Palo 2010 p10-126r

WHAT CONTROLS THE TAPHONOMY OF EXCEPTIONALLY PRESERVED
TAXA—ENVIRONMENT OR BIOLOGY? A CASE STUDY USING FROGS

FROM THE MIOCENE LIBROS KONSERVAT-LAGERSTÄTTE (TERUEL,
SPAIN)
Authors: MARIA E. MCNAMARA, PATRICK J. ORR, LUIS ALCALÁ, PERE ANADÓN, and
ENRIQUE PEÑALVER
Source: Palaios, 27(2) : 63-77
Published By: Society for Sedimentary Geology
URL: https://doi.org/10.2110/palo.2010.p10-126r
BioOne Complete (complete.BioOne.org) is a full-text database of 200 subscribed and open-access

titles in the biological, ecological, and environmental sciences published by nonprofit societies,
associations, museums, institutions, and presses.
Your use of this PDF, the BioOne Complete website, and all posted and associated content indicates
your acceptance of BioOne’s Terms of Use, available at www.bioone.org/terms-of-use.
Usage of BioOne Complete content is strictly limited to personal, educational, and non-commercial use.
Commercial inquiries or rights and permissions requests should be directed to the individual publisher
as copyright holder.
BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit
publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical
research.
Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM)
PALAIOS, 2012, v. 27, p. 63–77
Research Article
DOI: 10.2110/palo.2010.p10-126r
WHAT CONTROLS THE TAPHONOMY OF EXCEPTIONALLY PRESERVED TAXA—ENVIRONMENT

OR BIOLOGY? A CASE STUDY USING FROGS FROM THE MIOCENE LIBROS KONSERVAT-
LAGERSTÄTTE (TERUEL, SPAIN)
MARIA E. MCNAMARA,1,2* PATRICK J. ORR,2 LUIS ALCALÁ,3 PERE ANADÓN,4 and ENRIQUE PEÑALVER 5
1Department of Geology & Geophysics, Yale University, New Haven, Connecticut 06520-8109, USA, maria.mcnamara@yale.edu; 2UCD School of Geological Sciences,
University College Dublin, Belfield, Dublin 4, Ireland, patrick.orr@ucd.ie; 3Fundación Conjunto Paleontológico de Teruel-Dinópolis, Avenida Sagunto s/n, 44002 Teruel,
Aragón, Spain, alcala@dinopolis.com; 4Consejo Superior de Investigaciones Cientı́ficas, Institut de Ciències de la Terra ‘‘Jaume Almera,’’ Lluı́s Solé i Sabarı́s s/n 08028,
Barcelona, Spain, panadon@ija.csic.es; 5Museo Geominero, Instituto Geológico y Minero de España, Calle Rı́os Rosas, 23, E-28003, Madrid, Spain, epenalver@uv.es
ABSTRACT depositional context has been explored in detail in many studies of

skeletal taphonomy (e.g., Brett et al., 2006, 2009; Stigall et al., 2008;
The exceptional preservation of organisms is potentially influenced by
Foreman et al., 2011) but, surprisingly, is rarely combined with analysis
various factors that reflect either the environmental context or aspects of
of the fidelity of soft-tissue preservation in exceptionally preserved taxa.
the organisms’ biology. There has been no systematic investigation of the Previous studies of the relationship between sedimentological context
relative impact of such factors upon the fidelity of preservation of an and the taphonomy of exceptionally preserved taxa have identified
exceptionally preserved taxon. In this study, we present an integrated either very broad taphonomic patterns related to basin-scale facies
approach to taphonomic analysis of exceptionally preserved taxa using changes (e.g., Barton and Wilson, 2005; Allison et al., 2008) or fine-
primarily quantitative data to analyze the taphonomy of exceptionally scale sedimentary controls upon a single taphonomic feature (e.g.,
preserved frogs (Rana (Pelophylax) pueyoi) from the lacustrine-hosted, Wilby et al., 1996a; Skinner, 2005; Casey et al., 2007).
upper Miocene, Libros Konservat-Lagerstätte (Teruel, northeastern Taphonomic variation between specimens of a particular taxon, when
Spain). The frogs occur within several different laminated mudstone investigated using quantitative statistical techniques, has the potential to
facies and vary in their size, degree of completeness, degree of reveal the relative importance of different taphonomic controls but has
articulation, limb positions, and the extent, type, and fidelity of preserved not been exploited fully. Previous quantitative taphonomic studies have
soft tissues. For each specimen, we coded its physical and soft-tissue focused on patterns of skeletal preservation (e.g., Kemp and Unwin,
taphonomy using twelve indices, and identified its lithological context. 1997; Cambra-Moo and Buscalioni, 2003; Casey et al., 2007; McNamara
Systematic statistical analysis reveals no correlation between specimen et al., 2011) but have not assessed soft-tissue preservational features. We,
size, any taphonomic index, and lithological context: variations in the therefore, identify two gaps in our knowledge of the taphonomy of
taphonomy of the frogs are independent of centimeter-scale variations in exceptionally preserved fossils. Firstly, there has been no investigation of
lithology and, by inference, short-term fluctuations in environmental how the taphonomy of an exceptionally preserved taxon—considered
conditions. The consistently high fidelity of skeletal preservation and the as a whole, i.e., incorporating both skeletal- and soft-tissue data—
preservation of soft tissues, therefore, ultimately reflect the general nature is influenced by short-term (i.e., annual to decadal) variations in
of the sedimentary environment, i.e., burial within laminated organic- environmental conditions. The latter clearly occurred during the
rich muds below anoxic, monimolimnetic waters of a deep, stratified formation of many exceptional biotas as indicated by, for example,
meromictic lake. The statistical analysis did, however, identify significant centimeter-scale lithological changes in finely laminated sediments.
correlations between various taphonomic indices. These indicate that the Secondly, the relative importance of various environmental and
observed variation in the fidelity of frog preservation can be attributed, in biological factors upon the taphonomy of an exceptionally preserved
large part, to a suite of factors related to the biology of the frogs; the most taxon must be assessed in order to fully understand its preservation. Any
important of these is the original biochemistry of specific tissues, attempt to address these issues requires the integration of data treating
especially the skin, and the postmortem fate of the skin. the skeletal- and soft-tissue taphonomy and the precise depositional
context of individual specimens.
INTRODUCTION Herein we present such an integrated approach to taphonomic
analyses of exceptionally preserved taxa using the example of adult
Exceptionally preserved fossils provide critical evidence of the history frogs [Rana (Pelophylax) pueyoi (Ranidae)] from the late Miocene
of life on Earth but their paleobiological potential is compromised by Libros Konservat-Lagerstätte (Teruel, northeastern Spain) (Fig. 1).
our limited understanding of the taphonomic processes involved in their The frogs (n 5 73) are an ideal test case as they vary considerably in
preservation. These processes are controlled ultimately by decay their degree of completeness, degree of articulation, limb position,
mechanisms, the biochemistry of specific taxa and tissues, and the and the extent, type, and fidelity of preserved soft tissues (Figs. 2–3;
depositional context of the biota. Many studies have investigated the Supplementary Data 11), and occur within lithologies that vary on a
role of decay in exceptional preservation (Allison, 1986; Briggs and centimeter scale. Previous studies of the frog taphonomy have
Kear, 1993a, 1993b, 1994; Baas et al., 1995; Briggs et al., 1995; Kear considered only the mode of soft tissue preservation (Luque et al.,
et al., 1995; Hof and Briggs, 1997; Sagemann et al., 1999). Similarly, 1996; McNamara et al., 2006; McNamara et al., 2009). In this study we
taxonomic controls on taphonomic pathways—which indicate the investigate the sedimentological context, skeletal taphonomy, and
importance of the biology of an organism—have been documented in a extent of soft tissues of the Libros frogs.
range of Konservat-Lagerstätten (Voigt, 1988; Wilby and Whyte, 1995; Our approach uses fully and semiquantitative data to code for
Briggs et al., 1996; Wilby et al., 1996b; Wilby and Briggs, 1997; skeletal and soft-tissue taphonomic variables, coupled with other,
Gabbott, 1998; Liebig, 1998; McCobb et al., 2003). The significance of qualitative, taphonomic, and lithological data, to test two hypotheses.
* Corresponding author. 1 palaios.ku.edu
Copyright G 2012, SEPM (Society for Sedimentary Geology) 0883-1351/12/0027-0063/$3.00

64 MCNAMARA ET AL. PALAIOS
evaporative concentration of the lake probably occurred during brief

periods of hydrological closure (Anadón et al., 1992). The shelly fauna
and microflora indicate an alkaline, fresh to slightly saline, oligotrophic
epilimnion (Margalef, 1947; Gregor and Günther, 1985; Gil, 1986;
Anadón et al., 1992; Ortı́ et al., 2003). The monimolimnion was anoxic
and sulfidic, and there was intense bacterial sulfate reduction in
profundal sediments (Anadón et al., 1992; de las Heras et al., 2003; Ortı́
et al., 2003). The absence of such large aquatic organisms as fish
suggests the lake had few, if any, fluvial inlets during deposition of the
Libros Gypsum Unit.
MATERIAL AND METHODS
Sedimentology
Sections of the bituminous-calcareous subunit were logged in the

