International Society for the Study

and Conservation of Amphibians

(International Society of Batrachology)

SEAT

Laboratoire des Reptiles et Amphibiens, Muséum national d'Histoire naturelle,

5 rue Cuvier, 75005 Paris, France

BOARD FOR 1995

President: Raymond F. LAURENT (Tucumän, Argentina).

General Secretary: Annemarie OHLER (Paris, France).

Treasurer: Dominique PAYEN (Paris, France).

Deputy Secretaries: Brita GRILLITSCH (Wien, Austria); W. Ronald HEyER (Washington, U.S.A.):

Stephen J. RicHaRDs (Townsville, Australia).

Deputy Treasurers: Jean-Louis DENIAUD (Paris, France); Mark L. WyGoDA (Lake Charles, U.S.A.).

Councillors: Janalee P. CALDWELL (Norman, U.S.A.); Marissa FABREZI (Salta, Argentina); Sergius

KUZMIN (Moscow, Russia); Jon LOMAN (Lund, Sweden).

TARIFF FOR 1995

Individuals Institutions

Subscription to Alytes alone 250 FF / 50 $ (regular) 500 FF /100$

125 FF / 25 $ (student)

Subscription to Alytes + ISSCA + Circalytes 270 FF / 54 $ (regular) 540 FF /108$

135 FF / 27 $ (student)

Back issues of Alytes: single issue 60FF/125$ 120 FF/245$

Back issues of Ayu ne complete volume (4 issues) 200 FF / 40 $ 400 FF / 80 $

Back issues of 4/yu omplete set volumes 1 to 12 1920 FF / 38458 3840 FF / 768$

Five-years (1995-1999) individual subscription to Alytes: 1000 FF / 200 5.

Five-years (1995-1999) individual subscription to Alytes + ISSCA + Circalytes: 1350 FF / 270 $.

Life individual subscription to Alytes from 1995 on: 5000 FF / 1000 $.

Life individual subscription to Alytes + ISSCA + Circalytes from 1995 on: 6400 FF / 1280 $.

Patron individual subscription to Alytes from 1995 on: 10000 FF / 2000 $ or more.

Patron individual subscription to Alytes + ISSCA + Circalytes from 1995 on: 12800 FF / 2560 $ or

more.

Circalytes is the internal information bulletin of ISSCA. Back issues of this bulletin are also available:

prices can be provided upon request by our Secretariat.

Inclusive Section or Group affiliation to ISSCA: 250 FF / 50 $.

Individual subscription to the ISSCA Board Circular Letters: 200 FF / 40 $.

MODES OF PAYMENT

— In French Francs, by cheques drawn on a French bank payable to “ISSCA”, sent to our

Secretariat (address above)

— In French Francs, by direct postal transfer to our postal account: “ISSCA”, Nr. 1-398-91 L, Paris;

if you use this mode of payment, add 15 FF to your payment for postal charges at our end.

x MU! 5 Dollars, by cheques payable to “ISSCA”, sent to Mark L. WyGoDA, Department of

Biological and Environmental Sciences, P.O. Box 92000, MecNeese State University, Lake Charles,

Louisana 70609-2000, U.S.A

Source : MNHN, Paris

AILTTES

INTERNATIONAL JOURNAL OF BATRACHOLOGY

March 1995 Volume 13, N° 1

Alytes, 1995, 13 (1): 1-13. Bibliothèque Centrale Muséum

Description of a LR Amazonian

and Guianan tree frog, genus Osteocephalus

(Anura, Hylidae), with oophagous tadpoles

Karl-Heinz JUNGFER * & Luis Cesar SCHIESARI **

* Birkenweg 4, 74427 Fichtenberg, Germany

** Departamento de Zoologia, Instituto de Biociências, Universidade de Säo Paulo,

C.P. 20520, 01452-990 Säo Paulo-SP, Brazil

A new species of the hylid frog genus Osteocephalus is described from

Manaus, Amazonas, Brazil. It is a medium-sized species that resembles

©. taurinus, but is smaller and has shorter hind legs. In contrast to other

species of Osteocephalus, the dorsal skin is barely sexually dimorphic and the

male’s vocal sac is single, median and subgular. The tadpole is peculiar for

living in phytotelmes, feeding on conspecific eggs and having a labial tooth

row formula of 2(2)/3.

Most species of the neotropical hylid frog genus Osteocephalus Steindachner, 1862 are

identifiable by the presence of well-ossified skulls, the presence of paired lateral vocal sacs,

and rugose or warty dorsal skin in males and relatively smooth skin in females. TRUEB &

DUELLMAN (1971) reviewed and defined the genus and recognized five species: O. buckleyi

(Boulenger, 1882), O. leprieurii (Duméril & Bibron, 1841), O. pearsoni (Gaige, 1929), O.

taurinus Steindachner, 1862 and O. verruciger (Werner, 1901) (verrucigerus auct.).

DUELLMAN (1974) followed Core (1867) in placing Hyla langsdorffii Duméril & Bibron,

1841 in the genus. Topotypes of Hyla elkejungingerae Henle, 1981 were identified as O.

verruciger by D'UELLMAN (in litt. to W. BÔHME, pers. comm.) but placed into the synonymy

of O. taurinus by HOOGMOED (in FRosT 1985) and regarded as a distinct species, O.

elkejungingerae, by HENLE (1992). MARTINS & CARDOSO (1987) described O. subtilis from

Acre, Brazil, and DUELLMAN & HOOGMOED (1992) placed Hyla rodriguezi Rivero, 1968 in

Osteocephalus. AYARZAGÜENA et al. (1992a) described five species, O. aecii, O. edelcae, O.

galani, O. luteolabris and O. rimarum from Venezuelan table mountains and placed them

in the O. rodriguezi group. AYARZAGÜENA et al. (1992b) transferred this group to a

BIBL. DU

MUSEUM Source : MNHN, Paris

Dane

2 ALYTES 13 (1)

new genus, Tepuihyla. Thus, between 7 and 14 described species are currently comprised

in the genus.

An additional species from midwestern Amazonia is in the process of being described

by M. MarTINs and M. GORDo (pers. comm.). One more species from central Amazonia

and two Guyanese states has been known for much more than a decade and has appeared

in the literature many times as Osteocephalus sp. (HOOGMOED, 1979; ZIMMERMAN, 1983;

ZIMMERMAN & BOGART, 1984; MARTINS & CARDOSO, 1987; HERO, 1990; HôDL, 1990, 1993;

ZIMMERMAN & RODRIGUES, 1990; HOOGMOED & AViLA-PIRES, 1991; DUELLMAN &

HooGMoEp, 1992; WEYGOLDT & JUNGFER, 1993) or erroneously as Osteocephalus buckleyi

(ZIMMERMAN & BOGART, 1988). In the course of our independent work on the reproductive

biology and tadpole morphology of this species, we felt that the taxonomic status of the

frog needed to be resolved. Therefore we describe it here as

Osteocephalus oophagus sp. nov.

Holotype. — MZUSP 69852, an adult male, collected by K.-H. JUNGFER on 9 April 1993

at Reserva Florestal Adolfo Ducke (2°55'S, 59°59°W), situated at km 26 of the Rodovia

AM-010 (Manaus-Itacoatiara), Estado do Amazonas, Brazil.

Paratypes. — NMW 32925.1-2, collected by W. HôDL on 28 February 1978; MPEG

4845-4846, collected by M. J. HENZL, L. S. FORD and A. LiMA on 9 February 1992;

AMNH A.136183-136184, collected by M. J. HENZL and L. S. ForD on 3 March 1992;

MZUSP 69853, INPA 01446 and 01448, SMNS 10801-10802, ZFMK 57137-57138, all

collected by K.-H. JUNGFER between January and April 1993; all from Reserva Florestal

Adolfo Ducke. For sexes and Museum abbreviations see Table I.

Definition. — For purposes of comparisons, we follow TRUEB & DUELLMAN’s (1971)

standards. À medium sized species of Osteocephalus (maximum known snout-vent length

(SVL) 47.2 mm in males, 62.7 mm in females). Dorsal skin in males with a few non-spinous

tubercles or smooth, in females smooth. Skin on flanks smooth, slightly shagreened

posterior to the insertion of the arm in some specimens. Webbing on hand moderate, up

to first third of antepenultimate phalanx on inner edge of third finger. Dorsum brown to

grey with or without irregular tan flecks, spots, mottling or reticulation, with white spots

in some specimens. Venter white; in some specimens very light brown reticulation present

on the chest. Lips brown or grey, with or without a cream subocular spot extending to the

edge of the tympanum. Flanks brown or grey. White mottling or reticulation on posterior

half of the flanks; many white spots on dark ground present in some specimens.

Diagnosis. — Osteocephalus oophagus is distinguished from the frogs of the O. rodriguezi

group by substantial webbing on the hand (absent or rudimentary in the ©. rodriguezi

group), and from O. buckleyi by the lack of a conspicuous row of tubercles on the tarsus

and prominent supraorbital tubercles. O. verruciger and O. elkejungingerae are larger and

the dorsum of males is covered by many spinous tubercles. ©. leprieurii has less webbing

on the hands (web reaching base of antepenultimate phalanx on inner side of third finger)

and in some cases transverse lines or bars on the dorsum that are absent in ©. oophagus.

O. langsdorffii is larger and has scalloped dermal folds on the outer edges of hands and

Source : MNHN, Paris

JUNGFER & SCHIESARI 3

feet that are lacking in O. oophagus. The new species is distinguished from O. pearsoni by

the lack of brown reticulation on the venter (present only in chest area, if at all, in ©.

oophagus) and more webbing on the hands (to base of antepenultimate phalanx on inner

side of third finger in O. pearsoni). O. subtilis has a black iris (golden with black rays in

©. oophagus) and a glandular line above the vent (absent in ©. oophagus). An additional

undescribed species from midwestern Amazonia bears a conspicuous, complete white line

running around the whole of the upper lip (faint and interrupted in O. oophagus), and its

supratympanic fold is smooth (tubercular in ©. oophagus) (M. MARTINS, pers. comm.). O.

oophagus is most easily confused with ©. taurinus, with which it shares the golden iris with

radiating black lines. O. taurinus is a larger frog that reaches 104 mm. The dorsum in males

is covered by many spinous tubercles with keratinized tips. The webbing on the hands is

slightly more extensive than in O. oophagus, reaching the middle of the antepenultimate

phalanx on the inner side of the third finger. The legs are longer than in ©. oophagus.

When stretched forward and angled at 90° at the tibiotarsal articulation, the tarsus reaches

the tip of the snout in ©. oophagus and extends beyond that point in ©. taurinus.

Furthermore, adults of O. fauriius possess two elevated longitudinal bony ridges

(frontoparietal flanges) in the interorbital area, that are not visible in ©. oophagus, but

may be felt in large specimens when rubbing with the fingertips.

Description of holotype. — Measurements and proportions (following D'UELLMAN, 1970)

are given in Tables I and II. A male of 43.8 mm SVL (fig. la). Body wider than the head.

Diameter of the tympanum slightly wider than half the eye diameter. Head flat between

orbits, slightly concave in the intercanthal region and truncate in lateral and dorsal aspect.

Nostrils elevated, internarial region slightly depressed. Canthus rostralis slightly rounded,

loreal region concave, bearing a few rounded warts below the canthus. An elevated dermal

fold ascending posterior to the mid level of the eye to the area above the tympanum and

sloping to the insertion of the arm. Tympanum distinct and rounded. A weak axillary

membrane extending to less than one fifth the length of the humerus. A row of flat, barely

raised warts visible on the ventrolateral edge of the forearm, extending onto the proximal

half of the fourth finger. Fingers with moderately large discs, that of the third finger about

four fifths the diameter of the tympanum. Enlarged prepollex bearing an elliptical tubercle.

Nuptial pads absent (dark brown before), as the frog was no more in breeding condition

when preserved on 28 July 1993. Distal subpalmar tubercle bifid on finger IV. Webbing

basal between finger I and II and with a webbing formula (SAVAGE & HEYER, 1967; MYERS

& DuELLMAN, 1982) of 111,5-371112,5-2*IV in the others. Relative finger lengths of

adpressed fingers 3>4>2> 1 (fig. 2a). Legs relatively short. Tarsus reaching the tip of the

snout when stretched forward parallel to the body axis and bent at 90° at the tibiotarsal

articulation. Inner metatarsal tubercle large, flat and elliptical. No outer metatarsal

tubercle. Toe webbing formula 11*-2-111*-21111*-21V2-1*V. Relative toe lengths

4>5>3>2>1 (fig. 2b). Vent opening positioned posteriorly at the upper level of the

thighs.

Dorsally, skin weakly granular with low tubercles in the inter- and postorbital region

and few on the anterior part of the dorsum. Dorsal aspects of the arms weakly granular

and those of the legs smooth. Laterally, skin tubercular posteroventral to the tympanum,

shagreened on the anterior half of the flanks and smooth on the posterior half, tubercular

Source : MNHN, Paris

ALYTES 13 (1)

Table 1. - Measurements of type specimens of Osteocephalus oophagus. Measurements in mm:

ED, eye diameter; FL, foot length; HL, head length; HW, head width; SVL, snout-vent

length; TL, tibia length; TD, tympanum diameter. Museum abbreviations: AMNH,

American Museum of Natural History, New York; INPA, Instituto Nacional de Pesquisas

da Amazônia, Manaus; MPEG, Museu Paraense Emilio Goeldi, Belém; MZUSP, Museu de

Zoologia, Universidade de Säo Paulo; NMW, Naturhistorisches Museum Wien; SMNS,

Staatliches Museum für Naturkunde, Stuttgart; ZFMK, Zoologisches Forschungsinstitut und

Museum Alexander Koenig, Bonn.

