Content uploaded by Jonathan A Campbell
Author content
All content in this area was uploaded by Jonathan A Campbell on Jan 23, 2016
Content may be subject to copyright.
Herpetologists' League and Allen Press
are collaborating with JSTOR to digitize, preserve and extend access to
Herpetologica.
http://www.jstor.org
Herpetologists' League
A New Species of Abronia (Squamata: Anguidae) from the High Cuchumatanes of Guatemala
Author(s): Jonathan A. Campbell, Mahmood Sasa, Manuel Acevedo, Joseph R. Mendelson and III
Source:
Herpetologica,
Vol. 54, No. 2 (Jun., 1998), pp. 221-234
Published by: on behalf of the Allen Press Herpetologists' League
Stable URL: http://www.jstor.org/stable/3893428
Accessed: 23-01-2016 21:56 UTC
REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.org/stable/3893428?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
Herpetologica, 54(2), 1998, 221-234
? 1998 by The Herpetologists' League, Inc.
A NEW SPECIES OF ABRONIA (SQUAMATA: ANGUIDAE) FROM
THE HIGH CUCHUMATANES OF GUATEMALA
JONATHAN
A. CAMPBELL,' MAHMOOD SASA,1 MANUEL ACEVEDO,2
AND JOSEPH R. MENDELSON 11134
'Department of Biology, The University of Texas at Arlington, Arlington, TX 76019, USA
2Departamento de Biologfa, Universidad de San Carlos, Guatemala City, Guatemala
3Museum of Natural History, The University of Kansas, Lawrence, KS 66045, USA
ABSTRACT: We describe a striking new lizard of the genus Abronia from the Sierra de Los
Cuchumatanes of Guatemala. Adults of the new Guatemalan species have a unique body pattern
of white or yellow transverse markings on a black background and a suite of morphological char-
acters that differentiate them from all congeners. It is the only species of Abronia known from
Guatemala that lacks protuberant supra-auricular spines. The closest relatives of this new species
appear not to be any of the members of the genus previously known from Guatemala, but rather
may lie with certain species occurring in El Salvador and Honduras that previously have been placed
in two different subgenera (Abaculabronia and Lissabronia). Re-evaluation of phylogenetic hypoth-
eses of Abronia suggests that the subgenus Abaculabronia contains only two species (A. reidi and
A. ornelasi) and that A. montecristoi should be placed in the subgenus Lissabronia (formerly con-
taining only A. salvadorensis), along with the new species described herein.
The Sierra de Los Cuchumantanes, like many regions in Latin America, currently is undergoing
an ecological disaster. The demands by humans on the land and its forests have reduced much of
the range to barren, grassy slopes or fields of exposed lateritic clays punctuated with karstic pin-
nacles. Undoubtedly, this range still contains a great biodiversity but currently receives no effective
protection. With the burgeoning and uncontrolled human population growth, it appears likely that
the once magnificent forests of the Sierra de Los Cuchumatanes will be reduced to a few stands of
trees on the steeper slopes by early in the next decade.
Key words: Reptilia; Squamata; Anguidae; Abroniafrosti new species; Cuchumatanes; Huehue-
tenango; Guatemala
IT has been long recognized that the
isolated, cool forests covering the slopes of
the larger mountain systems of Central
America are analogous to islands. These
forests contain a wealth of species and,
more importantly, most of the species that
they harbor have small geographic distri-
butions, often restricted to a single slope
of a highland range or peak. The Sierra
de
Los Cuchumatanes in western Guatemala
is certainly among the most poorly ex-
plored ranges in Central America with re-
gard to its herpetofauna. Herpetological
investigations within this large mountain
system, which extends about 150 km and
which has several regions over 3600 m in
elevation, have been relatively few and
limited mostly to a few farms
or along sev-
eral unpaved roads
that traverse
the region
(Elias, 1984; Stuart, 1943).
4PRESENT ADDRESS: Department of Biology, Utah
State University, Logan, UT 84322, USA.
Many members of the genus Abronia,
for whatever reason, have managed to es-
cape detection until fairly recently. Al-
though the first species of Abronia were
described in an early work pertaining to
the herpetofauna of Mexico (Wiegmann,
1828), about half of the 27 known species
of the genus, including undescribed spe-
cies from Guerrero and Chiapas, have
been discovered within the last two de-
cades. In the brief time since the last com-
prehensive revision of the genus by Camp-
bell and Frost (1993), two additional spe-
cies have been described (Brodie and Sav-
age, 1993; Campbell, 1994) and the
existence of an undescribed species from
Chiapas, Mexico was reported (Campbell,
1994).
MATERIALS AND METHODS
Scale definitions and protocols for mak-
ing scale counts follow Bogert and Porter
(1967), Campbell (1982, 1984, 1994), and
221
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
222 HERPETOLOGICA [Vol. 54, No. 2
Campbell and Frost (1993). The number
of longitudinal rows of dorsal scales can be
particularly confusing because of the re-
duced size of one or two rows just above
the ventrolateral fold; we standardized by
counting only rows containing enlarged
scales (Campbell and Frost, 1993:11, their
footnote 3). Head measurements were
made to the nearest 0.1 mm using vernier
calipers held under a dissecting micro-
scope; measurements of the body were
made using a meter stick. Comparative
material was re-examined in The Univer-
sity of Texas at Arlington (UTA) collection,
which contains specimens representing
most of the species of Abronia. Other ma-
terial examined is listed in Campbell and
Frost (1993:Appendix 1). We consulted
the published descriptions of morphologi-
cal features in Brodie and Savage (1993),
Campbell (1982, 1984, 1994), Campbell
and Frost (1993), and Good (1988). Geo-
graphic coordinates were taken with a
global positioning system (Magellan? mod-
el 4000) and verified using a 1:50,000 scale
map for Guatemala. All specimens of the
new species were fixed in buffered for-
malin (diluted to 10% of stock solution)
and transferred within several months into
70% ethanol for permanent storage.
Drawing of the head was made using a
stereomicroscope and associated drawing
tube. Notes of color in life were taken
from field notes and photographs of live
specimens. Format and terminology in the
diagnosis and description follow that of
Campbell and Frost (1993).
SYSTEMATIC ACCOUNT
During the summer of 1996, we made
herpetological collections at various sites
throughout the Cuchumatanes. At one lo-
cality, we discovered a remarkable new an-
guid lizard, which we propose be known
as
Abroniafrosti sp. nov.
Holotype.-The University of Texas at
Arlington (UTA) R-41131, an adult male
from along road to Patalcal, 5.9 km (by
road) NW intersection of Guatemala Road
9N (near San Mateo Ixtatan) 2835 m, Si-
erra de Los Cuchumatanes, Huehueten-
ango, Guatemala (15?51'19" N, 91?31'02"
W). Collected by J. A. Campbell (original
field no. JAC 19078) on 19 June 1996 in
highland hardwood forest (Fig. 1).
Paratypes (four specimens).-UTA R-
41134 (Fig. 2) and 41135 (an adult female
and subadult, respectively), AMNH
142607 (an adult male, previously UTA R-
41132), and KU 223920 (an adult male,
previously UTA R-41133), all from the
type-locality, 2800-2835 m, collected on
18-19 June 1996 by a field party consisting
of M. Sasa Marin, M. Acevedo, J. R. Men-
delson III, and J. A. Campbell (Fig. 3).
Diagnosis.-A species of Abronia in
which (1) the supra-auricular scales of sub-
adults and adults are small, knoblike, and
nonprotuberant; (2) supranasals, if pres-
ent, are relatively small and unexpanded,
not in contact at dorsal midline; (3) fron-
tonasal scale present, not contacting fron-
tal; (4) posterior internasals moderately
large, 1.5-2 times larger than the anterior
internasals; (5) canthals discrete; (6) two
anterior temporals per side, both contact-
ing the postoculars; upper primary tem-
poral not in contact with parietal, excluded
by large scale in secondary temporal se-
ries; (7) parietal contacting median supra-
oculars; (8) occipital single; (9) posterior
head scales (of adults) not strongly convex
or knoblike; (10) anterior superciliary con-
tacting cantholoreal, similar in length to
other scales in series; (11) posterior sub-
ocular broadly separated from the lower
primary temporal by the penultimate su-
pralabial, the posteriormost scale in this
series to reach the orbit; (12) granular
preauricular scales in two or three rows;
(13) postmental single; (14) posterior in-
fralabial not elongate; (15) longitudinal nu-
chal scale in six rows; (16) dorsal scales in
28-32 transverse rows; (17) dorsal scales
in 12 longitudinal scale rows, arranged
parallel relative to ventrolateral fold, indi-
vidual scales in successive transverse rows
are staggered rather than in linear series;
(18) ventral scales in 14-16 longitudinal
rows; (19) adults black or dark blackish
brown with white or yellow transverse
markings; (20) subadults brown with yel-
lowish crossbands.
Abronia frosti differs from all other
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
June 1998] HERPETOLOGICA 223
FIG. I.-Abronia frosti, holotype (UTA R-4113 1), adult male, 100 mm SVL. Reproduced from UTA slide
no. 21194.
FIG. 2.-Abroniafrosti, paratype (UTA R-41134), adult female, 110 mm SVL. Reproduced from UTA slide
no. 21208.
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
224 HERPETOLOGICA [Vol. 54, No. 2
FIG. 3.-Type-series of Abroniafrosti showing variation of dorsal pattern; from left to right, UTA R-41131,
KU 223920, AMNH 142607, UTA R-41134, and UTA R-41135. Reproduced from UTA slide no. 21211.
members of the genus Abronia in having
only two primary temporals, the upper-
most of which does not contact either the
parietal or frontoparietal, and a unique
color pattern of a black background with
pale transverse markings on the sides and
dorsum of the body. The number of pri-
mary temporals in many species of Abron-
ia is three or four; in individuals that have
two, the upper primary temporal is in
broad contact with the parietal and fron-
toparietal and extends posteriorly well past
the supraocular series. Abronia frosti dif-
fers from all congeners east of the Isthmus
of Tehuantepec, except for A. bogerti, A.
ramirezi, A. montecristoi, and A. salvador-
ensis in lacking protuberant supra-auricu-
lar spines. Abronia bogerti and A. ramirezi
differ from A. frosti in having contact be-
tween prefrontal-anterior superciliary
scales, 38 or more transverse rows of dor-
sal scales, eight longitudinal rows of nuchal
scales, and an elongate, slender head and
body. Abronia montecristoi and A. salva-
dorensis differ from A. frosti in having up-
per primary temporals that preclude pari-
etals from contact with the median supra-
oculars, 3-5 occipitals, and 12-14 longi-
tudinal ventral scale rows. Abronia frosti
differs from all species of Abronia occur-
ring east of the Isthmus of Tehuantepec,
except A. gaiophantasma, in having 12
(versus 14 or more) dorsal longitudinal
scale rows and usually 16 (versus 12-14)
ventral longitudinal scale rows. The num-
ber of ventral longitudinal scale rows in A.
frosti is 16 in four specimens and 14 in
one, whereas five A. gaiophantasma have
14 ventral longitudinal scales and the
number is increased to 15 and 16 in one
specimen each. Abronia gaiophantasma
differs from A. frosti in having spinelike
supra-auricular scales, a parietal not in
contact with the median supraoculars, and
a brown dorsal coloration, with or without
dark crossbands. Abronia frosti is com-
pared with other Guatemalan members of
the genus in Table 1.
Description of holotype.-An adult male
having a snout-vent length (SVL) of 100
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
a
TABLE 1.-Selected features of species of Abronia inhabiting Guatemala. Abronia nontecristoi may also occur in Guatemala, but its presence has not been
confirmed. Data are from Campbell and Frost (1993), Brodie and Savage (1993), and additional specimens that we have examined.
Abronia Abronia Abronia Abronia Abronia Abronia Abronia
Character anzuetoi aurita camnpbelli fimnbriata frosti gaiophantasrna inatudai
Supranasal scales Not expanded Not expanded Not expanded Expanded, usual- Not expanded Not expanded Often partially ex-
ly meeting at panded
midline
Frontonasal scale Present Present Present Absent Present Present Present
Canthal scales Present Present Present Absent Present Present Present
Anterior temporals per 2-3 Usually 3, rarely Usually 3-4, Usually 2, some- 2 3-4 Usually 2
side 2 or 4 rarely 5 times 3
Parietal contacting medi- Variable, usually Variable Variable Usually yes Yes No Yes
an supraoculars no
Supra-auricular scales in Spinelike, well Spinelike, well Spinelike, well Spinelike, well Granular, not Spinelike, well Protuberant but
adults developed developed developed developed protuberant developed not spinelike
Posteriormost infralabial Elongate Elongate Elongate Not elongate Not elongate Not elongate Not elongate 0
Postmental scale Divided Variable Usually single Single Single Variable Variable
Transverse dorsal scale 27-28 27-31 29-31 27-30 28-32 28-30 33-37
rows
Dorsal scale ornamenta- Keeled Keeled Keeled Keeled Smooth Keeled Keeled
tion
Longitudinal dorsal scale 14 14 14, rarely 15 14 12 12 Usually 16, rarely
rows 14
Transverse ventral scale 35-37 35-38 32-35 36-39 33-35 35-38 36-37
rows
Longitudinal ventral 14 14 Usually 14, rare- Usually 12, rare- Usually 16, rare- Usually 14, rare- Usually 12, rarely
scale rows ly 12 ly 14 ly 14 ly 16 14
Adult male color pattern Uniformly green Greenish with Gray or brown Uniformly Black or dark Brownish with Greenish with
indistinct with indistinct brownish gray-black darker cross- dark lateral
black cross- dark dorsal with white or bands flecking
bands crossbands yellow trans-
verse markings
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
226 HERPETOLOGICA [Vol. 54, No. 2
mm. The length of the head from the ros-
tral to the upper anterior edge of the au-
ricular opening is 21.9 mm, and the head
width at the broadest point is 23.8 mm
(width/length = 1.09). The tail appears to
be fully regenerated, having a length of
137 mm and 69 caudal whorls.
There are no supranasals. An upper and
lower postnasal is present on each side; the
former is less than half the size of the na-
sals. Two pairs of internasals lie between
the rostral and canthals-frontonasal; the
posterior internasals are about twice as
large as the anterior internasals. A single
canthal on each side is situated mostly on
the dorsum of the snout; these scales are
slightly broader than long, and separate
the posterior internasals from the prefron-
tals, loreal, and cantholoreal. The cantho-
loreal is about as long as high. The fron-
tonasal is broadly separated from the fron-
tal by a pair of large prefrontals and, at its
broadest point, is over twice the width of
the canthals. There are 5/5 median supra-
oculars, 3/4 lateral supraoculars, and 6/6
superciliaries (left/right sides, respective-
ly). The first superciliary broadly contacts
the cantholoreal and preocular. There are
1/1 preoculars, 3/3 suboculars, and 3/4
postoculars. The frontal narrowly contacts
the interparietal; the posterior end of the
interparietal contacts a single large occip-
ital which is flanked laterally by two large
postparietals (narrow occipital-postparie-
tal contact on left side, contact precluded
on right by intervention of parietal), which
extend posteriorly well past the occipital.
Two transverse rows of scales separate the
occipital from the first transverse row of
nuchals. There are two primary temporals
on each side; the lower is longitudinally
elongate and broadly contacts the next to
the upper postocular and posterior two su-
pralabials; the upper is about twice the
size of the lower, horizontally elongate,
and broadly contacts the upper postocular.
There are 3/3 secondary temporals and 3/3
tertiary temporals. The parietal contacts
the median supraoculars. The upper pri-
mary temporal does not contact the pari-
etal, being excluded by a large scale in the
secondary temporal series. There are 9/9
supralabials; the penultimate is the poster-
FIG. 4.-Abroniafrosti, (upper) dorsal and (lower)
lateral aspects of head of holotype (UTA R-41131);
head length = 21.9 mm.
iormost to reach the orbit. There are 8/8
infralabials. The postmental is not divided
and is followed by three pairs of enlarged
chinshields (scales following postenior pair
are about half the size of the chin shields).
There are 5/5 scales in the sublabial series;
the anteriormost reaches only the third in-
fralabial and does not contact the post-
mental (Fig. 4).
The minimum number of nuchals in a
transverse series is six. There are 30 trans-
verse and 12 longitudinal rows of dorsal
scales that are arranged in parallel rows on
the sides; dorsal scales are mostly flat;
about four middorsal longitudinal rows on
the dorsum are slightly convex but cannot
be considered keeled. There are 35 trans-
verse and 16 longitudinal rows of ventral
scales. The scales of the lateral longitudi-
nal ventral scale rows tend to be the larg-
est ventral scales on the body. Osteoderms
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
June 1998] HERPETOLOGICA 227
are well developed in the dorsal scales, es-
pecially on the head, neck, and proximal
portion of the tail. The supra-auricular
scales are granular and nonprotruding.
There are about eight moderately-sized
granular
scales between the lateral nuchals
and the first large scales on the ventrolat-
eral surface of the neck; there are 13 an-
tebranchials, counted from the insertion of
the arm to the wrist. The ventrolateral fold
has 2-3 subgranular scales and several
small dorsal scales separating the large
dorsals from the ventrals. There are 21/18
subdigital
lamellae on the fourth toes.
In preservative (ethanol after formalin)
the dorsum and sides of the head, body,
and tail are black. There are eight white
vertical markings
on the sides of the body,
consisting of a series of spots on vertical
rows of scales; irregular
white dorsal cross-
bands are also present that are offset from
the lateral
markings.
About four horizontal
markings
are present on the proximal
por-
tion of the tail; distally the tail becomes
almost uniformly dark. The chin and
throat are mostly white with black spot-
ting; the venter of the body and tail are
dark gray.
There are about eight irregular
transverse series of whitish spots crossing
the venter of the body. There is a pair of
distinctive
white markings
on the posterior
of the head, extending from the post-pa-
rietal region across the posterolateral
edge
of the head to the auricular
region. Several
white spots are present on the side of the
head on the loreal, preocular, and poste-
rior two supralabials.
The side of the neck
is marked
with several irregular
white ver-
tical markings. Both the fore and hind-
limbs have four white crossbars; these
cross bars are present on the dorsal and
ventral surfaces of the hindlimbs, but re-
stricted to the dorsal surfaces of the fore
limbs where the ventral surfaces
are
white.
The visceral peritoneum is black.
Variation.-Three of the five specimens
of the type-series, including the holotype,
are adult males with SVLs
of 100-101 mm
and cephalic indices (= HW/HL) of 0.94-
1.09 (* = 1.01). In adult males, the head
is exceptionally broadly expanded in the
temporal region. The single adult female
paratype
is 110 mm in SVL
with a cephalic
index of 0.99. The single subadult female
has a SVL of 78 mm and a cephalic index
of 0.86. The tail is complete (unregener-
ated) in three specimens and is 1.39-1.50
(x = 1.43) times the body length with 69-
77 (x = 72.7) caudal whorls. The number
of caudal whorls in the two specimens (the
holotype and the adult female) with regen-
erated tails is 69 and 77.
There are 1/1 small, unexpanded su-
pranasal scales in four individuals; in the
holotype these scales are absent. The
number of postnasals in three individuals
is 2/2, but the lower postnasal is absent in
two specimens. The number of internasals
is 2/2 in the three males, but there are
2/3 or 3/2 in the two females. The fronto-
nasal is invariably present and it fails to
contact the frontal. There are 1/1 canthal
scales in all specimens, but in UTA R-
41134 the canthals are small and there is
an additional small scale on each side be-
tween the canthal and the loreal. There
are 5/5 median supraoculars, one occipital,
six nuchals in the shortest transverse row
across the neck, two temporals reaching
the postoculars, 2/2 anterior temporals, a
single postmental, 3/3 chinshields, and 12
longitudinal rows of dorsal scales in all
specimens. There are three lateral supra-
oculars per side in all individuals except
one, which has four on one side. There are
9/9 supralabials in all but one specimen,
which has 9/10. There are seven (10%),
eight (60%), or nine (30%) infralabials per
side. The supra-auricular scales are gran-
ular to moderate-sized, knobby scales and
not spinelike or protuberant. The upper
primary temporal is invariably absent and
the parietal contacts the medial supraocu-
lars, precluding contact of the frontopar-
ietal with the uppermost element in the
primary temporal series. The number of
transverse scale rows between the occipital
and the first nuchal row is two in all but a
single specimen which has one. The shape
of the cantholoreal varies from higher than
long in three individuals to longer than
high in two. The number of preoculars is
1/1 in all but one specimen in which this
scale is horizontally divided, producing
two scales per side. There are two (50%)
or three (50%) suboculars; three (60%) or
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
228 HERPETOLOGICA [Vol. 54, No. 2
four (40%) postoculars; five (30%), six
(50%), or seven (20%) superciliaries; three
(10%), four (20%), five (50%), or six (20%)
sublabials. There are 8-10 lateral neck
scales between the lateral nuchals and the
first large scales on the ventrolateral sur-
face of the neck, 11-13 (x = 12.0) ante-
branchials, 28-32 (x = 29.4) transverse
rows of smooth dorsal scales, 33-35 (x =
34.4) transverse rows of ventral scales, 4-
6 (x = 5.2) ventral scales between the
hindlimbs, and 18-21 (x = 19.3) subdigital
lamellae on the fourth toe. There are 16
longitudinal rows of ventral scales in all
but one specimen, which has 14. This high
number of longitudinal rows of ventral
scales represents a unique condition in the
genus (a single specimen of A. gaiophan-
tasma has 16, but this represents an un-
usual condition for this species, according
to Campbell and Frost, 1993).
Color in life.-In life, the dorsal color-
ation of adults of this species is black to
dark blackish brown. Seven or eight pale
dorsal crossbands are evident on the body,
and these are usually somewhat staggered
on the middorsum. Pale crossbands may
be white or yellow and may consist of se-
ries of small spots on scales or include
most of the scales in single transverse se-
ries (Fig. 3). The tail is dark and marked
with pale transverse bars or rings which
tend to be most prominent on the proxi-
mal portion of the tail. The top of the head
is marked with a number of whitish, yel-
lowish, or pale gray spots or blotches, the
most prominent of which is a pair of
blotches extending from about the parietal
or post-parietal region across the posterior
temporal region to the upper border of the
ear. The chin and throat are mostly whitish
or yellowish with black markings; the ven-
ter of the body is black or dark brown with
ventral crossbands. The iris is copper-
bronze.
In the subadult (UTA R-41135), the
darker coloration was medium brown
(rather than black) which was much less
extensively distributed than the black pig-
ment of adults. The dorsum of the head
and body was pale chartreuse with brown
blotches. Chartreuse bars extended from
the dorsum down the sides of the body to
just past the ventrolateral fold. Most of the
venter was straw yellow with a few dark
brown markings. The tail had alternating
pale brown and tan rings.
Etymology.-The species name is a
noun in the genitive case, formed in honor
of Darrel R. Frost, in recognition of his
significant contributions to the knowledge
of lizard systematics, including those of
Abronia, and who brings new meaning to
the term maverick.
Habits and habitat.-The highland for-
est inhabited by A. frosti (Fig. 5) consists
of hardwood species of trees; no conifers
are present in the vicinity. The larger trees
attain heights estimated to be at least 30
m and form a dense canopy heavily laden
with mosses. This kind of forest has often
been referred to as montane wet forest
(Holdrige, 1967). All of the specimens of
the type-series were collected on 18 (one
specimen) and 19 (four) June 1996, days
that were mostly bright and sunny except
for early morning and late afternoon fogs.
Additionally, the slowest member of our
party secured the autotomized tail of an
Abronia during the first afternoon that we
visited the area; this lizard escaped be-
neath a huge fallen tree trunk. Five spec-
imens (the paratypes and escapee) were
encountered on the ground, and one in-
dividual (the holotype) was found basking
on a clump of moss growing about 2.5 m
from the ground on the side of a huge
tree. A small ray of sunlight that penetrat-
ed thick forest fell on the clump of moss.
This species appears to be highly arboreal,
as are other members of the genus, and
spends most of its life high in trees. We
attribute the presence of lizards on the
forest floor to the recent cutting of many
trees in the area. Several woodcutters in
the area assured us that they usually found
these lizards in the debris of tall, recently
felled trees.
We returned to the area on 10-12 No-
vember 1996. At this time, conditions
were cold, rainy, and overcast. No other
specimens of Abronia were seen.
Other species of amphibians and rep-
tiles that we collected near the type-local-
ity of A. frosti were Bolitoglossa rostrata,
Pseudoeurycea rex, Bufo bocourti, Norops
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
June 1998] HERPETOLOGICA 229
FIG. 5.-Type-locality of Abronia frosti, along road to Patalcal, 5.9 km (by road) northwest intersection of
Guatemala Road 9N (near San Mateo Ixtatan) 2835 m, Sierra de Los Cuchumatanes, Huehuetenango, Gua-
temala (15051'19" N, 91?31"02"
W). Photograph taken on 19 June 1996.
. D X _ rL j~~9 88
94- 92 90
18W-
18- ~ ~ ~ ~ ~ 1 1
A. ramirezl X 6
?A. bogerb
roi
A ornobsi * 0->
?A lyffirocha g 6 0
'A. ochoterenai\ ,
*A. gmd7j !fl
vA matudai \ Q-<A'}carnpbelh
A aurita ? a
? A. flmbriata < . a
* A gaiophanftsma
* A, nzuetoi 1. '
-14- ? A %wntecristoi -
A sahvadorensi 's .'
'VA burobpi 92- 90
FIG. 6.-Distribution of species of Abronia in Nu-
clear Central America. Modified from Campbell and
Frost (1993) with the addition of species described
subsequent to that paper. Abronia ochoterenai and A.
leurolepis known only from the same type-locality.
Dashed line shows approximate extent of the Sierra
de Los Cuchumatanes.
crassulus, Mesaspis moreletii, Sceloporus
taeniocnemis, Thamnophis fulvus, and
Cerrophidion godmani.
Little forest cover remains at the type-
locality of A. frosti (Fig. 5), and this small
patch of forest is one of the few remaining
in the area. This forest is now under severe
attack by woodcutters who peddle their
firewood, the major fuel source in the Cu-
chumatanes, in the larger towns in the re-
gion, especially San Mateo Ixtatan. How
long this resource will survive is unknown,
but it will not be for long. We were im-
pressed by the amount of destruction that
occurred between our two visits to the
area-a span of <5 mo. Given the current
rate of deforestation, the forest at the
type-locality of A. frosti will very soon be
gone. Whether this happens in 2 yr or five
is irrelevant. With the destruction of this
forest, A. frosti will disappear from the
area and probably the planet.
Distribution.-Northwestern highlands
of Guatemala (Fig. 6); known only from
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
230 HERPETOLOGICA [Vol. 54, No. 2
the vicinity of the type-locality in the
northern portion of the Sierra de Los Cu-
chumatanes in the Department of Hue-
huetenango, Guatemala. This locality lies
in montane wet forest of the Cuchumatan
Subarea of Campbell and Vannini (1989).
The site lies near the crest of the moun-
tains, referred to as Montafia Yolcultac
Chemtetzat on the Instituto Geografico
Nacional map (1:50,000, hoja 1863 I). Al-
dea Chibazalum
is located about 1 km to
the WNW of the type-locality on a slope
facing the Grijalva Depression; this small
hamlet is not found on any map known to
us. To the east of the type-locality, slopes
are drained by the Rio Pacumal, an upper
tributary of the Rio Nenton, and, to the
west, slopes are drained by the Rio Chex-
joj, an upper tributary
of the Rio Ixcain.
DISCUSSION
Examination of the map delimiting the
geographic distribution of Abronia in
Campbell and Frost (1993) seems to re-
veal a biogeographic enigma-the absence
of the genus from most of the Sierra de
Los Cuchumatanes. This apparent absence
is curious given that the genus Abronia oc-
curs in every major mountain range from
the northern portion of the Sierra Madre
Oriental in Mexico (Martin, 1958), south-
ward through the Mesa Central of Mexico
(Sanchez-Herrera and L6pez-Forment,
1980), the Mesa del Sur (Bogert and Por-
ter, 1967) and Sierra Madre del Sur of Oa-
xaca and Guerrero (Davis and Dixon,
1961), the Meseta Central and Sierra Ma-
dre of Chiapas (Alvarez del Toro, 1982;
Martin del Campo, 1939; Smith and Al-
varez del Toro, 1963), through all of the
other highlands of Guatemala
(Brodie and
Savage, 1993; Campbell and Frost, 1993),
to northern El Salvador (Hidalgo, 1983)
and southern Honduras (Wilson et al.,
1986). The Sierra de Los Cuchumatanes
is
joined to the west with the Meseta Central
in Chiapas
where three species of Abronia
occur: A. lythrochila, A. ochoterenai, and
A. leurolepis.
To the east, the Sierra
de Los
Cuchumatanes
is separated
from the high-
lands of Alta Verapaz
and the Sierra de las
Minas by the Rio Chixoy Valley,
and these
northeastern Guatemalan highlands also
harbor three species of Abronia: A. aurita,
A. fimbriata, and A. gaiophantasma. Fi-
nally, the Sierra de Los Cuchumatanes are
connected with the highlands of southern
Guatemala via the broad Guatemalan Pla-
teau (Stuart, 1951) and at least four spe-
cies of Abronia occur across these high-
lands: A. aurita, A. matudai, A. anzuetoi,
and A. campbelli. It is significant that all
of the species of Abronia that inhabit
regions adjacent to the Cuchumatanes ap-
pear to be closely related and were placed
in the same subgenus, Auriculabronia
(Campbell and Frost, 1993). However, A.
frosti appears to be most closely related to
the two other species of Abronia in Cen-
tral America lacking protuberant supra-au-
ricular spines. We suspect that additional
field work may establish the occurrence in
the Cuchumatanes of several additional
species of Abronia whose relationships lie
with members of Auriculabronia possess-
ing prominant protuberant supra-auricular
spines.
For phylogenetic analysis, we used the
matrix of Campbell and Frost (1993: Ap-
pendix 4, characters numbered), but made
the following modifications. The parietals
are variously in contact or not with the me-
dian supraoculars in A. fimbriata (coded as
in contact by Campbell and Frost, 1993:
69). The upper scales in the primary tem-
poral series are absent in A. frosti with
only the lower two scales present. An un-
usual condition exists in this species where
no scale of the primary temporal series
makes contact with the parietal with the
exception of KU 223920 in which there is
very narrow contact on one side. Because
upper primary temporals are absent in A.
frosti, the character whereby parietals are
separated or not by these scales from the
median supraoculars is treated as unknown
(Character 10: no contact = 0, variable =
1, contact = 2, unknown = ?). We record
the number of occipital scales in A. mon-
tecristoi as three (Character 14: one = 0,
two = 1, three = 2); previously this species
was considered to have five occipitals
(Campbell and Frost, 1993), but re-ex-
amination suggests that the two posterior
lateral scales are more appropriately con-
sidered elements of the first transverse nu-
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
June 1998] HERPETOLOGICA 231
chal row. Previously, the circumorbital re-
gion of A. matudai was considered to be
the same color as the head (Campbell and
Frost, 1993:73). However, we now know
that the coloration of the circumorbital re-
gion in A. matudai is variable; we have
seen some adults in which this region as
green, similar to the rest of the side of the
head, and others in which it is bright yel-
low, but not as extensive as in A. anzuetoi
or A. aurita. Further, the circumorbital
coloration should be considered unknown
in A. ochoterenai (Character 29: circum-
orbital coloration same as head = 0, vari-
able = 1, consistently yellow = 2, un-
known = ?). The only known specimens of
A. ochoterenai examined by Campbell and
Frost (1993) were preserved over 50 yr
prior, and it is apparent that the pale cir-
cumoribtal coloration may become dark-
ened in preservative under certain circum-
stances. For the first time, we recently
have had the opportunity to examine live
adult specimens of A. matudai (since pre-
served and deposited at UTA); in all of
these specimens, the lip coloration was red
in life and not the ground color of the low-
er jaw (character 30: infralabials ground
color of lower jaw = 0, reddish = 1). Fi-
nally, we added data to the matrix based
on Brodie and Savage (1993) for A. camp-
belli, Campbell (1994) for A. ramirezi, and
data presented herein for A. frosti.
This revised data matrix was analyzed
using the PAUP Version 3.1 (Phylogenetic
Analysis Using Parsimony) program of
Swofford (1993). Heuristic search strate-
gies were employed to find optimal trees
using random stepwise additions with 100
repetitions, tree bisection-reconnection
(TBR), and MULPARS options. This anal-
ysis produced nine equally parsimonious
trees with 134 steps (CI = 0.381, RI =
0.672, RC = 0.256). The level of support
for each branch in the strict consensus of
the nine shortest trees (Fig. 7) was as-
sessed by calculating Bremer (decay) In-
dices (Bremer, 1988, 1994); these indices
were calculated using the program Auto-
decay ver. 3.0.3 (Eriksson and Wikstri5m,
1995). Abaculabronia and Scopaeabronia
are the most strongly supported subgenera
(Bremer Indices of four and nine, respec-
tively) in the present analysis;
the remain-
ing subgenera were each supported by a
Bremer Index of one. In this analysis,
A.
frosti forms a clade with several other
Central American species that lack protu-
berant supra-auricular
spines (i.e., A. mon-
tecristoi and A. salvadorensis), and this
clade forms the sister-lineage
to the larger
clade containing the Nuclear Central
American species placed in the subgenus
Auriculabronia (Campbell and Frost,
1993). Attempts at resolving relationships
of certain species of Central American
Abronia that lack supra-auricular
spines
have been problematical. Campbell and
Frost (1993) placed A. montecristoi in the
subgenus Abaculabronia with several oth-
er species, and A. salvadorensis was placed
by itself in the subgenus Lissabronia,
based largely
on the condition of the scales
in the temporal region. However, Camp-
bell and Frost (1993) cautioned that the
homologies of temporal scales, particularly
in rows 2 and 3, is frequently not clear.
Temporal scale contacts form a Rubic's
Cube analog, which can make simple as-
signments based on topography ambigu-
ous. In other words, with the kind of vari-
ation present in some species, it is difficult
enough to know how to code for number,
much less how to know whether a partic-
ular scale might have been lost.
On the basis of this new analysis and
consideration of new information,
we now
regard all of the species of Central Amer-
ican Abronia lacking protuberant supra-
auricular
spines to form a sister-group to
the subgenus Auriculabronia
(Fig. 7), rec-
ognized here as the subgenus Lissabronia.
Our notion of the infrageneric taxonomy
the genus Abronia conforms with that of
Campbell and Frost (1993), with the ex-
ception that Abronia nontecristoi is trans-
ferred to the subgenus Lissabronia, and
thus the subgenus Abaculabronia retains
only two species (A. reidi and A. ornelasi).
The subgenus Lissabronia comprises a
clade containing
A. frosti, A. montecristoi,
and A. salvadorensis. The diagnoses for
the various subgenera were provided by
Campbell and Frost (1993) and only that
for Lissabronia need be modified to ac-
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
232 HERPETOLOGICA [Vol. 54, No. 2
co
CZ cu C u
0 2 EZ
1, N- CoN ?5 ' co
CZ
tW
t;T
:3'
.
co co u-_CZI) u
J-C
Lissabronia
C I l ~~~~~~Auriculabronial
Scopaeabronia Abronia
Aenigmabronia
*-
Abaculabronia
FIG. 7.-Strict consensus of nine equally parsimonious trees (134 steps, CI = 0.381, RI = 0.672) derived
from unscaled data matrix modified from Campbell and Frost (1993: Appendix 4). The diagnosis for the
subgenus Lissabronia is given in the text; diagnoses for other subgenera are provided by Campbell and Frost
(1993). Unless otherwise indicated, Bremer (1988) indices for all branches are 1. The taxon "Guerrero"
represents an unnamed species; another undescribed species from Chiapas is not included in this analysis.
comodate the inclusion of additional spe-
cies into the subgenus.
We here provide a rediagnosis for the
subgenus Lissabronia. The diagnosis
should now read: Lissabronia is distin-
guished from all other subgenera in having
supranasal scales that are unexpanded (ex-
panded in Abaculabronia), in lacking fron-
tonasal-frontal scale contact (usually pres-
ent in Abaculabronia and a few other spe-
cies), in having anterior superciliary scales
of about similar length to following scales
(lanceolate in Scopaeabronia and often in
Abaculabronia), in having the posterior
subocular scale separated from the lower
primary temporal (in contact in Abacula-
bronia and Aenigamabronia), in lacking ex-
panded lower primary temporal scales (ex-
panded in Scopaeabronia), in lacking pro-
tuberant head shields on the posterolateral
"corners" of the head (head casquing in the
subgenus Abronia), in lacking protuberant
supra-auricular scales in adults (present in
Auriculabronia), in having <38 transverse
rows of dorsal scales (38 or more in Sco-
paeabronia), in having fewer than eight
longitudinal rows of nuchal scales (eight in
Scopaeabronia), and in having the lateral
most rows of ventral scales expanded (not
expanded in Aenigmabronia, Scopaeabron-
ia, and Abronia). The distribution of Lis-
sabronia includes the northern portion of
the Sierra de Los Cuchumatanes in Gua-
temala, the Cordillera de Alotepeque-Me-
tapan in northwestern El Salvador,
the Cor-
dillera de Nahuaterique in El Salvador, and
the Sierras de Opalaca and Montecillos in
Honduras. The elevational range for the
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
June 1998] HERPETOLOGICA 233
subgenus is from 1900-2835 m in mesic,
montane forests.
In attempting to identify A. frosti in the
key provided by Campbell and Frost
(1993), an impasse will be reached at cou-
plet 12 where the choices are "longitudinal
scale rows on side of body arranged in
oblique rows .. ., dorsal body scales very
slightly keeled, usually almost flat," versus
"longitudinal scale rows on sides of body
arranged parallel ..., dorsal body scales
with definite keel." Abronia frosti has dor-
sal scales arranged in parallel fashion on
the sides of the body relative to the ven-
trolateral fold and the dorsal body scales
are smooth. We provide below a revised
key, restricted to the species of Abronia
occurring in Guatemala, and which con-
tains A. frosti and A. campbelli, species
that were described subsequent to Camp-
bell and Frost (1993).
KEY TO THE SPECIES OF ABRONIA
OF
GUATEMALA
1. Transverse rows of dorsal scales 33 or
more; longitudinal rows of dorsal scales
usually 16, rarely 14 ....... A. matudai
Transverse rows of dorsal scales 32 or
fewer; longitudinal rows of dorsal
scales 12-14 .......... .... 2
2. Adults (and juveniles) lacking spinelike
supra-auricular scales; dorsal scales
smooth; in life, adults black or black-
ish brown with white or yellow trans-
verse markings ............. A. frosti
Adults with spinelike supra-auricular
scales; some dorsal scales keeled, al-
beit weakly; adult color pattern not as
above .............. 3
3. Longitudinal rows of dorsal scales 12 . .
.................. A. gaiophantasma
Longitudinal rows of dorsal scales 14 .. 4
4. Frontonasal absent; canthal scales ab-
sent; posteriormost infralabial not
elongate; circumorbital region same
color as head ........... A. fimbriata
Frontonasal present; canthal scales pres-
ent; posteriormost infralabial elongate;
in life, circumorbital region yellow or
tan, distinctly set off from rest of head
coloration .......5.......
5. In life, ground color of dorsum brown,
gray, to dull gray-green; second pri-
mary temporal usually broadly con-
tacting the enlarged lower tertiary
temporal .......... Abronia campbelli
In life, ground color of dorsum vivid
green, yellowish green, or blue-green;
second primary temporal rarely in
contact with lower tertiary temporal
....... ....................... 6
6. In life, body ground coloration pea-
green, yellow-green, or pale tur-
quoise; dorsal body scales heavily pig-
mented with black, with anterior
por-
tion of scales almost entirely dark ...
......................... A. aurita
In life, body ground coloration
dark
em-
erald green or dark
blue-green;
dorsal
body scales sparsely pigmented with
black, anterior portion of scales not
black . A. anzuetoi
RESUMEN
Se describe una Ilamativa nueva especie
de lagartija del genero Abronia de la Sierra
de Los Cuchumatanes, Guatemala. Los
adultos de esta nueva especie guatemal-
teca tienen un patron uinico de marcas
transversas blancas o amarillas sobre un
fondo negro, ademas de poseer un grupo
de caracteres morfologicos que los difer-
encian de sus congeneres. Esta es la uinica
especie de Abronia conocida de Guate-
mala que carece de espinas supra-auricu-
lares protuberantes. Al parecer, los parien-
tes mas cercanos a esta nueva especie no
son ninguno de los miembros del genero
previamente descritos para Guatemala,
sino ma's bien ciertas especies existentes
en El Salvador y Honduras, previamente
colocados en dos subgeneros diferentes
(Abaculabronia y Lissabronia). Una re-ev-
aluacion de la filogenia de Abronia sugiere
que el subgenero Abaculabronia contiene
solo dos especies (A. reidi y A. ornelasi) y
que tanto A. montecristoi como la nueva
especie descrita aqui deben ser ubicadas
en el subgenero Lissabronia (que antig-
uamente incluia solo a A. salvadorensis).
La Sierra de Los Cuchumatanes, como
otras muchas regiones en America Latina,
actualmente estai sufriendo un desastre
ecologico. Las demandas para el uso de la
tierra y bosques han reducido gran parte
de la sierra a pastizales inf6rtiles o a zonas
de arcillas lateriticas expuestas marcadas
con pinaculos calizos. Indudablemente,
esta sierra contiene aun gran biodiversi-
dad, pero actualmente no recibe ninguna
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
234 HERPETOLOGICA [Vol. 54, No. 2
proteccion. Con el aumento incontrolado
de la poblacion humana, los una vez es-
plendidos bosques de la Sierra de Los Cu-
chumatanes seran reducidos a unos cuan-
tos arboles en las pendientes m-as pron-
unciadas durante la proxima decada.
Acknowledgnents.-This material is based in part
upon work supported by the Texas Advanced Re-
search Program under grant 003656-001 to J. A.
Campbell and the Tinker Field Research Program of
The University of Kansas Center for Latin American
Studies to J. R. Mendelson III. We thank C. A. Sheil
for rendering Fig. 4. Permits for conducting research
in Guatemala were granted by officials of the Consejo
Nacional de Areas Protegidas (CONAP); we are es-
pecially grateful to Lic. Oscar F. Lara and Licda.
Mygdalia Garcia de Sol6rzano.
LITERATURE CITED
ALVAREZ DEL TORO, M. 1982. Los Reptiles de Chia-
pas, ed. 3. Instituto Zool6gico del Estado, Tuxtla
Guti6rrez, Chiapas, Mexico.
BOGERT, C. M., AND A. P. PORTER. 1967. A new spe-
cies of Abronia (Sauria, Anguidae) from the Sierra
Madre del Sur of Oaxaca, Mexico. Am. Mus. Novit.
2279:1-21.
BREMER, K. 1988. The limits of amino acid sequence
data in angiosperm phylogenetic reconstruction.
Evolution 42:795-803.
BREMMER, K. 1994. Branch support and tree stabil-
ity. Cladistics 10:295-304.
BRODIE, E. D., JR., AND R. F. SAVAGE. 1993. A new
species of Abronia (Squamata: Anguidae) from a
dry oak forest in eastern Guatemala. Herpetologica
49:420-427.
CAMPBELL, J. A. 1982. A new species of Abronia
(Sauria, Anguidae) from the Sierra Juarez, Oaxaca,
M6xico. Herpetologica 38:355-361.
. 1984. A new species of Abronia (Sauria: An-
guidae) with comments on the herpetogeography
of the highlands of southern Mexico. Herpetologi-
ca. 40:373-381.
. 1994. A new species of elongate Abronia
(Squamata: Anguidae) from Chiapas, Mexico. Her-
petologica 50:1-7.
CAMPBELL, J. A., AND D. R. FROST. 1993. Anguid
lizards of the genus Abronia: revisionary notes, de-
scriptions of four new species, a phylogenetic anal-
ysis, and key. Bull Am. Mus. Nat. Hist. 216:1-121.
CAMPBELL, J. A., AND J. P. VANNINI. 1989. Distri-
bution of amphibians and reptiles in Guatemala
and Belize. Proc. West. Found. Vert. Zool. 4:1-21.
DAVIS, W B., AND J. R. DIXON. 1961. Reptiles (ex-
clusive of snakes) of the Chilpancingo Region,
Mexico. Proc. Biol. Soc. Washington 74:37-56.
ELIAS, P. 1984. Salamanders of the northwestern
highlands of Guatemala. Contr. Sci. Nat. Hist. Mus.
Los Angeles Co. 348:1-20.
ERIKSSON T., AND N. WIKSTROM. 1995. Autodecay
ver. 3.0.3 (Computer program distributed by the
authors). Botaniska institutionen, Stockholm Uni-
versity, Stockholm, Sweden.
GOOD, D. A. 1988. Phylogenetic relationships among
Gerrhonotine lizards. Univ. California Publ. Zool.
121: 1-138.
HIDALGO, H. 1983. Two new species of Abronia
(Sauria: Anguidae) from the cloud forests of El Sal-
vador. Occas. Pap. Mus. Nat. Hist. Univ. Kansas
105:1-11.
HOLDRIDGE, L. R. 1967. Life Zone Ecology. Tropical
Science Center, San Jos6, Costa Rica.
MARTIN, P. S. 1958. A biogeography of reptiles and
amphibians in the Gomez Farias region, Tamauli-
pas, Mexico. Misc. Publ. Mus. Zool. Univ. Michi-
gan 101:1-102.
MARTiN DEL CAMPO, R. 1939. Contribuci6n al con-
ocimiento de los Gerrhonoti Mexicanos, con la pre-
sentaci6n de una nueva forma. An. Inst. Biol. Univ.
M6xico 10:353-361.
SANCHEz-HERRERA, O., AND W L6PEZ-FORMENT C.
1980. The lizard Abronia deppei (Sauria: Anguidae)
in the state of Mexico, with the restriction of its
type locality. Bull. Maryland Herpetol. Soc. 16:83-
87.
SMITH, H. M., AND M. ALVAREZ DEL TORO. 1963.
Notulae herpetologicae chiapasiae IV. Herpetolo-
gica 19:100-105.
STUART, L. C. 1943. Comments on the herpetofauna
of the Sierra de los Cuchumatanes of Guatemala.
Occas. Pap. Mus. Zool. Univ. Michigan 471:1-28.
. 1951. The herpetofauna of the Guatemalan
Plateau, with special reference to its distribution on
the southwestern highlands. Contr. Lab. Vert. Biol.
Univ. Michigan 49:1-71.
SWOFFORD, D. L. 1993. PAUP: Phylogenetic Analysis
Using Parsimony. Version 3.1.1. Computer pro-
gram and user's manual distributed by the Illinois
Natural History Survey, Champaign, Illinois, U.S.A.
WIEGMANN, A. F. A. 1828. Beitrage zur Amphibien-
kunde. Isis von Oken 21:364-383.
WILSON, L. D., L. PORRAS, AND J. R. MCCRANIE.
1986. Distributional and taxonomic comments on
some members of the Honduran herpetofauna.
Milwaukee Pub. Mus. Contr. Biol. Geol. 66:1-18.
Accepted: 29 June 1997
Associate Editor: John Wiens
This content downloaded from 129.107.108.172 on Sat, 23 Jan 2016 21:56:05 UTC
All use subject to JSTOR Terms and Conditions
