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 voliune 20 in 1971. Shorter research papers formerly pub-
lished in the above series are now published as Occasional
Papers, Museinn of Natural History. The Miscellaneous Publica-
tions, 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 extra-mural specialists; final acceptance is at the discretion
of the publications committee.

Institutional libraries interested in exchanging publications
may obtain the Occasional Papers and Miscellaneous Publica-
tions, by addressing the Exchange Librarian, The University of
Kansas Library, Lawrence, Kansas 66045. Individuals may pur-
chase separate numbers of all series. Prices may be obtained
upon request addressed to Publications Secretary, Museum of
Nattual Historv, The Universitv of Kansas, Lawrence, Kansas
66045.

Editor: Linda Trueb

PRINTED BY

UNIVERSITY OF KANSAS PRINTING SERVICE

LAWRENCE, KANSAS

OCCASIONAL PAPERS

of the

MUSEUM OF NATURAL HISTORY
The University of Kansas
Lawrence, Kansas

NUMBER 61, PAGES 1-24 NOVEMBER 19, 1976

THE SPECIES GROUPS OF THE SOUTH AMERICAN
FROGS OF THE GENUS ELEUTHERODACTYLUS

(LEPTODACTYLIDAE)coMP zooi^

L.IBRARY

John D. Lynch^ '^'''

HARVARO

The large number of names applied to frogs ot'tKe'genuS Eleu-
therodactijhis probably has contributed to a certain reluctance of
herpetologists to become involved with the systematics of the genus.
As presently defined (Lynch, 1971), the genus occurs throughout
the West Indies, in Central America from northern Mexico to the
Darien of Panama, and over the South American continent as far
south as northern Argentina.

The West Indian species ( 131 according to Schwartz and
Thomas, 1975; Schwartz, 1976) have been arranged into 4 to 6
species groups by Dunn (1925, 1926), Schwartz (1958), and Slii-eve
and Wilhams (1963). Although the limits of some of the species
groups are open to question ( Shreve and Williams, 1963; Schwartz,
1966), the arrangements have improved the situation regarding the
systematics of the West Indian species.

The Mexican and Middle American herpetofauna includes ap-
proximately 70 species not found in the West Indies or South Amer-
ica (Lynch, 1970; Savage, 1973, 1975; Smith and Taylor, 1948;
Stuart, 1963). Smith and Taylor's (1948) species-group arrange-
ment was modified by Firschein (1951) and Lynch (1965, 1968a,
1970); at present six species groups are recognized. Savage (1973)
recognized 9 groups for the Costa Rican fauna ( two of those groups
also are represented in Mexico ) .

^ Associate Professor of Zoology, School of Life Sciences, The University of
Nebraska, Lincoln, Nebraska 68588, U.S.A.; and Associate in Herpetology, The
University of Kansas Museum of Natural History.

2 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

The South American herpetofauna includes 141 recognized spe-
cies (see hsts in the accounts of the species groups). In spite of
the diversity of species, few authors have attempted to recognize
species groups. Lynch (196Sb) proposed a prehminary grouping
of the southeastern Brasihan species. Cochran and Coin (1970)
recognized four groups of the genus in Colombia. Lynch (1975b),
in anticipation of this paper, proposed recognition of two species
groups for the broad-headed Eleuthewdactijhis of Centi-al and
South America.

Many of the groups heretofore recognized in Middle and South
America are Artenkreisen largely defined on the basis of content.
The "groups" recognized in Mexico and Costa Rica are assemblages
of quite similar species, but the boundaries between groups have
not been defined. My attempts to define those boundaries have
forced me to adopt a generally more conservative stance toward
species groups; thus I combine the 9 groups Savage (1973) recog-
nized in Costa Rica into three groups. My current view of the
Mexican Eleutherodactijlus prompts me to combine most of the
mexicamis, pygmaetis, and rhodopis groups as a single group, the
rhodopis group. "Synonymies" for several of the species-groups
arrangements for Eleutherodactylus are summarized in Table 1.
The following account is restricted largely to the South American
species.

Acknowledgments. — For loan of type-specimens or provision of
working space at their respective institutions, I am grateful to J.
Bolhke, A. Capart, the late D. M. Cochran, G. F. deWitte, W. E.
Duellman, J. Eiselt, A. G. C. Grandison, J. Guibe, B. Hansson, A.
Leviton, E. Malnate, C. W. Myers, V. Parenti, the late J. A. Peters,
G. Vestegren, C. F. Walker, E. E. Williams, and R. G. Zweifel.
Travel to museums was supported by a Graduate School Honors
Fellowship from The University of Kansas, The University of Ne-
braska Research Council, and the Visiting Scientists program of the
Smithsonian Institution. I have profited from discussions with
Werner C. A. Bokermann, William E. Duellman, Hobart M. Smith,
Edward H. Taylor, and Charles F. Walker. Jaime Pefaur provided
the Spanish summary.

MATERIALS AND METHODS

For this paper, data were obtained for each nominate species by
one or more of the following: examination of the type-specimen
(holotype or syntypes), data derived only from the original de-
scription, study of referred specimens, or data extracted from re-
descriptions of the type or other specimens. Original descriptions
were studied for every species. I depended on redescriptions of
the type-specimens for only two names (affinis and f rater); in each

SPECIES GROUPS OF ELEUTHERODACTYLUS 3

case, the ledescription was tliat provided by Cochran and Coin
( 1970). If the holotype or syntypes were examined, that name bears
an asterisk in the hsts of species given for each species group.

The holotypes or syntypes of 116 of the 169 names and paratypes
of 9 other names have been examined personally. Of the remaining
44 names, I am aware of only 8 for which the types are lost (coniu-
tii.s, (lici(Ic)iiatus, g,aldi, henselii, lacrimosus, napaeiis, piiJcJtii<i,uhis,
and rubicttnchis) . Two names not considered here {Hylodes f^rav-
enhorsti Fitzingeri, 1861, and EJeutherodactylus conspiciUattts
guaijanensis Rivero, 1968) are nomena mida based on South Ameri-
can material. One other name long misapplied to South American
frogs is HyJa gri.seo Hallowell. The data provided bv Hallowell
(1860) and Cope (1862a, 1862b, 1863) clearly indicate that the
frog is a member of an assortment of species in Lower Central
America and Chocoan South America. Savage (1974) stabilized
the name by designating a common neotype for Hylodes fitzingeri
O. Schmidt and Hyla grisea Hallowell.

