l
f

f

(I

I
[/1

I
7

l
I

If!“ M

HUM!”

w
I

r’lWl'

I
f

I—
W“.
—_

‘ I“
H.’

 

 

A LATE PLEISTGCENE HERPETOFAUNA
FROM BAKER BLUFF CAVE.
SULLIVAN COUNTY, TENNESSEE

The“: for Hm Degree 0f M. S.
MICHIGAN STATE UNIVERSITY

George Henry Van Dam
1976

: I - -
M 1033‘: ‘ , ¢ "i. h)
1C11>AC3§ ”1 . ~ VI. w

. Un'

   
 
   
  

  

v: - 1:: ~. -
{Vb} 33W

   

 

 

 

ABSTRACT

A LATE PLEISTOCENE HERPETOFAUNA
FROM BAKER BLUFF CAVE,
SULLIVAN COUNTY, TENNESSEE

by

George Henry Van Dam

Baker Bluff Cave is located in extreme northeastern
Tennessee (Sullivan County), physiographically lying in
the southern portion of the Ridge and Valley Province of
the Appalachians, west of the Great Smokies (Unaka Range).
It is located in the Carolinean Biotic Province (Dice, 1943).

Due to the discovery of a caribou (Rangifer tarandus)

 

lower premolar, the Carnegie Museum decided to sample the
site at a depth below that of the amateur excavation. Four
Cl4 tests run by the Carnegie Museum, at different
stratigraphic levels, give datings ranging from 555:185 C14
years B.P. to 19,100i850 Cl4 years B.P.

The Baker Bluff Cave Herpetofauna consists of at least
five species of urodeles, five species of anurans, one
species of turtle, one species of lizard and 11 species of
snakes.

Perhaps the most striking thing about the herpetofauna

George Henry Van Dam

is that there is nothing that strongly indicates that the
climate or topography was any different than it is in the
area today.

The Baker Bluff Cave Herpetofauna exhibits many
similarities to the Ladds site (Late Pleistocene), of
northwestern Georgia (Holman, 1967).

Four major habitats are indicated at the Baker Bluff
Cave by the herpetological remains, but it is impossible to
conclude the exact nature of the environment immediate to
the cave because the bones were apparently derived from

wide-ranging predators.

A LATE PLEISTOCENE HERPETOFAUNA
FROM BAKER BLUFF CAVE,
SULLIVAN COUNTY, TENNESSEE

BY

George Henry Van Dam

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Geology

To Ann

ii

ACKNOWLEDGMENTS

I would like to thank my major professor, J. Alan
Holman, and the other members of my committee, Chilton E.
Prouty and Marvin M. Hensley, for their help in the
preparation of this paper. I would also like to thank
Thomas L. Kramer for giving me advice on various aspects of
the paper. Appreciation is also expressed to John E. Guilday
and the Carnegie Museum for letting me describe the Baker
Bluff Herpetofauna. Finally, my deepest appreciation goes
to my wife, Ann, who diligently spent hours of her time in

typing the manuscript.

iii

‘

LIST OF TABLES

INTRODUCTION .

TABLE

0 O O

SYSTEMATIC PALEONTOLOGY .

DISCUSSION . .

LITERATURE CITED

OF CONTENTS

iv

Page

27

31

LIST OF TABLES

Table Page

I. Minimum number of individuals - amphibians . . 29
II. Minimum number of individuals - reptiles . . . 29

III. Ecological preference chart . . . . . . . . . 30

INTRODUCTION

Baker Bluff Cave is located in extreme northeastern
Tennessee (Sullivan County) on the South Fork of Holston
River, a tributary of the Tennessee River at latitude
36°27'30" North, longitude 82°28' West approximately seven
miles south of the Tennessee-Virginia border and 40 miles
west of the Tennessee-North Carolina border. The cave is
on the west bank of the river three miles northwest
(downstream) of the junction of the South Fork Holston
River and the Watanga River at an elevation of approximately
1,550 feet above sea level.

Physiographically, the cave site lies in the southern
portion of the Ridge and Valley Province of the Appalachians,
west of the Great Smokies (Unaka Range). Biologically,
Baker Bluff Cave is located in the Carolinean Biotic
Province (Dice, 1943).

The cave is small, approximately 12 feet by 30 feet,
and prior to excavation was filled to within four feet of
the ceiling by sediments. The top three feet of sediment
were excavated by amateurs searching for Indian artifacts,

who informed the Carnegie Museum, Section of Vertebrate

2

Fossils, of the site. Due to their discovery of a caribou

(Rangifer tarandus) lower premolar, it was decided to sample

 

the site at a depth below that of the amateur excavation.
In 1970, the Carnegie Museum excavated a shaft four feet by
three feet and seven feet deep, beginning at the original
three foot level and extending down to the ten foot level.
Although time breaks apparently exist in this sequence (see
below), periods of sedimentation were probably continuous
as no stratification is evident.

