T

RA
OF THL

UNIVERSITY
OF ILLINOIS

5SO.S

FI

Return this book on or before the
Latest Date stamped below. A
charge is made on all overdue
books.

University of Illinois Library

M32

1HE LIBRARY OF THE

FEB10194Q

UNIVERSITY OF ILLINOIS

y OF THE

I 1940 ZOOLOGICAL SERIES

)F ILLINOIS OF

FIELD MUSEUM OF NATURAL HISTORY

Volume 24 CHICAGO, MAY 31, 1940 No. 12

NOTES ON TEXAN SNAKES OF THE
GENUS SALVADORA

BY KARL P. SCHMIDT

CURATOR OF AMPHIBIANS AND REPTILES

In drawing up a list of the amphibians and reptiles of the Big
Bend region in Trans-Pecos Texas, based on collections made by
Tarleton F. Smith for the National Park Service and by a party
from Field Museum and the Chicago Academy of Sciences in 1937,
I encounter some difficulty with snakes of the genus Salvadora.
For the study of the problem opened up by the west Texan collections,
it has been necessary to borrow additional material. Our thanks
are due to authorities of the Museum of Comparative Zoology, the
United States National Museum, the Museum of Zoology of the
University of Michigan, and the Chicago Academy of Sciences for
loans of specimens.

The problem at issue, the identification of the three forms which
I discern in Texan collections, has been discussed with C. M. Bogert,
who has reviewed the genus Salvadora for the United States, and with
Hobart M. Smith, who has reviewed the Mexican forms. The diver-
gence of the views in the present paper from those of Dr. Bogert
apparently rests on the study of our new material from Trans-Pecos
Texas. Without undertaking a revision of the genus, it is desired
to formulate this divergence of opinion and bring the new arrange-
ment proposed to the attention of our herpetological colleagues,
especially as Dr. Smith may be expected to reopen the whole prob-
lem on the basis of his Mexican collections, made as Walter Rathbone
Baker Scholar for the Smithsonian Institution. Dr. Smith (in litt.)
agrees with me in rejecting Bogert's treatment of the Texan forms.

Stejneger stated the problem as to the allocation of the names
hexalepis Cope and grahamiae Baird and Girard in his paper on the
reptiles of the Huachuca Mountains (1902). Although the type
locality of Salvadora grahamiae is stated as "Sonora," and hence prob-
ably southern Arizona (from Bogert's analysis of Graham's route,
in litt.), Stejneger assigns the name also to the southeast Texan form
which is abundant at Brownsville. Specimens obtained by Tarleton

No. 470 143

144 FIELD MUSEUM OF NATURAL HISTORY — ZOOLOGY, VOL. 24

Smith and myself in the Chisos Mountains, from the higher forested
areas, are in excellent agreement with specimens from the Huachucas
and with the type of grahamiae. They prove to be readily distin-
guishable from the Brownsville form. A second species in west
Texas from the desert plateaus surrounding the isolated mountain
forests proves to be distinguishable from both the Chisos form and
that of southeastern Texas; and, though it is directly allied to the
Arizonan hexalepis, it clearly represents an undescribed form. The
name bairdii Jan and Sordelli (1860, pi. 3, fig. 2) appears to refer to
a form closely allied to the one in southeastern Texas described
below as Salvadora lineata. I follow the advice of Dr. Smith in
reserving the name bairdii, at least until the type can be re-examined,
for the Mexican form defined in his recent paper (1939, p. 232).

Salvadora grahamiae Baird and Girard.

Salvadora grahamiae Baird and Girard, Cat. N. Amer. Kept., pt. 1, p. 104, 1853

Diagnosis. — A Salvadora with edges of rostral only slightly
raised; posterior chin shields in contact or narrowly separated; eight
upper labials (occasionally nine); ventrals fewer than in deserticola
and hexalepis, and caudals more numerous; caudals more numerous
than in lineata, from which it is further distinguished by the usual
absence of the lateral narrow line and by the broader dorsal light
band; no supra-anal keels in adult males.

Remarks. — In the Big Bend region of western Texas this species
is represented by a form which appears to be confined to the more
humid and timbered upper slopes of the mountains. Our few speci-
mens are insufficient for a detailed comparison with the typical
grahamiae of the Huachuca Mountains. It seems clear that the
Chisos Mountain form is directly allied to that of southern Arizona;
and it is believed that grahamiae is essentially a relict species with
more or less completely isolated populations in the mountains of the
Southwest. It is evident that the study of the distribution and vari-
ation of these populations represents a most interesting herpeto-
logical problem, but one that can scarcely be dealt with until more
extensive collections are available.

