2 1 0 1 1 ? 1 ? 2 ? 1 ? 1 1 2 1 1 1 1 1
Jaxartosaurus aralensis 0 1 ? 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Lambeosaurus laticaudus ? ? ? ? ? ? 1 ? 2 1 ? ? ? 1 2 1 1 1 1 ?
Lambeosaurus magnicristatus 2 1 0 ? 1 ? 1 1 ? 1 ? ? 1 1 ? ? ? ? ? 1
Nipponosaurus sachalinensis 0 ? ? ? ? 0 1 ? ? ? ? ? ? 1 2 0 1 1/2 ? 0
Olorotitan ararhensis 0 ? 0 ? ? 0 1 1 ? 1 2 ? 1 1 2 1 1 1 1 1
Pararhabdodon isonensis ? ? ? ? ? ? 1 ? 1 ? 1 ? ? ? ? ? ? ? ? 1
Parasaurolophus cyrtocristatus 2 1 0 1 ? ? 1 1 1 1 1 0 ? ? ? ? ? ? ? 1
Parasaurolophus tubicen 2 1 0 1 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Parasaurolophus walkeri 2 1 0 1 1 ? 1 1 1 1 1 ? ? 1 2 1 1 1 1 1
Sahaliyania elunchunorum ? ? ? ? ? ? ? ? ? ? ? ? 1 ? ? ? ? ? ? 1
Tsintaosaurus spinorhinus 0 2 0 1 1 ? 1 1 1 1 1 0 0 1 2 0 1 0 1 1
Velaphrons coahuilensis ? 1 0 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
250

Amurosaurus riabinini ? 1 1 ? 0 1 0 0 2 2 2 1 1 2 ? ? 1 ? ? ?
Charonosaurus jiayinensis ? 1 1 1 0 1 0 0 1 2 2 1 1 ? ? ? 1 0 0 ?
Hypacrosaurus stebingeri 1 1 1 0 0 1 0 0 2 2 2 0 1 2 ? ? ? ? ? ?
Lambeosaurus lambei 1 1 1 0 0 1 0 0 ? ? 2 0/1 1 1/2 1 1 1 1 1 1
Lambeosaurus laticaudus ? ? ? ? 0 1 0 ? 2 1 2 1 ? ? ? ? ? ? ? ?
Lambeosaurus magnicristatus 1 1 1 0 0 1 0 0 2 2 2 1 1 2 1 1 1 ? ? 1
Nipponosaurus sachalinensis 1 0 1 0 0 1 0 0 ? ? ? ? 1 ? ? 1 1 ? ? 1
Olorotitan ararhensis 1 1 1 1 0 1 0 0 2 2 ? 0 ? ? ? ? ? ? ? ?
Pararhabdodon isonensis ? ? ? ? 0 1 0 0 2 1/2 1/2 1 ? ? ? ? ? ? ? ?
Parasaurolophus cyrtocristatus 1 0 1 1 0 1 0 1 2 2 2 0 0 1 1 1 1 ? ? ?
Parasaurolophus tubicen ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Sahaliyania elunchunorum ? ? ? 0 0 1 0 0 2 2 2 1 ? ? ? ? ? ? ? ?
Tsintaosaurus spinorhinus 1 0 1 0 0 1 0 0 2 2 2 1 1 2 ? ? 1 1 1 ?
Velaphrons coahuilensis ? ? ? ? 0 1 0 ? ? 2 2 1 ? ? ? ? 1 ? ? ?
251

Amurosaurus riabinini ? ? 1 1 2 ? 1 2 3 1 2 0 ? 2 ? ? ? ? ? ?
Charonosaurus jiayinensis ? ? 1 1 2 0 1 2 3 ? 2 0 1 2 0 0 0 0 1 1
Hypacrosaurus stebingeri ? ? 1 1 2 0 1 2 3 1 2 0 1 2 1 0 0 0 1 1
Lambeosaurus laticaudus ? ? 1 1 2 0 1 2 3 1 2 0 1 2 ? ? ? ? 1 1
Lambeosaurus magnicristatus 1 1 1 1 2 0 1 2 3 1 2 0 1 2 1 0 0 0 1 1
Nipponosaurus sachalinensis 1 1 ? ? ? ? ? 1/2 3 1 2 0 ? ? ? ? ? 0 1 1
Olorotitan ararhensis ? ? 1 1 2 0 1 2 3 1 2 0 1 2 1 0 0 0 1 1
Parasaurolophus cyrtocristatus ? ? 1 1 2 1 1 3 3 1 2 0 1 2 0 0 0 0 1 1
Parasaurolophus tubicen ? ? ? ? ? ? ? ? ? ? ? 0 ? 2 ? ? ? ? ? ?
Sahaliyania elunchunorum ? ? 0 1 2 1 1 1 3 1 2 0 1 2 1 0 0 0 1 1
Tsintaosaurus spinorhinus 1 1 1 1 2 0 1 2 3 0 2 0 1 2 1 1 1 0 1 1
Velaphrons coahuilensis ? ? 1 1 2 0 1 1 3 1 2 0 1 2 0 1 2 0 1 1
252

Amurosaurus riabinini 1 1 1 0 1 1 0 1 1 1 1 ? 1 0 2 1 1 1 ? 2
Charonosaurus jiayinensis 1 ? 1 ? ? ? ? ? 1 1 1 0 1 ? 2 1 1 1 ? ?
Hypacrosaurus stebingeri 1 1 1 0 1 1 0 1 1 1 1 1 1 0 2 1 1 1 1 ?
Lambeosaurus lambei 1 1 1 0 1 ? 0 1 1 1 1 1 1 1 2 1 1 1 1 2
Lambeosaurus laticaudus 1 1 1 0 1 ? 0 1 1 1 1 1 1 0 2 1 1 1 1 2
Lambeosaurus magnicristatus 1 ? 1 0 1 1 0 1 1 1 1 ? 1 1 2 1 1 1 1 2
Nipponosaurus sachalinensis 1 1 1 ? ? ? ? ? ? ? ? ? ? ? 1 0 2 1 1 ?
Olorotitan ararhensis ? ? 1 ? ? ? ? ? 1 1 1 ? 1 ? 2 1 1 1 1 2
Pararhabdodon isonensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? 2 ? ? ? ? ?
Parasaurolophus cyrtocristatus 1 ? 1 0 1 1 0 1 1 1 1 1 1 0 2 0 0 0 1 1
Parasaurolophus tubicen ? ? ? 0 1 ? ? ? 1 1 1 ? ? ? ? ? ? ? ? ?
Parasaurolophus walkeri ? ? 1 0 1 1 0 1 1 1 1 ? 1 0 2 1 1 1 1 2
Sahaliyania elunchunorum 1 1 1 0 1 1 0 1 1 1 1 0 1 1 ? ? 1 1 1 2
Tsintaosaurus spinorhinus 1 ? 1 1 1 1 0 1 1 1 1 ? 1 0 2 1 1 1 1 1
Velaphrons coahuilensis 1 1 1 ? ? ? ? 1 1 1 1 1 1 ? 2 1 1 1 1 2
253

Amurosaurus riabinini 0 1 1 ? ? ? 1 1 1 0 ? ? 0 1 ? ? ? ? ?
Aralosaurus tuberiferus ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Charonosaurus jiayinensis 0 ? 1 1 1 1 1 1 1 1 1 ? 0 1 ? 1 ? ? ?
Corythosaurus casuarius 0 2 1 0 1 0 1 1 1 1 1 1 0 1 0 1 1 1 0
Corythosaurus intermedius 0 2 1 0 1 0 1 1 1 1 1 1 0 1 0 1 1 1 0
Hypacrosaurus altispinus 0 2 1 1 1 0 1 1 1 0 1 1 0 0 0 1 1 1 0
Hypacrosaurus stebingeri 0 1 1 0 1 0 1 1 1 ? 1 ? 0 0 0 1 1 1 0
Jaxartosaurus aralensis ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Lambeosaurus lambei 0 1 1 0 1 0 1 1 1 1 1 1 0 1 1 1 1 1 0
Lambeosaurus laticaudus 0 2 1 0 1 1 ? ? ? ? ? ? 0 ? ? ? ? ? ?
Lambeosaurus magnicristatus 0 2 1 0 1 0 1 1 1 1 ? 1 0 1 ? ? ? 1 0
Nipponosaurus sachalinensis 0 ? 1 0 ? 0 1 ? ? ? ? ? 0 ? 0 0 ? 1 0
Olorotitan ararhensis 0 1 1 0 ? 0 1 1 1 ? ? ? ? ? ? ? ? ? ?
Pararhabdodon isonensis ? ? 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Parasaurolophus cyrtocristatus 0 2 1 1 1 1 1 1 1 1 1 1 0 1 ? ? ? 1 0
Parasaurolophus tubicen ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Parasaurolophus walkeri 0 1 1 ? 1 1 1 1 ? ? ? 1 ? ? ? ? ? ? 0
Sahaliyania elunchunorum 0 1 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
Tsintaosaurus spinorhinus 0 1 1 0 1 0 1 1 1 1 1 ? 0 1 0 1 1 1 0
Velaphrons coahuilensis 0 1 1 0 0 0 ? ? ? ? ? ? 0 1 ? ? ? 1 0
254
APPENDIX B
DENTAL CHARACTERS
Dentary Dentition
Character 1: maximum number of tooth positions in the dentary dental battery

(Table B.1; Fig. B.1; modified from Horner et al., 2004, character 1)
(0): 30 or less (sample mean of 22 alveolar positions)
(1): 31 to 42 (sample mean of 37 alveolar positions)
(2): more than 42 (sample mean of 49 alveolar positions)
Comments: the increase in the number of teeth in the dentary is very consistent
throughout iguanodontoidean ornithopods, with only minor differences among taxa in the
rate by which the number of teeth increases throughout ontogeny (Figs. B.2-B.4).
Character 2: number of alveolar positions per cm of dental battery (Table B.2; Figs.
B.5-B.7)
(0): less than 1.1 teeth/cm of dental battery (sample mean of 0.9)
(1): 1.1 or more teeth/cm of dental battery (sample mean of 1.3)
Comments: the tooth density in the dentary decreases at a very similar rate
throughout ontogeny in all studied iguanodontoidean taxa (Figs. B6-B7).
Character 3: minimum number of teeth per alveoli arranged dorsoventrally at mid
length of the dental battery (Table B.3; Fig. B.8; modified from Horner et al., 2004,
character 2)
(0): two
(1): three
(2): four
(3): five or more
Character 4: maximum number of functional teeth exposed on the dentary occlusal
plane (Fig. B.9; modified from Horner et al., 2004, character 3)
(0): one
(1): one functional tooth rostrally and caudally, and up to two teeth at and
approaching the middle of the dental battery
255
(2): three functional teeth throughout most of the dental battery, gradually
decreasing to two near the rostral and caudal ends of the dentary.
Character 5: height/width ratio of the dentary tooth crowns in lingual aspect (Table
B.4; Fig. B.10)
(0): ratio up to 1.95 (sample mean ratio of 1.6)
(1); ratio from 1.95 to 2.7 (sample mean ratio of 2.4)
(2): ratio from 2.8 to 3.3 (sample mean ratio of 3.0)
(3): ratio greater than 3.3 (sample mean ratio of 3.7)
Comments: quantification of this character in maxillary teeth is problematic.
Unlike dentary tooth crowns, the maxillary ones are very rarely exposed. Instead, in the
vast majority of exemplars only the ventral-most teeth are exposed and the maxillary
tooth crowns appear worn in a variety of degrees throughout the dental battery. I
attempted to measure half the height of those crowns more extensively exposed, but this
rendered no pattern of variation, probably because of errors in capturing the actual half of
the crown height.
Character 6: maximum number of ridges on the enameled lingual side of dentary
tooth crowns (Table B.5; Fig. B.11; modified Horner et al., 2004, character 6)
(0): presence of a primary major ridge extending from the ventral to the dorsal
end of the crown, a rostral and slightly shorter secondary ridge and several (three or
more) subdisidary, faintly developed and short tertiary ridges
(1): presence of primary, secondary, and one or two tertiary ridges
(2): presence of a primary ridge and one or two faint and shorter ridges
(3): loss of all but the primary ridge
Character 7: dentary tooth crowns, position of the primary ridge (Fig. B.12;
modified from Hai-lu You et al., 2003, character 39)
(0): well offset caudally from the midline
(1): median for most teeth, although some teeth within the same dental battery
may display a slight caudal offset of the primary ridge
Character 8: shape of the primary ridge of dentary tooth crowns (Fig. B.13;
modified from Godefroit et al., 2000 character 23)
(0): straight in all teeth within the same dentition
256
(1): straight for some crowns and sinuous for others
Comments: in those taxa with teeth with sinuous primary ridges, the proportion in
which those crowns occur in the dentition is variable. For example, whereas most tooth
crowns bear sinuous ridges in the Corythosaurus specimen TMP 82-37-01, Godefroit et
al. (2004) reported that the sinuous carina are present only on the mesial crowns of
Amurosaurus riabinini.
Character 9: angle between the crown and the root of dentary teeth (Table B.6; Fig.
B.14; modified from Godefroit et al., 2000, character 24)
(0): up to 110º (sample mean angle of approximately 105º), curved root (0)
(1): more than 110º and up to 135º (sample mean angle of approximately 125º),
straight root
(2): more than 135º (sample mean angle of approximately 140º), straight root
Comments: this character was originally used by Sternberg (1936) to distinguish
hadrosaurine from lambeosaurine hadrosaurids. Isolated dentary teeth with preserved
roots that can be referred to particular genera or species are rare in the fossil record. Thus,
only a very limited number of measurements could be recorded for specific taxa. Brett-
Surman (1989) indicated that this character may be too variable to be of diagnostic
utility. However, measurements on several Brachylophosaurus teeth indicate that this
feature is rather consistent taxonomically at least for this taxon.
Character 10: overall morphology of the dentary marginal denticles (Table B.7; Fig.
B.15; Norman, 2002, character 31)
(0): wedge to tongue-shaped
(1): curved and mammillated asymmetrical ledge
(2): absent or very reduced to small papillae along the apical half of the dorsal
half of the crown
Comments: the maximum number of denticles per cm of crown margin ranges
from 8 to 11 in most taxa with states 0 and 1. This density of denticles increases to 20-30
in taxa with state 2 (Table B.7).
Character 11: structure of the dentary marginal denticles (Fig. B.16)
(0): denticles with three or more indentations in its apical margin
(1): denticles with three separate and rounded knobs aligned labiomesially
257
(2): each denticle consisting of a single and rounded knob
Comments: in all taxa denticles become smaller and more densely packed at the
apex of the crown. Thus, the character states listed above concern the morphology of the
denticles along the margins of the dorsal half of the crown to the exclusion of the apex
itself.
Character 12: differences in denticulation between the mesial and distal crown
margins (Fig. B.17)
(0): both mesial and distal margins have approximately the same quantity of
denticles; the size of the denticles in both carinae is also similar.
(1): the mesial margin has larger and either more or less denticles than the distal
one; when that occurs, crowns tend to appear imbricated, so that the mesial margin
overlaps the distal margin of the adjacent crown.
Character 13: thickness of the septa that separate the dentary alveolar sulci (Fig.
B.18)
(0): finger-like septa, as thick as the adjacent alveolar sulci or, at least, thicker
than half the width of the adjacent alveolus
(1): sheet-like septa, thinner than half the width of the adjacent alveolar sulci
Character 14: morphology of the alveolar sulci (Fig. B.19; Norman, 2002, character
33)
(0): shaped by dentary crowns
(1): narrow and parallel-sided sulci
Character 15: distribution of the enamel of dentary crowns (Norman, 2002,
character 30)
(0): presence of a thin veneer labially, thick lingually
(1): only present lingually
Maxillary Dentition
Character 16: maximum number of tooth positions in the maxillary dental battery
(Table B.8; Figs. B.20-B.22; modified from Horner et al., 2004, character 1)
(0): up to 32 tooth positions (sample mean of 23 teeth)
258
(1): from 33 to 44 tooth positions (sample mean of 40 teeth)
(2): 45 or more tooth positions (sample mean of 49 teeth)
Comments: the increase in the number of teeth in the maxilla is very consistent
throughout iguanodontoidean ornithopods, with only minor differences among taxa in the
rate by which the number of teeth increases throughout ontogeny (Figs. B.21-B.22).
Character 17: number of maxillary teeth per centimeter of dental battery (Table
B.9; Figs. B.23-B.24)
(0): less than 1.25 teeth per cm of dental battery (sample mean ratio of 1.05)
(1): 1.25 or more teeth per cm of dental battery (sample mean ratio of 1.44)
Character 18: increase in the number of tooth positions in the maxilla relative to the
dentary (Table B.10; Fig. B25; modified from Wagner, 2001)
(0): absent
(1): present, maxillary dental battery with 5 to 20% more tooth positions than the
dentary one
Character 19: maximum number of functional teeth per alveolus in the maxillary
occlusal plane (Fig. B26; modified from Horner et al., 2004, character 3)
(0): one
(1): one tooth for most of the dental battery, with the sporadic presence of a
second tooth forming the occlusal plane
(2): two functional teeth throughout most of the dental battery length, gradually
changing to one near the rostral and caudal ends of the maxilla
Character 20: maximum number of ridges on the enameled labial side of maxillary
tooth crowns (Fig. B.27; Horner et al., 2004 character 7)
(0): presence of a primary major ridge and three or more much fainter ridges
(1): loss of all but the primary ridge in all or, at least, most of the crowns (in the
latter situation a few crowns show a fainter secondary ridge)
Comments: because maxillary tooth crowns are poorly and/or incompletely
exposed in most hadrosaurs, there is not enough resolution in the data to distinguish taxa
with only a primary ridge in their maxillary crowns from those with a primary ridge
accompanied by a secondary one (e.g., from 15 Lambeosaurus lambei skulls, only in one
259
specimen could this condition be clearly observed). Hence, those two observations were
merged into a single state.
Character 21: maxillary tooth crowns, position of the primary ridge (Fig. B.28-B.29;
modified from Hai-lu You et al., 2003 character 36)
(0): the dental battery contains a mixture of teeth with primary ridge positioned
caudally and teeth with the ridge at the center of the crown.
(1): the majority of teeth in the dental battery have a primary ridge positioned at
the midline of the crown.
Character 22: shape of the primary ridge of maxillary tooth crowns (Fig. B.30;
modified from Godefroit et al., 2000, character 23)
(0): straight in all teeth within the same dentition
(1): straight for some crowns and sinuous for others
Character 23: overall morphology of the maxillary marginal denticles (Fig. B.31;
Norman, 2002, character 30)
(0): wedge to tongue-shaped
(1): curved and mammillated asymmetrical ledge
(2): absent or reduced to small papillae along the apical half of the dorsal half of
the crown
260
Table B.1. Character 1. Values of the number (N) of alveolar positions of the dentaries
used to investigate the distribution and patterns of variation of this character, with the
corresponding character state coding derived from the K-Means clustering.
Specimen number Taxon N K-Means Cluster

(State)
AEHM 1/12 Amurosaurus riabini 37 1
AMNH 6553 Bactrosaurus johnsoni 21 0
SBDE 95E5-12 Bactrosaurus johnsoni 21 0
MOR 1071-7-13-99-93 Brachylophosaurus canadensis 33 1
MOR 1071 Brachylophosaurus canadensis 34 1
FMNH 862 Brachylophosaurus canadensis 35 1
CMN 8893 Brachylophosaurus canadensis 39 1
CUST unrecorded no. Charonosaurus jiayinensis 34 1
ROM 776 Corythosaurus intermedius 35 1
ROM 871 Corythosaurus casuarius 36 1
CM 11376 Corythosaurus intermedius 38 1
MOR 549 Hypacrosaurus stebingeri 41 1
ROM 658 Edmontosaurus sp. 50 2
BMNH R4862 Edmontosaurus annectens 55 2
UCM 42764 Edmontosaurus annectens 49 2
MOR 003 Edmontosaurus annectens 51 2
AMNH 5879 Edmontosaurus annectens 54 2
CM 9970 Edmontosaurus annectens 55 2
USNM 4807 Edmontosaurus annectens 46 2
CMN 2289 Edmontosaurus regalis 48 2
ROM 46259 Edmontosaurus regalis 52 2
PIN 2549-1 Bactrosaurus johnsoni 22 0
PASAC-1 Hadrosauridae 51 2
AMNH 5465 Gryposaurus latidens 39 1
RAM 6797 Gryposaurus monumentensis 40 1
MSNM V354 Gryposaurus notabilis 35 1
ROM 873 Gryposaurus notabilis 39 1
ROM 702 Hypacrosaurus altispinus 40 1
BMNH 28660 Iguanodon sp. 18 0
BMNH R1831 Iguanodon sp. 18 0
Norman 1980 Iguanodon bernissartensis 25 0
ROM 1218 Lambeosaurus lambei 36 1
261
Table B.1—continued.
Specimen number Taxon N K-Means Cluster

(State)
CMN 351 Lambeosaurus lambei 37 1
USNM 10309 Lambeosaurus sp. 38 1
TMP 2005-00-28 Lambeosaurus lambei 41 1
YPM-PU 22405 Maiasaura peeblesorum 37 1
OTM F138 Maiasaura peeblesorum 38 1
GDF 300 Ouranosaurus nigeriensis 23 0
NMMNH P-25100 Parasaurolophus tubicen 39 1
IPS SRA 27 Pararhabdodon isonensis 35 1
MOR 447 Prosaurolophus maximus 37 1
MOR 447 WSQ86 Prosaurolophus maximus 40 1
MOR 553Sa Prosaurolophus maximus 45 2
CMN 8894 Prosaurolophus maximus 46 2
MOR 553Sb Prosaurolophus maximus 47 2
TMP 84-1-1 Prosaurolophus maximus 48 2
SMU 74582 Protohadros byrdi 28 0
ZPAL MgD-I 162 Saurolophus angustirostris 42 1
AMNH 5221 Saurolophus osborni 48 2
BMNH R3386 Telmatosaurus transsylvanicus 27 0
IVPP V723 Tsintaosaurus spinorhinus 37 1
262
50
Number of Alveolar Positions in the Dentary
40
30
−450
−470
BIC
−490
−510
20
2 4 6 8
Number of Clusters (States)
Am Ba Br Ch Coi Cos Ed Eda Edr Grl Grm Grn Hy Hys Ig Igb Lm Lml Ma Ou Pat Ph Pr Pt Saa Sao Te Ts cfLm x
Taxon
Figure B.1. Character 1. Boxplot showing the distribution of the number of dentary alveolar positions in a sample of iguanodontoidean
taxa. The inserted smaller graph shows the Bayesian information criterion (BIC) for the various numbers of clusters in which the data can
be classified.
263
38
Bactrosaurus johnsoni Brachylophosaurus canadensis
20
36
Number of Alveoli
Number of Alveoli
34
18
32
16
30
28
14
26
10 15 20 A 20 22 24 26 28 30 32 B
Dental Battery Length (cm) Dental Battery Length (cm)
35
30
Number of Alveoli
25
20
15
10
5 10 15 20 25 30 C
Dental Battery Length (cm)
relative to the dental battery length for three species of hadrosaur ornithopods.
264
Lambeosaurus lambei
40
Number of Alveoli
35
30
20 25 30 A
38
40
Corythosaurus casuarius “Hypacrosaurus” stebingeri

35
Number of Alveoli
36
Number of Alveoli
30
34
25
20
32
15
10
30
22 24 26 28 30 32 B 5 10 15 20 25 30 35 C
Figure B.3. Character 1. Scatterplots of the number of alveolar positions in the dentary relative to
the dental battery length for three species of hadrosaurids.
265
55
55
Edmontosaurus spp.
50
Number of Alveoli
50
Number of Alveoli
45
40
45
35
40
30
25
35
10 20 30 40 50 A 30 35 40 45 50 B
38
45
36
Number of Alveoli
Number of Alveoli
34
40
32
30
35
28
26
30
20 25 30 35 40 45 C 20 25 30 35 D
Figure B.4. Character 1. Scatterplots of the number of alveolar positions in the dentary versus
the dental battery length for four hadrosaurid taxa.
266
Table B.2. Character 2. Values of the number of teeth per cm of dental battery (tooth
density; D) for the dentaries used to investigate the distribution and pattern of variation of
this character. Included are the character state codings derived from K-Means clustering.
Specimen Taxon D K-Means

Cluster (State)
AEHM 1/12 Amurosaurus riabinini 1.29 1
SBDE 95E5-12 Bactrosaurus johnsoni 0.96 0
AMNH 6553 Bactrosaurus johnsoni 1.19 1
MOR 1071-7-13-99-93 Brachylophosaurus canadensis 1.03 0
FMNH 862 Brachylophosaurus canadensis 1.22 1
MOR 1071 Brachylophosaurus canadensis 1.25 1
CMN 8893 Brachylophosaurus canadensis 1.34 1
CUST unrecorded no. Charonosaurus jiayinensis 1.21 1
ROM 871 Corythosaurus casuarius 1.16 1
ROM 776 Corythosaurus intermedius 1.19 1
CM 11376 Corythosaurus casuarius 1.19 1
MOR 549 Corythosaurus stebingeri 1.2 1
ROM 658 Edmontosaurus sp. 1.19 1
AMNH 5879 Edmontosaurus annectens 1.15 1
MOR 003 Edmontosaurus annectens 1.16 1
CM 9970 Edmontosaurus annectens 1.25 1
USNM 4807 Edmontosaurus annectens 1.34 1
UCM 42764 Edmontosaurus annectens 1.4 1
BMNH R4862 Edmontosaurus annectens 1.48 1
ROM 46259 Edmontosaurus annectens 1.3 1
CMN 2289 Edmontosaurus regalis 1.37 1
PIN 2549-1 Bactrosaurus johnsoni 0.78 0
PASAC-1 Hadrosaurinae 1.06 0
AMNH 5465 Gryposaurus latidens 0.93 0
RAM 6797 Gryposaurus monumentensis 0.98 0
MSNM V354 Gryposaurus notabilis 1.01 0
ROM 873 Gryposaurus notabilis 1.11 1
ROM 702 Hypacrosaurus altispinus 1.31 1
BMNH 28660 Iguanodon sp. 0.49 0
BMNH R1831 Iguanodon sp. 0.58 0
ROM 1218 Lambeosaurus lambei 1.15 1
USNM 10309 Lambeosaurus sp. 1.19 1
267

Cluster (State)
CMN 8703 Lambeosaurus lambei 1.26 1
TMP 2005-00-28 Lambeosaurus lambei 1.32 1
YPM-PU 22405 Maiasaura peeblesorum 1.22 1
OTM F138 Maiasaura peeblesorum 1.36 1
GDF 300 Ouranosaurus nigeriensis 0.89 0
NMMNH P-25100 Parasaurolophus tubicen 1.27 1
IPS SRA 27 Pararhabdodon isonensis 1.46 1
MOR 447 WSQ86 Prosaurolophus maximus 0.97 0
MOR 553S Prosaurolophus maximus 0.98 0
MOR 447 Prosaurolophus maximus 1.01 0
CMN 8894 Prosaurolophus maximus 1.02 0
TMP 84-1-1 Prosaurolophus maximus 1.14 1
SMU 74582 Protohadros byrdi 1.01 0
ZPAL MgD-I 162 Saurolophus angustirostris 1.09 0
AMNH 5221 Saurolophus osborni 1.33 1
BMNH R3386 Telmatosaurus transsylvanicus 1.4 1
IVPP V723 Tsintaosaurus spinorhinus 1.32 1
268
Number of Teeth per cm of Dentary Dental Battery (Dentary Tooth Density)
1.4
1.2
1.0
20
0.8
0
BIC
−20
−40
0.6
−60
2 4 6 8
Number of Clusters
Am Ba Br Ch Coi Cos Ed Eda Edr Grl Grm Grn Hy Hys Ig Lml Ma Ou Pat Ph Pr Pt Saa Sao Te Ts cfLm x
Taxon
Figure B.5. Character 2. Boxplot showing the distribution of the dentary tooth density in a sample of iguanodontoidean taxa. The inserted
smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which these data can be classified.
269
1.6
1.5
Brachylophosaurus canadensis Gryposaurus notabilis
1.5
1.4
Tooth Density.
Tooth Density
1.4
1.3
1.3
1.2
1.2
1.1
1.1
1.0
20 22 24 26 28 30 32 A 20 25 30 35 B
Dental Battery
. Length Dental Battery Length
1.45
1.40
Tooth Density
1.35
1.30
1.25
1.20
1.15
30 35 40 45 50 C
Dental Battery Length
Figure B.6. Character 2. Scatterplots of the dentary tooth density (teeth per cm of dental battery)
relative to the dental battery length for three species of hadrosaurids.
270
1.6
Prosaurolophus maximus Lambeosaurus lambei
1.3
Tooth Density
1.5
Tooth Density
1.2
1.4
1.1
1.3
1.2
1.0
20 25 30 35 40 45 A 20 25 30 B
Dental Battery Length Dental Battery Length
2.6
2.4
1.6
Corythosaurus casuarius Hypacrosaurus stebingeri

2.2
Tooth Density
Tooth Density
1.5
2.0
1.4
1.8
1.6
1.3
1.4
1.2
1.2
22 24 26 28 30 32 C 5 10 15 20 25 30 35 D
Figure B.7. Character 2. Scatterplots of the dentary tooth density (teeth per cm of dental battery)
versus the dental battery length for four species of hadrosaurids.
271
Table B.3. Character 3. Values of the number of teeth per alveolus (N) for various
hadrosaurid dentaries, with the corresponding character state coding derived from K-
Means clustering analysis.
Specimen Taxon N K-Means Cluster

(State)
AEHM 1/12 Amurosurus ribinini 3 1
MOR 549 “Hypacrosaurus” stebingeri 3 1
USNM 10309 Lambeosaurus sp. 4 2
FMNH P27393 Parasaurolophus cyrtocristatus 4 2
MOR 447a Prosaurolophus maximus 4 2
MOR 447b Prosaurolophus maximus 4 2
272
Specimen Taxon N K-Means Cluster

(State)
MOR 447c Prosaurolophus maximus 4 2
MOR 553S Prosaurolophus maximus 5 3
273
6
Minimum Number of Teeth per Alveolus in the Dentary
5
−60
−80
BIC
−120
4
−160
E V
2 4 6 8
Number of Clusters
3
2
Am Ba Br Ch Co Hys Ed Eda Edr Eq Ba Gr Grl Grm Grn Hy Ig Lml Ma Ou Pac Pat Ph Pr Pt Saa Sao Te Ts x
Taxon
Figure B.8. Character 3. Boxplot showing the distribution of the number of teeth per alveolus in dentary of a sample of iguanodontoidean
taxa. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which these data can be
classified. The upper photograph shows four teeth per alveolus in Gryposaurus latidens (AMNH 5465).
274
Maximum Number of Teeth on the Dentary Occlusal Plane
3.0
2.5
2.0
1.5
A
1.0
Am Ba Br Ch Co Hys Ed Eda Edr Eq Gi Gr Grl Grm Grn Hy Ig Lml Ma Ou Pac Pat Ph Pr Pt Saa Sao Te Ts x
Taxon
D
B
Figure B.9. Character 4. A, boxplot showing the distribution of the number of teeth exposed in the
dentary occlusal plane of a sample of iguanodontoidean taxa. Teeth counted at mid-length of the dental
battery. B, occlusal plane of the left dentary of cf. Lambeosaurus sp. (USNM 10309), where arrows point
to individual teeth.
275
Table B.4. Character 5. Values of the height/width ratio (H/W) of tooth crowns for the
dentaries used to investigate the distribution and pattern of variation of this character,
with the corresponding character state coding derived from clustering using K-Means.
Specimen Taxon H/W K-means

Cluster (State)
CM 11376 Corythosaurus intermedius 2.9 2
MOR 549 Hypacrosaurus stebingeri 4.1 3
ROM 46259 Edmontosaurus regalis 3.3 2
IVPP V12534 Equijubus normani 1.6 0
IPS 915 Iguanodon sp. 1.4 0
FMNH P27393 Parasaurolophus cyrtocristatus 3.2 2
276
Specimen Taxon H/W K-means

Cluster (State)
MOR 447b Prosaurolophus maximus 2.8 2
MOR 447a Prosaurolophus maximus 3.0 2
MOR 447c Prosaurolophus maximus 3.0 2
277
4.0
Height / Width Ratio of Dentary Tooth Crowns
3.5
3.0
−100
−120
2.5
BIC
−140
−160
2.0
E V
2 4 6 8
Number of Clusters
1.5
Am Ba Br Coi Cos Eda Edr Eq Grl Grm Grn Hy Hys Ig Lml Ma Ou Pac Pat Pr Pt Saa Sao Ts cfLm
Taxon
Figure B.10. Character 5. Boxplot showing the distribution of the height/width ratio of dentary tooth crowns in a sample of iguanodontoidean
taxa. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which these data can be
classified. The upper photograph shows how this ratio was measured, exemplified in the dental battery of Edmontosaurus annectens (MOR 003).
278
Table B.5. Character 6. Values of the number of ridges (N) in the tooth crowns of several
hadrosaurid dentaries, with the character coding derived from K-Means clustering. A
“0.5” added to an integer value indicates the presence of a fainter secondary ridge.
Specimen Taxa N K-Means

Cluster (State)
CUST unrecorded no. Charonosurus jiayinensis 1 3
CUST unrecorded no. Charonosurus jiayinensis 1 3
CM 11376 Corythosaurus casuarius 1 3
MOR 549 “Hypacrosaurus” stebingeri 1.5 2
IVPP V12534 Equijubus normani 7 0
IPS 915 Iguanodon sp. 6 0
USNM 10309 Lambeosaurus lambei 1.5 2
279
Specimen Taxa N K-Means

Cluster (State)
MOR 447a Prosaurolophus maximus 1 3
MOR 447b Prosaurolophus maximus 1 3
MOR 447c Prosaurolophus maximus 1 3
280
281
state 0 state 0
A B
state 1 state 1
C D
Figure B.12. Character 7. Dentary teeth showing the position of the primary ridge (arrows) on
their lingual enameled sides. A, Hypacrosaurus stebingeri (MOR 549). B, Protohadros byrdi (SMU 74582).
C, Edmontosaurus regalis (CMN 2289). D, Prosaurolophus maximus (CMN 8894).
282
state 0 state 1
A B
state 1 state 1
C D
Figure B.13. Character 8. Dentary teeth showing the absence or present of sinuousity (arrows)
in the primary ridge on their lingual enameled sides. A, Saurolophus osborni (AMNH 5221).
B, Brachylophosaurus canadensis (FMNH 862). C, Edmontosaurus annectens(BMNH R4862).
D, cf. Lambeosaurus sp. (USNM 10309).
283
Table B.6. Character 9. Values of the crown-root angle (A) of the dentary teeth used to
investigate the distribution and patterns of variation of this character, with the
corresponding character state coding derived from K-Means clustering.
Specimen/Source Taxon A K-Means Cluster

(State)
Godefroit et al., 2004 Amurosaurus riabinini >145 2
Godefroit et al., 1998 Bactrosaurus johnsoni >145 2
Godefroit et al., 2001 Charonosarus jiayinensis >145 2
CMN 8676 Corythosaurus casuarius 142 2
Langston, 1960 Gryposaurus notabilis 130 1
IPS 915 Iguanodon sp. 105 0
Langston, 1960 Lophorhothon atopus .140 2
LACM 17713 “Lambeosaurus” laticaudus 142 2
USNM 10309 Lambeosaurinae 145 2
FMNH P27393 Parasauriolophus cyrtocristatus 140 2
Norman, 2002 Probactrosaurus gobiensis >145 2
Horner, 1992 Prosaurolophus maximums 120 1
284
140
Crown-Root Angle of Dentary Teeth
130
−160
120
−170
BIC
−180
−190
−200
E V
110
2 4 6 8
Number of Clusters
Am Ba Br Ch Co Eda Gr Ig Lh Lm Ou Pac Pb Pr Pt Ts x
Taxon
Figure B.14. Character 9. Boxplot showing the distribution of the angle between the root and the crown of the dentary teeth in a
sample of iguanodontoidean taxa. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters
in which the data can be classified. The two photographs show how this angle was measured, exemplified in A, Brachylophosaurus canadensis
(MOR 1071) and B, Iguanodon sp. (IPS 915).
285
state 0
A
state 1
B
state 2
Figure B.15. Character 10. Dentary teeth showing marginal denticles (inside the white rectangles)
on their lingual enameled sides. A, Iguanodon sp. (BMNH R1831). B, Lambeosaurinae from Dinosaur
Provincial Park (?Lambeosaurus; UALVP 11734). C, Edmontosaurus regalis (CMN 2289).
286
Table B.7. Character 7. Number of denticles per cm of crown margin (N) in the dentary
teeth of a sample of hadrosaurid dinosaurs. Ontogenetic stages are approximate based on
the relative size of the exemplars.
Specimen Taxon N Ontogenetic

State
AMNH 6581 Bactrosaurus johnsoni 15 Subadult
FMNH 862 Brachylophosaurus canandensis 9 Adult
MOR 1071-7-15-98-226 Brachylophosaurus canandensis 10 Subadult
MOR 1071 Brachylophosaurus canandensis 9 Adult
CUST K101 Charonosaurus jiayinensis 16 Unknown
ROM 868 Corythosaurus casuarius 8 Adult
MOR 549 “Hypacrosaurus” stebingeri 8 Adult
TMP 88-151-75 Hadrosauridae 8 Subadult
CEUM 9758 Eolambia caroljonesa 9 Adult
IVPP V12534 Equijubus normani 10 Adult
AMNH 5465 Gryposaurus latidens 10 Adult
BMNH R1831 Iguanodon sp. 6 Adult
BMNH R11521 Iguanodon sp. 12 Subadult
IPS 915 Iguanodon sp. 11 Adult?
AMNH 5799 Kritosaurus navajovius 9 Adult
AUMP 2295 Lophorhothon atopus 15 Subadult?
FMNH P27383 Lophorhothon atopus 13 Subadult
CMN 2869 Lambeosaurus lambei 8 Adult
CMN 8633 Lambeosaurus sp. 10 Subadult
ROM 794 Lambeosaurus lambei 9 Adult
UALVP 11734 Lambeosaurinae 8 Adult
USNM 10309 Lambeosaurus sp. 8 Adult
GDF 300 Ouranosaurus nigeriensis 11 Adult
FMNH P27393 Parasurolophus cyrtocristatus 9 Adult
NMMNH P25100 Parasaurolophus tubicen 10 Adult
SMU 74582 Protohadros byrdi 10 Adult
ZPAL MgDI-162 Saurolophus angustirostris >20 Adult
BMNHR3386 Tematosaurus transsylvanicus 9 Adult
IVPP V768 Tsintaosaurus spinorhinus 10 Adult
287
state 0 state 1 state 2
A B C
Figure B.16. Character 11. Structure of the marginal denticles of dentary teeth. A, Lambeosaurinae indeterminate from Dinosaur
Provincial Park; the arrows point to two indentations on a denticle. B, Lophorhothon atopus (AUMP 2295). C, Edmontosaurus cf. regalis
(unrecorded catalog number) from Roth Mason Quarry (South Dakota).
288
state 0 state 0
A B
state 1 state 1
C D
Figure B.17. Character 12. Comparison of the quantity and/or size of the denticles (white
rectangles) between the rostral and caudal margins of dentary tooth crowns. A, Bactrosaurus johnsoni
(AMNH 6581). B, Hadrosaurus foulkii (nomen dubium, ANSP 9201). C, Corythosaurus casuarius
(ROM 870). D, Lambeosaurinae indeterminate from Dinosaur Provincial Park (Alberta, Canada).
289
state 0
A
state 1
Figure B.18. Character 13. Mediolateral thickness of the dentary alveolar sulci (arrows).
A, Iguanodon sp. (BMNH R754). B, Brachylophosaurus canadensis (MOR 10717-25-98-405).
290
state 0
A
state 1
Figure B.19. Character 14. Shape of the dentary alveolar sulci (arrows). A, Iguanodon sp.
(BMNH R5764). B, Brachylophosaurus canadensis (MOR 1071-8-15-98-574).
291
Table B.8. Character 16. Values of the number of alveolar positions (N) of the maxillae
used to investigate the distribution and patterns of variation of this character, with the
Specimen Taxon N K-Means cluster

(state)
AEHM 1/12 Amurosaurus riabinini 39 1
SBDE 95E5 cast 1 Bactrosaurus johnsoni 21 0
SBDE 95E5 cast 2 Bactrosaurus johnsoni 20 0
UCMP 130139 Brachylophosaurus canadensis 41 1
USNM 10309 cf. Lambeosaurus sp. 43 1
TMP 82-37-01 Corythosaurus casuarius 35 1
MOR 601 Edmontosaurus sp. 51 2
CEUM 9758 Eolambia caroljonesa 30 0
MOR 553-7-24-8-68 Gryposaurus sp. 33 1
MOR 553S-8-26-9-55 Gryposaurus sp. 42 1
MOR 478-5-28-8-1 Gryposaurus latidens 33 1
CMN 8675 Hypacrosaurus altispinus 42 1
MOR 549-6-19-8-9 Hypacrosaurus stebingeri 45 2
BMNH R11521 Iguanodon atherfieldensis 20 0
AEHM 1/319 Kerberosaurus manakini 50 2
NMMNH P-16106 Kritosaurus navajovius 44 1
FMNH PR380 Lambeosaurus lambei 39 1
292
Specimen Taxon N K-Means cluster

(state)
MNA PI 520 Parasaurolophus tubicen 40 1
Unrecorded no. Parasaurolophus tubicen 43 1
IPS-SRA 22 Pararhabddon isonensis 35 1
MOR 454-6-24-6-2 Prosaurolophus maximus 44 1
MOR 447-WSQ 86 Prosaurolophus maximus 48 2
TMP 84.1.1 Prosaurolophus maximus 52 2
BMNH R4911 cf. Telmatosaurus 29 0
FGGUB R1010 cf. Telmatosaurus 31 0
293
50
Number of Maxillary Alveolar Positions
40
−380
BIC
−400
30
−420
−440
2 4 6 8
Number of Clusters
20
Am Ba Br Co Ed Edr Eo Ba Gr Grl Grm Grn Hy Hys Ig Iga Kb Krn Lml Ma Ou Pat Ph Pr Pt cfTe cf Lm
Taxon
Figure B.20. Character 16. Boxplot showing the distribution of the number of maxillary alveolar positions in a sample of
iguanodontoidean taxa. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters
in which the data can be classified.
294
23
22
Number of Alveoli
21
20
19
18
17
16
8 10 12 14 16 18 20
A
44
42
Number of Alveoli
40
38
36
34
32
18 20 22 24 26 28 30 32
B
relative to the dental battery length for two species of hadrosaur ornithopods.
295
55
50
Edmontosaurus spp. Prosaurolophus maximus
Number of Alveoli
Number of Alveoli
50
45
45
40
40
35
35
30
25
15 20 25 30 35 40 A 20 25 30 35 B
45
Lambeosaurus lambei
Number of Alveoli
40
35
30
20 25 30 C
relative to the dental battery length for three species of hadrosaurid ornithopods.
296
Table B.9. Character 17. Values of the number of teeth per cm of maxillary dental battery
(tooth density, D) for the maxillae used to investigate the distribution and pattern of
variation of this character. Included is the corresponding character state coding derived
from K-Means clustering analysis.

Cluster (State)
SBDE 95E5 Bactrosaurus johnsoni 1.06 0
SBDE 95E5 Bactrosaurus johnsoni 1.00 0
MOR 1071-7-6-98-79 Brachylophosaurus johnsoni 1.45 1
UCMP 130139 Brachylophosaurus johnsoni 1.39 1
CMN 8893 Brachylophosaurus johnsoni 1.41 1
USNM 10309 cf. Lambeosaurus sp. 1.39 1
CMN 8676 Corythosaurus casuarius 1.61 1
TMP 82-37-01 Corythosaurus casuarius 1.54 1
MOR 601 Edmontosaurus sp. 1.31 1
MOR 553-7-24-8-68 Gryposaurus sp. 1.32 1
MOR 553S-8-26-9-55 Gryposaurus sp. 1.35 1
MOR 478-5-28-8-1 Gryposaurus latidens 1.22 0
CMN 8675 Hypacrosaurus altispinus 1.50 1
MOR 549-6-19-8-9 Hypacrosaurus stebingeri 1.43 1
BMNH R11521 Iguanodon atherfieldensis 1.11 0
NMMNH P-16106 Kritosaurus navajovius 1.42 1
FMNH PR380 Lambeosaurus lambei 1.39 1
297

Cluster (State)
MNA PI 520 Parasaurolophus tubicen 1.43 1
Unrecorded no. Parasaurolophus tubicen 1.39 1
IPS-SRA 22 Pararhabdodon isonensis 1.30 1
MOR 447-WSQ 86 Prosaurolophus maximus 1.41 1
TMP 84.1.1 Prosaurolophus maximus 1.37 1
MOR 454-6-24-6-2 Prosaurolophus maximus 1.52 1
BMNH R4911 cf. Telmatosaurus 1.53 1
FGGUB R1010 cf. Telmatosaurus 1.72 1
298
Number of Teeth per cm of Maxillary Dental Battery (Maxilary Tooth Density)
1.6
1.4
20
1.2
BIC
0
−20
−40
1.0
2 4 6 8
Number of Clusters
Am Ba Br Co Ed Edr Gi Gr Grl Grm Hy Hys Ig Iga Kb Krn Lml Ma Ou Pat Ph Pr Pt cfTe cfLm
Taxon
Figure B.23. Character 17. Boxplot showing the distribution of the maxillary tooth density in a sample of iguanodontoidean taxa.
The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which these data can be classified.
299
1.8
Bactrosaurus johnsoni Brachylophosaurus canadensis
1.8
Maxillary Tooth Density

1.7
1.6
1.6
1.4
1.5
1.2
1.4
1.0
8 10 12 14 16 18 20 A 18 20 22 24 26 28 30 32 B
1.75
Lambeosaurus lambei
1.70
1.65
1.60
1.55
1.50
1.45
1.40
20 25 30 C
Dental Battery Length
Figure B.24. Character 17. Scatterplots of the maxillary tooth density (teeth per cm of dental
battery) relative to the dental battery length for three species of hadrosaurids.
300
Table B.10. Character 18. Comparison of the number of dentary and maxillary alveolar
positions in various hadrosaurid specimens for which both elements are preserved.
Specimen Taxon Number of Number of

maxillary teeth dentary teeth
CMN 8676 Corythosaurus intermedius 37 35
MOR 549-6-19-8-9 Hypacrosaurus stebingeri 45 41
OTM F138 Mariasaura peeblesorum 44 38
301
y = 0.8354x + 2.7194
50
Number of Maxillary Teeth
Rsq = 0.9299
40
30
20
20 25 30 35 40 45 50
Number of Dentary Teeth
Figure B.25. Character 18. Scatterplot of the number of dentary teeth and that of the
maxillary teeth a sample of iguanodontoidean specimens.
302
state 0
A
state 1
B
state 2
Figure B.26. Character 19. Maxillary occlusal planes of various iguanodontoidean ornithopods,
showing the number of funcitonal teeth (arrows) exposed per alveolus. A, Iguanodon atherfieldensis
(BMNH R11521). B, cf. Telmatosaurus (BMNH R4911). C, Corythosaurus intermedius (CMN 8676).
303
state 0 state 0
A B
state 1 state 1
C D
Figure B.27. Character 20. Maxillary tooth crowns showing the primary and, when present,
secondary and auxilliary ridges on their buccal enameled sides. A, Iguanodon atherfieldensis (BMNH
11521). B, Iguanodon sp. (BMNH R454). C, Brachylophosaurus canadensis (MOR 1071-8-15-98-573).
D, Lambeosaurus lambei (CMN 351).
304
state 0
A
state 0
B
state 1
(arrows) on their buccal enameled sides. A, Bactrosaurus johnsoni (juvenile specimen, AMNH 6390).
B, Bactrosaurus johnsoni (cast of SBDE 95/E5). C, Amurosaurus riabinini (cast of AEHM 1/12).
305
state 0 state 1
A B
Figure B.29. Character 21. Maxillary tooth crowns showing the position of the primary ridge (arrows) on their buccal enameled sides.
A, Equijubus normani (IVPP V12534), with asymmetrically positioned primary ridge. B, Gilmoreosaurus mongoliensis (AMNH 6551), with
symmetrically positioned primary ridge.
306
state 0 state 1
A B
Figure B.30. Character 22. Maxillary tooth crowns showing the absence (A) or presence (B) of sinuosity in the primary ridge (arrows).
A, Edmontosaurus regalis (CMN 2289). B, Brachylophosaurus canadensis (CMN 8893).
307
A B C
Figure B.31. Character 23. Maxillary tooth crowns showing the size and shape of the marginal denticles (arrows).A, Telmatosaurus
transsylvanicus (BMNH R3388). B, Gilmoreosaurus mongoliensis (AMNH 6551). C, Edmontosaurus regalis (CMN 2289).
308
APPENDIX C
MANDIBULAR CHARACTERS
Predentary
Character 24: ratio between the predentary maximum mediolateral width and the
maximum rostrocaudal length along the lateral process (Table C.1; Fig. C.1;
modified from Horner et al., 2004, character 13)
(0): less than 1.2 (sample mean ratio of 1)
(1): between 1.2 and 1.75 (sample mean ratio of 1.5)
(2): more than 1.75 (sample mean ratio of 2)
Character 25: ratio between the dorsoventral depth of the predentary rostral face
(excluding the median ventral process) and the length of the lateral process (Table
C.2; Fig. C.2)
(1): ratio of 0.38 or less (sample mean ratio of 0.33)
Character 26: orientation of the rostral surface relative to the dorsal margin of the
lateral process (Table C.3; Fig. C.3; modified from Horner et al., 2004, character
14)
(0): angle of 75º or greater (sample mean angle of 81º)
(1): angle between 56º and 74º (sample mean angle of 66º)
(3): angle of 40º or less, gently rounded rostral surface (sample mean angle of
34º)
Character 27: shape of the denticles of the predentary oral margin (Fig. C.4;
(0): triangular and pointed
(1): subrectangular to rectangular
Comments: denticle shape could only be appreciated in a fraction of the preserved
predentaries because in most exemplars the oral margin is abraded or incompletely
preserved. In those cases where denticles could be observed these were often only
309
partially preserved. Most complete articulated skulls have predentaries with oral margins
so abraded that denticles are missing. However, the finding of disarticulated predentaries
of taxa for which articulated skulls are recorded showed that denticles were indeed
present in all hadrosaurids (contra Weishampel et al., 1993 and Wagner, 2001). In several
cases denticle shape appeared subtriangular with truncated tips (e.g., B. canadensis, Fig.
C.4B). However, it was very difficult to assess how breakage and abrasion modified
denticle morphology. For this reason, I decided to consider only the two states described
above.
Character 28: predentary denticle spacing (measured between mid height of two
consecutive denticles (Fig. C4)
(0): the separation between two consecutive denticles is larger than or equal to the
basal width of each individual denticle
(1): the separation between two consecutive denticles is smaller than the basal
width of each individual denticle but larger than 25% of its basal bredth
(2): denticles are tightly arranged, with no substantial separation between their
bases (less than or with a separation equal to 25% the basal denticle width)
Character 29: number of predentary denticles in adult individuals lateral to the
median denticle (not included in the count) (Figs. C.5-C.7)
(0): maximum of five
(1): six
(2): more than six
Comments: the number of predentary denticles appeared to increase slightly
through ontogeny, at least in some taxa. For example, undescribed juvenile
Edmontosaurus remains from Alaska have only three or four denticles lateral to the
median one (adults probably had a minumum of five denticles, as in CMN 8399).
However, this ontogenetic variation may not be present in all hadrosaurids; for example,
both juveniles (TMP 83-64-3) and adults or larger subadults referable to Prosaurolophus
maximus have four denticles in their predentaries.
Character 30: extension of the predentary denticulate margin (Fig. C.5-C.7)
(0): denticles extending into the lateral process
(1): denticles limited to the rostral margin
310
Character 31: morphology of the predentary rostrolateral corner (Fig. C.5-
C.7; modified from Horner et al., 2004, character 13)
(0): gently rounded and continuous with the lateral process, giving the predentary
an arcuate dorsal profile
(1): subsquared rostrolateral corner
(2): subsquared, very broad and rostrolaterally projected
Character 32: development of a lateral shelf on the dorsal side of the
predentary lateral process: (Fig. C.5-C.7; modified from Wagner, 2001)
(0): absence of shelf, presence of a rostrocaudally short and shallow groove
limited to the distal region of the lateral process, bounded by a tall lateral wall
(1): short and shallow shelf limited to the laterocaudal region of the lateral
process
(2): short and well-incised shelf that is wider near the rostrolateral corner of the
predentary
(3): shelf extremely narrow mediolaterally and very long rostrocaudally
(4): shelf rostrocaudally long, deeply incised and mediolaterally broad, forming
half of the mediolateral breadth of the lateral process and becoming wider distally
Character 33: morphology of the caudal end of the predentary lateral
process (Fig. C.8)
(0): lingual margin extending more caudally than the lateral, shelf-bearing portion
(1): strongly bifid caudal end of the lateral process, with a lateral shelf-bearing
region that extends further caudally than the lingual margin
Character 34: ridge on the dorsal lingual, keel-like process of the predentary
(Fig. C.9)
(0): the process lacks a prominent median ridge on the lingual side of the rostral
region of the predentary, and, if present, the former forms and projects caudally from the
caudal margin of the predentary rostral region.
(1): the process has a well-developed ridge on the lingual surface of the rostral
segment of the predentary, from which the former extends further caudally to lie dorsal to
the dentary symphysis.
311
Character 35: length of the predentary ventral median process (including
undivided and bilobate portions (Fig. C.10)
(0): process longer than the depth of the rostral face of the predentary (excluding
denticles)
(1): process as long as, or shorter, than the depth of the rostral face of the
predentary
Character 36: ventral median process, degree of indentation of the split of
the process into two distinct lobes (Fig. C.11)
(0): short indentation and deep undivided portion, the splitting originates at a
distance from the predentary ventral margin that equals approximately half of the
mediolateral width of the ventral process;
(1): long indentation and shallow undivided portion, the splitting originates at a
distance from the predentary ventral margin that is less than the mediolateral width of the
process.
Dentary
A few features of the dentary were found intraspecifically variable and not useful
for phylogenetic inference. For example, the exposure of the Meckelian canal along the
lingual and ventral margin of the dentary was found variable intraspecifically (e.g.,
Brachylophosaurus canadensis MOR 1071 specimens). In outgroup taxa to hadrosaurids,
the Meckelian canal appeared wider. However, this was probably a consequence of the
dental battery not extending beyond the level of the coronoid process.
Within a specimen, the degree of concavity of the dentary occlusal plane was
found to vary from nearly flat to strongly concave. For example, in Prosaurolophus
maximus MOR 447 the occlusal plane was flat, whereas in CMN 8894 it was deeply
concave. Another example was found in Lambeosaurus lambei, from nearly flat in CMN
2869 to strongly concave in TMP 81-31-01.
The orientation of the dentary alveoli was also observed to be variable
intraspecifically. Among specimens of numerous taxa (e.g., Pararhabdodon,
Tsintaosaurus, Telmatosaurus, Brachylophosaurus, Maiasaura, Charonosaurus,
312
Lambeosaurus, Corythosaurus, Parasaurolophus, etc.) the alveoli were vertically
oriented along the caudal and median regions of the dental battery, but caudally inclined
along the rostral region of the battery as the ventral row of special foramina curved
dorsally.
In the sampled taxa, the occlusal plane sloped labioventrally forming an angle of
approximately 45º with the parasagittal plane of the coronoid process. Horner et al.
(2004) indicated that the occlusal plane of Prosaurolophus is the steepest among
hadrosaurids. However, this feature was found to be variable within a species (e.g.,
Brachylophosaurus canadensis FMNH 862 and CMN 8893, or Lambeosaurus lambei
CMN 8633 with ROM 794). Even within the same dentition, the orientation of the
occlusal plane was variable as the degree of wear varies along the tooth row. Thus, in
general the occlusal plane became steeper as wear increased, usually along the median-
caudal region of the tooth row (e.g., “Hypacrosaurus” stebingeri MOR 549 or even most
notably in Gryposaurus latidens AMNH 5465).
The apical curvature of dentary tooth crowns was regarded as autapomorphic for
Telmatosaurus transsylvanicus rather than its loss being synapomorphic for the
Euhadrosauria (Wagner, 2001).
Character 37: ratio between the length of the proximal edentulous slope of the
dentary and the distance between the rostralmost tooth position and the caudal
margin of the coronoid process (Table C.4; Fig. C.12)
(0): less than 0.20 (sample mean ratio of 0.11)
(1): ratio between 0.20 and 0.31 (sample mean ratio of 0.27)
Character 38: angle between the dentary proximal-most edentulous slope and the
horizontal (Table C.5; Fig. C.13)
(0): less than 150º (sample mean angle of 144º)
Character 39. Angle between the proximal slope of the edentulous margin of the
dentary that articulates with the predentary and the horizontal (Table C.6; Fig.
C14-C.16)
313
(0): angle greater than 128º (sample mean angle of 137º)
(1): angle between 113 and 128º (sample mean angle of 121º)
(2): angle less than 113º (sample mean angle of 107)º
Comments: this character showed positive allometry throughout ontogeny (Figs.
C.15-C.16). In Edmontosaurus annectens this feature increased with a relatively higher
rate (Fig. C.16B).
Character 40: angle of deflection of the rostral ventral margin of the dentary (Table
C.7 and Figs. C.17-C.19)
(0): angle less than 17º (sample mean angle of 13º)
Comments: the rate of increase in ventral deflection of the dentary throughout
ontogeny was found variable among hadrosaurids (Figs. C.18-C.19). Brachylophosaurus
canadensis appeared to show a decrease in ventral deflection later in ontogeny, whereas
the opposite was found in the closely related Maiasaura peeblesorum, as well as in
lambeosaurines such as Lambeosaurus lambei and “Hypacrosaurus” stebingeri.
Character 41: location of the origination of the ventral deflection the dentary
(measured as the ratio between the distance from the caudal margin of the coronoid
process to the inflexion point of the ventral margin and the distance from the caudal
margin the coronoid process to the rostralmost alveolus) (Table C.8; Fig. C.20-C.22)
(0): the deflection occurs near the rostral end of the dentary, ratio greater than
0.78 (sample mean ratio of 0.87)
(2): deflection originating near the middle of the dental battery, ratio of 0.65 or
less (sample mean ratio of 0.59)
Comments: in those taxa were the dentary ventral deflection increased with age
(Maiasaura peeblesorum, Lambeosaurus lambei, “Hypacrosaurus” stebingeri) the
position of the origin of the deflection shifted towards the middle of the dental battery
(Figs. C.21-C.22).
314
Character 42: lingual projection symphyseal region of the dentary (measured as a
ratio between the labiolingual extension of the symphyseal region and the maximum
labiolingual width of the dentary) (TableC.9; Figs. C.23-C.24)
(0): ratio up to 1.65 (sample mean ratio of 1.47)
(1): ratio greater than 1.65 and up to 2.85 (sample mean ratio of 1.87)
(2): extremely elongated rostral end of the dentary, ratio greater than 2.85 (sample
mean ratio of 3.03)
Character 43: orientation of the dentary symphysis (measured as the angle formed
by this surface and the lateral side of the rostral half of the dentary) (Table C.10;
Figs. C.25-C.26)
(1): angle up to 15º (sample mean angle of 10º)
Comments: this character showed a decrease throughout ontogeny.
Character 44. Degree of lingual curvature of the symphyseal region of the dentary
(Fig. C.27)
(0): curved lingually into a 90º arch, subhorizontal rostral-most region with
slightly convex or flat ventral side
(1): gently curved lingually forming a wide arch, labial convex side facing
labioventrally
Character 45: medial or lateral profile of the dorsal margin of the rostral
edentulous region of the dentary for articulation with the predentary (Fig. C.28)
(0): ranging from having a very subtle concavity (almost straight) to straight or
even displaying a subtle convexity
(1): having a well-pronounced concavity
Character 46: bulging of the ventral lateral margin of the dentary (Fig. C.29;
modified from Wagner, 2001)
(0): ventral dentary margin expanded ventral to the coronoid process
(1): margin straight along the caudal half of the dentary and slightly bowed along
the rostral half
(2): margin straight or slightly bowed rostral to the coronoid process
315
(3): margin with a wide and well-developed ventral bulge rostral to the coronoid
process
Character 47: angle between the long axis of the coronoid process and the dorsal
margin of the alveolar sulci of the dental battery (Table C.11; Fig. C.30)
(0): coronoid process subvertical or caudally inclined, angle greater than 82º
(sample mean angle of 101º)
(1): process rostrally inclined with an angle between 69º and 82º (sample mean
angle of 73º)
(2): coronoid process rostrally inclined with an angle up to 68º (sample mean
angle of 66º)
Comments: subadult and adult specimens showed no difference in the inclination
of the coronoid process.
Character 48: morphology of the apex of the coronoid process (in adults) (Fig. C.31;
(0): slightly expanded rostrocaudally, with very limited development of rostral
and caudal expansions resulting in an apex that is taller than wider
(1): well developed expansion of both the caudal and, especially, the rostral
margins
Comments: this character may be variable ontogenetically. For example, large
dentaries of Edmontosaurus have more expanded processes than undescribed juvenile
exemplars recently collected in Alaska. Lambeosaurines appeared to display a less
pronounced caudal margin of the process than other hadrosaurids. However, signs of
abrasion and breakage were practically always present, so that it remained uncertain
whether lambeosaurines possessed a caudally less expanded the coronoid process than,
for example, saurolophines such as Edmontosaurus.
Character 49: caudodorsal margin of the coronoid process projected dorsally into a
sharp point (Fig. C.31B)
(0): absent
(1): present
Character 50: thick and dorsoventrally elongated ridge on the medial side of the
coronoid process, located near the caudal margin of the process (Fig. C.32)
316
(0): absent, presence of fine striations;
(1): present, the ridge forms the rostral boundary of a depressed facet for
attachment of the rostrodorsal process of the surangular; coarse striations present rostral
to the ridge.
Character 51: lateral expansion of caudal region of the dentary, ventral to the base
of the coronoid process (measured as the angle between the lateral surface of the
dentary and that of the region caudoventral to the coronoid process) (Table C.12;
Fig. C.33):
(0): the lateral side of the dentary is only sightly expanded laterally ventral to the
coronoid process, with an angle greater than 165º (sample mean angle of 171º);
(1): well developed expansion of the lateral side of the dentary ventral to the
coronoid process, with an angle up to 165º (sample mean angle of 154º)
Character 52: orientation of the longitudinal axis of the dentary occlusal plane
relative to the lateral side of the bone (as seen dorsally and caudal to the edentulous
region) (Fig. C.34)
(0): diagonal axis, directed rostrolaterally and forming approximately 15º with the
lateral side of the dentary
(1): axis parallel to the lateral side of the dentary
Character 53: lingual arching of the occlusal plane (Fig. C.35; Horner et al., 2004,
character 12)
(0): present, lingually convex occlusal plane
(1): absent, rostrocaudally straight occlusal plane
Character 54: caudal extension of the dental battery (Fig. C.36; modified from
Horner et al., 2004, character 10)
(0): the caudal end of the dental battery is found rostral to the caudal margin of
the coronoid process.
(1): the caudal end of the dental battery is found flush with the caudal margin of
(2): the caudal end of the dental battery is found caudal to the caudal margin of
317
Comments: early in ontogeny, neonates (e.g., the lambeosaurine
(“Hypacrosaurus” stebingeri?) MOR 548 and probably in Maiasaura peeblesorum
YPM-PU 22400) have tooth rows flush with the caudal margin of the coronoid process.
Character 55: separation between the dentary tooth row and the coronoid process
(Norman, 2002, character 26; Fig. C.37)
(0): the coronoid process is laterally offset (but nearly in contact) with the tooth
row, lacking a platform in-between the tooth row and the base of the process.
(1): the coronoid process is laterally offset relative to the tooth row, with the
presence of a concave platform or, in some cases, a laterodorsal concave slope separating
the base of the process from the dental battery.
Surangular
Character 56: morphology of the rostrodorsal process of the surangular (Fig. C.38;
(0): rostrocaudally thick process, slightly reduced in thickess rostrally, extensively
exposed in lateral view
(1): rostrocaudally reduced in thickness, strap-like and wedging dorsally into a
thin sliver that becomes concealed in lateral view by the dorsal half of the caudal margin
of the coronoid process
Character 57: surangular foramen (Fig-. C.39; Norman, 2002, character, character
27)
(0): present
(1): absent
Character 58: accessory foramen located rostrodorsal to the main surangular
foramen (Fig. C.40; Kobayashi and Azuma, 1999, character 15)
(0): present
(1): absent
Character 59: orientation of the convex side of the lateral lap and the lateroventral
surface of the main body of the surangular (Fig. C.41)
(0): facing more laterally than ventrally
318
(1): facing more ventrally than laterally
Character 60: lateral curvature of the caudal process of the surangular (Fig. C.42)
(0): absent, process nearly straight rostrocaudally
(1): present, process laterally recurved
Character 61: angle between the medial margin of the proximal region of the
surangular and the medial margin of the proximal region of the caudal process of
the element (Fig. C.43)
Angular
Character 62. Position of the angular in the mandible (Fig. C.44; Weishampel et al.,
1993, character 26)
(0): positioned ventrally and slightly medially, exposed in lateral view
(1): positioned medially, not exposed in lateral view
Coronoid
Character 63. Coronoid bone attached to the medial, dorsal and part of the lateral
sides of the distal end of the dentary coronoid process (Fig. C.45; Wagner, 2001)
(0): absent
(1): present
Prearticular
Character 64. Prearticular bone (Fig. C.46; Wagner, 2001)

(0): absent;
(1): present.
319
Table C.1. Character 24. Values of the ratio (R) between the mediolateral and
rostrocaudal length of the predentary, with the corresponding character state coding
derived from K-Means clustering analysis.
Specimen Taxon R K-means

Cluster (State)
SBDE 95E5/36 Bactrosaurus johnsoni 1.04 0
AMNH 5338 Corythosaurus casuarius 1.06 0
USNM 16600 cf. Corythosaurus sp. 1.27 1
USNM 11893 cf. Corythosaurus sp. 1.32 1
MOR 553S-7-27-2-89 Hypacrosaurus stebingeri 1.21 0
UCMP fldAk Edmontosaurus sp. 1.43 1
AMNH 6369 Gilmoreosaurus mogoliensis 1.08 0
MOR 553S-7-25-8-78 cf. Gryposaurus sp. 1.37 1
CMN 2278 Gryposaurus notabilis 1.12 0
ROM 44770 Maiasaura peeblesorum 1.91 2
AEHM 2-845 Olorotitan ararhensis 1.67 1
MOR 553S-7-12-8-5 Prosaurolophus maximus 1.06 0
320
−30
2.0
−40
BIC
−50
Length/Width Ratio of the Predentary
1.8
−60
−70
1.6
2 4 6 8
Number of Clusters
1.4
1.2
1.0
Ba Br Coi Cos Ed Eda Gi Grm Grn Hys Iga Lml Ma Ol Ou Pr Pt Ts cfCo cfGr cfLm
Taxon
Figure C.1. Character 24. Boxplot showing the distribution of the length/width ratio of the predentary in a sample of iguanodontoidean
ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which the data can be
classified. The photograph shows how the length and width of the predentary were measured, exemplified in Protohadros byrdi (SMU 74582).
321
Table C.2. Values of the ratio (R) between the depth of the predentary rostral face and the
length of the lateral process, with the corresponding character state coding derived from
K-Means clustering analysis.
Specimen Taxon R K-means Cluster

(State)
CMN 8503 Corythosaurus intermedius 0.36 1
YPM 2182 Edmontosaurus annectens 0.44 0
MOR 553S Hypacrosaurus stebingeri 0.36 1
AMNH 5461 Hypacrosaurus stebingeri 0.36 1
IRSNB 1731 Iguanodon bernissartensis 0.43 0
IRSNB 1551 Iguanodon atherfieldensis 0.39 0
MOR 553S-7-12-8-5 Prosaurolophus maximus 0.34 1
322
Depth/Length Ratio of the Predentary
0.45
0.40
0.35
A
0.30
Ba Br Coi Cos Eda Eq Grm Grn Hys Ig Iga Igb Lml Ma Ou Pr Pt Ts cfLm
Taxon
80
BIC
60
40
B
C
20
E V
2 4 6 8
Number of Clusters
Figure C.2. Character 25. A, boxplot showing the distribution of the depth/length ratio of the predentary in
a sample of iguanodontoidean taxa. B, Bayesian information criterion (BIC) for various numbersof clusters
in which the data can be classified. C, Predentary of Gryposaurus notabilis (CMN 2278) showing how the
depth and length of this bone were measured.
323
Table C.3. Character 26. Angle (A) between the predentary rostral face and the lateral process,
with the corresponding character state coding derived from K-Means clustering analysis.
Specimen Taxon A K-Means Cluster

(State)
AMNH 5240 Corythosaurus casuarius 35 3
MOR 553S Hypacrosaurus stebingeri 32 3
MOR 553S-7-27-2-89 Hypacrosaurus stebingeri 32 3
UCMP 128372 V76142 Edmontosaurus annectens 62 1
YPM 2182 Edmontosaurus annectens 67 1
LACM 23502 Edmontosaurus regalis 68 1
AMNH 6369 Gilmoreosaurus mongoliensis 78 0
MOR 553-7-3-8-87 Gryposaurus sp. 72 1
CMN 2278 Gryposaurus notabilis 53 2
IRSNB 1731 Iguanodon bernissartensis 81 0
IRSNB 1551 Iguanodon atherfieldensis 85 0
ROM 44770 Maiasaura peeblesorum 46 2
AEHM 2-845 Olorotitan ararhensis 30 3
MOR 553S-7-12-8-5 Prosaurolophus maximus 46 2
324
Table C.3—continued.