field at a scale of 1:10. Only rare isolated anuran skeletal elements were
recovered from nonlaminated profundal deposits in the field (Supple-
mentary Data 21). Selected lithological samples and curated slabs of
laminated mudstones were sectioned for petrological analysis (Fig. 4).
Different lamina types—micritic, organic-rich and silty—were identi-
fied; percentages quoted in the text for components identified in thin
section are based on 250–300 point counts.
Fossil Material
FIGURE 1—Location of the Barrio de las Minas near Libros, Teruel with, insets, This study is based on 73 specimens from the following institutions
position of Teruel province within Spain and position of Spain within Europe (Supplementary Data 11): Forschungsinstitut und Naturmuseum
(modified from McNamara et al., 2009). Senckenberg, Frankfurt (FNS), Museo del Colegio La Salle, Teruel,
Spain (MCS), Museu de Geologia de Barcelona, Barcelona (MGB),
The first is that short-term variations in environmental conditions, as Institut Català de Paleontologia Miquel Crusafont, Sabadell, Spain
reflected in facies changes on a centimeter scale, impacted on the (IPS), Museu del Seminari, Barcelona (MSB), Museo Nacional de
physical taphonomy and soft tissue diagenesis of the fossils. The second Ciencias Naturales, Madrid (MNCN), and the Natural History
hypothesis is that some, or all, of the taphonomic features show a Museum, London (NHM). Each fossiliferous slab comprises laminated
statistically significant correlation with each other, and thus that their mudstones and bears a single specimen; the way up and precise field
occurrence is in some way related. The result is the first systematic context of each are unknown, as all specimens were recovered from
assessment of the relative importance of different environmental and the sulfur- and oil shale mines of the Libros Gypsum Unit in the early
biological factors on the fidelity of preservation of an exceptionally 20th century. Some or all of the fossiliferous horizons are no longer
preserved taxon. accessible; whether the curated specimens represent an unbiased sample
of the adult anuran assemblage is, therefore, impossible to test in the
GEOLOGICAL CONTEXT field. Certain taphonomic and sedimentological indices could not be
assessed for each specimen, for example, if a specimen was incomplete,
The Libros lacustrine sequence was deposited in the Teruel Basin in truncated by the edge of the slab, or obscured in part by sediment. The
northeastern Spain during the early Miocene-late Pliocene (Ortı́ et al., number of specimens assessed for each index is, therefore, indicated in
2003) and overlies Upper Triassic gypsum and Cretaceous carbonates the text in parentheses.
(Anadón et al., 1992). The sequence comprises up to 500 m of alluvial The position of any feature within a specimen is described using the
terrigenous strata, lacustrine carbonates, and evaporites. The deepest following terms: in plan view, proximal refers to structures in the
water deposit in the sequence is the 120-m-thick Libros Gypsum Unit thorax, abdomen, and in the proximal parts of limbs; distal, to those in
(Vallesian), which crops out in the Barrio de las Minas near Libros the outer half of each limb; and peripheral, to those close to the edges of
village. The unit is divided into, in ascending order, bituminous- the body and limbs.
calcareous, gypsiferous, and gypsum-carbonate subunits (Ortı́ et al.,
2003, 2010); native sulfur deposits occur in each. The 50-m-thick Lithological Context of Specimens
bituminous-calcareous subunit comprises intercalated charophytic
limestones (wackestones and packstones) and laminated mudstones To determine whether two or more fossil-bearing slabs include the
(including oil shales). Alternations of these lithologies indicate fluc- same horizon, the bar code–like pattern formed by the succession of
tuations between oxic (carbonate deposition) and anoxic (laminated laminae on the vertical face of each slab was compared as follows. Digital
mudstone deposition) conditions (Anadon et al., 1992). The exceptional images of the vertical edge of each slab were printed to the same scale and
biota includes amphibians, birds, snakes, insects, arachnids, and leaves the approximate position of the fossil-bearing horizon marked. Two
(Navás, 1922a, 1922b; Olson, 1995; Peñalver, 1996); it is hosted within slabs were excluded from the analysis, as the succession of laminae was
the deep-water laminated mudstones, which comprise several different indistinct. The printed images were laid on a flat surface and the lamina
facies. succession in one image was compared visually with that in all others—
During deposition of the Libros Gypsum Unit, the Libros paleolake similar to the technique used by Trewin (1986). The image was then
was a perennial, meromictic, hydrologically open, bench-type lake rotated by 180u and the process repeated, as the way up of each slab is
(Anadón et al., 1992) up to a maximum of 100 m deep (Ortı́ et al., 2003) unknown (McNamara et al., 2009). Slabs are likely to represent the same
and ,30 km2 in extent (de las Heras et al., 2003). Periods of increased stratigraphic interval if they exhibit an identical succession of laminae.

PALAIOS CONTROLS ON TAPHONOMY 65
FIGURE 2—Rana (Pelophylax) pueyoi from Libros, showing variations in completeness, articulation, limb positions and the extent of preserved soft tissues. A) MNCN 63778.
B) MCS 601b. C) MNCN 63805. D) MNCN 63763. E) MCS 604b. F) FNS A324. G) MSB 26217. H) MNCN 63694. Small white arrows in A, D, F indicate carbonate deposits
in the former position of the brain. Small black arrows in A, C indicate decay halo. Large black arrows in B–D indicate stomach contents. Scale, 50 mm.

FIGURE 3—Scanning electron micrographs (A, D, E) and light micrographs (B, C, F–H) of soft tissue features preserved in the Libros frogs. A) Bacteria in biofilm that
defines, in part, the body outline. B) Well-preserved mid-dermal Eberth-Katschenko (E-K) layer with polygonal texture. C) Poorly preserved, amorphous, E-K layer. b 5 bone;
bb 5 biofilm; s 5 sediment. D) Calcium carbonate crystals in masses defining former positions of brain and nerve cord. E) Granular masses of calcium phosphate in stomach.
F) Shell fragments in stomach. G) Lower dermal collagen fibers defined by discoids of calcium sulfate. H) Bone marrow within rib (5b).
Specimens were assigned to facies based on the lamina succession (a) the percentage of the total number of bones present in an adult Rana
exposed on the vertical edge of each slab. The lamina that contains the (i.e., 159) (percentage completeness), and (b) which skeletal element(s)
frog fossil is impossible to identify precisely: the way up of each are present (e.g., femur, phalanges). Some skeletal elements, for
specimen is unknown and the lamina subjacent to the specimen (as it example, the phalanges, are represented by more than one bone.
appears on the slab) may, therefore, have been deposited after, not Although the number of phalanges present was counted, identification
before, the specimen. To test for the significance of the distribution of of specific phalanges was not always possible. When assessing the
specimens among the various facies, expected values were calculated as presence-absence of the phalanges, they were, therefore, treated
follows: collectively for each fore- and hindlimb. Coding phalanges as absent
na |nb denotes specimens in which at least one phalanx (usually more) is
nabexp ~ absent. Similarly, the presence of individual cranial bones was difficult
nc
to assess as they are typically superimposed; these were also treated
where nabexp is the expected number of specimens in category a with collectively and coded as mentioned earlier. Percentage completeness
feature b, na is the number of specimens in category a, nb is the number for each specimen is plotted in Figure 5, and the percentage of
of total specimens with feature b, and nc is the total number of specimens in which each major skeletal element type is present is
specimens (see Supplementary Data 31 for a worked example). summarized in Figure 6. In order to assess whether the loss of skeletal
Differences between observed and expected values were tested using elements shows any pattern, the presence or absence of each skeletal
the chi-squared statistic. Expected values were weighted to account for element in each specimen was coded in a binary matrix that was seriated
the different relative abundances of each facies (see Supplementary in PAST (PAlaeontological STatistics; Hammer et al., 2001) using the
Data 31 for details). A significance level of p , 0.05 was used for all unconstrained algorithm (Fig. 7A); seriation is a scaling technique that
statistical tests. orders a list of items based on their similarity. The seriation was
assessed using a seriation criterion; the criterion inherent to PAST is
Physical Taphonomy that of Brower and Kyle (1988), which ranges between, and includes,
zero and one (a perfect seriation has a value of one). Qualitative data
The following taphonomic indices were analyzed for each specimen: for the type of skeletal element absent were used to define the following
(1) specimen size—included here as it is a physical attribute, albeit not a completeness categories: (1) complete specimens; (2) specimens lacking
taphonomic variable; (2) orientation with respect to bedding; (3) only phalanges; and (3) specimens lacking elements in addition to the
completeness; (4) articulation; and (5) limb position. phalanges.
Specimen Size.—Snout–urostyle length—the distance between the Articulation.—Articulation can be assessed only for those joints that
anterior tip of the snout and the posterior tip of the urostyle—was are present. Joints were coded as articulated if the relevant bones were
measured on each specimen (n 5 57). This is a standard proxy for the juxtaposed in life position. For each specimen (n 5 57), the degree of
size, but not age, of an anuran (Kumbar and Pancharatna, 2001). In articulation of those skeletal elements present was assessed in two ways:
addition, the ratio of tibiofibula length:femur length was calculated (n (1) the percentage of the total number of joints that are articulated
5 57). In extant frogs, a tibiofibula:femur length ratio .1 indicates a (percentage articulation) and (2) which of the major joints (e.g.,
saltatory mode of life and thus a predominantly terrestrial ecology shoulder, knee) are articulated. The percentage articulation of each
(Shubin and Jenkins, 1995). specimen is plotted in Figure 5. The percentage of specimens in which
Orientation with Respect to Bedding.—Specimens (n 5 64) are the bones at each of the major joints are articulated is summarized in
dorsoventral if the sagittal axis is perpendicular to bedding, oblique if Figure 6. In order to assess whether disarticulation of joints shows any
inclined to bedding, and lateral if parallel to bedding. pattern, articulation data for each specimen were coded in a binary
Completeness.—Bones were coded as present if in, or less than 20 mm matrix and seriated as previously stated (Fig. 7B). Qualitative data for
from, life position; complete specimens may, therefore, be disarticulat- the types of articulated joints present were used to define the following
ed. For each specimen (n 5 57), completeness was assessed in two ways: articulation categories: (1) fully articulated specimens; (2) specimens

FIGURE 5—Plot of percentage completeness and percentage articulation of

each specimen. Vertical and horizontal bars denote standard deviation and
mean, respectively.
angle possible at these joints is unlikely to have been underestimated as

the subject (understandably) preferred highly contracted limb positions.
The range of angles subtended by the bones at the hip and knee joints
was obtained from published data on Rana pipiens (Kargo et al., 2002)
(Fig. 8C).
Extent of Soft Tissues
The mode of soft tissue preservation of the Libros frogs has been
described previously (McNamara et al., 2006, 2009). In brief, the
outline of the soft tissues is defined by a brown, carbonaceous bacterial
biofilm (Fig. 3A) and the phosphatized remains of the mid-dermal
Eberth-Katschenko (E-K) layer (Figs. 3B–C). The biofilm occurs as
two layers: Layer 1 is internal to the E-K layer and represents the
degraded remains of the internal body contents; Layer 3 is external to
the E-K layer and represents the degraded remains of the epidermis and
upper dermis. The E-K layer is coded as either well preserved (solid,
cohesive, and exhibiting a microscopic polygonal texture and phos-
phatized collagen fibers) (Fig. 3B) or poorly preserved (friable and
FIGURE 4—Laminated mudstone facies. A, B) Photographs of Facies A and B, with

the succession of facies indicated by the schematic stratigraphic column in (B). C, D)
Light micrographs of Facies B1 (C) and B2 (D). m 5 micritic lamina; o 5 organic-rich
lamina; s 5 silty lamina.
in which joints distal of the tibiofibula and radioulna only are

disarticulated; (3) specimens in which joints proximal of the tibiofibula
and radioulna, but not abdominal joints (i.e., the urostyle-ischium,
urostyle-sacral vertebral and ilium/sacral vertebral joints) are also
disarticulated; and (4) specimens in which distal and proximal joints
(including abdominal joints) are disarticulated.
Limb Position.—The angle between the relevant bones at each of
the shoulder, elbow, hip, and knee joints was recorded as shown in
Figure 8A and plotted on a rose diagram (Fig. 8B) (n 5 64). Data for
both sides of the body are combined in Figure 8B as the left and right
sides of a specimen could not be determined, i.e., each specimen can
yield up to two values per joint. The range of angles subtended by the
bones at the shoulder and elbow joints was compared to that possible in
vivo in the terrestrial frog Rana temporaria (Fig. 8C) (UCD Animal
Research Ethics Committee protocol P-07-15-McNamara). As these
data were obtained using a live subject, manipulation of the limbs was
FIGURE 6—Outline drawing of a standard Rana skeleton illustrating the major
limited to those orientations in which the subject did not appear very skeletal elements and joints. For each element and joint, the percentage of specimens
distressed. The maximum angle possible at each of the forelimb joints in in which that element is complete, or the joint articulated, is indicated. Individual
vivo may, therefore, have been underestimated slightly. The minimum elements of the cranium are not shown for clarity.