I am indebted to Dr. Smith for the following manuscript notes on
the type of grahamiae, U. S. National Museum No. 2081 :

"A male specimen. Ventrals 180, anal divided, caudals 97, scale
rows 17-17-13; supralabials 9-9, infralabials 9-10; first supralabial
separated from posterior section of nasal; second supralabial in con-
tact with loreal on one side; preoculars 3-3; postoculars 2-2; loreal

1940 TEXAN SALVADORA— SCHMIDT 145

1-1; temporals 2+3, 2+2, posterior pair of chin shields slightly
shorter than anterior, separated medially by two scales along the
midgular line. Rostral well developed, with distinct, free lateral
edges. Median light stripe three scales wide on anterior two-thirds
of body, one and two half scale rows wide on posterior third. Adja-
cent dark band three scale rows wide on anterior two-thirds of body,
one and two half scale rows wide on posterior third. Each scale in
these dark bands has the base jet black; the antero-lateral edge of
the scales of the two upper rows in the dark bands is white. These
white spots are confined to the scales of the median row of the dark

FIG. 13. Pattern of Salvadora grahamiae Baird and Girard. Field Museum
F No. 27720, The Basin, Chisos Mountains, Brewster County, Texas.

stripe on the posterior fourth of the body. No other dark stripes.
Total length 690 mm., tail length 184 mm."

The four Huachucan specimens I have examined agree excellently
* with the type, except for the fact that the upper labials are 8-8 in
-; three and 8-9 in only one. All have dorsal scales 19-17-13, but the
drop to 17 takes place at about the twelfth ventral, so that the
anterior count is likely to be stated as 17. The ventrals in three
males are 180, 181, and 185; 186 in the single female. The corre-
sponding caudals are 107, 107, 97, and 92. Only one specimen
agrees with the type in number of preoculars, and this has two on
one side and three on the other. A narrow lateral dark line is faint
on the third scale row in Field Museum No. 2704, from "Tomb-
*- stone," collected by F. C. Willard, and hence only presumptively

1=

0

146 FIELD MUSEUM OF NATURAL HISTORY — ZOOLOGY, VOL. 24

from the Huachucas. The three remaining specimens agree exactly
in coloration with the type (see Baird, 1859, pi. 5, fig. 2).

Four specimens from the Chisos Mountains (two from the Mu-
seum of Zoology of the University of Michigan, one in Field Museum
and one in the United States National Museum [National Park
Service No. 138]) agree in having eight upper labials on each side, no
trace of a lateral line, and caudal count high, 101 to 112 (in both
sexes). The National Museum specimen, a female, has 197 ventrals,
while Field Museum No. 27720, also a female, from the Basin (in
the Chisos Mountains), has ventrals 190 and caudals 112; these two
specimens agree further in having three preoculars on each side.
Differences from the Huachuca series are slight. In the two males
and two females available, the ventrals are 179 and 186 in males, 190
and 197 in females; but the caudals are more numerous in females.
Should the Chisos and Huachuca populations prove distinguishable
as larger series become available, they would still be much more
closely allied inter se than with any other species of Salvadora.

A single male specimen from the Davis Mountains, Field Museum
No. 29497, collected by John M. Schmidt in 1938, has ventrals 183,
caudals 83, preoculars 3-3, upper labials 9-9, lower labials 11-11,
and a faint lateral line on the third scale row. It thus fits neither
hexalepis hexalepis, h. deserticola, grahamiae, nor lineata. It is best
left unidentified until a Davis Mountain series is available. It prob-
ably belongs with the mountain grahamiae series, as is suggested fur-
ther by the Tombstone specimen mentioned above and by one in the
Chicago Academy of Sciences from Silver City, New Mexico (No.
5247), which agrees with grahamiae in ventrals and caudals, but also
exhibits a lateral line on the third scale row.

Salvadora hexalepis deserticola subsp. nov.

Type from Government Spring, near Chisos Mountains, Brewster
County, Texas. No. 26615 Field Museum of Natural History.
Adult male. Collected 1935, by Tom Carney.

Diagnosis. — Allied to Salvadora hexalepis hexalepis in the separa-
tion of the posterior chin shields, enlargement of the rostral, nine
upper labials, in having keeled supra-anal scales, and in pale grayish
coloration; distinguished by the uniformly single loreal, lower ventral
count, and the situation of the narrow lateral line on the fourth scale
row instead of on the third and fourth.