(State)
325
80
Orientation of the Rostral Surfce of the Predentary
70
−400
−410
−420
BIC
60
−430
−440
−450
50
2 4 6 8
Number of Clusters
40
30
Ba Br Coi Cos Eda Edr Eq Gi Gr Grm Grn Hy Hys Ig Iga Igb Lml Ma Ol Ou Pr Pt Ts cfLm
Taxon
Figure C.3. Character 26. Boxplot showing the distribution of the angle between the rostral surface of the predentary and the frontal
plane of the skull in a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC)
for various numbers of clusters in which the data can be classified. The photograph shows how this angle (A) was measured, exemplified in
Prosaurolophus maximus (MOR 447-7-27-7-5).
326
Figure C.4. Characters 27 and 28. Predentary denticles of several iguanodontoidean ornithopods. A, Ouranosaurus nigeriensis (cast of GDF 300).
B, Brachylophosaurus canadensis (MOR 1071-7-28-98-299). C, Corythosaurus intermedius (ROM 777). D, B. canadensis (CMN 8893).
E, Bactrosaurus johnsoni (AMNH 6372, juvenile). F, Edmontosaurus sp. (UCMP unrecorded no., juvenile). G, Equijubus normani (IVPP 12534).
H, Edmontosaurus sp. (UCMP Fld 4.5-83-90, juvenile). I, Protohadros byrdi (SMU 74582). J, Edmontosaurus sp. (UCMP FldAK 83-V-292, juvenile).
K, cf. Corythosaurus ap. (USNM 11893, juvenile). L, Hypacrosaurus altispinus (CMN 8674). M, H. altispinus (CMN 8501). N, Gryposaurus notabilis
(MSNM V345). O, Gryposaurus notabilis (MOR 553S-7-25-8-78). P, H. stebingeri (MOR 553S). Q, Prosaurolophus maximus (MOR 447-7-27-7-5).
R, H. stebingeri (MOR 553S-7-27-2-9). S, cf. Lambeosaurus sp. (USNM 10309). T, Iguanodon sp. (BMNH R105). U, Maiasaura peeblesorum
(YPM-PU 22405). V, Tsintaosaurus spinorhinus (IVPP V723). W, Olorotitan ararhensis (cast of AEHM 2/845). X, C. intermedius (CMN 8676).
Y, Lambeosaurus lambei (TMP 2005-00-28).
327
Figure C.5. Characters 29-32. Dorsal views of several iguanodontoidean predentaries.
A, Iguanodon sp. (BMNH R105). B, I. atherfieldensis (BMNH R11521, juvenile). C, Bactrosaurus
johnsoni (AMNH 6372, juvenile). D. Gilmoreosaurus mongoliensis (AMNH 6369, juvenile).
E. Protohadros byrdi (SMU 74582). F, cf. Lambeosaurus sp. (USNM 10309). G, Ouranosaurus
nigeriensis (cast of GDF 300). H, Maiasaura peeblesorum (YPM-PU 22405). I, Brachylophosaurus
canadensis (MOR 1071-7-28-98-299).
328
Figure C.6. Characters 29-32. Dorsal views of several hadrosaurid predentaries. A, B, and C,
Edmontosaurus sp. (UCMP unrcorded no, UCMP Fld 4.5-83-90, and UCMP FldAK 83-V-292,
respectively; juveniles). D, Gryposaurus monumentensis (RAM 6797). E and F, Prosaurolophus
maximus (MOR 447-7-27-7-5 and MOR 553S-7-12-8-5, respectively). G and H, Gryposaurus
cf. notabilis (MOR 553S-7-3-8-87 and 553S-7-25-8-78). I, Hadrosauridae indeterminate
(UCMP V82223).
329
Figure C.7. Characters 29-32. Lambeosaurine predentaries in dorsal view (except I, ventral view).
A and B, Hypacrosaurus stebingeri (MOR 553S). C, H. stebingeri (MOR 553S-7-27-2-89).
D and E, cf. Corythosaurus sp. (USNM 11893 and 16600, respectively; juveniles). F, Tsintaosaurus
spinorhinus (IVPP V725). G and H, H. altispinus (CMN 8674 and 8501, respectively). I, Olorotitan
ararhensis (cast of AEHM 2/845).
330
state 0
state 1
Figure C.8. Character 33. Lateral views of two hadrosaurid predentaries. A, Brachylophosaurus
canadensis (MOR 1071-7-28-98-299), with an arrow pointing to the lingual margin as the more
caudally extended feature in the caudal end of the lateral process. B, Corythosaurus intermedius
(CMN 8676), with an arrow pointing to the lateral margin as the most caudally extended feature in
the caudal end of the lateral process.
331
state 0
A
state 0
B
state 1
C
state 1
Figure C.9. Character 34. Dorsolingual views of iguanodontoidean predentaries; arrows indicate
the ridge of the median lingual process. A, Iguanodon sp. (AMNH R105). B, Gilmoreosaurus
mongoliensis (AMNH 6369, juvenile). C, Brachylophosaurus canadensis (MOR 1071-7-28-98-299).
D, Hyppacrosaurus stebingeri (MOR-553S-7-27-2-89).
332
state 0
state 1
Figure C.10. Character 35. Ventral views of two hadrosaurid predentaries; arrows show the rostral
end of the ventral median process. A, Brachylophosaurus canadensis (MOR 794). B, Prosaurolophus
maximus (MOR 553S-7-12-8-5).
333
state 0
A
state 1
extent of the indentation of the split of the ventral process. A, Prosaurolophus maximus
(MOR 553S-7-12-8-5). B, Brachylophosaurus canadensis (MOR 794).
334
Table C.4. Character 37. Ratio (R) between the length of the proximal slope of the
dentary and that of the dental battery for the specimens used to investigate the
distribution and pattern of variation of this character. Included are also the corresponding
character state coding derived from K-Means cluster analysis.

(State)
TMP 90-104-1 Brachylophosaurus canadensis 0.54 3
CUST 04 Charonosaurus jiayinensis 0.1 0
UCM 42764 Edmontosaurus regalis 0.31 1
CEUM 9758 Eolambia caroljonesa 0.19 0
PASAC-1 Gryposaurus sp. 0.36 2
335

(State)
IVPP V12691 Jinzhousaurus yangi 0.09 0
UALVP 11734 Lambeosaurinae 0.24 1
MOR 553Sb Prosaurolophus maximus 0.26 1
MOR 553Sa Prosaurolophus maximus 0.3 1
CMN 8796 Saurolophus osborni 0.36 2
336
Ratio between the Proximal Edentulous Slope and Dental Battery Length
0.6
70
60
BIC
50
0.5
40
30
20
2 4 6 8
Number of Clusters
0.4
0.3
0.2
0.1
Am Ba Br Ch Coi Cos Ed Eda Edr Eo Eq Grl Grm Grn Hy Hys Ig Iga Igb Jz Lml Ma Ou Pa Pr Pt Saa Sao Te Ts x
Taxon
Figure C.12. Character 37. Boxplot showing the distribution of the ratio between the proximal slope of the dentary edentulous portion
and the length of the dental battery in a sample of iguanodontoidean taxa. The inserted smaller graph shows the Bayesian information criterion
(BIC) for various numbers of clusters in which the data can be classified. The photograph shows how the two variables of this raio were measured,
exemplified in Corythosaurus casuarius (CMN 8676).
337
Table C.5. Character 38. Aangle of deflection (A) of the proximal slope in the dentaries
used to investigate the distribution and pattern of variation in this character. Included are
also the corresponding character coding derived from K-Means clustering analysis.
Specimen Taxon A K-means Cluster

(State)
TMP 90-104-1 Brachylophosaurus canadensis 155 1
PASAC-1 Hadrosaurinae 147 0
TMP 80-22-1 Gryposaurus notabilis 143 0
IVPP V12691 Jinzhousaurus yangi 155 1
338
Specimen Taxon A K-means Cluster

(State)
NMMNH P-25100 Pararhabdodon tubicen 156 1
SMU 74582 Prosaurolophus maximus 145 0
CMN 8796 Saurolophus osborni 156 1
339
165
Angle of Ventral Deflection of the Dentary Proximal Edentulous Slope
160
155
150
145
−370
−380
BIC
140
−390
−400
135
2 4 6 8
Number of Clusters
Am Ba Br Ch Coi Cos Eda Eo Eq Grl Grm Grn Hy Hys Iga Igb Jz Lml Ma Ou Pat Pr Pt Saa Sao Te Ts x
Taxon
Figure C.13. Character 38. Boxplot showing the distribution of the angle between the proximal slope of the dentary edentulous portion
and the frontal plane of the skull in a sample of iguanodontoidean taxa. The inserted smaller graph shows the Bayesian information criterion
(BIC) for various numbers of clusters in which the data can be classified. The photograph shows how the two variables of this raio were measured,
exemplified in Lambeosaurus lambei (CMN 8633).
340
Table C.6. Character 39. Angle of deflection (A) of the proximal slope of the predentary
joint surface for the specimens used to investigate the distribution and pattern of variation
of this character. Included is also the corresponding character coding derived from K-
Specimen Taxon A K-means

cluster (state)
UALVP 11734 Lambeosaurinae 119 1
341

cluster (state)
AEHM 2/845 Olorotitan ararhensis 124 1
ROM 768 Parasaurolophus walkeri 105 2
342
Angle of Ventral Deflection of the Predentary Margin of the Dentary
−440
140
−460
BIC
−480
−500
130
−520
2 4 6 8
Number of Clusters
120
110
Am Ba Br Ch Co Hys Ed Eda Edr Eo Eq Gi x Grl Grm Grn Hy Ig Iga Jn Lml Ma Ol Ou Pac Pat Paw Ph Pr Pt Saa Sao Te Ts
Taxon
Figure C.14. Character 39. Boxplot showing the distribution of the angle formed by the slope of the rostral region of the dentary that
articulates with the predentaryand the frontal plane of the skull. The inserted smaller graph shows the Bayesian information criterion (BIC)
for various numbers of clusters in which the data can be classified. The photograph shows how the two variables of this raio were measured,
exemplified in Brachylophosaurus canadensis (MOR 1071).
343
112
Brachylophosaurus canadensis Prosaurolophus maximus
142
Angle of Slope for Predentary

Sq.R = 0.07 Sq.R = 0.64
110
140
138
108
136
106
134
104
132
102
130
20 22 24 26 28 30 32 A 20 25 30 35 40 45 B
126
Corythosaurus spp.
124
Sq.R = 0.26
122
120
118
116
114
20 22 24 26 28 30 C
margin of the dentary and the frontal skull plane relative to the dental battery length for three
species of hadrosaurid ornithopods.
344
138
Edmontosaurus spp.

Sq.R = 0.17
136
134
132
130
128
126
10 20 30 40 50
A
137
136
Sq.R = 0.70
135
134
133
132
131
130
30 35 40 45 50
B
margin of the dentary and the frontal skull plane relative to the dental battery length for the
hadrosaurid Edmontosaurus.
345
Table C.7. Character 40. Angle of deflection (A) of the ventral margin of the rostral
region of the dentary for the specimens used to investigate the distribution and pattern of
variation of this character. Included is the corresponding character coding derived from

(State)
AEHM 1/12 Amurosaurus riabini 43 2
CUST 04 Charonosaurus jiayinensis 21 1
BMNH R9527 Corythosaurus casuarius 32 2
346
Specimen Taxon D K-means Cluster

(state)
347
−490
Angle of Ventral Deflection of the Dentary Rostroventral Margin
40
−510
BIC
−530
−550
E V
30
2 4 6 8
Number of Clusters
20
10
Am Ba Br Ch Coi Cos Ed Eda Edr Eo Eq Grl Grm Grn Hys Ig Iga Igb Jz Lml Ma Ou Pat Paw Ph Pr Pt Saa Sao Te Ts x
Taxon
Figure C.17. Character 40. Boxplot showing the distribution of the angle of ventral deflection of the rostral edentulous region of
the dentary for a sample of iguanodontoidean taxa. The inserted smaller graph shows the Bayesian information criterion (BIC) for
various numbers of clusters in which the data can be classified. The photograph shows how this angle was measured, exemplified in
a dentary of Amurosaurus riabinini (AEHM 1/12).
348
24
28
23
Sq.R = 0.76
Angle of Deflection
Angle of Deflection
26
22
24
21
22
20
20
Sq.R = 0.36
19
18
18
16
17
20 22 24 26 28 30 32 A 5 10 15 20 25 30 B
14
Edmontosaurus spp.
Angle of Deflection
Angle of Deflection
25
Sq.R = 0.54
13
Sq.R = 0.14
20
12
15
11
10
10
10 20 30 40 50 C 30 35 40 45 50 D
margin of the dentary relative to the dental battery length for four species of hadrosaurid ornithopods.
349
20
Corythosaurus spp.
35
Sq.R = 0.01
Angle of Deflection
18
Angle of Deflection
30
16
25
14
Sq.R = 0.00
20
12
15
10
20 25 30 35 40 45 A 20 22 24 26 28 30 32 B
25
28
Lambeosaurus lambei
24
Sq.R = 0.43
26
Angle of Deflection
Angle of Deflection
23
24
22
22
21
Sq.R = 0.43
20
20
19
18
18
10 15 20 25 30 35 C 20 25 30 D
margin of the dentary relative to the dental battery length for four species of hadrosaurid ornithopods.
350
Table C.8. Character 41. Relative position of the inflexion point where the ventral margin
of the dentary originates (ratio R; Fig. C.20). The values are given for the specimens used
to investigate the distribution and pattern of variation of this character. Included is also
the corresponding character coding derived from K-Means clustering analysis.
Specimen Taxon R K-Means

Cluster (State)
CUST 04 Charonosaurus jiayinensis 0.71 1
CUST Charonosaurus jiayinensis 0.76 1
BMNH R9527 Corythosaurus casuarius 0.73 1
351

Cluster (State)
ROM 768 Parasaurolophus walkeri 0.59 2
MOR 553Sa Prosaurolophus maximus 0.8 0
MOR 553Sb Prosaurolophus maximus 0.83 0
352
1.1
80
60
Ratio of the Starting Posiion of Dentary Ventral Deflection
BIC
1.0
40
20
0.9
2 4 6 8
Number of Clusters
0.8
0.7
0.6
0.5
0.4
Am Ba Br Ch Coi Cos Eda Edr Eo Grl Grm Grn Hy Hys Ig Iga Igb Jz Lml Ma Ou Pat Paw Ph Pr Pt Saa Sao Te Ts cfLm x
Taxon
Figure C.20. Character 41. Boxplot showing the distribution of ratio of the starting position of the ventral deflection of the dentary
rostroventral margin. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which
the data can be classified. The photograph shows how the two variables of this raio were measured, exemplified in Lambeosaurus lambei
(CMN 351).
353
0.80
Brachylophosaurus canadensis Maiasaura peeblesorum

Sq.R = 0.00
Deflection Origin Ratio

Sq.R = 0.96
0.80
0.75
0.75
0.70
0.70
0.65
20 22 24 26 28 30 32 A 5 10 15 20 25 30 B
0.86
0.85

0.84
0.80
0.82
0.75
Edmontosaurus spp. Edmontosaurus annectens

0.80
Sq.R = 0.04 Sq.R = 0.02

0.70
0.78
0.65
10 20 30 40 50
C 30 35 40 45 50
D
Figure C.21. Character 41. Scatterplots of the ratio of the stating point of the ventral reflection
relative to the dental battery length for four species of hadrosaurid ornithopods.
354
Prosaurolophus maximus Corythosaurus spp.
0.90
0.85
Sq.R = 0.08 Sq.R = 0.10

0.88
0.80
0.86
0.75
0.84
0.70
0.82
0.65
0.80
0.60
20 25 30 35 40 45 A 20 22 24 26 28 30 32 B
0.78
0.65
0.76
Sq.R = 0.64
0.60
0.74
Lambeosaurus lambei
0.55
Sq.R = 0.07
0.72
0.50
0.70
0.45
0.68
5 10 15 20 25 30 35 C 20 25 30 D
Figure C.22. Character 41. Scatterplots of the ratio of the stating point of the ventral reflection
relative to the dental battery length for four species of hadrosaurid ornithopods.
355
Table C.9. Character 42. Values for the ratio (R) of the lingual projection of the dentaries
used to investigate the distribution and pattern of variation of this character. Included is
also the corresponding character coding derived from K-Means clustering analysis.

Cluster (State)
FMNH 862 Brachylophosaurus candensis 1.75 1
MOR 1071-8-15-98-574 Brachylophosaurus candensis 1.79 1
MOR 1071-7-13-99-93 Brachylophosaurus candensis 1.81 1
CMN 8893 Brachylophosaurus candensis 1.95 1
MOR 1071 Brachylophosaurus candensis 1.98 1

Cluster (State)
ROM 787 Prosaurolophus maximus 1.30 0
SMU 74582 Prosaurolophus maximus 1.92 1
3.0
Lingual Projection of the Sympphyseal End of the Dentary
−30
−40
−50
BIC
−60
−70
2.5
−80
E V
−90
2 4 6 8
Number of Clusters
2.0
1.5
Am Ba Br Ch Coi Cos Ed Eda Edr Eo Grm Grn Hys Ig Lml Ma Ou Pat Paw Ph Pr Saa Te Ts x
Taxon
Figure C.23. Character 42. Boxplot showing the distribution of the degree of lingual projection of the symphyseal end of the dentary
for a sample of iguanodontoidean taxa. The inserted smaller graph shows the Bayesian information criterion (BIC) for various
numbers of clusters in which the data can be classified. The photograph shows how the two variables for this character were measured,
exemplified in an indeterminate lambeosaurine dentary (? Lambeosaurus, UALVP 11734).
Lingual Projection of the Dentary Symphysis
2.1
Sq.R = 0.02
2.0
1.9
1.8
1.7
A
20 22 24 26 28 30 32
Lingual Projection of the Dentary Symphysis
2.2
Corythosaurus spp.
2.1
Sq.R = 0.61
2.0
1.9
1.8
1.7
22 24 26 28 30 32
B
Figure C.24. Character 42. Scatterplots of the degree of lingual projection of the symphyseal
end of the dentary relative to the dental battery length for two hadrosaurid taxa.
359
Table C.10. Character 43. Values of the angle of orientation of the symphysis (S) of the
Included are also the corresponding character coding derived from K-Means clustering.

(State)
ROM 787 Prosaurolophus maximus 13 1
360
30
−290 −280 −270 −260 −250 −240

Orientaton of the Dentary Symphysis (degres)
25
BIC
20
2 4 6 8
Number of Clusters
15
10
5
Ba Br Coi Cos Eda Edr Eo Eq Grm Grn Hys Ig Lml Ma Ou Pat Ph Pr Saa Te x
Taxon
Figure C.25. Character 43. Boxplot showing the distribution of the orientation of the dentary symphysis relative to the lateral wall
of the element for a sample of iguanodontoidean taxa. The inserted smaller graph shows the Bayesian information criterion (BIC) for
various numbers of clusters in which the data can be classified. The photograph shows how this character was measured, exemplified in
a dentary of Brachylophosaurus canadensis (MOR 1071-7-15-98-226).
361
Orientation of the Dentary Symphysis (º)
15
14
Corythosaurus spp.
14
12
13
Sq.R = 0.57
10
12
Sq.R = 0.20
8
11
10
6
20 22 24 26 28 30 32 22 24 26 28 30
Dental Battery Length (cm) A Dental Battery Length (cm) B

13
Lambeosaurus lambei
12
11
Sq.R = 0.99
10
9
20 25 30 C
Figure C.26. Character 43. Scatterplots of the orientation of the dentary symphysis
relative to the dental battery length for three hadrosaurid taxa.
362
A B C
Figure C.27. Character 44. Rostral views of dentaries showing the degree of lingual curvature of the symphyseal region; that area
curves approximately 90º in A and B, and gently forming a wide arch in C. A, Iguanodon sp. (BMNH 28660). B, Parasaurolophus
tubicen (NMMNH P-25100). C, Maiasaura peeblesorum (YPM-PU 22405).
363
state 0
te 0
A
A
state 0
B
state 1
C
state 1
Figure C.28. Character 35. Medial views of dentaries showing the morphology of the dorsal edge
of the rostral end of the dentary (i.e., absence or presence of a concave profile). A, Edmontosaurus
annectens (USNM 4807). B, Prosaurolophus maximus (MOR 447). C, Brachylophosaurus canadensis
(FMNH 862). D, Lambeosaurus lambei (CMN 8633).
364
state 0
A
state 1
B
state 2
C
state 3
Figure C.29. Character 46. Lateral views of hadrosaur dentaries, showing the development and
location of a ventral bulge (white rectangles) in this bone. A, Equijubus normani (IVPP V12534).
B, Jinzhousaurus yangi (IVPP V12691). C, Gryposaurus latidens (AMNH 5465). D, Prosaurolophus
maximusi (MOR 447-WSQ86).
365
Table C.11. Character 47. Angle of rostral inclination (A) of the coronoid process of the
Included is also the corresponding character coding derived from K-Means clustering.
Specimen Number Taxa A K-Means Cluster

(State)
CUST Charonosaurus jiayinensis 73 1
AMNH 5879 Edmontosaurus anenctens 65 2
366
Specimen Number Taxa A K-Means Cluster

(State)
367
Inclination Angle of the Coronoid Process (degrees)
100
−440
−460
90
BIC
−480
−500
E V
80
2 4 6 8
Number of Clusters
70
Am Ba Br Ch Coi Cos Ed Eda Eo Eq Grl Grm Grn Hys Ig Iga Igb Jz Lml Ma Ou Pat Pr Pt Saa Sao Te Ts x
Taxon
Figure C.30. Character 47. Boxplot showing the distribution of the angle of inclination of the dentary coronoid process for a sample
of iguanodontoidean taxa. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in
which the data can be classified. The photograph shows how the two variables for this character were measured, exemplified in a dentary
of Prosaurolophus maximus (CMN 8894).
368
state 0 state 1
A B
state 1 state 1
C D
Figure C.31. Characters 48 and 49. Medial views of the coronoid process of four iguanodontoidean
taxa. A, Iguanodon sp. (BMNH R754).]B, Brachylophosaurus canadensis (CMN 8893); the white
rectangle indicates the apomorphic caudodorsally pointed margin. B, Prosaurolophus maximus
(MOR 553S-7-12-8-5). C, Edmontosaurus annectens (BMNH R3656). D, Lambeosaurine from
Dinosaur Provincial Park (?Lambeosaurus, UALVP 11734).
369
A B C
Figure C.32. Character 50. Ridges and striations on the medial side of the coronoid process of the dentary, indicated by the
arrows and the white rectangle. A, Bactrosaurus johnsoni (BMNH 6581, juvenile specimen). B, Bactrosaurus johnsoni (PIN2540-1).
C, Parasaurolophus tubicen (NMMNH P-25100).
370
Table C.12. Character 51. Angle (A) between the lateral expansion of the base of the
coronoid process and the lateral side of the dentary for the specimens used to investigate
the distribution and pattern of variation of this character. Included is also the
corresponding character coding derived from K-Means clustering analysis.

(State)
AEHM 1-12 Amurosaurus riabinini 154 1
CUST 04 Charonosaurus jiayinensis 151 1
CUST Charonosaurus jiayinensis 152 1
371

(State)
372
175
Lateral Expansion of the Caudolateral Region of the Dentary
170
−310
−330
BIC
165
−350
−370
E V
160
2 4 6 8
Number of Clusters
155
150
Am Ba Br Ch Cos Ed Eda Edr Eo Eq Grm Grn Hy Hys Ig Igb Lml Ma Ol Ou Pat Paw Ph Pr Pt Saa Te Ts x
Taxon
Figure C.33. Character 51. Boxplot showing the distribution of the angle between the lateral surface and the caudoventral region of the
lateral side of the dentary for a sample of iguanodontoidean taxa. The inserted smaller graph shows the Bayesian information criterion (BIC)
for various numbers of clusters in which the data can be classified. The photograph shows how this character was measured, exemplified in
a dentary of Maiasaura peeblesorum (YPM-PU 22405).
373
state 0
A
state 1
Figure C.34. Character 52. Dorsal views of two hadrosaur dentaries showing the orientation
of the long axis of the occlusal plane relative to the lateral surface of the element. A, Protohadros
byrdi (SMU 74582). B, Prosaurolophus maximus (MOR 447).
374
state 0
A
state 1
Figure C.35. Character 53. Dorsal views of two iguanodontoidean dentaries showing the presence
and absence of lingual arching of the occlusal plane. A, Ouranosaurus nigeriensis (GDF 300).
B, Telmatosaurus transsylvanicus (BMNH R3386).
375
state 0
A
state 1
B
state 2
Figure C.36. Character 54. Medial views of three iguanodontoidean dentaries, with white
arrows showing the position of the caudal end of the dental battery relative to that of the
coronoid process. A, Iguanoodon atherfiendensis (BMNH R11521). B, Bactrosaurus johnsoni
(SBDE 95E5-12). C, Prosaurolophus maximus (MOR447-WSQ86).
376
state 0
A B
state 1
C D
Figure C.37. Character 55. Degree of separation (arrows) between the dentary tooth row and the coronoid process. A, Iguanodon
atherfieldensis (BMNH R11521), caudal view. B, I. atherfieldensis (BMNH R11521), dorsal view. C, Amurosaurus riabinini (AEHM 1/12),
caudal view. D, A. riabinini (AEHM 1/12), dorsal view.
377
A B C
state 1
Figure C.38. Character 56. Participation of the surangular in the coronoid process of the dentary
and shape of the rostrodorsal process (white rectangles) of the former. All images display lateral views.
A, Iguanodon atherfieldensis (BMNH R11521). B, cf. Bactrosaurus sp. (MAL 0329-002).
C, Edmontosaurus annectens (BMNH R4862). D, Prosaurolophus maximus (MOR 447-3-8-5-86).
378
state 0 state 0
A B
state 1 state 1
C D
Figure C.39. Character 57. Presence and absence of surangular foramen (arrows). A, Iguanodon
bernissartensis (IRSNB 1731), lateral view. B, Jinzhousaurus yangi (IVPP V12691), lateral view.
C, Brachylophosaurus canadensis (MOR 794), lateral view. D, Prosaurolophus maximus (TMP 84.1.1),
lateroventral view.
379
state 0
A
state 1
Figure C.40. Character 58. Presence and absence of surangular accessory foramen (black
arrow). A, Ouranosaurus nigeriensis (cast of GDF 300), lateral view. B, Parasaurolohus tubicen
(NMMNH P-25100), in lateral, and slightly ventral, view.
380
state 0
state 1
Figure C.41. Character 59. Orientation of the lateral quadrate lap (white rectangles)
and convex surface of the surangular. A, Equijubus normani (IVPP V12534), lateral view.
B, Brachylophosaurus canadensis (CMN 8893), lateral view.
381
state 0 state 1
A B
state 0
state 1
C
state 0
D
state 1
E F
Figure C.42. Character 60. Lateral curvature of the caudal process of the surangular (white rectangles); all images in dorsal view.
A, Brachylophosaurus canadensis (MOR 1071). B, Amurosaurus riabinini (AEHM 1/12). C, EdmontosaurusB (MOR 601). D, Lambeosaurus
lambei (CMN 2869). E, Gryposaurus monumentensis (RAM 6797). F, Parasaurolophus tubicen (NMMNH P-25100).
382
Table C.13. Character 61. Angle (A) between the proximal medial side of the caudal
process and the proximal medial side of the surangular in the specimens used to
investigate the distribution and patterns of variation of this character. Included is also the
corresponding character coding derievd from K-Means clustering.
Specimen Taxom A K-Means Cluster

(State)
SBDE 95/95ES Bactrosaurus johnsoni 160 0
SBDE 95/95ES Bactrosaurus johnsoni 159 0
MOR 1071-8-8-99-484-B Brachylophosaurus canadensis 157 0
TMP 90.104.1 Brachylophosaurus canadensis 150 0
UALVP unrecorded no. cf. Hypacrosaurus altispinus 161 0
TMP.88.151.86 cf. Hypacrosaurus stebingeri 160 0
CMN 8744 Edmontosaurus sp. 156 0
383
160
Angle Along Medial Margin of Surangular
155
−180
150
−200
BIC
145
−220
−240
140
2 4 6 8
Number of Clusters
135
Am Ba Br Cos Ed Eda Edr Grm Iga Lml Ma Pat Pr Pt cfHy cfHys
Taxon
Figure C.43. Character 61. Boxplot showing the distribution of the angle between the proximal medial margin of the surangular and
the proximal medial margin of the caudal process for a sample of iguanodontoidean taxa. The inserted smaller graph shows the Bayesian
information criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph shows how this angle was
measured, exemplified in a dentary of Lambeosaurus lambei (CMN 2869).
384
state 0
A
state 1
Figure C.44. Character 62. Lateral exposure of the angular (white arrows). A, Equijubus normani
(IVPP V12534), lateral view. B, Brachylophosaurus canadensis (TMP 104.90.1), lateral view.
385
state 0
A
state 1
Figure C.45. Character 63. Presence and absence of coronoid bone. A, Corythosaurus intermedius
(ROM 777), lateral view. B, Iguanodon bernissartensis (cast of IRSNB 1534), lateral view.
386
state 0
A
state 1
Figure C.46. Character 64. Presence and absence of prearticular bone. A, Brachylophosaurus
canadensis (CMN 8893), medial view. B, Iguanodon atherfieldensis (BMNH R11521), medial view.
Abbreviations: an = angular; dt = dentary; prar = prearticular; sa = surangular; spl = splenial.
387
APPENDIX D
FACIAL CHARACTERS
Premaxilla
Character 65: mediolateral expansion of the premaxillary oral margin (measured as

the ratio between the maximum mediolateral width of the premaxilla and the
minimum width at the narrowest point or post-oral constriction) (Table D.1 and Fig.
D.1; modified from Horner et al., 2004, character 22)
(0): relatively narrow, ratio less than 1.65 (mean ratio of 1.45)
(1): ratio between 1.65 and 2 (mean ratio of 1.84)
(2): very wide, with a ratio greater than 2 (mean ratio of 2.22).
Comments: the premaxilla becomes more expanded with age.
Character 66: Position of the premaxillary oral margin relative to the occlusal plane
of the dentition (Fig. D2; modified from Norman, 2002)
(0): premaxillary margin slightly ventrally offset from occlusal plane
(approximately, the dorsoventral distance between the occlusal plane and the level of the
premaxillary oral margin is less than the mean depth of the dentary);
(1): very strongly deflected ventrally (approximately, the dorsoventral distance
between the occlusal plane and the level of the premaxillary oral margin is equal to or
larger than the mean depth of the dentary).
Character 67: morphology of the oral margin of the premaxilla (Fig. D3-D6)
(0): moderately expanded border, dorsoventrally thicker towards the parasagittal
plane of the snout, and slightly deflected ventrally (Fig. D3)
(1): moderately expanded border, becoming thinner towards the parasagittal plane
of the snout, so that the narial fossa merges with the rostromedial region of the premaxilla
adjacent to the joint between premaxillae (Fig. D4)
(2): folded caudodorsally into a thin recurved margin (Fig. D5)
(3) ventrally deflected and dorsoventrally expanded, forming a very thick “lip-
like” margin (Fig. D6)
388
Character 68: morphology of the rostrolateral corner of the oral margin of the
premaxilla (Fig. D.7)
(0): rounded or slightly angulose
(1): triangular, dorsoventrally expanded, ventrally deflected and pointed
Character 69: premaxillary oral margin with a “double layer” morphology
consisting of an external denticle-bearing layer and an internal layer of thickened
bone set back slightly from the oral margin and separated from the denticular layer
by a deep sulcus bearing vascular foramina (Fig. D.8; Horner et al., 2004, character
25):
(0): absent
(1): present
Character 70: premaxillary foramen located rostrally and ventrolaterally to the
rostral margin of the external naris (Fig. D.9; Horner et al., 2004, character 23)
(0): absent
(1): present
Character 71: premaxillary accessory foramen entering rostrally through the outer
(rostral) narial fossa, located rostral to the premaxillary foramen (Fig. D.10; Horner
et al., 2004, character 24)
(0): absent
(1): present, empties into a common chamber with the premaxillary foramen
Character 72: premaxillary accessory narial fossa located rostral to the
circumnarial depression (Fig. D11; Horner et al., 2004, character 26)
(0): absent
(1): present, separated from circumnarial depression by a rostrocaudally wide
ridge
Character 73: premaxillary additional accessory fossa located lateral to the rostral
accessory fossa and rostrolateral to the circumnarial depression, parallel to the
lateral border of the oral margin (Fig. D.12)
(0): absent
(1): present
389
Character 74: elongation of premaxillary caudodorsal process (Fig. D.13; modified
from Horner et al., 2004, characters 27)
(0): the premaxillary caudodorsal process does not meet the caudoventral process
caudally
(1): elongate caudodorsal process that extends caudally to meet the caudoventral
process, forming the caudal margin of the external naris
Character 75: vertical groove on the caudoventral process of the premaxilla, located
rostral to the dorsal process of the maxilla and extending ventrally from a small
opening between the two premaxillary caudal processes; the groove is bounded
rostrally by a triangular ventral projection of the caudolateral process of the
premaxilla (Fig. D.14; Evans and Reisz, 2007, character 5)
(0): absent
(1): present
Character 76: elongation of the caudoventral process of the premaxilla (in adults)
(Fig. D.15; modified from Suzuki et al., 2004, character 4)
(0): relatively short, the caudoventral process extends caudodorsally to end dorsal
to the lacrimal or medidodorsal to the rostral end of the prefrontal.
(1): long, the caudoventral process extends to end medial to the dorsal region of
the prefrontal.
(2): very long, the caudoventral process extends caudodorsal to the prefrontal.
Character 77: morphology of the caudal region of the caudoventral process of the
adult premaxilla (Fig. D.16-D.18)
(0): mediolaterally compressed and triangular (Fig. D.16)
(1): dorsoventrally broad and directed caudally or caudally and slightly dorsally
(Fig. D.17)
(2): triangular and dorsoventrally expanded, laterally convex lobe, directed
rostrodorsally (Fig. D.18)
Character 78: premaxillary caudodorsal process has an accessory rostroventral
flange that overlaps the lateral surface of the nasal in the rostral region of a
supracranial crest (Fig. D.19; Evans and Reisz, 2007, character 18)
(0): absent
390
(1): present
Character 79: laterodorsal profile of the caudodorsal and caudoventral margins of
the external bony naris (Fig. D.20; Morris, 1978; modified from Evans and Resiz,
2007, character 4)
(0): subrectangular to subellipsoidal
(1): triangular and elongated, caudal constriction gradually closing caudodorsally
(2): lacriform and shorter than in state 1, caudal constriction occurs abruptly and it
is primarily composed of either a lateroventral expansion of the caudodorsal process or a
dorsal expansion of the caudoventral process of the premaxilla.
Character 80: relative thickness of the caudal region of the caudodorsal and
caudoventral processes (Fig. D.21)
(0): substantially much wider caudoventral process
(1): both processes are approximately equal in width
(2): substantially much wider caudodorsal process due to the presence of a
lateroventral expansion that constricts the bony naris caudall
Character 81: length/width ratio of the external naris (Table D.2; Fig. D.22)
(0): ratio less than 1.85 (mean ratio of 1.75)
(1): ratio between 1.85 and 2.85 (mean ratio of 2.43)
(2): ratio greater than 2.85 (mean ratio of 3.08)
Character 82: dorsolateral flange at approximately mid-length of the mediolaterally
compressed caudoventral process of the premaxilla (Fig. D.23; Gates and Sampson,
2007)
(0): absent
(1): present
Nasal
Character 83: location of the nasal bone and nasal cavity in the adult skull (Fig.
D.24; modified from Horner et al., 2004, character 33 and partially from Evans and
Reisz, 2007, character 7)
391
(0): the nasal extends from the rostral region of the skull roof to the rostrodorsal
region of the snout with the nasal cavity rostromedial to the orbit
(1): nasal retracted caudal to the rostrum and occupying a supracranial position in
the skull, with the ventral region of the nasal meeting the prefrontal rostral to the orbit,
resulting in a crest that extends supraorbitally
(2): retracted caudal to the rostrum and occupying a supracranial position in the
skull, with the ventral region of the nasal meeting the prefrontal caudal to the rostral
margin of the orbit, resulting in a convoluted narial passage and hollow crest that extend
supraorbitally
Character 84: curvature of the caudodorsal region of the nasal (Fig. D.25; modified
from Wagner, 2001 and partially from Evans and Reisz, 2007, character 7)
(0): absent, nasal straight caudodorsally
(1): present, nasal rotated and folded caudodorsally
Character 85: morphology of the rostral end of the nasal at the contact with the
dorsal process of the premaxilla (Figs. D.26 and D.27; states 3 and 4 modified from
Evans and Reisz, 2007, character 17)
(0): long and wedge-shaped rostral process, gradually decreasing in width
rostrally to a sharp point
(1): hook-like process, it becomes abruptly deep near its rostral end and then
wedges rostrally to a rostroventrally directed
(2): long and subrectangular process, with slightly rounded corners;
(3): small rostral process of the nasal fits along the ventral edge of the premaxilla,
the latter briefly overlapping the nasal
(4): the nasal bifurcates to meet the premaxilla in a W-shaped interfingering
suture, a long and finger-like process of the nasal has an extensive overlapping joint with
the caudodorsal process of the premaxilla; and additional, more caudally located shorter
process of the nasal abuts the premaxilla
Character 86: morphology of the nasal contact with the caudodorsal region of the
caudoventral premaxillary process at the caudal margin of the narial foramen (Fig.
D.28)
392
(0): the nasal forms a subrectangular flange exposed dorsal to the premaxillary
caudoventral process.
(1): the nasal forms a large hook-like rostroventral process, exposed dorsal to the
premaxillary caudoventral process.
(2): the nasal forms a greatly shortened and dorsoventrally narrow hook-like
rostroventral process, exposed dorsal to the premaxillary caudoventral process.
Character 87: location of the rostral end of the dorsal process of the nasal relative to
the rostral margin of the narial foramen (Fig. D.29)
(0): the rostral end of the rostrodorsal process of the nasal does not reach the
rostral margin of the narial foramen
(1): the rostral end of the rostrodorsal process of the nasal reaches the rostral
margin of the narial foramen
Comments: this character is linked to Gates and Sampson (2007) character 35
about the composition of the rostrodorsal corner of the narial foramen; is all formed by
pmx if process does not reach the rostral margin and formed by pmx and ns if it does.
Character 88: caudoventral region of the nasal ventrally recurved and hook-shaped,
with a rostral process that inserts under the caudoventral process of the premaxilla
(Fig. D.30)
(0): absent
(1): present
Character 89: caudal end of the nasals forming a pair of finger-like process on top
of the frontals and centered around the sagittal plane of the skull roof (Fig. D.31;
Gates and Sampson, 2007, character 65, in part)
(0): absent
(1): present
Character 90: caudal end of the nasals forming a pair of small and short processes
that insert between the frontals at the sagittal plane of the skull roof (Fig. D.32;
Gates and Sampson, 2007, character 65, in part)
(0): absent
(1): present
393
Character 91: position of the summit of the nasal arch crest relative to the
caudodorsal margin of the narial foramen (Fig. D.33)
(0): summit located dorsal to the caudal margin of the narial foramen
(1): summit located caudodorsal to the caudal margin of the narial foramen
Maxilla
Character 92: rostrodorsal process that is medially offset from the body of the
maxilla and extends also medial to the caudovenventral process of premaxilla to
form part of medial floor of external naris (Fig. D.34; partially after Wagner, 2001
and Horner et al., 2004, character 42):
(0): present
(1): absent, the rostral end of the maxilla forms a ventrally sloping rostrodorsal
shelf that underlies the premaxilla
Comments: a vestigial rostrodorsal process appears to be present in Tsintaosaurus
spinorhinus (IVPP V725).
Character 93: lateral exposure of the medial rostrodorsal process (Fig. D.35; Gates
and Sampson, 2007, character 45):
(0): not exposed through the narial foramen in lateral view
(1): exposed through the narial foramen in lateral view
Comments: the only Gryposaurus notabilis skull where this character was
observable was ROM 873. In this specimen, the laterodorsal flange of the caudoventral
process of the premaxilla was incomplete at the level of the rostrodorsal process of the
maxilla. If complete, the flange would have concealed the rostrodorsal process of the
maxilla. Based on this, the character was scored as absent in G. notabilis.
Character 94: pendant rostral end of the rostroventral process of the maxilla (Fig.
D.36)
(0): absent
(1): present
Character 95: angle between the dorsal margin of the rostroventral process or shelf
of the maxilla and the rostral segment of the tooth row (Table D.3; Fig. D.37)
394
(0): rostrodorsal region of the mxilla subconical in shape, dorsoventrally narrow,
forming an angle of 25º or less with the rostral tooth row (mean angle of 20º)
(1): dorsoventrally thicker, forming an angle greater then 25º and up to 39º with
the rostral tooth row (mean angle of 31º)
(2): rostroventral process dipping steeply ventrally, forming an angle of 40º or
greater with the tooth row (mean angle of 43º), rostral region of the maxilla appears
dorsally “swollen” and craniocaudally compressed.
Character 96: geometry of the lateral surface of the rostrodorsal region of the
maxilla (Fig. D.38)
(0): arcuate to subrectangular
(1): triangular and rostrocaudally compressed
(2): subsquared, taller than rostrocaudally wide, with slightly convergent rostral
and caudal margins
Character 97: elevation of the lateral surface of the rostrodorsal region of the
maxilla (modified from Weishampel et al., 1993, character 20)
(0): relatively low
(1): relatively high
Comments: this character is linked to the elevation of the dorsal maxillary process
and the migration of the antorbital fenestra dorsally near the premaxillary articulation
surface.
Character 98: position of the dorsal process and the dorsal margin of the
dorsolateral promontory of the maxilla (expressed as the ratio between the distance
from its summit to the rostral end of the maxilla and the craniocaudal length of the
element) (Table D.4; Fig. D.40)
(0): caudally located dorsolateral promontory (with a ratio greater than 0.57; mean
of 0.64), base of dorsal process positioned within the caudal third of the maxilla
(1): centrally located dorsolateral promontory (with a ratio between 0.47 and 0.57;
mean of 0.51), base of dorsal process positioned slightly caudal to the mid-length of the
maxilla
395
(2): dorsolateral promontory located slightly rostral to the mid-length of the
maxilla (with a ratio between 0.35 and 0.46; mean of 0.42), base of dorsal process
centered around the mid-length of the bone
(3): the base of dorsal process and dorsolateral promontory located rostral to the
mid-length of the maxilla, with a ratio less than 0.35 (mean of 0.28) for the relative
position of the rostrodorsal promontory.
Character 99: morphology of the apex of the dorsal process of the maxilla (Fig.
D.41; modified from Horner et al., 2004, character 48):
(0): subtriangular, not dorsoventrally taller than it is rostrocaudally wide
(1): dorsoventrally taller than it is wide, with a peaked and caudally inclined apex
Character 100: morphology of the jugal articulation surface (Fig. D.42-D.45)
(0): protruding lateral to the caudal third of the maxilla as a mediolaterally
compressed finger-like process directed caudolaterally, separated a short distance from
the lateral side of the element (Fig. D.42)
(1): process consisting on a promontory located dorsal and rostral to the
ectopterygoid shelf, bearing a concave and subtriangular, dorsolaterally-facing joint
surface for the jugal, with a caudolaterally directed corner (Fig. D.43)
(2): subtriangular joint surface for the jugal that is more laterally than dorsally-
facing, with a lateroventrally-directed pointed corner that is located adjacent and slightly
dorsal to the proximal end of the lateral ridge of the ectopterygoid shelf (Fig. D.44)
(3): dorsally elevated jugal joint (distance between the ventral margin of the jugal
joint and ectopterygoid shelf nearly equal to depth of the caudal segment of the maxilla),
caudal margin of the joint flush with the caudal margin of the rostrodorsal eminence of
the lateral side of the maxilla (Fig. D.45)
Character 101: arrangement of maxillary foramina ventral and rostral to the jugal
articulation (excluding large rostrodorsal or rostrolateral foramen) (Fig. D.46-D.49)
(0): positioned rostrocaudally and scattered throughout the lateral side of the
maxilla
(1): forming either a row or cluster that is oriented rostrodorsally
396
Character 102: number of maxillary foramina ventral and rostral to the jugal
articulation (excluding large rostrodorsal or rostrolateral foramen) (Table D.5; Fig.
D.50)
(0): seven or more
(1): six or less
Character 103: large rostral maxillary foramen (Fig. D.51 and D.52; Evans and
Reisz, 2007, character 22 modified after Horner et al., 2004, character 44)
(0): opening on the rostrolateral body of the maxilla, within the rostral half of the
rostrodorsal margin of the element and exposed in lateral view
(1): opening on the rostrolateral body of the maxilla, within the dorsal half of the
rostrodorsal margin of the element and exposed in lateral view
(2): opening on the dorsal surface of the maxilla along the maxilla-premaxilla
contact, not exposed laterally
Character 104: maxilla-lacrimal contact (Fig. D.53 and D.54; Evans and Reisz, 2007,
character 23)
(0): present externally
(1): largely covered externally by the jugal-premaxilla contact
Character 105: length of the ectopterygoid shelf relative to the total rostrocaudal
length of the alveolar margin of the maxilla (Table D.6; Fig. D.55; partially after
Hai-lu You et al., 2003, character 12)
(0): ratio between the length ectopterygoid shelf and the length of the rostrocaudal
alveolar margin up to 0.25 (mean ratio of 0.20)
(1): ratio greater than 0.25 and up to 0.35 (mean ratio of 0.30)
(2): ratio greater than 0.35 (mean ratio of 0.45)
Character 106: slope of the ectopterygoid shelf, measured as the angle between this
and the rostrocaudal axis of the caudal segment of the tooth row (Table D.7; Fig.
D.56)
(0): steeply inclined caudoventrally, with an angle greater than 21º (mean angle of
29º)
(1): shelf inclined with an angle greater than 10 and up to 21º (mean angle of 15º)
397
(2): slightly inclined shelf, with an angle greater than 4º and up to 10º (mean
angle of 8º)
(3): horizontal shelf, with an angle up to 4º
Character 107: morphology of the lateral emargination of the ectopterygoid
shelf (Fig. D.57; modified from Godefroit et al., 2000, character X)
(0): dorsoventrally thin ridge
(1): faint or dorsoventrally thin rostrally, then abruptly becoming dorsoventrally
thick along the caudal segment of the margin
(2): dorsoventrally thick continuous ridge, gradually thicker caudally than
rostrally.
Character 108: position of the central region of the arcuate row of special foramina
on the medial side of the dentary (Fig. D.58)
(0): ventral to or at the level of the mid-dorsoventral depth of the maxilla
(1): dorsal to the mid-dorsoventral depth of the maxilla
Lacrimal
Character 109. General morphology of the adult lacrimal in lateral view (Fig. D.59)
(0): triangular and rostrocaudally elongated, with a rostral process that is rostrally
(and slightly ventrally) directed
(1): triangular, rostrocaudally abbreviated with a relatively shorter and thinner
rostral process.
Comment: the lacrimal increases in length with age, as the skull becomes rostrally
elongated in adult hadrosaurids. Thus, this character is only applicable among adults.
Character 110: ventral margin of the lacrimal with a prominent convexity rostral to
the jugal notch (Fig. D.60)
(0): absent
(1): present
398
Jugal
Character 111: rostral apex of the rostral process of the jugal (Fig. D.61 and D.62)
(0): present, wedge-shaped, elongated and sharply pointed, positioned at mid
distance along the dorsoventral depth of the rostral process
(1): present, wedge-shaped, pointed and less elongated than in (0), positioned
within the dorsal half of the rostral process of the jugal; the dorsal magin of the apex
forms a steeper angle with the horizontal than in state (0)
(2): greatly reduced to a blunt convexity
(3): reduced to a short process, only slightly thinner rostrally and ending abruptly
(4): absent, straight nearly vertical rostral margin
Character 112: dorsoventral expansion of the caudodorsal margin of the rostral
process of the jugal (Fig. D.63; Weishampel et al., 1993, character 15)
(0): dorsoventrally narrow, rostrodorsally directed and forming little of the
rostroventral margin of the orbital rim
(1): dorsoventrally deep (about 60-90% as deep as the rostral jugal constriction),
dorsally or slightly recurved caudodorsally, forming the rostroventral corner of the orbital
rim
Character 113: morphology of the triangular caudoventral margin of the rostral
process of the jugal (Fig. D.64)
(0): shallow and rostrocaudally wide prominence (wider than deep)
(1): ventrally pointed, approximately as deep as or slightly deeper as its proximal
end is wide
(2): ventrally projected triangular narrow process, at least twice as deep as it is
wide, sharply pointed and often recurved caudally
Character 114: location of the caudoventral apex of the rostral process relative to
the caudodorsal articulation with the lacrimal (with longitudinal axis of the rostral
process oriented horizontally) (Fig. D.65)
(0): apex located caudoventral to the caudal margin of the lacrimal process
(1): apex located ventral to the caudal margin of the lacrimal process
399
Character 115: orientation of the medial articular surface of the rostral process of
the jugal (Fig. D.66)
(0): facing medioventrally, the articular surface forms a deep concavity bounded
dorsally and caudally by a laterally offset rim
(1): facing medially, the articular surface is bounded only caudally by a rim of
bone
Character 116: inclination of the bony rim that bounds caudally the medial
articulation surface of the jugal rostral process (this inclination is relative to the
rostrocaudal longer axis of the jugal) (Fig. D.67)
(0): slightly inclined rostrally (less than 100º) or nearly vertical
(1): strongly inclined rostrally, approximately 120º
Character 117: rostrocaudal width of the curvature of the ventral margin of the
jugal, between the caudoventral flange and the rostral process (modified from
Norman, 2002) (Fig. D.68)
(0): relatively wide and shallow embayment
(1): relatively narrow and deep embayment
Character 118: ventral expansion of the caudoventral jugal flange (measured as the
ratio between the dorsoventral depth of the flange and the minimum depth of the
caudal constriction of the jugal) (modified from Wagner, 2001) (Table D.8; Fig.
D.69)
(0): slightly expanded flange, ratio of 1.36 or less (mean ratio of 1.29)
(1): moderately expanded flange, ratio greater than 1.36 and up to 1.55 (mean
ratio of 1.44)
(2): greatly expanded flange, ratio greater than 1.55 (mean ratio of 1.68)
Character 119: lateral profile of the quadratojugal flange (Fig. D.70 and D.71)
(0): subconical, dorsoventrally tall and rostrocaudally narrow, with a nearly
vertical caudal margin
(1): auricular in shape, with subparallel concave to nearly straight dorsal and
convex ventral margins that converge dorsally into a short subconical point
(2): fan-like, with dorsal and ventral margins that are subparallel and diverge
caudodorsally; dorsal and ventral margins can be straight or slightly bowed dorsally
400
(3): auricular in shape, with subparallel concave to nearly straight dorsal and
convex ventral margins that converge dorsally into a recurved or dorsally directed tall
subconical extension. (Note: this state is similar to (1), but the dorsal region of the flange
is rostrocaudally narrower and taller).
Character 120: morphology of the ventral margin located between the caudoventral
and quadratojugal flanges (Fig. D.72; modified from Weishampel et al., 1993,
character 18)
(0): relatively short and shallow concavity
(1): relatively wide and well pronounced concavity
Character 121: relative depth of the caudal and rostral constrictions (in adults)
(rostral constriction: region located between the rostral and postorbital processes;
caudal constriction: region located between the postorbital process and the
caudoventral flange) (Table D.9; Fig. D.73)
(0): deeper rostral constriction, ratio of the depth of the caudal constriction
relative to the rostral of 1 or less (mean ratio of 0.93)
(1): deeper caudal constriction, with a ratio greater than 1 and less than 1.35
(mean ratio of 1.18, resulting from merging two K-means clusters with mean ratios of
1.13 and 1.24)
(2): much deeper caudal constriction, with a ratio greater than 1.35 (mean ratio of
1.43)
Comments: in juveniles the orbital margin is wider and the rostral constriction
much shallower than the caudal constriction, probably due to the proportionally larger
size of the orbit. For example, this could be observed in juveniles of Lambeosaurus such
as ROM 758 (ratio of 1.40), ROM 869 (ratio of 1.42) and AMNH 5340 (ratio of 1.54). In
subadult Edmontosaurus, such as CMN 8509, both the rostral and caudal constrictions
are shallower than in adults, where the caudal one becomes particularly deep.
Character 122: jugal overall robustness (in adults), measured as the ratio between
the minimum depth of the caudal constriction and distance between the point of
maximum curvature of the infratemporal margin and the caudal margin of the
lacrimal process (Fig. D.74; modified from Weishampel et al., 1993, character 13)
(0): relatively gracile jugal, ratio less than 0.60 (mean ratio of 0.49).
401
(1): relatively robust jugal, ratio of 0.60 or greater (mean ratio of 0.72).
Comments: in juveniles of at least Edmontosaurus, the above ratio had a lower
value relative to that in adults. However, in Lambeosaurus there was no substantial
change in the robustness of the jugal through ontogeny. For example, subadults ROM
758 with ratio of 0.69, ROM 869 with 0.70 and AMNH 5340 with 0.67, have ratios
within the range of those found in adults.
Character 123: relative width and lateral profiles of the orbital and infratemporal
margins of the jugal (Fig. D.75)
(0): wider orbital margin and relatively constricted ventral margin of the
infratemporal fenestra
(1): orbital and infratemporal margins are nearly equally wide
(2): wider infratemporal margin
Quadrate
Character 124: degree of curvature of the caudal margin of the quadrate (Wagner,
2001) (Table D.10; Fig. D.76)
(0): the caudal margin of the dorsal half or third of the quadrate displays a slight
curvature relative to the ventral half of the element, with an angle of 150º or greater
(mean angle of 161º)
(1): the caudal margin of the dorsal half or third of the quadrate is strongly curved
caudally relative to the ventral half of the element, with an angle less than 150º (mean
angle of 143º)
Character 125: position of the quadratojugal (paraquadrate) notch along the
dorsoventral length of the quadrate (measured as the ratio between the distance
from the mid-length of the notch to the qudadrate head and the dorsoventral length
of the element) (Table D.11; Fig. D.77)
(0): the mid point of the notch is located near the mid-length of the quadrate, ratio
less than 0.60 (mean ratio of 0.54).
(1): the mid point of the notch is located ventral to the mid-length of the quadrate,
ratio of 0.60 or greater (mean ratio of 0.64).
402
Character 126: orientation of the dorsal margin of the quadratojugal notch of the
quadrate (measured as the angle between this and the caudal margin of the element)
(Table D.12; Fig. D.78)
(0): angle greater than 45º (mean angle of 52º)
(1): angle up to 45º (mean angle of 28º)
Character 127: morphology of the lateral profile of the quadratojugal notch of the
quadrate (Fig. D.79)
(0): subcircular, with a ventral half of the notch that is recurved and has a
horizontal rostral segment
(1): wide arcuate and asymmetrical, with the ventral half of the notch having a
short horizontal rostral segment
(2): wide arcuate and symmetrical, the ventral half of the notch being
rostroventrally-directed, nearly straight as it is the dorsal half
Character 128: development of the squamosal buttress on the caudal margin of the
dorsal end of the of the quadrate (Fig. D.80)
(0): absent or poorly developed as a gentle convexity
(1): present, the buttress is a sharp protuberance hanging from the caudal side of
the dorsal fourth of the quadrate, near the head of the element
Character 129: morphology of ventral surface of the quadrate (Table D.13; Fig.
D.81 and D.82; Weishampel et al., 1993, character 22)
(0): mediolaterally broad and rostrocaudally compressed, lateral condyle slightly
larger than the medial one (mean ratio between the rostrocaudal width of the lateral
condyle and the mediolateral width of the ventral end of the quadrate of 0.59); the ventral
surface of the lateral condyle is only slightly offset ventrally relative to the ventral
surface of the medial condyle;
(1): subtriangular in ventral view, lateral condyle rostrocaudally expanded and
much larger than the medial one (mean ratio between the rostrocaudal width of the lateral
condyle and the mediolateral width of the ventral end of the quadrate of 0.90); the ventral
surface of the lateral condyle is well offset ventrally relative to the ventral surface of the
medial condyle.
403
Prefrontal
Character 130: dorsomedial margin of the prefrontal developed into a