No evidence of human occupation was seen during the
museum excavation.

It appears that all of the bones, seeds, shells, etc.,
were either predator-derived, brought in by rodents, or were
remains of animals that lived and died in the cave. Large
bones were scarce and extensively rodent-gnawed.

Four Cl4 tests run by the Carnegie Museum at Baker Bluff
Cave on uncharred bone fragments give datings ranging from
5551185 C14 years B.P. to 19,0001850 Cl4 years B.P. The

relationship of these Cl4 dates to the levels are as

follows:
4-5 Ft. Level 555:185 Cl4 years B.P.
6-7 Ft. Level 10,56oi220 cl4 years B.P.

11,64oi250 C14 years B.P.

9-10 Ft. Level 19,100i850 C14 years B.P.

Dates below the six foot level indicate a late Wisconsinan
age.

Fossils of the Baker Bluff Cave Fauna include
gastropoda, fishes, amphibians, reptiles, birds, and
mammals. The present report deals with the herpetofauna of
Baker Bluff Cave.

Several people are involved in identifying the
vertebrates and invertebrates from Baker Bluff Cave,
Frederick C. Hill, University of Louisville, is studying
the fish remains; Paul W. Parmalee, University of Tennessee,
is studying the birds; Elaine Anderson, Maryland Academy of
Science, and John E. Guilday, Carnegie Museum, are studying
the mammals. In addition, gastropoda are being identified

by Leslie Hubricht of Meridian, Mississippi.

SYSTEMATIC PALEONTOLOGY

This section consists of an annotated list of all
species identified from the Baker Bluff Cave Fauna. Criteria
for osteological identification of the species are discussed;
habitats are also noted.

Except for the presence of Lampropeltis getulus and

 

possibly Ambystoma tigrinum, all species are found in the

 

area today.

A checklist of the species discussed is given below.

Class Amphibia
Order Urodela
Family Ambystomatidae
Ambystoma opacum (Gravenhorst)
Ambystoma maculatum (Shaw)
Ambystoma sp. indet.
Family Proteidae
Necturus maculosus (Rafinesque)
Family Cryptobranchidae
Cryptobranchus alleganiensis (Daudin)
Family Plethodontidae
Desmognathus sp. indet.
Order Anura
Family Bufonidae
Bufo americanus Holbrook
Bufo woodhousi fowleri Girard
Bufo sp. indet.
Family Hylidae
Hyla sp. indet.
Family Ranidae
Rana sylvatica LeConte
Rana catesbeiana Shaw
Rana sp. indet.

 

 

 

 

 

 

 

 

 

 

 

Class Reptilia
Order Testudines
Family Emydidae
Graptemys geographica (LeSueur)
Order Squamata
Suborder Sauria
Family Scincidae
Eumeces fasciatus (Linnaeus)
Suborder Serpentes
Family Viperidae
Crotalus horridus Linnaeus
Family Colubridae
Subfamily Xenodontinae
Heterodon platyrhinos Latreille
Subfamily Colubrinae
Diadophis punctatus (Linnaeus)
Carphophis amoenus (Say)
Coluber or Masticophis sp. indet.
Lampropeltis triangulum (Lacepede)
Lampropeltis getulus (Linnaeus)
Elaphe sp. indet.
Subfamily Natricine
Natrix sipedon (Linnaeus)
Natrix sp. indet.
Thamnophis sirtalis (Linnaeus)
Thamnophis sauritus (Linnaeus)
Thamnophis sp. indet.

 

 

 

 

 

 

 

 

 

 

Fossils are from the collection of the Carnegie

Museum of Natural History (CM).

Class Amphibia
Order Urodela
Family Ambystomatidae Hallowell
Assignment to the family Ambystomatidae is based on
characters discussed by Holman (1962): centrum amphicoelous,
weakly constricted ventrally, without spine produced from
its posteroventral surface; neural spine obsolete and single

throughout.

Ambystoma opacum (Gravenhorst)

 

Material: CM29754-CM29757. Two pre-caudal vertebrae

 

(4-5'); one pre-caudal vertebra (5-6'); five pre-caudal
vertebrae (6-7'); one pre-caudal vertebra (8-9').

Remarks: Tihen (1958) pointed out that the
Ambystomatidae can be divided up into major groups using
vertebral ratios. He stated that the most useful ratios
were (1) the length of the centrum divided by its width at
the anterior end, and (2) the combined zygapophyseal width
divided by the zygapophyseal length. In addition, Tihen

(1958) also noted that in the A, maculatum group and in the

 

subgenus Linguaelapsus (at least in the posterior part of
the trunk) the postzygapophyses extends as far, usually
farther posteriorly, than does the neural arch.

In my examination of the material, I used the length of
the centrum divided by its width at the anterior end for

species determination.

Ambystoma maculatum (Shaw)
Material: CM29758-CM29760. Six pre-caudal vertebrae
(4-5'); three pre-caudal vertebrae (5-6'); four pre-caudal
vertebrae (6—7').