Description of type. — Head a little wider than neck, notably
truncate anteriorly on account of the enlarged rostral with its
raised edges; body slender, tail alternate.

1940 TEXAN SALVADORA— SCHMIDT 147

Portion of rostral seen from above equal to its distance from the
frontal; internasals sub triangular, pref rentals nearly rectangular;
frontal nearly as long as the parietals, which are truncate posteriorly;
upper labials 9-10; lower labials 11-11; anterior chin shields longer
than posterior; posterior chin shields separated by two scales; nasal
divided; loreal single; oculars 2-2; temporals 2-3; ventrals 190,
anal divided, caudals 78; dorsal scales 17-17-13, smooth; faint keels
present on the supra-anal region.

FIG. 14. Pattern of Salvadora hexalepis deserticola subsp. nov. Field Museum
No. 26615, type.

Light ground color gray, involving the three median scale rows,
and parts of the adjacent rows, the three lateral scale rows and the
upper half of the fourth and all of the fifth on each side; a dark
dorsolateral band on each side on the sixth and seventh scale rows,
with a narrow lateral line on the lower half of the fourth scale row;
dorsolateral band extending forward to the postoculars; lateral line
merged with the dorsolateral band on the neck, on the third scale
row on the posterior third of the body.

Measurements of type. — Length 854, tail 189.

Notes on paratypes. — Other specimens in Field Museum collec-
tions that are referred to deserticola are No. 26620, male, Mesa de
Anguilla, Chisos Mountains, collected in 1937 by A. E. Borell; and
two females, Nos. 25210 and 27719, collected near the Chisos Moun-
tains by Tarleton F. Smith. Blanchard Collection No. 284 from
Hayes Ridge, Chisos Mountains (collected by F. N. Blanchard
and Howard K. Gloyd), and University of Michigan No. 66023 from

148 FIELD MUSEUM OF NATURAL HISTORY — ZOOLOGY, VOL. 24

Glenn Spring, also in the vicinity of the Chisos Mountains, are also
females. Ventrals and caudals in the male paratype number 195 and
82; in the female specimens they range from 187 to 190 and from 71
to 79. The loreal is invariably single, the preoculars invariably two;
upper labials 9-9 in all but one, which has 10-9; lower labials 10-10 in
three, 11-10 in one, 11-11 in one, and 12-12 in one; the postoculars
occasionally three; the temporals variable, one to three in the first
row and two or three in the second; the posterior chin shields are
invariably widely separated. All agree in having the dark lateral
line confined to the fourth scale row on the anterior two-thirds of
the body.

Remarks. — The subspecies deserticola may be associated with the
west Texan creosote bush desert. Its relation to the Chisos popula-
tion of grahamiae is evidently the same as that of hexalepis hexalepis to
the Huachucan grahamiae. It may be pointed out that the range of
h. hexalepis as thus restricted includes two very different vegetational
regions, the Sonoran desert of western Arizona to southwestern Utah
and Lower California, and the grassy region of southeastern Arizona.
Salvadora hexalepis virgultea occupies a more uniform faunal area.

Salvadora lineata sp. nov.

Type from Kingsville, Kleberg County, Texas. No. 28605 Field
Museum of Natural History. Adult female. Collected 1938, by
Professor J. C. Cross.

Diagnosis. — A Salvadora with rostral edges little raised, pos-
terior chin shields in contact or narrowly separated, upper labials
almost invariably 8-8, lower labials more frequently 9 than 10,
ventrals few, 179 to 192 in males, 189 to 194 in females, caudals 90
to 103 in males, 91 to 93 in females. Supra-anal scales not keeled.
In most of these characters lineata agrees with grahamiae; it is
sharply distinguished from that form by having a well-defined lateral
line, which is on the third scale row anteriorly (the second posteri-
orly). The anterior section of the nasal is usually in contact with the
second labial, and the dorsolateral dark stripe passes over the
temporal region to the eye.

Description of type. — Head a little wider than body, body slender,
tail attenuate. Snout rounded, edges of rostral slightly raised (worn
in type specimen).

Portion of rostral seen from above about equal to the internasal
suture, which is nearly as long as the preocular suture; frontal longer
than its distance from the end of the snout, about as long as the

1940

TEXAN SALVADORA— SCHMIDT

149

parietals, which are truncate posteriorly; upper labials 8-8; lower
labials 10-10; nasal divided; loreal single; oculars 2-2; temporals
2-2 and 2-3; ventrals 201, anal divided, caudals 93; dorsal scale
rows 19-17-13, smooth.