FIGURE 7—Seriated data matrices for completeness (A) and articulation (B).
Specimens in rows, skeletal elements missing (A) and joints disarticulated (B) in
columns. Black squares denote absence of elements in A and disarticulation of bones
at specific joints in B. ac 5 astragalus-calcaneum; c 5 carpals; f 5 femur; fp 5
forelimb phalanges; h 5 humerus; hp 5 hindlimb phalanges; i 5 ilium; r 5 radioulna;
sv 5 sacral vertebra; t 5 tarsals; tf 5 tibiofibula; u 5 urostyle; v 5 vertebra.
amorphous) (Fig. 3C). The E-K layer and Layer 3 occur in the same
areas in each specimen; only the former is considered further and is
hereafter referred to as the skin. Layer 1 of the biofilm is hereafter FIGURE 8—Limb positions. A) Outline drawing of a standard Rana skeleton
referred to as the biofilm. illustrating method for measuring the angle between the bones at the shoulder, elbow,
Problematically, the visible extent of soft tissues in a specimen can be hip, and knee joints in each specimen. The vertebral column was chosen as a proxy for
the long axis of the body. B) Rose diagrams of the range of angles exhibited by the
controlled by the position of the edge, and plane of splitting, of the slab.
bones at each of the shoulder, elbow, hip, and knee joints. C) Range of angles between
In 6 specimens, more than one pes, or the head, is truncated by the edge bones at each of the shoulder, elbow, hip, and knee joints exhibited by living Rana.
of the slab; data from these specimens are, therefore, omitted. In 33
specimens, the plane of splitting occurs predominantly within the
sediment. In such cases, much of the soft tissues is obscured by biofilm or skin was determined for each cell, and the cumulative
sediment (Figs. 2A, C, F, H) or in the counterpart of the specimen; the frequency for each cell calculated. These data were displayed as three-
apparent extent of soft tissues in each may therefore be considerably dimensional plots constructed using Surfer 9 (Figs. 9B–C). The extent
less than the true extent. Data from these specimens are also omitted of the skin and biofilm preserved in each specimen is expressed as a
from this study. In the remaining specimens (n 5 32), the plane of percentage of the theoretical maximum.
splitting occurs wholly (or predominantly) within the vertical thickness For each specimen, the presence or absence of the following five soft
of the soft tissues, usually along the boundary between the biofilm and tissue features was noted where possible (see also McNamara et al.,
the skin (Figs. 2D, G). Soft tissues are of similar extent in both part and 2006, 2009): definition of the brain and nerve cord (by calcium
counterpart of such specimens (where available). The impact of having carbonate (51 specimens; Fig. 3D)), the stomach (by calcium phosphate
only one half of the specimen is, therefore, minimal. (27 specimens; Fig. 3E) and or shell fragments (32 specimens; Fig. 3F)),
The areas occupied by the skin and biofilm differ and were assessed and collagen fibers of the lower dermis (by discoids of calcium sulfate
separately. A theoretical maximum value for the extent of soft tissues (26 specimens; Fig. 3G)), bone marrow (organically preserved (seven
was derived by combining evidence from several specimens with specimens; Fig. 3H)), and a decay halo: a subtle, dark-colored,
extensive soft tissues. This was mapped onto an outline drawing of a carbonaceous stain that extends up to 10 mm from the specimen and
Rana skeleton (Fig. 9A). The areas occupied by the skin and biofilm is most common adjacent to disarticulated joints (13 specimens; small
in each specimen were mapped (by eye) onto this standard. A grid arrows in Figs. 2A, C). This feature has not been described previously.
containing 5504 equally sized square cells was then overlain on the The decay halo appears darker in backscattered electron images than
standard; for each specimen, the presence or absence of each of the the adjacent sediment; this likely reflects a higher abundance of carbon

FIGURE 9—Extent of soft tissues. A) Theoretical maximum extent of soft tissues present in the Libros frogs mapped onto a Rana skeleton. B, C) 3-D surface maps showing the
cumulative frequency of skin (B) and bacterial biofilm (C) (expressed as a percentage of specimens, n 5 32) over the theoretical maximum extent of soft tissues.
in the former (see Orr et al., 2002). The stain is superficially similar to specimens with each feature was determined for each facies; note that
the dark stain associated with some metazoans (e.g., Aysheaia and one specimen may exhibit more than one feature. The Pearson’s
Marella) from the Burgess Shale (Hou and Berström, 1997; Whittington, correlation coefficient was used to test for similarity between the
1971, 1977), and with naraoiids (Arthropoda) from the middle Cambrian number of specimens in each facies with a particular soft tissue feature,
Kaili Biota (Lin, 2006). The stain in these Cambrian organisms often and the total number of specimens in each facies. The coefficient has a
occurs close to the mouth, or the posterior of the abdomen (Whittington, value between 0 and 1; lower values indicate less similarity between the
1971, 1977; Hou and Berström, 1997; Lin, 2006) and is probably dataset for each facies and therefore greater significance in this study.
analogous to the purge fluid (the decomposed lining of the digestive
tract) that seeps from the orifices of the digestive tract in vertebrates a few Interrelationships among Taphonomic Indices
weeks after death (Clark et al., 1997). In the Libros frogs, however, the
decay halo never occurs close to the head or the former position of the The presence of any significant relationships between any two of the
cloaca. Instead, the decay halo is repeatedly associated with disarticu- taphonomic indices was analyzed using the following techniques
lated joints, suggesting it represents decay fluids from degraded internal (Table 1): (1) Analyses involving only fully quantitative indices (i.e.,
body contents, not just the digestive organs, that seeped through tears in percentage data and size measurements) used standard X-Y plots; the
the skin. A similar phenomenon has been described in frog carcasses extent of any correlation was assessed using the R2 value for the best fit
degraded experimentally (Wuttke, 1983). regression line. (2) Except for a single exception (see (3)), analyses
involving one fully quantitative and one qualitative (with mutually
Impact of Lithology on Taphonomy exclusive categories, e.g., the type of skeletal element absent) index used
X-Y plots where the mean, and standard deviation range, of the
The taphonomic indices assessed are listed in Table 1. The quantitative data were plotted for each qualitative category. For each
significance of the presence or absence of a decay halo and bone quantitative variable, equality of the means of the data for each
marrow were not assessed statistically as the sample size for each is category of the qualitative variable was tested using one-way ANOVA
small (n 5 13 and n 5 7, respectively). or the unequal variance Welch’s ANOVA test. (3) For analyses
The significance of any variation between facies type and the various involving limb positions and the completeness and articulation
taphonomic indices was tested statistically (Table 1). The choice of categories, the angle between the bones at the two forelimb joints were
statistical test reflects the nature of the data. (1) For fully quantitative plotted against each other and specimens coded according to the
data (percentages, size measurements, and angles at limb joints), data various categories. Variation in the distribution of data for the various
were plotted for each facies and the mean and standard deviation categories was summarized as the smallest area polygon that enclosed
calculated. The equality of the means of the data for each facies was all relevant specimens. Minimal or no overlap between any two
tested using one-way ANOVA or the unequal variance Welch’s polygons are considered to merit further investigation. (4) Analyses
ANOVA test. (2) For qualitative data, e.g., the categories defined for involving two qualitative indices (one comprising nonexclusive catego-
completeness and articulation, each specimen was assigned to the ries, e.g., the type of soft tissue feature present) used Pearson’s test of
relevant category and facies. The resulting matrix of observed values statistical significance. (5) For presence-absence data for the type of soft
was compared to a matrix of expected values that were calculated as tissue feature present, a Raup-Crick similarity index was calculated
described above (see Lithological Context of Specimens). Differences (using PAST) for each pair of co-occurring features. This nonmetric
between the observed and expected matrices were tested using Fisher’s probabilistic index includes absences (and is, therefore, not biased
exact test; the x2 test could not be used as at least one expected value in toward co-occurrences of features) and has a value between 0 and 1;
each matrix was less than five. (3) For semiquantitative data, e.g., higher values have higher statistical significance. (6) For analysis of the
the presence or absence of various soft tissue features, the number of type of skeletal element present versus the type of disarticulated joint