Remarks: Material was assigned to A, maculatum based

 

on criteria discussed under A, Opacum.

Ambystoma sp. indet.

 

Material: CM29761—CM29767. Three pre-caudal vertebrae
(3-4'); 15 pre-caudal vertebrae (4-5'); seven pre-caudal
vertebrae (5-6'); seven pre-caudal vertebrae (6-7'); three
pre-caudal vertebrae (7-8'); seven pre-caudal vertebrae
(8-9'); one pre-caudal vertebra (9-10').

Remarks: Material which was too fragmentary for
measurement or those whose ratios fell within the ranges

of both A, maculatum and A, opacum were assigned to genus

 

only.

Some specimens referred to Ambystoma species exhibited

 

the back-swept neural arch characteristic of A, tigrinum as
pointed out by Holman (1969); but, these could not be
assigned to species because anterior thoracic vertebrae of

A, maculatum and A, opacum have this characteristic. Conant

 

(1975) states that Ambystoma opacum occurs in a variety of

 

habitats, ranging from moist, sandy areas to dry hillsides.

A, maculatum is occasionally found (from spring to autumn)

 

beneath stones or boards.

Family Proteidae Tschudi
Specimens were assigned to the Proteidae by comparison
with recent specimens and using criteria of Holman (1968)
who noted that their vertebrae are amphicoelous and the

transverse processes are undivided.

Necturus maculosus (Rafinesque)

 

Material: CM29768-CM29773. Fifteen pre-caudal
vertebrae (3-4'); 22 pre-caudal and two caudal vertebrae
(4-5'); five pre-caudal vertebrae (5—6'); three pre-caudal
vertebrae (6-7'); five pre-caudal vertebrae (7-8'); three
pre-caudal vertebrae (8-9').

Remarks: The fossils were indistinguishable from Recent

material of A, maculosus. Necturus maculosus habitats

 

 

include lakes, ponds, rivers, streams and other permanent

bodies of water (Conant, 1975).

Family Cryptobranchidae Cope

Cryptobranchus alleganiensis (Daudin)

 

Material: CM29774-CM29779. Thirty pre-caudal, three

 

caudal vertebrae and one right dentary (3-4'); 73 pre-caudal
vertebrae and two vomers (4-5'); 33 pre-caudal vertebrae
(5-6'); nine pre-caudal vertebrae (6-7'); one pre-caudal
vertebra (7-8'); four pre-caudal vertebrae (9-10').

Remarks: Meszoely (1967) gave characteristics for

the identification of Cryptobranchus: (l) deeply

 

amphicoelous cotyles, circular in outline; (2) centrum
relatively short in respect to the diameter of the cotyle;
(3) ventral surface of the centrum rounded without keel or
processes; (4) large lateral fossa anteriad to the base of
the transverse process; and (5) very large size. In
addition, Holman (1968) notes that the transverse processes
are undivided.

Due to the fragmentary nature of much of the material
I found it sometimes difficult to separate vertebrae of

Cryptobranchus alleganiensis and Necturus maculosus. I have

 

 

found the following criteria useful in differentiating these

two: (1) in g, alleganiensis the sides of the centrum are

 

10

more sculptured than in N. maculosus, (2) in g, alleganiensis

 

 

the upper transverse process is heavy and cylindrical in
shape (A, maculosus has a very wing-like upper transverse
process) and, (3) g, alleganiensis has the articular facets
of the transverse processes exhibiting a single opening

whereas in N, maculosus there are usually two distinct

 

openings. 9, alleganiensis is always found in rivers and
larger streams where water is running and ample shelter is

available in the form of large rocks, snags or debris

(Conant, 1975).

Family Plethodontidae Gray

Desmognathus Sp. indet.

 

Material: CM29780-CM29785. Twenty-two vertebrae
(3-4'); 158 vertebrae (4-5'); 95 vertebrae (5-6'); 81
vertebrae (6-7'); 19 vertebrae (7-8'); 21 vertebrae (8-9').

Remarks: These specimens have been assigned to
Desmognathus based on Soler (1950) who states that their
vertebrae are opisthocoelous and have pointed processes
arising from the dorsal surfaces of the postzygapophyses.

I am unable to carry the identification any further due to

lack of Recent comparative material.

 

Desmognathus fuscus, Q, quadramaculatus, 2, monticola

 

 

and Q, wrighti are all present in northeastern Tennessee

11

today and it is not unlikely that all four are present in

the collection. Desmognathus fuscus occurs in brooks, near

 

springs, and in seepage areas, most commonly along edges of
small woodland streams, where stones, chunks of wood and
miscellaneous debris provide ample shelter both for the
salamanders and for their food. 2, quadramaculatus is
abundant in boulder-strewn brooks and also found near
waterfalls or other places where cold water drips or flows.

