• TABLE OF CONTENTS
HIDE
 Front Cover
 Synopsis
 Introduction
 Acknowledgments and abbreviati...
 Osteological resume of the recent...
 Shell
 Literature cited
 Figures 1-33
 Back Cover






Title: genus Gopherus (Testudinidae)
CITATION DOWNLOADS THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00027851/00001
 Material Information
Title: genus Gopherus (Testudinidae)
Physical Description: Book
Creator: Auffenberg, Walter.
Publisher: Universiy of Florida,
Copyright Date: 1976
 Record Information
Bibliographic ID: UF00027851
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: 003318781 - Electronic_Aleph
60779406 - Electronic_OCLC

Downloads
Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Synopsis
        Page 47
    Introduction
        Page 48
        Page 49
        Page 50
    Acknowledgments and abbreviations
        Page 51
    Osteological resume of the recent species
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
    Shell
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
    Literature cited
        Page 82
        Page 83
    Figures 1-33
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
    Back Cover
        Page 111
        Page 112
Full Text







of the
FLORIDA STATE MUSEUM
Biological Sciences

Volume 20 1976 Number 2




THE GENUS GOPHERUS (TESTUDINIDAE):
PT. I. OSTEOLOGY AND RELATIONSHIPS OF EXTANT SPECIES

d WALTER AUFFENBERG







FSM
UNIVERSITY OF FLORIDA GAINESVILLE

UNIVERSITY OF FLORIDA GAINESVILLE










Numbers of the BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL
SCIENCES, are published at irregular intervals. Volumes contain about 300 pages and
are not necessarily completed in any one calendar year.






CARTER R. GILBERT, Editor
RHODA J. RYBAK, Managing Editor



Consultants for this issue:


JAMES L. DOBIE

J. ALAN HOLMAN

SAM R. TELFORD, JR.









Communications concerning purchase or exchange of the publications and all man-
uscripts should be addressed to the Managing Editor of the Bulletin, Florida State
Museum, Museum Road, University of Florida, Gainesville, Florida 32611.





This public document was promulgated at an annual cost of
$2,561.67 or $2.328 per copy. It makes available to libraries,
scholars, and all interested persons the results of researches in
the natural sciences, emphasizing the Circum-Caribbean region.


Publication date: January 23, 1976 Price $2.35


































THE GENUS GOPHERUS (TESTUDINIDAE):
PT. I. OSTEOLOGY AND RELATIONSHIPS OF EXTANT SPECIES

WALTER AUFFENBERG1

SYNOPSIS: Adult skeletons of the extant species of the genus Gopherus were studied
to determine the kind and level of similarities and differences between them and to
form a comparative base for studies of fossil members of the genus. The skeleton
and its variation in each of the species is described and/or figured and analyzed.
The four extant species form two species groups, based on a number of osteological
characters. One group includes G. polyphemus and G. flavomarginatus; the other
C. berlandieri and G. agassizi. G. polyphemus and berlandieri represent the morpho-
logical extremes of each group. G. flavomarginatus and agassizi overlap in a number
of characters, and may be close to the evolutionary base of both groups.










1 The author is Curator in Herpetology, Florida State Museum, University of Florida,
Gainesville, Florida 32611. Manuscript accepted 15 January 1973.

AUFFENBERG, WALTER. 1975. The genus Gopherus: Pt. I. Osteology and relation-
ships of extant species. Bull. Florida State Mus., Biol. 20(2): 47-110.





5-10. 9


48 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

e- TABLE OF CONTENTS

INTRODUCTION ----...--.....------.------------------- ------------------------ 48
ACKNOWLEDGEMENTS AND ABBREVIATIONS .------------.------------------.----------------- 51
ANATOMICAL RESUME OF THE RECENT SPECIES ...-..---.-------- -------------------52 52
The Skull ..------ ..---------.------- ------------------- .--- 52
Cervical Vertebrae ----_._-----------._.. ....-----..-- ---- ------ 5 57
Caudal Vertebrae ..-------------- ---------------------------------- 57
Humerus .--------- --- ------------------.-- ----------- 58
Pelvic Girdle .-----.--------------- ----------- -------- 59
Femur .---. --------------- ----.----------.--------- 60
Carpus .---__--- ------------------ ------------ 60
Tarsus ..---------.. .... ...------------------ ---------- 62
Dermal Armor .-------.___..-. ----. ----------.----------------------- 62
Radius and Ulna -...-.----_.. --------------- ----_.--..---- 64
THE SHELL .....------ ------ -..-. .....--------------.---------.. 65
Plastron _.----.--.----..--...-------..--- ------------ 75
Carapace .. ----------... ---------. ----..--..----------. 75
LITERATURE CITED ...------..........--.----. ....- -----............... 82



INTRODUCTION

The present osteological study of extant populations of the genus
Gopherus is the first of several intended to treat both Recent and fossil
populations. It has two purposes: (1) to determine the systematic
relations of these populations to one another, and (2) as a basis of com-
parison for studies of the earlier fossil members of the genus.
The earliest known geologic occurrence of the genus is in Lower Oli-
gocene rocks of the White River Formation in Colorado, Nebraska, Wy-
oming, and South Dakota (Gopherus laticunae and Gopherus praextans).
From that period to the present the fossil history is unbroken, although
the relationships of the extinct populations to one another are not clear.
What is certain, however, is that the genus Gopherus is closely related
to the extinct genus Stylemys. This is based on certain features of the
shell and skull (Williams 1950a; Hay 1908). The most important cor-
relative character is the presence of a median premaxillary ridge, which
is found only in these two genera of land tortoises. The presumed
phalangeal differences between these genera mentioned by several earlier
workers has been shown to be incorrect (Auffenberg 1961). However,
certain differences in rate of shell ossification during ontogenetic develop-
ment and the shape of particular parts of the skeleton in adults seem
sufficiently great to consider these two related phyletic lines distinct at








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 49

the generic level (Auffenberg 1963), at least since Oligocene time.
On the basis of morphological similarity, and even intermediacy in
some cases, the genus Gopherus is believed to have evolved from an
early member of Stylemys, probably in western North America during
the Late Eocene. The genus Gopherus is, and has always been, re-
stricted to the Nearctic Realm. Its failure to migrate over Cenozoic
land bridges to either Asia or South America was probably due to eco-
logical factors.
The salient morphologic characters of the genus Gopherus are: a
continuation of the palatal vomerine ridge upon the premaxillaries at
their symphyses; short cervical vertebrae with an articular arrangement
considered normal within the crypotodires; length of hypoplastron and
hyoplastron about equal; forelimbs and claws flattened and widened;
antebrachial scales juxtaposed, flattened, and broad; head scales small,
flat, and rounded; eyes moderate in size; usually a single, large supra-
caudal shield; alveolar surface of the upper jaw with an inner ridge run-
ning parallel with the cutting edge; alternate anterior neurals usually
octagonal; anterior palatine foramina small; distal and proximal ends of
humerus considerably expanded; tail very short; caudal vertebrae short
and broad, without aberrant processes; skull wide, with a well developed
post-orbital bridge, quadrate enclosing stapes; pro-otic exposed; shell
broad, moderate in height; costal scutes low and broad; vertebral scutes
wide; nuchal scute usually as wide, or wider, than long; hind foot short,
elephantine, with short metatarsals; tibiale and fibulaire usually fused;
carpus with a distinct proximal central, at least in the young, but often
fused with an adjacent central in adults.
At the present time the genus is comprised of four allopatric popula-
tions that have long been considered distinct at the species level. The
relegation of these populations to subspecific rank (Gray 1873; Mertens
and Wermuth 1955) implies a degree of morphological similarity and
an evolutionary history that is not consistent with our present knowledge.
As will be shown below, the four Recent populations are morphologically
distinct, although their superficial similarities often obscure important
differentiating characters. Furthermore, fossil evidence based both on
distinctness of form and at least partial overlap of geographic range
during the same geologic period (Sangamon Pleistocene) suggests that
the populations are distinct at the species level, and that they have been
genetically separated from one another for a considerable length of time
(Auffenberg, study in progress).
Hybrids resulting from a cross between captive Gopherus agassizi
and Gopherus berlandieri have been described (Woodbury 1952; Mer-
tens 1964), but the origin of at least one of the parents was uncertain








50 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

and its identification in doubt. More importantly, it is not known
whether the young were infertile or not. Furthermore, hybridization be-
tween these two forms, even if proven, does not affect the relationship
of either of them to G. polyphemus and/or G. flavomarginatus; inasmuch
as G. agassizi and G. berlandieri are clearly more similar morphologically
than either are to G. polyphemus or G. flavomarginatus. This view, held
by several previous authors, is based on several morphological characters.
Thus, G. berlandieri is more similar to G. agassizi than either is to G.
polyphemus in alveolar angle, hind foot diameter, head width and pro-
portionate shell height (True 1882; Bogert and Oliver 1945; Williams
1950a, 1952; Carr 1957; Legler 1959; Brattstrom 1961; Legler and Webb
1961; et al.). Gopherus flavomarginatus was considered more closely
related to G. polyphemus than to either of the two remaining species by
Legler (1959) and Legler and Webb (1961) on the basis of these char-
acters and by Auffenberg (1966a) on the basis of carpal architecture.
This conclusion is supported by the studies of Rose, Drotman, and
Weaver (1969) on electrophoretic composition and separation of chin
gland exudates. Grant's purely speculative statement (1960) that flavo-
marginatus may be a synonym of agassizi lacks supporting evidence.
The lack of knowledge concerning morphological variation in tortoise
species is due largely to the difficulty of obtaining sufficiently large
skeletal series. This is particularly true of fossil tortoises, where previous
work usually has been based on the study of single specimens or very
small samples. In this study large numbers of specimens of each of the
four allopatric Recent populations were accumulated for osteological
comparisons: G. polyphemus, 63 (Florida, Georgia, Alabama, and Lou-
isiana); G. agassizi, 42 (Arizona, California, Utah, and Sonora); G. ber-
landieri, 54 (Texas, Tamaulipas, Coahuila and Nuevo Leon); and G.
flavomarginatus, 24 (Durango).
The first and most important step in an investigation of this type is
the selection of characters to be used. The concepts of size and shape
are fundamental to such analyses. Shape tends to provide more reliable
indications of relationships among reptiles than size (Jolicoeur and Mosi-
mann 1960; et al.). Careful study and comparison of the available
Recent and fossil material indicate that certain structural details of the
skeleton, scute proportions, and skeletal dimensions of the shell, limbs,
and skull are probably the most useful and significant taxonomic char-
acters. The exact nature of these and the method in which they were
measured are identified and discussed below. These measurements are
then analyzed.
Statistical treatment of linear measurements on reptiles is often com-
plicated by the fact that objective criteria for identifying the growth








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 51

stage are lacking. In many studies on reptiles such difficulties are be-
lieved avoided by considering the relative size of single or multiple
groups of unit characters. Most often this approach is essentially a uni-
variate characterization, expressed as a ratio in which one variable is
considered independent of the other. In many cases this independence
is improperly assumed; in others it is difficult to establish. Statistically
it is considered valid only if the variances of the two variables are similar
(Kermock and Holdon 1950; et al.). To establish the best possible dis-
crimination technique, a number of characters were analyzed by ratios,
log-ratios, and actual measurements. The results of the comparative
analysis clearly showed that log-ratio and simple ratio comparisons are
less discriminatory than actual measurements in adults (see Tables 13-
14). Three major types of statistical analyses based on actual measure-
ments are employed in this paper: (1) univariate characterization,
establishing the mean and standard deviation of all characters analyzed,
(2) multivariate characterization of a number of differently combined
sets of 36 variables on the shell, and (3) linear discrimination by means
of the BIOMED 005 technique (Kendall 1951). Details of analyses will
be found in specific sections below.

ACKNOWLEDGMENTS AND ABBREVIATIONS
Unless indicated otherwise all the specimens of extant populations upon which
this study is based are in the collections of the Florida State Museum, University
of Florida (UF). Fossil materials were examined from this same collection, plus
that of the American Museum of Natural History (AMNH), Chicago Natural History
Museum (CNHM), United States National Museum (USNM), University of Michi-
gan, Museum of Zoology (UMMZ), Los Angeles County Museum (LACM), Uni-
versity of Arizona (UA), University of Texas (UTBEG), Academy of Natural Sci-
ences of Philadelphia (ANSP), Panhandle Plains Historic Museum (PPHM), Uni-
versity of California, Museum of Paleontology (UCMP), Museum of Comparative
Zoology (MCZ), University of Colorado Museum (UCM), University of Kansas,
Museum of Natural History (UK),.Charleston Museum (ChM), and Tulane Uni-
versity (TU). I wish to thank all of the persons in charge of these collections for
allowing me to examine specimens in their care.
Special acknowledgment is due the National Science Foundation (Grant GB
1362), the Graduate School of the University of Florida (Grant 211*A56), and the
American Philosophical Society (Grant # 406, Johnson Fund) for their financial sup-
port.
For gifts of specimens and data I would like particularly to thank W. T. Neill,
New Port Richey, Florida; J. D. Woolever, Sarasota, Florida; G. Einem, Melbourne,
Florida; J. Legler, University of Utah; J. McCoy, Carnegie Museum; and N. Tess-
man, Wm. Weaver, R. Woodruff, G. Gourley, and L. D. Ober, all formerly of the
University of Florida. I would also like to indicate my appreciation to those artists
who prepared many of the illustrations in this publication: Jane Larson, formerly
of the BSCS, University of Colorado; Ross Norris, University of Wisconsin; Erik
Speyer, University of Florida; Robert Nesby, Brown University; and Margaret Estey,
formerly of the Florida State Museum. Elizabeth S. Wing, Florida State Museum,
and F. Brandt, University of Florida Computing Center, deserve special recognition
for their assistance in data analysis.








