[Pg 33]

Mammals Obtained by Dr. Curt von Wedel from

the Barrier Beach of Tamaulipas, Mexico

BY




E. RAYMOND HALL







University of Kansas Publications

Museum of Natural History



Volume 5, No. 4, pp. 33–47, 1 figure in text
October 1, 1951





University of Kansas

LAWRENCE

1951

[Pg 34]University of Kansas Publications, Museum of Natural History





Editors: E. Raymond Hall, Chairman, A. Byron Leonard,

Edward H. Taylor, Robert W. Wilson



Volume 5, No. 4, pp. 33–47, 1 figure in text



October 1, 1951





University of Kansas

Lawrence, Kansas





PRINTED BY

FERD VOILAND, JR., STATE PRINTER

TOPEKA, KANSAS

1951



23–7414

[Pg 35]

Mammals Obtained by Dr. Curt von Wedel from the Barrier Beach of

Tamaulipas, Mexico

BY



E. RAYMOND HALL

WHAT species of mammals occur on the "coastal island", barrier beach, of
Tamaulipas? Are the closest relatives of these mammals on Padre and
Mustang islands of Texas, instead of on the mainland of Tamaulipas, or
are the mammals on the barrier beach distinct from all others? These
were questions that Dr. von Wedel of Oklahoma City and I asked ourselves
in March of 1950 when we were in southern Texas. With the aim in mind of
answering these questions, Dr. von Wedel arranged round-trip
transportation, by air, for the two of us between Brownsville, Texas,
and Boca Jesús María. The latter place is a "pass", tidal inlet, through
the long barrier beach. The waters of the Gulf of Mexico and of the
lagoon behind the beach flow back and forth with the changing tides
through the inlet.

We arrived at Boca Jesús María on March 18, 1950, and left on March 22,
1950. Our headquarters there were in one of the four one-story buildings
immediately north of the inlet. This place is approximately 89½ miles
south, and 10 miles west, of Matamoros, Mexico. Most of our collecting
was done on the sand dunes one and one-half miles north of the buildings
but on the evening of March 20 we made a round-trip, by boat of course,
to the sand dunes on the south side of the inlet to set traps; these
traps, and the Dipodomys that were caught in them, were picked up the
following morning.

At the time of our visit, the part of the barrier beach south of the
tidal inlet was connected with the mainland. The connection was far to
the southward, according to our pilot, Mr. Kagy of Brownsville, and also
according to the testimony of the Mexicans at the fishing camp where we
stayed on the north side of the inlet. The barrier beach which lay to
the north of the inlet extended sixty-odd miles northward to the delta
of the Río Grande and had, we were told, eight "passes," including Paso
Jesús María. At the time of our visit, however, only three of these
tidal inlets were open, it was said; the five others were thought to be
filled in with sand, which permitted terrestrial animals to move from
one part of the beach to another. Dr. von Wedel and I saw two tidal
inlets that were open when we were being flown back to Brownsville.

[Pg 36]

Fig. 1. Diagram of physiographic features of the barrier
beach of Tamaulipas. Top view looking down, as from an airplane, on the
beach. Bottom view is profile.

[Pg 37]

The long, low, sandy island, technically a barrier beach, irrespective
of tide varied in width from a quarter of a mile to as much as a mile
and was separated from the mainland by the Laguna Madre, which was four
miles wide opposite our trapping station. To the northward the width of
the lagoon gradually increased until, at a place thirty miles north of
our trapping station, the lagoon was almost 20 miles wide.

The island was perhaps four feet above high tide. Superimposed on this,
in places, there were sand dunes, technically barchans, so arranged that
the end of one touched the end of the next. The tops of some were as
much as 20 feet above high tides and the chain of these connected-dunes
on which we trapped was approximately a mile long. Incipient tidal
inlets were frequent; they were where storm-driven waves of high tides
had broken across the island between the adjacent ends of two dunes. The
windward side of a dune was toward the Gulf and the slope of that side
was gentler than that on the leeward side. According to the cycle
described by Davis (Proc. Amer. Acad. Arts and Sciences, 22:303–332,
1896) and recently figured on page 364 by Lobeck (Geomorphology, 1st
ed., xii + 731 pp., 1939, McGraw Hill Book Co., Inc., New York) the
barrier beach concerned was in the early part of the "Middle Youth
Stage".