caudodorsally-oriented crest (Fig. D.83 and D.84; Godefroit et al., 2004, character
16)
(0): absent
(1): present, not extending caudal to the prefrontal-frontal articulation
(2): present, the crest extends caudally over the dorsal surface of the frontal and
above the prefrontal-postorbital articulation in lateral view in adults
Character 131: lateral profile of the rostrodorsal margin of the prefrontal (Fig.
D.85; modified from Horner et al., 2004, character 50)
(0) subarcuate to smoothly curved, the rostral margin is rostroventrally oriented
and forming an obtuse angle with the dorsal orbital margin
(1): rostromedially broad with subsquared rostrodorsal corner, the rostral margin
is ventrally oriented and forms a 90º angle with the dorsal orbital margin.
Character 132: mediolateral breadth of the exposed rostroventral region of the
prefrontal (Fig. D.86; partially after Horner et al., 2004, character 50)
(0): the rostroventral region is mediolaterally expanded
(1): the exposed rostroventral region is mediolaterally compressed and narrow
Character 133 (PF4). Inclusion of the prefrontal in the circumnarial fossa (Figs.
D.87 and D.88; Wagner, 2001)
(0): absent
(1): present
Character 134 (PF5). Outward flaring of the rostrodorsal orbital margin of the
prefrontal (Fig. D.89; Horner et al., 2004, character 49)
(0): absent, the prefrontal lies flush with the surrounding lacrimal and postorbital
(1): present, the prefrontal flares dorsolaterally forming a thin and everted wing-
like rim around the rostrodorsal margin of the orbit
Character 135 (PF6). Exposure of the prefrontal-nasal contact in lateral and/or
dorsal view (Fig. D.90; modified from Wagner, 2001)
(0): contact totally exposed in lateral and/or dorsal view
404
(1): contact visible in lateral view along the caudal and half of the dorsal margin
of the prefrontal
(2): contact visible in lateral view only along the caudal region of the prefrontal in
adults, due to the invasion of the premaxilla along the medial side of the prefrontal.
Postorbital
Character 136: dorsal promontorium on the rostral process of the postorbital (Figs.
D.91 and D.92; Godefroit et al., 2004, character 17)
(0): absent, the dorsal surface of the postorbital above the jugal process is
horizontal or slightly concave
(1): present in adult specimens, the articular margin for the prefrontal is elevated
and the dorsal surface of the postorbital above the jugal process is deeply depressed
Godefroit et al. (2004) incorrectly reported the absence of this process in
Amurosaurus riabinini. When a dorsal promontory is present, the dorsal surfaces of the
caudal and rostral processes of the postorbital are steeply inclined towards the dorsal
surface of the central body of the bone.
In Charonosaurus jiayinensis, the relatively small size of CUST JIV 1223
indicates that this character is probably present in subadult specimens as well as in adults,
although probably not developed in juveniles. However, sudadult individuals do not
display a deeply depressed dorsal surface, as in Lambeosaurus lambei (ROM 758) and
Corythosaurus casuarius (CMN 8676). In Parasaurolophus cyrtocristatus (UCMP
143270) there is a small dorsal promontorium and the inclination of the caudal process of
the postorbital, while not horizontal, is not as steep as in those taxa with postorbitals
having depressed dorsal margins.
Character 137: rostrocaudal constriction of the dorsal region of the infratemporal
fenestra (Fig. D.93; modified from Evans and Reisz, 2007, character 36)
(0): absent, caudal (squamosal) process of the postorbital elongate over the
infratemporal fenestra (broad and subrectangular dorsal region of the fenestra)
405
(1): present and caused by the presence of a nearly straight and oblique
caudoventral margin of the caudodorsal region of the postorbital (dorsal region of
infratemporal fenestra typically subtriangular)
(2): present and caused by rostrocaudal shortening of the caudal process of the
postorbital (dorsal region of infratemporal fenestra typically oval)
Comments: this character appeared invariable throughout ontogeny, as in for
example Saurolophus agustirostris. In this species state 2 is was in subadults (e.g., ZPAL
MgD-I 159) as well as in adults (e.g., MPC-D100/706). In Edmontosaurus spp, the
constriction of the infratemporal fenestra is further accentuated by the “swelling” of the
postorbital. Nothwithstanding this “swelling”, the condition in the genus is as in state 1.
Character 138: morphology of the central body of the postorbital (Fig. D.94)
(0): triangular, craniocaudally broad, expanded rostroventrally to form a straight
and obliquely oriented caudodorsal orbital margin
(1): triangular, with a caudodorsal orbital margin that ranges in lateral profile
from semicircular to subsquared
(2): rostrocaudally expanded, rostrally excavated and and bulging laterally
(“inflated”), containing a hollow inner cavity (in adults)
Comments: state 2 was not expressed in subadults and was observed to increase in
development with increasing age of the specimens.
Character 139: length of the jugal process of the postorbital (Fig. D.95)
(0): relatively short, approximately as long as the craniocaudal width of the orbit,
hook-like in lateral profile
(1): relatively long, longer than the craniocaudal width of the orbit, nearly
straight, only slightly recurved rostrally
Character 140: morphology of the caudal end of the caudal process of the
postorbital at its articulation with the squamosal (Fig. D. 96; Evans and Reisz,
2007, character 35)
(0): oblong or wedge-shaped
(1): bifid
406
Character 141: caudal extension of the caudal ramus of the postorbital that overlaps
the laterodorsal surface of the squamosal (Fig. D.97; modified from Godefroit et al.,
2000)
(0): the caudal end of the postorbital caudal ramus extends to a point rostral to the
quadrate cotylus and does not overlap the latter.
(1): the caudal end of the postorbital caudal ramus extends caudodorsal to the
precotyloid process and over as much as the rostral half of the quadrate cotylus.
(2): the caudal end of the postorbital caudal ramus completely overlaps the
laterodorsal side of the squamosal quadrate cotylus.
Squamosal
Character 142: length of the precotyloid process of the squamosal (measured as the
ratio of its length relative to the width of the quadrate cotylus) (Table D.14 and
Fig.D.98)
(0): very short precotyloid process, ratio less than 0.95 (mean ratio of 0.74)
(1): moderately long precotyloid process, ratio between 0.95 and 1.25 (mean ratio
of 1.13)
(2): very long precotyloid process, ratio greater than 1.25 (mean ratio of 1.41)
Character 143: dorsoventral expansion of the caudolateral surface of the squamosal
(Fig. D.99; Horner et al., 2004, character 64)
(0): unexpanded, shallowly exposed in caudal view
(1): greatly expanded dorsomedially, forming a deep, near vertical, well-exposed
face in caudal view (in adults)
Character 144: separation of the squamosals at the occipital margin of the skull roof
(Fig. D.100; Suzuki et al., 2004, character 33)
(0): completely separated by the parietal
(1): the squamosal approach the sagittal plane of the skull, separated by a narrow
band of parietal
(2): extensive intersquamosal joint present at the midline, parietal completely
excluded from the sagittal plane of the skull at that particular spot (in adults)
407
Character 145: rostromedial indenture of the medial ramus of the squamosal (Fig.
D.101; modified from Godefroit et al., 1998)
(0): present, medial ramus of the squamosal curves rostromedially, so that the
back of the skull appears to be deeply indented rostrally when viewed dorsally
(1): absent, medial ramus of the squamosal extends medially, forming a
subsquared caudolateral border of the skull roof
408
Table D.1. Character 65. Ratio (R) between the maximum width of the premaxillary oral
margin and the minimum width of the skull at the post-oral constriction for a sample of
iguanodontoidean ornithopods. Included is the corresponding character coding derived
from K.Means clustering analysis.
Specimen Taxon R K-Means Cluster

(State)
MAL 0329-002 cf. Bactrosaurus sp. 1.87 1
TMP 90.104.1 Brachylophosaurus canadensis 2.22 2
SM R4036 Edmontosaurus annectens 2.24 2
CM 26258 Edmontosaurus regalis 2.17 2
BMNH R8297 Edmontosaurus regalis 2.32 2
TMP 81.37.1 Lambeosaurus lambei 1.85 1
409
Rato of the Mediolateral Expansion of the Premaxilla
2.2
2.0
1.8
−20
−30
1.6
−40
BIC
−50
−60
E V
−70
1.4
2 4 6 8
Number of Clusters
Br Coi Cos Ed Eda Edr Grl Grn Hy Iga Igb Lml Ma Ou Pr Pt Ts cfBa
Taxon
Figure D.1. Character 65. Boxplot showing the distribution of the ratio of the medioteral expansion of the premaxilla in a sample of
iguanodontoidean taxa. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which the
data can be classified. The photograph shows how the two variables of this raio were measured, exemplified in the premaxilla of Gryposaurus
notabilis (CMN 2278).
410
state 0
A
state 1
B
B
Figure D.2. Character 66. Position of the premaxillary oral margin relative to the occlusal
plane of the dentition. A, Equijubus normani (IVPP V12534), lateral view. B, Edmontosaurus
regalis (CMN 2288), lateral view.
411
A B
C D
Figure D.3. Character 67. Morphology of the oral margin (arrows) of the premaxilla. State 0: moderately
expanded, dorsoventrally thicker towards the parasagittal plane of the snout and slightly deflected
ventrally. A, Iguanodon bernissartensis (IRSNB 1731), lateral view. B, Ouranosaurus nigeriensis (cast of
GDF 300), lateral view. C, Iguanodon atherfieldensis (BMNH R11521), lateral view. D, Telmatosaurus
transsylvanicus (BMNH R3386), lateral view.
412
A B
C D
Figure D.4. Character 67. Morphology of the oral margin (arrows) of the premaxilla. State 1: moderately
expanded, becoming thinner towards the parasagittal plane of the snout, so that the narial fossa
merges with the rostromedial region of the premaxilla adjacent to the contact between both premaxillae.
A, Corythosaurus intermedius (CMN 8676), lateral view. B, Hypacrosaurus altispinus (CMN 8501),
lateral view. C, “Lambeosaurus” laticaudus (LACM 17715), lateral view. D, Lambeosaurus lambei
(CMN 2869), lateral view.
413
A B C
Figure D.5. Character 67. Morphology of the oral margin of the premaxilla. State 2: oral margin (arrows) folded caudodorsally, thin
and recurved. A, Gryposaurus notabilis (ROM 873), lateral view. B, Prosaurolophus maximus (CMN 2277), lateral view. C, Saurolophus
osborni (AMNH 5220), lateral view.
414
A B C
D E F
Figure D.6. Character 67. Morphology of the oral margin of the premaxilla. State 3: ventrally deflected and dorsoventrally expanded margin
(arrows), forming a thick “lip-like” border. A, Brachylophosaurus canadensis (MOR 794), lateral view. B, Edmontosaurus regalis (CMN 2288),
lateral view. C, Hypacrosaurus stebingeri (MOR 549), lateral view. D, Maiasaura peeblesorum (OTM F138), lateral view. E, Parasaurolophus
walkeri (ROM 768), lateral view. F, Tsintaosaurus spinorhinus (IVPP V725), lateral view.
415
state 0
A
state 1
Figure D.7. Character 68. Morphology of the rostrolateral corner (arrows) of the premaxillary
oral margin. A, Equijubus normani (IVPP V12534), lateral view. B, Lambeosaurus lambei (ROM 794),
lateral view.
416
state 0
A
state 0
B
state 1
C
state 1
Figure D.8. Character 69. Presence and absence of a double denticle layer (arrows) ventral to the
premaxillary oral margin. A, Iguanodon atherfieldensis (BMNH R11521), ventral view. B, Telmatosaurus
transsylvanicus (BMNH R3386), ventral view. C, Bactrosaurus johnsoni (AMNH 6501), lateral view.
D, Gryposaurus latidens (AMNH 5465), lateral view.
417
state 0 state 0
A B
state 1 state 1
C D
Figure D.9. Character 70. Presence (arrows) and absence of premaxillary foramen. A, Iguanodon
bernissartensis (IRSNB 1731), caudodorsolateral view. B, Lambeosaurus lambei (CMN 8633), dorsolateral
view. C, Brachylophosaurus canadensis (CMN 8893), dorsolateral view. D, Edmontosaurus annectens
(SM R4036), lateral view.
418
state 0
A
state 1
Figure D.10. Character 71. Presence and absence of accessory premaxillary foramen.
A, Iguanodon bernissartensis (CMN 2278), lateral view. B, Edmontosaurus annectens (CMN 8509),
lateral view. Abbreviations: pmxf = premaxillary foramen; apmxf = accessory premaxillary foramen.
419
A B C
Figure D.11. Character 72. Presence (arrow) and absence of premaxillary accessory fossa. A, Equijubus normani (IVPP V12534), lateral
view. B, Gryposaurus sp. (MOR 553S-7-23-8-57), lateral view. C, Maiasaura peeblesorum (OTM F138), lateral view.
420
state 0
A
state 1
Figure D.12. Character 73. Presence (arrow) and absence of an additional accessory fossa in the
premaxilla. A, Brachylophosaurus canadensis (TMP 90.104.1), dorsolateral view. B, Edmontosaurus
regalis (CM 26258), dorsolateral view.
421
state 0 state 1
A B
state 1
state 0
C D
Figure D.13. Character 74. Elongation (arrows) of the premaxillary caudodorsal process over the orbit. A, Gryposaurus sp.
(MOR553S-7-18-91-107), lateral view. B, Corythosaurus sp. (CMN 11375), lateral view. C, Gryposaurus notabilis (ROM 873), lateral view.
D, Lambeosaurus lambei (CMN 2869,lateral view.
422
state 0
A
state 1
Figure D.14. Character 75. Presence and absence of vertical groove (arrow) on the
caudoventral process of the premaxilla. A, Parasaurolophus walkeri (ROM 768), lateral view.
B, Corythosaurus casuarius (AMNH 5338), lateral view.
423
A B C
Figure D.15. Character 67. Elongation of the caudoventral process of the adult premaxilla; caudal end of this process indicated by arrows.
A, Brachylophosaurus canadensis (TMP 90.104.1), lateral view. B, Saurolophus osborni (AMNH 5220), lateral view. C, Lambeosaurus lambei
(CMN 8703), lateral view. Abbreviations: lc = lacrimal; ns = nasal; pf = prefrontal; pmx = premaxilla.
424
A
the adult premaxilla. State 0: mediolaterally compressed and triangular. A, Brachylophosaurus canadensis
(CMN 8893), lateral view. B, Parasaurolophus walkeri (ROM 768), lateral view.
Abbreviations: cddrs = caudodorsal; cdvnt = caudoventral; ns = nasal; pmx = premaxilla.
425
A
Figure D.17. Character 77. Morphology of the caudal region of the caudoventral process of the adult
premaxilla. State 1: dorsoventrally broad and directed caudally or caudally and slightly dorsally.
A, Hypacrosaurus stebingeri (MOR 549), lateral view. B, Corythosaurus casuarius (ROM 871), lateral view.
Abbreviations: cddrs = caudodorsal; cdvnt = caudoventral; ns = nasal; pmx = premaxilla.
426
A
Figure D.18. Character 77. Morphology of the caudal region of the caudoventral process of the adult
premaxilla. State 2: triangular and dorsoventrally expanded, laterally convex lobe directed
rostrodorsally. A, Hypacrosaurus stebingeri (MOR 549), lateral view. B, Corythosaurus casuarius (ROM 871),
lateral view. Abbreviations: cddrs = caudodorsal; cdvnt = caudoventral; ns = nasal; pmx = premaxilla.
427
Figure D.19. Character 78. Presence of an accessory rostroventral flange (black rectangle) in the
caudodorsal premaxillary process that overlaps the lateral surface of the nasal in the rostral region of a
supracranial crest. Exemplified here in a specimen of Lambeosaurus lambei (TMP 81.37.1), lateral view.
428
state 0 state 0
A B
state 1 state 2
C D
Figure D.20. Character 79. Laterodorsal profile of the caudodorsal and caudoventral margins of
the external bony naris. A, Gryposaurus notabilis (ROM 873), lateral view; the white rectangle surrounds the
area regarded homologous to the narial opening displayed in B, C and D. B, Tsintaosaurus spinorhinus
(IVPP V725), lateral view. C, Hypacrosaurus stebingeri (MOR 549), lateral view. D, Parasaurolophus walkeri
(ROM 768), lateral view.
429
A B C
Figure D.21. Character 80. Relative thickness of the caudal region of the caudodorsal and caudoventral processes (indicated by the
double-headed arrows). A, Parasaurolophus walkeri (ROM 768), lateral view. B, “Lambeosaurus” laticaudus (LACM 17715), lateral view.
C, Hypacrosaurus altispinus (ROM 702), lateral view. Abbreviations: cddrs = caudodorsal; cdvnt = caudoventral; pmx = premaxilla.
430
3.0
Length/Width Ratio of the External Bony Naris
0
−10
2.5
BIC
−20
−30
−40
E V
2 4 6 8
Number of Clusters
2.0
Coi Cos Hy Hys Lml Lmm Ol Paw Ve lc
Taxon
Figure D.22. Character 81. Boxplot showing the distribution of the length/width ratio of the external bony naris of the premaxilla in a sample
of lambeosaurine hadrosaurids. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in
which the data can be classified. The photograph shows the two variables of the ratio were measured in a premaxilla of Corythosaurus
casuarius (ROM 1933).
431
Table D.2. Character 81. Length/width ratio of the external nares in a sample of
lambeosaurine hadrosaurs. Included is the corresponding character coding derived from

(State)
LACM 11775 "Lambeosaurus" laticaudus 2.41 1
TMP 84.121.1 Corythosaurus casuarius 3.02 2
CMN 8705 Lambeosaurus magnicristatus 3.07 2
TMP 66.4.1 Lambeosaurus magnicristatus 3.12 2
AEHM 2/845 Olorotitan ararhensis 2.31 1
CPC-59 Velafrons coahuilensis 2.56 1
432
state 0
A
state 1
Figure D.23. Character 82. Presence and absence of a well-developed dorsolateral flange (rectangles)
of the caudoventral process of the premaxilla. A, Brachylophosaurus canadensis (MOR 794),
dorsolateral view. B, Gryposaurus notabilis (ROM 873), dorsolateral view. Abbreviations: mx = maxilla;
ns = nasal; pmx = premaxilla.
433
A B C
Figure D.24. Character 83. Location of the nasal bone and nasal cavity in the adult skull. A, Gryposaurus notabilis (CMN 2278), lateral view.
B, Tsintaosaurus spinorhinus (IVPP V725), lateral view. C, Corythosaurus casuarius (ROM 871), lateral view. Abbreviation: ns = nasal.
434
state 0
state 1
Figure D.25. Character 84. Presence and absence of curvature of the caudodorsal region of the nasal.
A, Edmontosaurus annectens (MOR 003), lateral view. B, Hypacrosaurus altispinus (CMN 8501),
lateral view. Abbreviation: ns = nasal.
435
A B C
Figure D.26. Character 85. Morphology of the rostral end of the nasal at the contact with the dorsal process of the premaxilla.
A, Gryposaurus notabilis (ROM 873), lateral view. B, Maiasaura peeblesorum (ROM 44770), lateral view. C, Prosaurolophus maximus
(ROM 1928), lateral view. Abbreviations: ns = nasal; pmx = premaxilla.
436
state 3
A
state 4
Figure D.27. Character 85. Morphology of the rostral end of the nasal at the contact with the dorsal
process of the premaxilla. A, Lambeosaurus lambei (ROM 869), lateral view. B, Corythosaurus
(ROM 870), lateral view. Abbreviations: ns = nasal; pmx = premaxilla.
437
state 0 state 1
A B
state 2 state 2
state 0
C D
Figure D.28. Character 86. Morphology of the nasal contact with the caudodorsal region of the
caudoventral premaxillary process at the caudal margin of the narial foramen.A, Equijubus normani
(IVPP V725), lateral view. B, Brachylophosaurus canadensis (CMN 8893), lateral view.
C, Edmontosaurus sp. (BMNH R8927), lateral view. D, Prosaurolophus maximus (CMN 2277),
lateral view. Abbeviations: ns = nasal; pmx = premaxilla.
438
state 0
A
state 1
Figure D.29. Character 87. Location of the rostral end of the dorsal process of the nasal relative to
the rostral margin of the narial foramen (indicated by the separation between the pairs of white arrows).
A, Gryposaurus notabilis (ROM 873), lateral view. B, Prosaurolophus maximus (ROM 1928), lateral view.
Abbreviations: ns = nasal; pmx = premaxilla.
439
Figure D.30. Character 88. Presence of a caudoventral region of the nasal ventrally recurved and
hook-shaped, with a rostral process that inserts under the caudoventral process of the premaxilla.
Exemplified here in a specimen of Lambeosaurus lambei (CMN 8703), lateral view.
Abbreviations: ns = nasal; pmx = premaxilla.
440
A
Figure D.31. Character 89. Caudal end of the nasals forming a pair of finger-like process
(arrows) on top of the frontals and centered around the sagittal plane of the skull roof.
Exemplified here in A, a subadult specimen of Kritosaurus navajovius (NMMNH P-16106,
dorsal view) and B, Prosaurolophus maximus (CMN 2870). Abbreviations: f = frontal;
ns = nasal; pf = prefrontal; po = postorbital.
441
Figure D.32. Character 90. Caudal end of the nasals forming a pair of small and short processes
(arrows) that insert between the frontals at the sagittal plane of the skull roof. Exemplified here
in a specimen of Gryposaurus notabilis (AMNH 5350), dorsal view. Abbreviations: f = frontal;
ns = nasal; pf = prefrontal; po = postorbital.
442
state 0
state 1
Figure D.33. Character 91. Position of the summit (asterisk) of the nasal arch relative to the
caudodorsal margin of the narial foramen (rectangle). A, Gryposaurus latidens (AMNH 5465),
lateral view. B, G. notabilis (CMN 2278), lateral view.
443
state 0 state 0
A B
state 1 state 1
C D
Figure D.34. Character 92. Rostrodorsal process that is medially offset from the body of the maxilla
and extends medial to the caudovenventral process of premaxilla, shown in A, Brachylophosaurus
canadensis (MOR-1071-7-6-98-79), lateral view; and B, medial view of the same specimen. Rostral end
of the maxilla formed by a ventrally sloping rostrodorsal shelf that underlies the premaxilla, show in
C, Corythosaurus intermedius (CMN 8676), lateral view; and D, medial view of the same specimen.
444
Figure D.35. Character 93. Absence and presence of lateral exposure of the rostrodorsal process of the
maxilla (arrow). A, Saurolophus osbori (CMN 8796), lateral view. B, rachylophosaurus canadensis
(TMP 90.104.1), lateral view.
445
state 0
A
state 1
Figure D.36. Character 94. Presence and absence of a pendant rostroventral process of
the maxilla. A, Velafrons coahuilensis (CPC-59), lateral view. B, “Lambeosaurus” laticaudus
(LACM 17715), lateral view.
446
Table D.3. Character 95. Angle (A) between the dorsal margin of the maxillary
rostroventral region and the rostral segment of the maxillary tooth row for a sample of
iguanodontideans. Included is the corresponding character coding derived from K-Means
clustering analysis.
Specimen Taxon A K-Means

Cluster (State)
IRSNB SBDE/95E5 Bactrosaurus johnsoni 30 1
BMNH R3638 Edmontosaurus sp. 20 0
MOR549-6-19-8-9 Hypacrosaurus stebingeri 17 0
BMNH R5764 Iguanodon atherfieldedensis 19 0
BMNH R11521 Iguanodon atherfieldedensis 18 0
IPS-SRA 22 Pararhabdodon isonensis 23 0
447
Table D.3—continued.

Cluster (State)
BMNH R4911 cf. Telmatosaurus 29 1
FGGUB R1010 cf. Telmatosaurus 31 1
CPC-95 Velafrons cohauilensis 23 0
TMM 42452 Hadrosauridae 21 0
448
45
−320
−340
40
BIC
Angle of Inclination of the Rostral Maxilla
−360
−380
E V
35
2 4 6 8
Number of Clusters
30
25
20
15
Am Ba Br Coi Cos Ed Eda Edr Gi Gr Grl Grm Hy Hys Iga Lml Ma Ol Ou Pat Ph Pr Pt Sao Ts Ve cfTe x
Taxon
Figure D.37. Character 95. Boxplot showing the distribution of the angle between the dorsal margin of the rostroventral process or shelf
of the maxilla and the rostral segment of the tooth row. The inserted smaller graph shows the Bayesian information criterion (BIC) for various
numbers of clusters in which the data can be classified. The photograph shows how this angle was measured, exemplified in a maxilla of
Corythosaurus casuarius (ROM 1933).
449
A B C
Figure D.38. Character 96. Lateral profile of the lateral surface of the rostrodorsal region (rectangles) of the maxilla. A, Saurolophus osborni
(AMNH 5221), lateral view. B, Corythosaurus casuarius (ROM 1933), lateral view. C, Pararhabdodon isonensis (IPS-SRA 22), lateral view.
B
450
state 0
A
state 1
Figure D.39. Character 93. Elevation of the lateral surface of the rostrodorsal region of the maxilla.
B
A, Brachylophosaurus canadensis (MOR 1071-7-6-98-79), lateral view. B, Gryposaurus sp. (MOR
553S-8-26-9-55), lateral view.
451
Table D.4. Character 98. Position of the rostrolateral region of the maxilla (expressed as
the ratio (R) between the distance from its dorsal margin to the rostral end of the maxilla
and the rostrocaudal distance of the element) for a sample of iguanodontoideans.
Included is the corresponding character state coding derived from K-Means clustering.

(State)
AEHM1/12 Amurosaurus riabinini 0.42 2
IRSNB SBDE/95E5 Bactrosaurus johnsoni 0.49 1
IRSNB SBDE/95E5 Bactrosaurus johnsoni 0.51 1
BMNH R3638 Edmontosaurus sp. 0.44 2
CMN 8744 Edmontosaurus sp. 0.40 2
AMNH 6551 Gilmoreosaurus mongoliensis 0.39 2
AMNH 5472 Hadrosauridae 0.38 2
AEHM 2/845 Olorotitan arharensis 0.51 1
452

(State)
453
100
Ratio of the Position of the Rostrodorsal Eminence of the Maxilla
90
80
0.6
BIC
70
60
50
0.5
2 4 6 8
Number of Clusters
0.4
0.3
Am Ba Br Coi Cos Ed Edr Gi Gr Grl Grn Hy Hys Iga Lml Ma Ol Ou Pat Pr Pt Sao Ts cfTe x
Taxon
Figure D.40. Character 98. Boxplot showing the distribution of the position of the dorsal process and the dorsal margin of the dorsolateral
eminence of the maxilla for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion
(BIC) for various numbers of clusters in which the data can be classified. The photograph shows how the two variables of this raio were
measured, exemplified in a maxilla of Gryposaurus sp. (MOR 553S-8-26-9-55).
454
A B C
D E F
Figure D.41. Character 99. Morphology of the dorsal process of the maxilla. A, Bactrosaurus johnsoni (SBDE 95/E5), lateral view.
B, Maiasaura peeblesorum (OTM F138), lateral view. C, Prosaurolophus maximus (TMP 84.1.1), lateral view. D, Amurosaurus riabinini
(AEHM 1/12), lateral view. E, Hypacrosaurus altispinus (CMN 8674), lateral view. F, Tsintaosaurus spinorhinus (IVPP V725), lateral view.
455
A
Figure D.42. Character 1000. Morphology of the jugal articulation surface of the maxilla. State 0:
jugal joint protruding lateral to the caudal third of the maxilla as a mediolaterally compressed
finger-like process (rectangle and arrow). Exemplified here in a specimen of Iguanodon atherfieldensis
(BMNH R5764) in A, lateral and B, dorsal view.
456
A
State 1: joint located on a promontory (rectangles) dorsal and rostral to the ectopterygoid shelf, with
a caudolaterally directed corner (arrow). Exemplified here in a specimen of Bactrosaurus johnsoni
(BMNH 6553) in A, lateral and B, dorsal view. C, cf. Telmatosaurus sp. (FGGUB R1010), lateral view.
457
A
subtriangular joint surface (rectangles) for the jugal that is more laterally than dorsally-facing,
with a lateroventrally-directed pointed corner (arrow) located adjacent and slightly dorsal to the the
lateral ridge of the ectopterygoid shelf. Exemplified here in a specimen of Brachylophosaurus canadensis
(MOR 1071-8-13-98-554) in A, lateral and B, dorsal view. C, Corythosaurus intermedius (CMN 8676),
lateral view.
458
A
dorsally elevated jugal joint (distance between the ventral margin of the jugal joint and
ectopterygoid shelf nearly equal to depth of the caudal segment of the maxilla), caudal margin of the
joint flush with the caudal margin of the rostrodorsal eminence of the lateral side of the maxilla
Exemplified here in a specimen of Tsintaosaurus spinorhinus (IVPP V725) in A, lateral and B, dorsal
view. C, Pararhabdodon isonensis (IPS SRA 22), lateral view.
459
state 0 state 0
A B
state 0 state 0
C D
state 1 state 1
E F
Figure D.46. Character 101. Arrangement of maxillary foramina (arrows) ventral and rostral to the
jugal articulation, shown in lateral view. A, Bactrosaurus johnsoni (cast of PIN 2549-1). B, Equijubus
normani (IVPP V12534). C, Gilmoreosaurus mongoliensis (AMNH 6551). D, Iguanodon atherfeldensis
(BMNH R11521). E, Brachylophosaurus canadensis (CMN 8893). F, B. canadensis (MOR 1071-7-6-98-79).
460
state 1 state 1
A B
state 1 state 1
C D
state 1 state 1
E F
Figure D.47. Character 101. Arrangement of maxillary foramina (arrows) ventral and rostral to
the jugal articulation, shown in lateral view. A, Brachylophosaurus canadensis (MOR 1071-8-13-98-559).
B, B. canadensis (MOR 1071-8-15-98-573). C, B. canadensis (MOR 794). D, B. canadensis (TMP 90.104.1).
E, Lambeosaurus lambei (CMN 2869). F, L. lambei (TMP 81.37.01).
461
state 1 state 1
A B
state 1 state 1
C D
state 1 state 1
E F
the jugal articulation, shown in lateral view. A, Edmontosaurus regalis (AMNH 5254). B, E. regalis
(CMN 2289). C, E. regalis (ROM 801). D, Saurolophus osborni (AMNH 5220). E, S. osborni (AMNH
5221, left maxilla). F, S. osborni (AMNH 5221, right maxilla).
462
state 1 state 1
A B
state 1 state 1
C D
state 1 state 1
E F
the jugal articulation, shown in lateral view. A, Prosaurolophus maximus (CMN 2277). B, Corythosaurus
intermedius (CMN 8676). C, C. intermedius (ROM 776). D, C. intermedius (ROM 777). E, C. casuarius
(ROM 1933). F, C. casuarius (TMP 82.37.01).
463
Table D.5. Character 102. Number of foramina rostral and ventral to the jugal articulation
(excluding the large dorsal or rostrodorsal foramen) in a sample of iguanodontoidean
ornithopods. Abbreviations: lt = left; rt = right.
Specimen Taxon Number of

Foramina
AEHM1/12 (lt) Amurosaurus riabinini 3
PIN 2549-1 (lt) Bactrosaurus johnsoni 7
CMN 8893 (rt) Brachylophosaurus canadensis 3
MOR 1071-7-6-98-79 (rt) Brachylophosaurus canadensis 5
MOR 1071-8-13-98-554 (lt) Brachylophosaurus canadensis 5
MOR 1071-8-13-98-559 (rt) Brachylophosaurus canadensis 5
MOR 1071-8-15-98-573 (lt) Brachylophosaurus canadensis 5
MOR 794 (rt) Brachylophosaurus canadensis 4
TMP 90.104.1 (rt) Brachylophosaurus canadensis 3
CMN 8676 (rt) Corythosaurus intermedius 3
ROM 776 (rt) Corythosaurus intermedius 3
ROM 777 (lt) Corythosaurus intermedius 5
ROM 1933 (lt) Corythosaurus casuarius 5
TMP 82.37.01 (rt) Corythosaurus casuarius 4
CMN 8704 (lt) Corythosaurus intermedius 3
TMP 80.23.4 (lt) Corythosaurus casuarius 2
CMN 8509 (rt) Edmontosaurus annectens 2
ROM 867 (rt) Edmontosaurus annectens 3
SM R4036 (lt) Edmontosaurus annectens 4
AMNH 5254 (lt) Edmontosaurus regalis 4
CMN 2289 (rt) Edmontosaurus regalis 3
ROM 801 (lt) Edmontosaurus regalis 2
UCM 20892 (lt) Edmontosaurus regalis 2
BMNH R3638 (lt) Edmontosaurus sp. 3
IVPP V12534 (lt) Equijubus normani 8
AMNH 6551 (rt) Gilmoreosaurus mongoliensis 6
CMN 2278 (lt) Gryposaurus notabilis 3
ROM 764 (rt) Gryposaurus notabilis 2
ROM 873 (rt) Gryposaurus notabilis 2
TMP 80.22.1 (lt) Gryposaurus notabilis 2
MOR 553S-7-24-8-68 (rt) Gryposaurus sp. 4
AMNH 5472 (lt) Hadrosauridae 3
CMN 8501 (rt) Hypacrosaurus altispinus 3
CMN 8673 (lt) Hypacrosaurus altispinus 3
464
Specimen Taxon Number of

Foramina
MOR 549-6-19-8-9 (rt) Hypacrosaurus stebingeri 3
BMNH R11521 (lt) Iguanodon atherfieldensis 12
BMNH R454 (lt) Iguanodon sp. 7
NMMNH P-16106 (rt) Kritosaurus navajovius 4
CMN 8703 (rt) Lambeosaurus lambei 4
CMN 8703 (lt) Lambeosaurus lambei 3
ROM 758 (lt) Lambeosaurus lambei 2
ROM 1218 (lt) Lambeosaurus lambei 3
TMP 81.37.1 (rt) Lambeosaurus lambei 3
YPM 3222 (lt) Lambeosaurus lambei 3
LACM 17715 (rt) "Lambeosaurus" laticaudus 2
OTM F138 (lt) Maiasaura peeblesorum 3
ROM 44770 (lt) Maiasaura peeblesorum 4
TCMI 2001.89.2 (lt) Maiasaura peeblesorum 3
AEHM 2/845 (lt) Olorotitan arharensis 3
GDF 300 (rt) Ouranosaurus nigeriensis 5
IPS SRA 22 (lt) Pararhabdodon isonensis 3
Unrecorded no. (rt) Parasaurolophus tubicen 4
MNA PI 529 (rt) Parasaurolophus tubicen 6
NMMNH P-25100 (lt) Parasaurolophus tubicen 4
CMN 2870 Prosaurolophus maximus 5
MOR 447 (rt) Prosaurolophus maximus 4
ROM 787 (rt) Prosaurolophus maximus 4
CMN 2277 (rt) Prosaurolophus maximus 5
TMP 84-1-1 (lt) Prosaurolophus maximus 5
TMP 84-1-1 (rt) Prosaurolophus maximus 4
SMU 74582 (rt) Protohadros byrdi 5
MPCA SM-12 (rt) Salitral Moreno hadrosaurid 2
ZPAL MgD-I159 (lt) Saurolophus angustirostris 4
AMNH 5221 (lt) Saurolophus orboni 6
AMNH 5221 (rt) Saurolophus orboni 5
CMN 8796 (rt) Saurolophus orboni 3
AMNH 5220 (rt) Saurolophus osborni 3
IVPP V725 Tsintaosaurus spinorhinus 4
465
12
−100
−150
BIC
−200
10
−250
Number of Foramina
−300
8
2 4 6 8
Number of Clusters
6
4
2
Am Ba Br Co Ed Eda Edr Eq Gi Gr Grn Hy Hys Ig Iga Krn Lml Ma Ol Ou Pat Ph Pr Pt Saa Sao Ts lc sal x
Taxon
Figure D.50. Character 102. Boxplot showing the distribution of the number of maxillary foramina ventral and rostral to the jugal articulation
for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers
of clusters in which the data can be classified. The photograph shows the shape and location of these foramina in a maxilla of
Brachylophosaurus canadensis (MOR 1071-8-13-98-554).
466
state 0
A
state 0
B
state 0
CB
A, Equijubus normani (IVPP V725), lateral view. B, Jinzhousaurus yangi (IVPP V12691),
lateral view. B, Bactrosaurus johnsoni (SBDE 95/E5), lateral view.
467
state 1 state 1
A B
state 2 state 2
C D
B
A, Brachylophosaurus canadensis (MOR 1071-8-13-98-559), rostral view. B, lateral view of the same
specimen. C, Corythosaurus sp. (TMP 82.37.01), rostral view. D, lateral view of the same specimen
468
A B
C D
Figure D.53. Character 104. State 0: presence of maxilla-lacrimal contact. A, Gryposaurus notabilis
(CMN 2278), lateral view. B, Prosaurolophus maximus (ROM 1928), lateral view. C, Parasaurolophus
walkeri (ROM 768), lateral view. D, Saurolophus osborni (AMNH 5220), lateral view.
Abbreviations: j = jugal; lc = lacrimal; mx = maxilla; ns = nasal; pf = prefrontal; pmx = premaxilla.
469
A B
C D
Figure D.54. Character 104. State 1: absence of maxilla-lacrimal contact. A, Brachylophosaurus

canadensis (TMP 90.104.1), lateral view. B, Corythosaurus casuarius (AMNH 5338), lateral view.
C, Edmontosaurus annectens (SM R4036), lateral view. D, Maiasaura peeblesorum (ROM 44770),
lateral view. Abbreviations: j = jugal; lc = lacrimal; mx = maxilla; ns = nasal; pmx = premaxilla.
470
Table D.6. Character 105. Ratio (R) between the length of the ectopterygoid shelf and the
length of the alveolar margin of the maxilla for a sample of iguanodontoidean
ornithopods. Included is the corresponding character coding derived from K-Means

Cluster (State)
AEHM1/12 Amurosaurus riabinini 0.45 2
IRSNB cast Bactrosaurus johnsoni 0.31 1
BMNH R5764 Iguanodon atherfieldensis. 0.18 0
471

Cluster (State)
MPCA SM-10 Salitral Moreno hadrosauridae 0.49 2
AMNH 5221 Saurolophus orborni 0.46 2
MACN-RN2 Secernosaurus koerneri 0.46 2
BMNH R4911 cf. Telmatosaurus 0.40 2
472
0.50
0.45
Ratio of the Length of the Ectopterygoid Shelf
0.40
120
110
0.35
100
BIC
90
0.30
80
70
E V
0.25
2 4 6 8
Number of Clusters
0.20
Am Ba Br Coi Cos Ed Edr Gi Gr Grl Grm Hy Hys Ig Iga Krn Lml Ma Ou Pat Ph Pr Pt Sao Se Ts cfTe lc sal x
Taxon
Figure D.55. Character 105. Box plot showing the distribution of the relative length of the ectopterygoid shelf. The inserted smaller
graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph to
the lower right corner shows how the two variables of this raio were measured, exemplified in a maxilla of Maiasaura peeblesorum (OTM F138).
473
Table D.7. Character 106. Angle (A) between the ectopterygoid shelf and the caudal
segment of the tooth row of the maxilla for a sample of iguanodontoidean ornithopods.

Cluster (State)
IRSNB SBDE/95E Bactrosaurus johnsoni 27 0
SBDE 95/E5 Bactrosaurus johnsoni 32 0
ZPAL MgD-I 162 cf. Saurolophus angustirostris 0 3
MOR549-6-19-8-9 Hypacrosaurus stebingeri 1 3
IPS-SRA 22 Pararhabdodon isonensis 0 3
474

Cluster (State)
AMNH 5221 Saurolophus orborni 0 3
MACN-RN 2 Secernosaurus koerneri 16 1
BMNH R4911 cf. Telmatosaurus transsylvanicus 14 1
FGGUB R1010 cf. Telmatosaurus transsylvanicus 13 1
AMNH 5472 Hadrosauridae 0 3
UMNH 12265 Hadrosauridae 0 3
TMM 42452 Hadrosauridae 14 1
475
30
Angle of Inclination of the Ectopterygoid Shelf
25
−320
−340
20
BIC
−360
−380
15
−400
E V
2 4 6 8
Number of Clusters
10
5
0
Am Ba Br Coi Ed Eda Edr Gi Gr Grl Grm Hy Hys Iga Lml Ma Ou Pat Ph Pr Pt Sao Se Te Ts cfSaa cfTe x
Taxon
Figure D.56. Character 106. Boxplot showing the distribution of the angle of inclination of the ectopterygoid shelf. The inserted smaller
graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph shows
how this angle was measured, exemplified in a maxilla of Brachylophosaurus canadensis (MOR1071-7-6-98-79).
476
state 0 state 1
A B
state 1 state 2
C D
Figure D.57. Character 107. Morphology of the lateral emargination (arrows) of the ectopterygoid
shelf. A, Iguanodon atherfieldensis (BMNH R5764), lateral view. B, Protohadros byrdi (SMU 74582),
lateral view. C, cf. Telmatosaurus transsylvanicus (BMNH R4911), lateral view. D, Edmontosaurus regalis
477
state 0
A
state 1
Figure D.58. Character 108. Position of the arcuate row of special foramina on the medial side of
the dentary. A, Iguanodon atherfieldensis (BMNH R5764), lateral view. B, Brachylophosaurus canadensis
(MOR 1071-8-13-98-559), lateral view.
478
Figure D.59. Character 109. Morphology of the lacrimal.. A, Maiasaura peeblesorum (ROM 44770),
lateral view. B, Lambeosaurus lambei (ROM 1218), lateral view. Abbreviations: lc = lacrimal.
479
state
state00
A
state 1
Figure D.60. Character 110. Absence (A) and presence (B) of a prominent convexity (arrow),
rostral to the jugal notch, in the ventral margin of the lacrimal. A, Edmontosaurus annectens
(SM R4036), lateral view. B, Gryposaurus monumentensis (RAM 6797), lateral view.
480
state 0 state 0
A B
state 1 state 1
C D
Figure D.61. Character 111. Morphology of the apex (arrow) of the rostral process of the jugal.
A, cf. Bactrosaurus johnsoni (MAL 0329-002), lateral view. B, Maiasaura peeblesorum
(ROM 44770), lateral view. C, Edmontosaurus annectens (SM R4036), lateral view. D, Gryposaurus
notabilis (ROM 873), lateral view.
481
state 2 state 2
A B
state 3 state 4
C D
Figure D.62. Character 111. Morphology of the apex (arrow) of the rostral process of the jugal.
A, Corythosaurus intermedius (ROM 777), lateral view. B, Lambeosaurus lambei (ROM 1218),
lateral view. C, Parasaurolophus tubicn (NMMNH P-25100), lateral view. D, Hypacrosaurus
altispinus (ROM 789), lateral view.
482
state 0
A
state 1
Figure D.63. Character 112. Dorsoventral expansion of the caudodorsal margin of the rostral
process of the jugal (rectangles). A, Equijubus normani (IVPP V12534), lateral view.
B, Lambeosaurus lambei (CMN 2869), lateral view.
483
A B C
Figure D.64. Character 113. Morphology of the triangular caudoventral margin of the rostral process of the jugal (rectangles).
A, Prosaurolophus maximus (TMP 84.1.1), lateral view. B, Brachylophosaurus canadensis (TMP 90.104.1), lateral view.
C, Corythosaurus casuarius (ROM 1933), lateral view.
484
state 0 state 0
A B
state 0 state 1
C D
Figure D.65. Character 114. Location of the caudoventral apex of the rostral process relative to
the caudodorsal articulation with the lacrimal (circles) (having as reference the longitudinal axis of the
rostral process oriented horizontally). A, Edmontosaurus sp. (BMNH R8297), lateral view.
B, Gryposaurus notabilis (ROM 764), lateral view. C, Prosaurolophus maximus (ROM 1928), lateral
view. D, Brachylophosaurus canadensis (TMP 90.104.1), lateral view.
485
state 0
A
state 1
Figure D.66. Character 115. Orientation of the medial articular surface (rectangles) of the rostral
process of the jugal. A, Tanius sinensis (PMU R240), medial view. B, Edmontosaurus sp. (MOR 601),
medial view.
486
state 0
state 1
Figure D.67. Character 116. Inclination of the bony rim that bounds caudally the medial articulation
surface of the rostral process of the jugal (this inclination is considered relative to the rostrocaudal longer
axis of the jugal). A, Maiasaura peeblesorum (OTM F138), medial view. B, Prosaurolophus maximus
(MOR 454-6-24-6-2), medial view.
487
state 0
state 1
Figure D.68. Character 117. Rostrocaudal width of the curvature of the ventral margin of the jugal,
between the caudoventral flange and the rostral process. A, Prosaurolophus maximus (MOR 447-7-6-7-1),
lateral view. B, Charonosaurus jiayinensis (CUST unrecorded number), lateral view.
488
Table D.8. Character 118. Ratio (R) between the dorsoventral depth of the caudoventral
flange and the minimum depth of the caudal constriction for a sample of
iguanodontoidean ornithopods. Included is the corresponding character state coding

Cluster (State)
MOR 1071-7-16-98-248G Brachylophosaurus canadensis 1.64 2
MOR 1071-7-31-99-281-O Brachylophosaurus canadensis 1.74 2
UCMP 130139 Brachylophosaurus canadensis 1.74 2
MAL 0329-002 cf. Bactrosaurus johnsoni 1.68 2
IVPP V830 cf. Tsintaosaurus spinorhinus 1.54 1
CUST unrecorded no. Charonosaurus jianyinensis 1.22 0
CMN 8399 Edmontosaurus annectens 1.20 0
489

Cluster (State)
LACM 23502 Edmontosaurus regalis 1.22 0
AMNH 5254 Edmontosaurus regalis 1.25 0
FMNH P15004 Edmontosaurus regalis 1.29 0
TMP 80.22.1 Gryposaurus notabilis 1.40 1
TMP 82.10.1 Hypacrosaurus altispinus 1.34 0
AMNH 5799 Kritosaurus navajovius 1.70 2
AMNH 5373 Lambeosaurus lambei 1.30 0
YPM 3222 Lambeosaurus lambei 1.42 1
490

Cluster (State)
TMM 41262-1 Prosaurolophus maximus 1.39 1
PMU R240 Tanius sinensis 1.22 0
CPC-59 Velafrons cohauilensis 1.40 1
TMM 42452-1 Hadrosauridae 1.31 0
491
100
1.7
80
Ratio of the Expansion of the Ventral Jugal Flange
BIC
60
40
1.6
20
2 4 6 8
Number of Clusters
1.5
1.4
1.3
1.2
Am Ba Br Ch Coi Cos Ed Eda Edr Eo Eq Grm Grn Hy Iga Jz Krn Lml Lmm Ma Ol Pat Paw Pr Pt Saa Sao Ta Ve cfBa cfTs x
Taxon
Figure D.69. Character 118. Boxplot showing the distribution of the ratio between the dorsoventral depth of the flange and the minimum
depth of the caudal constriction of the jugal) . The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of
clusters in which the data can be classified. The photograph how the two variables of this raio were measured, exemplified in a jugal of
Brachylophosaurus canadensis (MOR1071-7-6-98-79).
492
Figure D.70. Character 119. Morphology of the quadratojugal flange of the jugal.
A, Equijubus normani (IVPP V12534), lateral view. B, Corythosaurus intermedius (ROM 845),
lateral view. C, Brachylophosaurus canadensis (MOR 794), lateral view. D, Prosaurolophus
maxmimus (MOR 447-7-6-7-1), lateral view.
493
state 0 state 2
A B C D E F G H
state 1
I J K L M N Ñ O P
Q R S T U V W X Y Z
state 3
AA BB CC DD EE FF GG HH II
JJ KK LL MM NN ÑÑ OO PP QQ
Figure D.71. Character 119. Lateral outlines of the quadratojugal flange of the jugal. A, Equijubus normani (IVPP V12534). B, Iguanodon
atherfieldensis (BMNH R5764). C, Protohadros byrdi (SMU 74582). D, Brachylophosaurus canadensis (MOR 794). E, B. canadensis
(MOR 1071-7-27-98-453). F, B. canadensis (MOR 1071-7-31-99-281-O). . G, Maiasaura peeblesorum (ROM 44770). H, B. canadensis (CMN 8893).
I, Bactrosaurus johnsoni (AMNH6379). J, B. johnsoni(AMNH 6396). K, Amurosaurus riabinini (AEHM 1/12). L, Corythosaurus casuarius
(AMNH 5240). M, C. casuarius (AMNH 5382). N, C. intermedius (CMN 8704) Ñ, C. intermedius (ROM 777). O, C. intermedius (ROM 845).
P, C. casuarius (ROM 868). Q, Corythosaurus sp. (CMN 34845). R, Corythosaurus sp. (ROM 1947). S, H. altispinus (CMN 8501). T, H. stebingeri
(AMNH 5461). U, Lambeosaurus lambei (CMN 8703). V, L. lambei (AMNH 5373). W, L. lambei (FMNH PR380). X, L. lambei (AMNH 5340).
Y, L. lambei (CMN 8633). Z, Velafrons coahuilensis (CPC-59). AA, Edmontosaurus regalis (AMNH 5254). BB, Edmontosaurus sp. (BMNH R8297).
CC, E. regalis (CM 26258. DD, E. regalis (CMN 2288). EE, E. regalis (FMNH P15004). FF, Prosaurolophus maximus (MOR 447). GG, E. regalis
(MOR 801). HH, E. annectens (CMN 8509). II, E. annectens (UCM 20892). JJ, Edmontosaurus sp. (AMNH 5046). KK, Gryposaurus notabilis
(TMP 80.22.1). LL, G. notabilis (MSNM V345). MM, P. maximus (CMN 2870). NN, P. maximus (AMNH 5386). ÑÑ, P. maximus (ROM 787).
OO, P. maximus (TMM 41262-1). PP, P. maximus (ROM 1928). QQ, S. osborni (AMNH 5220).
494
state 0 state 0
A B
state 1 state 1
C D
Figure D.72. Character 120. Morphology of the ventral margin (rectangles) located between the caudo-
ventral and quadratojugal flanges of the jugal. A, Hypacrosaurus altispinus (ROM 702), lateral view.
B, Velafrons coahuilensis (CPC-59), lateral view. C, Brachylophosaurus canadensis (MOR 1071-7-31-
99-281O), lateral view. D, Gryposaurus notabilis (ROM 873), lateral view.
495
Table D.9. Character 121. Ratio (R) between the minimum depth of the caudal and rostral
constrictions for the jugals sampled to study the variation of this character, with the

Cluster
IRSNB cast Bactrosaurus johnsoni 0.89 0
FMNH PR862 Brachylophosaurus canadensis 1.13 1
MOR 1071-7-31-99-281-O Brachylophosaurus canadensis 1.16 1
1071-7-16-98-248G Brachylophosaurus canadensis 1.30 1
MAL 0329-002 cf. Bactrosaurus johnsoni 0.93 0
MNA PI 529 cf. Parasaurolophus tubicen 1.30 1
Unrecorded no. cf. Parasaurolophus tubicen 1.35 2
496

Cluster
ZPAL MgD-I162 Saurolophus angustirostris 0.98 0
IVPP V830 cf. Tsintaosaurus spinorhinus 0.98 0
497
1.5
Relative Depth of the Jugal Rostral and Caudal Constrictions
1.4
1.3
1.2
1.1
30
20
10
1.0
BIC
0
−10
−20
0.9
2 4 6 8
Number of Clusters
Am Ba Br Ch Coi Cos Ed Eda Edr Eq Grn Hy Iga Krn Lml Ma Pat Paw Pr Pt Saa Sao Ta Ve cfBa cfPat cfTs
Taxon
Figure D.73. Character 121. Boxplot showing the distribution of the relative depth of the caudal and rostral constrictions of the adult jugal
in a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of
clusters in which the data can be classified. The photograph shows how the two variables of this raio were measured, exemplified in a jugal of
Prosaurolophus maximus (MOR 447-7-6-7-1).
498
0.9
0.8
Ratio of Jugal Robustness
0.7
0.6
80
60
0.5
BIC
40
20
0.4
E V
0
2 4 6 8
No. of Clusters
Am Ba Br Ch Coi Cos Ed Eda Edr Eq Grm Grn Hy Iga Igb Krn Lml Ma Ou Pat Paw Pr Pt Saa Sao Ta Ve cfBa cfPat cfTs
Taxon
Figure D.74. Character 122. Boxplot showing the distribution of the ratio between the minimum depth of the caudal constriction and
he distance between the point of maximum curvature of the infratemporal margin and the caudal margin of the lacrimal process .The inserted
smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph
at the lower right corner shows how the two variables of this raio were measured, exemplified in a jugal of Edmontosaurus sp. (MOR 601).
499
state 0
state 2
state 1
A C
Figure D.75. Character 123. Relative width of the orbital and infratemporal margins of the jugal. A, Velafrons coahuilensis (CPC-59),
lateral view. B, Maiasaura peeblesorum (OTM F138), lateral view. C, Prosaurolophus maximus (MOR 447-7-6-7-1), lateral view.
500
Table D.10. Character 124. Angle (A) of curvature of the caudal margin of the quadrate
of a sample of iguanodontoidean ornithopods. Included is also the corresponding
character state coding derived from K-Means clustering analysis.

Cluster (State)
FMNH PR682 Brachylophosaurus canadensis 156 0
MOR 1071-8-13-98-559D Brachylophosaurus canadensis 154 0
CMN 8509 Edmontosaurus annectens 156 0
ROM 867 Edmontosaurus annectens 168 0
SM R4036 Edmontosaurus annectens 163 0
AMNH 5254 Edmontosaurus regalis 153 0
BMNH R8297 Edmontosaurus regalis 155 0
CM 26258 Edmontosaurus regalis 154 0
501
Table Q1—continued.

Cluster (State)
AMNH 5353 Lambeosaurus lambei 146 1
TMP 81.37.1 Lambeosaurus lambei 145 1
TMP 66-04-01 Lambeosaurus magnicristatus 140 1
Unrecorded no. cf. Parasaurolophus tubicen 142 1
AMNH 5386 Prosaurolophus maximus 158 0
IVPP K68 Tsintaosaurus spinorhinus 145 1
CPC-59 Velafrons cahuilensis 139 1
502
170
Angle of Quadrate Curvature
160
150
−560
−580
140
BIC
−600
−620
2 4 6 8
Number of Clusters
130
Ba Br Coi Cos Ed Eda Edr Eq Grm Grn Hy Hys Iga Igb Lml Lmm Ma Ou Paw Pr Pt Sao Ts Ve cfPat
Taxon
Figure D.76. Character 124. Boxplot showing the distribution of the angle of curvature of the long axis of the quadrate. The inserted
smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which the data can be classified.
The photograph shows how this angle was measured, exemplified in a quadrate of Iguanodon atherfieldensis (BMNH 11521).
503
Table D.11. Character 125. Ratio (R) between the distance from the mid-length of the
notch to the quadrate head and the total length of the element, for the quadrates sampled
to study the variation of this character Included is the corresponding character state
coding derived from K-Means clustering analysis.

Cluster (State)
IRSNB SBDE 95/E5 Bactrosaurus johnsoni 0.64 1
MOR 1071-8-13-98-559D Brachylophosaurus canadensis 0.64 1
USNM 4807 Edmontosaurus sp. 0.61 1
UCMP 128372 V76142 Edmontosaurus annectens 0.69 1
DMNH 1943 Edmontosaurus annectens 0.63 1
BMNH R8297 Edmontosaurus regalis 0.66 1
504

Cluster (State)
TMP 80-22-1 Gryposaurus notabilis 0.66 1
TMP 66-04-01 Lambeosaurus magnicristatus 0.54 0
AMNH 5386 Prosaurolophus maximus 0.62 1
505

Cluster (State)
IVPP K68 Tsintaosaurus spinorhinus 0.56 0
506
220
Ratio of the Position of the Quadratojugal Notch
200
0.65
BIC
180
160
2 4 6 8
0.60
Number of Clusters
0.55
0.50
Ba Br Coi Cos Ed Eda Edr Eq Gi Grm Grn Hy Hys Iga Igb Lml Lmm Ma Ou Paw Pr Pt Sao Ts Ve
Taxon
Figure D.77. Character 125. Boxplot showing the distribution of the ratio of the position of the quadratojugal notch of the quadrate. The
inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which the data can be classified. The
photograph shows how the two variables of this raio were measured, exemplified in a quadrate of Prosaurolophus maximus (MOR 454-6-24-6-2).
507
Table D.12. Character 126. Angle (A) formed by the rostral margin of the dorsal half of
the quadratojugal notch with the caudal margin of the quadrate for a sample of
derived from K-Means clustering.