52 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

OSTEOLOGICAL RESUME OF THE RECENT SPECIES

THE SKULL
Among testudinine turtles Gopherus has been recognized as posses-
sing a distinctive skull, and it was on this basis that Gray (1873) estab-
lished the Tribe Xenobatiana. It presents a complex of both primitive
and specialized structures. The unique median premaxillary ridge is
shared with the extinct genus Stylemys. In Gopherus the exposure of the
pro-otic varies from completely visible to completely covered. An ex-
posed pro-otic is generally considered a primitive condition (Loveridge
and Williams 1957). An Os transiliens has been reported in all extant
species (Ray 1959; Legler 1962; and Bramble 1974). The prefrontals
are short (specialized), but still enter the orbit (generalized). The re-
duced postorbital in this genus is believed to be a specialization. The
temporal region is broadly emarginate posteriorly. Although the jugal
is small, the quadratojugal is well developed and frequently in contact
with the maxillary. The quadrate is considered specialized because it
encloses the stapes. The maxillary is specialized, and usually possesses
three ridges: an outer cutting edge, one near the posterior edge of the
expanded alveolar surface, and another between these. Although the
anterior palatal foramina are large, they are somewhat concealed (primi-
tive?), because the palate is highly vaulted (specialized). The basisphe-
noid is moderately small, and the ethmoid fissure moderately deep. The
supraoccipital crest is short, and the orbits are placed well forward on
the skull. A parietal foramen has been reported in Copherus polyphe-
mus (Zangerl 1957). In this study it was found in seven percent of G.
polyphemus, and in five percent of G. agassizi specimens examined, but
was never found in G. berlandieri or G. flavomarginatus. The lower jaw
has lingual and lateral ridges, separated by the linear alveolar concavity.
Both ridges are interrupted dorsal to the symphysis by a shallow antero-
posterior groove. Eleven measurements (Fig. 1) were taken on the
skulls of 133 specimens of Gopherus (6 flavomarginatus, 22 agassizi, 38
berlandieri and 67 polyphemus).
SEXUAL VARIATION.-There are no significant differences between the
skull proportions of males and females in any of the Recent species of
Gopherus; however, G. berlandieri males have proportionally larger
heads than do females (Fig. 2).
ONTOGENETIC VARIATION.-The most obvious changes with growth
are associated with the proportionate growth of the pre-and postorbital
areas. The postorbital area exhibits considerably more growth than the
preorbital area (Fig. 3). There is no significant interspecific difference








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 53

in allometric skull growth, although the skulls of G. polyphemus and
G. flavomarginatus are considerably wider than those of G. agassizi and
G. berlandieri.
INTRASPECIFIC POPULATION COMPARISONs.-Only in G. polyphemus
is there sufficient material to make population comparisons. Two popu-
lations were compared, of which one comprised a sample of 44 specimens
(from north-central Florida and the east coast of the Florida peninsula)
and the other 23 specimens (from the Florida panhandle, southeastern
Georgia, southern Alabama, and Mississippi). Specimens in the western
sample have slightly wider heads than those in the eastern (peninsula)
sample, a situation already suggested by Legler (1959) (greatest width
of skull as percentage of condylobasilar length: western X=0.88,
eastern X 0.95; P=0.08). The alveolar angle is wider in the eastern
(X 82.5) than in the western (X 67.5) sample. When the series of
berlandieri is separated into two samples (one north of the Rio Grande
and another south of that river), the northern specimens seem to have a
wider interorbital area, although the difference is not statistically sig-
nificant (P = 0.42). It is reasonable to assume that when more specimens
are examined the character will be found to be geographically variable in
berlandieri, perhaps even clinal.
INTERSPECIFIC DIFFERENCES.-It is shown in Table 1 that many struc-
tural skull features are shared by G. polyphemus and G. flavomarginatus
on the one hand, and by G. agassizi and G. berlandieri on the other.
The most obvious of these is the proportionate width of the skull, as
measured at the widest point (across the postorbital bridge). Gopherus
polyphemus and G. flavomarginatus have wide skulls, whereas G. agassizi
and G. berlandieri have relatively narrow skulls. Mean proportional
measurements (greatest skull width/condibasilar length) of the four
species are as follows: polyphemus, 0.88; flavomarginatus, 0.82; agassizi,
0.75; and berlandieri, 0.76. The difference between the means of poly-
phemus and flavomarginatus is not significant (P=0.32), as is also true
in means of agassizi and berlandieri (P = 0.71). When the data for both
species in each group is grouped (polyphemus+flavomarginatus versus
agassizi+ berlandieri), however, the differences in combined means is
highly significant (P=0.01). There is no significant difference in the
interorbital width of these four species (Table 1).
Skull height is not significantly different in the four Recent species
(Table 1), although Mexican specimens of G. berlandieri have propor-
tionately higher skulls than those from Texas.
The postorbital bar shows considerable variation with regard to the
shape of its constituent elements. To a certain extent, some variational


















TABLE 1.-SKULL VARIATION IN Gopherus SPECIES.
Species
agassizi berlandieri flavomarginatus polyphemus
Ratios and Measures X OR X OR X OR X OR

1. prmx at symph.
condibasilar 1. 0.20 0.17-0.23 0.24 0.15-0.26 0.18 0.16-0.20 0.18 0.14-0.21 0
gr. skull width
condibasilar 1. 0.75 0.65-0.88 0.76 0.67-0.83 0.82 0.77-0.86 0.88 0.77-0.97 C
interorbital w.
condibasilar 1. 0.24 0.19-0.29 0.21 0.19-0.24 0.25 0.24-0.27 0.28 0.21-0.31
skull height
condibasilar 1. 0.43 0.36-0.48 0.45 0.37-0.59 0.48 0.42-0.54 0.46 0.41-0.49
alveolar angle 72 69-79 74 66-81 77 74-80 79 76-84


o



to








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 55

patterns are species characteristic (Fig. 4). The same is true of the
shape of the exocciptal bones and the position and size of their included
foramina (Fig. 4).
The width of the alveolar surface of G. polyphemus and G. flavomar-
ginatus is identical and is very wide when compared with that of the
other two species. The alveolar surface of berlandieri is the narrowest of
the four species, whereas agassizi is intermediate in this respect. Two
alveolar ridges occasionally occur as an individual variant in specimens
of all species. This somewhat primitive condition which seems most
common in flavomarginatus, may be seen in an illustration by Gray
(1873: pl. LX).
The outer alveolar angle of G. polyphemus and G. flavomarginatus
is wider than that of G. agassizi and G. berlandieri (Table 1 and Figs.
5-8): agassizi has the least angle, with berlandieri intermediate between
this species and polyphemus-flavomarginatus. The angle formed by the
outer surface of the alveolar surface is directly related to the width of
the skull. The above data are in agreement with those of other authors
and confirm the idea that the differences in skull measurements are of a
specific nature (Legler 1959; et al.).
Other more subjective characters are present that suggest the same
general relationship (polyphemus-flavomarginatus, agassizi-berlandieri).
These are (see also Figs. 5-8):
polyphemus-flavomarginatus agassizi-berlandieri group
group
1. Median longitudinal alve- 1. Longitudinal maxillary al-
olar maxillary ridge always veolar ridge frequently fails
crosses premaxilla to con- to cross premaxilla (especi-
tact median premaxillary ally in berlandieri).
ridge.
2. Interpterygoid width 2. Interpterygoid width less.
greater.
3. Basisphenoid shorter, an- 3. Basisphenoid usually long-
terior angle less acute (87- er, anterior angle usually
460). more acute (52-280).
4. Maxilla proportionately 4. Maxilla usually proportion-
lower. ately higher.
5. When viewed through na- 5. When viewed through na-
rial opening: rial opening:
(a) dorsal projection of (a) dorsal projection of vo-
vomer longer. mer generally shorter.








56 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

(b) anterior median (b) anterior median vomer
groove on vomer groove longer, running
shorter. more anteriorly.
(c) palatine foramina usu- (c) palatine foramina usu-
ally smaller, ally larger (especially
in berlandieri).
(d) dorsal, median fora- (d) dorsal, median fora-
men usually larger. men usually smaller
(especially in berlandi-
eri).
6. When otic area is viewed 6. When otic area is viewed
from side (Fig. 9): from side (Fig. 9):
(a) lateroposterior wall of (a) lateroposterior wall of
quadrate usually thin- quadrate usually thick-
ner. er (especially in ber-
landieri).
(b) Stapes in groove to (b) Stapes not in groove to
edge of tympanic ex- edge of tympanic exca-
cavation. ovation.
(c) Tympanic excavation (c) Inner wall of tympanic
with a rather smooth excavation usually fur-
inner wall. rowed and ridged.
7. Quadrate practically verti- 7. Quadrate usually angled
cal when seen from side. anteriorly (least in ber-
landieri).
8. Pro-otic usually well ex- 8. Pro-otic usually not ex-
posed anteriorly (Table 2). posed anteriorly (Table 2).
9. At symphysis of lower jaw: 9. At symphysis of lower jaw:
(a) Median symphyseal (a) Median symphyseal
length shorter, length longer.
(b) Dorsal symphyseal (b) Dorsal symphyseal
groove (to receive me- groove usually much
dian premaxillary wider (especially in
ridge) narrower, berlandieri).

With regard to the basis of the skull and lower jaw, G. agassizi is inter-
mediate in several important characters between G. berlandieri and G.
polyphemus-flavomarginatus. G. berlandieri is the most specialized in
many respects, although it is obviously closer to G. agassizi than to either
of the remaining species.









1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 57

TABLE 2.-CONDITION OF THE PRO-OTIC BONE IN Gopherus SPECIES.

Partly Completely
Exposed Concealed Concealed
agassizi 4 12 14
berlandieri 5 11 19
flavomarginatus 5 1 0
polyphemus 53 2 1

CERVICAL VERTEBRAE
In most living species of turtles the cervical vertebrae are quite vari-
able in a number of respects. Often this variation is intraspecific in
nature. It may include the number of vertebrae, length/width ratios,
development of neural spines, transverse processes and ventral keels,
number and degree of development of intervertebral ossicles, placement
and shape of the zygapophyses, and in the number and placement of con-
vexities and concavities forming the central joints. Of these, the last has
been most extensively studied (Williams 1950b). Williams has clearly
shown that the normal central-joint pattern (occipito-cervical joint ex-
cluded) is (2 (3 (4) 5) 6) 7 (8).1 This is believed to be the primitive
condition and was the only pattern he found in G. agassizi and G. poly-
phemus. Of 19 specimens of berlandieri, however, six had the pattern (2
(3) 4) 5) 6) 7 (8) and three had the pattern (2 (3) 4) 5) 6) 7) 8).
Both variations are specializations tending toward procoely. Additional
data obtained during the course of this study substantiate the major fea-
tures of the intraspecific differences in this genus suggested by Williams
(Table 3). It is significant that divergence from the normal pattern occurs
only in agassizi and berlandieri and is of a similar type. There is no
divergence from the normal pattern in polyphemus and flavomarginatus.
TABLE 3.-CERVICAL JOINT PATTERN VARIATION (IN ) WITHIN Gopherus SPECIES.

Patterns
4th joint 7th joint
Species Normal variation variation
agassizi 77.7 16.7 5.6
berlandieri 51.0 32.3 16.7
flavomarginatus 83.3 16.7 0.0
polyphemus 94.4 5.6 0.0

CAUDAL VERTEBRAE
The number of caudal elements in Gopherus vary from 12 to 19.
The first five to six elements are often incompletely ossified, so that the
1 ( = convexity anteriorly; ) = convexity posteriorly; and ( ) = convexity both an-
teriorly and posteriorly.








58 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

neural arch pedicles are not fused to the centra. The first three centra
are platycoelous; the rest are procoelous. At the anterior end of the
series the vertebrae are somewhat elongated and as high as wide. Poster-
iorly they become noticeably flattened and much wider than long. The
last two or three centra are often fused and may lack transverse proc-
esses. Pre- and postzygapophyses are well developed throughout the
series, with the exception of the last few fused members. The transverse
processes are rarely firmly ankylosed to the pedicle-centrum suture area.
In the most posterior 10 to 11 elements the transverse process is also in
contact with the posterolateral corner of the preceding centrum. Ad-
jacent transverse processes are often fused near the posterior end of the
series. An interpostzygapophyseal notch is always present. The trans-
verse processes are normal in length from the anterior part of the series
to near the point of the base of the external tail. Here the processes are
most elongated, decreasing rapidly in length to the terminus of the tail.
If the terminal tips of the transverse processes are considered as an out-
line of the bony skeleton of the tail, the overall shape is broadly spatu-
late. Subcaudal dermal ossicles are often fused to the underside of the
last 8 to 10 vertebrae. Differences observed in the caudal series of
Gopherus (Recent and Pleistocene only) and the closely related genus
Stylemys are as follows:
Gopherus Stylemys
Interpostzygapophyseal Interpostzygapophyseal
notches present throughout se- notches usually absent from
ries. most vertebrae, and always ab-
sent from some of the middle
members of the series.
Neural spine, boss, or keel ab- Neural spine, boss, or keel ab-
sent from all caudal vertebrae, sent from only the anterior
caudal vertebrae.
Post- and prezygapophyseal Post- and prezygapophyseal ar-
articular surfaces normal, sepa- ticular surfaces transversely
rated, distinct, elongate, often continuous.
Unfortunately, the tail tip in Stylemys remains unknown. I have
found no significant interspecific differences among the extant popula-
tions of the genus Gopherus.
HUMERUS
The humerus of Gopherus is like that of most other testudinid turtles.
Proximal condyle round, rarely slightly compressed vertically; shaft
slightly compressed, noticeably bent in the middle. Distal end com-
pressed anterolaterally, broadly widened laterally. Medial proximal tu-








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 59

berosity larger than in most testudinines; not greatly projecting medially
and higher than lateral process. Lateral process directed lateroposteri-
orly, with the angle between it and the medial tuberosity small. Posterior
intertubercular excavation completely open and well developed. No
humeral depression for attachment of the deltoid muscle, which inserts
on a low ridge below and slightly lateral to the condyle. Entepicondylar
foramen slit-like, sometimes obliterated. Radial and ulnar condyles well
developed, but not clearly differentiated from each other.
The extant species do not differ greatly in humeral proportions
(Table 4), although the humeral shaft is slightly thicker in G. flavomar-
ginatus than in the remaining species, and the distal expansion is pro-
portionately narrower. The distal width of the humerus is identical in
G. agassizi and G. berlandieri, although agassizi has a slightly wider
proximal width.