The 169 ti-i\'ial names are listed in Gorham's (1966) checklist as
EleutJterodoctyJus with the follo\\'ing exceptions (listed in other
genera): in Basanitia {hoJhodactyla, gehrtii, loctea, nigroventris),
in Lepfodactyhis (pumdio), in Fhrynanodus {nanus), in Syr-
rhophus (areolatns, cJialceus, coendeus, festae) and in Trachy-
phrynus (niyersi). The following are names either not included in
Gorham's (1966) checklist or named since 1963: Craugasior ptd-
chrigulus Cope, 1862, EJeutherodactylus anderssoni Lynch, 1968,
E. hiUneatus Bokermann, 1974, E. hockermanni Donoso-Barros,
1970 (here emended to E. hokennanni) , E. hoconoeivsis Ri\'ero and
Mayorga, 1973, E. calcarulatus Lynch, 1976, E. celafor Lynch, 1976,
E. cerastes Lynch, 1975, E. chloronotus Lynch, 1970, E. crenunguis
Lynch, 1976, E. crepitans Bokermann, 1965, E. croceoinguinis
Lynch, 1968, E. elassodiscus Lynch, 1973, E. gualteri Lutz, 1974, E.
Janthanites Lynch, 1975, E. latidiscus tamsitti Cochran and Coin,
1970, E. lehmanvaJenciae Thornton, 1965, E. leucopus Lynch, 1975,
E. luteolateralis Lynch, 1976, E. martiae Lynch, 1974, E. necerus
Lynch, 1975, E. nicefori Cochran and Coin, 1970, E. nyctophylax
Lynch, 1976, E. octavioi Bokermann, 1965, E. orcesi Lynch, 1972,
E. orphnolaimus Lynch, 1970, E. paulodutrai Bokermann, 1974, E.
pauhdus Lynch, 1974, E. parvillus Lynch, 1976, E. pugnax Lynch,
1973, E. puhinatus Ri\'ero, 1968, E. pusiUus Bokermann, 1967, E.
quaquaversus Lynch, 1974, E. surdus cabreroi Cochran and Coin,
1970, E. thectopternus Lynch, 1975, E. thymaJopsoides Lynch, 1976,
E. thymeJensis Lynch, 1972, E. trepidotus Lynch, 1968, E. variabilis
Lynch, 1968, E. vinhai Bokermann, 1974, E. ualkeri Lynch, 1974,
Eupsophus ginesi Rivero, 1964, Hyla chimboe Fowler, 1913, and
Pseudohyla nigrogrisea Andersson, 1945.

4 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

THE INFRAGENERIC UNITS AND SPECIES GROUPS
History of the Problem

In treating the comparatively small Colombian Eleiitherodacty-
liis fauna, Cochran and Coin (1970) proposed four species groups.
The groups were defined as follows:

"Group I: long-legged, smooth-bellied frogs with smooth
heads, fairly large in size; no bony ridges on head.

Group II: Granular-bellied, shorter-legged frogs usually hav-
ing a frontoparietal depression; bodies moderately long;
no bony ridges on head.

Group III: Small frogs, usually with U- or )( -shaped gular
folds in the male; often with pointed toes; no bony ridges
on head.

Group IV: Medium-sized to large frogs, usually with bony
ridges on head or some ossification of head skin and skull."

This arrangement must be viewed as the first attempt to define
species groups for the South American Eleuthewdactylus even
though the study dealt only with Colombian frogs. Cochran was
the author of the leptodactylid section of the "Frogs of Colombia"
and had considerable experience with EleutJierodactyhis from all
parts of the range of the genus. A number of authors have men-
tioned groups, listed included taxa, and in some cases defined the
group(s) involved. However, these attempts dealt with restricted
geographic areas and faunae (e.g., Lynch, 1968b).

The taxonomy of eleutherodactyline frogs is difficult owing to
the diversity of the group and the often subtle differences evident
among preserved examples of supposed difi^erent species. Many
differences apparent in living or freshly preserved material are lost
after relatively short periods in alcohol. The problem of species
identification is not mitigated by the plethora of species group
designations used by authors (including me) dealing with the
genus. A recent opportunity to examine the holotypes of nearly
two-thirds of the nominate South American species and the results
of those studies prompts this report.

Species groups have been defined on a diverse set of charac-
teristics. Most groups are in fact defined (or diagnosed) on single
traits, e.g., the biporcatus-cornutiis group on the basis of cranial
crests, or defined on the basis of content and a general homogeneity
of character states ( not specified ) . In my own papers ( 1968-74 )
I have used the following group names with too little concern for
group definition: biporcatus, hinotatus, conuitus, cuiiipes, fitzing-
eri, galdi, gtientheii, lacteiis, parvus, rugidosus\ siirdus, and iini-
strigatus.

SPECIES CROUPS OF ELEUTHERODACTYLUS 5

As a part of my studies on the Ecuadorian species of EleutJiero-
dactylus, I have examined the type specimens of 116 of the 169
names appHed to frogs from South America, as well as several
thousand additional specimens. Reviews of the literature, princi-
pally original descriptions or translations of original descriptions,
have revealed a fair number of erroneous statements about various
taxa; those statements have resulted in some erroneous conclusions
being drawn. My study of these frogs and the literature has con-
vinced me that it is desirable to define the several species groups
which I have recognized for the EleutlicrodacfijJus found in main-
land South America. I pointed out (Lynch, 1971) the possibility
of two major divisions within the genus defined on osteological
grounds — an Alpha division for the species of the West Indies
(exclusive of the inoptatiis group of Hispaniola) and the genera
SmintliiJhis of Cuba and Syrrhophus and Tomodactyhis of Mexico
and Guatemala, and the Beta division for the species in Mexico,
Central America, South America, and the Hispaniolan inoptatiis
group. I still have not acquired enough data to propose formal
recognition of these two divisions.

The most recent synthesis of the species of Eleutherodactylus
is that of Boulenger ( 1882) who recognized 45 species of the genus
over its entire distributional area. At present, more than 400 names
have been applied to the group. The genus has been defined
(Lynch, 1971), although some question remains concerning the
advisability of generic separation of Synlwplius and Tomodactyhis,
but the major question involving the genus is the definition and
delineation of the units within the genus. For convenience, the
genus is divided into its three geographic subunits: the West
Indian section (Florida to Trinidad), the xVIiddle American section
(Mexico to the Canal Zone in Panama), and the South American
section ( Panama east of the isthmus and mainland South America ) .

One hundred sixty-nine names have been proposed based on ma-
terial known or thought to have originated from w ithin the South
American section. Many of these names have been synonyms in
the zoological literature from essentially the time of their proposal
and never have been used in the combination with Eleutherodac-
tylus. Thus the spelling of names in the following account reflects
their original orthography and is not necessarily in agreement in
gender with Eleutherodactylus.

The Infrageneric Units

In comparing the data from original descriptions and from
examinations of the type-specimens, I noted a general concordance
of the character states of two characteristics among most of the
South American species of the genus. Most species have the first
finger shorter than the second and the skin of the venter aerolate.

6 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

The characteristics and their character states are the relative lengths
of the two inner fingers ( first finger longer than second / first finger
shorter than second) and the texture of the skin of the venter
(smooth or feebly granular / coarsely areolate). The combinations
of these character states are as follows:

lA: first finger longer than second; skin of ^'enter smooth or

feebly granular.
IB: first finger longer than second; skin of venter coarsely

areolate.
2A: first finger shorter than second; skin of venter smooth.
2B: first finger shorter than second; skin of venter coarsely

areolate.

The species can be arranged into groups using other charac-
teristics but the groups thus generated appear to be more artificial
than the groupings resulting from these two characteristics. One
does find a homogeneity of several character states among the
groups generated by simultaneous consideration of fingers lengths
and the texture of the skin of the venter. The characteristics utilized
for West Indian species by Dunn (1925, 1926), Schwartz (1958,
1964), and Shreve and Williams (1963) do not result in homoge-
neous groups among the South American species of the genus, nor
is the homogeneity amongst West Indian species high. Among the
West Indian species of the genus there is considerable variation
in the size and shape of the prevomerine dentigerous processes.
The variation in this characteristic among South American (and
Centi-al American) species of the genus is insignificant.

These four units ( lA, IB, 2A, and 2B) are subdi\'ided into sev-
eral species groups. With the exception of a single taxon included
in 2B, all species are evidently members of what I have termed
the Beta division of EleutJwrodactylns; the osteological character-
istics used in defining the Alpha and Beta divisions have not been
checked for each taxon but are used in the species group definitions
to distinguish externally identical species groups (e.g., aurictilatus
and iinistrig,atiis groups).

The four "infrageneric units" should not be viewed as more than
convenient subdivisions. I do not consider the species groups in-
cluded in infrageneric unit IB to be more closely related to one
another than either is to some other species group here listed in
another infrageneric unit. In the accounts given below the ten
species groups are listed in alphabetical order within the four
infrageneric units.