Typically, on the center of the area in the lee of a dune there was a
patch of plum brush, almost five feet tall and so dense that a person
could not penetrate it. A belt of grass, 20 to 100 feet wide, surrounded
the plum brush. The grass was approximately 20 inches high. Outside the
area of grass, there were widely-spaced xerophitic shrubs which grew
also on the dunes. The diagram (fig. 1) shows these prominent features
as a person might see them if he looked directly down from an airplane.

We obtained specimens of the spotted ground squirrel (Citellus
spilosoma), Ord kangaroo rat (Dipodomys ordii), hispid cotton rat
(Sigmodon hispidus) and black-tailed jack rabbit (Lepus
californicus). Tracks and other sign of the coyote (Canis latrans)
were seen. So far as we could ascertain, by our own investigations and
from our Mexican hosts at the fishing camp, no other kinds of native
mammals lived on the island. The ground squirrel and kangaroo rat were
found by us on only the sandy areas where there were[Pg 38] xerophitic shrubs.
The cotton rat was found only in the grass. The jack rabbit and coyote
ranged over the whole of the island excepting the areas of plum brush in
which we saw no sign of any mammal.

To answer the second of our initial questions: The affinities of the
mammals of the barrier beach of Tamaulipas are approximately equally
divided between those of the mainland and those of Padre Island. The
ground squirrel is indistinguishable from the subspecies which occurs
both on the mainland and Padre Island to the northward; the other three
kinds of mammals of which we obtained specimens prove to be
subspecifically distinct from any previously named kinds and seem to be
confined to the off-shore beach. Accounts of these four mammals and of a
previously unnamed subspecies of kangaroo rat on Mustang Island, Texas,
follow.

Citellus spilosoma annectens (Merriam)

Spotted Ground Squirrel

Thirteen specimens (Nos. 35441–35453) were collected. All are from the
north side of the tidal inlet. Although the ground squirrels were easily
trapped, it was difficult to obtain a perfect skin because the gulls
(Larus sp.) pulled the skin off of the distal part of the tail as soon
as a squirrel was secured in a trap. The specimens seem not to differ
from Texan specimens from the type locality and Mustang Island.

Dipodomys ordii parvabullatus new subspecies

Ord Kangaroo Rat

Type.—Male, adult, skull and skin, No. 35454, Mus. Nat. Hist. Univ.
Kansas, from island, 88 miles south and 10 miles west of Matamoros,
Tamaulipas, Mexico; obtained 19 March 1950 by E.R. Hall and Curt von
Wedel; original No. 6778 E.R. Hall.

Range.—Islands along coast of Tamaulipas, Mexico.

Diagnosis.—Size small (see measurements). Color pale; entire dorsal
surface Light Ochraceous-Buff (Capitalized color terms according to
Ridgway: Color Standards and Color Nomenclature, Washington, D.C.,
1912), purest on sides and flanks, upper parts lightly suffused with
black; cheeks white; plantar surfaces of hind feet, dorsal and ventral
stripe of tail, and anterior face of ear brownish. Skull small; auditory
bullae smaller (actually and relative to remainder of skull) than in any
other known kind of Dipodomys, excepting the one from Mustang Island,
Texas (named beyond) in which the breadth is approximately the same;
rostrum and interorbital region narrow.

[Pg 39]

Comparisons.—From Dipodomys ordii sennetti (Allen), of the
mainland of Texas, D. o. parvabullatus differs in: Color paler on
pigmented areas; white areas more extensive; skull smaller, in all parts
measured, except the nasals which are slightly longer. From Dipodomys
ordii compactus of Padre Island, Texas, D. o. parvabullatus differs
in: Tail and hind foot shorter; skull smaller in all parts measured,
especially so in breadth across maxillary processes of zygomatic arches.

Remarks.—D. o. parvabullatus resembles D. o. sennetti in external
proportions and D. o. compactus in cranial proportions.