TABLE 4.-HUMERAL PROPORTIONS IN Gopherus SPECIES.

Greatest Proximal Width/Least Shaft Width:
agassizi OR 2.4-3.5 X 3.0 SD0.22; N 20
berlandieri OR 2.4-3.5 X 2.9 SD0.32; N 35
flavomarginatus OR 2.5-3.0 X 2.7 SD 0.43; N 13
polyphemus OR 2.6-3.8 X 3.1 SD0.24; N 28
Greatest Distal Width/Least Shaft Width:
agassizi OR 2.5-3.8 X 3.1 SD 0.31; N 20
berlandieri OR 2.3-3.8 X 3.1 SD 0.33; N 35
flavomarginatus OR 2.6-3.3 X 2.9 SD 0.36; N 13
polyphemus OR 2.8-3.8 X 3.3 SD 0.28; N 28
Greatest Proximal Width/Greatest Distal Width:
agassizi OR 0.80-1.05 X 0.98 SD 0.42; N 20
berlandieri OR 0.80-1.02 X 0.93 SD 0.38; N 35
flavomarginatus OR 0.78-1.01 X 0.91 SD 0.32; N 13
polyphemus OR 0.86-1.21 X 0.98 SD 0.61; N 28

PELVIC GIRDLE
The species of living Gopherus differ little in respect to their pelvic
girdle proportions. In all of them the obturator foramen is large, and
usually more oval in G. polyphemus and G. flavomarginatus than in the
other two species. The epipubic area is well developed anteriorly, being
widest in polyphemus, narrowest in agassizi and berlandieri, and inter-
mediate in flavomarginatus (Figs. 10-13). The pubic process is not as
well developed as in most turtles, and typically is variable in its length
and orientation. In all species, except flavomarginatus, it is located ap-
proximately halfway between the acetabulum and the anteromedial
corner of the pubis. In flavomarginatus the process is located much








60 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

closer to the epipubic process. The remainder of the girdle is similar
in all four recent species.
FEMUR
The femur of testudinid turtles is distinguished from that of emydid
turtles by a ridge that runs nearly on the level of the proximal condyle
and connects the medial and lateral trochanters. Between this ridge
and the condyle there is a pit, the fossa intertrochanterica. The proximal
condyle is set at an oblique angle, is oval in outline, and is more com-
pressed in G. agassizi and G. berlandieri than in the two remaining
species (Figs. 10-13). The shaft is slightly curved and not compressed.
It is proportionately widest in flavomarginatus. The distal end is greatly
expanded laterally and compressed anteroposteriorly. In adults the distal
condyles are well developed, separated by an intertubercular sulcus.
They are most clearly differentiated in flavomarginatus and polyphemus
and weakly developed in agassizi and berlandieri.
CARPUS
Primitively the turtle carpus contained 10 elements (radius, inter-
medium, ulnare, central, pisiform, and five carpals). Apparently the
central was not in contact with either the radius or the ulna. In pre-
sumably primitive tortoises, such as Geochelone denticulata, the carpus
is composed of 10 elements, at least in the embryonic stage (Hoffman
1890). In this species the central has moved distally, so that it contacts
the radius; the pisiform is small and in contact with the ulna and carpal
5, instead of only the latter. This pattern is apparently basic in land
tortoises. Its modifications include fusion or loss of elements, both of
which are individually and ontogenetically variable.
In cleared and stained forelimbs of very young Gopherus, I have
never found more than nine carpal elements. The pisiform is apparently
lost. Furthermore, there is considerably more fusion of the elements in
adult Gopherus than in any other genus I have studied (Auffenberg
1966a). This is probably associated with the use of front limbs for ex-
cavating burrows and shelters.
An examination of many articulated feet of the species of this genus
suggests that, although there is considerable variation in the degree of
fusion of the component distal elements, two basic patterns are discern-
ible (Auffenberg 1966a). One is found in the G. polyphemus-flavo-
marginatus group, where six carpal elements are in contact with the ulna
and radius. In the G. berlandieri-agassizi group only four or five ele-
ments are in contact with the two brachial elements.
The carpus of Gopherus has been shown to represent a primitive type
among tortoises (Auffenberg 1966a). It is distinguished by two features:








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 61

a separate proximal central, at least in the juveniles; and the position
and shape of carpal 4, which completely separated the ulnare from
medial 3.
Carpi were examined in 32 specimens of G. polyphemus, 7 of G.
flavomarginatus, 31 of G. agassizi, and 37 of G. berlandieri with the fol-
lowing results:
Gopherus polyphemus-Except for a slight difference in degree of
phalangeal fusion in digit 5, there is little variation correlated with tor-
toise size. The proximal central is present in all specimens. Carpal 1
is fairly large and articulates broadly with the radius, the radius being
bordered below by four elements (carpal 1, radiale + mediale 2, proximal
central, and intermedium). There are no intercarpal fusions and the
metacarpals are typically fused to adjacent phalanges (14 of 32 speci-
mens in all 5 digits, 18 in all digits except number 5). Carpal 4 is
wedged between the ulnare and mediale 3, but not so extensively as in
G. agassizi and G. berlandieri. Carpal 5, the ulnare, and intermedium
are all in broad contact with the ulna. The pisiform is absent in all but
three specimens.
Gopherus flavomarginatus-The carpus in this species is similar to
that of G, polyphemus, with four subradial elements (carpal 1, radiale+
mediale 2, proximal central, and intermedium), and three subulnar ele-
ments intermediumm, ulnare and carpal 5). The metacarpals are fused to
adjacent phalanges.
Gopherus agassizi-The arrangements of elements in G. agassizi is
different from those of G. polyphemus and G. flavomarginatus and similar
to that of G. berlandieri. The proximal central sometimes is a separate
element, as in G. polyphemus; this was observed in four of eight speci-
mens smaller than 150 mm shell length and in 11 of 23 larger specimens.
In those smaller specimens lacking a separate proximal central, the bone
is fused to the intermedium. In all 11 larger specimens having a separate
proximal central, carpal 1 is fused to the composite radiale, whereas in
10 of the other 12 specimens (i.e. in which the proximal central was
fused to the intermedium), carpal 1 is free in 21 of the 23 specimens
over 150 mm, although in some the fusion is not complete. No such
fusion can be seen in six of eight specimens less than 150 mm in shell
length. The radiale and medialia 2 and 3 are solidly fused together in
11 specimens over 150 mm, and in all individuals carpals 1 and 2 are
also fused into this mass. Carpals 1 and 2 are fused in 21 of the speci-
mens over 150 mm. All the metacarpals are fused to the adjacent pha-
langes in all specimens examined, regardless of size. Carpal 5 is fused
to both the pisiform and ulnare in 13 of the 23 specimens over 150 mm;
in every larger individual in which carpal 5 is free, the pisiform is fused








62 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

to the ulnare. The pisiform is free in all the smaller specimens examined.
These data show that the carpus of G. agassizi is more variable than
that of G. polyphemus, particularly in the lateral elements of the sub-
radial and subulnar complexes.
Gopherus berlandieri-The lack of a separate proximal central, plus
several other anatomical features, suggest that G. berlandieri is closer to
G. agassizi than to the other two extant species. In six specimens under
100 mm in shell length, mediale 2 is not fused to the composite radiale,
as it is in all specimens over 100 mm in shell length. In no specimen
is the proximal central separate from the intermedium. Carpal 1 is
separate and in contact with the radius in all specimens and is never
fused to the composite radiale. Metacarpals 3 and 4 are always fused
to the adjacent phalanges, but metacarpals 1, 2, and 5 are fused to the
adjacent phalanges in only the larger specimens. The pisiform is al-
ways present. Carpal 5 is fused to the ulnare in 22 of 41 individuals
over 100 mm, and carpal 4 separates mediale 3 and the ulnare in all
specimens. The radiale is fused to mediale 2 in all individuals, and these
are fused to mediale 3 in all adult specimens.
The major difference between the carpus of the polyphemus and
berlandieri groups seems to lie in the position of the central. In the
former it is in its primitive tortoise position, i.e. in contact with the radius.
In the berlandieri group it is excluded from contact with the radius by
the intermedium (Fig. 14). In the polyphemus group loss and fusion of
carpals and metacarpals occur along the axis of digits II, III, IV and V.
When fusion occurs in the berlandieri group, it does so along the axis of
digits I, IV and V (Table 5).
TARSUS
The tarsus of tortoises is evidently more conservative than the carpus.
In Gopherus I found no significant interspecific differences in the tarsus.
The tibiale, intermedium, and central are always completely fused in
adults, and the fibulaire is usually separate. In most other tortoises this
element is also fused to the large composite element. The tarsals are
either four or five in number; if the former, then tarsals 4 and 5 are fused,
rather than 3 and 4 as in most turtles (Hoffman 1890). All five meta-
tarsals are distinct, with number 1 being the most robust, as is generally
true in all turtles. Metatarsals IV and V articulate with separate tarsals,
or with only one if tarsals 4 and 5 are fused. Two phalanges are always
found on digits I through IV. The phalanges of digit V are often mis-
sing, although one is sometimes present (Fig. 14).
DERMAL ARMOR
Dermal ossicles in Copherus are mainly developed in three areas:

















TABLE 5.-CARPAL AND METACARPAL FUSION AND/OR LOSS PATTERNS IN Gopherus.

berlandieri-agassizi polyphemus-flavomarginatus

axis of digit I usually all present, but C' frequently fuses with M1 all present
axis of digit II all present, though C2 may fuse to M1 and C3, all present, though the radiale may be fused to C
never to the radiale and C3 in adults.
axis of digit III Centrale and intermedium always fused. C3 may Ca in contact with central and the intermedium 0
be fused to C2 (C3 may be fused to radiale and (fused to these and the radiale in large adults).
intermedium in adults)
axis of digit IV C4 missing (if ulnare fused, then to C5) C4 missing (ulnare fused to intermedium in adults) t
axis of digit V C5 present in young, fused to ulnare in adults Cs missing in young and adults

0
03








64 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

the posterior surface of the thighs, the hind foot, and the forearm. These
ossicles, which are covered by horny scales, vary ontogenetically, sex-
ually, individually, interspecifically, and geographically; the most obvious
variation is interspecific. The anterior forelimb ossicles of both G. poly-
phemus and G. flavomarginatus are usually flattened, and though variable
in shape, are usually hexagonal or pentagonal (Fig. 16). Those of adult
flavomarginatus are often very thick and sometimes partially fused. A
single scale on the anteromedial surface opposite the elbow is noticeably
enlarged, particularly in polyphemus and flavomarginatus. This scale
may be important in courtship (Auffenberg 1966b). The scales at the
outer edge of the forearm are usually longer and more spine-like in all
the species. In G. agassizi and G. berlandieri the ossicles of the anterior
surface of the forearm are more imbricate and are often keeled and spine-
like. G. agassizi is much more variable than G. berlandieri in this char-
acter, and the variation may be geographically correlated. The material
available is insufficient to prove this, however.
The dermal ossicles of the thigh are usually best developed in large
males. They are often arranged in a circular patch, with one or two
central members largest. They are best developed in G. flavomarginatus,
where they are cone-like in adults. In G. polyphemus they are often
completely flattened.
The dermal ossicles of the hind foot are best developed on the pos-
terior surface of the foot at the heel. Males tend to have larger heel
spurs than females. They are best developed in G. flavomarginatus.

RADIUS AND ULNA
The radius is always the larger of the two elements. Its curvature
and proportions are slightly different in G. polyphemus and G. agassizi,
although there is considerable overlap and single elements of the two
species usually cannot be separated. In general, the radii in both ber-
landieri and polyphemus are flatter and less curved than that of agassizi
(Figs. 10-13). G. flavomarginatus has a radius similar to that of poly-
phemus. In large adults it is often possible to discern the number of
carpal elements that had articulated with the distal end of the radius.
Four facets are usually evident in polyphemus and flavomarginatus, and
two or three in berlandieri and agassizi. This is of particular importance
in the study of fossil radii.
The ulna in all land tortoise genera is considerably shorter than the
radius. The ulna is very similar in all the extant species but occasionally
is more twisted in berlandieri and agassizi than in polyphemus and flavo-
marginatus.