The Species Groups
Infrageneric Unit lA

The 47 taxa included in this contingency include diverse types.

SPECIES GROUPS OF ELEUTHERODACTYLUS 7

Ten of the 47 are considered synonyms of other taxa. Lynch ( 1975a)
considered brachypodius, ileamazonicus, and rosmeliims synonyms
of vilarsi. Cochran and Coin ( 1970 ) considered biiergeri a synonym
of ic-nignim. Lynch (1969) considered carrioni a synonym of
lytuani. Cochran (1955) considered divisa a synonym of guentheri.
Lynch (1975b) considered napaeus a synonym of corntitus. Lynch
(1975a) considered peruvianm a synonym of conspiciUatiis. Savage
(1974) made pulchrigidus an objective synonym of jxtzingeri by
designating a common neotype. Peters (1872) considered rugido-
sus Peters a synonym of binotatus.

I consider sexeral of the 37 currently recognized taxa to be syno-
nyms but those documentations will be dealt with separatelv in
appropriate papers. The 37 recognized taxa are assigned here to
four species groups. The name of each group is taken from the
oldest species of the group. The members of all four groups share
the following features: skin of venter smooth; first finger longer
than second; tympanum prominent, annulus not concealed; pre-
vomerine odontophores prominent, triangular in shape. Those spe-
cies for which skeletons are a\ailable ha\'e large nasal bones in
median contact, frontoparietals not synosteotically united with the
prootics, and the median pterygoid ramus overlapping the para-
sphenoid ala.

The binotatus Group

Definition. — Traits of unit lA and the following: head narrow
(HW/SVL < 0.43); no cranial crests; ungual flap indented (Fig.
1); digits bearing pads or not {octavioi); toes lacking webs.

Content. — binotatus (nigidostis), gualteri, guentlwri^ (divi.m),
namitus, octavioi, and pliciferus.'^ Data are inadequate to definitely
assign hemeJii and hoehnei to the group. The holotype of hemelii
is now lost (C. Peters, in litt.).

Distribution. — Coastal Brasil from Pernambuco to Rio Grande
do Sul (Fig. 2).

Remarks. — The six species of the group form 3 sets. E. octavioi
differs markedly from the other five in lacking pads on the digits.
E. guentheri and E. nasiitu.s are very similar (Cochran, 1955) with
apically rounded discs; the first finger, although longer than the
second, is not markedly longer. E. binotatus and E. pliciferus have
pointed discs and very long first finger. Both have numerous dorsal
ridges.

The biporcatus Group

Definition. — Traits of unit lA and the following: head broad
(HW/SVL 0.44-0.58); cranial crests present; ungual flap not in-
dented; digits bearing narrow pads; toes lacking webs.

Content. — Lynch (1975b) recognized five species: biporcatus.

OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

Fig. 1. Dorsal views of digital pads. A) E. guentlieri, 3id finger right
hand, KU 92816. B) E. nigrovcntris, 4th toe right foot, KU 92734. C) E.
octavioi, 4th toe left foot, KU 92827. D) E. parvus, 4th toe left foot, KU 92831.
E) E. fenestrattis, 3rd finger left hand, WACB 9841. Lines equal 1 mm.
Abbreviations: KU — The University of Kansas Museiun of Natural History;
WACB — Werner C. A. Bokermann (private collection).

hufonijormis,^ cerastes,^ cornutus (napaeiis), and necerus.^

Distribution. — Lowlands of Central America from Honduras to
eastern Panama; Pacific lowlands and Andean \'ersant in Colombia
and Ecuador; Amazonian Andean versant in Ecuador (Fig. 2).

Re77iarks. — E. anomalus (Boulenger) is assigned to the fitzin<i,eri
group (see below) but approaches the hiporcatus group in that the
head is moderately broad (HW/SVL 0.42-0.48). E. anomalus lacks
cranial crests and has fully webbed toes but is otherwise similar
to E. hufoniformis.

SPECIES GROUPS OF ELEUTHERODACTYLUS

Fig. 2. Distributions of the species groiips of Eleiithewdactyhis in South
America. Open circles and squares denote t\pe-loca]ities, whereas triangles
denote \'ague t\pe-loca!ities. The distribution of the group is indicated with
hatcliing. A) E. auriculaius group (vertical hatching); E. hinotatus group
(horizontal hatching); £. discoidalis group (right oblique hatching); E. sul-
catus group (left oblique hatching). B) E. fitzingeri group. C) E. hiporcatus
group (horizontal hatching); E. parvus group (vertical hatching); E. ramagii
group (obhque hatching). D) E. lactcus group (vertical hatching); E. uiii-
strigatus group (oblique hatching).

The discoidalis Group

Definition. — Traits of unit L\ and the following: head narrow;
no cranial crests; ungual flap not indented; digits lacking pads;
toes lacking webs.

Content. — cruralis,^ discoidalis,'^ eJassodiscus,'* gronidosus,'^ and
nigrovittatus.^

Distribution. — Amazonian Andean versant in Bolivia and south-
ern Peru (cnimlis and granulosus) and northern Ecuador (elasso-
discus and nigrovittatus); Andean foothills in northern Argentina
{discoidalis).

Remarks. — E. elassodiscus and E. nigrovittatus have pointed
discs (thus resembling E. hinotatus and E. pliciferus of the hinota-
tus group), whereas E. cruralis, E. discoidalis, and E. granulosus

10 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

have apically rounded discs. These three species resemble E. oc-
tavioi {hinotatus group) in having large outer metatarsal tubercles.
I consider this group annectant between the hinotatus and jitzingeri
groups.

The fitzingeri Group

Definition. — Traits of unit lA and the following: head narrow
(HW/SVL 0.32-0.43 in most; 0.36-0.48 in some species, see re-
marks); no cranial crests; ungual flap not indented; digits bearing
pads; toes lacking webs or moderately to fully webbed.

Content (South American taxa only). — achatinus,** anomahis,'*
hiUneatus, bisignatiis\'^ hrederi,'^ conspiciUatiis'' {pertwianiis'^),
crenunguis,'* crepitans, fenestmtus,^ fitzingeri (pidclirigiihis), het-
erodactijlum, insignitiis,'* lanthanites,'* hngirostris,^ hjnuini'^ {car-
riotn"^), pagmae,'^ ranifonnis,'^ terraehoUvaris,^ thectopternus,"^
vilarsi^ {JyrocJu/podiiis," ileamazonicus,'^ rosmelimis^ ) , and tc-
nigruni {hiiergeri'^).

Distribution. — The forested Amazon basin as far south as Mato
Grosso and Bolivia; the Andean slopes in Colombia, Ecuador, and
extreme northern Peru; the Pacific lowlands of Colombia and Ecua-
dor; the Coastal Range of Venezuela. The group occurs in Central
America from Panama west and north into Mexico (Fig. 2).

Remarks. — I tentatixely include the Central American frogs
placed in the goUmeri group in this group as well as the frogs
Savage (1975) considered to be members of the rugidosus group.
The several groups of Mexican Eleutlierodactylus recognized by
Smith and Taylor (1948) include some representatives of the
fitzingeri group as used here ( viz., laticeps group, in part; rhodopis
group, in part; rugidosus group). The Mexican and Guatemalan
species pre\'iously assigned to the laticeps and rhodopis groups but
here considered fitzingeri group frogs include E. anzuetoi Stuart,
E. laticeps (Dumeril), E. hneatus (Brocchi), E. macdougaUi Tay-
lor, E. rostraJis (Werner), E. stantoni Schmidt, and E. werleri
Lynch and Fritts. These seven nominate taxa are closely allied to
the Middle American E. goUmeri (Peters), E. mimus Taylor, and
E. nohjei Barbour and Dunn.