2, monticola prefers boggy spots in cool, well-shaded

 

ravines and banks of mountain brooks. Q, wrighti is today
a resident chiefly of high spruce-fir forests and lives
under moss and bark on rotting logs or beneath rotting wood
or litter on the forest floor near seepage areas (Conant,

1975).

Order Anura
Family Bufonidae Fitzinger
Fossil §3f9_as pointed out by Holman (1962) may be
identified by the following characteristics: ilium with
dorsal blade absent; dorsal prominence produced dorsally,
well developed, grooved or irregular in shape; sacral
vertebrae procoelus, with one anterior and two posterior
condyles; sacrum free from urostyle, its diapophyses

moderately expanded.

12

Bufo americanus Holbrook

 

Material: CM29786-CM29792. Twelve right and seven
left ilia (3-4'); 17 right and 24 left ilia (4-5'); ten
right and ten left ilia (5-6'); three right and seven left
ilia (6-7'); two right ilia (7-8'); two right ilia (8-9');
three left ilia (9-10').

Remarks: Holman (1967) pointed out that the ilium of

Bufo woodhousi fowleri has the base of the dorsal

 

protuberance narrower than in equal-sized A, americanus.

 

Habitats seem to be shallow bodies of water in which to
breed (temporary pools or ditches or shallow portions of
streams, for example), shelter in the form of hiding places
where there is some moisture, and an abundant food supply

of insects and other invertebrates (Conant, 1975).

Bufo woodhousi fowleri Girard

 

Material: CM29793-CM29795. Three right ilia (3-4');
two right and three left ilia (4-5'); one left ilia (5-6').

Remarks: Assignment of material to §2§g_w, fowleri was
based on criteria given in the discussion of A, americanus.
§2£g_w, fowleri occurs chiefly in sandy areas, around shores

of lakes or in river valleys (Conant, 1975).

13

Bufo sp. indet.

Material: CM29796-CM29801. Two left ilia, four sacral

 

vertebrae (3-4'); three right and two left ilia, 15 sacral
vertebrae and ten fronto-parietals (4-5'); four sacral
vertebrae and six fronto-parietals (5-6'); two fronto-parietals
(6-7'); one sacral vertebra and one fronto-parietal (7-8');

two sacral vertebrae and two fronto-parietals (8-9').

Remarks: Ilia were assigned to AA£2_sp. when: (l) the
anterior or posterior portions of the prominence were missing,
thus making it impossible to examine the prominence-protuberance
relationship, or (2) when the boundaries of the dorsal
protuberance were not clearly defined within the prominence.
Tihen (1962) pointed out that the fronto-parietal is the
most reliable single element for identification of the
greatest number of New World AAAQ, I could not find any
distinct differences between the fronto—parietals of

A, americanus and A, A, fowleri.

 

Family Hylidae Hallowell
Hyla sp. indet.
\

Material: CM29802. One left ilium (3-4').

 

Remarks: Specimen is assigned to the genus Hyla based
on characters given by Holman (1962): ilium with dorsal

blade absent; dorsal prominence produced dorsolaterally, well

14

developed, usually round and smooth.
The bone is too fragmentary for specific identification,
but in comparison with Recent material, it most closely

resembles Ayla chrysoscelis and A, versicolor in the shape

 

 

of the dorsal prominence.

Family Ranidae Bonaparte
All specimens are assigned to the family Ranidae based
on the following characteristics_given by Holman (1962):
ilium with dorsal blade well developed and arising anterior
to dorsal prominence, without lateral deflection, and with

deep notch between it and dorsal acetabular expansion.

Rana sylvatica LeConte

 

Material: CM29803-CM29807. Five right and four left

 

ilia (3-4'); six right and six left ilia (4-5'); eight right
and three left ilia (5-6'); one right and three left ilia
(6-7'); two left ilia (9-10').

Remarks: Assignment to Rana sylvatica is based on

 

Holman (1967) who noted that Rana palustris LeConte,

 

A, pipiens and A, sylvatica LeConte may be distinguished

 

from A, catesbeiana Shaw and A, clamitans Latreille in that

 

 

the posterodorsal border of the ilial shaft slopes more

gently into the dorsal acetabular expansion in the former

15

group than in the latter. Furthermore, A, pipiens and

A, palustris may be separated from A, sylvatica in that the

 

 

prominence for the origin of the vastus externus head of the
triceps femoris muscles is larger, less produced, and less

roughened than in A, sylvatica. Based on this criteria, the

 

specimens are assigned to A, sylvatica.

 

In examination of nine Recent specimens of A, sylvatica

 

and two of A, palustris, the above characteristics hold in

 

separating the two species. Rana sylvatica is usually

 

encountered in or near moist wooded areas, but it often

wanders considerable distances from water (Conant, 1975).

Rana catesbeiana Shaw

 

Material: CM29808. One left ilium (4-5').