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 65

SHELL
At hatching the young gopher tortoise possesses only the most rudi-
mentary elements of the bony shell (Fig. 18). These are the vertebrae,
the ribs (which extend to the edge of the costal scutes), and the small
centers of ossification at the posterior lower corner of each of the anterior
and posterior marginal scutes. The plastral buttresses are long and pro-
vide maximum support until the shell is completely ossified. The axil-
lary buttress extends to near the junction of the lateral anterior corer
of vertebral scute 1 and costal scute 1. The posterior inguinal buttress
lies just anterior to the sixth rib, and extends approximately one-fourth
of the way medially. Later in development the ends of the buttresses
come to be located closer to the distal rib ends (Fig. 18). All of the
plastral bones, except for the entoplastron, are more heavily ossified than
those of the carapace. The epiplastral lip is not greatly projected, and
there is no deep excavation behind.
With increasing age and progressive ossification of the shell, the
shallow grooves on the dorsal edge of the peripherals (in which distal
ends of the rib were originally located) become modified to shallow pits
(Fig. 18). In addition, there are important changes in the hypoplastral-
xiphiplastral articulation.
In general, ontogenetic change in shell shape (viewed from above)
is from a more rounded to a more elongate outline. This is much more
evident in polyphemus than in berlandieri or agassizi (Fig. 19). I have
not seen any hatchlings of flavomarginatus.
The general shell characteristics of Late Tertiary, Pleistocene, and
Recent adult specimens of Gopherus are: shell of moderate height, with
moderate anterior and posterior openings; plastron extending beyond an-
terior margin of carapace; anterior edge of carapace transverse to con-
cave; marginals with entire edges, not deeply serrated but often with
small pointed projections at scute sulci; no hinge in either carapace or
plastron; width of vertebral scutes greater than length and about equal to
height of costal scutes, which in turn are about equal to bridge marginals;
last vertebral scute almost always as wide as caudal scute and two last
marginals; bridge moderately long, without a well developed continuous
keel; bones of shell thin; usually two suprapygals, the anterior one either
A-shaped and embracing the posterior one, or placed on top of the other;
one supracaudal scute, which is wider than the pygal bone; 11 marginal
scutes; 11 peripheral bones; pleurals alternately wider and narrower
medially and laterally; neural formula usually 4-8-4-8-4-(6-(6-(6, with
the vertebral sulci usually crossing neurals 1, 3, 5 and 8; nuchal scute
usually as wide or wider than long; rib heads slender, but relatively








66 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

short; epiplastral lip projecting, frequently with notches anteriorly and
deeply excavated behind; inguinal and axillary scales present; inguinal
buttress contacting underside of pleural 6 near distal end of rib (in
adults); hyoplastral median length divided by hypoplastral median
length, almost always 1.0 or more (Table 6).
TABLE 6.-HYO-HYPOPLASTRAL RATIOS IN Gopherus SPECIES.

Species Hyo-Hypoplastral
agassizi = 1.24 SD = + 0.07
berlandieri X= 1.24 SD= + 0.03
flavomarginatus X = 1.36 SD = 0.58
polyphemus X= 1.16 SD = 0.02

Adult specimens show many sexual differences in the shell, most of
which will be discussed below. The most noticeable is the shape of the
plastral surface. In adult males there is a fairly large, slightly concave
area centered near the xiphi-hypoplastral suture (least developed in
polyphemus, best developed in berlandieri). When it occurs, this dif-
ference can usually be detected in specimens with a shell length of 150
mm or more. In females and young the plastron is flat.
SCUTE AND BONE ABNORMALITIES.-Sufficient skeletal material of the
Recent species of Gopherus is available to establish the occurrence of at
least the more common abnormalities of the shell. These are much more
common than generally presumed, only 25 percent of all Recent Gopherus
examined being free of any abnormalities.
Fourteen different anomalies are noted, seven in the scutes and seven
in the bony plates (Fig. 20). Abnormalities in the bony plates are far
more common than in the scutes. Scute abnormalities of several types
occur in six different scutes: abdominal (partial and complete division),
nuchal (deletion or addition of scutes), supracaudal (vertically divided),
gular (divided, or failing to reach the entoplastron), vertebral (divided
transversely or diagonally), and costal (additional scute, or divided
transversely).
Abnormalities in the bony plates include various additions or deletions
in the area of the last few neurals, the first supracaudal, second supra-
caudal; and additional elements at the proximal end of any of the last
few pleurals, the first pair of pleurals, or the first neural (Fig. 20).
Anomalies in the area of the first and second suprapygals are most com-
mon (Table 7), accounting for 57 percent of all the anomalous speci-
mens.
The Recent species of Gopherus differ from one another with regard
to percentage of abnormal specimens recorded and types of abnormali-









1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 67

TABLE 7.-SHELL ABNORMALITIES IN Gopherus SPECIES.

Abnormalities agassizi berlandieri flavomarginatus polyphemus Totals
BONES
3 suprapygals 6 1 0 1 8
Suprapygal 1 not
embracing 2 20 6 0 10 36
Suprapygals
1+2 fused 1 0 0 0 1
Divided suprapygal 2 0 1 0 0 1
Divided last neural 0 1 0 1 2
Divided first neural 2 0 0 1 3
Divided pleurals 1 0 0 0 1

SCUTES
No nuchal 0 3 0 0 3
Extra nuchals 0 0 2 0 2
Extra or fused
marginals 11 6 1 4 22
Divided supracaudal 0 2 0 0 2
Extra abdominals 2 0 0 0 2
Divided vertebrals 1 0 0 0 1
gulo-pectoral sulcus
fails to enter
entoplastron 11 4 0 1 15
Extra or
fused costals 4 1 0 0 5
Total Normal 4 11 1 13
Total Abnormal 31 24 3 25

ties found. At least one abnormality occurs in either the scutes or bones of
the following percentages of specimens examined: G. polyphemus 52%;
G. agassizi 80%; berlandieri 72%; and flavomarginatus 66% (Table 8).
The different abnormalities are not completely independent of one an-
other; for example, in G. agassizi there is a significant correlation be-
tween the failure of the gular scute to overlap the entoplastron and the
presence of one of the other abnormalities (chi square= 0.02).

NON-SPECIES CORRELATED QUALITATIVE VARIATION.-The neural form-
ula (the sequence of quadragonal, hexagonal, or octagonal members in
the neural series) is one of the most characteristic features of the family
Testudinidae. Especially important is the fact that the series contains
octagonal members that alternate with quadragonal members. The
formula shows greater variation in Gopherus than in other genera of the
family studied so far, and this variation is both bilateral and serial.
The most important of the serial variations is that existing between
the anterior and posterior members of the series. Of these two areas,
only the anterior is characteristically composed of elements that are al-








68 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

ternately quadrangular and octagonal. The posterior members are
normally 4- or 6-sided; if the latter, the broad end of the hexagon faces
anteriorly. All the neurals are generally hexagonal, with the broad end
facing anteriorly in the emydid turtles. This arrangement is thought
to represent the ancestral condition.
TABLE 8.-CORRELATED SHELL ABNORMALITIES IN Gopherus SPECIES.

Neural Pleural Epiplastral Plastral
or or or or
Vertebral Costal Gular Abdominal
agassizi
Neural or Vertebral 2 1 7 1
Pleural or Costal 0 1 0
Epiplastral or Gular 0 1
Plastral or Abdominal 0
berlandieri
Neural or Vertebral 3 0 2 0
Pleural or Costal 0 0 0
Epiplastral or Gular 0 0
Plastral or Abdominal 0
polyphemus
Neural or Vertebral 2 0 0 0
Pleural or Costal 0 0 0
Epiplastral or Gular 0 0
Plastral or Abdominal -- 0

During the early phases of carapaceal ossification a single pair of
ribbon-like pleurals is associated with each of the still undifferentiated,
plate-like neurals. After this developmental stage is reached, the growth
pattern of the proximal end of the pleural is exceedingly important in
the final configuration of the neural (Fig. 21). Thus, in emydids and
primitive testudinids the growth pattern of the proximal end of the
pleural is such that posteriorly it grows more rapidly than it does dorso-
anteriorly. This produces the neural formula 4-(6-(6-(6-(6-(6-(6, with
the broad ends of the hexagons directed anteriorly (direction shown by
"(") (Fig. 21). If the growth pattern of the proximal pleural end is
such that the anterior edge grows faster than the posterior edge, the
neural formula eventually produced is 4-6)6)6)6)6)6); if lateral growth
in both directions is equal, the neural formula will be 4-8-4-8-4-8-4 (the
typical, more specialized tortoise pattern); if every other pleural grows
more posteriorly than anteriorly, the formula will be 4-6)4-6)4-6)4; and
if every other pleural grows more anteriorly than posteriorly, the formula
will be 4-(6-4(6-4-(6-4. All these patterns can be found in Gopherus,
often in serial or bilateral combination in one specimen. There is a








"0-4
Q0





TABLE 9.--VARIATION IN THE SHAPE OF THE FIRST FIVE NEURALS IN THREE SPECIES OF Gopherus.1

berlandieri agassizi polyphemus
(D) 4-8-4-8-4 (70%) (D) 4-8-4-8-4 (50%) (D) 4-8-4-8-4 (47%)
(%D) 4-7-4-7-5 (5%) (%D+D) 4-7-5-8-4 (3%%) (D+%E) 4-8-4-8-5 (8%) P
(D+E) 4-8-4-8(6 (10%) (D+C) 4-8-4-6)6) (3-'1%) (D+ 2B) 4-8-5-5-4 (8%)
(D+E) 4-8-4(6-4 (5%) (D+%D) 4-8-5-7-4 (3%%) (YD) 5-7-4-7-4 (8%)
(B+%D+D)(6-7-4-8-4 (5%) (D+'/C) 4-8-5-5-4 (3%%) (D+E) 4-8-4-8(6 (13%)
(%B-B+ D) 4-5(6-8-4 (5%) (D+C+%zC) 4-8-6-5-5 (3y%) (D+F) 4-8-6-4-6) (8%)
(D+F+%E) 4-8-6-4-5 (3Y2%) (D+A+F) 4-8-4-4-6) (8%)
(B+D) 4(6(6-4-8 (3-Y2%) O
(C+D) 6)6)6)8-4 (3-%2%)
(C+% C+E) 6)5-6)6)4 (3-1/2%)
(D+E) 4-8-4(6-4 (7%)
(yzC+C+%C+C) 5-6)6)5-6) (3-%%)
(B+%B) 4(6(6-5(6 (3-%%)
(B+F) 4(6(6-4(6 (3-/2%)
O
A each pleural grows laterally at same rate (4-4-4-4-4); B = each pleural grows anteriorly (4(6(6(6(6); C = each pleural grows posteriorly (4-6)6)6)6); D= alternate pleurals grow laterally at same rate
(4-8-4-8-4); E= alternate pleurals grow anteriorly (4(6-4(6-4); F= alternate pleurals grow posteriorly (4-6)4-6)4); 1/2 = suture characteristic on one side only. For additional explanation see text. 0


0


0'








70 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

definite tendency for the anterior neurals to show greater differentiation
than the posterior members, and thus the familial neural characteristics
of the Testudinidae are usually restricted to the first few neurals. How-
ever, in Gopherus even the anterior neurals show considerable variation
in growth pattern (Table 9), although some conceivable combinations
of growth patterns have not yet been found. There is no significant
difference between the neural growth patterns exhibited in the four
species of Gopherus (Table 10). It can be seen in Figure 22 that the
pleural growth pattern (in which there is alternate equal growth of the
proximal ends of the pleurals) is the most common type when each
neural is taken as an independently developed unit.
TABLE 10.-PLEURAL GROWTH PATTERNS IN SPECIES OF Gopherus INFERRED FROM
NEURAL PATTERNS.

Species Pleural Growth Patterns
A B C D E F
agassizi 0 3 6 22 2 2
berlandieri 0 2 0 28 3 0
flavomarginatus 0 0 0 3 1 0
polyphemus 1 0 0 15 3 1

It has been said that in all New World tortoises, including Gopherus,
the usual suprapygal condition is one in which the first suprapygal em-
braces the second. My findings do not entirely agree, since this condi-
tion was not found in about 39% (71) of the 183 Gopherus specimens
I examined (polyphemus 28 [44%], agassizi 26 [62%], berlandieri 15
[28% ], flavomarginatus 2 [8% ]). It is clear that in Gopherus this varia-
tion is the direct result of a rather simple developmental phenomenon.
In specimens with a carapace length of about 150 mm, in which the
suprapygals are not yet fully ossified, there is a fenestrum on either side
of suprapygal 2. These fenestrae eventually become filled with bone
originating from only one of two centers of ossification (i.e. either one
or the other of the two suprapygals). Should the ventro-lateral corners
of the first suprapygal grow downward to fill the fenestrae, an embrasure
will be formed. However, if during ossification the fenestrae are filled
from the second, lower center of ossification, the first suprapygal will not
embrace the second (Fig. 23).
All emydids and some of the testudinids (Testudo, etc.) possess the
non-embracing type of pattern (Loveridge and Williams 1957). The
earliest known fossil testudinids (Geochelone [subgenus Hadrianus] and
Stylemys) possess the embrasure. So do almost all the available Tertiary
specimens of Gopherus. Thus, the lack of an embrasure in Recent
Gopherus is presumed to be secondary, and due to a simple change in








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 71

the pattern of ossification. This pattern is coincidentally identical to
the more primitive, pre-testudinine pattern because only two major types
are possible. Fenestral ossification is usually bilaterally correlated, so
that if one center of ossification fills the fenestra on one side, the same
center tends to do so on the other side as well. In some specimens the
right and left sides do not develop identically (Fig. 24).