Dorsal stripe pale yellowish, less than three scale rows wide
except on the neck; adjacent dorsolateral dark brown band on the
fifth, sixth, seventh, and part of the eighth scale rows, very sharply
defined, with light bluish spots on the lower anterior borders of the
dark scales; dark band continued forward to the loreal region; sides
greenish gray; a narrow dark line on the middle of the third scale
row, beginning on the neck about the length of the head behind the

FIG. 15. Pattern of Salvadora lineata sp. nov. Field Museum No. 6903, Point
Isabel, near Brownsville, Texas.

rictus, continued to the anus; the lateral line is on the third scale
row on the posterior third of the body; top of head brown.

Measurements of type. — Length 753, tail 192.

Notes on paratypes. — Twelve male and seven female specimens
may be regarded as paratypic. Of these Field Museum Nos. 3395,
3395a-c, 5493, 6795-98, Chicago Academy of Sciences No. 5239,
and Museum of Comparative Zoology No. 13823 are from Browns-
ville, Texas; Field Museum No. 6903 is from Point Isabel, near
Brownsville; Field Museum No. 25742 is a specimen received through
the Lincoln Park Zoo, hence probably from W. A. King, of Browns-
ville; U. S. National Museum No. 26441 is from Kinney County;
Chicago Academy of Sciences No. 5240 is from San Antonio, Bexar
County, as is No. 67363 of the National Museum collection; No.

150 FIELD MUSEUM OF NATURAL HISTORY— ZOOLOGY, VOL. 24

55798, also in the National Museum, is from Waco, McLennan
County; and Museum of Comparative Zoology No. 43910 is from
Georgetown, Williamson County. A specimen from Ernest G. Marsh
(field No. 85) is from Musquiz, Coahuila. This series is remarkably
uniform in coloration, and the narrow lateral line is sharply defined
on the third scale row in all. The ventrals in twelve males range from
179^ to 192 (average 187), and in eight females (including the type)
from 187 to 201 (average 193). The caudals range from 90 to 103 in
nine males (average 93), and from 85 to 93 in six females (average 90).
The dorsal scale formula is 19-17-15 in six of the female paratypes,
while in nine out of twelve males the posterior count is 13; upper
labials 8-8 in all but one specimen, lower labials 9-9 in twelve, 10-10
in eight; preoculars 2-2 in thirteen, 1-2 in one, 2-3 in two, 3-3 in
four, and 4-3 in one; postoculars 2-3 in one, 3-3 in two, 2-2 in seven-
teen; temporals variable, 2-2-3 to 3-4-4. The supra-anals are not
keeled in adult males.

Remarks. — Salvadora lineata appears to be abundant in the lower
Rio Grande Valley, whence we have fourteen specimens. Musquiz,
in the basin of the Rio Alamos, is obviously a normal extension from
the Rio Grande Valley. It is clearly this form which is recorded
from southeastern Texas by Strecker (1915, p. 36) from Kerr,
Kendall, Comal, McLennan, and Matagorda counties.

REFERENCES
BAIRD, S. F.

1859. Reptiles of the Boundary (No. [3], pp. 1-35, pis. 1-41) in Emory, W. H.,
Report on the United States and Mexican Boundary Survey Made under
the Direction of the Secretary of the Interior. 2, part II, 62 +32 +35+11 pp.,
27+25+35+41 pis. Washington.

, and GIRARD, CHARLES

1853. Catalogue of North American Reptiles in the Museum of the Smithsonian
Institution. Part I. — Serpents. 8vo, XVI + 172 pp. Smithsonian Institution,
Washington.

BOGERT, C. M.

1939. A Study of the Genus Salvadora, the Patch-nosed Snakes. Pub. Univ.
Calif. Los Angeles, Biol., 1, pp. 177-236, figs. A-B, pis. 3-7, maps 1-2.

JAN, GIORGIO, and SORDELLI, FERNANDO

1860. Iconographie generate des ophidiens. Livr. 1, pis. 1-6.
SMITH, HOBART M.

1939. Notes on the Snakes of the Genus Salvadora. Univ. Kans. Sci. Bull.,
25, pp. 229-237, pi. 22 (1938).

STEJNEGER, LEONHARD

1902. The Reptiles of the Huachuca Mountains, Arizona. Proc. U. S. Nat.
Mus., 25, pp. 149-158.

STRECKER, J. K.

1915. Reptiles and Amphibians of Texas. Baylor Bull., 18, pp. 1-82.

THE LIBRARY OF THE

JUN131940

UNIVERSITY OF ILLINOIS

UNIVERSITY OF ILLINOIS-URBANA