E. anonudus is perhaps the most di\ergent member of the group
in having no pads on the fingers with \'ery poorly developed discs
and having nearly fully webbed feet. E. fitzingeri, E. longirostris,
and E. raniformis have appreciable webbing of the toes as well but
presently I am not willing to separate these four species from the
other 17 on the basis of toe webbing. I am aware of one or two
undescribed taxa from the upper Amazon basin (Colombia, Ecua-
dor, and Peru) having nearly as great but obser\ably less webbing.
Additionally, Savage's (1975) studies of Central American species

SPECIES GROUPS OF ELEUTHERODACTYLUS 11

suggest that variation in toe webbing is not a desirable base for
species group separations.

Bokermann's (1974) E. hilincatus is placed here only tentatively.
On the basis of the description and illustrations, the frog is similar
to a frog now under study by M. S. Iloogmoed; if E. hilineatii.s is
congeneric with Hoognioeds frog, it is not a member of the genus
Eleutherodactijhis.

Infrageneric Unit IB

The se\'en nominate species of this contingency readily fall into
three groups. Lynch (1975b) considered koki and macrocephalus
synonyms of siilcatus.

Cochran (1956) described E. J)eebei from Kartabo, British
Guiana (= Guyana) noting that the frog was similar to the Haitian
E. inoptotus. I re-examined the unique holotype (USNM 129526)
and found a number of discrepancies with the original description.
The holotype is an adult female containing o\arian eggs (not a
male with vocal slits as reported by Cochran). Cochran cited the
presence of dermal appendages on the upper eyelid, elbow, knee,
and heel and illustrated a tarsal fold. Although dermal horns are
present on the upper eyelid no trace of such structures are found
on the elbow, knee, or heel; likewise no tarsal fold is present.
Lastly, the illustration of the holotype shows vertical pupils; the
pupils of the holotype are horizontal.

The holotype of E. heehei is not distinguishable from equal-
sized adult females of E. inopfotus; in the absence of differences,
E. heehei Cochran is referred to the synonymy of E. itwpfatus
(Barbour). I seriously doubt that the holotype originated from
Guyana in spite of the notes of William Beebe (quoted by Cochran,
1956:12). The frogs of the E. inoptatus group (E. chlorophenax,
E. hypostenor, E. inoptatus, E. nortoni, and E. ruthae) are restricted
to Hispaniola (Schwartz, 1965, 1976) and although the Kartabo
region has not been so extensixely collected as some areas in South
America it seems unlikely that so large a frog would continue to
escape notice.

The ramagii Group

Definition. — Skin of venter areolate; first finger longer than sec-
ond; tympanum prominent, annulus not concealed; prevomerine
odontophores small, oblique; head narrow; no cranial crests; ungual
flap not indented; digits bearing large pads; toes lacking webbing.
No osteological data available.

Content. — paulodutrai, ramagii."

Distrihution. — Extreme eastern Brasil (Bahia, Pernambuco).

Remarks. — These two species differ from all other South Amer-
ican Eleutlierodactijlus in ha\ing a long first finger and large pads

12 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

Fig. 3. Palmar \ie\vs of hands and digits of Elcidherodactijlus. A) E.
ranmgii, KU 137752. B) E. gularis, LACM 73239. C) E. altamazonicus,
3rd finger left hand, KU 148772. Line equals 2 mm. Ahhreviations: KU —
The University of Kansas Museum of Natural History; LACM — Los Angeles
County Museum.

and discs on the innermost digits (Fig. 3). These traits in combina-
tion with the insignificant pre\'omerine odontophores are suggestive
of an affinity with West Indian Eleutlwrodactylus but until osteo-
logical data are avaihible I am reluctant to assert a West Indian
affinity for the group.

The sulcatus Group

Definition. — Skin of venter coarsely areolate; first finger longer
than second; tympanum prominent, annulus not concealed; pre-
vomerine odontophores prominent, arch-like; head broad (HW/
SVL 0.44-0.55); cranial crests present; ungual flap absent or not
indented; fingers lacking pads; toes lacking webbing; nasal bones
large, in median contact; frontoparietal not fused to prootic; median
ramus of pterygoid overlapping parasphenoid ala.

Content. — Lynch (1975b) recognized two species: matissi,'^
sulcatus'* (/voA:/,* macrocep]\alus'* ) .

Disirihution. — Coastal Range of Venezuela and upper Amazon
basin in western Brasil and eastern Ecuador and Peru (Fig. 2).

Remarks. — The frogs of the .sulcatus group are quite similar to
those of the hiporcotus group but, perhaps, are most closely related
to the eleutherodactyline genus Anil)Iy))J)n/niis. In spite of the
similarities of the biporcatus and sulcatus groups, I am reluctant

SPECIES GROUPS OF ELEUTHERODACTYLUS 13

to combine the two groups or to advocate generic separation from
Eleutherodactijlus.

Infrageneric Unit 2A

Nine nominate species are included in this contingency and are
separable into two species groups. One of the included species,
E. racenm Rivero, is probably incorrectly referred to Unit 2A but
until the holotype is re-examined and the inadequate original de-
scription verified, the species is included here; it is not assigned to
either the lacteus group or parvus group.

Of the eight other names, Cochran (1955) considered fieliiiii
a synonym of hoJhodaciijla. Lynch (1968b) assigned nanus and
Heyer (1972) assigned pumilio to the synonymy of parvus.

In addition to having the skin of the venter smooth and the first
finger shorter than the second these five species have indented
ungual flaps (as in the E. Innoiatus group). Those species for
which skeletal data are available have large nasal bones, fronto-
parietals not fused to the prootics, and the median ramus of the
pterygoid overlapping the parasphenoid ala.

The lacteus Group

Definition. — Traits of unit 2A (except that £. nigroventris has
an areolate venter) and the following: digits bearing large pads
and apically rounded discs; tympanum prominent, annulus not con-
cealed; prevomerine odontophores low% oblique; nasal bones in
median contact.

Content. — bolbodactyl-a (gelirtii), lacteus, nigroventris (Iwlti"),
venancioi.

Distribution. — Mountains of southeastern Brasil (Fig. 2).

Remarks. — E. nigroventris (Lutz) [and its synonym, E. iinistri-
gatus holti Cochran {fide Bokermann, 1966)] differs from the other
species of the group in having areolate skin on the ^'enter. In spite
of this difference I consider nigroventris a member of the lacteus
group because of the agreement in digit morphology, prevomerine
dentition, and the visibility of the tympanum.

The parvus Group

Definition. — Traits of unit 2A and the following: digits bearing
narrow pads and apically pointed discs; tympanum indistinct or
concealed, annulus not or only partially xisible; prevomerine odon-
tophores prominent, triangular in outline. Nasal bones narrowly
separated.

Content. — parvus (nanus,'^ pumilio'^), pusiUus.

Distribution. — Mountains of southeastern Brasil (Fig. 2).

Remarks. — These frogs once were confused with E. rhodopis

14 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

( Miranda-Ribeiro, 1926); the similarity largely stems from the
pointed discs in each group. The pointed discs of the frogs of the
E. parvus group are also shared with E. nifirovittattis (an Ecua-
dorian frog of the cliscoidalis group) and E. stenodiscus (a Vene-
zuelan frog of the unistri^atus group). Neither of the last two spe-
cies has indented ungual flaps.