TABLE 1—Matrix summarizing statistical methods and results. Except for two cells containing a dash (-), cells indicate how the relationship between each pair of indices used
to characterize the taphonomy and lithological context of the frogs was analyzed. Analyses returning statistically significant correlations are indicated in bold, italicized text
within boxes. Analyses returning correlations that are not statistically significant but discussed in the text are indicated within normal text within boxes. A 5 parametric
ANOVA; F 5 Fisher exact test; o 5 other method (see text); P 5 Pearson correlation coefficient; R-C 5 Raup-Crick test; R2 5 X-Y plot using R2 value of trendline; W 5
Welch’s unparametric ANOVA; X-Y 5 X-Y plot where specimens are coded according to a qualitative index.
present, the following were calculated: (i) the percentage of specimens in 1. Facies A comprises intervals (typically 0.6–1.5 m thick) of white to
which each of the major limb joints is disarticulated and in which the light gray laminated mudstones (Fig. 4A). Micritic laminae predominate
bone(s) immediately distal of that joint is (are) missing, and (ii) the and average 2.4 mm thick; silty and organic-rich laminae are rare and, when
percentage of specimens in which any trunk vertebra(e), or abdominal present, very thin (#0.3 mm thick). This facies represents deposition during
joint, is disarticulated and one, or both, bone(s) at that joint is (are) relatively warm, arid, climatic periods when primary productivity was low.
missing. 2. Facies B comprises intervals (typically 0.4–1 m thick) of organic-
In total, 76 pairs of taphonomic indices were tested; the methodology rich laminated mudstones (oil shales) that are characterized by the high
used for each, and the conclusion reached, are summarized in Table 1. abundance of organic-rich laminae (Fig. 4B). Further subdivision of this
facies is based on the relative abundance of micritic and silty laminae.
RESULTS AND INTERPRETATIONS 3. Facies B1 comprises alternations of micritic (average 0.83 mm
thick) and organic-rich laminae (average 0.65 mm thick) (Fig. 4C); silty
Sedimentology laminae are rare. This facies represents deposition during relatively
humid periods characterized by seasonal environmental fluctuations
The sedimentology of the laminated mudstones is summarized below; between periods of low organic input (micritic laminae: early spring
a more detailed description and interpretation is provided in deposition) and periods of high organic input (organic-rich laminae:
Supplementary Data 21. The laminated mudstones comprise the late summer-autumn deposition).
following lamina types (Fig. 4): 4. Facies B2 comprises alternations of silty (average 0.96 mm thick) and
organic-rich (average 0.75 mm thick) laminae (Fig. 4D); micritic laminae
1. Micritic laminae (0.4–10 mm thick) comprise predominantly are rare and, if present, ,1.2 mm thick. This facies represents deposition
micrite (80%–93%) and represent deposition of epilimnetic carbonate during cooler, more humid conditions than Facies A, in a relatively deeper,
precipitates during early spring. These laminae are usually cream more dilute lake; carbonate precipitation was suppressed and organic
colored but can be light gray or light blue-gray in color where plant matter flux to the lake floor was high for much of the year.
fragments or organic matter approach a combined value of 20% of the
total area. Overall, the laminated mudstones represent deposition during high lake
2. Organic-rich laminae (0.2–2 mm thick) are dark gray to black, levels and, therefore, relatively more humid climatic conditions than during
comprise predominantly organic matter (72%–83%) and represent deposition of other sediments in the bituminous-calcareous subunit (see
deposition during high rates of production and or accumulation of Supplementary Data 21). Laminated mudstone facies B2 represents
organic detritus, primarily by algae (with minor inputs from higher deposition during the highest lake levels and most humid climatic
plant remains and bacteria (de las Heras et al., 2003)), during mid- conditions. Transitions between facies occur on a centimeter scale in
summer to autumn. outcrop; assuming an average lamina thickness of ,0.8 mm, these facies
3. Silty laminae (0.2–2 mm thick) are medium to dark gray and transitions indicate that climatic conditions fluctuated on decadal timescales.
comprise organic matter (48%–69%), micrite (20%–2%), bioclasts The relative abundance of each facies was calculated from the total
(12%–20%), and detrital quartz (5%–8%). Silt-sized quartz is typically thickness of the intervals of each recorded during logging (Facies A
concentrated in discrete horizons at, or close to, the base of these (52.44%), Facies B1 (17.94%), and Facies B2 (29.62%).
laminae. These laminae reflect increases in precipitation or runoff into
the lake during autumn or winter flood events. Lithological Context of Specimens
Variation in the relative proportions of these laminae is used to The lamina succession of three frog-bearing slabs was indistinct due
define the following facies: to their having irregular vertical surfaces. Comparison of the vertical

faces of the remaining slabs (n 5 70) reveals no correlation between any is markedly lower, and is lowest for the phalanges; these are complete in
examples. The frogs are, therefore, from separate horizons; specimens the forelimbs of only 25% of specimens, and the hindlimbs of 28%
were introduced to the site of deposition on an on-going basis over an (Fig. 6). 62% of incomplete specimens lack only (some or all)
extended time interval. phalanges; 84% lack a combination of phalanges, tarsals and carpals
Six, 33, and 31 specimens (n 5 70) occur within Facies A, Facies B1, (Fig. 7A). The forelimb and or hindlimb phalanges are always
and Facies B2, respectively. When the relative abundance of each facies incomplete if other elements of the relevant limb(s) are absent
is taken into account, 36.7, 12.6, and 20.7 specimens should occur in (Fig. 7A). The high value of the seriation criterion (0.75) indicates
Facies A, Facies B1, and Facies B2. The difference between the observed significant sequential structure in the data. Collectively, these data
and expected values is statistically significant (61.54; d 5 2; x22 5 13.82; indicate that the loss of skeletal elements was initiated peripherally,
p , 0.001), i.e., the frogs are less abundant in Facies A, and more then occurred at progressively more proximal positions within the
abundant in Facies B1 and B2, than expected. skeleton.
Articulation.—Percentage articulation of specimens varies between
Cause of Death 0% and 100%; most values are greater than 70% (Fig. 5). Articulation
is lowest for the forelimb and hindlimb phalanges (21% and 27.5% of
The primary causes of death of modern adult frogs are (1) specimens, respectively) (Fig. 6). More proximal limb joints exhibit
environmental stress; (2) predation; (3) starvation during hibernation notably higher values for articulation, ranging between 64% and 95%;
(if body fat reserves are insufficient); (4) drowning during or following all vertebrae are articulated in 84% of specimens. Articulation of the
mating (females are at particular risk due to their position beneath the urostyle with the sacral vertebra and ischium is unexpectedly low
males during amplexus); (5) depletion or exhaustion of energy reserves (71.9% and 75% of specimens, respectively). When seriated, the
following mating (females are at particular risk (Beebee and Griffiths, disarticulation data are not ordered according to a simple distal-to-
2000)); and (6) old age (Duellmann and Trueb, 1986; Tyning, 1990). proximal sequence (Fig. 7B); the value of the criterion is low (0.45),
Environmental stress-induced mortalities can result from seasonal indicating that much information is not accounted for by the seriation
overturn of lake waters (Franzen, 1985; Wilson and Barton, 1996), (Fig. 7B).
degassing of sulfides (Schaal and Ziegler, 1992), salinity- or temperature- Low articulation values for bones at joints in the abdomen of the
induced anoxia (Bernier et al., 1991; Martill and Wilby, 1994), loss of a frogs are not attributed primarily to abdominal rupture (i.e., the violent
habitable zone (Elder and Smith, 1988), or release of toxic metabolites release of decay gases from the abdomen; see Elder and Smith, 1988;
during algal blooms (Boyer, 1981; Buchheim and Surdam, 1981). Each of Barton and Wilson, 2005). In only three specimens (discussed later),
these phenomena induce death by asphyxiation (Braun and Pfeiffer, disarticulated abdominal and vertebral joints are displaced from life
2001) and typically generate mass mortality events. There is no evidence position by $3 mm. The low articulation values more likely reflect the
for mass mortality, or any of the aforementioned environmental orientation of the bones in the abdominal region in vivo. The urostyle,
phenomena, in the Libros strata. The nature of preservation of the frogs ilia, and sacral vertebra form a three-dimensional, prism-like structure;
excludes predation as a cause of death. The Libros frogs are also unlikely decay of the surrounding soft tissues (perhaps exacerbated by later
to represent mortalities resulting from starvation during hibernation: compaction) would result in reorientation and, as a result, disarticu-
many specimens contain stomach contents in which specific biological lation of the abdominal skeletal elements.
components can be identified. Further, frogs hibernate in mud in low- The only robust case for abdominal rupture in the frogs from Libros
energy, shallow-water environments. The exhumation and transport of is the specimen illustrated in Fig. 2H. Disarticulation of the abdomen
carcasses that had starved to death would have required extensive and vertebral column is extensive, and the limbs are notably well
sediment agitation, with the carcasses (but not associated sediment) being articulated. Disarticulated elements are displaced by up to 60 mm from
transported to profundal regions; this is considered unlikely. life position and exhibit unidirectional scattering, as expected. The
The presence of a dermal E-K layer strongly indicates a predomi- other two atypical specimens comprise accumulations of entirely
nantly terrestrial ecology (Elkan, 1968). In modern terrestrial adult disarticulated elements and probably reflect disarticulation of carcasses
frogs, reproduction is the primary, if not sole, impetus for returning to at the water surface, possibly following brief initial sinking. Experi-
the water (Tyning, 1990). Furthermore, modern frogs that breed in mental decay of frogs demonstrates that fresh carcasses sink initially at
permanent waterbodies usually exhibit strong breeding site fidelity temperatures up to 24 uC but can refloat subsequently due to the
(Blaustein et al., 1994). Many, possibly most, of the Libros frogs are accumulation of decay gases (Wuttke, 1983). These patterns of
best interpreted as mating-related mortalities. In modern frogs, most disarticulation are similar to those in exceptionally preserved fish from
such mortalities are of females, but the gender distribution of the Libros the Eocene Horsefly Formation (British Colombia); the latter have
specimens cannot be determined unequivocally. been attributed to variation in the temperature of the hypolimnion,
whereby lower temperatures retard bacterial metabolism and, thus,
Physical Taphonomy inhibit the generation of decay gases (Barton and Wilson, 2005). The
patterns of disarticulation exhibited by the three atypical Libros frog
Specimen Size.—Snout–urostyle length varies between 22.4 and specimens could imply relatively high monimolimnetic temperatures
70.3 mm; it is usually (75.5% of specimens; n 5 49) between 43.2 and that allowed either partial flotation (where the carcass does not lift fully
65 mm. The ratio of femur length:tibiofibula length varies between 1.0 from the lake bed) in the case of the specimen in Fig. 2H, or refloating
and 1.27; in most (77%; n 5 48) specimens it is between 1.03 and 1.14. at the water surface, in the case of the entirely disarticulated specimens.
This supports soft-tissue evidence (McNamara et al., 2009) that R. Abdominal rupture could also have been influenced by other factors, in
pueyoi was predominantly terrestrial and our hypothesis that the frogs particular, the depth of the lake at the time of carcass deposition.
represent deaths during mating-related drowning. Taphonomic experiments by Smith and Elder (1985) demonstrated that
Orientation.—All specimens are in dorsoventral aspect. This posi- the generation of sufficient decay gases to cause flotation can be
tions the limbs in approximately the horizontal plane and is the most suppressed when carcasses are deposited in deep water, due in part to
hydrodynamically stable orientation. increased hydrostatic pressure, as well as cooler water temperatures at
Completeness.—Percentage completeness of specimens varies widely depth.
(20%–100%); most values are greater than 70% (Fig. 5). The cranium, Limb Position.—Limb position is highly variable (Fig. 8B). Notably,
femur, humerus, ilium, tibiofibula, and urostyle are present in every limbs are always lateral to, never overlapping, the remainder of the
specimen (Fig. 6). The completeness of the distal elements of each limb carcass. The shoulder, hip, and knee joints do not exhibit angles greater