Len SVL HL HW ED TD TL FL

Males

SMNS 10802 35.7 12.2 10.7 3.90 2.16 19.5 13.8

MPEG 4846 41.6 14.5 13.7 4.80 3.12 23.5 17.5

INPA 1448 423 142 13.1 5.10 2.82 22.8 15.8

AMNH 136184 42.5 15.2 13.8 5.10 2.86 242 17.8

NMW 32925.2 43.2 14.6 13.9 4.86 2.82 22.7 17.0

ZFMK 57138 43.3 15.5 14.5 5.22 3.24 24.3 17.9

MZUSP 69852 43.8 14.7 13.0 4.56 2.52 23:3 17.2

SMNS 10801 47.2 16.6 14.6 6.00 3.66 23.5 18.3

Females

AMNH 136183 46.2 16.0 13.7 5.34 3.20 23.8 18.1

ZFMK 57137 49.8 17.2 16.9 5.34 3.48 27.7 20.3

MPEG 4845 53.2 18.2 16.9 5.52 3.96 29.5 22.5

MZUSP 69853 53.6 17.9 16.7 5.82 3.90 30.8 21.2

NMW 32925.1 54.6 18.9 16.5 5.46 4.32 31.2 22.0

INPA 1446 55.6 17.7 16.8 5.10 3.72 29.5 21.6

Table II. - Proportions of male and female Osteocephalus oophagus. Abbreviations: see Table I.

Locality, sex HL/SVL |HW/SVL| TD / ED | TL/SVL | FL/SVL

Reserva Ducke, males

min. | 0.336 0.297 0.553 0.498 0.374

max.| 0.358 0.335 0.650 0.569 0.421

mean| 0.346 0.316 0.585 0.542 0.399

Reserva Ducke, females

min. 0.318 0.297 0.599 0.515 0.388

max.| 0.346 0.339 0.791 0.575 0.423

mean| 0.339 0.312 0.693 0.551 0.402

Rio Urucu, females

min. | 0.348 0.297 0.735 0.557 0.403

max.| 0.357 0.301 0.835 0.574 0.418

mean| 0.353 0.299 0.785 0.566 0.411

Source : MNHN, Paris

JUNGFER & SCHIESARI S

below the vent. Skin on the belly and the posteroventral thigh surfaces granular, the other

ventral surfaces smooth. Tongue round. Prevomers angular, with 8 and 10 prevomerine

teeth. Vocal slits extending postero-laterally from the middle of the tongue. Vocal sac

subgular, median and weakly distensible during call.

Colour in alcohol. — Dorsal surfaces brown with tan flecks and blotches. Flanks light

brown with brown mottling on the posterior half. Upper and lower lips bordered by faint

creamy stripes often interrupted by tan spots. Venter creamy with some faint light brown

mottling on the chest. Throat creamy white with light brown mottling. Ventral surfaces of

the arms creamy with a brown hue, those of the legs pale light brown. Posterior thigh

surfaces brown. Bones green and visible through the skin. Iris yellowish white with many

radiant black stripes and a horizontal black bar on each side of the pupil.

Colour in life. — Depending on illumination and colour of the substrate the frog was

sitting on, the dorsal colour varied from light brown with indistinct darker flecks and

blotches to dark brown with almost black flecks and blotches. Flanks were brown to

creamy brown with darker brown-mottling on the posterior half. Ventral surfaces were

white. Bones were green and the iris golden yellow with radiant black stripes and

horizontal black bars.

Variation. — The largest male of the type series has an SVL of 47.2 mm, the largest female

of 55.6 mm. Thirty marked and released males from the type locality bearing nuptial pads

ranged from 35.9 to 45.5 mm, and seventeen ovigerous females from 49.8 to 60.6 mm. One

female (INPA 01442) from the Rio Urucu is 62.7 mm. There is little variation in

proportions between the sexes, except that the tympana of females are slightly larger than

those of males (Table II). The dorsal skin varies from smooth to weakly granular, bearing

none or a few raised tubercles. There is a tendency of males to bear a few more tubercles

on head and dorsum than females, but some males have both surfaces smooth, while some

females have a few tubercles on the head and sometimes on the dorsum. This variation is

found in the preserved material at hand, and was also obvious in the many live frogs we

saw. Dorsal tuberculation is not a reliable sexual dimorphic character. Breeding males bear

horny dark brown nuptial pads on the prepollices.

A lot of variation was observed in coloration. In preservative, the dorsal colour

ranges from grey to dark brown with or without darker blotches or flecks (figs. 1a-b).

Also, the amount of cream spots or mottling dorsally, laterally and around the vent is

variable (figs. 1c-d). The ventral brown mottling on chest and throat may be lacking or

reduced to the area under the lower jaw.

Habitat. — Reserva Florestal Adolfo Ducke, a reserve managed by the Instituto Nacional

de Pesquisas da Amazônia (INPA), consists of hilly terra firme lowland forest at an

altitude of about 50 m. Frogs usually migrate at night on vertical stems of young trees and

bushes normally less than 2 cm in diameter at heights of 0.5 to 2 m. Males call from near

or inside phytotelmes.

Reproductive biology. — Pairs of O. oophagus deposit their eggs in phytotelmes, such as

epiphytic or ground bromeliads, Buriti Palm (Mauritia flexuosa) leaf axils, water-filled

palm bracts lying on the ground or tree holes up to heights of about 35 m. Females return

to the deposition sites regularly in intervals of about five days, usually clasped by the same

Source : MNHN, Paris

ALYTES 13 (1)

Fig. 1. — Osteocephalus oophagus sp. nov. Specimens from the type locality. (a) Holotype MZUSP

69852, adult male. (b) Male showing numerous tan dorsal blotches. (c) Subadult female with

numerous small white spots. (d) Adult female showing extreme extent of lateral and dorsal white

spots.

Source : MNHN, Paris

JUNGFER & SCHIESARI 4

Fig. 2. — Hand (a) and foot (b) of a male Osteocephalus oophagus (paratype INPA 01448). Line

equals 5 mm.

Source : MNHN, Paris

8 ALYTES 13 (1)

males. If there are larvae already present, they consume the newly laid eggs. The eggs not

eaten hatch. The larvae starve if the mother fails to return and if they are not provided

with trophic eggs (WEYGOLDT & JUNGFER, 1993). A detailed study on the reproductive

biology is in progress.

Calls. — The advertisement call usually consists of one to six croaking notes emitted at

night and occasionally during the day (type A). At night it is very often followed by one

to four (usually two) distinctly different notes that sound like “ka kä” (type B).

Type A of frogs from the INPA-WWF MCSEP reserves and from Parque Nacional

Tapajés were described in detail by ZIMMERMAN & BOGART (1984), who recorded an

emphasized frequency of 1.62 + 0.65 kHz, low frequency range of 0.89 + 0.33 kHz, high

frequency range of 2.84 + 0.42 kHz, and call duration per note of 0.24 + 0.07 s.

ZIMMERMAN & BOGART (1988) again described calls and calling of this frog, under the

name of Osteocephalus buckleyi. They noted the high intraspecific variability and other

characteristics of the call.

Description of the tadpole. — Tadpoles were collected by L. C. SCHIESARI on 5 March 1993

in water-filled plastic basins used as egg-laying sites by O. oophagus near the Igarapé

Acarà, Reserva Florestal Adolfo Ducke, and preserved in 10 % formalin. Two tadpoles

were raised until metamorphosis. Measurements were taken with the optical measuring

unit Wild MMS 235. Drawings (figs. 3-4) were made with a camera lucida attached to a

stereomicroscope. Measurements were made according to GRILLITSCH et al. (1993).

Developmental stages were determined following GosnER (1960). Labial tooth row

formula (LTRE) is after ALTIG (1970). The following description is based on 10 tadpoles

ranging from stage 35 to stage 38. Measurements of larvae are given in Table III.

Tadpole of ORToN’s (1953) type IV. Body slightly depressed, ovoid in dorsal view.

Snout nearly truncate from above and, in profile, acutely rounded. Eyes positioned dorsola-

terally; interorbital distance 1.5-1.7 times the internarial distance, which in turn equals width

of oral disc. Nares rimmed, rounded, and directed anterolaterally. Their distance to the eyes

about half of their distance to tip of the snout. Spiracular tube sinistral, ventrolateral, and

directed posterodorsally. Spiracular opening slightly posterior to mid length of body, to

which it is tightly attached. Dorsal and ventral margin of caudal musculature parallel in

proximal third, then gradually narrowing and almost reaching tip of the tail. Ventral fin

slightly lower than dorsal one, fairly paralleling ventral margin of caudal musculature.

Dorsal fin extending a short distance onto body. Oral disc anteroventral and not

emarginate. One row of moderately sized marginal papillae with a medial gap in upper

labium which may be visible in dorsal view. Submarginal papillae in some specimens in

one discontinuous row in lower labium as well as scattered in ventrolateral portion of oral

disc. Two rows of denticles on the upper labium, the second one showing a distinct median

gap, and three rows on the lower one (LTRF 2(2)/3). Upper jaw sheath arched, lower one

V-shaped, both black and without serrations (magnification 50 x). Colour of measured

tadpoles chestnut brown dorsally; venter and caudal musculature lighter.

Ontogenetic change. — Newly metamorphosed frogs 12-13 mm in SVL and completely

different in colour from the adults. Dorsal and lateral surfaces grey except for a black

canthal stripe continuing as a supratympanic stripe posterior to eye to insertion of the arm.

Dorsal surfaces of upper arm and proximal half of lower arm white. An orange spot on

Source : MNHN, Paris

JUNGFER & SCHIESARI 9

Fig. 3. — Dorsal (a) and lateral (b) view of a tadpole of Osteocephalus oophagus in stage 36 (GOSNER,

1960) of typical proportions. Line equals 5 mm.

Fig. 4. — Oral disc of a tadpole of Osteocephalus oophagus (same individual as in fig. 3). Line equals

14 mm.

Source : MNHN, Paris

10 ALYTES 13 (1)

Table III. - Measurements of larvae of Osteocephalus oophagus (mean values, ranges in

parentheses). Measurements in mm: DT, maximum height of tail; LF, maximum height of

lower (ventral) tail fin, SVL, snout-vent length; TL, total length; UF, maximum height of

upper (dorsal) tail fin; VT, tail length.

19 |1 4.66

20 | 4 |4.73 (4.35-5.03)

23 | 5 |5.98 (5.70-6.48)|

27 |1 16.7 72 9.5 3.71 1.20

28 |1 19.5 8.6 10.9 4.68 1.61

30 |1 213 9. 12.2 4.97 1.60

31|1 242 10.2 14.0 521 1.61

33 | 2 |26.2(25.4-27.0)|11.7 (11.6-11.8)/14.5 (13.8-15.2))6.02 (5.93-6.10)]1.98 (1.86-2.10)/1.

34 | 2 [23.7 (22.5-24.9)/10.8 (10.1-11.4)] 12.9(12.4-13.5) |5.83 (5.31-6.34)| 1.74(1.59-1.89)

35 | 5 |27.0(25.0-28.2)/12.5 (10.0-13.3)/14.5 (12.8-16.0)|6. 14 (5.84-6.44)|1.99 (1.90-2.14)|

36 |1 28.9 142 14.7 6.49 2.08

37 |1 27.8 12.9 14.9 6.55 2.08

38 | 3 [29.5 (26.4-31.6)/13.9 (13.6-14.5)/15.5 (11.9-17.8)]6.44 (6.44-6.44)/2.06 (1.98-2.13)|1.

39 | 3 [30.0 (28.4-31.4)/13.8 (13.4-14.3)/16.2 (15.0-17.6)|6.96 (6.27-7.85)/2.07 (1.84-2.35)|1.

40 | 2 |34.6(32.9-36.2)/15.1 (15.0-15.1)/ 19.5 (17.8-21.2))6.78 (6.65-6.90)[1.96 (1.57-2.35)|1.

elbow, distal dorsal surface of lower arm and proximal dorsal surface of fourth finger. A

large white spot, capped by a smaller orange one, on heel. Another orange spot on knee

and in some specimens on outer edge of the metatarsus. Finger- and toe-discs orange. Iris

bright red without radiating black stripes. With this colour pattern, they are virtually

indistinguishable from sympatric O. taurinus (juveniles described as ©. taurinus by

DUELLMAN & LESCURE, 1973, are most likely those of Hyla geographica).

A specimen of 21 mm SVL already had the typical adult pattern, except that the white

spot on the upper arm and proximal lower arm was still present. The finger- and toe-discs,

as well as the iris, already with black rays and horizontal bars, had an orange hue.

Distribution. — Apart from the Reserva Ducke, about 25 km north of Manaus, we have

heard the distinctive call of this species in the forest of the Universidade do Amazonas

campus within the city of Manaus and in Reserva Florestal Walter Egler, about 50 km

north-northeast of Manaus. The frog is well known from the INPA-WWF Minimal

Critical Size of Ecosystems Project (MCSEP) reserves, approximately 80 km northeast of

Manaus (ZIMMERMAN, 1983; ZIMMERMAN & BOGART, 1984, 1988; ZIMMERMAN & RoODRI-

GUES, 1990). HooGMoED (1979) listed it from ‘“eastern Guiana” and HOOGMOED &

AVILA-PIRES (1991) recorded it from Petit Saut, French Guyana, and remarked that it was

also known from Suriname and Brazil. Those records are all in the Guianan biogeo-

graphical region (see HOOGMOED, 1979 for a delimitation). We have not found or heard the

species just south of the Amazon in the forests along the Manaus-Humaitä road, but M.