No difference was detected between specimens from the two sides of the
tidal inlet 89 miles south of Matamoros. Only one of the 14 specimens is
of the light color phase (upper parts Cartridge Buff). This pale
specimen is from the north side of the inlet. The brownish stripe on the
ventral side of the tail is absent on the distal two-fifths of the tail
and the specimens are uniform in this respect. On the occlusal surfaces
of the cheek-teeth, the enamel surrounding the dentine is incomplete on
both the lingual and labial sides of the teeth of five individuals and
is incomplete on the labial side of some of the teeth of a sixth
specimen.

In the snap traps, all of which were baited with rolled oats, more than
twice as many land crabs as kangaroo rats were taken. Judging from
tracks in the sand, land crabs greatly outnumbered kangaroo rats. The
parietal bones in two of the 13 skulls are much eroded by some parasite
(seemingly nematode worms) and in one of these two specimens the roof of
the left tympanic cavity is perforated. As regards life-zones, the
occurrence of Dipodomys ordii in the lower part of the Lower Sonoran
Life-zone on the off-shore beach 88 and 90 miles south of Matamoros is
low zonally and perhaps is at or near the zonal margin of the range of
the species. The crabs and worms conceivably are two of the
environmental features inhospitable to the rats.

Specimens examined.—Total, 14, all from Tamaulipas, Mexico, as
follows: 88 mi. S and 10 mi. W Matamoros, 7; 90 mi. S and 10 mi. W
Matamoros, 7.

When Setzer (Univ. Kansas Publ., Mus. Nat. Hist., 1:473–573, December
27, 1949) reviewed the subspecies of Dipodomys ordii he lacked
specimens of Dipodomys ordii compactus from the type locality or from
anywhere else on Padre Island. He used as representative of D. o.
compactus specimens from Mustang Island, Texas, the island next
northeast of Padre Island. Through the courtesy of Mr. Stanley P. Young,
Dr. Hartley H.T. Jackson and Miss Viola S.[Pg 40] Schantz, of the United
States Biological Surveys Collection, I have examined topotypes of D.
o. compactus from Padre Island. This examination discloses that the
kangaroo rats on Padre Island and Mustang Island are significantly
different. Those from Mustang Island may be named and described as
follows:

Measurements (in millimeters) of adult males of four subspecies of

Dipodomys ordii

Dipodomys ordii largus new subspecies

Ord Kangaroo Rat

Type.—Female, adult, skull and skin, No. 27234, Mus. Nat. Hist.,
Univ. Kansas, from Mustang Island, 14 mi. SW Port Aransas, Aransas
County, Texas; obtained 30 June 1948 by W.K. Clark; original No. 543.

Range.—Known from Mustang Island only.

Diagnosis.—Size medium (see measurements). Color pale, and as
described for D. o. parvabullatus. Skull small; auditory bullae
(actually and relative to remainder of skull) smaller than in any other
known kind of Dipodomys, except D. o. parvabullatus in which breadth
across bullae is approximately the same; notably narrow across maxillary
processes of zygomatic arches.

[Pg 41]

Comparisons.—From Dipodomys ordii sennetti (J.A. Allen) of the
mainland, D. o. largus differs in: Color paler on pigmented areas;
white areas more extensive; skull averaging smaller except in basilar
length and length of nasals which are approximately the same as in D.
o. sennetti. From Dipodomys ordii compactus True of Padre Island, D.
o. largus differs in: Body longer; tail shorter; skull narrower across
tympanic bullae and across maxillary processes of zygomatic arches;
nasals shorter. From Dipodomys ordii parvabullatus of the coastal
island south of Padre Island, along the gulf coast of Tamaulipas, D. o.
largus differs in: Body and tail longer; basilar length of skull
averaging less; breadth across maxillary processes of zygomatic arches
greater; premaxillae not extending so far behind nasals.

Remarks.—D. o. largus resembles D. o. compactus in external
proportions and D. o. parvabullatus in cranial proportions. The degree
of difference between D. o. compactus and D. o. largus is less than
between D. o. compactus and D. o. parvabullatus. To me, the three
subspecies mentioned in the preceding sentence are indistinguishable in
color.