 

Remarks: This specimen is assigned to Rana catesbeiana
on the basis of characters discussed above and also on

observations of Tihen (1954) who says that A, catesbeiana

 

ilia appear to be highly sculptured. A, catesbeiana is an

 

aquatic frog that prefers larger bodies of water than most
other frogs. It is a resident of lakes, ponds, bogs and

sluggish portions of streams (Conant, 1975).

16

3222.59- indet.

Material: CM29809-CM29813. Three sacral vertebrae
(3-4'); two sacral vertebrae (4-5'); two sacral vertebrae
(5-6'); one sacral vertebra (6-7'); one sacral vertebra
(8—9').

Remarks: I am unable to differentiate these vertebrae
to species but they do have the diplasiocoelous condition
with cylindrical rather than expanded diapophyses (Holman,
1962). The sacral vertebrae more closely resemble those of
A, sylvatica in the shape of the neural canal and in their

small size.

Class Reptilia
Order Testudines Batsch
Family Emydidae

Graptemys geographica (LeSueur)

 

Material: CM29814. One pro-neural bone (4-5').

 

Remarks: Material is assigned to A, geographica based

 

on the shape, location of shield impressions and surface

sculpturing to that of Recent material. Graptemys geographica

 

occurs in large bodies of water. It prefers rivers rather

than creeks, and lakes rather than ponds (Conant, 1975).

17

Order Squamata
Suborder Sauria
Family Scincidae

Eumeces fasciatus (Linnaeus)

 

Material: CM29815-CM29817. One pre-caudal vertebra
(3-4'); three pre-caudal vertebrae (4-5'); one pre-caudal
vertebra (6-7').

Remarks: A, fasciatus has a more backswept neural

 

spine than A, laticeps. Eumeces fasciatus lives in rock

 

piles and decaying debris in or near woods. The habitat is

usually damp (Conant, 1975).

Suborder Serpentes Linnaeus
Family Viperidae

Crotalus horridus Linnaeus

 

Material: CM29818-CM29824. Twenty-seven vertebrae
(3-4'); 50 vertebrae (4-5'); 27 vertebrae (5-6'); nine
vertebrae (6-7'); four vertebrae (7-8'); two vertebrae
(8-9'); two vertebrae (9-10').

Remarks: Holman (1963) gives characters to differentiate

Crotalus from Agkistrodon. In Agkistrodon a distinct pit

 

 

usually occurs on either side of the cotyle of the centrum.
Each of these pits contains one moderately large fossa. In

Crotalus the distinct pits are usually absent and the one or

18

more fossae that occur on either side of the cotyle of the

centrum are minute. In addition, Crotalus horridus has a

 

lower neural spine than either A, adamanteus or Agkistrodon

 

 

piscivorus (Holman, 1967). The material most closely

 

resembles Crotalus horridus in these characters and is

 

therefore assigned to it. Crotalus horridus lives in

 

timbered terrain; usually it is common in second-growth

where rodents abound (Conant, 1975).

Family Colubridae
Subfamily Xenodontinae
Holman (1973b) states that members of this subfamily
lack hypapophyses on their lumbar vertebrae and have
depressed vertebral neural arches and wide vertebral hemal
keels. Based on this criteria the material is assigned to

this subfamily.

Heterodon platyrhinos Latreille

 

Material: CM29825-CM29827. One pre-caudal vertebra
(5—6'); one pre-caudal vertebra (6-7'); one pre-caudal
vertebra (8-9').

Remarks: Holman (1962) states that the genus

Heterodon Latreille may be diagnosed by the following strong

 

characters: hypapophyses absent; vertebrae wider than long

19

through zygapophyses; neural arch flat; neural spine longer
than high, usually thickened dorsally, and with its anterior
and posterior borders concave; prezygapophyseal processes
large, pointed or truncated; epizygapophyseal spines absent;
hemal keel very broad and indistinct on many thoracic
vertebrae.

In addition, Holman (1963) was able to differentiate

between A, platyrhinos and A, nasicus in that in the former,

 

the anterior zygapophyseal faces are more elongate, and in
dorsal view, their anterior margins are much flatter than in
the latter species. This material best compares to the

characteristics of A, platyrhinos. Heterodon platyrhinos

 

 

lives in sandy areas (Conant, 1975).

Subfamily Colubrinae

Colubrinae never bear lumbar hypapophyses as in the

 

subfamily Natricine, and they lack the combination of the

 

depressed neural arch and the very wide hemal keel of the

Xenodontinae (Holman, 1973b).

 

Diadophis punctatus (Linnaeus)

 

Material: CM29828-CM29830. Twenty-six pre-caudal

 

vertebrae (3-4'); 19 pre-caudal vertebrae (4-5'); six

pre-caudal vertebrae (5-6').

20

Remarks: Holman (1967) gives characters to distinguish

between the vertebrae of Diadophis punctatus and Carphophis

 

 

amoenus. He states that the most consistent way to tell the
two forms apart is that the neural spine is higher, thicker,
and usually with more of a posterior overhang in the former
than in the latter species. The above fossils more closely

resemble those of A, punctatus in these characters.