MaRTINS and M. GORDO (pers. comm.) found specimens at the Rio Urucu about 100 km

Source : MNHN, Paris

JUNGFER & SCHIESARI 11

SSE of Coari. It has also been recorded from the Parque Nacional do Tapajés near

Itaituba, southwest of Santarém, Parä (ZIMMERMAN & BOGART, 1984).

Discussion. — The new species possesses a number of characters new or unusual for the

genus. In their definition, TRUEB & DUELLMAN (1971) (slightly altered by DUELLMAN &

HooGMoEp, 1992) found Osteocephalus to have “vocal sacs paired, posterior, and when

inflated protruding posteroventral or posterolateral to angles of jaws”. Osteocephalus

oophagus has a single median subgular vocal sac, a character shared with some frogs of

the O. rodriguezi group (sensu AYARZAGÜENA et al., 1992a) (AYARZAGÜENA et al., 1992b).

Also, it is similar to some frogs of that group in that the dorsal skin structure is not a

reliable sexually dimorphic character (AYARZAGÜENA et al., 1992a). Although males tend

to be slightly more granular, there are smooth skinned males both alive and especially in

preservative.

Tadpoles have two upper and three lower tooth rows like most lentic hylid larvae. In

other species of Osteocephalus, however, there are 2-3 upper and 5-8 lower tooth rows

CHENLE, 1981; HERO, 1990). The reduction of denticle rows in ©. oophagus may be an

adaptation for oophagy as a special case of macrophagy. The tadpoles do not need to rasp

their food as grazers. This might also be the reason for the absence of beak serrations. The

reduction of denticle rows is known for many other arboreal tadpoles (LANNOO et al.,

1987) and does not oppose its inclusion in the genus Osteocephalus.

Despite the differences mentioned above, O. oophagus shares important characters

with ©. taurinus: the juvenile colorations and the colour of the iris in adults are identical.

The frontoparietal flanges are present, though less conspicuous in ©. oophagus. For these

reasons the new species may be most closely related to O. taurinus.

Derivatio nominis. — The specific name oophagus is a compound of the Greek oon (egg)

and phagein (to eat) and refers to the larval habit of eating conspecific eggs.

RESUMEN

Osteocephalus oophagus sp. nov. de Reserva Ducke, Manaus, Amazonas, Brasil,

especie mediogrande del género, es semejante a O. taurinus pero mäs pequeña y con las

piernas posteriores mäs cortas. En contraste a otros Osteocephalus, la piel dorsal no

muestra dimorfismo sexual y el saco vocal del macho es impar, mediano y subgular. El

renacuajo es atipico para este género, porque vive en fitotélmata, alimentändose de huevos

coespecificos y teniendo una formula de denticulos labiales de 2(2)/3. La especie es

conocida de las bajas Guianas y de la baja Amazonia.

ACKNOWLEDGEMENTS

K.-H. JUNG stay at the Instituto Nacional de Pesquisas da Amazônia (INPA), Manaus, was

supported by a grant of the Deutsche Forschungsgemeinschaft (Peter WEYGOLDT, principal

investigator). L. C. SCHIESARI was partially granted by FAPESP (Säo Paulo Government Funding

Agency), process 91/1131-0. We are grateful to Cornelia ADELMANN and Gustavo DE MATTOS

Source : MNHN, Paris

12 ALYTES 13 (1)

ACCACIO for executing the drawings, to Martin J. HENZL for the opportunity of searching with him

south of the Amazon, to Wolfgang BÔHME, Alain Dumois, Britta GRILLITSCH, Marcio MARTINS and

Andreas SCHLÜTER, who also translated the summary, and two anonymous reviewers, for critically

reviewing the manuscript, to T. C. S. AviLA PRES (MPEG), S. G. EGLER (INPA), C. W. Myers

(AMNH) and F. TIEDEMANN (NMW) for the loan of specimens in their care. Karin and Helmut

MÂGDErRAU helped once again with unavailable literature. Logistical support was received through

the INPA Department of Ecology. Our special thanks go to Walter HôDL, who is responsible for and

initiated our interest in this frog.

LITERATURE CITED

ALTIG, R., 1970. — A key to the tadpoles of the Continental United States and Canada.

Herpetologica, 26 (2): 180-207.

AYARZAGÜENA, J., SENARIS, J. C. & GORZULA, S., 1992a. — El grupo Osteocephalus rodriguezi de las

tierras altas de la Guayana venezolana: descripciôn de cinco nuevas especies. Mem. Soc. Cienc.

nat. La Salle, 52 (137): 113-142.

— 1992b. — Un nuevo genero para las especies del “grupo Osteocephalus rodriguezi” (Anura:

Hylidae). Mem. Soc. Cienc. nat. La Salle, 52 (138): 213-221.

Copr, E. D., 1867. — On the families of the raniform Anura. Proc. Acad. nat. Sci. Phila., 2: 189-206.

DuELLMAN, W. E., 1970. — Hylid frogs of Middle America. Monogr. Mus. nat. Hist. Univ. Kansas,

1: 1-753.

----- 1974. — A reassessment of the taxonomic status of some neotropical hylid frogs. Occ. Pap. Mus.

nat. Hist. Univ. Kansas, 27: 1-27.

DUELLMAN, W. E. & HOOGMoED, M. S., 1992. — Some hylid frogs from the Guiana Highlands,

northeastern South America: new species, distributional records, and a generic reallocation.

Occ. Pap. Mus. nat. Hist. Univ. Kansas, 147: 1-21.

DuELLMAN, W. E. & LESCURE, J., 1973. — Life history and ecology of the hylid frog Osteocephalus

taurinus, with observations on larval behavior. Occ. Pap. Mus. nat. Hist. Univ. Kansas, 13: 1-12.

FROsT, D. R. (ed.), 1985. — Amphibian species of the world. Lawrence, Allen Press & Assoc. Syst.

Coll.: [i-iv] + i-v + 1-732.

GRILLITSCH, B., GRILLITSCH, H., DUBOIS, A. & SPLECHTNA, H., 1993. — The tadpoles of the brown

frogs Rana [graeca] graeca and Rana [graeca] italica (Amphibia, Anura). Alytes, 11 (4):

117-139.

Goser, K. L., 1960. — A simplified table for staging anuran embryos and larvae with notes on

identification. Herpetologica , 16 (3): 183-190.

HENLE, K., 1982. — Hyla elkejungingerae, ein neuer Hylide aus dem peruanischen Regenwald

(Amphibia: Salientia: Hylidae). Amphibia-Reptilia, 2: 123-132.

= 1992. — Zur Amphibienfauna Perus nebst Beschreibung eines neuen Eleutherodactylus. Bonn.

zool. Beitr., 43 (1): 79-129.

HERO, J. M., 1990. — An illustrated key to tadpoles occurring in the Central Amazon rainforest,

Manaus, Amazonas, Brasil. Amazoniana, 11 (2): 201-262.

HôbL, W., 1990. — Reproductive diversity in Amazonian lowland frogs. In: W. HANKE (ed.), Biology

and physiology of amphibians, Fortschr. Zool., 38: 41-60.

Le 1993. — Amazonien aus der Froschperspektive. In : OÙ. Landesmuseum (ed.), Amerika,

Kataloge des OÙ. Landesmuseums, 61: 499-545.

HooGmorp, M. S., 1979. — The herpetofauna of the Guianan region. /n: W. E. DUELLMAN (ed.), The

South American herpetofauna: its origin, evolution, and dispersal. Monogr. Mus. nat. Hist. Univ.

Kansas, T: 241-279.

HooGMoED, M. S. & AviLa-Pires, T. C., 1991. — Annotated checklist of the herpetofauna of Petit

Saut, Sinnamary River, French Guiana. Zool. Med., 65 (5): 53-88.

LANNOO, M. J., TOWNSEND, D. S. & WASsERSUG, R. J., 1987. — Larval life in the leaves: arboreal

tadpole types, with special attention to the morphology, ecology, and behavior of the

oophagous Osteopilus brunneus (Hylidae) larva. Fieldiana Zool., (n.s.), 38: 1-31.

Source : MNHN, Paris

JUNGFER & SCHIESARI 13

MaRTINs, M. & CARDOSO, A. J., 1987. — Novas espécies de hilideos do Estado do Acre (Amphibia:

Anura). Rev. brasil. Biol., 47 (4): 549-558.

Myers, C. W. & DUELLMAN, W. E., 1982. — A new species of Hyla from Cerro Colorado, and other

tree frog records and geographical notes from western Panama. 4mer. Mus. Novit., 2752: 1-32.

ORTON, G., 1953. — The systematics of vertebrate larvae. Syst. Zool., 2: 5.

SAVAGE, J. M. & HEYER, W. R., 1967. — Variation and distribution in the tree-frog genus

Phyllomedusa in Costa Rica, Central America. Beitr. neotrop. Fauna, 5 (2): 111-131.

TRUEB, L. & DUELLMAN, W. E., 1971. — A synopsis of neotropical hylid frogs, genus Osteocephalus.

Occ. Pap. Mus. nat. Hist. Univ. Kansas, 1: 1-47.

‘WEYGOLDT, P. & JUNGFER, K.-H., 1993. — Oviphagy and the evolution of complex parental care in

frogs. In: Second World Congress of Herpetology Abstracts, Univ. Adelaide: 281-282.

ZIMMERMAN, B. L., 1983. — A comparison of structural features of calls of open and forest habitat

frog species in the Central Amazon. Herpetologica, 39 (3): 235-246.

ZIMMERMAN, B. L. & BOGART, J. B., 1984. — Vocalizations of primary forest frog species in the

Central Amazon. Acta Amazonica, 14 (3-4): 473-519.

ue 1988. — Ecology and calls of four species of Amazonian forest frogs. J. Herpet., 22 (1): 97-108.

ZIMMERMAN, B. L. & RODRIGUES, M. T., 1990. — Frogs, snakes, and lizards of the INPA-WWF

reserves near Manaus, Brazil. /n: A. GENTRY (ed.), Four neotropical rain forests, New Haven,

Yale Univ. Press: 426-454.

Corresponding editors: Ronald G. ALTIG & Alain DuBois.

Source : MNHN, Paris

Alytes, 1995, 13 (1): 14-44.

Reassessment of central

Peruvian Telmatobiinae

(genera Batrachophrynus and Telmatobius).

L. Morphometry and classification

Ulrich SiNsCH * , Antonio W. SALAS ** & Veronica CANALES *

* Institut für Biologie, Universität Koblenz-Landau, Rheinau 1, 56075 Koblenz, Germany

** Asociaciôn de Ecologia y Conservaciôn, Aptdo. 0359, Lima 18, Perû

The taxonomic status of the central Peruvian Telmatobiinae is reassessed

by analyzing the intraspecific variation of 18 morphometric measures among

the currently recognized taxa (3 genera, 6 species, 9 subspecies). Cluster

analysis, principal component analysis and discriminant analysis lead to the

recognition of two genera (Batrachophrynus and Telmatobius) including six

species (Batrachophrynus brachydactylus, B. macrostomus, Telmatobius bre-

virostris, T. carrillae, T. jelskii, T. rimac) without segregation in sul (es.

Two diagnostic external features distinguish Batrachophrynus from Telmato-

bius species, another two characters are convergent adaptations distin-

guishing the stream-inhabiting ecotype from the lake-inhabiting one.

INTRODUCTION

Leptodactylid frogs of the genus Telmatobius Wiegmann, 1835 inhabit the whole

range of the Andes from Ecuador in the north to Chile/Argentina in the south (FRosT,

1985; Cri, 1986). However, in the high-Andean regions of central Perû around Lake Junin

(FsLpsA, 1983), two endemic Te/matobius-like species have been described as members of

the genus Batrachophrynus Peters, 1873: the large lake-inhabiting Junin frog B. macro-

stomus and the stout stream-inhabiting B. brachydactylus. PETERS’s (1873) distinction was

based solely on the presence (Telmatobius) or absence (Batrachophrynus) of maxillary and

prevomerine teeth. Yet, T. brevipalmatus, T. edentatus and T. intermedius lack maxillary

teeth (LyNCH, 1971). Differences in several osteological characters other than teeth

between Batrachophrynus and Telmatobius confirmed the generic distinction (LYNCH,

1978) and indicated an early separation of Batrachophrynus from the Telmatobius stock

(Cr, 1986). LAURENT (1983) assigned B. brachydactylus to a monotypic third genus

Lynchophrys based on morphometric differences from B. macrostomus. At present, this

view on taxonomy is maintained, though independent support is absent (DUELLMAN, 1993;

FRosT, 1985; LaviLLA, 1988a). According to LAURENT'S opinion, Lynchophrys is more

recently derived from the Telmatobius stock than is Batrachophrynus. Thus, the

Source : MNHN, Paris

SINSCH, SALAS & CANALES 15

phylogenetic relationship among the two species B. brachydactylus and B. macrostomus

and of both with the genus Telmatobius are controversial.

The central Peruvian Telmatobius presently include four stream-inhabiting species, T.

brevirostris Vellard, 1955, T. carrillae Morales, 1988, T. jelskii (Peters, 1873) and T. rimac

Schmidt, 1954, whereas T. juninensis (Shreve, 1938) was an incorrectly classified Phrynopus

(DUELLMAN, 1993). VELLARD (1955) proposed three subspecies for T. brevirostris, four for

T. jelskii and two for T. rimac, although the significance of the morphometric differences

between the proposed subspecies was not tested for, and the existence of intermediate

individuals was stated. Field studies in the Mantaro Valley (Department of Junin, Peru)

on T. jelskii raised doubts on the subspecific classification because morphs pertaining to

different “‘subspecies” were found within the same population (SINsCH, 1985, 1986, 1990).

Part of the taxonomic confusion in this group is probably due to morphological

convergence among the species of each ecotype: the stream-inhabiting, semiaquatic frogs

are stout and usually moderate-sized or small, whereas the lake-inhabiting, fully aquatic

frogs are large (SINSCH, 1986, 1990). Therefore, phenotypic similarity between different

species may reflect convergent adaptation to the same environmental factor rather than

phylogenetic relationship. On the other hand, subtle differences among populations of the

same species could result from direct responses to such factors as the amount of water flow

in different streams. To test for environmental influences on morphological traits, we

complemented our analysis of the two Batrachophrynus species and four stream-inhabiting

Telmatobius species from central Perû with the lake-inhabiting Titicaca frog Telmatobius

culeus (Garman, 1875) from southern Perü.

It is obvious that the validity of the central Peruvian taxa of Telmatobiinae needs to

be confirmed by a thorough investigation. The definition of most taxa is exclusively based

on morphological and morphometric characters (e.g. VELLARD, 1951, 1953, 1955) without

an appropriate assessment of the intraspecific variation (TRUEB, 1979). Currently, the

identification of most species is only possible by comparison with type specimens.

Consequently, our investigation evaluates VELLARD’S type material in MHNSM (Museum

of Natural History “Javier Prado”, Lima, Perü) in comparison with the preserved

specimens in URP (Museum of Natural History of the Ricardo Palma University, Lima,

Perü) and with specimens collected in the field in 1992. We use cluster analyses to detect

intraspecific morphological traits within the data sets of phenotypes described by

morphometric characters. At the species level we widely follow the procedures used by

WIENS (1993) in his recent taxonomic revision of the Telmatobiinae from northern Perü.

Principal component and discriminant analyses are applied to quantify the differentiation

among the seven nominal species. Finally, the differences among the genera and the

ecotypes (stream- versus lake-inhabitants) are surveyed in discriminant analyses. The aims

of our study are to provide an objective basis for the identification of central Peruvian

Telmatobiinae and to distinguish convergent morphological traits from morphometric

features which characterize species and/or genera.

Source : MNHN, Paris

ALYTES 13 (1)

Source : MNHN, Paris

SINSCH, SALAS & CANALES 17

Fig. 1. — Batrachophrynus brachydactylus (A, dorsal view: E, ventral view); B. macrostomus (B, F):

Telmatobius jelski (C, G}; T. rimac (D, H).

Source : MNHN, Paris

18 ALYTES 13 (1)

MATERIAL AND METHODS

The material examined included a total of 280 adult frogs pertaining to the seven

nominal species Batrachophrynus brachydactylus (figs. 1A, 1E), B. macrostomus (figs. 1B,

1F), Telmatobius brevirostris, T. carrillae, T. culeus, T. jelskii (figs. 1C, 1G) and T. rimac

(figs. 1D, 1H). The frogs had been collected at 32 localities (fig. 2). The assignment of

specimens to subspecies and to localities is listed in Appendix I, the number of individuals

studied and the sex-specific size are given in Table I.

Standard morphometric measurements (nearest 0.1 mm) include: (1) snout-vent length

(SVL); (2) height of body at the pectoral girdle (BH); (3) maximum width of head

(HWID); (4) eye diameter (EYE); (5) interorbital distance (10D); (6) eye-nostril distance,

from anterior margin of eye to posterior edge of naris (ENOSE); (7) distance between the

eye and the tip of the snout (ESNOUT); (8) humerus length (HUML); (9) radioulnar

length (RADL); (10) hand length (HNDL); (11) length of the third finger (FG3L); (12)

femur length (FEML); (13) tibia length (TIBL); (14) foot length, from union with tibia to

the tip of fourth toe (FOOTL); (15) length of first toe (TOEIL); (16) length of fourth toe

(TOEA4L); (17) length of callus internus (CIL); (18) maximum length of toe web (WEBL).

Multivariate analyses were performed on log;,-transformed data (BOOKSTEIN et al.,

1985) and morphometric ratios. The empiric measurements were transformed to ratios

(range: 0-1) by calculating measures relative to SVL (SCHNEIDER et al., 1992, 1993).

Moreover, two indices were used for further analysis: CIL/TOEIL and FEML/TIBL.

Table II gives the means (and corresponding standard deviation) of these 19 relative

measures for each species.

Due to the low number of individuals assigned to type material, we analysed the

morphometric similarity between individuals in a cluster analysis using hierarchical

grouping of the SVL-standardized phenotypes (WARD, 1963). This procedure subsequently

reduces the number of groups by joining that specimen to another one or to a cluster

which originates the lowest error sum of square. The result is a dendrogram based on

phenetic similarity. As the proposals of subspecies by VELLARD were exclusively based on

external morphology, valid subspecific taxa are expected to form homogeneous clusters.

At the species level, sets of the log,,-transformed data were subjected to principal

component analysis to explore the morphometric variability independent of taxonomic

assignment. Principal components (PC) are linear combinations of the measured variables,

uncorrelated with each other and explaining the maximum amount of variation. The first

principal component (PCI) of morphometric data generally describes differences in size,

but size effects may be present in subsequent principal components (HUMPHRIES et al.,

1981). Techniques such as shearing have been developed to correct PC2 and PC3 for

possible size effects (BOOKSTEIN et al., 1985), but they are controversial and size effects may

still persist (ROHLF & BOokSTEIN, 1987). Moreover, we applied canonical discriminant

analysis to distinguish between the taxonomic groups delimited a priori. The resulting

discriminant functions (CAN — canonical variables) are linear combinations of the

measured variables that maximize the differences between the groups. Discriminant

Source : MNHN, Paris

SINSCH, SALAS & CANALES 19

Rio Huallaga

TROPICAL

LOWLANDS

C Le

po patviles

jus

go 27

Rio mise

2 Q

& Q

NORTH Zn ï }

vin

FF © Rio Pampas

A |

# MN

st =

te de

PACIFIC OCEAN

Rio Acari ren

Rio Vauca Parinacochas|

Fig, 2. — Distribution of the central Peruvian Telmatobiinae: Batrachophrynus brachydactylus

(inverted triangle); 8. macrostomus (open triangle); Telmatobius brevirostris (* ); T. carrillae (+;

T. jelskät (circles); T. rimac (dots). Localities are approximated from distances by roads; multiple

localities in close proximity are represented by a single symbol. The main Andean river systems

are indicated.

Source : MNHN, Paris

Table 1. - Morphometric data for Batrachophrvnus brachydactylus, B. macrostomus. Telmatobius brevirostris, T. carrillae, T. culeus, T. jelskit and T.

rimac. The first line is mean + 1 SD; second line is range. All values are in millimeters: see text for abbreviations of variables.

B. brachydacrylus B. macrostomus T. brevirostris T. carrillae T. culeus T. jelskit

Character

N=23 | N=6 | N=7 | N-2 | N N=32 | N=21 | N=19 | N=23 | N=37 =35

sv |s65+63|s73:56l132004+s80h130+17) 577 504 |s33247/ 70237034 2160052235] s44441[ 58454 E Es

477-719 | 449-694 117.5-141[117.7-1701 57.0-58.3| 504-763 | 35.3-524| 412-550 570-1164] 50.7- 119,0) 46.7-619| 439-712 | 420-572 | 470-869

BH |111422/106+29/2304246/234+48| 164 17.1 +15/109212/262+47|279+92|1364#14|139416|118416| 143 +34

71-150 | 74-179 - 183-330/ 152-175 130-232] 60-140 | 83-127 | 163-350/130-412| 102-159] 112-164] 86-147 | 108-23.5

Hwin [i2isio|174+16|508+47|554+93| 187 218 |is3+1s| ir +12) 369 +77/366 + 11.4] 19.6 + 1.8] 202 + 23| 16.6 + 1.4] 20.0 + 50

150-243/ 152-205] 461-574 | 464-753| 174-200 | 170-301 | 104-162| 115-170 168-517] 15.7-23.3| 151-257] 13.8-183| 144-316

EVE |49+07|47406|82+14|76+07| 58 62 |38+07| 3940 68+12| 53408 | 55408 | 54406 | 57409

32-68 | 37-62 | 60-100 | 66-88 | 58-58 | 51-83 | 27-61 | 30-56 45-83 | 42-70 | 35-70 | 46-69 | 48-81

1oD |1374#10/133+09|3084+24/320+48| 164 164 |108+11|110+08 a6454/1454+16/188414/142#11/1594+28

17-156| 117-152] 27.7-330| 284-425 | 159-169 | 144-204] 74-134 | 91-126 120-28.7| 114-202] 11.0-17.7 3.8

ENOsE | 87413| 83206 |182+14/1924226| 95 100 |69+13|68+05 127428] 9.0 +0.7 | 91408

6-139 | 67-04 |162-198|168-28| 90-100 | 83-133 | 55-129 | 56-77 80-167 | 80-107 | 75-110

EsNouT|117+10|115+09|274+19|283+37| 137 41 [94208 |96205 189 +43] 129 4 09] 130 4 1.0

4.138 | 96-134 | 249-207|250-362| 125-148 | 117-186] 73-113 | 89-108 117-230 | 114-153] 108-146

HUML [154423 142+17/434+25|4884+70| 174 173 |108416| 99416 272476/156+21| 15.4 + 24

10-19,5|10.5-16.7| 448-51.7| 422-628 | 167-180| 160-195] 82-141 | 77-144 | 177-377| 137-385 | 113-108

RADL |115+12/108412|342431|3464+43| 130 133 |94+12/1014#12|/216+40|203+58|124+14

97-144 | 82-127 | 297-373] 308-420| 128-131 | 105-177] 74-125 | 84-127 | 140-287| 101-273] 97-155

anoL |ussiilitiæiol38s+23/390451| 150 162 5 240+43/234+61/136+10| 141410

91-139 | 94-131 | 361-425 361-510 | 148-152| 146-190 152-310/129-33.7| 11.6-16.3| 123-173

FGL | 63407| 64209 [251416245431] 94 10.9 1W6+30/146+43| 8.8 +13 | 90 +09

50-80 | 50-83 214-304] 94-04 | 94-139 77-189 | 60-211 | 69-113 | 73-106 | 60-106 | 70-180

FEML | 25.5 + 3.4/ 243 + 32 GSs+40| 252 28.4 432 4 7.839.8 + 10.5| 24,8 + 24

204-348 | 195-327 578-680| 244-259 | 250-315 268-536 | 18.3-52.4 | 18.7- 29.0

TIBL 28426 52465| 273 281 44.5 + 80414 + 10.9] 24.5 + 1.5] 24.2 + 1.6] 25.1 + 1.5] 27.6 + 43

189-294 534-730 | 258-288 | 241-350 271-568 | 204-575] 214-268 | 205-270 | 210-283| 214-368

FOOTL 91467] 413 46 |316+38| 329 + 30|710 + 12.9/67.2 + 171 38.1 4 28| 43.8 4 70

910-1123] 403-422| 383-53.7| 25.5-40.5| 293-410 | 46.2-864 | 35.3-00,0| 290-423 328-427] 372-624

TOEIL +16/175424| 64 66 |44405| 50407 |102422| 98430 58406 | 68412

136-183 6| 54-73 | 55-86 | 36-55 | 37-60 | 58-134 | 50-144 43-67 | 54-103

TOESL ss+20|c8+32| 273 292 |206+23|202+19|477+8.8|45.6 + 12.1 29.9 +49

556-69|s60-652| 255-290 265-336 176-265 | 194-267 | 202-577] 251-677 244-417

C1 46+07/452+00| 36 34 22403/ 4541043212

37-52 | 34-55 | 33-38 | 32-36 17-31 | 27-65 | 26-65

WEBL 31/271439| 76 50 78210 |19.1444| 18.1 46.5

M5-336|222-345 83

101 | 9.5-25.2 | 7.3-274

(D €T SALATV

Source : MNHN, Paris

Table IL. - Ratios of morphometric data for Batrachophrynus brachydactylus, B. macrostomus, Telmatobius brevirostris, T. carrillae, T. culeus, T.

jelskii and T. rimac. Data are given as mean + 1 SD. See text for abbreviations of variables.

ne B. brachydactylus | B. macrostomus | T. brevirostris T. carrillae T. culeus T. jelskit T. rimac

atio

N=53 N=13 N=5 N=53 N =42 N=72 N=4

BH/SVL 0.194 + 0.047 | 0.176 + 0.025 | 0.285 + 0.023 | 0.231 + 0.020 | 0.292 + 0.032 | 0.252 + 0.028 | 0.240 + 0.029

HWID/SVL 0.314 + 0.026 | 0.404 + 0.024 | 0.347 + 0.034 | 0.302 + 0.021 | 0.397 + 0.037 | 0.364 + 0.026 | 0.337 + 0.026

EYE/SVL 0.085 + 0.014 | 0.060 + 0.007 | 0.103 + 0.004 | 0.085 + 0.012 | 0.076 + 0.009 | 0.100 + 0.015 | 0.106 + 0.012

IOD/SVL 0.240 + 0.021 | 0.238 + 0.010 | 0.281 + 0.011 | 0.242 + 0.019 | 0.240 + 0.014 | 0.270 + 0.026 | 0.282 + 0.022

ENOSE/SVL 0.151 + 0.015 | 0.143 + 0.005 | 0.167 + 0.006 | 0.152 + 0.022 | 0.138 + 0.012 | 0.167 + 0.014 | 0.170 + 0.013

ESNOUT/SVL | 0.206 + 0.016 | 0.212 + 0.007 | 0.236 + 0.013 | 0.210 + 0.014 | 0.211 + 0.018 | 0.238 + 0.015 | 0.243 + 0.016

HUML/SVL 0.262 + 0.034 | 0.373 + 0.020 | 0.299 + 0.027 | 0.233 + 0.037 | 0.306 + 0.025 | 0.285 + 0.043 | 0.280 + 0.028

RADL/SVL 0.195 + 0.022 | 0.262 + 0.017 | 0.227 + 0.014 | 0.215 + 0.017 | 0.227 + 0.015 | 0.226 + 0.023 | 0.218 + 0.021

HNDL/SVL 0.199 + 0.021 | 0.299 + 0.017 | 0.270 + 0.020 | 0.217 + 0.027 | 0.257 + 0.017 | 0.255 + 0.023 | 0.260 + 0.022

FG3L/SVL 0.113 + 0.015 | 0.189 + 0.017 | 0.176 + 0.012 | 0.128 + 0.017 | 0.158 + 0.019 | 0.163 + 0.025 | ‘0.166 + 0.024

FEML/SVL 0.440 + 0.038 | 0.481 + 0.039 | 0.468 + 0.062 | 0.445 + 0.037 | 0.452 + 0.032 | 0.453 + 0.043 | 0.475 + 0.039

TIBL/SVL 0.410 + 0.035 | 0.441 + 0.019 | 0.475 + 0.021 | 0.457 + 0.027 | 0.467 + 0.034 | 0.448 + 0.033 | 0.495 + 0.037

FOOTL/SVL | 0.640 + 0.047 | 0.729 + 0.039 | 0.741 + 0.051 | 0.713 + 0.042 | 0.751 + 0.062 | 0.695 + 0.070 | 0.763 + 0.044

TOEIL/SVL 0.088 + 0.010 | 0.127 + 0.010 | 0.110 + O.011 | 0.104 + 0.013 | 0.108 + 0.013 | 0.089 + 0.011 | 0.117 + 0.012

TOEA4L/SVL 0.418 + 0.030 | 0.458 + 0.033 | 0.489 + 0.047 | 0.469 + 0.029 | 0.506 + 0.042 | 0.454 + 0.041 | 0.519 + 0.047

CIL/SVL 0.041 + 0.009 | 0.035 + 0.006 | 0.060 + 0.009 | 0.049 + 0.006 | 0.047 + 0.007 | 0.047 + 0.011 | 0.051 + 0.008

WEBL/SVL 0.126 + 0.026 | 0.214 + 0.027 | 0.101 + 0.053 | 0.159 + 0.027 | 0.198 + 0.036 | 0.187 + 0.072 | 0.191 + 0.087

CIL/TOEIL 0.472 + 0.107 | 0.275 + 0.050 | 0.555 + 0.123 | 0.484 + 0.092 | 0.440 + 0.066 | 0.536 + 0.144 | 0.442 + 0.056

FEML/TIBL 1.072 + 0.074 | 1.092 + 0.084 | 1.002 + 0.090 | 0.975 + 0.075 | 0.972 + 0.074 | 1.013+ 0.080 | 0.960 + 0.058

STIVNVO ®@ SVIVS ‘HOSNIS

Source : MNHN, Paris

2 ALYTES 13 (1)

functions were derived from the log,,-transformed data and from the ratios. The degree

of separation of taxa was almost identical in both analyses and therefore we present the

results using the log,,-transformed data only (analogous to WiEns, 1993).

AIl calculations were performed on a PC using the FORTRAN77 program

CLUSTER and the program package STATGRAPHICS, version 5.5.

RESULTS

INTRASPECIFIC VARIATION

Among the seven nominal species studied in this paper, four are thought to segregate

in subspecies: Telmatobius brevirostris, T. culeus, T. jelskii and T. rimac (VELLARD, 1951,

1953, 1955). We approached the problem of morphometric variation within a nominal

species by applying cluster analysis on the phenotypes of all conspecific individuals

available, including those which VELLARD assigned as type material for the proposed

subspecies. Phenotypes are described by 19 SVL-standardized morphometric ratios to

minimize size effects.

The phenograms of all species (including those without subspecific segregation)

revealed existence of different intraspecific morphological traits documented by groups of

specimens which joined to the same cluster with an error sum of squares of less than 0.1

(gs. 3-6, data on Batrachophrynus and T. carrillae not shown). We treat these groups of

remarkably similar specimens as “morphs”, without intending a taxonomic implication.

Generally, intraspecific morphs did not reflect polymorphism related to sex or size.

Telmatobius brevirostris Vellard, 1955

VELLARD (1955) recognized three subspecies: Telmatobius b. brevirostris, T. b. parvulus

and T. b. punctatus. Unfortunately, we only found in the MHNSM collection two

specimens of the first two subspecies, and one of the third. The original descriptions were

based on only three adults of T. b. brevirostris, six of T. b. parvulus and two of T. b.

Punctatus — à prohibitively small sample size to define reliable subspecific taxa. The

phenogram (fig. 3) of the five specimens available for analysis shows that: (1) the two type

specimens of T. b. brevirostris from Chasqui join different clusters; (2) the morphometri-

cally most similar specimens are one T. b. brevirostris individual and the T. b. punctatus

male from Santa Maria del Valle; (3) the two individuals of T. b. parvulus from Caina are

very alike, but form together with one T. b. brevirostris a main cluster with a error sum

of squares far below the level of morph distinction. AIl specimens join to one group at an

error sum of squares of only 0.146. Neither the association of the five phenotypes to

groups nor the degree of morphological differentiation between them support a subspecific

segregation within the nominal species T. brevirostris.

Source : MNHN, Paris

SINSCH, SALAS & CANALES 23

Chasqui 1

Sta. Maria

Caina 1

Caina 2

Chasqui 2

PE UN PET CRE TRE

0.150 0.100 0.050 0.000

error sum of squares

Fig. 3. — Phenogram of morphometric similarity between adult Telmatobius brevirostris which

VELLARD (1955) assigned to different subspecies. Each specimen is identified by the locality of

collection and an individual number. Similarity is based on 19 ratios and computed by

hierarchical grouping in a cluster analyis. T. b. brevirostris: Chasqui, T. b. parvulus: Caina;

T. b. punctatus: Santa Maria del Valle.

Telmatobius culeus (Garman, 1875)

VELLARD (1953) recognized six subspecies: Telmatobius c. culeus, T. c. dispar

(redescribed by LAVILLA, 1988b), T. c. escomeli, T. c. exsul, T. c. fluviatilis and T. c.

lacustris. The phenogram (fig. 4) calculated for 42 adults assigned as type material

(MHNSM) reveals that: (1) there exist five morphs within the nominal species; (2) these

morphs do not coincide with any of the subspecies proposed by VELLARD; (3) different

morphs occur at the same locality. Joining all specimens to one group causes an error sum

of square of 0.453, that is three times greater than in T. brevirostris. Again, from the

morphometric point of view there is no evidence that the morphological traits within T.

culeus agree with the proposed subspecific differentiation.

Telmatobius jelskii (Peters, 1873)

VELLARD (1955) recognized four subspecies: Telmatobius j. jelskü, T. j. bufo

(redescribed by LAvILLA, 1988b), T. j. longitarsis and T. j. walkeri. The phenogram (fig. 5)

of 72 adults including VELLARD’Ss type specimens (MHNSM) shows that: (1) there exist

seven morphs within the nominal species; (2) these morphs do not coincide with any of the

subspecies proposed by VELLARD; (3) different morphs inhabit the same locality. Joining

all specimens to one group causes an error sum of squares of 1.444, that is three times

greater than in T. culeus or in T. rimac and ten times greater than in T. brevirostris.

Nevertheless, the occurrence of different morphs at all sites — though in different

frequencies — does not support the validity of the four subspecies proposed by VELLARD.

Telmatobius rimac Schmidt, 1954

VELLARD (1955) recognized two subspecies: Telmatobius r. rimac and T. r. meridio-

nalis. The phenogram (fig. 6) of 42 adults including VELLARD’s type specimens shows that:

(1) there are five morphs; (2) the type material from Ocros and Tupe forms one of two

Source : MNHN, Paris

24 ALYTES 13 (1)

Isla del Sol 1

Ocama 3

Lagunillas 1

Ocama 2

Ocema 1

lave 2

Lagunillas 2

Chécayani 2

Umayo

Julinca 2

Juliaca 3

lave 5

Lagunillas 7

Lagunilles 10

Tave 1

Juliaca 6

Lagunillas 8

Ocama 6

Chuculto 1

Conta 1

Gueule 2

Lagunillas

0.453 laguniles ©

move 4

Iave 3

Hunyllata 1

Huayllata 5

Lagunillas 5

Lagunillas 3

Hunyllata 2

QE als PRES PRELS pl

.200 0.150 0.100 0.050 0.000

error sum of squares

Fig. 4. — Phenogram of morphometric similarity between adult Telmatobius culeus which VELLARD

(1953) assigned to different subspecies. Presentation of data analogous to that in fig. 3. T. c.

culeus: Ocama, Ilave, Isla del Sol; 7. c. dispar: Coata, Juliaca; T. c. escomeli Lagunillas; T. c.

exsul: Yura; T. c. fluviatilis: Chucuito, Huayllata, Ilave; T. c. lacustris: Checayani, Umayo.

main clusters including two mixed morphs; (3) the more recently collected material from

Obrojillo and Huaytara (situated at the same distance to Ocros in the north and to Tupe

in the south) forms the second main cluster and segregates into three more morphs; (4)

only one individual (Obrojillo 3) directly joins the cluster of the type specimens. This

unexpected structure of the phenogram does not support the validity of the proposed

subspecies, but demonstrates suspicious differences between the external morphology of T.

rimac which were collected forty years ago and those presently inhabiting the streams of

the Pacific slope of the Andes.

INTERSPECIFIC VARIATION

In order to compare the intraspecific variation with the morphometric differences

among the nominal species, we performed principal component analysis and discriminant

analysis on two groups of geographically neighbouring species: (1) the northern Tel-

Source : MNHN, Paris

SINSCH, SALAS & CANALES 25

pri

0.432 El

Palin 6

sus 10

0.250 0.200 0.150 0.100 0.050 0.000

error sum of squares

Fig. 5. — Phenogram of morphometric similarity between adult Telmatobius jelskit which VELLARD

(1955) assigned to different subspecies. Presentation of data analogous to that in fig. 3. 7. j

jelskii: Acolla; T. j. bufo: Tambo: T. j. longitarsis: Huancavelica, Puquio: T. j. walkeri: Ayacucho

Unclassified T. jelskii were collected in Palian, Cuyrohuasi and Parinacochas.

Source : MNHN, Paris

FOPawRom Na" +

eOTT

LÉ

ÉÉÉÉLEU

0.547 users Le

(D) € SALATV

SE SET ES

0.200 0.150 0.100 0.050 0.000

error sum of squares

Fig. 6. — Phenogram of morphometric similarity between adult Te/matobius rimac which VELLARD

(1955) assigned to different subspecies. Presentation of data analogous to that in fig. 3. T.r.

rimac: Ocros; T. r. meridionalis: Tupe. Unclassified T. rimac were collected in Obrojillo and

Huaytara.

Source : MNHN, Paris

SINSCH, SALAS & CANALES 27

matobiinae including Batrachophrynus brachydactylus, B. macrostomus, Telmatobius

brevirostris and T. carrillae; (2) the southern Telmatobius species T. culeus, T. jelskii and

T. rimac. The main reason for the subdivision of the complete data set on phenotypes into

groups of three and four species, respectively, was to reduce the number of significant

canonical variables to two, which permits the distinction of these taxa in two-dimensional

scatter plots (figs. 7-8). Here, we present only the results based on log,,-transformed data

because the separation of taxa by discriminant analysis was almost identical in data sets

of 18 log,,-transformed distances and in those consisting of 19 morphometric ratios.

Generally, the interspecific differences in size (PCI) by far exceeded those in shape

(PC2, PC3). The size effects on PC2 and PC3 appeared to be small, because shearing did

not notably improve the separation of taxa. Discriminant analysis led to an almost optimal

separation of species by combining differences in size and shape.

In the northern group of central Peruvian Telmatobiinae, the first three principal

components explained 95.0 % of the total variance. PCI distinguishes the large B.

macrostomus from the smaller three species. The plot of PC2 and PC3 scores (fig. 7A)

shows a wide overlap between B. macrostomus, B. brachydactylus and T. carrillae, whereas

PC3 distinguishes four of the five T. brevirostris from the other species. An almost

complete separation of the four taxa was obtained by discriminant analysis, only 2 out of

53 B. brachydactylus were confounded with T. carrillae (fig. 7B, Table III). The separation

of taxa is mainly based on size (SVL) and interorbital distance (I0D) in CANI, and on

size (SVL), head shape (HWID, EYE, IOD) and humerus length (HUML) in CAN2.