(State)
IRSBN SBDE 95/E5 Bactrosaurus johnsoni 31 1
FMNH PR862 Brachylophosaurus canadensis 23 1
MOR 1071-8-13-98-559D Brachylophosaurus canadensis 28 1
TMP 90-104.1 Brachylophosaurus canadensis 25 1
USNM 8629 cf. Kritosaurus sp. 24 1
Unrecorded no. cf. Parasaurolophus tubicen 49 0
UCMP fldAK83-V242 Edmontosaurus sp. 28 1
TMP 80.22.1 Gryposaurus notabilis 25 1
BMNN R11521 Iguanodon atherfieldensis 52 0
NMMNH P-16106 Kritosaurus navajovius 31 1
508

(State)
MOR 241 Hadrosauridae 30 1
509
55
Orientation Angle of the Dorsal Margin of the Quadratojugal Notch
50
−300
45
−320
BIC
40
−340
−360
E V
35
2 4 6 8
Number of Clusters
30
25
Ba Br Coi Cos Ed Eda Edr Eq Grm Grn Hys Iga Krn Lml Ma Ou Pr Pt Sao Ts cfKr cfPat x
Taxon
Figure D.78. Character 126. Boxplot showing the distribution of the angle of inclination of the dorsal margin of the quadrate notch of the
quadrate. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which the data can be
classified. The photograph shows how the two variables of this raio were measured, exemplified in a quadrate of Prosaurolophus maximus
(TMP 98.50.1).
510
A B C
Figure D.79. Character 127. Morphology of the lateral profile of the quadratojugal notch of the quadrate. A, Hypacrosaurus stebingeri (MOR 455),
lateral view. B, Tsintaosaurus spinorhinus (IVPP V725), lateral view. C, Edmontosaurus sp. (ROM 658), lateral view.
511
state 0
A
state 1
Figure D.80. Character 128. Development of the squamosal buttress on the caudal margin of the
dorsal end of the of the quadrate. A, Prosaurolophus maximus (CMN 2870), lateral view. B, Gryposaurus
notabilis (TMP 80.22.1), lateral view.
512
Table D.13. Character 129. Ratio (R) between the rostrocaudal width of the lateral
condyle and the mediolateral width of the ventral end of the quadrate for a sample of
iguanodontoidean specimens. Included is also the character state coding derived from K-
Means clustering.

Cluster (State)
SBDE 95/E5 Bactrosaurus johnsoni 0.65 0
MOR 1071-8-13-98-559D Brachylophosaurus canadensis 0.90 1
Unrecorded no. cf. Parasaurolophus tubicen 1.00 1
MOR 399 Gryposaurus sp. 0.93 1
MOR 241 Hadrosauridae 0.85 1
ZPAL MgD-I162 Saurolophus angustirostris 0.93 1
IVPP K68 Tsintaosaurus spinorhinus 0.82 1
PASAC-1 Hadrosauridae 0.83 1
513
1.0
Width/Length Ratio of the Distal Surface of the Quadrate
0.9
0.8
20
0.7
10
BIC
0
0.6
−10
2 4 6 8
Number of Clusters
0.5
Ba Br Ed Edr Eq Gi Gr Hy Hys Iga Ou Pt Saa Sao Te Ts Ve cfLm cfPat x
Taxon
Figure D.81. Character 129. Boxplot showing the distribution of the width/length ratio of the distal surface of the quadrate. The inserted
smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph
shows how the two variables of this raio were measured, exemplified in a quadrate of Edmontosaurus sp. (ROM 658).
514
state 0
A
state 1
Figure D.82. Character 129. Degree of ventral offset of the lateral condyle of the distal end of
the quadrate relative to the medial condyle (correlated to the width/length ratio of the distal surface of
the quadrate). A, Bactrosaurus johnsoni (SBDE 95/E5), lateral view. B, Brachylophosaurus canadensis
(MOR 1071-8-13-98-559D), lateral view.
515
state 0 state 0
A B
state 1 state 2
C D
Figure D.83. Character 130. Development of a prefrontal crest (arrows). A, Edmontosaurus sp.
(AMNH 427), lateral view. B, dorsal view of the same specimen. C, Jaxartosaurus aralensis (PIN 1/50009),
dorsolateral view. D, Corythosaurus casuarius (ROM 871), lateral view.
516
state 2 state 2
A B
state 2
Figure D.84. Character 130. Presence of a well-developed prefrontal crest (arrows). A, Hypacrosaurus
altispinus (ROM 702), lateral view. B, dorsal view of the same specimen. C, H. stebingeri (MOR 553S-7-
27-2-93), dorsal view.
517
state 0
A
state 1
Figure D.85. Character 131. Lateral profile of the rostrodorsal margin (rectangles) of the prefrontal.
A, Lambeosaurus lambei (ROM 1218), lateral view. B, Kritosaurus navajovius (NMMNH P-16106),
lateral view.
518
state 0
A
state 1
Figure D.86. Character 132. Mediolateral breadth of the exposed rostroventral region (rectangles)
of the prefrontal. A, Edmontosaurus annectens (SM R4036), lateral view. A, Lambeosaurus lambei
519
A
Figure D.87. Character 133. State 0: exclusion of the prefrontal from the circumnarial fossa.
A, Gryposaurus notabilis (ROM 873),lateral view. B, G. notabilis (AMNH 5350), rostrodorsolateral view.
Abbreviations: bcf = boundary of the circumnarial fossa; ns = nasal; pf = prefrontal.
520
A B
Figure D.88. Character 133. State 1: inclusion of the prefrontal in the circumnarial fossa, exemplified in a skull roof of Maiasaura peeblesorum
(OTM F138). A, rostrodorsolateral view. B, right lateral view. Abbreviations: bcf = boundary of the circumnarial fossa; ns = nasal; pf = prefrontal.
521
state 0
A
state 1
Figure D.89. Character 134. Absence and presence of eversion (arrow) of the rostrodorsal
margin of the prefrontal. A, Gryposaurus notabilis (ROM 873), lateral view. A, Brachylophosaurus
canadensis (MOR 794), lateral view.
522
state 0
A
state 1
B
state 2
Figure D.90. Character 135. Exposure of the prefrontal-nasal contact. A, Brachylophosaurus canadensis
(MOR 794), lateral view. B, Lambeosaurus lambei (CMN 2869), lateral view. C, Corythosaurus casuarius
(ROM 871), lateral view. Abbreviations: ns = nasal; pf = prefrontal; pmx = premaxilla.
523
state 0 state 0
A B
state 0 state 1
C D
Figure D.91. Character 136. Absence and presence of a dorsal promontorium (arrow) on the rostral
process of the postorbital. A, Iguanodon atherfieldensis (BMNH R11521), lateral view. B, Brachylophosaurus
canadensis (TMP 90.104.1), lateral view. C, Gryposaurus notabilis (RAM 6797), lateral view. D, Saurolophus
osborni (AMNH 5220), lateral view.
524
state 1 state 1
A B
state 1 state 1
C D
Figure D.92. Character 136.Presence of a dorsal promontorium (arrows) on the rostral process
of the postorbital. A, Charonosaurus jiayinensis (CUST JIV1223), lateral view. B, Parasaurolophus
cyrtocristatus (UCMP 143270), lateral view. C, Lambeosaurus lambei (YPM 3222), lateral view.
D, L. lambei (CMN 2869), lateral view.
525
state 0
A
state 1
B
state 2
Figure D.93. Character 137. Rostrocaudal constriction of the dorsal region of the infratemporal
fenestra (rectangle) and length of the caudal process of the postorbital. A, Corythosaurus casuarius
(ROM 776), lateral view. B, Brachylophosaurus canadensis (cast of MOR 794). C, Parasaurolophus
tubicen (NMMNH P-25100), lateral view.
526
Figure D.94. Character 138. Morphology of the central body of the postorbital (arrows and rectangle). A, Bactrosaurus johnsoni
(AMNH 6365),lateral view. B, Equijubus normani (IVPP V12534), lateral view. C, Iguanodon bernissartensis (IRSNB 1731), lateral view.
D, Amurosaurus riabinini (AEHM 1/12), lateral view. E, Edmontosaurus regalis (CMN 2288), lateral view. F, rostral view of the same specimen.
527
state 0
A
state 1
Figure D.95. Character 139. Length of the jugal process (rectangles) of the postorbital.
A, Corythosaurus casuarius (AMNH 5338), lateral view. B, Brachylophosaurus canadensis
528
state 0
A
state 1
Figure D.96. Character 140. Morphology of the caudal end of the caudal process of the postorbital
at its articulation with the squamosal A, Jinzhousaurus yangi (IVPP V12691), dorsolateral
view. B, Bactrosaurus johnsoni (PIN 2549-1), dorsolateral view.
529
state 0
A
state 1
B
state 2
Figure D.97. Character 141. Caudal extension of the caudal ramus of the postorbital (rectangles)
overlapping the laterodorsal surface of the squamosal A, Gryposaurus notabilis (CMN 2278), lateral
view. B, Brachylophosaurus canadensis (CMN 8893), lateral view. C, Parasaurolophus cyrtocristatus
(UCMP 143270), lateral view. Abbreviations: po = postorbital; qc = quadrate cotylus; sq = squamosal.
530
Table D.14. Character 142. Ratio (R) between prequadratic process length and the
quadrate cotylus with of the squamosal, with the corresponding character state coding

(State)
PIN 1/5009 Jaxartosaurus aralensis 1.32 2
FMNH P27383 Lophorhothon atopus 1.33 2
UCMP 143270 Parasaurolophus cyrtocristatus 1.36 2
PMU R222 Parasaurolophus tubicen 1.34 2
531
Ratio of the Length of the Precotyloid Process of the Quadrate
1.4
1.2
0
−10
BIC
−20
−30
1.0
−40
2 4 6 8
Number of Clusters
0.8
Am Br Coi Cos Eda Edr Grn Hys Jx Krn Lh Lml Ma Pac Pat Paw Pr Sao
Taxon
Figure D.98. Character 142. Boxplot showing the distribution of the ratio of the length of the precotyloid process of the squamosal for
a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of
clusters in which the data can be classified. The photograph shows how the two variables of this raio were measured, exemplified in a squamosal
of Parasaurolophus walkeri (ROM 768).
532
state 0 state 1
A B
Figure D.99. Character 143. Dorsoventral expansion of the caudolateral surface (rectangles) of the squamosal. A, Brachylophosaurus
canadensis view (CMN 8893). B, Hypacrosaurus stebingeri (MOR 455).
533
state 0 state 0
A B
state 1 state 2
C D
Figure D.100. Character 144. Presence (arrows) and absence (rectangle) of separation of the
squamosals at the occipital margin of the skull roof. A, Kritosaurus navajovius (NMMNHP-16106),
dorsal view. B, Tsintaosaurus spinorhinus (IVPP V725), dorsal view. C, Brachylophosaurus canadensis
(MOR 1071-7-7-98-86), dorsal view. D, Hypacrosaurus altispinus (CMN 8501), dorsal view.
534
state 0 state 1
A B
Figure D.101. Character 145. Presence and absence of rostromedial indenture of the medial ramus of the squamosal . A, Bactrosaurus
johnsoni (AMNH 6365), dorsal view. B, Maiasaura peeblesorum (OTM F138), dorsal view.
535
APPENDIX E
NEUROCRANIAL CHARACTERS
Frontal
Character 146: bifurcation of the rostromedial margin of the frontals at the sagittal
plane of the skull roof, leaving a V-shaped space in between (Fig. E.1)
(0): absent
(1): present
Character 147: frontal fontanelle, present at least at one stage during ontogeny (Fig.
E.2; Langston, 1960; Wagner, 2001)
(0): absent
(1): present
Character 148: nasal articulation surface of the frontal shaped into a
rostroventrally-slopping platform (Fig. E.3; Godefroit et al., 2004b, character 7, in
part; Evans and Reisz, 2007, character 40, in part)
(0): absent
(1): present
Character 149: nasal articulation surface of the frontal shaped into a dorsoventrally
thickened, tongue-like platform that projects caudodorsally to overhang the parietal
in adults (Fig. E.4; Godefroit et al., 2004b, character 7, in part; Evans and Reisz,
2007, character 40, in part, and character 41)
(0): absent
(1): present
Comments: that by separating F4 from F3 I don’t assume that elevated platform
could be derivced from the rostral platform. I consider them as two separate structures
Character 150: median cleft separating the two striated tongues of the frontal
platform (Fig. E.5; Evans and Reisz, 2007, character 40, in part)
(0): absent
(1): present
Character 151: exposure of the frontal along the dorsal margin of the orbit (Fig.
536
E.6; modified from Horner et al., 2004, character 57)
(0): frontal exposed, forming part of the dorsal orbital margin in between the
prefrontal and postorbital
(1): the frontal projects laterally into a triangular narrow apex that inserts in
between a narrow gap left by the prefrontal and postorbital (in some exemplars, although
the prefrontal and postorbital do not meet at the orbital margin, the apex of the frontal
does not reach the lateral margin of the orbit at the same level than the former two
elements)
(2): frontal completely excluded from the orbital margin by an extensive
articulation between the prefrontal and postorbital;
Character 152: frontal upward doming dorsal to the braincase of subadult (and
perhaps young adult) specimens (Fig. E.7; modified after Horner et al., 2004,
character 58)
(0): absent
(1): present
Comments: this character appears to be absent in the largest adult specimens, as
pointed out by Godefroit et al. (2008). An example is the relatively large braincase ROM
1940. The character is retained here because subadult hadrosaurines do not have a dome.
Character 153: length/width ratio of the ectocranial surface of the frontal (Fig. E.8;
Evans and Reisz, 2007, character 42)
(0): relatively elongated ectocranial surface, with a ratio greater than 0.8
(1): relatively short ectocranial surface, with a ratio of 0.8 or less, but greater than
0.4
(2): greatly shortened ectocranial surface, with a ratio less than 0.4
Comments: this ratio was measured by Evans (2007) between the lengths of the
external interfrontal suture caudal to the base of a crest (if present) over the maximum
transverse width of the frontal perpendicular to the sagittal axis of the skull.
Character 154: morphology of the ventral annular ridge that defines the rostral
extent of the cerebral fossa (Fig. E.9; Evans and Resiz, 2007, character 43)
(0): long, low and gently rounded in medial view
(1): sharp annular ridge
537
Parietal
Character 155: maximum length/minimum width proportions of the adult parietal

(Table E.1; Fig. E.10; modified from Godefroit et al., 2004, character 18)
(0): very short, length/width ratio less than 1.40 (sample mean ratio of 1.19)
(1): short, ratio between 1.40 and 2.35 (sample mean ratio of 1.98)
(2): relatively long, ratio greater than 2.35 (sample mean ratio of 2.75)
Comments: This ratio decreases during ontogeny, at least in lambeosaurines such
as Lambeosaurus spp. For example, the ratio is 1.84 in ROM 758, 1.69 in the larger
CMN 8633 and finally reaches 1.16 in ROM 794. Similarly, the subadult Corythosaurus
ROM 1947 has a long parietal (ratio of 1.90). However, the caudal inclination of the
parasagittal crest is consistent through ontogeny. In other hadrosaurids, the parietal
appeared to lengthen with ontogeny. For example, in Brachylophosaurus MOR 1071-7-
13-99-87-I the ratio is 2.04 and increases to 2.50 in the adult specimen UCMP 130139.
Horner et al. (2004; character 70) and Evans and Reisz (2007; character 45)
include the length of the parietal sagittal crest as a character. This character is not
considered here because it is linked to the length of the entire parietal: shorter parietals
have shorter sagittal crests and vice versa. Likewise, no difference in the degree to which
the proximal region of the sagittal crest narrows was appreciated among the taxa under
study.
Character 156: orientation of the parietal midline crest (Fig. E.11; Horner et al.,
2004; character 69; Evans and Resiz, 2007, character 44)
(0): straight and level with the skull roof or slightly down-warped along its length
(1): the sagittal crest deepens caudally and is strongly down-warped
Character 157: morphology of the rostromedian process of the parietal that forms a
crenulated suture in between the caudomedian margin of the frontals (Fig. E.12)
(0): rectangular, rostrocaudally short and mediolaterally expanded
(1): rostrocaudally short and subtriangular to arcuate or absent
(2): rostrocaudally elongate and mediolaterally narrow
Character 158: rostral extension of the sagittal crest along the dorsal surface of the
parietal (Fig. E.13)
538
(0): the sagittal crest extends along the entire length of the parietal and remains
sharp and well defined at the rostral region
(1): the sagittal crest extends along the entire length of the parietal but its
sharpness fades away at the rostral region where the parietal is rostrocaudally shorter than
it is wide
(2): the sagittal crest only extends along the caudal half of the parietal and the
rostral half of the dorsal surface of the parietal is flattened, lacking any ridge or
mediolateral compression
Basioccipital
Character 159: participation of the basioccipital in the ventral margin of the

foramen magnum (Fig. E.14; Weishampel et al., 1993, character 24)
(0): absent, the exoccipital condyloids nearly or completely exclude the
basioccipital from the ventral margin of the foramen magnum
(1): present, the exoccipitals are separated at the sagittal plane of the braincase
and allow the basioccipital to become part of the ventral margin of the foramen magnum
Comments: exclusion of bo from foramen magnum is not syanpomrphic for
hadrosaurines; on the contrary, all hadrosaurids have bo participating in floor of foramen
magnum.
Character 160: orientation of the articulation margin of the occipital condyle
relative to the convex caudoventral border of the basioccipital (Fig. E.15; modified
from Godefroit et al., 2000, character 1)
(0): the articulation margin of the occipital condyle faces caudoventrally and is
continuous with the caudal border of the basioccipital
(1): the articulation margin of the occipital condyle faces caudally and is divided
from the caudal border of the basioccipital by a shallow cleft
Character 161: length of basioccipital constriction (Fig. E.16; Godefroit et al., 2001)
(0): relatively long and well developed
(1): poorly developed and relatively short constriction
539
Basisphenoid
Character 80 form Gates and Sampson (2007) was excluded because in all taxa
the basisphenoid was found contributing to the delimitation of the trigeminal foramen.
Character 162: orientation of the basipterygoid process of the basisphenoid
(measured as the angle between the ventral margins of both processes) (Table E.2;
Fig. E.17; modified from Godefroit et al., 2001, character 2)
(0): angle less than 100º (sample mean angle of 79º)
Comments: is substantial variation in the degree of rostral orientation of the
basispyerygoid proceses. E.g., Hy ROM 703 are rostrolaterally directed buy laterally in
CMN 8765; in Cos CMN 8676 are laterally directed and rostrolaterally in Coi ROM 776.
Character 163: developement of the alar process of the basisphenoid (Fig. E.18)
(0): moderately developed
(1): very well developed, relatively large in size
Character 164: ventral transverse ridge between the basipterygoid processes of the
basisphenoid (Fig. E.19; Gates and Sampson, 2007, character 78)
(0): absent or very poorly developed
(1): present, sharply defined ridge
Character 165: short median ventral process located between the basipterygoid
processes of the basisphenoid (Fig. E.20; Gates and Sampson, 2007, character 79)
(0): absent
(1): present, ventrally or rostroventrally inclined
Character 166: development of the rostral constriction of the basisphenoid, caudal
to the basipterygoid processes (measured as the ratio between the minimum
mediolateral width of the rostral constriction and the maximum width of the
basisphenoid across the spheno-occipital tubercles) (Table E.3; Fig. E.21)
(0): very thick constriction, ratio less than 1.45 (sample mean ratio of 1.37)
(1): moderately developed constriction, ratio between 1.45 and 1.90 (sample mean
ratio of 1.72)
(2): very thin constriction, ratio greater than 1.90 (sample mean ratio of 2.25).
540
Laterosphenoid
Character 167: complete lateral osseous closure of the ophthalmic sulcus (V1) of the
laterosphenoid (Fig. E.22; Evans and Reisz, 2007, character 51)
(0): absent
(1): present
Character 168: extreme reduction of the length of the postorbital process of the
laterosphenoid to 25% or less the length of the mediodorsal flange of this element
(Fig. E.23; modified from Prieto-Marquez et al., 2006, character 76)
(0): absent
(1): present
Supraoccipital
Character 169: orientation of the caudal surface of the supraoccipital (Fig. E.24;
(0): nearly vertical
(1): rostrally inclined and facing caudodorsally
Character 170: lateroventral corner of the supraoccipital deeply inset into the
exoccipital, so that the latter is “locked” between two short flanges that project
medially above lateral end of the supraoccipital-exoccipital contact (Fig. E.25;
(0): absent
(1): present
Exoccipital-Opisthotic
Character 171: caudal extension of the exoccipital-supraoccipital shelf above the

foramen magnum (Fig. E.26; modified from Godefroit et al., 2004b, character 24)
(0): very short rostrocaudal length, approximately less than half the diametyer of
541
the foramen magnum
(1): moderately long, approximately more than half but less than the diameter of
the foramen magnum
(2): very long, substantially longer (often twice or more) than the diameter of the
foramen magnum
Character 172: orientation of the distal portion of the paroccipital process (Fig.
E.27; Horner et al., 2004, character 62)
(0): ventrally directed
(1): rostroventrally directed
Character 173: elevation of the proximodorsal region of the quadrate wing of the
pterygoid (Fig. E.28; Prieto-Marquez et al., 2006, character 72)
(0): absent
(1): present
Character 174: ventral extension of the lamina located ventral to the central
buttress of the pterygoid (Fig. E.29; Prieto-Marquez et al., 2006, character 75)
(0): lamina of moderate size, a relatively large portion of the ventral quadrate
process and the rostroventral process extends beyond the ventral margin of the lamina;
(1): extensive lamina, only a relatively small portion of the ventral quadrate
process and the rostroventral process extends beyond the ventral margin of the lamina.
Comments: this character was only applied to adults only, since in juvenile
Edmontosaurus from Alaska the lamina is not particularly extended. This character is
here regarded as synapomorphic for species of Edmontosaurus. The straight versus
arcuate morphology observed in the ventral margin of the lamina was not considered
because it showed intraspecific variation (e.g., compare Brachylophosaurus specimens
MOR 1071-8-1-99-315 and MOR 1071-7-23-98-387).
Character 175: ectopterygoid–jugal contact (Fig. E.30; Godefroit et al., 2001,
character 12)
(0): present, the ectopteryoid contacts the medial side of the jugal
(1): absent, the jugal lacks an articular facet for the ectopterygoid
542
state 0
A
state 0
B
state 1
C
state 1
D
state 1
Figure.E.1. Character 146. Absence and presence (rectangles) of bifurcating rostromedian region
of the frontals. All specimens in dorsal view. A, Tsintaosaurus spinorhinus (IVPP V725). B, Maiasaura
peeblesorum (YPM-PU 22405). C, Tanius sinensis (PMU R240). D, Secernosaurus koerneri (MACN-RN
144). E, Hypacrosaurus altispinus (CMN 8675). Abbreviations: f =frontal; ns = nasal; po = postorbital.
543
state 0
A
state 1
Figure E.2. Character 147. Absence and presence (rectangle) of frontal fontanelle. A, Prosaurolophus
maximus (CMN 2870). B, Lophoehothon atopus (FMNH P27383). Abbreviations: f =frontal; ns = nasal.
544
A
Figure E.3. Character 148. Presence of a frontal platform. A, Amurosaurus riabinini

(cast of AEHM 1/12). B, Brachylophosaurua canadensis (MOR 1071-7-13-99-87-I, subadult
specimen). Abbreviation: fp = frontal platform.
545
A
Figure E.4. Character 149. Presence of a frontal platform. A, Charonosaurus jiayinensis

(CUST JV-12512). B, Parasaurolophus tubicen (PMU R222). C, Saurolophus osborni (AMNH 5221).
Abbreviation: fp = frontal platform.
546
state 1 state 0
B
state 1
A C
Figure E.5. Character 150. Absence and presence (arrows) of median cleft in the frontal platform. A, Saurolophus osborni (AMNH 5221),
dorsal view. B, Parasaurolophus tubicen PMU R222), dorsal view. C, Hypacrosaurus altispinus (CMN 8675), dorsal view.
Abbreviation: fp = frontal platform.
547
A B C
Figure E.6. Character 151. Absence and degree of exposure (rectangles) of the frontal in the dorsal orbital margin. A, Edmontosaurus sp.
(YPM 618), dorsal view. B, Gryposaurus notabilis (ROM 764), dorsal view. C, cf. Corythosaurus sp. (AMNH 5433), dorsal view.
Abbreviations: f = frontal; pf = prefrontal; po = postorbital.
548
state 0
A
state 1
Figure E.7. Character 152. Absence and presence of frontal doming (rectangle). A, Gryposaurus
notabilis (AMNH 5350). B, Amurosaurus riabinini (cast of AEHM 1/12). Abbreviation: f = frontal.
549
state 0
A
state 1
B
state 2
Figure E.8. Character 153. Length/width proportions of the ectocranial surface (rectangles) of the
frontal.. A, Tanius sinensis (PMU R222), dorsal view. B, cf. Corythosaurus sp. (AMNH 5433), dorsal
view. C, Charonosaurus jiayinensis (CUST JV-12512), dorsal view. Abbreviation: f = frontal.
550
state 0 state 1
A B
Figure E.9. Character 154. Absence and presence of annular ridge. A, Brachylophosaurus canadensis (FMNH 862), ventral view.
B, cf. Corythosaurus sp. (USNM 11893), ventral view.
551
Table E.1. Character 155. Ratio (R) between the length and the width of the parietal for
the specimens sampled to study the variation of this character, with the corresponding
character state coding derived from K-Means cluster analysis.

Cluster (State)
CUST JV1251-57 Charonosaurus jiayinensis 2.03 1
AMNH 5433 cf. Corythosaurus sp. 1.16 0
AMNH 427 Edmontosaurus sp. 2.84 2
AMNH 5350 Gryposaurus notabilis 2.77 2
AMNH 5248 cf. Hypacrosaurus altispinus 1.23 0
BMNH 11521 Iguanodon atherfieldensis 2.03 1
AMNH 5799 Kritosaurus navajovius 2.20 1
552
3.0
Length/Width Ratio of the Parietal
−50
2.5
−60
BIC
−70
−80
E V
2.0
2 4 6 8
Number of Clusters
1.5
Am Ba Br Ch Ed Eda Edr Grn Hy Hys Iga Igb Krn Lml Ma Ou Pat Pr Saa Ta Ts cfCo cfHy
Taxon
Figure E.10. Character 155. Boxplot showing the distribution of the ratio of the length/width ratio of the parietal for a sample of
iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in
which the data can be classified. The photograph shows how the two variables of this raio were measured, exemplified in the parietal of Tanius
sinensis (PMU R240).
553
state 0
A
state 1
Figure E.11. Character 156. Orientation of the parietal midline crest (rectangles). A, Gryposaurus
notabilis (AMNH 5350). B, Hypacrosaurus stebingeri (MOR553S-7-27-2-93).
554
state 0
A
state 1
B
state 2
Figure E.12. Character 157. Morphology of the rostromedian process of the parietal (rectangle)
that forms a crenulated suture in between the caudomedian margin of the frontals. A, Tanius sinensis
(PMU R222), dorsal view. B, cf. Corythosaurus sp. (AMNH 5433), dorsal view. C, Edmontosaurus sp.
(AMNH 427), dorsal view.
555
state 0 state 0
A B
state 1 state 1
C D
state 2 state 2
E F
Figure E.13 Character 158. Rostral extension (rectangles) of the sagittal crest along the dorsal surface
of the parietal. Specimens shown in dorsal view. A, Edmontosaurus regalis (CMN 2289). B, Gryposaurus
notabilis (AMNH 5350). C, Hypacrosaurus stebingeri (MOR 553S-7-27-2-93). D, Corythosaurus intermedius
(ROM 776). E, Amurosaurus riabinini (AEHM 1/12). F, Charonosaurus jiayinensis (CUST JV1251-57).
556
state 0 state 1
A B
state 1 state 1
C D
Figure E.14. Character 159. Participation of the basioccipital in the ventral margin of the foramen
magnum. All specimens shown in caudal view. A, Equijubus normani (IVPP V12534). B, Gryposaurus
notabilis (AMNH 5350). C, Hypacrosaurus altispinus (CMN 2246, juvenile specimen). D, Corythosaurus
intermedius (CMN 8676). Abbreviations: bo = basioccipital; ex = exoccipital.
557
state 0
A
state 1
B
state 1
Figure E.15. Character 160. Orientation of the articulation margin of the occipital condyle relative to
the convex caudoventral border of the basioccipital. The arrows point to a shallow cleft dividing the
articulation margin of the occipital condyle from the caudal margin of the basioccipital. A, Ouranosaurus
nigeriensis (cast of GDF 300), lateral view. B, Bactrosaurus johnsoni (AMNH 6365), lateral view.
C, Amurosaurus riabinini (cast of AEHM 1/12), lateral view.
558
state 0 state 0
A B
state 1 state 1
C D
Figure E.16. Character 161. Length of the basioccipital constriction (arrows). A, Bactrosaurus johnsoni (AMNH 6365), ventral view.
B, Ouranosaurus nigeriensis (cast of GDF 300), ventral view. C, Gryposaurus notabilis (CMN 2289), ventral view. D, Charonosaurus
jiayinensis (CUST JV1251-57), ventral view.
559
Table E.2. Character 162. Angle (A) between the ventral margins of the pterygoid
processes for the braincases sampled to study the variation of this character, with the
corresponding character state coding derived from K-Means clustering analysis.

Cluster (State)
AEHM 1/232 Amurosaurus riabibini 64 0
IRSNB SBDE 95/E5 Bactrosaurus johnsoni 77 0
CUST JV1251-57 Charonosaurus jiayinensis 80 0
AMNH 5433 cf. Corythosaurus sp. 72 0
AMNH 427 Edmontosaurus sp. 108 1
UCMP 130156 Edmontosaurus sp. 128 1
YPM 618 Edmontosaurus sp. 110 1
MOR 702 Hypacrosaurus altispinus 88 0
ROM 667 cf. Prosaurolophus 107 1
PMU R222 Parasaurolophus tubicen 87 0
MOR 1155 Hadrosauridae 113 1
UCMP 143270 Parasaurolophus cyrtocristatus 91 0
MOR 553S-7-27-2-93 Hypacrosaurus stebingeri 107 1
560
130
120
Angle between Pterygoid Processes
110
−180
100
−190
BIC
−200
−210
90
−220
2 4 6 8
80
Number of Clusters
70
Am Ba Br Ch Ed Grm Hy Hys Ou Pat Te cfCo cfPr x
Taxon
Figure E.17. Character 162. Boxplot showing the distribution of the orientation of the basipterygoid process of the basisphenoid for a
sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers
of clusters in which the data can be classified. The photograph shows how the two variables of this raio were measured, exemplified in the
parietal of Charonosaurus jiayinensis (CUST JV1251-57).
561
state 0
A
state 1
Figure E.18. Character 163. Developement of the alar process (rectangles) of the basisphenoid.
A, Hypacrosaurus stebingeri (MOR 455). B, Brachylophosaurus canadensis (MOR 1071-7-7-98-86).
562
state 0
A
state 0
B
state 1
C
state 1
Figure E.19. Character 164. Absence and presence (arrows) of a ventral transverse ridge between the
basipterygoid processes of the basisphenoid. All specimens shown in ventral view. A, Amurosaurus
riabinini (cast of AEHM 1/12). B, Hypacrosaurus altispinus (CMN 8675). C, Parasaurolophus tubicen
(PMU R222). D, Edmontosaurus regalis (CMN 2289).
563
state 0 state 0
A B
state 1 state 1
C D
Figure E.20. Character 165. Absence and presence (rectangles) of a median ventral process between
the basipterygoid processes of the basisphenoid. All specimens shown in caudal view. A, Corythosaurus
intermedius (ROM 776). B, Hypacrosaurus stebingeri (MOR 455). C, H. altispinus (ROM 702).
D, Ouranosaurus nigeriensis (cast of GDF 300).
564
Table E.3. Character 166. Ratio (R) between the minimum mediolateral width of the
basisphenoid constriction and the maximum width of this element across the spheno-
occipital tubercles, for the braincases sampled to study the variation of this character.

Cluster (State)
CUST JV Charonosaurus jiayinensis 1.67 1
UCMP 130156 Edmontosaurus sp. 1.87 1
YPM 618 Edmontosaurus sp. 1.66 1
MOR unrecorded no. Hadrosaurinae lower Two Medicine Fm 2.21 2
MOR 553S Hypacrosaurus stebingeri 1.42 0
UCMP 143270 Parasaurolophus cyrtocristatus 1.68 1
−20
2.4
−30
Ratio of the Rostral Basisphenoid Constriction
BIC
−40
2.2
−50
−60
E V
2.0
2 4 6 8
Number of Clusters
1.8
1.6
1.4
Am Ba Br Ch Coi Cos Ed Edr Grm Hy Hys Ma Pac Pat Ta Ts x
Taxon
Figure E.21. Character 166. Boxplot showing the distribution of the the ratio between the minimum mediolateral width of the rostral
constriction and the maximum width of the basisphenoid across the spheno-occipital tubercles for a sample of iguanodontoidean ornithopods.
The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which the data can be classified.
The photograph shows how the two variables of this raio were measured, exemplified in the parietal of Edmontosaurus sp. (AMNH 427).
Grey horizontal lines delimit character states.
state 0
A
state 1
Figure E.22. Character 167. Absence and presence (rectangle) of complete lateral osseous closure
of the ophthalmic sulcus of the laterosphenoid. A, Amurosaurus riabinini (cast of AEHM 1/12), lateral view.
B, Hypacrosaurus altispinus (CMN 8675), dorsal view. Abbreviations: V1 = ophtalmic sulcus; V3 = mandibular
sulcus.
567
Figure E.23. Character 168. Great reduction of the length of the postorbital process (rectangle) of the
laterosphenoid. Neurocranium of Edmontosaurus regalis (CMN 2289) in rostroventral view.
568
state 0
A
state 1
Figure E.24. Character 169. Orientation of the caudal surface (arrows) of the supraoccipital.
A, Tanius sinensis (PMU R240), caudodorsolateral view. B, Brachylophosaurus canadensis
(TMP 90.104.1), caudodorsolateral view.
569
state 0
A
state 1
Figure E.25. Character 170. Absence and presence of a lateroventral corner of the supraoccipital
that is deeply inset into the exoccipital. A, Bactrosaurus johnsoni (AMNH 6365), caudal view.
B, Corythosaurus intermedius (CMN 8676), caudal view.
570
A B C
Figure E.26. Character 171. Caudal extension of the exoccipital-supraoccipital shelf above the foramen magnum. A, Hypacrosaurus
altispinus (CMN 8675), ventral view. B, Brachylophosaurus canadensis (MOR 1071-7-7-98-86), ventral view. C, Edmontosaurus regalis
(CMN 2289), ventral view.
571
state 0
A
state 1
Figure E.27. Character 172. Absence and presence (rectangle) of complete lateral osseous closure of the
ophthalmic sulcus of the laterosphenoid. A, Charonosaurus jiayinensis (CUST JV1251-57), caudal view.
B, Tsintaosaurus spinorhinus (IVPP V725), caudal view. Abbreviations: pp = paroccipital process.
572
state 0
state 1
Figure E.28. Character 173. Elevation (rectangles) of the proximodorsal region of the quadrate
wing of the pterygoid. A, Edmontosaurus regalis (CMN 2289), mediorostral view. B, Brachylophosaurus
canadensis (MOR 1071-7-23-98-387), mediorostral view.
573
state 0
state 1
Figure E.29. Character 174. Ventral extension of the lamina (rectangles) located ventral to the
central buttress of the pterygoid. A, Gryposaurus sp. (MOR 399), mediorostral view. B, Edmontosaurus
regalis (CMN 2289), mediorostral view.
574
state 0
state 1
Figure E.30. Character 175. Presence and absence of ectopterygoid –jugal contact, as
evidenced on the medial side of the jugal. A, Iguanodon atherfieldensis (BMNH R5764), medial
view. B, Bactrosaurus johnsoni (CMN 6373), medial view.
575
APPENDIX F
REGIONAL CRANIAL CHARACTERS
Character 176: lateral profile of the dorsal margin of the rostrum above the
external naris (Fig. F.1)
(0): convex and arcuate
(1): straight or slightly concave
Character 177: angle between the dorsal margin of the rostrum parallel to the long
axis of the external naris and the maxillary tooth row (adults only) (Table F.1; Fig.
F.2)
(1): angle greater than 30º and up to 40º (sample mean angle of 34º)
Character 178: exposure of the nasal passage (Fig. F.3; modified from Norman,
2002, character 5)
(0): present, nasal passage open and exposed on the lateral side of the rostrum
(1): absent, nasal passage nearly or completely enclosed by bone and formation of
internal cavities and passages (such as lateral diverticula and a common median chamber)
Comments: I differentiated between narial foramen (Wagner, 2001) and the bony
naris (Evans, 2006). Functionally, both structures formed the external naris. The
iguanodontoidean narial foramen was regarded as homologous to the common median
chamber of lambeosaurines (Hopson, 1975; Wagner, 2004). Because of caudodorsal
migration of the external passage to become enclosed in a supracranial crest forming the
common median chamber in lambeosaurines, what remained on the laterodorsal surface
of the premaxilla in the latter group was not regarded as structurally homologous to the
narial foramen.
Character 179: morphology of the narial foramen (Fig. F.4 and F.5; Evans and
Reisz, 2007, character 15)
(0): broad and subellipsoidal in lateral profile
(1): narrow and subellipsoidal in lateral profile
(2): extremely narrow, cleft-like, in lateral profile
576
Character 180: degree of closure of the nasal passage on the lateral crest surface
between the caudoventral process of te premaxilla and the nasal (Fig. F.6; Evans,
2007, character 15)
(0): present, premaxilla-nasal fontanellae persist into late ontogenetic stages;
(1): absent, fontanellae completely closed in adults.
Comments: As indicated by Evans and Reisz (2007), the nasal passage is entirely
closed in subadult specimens of Hypacrosaurus altispinus and H. stebingeri.
Character 181: ratio between the length of the narial foramen and the distance
between the rostroventral corner of the premaxilla and the rostroventral margin of
the prefrontal (Table F.2 and Fig. F.7)
(0): very short narial foramen, ratio up to 0.40 (sample mean ratio of 0.32)
(1): moderately long narial foramen, ratio greater than 0.40 but less than 0.60
(sample mean ratio of 0.49)
(2): elongated narial foramen, ratio between 0.60 and 0.65 (sample mean ratio of
0.62)
(3): extemely long narial foramen, ratio greater than 0.65 (sample mean ratio of
0.72)
Character 182: nasal vestibule folded into an S-loop in the enclosed premaxillary
passages rostral to the dorsal process of the maxilla (Weishampel, 1981; Evans,
2007, character 8)
(0): absent (Weishampel, 1981, Fig. 6)
(1): present (Weishampel, 1981, Figs. 4-5)
Cnaracter 183: location of the lateral diverticulum relative to the common median
chamber (Weishampel, 1981)
(0): lateral to the common median chamber (Weishampel, 1981, Fig. 4 and 6)
(1): rostral to the common median chamber (Weishampel, 1981, Fig. 10)
(2): caudodorsal to the common median chamber (Weishampel, 1981, Fig. 8)
Character 184: communication between the external bony naris, the lateral
diverticulum and the common median chamber (Evans, 2006)
577
(0): a tubular premaxillary passage extends caudodorsally from the bony naris to
the lateral diverticulum, that is then connected to the common median chamber
(Weishampel, 1981, Figs. 4-6)
(1): a tubular premaxillary passage connects directly the bony naris to the
common median chamber (Weishampel, 1981, Fig. 7-8).
Character 185: caudal extent of the nasal passage dorsal and/or caudal to the orbit
(Fig. F.8; modified from Evans, 2007, character 9)
(0): absent, nasal passage restricted to the antorbital region of the skull
(1): present, but not extending caudal to the occiput, with a nasal vestibule that
flanks a common median chamber
(2): present, nasal vestibule extending caudodorsal to the occipital region of the
skull
Character 186: composition of the caudal margin of the functional external naris
(Fig. F.9; modified from Horner et al., 2004, character 29)
(0): formed by the nasal dorsally and the premaxilla ventrally
(1): formed entirely by the nasal
(2): formed entirely by the premaxilla
Character 187: circumnarial fossa on the lateral surface of the facial region of the
skull (Fig. F.10; Wagner, 2001; Horner et al., 2004, character 31)
(0): absent, circumnarial structure entirely enclosed
(1): present
Character 188: caudodorsal extension circumnarial fossa (homologous to the lateral
diverticulum inside hollow supracranial crests) (Fig. F.11 and F.12; Hopson, 1975;
Wagner, 2001; 2004; modified from Horner et al., 2004, characters 32):
(0): the fossa does not reach the caudal margin of the narial foramen and, thus,
lacks a caudal margin.
(1): the fossa extends as far as to surround the caudal margin of the narial
foramen, but does not reach the orbit.
(2): the fossa extends as far as the rostrodorsal region of the orbit.
(3): the fossa extends beyond the orbit, caudodorsal to its caudal margin.
578
Character 189: degree of excavation of the caudal region of the circumnarial fossa
(Fig. F.13; Horner et al., 2004, character 32)
(0): lightly incised
(1): deeply incised and invaginated
Character 190: elevation of the skull roof dorsal to the ancestral lateral profile (i.e.,
presence of supracranial crest) (Figs. F.14; Wagner, 2004)
(0): absent
(1): present
Character 191: composition of the supracranial crest (or the homologous region of
the skull from which the crest forms) (excluding supporting elements) (Fig. F.15;
(0): composed exclusively of the nasals
(1): primarily composed of the nasals and frontals
(2): primarily composed of the nasals and premaxillae
Character 192: relative contribution of the nasal and premaxilla in the formation of
hollow supracranial crests (Fig. 16; Wagner, 2001 and modified in part from Evans
and Reisz, 2007, character 11)
(0): the nasals constitute half or a larger portion of the crest in the form of a
caudal plate-like surface.
(1): the nasals form a smaller portion of the crest relative to the surrounding
premaxillae.
Character 193: general shape of the supracranial crest (Figs. F.17-F.20; modified
from Horner et al., 2004, character 36)
(0): dome-like broad and low protuberance
(1): mediolaterally compressed arcuate protuberance, rostral or, in adults, dorsal
to the level to the orbits
(2): paddle-like and caudally (as well as slightly dorsally) directed solid blade of
bone
(3): mediolaterally narrow and paddle-like, extending caudal to the occiput
(4): rostrally excavated and rostrally-facing protuberance, rostrodorsal to the orbit
579
(5): nasal fold that rises dorsally or caudodorsally to form a laterally excavated
promontory, with a caudal region that rests over the frontals
(6): raised into a large vertical fan, formed by a solid plate-like extension of the
premaxilla (“cockscomb”) above the nasal passages in the rostral region of the crest
(7): long and tubular, caudodorsally directed beyond the occiput and slightly
arched
Character 194: hollow crest-snout angle along the dorsal margin of the premaxilla
in lateral view (in adults) (Fig. F.21; Evans, 2007, character 13)
(0): absent, the lateral profile of the snout is continuous with the lateral profile of
the dorsal premaxillary margin of the crest
(1): facial profile shallowly concave in lateral view, angle greater than 140º.
(2): angle between 110º and 140º
(3): crest procumbent and rostrally inclined, angle less than 110º
Character 195: caudal extension of the hook-like nasal process on the caudoventral
region of helmet-shaped hollow supracranial crests (Fig. F.22)
(0): rostral to or at the level of the caudal margin of the occiput
(1): extended caudal to the level of the caudal margin of the occiput
Character 196: palpebral (supraorbital) bone (Fig. F.23; Norman, 2002, character
13)
(0): absent or fused to the orbital margin
(1): present
Character 197: length/width proportions of the orbit (Fig. F.24; Wagner, 2001)
(0): nearly circular, approximately as wide as it is deep
(1): elongated, dorsoventrally deeper than it is wide
Character 198: presence of a gap (paraquadratic foramen) between the
quadratojugal and the jugal (Fig. F.25)
(0): absent
(1): present
Character 199: size of the infratemporal fenestra relative to that of the orbit (Fig.
F.26; modified from Gates and Sampson, 2007)
580
(0): infratemporal fenestra both rostrocaudally wider and dorsoventrally deeper
than the orbit
(1): infratemporal fenestra rostrocaudally narrower than or approximately as wide
as the orbit
Character 200: shape and rostrocaudal width of the dorsal margin of the
infratemporal fenestra relative to that of the dorsal orbital margin (Fig. F.27-F.28)
(0): subrectangular, with a dorsal infratemporal margin that is approximately as
wide as the ventral margin
(1): subtriangular, with a dorsal infratemporal margin that is narrower than the
ventral margin
Comments: within state (1), there was considerable variation in the width of the
infratemporal fenestra within species of Corythosaurus and Lambeosaurus. The
narrowest width among lambeosaurines was found in Hypacrosaurus altispinus, where
the latter was approximately four times longer than wide.
Character 201: location of the dorsal margin of the infratemporal fenestra relative
to the dorsal margin of the orbit (Fig. F.29; modified from Gates and Sampon, 2007,
character 86)
(0): the dorsal margin of the infratemporal fenestra lies approximately at the same
level than the dorsal margin of the orbit and the caudal region of the skull roof is
subhorizontal or slightly slopping rostroventrally relative to the frontal plane.
(1): the dorsal margin of the infratemporal fenestra is substantially more dorsally
located than the dorsal margin of the orbit and the caudal region of the skull roof is
rostroventrally inclined relative to the frontal plane.
(2): the dorsal margin of the infratemporal fenestra lies slightly or substantially
below the level of the dorsal margin of the orbit and the caudal region of the skull roof is
subhorizontal or slightly slopping caudoventrally relative to the frontal plane.
Character 202: morphology of the dorsal outline of the supratemporal fenestra (Fig.
F.30)
(0): subrectangular, with the long axis directed rostrally
(1): oval, with the long axis directed rostrolaterally
(2): oval and wider mediolaterally than rostrocaudally
581
Character 203: length of the skull (measured from the caudal margin of the
quadrate to the rostral end of the oral margin of the premaxilla) relative to its
height (measured from the ventral surface of the quadrate to the dorsal border of
the squamosal) (Table F.3; Fig. F.32; modified from Hai-lu You et al., 2003,
character 1)
(0): ratio up to 2 (sample mean ratio of 1.82)
(1): relatively elongated skull, ratio greater than 2 (sample mean ratio of 2.18)
Character 204: maximum transverse width of the cranium in dorsal view across the
postorbitals relative to the width across the quadrate cotylus of the squamosals
(Table F.4; Fig. F.32; modified from Horner et al., 2004, character 67)
(0): the skull is more than 25% wider across the postorbitals (sample mean ratio
of 0.35).
(1): the skull is up to 25% wider across the postorbitals (sample mean ratio of
0.14).
Character 205: shape of the occiput in caudal view (Fig. F.33; Horner et al., 2004,
character 68)
(0): rectangular
(1): subtriangular, with the quadrates laterally splayed distally
582
state 0
A
state 1
Figure F.1. Character 176. Lateral profile of the dorsal margin of the rostrum (brackets)
above the narial foramen or the bony naris. A, Brachylophosaurus canadensis (TMP 90.104.1),
lateral view. B, Edmontosaurus regalis (CM 26258), lateral view.
583
Table F.1. Character 177. Angle (A) between the long axis of the external naris and the
maxillary tooth row for a sample of iguanodontoidean ornithopods. Included is the

Cluster (State)
TMP 80.40.1 Corythosaurus casuarius 34 1
YPM 2182 Edmontosaurus annectens 23 0
DMNH 1943 Edmontosaurus annectens 27 0
FMNH P15004 Edmontosaurus regalis 26 0
TMP 80.22.1 Gryposaurus notabilis 49 2
AMNH 5278 Hypacrosaurus altispinus 34 1
584
Table F.1—continued.

Cluster (State)
IVPP 12691 Jinzhousaurus yangi 34 1
YPM 3222 Lambeosaurus lambei 35 1
CMN 8705 Lambeosaurus magnicristatus 34 1
TMP 66.04.01 Lambeosaurus magnicristatus 37 1
AEHM 2/845 Olorotitan arharensis 32 1
MPC-D100/706 Saurolophus angustirostris 36 1
PIN unrecorded no. Saurolophus angustirostris 37 1
585
50
−410
−430
45
BIC
Inclination Angle of the Rostrum
−450
40
−470
E V
2 4 6 8
Number of Clusters
35
30
25
Br Coi Cos Eda Edr Eq Grn Hy Hys Iga Igb Jz Lml Lmm Ma Ol Paw Pr Pt Saa Sao
Taxon
Figure F.2. Character 177. Boxplot showing the distribution of the angle between the dorsal margin of the rostrum and the maxillary tooth
row for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for various
numbers of clusters in which the data can be classified. The photograph shows how this angle was measured, exemplified in a subadult skull
of Edmontosaurus annectens (CMN 8509).
586
state 0
state 1
Figure F.3. Character 178. Exposure (A) and closure (B) of the nasal passages. A, Prosaurolophus
maximus (CMN 2277), lateral view. B, Corythosaurus intermedius (ROM 776), lateral view.
587
300
●
●
● ●
● ●
●
●
●
200
BIC
100
0
EII VVI
VII EEE
● EEI EEV
−100
VEI VEV
EVI VVV
2 4 6 8 A
Number of Clusters
0.00 0.05 0.10 0.15
Axis2
−0.10
−0.4 −0.2 0.0 0.2 0.4
Axis1 B
Figure F.4. Character 179. Geometry of the narial foramen. A, BIC plot for various clustering models of the
NMMDS derived from the GDA dissimilariy matrix. B, coordinate projection of the NMMDS data
showing chosen clustering into three states. Abbreviations of clustering models: EII = spherical, equal volume and
shape; VII = sperical, variable volume, equal shape; EEI = diagonal, equal volume and shape; VEI = diagonal, variable
volume, equal shape; EVI = diagonal, equal volume, variable shape; VVI = diagonal, variable volume and shape;
EEE = ellipsoidal, equal volume, shape and orientation; EEV = ellipsoidal, equal volume and shape, variable orientation;
VEV = ellipsoidal, variable volume, equal shape; VVV = ellipsoidal, variable volume and shape (Fraley and Raftery, 2003).
588
state 0
A
state 1
state 2
Figure F.5. Character 179. Geometry of the narial foramen (rectangle). A, Brachylophosaurus
canadensis (CMN 8893), lateral view. B, Edmontosaurus annectens (UCM 20892), lateral view.
C, Prosaurolophus maximus (TMP 84.1.1), lateral view .
589
state 0
A
state 1
Figure F.6. Character 180. Degree of closure of the nasal passage on the lateral crest surface between
the caudoventral process of te premaxilla and the nasal. Arrows point to fontanellae. A, Corythosaurus
intermedius maximus (ROM 776), lateral view. B, Hypacrosaurus altispinus (CMN 8501), lateral view.
590
Table F.2. Character 181. Ratio (R) between the length of the narial foramen and the
distance between the rostroventral corner of the premaxilla and the rostroventral margin
of the prefrontal, for the specimens sampled to study the variation of this character.

Cluster (State)
IRNSB 1731 Iguanodon bernissartensis 0.28 0
591
30
20
0.7
BIC
Ratio of the Length of the Narial Foramen
10
0
−10
−20
0.6
2 4 6 8
Number of Clusters
0.5
0.4
0.3
Br Ed Eda Edr Eq Grn Igb Ma Pr Pt Sao
Taxon
Figure F.7. Character 181. Boxplot showing the distribution of the ratio between the length of the narial foramen and the distance between the
rostroventral corner of the premaxilla and the rostroventral margin of the prefrontal for a sample of iguanodontoidean ornithopods.
The photograph shows how the two variables of this raio were measured, exemplified in the skull of Gryposaurus notabilis (CMN 2278).
592
state 0
state 1
B
state 2
Figure F.8. Character 185. Caudal extent (arrows) of the nasal passages. A, Brachylophosaurus
canadensis (CMN 8893), lateral view. B, Corythosaurus intermedius (UALVP 13), lateral view.
C, Parasaurolophus cyrtocristatus (UCMP 143270), lateral view. Abbreviation: cnsp = caudal extent
of the nasal passage.
593
state 0
A
state 1
B
state 2
Figure F.9. Character 186. Composition of the caudal margin (rectangles) of the external naris.
A, Equijubus normani (IVPP V12534), lateral view. B, Brachylophosaurus canadensis (MOR 794), lateral
view. C, Lambeosaurus lambei (YPM 3222), lateral view. Abbreviations: ns = nasal; pmx = premaxilla.
594
state 0
state 1
Figure F.10. Character 187. Absence and presence of circumnarial fossa (rectangle).
A, Corythosaurus intermedius (ROM 845), lateral view. B, Edmontosaurus regalis (ROM 801),
lateral view.
595
state 0
A
state 1
B
state 2
lateral and caudal margins of the circumnarial fossa A, Iguanodon bernissartensis (IRSNB 1731), lateral
view. B, Edmontosaurus regalis (ROM 801), lateral view. C, Maiasaura peeblesorum (TCMI 2001.89.2),
laterodorsal view (courtesy of Terry Gates).
596
state 2
state 2 A
Figure F.12. Character 188. Caudal extent of the circumnarial fossa. The arrows indicate the medial,
lateral and(or caudal margins of the circumnarial fossa A, Brachylophosaurus canadensis (MOR 794),
dorsolateral view. B, Saurolophus angustirostris (unrecorded catalog number), lateral view
(photo courtesy of David Evans).
597
state 0
A
state 1
B
state 1
Figure F.13. Character 189. Degree of excavation of the caudal region (delimited by the arrows)
of the circumnarial fossa. A, Gryposaurus notabilis (ROM 873), lateral view. B, Edmontosaurus regalis
(FMNH P15004), lateral view. C, Prosaurolophus maximus (TMP 84.1.1), lateral view.
598
state 0
state 1
Figure F.14. Character 190. Absence and presence (rectangle) of supracranial crest.
A, Edmontosaurus annectens (AMNH 5730), lateral view. B, Gryposaurus notabilis (MSNM V345),
lateral view.
599
state 0
A
state 1
B
state 2
Figure F.15. Character 191. Composition of the supracranial crest. A, Brachylophosaurus

canadensis (MOR 1071-7-7-98-86), lateral view. B, Maiasaura peeblesorum (OTM F138), lateral
view. C, Hypacrosaurus stebingeri (MOR 549), lateral view. Abbreviations: f = frontal; ns = nasal;
pmx = premaxilla.
600
state 0
A
state 1
Figure F.16. Character 192. Relative contribution of the premaxilla and nasal bones to the
formation of hollow supracranial crests. A, Corythosaurus casuarius (ROM 871), lateral and slightly
caudal view. B, Lambeosurus magnicristatus (CMN 8705), lateral view. Abbreviations: ns = nasal;
pmx = premaxilla.
601
state 0
A
state 1
B
state 2

A, Ouranosaurus nigeriensis (GDF 300; photo courtesy of David B. Weishampel), lateral
view. B, Gryposaurus notabilis (CMN 2278), lateral view. C, Brachylophosaurus canadensis
(CMN 8893), laterodorsal view.
602
state 3
A
state 4

A, Saurolophus osborni (AMNH 5220), lateral view. B, Maiasaura peeblesorum (OTM F138),
lateral view. C, Brachylophosaurus canadensis (CMN 8893),
603
B
state 5
A
state 5 state 5
B C

A, Prosaurolophus maximus (CMN 2277), lateral view. B, Kritosaurus navajovius (BYU 12950),
lateral view.
604
B
state 6
A
state 7
Figure F.20. Character 193. General morphology of supracranial crests (arrows). A, Corythosaurus
casuarius (ROM 1933), lateral view. B, Parasaurolophus tubicen (NMMNH P-25100), lateral view.
605
state 0 state 1
A B
state 2 state 3
C D
Figure F.21. Character 194. Hollow crest-snout angle along the dorsal margin of the premaxilla in lateral view (in adults).
A, Parasaurolophus walkeri (ROM 768), lateral view. B, Hypacrosaurus altispinus (CMN 8501), lateral view. C, Corythosaurus
casuarius (TMP 84.121.1), lateral view. D, Lambeosaurus lambei (FMNH PR380), lateral view.
606
state 0 state 0
20 cm 20 cm
A B
state 1 state 1
20 cm 20 cm
C D
Figure F.22. Character 195. Caudal extension of the caudoventral corner of the nasal in a
dome-shaped hollow supracranial crest. A, Hypacrosaurus altispinus (CMN 8501), lateral view.
B, Corythosaurus intermedius (ROM 776), lateral view. C, C. casuarius (ROM 1933), lateral view.
D, “H.” stebingeri (MOR 549), lateral view.
607
A
Figure F.23. Character 196. Presence of palpebral bone (arrows). A, Iguanodon

bernissartensis (IRSNB 1726), lateral view. B, I. atherfieldensis (IRSNB 1551), lateral view.
608
state 0
A
state 1
Figure F.24. Character 197. Length/width proportions of the orbit. A, Iguanodon atherfieldensis
(BMNH 11521), lateral view. B, Brachylophosaurus canadensis (MOR 794), lateral view.
Abbreviation: O = orbit.
609
state 0
A
state 1
Figure F.25. Character 198. Absence and presence of paraquadratic foramen (rectangle).
A, Maiasaura peeblesorum (ROM 44770), lateral view. B, Equijubus normani (IVPP V12534),
lateral view.
610
state 0
A
state 1
Figure F.26. Character 199. Height and width of the infratemporal fenestra relative to
those of the orbital cavity. A, Gryposaurus notabilis (CMN 2278), lateral view. B, Corythosaurus
intermedius (CMN 8503), lateral view. Abbreviations: itf = infratemporal fenestra; o = orbit.
611
A B C D E F
G H I J K L
M N O P Q R
S T U V W
infratemporal fenestra relative to that of the dorsal orbital margin. State 0:subrectangular lateral
profile, with a dorsal infratemporal margin that is approximately as wide as the ventral margin.
This figure shows the outlines of a number of hadrosaurid taxa in right lateral view. A, Corythosaurus
intermedius (CMN 8503). B, C. intermedius (CMN 8704). C, C. intermedius (ROM 845).
D, C. intermedius (TMP 80.23.4). E, Corythosaurus sp. (ROM 870). F, C. casuarius (AMNH 5240).
G, C. casuarius (AMNH 5338). H, C. casuarius (ROM 871). I, Gryposaurus monumentensis (RAM 6797).
J, G. notabilis (CMN 2278). K, G. notabilis (MSNM V345). L, G. notabilis (ROM 873). M, G. notabilis
(TMP 80.22.1). N, Hypacrosaurus altispinus (CMN 8501). O, H. altispinus (ROM 789). P, Kritosaurus
navajovius (AMNH 5799). Q, K. navajovius (NMMNH P-16106). R, Lambeosaurus lambei (AMNH 5340,
juvenile). S, L. lambei (CMN 8703). T, L. lambei (CMN 8633, subadult). U, L. lambei (AMNH 5353).
V, L. lambei (AMNH 5373). W, L. lambei (ROM 1218).
612
A B C D E
F G H I J
K L M N
infratemporal fenestra relative to that of the dorsal orbital margin. State 1: subtriangular, with a
dorsal infratemporal margin that is narrower than the ventral margin. This figure shows the outlines
of a number of hadrosaurid taxa in right lateral view. A, Brachylophosaurus canadensis (MOR 794).
B, Edmontosaurus annectens (CMN 8399). C, E. annectens (CMN 8509, subadult). D, E. annectens
(DMNH 1943). E, E. annectens (SM R4036). F, E. regalis (AMNH 5254). G, E. regalis (CM 26258).
H, E. regalis (CMN 2288). I, E. regalis (ROM 801). J, E. regalis (FMNH P15004). K, Parasaurolophus
walkeri (ROM 768). L, Prosaurolophus maximus (CMN 2870). M, P. maximus (TMP 84.1.1).
N, Saurolophus osborni (AMNH 5220).
613
state 0
A
state 1
B
state 2
Figure F.29. Character 201. Location of the dorsal margin of the infratemporal fenestra relative to
the dorsal margin of the orbit . A, Brachylophosaurus canadensis (TMP 90.104.1), lateral view.
B, Gryposaurus notabilis (CMN 2278), lateral view. C, Hypacrosaurus altispinus (CMN 8501),
lateral view.
614
state 0
A B C D E F G
state 0
H I J K L M N
state 1
O P Q R S T U
state 2
Figure 30. Character 202. Shape of the supratemporal fenestra. Outlines drawn in dorsal view; the lateral side of the skull is to the
left of the drawings. A, Bactrosaurus johnsoni (PIN 2549-1. B, Edmontosaurus annectens (AMNH 5730). C, E. annectens (ROM 867, subadult).
D, E. annectens (SM R4036). E, Gryposaurus notabilis (AMNH 5350). F, G. notabilis (CMN 2278). G, G. notabilis (ROM 764). H, Kritosaurus
navajovius (NMMNH P-16106, subadult). I, Maiasaura peeblesorum (OTM F138). J, M. peeblesorum (YPM-PU 22405). K, Prosaurolophus
maximus (CMN 2870). L, P. maximus (ROM 787). M, P. maximus (TMP 84.1.1). N, Secernosaurus koerneri (MACN-RN 144). O, Aralosaurus
tuberiferus (PIN 2229). P, cf. Lambeosaurus (USNM 10309). Q, Hypacrosaurus altispinus (CMN 8675). R, H. altispinus (ROM 702).
S, H. stebingeri (MOR 549). T, Jaxartosaurus aralensis (PIN 1/50009). U, Ouranosaurus nigeriensis (GDF 300). V, H. altispinus (IVPP V725).
615
Table F.3. Character 203. Ratio (R) between the basal length of the skull (from the caudal
margin of the quadrate to the rostral end of the oral margin of the premaxilla) and its
height (from the ventral margin of the quadrate to the dorsal border of the squamosal) for
a sample of iguanodontoidean taxa. Included is the corresponding character state coding

(State)
UCMP 128372 V76142 Edmontosaurus annectens 2.22 1
FMNH 15004 Edmontosaurus regalis 1.77 0
616

(State)
TMCI 2001.89.2 Maiasaura peeblesorum 1.78 0
PIN unrecorded no. Saurolophus angustirostris 2.34 1
617
2.3
30
20
2.2
BIC
Height/Length Ratio of the Skull
10
0
2.1
−10
2 4 6 8
Number of Clusters
2.0
1.9
1.8
1.7
Br Coi Cos Eda Edr Eq Grm Grn Hy Hys Igb Jz Lml Lmm Ma Pat Pr Saa Sao
Taxon
Figure F.31. Character 203. Boxplot showing the distribution of the ratio between the height and length of the skull for a sample of
iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in
which the data can be classified. The photograph shows how the two variables of this raio were measured, exemplified in the skull of
Corythosaurus intermedius (CMN 8704).
618
Table F.4. Character 204. Ratio (R) between the maximum mediolateral width of the
skull across the postorbitals and the mediolateral width across the quadrate cotylae, for
the crania sampled to study the variation of this character. Included is the corresponding
character state coding derived from K-Means clustering.