FIGURE 10—Analyses of articulation and completeness. A) Percentage completeness of specimens in each completeness category, with results of ANOVA analysis. B) Outline
drawing of the left side of a standard Rana skeleton illustrating the frequency with which skeletal elements are lost from disarticulated joints in the fossils. Shown for each joint
is the percentage of specimens in which the joint is disarticulated and (1) (for limb joints) the bone(s) immediately distal of the joint is absent, (2) (for vertebrae and abdominal
joints) the bones at the joint are absent. C) Percentage extent of the skin of specimens in each articulation category, with results of nonparametric Welch’s ANOVA analysis. D)
Percentage completeness for specimens in which each soft tissue feature is present (left-hand column of data for each feature) and absent (right-hand column of data for each
feature), respectively, with results of ANOVA and nonparametric Welch’s ANOVA analyses. (n 5 x / n 5 y) indicates that a feature is present in x specimens, and absent in y
specimens. cc 5 carbonate in the cranium; ph 5 phosphate in the stomach; sd 5 calcium sulfate discoids; sh 5 shell fragments in the stomach; wps 5 well-preserved skin.
Vertical and horizontal bars denote standard deviation and mean, respectively. Significant ANOVA results are indicated by bold, italicized text. compl. 5 completeness;
disartic. 5 disarticulated; excl. 5 excluding; incl. 5 including.
than or, in the case of the hindlimb joints, even close to the range All joints exhibit angles less than the minimum value possible in vivo
possible in vivo (Figs. 8B, C). Angles greater than the maximum (shoulder, 14%; elbow, 6.5%; hip, 13%; knee, 18%). Current activity
possible in vivo do occur for the elbow joint, but this may be an artifact would be unlikely to generate only angles less than the minimum
of underestimating the maximum angle possible in vivo (see Methods). possible in vivo. This feature almost certainly results from either

tetany—postmortem muscular contraction following fatal environmen-

tal stress due to changes in oxygen levels, temperature, salinity or
alkalinity—or osmotic dehydration. Although tetany is well documented
in the fossil record (e.g., Elder and Smith, 1988; Ferber and Wells, 1995;
Barton and Wilson, 2005), evidence for this process in the Libros frogs is
inconclusive. Tetany has been described in detail only from fossil fish
(Smith and Elder, 1985; Elder and Smith, 1988; Ferber and Wells, 1995);
many characteristic features (e.g., distended gills, raised opercula and
stiffened fins) do not apply. Other features indicative of tetany in fish
(e.g., gaping mouths and curvature of the spine in the sagittal plane) may
be obscured by the dorsoventral orientations of the frog specimens. Even
if tetany did occur, the preservation of evidence thereof is problematic.
Some authors have suggested that monimolimnetic anoxia may facilitate
preservation of postmortem muscular contraction (Elder and Smith,
1988), but experimental decay of frogs demonstrates that this feature
persists for only 4 to 5 days under anoxic conditions (Wuttke, 1983).
Highly contracted limb positions can also be generated postmortem
via osmotic dehydration of tissues of carcasses decaying in waters with
high ionic concentrations. This process has been invoked to explain the
death posture of Archaeopteryx, fish, crustaceans, and crinoids from
the Solnhofen fauna (Kemp and Unwin, 1997; Etter, 2002). The Libros
lake was highly alkaline (Ortı́ et al., 2003); the contracted limb positions
exhibited by many of the adult frogs are therefore likely to reflect
postmortem osmotic dehydration of tissues. This process may, of
course, have affected other specimens, but not generated angles less
than those possible in vivo.
Extent of Soft Tissues
Skin and biofilm are most commonly preserved in the thorax

(Figs. 9B, C). Most specimens also contain both skin and biofilm in the
abdomen and upper thighs, but few specimens contain either feature in
the distal parts of the limbs. These data indicate that complete decay of
soft tissues proceeded from the periphery of the carcasses, notably the
limb extremities, toward the core of the body. This does not, however,
imply that decay was initiated in peripheral areas: the persistence of a
biofilm in proximal regions of specimens could reflect the greater
volume of soft tissues available for bacterial consumption in the central
regions of a carcass.
Effect of Lithology on Taphonomy
The analytical results are summarized in Table 1. The results of analyses

returning insignificant results are available in Supplementary Data 41.
There is no statistical evidence that any of the taphonomic indices
varies systematically among the different facies (Supplementary Data 41).
There is no statistical support for the preferential occurrence of any
soft tissue feature in any facies. The taphonomy of the frogs is, there-
fore, independent of centimeter-scale fluctuations in lithology and,
by inference, variations in environmental conditions over decadal
timescales.
Interrelationships among Taphonomic Variables
Completeness and Articulation.—Percentage completeness, and the

type of skeletal element lost, are positively correlated (Fig. 10A). Most
specimens lacking only phalanges exhibit high percentage completeness.
FIGURE 11—Analyses of the extents of the skin and biofilm. A) Percentage extent of the The phalanges represent 63.5% of the total number of skeletal elements;
biofilm versus that of the skin, with results of ANOVA and nonparametric Welch’s
ANOVA analyses. B, C) Percentage extent of the skin (A) and percentage extent of the loss of most phalanges would result in percentage completeness values
biofilm (B) for specimens in which each soft tissue feature is present (left-hand column of of ,50%. Most (88%, n 5 34) specimens lacking only phalanges,
data for each feature) and absent (right-hand column of data for each feature), respectively. however, are more than 80% complete; few specimens, therefore, lack
(n 5 x / n 5 y) indicates that a feature is present in x specimens, and absent in y specimens. many, or most, phalanges. Skeletal element loss, therefore, involved at
cc 5 carbonate in the cranium; ph 5 phosphate in the stomach; sd 5 calcium sulfate
least two distinct stages: initial loss of a limited number of phalanges,
discoids; sh 5 shell fragments in the stomach; wps 5 well-preserved skin. Vertical and
horizontal bars denote standard deviation and mean, respectively. Significant ANOVA followed by the loss of additional phalanges and more proximal
results are indicated by bold, italicized text within boxes. elements, usually from the remainder of the limbs.

TABLE 2—Raup-Crick similarity matrix for the soft tissue features showing the
Raup-Crick index for each pair of co-occurring features. Indices .0.95 are highlighted
in bold, italicized text inside boxes.
Percentage articulation does not correlate with the type of joint

disarticulated (Supplementary Data 41). In particular, there is no
tendency for the bones at specific joints to be disarticulated in
specimens that have low percentage articulation values. This supports FIGURE 12—Summary of relationships between taphonomic indices. Arrows
indicate the direction in which a particular index is dependent on another. Connectors
earlier interpretations (see Physical Taphonomy) that patterns of
without arrowheads denote relationships for which neither index is co-dependent.
disarticulation are more complex than those of skeletal element loss.
Percentage articulation and completeness are strongly, and positive-
ly, correlated, but only if an outlier comprising two specimens that are most recalcitrant nonbiomineralized tissues (Wuttke, 1983). Thus decay
20% complete and 0% articulated is included (Supplementary Data 41). of the skin, and retreat of the biofilm from peripheral regions of the
Excluding these data returns a much weaker positive correlation carcass, would have occurred in tandem. Neither the extent of the
between percentage completeness and articulation (R2 5 0.348). skin nor the biofilm varies systematically with the presence or absence
Disarticulation need not, therefore, result in loss of skeletal elements, of carbonate in the cranium (Figs. 11B, C). There are statistically
and the loss of elements does not imply that the rest of the carcass is
significant differences in the extent of both the skin and biofilm and
disarticulated. Disarticulation of bones at joints in the distal parts of
whether specimens have well-preserved skin, phosphate in the stomach,
the limbs is usually accompanied by loss of the bone(s) distal of the
and calcium sulphate discoids (Figs. 11B, C): the skin and biofilm tend
joint (56%–96% of specimens) (Fig. 10B). Disarticulation of bones at
to be more extensive in specimens containing stomach contents and in
joints in the thorax, abdomen, and in the proximal parts of the limbs,
which the skin is well preserved. The biofilm also tends to be more
however, is rarely associated with loss of those elements (0%–10% of
extensive in specimens containing stomach contents (Fig. 11C).
specimens). Loss of skeletal elements, therefore, reflects position within
Soft Tissue Features.—The presence of carbonate in the cranium does
the skeleton, and requires (but does not automatically follow)
disarticulation at a particular joint. not correlate with the presence of any other soft tissue feature (Table 2).
Statistically significant positive correlations exist between the presence
Percentage articulation and percentage completeness do not correlate
with the extent of the skin or biofilm (Supplementary Data 41). of calcium sulfate discoids and skin being well preserved (Table 2). This
Similarly, the type of skeletal element lost does not correlate with the likely reflects decay of the mid- and lower dermis in tandem: where
extent of the skin or biofilm (Supplementary Data 41). Which joints are decay is incomplete, evidence of the collagen fibers of both the mid-
disarticulated, however, does correlate with the extent of the skin dermis (well-preserved skin) and the lower dermis (calcium sulfate
(Fig. 10C): the latter is more extensive in fully articulated specimens discoids) is likely to be preserved. There are also correlations between
than in disarticulated specimens. The skin is mainly absent in peripheral the presence of phosphate in the stomach and the skin being well
parts of specimens, notably the lower half of each limb. This is preserved, and between the presence of phosphate and shell fragments
considered a prerequisite for disarticulation of distal joints. in the stomach; the latter, although having a p value slightly less than
Percentage articulation, the type of skeletal element disarticulated the threshold used herein (p , 0.05), is sufficiently strong to merit
and the type of skeletal element absent do not correlate with the consideration. We do not, however, consider it likely that all three
presence or absence of any of the soft tissue features (Supplementary indices are interconnected. Specifically, the phosphate in the stomach
Data 41). Specimens with well-preserved skin, however, are more and that replicating collagen fibers in well-preserved skin are unlikely to
complete than would be expected (Fig. 10D). Persistence of the skin have had the same source: these processes occurred during initial, and
during decay and subsequent mineralization of specific layers may have later, stages of decay, respectively (McNamara et al., 2009). Notably,
served to hold in place skeletal elements; the earlier this holding in poorly preserved skin is also replicated in calcium phosphate. The
place occurred, the greater both the degree of skeletal completeness, presence of phosphate in the stomach indicates rapid deposition
and fidelity of preservation of the skin. Specimens containing shell postmortem, i.e., minimal decay of specimens prior to deposition
fragments in the stomach are also more complete than expected (McNamara et al., 2009). Decaying organic matter is a source of
(Fig. 10D). Such specimens clearly died, and were deposited, soon after phosphate ions (Briggs and Kear, 1993b, 1994; Wilby and Whyte, 1995;
ingestion of a meal. Freshly killed carcasses are capable of withstanding Hof and Briggs, 1997). Shell fragments in the stomach of a frog
transport without significant disarticulation (Allison, 1986). specimen would have been associated with labile soft tissues. Decay of
Extent of Skin and Bacterial Biofilm.—Although the mode of this organic matter within the stomach would have released phosphate
preservation of each is different, the percentage extent of skin and ions and likely promoted precipitation of calcium phosphate (McNa-
bacterial biofilm correlate positively (Fig. 11A). This correlation, mara et al., 2009). The correlation, therefore, simply indicates that
although having an R2 value slightly less than the statistically decay was less advanced in specimens with skin than others.
significant threshold used herein (p , 0.05), is considered sufficiently Limb Position.—Variation in the angle between the bones at the
strong to merit further discussion. The persistence of the biofilm shoulder, elbow, hip, and knee joints does not correlate with any other
depends on the presence of tissues to degrade. The skin is among the taphonomic feature (Supplementary Data 41). In particular, there is no