 

Diadophis punctatus is a woodland snake, usually most common

 

in cutover areas that include an abundance of hiding places
in the form of stones, logs, bark slabs, or other rotting
wood. Rocky, wooded hillsides are also favored (Conant,

1975).

Carphophis amoenus (Say)

 

Material: CM29831-CM29832. Sixteen pre-caudal
vertebrae (3-4'); seven pre-caudal vertebrae (4-5').
Remarks: Fossils are assigned to A, amoenus based on

criteria discussed under Diadophis punctatus. Carphophis

 

 

amoenus is partial to moist earth and disappears deep

underground in dry weather (Conant, 1975).

Coluber or Masticophis Linnaeus

 

Material: CM29833-CM29837. One pre-caudal vertebra

(4-5'); three pre-caudal vertebrae (5—6'); seven pre-caudal

21

vertebrae (6-7'); one pre-caudal vertebra (7-8'); three
pre-caudal vertebrae (8-9').

Remarks: Holman (1962) gives characteristics of the
lumbar vertebrae for the genus Coluber: hypapophyses
absent; vertebrae longer than wide through zygapophyses;
neural arch vaulted; neural spine about as high as long, thin
and delicate, not beveled anteriorly; epizygapophyseal spines
usually well developed; hemal keel narrow throughout.

The vertebrae of Coluber, Masticophis and Opheodry§_are

 

 

similar in that they are elongate and the neural spine is
thin and delicate. But the former two genera differ in that
they are larger, the neural spine is higher, and a well
developed epizygapophyseal spine is almost always present.
The fossils resemble the characters of the former two genera
in this respect. Based on the present geographic ranges of

Masticophis and Coluber it would appear that the fossils

 

represent Coluber, but I am unable to separate the two

genera on vertebral remains.

Lampropeltis Fitzinger

 

Remarks: The vertebrae of Pituophis, Elaphe and

Lampropeltis are very similar but have been separated on the

 

basis of characters of Holman (1965). Pituophis is distinct

 

from the other two genera in having a higher neural spine

22

with an indented anterior edge. The genera Elaphe and
Lampropeltis can be separated by the more depressed neural

arch of the latter. The fossils resemble Lampropeltis in

 

this respect.

Lampropeltis triangulum (Lacepede)

Material: CM29838-CM29843. Eight pre-caudal vertebrae
(3—4'); 33 pre-caudal vertebrae (4-5'); 11 pre-caudal
vertebrae (5-6'); five pre-caudal vertebrae (6-7'); one
pre-caudal vertebra (7-8'); three pre-caudal vertebrae
(8-9').

Remarks: Fossils are assigned to A, triangulum because

 

the vertebrae possess lower neural spines than those of

A, getulus. Also A, getulus vertebrae are quite robust
with thick neural spines and neural arches, and the hemal
keels and sub-central ridges are usually quite strong, with

the valleys between them quite deep (Holman, 1965).

Lampropeltis getulus (Linnaeus)

Material: CM29844-CM29848. One pre-caudal vertebra
(3-4'); two pre-caudal vertebrae (4-5'); six pre-caudal
vertebrae (5-6'); two pre-caudal vertebrae (8-9'); two
pre-caudal vertebrae (9-10').

Remarks: The fossils have been assigned to A, getulus

23

based on criteria discussed under A, triangulum. A, getulus

 

occurs regionally, but has not been recorded in the immediate
area. It is possible that Recent A, getulus may be collected

in the area in the future.

Elaphe sp. indet.

Material: CM29849. Three pre-caudal vertebrae (6-7').
Remarks: Material is assigned to the genus Elaphe sp.
indet. because they have a more vaulted neural arch than

Lampropeltis, but less vaulted than Pituophis (Holman,

 

 

1973a). In addition, Pituophis exhibits strongly developed

 

epizygapophyseal spines which are lacking in the fossils
(Auffenberg, 1963).
The material is too fragmentary to assign it to the

specific level.

Subfamily Natricine

Material is assigned to the subfamily Natricine on

 

characters given by Holman (1973b): hypapophyses on their
lumbar vertebrae, and Auffenberg (1965): epizygapophyseal

spines are usually present.

24

Natrix sipedon (Linnaeus)

 

Material: CM29850-CM29852. Two pre-caudal vertebrae

 

(3-4'); one pre-caudal vertebra (4-5'); three pre-caudal
vertebrae (5-6').
Remarks: Brattstrom (1967) discusses criteria used in

distinguishing Natrix vertebrae from Thamngphis. In general,

 

Thamnophis vertebrae are elongate when viewed from above,

 

while Natrix vertebrae are almost square. In addition,
Natrix vertebrae tend to have higher neural spines (Holman,
1962). The fossil material most closely resembles Natrix
in these characters.

