• TABLE OF CONTENTS
HIDE
 Front Cover
 Main
 Back Cover






Group Title: Bulletin of the Florida State Museum
Title: The colubrid snake genus Thamnophis
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00001515/00001
 Material Information
Title: The colubrid snake genus Thamnophis a revision of the Sauritus group
Series Title: Bulletin of the Florida State Museum
Physical Description: 100-178 p. : illus., maps, tables. ; 23 cm.
Language: English
Creator: Rossman, Douglas Athon, 1936-
Publisher: University of Florida
Place of Publication: Gainesville
Publication Date: 1963
 Subjects
Subject: Garter snakes   ( lcsh )
Genre: bibliography   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
non-fiction   ( marcgt )
 Notes
Bibliography: "Literature cited:" p. 171-178.
General Note: Cover title.
General Note: Revised version of thesis, University of Florida.
 Record Information
Bibliographic ID: UF00001515
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: ltqf - AAA0826
notis - ACK0907
alephbibnum - 000440441
oclc - 05067190
lccn - a 63007825

Table of Contents
    Front Cover
        Page 97
        Page 98
    Main
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
    Back Cover
        Page 179
Full Text





BULLETIN

OF THE

FLORIDA STATE MUSEUM


BIOLOGICAL SCIENCES


Volume 7


Number 3


THE COLUBRID SNAKE GENUS THAMNOPHIS:
A REVISION OF THE SAURITUS GROUP

Douglas A. Rossman


UNIVERSITY OF


FLORIDA


Gainesville
1963










Numbers of the BULLETIN OF THE FLORIDA STATE MUSEUM are pub-
lished at irregular intervals. Volumes contain about 300 pages and are not nec-
essarily completed in any one calendar year.















WILLIAM J. RIEMER, Managing Editor
OLIVER L. AuSTIN, JR., Editor


Consultants for this issue:
Phillip W. Smith
Charles C. Carpenter





















Communications concerning purchase or exchange of the publication and all man-
uscripts should be addressed to the Managing Editor of the Bulletin, Florida State
Museum, Seagle Building, Gainesville, Florida.


Price for this issue $1.05


Published 10 June 1963










THE COLUBRID SNAKE GENUS THAMNOPHIS:
A REVISION OF THE SAURITUS GROUP


DOUGLAS A. ROSSMAN1







SYNOPSIS: The Sauritus group of the gartersnakes is composed of two polytypic
species, Thamnophis sauritus (Linnaeus) and Thamnophis proximus (Say), com-
monly called the ribbonsnakes.
Both species vary geographically in number of ventrals, number of subcaudals,
relative tail length, and several features of the color pattern. Supralabial number
is subject to geographic variation in T. sauritus but not in T. proximus. Most of
the geographic variation is discordant, transition zones in one character seldom
corresponding with those of other characters. As color pattern fluctuates less
locally than do the meristic characters, it is more reliable in defining geographic
races.
Four subspecies of T. sauritus are recognized, two of them previously un-
described (T. s. septentrionalis of the Great Lakes region, northern New England,
and Nova Scotia; T. s. nitae of northwestern peninsular Florida). Of six races of
T.proximus four are new (T. p. orarius of coastal Louisiana, Texas, and northern
S Tamaulipas; T. p. rubrilineatus of the Edwards Plateau region of central Texas;
T. p. diabolicus of New Mexico, trans-Pecos Texas, Coahuila, and Nuevo Leon;
T. p. alpinus of the Chiapas Highlands in southern Mexico).
The evolutionary history of this species group and the trends within each
S species are discussed. Members of the group are highly specialized and occupy
S an advanced position within the genus.









xThe author, Instructor in Zoology at the University of North Carolina. Chapel
Hill, completed this study while a graduate student at the University of Florida,
Gainesville. An earlier version formed his doctoral dissertation, which was
accepted by the Graduate School in August 1961. Manuscript submitted 18
January 1962.-ED.

Rossman, Douglas A. 1963. The colubrid snake genus Thamnophis: A revision
of the Sauritus group. Bull. Florida State Mus., vol. 7, no. 3, pp. 99-178.






v. m7


100 BULLETIN OF THE FLORIDA STATE MUSEUM Vol. 7



TABLE OF CONTENTS

Introduction ................. ...................... ........... 101
Acknowledgments ........................................... 101
Materials and methods .......................................... 102
The Sauritus group .................. ....................... 104
Definition .................. ................ .............. 104
Morphology ................. ....... ........................... 104
Relationships .................. ......... ................ 107
Key to the species of the Sauritus group ................. ......... 108
The WesterrfRibbonsnake ......................... .... .. . 109
Variation ............... .. ....... .. .. ........... ......... 111
Sexual dimorphism ........................... ................. 111
Ontogenetic variation ................. .... ... .............. 114
Individual variation ......... ....... ............. ........... 114
Geographic variation ........................................ 117
-Taxonomy ......................... ........... ............. 128
Subspecies of Thamnophis proximus ............................ 130
Thamnophis proximus proximus .............................. 131
Thamnophis proximus orarius .............................. 132
Thamnophis proximus rubrilineatus ............................ 134
Thamnophis proximus'diabolicus ...... ....................... 135
Thamnophis proximus rutiloris ............................... 138
Thamnophis proximus alpinus ................ ........... 140
,Natural history ........... ............... ................... 142
-The Eastern Ribbonsnake ...................................... 145
-Variation ............................................... .. 147
Sexual dimorphism ............................ ............. 147
Ontogenetic variation ............... ...................... 148
Individual variation ........................................ 149
Geographic variation ............................ ............ 150
-Taxonomy ............................. ............ ........ . 157
Subspecies of Thamnophis sauritus ............................. 157
Thamnophis sauritus sauritus ................ ............... 157
Thamnophis sauritus septentrionalis .......................... 159
Thamnophis sauritus sackenii ................................ 161
Thamnophis sauritus nitae ..... ............................. 163
-Natural history ............... ..... ............ ........... 166
Conclusions ......... .......... .. .. ........... .'. ......... 168
Literature cited ............... ................ ........... 171







ROSSMAN: GENUS THAMNOPHIS


INTRODUCTION

More than half a century has elapsed since Ruthven (1908) produced
his classic monograph of the gartersnake genus Thamnophis, a monu-
mental work which has served as a model for most subsequent studies
of geographic variation in ophidians. In that study he arranged the 12
species of Thamnophis then recognized into two primary phylogenetic
divisions, each containing two species groups which he named for the
"best known" species in each. Thus, his Division I comprised the Radix
group and the Sauritus group, and his Division II the Elegans group
and the Sirtalis group.
Ruthven based his revision on a study of approximately 3000 speci-
mens. While impressive for that time, this sample was hardly adequate
to show detailed variation when divided among the 12 species he
recognized; moreover it apparently contained specimens of five or six
additional species of Thamnophis then unrecognized. As additional
material accumulated in American museums during the first half of
the 20th century, more refined analyses of geographic variation be-
came possible for many species (Fitch, 1940; Mittleman, 1949; A. G.
Smith, 1949; and Milstead, 1953 for example). In the recent flurry of
activity by students of gartersnake variation the Sauritus group has
been overlooked except for- a few brief and scattered notes buried in
papers on other subjects. This general lack of interest in the ribbon-
snakes is surprising when one considers their abundance and their
extensive geographic range-Nova Scotia and the Great Plains to
Costa Rica-and the large amount of material available for study-
now more than 2400 specimens. The present investigation was under-
taken to provide a more adequate definition of the Sauritus group as
a possible basis for future comparative studies of all members of the
genus Thamnophis (of which T. sauritus is the type species) by an-
alyzing individual and geographic variation in the two species com-
prising it.

ACKNOWLEDGMENTS
I am grateful to the following persons for the loan or gift of specimens,
for information or advice, or for other assistance during the course of
this study: K. K. Adler, R. M. Bailey, R. J. Baldauf, J. A. Bartley, J. F.
Branham, R. Conant, J. R. Dixon, F. L. Downs, W. E. Duellman,
M. J. Fouquette, Jr., F. R. Gehlbach; H. C. Gerhardt, S. D. Lee, J. C.
List, B. McBride, S. A. Minton, Jr., F. E. Potter, Jr., J. A. Quinby,
N. J. Rossman, B. Rothman, N. Rothman, M. Sabath, C. R. Shoop,
P. W. Smith, L. C. Stuart, D. W. Tinkle, T. M. Uzzell, Jr., R. D.


1963







BULLETIN OF THE FLORIDA STATE MUSEUM


Worthington, the curators of the institutions cited below, and the
many helpful friends who cannot be mentioned individually because
of space limitations. T. T. Allen prepared the drawings of the ribbon-
-snake heads, figure 1.
Special thanks for their constant encouragement and ready advice
are due A. B. Grobman, under whose guidance this study was initiated,
and A. F. Carr and W. J. Riemer, who supervised its completion.

MATERIALS AND METHODS

Although it was impossible to examine every specimen of Thamnophis
sauritus and T. proximus in North American collections, the sample
studied included more than 2400 specimens from all parts of the
ranges of both species from the following institutional collections:


AMNH American Museum of
Natural History
ANSP Academy of Natural
Sciences of Philadel-
phia
CAS Chicago Academy of
Sciences
CM Carnegie Museum
CMNH Cincinnati Museum of
Natural History
CNHM Chicago Natural History
Museum
CUM University of Colorado
Museum
INHS Illinois State Natural His-
tory Survey
KU University of Kansas
Museum of Natural
History
LSUMZ Louisiana State Univer-
sity Museum of
Zoology
MCZ Museum of Comparative
Zoology
MGFCM Mississippi Game and
Fish Commission
Museum
MPM Milwaukee Public
Museum
MSC Mississippi Southern
College


NMC National Museum of
Canada
NSMS Nova Scotia Museum of
Science
NYSM New York State Museum
OSM Ohio State Museum
TCWC Texas Cooperative Wild-
life Collection
TNHC University of Texas Nat-
ural History Collection
TTC Texas Technological
College
TU Tulane University
UADZ University of Arkansas
Department of Zoology
UF University of Florida
Collections
UG University of Georgia
UIMNH University of Illinois
Museum of Natural
History
UK University of Kentucky
UMMZ University of Michigan
Museum of Zoology
USCM University of South Car-
olina Museum
USNM United States National
Museum
UV University of Vermont


In addition to preserved material, I have been fortunate in having
available for examination live ribbonsnakes of all the subspecies rec-


102'


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


ognized in this study from many parts of the United States and Mexico.
These animals revealed several distinctive and diagnostic color-pattern
characters not apparent in preserved specimens. A small series of
articulated and disarticulated skeletons was prepared for comparative
purposes, and the left maxilla was removed from a number of the
preserved specimens for tooth counts. Sex was determined by dissec-
tion, except when males had the hemipenes everted.
The following measurements of cephalic scales were made: parietal
length, frontal length and width, muzzle length, internasal-rostral
suture length, and nasal-rostral suture length. Muzzle length, as used
herein, is equivalent to the combined length of one prefrontal and the
adjacent internasal when measured along the middorsal suture. These
measurements are the basis of the following ratios: frontal length/
parietal length; frontal width/frontal length; muzzle length/frontal
length; internasal-rostral contact/nasal-rostral contact. Measurements
of head length, body length (including the head), and tail length
permitted calculation of head length/body length and tail length/
total length ratios. Absence or distortion of the tail in many specimens
reduced the number of tail length/total length values obtained. A
few ratios (frontal length/parietal length and frontal width/frontal
length) that showed no significant variation are not discussed further.
Standard counts were made of such meristic characters as number
of supralabials and infralabials, preoculars and postoculars, temporals,
intergenials, ventrals, and subcaudals, as well as of any anomalies
occurring in these characters. The number of dorsal scale rows was
determined at three points on the body: one head length behind the
head, at midbody, and above the last ventral scute. The point of
dorsal scale row reduction oni the body was determined for a small
number of specimens; caudodorsal reduction in a few individuals.
The number of teeth on each of the dentigerous bones was noted,
empty sockets being included in the total count.
The state of certain elements of pattern that vary significantly in
the Sauritus group was observed in each specimen: relative size and
shape of the light-colored parietal spots, extent of the light spot on
the postoculars, presence or absence of a black postorbital vitta, width
of the vertebral stripe, presence or absence of a black paravertebral
stripe, width of the light lateral stripe, degree of spotting between the
vertebral and lateral stripes, and width of the dark ventrolateral
stripe occupying the lateral margin of the ventral scutes on each side.
Color notes were made on living specimens, with consistent atten-
tion to the labials, vertebral and lateral stripes, the dorsum, and the
venter. In most cases these evaluations were subjective, but, with only


1963








104 BULLETIN OF THE FLORIDA STATE MUSEUM


two exceptions, at least one specimen of each geographic race was
compared with a color atlas (Maerz and Paul, 1950). All color descrip-
tions in this study are based on freshly killed, freshly shed specimens,
unless otherwise stated.

THE SAURITUS GROUP
As Thamnophis proximus, long thought to be a subspecies of T. sauri-
tus, is a valid species (Rossman, 1962), the Sauritus group, long
recognized as distinct from the other gartersnakes, is now known to be
composed of two closely related species. Holbrook (1842) placed the
species sauritus (including the unrecognized proximus) in Leptophis,
a genus of slender, semiarboreal colubrids; the other gartersnake
species,were not thus allocated. Fitzinger (1843) erected the genus
Thamnophis, in which later authors placed all gartersnakes, solely to
accommodate the ribbonsnake. Even as late as 1893, Dumeril and
Bocourt included only sauritus in Thamnophis, allocating the other
species to Eutaenia. While Ruthven (1908) placed sauritus in Tham-
nophis, he acknowledged its distinctive nature by designating a species
group solely fr that species.

DEFINITION OF THE SAURITUS GROUP
These are snakes of the genus Thamnophis possessing a small hemi-
penis that extends to the 7th or 8th subcaudal in the inverted position.
The teeth are numerous, usually 29 to 32 on each maxilla, 20 to 22
on each palatine, 33 to 36 on each pterygoid, and $6 to 38 on each
dentary. The basal portion of the tongue is red; the forked tips are
black. The tail is long, usually constituting 27 percent or more of the
total length of adults. The maximum number of dorsal scale rows
rarely exceeds 19. The ventrals range from 141 to 181 (both extremes
in T. proximus), the subcaudals from 82 (T. proximus) to 136 (T.
sauritus). In both species the males have a slightly greater number of
scutes than the females. The lateral stripe is usually confined to dorsal
scale rows 3 and 4. The supralabials always lack vertical black bars.

MORPHOLOGY
FORM AND PROPORTIONS. Both species are relatively slender, long-tailed
snakes in comparison with other members of the genus. Thamnophis
proximus may exceed 1250 mm. in total length, equaled among the
gartersnakes by Thamnophis sirtalis and surpassed only by Thamno-
phis elegans gigas. T. sauritus is smaller; the largest adult examined
measured only slightly over one meter. The relatively narrow head


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


makes up 3.5 to 4.6 (mean 4.1) percent of the body length in adults.
The muzzle is narrow, apparently reflecting the. semiaquatic habits
of these two species (Fitch, 1940), the internasal-rostral contact/nasal-
rostral contact ratio exceeding 1.00 in only one of 29 adults examined
for this character (mean 0.82). The tail is longer than in any other
species of Thamnophis and constitutes 25.2 to 33.6 percent of the
total length in adult T. proximus, and 28.8 to 38.8 percent in adult
T. sauritus. Both species show sexual dimorphism in proportionate tail
length.
OsTEOLOGY. Hypapophyses are present throughout the length of the
vertebral column, a condition generally thought to be characteristic
of natricine snakes (Malnate, 1961, however, has described a new
genus apparently closely related to the natricine genus Amphiesma
but lacking hypapophyses on the posterior trunk vertebrae). The
skull is of the generalized colubrid type and similar to those of other
species in the genus; although certain differences do exist, these must
be analyzed by a comparative study of the skulls of all the species of
Thamnophis, which is beyond the scope of the present investigation.
Members of the Sauritus group have more teeth on each of the
dentigerous bones than any other species in the genus for which data
are available. The number of maxillary teeth ranges from 27 to 84,
palatine from 17 to 25, pterygoid from 26 to 38, and dentary from
31 to 40. The means for T. proximus and T. sauritus are presented
in table 1. As no sexual dimorphism in tooth number is apparent, the
counts for both sexes are combined. Maxillary counts were taken both

TABLE 1. Meristic dentitional variation in the Sauritus group
(N = number of counts; M = mean; R = range)

T. proximus T. sauritus

N M R N M R

Maxilla..................... 37 30.0 27-34 49 30.4 27-34
Palatine..................... 10 19.7 17-22 37 21.0 18-25
Pterygoid. ................... 11 32.7 26-37 38 34.3 28-38
Dentary..................... 13 33.8 31-36 36 37.1 33-40

from preserved specimens (only the left maxilla examined) and
from skeletons (both maxillae examined); all other counts were taken
from skeletal material only. Because the dentigerous bones are paired
the number of individual counts differs from the number of individual
snakes examined. The number of teeth on the two bones of any given


1963








106 BULLETIN OF THE FLORIDA STATE MUSEUM


pair often vary;, the same number occurred on both maxillae in only
43.5 percent of the specimens examined, on both palatines in 39.1
percent, on both pterygoids in 30.4 percent, and on both dentaries
in 28.6 percent. Differences in tooth count between the two members
of a given pair range from 1 to 3 on the maxillae, 1 to 3 on the pala-
tines, 1 to 7 on the pterygoids, and 1 to 2 on the dentaries.
Examination of the maxillae of western ribbonsnakes from Nica-
ragua, Chiapas, Veracruz, Tamaulipas, Texas, and Louisiana revealed
no geographic variation in number of teeth. This may not prevail,
however, when information on the more northern populations becomes
available. Too few of the other dentigerous bones have been available
to study geographic trends in their tooth counts. Southern T. sauritus
do appear to have more maxillary teeth than their northern counter-
parts; 38 counts from a peninsular Florida series range from 28 to 34
(mean 31.1) while 13 counts from Pennsylvania ribbonsnakes range
from 27 to 30 (mean 28.5). More data are needed to determine the
exact nature and extent of this variation. The other toothed bones
are represented by counts from Florida specimens only.
Maxillary teeth increase slightly in length posteriorly; the last three
teeth are abruptly enlarged. Teeth on the other three pairs of bones
decrease slightly in length posteriorly.
HEMIPENIS. The hemipenes of T. proximus and T. sauritus are vir-
tually identical in all respects and are the smallest known for the
genus (information is not available for many species). The inverted
hemipenis usually extends to the level of the eighth subcaudal (see

TABLE 2. Hemipenis length in the Sauritus group

Hemipenis extends Hemipenis extends
Species to subcaudal Species to subcaudal

6(11)* 6(2)
T. proximus....... 7(15) T. sauritus......... 7
8(16) 8(10)
9
Figures in parentheaee in this and subsequent tables indicate the number of individuals when
there is more the one.

table 2 for individual variation); the m. retractor penis magnus is
undivided and inserts at the level of the 28th subcaudal. The everted
organ is single and bears a simple straight sulcus spermaticus which
terminates at the apex. Numerous rows of small spines occupy the
distal half of the ornamented portion of the hemipenis, the margins


Vol. 7







ROSSMAN: GENUS THAMNOPHIS


of the sulcus, and the extreme basal area. Those in the latter region
are proximal to the two pairs of enlarged basal hooks (one pair on
each side of the sulcus). Laterad to each pair of basal hooks is an
unornamented area composed of nude patches separated from each
other and from the many distal rows of small spines by several oblique
rows of large spines which arise on the medial surface of the organ
near the basal hooks and converge on the lateral surface. The most
proximal of these spines are almost half as large as the basal hooks.
The apical region of the hemipenis lacks ornamentation.
SCUTELLATION. The most distinctive feature in the scutellation of the
Sauritus group is the unusually large number of subcaudals in both
species. These range from a minimum of 82 in T. proximus to a maxi-
mum df 136 in T. sauritus, almost always exceeding 100 in T. sauritus,
less frequently in T. proximus. Only rarely do other species in the
genus have more than 100 subcaudals, and usually the number is
much less. Other important scale characters include: a single pre-
ocular; 2 supralabials entering the orbit; a maximum qf 19 dorsal scale
rows in 98.6 percent of the T. proximus and 99.8 percent of the T.
sauritus examined; all dorsal scales strongly keeled, those of the outer-
most row not greatly enlarged.
COLORATION. Both T. proximus and T. sauritus lack vertical black bars
on the labials and black markings on the venter, though in all T.
sauritus and in certain populations of T. proximus brown pigment
extends onto the lateral margins of the ventral scutes to form a dark
ventrolateral stripe. A light lateral stripe is always present on dorsal
scale rows 3 and 4 (confined to row 3 posteriorly in one race of T.
proximus and including row 2 anteriorly in northern populations of
T. sauritus), but the vertebral stripe is frequently lacking in some
geographic areas. The lateral stripe is usually a shade of yellow, white,
or blue, the vertebral stripe a shade of yellow, orange, red, or brown.
The color between the stripes ranges from olive-gray and light brown
to black. On each preocular is a broad, vertical yellow bar.

RELATIONSHIPS
T. proximus and T. sauritus form a closely knit species group which
shows no particularly close affinities to the other species complexes in
the genus. Though Ruthven (1908) suggested a relationship to the
Radix group-more specifically to Thamnqphis eques-this can be
evaluated properly only by a comprehensive study of the interspecific
relationships of the entire genus. For the present we can say that the
relatively long tail, extreme reduction in spotting of the dorsum, and


1963








108 BULLETIN OF THE FLORIDA STATE MUSEUM


transitional syncranterian maxillary dentition2 indicate that members
of the Sauritus group are highly specialized, and that the group itself
occupies an advanced evolutionary position within the genus Tham-
nophis. The large number of teeth present in T. sauritus and T. prox-
imus is interpreted here as a secondary modification rather than a
primitive condition.
KEY TO THE SPECIES OF THE SAURITUS GROUP
Parietal spots almost always present, fused, bright, and usually fairly large
(fig. 1A, left); brown pigment usually not extending onto the ventral scutes
to form a dark ventrolateral stripe on each side or, if present, covering less


PROXIMUS SAURITUS
FIGURE 1. Head patterns in the two species of the Sauritus group in dorsal
view (A) and lateral view (B).
'Transitional syncranterian is equivalent to the "continuous-abrupt" arrangement
characterized by Malnate (1960) and thought to represent an, evolutionary
transition between the more generalized syncranterian and the more specialized
diacranterian conditions.