evidence that more incomplete or disarticulated specimens are among the taphonomic indices (Table 1). Ten of these correlations
associated with a wider range of limb orientations. relate to the preserved extent or fidelity of the skin, indicating that this
Specimen Size.—This does not correlate with any other taphonomic anatomical feature is the primary biological factor influencing the
index (Supplementary Data 41). The taphonomic histories of the frogs taphonomy of the frogs.
are independent of specimen size. Articulation.—This is controlled primarily by the extent of preserva-
Relationships among the taphonomic indices are summarized in tion of the skin (i.e., the E-K layer): specimens in which the skin is
Figure 12. extensive and well preserved are better articulated than others.
Extensive or complete degradation of the skin in peripheral parts of
DISCUSSION the body was a prerequisite for, but did not necessarily result in,
disarticulation of distal skeletal elements. Disarticulation of abdominal
Lithological Context of Specimens joints is predominantly related to collapse of the carcass during decay
and or compaction, and, in a single example, by abdominal rupture.
Frogs occur preferentially in Facies B and are rare in Facies A. We Completeness.—Loss of skeletal elements is coupled strongly with
consider that this does not reflect, at least entirely, a sampling bias. their position within the carcass and the fidelity of preservation of the
Facies A occurs interbedded with Facies B; bulk extraction of oil shale skin. Loss of skeletal elements at proximal joints is extremely rare; the
is unlikely to have selectively removed the latter, except where intervals skin and biofilm persist in the thorax, abdomen and proximal parts of
of Facies A were particularly thick. The biased facies distribution of the the limbs for longer during decay. Conversely, loss of skeletal elements
frogs could, however, reflect prevailing climatic conditions. Relatively is most prevalent at distal joints following disarticulation of the relevant
higher precipitation levels during deposition of Facies B would have skeletal elements. This would be favored if, as the positive correlation
resulted in a higher lake level. Given the bench-type morphology of the implies, soft tissues, especially the skin, had degraded completely in
Libros lake, such an increase in lake level would have resulted in these parts of the carcass.
significant lateral migration of shorelines and expansion of shallow- Limb Position.—This does not correlate with any other taphonomic
water habitats landward. The combination of this, and the generally feature or with sedimentary facies. The observed variation in limb
wetter climate, could have facilitated frog population expansions. position among specimens is hypothesized to imply, in part, decay of
Assuming constant mortality rates throughout deposition of the carcasses within waters with high ionic concentrations in the ionic
laminated mudstones, more deaths would, therefore, have occurred concentration of the hypolimnion.
during deposition of Facies B than during deposition of an equivalent Preserved Extent and Fidelity of the Skin.—Preservation of the skin is
thickness of Facies A. facies-independent; this implicitly supports the previous conclusion that
the primary source of phosphate ions was the skin itself (McNamara
Controls upon Taphonomy et al., 2009). Variation in the fidelity of preservation of the skin between
specimens is therefore unexpected as it requires that the timing of
There are no appreciable differences in the fidelity of preservation of mineralization differed among specimens. The presence of well-
frogs among the different facies and, thus, a general model of the preserved and or extensive skin implies decay prior to mineralization
frogs’ taphonomy can be constructed. Surficial water temperatures was limited. This explains the positive correlation of well-preserved skin
were usually sufficiently low to allow sinking of carcasses shortly after with specimens that are near complete, and two other taphonomic
death. The absence of such large aquatic predators as fish prevented features that originated before degradation of the carcass was well
scavenging of intact carcasses before deposition. Deposition of advanced: the presence of phosphate in the stomach, and calcium
carcasses in profundal regions of the lake beneath an anoxic sulphate discoids. The restriction of a calcium phosphate infill to the
monimolimnion prevented disruption by scavenging and bioturbation. stomach implies that its wall was present at the time of mineralization.
Subsequent refloating was usually prevented by the pressure of the The positive correlation between the presence of calcium phosphate,
overlying water column and by low monimolimnetic temperatures. and shell fragments, in the stomach of the frogs suggests that the latter,
The high alkalinity of the lakewaters promoted osmotic dehydration comprising ingested, decaying, organic matter, was both the source of
of tissues that, in at least some specimens, modified the positions of relevant ions and promoted precipitation of calcium phosphate, most
limbs. likely by mediating pH values within the stomach. Other factors that
Exceptional preservation of the Libros frogs, therefore, reflects the could have been involved, but that are impossible to test, include diet,
general hydrological, ecological, and sedimentological characteristics the time since the last meal, and the abundance or activity of gastric
of the lacustrine environment during deposition of the laminated enzymes; the effect of these variables is likely to have been subtle
mudstones. These broad-scale environmental factors in turn reflect: (1) relative to whether or not the stomach contained food contents at the
the prevailing climatic and tectonic regimes, which determine water time of death.
depth, water temperature, and, therefore, position of the monimolim- The calcium sulphate discoids precipitated on the template provided
nion and the extent to which profundal sediments are bioturbated; (2) by collagen fibers of the lower dermis (McNamara et al., 2009), and
the faunal composition of the lake, which determines the extent to must, therefore, have occurred before degradation of the latter.
which carcasses are scavenged; and (3) the composition of the parent Authigenic mineralization of soft tissues is restricted to the lower
material rocks in the drainage area, the main determinant of lake dermis, reflecting the biochemistry of the latter, especially its inherent
chemistry. Short-term variations in environmental conditions, as sources of calcium and sulphate ions and the propensity of collagen
documented by centimeter-scale lithological variations, therefore, had fibers to adsorb calcium ions (McNamara et al., 2009).
no discernable impact upon the taphonomy of the frogs. Critically, The extent of the biofilm correlates positively with both the extent
despite such environmental changes, specimens were consistently and fidelity of preservation of the skin, indicating the three are
deposited a short time after their death in the profundal regions of interlinked. Fossilization of the bacteria, and thus termination of decay,
the lake beneath a persistent anoxic monimolimnion. The result is early in the diagenetic history would result in an extensive biofilm; large
consistently well articulated skeletons enveloped by a carbonaceous areas of skin that had yet to degrade extensively would also still be
biofilm. present. This relationship also explains the correlation between the
Superimposed on this general pattern, however, are taphonomic extent of the biofilm and the presence of phosphate in the stomach and
variations that can be linked to the frogs’ biology and its effect on the calcium sulphate discoids, which are other indicators of incomplete
pattern of decay. In total, sixteen positive correlations are identified decay.

