HARVARD UNIVERSITY

Library of the

Museum of

Comparative Zoology

I

OCCASIONAL PAPERS

«f the ^

MUSEUM OF NATURAL HISTORY ^- Oo^p ^
The University of Kansas ®'^4^y ^

Lawrence, Kansas *^' 4 ^qq^

NUMBER no, PAGES 1-9 APRIL 17, 1984
^TT-

A NEW FROG OF THE GENUS ELEUTHERODACTYLUS
(LEPTODACTYLIDAE) FROM GUATEMALA

By

Linda S. Ford' and Jay M. Savage^

Among the biotically least known regions of Central America are the
Sierras de Chama, Pocolha, Xucaneb, Chuaciis, las Minas, and Mico, a
series of transverse ranges that extend in a generally easterly direction
from the main Guatemalan highland toward the Caribbean coast. The
ranges form a group of high ridges that rarely exceed 2100 m in elevation,
catch rain from the prevailing northeast tradewinds, and support a
luxuriant cloud forest from around an elevation of 1500 m to their tops.

Recently, a new road running south from Purulha, Baja Verapaz, into
the western edge of the Sierra de las Minas has been opened so that
collectors from the University of California, Berkeley, and the University
of Texas, Arlington, have been able to sample the herpetofauna from the
region. More or less simultaneously, primarily through the efforts of the
distinguished Guatemalan biologist. Lie. Mario Dary Rivera, a section of
undisturbed forest along the new road between Purulha and La Union
Barrios has been set aside as a nature preserve. While the principal goal of
the preserve is to protect typical habitat for the Quetzal, it also has modest
support facilities for scientific field studies of other members of the biota.

It seems appropriate, under these circumstances, that a distinctive new
frog originally collected in this region and subsequently found in the Sierra
Xucaneb to the north, be called:

Eleutherodactylus daryi new species
Figure 1

Eleutherodactylus greggi: Savage, 1975.

Holotype.—KV 186202, an adult male from 3.8 km (by road) SE

' Department of Biology, University of Texas at Arlington, Arlington, Texas 76019.
(Present address: Museum of Natural History, University of Kansas, Lawrence. Kansas
66045.)

2 Department of Biology, University of Miami, Coral Gables, Florida 33124.

2 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

Purulha, Baja Verapaz. Guatemala, 1585 m; collected by J. A. Campbell
and L. S. Ford on 4 July 1980.

Paranpes.— GUATEMALA, BAJA VERAPAZ: UTACV A-5990
(adult male), 3.4 km SE (by rd) Purulha, 1600 m; UTACV A-5985-86,
8266,8505 (adult females), 3.7 km SE (by rd) Purulha, 1615 m; UTACV
A-5603 (adult female), 7.7 km SSE (by rd) Purulha, 1615 m; UTACV
A-6181-82,6184,6186-87 (adult females), A-6813 (adult male), vicinity of
La Union Barrios, 1524-2100 m; MVZ 130785 (adult male), 4.2 km S
Purulha. 1760 m; MVZ 150933 (aduU female), 4 km NE Chilasco, 1900
m: MVZ 144529 (adult female), 4 km ENE Chilasco, 1829 m; MVZ
150936 (aduU male), 5 km NE Chilasco. 1900 m; MVZ 175813 (aduh
female), 2.5-4.8 km S Purulha, 1630-1720 m. Additional (immature or
juvenile males and females) specimens not included with measurements of
adults. -GUATEMALA, BAJA VERAPAZ: UTACV A-5540,8285, 3.7
km SE (by rd) Purulha, 1615 m; UTACV A-5604, 7.7 km (by rd) SSE
Purulha, 1615 m; UTACV A-5580,8008, Cerro Verde, 1707 m; UTACV
A-6185, vicinity of La Union Barrios, 1524-2100 m; KU 189642-43, 3.8
km (by rd) SE Purulha, 1585 m and 1650 m, respectively; MVZ
130784,130786, 4.2 km S Purulha, 1760 m; ALTA VERAPAZ: MVZ
104648-49, Finca Chichen, 12 km (airline) due S Coban, about 1850 m.

Diagnosis.— The new species is readily separable from almost all
Mexican and Central American members of the genus in having the
following combination of features: no tarsal fold; no toe webs; an areolate
venter; finger II slightly longer than finger I; non-emarginate finger disks;
eyelids, dorsum, flanks, and upper limb surfaces extremely rugose and

Figure \ .—Eleutherodactylus duryi
graph courtesy of J. A. Campbell.

adult male holotype (KU 186202), 22.7 mm. Photo-

A NEW ELEUTHERODACTYLUS FROM GUATEMALA 3

covered with numerous enlarged and elevated pustules and ridges which
frequently are tuberculate and tipped with white glandular areas; and a
distinct light para-anal bar on each side lateral to the dark anal patch.

It most closely resembles Eleutherodactylus greggi (Bumzahem, 1955)
of the Pacific versant of extreme eastern Chiapas, Me'xico, and adjacent
western Guatemala, but the two are readily distinguished by the following
features (the characteristics for the new form in parentheses): dorsal
surfaces smooth to rugose, never with most surfaces covered by pustules
and/or white tubercles and ridges (dorsal surfaces covered with prominent
pustules, tubercles and ridges that are often tipped with white glandular
areas), males without thenar nuptial pad (a distinct white thenar nuptial
pad in males) and size, adult males 28-36 mm in snout-vent length,
females 34-46 mm, in greggi (males 18-23 mm, females 26-35 mm).

The only other form with which E. daryi might be confused in
Guatemala is E. matudai which has a geographic range similar to that off.
greggi, in eastern Chiapas, Mexico, and western Guatemala, but at lower
elevations along the Pacific slope. Eleutherodactylus daryi resembles
matudai in having an extreme development of dorsal pustules and
tuberculations, but cannot be confused with the latter species which has a
smooth venter, definite toe webs, finger I longer than finger II and a much
larger size (adult males 36-50 mm in SVL, adult females to 65 mm).

Description.— WQdid about as wide as long, broader than body. Snout
subelliptical in dorsal view, rounded in lateral profile. Nostrils slightly
protuberant, projecting laterally; almost terminal. Canthus rostralis sharp,
ridgelike, concave in dorsal view. Loreal region concave, sharply angled.
Eyes large; diameter about 'A wider than loreal length. Choanae small,
oval to round. Vomerine processes triangular, about twice size of choanae,
separated posteriorly by V2 width of processes; one row of teeth on
posterior edge. Tongue ovoid, V4 length attached. No vocal slits or sacs.
Tympanum distinct, round; width V2 diameter of eye in females, -/? eye
diameter in males.

Skin of dorsum, head and dorsal surface of limbs strongly pustulate to
tuberculate with prominent scattered complex warts and ridges, often each
tipped with a light glandular area; supratympanic fold extends from lower
portion of eye to posterior part of lower tympanum; postocular fold
extends from upper eyelid, across scapula, posteriorly on dorsum,
terminating either behind arm, midbody or at vent. Eyelid extremely
warty; prominent ridge-like tubercle posterior to jaw articulation. Definite
small inguinal gland present. Venter granulate.

Finger II slightly longer than I, finger disks not emarginate; I and II
rounded. III and IV slightly expanded. Whitish nuptial pads on thumbs of
males. Subarticular tubercles round, flat to globular in profile; no
supernumerary tubercles. Palmar tubercle flattened, ovoid; thenar tuber-
cle elongate, globular; small flat round acessory palmar tubercle below
basal phalanx on finger II and IV, two tubercles below basal phalanx on
III.

Toe disks not emarginate; I, II, V rounded. III and IV slightly expanded

4 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

laterally. No toe webbing. Subarticular tubercles round to ovoid, globular
to obtuse in profile; no supernumerary tubercles. Outer metatarsal
tubercle rounded, flattened to globular; V3 size of elongated, obtusely
protruding inner tubercle. No tarsal folds, but outer edge of tarsal segment
extremely warty. Sides and posterior thighs below vent areolate.

Jaw musculature.— T\\c depressor mandibulae originates as three dis-
tinct slips, one each from the dorsal fascia, the squamosal and the annulus
tympanicus (DFSQAT, according to the system of Starrett, 1968). An
adductor mandibulae externus superficialis is present (e). (Two specimens
examined: UTACV A-5986,8266).

Coloration in ///e.— Dorsum and limbs light to medium brown, tuber-
cles and ridges slightly darker; in preservative, brown base color darkens,
tubercles and ridges less contrasted than in life. Lips with blotched pattern;
supratympanic and postocular ridge usually dark; snout color same as
dorsum, much lighter in some specimens. Dorsum uniform with light
middorsal stripe or blotch behind scapula on some specimens. Snout and
flanks of one adult female and one juvenile cream, head and dorsum dark
brown. First and second finger with white margin. Dorsal forearm, shank
and thigh faintly to prominently barred. Posterior thigh surface brown;
post-anal light bar separating dark brown anal patch from posterior thigh
area on each side and forming a distinct chevron-shaped light mark.
Posterior venter and underside of limbs mottled rusty color; upper chest
mottled white in life. Throat uniform brown, slight stippling. Venter
cream-colored in some preserved specimens. Iris gold, brightest dorsally
with fine black specks and reticulations (in life); bronze wedge extends
anteriorly and posteriorly from pupil.

Measurements.— Sx\o\xX-\tr\i lengths are given in millimeters; other
measurements are expressed as percentages of the snout-vent. The mean is
followed by the range in parentheses.

Snouth-vent length, adult males (N = 5) 20.9 (18.7-22.7), adult females
(N= 14) 29.8 (25.8-34.8); head length, males 44.2 (43.3-45.3), females
43.4 (39.9-46.5); head width, males 45.1 (43.2-50.2), females 43.3
(40.6-46.9); eye width, males 14.8 (13.1-16.3), females 13.4 (12.2-
14.8); snouth length, males 19.9 (18.3-20.7), females 19.4 (17.2-22.2);
loreal length, males 11.5 (9.8-12.3), females 10.5 (9.5-12.4); tympanum
width, males 9.7 (9.6-9.8), females 7.0 (6.3-7.6); tibia length, males 64.9
(61.6-70.9), females 65.2 (61.2-69.0); foot length, males 59.9 (57.6-
64.2), females 59.7 (54.4-66.5); hind leg length, males 204.8 (192.8-
217.6), females 200.8 (185.3-216.7). There is obvious sexual dimorphism
in snout-vent length and tympanum width. Juveniles ranged in snout-vent
length between 8.5-24.7 mm.

Measurements of holotype in millimeters: Snout-vent length 22.7; head
length 10.1; head width 11.4; eye width 3.7; snout length 4.6; loreal
length 2.6; tympanum width 2.2; tibia length 14.2; foot length 13.5; hind
leg length 45.1.

Habitat. —The male holotype was collected during the day in moist leaf
litter along a stream in cloud forest. Most of the UTACV and KU

A NEW ELEUTHERODACTYLUS FROM GUATEMALA 5

specimens were sitting by day on rocks in shallow water or in damp leaf
litter along small mountain streams surrounded by heavy foliage.

Distribution.— Lov/er montane rainforests of the Sierra Xucaneb, Alta
Verapaz, and Sierra de las Minas, Baja Verapaz, Guatemala; 1500-2100 m
(Fig. 2).

Relationships

Savage and DeWeese (1979, 1981) have commented at some length on
the difficulties inherent in establishing relationships among the many
(approximately 400) species of the genus Eleutherodactylus. Essentially,
they argue that contemporary attempts to cluster species into larger
divisions as proposed by Savage (1976, 1980) and Lynch (1976) are
inadequate, since they are based upon trivial features of morphology that
are subject to many convergences in the genus. Lynch's divisions, here
called series, are based primarily on the "key" characters of relative
finger (I versus II) lengths and whether the venter is smooth or coarsely
areolate. Savage's species groups are comprised of morphologically very
similar species that are probably related evolutionarily and would form
clusters within the series defined by Lynch. While both these approaches
are useful for taxonomic sorting of the many species, it seems likely that
they result in non-monophyletic groupings in many instances. As pointed
out by Savage and DeWeese (1979, 1981), resolution of this problem
awaits fuller utilization of the features of jaw musculature (Starrett, 1968),
karyology (DeWeese, 1976), and electrophoretic analysis of genetic
features (Harris, 1973; Miyamoto, 1981).

Within the genus, Eleutherodactylus daryi most closely resembles in
morphology E. greggi of extreme southeastern Chiapas, Mexico and
adjacent southwestern Guatemala and E. omiltemanus of the Sierra Madre
del Sur of Guerrero, Mexico. They share the features of having narrow
non-emarginate finger and toe disks, no tarsal fold or tubercle, no toe
webbing, finger I shorter than II, a strongly granulate (areolate) venter, a
distinct sub-integumentary inguinal gland, and no vocal slits in adult
males. In his early work. Lynch (1970) regarded greggi and omiltemanus
as allies of E. mexicanus, but substantially modified this position to
conform with his emphasis (Lynch, 1976) on relative finger lengths and
venter texture in establishing his series. According to Lynch's latter
arrangement mexicanus and its relatives belong in the rhodopis series,
including rhodophis, hobartsmithi, and pygmaeus of Mexico and northern
Central America and bransfordii of Costa Rica. Members of this series
have smooth venters and fingers I-II equal in length but otherwise show
considerable similarity to greggi and omiltemanus.

Lynch then suggested that greggi and omiltemanus conformed so closely
with his definition of the speciose unistrigatus series that they should be
placed there. This series has nearly 100 described forms in South and
lower Central America, with one undoubted member, E. ridens, ranging
to eastern Hondurus. The distributions of the omiltemanus-greggi-daryi
group are separated from the range of ridens by a distance of some 600
km.

OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

3

O

■«-»

c
o
o

1)

C8

•3
c

u

c

T3
1)

o
•o

u

c

So
s

S
o

s

s

a
S3

c
o

3

O

A NEW ELEUTHERODACTYLUS FROM GUATEMALA 7

The important features provided by jaw musculature and karyology
throw considerable light upon this situation. As pointed out by Starrett
(1968) the most consistent and evolutionarily significant aspect of these
muscles is in the condition of the adductor mandibulae. Three states of this
character can be recognized: s + e, both an adductor mandibulae posterior
subexternus and an externus superficialis present; s, only the subexternus
present; or e. only the externus superficialis present. Eleutherodactylus
daryi, greggi, and omiltemanus all agree in having condition e. All
members of the imistrigatus series of Lynch (1976) in which the
musculature is known have condition s.

The features of the depressor mandibulae muscles also provide signifi-
cant data for establishing relationships within the genus. Starrett (1968)
recognized 10 patterns of depressor mandibulae origins in frogs. Six of
these occur in the Leptodactylidae and four are known in Eleutherodac-
tylus. The conditions are: dfsq, a single slip primarily from the dorsal
fascia, but with a few fibers from the squamosal; dfsqat, similar to the
above, but with a few fibers from the annulus tympanicus; DFSQAT, three
distinct slips from the dorsal fascia, squamosal, and annulus; DFSQjAT,
three distinct slips with superficial slips from the dorsal fascia and annulus
covering a deeper slip from the squamosal. Savage and DeWeese (1979)
have pointed out that the first two conditions may be regarded as slight
variants in a single character state. The depressor mandibulae of daryi,
greggi, and omiltemanus are identical and of the DFSQAT condition. In all
members of the unistrigatus series where the muscles are known (17 of
approximately 100 species), the condition is DFSQ^AT.

Karyologically. members of the unistrigatus series for which data are
available have 2N = 26,32,34 and a nombre fundamental (N.F.) =
32,36,46 (11 of approximately 100 species). Conversely in E. greggi,
2N = 22 and the N.F. =40 (DeWeese, 1976).

These data convince us that the daryi-greggi-omiltemanus group forms a
natural unit that cannot be associated with Lynch's (1976) unistrigatus
series because of the consistent and marked differences in jaw musculature
and the suggestive difference in greggi's karyotype. Although Lynch
(1976) placed two other Mexican forms, E. batrachylus Taylor of
Tamaulipas and E. glaucus Lynch of Chiapas, in the unistrigatus series,
the status of these taxa are unclear. The former is known from a single
specimen, the latter from one adult and two juveniles. While we question
the assignment of batrachylus and glaucus to any species group, since they
are inadequately known, it seems very unlikely that they belong with
unistrigatus. Whether they are related to daryi and its allies remains to be
determined when additional material becomes available that will allow
examination of jaw musculature and karyotypes as well as direct compari-
son with other species.

In conclusion, daryi, greggi, and omiltemanus form a distinctive and
compact group of related frogs which may be called the omiltemanus
group. This stock is characterized by: narrow, non-emarginate disks; no
tarsal fold or tubercle; no toe webs; finger I shorter than II; an inguinal

8 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

gland; no vocal slits in adult males; jaw muscle formula DFSQAT + e;
karyotype, 2N = 22, N.F. =40, in the one known species (greggi). The
relationships of this distinctive assemblage to other stocks of Eleuthero-
dactylus remains, as with much else in the genus, enigmatical.

ACKNOWLEDGEMENTS

We wish to thank J. A. Campbell for his collection of most of the
specimens and invaluable assistance in Guatemala; the Instituto Nacional
Forestal (INAFOR) for graciously issuing permits; Dr. W. F. Pyburn for
his helpful suggestions concerning the manuscript; and Drs. W. E.
Duellman of the University of Kansas Museum of Natural History, D. B.
Wake and H. W. Greene of the Museum of Vertebrate Zoology for loan of
specimens. Partial financial support for the senior author was provided by
the Society for the Study of Amphibians and Reptiles (SSAR) graduate
student grant-in-herpetology. Society for Sigma Xi grant-in-aid of re-
search, and the UTA Department of Biology Research Assistantship.
Finally we acknowledge our debt of gratitude to the late Lie. Mario Dary
Rivera, who was the Rector of the Universidad de San Carlos of
Guatemala at the time of his death. We mourn the senseless assassination
of this scholar, statesman, and friend. We are honored to name this
distinctive new frog after him in recognition of his outstanding contribu-
tions and dedication to conservation, especially the establishment of the
Biotopo Universitario para la Consen'acion del Quetzal (posthumously
renamed Biotopo Mario Dary) which is included in the range of this new
species.

RESUMEN

Una nueva especie de leptodactflido, Eleutherodactylus daryi, es
descrita de el bosque lluvioso montano bajo en la Sierra Xucaneb, Alta
Verapaz, y la Sierra de las Minas, Baja Verapaz, Guatemala. Esta forma se
distingue de los otros miembros del genero de Centro America por la
siguiente combinacion de caracteres: sin pliegue tarsal; sin membranas
interdigitales en las patas; vientre areolado; dedo II ligeramente mas largo
que el dedo I; parpados, dorso, y flancos de la superficie de las
extremidades superiores muy rugosos y cubiertos con pustulas numerosas,
agrandadas, y elevadas y con pliegues que son frecuentemente tuber-
culados y cubiertos con areas blancas y glandulares; y una banda clara
para-anal a ambos lados de la mancha obscura anal. Esta nueva especie es
considerada un miembro del grupo omiltenmnus que es descrito en esta
publicacion.

LITERATURE CITED

BUMZAHEM, C. B. 1955. A new species of frog in the genus Eleutherodactylus from
Guatemala. Copeia 1955: 118-119.

DeWeese, J. E. 1976. The karyotypes of Middle American frogs of the genus Eleutherodac-
tylus (Anura: Leptodactylidae): A case study of the significance of the karyologic
method. Ph.D. diss., Univ. Southern California: 1-210.

A NEW ELEUTHERODACTYLUS FROM GUATEMALA 9

Harris,' R. T. 1973. Comparative serum-protein electrophoresis and protein characterization
in the systematics of Costa Rican frogs genus Eleutherodactylus (Anura: Leptodac-
tylidae). Ph.D. diss., Univ. Southern California: 1-247.

Lynch, J. D. 1970. Ta.xonomic notes on some Mexican frogs {Eleutherodactylus: Leptodac-
tylidae). Herpetologica 26: 172-180.

. 1976. The species groups of the South American frogs of the genus Eleuthero-
dactylus (Leplodactylidae). Occ. Pap. Mus. Nat. Hist. Univ. Kansas 61: 1-24.

Miyamoto, M. M. 1981. Cladistic studies of Costa Rican frogs genus Eleutherodactylus:
Phylogenetic relationships based on multiple character sets. Ph.D. diss., Univ.
Southern California: 1-217.

Savage, J. M. 1975. Systematics and distribution of the Mexican and Centra! American
stream frogs related to Eleutherodactylus rugulosus. Copeia 1975: 254-306.

. 1976. A preliminary handlist of the herpetofauna of Costa Rica. Second edition.

Editorial Univ. de Costa Rica: 1-19.

1980. A preliminary handlist of the herpetofauna of Costa Rica. Third edition.

Hancock Foundation, Univ. Southern California: 1-21.
Savage, J. M. and J. E. DeWeese. 1979. A new species of leptodactylid frog, genus

Eleutherodact\lus , from the Cordillera de Talamanca, Costa Rica. Bull. Southern

California Acad. Sci. 78: 107-115.
. 1981. The status of the Central American leptodactylid frogs Eleutherodactylus

melanostictus (Cope) and Eleutherodactylus platyrhynchus (Gunther). Proc. Biol.

Soc. Wash. 93(4): 928-942.
Starrett, p. H. 1968. The phylogenetic significance of the jaw musculature in anuran

amphibians. Ph.D. diss., Univ. Michigan: 1-179.

University of Kansas Publications

MUSEUM OF NATURAL HISTORY

The University of Kansas Publications, Museum of Natural
History, beginning with volume 1 in 1946, was discontinued with
volume 20 in 1971. Shorter research papers formerly published in the
above series are now published as Occasional Papers, Museum of
Natural History. The Miscellaneous Publications, Museum of Natural
History, began with number 1 in 1946. Longer research papers are
published in that series. Monographs of the Museum of Natural
History were initiated in 1970. All manuscripts are subject to critical
review by intra- and extramural specialists; final acceptance is at the
discretion of the publications committee.

Institutional libraries interested in exchanging publications may
obtain the Occasional Papers and Miscellaneous Publications by
addressing the Exchange Librarian, The University of Kansas Li-
brary, Lawrence, Kansas 66045. Individuals may purchase separate
numbers of all series. Prices may be obtained upon request addressed
to Publications Secretary, Museum of Natural History, The Univer-
sity of Kansas, Lawrence, Kansas 66045.

Editor: E. O. Wiley
Managing Editor: Joseph T. Collins

PRINTED BY

UNIVERSITY OF KANSAS PRINTING SERVICE

LAWRENCE, KANSAS

7 07 0 \\h

3 2044 093 361 699

DATE DUE

APD 1 ^ ...Qiftfiji

HM<-JL o £(Ju4

DEMCO, INC. 38-2931