Two of the eleven specimens of D. o. largus are of the light color
phase (upper parts Cartridge Buff) whereas all but two of the eleven
specimens of D. o. compactus are of the light color phase. Each of the
cheek-teeth of the upper jaw of D. o. largus has a complete ring of
enamel around the dentine of the occlusal surface, as described by
Setzer (Univ. Kansas Publ., Mus. Nat. Hist., 1:517, December 27, 1949)
for D. o. compactus. The upper dentitions of ten specimens of D. o.
compactus examined by me in this respect reveal a total of only five
teeth (in four individual animals) that have the enamel ring incomplete;
one premolar and three molars are incomplete on the lingual side and one
molar is incomplete on the labial side.

Two specimens from Bagdad, Tamaulipas, in the delta of the Río Grande
(Nos. 116485 and 11487, U.S.N.M., Biol. Surv. Coll.), are referred to
D. o. compactus on basis of long body and long tail. The specimens,
both Light Ochraceous-Buff, are so young that not all of the enamel is
worn off the crowns of the cheek-teeth. Specimens of D. o. compactus,
D. o. parvabullatus and D. o. sennetti of comparable age are not
available, and it, therefore, is impossible to know whether size and
shape of the skull in the population at Bagdad are the same as they are
in D. o. compactus of Padre Island.

Specimens examined.—Total, 11, all from Texas. Aransas County:
Mustang Island, 14 mi. SW Port Aransas.

[Pg 42]

Sigmodon hispidus solus new subspecies

Hispid Cotton Rat

Type.—Male, adult, skull and skin; No. 35468, Mus. Nat. Hist., Univ.
Kansas; from island, 88 mi. S and 10 mi. W Matamoros, Tamaulipas,
Mexico; 22 March 1950; obtained by E.R. Hall and Curt von Wedel;
original No. 6806 E.R. Hall.

Range.—Known from the type locality only but probably occurring on
most of the chain of islands off the coast of Tamaulipas.

Diagnosis.—Small; hind foot short; rostrum broad.

Comparison.—From its nearest relative, geographically and
morphologically, Sigmodon hispidus berlandieri Baird of the adjacent
mainland, S. h. solus differs in smaller size, and a rostrum that is
broader in relation to the length of the skull.

Remarks.—On the last night of our stay on the island, traps set in
grass approximately 20 inches high, yielded one pair of Sigmodon. The
color is lighter than in the average of specimens from the mainland (for
instance those from Victoria and Soto la Marina) but can be matched by
selected specimens. In animals of equal age, the hind foot and basilar
length are shorter in S. h. solus than in berlandieri. The
broadening of the rostrum, which occurs with advanced age, is attained
in solus when the skull is yet short; the maximum breadth of the
rostrum in the adults is more, instead of less, than a fourth of the
basilar length.

Measurements.—The following measurements are of specimens in which
the occlusal face of each molar tooth is worn flat. The first
measurement is of the holotype followed by the corresponding measurement
of a male of T. b. berlandieri, No. 116466 from Camargo, Tamaulipas,
in parentheses. The third measurement is that of the female from the
type locality of S. h. solus and it is followed by the corresponding
measurement of a female of T. b. berlandieri, No. 116462 from Camargo,
Tamaulipas. Total length, 266 (298),—(293); length of tail, 113
(135),—(137); length of head and body, 153 (163), 155 (156); length of
hind foot, 30 (35), 30 (33); basilar length of Hensel, 28.2 (28.9); 27.9
(29.0); zygomatic breadth, 19.5 (—), 19.0 (20.8); mastoidal breadth,
13.9 (14.4), 13.9 (14.8); greatest breadth of rostrum, 7.2 (7.3), 7.8
(7.2); length of nasals, 14.6 (14.1), 13.4 (14.2); crown length of upper
molar teeth, 6.3 (6.1), 6.3 (5.9).

Specimens examined.—Two from the type locality.

Lepus californicus curti new subspecies

Black-tailed Jack Rabbit

Type.—Female, adult, skull and skin, No. 35470, Mus. Nat. Hist.,
Univ. Kansas; from island, 88 miles south and 10 miles west of
Matamoros, Tamaulipas, Mexico; obtained 19 March 1950 by E.R. Hall;
original No. 6783.