(State)
PIN 2229 Aralosaurus tuberiferus 1.09 1
USNM 10309 cf. Lambeosaurus 1.14 1
ROM 1940 Corythosaurus sp. 1.17 1
PIN 1/5009 Jaxartosaurus aralensis 1.18 1
OTM F138 Maiasaura peeblerosum 1.36 0
PIN 2232/36-1 Probactrosaurus gobiensis 1.09 1
GMV 1780 Shantungosaurus giganteus 1.09 1
619

(State)
MOR 1155 Wahweap hadrosaurid 1.30 0
620
60
50
40
1.4
BIC
30
Width Ratio of the Skull
20
10
0
1.3
2 4 6 8
Number of Clusters
1.2
1.1
Ar Ba Br Co Coi Eda Edr Grn Hy Hys Iga Igb Jx Krn Lml Ma Ou Pat Pb Pr Saa Sh Ta Ts Ve cfLm x
Taxon
Figure F.32. Character 204. Boxplot showing the distribution of the ratio between the maximum transverse width of the cranium across
the postorbitals and the width across the quadrate cotylus of the squamosals for a sample of iguanodontoidean ornithopods. The inserted smaller
graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph shows
how the two variables of this raio were measured, exemplified in a skull roof of Hypacrosaurus altispinus (CMN 8675).
621
state 0 state 1
A B
Figure F.33. Character 205. Shape of the occiput and orientation of the quadrates in caudal view. A, Iguanodon bernissartensis
(cast of IRSNB 1534), caudal view. B, Brachylophosaurus canadensis (CMN 8893), caudal view.
622
APPENDIX G
VERTEBRAL AND OTHER AXIAL CHARACTERS
Character 73 of Horner et al. (2004) about shortening of the central of cervical

vertebrae is excluded from this analysis because no significant difference was found in
this character among the taxa under study. Additionally, the position of the vertebra in the
series and postdepositional deformation add to the variation see in the sample examined.
Character 206: morphology of the dorsal flange of the axis (Fig. G.1; modified from
Campione et al., 2007)
(0): dorsally convex flange extending beyond or to the level of the cranialmost
region of the postzygapophyses
(1): presence of short cranial flange separated from the postzygapophyseal region
by a prominent embayment
Character 207: development of the postzygapophyseal proceses of cranial and
middle cervical vertebrae (Fig. G.2; modified from Horner et al., 2004, character
74)
(0): relatively low and relatively short, less than three times the rostrocaudal
breadth of the neural arch
(1): relatively high and relatively long, three times or more longer than the
breadth of the neural arch
Character 208: number of cervical vertebrae (Horner et al., 2004, character 72)
(0): 11 or less
(1): 12 or more
Dorsal Vertebrae
Character 209: height of the neural spine relative to that of the centrum of the
tallest posterior dorsal or sacral vertebrae (in adults) (Table G.1; Fig. G.3; modified
from Norman, 2002, character 41)
(0): short neural spine, ratio up to 2.10 (mean ratio of 1.79)
(1): ratio greater than 2.10 and up to 3.25 (mean ratio of 2.57)
623
(2): very long neural spine, ratio greater than 3.25 (mean ratio of 3.97)
Character 210: slightly elongated neural spines in the cranial dorsal vertebrae,
forming a “wither-like” region above the pectoral girdle (Fig. G.4; Wagner, 2001)
(0): absent
(1): present
Sacral Vertebrae
In the past, several authors regarded the presence of a haemal ridge as

synapomorphy of lambeosaurine hadrosaurids (Weishampel et al., 1993; Kirkland, 1998).
Horner (1990) pointed out that his Hadrosauria (composed of Iguanodon and non-
lambeosaurine hadrosaurids) could be characterized by a ventrally ridged sacrum. Brett-
Surman (1979) and Weishampel and Horner (1990) regarded the ridged sacrum as
synapomorphic for hadrosaurines. My observations support Godefroit et al. (2001) in that
the presence of haemal sulcus and/or ridge is not phylogenetically informative. A haemal
sulcus is commonly present on the caudal half of the ventral side of the sacrum in the
majority of observed iguanodontoidean specimens. Additionally, in many exemplars the
cranial half of the sacrum showed a longitudinal ridge (Table G.2).
Character 211: minimum count of co-ossified vertebrae in the sacral region
(including single dorsal and caudal contributions (Table G.3; Godefroit et al., 2000,
character 27)
(0): seven or fewer
(1): eight or more
Comments: Brett-Surman (1989) suggested that the number of sacral vertebrae
may increase with age. This idea is supported by the observation the addition of two
caudal vertebrae to the sacrum of a very large Shantungosaurus, for a total of ten sacral
vertebrae. However, some small hadrosaurid individuals such as ZPAL MgD-I 159 have
as much as nine sacrals, compared to the eight sacrals observed in adult specimens such
as MPC-D 100/706. The small exemplar of Lambeosaurus, AMNH 5340, showed the
relatively large count of nine sacrals.
624
Caudal Vertebrae
Character 212: chevron length relative to the length of the neural spines in the
caudal vertebrae of the proximal half of the tail (Fig. G.6; Wagner, 2001)
(0): chevrons shorter or nearly as long as the neural spines
(1): chevrons longer than the neural spines
Sternal
Character 101 by Prieto-Marquez et al. (2006) regarding the development of the

caudoventral process of the craniomedial plate of the sternal was excluded from this
analysis because of the substantial intraspecific variation of this feature (Fig. G.7). For
example, the caudoventral process was relatively small in Brachylophosaurus
canandensis MOR 794 and MOR 1071 but very large in UCMP 130139. Likewise, the
process was poorly developed in some exemplars of Edmontosaurus annectens (SM
R4036) but very large in others (AMNH 5060). In many lambeosaurines the process
appeared to be relatively large (Lambeosaurus sp., AMNH 5340 or cf. Hypacrosaurus
stebingeri, MOR 548) but poorly developed in others (Olorotitan arharensis AEHM
2/845 and Amurosaurus riabinini, AEHM 1/12).
Character 213: length of the “handle-like” caudolateral process of the sternal
relative to that of the craniomedial plate (excluding the caudoventral process) (Fig.
G.8; modified from Prieto-Marquez et al., 2006, character 100)
(0): caudolateral process slightly shorter or as long as the craniomedial plate
(1): caudolateral process longer than the craniomedial plate
625
state 0
df
A
state 1
df
Figure G.1. Character 206. Morphology of the dorsal flange of the axis. The arrow indicates an
embayment that separates the dorsal falnge from the postzygapophyseal region of the axis.
A, Indeterminate hadrosaurine (USNM 8629), lateral view. B, Brachylophosaurus canadensis
(MOR 794), lateral view. Abbreviation: df = dorsal flange.
626
state 0
state 1
Figure G.2. Character 207. Development of the postzygapophyseal procesess of cranial

and middle cervical vertebrae. A, Gilmoreosaurus mongoliensis (AMNH 6551), dorsal view.
B, Brachylophosaurus canadensis (CMN 8893), dorsal view. Abbreviations: na = neural arch length;
po = postzygapophyseal process length.
627
Table G.1. Character 209. Ratio (R) between the height of the neural spine and the height
of the centrum of the tallest vertebra from the posterior dorsals or sacral vertebrae from a
sample of iguanodontoidean ornithopods. Included is the corresponding character state
coding derived from K-Means clustering analysis.

(State)
ZPAL MgD-I 110 Barsboldia sicinskii 4.18 2
UTEP P37.7.250 Hadrosauridae 1.78 0
MPCA unrecorded no. Hadrosauridae Salitral Moreno 2.42 1
AMNH 5206 Hypacrosaurus altispinus 3.89 2
MACN RN 826 Secernosaurus koerneri 2.41 1
628
Ratio of the Height of the Neural Spine of Dorsal and Sacral Vertebrae
−60
4.0
−70
BIC
−80
−90
3.5
−100
2 4 6 8
Number of Clusters
3.0
2.5
2.0
1.5
Ba Bb Br Eda Edr Grn Hy Hys Iga Igb Lml Ma Ol Pac Paw Ph Pr Se Sh Ta Ts sal x
Taxon
Figure G.3. Character 209. Boxplot showing the distribution of the ratio between the height of the neural spine and that of the centrum
of the tallest posterior dorsal or sacral vertebrae (in adults) for a sample of iguanodontoidean ornithopods. The inserted smaller graph
shows the Bayesian information criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph
shows how the two variables of this raio were measured, exemplified in a dorsal vertebra of Bactrosaurus johnsoni (AMNH 6553).
629
state 0
A
state 0
B B
Figure G.4. Character 210. Absence (A) and presence (B) of slightly elongated neural spines
in the cranial dorsal vertebrae (rectangles), forming a “wither-like” region above the pectoral girdle.
A, Edmontosaurus annectens (AMNH 5730; photo courtesy of David B. Weishampel), lateral view.
B, Lambeosaurus lambei (TMP 82.38.1), lateral view.
630
Figure G.5. Ventral view of the sacrum of Iguanodon bernissartensis (IRSNB 1723) showing the co-occurrence of a ventral sulcus and
ridge (arrows) in the same specimen. Abbreviations: r = ridge; s = sulcus.
631
Table G.2. Presence/absence of haemal sulcus and longitudinal ridge on the ventral surface of the sacral vertebrae in various
iguanodontoidean taxa. When a sulcus is present, it is found along the caudal region of the sacrum (exceptions are a subadult
Saurolophus angustirostris, ZPAL MgD-I 159, where the feature is found cranially, and Charonosaurus jiayinensis, that
showed it running along the whole length of the sacrum).
Specimen Taxon Haemal Ridge Source

sulcus
SBDE 95E5/39 Bactrosaurus johnsoni yes no Godefroit et al., 1998
SBDE 95E5/40 Bactrosaurus johnsoni no yes Godefroit et al., 1998
AMNH 6553 Bactrosaurus johnsoni no yes Godefroit et al., 1998
ZPAL MgD-I 110 Barsboldia sicinskii yes yes Maryanska and Osmolska, 1982
MOR 794 Brachylophosaurus canadensis yes yes personal observation
MOR 547 cf. Maiasaura peeblesorum ? yes personal observation
CUST unrecorded no. Charonosaurus jiayinensis yes no Godefroit et al., 2001
YPM 1190 Claosaurus agilis no yes Carpenter et al., 1995
ROM 845 Corythosaurus intermedius yes yes? personal observation
AMNH 5730 Edmontosaurus annectens yes? ? personal observation
CMN 8893 Edmontosaurus annectens yes ? personal observation
OMNH 27749 Eolambia caroljonesa yes no Kirkland, 1998
IVPP V12534 Equijubus normani no no Hai-lu You et al., 2003
AMNH 6551 Gilmoreosaurus mongoliensis yes yes personal observation
MSNM V345 Gyposaurus notabilis yes ? personal observation
CMN 2278 Gyposaurus notabilis yes no personal observation
PASAC-1 Hadrosauridae no? yes personal observation
AMNH 5202 Hypacrosaurus altispinus no yes personal observation
AMNH 5217 Hypacrosaurus altispinus no yes Brown, 1913
MOR 773 Hypacrosaurus stebingeri yes yes personal observation
BMNH R11521 Iguanodon atherfieldensis ? yes personal observation
BMNH R5764 Iguanodon atherfieldensis yes yes Norman, 1986
IRSNB 1723 Iguanodon bernissartensis yes yes personal observation
632
Table G.2—continued.
Specimen Taxon Haemal Ridge Source

sulcus
IRSNB 1722 Iguanodon bernissartensis yes yes Norman, 1980
IRSNB 1535 Iguanodon bernissartensis yes yes Norman, 1980
IRSNB 1728 Iguanodon bernissartensis yes no personal observation
TMP 82.38.1 Lambeosaurus lambei yes yes personal observation
ROM 1218 Lambeosaurus lambei yes yes personal observation
GDF 300 Ouranosaurus nigeriensis ? yes Head, 2001
IPS SRA-24 Pararhabdodon isonensis yes yes personal observation
ROM 768 Parasaurolophus walkeri ? yes personal observation
PIN 2232/10 Probactrosaurus gobiensis no yes Norman, 2002
ZPAL MgD-I 159 Saurolophus angustirostris yes no personal observation
GMV 1780 Shantungosaurus giganteus yes ? Hu, 1972
BMNH R4911 Telmatosaurus transsylvanicus ? yes personal observation
IVPP V724 Tsintaosaurus spinorhinus yes ? Young, 1958
633
Table G.3. Character 211. Minumum number (N) of co-ossified vertebrae (sacrodorsal
and sacrocaudal contributions included) found in the sacral region of various
iguanodontoidean ornithopods.
Specimen or Source Taxon N

Godefroit et al., 1998 Bactrosaurus jonsoni 7
ZPAL MgD-I 110 Barsboldia sicinskii 9
MOR 794 Brachylophosaurus candensis 9
Godefroit et al., 2001 Charonosaurus jiayinensis 9
Carpenter et al., 1995 Claosaurus agilis 7
ROM 845 Corythosaurus intermedius 10
ROM 867 Edmontosaurus sp. 9
AMNH 5730 Edmontosaurus annectens 9
YPM 2182 Edmontosaurus annectens 9
CMN 2289 Edmontosaurus regalis 9
Kirkland et al., 1998 Eolambia caroljonesa 7
IVPP V12534 Equijubus normani 7
Parks, 1920 Gryposaurus notabilis 10
CMN 2278 Gryposaurus notabilis 9
AMNH 5465 Gryposaurus latidens 9
AMNH 5217 Hypacrosaurus altispinus 8
CMN 8501 Hypacrosaurus altispinus 8
MOR 773 Hypacrosaurus stebingeri 10
Norman, 1986 Iguanodon atherfieldensis 7
Norman, 1980 Iguanodon bernissartensis 8
AMNH 5340 Lambeosaurus sp. 9
ROM 1218 Lambeosaurus lambei 9
Horner et al., 2004 Maiasaura peeblesorum 10
Godefroit et al., 2003 Olorotitan arharensis 15
Taquet, 1976 Ouranosaurus nigeriensis 6
FMNH P27393 Parasaurolophus cyrtocristatus 8
ROM 768 Parasaurolophus walkeri 8
Norman, 2002 Probactrosaurus gobiensis 6
IPS SRA 24 Pararhabdodon isonensis 8
Parks, 1924 Prosaurolophus maximus 10
MOR 454 Prosaurolophus maximus 10
ZPAL MgD-I 159 Saurolophus angustirostris 9
MPC-D 100/706 Saurolophus angustirostris 8
Brown, 1913 Saurolophus osborni 8
GMV 1780 Shantungosaurus giganteus 10
Young, 1958 Tsintaosaurus spinorhinus 10
UTEP P37.7.291 Hadrosauridae 7
634
state 0
A
state 1
Figure G.6. Character 212. Chevron length relative to the length of the neural spines in the
caudal vertebrae of the proximal half of the tail. A, Parasaurolophus cyrtocristatus (FMNH P27393),
lateral view. B, Edmontosaurus annectens (AMNH 5730), lateral view. Abbreviations: c = chevron
length; n = neural spine length.
635
A B C D E F
G H I J K L
M N O P Q R
S T U V
Figure G.7. Shape of the craniomedial plate and the caudoventral process of various iguanodontoidean sternals. Outlines drawn in
cranioventral view.. A, Lambeosaurinae indeterminate (ANSP 16969). B, Bactrosaurus johnsoni (). C, Brachylophosaurus canadensis
(MOR 794). D, B. canadensis (MOR 1071). E, B. canadensis (UCMP 130139). F, Corythosaurus sp. (CMN 36141). G, Edmontosaurus sp.
(LACM 23504, subadult). H, Lambeosaurinae indeterminate (USNM 11950). I, Corythosaurus sp. (ROM 1947). J, E. annectens (YPM 2182).
K, E. regalis (CMN 2289). L, Hadrosauridae indeterminate (USNM 2413). M, Iguanodon atherfieldensis (IRSNB 1551). N, I. bernissartensis
(IRSNB 1534). O, I. bernissartensis (IRSNB 1535). P, I. bernissartensis (IRSNB 1725). Q, Iguanodon sp. (BMNH R2218). R, Lophorhothon
atopus (AUMP 2295). S, Lambeosaurus lambei (AMNH 5340, juvenile). T, Olorotitan ararhensis (AEHM 2/845). U, Prosaurolphus maximus
(ROM 787). V, Tsintaosaurus spinorhinus (IVPP V725).
636
state 0
A
state 1
Figure G.8. Character 213. Length of the “handle-like” caudolateral process of the sternal relative
to that of the craniomedial plate (excluding the caudoventral process). A, Iguanodon sp. (FMNH R2218),
rostroventral view. B, Olorotitan ararhensis (AEHM 1/845), rostroventral view.
637
APPENDIX H
PECTORAL AND FORELIMB CHARACTERS
Coracoid
Character 214: coracoid size relative to the length of the scapula (Fig. H.1; Horner
et al., 2004, character 77)
(0): relatively large coracoid, ratio between craniocaudal length of coracoid and
length of scapula of approximately
(1): coracoid reduced in size relative to the scapula
Character 215: ratio between the length of the lateral margin of the facet for the
scapular articulation and the length of the lateral margin of the glenoid (Table H.1;
Fig. H.2)
(0): longer scapular facer, with a ratio greater than 1.30 (sample mean ratio of
1.48);
(1): slightly longer scapular facet, ratio greater than 1 and up to 1.30 (sample
mean ratio of 1.14)
(2); glenoid longer than the scapular facet, with a ratio up to 1 (sample mean ratio
of 0.75)
Character 216: angle between the lateral margins of the facet for scapular
articulation and the glenoid (Table H.2; Fig. H.3)
Character 217: morphology of the craniomedial margin of the coracoid (Fig. H.4;
(0): convex or straight, associated to a moderate development and slight
projection biceps tubercle
(1): concave, associated to a relatively large and lateroventrally-projected biceps
tubercle
638
Character 218: development of the “hook-like” ventral process of the coracoid,
measured as the ratio between the dorsoventral depth and the breadth of the
process (Table H.3; Fig. H.5; modified from Godefroit et al., 2000, character 25)
(0): relatively short, ratio less than 0.65 (sample mean ratio of 0.55)
(1): ratio between 0.65 to 0.80 (sample mean ratio of 0.71)
(2): long process, nearly as deep as it is wide, with a ratio greater than 0.80
Character 219: curvature of the ventral “hook-like” process of the coracoid (Fig.
H.6; Godefroit et al., 2000, character 25)
(0): ventrally directed
(1): recurved, so that the process is caudoventrally directed
Scapula
Character 220: lateral profile of the dorsal margin of the scapula (Fig. H.7;
modified from Sereno, 1986)
(0): craniocaudally straight from the cranial margin of the coracoid facet to the
distal end of the blade
(1): curved, dorsally convex, curvature originating at the level of the dorsal
margin of the acromion process and is most pronounced over the dorsoventral
constriction
Character 221: scapular length, ratio between the craniocaudal length of the
scapula (from the cranial end of the acromion process to the distal margin of the
blade) and the dorsoventral depth of the cranial end (from the cranial end of the
acromion process to the ventral apex of the glenoidal facet) (Table H.4; Fig. H.8)
(0): relatively short scapula, ratio up to 4 (sample mean ratio of 3.54)
(1): relatively long scapula, ratio greater than 4 (sample mean ratio of 4.64)
Character 222: development of a ventral convexity along the distal half of the blade,
caudal to the proximal constriction (Fig. H.9)
(0): absent, straight margin or very slight convexity
(1): present, well-developed convexity
639
Character 223: dorsoventral expansion of the distal region of the scapular blade
(measured as a ratio between the depth of the distal end of the blade and the depth
of the proximal region) (Table H.5; Fig. H.10)
(0): ratio less than 1 (sample mean ratio of 0.80)
(1): ratio of 1 or greater (sample mean ratio of 1.15)
Character 224: proximal constriction (scapular “neck”), ratio between the
dorsoventral width of the proximal constriction and the dorsoventral depth of the
cranial end of the scapula (Table H.6; Fig. H.11)
(0): narrow “neck”, ratio up to 0.60 (sample mean ratio of 0.53)
(1): relatively broad “neck”, ratio greater than 0.60 (sample mean ratio of 0.68)
Character 225: morphology and orientation of the acromion process of the scapula
(Fig. H.12; modified from Horner et al., 2004, character 80)
(0): recurved, so that the cranial region is dorsally or craniodorsally directed
(1): horizontal, occasionally with minor and subtle dorsal or ventral curvatures, so
that the cranial region is cranially or mostly cranially directed
Character 226: degree of curvature of the dorsally oriented acromion process of the
scapula (Fig. H.12; modified from Horner et al., 2004, character 80)
(0): strongly recurved, so that the cranial region of the process is dorsally oriented
(1): slightly recurved, with concave lateral profile of the dorsal margin, so that the
cranial region of the process is craniodorsally oriented
Character 227: cranial extension of the craniodorsal region of the scapula (bearing
the coracoid facet), measured as a ratio between the distance from the coracoid joint
and the cranial end of the acromion process and the height between this and the
ventral apex of the glenoidal facet (Table H.7; Fig. H.13)
(0): short craniodorsal region, ratio less than 0.45 (sample mean ratio of 0.35)
(1): long craniodorsal region, ratio of 0.45 or greater (sample mean ratio of 0.53)
Character 228: development of the deltoid ridge (Fig. H.14)
(0): dorsoventrally narrow convexity limited to the proximal region of the
scapula, near the acromion process from which it develops, with a poorly demarcated
ventral margin
640
(1): dorsoventrally deep and craniocaudally long, with a well demarcated ventral
margin
Comments: the degree to which the ventral margin of the deltoid ridge reaches the
ventral margin of the scapula varied intraspecifically and did not vary according to age
(e.g., Brachylophosaurus canadensis, MOR 794 and 1071 specimens).
Humerus
Comparison of the distal condyles of a wide range of taxa from Iguanodon to

dervided hadrosaurids showed no difference in the mediolateral flaring of the dondyles.
Furthermore, postdepositional mediolateral compression was commonly observed in
humeri. Thus, character 84 of Horner et al. (2004) was excluded from this analysis.
Likewise, no difference was found in the development of each distal humeral
condyle. The ulnar condyle is larger in all taxa and thus a character conerning the
difference in size between the distal condyles of the humerus (Wagner, 2001) was not
included.
Character 106 of Prieto-Marquez et al. (2006) considered the position (centered
versus ventral) of the caudal tuberosity of the humerus. Observation of additional
specimens falsified the phylogenetic utility of this character. The tuberosity was found
between the center of the proximal half and the middle of the humerus in all outgroup and
ingroup taxa considered in this study.
Character 229: length of the deltopectoral crest of the humerus (measured as the
ratio between the proximodistal length of the crest and the proximodistal length of
the humerus) (Table H.8; Fig. H.15; modified from Godefroit et al., 2000, character
26)
(0): proximodistally short crest, ratio less than 0.48 (sample mean ratio of 0.44)
(2): very long crest, ratio greater than 0.55 (sample mean ratio of 0.59)
Character 230: lateroventral expansion of the deltopectoral crest of the humerus
(measured as the ratio between the width of the humerus across the distal fourth of
641
the deltopectoral crest and the width of the distal shaft at the point of maximum
curvature) (Table H.9; Fig. H.16; modified from Horner et al., 2004)
(0): poorly expanded deltopectoral crest, ratio less than 1.65 (sample mean ratio
of 1.53)
(2): very expanded deltopectoral crest, ratio greater than 1.90 (sample mean ratio
of 2).
Comments: no substantial increase in the lateroventral expansion of the
deltopectoral crest was found through ontogeny. For example, juveniles of Bactrosaurus
johnsoni, AMNH 6378 and AMNH 30239, showed a ratio of 1.66 and 1.76, respectively,
that are within the range of the 1.74 value measured for a much larger specimen
(SBDE/95E). The same applied to other taxa: Hypacrosaurus altispinus (juvenile TMP
82.10.1 with ratio of 2.01 and adult CMN 8501 with ratio of 2.06); Prosaurolophus
maximus (juvenile TMP 86.34.3 with ratio of 1.79 and adult TMP 84.1.1 with ratio of
1.80); and Corythosaurus (juvenile ROM 1947 with ratio of 1.98 and adult ROM 845
with ratio of 2.03).
Character 231: morphology of the distal margin of the deltopectoral crest of the
humerus, quantified as the angle between the cranial margin of the crest and the
proximalmost margin of the cranial curvature of the humeral shaft (Table H.10;
Fig. H.17; modified from Weishampel et al., 1993, character 37)
(0): wide and rounded margin, angle of 130º or greater (sample mean angle of
141º)
(1): wide and rubrectangular, angle between 129º and 115º (sample mean angle of
125)º
(2): narrow and subsquared, angle less than 115º (sample mean angle of 98)º.
Character 232: overall proportions of the humerus (measured as a ratio between the
total length and the width of the lateral surface of the proximal end of the humerus)
(Table H.11; Fig. H.18modified from Weishampel et al. 1993, character 36)
(0): relatively short and stocky humerus, ratio less than 4.25 (mean ratio of 3.85)
(2): relatively long and thin humerus, ratio greater than 4.90 (mean ratio of 5.40
642
Ulna
Character 233: length of the ulna relative to its dorsoventral thickness (measured at
mid-shaft) (Table H.12; Fig. H.19)
(0): ratio length/width less than 10
(1): ratio length/width equal or larger than 10
Character 234: ulnar length relative to humeral length (Table H.13; Fig. H.20;
Norman, 2002, character 47)
(0): ulna shorter than or as long as the humerus
(1): longer ulna, up to 20% longer than the humerus
(2): longer ulna, being more than 20% longer than the humerus
Comments: the ulna is nearly as long as the radius in the subadult specimen of
Lambeosaurus, AMNH 5340. This indicated that with age the ulna increases in length
more than the radius, at least in this taxon and possibly also in other lambeosaurines.
Manus
Character 111 of Prieto-Marquez (2006) concerning the relative lengths of

metacarpals II and III-IV was found to be phylogenetically uninformative. The length of
metacarpal II relative to that of metacarpals III and IV was measured in a number of
specimens spanning a variety of outgroup and ingroup taxa. The character was found
constant (ranging from 1.19 to 1.32 intraspecifically) and not significatively variable
among taxa. An exception was the mounted skeleton of Iguanodon atherfieldensis,
IRSNB 1551, where all three metacarpals have approximately the same length (that is, if
all those three elements displayed in the mounted manus are really metacarpals). In
contrast, BMNH R5764, referred to the same species, displayed a much shorter
metacarpal II as in all other studied Iguanodontoidea.
Another excluded character was the position of the proximal end of metacarpal III
relative to that of the proximal ends of metacarpals II and IV (Horner et al., 2004,
character 88; Prieto-Marquez et al., 2006, character 110, regarding distal offset). That
643
metacarpal III might appear proximally or distally offset relative to metacarpal II was due
to their difference in length. However, metacarpal III was not observed to be substantially
offset from metacarpal IV in the specimens sampled.
Character 90 of Horner et al. (2004) was excluded because wedge-shaped
penultimate phalanges in digits II and III were found in all taxa under study, including
Iguanodon bernissartensis (e.g., IRSNB 1534) and I. atherfieldensis (IRSNB 1551).
Character 235:composition of the carpus (Fig. H.21; adapted from Horner et al.,
2004, character 86)
(0): presence of fused ulnare, radiale, intermedium and distal carpals;
(1): number of carpal bones reduced to a maximum of two unfused elements.
Character 236: manual digit I (Fig. H.22; Norman, 2002, character 49)
(0): presence of metacarpal I and one ungual phalanx
(1): entire digit I absent
Character 237: elongation of the manus exemplified by elongation of metacarpals II
through IV, measured as the ratio between the length of metacarpal III and the
width of its mid-shaft (Table H.14; Fig. H.23; modified from Horner et al., 2004,
character 84)
(0): relatively short and blocky, ratio up to 5 (sample mean ratio of 4.25)
(1): relatively long and slender, ratio greater than 5 (sample mean ratio of 8.54)
Character 238: elongation of metacarpal V, so that it is more than twice as long as it
is proximally wide (Fig. H.24)
(0): absent
(1): present
Character 239: expansion of the proximal surface of metacarpal V (Fig. H.25)
(0): moderately expanded relative to the distal end
(1): greatly expanded relative to the distal end
Character 240; length/width proportions of manual phalanx III1 (Fig. H.26;
modified from Prieto-Marquez et al., 2006, character 114)
(0): proximodistally compressed, mediolaterally wider than it is long
(1): slightly longer proximodistally than it is wide mediolaterally
644
(2): very elongated, proximodistal length that is at least twice its mediolateral
width at the middle of its longitudinal axis
Character 241: shape of manual ungual II (Fig. H.27; Norman, 2002, character 53,
in part)
(0): claw-like
(1): hoof-like
Character 242: proximodistal length of manual palanx II1 relative to that of II2
(Fig. H.28; modified from Hai-lu You et al., 2003, chraracter 55)
(0): phalanx II1 less than three times longer than phalanx II2
(1): phalanx II1 three times or more longer than phalanx II2
Comments: this character may appear similar to the elongation of the manus
(character 237). However, the length of phalanx II2 is proportionally less than that of
phalanx II1, so that it would appear that phalanx II1 evolved to a greater rate.
645
state 0
A
state 1
Figure H.1. Character 214. Coracoid size relative to the length of the scapula. A, Iguanodon
bernissartensis (IRSNB 1534), lateral view. B, Gryposaurus latidens (AMNH 5465), medial view.
Abbreviations: cor = coracoid; scp = scapula.
646
Table H.1. Ratio (R) between the length of the lateral margins of the scapular and glenoid
facets of the coracoid for a sample of iguanodontoidean ornithopods. Included is also the
corresponding character state coding derived from K-means clustering analysis.

Cluster (State)
AUMP 2295 Lophorhothon atopus 1.10 1
MACN-RN 2 Secernosaurus koerneri 1.17 1
MPCA Salitral Moreno hadrosaurid 1.13 1
CUST K2 Hadrosauridae 0.68 2
USNM 2413 Hadrosauridae 0.76 2
TMP 92.36.983 Hadrosauridae 0.73 2
647
Ratio of the Width between Coracoid Articular Facets
20
1.4
10
BIC
0
−10
1.2
−20
E V
2 4 6 8
Number of Clusters
1.0
0.8
Br Co Ed Eda Gi Grl Grn Hy Igb Lh Ol Ou Pr Se Ts lc sal x
Taxon
Figure H.2. Character 215. Boxplot showing the distribution of the ratio between the length of the lateral margin of the facet for
the scapular articulation and the length of the lateral margin of the glenoid for a sample of iguanodontoidean ornithopods. The inserted smaller
graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph
shows how the two variables of this raio were measured, exemplified in an indeterminate hadrosaurid coracoid (TMP 80.29.1).
648
Table H.2. Character 216. Angle (A) between the lateral margins of the scapular and
glenoid facets of the coracoid for a sample of iguanodontoidean ornithopods. Included is
the corresponding character state coding derived from K-Means clustering.

Cluster (State)
LACM 17755 "Lambeosaurus" laticaudus 101 1
ROM 1947 Corythosaurus sp. 108 1
CUST K2 Hadrosauridae 118 0
UTEP P37.7.298 Hadrosauridae 92 1
TMP 1980-30-49 Hadrosauridae 97 1
AMNH 5272 Hadrosauridae 98 1
NMMNH P-22971 Hadrosauridae 102 1
USNM 2413 Hadrosauridae 103 1
IRNSB 1731 Iguanodon bernissartensis 121 0
IRNSB 1535 Iguanodon bernissartensis 125 0
AUMP 2295 Lophorhothon atopus 126 0
MPCA Salitral Moreno hadrosaurid 124 0
649
130
−230
Angle between Coracoid Articular Facets
−240
BIC
−250
120
−260
−270
E V
2 4 6 8
Number of Clusters
110
100
Ba Br Co Ed Eda Edr Grl Hy Igb Lh Ol Ou Se Ts lc sal x
Taxon
Figure H.3. Character 216. Boxplot showing the distribution of the angle between the lateral margins of the scapular facet and the glenoid
for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for various
numbers of clusters in which the data can be classified. The photograph shows how the two variables of this raio were measured,
exemplified in an indeterminate hadrosaurid coracoid (TMP 80.29.1).
650
state 0 state 0 state 0 state 1
5 cm 55 cm 10 cm
10 cm
A B C D
5 cm
5 cm 5 cm 5 cm
E F G H
Figure H.4. Character 217. Morphology of the craniomedial margin (arrows) of the coracoid. Specimens shown in lateral view.
A, Lophorhothon atopus (AUMP 2295). B, Iguanodon sp. (BMNH R6471). C, I. bernissartensis (IRSNB 1534). D, “Lambeosaurus laticaudus“
(LACM 17715). E, Secernosaurus koerneri (MACN-RN 2). F, cf. Hypacrosaurus stebingeri (MOR 355-8-28-5-12). G, Olorotitan ararhensis
(AEHM 2/845). H, Lambeosaurine indeterminate (USNM 11950).
651
Table H.3. Character 218. Ratio (R) between the dorsoventral length and the breadth of
the ventral process of the coracoid for a sample of iguanodontoidean ornithopods.

Cluster (State)
IRSNB SBDE/95 Bactrosaurus johnsoni 0.54 0
TMP 90.104-1 Brachylophosaurus canadensis 0.87 2
MPCA Salitral Moreno hadrosaurid 0.74 1
CUST K2 Hadrosauridae 0.58 0
NMMNH P-22971 Hadrosauridae 0.75 1
652
20
1.0
Ratio of the Depth/Width of the Coracoid Ventral Process
10
0
BIC
0.9
−10
−20
−30
E V
0.8
2 4 6 8
Number of Clusters
0.7
0.6
0.5
Ba Br Co Ed Eda Edr Gi Grl Grn Hy Igb Lh Ol Ou Pr Se Ts lc sal x
Taxon
Figure H.5. Character 218. Boxplot showing the distribution of the ratio between the dorsoventral depth and the breadth of the ventral
process of the coracoid for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC)
for various numbers of clusters in which the data can be classified. The photograph shows how the two variables of this raio were measured,
exemplified in an indeterminate hadrosaurid coracoid (TMP 92.36.983).
653
A B C
D E F
Figure H.6. Character 219. Degre of curvature of the ventral process of the coracoid (arrows). Specimens shown in lateral view.
A, Iguanodon sp. (BMNH R118). B, Ouranosaurus nigeriensis (cast of GDF 300). C, Brachylophosaurus canadensis (CMN 8893).
D, B. canadensis (UCMP 130139). E, Edmontosaurus sp. (MOR 601). F, cf. Hypacrosaurus sp. (MOR 609-89-177).
654
state 0
A
state 0
state 1
B
Figure H.7. Character 220. Lateral profile of the dorsal margin of the scapula. A, Bactrosaurus
johnsoni (AMNH 6553), lateral view. B, Brachylophosaurus canadensis (MOR 1071-7-18-98-298),
lateral view.
655
depth of the cranial end of the scapula, for a sample of iguanodontoidean ornithopods.
Included is the corresponding character state coding derived from K-Means clustering
analysis.

Cluster (State)
DMNN 1943 Edmontosaurus annectens 4.70 1
IRNSB 1551 Iguanodon atherfieldensis 3.10 0
MPC-D 100/706 Saurolophus angustrirostris 4.22 1
656
Table H.4—continued.

Cluster (State)
USNM 2414 Edmontosaurus 4.20 1
ANSP 16969 Hadrosauridae 4.64 1
657
5.5
−80
−90
BIC
−110
Ratio of the Length of the Scapula
5.0
−130
2 4 6 8
Number of Clusters
4.5
4.0
3.5
Ba Br Co Eda Edr Grl Grn Hy Iga Igb Lh Lml Lmm Ma Ol Ou Pac Paw Pr Saa Sao Se Sh Ta Ts x
Taxon
Figure H.8. Character 221. Boxplot showing the distribution of the ratio between the dorsoventral depth of the proximal region of the
scapula and its craniocaudal length for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information
criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph shows how the two variables of this raio were
measured, exemplified in a scapula of Secernosaurus koerneri (MACN-RN 142).
658
state 0
state 1
Figure H.9. Character 222. Absence and presence of a convexity along the ventral margin of
the distal blade of the scapula. A, Secernosaurus koerneri (MACN-RN 142), lateral view.
B, Corythosaurus intermedius (ROM 845), lateral view.
659
Table H.5. Character 223. Ratio (R) between the maximum dorsoventral width of the
scapular blade and the proximal region of the scapula for a sample of iguanodontoidean
ornithopods. Included is the corresponding character state coding K-Means clustering.

Cluster (State)
ZPAL MgD-I 161 cf. Saurolophus angustirostris 1.10 1
YPM 1190 Claosaurus agilis 1.04 1
MPC-D 100/706 Saurolophus angustirostris 1.16 1
660

cluster (State)
GMV 1780 Shantungosaurus australis 0.80 0
ANSP 16969 Lambeosaurinae 1.15 1
661
1.4
Ratio of the Distal Expansion of the Scapular Blade
1.2
1.0
10
0
−50 −40 −30 −20 −10
BIC
0.8
2 4 6 8
Number of Clusters
0.6
Am Ba Br Cl Coi Cos Eda Edr Gi Grl Grn Hy Ig Iga Igb Lh Lml Lmm Ma Ol Ou Pac Paw Saa Sao Se Sh Ta Ts cfSaa x
Taxon
Figure H.10. Character 223. Boxplot showing the distribution of the ratio between the dorsoventral depth of the distal end of the scapular
blade and the depth of the proximal region of the scapula for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the
Bayesian information criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph shows how the two
variables of this raio were measured, exemplified in a scapula of Lambeosaurus lambei (ROM 1218).
662
Table H.6. Character 224. Ratio (R) between the dorsoventral depth of the proximal
constriction and the dorsoventral depth of the cranial end of the scapula, for a sample of
Specimen Taxon R K-Means cluster

(State)
CUST JII1398 Charonosaurus jiayinensis 0.69 1
BMNH 2196A Iguanodon sp. 0.51 0
663

(State)
MPCA unrecorded no. Salitral Moreno hadrosaurid 0.69 1
664
100
0.75
80
Ratio of the Proximal Constriction of the Scapula
BIC
60
0.70
40
20
0.65
2 4 6 8
Number of Clusters
0.60
0.55
0.50
0.45
Ba Br Ch Cl Co Eda Edr Gi Grl Grn Hy Hys Ig Iga Igb Lh Lml Lmm Ma Ol Ou Pac Paw Pr Saa Sao Se Sh Ta Ts cfSaa sal x
Taxon
Figure H.11. Character 224. Boxplot showing the distribution of the ratio between the dorsoventral width of the proximal constriction
and the dorsoventral depth of the cranial end of the scapula for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the
Bayesian information criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph shows how the two
variables of this raio were measured, exemplified in a scapula of Gryposaurus latidens (AMNH 5465).
665
A B C D E
F G H I J
Figure H.12. Characters 225 and 226. Morphology and orientation of the acromion process (arrows) of the scapula. Specimens shown in
lateral view. A, Iguanodon atherfieldensis (IRSNB 1551). B, Parasaurolophus walkeri (ROM 768). C, Tanius sinensis (PMU R241).
D, Bactrosaurus johnsoni (AMNH 6553). E, Corythosaurus intermedius (CMN 8704). F, Charonosaurus jiayinensis (CUST JII1398).
G, Lambeosaurinae (? Hypacrosaurus stebingeri, TMP 87.79.220; juvenile specimen). H, Maiasaura peeblesorum (ROM 44770).
I, Brachylophosaurus canadensis (UCMP 130139). J, Edmontosaurus annectens (SM R4036). Exemplars A through G document state 0
of character 225; exemplars H through J document state 1 of character 225. Specimens A through C document state 0 of character 225;
specimens D through G document state 1 of character 226.
666
Table H.7. Character 227. Ratio (R) between the maximum craniocaudal length of the
craniodorsal region of the scapula (cranial to the acromion process) and the dorsoventral
height from the cranial end of the acromion process to the ventral apex of the glenoidal
facet, for a sample of iguanodontoidean scapulae. Included is the corresponding character
state coding derived from K-Means clustering.

(State)
CUST JII1398 Charonosaurus jiayinensis 0.35 0
667

(State)
668
0.60
100
Ratio of the Craniodorsal Extension of the Proxima Scapula
90
0.55
80
BIC
70
0.50
60
50
40
E V
0.45
2 4 6 8
Number of Clusters
0.40
0.35
0.30
Br Ch Co Eda Edr Grl Grn Hy Hys Iga Igb Lh Lml Ma Ol Ou Pac Paw Ph Pr Sao Se Sh Ta Ts Ve cfSaa x
Taxon
Figure H.13. Character 227. Boxplot showing the distribution of the ratio between the distance from the coracoid joint and the cranial end
of the acromion process and the height between this and the ventral apex of the glenoidal facet for a sample of iguanodontoidean ornithopods.
The photograph shows how the two variables of this raio were measured, exemplified in a scapula of Secernosaurus koerneri (MACN-RN 142).
669
state 0
A
state 1
Figure H.14. Character 228. Development of the deltoid ridge (arrows) of the scapula.
A, Hypacrosaurus stebingeri (MOR 549), lateral view. B, Secernosaurus koerneri (MACN-RN 146),
lateral view.
670
Table H.8. Character 229. Ratio (R) between the proximodistal length of the
deltopectoral crest and the proximodistal length of the humerus for a sample of

(State)
SBDE 95E/24 Bactrosaurus johnsoni 0.46 0
AMNH 5893 cf. Parasaurolophus 0.60 2
TMP 1992.53.21 cf. Parasaurolophus 0.62 2
CUST JVIII1247 Charonosaurus jiayinensis 0.54 1
MNHN AMN 18 Edmontosaurus sp. 0.51 1
MOR 478 Gryposaurus latidens 0.51 1
ANSP 10005 “Hadrosaurus foulkii" 0.46 0
671

(State)
MOR 758 Maiasaura peeblesorum 0.51 1
MPCA Hadrosauridae Salitral Moreno 0.56 2
672
140
0.60
120
BIC
Ratio of the Length of the Deltopectoral Crest
100
80
0.55
2 4 6 8
Number of Clusters
0.50
0.45
Am Ba Br Ch Coi Cos Ed Eda Edr Gi Grl Grn Hd Hy Ig Iga Igb Kn Lh Ma Ol Ou Pac Paw Pr Saa Sh Ta Ts cfPa lc sal x
Taxon
Figure H.15. Character 229. Boxplot showing the distribution of the ratio between the proximodistal length of the deltopectoral crest
and the proximodistal length of the humerus for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian
information criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph shows how the two variables
of this raio were measured, exemplified in a humerus of Gryposaurus latidens (AMNH 5465).
673
Table H.9. Character 230. Ratio (R) between the width of the proximal half of the
humerus (across the distal fourth of the deltopectoral crest) and the minimum breadth of
the distal shaft, for a sample of iguanodontoidean ornithopods. Included is the

(State)
TMP 1992.53-21 cf. Parasaurolophus sp. 2.00 2
AMNH 5893 cf. Parasaurolophus sp. 2.04 2
TMP 82.10.1 Hypacrosaurus altispinus 2.01 2
MOR 758 Maiasaura peeblesorum 1.60 0
674

(State)
PIN 2232-2/8 Probactrosaurus gobiensis 1.53 0
MPCA Hadrosauridae Salitral Moreno 1.71 1
ANSP 10005 “Hadrosaurus foulkii” 1.62 0
675
2.1
2.0
Ratio of the Expansion of the Deltopectoral Crest
10
1.9
0
−10
BIC
1.8
−30
−50
E V
1.7
2 4 6 8
Number of Clusters
1.6
1.5
1.4
Ba Br Ch Co Ed Eda Edr Gi Grl Grn Hd Hy Hys Ig Iga Igb Lh Ma Ol Ou Pac Paw Pb Pr Saa Sao Sh Ta Ts cfPa lc sal x
Taxon
Figure H.16. Character 230. Boxplot showing the distribution of the ratio between the width of the humerus across the distal fourth of the
deltopectoral crest and the width of the distal shaft at the point of maximum curvature for a sample of iguanodontoidean ornithopods.
The photograph shows how the two variables of this raio were measured, exemplified in a humerus of Brachylophosaurus canadensis (CMN 8893).
676
Table H.10. Character 231. Angle (A) between the cranial margin of the deltopectoral
crest and the proximalmost margin of the cranial curvature of the humeral shaft, for the
specimens sampled to study the variation of this character Included is the corresponding

(State)
SBDE 95E Bactrosaurus johnsoni 127 1
SBDE 95E/24 Bactrosaurus johnsoni 136 0
UCMP 130139 Brachylophosaurus canadensis 127 1
USNM 2414 Edmontosaurus sp. 125 1
BMNH R3640 cf. Edmontosaurus sp. 126 1
AMNH 5893 cf. Parasaurolophus sp. 94 2
TMP 1992.53-21 cf. Parasaurolophus sp. 98 2
CUST JVIII1247 Charonosaurus jiayinensis 100 2
YPM 1190 Claosaurus agilis 144 0
MNHN AMN 18 Edmontosaurus sp. 126 1
MOR 478 Gryposaurus latidens 126 1
AMNH 5350 Gryposaurus notabilis 100 2
MOR 758 Maiasaura peeblesorum 123 1
AEHM 2/845 Olorotitan arharensis 99 2
677

(State)
PIN 2232-2/8 Probactrosaurus gobiensis 139 0
MPC-D 100/706 Saurolophus angustirostris 106 2
GMV 1780 Shantungosaurus giganteus 94 2
PMU R235 Tanius sinensis 125 1
UTEP P32.7.047 Hadrosauridae 101 2
ANSP 10005 “Hadrosaurus foulkii” 126 1
678
160
−420
Craniodistal Angle of the Deltopectoral Crest
−430
BIC
−440
140
−450
E V
2 4 6 8
Number of Clusters
120
100
Am Ba Br Ch Cl Co Ed Eda Edr Gi Grl Grn Hd Hy Hys Ig Iga Igb Lh Ma Ol Ou Pac Paw Pb Pr Saa Sao Sh Ta Ts cfEd cfPa lc x
Taxon
Figure H.17. Character 231. Boxplot showing the distribution of the angle between the cranial margin of the crest and the proximalmost
margin of the cranial curvature of the humeral shaft for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian
information criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph shows how this angle was measured,
exemplified in a humerus of Tanius sinensis (PMU R235).
679
Table H.11. Character 232. Ratio (R) between the total length and the width of the lateral
surface of the proximal end of the humerus, for the specimens sampled to study the
variation of this character. Included is the corresponding character state coding derived

(State)
BMNH R3640 cf. Edmontosaurus sp. 4.49 1
AMNH 5893 cf. Parasaurolophus sp. 3.95 0
TMP 1992.53-21 cf. Parasaurolophus sp. 4.16 0
680

(State)
681
6.0
−70
−80
−90
BIC
Robustness Ratio of the Humerus
−100
5.5
−110
−120
E V
2 4 6 8
Number of Clusters
5.0
4.5
4.0
3.5
Ba Br Ch Coi Ed Eda Edr Gi Grl Grn Hy Hys Ig Iga Igb Lh Ou Pac Paw Pr Saa Ta Ts cfEd cfPa lc x
Taxon
Figure H.18. Character 232. Boxplot showing the distribution of the ratio between the total length of the humerus and the width of
the lateral surface of the proximal end of the element for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the
Bayesian information criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph shows how this angle
was measured, exemplified in a humerus of Tsintaosaurus spinorhinus (IVPP V725).
682
Table H.12. Character 233. Ratio (R) between the craniocaudal length and the
dorsoventral width (at mid-length) of the ulna, for the exemplars used to investigate the
distribution and pattern of variation of this character. Included is the corresponding
character state coding derived from K-Means clustering.

cluster (state)
MOR 794 Brachylophosaurus candensis 14.6 1
CMN 8893 Brachylophosaurus candensis 12.9 1
TMP 90.104.1 Brachylophosaurus candensis 13.2 1
TMP 92.53.21 cf. Parasaurolophus sp. 12.3 1
CUST JIII1715 Charonosaurus jiayinensis 13.4 1
MNHN AMN 18 Edmontosarus sp. 10.9 1
USNM 3814 Edmontosaurus anenctens 9.8 0
CMN 8399 Edmontosaurus anenctens 13.9 1
DMNH 1943 Edmontosaurus anenctens 10.8 1
YPM 2182 Edmontosaurus anenctens 9.8 0
AMNH 5730 Edmontosaurus anenctens 9.7 0
PIN 2232/10 Probactrosaurus gobiensis 14.9 1
683

cluster (state)
684
15
−190
−210
14
BIC
−230
Length/width Ratio of the Ulna
13
−250
E V
2 4 6 8
Number of Clusters
12
11
10
9
8
Br Ch Coi Cos Ed Eda Edr Gi Grl Grn Hy Hys Iga Igb Lh Lml Lmm Ma Pac Paw Pb Pr Saa Sao Sh Ts cfPa x
Taxon
Figure H.19. Character 233. Boxplot showing the distribution of the length of the ulna relative to its dorsoventral thickness (measured
at mid-shaft) for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for
various numbers of clusters in which the data can be classified. The photograph shows how the two variable of this ratio were measured,
exemplified in a ulna of Parasaurolopus walkeri (ROM 768).
685
Table H.13. Character 234. Ratio (R) between the length of the ulna and the length of the
humerus for a sample of iguanodontoidean ornithopods. Included is the corresponding

Cluster (State)
686
10
0
1.4
Ratio of the Ulnar Length/Humeral Length
−50 −40 −30 −20 −10

BIC
1.3
2 4 6 8
Number of Clusters
1.2
1.1
1.0
0.9
Br Cl Coi Cos Eda Edr Grn Hd Hy Iga Igb Lh Lmm Ou Pac Paw Pr Sao
Taxon
Figure H.20. Character 234. Boxplot showing the distribution of the ratio between the length of the ulna and the length of the humerus
of clusters in which the data can be classified. The photograph shows how the two variable of this ratio were measured, exemplified in a forelimb
of Brachylophosaurus canadensis (MOR 794).
687
state 0
A
state 1
Figure H.21. Character 235. Composition of the carpus (rectangle and arrows). A, Iguanodon
bernissartensis (IRSNB 1535), dorsal view. B, Brachylophosaurus canadensis (MOR 794),
proximolateral view.
688
state 0
A
state 1
Figure H.22. Character 236. Absence and presence of manual digit I. A, Iguanodon
bernissartensis (IRSNB 1535), dorsal view. B, Hadrosauridae indeterminate (TMP 84.36.29).
689
Table H.14. Character 237. Ratio (R) between the total length and the width, at mid-shaft,
of metacarpal III for the specimens sampled to study the variation of this character.

(state)
AEHM unrecorded no. Amurosaurus riabinini 8.25 1
SBDE 95E Bactrosaurus johnsoni 7.10 1
UTEP P37.7 Hadrosauridae 7.95 1
FMNH P13423 Secernosaurus koerneri 8.25 1
690
9
Length/width Ratio of Metacarpal III
8
7
−110
−120
6
BIC
−130
−140
5
−150
E V
2 4 6 8
Number of Clusters
4
Am Ba Br Coi Cos Ed Eda Edr Grl Grn Hy Igb Lh Lml Lmm Ma Pac Paw Pr Se Ts cfCo cfPa x
Taxon
Figure H.23. Character 237. Boxplot showing the distribution of the ratio between the length and the width at mid-shaft of metacarpal III for
a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of
clusters in which the data can be classified. The photograph shows how the two variable of this ratio were measured, exemplified in a
metacarpal III of Tsintaosaurus spinorhinus (IVPP V725).
691
state 0 state 1
A B
Figure H.24. CHaracter 238. Proximodistal elongation of metacarpal V.

A, Bactrosaurus johnsoni (SBDE 95E). B, Maiasaura peeblesorum (ROM 44770).
692
state 0
state 1
Figure H.25. Character 239. Expansion of the proximal surface of metatarsal V.