In the southern group of Telmatobius species the first three principal components

accounted for 92.9 % of the total variance. PCI distinguishes T. culeus from the other two

species. The plot of principal component scores (fig. 8A) shows a considerable separation

of T. jelskii from T. rimac based on PC3, but a complete overlap of T. culeus with both

other species. An almost complete separation of the three taxa was obtained by canonical

discriminant analysis, only 1 out of 42 T. culeus was confounded with T. jelskii, and 1 out

of 72 T. jelskii with T. rimac (fig. 8B, Table IV). The separation of species is mainly based

on size (SVL, BH) and head shape (ENOSE, ESNOUT) in CANI, and on size (SVL), head

shape (HWID, ESNOUT) and extremity length (HANDL, TIBL) in CAN.

INTERGENERIC VARIATION

The next step of analysis concerns the morphometric features of each genus and the

possibility to identify genus-specific morphometric characters or ratios. The data of the

two Batrachophrynus species form one group, those of the five Te/matobius taxa the other.

Again, log,,-tranformed data and ratios provided the same degree of group separation. A

highly significant discriminant function was obtained which correctly classifies 80 % of the

Batrachophrynus and 96 % of the Telmatobius (log, ,-transformed data, Table V). Distinc-

tive morphometric ratios (ANOVA, P < 0.01) were BH/SVL (fig. 9A) and FEML/TIBL

(fig. 9B).

Source : MNHN, Paris

Principal Component 3

LU]

(e]

e Batrachophrynus brachydactylus o Telmatobius brevirostris

a Batrachophrynus macrostomus à o Telmatobius carrillae

N+2

ne)

7 0

je]

—4

=2 1 0 +1 +2 +3 6 () +6 +12 +18

Principal Component 2 Canonical Variable 1

Fig. 7. — Plot of (A) principal component scores and (B) discriminant function scores of the four

northern telmatobiine species: Batrachophrynus brachydactylus, B. macrostomus, Telmatobius

brevirostris and T. carrillae. Discriminant functions and classification success are given in

Table III.

(D €I SALATV

Source : MNHN, Paris

e Telmatobius culeus o Telmatobius rimac

a Telmatobius jelskii

+3 +6

Ce]

? ax+T4+

eo] (

g +1 = g

ê É+2 4

Ê S ê

Q O0 A

O 4 >

% à Ô ë

& S æ

e £ o

£ ec) >

Dal Z

ä 9 _2 £

_ È

—2 3

—3 —4

—3 —e 1 O0 EL +2 +3 —4 —2 O0 +2 +4 +6

Principal Component 2 Canonical Variable 1

Fig. 8. — Plot of (A) principal component scores and (B) discriminant function scores of the three

southern telmatobiine species: Telmatobius jelskii, T. rimac and T. culeus. Discriminant functions

and classification success are given in Table IV. &

Source : MNHN, Paris

Table II. - Discri

ALYTES 13 (1)

minant functions to distinguish among Batrachophrynus

macrostomus, B. brachydactylus, Telmatobius brevirostris and T. carrillae based

on 18 logi, transformed morphometric characters.

A. Statistical significance

: Canonical Wilks ; Degrees of

32.10 0.985 0.0042 613.5

43 0.842 0.1383 221.5

Lil 0.725 0.4741 583.6

B. Unstandardized discriminant function coefficients

Coefficients

Character

CAN 1 CAN2 CAN3

- 11.64 18.89 - 6.59

- 3.60 - 0.10 4.81

2.91 17.56 - 9.19

- 6.78 10.25 2.81

25.39 -1221 14.51

-442 -4.49 - 10.26

6.79 2775 7.84

1.64 11.01 0.92

5.77 - 7.09 - 3.67

6.99 -5.13 - 1.41

6.92 - 1.94 6.82

9.23 6.85 -4.21

5.42 - 8.20 6.23

- 8.38 121 -4.85

4.57 -7.16 1.60

2.50 - 8.12 7.13

-2.14 - 0.50 3.27

- 1.6 - 3.84 - 9.81

Constant - 25.08 -13.17 0.75

C. Classification success

Predicted group

Actual group Batrachophrynus | Batrachophrynus | Telmatobius Telmatobius

brachydactylus | macrostomus | brevirostris carrillue

B. brachydactylus | 51 (96%) 0 0 24%)

B. macrostomus 0 13 (100%) 0 0

T. brevirostris 0 0 5 (100%) 0

T. carrillae 0 0 0 53 (100%)

D. Group centroids

Species CAN I CAN2 CAN3

B. brachydactylus -0.71 1.75 oi |

B. macrostomus 15.59 - 0.98 - 0.58

T brevirostris -341 - 1:49 - 0.17

T. carrillae 3.17 -0.24 5.02

Source : MNHN, Paris

SINSCH, SALAS & CANALES

Table IV. - Discriminant functions to distinguish among Telmatobius culeus, T. jelski

and T. rimac based on 18 log;, transformed morphometric characters.

A. Statistical significance

: Canonical Wilks : Degrees of

Eigenvalue | Correlation Lambda | Chisquared | “féedom L

0.0416 440.3 36 < 0.00001

694 181.7 17 < 0.00001

Coefficients

Constant

C. Classification success

Predicted group

RERALEroNP Telmatobius culeus

Telmatobius jelskii

Telmatobius rimac

Telmatobius culeus 41 (98%)

Telmatobius jelskii 0

Telmatobius rimac 0

D. Group centroids

Species

Telmatobius culeus

Telmatobius jelskii

Telmatobius rimac

1(2%) 0

71 (99%) 1 (1%)

0 42 (100%)

CAN2

- 0.06

- 1.42

2.46

Source : MNHN, Paris

ALYTES 1

3 (1)

Table V. - Discriminant functions to distinguish the genera Batrachophrynus and

Telmatobius based on 18 logio transformed morphometric characters.

A. Statistical significance

Canonical

Eigenvalue

correlation

hi-squared

Degrees of

freedom

1.63 0.788

254.9

B. Unstandardized discriminant function coefficients

Coefficients

Character

CAN 1

Constant

C. Classification success

BRSBCISRIRRGAISSS

RENTE ENS

Predicted group

Actual group

D. Group centroids

Genus

Batrachophrynus Telmatobius

53 (80%) 13 (20%)

8 (4%) 200 (96%)

Batrachophrynus

Telmatobius

Source : MNHN, Paris

STIVNVI ®@ SVIVS ‘HOSNIS

B.b. B.m. T.br. T.ca T.cu T.j Tr.

Fig. 9. — Box- and whisker-plot of morphometric ratios which permit the distinction between the

genera Batrachophrynus and Telmatobius. (A) BH/SVL; (B) FEML/TIBL. The dotted line

represents the means of each genus. B.b.: Batrachophrynus brachydactylus; B.m.: B. macrostomus;

T.b.: Telmatobius brevirostris; T.ca.: T. carrillae; T.cu.: T. culeus; T.j.: T. jelskü, T.r.: T. rimac.

£E

Source : MNHN, Paris

34 ALYTES 13 (1)

VARIATION BETWEEN ECOTYPES

Finally, we examined the relationship of morphological traits caused by the

convergent adaptation to a specific type of habitat (stream/lake) with the phylogenetic

relationships between the seven taxa. For this purpose, we pooled the data of B.

macrostomus and T. culeus forming the lake group and compared them with the stream

group formed by the remaining species. Again, there was no difference between the

separation of groups based on log,,-transformed data and that based on ratios. The highly

significant discriminant function correctly classifies 100 % of the stream-inhabitants and

92 % of the lake-inhabitants (log, ,-transformed data, Table VI). Distinctive morphometric

ratios (ANOVA, P < 0.01) between lake- and stream-inhabitants are HWID/SVL (fig.

10A) and EYE/SVL (fig. 10B). Moreover, SVL of adult lake-inhabitants is considerably

larger than that of stream-inhabitants (Table I).

DIsCUSSION

The general similarity in size and shape of the riparian central Peruvian Telmatobii-

nae does not facilitate a reliable and objective definition of taxa. TRUEB (1979) stated that

most taxonomic descriptions of Telmatobius are inadequate because many proposed

diagnostic characters are so subjective that an identification without comparison with type

material is almost impossible. A recent study on the Telmatobius species of northern Perû

used for the first time multivariate statistics on morphometric data to obtain objective

criteria for the classification of telmatobiine frogs (WiENs, 1993). We followed this

approach and successfully applied cluster, principal component and discriminant analyses

to assess the morphometric variation within and among the central Peruvian taxa of

Telmatobiinae. Consequently, we offer discriminant functions based on external characters

which permit the identification of adults with a very low rate of erroneous classifications.

Moreover, we identified convergent morphological traits distinguishing riparian and

lake-inhabiting telmatobiine frogs which in the future should be avoided for taxonomic

conclusions.

REASSESSMENT OF THE SUBSPECIFIC TAXA

Within wide ranging species such as T. jelskii and T. rimac we are confronted with the

problem of interpopulational morphological differentiation, due to the partial geographi-

cal isolation between different hydrographic systems and valleys. VELLARD (1951, 1953,

1955) attempted to solve this problem by naming more than half of the populations

studied up to 1955 at the subspecific level. However, the existence of a complete series of

intermediate specimens (VELLARD, 1955) between all subspecies emphasizes the more or

less arbitrary nature of their definition. Nevertheless, due to VELLARD’s proposal, the

segregation of T. brevirostris, T. culeus, T. jelskii and T. rimac into a total of fifteen (!)

subspecies is still recognized (FRosr, 1985).

Our attempt to deal with interpopulational variability in taxonomic terms is based on

the hierarchical grouping of individual phenotypes according to their similarity in external

Source : MNHN, Paris

SINSCH, SALAS & CANALES 35

Table VI. - Discriminant function to distinguish between the stream- and lake-

inhabiting species of the genera Batrachophrynus and Telmatobius based on 18

logo transformed morphometric characters.

A. Statistical significance

Canonical

correlation

Degrees of

freedom

Eigenvalue

Chi-squared

4.58 0.906 452.1

C. Classification success

Predicted group

Actual group

Stream-inhabitant Lake-inhabitant

Stream-inhabitant 224 (100%) 0

Lake-inhabitant 4(8%) 46 (92%)

D. Group centroids

Genus

Stream-inhabitant - 1.01

Lake-inhabitant

Source : MNHN, Paris

0.50

0.08

0.04

A HWID/SYL

B.m. T.cu. B.b. T.br. T.ca. T.j.

Fig. 10. — Box- and whisker-plot of morphometric ratios which permit the distinction between

inhabitants of lakes and of streams. (A) HWID/SVL; (B) EYE/SVL. Presentation of data

analogous to fig. 9.

T.r.

9€

(D €T SALATV

Source : MNHN, Paris

SINSCH, SALAS & CANALES 37

morphology. To avoid major biasses due to size, the phenotypes are described by 19

unweighted morphometric ratios. The phenotypic grouping in a cluster analysis permits an

objective assessment of similarity between conspecific individuals, even if the sample size

is small as usually for type material. Principal component and discriminant analyses, in

contrast, require about 20 cases per predictor which imply sample sizes exceeding by far

the numbers of type specimens.

VELLARD’S proposal of subspecies is exclusively based on differences in the external

morphology of specimens from different localities. Therefore, if the proposed subspecies

were valid, the phenograms of conspecific individuals collected from different localities (=

reproductively isolated populations) should show the following structure: (1) specimens

pertaining to the material originating from one type locality should form a homogeneous

group (cluster); (2) material assigned to different subspecies should be represented in

different clusters; (3) conspecific specimens originating from one population without their

own subspecific status should join as a group one cluster formed by type specimens.

The phenograms obtained for T. brevirostris, T. culeus, T. jelskii and T. rimac do not

show structures compatible with VELLARD’s taxonomic suggestions. The morphometric

differentiation between the five specimens of T. brevirostris which VELLARD assigned to

three subspecies is low and the most similar pair of individuals belongs to different

subspecies (fig. 3). The greatest morphometric difference between the 42 T. culeus exceeds

three times that between the T. brevirostris specimens but none of the five morphs

identified within this data set coincides with any of the proposed subspecies (fig. 4). The

same applies to the seven and five morphs, respectively, found within the T. jelskii (fig. 5)

and T. rimac (fig. 6) data sets. As objective morphometric similarity does not correspond

to VELLARD’s subjective grouping of conspecific specimens to subspecies, we conclude that

the proposed subspecific segregation is invalid in the four species studied. This conclusion

does not rule out that the definition of subspecies may be useful to describe the speciation

processes due to the geographic isolation of populations of wide ranging species. However,

if considered useful, the definition should be based on conceivable, objective criteria better

than those presented so far.

What is the meaning of the different morphological traits (morphs = groups of

conspecific specimens which are morphometrically very similar) which we detected in all

seven species? As different morphs usually occur at the same locality, i.e. within the same

population, they probably do not represent taxonomically relevant units. Neither do they

reflect sexual dimorphism in shape, or size-specific polymorphism. The varying frequencies

of morphs at different localities indicate that the adaptive value of a morph for the

increase of individual fitness probably differs among the localities. Thus, intraspecific

polymorphism may have evolved in response to local environmental factors such as

altitude, water flow and competition. The occurrence of different morphs at the same

locality may be due to temporal changes in the local environments. This interpretation is

supported by the conspicious change of the frequencies of T. jelskii morphs which took

place in the Mantaro Valley during the last forty years: 9 out of 10 specimens collected

near Acolla in the early fifties (VELLARD, 1955), but only 1 out of 7 frogs recently collected

near Palian join the same main cluster, whereas only 1 Acolla specimen joins the remaining

6 Palian specimens. Even more impressive is the same tendency in T. rimac though

Source : MNHN, Paris

38 ALYTES 13 (1)

referring to geographically distant localities: only one recently collected individual pertains

to the morphs dominating about forty years before (VELLARD, 1955). Attributing these

changes to the different duration of preservation seems too simple because older morphs

still occur (rarely) in contemporary populations, and presently dominating morphs also

existed (rarely) forty years ago. Instead, there have been dramatic changes in the riparian

habitats of these species during this period due to the enormous increase in human

population and the resulting pollution of the rivers and streams.

In summary, we reject the proposal of subspecies for T. brevirostris, T. culeus, T.

jelskii and T. rimac.

REASSESSMENT OF THE SPECIES STATUS

The large numbers of studied specimens in most species permit a reliable estimate of

the intraspecific morphometric variability and the comparison between different species.

Principal component analysis of the log,,-transformed data did not substantially

contribute to the resolution between different taxa: the usually large overlap between

different species reflects their similarity in many aspects of shape. The only exception from

the rule is T. brevirostris which considerably differs from the neighbouring species (fig.

7A). However, WIENS (1993) obtained a similar low degree of resolution between eight

Telmatobius species from northern Perü, indicating that PCA is not a powerful tool to

distinguish between Andean Telmatobiinae.

In contrast, if groups (= species) are determined a priori, and if the differences

between them are maximized by canonical discriminant analysis, the same morphometric

data sets can distinguish between the taxa. It is noteworthy that, despite the bad image of

morphometric ratios (e.g. BOOKSTEIN et al., 1985), discriminant analysis using logio-

transformed data or ratios give very similar results and reach the same classification

success. The convergent morphological adaptation to similar environmental constraints

prevents an unequivocal identification at the level of individuals, but the rate of

erroneously classified individuals is low: 2 B. brachydactylus, 1 T. culeus and 1 T. jelskü

out of a total of 280 specimens. WIENs (1993) did not provide results of classification

success in his analysis, but he also stated that discriminant analysis provided an objective

base for the distinction between Andean species of Telmatobiinae. In summary, the six

central Peruvian species as well as the one from southern Perü proved to be well-defined

taxonomic units which possess external features allowing an objective diagnosis.

REASSESSMENT OF THE GENERIC STATUS

LAURENT (1983) recognized three genera within the central Peruvian Telmatobiinae:

the monotypic genera Batrachophrynus and Lynchophrys, and Telmatobius with four

species. However, the evidence presented for the change of Batrachophrynus brachydactylus

to the genus Lynchophrys is weak: the main differences from Batrachophrynus macrostomus

put forward are: (1) shorter third finger; (2) smaller size; (3) the statement that male B.

macrostomus lack nuptial pads. Points (1) and (2) are convergent morphological

adaptations to the stream habitat which are shared with all riparian Telmatobius, whereas

point (3) is simply an error (see figure 16 in SINsCH, 1990). Moreover, both species share

two morphometric features which are diagnostic for the genus Batrachophrynus despite

Source : MNHN, Paris

SINSCH, SALAS & CANALES 39

their adaptation to different habitats: flat body and femur length exceeding tibia length.

Finally, the analysis of allozymes of these species and of three Telmatobius species (SINSCH

& JURASKE, 1995) clearly demonstrates that B. brachydactylus and B. macrostomus are

closely related species as originally proposed by PETERS (1873).

In summary, we do not see any conceivable reason to split the genus Batrachophrynus

and reject the proposal of Lynchophrys.

SUMMARY OF TAXONOMIC PROPOSALS

We summarized in Table VII the reassessment of the taxonomic status of central

Peruvian Telmatobiinae. Three genera including seven nominal species which segregate

into 15 subspecies are currently reduced to two genera Batrachophrynus and Telmatobius

which include seven species without subspecific segregation.

Finally, we wish to comment on DUELLMAN’s (1993) statement that the reading of

Batrachophrynus macrostomus Peters, 1873 should be changed to B. microstomus. We do

not agree because there is no doubt about the original naming by PETERS (1873), and in

all research papers dealing with this species (AVILA RAMON, 1953; CAMARENA, 1953;

Dusois, 1984; FIELDSA, 1983; GORHAM, 1966; LAURENT, 1983; LAVILLA, 1988a; LYNCH,

1971, 1978; MACEDO, 1950, 1960, 1976; MORALES, 1988; SINsCH, 1986, 1990; VELLARD,

1951) the original name has been used consequently.

CONVERGENT MORPHOLOGICAL TRAITS

Stream habitats require frogs to evolve morphological adaptations which enable them

to move within the permanent current of water. Therefore, it is not surprising that the

riparian B. brachydactylus shares two diagnostic features with the riparian Telmatobius:

slim head and large eye diameter. Moreover, adults of all riparian species are smaller-sized

than those of the lake-inhabiting B. macrostomus and T. culeus. Confusion of these

convergent morphological adaptations to the same type of habitat with similar morphol-

ogy due to phylogenetic relationship has led to the creation of the genus Lynchophrys by

LAURENT (1983). In the Andean Telmatobiinae any taxonomic conclusion based on

external morphology should be backed up by other kinds of characters because the rate

of erroneous classification in the morphometric distinction of the genera Batrachophrynus

and Telmatobius is considerably greater than that in the distinction between stream- and

lake-inhabitants: 20 % versus 8 %. Thus, the effect of convergent lines of development is

probably great in this group of frogs.

GEOGRAPHICAL DISTRIBUTION

The geographical range (fig. 2) of the six central Peruvian species is still relatively

unknown. We know little about the northern extension of the ranges of T. brevirostris, T.

carrillae and T. rimac, and about the southern range of T. jelskü. The northern gap

between the species surveyed by WIENs (1993) and those in this study is subject to a recent

study (SALAS, in prep.). Further attention should be paid to the exact limits of distribution

of the riparian B. brachydactylus in relation to those of the neighbouring Telmatobius

species. We do not even know if B. brachydactylus and T. jelskiü, both present in the

hydrographic system of the Mantaro river, can coexist at the same locality.

Source : MNHN, Paris

Table VII. - Alphabetical synonymy of the telmatobine species revised in this study.

Names in use

This study

Lynchophrys brachydactyla (Peters, 1873)

Batrachophrynus macrostomus Peters, 1873

Telmatobius brevirostris Vellard, 1955

Telmatobius carrillae Morales, 1988

Telmatobius culeus (Garman, 1875)

Telmatobius jelskii (Peters, 1873)

Telmatobius rimac Schmidt, 1954

T. b. brevirostris Vellard, 1955

T. b. parvulus Vellard, 1955

T. b. punctatus Vellard, 1955

T. culeus culeus (Garman, 1875)

T. culeus dispar Vellard, 1953

T. culeus escomeli Angel, 1923

T. culeus exsul Vellard, 1951

T. culeus fluviatilis Vellard, 1953

T. culeus lacustris Vellard, 1953

T. jelskit jelskit (Peters, 1873)

T. jelskit bufo Vellard, 1955

T. jelskit longitarsis Vellard, 1955

T. jelski walkeri (Shreve, 1941)

T. rimac rimac Schmidt, 1954

T. rimac meridionalis Vellard, 1955

Batrachophrynus brachydactylus Peters, 1873

Batrachophrynus macrostomus Peters, 1873

Telmatobius brevirostris Vellard, 1955

Telmatobius carrillae Morales, 1988

Telmatobius culeus (Garman, 1875)

Telmatobius jelskii (Peters, 1873)

Telmatobius rimac Schmidt, 1954

(D ET SALATV

Source : MNHN, Paris

SINSCH, SALAS & CANALES 41

Nevertheless, there can be little doubt that the genus Batrachophrynus is endemic to

central Perü, as already mentioned in the original description and later in VELLARD (1951).

The comment on distribution in FRosT (1985) — “Andes of southern Peru and Bolivia”

— is obviously wrong, as is the citation of “Lynchophrys brachydactyla” occurring in the

northern Andes of Bolivia (DE LA Riva, 1990).

CONCLUSION

At this stage of morphometric analysis we refrain from phylogenetic considerations

because of the convergent adaptations of the external morphology to the same type of

habitat. In the next step of analysis, we use allozyme variation within and between

telmatobiine species to approach the phylogenetic relationships and to compare them with

phenetic relationships based on morphometry (SINSCH & JURASKE, 1995).

In conclusion, this paper provides an objective, diagnostic method to assign central

Peruvian Telmatobiinae to the presently known species, based exclusively on external

characters which are easy to measure. Thus, multivariate statistics, specifically canonical

discriminant analysis, have proven again to be an useful tool in the classification of

amphibians.

RESUMEN

Se revisa el estado taxonémico de los Telmatobiinae de la regiôn central del Perü, en

base a la variaciôn intraespecifica de 18 parametros morfométricos que presentan las seis

especies (con nueve subespecies) reconocidas actualmente, las cuales se agrupan en tres

géneros (Batrachophrynus, Lynchophrys y Telmatobius). Aplicando los anälises de

componentes principales y de discriminaciôn se reconocen dos géneros (Batrachophrynus

y Telmatobius) incluyendo seis especies (Batrachophrynus brachydactylus, B. macrostomus,

Telmatobius brevirostris, T. carrillae, T. jelskü, T. rimac) y ninguna subespecie. Dos

caracteres diagnôsticos y externos distinguen las especies de Batrachophrynus y de

Telmatobius, otros dos son considerados adaptaciones convergentes que diferencian a los

habitantes de arroyos de aquellos que habitan las lagunas.

ACKNOWLEDGEMENTS

We are grateful to Lic. J. CORDOVA, curator of the herpetological section of the MHNSM, Lima,

and to Dr. M. Ortiz, Director of the Natural History Museum of the URP, Lima, for permitting us

access to the Telmatobiinae of the local collections, especially to the type specimens assigned by J.

VELLARD and V. MORALES. M. ANTIGNANI, C. S. ARIAS, J. and F. TUEROS, and J. ICOCHEA helped us

to collect frogs in the field. Finally, the technical assistance of B. NiLow is acknowledged. Two

anonymous reviewers provided valuable comments on an earlier draft of this paper.

Source : MNHN, Paris

4 ALYTES 13 (1)

LITERATURE CITED

AVILA RAMON, M. N., 1953. — Biometria de Batrachophrynus macrostomus. Thesis, Univ. Nac. San

Marcos, Lim:

BOOKSTEIN, F. L., CHERNOFF, B. C., ELDER, R. L., HUMPHRIES, J. M., SMITH, G. R. & STRAUSS, R.

E., 1985. — Morphometrics in evolutionary biology. Acad. nat. Sci. Philad. special Publ., 15:

1-277.

CAMARENA, E., 1953. — Consideraciones sobre la morfologia de Batrachophrynus macrostomus.

Thesis, Univ. Nac. San Marcos, Lima.

Cm, J. M. 1986. — Speciation and adaptive radiation in Andean Telmatobius frogs. Im: F.

VUILLEUMIER & M. MONASTERIO (eds.), High altitude tropical geography, New York, Oxford

Univ. Press: 374-386.

DE LA Riva, L., 1990. — Lista preliminar comentada de los anfibios de Bolivia con datos sobre su

distribuciôn. Boll. Mus. reg. Sci. nat. Torino, 8: 261-319.

Dusois, A., 1984. — La nomenclature supragénérique des Amphibiens Anoures. Mém. Mus. natn.

Hist. nat., (A), 131: 1-64.

DuELLMAN, W. E., 1993. — Amphibian species of the world: additions and corrections. Spec. Publ.

Mus. nat. Hist. Univ. Kansas, 21: 1-372.

FyeLpsa, J., 1983. — Vertebrates of the Junin area, central Peru. Steenstrupia, 8: 285-298.

Frost, D. E., 1985. — Amphibian species of the world. A taxonomic and geographic reference.

Lawrence, Allen Press & Assoc. Syst. Collections: 1-732.

GorHaM, S. W., 1966. — Liste der rezenten Amphibien und Reptilien. Ascaphidae, Leiopelmatidea

(sic), Pipidae, Discoglossidae, Pelobatidae, Leptodactylidae, Rhinophrynidae. Das Tierreich,

85: i-xvi + 1-222.

HUMPHRIES, J. M., BOOKSTEIN, F. L., CHERNOFF, B. C., SMITH, G. R., ELDER, R. L. & Poss, S. G.,

1981. — Multivariate discrimination by shape in relation to size. Syst. Zool., 30: 291-308.

LAURENT, R. F., 1983. — Heterogeneidad del género Batrachophrynus Peters (Leptodactylidae). Acta

zool. lilloana, 37: 107-113.

LAvILLA, E. O., 1988a. — Lower Telmatobiinae (Anura: Leptodactylidae): generic diagnoses based on

larval characters. Occ. Pap. Mus. nat. Hist. Univ. Kansas, 124: 1-19.

-- 1988b. — Telmatobius (Anura: Leptodactylidae): the name-bearing types of five Vellard’s taxa.

Alytes, 7: 6-18.

LyNCH, J. D., 1971. — Evolutionary relationships, osteology, and zoogeography of leptodactyloid

frogs. Univ. Kansas Mus. nat. Hist. misc. Publ., 53: 1-238.

1978. — A re-assessment of the telmatobiine leptodactylid frogs of Patagonia. Occ. Pap. Mus.

nat. Hist. Univ. Kansas, 72: 1-57.

MACeDO, H. DE, 1950. — Anotaciones para el conocimiento zoolôgico del género Batrachophrynus

(Amphibia: Salientia). Thesis, Univ. Nac. San Marcos, Lima.

--- 1960. — Vergleichende Untersuchungen an Arten der Gattung Telmatobius. Z. wiss. Zool., 163:

355-396.

= 1976. — Aspectos ecomorfolégicos de los aparatos respiratorios y circulatorios en anfibios andinos.

Ph. D. thesis, Univ. Nac. San Marcos, Lima.

MORALES, V. R., 1988. — Una nueva especie de Telmatobius (Anura, Leptodactylidae), de Ancash,

Peru. Rev. bras. Zool., 5: 603-608.

Peters, W., 1873. — Über neue oder weniger bekannte Gattungen und Arten von Batrachiern.

Monatsb. künigl. preuss. Akad. Wiss. Berlin, 1873: 411-418.

ROHLF, F. J. & BOOKSTEIN, F. L., 1987. — A comment on shearing as a method for “size correction”.

Syst. Zool., 36: 356-367.

SCHMIDT, K. P., 1954. — Notes on frogs of the genus Telmatobius. Fieldiana: Zool., 34: 277-284.

SHREVE, B., 1941. — Notes on Ecuadorian and Peruvian reptiles and amphibians. Proc. New England

zool. Club, 18: 71-83.

SCHNEIDER, H., SINSCH, U. & NEVO, E., 1992. — The lake frogs in Israel represent a new species. Zool.

Anz., 228: 97-106.

SCHNEIDER, H., SINSCH, U. & SorIANIDOU, T., 1993. — The water frogs of Greece: bioacoustic

evidence for a new species. Z. syst. Zool. Evol.-forsch., 31: 47-63.

Source : MNHN, Paris

SINSCH, SALAS & CANALES 43

Sinscn, U., 1985. — Die Reproduktionsbiologie eines bachbewohnenden Frosches, Telmatobius

jelski, in einem Andenhochtal Zentralperus. Verh. disch. zool. Ges. Wien, 18: 265.

_ 1986. — Anfibios de la sierra central del Perü. Una clave de identificaciôn para adultos y larvas.

Boletin de Lima, 45: 23-33.

-— 1990. — Froschlurche (Anura) der zentralperuanischen Anden: Artdiagnose, Taxonomie,

Habitate, Verhaltensôkologie. Salamandra, 26: 177-214.

SmscH, U. & JURASKE, N., 1995. — Reassessment of central Peruvian Telmatobiinae (genera

Batrachophrynus and Telmatobius). 11. Allozymes and phylogenetic relationships. Alytes, in

press.

Truss, L., 1979. — Leptodactylid frogs of the genus Telmatobius in Ecuador with the description of

a new species. Copeia, 1979: 714-733.

VeLLARD, J., 1951. — Estudios sobre batracios andinos. 1. El grupo Telmatobius y formas afines.

Mem. Mus. Hist. nat. J. Prado, 1: 1-89.

1953. — Estudios sobre batracios andinos. I. El grupo marmoratus y formas afines. Mem. Mus.

Hist. nat. J. Prado, 2: 1-53.

ES 1955. — Estudios sobre batracios andinos. III. Los Telmatobius del grupo jelskii. Mem. Mus.

Hist. nat. J. Prado, 4: 1-28.

War, J. H., 1963. — Hierarchical grouping to optimize an objective function. J. am. stat. Assoc,

58: 236-244.

Wiens, J., 1993. — Systematics of the leptodactylid frog genus Telmatobius in the Andes of northern

Peru. Occ. Pap. Mus. nat. Hist. Univ. Kansas, 162: 1-76.

Corresponding editor: W. Ronald HEYER.

APPENDIX I

SPECIMENS EXAMINED

Batrachophrynus brachydactylus

PERÜ: (1) Departamento Cerro de Pasco: Caza Pato, 9 males, 3 females, MHNSM

1311, 1314, 1316, 1319, 1324-1325, 1331-1336; (2) Departamento Junin: brook near

Ondores: (a) first sample, 9 males, 10 females, URP 054-072; (b) second sample, 11 males,

11 females, collected by A. SALAS, March 26, 1992.

Batrachophrynus macrostomus

PERÛ: Departamento Junin: Junin Lake, Carhuamayo, 6 males, 7 females, 8 juveniles,

collected by M. ANTIGNANI, February 20, 1992.

Telmatobius brevirostris

PERÜ: Departamento Huanuco: (1) Ambo, Chasqui, 1 male, 1 female, MHNSM 3736,

7676 (syntypes of T. b. brevirostris), (2) Ambo, Caina, 2 females, MHNSM 7666-7667

(syntypes of T. b. parvulus); (3) Santa Maria del Valle, 1 male, MHNSM 7681 (holotype

of T. b. punctatus).

Telmatobius carrillae

PERU: Departamento Ancash: (1) Yuracyacu: (a) first sample, 4 males, 2 females,

MHNSM 1528 (holotype), 1544-1545, 3932-3934 (paratypes); (b) second sample, 9 males,

Source : MNHN, Paris

44 ALYTES 13 (1)

12 females, URP 001-021; (2) Huikia: (a) first sample, 5 males, 1 female, MHNSM

6681-6687; (b) second sample, 6 males, 2 females, URP 022-029; (3) Huaychopampa, 8

males, 4 females, URP 030-041.

Telmatobius culeus

BoLivia: Lake Titicaca, Isla del Sol, 1 male, 1 female, MHNSM 7769-7770 (assigned

to T. c. culeus).

PERÜ: (1) Departamento Arequipa: Yura, Arequipa, 1 female, MHNSM 7678

(syntype of T. c. exsul); (2) Departamento Puno: (a) Azangaro, Hacienda Checayani, 1

male, 1 female, MHNSM 7673-7674 (syntypes of T. c. lacustris); (b) Lake Lagunillas, 6

males, 4 females, MHNSM 7768, 7776-7777, 7185-7786, 7806-7807, 7823-7825 (assigned to

T. c. escomeli);, (c) Lake Titicaca, Ocama, 3 males, 3 females, MHNSM 7779-7784

(assigned to T. c. culeus); (d) Rio Coata, Juliaca, 2 males, MHNSM 7771-7772 (syntypes

of T. c. dispar); (e) Rio Ilave, Chucuito, 1 male, 1 female, MHNSM 7754-7755 (syntypes

Of T. c. fluviatilis); (f) Rio Ilave, Huayllata, 1 male, 2 females, MHNSM 7812-7814; (g)

mouth of Rio Ilave, 2 females, MHNSM 7773-7774 (assigned to T. c. culeus), and 3

females, MHNSM 7812-7814 (assigned to T. c. fluviatilis); (h) Rio Juliaca, Puno, 4 males,

4 females, MHNSM 7766-7767, 7787-7789, 7793-7794 (assigned to T. c. dispar); (i)

Umayo, 34 km NO of Puno, 1 female, MHNSM 7811 (assigned to T. c. lacustris).

Telmatobius jelskii

PERÜ: (1) Departamento Ayacucho: (a) Ayacucho, 4 males, 6 females, MHNSM

12202, 12206, 12213-12214, 12217, 12219-12220, 12222, 12225, 12899 (assigned to T. j.

walkeri); (b) Parinacochas, 6 males, 6 females, MHNSM 12838, 12841, 12883, 12901,

12904-12910; (c) Puquio, 2 males, 1 female, MHNSM 7642-7643, 7645 (syntypes of T. j.

longitarsis), (d) Tambo, 2 males, 4 females, MHNSM 7646-7651 (syntypes of T. jelskii

bufo), (2) Departamento Huancavelica: Huancavelica, 4 males, 1 female, 2 juveniles,

MHNSM 7639-7641, 7660-7661, 7663-7664; (syntypes of T. j. longitarsis); (3) Departa-

mento Junin: (a) Huancayo, Acolla, 3 males, 7 females, MHNSM 6903-6906, 6909-6914

(assigned to T. j. jelskü); (b) Huancayo, Palian, Rio Shullcas, 5 males, 2 females, 14

juveniles, collected by U. SINsCH & V. CANALES, February 19-25, 1992; (c) Tarma,

Cuyrohuasi: (i) first sample, 5 males, 5 females, URP 90066-90075; (ii) second sample, 6

males, 3 females, collected by A. SALAS, March 24, 1992.

Telmatobius rimac

PERÜ: (1) Departamento Ancash: Ocros, 7 males, 3 females, MHNSM 6935-6936,

6941-6942, 6944-6945, 6950-6951, 6953-6954 (assigned to T. r. rimac); (2) Departamento

Lima: (a) Canta, Obrojillo, Rio Chillon, 8 males, 7 females, 7 juveniles, collected by J.

ICOCHEA, March 8-9, 1992; (b) Canta, Quebrada Huaytara, 9 males, 4 females, 2 juveniles,

collected by J. ICOCHEA, March 8-9, 1992; (c) Tupe, 2 males, 2 females, MHNSM

7656-7659 (syntypes of T. r. meridionalis).

BIBL. DU

MUSÉUM Source : MNHN, Paris

PARIC

AINYTES

International Journal of Batrachology

published by ISSCA

EDITORIAL BOARD FOR 1995

Chief Editor: Alain Dumois (Laboratoire des Reptiles et Amphibiens, Muséum national d'Histoire

naturelle, 25 rue Cuvier, 75005 Paris, France).

Deputy Editor: Janalee P. CALDWELL (Oklahoma Museum of Natural History, University of

Oklahoma, Norman, Oklahoma 73019, U.S.A.).

Editorial Board: Ronald G. ALTiG (Mississippi State University, U.S.A.); Emilio BALLETTO (Torino,

Italy); Alain COLLENOT (Paris, France); Günter GOLLMANN (Wien, Austria); Tim HALLIDAY

(Milton Keynes, United Kingdom); W. Ronald HEYER (Washington, U.S.A.); Walter HôDL

(Wien, Austria); Pierre JOLY (Lyon, France): Masafumi MaTsuI (Kyoto, Japan: Jaime

E. PÉFAUR (Mérida, Venezuela), J. Dale ROBERTS (Perth, Australia); Ulrich SINsCH (Koblenz,

Germany); Marvalee H. Wake (Berkeley, U.S.A.).

Technical Editorial Team (Paris, France): Alain DUBois (texts); Roger BoUR (tables); Annemarie

OHLER (figures).

Index Editors: Annemarie OHLER (Paris, France); Stephen J. RICHARDS (Townsville, Australia).

GUIDE FOR AUTHORS

Alytes publishes original papers in English, French or Spanish, in any discipline dealing with

amphibians. Beside articles and notes reporting results of original research, consideration is given for

publication to synthetic review articles, book reviews, comments and replies, and to papers based

upon original high quality illustrations (such as color or black and white photographs), showing

beautiful or rare species, interesting behaviors, etc.

The title should be followed by the name(s) and address(es) of the author(s). The text should

be typewritten or printed double-spaced on one side of the paper. The manuscript should be organized

as follows: English abstract, introduction, material and methods, results, discussion, conclusion,

French or Spanish abstract, acknowledgements, literature cited, appendix.

Figures and tables should be mentioned in the text as follows: fig. 4 or Table IV. Figures should

not exceed 16 * 24 cm. The size of the lettering should ensure its legibility after reduction. The

legends of figures and tables should be assembled on a separate sheet. Each figure should be

numbered using a pencil.

References in the text are to be written in capital letters (SOMEONE, 1948; So & So, 1987;

EveryBoDY et al., 1882). References in the literature cited section should be presented as follows:

BOURRET, R., 1942. - Les batraciens de l'Indochine. Hanoi, Institut Océanographique de l’Indochine:

ix + 1-547, pl. I-IV.

GRAF, J.-D. & PoLLs PELAZ, M., 1989. - Evolutionary genetics of the Rana esculenta complex. In:

R. M. DAWLEY & J. P. BOGART (eds.), Evolution and ecology of unisexual vertebrates, Albany, The

New York State Museum: 289-302.

INGER, R. F., Voris, H. K. & Voris, H. H., 1974. - Genetic variation and population ecology of some

Southeast Asian frogs of the genera Bufo and Rana. Biochem. Genet. 12: 121-145.

Manuscripts should be submitted in triplicate either to Alain DuBois (address above) if dealing

with amphibian morphology, systematics, biogeography, evolution, genetics or developmental

biology, or to Janalee P. CALDWELL (address above) if dealing with amphibian population genetics,

ecology, ethology or life history. Acceptance for publication will be decided by the editors following

review by at least two referees.

If possible, after acceptance, a copy of the final manuscript on a floppy disk (3 % or 5 %)

should be sent to the Chief Editor. We welcome the following formats of text processing: (1)

preferably, MS Word (1.1 to 6.0, DOS or Windows), WordPerfect (4.1 to 5.1, DOS or Windows) or

WordStar (3.3 to 7.0); (2) less preferably, formated DOS (ASCII) or DOS-formated MS Word for the

Macintosh (on a 3 2 high density 1.44 Mo floppy disk only).

No page charges are requested from the author(s), but the publication of color photographs is

charged. For each published paper, 25 free reprints are offered by Alytes to the author(s). Additional

reprints may be purchased.

Published with the support of AALRAM

(Association des Amis du Laboratoire des Reptiles et Amphibiens

du Muséum National, d'Histoire Naturelle, Paris, France).

Directeur de la Publication: Alain Dusois.

Numéro de Commission Paritaire: 64851.