WIDER IMPLICATIONS ACKNOWLEDGMENTS
Statistical analysis of quantitative and qualitative taphonomic data We thank Drs. Cormac O’Connell and Dave Cottell for assistance
for exceptionally preserved taxa has the ability to identify interrela- using TEM facilities at University College Dublin, Eric Callaghan for
tionships between different taphonomic features and the principal assistance with preparation of histological sections and supply of Rana
controls on the fidelity of preservation. Critically, while many of the temporaria specimens, Seán Burke for field assistance, the Aragón
latter are inferred, the basis for linking two or more variables is based Government (exp. 103/2003 from the Dirección General de Patrimonio
exclusively in objective analysis of the data. This integrative statistical Cultural de Aragón and FOCONTUR project) for access to the Libros
approach to taphonomic analysis of exceptionally preserved taxa used site, Prof. Derek Briggs for useful discussion, and several reviewers
herein could be developed further, e.g., via the use of multivariate for comments that improved the manuscript. Dr. Jordi Agustı́ and Dr.
statistical techniques, in order to provide additional insights into the Ángel Galobart (Institut Català de Paleontologia Miquel Crusafont,
interrelationships among taphonomic features. The approach could be Sabadell, Spain), Dr. Sebastià Calzada (Museu de Geologia del
readily applied to investigations of the taphonomy of exceptionally Seminari, Barcelona, Spain), Dr. Sandra Chapman (Natural History
preserved vertebrates in various depositional settings. Notably, the Museum, London), Hermano Miguel Pérez (Museo del Colegio La
general depositional setting of the Libros Konservat-Lagerstätte is Salle, Teruel, Spain), Dr. Eberhard Schindler (Forschungsinstitut
broadly similar to that of many other Mesozoic and Cenozoic Senckenberg), and Dra. Begoña Sánchez (Museo Nacional de Ciencias
Konservat-Lagerstätten, e.g., the Eocene Grube Messel biota of Naturales, Madrid, Spain) provided access to specimens. Funded by
Germany (Schaal and Ziegler, 1992), the Eocene Green River biota Enterprise Ireland Basic Research Grant SC/2002/138 to PJO.
of the United States (Grande, 1984), the Miocene Shanwang biota of
China (Yang and Yang, 1994), and the Pliocene Shiobara biota of REFERENCES
Japan (Allison et al., 2008). This study, therefore, has significant
broader implications. In particular, given that short-term environmen- ALLISON, P.A., 1986, Soft-bodied animals in the fossil record: The role of decay in
tal fluctuations do not impact on the taphonomy of the Libros frogs, we fragmentation during transport: Geology, v. 14, p. 979–981.
ALLISON, P.A., MAEDA, H., TUZINO, T., and MAEDA, Y., 2008, Exceptional
identify three hypotheses for further testing: (1) We anticipate that
preservation within Pliocene lacustrine sediments of Shiobara, Japan: PALAIOS,
vertebrate taxa from other Lagerstätten exhibit similar taphonomic v. 23, p. 260–266.
patterns to the Libros frogs, i.e., where the primary taphonomic control ANADÓN, P., ROSELL, L., and TALBOT, M.R., 1992, Carbonate replacement of
is the general depositional setting and is maintained despite short-term lacustrine gypsum deposits in two Neogene continental basins: Sedimentary
variations in the depositional environment. (2) Where these broad-scale Geology, v. 78, p. 201–216.
environmental conditions are sustained, taphonomic features linked BAAS, M., BRIGGS, D.E.G., VANHEEMST, J.D.H., KEAR, A.J., and DELEEUW, J.W.,
to the organisms’ biology will superimpose the general taphonomic 1995, Selective preservation of chitin during the decay of shrimp: Geochimica et
Cosmochimica Acta, v. 59, p. 945–951.
pattern. Any taphonomic differences between taxa are most likely to BARTON, D.G., and WILSON, M.V.H., 2005, Taphonomic variations in Eocene fish-
reflect differences in the animals’ biology and, potentially, ecology; the bearing varves at Horsefly, British Columbia, reveal 10,000 years of environmental
latter could impact on the distance transported from the point of death change: Canadian Journal of Earth Sciences, v. 42, p. 137–149.
to the site of deposition. (3) Other taphonomic pathways in lacustrine BEEBEE, T.J.C., and GRIFFITHS, R.A., 2000, Amphibians and Reptiles: A Natural
systems that were not encountered in this study may be more sensitive History of the British Herpetofauna, The New Naturalist: Harper Collins, London,
to short-term environmental fluctuations. Finally, the sensitivity of 270 p.
BERNIER, P., GAILLARD, C., GALL, J.-C., BARALE, G., BOURSEAU, J.P., BUFFETAUT, E.,
taphonomic pathways to short-term environmental fluctuations in and WENZ, S., 1991, Morphogenetic impact of microbial mats on surface structures
other, nonlacustrine, depositional settings remains to be tested. of Kimmeridgian micritic limestones (Cerin, France): Sedimentology, v. 38, p. 127–
136.
CONCLUSIONS BLAUSTEIN, A.R., WAKE, D.B., and SOUSA, W.P., 1994, Amphibian declines: Judging
stability, persistence, and susceptibility of populations to local and global
The Libros frogs occur preferentially in laminated mudstones deposited extinctions: Conservation Biology, v. 8, p. 60–71.
BOYER, B.W., 1981, Tertiary lacustrine sediments from Sentinel Butte, North Dakota,
under relatively humid conditions, when the extent of anuran littoral and the sedimentary record of ectogenic meromixis: Journal of Sedimentary
habitats was at a maximum. Most specimens likely reflect mating-related Petrology, v. 51, p. 429–440.
drowning. The taphonomy of the frogs is independent of centimeter-scale BRAUN, A., and PFEIFFER, T., 2001, Cyanobacterial blooms as the cause of a
lithological variation, and thus short-term environmental fluctuations. Pleistocene large mammal assemblage: Paleobiology, v. 28, p. 139–154.
The consistently high fidelity of skeletal preservation and the preservation BRETT, C.E., ALLISON, P.A., TSUJITA, C.J., SOLDANI, D., and MOFFAT, H.A., 2006,
of soft tissues reflect primarily the broad-scale depositional setting, i.e., Sedimentology, taphonomy, and paleoecology of meter-scale cycles from the
Upper Ordovician of Ontario: PALAIOS, v. 21, p. 530–547.
deposition shortly after death in profundal zones of a deep, fishless,
BRETT, C.E., ALLISON, P.A., DESANTIS, M.K., LIDDELL, W.D., and KRAMER, A., 2009,
stratified, alkaline lake. Superimposed upon this very general environ- Sequence stratigraphy, cyclic facies, and Lagerstätten in the Middle Cambrian
mental control are a suite of factors that relate specifically to the biology of Wheeler and Marjum Formations, Great Basin, Utah: Palaeogeography,
the frogs. The most important of these is the inherent biochemistry of Palaeoclimatology, Palaeoecology, v. 277, p. 9–33.
specific tissues, in particular, the high recalcitrance of the skin. Specimens BRIGGS, D.E.G., and KEAR, A.J., 1993a, Decay and preservation of polychaetes:
with extensive and or well-preserved skin tend to be more complete and Taphonomic thresholds in soft-bodied organisms: Paleobiology, v. 19, p. 107–135.
BRIGGS, D.E.G., and KEAR, A.J., 1993b, Fossilization of soft-tissue in the laboratory:
better articulated than others. Disarticulation and loss of bones affects the
Science, v. 259, p. 1439–1442.
periphery of the specimen preferentially and more extensively. These are BRIGGS, D.E.G., and KEAR, A.J., 1994, Decay and mineralization of shrimps:
areas in which the skin degrades completely first. Specimens that exhibit PALAIOS, v. 9, p. 431–456.
extensive and or well-preserved skin also tend to possess other soft-tissue BRIGGS, D.E.G., KEAR, A.J., BAAS, M., DE LEEUW, J.W., and RIGBY, S., 1995, Decay
features that imply limited degradation at the time of their formation; and decomposition of the hemichordate Rhabdopleura: Implications for the
these include the presence of an extensive biofilm, calcium sulphate taphonomy of graptolites: Lethaia, v. 28, p. 15–23.
discoids, and calcium phosphate in the stomach. The ultimate cause of the BRIGGS, D.E.G., SIVETER, D.J., and SIVETER, D.J., 1996, Soft-bodied fossils from a
Silurian volcaniclastic deposit: Nature, v. 382, p. 248–250.
taphonomic variations exhibited by the frogs can, therefore, be traced back BROWER, J.C., and KYLE, K.M., 1988, Seriation of an original data matrix as applied
to these biological controls. The ultimate biological controls upon to palaeoecology: Lethaia, v. 21, p. 79–93.
taphonomy are therefore the biochemistry and recalcitrance of specific BUCHHEIM, H.P., and SURDAM, R.C., 1981, Palaeoenvironments and fossil fishes of the
tissues, most critically, the skin. Laney Member, Green River Formation, Wyoming, in Gray, J., Boucot, A.J., and