Natrix septemvittata and A, sipedon occur in the area

 

today. A, septemvittata possesses a long, low neural spine

 

and A, sipedon possesses a much higher one (Auffenberg, 1963).

The fossils resemble the latter in this respect.

Natrix sp. indet.

Material: CM29853. One pre-caudal vertebra (5-6').

 

Remarks: The fossil is too fragmentary for specific
identification but genera was determined using criteria

discussed under Natrix sipedon.

 

Thamnophis sauritus (Linnaeus)

 

Material: CM29854-CM29856. One pre-caudal vertebra

 

25

(3-4'); five pre-caudal vertebrae (4-5'); one pre-caudal
vertebra (8-9').

Remarks: Criteria for assignment to Thamnophis was

 

discussed under Natrix sipedon. Material is assigned to

 

A. sauritus in that the accessory processes are oblique
to the longitudinal axis of the centrum; in A, sirtalis the
accessory processes are at right angles to the longitudinal

axis of the centrum (Holman, 1962).

Thamnophis sirtalis (Linnaeus)

 

Material: CM29857-CM29860. Eleven pre-caudal
vertebrae (3-4'); six pre-caudal vertebrae (4-5'); two
pre-caudal vertebrae (5-6'); one pre-caudal vertebra (6-7').

Remarks: The fossils were assigned to A, sirtalis

based on criteria discussed under A, sauritus.

Thamnophis sp. indet.

 

Material: CM29861-CM29866. Fifteen pre-caudal
vertebrae (3-4'); 20 pre-caudal vertebrae (4-5'); three
pre-caudal vertebrae (5-6'); two pre-caudal vertebrae (6-7');
one pre-caudal vertebra (8-9'); one pre-caudal vertebra
(9-10').

Remarks: The material was too fragmentary for specific

identification but could be assigned to genus based on

26

characters discussed under Natrix sipedon.

 

DISCUSSION

The Baker Bluff Cave fauna consists of at least five
species of urodeles, five species of anurans, one species
of turtle, one species of lizard and 11 species of snakes.

All of these forms, as far as can be determined, are

living in the area today. Only Lampropeltis getulus, which

 

occurs regionally, is not found in the immediate area today.

Perhaps the most striking thing about the herpetofauna
is that there is nothing that strongly indicates that the
climate or topography was any different than it is today in
northeastern Tennessee. In addition, inspection of Tables
I and II giving minimum numbers of individuals from each
level show no discernible trends in the herpetofauna that
indicate that climatic or ecological conditions changed
markedly from approximately 20,000 years B.P. to approximately
600 years B.P.

The Baker Bluff Cave Herpetofauna exhibits many
similarities to the Late Pleistocene Herpetofauna from
Ladds, Georgia, located in the northwestern part of the
state (Holman, 1967). At least ten species, mostly snakes,
from the Ladds, Georgia, site are also present at Baker

Bluff Cave.

27

28

The Baker Bluff Cave fauna is indicative of four
major ecological preferences (Table III). (1) A permanent

aquatic habitat based on the evidence of Rana'catesbeiana

 

and Graptemys geogrgphica; (2) a marsh-stream border

 

situation indicated by the water snakes Natrix and

Thamnophis and the toad Bufo A, fowleri; (3) an Open, sandy

 

area indicated by Heterodon platyrhinos; and (4) a moist

 

woodland habitat where the majority of the identified

Colubrinae, Crotalus horridus, Ambystoma opacum and

 

 

A, maculatum lived.

 

The diversity of habitats exhibited by the
herpetofauna strongly indicates that the fossil remains
were probably deposited by raptors, most likely owls. As a
result of this type of accumulation, it is impossible to
conclude what the exact nature of the environment was in
immediate proximity to the cave site, but the herpetological
remains show that the four major habitats discussed were

present in the area.

29

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE I
MINIMUM NUMBER OF INDIVIDUALS
AMPHIBIANS
Species 3-4' 4-5' 5-6' 6-7' 7-8' 8-9' 9-10' Total
Ambystoma Qpacum 0 1 l l 0 1 0 4
Ambystoma maculatum 0 l l l 0 0 0 3
Ambystoma sp. 1 l l 1 l l 1 7
Necturus maculosus 1 l l l l l 0 6
A, alleganiensis l l l 1 l 0 1 6
Desmggnathus sp. 1 1 l l 1 l 0 6
Bufo americanus 12 24 10 7 2 2 3 60
Bufo A, fowleri 3 3 l 0 0 0 0 7
Bufo sp. 4 15 4 l 1 2 0 27
Hyla sp. 1 0 0 0 0 0 0 1
Rana §ylvatica 5 6 8 3 0 0 2 24
Rana catesbeiana 0 l 0 0 0 0 0 1
Rana sp. 3 2 2 l 0 1 0 9
TABLE II
MINIMUM NUMBER OF INDIVIDUALS
REPTILES
Species 3-4' 4—5' 5-6' 6-7' 7-8' 8-9' 9-10' Total
Graptemys geographica 0 1 0 0 0 0 0 l
Eumeces fasciatus l l 0 l 0 0 0 3
Crotalus horridus l l l 1 l l l 7
Heterodon platyrhinos 0 0 l l 0 l 0 3
Diadophis punctatus l l 1 0 0 0 0 3
Cagphophis amoenus l l 0 0 0 0 0 2
Coluber or Masticophis 0 1 l 1 1 l 0 5
A, triangulum l l l l 1 l 0 6
A, getulus l l l 0 0 l l 5
Elaphe sp. 0 0 0 l 0 0 0 1
A, sipedon l l l 0 0 0 0 3
Natrix sp. 0 0 l 0 0 0 0 l
Thamnophis sauritus 1 l 0 0 0 l 0 3
Thamnophis sirtalis l l l 1 0 0 0 4
Thamnophis sp. 1 l l l 0 l l 6