Vol. 7


IA








ROSSMAN: GENUS THAMNOPHIS


than 2/5 of the area of each scute (except in the Chiapas Highlands); mean
muzzle length/frontal length for any given population usually greater
than 0.700 -----.....------.. -- ------......._- ------ -------- -... T. proximus
Parietal spots often lacking, when present small and rarely fused or bright
(fig. 1A, right); brown pigment always extending onto ventral scutes and
usually covering 2/5 or more of the area of each scutee; mean muzzle
length/frontal length usually less than 0.700 ----- ----- --T. sauritus


THE WESTERN RIBBONSNAKE
THAMNOPHIS PROXIMUS (SAY)

Coluber proximus Say, 1823:339
Tropidonotus proximus: Boie, 1827:535
Eutaimia Faireyi Baird & Girard, 1853:25
E*tainia proxima: Baird & Girard, 1853:25
Eutaenia proxima: Baird, 1859:16
Eutaenia faireyi: Cooper, 1860:299
Eutaenia rutiloris Cope, 1885:388
Thamnophis proxima: S. Garman, 1892:105
Thamnophis proximus: Strecker, 1909:8
Eutainia rutilorus: Cochran, 1961:182

HOLOTYPE. According to Smith and Taylor (1945), the holotype is
lost. It was probably deposited in the Academy of Natural Sciences
of Philadelphia, though this is not a certainty (H. M. Smith, personal
communication). The specimen was collected in Nebraska at a stone
quarry on the west side of the Missouri River, 3 miles above the
mouth of Boyer's River (Iowa) by Long's expedition to the Rocky
Mountains in 1819 or 1820. The type locality lies approximately 3
miles ENE Fort Calhoun, Washington County, Nebraska.
DEFINITION. A large, long-tailed member of the genus Thamnophis
characterized by: 19-19-17 dorsal scale rows; a single preocular; typi-
cally 8 supralabials, the 4th and 5th entering the orbit; 141 to 181
ventrals; 82 to 131 subcaudals; lateral stripe on dorsal scale rows 3
and 4, at least anteriorly; labials and ventrals without black markings;
dark ventrolateral stripe absent or narrow in most populations; parietal
spots fused, brightly colored, and usually large; hemipenis short, usu-
ally extending to the seventh or eighth subcaudal when inverted;
teeth numerous, averaging about 30 to each maxilla, 34 to each den-
tary, 20 to each palatine, and 33 to each pterygoid.
RANGE. From southern Wisconsin, Indiana, and the Mississippi Valley
west through the Great Plains to southeastern Colorado and eastern
New Mexico, and south through eastern Mexico to central Costa


109


1963









110 BULLETIN OF THE FLORIDA STATE MUSEUM


Rica (figs. 2 and 3). It occurs on the Pacific coast of Mexico near
Acapulco, Guerrero, and on the Isthmus of Tehuantepec in Oaxaca
(see T. p. diabolicus for discussion of an early published record for


S190 90 390 49Q
WLEG


FIGURE 2. Distribution of Thamnophis proximus in the United States. Solid dots
represent specimens examined; open circles are published records. Stippled areas
represent the probable zones of intergradation between subspecies.

the Rio Grande Valley of western New Mexico). A western ribbon-
snake (ANSP 6179) collected in the 1800's and labelled simply
"Minnesota" requires authentication by further collecting. Another spe-
cimen (KUJ 21462), supposedly taken in Hamilton County, Ohio, is


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


probably incorrectly labelled. Wheeler (1947) recorded several speci-
mens of T. proximus from north-central North Dakota; Thomas Uzzell
examined these specimens (UMMZ 74337, 74339-342) at my request
and reports they are actually Thamnophis radix. Apparently no valid
record of T. proximus exists for the Great Plains north of Thurston
County, Nebraska.


S9 190 t9o 390 4 -


FIGURE 3. Distribution of Thamnophis proximus in Mexico and Central America.
Solid dots represent specimens examined; open circles are published records.
Stippled areas represent the probable zones of intergradation between subspecies.

VARIATION
Sexual Dimorphism
Distinct sexual differences exist in number of ventrals, number of
subcaudals, and relative tail length. The amount of dimorphism in
each of these characters varies from one local population to the next
with no apparent geographic correlation.
The difference in mean number of ventrals ranges from 0.3 of a
scale in favor of the females in the sample from the Staked Plains of


1963








112 BULLETIN OF THE FLORIDA STATE MUSEUM


western Texas (the only sample with at least five individuals of each
sex in which females have more ventrals than males), to more than
5 in favor of the males. Those populations for which at least 15 adults
of each sex are available are compared in table 3. This shows a marked

TABLE 3. Sexual dimorphism in ventrals of Thamnophis proximus

Males Differences Females
Population
N M DM CD% N M

New Orleans area... 35 173.60.50 3.2 1.860.05 52 170.4-0.39
East-central Texas... 18 168.90.73 2.8 1.67i-0.09 18 166.1-0.61
Edwards Plateau.... 34 167.80.46 3.9 2.350.10 30 163.9=0.74
Southern Texas..... 46 166.60.40 2.9 1.760.04 49 163.70.36
Chiapas Highlands... 15 161.70.59 3.7 2.310.08 22 158.00.53

Y = number of individuals; M = mean; DM = difference between meaxis; CD% = coefficient
of divergence expressed as a percent.

lack of geographic variation in degree of dimorphism as expressed
by the coefficient of divergence, which is slight in members of the
Sauritus group in comparison to most of the other forms for which
comparable data exist (Klauber, 1943, table 3). Thamnophis proximus
shows greater sexual dimorphism in number of ventrals than does
Thamnophis ordinoides, less dimorphism than T. radix and T. elegans,
and about the same amount as Thamnophis brachystoma and T. sauri-
tus (table 4).

TABLE 4. Coefficients of sexual divergence in six species of Thamnophis

Ventrals Subcaudals
Species Origin of Sample CD% CD%

T. brachystoma1..... Western Pennsylvania ...... 2.00 ? 14.99= ?
T. elegans2....... .. San Diego Co., Calif........ 4.170.20 10.940.32
T. ordinoides2....... Western Oregon............ 1.060.35 12.480 64
T. proximus........ New Orleans area .......... 1.864-0.05 9-054:-0.24
East-central Texas .......... 1.670.09 ...
Edwards Plateau........... 2.350.10 5.14+018
Southern Texas............ 1.760.04 6.68 0.26
Chiapas Highlands. ........ 2.310.08 ........
T. radix2.......... Cook Co., Ill............... 4.010.36- 12.240.71
T. sauritus.......... Southwestern Florida....... 3.330.13 6.05 =0.39
North-central Florida ....... 2.310.19 6.4540.39
Southern Michigan......... 1.480.09 5.020.38

1 Data from Barton (1956). 2 Data from Klauber (1943).


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


Somewhat greater sexual differences exist in mean number of sub-
caudals. They range from 0.8 in favor of the females in the Galveston
Bay, Texas, sample (the only one in which the males have fewer
subcaudals than the females) to as much as 15.2 in favor of the males.
Table 5 shows counts in those populations for which at least 15 adults

TABLE 5. Sexual dimorphism in subcaudals of Thamnophis proximus

Males Differences Females
Population
N M DM CD% N M

New Orleans area .... 19 117.81.04 10.2 9.05+0.24 33 107.60.91
Edwards Plateau..... 22 108.00.82 5.9 5.140.18 22 102.10.87
Southern Texas ...... 16 108.31.17 7.0 6.680.26 18 101.3=0.88
N= number of individuals; M = mean; DM = difference between means; CD% = coefficient
of divergence expressed as a percent.

of each sex are available. The amount of sexual dimorphism in sub-
caudal numbers is not great either in this species or in T. sauritus
compared to those of many other colubrid snakes (Klauber, 1943;
Auffenberg, 1955; Duellman, 1958; Dowling, 1960; and others). The
coefficients of sexual divergence are even less, in fact, than in other
species of Thamnophis (table 4), which is in keeping with the hypoth-
esis that sexual dimorphism in subcaudals tends to be greater in short-
tailed than in long-tailed species (Klauber, 1943).
Females average longer than males but usually have a proportion-
ally shorter tail. The largest female examined (USNM 761 from Cam-
eron County, Texas) has a body length of 900 mm. plus a 332 mm.
incomplete tail. The largest male (UF 12161.4 from St. Charles Parish,
Louisiana) has a body length of 553 mm. and an incomplete tail.
Wright and Wright (1957) reported animals with greater total
lengths (1268 mm. female, 1119 mm. male), but gave no locality data.
Adults of both sexes average well below these maxima.
The mean difference in adult tail length/total length ratios ranges
from 0.004 in favor of the females (at Galveston Bay, Texas, the only
area where the female ratio is higher than that for males) to"0.022 in
favor of the males. Mean tail length/total length values for males
range from 0.275 to 0.318, for females from 0.270 to 0.313. The muzzle
length/frontal length ratio shows females apparently have slightly
longer muzzles than do males.
Caudodorsal scale row reduction reflects sexual :dimorphism, males
having the more posterior reduction. This, is almost surely a conse-


1963


113









114 BULLETIN OF THE FLORIDA STATE MUSEUM Vol. 7

quence of the presence of the hemipenes and their retractor muscles
in the proximal part of the tail. The reduction formulas for CAS 3091,
an adult male from Comanche County, Oklahoma, and CAS 5099, an
adult female from St. Charles County, Missouri, are presented below
as examples. The formulas are slightly modified from the system pro-
posed by Dowling (1951) in that the figures above the line represent
the left side of the animal, those below the line the right.

CAS 3091 2(2) 15-9(3) 2+3(4) -3(7) -4(17) -3(36)
A 17- 15- 14 12 10 8-
S 2(1) 3(4) -3(7) 4(17) -3(32)
2(66) 2(100)2(113)
6--- 4 2(113)
2(64) 2(102)


CAS 5099 1(1) 1(2) -7(3) -2(3) 4(10) 8-3(33)
17 35- 3--- 12 1 8 -__ _'-
9 -1(1) -1(2) 2(4) -4(10) -3(33)
2(66) 2(96)
6--- 4-: 2(99)
--2(69) 2(96)

Ontogenetic Variation

Newly born T. proximus have proportionally shorter tails than adults
(table 6), but differential growth must be rapid during the first year,

TABLE 6. Ontogenetic change of tail length/total length ratio
in the Sauritus group

Males Females COD%
Species and Population -- -
Juv. Adult Juv. Adult Male Female

T. proximus
New Orleans area............... .299 .317 .288 .302 5.84 4.75
T. sauritus
Southern Florida ................ 334 .347 .338 .340 3.82 0.59
Northwestern peninsular Florida... .322 .343 .320 .340 6.32 6.06
Northern New Jersey............ 329 .353 .325 .341 7.04 4.80

COD% = coefficient of ontogenetic divergence expressed as a percent.

for subadults have essentially the same tail length/total length values
as adults of the same population.

Individual Variation

Characters that vary neither geographically, nor sexually, nor onto-








ROSSMAN: GENUS THAMNOPHIS


genetically are discussed below. Data on head scales refer to one
side only. Individual variation in scutellation is summarized in table 7.

TABLE 7. Individual variation in scutellation of Thamnophis proximus

Pre- Post- Anterior Posterior Supra- Infra- Inter-
oculars oculars temporals temporals labials labials genial

c"ci 1(653)* 2 (3) 1 (586) 1 (14) 6 8,9 1 (22)
2,3 (8) 1,2 (34) 1,2 (9) 7 (12) 9 (14) 2 (543)
3 (605) 2 (30) 2 (548) 7,8 (35) 9,10(20) 3 (71)
3,4 (32) 2,3 (55) 8 (600) 10(579) 4 (2)
4 (5) 3 (24) 8,9 (8) 10,11(16) 5
11 (14)
99 1(792)*2 (8) 1 (671) 1 (10) 6 8 (2) 1 (29)
"1,2 2,3 (5) 1,2 (64) 1,2 (15) 7 (15) 9 (11) 2 (590)
3 (704) 2 (54) 2 (573) 7,8 (20) 9,10(35) 3 (140)
3,4 (63) 2,3 (127) 8 (746) 10 (693) 4 (11)
4 (12) 3 (63) 8,9 (13) 10,11(30) 5
4,5 9 (5) 14 (15) 6 (2)
11,12 8
c" ci Dorsal scale rows 9 9

17-18-17 17-19-17 (2)
17-19-15 18-19-16
17-19-17 (2) 19-19-17 (613) 18-19-17 19-19-17 (754).
18-19-17 (2) 19-19-18 (4) 19-17-17 (5) 19-19-18 (5)
19-17-16 19-19-19 19-18-17 19-21-17
19-17-17 (2) 19-21-17 19-19-15 19-21-19
19-18-17 20-19-17 19-19-16 20-19-17 (2)
19-19-15 (3) 21-19-17 (3) 21-19-17 (9)
19-19-16 21-19-19 21-21-17
Single numbers indicate that both sides of the head have the same count.
OcuLARs. There is 1 preocular and usually 3 postoculars. The most
frequent variation is the appearance of a 4th postocular on one or
both sides. Only one individual of 1446 T. proximus has an extra pre-
ocular, and that on one side only. It is interesting to note the stability
of this character in the Sauritus group in view of its extreme varia-
bility in other species of Thamnophis (Ruthven, 1908; Fitch,--1940).
TEMPORALS. One large anterior temporal and 2 slightly smaller pos-
terior temporals is the usual condition, but this character shows greater
individual variation than any other, most often an increase in number.
Although the occurrence of 2 anterior temporals is common, only the
lower is ever in contact with the postoculars. In addition to an increase
in the number of scales, the anterior portion of the anterior temporal


1963








116 BULLETIN OF THE FLORIDA STATE MUSEUM


is occasionally fragmented to form 1 or 2 small scales which are rarely
much larger than the adjacent postocular scales. In the Chiapas High-
lands 42 percent of the animals examined have this fragmentation of
the anterior temporal on one or both sides of the head, a higher inci-
dence of this variation than in any other area.
SUPRALABIALS. The usual number is 8, the 4th and 5th entering the
orbit; 7 apparently occurs by loss of a preorbital supralabial, 9 by addi-
tion of a scale posterior to the orbit. Infrequently small scales may be
wedged between two normal scales either at their upper or lower
margins. Rarely an individual supralabial is partly divided by an
incomplete suture.
Although marked geographic variation is not apparent, a number
of populations in and east of the Mississippi Valley have a somewhat
higher incidence of 7 supralabials than does the species as a whole.
INFRALABIALS. The usual number is 10, 9 and 11 being equally com-
mon as variations; 9 usually results from the elimination of the suture
between the 3rd and 4th or the 8th and 9th infralabials, in that order
of frequency; 11 usually results from a division of the 3rd infralabial.
As in the supralabials, small anomalous scales are occasionally inter-
posed between normal ones.
INTERGENIALS. Usually 2 small scales lie one behind the other between
the genials, occasionally only 1 or as many as 8 are present. When
several are present, some may be paired instead of in a linear series
of single scales. In the Chiapas Highlands sample 50 percent of the
individuals have more than two intergenials, an unusually high inci-
dence. These animals also have a high frequency of anterior tem-
poral fragmentation; over 76 percent of the series examined have one
or both of these anomalies.
ANAL PLATE. In all but a few of the specimens examined the anal
plate is undivided; it is completely divided in TCWC 865 (5 km. E
Las Vigas, Veracruz) and CNHM 4441 (British Honduras), and
partly divided in UIMNH 4499 (Crockett County, Texas). The anal
plate is creased but not actually divided in CNHM 4228 (British
Honduras).
VENTRALS. Anomalous divisions or partial duplications of ventral scutes,
largely caused by partial duplication of vertebrae, occur in many
groups of snakes (King, 1959), and the ribbonsniakes are no excep-
tion. All the types Peters (1960) illustrated in his study of the colu-
brid subfamily Dipsadinae occur in both T. proximus aiad T. sauritus,
the commonest being that with one or more anomalous -partial scutes


Vol. 7







ROSSMAN: GENUS THAMNOPHIS


inserted between the terminal ventral and the anal plate, which occurs
in 9.5 percent of the T. proximus examined. An additional 4.3 percent,
including a few of the same animals, have partial scutes elsewhere
on the venter. Barton (1956) reported that 19.5 percent of the speci-
mens of T. brachystoma he examined showed ventral or subcaudal
anomalies.
SUBCAUDALS. Of the individuals examined, 2.3 percent have one or
more sets of fused subcaudals (with no longitudinal suture separating
the adjacent members of a pair).
DORSAL SCALE ROWS. Only 4 percent of the snakes examined differ
from the typical formula of 19-19-17. A large majority of the individ-
uals having more than 19 rows at any point on the body are from the
delta region of the Mississippi River, but these make up only a small
part of that population. Of the specimens examined from Orizaba,
Veracruz, 31 percent have reductions in the dorsal scale row formula
(i.e., 17-18-17, 18-19-16),
The normal posterior reduction in the number of dorsal scale rows
is, in all but 4 of the 18 specimens checked for this character, the
result of loss of row 4 on both sides of the body. In the four exceptions
row 5 is lost on one or both sides.
Among males the reduction takes place opposite ventrals 83 through
105, mean 93.8, among females from ventral 78 through 107, mean
90.0. There appears to be a positive correlation between the total
number of ventrals and the point of reduction, the greater the num-
ber the more posterior the reduction. Consequently the point of re-
duction varies geographically to some degree and also sexually, as
males almost invariably have more ventrals than do females.
PATTERN AND COLORATION. Size of the parietal spots varies individually;
some individuals have streaks of light pigment extending anteriorly
from the parietal'spots along the frontal-parietal suture and occasion-
ally onto the frontal itself. Size and even presence of the light post-
ocular spot and of the black postorbital vitta vary individually as well
as geographically. Individual variation is also manifest-in the relative
darkness of the dorsal ground color and in the presence or absence of
a row of spots above the lateral stripe or of a black margin along the
vertebral stripe.
Geographic Variation
In an animal whose range extends about 3000 miles from southern
Wisconsin to central Costa Rica, considerable geographic variation
is not surprising. Because of the lack of concordance among most of


1968









118 BULLETIN OF THE FLORIDA STATE MUSEUM


the geographically variable characters, each is considered separately.
In the following discussion the term "population" is used in the bio-
logical, not the statistical sense. "Deme" or "local population" would
be more nearly equivalent to "population" as I use it. A group of
morphologically similar, geographically contiguous populations con-


Fso W6ss58(ia ifM19 4. 145)
,.6(8) 168940 171.400
167.7(3) B 2
9 iqo 9ao 390 42o
MILES


FIGURE 4. Geographic variation of ventral number in Thamnophis proximus in
the United States. Black blotches represent the areas from which samples were
taken. The upper numbers associated with each blotch are the sample mean and
sample size (in parentheses) for males; the lower numbers present the same
data for females. When a line appears in place of either the upper or lower
numbers, it signifies that no specimens of that sex were present in the sample.


VoL 7







ROSSMAN: GENUS THAMNOPHIS


statutes a subspecies. Meristic and proportional values discussed below
are based on population samples large enough to represent fairly the
population from which they were collected.
VENTRALS. The mean ventral counts are remarkably constant east of
the Mississippi River from western Indiana to north of Lake Ponchar-
train in Louisiana (fig. 4). The means for males fall between 170.4
and 171.8, for females between 166.9 and 169.0. South of Lake Pon-
chartrain in and adjacent to the Mississippi delta region the number
of ventrals increases slightly, the means ranging from 171.9 to 174.1
in males, and from 168.9 to 171.4 in females. The highest means in
both sexes occur near the mouth of the Mississippi River.
West and southwest, from the Mississippi Valley to the Rio Grande
Valley, The mean number of ventrals decreases gradually and rela-
tively uniformly, although with some interpopulational variation. A
population in western Louisiana (fig. 4, B), however, shows more
than the expected divergence in that the sample mean in females
(only one male is available) is from 7 to 9 scales less than in females
from populations less than 25 miles distant. It is perhaps of significance
that all the localities for specimens of this population are relict prairie
areas.
Southward from the Rio Grande Valley the mean ventral counts
decrease only slightly in the Sierra Madre Oriental, but drop sharply
in the coastal plain of Tamaulipas (fig: -5). Females from the lower
Rio Grande Valley show a mean difference of almost 10 ventrals
(males about 8) from those from southern Tamaulipas, little more
than 200 miles distant. Essentially no change is manifest in the mean
for females and only a slight decrease in that for males over the 200-
mile stretch from southern Tamaulipas to central Veracruz, but within
the next 50 to 75 miles southward along the coast to the vicinity of
Ciudad Veracruz; the means for both sexes decrease by almost 4
scales.
The number of ventrals is lowest in the Tabasco-Campeche low-
lands where the mean for males is 146.6, for females 144.9. This repre-
sents a decrease of almost 6% scales in males and 4% in females over
the 800-mile distance from south-central Veracruz. Counts in a small
series from isthmian Oaxaca indicate that ventral numbers in this
region probably average somewhat higher than those in adjacent
areas. From the Tabasco-Campeche region to the northern tip of the
Yucatan Peninsula there is a slight increase in the mean number of
ventrals, which then remains essentially constant .along the 1000
miles of Caribbean coastland from Cozumel Island to Boca del Rio
Colorado, Costa Rica.


1963








120 BULLETIN OF THE FLORIDA STATE MUSEUM
y


q q0 "90 890 490


FIGURE 5. Geographic variation of ventral number in Thamnophis proximus in
Mexico and Central America. Black blotches represent the areas from which sam-
ples were taken. The upper numbers associated with each blotch are the sample
mean and sample size (in parentheses) for males; the lower numbers present the
same data for females. When a line appears in place of either the upper or lower
numbers, it signifies that no specimens of that sex were present in' the sample.


The western ribbonsnake is not limited to the coastal plain in tropi-
cal America, but also occurs in the Sierra Madre Oriental at altitudes
up to 8000 feet, in the Chiapas Highlands up to 7500 feet, and on
the Pacific side of the Central American uplands up to-5000 feet.
Most of these high-altitude populations have mean ventral numbers
markedly different from those in the geographically adjacent "row-
lands. Rather than decreasing, as one might predict on the basis of
temperature-scutellation studies on T. elegans (Fox, 1948; Fox,
Gordon,, and Fox, 1961), the number of ventrals increases. Further-
more, in certain areas the amount of increase seemsroughly corre-
lated with the amount of increase in altitude. Three populations in
south-central Veracruz, only 85 airline miles apart, range in altitude


Vol. 7







ROSSMAN: GENUS THAMNOPHIS


from sea level to 5500 feet. Near sea level in the vicinity of Ciudad
Veracruz, males have 153.0 mean ventrals, females have 149.3. Near
Cordoba (2000-3000 feet) the means are 5 scales greater in both
sexes. In the sample from the Orizaba area (4000-5500 feet) the mean
for males is only 1 scale more than at Cordoba; in females it is about
2% scales more.
The population in the Chiapas Highlands has a greater mean ven-
tral number (almost 8 higher in both sexes in the series examined)
than the Central American upland population. As the two are widely
separated geographically, factors other than altitude may be involved.
On the other hand, the means for samples from the Central American
upland population are 3 to 4 scales greater than means for the Carib-
bean lowland animals, and means for the Chiapas Highlands series
ar6 13 to 15 scales more than for the Tabasco-Campeche lowlands
population, 100 miles to the north and some 7000 feet lower in eleva-
tion. This relationship between high ventral count and high altitude
does not appear to hold true north of the Orizaba-Cordoba area. A
small series from the vicinity of Jalapa, Veracruz (3500-4500 feet), a
juvenile female (CNHM 103445) from 6 miles S Zacualtipan, Hidalgo,
and a subadult male (UIMNH 18919) from near Tezuitlan, Puebla
(approximately 6500 feet), agree well with coastal specimens at the
same latitudes. An adult male (TCWC 865) collected about 10 miles
NW Jalapa at an altitude of 8000 feet further complicates the situation
by having a much higher number of ventrals (166) than does any of
the animals taken at lower altitudes.
Some other populations, unfortunately represented by only a few
specimens, do not fit well into the variational pattern described above.
Four females from eastern San Luis Potosi agree in mean number of
ventrals with the south-central Veracruz females rather than with
those in the intervening area. Two females (UMMZ 41555-556)
from Cuatotolapam in southern Veracruz have a mean almost 6 scales
higher than that of the population less than 50 miles to the northwest,
and 10 scales greater than in the Tabasco-Campeche population barely
200 miles to the east.
SUBCAUDALS. Although the general pattern of reduction in subcaudal
numbers is similar to that of the ventrals, many of the details are
strikingly dissimilar. Whereas in ventrals the populations throughout
the Mississippi Valley have essentially the same mean, and reduction
in number occurs southwestward, in subcaudals the highest means
are centered in southeastern Louisiana (both to the north and south
of Lake Ponchartrain), and the number of subcaudals decreases fairly
uniformly and rapidly northward east of the Mississippi River, north-


1963









122 BULLETIN OF THE FLORIDA STATE MUSEUM


westward into the Great Plains, westward through central Texas, and
along the Gulf Coast into Mexico (see figs. 6 and 7). The mean num-
ber of subcaudals in western Indiana is about 12 to 14 scales less than
in southeastern Louisiana; means in central Kansas are 10 to 12 less;
and the series representing the Edwards Plateau population has 8


I" t992W5)
g 199 290 190


FIGURE 6. Geographic variation of subcaudal number in Thamnophis proximus
in the United States. Black blotches represent the areas from which samples were
taken. The upper numbers associated with each blotch are the sample mean and
sample size (in parentheses) for males; the lower numbers present the same data
for females. When a line appears in place of either the upper or lower numbers,
it signifies that no specimens of that sex were present in the sample.


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


to 9 fewer mean subcaudals in females (10 to 12 in males), as does
the sample from southern Texas.
West of the Pecos River the number of subcaudals decreases sharply,
the mean for females being 10 scales less than on the Edwards Plateau
less than 200 miles to the east. The Coahuilan desert ribbonsnakes


9 .o9 tg9o 3p 49o
WAnS.


FIGURE 7. Geographic variation of subcaudal number in Thamnophis proximus
in Mexico and Central America. Black blotches represent the areas from which
samples were taken. The upper numbers associated with each blotch are the
sample mean and sample size (in parentheses) for males; the lower numbers
present the same data for females. When a line appears in place of either the
upper or lower numbers, it signifies that no specimens of that sex were present
in the sample.

are similar to the trans-Pecos population in subcaudal number moreso
in fact than to the geographically nearer Nueyo Ceon plo atitni / ,
Along the Gulf Coast of Texas, two populations show a rather anoma'- .'
lous variance from the pattern described above: The sample rn ap -
for Galveston Bay area females fits rather well ito the preyviusIt -
mentioned sequence of reduction along the coastb) bt theft~ en for



V -----C- :
-*. ^


1963








124 BULLETIN OF THE FLORIDA STATE MUSEUM


males is almost 7 scales less than for the populations on either side of
the Galveston Bay area; it is even less than the mean for females from
the same population. Only 150 miles farther down the coast the situa-
tion is completely reversed; the males have the expected comple-
ment of subcaudals while the mean for females is from 52 to 8%
scales less than in adjacent populations.
In south-central Tamaulipas the number of subcaudals drops sharply,
the means decreasing by 8 or 9 scales within less than 100 miles. From
thence southward and eastward to the Tabasco-Campeche area the
decline in subcaudal number again appears to be fairly uniform and
gradual (the drop in the mean for males may be fairly steep between
south-central Veracruz and Tabasco-Campeche, but only one specimen
with a contplete tail is available from the latter population). The
lowest mean subcaudal number in T. proximus apparently occurs in
northern Yucatan. It increases markedly in the Caribbean lowlands,
but more comparative material is needed from the Gulf side of the
Yucatan Peninsula to determine the degree of increase.
Subcaudal number apparently is less well correlated with altitude
than is the number of ventrals. The means of males from sea level
show virtually no difference from those at an altitude of 5000 feet in
south-central Veracruz, and those of the high-altitude females increase
only by 2 to 4 scales. Females from the Caribbean lowlands, Central
American uplands, and the Chiapas Highlands show no major differ-
ences although the high-altitude populations do have slightly higher
means. The males from these populations may demonstrate altitudi-
nal variation more clearly, although the Chiapas snakes are separated
from the other two groups latitudinally as well as altitudinally. The
fact that the isthmian Oaxaca series (all from less than 700 feet alti-
tude) has means almost identical to those of the high-altitude Chiapas
snakes emphasizes the need for caution in interpretation.
PROPORTIONS. Though many populations are represented by only a
few specimens of either sex with complete tails, relative tail length
(as expressed by the tail length/total length ratio) appears to be, with
a few notable exceptions, fairly uniform over most of the- range of
T. proximus.
SThe snakes with the proportionally longest tails occur in the lower
,Mississippi V.lley, where the means for males range from 0.312 to
S.-Igancl-; fr females from 0.301 to 0.313 (except for those females
-"from southweh, of New Orleans and in the Atchafalaya delta region
- whcja have 0.246). Here the correlation between relative tail length
S lg th6 number ofsaubcaudals, weak or nonexistent in many parts of
the range, is strongEast of the Mississippi River the means vary within
/-e -
-


Vol. 7







ROSSMAN: GENUS THAMNOPHIS


these extremes as far north as the Ohio River. Populations west of the
Mississippi River having comparable means extend from the lower
Mississippi Valley northward and westward through Arkansas and
southern Missouri to the limits of the Ozark Plateau (the means for
females, however, decrease in the Ozark region).
Beyond this circumscribed area the snakes have somewhat shorter
tails, all but one of the means for males falling between 0.290 and
0.308 (the majority between 0.296 and 0.303). The one notable ex-
ception is the Chiapas Highlands population in which the nine males
examined have a mean of 0.275. The lowest mean tail length/total
length value for females, 0.270, also occurs in this population. Females
from outside of the 'long-tailed" region have means ranging from
0.279,to 0.298 (in most cases from 0.289 to 0.297), but in addition
to the Chiapas females already mentioned, those from trans-Pecos
Texas (also 0.270) have shorter tails than the adjacent populations.
Relative muzzle length as expressed by the muzzle length/frontal
length ratio is apparently subject only to interpopulation variation
throughout most of the range of T. proximus. The Chiapas Highlands
population, however, in this as in so many other features, is different
from all other populations, particularly those adjacent to it. Chiapas
males have a mean of 0.681, females 0.694. Means for other popula-
tions with four or more adults of each sex range from 0.708 to 0.849
(mean 0.744) in males,, from 0.711 to 0.838 (mean 0.759) in females.
PATTERN. The fused parietal spots attain their largest relative size in
northern Mexico and adjacent Texas; in some individuals these oc-
cupy half the length of the interparietal suture, as in AMNH 67336
from Nuevo Leon. Snakes in this region also exhibit a high incidence
of light areas or streaks extending forward from the parietal spots and
forking out for varying distances along the frontal-parietal suture. In
the Chiapas Highlands the parietal spots are incompletely fused and
their anterior edges diverge at about a 30-degree angle from the inter-
parietal suture, giving the entire arrangement the appearance of a
thick chevron. This condition, which sporadically occurs in individuals
elsewhere in the range of T. proximus, is present in 70 percent of the
Chiapas specimens.
In most populations the light postocular spot occupies all of the
lower and varying amounts of the middle postocular. In the southern
part of the range from trans-Pecos Texas to, Coahuila and Nuevo
Leon, and in the coastal plain from the Tropic of Cancer to Central
America, the postocular spot is frequently absent and when present
is usually not brightly colored. The spot is also talent in an occasional
individual from more northern populations.


1963








126 BULLETIN OF THE FLORIDA STATE MUSEUM


The broad black postorbital vitta typical of T. sauritus is usually
lacking in tropical American specimens of T. proximus. When present
the stripe is usually narrow and does not extend onto the postocular
scales. Throughout the rest of the range the presence and extent of
the vitta vary individually.
In the lowlands of Mexico and Central America and along the coast
to the lower Mississippi Valley, the vertebral stripe is usually broad
and covers the entire vertebral and more than half, often all, of both
paravertebral scale rows. INHS 9469 from near Ciudad Mante,
Tamaulipas, has the vertebral stripe covering the entire vertebral-
paravertebral series and a small portion of row 8 as well. Outside
this region the stripe rarely includes more than half the width of each
paravertebral row, and it is often somewhat narrower, particularly
in the extreme northern part of the range.
Mexican and Central American specimens are also characterized
by a narrow lateral stripe, which occupies less than half of 1 of the 2
scale rows bearing it and varying amounts of the other row, usually
slightly more than half. The stripe becomes even narrower beyond
the point of dorsal scale row reduction, where it is largely confined
to the 3rd row and a small portion of row 4. The Chiapas Highlands
population characteristically has the stripe involving only the 3rd row
posteriorly. Only about half the snakes in the Nuevo Leon series
have this reduction of the lateral stripe. North of the Rio Grande the
lateral stripe almost invariably occupies more than half of both rows
3 and 4 and does not become noticeably reduced posterior to the loss
of row 4.
The dark ventrolateral stripe, which is found in all specimens of
T. sauritus, and is the most reliable single character for distinguishing
between the two species in the areas of sympatry (Rossman, 1962),
is well developed in certain populations of T. proximus. It is broadest
in the Chiapas Highlands where the width of each stripe is 0.14 to
0.25 (usually 0.17 to 0.20) the width of a ventral scale. The stripe is
slightly narrower in the Central American uplands, still narrower
in the lowlands of Central America and Mexico, and usually disap-
pears altogether north and east of central Texas.
COLORATION. Cope (1885) described a ribbonsnake from Cozumel
Island, Quintana Roo, Mexico, as having reddish-orange labials. While
I have not seen one of that particular color, the labials of living speci-
mens from British Honduras, Chiapas, Veracruz, and eastern San Luis
Potosi are yellow-orange instead of greenish white or greenish yellow
as in individuals from Coahuila, Nuevo Leon, and the United States.
Because of the dearth of color descriptions in the literature or on


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


museum specimen labels, and of live captive snakes for study from
the entire range of the species, the geographic variation in the colors
of the dorsum and of the vertebral and lateral stripes can be dis-
cussed only generally and approximately. A more precise delineation
of the geographical extent of each color variation awaits analysis from
more adequate material.
Specimens from Coahuila, Nuevo Leon, and apparently trans-Pecos
Texas (which were not seen in life) have an olive-gray dorsum.
Throughout the rest of Mexico and Central America the dorsum is
olive-brown. Chiapas Highlands examples are an exception; their
dorsa are dark brown. The olive-brown of Central American snakes
and several other shades of brown occur in southern and central Texas,
along the Gulf coast of Texas and Louisiana, and apparently north-
ward through the Texas panhandle to southeastern Colorado. The
black-backed form ranges from central Kansas, central Oklahoma, and
eastern Texas northward and eastward throughout the rest of the
range.
The orange vertebral stripe has the most extensive geographical
range, replaced by other colors only in tropical America, on the Ed-
wards Plateau, and in coastal Texas and Louisiana. Ribbonsnakes on
and about the Edwards Plateau have a bright red vertebral stripe.
Where brown-backed populations occur along the Gulf Coast of
Louisiana and Texas, the stripe is usually gold or greenish gold, but
may have an orange cast. The stripe is also gold in the Chiapas High-
lands. A fresh specimen (UF 12213) from near Acapulco, Guerrero,
has a stripe that appears pale yellow, but may have been gold in life.
Snakes from Central America and the southern and central Gulf Coast
and adjoining uplands of Mexico have a grayish-tan vertebral stripe,
sometimes with a metallic luster.
Throughout most of the range of T. proximus, the lateral stripe is
usually some shade of light yellow, generally paler in lowland Mexico
and Central America. Two specimens from Brazos County, Texas
(UF 12193), are unique in having a lateral stripe that in life was
bluish green on the neck and greenish white posteriorly. Whether
this condition is characteristic of the entire population or merely an
individual variant is not known. One of these two specimens also had
a gold vertebral stripe; the other had the orange stripe typical of this
area.
DISCUSSION. Obviously much of the geographic variation observed in
T. proximus is of the discordant type. Almost all the characters I have
discussed vary independently of each other; some are constant
throughout a geographical range in which one or several other char-


1963








BULLETIN OF THE FLORIDA STATE MUSEUM


acters undergo radical change. This discordant variation is to be ex-
pected, however, unless all these characters are genetically linked. If
this were so, certain combinations could be selected as a single unit.
Otherwise, as apparently in this case, each character must be subject
to independent selection by the environment (Tilden, 1961).
In two areas marked changes do take place simultaneously in
several characters-the Chiapas Highlands and southern Tamaulipas.
In the Chiapas Highlands almost every character that varies geograph-
ically (except subcaudal number and labial color) differs from that in
the adjoining populations. The high-altitude Chiapas population prob-
ably has been at least semi-isolated from the lowland populations for a
long time. This isolation, whether partial or complete, coupled with
adaptations to the rigors of life at high altitudes and perhaps furthered
by genetic drift, has allowed these snakes to differ in many respects
from adjoining populations. It is not understood why the high-altitude
populations in the Sierra Madre Oriental do not differ from adjacent
lowland populations to the same marked degree. The abrupt transition
of characters in southern Tamaulipas coincides with the transition be-
tween the Neotropical and Nearctic faunal regions, long-recognized
as a major faunal boundary (Darlington, 1957). Ventral number, sub-
caudal number, labial color, vertebral stripe color, and lateral stripe
color and width all change markedly at this point.
In southern Louisiana lies a third and somewhat less important
zone of concordance. Here dorsum color, and vertebral stripe width
and color, change over a relatively narrow zone between the coast
and the interior.
In studying geographic variation in these snakes, one is struck by
the fact that the counts of ventrals and subcaudals are much more
apt to fluctuate locally than are the elements of color pattern. The
reason for this is not clear, but the relative instability of the meristic
characters suggests they may be subject to less selective pressure
than is color pattern.

TAXONOMY
Many arguments for and against the trinomen have been presented
in recent years, most notably in the journal Systematic Zoology. In
recognizing subspecies in this study I am not arguing for the "reality"
of these or of any other geographic races as evolutionary units. With
most geographic variation being of the discordant type, apparently it
is the individual characters (or small groups thereof) that are under-
going evolutionary selection. The systematist is thus forced to select
one or more characters to define each geographic population he recog-


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


nizes taxonomically. The subspecies is thus an arbitrary entity albeit,
I believe, a useful one. It is a convenient tool which allows us to locate
geographically and to denote the nature of distinctive intraspecific
populations with a single word. This, to me, seems ample justifica-
tion for its existence and continued usage.
The population samples discussed above group themselves into 6
fairly well defined major populations which occupy discrete geo-
graphic areas and appear to deserve taxonomic recognition. While
some of these populations can be distinguished by differences in scale
counts as well as in color pattern, I have placed more emphasis on
color because it is usually less variable over wide geographic areas,
a common phenomenon in the genus Thamnophis (Fitch, 1941, and
others). Fox, Gordon, and Fox (1961) have shown that color patterns
in individual T. elegans subjected to low temperatures during em-
bryonic development do not differ from those of animals that devel-
oped at normal temperatures, while ventral and subcaudal numbers
differ strikingly between the two groups.
Recent systematic work on the gartersnakes generally fits into one
of two categories: recognition of intraspecific populations primarily
on variations in color pattern, or mainly on differences in the number
of ventral and subcaudal scales. This dichotomy of viewpoints is
equally manifest in the work on most other snake genera that have
received much taxonomic attention. A recent paper on T. sirtalis
parietalis and its allies (Fitch and Maslin, 1961), which dismissed
the number of ventrals and subcaudals with two brief sentences on
their general clinal trends, and an earlier paper on Thamnophis mar-
cianus (Mittleman, 1949), which recognized two subspecies solely
on meristic differences, represent extremes of the two approaches. As
variations in color pattern and in meristic characters are often dis-
cordant, the geographic races recognized by the two schools of inter-
pretation are seldom equivalent. If the two schools were to study the
same material, their respective delineation of range limits of the
included subspecies would certainly differ even if both recognized the
same number of races-which is unlikely.
The chaos inherent in such a situation provides opponents of the
subspecies concept with an excellent argument for discontinuing the
trinomial system of nomenclature. Neither interpretative approach can
be said to be the only correct one, inasmuch as the characters em-
phasized in each case are merely different expressions of the same
genetic system. However, if the subspecies is to play a meaningful
role in ophidian systematics, a uniform basis for interpretation must
be established. A compromise of sorts might be achieved by applying


1963








BULLETIN OF THE FLORIDA STATE MUSEUM


subspecific names only to those populations that display concordance
in both color pattern and meristic characters, but this apparently
occurs so infrequently that many distinctive populations would go
unrecognized. Consequently I suggest that geographic races in the
gartersnakes be based on color pattern when variation is discordant.
Application of this suggestion to other genera should await a careful
evaluation of each individual genus. Meristic characters such as ven-
trals and subcaudals must not be ignored, however, but should be
thoroughly and completely investigated, analyzed, and reported.

Subspecies of Thamnophis proximus

The following key for distinguishing the subspecies of T. proximus
is designed for use with living or freshly killed specimens. As pre-
servatives alter certain colors quickly and drastically, emphasis should
be placed on "geographic probability" when only preserved material
is available.

1. Dorsum black; vertebral stripe orange and usually narrow; dark ventro-
lateral stripes lacking ------____---.-- --------. --------------- T. p. proximus
Dorsum olive-gray or some shade of brown ..-----------------------..------. 2
2. Dorsum dark brown; vertebral stripe gold, of medium width; dark ventro-
lateral stripe broad, usually 0.17 to 0.20 of a ventral scale wide; tail
relatively short (mean tail length/total length in males 0.275, females
0.270); parietal spots usually chevron-shaped ------------- T. p. alpinus
Dorsum olive-gray or brown (not dark); dark ventrolateral stripe relatively
narrow or absent; tail relatively long (mean tail length/total length
usually exceeding 0.290 in males, 0.280 in females); parietal spots rarely
chevron-shaped -----..------------------------- ---------------------------. 3
3. Vertebral stripe bright red, of medium width; ventrolateral stripe narrow
or lacking _---. _____-------------------- _T. p. rubrilineatus
Vertebral stripe orange, grayish tan, or some shade of gold --- ------- 4
4. Vertebral stripe grayish tan, broad; labials yellow-orange; lateral stripe
narrow anteriorly, further reduced posteriorly; mean number of ventrals
not exceeding 160 in males or 157 in females ----___ __------T. p. rutiloris
Vertebral stripe orange or some shade of gold; greenish-white or greenish-
yellow labials; lateral stripe not narrow, at least anteriorly; mean number
of ventrals exceeding 162 in males and 158 in females --__ ----- 5
5. Vertebral stripe orange, of medium width; dorsum usually olive-gray; lateral
stripe frequently reduced posteriorly; narrow ventrolateral stripe fre-
quently present __----- -- --- ------- T. p. diabolicus
Vertebral stripe usually some shade of gold, broad; dorsum usually olive-
brown; lateral stripe rarely reduced; ventrolateral stripe usually absent
----------............................................--------------------------------- T. p. orarius


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


As geographic variation has already been discussed in detail, only
those features pertinent to a formal description of the subspecies are
presented below.

Thamnophis proximus proximus (Say), new combination
Tropidonotus saurita var. Faireyi: Dumeril, Bibron, & Dumeril, 1854:585 [by
implication]
Tropidonotus saurita var. proxima: Jan, 1863:70
Eutainia saurita var. faireyi: H. Garman, 1892:264
Eutainia saurita var. proxima: H. Garman, 1892:264
Thamnophis saurita var. proxima: Dumeril & Bocourt, 1893:757
Thamnophis saurita var. faireyi: Dumeril & Bocourt, 1893:758
Thamnophis sauritus proximus: Ruthven, 1908:98
Thamnophis proxima faireyi: Hurter, 1911:159
Thamnophis sirtalis proximus: Klauber, 1948:9
HOLOTYPE. The information presented for the species applies here.
Although the holotype is thought to be lost, a neotype has not been
designated because the original description (Say, 1823) and the type
locality are adequate to associate the name with a specific population.
DEFINITION. A subspecies of Thamnophis proximus characterized by a
black dorsum, a narrow orange vertebral stripe, and lack of a dark
ventrolateral stripe.
RANGE. From Indiana and southern Wisconsin south in the Mississippi
Valley to, but not including, southern Louisiana, west through south-
ern Iowa and extreme eastern Nebraska to central Kansas, central
Oklahoma, and eastern Texas (exclusive of the coastal region).
DESCRIPTION. In the absence of the holotype, the following,description
is based on UF 12160.1, an adult male collected 6.5 miles S Safford-
ville, Chase County, Kansas, by William E. Duellman.
Scutellation: Oculars 1 + 3, temporals 1 + 2, supralabials 8, infrala-
bials 10, intergenials 2, dorsal scale rows 19-19-17, ventrals 167, sub-
caudals 108. Body length is 392 mm., tail length 166, tail length/total
length ratio 0.297.
The dorsum is black over most of the body, becoming dark brown
posteriorly, and dark brown below the lateral stripe. The vertebral
stripe is pale orange (Maerz & Paul 11L7), the lateral stripe light
yellow. The venter is yellowish green with a pale orange tinge along
the lateral margin of each scute, the dark ventrolateral stripe lacking
entirely. The pale yellow postocular spot occupies all of the lower
and the anterior half of the middle postocular; the fused parietal spots
are moderately prominent and bright.


1963








132 BULLETIN OF THE FLORIDA STATE MUSEUM


DIsCUSSION. Counts and measurements of this subspecies may be sum-
marized as: postoculars 2 to 4, temporals 1 + 1 to 2 + 3, supralabials
7 to 9, infralabials 8 to 11, intergenials 1 to 6, dorsal scale rows 17-19-15
to 21-19-17, male ventrals 151-178, female ventrals 154-176, male sub-
caudals 99-131, female subcaudals 90-120. Body length in males 122 to
522.5 mm., in females 128.5 to 680 mm. Total length in males 174 to
755 mm., in females 177.5 to 956 mm. Adult tail length/total length
ratio in males 0.284 to 0.333, in females 0.267 to 0.336.
An adult female (KU 21462) supposedly collected in Hamilton
County, Ohio, by R. Goodpaster was sent to the University of Kansas
Museum of Natural History in a lot that included specimens from
Reelfoot Lake, Tennessee, as well as from Hamilton County, Ohio
(W. E. Duellman, personal communication). This snake agrees well
in all respects with a series of females from the vicinity of Reelfoot
Lake and, as no other western ribbonsnake has been taken within 80
miles of Hamilton County, I think this specimen was actually collected
at Reelfoot Lake.
I have seen no intergrades between T. p. proximus and either
T. p. rubrilineatus or T. p. diabolicus in life, but the few preserved
specimens from regions where the ranges of the subspecies meet sug-
gest that the black-backed T. p. proximus apparently intergrades with
one brown race (diabolicus) over a broad zone in southwestern Kan-
sas, western Oklahoma, and the eastern part of the Texas panhandle,
with still another brown form (rubrilineatus X diabolicus intergrades)
in north-central Texas, and finally with T. p. rubrilineatus in east-
central Texas. Intergradation between T. p. proximus and T. p. orarius
is discussed under the latter form.
The black dorsum of the nominate race and the narrowness of its
light vertebral stripe may have adaptative value in increasing the
absorption of solar radiation in the cooler, northern part of the spe-
cies' range. The presence of the black-dorsum subspecies in the south-
central United States where this adaptation would seem of little im-
portance today may indicate that this race originally evolved in the
South when the climate was much cooler, and has since extended its
range northward.
Thamnophis proximus orarius, new subspecies
HOLOTYPE. TU 11764, an adult female from Waggaman, Jefferson
Parish, Louisiana, collected 28 February 1950 by Horace Whitten.
DEFINITION. A subspecies of Thamnophis proximus characterized by
an olive-brown dorsum, a broad gold vertebral stripe, and lack of a
dark ventrolateral stripe.


Vol. 7







ROSSMAN: GENUS THAMNOPHIS


RANGE. From Cat Island in Mississippi Sound, and southwestern
Hancock County, Mississippi, through southern St. Tammany Parish,
Louisiana, south and west along the Gulf Coast, probably as far as
northern Tamaulipas.
DESCRIPTION OF HOLOTYPE. Scutellation: Oculars 1 + 3, temporals
1 + 2, supralabials 8, infralabials 10, intergenials 2, dorsal scale rows
19-19-17, ventrals 169, subcaudals 106. Body length is 463 mm., tail
length 200 mm., tail length/total length ratio 0.302.
The following color description is based on UF 12165, an adult
female collected 3 miles E Boutte, St. Charles Parish (approximately
7.5 miles W Waggaman), by H. A. Dundee. Dorsum olive-brown
(Maerz & Paul 15H9), grayish brown below the lateral stripe (15E4);
vertebral stripe gold (13K4), broad; lateral stripe yellow (10E1);
venter pale green medially (18D3), lateral margins orange (11H6);
dark ventrolateral stripe lacking entirely; chin shields white; suprala-
bials pale yellowish green, anterior three tan. The pale yellow post-
ocular spot occupies all of the lower and middle postoculars and the
leading edge of the upper one; the fused parietal spots are moderately
prominent and bright.
DIscussioN. Counts and measurements of this subspecies may be sum-
marized as: postoculars 2 to 4, temporals 1 + 1 to 2 + 3, supralabials
7 to 8, infralabials 8 to 11, intergenials 1 to 4, dorsal scale rows 17-19-17
to 21-21-17, male ventrals 161-181, female ventrals 158-177, male sub-
caudals 97-127, female subcaudals 87-124. Body length in males 139
to 553 mm., in females 170 to 900 mm, Total length in males 243 to
762 mm., in females 241.5 to 1232 + mm. Adult tail length/total length
ratio in males 0.280 to 0.328, in females 0.273 to 0.329.
I have seen living T. p. proximus X orarius intergrades from Harris
County and extreme southern Liberty County in Texas, and from St.
James and northern St. Charles Parishes in Louisiana. All have verte-
bral stripes more or less typical of orarius. The St. Charles Parish speci-
mens (UF 12161 from 1 mile N St. Rose) range from fairly typical
brown-dorsum orarius to almost black. The keels of the dorsal scales
are brown, but varying amounts of the remaining area of each scale
are black. Specimens from Refugio, San Patricio, and Victoria Coun-
ties, Texas, also have varying amounts of black pigment on essentially
brown scales. Most of the St. James Parish individuals are entirely
black.
The situation in Harris County is not clear. Two adults (UF 12203)
from Channelview at the head of Galveston Bay have entirely black
dorsa and gold vertebral stripes, but a subadult from the same locality


1963


133







BULLETIN OF THE FLORIDA STATE MUSEUM


appears typical of T. p. orarius, and I have seen five typical orarius
from 9 miles W Tomball in northwestern Harris County, almost 40
miles inland from Channelview. Typical T. p. proximus have been
found less than 70 miles northeastward in Liberty County.
Recognition of intergrades is complicated by the occasional appear-
ance of vertebral stripes with a definite orange tinge as an individual
variation in populations of T. p. orarius well removed from any possi-
ble contact either with T. p. proximus or with intergrades.
I have seen no live intergrades with other subspecies but I suspect
that all of inland southern Texas is an area of intergradation between
T. p. orarius, T. p. rubrilineatus, and T. p. diabolicus. Specimens from
south-central Tamaulipas appear to be intergrades between orarius
and rutiloris, diabolicus and rutiloris, or all three subspecies.
The subspecific name orarius (Latin, of the coast) refers to the fact
that this race lives along the coast.
Thamnophis proximus rubrilineatus, new subspecies
HOLOTYPE. UF 12188.3, an adult male from the State Fish Hatchery,
8.2 miles NW Ingram, Kerr County, Texas, collected 26 June 1960
by M. J. Fouquette, Jr., and D. A. Rossman.
DEFINITION. A subspecies of Thamnophis proximus characterized by
an olive-brown to olive-gray dorsum, a bright red vertebral stripe, and
a narrow dark ventrolateral stripe or none at all.
RANGE. Throughout the Edwards Plateau region of central Texas and
northward an undetermined distance, probably not far beyond San
Angelo or Waco.
DESCRIPTION OF HOLOTYPE. Scutellation: Oculars 1 + 3, temporals 1 +
2, supralabials 8, infralabials 10, intergenials 2, dorsal scale rows 19-19-
17, ventrals 170, subcaudals 107. Body length is 353 mm., tail length
141 mm., tail length/total length ratio 0.285.
The dorsum is olive-brown (Maerz & Paul 15J5), a lighter shade of
brown below the lateral stripe (15L2). The vertebral stripe is red
(4K12) on the posterior three-fourths of the body, orange (9111) on
the neck; the lateral stripe is light yellow (10F1). The venter is pale
green (18F4), the posterior edge of each ventral scute cream yellow;
the dark ventrolateral stripe is lacking. The chin shields are white,
the supralabials pale greenish white. The pale yellow postocular spot
occupies all the lower and leading edge of the middle postocular;
the fused parietal spots are moderately prominent and bright.
DISCUSSION. Counts and measurements of this subspecies may be sum-
marized as: postoculars 2 to 4, temporals 1 + 1 to 2 + 3, supralabials


134


Vol. 7







ROSSMAN: GENUS THAMNOPHIS


7 to 9, infralabials 9 to 11, intergenials 1 to 4, dorsal scale rows 19-19-15
to 19-19-18, male ventrals 163-174, female ventrals 158-175, male sub-
caudals 98-119, female subcaudals 93-116. Body length in males 223
to 493 mm., in females 169 to 728 mm. Total length in males 319 to
711 mm., in females 234 to 994 mm. Adult tail length/total length
ratio in males 0.260 to 0.322, in females 0.268 to 0.326.
In a series of 15 individuals from 8 miles S Oak Hill, Hays County,
Texas, 14 have the usual bright red vertebral stripe; on the other
specimen (UF 12204.1) the stripe is orange its entire length. A series
of 5 snakes from Fern Bank Springs, Hays County, another 5 from a
few miles NW Austin, Travis County, and 7 individuals from scattered
localities in Blanco County all have the red stripe, as do 7 collected,
and others seen but not collected, at the type locality in Kerr County.
One of two specimens from the Devil's River in northern Val Verde
County agrees well with them; in the other (UF 12189.2), the verte-
bral stripe is orange (Maerz & Paul 3B12) on the anterior half of the
body, and bright red (3112) thereafter.
In a series of 17 animals from 6.5 miles S Colorado City, Mitchell
County, 14 have orange vertebral stripes, 1 has a reddish-orange stripe,
and the remaining 2 have red stripes. This series apparently repre-
sents a population of T. p. rubrilineatus X diabolicus intergrades with
a heavy genetic influence from diabolicus.
Stone (1903) reported a red-striped individual from the edge of
the Balcones Escarpment at New Braunfels, Comal County, and Gloyd
(1935) described another from near Floresville, Wilson County. As
the latter locality is some distance from the Escarpment, the animals
may be intergrades rather than typical T. p. rubrilineatus. Strecker
(1908) indicated that the ribbonsnakes in McLennan County usually
have a red vertebral stripe. Brown (1901) reported a red-striped
snake from Pecos, Reeves County, more than 100 miles west of the
westernmost verified localities for red-striped individuals. As all other
records from trans-Pecos Texas are of animals with orange vertebral
stripes, the locality or color data may be inaccurate.
The name rubrilineatus (Latin, red-lined) refers to the bright red
vertebral stripe which characterizes this subspecies.

Thamnophis proximus diabolicus, new subspecies
HOLOTYPE. UF 12210, a large adult female from the Rio Nadadores,
3 miles W Nadadores, Coahuila, Mexico, collected 3 August 1960 by
Roger Conant.
DEFINITION. A subspecies of Thamnophis proximus characterized by
an olive-gray to olive-brown dorsum, an orange vertebral stripe, a







136 BULLETIN OF THE FLORIDA STATE MUSEUM


narrow dark ventrolateral stripe, and frequent reduction in the width
of the lateral stripe on the posterior portion of the body in the southern
part of the range.
RANGE. From west-central Tamaulipas, Nuevo Leon, and eastern Coa-
huila north-northwestward through the Pecos Valley of Texas and
New Mexico and, possibly, the Staked Plains of western Texas and
adjacent New Mexico to extreme southeastern Colorado (Baca
County). Isolated populations may exist in the Rio Grande Valley in
western New Mexico (see Discussion).
DESCRIPTION OF HOLOTYPE. Scutellation: Oculars 1 + 3, temporals
1 + 3, supralabials 8, infralabials 10, intergenials 2, dorsal scale rows
19-19-17, ventrals 160, subcaudals 99. Body length is 771 mm., tail
length 309 mm., tail length/total length ratio 0.286.
Color pattern: Dorsum olive-gray; vertebral stripe of moderate
width, yellowish orange immediately behind the head, a bright orange
thereafter; lateral stripe reduced in width posteriorly; venter grayish
white medially, pale orange marginally; dark ventrolateral stripe occu-
pies 0.10 the width of a ventral scute on each side; chin shields white;
supralabials pale greenish white. The light postocular spot is not bright
and is confined to the lower postocular; the fused parietal spots are
prominent, bright yellow, and broadly margined with black; a black
postorbital vitta is lacking.
Because no color atlas was available when the above notes were
made, the following supplementary notes are presented for a sub-
adult female (UF 12217.1) collected in Huasteca Canyon, near Santa
Catarina, Nuevo Leon, by M. J. Fouquette, Jr., and D. A. Rossman.
Dorsum olive-brown (Maerz & Paul 15J7); vertebral stripe bright
orange (10110); lateral stripe bright yellow (10F1); venter pale
yellow-green (18E1).
DIscussioN. Counts and measurements of this subspecies may be sum-
marized as: postoculars 3 to 4, temporals 1 + 1 to 2 + 3, supralabials
8 to 9, infralabials 9 to 12, intergenials 1 to 4, dorsal scale rows 19-19-15
to 21-19-17, male ventrals 161-171, female ventrals 154-169, male
subcaudals 98-114, female subcaudals 87-102. Body length in males
185 to 415 mm., in females 179 to 771 mm. Total length in males
259 to 598 mm., in females 246 to 1080 mm. Adult tail length/total
length ratio in males 0.285 to 0.307, in females 0.268 to 0.304.
The populations in Nuevo Leon, Coahuila, trans-Pecos Texas, and
southern New Mexico are typical of the subspecies. A single specimen
(USNM 86920) from Union County in extreme northeastern New
Mexico also seems to belong with this group. KU 49585 from Hartley


Vol. 7







ROSSMAN: GENUS THAMNOPHIS


County, Texas, and CUM 11669-673 from Baca County, Colorado,
apparently represent populations of T. p. diabolicus, although the
dorsum may be somewhat darker than usual. The northeastern New
Mexico and Texas individuals were collected on tributaries of the
Canadian River, the Colorado series from a tributary of the Cimar-
ron River. Both these streams flow eastward, yet the snakes show
much less affinity to the populations in the eastern parts of the Cana-
dian and Cimarron drainages than they do to the animals of the Pecos
River drainage. As ribbonsnakes in arid regions are generally re-
stricted to the immediate vicinity of permanent springs and water-
courses, these northern populations of T. p. diabolicus are probably
effectively isolated from those in the Pecos drainage and may be relicts
of an old population that was continuously distributed when condi-
tions were more mesic.
A similar disjunct distribution in this region occurs in Thamnophis
sirtalis (Fitch and Maslin, 1961) which has an isolated population
in the Rio Grande Valley of New Mexico. Possibly T. proximus has
one there too, but the several old records for the area are of question-
able validity. Baird and Girard (1853) listed two T. proximus from
Sabinal and Medina, New Mexico, and discuss in the text specimens
from the Rio Grande Valley which, as they list no other specimens
from anywhere near the Rio Grande, might be the "New Mexico"
specimens. I cannot locate a Medina anywhere in New Mexico, but
there is a Sabinal in northern Socorro County on the west side of
the Rio Grande. F. R. Gehlbach, who is studying the herpetofauna
of the state, suggests these snakes may actually have been collected
in Texas at Sabinal, Uvalde County, and Medina, Bandera County.
The specimens themselves apparently are no longer extant. Two T.
proximus (UMMZ 68751) from "east of Alburquerque" are probably
from the Pecos drainage in San Miguel County where their collector,
an entomologist, took other material now also in the University of
Michigan Museum of Zoology (F. R. Gehlbach, personal communica-
tion). Thus there is no conclusive evidence that the western ribbon-
snake occurs in the Rio Grande Valley.
Specimens from the eastern edge of the Staked Plains appear to be
intergrades between T. p. diabolicus and proximus in the north, dia-
bolicus and rubrilineatus in the south. North-central Texas populations
may be intermediate between all three races.
Intergradation between T. p. diabolicus, rubrilineatus, and orarius,
apparently occurs in varying combinations throughout most of inland
southern Texas. Individuals from the eastern foothills of the Sierra
Madre Oriental in west-central Tamaulipas (as far south as Ciudad


1963







138 BULLETIN OF THE FLORIDA STATE MUSEUM


Victoria) are here considered to represent the southernmost popula-
tion of diabolicus, but living material may show them to be diabolicus
x orarius intergrades. Possible intergrades between diabolicus and
rutiloris are discussed under T. p. rutiloris.
The name diabolicus (Gr. diabolikos, devilish) alludes to a truly
horned inhabitant of hot places. The ribbonsnake's "horns" are strictly
imaginative, suggested by the forward-protruding forked arms of the
parietal spots in many individuals of this inhabitant of arid regions.

Thamnophis proximus rutiloris (Cope), new combination
Thamnophis sauritus rutiloris: Smith, 1938:5
Thamnophis sauritus chalceus: Dunn, 1940:192
Thamnophis sirtalis chalceus: Klauber, 1948:9
HOLOTYPE. USNM 13906, an adult female from Cozumel Island, Quin-
tana Roo, Mexico, collected 23-29 January 1885 by personnel of the
U.S. Fish Commission vessel Albatross (Cochran, 1961).
DEFINITION. A subspecies of Thamnophis proximus characterized by
an olive-brown dorsum, a broad grayish-tan vertebral stripe, yellow-
orange labials, a narrow lateral stripe, and a moderately narrow ven-
trolateral stripe.
RANGE. From extreme southern Tamaulipas south along the coastal
plain, eastern rim of the Sierra Madre Oriental, and Central Ameri-
can highlands to Cartago, Costa Rica. It occurs in the Pacific lowlands
only at the Isthmus of Tehuantepec and in central Guerrero. There
are no records for the highlands of Guatemala.
DESCRIPTION OF HOLOTYPE. Scutellation: Oculars 1 + 3, temporals
1 + 1, 2 (the left anterior temporal is fragmented anteriorly), suprala-
bials 8, infralabials 10, intergenials 2, dorsal scale rows 19-19-17, ven-
trals 147, subcaudals 97 (tip damaged but apparently complete-in
the original description Cope reported 92 subcaudals). Body length
is 520 mm., tail length 215 mm., tail length/total length ratio 0.293.
(Cope, in 1885, recorded a body length of 539 mm. and a tail length
of 214 mm., the discrepancy doubtless resulting from shrinkage during
75 years storage in preservative.)
Cope (1885) described the holotype as having a brownish-olive
dorsum, vertebral stripe existing only as a faint trace, lateral stripe
pale olive, venter pale olive, supralabials, infralabials, and first 3 ven-
trals reddish orange or salmon, a pair of light parietal spots present.
Whether or not this description was of the living animal is not clear,
hence the following description of UF 12215, an adult female collected
13 miles WNW Veracruz, Veracruz, by M. J. Fouquette, Jr., and


Vol. 7







ROSSMAN: GENUS THAMNOPHIS


D. A. Rossman. Dorsum olive-tan in life, somewhat lighter below the
lateral stripe; vertebral stripe grayish tan, broad; lateral stripe pale
yellow, narrow, and slightly reduced in width posteriorly; venter
pale yellow at neck becoming pale green posteriorly; dark ventro-
lateral stripe narrow, occupying 0.10 the width of a ventral on each
side; chin shields and lower margin of infralabials white; supralabials
and upper margin of infralabials bright yellow (10 other live T. p.
rutiloris have yellow-orange labials). The light postocular spot occu-
pies the lower and leading edge of the middle postocular and is not
brightly colored; the fused parietal spots are prominent, bright yellow,
and broadly margined with black. The black postorbital vitta is nar-
row, irregular, and confined to the temporals.
DISCUSSION. Counts and measurements of T. p. rutiloris may be sum-
marized as: postoculars 2 to 5, temporals 1 + 1 to 2 + 3, supralabials
7 to 9, infralabials 8 to 11, intergenials 1 to 6, dorsal scale rows
17-18-17 to 21-19-17, male ventrals 1422-164, female ventrals 141-160,
male subcaudals 91-104, female subcaudals 82-98. Body length in males
157.5 to 445 mm., in females 153 to 556 mm. Total length in males
222 to 625 mm., in females 210.5 to 776 mm. Adult tail length/total
length ratio in males 0.286 to 0.307, in females 0.275 to 0.306.
Three specimens from between El Limon and Chamal in extreme
southwestern Tamaulipas apparently represent intergrades between
T. p. rutiloris and diabolicus. The two females (UMMZ 101222,
101908) have the broad vertebral stripe characteristic of rutiloris, but
in life it almost surely was of some color other than tan (presumably
orange). Their ventral counts (158, 160) are well within the normal
range of variation for Mexican diabolicus. UMMZ 110815, a male, has
162 ventrals and a vertebral stripe which apparently was tan, the
former characteristic of diabolicus, the latter of rutiloris.
A series of snakes (CNHM 114582-586, UMMZ 102893-894,
UIMNH 18926-928, USNM 105305) from the Sierra de Tamaulipas on
the Tropic of Cancer appears to combine the color pattern of T. p.
orarius (or orarius X diabolicus intergrades) with ventral counts much
closer to those of rutiloris. Snakes from the immediate vicinity of
Ciudad Mante appear to have the color pattern of rutiloris, but the
males have a rather high ventral count (the females have a ventral
count comparable to those of female rutiloris 200 miles to the south).
The populations in eastern San Luis Potosi and near Tampico are
considered to be rutiloris.
The only available specimen of the possibly relict population in
coastal Guerrero (UF 12213) has a moderately wide vertebral stripe
that definitely was not tan in life, but rather seems to have been a


1963







BULLETIN OF THE FLORIDA STATE MUSEUM


pale yellow. Furthermore, the lateral stripe is largely confined to row
3 as in T. p. alpinus. On the other hand, the specimen agrees with
lowland rutiloris in scutellation, color of dorsum, width of ventro-
lateral stripe, and proportional muzzle length. Without additional
material (particularly live specimens) to confirm the distinctiveness
of this population, it is pointless to give it taxonomic recognition.
An adult male (TCWC 865) taken near Las Vigas, Veracruz, at an
altitude of 8000 feet has been tentatively assigned to this subspecies
largely on the basis of locality. The animal has 166 ventrals, 110 sub-
caudals, a tail length/total length value of 0.313, an unwidened ver-
tebral stripe, and a large pair of fused parietal spots. Preservation has
so darkened the specimen that nothing can be determined about its
original color. The tail is longer than in any other specimen examined
from south of the Rio Grande, and no other rutiloris possesses so high
a ventral or subcaudal count or as large a pair of parietal spots.
Whether this individual is aberrant or represents a distinctive high-
altitude population remains to be determined.
A specimen (ANSP 20046) from the mountains of south-central San
Luis Potosi near Zaragoza (km. 42, Potosi-Rio Verde Railroad) is
rather far removed from the range of T. p. rutiloris as presently under-
stood, and this record needs substantiation.
Found all the way from sea level to an altitude of 8000 feet, this
neotropical subspecies of T. proximus appears to have an astounding
range of temperature tolerance.

Thamnophis proximus alpinus, new subspecies
HOLOTYPE. UF 12216, an adult male from the trout hatchery at San
Cristobal Las Casas, Chiapas, Mexico, collected 1 August 1960 by
M. J. Fouquette, Jr., and D. A. Rossman.
DEFINITION. A subspecies of Thamnophis proximus characterized by
a dark brown dorsum, a gold vertebral stripe, yellow-orange labials,
a narrow lateral stripe, chevron-shaped parietal spots, a broad ven-
trolateral stripe, and a relatively short tail.
RANGE. Chiapas Highlands from San Cristobal Las Casas to Comitan
at elevations above 5000 feet.
DESCRIPTION OF HOLOTYPE. Scutellation: Oculars 1 + 3, temporals
1 + 2 (the anterior temporal is fragmented anteriorly on both sides
of the head), supralabials 8, infralabials 10, intergenials 2, dorsal scale
rows 19-19-17, ventrals 163, subcaudal series and tail incomplete. Body
length is 499 mm.


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


The dorsum is dark brown (Maerz & Paul 8L12), olive-brown below
the lateral stripe (15J10). The vertebral stripe is gold (12K6) and
only moderately wide; the lateral stripe is light yellow (10H1) over-
laid with brown pigment, narrow, and present only on row 3 on the
posterior part of the body. The venter is pale green (17E4), pale
yellow at the neck; the dark ventrolateral stripe is broad, 0.25 the
width of a ventral scute on each side (on the tail it covers the entire
surface of the subcaudals). The chin shields are white, irregularly
spotted with pale orange; the supralabials and infralabials are pale
yellow-orange, but on the supralabials much brown pigment overlays
the yellow-orange (the anterior 3 plates are completely overlaid and
varying amounts are irregularly distributed on the posterior 5). The
light postocular spot occupies all of the lower and the anterior part
of the middle postocular, and is overlaid with brown pigment; the
fused, chevron-shaped parietal spots are prominent and pale yellow
overlaid with brown.
DISCUSSION. Counts and measurements of this subspecies may be
summarized as: postoculars 2 to 4, temporals 1 + 1 to 2 + 3, supra-
labials 7 to 9, infralabials 9 to 11, intergenials 2 to 5, dorsal scale rows
19-19-17 to 20-19-17, male ventrals 157-165, female ventrals 153-163,
male subcaudals 99-102, female subcaudals 86-99. Body length in males
175 to 499 mm., in females 141.5 to 528. Total length in males 241.5
to 642 mm., in females 194 to 722 mm. Adult tail length/total length
ratio in males 0.268 to 0.280, in females 0.252 to 0.283.
I have seen no intergrades between T. p. alpinus and rutiloris, but
they should occur at lower elevations on the northern and western
slopes of the Chiapas Highlands. T. p. alpinus shows its relationship
to the widespread rutiloris in supralabial color and lateral stripe width.
Its darker dorsum, reduced width of vertebral stripe, and the dark
pigment overlaying the lighter markings are apparently adaptations
to increase absorption of solar radiation in an environment where low
temperatures severely restrict the activity of ectotherms.
The frequency of such scutellation variations as three intergenials
and fragmentation of the anterior temporal may result either from
exposure of the mother to low temperatures during the gestation
period (as demonstrated for Thamnophis elegans by Fox, 1948) or
from the genetic intensification of a normally rare variation in an
isolated population.
The subspecific name alpinus (Latin, of the Alps) refers to the high
altitudes where this population lives.








142 BULLETIN OF THE FLORIDA STATE MUSEUM


NATURAL HISTORY
This brief summary of the known ecology of T. proximus is by no
means complete or exhaustive. Many aspects are treated in much
greater detail elsewhere, particularly by Tinkle who has produced
(1957) the most comprehensive study to date.
HABITAT. My personal experience has been that T. proximus is rather
closely confined throughout its range to the vicinity of permanent or
semipermanent bodies of water, either running or standing, which
most literature accounts substantiate. Neill and Allen (1959), how-
ever, collected a large adult beneath a log in a high, dry savanna in
British Honduras. The species seems to prefer open grassy situations,
but I have also found it in wet woodlands, at a desert waterhole, and
in rush-lined ponds. Ruthven (1908) reported ribbonsnakes from a
brackish mangrove swamp at Progreso, Yucatan. Few authors have
referred to the semiarboreal habits of this species. Specimens of
T. proximus have been observed in bushes in British Honduras
(Dumeril and Bocourt, 1893), Kansas (Gloyd, 1926), and Nebraska
(Hudson, 1942). I have collected individuals several feet above the
water in rushes, one more than 6 feet up, in Chiapas and San Luis
Potosi.
FooD. The diet of the western ribbonsnake is chiefly confined to fishes
and to amphibians and their larvae. Table 8 lists species Thamnophis
proximus is known to have eaten in the wild. Taylor (1892) and
Fouquette (1954) recorded insects in the diet, but these may have
been secondary ingestions. Wright and Wright (1957) mentioned
an account of T. proximus eating mice, which is probably based
on Trowbridge's (1937) report of finding a T. pro.imuiis in a mouse
nest in a log-the snake contained Rana pipiens, however, not mice.
Endothermic vertebrates, lizards, earthworms and other inverte-
brates, all of which one or more species of gartersnake eat readily,
are apparently not normal prey of T. proximus. Moreover, while
captive ribbonsnakes readily accept newly transformed toads, they
almost always refuse adult Bufo. Possibly members of the Sauritus
group are not so tolerant as other gartersnakes to the toxic parotoid
secretions of adult toads.
When food is plentiful, a ribbonsnake may continue eating as long
as the supply lasts. Laughlin (1959) reported that a single T.
proximus contained 64 metamorphosing toads of an unidentified spe-
cies; and Klein (1949) observed that ribbonsnakes (and several species
of watersnakes) feeding on aggregations of fish, tadpoles, and frogs
in pools left by a drop in water level of the San Antonio River (Texas)


Vol. 7









ROSSMAN: GENUS THAMNOPHIS


gorged themselves to the
the juveniles, died.


point that many of the snakes, especially


TABLE 8. Food of Thamnophis proximus


Source


Anurans
Scaphiopus bombifrons
Bufo woodhousei
Acris crepitans
Acris gryllus
Hyla crucifer
Hyla versicolor
Smilisca baudini
Pseudacris clarki
Pseudacris streckeri
Pseudacris triseriata
Rana catesbeiana
Rana clamitans
Rana pipiens
Leptodactylus labialis
Hypopachus cuneus
Microhyla carolinensis
Microhyla olivacea (larvae only)

Salamanders
Ambystoma texanum
Plethodon cinereus

Fishes
Salmo gairdneri
Gambusia afinis
Lepomis megalotis
Hericthys cyanoguttatus


Fouquette, 1954
Fouquette, 1954
Fouquette, 1954
Boyer and Heinze, 1934
Fouquette, 1954
Gloyd, 1926; Fouquette, 1954
M. J. Fouquette, Jr., personal communication
Fouquette, 1954
Fouquette, 1954
Gloyd, 1926; Fouquette, 1954
Fouquette, 1954
Fouquette, 1954
Trowbridge, 1937; Clark, 1949; Fouquette, 1954
personal observation
Wright and Wright, 1957
personal observation
Fouquette, 1954


Fouquette, 1954
C. W. Myers, personal communication


Myers, 1957
Fouquette, 1954
Fouquette, 1954
Fouquette, 1954


Ribbonsnakes are as active by night as by day where nocturnal
temperatures permit, and apparently feed as readily at one time as
the other. Near Ciudad Mante in Tamaulipas, M. J. Fouquette, Jr.
(personal communication) watched a T. proximus stalk a calling
Smilisca baudini. When the frog's vocal sacs moved, the snake moved
closer, stopping each time the frog stopped calling.

REPRODUCTION. Tinkle (1957) found that ribbonsnakes in southeastern
Louisiana reach sexual maturity in approximately 2 years and at mini-
mum body lengths of 485 mm. in females and 410 mm. in males.
Gloyd (1926) reported a Kansas female that bore young at 435 mm.
body length, and Carpenter (1958) noted one from Oklahoma bearing


Species


1963








144 BULLETIN OF THE FLORIDA STATE MUSEUM Vol. 7

at 460 mm. Clark (1949) recorded a brood from a 20-inch female
from northern Louisiana; if this is total length and the animal's tail
was complete, its body length was probably between 350 and 360 mm.
The smallest gravid females I have examined measured 355 (British
Honduras), 385 (Veracruz), 390 (Veracruz), 442 (southeastern
Texas), 450.5 (Veracruz), and 454 (San Luis Potosi) mm. These
measurements suggest that ribbonsnakes achieve sexual maturity
more rapidly and at smaller sizes in the tropics. Extrapolating from
these and Tinkle's data, males in Mexico and Central America proba-
bly mature at body lengths of 300 to 350 mm.
Where ribbonsnakes are inactive in winter, mating apparently takes
place soon after the spring emergence. Carpenter (1958), Neill and
Allen (1959), and Tinkle (1957) all reported females gravid in April.
The young are usually born in late July or early August, but dates
range from 23 June (R. D. Worthington, personal communication)
through 14 September (Carpenter, 1958). Brood size tends to increase
with the size of the female, and ranges from 4 (Carpenter, 1958) to
27 (Klein, 1949), the mean for 34 broods being 12.8. Table 9 sum-
marizes available data on body length of newly born T. proximus.
Size of individuals at birth varies greatly between broods, but no

TABLE 9. Body length of newly born Thamnophis proximus

Locality N M R

St. Charles Parish, La.1 ................ 22 202.6mm 187 -220mm
St. Charles Parish, La.2 .................. 5cc 201.7 188 -214
39 9 198.0 190.5-206

2dc 177.8 176.5-179
29 9 173.3 170 -176.5
Alexander Co., Illinois3. ............... 8 157.3 151 -162

Southern Oklahoma4................... 5cc 206 192 -225
79 9 201 199 -204

40'c 170 165 -174

7cc 159 156 -160
29 9 152 148 -156

Eastern San Luis Potosi2. ............... l. 1 166
49 9 166 164 -168

N = number of individuals; M = mean; R = range
1 Data from Tinkle, 1957 3 Data from N. J. Rossman, 1958
2 Personal observation 4 Data from Carpenter, 1958








ROSSMAN: GENUS THAMNOPHIS


geographical correlation is readily apparent. Males, though smaller as
adults, average larger than females at birth.

PARASITES. Harwood (1932) reported finding the helminths Rhabdias
vellardi and Kalicephalus agkistrodontis in the lungs and stomach of
Thamnophis proximus. On MCZ 53912, an adult female proximus
taken near Merida, Yucatan, I found a small tick embedded between
scales on the left side of the head. George Anastos identified it as a
nymphal stage of the genus Ixodes (species undetermined).

THE EASTERN RIBBONSNAKE
THAMNOPHIS SAURITUS (LINNAEUS)
Coluber saurita Linnaeus, 1766:385
Natrix saurita: Merrem, 1820:122
Tropidonotus saurita: Boie, 1827:535
Leptophis sauritus: Holbrook, 1842:21
Thamnophis saurita: Fitzinger, 1843:26 [by implication]
Eutainia saurita: Baird & Girard, 1853:24
Eutaenia saurita: Kennicott, 1859:98
Eutaenia sackenii Kennicott, 1859:98
Prymnomiodon chalceus Cope, 1860:558
Eutaenia sackeni: Davis & Rice, 1883:39
Thamnophis sackenii: Loennberg, 1894:329
Thamnophis sauritus: Stone, 1906:164
Thamnophis sackeni: Deckert, 1918:31
Thamnophis sirtalis: Klauber, 1948:9
Klauber showed that Linnaeus's description of Coluber sirtalis does not fit the
common gartersnake, and that the name probably should be applied to the
eastern ribbonsnake. For several years some authors followed Klauber; others
did not. The resulting confusion led to a request that the International Com-
mission on Zoological Nomenclature conserve the pre-Klauber arrangement
in the interest of nomenclatural stability. The Commission complied and neo-
types were subsequently designated (Schmidt and Conant, 1956-57).
NEOTYPE. As the holotype (collected in "Carolina") is presumed
lost, Schmidt and Conant (1956-57) designated as neotype an adult
female, CNHM 73119, collected 15 miles NNE Charleston, Berkeley
County, South Carolina, 20 July 1953, by John Quinby.

DEFINITION. A moderately large, long-tailed member of the genus
Thamnophis characterized by: 19-19-17 dorsal scale rows; a single
preocular; 7 or 8 supralabials, the 3rd and 4th or 4th and 5th, respec-
tively, entering the orbit; 143 to 177 ventrals; 94 to 136 subcaudals;
lateral stripe on dorsal scale rows 3 and 4 (row 2 is occasionally in-
volved); labials and ventrals without black markings; dark ventro-
lateral stripe always present and broad; parietal spots frequently


145


1963









146 BULLETIN OF THE FLORIDA STATE MUSEUM


lacking, when present they are small and rarely in contact or brightly
colored; hemipenes short, extending to the 8th subcaudal when in-
verted; teeth numerous, averaging about 30 to each maxilla, 37 to
each dentary, 21 to each palatine, and 34 to each pterygoid.
RANGE. From southern Ontario and southern Maine southward east
of the Mississippi River to the Florida Keys and to the northern side
of Lake Ponchartrain in Louisiana (fig. 8). A colony is isolated in


9 190 290 390 490
MILES


FIGURE 8. Distribution of Thamnophis sauritus in the United States and Canada.
Solid dots represent specimens examined; open circles are literature records.
Stippled areas represent probable zones of intergradation between subspecies.


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


central Nova Scotia, and perhaps another in northern Wisconsin. The
species apparently is rare in much of the interior of the eastern United
States.
VARIATION
Sexual Dimorphism
Thamnophis sauritus shows sexual dimorphism in the same characters
as does T. proximus, and to about the same degree (see table 3). In
populations for which at least five snakes of each sex are available, the
differences in mean number of ventrals range from 0.8 scales in favor
of the females (Charleston County, South Carolina) to 5.7 in favor
of the males. While the greatest dimorphism occurs in western penin-
sular Florida, the degree of dimorphism does not appear to vary
geographically. Populations represented by at least 15 adults of each
sex are compared in table 10.

TABLE 10. Sexual dimorphism in ventrals of Thamnophis sauritus

Males Differences Females
Population
N M DM CD% N M

Southwestern Florida...... 20 164.70.83 5.4 3.330.13 27 159.30.63
North-central Florida...... 15 157.80.99 3.6 2.31-0.19 29 154.20.74
Southern Michigan........ 16 163.1-0.58 2.4 1.480.09 25 160.70.60
N = number of individuals; M mean; DM = difference between means; CD% = coefficient
of divergence expressed as a percent.

Sexual dimorphism in subcaudal number also agrees well with that
exhibited by T. proximus (table 5), but does not vary so widely. In
no samples with at least five specimens of each sex do males average
fewer subcaudals than the females. The mean differences range from
3.4 in south-central Mississippi to 10.2 in southeastern Pennsylvania.
This character does not appear to vary geographically. Populations
represented by at least 10 adults of each sex are compared in table 11.

TABLE 11. Sexual dimorphism in subcaudals of Thamnophis sauritus

Males Differences Females
Population -- -- -------
N M DM CD% N M

Southwestern Florida...... 11 126.31.52 7.9 6.050.39 16 118.40.96
North-central Florida ...... 12 123.31.23 7.7 6.450.39 18 115.61.26
Southern Michigan........ 10 108.31.16 5.3 5.020.38 16 103.01.29
N = number of individuals; M = mean; DM = difference between means; CD% = coefficient
of divergence expressed as a percent.


1963









BULLETIN OF THE FLORIDA STATE MUSEUM


Females average longer than the males but have proportionally
shorter tails. The largest female examined (UF 12120 from Dade
County, Florida) has a body length of 670 mm., a tail of 348 mm. The
largest male (UMMZ 108313), also from Dade County, has a body
537 mm. long and a tail 283 mm. long. Wright and Wright (1957)
cited a New Jersey male with a total length of 842 mm.
The mean differences in adult tail length/total length ratios ranges
from 0.007 to 0.014. Males in all populations have proportionally
longer tails than the females. Mean tail length/total length values
range from 0.313 to 0.378 in males, 0.299 to 0.350 in females.
Although females have proportionally longer muzzles than the males
in 5 of 7 populations in which muzzle length/frontal length values are
available for at least 4 adults of each sex, the means for all popula-
tions measured are almost identical. In males they range from 0.628
to 0.715 (mean 0.666), in females from 0.607 to 0.707 (mean 0.669).
Caudodorsal scale row reduction occurs more posteriorly in males
than in females. The reduction formulas for UF 12187.4, an adult male
from Charleston County, South Carolina, and UF 12183, an adult
female from Hyde County, North Carolina, are presented below.

UF 12187.4 2(1) 2 ? 3(3) 7(4) 2 ? 3(5) -4(18) -3(41)
17 15- 13 12 10---- 8
0 3(1) 2(3) 2 ? 3(5) 4(18) 3(42)
2+3(85) --2(112)
6---- 4- 2(117)
3(85) 2(111)

UF 12183 1(1 2(2) 3(6) 3(27) 3(70) 2(100)
14-- 12 -10_-- 8 6- -- 4
9 1(2) 2? 3(3) 3+ 4(7) 3(30) 3(69) 2(100)
2(107)

Ontogenetic Variation
Newly born T. sauritus have proportionally shorter tails than do adults
(table 6). Coefficients of ontogenetic divergence are comparable with
those obtained for one population of T. proximus except for southern
Florida females which exhibit almost no change in proportional tail
length from birth to adulthood (table 6). Females of two broods from
southern Florida (Broward and Collier Counties) are unique in hav-
ing proportionally longer tails than the males of the same brood; the
situation is reversed in the adults.
All juveniles have a brown dorsum; there is a progressive darken-
ing with age in those populations in which adults have either a black
or dark brown dorsum.


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


Individual Variation
Characters which vary neither geographically, nor sexually, nor onto-
genetically are discussed below. Data on head scales refer to one side
only. Individual variation in scutellation is summarized in table 12.
OCULARS. None of the 957 T. sauritus examined has more than one
preocular. Postoculars are usually 3, with 4 the most frequent variation.
TEMPORALS. There are usually 1 anterior and 2 posterior temporals,
but this character is highly variable. Unlike T. proximus, no population
shows a high incidence of fragmentation of the anterior portion of the
anterior temporal.
SUPRALABIALS. Though the number of supralabials varies geographi-
cally, the infrequent interposition of small scales between two of the
normal supralabials does not.

TABLE 12. Individual variation in scutellation of Thamnophis sauritus
Pre- Post- Anterior Posterior Supra- Infra- Inter-
oculars oculars temporals temporals labials labials genials
ce"l*(363) 2,3 (2) 1 (325) 1 (5) 6,7 (2) 7,8 1 (21)
3 (339) 1,2 (14) 1,2 (7) 7 (228) 8 (2) 2 (293)
3,4 (17) 2 (21) 1,3 7,8 (37) 8,9 (3) 3 (28)
3,5 2 (301) 8 (90) 8,10 4
4 (4) 2,3 (33) 9 9 (11)
3 (11) 9,10(32)
3,4 (2) 9,11 (3)
10 (288)
10,11(14)
9 9 1*(594) 2 1 (481) 1 (17) 6 8 (4) 0 (2)
2,3 (10) 1,2 (56) 1,2 (13) 6,7 (3) 8,9 (6) 1 (34)
3 (537) 2 (50) 1,3 (2) 7 (327) 8,10 (4) 2 (436)
3,4 (32) 2 (402) 7,8 (51) 9 (34) 3 (68)
4 (13) 2,3 (105) 8 (209) 9,10(56) 4 (2)
4,5 3 (47) 8,9 9,11
3,4 10 (464)
10,11(11)
10,12
11
c&c' Dorsal scale rows 9 9
17-19-17 (3) 17-18-17
19-17-17 (2) 18-19-17
19-18-17 19-17-16 19-19-17 (536)
19-19-16 (3) 19-19-18 19-17-17 (19) 19-19-18
19-19-17 (327) 19-19-19 19-19-16 (5) 19-19-19
19-21-18 21-19-17
Single numbers indicate that both sides of the head have the same count.


1963


149








150 BULLETIN OF THE FLORIDA STATE MUSEUM


INFRALABIALS. Ten is the usual number, 9 the most frequent variation.
Infralabial number is more variable in sauritus than in proximus;
10 individual sauritus have two more infralabials on one side than
on the other; no proximus shows so large a bilateral difference.
INTERGENIALS. Two is the usual condition, three the most frequent
variation, but of low incidence in all populations.
ANAL PLATE. In no specimen examined is the anal plate divided.
VENTRALS. Anomalous scutes between the terminal ventral and the
anal plate occur in 9.0 percent of the specimens examined; 4.6 per-
cent, including a few of these, have partial scutes elsewhere on the
venter.
SUBCAUDALS. Fusion of one or more pairs of subcaudals occurs in 2.2
percent of the material examined.
DORSAL SCALE ROWS. Almost all individuals examined have the typical
19-19-17 formula; only two animals have more than 19 rows. Reduction
takes place by loss of row 4 on both sides of the body in 36 of the 58
specimens examined, and the 4th row is involved in the reduction on
one side in all but 3 of the rest, which lost the 5th row on both sides.
One snake dropped row 6 on one side of the body. The reduction
occurs opposite ventrals 76 through 106 (mean 89.8) in males, 71
through 101 (mean 87.0) in females. Correlation between total num-
ber of ventrals and point of dorsal scale row reduction appears roughly
positive, but not so strongly as in T. proximus.
PATTERN AND COLORATION. Presence or absence and size of parietal
spots vary individually; so do the extent of the black postorbital vitta
and the width of the black paravertebral stripes.

Geographic Variation
Although T. sauritus has a less extensive range than T. proximus, it
exhibits considerable geographic variation; some of its patterns of
character distribution are even more complex than those in proximus.
As in the discussion of the western ribbonsnake, all counts, measure-
ments, and proportions are based on population samples.
SUPRALABIALS. Throughout peninsular Florida to the vicinity of Talla-
hassee in the west and the Okefenokee Swamp region in the north,
and through coastal Georgia north to Jasper County, South Carolina,
91.8 percent of the snakes have 8 supralabials on both sides and an
additional 5.5 percent have 8 on one side. Throughout the rest of the
species' range, only 9.1 percent have 8 on both sides; another 10.8
percent have 8 on one side, 7 on the other. Almost all of the remaining


Vol. 7









ROSSMAN: GENUS THAMNOPHIS


80.1 percent have 7 supralabials on both sides. The few that do not,
have 6 on one or both sides, and all but one of these animals are from
Michigan.
VENTRALS. Variation in number of ventrals is exceedingly complex
and can be best visualized by reference to figure 9. In southeastern


9 IPo 290 30 490 o8.1 (II)
MILES 1646 4
I 159.5(15
---- ^" "


pyV. '~


FIGURE 9. Geographic variation of ventral number in Thamnophis sauritus.
Black blotches represent the areas from which samples were taken. The upper
numbers associated with each blotch are the sample mean and sample size (in
parentheses) for males; the lower numbers present the same data for females.
When a line appears in place of either the upper or lower numbers, it signifies.
that no specimens of that sex were present in the sample.


1963








BULLETIN OF THE FLORIDA STATE MUSEUM


Louisiana and adjacent Mississippi, males have a mean of 158.7
ventrals, females 157.5. Population samples from southeastern Missis-
sippi (exclusive of the Biloxi area) and the Mobile Bay-Pensacola
Bay area agree closely, but the snakes in southwestern Mississippi
have a mean of 6 to 7 more ventrals in males, 5% in females. From
thence north-northeastward to northern Indiana (exclusive of central
Tennessee) the mean number of ventrals in females is almost invari-
able. The mean for males increases slightly over this distance and
reaches its peak (167.6) in central Indiana. In northern Indiana males
the mean is about 2% scales less, and it decreases gradually but steadily
northward in Michigan to the Straits of Mackinac, the mean in
extreme northern Michigan being another 7 scutes lower. The female
mean ventral count drops by 3 scales between northern Indiana and
southern Michigan, then lessens more gradually northward, being
another 5 scales lower in extreme northern Michigan. A series of
females (no males available) from extreme southern Ontario agrees
well with the northern Indiana females, and those from central Ohio
have only slightly lower means. Northeastward from these popula-
tions the mean number of ventrals in both sexes decreases abruptly,
by about 4 scales in populations south of Lake Erie, and then tapers
off gradually in western New York and eastern Ontario.
A series of 7 females (no males available) from the vicinity of
Pittsburgh agrees closely with females from the northeastern Ohio-
northwestern Pennsylvania area in mean ventral number. One hun-
dred miles to the south in eastern West Virginia, the mean number of
ventrals in both sexes decreases abruptly. A clinal reduction in ven-
tral number is also apparent from central Ohio into West Virginia and
eastern Kentucky. The reduction in this case is also abrupt, a decrease
of 11% (females) to 13 (males) taking place over a distance of less
than 200 miles.
The snakes in central Tennessee have means at least 9 scales less
than those of the populations to the west of the Tennessee River. A
series from the vicinity of Biloxi, Mississippi, has a mean of 5 to 6
fewer ventrals than the populations surrounding it and no more than
50 miles distant. Snakes from southeastern Alabama and southern
Georgia (exclusive of the coast) appear to be similar to the Biloxi
animals in mean ventral count, but more specimens are needed to
determine just how well they agree, and whether the two are con-
nected by populations in southwestern Alabama north of Mobile Bay.
In the Apalachicola River Valley 150 miles east of the Mobile Bay-
Pensacola Bay area the mean number of ventrals decreases by more
than 9 scales. Populations just to the north in southern Georgia have


Vol. 7







ROSSMAN: GENUS THAMNOPHIS


means less than 4 scales higher. A single adult female (UF 7385)
from St. Vincents Island off the mouth of the Apalachicola River has
161 ventrals, almost 13 more than the mean for the Apalachicola Val-
ley series. From thence eastward and southward in the western coastal
lowlands to southern Florida, and northward in the eastern coastal
lowlands to about the latitude of Daytona Beach, the mean number
of ventrals is remarkably constant, the majority ranging from 163.3
to 164.8 in males, 159.2 to 160.7 in females (females of the west-
central Florida, and males of the east-central Florida, populations
have means 1 to 2 scales less than those just mentioned).
In the central highlands of Florida and in the eastern coastal low-
lands north of Daytona Beach, ventral counts decrease abruptly, the
means ranging from 156.4 to 157.8 in males, 153.6 to 154.9 in females.
Ventral number falls off sharply to the north in inland Georgia, but
on the coast it varies from the mean range stated for northern Florida
by less than 1 scute in females and less than 3 in males (larger series
of males may somewhat alter this picture) from northeastern Florida
to central New Jersey, thence inland up the west side of the Hudson
River to the vicinity of Albany, New York. On and adjacent to the
Piedmont from the Savannah River to southeastern Pennsylvania, mean
ventral counts are 2 to 4 scales less than on the outer Atlantic Coastal
Plain. In these inland populations the number of mean ventrals in-
creases to the north, but by less than 1% scales in females, somewhat
more in males. These populations agree rather well in mean number
of ventrals with those located in extreme northern Georgia and eastern
West Virginia.
On Long Island and at scattered localities in New York east of the
Hudson River as far north as Lake George, the ventral count in
females rises abruptly to a mean of 158.0; there is virtually no change
in males. As counts in New England females are somewhat lower,
this change does not appear to be associated with any clinal trends.
In central New England means for females range from 154.4 to 156.1;
only the southern New Hampshire sample has a sufficient number of
males to place any reliance on the resulting mean (i.e. 158.0). Thus,
these populations show a slight increase in ventral number over the
relatively stable Atlantic Coastal Plain series, but this trend is abruptly
reversed in southeastern Massachusetts, Maine, and Nova Scotia,
where the means fall between 152.1 and 154.0 in males, 149.3 and
152.0 in females.
SUBCAUDALS. Variation in subcaudal number appears to be almost
chaotic when viewed in terms of local populations, and a meaningful
evaluation is further hampered by having so few individuals with


1963









154 BULLETIN OF THE FLORIDA STATE MUSEUM


complete tails in most series. A generalized pattern can be described,
but see figure 10 for details. Mean subcaudal counts in males almost
always exceed 120 in the area south of the Ohio River on the west,
and as far north as southeastern Pennsylvania, northern New Jersey,
and possibly Connecticut on the east. Within this area the means


9 190 290 390 490
MILES


117.4(80 t.. -- 2 .. N u
^^ \ 1 \J


FIGURE 10. Geographic variation of subcaudal number in Thamnophis sauritus.
Black blotches represent the areas from which samples were taken. The upper
numbers associated with each blotch are the sample mean and sample size (in
parentheses) for males; the lower numbers present the same data for females.
When a line appears in place of either the upper or lower numbers, it signifies
that no specimens of that sex were present in the sample.


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


exceed 125 in eastern Virginia, coastal Georgia, the southern 4/5
of peninsular Florida, the Mobile Bay-Pensacola Bay area (the highest
mean, 131.8, occurs here), southwestern Mississippi, and northern
Mississippi. The mean number decreases to the north, males from
populations in northern Indiana, Michigan, Ontario, and Maine hav-
ing not over 112. Females from this northern area have less than 105
mean ventrals. The mean number in females increases southward,
rarely being less than 112 south of New Hampshire, southeastern
New York, Pennsylvania, Ohio, and central Indiana. The mean ex-
ceeds 120 in females from northeastern Florida, southeastern Florida,
west-central Florida (the highest mean, 122.3, occurs here), Levy
County, Florida, the Mobile Bay-Pensacola Bay area, and northern
Mississippi.
PROPORTIONS. Relative tail length is fairly constant in the area south
of the Ohio River, and in southern Pennsylvania, New Jersey, and
southeastern New York. Means of tail length/total length values in
males usually fall between 0.345 and 0.355, in females between 0.335
and 0.345. Means are in excess of 0.360 in males from central Tennes-
see and the mid-Savannah River area, and as high as 0.378 for a small
series from eastern West Virginia. In three New England series the
mean for males is 0.343 as compared to 0.352 in New Jersey, in
females the respective means are 0.330 and 0.342. The one male avail-
able from'Nova Scotia (NSMS unnumbered) has a value of 0.310.
Males from western New York, northwestern Pennsylvania, Ohio, and
southern Indiana have means ranging from 0.327 to 0.336; means in
females range from 0.317 to 0.319. In northern Indiana and Michigan
means in males range from 0.310 to 0.315, in females from 0.299 to
0.308. In eastern Ontario the mean for females is 0.312.
PATTERN. The vertebral stripe is frequently poorly developed or lack-
ing in the snakes inhabiting peninsular Florida and southeastern
Georgia. As the stripe in this area is tan, often only a shade lighter
than the dorsum, it tends to disappear with the loss of its narrow,
black borders, though it is almost always at least faintly indicated
for a short distance immediately behind the head. Throughout the rest
of the species' range the vertebral stripe is always present and dis-
tinctly margined by the black paravertebral stripes, which may vary
greatly in width. In the Great Lakes region, where the dorsum is
usually black or nearly black, the stripe is often narrower, involving
in addition to all of the vertebral row, somewhat less than half the
width of each paravertebral row. The vertebral stripe in T. proximus
is also often narrower in the northern part of the range.


1963


155








BULLETIN OF THE FLORIDA STATE MUSEUM


In peninsular Florida and southeastern Georgia the lateral stripe
is narrow, always occupying less than half of one of the two rows
bearing it, and usually less than half of both rows. In the Great Lakes
region frequently the lateral stripe slightly involves dorsal scale row 2
on the anterior part of the body in addition to the normal 3 and 4,
thus paralleling the condition typical of Thamnophis butler, which
inhabits part of this same area. While the condition is not entirely
characteristic of the eastern ribbonsnake in this region, it does occur
here with much higher frequency than elsewhere in the species' range.
Individuals with an anteriorly widened lateral stripe have also been
observed in samples from central Tennessee, eastern West Virginia,
and western Pennsylvania, where they are not associated with a
black dorsum. CAS 15680 from West Virginia and NMC 3090.2 from
Nova Scotia have rows 3, 4, and 5 incorporated in the lateral stripe.
COLOR. From western Levy County, Florida, northwestward along
the Gulf Coast to a point almost due south of Tallahassee, the dorsum
is usually dark brown to black. It is also black or dark brown in Nova
Scotia, Maine, and the Great Lakes region. Throughout the rest of
the range, except in peninsular Florida and coastal Georgia where it
is tan or light brown, the dorsum is a reddish brown.
In ribbonsnakes from peninsular Florida and southeastern Georgia
the vertebral stripe is usually some shade of tan, occasionally with a
metallic luster, or it is lacking. Outside this area it is usually some
shade of yellow, occasionally overlaid with dark pigment in the Great
Lakes region.
The lateral stripe is yellow throughout most of the species' range;
it is occasionally white at scattered localities on the Florida peninsula.
In western Levy County and in the coastal lowlands to the northwest
it is usually light blue or bluish white.
DIscussION. Geographic variation is, for the most part, of the dis-
cordant type discussed previously under T. proximus. Like that species,
T. sauritus has two major zones of concordant character transition.
An abrupt change in supralabial number, lateral stripe width, ver-
tebral stripe color, and a slighter one in color of the dorsum takes
place between the peninsular Florida-southeastern Georgia area and
the region to the north and west. This transition may be related to the
postulated isolation of much of peninsular Florida from the mainland
in the form of islands during shifts in the level of Pleistocene seas
(see Neill, 1957, for a detailed discussion of the effects of this isolation
on vertebrate evolution).
The other transition zone appears associated to some degree with


156


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


the southern limits of Pleistocene glaciation. The animals occupying
glaciated terrain have darker dorsa, relatively shorter tails, and fewer
subcaudals than the populations south of the glacial boundary. This
correlation breaks down in northern New Jersey, eastern New York,
and southern New England where long-tailed snakes with brown
dorsa and high subcaudal counts range well into glaciated territory.
These may, however, represent a fairly recent invasion.
The elements of color pattern have less observable geographic vari-
ability in T. sauritus than do the numbers of ventrals or subcaudals.

TAXONOMY
The population samples examined appear to represent four major
populations which replace each other geographically. These are de-
fined chiefly on the basis of differences in color pattern, to a lesser
degree in scutellation and proportions. The salient features of these
populations are discussed below within their formal descriptions as
subspecies.
Subspecies of Thamnophis sauritus
The following key for distinguishing adults of the four races of T.
sauritus was designed primarily for use with live material. If only
preserved specimens are available, attention should be paid to the
geographical source of the sample.
1. Supralabials 7, occasionally 8; vertebral stripe some shade of yellow; lateral
stripe occupies % or more of both rows 3 and 4 -------- --------2
Supralabials 8, rarely 7; vertebral stripe tan, or lacking; lateral stripe narrow,
occupying less than 3 of either row 3 or 4, or both 3--------------3
2. Dorsum reddish brown; lateral stripe rarely widened anteriorly; tail rela-
tively long (mean tail length/total length usually exceeding 0.340 in
males, 0.330 in females) __--_-- ___----- ---------- T. s. sauritus
Dorsum velvety black or dark brown; lateral stripe frequently widened an-
teriorly, involving row 2; tail relatively short (mean tail length/total length
usually not exceeding 0.335 in males, 0.325 in females) T. s. septentrionalis
3. Dorsum tan or brown, rarely dark; lateral stripe yellow, occasionally
white .---------T----------------------------- T. s. sackenii
Dorsum dark brown or black; lateral stripe bluish white or light blue, occa-
sionally white ----------- ---------- --------- ........ -T. s. nitae

Thamnophis sauritus sauritus (Linnaeus)
Thamnophis sauritus sauritus: Ruthven, 1908:122 [by implication]
Thamnophis sirtalis sirtalis: Klauber, 1948:9
NEOTYPE. The data presented for the species apply here.
DEFINITION. A subspecies of Thamnophis sauritus characterized by 7


1963








BULLETIN OF THE FLORIDA STATE MUSEUM


supralabials, a reddish-brown dorsum, a golden yellow vertebral
stripe, and a relatively long tail.
RANGE. From southern New England, eastern New York, New Jersey,
and eastern Pennsylvania, west-southwestward through southern Ohio
and southern Indiana to the Mississippi River, south to the northern
edge of Lake Ponchartrain on the west and to northwestern Florida
and Georgia (exclusive of the southeastern third of the state) on the
east.
DESCRIPTION OF NEOTYPE. Scutellation and proportions (from Schmidt
and Conant, 1956-57): Oculars 1 + 3, temporals 1 + 3, supralabials
7, infralabials 10, dorsal scale rows 19-19-17, ventrals 156, subcaudals
113. Body length 388 mm., tail length 202 mm., tail length/total length
ratio 0.342.
As Schmidt and Conant's (1956-57) description of the neotype is
based on a preserved specimen, the following color notes are pre-
sented for UF 12187.1, an adult male collected 1.5 miles N Ravenel,
Charleston County, South Carolina, by N. J. and D. A. Rossman. Dor-
sum reddish brown (Maerz and Paul 14B12), slightly lighter below
the lateral stripe; vertebral stripe golden yellow (12J2); lateral stripe
light yellow (10H1); venter cream (9F1); dark ventrolateral stripe
broad, occupying 0.25 the width of a ventral scute on each side; chin
shields and infralabials white, supralabials yellowish white. The light
postocular spot is confined to the lower postocular; the black postorbi-
tal vitta is broad. The tiny parietal spots are separate and yellow,
margined narrowly with black.
DISCUSSION. Counts and measurements of the nominate race may be
summarized as: postoculars 2 to 4, temporals 1 + 1 to 2 + 3, suprala-
bials 6 to 8, infralabials 8 to 11, intergenials 1 to 4, dorsal scale rows
17-18-17 to 19-19-17, male ventrals 145-166, female ventrals 143-169,
male subcaudals 106-136, female subcaudals 96-130. Body length in
males ranges from 124.5 to 451 mm., in females 129 to 636 mm. Total
length in males 185 to 665 mm., in females 190 to 956 mm. Adult tail
length/total length ratio in males 0.333 to 0.388, in females 0.317 to
0.388.
My concept of the color pattern in T. s. sauritus is based on living
specimens from Cape May, Sussex, and Warren Counties, New Jer-
sey; Hyde County, North Carolina; Charleston County, South Caro-
lina; Barbour County, Alabama; and Forrest County, Mississippi, as
well as on a colored slide of a snake from Burlington County, New
Jersey, and color notes on one from Tolland County, Connecticut.
Intergradation involving the nominate race is discussed later.


158


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


The subspecies sauritus is wide ranging and apparently thrives
under a great many environmental conditions. It may have evolved
on the mainland when the ancestral stock of the species (of which
T. s. sackenii probably is closest to being the present-day representa-
tive) was isolated on one or more of the islands that constituted Flor-
ida at various times during the Pleistocene.

Thamnophis sauritus septentrionalis, new subspecies
HOLOTYPE. UF 12179.2, an adult female from Michigan Hollow, near
Ithaca, Tompkins County, New York, collected 20 May 1959 by J. A.
Bartley.
DEFINITION. A subspecies of Thamnophis sauritus characterized by 7
supralabials, a velvety black or dark brown dorsum, a yellow vertebral
stripe often overlaid with brown pigment, and a relatively short tail.
RANGE. From Nova Scotia, Maine, and possibly central New Hamp-
shire westward through southern Ontario and western New York to
extreme northern Michigan and central Indiana. There may be an
isolated colony in northeastern Wisconsin (see Discussion).
DESCRIPTION OF HOLOTYPE. Scutellation: Oculars 1 + 3, temporals
1 + 2, supralabials 7, infralabials 10, intergenials 2, dorsal scale rows
19-19-17, ventrals 156, subcaudals 108. Body length is 574 mm., tail
length 271 mm., tail length/total length ratio 0.321.
The dorsum is velvety black on the anterior % of the body, dark
brown posteriorly; below the lateral stripe the dorsum is dark brown
(Maerz and Paul 7E12). The vertebral stripe is yellow, but so heavily
overlaid with brown that it is impossible to determine the exact shade;
the lateral stripe is light yellow (10J1). The venter is pale green
suffused throughout with tan; the dark ventrolateral stripe is broad,
occupying 0.20 the width of a ventral on each side. The chin shields
and infralabials are white, the supralabials a pale greenish white. The
head is black, hence no postorbital vitta is visible; the parietal spots
are lacking.
DIscussioN. Counts and measurements of this subspecies are: postocu-
lars 2 to 5, temporals 1 + 1 to 2 + 3, supralabials 6 to 8, infralabials 7
to 11, intergenials 1 to 3, dorsal scale rows 18-19-17 to 19-21-18, male
ventrals 152-177, female ventrals 148-169, male subcaudals 98-123,
female subcaudals 94-114. Body length in males ranges from 125.5
to 438 mm., in females 128 to 574 mm. Total length in males 179 to
652 mm., in females 184 to 845 mm. Adult tail length/total length
ratio in males 0.298 to 0.333, in females 0.288 to 0.335.


1963








BULLETIN OF THE FLORIDA STATE MUSEUM


In defining T. s. septentrionalis and in attempting to delimit its
range, I have had access to living ribbonsnakes from Monroe and
Tompkins Counties, New York; Trumbull County, Ohio; Hillsdale and
Washtenaw Counties, Michigan; and Marshall County, Indiana, as
well as to colored slides of two live individuals from Honey Harbour,
Ontario, and color notes on a series of specimens from Rondeau
Provincial Park, Ontario. Each of these animals has the dorsum black
or dark brown on at least the anterior half of the body and usually
more.
Locating the zone of intergradation between T. s. septentrionalis
and sauritus by examination of preserved specimens is hampered by
the fading of what was a black dorsum in life to a brown one after
several years in preservative, particularly if the animal was approach-
ing ecdysis at the time of preservation. Nevertheless, a cautious evalu-
ation of color in preserved specimens coupled with relative tail length
values and the color data from live material permits an estimate of the
subspecies' southern range limits. A more accurate delimitation awaits
color notes on more live specimens from critical points.
The northern Indiana population clearly belongs to T. s. septen-
trionalis, the central Indiana snakes appear close to it (though they
may possibly be intergrades), and the series from southwestern In-
diana-southeastern Illinois represents either sauritus X septentrionalis
intergrades or sauritus. Two adult males (UMMZ 107916, 108092)
from the Muscatatuck River in southeastern Indiana near the glacial
boundary appear to be T. s. sauritus. The zone of intergradation proba-
bly extends through central Ohio. The color pattern of a live adult
female from Licking County in south-central Ohio is typical of T. s.
sauritus, but its tail is relatively short. OSM 758 from Jackson County
in the southern part of the state is definitely sauritus. Northeastern
Ohio ribbonsnakes are members of the northern race, and so pre-
sumably are the populations in adjacent Pennsylvania, although their
exact status is not clear. Ribbonsnakes from the vicinity of Pittsburgh
probably are intergrades. All the West Virginia animals examined
appear to be typical sauritus, with the possible exception of CM 28127
from Preston County, which may have had a black dorsum in life,
although it agrees with the nominate form in all other characters.
The eastern New York and southern and central New England
ribbonsnakes all appear to be T. s. sauritus, but the southern New
Hampshire sample may be representatives of an intergrade popula-
tion. I have included the small series from Maine in T. s. septentrionalis
because these animals appear to have had black dorsa in life, and at
least one of them (MCZ 699) shows anterior widening of the lateral


160


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


stripe. This population seems to be disjunct from the main body of
the range of septentrionalis, but the literature records for Essex and
Franklin Counties, New York (Weber, 1928, and Evermann, 1918,
respectively), probably are based on individuals of this race, and
future collecting in central New Hampshire and central Vermont may
well close the gap. Bleakney (1951) noted that the T. sauritus he
collected in Nova Scotia were black with yellow stripes. The only
specimen of that series with a complete tail (NSMS unnumbered)
has the low tail length/total length value characteristic of T. s. septen-
trionalis. This population is unquestionably a relict; Bleakney (1958)
has discussed the probable cause of the disjunction.
Dickinson (1950) reported that H. K. Suzuki observed a ribbon-
snake at Bass Lake, Marinette County, Wisconsin, but that the speci-
men was not preserved. Nevertheless, a specimen in the Milwaukee
Public Museum (MPM 1829) is listed as collected by Suzuki in
Marinette County at Bass Lake. It agrees fairly closely with Michigan
T. s. septentrionalis in details of scutellation, and dorsal scale row 2
is clearly involved anteriorly in the lateral stripe. Hence an isolated
colony of septentrionalis quite possibly exists in northeastern Wisconsin
and the upper peninsula of Michigan.
The black dorsum of septentrionalis and the frequent narrowing
and darkening of its vertebral stripe probably are adaptations for
absorption of solar radiation paralleling those in the northern race of
T. proximus. As the brown young are almost identical in color to the
young of T. s. sauritus, it seems fairly certain that septentrionalis
represents a northern, more cold-tolerant derivative of the nominate
race.
The subspecific name septentrionalis (Latin, of the north) refers
to the northern distribution of this race in relation to the range of
the species.