A, Iguanodon bernissartensis (IRSNB 1535). B, Parasaurolophus walkeri (ROM 768).
693
A B C
Figure H.26. CHaracter 240. Length/width proportions of manual phalanx III1. A, Parasaurolophus walkeri (ROM 768).
B, Corythosaurus intermedius (ROM 845). C, Maiasaura peeblesorum (ROM 44770).
694
state 0
A
state 1
Figure H.27. Character 241. Shape of manual ungual II. A, Iguanodon

bernissartensis (IRSNB 1534). B, Brachylophosaurus canadensis (MOR 794).
695
state 0
A
state 1
Figure H.28. Character 242. Proximodistal length of manual palanx II1 relative to that of II2.
A, Iguanodon atherfieldensis (IRSNB 1551). B, Lambeosaurus lambei (CMN 8703).
696
APPENDIX I
PELVIC CHARACTERS
Ilium
Character 243: angle of ventral deflection of the preacetabular process of the ilium
(Table I.1; Fig. I.1; modified from Suzuki et al., 2004, character 69)
(1): angle of 150º or less (sample mean angle of 143º)
Character 244: length of the preacetabular process of the ilum, measured as the
ratio between the length of the process and the craniocaudal length of the central
body of the ilum (Table I.2; Fig. I.2)
(0): extremely long preacetabular process, twice or more the length of the central
plate of the ilium
(1): preacetabular process less than twice as long as the central plate of the ilium
Comments: both the model clustering analysis and the boxplot distribution
resulted in no clear pattern of variation in all taxa underconsideration, with the exception
of Ouranosaurus. For the latter, the length of the ilium is autapomorphic (ratio of 2.15) in
the context of the sampled taxa. Still, the character was retained here for the sake of
documentation of this feature and to allow the scoring of future discoveries where the
condition may be as in Ouranosaurus, or at least different from state (1).
In the few ontogenetic series that I could examine, the length of the preacetabular
process increased slightly with age (e.g., subadults of Brachylophosaurus have a ratio of
1.69 for MOR 1071-7-15-98-215 and 1.74 for MOR 1071-7-10-99-50, compared to the
1.80 in the adult MOR 794 and 1.81 in MOR 1071-8-2-98-469). Similarly, the juvenile
Lambeosaurus AMNH 5340 has a ratio of 1.60, in contrast with the adult ROM 1218 hat
had a ratio 1.77. However, the subadult Saurolophus angustirostris ZPAL MgD-I 159
had a ratio of 1.52, which did not differ much from the 1.54 in S. osborni AMNH 5220.
GDA was also conducted to explore whether the lateral profile of the
preacetabular process contained a phylogenetic signal. Unfortunately, not pattern of
variation yielding phylogenetically informative characters was found (Table I.3; Fig. I.3)
697
Character 245: dorsoventral depth of the proximal region of the preacetabular
process (measured as a ratio between this and the dorsoventral distance between the
pubic peduncle and the dorsal margin of the ilium) (Table I.4; Fig. I.4)
(0): shallow, less than half the depth of the cranial central blade, ratio less than
(1): approximately as deep as the cranial central blade depth, ratio between 0.50
and 0.55 (sample mean ratio of 0.51)
(2): deeper than half the depth of the cranial central blade, ratio greater than 0.55
Character 246: dorsoventral depth of the central blade of the ilium (expressed as a
ratio between this and the distance between the pubic peduncle and the caudodorsal
prominence of the ischial peduncle) (Table I.5; Fig. I.5)
(0): ratio of 0.80 or greater (sample mean ratio of 0.90)
(1): ratio less than 0.80 (sample mean ratio of 0.71)
Character 247: position of the ventral-most margin of the supraacetabular process
relative to the caudoventral margin of the lateral ridge of caudal protuberance of
the ischial peduncle of the ilium (Brett-Surman and Wagner, 2007) (Fig. I.6)
(0): apex located caudodorsally
(1): apex located craniodorsally.
Character 248: development of the lateroventral projection of the supraacetabular
process of the ilium (Fig. I.7; modified from Horner et al., 2004, character 91)
(0): forms a longitudinal and continuous “swelling” or reflected border along the
dorsal margin of the central blade and the proximal region of the postacetabular process,
with a depth up to 25% the depth of the ilium
(1): projected lateroventrally at least 25% (but less than half) the depth of the
ilium
(2): projects lateroventrally between half and three quarters of the dorsoventral
depth of the ilium
(3): projects lateroventrally to overlap totally or at least half of the lateral ridge of
the caudal prominence of the ischial peduncle.
698
Character 249: craniocaudal breadth of the supraacetabular process, measured as
the ratio between the breadth of the process across its dorsal region and the
craniocaudal length of the central iliac blade from the caudal ischial peduncle to the
pubic one (Table I.6; Fig. I.8)
(0): craniocaudally wider than the central blade of the ilium, ratio greater than
(1): craniocaudally broad, ratio between 0.70 and 0.85 (sample mean of 0.73)
(2): slightly broader than half the length of the central iliac blade, ratio between
0.55 and 0.69 (sample mean ratio of 0.62)
(3): short, ratio less than 0.55 (sample mean ratio of 0.48)
Character 250: symmetry of the lateral profile of the supraacetabular process (Fig.
I.9)
(0): asymmetrical, with a caudally skewed lateral profile
(1): symmetrical or with a slightly caudally skewed profile.
Character 251: morphology of the lateroventral margin of the supraacetabular
process (Fig. I.10)
(0): craniocaudally sinuous
(1): widely arched
(2): U or V-shaped
(3): subrectangular, with a shallow notch that divides the ventral margin in two
poorly demarcated lobes
Character 252: demarcation of the caudodorsal margin of the lateroventral rim of
the supraacetabular process (Fig. I.11 and I.12)
(0): the caudodorsal margin is a well-defined ridge that is continuous with the
dorsal margin of the proximal region of the postacetabular process
(1): the caudodorsal margin is poorly defined and appears discontinuous with the
dorsal margin of the proximal region of the postacetabular process due to the lack of a
well demarcated caudodorsal ridge
Character 253: morphology of the pubic peduncle of the ilium (modified from
Sereno, 1986; Horner et al., 2004; character 92) (Table I.7; Figs. I.13 and I.14)
699
(0): relatively large and dorsoventrally deep (longer than wide), subconical, with a
proximal region that is only slightly craniocaudally wider than the distal end of the
process
(1): relatively shorter (wider or as wide as long) and triangular, with a proximal
region that is much craniocaudally wider than the distal end
Character 254: morphology of the ischial peduncle of the ilium (modified from
Godefroit et al., 2001, character 30) (Table I.8; Figs. I.15 and I.16)
(0): formed by a single and large, oval ventral protrusion
(1): composed of a large and oval ventral protrusion and by a smaller,
caudodorsally located prominence emergin from the caudodorsal ridge
(2): formed by two protrusions of similar size, the caudal-most one located slighty
caudodorsally
Character 255: ratio between the craniocaudal length of the postacetabular process
and the craniocaudal length of the central blade of the ilium (Table I.9; Fig. I.17)
(0): short postacetabular process, ratio up to 0.80 (sample mean ratio of 0.70)
(1): postacetabular process nearly as long as the central blade, ratio greater than
0.80 but less than 1.1 (sample mean ratio of 0.90)
(2): postacetabular process substantially longer than the central blade, ratio of 1.1
or greater (sample mean ratio of 1.23)
Character 256: brevis shelf at the base of the postacetabular process of the ilium
(modified in part from Horner et al., 2004, character 93; Fig. I.18)
absent (0)
present (1)
Comment: a number of hadrosaurid taxa showed a structure that could only be
compared to the brevis shelf of non-hadrosaurid iguanodontoideans. The structure present
in the hadrosaurids Secernosaurus koerneri, the Salitral Moreno OTU, Hypacrosaurus
altispinus (AMNH 5204) and Velafrons coahuilensis differed from that in non-
hadrosaurid taxa in that it appeared to have originated from dorsomedial rotation of the
postacetabular process, rather than being the ventral surface of a mediolaterally thickened
process. However, no such distinction was included in the definition and coding of this
structure in order to let the phylogenetic analysis infer whether those two types of shelves
700
were homologous or not. The results of the phylogenetic analysis indicated that the
brevis shelf in hadrosaurids was evolved independently form that in outgroup
iguanodontoideans.
Character 257: medioventral ridge on the medial side of the postacetabular process,
crossing the bone surface from the proximoventral to the caudodorsal margins and
orientation of the brevis shelf (Fig. I.19)
(0): absent or presence of a faint ridge
(1): well-defined ridge bounding the medial margin of the brevis shelf; the latter
faces medioventrally and in medial view appears restricted to the caudal region of the
postacetabular process
(2): well-defined ridge forming the medial margin of a medioventrally-facing
shelf, with a postacetabular process that is progressively expanded mediolaterally
towards the caudal end
(3): well-developed, oblique and expanded flange forming the medial margin of
an extensive brevis shelf that faces more ventrally than medially
Character 258: craniocaudally-oriented median ridge on the laterodorsal surface of
the postacetabular process (Fig. I.20)
(0): absent
(1): present
Character 259: geometry of the lateral profile of the postacetabular process of the
ilium (modified from Horner et al., 2004, character 93; Table I.10; Figs. I.21 and
I.22)
(0): the ventral margin converges caudodorsally to meet the horizontal dorsal
margin, forming a tapering caudal end and producing a triangular lateral profile of the
process
(1): dorsal and ventral margins parallel or slightly convergent, forming a distinct
(rectangular or subcircular) caudal margin
Character 260: orientation of the dorsal margin of the postacetabular process
relative to the acetabular margin (Fig. I.23)
(0): horizontal dorsal marging, parallel or nearly parallel to the acetabular margin
701
(1): caudodorsally oriented dorsal margin, rising dorsally relative to the acetabular
margin
Character 261: position of the medial sacral ridge within the medial surface of the
central plate of the ilium (Fig. I.24)
(0): ridge well separated ventrally from the dorsal margin of the ilium, set
between 50% and 30% the dorsoventral depth of the central blade;
(1): ridge located more dorsally, closer to the dorsal margin, within the dorsal
third (less than 30% the depth) of the central blade
Character 262: orientation of the medial sacral ridge of the ilium (Fig. I.25)
(0): craniocaudally directed, parallel to the dorsal margin of the ilium, and ending
caudal to the level of the ischial peduncle, well into the proximal region of the
postacetabular process;
(1): diagonal, cranioventrally to caudodorsally oriented, concave ventrally and
converging with the dorsal margin at the level of the ischial peduncle
Character 263: lateral profile of the dorsal or laterodorsal margin of the ilium (Fig.
I.26; modified from Horner et al., 2004, character 100)
(0): straight or slightly convex
(1): distinctly depressed over the supraacetabular process and dorsally bowed over
the proximal region of the preacetabular process
Pubis
Character 264: orientation of the dorsoventral expansion of the prepubic process

(Fig. I.27)
(0): the dorsal region of the expansion is more expanded than the ventral region,
so that distally the process is dorsally-directed
(1): the ventral region is more expanded than the dorsal region, so that the distal
expansion is ventrally-directed
Character 265: geometry of the dorsoventral expansion of the prepubic process of
the pubis (in lateral or medial views) (modified in part from Wagner, 2001) (Table
I.11; Figs. I.28-I.30)
702
(0): circular to oval expansion, extensive and convex ventral margin
(1): subsquared distal dorsal margin, expansion dorsoventrally taller than
cranioventrally long, very pronounced proximal dorsal concavity and nearly straight
distal ventral margin
(2): ellipsoidal, expansion craniocaudally longer than dorsoventrally tall, well
pronounced concavities of the dorsal and ventral proximal margins
(3): oval expansion, dorsoventrally taller than craniocaudally long, well
pronounced concave profiles of dorsal and ventral proximal margins
(4): rectangular, craniocaudally longer than dorsoventrally tall, nearly straight
profiles of the dorsal and ventral proximal margins
Comments: the shape of the prepubic process of the pubis remained constant
throughout ontogeny in the few ontogenetic series recorded (e.g., lambeosaurine taxon,
cf. Hypacrosaurus stebingeri, juveniles MOR 548 and adult 549; Maiasaura
peeblesorum, subadults MOR 547 and adult ROM 44770; Edmontosaurus spp., subadults
LACM 7241, LACM 23504 and adults CMN 8399, DMNH 1943, SM 3046, CMN 2289,
etc., to name just a few examples).
It was also observed that, within the characteristic “fist-like” shape of the
lambeosurine pubis, Corythosaurus (e.g., AMNH 5240 and possibly AMNH 3971) and
Lambeosaurus (e.g., TMP 66.4.1) showed a more elongated prepubic constriction relative
to that of other forms such as Hypacrosaurus altuspinus (e.g., CMN 8501), H. stebingeri
(MOR 549) and “Lambeosaurus” laticaudus (e.g., LACM 26874).
Character 266: depth of the dorsoventral expansion of the distal region of the
prepubic process relative to the width of the acetabular margin of the pubis (Fig.
I.31)
(0): distal expansion as wide as or shallower than width of the acetabular margin
(1): dorsoventral expansion deeper than the acetabular margin
Character 267: craniocaudal length of the proximal constriction of the prepubic
process of the pubis relative to length of the dorsoventral expansion (Fig. I.32;
(0): constriction longer than the dorsoventral expansion, that is restricted to the
distal region of the process
703
(1): constriction and distal expansion have approximately the same length
(2): constriction slightly shorter than the dorsoventral expansion, that begins at the
proximal region of the process
Character 268: relative position of maximum concavity of the dorsal and ventral
margins of the prepubic process of the pubis (Fig. I.33)
(0): maximum ventral concavity achieved adjacent to the proximal region of the
postpubic process, maximum dorsal concavity located further distally
(1): maximum ventral concavity located ventral to or slightly caudal to the
maximum dorsal concavity
Character 269: morphology of the acetabular margin, ventral to the lateral edge of
the iliac peduncle (Fig. I.34)
(0): the lateral margin of the iliac peduncle extends ventrally forming a prominent
ridge that merges with the proximal region of the ischial peduncle
(1): the lateral margin of the iliac peduncle progressively disappears ventrally into
the lateral surface of the region adjacent to the acetabular margin
Character 270: pubic obturator foramen (Fig. I.35; modified from Horner et al.,
2004, character 97)
(0): present, proximal postpubic ramus has a caudodorsally oriented short process
that contacts totally or partially with the ischial peduncle to form the obturator foramen
(1): absent, proximal postpubic ramus lacks a dorsocaudally oriented process
Comments: although Bactrosaurus has an obturator “gutter” (Godefroit et al.,
1998), this is not enclosed to form a foramen.
Character 271: length/width ratio of the ischial peduncle of the pubis (Table I.12;
Fig. I.36)
(0): very short ischial peduncle, ratio less than 1.85 (sample mean ratio of 1.5)
(1): ratio ranging from 1.85 to less than 3 (sample mean ratio of 2.4)
(2): very long, ratio of 3 or more (with a sample mean ratio of 4)
Character 272: lateroventral protuberance on the proximal region of the ischial
peduncle of the pubis (Fig. I.37)
(0): absent or very faintly developed
(1): present
704
Character 273: depth/width proportions of the the iliac peduncle of the pubis (Fig.
I.38)
(0): craniocudally broader than dorsoventrally tall
(1): taller dorsoventrally than broad craniocaudally
Character 274: total length of the pubis, as the ratio between the craniocaudal
distance from the acetabular margin to the distal margin of the prepubic process
and the distance from the dorsal margin of the iliac peduncle and the ventral
margin of the proximal postpubic shaft (Table I.13; Fig. I.39)
(0): short, ratio less than 2.70 (sample mean raio of 2.30)
(1): long, ratio between 2.70 and 3 (sample mean ratio of 2.84)
(2): very long, ratio greater than 3 (sample mean ratio of 3.53)
Ischium
A few characters used by previous authors were not found phylogenetically

informative. One of these features was the position of the obturator process on the ischial
shaft, that Sereno (1986) reported as proximally positioned in hadrosaurs. Sereno (1986)
also diagnosed hadrosaurs (his Hadrosauroidea) based on the presence of a deeper
obturator notch between the obturator process and pubic peduncle of the ischium.
Actually, when found complete, the notch is as deep in non-hadrosaur iguanodontoideans
as it is hadrosaurs.
Another uninformative character was the shape of ischial shaft in lateral view
(strongly curved downward or nearly straight; Horner et al. 2004, character 98). This
character was excluded because of substantial intraspecific variation (e.g., Iguanodon
bernissartensis showed a curved shaft in IRSNB 1534, but straight ones in IRSNB 1535
and 1726. In addition, Prieto-Marquez et al. (2006a) discussed the susceptibility of the
ischial shaft to post-depositional deformation.
Character 275: development of a caudal curvature of the distal margin of the iliac
peduncle of the ischium (Table I.14; Figs. I.40 and I.41)
(0): curvature absent, distal margin slightly rounded and, in some exemplars,
slightly curved cranially
705
(1): presence of a very short and slight curvature in the caudodorsal corner
(2): presence of a well developed curvature in the caudodorsal corner, so that the
peduncle appears “thumb-like” in lateral and medial profiles
Character 276: orientation of the distal articular surface of the iliac peduncle,
measured as the angle that it forms with the acentabular margin of the peduncle
(Table I.15; Fig. I.42)
(0): angle up to 115º (sample mean angle of 95º), angular craniodistal edge of the
iliac peduncle
(1): angle greater than 115º (sample mean angle of 125º), the distal surface of the
peduncle is further cranially oriented
Character 277: elongation of the iliac peduncle of the ishium (ratio between the
proximodistal length and the craniocaudal width of the distal margin) (Table I.16;
Fig. I.43)
(0): relatively short peduncle, ratio less than 1.5 (sample mean ratio of 1.27)
(1): ratio between 1.5 and 2 (sample mean ratio of 1.78)
(2): relatively long peduncle, ratio greater than 2 (sample mean ratio of 2.30)
Character 278: relative orientation of the acetabular and caudodorsal margins of
the iliac peduncle of the ischium (Fig. I.44)
(0): margins are either parallel or slightly convergent relative to each other
(correlated with a greater expansion of the craniodorsal corner of the peduncle)
(1): margins become slightly to greatly divergent near the proximal region of the
peduncle
Character 279: orientation of the craniocaudal axis of the pubic peduncle
(perpendicular to its articular margin) relative to the ischial shaft) (Table I.17; Fig.
I.45)
(0): ventrally inclined, angle up to 130º (sample mean angle of 118º)
(1): slightly inclined ventrally, angle greater than 130 and up to 170º (sample
mean angle of 157º)
(2): pubic peduncle parallel to the ischial shaft
Character 280: length/width proportions of the pubic peduncle of the ishium (Fig.
I.46)
706
(0): proximodistally longer than the distal articular surface is wide
(1): approximately as long proximodistally as the distal articular surface is
dorsoventrally wide
(2): proximodistally shorter than the dorsoventral width of the distal articular
surface
Character 281: relative position of the dorsal acetabular margin of the pubic
peduncle (Fig. I.47)
(0): ventral to or at the same level as the dorsal margin of the ischial shaft
(1): peduncular margin set dorsal to the dorsal margin of the ischial shaft
Character 282: dorsoventral thickness of the mid-shaft of the ischium (measured as
a ratio between this and the length of the entire shaft) (Table I.18; Fig. I.48;
(0): very thin shaft, up to 5% the length of the ischial shaft (sample mean of 4.6%)
(1): thickness ranging from more than 5% and up to 7.5% the length of the ischial
shaft (sample mean of 6.7%)
(2): very thick shaft, thickness greater than 7.5% the length of the ischial shaft
(sample mean of 8.4%)
Character 283: morphology of the distal region of the ischial shaft (Table I.19; Fig.
I.49 and I.50; modified from Godefroit et al., 2001, character 31)
(0): slightly expanded into a blunt end
(1): ventrally expanded forming a large “foot” or “boot-like” process
Character 284: degree of ventral projection of the distal expansion of the ischium
(expressed as the ratio between the length of the ischial shaft and the length of the
distal ventral expansion) (Table I.20; Fig. I.51; modified from Evans and Reisz,
2007, character 90)
(0): ratio less than 0.25 (sample mean ratio of 0.18)
(1): ratio of 0.25 or greater (sample mean ratio of 0.28)
Character 285: morphology of the cranial margin of the ventral expansion of the
distal ischial shaft (Fig. I.52)
(0): slightly concave and directed caudoventrally to meet the caudal margin to
nearly a point
707
(1): strongly concave, recurved cranial margin
Comments: the small juveniles of MOR 548 showed that condition (0) may be
present in the earlier stages of ontogeny in lambeosaurines.
Character 286: orientation of the long axis of the distal “foot” relative to the ischial
shaft (Fig. I.53)
(0): straight, ventrally directed
(1): cranioventrally directed, the inclination starting at the dorsal margin of the
“foot”
708
Table I.1. Character 243. Angle (A) of ventral deflection of the preacetabular process of
the ilium for a sample of iguanodontoidean ornithopods. Included is also the

Cluster (state)
AMNH 3971 cf. Corythosausus casuarius 143 1
TMP 92-53-21 cf. Parasaurolophus sp. 143 1
YPM 1190 Claosaurus agilis 163 0
AMNH 5240 Corythosausus casuarius 138 1
ROM 845 Corythosausus intermedius 140 1
BMNH 8821 Edmontosaurus sp. 142 1
USNM 22102 Edmontosaurus sp. 145 1
IRNSB 1551 Iguanodon atherfieldensis 159 0
BMNH unrecorded no. Iguanodon sp. 162 0
ANSP 16969 Lambeosaurinae 139 1
CMN 8705 Lambeosaurus magnicristatus 138 1
709
Table I.1—continued.

Cluster (state)
PIN 2232/19-1 Probactrosaurus gobiensis 164 0
PIN 12017 Probactrosaurus gobiensis 169 0
PMU R242 Tanius sinensis 148 1
710
170
−390
Deflection Angle of the Preacetabular Process
BIC
−410
160
−430
E V
2 4 6 8
Number of Clusters
150
140
130
Am Ba Br Ch Cl cfCo Co Ed Eda Edr Eo Gi Grl Grn Ig Iga Igb Lh Lml Lmm Ma Ol Ou Pac Paw Pb Pr Saa Sao Se Ta Ts Ve cfPa lc x
Taxon
Figure I.1. Character 243. Boxplot showing the distribution of the angle of ventral deflection of the preacetabular process of the ilium
of clusters in which the data can be classified. The photograph shows how this angle was measured, exemplified in the ilium of Barsboldia
sicinskii (ZPAL MgD-I 110).
711
Table I.2. Character 244. Ratio (R) between the craniocaudal length of the preacetabular
process and the craniocaudal length of the central plate of the lium for a sample of

cluster (state)
GMH HIj77 Charonosaurus jiayinensis 1.64 1
BMNH 8821 Hadrosauridae 1.60 1
MPCA unrecorded no. Salitral Moreno Hadrosauridae 1.65 1
712

cluster (state)
713
1.8
Ratio of the Lengtn of the Preacetabular Process
1.7
60
1.6
50
40
BIC
30
20
10
1.5
E V
0
2 4 6 8
Number of Clusters
Ba Br Ch Cos Eda Edr Gi Grl Grn Hy Iga Igb Lml Ma Ol Ou Pac Paw Pr Sao Se Sh Ta Ts cfPa sal x
Taxon
Figure I.2. Character 244. Boxplot showing the distribution of the ratio between the length of the preacetabular process and that of the
central plate of the ilium for a sample of iguanodontoidean ornithopods. The ilium of Ouranosaurus nigeriensis (GDF 300) is excluded from
the the cluster analysis because it is very autapomorphic. The inserted smaller graph shows the Bayesian information criterion (BIC) for various
numbers of clusters in which the data can be classified. The photograph shows how the two variable of this ratio were measured, exemplified
in an ilium of Brachylophosaurus canadensis (MOR 794).
714
Table I.3. List of specimens used to conduct a GDA analysis of the geometry of the
lateral profile of the preacetabular process of the ilium. The numbers correspond to the
specimens plotted in Fig. I.3.
Specimen Taxon Number on

Fig. I.3
YPM 1190 "Claosaurus agilis" 21
MOR 773 "Hypacrosaurus" stebingeri 13
AEHM 1/264 Amurosaurus riabinini 1
SBDE/95E Bactrosaurus johnsoni 5
MOR 794 Brachylophosaurus canadensis 2
MOR 1071-215 Brachylophosaurus canadensis 3
MOR 548 cf. "Hypacrosaurus" stebingeri 37
MOR 547 cf. Maiasaura peeblesorum 18
CUST JIII430 Charonosaurus jiayinensis 57
AMNH 5240 Corythosaurus casuarius 22
CMN 8399 Edmontosaurus annectens 25
DMNH 1943 Edmontosaurus annectens 26
ROM 801 Edmontosaurus regalis 27
BMNH 8821 Edmontosaurus sp. 9
USNM 3814 Edmontosaurus sp. 10
AMNH 6551 Gilmoreosaurus mongoliensis 32
ROM 764 Gryposaurus notabilis 35
LACM 53917 Hadrosauridae 8
UCMP 172239 Hadrosauridae 29
BMNH R11521 Iguanodon atherfieldensis 14
IRSNB 1551 Iguanodon atherfieldensis 39
IRSNB 1534 Iguanodon bernissartensis 15
BMNH R9290 Iguanodon sp. 16
715

Fig. I.3
YPM-PU 16970 Iguanodon sp. 17
YPM-PU 16970 Iguanodon sp. 17
ANSP 16969 Lambeosaurinae 7
FMNH UC1479 Lambeosaurinae 31
CMN 8705 Lambeosaurus magnicristatus 42
ROM 44770 Maiasaura peeblesorum 44
AEHM 2/845 Olorotitan ararhensis 45
GDF 300 Ouranosaurus nigeriensis 46
IVPP unrecorded no. Probactrosaurus sp. 54
ROM 787 Prosaurolophus maximus 51
AMNH 5220 Saurolophus osborni 52
FMNH P13423 Secernosaurus koerneri 53
PMU R242 Tanius sinensis 12
716
0.15
Gi(34)
x(17)
0.1
Ed+(29) Ch(57)
Am(1) Ed(26) Ts(56)
Hy+(38) Ba(6)
Ed(9) Cl(21) Pb(54)
Gr(35)
0.05 Pr(50) Ed(11)
Ed(10)
Ig(39)
Pa(48) Pr(51) Pb(55)
Ma(44) Lm+(41) Pr(49) Ba(5)
Ig(16)
Hy+(37) Ol(45) Ma+(20) Ed(25)
Co(22) Ma+(18) Ma+(19)
x(31) Co+(23)
0 x+(8)
Lm(42) Br(4)
Lm(43) Pa(47)
Br+(3)
Se+(53) Gi(33)
900
x(7) Sa(52) ●
Ed(24) Sh(58) Ou(46) ●
Ig(14) ●
●
Ed(28) Br(2) Ed(30) ●
800
●
Hy(36) ● ●
−0.05 ●
Gi(32) Ta(12)
700
Ig(15) Ed(27)
BIC
600
−0.1
500
Gr(40) EII VVI
VII EEE
400
● EEI EEV
x(13) VEI VEV
EVI VVV
300
−0.15
2 4 6 8
Number of Clusters
−0.15 −0.1 −0.05 0 0.05 0.1 0.15
Figure I.3. Bivariate plot between two axes of the NMMDS analysis of a GDA dissimilarity matrix of the geometry of the preacetabular
process of the ilium. Each dot is surrounded by the outline of the preacetabular process of the specimen that it represents. Colors indicate
the following groups: green = non-hadrosaurid Iguanodontoidea; blue = non-lambeosaurine Hadrosauridae; red = Lambeosaurinae;
yellow = Hadrosauridae indeterminate. The inserted smaller plot shows the Bayesian Information Criterion (BIC) for various numbers of
clusters in which the data may be classified. Cluster model identifiers in the BIC plot are as in Fig. F.4.
717
Table I.4. Character 245. Ratio (R) between the dorsoventral depth of the proximal region
of the preacetabular process and the dorsoventral depth of the proximal region central
plate of the ilum for a sample of iguanodontoidean ornithopods. Included is the
corresponding character state coding derived from K-Means cluster analysis.

Cluster (State)
ZPAL MgD-I 110 Barsboldia csinskii 0.64 2
TMP 92-53-21 cf. Parasaurolophus sp. 0.63 2
BMNH unrecorded no. Iguanodon sp. 0.41 0
718

Cluster (State)
FMNH UC1479 Lambeosaurinae 0.6 2
719
0.70
Depth Ratio of the Proximal Region of the Preacetabular Process
0.65
0.60
0.55
130
0.50
120
110
BIC
100
0.45
90
80
70
E V
0.40
2 4 6 8
Number of Clusters
Ba Bb Br Ch Cl Co Eda Eq Gi Grn Hy Ig Iga Igb Lml Ma Ol Ou Pac Paw Pb Pr Saa Sao Se Sh Ta cfPa sal x
Taxon
Figure I.4. Character 245. Boxplot showing the distribution of the a ratio between the depth of the proximal region of the preacetabular
process and the dorsoventral distance between the pubic peduncle and the dorsal margin of the ilium for a sample of iguanodontoidean ornithopods.
The photograph shows how the two variable of this ratio were measured, exemplified in an ilium of Prosaurolophus maximus (ROM 787).
720
Table I.5. Character 246. Ratio (R) between the dorsoventral depth and the craniocaudal
length of the central iliac plate for a sample of iguanodontoidean ornithopods. Included is
the character state coding derived from K-Means clustering analysis.

Cluster (State)
GMH Hlj77 Charonosaurus jiayinensis 0.83 0
BMNH 8821 Edmontosaurus sp. 0.69 1
PIN 12017 Probactrosaurus gobiensis 0.88 0
PIN unrecorded no. Saurolophus angustirostris 0.68 1
721

Cluster (State)
IVPP V725 Tsintaosaiurus spinorhinus 0.86 0
722
Ratio of the Depth of the Central Plate of the Ilium
0.9
60
50
40
BIC
30
20
10
0
0.8
2 4 6 8
Number of Clusters
0.7
0.6
Ba Br Ch Cl Coi Cos Ed Eda Edr Gi Grn Hy Hys Ig Iga Igb Lml Ol Ou Pac Paw Pb Pr Saa Sao Se Ta Ts
Taxon
Figure I.5. Character 246. Boxplot showing the distribution of the ratio between the depth of the central plate of the ilium and the distance
between the pubic peduncle and the caudodorsal prominence of the ischial peduncle, for a sample of iguanodontoidean ornithopods.
The photograph shows how the two variable of this ratio were measured, exemplified in the ilium of Secernosaurus koerneri (MACN-RN 2).
723
state 0
A
state 1
Figure I.6. Character 247. Position of the apex (X) of the supraacetabular process relative to that of
the caudal tuberosity (white solid circle) of the ischial peduncle. A, Gilmoreosaurus mongoliensis
(AMNH 6551), lateral view. B, Brachylophosaurus canadensis (MOR 794), lateral view.
724
state 0 state 1
A B
state 2 state 3
C D
Figure I.7. Character 248. Lateroventral expansion of the supraacetabular process of the ilium. A, Ouranosaurus nigeriensis
(GDF 300), lateral view. B, Bactrosaurus johnsoni (SBDE 95E), lateral view. C, Prosaurolophus maximus (ROM 787), lateral view.
D, Parasaurolophus cyrtocristatus (FMNH P27393), lateral view. Abbreviations: S = supraacetabular process.
725
Table I.6. Character 249. Ratio (R) between the craniocaudal width of the dorsal region
of the supraacetabular process and the craniocaudal length of the central iliac plate for a
sample of iguanodontoidean ornithopods. Included is the character state coding derived

Cluster (State)
ZPAL MgD-I 110 Barsboldia csinskii 0.47 3
TMP 92-53-21 cf. Parasaurolophus sp. 0.53 3
726

Cluster (State)
AMMH 5220 Saurolophus osborni 0.61 2
MCD 4791 Hadrosauridae 0.47 3
CM 1071 Hadrosauridae 0.52 3
BMNH 8821 Hadrosauridae 0.58 2
727
1.2
50
Ratio of the Breadth of the Supraacetabular Process
40
BIC
30
1.0
20
10
E V
2 4 6 8
0.8
Number of Clusters
0.6
0.4
Am Ba Bb Br Ch Cl Co Eda Edr Gi Grl Grn Hy Hys Ig Lml Ma Ol Ou Pac Paw Pr Saa Sao Se Sh Ta Ts Ve cfCo cfPa lc sal x
Taxon
Figure I.8. Character 249. Boxplot showing the distribution of the ratio between the breadth of the supraacetabular process along
its dorsal region and the craniocaudal length of the central iliac plate from the caudal ischial peduncle to the pubic peduncle, for a sample
of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers
of clusters in which the data can be classified. The photograph shows how the two variable of this ratio were measured, exemplified in
a right ilium of Prosaurolophus maximus (MOR 787).
728
state 0 state 0
A B
state 1 state 1
C D
Figure I.9. Character 250. Asymmetry and symmetry of the lateroventral margin of the supraacetabular process of the ilium.
A, Maiasaura peeblesorum (ROM 44770), lateral view. B, Tanius sinensis (PMU R242), lateral view. B, Gryposaurus latidens
(AMNH 5465), lateral view. C, Hypacrosaurus stebingeri (MOR 549), lateral view.
729
state 0 state 1
A B
state 2 state 3
C D
Figure I.10. Character 251. Morphology of the lateroventral margin of the supraacetabular process of the ilium. A, Iguanodon
bernissartensis (IRSNB 1551), lateral view. B, Bactrosaurus johnsoni (SBDE 95E), lateral view. C, Prosaurolophus maximus (TMP 84.1.1),
lateral view. D, Brachylophosaurus canadensis (MOR 794), lateral view; arrows point to the two poorly defined lobes of the process.
730
state 0 state 0
A B
state 0 state 0
C D
Figure I.11. Character 252. Presence of a ridge (arrows) uniting the caudodorsal margin of the supraacetabular process and the
proximodorsal margin of the postacetabular process of the ilium. A, Gilmoreosaurus mongoliensis (AMNH 6551), lateral view.
B, “Hypacrosaurus” stebingeri (MOR 549), lateral view. C, Secernosaurus koerneri (MACN-RN 2), lateral view. D, cf. Parasaurolophus sp.
731
state 1 state 1
A B
state 1 state 1
C D
Figure I.12. Character 252. Absence of demarcated margin between the caudal margin of the supraacetabular process and the proximodorsal
margin of the postacetabular process. A, Brachylophosaurus canadensis (MOR 794), lateral view. B, Gryposaurus notabilis (ROM 764), lateral view.
C, Saurolophus angustirostris (unrecorded number), lateral view. D, Hadrosauridae indeterminate (YPM-PU 16970; ?Gryposaurus), lateral view.
732
lateral profile of the pubic peduncle of the ilium.

Fig I.13
ANSP 10005 "Hadrosaurus foulkii" 1
AMNH 6579 Bactrosaurus johnsoni 2
51_BbZPAL110 Barsboldia sicinskii 51
AMNH 5469 cf. "Hypacrosaurus" stebingeri 9
UCMP 172239 Edmontosaurus sp. 34
USNM 214591 Hadrosauridae 11
PMU R216 Hadrosauridae 23
UTEP 377222 Hadrosauridae 24
YPM 16970 Hadrosauridae 25
CMN 8330 Hadrosauridae 31
BMNH R11521 Iguanodon atherfieldensis 17
733

Fig I.13
GFD 300 Ouranosaurus nigeriensis 44
734
735
0.3
Pa(48)
Ta(45)
0.2
Ig(20) x(12)
x(31)
Ba(2) x(1)
Ba(6)
Hy(37) Hy+(38)
0.1 Gi(35)
Se(46) Hy+(39)
Bb(51)
x+(9) Gi(36)
Ba(8) Ma+(27)
Lm+(42) x+(10)
Br(5) Br(3) x(24)
x(23) Pr(47) Ed+(34)
0 Ig(19) Ou(44) x(13) x+(32)
x(15) x+(11)
Ba(7) Lm(43)
Ig(16) Br+(4)
Ma+(28) Ed(33) x+(41)
Ig(18)
Ig(17)
Cl(26) Sh(50) Co+(30)
500
−0.1 ●
●
Hy(14) ● ● ●
●
Pb(49) x(25) ●
400
●
●
300
−0.2
BIC
Ig(21)
Ig(22)
200
−0.3
100
EII VVI
VII EEE
● EEI EEV
VEI VEV
0
EVI VVV
−0.4 Ig(40)
2 4 6 8
Number of Clusters
−0.6 −0.4 −0.2 0 0.2 0.4 0.6
Figure I.13. Bivariate plot of two axes of the NMMDS analysis of a GDA dissimilarity matrix of the geometry of the pubic peduncle of the ilium
of 52 iguanodontoidean specimens. Each dot is surrounded by the outline (rotated 90º) of the pubic peduncle of the specimen that it represents.
Colors indicate the following groups: green = non-hadrosaurid Iguanodontoidea; blue = non-lambeosaurine Hadrosauridae; red = Lambeosaurinae;
yellow = Hadrosauridae indeterminate. The inserted smaller plot shows the Bayesian Information Criterion (BIC) for various numbers of clusters
in which the data may be classified. Abbreviations in the BIC plot are as in Fig. F.4.
736
state 0
A
state 1
Figure I.14. Character 253. Morphology of the pubic peduncle of the ilium. A, Iguanodon sp.
(BMNH R9290), lateral view. B, Brachylophosaurus canadensis (MOR 794), lateral view.
737
lateral profile of the ischial peduncle of the ilium.
Specimen Taxon Number on Fig

I.15
ZPAL MgD-110 Barsboldia sicinskii 53
AMNH 5469 cf. "Hypacrosaurus" stebingeri 6
UCMP V81235 Hadrosauridae 10
ROM unrecorded no. Hadrosauridae 22
TMM 423094 Hadrosauridae 23
YPM-PU 16970 Hadrosauridae 27
BMNH 11521 Iguanodon atherfieldensis 14
BMNH 6462 Iguanodon sp. 13
738
Specimen Taxon Number on Fig.

I.15
BMNH unrecorded no. Iguanodon sp. 21
739
500
●
●
●
● ●
●
● ●
400
●
300
BIC
200
100
EII VVI
VII EEE
● EEI EEV
VEI VEV
x(28)
0
EVI VVV
2 4 6 8 Sh(52)
Co(32)
Number of Clusters
Ts(51)
Ma+(29)
x(24)
0.25 x(11)
Ba(1) Ba(5) Pb(50)
Br(4)
x+(9) x(27) Ed(35) Ou(42)
0.2
Br(2)
Cl(31) Ed(36) x+(6) x+(10) Ig(20)
x(34) Co+(33) Ig(17) Ig(21)
0.15 Gi(38)
Ta(48) Ig(14) Ig(13)
Hy(40)
0.1 Br+(3) x(26) Se(47) Pr(45)
Lm(41) x+(8) x(12)

0.05 Hy+(39)
x(23) Ig(18)
x+(7) Ig(15)
0 x+(25) Pa(43) −0.4
Ed(37)
Pr(44)
Gr(22) Ig(16)
Ma+(30) −0.3
−0.05
Sa(46)
−0.2
Hy+(49)
−0.1
Bb(53) Ig(19) −0.1
−0.15 0
0.1
−0.2
0.2
−0.25
0.3
−0.3 −0.2 −0.1 0 0.1 0.2 0.3 0.4
Figure I.15. 3D plot of three axes of the NMMDS analysis of a GDA dissimilarity matrix of the geometry of the ischial peduncle of the ilium in
53 iguanodontoidean specimens. Each dot is surrounded by the outline of the preacetabular process of the specimen that it represents. Colors
indicate the following groups: green = non-hadrosaurid Iguanodontoidea; blue = non-lambeosaurine Hadrosauridae; red = Lambeosaurinae;
740
A B C
D E F
Figure I.16. Character 254. Morphology of the ischial peduncle (arrows) of the ilium. A, Iguanodon sp. (BMNH R113), lateral view.
B, Iguanodon sp. (BMNH R114), lateral view. C, Tanius sinensis (PMU R242), lateral view. D, Bactrosaurus johnsoni (SBDE 95E/25), lateral view.
E, Gilmoreosaurus mongoliensis (AMNH 6551), lateral view. F, Lambeosaurinae indeterminate (USNM 11950; ?Corythosaurus sp.), lateral view.
741
Table I.9. Character 255. Ratio (R) between the craniocaudal length of the postacetabular
process and the craniocaudal length of the iliac central plate for a sample of
iguanodontoidean ornithopods. Included is the character state coding derived from K-

Cluster (State)
GMH Hlj77 Charonosaurus jiayinensis 0.70 0
BMNH 8821 Edmontosaurus sp. 0.97 1
Unrecorded no. Hadrosauridae 0.82 1
PIN 12017 Probactrosaurus gobiensis 0.62 0
742

Cluster (State)
743
20
Ratio of the Length of the Postacetabular Process
10
1.2
0
BIC
−40 −30 −20 −10
E V
1.0
2 4 6 8
Number of Clusters
0.8
0.6
Ba Br Ch Cl Co Ed Eda Edr Gi Grn Hy Hys Ig Iga Igb Lml Ma Ol Ou Pac Paw Pb Pr Saa Sao Se Ta Ts x
Taxon
Figure I.17. Character 255. Boxplot showing the distribution of the ratio between the length of the postacetabular process and that of the
central plate of the ilium, for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion
(BIC) for various numbers of clusters in which the data can be classified. The photograph shows how the two variable of this ratio were
measured, exemplified in an ilium of Brachylophosaurus canadensis (MOR 794).
744
state 0 state 0
A B
state 1 state 1
C D
Figure I.18. Character 256. Presence and absence of brevis shelf (arrow and rectangle) at the base of the iliac postacetabular process.
A, Bactrosaurus johnsoni (SBDE 95E), medial view. B, ventral view of SBDE 95E. C, Brachylophosaurus canadensis (MOR 1071-8-2-98-469),
medial view. D, ventral view of MOR 1071-8-2-98-469.
745
state 0 state 1
A C
state 2 state 3
D F
E G
Figure I.19. Character 257. Medial ridge (arrows) of the postacetabular process and orientation of the
brevis shelf (when present). The rectangles highlight the caudal region of the postacetabular process in
dorsal view. A, Hypacrosaurus stebingeri (MOR 549), medial view. B and C, Gilmoreosaurus mongoliensis
(AMNH 6551), medial and dorsal views, respectively. D and E, Hypacrosaurus stebingeri (AMNH 5204),
medial and dorsal views, respectively. F and G, Secernosaurus koerneri (MACN-RN 2), medial and dorsal
views, respectively.
746
state 0
A
state 1
Figure I.20. Character 258. Absence and presence of a median ridge on the laterodorsal
surface of the postacetabular process of the ilium. A, Hypacrosaurus altispinus (AMNH 5204),
caudal view. B, Secernosaurus koerneri (MACN-RN 2), caudal view.
747
lateral profile of the postacetabular process of the ilium.
Specimen Taxon Number on Plot

form Fig. I.21
AMNH 3971 ?Corythosaurus sp. 8
CUST JIII430 Charonosaurus jiayinensis 69
CMN 44845 Corythosaurus sp. 36
LACM 23504 Edmontosaurus sp. 15
PMU R216 Hadrosauridae 27
YPM 16970 Hadrosauridae 29
748
Specimen Taxon Number on Plot

from Fig. I. 21
CMN 41000 Hadrosauridae 31
FMNH PR1922 Hadrosauridae 42
BMNH 11521 Iguanodon atherfieldensis 21
749
0.6
Hy(49)
Pa(56)
0.4
Hy+(50) Pa(57)
Co+(37) x(12)
Ig(26)
Gi(45)
0.2 x+(10)
Ig(20)
Ed+(38) Gi(43)x(14) Hy(47)
Hy+(48) x(63)
Ba(6)
x(8) Ed(40) x(19)
Ch(69)
x(16) Gr(51)
Ed(39) Br(1) Sh(70)
Ig(23) x+(7)
x(18) Br(2)
x(17) x+(28)
Lm(53)
0 Ig(25) Ig(24)
x(64)x+(11)
x(42) Ma+(32) Lm+(52) Ma(54)
Ig(21) x+(65) x(31)
Co(36) Ed+(15)
Gi(44) Ed(41)
Ma+(33)
x(30) Co(35)x(29)
Ou(55) Br+(3) Sa(61)
x(13)
x(27) Pr(59)
Ba(4)
x(9)
Ig(22) Ta(62) Pr(60)
−0.2 ● ●
●
900
●
Pr(58)
Gr(46) ●
●
●
Pb(66)
●
Cl(34)
800
●
Ba(5)
Ts(68)
700
BIC
−0.4
600
500
EII VVI
Pb(67) VII EEE
400
● EEI EEV
VEI VEV
−0.6 EVI VVV
2 4 6 8
Number of Clusters
−0.8 −0.6 −0.4 −0.2 0 0.2 0.4
Figure I.21. Character 259. Bivariate plot between two axes of the NMMDS analysis of a GDA dissimilarity matrix of the geometry of the
postacetabular process of the ilium. Each dot is surrounded by the outline of the preacetabular process of the specimen that it represents.
Colors indicate the following groups: green = non-hadrosaurid Iguanodontoidea; blue = non-lambeosaurine Hadrosauridae; red = Lambeosaurinae;
yellow = Hadrosauridae indeterminate. The inserted smaller plot shows the Bayesian Information Criterion (BIC) for various numbers of
clusters in which the data may be classified. Abbreviations in the BIC plot are as in Fig. F.4.
750
state 0
A
state 1
Figure I.22. Character 259. Lateral profile of the postacetabular process of the ilium.
A, Iguanodon atherfieldensis (IRSNB 1551), lateral view. B, Lambeosaurus lambei (ROM 1218).
751
state 0
A
state 1
Figure I.23. Character 260. Relative orientation of the dorsal and ventral margins of the lateral
or laterodorsal surface of the postacetabular process of the ilium. A, Hadrosauridae indeterminate
(UTEP P37.7.222), lateral view. B, Tsintaosaurus spinorhinus (IVPP V725).
752
state 0
A
state 1
Figure I.24. Character 261. Position of the medial sacral ridge (arrows) within the medial
surface of the central iliac plate. A, Iguanodon sp. (BMNH R113), medial view. B, Bactrosaurus
johnsoni (cast of SBDE 95E), medial view.
753
state 0 state 0
A B
state 1 state 1
C D
Figure I.25. Character 262. Orientation of the medial sacral ridge (arrows) of the ilium. A, Ouranosaurus nigeriensis (GDF 300), medial view.
B, Tanius sinensis (PMU R242), medial view. C, Barsboldia sicinskii (ZPAL MgD-I 110), medial view. D, Brachylophosaurus canadensis
(MOR 1071-8-2-98-469), medial view.
754
state 0
A
state 1
Figure I.26. Character 263. Lateral profile of the dorsal margin of the ilium above the supraacetabular
process. The arrow indicates the presence of a concave dorsal margin. A, Iguanodon bernissartensis
(IRSNB 1534), lateral view. B, Charonosaurus jiayinensis (CUST JIII430).
755
state 0
A
state 1
Figure I.27. Character 264. Orientation of the dorsal margin of the prepubic process of the
pubis. A, Ouranosaurus nigeriensis (GDF 300), lateral view. B, Prosaurolophus maximus
(MOR 454).
756
lateral profile of the prepubic process of the pubis.
Specimen Taxon Number form

Fig. I.28
MOR 549 "Hypacrosaurus stebingeri" 21
LACM 25874 "Lambeosaurus" laticaudus 29
AMNH 3971 cf. Corythosaurus sp. 14
TMP 92.53.21 cf. Parasaurolophus sp. 45
USNM 4737 Edmontosaurus annectens 46
BMNH R5915 Edmontosaurus annectens 47
CM 38333 Edmontosaurus sp. 44
TMP 80.22.1 Gryposaurus notabilis 49
BMNH 5915 Hadrosauridae 20
YPM-PU 16970 Hadrosauridae 51
BENC 18-1-0013 Hadrosauridae 52
PASAC-1 Hadrosauridae 53
BMNH 1534 Iguanodon bernissartensis 36
2MOR 454 Prosaurolophus maximus 27
MACN-RN 2 Secernosaurus koerneri 37
757
●
500
●
● ●
●
●
● ●
●
0.2
400
BIC
300
0.15
Ig(35)
200
EII VVI
VII EEE Ig−R(36)
0.1 ● EEI EEV
VEI VEV Ig(5)
100
EVI VVV
Pa(11)
2 4 6 8
0.05 Number of Clusters Ig(22) Pa(6)
x(14)
Co+R(23)
Ou(13)
Ma+R(26) Pr−R(12) Pr(27)
0 Co(4)
Hy−R(21)
Br(28) x+(19) Br−R(24) Gr(10)
x(30)
x(17) Br(16)
−0.05 Hy(32) Ed(8) Sh(7)

x(1) Hy++(34) Se(29) Ed+(18)
state 1 x(31) Ed−R(3) x(20)
Hy++(33)
−0.1 Ma(25)
Ed(2)
Ed−R(9)
−0.15 Ed(15)
0.4
0.2
−0.2 0.1 0.15 0.2

−0.05 0 0.05
−0.2 −0.15 −0.1
−0.3
−0.4 −0.25
Figure I.28. 3D plot of three axes of the NMMDS analysis of a GDA dissimilarity matrix of the geometry of the prepubic process of the pubis
for 36 iguanodontoidean specimens. Each dot is surrounded by the outline of the preacetabular process of the specimen that it represents. Colors
yellow = Hadrosauridae indeterminate. Pubic profiles for each point are facing rostrolaterally towards the reader. The inserted smaller plot shows
the Bayesian Information Criterion (BIC) for various numbers of clusters in which the data may be classified. Abbreviations in the BIC plot are
as in Fig. F.4.
758
state 0
A
state 1
state 1
Figure I.29. Character 265. Lateral profile of the prepubic process of the pubis. A, Iguanodon sp.
(BMNH R2194), lateral view. B, “Lambeosaurus” laticaudus (LACM 26874).
759
state 2
A
state 3
B
state 4
state 1
Figure I.30. Character 265. Geometry of the distal blade of the prepubic process of the pubis.
A, Brachylophosaurus canadensis (MOR 1071- 8-20-98-587), lateral view. B, Edmontosaurus
regalis (CMN 2289), lateral view. B, Gryposaurus notabilis (ROM 764), lateral view.
760
state 0
A
state 1
state 1
Figure I.31. Character 266. Depth of the dorsoventral expansion of the distal region of the prepubic
process relative to the width of the acetabular margin of the pubis. A, Secernosaurus koerneri
(MACN-RN 2), right medial view. B, cf. Corythosaurus sp. (AMNH 3971).
761
state 0
A
state 1
B
state 2
Figure I.32. Character 267. Length of the distal blade (d) of the prepubic process relatve to that
of the prepubic constriction (n). A, Lambeosaurinae (?Corythosaurus sp.; GDF 300), lateral view.
B, Secernosaurus koerneri (MACN-RN 2), lateral view. C, Bactrosaurus johnsoni (SBDE 95E5/43),
lateral view.
762
state 0
state 1
state 1
Figure I.33. Character 268. relative position of maximum concavity of the dorsal (asterisk) and ventral
(circle) margins of the prepubic process of the pubis. A, Ouranosaurus nigeriensis (GDF 300), lateral
view. B, Gryposaurus latidens (AMNH 5465).
763
state 0
A
state 1
state 1
Figure I.34. Character 269. Morphology of the acetabular margin of the pubis. A, Gilmoreosaurus
mongoliensis (AMNH 6551), lateral view. The rectangle surrounds a rudge that connects the iliac peduncle with the
proximal region of the ischial peduncle. B, cf. Tsintaosaurus spinorhinus (IVPP K149).
764
state 0
A
state 1
state 1
Figure I.35. Character 270. Presence and absence of an obturator gutter or foramen ventral to the
ischial peduncle of the pubis. A, Iguanodon sp. (BMNH R28504), lateral view. B, Edmontosaurus sp.
(LACM 7241). Abbreviations: ip = ischial peduncle; og = obturator gutter.
765
Table I.12. Character 271. Length/width ratio (R) of the ischial peduncle of the pubis for
a sample of iguanodontoidean ornithopods. Included is the character state coding derived

Cluster (State)
MOR 1071- 8-20-98-587 Brachylophosaurus canadensis 2.36 1
TMM 42309 cf. Gryposaurus sp. 3.70 2
AMNH 5202 cf. Hypacrosaurus altispinus 2.32 1
TMP 88-151-26 Hypacrosaurus stebingeri 2.43 1
LACM 7241 Edmontosaurus annectens 2.30 1
MOR 547 cf. Maiasaura peeblesorum 2.41 1
MOR 547 cf. Maiasaura peeblesorum 2.50 1
TMP 1998-93-181 Hadrosauridae 2.11 1
CM 1071 Hadrosauridae 2.54 1
LACM 26874 Hadrosauridae 2.68 1
766
4.5
−60
Lenggth/Width Ratio of the Ischial Peduncle
−70
4.0
BIC
−80
−90
−110 −100
3.5
E V
2 4 6 8
Number of Clusters
3.0
2.5
2.0
1.5
Br Co Cos Eda Edr Ig Ma Pac Pr Se cfGr cfHy cfVe x

Taxon
Figure I.36. Character 271. Boxplot showing the distribution of the length/width ratio of the ischial peduncle of the pubis for a sample
of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in
which the data can be classified. The photograph shows how the two variable of this ratio were measured, exemplified in a pubis of an
indeterminate hadrosaurid (TMP 1966.27.24).
767
Table I.13. Character 274. Ratio (R) between the craniocaudal length of the pubis
(distance from the acetabular margin to the distal margin of the prepubic process) and the
distance from the dorsal margin of the iliac peduncle to the ventral margin of the
proximal postpubic shaft, for a sample of iguanodontoidean ornithopods. Included is the

Cluster (State)
MOR 1071- 8-20-98-587 Brachylophosaurus canadensis 2.87 1
CM 38333 Edmontosaurus sp. 3.61 2
TMP 66.04.01 Lambeosaurus magicristatus 2.86 1

Cluster (State)
UALVP unrecorded no. Lambeosaurinae 2.23 0
TMP 1997.12.147 Lambeosaurinae 2.27 0
−60
−70
−80
BIC
3.5
−90
−110 −100
Pubis Total Length Ratio
E V
2 4 6 8
Number of Clusters
3.0
2.5
2.0
Ba Br Co Ed Eda Edr Grn Hys Ig Iga Igb Lml Lmm Ma Ou Pac Paw Pr Se Sh Ts cfCo cfPa lc sal x
Taxon
Figure I.39. Character 274. Boxplot showing the distribution of the ratio between the distance from the acetabular margin to the distal
border of the prepubic process and the distance from the dorsal margin of the iliac peduncle to the ventral border of the proximal postpubic shaft,
of clusters in which the data can be classified. The photograph shows how the two variable of this ratio were measured, exemplified in a pubis of
Bactrosaurus johnsoni (SBDE 95E5/43).
state 0
A
state 1
state 1
Figure I.37. Character 272. Lateroventral protuberance (arrow) on the proximal region of the ischial
peduncle of the pubis A, Prosaurolophus maximus (MOR 454), lateral view. B, Hypacrosaurus
stebingeri (MOR 549).
768
state 0
A
state 1
state 1
Figure I.38. Character 273. Depth/width proportions of the iliac peduncle (rectangles) of the
pubis. A, Iguanodon sp. (BMNH R2194), lateral view. B, Corythosaurus intermedius (ROM 845).
769
Table 14. List of specimens used to conduct a GDA analysis of the geometry of the iliac
peduncle of the ischium.
Specimen Taxon
ANSP 10005 Hadrosaurus foulkii
MOR 549 Hypacrosaurus stebingeri
LACM 17715 Lambeosaurus laticaudus
SBDE95/E26 Bactrosaurus johnsoni
AMNH 6553 Bactrosaurus johnsoni
MOR 794 Brachylophosaurus canadensis
TMP 9253-21 cf. Parasaurolophus sp.
CUST 679 Charonosaurus jiayinensis
ROM 845 Corythosaurus intermedius
CMN 8399 Edmontosaurus annectens
CMN 8509 Edmontosaurus annectens
SM 4036 Edmontosaurus annectens
CMN 2289 Edmontosaurus regalis
AMN 17 Edmontosaurus sp.
ROM 867 Edmontosaurus sp.
AMNH 6551 Gilmoreosaurus mongoliensis
MSNM V345 Gryposaurus notabilis
PMU R95 Hadrosauridae
USNM 15582 Hadrosauridae
LACM 20875 Hadrosauridae
UCMP V86026 Hadrosauridae
UCMP 137273 Hadrosauridae
TMP 1981.14.47 Hadrosauridae
PASAC-1 Hadrosauridae
CMN 8501 Hypacrosaurus altispinus
IRSNB 1534 Iguanodon bernissartensis
IRSNB 1535 Iguanodon bernissartensis
BMNH R115 Iguanodon sp.
BMNH 2158 Iguanodon sp.
IRSNB 3708 Iguanodon sp.
FMNH UC1479 Lambeosaurinae
ROM 1218 Lambeosaurus lambei
CMN 8703 Lambeosaurus lambei
ROM 44770 Maiasaura peeblesorum
IRSNB 2/845 Olorotitan ararhensis
FMNH P27393 Parasaurolophus cyrtocristatus
773
Specimen Taxon
IVPP unrecorded no. Probactrosaurus sp.
IVPP unrecorded no. Probactrosaurus sp.
MOR 454 Prosaurolophus maximus
MACN-RN 2 Secernosaurus koerneri
IVPP V725 Tsintaosaurus spinorhinus
774
●
300
● ●
● ●
●
●
●
200
BIC
100
0
EII VVI
VII EEE
−100
● EEI EEV
VEI VEV
EVI VVV
2 4
Number of Clusters
6 8
A
0.3
state 1
0.2
0.1
Axis2
0.0
−0.1
−0.2
−0.4 −0.2 0.0 0.2 0.4 0.6

Axis1 B
Figure I.40. Character 275. Morphology of the distal region of the iliac peduncle of the ischium. A, BIC
plot for various clustering models of the NMMDS derived from the GDA dissimilariy matrix. B, coordinate
projection of the NMMDS data showing chosen clustering into three states. Abbreviations of clustering models.
Abbreviations for the BIC plot are as in Fig. F.4.
775
state 0
A
state 1
state 1
B
state 2
Figure I.41. Character 275. Degree of development of a curved distal margin (rectangles) of the iliac
peduncle of the ischium. A, Brachylophosaurus canadensis (MOR 794), lateral view. B, Iguanodon
atherfieldensis (BMNH 11521), lateral view. C, B, Tsintaosaurus spinorhinus (IVPP V725), lateral view
776
Table I.15. Character 276. Orientation (angle A) of the distal articular surface of the iliac
peduncle of the ischium for a sample of iguanodontoidean ornithopods. Included is the

Cluster (State)
SBDE 95E5/26 Bactrosaurus johnsoni 121 1
TMP 92.53.21 cf. Parasaurolophus sp. 130 1
CUST JV679 Charonosaurus jiayinensis 119 1
SM 4036 Edmontosaurus annectens 120 1
MSNM 345 Gryposaurus notabilis 121 1
BMMNH R110 Iguanodon sp. 103 0
IRSNB 1535 iguanodon bernissartensis 80 0
ROM 44770 Maiasaura peeblessorum 118 1
IRSNB Olorotitan ararhensis 118 1
IVPP unrecorded no. cf. Probactrosaurus gobiensis 102 0
IVPP unrecorded no. cf. Probactrosaurus gobiensis 99 0
PMU R95 Hadrosauridae 130 1
UCMP 137273 Hadrosauridae 128 1
FMNH UC1479 Hadrosauridae 128 1
UCMP V86026 Hadrosauridae 106 0
777

Cluster (State)
ANSP 10005 “Hadrosaurus foulkii” 124 1
778
130
Orientation of the Articular Surface of the Iliac Peduncle
120
110
−310
−320
100
BIC
−330
−340
90
−350
E V
2 4 6 8
Number of Clusters
80
Ba Br Ch Coi Eda Edr Gi Grn Hd Hy Hys Ig Igb Lml Ma Ol Pac Pr Se Ts cfPa cfPb lc x
Taxon
Figure I.35. Character 276. Boxplot showing the distribution of the angle between the distal articular surface of the iliac peduncle of the
ischium and the acentabular margin of the peduncle, for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the
Bayesian information criterion (BIC) for various numbers of clusters in which the data can be classified. The photograph shows how this angle
was measured, exemplified in an ischium of Lambeosaurus lambei (ROM 1218).
779
Table I.16. Character 277. Ratio (R) between the proximodistal length and the
craniocaudal width of the distal margin of the iliac peduncle of the ischium for a sample
of iguanodontoidean taxa. Included is the corresponding character state coding derived
from K-means clustering analysis.

Cluster (State)
CUST JV679 Charonosaurus jiayinensis 1.85 1
SM 4036 Edmontosaurus annectens 2.38 2
MSNM 345 Gryposaurus notabilis 1.55 1
BMMNH R110 Iguanodon sp. 1.65 1
IRSNB 1535 iguanodon bernissartensis 1.21 0
ROM 44770 Maiasaura peeblessorum 1.82 1
IVPP unrecorded no. cf. Probactrosaurus gobiensis 1.72 1
IVPP unrecorded no. cf. Probactrosaurus gobiensis 1.72 1
FMNH UC1479 Hadrosauridae 1.72 1
780

Cluster (State)
UCMP 137273 Hadrosauridae 2.14 2
UCMP V86026 Hadrosauridae 2.14 2
781
2.6
−40
Length/Width Ratio of the Iliac Peduncle of the Ischium
2.4
−60
BIC
−80
2.2
−100
2 4 6 8
2.0
Number of Clusters
1.8
1.6
1.4
1.2
Ba Br Ch Co Cos Eda Edr Gi Grn Hy Ig Igb Lml Ma Ol Pa Pac Pb Pr Saa Se Ts Ve lc x
Taxon
Figure I.43. Character 277. Boxplot showing the distribution of the ratio between the length and the width of the iliac peduncle of the
ischium for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for various
numbers of clusters in which the data can be classified. The photograph shows how the two variable of this ratio were measured, exemplified
in an inchium of Charonosaurus jiayinensis (CUST JV679).
782
state 0
A
state 1
state 1
Figure I.44. Character 278. Relative orientation of the acetabular and caudodorsal margins of the
iliac peduncle of the ischium A, Iguanodon sp. (BMNH R51623), lateral view. B, Brachylophosaurus
canadensis ( MOR 454), lateral view
783
Table I.17. Character 279. Orientation of the pubic peduncle of the ischium, measured as
the angle (A) between the ischial shaft and the axis perpendicular to the articular margin
of the pubic peduncle, for a sample of iguanodontoidean taxa. Included is also the

Cluster (State)
SBDE 95E5/26 Bactrosaurus johnsoni 165 1
ROM 867 Edmontosaurus annectens 178 2
PIN 2232/29-2 Probactrosaurus gobiensis 147 1
GMV 1780 Shantungosaurus giganteus 180 2
CPC-59 Velafrons cahuilensis 162 1
ANSP 16969 Hadrosauridae 163 1
784
200
−260
−270
Orientation of the Pubic Peduncle of the Ischium
BIC
180
−280
−290
2 4 6 8
Number of Clusters
160
140
120
Ba Br Coi Eda Edr Gi Grn Hy Ig Igb Lml Ou Pb Saa Se Sh Ts Ve x
Taxon
Figure I.45. Character 279. Orientation of the craniocaudal axis of the pubic peduncle (perpendicular to its articular margin) relative to
the ischial shaft for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for
various numbers of clusters in which the data can be classified. The photograph shows how the two variable of this ratio were measured, exemplified
in an ischium of Corythosaurus intermedius (ROM 845).
785
state 0
A
state 1
B
state 2
Figure I.46. Character 280. Length/width proportions of the pubic peduncle of the ishium.
A, Iguanodon atherfieldensis (BMNH R11521), lateral view. B, Tsintaosaurus spinorhinus
(IVPP V725). C, Secernosaurus koerneri (MACN-RN 2), lateral view.
786
state 0
A
state 1
Figure I.47. Character 281. Position of the dorsal acetabular margin of the pubic peduncle
relative to the dorsal margin of the ischial shaft. A, Velafrons coahuilensis (CPC-59), lateral view.
B, Brachylophosaurus canadensis (MOR 794), lateral view.
787
Table I.18. Character 282. Ratio (R) between the dorsoventeal thickness of the ischial
shaft and its length for a sample of iguanodontoidean taxa. Included is also the
corresponding character state coding derived from K-means clustering analysis,

Cluster (State)
OMNH 24389 Eolambia caroljonesa 6.94 1
PIN 2232-29-2 Probactrosaurus gobiensis 7.04 1
USNM 15581 Lambeosaurinae 7.46 1
788
9
Ratio of the Thickness of the Ischial Shaft
8
7
−140
6
−150
BIC
−160
−170
5
−180
2 4 6 8
Number of Clusters
4
Ba Br Co Hys Eda Edr Eo Gi Grn Hy Ig Igb Lml Lmm Ol Pac Pb Pr Saa Sao Sh Ts x
Taxon
Figure I.48. Character 282. Boxplot showing the distribution of the ratio between the dorsoventral width of the midshft of the ischium and the
length of the ischial shaft for a sample of iguanodontoidean ornithopods. The proximal end of the shaft is considered to be located where the dorsal
profile of the ischium begins to rise to form the base of the iliac peduncle.The inserted smaller graph shows the Bayesian information criterion (BIC)
for various numbers of clusters in which the data can be classified. The photograph shows how the two variables of this ratio were measured,
exemplified in an ishium of Parasaurolophus cyrtocristatus (FMNH P27393).
789
lateral profile of the distal expansion the ischium.
Specimen Taxon Number of Fig.

I.49
LACM 17715 "Lambeosaurus" laticaudus 16
SBDE 95 Bactrosaurus johnsoni 2
AMNH 5225 cf. Hypacrosaurus altispinus 20
CUST K2 cf. Tsintaosaurus spinorhinus 8
CUST 1225 Charonosaurus jiayinensis 6
CMN8673 Corythosaurus sp. 9
BMNH 2158 Hadrosauridae 13
BMNH R811 Hadrosauridae 14
SC (Serrat del Corb) Hadrosauridae 22
TMM 42309 Hadrosauridae 29
IRSMB 1534 Iguanodon bernissartensis 28
CMN 351 Lambeosaurus lambei 17
CMN 8703 Lambeosaurus lambei 23
ZPAL MgD-I 159 Saurolophus angustirostris 21
790
0.1
0.05
Ba(3)
Lm(19)
Br(4)
x(22)
Hy(25)
0
x(5) Gr(35)
Sa+(21)
Lm(17) x(29)
Pr(36) Ed(32)
Hy(12) Co(18) Sh(31)
Ba(2)
−0.05 Co(9) Ig+(13)x(23) Co(27)
Ig(28) Ed(34) x(15)
x(11) Ts(8) Ed(1)
Ig(14)
Lm(16) Ed(33)
−0.1 Pb(26)
Ig(10) x(7) Pa(30)
x(24) x(20)
−0.15
500
0.15 ●
● ● ● ● ●
●
●
●
400
0.1
BIC
300
0.05 Ch(6)
200
0
0.4
EII VVI 0.3
VII 0.2
EEE
100
−0.05 0.1 ● EEI EEV
0 VEI VEV
−0.1 EVI VVV
−0.1 −0.2
−0.3 2 4 6 8
Number of Clusters
Figure I.49. 3D plot of three axes of the NMMDS analysis of a GDA dissimilarity matrix of the geometry of the prepubic process of the pubis
for 36 iguanodontoidean specimens. Each dot is surrounded by the outline of the preacetabular process of the specimen that it represents. Colors
791
state 0
A
state 0
state 1
Figure I.50. Character 283. Morphology of the distal end of the ischium. A, Shantungosaurus
giganteus (GMV 1780), lateral view. B, Hypacrosaurus altispinus (CMN 8501), lateral view.
792
Table I.20. Ratio (R) between the length of the distal “foot” and the craniocaudal length
of the shaft of the ischium for a sample of iguanodontoidean ornithopods. Included is also
the corresponding character state coding derived from K-means clustering analysis.