The coincidence of notched or indented ungual flaps in all three
groups found in southeastern Brasil is suggestive that they not be
separated. All three also share smooth skin on the venter and the
osteological traits cited above for unit 2A but vary in relative lengths
of the first two fingers, prominence of the pre\omerine odonto-
phores, and prominence of the tympanum. I place little significance
on the last trait but consider finger lengths and prominence of
prevomerine odontophores relatively important features.

Infrageneric Unit 2B

This is the largest of the four contingencies, including 104
nominate species. The 104 taxa are here assigned to two groups:
uriclii is placed in the auriculatus group and the other 103 in the
unisfrigatiis group. Eleven of the 104 are considered synonyms of
other names. Lynch (1974) considered onderssoni a synonym of
ockendeni and brevicnis a synonym of oltamazonicus. Lynch and
Schwartz (1972) considered Inijonius a synonym of diodematiis.
Barbour and Noble (1920) consider equatorialis a synonym of tiui-
strigatus. Dunn (1944) considered fasslianus a synonym of bogo-
tensis. Lynch (1974) considered Hylodes festae a synonym of gold i
and pointed out ( 1975c ) that PaJiidicola festae was identical to
trepidotus but as a secondary homonym of Hylodes festae was not
available. Dunn (1944) considered fiiJiiinanni a synonym of ele-
gans. Savage (1968) considered lelDuanvalenciae a synonym of
mom. Lynch (1969) considered margaritifer a synonym of gald'i.

Dunn (1942) considered gularis and huigrae synonyms of dia-
stema but both differ from diastema in ha\ing papillae at the digit
tips. Although I here remoxe them from the unjustified synonymy
u'ith diastema I consider huigrae a svnonvm of gidaris.

The remaining 93 names are here treated as valid although some
surely are not. My use of chimhoe, grandoculis, melini, ventrivit-
tatus, and whymperi is tentative; these names will be placed in the
synonymies of other species in appropriate papers.

The 93 species here recognized include one (tiricJn) placed in
the West Indian auricidatus group by Schwartz (1967) as well as
the frogs Cochran and Coin (1970) placed in groups II and III.
Their group III is equivalent to a sometimes recognized Central
American species group, the diastema group (Table 1).

In addition to haxing the skin of the venter coarsely areolate

SPECIES GROUPS OF ELEUTHERODACTYLUS 15

and the first finger shorter tlian the seeond, these frogs have narrow
heads (HW/SVL 0.30-0.42) and the ungual flap is not indented.
Compared to the other infrageneric units, 2B is heterogeneous. The
tympanum is prominent (annuhis not concealed) in many species
but an equal number have partially or entirely concealed tympana
and in three species (anotis, ):ttii.nox, and surdus) the tympanic
annulus, cavum tympanicum, and plectrum are absent. The pre-
vomerine odontophores range from prominent, triangular-shaped
structures (e.g., erytliropleurus, quoqtiaversus) to obscure, thin,
oblique structures (e.g., curtipes, nigrogriseus, ockendeni, unistri-
gatiis). In coertileus and orcesi odontophores are absent and those
of oreolattis are nearlv indistinguishable. Cranial crests usuallv are
lacking in these frogs but are well developed in huckleiji, curtipes,
deviUei, galdi, and surdtis and prominent, although less well devel-
oped in chloronotus, glandidosus, and vertebralis. Many species
have very slight upturning of the lateral edges of the frontoparietals
which might be termed incipient or poorly-developed cranial crests.

The variation in digit morphology ranges from those lacking
pads (e.g., ginesi, nujersi, and trepidotus) to those having greatly
dilated pads (e.g., diadematiis, kitidiscus, and ruhicundus) with a
complete array of intermediates. In those ha\ ing digital pads, the
pads on the inner fingers are smaller than those of the outer fingers.
Two species (areoIatus\ gularis) have triangular discs (apex distal)
with prominent papillae abo\'e the disc (Fig. 3). In most species
the discs are much broader than long (Fig. 3) but in the species
lacking pads as well as those having triangular discs the disc length
is equal to or exceeds disc width. However, disc shape (i.e., length
vs width) is a continuous variable.

Most frogs of this group lack toe webbing but many have lateral
fringes on the toes. One (piignax) has appreciable webbing but the
extent of the web is much less than that seen in certain species of
the fitzingeri group.

I have examined skeletons of only 17 of the named species in
unit 2B {acuminatus, altamazonicus, areolatits, bogotemis, cajamar-
censis, chloronotus, croceoinguinis, curtipes, deviUei, galdi, Jacrimo-
sus, nicefori, ockendeni, riveti, stenodiscus, surdus, thymelensis,
trepidotus, unistrigatus, and variaJ)ilis) . The frontoparietals are
fused to the prootics in curtipes and tJujmelensis and may be fused
in riveti and trepidotus but the specimens examined do not permit
definite characterization. The median ramus of the pterygoid over-
laps the parasphenoid ala in all except acuminatus and cajamarcen-
sis. The nasals are large and in median contact in all species al-
though one might describe the nasals of hogotensis, riveti, thymelen-
sis, unitrigatus, and variabilis as narrowly separated. The nasals of
these five species are less separated than are those of E. parvus
{parvus group, unit 2A).

16 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

Lynch ( 1971 ) suggested dividing Eleutlierodactylus into Alpha
and Beta divisions on the basis of three osteological features. The
Alpha division was characterized in having the frontoparietals and
prootics fused, the median pterygoid ramus not o\erlapping the
parasphenoid ala, and the prevomers small and separated medially.
At least two of the 17 species in unit 2B have frontoparietal-prootic
fusions but also have pterygoid-parasphenoid contact. Two others
present the inverse character states. In view of the variability in
osteological characteristics, separation of the South American and
West Indian species of unit 2B as distinct species groups is difficult
to defend. That skeletal data of only 17 South American and 16
Antillean species currently are available also renders any such
separation tenuous. Osteological preparations are not available for
E. urichi (putati\'e E. auriculatus group member) or for most
species in unit 2B from Venezuela. However, until more species of
unit 2B arc known osteologically, I propose to recognize Antillean
and mainland groups.

The auriculatus Group

Definition. — skin of venter coarsely areolate; first finger shorter
than second; heads narrow; ungual flap not indented; frontoparietal
bone fused to prootic; median ramus of pterygoid not overlapping
parasphenoid ala; large nasal bones in median contact or narrowly
separated; prevomerine odontophores narrow (patch-like, not
arched ) .

Content. — 39 species (see Schwartz, 1969, 1973); urichi occurs
on Trinidad and the adjacent mainland (Venezuela, Guyana).

Distribution. — Throughout the Antilles possibly excepting Ja-
maica (Schwartz, 1969).

The unistrigatus Group

Definition. — skin of venter coarsely areolate; first finger shorter
than second; heads narrow; ungual flap not indented; frontoparietal
bone rarely fused to prootic; median ramus of pterygoid usual!)'
overlapping parasphenoid ala; large nasal bones in median contact
or narrowly separated; prevomerine odontophores narrow (not
arch-like).

Content. — 92 species having South American type-localities;
the type-locality for moro is extra-limital: acnminatus,'^ ajfinis,
altamazoniciis'* (brevicrus), anonyinns, anotis,'* oppendicula-
tii.s;'^ areohitus,'^ Ijicumulus,'* bofiotensi.s (fasslianus), boconoen-
sis, bokermanni, briceni,"* buckleyi," cabrerai,'^ cajamarcen.sis,'^
ccdcaratus,'^ calcarulatus,** carmelitae,'^ carvaUwi, relator,^ chal-
ceus, chimboe,'^ cJiloronotus,'* coeruleus,'^ croceoin<ininis,'^ criici-
fer,'^ curtipes,'* delicatus,*' devillei,'* diadematus (btifonius'^).