Berry, W.B.N., eds., Communities of the Past: Dowden, Hutchinson and Ross, MCCOBB, L.M.E., BRIGGS, D.E.G., CARRUTHERS, W.J., and EVERSHED, R.P., 2003,
Stroudsburg, Pennsylvania, p. 415–452. Phosphatisation of seeds and roots in a Late Bronze Age deposit at Potterne,
CAMBRA-MOO, O., and BUSCALIONI, A.D., 2003, Biostratinomic patterns in archosaur Wiltshire, UK: Journal of Archaeological Science, v. 30, p. 1269.
fossils: Influence of morphological organization on dispersal: Journal of MCNAMARA, M., ORR, P.J., KEARNS, S.L., ANADÓN, P., ALCALÁ, L., and PEÑALVER-
Taphonomy, v. 1, p. 247–296. MOLLÀ, E., 2006, High-fidelity organic preservation of bone marrow in c. 10
CASEY, M.M., FRASER, N.C., and KOWALEWSKI, M., 2007, Quantitative taphonomy of million year old amphibians: Geology, v. 34, p. 641–644.
a Triassic reptile Tanytrachelos anhynis from the Cow Branch Formation, Dan MCNAMARA, M., ORR, P.J., KEARNS, S.L., ANADÓN, P., ALCALÁ, L., and PEÑALVER-
River Basin, Solite Quarry, Virginia: PALAIOS, v. 22, p. 598–611. MOLLÀ, E., 2009, Soft tissue preservation in Miocene frogs from Libros (Spain):
CLARK, M.A., WORRELL, M.B., and PLESS, J.E., 1997, Postmortem changes in soft Insights into the genesis of decay microenvironments, PALAIOS, v. 24, p. 104–117.
tissues, in Haglund, W.D., and Sorg, M.H., eds., Forensic Taphonomy: The MCNAMARA, M.E., ORR, P.J., MANZOCCHI, T., ALCALÁ, L., ANADÓN, P., and
Postmortem Fate of Human Remains: CRC Press, Boca Raton, p. 151–164. PEÑALVER, E., 2011, Biological controls upon the physical taphonomy of
DE LAS HERAS, F.X.C., ANADÓN, P., and CABRERA, L., 2003, Biomarker record exceptionally preserved salamanders from the Miocene of Rubielos de Mora,
variations in lacustrine coals and oil shales: Contribution from Tertiary basins in northeast Spain: Lethaia, doi: 10.1111/j.1502-3931.2011.00274.x.
NE Spain, in Valero Garcés, B.L., ed., Limnogeology in Spain: A Tribute to Kerry NAVÁS, L., 1922a, Algunos fósiles de Libros (Teruel): Boletı́n de la Sociedad Ibérica
Kelts: Biblioteca de Ciencias: Spanish Research Council (Consejo Superior de de Ciencias Naturales, v. 21, p. 52–61.
Investigaciones Cientı́ficas (CSIC)), Madrid, p. 187–228. NAVÁS, L., 1922b, Algunos fósiles de Libros (Teruel). Adiciones y correcciones:
DUELLMANN, W.E., and TRUEB, L., 1986, Biology of Amphibians: McGraw-Hill, New Boletı́n de la Sociedad Ibérica de Ciencias Naturales, v. 21, p. 172–175.
York, 670 p. OLSON, S.L., 1995, Thiornis sociata NAVÁS, a nearly complete Miocene Grebe (Aves:
ELDER, R.L., and SMITH, G.R., 1988, Fish taphonomy and environmental inference in Podicipedidae): Courier Forschungsinstitut Senckenberg, v. 181, p. 131–140.
palaeolimnology: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 62, p. ORR, P.J., KEARNS, S.L. and BRIGGS, D.E.G., 2002, Backscattered electron imaging of
577–592. fossils exceptionally-preserved as organic compressions: PALAIOS, v. 17, p. 110–117.
ELKAN, E., 1968, Mucopolysaccharides in the anuran defense against dessication: ORTÍ, F., ROSELL, L., and ANADÓN, P., 2003, Deep to shallow lacustrine evaporites in
Journal of Zoology, London, v. 155, p. 19–53. the Libros gypsum (southern Teruel Basin, Miocene, NE Spain): An occurrence of
ETTER, W., 2002, Plattenkalk preservation with Archaeopteryx, in Bottjer, D.J., Etter, pelletal gypsum rhythmites: Sedimentology, v. 50, p. 361–386.
W., Hagadorn, J.W., and Tang, C.M., eds., Exceptional Fossil Preservation: ORTÍ, F., ROSELL, L., and ANADÓN, P., 2010, Diagenetic gypsum related to sulphur
Critical Moments and Perspectives in Earth History and Paleobiology: Columbia deposits in evaporites (Libros Gypsum, Miocene, NE Spain): Sedimentary
University Press, New York, p. 327–352. Geology, v. 228, p. 304–318.
FERBER, C.T., and WELLS, N.A., 1995, Paleolimnology and taphonomy of some fish PEÑALVER, E., 1996, Los yacimientos con insectos fósiles de Aragón (España):
deposits in ‘‘Fossil’’ and ‘‘Uinta’’ lakes of the Eocene Green River Formation, PaleoEntomologı́a, v. 16, p. 139–146.
Utah, Wyoming: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 117, p. SAGEMANN, J., BALE, S.J., BRIGGS, D.E.G., and PARKES, R.J., 1999, Controls on the
185–210. formation of authigenic minerals in association with decaying organic matter: An
FOREMAN, B.Z., FRICKE, H.C., LOHMANN, K.C., and ROGERS, R.R., 2011, experimental approach: Geochimica et Cosmochimica Acta, v. 63, p. 1083–1095.
Reconstructing paleocatchments by integrating stable isotope records, sedimen- SCHAAL, S., and ZIEGLER, W., 1992, Messel: An Insight into the History of Life and of
tology, and taphonomy: A late Cretaceous case study (Montana, United States), the Earth: Oxford, Clarendon Press, 328 p.
PALAIOS, v. 26, p. 545–554. SHUBIN, N.H., and JENKINS, F.A.J., 1995, An early Jurassic jumping frog: Nature, v.
FRANZEN, J.L., 1985, Exceptional preservation of Eocene vertebrates in the lake 377, p. 49–52.
deposit of Grube Messel (West Germany): Philosophical Transactions of the Royal SKINNER, E.S., 2005, Taphonomy and depositional circumstances of exceptionally
Society of London B, v. 311, p. 181–186. preserved fossils from the Kinzers Formation (Cambrian), southeastern Pennsyl-
GABBOTT, S.E., 1998, Taphonomy of the Ordovician Soom Shale Lagerstätte: An vania: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 220, p. 167–192.
example of soft tissue preservation in clay minerals: Paleontology, v. 41, p. 631–667. SMITH, G.R., and ELDER, R.L., 1985, Environmental interpretation of burial and
GIL, M.J.Q., 1986, Organic microscopic remains in Miocene lacustrine sediments near preservation of Clarkia fishes., in Smiley, C.J., ed., Late Cenozoic History of the
Libros (Teruel, Spain): Hydrobiologia, v. 143, p. 209–212. Pacific Northwest: Pacific Division AAAS, California Academy of Science, San
GRANDE, L., 1984, Paleontology of the Green River Formation with a review of Francisco, p. 85–93.
thefish fauna: The Geological Survey of Wyoming, Bulletin 63, 333 p. STIGALL, A.L., BABCOCK, L.E., BRIGGS, D.E.G., and LESLIE, S.A., 2008, Taphonomy
GREGOR, H.-J., and GÜNTHER, T., 1985, Neue pflanzenfunde aus dem Vallesium of lacustrine interbeds in the Kirkpatrick Basalt (Jurassic), Antarctica: PALAIOS,
(jüngeres Neogen) von Libros (Becken von Teruel, Spanien): Mitteilungen v. 23, p. 344–355.
Badischen Landesverein für Naturkunde und Naturschutz, v. 13, p. 297–309. TREWIN, N.H., 1986, Palaeoecology and sedimentology of the Achanarras fish bed of
HAMMER, Ø., HARPER, D.A.T., and RYAN, P.D., 2001, PAST: Paleontological the Middle Old Red Sandstone, Scotland: Transactions of the Royal Society of
Statistics software package for education and data analysis: Palaeontologia Edinburgh: Earth Sciences, v. 77, p. 21–46.
Electronica, v. 4, p. 9. TYNING, T., 1990, Stokes Nature Guides: A Guide to Amphibians and Reptiles: Little,
HOF, C.H.J., and BRIGGS, D.E.G., 1997, Decay and mineralization of mantis shrimps Brown and Company, Boston, Massachusetts, 416 p.
(Stomatopoda: Crustacea): A key to their fossil record: PALAIOS, v. 12, p. 420–38. VOIGT, E., 1988, Preservation of soft tissues in the Eocene lignite of the Geiseltal near
HOU, X.G., and BERGSTRÖM, M.J., 1997, Arthropods of the Lower Cambrian Halle/S.: Courier Forschungsinstitut Senckenberg, v. 107, p. 325–343.
Chengjiang fauna, southwest China: Fossils and Strata, v. 45, p. 1–116. WHITTINGTON, H.B., 1971, Redescription of Marrella splendens (Trilobitoidea) from
KARGO, W.J., NELSON, F., and ROME, L.C., 2002, Jumping in frogs: Assessing the the Burgess Shale, Middle Cambrian, British Columbia: Geological Survey of
design of the skeletal system by anatomically realistic modeling and forward Canada Bulletin, v. 209, p. 1–31.
dynamic simulation: Journal of Experimental Biology, v. 205, p. 1683–1702. WHITTINGTON, H.B., 1977, The lobopod animal Aysheaia pedunculata Walcott,
KEAR, A.J., BRIGGS, D.E.G., and DONOVAN, D.T., 1995, Decay and fossilization of Middle Cambrian, Burgess Shale, British Columbia: Philosophical Transactions of
non-mineralized tissue in coleoid cephalopods: Palaeontology, v. 38, p. 105–131. the Royal Society of London, B, v. 284, p. 165–197.
KEMP, R.A., and UNWIN, D.M., 1997, The skeletal taphonomy of Archaeopteryx: A WILBY, P.R., and BRIGGS, D.E.G., 1997, Taxonomic trends in the resolution of detail
quantitative approach: Lethaia, v. 30, p. 229–238. preserved in fossil phosphatized soft tissues: Geobios, v. 20, p. 493–502.
KUMBAR, S.M., and PANCHARATNA, K., 2001, Determination of age, longevity and age WILBY, P.R., and WHYTE, M.A., 1995, Phosphatized soft tissues in bivalves from the
at reproduction of the frog Microhyla ornata by skeletochronology: Journal of Portland Roach of Dorset (Upper Jurassic): Geological Magazine, v. 132, p. 117–120.
Bioscience, v. 26, p. 265–270. WILBY, P.R., BRIGGS, D.E.G., BERNIER, P., and GAILLARD, C., 1996a, Role of
LIEBIG, K., 1998, Fossil microorganisms from the Eocene Messel Oil Shale of microbial mats in the fossilization of soft tissues: Geology, v. 24, p. 787–790.
Southern Hesse, Germany: Kaupia, v. 7, p. 1–95. WILBY, P.R., BRIGGS, D.E.G., and RIOU, B., 1996b, Mineralization of soft-bodies
LIN, J.-P., 2006, Taphonomy of naraoiids (Arthropoda) from the Middle Cambrian invertebrates in a Jurassic metalliferous deposit: Geology, v. 24, p. 847–850.
Kaili Biota, Guizhou Province, South China: PALAIOS, v. 21, p. 15–25. WILSON, M.V.H., and BARTON, D.G., 1996, Seven centuries of taphonomic variation
LUQUE, L., MERINO, L., SANCHIZ, B., and ALCALÁ, L., 1996, Procesos de fosilización de in Eocene freshwater fishes preserved in varves: Paleoenvironments and temporal
las ranas miocenas de Libros (Teruel): Comunicaciones de la II Reunión de averaging: Paleobiology, v. 22, p. 535–542.
Tafonomı́a y Fosilización, Zaragoza, Spain, p. 169–174. WUTTKE, M., 1983, Aktuopaläontologische studien über den Zerfall von Wirbeltieren.
MARGALEF, R., 1947, Observaciones micropaleontológicas sobre los sedimentos Teil 1: Anura: Senckenbergiana lethaea, v. 64, p. 529–560.
lacustres miocénicos de Libros (Teruel): Estudios Geológicos, v. 5, p. 171–177. YANG, H. and YANG, S., 1994, The Shanwang fossil biota in eastern China: A
MARTILL, D.M., and WILBY, P.R., 1994, Lithified prokaryotes associated with fossil Miocene Konservat-Lagerstatten in lacustrine deposits: Lethaia, v. 27, p. 345–354.
soft organisms from the Santana Formation (Cretaceous) of Brazil: Kaupia, v. 4, p.
71–77. ACCEPTED NOVEMBER 21, 2011


Palo 2010 p10-126r

Uploaded by

Document Information

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

Palo 2010 p10-126r

Uploaded by

Copyright:

Available Formats

WHAT CONTROLS THE TAPHONOMY OF EXCEPTIONALLY PRESERVED

TAXA—ENVIRONMENT OR BIOLOGY? A CASE STUDY USING FROGS

BioOne Complete (complete.BioOne.org) is a full-text database of 200 subscribed and open-access

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

WHAT CONTROLS THE TAPHONOMY OF EXCEPTIONALLY PRESERVED TAXA—ENVIRONMENT

ABSTRACT depositional context has been explored in detail in many studies of

* Corresponding author. 1 palaios.ku.edu

Copyright G 2012, SEPM (Society for Sedimentary Geology) 0883-1351/12/0027-0063/$3.00

evaporative concentration of the lake probably occurred during brief

MATERIAL AND METHODS

Sections of the bituminous-calcareous subunit were logged in the

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

FIGURE 5—Plot of percentage completeness and percentage articulation of

angle possible at these joints is unlikely to have been underestimated as

Extent of Soft Tissues

FIGURE 4—Laminated mudstone facies. A, B) Photographs of Facies A and B, with

in which joints distal of the tibiofibula and radioulna only are

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

tetany—postmortem muscular contraction following fatal environmen-

Extent of Soft Tissues

Skin and biofilm are most commonly preserved in the thorax

Effect of Lithology on Taphonomy

The analytical results are summarized in Table 1. The results of analyses

Interrelationships among Taphonomic Variables

Completeness and Articulation.—Percentage completeness, and the

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

Percentage articulation does not correlate with the type of joint

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

WIDER IMPLICATIONS ACKNOWLEDGMENTS

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

Downloaded From: https://bioone.org/journals/Palaios on 09 Oct 2019

You might also like