Cluster (State)
UHR 6590 Nipponosaurus sachalinensis 0.20 0
USNM 15581 Lambeosaurinae 0.18 0
FMNH UC1479 Lambeosaurinae 0.19 0
LACM 17715 “Lambeosaurus” laticaudus 0.18 0
793
0.30
90
80
70
0.28
Ratio of the Length of the Ischial “Foot”
BIC
60
50
0.26
40
30
E V
0.24
2 4 6 8
Number of Clusters
0.22
0.20
0.18
0.16
Ba Co Hy Hys Ig Lml Lmm Np Pac Ts Ve lc x
Taxon
Figure I.51. Character 284. Boxplot showing the distribution of the ratio between the length of the distal expansion of the ischium and that
of the ischial shaft for a sample of iguanodontoidean ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC)
for various numbers of clusters in which the data can be classified. The photograph shows how the two variables of this ratio were measured,
exemplified in a ischium of “Lambeosaurus” laticaudus (LACM 17715).
794
state 0
A
state state
1 1
Figure I.52. Character 285. Morphology of the cranial margin of the ventral expansion of the distal
ischial shaft. A, Iguanodon sp. (BMNH R9290), lateral view. B, Parasaurolophus cyrtocristatus
(FMNH P27393), lareral view.
795
state 0
A
state 1
Figure I.53. Character 286. Orientation of the long axis of the distal expansion of ths ischium
relative to the ischial shaft. A, Lambeosaurus lambei (ROM 1218), lateral view. B, Lambeosurinae
(?Corythosaurus sp.; USNM 15581), lateral view
796
APPENDIX J
HINDLIMB CHARACTERS
Femur
The development of an intercondylar extensor channel due to rostromedial fusion

of the distal condyles has been regarded by several authors (Weishsmapel and Horner,
1990; Head, 2001; Norman, 2002; Hai-lu You et al., 2003; Horner et al., 2004) as
synapomorphic for hadrosaurids and the most closely-related outgroup taxa such as
Bactrosaurus and Telmatosaurus. However, I have observed this condition in large
specimens of Iguanodon bernissartensis (IRSNB 1534) and smaller ones from I.
atherfieldensis (BMNH 11521). All other outgroup taxa sampled for this studyed a closed
or nearly closed intercondylar groove as in hadrosaurids. Thus, this character was
excluded from the present analysis.
Similarly, no substantial difference in the craniocaudal expansion of the distal
condyles (Norman, 2002, character 65) was found among the outgroup and ingroup taxa
of this study. For example, the condyles in Iguanodon bernissartensis (IRSNB 1731 and
IRSNB 1534) were not less expanded than in Bactrosaurus johnsoni (AMNH 6553) or a
typical hadrosaurid such as Brachylophosaurus canadensis (MOR 794) and
Lambeosaurus lambei (ROM 1218). Thus, this character was also excluded for being
uninformative.
Character 287: degree of curvature of the distal half of the femoral shaft (Fig. J.1;
modified from Norman, 2002, character 62)
(0): slightly curved caudomedially
(1): absence of curvature, straight distal shaft
Character 288: lateral profile of the caudoventral margin of the fourth trochanter of
the femur (Fig. J.2; modified from Wagner, 2001)
(0): triangular and ending in a caudally, and slightly ventrally, directed point
(1): smooth and arcuate
797
Tibia
Character 289: extension of the cnemial crest of the tibia (Fig. J.3; Godefroit et al.,
2000, character 31)
(0): the cnemial crest is restricted to the proximal end of the tibia
(1): cnemial crest further extended along the cranial surface of the proximal half
of the diaphysis
Fibula
Character 290: shape of the cranially expanded distal end of the fibula (Fig. J.4;
modified from Godefroit et al., 2000, character 32)
(0): subtriangular, with a straight or slightly concave cranial margin;
(1): club-shaped, with a concave cranial margin and prominent cranial expansion.
Comments: Godefroit et al. (2000) regarded state (1) of this character as
synapomorphic for Parasaurolophus cyrtocristatus and Charonosaurus jiayinensis.
However, a club-like distal end of the fibula is also present in other taxa, such as
Corythosaurus intermedius (e.g., ROM 845) and even in a juvenile cf. Hypacrosaurus
stebingeri (MOR 548).
Pes
Character 291: development of the medial platform of the astragalus (Fig. J.5)
(0): it extends medially to completely underlie the medial malleolus of the tibia
(1): short, wedges laterally underlying only part of the medial malleolus of the
tibia
Character 292: distal tarsals II and III (Horner et al., 2004, character 102)
(0): present
(1): absent
Character 293: metatarsal I (modified from Norman, 2002, character 66)
(0): absent
798
(1): slender and splint-like element
Character 294: length/width proportions of metatarsal III (Table J.1; Fig. J.6;
measured as the ratio between its proximodistal length and its mediolateral breadth
at mid-shaft)
(0): ratio less than 4.50 (mean ratio of 4.02)
(1): elongated, ratio of 4.50 or greater (mean ratio of 5.12)
Character 295: length/width proportions of pedal phalanx II2 (Fig. J.7)
(0): proximodistally shortened, being twice as wide mediolaterally as it is
proximodistally long
(1): subsquared, only slightly shorter proximodistally than it is wide
mediolaterally
Character 296: length/width proportions of the disc-shaped pedal phalanx III2-III3
(Fig. J.8; modified from Horner et al., 2004, character 104)
(0): up to three times (or less) wider than they are proximodistally long
(1): more than three times wider than they are proximodistally long
Character 297: length/width proportions of the disc-shaped pedal phalanges IV2-
IV4 (Fig. J.9; modified from Horner et al., 2004, character 104)
(0): up to three times (or less) wider than they are proximodistally long
(1): more than three times wider than they are proximodistally long
Character 298: morphology of the pedal unguals (Fig. J.10; Norman, 2002,
character 67)
(0): proximodistally elongated and arrow-shaped, with a bluntly truncated tip and
prominent claw grooves
(1): mediolaterally broad and proximodistally shortened, rounded shield or hoof-
like shaped, with reduced or absent claw grooves
Character 299: ridge on the plantar surface of pedal unguals (Fig. J.11; Prieto-
Marquez, 2005, character 47)
(0): absent
(1): present
799
state 0 state 1
A B
Figure J.1. Character 287. Degree of curvature of the distal half of the femoral shaft.
A, Iguanodon sp. (BMNH R2734), laterocaudal view. B, “Hadrosaurus foulkii” (ANSP 10005),
lateral view.
800
state 0
A
state 1
Figure J.2. Character 288. Lateral profile of the caudoventral margin (rectangles) of the fourth
trochanter of the femur. A, Iguanodon bernissartensis (IRSNB 1731), lateral view. B, Bactrosaurus
johnsoni (AMNH 6553), lateral view.
801
state 0 state 1
A B
Figure J.3. Character 289. Distal extension of the cnemial crest (rectangles) of the tibia.
A, Iguanodon bernissartensis (unrecorded catalog number), medial view. B, Parasaurolophus
cyrtocristatus (FMNH P27393), lateral view.
802
state 0 state 1
A B
Figure J.4. Character 290. Expansion of the distal end (rectangles) of the fibula. A, Bactrosaurus
johnsoni (AMNH 6553), lateral view. B, Tsintaosaurus spinorhinus (IVPP V725), lateral view.
803
A B C
D E F
Figure J.5. Character 291. Development of the medial platform (rectangles) of the astragalus. Specimens displayed in cranial view.
A, Iguanodon bernissartensis (IRSNB 1534). B, Edmontosaurus annectens (CMN 8399). C, Brachylophosaurus canadensis (MOR 794).
D, Parasaurolophus cyrtocristatus (FMNH P27393). E, Lambeosaurus lambei (ROM 1218). F, Tsintaosaurus spinorhinus (IVPP V725).
Abbreviations: as = astragalus; cc = calcaneum; fb = fubula; tb = tibia.
804
Table J.1. Character 294. Ratio (R) between the proximodistal length and the
mediolateral width at midshaft of metatarsal III, for the specimens sampled to study the
variation of this character. Included is the character state coding derived from K-Means

(State)
AMNH 5271 cf. Saurolphus osborni 4.16 0
CUST JV1236 Charonosaurus jiayinensis 4.65 1
PIN 2232/10 Probactrosaurus gobiensis 4.85 1
FMNH P27383 Lophorhothon atopus 4.73 1
MPCA unrecorded no. Hadrosauridae Salitral Moreno 4.15 0
805
Table J.1—continued.

(State)
SEPCCP 1-406 Velafrons cohauliensis 4.84 1
806
6.0
−80
−90
BIC
Length/width Ratio of Metatarsal III
−110
5.5
−130
2 4 6 8
Number of Clusters
5.0
4.5
4.0
3.5
Am Ba Br Ch Cl Coi Cos Eda Edr Gi Grn Hy Hys Iga Igb Lh Lml Ma Ou Pac Pb Pr Saa Se Sh Ts Ve cfSao sal x
Taxon
Figure J.6. Character 294 Boxplot showing the distribution of the length/width ratio of metatarsal III for a sample of iguanodontoidean
ornithopods. The inserted smaller graph shows the Bayesian information criterion (BIC) for various numbers of clusters in which the data
can be classified. The photograph shows how the two variables of this ratio were measured, exemplified in a third metatarsal of Charonosaurus
jiayinensis (CUST JV1236).
807
A B C
Figure J.7. Character 295. Length/width proportions of pedal phalanx II2 (rectangles). A, Iguanodon bernissartensis (IRSNB 1723),
dorsal view. B, Corythosaurus intermedius (ROM 845) dorsal view. C, Brachylophosaurus canadensis (MOR 1071), dorsal view.
808
A B C D
Figure J.8. Character 296. Length/width proportions of pedal phalanges III2 and III3. Specimens displayed in dorsal view.
A, Shantungosaurus giganteus (GMV 1780). B, Edmontosaurus regalis (CMN 2289). C, Hypacrosaurus altispinus (CMN 8501).
D, Saurolophus angustirostris (ZPAL MgD-I 157, subadult specimen).
809
A B C D
Figure J.9. Character 297. Length/width proportions of pedal phalanges IV2-IV4. A, Brachylophosaurus canadensis (MOR 1071),
dorsal view. B, Edmontosaurus anenctens (SM R3046), dorsal view. C, Gryposaurus notabilis (ROM 764), dorsal view. D, Shantungosaurus
giganteus (GMV 1780), dorsal view.
810
state 0
state 1
Figure J.10. Character 298. Morphology of the pedal ungual phalanges. A, Gilmoreosaurus
mongoliensis (AMNH 6551), dorsal view. B, Brachylophosaurus canadensis (MOR 1071-1071-
7-31-99-273), dorsal view.
811
state 0
state 1
Figure J.11. Character 299. Absence and presence of plantar ridge (arrow) in pedal ungual phalanges.
A, Velafrons coahuilensis (SEPCCP 1-4061), cranial view. B, Brachylophosaurus canadensis
(MOR 1071-1071-7-31-99-273), cranial view.
812
APPENDIX K
DEFINITIONS AND SYNAPOMORPHIES FOR THE CLADES INFERRED IN THE
WEIGHTED PARSIMONY ANALYSIS
Clade 1. Hadrosauria Von Huene, 1956

Emended definition: the most inclusive clade containing Equijubus normani, but not
Iguanodon spp. and Ouranosaurus nigeriensis.
Recorded temporal range: late Barremian-Aptian to late Maastrichtian
Unambiguous synapomorphies: rostral end of the rostrodorsal process of the nasal
reacheing the rostral margin of the narial foramen (character 87-1; reversed in Kritosaurus
navajovius, Gryposaurus, Secernosaurus koerneri and the Salitral Moreno and Big Bend UTEP
OTU’s); centered dorsolateral promontory of the maxilla, base of dorsal process positioned
slightly caudal to the mid-length of the element (character 98-1); articulation margin of the
occipital condyle facing caudally and divided from the caudal border of the basioccipital by a
shallow cleft (character 160-1); palpebral bone absent or fused to the orbital margin (character
196-0).
Clade 2
Definition: the clade stemming from the most recent common ancestor of
Probactrosaurus spp. and Parasaurolophus walkeri.
Recorded temporal range: Albian to late Maastrichtian
Unambiguous synapomorphies: dentary teeth with primary, secondary and one or two
additional ridges (character 6-1); predentary with short and shallow shelf limited to the
laterocaudal region of the lateral process (character 32-1); ventral margin of the dentary straight
or slightly bowed rostral to the coronoid process (character 46-2; reversed in Eolambia
caroljonesa); and skull up to 25% wider across the postorbitals (character 204-1; reversed in
Bactrosaurus johnsoni, Edmontosaurus and “maiasaurs”, except Wulagasaurus dongi).
Clade 3
Definition: the clade stemming from the most recent common ancestor of Eolambia
caroljonesa and Parasaurolophus walkeri.
Recorded temporal range: early Cenomanian to late Maastrichtian
813
Unambiguous synapomorphies: Unambiguous synapomorphies: dentary marginal
denticles forming a curved and mammillated asymmetrical ledge (character10-1); preacetabular
process of the ilum deflected ventrally 150º or less (character 243-1; reversed in Claosaurus
agilis and Sahaliyania elunchunorum); relatively short pubic peduncle of the ilium and
triangular, with a proximal region that is much craniocaudally wider than the distal end
(character 253-1); iliac peduncle of the ischium forming an angle greater than 115º with the
acetabular margin (character 276-1; reversed in the Big Bend UTEP hadrosaurid); iliac peduncle
of the ischium with acetabular and caudodorsal margins slightly to greatly divergent near the
proximal region of the peduncle (character 278-1).
Clade 4
Definition: the clade stemming from the most recent common ancestor of Protohadros
byrdi and Parasaurolophus walkeri.
Recorded temporal range: middle Cenomanian to late Maastrichtian
Unambiguous synapomorphies: dentary denticles composed of three separate and
rounded knobs aligned labiomesially (character 11-1; reversed in euhadrosaurs, except
Edmontosaurus); mid point of the quadratojugal notch of the quadrate located ventral to the mid-
length of the quadrate (character 125-1; reversed in Maiasaura peeblesorum and lambeosaurines,
except Sahaliyania elunchunorum and Velafrons coahuilensis)
Clade 5
Definition: the clade stemming from the most recent common ancestor of Tanius sinensis
and Parasaurolophus walkeri.
Recorded temporal range: early-middle Coniacian-late Maastrichtian
Unambiguous synapomorphies: absence of ectopterygoid –jugal contact, evidenced by
the lack of an articular facet in the medial side of the jugal (character 175-1; reversed in
Lophorhothon atopus); and moderately developed deltopectoral crest, ratio between the width of
the humerus across the distal fourth of the deltopectoral crest and the width of the distal shaft at
the point of maximum curvature between 1.65 and 1.90 (character 230-1; reversed in
Hadrosaurus foulkii, Claosaurus agilis, Lophorhothon atopus, Brachylophosaurus canadensis
and Maiasaura peeblesorum).
814
Clade 6
Definition: the clade stemming from the most recent common ancestor of Bactrosaurus
johnsoni and Parasaurolophus walkeri.
Recorded temporal range: early-middle Coniacian to late Maastrichtian
Unambiguous synapomorphies: quadratojugal flange of the jugal auricular in shape, with
subparallel concave to nearly straight dorsal and convex ventral margins that converge dorsally
into a short subconical point (character 119-1); caudal surface of the supraoccipital rostrally
inclined and facing caudodorsally (character 169-1); decrease or absence of the dorsal curvature
of the acromion process of the scapula (character 226-1; reversed in Parasaurolophus walkeri)
and supraacetabular process projected lateroventrally between 25% and less than 50% the depth
of the central plate of the ilium (character 248-1).
Clade 7
Definition: the clade stemming from the most recent common ancestor of Gilmoreosaurs
mongoliensis and Parasaurolophus walkeri.
Recorded temporal range: early middle Coniacian to late Maastrichtian
Unambiguous synapomorphies: small lingual projection of the symphyseal region od the
dentary, with a ratio between the labiolingual extension of the rostral end and the maximum
labiolingual width of the dentary up to 1.65 (character 42-0; convergent in Iguanodon and
Ouranosaurus nigeriensis); dorsolateral region of the rostral half of the maxilla located slightly
rostral to the mid-length of the maxilla, base of dorsal process centered around the mid-length of
the maxilla (character 98-2; reversed in Sahaliyania elunchunorum, “parasaurolophs”,
“tsintaosaurs” and Olorotitan ararhensis); ectopterygoid shelf inclined with an angle between
11º and 21º (character 106-1); and ischial peduncle of the ilium formed by two protrusions of
similar size, the caudal one located slighty caudodorsally relative to the other protrusion
(character 254-2)
Clade 8
Definition: the clade stemming from the most recent common ancestor of Telmatosaurus
transsylvanicus and Parasaurolophus walkeri.
Recorded temporal range: Coniacian to late Maastrichtian
Unambiguous synapomorphies: at least three teeth per alveolus arranged dorsoventrally
at mid length of the dental battery (character 2-1); 1.25 or more teeth per cm of maxillary dental
815
battery (character 17-1; reversed in Gryposaurus, Saurolophus angustirostris and
Charonosaurus jiayinensis); medial side of the coronoid process ornamented with only fine
striations (character 50-0); absence of lingual bowing of the occlusal plane, its long axis being
parallel to the lateral side of the dentary (character 53-1; convergent in Protohadros byrdi);
ectopterygoid shelf more than 35% the length of the alveolar margin of the maxilla (character
105-2; reversed in Kerberosaurus manakini and Sahaliyania elunchunorum); subtriangular
ventral surface of the quadrate, lateral condyle rostrocaudally expanded and much larger than the
medial one, the lateral condyle being well offset ventrally relative to the ventral surface of the
medial condyle (character 129-1); and relatively high and long postzygapophyseal processes of
cervical vertebrae, three times or more longer than the width of the neural arch (character 207-1;
reversed in Shantungosaurus giganteus)
Clade 9
Definition: the clade stemming from the most recent common ancestor of Lophorhothon
atopus and Parasaurolophus walkeri.
Recorded temporal range: late Coniacian to late Maastrichtian
Unambiguous synapomorphies: primary ridge of dentary teeth positioned at the middle of
the crown for most teeth, with some teeth within the same dentition displaying a slight caudal
offset of the primary ridge (character 7-1); and rostrodorsal region of the mxilla subconical in
shape, dorsoventrally narrow and forming an angle of 25º or less with the rostral tooth row
(character 95-0)
Clade 10
Definition: the clade stemming from the most recent common ancestor of Claosaurus
agilis and Parasaurolophus walkeri.
Recorded temporal range: late Coniacian to late Maastrichtian
Unambiguous synapomorphies: loss of all but the primary ridge of dentary tooth crowns
(character 6-3; convergent in Eolambia caroljonesa and reversed in Kritosaurus navajovius,
Gryposaurus notabilis, “amurosaurs”, Olorotitan ararhensis and “parasaurolophs”, except
Charonosaurus jiayinensis); ventral-most margin of the supraacetabular process located
craniodorsal to the caudal protrusion of the ischial peduncle (character 247-1); and
proximodistally shortened pedal phalanx II2, being twice as wide mediolaterally as it is
proximodistally long (character 295-0; convergent in Iguanodon)
816
Clade 11. Hadrosauridae Cope, 1869
Definition: the clade stemming from the most recent common ancestor of Hadrosaurus
foulkii and Parasaurolophus walkeri.
Recorded temporal range: middle-late Santonian to late Maastrichtian
Unambiguous synapomorphies: iliac supraacetabular process that projects lateroventrally
between half and three quarters of the dorsoventral depth of the central plate of the iilium
(character 248-2; convergent in Lophorhothon atopus); and craniocaudally short supraacetabular
process, less than 55% the length of the central plate of the ilium (character 249-3; reversed in
saurolophines, except Edmontosaurus and Shantungosaurus giganteus).
Clade 12. Euhadrosauria Weishampel et al., 1993 (Saurolophinae and Lambeosaurinae)
Definition: the clade stemming from the most recent common ancestor of Edmontosaurus
annectens and Parasaurolophus walkeri.
Unambiguous synapomorphies: dentary denticles formed by two, three or more
indentations in their apical margins (character 11-0); long deltopectoral crest that is more than
55% the total length of the humerus (character 229-2; reversed in “maiasaurs”, “kritosaurs”,
except the Salitral Moreno OTU, Prosaurolophus maximus and Charonosaurus jiayinensis);
narrow and subsquared laterodistal margin of the deltopectoral crest of the humerus with an
angle less than 115º (character 231-2; reversed in Brachylophosaurus and Maiasaura); and
caudodorsally oriented dorsal margin of the postacetabular process of the ilium (character 260-1;
reversed in the Big Bend UTEP hadrosaurid).
Clade 13. Saurolophinae Brown, 1914
Definition: the most inclusive clade containing Edmontosaurus annectens but not
Lambeosaurus lambei.
Unambiguous synapomorphies: more than 42 alveolar positions in the dentary (character
1-2; reversed in “maiasaurs” and “kritosaurs”); five or more teeth per alveoli arranged
dorsoventrally at mid length of the dentary dental battery (character 3-3; reversed in “maiasaurs”
and “kritosaurs”, except Gryposaurus); dentary marginal denticles absent or very reduced to
small papillae along the apical half of the dorsal half of the crown (character 10-2; convergent in
Bactrosaurus johnsoni, Charonosaurus jiayinensis and “hypacrosaurs”); 45 or more tooth
817
positions in the maxillary dental battery (character 16-2; reversed “maiasaurs” and “kritosaurs”,
except Kritosaurus navajovius, and convergent in Hypacrosaurus stebingeri and Olorotitan
ararhensis); wide arcuate and symmetrical quadratojugal notch of the quadrate, the ventral half
of the notch being rostroventrally-directed, nearly as straight as its dorsal half (character 127-2;
reversed in “saurolophs”); relatively long parietal, length/width ratio greater than 2.35 (character
155-2; reversed in the Wahweap hadrosaurid and Kritosaurus navajovius); relatively long
exoccipital-supraoccipital shelf above the foramen magnum, being substantially longer (often
twice or more) than the diameter of the foramen magnum (character 171-2; reversed in
“maiasaurs”); caudodorsal margin of the iliac supraacetabular process poorly defined and
discontinuous with the dorsal margin of the proximal region of the postacetabular process, due to
the lack of a well demarcated caudodorsal ridge (character 252-1; reversed in Gryposaurus
latidens, Secernosaurus koerneri and the Salitral Moreno OTU); long iliac peduncle of the
ischium, with a length/width ratio greater than 2 (character 277-2; convergent in Lophorhothon
atopus and reversed in Brachylophosaurus canadensis, Maiasaura peeblesorum and
Gryposaurus notabilis); pubic peduncle of the ischium oriented parallel to the ischial shaft
(character 279-2; reversed in Saurolophus angustirostris); and dorsal acetabular margin of the
pubic peduncle of the ischium located above the level of the dorsal margin of the ischial shaft
(character 281-1; reversed in the Salitral Moreno OTU and S. angustirostris)
Clade 14
Definition: the clade stemming from the most recent common ancestor of Kerberosaurus
manakini and Gryposaurus notabilis.
Recorded temporal range: late Santonian to late Maastrichtian
Unambiguous synapomorphies: dorsal margin of the rostroventral process (or shelf) of
the maxilla forming an angle greater than 25º and up to 39º with the rostral segment of the tooth
row (character 95-1); and rostrodorsal margin of the prefontal rostromedially broad, with
subsquared rostrodorsal corner, the rostral margin being ventrally oriented and forming a 90º
angle with the orbital margin (character 131-1; reversed in Brachylophosaurus canadensis,
Maiasaura peeblesorum and Saurolophus)
Clade 15 (“saurolophs” and “wulagasaurs”)
annectens and Gryposaurus notabilis.
818
Recorded temporal range: late Santonian to late Maastrichtian
Unambiguous synapomorphy: mediolaterally spread out basipterygoid processes of the
basisphenoid, forming a ventral angle of 100º or greater (character 162-1; convergent in
Ouranosaurus nigeriensis, Tanius sinensis and Tsintaosaurus spinorhinus)
Clade 16 (“saurolophs”)
annectens and Saurolophus osborni.
Recorded temporal range: late Campanian to late Maastrichtian
Unambiguous synapomorphies: ventral margin of the dentary with a wide and prominent
bulge rostral to the coronoid process (character 46-3); rostral end of the nasal forming a long and
subrectangular process, with slightly rounded corners, at the contact with the dorsal process of
the premaxilla (character 85-2); rostrocaudally constricted dorsal region of the infratemporal
fenestra caused by a nearly straight and oblique caudoventral margin of the caudodorsal region
of the postorbital (dorsal region of infratemporal fenestra typically subtriangular) (character 137-
1); caudal region of the circumnarial fossa deeply incised and invaginated (character 189-1);
short neural spines in caudal dorsal and sacral vertebrae, being up to 2.10 times longer than the
depth of their centra (character 209-0; convergent in Gryposaurus and the Big Bend UTEP
hadrosaurid); distal region of the scapular blade as deep as or deeper than the proximal region of
the scapula (character 223-1; convergent in Bactrosaurus johnsoni, Brachylophosaurus
canandensis, Maiasaura peeblesorum and lambeosaurines); and absent or very faintly developed
lateroventral protuberance of the proximal region of the ischial peduncle of the pubis (character
272-0; convergent in Iguanodon, Ouranosaurus nigeriensis, “kritosaurs”, Sahaliyania
elunchunorum and Charonosaurus jiayinensis)
Clade 17
annectens and E. regalis.
Recorded temporal range: early to late Maastrichtian
Unambiguous synapomorphies: ratio between the maximum mediolateral width of the
predentary and the maximum rostrocaudal length along the lateral process of less than 1.2
(character 24-1; convergent in “hypacrosaurs”, “parasaurolophs”, Olorotitan ararhensis and
“tsintaosaurs”); predentary rostral surface forming an angle between 56º and 74º with the dorsal
819
margin of the lateral process (character 26-1; convergent in Iguanodon, Ouranosaurus
nigeriensis, Bactrosaurus johnsoni, Kritosaurus navajovius and Gryposaurus monumentensis);
premaxilla with an accessory foramen entering rostrally through the outer (rostral) narial fossa,
located rostral to the premaxillary foramen (character 71-1); premaxilla with an additional
accessory fossa located lateral to the rostral accessory fossa and rostrolateral to the circumnarial
depression, parallel to the lateral border of the oral margin (character 73-1); rostrodorsal region
of the mxilla subconical in shape, dorsoventrally narrow and forming an angle of 25º or less with
the rostral tooth row (character 95-0; convergent in Iguanodon, O. nigeriensis, Lophorhothon
atopus, Claosaurus agilis, Hadrosaurus foulkii and lambeosaurines, except “parasaurolophs”);
ventral margin of the jugal with a relatively wide and shallow embayment (character 117-1;
convergent in “maiasaurs”, Gryposaurus monumentensis and lambeosaurines); central body of
the adult postorbital rostrocaudally expanded, rostrally excavated and and bulging laterally,
containing a hollow inner cavity (character 138-2); moderately long precotyloid process of the
squamosal, with ratio ratio of its length relative to the width of the quadrate cotylus between 0.95
and 1.25 (character 142-1; convergent in iguanodontoidean outgroup taxa to clade 7, the
Wahweap hadrosaurid and “amurosaurs”); rostrocaudally elongate and mediolaterally narrow
rostromedian process of the parietal at the contact with the frontal (character 157-2); extreme
reduction of the length of the postorbital process of the laterosphenoid to 25% or less the length
of the mediodorsal flange of this element (character 168-1); pterygoid with extensive ventral
lamina, so that only relatively small portion of the ventral quadrate process and the rostroventral
process extends beyond the ventral margin of the lamina (character 174-1); dorsal margin of the
rostrum, parallel to the long axis of the external naris, forming an angle up to 30º with the
maxillary tooth row (character 177-0; convergent in O. nigeriensis); absence of cranial crest
(character 190-0; convergent in iguanodontoidean outgroup taxa to clade 7, the Wahweap
hadrosaurid and Shantungosaurus giganteus); and skull more than 25% wider across the
postorbitals relative to its minimum width across the quadrate cotylus of the squamosals
(character 204-0; convergent in Iguanodon, O. nigeriensis, Equijubus normani, B. johnsoni and
“maiasaurs”)
Clade 18
Prosaurolophus maximus and Saurolophus osborni.
820
Recorded temporal range: late Campanian to early-middle Maastrichtian
Unambiguous synapomorphies: tooth crown-rooth angle greater than 110º and up to 135º
(character 9-1; convergent in Protohadros byrdi, Hadrosaurus foulkii, Gryposaurus,
Hypacrosaurus stebingeri and Tsintaosaurus spinorhinus); oral margin of the premaxilla folded
caudodorsally into a thin and recurved margin (character 67-2; convergent in Gryposaurus);
ventral margin of the lacrimal with a prominent convexity rostral to the jugal notch (character
110-1; convergent in Iguanodon and Gryposaurus); quadratojugal notch of the quadrate wide
arcuate and asymmetrical, with the ventral half of the notch having a short horizontal rostral
segment (character 127-1; convergent in Bactrosaurus johnsoni, Gilmoreosaurus mongoliensis,
Telmatosaurus transsylvanicus, Lophorhothon atopus, Claosaurus agilis, Hadrosaurus foulkii,
“tsintaosaurs”, Jaxartosaurus aralensis and Aralosaurus tuberiferus); inclusion of the prefrontal
in the circumnarial fossa (character 133-1; convergent in Maiasaura peeblesorum);
postacetabular process of the ilium at least 10% longer than the central iliac plate (character 255-
2; convergent in Gryposaurus); and pesal phalanges III2-III3 more than three times wider than
they are proximodistally long (character 296-1; convergent in G. notabilis and lambeosaurines)
Clade 19
Definition: the clade stemming from the most recent common ancestor of Saurolophus
osborni and S. angustirostris.
Recorded temporal range: early to middle Maastrichtian
Unambiguous synapomorphies: moderatelly expanded oral margin of the premaxilla,
between 1.65 and twice the minimum width of the skull at its narrowest point (post-oral
constriction) (character 65-1; convergent in Protohadros byrdi, Tanius sinensis, Bactrosaurus
johnsoni, Gilmoreosaurus mongoliensis, Telmatosaurus transsylvanicus, Lophorhothon atopus,
Claosaurus agilis, Hadrosaurus foulkii, Gryposaurus and lambeosaurines, except
“tsintaosaurs”); long caudoventral process of the premaxilla, extending to end medial to the
dorsal region of the prefrontal (character 76-1); subarcuate to smoothly curved rostrodorsal
margin of the prefrontal, with a rostral margin that is rostroventrally oriented and forming an
obtuse angle with the dorsal orbital margin (character 131-0; convergent in Maiasaura
peeblesorum, Brachylophosaurus canadensis, Shantungosaurus giganteus and non-saurolophine
iguanodontoideans); dorsal promontorium on the rostral process of the adult postorbital
(character 136-1; convergent in “amurosaurs” and “parasaurolophs”); (character 137-2;
821
convergent in Jaxartosaurus aralensis, Parasaurolophus walkeri and P. tubicen), 143-1
(convergent in “amurosaurs” and “parasaurolophs”) , nasal articulation surface of the frontal
shaped into a dorsoventrally thickened, tongue-like platform that projects caudodorsally to
overhang the parietal in adults (character 149-1; convergent in “parasaurolophs”); rostrocaudally
constricted dorsal region of the infratemporal fenestra, caused by rostrocaudal shortening of the
caudal process of the postorbital (dorsal region of infratemporal fenestra typically oval); frontal
completely excluded from the orbital margin by an extensive articulation between the prefrontal
and postorbital (character 151-2; convergent in lambeosaurines); mediolaterally narrow and
paddle-like supracranial nasal crest, extending caudal to the occiput (character 193-3); slightly
elongated neural spines in the cranial dorsal vertebrae, forming a “wither-like” region above the
pectoral girdle (character 210-1; convergent in Ouranosaurus nigeriensis, Shantungosaurus
giganteus and lambeosaurines); extremely expanded deltopectoral crest, ratio between the width
of the humerus across the distal fourth of the deltopectoral crest and the width of the distal shaft
at the point of maximum curvature greater than 1.90 (character 230-2; convergent in
Wulagasaurus dongi and lambeosaurines, except Lambeosaurus laticaudus); length/width ratio
of the ulna less than 10 (character 233-0; convergent in Iguanodon, B. johnsoni, C. agilis,
Gryposaurus latidens and Parasaurolophus); and total length/depth ratio of the pubis between
2.70 and 3 (character 274-1; convergent in P. byrdi, T. sinensis, B. johnsoni, G. mongoliensis, T.
transsylvanicus, L. atopus, Maiasaura peeblesorum, Brachylophosaurus canadensis,
Edmontosaurus regalis, Sahaliyania elunchunorum, Corythosaurus and Lambeosaurus)
Clade 20
Definition: the clade stemming from the most recent common ancestor of the Sabinas
OTU and Gryposaurus notabilis.
Recorded temporal range: late Santonian-early Campanian to middle-late Maastrichtian
Unambiguous synapomorphy: presence of squamosal buttress on the caudal margin of the
dorsal end of the of the quadrate, forming a sharp protuberance that hangs from the caudal side
of the dorsal fourth of the quadrate, near the head of the element (character 128-1; convergent in
iguanodontoidean outgroup taxa to clade 11)
Clade 21 (“wulagasaurs”)
Definition: the clade stemming from the most recent common ancestor
Brachylophosaurus canadensis and Gryposaurus notabilis.
822
Recorded temporal range: late Santonian-early Campanian to middle-late Maastrichtian
Unambiguous synapomorphies: dentary dental battery with 31 to 42 alveolar positions
(character 1-1; convergent in the Lambeosaurinae) and at least four teeth per alveoli arranged
dorsoventrally at mid length of the dentary dental battery (character 3-2; convergent in
Corythosaurus and Parasaurolophus)
Clade 22 (“maiasaurs”)
Definition: the clade stemming from the most recent common ancestor Wulagasaurus
dongi and Brachylophosaurus canadensis.
Recorded temporal range: early Campanian to middle-late Maastrichtian
Unambiguous synapomorphies: slightly inclined ectopterygoid shelf, forming an angle
grater than 41º and up to 10º with the caudal segment of the tooth row (character 106-2;
convergent in the Salitral Moreno hadrosaurid); and presence of a rostral apex in the rostral
process of the jugal that is wedge-shaped, elongated, sharply pointed and positioned at mid
distance along the dorsoventral depth of the rostral process (character 111-0; convergent in
iguanodontoidean outgroup taxa to hadrosaurids)
Clade 23
Definition: the clade stemming from the most recent common ancestor of the Wahweap
OTU and Maiasaura peeblesorum.
Recorded temporal range: early to late Campanian
Unambiguous synapomorphies: relatively low dorsolateral region of the rostral half of the
maxilla (character 97-0; convergent in outgroup iguanodontoideans taxa to euhadrosaurs,
Kerberosaurus manakini and Sahaliyania elunchunorum); and caudoventral apex of the rostral
process of the jugal located ventral to the caudal margin of the lacrimal process (character 114-1;
convergent in Olorotitan ararhensis, “amurosaurs” and “parasaurolophs”)
Clade 24
Defnition: the clade stemming from the most recent common ancestor of the
Brachylophosaurus canadensis and Maiasaura peeblesorum.
Recorded temporal range: late Campanian
Unambiguous synapomorphies: subarcuate to smoothly curved rostrodorsal margin of the
prefrontal, with a rostral margin that is rostroventrally oriented and forming an obtuse angle with
the dorsal orbital margin (character 131-0; convergent in Shantungosaurus giganteus,
823
Saurolophus and non-saurolophine iguanodontoideans); and outward flaring of the rostrodorsal
orbital margin of the prefrontal, in the form of a thin and everted wing-like rim around the
rostrodorsal margin of the orbit (character 134-1; convergent in Saurolophus)
Clade 25 (“kritosaurs”)
Definition: the clade stemming from the most recent common ancestor of Kritosaurus
navajovius and Secernosaurus koerneri.
Recorded temporal range: early Campanian to late Campanian-early Maastrichtian
Unambiguous synapomorphies: rostral end of the rostrodorsal process of the nasal not
reaching the rostral margin of the narial foramen (character 87-0; convergent in Iguanodon and
Ouranosaurus nigeriensis); rostroventral process dipping steeply ventrally, forming an angle of
40º or greater with the tooth row, rostral region of the maxilla appearing dorsally “inflated” and
craniocaudally compressed (character 95-2); caudal end of the postorbital caudal ramus
extending to a point rostral to the quadrate cotylus, without overlapping the latter (character 141-
0; convergent in Iguanodon, Equijubus normani and Probactrosaurus); and dorsal margin of the
infratemporal fenestra located substantially more dorsally than the dorsal margin of the orbit,
with the caudal region of the skull roof inclined rostroventrally relative to the frontal plane
(character 201-1).
Clade 26
Definition: the clade stemming from the most recent common ancestor of Gryposaurus
notabilis and Secernosaurus koerneri.
Recorded temporal range: early Campanian to late Campanian-early Maastrichtian
Unambiguous synapomorphy: base of dorsal process and dorsolateral region of the rostral
half of the maxilla located rostral to the mid-length of the maxilla (character 98-3)
Clade 27
notabilis and G. latidens.
Recorded temporal range: early to late Campanian
Unambiguous synapomorphies: five or more teeth per alveoli arranged dorsoventrally at
mid length of the dental battery (character 3-3; convergent in the Sabinas hadrosaurid,
“saurolophs”, Kerberosaurus manakini and Shantungosaurus giganteus); angle between the
crown and the root of dentary teeth greater than 110º and up to 135º (character 9-1; Protohadros
824
byrdi, Hadrosaurus foulkii, Prosaurolophus maximus, Saurolophus, Hypacrosaurus stebingeri
and Tsintaosaurus spinorhinus); less than 1.25 teeth per cm of dental battery (character 17-0;
convergent in iguanodontoidean outgroup taxa to clade 8); and postacetabular process at least
10% longer than the central plate of the ilium (character 255-2; convergent in Prosaurolophus
maximus and Saurolophus)
Clade 28
notabilis and G. monumentensis.
Unambiguous synapomorphy: moderatelly expanded oral margin of the premaxilla,
between 1.65 and twice the minimum width of the skull at its narrowest point (post-oral
constriction) (character 65-1; convergent in hadrosaur outgroup taxa to euhadrosaurs, except
Equijubus normani and Probactrosaurus, and Saurolophus and lambeosaurines); and nasal arch
with its summit located caudodorsal to the caudal margin of the narial foramen (character 91-1)
Clade 29
Definition: the clade stemming from the most recent common ancestor of Secernosaurus
koerneri and the UTEP OTU.
Recorded temporal range: late Campanian-early Maastrichtian
Unambiguous synapomorphies: coracoid with slightly longer scapular facet, ratio
between the length of the lateral margin of the facet for the scapular articulation and the length of
the lateral margin of the glenoid greater than 1 and up to 1.30 (character 215-1; convergent in
Probactrosaurus, Lophorhothon atopus and the Sabinas OTU); brevis shelf at the base of the
postacetabular process of the ilium (character 256-1; convergent in the iguanodontoidean
outgroup taxa to clade 9, Tsintaosaurus spinorhinus and the “hypacrosaurs”); well-developed
medioventral ridge on the medial side of the postacetabular process, forming an oblique and
expanded flange forming the medial margin of an extensive brevis shelf that faces more ventrally
than medially (character 257-3); and presence of a craniocaudally-oriented median ridge on the
laterodorsal surface of the postacetabular process (character 258-1)
Clade 30
Definition: the clade stemming from the most recent common ancestor of the Salitral
Moreno OTU and Secernosaurus koerneri.
825
Unambiguous synapomorphies: angle between the lateral margins of the facet for
scapular articulation and the glenoid greater than 115º (character 216-0; convergent in outgroup
taxa to euhadrosaurs and “tsintaosaurs”, Jaxartosaurus aralensis and Aralosaurus tuberiferus);
asymmetrical iliac supraacetabular process, with a caudally skewed lateral profile (character 250-
0; convergent in Brachylophosaurus canadensis, Maiasaura peeblesorum, outgroup taxa to
euhadrosaurs, “tsintaosaurs”, A. tuberiferus and J. aralensis); caudodorsal margin of the
lateroventral rim of the supraacetabular process forming a well-defined ridge that is continuous
with the dorsal margin of the proximal region of the postacetabular process (character 252-0;
convergent in Gryposaurus latidens, Prosaurolophus maximus and all other iguanodontoideans,
except remaining saurolophines); and thick ischial shaft, thickness greater than 7.5% its length
(character 282-2; convergent in Bactrosaurus johnsoni, Hypacrosaurus altispinus,
Lambeosaurus laticaudus, Corythosaurus and Parasaurolophus cyrtocristatus)
Clade 31. Lambeosaurinae Parks, 1923
Definition: the most inclusive clade containing Lambeosaurus lambei but not
Edmontosaurus annectens.
Unambiguous synapomorphies: absence of maxillary rostrodorsal process, the rostral end
of the maxilla formsing a ventrally sloping rostrodorsal shelf that underlies the premaxilla
(character 92-1); dorsal maxillary process dorsoventrally taller than it is wide, with a peaked and
caudally inclined apex (character 99-1); ventral margin of the jugal with a relatively narrow and
deep embayment (character 117-1; convergent in “maiasaurs” except Wulagasaurus dongi and in
Gryposaurus monumentensis and Edmontosaurus) and oval supratemporal fenestra, wider
mediolaterally than rostrocaudally (character 202-1)
Clade 32
Definition: the clade stemming from the most recent common ancestor of Jaxartosaurus
aralensis and Parasaurolophus walkeri.
Unambiguous synapomorphies: frontal upward doming dorsal to the braincase of
subadult (and possibly also young adult) specimens (character 152-1) and very short parietal,
with a length/width ratio less than 1.40 (character 155-0)
826
Clade 33
Definition: the clade stemming from the most recent common ancestor of Tsintaosaurus
spinorhinus and Parasaurolophus walkeri.
Recorded temporal range: Campanian to late Maastrichtian
Unambiguous synapomorphies: relatively short ectocranial surface of the frontal, with a
length/width ratio greater than 0.4 and up to 0.8 (character 153-1) and sagittal crest of the
parietal that extends along the entire length of the element, fading away at the rostral region
where the parietal is rostrocaudally shorter than it is wide (character 158-1; reversed in
Hypacrosaurus altispinus)
Clade 34 (“tsintaosaurs”)
Definition: the clade stemming from the most recent common ancestor of Tsintaosaurus
spinorhinus and Pararhabdodon isonensis.
Recorded temporal range: Campanian to late Maastrichtian
Unambiguous synapomorphies: great medial projection of the symphyseal region of the
dentary, with a ratio between the labiolingual extension of the rostral end and the maximum
labiolingual width of the dentary greater than 2.85 (character 42-2); rostral edentulous region of
the dentary for articulation with the predentary ranging from having a very subtle concavity
(almost straight) to straight or even displaying a subtle convexity (character 45-0; convergent in
Ouranosaurus nigeriensis, the Sabinas hadrosaurid and “saurolophs”) and dorsally elevated jugal
joint characterized by a distance between the ventral margin of the jugal joint and ectopterygoid
shelf that is nearly equal to the depth of the caudal segment of the maxilla and a caudal margin of
the joint flush with the caudal margin of the rostrodorsal promontory of the lateral side of the
maxilla (character 100-3)
Clade 35
Definition: the clade stemming from the most recent common ancestor of Olorotitan
ararhensis and Parasaurolophus walkeri.
Unambiguous synapomorphies: dentary tooth crowns with a primary ridge accompanied
by one or two faint and shorter ridges (character 6-2; convergent in Talmatosaurus
transsylvanicus, Lophorhothon atopus, Kritosaurus navajovius and Gryposaurus notabilis); ratio
827
between the dorsoventral depth of the predentary rostral face (excluding the median ventral
process) and the length of the lateral process of 0.38 or less (character 25-1; convergent in
Protohadros byrdi); predentary rostral surface forming an angle of 40º or less with the dorsal
margin of the lateral process (character 26-3; convergent in Saurolophus osborni and reversed in
Velafrons coahuilensis); laterally curved caudal process of the surangular (character 60-1;
convergent in Ouranosaurus nigeriensis and Bactrosaurus johnsoni); medial margin of the
proximal region of the surangular forming an angle up to 148º with the medial margin of the
proximal region of the caudal process (character 61-1; reversed in Hypacrosaurus stebingeri);
rostrolateral corner of the oral margin of the premaxilla triangular, dorsoventrally expanded,
ventrally-deflected and pointed (character 68-1; convergent in Eolambia caroljonesa, B. johnsoni
and “maiasaurs”, except Wulagasaurus dongi); lacriform and relatively short external
premaxillary naris, caudal constriction occurring abruptly and primarily composed of either a
lateroventral expansion of the caudodorsal process or a dorsal expansion of the caudoventral
process of the premaxilla (character 79-2; reversed in “amurosaurs”, except H. altispinus);
triangular and rostrocaudally compressed dorsolateral region of the rostral half of the maxilla
(character 96-1); dorsomedial margin of the prefrontal developed into a caudodorsally-oriented
crest that extends caudally over the dorsal surface of the frontal and above the prefrontal-
postorbital articulation in adults (character 130-2); and well-developed ventral convexity along
the distal half of the scaular blade (character 222-1; convergent in Prosaurolophus maximus and
reversed in Parasaurolophus cyrtocristatus)
Clade 36 (“amurosaurs” and “parasaurolophs”)
Definition: the clade stemming from the most recent common ancestor of Corythosaurus
casuarius and Parasaurolophus walkeri.
Unambiguous synapomorphies: rostral apex of the rostral process of the jugal reduced to
a blunt convexity (character 111-2) and dorsal margin of the infratemporal fenestra lying slightly
or substantially below the level of the dorsal margin of the orbit, the caudal region of the skull
roof being subhorizontal or slightly slopping caudoventrally relative to the frontal plane of the
skull (character 201-2)
Clade 37 (“amurosaurs”)
828
casuarius and Hypacrosaurus altispinus.
Unambiguous synapomorphies: moderately expanded oral margin of the premaxilla,
becoming thinner towards the parasagittal plane of the snout, so that the narial fossa merges with
the rostromedial region of the premaxilla adjacent to the contact between both premaxillae
(character 67-1); vertical groove on the caudoventral process of the premaxilla, located rostral to
the dorsal process of the maxilla and extending ventrally from a small opening between the two
premaxillary caudal processes; the groove is bounded rostrally by a triangular ventral projection
of the caudolateral process of the premaxilla (character 75-1); caudal region of the caudoventral
premaxillary process triangular and dorsoventrally expanded, forming a laterally convex lobe,
directed rostrodorsally (character 77-2; reversed in Corythosaurus and Hypacrosaurus
stebingeri); laterodorsal profile of the caudodorsal and caudoventral margins of the external naris
triangular and elongated, with a caudal constriction that gradually closes caudodorsally
(character 79-1); moderately long precotyloid process of the squamosal, being between 0.95 and
1.25 as long as the squamosal cotylus is wide (character 142-1; convergent in Edmontosaurus,
the Wahweap hadrosaurid and iguanodontoidean outgroup taxa to clade 7); and nasal articulation
surface of the frontal shaped into a rostroventrally-slopping platform (character 148-1;
convergent in Brachylophosaurus canadensis and Jaxartosaurus aralensis)
Clade 38 (“corythosaurs”)
casuarius and Lambeosaurus lambei.
Unambiguous synapomorphies: mesial margin of dentary tooth crowns with larger and
either more or less denticles than the distal one (when that occurs, crowns tend to appear
imbricated, so that the mesial margin overlaps the distal margin of the adjacent crown; character
12-1, convergent in Telmatosaurus transsylvanicus); ratio between the maximum mediolateral
width of the predentary and the maximum rostrocaudal length of the lateral process less than 1.2
(character 24-0; convergent in non-lambeosaurine iguanodontoideans, except Ouranosaurus
nigeriensis, Brachylophosaurus canadensis, Maiasaura peeblesorum and Edmontosaurus);
length/width ratio of the external bony naris of the premaxilla greater than 2.85 (character 81-2);
829
and nasal vestibule folded into an S-loop in the enclosed premaxillary passages rostral to the
dorsal process of the maxilla (character 182-1)
Clade 39
casuarius and Hypacrosaurus stebingeri.
Unambiguous synapomorphies: caudal region of the caudoventral provess of the adult
premaxilla dorsoventrally broad and directed caudally (or caudally) and slightly dorsally
(character 77-1); absence of the short median ventral process located between the basipterygoid
processes of the basisphenoid (character 165-0; convergent in Parasaurolophus cyrtocristatus);
and relatively short and stocky humerus, length/width ratio less than 4.25 (character 232-0;
convergent in Tanius sinensis, Saurolophus angustirostris, Olorotitan ararhensis,
Hypacrosaurus altispinus and Parasaurolophus)
Clade 40
casuarius and C. intermedius.
Unambiguous synapomorphies: at least four teeth per alveoli arranged dorsoventrally at
mid length of the dentary dental battery (character 3-2; convergent in “maiasaurs”, “kritosaurs”
except Kritosaurus navajovius and Parasaurolophus); supracranial crest forming an angle
between 110º and 140º with the dorsal margin of the rostrum (character 194-2); relatively broad
proximal constriction of the scapula, more than 60% the depth of the cranial end of the scapula
(character 224-1; convergent in saurolophines, “parasaurolophs” and Olorotitan ararhensis); and
thick ischial shaft, thickness greater than 7.5% its length (character 282-2; convergent in
Bactrosaurus johnsoni, Saurolophus, Hypacrosaurus altispinus, Lambeosaurus laticaudus, L.
magnicristatus and P. cyrtocristatus)
Clade 41
Definition: the clade stemming from the most recent common ancestor of Lambeosaurus
lambei and L. magnicristatus.
830
Unambiguous synapomorphies: premaxillary caudodorsal process with an accessory
rostroventral flange that overlaps the lateral surface of the nasal in the rostral region of a
supracranial crest (character 78-1); prefrontal-nasal contact visible in lateral view along the
caudal and half of the dorsal margin of the prefrontal (character 135-1); complete lateral osseous
closure of the ophthalmic sulcus (V1) of the laterosphenoid (character 167-1; may be convergent
in the Sabinas hadrosaurid); nasals forming a smaller portion of the supracranial crest relative to
the surrounding premaxillae (character 192-1; convergent in Parasaurolophus) and supracranial
crest procumbent and rostrally inclined, forming an angle of less than 110 with the dorsal margin
of the rostrum (character 194-3)
Clade 42 (“hypacrosaurs”)
Definition: the clade stemming from the most recent common ancestor of Amurosaurus
riabinini and Velafrons coahuilensis.
Unambiguous synapomorphies: dentary denticles absent or very reduced to small papillae
along the apical half of the dorsal half of the crown (character 10-2; convergent in
Charonosaurus jiayinensis, all saurolophines except Gryposaurus latidens and the Wahweap
hadrosaurid and Bactrosaurus johnsoni); absence of very poor development of the ventral
transverse ridge between the basipterygoid processes of the basisphenoid (character 164-0); and
subtriangular distal end of the fibula, with a straight or slightly concave cranial margin (character
290-0; convergent in all other iguanodontoideans, except Tanius sinensis)
Clade 43
Definition: the clade stemming from the most recent common ancestor of Amurosaurus
riabinini and Sahaliyania elunchunorum.
Recorded temporal range: late Maastrichtian
Unambiguous synapomorphies: the proximal edentulous slope of the dentary with a
length between 32% and 45% the length of the dental battery (character 37-2; convergent in
Maiasaura peeblesorum, the Wahweap hadrosaurid, Gryposaurus, Saurolophus osborni,
Shantungosaurus giganteus and Olorotitan ararhensis); rostral end of the dentary ventrally
deflected more than 25º relative to the occlusal plane (character 40-2; convergent in Protohadros
byrdi, Gryposaurus, Kritosaurus navajovius and “tsintaosaurs”); ventral deflection of the dentary
originating near the mid-length of the dental battery (character 41-2; convergent in
831
Ouranosaurus nigeriensis, Eolambia caroljonesa, P. byrdi, G. latidens, S. giganteus,
Hypacrosaurus stebingeri, Parasaurolophus and “tsintaosaurs”); and sagittal crest of the parietal
that only extends along the caudal half of the parietal, with a flattened rostral half of the dorsal
surface of the parietal (character 158-2; convergent in Charonosaurus jiayinensis)
Clade 44
Definition: the clade stemming from the most recent common ancestor of Hypacrosaurus
altispinus and Velafrons coahuilensis.
Unambiguous synapomorphies: presence of brevis shelf at the base of the iliac
postacetabular process (character 256-1; convergent in Tsintaosaurus spinorhinus,
Secernosaurus koerneri, the Salitral Moreno and UTEP OTU’s and iguanodontoidean outgroup
taxa to clade 8) and well-defined ridge forming the medial margin of a medioventrally-facing
shelf in the postacetabular process, that is progressively expanded mediolaterally towards the
caudal end (character 257-2)
Clade 45
Definition: the clade stemming from the most recent common ancestor of Hypacrosaurus
altispinus and Lambeosaurus laticaudus.
Recorded temporal range: Campanian to early Maastrichtian
Unambiguous synapomorphy: thick ischial shaft, thickness greater than 7.5% its length
(character 282-2; convergent in Bactrosaurus johnsoni, Secernosaurus koerneri, the Salitral
Moreno OTU, Corythosaurus, Lambeosaurus magnicristatus and Parasaurolophus
cyrtocristatus)
Clade 46 (“parasaurolophs”)
Definition: the clade stemming from the most recent common ancestor of Charonosaurus
jiayinensis and Parasaurolophus walkeri.
Unambiguous synapomorphies: jugal with wider orbital margin and relatively constricted
ventral margin of the infratemporal fenestra (character 123-0; convergent in Iguanodon,
Ouranosaurus nigeriensis, Telmatosaurus transsylvanicus, Sahaliyania elunchunorum and
Velafrons coahuilensis); nasal articulation surface of the frontal shaped into a dorsoventrally
thickened, tongue-like platform that projects caudodorsally to overhang the parietal in adults
832
(character 149-1; convergent in Saurolophus); greatly shortened ectocranial surface of the
frontal, with a length/width ratio less than 0.4 (character 153-2); relatively short metacarpal V, so
that it is less than twice as long as it is proximally wide (character 238-0; convergent in
iguanodontoidean outgroup taxa to clade 10); short postacetabular process, up to 80% the length
of the central iliac plate (character 255-0; convergent in Equijubus normani, Probactrosaurus,
Gilmoreosaurus mongoliensis and Velafrons coahuilensis); long ischial “foot”, 25% or more the
length of the shaft of the ischium (character 284-1; convergent in Hypacrosaurus altispinus); and
cranioventrally directed long axis of the ischial “foot” (character 286-1; convergent in
Bactrosaurus johnsoni and Lambeosaurus laticaudus)
Clade 47
Parasaurolophus cyrtocristatus and P. walkeri.
Unambiguous synapomorphies: at least four teeth per alveoli arranged dorsoventrally at
mid length of the dentary dental battery (character 3-2; convergent in Corythosaurus,
“kritosaurs” except Gryposaurus and “maiasaurs”); dorsoventrally deep and craniocaudally long
deltoid ridge of the scapula, with a well demarcated ventral margin (character 228-1; convergent
in in Bactrosaurus johnsoni, Gilmoreosaurus mongoliensis, and saurolophines, except
Shantungosurus giganteus); relatively short and stocky humerus, length/width ratio less than
4.25 (character 232-0; convergent in Tanius sinensis, Saurolophus angustirostris, Olorotitan
aralensis, Corythosaurus and Hypacrosaurus stebingeri); length/width ratio of the ulna less than
10 (character 233-0; convergent in Iguanodon, Bactrosaurus johnsoni, Claosaurus agilis,
Gryposaurus latidens and Saurolophus) and supraacetabular process of the ilium that projects
lateroventrally to overlap totally or at least half of the lateral ridge of the caudal protuberance of
the ischial peduncle (character 248-3; convergent in Saurolophus)
Clade 48
Parasaurolohus walkeri and P. tubicen.
Unambiguous synapomorphy: rostrocaudally constricted dorsal region of the
infratemporal fenestra, caused by rostrocaudal shortening of the caudal process of the postorbital
833
(dorsal region of infratemporal fenestra typically oval) (character 137-2; convergent in
Jaxartosaurus aralensis and Saurolophus)
834
REFERENCES
Abramoff, M.D., Magelhaes, P. J., Ram, S. J. 2004. Image Processing with ImageJ.
Biophotonics International 11: 36-42.
Alba, D. M., Agustí, J., and Moyà-Solà, S. 2001. Completeness of the mammalian fossil
record in the Iberian Neogene. Paleobiology 27: 79-83.
Alonso, R. 1980. [Dinosaur footprints (Ornithopoda, Hadrosauridae) in the Upper

Cretaceous of northern Argentina.] Acta Geológica Lilloana 15: 55-63 [Spanish].
Alonso, R., and Marquillas, R. 1986. [New locality of dinosaur footprints and the finding of
bird footprints in the Yacoraite Formation (Maastrichtian of Argentina).] Cuarto
Congreso de Paleontología y Estratigrafía, Actas 2: 33-41. [Spanish].
Anonymous, 1922. The first mounted skeleton of a dinosaur, Hadrosaurus. Natural History
22: 382-393.
Apesteguia, S. 2005. A late Campanian sphenodontid (Reptilia, Diapsida) from northern

Patagonia. Comptes Rendus Palevol 4: 663-669.
Averianov, A. O., and Archibald, J. D. 2003. Mammals from the Upper Cretaceous Aitym
Formation, Kyzylkum desert, Uzbekistan. Cretaceous Research 24: 171-191.
Averianov, A. O. 2007. Theropod dinosaurs from Late Cretaceous deposits in the

northeastern Aral Sea region, Kazakhstan. Cretaceous Research 28: 532-544.
Averianov, A. O., and Nessov, L. 1995. A new Cretaceous mammal from the Campanian of
Kazakhstan. Neues Jahrbuch für Geologie und Paläontologie, Montshefte 1995: 65-
74.
Benton, M. J. and Storrs, G. W. 1994. Testing the quality of the fossil record:
paleontological knowledge is improving. Geology 22: 111-114.
Berkey, C. P. and Morris, F. K. 1927. Natural History of Central Asia, Volume II. Geology
of Mongolia. 475 pp. The American Museum of Natural History, New York.
Blakey, R. C. 2001. Regional paleogeographic views of Earth history. University of

Northern Arizona Geology Department Web site, http://jan.ucc.nau.edu/-
rcb7/globaltext.html.
Bolotsky, Y., and Kurzanov, S.K. 1991. The hadrosaurs of the Amur Region [in Russian].
In: V. G. Moiseyenko, [Geology of the Pacific Ocean Border], 94-103. Amur
KNII, Blagoveschensk. [In Russian]
835
Bolotsky, Y. L., and Godefroit, P. 2004. A new hadrosaurine dinosaur from the Late
Cretaceous of Far Eastern Russia. Journal of Vertebrate Paleontology 24: 351-365.
Bonaparte, J. F. 1996. Cretaceous tetrapods from Argentina. Münchner

Geowissenschaftliche Abhandlungen 30: 73-130.
Bonaparte, J. F., and Rougier, G. 1987. The late Cretaceous fauna of Los Alamitos,
Patagonia, Argentina, Part 7 – the hadrosaurs. Revista del Museo Argentino de
Ciencias Naturales 3: 155-161.
Bonaparte J.F., Franchi, M. R., Powell, J. E., Sepulveda, E. G. 1984. [The Alamitos
Formation (Campanian-Maastrichtian) from southeastern Rio Negro, with
description of Kritosaurus australis n. sp. (Hadrosauridae). Paleogeographical
meaning of the vertebrates]. Revista de la Asociacion Geologica de Argentina 39:
284–299. [In Spanish]
Bond, J. E., and Opell, B. D. 1998. Testing adaptive radiation and key innovation
hypotheses in spiders. Evolution 52: 403-414.
Bookstein, F. L., Chernoff, B. C., Elder, R. L., Humphries, J. M., Smith, G. R., and Strauss,
R. E. 1985. Morphometrics in evolutionary biology. Academy of Natural Sciences
of Philadelphia Special Publication 15: 1-277.
Bremer, K. 1988. The limits of amino acid sequence data in angiosperm phylogenetic
reconstruction. Evolution 42: 795-803.
Brett-Surman, M. K. 1979. Phylogeny and palaeobiogeography of hadrosaurian dinosaurs.

Nature 277: 560-562.
Brett-Surman, M. K. 1989. A revision of the Hadrosauridae (Dinosauria: Ornithischia)

and their evolution during the Campanian and Maastrichtian. 272 pp. Unpublished
Ph. D dissertation, George Washington University, Washington D.C.
Brett-Surman, M. K., and Wagner, J. 2007. Discusión of carácter análisis of the

appendicular anatomy in Campanean and Maastrichtian North American
hadrosaurids – variation and ontogeny. In Carpenter, K. (ed.), Horns and Beaks:
Ceratopsian and Ornithopod Dinosaurs, 135-169. Indiana University Press,
Bloomington.
Brown, B. 1910. The Cretaceous Ojo Alamo beds of New Mexico with description of the
new dinosaur genus Kritosaurus. American Museum of Natural History Bulletin 28:
267-274.
Brown, B. 1912. A crested dinosaur from the Edmonton Cretaceous. American Museum of
Natural History Bulletin 31: 131-136.
836
Brown, B. 1913. A new trachodont dinosaur, Hypacrosaurus, from the Edmonton
Cretaceous of Alberta. American Museum of Natural History Bulletin 32: 395-406.
Brown, B. 1914. Corythosaurus casuarius, a new crested dinosaur from the Belly River
Cretaceous, with provisional classification of the family Trachodontidae. Bulletin of
the American Museum of Natural History 33: 559-565.
Brown, B. 1916. A new crested trachodont dinosaur Prosaurolophus maximus. Bulletin of

Buffetaut, E. 1995. An ankylosaurid dinosaur from the Upper Cretaceous of Shandong

(China). Geological Magazine 132: 683-692.
Buffetaut E., and Tong-Buffetaut, H. 1993. Tsintaosaurus spinorhinus Young and Tanius
sinensis Wiman: a preliminary comparative study of two hadrosaurs (Dinosauria)
from the Upper Cretaceous of China. Comptes Rendus de l'Academie de Sciences II
317: 1255–1261.
Cantino, P. D., and de Queiroz, K. 2007. PhyloCode: a phylogenetic code of biological

nomenclature. Version 4b. http://www.ohiou.edu/phylocode.html
Carpenter, K., Dilkes, D., and Weishampel, D. B. 1995. The dinosaurs of the Niobrara
Chalk Formation (Upper Cretaceous, Kansas). Journal of Vertebrate Paleontology
15: 275-297.
Casal, G., Martínez, R., Luna, M., Sciutto, J. C., and Lamanna, M. 2007. [Aeolosaurus
colhuehuapensis sp. nov. (Sauropoda, Titanosauria) from the Bajo Barreal
Formation, Upper Cretaceous of Argentina]. Revista Brasileira de Paleontologia
10:53-62. [In Spanish]
Casamiquela, R. M. 1964. [On a hadrosaurid dinosaur from Argentina.] Ameghiniana

3:285-308 [Spanish].
Casamiquela, R. M. 1980. [Ecological and zoogeographical considerations on the

vertebrates from the Maastrichtian of northern Patagonia.] Memoir de la Societé
Géologique de France, Nouvelle série 139: 53-55 [French].
Casanovas-Cladellas, M. L., Santafé-Llopis, J. V., and Isidro-Llorens, A. 1993.