SPECIES GROUPS OF ELEUTHERODACTYLUS 17

elegans (fuhiDianni), c'n/flirojjlciiiiis," f rater, <guldi {festae,'^ mar-
garitifer" ) , ginesi, ghDuliilosus," graiulociilis, giilaris'* (huigrae'' ),
inguinaUs,*' lacrimosus, lancinii, htidiscus,'* leJimanni, leucoptis,^
luteolate rails, *" marmoratus,^ maHiae,'^ megalops,'^ melini, moro
(lelimanvalenciae'* ) , myersi,'* nicefori,'^ nigrogriseus,'* nijctophy-
lax,^ ockcndeni'^ {anderssoni^ ), orcesi,'* ornatissimus, orocostalis,
orphnolainuis,'* palmeru'^ parviUus,'^ pastazensis,^ pauhdus,*' platy-
dactylus,^ pseudoacuininatus,'^ pugnax,"^ pulvinaius,'^ quaquaver-
sus,^ reticulatiis,'^ riveti,'^ roseus,^ rozei, ruhicundus, sanctaemar-
tae,^ stenodiscus,^ sidjsigillatus,** surdiis,^ taeniatus,^ iamsitti,
thymalopsoides,'^ tJiymeletisis,'* trachyhlepharis,'* trepidoius^ (Pahi-
dicola festae'^), turumiquirensis,'* itnistrigatus'* {equatoriolis'^ },
variabilis,'^ ventrimannoratiis,'* ventrivittatiis; vertebraJis,'^ vinhai,
icalkeri,^ whymperi,'^ and icilliamsi.

Distribution. — Throughout forested Colombia and Ecuador as
well as into the high altitude grasslands; along the western edge of
the Amazon basin in Peru and adjacent Bolivia; the northern
Amazon basin in Brasil and Venezuela; Guy anas; Merida Andes
and the Coastal Range of Venezuela ( Fig. 2 ) . Species of the group
occur in Central America at least to eastern and southern Mexico.
Insofar as I can judge from the description (Bokermann, 1974),
E. vinhai, in eastern Brasil, is a member of this group.

Remarks. — As noted above, the imistrigatus group may not prove
separable from the Antillean auricidatus group. Likewise, some of
the species listed above may be auricidatus group species (i.e.,
those from the Coastal Range of Venezuela and the Sierra Nevada
de Santa Marta in Colombia ) .

The imistrigatus group has its center of di\"ersity in the Andes
of Colombia and Ecuador. The distributional rami of this group
extending south along the east face of the Andes and cast across
the northern Amazon basin contain few ( 3-4 ) species. The ancient
Guyana highlands have produced only 2 species. The Coastal
Range of Venezuela and the Santa Martas of Colombia each have
a half-dozen species. Sa\'age's (1973) checklist includes fewer than
a dozen species in Costa Rica. By way of contrast, the region about
Santa Cecilia in eastern Ecuador harbors 12 species of this gioup.
Most localities on the Andean slopes of Ecuador support 5-S species
of the unistrigatus group.

COMPARISONS WITH ANTILLEAN GROUPS

Schwartz generally has recognized 6 species groups for Antil-
lean EleutJierodactylus even though a number of species has not
been assigned to any species group. By and large, he used the same
features E. R. Dunn proposed to separate groups. Shreve and
Williams ( 1963 ) proposed using some additional features ( i.e..

18 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

webbing, body glands) but Schwartz (1966, 1967) has objected
adequately to the use of such features. Neither Dunn nor Schwartz
proposed using finger lengths as a means of distinguishing groups,
perhaps because all Cuban forms are identical with regard to this
feature. In perusing Cochran's ( 1941 ) Hispaniolan herpetology,
Stejneger's (1904) and Schmidt's (192S) Puerto Rican studies,
Lynn and Grant's (1940) Jamaican study, as well as the many
papers by Schwartz I noted that only 5 West Indian Eleutlwrodoc-
tyliis have the thumb longer than the second finger. Schwartz
(1965, 1976) proposed separating these from the auricidatus group
as the inoptatiis group and in so doing noted that the frogs of this
group were quite dissimilar to all other Antillean Eleutlwrodacttjlus.
That dissimilarity is reflected osteologically as well; unlike all other
Antillean EIeutJierodactijJus\ the frontoparietal and prootic bones
are not fused in frogs of the inopfatus group (Lynch, 1971).

As already noted, the Antillean auricidatus group is \'ery similar
to the mainland uiustiiis,atus group. The Antillean equivalent to the
unistrigattts group includes the aurkukiius and varleyi groups of
Dunn and Schwartz. Schwartz (1969) termed the digits of E.
jiavescens and E. poolei "notched " suggesting a possible relationship
of these otherwise normal auriculatus group frogs to some south-
eastern Brasilian groups (e.g., lacteus and parvus groups). The
digital pads of E. favescens and E. poolei are not notched in the
same way that the pads are notched in the southeastern Brasilian
frogs but are more appropriately termed emarginate and thus re-
semble the pads of certain Mexican frogs (alfredi group) as well as
some members of the unisfrigatus group (i.e., E. anotis of Venezuela
and E. latidiscus of Colombia and Ecuador).

The other four groups currently recognized by Schwartz ( in Hit;
emiliae, gossei, ricordi, and sipningtoni) do not have readily identi-
fied South American equivalents. The four groups are distinguished
in the breadth of the prevomerine odontophores (very broad in
gossei and ricordi groups, patch-like in emiliae and symingtoni
groups), texture of the skin (smooth in eiudiae and gossei groups,
rugose in ricordi and symingtoni groups), and body size (syming-
toni group frogs are quite large). The only South American groups
having broad prevomerine odontophores (some hiporcatus and
sulcatus group species) are markedly difterent large frogs having
broad heads, prominent cranial crests, and the thumb longer than
the second finger. The South American ramagii group reminds me
of some ricordi group frogs until the prevomerine dentitions are
compared.

The Antillean inoptatus group resembles the South American
sulcatus group in having broad prevomerine dentition (sHghtly
narrower than those of frogs of the emiliae, gossei, and ricordi
groups), long first fingers, cranial crests, and lack of frontoparietal-

SPECIES GROUPS OF ELEUrHERODACTYLUS 19

prootic fusion. The groups difier in that the siilcatus group frogs
have broad heads, markedly (rather than feebly) areolate skin on
the venter, and contact between the pterygoid and parasphenoid
bones.

COMPARISON WITH MEXICAN AND CENTRAL
AMERICAN GROUPS

As mentioned above, the biporcatus, fitzingeri, and unistrigatiis
groups extend easterly and northeasterly into Central America. The
biporcatus group is represented in Central America by two species
also found in Colombia (£. biporcatus and E. bujoniforniis). The
fitzingeri group includes Central American frogs placed in the
goUmeri and rugulosus groups of Savage (1973, 1975) and Smith
and Taylor (1948) as well as some species previously placed in the
rhodopis group (E. anzuetoi, E. lineaius, E. macdougaUi, E. ros-
tralis, and E. werleri) and laticcps group (£. laticeps and E.
stantoni). The unistrigatus group includes Savage's (1973) cruen-
tus, diastema, and mclanostictus groups as well as some species
previously placed in the alfredi group (£. batrachylus Taylor and
E. glaiicus Lynch) and mexicamis group [E. greggi Bumzahem and
E. omiltemanus (Giinther)].

For the present, the remaining species of Smith and Taylor's
(1948) alfredi group are left in that group. I am not convinced
that these remaining eleven species are closely related. The alfredi
group is partially characterized in having smooth skin on the
venter, the first finger shorter than the second, and large, emarginate
digital pads. The species of the group occur in southern Mexico
and Guatemala.

In the rhodopis group I include the species Lynch ( 1970) placed
in the mexicanus and pijgmaeus groups (except for greggi, occi-
dentalis and omiltenuums) as well as most species Smith and Taylor
( 1948 ) included in their rhodopis group. Frogs of this group have
smooth skin on the venter, the first finger as long as the second,
small digital pads, and lack discs on the innermost fingers and toes.
Males have markedly larger tympana (nearly as large as or larger
than the eye) than do females. These seven species occur in south-
ern Mexico and Guatemala and along the Pacific versant to Nica-
ragua (Stuart, 1963). Eleutherodactijlus bransfordi of Nicaragua,
Costa Rica, and Panama also appears to belong to this group.

The alfredi and rhodopis groups superficially resemble the An-
tillean dimidiatus, ricordi, and syuiingtoni groups but differ in
lacking the frontoparietal-prootic fusion and in ha\ing longer me-
dian pterygoid rami. The southeastern Brasilian parvus group
superficially resembles some rhodopis group members. The peculiar
digital structure of parvus group frogs does not suggest a close

20

OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

relationship. At least for the present, it appears that the two en-
demic Middle American groups are distinguishable from the 9
South American and 6 Antillean groups.

Table 1. — Summary of Species-Croup Arrangements for Eleutherodactyliis

Mexico'

Costa Rica^

Colombia* South America^ West Indies*

alfredi

mexicanus

ptjgmaeiis

iliodopis

laticeps

nin.ulosus

[mexicanus ~1
ptjgmaeus I
rliodopi.s _J

golbneri

riigulosits

fitzingeri

gaigeae

btifoniformis

biporcatus

cruentus

melanostictus

diastema

>- Group I

y

roup II

Group III
Group IV

}

fitzingeri

biporcatus

iiuistrigatiis

siilcafus

binotatus

discoidaJis

lacteus

parvus

ramagii

auriculatus

auriculatus

emiliae

gossei

inoptatus

ricordi

symingtoni

varleyi

'Modified after Smith and Taylor (1948). "Combined herein; see text.
'After Savage (1973). 'After Cochran and Goin (1970). ''Proposed herein.
" Following Schwartz.

SPECIES GROUPS OF ELEUTHERODACTYLUS 21

SUMMARY

The South American species of the genus Eleutlierodactylus
are partitioned into 10 species groups. Three groups are centered
in southeastern Brasil {hinotatus, lacteus, and parvus groups). The
ramagii group is found in extreme eastern Brasil. The discoidalis
group occurs on the interior Andean slopes in Ecuador, southern
Peru and Bolivia, and nothern Argentina. The fitzingeri group is
the most widespread species group; frogs of this group are primarily
lowland species and range throughout the forested Amazon liasin,
the Guyanas, and the Choco as well as into Central America as far
as Mexico. The biporcattis group occurs in lower Centi-al America,
western Colombia, and Ecuador. The sulcatus group occurs in the
Coastal Range of Venezuela and the upper Amazon basin. The
tinistrigatus group occurs primarily in Colombia and Ecuador but
species are also found along the northern and western edges of the
Amazon basin and the Coastal Range of Venezuela. Also the group
is moderately well represented in lower Central America; four spe-
cies occur in Guatemala and Mexico. The auriculatus group occurs
over most of the West Indies and is represented in South America
by E. urichi in Guyana and Venezuela.

One hundred and forty-one species are currently recognized
based on 169 names proposed for South American Eleutherodac-
tyltis. The numbers of species in each of the 10 species groups are:
auriculatus (1), hinotaiiis (6), hiporcatus (5), discoidalis (5),
fitzingeri (21), lacteus (4), parvus (2), ramagii (2), sidcatus (2),
and tinistrigatus (92).

The approximately 340 recognized species of Eleutlierodactylus
are currently placed in 6 primarily Antillean groups, 9 groups in
Central and South America, and 2 groups in Guatemala and Mexico.

RESUiMEN

Las especies sudamericanas del genero Eleutlierodactylus estan
repartidas en 10 grupos de especies. Tres grupos se encuentran
ubicados en el sudeste brasileiio {hinotatus, lacteus, y parvus). El
grupo ramagii se encuentra en el extremo este del Brasil. El grupo
discoidalis ocurre en el interior de las laderas andinas en Ecuador,
sur del Peru y Bolivia, y norte de Argentina. El grupo -fitzingeri
es el de mas amplia distribucion entre todos; los sapos de esta
grupo son primariamente de tierras bajas y se extienden a lo largo
de la cuenca amazonica, las Guayanas, y el Choco (colombiano)
asi como tambien en Centroamerica hasta Mexico. El grupo hipor-
catus se encuenti-a en las partes bajas de America Central, en el
oeste de Colombia, y en Ecuador. Al grupo sulcatus se lo encuentra
en la region costera de Venezuela y en el Alto Amazonas. El grupo
unistrigatus ocurre primariamente en Colombia y Ecuador pero

22 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

tambien sus especies se encuentran a lo largo de los bordes norte
y oeste de la Cuenca Amazonica y en zonas costeras de Venezuela;
este grupo esta tambien presente en los bajios de la America Cen-
tral. El grupo atiricuJatus se dispersa sobre la mayor parte de las
Indias Occidentales y esta representado en Sud- America por E.
urichi en Venezuela y Guyana.

Para este grupo de anfibios actualmente se reconocen 141 es-
pecies basadas en 169 nombres propuestos para los Eleutlierodac-
tijliis sudamericanos. El numero de especies en cada uno de los
diez grupos (incluyendo un group de las Indias Occidentales) son:
atiriculatiis (1), hinotatiis (6), biporcatus (5), discoidaUs (5),
■fttzmgeri (21), lacteus (4), parvus (2), ramagii (2), sulcatus (2),
and unistrigatus (92).

LITERATURE CITED

Barbour, T., Noble, G. K. 1920. Some amphiliians from northwestern Peru,

with a re\'ision of the genera PJiyllobatcs and Telmatobius. Bull. Mus.

Comp. Zool., 63:393-427.
BoKERMAXx, W. C. A. 1966. Lista annotada das localidades tipo de anfibios

brasileros. Servico documentacao, Sao Paulo, Brasil, RUSP.
BoKERMANNT, W. C. A. 1974. Tres especies novas de Eleutherodactylus do

sudeste da Bahia, Brasil (Anura, Leptodactylidae ) . Rev. Brasil. Biol.,

34:11-18.
BouLEXGER, G. A. 1882. Catalogue of the Batrachia Salientia s. Ecaudata in

the collections of the British Museum (Natural History). 2nd. ed., London.
CocHRAX, D. M. 1941. The herpetolog>' of Hispaniola. Bull. U. S. Natl. Mus.,

(177): 1-398.
CocHRAx, D. M. 1955. Frogs of southeastern Brazil. Ibid., (206): 1-423.
CocHRAX, D. M. 1956. A new species of frog from Kartabo, British Guiana.

Zoologica, 41:11-12.
CocHRAx, D. M., Goix, C. T. 1970. Frogs of Colombia. Bull. U. S. Natl.

Mus., (288): 1-655
Cope, E. D. 1862a. On some new and little known American Anura. Proc.

Acad. Nat. Sci. Philadelphia, 14:151-159.
Cope, E. D. 1862b. Catalogues of the reptiles obtained during the explorations

of the Parana, Paraguay, Vermejo and Uruguay rivers by Capt. Thos. J.

Page, USN; and of those procured by Lieut. N. Michler, US top. Eng.,

Commander of the expedition conducting the sur\'ev of the Atrato ri\er.

Ibid., 14:346-3.59.
Cope, E. D. 1863. On Trachtjcephalus, Scaphiopus and other American Ba-
trachia. Ibid., 15:43-54.
Duxx, E. R. 1925. New frogs from Cuba. Occas. Papers Boston Soc. Nat.

Hist., 5:163-166.
Duxx, E. R. 1926. AddiHonal frogs from Cuba. Ibid., 5:209-215.
DuNX, E. R. 1942. A new species of frog {Eleutherodactylus) from Costa

Rica. Notulae Naturae, ( 104 ) : 1-2.
Duxx, E. R. 1944. Heipetolog>- of the Bogota area. Rev. Acad. Colombiana

Cien. Ex. Fis. Nat., 6:68-81.
Firscheix, I. L. 1951. Redisco\ery of the broad-headed frog, Eleutherodac-
tylus laticeps (Dumeril), of Mexico. Copeia, 1951(4) :268-274.
GoRHAM, S. 1966. Liste der rezenten Amphibien und Reptilien. Ascaphidae,

SPECIES GROUPS OF ELEUTHERODACTYLUS 23

Leiopelmatidea (sic), Pipidae, Discoglossidae, Pelobatidae, Leptodactyli-
dae, Rhinophrynidae. Das Tieneich, 85:1-222.

Hallowell, E. 1860. Report on the Reptilia of the North Pacific Expedition,
under the command of Capt. John Rogers, USN. Proc. Nat. Sci. Phila-
delphia, 6:480-509.

Heyer, W. R. 1972. The status of Leptodactyhts pinnilio Boulenger (Am-
phibia, Leptodactylidae ) and the description of a new species of LcjHo-
dactyhis from Ecuador. Contrib. Sci. Nat. Hist. Mus. Los Angeles Co.,
(231):l-8.

Lynch, J. D. 1965. A review of the eleutherodactylid frog genus Micro-
batrachijlus (Leptodactylidae). Nat. Hist. Misc., (182):1-12.

Lynch, J. D. 1968a. Genera of leptodactylid frogs in Mexico. Univ. Kansas
Publ. Mus. Nat. Hist., 17:503-515.

Lynch, J. D. 1968b. The status of the nominal genera Basanitia and Phry-
nanodus from Brazil (Amphibia; Leptodactylidae). Copeia, 1968(4):
875-876.

Lynch, J- D- 1969. Taxonomic notes of Ecuadorian frogs (Leptodactylidae:
Eh'utliciodactyhis) . Herpetologica, 25:262-274.

Lynch, J. D. 1970. Taxonomic notes on some Mexican frogs { Eletitherodac-
tyhis: Leptodactylidae). Ibid., 26:172-180.

Lynch, J. D. 1971. Evolutionary relationships, osteology, and zoogeography
of leptodactyloid frogs. Univ. Kansas Mus. Nat. Hist. Misc. Publ., ( 53 ) :
1-238.

Lynch, J. D. 1974. New species of frogs (Leptodactylidae: Eletitherodac-
tylus) from the Amazonian lowlands of Ecuador. Occas. Papers Mus. Nat.
Hist. Univ. Kansas (31): 1-22.

Lynch, J. D. 1975a. The identity of Elctitherodactyhis conspicillatus (Giin-
ther), with the descriptions of two related species from northwestern South
America (Amphibia, Leptodactylidae). Contrib. Sci. Nat. Hist. Mus. Los
Angeles Co., (272): 1-19.

Lynch, J. D. 1975b. A review of the broad-headed eleutherodactyline frogs
of South America (Leptodactylidae). Occas. Papers Mus. Nat. Hist. Kan-
sas, (38): 1-46.

Lynch, J. D. 1975c. A review of the Andean leptodactylid frog genus
Phrynopus. Ibid., ( 35 ) : 1-51.

Lynch, J. D., Schwartz, A. 1972. Taxonomic disposition of some 19th cen-
tury leptodactylid frog names. J. Herp., 5:103-114.

Lynn, W. G., Grant, C. 1940. The herpetology of Jamaica. Bull. Inst.
Jamaica, Sci. Ser., 1:1-148.

MiRANDA-RiBEiRO, A. 1926. Notas para servirem ao estao dos Gymnoba-
trachios (Anura) brasileiros. Arq. Mus. Nac, 27:1-227.

Peters, W. 1872. Uber die von Spix in Brasilien gesammelten Batrachier des
Konigl. Naturalienkabinets zu Miinchen. Monatsberichte k. Akademie der
Wissenschaften zu Berlin, 1872:196-227.

Savage, J. M. 1968. The distribution and synonym\' of the Neotropical frog,
Eleutherodachjhis mow. Copeia, 1968(4) : 878-879.

Savage, J. M. 1973. A preliminary handlist of the herpetofauna of Costa
Rica. Privately published. 17 pp.

Savage, J. M. 1974. On the leptodactylid frog called Eleutherodactylus pal-
viaius (Boulenger) and the status of Hylodes fitzingeri O. Schmidt. Her-
petologica, 30:289-299.

Savage, J. M. 1975. Systematics and distribution of the Mexican and Central
American stream frogs related to Eleutherodactylus rugulosus. Copeia,
1975(2) :254-306.

Schmidt, K. P. 1928. Ampliibians and land reptiles of Puerto Rico, with a

24 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY

list of those reported from tlie Virgin Islands. Ann. New York Acad. Sci.,
10:1-160.

Schwartz, A. 1958. Four new frogs of the genus Eleutherodactijlits ( Lepto-
dactyHdae) from Cuba. Amer. Mus. Novitates, (1873): 1-20.

Schwartz, A. 1964. Three new species of frogs ( Leptodactylidae, Eleuthero-
dacttjlus) from Hispaniola. Breviora, (208): 1-15.

Schwartz, A. 1965. Variation and natural history of Eleutherodactylus riithae
on Hispaniola. Bull. Mus. Comp. Zool., 132:479-508.

Schwartz, A. 1966. The relationships of four small Hispaniolan Eleutherodac-
tylus (Leptodactylidae). Ibid., 133:369-399.

Schwartz, A. 1967. Frogs of the genus Eleutherodactylus in the Lesser An-
tilles. Stud. Fauna Curacao and other Carib. Isl., 24 (91): 1-62.

Schwartz, A. 1969. The Antillean Eleutherodactijlus of the auriciilatus group.
Ibid., 30(1 14): 99-1 15.

Schwartz, A. 1973. Six new species of Eleutherodactylus (Anura, Leptodac-
tylidae) from Hispaniola. J. Herp., 7:249-273.

Schwartz, A. 1976. Two new species of Hispaniolan Eleutherodactylus (Lep-
todactylidae). Herpetologica. 32:163-171.

Schwartz, A., Thomas, R. 1975. A checklist of West Indian amphibians and
reptiles. Carnegie Mus. Nat. Hist., Spec. Publ., (1):1-216.

Shreve, B., Williams, E. E. 1963. The frogs, pp. 302-342. In Williams,
E. E., Shreve, B. and Humphrey, P. S., The heipetolog>' of the Port-au-
Prince region and Gonave Island, Haiti. Parts I-II. Bull. Mus. Comp. Zool.,
129:291-342.

Smith, H. M., Taylor, E. H. 1948. An annotated checklist and key to the
Amphibia of Mexico. Bull. U. S. Natl. Mus., (194): 1-118.

Stejneger, L. 1904. The herpetology of Puero Rico. Ann. Rept. U. S. Natl.
Mus., 1902:549-724.

Stuart, L. C. 1963. A checklist of the herpetofauna of Guatemala. Misc.
Publ. Mus. Zool. Univ. Michigan, (122): 1-150.