Thamnophis sauritus sackenii (Kennicott)
Thamnophis saurita var. sackenii: Dumeril and Bocourt, 1893:759
Thamnophis sauritus sackeni: Ruthven, 1908:107
Thamnophis sauritus sackenii: Blanchard, 1925:18
Thamnophis sirtalis sackenii: Klauber, 1948:9
HOLOTYPE. The holotype was collected in "Florida" by Baron Osten
Sacken and was originally deposited in the collection of Northwestern
University, but Orlando Park informs me it is no longer there. As
letters to a number of museums have failed to reveal it, we may pre-
sume it lost. A neotype has not been designated because the original
description (Kennicott, 1859) and the type locality serve adequately


1963








162 BULLETIN OF THE FLORIDA STATE MUSEUM Vol. 7

to associate the name with a specific population. Schmidt (1953)
restricted the type locality to Gainesville, Alachua County, Florida.
DEFINITION. A subspecies of Thamnophis sauritus characterized by 8
supralabials, a tan to brown dorsum, a tan vertebral stripe often with
a metallic luster, a narrow lateral stripe, and a relatively long tail.
RANGE. Throughout peninsular Florida (exclusive of western Levy
County and the coastal area to the northwest of that county) west
to the vicinity of Tallahassee, and north through the Okefenokee
Swamp and coastal Georgia to Jasper County, South Carolina. Ap-
parently a small colony is isolated on St. Vincents Island just west of
the mouth of the Apalachicola River. Contrary to virtually every re-
cent statement of the western range limits of this subspecies, T. s.
sackenii does not occur in Mississippi, Alabama, or in Florida west of
the Apalachicola River.
DESCRIPTION. In the absence of the holotype, the following description
is based on UF 12145.2, an adult female from 2.5 miles E Moss Bluff,
Marion County, Florida, collected by W. T. Neill and S. R. Telford,
Jr. Scutellation: Oculars 1 + 3, temporals 1 + 2, supralabials 8, infra-
labials 10, intergenials 2, dorsal scale rows 19-19-17, ventrals 151, sub-
caudals 114. Body length is 458 mm., tail length 223 mm., tail length/
total length ratio 0.327.
The dorsum is brown. The vertebral stripe is metallic tan, the lateral
stripe pale yellow and narrow, the stripe occupying slightly less than
3 the width of a scale on rows 3 and 4. The venter is pale greenish
yellow, pale green on the tail; the dark ventrolateral stripe is broad,
occupying slightly more than 0.25 the width of a ventral scute on each
side. The chin shields and lower part of the infralabials are white,
the supralabials and upper part of the infralabials pale yellow. Parietal
spots are lacking.
DIscussION. Counts and measurements of this subspecies may be sum-
marized as: postoculars 2 to 5, temporals 1 + 2 to 2 + 3, supralabials
7 to 9, infralabials 8 to 12, intergenials 0 to 3, dorsal scale rows 19-17-17
to 19-19-17, male ventrals 150-171%, female ventrals 146-166, male
subcaudals 118-134, female subcaudals 105-131. Body length in males
ranges from 144 to 537 mm., in females 137 to 670 mm. Total length
in males 215 to 820 mm., in females 207 to 1018 mm. Adult tail length/
total length ratio in males 0.328 to 0.382, in females 0.317 to 0.364.
I have seen living examples of this subspecies from Alachua,
Brevard, Broward, Collier, Dade, Duval, Hendry, Lake, Levy, Mana-
tee, Marion, Martin, and Palm Beach Counties, Florida, and from


___


__________ _








ROSSMAN: GENUS THAMNOPHIS


Jasper County, South Carolina, as well as a colored slide of one from
Charlton County, Georgia. The South Carolina specimen, from the
Savannah National Wildlife Refuge, is typical of T. s. sackenii in
every respect. A preserved specimen (UF 12186) from the same
locality also appears assignable to the southern race. A living specimen
(unfortunately no longer available) from Ridgeland, some 20 miles
farther north in Jasper County, appeared to be either a sackenii or an
intergrade. As individuals from less than 50 miles up the coast in
Charleston County are characteristic of the nominate subspecies, in-
tensive collecting in the intervening area should reveal intergrades.
Inland, a series from the Okefenokee Swamp I consider unques-
tionably sackenii despite a high incidence of 7 supralabials. Neill
(1954) expressed the opinion that animals from as far inland as Eman-
uel County, Georgia, and Barnwell County, South Carolina, are closer
to sackenii than to T. s. sauritus, and that sauritus does not occur below
the Fall Line in eastern Georgia. I have not seen specimens from
Emanuel County, but I find the sample from just below the Fall Line
in eastern Georgia and adjacent South Carolina to be fairly typical of
sauritus.
Two adult females (UMMZ 106234-235) from Lake Miccosukee,
Jefferson County, Florida, are T. s. sackenii, although one of them has
7 supralabials on both sides of the head. They represent the west-
ernmost locality for this race, except for UF 7385, an adult female
from St. Vincents Island, Franklin County, Florida. This specimen
definitely appears to be sackenii, but all the specimens collected
on the adjacent mainland are good examples of sauritus. Only further
collecting can resolve the situation. All specimens available from
southwestern Georgia appear to be sauritus. Intergrades between
sackenii and T. s. nitae are discussed under nitae.
Cliburn (1961) recently demonstrated that T. s. sackenii is not
found in southern Mississippi as had been generally accepted, but
he stated that the genetic influence of sackenii was apparent in the
slightly higher tail length/total length values in extreme southeastern
Mississippi. This implies that sackenii has a relatively longer tail than
T. s. sauritus, a contention not supported by my data. The south-
eastern Mississippi material I have examined is typical sauritus.

Thamnophis sauritus nitae, new subspecies
HOLOTYPE. UF 12150.2, an adult female from McDonald Slough, St.
Marks National Wildlife Refuge, Wakulla County, Florida, collected
16 May 1959 by N. J. and D. A. Rossman.








BULLETIN OF THE FLORIDA STATE MUSEUM


DEFINITION. A subspecies of Thamnophis sauritus characterized by 8
supralabials, a dark brown or black dorsum on at least the anterior
part of the body, the vertebral stripe obscure or absent, a bluish-white
or light blue lateral stripe, and a relatively long tail.
RANGE. The coastal lowlands of northwestern peninsular Florida from
the Withlacoochee River on the south to central Wakulla County on
the west.
DESCRIPTION OF HOLOTYPE. Scutellation: Oculars 1 + 3, temporals
1 + 2, supalabials 8, infralabials 10, intergenials 3, dorsal scale rows
19-19-17, ventrals 160, subcaudal series and tail incomplete. Body
length is 621 mm.
The dorsum is black for a short distance behind the head, becoming
dark brown posteriorly (Maerz and Paul 15L12 at midbody) with
much black pigment overlaying the brown. The vertebral stripe is
lacking; the narrow lateral stripe is light blue (34G1). The venter is
bluish white anteriorly, becoming pale tannish yellow thereafter (the
posterior margin of each scute is light bluish green); the dark ventro-
lateral stripe is broad, occupying 0.20 the width of a ventral on each
side. The chin shields are white, and supralabials light blue at least
posteriorly. The head is black, hence no postorbital vitta is visible;
the parietal spots are lacking.
DISCUSSION. Counts and measurements of this subspecies are: postocu-
lars 3 to 4, temporals 1 + 2 to 2 + 2, supralabials 7 to 8, infralabials 9
to 11, intergenials 1 to 3, dorsal scale rows 17-19-17 to 19-19-17, male
ventrals 1612-169, female ventrals 156-166; male subcaudals 119-131,
female subcaudals 105-130. Body length in males ranges from 150 to
534 mm., in females 146.5 to 621 mm. Total length in males 220 to
756 mm., in females 216 to 701 mm. Adult tail length/total length
ratio in males 0.335 to 0.358, in females 0.331 to 0.372.
I have examined living examples of this race from Jefferson, Levy,
Taylor, and Wakulla Counties, and a freshly killed specimen from
Dixie County, all in Florida. All those from localities northwest of the
Suwannee River are fairly typical of this subspecies as defined above.
The situation in Levy County is confusing: From about the vicinity
of Bronson eastward, only T. s. sackenii occurs. Specimens fairly
characteristic of nitae have been collected at or near Vista, Sumner,
Otter Creek, Gulf Hammock, Lebanon Station, and Inglis; I have
seen individuals intermediate in color pattern between T. s. nitae and
sackenii from the vicinity of Lukens, Sumner, and Yankeetown. One
subadult (UF 12141) from 4 miles WSW Sumner is assignable to


164


Vol. 7








ROSSMAN: GENUS THAMNOPHIS


sackenii on its color pattern. Those ribbonsnakes from western Levy
County have a higher mean number of ventrals than does a series
from eastern Levy and adjacent Alachua Counties. This mean, how-
ever, does not differ from those of coastal lowland populations of
T. s. sackenii which lie to the south. Insufficient fresh material is avail-
able to determine adequately, or perhaps even accurately, the status
of the western Levy County snakes, but pending the acquisition of
large series of live specimens from throughout this area, I am inclined
to assign this population to T. s. nitae, as the majority of the speci-
mens examined appear to be much closer to this form than to sackenii.
The possibility that it is an intergrade population cannot be ruled
out and should not be overlooked; if such is the case, it shows a
heavier genetic influence from nitae. Several individuals from the
vicinity of Lake Tsala Apopka in Citrus County are considered to be
sackenii x nitae intergrades.
I have seen a few unusually dark T. s. sackenii from the Everglades
of southern Florida, but such individuals appear only rarely in the
populations of that region and lack a continuous distribution.
If the present coloration of CM 12419, an adult female, at all closely
approximates its colors in life, the animal is probably a T. s. sackenii
X nitae intergrade. The collecting data, however, show the snake was
taken 40 miles W Tallahassee in Calhoun County, Florida. The mileage
places the locality at Telogia Creek, which is in Liberty County, not
Calhoun. If these data are accurate, then sackenii, or a derivative,
penetrates the range of typical T. s. sauritus (which occurs 3 miles
WNW Hosford, a locality less than 5 miles east of Telogia Creek)
with no indication of intergradation. The specimen in question has 8
supralabials, 161 ventrals, and a tan vertebral stripe; the Hosford
specimen (UF 9398, a male) has 7 supralabials, 148 ventrals, and
agrees equally well with typical sauritus in all other respects. I be-
lieve the locality data are in error.
Within the range of Thamnophis sauritus nitae, the common gar-
tersnake, T. sirtalis, exhibits a remarkable similarity to the ribbon-
snake in color pattern (Neill, 1958). The adaptive significance of the
dark dorsum and blue lateral stripe present in these gartersnakes is
not readily apparent, if indeed there be any. To human eyes at least,
T. s. nitae is much more conspicuous in each of the several habitats
it occupies than would be the comparatively drab T. s. sackenii.
The subspecific name nitae is in honor of my wife Nita, who col-
lected the first living example I saw of this strikingly beautiful race.


1963








166 BULLETIN OF THE FLORIDA STATE MUSEUM Vol. 7

NATURAL HISTORY
This summary of our knowledge of the ecology of T. sauritus is
intentionally brief. For a detailed account of this species in the north-
ern part of its range see Carpenter (1952a, 1952b, and 1953).
HABITAT. The eastern ribbonsnake occupies a great variety of habitats;
the only basic requirement seems to be accessibility to permanent or
semipermanent bodies of water. Its habitats include such diverse situ-
ations as sandy beach, salt marsh, roadside ditch, pine flatwoods,
grassy stream bank, marshy meadow, rocky woodlands, sphagnum
bog, and tamarack swamp. I have collected T. sauritus in most of
these habitats; other records are from the literature and from field
notes of colleagues.
Many authors have commented on the semiarboreal tendencies of
this species (the generic name Thamnophis, which originally applied
only to the species sauritus, means "bush snake"), which, with its
relatively longer tail, is perhaps better adapted for climbing than is
T. proximus. Carpenter (1952a) reported that 61 percent of the
T. sauritus collected on his study area in southern Michigan during
the summer were up in bushes. The night of 15 August 1961 I saw
more than 30 T. sauritus sackenii along a sandy road running through
a marshy "prairie" in north-central Florida; all were 2 to 4 feet above
the ground in the vegetation bordering the road, not an unusual
phenomenon in my experience.
FooD. The diet of the eastern ribbonsnake consists chiefly of fishes
and adult and larval amphibians. Table 13 lists species that Thamno-
phis sauritus definitely is known to eat in the wild. Several authors
(Surface, 1906; Carpenter, 1952a; Hamilton and Pollack, 1956) have
found invertebrates in the stomach contents of ribbonsnakes, and
Fowler (1907) termed sauritus as largely insectivorous. On the basis
of present knowledge, Fowler's statement does not seem justified, for
the few instances in which invertebrates actually have been found
in sauritus are probably cases of secondary or accidental ingestion.
The eastern ribbonsnake does not seem any more prone than its
western counterpart to eat invertebrates, lizards, or endothermic verte-
brates voluntarily.
Nocturnal prowling and feeding is extremely common in the spe-
cies, especially in the southern part of the range and during the warm
months.
REPRODUCTION. Carpenter (1952b) found that female ribbonsnakes
in southern Michigan attain sexual maturity at between 2 and 3 years,
when they attain a minimum body length of 421 mm.; he did not









ROSSMAN: GENUS THAMNOPHIS


TABLE 13. Food of Thamnophis sauritus


Species


Source


Anurans
Bufo quercicus
Bufo terrestris
Acris gryllus
Hyla cinerea
Hyla crucifer
Hyla ocularis
Hyla squirella
Hyla versicolor
Pseudacris nigrita
Pseudacris triseriata
Rana areolata (larvae only)'
Rana clamitans
Rana palustris
Rana pipiens
Rana sylvatica

Salamanders
Diemictylus viridescens
Ambystoma maculatum
Desmognathus ochrophaeus

Fishes
Umbra limi
Gambusia ainis
Heterandria formosa


Duellman and Schwartz, 1958
Duellman and Schwartz, 1958
personal observation
Duellman and Schwartz, 1958
Carpenter, 1952a
Carr, 1940
Duellman and Schwartz, 1958
Carpenter, 1952a
Duellman and Schwartz, 1958
Carpenter, 1952a
personal observation
Carpenter, 1952a
Carpenter, 1952a
Carpenter, 1952a
Carpenter, 1952a; Conant, 1951


Carpenter, 1952a
Carpenter, 1952a
Conant, 1951


Carpenter, 1952a
personal observation
personal observation


1 Identified by Sam R. Telford, Jr.

present corresponding data for males. McCauley (1945) reported
a gravid female from Maryland 341 mm. in body length. The seven
smallest gravid female T. sauritus I have measured had body lengths
of 332 (Pennsylvania), 346 (Georgia), 350 (West Virginia), 365
(Nova Scotia), 371 (Georgia), 374 (West Virginia), and 410 mm.
(Ohio) respectively. Thamnophis sauritus appears to mature at a
smaller body size than does T. proximus.
Mating apparently occurs in April in the northern part of the
range where spring emergence takes place in the latter part of March;
in the extreme southern portion of the range where no true winter
hibernation takes place, mating probably occurs earlier. The young
are usually born in August in the north, July in the south, with con-
siderable overlap. Dates range from 2 July through 4 October (per-
sonal observations). The latter date, surprisingly enough, is for a


1963


167







BULLETIN OF THE FLORIDA STATE MUSEUM


female from northern Florida, and it might possibly have been her
second brood of the year. Tinkle (1957) has indicated the possibility
of a second brood in a female T. proximus from southeastern Louisi-
ana, and Telford (1952) reported a female T. sauritus sackenii that
gave birth to 12 normal young on 9 July and bore four more on 16
September which were malformed and born dead. Several female
sackenii I collected in north-central Florida 15 August 1961 contained
early embryos that almost certainly would not have been born until
late September, and possibly represented second broods. This ques-
tion certainly deserves more thorough investigation.
Brood size ranges from 3 (Ditmars, 1896) to 26 (Telford, 1952),
the mean for 69 broods being 11 individuals. Geographic variation
in brood size is pronounced. The mean number of individuals is 15.8
in 12 broods from Florida, 12.9 in 9 broods from southern Ontario,
10.7 in 7 broods from Maryland (McCauley, 1945), 10.0 in 5 broods
from southern Michigan (Carpenter, 1952a), and 6.0 in 5 broods from
northern Michigan (Burt, 1928). As the size of the brood tends to
increase with the size of the female, these means may not be strictly
comparable. Available data on body length of newly born T. sauritus
are summarized in table 14. Males average markedly larger than
females in the southern part of the range, slightly, if at all larger, in
the North. Individuals of T. sauritus tend to average smaller at birth
than T. proximus.
Burt (1928) observed twice as many embryos in the right oviduct
than in the left one in 5 female T. sauritus she examined, and Tinkle
(1957) found a similar ratio in 37 female T. proximus.
PARASITES. Langmann (1899) reported an eastern ribbonsnake in-
fected with haemosporidia.

CONCLUSIONS
Thamnophis sauritus is closely related to T. proximus and probably
evolved from proximus or its prototype. Certainly sauritus, with its
longer tail and reduced number of supralabials, appears the more
specialized form. Speciation may have occurred, as in so many other
species pairs (Blair, 1958), during division in Pleistocene time into
eastern and western populations confined respectively to Floridian
and Mexican refugia. In any case, to judge by present distribution of
the two species, the Mississippi River, when it was wider than at pres-
ent, apparently played an important role in the speciation process,
possibly by delaying contact between the two populations as they
expanded their ranges during glacial retreat. When the two popula-


Vol. 7









ROSSMAN: GENUS THAMNOPHIS


TABLE 14. Body length of newly born Thamnophis sauritus

Locality N M R


Alachua County, Florida ..............

Broward County, Florida .............

Collier County, Florida ...............

Wakulla County, Florida. .............

Sussex County, New Jersey............

Trumbull County, Ohio ...............

Marshall County, Indiana .............

Washtenaw County, Michigan.........

Rondeau Provincial Park, Ontario*.....


3cd
699
60cN
79 9
ice
59 9
14Mc'c
119 9
5ce
499
13cd'c
799
5ee
599
2c0N
499
12
11
9
21
9
13
9
11
15


170.7mm
155.7
168.0
160.4
181.0
172.6
158.9
153.3
135.0
140.6
158.4
158.0
141.0
140.4
149.8
141.8
159.2
164.0
149.8
149.3
141.1
154.4
165.3
152.2
163.7


144 -193 mm
137 -174
159.5-176
157 -164.5

170.5-175
150 -184.5
146.5-157.5
124.5-147.5
134.5-144.5
148 -169
150.5-165
125.5-146
131.5-149.5
142 -157.5
138 -145
150 -166
153 -170
144 -159
142 -156
130 -149
142 -163
160 -171
140 -160
149 -172


N = number of individuals; M = mean; R = range
Personal communication from B. McBride, all others from personal observation.

tions finally did meet, genetic isolation was either complete or nearly
so. The great similarity between the two species in diet and habitat
preference caused sufficient competition to prevent extensive sym-
patry. The overlap in ranges in Indiana may be the result of an east-
ward movement of proximus in the Prairie Peninsula during 'the
postglacial Xerothermic period (Smith, 1957); the populations in
Indiana, and perhaps those in eastern Illinois and southeastern Wis-
consin, probably represent relicts of that earlier, more eastward,
distribution.
Except for its long tail and small, separate, parietal spots, T. sauri-
tus sackenii bears a striking resemblance to T. proximus rutiloris of
Gulf and Caribbean Middle America. The similarity is so great that
some taxonomic confusion has revolved about a specimen of sackenii


1963








BULLETIN OF THE FLORIDA STATE MUSEUM


(ANSP 5826) which bears incorrect locality data (Rossman, 1961).
Two alternative hypotheses present themselves: either sackenii and
rutiloris evolved independently and their similarities are due to
convergence, possibly in response to similar environmental condi-
tions (a supposition which disregards the existence of the montane
populations of rutiloris), or sackenii developed from rutiloris prior
to the partition of the ribbonsnake into two species (sackenii in this
case is considered to resemble most closely the prototype of the species
sauritus, and rutiloris the prototype of T. proximus). The constancy
with which the presumably ancestral supralabial number (8) is main-
tained in sackenii seems to indicate that it is the most primitive race
of T. sauritus. If sackenii developed in this manner it explains how,
other than through convergent evolution, the species sauritus acquired
the dark ventrolateral stripe present in Mexican and Central American
T. proximus, but lacking in those races now in contact with T. sauritus.
The ranges of sackenii and rutiloris are now separated by over 1000
miles and by the ranges of two other subspecies, T. s. sauritus and
T. p. orarius, but roughly comparable disjunctions occur between the
Mexican and eastern United States subspecies of Drymarchon corais
and Storeria occipitomaculata. However, only in the latter species
has an additional subspecies been interposed between the pair which
became separated (assuming that, as certain elements of color pattern
seem to indicate, S. o. obscura of Florida is more closely related to
S. o. hidalgoensis of eastern Mexico than either are to the nominate
race). The seemingly close relationship between Storeria dekayi victa
of Florida and Storeria tropical (Anderson, 1961) presents an even
closer parallel to the sackenii-rutiloris situation. This existence of close
affinities between Floridian and Mexican forms seems to occur fre-
quently among vertebrates (Neill, 1957; Neill and Allen, 1959).
The members of the Sauritus group are, ecologically speaking,
both semiarboreal and semiaquatic. They have no peers within the
genus in the former category, and in the latter must surely be con-
sidered the ecological equivalents in eastern Mexico and eastern
United States of Thamnophis angustirostris, T. eques, and the aquatic
group of T. elegans of western North America. Several studies (Car-
penter, 1952a; Fouquette, 1954) have compared the diet and habitat
preference of the ribbonsnakes with those of other species of Thamno-
phis with which they are sympatric, and have shown that of the
sympatric species the common gartersnake, T. sirtalis, is most like the
ribbonsnakes in habitat preference, although not in diet. These studies
indicate that the ecological niches occupied by these sympatric species
are different enough to prevent serious interspecific competition.


Vol. 7


170








ROSSMAN: GENUS THAMNOPHIS


LITERATURE CITED

Anderson, Paul K.
1961. Variation in populations of brown snakes, genus Storeria, bordering the
Gulf of Mexico. Amer. Midland Nat., vol. 66, pp. 235-249.
Auffenberg, Walter
1955. A reconsideration of the racer, Coluber constrictor, in eastern United
States. Tulane Studies Zool., vol. 2, pp. 89-155.
Baird, Spencer F.
1859. Reptiles of the boundary. United States and Mexican Boundary Survey.
Washington, pp. 1-35.
Baird, Spencer F., and Charles Girard
1853. Catalogue of North American reptiles in the Museum of the Smith-
sonian Institution, part I. Serpents. Smithsonian Misc. Coll., vol. 2,
no. 5, pp. 1-172.
Barton, A. J.
1956. A statistical study of Thamnophis brachystoma (Cope) with comments
on the kinship of T. butler (Cope). Proc. Biol. Soc. Washington, vol.
69, pp. 71-82.
Blair, W. Frank
1958. Distributional patterns of vertebrates in the southern United States in
relation to past and present environments, pp. 433-468. In Zoogeog-
raphy. Amer. Assoc. Advance. Sci., Publication 51.
Blanchard, Frank N.
1925. A key to the snakes of the United States, Canada, and Lower Cali-
fornia. Papers Michigan Acad. Sci., Arts, Letters, vol. 4, no. 2, pp. 1-65.
Bleakney, J. Sherman
1951. A new snake record for Nova Scotia, Thamnophis sauritus sauritus
Linnaeus. Canadian Field Nat., vol. 65, pp. 118-119.
1958. A zoogeographical study of the amphibians and reptiles of eastern
Canada. Natl. Mus. Canada Bull. 155, pp. 1-119.
Boie, Friedrich
1827. Bemerkungen iiber Merrem's Versuch eines Systems der Amphibien.
Isis, vol. 20, pp. 508-566.
Boyer, Dorothy A., and Albert A. Heinze
1934. An annotated list of the amphibians and reptiles of Jefferson County,
Missouri. Trans. Acad. Sci. St. Louis, vol. 28, pp. 185-201.
Brown, Arthur E.
1901. A review of the genera and species of American snakes, north of
Mexico. Proc. Acad. Nat. Sci. Philadelphia, vol. 53, pp. 10-110.
Burt, M. D.
1928. The relation of size to maturity in the garter snakes, T. s. sirtalis (L.)
and T. s. sauritus (L.). Copeia, no. 166, pp. 8-12.


1963









172 BULLETIN OF THE FLORIDA STATE MUSEUM Vol. 7

Carpenter, Charles C.
1952a. Comparative ecology of the common garter snake (Thamnophis s.
sirtalis), the ribbon snake (Thamnophis s. sauritus), and Butler's garter
snake (Thamnophis butler) in mixed populations. Ecol. Monogr.,
vol. 22, pp. 235-258.
1952b. Growth and maturity of the three species of Thamnophis in Michigan.
Copeia, 1952, pp. 237-243.
1953. A study of hibernacula and hibernating associations of snakes and
amphibians in Michigan. Ecology, vol. 34, pp. 74-80.
1958. Reproduction, young, eggs and food of Oklahoma snakes. Herpetologica,
vol. 14, pp. 113-115.

Carr, Archie F., Jr.
1940. A contribution to the herpetology of Florida. Univ. Florida Publ., Biol.
Sci. Ser., vol. 3, no. 1, pp. 1-118.

Clark, R. F.
1949. Snakes of the hill parishes of Louisiana. Jour. Tennessee Acad. Sci.,
vol. 24, pp. 244-261.

Cliburn, J. William
1961. The ribbon snakes of southern Mississippi. Herpetologica, vol. 17, pp.
211-212.

Cochran, Doris M.
1961. Type specimens of reptiles and amphibians in the United States Na-
tional Museum. U.S. Natl. Mus. Bull. 220, pp. 1-291.

Conant, Roger
1951. The reptiles of Ohio. 2nd edition. Amer. Midland Nat., vol. 20, pp.
1-284.

Cooper, J. G.
1860. Reports of explorations and surveys to ascertain the most practicable
and economical route for a railroad from the Mississippi River to the
Pacific Ocean. U.S. Senate Rept., vol. 12, no. 4, pp. 292-306.

Cope, Edward D.
1860. Catalogue of the Colubridae in the Museum of the Academy of Natural
Sciences of Philadelphia, part 3. Proc. Acad. Nat. Sci. Philadelphia,
vol. 12, pp. 553-566.
1885. A contribution to the herpetology of Mexico. Proc. Amer. Philos. Soc.,
vol. 22, pp. 379-404.

Darlington, Philip J., Jr.
1957. Zoogeography: the geographical distribution of animals. John Wiley &
Sons, Inc., New York, pp. 1-675.

Davis, N. S., Jr., and Frank L. Rice
1883. North American Batrachia and Reptilia, found east of Mississippi River.
Bull. Illinois State Lab. Nat. Hist., no. 5, pp. 1-66.

Deckert, Richard F.
1918. A list of reptiles from Jacksonville, Florida. Copeia, no. 54, pp. 30-33









1963 ROSSMAN: GENUS THAMNOPHIS 173

Dickinson, William E.
1950. Recent additions to the records of the distribution of the reptiles in
Wisconsin. Trans. Wisconsin Acad. Sci., Arts, Letters, vol. 40, pp.
71-77.

Ditmars, Raymond L.
1896. The snakes found within fifty miles of New York City. Proc. Linnaean
Soc. New York, no. 8, pp. 9-24.

Dowling, Herndon G.
1951. A proposed method of expressing scale reductions in snakes. Copeia,
1951, pp. 131-134.
1960. A taxonomic study of the ratsnakes, genus Elaphe Fitzinger, VII. The
Triaspis Section. Zoologica, vol. 45, pp. 53-80.

Duellman, William E.
1958. A monographic study of the colubrid snake genus Leptodeira. Bull.
Amer. Mus. Nat. Hist., vol. 114, pp. 1-152.

Duellman, William E., and Albert Schwartz
1958. Amphibians and reptiles of southern Florida. Bull. Florida State Mus.,
vol. 3, pp. 181-324.

Dumeril, A.-M.-C., G. Bibron, and A. H. A. Dumeril
1854. Erpetologie g6nerale ou histoire naturelle des reptiles. Librairie Ency-
clopedique de Rout, Paris, vol. 7, pt. 1, pp. 1-780.

Dumeril, A., and M.-Firmin Bocourt
1893. Recherches zoologiques pour servir A l'histoire de la faune de l'Amerique.
Rtudes sur les reptiles. Mission scientifique au Mexique et dans
l'Amerique Centrale, pt. 3, sect. 1, pp. 1-1012.

Dunn, Emmet R.
1940. Notes on some American lizards and snakes in the Museum of Gote-
borg. Herpetologica, vol. 1, pp. 189-194.

Evermann, Barton W.
1918. Notes on some Adirondack reptiles and amphibians. Copeia, no. 56,
pp. 48-51.

Fitch, Henry S.
1940. A biogeographical study of the ordinoides artenkreis of garter snakes
(genus Thamnophis). Univ. California Publ. Zool., vol. 44, pp. 1-150.
1941. Geographic variation in garter snakes of the species Thamnophis sirtalis
in the Pacific coast region of North America. Amer. Midland Nat., vol.
26, pp. 570-592.

Fitch, Henry S., and T. Paul Maslin
1961. Occurrence of the garter snake, Thamnophis sirtalis, in the Great
Plains and Rocky Mountains. Univ. Kansas Publ. Mus. Nat. Hist., vol.
13, pp. 289-308.


________ ______ ___ _~__ ____









BULLETIN OF THE FLORIDA STATE MUSEUM


Fitzinger, Leopoldo
1843. System reptilium. Fasciculus primus, Amblygossae. Vienna, vol. 1,
pp. 1-106.

Fouquette, M. J., Jr.
1954. Food competition among four sympatric species of garter snakes, genus
Thamnophis. Texas Jour. Sci., vol. 6, pp. 172-188.

Fowler, Henry W.
1907. The amphibians and reptiles of New Jersey. Ann. Rept. New Jersey
State Mus. (1906), pt. 2, pp. 23-250, 351-365, 367-408.

Fox, Wade
1948. Effect of temperature on development of scutellation in the garter
snake, Thamnophis elegans atratus. Copeia, 1948, pp. 252-262.

Fox, Wade, Charles Gordon, and Marjorie H. Fox
1961. Morphological effects of low temperatures during the embryonic devel-
opment of the garter snake, Thamnophis elegans. Zoologica, vol. 46,
pp. 57-71.

Garman, Harrison
1892. A synopsis of the reptiles and amphibians of Illinois. Bull. Illinois State
Lab. Nat. Hist., vol. 3, pp. 215-388.

Garman, Samuel
1892. On Texan reptiles. Bull. Essex Inst., vol. 24, pp. 98-109.

Gloyd, Howard K.
1926. The amphibians and reptiles of Franklin County, Kansas. Trans. Kan-
sas Acad. Sci., vol. 31, pp. 115-141.
1935. Some aberrant color patterns in snakes. Papers Michigan Acad. Sci.,
Arts, Letters, vol. 20, pp. 661-668.

Hamilton, W. J., Jr., and Joseph A. Pollack
1956. The food of some colubrid snakes from Fort Benning, Georgia. Ecology,
vol. 37, pp. 519-526.

Harwood, Paul D.
1932. The helminths parasitic in the Amphibia and Reptilia of Houston,
Texas, and vicinity. Proc. U.S. Natl. Mus., vol. 81, no. 17, pp. 1-71.

Holbrook, John E.
1842. North American herpetology. Vol. IV, pp. 1-138. 2nd edition. Phila-
delphia.

Hudson, George E.
1942. The amphibians and reptiles of Nebraska. Nebraska Conserv. Bull.,
no. 24, pp. 1-146.

Jan, Georges
1863. Elenco sistematico degli ofidi descritti e desegnati per l'iconografia
general. Milan, pp. 1-143.


Vol. 7









ROSSMAN: GENUS THAMNOPHIS


Kennicott, Robert
1859. Notes on Coluber calligaster of Say, and a description of new species
of serpents in the collection of the North Western University of Evans-
ton, Ill. Proc. Acad. Nat. Sci. Philadelphia, vol. 11, pp. 98-100.

King, Wayne
1959. Vertebra duplication, an osteological anomaly widespread in snakes.
Herpetologica, vol. 15, pp. 87-88.

Klauber, Laurence M.
1943. Tail-length differences in snakes with notes on sexual dimorphism and
the coefficient of divergence. Bull. Zool. Soc. San Diego, no. 18, art. 1,
pp. 1-60.
1948. Some misapplications of the Linnaean names applied to American
snakes. Copeia, 1948, pp. 1-14.

Klein, Theodore A., Jr.
1949. A record litter of Thamnophis sirtalis proximus. Herpetologica, vol. 5,
p. 17.

Langmann, Gustav
1899. On haemosporidia in American reptiles and batrachians. New York
Med. Jour. for Jan. 7, 1899, pp. 1-17.

Laughlin, Harold E.
1959. Stomach contents of some aquatic snakes from Lake McAlester, Pitts-
burgh County, Oklahoma. Texas Jour. Sci., vol. 11, pp. 83-85.

Linnaeus, Carolus
1766. System naturae. Editio duodecima. Stockholm, pp. 1-532.

Loennberg, Einar
1894. Notes on reptiles and batrachians collected in Florida in 1892 and
1893. Proc. U.S. Natl. Mus., vol. 17, no. 1003, pp. 317-339.

McCauley, Robert H., Jr.
1945. The reptiles of Maryland and the District of Columbia. Privately pub-
lished, 194 pp.

Maerz, Aloys J., and M. Rea Paul
1950. A dictionary of color. 2nd edition. McGraw-Hill, New York, 208 pp.,
56 pl.
Malnate, Edmond V.
1960. Systematic division and evolution of the colubrid snake genus Natrix,
with comments on the subfamily Natricinae. Proc. Acad. Nat. Sci.
Philadelphia, vol. 112, pp. 41-71.
1961. Amphiesmoides, a new genus for Tropidonotus ornaticeps Werner
(Serpentes: Colubridae). Notulae Naturae, no. 341, pp. 1-7.

Merrem, Blasius
1820. Versuch eines Systems der Amphibien. Johann Christian Krieger, Mar-
burg, pp. 1-191.


1963









BULLETIN OF THE FLORIDA STATE MUSEUM


Milstead, William W.
1953. Geographic variation in the garter snake, Thamnophis cyrtopsis. Texas
Jour. Sci., vol. 5, pp. 348-379.

Mittleman, M. B.
1949. Geographic variation in Marcy's garter snake, Thamnophis marcianus
(Baird and Girard). Bull. Chicago Acad. Sci., vol. 8, pp. 235-249.

Myers, Charles W.
1959. Amphibians and reptiles of Montauk State Park and vicinity, Dent
County, Missouri. Kansas Acad. Sci., vol. 62, pp. 88-90.

Neill, Wilfred T.
1954. Ranges and taxonomic allocations of amphibians and reptiles in the
southeastern United States. Publ. Res. Div. Ross Allen's Reptile Inst.,
vol. 1, pp. 75-96.
1957. Historical biogeography of present-day Florida. Bull. Florida State
Mus., vol. 2, pp. 175-220.
1958. The occurrence of amphibians and reptiles in saltwater areas, and a
bibliography. Bull. Marine Sci. Gulf and Caribbean, vol. 8, pp. 1-97.

Neill, Wilfred T., and Ross Allen
1959. Studies on the amphibians and reptiles of British Honduras. Publ.
Res. Div. Ross Allen's Reptile Inst., vol. 2, pp. 1-76.

Peters, James A.
1960. The snakes of the subfamily Dipsadinae. Misc. Publ. Mus. Zool., Univ.
Michigan, no. 114, pp. 1-224.

Rossman, Douglas A.
1961. Nomenclatural status of the neotropical subspecies of the colubrid
snake, Thamnophis sauritus. Notulae Naturae, no. 340, pp. 1-2.
1962. Thamnophis proximus (Say), a valid species of garter snake. Copeia,
1962, pp. 741-748.

Rossman, Nita J.
1958. Birth, growth, and feeding records of a brood of western ribbon
snakes, Thamnophis sauritus proximus. Southern Illinois Herpetol. Soc.
Bull., vol. 1, no. 1, pp. 4-5.

Ruthven, Alexander G.
1908. Variation and genetic relationships of the garter-snakes. U.S. Natl.
Mus. Bull. 61, pp. 1-201.

Say, Thomas
1823. In Edwin James, Account of an expedition from Pittsburgh to the
Rocky Mountains, performed in the years 1819, 1820. Vol. 1, pp. 1-344.
Longman, Hurst, Rees, Ovme, and Brown, London.

Schmidt, Karl P.
1953. A check list of North American amphibians and reptiles. Sixth edition.
Amer. Soc. of Ichthyologists and Herpetologists, pp. 1-280.


176


Vol. 7









ROSSMAN: GENUS THAMNOPHIS


Schmidt, Karl P., and Roger Conant
1956-57. Appendix to Opinion 385. Request that the International Commis-
sion on Zoological Nomenclature should direct that the nominal species
"Coluber sirtalis" Linnaeus 1758, and "Coluber saurita" Linnaeus 1766,
should be interpreted by the neotypes here designated by the present
authors instead of, as hitherto proposed, by reference to specified pre-
viously published figures and descriptions. Pp. 224-230. Opinion 385.
Opinions and Declarations Rendered by the International Commission
on Zoological Nomenclature, vol. 12, pt. 6, edited by Francis Hemming,
London, pp. 191-230.

Smith, Albert G.
1949. The subspecies of the plains garter snake, Thamnophis radix. Bull.
Chicago Acad. Sci., vol. 8, no. 14, pp. 285-300.

Smith, Hobart M.
1938. Notes on reptiles and amphibians from Yucatan and Campeche, Mex-
ico. Occ. Papers Mus. Zool., Univ. Michigan, no. 388, pp. 1-22.

Smith, Hobart M., and Edward H. Taylor
1945. An annotated checklist and key to the snakes of Mexico. U.S. Natl.
Mus. Bull. 187, pp. 1-239.

Smith, Philip W.
1957. An analysis of post-Wisconsin biogeography of the prairie peninsula
region based on distributional phenomena among terrestrial vertebrate
populations. Ecology, vol. 38, pp. 205-218.

Stone, Witmer
1903. A collection of reptiles and batrachians from Arkansas, Indian Territory,
and western Texas. Proc. Acad. Nat. Sci. Philadelphia, vol. 55, pp.
538-542.
1906. Notes on reptiles and batrachians of Pennsylvania, New Jersey and
Delaware. Amer. Nat., vol. 40, no. 471, pp. 159-170.

Strecker, John K., Jr.
1908. The reptiles and batrachians of McLennan County, Texas. Proc. Biol.
Soc. Washington, vol. 2, pp. 69-84.
1909. Notes on the herpetology of Burnet County, Texas. Baylor Univ. Bull.,
vol. 12, no. 1, pp. 1-9.

Surface, H. A.
1906. The serpents of Pennsylvania. Bull. Pennsylvania State Dept. Agric.,
Div. Zool., no. 4, pp. 133-208.

Taylor, W. Edgar
1892. Catalogue of the snakes of Nebraska with notes on their habits and
distribution. Amer. Nat., vol. 26, Sept., pp. 742-752.

Telford, S. R., Jr.
1952. A herpetological survey in the vicinity of Lake Shipp, Polk Co., Flcrida.
Quart. Jour. Florida Acad. Sci., vol. 15, pp. 175-185.


1963


177










BULLETIN OF THE FLORIDA STATE MUSEUM


Tilden, J. W.
1961. Certain comments on the subspecies problem. Syst. Zool., vol. 10,
pp. 17-23.

Tinkle, Donald W.
1957. Ecology, maturation and reproduction of Thamnophis sauritus proximus.
Ecology, vol. 38, pp. 69-77.

Trowbridge, Albert H.
1937. Ecological observations on amphibians and reptiles collected in south-
eastern Oklahoma during the summer of 1934. Amer. Midland Nat.,
vol. 18, pp. 285-303.

Weber, Jay A.
1928. Herpetological observations in the Adirondack Mountains, New York.
Copeia, no. 169, pp. 106-112.

Wheeler, G. C.
1947. The amphibians and reptiles of North Dakota. Amer. Midland Nat.,
vol. 38, pp. 162-190.

Wright, Albert H., and Anna A. Wright
1957. Handbook of snakes of the United States and Canada. Vol. 2, pp.
565-1105. Comstock Publishing Associates, Ithaca, New York.


178


Vol. 7









Contributions to the BULLETIN OF THE FLORIDA STATE MUSEUM may be in any
field of biology. Manuscripts dealing with natural history or systematic problems
involving the southeastern United States or the Caribbean area are solicited
especially.
Manuscripts should be of medium length-50 to 200 pages. Examination for
suitability is made by an Editorial Board.
The BULLETIN is distributed worldwide through institutional subscriptions and
exchanges only. It is considered the responsibility of the author to distribute his
paper to all interested individuals. To aid in this, fifty copies are furnished the
author without cost.
PREPARATION OF MANUSCRIPT
Highly recommended as a guide is the volume:

Conference of Biological Editors, Committee on Form and Style.
1960. Style manual for biological journals.
Amer. Inst. Biol. Sci., Washington. 92 p.
Manuscripts should be typewritten with double spacing throughout, with ample
margins, and on only one side of the paper. The author should keep a copy; the
original and a carbon must be submitted. Tables and legends of figures should
be typed on sheets separate from the text. Several legends or tables may be
placed on a single sheet.
Illustrations, including maps and photographs, should be referred to as "figures."
All illustrations are reduced to a maximum of 4-1/4 by 7-1/8 inches. Size scales,
wherever they are necessary, should be incorporated into the figure.
References to literature should conform with the bibliographic style used in recent
numbers of the BULLETIN. Spell out in full the titles of non-English serials and
places of publication.
Footnote material should be kept to a minimum. However, provide copy for a
footnote detailing the title, affiliations, and address of the author (see recent num-
bers of the BULLETIN).
Manuscripts must be accompanied by a synopsis-a brief and factual summary
(not a mere description) of the contents and conclusions, which points out the
presence of any new information and indicates its relevance. In it list all new
organisms described and give their ranges; indicate all taxonomic changes pro-
posed. The synopsis, written in full sentences, should be concise, but completely
intelligible in itself without references to the paper, thereby enabling the busy
reader to decide more surely than he can from the title alone whether the paper
merits his reading. The synopsis will be published with the paper. It does not
replace the usual conclusions or summary sections. It may also serve as copy
for the abstracting services.
Manuscripts and all editorial matters should be addressed to:

Managing Editor of the BULLETIN
Florida State Museum
Seagle Building
Gainesville, Florida




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