QUANTITATIVE SHELL VARIABLES.-The purpose of this portion of the
study was to determine those combinations of characters that would best
separate the adult specimens of the four species of Gopherus, and to
establish how widely they are separated. The sample analyzed con-
sisted of 183 adult Gopherus, including an approximately equal number
of males and females of each species. From each of the shells available
for study 49 measurements were obtained, 25 on the plastron (Fig. 25)
and 24 on the carapace (Fig. 26). To simplify computation and com-
parison, these two parts of the shell were analyzed independently. Other
variables considered early in the study were sex and population (=
species). Accumulated data are provided in Table 11.
Relative variability of the characters investigated was first established
by analysis of the coefficients of variation (=CV) exhibited by the
measurements. These are summarized as follows:

PLASTRON
1) Different parts of the plastron are differentially variable (Range CV=11.13 to
57.54; CV= 21.31).
2) In different species the same characters do not vary in the same direction, or to
the same degree.
3) Variation of scute length along median line (CV=22.68) does not differ signifi-
canty from variation in bone length along median line (CV = 23.80) (Fig. 27).
4) Plastral variation in males is not significantly different from that of females
(males CV=21.19, females CV=22.45) (Table 12). Within the species poly-
phemus, females are significantly more variable than males (males CV=18.52,
females CV= 26.80), but sexual variation is not significantly different in the three
remaining secies (agassizi males CV= 22.60, females CV= 22.70; flavomargina-
tus males CV= 23.15, females CV= 24.05; berlandieri males CV= 21.81, females
CV= 18.79).
5) None of the species differ significantly from one another in total plastral variability
(CV for all plastral measurements in polyphemus = 22.76, flavomarginatus = 25.06,
agassizi= 23.79, berlandieri= 26.26) (Table 12).

CARAPACE
1) Different parts of the carapace are differentially variable (Range CV=2.00-
162.94, CV= 29.80).
2) Variation in different parts of the carapace are species-dependent.
3) Variation of carapace scutes (CV= 38.36) is not significantly different from varia-
tion of carapace bones (CV= 30.87) (Table 12).
4) Males (CV= 29.01) are significantly less variable than females (CV=36.63)
when all species are combined. There is no significant difference in the mean
variance of males (CV=29.80) and females (CV=32.27) in berlandieri. How-








72 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2




TABLE 11.-SHELL STATISTICS, Copherus agassizi.

Measure- Males (N=11) Females (N=13) Combined (N=24)
ments' XRSD CV X+-SD CV X+SD CV
PL 191.2039.36 20.59 175.76 32.46 18.47 181.48 35.24 19.42
G 31.20+ 8.31 26.63 25.59 5.33 20.83 27.67 7.00 25.30
H 50.90+ 10.16 19.96 42.82 + 9.28 21.67 45.81 + 10.22 22.31
P 15.20+ 3.65 24.01 14.47+ 4.35 30.06 14.74 4.04 27.41
AB 58.1010.01 17.23 56.1810.40 18.51 56.89_+10.10 17.75
F 27.00 5.98 22.15 25.35 5.17 20.39 25.96+ 5.43 20.92
A 17.80+ 3.52 19.78 17.88+ 3.97 22.20 17.85 3.74 20.95
ND 12.20+ 4.32 35.41 8.06+ 2.51 31.14 9.59 3.81 39.73
NW 22.70 5.52 24.32 17.82 + 6.50 36.48 19.63 6.51 33.16
XH 9.60 1.96 20.42 9.06 2.33 25.72 9.26 2.18 23.54
XL 41.50+ 6.82 16.43 38.06 6.92 18.18 39.33+ 6.96 17.70
XW 46.80+ 9.59 20.49 44.41+ 8.47 19.07 45.30 4.15 19.43
POL 56.1012.91 23.01 50.2410.40 20.70 52.4111.51 21.96
HYP 38.60+ 6.60 17.10 38.94 7.03 18.05 38.81- 6.75 17.39
HYO 48.00+ 9.94 20.71 43.53 8.65 19.87 45.19+ 9.23 20.43
EL 40.00 7.72 19.30 36.94 + 7.77 21.03 38.07- 7.75 20.36
EW 38.80 9.76 25.16 35.82+ 8.41 23.48 36.93 8.87 24.02
ME 28.30 10.12 35.76 20.24 4.22 20.85 23.22 7.89 33.97
GE 48.70 13.39 27.50 38.82 + 8.74 22.51 42.48 11.52 27.12
AL 72.9018.46 25.32 58.29 11.20 19.21 63.7015.71 24.66
LT 13.10+ 2.13 16.25 11.71+ 2.54 21.69 12.22 + 2.45 20.05
LL 28.60+ 6.90 24.13 23.53 6.16 26.18 25.41+ 6.78 26.68
CL 208.1041.02 19.71 188.0041.24 21.94 195.44 +41.56 21.27
CW 157.00 30.24 19.26 171.3542.95 54.25 166.04- 75.38 45.40
CH 92.80 17.78 19.16 96.00 56.28 58.63 94.81 +45.40 47.88
V2L 42.60 6.08 14.27 44.65 14.85 33.28 43.89 12.23 27.87
V2W 60.00 10.35 17.25 54.29 10.50 19.34 56.41 10.62 18.82
V4L 47.00 10.03 21.34 42.00+ 9.30 22.14 43.85+ 9.70 22.12
V4W 58.60 11.52 19.66 56.00+ 11.81 21.07 57.00+ 11.55 20.26
DPW 35.20 8.36 23.75 30.82 6.77 21.97 32.44-+ 7.55 23.27
VPW 18.60+ 4.60 24.73 11.59+ 5.15 44.44 14.19_+ 5.96 42.00
LPY 27.00 6.78 25.11 21.53 + 6.58 30.56 23.56 + 7.06 29.97
AW 16.60+ 3.63 21.87 16.71+ 5.14 30.76 16.67+ 4.57 27.41
VW 32.50 11.37 34.98 35.12 12.67 36.08 34.15+ 12.05 35.29
S1L 29.30 + 7.90 26.96 29.65 15.01 50.62 29.52 12.66 42.89
S2L 19.40 7.11 36.65 15.76+ 6.45 40.93 17.11+ 6.80 39.74
S2W 35.80+12.65 35.34 34.18+ 17.67 51.70 34.78+ 15.75 45.29
ANBW 36.60 9.99 27.30 36.94 15.54 42.07 36.81+ 13.54 36.78
NBW 49.70 9.36 18.83 44.24 + 9.08 20.52 46.26 + 9.40 20.32
NBL 37.00+ 8.45 22.84 32.88 12.74 38.75 34.41+11.35 32.98
NL 17.70 28.84 162.94 10.71 + 10.27 95.89 13.30 19.09 143.53
N 10.90_+ 5.04 46.24 15.88 +17.12 107.81 14.04 13.97 99.50
PP3 38.60 5.25 18.36 26.41 5.30 20.06 27.22+ 5.29 19.43
DP, 9.00 4.69 52.11 12.76 13.09 102.59 11.37+ 10.80 94.99
LP, 64.8011.01 16.99 58.06 -13.89 23.92 60.56 13.10 21.63
See Figs. 25-26 for abbreviations.









1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 73




TABLE 11 (CONTINUED). SHELL STATISTICS, Gopherus berlandieri.

Measure- Males (N=17) Females (N=25) Combined (N=42)
ments1 XSD CV X+SD CV X+SD CV
PL 147.7524.05 16.28 112.00 16.72 14.93 132.4327.54 20.30
G 28.92+- 6.11 21.13 20.89 3.06 14.65 25.48+ 6.39 25.08
H 32.00 6.62 20.69 23.33 3.20 13.72 28.29 6.89 24.36
P 12.50- 3.42 27.36 12.11 3.02 24.94 12.33 3.18 25.79
AB 45.83 9.02 19.68 35.56 6.39 17.97 41.43 9.39 22.67
F 22.17+ 4.02 18.13 18.44- 4.00 21.69 20.57 4.34 21.10
A 13.33 3.47 12.00 9.00 1.73 19.22 11.48 3.56 31.01
ND 11.00 3.77 34.27 6.67- 1.73 25.94 9.14 3.72 40.70
NW 19.17 4.93 25.72 12.33 3.35 27.17 16.24 5.47 33.68
XH 11.50+ 2.91 25.30 6.44 1.74 27.02 9.33 3.53 37.83
XL 32.25 4.49 13.92 25.56 5.22 20.42 29.38- 5.79 19.71
XW 38.42+ 5.58 14.52 29.22 + 5.29 18.10 34.48 7.08 20.53
POL 38.67 14.28 36.93 32.56+ 4.69 14.34 36.05 11.43 31.71
HYP 29.83+ 5.72 19.18 23.33 + 2.92 12.52 27.05+ 5.68 21.00
HYO 34.17- 6.06 17.73 29.11+ 6.53 22.43 32.00 6.62 20.69
EL 32.92 6.44 19.56 25.44+ 3.43 13.48 29.71+ 6.47 21.78
EW 29.67+ 6.05 20.39 22.44+ 4.69 20.40 26.57+ 6.51 24.50
ME 21.33 7.10 33.29 13.00+ 2.35 18.08 17.76 6.91 39.91
GE 36.33 7.16 19.70 25.44+ 4.50 17.69 31.67 8.18 25.83
AL 53.33+ 10.21 19.14 36.00+ 5.74 15.95 45.90+12.16 26.49
LT 14.50+ 2.61 18.00 10.11+ 2.09 20.67 12.62 3.23 25.59
LL 23.33 4.81 20.62 15.89+ 2.20 13.85 20.14+ 5.38 26.71
CL 152.33 23.74 15.58 120.22+ 17.48 14.54 138.5726.41 19.06
CW 134.33 17.06 12.70 103.00+ 16.42 15.94 120.9022.81 18.87
CH 84.42 + 16.87 2.00 62.67 12.09 1.92 75.10 18.35 24.43
V2L 32.83 4.15 12.64 26.89+ 2.98 11.08 30.29+ 4.70 15.52
V2W 51.33 7.40 14.42 40.33 + 4.53 11.23 46.62 8.33 17.68
V4L 39.83 5.57 13.98 30.89 + 4.34 14.04 36.00 6.72 18.67
V4W 51.92 8.87 17.08 39.78 + 5.43 13.65 46.71 9.64 20.64
DPW 31.25+ 5.61 17.95 20.56+ 8.50 41.34 26.67 8.70 32.62
VPW 12.83 7.41 57.76 9.78+ 3.90 39.88 11.52+ 6.22 53.99
LPY 21.92+ 4.96 22.63 12.67+ 5.70 44.99 17.95 6.97 58.83
AW 13.33 4.64 34.81 12.78+ 4.99 39.05 13.10 4.68 35.73
VW 32.08 7.20 22.44 23.56+ 9.63 40.87 28.43 9.18 32.29
SIL 28.33 10.47 36.96 18.78 8.38 44.62 24.2410.58 43.65
S2L 113.50 4.93 36.53 9.11 4.04 44.35 11.62 4.98 42.86
S2W 33.92+ 9.19 27.09 20.67 9.46 45.77 28.24 11.29 39.79
ANBW 25.08 9.52 37.96 22.33 + 3.08 13.79 23.90 7.46 31.21
NBW 37.17 13.09 35.22 31.78 4.29 13.50 34.8610.44 29.95
NBL 27.83 9.92 35.64 22.78 3.31 14.53 25.67 8.06 31.40
NL 2.42 1.88 77.69 2.89+ 1.62 56.06 2.62 1.75 66.70
N 6.75 3.62 53.63 5.89+ 2.57 43.63 6.38 3.17 49.69
PP1 21.42 7.12 33.24 17.89 7.74 43.26 19.90 7.42 37.29
DP, 4.17+ 2.12 50.84 2.67 2.55 99.51 3.52 2.38 67.14
LP, 54.25 8.95 16.50 39.22 15.16 38.65 47.8113.93 29.14
'See Figs. 25-26 for abbreviations.








74 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2



TABLE 11 (CONTINUED). SHELL STATISTICS, Gopherus flavomarginatus.