[Pararhabdodon isonense n.gen. n.sp. (Dinosauria). Morphology, radio-
tomographic study and biomechanic considerations]. Paleontologia i Evolució 26-
27: 121-131. [In Spanish]
Casanovas-Cladellas, M. L., Santafé-Llopis, J. V., and Pereda-Suberbiola, X. 1997. [New

remains of Pararhabdodon isonensis (Dinosauria, Hadrosauridae) and a synthesis
of the assemblage of material discovered in the Upper Cretaceous of Catalonia].
837
Second Workshop of Vertebrate Paleontology, Abstracts (unpaginated). Espéraza-
Quillan. [In French]
Casanovas, M.L, Pereda-Suberbiola, X., Santafé, J.V., and Weishampel, D.B. 1999. First
lambeosaurine hadrosaurid from Europe: palaeobiogeographical implications.
Geological Magazine 136: 205-211.
Case, J. A., Martin, J E., Chaney, D. S., Reguero, M., Marenssi, S. A., Santillana, S. M.,
and Woodburne, M. O. 2000. The first duck-billed dinosaur (Family Hadrosauridae)
from Antarctica. Journal of Vertebrate Paleontology 20: 612-614.
Chapman, R. E. 1997. Technology and the study of dinosaurs. In: J. O. Farlow and M. K.
Brett-Surman (eds.), The Complete Dinosaur, 112-135. Indiana University Press,
Bloomington.
Chapman, R. E., and Brett-Surman, M. K. 1990. Morphometric observations on

hadrosaurid ornithopod ornithopods. In: K. Carpenter and P. J. Currie, Dinosaur
Systematics: Approches and Perspetives, 163-178. Cambridge University Press,
Cambridge.
Chen, P. 1983. A survey of the non-marine Cretaceous of China. Cretaceous Research 4:

123-143.
Clark, J. M., Maryanska, T., and Barsbold, R. 2004. Therizinosauroidea. In: D.B.
Weishampel, P. Dodson and H. Osmólska (eds.), The Dinosauria, second edition,
151-164. University of California Press, Berkeley.
Company, J., Galobart, A., and Gaete, R. 1998. First data on the hadrosaurid dinosaurs
(Ornithischia, Dinosauria) from the Upper Cretaceous of Valencia, Spain. Oryctos
1: 121-126.
Cootes, T. F., Taylor, C. J., Cooper, D. H., and Graham, J. 1995. Active shape models, their
training and application. Computer Vision and Image Understanding 61: 38-59.
Cope, E. D. 1869. Sinopsis of the extinct Batracia, Reptilia and Aves of North America.
Transactions of the American Philoshopical Society 14: 1-252.
Cope, E. D. 1883. On the characters of the skull in the Hadrosauridae. Proceedings of rhe
Academy of Natural Sciences of Philadelphia 35: 97-107.
Coria, R. A. 1999. Ornithopod dinosaurs from Neuquén Group, Patagonia, Argentina:

phylogeny and biostratigraphy. In: Y. Tomida, T. H. Rich, and P. Vickers-Rich
(eds.), Proceedings of the Second Gondwanan Dinosaur Symposyum, National
Science Museum Monographs, 15: 47-60.
838
Cox, C. B. 2000. Plate tectonics, seaways and climate in historical biogeography of
mammals. Memorias del Instituto Oswaldo Cruz, Rio de Janeiro 95: 509-516.
Creisler, B. S. 2007. Deciphering duckbills: a history in nomenclature. In: K. Carpenter

(ed.), Horns and Beaks: Ceratopsian and Ornithopod Dinosaurs, 185-210. Indiana
University Press, Bloomington.
Crisci, J. V., L. Katinas and P. Posadas. 2003. Historical Biogeography. An Introduction.

250 pp. Harvard University Press, Cambridge.
Cruzado-Caballero, P., Canudo, J. I., and Ruíz-Omeñaca, J. I. 2005. [New evidence of the
presence of lambeosaurine hadrosaurs (Dinosauria) in the Upper Maastrichtian of
the Iberian Peninsula (Arén, Huesca)]. Geogaceta 38: 3-6. [In Spanish]
Cruzado-Caballero, P., Canudo, J. I., and Ruíz-Omeñaca, J. I. 2007. [Hadrosaurid dinosaurs

(Ornithischia: Ornithopoda) from the later Maastrichtian (Late Cretaceous) of Arén,
(Huesca, Spain)]. IV Jornadas Internacionales sobre Paleontología de Dinosaurios
y su Entorno, Salas de los Infantes, Burgos 55-56. [In Spanish]
Currie, P. J. and Eberth, D. A. 1993. Paleontology, sedimentology and paleoecology of the

Iren Dabasu Formation (Upper Cretaceous), Inner Mongolia, People’s Republic of
China. Cretaceous Research 14: 127-144.
Currie, P. J., Nadon, G. C., and Lockley, M. G. 1991. Dinosaur footprints with skin
impressions from the Cretaceous of Alberta and Colorado. Canadian Journal of
Earch Sciences 28: 102-115.
Dalla Vecchia, F. 2006. Telmatosaurus and the other hadrosaurids of the Cretaceous
European Archipelago. An overview. Natura Nascosta 32: 1-55.
Dalla Vecchia, F. 2007. European hadrosauroids. IV Jornadas Internacionales sobre

Dinosaurios y su Entorno, Salas de los Infantes 25-28.
Davies, K. L. 1983. Hadrosaurian dinosaurs of Big Bend National Park. 231 pp.
Unpublished Master’s Thesis, university of Texas at Austin, Austin.
Donoghue, M. J., Olmstead, R.G., Smith, J.F., and Palmer, J.D. 1992. Phylogenetic
relationships of Dipsacales based on rbcL sequences. Annals of the Missouri
Botanical Garden 79: 672-685.
Dodson, P. 1971. Sedimentology and taphonomy of the Oldman Formation (Campanian),

Dinosaur Provincial Park, Alberta (Canada). Palaeogeography, Palaeoclimatology,
Palaeoecology 10: 21-74.
Dodson, P. 1975. Taxonomic implications of relative growth in lambeosaurine hadrosaurs.

Systematic Zoology 24: 37-54.
839
Dryden, I. L., and Mardia, K. V. 1998. Statistical Shape Analysis. John Wiley and Son.
Duellman, W. E. 2001. The Hylid Frogs of Middle America, 2 volumes. Society for the
Study of Amphibians and Reptiles, Ithaca, New York.
Dyke, G. J., and Malakhov, D. V. 2004. Abundance and taphonomy of dinosaur teeth and
other vertebrate remains from the Bostobynskaya Formation, north-east Aral Sea
region, Republic of Kazakhstan. Cretaceous Research 25: 669-674.
Egi, N., and Weishampel, D. B. 2002. Morphometric analyses of humeral shapes in

hadrosaurids (Ornithopoda, Dinosauria). Senckenbergiana Lethaea 82: 43-58.
Evans, D. C. 2006. Nasal cavity homologies and cranial crest function in lambeosaurine
dinosaurs. Paleobiology 32: 109-125.
Evans, D. A. 2007a. Ontogeny and evolution of lambeosauine dinosaurs (Ornithischia:

Hadrosauridae). 529 pp. Unpublishd Ph. D Dissertation, University of Toronto,
Toronto.
Evans, D. A. 2007b. Phylogeny of lambeosaurine dinosaurs using parsimony and Bayesian

likelihood approaches. Journal of Vertebrate Paleontology 27: 72A.
Evans, D. C., and Reisz, R. R. 2007. Anatomy and relationships of Lambeosaurus

magnicristatus, a crested hadrosaurid dinosaur (Ornithischia) from the Dinosaur
Park Formation, Alberta. Journal of Vertebrate Paleontology 27: 373-393.
Evans, D. C., Forster, C. A., and Reisz, R. R. 2005. The type specimen of Tetragonosaurus
erectofrons (Ornithischia: Hadrosauridae), and the taxonomic identification of
juvenile lambeosaurines form the Dinosaur Park Formation, Alberta, Canada. In: P.
J. Currie and E. B. Koppelhus (eds.), Dinosaur Provincial Park: A Spectacular
Ancient Ecosystem Revealed, 349-366. Indiana University Press, Bloomington.
Everhart, M. J., and Ewell, K. 2006. Shark-bitten dinosaur (Hadrosauridae) caudal

vertebrae from the Niobrara Chalk (Upper Coniacian) of western Kansas.
Transactions of the Kansas Academy of Science 109: 27-35.
Felsenstein, J. 1985. Confidence limits on phylogenies: An approach using the bootstrap.

Evolution 39: 783-791.
Fitch, W. M. 1971. Toward defining the course of evolution: minimum change for a
specific tree topology. Systematic Zoology 20: 406-416.
Foote, M., and Raup, D. M. 1996. Fossil preservation and the stratigraphic ranges of taxa.
Paleobiology 22: 121-140.
840
Foulke, W. P. 1859. Remarks on fossil bones shells, and wood, particularly historical
remarks on the collection of Hadrosaurus foulkii Leidy. Proceedings of the
National Academy of Natural Sciences of Philadelphia 10: 213-215.
Forster, C. A., and Sereno, P. C. 1994. Phylogeneny of the Hadrosauridae. Journal of

Vertebrate Paleontology 14: 25A.
Forster, C. A. 1997a. Phylogeny of the Iguanodontia and Hadrosauridae. Journal of

Vertebrate Paleontology 17: 47A.
Forster, C. A. 1997b. Hadrosauridae. In: P. J. Currie and K. Padian (eds.), Encyclopedia of

Dinosaurs, 293-299. Academic Press, San Diego.
Fraley, C., and Raftery, A. E. 2002. Model-based clustering, discriminant analysis, and
density estimation. Journal of rhe American Statistical Association 97: 611-630.
Fraley, C., and Raftery, A. E. 2003. Enhanced model-based clustering, density estimation
and discriminant analysis software: MCLUST. Journal of Classification 20: 263-
286.
Fraley, C., and Raftery, A. E. 2006. MCLUST Version 3 for R: Normal Mixture Modeling
and Model-Based Clustering, Technical Report no. 504, Department of Statistics,
University of Washington.
Fraley, C., Raftery, A. E., and Wehrens, R. 2005. Incremental Model-Based clustering for
large datasets with small clusters. Journal of Computational and Graphical
Statistics 14: 1-18.
Gates, T. A. 2007. Taxonomy, Biogeography, and Paleoecology of North American

Hadrosaurid (Ornithopoda) Dinosaurs. 299 pp. Unpublished Ph. D Dissertation,
The University of Utah, Salt Lake City.
Gates, T. A, and Evans, D. A. 2005. Biogeography of Campanian hadrosaurid dinosaurs

from western North America. In: D. R. Braman, F. Therrien, E. B. Koppelhus and
W. Taylor, W. (eds.), Dinosaur Park Symposium short papers, abstracts, and
programs, 33-39. Royal Tyrrell Museum of Paleontology, Drumheller.
Gates, T. A. and Sampson, S. D. 2007. A new species of Gryposaurus (Dinosauria:

Hadrosauridae) from the late Campanian Kaiparowits Formation, southern Utah,
USA. Zoological Journal of the Linnean Society 151: 351-376.
Gates, T. A., Sampson, S. D., Delgado de Jesús, C. R., Zanno, L. E., Eberth, D.,
Hernández-Rivera, R., Aguillón Martínez, M. C., and Kirkland, J. I. 2007.
Velafrons coahuilensis, a new lambeosaurine hadrosaurid (Dinosauria:
Ornithopoda) from the late Campanian Cerro del Pueblo Formation, Coahuila,
Mexico. Journal of Vertebrate Paleontology 27: 917-930.
841
Genise, J. F., Sciutto, J. C., Laza, J. H., González, M. G., and Bellosi, E. S. 2002. Fossil bee
nests, coleopteran pupal chambers and tuffaceous paleosols from the Late
Cretaceous Laguna Palacios Formation, Central Patagonia (Argentina).
Palaeogeography, Palaeoclimatology, Palaeoecology 177: 215-235.
Gilmore, C. W. 1924. A new species of hadrosaurian dinosaur from the Edmonton

Formation (Cretaceous) of Alberta, Canada. Bulletin of the Department of Mines,
Geological Survey of Canada (Geological Series 43) 38: 13-26.
Gilmore, C. W. 1933. On the dinosaurian fauna of the Iren Dabasu Formation. Bulletin of
Godefroit, P., Dong, Z-M., Bultynck, P., Li, H., Feng, L. 1998. Sino-Belgian Cooperative
Program. Cretaceous Dinosaurs and Mammals from Inner Mongolia: 1) New
Bactrosaurus (Dinosauria: Hadrosauroidea) material from Iren Dabasu (Inner
Mongolia, P.R. China). Bulletin de l’Institute Royal des Sciences Naturelles du
Belgique, 68: 1-70.
Godefroit, P., Zan, S., and Jin, L. 2000. Charonosaurus jiayinensis n.g., n.sp., a
lambeosaurine dinosaur from the Late Cretaceous of northeastern China. Comptes
Rendus de l’Academie des Sciences de Paris, Sciences de la Terre et des Planetes
330: 875-882.
Godefroit, P., Zan S., and Jin, L. 2001. The Maastrichtian (Late Cretaceous) lambeosaurine
dinosaur Charonosaurus jiayinensis from north-eastern China. Bulletin de l'Institut
Royal des Sciences Naturelles du Belgique, Sciences de la Terre 71: 119-168.
Godefroit, P., Bolotsky, Y. L., and Alifanov, V. 2003. A remarkable hollow-crested

hadrosaur from Russia : an Asian origin for lambeosaurines. Comptes Rendus
Palevol 2 :143-152.
Godefroit, P., Bolotsky, Y. L., and Van Itterbeeck, J. 2004a. The lambeosaurine dinosaur
Amurosaurus riabinini, from the Maastrichtian of Far Eastern Russia. Acta
Palaeontologica Polonica 49: 585-618.
Godefroit, P., Alifanov, V., and Bolotsky, Y. 2004b. A re-appraisal of Aralosaurus

tuberiferus (Dinosauria, Hadrosauridae) from the Late Cretaceous of Kazakhstan.
Bulletin de l’Institute Royal des Sciences Naturelles du Belgique 74: 139-154.
Godefroit, P., Li, H., and Shang, C.-Y. 2005. A new primitive hadrosauroid dinosaur from
the Early Cretaceous of Inner Mongolia (P. R. China). Comptes Rendus Palevol 4:
697-705.
842
Godefroit, P., Shulin, H., Tingxiang, Y, and Lauters, P. 2008. New hadrosaurid dinosaurs
from the uppermost Cretaceous of northeastern China. Acta Palaeontologica
Polonica 53: 47-74.
Goloboff, P. 1993. Estimating character weights during tree search. Cladistics 9: 83-91.
Gonzalez Riga, B., and Casadio, S. 2000. [First record of Dinosauria (Ornithischia,
Hadrosauridae] in the province of La Pampa (Argentina) and its paleobiogeographic
implications]. Ameghiniana 37: 341-351 [In Spanish].
Gradstein, F. M., Ogg, J. G., and Smith, A. G. 2004. A Geologic Time Scale. Official
website of the International Commission on Stratigraphy (ICS),
http://www.stratigraphy.org.
Hai-lu You and Dodson, P. 2004. Basal Ceratopsia. In: D.B. Weishampel, P. Dodson and
H. Osmólska (eds.), The Dinosauria, second edition, 478-493. University of
California Press, Berkeley.
Hai-lu You, Luo, Z., Shubin, N. H., Witmer, L. M., Tang, Z.-L., and Tang, F. 2003. The
earliest-known duck-billed dinosaur from deposits of late Early Cretaceous age in
northwestern China and hadrosaur evolution. Cretaceous Research 24: 347-355.
Haq, B. U., Hardenbol, J., and Vail, P. R. 1987. Chronology of fluctuating sea levels since
the Triassic. Science 235: 1156-1167.
Hawkins, J. A. 2000. A survey of primary homolgy assessment: different botanists perceive

and define characters in different ways. In: R. Scotland and T. Pennington (eds.),
Homology and Systematics, 22-53. The Systematics Association Special Volume
Series 58, London and New York.
Hawkins, J. A., Hughes, C. E., and Scotland, R. W. 1997. Primary homology assessment,
characters and character states. Cladistics 13: 275-283.
Hay, O. P. 1902. Bibliography and catalogue of the fossil Vertebrata of North. United
States Geological Survey Bulletin 179.
Hay, W. W., DeConto, R. M., Wold, C. N., Wilson, K. M., Voigt, S., Schulx, M., Wold, A.
R., Dullo, W.-C., Ronov, A. B., Balukhovsky, A. N., and Soding, E. 1999.
Alternative global Cretaceous paleogeography. In: E. Barrera and C. C. Johnson
(eds.), Evolution of the Cretaceous Ocean-Climate System. Geological Society of
America, Special Paper 332: 1-48.
Head, J. J. 1998. A new species of basal hadrosaurid (Dinosauria, Ornithischia) from the
Cenomanian of Texas. Journal of Vertebrate Paleontology 18: 718-738.
843
Head, J. J. 2001. A reanalysis of the phylogenetic position of Eolambia caroljonesa
(Dinosauria, Iguanodontia). Journal of Vertebrate Paleontology 21: 392-396.
Hedges, S. B. 2006. Paleogeography of the Antilles and origin of West Indian terrestrial
vertebrates. Annals of the Missouri Bonatical Garden 93: 231-244.
Hennig, W. 1966. Phylogenetic Systematics. University of Illinois Press, Urbana.
Hillis, D. M. 1996. Inferring complex phylogenies. Nature 383: 130-131.
Holtz, T. R. Jr., Chapman, R. E., and Lamanna, M. C. 2004. Mesozoic biogeography of

Dinosauria. In: D. B. Weishampel, P. Dodson, and H. Osmólska (eds.), The
Dinosauria, second edition, 627-642. University of California Press, Berkeley.
Hong, J. and Miyata, T. 1999. Strike-slip origin of Cretaceous Mazhan Basin, Tan-Lu fault
zone, Shandong, east China. The Island Arc 8: 80-91.
Hopson, J. A. 1975. The evolution of cranial display structures in hadrosaurian dinosaurs.

Horne, G. S. 1994. A mid-Cretaceous ornithopod from the central Honduras. Journal of

Vertebrate Paleontology 14: 147-150.
Horner, J. R. 1982. Evidence of colonial nesting and “site fidelity” among ornithischian
dinosaurs. Nature 297: 675-676.
Horner, J. R. 1983. Cranial osteology and morphology of the type specimen of Maiasaura
peeblesorum (Ornithischia: Reptilia) with discussion of its phylogenetic position.
Journal of Vertebrate Paleontology 3: 29-38.
Horner, J. R. 1985. Evidence of polyphyletic origination of the Hadrosauridae (Reptilia:

Ornithischia). Proceedings of the Pacific Division of the American Association for
the Advancement of Science 4: 31-32.
Horner, J. R. 1988. A new hadrosaur (Reptilia, Ornithischia) from the Upper Cretaceous
Judith River Formation of Montana. Journal of Vertebrate Paleontology 8: 314-
321.
Horner, J. R. 1990. Evidence of diphyletic origination of the hadrosaurian (Reptilia:

Ornithischia) dinosaurs. In: K. Carpenter and P. J. Currie (eds.), Dinosaur
Systematics: Perspectives and Approaches, 179-187. Cambridge University Press,
New York.
Horner, J. R. 1992. Cranial morphology of Prosaurolophus (Ornithischia: Hadrosauridae)

with descriptions of two new hadrosaurid species and an evaluation of hadrosaurid
phylogenetic relationships. Museum of the Rockies, Occasional Paper 2: 1-119.
844
Horner, J. R. 2000. Dinosaur reproduction and parenting. Annual Review of Earth and
Planetary Sciences 28: 19-45.
Horner, J. R., and Makela, R. 1979. Nest of juveniles provides evidence of family structure
among dinosaurs. Nature 282: 296-298.
Horner, J. R., and Currie, P. J. 1994. Embryonic, neonatal morphology and ontogeny of a
new species of Hypacrosaurus (Ornithischia, Lambeosauridae) from Montana and
Alberta. In: K. Carpenter, K. F. Hirsch, and J. R. Horner (eds), Dinosaur Eggs and
Babies, 312-336. Cambridge University Press, Cambridge.
Horner, J. R., Ricqlès, A. de, and Padian, K. 2000. Long bone histology of the hadrosaurid
Maiasaura peeblesorum: growth dynamics and physiology based on an ontogenetic
series of skeletal elements. Journal of Vertebrate Paleontology 20: 109-123.
Horner, J. R., Weishampel, D. B., and Forster, C. A. 2004. Hadrosauridae. In: D.B.
Hu, C. 1972. A new hadrosaur from the Cretaceous of Chucheng, Shantung. Acta
Geologica Sinica 2: 179-206.
Hu, C., Zhengwu, C., Qiqing, P. and Xiaosi, F. Shantungosaurus giganteus. 135 pp. Hauyu
Nature Trade, Beijing, China.
Huelsenbeck, J. P. 1994. Comparing the stratigraphic record to estimates of phylogeny.

Huelsenbeck, J. P. and Ronquist, F. 2001. MRBAYES: Bayesian inference of phylogeny.

Bioinformatics 17: 754-755.
von Huene, F. R. 1956. Paläontologie und Phylogenie der Niederen Tetrapoden. 716 pp.
Veb Gustav Fischer Verlag, Jena, Germany.
Hunt, A. P., and Lucas, S. G. 1993. Cretaceous vertebrates of New Mexico. New Mexico
Museum of Natural History and Science Bulletin 2: 77-91.
Hunter, J. P., and Jernvall, J. 1995. The hypocone as a key innovation in mammalian
evolution. Proceedings of the National Academy of Siences 92: 10718-10722.
Hwang, K-G., Huh, M., Lockley, M. G., Unwin, D. M., and Wright, J. L. 2002. New
pterosaur tracks (Pteraichnidae) from the Late Cretaceous Uhangri Formation,
southwestern Korea. Geological Magazine 139: 421-435.
845
Ihaka, R., and Gentleman, R. 1996. R: a computer language for data analysis and graphics.
Journal of Computational and Grafical Statistics 5: 299-314.
Jerzykiewicz, T. and Russell, D. A. 1991. Late Mesozoic stratigraphy and vertebrates of the
Gobi Basin. Cretaceous Research 12: 345-377.
Klassen, E., Srivastava, A., Mio, W., and Joshi, S. 2004. Analysis of planar shapes using
geodesic paths on shape spaces. Institute of Electrical and Electronic Engineers,
Transactions on Pattern Analysis and Machine Intelligence 26: 372-383.
Kirkland, J. I. 1998. A new hadrosaurid from the Upper Cretaceous Cedar Mountain
Formation (Albian-Cenomanian: Cretaceous) of Eastern Utah – the oldest known
hadrosaurid (Lambeosaurinae?). New Mexico Museum of Natural History and
Science Bulletin 14: 283-295.
Kirkland, J. I., Hernández-Rivera, R., Gates, T., Paul, G. S., Nesbitt, S., Inés-Serrano, C.,
and Garcia-de la Garza, J. P. 2006. Large hadrosaurine dinosaurs form the latest
Campanian of Coahuila, Mexico. New Mexico Museum of Natural History and
Kluge, A. G. and Farris, J. F. 1969. Quantitative phyletics and the evolution of anurans.
Systematic Zoology 18: 1-32.
Kordikova, E. G., Polly, P. D., Alifanov, V. A., Rocek, Z., Gunnell, G. F., and Averianov,
A. O. 2001. Small vertebrates form the late Cretaceous and early Tertiary of the
northeastern Aral Sea Region, Kazakhstan. Journal of Paleontology 75: 390-400.
Lambe, L. M. 1902. New genera and species from the Belly River series (Mid-Cretaceous).
In: E. D. Cope (ed.), On vertebrata from the Tertiary and Cretaceous Rocks of the
North West Territory, 25-78. Contributions to Canadian Paleontology, volume 3,
Geological Survey of Canada.
Lambe, L. M. 1914a. On Gryposaurus notabilis, a new genus and species of trachodont

dinosaur from the Belly River Formation of Alberta. Ottawa Naturalist 27: 145-
155.
Lambe, L. M. 1914b. On a new genus and species of carnivorous dinosaur from the Belly
River Formartion of Alberta, with a description of the skull of Stephanosaurus
marginatus from the same horizon. Ottawa Naturalist 27: 13-21.
Lambe, L. M. 1917. A new genus and species of crestless hadrosaur from the Edmonton
Formation of Alberta. Ottawa Naturalist 31: 65-73.
Lambe, L. M. 1918. On the genus Trachodon of Leidy. Ottawa Naturalist 31: 135-139.
846
Lambe, L. M. 1920. The hadrosaur Edmontosaurus from the Upper Cretaceous of Alberta.
Geological Survey Memoir, Canada Department of Mines 120: 1-79.
Langston, W. D. 1960. The vertebrate fauna of the Selma Formation of Alabama. Part 6:
the dinosaurs. Fieldiana: Geology Memoirs 3: 313-363.
LaRock, J. W. 2000. Sedimentology and taphonomy of a dinosaur bonebed from the Upper
Cretaceous (Campanian) Judith River Formation of North Central Montana. 61 pp.
Unpublished Master’s Thesis, Montana State University, Bozeman.
Lee, M. S. Y. 2005. Molecular evidence and marine snake origins. Biological Letters 1
:227-230.
Leidy, J. 1856a. Notices of remains of extinct reptiles and fishes, discovered by Dr. F. V.
Hayden in the badlands of Judith River, Nebraska Territory. Proceedings of the
Academy of Natural Sciences of Philadelphia 8: 72-73.
Leidy, J. 1856b. Notices of extinct Vertebrata discovered by Dr. F. V. Hayden, during the
expedition to the Sioux country the command of Lieunt. G. K. Warren. Proceedings
of the Academy of Natural Sciences of Philadelphia 8: 311-312.
Leidy, J. 1858. Hadrosaurus foulkii, a new saurian from the Cretaceous of New Jersey,
related to Iguanodon. Proceedings of the Academy of Natural Sciences of
Philadelphia 10: 213-218.
Leidy, J. 1859. Hadrosaurus foulkii, a new saurian from the Cretaceous of New Jersey,
related to Iguanodon. American Journal of Science 77: 266-270.
LeLoeuff, J., Buffetaut, E., Mechin, P., and Mechin-Salessy, A. 1992. The first record of
dromaeosaurid dinosaurs (Saurischia, Theropoda) in the Maastrichtian of southern
Europe: paleobiogeographical implications. Bulletin de la Societe Geologique de
France 163: 337-343.
Lewis, P. O. 2001. A likelihood approach to estimating phylogeny from discrete

morphological character data. Systematic Biology 50: 913-925.
Lieberman, B. S. 2000. Paleobiogeography. Kluwer/Plenum, New York.
Lieberman, B. S. 2003. Paleobiogeography: the relevance of fossils to biogeography.

Annual Review of Ecology, Evolution and Systematics 34: 51-69.
Liu, J. and Wu, X. 1990. Cretaceous-Tertiary charophyte fossils in northern Xinjiang. In:
Study on the Micropaleobotany from the Cretaceous-Tertiary of the Oil-bearing
Basins in Some Regions of Qinghai and Xinjiang, 181-229. China Environment
Science Press.
847
López-Martínez, N., Canudo, J.I., Ardévol, L., Pereda Suberbiola, X., Orue-Etxebarría, X.,
Cuenca-Bescós, G., Ruiz Omeñaca, J. I., Murelaga, X & Feist, M. 2001. New
dinosaur sites correlated with Upper Maastrichtian pelagic deposits in the Spanish
Pyrenees: implications for the dinosaur extinction pattern in Europe. Cretaceous
Research 22: 41-61.
Lucas, S., Spielmann, J. A., Sullivan, R. M., Hunt, A. P. and Gates, T. A. 2006.
Anasazisaurus, a hadrosaurian dinosaur from the Upper Cretaceous of New Mexico.
New Mexico Museum of Natural History and Science Bulletin 35: 293-297.
Lull, R. S., and Wright, N. E. 1942. Hadrosaurian dinosaurs of North America. Geological
Society of America Special Papers 40: 1-242.
Luna, M., Casal, G., Martínez, R., Lamanna, M., Ibiricu, L., and Ivany, E. 2003. [The
presence of an Ornithoopda (Dinosauria: Ornithischia) in the upper member of the
Bajo Barreal Formation (Campanian-Maastrichtian?) of southern Chubut] Jornadas
Argentinas de Paleontología de Vertebrados 19: 61 [In Spanish]
Lund, E. K., and Gates, T. A. 2006. A historical and biogeographical examination of

hadrosaurian dinosaurs. In: S. G. Lucas R. M. Sullivan (eds.), Late Cretaceous
vertebrales from the Western Interior. New Mexico Museum of Natural History and
Lydekker, R. 1888. Catalogue of the Fossil Reptilia and Amphibia in the British Museum
(Natural History), Cromwell Road, S. W. Part I. Containing the Orders
Ornithosauria, Crocodilia, Dinosauria, Squamata, Rhynchocephalia, and
Protosauria. Trustees. 309 pp. British Museum (Natural History), London.
Makovicky, P. J. and Norell, M. A. 2004. Troodontidae. In: D.B. Weishampel, P. Dodson

and H. Osmólska (eds.), The Dinosauria, second edition, 184-195. University of
California Press, Berkeley.
Maddison, W. P., Donoghue, M. J., and Maddison, D. R. 1984. Outgroup analysis and
parsimony. Systematic Zoology 33: 83-103.
Maddison, D. R., and Maddison, W. P. 2003. MacClade version 4.0. Sinauer Associates,
Inc., Sunderland, Massachusetts.
Mantell, G. 1825. Notice on the Iguanodon, a newly discovered fossil reptile, from the
sandstone of the Tilgate Forest, in Sussex. Philosophical Transactions of the Royal
Society 115: 179-186.
Marsh, O. C. 1872. Notice on a new species of Hadrosaurus. American Journal of Science

3: 301.
848
Seeley, H. G. 1888. On the classification of the fossil animals commonly named
Dinosauria. Proceedings of the Royal Society of London 43: 165–171.
Marsh, O. C. 1890. Description of new dinosaurian reptiles. American Journal of Science

39: 81-86.
Marsh, O. C. 1892. Notice of new reptiles from the Laramie Formation. American Journal
of Science 257: 449-453.
Maryanska, T., and Osmolska, H. 1981. First lambeosaurine dinosaur from the Nemegt
Formation, Upper Cretaceous, Mongolia. Acta Palaeontologica Polonica 26: 243-
255.
Maryanska, T., Chapman, R. E., and Weishampel, D. B. Pachycephalosauria. In: D.B.

Milner, A. R., and Norman, D. B. 1984. The biogeography of advanced ornithopods

dinosaurs (Archosauria: Ornithischia) – a cladistic-vicariance model. In: W.-E. Reif
and F. Westphal (eds.), Third Symposyum on Mesozoic Terrestrial Ecosystems,
Short Papers, 145-150. Attempto Verlag, Tubingen.
Mio, W., Srivastava, A., and Joshi, S. H. 2007. On shape of plane elastic curves.
Internacional Journal of Computer Vision 73: 307-324.
Morrone, J. J., and Crisci, J. V. 1995. Historical biogeography: introduction to methods.

Annual Review of Ecology and Syatematics 26: 373-401.
Morris, W. J. 1970. Hadrosaurian dinosaur bills: morphology and function. Contributions

in Science, Los Angeles County Museum of Natural History 193: 1-15.
Morris, W. J. 1973. A review of Pacific coast hadrosaurs. Journal of Paleontology 47: 551-
561.
Morris, W. J. 1981. A new species of hadrosaurian dinosaur form the Upper Cretaceous of
Baja California –?Lambeosurus laticaudus. Journal of Paleontology 55: 453-462.
Muller, J. and Reisz, R. R. 2006. The phylogeny of early eurepties: comparing parsimony
and Bayesian approaches to the investigation of a basal fossil clade. Systematic
Biology 55: 503-511.
Murphy, N. L., Trexler, D., and Thompson, M. 2007. “Leonardo”, a mummified

Brachylophosaurus (Ornithischia: Hadrosauridae) from the Judith River Formation
of Montana. In: K. Carpenter (ed.), Horns and Beaks: Ceratopsian and Ornithopod
Dinosaurs, 117-133. Indiana University Press, Bloomington.
849
Nagao, T. 1936. Nipponosaurus sachalinensis, a new genus and species of Trachodont
dinosaur from Japanese Saghalien. Journal of Faculty Science of Hokkaido Imperial
University 4: 185-220.
Nessov, L. A. 1995. Dinosaurs of northern Eurasia: new data about assemblages, ecology
and paleobiogeography. 156 pp. St. Petersburg State University, St. Petersburg.
Nessov, L. A. and Khissarova, G. D. 1988. New data on vertebrates from the Late
Cretaceous of Shakh-Shakh and Baybolat (north-eastern Aral Sea region). In:
Gavrilov, E. I., Kozhamkulova, B. S., Tleuberdina, P. A., Savinov, P. F., Kochnov,
V. G. (eds), Materials on the History of Fauna and FLora of Kazahstan, 5-14.
Zoological Institute, Almaty.
Nopcsa, F. 1900. Dinosaurierreste aus Siebenbürgen. I. Schädel von Limnosaurus

transsylvanicus nov. gen. et spec. Denkschriften der königlichen Akademie der
Wissenschaften, Wien 68: 555-591.
Nopcsa, F. 1903. Telmatosaurus, new name for the dinosaur Limnosaurus. Geological
Magazine 10: 94-95.
Nopcsa, F. 1928. The genera of reptiles. Palaeobiologica 1: 163-188.
Norell, M. A. and Novacek, M. J. 1992. Te fossil record and evolution: comparing cladistic
and paleontologic evidence for vertebrate history. Science 255: 1690-1693.
Norell, M. A. and Makovicky, P. J. 2004. Dromaeosauridae. In: D.B. Weishampel, P.

Dodson and H. Osmólska (eds.), The Dinosauria, second edition, 196-209.
University of California Press, Berkeley.
Norman, D. B. 1984. On the cranial morphology and evolution of ornithopod dinosaurs.

Symposium of the Zoological Society of London 52: 521-547.
Norman, D. B. 2002. On Asian ornithopods (Dinosauria: Ornithischia). 4. Probactrosaurus

Rozhdestvensky, 1966. Zoological Journal of the Linnean Society 136: 113-144.
Norman, D. B. 2004. Basal Iguanodontia. In: D.B. Weishampel, P. Dodson and H.

Osmólska (eds.), The Dinosauria, second edition, 413-437. University of California
Press, Berkeley.
Norman, D. B., and Weishampel, D.B. 1985. Ornithopod feeding mechanisms: their
bearing on the evolution of herbivory. The American Naturalist 126: 151-164.
Norman, D. B., and Sues, H.-D. 2000. Ornithopods from Kazakhstan, Mongolia and
Siberia. In: Benton, M. J., Shishkin, M. A., Unwin, D. M. and Kurochkin, E. N.
(eds.), The Age of Dinosaurs in Russia and Mongolia, 462-479. Cambridge
University Press, Cambridge.
850
Novas, F. 1997. South American dinosaurs. In: P. J. Currie and K. Padian (eds.),
Encyclopedia of Dinosaurs, 678-689. Academic Press, San Diego.
Nylander, J. A. A., Ronquist, F., Huelsenbeck, J., and Nieves-Aldrey, J. L. 2004. Bayesian
phylogenetic analysis of combined data. Systematic Biology 53: 47-67.
Osborn, H. F. 1912. Integument of the iguanodont dinosaur Trachodon. Memoirs of the

American Museum of Natural History. New Series 1: 35-54.
Ostrom, J. H. 1961. Cranial morphology of the hadrosaurian dinosaurs of North America.

Bulletin of the American Museum of Natural History 122: 33-186.
Ostrom, J. H. 1962. The cranial crests of hadrosaurian dinosaurs. Postilla 62: 1-29.
Owen, R. 1842. Report on British fossil reptiles. Part II. Report of the eleventh meeting of
the British Association for the Advancement of Science, July 1841: 66–204.
Parks, W. A. 1920. The osteology of the trachodont dinosaur Kritosaurus incurvimanus.

University of Toronto Studies, Geological Series 11: 1-74.
Parks, W. A. 1922. Parasaurolophus walkeri, a new genus and species of crested

trachodont dinosaur. University of Toronto Studies, Geological Series, 13: 1-32.
Parks, W. A. 1923. Corythosaurus intermedius, a new species of trachodont dinosaur.

University of Toronto Studies, Geological Series 15: 5-57.
Parks, W. A. 1924. Dyoplosaurus acutosqameus, a new genus and species of armored

dinosaur; and notes on a skeleton of Prosaurolophus maximus. University of
Toronto Studies, Geological Series 18: 1-35.
Patterson, C. 1982. Morphological characters and homology. In: K.A. Joysey and A.E.
Friday (eds.), Problems of phylogenetic reconstruction, 21-74. Academic Press,
London and New York.
Paul, G. S. 2008. A revised taxonomy of the iguanodont dinosaur genera and species.
Cretaceous Research 29: 192-216.
Persoon, E., and Fu, K. S. 1986. Shape discrimination using Fourier descriptors. Institute of
Electrical and Electronic Engineers, Transactions on Pattern Analysis and Machine
Intelligence 8: 388-397
Pimentel, R.A., and Riggins, R. 1987. The nature of cladistic data. Cladistics 3: 201-209.
Pindell, J. L., and Kennan, L. 2002. Plate model for the Caribbean:
http://www.ig.utexas.edu/CaribPlate/forum/pindell/pindell_kennan.htm.
851
Pleijel, F. 1995. On character coding for phylogenetic reconstruction. Cladistics 11: 309-
315.
Pol, D. and Norell, M. A. 2001. Comments on the Manhattan Stratigraphic Measure.

Cladistics 17: 285-289.
Powell, J. E. 1987. [Finding of a hadrosaurid dinosaur (Ornithischia, Ornithopoda) in the

Allen Formation (Upper Cretaceous) of Salitral Moreno, Rio Negro Province,
Argentina]. Décimo Congreso Geologico Argentino, San Miguel de Tucuman, Actas
3: 149-152. [In Spanish]
Pough, F. H., Andrews, R. M., Cadle, J. E., Crump, M. L., Savitzky, A. H., and Wells, K.
D. 2004. Herpetology, third edition. 736 pp. Prentice-Hall, London.
Prieto-Márquez, A. 2005. New information on the cranium of Brachylophosaurus

canadensis (Dinosauria: Hadrosauridae) with a revision of its phylogenetic position.
Prieto-Márquez, A., Gaete, R., Galobart, A., and Ardevol, L. 2000. A Richardoestesia-like
theropod tooth from the Late Cretaceous foredeep, south-central Pyrenees, Spain.
Eclogae Geologicae Helvetiae 93: 497-501.
Prieto-Márquez, A., Weishampel, D. B., and Horner, J. R. 2006a. The hadrosaurid dinosaur
Hadrosaurus foulkii from the Campanian of the East coast of North America, with a
review of the genus. Acta Palaeontologica Polonica 51: 77-98.
Prieto-Márquez, A., Gaete, R., Rivas, G., Galobart, A., and Boada, M. 2006b.
Hadrosauroid dinosaurs from Western Europe: Pararhabdodon isonensis revisited
and Koutalisaurus kohlerorum n. gen. et sp. Journal of Vertebrate Paleontology 26:
929-943.
Prieto-Márquez, A., Gignac, P. M., and Joshi, S. 2007. Neontological evaluation of pelvic
skeletal attributes purported to reflect sex in extinct non-avian archosaurs. Journal
of Vertebrate Paleontology 27: 603-609.
Riabinin, A. N. 1939. [The Upper Cretaceous vertebrate fauna of south Kazakhstan. I. part
1. Ornithischia.] Centralnyj Naucno-issledovatelnyj geologiceskij Institut, Trudy
118: 1-40. [In Russian]
Ronquist, F. 1996. DIVA version 1.1. Computer program and manual available by
anonymous FTP from Uppsala University (ftp.uu.se or ftp.systbot.uu.se).
Ronquist, F. 1997. Dispersal-vicariance analysis: a new approach to the quantification of

historical biogeography. Systematic Biology 46: 195-203.
852
Rozhdestvensky, A. K. 1952. The unveiling of an Iguanodon in Mongolia. Comptes Rendus
of the Academy of Sciences of Moscow 84: 1243-1246.
Rozhdestvensky, A. K. 1957. Duck-billed dinosaur—Saurolophus from Upper Cretaceous

of Mongolia. Vertebrata Palaasiatica 1: 129-148
Rozhdestvensky, A. K. 1966. New iguanodonts from Central Asia: phylogenetic and

taxonomic relationships between late Iguanodontidae and early Hadrosauridae.
Paleontological Journal 1966: 103-116.
Rozhdestvensky, A. K. 1967. New iguanodonts from central Asia. International Geology

Review 9: 556-566.
Rozhdestvensky, A. K. 1968. [Hadrosaurs of Kazakhstan]. In: L. P. Tatarinov (ed.), [Upper

Paleozoic and Mesozoic Amphibians and Reptiles], 97-141. Akademia Nauk
S.S.S.R. Moscow. [In Russian]
Rozhdestvensky, A. K. 1977. The study of dinosaurs in Asia. Journal of the

Palaeontological Society of India 20: 102-119.
Ruschenberger, W. S. W., and Tryon, G. W. 1879. Guide to the Museum of the Academy of
Natural Sciences of Philadelphia. 128 pp. Academy of Natural Sciences,
Philadelphia.
Russell, D. A. 1993. The role of Central Asia in dinosaurian biogeography. Canadian

Journal of Earth Science 30: 2002-2012.
Salgado, L., Coria, R. A., Magalhaes Ribeiro, C. M., Garrido, A., Rogers, R., Simón, M. E.,
Arcucci, A. B., Curry-Rogers, K., Carabajal, A. P., Apesteguia, S., Fernandez, M.,
Garcia, R. A., and Televi, M. 2007. Upper Cretaceous dinosaur nesting sites of Río
Negro (Salitral Ojo de Agua and Salinas de Trapalcó-Salitral de Santa Rosa),
northern Patagonia, Argentina. Cretaceous Research 28: 392-404.
Salinas, G. C., Juárez Valieri, R. D., and Fiorelli, L. E. 2005. [The ornithischian dinosaurs of
South America and their paleobiogeographic importance.] Boletin de Resumos do II
Congresso Latino-Americano de Paleontología de Vertebrados, Río de Janeiro,
Brasil: 240 [Spanish].
Sanmartín, I., and Ronquist, F. 2004. Southern Hemisphere biogeography inferred by

event-based models: plant versus animal patterns. Systematic Biology 53: 216-243.
Seeley, H. G. 1888. On the classification of the fossil animals commonly named

Dinosauria. Proceedings of the Royal Society of London 43: 165–171.
Sereno, P. C. 1986. Phylogeny of the bird-hipped dinosaurs (Order Ornithischia). National

Geographic Research 2: 234-256.
853
Sereno, P. C. 1997. The origin and evolution of dinosaurs. Annual Review of Earth and
Planetary Sciences 25: 435-489.
Sereno, P. C. 1999a. The evolution of dinosaurs. Science 284: 2137-2147.
Sereno, P. C. 1999b. Dinosaurian biogeography: vicariance, dispersal, and regional

extinction. In: Y. Tomida, T. Rich, and R.-V. Rich (eds.). Proceedings of the
Second Gondwanan Dinosaur Symposium, National Science Museum Monographs
15: 249-257. Tokyo.
Sereno, P. C., Wilson, J. A., and Conrad, J. L. 2004. New dinosaurs link southern
landmasses in the Mid-Cretaceous. Proceedings of the Royal Society of London,
Series B 271: 1325-1330.
Seitz, V., Garcia, S. O., and Liston, A. 2000. Alternative coding strategies and the
inapplicable data coding problem. Taxon 49: 47-54.
Siddall, M. E. 1998. Stratigraphic fit to phylogenies: a proposed solution. Cladistics 14:

201-208.
Smith, A. B. and Littlewood, D. T. J. 1994. Paleontological data and molecular

phylogenetic analysis. Paleobiology 20: 259-273.
Smith, A. G., Smith, D. G., and Funnell, B. M. 1994. Atlas of Mesozoic and Cenozoic
Coastlines. 109 pp. Cambridge University Press, Cambridge.
Snively, E., Russell, A. P. and Powell, G. L. 2004. Evolutionary morphology of the

coelurosaurian arctometatarsus: descriptive, morphometric and phylogenetic
approaches. Zoological Journal of the Linnean Society 142: 525-553.
Srivastava, A., Joshi, S. H., Kaziska, D., and Wilson, D. C. 2005. Planar shape analysis and
its applications in image-based inferences. In: N. Paragios, Y. Chen, and O.
Faugeras (eds.), Matematical Models of Computer Vision: The Handbook, 189-203.
Springer-Verlag.
Sternberg, C. M. 1926. A new species of Thespesius from the Lance Formation of

Saskatchewan. Bulletin of the Department of Mines, Geological Survey of Canada
44: 73-84.
Sternberg, C. M. 1935. Hooded hadrosaurs of the Belly River Series of the Upper
Cretaceous: a comparison, with descriptions of new species. Bulletin of the National
Museum of Canada 52: 1-37.
Sternberg, C. M. 1953. A new hadrosaur from the Oldman Formation of Alberta:

discussion of nomenclature. National Museum of Canada, Bulletin 128: 275-286.
854
Sternberg, C. M. 1954. Classification of North American duck-billed dinosaurs. Journal of
Paleontology 28: 382-383.
Strong, E. E., and Lipscomb, D. 1999. Character coding and inapplicable data. Cladistics
15: 363-371.
Suzuki, D., Weishampel, D. B., and Minoura, N. 2004. Nipponosaurus sachalinensis

(Dinosauria: Ornithopoda): anatomy and systematic position within Hadrosauridae.
Swofford, D. L. 2002. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other

Methods). Version 4.0b10. Sinauer Associates, Sunderland, Massachusetts.
Swofford, D. L., Olsen, G. J., Waddell. P. J., and Hillis, D. M. 1996. Phylogenetic
inference. In: D.M. Hillis, C. Moritz, and B. Mable (eds.), Molecular Systematics,
second edition, 407-514. Sinauer Associates, Sunderland, Massachusetts.
Sullivan, R. M. 1999. Nodocephalosaurus kirtlandiensis gen. et sp. nov., a new

ankylosaurid dinosaur (Ornithischia: Ankylosauria) from the Upper Cretaceous
(Late Campanian) Kirtland Formation of New Mexico. Journal of Vertebrate
Tang, F., Luo, Z-X-. Zhou, Z.-H., You, H.-L-. Georgi, A., Tang, Z.-L., and Wang, X.-Z.
2001. Biostratiography and palaeonvironment of the dinosaur-bearing sediments in
Lower Cretaceous of Mazongshan area, Gansu Province, China. Cretaceous
Research 22: 115-129.
Taquet, P. 1976. [Geology and paleontology of the Gadoufaoua fossil locality. Aptian of
Niger]. Cahiers de Paleontologie. 191 pp. Centre National de la Recherche
Scientifique, Paris. [In French]
Therrien, F. 2005. Palaeoenvironments of the latest Cretaceous (Maastrichtian) dinosaurs of

Romania: insights from fluvial deposits and paleosols of the Transylvanian and
Hateg basins. Palaeogeography, Palaeoclimatology, Palaeoecology 218: 15-56.
Turner, A. H. 2004. Crocodyliform biogeography during the Cretaceous: evidence of

Gondwanan vicariance from biogeographical analysis. Proceedings of the Royal
Society of London, Series B 271: 2003-2009.
Umazano, A. M., Bellosi, E. S., Visconti, G., and Melchor, R. N. 2008. Mechanisms of
aggradation in fluvial systems influenced by explosive volcanism: an example from
the Upper Cretaceous Bajo Barreal Formation, San Jorge Basin, Argentina.
Sedimentary Geology 203: 213-228.
855
Upchurch, P., Hunn, C. A., and Norman, D. B. 2002. An analysis of dinosaurian
biogeography: evidence for the existence of vicariance and dispersal patterns caused
by geological events. Proceedings of the Royal Society of London, Series B 269:
613-621.
Van Itterbeek, J. V., Horne, D. J., Bultynck, P, and Vandenberghe, N. 2005. Stratigraphy
and palaeonvironment of the dinosaur-bearing Upper Cretaceous Iren Dabasu
Formation, Inner Mongolia, People’s Republic of China. Cretaceous Research 26:
699-725.
van Veller, M. G. P., Kornet, D. J., and Zandee, M. 2000. Methods in vicariance
biogeography: assessment of the implementations of assumptions 0, 1, and 2.
Cladistics 16: 319-345.
Varricchio, D. J., and Horner, J. R. 1993. Hadrosaurid and lambeosaurid bone beds from
the Upper Cretaceous Two Medicine Foramtion on Montana; taphonomic and
biologic implications. Canadian Journal of Earth Sciences 30: 997-1006.
Wagner, J. R. 2001. The hadrosaurian dinosaurs (Ornithischia: Hadrosauria) of Big Bend

National Park, Brewster County, Texas, with implications for Late Cretaceous
Paleozoogeography. 417 pp. Unpubished Master’s Thesis, Texas Tech University,
Austin.
Warren, L. 1998. Joseph Leidy: The Last Man Who Knew Everything. 352 pp. Yale
University Press, New Haven.
Weil, A. 2002. Late Cretaceous and early Paleocene mammalian faunal exchange between
Asia and North America. The Geological Society of America Annual Meeting,
http://gsa.confex.com/gsa/2002AM/finalprogram/abstract_41476.htm
Weishampel, D. B. 1981. The nasal cavity of lambeosaurine hadrosaurids (Reptilia:

Ornithischia): comparative anatomy and homologies. Journal of Paleontology 55:
1046-1057.
Weishampel, D. B. 1984. Evolution of jaw mechanisms in ornithopod dinosaurs. Advances

in Anatomy, Embriology and Cell Biology 87: 1-110.
Weishampel, D. B. 1990. Dinosaur distributions. In: D.B. Weishampel, P. Dodson, and H.

Osmólska (eds.), The Dinosauria, 534-561. University of California Press,
Berkeley.
Weishampel, D. B., and Weishampel, J. B. 1983. Annotated localities of ornithopod

dinosaurs; implications to paleobiogeography. The Mosasaur 1: 43-88.
856
Weishampel, D. B. and Horner, J. R. 1986. The hadrosaurid dinosaurs of the Iren Dabasu
Formation (People’s Republic of China, Late Cretaceous). Journal of Vertebrate
Weishampel, D. B. and Horner, J. R. 1990. Hadrosauridae. In: D. B. Weishampel, P.

Dodson, and H. Osmólska (eds.), The Dinosauria, 534-561. University of California
Press, Berkeley.
Weishampel, D. B. and L. Young. 1996. Dinosaurs of the East Coast. 275 pp. Johns
Hopkins University Press, Baltimore.
Weishampel, D. B. and N. White. 2003. The Dinosaur Papers: 1676-1906. 524 pp.
Smithsonian Institution Press, Washington, D. C.
Weishampel, D. B., Grigorescu, D., and Norman, D.B. 1991. The dinosaurs of
Transylvania. National Geographic Research and Exploration 7: 196-215.
Weishampel, D. B., Norman, D.B, and Grigorescu, D. 1993. Telmatosaurus

transsylvanicus from the Late Cretaceous of Romania: the most basal hadrosaurid.
Palaeontology 36: 361-385.
Weishampel, D. B., Dodson, P., and Osmolska, H. 2004. Introduction. In: D.B.
Weishampel, P. Dodson, and H. Osmólska (eds.), The Dinosauria, second edition,
Wiens, J. J. 1998. Does adding characters with missing data increase or decrease
phylogenetic accuracy? Systematic Biology 47: 625-640.
Wiens, J. J. 2003. Incomplete taxa, incomplete characters, and phylogenetic accuracy: is

there a missing data problem? Journal of Vertebrate Paleontology 23: 297-310.
Wiley, E. O. 1981. Phylogenetics: The Theory and Practice of Phylogenetic Systematics.

439 pp. John Wiley & Sons, New York.
Wiens, J. J. 2003. Incomplete taxa, incomplete characters, and phylogenetic accuracy: is

there a missing data problem? Journal of Vertebrate Paleontology 23: 297-310.
Wiens, J. J., Bonett, R. M. and Chippindale, P. T. 2005. Ontogeny discombobulates

phylogeny: paedomorphosis and higher-level salamander reationships. Systematic
Biology 54: 91-110.
Wilkinson, M. 1995. A comparison of two methods of character construction. Cladistics

11: 297-308.
Wilkinson, M. 2003. Missing trees and multiple trees: instability, relationships, and support
in parsimony analysis. Journal of Vertebrate Paleontology 23: 311-323.
857
Williamson, T. E. 2000. Review of Hadrosauridae (Dinosauria: Ornithischia) from the San
Juan Basin. New Mexico Museum of Natural History and Science Bulletin 17: 191-
213.
Wilson, J. A. 2005. Redescription of the Mongolian sauropod Nemegtosaurus mongoliensis

Nowinski (Dinosauria: Saurischia) and comments on late Cretaceous sauropod
diversity. Journal of Systematic Palaeontology 3: 283-318.
Wiman, C. 1929. Die Kreide. Dinosaurier aus Shantung. Palaeontologia Sinica, Series C 6:
1-67.
Wiman, C. 1931. Parasaurolophus tubicen (n. sp.) aus der Kreide in New Mexico. Nova
Acta Regiae Societatis Scientiarum Upsaliensis 7:1-11.
Windhausen, A. 1918. The problem of the Cretaceous-Tertiary boundary in South America

and the stratigraphic position of the San Jorge Formation in Patagonia. American
Journal of Science 45: 2-53.
Young, C. C. 1958. The dinosaurian remains of Laiyang, Shantung. Palaeontologica

Sinica, New Series C 16: 53-138.
Young, F. W. 1985. Multidimentional scaling. In: S. Kotz and N. L. Johnson (eds.),

Encyclopedia of Statistical Sciences, volume 5, 649-658. John Wiley and Sons.
Zelditch, M. L., Swiderski, D. L., and Fink, W. L. 2000. Discovery of phylogenetic

characters in morphological data. In : J. J. Wiens (ed.), Phylogenetic Analysis of
Morphological Data, 37-83. Smithsonian Institution Press, Washington.
Zharkov, M. A., Murdmaa, I. O., and Filatova, N. I. 1998. Paleogeography of the

Coniacian-Maastrichtian ages of the Late Cretaceous. Stratigraphic and Geological
Correlation 6: 209-221.
Zhang, Y., Dong, S., and Shi, W. Cretaceous deformation history of the middle Tan-Lu
fault zone in Shandong Province, eastern China. Tectonophysics 363: 243-258.
Zhang, J., Zhang, H.-F., Yiang, J.-F., Tang, Y.-J., and Niu, L.-F. 2008. Contribution of
subducted Pacific slab to late Cretaceous mafic magmatism in Qingdao region,
China: a petrological record. The Island Arc 17: 231-241.
858
BIOGRAPHICAL SKETCH
Albert Prieto-Marquez was born in 1974 in Sant Vicenç dels Horts, Barcelona, Spain. He
earned a Bachelor’s degree in Geological Sciences from the University of Barcelona in 1996,
and spent two years conducting paleontological research at the Catalan Institute of Paleontology
of Sabadell, Barcelona, Spain. In 2001 he earned a Master’s of Science degree in Earth Sciences
at Montana State University, Bozeman, Montana. In 2008 he earned a Doctor of Phylosophy
degree in Biological Sciences at Florida State University. He has accepted a postdoctoral
fellowship at the American Museum of Natural History, New York City.
859

Prieto-Marquez 2008 - PHD Dissertation

Uploaded by

Document Information

Original Description:

Original Title

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

Prieto-Marquez 2008 - PHD Dissertation

Uploaded by

Copyright:

Available Formats

THE FLORIDA STATE UNIVERSITY

COLLEGE OF ARTS AND SCIENCES

PHYLOGENY AND HISTORICAL BIOGEOGRAPHY OF

A Dissertation submitted to the

A mamá, papá, mi hermana... y la pequeña Noa

List of Tables .................................................................................................................. x

1. HISTORICAL REVIEW OF STUDIES ON HADROSAURID SYSTEMATICS........ 5

First Discoveries ......................................................................................................... 5

2. PHYLOGENY OF HADROSAURID DINOSAURS ................................................ 43

3. HISTORICAL BIOGEOGRAPHY OF HADROSAURID DINOSAURS................ 122

Abstract .................................................................................................................. 122

4. OSTEOLOGY AND SYSTEMATICS OF THE HADROSAURID DINOSAUR

Abstract .................................................................................................................. 153

5. PARARHABDODON ISONENSIS AND TSINTAOSAURUS SPINORHINUS: A NEW

Abstract .................................................................................................................. 195

APPENDIX A. TAXON-CHARACTER STATE MATRIX........................................ 210

APPENDIX B. DENTAL CHARACTERS ................................................................. 255

APPENDIX C. MANDIBULAR CHARACTERS....................................................... 309

APPENDIX D. FACIAL CHARACTERS................................................................... 388

APPENDIX E. NEUROCRANIAL CHARACTERS................................................... 536

APPENDIX F. REGIONAL CRANIAL CHARACTERS ........................................... 576

APPENDIX G. VERTEBRAL AND OTHER AXIAL CHARACTERS...................... 623

APPENDIX H. PECTORAL AND FORELIMB CHARACTERS............................... 638

APPENDIX I. PELVIC CHARACTERS..................................................................... 697

APPENDIX J. PELVIC CHARACTERS .................................................................... 797

APPENDIX K. SYNAPOMORPHIES AND CLADE DEFINITIONS........................ 813

REFERENCES ........................................................................................................... 834

BIOGRAPHICAL SKETCH....................................................................................... 858

Figure 1.4. Phylogenetic hypotheses of the evolutionary relationships among hadrosaurid

Figure 1.5. Phylogenetic hypotheses of the evolutionary relationships among hadrosaurid

Figure 1.7. Phylogenetic hypotheses of hadrosaurid relationships by John R. Horner. A,

Figure 1.8. Phylogenetic hypotheses of hadrosuaroid and hadrosaurid relationships by A,

Figure 1.9. Phylogenetic hypotheses of hadrosuaroid and hadrosaurid relationships by

Figure 1.11. Phylogenetic hypotheses of hadrosaur relationships by A, Norman (2002) and B,

Figure 1.13. Phylogenetic hypotheses of hadrosaurid relationships by A, Prieto-Marquez

Figure 2.2. Signs of postdepositional dorsoventral compression (arrows) in a specimen of

Figure 2.18. Ancestral state reconstructions of ischial characters. A, cranioventral orientation

Figure 3.3. Paleogeographic reconstruction of continental coastlines during the Coniacian.

Figure 3.4. Paleogeographic reconstruction of continental coastlines during the Santonian.

Figure 3.5. Paleogeographic reconstruction of continental coastlines during the Campanian.

Figure 3.6. Paleogeographic reconstruction of continental coastlines during the

Figure 4.2. Mandibular elements of Secernosaurus koerneri. A and B, predentary, MACN-

Figure 4.7. Secernosaurus koerneri, fragment of basioccipital and basisphenoid, holotype

Figure 4.15. Pelvic elements of a juvenile specimen of Secernosaurus koerneri, holotype

Figure 4.16. Pelvic elements of Secernosaurus koerneri. A, left ilium, MACN-RN 2, in

Figure B.26. Character 19. Maxillary occlusal planes of various iguanodontoidean

Figure C.4. Characters 27 and 28. Predentary denticles of several iguanodontoidean

Figure C.8. Character 33. Lateral views of two hadrosaurid predentaries. A,

Figure C.8. Character 33. Lateral views of two hadrosaurid predentaries. A,

Figure C.9. Character 34. Dorsolingual views of iguanodontoidean predentaries; arrows

Figure D.53. Character 104. State 0: presence of maxilla-lacrimal contact. A, Gryposaurus

Figure D.54. Character 104. State 1: absence of maxilla-lacrimal contact. A,

Figure D.83. Character 130. Development of a prefrontal crest (arrows). A, Edmontosaurus

Figure D.84. Character 130. Presence of a well-developed prefrontal crest (arrows). A,

Figure D.92. Character 136.Presence of a dorsal promontorium (arrows) on the rostral

Figure E.3. Character 148. Presence of a frontal platform. A, Amurosaurus riabinini

Figure E.4. Character 149. Presence of a frontal platform. A, Charonosaurus jiayinensis

Figure E.16. Character 161. Length of the basioccipital constriction (arrows). A,

Figure F.15. Character 191. Composition of the supracranial crest. A, Brachylophosaurus

Figure F.17. Character 193. General morphology of supracranial crests (rectangles).

Figure F.18. Character 193. General morphology of supracranial crests (rectangles).

Figure F.19. Character 193. General morphology of supracranial crests (rectangles).