Range.—Islands along coast of Tamaulipas, Mexico.

Diagnosis.—Color pale; size small; ears short; tympanic bullae small.

[Pg 43]

Comparisons.—From Lepus californicus merriami Mearns (specimens
from Fort Clark, Brownsville and intermediate localities), L. c. curti
differs in paler color, lesser size except ear that is of almost same
length and except interorbital breadth that is approximately same in the
two subspecies; tympanic bullae notably smaller. From Lepus
californicus altamirae Nelson, L. c. curti differs in having the
black patch on the nape less definitely divided by a median,
longitudinal band of buffy color, and lesser size. Exception is to be
made for the ear and tympanic bullae, which are of approximately the
same size in the two subspecies.

Remarks.—The subspecific part of the name Lepus californicus curti
is proposed in honor of Dr. Curt von Wedel who shared the pleasure of
collecting on the islands where this handsome hare lives.

The specimens of L. c. curti are all females, which, in the genus
Lepus, average larger than the males. Comparison of the measurements
recorded below with those in the account by Nelson (N. Amer. Fauna,
29:129, 1909) may not reveal the full measure of difference in size
between L. c. curti and other subspecies because Nelson (op. cit.)
pooled males and females in obtaining the average measurements that he
records. For example, he used three males and two females of Lepus
altamirae in obtaining an average (op. cit.:117). The specimens of
L. c. curti here recorded are thought to be of full size inasmuch as
the degree of fusion of bones in the skull, and the density of the
cranial bones indicate full adulthood for each specimen.
Reproductive-wise, there is no question as to adulthood; each of the
four females was pregnant. One specimen had two embryos (each 30
millimeters long in crown-rump measurement) and each of the other
specimens contained one embryo. These three embryos were 55, 60, and 105
mm. long.

Three of our specimens, including the holotype, were obtained north of
the eighth pass and the other specimen, No. 35473, was obtained a few
hundred yards south of the pass. Because the part of the barrier beach
south of the pass was connected to the mainland, it is likely that the
newly named subspecies occurs also on the adjacent mainland; however, we
have examined no specimens of Lepus californicus from the opposite
mainland except from Matamoros, ninety miles to the north, and from
Altamira, approximately one hundred and fifty miles south of our
collecting locality. A specimen from Matamoros, Tamaulipas, and several
from Brownsville, Texas, in size of auditory bullae, larger overall size
and darker color clearly are L. c. merriami and not L. c. curti.

The small tympanic bullae of the specimens from Padre Island[Pg 44] were
commented upon by Nelson (op. cit.:149) who found smallness of bullae
to characterize many of the specimens from the eastern part of the
geographic range of L. c. merriami. In the northeastern part of the
geographic range of L. c. merriami, as Nelson pointed out, the small
size of the tympanic bullae was one of several evidences of
intergradation there with Lepus californicus melanotis, the subspecies
next adjacent to the north. In the light of present information, it
seems that the smallness of the tympanic bullae in the specimens (3)
from Padre Island may be an independent development—an adaptation to
environmental conditions that reaches its fullest development on the
same chain of islands eighty-odd miles southward of Matamoros. The
specimens from Padre Island, although possessing small bullae, in other
features, for example, larger size of other parts, are merriami.

The four specimens of L. c. curti are in worn winter pelage and the
new pelage is coming in on the thighs. Most of the specimens (6) of the
L. c. altamirae are in the same condition of pelage. In color and
color pattern, the two subspecies are, to me, indistinguishable except
that the black patch on the nape is less widely and less definitely
separated into two parts by a median, longitudinal, band of buffy color.