Measure- Males (N=31) Females (N=23) Combined (N=54)
ments1 X+-SD CV X-SD CV XSD CV
PL 263.5031.05 11.78 258.0033.01 12.79 261.6741.01 15.67
G 57.50+ 4.95 8.61 61.03 5.91 9.68 58.63 8.03 13.70
H 43.50+ 3.54 8.14 41.00 4.62 11.28 42.76+ 4.76 11.13
P 25.00+ 4.41 17.64 29.03+ 5.03 17.33 26.30- 6.75 25.67
AB 73.00 -10.83 14.83 67.06 9.81 14.63 71.12 17.75 24.95
F 43.00-+ 6.01 13.98 41.00 5.32 12.99 42.33+ 6.80 16.06
A 26.01+ 4.30 16.53 22.61- 8.31 41.18 24.67+ 10.68 43.29
ND 17.50 + 3.12 17.83 22.00+ 5.63 25.59 19.01- 6.17 32.46
NW 27.50 -14.95 54.36 26.81- 12.01 44.80 27.13 15.61 57.54
XH 11.50-- 3.71 23.56 11.61 3.06 26.64 11.33+ 3.58 31.60
XL 62.00- 8.41 13.56 56.00+ 9.03 16.13 60.17-10.61 17.63
XW 65.50+ 8.12 12.40 74.06 10.21 13.79 68.37 11.37 16.63
POL 86.50 + 8.12 9.39 79.00 9.63 12.19 84.92 10.58 12.45
HYP 48.50 + 9.16 18.89 45.00 10.18 22.62 47.38 9.92 20.94
HYO 66.00 9.28 14.06 67.00 10.00 14.93 68.18+ 13.06 19.16
EL 52.00-- 7.24 13.94 50.13 8.31 16.58 51.86+ 9.63 18.57
EW 51.50+ 7.12 13.83 46.86 7.33 15.64 49.57+ 8.51 17.17
ME 41.00+ 8.63 21.05 42.91 8.38 19.63 41.11+- 9.42 22.91
GE 58.00 + 9.41 16.22 63.61 9.38 14.75 59.13+10.06 17.01
AL 89.00- 17.07 19.18 91.0315.09 16.58 89.91-+18.13 20.16
LT 17.00+- 6.16 36.24 22.66+- 8.14 35.92 18.17-- 9.33 51.34
LL 34.00 5.41 15.91 34.16 6.33 18.53 34.17-+ 8.19 23.39
CL 288.00 61.09 21.21 268.1658.13 22.22 281.33+ 63.19 22.46
CW 238.50+ 31.19 13.08 233.03+ 29.61 12.71 236.51 +33.07 13.98
CH 130.26+28.15 21.23 145.06 27.51 18.96 138.66 +38.18 27.53
V2L 55.50+- 8.63 15.55 52.16+ 8.96 17.18 54.39+ 10.15 18.47
V2W 95.61+10.12 10.58 96.21 9.68 10.06 95.76 +12.61 13.17
V4L 68.91 10.77 15.63 68.03 10.93 16.07 69.71 +15.56 22.32
V4W 98.00 10.03 10.23 95.28 9.13 9.58 97.03+ 11.63 11.99
DPW 52.19 4.24 8.12 54.43- 5.30 9.74 52.67+ 6.31 11.98
VPW 24.62 + 6.41 26.04 27.23 + 5.01 18.40 25.28 + 8.45 33.43
LPY 36.00+ 5.71 15.86 38.51 6.63 17.22 36.78+- 8.92 24.25
AW 22.51+- 3.66 16.26 25.68 + 4.34 16.70 23.81 + 6.03 25.33
VW 74.63+10.13 13.57 69.31 9.10 13.21 63.93+19.07 29.83
S1L 52.01+12.41 23.86 49.93 18.42 36.89 57.83 -21.16 36.59
S2L 32.50- 8.77 26.99 31.00 9.91 31.97 32.48+ 10.07 31.00
S2W 56.88 9.51 16.72 60.91 10.15 16.66 57.83 12.73 22.36
ANBW 54.50+ 6.41 11.76 67.36 8.91 13.23 58.67+ 9.98 17.01
NBW 73.21 + 7.12 9.73 79.01 8.68 10.99 75.06+ 9.16 12.20
NBL 50.00+ 6.83 13.66 45.13 8.73 19.34 48.33+ 10.03 20.75
NL 12.53+ 3.71 29.60 10.13+ 2.82 27.84 11.67+- 5.53 47.38
N 21.01+ 3.96 18.85 17.62 3.17 18.00 19.78 5.69 28.77
PP, 40.38 + 4.91 12.16 45.51+ 3.01 6.61 41.07+ 5.06 12.32
DP, 9.03-+ 1.41 15.61 4.93+ 2.61 52.94 7.83+- 3.09 39.46
LP, 89.51+ 10.17 11.36 95.89- 13.07 13.63 91.02+- 15.81 17.37
'See Figs. 25-26 for abbreviations.









1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 75



TABLE 11 (CONTINUED). SHELL STATISTICS, Gopherus polyphemus.

Measure- Males (N=28) Females (N= 35) Combined (N= 63)
ments' X+SD CV X+SD CV X+SD CV
PL 228.9527.05 11.82 218.7042.40 19.39 217.00+51.99 23.87
G 37.85 8.19 4.62 30.40 12.42 42.50 35.48 10.31 29.06
H 40.10 5.69 14.20 42.20 7.54 17.87 40.36+ 6.86 17.00
P 22.85 5.24 22.93 22.30 3.77 16.91 23.00+ 4.79 20.83
AB 77.60 12.97 16.71 77.80 17.53 22.53 77.48+ 14.43 18.62
F 38.70 5.59 14.44 35.60 5.19 14.58 37.61+ 6.24 16.59
A 14.45 3.30 22.83 14.90 4.43 29.73 14.82+ 3.68 24.90
ND 15.25 3.01 19.74 15.10 3.75 24.83 14.97+ 3.50 23.38
NW 24.85 4.26 17.14 21.40 5.23 24.44 23.45 + 4.96 21.15
XH 13.70 2.79 20.37 13.20+ 3.52 26.67 13.55+ 3.17 23.39
XL 45.95 6.74 14.67 45.20 7.66 16.95 45.64 7.21 15.80
XW 54.30 7.77 14.31 52.9010.40 19.66 53.73+ 9.15 17.03
POL 64.55 7.83 12.13 61.30+11.11 18.12 64.0611.31 17.60
HYP 49.00 9.34 19.06 48.90 8.12 16.60 49.45+ 9.12 18.44
HYO 65.80 9.19 13.97 63.20+12.18 19.27 64.79+10.51 16.22
EL 43.20 8.38 19.40 41.90+ 8.01 19.12 42.64 + 8.22 19.28
EW 48.15 7.58 15.74 45.70+11.11 24.31 47.61+ 9.41 19.76
ME 27.20 5.88 21.62 22.50 11.00 48.89 25.15 8.12 32.29
GE 48.55 8.89 18.31 41.10+ 18.12 44.90 45.97 12.74 27.71
AL 72.30 11.00 15.24 61.40+26.58 43.29 68.36+18.15 26.55
LT 18.10 5.17 28.56 15.80+ 7.11 45.00 17.55+ 5.85 33.33
LL 28.85 5.85 20.28 23.20 10.59 45.65 26.88+ 7.98 29.69
CL 233.75 63.92 27.35 194.40+ 110.52 56.59 214.70 89.26 41.58
CW 179.30 49.80 27.77 163.60+ 67.39 41.19 169.21+ 62.17 36.74
CH 101.15 28.24 27.92 90.80 37.81 41.64 94.82 +34.90 36.81
V2L 51.05 8.14 16.95 44.30 17.31 39.07 47.2114.63 30.99
V2W 75.05 9.63 12.83 67.70+27.72 40.95 70.24 +21.55 30.68
V4L 55.70 8.77 15.76 48.20 20.53 42.59 51.73 16.78 32.44
V4W 73.60 -10.03 13.63 64.10 +27.40 42.75 68.42 22.50 31.42
DPW 36.10 5.14 14.24 34.50 8.53 24.72 34.33 + 9.09 26.48
VPW 19.15 6.56 34.26 19.40 5.08 26.19 18.82+ 6.85 36.40
LPY 25.90 5.00 19.31 25.80 + 9.70 37.60 25.24 + 8.40 33.28
AW 16.35 3.12 19.08 16.40-+ 3.81 23.23 16.12+ 3.32 20.60
VW 43.60 9.95 22.82 34.10 +14.96 43.87 40.55 +12.73 31.39
S1L 34.50 11.62 33.68 27.40 14.32 52.26 32.52+ 12.53 38.53
S2L 21.60 4.33 20.00 19.40 5.30 27.32 20.06+ 5.92 29.51
S2W 47.40 8.52 17.97 41.90 12.61 30.10 43.67-+13.42 30.73
ANBW 41.80+ 5.99 14.33 41.10 10.34 20.33 41.55 7.47 17.98
NBW 54.35 6.38 11.74 54.50+13.48 24.73 54.24- 9.43 17.39
NBL 42.40 5.97 14.08 41.10+ 8.69 21.14 41.88 7.18 16.91
NL 10.80 2.17 20.09 11.10- 3.60 35.14 10.82+ 2.64 24.40
N 12.85+ 3.41 26.54 13.90 3.21 23.09 13.18 3.29 24.96
PP3 35.30 3.96 11.22 35.10 6.67 19.00 35.00- 5.36 15.31
DP3 15.25 3.58 23.48 15.80+ 5.92 37.47 15.24 4.63 30.38
LP3 76.25 9.97 13.08 74.00+-16.47 22.26 75.06 12.56 16.73
'See Figs. 25-26 for abbreviations.

















TABLE 12.-SEXUAL VARIABILITY.

MALE FEMALE BOTH SEXES
Length Range, all CV Length Range, all CV Length Range, all CV
TAXA measures measures measures
PLASTRAL
agassizi 20.59 16.43-35.76 22.60 18.47 18.05-36.48 22.70 17.42 17.39-39.73 23.79 t
berlandieri 16.28 13.92-36.93 21.81 14.93 12.52-27.17 18.79 20.80 17.34-40.70 26.26
flavomarginatus 15.67 12.08-33.61 23.15 8.03 15.03-39.01 24.05 13.14 15.06-51.03 25.06
polyphemus 11.82 11.82-30.84 18.52 19.39 14.58-48.89 26.80 23.87 15.80-33.33 22.76 >
TOTAL PLASTRON CV=23.13 H
CARAPACEAL
agassizi 19.71 14.29-162.94 31.15 21.94 19.26-107.81 43.01 21.27 18.83-143.53 33.04 g
berlandieri 15.52 02.00-77.69 29.80 14.54 01.92-95.51 32.27 19.06 15.24-67.14 34.53 C
flavomarginatus 22.46 12.03-47.83 28.02 11.61 14.38-48.07 32.03 18.06 18.63-76.16 31.08
polyphemus 27.35 11.22-34.28 19.98 56.59 19.00-56.59 33.79 41.57 15.31-41.57 28.33 c
TOTAL CARAPACE CV =31.80



o

IZ








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 77

ever, significant differences do occur between the sexes of the other three species,
with males being less variable than females (polyphemus males CV = 19.98, fe-
males 33.79; flavomarginatus males 28.02, females 32.03; agassizi males 31.15, fe-
males 43.01) (Table 12).
5) None of the species differ significantly from any other in total carapace variability
(berlandieri CV= 34.53, polyphemus CV= 28.33, agassizi CV= 33.04, flavomar-
ginatus CV=31.08).
6) Patterns of character variances differ in each of the four extant species.
The correlation of size and sex was investigated by means of an
analysis of plastral length variance (o) of adult males and females in
each population. This showed that the deviation of each sample is homo-
geneous and within expected limits. The results of Bartlett's test of Chi-
Square (DF=77) strongly suggest (F=24.1) that the observed differ-
ences in intersexual plastral length variances are due to differences in
the mean lengths alone, and not to the observed variability of the sexual
subpopulations.
An identical analysis of carapace length variances in adult males and
females of each population shows that in this instance the observed dif-
ferences in carapace variances are not homogeneous, inasmuch as the
variance in polyphemus females is much greater than any other sample.
Females tend to be more variable in other shell characters as well (No. 4
of both plastral and carapace sections above).
Osteological studies normally make use of one of three comparative
techniques: ratios, log-ratios, or actual measurements. To determine
which of these techniques would be most discriminatory in adult Go-
pherus shells, a series of discrimination analyses were completed, utilizing
a stepwise regression technique (BioMed 005, UCLA Med. Center).
Two tests analyzing discrimination ability of different techniques were
applied to one dependent variable and four (and in another case to 10)
independent morphological variables of the xiphiplastron in three Go-
pherus species (Table 13). Still another technique test utilized the same
TABLE 13.-COMPARISON OF ANALYSIS TECHNIQUES.
Ratios Log-ratios Measurements
C.D.1 D.D2 I.C.3 C.D.1 D.D2 I.C.3 C.D.1 D.D2 I.C.3
4 INDEPENDENT VARIABLES
polyphemus x agassizi (n=40) 11 21 14 14 23 3 18 16 2
berlandieri x agassizi (n=36) 6 30 0 7 29 0 32 4 0
polyphemus x berlandieri (n = 38) 3 34 1 7 30 1 31 7 0
10 INDEPENDENT VARIABLES
polyphemus x agassizi (n=40) 35 5 0 11 28 1 36 4 0
polyphemus x berlandieri (n= 38) 19 8 11 17 1 20 21 8 9
berlandieri x agassizi (n=30) 14 11 5 10 2 18 27 3 0
'C.D.= Correctly differentiated.
"D.D. Doubtfully differentiated.
I.D. = Incorrectly differentiated.








78 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

independent variables, combined the species, and used sex as the single
dependent variable (Table 14). The results clearly show that, for adults,
analyses based on log-ratio comparisons are less discriminatory than
those using the actual measurements or ratios. The necessity of using
variables with the same variances in ratio analyses, the danger of com-
paring dependent variables, and the fact that only adults were being
studied, suggest using only actual measurements and a restricted suite
of ratios as a basis of comparison throughout most of the remaining
analyses.