 

 

30

TABLE III
ECOLOGICAL PREFERENCE CHART

 

Permanent
Aquatic

Apecies

Marsh-
Stream
Border

Moist Open
Woodland Wooded

 

Ambystoma opacum

Ambystoma maculatum

Ambystoma sp.

Necturus maculosus X
Cryptobranchus alleganiensis X
Desmognathus X
Bufo americanus

fowleri

 

 

 

 

 

 

 

 

E'p .
sp.
sylvatica

catesbeiana X
sp.

Graptemys geographica X
Eumeces fasciatus

Crotalus horridus

Heterodon platyrhinos

Diadophis ppnctatus

Carphophis amoenus

Coluber or Masticophis
Lamprgpeltis triangulum
Lampropeltis getulus

Elaphe sp.

Natrix sipedon

Natrix sp.
Thamngphis
Thamnophis
Thamnophis

 

 

 

wwwm
OJDJDJ

:3be
WWW!»

 

 

 

 

 

 

 

 

 

 

sauritus
sirtalis

sp.

 

 

 

Ecological preferences from Holman

X
X

(1972) and Conant (1975).

LITERATURE CITED

LITERATURE CITED

Auffenberg, W. 1963. The fossil snakes of Florida. Tulane
Studies in Zoology 10(3):129-216.

Brattstrom, B. H. 1967. A succession of Pliocene and
Pleistocene snake faunas from the High Plains of the
United States. Copeia 1967(1):188-202.

Conant, R. 1975. A Field Guide to the Reptiles and
Amphibians of Eastern and Central North America.
Houghton Mifflin Co., Boston.

Dice, L. R. 1943. The Biotic Provinces of North America.
Univ. Michigan Press, Ann Arbor: 1-78.

Holman, J. A. 1962. A Texas Pleistocene Herpetofauna.
Copeia l962(2):255-26l.

. 1963. Late Pleistocene amphibians and reptiles
of the Clear Creek and Ben Franklin Local Faunas of
Texas. Jour. Grad. Res. Center 31(3):152-167.

. 1965. A Late Pleistocene Herpetofauna from
Missouri. Trans. Ill. Acad. Sci. 58(3):l90-l94.

 

. 1967. A Pleistocene Herpetofauna from Ladds,
Georgia. Bull. Georgia Acad. Sci. 25(3):154-166.

. 1968. Lower Oligocene Amphibians from Saskatchewan.
Quart. Jour. Fla. Acad. Sci. 31(4):273-289.

. 1969. Herpetofauna of the Pleistocene Slaton
Local Fauna of Texas. The Southwestern Naturalist
14(2):203-212.

. 1972. Herpetofauna of the Kanopolis Local Fauna
(Pleistocene:Yarmouth) of Kansas. Mich. Academician
5(1):87-98.

 

. 1973a. A new Pliocene snake, genus Elaphe, from
Oklahoma. Copeia 1973(3):574-580.

 

31

32

. 1973b. Reptiles of the Egelhoff Local Fauna
(Upper Miocene) of Nebraska. Contrib. Mus. Paleontol.,
Univer. Mich. 24(12):125-l34.

Meszoely, C. A. M. 1967. A new cryptobranchid salamander
from the Early Eocene of Wyoming. Copeia 2:346-349.

Soler, E. I. 1950. On the status of the family Desmognathidae.
Univ. Kansas Sci. Bull. 33(12):459-480.

Tihen, J. A. 1954. A Kansas Pleistocene Herpetofauna.
Copeia l954(3):217-221.

. 1958. Comments on the osteology and phylogeny of.
ambystomatid salamanders. Bull. Fla. State Mus. 3:1-50.

. 1962. A review of new world fossil bufonids.
Amer. Mid. Nat. 68(1):l-50.

.Illlllll II. Ill 10“

I- .1 L
1 if ilk!!! 1‘

(11"?! ‘1 ti‘l‘l‘. I