Lepus californicus altamirae was named by Nelson (Proc. Biol. Soc.
Washington, 17:109, May 18, 1904) as a black-tailed jack rabbit, Lepus
merriami altamirae, but was later transferred by Nelson (N. Amer.
Fauna, 29:124, 1909) to the white-sided section of the genus and
arranged as a full species, Lepus altamirae. In making this transfer,
Nelson (op. cit.:125) wrote that in "This well marked species ... the
lack of a black patch on the posterior half of the ear at the tip and
the white flanks (somewhat obscured in some of the original specimens)
are strong characters which place it in the callotis group."
"Posterior half of ears white without any trace of black at tip", was
the way Nelson (op. cit.:124) described the ears in L. altamirae. My
examination of the original series including the type, reveals that the
ears do have some black at the tip of the posterior half in three of the
specimens, some brown in one other specimen, and only a dusky tinge in
two others. In the four specimens of L. c. curti the tip of the ear is
faintly brownish in one animal and dusky in three. The extent of the
white flanks seems to be identical in the two series. Fortunately they
are in the same pelage and same stage of molt on the hind legs. The one
difference that I can detect is in the coloration of the nape. In each
of the specimens of L. altamirae the coloration is as described by
Nelson (op. cit.:124):[Pg 45] "nape with two lateral black bands extending
back from base of ears, and separated by a median band of buffy." In L.
c. curti the nape is all black in one specimen and the median band of
buffy is present in the other three but is narrower and more dusky than
in L. altamirae. Since the characters (color of tip of ear and extent
of white on the flank) relied upon by Nelson for placing L. altamirae
in the callotis group are duplicated in the californicus group, in
L. c. curti, there is reason for questioning whether altamirae is
correctly placed, taxonomically, in the L. callotis group.

Cursory examination of skulls of the callotis group and the
californicus group indicates that the prepalatal spine (the part of
the palate which extends anteriorly toward the vomer) is longer in L.
californicus than in L. callotis, L. gaillardi and L. alleni. In
this feature, L. altamirae agrees with Lepus californicus and
differs from members of the Lepus callotis group. Furthermore, the
newly named L. c. curti, in color of ear and color of nape, is
intermediate between L. altamirae and L. c. merriami. Consequently,
Lepus merriami altamirae Nelson, it seems, should stand as Lepus
californicus altamirae.

Mention should be made here of the view of Shamel (Proc. Biol. Soc.
Washington, 55:25–26, May 12, 1942) that the californicus group should
be divided into two groups (each group possibly amounting to something
more than a species and something less than a subgenus) on the basis of
a white rump and complex infolding of the enamel layer of the front of
the first upper incisor versus a dark rump and simple infolding of the
mentioned layer of enamel. He placed Lepus californicus merriami
Mearns, among other subspecies, in a group different from the one in
which he placed several other subspecies of Lepus californicus.

Specimens (skulls with accompanying skins) of the species Lepus
californicus in the Biological Surveys Collection of the United States
National Museum, representative of a gradual transition from the dark
rump and simple fold in the enamel to the white rump and complex fold in
the enamel are as follows: L. c. deserticola, No. 29733/41808,
Paharanagat Valley, Nevada; Nos. 117463 and 156744, Beals Spring,
Arizona. L. c. texianus, No. 24635/32031, Springerville, Arizona; No.
97453, Roswell, New Mexico; No. 118751, Toyah, Texas; No. 118749,
Valentine, Texas; and No. 108700, Terlingua Creek, Texas. In the
continuously distributed species Lepus californicus, along the
northwest to southeast line provided by the localities of occurrence
listed immediately above, there is a gradual transition from one kind of
fold to the other kind and from one color[Pg 46] of rump to the other color.
It is clear that Shamel (op. cit.) was in error in his conclusions;
the kinds of black-tailed jack rabbits to which Shamel (op. cit.)
applied the name Macrotolagus should stand as given below.

Measurements (in millimeters) of adults of two subspecies of
Lepus
californicus

The cranial measurements given above are taken, in so far as possible,
in the same way that the measurements recorded by Nelson in his North
American Fauna (No. 29, 1909) were taken. In that publication he records
mostly average measurements but he records also some measurements of
individual specimens. Two of these specimens are the holotypes of
Sylvilagus mansuetus Nelson and Romerolagus nelsoni Merriam. By
attempting to duplicate Nelson's measurements on these specimens, the
following opinions were formed.

[Pg 47]

Specimens examined.—Total, four, all from Tamaulipas, Mexico, as
follows: 88 mi. S and 10 mi. W Matamoros, 3; 90 mi. S and 10 mi. W
Matamoros, 1.

Transmitted February 20, 1951.

23–7414