TABLE 14.-COMPARISON OF ANALYSIS TECHNIQUES (4 Independent Variables, de-
pendent variable sex [n=57].
Ratios Log-ratios Actual Measurements
Correct 15 9 13
Doubtful
(with probable
error) 42 48 42
Clearly incorrect 0 0 2

The mean (X), standard deviation (SD), and coefficient of variation
(CV) of the measured characters were computed (Table 11). Each of
the measurements was roughly plotted against plastral or carapace
lengths to eliminate the problem of size differences and to estimate re-
gression lines. Of the 46 possibly significant characters, 23 were con-
cerned with carapace shape and 24 with plastron shape. The diagnostic
value of each character in distinguishing a pair of species was tested by
computing single character distances for the pair, and dividing the dif-
ference in means for the two populations by the average standard devia-
tion.
From this analysis 5 carapace and 12 plastral measurements were
found to be most diagnostic, although there was overlap between all
species pairs in every character studied. These measurements, which
were used in the linear discrimination tests, are: carapace length, cara-
pace height, carapace width, distal and proximal widths of pleural 3;
plastral length, xiphiplastron length and width, posterior lobe width,
epiplastral lip thickness and length, epiplastral bone length and width,
entoplastral length and width, and anterior lobe length and width.
The linear discrimination technique has been described and its use
explained by Kendall (1951), Jolicoeur (1959), and Lawrence and Bos-
sert (1967). Rao (1952) extended its usefulness through certain cluster-
ing statistics. The discrimination tests find the weighted sum of char-
acters that best separates the populations (where the sum= the discrimi-








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 79

nant function DF, the weights=the discriminant coefficients DC, and
the population mean of DF is the DF X DC for each character). The
relationship of tested population pairs depends on an evaluation of the
DF separating the population pair. Population pairs that are most simi-
lar have similar DF values. The accuracy of the expression of relation-
ship depends on the degree to which the populations are separated by
DF. In addition, the multiple character difference between Gopherus
species was also estimated using Mahalonobis' DZ statistic, which is a
general extension of distance comparisons for single characters. For this
study the discriminant coefficients and the D2 statistic for each pair of the
selected populations of Gopherus polyphemus, G. berlandieri, G. agassizi,
and G. flavomarginatus were computed using the 17 selected adult shell
characters (Table 15). We see that polyphemus differs by 2.4 and 13.6
standard deviations from flavomarginatus and berlandieri, respectively,
whereas agassizi and berlandieri differ by only 3.1 standard deviations.
These tests serve to emphasize the separation of Gopherus into two major
groups.

TABLE 15.-GENERALIZED D2 DISTANCE (AND STANDARD DEVIATIONS) BETWEEN
EXTANT POPULATIONS OF Gopherus.
agassizi
polyphemus 64.8 (08.3) polyphemus
flavomarginatus 55.8 (12.6) 8.1 (02.4) flavomarginatus
berlandieri 21.3 (03.1) 82.3 (13.6) 92.7 (18.9)


The analysis of relationship was checked by tentatively assigning
each specimen to one of the populations by use of the functions, and then
checking each assignment a posteriori against the actual identification.
With the proper choice of characters, most of the specimens can be cor-
rectly identified, though there is considerable overlap in species DF
values. The results of these analyses are plotted in Figure 28. Although
the results are less clear, G. polyphemus and G. flavomarginatus are
closer to one another in both coordinates, whereas berlandieri is clearly
different. Gopherus agassizi overlaps the characters of the remaining
three species in both coordinates.
QUALITATIVE SPECIES-CORRELATED SHELL CHARACTERS.-To date most
efforts to measure differences between the shells of Gopherus populations
or to identify fossil specimens have used few specific measurements of
any kind. They have relied mainly on size and general proportions.
However, the overlap is so great that many individuals, particularly fos-








80 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

sils, cannot be reliably identified. Species characteristics of the shell
depend on complex differential development of the component parts of
the shell. In their extreme form, these differences are easily seen. In
most cases, these are the characters upon which paleontological work
depends, because large series of specimens from restricted sites are not
available. Thus, from a practical standpoint, identifying fragmentary
fossil specimens using multivariate analysis techniques is often impracti-
cal. Identifications must instead be based on a simple system, without
resorting to the kind of analysis described above. The simplest system
is usually a direct comparison of fossil shell fragments with the bones of
known species.
Certain parts of the shell are quite diagnostic, and, when used in
combination, will serve to identify most reasonably complete adult fossil
specimens. A graphic summary of these general differences (which are
usually best developed in adult males) between species of the genus
Gopherus is provided in Figure 32. In identification of fossils, reference
should also be made to shells of typical adults of all extant species (Figs.
29-32).
A short description of the salient differentiating characteristics of the
shells of most adult specimens of each species follows. The quantitative
parameters of the shells of extant populations of Gopherus are given in
Table 12. Reference should also be made to appropriate figures in the
text. In considering species-correlated characters it is important to re-
iterate that certain secondary sex characters are sometimes confused with
specific characters. Thus, adult males of all species generally have a
more concave plastron, the gular projection is proportionately longer, and
the posterior xiphiplastral projection is thickened and directed more
laterally, with more blunted tips than in adult females.
Qualitative shell characters found to be useful in separating the four
extant species are:
1. Adjacent pleural bones (Figs 29-32) typically are alternately wider
aid narrower in almost all tortoises having high shells. This arrangement
apparently strengthens the dome. As might be expected, it is best de-
veloped in berlandieri, in which the shell is proportionately higher than in
the other species. This character is expressed in the ratio of the measure-
ments: Distal width costal scute 3, Proximal width costal scute 3 (X
ratio: berlandieri =5.65, CV=67.14; agassizi=2.39, CV=94.99; flavo-
marginatus 5.22, CV=28.31; polyphemus=1.78, CV= 30.38).
In berlandieri some pleurals actually fail to reach the peripherals.
This character is most variable in berlandieri (CV=67.14) and agassizi
(CV= 94.99).








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 81

The pygal is generally more triangular in berlandieri than in the other
species. G. polyphemus is broadest caudally (Fig. 33, I).
2. The keel on the bridge peripherals is best developed in berlandieri,
most weakly developed in polyphemus, and intermediate in the other two
species (Fig. 33, II).
3. The outward-shell flare over the hind legs is stronger in males
than in females. Within each sex it tends to be best developed in agassizi
and berlandieri, intermediate in flavomarginatus, and least developed in
polyphemus (Fig. 33, III).
4. In males, the curve (as seen from above) formed by the posterior
pleurals and suprapygals is broadly rounded in polyphemus and flavo-
marginatus, rounded to slightly angled in agassizi, and usually strongly
angled in berlandieri. This is related to the degree of doing of the
shell, the flaring over the hind legs, and the downward twisting of the
posterior part of the carapace (Fig. 33, IV).
5. The shell height is proportionately highest in G. berlandieri
(CH/CL = 1.84), intermediate in agassizi (X = 2.07) and flavomarginatus
(X= 2.04), and lowest in polyphemus (X= 2.68).
6. The highest part of the shell is usually behind the midpoint in
berlandieri, usually at the middle in agassizi, and at the middle or an-
terior to it in polyphemus and flavomarginatus.
7. The widest part of the shell is usually well behind the center in
berlandieri; behind the center, but less markedly so, in agassizi; center
(sometimes behind it) in flavomarginatus; and also at the center, but
with rather parallel sides, in polyphemus.
8. In both berlandieri and agassizi, the upper surface of the epi-
plastral lip is often concave. It is rarely so in flavomarginatus or poly-
phemus, particularly the latter. The lip is usually proportionately longer
in berlandieri and agassizi (Fig. 33, VIII).

9. The bridge is shortest in flavomarginatus X 2.99
SPlastron L
longest in polyphemus (2.57), and intermediate in agassizi (2.77) and
berlandieri (2.67). The anterior and posterior plastral lobes are nearly
equal in polyphemus (Ant. Lobe L/Plastral L X=3.25, Post. Lobe L/
Plastral L X=3.37) and flavomarginatus (3.26 and 3.41), and more un-
equal in the two remaining species (agassizi 2.84 and 3.43; berlandieri
2.58 and 3.66 respectively). The bridge in polyphemus and flavomargina-
tus is almost always larger than both the anterior and the posterior lobes.
In agassizi and berlandieri the bridge length usually equals the anterior
lobe length and is larger than the posterior lobe (polyphemus Bridge








82 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

L/Plastral L X=2.57; flavomarginatus 2.99; agassizi 2.77; berlandieri
2.67) (Fig. 33, IX).
11. The entoplastron is very often narrow and long in berlandieri
(Entoplastral L/Entopl. W X=0.89), with the lateral apices of the penta-
gonally-shaped bone often in the anterior half. In the remaining species
these apices are always in the posterior half. The element is usually
widest in polyphemus (1.11), and intermediate in flavomarginatus (0.95)
and agassizi (1.03) (Fig. 33, X).
12. In all adult tortoises, the periphero-pleural articulation is com-
pletely fused, whereas in subadults this is not so. This fusion occurs at
different sizes in each of the Gopherus species, and is obviously directly
correlated with adult size attained by each of the species. In berlandieri
fusion occurs at a shell length of about 130 mm, in agassizi about 180
mm, in polyphemus about 225 mm, and in flavomarginatus about 240 mm
(Fig. 33, XI).
13. From the front, the anterior opening of the shell is more concave
dorsally in berlandieri than in the remaining three species (Fig. 33, XII).
14. In adult males the epiplastral projection is narrowest and longest
in berlandieri, and is often quite bifurcated. In agassizi it is quite vari-
able, ranging from this extreme to a wide, medium-length process with
a shallow notch. The projection in polyphemus is proportionately widest
and shortest, is never bifurcate, and often lacks even the median anterior
notch. G. flavomarginatus broadly overlaps the lower end of the range
of variation in agassizi and overlaps almost the entire range of variation
of polyphemus (Fig. 33, XIII).


LrrERATURE CITED

Auffenberg, W. 1961. A correction concerning the phalangeal formula of the turtle
Stylemys nebrascensis Leidy. Copeia 1961(4): 496-498.
.1963. A redefinition of the fossil tortoise genus Stylemys Leidy. J.
Paleontol. 38(2): 316-324.
1966a. The carpus of land tortoises (Testudinata). Bull. Florida
State Mus., Biol. Sci. 10(5): 159-191.
1966b. On the courtship of Gopherus polyphemus. Herpetologica
22(2): 113-117.
Bogert, C. M., and J. Oliver. 1945. A preliminary analysis of the herpetofauna of
Sonora. Bull. Amer. Mus. Nat. Hist. 83(6): 297-426.
Bramble, D. M. 1974. Occurrence and significance of the os transiliens in gopher
tortoises. Copeia 1974(1): 102-109.
Brattstrom, B. 1961. Some new fossil tortoises from western North America, with
remarks on the zoogeography and paleoecology of tortoises. J. Paleontol. 35(3):
543-560.
Carr, A. F., Jr. 1957. Handbook of turtles. New York: Constock Publ. Assoc.
542 p.









1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 83

Grant, C. 1960. Gopherus. Herpetologica, 16: 29-31.
Gray, J. E. 1873. On the skulls and alveolar surfaces of land tortoises (Testudi-
nata). Proc. Zool. Soc. London. pp. 722-728.
Hay, O. P. 1908. The fossil turtles of North America. Carn. Inst. Wash. Publ.
(75): 1-555.
Hoffmann, C. K. 1890. Schildkroten, In Bronn's, Klassen und Ordnungen des
Thier-Reichs. Leipzig. 442 p.
Jolicoeur, P. 1959. Multivariate geographical variation in the wolf, Canis lupus
L. Evolution 13: 283-299.
and J. E. Mosimann. 1960. Size and shape variation in the painted
turtle: A Principal Component Analysis. Growth, 24: 339-354.
Kendall, M. G. 1951. The advanced theory of statistics, Vol. II. New York:
Hafuer House. 521 p.
Kermack, K. A., and J. B. S. Haldane. 1950. Organic correlation and allometry.
Biometrika 37: 30-41.
Lawrence, B., and W. H. Bossert. 1967. Multiple character analysis of Canis lupus,
latrans, and familiaris, with a discussion of the relationships of Canis niger.
Amer. Zool. 7: 223-232.
Legler, J. M. 1959. A new tortoise, genus Gopherus, from northcentral Mexico.
Univ. Kansas Publ., Mus. Nat. Hist. 11(5): 335-343.
1962. The os transiliens in two species of tortoises, genus Gopherus.
Herpetologica 18(1): 68-69.
Legler, J. M., and R. C. Webb. 1961. Remarks on a collection of bolson tortoises,
Gopherus flavomarginatus. Herpetologica 17(1): 26-37.
Loveridge, A., and E. E. Williams. 1957. Revision of the African tortoises and
turtles of the Suborder Cryptodira. Bull. Mus. Comp. Zool. 115: 163-557.
Mertens, R. 1964. Uber Reptilienbastarde, III. Nat. Mus. Forschungs-Inst. Sencken-
berg 45(1): 33-49.
Mertens, R., and H. Wermuth. 1955. Die rezenten Schildkroten Krokodile und
Bruckenechsen. Zool. Yarhb. Syst. 83: 323-440.
Rao, C. R. 1952. Advanced statistical methods in biometric research. New York:
John Wiley. 390 p.
Ray, C. E. 1959. A sesamoid bone in the jaw musculature of Gopherus polyphemus.
Anat. Anz. 107: 85-91.
Rose, F. L., R. Drotman, and W. G. Weaver. 1969. Electrophoresis of chin gland
extracts of Gopherus (Tortoises). Comp. Biochem. Physiol. 29: 847-851.
True, F. W. 1882. On the North American land tortoises of the genus Xerobates.
Proc. U. S. Natl. Mus. 4: 434-449.
Williams, E. E. 1950a. Testudo cubensis and the evolution of western hemisphere
tortoises. Bull. Amer. Mus. Nat. Hist. 95(1): 1-36.
.1950b. Variation and selection in the cervical central articulations of
living turtles. Bull. Amer. Mus. Nat. Hist. 94(9): 505-561.
S1952. A new fossil tortoise from Mona Island, West Indies, and a
tentative arrangement of the tortoises of the world. Bull. Amer. Mus. Nat. Hist.
94: 541-560.
Woodbury, A. M. 1952. Hybrids of Gopherus berlandieri and G. agassizi. Herpe-
tologica 8(1): 33-36.
Zangerl, R. 1957. A parietal foramen in the skull of a recent turtle. Proc. Zool.
Soc. Calcutta. Mookerjee Mem. Vol.: 269-272.








84 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

------ c-------- --------i _^_----




a b g hi




FIGURE 1.-Skull measurements utilized in this study. In addition, k= greatest skull
height, f= outer alveolar angle.



220 ,
0
Q*
200 -

E 0 0 0
180- 0
000

24 28 32 0 60 0 4
O O
0 0 0
S160 00. 0 0











SKULL LENGTH J (IN MM)
"FIGUE 2.-The skulls of Gopherus berlander males are proportionately longer than





those of females.
those of females.








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 85

60

56

0 52





pr po
40

10 20 30 40 50 60
skull length J
FIGURE 3.-During growth the preorbital area of the skull increases faster than the
postorbital area. a =agassizi, b =berlandieri, f = flavomarginatus, p =polyphemus.







a 44 45 11
b 31 60 9
f 16 28 56
p 80 18 2







a 0 13 87
b 0 9 91
f 60 32 8
p 98 2 0
FIGURE 4.-Upper row, major bone patterns of the postorbital bar, clear = postorbital,
black= jugal, stippled= quadratojugal. Lower row, patterns of exoccipital shape
(stippled) and included foramina. a= G. agassizi, b= G. berlandieri, f= G. flavo-
marginatus, and p = G. polyphemus.









86 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2









































FIGURE 5.-Skull, Gopherus polyphemus, UF 11110.









1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 87






































FIGURE 6.-Skull, Gopherus agassizi, W. A. Field No. 75.









88 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2








































FIGURE 7.-Skull, Gopherus flavomarginatus, W. A. Field No. 200.









1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 89








































FmIGnE 8.-Skull, Gopherus berlandieri, W. A. Field No. 21.








90 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2


















A

















FIGURE 9.-Otic area, A. Gopherus flavomarginatus, B. G. polyphemus, C. G. agassizi,
D. G. berlandieri. Bars represent 1 cm.









1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 91


















































FIGcRE 10.-Appendicular skeleton, Gopherus polyphemus, UF 11110. Upper left,
femoral views; upper right, pelvis; lower left, scapular articulation and coracoid;
lower right, humeral views.









92 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2





















































FIGURE 11.-Appendicular skeleton, Gopherus berlandieri, W. A. Field No. 21. Same
views as Figure 10.









1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 93




















































FIGURE 12.-Appendicular skeleton, Gopherus flavomarginatus, W. A. Field No. 63.
Same views as Figure 10.









94 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2




















































FIGURE 13.-Appendicular skeleton, Gopherus agassizi. W. A. Field No. 83. Same
views as Figure 10.







1976 AUFFENBERG:. RECENT GOPHERUS OSTEOLOGY 95




R






A B







U+p+

3' CMt 5.



pC+







C +m2 3CIr+mZ + m3
C 30



FIGURE 14.-Carpi (diagrammatic) of Gopherus species: A. Adult G. polyphemus,
B. Adult G. flavomarginatus, C. Juvenile G. agassizi, D. Adult G. agassizi, E. Juvenile
G. berlandieri, F. Adult G. berlandieri.








96 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2


































I CM










FIGURE 15.-Manus (left), and pes (right) of Gopherus polyphemus, UF 11110.









1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 97
























FIGURE 16.-Dermal armor of the outer surface of the manus (left) and heel of the
pes (right) of Gopherus polyphemus.













-- ----- -








A B C D
FIGURE 17.-Ossification of the carapace in Gopherus polyphemus. A) Shell length
72 mm; B) Shell length 109 mm; C) Shell length 182 mm; D) Shell length 204 mm.









98 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2




A ^A B B

4







FIGURE 18.-Left, peripheral-pleural articulation in a young Gopherus polyphemus
(W. A. Field No. 70), carapace length 127 mm. Upper, inner surface of anterior
bridge peripherals (numbers 2-4), showing relationship of rib ends and peripherals.
Lower, cross section through AA above showing rib ends in shallow peripheral grooves.
Right, Same in a young adult Gopherus polyphemus (UF 10944), carapace length
171 mm. Rib ends are now in shallow peripheral pits.
































FIGURE 19.-Carapace growth patterns in, upper, G. polyphemus; lower, G. ber-
landieri. Hatchling on left, adults on right.








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 99




















FIGURE 20.-Major types of anomalies recorded in the bones and scutes of Gopherus
(all species combined). Bones are on the left of each figure, scutes on right.




4 6 6 )6 76 B


undifferentiated








4 8 4 8 4l D


FIGURE 21.-Neural differentiation in turtles. Black arrows indicate major growth
direction in pleural bones. A, undifferentiated pattern of most hatchling tortoises.
B, typical neural pattern of most emydine turtles and primitive testudinines. C, pat-
tern variant found in several turtle groups. D, typical pattern of advanced tortoise
groups.








100 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

ANTERIOR
A -- r B C
3.6% 4.7% 9%







D E F
75% 4.7% 3.6%







FIGURE 22.-Distribution of proximal pleural growth patterns (A-F) in Gopherus
(all species combined) when each pleural-neural pair is considered as a single in-
dependent unit.










C "









A
FIGURE 23.-Ossification patterns in pygal area of Gopherus. A, hatchling condition
with fenestra shown in black. B and C, derived adult bone patterns, depending on
whether suprapygals 1 or 2 fill the fenestra. p=pygal bone.









1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 101






















--- /---
2













1%
2/ 00/0


















"P 43%






FIGURE 24.-Variation in ossification patterns of Gopherus (all species combined).
1 and 2= suprapygals 1 and 2, p=pygal.









102 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2





PL












G- H P AB ---t-A

S _-E L HYO HYP XL-
z
-4-

-GE N



AL PO L-
AP.





xw-4


FIGURE 25.-Plastral measurements used in this study. Stippled drawings are cross
sections of plastral lip (left) and xiphiplastron (right). Abbreviations: A, anal; AB,
abdominal; AL, anterior lobe length; AP, anterior projection length; APW, anterior
projection width; EL, entoplastron length; EW, entoplastron width; F, femoral; G,
gular scute length; GE, greatest epiplastron length; GW, gulo-humeral sulcus; H,
humeral; HYO, hyoplastron; HYP, hypoplastron; LL, epiplastron lip length; LT, lip
thickness; ME, median epiplastral length; ND, anal notch depth; NW, anal notch
width; P, pectoral; PL, plastron length; POL, posterior lobe length; XH, xiphiplastron
height of outer surface; XL, xiphiplastron length; XW, xiphiplastron width.








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 103






NBW
ANBW
AWI t N |I N4L

SII

PP3
S2 PP4

W
VW------

DPW


\ V4W


VPW4 V-DP4-4- -DP3-
FIGURE 26.-Carapace measurements used in this study. Upper left, suprapygals 1
and 2. Lower left, pygal. Upper middle, nuchal bone. Lower middle, 4th vertebral
scute. Right, neural and pleural bones. Abbreviations: ANBW, anterior nuchal bone
width; AW, dorsal width suprapygal 1; CL, carapace length; CW, carapace width;
CH, carapace height at highest point; DP,, distal width pleural 4; DPW, dorsal
width of pygal bone; LP3, length pleural 3; LPY, length pygal bone; NBL, nuchal
bone length; NBW, nuchal bone width; NL, nuchal scute length; NW, width nuchal
scute; N4L, neural 4 length; PP., proximal width pleural 3; PP4, proximal width
pleural 4; SlL and S2L, suprapygal 1 and 2 lengths; S1W and S2W, suprapygals 2
and 3 widths; V2W, vertebral scute 2 width; V2L, vertebral scute 2 length; V4W,
vertebral scute 4 width; V4L, vertebral scute 4 length; VPW, ventral pygal bone
width; VW, ventral width suprapygal 1.









104 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2



25.3 33.9 25.1 40.0


274 20.4 25.8 21.8



22.3 20.4 24.4 20.7




20.9 31.0
17.4 21.0



20.9 1 721.1
177 19.7
24.6 26.5


A B




13.7 22.9 29.1 32.3


25.7 18.6 20.8 19.7



11.3 19.2 170 16.2




25.0 18.6
20.9 18.4



16.1 16.6
176 15.8
43.3 24.9


C D
FIGURE 27.-Comparison of coefficients of variation in median lengths of plastral
scutes (right) and bones (left) in the extant species of Gopherus. A) agassizi; B)
berlandieri; C) flavomarginatus; D) polyphemus.








1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 105







7-i
BERLANDIERI

6-


5-
POLYPHEMUS

4- +


3-
AGASSIZI

2-
FLAVOMARGINATUS



8 9 10 11 12 13 14 15 16
FIGURE 28.-Linear discrimination of Gopherus species. The contours indicate the
extreme range of individuals in the populations used. The berlandieri-polyphemus
discriminant function is used as the ordinate, the berlandieri-agassizi discriminant
function as the abcissa.









106 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2













































S








FIGuRE 29.--Shell of adult Gopherus berlandieri.









1976 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 107











































31











FIGUxE 30.-Shell of adult Gopherus agassizi.









108 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2




















































FIGURE 31.-Shell of adult Gopherus flavomarginatus.










1970 AUFFENBERG: RECENT GOPHERUS OSTEOLOGY 109






















f



























S3-Shell of adult oheu olhes.







-r -- 7 'j f 7
r.-.









110 BULLETIN FLORIDA STATE MUSEUM Vol. 20, No. 2

BERLANDIERI AGASSIZI FLAVOMARGINATUS POLYPHEMUS


I PYGAL SHAPE 765 '7 \ 7 2



II BRIDGE MARGINALS
SECTIONO)


SHELL FLAIRLEGS
"OVER HIND LEGS I



IV SHELL POSTERIOR
FROM ABOVE


V SHELL X-SECTION MOD-SLIGHTLY SLIGHTLY
HIGH DEPRESSED DEPRESSED DE

VI HIGHEST POINT BEHIND USUALLY MIDDLE OR MIDDLE OR
ON SHELL MIDDLE AT MIDDLE ANT. TO MID. ANT. TO M.D.
USUALLY AT MID.
VII WIDEST POINT BEHIND BEHIND MID.. M DLE TO SIDES OFTEN
ON SHELL MIDDLE BUT LESS MARKED BEHIND MID PARALLEL



LIP X-SECTION




IX PLASTRON





X ENTOPLASTRON O


XI PERIPHERO- GENERALLY OFTEN SMALL
PLEURAL FIRMLY FUSED FENESTRA
ARTICULATION FIRMLY FUSED FENESTRAL COMMON


XII TRANS-X-SEC. t I
SHELL C f C



XIII EPIPLASTRAL / f
PROJECTION ((3)
FIGuRE 33.-Qualitative shell differences in the extant Gopherus species. Numbers
adjacent to each of the pygals are the average values for anterior/posterior widths.













Contributions to the BULLETIN OF THE FLORIDA STATE MUSEUM, BI-
OLOGICAL SCIENCES SERIES, may be in any field of biology. Manuscripts deal-
ing with natural history of systematic problems involving the southeastern United
States or the New World tropics are solicited especially. Manuscripts should be of
medium length-circa 35 to 200 pages (10,500-16,000 words). Examination for
suitability is made by an Editorial Board.

The BULLETIN is distributed worldwide through institutional subscriptions and
exchanges. It is considered the responsibility of the author to distribute his paper
to all interested individuals. To aid in this the authors) receives(s) 50 copies free,
and he may purchase additional separates at cost if ordered when page proof is re-
turned. The author is also responsible for any charges incurred for alterations made
by him on galley or page proofs. The Museum will send an invoice to the author
for this amount upon completion of publication.


PREPARATION OF MANUSCRIPT

Contributors should consult recent numbers of the BULLETIN for preferred style
and format. Highly recommended as a guide is the CBE Style Manual, 3rd Edition,
1972 (Washington, D.C., Amer. Inst. Biol. Sci.).

MSS must be submitted in duplicate (please no onionskin) and satisfy the following
minimal requirements: They should be typewritten, double-spaced (especially tables,
figure captions, and "Literature Cited"), on one side of numbered sheets of standard
(8-1/2x11 in.) bond paper, with at least one-inch margins all around. Tables
(which should be unruled) and figure legends should be typed on separate sheets.
All illustrations are referred to as figures. They must comply with the following
standards: Photographs should be sharp, with good contrast, and printed on glossy
paper. Drawings should be made with dense black waterproof ink on quality paper
or illustration board. All illustrations should have a cover sheet. All lettering will
be medium weight, sans-serif type (e.g. Futura Medium, News Gothic) in cutout,
dry transfer or lettering guide letters. Make allowance so that after reduction no
lowercase letter will be less than 1 mm high (2 mm is preferred) nor any capital
letter greater than 5 mm high. The maximum size for illustrations is 8-5/8x14 in.
(twice typepage size); illustrations should not be less than typepage width (4-5/16
in.). Designate the top of each illustration and identify on the back with soft
pencil by author's name, MS title, and figure number.



Manuscripts and all editorial matters should be addressed to:
Managing Editor of the BULLETIN
Florida State Museum
Museum Road
University of Florida
Gainesville FL 32611

































































































































































I





University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs