ISSCA

International Society for the Study

and Conservation of Amphibians

(International Society of Batrachology)

SEAT

Laboratoire des Reptiles et Amphibiens, Muséum national d'Histoire naturelle, 25 rue Cuvier, 75005

Paris, France. — Tel.: (33).(0)1.40.79.34.87. — Fax: (33).(0)1.40.79.34.88. — E-mail: dubois@mnhn.fr.

BOARD

President: W. Ronald HEYER (Washington, USA).

General Secretary: Alain Durors (Paris, France).

Deputy Secretary: Annemarie OHLER (Paris, France).

Treasurer: Jean-Louis DENIAUD (Paris, France).

Deputy Treasurer: Rafael De S4 (Richmond, USA).

Councillors: C. Kenneth Dopp, Jr. (Gainesville, USA); Britta GriLLrrscH (Wien, Austria); Julio Mario

Hoyos (Bogotä, Colombia); Thierry Lopé (Angers, France); Jon LOMAN (Lund, Sweden); Alain

PAGANO (Angers, France); Stephen J. RICHARDS (Townsville, Autralia).

TARIFF FOR 2002

INDIVIDUALS

[Regular 2002 subscription to Alpres (volume 20) + ISSCA + Circalptes 46€ 760$

Student 2002 subscription to Apres (volume 20) + ISSCA + Circalyres 23€130$

SrectaLorrer:gifthalf-pricesubscriptionof one year foracolleagueof yourchoice: Halfofpriceaboye

Regular 2002 subscription to Afjres (volume 20) alone FETES

Student 2002 subscription to Alpes (volume 20) alone 22€/28S

Back issues of Alyres: any single issue 11/1458

Back issues of A/pies: any double issue 2€/28S

Back issues of Alyres: any complete volume (4 issues) 35€/45$

Back issues of Alptes: complete set of volumes 1 to 19 (1982-2001) 532€ / 697$

Regular five-year (2002-2006, volumes 20 to 24) individual subscription to Alytes +

ISSCA + Circalytes 229€ 13008

Regular five-year (2002-2006, volumes 20 to 24) individual subscription to Alyres 171€/224$

SPECIAL OFF

Circalytes, with complete set of back issues of Alytes (1982-2001, volumes 1 to 19)

: five-year (2002-2006, volumes 20 to 24) individual subscription to A/ytes + ISSCA +

534€/700$

Life individual subscription to Alytes from 2002 on 900 €/1180$

Life individual subscription to Alytes + ISSCA + Circalytes from 2002 on 1100 € / 1440 $

Patron individual subscription to Alytes from 2002 on 1760 € / 2304 $ or more

Patron individual subscription to Alytes + ISSCA + Circalytes from 2002 on 2200 € / 2880 $ or more

Important notice: from 1996 on, any new life or patron individual subscriber to Alyres is offered a free

complete collection of back issues of Alytes from the first issue (February 1982) until the start of her/his

subscription.

INSTITUTIONS

2002 subscription to Alyres (volume 20) + ISSCA + Circalpres 92€/120$

2002 subscription to Afpres (volume 20) alone 88e/112S

Back issues of Alpes: any single issue 22€/28S

Back issues of Alptes: any double issue HEI56S

Back issues of A/ptes: any complete volume (4 issues) 70€/90$

Back issues of 4/pres: complete set volumes 1 to 19 1064 € / 13945

Sreciar orre: five-vear (2002-2006, volumes 20 to 24) subscription to Alpes + ISSCA + Circalyres,

with complete set of back issues of Alpes (1982-2001, volumes 1 to 19) 534€ / 100 $

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

prices can be provided upon request by our Secretariat.

MODES OF PAYMENT

— In Euros, by cheques drawn on a French bank payable to “ISSCA”’, sent to our secretariat (address

above).

Se by direct postal transfer to our postal account: “ISSCA”, Nr. 1-398-91 L, Paris: if you use this

mode of payment, add 2.30 € to your payment for postal charges at our end.

- In US Dollars, by cheques payable to “ISSCA”, sent to Rafael O. De S4, Associate Professor.

Department of Biology, University of Richmond, Richmond, VA 23173, USA (e-mail.

rdesa@richmond.edu; fax: (804)289-8233).

Source : MNHN, Paris

AITES

INTERNATIONAL JOURNAL OF BATRACHOLOGY

December 2001 Volume 19, N° 2-4

Alytes, 2001, 19 (2-4):

À new genus and species

of Ranidae (Amphibia, Anura)

from south-western India

Alain Dugois*, Annemarie OHLER* & S. D. Buu**

* Laboratoire des Reptiles et Amphibiens,

Muséum National d'Histoire Naturelle,

25 rue Cuvier, 75005 Paris, France

** Tropical Botanic Garden and Research Institute,

Thiruvananthapuram 695 562, Kerala, India

À new genus and species of Ranidae is described from Karnataka and

Kerala in south-western India. The new genus appears to belong in the

subfamily Dicroglossinae. It shares with the genus Fejervarya Bolkay,

1915 the presence of fejervaryan lines on both sides of the belly, but differs

from the latter genus in several respects, particularly in possessing a rictal

gland at the mouth commissure and a white horizontal band along the upper

lip, two characters that are common in the Indian Raninae but otherwise

absent in the Dicroglossinae. It appears to be the fifth genus/subgenus of

Ranidae endemic of southern India.

ABBREVIATIONS

Measurements. — EL, eye length (eye horizontal diameter); EN, dis

FETE, distance from maximum incurvation of web betw

femur length (from vent to knee); FLL, forelimb length (from elbow to base of outer palmar tubercle):

FOL, foot length (from base of inner metatarsal tubercle to tip of fourth toe); FTL, length of fourth toe

from basal border of proximal subarticular tubercle; HAL, hand length (from base of outer palmar

tubercle to tip of third finger); HL, head length (from back of mandible to tip of snout); HW, head width:

IBE, distance between back of eyes; IFE, distance between front of eyes: IMT, length of inner metatarsal

tuberele: IN, internarial space; ITL, inner toe length; IUE, minimum distance between upper eyelids;

MBE, distance from back of mandible to back of eye: MFE, distance from back of mandible to front of

eye: MN, distance from back of mandible to nostril, MTF

tance from distal edge of metatarsal

tubercle to maximum incurvation of web between fourth and fifth toe; MTTE, distance from distal edge

ET Bibliothèque Centrale Muséum

AT Il I | | II |. Paris

/ 3 0143745 7

3001 à

nce from front of eye to nostril;

n fourth and fifth toe to tip of fourth toe; F

54 ALYTES 19 (2-4)

of metatarsal tubercle to maximum incurvation of web between third and fourth toe; NS, distance from

nostril to tip of snout; SL, distance from front of eye to tip of snout: SVL, snout-vent length; TFL, length

of third finger from basal border of proximal subarticular tubercle; TFOL, length of tarsus and foot

(from base of tarsus to tip of fourth toe); TFTF, distance from maximum incurvation of web between

third and fourth toe to tip of fourth toe; TL, tibia length; TW, maximum “tibia” (actually shank) width;

TYD, maximum tympanum diameter; TYE, tympanum-eye distance; UEW, maximum width of upper

eyelid.

Museums, collections and persons. - AD, Alain Dubois; AMO, Annemarie Ohler, BMNH, Natural

History Museum, London, United Kingdom; FMNH, Field Museum of Natural History, Chicago,

Illinois, USA; MNHN, Muséum National d'Histoire Naturelle, Paris, France; MSNG, Museo Civico di

Storia Naturale Giacomo Doria, Genova, ltaly; MV, Michael Veith collection, Mainz, Germany: NMW,

Naturhistorisches Museum, Wien, Austria; SDB, S. D. Bju; TBGRI, Tropical Botanic Garden and

Research Institute, Thiruvananthapuram, Kerala, India; ZSI/SRS, Zoological Survey of India, Southern

Regional Station, Madras, Tamil Nadu, India.

INTRODUCTION

Southern India, especially in its western part (Western Ghats or Sahyadris), is one of the

richest biogeographic areas of the Oriental region. Myers (1990) identified the Western Ghats

as one among the 18 biodiversity hotspots of our planet. The amphibian fauna of this region

is rich both in terms of species number and endemicity (INGER et al., 1987; Buu, 2000). It also

contains several endemic genera, in particular of the family Ranidae Rafinesque-Schmaltz,

1814 (sensu Dugois, 1992, ï.e. including the Rhacophorinae Hoffman, 1932 as a subfamily) or

epifamily Ranoïidae (sensu VENCES & GLAW, 2001, ie. including a family Ranidae and a

family Rhacophoridae). In a traditional approach, supported by the latter proposal, the taxon

Ranidae is understood as including only groups that lack intercalary elements between the

penultimate and last phalanx of digits.

According to the highly provisional current working taxonomy of this family (see

Dugois, 1999), the endemic ranid genera of the Western Ghat ranges include /ndirana

Laurent, 1986, Micrixalus Boulenger, 1888 and Nyctibatrachus Boulenger, 1882, three genera

which belong in three different subfamilies. The genus /ndirana is a member (and currently

only genus: see BossuyT & MILiNKOVICH, 2000, and VENCES et al., 2000b) of the Ranixalinae

Dubois, 1987 (type-genus Ranixalus Dubois, 1986, a junior subjective synonym of /ndirana:

see DuBois, 1987b; subfamily sometimes incorrectly referred to as Indiraninae Blommers-

Schlôsser, 1993: see DuBois, 1999). The genus Micrixalus is the type-genus and only genus of

the subfamily Micrixalinae (see BossuyT & MiLiNKOVICH, 2001); the nomen of the latter

taxon, published without any diagnosis, is a nomen nudum (ANONYMOUS, 19994); in tab. 1, we

provide a diagnosis for this taxon. Finally, the genus Nyctibatrachus (synonym Nannobatra-

chus Boulenger, 1882; see Dugois, 1987a) is the type-genus and only genus of the subfamily

Nyctibatrachinae Blommers-Schlôsser, 1993 (see BLOMMERS-SCHLÔSSER, 1993; VENCES et al.,

2000b).

In southern India, the nominative subfamily Raninae also occurs. It is represented there

by a few species traditionally referred to the genus Rana Linnaeus, 1758, by some authors

Source : MNHN, Paris

Duois, OHLER & BuU 55

(e.g.. BOULENGER, 1920; Durra, 1997) to the subgenus Hy/arana Tschudi, 1838 of this genus,

and by Dugois (1992) to three provisional subgenera of this genus, two of which (Hydrophy-

lax Fitzinger, 1843 and Sy/virana Dubois, 1992) also occur in other regions, but the third of

which (Clinotarsus Mivart, 1869) is also an endemic of southern India.

We here report on the existence of a fifth group that also appears to be endemic of

southern India, and that belongs to a fifth provisional subfamily of Ranidae, the Dicroglos-

sinae Anderson, 1871. Before proceeding further, a few words are necessary concerning the

tribal taxonomy of this subfamily. Dugois (1992) recognized four tribes in the latter: an

African one with tadpoles (Conrauini Dubois, 1992), and three mostly Asian and Oriental

ones, one with direct developing genera (Ceratobatrachini Boulenger, 1884), and two with

tadpoles (Dicroglossini and Limnonectini Dubois, 1992). Recent data, especially molecular,

lead to a re-evaluation of this taxonomy. First of all, MARMAYOU et al. (2000) showed that the

genera Occidozyga Kuhl & Van Hasselt, 1822 and Phrynoglossus Peters, 1867 should be

excluded from this subfamily, but the genus Euphlyctis Fitzinger, 1843, which is cladistically

closely related to Hoplobatrachus Peters, 1863 and several other genera (BossuyT & MiILIN-

KOVITCH, 2000; VENCES et al., 2000a-b; Kosucu et al., 2001) should be maintained in the

Dicroglossinae; therefore the subfamily including the genera Occidozyga and Phrynoglossus

should be called Occidozyginae Fei, Ye & Huang, 1991 (see Dupois, 1992). Secondly, a whole

set of concordant data (BossuyT & MiLiINKOvITCH, 2000; EMERSON et al., 2000; MARMAYOU et

al., 2000; VENCES et al., 2000a-b; KosuCH et al., 2001; DELORME et al., submitted) suggest that

at least three clades exist among Asian and Oriental Dicroglossinae. The first clade, for which

the nomen Dicroglossini is available, includes the genera Euphlyctis Fitzinger, 1843, Fejerva-

rya Bolkay, 1915, Hoplobatrachus, Nannophrys Günther, 1869 and Sphaerotheca Günther,

1859. The second clade includes one genus with tadpoles, Limnonectes Fitzinger, 1843, and

one with direct development, Taylorana Dubois, 1987. MARMAYOU et al. (2000) showed that

the origin of direct development was independent in the latter genus and in the genus

Philautus Gistel, 1848 (Rhacophorinae/dae), but they did not study the cladistic relationships

of the other direct developing genera placed by Dugois (1992) in the Ceratobatrachini. In a

recent work, DELORME et al. (submitted) provided evidence that at least two of these genera

(Ceratobatrachus Boulenger, 1884 and /ngerana Dubois, 1987) were not closely related to

Limnonectes and Tuylorana, nor to the other Dicroglossinae, and deserve recognition as an

independent clade, for which the nomen Ceratobatrachinae is available. In the absence of

additional evidence, we here transfer the genus Zaylorana to the Limnonectini, but we

maintain all other developing genera in the Ceratobatrachinae. This is however a provisional

solution, until the cladistic position of all of them has been ascertained, as some of them

might later prove to belong in fact in the Limnonectini. Finally, the data of DELORME et al.

(submitted) also suggest that the group recognized by Duois (1992) as a tribe Paini of the

Raninae should be transferred to the Dicroglossinae, as a fourth tribe (including species that

differ from those of all other tribes of this subfamily by their unforked omosternum). The

subfamilial and tribal classification of the Ranidae will be discussed in more detail elsewhere

(Dusois & OHLER, in preparation), but for the time being we just provide in tab. 1 some major

diagnostic morphological characters for the five subfamilies mentioned above and for the two

Asian tribes of Dicroglossinae with forked omosternum that include species with tadpoles

(Dicroglossini and Limnonectini). The data summarized in this table will be useful to establish

the place of the new taxon described below in the Ranidae.

Source : MNHN, Paris

ALYTES 19 (2-4)

Table 1. - Some diagnostic morphological characters of three subfamilies of the family Ranidae

present in southern India. See GRANT et al. (1997) for a definition and discussion of the

median lingual process. See DUBOIS (1995) for the definition of the tadpole’s condensed

collective keratodont formula (CCKF), i.e. minimum-maximum numbers of keratodont rows

on upper/lower lips of tadpoles observed in the taxon. In “generic contents”, genera present

in southern India are marked with an asterisk.

; Dicroglossinae Dicroglossinae Micrixalinae

Subfamily Anderson, 1871 Anderson, 1871 subfam. nov.

Te Dieroglossini Limnonectini r

Anderson, 1871 Dubois, 1992

Type-genus Dicroglossus Günther, | Limnonectes Fitzinger, | Micrixalus Boulenger,

1860 (junior subjective 1843 1888

synonym of Euphlyctis

Fitzinger, 1843)

Generic contents Euphlyctis* Fitzinger, | Limnonectes Fitzinger, Micrixalus*

1843; Fejervarya*

Bolkay, 1915;

Hoplobatrachus*

Peters, 1863;

Nannophrys Günther,

1843; Taylorana

Dubois, 1987

Boulenger, 1888

1869; Sphaerotheca*

Günther, 1859

Base of omosternum Forked Forked Unforked

Vomerine teeth Present Present Absent

Median ligual process Absent Absent Present or absent

Digital disks Absent Absent or present Present

Femoral glands Absent or present Absent Absent

Vocal sacs in male Present Present or absent Present

Nuptial pads in males | Present or absent Absent Present

Tadpole type Aquatic or terrestrial | Aquatic or direct Aquatic

development

Tadpole’s CCKF 1-5/2-6 1-3/2-3 1/0

References for BOLKkAY, 1915; DECKERT, 1938; BOULENGER, 1882,

characters DECKERT, 1938; LAURENT, 1986; 1890; SMITH, 1924;

LAURENT, 1950;

KIRTISINGHE, 1958;

CLARKE, 1981, 1983;

DuBois, 1987a, 1992;

FEI, 1999

Dugois, 1987a, 1992;

FEI, 1999

DECKERT, 1938;

MYERS, 1942b; INGER

etal., 1984; DUBOIS,

1987a, 1992

Source : MNHN, Paris

Table 1. (continued)

Dugois, OHLER & Bu

Raninae

) Nyctibatrachinae Rarilalinaë

Subfamily Blommers-Schlüsser, | Rafinesque-Schmaltz, :

Dubois, 1987

1993 1814

: Ranini

Tribe = Rafinesque-Schmaltz, Q

1814

Type-genus Nyctibatrachus | Rana Linnaeus, 1758 | Ranixalus Dubois,

Boulenger, 1882

1986 (junior subjective

synonym of /ndirana

Laurent, 1986)

Generic contents Nyctibatrachus* | Amolops Cope, 1865; | Indirana* Laurent,

Boulenger, 1882 Batrachylodes 1986

Boulenger, 1887;

Nanorana Günther,

1896; Rana* Linnaeus,

1758; Staurois Cope,

1865

Base of omosternum Forked Unforked Forked

Vomerine tecth Present Present Present

Median ligual process Present Absent Present

Digital disks Present Absent or present Present

Femoral glands Present Absent Present

Vocal sacs in male Absent Present Present or absent

Nuptial pads in males Present Present or absent Present

Tadpole type Aquatic Aquatic Terrestrial

Tadpole’s CCKF 0/0 1-12/2-9 3-5/3-4

References for BOULENGER, 1882, Dugois, 1992 ANNANDALE, 1918;

characters 1890; ANNANDALE, BOULENGER, 1920;

1918, 1919; MYERS,

1942a; BHADURI &

KRIPALANI, 1955;

PILLAI, 1978; INGER et

al., 1984; DUBOIS,

1987a, 1992;

SHAFFER, 1988

RAO, 1920; INGER et

al., 1984; LAURENT,

1986; DUBOIS, 1987a,

1992; SEKAR, 1992;

BLOMMERS-

SCHLÔSSER, 1993

MATERIAL AND METHODS

Specimens were collected in the field, fixed in 4 % formalin shortly after capture and

stored in 70 %ethanol. The list of specimens examined and measured is given below under the

description of the new species and in app. 1 for all other specimens, belonging to other species,

used as comparative material.

Source : MNHN, Paris

58 ALYTES 19 (2-4)

Thirty-two measurements of adult and young specimens were taken by AMO with a slide

calliper to the nearest 0.1 mm, or, for values below 5 mm, with an ocular micrometer to the

nearest 0.01 mm. The list of measurements is given above under Abbreviations.

In order to facilitate comparisons, the description’s methodology and plan were the same

as those used in previous works on Asian anurans (Dugois & OHLER, 1998, 1999, 2000; OHLER

& Dusois, 1999; BossuyT & Dugois, 2001; VerrH et al., 2001). The webbing formula is given

according to MyERs & DUELLMAN (1982) and the tadpole keratodont formula according to

Dusgois (1995). À male specimen (MNHN 2000.3033) was partially dissected to ventrally

examine the pectoral girdle. Drawings of an adult were made by AMO using a camera lucida

(Wild Heerbrugg type 256576).

Morphometrical analyses and graphs were made using the SPSS statistical programs for

personal computers (Norusis, 1992; ANONYMOUS, 1999b). We used principal component

analysis with varimax rotation (ANONYMOUS, 1999b: 426) to show morphological distinctive-

ness of the new genus and canonical discriminant analysis to indicate morphological discri-

mination from Fejervarya and Sphaerotheca. One-way Tukey type b tests were performed on

ranked ratios of all measurements between the six genera of Dicroglossinae. Detailed results

of this analysis can be communicated upon request by the first authors, but are not given here

because of space limitations.

RESULTS AND DISCUSSION

The data presented in detail below suggest that the new taxon discussed here is a new

species that is the first known representative of a new genus of the Dicroglossinae Dicroglos-

sini. We provide below a definition of the genus, followed by a comparison with other genera

and a discussion of its relationships, and a detailed description of the species.

Minervarya gen. nov.

Type-species. — Minervarya sahyadris sp. nov.

Diagnosis. — Size small (SVL 17.6-19.2 mm in adult males, 20.6-23.0 mm in adult females):

omosternum forked at base; vomerine teeth present; median lingual process absent: Il

gland present; digital extremities rounded, not dilated; webbing rudimentary; inner metatar-

sal tubercle very short and conical: external metatarsal tubercle present; tarsal ridge present:

femoral glands absent; dorsal skin with longitudinal folds; lateral-line system absent in adult;

fejervaryan lines present; upper lip with white horizontal band; nuptial pads present on

prepollex and finger I of breeding male; vocal sac present, marked by glandular skin on

middle of throat of adult male; aquatic tadpole with a keratodont formula of 2/3, keratodont

rows simple; eggs of rather small size, white and brown colored.

Phenetic comparisons. —- The combination of character states of the diagnosis above clearly

excludes Minervarya from all subfamilies and tribes listed in tab. 1, except the tribe Dicro-

glossini of the Dicroglossinae. The tribes Conrauini and Paini can also be readily excluded,

Source : MNHN, Paris

Dugois, OHLER & BHU 59

the first one, among other characters, because of the keratodont formula of its tadpoles

(CCKF 7-8/6-11; LAMOTTE & PERRET, 1968), and the second one, among other characters, by

its unforked omosternum (DuBois, 1975, 1992). As a result of these comparisons, we propose

to place the new genus in the tribe Dicroglossini. However, comparisons of the new species

with the five genera referred here to this tribe point to the uniqueness of this species, that

suggests that it belongs to a new, distinct genus.

The endemic Sri Lankan genus Nannophrys can readily be excluded from these compa-

risons, not only because of the very peculiar morphology of the adult, that has nothing to do

with that of the new taxon (see eg. KIRTISINGHE, 1957; CLARKE, 1983; DUTTA &

MANAMENDRA-ARACHCHI, 1996), but also of its very peculiar tadpole (KIRTISINGHE, 1958).

Remain four genera (tab. 1), all of which do occur in southern India, for which we provide

detailed comparisons in tab. 2. For more security, in this table we also extended comparisons

to the genus Limnonectes, that was placed by DuBois (1992) in the same tribe as Fejervarya

and Hoplobatrachus, and to the three subgenera of Rana that are known to occur in southern

India.

The data of tab. 2 show clearly that the new taxon does not fit by its combination of

characters with any of the four other genera of Dicroglossini. Among them, of particular

relevance is a comparison with the genus Fejervarya, with which it shares a rare character, the

presence of fejervaryan lines (as defined by Dugois & OHLER, 2000: 35) on both sides of the

belly (fig. 7). But the new taxon differs from the genus Fejervarya in several other important

characters. In particular, the presence in this taxon of a rictal (mandibular) gland just

posterior to the corner of the mouth (fig. 5) is unique in the subfamily Dicroglossinae, while

this character is common in the subfamily Raninae (personal observations), including the

Indian subgenera of Rana (fig. 6: tab. 2). Another important difference is the coloration of the

upper lip of the adults (bright white) (fig. 5), that is not to be found in any other Dicroglos-

sinae but is common in the Raninae, including the Indian subgenera of Rana (fig. 6; tab. 2).

However, the forked omosternum of the new genus definitely excludes it from the Raninae as

currently understood.

Minervarya differs from Fejervarya by the two characters mentioned above, and by its

smaller adult size, the proportions of its head and hindlimbs, the aspect of its dorsal folds and

of its vocal sacs. Beside the presence of fejervaryan lines and of rictal glands and the

coloration of its upper lip, the new genus differs from all other Dicroglossinae, and in

particular of Dicroglossini, by a number of characters (see tab. 1-2), among which only the

most striking ones need to be mentioned here: from Euphlyctis, the new genus differs by its

size, skin structure, webbing and tadpole’s CCKF; from Hoplobatrachus, it differs by its size,

webbing and tadpole’s characters: from Sphaerotheca, it differs by its size, dorsal skin, shape

of internal metatarsal tubercle and absence of femoral glands; from Nannophrys, it differs by

its general shape and the whole morphology of tadpoles; finally, from Limnonectes (a genus

member of the tribe Limnonectini of the Dicroglossinae), Minervarya differs by its size,

undilated digital tips, webbing and presence of nuptial pads in adult breeding males.

Multivariate comparisons. - Morphometric data also confirm the uniqueness of this taxon. In

several zoological groups, genera can be viewed as “shape groups”, among which species are

more similar in shape than with species of other genera (LEMEN & FREEMAN, 1984; DuBoIs,

1988a-b). This is particularly true in anuran amphibians, where morphometric differences

Source : MNHN, Paris

ALYTES 19 (2-4)

Table 2. — Some diagnostic morphological characters of nine genera or subgenera of the

subfamilies Dicroglossinae and Raninae of the family Ranidae. See OULER & DUBOIS

(1989) for a definition of the different kinds of digital dises recognized here.

Fe Minervarya Fejervarya Sphaerotheca

gen. nov. Bolkay, 1915 Günther, 1859

Adult male SVL 17-20 25-80 30-55

Adult female SVL 20-23 30-90 35-60

Head proportions Longer than wide Longer than wide Shorter than wide

CŒHW:HL)

Rictal gland Present Absent or present Absent or present

Digit tips Rounded Pointed Rounded

Webbing on feet Rudimentary Small to medium Small

Internal metatarsal Rather long, Long and narrow, | Short, shovel-shaped

tubercle cylindrical cylindrical or

shovel-shaped

External metatarsal Present Absent or present Absent

tubercle

Tarsal ridge Absent Faint and short, or Absent

absent

Dorsal skin With several With numerous Smooth

Upper lip coloration

Dorsal chevron

Shoulder spots

Tympanum coloration

Lateral line system in

adult

Fejervaryan lines

Femoral glands

Vocal sacs in male

Nuptial pads in male

longitudinal folds

With white horizontal

band

Absent

Dark brown with

its inferior border

white

Absent

Present

Absent

Marked by darker

coloration and skin

differentiation on

throat and chest

Present on prepollex

longitudinal folds

With vertical brown

bars

Present or absent

Dark spot on upper

posterior part

Absent

Present

Absent

Marked by darker

coloration, and

sometimes also by

longitudinal folds, on

sides of throat

Present on prepollex

With vertical brown

bars

Absent

Present or absent

Uniform or marbled,

no distinct spot

Absent

Present

Marked by darker

coloration and folds on

sides of throat

Present on prepollex

and finger I and finger I and finger 1

Humeral glands in Absent Absent Absent

males

Tadpole’s CCKF 2/3 2/3 2/3

Keratodont rows Simple Simple Simple

References for This paper BOULENGER, 1920; BOULENGER, 1920;

characters DuBois & OHLER, KIRTISINGHE, 1958;

2000; VErrH et al., DUTTA &

2001 MANAMENDRA-

ARACHCHI, 1996

Source : MNHN, Paris

Dugois, OHLER & BU 61

Table 2. (continued)

us Euphlyctis Hoplobatrachus Limnonectes

Fitzinger, 1843 Peters, 1863 Fitzinger, 1843

Adult male SVL 40-95 75-130 35-150

Adult female SVL 45-130 65-140 35-135

Head proportions As wide as long As wide as long As wide as long

ŒW:HL)

Rictal gland Absent Absent Absent

Digit tips Pointed Rounded Rounded and dilated,

sometimes with dorso-

terminal grooves

Webbing on feet Complete Large Medium to large

Internal metatarsal Pointed, cylindrical, Rather long, Rather long,

tubercle digit-like cylindrical or cylindrical

shovel-shaped

External metatarsal Absent Absent Absent

tubercle

Tarsal ridge Faint or absent Long, distinct Present, usually faint,

or absent

Dorsal skin Smooth with horny With numerous Smooth or with

granules longitudinal folds longitudinal folds

Upper lip coloration With or without With vertical brown Without special

vertical brown bars bars coloration

Dorsal chevron Absent Absent Present or absent

Shoulder spots Absent Absent Absent

Tympanum coloration Uniform, light Uniform, light Uniform or with an

indistinct darker spot

Lateral line system in Present Absent Absent

adult

Fejervaryan lines Absent Absent Absent

Femoral glands Absent Absent Absent

Vocal sacs in male Dark-coloured, Marked by colour and Present or absent

projecting through two | longitudinal folds on

slits on sides of throat sides of throat

Nuptial pads in male Absent Present on prepollex Absent

and finger I

Humeral glands in Absent Absent Absent

males

Tadpole’s CCKF 12 2-5/3-6 1-3/2-3

Kcratodont rows Simple Double Simple

References for BOULENGER, 1920; BOULENGER, 1920; BOULENGER, 1920;

characters KIRTISINGHE, 1958; DuBois, 1992; FEI, BOURRET, 1942;

DUTTA & 1999; KOSUCH et al., INGER, 1966, 1985;

MANAMENDRA- 2001 OHLER & DUBOIS,

ARACHCHI, 1996 1999

Source : MNHN, Paris

Table 2. (continued)

ALYTES 19 (2-4)

Dorsal skin

Smooth, with dorso-

lateral folds

Genus Rana Linnaeus, 1758 | Rana Linnaeus, 1758 | Rana Linnaeus, 1758

Clinotarsus Mivart, Hydrophylax Sylvirana Dubois,

Subgenus 1869 ritzinger, 1843 71992

Adult male SVL 42-55 59-65 34-80

Adult female SVL 63-85 40-70 30-83

Head proportions | As wide as long or As wide as long As wide as long or

ŒW:HL) wider than long longer than wide

Rictal gland Absent Present Present

Digit tips Rounded and dilated, | Rounded and slightly | Rounded and dilated,

sometimes with dilated usually with latero-

indistinct latero- ventral groove

ventral groove

Webbing on feet Large Medium Medium or large

Internal metatarsal | Rather long, oval, very | Rather long, oval, very | Rather short, oval or

tubercle prominent prominent elliptic

External metatarsal Absent Present Present

tubercle

Tarsal ridge Absent Absent Absent

Smooth or granular,

with indistinct dorso-

lateral folds

Smooth, with dorso-

lateral folds

Upper lip coloration With a blackish With white horizontal | With white horizontal

margin band band

Dorsal chevron Absent Absent Absent

Shoulder spots Absent Absent Absent

Tympanum coloration Uniform, dark Uniform, light-colored | Uniform, light-colored

Lateral line system in Absent Absent Absent

adult

Fejervaryan lines Absent Absent Absent

Femoral glands Absent Absent Absent

Vocal sacs in male Present Present Present

Nuptial pads in male | Present on prepollex | Present on prepollex | Present on prepollex

and finger I and finger I and finger I

Humeral glands in Absent Present Present

males

Tadpole’s CCKF 8/6-8 1/2-3 1-2/2-3

Keratodont rows Simple Simple Simple

References for BOULENGER, 1920; | ANDERSSON, 1937; | BOULENGER, 1920;

characters CHaRt, 1962 CHARI, 1962 KIRTISINGHE, 1957;

DuBois, 1992; DUTTA

& MANAMENDRA-

ARACHCHI, 1996

Source : MNHN, Paris

Dusois, OHLER & BuU 63

o 1

à o ;

» Co° el

HET : TAXON

>> de a

| «4 ® Sphaerotheca

| Êl

0! (o] 4 & < Hoplobatrachus

> À .

La É à

. .

Fe &e # Euphlyctis

5 a © Fejervarya

5 |

ww 2] © Limnonectes

15 “10 -5 0.0 .5 10 15 20

Factor 1

Fig. 1. - Plots of factors 1 and 2 of principal component multivariate analysis based on varimax rotated

coefficients from log-transposed characters (25 measurements) for the following genera: Euphlvetis,

Fejervarva, Hoplobatrachus, Limnonectes, Minervarya and Sphaerotheca.

between species within genera are often very slight, if not absent, while differences between

genera are often very clear-cut, thus allowing generic allocation of isolated species or

individuals (HEYER, 1994; Onrer, 1996; Dumois & OuLer, 1999; Verru et al., 2001), or

recognition of new genera (DUBOIS, 1980; Our, 1999). The validity of this approach is once

again confirmed in this study (tab. 3). As shown in fig. 1, the first and second factors of the

principal component analysis based on 25 measurements sorted all genera of Dicroglossini,

and also Limnonectes, as discrete groups. The new taxon appears as a new, discrete group,

quite distinct from the other five groups, at least as distinct as they are between themselves.

Morphologically it is closest to Fejervarya and Sphaerotheca. A discriminant analysis includ-

ing these three genera (tab. 4; fig. 2) allowed to show that the measurements that best explain

variation between groups are size, head shape, hind leg length and webbing.

One-way analysis including all six genera showed significant differences of the new genus

in all but two measurements (HAL, TFL) to at least one of the other genera. The new taxon

is the smallest species (SVL), it has the largest internarial distance (IN) and the shortest

metatarsal tubercle (IMT). The new genus can be distinguished from Fejervarya by its wider

head as expressed by IFE and IBE, its shorter head measurements (MFE, EN), larger eyes

Source : MNHN, Paris

64 ALYTES 19 (2-4)

€ .

a TAXON

-10: Le

| Na + Group Centroids

N 4 Sphaerotheca

S .

5 + Minervarya

Fr SOUSSE es 2e 7, : : ° Fejervarya

-30 -20 -10 0 10 20 30

Function 1

Fig. 2. — Plots of canonical discriminant function scores using minimization of Wilk's lambda of

morphometrie log-transposed characters (25 measurements) for the genera Fejervarya, Minervarya

and Sphaerotheca.

(EL), shorter shank (TL), foot (FOL) and inner toe (ITL), and its reduced webbing (MTTEF,

MTFF). It can be differentiated from Limnonectes by its narrower (HW) and shorter (HL,

MN, MBE, EN, TYE) head, its shorter forearm (FLL), shank (TL) and inner toe (ITL), but

its longer fourth toe (FTL), and more reduced webbing (MTTF, MTFF, TETE, FFT). It

differs from Hoplobatrachus by its head shape (MFE, IFE, IBE, TYE), its longer fourth toe

(FTL), larger eyes (EL) and reduced webbing (MTTF, MTFE, TETE, FETE). It is distin-

guished from Euphlyctis by its wider head (IFE, IBE), larger eyes (EL) and reduced webbing

(MTTE, MTFE, TETE, FFTF). The morphological differences to Sphaerotheca are its

narrower (HW, IFE) and longer head (MFE, MN, EN), longer shank (TL), foot (FOL) and

fourth toe (FTL), and its reduced webbing (TFTF, FFTF).

Cladistic relationships. — AS mentioned above, two important characters of the new taxon are

not to be found in any known member of the Dicroglossinae but are common in the Raninae:

(1) the presence of a rictal gland: (2) the coloration of the upper lip (bright white). However,

these two characters are not unique to the Raninae, but can be observed in various other

anuran taxa, where they probably appeared by convergence. On the other hand, the forked

omosternum of the new species excludes it from the Raninae as currently understood. For the

Source : MNHN, Paris

Dusois, OHLER & BUU 65

Table 3. — Results of principal component analysis based on varimax rotated coefficients from log-

transposed characters (25 measurements) for specimens referred to the genera Euphlyctis,

Fejervarya, Hoplobatrachus, Limnonectes, Minervarya and Sphaerotheca.

Initial Eigenvalues

Component Total % of Variance Cumulative %

1 22.639 90.558 90.558

2 0.799 3.196 93.754

3 0.696 2.783 96.537

Rotation Sums of Squared Loadings

Component Total % of Variance Cumulative %

1 10.152 40.610 40.610

2 9.597 38.390 78.999

3 4384 17.538 96.537

Components for rotated component matrix

Variable 1 2 3

SVL 0.649 0.621 0.422

HW 0.617 0.706 0.337

HL 0.673 0.647 0.340

MN 0.667 0.646 0.330

MFE 0.649 0.674 0.322

MBE 0.639 0.683 0.312

IFE 0.505 0.768 0.371

IBE 0.553 0.757 0.368

FLL 0.589 0.682 0.419

HAL 0.661 0.653 0.346

TE 0.732 0.530 0410

FOL 0.709 0.534 0.456

IN 0.235 0.817 0.471

EN 0.698 0.592 0.353

EL 0.599 0.691 0.351

TYD 0.712 0.487 0.367

TYE 0.449 0.773 0.223

TFL 0.654 0.635 0.364

FIL 0.757 0415 0.484

MTTF 0.836 0.453 0.299

TETE 0.349 0.349 0.859

MTFF 0.830 0.463 0.297

FFTF 0.432 0.402 0.788

IMT 0.419 0.797 0.293

ITL 0.873 0.304 0.283

Source : MNHN, Paris

66 ALYTES 19 (2-4)

Fig. 3.- Holotype of Minervarya sahyadris, MNHN 2000.3031, adult male (SVL 19.1 mm). Top left, head

from above: top right, right hand from below; bottom left, head from right side; bottom right, right

foot from below.

Source : MNHN, Paris

Dusois, OHLER & Buu

Table 4. — Results of canonical discriminant analysis between specimens referred to the genera

Fejervarya, Minervarya and Sphaerotheca.

A. Statistical significance

56.426 0.991 0.017

Bigenvalue | Canonical | \yiiks Lambda| Chi-square Degrees P

correlation of freedom

298.016 0.998 0.000 146.265 50 0.000

60.757 24

B. Standardized canonical discriminant function coefficients

Morphometric Function Werromede Function

character 1 2 character n 2

SVL -5472 - 0.465 FIL - 0.092 -3.222

HW 2411 -1235 IN - 6213 2.296

HL 5.586 - 1.868 EN 0.992 0.957

MN -4.922 1.529 EL 3.284 -2374

MFE 1.366 2.573 TYD 3.701 -2.020

MBE - 0.320 - 2.749 TYE 2.179 -0.410

IFE -0.381 - 1.919 MTTF - 2.364 3.692

IBE -0.339 1.960 MTFF 4273 2.344

FLL -1372 - 0.188 TETE = 6.194 3.013

HAL 2.392 0.982 FFTF 2.974 1.112

TFL - 1.859 -0410 IMT -3.768 -0.743

TL 5.54 -3.821 ITL - 0.309 1.055

FOL -0.443 - 0.878

C. Classification success

Predicted group membership

Actual group -

Fjervarya Minervarya Sphaerotheca

Fejervarya 10 (100%) 0 0

Minervarya 0 10 (100%) 0

Sphaerotheca 0 0 10 (100%)

Source

MNHN, Paris

Table 5. - Some measurements and ratios of 6 adult males (the holotype* and 5 paratypes) of Minervarya sahyadris from the type-locality Gundia in

Kamataka and two localities in Kerala (Mukkam, Thiruvanpady). SVL is given in mm; all other measurements are given as per thousands of SVL.

Values given in the last two columns: mean + standard deviation (range).

Collection MNEN MNHN MNHN TBGRI TBGRI MNHN Total 3 adult males tal 3 adult males

number 2000.3030 2000.3031* 2000.3033 2001.0006 2001.0007 2000.3036 from Karnataka from Kerala

Locality | Gundia Gundia Gundia | Mukkam | Mukkam |Thiruvanpady Gundia seront

| | Thiruvanpady

SVL 18.9 19.1 18.0 172 18.5 19.0 18.7 + 0.59 (18.0-19.1) | 18.2 +0.93 (17.2-19.0)

HW 344 319 317 366 341 358 3274 15.0 (317-344) | 355 + 13.1 (341-366)

HL 365 393 428 448 384 374 395431.4(365-428) | 402 + 40.1 (374-448)

357 353 371 428 357 331 360 + 6.6 (353-371) 372 + 50.4 (331-428)

MFE 295 282 292 359 291 266 289 +6.8 (282-295) 305 + 48.2 (266-359) >

MBE 178 153 173 206 172 147 168 + 13.4(153-178) | 175 + 29.9 (147-206) C

IFE 199 170 180 211 196 194 183 +14.9 (170-199) 200 +9.2 (194-211) al

IBE 285 258 267 302 281 293 270 + 13.5 (258-285) | 292 + 10.7 (281-302) D

FLL 209 186 205 234 217 225 200 + 12.1 (186-209) 225 + 8.2 (217-234) re

HAL 223 207 194 237 238 232 208 + 14.2 (194-223) 236 + 3.3 (232-238) Le

TFL 96 119 112 132 133 140 109 +11.8 (96-119) 135 + 4.4 (132-140) =

TL 434 424 428 459 465 458 429 + 4.9 (424-434) 461 + 3.7 (458-461) È

FOL 534 456 483 494 486 537 4914 40.0 (456-534) | 506 + 27.1 (486-537)

FTL 336 298 263 316 301 331 299 + 36.6 (263-336) | 316+ 15.0 (301-331)

IN 96 92 106 105 101 100 98 + 7.2 (92-106) 102 + 2.4 (100-105)

EN 76 68 92 78 74 75 78 + 12.0 (68-92) 76 + 2.0 (74-78)

EL 137 139 129 141 135 133 135 + 5.1 (129-139) 136#4.5 (133-141)

TYD 69 61 61 67 57 69 64 + 4,6 (61-69) 64+ 6.2 (57-69)

TYE 21 20 29 20 25 25 23 + 5.1 (20-29) 23 #3.1 (20-25)

IMT 41 38 43 40 39 41 414 2.9 (38-43) 40 4 1.1 (39-41)

ITL | 102 83 84 99 99 93 90 + 10.6 (83-102) 97 + 3.6 (93-99)

MTTF 202 173 177 188 186 187 184+15.8 (173-202) 187 + 1.2 (186-188)

MTFF 202 176 180 188 189 191 186 +13.9 (176-202) 190 & 1.4 (188-191)

TFTF 305 302 263 302 291 331 290 + 23.7 (263-305) | 308 + 20.8 (291-331)

FETE 315 312 277 320 305 334 3024 21.1 (277-315) | 320 + 14.7 (305-334)

Source : MNHN, Paris

Dugois, OHLER & BuU 69

Table 6. — Some measurements and ratios of 4 adult female paratypes of Minervarya sahyadris

from the type-locality Gundia in Kamataka. SVL is given in mm; all other measurements

are given as per thousands of SVL. Values given in the last column: mean + standard

deviation (range).

Collection | MNHN | MNHN | MNHN | MNHN Total 4 adult females

number | 2000.3027 | 2000.3028 | 2000.3029 | 2000.3034 from Karnataka

SVL 23 23.0 219 20.6 22.0 # 1.01 (20.6-23.0)

HW 327 326 329 316 324 + 6.0 (316-329)

HL 345 352 370 384 363 à 17.3 (345-384)

MN 302 313 352 346 328 + 24.5 (302-352)

MFE 270 242 281 270 266 + 16.7 (242-281)

MBE 169 147 157 167 160 + 10.2 (147-169)

IFE 171 169 184 183 1774 7.5 (169-184)

IBE 273 253 263 270 265 + 8.7 (253-273)

FLL 189 194 210 201 199 + 9.2 (189-210)

HAL 212 217 219 201 212 + 7.9 (201-219)

TFL 134 127 130 132 1314 2.9 (127-134)

TL 435 417 438 403 423 + 16.5 (403-438)

FOL 507 474 sil 456 487 + 26.4 (456-511)

FIL 328 304 320 293 3114 15.8 (293-328)

IN 93 93 98 100 96 + 3.4 (93-100)

EN 76 68 80 79 76 + 5.4 (68-80)

EL 131 130 133 133 1324 1.8(130-133)

TYD 67 & @ @ 63+ 2.3 (62-67)

TYE 17 25 26 19 22444 (17-26)

IMT 39 40 4 30 384 5.5 (30-42)

IL 100 100 100 94 98 43.2 (94-100)

MTTF 192 177 180 164 178 + 11.6 (164-192)

MTFF 186 183 195 176 185 4 8.0 (176-195)

TETE 320 302 305 283 302 4 15.1 (283-320)

FFTF 320 310 305 299 308 + 8.8 (299-320)

time being, the cladistic significance of the state of the omosternum in the Ranidae

(forked/unforked) is not yet clarified, especially as the recent data mentioned above in the

Introduction, and especially those of DELORME et al. (submitted), suggest that a group of

genera with unforked omosternum (the Paini) is cladistically nested within a clade of genera

with forked omosternum (the Dicroglossinae), while other groups with forked omosternum

(the Occydozyginae and Ceratobatrachinae) are apparently not part of the Dicroglossinae

lincage. However, for the time being there exists no evidence of a genus with forked omoster-

num belonging in the well-identified clade to which the nomen Raninae applies.

s forked omosternum, our main reason for referring the new genus to the tribe

is because it shares with the genus Fejervarva Bolkay, 1915 à unique character

Source : MNHN, Paris

70 ALYTES 19 (2-4)

Fig. 4. — À male specimen of Minervarya sahyadris (paratype, TBGRI 2001.0002) photographed in life on

24 August 1999 at Mukkam, Kozhikodu District, Kerala (photo S. D. Biju).

Fig. 5 — Lateral view of head and body of Minervaya sahyadris: (a) MNHN 2000.3031, adult male,

holotype, Gundia, Karnataka; (b) TBGRI 2001.0007, adult male, Mukkam, Kerala. b, white band

on upper lip: g, rectal gland.

Source : MNHN, Paris

Dugois, OHLER & BU 71

Fig. 6- Lateral view of head and body of: (a) Rana (Hydrophylax) malabarica, MNHN 771, adult

fornalesmitype: Malabat: b) Rana (Sylvirana) temporalis, MNHN 2000.0613, adult male,

Kitulgala, Sri Lanka. b, white band on upper lip: g, rictal gland.

Source : MNHN, Paris

72 ALYTES 19 (2-4)

Fig. 7. - Ventral view of body of: (a) Minervarya sahyadris, MNHN 2000.3029, adult female, paratype,

Gundia, Karnataka; (b) Fejervarya limnocharis, MNHN 1999.5723, adult female, Sapa, Vietnam. !,

fejervaryan line.

among anurans, the presence on both sides of the belly of a fejervaryan line. We do not know

of a single other genus of Anura showing this character. As both Fejervarya and Minervarya

occur in southern India, we think it is more parsimonious to hypothesize a common origin of

this character in these two genera than its independent occurrence by homoplasy. In the

absence of contradictory information, we regard this character as a synapomorphy of both

genera, that suggests that they are sister-genera. Molecular works are currently in progress to

test this hypothesis.

Contents. — For the time being, Minervarya sahyadris is the only known species of the genus

Minervarya.

Distribution. — The new genus is known from the states of Karnataka and Kerala in

south-western India.

Vernacular name. — We propose to use the name “minervarya” as vernacular name for these

frogs, and “fejervarya” for frogs of the genus Fejervarya.

Etymology of the generic nomen. — The generic nomen, derived from the Latin adjective

minimus, “very small, the smallest”, and from the zoological generic nomen Fcjervarva,

suggests that the new frog looks like a very small fejervarya. Furthermore, the new generic

nomen includes the word Minerva, the Latin name of the Greek goddess of wisdom Athena,

daughter of Zeus (Jupiter in Latin), who, according to her priests, leapt fully armed from the

skull of her father: this evokes the behaviour of these tiny frogs, which suddenly jump,

apparently from the mud or the ground, just like under the feet of a walking person.

Source : MNHN, Paris

Duois, OHLER & BU 73

Minervarya sahyadris sp. nov.

Holotype. - MNHN 2000.3031, adult male (SVL 19.1 mm), collected on 25 July 1984 by Alain

Dubois along a rivulet in forest near the village of Gundia (13°05°N, 76°07'E:; altitude about

200 m), Kempholey forest, Hassan District, Karnataka, India.

Paratopotypes. - MNHN 2000.3026-3030, 2000.3032-3035, 5 adult males and 5 adult

females, collected from 24 to 28 July 1984 by Alain Dubois at the same locality as the

holotype.

Other paratypes. —- TBGRI 2001.0002-0007 and ZSI/SRS VA.1105, seven males, collected on

24 August 1999 byS. D. Biju at Mukkam (11°15°N, 75°43'E; altitude about 40 m), Kozhikodu

District, Kerala, India; MNHN 2000.3036 and TBGRI 2001.0001, one adult male and one

adult female, collected by S. D. Biju on 20 July 2001 in Thiruvanpady area (altitude about

30 m), about 30 km away from Kozhikodu, Kerala, India.

Description of holotype. - MNHN 2000.3031, adult male (fig. 3, 5), from Gundia (Karnataka,

India).

(A) Size and general aspect. — (1) Specimen of small size (SVL 19.1 mm), body mode-

rately slender.

(B) Head. - (2) Head of moderate size, longer (HL 6.1 mm) than wide (HW 7.5 mm; MN

6.74 mm; MFE 5.38 mm; MBE 2.92 mm), convex. (3) Snout oval, protruding, its length (SL

3.05 mm) longer than horizontal diameter of eye (EL 2.66 mm). (4) Canthus rostralis

rounded, loreal region concave, angle to upper surface of snout scarcely obtuse. (5) Interor-

bital space flat, larger ([UE 1.75 mm) than upper eyelid (UEW 1.62 mm) and as large as

internarial distance (IN 1.75 mm); distance between front of eyes (IFE 3.24 mm) two third of

distance between back of eyes (IBE 4.93 mm). (6) Nostrils rounded, with flap of skin laterally,

closer to eye (EN 1.30 mm) than to tip of snout (NS 1.43 mm). (7) Pupil indistinct. (8)

Tympanum (TYD 1.17 mm) distinct, rounded; about half of eye diameter, tympanum-eye

distance (TYE 0.39 mm) one third its diameter. (9) Pineal ocellus absent. (10) Vomerine ridge

present, bearing few small teeth, between posterior parts of choanae, with an angle of 40° to

body axis, closer to choanae than to each other, shorter than distance between them. (11)

Tongue moderate, oval, slightly emarginate; median lingual process absent. (12) Supratym-

panic fold prominent, from eye to shoulder. (13) Parotoid glands absent. (14) Cephalic ridges

absent. (15) Co-ossified skin absent.

3.95 mm), not enlarged. (17) Fingers rather long, thin (TFL 2.27 mm). (18) Relative length of

fingers, shortest to longest: IV < II < 1 < HI. (19) Tips of fingers bluntly rounded, not

enlarged. (20) Fingers without dermal fringe; webbing absent. (21) Subarticular tubercles

prominent, rounded, single, all present. (22) Prepollex oval, prominent: two round, distinct

palmar tubercles; supernumerary tubercles absent.

(D) Hindlimbs. — (23) Shanks three times longer (TL 8.1 mm) than wide (TW 2.92 mm),

longer than thigh (FL 7.2 mm), but shorter than distance from base of internal metatarsal

tubercle to tip of toe IV (FOL 8.7 mm). (24) Toes long, thin; toe IV long (FTL 5.70 mm) more

Source : MNHN, Paris

74 ALYTES 19 (2-4)

than one third of distance from base of tarsus to tip of toe IV (TFOL 13.5 mm). (25) Relative

length of toes, shortest to longest: I < II < V < III < IV. (26) Tips of toes rounded, not

enlarged. (27) Webbing rudimentary: 1 2-2 V4 112-3 1/3 111 3-41V 4-22/3 V (MTTF 3.31

mm; MTFF 3.37 mm; TFTF 5.77 mm; FETE 5.96 mm). (28) Dermal fringe along toe V

absent. (29) Subarticular tubercles prominent, oval, simple, all present. (30) Inner metatarsal

tubercle rather long, very prominent, spike-shaped; its length (IMT 0.72 mm) 2.22 times in

length of toe I (ITL 1.59 mm). (31) Inner tarsal ridge present. (32) Outer metatarsal tubercle

present, small, rounded; supernumerary tubercles absent; tarsal tubercle absent.

(E) Skin. — (33) Dorsal and lateral parts of head and body: snout, between the eyes and

side of head smooth; back with indistinct, interrupted longitudinal glandular folds: upper

part of flanks with glandular warts; lower part of flanks smooth. (34) Latero-dorsal folds

absent (one of the dorsal folds in about the same position). (35) Dorsal parts of limbs:

forelimbs smooth; thigh, shank and tarsus with indistinct glandular warts. (36) Ventral parts

of head, body and limbs: throat and chest with dense glandular warts; belly smooth; thigh

ventrally smooth, zone surrounding vent and posterior part of thigh with dense glandular

warts. (37) Rictal gland present, just behind mouth commissure.

(F) Coloration in alcohol. — (38) Dorsal and lateral parts of head and body: dorsal parts

of head and dorsum fawn colored, with indistinct mid-dorsal line and lighter longitudinal

bands; a large light beige band from posterior border of eye to groin, underlined ventrally by

a brown band; loreal region, tympanic region, supratympanic fold and tympanum brown;

upper lip with a whitish horizontal band; rictal gland white. (39) Dorsal parts of limbs:

forelimbs, thigh, shank and foot beige with brown bands: posterior part of thigh light brown.

(40) Ventral parts of head, body and limbs: throat light grey; margin of throat yellowish

without spots or bands; chest, belly and thigh yellowish. Vocal sac light grey in its anterior

part and yellowish in its posterior part (including throat and chest).

(G) Male secondary sexual characters. - (41) Nuptial spines present, one single patch on

prepollex and finger I up to half penultimate phallange: numerous, very small, whitish spines.

(42) Vocal sac present, unique subgular pouch, marked by glandular skin on middle of throat

and anterior chest; a pair of rounded openings in rather posterior part of mouth floor. (43)

No other male secondary characters.

Variation. - Measurements of minervarya specimens are given in tab. 5-6. Due to the small

number of specimens no statistical analysis has been performed. AI16 adult males measured are

smaller (SVL 17.2-19.1 mm) than the 4 adult females (SVL 20.6-23.0 mm), a sex size dimor-

phism which, although not testable statistically (DUBOIs, 1984), is probably significant and

general in the species. Individuals of different origin seem very similar in their measurements,

except concerning the limbs: the forelimb (FLL), hand (HAL), third finger (TFL) and tibia

(TL) appear shorter in the specimens of both sexes from Karnataka than in the males from

Kerala: at least the two series of measurements do not overlap in range. Such a trend would

need to be confirmed on much larger series of specimens. In the specimens from Gundia in life,

when the hind leg was extended anteriorly along flank, the heel reached a point situated from

back of tympanum to slightly beyond back of eye. The series from Kerala is much clearer in

coloration, but shows no differentiation in color pattern. This variation may be due to storage

and age of the specimens. The males from Kerala have a greyish pattern on the throat which

Source : MNHN, Paris

Duois, OHLER & BUU 75

reminds the gular “W” of fejervarya males. In the minervarya specimens from Karnataka, the

greyish zone is much more indistinct. All males have a glandular skin under the vocal sac which

extends to the anterior part of the chest. The rictal gland can be observed on all specimens.

Colours in life (fig. 4). — Iris clear golden in its upper part, darker in its lower part. Pupilla

horizontal oval, continued anteriorly and posteriorly in iris by a dark horizontal bar, and

inferiorly by a narrow dark vertical line. Sides of head dark brown, darker in tympanic region.

Upper lip with a bright white bar, from snout to below tympanum or posteriorly, particularly

bright in its anterior part. Mid dorsum brick red, reddish, reddish brown, brown, greyish or

golden, often with a creamish, golden, yellowish or reddish mid-dorsal line or band (indistinct

after fixation). Indistinct, discontinuous colored bands on sides of back, especially in its

posterior part, in the usual place of latero-dorsal folds in frogs, but supported by folds that are

not different from or more prominent than the other longitudinal folds of back. Upper flanks

dark greyish with indistinct clearer zones, lower flanks darker. Upper arm brick red or brown.

A clear median line sometimes present on upper part of hind leg from vent to heel. Throat

translucent or yellowish in its anterior part, followed by a greyish zone, and by a yellow

posterior part. Chest and belly bright white, yellow or yellowish. Translucent fejervaryan lines

quite distinct. Inferior part of thighs translucent or clear yellowish.

Natural histo: In Karnataka the species was collected in open habitat in forested area (Kem-

pholey forest); in Kerala it was found near paddy fields and in an abandoned quarry. During

monsoon (June-July), the males call from terrestrial calling sites, never in water, but always tur-

ned towards water (small ponds or rivulets). Calls are rapid sequences of high-pitched “chick.

chick...”, reminding the jingling of a bunch of keys. Amplexusis axillary. Eggs masses are atta-

ched to aquatic vegetation. They arecomposed of 20-35 eggs that are 2mmin diameter and pig-

mented. Tadpoles (that will be described in detail elsewhere) are typical ranid tadpoles, with

unspecialized, ventrally directed mouthparts and a keratodont formula of 1:1+1/1+1:2. In

captive condition they were observed to reach metamorphosis within 28 days.

Etymology of the specific nomen. — The specific nomen is the Sanskrit name Sahyadris,

meaning “the mountains” (adri) “of the Western Ghats” (Suhyan): it refers to the area of

distribution of the species.

ACKNOWLEDGEMENTS

For the help received in India in 1984, AD is grateful to K. Sundar Naik (Hass

(Sakleshpun). SDB would like to thank the Kerala Forest Department for giving necessary permission to

survey, the Tropical Botanic Garden and Research Institute (TBGRI) for extending facilities to conduct

this study, and Anil Zachariah (Wayanad) for support in the collection of specimens from Kerala. This is

publication N°46 of the PPF “Faune et flore du sud-est asiatique” (N°45, see DaviD et al., 2001).

LITERATURE CITED

ANONYMOUS [International Commission on Zoological Nomenclature], 19994. - International code of

zoological nomenclature. Fourth edition. London, International Trust for zoological Nomencla-

ture: i-xxix + 1-306.

Source : MNHN, Paris

76 ALYTES 19 (2-4)

ANONYMOUS, 19994. — SPSS advanced models 9.0. Chicago, SPSS Inc.: i-xv + 1-497.

ANDERSSON, L. G., 1937. — Reptiles and Batrachians collected in the Gambia by Gustav Svensson and

Birger Rudebeck (Swedish expedition 1931). 4rk. Zool., 29A (16): 1-28.

ANNANDALE, N., 1918. - Some undescribed tadpoles from the hills of southern India. Rec. indian Mus.,

15: 17-24, pl. 1.

—— 1919. - The tadpoles of Nyctibatrachus pygmaeus and Ixalus variabilis: a correction. Rec. indian

Mus., 16: 303.

BHADURI, J. L. & KRIPALANI, M. B., 1955. - Nyctibatrachus humayuni, a new frog from the Western

Ghats, Bombay. J Bombay nat. Hist. Soc., 52: 852-859.

Buu, S. D., 2000. - Two frogs and hundred years. Biodiversity India, ISCB Newsletter, 8-12: 10-11.

BLOMMERS-SCHLÔSSER, R. M. A., 1993. — Systematic relationships of the Mantellinae Laurent, 1946

(Anura, Ranoïdea). Ethol. Ecol. Evol., 5: 199-218.

BozKay, S. J., 1915. - Beiträge zur Osteologie einiger exotischer Raniden. Anat. Anz., 48: 172-183.

BossuyT, F. & Dugois, A., 2001. — A review of the frog genus Philautus Gistel, 1848 (Amphibia, Anura,

Ranidae, Rhacophorinae). Zeylanica, 6 (1): 1-112.

Bossuyr, F. & MinnkovircH, M. C., 2000. - Convergent adaptive radiations in Madagascan and Asian

ranid frogs reveal covariation between larval and adult traits. Proc. nat. Acad. Sci. USA, 97 (12):

6585-6590.

ue 2001. Amphibians as indicators of early Tertiary “out of India” dispersal of vertebrates. Science,

292: 93-95.

BOULENGER, G. A., 1882. - Catalogue of the Batrachia Salientia s. Ecaudata in the collection of the British

Museum. London, Taylor & Francis: i-xvi + 1-503, pl. 1-30.

ee 1890. — The fauna of the British India, including Ceylon and Burma. Reptilia and Batrachia. London,

Taylor & Francis: i-xviii + 1-541.

_— 1920. - À monograph of the South Asian, Papuan, Melanesian, and Australian frogs of the genus

Rana. Rec. indian Mus., 20: 1-226.

BouRRET, R., 1942. - Les Batraciens de l'Indochine. Hanoi, Institut océanographique de l’Indochine: i-x

+ 1-547,4 pl.

CHaRi, V. K., 1962. — A description of the hitherto undescribed tadpole of, and some field notes on the

fungoid frog, Rana malabarica Bibron. J. Bombay nat. Hist. Soc., 59: 71-76, 1 pl.

Ciarke, B. T., 1981. - Comparative osteology and evolutionary relationships in the African Raninae

(Anura Ranidae). Monit. zoo. ital., (n.s.), suppl. 15: 285-331.

— 1983. - À morphologic. xamination of the frog genus Nannophrys (Anura: Ranidae) with

comments on its biology, distribution and relationships. Zoo!. J Linn. Soc., 79: 377-398.

Davi, P., VIDAL, N. & PAUWELS, O. S. 001. — À morphological study of Stejneger’s pitviper

Trimeresurus stejnegeri (Serpentes, Viperidae, Crotalinae), with the description of a new species

from Thailand. Russian J. Herp., 8 (3): 21

Decker, K., 1938. - Beitrâge zur Osteologie und

Freunde Berlin, 1938: 127-184.

DELORME, M, DUBois, A., KosUCH, JL & VENCES, M., submitted. - Molecular evidence for higher-level

endemism among South Asian amphibians: the relationships of ‘Rana” corrugata from Sri

Lanka.

Dusois, A. 1975. - Un nouveau sous-genre (Paa) et trois nouvelles espèces du genre Rana. Remarques

sur la phylogénie des Ranidés (Amphibiens, Anoures). Bull. Mus. natn. Hist. nat., (3), 324 (Zool.

231): 1093-1115.

e 1980. - Notes sur la systématique et la répartition des Amphibiens Anoures de Chine et des régions

avoisinantes. IV. Classification générique et subgénérique des Pelobatidae Megophryinae. Bull.

Soc. linn. Lyon, 49: 469-482.

re 1984. - Sample-size constraints in the use of the nonparametrie Mann-Whiney U test for the

stematik ranider Froschlurche. Sber. Ges. Naturf.

987b. - Miscellanea nomenclatorica batrachologica (XV). Alytes, 5 (4): 175-176.

988a. - Some comments on the genus concept in zoology. Monit. ool. ital., (n. s.), 22: 27-44.

9886. - The genus in Zoology: a contribution to the theory of evolutionary systematics. Mém. Mus.

nan, Hist, nat. (A), 140: 1-123.

Source : MNHN, Paris

Dugois, OBLER & BuU 77

ES 1992. - Notes sur la classification des Ranidae (Amphibiens, Anoures). Bull. Soc. linn. Lyon, 61 (10):

305-352.

Je 1995. - Keratodont formulae in anuran tadpoles: proposals for a standardization. J 2001. Syst. Evol

Res., 33 (1): I-XV.

_—. 1999. - Miscellanea nomenclatorica batrachologica. 19. Notes on the nomenclature of Ranidae and

related groups. A/ytes, 17 (1-2): 81-100.

--- 2000. - The influence of man on the distribution of amphibians in the Himalayas of Nepal: an

example of critical evaluation of biogeographical data. In: G. MIEHE & Y. ZHANG (ed.), Environ

mental changes in high Asia, Marburger geogr. Schrifien, 135: 326-345.

Duois, A. & OHLER, A., 1998. —- A new species of Leptobrachium (Vibrissaphora) from northern

Vietnam, with a review of the taxonomy of the genus Leptobrachium (Pelobatidae, Megophryi-

nae). Dumerilia, 4 (1): 1-32.

= 1999. — Asian and Oriental toads of the Bufo melanostictus, Bufo scaber and Bufo stejnegeri groups

(Amphibia, Anura): a list of available and valid names and redescription of some name-bearing

types. J South Asian nat. Hist., 4 (2): 133-180.

_—— 2000. - Systematics of Fejervarya limnocharis (Gravenhorst, 1829) (Amphibia, Anura, Ranidae) and

related species. 1. Nomenclatural status and type-specimens of the nominal species Rana limno-

charis Gravenhorst, 1829. Alytes, 18 (1-2): 15-50.

Durra, S. K., 1997. — Amphibians of India and Sri Lanka. (Checklist and bibliography). Bhubaneswar,

Odyssey Publishing House: [iii] + i-xiii + 1-342 + i-xxii,

Durra, S. K. & MANAMENDRA-ARACHCHI, K., 1996. — The amphibian fauna of Sri Lanka. Colombo,

Wildlife Heritage Trust of Sri Lanka: 1-232.

EMERSON, S. B., INGER, R. F. & IskANDAR, D., 2000. - Molecular systematics and biogeography of the

fanged frogs of southeast Asia. Mol. Phyl. Evol., 16 (1): 131-142.

Ft, L. (ed.), 1999. — Alas of amphibians of China. Zhengzhou (China), Henan Press of Science and

Technology: [i-ii] + 1-432. [In Chinese.]

GRANT, T., HUMPHREY, E.C. & Myers, C.W., 1997. — The median lingual process of frogs: a bizarre

character of Old World ranoïds discovered in South American dendrobatids. Am. Mus. Novit..

3212: 1-40.

Hey, W.R., 1994. - Variation within the Leptodactylus podicipinus-wagneri complex of frogs (Amphi-

ia: Leptodactylidae). Smithsonian Contrib. Zool., 546: i-iv + 1-124.

ER, R. F,, 1966. The systematics and zoogeography of the Amphibia of Borneo. Fieldiana: Zool., 52:

1-402.

1985. - Tadpoles of the forested regions of Borneo. Fieldiana: Zool., (n.s.), 26: i-v + 1-89.

INGER, R. F., SHAFFER, H. B., KosHy, M. & BAKDE, R., 1984. — A report on a collection of amphibians and

reptiles from the Ponmudi, Kerala, South India. J Bombay nat. Hist. Soc., 81 (2): 406-427, pl. 1-3.

1987, — Ecological structure of a herpetological assemblage in south India. Amphibia-Reptilia, 8:

189-20:

KIRTISINGE , 1957. - The Amphibia of Ceylon. Colombo: i-xiii + 1-12, 1 pl.

1958. - Some hitherto undescribed anuran tadpoles. Ceylon J. Sci., (Biol. Sci. ), 1: 171-176.

KosucH, J., VENCES, M., Dugois, A., OHLER, A. & BÔHME, W., 2001. — Out of Asia: mitochondrial DNA

evidence for an Oriental origin of tiger frogs, genus Hoplobatrachus. Mol. Phyl. Evol., 21 (2), in

press.

LAMOTTE, M. & PERRET,

1603-1644.

LAURENT, R., 1950. — Reptiles et batraciens de la région de Dundo (Angola du Nord-Est). (Première

note). Publço: Diam. Angola,

—— 1986. - Sous-cla mphibiens (Lissamphibia). Systématique. /n: P.-P. GRAssÉ & M. DELSOL

(ed.), Traité de Zoologie, 4, Amphibiens, fase. 1-B, Paris, Masson: 594-796.

LEMEN, C. A. & FREEMAN, P. W., 1984. - The genus: a macroevolutionary problem. Evolution, 38:

1219-1237.

MaRMAYOU, J., DUBOIS, A., OHLER, PAsQUET, E. & TILLIER, A., 2000. — Phylogenetic relationships in

the Ranidae. Independent origin of direct development in the genera Philautus and Taylorana. C.

r Acad. Sci, Sci. Vie, 323: 287-297.

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. 4m. Mus. Novit., 2752: 1-32.

J.-L., 1968. — Révision du genre Conraua Nieden. Bull. IF A.N., (A), 30:

Source : MNHN, Paris

78 ALYTES 19 (2-4)

MYERS, G.S., 1942a. — A new frog from the Anamallai Hills, with notes on other frogs and some snakes

from south India. Proc. biol. Soc. Wash., 55: 49-55.

--- 1942b, — À new frog of the genus Micrixalus from Travancore. Proc. biol. Soc. Wash., 55: 71-74.

Myers, N., 1990. - The biodiversity challenges: expanded Hot Spots analysis. The Environmentalist, 10:

243-256.

Norusis, M. J., 1992. — SPSS for Windows. Professional Statistics Release 5. Chicago, SPSS Inc.: [i-xii] +

1-348.

OuLer, A., 1996. — Systematics, morphometries and biogeography of the genus Aubria (Ranidae,

Pyxicephalinae). Alytes, 13 (4): 141-166.

ee 1999. - The identity of Dendrobatorana Ahl, 1927 (Amphibia, Ranoïdea). Mitt. Mus. Naturkd.

Berlin, 75: 37-45.

OuLeR, A. & Dunois, A., 1989. - Démonstration de l'origine indépendante des ventouses digitales dans

deux lignées phylogénétiques de Ranidae (Amphibiens, Anoures). C. r Acad. Scï., (3), 309:

419-422.

== 1999. — The identity of Elachyglossa gyldenstolpei Andersson, 1916 (Amphibia, Ranidac), with

comments on some aspects of statistical support to taxonomy. Zoologica Scripta, 28 (3-4):

269-279.

PiLLAI, R. S., 1978. — On Nyctibatrachus major Boul. (Ranidae) with a description of its tadpole. Bull.

zool. Surv. India, 1: 135-140.

RA40, C. R. N., 1920. - Some south Indian batrachians. J Bombay nat. Hist. Soc., 27: 119-127, pl. 1-2.

SEKAR, A. G., 1992. - Morphometry, habitat, behaviour and food of the tadpoles of Leith’s frog Rana

leithüi. J Bombay nat. Hist. Soc., 89: 259-261.

SHAFFER, H. B., 1988. — Size and scaling in the Indian frogs Nyctibatrachus and Nannobatrachus

(Ranidae). Fieldiana: Zool., (n.s.), 46: i-iüi + 1-10.

Surru, M.A., 1924. - Descriptionsof Indian and Indochinese tadpoles. Rec. indian Mus.,26: 131-143, pl. 7.

Vera, M. Kosucn, J., OuLer, A. & Dumois, À., 2001. - Systematics of Fejervarya limnocharis

(Gravenhorst, 1829) (Amphibia, Anura, Ranidae) and related species. 2. Morphological and

molecular variation in frogs from the Greater Sunda Islands (Sumatra, Java, Borneo) with the

definition of two spel lytes, 19 (D): 5-28.

VENCEs, M. & GLAW, F,, 2001. - When molecules claim for taxonomic changes: new proposals on the

classifeation of Old World treefrogs. Spixiana, 24: 85-92.

Vences, M. GLAW, F., KosucH, J., Das, L. & Verrn, M., 2000a. - Polyphyly of Tomopterna (Amphibia:

Ranidae) based on sequences of the mitochondrial 16S and 12S rRNA genes, and ecological

biogeography of Malagasy relict amphibian groups. R. LOURENÇO & $, M. GOODMAN (ed.),

Diversité et endémisme à Madagascar, Mém. Soc. Biogéogr, Paris: 229-242.

VENCES, M., WaANKE, S., ODIERNA, G., KosucH, J. & VErrH, M., 2000b. — Molecular and karyological

data on the south Asian ranid genera /ndirana, Nyctibatrachus and Nannophrys (Anura: Ranidae).

Hamadryad, 25 (2): 75-82.

APPENDIX 1

COMPARATIVE MATERIAL EXAMINED

Specimens marked with an asterisk (*) were included in the morphometric analyses (tab. 3-4,

fig. 1-2).

Conraua alleni (Barbour & Loveridge, 1927). — SizRRA LEONE: Region of Loma: MNHN 1979.6136-6146.

Conraua beccari (Boulenger, 1911). - ErmiOpia: Gondar, 2000 m: MNHN 1933.0021.

Conraua crassipes (Buchholz & Peters, 1875). - NiGErA: Jdanre: MNHN 1995.5704-5708, 19955711,

1995.5715.

Conraua derooi Hulselmans, 1972. - ToGo: Dangi Atigba: MNHN 1978.2026-2031; Misahühe: MNHN

1993.4084-4087.

Source : MNHN, Paris

Dugois, OHLER & BuU 79

Conraua goliath (Boulenger, 1906). - EQUATORIAL GUINEA: MNHN 1992.5316.

Euphlyctis cyanophyetis (Schneider, 1799). - NepaL: Dillikot, 2400 m: MNHN 1975.2164*, 1975.2182*-

2183*, 1975.2194*, 1975.2196*; Sanichare: MNHN 1977.1364-1403; Suikher, 900 m: MNHN 1996.9274-

9280; Tatopani Khola, 2200 m: MNHN 1975.2250-2273.

Euphlyctis hexadactylus (Lesson, 1834). — InDia: NMW 2512.1*-5*, 25121*,

Fejervarya greeni (Boulenger, 1904). - Sri Lanka: Ninara Eliya: MNHN 2000.0617.

Fejervarya keralensis (Dubois, 1981). — INDIA: Coonoor, Nilgiris: MNHN 1902.0124-0127; Genji, Coro-

mandel: MNHN 1902.0172.

Fejervarya limnocharis (Gravenhorst, 1829). — INDONESIA: Java, Bogor: FMNH 256721*-256724*; Java,

Chianjur: FMNH 256727*-256728*, 256732*; Sumatra, Medan: FMNH 256758*; Sumatra, Seleuk

FMNH 256769* ; Sumatra, Sidikalang: FMNH 256762*. — LaOS: Luang Prabang: MNHN 1999.6093. —

THAILAND: Khao Chong, Trang: MNHN 1987.2357-2385. — VirraM: Sapa : MNHN 1996.3373-3376.

Fejervarya nepalensis (Dubois, 1975). - NepaL: Godavari: MNHN 1975.1607-1624; Joubari: MNHN

1975.1640-1645.

Fejervarya nilagirica (Jerdon, 1853). - Ivoua: Udhagamangalam, Nilgiris: MNHN 1984.2334-2342.

Fejervarya pierrei (Dubois, 1975). - NepaL: Birtamode: MNHN 1975.1681-1710.

Fejervarya rufescens (Jerdon, 1853). — INDIA: Gundia, Karnataka: MNHN 1984.2349-2355.

Fejervarya syhadrensis (Annandale, 1919). - NEPAL: Dongison: MNHN 1975.2038-2062; Tir Pani Khola:

MNHN 1996.9259-0265.

Fejervarya teraiensis (Dubois, 1984). - NepaL: Birtamode: MNHN 1976.1029-1040.

Fejervarva vittigera (Wiegmann, 1834). - PriLiprines: Manilla: MNHN 1900.0435-0437, 1994.0568.

Hoplobatrachus chinensis (Osbeck, 1765) [valid nomen of the species often referred to as Rana rugulosa

Wiegmann, 1834: see KosucH et al., 2001, and Dumois & OHLER, in preparation]. -CHiNA: “Canton”:

BMNH 1933.12.4.18*, Cap Syngmore: ZMB 3271* ; Hong Kong: NMW 2614.5*. - Laos: Vientiane,

market: MNHN 1997.4900%. - Myanmar: Pegu: BMNH 1868.4.3.90*.

Hoplobatrachus crassus (Jerdon, 1853). — INDIA: Madras: BMNH 1872.4.17.245*.

Hoplobatrachus occipitalis (Günther, 1859). — Gamma: BMNH 1947.2.29.8*, — Ivory Coasr: Soubré:

MNHN 1990.4428; Soubré-Sanandra: MNHN 1993.2845 ; Tabou: MNHN 1990.4386-4389. — LIBERIA:

Nimba, Grassfield: MNHN 1990.4380-4385.

Indirana brachytarsus (Günther, 1876). - INDIA: Ponmudi, Trivandrum District, Kerala: FMNH 217954.

Indirana diplosticra (Günther, 1876). — INDIA: Malabar: BMNH 1947.2.3.27.

Limnonectes kuhlit (Tschudi, 1838). — INDONESIA: Sumatra, Sidikalang: MV 80*, MV 82*, MV 102*-103*,

MV 105*, MV 108%, MV 111#-112*, MV 117*, MV 127%.

Rana (Clinotarsus) curtipes Jerdon, 1853. - INDIA: Thekaddi, Periy

1989.2751.

Rana (Hydrophylax) malabarica Tschudi, 1838. — INDIA: Bengal: MNHN 4439, 1989.3448; Malabar:

MNHN 771, 4440, 1989.3451-3452

Gravenhorst, 1829. - Ski LANKA: Belihuloya: MNHN 2000.0614.

yhirana) temporalis Günther, 1864. — INDIA: Coonoor, Nilghiris: MNHN 1902.0128, Gundia,

Karnataka: MNHN 1985.584-586. - Sri LaNKA: Kitulgala: MNHN 20000613.

Sphacrotheca pluvialis (Jerdon, 1853) [valid nomen of the species often referred to as Rana breviceps

Schneider, 1799: see DuBoïs, 2000, and Dugois & OHLER, in preparation]. - INDIA: Madras: BMNH

1874.4.29+, 1947.2.28,55*-56*, 1947.2.28.58*, MSNG 28519*. - Myanmar: North Chin Hills: BMNH

1893.11.17.4%. - NépaL: Burimorang: MNHN 1997.5100*-5119*; Tékunala: MNHN 1983.0808*-0812*.

= PaRISTAN: Lower Hub River: MNHN 9624*, 9626*. - Sri LaNKA: BMNH 1877.3.9.8*; Kurunegala:

1973.3024*.

Tiger Reserve, Kerala: MNHN

Sylvirana graci

Corresponding editor: Miguel VENCHS.

Source : MNHN, Paris

ALYTES 19 (2-4)

Association des Amis du Laboratoire des Reptiles et Amphibiens du Muséum

(AALRAM)

SECRÉTARIAT

Laboratoire des Reptiles et Amphibiens, Muséum national d'Histoire naturelle, 25 rue Cuvier, 75005 Paris, France

Tél: (83) 1 40 79 34 87- Fax: (33) 1 40 79 34 88 - E-mail: reptamph@mnhn.fr

EXTRAIT DES STATUTS DE L'AALRAM

Les buts de association sont de défendre par tous les moyens les intérêts matériels et moraux du Laboratoire de Zoologie

(Reptiles et Amphibiens) du Muséum national d'Histoire naturelle, de soutenir son action dans le domaine de l'étude scientifique

et de la protection des Reptiles et Amphibiens et de leurs milieux, son rayonnement et enrichissement des collections nationales

dont il est le dépositaire.

Dumerilia

Le nom Dumerilia est un hommage à la mémoire d'André-Marie-Constant DumriL (1774-1860) l'un des fondateurs

de l'herpétologie, premier auteur du magistral traité intitulé Erpérologie générale (1834-1854).

Le périodique Dumerilia publie les résultats des travaux des chercheurs, associés et correspondants, du Laboratoire

des Reptiles et Amphibiens du Muséum national d'Histoire naturelle de Paris, ainsi que les articles d'autres collègues s'appuyant,

au moins en parie, sur ses collections, ou décrivant de nouveaux taxons dont es Aolotypes sont déposés dans ses collections.

TARIFS POUR 2002

Euros (€) US Dollars (USS)

Cotisation à 'AALRAM (membre act) 3 Ë

Cotisation à l'AALRAM (membre bienfaiteur) 1 2

Dumerilia (volumes précédents et souscription)

VOLUME 1, juin 1994, 128 pages. a

Patrick DAVID: Liste des reptiles actuels du monde. 1, Cheloniï

VOLUME 2, octobre 1996, 124 pages. s6

6 articles par 7 auteurs

VOLUME 3, février 1999, 500 pages.

Patrick DAVID & Ivan INEICH: Les Serpents venimeux du monde:

systématique et répartition

VOLUME 4, octobre 1998 - Juin 2001, 220 pages

Fascicule 4 (1), 32 pages.

Alain DUBOIS et Annemarie OHLER: À new species of Leprobrachiunr

(ibrissaphora) from northem Vietnam, with a review ofthe

taxonomy of the genus Leptobrachium (Pelobatidae, Megophryinae)

Fascicule 4 (2), 124 pages

4 amies par 13 auteurs

Fascicule 4 (3), 64 pages.

France DE LAPPARENT D BROIN: The European tuile fauna

From 1he Triasie 10 ihe Present: a short review

MODES DE PAIEMENT

Paiement en Francs Français: par chèque à l'ordre de “AALRAM”, adressé à: AALRAM, c/o Alain DUBOIS, Laboratoire des.

Reptiles et Amphibiens, Muséum national d'Histoire naturelle, 25 rue Cuvier, 75005 Paris, France

Payment in U.S. Dollars: by cheque payable to *AALRAM"”, sent io: Dr. John B.IVERSON, Department of Biology, Elham College,

Richmond IN 47374-409$, USA.

Source

MNHN, Paris

Alytes, 2001, 19 (2-4): 81-106. 81

À new genus for an aquatic ranid

(Amphibia, Anura) from Sri Lanka

Alain DUBoIs & Annemarie OHLER

Laboratoire des Reptiles et Amphibiens,

Muséum National d'Histoire Naturelle,

25 rue Cuvier, 75005 Paris, France

A new monotypic genus is erected for the species Rana corrugata

Peters, 1863, a ranid endemic of the island of Sri Lanka. This species shares

the paedomorphic condition of the retention of a lateral-line system in

adults with two other genera of Ranidae, Euphlyctis and Occidozyga. lt

shares with many species of the genus Limnonectes the presence of

odontoïd “fangs” on the anterior lower jaw of males. Despite these

similarities, the new genus is not closely related to the three genera above,

and appears to represent a new, independent lineage within the South

Indian ranids. It is here provisionally referred to a new subfamily of the

family Ranidae. This study provides an opportunity for a brief review of the

distribution of several interesting characters among several genera of

Ranidae, including characters related with an aquatic mode of life (general

morphology, lateral-line system, coloration of the back of thighs), male

secondary characters (fangs, large head, dorsal cephalic knob, size dimor-

phism, nuptial pads, vocal sacs, advertisement and territorial calls), amplec-

tic position, parental care and egg coloration.

ABBREVIATIONS

Measurements. — EL, eye length: EN, distance from front of eye to nostril; FFTF, distance from

maximum incurvation of web between fourth and fifth toe to tip of fourth toe; FL, femur length (from

vent to knee); FLL, forelimb length (from elbow to base of outer palmar tubercle); FOL, foot length

(from base of inner metatarsal tubercle to tip of fourth toe); HAL, hand length (from base of outer

palmar tubercle to tip of third finger); HL, head length (from back of mandible to tip of snout); HW, head

width; IBE, distance between back of eyes; IFE, distance between front of eyes; IMT, length of inne

metatarsal tubercle: IN, internarial space: ITL, inner toe length; LUE, minimum distance between upper

eyelids; MBE, distance from back of mandible to ba *E, distance from back of mandible to

front of eye; MN, distance from back of mandible to nostril, MTFE, distance from distal edge of

metatarsal tubercle to maximum incurvation of web between fourth and fifth toe; MTTE, distance from

distal edge of metatarsal tubercle to maximum incurvation of web between third and fourth toe; nn, no

measurement taken; NS, distance from nostril to tip of snout; SL, distance from front of eye to tip of

snout: SVL, snout-vent length; TFOL, length of tarsus and foot (from base of tarsus to tip of fourth toe);

TETE, distance from maximum incurvation of web between third and fourth toe to tip of fourth toe; TL,

tibia length: TW, maximum tibia width: UEW, maximum width of upper eyclid.

Source : MNHN, Paris

82 ALYTES 19 (2-4)

Museums, collections and persons. - AD, Alain Dubois; AMO, Annemarie Ohler;, BMNH, Natural

History Museum, London, United Kingdom; MNHN, Muséum National d'Histoire Naturelle, Paris,

France; MV, Michael Veith collection, Mainz, Germany; NHMB, Naturhistorisches Museum Basel,

Basel, Switzerland; NMW, Naturhistorisches Museum, Wien, Austria; NRM, Swedish Museum of

Natural History, Stockholm, Sweden; ZMB, Zoologisches Museum, Berlin, Germany.

INTRODUCTION

Taxonomy of the family Ranidae Rafinesque-Schmaltz, 1814 is among the most chal-

lenging in anuran amphibians. This is due in part to the existence of groups including many

sibling species (see e.g.: DuBois, 1977; Marsur et al., 1993; EMERSON & WARD, 1998), and of

numerous cases of convergence between species belonging to distinct lineages (see e.g.: OHLER

& Dugois, 1989; BossuyT & MILINKOVITCH, 2000; MARMAYOU et al., 2000).

Particularly confused and controversial is the taxonomy of the so-called genus Rana

Linnaeus, 1758 (sensu BOULENGER, 1920), that has until now been used to group frogs

belonging to various lineages but showing “generalized” morphologies and unspecialized

plesiomorphic characters. As far back as in 1915, however, BOLKAY had proposed to remove

from this genus several species with a forked omosternum and other unusual characters for the

genus Rana, and to place them in the new genus Fejervarya. A similar proposal was made by

DECKERT (1938), followed by LAURENT (1950) and others, who used the generic nomen

Dicroglossus Günther, 1860 for these frogs: in particular, for several decades, the common

African frog now known as Hoplobatrachus occipitalis (see e.g. KosucH et al., 2001) was

referred to under the nomen Dicroglossus occipitalis. However, this was ignored by many

other authors, especially those working on the Asian fauna (see e.g.: BOURRET, 1942; Liu,

1950; INGER, 1954, 1966, 1985; Liu & Hu, 1961; TAYLOR, 1962; etc.). DuBois (1974) was the

first author to use the nomen Dicroglossus for Asian frogs, before showing (DUBoIs, 1975) that

this nomen was a strict junior synonym of Euphlyctis Fitzinger, 1843. The same author

subsequently distributed the ranid species with forked omosternum in several subgenera of

Rana (DuBois, 1981), and later in several distinct genera (DuBoIs, 1987b, 1992). Among the

five subgenera he recognized in Limnonectes Fitzinger, 1843 in 1987, Dugois (1992) raised

Hoplobatrachus Peters, 1863 and Taylorana Dubois, 1987 to the rank of distinct genera, and

Dugois & OHLER (2000) did the same for Fejervarya. OuLer & DuBois (1999) showed that

Bourretia Dubois, 1987 was a junior synonym of Ælachyglossa Andersson, 1916. Therefore,

according to these latter authors, the genus Limnonectes is now understood as including two

subgenera, Elachyglossa and Limnonectes, the latter with three species-groups (DUBOIS,

1987b: 63).

In the genus Limnonectes, the Limnonectes kuhlii group corresponds to the Ranae

kuhlianae of BOULENGER (1920). Adult males of these frogs are devoid of vocal sacs and

nuptial pads, but have very enlarged heads and strong tooth-like prominences (“fangs”) on

the anterior lower jaw. AIl species of this group occur in South-East Asia (Indonesia,

Malaysia, Indochina and southern China), except for one, originally described by PETERS

(1863) as Rana corrugata, which lives only in Sri Lanka. When he first erected the uhlii

group, Dugois (1987b) followed BOULENGER (1920) in including this Sri Lankan species in this

group, although he had never had an opportunity to examine à specimen of this species.

Source : MNHN, Paris

Dugois & OHLER 83

However, as soon as he first saw this species alive in the field, on 30 June 1999 in Morningside

in Sri Lanka, he realized that all published descriptions of this species (PETERS, 1863;

GÜNTHER, 1864; BOULENGER, 1882, 1890, 1920; KiRTISINGHE, 1957; DUTTA & MANAMENDRA-

ARACHCHI, 1996) were incomplete or even partly inaccurate, and that the external characters

of this species (see below) were in several respects quite different from those of the L. kuhlii

group and justified the exclusion of this species not only from this group but also from the

genus Limnonectes.

Recently, some molecular cladistic data were published concerning these frogs. After an

analysis of parts of the mitochondrial ribosomal 12S and 16S genes of several species,

EMERSON et al. (2000: 136) wrote that “the fanged frogs constitute a monophyletic group” and

that “it seems appropriate, in the future, to refer to these frogs as members of the genus

Limnonectes”. While doing so, however, they did not provide a list of taxa that they referred

to this genus, so that one can infer that they probably adopted Dugois's (1992) concept of the

latter. However, they provided (EMERSON et al., 2000: 131) a “definition” of “fanged frogs”?

that does not apply to all species or species-groups of this genus. AÏl characters listed in this

“definition” either apply to some of these taxa only (see e.g.: BOULENGER, 1920; Dugois,

1987b, 1992), such as fangs and “voicelessne: (see below) in adult males, sexual size

dimorphism or parental care. However, using this “definition”, it is quite clear that Rana

corrugata should be included in the genus Limnonectes. These authors did not, however,

consider this species in their study.

Other recent studies provided additional data in this respect. Using mitochondrial 12S

and 16$ rRNA gene sequences, VENCES et al. (2000) and DELORME et al. (submitted) found

that Rana corrugata is not cladistically a member of the group including L. kuhlii, the

type-species of Limnonectes. BOSSUYT & MILINKOVITCH (2000) found a similar result using

the same genes but also two nuclear DNA gene sequences. These data confirm the morpho-

logical interpretation of R. corrugata as not belonging in the genus Limnonectes.

On the basis of the molecular phylogenetic data mentioned above, DELOR:

(submitted) followed DuBois (1992) in recognizing in the Ranidae a subfamily Dicroglossi

Anderson, 1871, and, within the latter, a tribe Limnonectini Dubois, 1992 for the genera

Limnonectes and Taylorana. They excluded Rana corrugata from this tribe, suggesting that the

latter deserves erection of a new genus, that represents an hitherto unsuspected new clade

Within the Ranidae. In the frame of the current “working taxonomy” of the latter family

(Dusois, 1999), we suggest that this clade be recognized provisionally as a new subfamily, and

we hereby propose a diagnosis and a nomen both for this subfamily and for its unique genus.

In order to faciltate the discussion below, we introduce the new nomina first, so that we can use

them in the rest of the paper. According to Kelum Manamendra-Arachchi (personal com-

munication), some differences exist between low and high altitude populations currently

referred to this species, so that later two distinct taxa (species or subspecies) might have to be

distinguished. In order to clarify the nomenclatural decisions that might have to be taken in

t pect, we provide a detailed redescription of one of the three original syntypes of this

nominal species, that we hereby designate as lectotype. In the final part of the paper, we

discuss the distribution of some morphological characters among several genera of Ranidae,

that give support to our taxonomic decisions.

Source : MNHN, Paris

84 ALYTES 19 (2-4)

MATERIAL AND METHODS

The list of specimens examined and measured is given below in tab. 4 and in app. 1.

Twenty-two measurements of adult and young specimens were taken by AMO with a

slide calliper to the nearest 0.1 mm, or, for values below 5 mm, with an ocular micrometer to

the nearest 0.01 mm. The list of measurements is given above under Abbreviations.

In order to facilitate comparisons, the description’s methodology and plan used in

the lectotype description below were the same as those used in previous works on Asian

anurans (DUBois & OHLER, 1998, 1999, 2000; OnLer & DuBois, 1999; BossuyT & DuBois,

2001; Verrx et al., 2001; Dupois et al., 2001). The webbing formula is given according

to Myers & DUELLMAN (1982) and the tadpole keratodont formula according to DuBois

(1995).

Morphometric analyses and graphs were made using the SPSS statistical programs

for personal computers (Norusis, 1992; ANONYMOUS, 1999). We used principal compon-

ent analysis using varimax rotation (ANONYMOUS, 1999: 426) to show morpholog-

ical distinctiveness of the new genus and canonical discriminant analysis to indicate morpho-

logical discrimination from the subgenera and species-groups of Limnonectes. Oneway

analysis using Scheffe tests were performed on ranked ratios of all measurements between the

seven genera of Ranidae compared in tab. 1. Detailed results of this analysis can be

communicated upon request by the authors; they are not provided here because of space

limitations.

TAXONOMIC NOVELTIES

Subfamily Lankanectinae nov.

Type-genus, by present designation. - Lankanectes gen. nov.

Diagnosis. — This subfamily is distinguished from all other subfamilies of Ranidae by the

following combination of characters: (1) omosternum forked at base; (2) vomerine teeth

present; (3) median lingual process absent; (4) femoral gland absent; (5) extremities of digits

pointed or slightly rounded, not enlarged; (6) tarsal fold present; (7) lateral-line system

present in adult; (8) head and back covered by a network of ridges:; (9) adult male without

nuptial pads, but with fangs and internal vocal sacs: (10) eggs pigmented; (11) tadpole with

ventral mouthparts, keratodont formula 2/3.

Distribution. — So far, this subfamily is known only from the island of Sri Lanka.

Source : MNHN, Paris

DuBois & OHLER 85

Genus Lankanectes nov.

Type-species, by present designation. — Rana corrugata Peters, 1863.

Diagnosis. — This genus is distinguished from all other genera of Ranidae by the following

combination of characters: (1) omosternum forked at base; (2) size medium (adult SVL 33-

65 mm); (3) internarial distance shorter than distance between upper eyelids; (4) upper eyelids

covered with numerous round warts; (5) canthus rostralis indistinct, loreal region slightly

convex; (6) edge of lower jaw without transverse bands; (7) tympanum indistinct; (8) vomerine

teeth present; (9) median lingual process (see GRANT et al., 1997) absent; (10) extremities of

fingers pointed, of toes slightly rounded; (11) finger IT longer than finger [; (12) no distal

subarticular tubercles on fingers III and IV; (13) inner palmar tubercle very small, rounded,

on base of metacarpus; (14) outer palmar tubercle very small, rounded, similar and of same

size as inner; (15) legs strong, heels far apart when hind legs are placed at right angle with

body; (16) tarsal fold present, well developed: (17) inner metatarsal tubercle flat, elongate; (18)

outer metatarsal tubercle absent; (19) tarsal tubercle absent; (20) femoral glands absent; (21)

lateral-line system present in adult; (22) dorsal parts covered with a network of ridges: (23)

fejervaryan line (see Dusois & OuLER, 2000, and DuBois et al., 2001) absent; (24) rear part of

thighs marbled, without longitudinal white and dark lines; (25) adult male with fangs and

internal vocal sacs, without nuptial pads:; (26) eggs pigmented; (27) tadpole with ventral

mouthparts, keratodont formula 2/3.

Comparisons. - Detailed comparisons of this genus with six other genera of Asian Ranidae

with forked omosternum are provided in tab. 1. Of particular relevance are the comparisons

with three of them, which in several characters rather closely resemble the new genus.

Lankanectes shares several characters with the Asian ranid genera Euphlvctis and Occidozyga

Kuhl & Van Hasselt, 1822; in particular, in these three genera a lateral line system is present on

the body of adults, a rare character in the Ranidae (see below). It is distinguished from these

two genera by a combination of characters (see tab. 1), among which the following ones in

particular may be highlighted: (1) internarial distance shorter than distance between upper

eyelids (instead of subequal or longer); (2) loreal region slightly convex (instead of slightly

concave); (3) network of numerous transverse folds on the whole of back and head (absent in

the other two genera); (4) inner metatarsal tubercle flat (instead of digit-like);

(5) rear part of thighs marbled (instead of showing a longitudinal white line underlined by a

dark line); (6) adult male with fangs on the anterior lower jaw. This last character is shared by

the new genus and some species of the genus Limnonectes, but Lankanectes differs from the

latter in several other characters (see tab. 1), including: (1) internarial distance shorter than

distance between upper eyelids (instead of subequal or longer); (2) loreal region slightly

convex (instead of concave); (3) network of numerous transverse folds on the whole of back

and head (absent in Limnonectes); (4) upper eyelids covered with numerous round warts

(instead of bearing only a few round warts in their rear part); (5) finger II longer than finger

I (instead of shorter or subequal); (6) lateral-line system present in adult (instead of absent).

Generic content and distribution. — For the time being, a single species, Lankanectes

(Peters, 1863), an endemic of the island of Sri Lankan, is known in this genus. However, as

mentioned above, this species might prove later to be heterogeneous and to consist in fact of

Source : MNHN, Paris

Table 1. - Some diagnostic morphological characters of seven Asian genera of the family Ranidae with omosternum forked at base. See DUBOIS (1995) for the

definition of the tadpole’s condensed collective keratodont formula (CCKF), ie. minimum-maximum numbers of keratodont rows on upper/lower lips of

tadpoles observed in the taxon. See OHLER & DUBOIS (1999) for the definition of categories of digital disks in the Ranidae.

sottolh Dicroglossinae Dicroglossinae Dicroglossinae Occidozyginae Occidozyginae Nyctibatrachinae Lankanectinae

REY Anderson, 1871 Anderson, 1871 Anderson, 1871 Fei, Ye & Huang, Fei, Ye & Huang, | Blommers-Schlüsser, subfam. nov.

1991 1991 1993

Tribe Dicroglossini Limnonectini Limnonectini L 3 s, 5

Anderson, 1871 Dubois, 1992 Dubois, 1992

& Euphlyets Limnonectes Taylorana Occidozyga Phrynoglossus Myctibatrachus Lankanectes

pus Fitzinger, 1843 Fitzinger, 1843 Dubois, 1987 Kuhl & Ni Peters, 1867 Boulenger, 1882 gen. nov.

Rana leschenaultit Rana kuhli Polypedates hascheanus Rana lima Phrnoglossus martensit | Nyctibatrachus major Rana corrugata

(FrrznGeR, 1843:31)_ | (France, 1843:31) | (DUBOIs, 19872: 63) |" STEGER (1925: 33) 29) MYERS (1942: 54) >

“Aduït male SVL (mm) 40-95 35-150 25-39 19-26 18-30 1346 33-65 (=)

Adult female SVL (mm) 45-130 35413 2437 2635 2245 1447 459 ee

eyelids =

round wars wars in their rear part | wars in tir rear part round wars indistinetround wars | with numerous round round wars S

wars È

Canthus rostralis Indistinet Distinet or te distinct Lite distinet Indistinet Indistinet Indistnct or te disinet Indistinet Ÿ

Loreal region lights concave Concave Fat Stighily concave Stighty convex Stighcly convex Stighty convex

lower jaw

Tympanum Distinet Distinet or indistinet Distinet Indistinct Indistinet Indistinet or te distinct Indistinet

sometimes dilted as | rudimentary dorso- slighty enlarged terminal fods pointed, or oes slightly

small isks bearing a terminal old rounded

Relative length of fingers

Land 1

Distl subarticular

tubercles on fingers LI

andiV

Inner palmar tubercle

Outer palmar tubercle

Finger I longer than

Singer!

Small

Medium, oval, on the

base of metacarpus

Indistinet

dorso-terminal fold

Finger Il shortr than or

25 long as finger 1

Indistinet

Medium or large, on base

‘of metacarpus or on the

whole of it

Elongate, half smaller

han inner palmar

tubercle or of same size

Finger IL shortr than

finger 1

Small

Medium, oval, on halfof

metacarpus

Oval, a little smaller than

Fingers Land 11 subequal

Absent

Small, rounded,

prominent, on base of

metacarpus

Small, rounded,

prominent, of same size

5 inner palmar tubercle

Finger IL shorter than

finger

Absent

Small, oval, on base of

metacarpus

Small, oval, of same size

5 inner palmar tubercle

Finger I longer than

finger |

distinct

Small, val, rather

prominent, on base of

metacarpus

Oval, about half-size of

inner palmar tubercle

Finger I longer than

finger L

Absent

Very small, rounded, on

base of metacarpus

Very small, rounded, of

ame size as inner palmar

tubercle

Source : MNHN, Paris

Table 1. (continued)

at right angle with body

ë Euphlyctis Limnonectes Taylorana Occidozyga Phrmoglossus Nyctibatrachus Lankanectes

NUE Fitzinger, 1843 Fitzinger, 1843 Dubois, 1987 Kuhl & an Peters, 1867 Boulenger, 1882 gen. nov.

Hind legs Rather strong and short | Strong or narrow, short or | Moderately strong, rather | Rather strong, short Rather Strong, Short Strong, shot Very strong, short

long. Short

Distance berween hecls Heels far apart Heels in contact or Heels in contact Heels far apart Heels far apart Hecls far apart Heels far apart

when hind legs are placed overlapping

tem feurdes À Pésorndn | -Praisacitauioité indisinct Present, modenue | Present, well developes | Present well eveloped | Present well developed

Promise

Outer metaarsal uberce Absent Absent Absent Present Absent Absent Avsent

Tarsal ubercle pro Absent Absent Present Absent Absent Avant

Femoral glands Absent Absent Absent Absent Absent Present Absent

Laertie sysam in Present Absent Abtent Present Abeent Absent Present

suit

Longitudinal dorsal Absent Present or absent Present Absent Absent Absent Absent (w]

glandular folds =

Nero fige on Assent Absent Absent Absent Absent Absent Present a

back and head 8

Colortion of enr partof | Longidina vite ne Maled Maeé Login | natucrpos Maried Male &

ighs atinéd y ar ns sms y kite

Sex size dimorphism Males smaller than | _ Absent or males larger Absent Males smaller than Males smaller than Absent Absent ©

feras han Femaes feras ras È

Enlrgement of bed in Absent Present or absent Present Absent A Absent Absent E

él male

Fa aduit raie Avsent Present or ae Present, smal Absent Absent Absent Present

D die aveu | init vid Elo Fos ontroat Fos on toat Fos on ont Fois on oat

hs on boat “out

Male advertisement cal Present Absent or present Present Present Present Present Present

up pas in adult Abent Absent Abéent Present Present Present absent

Lens

At paré Axis ; : Anitary Labs : :

Egg coloration Pigmented Pigmented Unpigmented Pigmented Unpigmented Pigmented or not Pigmented

Moëe of development Tadpole Tadpoie or endowoph Endoroph Tadpole Tadpole Tadpole Tadpole

Parenal are Absent Absentor present Present Absent Absent Absent Absent

Tadpoles CCKF 12 EE k 00 o0 o0 23

References BOULENGER, 1920; BOULENGER, 1920; BOULENGER, 1920; BOULENGER, 1890; DECKERT, 1938; INGER, | CLARKE, 1983; INGER et BOULENGER, 1920;

ioréb 1991, OMLER ea, 1999 iooi KRSNGNE, 195

Durta&

MANAMENDRA-

ARacIOM, 1996 æ

Source : MNHN, Paris

88 ALYTES 19 (2-4)

two species or subspecies. In order to facilitate further works in this respect, we provide below

a detailed redescription of the lectotype, designated herein, of Rana corrugata Peters, 1863.

Vernacular name. — We propose to use the name “lankanects” as vernacular name for these

frogs, and “limnonects” for frogs of the genus Limnonectes.

Etymology of the generic nomen. — The new generic nomen, of masculine grammatical gender,

is derived from the frog generic nomen Limnonectes Fitzinger, 1843, and from the name of the

island of Sri Lanka. It suggests that these frogs are limnonect-like frogs endemic of this island.

LECTOTYPE DESCRIPTION

Lectotype, by present designation, of Rana corrugata Peters, 1863: ZMB 4897, adult

male (fig. 1), collected by J. Nietner in “Rambodde” (Ramboda; 07°03'N, 80°14’E; 1310 m)

(Durra & MANAMENDRA-ARACHCHI, 1996: 12), Sri Lanka.

(A) Size and general aspect. — (1) Specimen of moderate size (SVL 44.0 mm), body stout.

(B) Head. — (2) Head rather large, wider (HW 17.2 mm) than long (HL 16.8 mm;

MN15.3 mm; MFE 12.8 mm; MBE 8.3 mm), convex. (3) Snout rounded, slightly protruding;

its length (SL 6.03 mm) longer than horizontal diameter of eye (EL 5.25 mm). (4) Canthus

rostralis indistinct, loreal region convex: angle of loreal region with upper face of head flared.

(5) Interorbital space flat, broader (TUE 4.02 mm) than upper eyelid (UEW 2.01 mm) and

than internarial distance (IN 2.46 mm); distance between front of eyes (IFE 6.68 mm) about

half of distance between back of eyes (IBE 12.76 mm). (6) Nostrils oval, with small flap of

skin laterally; closer to eye (EN 2.66 mm) than to tip of snout (NS 3.37 mm). (7) Pupil not

observable. (8) Tympanum indistinct (TYD nm, TYE nm). (9) Pineal ocellus absent.

(10) Maxillary teeth present; vomerine ridge present, bearing 2 small teeth, posterior to

choanae, with an angle of 40° relative to body axis, closer to each other than to choanae,

longer than distance between them. (11) Tongue chordate, deeply emarginate, without lingual

process, covered by numerous small papillae. (12) A dermal, non glandular supratympanic

fold, distinct, from eye to shoulder. (13) Parotoid glands absent. (14) Cephalic ridges absent.

(15) Co-ossified skin absent.

8.8 mm), not enlarged. (17) Fingers short and rather strong (TFL 4.21 mm). (18) Relative

length of fingers, shortest to longest: I < IV < IT < IL. (19) Tips of fingers pointed, bearing

small, rounded terminal notch, not enlarged. (20) Fingers without dermal fringe and web-

bing. (21) Subarticular tubercles prominent, conical, single; distal tubercle of finger IT and IV

absent. (22) Prepollex small (size of subarticular tubercles), rounded, distinct; a single, small,

round inner palmar tubercle on the base of metacarpus; outer palmar tubercle similar and of

same size as inner; supernumerary tubercles absent.

(D) Hindlimbs. — (23) Shank two times longer (TL 18.7 mm) than wide (TW 10.1 mm),

shorter than thigh (FL 19.7 mm) and than distance from base of internal metatarsal tubercle

to tip of toe IV (FOL 19.8 mm). (24) Toes short, rather thin, toe IV (FTL 10.8) longer than

third of distance from base of tarsus to tip of toe IV (TFOL 28.5 mm). (25) Relative length of

Source : MNHN, Paris

DuBois & OHLER 89

Fig. 1. - Lectotype of Rana corrugata Peters, 1863, ZMB 4897, adult male (SVL 44.0 mm). (a) Dorsal

view. (b) Ventral view.

toes, shortest to longest: I < II < V < III < IV. (26) Tips of toes pointed, bearing enlarged knob

terminally. (27) Webbing complete: I 0 —0 II 0 —0 II 0-0 IV 0-0 V(WTF 7.24 mm; WFF

6.32 mm; WI 6.58 mm; WII 5.00 mm; MTTF 13.68 mm; MTFF 14.47 mm; TFTF 5.66 mm;

FFTF 8.42 mm). (28) Dermal fringe along toe V absent. (29) Subarticular tubercles conical,

all present. (30) Inner metatarsal tubercle elongate, very prominent, shovel-shaped, its length

(MT 3.18 mm) 2 times in length of toe I (ITL 6.35 mm). (31) Tarsal fold present, from inner

metatarsal tubercle to before tibio-tarsal articulation. (32) Outer metatarsal tubercle, super-

numerary tubercles and tarsal tubercle absent.

(E) Skin. — (33) Dorsal and lateral parts of head and body: smooth skin forming

numerous regularly arranged folds, transversally arranged on back, longitudinally arranged

between eyes; between these folds presence of round indistinet warts; laterally on posterior

part of back, 4-5 symmetrically arranged pairs of prominent, medium sized warts; flanks

smooth. (34) Latero-dorsal folds absent. (35) Dorsal parts of limbs: forelimb with transversal

foldings; thigh with flat warts; leg and tarsus with glandular warts bearing horny spinules.

(36) Ventral parts of head, body and limbs: throat with longitudinal foldings; chest, belly

and thigh smooth; fejervaryan line absent; lateral-line system indistinct. (37) Macroglands

absent.

Source : MNHN, Paris

90 ALYTES 19 (2-4)

(F) Coloration in alcool. - (38) Dorsal and lateral parts of head and body: dorsal parts

of head and dorsum and upper part of flank dark brown with top of folds whitish (discolor-

ation); a blackish brown band between eyes. (39) Dorsal parts of limbs: dorsal part of

forelimb, of thigh, of shank and of foot dark brown with indistinct darker brown bands:

posterior part of thigh brown with blackish triangle around vent. (40) Ventral parts of head,

body and limbs: throat and margin of throat dark brown; chest and belly whitish with some

dark brown spots; thigh whitish; webbing whitish with dark brown marblings.

(G) Male secondary sexual characters. - (41) Nuptial spines absent. (42) Vocal sacs

present, indistinct on throat; distinct, rounded, paired openings, posteriorly on mouth floor.

(43) Other male secondary sexual characters: toothlike projections (fangs) at the front of

lower jaw.

DISCUSSION

MORPHOLOGICAL AND MORPHOMETRIC ANALYSES

Morphological comparison between the genus Lankanectes and 6 other genera of Asian

Ranidae with omosternum forked at base is given in tab. 1. Some of the major differences

between the new genus and these genera were already mentioned in the diagnoses of the new

taxa given above, and are not repeated here.

Morphometric comparisons also support the distinctiveness of the new taxon. As we

have already stressed elsewhere (e.g., DuBois et al., 2001), in many anuran groups the general

“body shape” gives good clues regarding generic classification and allocation of species to

genera. Once again we confirm this statement in the present study. On the basis of 22

measurements (see Material and methods above), we compared Lankanectes corrugatus with

several species belonging to the four subgenera and species-groups currently recognized in the

genus Limnonectes. Besides, we also thought useful to compare this species with members of

several other genera discussed above (Euphlyctis, Occidozyga and Phrynoglossus) and also

with the genus Nyctibatrachus, an endemic of southern India. The results are shown in tab. 2

and fig. 2. Lankanectes corrugatus appears as a well-distinguished group, as much as the other

genera considered here. This result is confirmed by the canonical discriminant analysis based

on 19 measurements and involving Lankanectes and the four subgroups (subgenera or

species-groups) currently recognized in the genus Limnonectes (tab. 3, fig. 3).

Oneway analysis using the Scheffe test shows significant differences in various characters

between Lankanectes Specimens and specimens of the 8 other taxa studied. The new genus can

be distinguished from all 4 subgroups of Limnonectes studied by a shorter head (HL), shorter

eye-nostril distance (EN) and shorter shank (TL). Members of the subgenus Elachyglossa

also have larger head (HW), greater internarial distance (IN) and more developed webbing

(TETE). The frogs of the grunniens species-group are significantly larger (SVL) than Lanka-

nectes specimens and show differences in eye position (MFE, IBE). As compared to the kuhlii

species-group, the new genus has significantly smaller (HW) and shorter head (beside HL,

MN is significantly shorter), shorter forearm (FLL) and less developed webbing (MTFF).

Source : MNHN, Paris

Dugois & OHLER 91

2 ———— 2

TAXON

1 F2 | x L. gr grunniens

eve # L.gr microdisous

L #4 | 5 Nyctbatrachus

« + Lankanectes

4 L < Euphiyctis

+ Occidezyga

2. ÿ 4 Phrynoglossus

a o Û

5 5 CR à Elachyglossa

3 G ER

È DS De, ALES © L. gr kuhl

3 2 ñ ô 2

Factor 1

Fig. 2. - Plots of multivariate analysis (first three axes) based on 22 measurements for the following nine

genera, subgenera and species-groups of Asian Ranidae: Euphlyctis, Lankanectes, Limnonectes

(Elachyglossa), Limnonectes (Limnonectes) gr. grunniens, Limnonectes (Limnonectes) gr. kuhlii,

Limnonectes ( Limnonectes) gr. microdiseus, Nyctibatrachus, Occidozyga and Phrynoglossus.

20 ——

‘

10 .

TAXON

D Group Cents

* L.gr grumniens

LA = L. gr microdisous

x | Lantanectes

É © Elsengossa

É | L Lou

2 0 ô o] ET

Function 1

Fig. 3. - Plots of discriminant function scores using minimization of Wilk’s lambda of morphometric

log-transposed characters (19 measurements) for the following five genera, subgenera and species-

groups of Asian Ranidae: Lankanectes, Linmonectes (Elachyglossa), Limnonectes ( Limnonectes )

gr. grunniens, Limnonectes ( Limmonectes) gr. kuhlii and Limnonectes { Limnonectes) gr. microdis-

eus.

Source : MNHN, Paris

92 ALYTES 19 (2-4)

Table 2. - Results of principal component analysis based on 22 In-transposed measurements

including specimens referred to the genera Euphlyctis, Lankanectes, Limnonectes,

Nyctibatrachus, Occidozyga, and Phrynoglossus.

Initial Eigenvalues Rotation Sums of Squared Loadings

“ici 9 ï of C lative

1 19.991 90.870 90.870 10.520 47.819 47.819

2 1.148 5.219 96.089 7.154 32.518 80.338

3 0.260 1.181 97.270 3.725 16.932 97.270

Components for rotated Components for rotated

Variable component matrix Variable component matrix

ll 2 3 1 2 é

SVL 0.670 0.635 0.374 FOL 0.630 0.623 0.427

HW 0.719 0.599 0.333 IN 0.951 0.230 0.132

HL 0.711 0.608 0.341 EN 0.699 0.571 0.397

MN 0.697 0.593 0.383 EL 0.702 0.605 0.460

MFE 0.698 0.583 0.388 TFL 0.564 0.720 0.366

MBE 0.605 0.596 0.468 MTTF 0.475 0.692 0.525

IFE 0.838 0.451 0.286 TFTF 0.814 0.497 0.334

IBE 0.831 0.504 0.165 MTFF 0.452 0.713 0.520

FLL 0.743 0.569 0.334 FFTF 0.852 0.405 0.208

HAL 0.589 0.694 0.392 IMT 0.700 0.483 0.460

TL 0.706 0.590 0.374 ITL 0.108 0.283 0.949

Specimens of the microdiscus species-group have a larger distance between eyes (IFE) and

nostrils (IN) and their webbing is more incurved (TFTF). As to the differences existing to the

other genera studied here, Lankanectes is larger (SVL) than Nyctibatrachus, its notrils are

more distantly separated; the inner metatarsal tubercle is smaller (IMT) in Nyctibatrach

is the webbing of the feet (MTTE, MTFE, TFTF). Phrynoglossus can be distinguished from

the new genus by its smaller body size (SVL), its larger internarial distance (IN), its larger

inner metatarsal tubercle (IMT) and its smaller webbing (MTFF, TFTF). Members of the

genus Occidoz yga show smaller distance between the eyes (IBE), longer hand length (HAL),

a shorter inner metatarsal tubercle (IMT) and a longer inner toe (ITL). A smaller distance

between the eyes and a smaller inner metatarsal tubercle separates Lankanectes from the

members of the genus Euphlvctis.

, as

Source : MNHN, Paris

Duois & OHLER

Table 3. — Results of principal component analysis based on varimax rotated cocfficients from log-

transposed characters (25 measurements) for specimens referred to the gencra Euphlyctis,

Fejervarya, Hoplobatrachus, Limnonectes, Minervarya and Sphaerotheca.

Initial Eigenvalues

Component Total % of Variance Cumulative %

1 22.639 90.558 90.558

2 0.799 3.196 93.754

2.783 96.537

Rotation Sums of Squared Loadings

Component Total % of Variance Cumulative %

1 10.152 40,610 40.610

2 9.597 38.390 78.999

3 4384 17.538 96.537

Components for rotated component matrix

Variable Ll F4 S

SVL 0.649 0.621 0.422

HW 0.617 0.706 0.337

HL 0.673 0.647 0.340

MN 0.667 0.646 0.330

MFE 0.649 0.674 0.322

MBE 0.639 0.683 0312

IFE 0.505 0.768 0.371

IBE 0.553 0.757 0.368

FLL 0.589 0.682 0.419

HAL 0.661 0.653 0.346

TL 0.732 0.530 0410

FOL 0.709 0.534 0.456

IN 0.235 0.817 0.471

EN 0.698 0.592 0.353

EL 0.599 0.691 0.351

TYD 0.712 0.487 0.367

TYE 0.449 0.773 0.223

TFL 0.654 0.635 0.364

FTL 0.757 0.415 0.484

0.836 0453 0.299

0.349 0.349 0.859

MTFF 0.830 0463 0.297

FFTF 0.432 0.402 0.788

IMT 0.419 0.797 0.293

ITL 0.873 0.304 0.283

Source : MNHN, Paris

94 ALYTES 19 (2-4)

Table 4. - Some measurements and ratios of four specimens of Lankanectes corrugatus, including

the lectotype (ZMB 4897) and the two paralectotypes (ZMB 62771-62772) from Rambodde

(Sri Lanka), and a fourth specimen (MNHN 2000.0616) from Kandy (Sri Lanka). SVL is

given in mm; all other measurements are given as per thousands of SVL. Sex and stages: À,

adult; J, juvenile; F, female; M, male.

Collection ZMB ZMB ZMB MNHN

number 4897 62771 62772 2000.0616

Locality Rambodde Rambodde Rambodde Kandy

Sex and stage AM JF AM AF

SVL 44.0 372 33.5 44.4

HW 391 363 337 338

HL 382 379 379 354

MN 348 333 333 302

MFE 291 280 280 243

MBE 189 177 177 164

IFE 152 153 153 158

IBE 290 298 298 264

IN 56 70 66 66

EN 60 70 66 70

EL 119 138 106 108

FLL 195 210 185 191

HAL 200 199 224 218

TFL 96 127 120 115

TL 425 414 394 405

FOL 457 465 421 462

FTL 245 242 242 248

IMT 72 63 79 77

ITL 144 148 132 150

MTTF 311 328 310 296

MTFF 329 336 310 329

TFTF 129 124 141 139

FFTF 191 177 189 184

DISTRIBUTION OF SOME CHARACTERS AMONG SEVERAL GENERA OF RANIDAE

Characters related with an aquatic mode of life

All anuran tadpoles show a lateral-line system on body and head, similar to that of fishes,

i.e. composed of rows of small pores opening on sense cells or neuromasts that are sensible to

vibrations of low frequency in water (NOBLE, 1931: 318-321; DUELLMAN & TRUEB, 1985:

Source : MNHN, Paris

Dugois & OHLER 95

378-379). Most anuran species lose this system at metamorphosis, but it remains present in

adults of a few anuran groups that have a mainly aquatic mode of life. This retention of a

larval character in otherwise adult specimens is a case of partial paedomorphism (DuBoIs,

1987a). This is observed in several aquatic genera of anurans, distributed in various families,

including the Discoglossidae (Barbourula Taylor & Noble, 1924; Bombina Oken, 1816), the

Leptodactylidae (Lepidobatrachus Budgett, 1899) and the Pipidae (all genera) (fig. 4). In the

Ranidae, which include various aquatic groups, some of them show the paedomorphic

retention of lateral-line systems in adults, while others, which may seemingly appear as

aquatic as the former ones, do not show this phenomenon. Three genera of Ranidae are

known to retain the lateral-line system in adults: Euphlyctis (see e.g.: BOULENGER, 1920;

Dusois, 1987b, 1992), Occidozyga Kuhl & Van Hasselt, 1822 (see e.g. DuBois, 1987b, 1992)

and the new genus Lankanectes (fig. 4). To the best of our knowledge, the presence of a

lateral-line system in adults of L. corrugatus has never been mentioned in the scientific

literature, although these lines are quite obvious in live specimens (AD, personal observa-

tions) and usually remain visible, although not so easily, in fixed specimens.

Dusois (1987b) had considered the presence of a lateral-line system in adults as a

synapomorphy of Euphlyctis and Occidozyga, which had led him to regard these two taxa as

sister-groups and to treat them as subgenera of a single genus. Other characters which had

supported this interpretation were the general body shape (0. lima looking almost exactly as

a miniaturized £. cyanophlyctis), the shapes of the foot and of the extremities of digits, and the

presence of continuous longitudinal white and dark stripes all along the rear part of the thighs

(fig. 5). However, molecular cladistic data provided by MARMAYoOU et al. (2000), Kosucx et al.

(2001) and DELORME et al. (submitted) strongly suggest that Occidozyga and Euphlyctis are

not sister-groups, and that all or most of the characters listed above are convergences related

to aquatic life. As a matter of fact, as mentioned above the lateral-line system is retained in

adults of several aquatic frogs of other families and this is the case also of pointed digits and

of fully webbed feet with a relatively short fourth toe.

As concerns the last character of the list above, the presence of longitudinal white and

dark stripes on the posterior thigh is also observed in aquatic South-American hylid frogs of

the genus Pseudis Wagler, 1830 (fig. 5) and, although less strikingly, in Chinese Ranidae that

are also largely aquatic, i.e. Rana (Pelophylax) plancyi Lataste, 1880 and Rana ( Pelophylax)

hubeiensis Fei & Ye, 1982 (see e.g.: PoPe, 1931: 511; Ft, 1999: 161). The meaning of this

coloration character is not quite clear, but the fact that it appeared independently in several

unrelated anuran groups having a largely aquatic life suggests that it also has an adaptive

value for frogs with such a mode of life, probably as a camouflage device towards aquatic or

aerial potential predators. In terrestrial frogs that live in grassland habitats, a striped dorsal

pattern is often observed. These frogs have longitudinal lines either all along the middle of the

back (vertebral stripe or band, present in many groups of frogs), or as several subparallel dark

stripes on a brown dorsum. The latter, although perhaps less common, is also a rather widely

distributed phenotype in frogs, observed e.g. in the Hyperoliidae (e.g.. some Afrixalus

Laurent, 1944 or Hyperolius Rapp, 1842), in the Ranidae (e.g., some Prychadena Boulenger,

1917 or Strongylopus Tschudi, 1838), or in the Rhacophorinae/dae (e.g., some Chirixalus

Boulenger, 1893 or Polypedates Tschudi, 1838). Such patterns can clearly contribute to a

camouflage among herbs or elongated leaves. However in such frogs the rear parts of the

thighs do not show longitudinal stripes. In terrestrial frogs the legs are not kept extended at

Source : MNHN, Paris

96 ALYTES 19 (2-4)

g. 4. Lateral-line system (1) in several examples of aquatic anurans: (a) Sélurana tropicalis Gray, 1864

(Pipidae, Siluraninae), MNHN 1994.1915, adult male, Guinea: (b) Occidozpga lima (Gravenhorst,

1829) (Ranidue, Occidozyginae), MNHN 19996418, adult female, Yunnan: (c) Lankanectes corru-

gatus (Peters, 1863), MNHN 2000.0616, adult female, Sri Lanka.

Source : MNHN, Paris

Dusois & OHLER 97

à) Pseudis el ous 175$) (Hylidue, Pseudinae) MNHN

1983.0390, juve É zuela: lines present; (b) Occidozyga lima (Gravenhorst, 1829)

(Ranidae, Occidozyginae), MNHN 1999.6418, adult female, Yunnan: lines present: (c) Lankanectes

corrugatus (Peters, 1863) (Ranidae, Lankanectinae), MNHN 2000.0616, adult female, Sri Lanka:

lines absent.

Source : MNHN, Paris

98 ALYTES 19 (2-4)

rest, so that the coloration of the back of thighs is not exposed: it is shown only during

movements. In aquatic frogs, the situation may be different. These frogs, like Euphlyctis or

Occidozyga, often remain suspended floating in water for some minutes or more, using the

four limbs extended in the prolongation of the body or feebly bended laterally as balancers. In

such a position the posterior surface of hindlegs is visible. If such a frog is then hidden within

long and narrow aquatic vegetal structures, the longitudinal lines at the back of thighs might

contribute to the camouflage, especially if it follows some other linear structures or coloration

on the flank or dorsum of the frog, as well exemplified in the figure 16 of Pope (1931: 511).

Although the new genus Lankanectes shares with Occidozyga and Euphlyctis the retention of

lateral-line system in adults, it does not show the longitudinal stripes on the rear parts of the

thighs (fig. 5) and this is a significant difference between the two genera (as well as between the

closely related Occidozyga and Phrynoglossus). Perhaps this is connected to the fact that L.

corrugatus usually inhabits shallow, mud-substrate (as opposed to gravel- or rock-substrate)

streams, poor in vegetation (Pethiyagoda, personal communication).

Male secondary sex characters

Anurans display a large diversity of male secondary sex characters, including various

kinds of spines, asperities and glands, vocal sacs and adult morphometric differences. The

taxonomic significance of such dimorphic characters has no generality over the whole of

anurans. In several groups, male secondary sex characters are diagnostic of species-groups,

subgenera or genera, or even of higher taxa: this is e.g. the case of the pectoral plates of the

megophryid tribe Oreolalagini (see DELORME & Dugois, 2001). In some other cases however,

differences in such characters are species-specific and can even separate very similar and

closely related species: examples include the presence/absence of nuptial spines in Paa liebigii

(Günther, 1860) and Paa vicina (Stoliczka, 1872) (Dugois, 1976a, 1980) and the

presence/absence of vocal sacs in Polypedates leucomystax (Gravenhorst, 1829) and Polype-

dates mutus (Smith, 1940) (Smrrx, 1940; Liu & Hu, 1961).

The major son, besides general morphological resemblance, that apparently led

BOULENGER (1920) to include Rana corrugata in the same group as Rana kuhlii seems to have

been the presence in both species of “fangs” at the front of the lower jaw (fig. 6). This

character was also used by EMERSON & WaRD (1998) and EMERSON et al. (2000) as the basis for

the vernacular name of “fanged frogs” which they gave to the genus Limnonectes. However,

not all frogs of this group possess fangs (see e.g.: BOULENGER, 1920; SMirH, 1922a-b; BOURRET,

1942), and this vernacular name does not appear more appropriate for these frogs than that of

“voiceless frogs”, the previous name used by the same authors (EMERSON & Voris, 1992;

EMERSON & BERRIGAN, 1993). As a matter of fact, even if several species of South-East Asian

frogs of this group are devoid of structurally differentiated vocal sacs, they are not voiceless,

as their males can emit advertisement calls, as was observed in Limnonectes blythii (Matsut,

1995), or at least loud territorial calls, as was observed in Limnonectes cf. kuhlii (AD & AO,

unpublished observations, see below). As for Lankanectes corrugatus, males show differenti-

ated vocal sacs and emit loud calls (AD, personal observations, Sri Lanka, June 1999) whose

function has to be clarified, given that they persist after the breeding season (Pethiyagoda,

personal communication), but which probably can have an advertisement function, possibly

combined with a territorial one.

Source : MNHN, Paris

Dunois & OHLER 99

Fig. 6.— Fangs (D at front of lower jaw in two Asian ranid groups: (a) Limnonectes { Limnonectes) cf.

kuhlit (Tschudi, 1838) (Ranidae, Dicroglossinae), MNHN 19380030, adult male, Vietnam: (b)

Lankanectes corrugatus (Peters, 1863) (Ranidae, Lankanectinae), ZMB 4897, adult male, lectotype,

Sri Lanka.

As mentioned above, fangs are present in adult males of some species of Limnonectes

only. These species also tend to have a much wider head in males than in females, and in some

of them (some of the members of the subgenus Elachyglossa) they also show a knob on the

dorsal back of head, starting between the eyes and extending beyond them (SMrrH, 1922a-b;

Source : MNHN, Paris

100 ALYTES 19 (2-4)

BoURRET, 1942). We regard all these characters (fangs, wide head, dorsal cephalic knob) as

related to agonistic behaviour between males, like in several other cases of spines and other

differentiated structures on the heads of males (see e. HINE, 1979; DuBois & OHLER, 1998).

In the Siriphum agricultural station of the Doi Inthanon in northern Thailand, in the night of

24 September 1986 we had the opportunity to observe an adult male of Limnonectes cf. kuhlii

that, disturbed by our foraging in water with a small net to collect tadpoles, suddenly emitted

a loud and deep territorial call, “koaa, koaa”, and repeated it several times. While emitting this

guttural whistling, this frog had its body immersed in water but its head was raised above the

water level, and quite voluntary so as the frog was leaning on a rock and the fore part of its

body was raised on its stretched arms. Seen from the level of the surface of water, this large

head evoked a much larger frog than the actual size of this male (MNHN 1987.3197; SVL 63.7

mm; HW 28.6 mm, HL 29.6 mm). On several occasions, in Thailand, Laos and Vietnam, we

had the opportunity to observe and collect very large-headed males of Limnonectes cf. kuhlii.

However, on every occasion we were struck by the fact that, in a given station (e.g., a small

pond, or a portion of several meters along a small stream), we never found more than one such

large-headed male, although other males may have been seen there: all other males collected

along with the latter had a “normal” or only slightly enlarged head, although some of them

were of a size similar to that of the large-headed male of the station. We suggest a possible

interpretation for these observations: in each station, a single male might occupy the hierar-

chical position of a dominant male. This male would develop a very enlarged head but its

presence, and most likely also its behaviour (with territorial calls and possibly also fighting

with other males) would inhibit the development of enlarged head in all other males nearby.

The existence of such an inhibition in dominated males, which could likely be implemented

through a hormonal mechanism, could rather easily be submitted to experimental testing,

and this could be done by scientists living in countries where these frogs occur.

These observations suggest that, unless large series of specimens are available for study,

it is impossible to be sure of the “maximum” development of male sex characters (including

the length of the fangs, the width of the head or the size of the cephalic knob) in any species

or population of Limnonectes. It is therefore advisable to look for other characters to

distinguish species, because, when only the development of male sex characters is available in

this respect, these characters may be misleading, being in part due to the studied males

occupying a dominant or dominated position in the hierarchy of the group. This remark holds

particularly true for frogs of the subgenus Elachyglossa, that show a very variable develop-

ment of the cephalic knob (Smrrn, 1922a-b; BOURRET, 1942). However, despite these remarks,

it should be stressed that, even in those males that do not show a “much enlarged head”, the

head is significantly proportionally wider in males than in females (OnLER & DuBois, 1999).

As concerns the species Lankanectes corrugatus, very few museum specimens are availa-

ble for study outside Sri Lanka, and little is known on its variation, including sexual

dimorphism in size and other measurements. No detailed measurements of this species were

provided in the two books dedicated to the Sri Lankan frogs by local zoologists (KIRTISINGHE,

1957; DUTrA & MANAMENDRA-ARACHCHI, 1996). BOULENGER (1920) provided meas-

urements for 4 specimens in the London museum, including 3 adult males and 1 female. We

provide in tab. 1 our measurements of 4 other specimens, 2 males and 2 females, in the Berlin

and Paris museums, including the 3 original syntypes of the species. According to this very

limited material, no sex dimorphism appears to exist in this species for either the total size or

Source : MNHN, Paris

Dugois & OHLER 101

the size of head, but this sample is much too small to permit definitive statements in this

respect. However, given the limited information currently available, this species would appear

to differ from Limnonectes in not exhibiting sex dimorphism in the size of head.

While the male secondary sex characters discussed above are exceptional in anurans, two

other male characters are very widely distributed in many anuran groups, i.e. the presence of

nuptial pads (usually covered with a layer of minute spines) on the first finger (and sometimes

also on the prepollex, the second and the third fingers) and the presence of vocal sacs with

openings on the sides of the mouth floor. The first of these characters is absent both in

Lankanectes and in all frogs of the tribe Limnonectini of the Dicroglossinae, i.e. the genera

Limnonectes and Taylorana. As for the second character, as mentioned above, so-called

“voiceless” frogs of the genus Limnonectes can emit loud territorial calls whose function is

probably to keep other males at distance. Some at least of them are known to emit also

advertisement calls, i.e. calls whose function is to attract females during breeding. Some

members of this genus, as currently understood, do have internal vocal sacs, while others lack

them. Lankanectes corrugatus produces dull advertisement calls that are very striking for

anyone who meets these frogs in the field and that can be heard from several meters in the

forest habitat (AD, personal observations); they are evoked by DUTTA & MANAMENDRA-

ARACHCHI (1996: 82) as “Urrm...”. The presence of vocal sacs in these frogs, which had been

ignored by GÜNTHER (1864) and BOULENGER (1890, 1920), was noted by KIRTISINGHE (1957)

and DUTTA & MANAMENDRA-ARACHCHI (1996).

Other characters

Three final characters related to reproduction may briefly be mentioned here. The first

one is parental care, listed by EMERSON et al. (2000) in their “definition” of the genus

Limnonectes. However, according to currently published observations (ALCALA, 1962; INGER,

1966, 1985; INGER et al., 1986; INGER & Voris, 1988; EMERSON, 1996; INGER & STUEBING,

1997; BROWN & ISKANDAR, 2000), parental care is only known in some species of the

Limnonectes microdiscus group of the nominative subgenus Limnonectes (sensu DuBois,

1987b) and cannot be included among the characters diagnostic of the whole genus Limno-

nectes, at least as currently understood. According to ISKANDAR (in EMERSON, 1996; see

Dugois, 1999), a species of this group shows endotrophic development of embryos within the

genital tract of the female. In another group of Limnonectini, the genus Taylorana, direct

development occurs in eggs laid in terrestrial nests (TAYLOR, 1962; OnLer et al., 1999). AIT

these observations confirm the tendency that exists in this group for correlative increase of the

size of eggs with reduction of their numbers, leading eventually to direct development or

ovoviviparity, a tendency already identified by DuBois (1975).

A second interesting character is the position of the arms of the male during amplexus.

Although this has never been mentioned in the literature, we observed on various occasions

(AD & AO, unpublished observations, briefly mentioned in MARMAYOU et al., 2000: 295) that

in the species Phrynoglossus martensii Peters, 1867, type-species of Phrynoglossus Peters, 1867,

amplexus is lumbar, not axillary. This is a Strong reason, added to the morphological ones

(Smiru, 1931; TAYLOR, 1962; OuLer & DuBois, 1999) for considering Occidozyga as a genus

distinct from Phrynoglossus, and not as a synonym of the latter, as suggested by some authors

(INGER, 1954, 1966, 1996), or even as a subgenus of Rana (EMERSON & BERRIGAN, 1993).

Source : MNHN, Paris

102 ALYTES 19 (2-4)

Another peculiarity of Phrynoglossus is its unpigmented eggs, that most likely are deposited

under some shelter, but, to the best of our knowledge, reproduction and egg-laying has never

been described in this genus. The amplectic position of Lankanectes corrugatus has never been

observed so far (Pethiyagoda, personal communication). As for the eggs, in this species they

are pigmented, thus differing from those of Phrynoglossus.

CONCLUSION: TAXONOMIC ALLOCATION OF THE NEW GENUS

We presented above in tab. 1 a list of characters that we consider diagnostic of the genera

Euphlyctis, Lankanectes, Limnonectes, Nyctibatrachus, Occidozyga, Phrynoglossus and Tay-

lorana. Al these genera have in common the presence of a forked omosternum, that distin-

guishes them from the Raninae. However, the phylogenetic data recently provided by several

teams (BossuyT & MiLINKOvITCH, 2000; VENCES et al., 2000; DELORME et al., submitted)

suggest that these seven genera must be referred to several subclades within the Ranidae,

which we taxonomically treat as distinct provisional subfamilial taxa: (1) Euphlyctis, Limno-

nectes and Taylorana are members of the Dicroglossinae Anderson, 1871: (2) Occidozyga and

Phrynoglossus are members of the Occidozyginae Fei, Ye & Huang, 1991: and (3) Nycriba-

trachus is a member of the Nyctibatrachinae Blommers-Schlôsser, 1993. As for Lankanectes

corrugatus, the cladistic data available (BossUYT & MiLINKOVITCH, 2000; VENCES et al., 2000;

DELORME et al., submitted) suggest not only that it belongs in a genus distinct from Limno-

nectes, but also that it cannot be maintained in the subfamily Dicroglossinae. For the time

being, given the data of Bossuyr & MILINKOVITCH (2000) and VENCES et al. (2000), the closest

relatives of this genus would appear to be the subfamilies Raninae and Nyctibatrachinae, but

both groupsexhibit characters widely different from those of Lankanectes. From the Raninae,

Lankanectes differs readily by its forked omosternum and by a completely different general

habitus. As for the Nyctibatrachinae, except for the forked omosternum the new genus only

shares with Nyctibatrachus a few derived characters presumably related to the aquatic mode

of life of both genera (general body shape, short legs, short internarial distance), but both

genera show significant differences in a number of other characters (extremities of digits,

lateral-line system in adults, network of ridges on dorsal parts, femoral glands, fangs and

nuptial pads in males, tadpole keratodont formula), which do not support the inclusion of the

new genus in the Nyctibatrachinae. The only solution for the time being is to refer the new

genus to a new provisional suprageneric taxon, which, as well as all other such taxa, will have

to be tested by subsequent works (for more details, see DUBoIs, 1999).

ACKNOWLEDGEMENTS

For the help received in Thailand in 1986 and in Sri Lanka in 1999, AD is grateful to Jarujin

Nabhitabhata (Bangkok) and Paitoon Leksawasdi (Chiangmai), and to Kelum Manamendra-Arachchi

and Rohan Pethiyagoda (Colombo). For the loan of specimens, we thank Barry Clarke (London), Heinz

Grillitsch (Wien) and Rainer Günther (Berlin). For comments on a previous draft of the manuscript, we

are grateful to Franco Andreone (Torino) and Rohan Pethiyagoda (Colombo). This is publication N°47

of the PPF “Faune et flore du sud-est asiatique” (N°46, see DuBois et al., 2001).

Source : MNHN, Paris

Duois & OHLER 103

LITERATURE CITED

ANONYMOUS, 1999. — SPSS advanced models 9.0. Chicago, SPSS Inc. i-xv + 1-497.

ALCALA, À. C., 1962. - Breeding behavior and early development of frogs of Negros, Philippine islands.

Copeia, 1962: 679-726.

BoLKaY, S. J., 1915. — Beiträge zur Osteologie einiger exotischer Raniden. Anat. Anz., 48: 172-183.

BossuyT, F. & Dugois, A., 2001. — A review of the frog genus Philautus Gistel, 1848 (Amphibia, Anura,

Ranidae, Rhacophorinae). Zeylanica, 6 (1): 1-112.

Bossuvr, FE. & MiikovrrcH, M. C., 2000. - Convergent adaptive radiations in Madagascan and Asian

ranid frogs reveal covariation between larval and adult traits. Proc. nat. Acad. Sci. USA, 97 (12):

6585-6590.

BOULENGER, G. A. 1882. - Catalogue of the Batrachia Salientia s. Ecaudata in the collection of the British

Museum. London, Taylor & Francis: i-xvi + 1-503, pl. 1-30.

a 1890. - The fauna of the British India, including Ceylon and Burma. Reptilia and Batrachia. London,

Taylor & Francis: i-xviii + 1-541.

= 1920. - A monograph of the South Asian, Papuan, Melanesian, and Australian frogs of the genus

Rana. Rec. indian Mus., 20: 1-226.

BourRET, R., 1942. — Les Batraciens de l'Indochine. Hanoï, Institut océanographique de l'Indochine: i-x

+ 1-547, 4 pl.

BROWN, R. F. & IsKANDAR, D. T., 2000. - Nest site selection, larval hatching, and advertisement calls, of

Rana arathooni from Southwestern Sulawesi (Celebes) Island, Indonesia. J Herp., 34 (3): 404-413.

CLARKE, B. T, 1983. - A morphological re-examination of the frog genus Namophrys (Anura: Ranidae)

with comments on its biology, distribution and relationships. Zoo. J. Linn. Soc., 79: 377-398.

DECKERT, K., 1938. — Beiträge zur Osteologie und Systematik ranider Froschlurche. Sher. Ges. Naturf.

Freunde Berlin, 1938: 127-184.

DeLorE, M. & Duois, A., 2001. - Une nouvelle espèce de Scutiger du Bhutan, et quelques remarques

sur la classification subgénérique du genre Scutiger (Megophryidae, Leptobrachiinae). Aytes, 19

(2-4): 141-153.

DELORME, M, DuBois, A., KOsUCH, J. & VEN: M. submitted. - Molecular evidence for higher-level

endemism among South Asian amphibians: the relationships of Rana corrugata from Sri Lanka.

Dugois, A., 1974. - Liste commentée d’Amphibiens récoltés au Népal. Bull. Mus. natn. Hist. nat., (3), 213

(Zool.143): 341-411.

Dune 1975. - Un nouveau sous-genre (Paa) et trois nouvelles espèces du genre Rana. Remarques sur la

phylogénie des Ranidés (Amphibiens, Anoures). Bull. Mus. natn. Hist. nat., (3), 324 (Zool. 231):

1093-1115.

> 1976a. - Les Grenouilles du sous-genre Paa du Népal (famille Ranidae, genre Rand). Cahiers népalais

Documents, Paris, CNRS, 6: i-vi + 1-275.

se 1976b. - Deux Rana cyanophlyctis du Népal aux yeux noirs (Amphibiens, Anoures). Bull. Soc. linn.

Lyon, 45: 303-307.

LAMOTTE (ed.), Les problèmes de l'espèce dans le règne animal, 2, Mém. Soc g

me 1980. — L'influence de l'homme sur la répartition des Amphibiens dans l'Himalaya central et

idental. C. r Soc. Biogéogr . 55: 155-178.

_— 1981. - Liste des genres et sous-genres nominaux de Ranoidea (Amphibiens, Anoures) du monde,

avec identification de leurs espèces-types: conséquences nomenclaturales. Monit. zool. ital., (n.s).

15, suppl.: 225-284.

1987a. - Neoteny and associated terms. A/yies, 4: 122-130.

1987b. - Miscellane: taxinomica batrachologica (1). Alytes, 5 (1-2): 7-95.

lassification des Ranidae (Amphibiens, Anoures). Bull. Soc. linn. Lyon, 61 (10):

Keratodont formulae in anuran tadpoles: proposals for a standardization. Z zool. Syst. Evol.

3 (1): I-XV.

Source : MNHN, Paris

104 ALYTES 19 (2-4)

ee 1999. - Miscellanea nomenclatorica batrachologica. 19. Notes on the nomenclature of Ranidae and

related groups. Alytes, 17 (1-2): 81-100.

Dumois, A. & OuLrr, A., 1998. — A new species of Leprobrachium (Vibrissaphora) from northern

Vietnam, with a review of the taxonomy of the genus Leptobrachium (Pelobatidae, Megophryi-

nae). Dumerilia, 4 (1): 1-32.

- 1999. — Asian and Oriental toads of the Bufo melanostictus, Bufo scaber and Bufo stejnegeri groups

(Amphibia, Anura): a list of available and valid names and redescription of some name-bearing

types. J South Asian nat. Hist., 4 (2): 133-180.

2 2000. - Systematies of Fejervarya limnocharis (Gravenhorst, 1829) (Amphibia, Anura, Ranidae) and

related species. 1. Nomenclatural status and type-specimens of the nominal species Rana limno-

charis Gravenhorst, 1829. A/ytes, 18 (1-2): 15-50.

Dusois, A., OHLER, A. & Buu, S. D., 2001. — A new genus and species of Ranidae (Amphibia, Anura)

from south-western India. Alytes, 19 (2-4): 53-79.

DuELLMAN, W. E. & TRUEB, L., 1985. — Biology of amphibians. New York, McGraw-Hill, “1986”: i-xix +

1-670.

Durra, S. K. & MANAMENDRA-ARACHCHI, K., 1996. — The amphibian fauna of Sri Lanka. Colombo,

Wildlife Heritage Trust of Sri Lanka: 1-232.

EMERsON, $. B., 1996. - Phylogenies and physiological processes. The evolution of sexual dimorphism in

Southeast Asian frogs. Syst. Biol., 45 (3): 278-289.

EMERSON, S. B. & BERRIGAN, D., 1993. - Systematics of Southeast Asian ranids: multiple origins of

voicelessness in the subgenus Limnonectes (Fizinger). Herpetologica, 49 (1): 22-31.

EMrRsoN, S. B., INGER, R. F. & IskANDAR, D., 2000. - Molecular systematics and biogeography of the

fanged frogs of Southeast Asia. Mol. Phyl. Evol., 16 (1): 131-142.

EMERSON, $. & Vois, H., 1992. - Competing explanations for sexual dimorphism in a voiceless Bornean

frog. Funct. Ecol., 6: 654-660.

EMERSON, S. B. & WarD, R., 1998. - Male secondary sexual characteristics, sexual selection, and

molecular divergence in fanged ranid frogs of Southeast Asia. Zool. J Linn. Soc., 122: 537-553.

FEt, L. (ed.), 1999. — Arlas of amphibians of China. Zhengzhou (China), Henan Press of Science and

Technology: [i-ii] + 1-432. [In Chinese.]

FITzINGER, L., 1843. - Systema Reptilium. Fasc. 1. Amblyglossae. Vindobonae, Braumüller & Seidel:

1-106 + i-ix.

GRANT, T., HUMPHREY, E. C. & Myers, C. W., 1997. - The median lingual process of frogs: a bizarre

character of Old World ranoïds discovered in South American dendrobatids. Am. Mus. Novit.,

3212: 1-40.

GÜNTHER, À. 1864. — The reptiles of British India. London, Ray Society: i-xxvii + 1-452, pl. 1-26.

INGER, R. F, 1954. - Systematies and zoogeography of Philippine amphibia. Fieldiana: Zool., 33:

181-531.

966. - The systematies and z0ogeography of the Amphibia of Bornco. Fieldiana: Zool., 52: 1-402.

985. - Tadpoles of the forested regions of Borneo. Fieldiana: Zool., (n.s.), 26: i-v + 1-89.

1996. - Commentary on a proposed classification of the family Ranidae. Herpetologica, 52 (2):

241-246.

INGER, R. F., SHAFFER, H. B., KosHY, M. & BAKDE, R., 1984. — A report on a collection of amphibians and

reptiles from the Ponmudi, Kerala, South India. J. Bombay nat. Hist. Soc., 81 (2): 406- pl. 1-3.

INGER, R. F. & STUEBING, 1997. — À field guide to the frogs of Borneo. Natural History Publications and

Science and Technology Unit, Kota Kinabalu

INGER, R. & Voris, H., 1988. Taxonomic status and reproductive biology of Bornean tadpole carrying

frogs. Copeia, 1988: 1060-1061.

INGER, R.F., VORIS, H. K. & WALKER, P. 1986. Larval transport in a Bornean ranid frog. Copeia, 1986:

523-525

KIRTISINGHE, P., 1957. — The Amphibia of Ceylon. Colombo: i-xiii + 1-12, 1 pl.

Kosucu, J, VENCES, M., DUBOIS, À., OHLER, A. & BÜHME, W,, 2001. - Out of Asia: mitochondrial DNA

evidence for monophyly and Oriental origin of tiger frogs, genus Hoplobatrachus. Mol. Phyl. Evol.,

in press.

LAURENT, R., 1950. - Reptiles et batraciens de la région de Dundo (Angola du Nord-Est). (Première

note). Publçoes cul. Co. Diam. Angola, 6: 7-17.

Liu, C.-C., 1950. - Amphibians of western China. Fieldiana: Zool. Mem., 2: 1-400, pl. 1-10.

Source : MNHN, Paris

Dusois & OHLER 105

Liv, C.-C. & Hu, S.-C., 1961. - The tailless amphibians of China. Beijing, Science Press: [i-fi] + i-xvi +

1-364, pl. 1-6 + 1-28. [In Chinese].

MARMAYOU, J., DUBOIS, A., OHLER, A., PASQUET, E. & TILLIER, A., 2000. — Phylogenetic relationships in

the Ranidae. Independent origin of direct development in the genera Philautus and Taylorana. C.

r. Acad. Sci., Sci. Vie, 323: 287-297.

Marsui, M. 1995. - Calls produced by a “voiceless” frog, Rana blythi Boulenger, 1920, from Peninsular

Malaysia (Amphibia Anura). Tropical Zoology, 8: 325-331.

MaTsuI, M., Wu, G.-F. & SONG, M.-T.,, 1993. - Morphometric comparisons of Rana chensinensis from

Shaanxi with three Japanese brown frogs (genus Rana). Jap. J Herp.. 14: 29-36.

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

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

Myers, G.S., 1942. — A new frog from the Anamallai Hills, with notes on other frogs and some snakes

from south India. Proc. biol. Soc. Wash., 55: 49-55.

NoBLE, G. K., 1931. — The biology of the Amphibia. New York, Dover: i-xviii + 1-577.

Norusis, M. J., 1992. - SPSS for Windows. Professional Statistics Release 5. Chicago, SPSS Inc: [i-xii] +

1-348.

Our, A. & Dugois, A., 1989. - Démonstration de l'origine indépendante des ventouses digitales dans

deux lignées phylogénétiques de Ranidae (Amphibiens, Anoures). Cr Acad. Sci., 309 (3): 419-

422.

es 1999. — The identity of Elachyglossa gyldenstolpei Andersson, 1916 (Amphibia, Ranidae), with

comments on some aspects of statistical support to taxonomy. Zoologica Scripta, 28 (3-4):

269-279.

OHLER, A., GROSIEAN, S. & Hoyos, J. M. 1999. - Observation de la construction d’un nid chez Taylorana

hascheana (Anura: Ranidae). Revue fr. Aquariol., 26: 67-70.

Perers, W., 1863. - Mittheilung über eine neue Schlangengattung, Styporhynchus, und verschiedene

andere Amphibien des zoologisches Museums.

-- 1867. - Herpetologische Notizen. Monatsb. Akad. Wiss. Berlin, 1867: 13-37.

Porr, C.H., 1931. - Notes on amphibians from Fukien, Hainan and other parts of China. Bull. am. Mus.

Nat. Hist., 61: 397-611, pl. 1-10.

SHINE, R., 1979. — Sexual selection and sexual dimorphism in the Amphibia. Copeia, 1979 (2): 297-

306.

SmrrH, M. A. 19224. - The frogs allied to Rana doriae. J nat. Hist. Soc. Siam. 4: 215-225, pl. 9.

= 1922b. - The frogs allied to Rana doriae. Addendum. Z nat. Hist. Soc. Siam., 4: 227-229.

= 1931. — The herpetology of Mt. Kinabalu, North Borneo, 13,455 ft. Bull. Raffles Mu:

pl. 1-2.

== 1940. - The amphibians and reptiles obtained by Mr. Ronald Kaulback in Upper Burma. Rec. indian

Mus., 42: 465-486, pl. 8.

STEINEGER, L., 1925. - Chinese amphibians and reptiles in the United States National Museum. Proc. U.

S. nat. Mus., 66 (25): 1-115.

TayLoR, E. H., 1962. - The amphibian fauna of Thailand. Uni. Kansas Sci. Bull, 63: 265-599.

Vera, M. KosucH, J., OuLer, À. & Dunois, A. 2001. — Systematies of Fejervarya limnocharis

(Gravenhorst, 1829) (Amphibia, Anura, Ranidae) and related species. 2. Morphological and

molecular variation in frogs from the Greater Sunda Islands (Sumatra, Java, Borneo) with the

definition of two species. A/ytes, 19 (1): 5-28.

, M., WaANKE, S., ObiERNA, G., KosuCH, J. & VEITH, M., 2000. - Molecular and karyological data

on the South Asian genera /ndirana, Nyctibatrachus and Nannophrys (Anura: Ranidae). Hama-

dryad, 25 (2): 75-82

YANG, D. (ed.), 1991. - The Amphibia fauna of Yunnan. Kunming, China Forestry Publishing House:

fi-vii] + i-iv + 1-259. [In Chinese].

: 8-32,

Source : MNHN, Paris

106 Duois & OHLER

APPENDIX I

COMPARATIVE MATERIAL EXAMINED

Specimens marked with an asterisk (*) are those which were used for the morphometric analyses (tab.

2-3, fig. 2-3).

Euphlyctis cyanophyetis (Schneider, 1799). - NepaL: Dillikot, 2400 m: MNHN 1975.2164*, 1975.2182*-

2183*, 1975.2194*, 1975.2196*; Sanichare: MNHN 1977.1364-1403; Sukhet, 900 m: MNHN 1996.9274-

9280; Tatopani Khola, 2200 m: MNHN 1975.2250-2273.

Euphlyctis hexadactylus (Lesson, 1834). — INDia: NMW 2512.1*-5*, 25121*.

Limnonectes (Elachyglossa) doriae (Boulenger, 1887). - MYANMAR: Mount Carin, 900-1000m: MNHN

1893.435*-437*.

Limnonectes (Elachyglossa) gyldenstolpei (Andersson, 1916). - Laos: Ban Tup, Bokeo: MNHN

1997.4149%-4152*. - THAILAND: Bang Hue Pong, Koon Tan Mountains, Lamphun Province: NRM 1656*,

holotype; Phu Kradung, Samkokpai, 860-870 m, Loei Province: MNHN 1987.3132*.

Limnonectes (Elachyglossa) toumanoffi (Bourret, 1941). - CamBobia: MNHN 1948.126*, holotype.

Limnonectes (Limmonectes) (gr. grumniens) blythit (Boulenger, 1920). - THaILAND: Khao Phra Tiu:

MNHN 1986.3154*-3168*.

Limnonectes ( Limnonectes) (gr. kuhlit) cf, kuhlii (Tschudi, 1838). — INDONESIA: Sumatra, Sidikalang: MV

80*, MV 82*, MV 102*-103*, MV 105*, MV 108*, MV 111*-112*, MV 117*, MV 127*.- THAILAND: Doi

Inthanon: MNHN 1987.3197. — VierNaM: Tonkin: MNHN 1938.0030.

Limnonectes (Limnonectes) (gr. microdiscus) leytensis (Boettger, 1893). — PHiLippiNEs: Dumaguete:

MNHN 1964.0283*.

Limnonectes (Limnonectes) (gr. microdiscus) woodworthi (Taylor, 1923) — PHILIPPINES: Baybay: MNHN

2000.0611*-0612*.

Nyctibatrachus beddomei Boulenger, 1882. - INDIA: Tinnevelly: BMNH 1882.2.10.27-30*; NHMB 1271*.

Nyctibatrachus humayuni Bhaduri & Kripalani, 1955. — INDIA: BMNH 1958.1.4.25*-26*.

Nyctibatrachus deccanensis Dubois, 1984. — INDIA: Anamallays: BMNH 1947.2.4.47*, 1947.2.4.49*,

1947.2.4.52*, 1947.2.4.55*, syntypes of Rana pyemaea Günther, 1876

Occidozyga lima (Gravenhorst, 1829). — CamBoia: BMNH 1861.4.12.31*-32*, — China: BMNH

1932.5.1.2*, holotype of Houlema obscura Gray, 1831. —INDONESIA: Java: BMNH 1844.2.22.94A+-94C*.

= THAILAND: Siam: BMNH 1859.7.1.36*-39*. - Cuina: Jinghong, Yunnan Province: MNHN 1999.6416*-

6422.

Phrynoglossus magnapustulosus Taylor & Elbel, 1958: — CHina: Jinghong, Yunnan Province: MNHN

1999.6442-6453.

Phrynoglossus martensii Peters, 1867. — THAILAND: Khao Chong, Trang Province: MNHN 1987.2894*,

1987.2898*, 1987.2907*, 1987.2915*, 1987.2925*, 1987.2934*, 1987.2938*, 1987.2940*, 1987.2958*,

1987.2960*

Pipa carvalhoi (Miranda-Ribeiro, 1937). — BRAZIL: Bahia: MNHN 1981.298-299.

Pseudis paradoxus (Linnaeus, 1758). - VENEZUELA: Montecal: MNHN 1983.0390.

Rana ( Pelophylax) hubeiensis Fei & Ye, 1982. - CHiNa: Zhejiang: MNHN 1931.0064-0066.

Silurana tropicalis Gray, 1864 — GUINEA: Mount Nimba, Region of N20: MNHN 1944.0162-0164,

1994.1907-1927.

Corresponding editor: Franco ANDREONE

Source : MNHN, Paris

Alytes, 2001, 19 (2-4): 107-139.

Address for correspond

rallee 160, 53113 Bonn, Germa:

Systematic review

of the Mantidactylus asper group

(Amphibia, Mantellidae)

Miguel VENCES*! & Frank GLAW**

* Muséum National d'Histoire Naturelle, Laboratoire des Reptiles et Amphibiens,

25 rue Cuvier, 75005 Paris, France

atssammlung, Münchhausenstr. 21, 81247 München, Germany

<Frank.Glaw@zsm.mwn.de>

** Zoologische Si

The taxonomy and distribution of Malagasy frogs of the Mantidactylus

asper group (included in the subgenus Gephyromantis) is revised. The

group is considered to include Mantidactylus asper, M. spinifer, M.

luteus, M. plicifer, M. sculpturatus (which is resurrected from the syn-

onymy of M. luteus) and a new species described herein. Lectotypes are

designated for Rana aspera Boulenger, 1882 (Mantidactylus asper),

Mantidactylus ceratophrys Ahl, 1929 (junior synonym of M. asper) and

Rana plicifera Boulenger, 1882 (Mantidactvlus plicifer). M. asper and M.

spinifer are characterized, among other features, by a distinct black-brown

contrasted ventral pattern, presence of an outer metatarsal tubercle, and a

moderate amount of webbing. They are apparently allopatrically dis!

uted, M. spinifer occurring in south-eastern Madagascar and M. asper

inhabiting eastern and north-eastern rainforests. M. luteus, M. plicifer and

M. sculpturatus have a largely uniform light venter, lack the outer meta-

tarsal tubercle and have more extended webbing. A reliable distinction of

these three species is only possible in adult males, and is based on

differences in femoral gland size and advertisement calls. M. luteus is

mainly distributed in lowlands along the Malagasy east coast, while M.

sculpturatus appears to be restricted to mid-altitudes. M. plicifer has been

found sympatrically with M. sculpturatus and M. luteus, and is known from

the south-east. The new species described herein shares characters with M.

asper and M. spinifer (presence of an outer metatarsal tubercle) and with

M. luteus, M. plicifer and M. sculpturatus (uniform venter, extended

webbing). It is only known from Montagne d'Ambre in far northern Mada-

gascar.

À molecular phylogenetic analysis based on partial sequences of the

mitochondrial 16$ rRNA gene supported monophyly of the M. granulatus

group and of the M. pseudoasper group in the subgenus Phylacomantis,

and of a clade containing M. luteus, M. plicifer and M. sculpturatus. In

contrast, the M. asper group and the subgenus Gephyromantis as a whole

appeared to be paraphyletic. The obtained trees indicated a possible

evolution of the direct-developing lineage from brook breeding ancestors,

and a reversal from direct development in M. granulatus. Although these

; <m.vences(@t-online.de>.

107

and molecular phylogenetic relationships

of the direct developing Malagasy anurans

Zoologisches Forschungsinstitut und Museum Alexander Koenig, Adenaue-

Source : MNHN, Paris

108 ALYTES 19 (2-4)

aspects received no relevant bootstrap support, they constitute hypotheses

of great interest for general questions on amphibian evolution and should

be tested with extended data sets.

INTRODUCTION

The Malagasy genus Mantidactylus currently contains more than 70 species which show

a large diversity in morphology and reproductive biology. DUBoIs (1992) and GLAW & VENCES

(1994) divided the genus into a total of 12 subgenera, reflecting this diversity. Major charac-

teristics of all Mantidactylus include the absence of nuptial pads in males (and, as far as

known, of a strong mating amplexus), and the deposition of eggs outside the water in

all species studied so far (BLOMMERS-SCHLÔSSER, 1979). Most species additionally have

specialized femoral glands on the ventral surface of thighs (GLAW et al., 2000), especially in

males.

GLAwW & VENCES (1994) distinguished three major clades within Mantidactylus. One

group contains more or less arboreal species which mostly deposit their eggs on leaves above

the water surface; their tadpoles are rather generalized (subgenera Blommersia, Guibemantis,

Pandanusicola, Spinomantis). À second group consists of brook-edge-dwelling species, the

larvae of which often show specialized mouthparts (subgenera Brygoomantis, Chonomantis,

Hylobatrachus, Mantidactylus, Ochthomantis). The third assemblage contains three subge-

nera (Gephyromantis, Laurentomantis, Phylacomantis): while Phylacomantis males usually

call along brooks and at least some species have free-swimming larval stages, calling males of

most Laurentomantis and Gephyromantis do not aggregate around water bodies, and direct

development without free-swimming larval stages has been demonstrated in two species,

Mantidactylus asper and M. eiselti (BLOMMERS-SCHLÔS: 1979; GLAW & VENCES, 1994).

Due to this reproductive diversity, studies on these frogs have the potential to contribute to the

understanding of the evolution of direct development and other specializations in anuran

reproductive biology.

One basic pre-requisite for such studies, however, is a detailed basic knowledge on the

species’ taxonomy and distribution (GLAW & VENCES, 2000). Distributional data of Malagasy

frogs are largely based on the monograph of BLOMMERS-SCHLÔSSER & BLANC (1991) who,

however, mostly did not recognize sibling species and gave no voucher specimens for the

localities plotted on their distribution maps. Apart from type specimens from other collec-

tions, their work was based almost exclusively on the collections housed at Amsterdam and

Paris.

In the present paper, we review the Mantidactylus asper group, a phenetic species

assemblage in the subgenus Gephyromantis, distinguished from other Mantidactylus by

reproduction independent from water, mainly nocturnal calling behaviour, largely separated

lateral metatarsalia, and black paired subgular vocal n males. We re-examined the

material available to BLOMMERS-SCHLÔSSER & BLANC (1991), and complemented this infor-

mation by own field observations.

Source : MNHN, Paris

VENCES & GLAW 109

MATERIALS AND METHODS

ABBREVIATIONS AND MEASUREMENTS

Vocalizations were recorded using portable tape recorders with an external microphone

(Vivanco EM 238) and were analyzed with the MEDAV sound analyzing system Spektro 3.2.

The following morphological measurements were taken with a caliper to the nearest

0.1 millimeter: SVL, snout-vent length; HW, head width; HL, head length; ED, horizontal eye

diameter; END, eye-nostril distance; NSD, nostril-snout tip distance; NND, nostril-nostril

distance; TD, horizontal tympanum diameter; HAL, hand length: FORL, forelimb length;

HIL, hindlimb length; FOL, foot length; FOTL, foot length including tarsus; IMTL and

IMTH, length and height of inner metatarsal tubercle; TLI, length of first toe. Statistical

analyses were carried out using SPSS for Windows, version 10. We performed Mann-Whitney

U tests to test significance of intersexual differences in size and morphometric ratios

(TD/SVL, relative tympanum diameter; IMTL/SVL and IMTH/SVL, relative size of inner

metatarsal tubercle; FORL/SVL and HIL/SVL, relative length of fore- and hindlimbs), and

of interspecific differences in selected morphological variables and ratios. Measurements are

given as range, with mean + standard deviation in parentheses.

INSTITUTIONAL ABBREVIATIONS

BMNH, The Natural History Museum, London (formerly British Museum of Natural

History); MNHN, Muséum National d'Histoire Naturelle, Paris: MRSN, Museo Regionale

di Scienze Naturali, Torino; MSNG, Museo Civico “G. Doria” di Storia Naturale, Genova;

MTKD, Museum für Tierkunde, Dresden; TM, Transvaal Museum, Pretoria; UADBA,

Université d’Antananarivo, Département de Biologie Animale; ZFMK, Zoologisches Fo:

chungsinstitut und Museum Alexander Koenig, Bonn; ZMA, Zoëlogisch Museum, Amster-

dam; ZMB, Museum für Naturkunde, Berlin, ZSM, Zoologische Staatssammlung, Mün-

chen. The catalogue numbers of voucher specimens housed in the ZMA are given as the jar

number followed by the field number of R. Blommers-Schlôsser, since they bear no individual

ZMA tags.

TAXONOMY

To avoid confusion by introducing working definitions (operational taxonomic units)

gning them to specific names in a second step, we decided to anticipate our taxonomic

ls and use consistent names throughout this paper. This mainly regards: (1) the

recognition of the Montagne d'Ambre population previously considered as Mantidactylus

plicifer by BLOMME r & BLANC (1991) or as M. cf. asper by GLAW & VENCES (1994)

as a new species which is described herein; (2) the re-definition of Mantidactylus plicifer as a

Source : MNHN, Paris

110 ALYTES 19 (2-4)

species of usually rather large body size and with large and distinct femoral glands from

south-eastern Madagascar; (3) the recognition of mid-altitude eastern populations previously

assigned to M. luteus by GLAW & VENCES (1994) as a distinct species M. sculpturatus. These

decisions are largely corroborated by high genetic divergence levels between the species

recognized, by the morphological differentiation of the new species from Montagne d’Ambre

and by the morphological and bioacoustic differentiation and syntopic occurrence of M.

sculpturatus and M. plicifer at Ranomafana. More detailed justifications are given in the

respective /dentity and Diagnosis sections below.

MORPHOLOGICAL TERMINOLOGY

Webbing formula is given according to BLOMMERS-SCHLÔSSER (1979). Femoral gland

morphology is described according to GLAW et al. (2000). Most Gephyromantis species are

characterized by a number of dermal spines, tubercles and ridges. The arrangement and

degree of expression of these structures is often important for species definitions and

probably also bears relevance for the assessment of phylogenetic relationships among species

and subgenera in the genus Mantidactylus. To refer unequivocally to these structures, we here

define a number of terms (fig. 1):

() {nter-ocular tubercles.— On the upper surface of the head, between the eyes, a number

of tubercles are present in many species. These are generally arranged symmetrically, either as

one pair or as two pairs, and should not be mistaken with the unelevated black inter-ocular

spots as present in M. leucomaculatus (Phylacomantis). In several Phylacomantis (M. cornu-

tus, M. redimitus, M. tandroka, M. tschenki), one pair of rounded, black tubercles are

generally present. On the contrary, in M. asper, M. spinifer and the new species described

herein (Gephyromantis), the tubercles are generally not rounded but rather longitudinal and

ridge-like; often, two pairs of such tubercles are present which sometimes appear to be a

discontinuous anterior continuation of the inner dorsolateral ridges; and sometimes, in M.

spinifer, these ridge-like tubercles are fused to form a symmetrical figure (fig. 1).

(2) Inner dorsolateral ridges.— As a constant state in all species of the M. asper group, two

largely continuous ridges start above or up to 4 mm behind the eyes and run medially onto the

anterior back. Here they either continue straight dorsolaterally onto the posterior fourth of

the back, or curve slightly towards the flanks and fade.

(3) Outer dorsolateral ridges. — In all species of the M. asper group, a second pair of

dorsolateral ridges runs laterally of the inner dorsolateral ridges. Often this second pair is not

continuous and poorly defined.

(4) Connecting dorsal ridge. — In some M. spinifer, at the point of maximum convergence

of the inner dorsolateral ridges on the anterior dorsum, these are connected by a short

transversal ridge.

(S) Supraocular spines. — Above the ey:

species of the M. asper group (and in several Phylacomantis: e.g.

M. tschenki), although they can be small and indistincet in the new species described ee In

contrast to the rather rigid dorsal ridges, these spines are flexible.

Source : MNHN, Paris

VENCES & GLAW 111

Fig. 1. - Partial dorsal views (head and anterior body and hindlimb) of three representative species of the

Mantidactylus asper group: Mantidactylus spinifer (left, female MNHN 1972.1444; M. plicifer

(center), female MNHN 1972.1431; and M. ambohitra (right), female MNHN 1893.245. The arrows

mark dermal structures which are described in the text: CDR, connecting dorsal ridge; HSP, heel

spine: IDR, inner dorsolateral ridge; OT, inter-ocular tubercles, ODR, outer dorsolateral ridge:

SOSP, suprocular spines; SOT, supraocular tubercles; TASP, tarsal spines. Not to scale.

(6) Supraocular tubercles. — These are homologous to supraocular spines but less elevated

and not pointed.

(7) Heel spine. — A long or short spine may be present on the heel. Similar to the

supraocular spines, the heel spine is a dermal, flexible structure. Beside the M. asper group, it

is also found in several Phylacomantis and in species of the genus Boophis (e.g., B. madagas-

cariensis).

(8) Tarsal spines. — À number of smaller dermal spines are sometimes arranged at the

posterior edge of tarsus. Species with tarsal spines always bear also a distinct heel spine.

() Humeral protuberance. — À well known synapomorphy of Mantidactylus species are

the femoral glands on the ventral surface of the femur (GLAW et al., 2000). In several species

of the M. asper group (M. luteus, M. plicifer, M. spinifer), we observed a prominent structure

on the ventral side of the humerus, too (fig. 2). By dissection (internal view; GLAW et al.,

2000), we noted that this prominence was not caused by enlargement of skeletal or muscular

tissues but largely dermal, and sometimes contained a gland-like element at its most promi-

nent part. Without further histological analyses we are unable to state whether this structure

is actually a gland, and thus here refer to it as humeral protuberance. It is best visible in

Source : MNHN, Paris

112 ALYTES 19 (2-4)

Fig. 2. - Left: ventral views of a male (above; MNHN 1972.1462) and a female (below; MNHN

1972.1443) of the forelimb of Mantidactylus spinifer. The arrows mark the humeral protuberances.

Note also the strongly contrasted ventral pattern in both sexes. Right: ventral views of a male (above:

MNHN 1972.1436) and a female (below: MNHN 1972.1431) of the forelimb of M. plicifer. In this

species, the humeral protuberance is very distinct in males but apparently absent in females.

Mantidactylus plicifer and M. luteus, where a clear sexual dimorphism exists (no protuberance

recognizable in females; fig. 2). In other species, e.g. M. spinifer, a weakly expressed humeral

protuberance appears to be present in males and females (fig. 2).

MOLECULAR AND PHYLOGENETIC METHODS

To sample data on genetic differentiation and phylogenetic relationships, we sequenced

fragments of the mitochondrial 16 rRNA gene of up to 567 base pairs (bp), using primers

and protocols given by VENCES et al. (2000), in species of the M. asper group and of related

taxa. The corresponding voucher specimens and EMBL/Genbank accession numbers are:

Mantidactylus asper, UADBA-FG/MV 2000.17, AJ314802; M. luteus, ZFMK 66674,

AF215313; M. plicifer, ZFMK 62306, AJ314800; M. sculpturatus, ZFMK 62304, AJ314799;

M. boulengeri, ZFMK 66672, AF215 M. cor ZFMK 70494, AF215

pseudoasper, MSNG 49087, AJ314803; M. malagasius, MRSN A1991, AJ314797; M. af.

malagasius, MRSN A1938, AJ314796; M. granulatus, ZSM 645.2001, AJ314794; M. sp. n.

(Tsaratanana; new species close to M. granulatus), ZSM 627.2001, AJ314795; M. leucomacu-

latus, ZFMK 59953, AJ314805: M. cornutus, ZSM 308.2000, AJ314798; M. 1schenki, ZFMK

62296, AJ314806: M. tandroka, ZFMK 59894, AJ314803. Boophis xerophilus (ZFMK 66705,

Source : MNHN, Paris

VENCES & GLAW 113

AF215335) and Mantidactylus liber (subgenus Guibemantis, ZSM 491.2000, AJ314801) were

used as outgroups. Sequences of the new species described below were obtained from the

specimen MTKD 37424 and have the Genbank accession number AF215327.

Sequences were aligned using the CLUSTAL algorithm in SEQUENCE NAVIGATOR

(Applied Biosystems); subsequently, the alignment was adjusted by eye. A total of 33 base

pairs of the hypervariable region could not be reliably aligned and were excluded from further

analysis. The aligned sequences were submitted to analysis using PAUP*, version 4.0 (SwWor-

FORD, 2001). We used MODELTEST (Posapa & CRANDALL, 1998) to estimate the model of

sequence evolution for a Maximum Likelihood (ML) phylogenetic reconstruction. We addi-

tionally performed Maximum Parsimony (MP) analyses, using the heuristic search option

with tree-bisection reconnection branch-swapping, and Neighbor-joining (NJ) analyses, with

LogDet distances which are robust against possible variation of sequence evolution among

lineages (LOCKHART et al., 1994). In the MP analyses, we coded gaps as fifth state, but also

performed additional searches excluding all characters with gaps in one or more species. We

used PAUP* to test for the presence of a significant phylogenetic signal by a permutation-

tailed-probability (PTP) test with 100 replicates, and for homogeneity of base frequencies

across sequences.

RESULTS

SPECIES ACCOUNTS

Mantidactylus asper (Boulenger, 1882)

(fig. 3a-b)

Rana aspera Boulenger, 1882. — Name-bearing type: lectotype, by present designation, BMNH

1882.3.16.80, adult male collected by W. D. Cowan, SVL 27.8 mm. - Tÿpe localit s

according to original description. - Osher types: paralectotypes, following present lectotype designa-

tion, BMNH 1882.3.16.81-90, same collecting data as lectotype.

Mantidactylus ceratophrys Ahl, 1929. — Name bearing type: lectotype, by present designation, ZMB

10443, adult female, collected by J. M. Hildebrandt according to original description. — Type locality:

“Betsileo” according to original description. — Other types: paralectotypes, following present

lectotype designation, ZMB 10444 and 50501-50502, three adult females with same collecting data

as lectotype.

Comments. —(1) Considering the existence of at least one new species in the M. asper group as

described below it appears important for taxonomic stability to define single name-bearing

types for all taxa in the group. Following this rationale, we here designate lectotypes for M.

asper and its junior synonym M. ceratophrys. Detailed morphological measurements of these

lectotypes are given in tab. 1.—(2) BLOMMERS-SCHLÔSSER & BLANC (1991) listed the specimens

“BMNH 1882.316.80-82, 83-86” as syntypes of M. asper. The number “316° in this mention

is certainly a typing error for “3.16”. Beside BMNH 1882.3.16.80-81, all specimens of the

Source : MNHN, Paris

114 ALYTES 19 (2-4)

Fig, 3.- Photographs of Mantidactylus asper, male from Ankeniheny (ZFMK 60789), in dorsolateral and

ventral views (a-b); of Mantidactylus ambohitra, male holotype from Montagne d'Ambre (ZSM

1084.2001) in dorsolateral and ventral views (c-d); and of two additional specimens of M. ambohitra:

female ZFMK 57419 (e) and a calling male specimen (not collected), photographed by J. Kôhler (M.

both from Montagne d'Ambre.

series BMNH 1882.3.16.83-90 are labeled as syntypes of the taxon in the London museum.

We examined four specimens of this series which were not individually numbered. We could

not locate the specimen BMNH 1882.3.16.82 in the London collection; it may have been

exchanged or used for osteological examinations. It can be assumed that the series of

paralectotypes (according to present lectotype designation) consists of at least nine, possibly

ten specimens. — (3) BLOMMERS-SCHLÔSSER & BLANC (1991) stated that the type of Mantidac-

Source : MNHN, Paris

VENCES & GLAW 115

tylus ceratophrys Ahl, 1929 was lost. In the meantime, four original syntype specimens have

been rediscovered in the ZMB collection (lectotype and paralectotypes according to present

designation).

Material examined. - BMNH 1882.3.16.80-81, 1882.3.16.83-86 (East Betsileo, lectotype and paralecto-

types); BMNH 1925.7.2.29 (Antsihanaka); BMNH 1928.5.9.1 (Brickaville); MNHN 1972.557-558 (Tsa-

ratanana); MNHN 1972.559-560 (Mandraka); MNHN 1972.574-576, 1972.578-583, 1972,586-591 (Tsa-

ratanana); MNHN 1972.584-585 (locality uncertain); MNHN 1973.905 (Marojejy, alt. 1300 m); MNHN

1975.313 (Marojejy); MNHN 1975.317 (no precise locality); MNHN 1975.318 (Marojeiy, alt. 1300 m);

MNHN 1975.319-320 (Marojejy, alt. 2000 m); MNHN 1975.321 (Mandraka); ZFMK 62236-62237,

62240 (Mantady); ZFMK 62303 (Ranomafana); ZFMK 60789 (Ankeniheny): ZMA 6867.988-990 and

6868.470 (Andasibe); ZMA 6894.36, 6895.376-383 and 6895.480-482 (Mandraka); ZMA 6897.906

(Andasibe): ZMB 50501-50502 and 10443-10444 (Betsileo, lectotype and paralectotypes of Mantidacty-

lus ceratophrys); ZSM 401.2000 and UADBA-FG/MV 2000.17 (Mandraka).

Morphology and diagnosis. — The following morphological description is based on the type

series, and on specimens from central eastern Madagascar (Mandraka, Andasibe, Mantady,

Antsihanaka, Brickaville). Specimens from these localities form a well-defined homogeneous

cluster, although several characters (expression of dorsal tubercles and ridges, relative hind-

limb length) are subject to considerable individual variation. Other attributed specimens are

discussed in the section on distribution.

Inner dorsolateral ridges prominent; either continuous (e.g. in ZMA 6867.990) or

discontinuous (e.g. in ZMA 6867.989), starting 1-3 mm behind eyes. No connecting dorsal

ridge in the specimens examined by us. Outer dorsolateral ridges generally present, but often

not continuous, appearing as an irregular series of short folds and tubercles. One or two pairs

of distinct more or less prominent ridge-like interocular tubercles. Additional smaller tuber-

cles and short ridges on the dorsum of most specimens. Distinct supraocular spines in all

specimens, two of these being especially large. A distinct heel spine; tarsal spines generally

reduced to small tubercles, sometimes not recognizable (e.g. in several specimens of the type

series). Femoral glands usually visible in males, but not very distinct or prominent. Vocal sac

paired subgular. No clearly recognizable humeral protuberance in either sex. Webbing in most

specimens reaching slightly beyond the first subarticular tubercle of the fifth toe: webbing

formula 5(0.5) to 5(0.75). One inner and two outer metacarpal tubercles, the latter in contact

with each other. A large (males) or medium-sized (females) inner metatarsal tubercle and a

distinct, small to medium-sized outer metatarsal tubercle.

Significant or near-significant intersexual differences were found in SVL(Mann-Whitney

Utest, P <0.06), in relative length of the inner metatarsal tubercle (P <0.005), and in relative

tympanum diameter (P < 0.05), but not in relative height of the inner metatarsal tubercle or

limb length. Males had larger relative tympanum sizes and longer inner metatarsal tubercles

than females. Male/female size ratio was 97 %.

Coloration. — In-preservative, dorsal coloration generally brown, with more or less symmetri-

cal light brown or dark brown markings. In ZMA 6867.988, dark brown vertebral area

enclosed between inner (anterior dorsum) and outer (posterior dorsum) dorsolateral ridges,

and beige flanks and areas lateral to the ridges, the beige color starting as narrow dorsolateral

bands above the eyes. Hindlimbs brown with dark brown crossbands. A thin light vertebral

line in ZFMK 62236. Head laterally brown, including the lips, with some rather indistinct

Source : MNHN, Paris

116 ALYTES 19 (2-4)

dark brown markings. Ventral side white on the chest and, in females, on the throat, more

cream on the posterior belly. Throat in males light brown with a distinct central white

longitudinal stripe, and with black lateral color coinciding with the inflatable parts of the

vocal sac. In females, indistinct but sharply delimited brown vermiculations on throat and

chest. Limbs ventrally light brown, with some darker pattern at the edges.

Distribution. - Beside the specimens from central eastern Madagascar, we also assign to M.

asper rather than to M. spinifer one subadult male from Ranomafana (ZFMK 62303; vocal

sac not visible, but throat laterally already slightly black colored) based on its smaller size,

ventral pattern, and expression of ridges and tubercles. The situation is more difficult for the

available material from the Marojejy and Tsaratanana massifs in northern Madagascar. These

specimens are larger than typical M. asper, have more weakly expressed spines and ridges

(supraocular spines reduced to tubercles in most specimens), and a very weak ventral pattern.

In part, they thus remind specimens from Montagne d’Ambre which are described below as a

new species. However, the low amount of webbing (at fifth toe consistently 0.75-1), as well as

other characters constitute a distinct difference to that species, and support the tentative

inclusion of the Tsaratanana and Marojejy populations in M. asper. Despite low number of

females in this sample (only two specimens), the inner metatarsal tubercle is significantly

relatively longer in males (P < 0.05; tab. 2-3). Male/female size ratio is 99 %.

Summarizing, the species is known from the following precise localities: (1) Tsaratanana;

(2) Marojejy; (3) Antsihanaka; (4) Brickaville; (5) Mantady; (6) Andasibe; (7) Ankeniheny; (8)

Mandraka; (9) Ranomafana. Except for Brickaville, which is located close to sea level at

the east coast, all localities in the central east are at mid-elevations, ranging from ca. 700 to

1200 m. Marojejy specimens are catalogued as originating from an elevational range of 1300-

2000 m.

BLOMMERS-SCHLÔSSER & BLANC (1991) additionally recorded the species from Les

Roussettes (Montagne d’Ambre) and Ambatofitoharanana. Specimens from the former

locality are here attributed to the new species described below, while the voucher specimen

from Ambatofitoharanana (MNHN 1975.315) cannot be reliably attributed to any described

species (see section at the end of the species accounts).

Natural history. - Calling males were observed during the day on the ground (at Ankeniheny)

and after dusk from branches about 50 cm above the ground (at Mantady and Mandraka). At

Ankeniheny, they were mainly found close to a swampy brook, while they called far from

water bodies at Mantady, and close to a tiny brook at Mandraka. Vocal sacs did not remain

inflated between notes. Each note was one expiration.

Advertisement call. - Calls were recorded at Ankeniheny on 18 February 1994, 17.15 h, at

23.5°C air temperature. They consisted of series of single notes or series of note groups of

2-4 notes each (fig. 4). Note duration was 5-13 ms (10 + 3 ms, » = 8), duration of intervals

between notes was 56-80 ms (66 + 11 ms, 7 = 6). Frequency was 1700-7700 Hz, dominant

frequency 3200-4200 Hz.

Similar calls were heard after dusk on 10 February 1996 at Mantady. Only series of single

notes were heard (note repetition rate 1.3-1.4 per second). Frequency was about 2000-

5000 Hz, dominant frequency 3500-4500 Hz.

Source : MNHN, Paris

VENCES & GLAW 117

Frequency (kHz)

4 = = =

0+ ï ï L !

0 100 200 300 400 500ms

Fig. 4. - Sonagram and oscillogram of a call (series of four notes) of Mantidactylus asper from

Ankeniheny.

Calls from Mandraka (recorded by R. Blommers-Schlôsser in December 1972 at 17.00 h)

were series of note groups (up to 7 notes per group). Note duration was 13-20 ms (16 + 2ms,

ñ=9), duration of intervals between notes 72-80 ms (75 + 2 ms, n = 7). Note groups with 4-7

notes had a duration of 323-494 ms (386 + 69 ms, n = 5), duration of intervals between note

groups was 626-733 ms (684 + 44 ms, n = 4). Frequency was 1800-6000 Hz without

recognizable dominant frequency.

Mantidactylus spinifer Blommers-Schlôsser & Blanc, 1991

Mantidactylus spiniferus Blommers-Schlësser & Blanc, 1991. - Mame-bearing type: holotype, by original

designation, MNHN 1972.1450, adult male collected by C. P. Blanc in November-December 1971

according to MNHN catalogue. — Tÿpe localit: “Chaînes Anosyennes” according to original

description: “Camp IV, Chaînes Anosyennes” according to MNHN catalogue. - Paratypes: MNHN

1972.1440 and 1972.1470, adult female and male.

Material examined. -(1) types and topotypical material: MNHN 1972.1450 (holotype, Camp IV, Chaînes

Anosyennes): MNHN 1972.1440 (paratype, Camp IV, Chaînes Anosyennes); MNHN 1972.1470 (para-

type, Ambana-Camp IV, Chaines Anosyennes); MNHN 1972.1439, 1972.1443-1468 (Camp LV, Chaînes

Anosyennes); (2) further material: MNHN 1972.551-554 (Ivohibe, Marovitsika forest).

nues and diagnosis. — Beside the three type specimens designated by BLOMMERS-

ôsser & BLANC (1991), nine males, thirteen females and four juveniles from the same

Source : MNHN, Paris

118 ALYTES 19 (2-4)

series and general locality were available. Considering this material, the following updated

description of M. spinifer (based on the topotypic material only) can be provided. Identity of

other specimens is discussed in the section on distribution.

Inner dorsolateral ridges prominent, starting 0-2 mm behind the eye. Dorsal connecting

ridge present in some specimens and completely absent in others. Outer dorsolateral ridges

generally present, but mostly not continuous and poorly defined, often appearing as an

irregular series of short folds and tubercles. One pair of distinct and very prominent ridge-like

interocular tubercles, a second, less distinct anterior pair often being also visible; the two pairs

sometimes fusing to a single symmetrical structure (fig. 1). A large number of additional

irregular tubercles and short ridges scattered on the dorsum. Distinct supraocular spines in all

specimens, two of these being especially large. A distinct heel spine and a line of 2-7 distinct

tarsal spines. Males with very distinct and rather prominent femoral glands. Vocal sac paired

subgular. À weakly expressed humeral protuberance in both sexes (fig. 2). Webbing in most

specimens reaching distinctly beyond last subarticular tubercle of fifth toe, but not reaching

toe disc: webbing formula 5(0.25) to 5(0.75). One inner and two outer metacarpal tubercles,

the latter two in contact with each other. A large inner and a distinct, small to medium sized

outer metatarsal tubercle.

Morphometric measurements are given in tab. 1-3. No significant sexual dimorphism

was found in relative forelimb length (P > 0.5), but the remaining variables tested showed

significant differences between sexes: females were significantly larger (mean male/female size

ratio 96%: P < 0.05), had shorter relative hindlimbs (P < 0.05), and smaller relative

tympanum sizes (P < 0.01). Relative length and height of the inner metatarsal tubercle were

also distinctly smaller in females (P < 0.001).

Coloration. — The following description is based on preserved specimens only, as no color

pictures of living specimens were available. Dorsal coloration with various tones of brown,

generally arranged in longitudinal patterns. Narrow or broad brown bands sometimes

arranged dorsolaterally, following the dermal ridges, on a beige or light brown dorsum

(MNHN 1972.1443, 1972.1449), or dark brown central dorsum with cream-beige flanks

(MNHN 1972.1458, 1972.1461). In most specimens a rather indistinct mixture of dark and

light brown.

Allspecimens ventrally with a very distinct coloration, the dark flank color reaching onto

the posterior venter and ending abruptly, without any fading, bordering onto the ventral color

which (in preservative) is bright white (fig. 2). Males with a brownish throat (laterally black,

corresponding to the inflatable parts of the vocal sac) with a more or less distinct median

white stripe. Females with a white throat and a brown border along the lower lip. Limbs

ventrally more or less distinctly marbled with dark brown and white. This conspicuous ventral

color pattern is already visible in juveniles (MNHN 1972.1452, 1972.1466-1468) of 14-18 mm

SVL.

Distribution. — Beside the specimens from the type locality Chaînes Anosyennes, BLOMMERS-

ScHLüsser & BLANC (1991) included three additional localities of M. spinifer in their distri-

bution map: Marojejy, Marovitsika and Andringitra. The MNHN vouchers from Marojejy

were herein assigned to M. asper and we did not locate vouchers from Andringitra assignable

to the species. However, a series from the Marovitsika forest (Pic d'Ivohibe) could be included

Source : MNHN, Paris

VENCES & GLAW 119

Table 1. Morphometric measurements (al in mm) of name-bearing types in the Mantidacylus asper group. For abbreviations of measured variables, see

Materials and methods other abbreviations used are: M (male), F (female), RHL (relative hindlimb length, gives the position reached by the

tibiotarsal articulation when the hindlimb is adpresscd along the body).

D EE EEE

Boteger: 12 Abl, 1929 & Blanc, 1991 sp. pre hi Abl, 1929

Status Lectotype Lectotype Holotype Holotype Lectotype Holotype

nl un, lon Manual lame

‘Table 2. - Morphometric measurements (all in mm) of male specimens of species in the Mantidactylus asper group. Abbreviations as given in Materials and

methods and in caption to tab 1. The measurements of M. asper are given separatelÿ for populations from north-castern (NE; Tsaratanana, Marojejy)

and central eastem (CE) Madagascar. n is he number of measured specimens. Measurements are given as mean à standard deviation (range in

parentheses) Relative hindlimb length (RHL) is coded as follows: when adpressed along body, the tibiotarsal articulation reaches (1) the nostrl, (2)

te snout tip, (3) beyond the snout tip or (4) widely beyond the snout tip.

Species | A asper(CE) | A asper(NE) M spimfer M ambokitra M tueus M plicifer M sculpturans

" mn 15 n 5 n 2 4

svL [243409066302 | 8216008355 | 1410025250 | asoa12çu0s72 | mo+22057200 |aso:25 88482 | 403224679450)

Hu |iogaospons |ssosairm | 9105022139) | 134404028138) | iéoaorqaern | m7+1aaraon | 163212046172

TD | 23+050029 | 2940504359 | 27+010s29 | 27.050430 | 22020539 | 2740202429) | 32.050739

en | s4s02003n | s8+020340 | 4020205843 | 4240466409 | assosuesn | 5240544759 | s04034655

END | 10:020635 | 3420210136 | 252020240 | 32020937 | «24030749 | 4510406449 | 45202434

np | isso2us2n | 2240208206 | 2420202206 | 21+020020 | 24102002n | 30+030435 | 26205450

NND | 2320201829 | 2740350039 | 25202327 | 292030234 | 4030739 | s4ao2goan | 54016739

FoRL | 1ssiaqésnn | nisiacoi2sn |22413000255) | 238409 @ 7246 | 134100332606 | 9421025110 | 254113044272

rorL |as1i@i22sn | 3412063209 | 286413062209 | 2340908000 | ma+15096352) | +28 ç00407 | 353219036380)

nai | assoesan | 374046240 | 272050242) | 40.020742 | 4214046550 | sis060asm | 46s04uiso

Fou | sosoou2so | 4920655 | ssaos(sos) | siaosusso | sososuérn | 542070505 7040

rGw | issosuéan | 1sso2620 | 23403420) | 21s060725 | 1é«os2n | 32.052537 16@=1)

RHL 14 34 34 34 4 14 34

Source : MNHN, Paris

120 ALYTES 19 (2-4)

“Table 3. - Morphomenrie messurement (ll in mm) of female specimens of species in he Mantidactylus asper group. Abbreviations as given in Materials

and methods and in he eaption 1 tb, 1, The mexsurements oF M. asper a given separatcly for populations from north-caster (NE; Tsaratanana,

Maroejy) and central eastem (CE) Madagascar. is he number of measured specimens. Measurement are given as mean à standard deviation (range

in parentheses), Relative hindlimb length (RHL) s coded as follows: when adpressed along body, the tibiotarsal articulation reaches (1) the nostil, (2)

Ah snout ip, (5) beyond the snout ip or (4) widely beyond the snout tip.

Species | M asper(C) M asper (NE) M spinfer M ambohitra M luteus M plier

“ nl z mn En n D

sv | 2wi0s@2209 | sasoiinss | s4s15ç2ia8n | 352208024189 | ais26uosaro | 469419046408)

mw | o204100 | niso7gosro | nissosoris | nussooqoons | iéos12(sins | 1607059180

mu | iosscaozns | moso6qéi2s | 1340502510 | nssosqionumn | i2ei2qrauss | 1610074108

m 2140201824) 2920207230) 2520202228) 2620202330) 3140302834 2642020431)

ED 3542026238) 3382026639) 4120206743) 41203 0649 54204(5060 56404860)

END | 30+020832) 33201 6233) 35202341) 3340202839 48202(4650) 48203453)

NsD | 1940201523) 20205(1623) 2520202127 21402(1620 28010730) 31403073

np | 2134030529 302006030) 2520202230) 282020532 37+026539) 45202250)

FORL | 1521005208 | 231406026239 | 2140902229 | 2921408269 | nossiqsins | 2741281317

MAL | 90405190 | iossoaçosiio | iossosqomnin | nssosqorno | nisiiqr2uo | 1040602819

mn | sésisciess | éi72066245n | éos2sçoscs | nossitrarsm | mésasçoaso) | 906430451962

Fou | 1ssa0scou72 | 195202003199 | 187207075200 | ni+0500024 | 24sis@mon2 | 2741103220

IMTL | 114020819 12402(1013) 1420201217) 1320209139) 17202 (15-19) 1840301422)

mmTw | 07+020510) 02020600) 09201 (07-11) 07401 0510 09201710) 1020100912)

TL 304030539 39203674) 372036249 40203659 s02064159 s12044656)

RHL 14 34 14 34 54 24

with some reservations in M. spinifer (MNHN 1972.551-554). It contains specimens larger

than typical M. asper, with more distinctly expressed spines and tubercles (two tarsal spines

clearly recognizable) and a more distinct ventral contrast of white and dark coloration. These

character states remind those of M. spinifer (see below). However, the specimens are still

smaller than typical representatives of the species, and the ventral pattern contrast is less

expressed. We here tentatively follow their assignation to M. spinifer as suggested by

BLOMMERS-SCHLÔSSER & BLANC (1991), but it should be kept in mind that they are interme-

diate between M. spinifer and M. asper in size and expression of ventral pattern contrast. The

species is therefore known from (1) the Chaînes Anosyennes and (2) Ivohibe.

Natural history and advertisement call. - Unknown.

Mantidactylus ambohitra n. sp.

(fig. 3c-f)

Holotype. - ZSM 1084.2001 (originally ZFMK 57418), adult male, Montagne d’Ambre,

collected by Frank Glaw, Nirhy Rabibisoa and Olivier Ramilison on 14-17 March 1994.

Paratypes. —- MNHN 1893.244-245 (two females) from Montagne d’Ambre, MNHN

1893.246 (female), 1893.248 (male), 1893.249-250 (two females), 1893.252 (female), 1893.253

(juvenile), 1991.3148 (previously 1893.246A; female), all from Mararaomby (Montagne

d’Ambre), sent to the Paris museum by Alluaud and Belly in 1893; MNHN 1972.573 (female)

from Montagne d’Ambre; MNHN 1975.314 (female), 1975.322 (female), 1975.325 (female),

197 9 (female), 1975.330-331 (two males) from Les Roussettes (Montagne d'Ambre), all

without precise collecting data; MTKD 37424 (male) from Montagne d'Ambre; ZFMK

57419 (female) from Montagne d’Ambre, with same collecting data as holotype: ZFMK

Source : MNHN, Paris

VENCES & GLAW 21

62204-62205 (two subadults), collected by J. Steinbrecher on 26 November 1995 at Montagne

d’Ambre.

Diagnosis. — À member of the genus Mantidactylus based on the presence of femoral glands

and absence of nuptial pads in males. Assigned to the Mantidactylus asper group in the

subgenus Gephyromantis based on: (1) blackish paired subgular vocal sacs in males; (2) inner

and outer outer dorsolateral ridges; (3) (small) heel spines: and (4) general similarity with

other species of the group. The species is in several characters intermediate between M. asper

and M. spinifer on the one hand, and M. luteus, M. plicifer and M. sculpturatus on the other

hand. Itis distinguished from M. asper and M. spinifer by its largely uniformly colored ventral

side, a lower amount of dermal spines, less extended webbing and less prominent dorsolateral

ridges. It is distinguished from M. luteus, M. plicifer and M. sculpturatus by smaller body size

(male SVL 34-37 mm vs. 36-48 mm), less prominent inner dorsolateral ridges, small size of

heel spines and supraocular spines, presence of an outer metarsal tubercle and occasional

occurrence of interocular tubercles.

Etymology.— Derived from Ambohitra, the Malagasy name for the Amber Mountain (Mon-

tagne d’Ambre), the type locality of the species. The name is used as invariable noun in

apposition to the generic name.

Description of the holotype.-— Adult male, SVL 35.8 mm (fig. 3c-d). For measurements, see tab.

1. Body slender; head slightly longer than wide, distinctly wider than body; snout rounded in

dorsal and lateral views; nostrils directed laterally, slightly protuberant, much nearer to tip of

snout than to eye; canthus rostralis distinct, concave; loreal region concave; tympanum

distinct, elliptical (slightly higher than wide), 67 % of eye diameter; supratympanic fold very

distinct, straight; tongue ovoid, distinctly bifid posteriorly; vomerine teeth distinct, in two

rounded aggregations, positioned posterolateral to choanae; choanae rounded. Arms slender,

subarticular tubercles single; one outer, central, and inner metacarpal tubercles present;

fingers without webbing; relative length of fingers 1 < 2 < 4 < 3, second finger distinctly

shorter than fourth; finger disks distinctly enlarged; nuptial pads absent. Hindlimbs slender;

tibiotarsal articulation reaching widely beyond snout tip when hindlimb is adpressed along

body; lateral metatarsalia separated by webbing; inner metatarsal tubercle distinct, outer

metatarsal tubercle small but recognizable; webbing formula between toes 1(1), 2i(2), 2e(0.75),

3i(2), 3e(1), 4i(2), 4e(1.75), 5(0.25); relative toe length 1 <2 <3 <5<4, third toeclearly shorter

than fifth toe. Skin on the upper surface smooth; dorsolateral folds present but weakly

expressed and not very prominent; inner dorsolateral folds present, starting ca. 2.5 mm

behind the eyes and fading on the anterior back: outer dorsolateral folds running from ca. 4

mm posterior to the supratympanic fold to the inguinal region. Two distinct, longitudinal

interocular tubercles, of same color as surrounding skin; supraocular tubercles present:

Supraocular spines and heel spine present but small. Ventral skin smooth on throat, slightly

granular on posterior belly. Femoral glands very poorly delimited and very indistinct from

both external and internal views; a patch of indistinct granules visible from internal view.

Dorsal coloration in preservative dark brown with a weakly defined vertebral region of

lighter beige-brown color. Posterior fifth of the dorsum and dorsal surface of the hindlegs

light brown. One distinet and four indistinct dark brown crossbands on femur, one indistinet

and three distinct bands on tibia, and five indistinet bands on tarsus and foot. Dorsal color of

Source : MNHN, Paris

122 ALYTES 19 (2-4)

forelimbs irregularly dark and light brown. On the flanks, the dark dorsal color fading

gradually into the light ventral color. Head sides dark brown except an ill-defined light band

running from the anterior eye corner to the upper lip. Ventrally cream-white with some weak

symmetrical brown markings in the chest region. Throat cream-white, the inflatable lateral

parts of the vocal sac black. In life, color was similar to that in preservative, the iris being light

brown in its upper part, reddish brown laterally, and greyish brown in its lower part.

Variation. — The available specimens are largely in mediocre state of preservation, especially

the MNHN specimens that were collected more than 100 years ago. Nevertheless, a relatively

large variability of morphology (dorsal dermal structures) and coloration could be assessed.

ZFMK 57419 (fig. 3e) has a strongly contrasted dorsal pattern of dark dorsolateral stripes on

a light brown back. A further specimen (fig. 3f; not collected) had a light reddish brown head

surface with a triangular posterior end, a pattern also known in Mantidactylus luteus (GLAW

& VENCES, 1994: color plate 93) and other representatives of the genus. Another specimen (not

collected) had a broad and sharply delimited median light stripe on the dorsum. The inner and

outer dorsolateral ridges are always present but usually weakly expressed and often discontin-

uous. The inner dorsolateral ridges generally begin 2-3 mm behind the eyes. In some

specimens, one or two pairs of indistinct ridge-like interocular tubercles are seen, but usually

they are absent. Heel spine and supraocular spines are small, and often reduced to tubercles:

tarsal spines are absent. Males have rather indistinct femoral glands. Humeral protuberances

are not visible in the available material. Webbing in most specimens is rather developed,

reaching almost the fifth toe disk, although it is less extended in other specimens (formula

0.75-0). The outer metatarsal tubercle is always visible though generally small. The three

available juvenile paratypes already show some of the characteristics of M. ambohitra (e.g.,

not very prominent dorsolateral ridges, presence of ridge-like interocular tubercles).

Significant intersexual differences (Mann-Whitney U tests) were detected in relative

length of fore-(P < 0.05) and hindlimbs (P < 0.005) and in relative length (P <0.05) and height

(P < 0.005) of the inner metatarsal tubercle, but not in relative tympanum diameter or SVL.

Limbsare longer in females than in males, which is a rather uncommon state in Malagasy frogs:

the inner metatarsal tubercle is longer and higher in males. Male/female size ratio is 102 %.

Further material. - MNHN 1893.243 (Montagne d'Ambre) is a large male specimen which

reminds Mantidactylus plicifer by size (SVL 44.8 mm) and by its large, distinct femoral glands.

The probably subadult female MNHN 1975.324 (Les,Roussettes) reminds M. luteus rather

than M. ambohitra. As both specimens are in poor states of preservation, it is not possible to

make a definitive statement on their identity. It is possible, however, that more than one

representative of the M. asper group occurs at Montagne d’Ambre.

Three further specimens agree morphologically with M. ambohitra but are not included

in the type series due to dubious or lacking information on their collecting locality. The male

MNHN 1975.326 is labeled as originating from “Ifaty”rbut we consider it as very improbable

that this refers to Ifaty in extremely arid south-western Madagascar (close to Toliara). The

female MNHN 1975.332 has no locality information: Both these specimens bear numbers

subsequent to small s originating from the Montagne d'Ambre region. The female

MNHN 1973.896 was collected ording to the MNHN catalogue, by C. P. Blanc on 16 July

1972 at Marojejy (300 m elevation); however, its “hard” state of fixation is similar to that of

Source : MNHN, Paris

VENCES & GLAW 123

specimens from Montagne d’Ambre (e.g. MNHN 1975.332) rather than to the more flexible,

obviously formalin-fixed M. luteus specimens from Marojejy, collected also by C. P. Blanc in

July 1972 (MNHN 1973.897-902).

Distribution. - The species is so far reliably known only from (1) its type locality, Montagne

d’Ambre in far northern Madagascar.

Natural history. — We observed specimens during the day on the forest floor. J. Kühler (pers.

comm.) observed calling males in November in the evening (20-22 h and later) during light

rain, calling from bushes at the forest edge, ca. 60-150 cm above the ground. Inflated vocal sacs

were paired subgular (fig. 3f).

Advertisement call. —- Unknown.

Mantidactylus luteus Methuen & Hewitt, 1913

(fig. Sa-b)

Mantidactylus luteus Methuen & Hewitt, 1913. — Name-bearing type: holotype, by original designation,

TM 10077, adult male collected by Herschell-Chauvin. — Type locality: Folohy. — Other types:

paratypes, TM 10095-10101, two adult males, two adult females and three juveniles, and TM 10094

{exchanged with another collection according to TM catalogue)

Identity. — The type series, as examined by us in 1998, agrees morphologically with the

populations attributed to M. luteus and inhabiting the low-elevation areas of the Malagasy

east coast. This form is morphologically recognizable by: (1) rather small size; (2) distinct but

rather small femoral glands in males; and (3) generally uniformly whitish venter without

distinct dark chest markings. The holotype has a SVL of 41.3 mm, FGL of 6.4-6.8 mm and

FGW of 1.6-1.9 mm (W. Haacke, pers. comm. in 2001), thus within the range of other male

specimens herein assigned to the species (tab. 2). We will, in the following, provide a detailed

morphological description of M. luteus, and only mention differences from this pattern in the

subsequent sections on M. plicifer and M. sculpturatus.

Material examined. - MNHN 1975.323 (Kianjavato); MNHN 1972.1435 (Ambana-Soavala, Chaînes

Anosyennes): MNHN 1972.1418 (Camp V, Chaînes Anosyennes); MNHN 1973.897-902, 907, 908, 910

(Marojejy, alt. 300 m); MNHN 1973.906 (Marojeiy, alt. 1300 m); MNHN 1973.909 (Marojejy. alt. 600

m); TM 10077, 10095-10101 (holotype and paratypes: Folohy); ZFMK 47222, 47290-47291, 52711,

52713-52714 (Nosy Boraha); ZFMK 52715, 66674 (Nosy Mangabe).

Morphology. —- The description provided here refers to male specimens from Marojejy

(MNHN 1973.897, 1973.899-902, 1973.906-908), Nosy Mangabe (ZFMK 66674) and Nosy

Boraha (ZFMK 52714) which form a morphologically homogeneous group. Morphology

and attribution of other specimens is discussed in the section on distribution. Inner and outer

dorsolateral ridges present and usually continuous, the inner ridge being more prominent

than the outer ridge. Inner dorsolateral ridges generally beginning above the eye and running

onto the central dorsum, ending in various modes, either fading or curving medially or

laterally, sometimes continued as indistinet undulating structures towards the inguinal region.

Outer dorsolateral ridges generally beginning in the shoulder region where the inner ridges

end (fig. 1). No connecting dorsal ridge, no interocular tubercles. Supraocular tubercles and a

Source : MNHN, Paris

124 ALYTES 19 (2-4)

Fig. 5.- Photographs of Mantidactylus luteus, male from Nosy Boraha, in dorsolateral and ventral views

(a-b); M. plicifer, male from Ranomafana (ZFMK 62305) in dorsolateral and ventral views (c-d); and

M. sculpturatus, male from Andasibe (ZFMK 53688) in dorsolateral and ventral views (e-f).

few small supraocular spines present. Usually one distinct supraocular spine coinciding with

the origin of the inner dorsolateral ridge. A distinct heel spine; no tarsal spines. Males with

small but distinct femoral glands (see tab. 2); in MNHN 1973.899 (Marojejy), the gland on

one femur, in internal view, composed of 16 granules of 0.4-0.9 mm diameter each; in ZFMK

66674 (Nosy Mangabe), of 29 granules of 0.4-0.6 mm. Vocal sac paired subgular. A strongly

expressed humeral protuberance in males, lacking in females. Webbing in most specimens

reaching the disc of the fifth toe, resulting in webbing formulas of 5(0) or 5(0.25). One inner

and two rather weakly expressed, similarly-sized outer metacarpal tubercles, the latter two in

contact with each other. A distinct inner but no outer metatarsal tubercle.

Source : MNHN, Paris

VENCES & GLAW 125

Significant intersexual differences in the whole sample attributed to M. luteus were found

in SVL and in relative length (Mann-Whitney U test, P < 0.05) and height (P < 0.005) of the

inner metatarsal tubercle (longer and higher in males), but not in relative tympanum diameter

or relative length of limbs. Male/female size ratio is 89 %.

Coloration. — Dorsally generally uniformly greyish brown, sometimes with some shades and

indistinct markings of lighter or darker brown. Usually with at least a partial black bordering

of the inner dorsolateral ridges towards the flanks. Two further black spots often present at

the origin of the outer dorsolateral ridges. Indistinct but often well-delimited crossbands on

the hindlimbs (5-7 on femur, 3-7 on tibia, 4-10 on tarsus and foot). Ventral color uniformly

cream, including the limbs. Only some specimens with faint brownish markings on chest and

(in males) on throat. Lateral, inflatable parts of the (paired subgular) vocal sacs black.

Distribution. — Beside male specimens from Marojejy, Nosy Mangabe and Nosy Boraha, we

also attribute several vouchers to M. luteus. Two males from the Chaînes Anosyennes

(MNHN 1972.1418 and 1972.1435), which were collected syntopically with M. plicifer (see

below), differ from that species by their completely uniform chest, smaller size and smaller

femoral glands (consisting of 24 granules of 0.4-0.6 mm diameter in MNHN 1972.1435). The

male MNHN 1975.323 (Kianjavato) also agrees in having distinct femoral glands with 21

large granules. The females ZFMK 52711 (Nosy Boraha) and ZFMK 52715 (Nosy Mangabe)

are rather large (SVL 45.0 and 47.0 mm, respectively), and would therefore agree better with

M. plicifer by size. However, the almost uniform whitish venter of ZFMK 52711 and the very

faint chest mottling of ZFMK 52715 agree with the state in M. luteus. As so far no records of

M. plicifer exist for central-eastern or north-eastern Madagascar, we assign these female

specimens to M. luteus. We attribute also the juveniles MNHN 1972.909 and 910 (Marojejy)

to this species, as neither M. plicifer nor M. sculpturatus are known from this region. A further

locality is Foulpointe which is based on a single voucher (ZMA 6725) which agrees with M.

luteus in body size and femoral gland proportions according to a personal communication of

FE Andreone. Summarizing, specimens attributable to M. luteus are from: (1) Marojejy; (2)

Nosy Mangabe: (3) Nosy Boraha; (4) Foulpointe; (5) Folohy (type locality); (6) Kianjavato;

and (7) the Chaînes Anosyennes (including Ambana-Soavala and Camp V). All these

localities are at low elevations along the Malagasy east coast: only one specimen (MNHN

1973.906) is catalogued as originating from 1300 m on the Marojejy massif, while the other

specimens from this locality were collected at 300-600 m elevation.

BLOMMERS-SCHLÔSSER & BLANC (1991) additionally recorded the species from a number

of further localities, namely Les Roussettes (Montagne d’Ambre), Ambalamarovandana,

Marovitsika and Antsalova. The available Montagne d’Ambre specimens are here all referred

to the new species M. ambohitra, although at least two of these are actually morphologically

similar to M. luteus, and the possible occurrence of this or a similar species at Montagne

d'Ambre warrants further research (see section on M. ambohitra). The available material from

Ambalamarovandana may be attributable to M. sculpturatus or M. luteus (see below). The

locality Marovitsika forest (Pic Ivohibe) is based on juvenile specimens only. The locality

Antsalova (Antsingy) is corroborated by one MNHN voucher from this site catalogued as M.

luteus. This specimen (MNHN 1975.327), however, has no dorsolateral ridges and clearly

agrees in general morphology with Mantidactylus corvus and M. pseudoasper (subgenus

Phylacomantis), two morphologically similar species of which one (M. corvus) has been

Source : MNHN, Paris

126 ALYTES 19 (2-4)

Frequency (kHz)

se. RE '

0 200 400 600 800 1000 ms

Fig. 6. - Sonagram and oscillogram of part of a call (five notes of a note series) of Mantidactylus luteus

from Marojeiy.

described from western Madagascar (Isalo). Occurrence of M. cf. pseudoasper in the Antsingy

forest is also corroborated by MSNG 49087 which was the origin of the tissue samples used

for DNA analyses herein.

Natural history. Calling activity was mainly nocturnal but sometimes started before dusk. In

such cases, the calls were emitted from the ground. Calling activity increased with progress of

dusk, and at night specimens generally called from leaves 1-2 m above the ground. We never

observed calling specimens gathering at or close to water bodies.

Advertisement call. - Calls were serie:

of short, relatively melodious notes (fig. 6). Each note

did correspond to one expiration. Between notes, the vocal sac did not remain distinctly

inflated, although a slight inflation was noted when a call series was about to start. Calls from

Marojejy (Camp 1) were recorded on 27 March 1994, 17.30 h, at 22°C air temperature. They

5), duration of intervals

contained up to 21 notes. Note duration was 22-24 ms (23 + | ms,”

between notes was 170-180 ms (173 + 5 ms, n = 4). Note repetition rate was 5.0-5.5 ms. C.

from Nosy Boraha as described by GLAW & VENCES (1992) were similar and are here

re-described. They were recorded on 8 March 1991 at ca. 22°C air temperature. Note duration

was 44-59 ms (51 + 5 ms, x = 8), interval duration 127-236 s (156 + 40 ms, n = 7), and note

repetition rate 4.3

On 28 March 1994, distress calls were emitted by a specimen from Marojejy when

handled. Two calls had a duration of 1600 ms and 1663 ms. Frequency was 1650-10000 Hz,

5.0 per second.

Source : MNHN, Paris

VENCES & GLAW 127

respectively 0-10000 Hz, and showed the frequency band pattern typical for anuran distress

calls.

Mantidactylus plicifer (Boulenger, 1882)

(fig. 5c-d)

Rana plicifera Boulenger, 1882. — Name-bearing type: lectotype, by present designation, BMNH

1882.3.16.58, adult male collected by W. D. Cowan. — Type locality: “East Betsileo” according to the

original description - Paralectotypes: BMNH 1882.3.16.57, female and 1882.3.16.56 and 59, two

subadult specimens, with same collecting data as lectotype.

Comment. The current re-definition of the species M. plicifer is based on characters (size and

structure of femoral glands) recognizable in males only. As only one of the original syntypes

of Rana plicifera is a male, its present designation as lectotype is necessary to reach taxonomic

stability in a group of extremely similar species (M. luteus, M. plicifer, M. sculpturatus).

Identity. Our field observations demonstrated syntopic occurrence of a smaller and a larger

M. luteus-like species at Ranomafana, males of the larger species reaching 44.2-44.4 mm SVL.

Also in the MNHN sample from the Chaînes Anosyennes, a larger and a smaller form could

be distinguished, males of the larger form ranging from 43.6 to 48.3 mm SVL. One distinctive

character of the large form at both sites were its consistently large and distinct femoral glands.

Based on this character, we consider the large specimens as Mantidactylus plicifer: fig. 7 shows

that they agree with the lectotype of Rana plicifera in relative femoral gland size, although the

type is distinctly smaller (38.8 mm SVL: see tab. 1-2). The difference both in SVL and in

relative length and width of femoral glands (ratio FGL/SVL and FGW/SVL) was highly

significant between M. plicifer and M. luteus (P < 0.005).

Material examined. - BMNH 1882.3.16.56-59 (East Betsileo; lectotype and paralectotypes): MNHN

1972.1404 (Ambana-Soavala, Chaînes Anosyennes); MNHN 1972.1405, 1972.1407-1408, 1972.1410,

1972.1412-1414, 1972.1419-1422, 1972.1424-1425, 1972.1427, 1972.1429, 1972.1431 (Camp V, Chaînes

Anosyennes); MNHN 1972.1436-1437 (Camp IV, Chaînes Anosyennes); ZFMK 62305-62306 (Rano-

mäfana).

Morphology and diagnosis. — Beside the larger size of most specimens as compared to M.

luteus, the most important morphological character to identify M. plicifer is the large size of

its femoral glands (see /dentity section). In internal view, a gland of the lectotype contained

about 45 granules. In one specimen from Ranomafana (ZFMK 62306), a gland was composed

of 47 granules of 0.5-0.7 mm in diameter. In one specimen from the Chaînes Anosyennes, a

gland was composed of 55 granules of 0.4-0.8 mm in diameter.

Beside this character, the morphology of M. plicifer is extremely similar to M. luteus. The

differences mentioned by BLOMMERS-SCHLÔSSER & BLANC (1991) in their key to Mantidacty-

lus, namely webbing (less developed in M. plicifer) and heel spine (only a tubercle present in

M. plicifer) are not suited to distinguish both species. The webbing is variable in M. plicifer.

For instance, it reaches the disk of the fifth toe in ZFMK 62306, but only inbetween the disk

and the external subarticular tubercle in ZFMK 62305, resulting in webbing formulas for the

fifth toe of 5(0) vs. 5(0.5). And the presence of a heel spine appears to be highly dependent on

Source : MNHN, Paris

128 ALYTES 19 (2-4)

* M. luteus

> M. plicifer

e M. sculpturatus

0.05 0.07 0.09

Mantidactylus luteus, M. plicifer and M. sculpturatus. Note the clear distinction of Mantidactylus

plicifer from the other two species by its larger femoral gland size. Most M. plicifer can also be

distinguished by their larger SVL: the lectotype of M. plicifer (marked by an arrow), which in

small-sized, shares with the other specimens its large femoral glands.

the state of fixation and preservation of the specimens. So, a heel spine is clearly visible in the

formalin-fixed M. plicifer from Chaînes Anosyennes, but this structure is only rudimentary

and faintly recognizable in the ethanol-fixed specimens from Ranomafana. The outer meta-

tarsal tubercle, absent in M. luteus, is sometimes visible as small rudiment in formalin-fixed

M. plicifer. The pair of outer metacarpal tubercles are often indistinet in M. plicifer as in M.

luteus, but when recognizable, the outermost tubercle is distinctly smaller and more elongate.

Color and pattern variation is as follows. Many specimens with a tendency of a lighter

lateral coloration and a darker dorsal coloration, more or less sharply separated by the

dorsolateral ridges: this contrasted pattern very distinct in a few specimens (e.g. MNHN

1972.1427) which have light brown to beige flanks and dark brown central dorsum and head

su , both colors sharply separated by the dorsolateral ridges. In this specimen, head also

laterally beige, with a broad vertical band running from the eye to the upper lip. In some

specimens, head laterally with a broad light beige band along the upper lip, bordered sharply

Source : MNHN, Paris

VENCES & GLAW 129

by a dark brown band along the canthus rostralis. Venter cream whitish, with a few brownish

markings usually present in the chest region. Throat largely covered by faint brown pigment

in many males (except a longitudinal light median band). Vocal sacs blackish.

Intersexual differences in SVL and relative tympanum size are close to significance

(Mann-Whitney U test; P < 0.07), males being smaller and having relatively larger relative

tympanum sizes than females. Highly significant differences were found in relative length and

height of the inner metatarsal tubercle (P < 0.001; longer and higher in males), while relative

limb length did not differ significantly between sexes. Mean male/female size ratio is 96 %.

Distribution. — Beside the unprecise type locality East Betsileo, the species is reliably known

from (1) Ranomafana and (2) the Chaïnes Anosyennes. BLOMMERS-SCHLÔSSER & BLANC

(1991) listed six additional localities for the species: Tsaratanana, Marojejy, Ambohitantely,

Mandraka, Ambalamarovandana, Marovitsika. We could not locate any voucher for the

Ambohitantely locality in the MNHN or ZMA collections. The specimens from Tsaratanana,

Marojejy and Mandraka, determined as M. plicifer by R. Blommers-Schlôsser according to

the MNHN catalogue, are here all referred to M. asper (see also GLAW & VENCES, 1994, for the

Tsaratanana vouchers). The specimens from Marovitsika are juveniles that cannot be reliably

determined, while those from Ambalamarovandana are here assigned to M. sculpturatus in a

preliminary way (see below).

Natural history. — Calls were heard during dusk from the vegetation in rainforest. Calling

males were sitting ca. 50 cm above the ground. No water bodies were observed in the

surroundings.

Advertisement call. - Recordings were done on 2 March 1996, 18.15 h, at 23°C air temperature

in the Ranomafana National Park. Single notes as well as short series of up to five notes

(fig. 8) were emitted. Note duration was 121-148 ms (134 + 10 ms, 7 = 5), duration of intervals

between notes was 492-559 ms (535 + 31 ms, n = 4). Since intensity faded continuously at the

end of each note, measuring of note duration was difficult, and it would also be possible to

consider note durations as longer (and, correspondingly, interval durations as shorter).

Frequency was 1200-4000 Hz, dominant frequency 2700-3300 Hz.

Mantidactylus sculpturatus Ahl, 1929

(fig. 5e-f)

Mantidactylus sculpturatus Ab, 1929. - Name-bearing type: holotype, by monotypy, ZMB 30515.- Type

locality: “Nordwest-Madagascar” according 10 original description (probably erroneous). - Other

Lypes: none

Identity. — At mid-altitude localities of eastern Madagas

extremely similar to M. luteus by morphology. It is, however, distinguished by its unharmo-

nious (vs. melodious) advertisement calls, its indistinet femoral glands, and a relatively high

genetic divergence (see below). We here revalidate the name Mantidactylus sculpturatus

for this species based on the following rationale. (1) The type of M. sculpturatus is a female of

37.5 mm SVL: its morphology totally corresponds to species previously assigned to M. luteus,

, à species occurs which is

Source : MNHN, Paris

130 ALYTES 19 (2-4)

pi Frequency (kHz)

44 à

24 Li

D ———_—Î2————— 7 ————l

0 800 1600 2400 3200 4000 ms

Fig. 8. - Sonagram and oscillogram of part of a call (five notes of a note series) of Mantidactylus plicifer

from Ranomafana.

and its size agrees with the mid-altitude /ureus-like species but not with the larger species M.

Plicifer. (2) AuL (1929) described M. sculpturatus from a specimen that was collected by J. M.

Hildebrandt, and that was reported to originate from northwestern Madagascar. However, no

species close to M. luteus is so far known from this region, except an undescribed species from

the Manongarivo Reserve collected by C. J. Raxworthy (deposited in the London museum)

and by D. Rakotomalala (pers. comm.). These Manongarivo specimens, however, have

distinct interocular tubercles in males and females which are absent in the type of M.

sculpturatus. As With other anuran species collected by J. M. Hildebrandit in this region, it is

probable that some labeling errors occurred either by Hildebrandt itself or when the speci-

mens were entered in the Berlin Museum after Hildebrandt's death in Madagascar in 1881

(VENCES & GLAW, submitted). (3) Beside localities in northwestern Madagascar, Hildebrandt

collected at sites in central eastern Madagascar but apparently not at low altitudes along the

east coast (BEENTIE, 1998). It is therefore probable that the M. sculpturatus type was collected

at a mid-altitude eastern locality. We are aware that these arguments do not suffice to fully

clarify the identity of M. sculpturatus, but its present re-definition appears as most parsimo-

nious solution and avoids the description of the mid-altitude lureus-like form as a new species.

Note. — BLOMMERS-SCHLÔSSER & BLANC (1991) stated that the “type” of M. sculpturatus was

lost. In the meantime, the holotype has been rediscovered in the Berlin museum

Material examined. - ZFMK 53688-53689 (Andasibe); ZFMK 62304 (Ranomafana); ZMB 30515

(holotype).

Source : MNHN, Paris

VENCES & GLAW 131

Frequency (kHz)

4 us de SE

st & l

2 sr EST Re

ü à , PAS

0 200 400 600 800 1000 ms

Fig. 9. - Sonagram and oscillogram of a call (three notes of a note series) of Mantidactylus sculpturatus

from Ranomafana.

Morphology and diagnosis. — By size and general morphology, this species is similar to M.

luteus. Femoral glands are indistinct in most specimens: in ZFMK 62304 (Ranomafana), an

adult male collected while calling and therefore in sexually mature and active state, no femoral

glands at all were recognizable in external or internal view. The male ZFMK 53688 (Anda-

sibe) had 26 scattered granules visible in internal view of a gland. These granules, however,

were smaller than in M. luteus (0.3-0.4 mm in diameter).

In the available material, no recognizable outer metatarsal tubercle. Webbing reaching

the disk of the fifth toe or slightly below, resulting in a webbing formula of 5(0)-5(0.25). A

small heel spine, and two external metacarpal tubercles of approximately similar size. General

coloration as in M. luteus, the specimen ZFMK 62304 from Ranomafana having a broad

white streak along the upper lip bordered by a blackish streak running under the canthus

rostralis. This specimen additionally with a curved white streak running transversally above

the cloacal region. Venter uniformly light, with very faint brownish pigment on the chest in

some specimens.

Distribution. — The species is reliably known from: (1) Andasibe; (2) Ankeniheny (no voucher

specimens collected); and (3) Ranomafana.

Natural history. — Calling males were observed at night, not concentrated around water

bodies, 1-2 m high in the vegetation, in rainforest.

Advertisement call. — Cal

from Ankeniheny were recorded on 18 December 1994, 21.45 h, at

ca 22°C air temperature. They consisted of up to 22 unharmonious notes, but at the beginning

Source : MNHN, Paris

132 ALYTES 19 (2-4)

of calling activity single-note calls were common. Following several of these, note-series were

successively composed of an increasing number of notes. After such a call series, calling

activity was often interrupted for some minutes; one call series from Andasibe lasted 20 s and

showed the following pattern: 1-1-2-3-15 (numbers referring to notes per call). However,

multi-note calls were emitted spontaneously as well. Note duration at Ankeniheny was

154-180 ms (168 + 13 ms, n = 5), duration of intervals between notes was 126-163 ms (143 +

18 ms, n = 4). Note repetition rate was 3.0-3.3 per second. Calls from Andasibe (recorded on

12 January 1992 at 22°C air temperature) were similar, with a note duration of 90-120 ms, an

interval duration of 250-260 ms, and a note repetition rate of 2.8/s (GLAW & VENCES, 1994).

Calls from Ranomafana (fig. 9) were similar, too. Two analysed calls consisted of 15-17

notes and had a duration of 4463-5127 ms. Note repetition rate was 3.3-3.4/s. Note duration

was 150-196 ms (173 + 13 ms, = 10), inter-note interval duration was 122-158 ms (139 + 12

ms, 7 = 10). Frequency bands were recognizable between 1500 and 2000 Hz, 2800 and 4000

Hz, and 4800 and 5200 Hz. Dominant frequency was between 3200 and 3400 Hz.

FURTHER SPECIMENS OF UNCERTAIN ATTRIBUTION

While the specimens listed in the respective sections above could be attributed to the

recognized species with a certain reliability and were used to delimit distribution patterns

(fig. 10), the following juvenile specimens (and thus the localities Vondrozo, Ivohibe, Sanga

Sanga forest and Tolongoina) could be assigned to the complex of the three species M. luteus,

M. plicifer or M. sculpturatus, but a specific determination was not possible: MNHN 1930.414

and MNHN 1991.3149-3150 (previously 1930.414A-B) (Vondrozo); MNHN 1972.556 (Ivo-

hibe, Marovitsika forest); MNHN 1972.1434 (Ambana-Soavala); MNHN 1972.1406,

1972.1409, 1972.1411, 1972.1415-1417, 1972.1423, 1972.1426, 1972.1428, 1972.1430,

1972.1432-1433 (Camp V, Chaînes Anosyennes); MNHN 1972.1469 (Ambana-Camp IV,

Chaînes Anosyennes); MNHN 1975.328 (Sanga Sanga forest); ZFMK 47252-47253 (Tolon-

goina). The same applies to an adult female (MNHN 1930.413) from Fort Carnot, (SVL

42.1 mm) which is in poor state of preservation and with largely faded pattern.

A series from Ambalamarovandana (Andringitra, at 1530 m elevations MNHN

1972.595-599) consists of two juveniles, one subadult, and two adult females with immature

oocytes. By the size of the females (SVL 40.8 mm, MNHN 1972.595; 40.6 mm, MNHN

1972.596) and complete lack of dark pattern on throat and chest, they can be assigned to

either M. luteus or M. sculpturatus rather than to M. plicifer.

Three examined specimens probably belong to the Mantidactylus asper group but can-

not be determined further at present. (1) MNHN 1975.315 (Ambatofitoharanana) is a rather

large female (SVL 42.9 mm) reminding Mantidactylus asper and M. ambohitra. has very

weakly expressed, straight dorsolateral ridges and apparently no markings on the venter. The

locality Ambatofitoharanana is located close to Fianarantsoa, at a rather high altitude of ca.

1600 m. The specimen may represent a hitherto unknown species. (2) MNHN 1975.316

(Marojejy, no elevational information), an adult female (SVL 31.3 mm) with maturing

oocytes recognizable by dissection, has two distinct pairs of ridge-like interocular tuberc

but only rudiments of dorsolateral ridges; instead, the dorsum is covered with irregular small

Source : MNHN, Paris

VENCES & GLAW 133

44° 46° 48° 50° 447 46° 48° 50°

12 12 1240 12

14 14° 14 14°

16° 16° 16° 16°

18 18 18° 18°

20° 20° 20° 20°

| >

22° æ 22 22°

de @ Mambonina| Loue ge # M lutous ES

I M asper © M pliciter

M M spinier © M scupturatus

= — _ i =

44° 46° 48° 50° 44 46° 48° 50°

Fig. 10. - Distribution maps of species in the Mantidactylus asper group. Locality numbers refer to those

used in the text.

tubercles. The venter is uniformly light except two faint symmetrical brownish markings on

the chest. This individual may represent a new species of Mantidactylus (Gephyromantis), but

a description does not seem justified until further collections yield also male specimens. (3)

MNHN 1895.257 is a juvenile specimen in poor state of preservation that reminds M. asper.

Itis remarkable because according to the MNHN catalogue it was collected on the west coast

(“Côte Ouest”). However, we consider this locality information as dubious until the occur-

rence of the group at western localities is confirmed.

KEY TO SPECIES IN THE MANTIDACTYLUS ASPER GROUP

The following key allows identification of most specimens belonging to the Mantidacty-

lus asper group. However, in some taxa (e.g., M. luteus, M. plicifer. M. sculpturatus), a reliable

determination is only possible in adult males (after examination of femoral glands or

advertisement calls). Considering the important variability of characters such as dorsolateral

ridges within populations (e.g., of M. asper or M. spinifer), determinations based on single

specimens should be always considered with caution.

Source : MNHN, Paris

134 ALYTES 19 (2-4)

la. Moderately sized to large species (SVL 32-50 mm); ventrally uniformly cream-whitish,

sometimes with brown mottling on chest and throat; tarsal spines and connecting dorsal ridge

absent; interocular tubercles usually absent; outer metatarsal tubercle present or absent;

webbing generally reaching closer to the disk of the fifth toe than to the first subarticular

tubercle: webbing formula 5(0) to 5(0.5)...

1b. Moderately sized species (SVL 27-38 mm); ventrally usually with distinct dark rotin

pattern (laterally delimiting a central bright white area on chest and sometimes belly; throat in

males brown with a light median stripe); tarsal spines sometimes present; interocular tubercles

generally present; outer metatarsal tubercle present; webbing generally closer to the first

subarticular tubercle than to the disk of the fifth toe: webbing formula 5(0.5) to 5(0.75), rarely

5(0.25)

2a. Moderately sized species (SVL 32-39 mm); interocular tubercles sometimes present:

supraocular spines small or in the form of tubercles only; dorsolateral ridges relatively

irregular and indistinct; heel spine small and indistinct; outer metatarsal tubercle present;

only known from far northern Madagascar (Montagne d'Ambre).........................

2b. Larger species (SVL 36-50 mm); interocular tubercles always absent; supraocular spines

distinct, with one relatively large spine at the beginning of the inner dorsolateral ridge; inner

dorsolateral ridge always distinct; heel spine distinct; outer metatarsal tubercle absent or

rudimentary

3a. SVL 39-50 mm; femoral glands distinct, composed of 47-55 granules; calls composed of

a slow series of unharmonious notes … Mantidactylus plicifer

3b. SVL 36-47 mm; femoral glands small but distinct, composed of 16-29 granules; calls

composed of a rapid series of melodious notes; known from eastern lowlands .............

Mantidactylus luteus

3c. SVL 38-43 mm; femoral glands small and often indistinct or not recognizable, composed

of 26 granules (if visible); calls composed of a rapid series of unharmonious notes; known

from eastern mid-altitudes Mantidactylus sculpturatus

4a. Very granular dorsum with highly elevated and sharp ridges; connecting dorsal ridge often

present; tarsal spines present and distinct; very distinct sharp dark brown-white contrast in

the ventral pattern, extending onto the belly Mantidactylus spinifer

4b. Less granular dorsum; no connecting dorsal ridge; tarsal spines usually reduced to

tubercles or absent; sharp brown-white contrast in the ventral pattern usually restricted to

chest and/or throat 5.42 ee nue Mantidactylus asper

ANALYSIS OF DNA SEQUENCI

A chi-square test did not contradict homogencity of base frequencies across taxa (d/=

51; P > 0.9). The PTP test resulted in a significant difference (P = 0.01) between the most

parsimonious tree and trees generated from random permutations of the data matrix,

demonstrating presence of significant phylogenetic signal. Of the total of 556 included

characters, 358 were constant, 59 variable but parsimony-uninformative, and 139 variable and

Source : MNHN, Paris

VENCES & GLAW 135

DERe M. (Phylscomants granuiatus QT ———;;50

= M (Phylacomantis) sp. n. (Tsaratanana) @e*

#7 ---- M (Phylacomantis) comutus 7?

100 100/100 | so/gs

= M (Pnyiacomants)tschenti ? ——

= - M (Phylacomanté) tendroka

gel M (Phylacomantis) eucomacutates (Æ) ?

af —— + M (Laurentomants) af. malsgasius(Mascae) 2

_—__-- M. (Laurentomantis) malagasius (Masoats) (Æ) ?

400 — M: (Gephyromantis) scupturatus (Ranomatana) (Æ) ? ——— ;o5109

M. (Gephyromants) luteus (Nosy Mangabc) (Æ) ? ———| su8s

— - M. (Gephyromantis)pliciter(Ranomafana) (Æ) ?

eee M. (Gephyromantis) ambohitra sp. n. (Æ) ?

are À Canet air nat D |

M (Gephyromanis) boulengeri (N. Mangabe) (À)?

rés M (Phylacomantis) corvus (Isalo) @e —— 674

= M (Phyacomants) cf pseudoasper (Antsngy) @e— ——

drsestdesiunests M. (Guibemantis) iber @e

— Boophis xerophlus Qi

Fig. 11. - Results of a Neighbor-joining (NJ: left) and a Maximum Parsimony (MP: right) analysis of

556 bp of a fragment of the mitochondrial 168 rRNA gene in 16 species of Mantidactylus belonging

to the subgenera Gephyromantis, Laurentomantis and Phylacomantis. Species of the M. asper group

are printed in bold. The MP cladogram is a strict consensus of eight equally most parsimonious trees

(605 steps; consistency index 0.54, retention index 0.48). Numbers above branches are bootstrap

values in percent of Neighbor-joining (left) and Maximum Parsimony and Maximum Likelihood

(right) analyses, respectively (2000, 2000 and 100 replicates). Boophis xerophilus was used as the

outgroup. The symbol refer to presence of free-swimming or non feeding larvae (tadpole symbol) or

direct development within the egg (egg symbol). Symbols with question-marks refer to indirect

evidence from calling behaviour; species which call around water bodies (brooks) may have tadpoles,

while species calling independentiy from water probably have direct development.

parsimony-informative. MODELTEST proposed a Tamura-Nei substitution model (TrN + I

+ G) as best fitting the data, with a proportion of invariable sites of 0.4764, à gamma shape

distribution parameter of 0.9064, and empirical base frequencies (A: 0.3370; C: 0.2328;

G:0.1765; T:0.2536) and substitution rates (A-G: 5.3583; C-T: 8.5036; all other S:

Maximum Parsimony analysis resulted in eight equally most parsimonious trees (505 steps;

consistency index 0.54, retention index 0.48). A strict consensus of these is shown in fig. 11.

Most splits in the topology were unresolved, resulting in several polytomies. The following

clades were resolved: (1) a lineage containing the species of the Mantidactylus granulatus

group in the subgenus Phylacomantis (sensu GLAW & VENCES, 1994); (2) within the M.

Source : MNHN, Paris

136 ALYTES 19 (2-4)

granulatus group, one clade containing M. granulatus and an undescribed species from

Tsaratanana, and a second clade containing M. cornutus and M. tschenki; (3) a lineage

containing the two species of the subgenus Laurentomantis included; (4) a clade containing

M. luteus, M. sculpturatus and M. plicifer; (5) a clade containing M. sculpturatus and M.

luteus: (6) a clade with the two species of the Mantidactylus pseudoasper group in the subgenus

Phylacomantis, M. pseudoasper and M. corvus.

An identical topology was obtained by a second MP analysis after exclusion of all

characters with gaps in one or more taxa. The NJ analysis with LogDet distances (fig. 11) and

the ML analysis using the substitution model suggested by MODELTEST (not shown)

supported the same clades. Both in the NJ and ML trees, the M. pseudoasper group was

arranged basal to all other Gephyromantis, Laurentomantis and Phylacomantis species;

sequentially, the species of Gephyromantis split off the phylogram, while the two Laurento-

mantis Were placed as sister group of the Mantidactylus granulatus group.

Bootstrap analyses (FELSENSTEIN, 1985) with 2000 replicates provided rather high sup-

port for most of the clades defined above (1-6); the sole exceptions were clade 2 which was not

supported by values higher than 50 % in the NJ bootstrapping, and clade 3 which received low

support (51 %) in the MP bootstrapping.

Pairwise sequence divergences among species of the subgenera Gephyromantis, Phylaco-

mantis and Laurentomantis ranged from 3.8 to 16.4 %. Even the lowest values (3.8 % and

4.7 % between the species pairs M. cornutus — M. tschenki and M. luteus — M. sculpturatus)

were distinctly higher than values so far identified between conspecific Malagasy frog popu-

lations (up to ca. 1.5 % in M. granulatus; M. Vences, unpublished).

DISCUSSION

The distinctness of the three species M. luteus, M. plicifer and M. sculpturatus as

re-defined here is well corroborated by their bioacoustic and genetic differentiation. Previous

works (e.g., BLOMMERS-SCHLÔSSER & BLANC, 1991) also recognized two of these species (M.

luteus and M. plicifer), but their concept of M. plicifer was equivocal, and in large part based

on specimens attributable to M. asper. We observed syntopic occurrence of two species (M.

plicifer and M. sculpturatus) in the field at one site (Ranomafana), and preserved material

provided evidence for syntopic occurrence of M. plicifer and M. luteus at the Chaînes

Anosyennes. On the other hand, the morphological differences between these species are

extremely faint, and at present no reliable characters are known which could diagnose their

females or juveniles. This is in agreement with the situation in almost all groups of anurans in

Madagascar; the emerging picture indicates that all forms which differ bioacoustically have a

high genetic differentiation and must be regarded as valid species both under evolutionary

and biological species concepts. Among M. luteus, M. plicifer and M. sculpturatus, the lowest

genetic differences are found between M. luteus and M. sculpturatus, which were reliably

grouped as sister taxa by the phylogenetic analysis. The apparent allopatric distribution of

these species indicates an altitudinal segregation, M. sculpturatus inhabiting mid-elevations

and M. luteus low-elevations along the eastern coast. This example may indicate that sibling

Source : MNHN, Paris

VENCES & GLAW 137

species that segregate along elevational gradients in Madagascar are phylogenetically younger

than those which occur in close sympatry (as M. plicifer - M. luteus or M. plicifer - M.

sculpturatus).

The presence of a new species, Mantidactylus ambohitra, at Montagne d’Ambre in far

northern Madagascar has already been suspected by GLAW & VENCES (1994), who figured a

specimen from this locality as “Mantidactylus cf. asper”. The isolated Montagne d’Ambre

rainforest shares a rather large number of species with the eastern and north-eastern rainfor-

ests which, on the contrary, are not present in the north-western (Sambirano) region

(RAXWORTHY & NUSSBAUM, 1994; RASELIMANANA et al., 2000), but the taxonomy of the

Ambre populations has so far not been studied in detail. Our unpublished bioacoustic and

genetic data indicate substantial differentiation of the Ambre populations of at least three

other species shared with the east, namely Heterixalus betsileo, Mantidactylus curtus and

Mantidactylus liber. At least some of these populations possibly merit a status as separate

species, and may have originated by allopatric speciation in the isolated rainforest patch of the

Ambre massif.

As the occurrence of Mantidactylus luteus in Antsalova was based on a misidentification

of MNHN 1975.327 (actually belonging to M. pseudoasper or M. corvus), no reliable records

of any species of the M. asper group and of the whole subgenus Gephyromantis exist at present

from western Madagascar. Similarly, the record of M. granulatus from Antsingy (BLOMMERS-

SCHLÔSSER & BLANC, 1991) is in need of confirmation as we did not find any voucher specimen

in the MNHN and ZMA collections on which the distribution data of these authors were

largely based. EMANUELI & JESU (1995) did not record M. granulatus from this area. The only

records of the whole Gephyromantis-Laurentomantis-Phylacomantis assemblage occurring in

western Madagascar refer therefore to M. corvus and possibly M. pseudoasper This pattern

correlates with reproductive modes. M. corvus and M. pseudoasper have tadpoles developing

normally in brooks (though of derived morphology:; see GLAW & VENCES, 1994); Gephyro-

mantis have direct development (BLOMMERS-SCHLÔSSER, 1979; GLAW & VENCES, 1994); direct

development is also probable in some Laurentomantis (GLAW & VENCES, 1994); and the

species of the M. granulatus group may have either direct development (e.g. M. leucomacula-

tus) or non-feeding tadpoles (M. granulatus), although no detailed observations have so far

been published. The absence of species with direct development and similarly derived

reproductive modes from western Madagascar is probably a consequence of the need of

continuous humidity for the development of terrestrial nests. The seasonal environments in

western Madagascar possibly do not provide the climatic continuity for successful reproduc-

tion of direct-developing in anurans or for the survival of their diminutive juveniles. This may

also be the reason for the absence of cophyline microhylids of the genus Plethodontohyla (with

non-feeding tadpoles developing in terrestrial nests) from western Madagascar.

In the molecular analyses, the subgenus Gephyromantis (including the Mantidactylus

asper group) Was a paraphyletic assemblage along the branch leading towards the subgenera

Laurentomantis and partly Phylacomantis (the M. granulatus group), while another section of

Phylacomantis was placed basal to all these taxa (M. corvus and M. pseudoasper). Rela-

tionships of Laurentomantis to M. granulatus were also obvious from the cladograms of

RICHARDS et al. (2000). If this topology was confirmed by further molecular or morphological

data, it would indicate (beside the obvious diphyly of Phylacomantis) that the assemblage of

Source : MNHN, Paris

138 ALYTES 19 (2-4)

direct-developing Mantidactylus species evolved from brook-breeding ancestors such as M.

corvus and M. pseudoasper. If new field data confirm that M. granulatus has (possibly

non-feeding) tadpoles rather than true direct development, the phylogenetic topology would

furthermore suggest reversal of the reproductive mode (from direct development back to

tadpoles) in this species. Such a (multiple) reversal has also been suggested for Neotropical

hylid frogs of the genus Gastrotheca (e.g., SCANLAN et al., 1980; DEL PINo & EscoBAR, 1981;

DUELLMAN & Hiris, 1987; see also discussion in Dugois, 1987): in this genus, direct

development appears to be the ancestral state, and montane species from the Andes are

hypothesized to have re-acquired a free swimming tadpole stage through developmental arrest

in the course of their evolution. However, some Andean species of direct development may

have returned again to this reproductive mode (DUELLMAN & HizLis 1987), which would

indicate a high plasticity of this trait in this frog group. The testing of these hypotheses

appears highly attractive to address general questions of amphibian evolution, and may

qualify — beside Gastrotheca — the Gephyromantis-Laurentomantis-Phylacomantis lineage as a

suited model group for such studies.

ACKNOWLEDGEMENTS

We are grateful to F. Andreone, FE Maitioli, N. Rabibisoa, D. Rakotomalala, O. Ramilison,

E. Ranaivojaona, and J. Randrianirina for their help in the field. W. Bôhme (Bonn), B. Clarke and

C. MeCarthy (London), R. Günther (Berlin), B. van Tuijl (Amsterdam) and A. Dubois and A. Ohler

(Paris) allowed examination of specimens held in their care. W. Haacke (Pretoria) kindly provided

measurements of the M. luteus holotype. R. Blommers-Schlôsser made call recordings of M. asper

available, 1. Kôhler and D. Rakotomalala contributed their personal observations. We are indebted to the

Malagasy authorities for research and export permits. R. Jesu and G. Schimmenti (Genova), K. Schmidt

(München) and $. Wanke (Bonn) provided tissue samples for genetic analysis; J. Kosuch (Mainz)

contributed some of the sequences. The work in Madagascar was made possible by a cooperation accord

between the Zoologisches Forschungsinstitut und Museum A. Kocnig, Bonn, respectively the Zoologis-

che Staatssammlung München, with the Département de Biologie Animale, Université d'Antananarivo,

and financially supported by the Deutscher Akademischer Austauschdienst (DAAD).

LITERATURE

BEëNTE, H. J., 1998. —J. M. Hildebrandt (1847-1881): notes on his travels and plant collections. Kew Bull,

53: 835-856.

BLOMMERS-SCHLOSSER, R.M.A., 1979. - Biosystematies of the Malagasy frogs. I. Mantellinae (Ranidae).

Beaufortia, 352 (2): 1-77.

BLOMMERS-SCHLOSSER, R. M. A. & BLANC, C. P.. 1991. - Amphibiens (première partie). Faune de

Madagascar, 75 (1): 1-379.

De. Po, E. M. & EscoBar, B., 1981. - Embryonic stages of Gastrotheca riobambae (Fowler) during

maternal incubation and comparison of development with that of other egg-brooding hylid frogs

J. Morph., 167: 277-295.

Dumors, À., 1987. - Miscellanea taxinomica batrachologica (1. Alvres, 5 (1-2): 7-95.

. 1992. - Notes sur la classification des Ranidae (Amphibiens Anoures). Bull. mens. Soc. lin. Lyon, 61:

305-352.

Max, WE. & Hi

Ecuadorian Andes: r

gica, 43: 141-173.

D. M, 1987. - Marsupial frogs (Anura: Hylidae: Gastrotheca) of the

olution of taxonomie problems and phylogenetic relationships. Herperolo-

Source : MNHN, Paris

VENCES & GLAW 139

EMANUELI, L. & JESU, R., 1995. - The herpetofauna of the World Heritage Site “Tsingy de Bemaraha”

(Western Madagascar). In: G. A. LLORENTE, A. MONTORI, X. SANTOS & M. A. CARRETERO (ed.),

Scientia Herpetologica, Barcelona, Societas Europaea Herpetologica: 341-348.

FELSENSTEIN, J., 1985. - Confidence limits on phylogenies: an approach using the bootstrap. Evolution, 39:

783-791.

GLAw, FE. & VENCES, M., 1992. - Zur Kenntnis der Gattungen Boophis, Aglyptodactylus und Mantidac-

tylus (Amphibia: Anura) aus Madagaskar, mit Beschreibung einer neuen Art. Bonner zool. Beitr.,

43 (1): 45-77.

LE 1994. — 4 ficldguide to the amphibians and reptiles of Madagascar. 2nd edition. Kôln, Vences & Glaw

Verlag: 1-480, 48 pl.

De 2000. - Current counts of species diversity and endemism of Malagasy amphibians and reptiles. /n:

W.R. LOURENÇO & S. M. GooDMAN (ed.), Diversité et endémisme de Madagascar, Mémoires de la

Société de Biogéographie, Paris: 243-248.

GLAW, F, VENCES, M. & GOssManN, V., 2000. - A new species of Mantidactylus (subgenus Guibemantis)

from Madagascar, with a comparative survey of internal femoral gland structure in the genus

(Amphibia: Ranidae: Mantellinae). J nat. Hist., 34: 1135-1154.

LOCKHART, P.J., STEEL, M. A., HENDY, M. D. & PENNY, D., 1994. - Recovering evolutionary trees under

a more realistic model of sequence evolution. Mol. Biol. Evol., 11: 605-612.

METHUEN, P. A. & HEWITT, J., 1913. — On a collection of Batrachia from Madagascar made during the

year 1911. Annals Transvaal Museum, (4), 2: 49-64.

PosaDA, D. & CRANDALL, K.A., 1998. - MODELTEST: testing the model of DNA substitution.

Bioinformatie 817-818.

RASELIMANANA, A. P., RAXWORTHY, C. J. & NUSSBAUM, R. A., 2000. — Herpetofaunal species diversity

and elevational distribution within the Parc National de Marojejy, Madagascar. . M. Goop-

MAN (ed.), À floral and faunal inventory of the Parc National de Marojejy: with reference to

elevational variation, Fieldiana: Zool., 92: 157-174.

RAXWORTHY, C. J. & NUSSBAUM, R. A., 1994. — A rainforest survey of amphibians, reptiles and small

mammals at Montagne d’Ambre, Madagascar. Biol. Conservation, 69: 65-73.

RICHARDS, C. M., NUSSBAUM, R. A. & RAxWORTHY, C. J., 2000. — Phylogenetic relationships within the

Madagascan boophids and mantellids as elucidated by mitochondrial ribosomal genes. African J.

Herp., 49: 23-32.

SCANLAN, B. E., MAxsON, L. R. & DUELLMAN, W. E., 1980. - Albumin evolution in marsupial frogs

(Hylidae: Gastrotheca). Evolution, 34: 222-22

Sworrorp, D. L., 2001.-PAUP*. Phylogenetic Analysis Using Parsimony (* and other methods). Version

4. Sunderland, Massachusets, Sinauer Associates.

VENCEs, M. & GLAW, F, submitted. - Revision of the subgenus Chonomantis (Anura: Mantellidae:

Mantidactylus) from Madagascar, with description of two new species. J nat. Hist.

VENCES, M., KOSUCH, J., LÔTTE WIDMER, A., JUNGI K.-H., KÔHLER, J. & VerrH, M., 2000. —

Phylogeny and classification of poison frogs (Amphibia: Dendrobatidae), based on mitochondrial

168 and 12S ribosomal RNA gene sequences. Mol. Phyl. Evol., 15: 34-40.

Corresponding editor: Alain DURois

CA 2001

Source : MNHN, Paris

140 ALYTES 19 (2-4)

Deutsche Gesellschaft für

Herpetologie und Terrarienkunde e.V. - Bundesverband

The name “Deutsche Gesellschaft für Herpetologie und Terrarienkunde (DGHT)” means “German Society for Herpetology

and Amphibian and Reptile Husbandry”. It was founded in 1964 as a successor to Salamander, founded in 1918.

Our members work with amphibians and reptiles:

- in science and research,

in husbandry, and

- on conservation of nature and species.

Meanwhile, the DGHT has more than 8000 members in more than 30 countries. Most of our members are from German

speaking countries including Austria and Switzerland. Thus, it is the largest society in its field worldwide. It unites

knowledge and competence in all disciplines of herpetology.

plies, the DGHT unites scientists and non-professionals in herpetology and husbandry

This creates the particular strength of our society.

The DGHT offers a common platform for information, communication, and publication for both, the scientist working with

amphibians and reptiles as well as the layman who keeps and breeds amphibians and reptiles with enthusiasm.

one society.

Herpetologie:

Members of our society

= work in research and education at museums, z00s, universaties, and schools. They comprise specialized authorities on

systematics, biology, or ecology of amphibians and reptiles both in the laboratory and in the field

= work in Germany and abroad on conservation projects saving populations, evaluate causes of potential threats forthe

protection of amphibians and reptiles and their habitats. Due to these activitiesit has been approved according 10 $ 29 of the

Federal Law for the Protection of Nature conservation organisation. Dues and donations 10 it are tax deductible.

+ cooperate as experts and advisors in matters of trade in protected species with the respective federal and regional

authorities.

Amphibian and Reptile Husbandry

Our members

- keep and breed with enthusiasm amphibians and reptiles as an interesting hobby.

- contribute by observation, research, and publication to a better understanding of the ethology and biology of amphibians

and reptiles.

+ participate - privately or in 200 - in speci

reserves.

establish together with the respective authorities guidelines for biologically sound husbandry of amphibians and reptiles.

programmes 10 breed endangered species, thus helping tosecure genetic

Our periodicals and scientific publications:

Unlike most other societies, the DGHT offers its members several types of publication:

“SALAMANDRA" is publishing quarterly original papers on amphibians and reptiles. Its international reputation is

outstanding.

“MERTENSIELLA"

herpetological topics.

“ELAPHE N.F." is a quartely publication to inform our members on activities of interest like bre s, protection

measures, programs of regional and specialised working groups etc. A special part is reserved for papers with practical

information on husbandry.

is a scientific supplement 10 “Salamandra”. It publishes monographically selected papers on certain

Dedicated publications include

“RADIATA" - Zeitschrift der DGHT-AG Schildkrôten

“RADIATA" - E Edition

“DIE EIDECHSE" - Zeitschrift der DGHT-AG Lacertiden

“IGUANA” - Zeitschrift der DGHT-AG Leguane

“AMPHIBIA" - Zeitschrift der DGHT-AG Anuren / DGHT-AG Urodela

Apart from “Mertensiella” which can be puchased at special prices, all other publications are included in the annual

membership fee. Most of our publications are in English!

Sponsoring outstanding work in herpetology:

‘The “DGHT-FONDS” sponsors brecding, research, or conservation projects.

If you are interested to get more information on the DGHT, please contact:

DGHT-Geschäfisstelle - Box 1421, D-53351 Rheinbach- http://www.dght.de

msn —

Source : MNHN, Paris

Alytes, 2001, 19 (2-4): 141-153. 141

Une nouvelle espèce de Scutiger du Bhutan,

et quelques remarques

sur la classification subgénérique

du genre Scutiger (Megophrvyidae,

Leptobrachiinae)

Magali DELORME & Alain DUBOIS

Laboratoire des Reptiles et Amphibiens,

Muséum national d'Histoire naturelle,

25 rue Cuvier, 75005 Paris, France

A new species of the genus Scutiger from Bhutan is described on the

basis of two specimens without precise locality. The new species is placed in

the Scutiger glandulatus group. Within this group, it differs from Scutiger

jiulongensis by the large size of the spines on its breast, of Scutiger

glandulatus by its important webbing and by its flattened warts on the

back, and from Scutiger tuberculatus, which seems to be its closest known

relative, by its smaller size and the absence of spines under the arms. À

tentative phylogenetic analysis of a representative set of 17 species of

Megophrvidae, based on 54 external morphological characters, suggests

that, in the genus Scutiger, the subgenus Scutiger is paraphyletic relative

to the subgenus Aelurophryne. As these data are only preliminary, we

suggest to keep using these two subgenera for the time being. We provide

new diagnoses for these two subgenera, with lists of included species-

groups and species.

INTRODUCTION

Le genre Scutiger Theobald, 1868 (Megophryidae, Leptobrachiinae) comporte une

quinzaine d'espèces connues (voir GLAW et al., 1998, et Fer, 1999) de crapauds torrenticoles,

de moyenne et haute altitude, de l'Himalaya, du plateau tibétain et des montagnes de l’ouest

et du centre de la Chine. Deux sous-genres peuvent être reconnus au sein de ce genre (voir p.

ex.: FEI et al., 1991; YE et al., 1993; Dusois & OHLER, 1998; Fer, 1999): Scutiger Theobald,

1868 et Aelurophryne Boulenger, 1919. Etant données leur distribution et leur mode de vie, les

et il est probable qu'un certain nombre en

espèces de ce groupe sont encore mal connu

restent à découvrir. C'est ce qu'illustre la présente description d'une nouvelle espèce de ce

genre provenant du Bhutan, petit royaume est-himalayen où jusqu’à présent fort peu de

récoltes et d'observations batrachologiques ont été effectuées. Le genre Seuriger n'y avait pas

Source : MNHN, Paris

142 ALYTES 19 (2-4)

encore été signalé, mais la présence de Scutiger sikimmensis (Blyth, 1854) dans le sud-est du

Sikkim, près de la frontière du Bhutan (Dugois, 1987a: 20) permettait de prévoir la présence

de cette espèce dans ce pays. Toutefois, les deux seuls spécimens de ce genre qui figurent dans

la collection d’Amphibiens effectuée en 1972 au Bhutan par une équipe du Muséum de Bâle

(voir BARONI URBANI et al., 1973) s'avèrent appartenir à une espèce distincte, et dont

l’inhabituelle combinaison de caractères suscite une réflexion sur la taxinomie subgénérique

actuelle de ce genre. Malheureusement, aucune information précise sur la localité de capture

de ces deux spécimens n’a été enregistrée dans le catalogue du Muséum de Bâle où ces

spécimens ont été déposés (Rainer Günther, lettre à AD du 2 août 1988), si bien que la

localité-type de l'espèce doit rester vague, au même titre que dans certains travaux des

zoologistes des siècles passés! Toutefois cette localité se trouve nécessairement sur l'itinéraire

de l'expédition de 1972 du Muséum de Bâle au Bhutan (voir BARONI URBANI et al., 1973):

parcourir soigneusement cet itinéraire pourrait être une manière de retrouver cette espèce

dans la nature. Après avoir repoussé pendant longtemps la description de cette nouvelle

espèce dans l'espoir, jusqu’à présent resté vain, que des spécimens provenant de localités

précises en seraient récoltés, il nous paraît utile de signaler l'existence de cette espèce

inhabituelle à la communauté scientifique, ce qui suscitera peut-être de nouvelles recherches

batrachologiques dans ce pays mal connu à cet égard, et la redécouverte de cet animal.

MÉTHODES ET ABRÉVIATIONS

Afin de faciliter les comparaisons, le canevas et la numérotation des caractères de la

description qui suit sont les mêmes que ceux utilisés dans un récent article portant sur la même

famille des Megophryidae (DUBois & OHLER, 1998), dont nous avons également repris les

abréviations désignant les mesures effectuées sur les spécimens: EL, longueur de l’œil; EN,

distance entre le coin antérieur de l’œil et la narine; FL, longueur du fémur (entre l’anus et le

genou); FLL, longueur de la patte antérieure (entre le coude et la base du tubercule palmaire

externe); FOL, longueur du pied (entre la base du tubercule métatarsien interne et l'extrémité

du quatrième orteil); HAL, longueur de la main (entre la base du tubercule palmaire externe

et l'extrémité du troisième doigt); FTL, longueur du quatrième orteil; HL, longueur de la tête

(entre le coin postérieur de la mâchoire et l'extrémité du museau); HW, largeur de la tête; IBE,

distance entre les coins postérieurs des yeux; IFE, distance entre les coins antérieurs des yeux;

IMT, longueur du tubercule métatarsien interne; IN, distance entre les narines: ITL, longueur

de l’orteil interne; TUE, distance minimale entre les paupières; MBE, distance entre le coin

postérieur de la mâchoire et le coin postérieur de l'œil; MFE, distance entre le coin postérieur

de la mâchoire et le coin antérieur de l'œil; MN, distance entre le coin postérieur de la

mâchoire et la narine; MTFF, distance entre l’incurvation maximale de la palmure entre le

troisième et le quatrième orteil et l'extrémité du quatrième doigt; NS, distance entre la narine

et l'extrémité du museau; SL, distance entre le coin antérieur de l'œil et l'extrémité du museau;

SVL, longueur entre l'extrémité du museau et l'anus: TFL, longueur du troisième doigt:

TFOL, longueur du tibia et du tarse (à partir de la base du tarse); TL, longueur de la jambe:

TW, largeur de la jambe; UEW, largeur maximale de la paupière. Les mesures ont été prises

par MD, soit au pied à coulisse avec une précision de 0,1 mm, soit, pour les mesures inférieures

à 6 mm, au micromètre oculaire avec une précision de 0,1 mm.

Source : MNHN, Paris

DELORME & DuBois 143

L'analyse cladistique a été effectuée avec le programme PAUP (SWoFFORD, 1993) en

prenant les spécimens Scaphiopus couchii et Pelodytes punctatus comme outgroups.

Les spécimens étudiés dans ce travail sont déposés dans les collections de trois muséums:

FMNH, Field Museum of Natural History, Chicago, USA; MNHN, Muséum National

d'Histoire Naturelle, Paris, France; NHMB, Naturhistorisches Museum Basel, Bâle, Suisse.

DESCRIPTION DE LA NOUVELLE ESPÈCE

Scutiger (Aelurophryne) bhutanensis sp. nov.

Holotype. - NHMB 17551, mâle adulte (SVL = 53,0 mm).

Paratype. - NHMB 17550, mâle adulte (SVL = 53,4 mm).

Localité-type. — Bhutan. Les deux spécimens ont été récoltés, semble-t-il dans la même

localité, en 1972 le long de l'itinéraire de l'expédition zoologique au Bhutan du Muséum de

Bâle (voir Zntroduction ci-dessus, et BARONI URBANI et al., 1973).

Description de l'holotype. - (A) Taille et aspect général (fig. 1). - (1) Taille moyenne (SVL

53,0 mm), aspect ramassé.

(B) Tête (fig. 2). - (2) Tête plutôt carrée (HW 18,9 mm; HL 18,1 mm; MN 14,7 mm; MFE

12,8 mm; MBE 6,5 mm), sans coossification. (3) Museau arrondi, non protubérant, plus long

(SL 12,8 mm) que le diamètre de l’œil (EL 6,5 mm). (4) Canthus rostralis peu distinct, région

loréale légèrement concave. (5) Espace interorbital plat, plus petit ([UE 3,7 mm) que la largeur

de la paupière (UEW 6,5 mm) et la distance internasale (IN 4,6 mm), distance entre les bords

antérieurs des yeux (IFE 7,5 mm) environ la moitié de celle entre les bords postérieurs des

yeux (IBE 14,5 mm). (6) Narines plus proches des yeux (EN 2,7 mm) que de l'extrémité du

museau (NS 3,6 mm). (7) Forme de la pupille indistincte. (8) Tympan absent. (9) Ocelle pinéal

absent. (10) Dents vomériennes absentes. (11) Dents maxillaires absentes. (12) Langue étroite,

libérée sur plus d’un tiers, avec une encoche postérieure très réduite. (13) Repli supratympa-

nique différencié en une glande parotoïde peu épaisse.

que la main (HAL 12,8 mm). (15) Doigts courts (TFL 6,8 mm). (16) Longueur relative des

doigts, par ordre croi Let II subégaux < IV < IL. (17) Extrémités des doigts arrondies,

non élargies, sans ventouses, et de couleur claire. (18) Doigts sans franges dermiques; palmure

absente. (19) Tubercules subarticulaires absents. (20) Tubercule palmaire interne arrondi, ne

débordant pas sur le métacarpe; tubercule palmaire externe de taille équivalente; pas de

tubercule palmaire médian, de tubercules surnuméraires ni de crêtes glandulaires sous les

doigts.

(D) Pattes postérieures (fig. 4). — (21) Jambe (TL 18,9 mm; TW 6,53 mm) à peine plus

courte que la cuisse (FL 20,2 mm) et que la distance entre la base du tubercule métatarsien

interne et l'extrémité de l'orteil IV (FOL 20,4 mm). (22) Orteils courts; orteil IV (FTL

10,5 mm) mesurant environ la moitié de la longueur totale du tibia et du tarse (TFOL

20,3 mm). (23) Longueur relative des orteils, par ordre croissant: | < II < IL et V subégaux <

Source : MNHN, Paris

Lai

(p-0) 61 SALATV

Fig. 1. Scutiger bhutanensis, holotype, NHMB 17551, mâle adulte, En haut à gauche, vue dorsale: en haut à droite, vue ventrale;

en bas, vue latérale.

Source : MNHN, Paris

DELORME & DuBoIs 145

à :

1: »°5%,

Fig. 2. - Scutiger bhutanensis, holotype, NHMB 17551, mâle adulte. Profil du museau.

Fig. 3. - Scutiger bhutanensis, holotype, NHMB 17551, mâle adulte. Main droite, vue de dessous.

Source : MNHN, Paris

146 ALYTES 19 (2-4)

Fig. 4. — Scutiger bhutanensis, holotype, NHMB 17551, mâle adulte. Pied gauche, vue de dessous.

Source : MNHN, Paris

DELORME & DUBOIS 147

IV. (24) Extrémités des orteils arrondies, non élargies et sans ventouses. (25) Palmure absente.

(26) Pas de bordure dermique le long des orteils. (27) Tubercules subarticulaires indistincts.

(28) Tubercule métatarsien interne distinct, plus petit (IMT 4,0 mm) que la longueur du

premier orteil (ITL 3,5 mm). (29) Pli tarsien très marqué. (30) Pas de tubercule métatarsien

externe ni de tubercules surnuméraires.

(Ë) Peau. — (31) Museau et zone entre les yeux à peine verruqueuse. (32) Absence de

réseau sur le dos; verrues de toutes tailles couvrant le dos, petites et rondes sur les membres.

(33) Pas de glandes sur le ventre, ni de lignes glandulaires latéro-ventrales. (34) Pas de

macroglandes visibles extérieurement à la base des bras et sur les cuisses.

(F) Coloration en alcool. — (35) Taches sur le dos distribuées de façon aléatoire; trait

sombre sur la tête reliant les yeux; tache importante entre les épaules. (36) Taches plutôt que

bandes transversales sur les membres. (37) Anus plus foncé sans délimitation nette de

coloration.

(G) Caractères sexuels secondaires mâles. — (38) Glandes axillaires, couvertes d’épines

noires, situées sur les côtés de la poitrine; glandes pectorales de taille similaire, couvertes

d’épines noires. (39) Coussinets nuptiaux, avec des épines épaisses, noires et très kératinisées,

du bord du métacarpe jusqu’à la première articulation sur le doigt I (18 épines à droite, 17 à

gauche) et tout le long du bord interne du doigt II (17 épines à droite, 16 à gauche); pas

d’épines sur le doigt LIT. (40) Pas d'ouvertures de sacs vocaux sur les côtés du plancher buccal.

(41) Épines noires, bien kératinisées, accompagnées d’aspérités, sur plusieurs parties du corps:

lèvre supérieure et région loréale, très concentrées près des commissures de la lèvre inférieure,

et sur la totalité du dos, cloaque inclus: épines plus fines sur les jambes et les pieds, absentes sur

les bras.

Variation. — Les deux spécimens disponibles sont très semblables. Toutefois, les glandes

porteuses d’épines sur l’ensemble du corps (dos, tempes, museau) sont moins développées

chez le paratype. Chez celui-ci, les coussinets nuptiaux sont aussi moins développés, sur le

doigt I (15 épines à droite, 12 à gauche) comme sur le doigt II (14 épines à droite, 12 à gauche).

Etymologie du nom spécifique. - De Bhutan, nom du pays où cette espèce a été découverte.

DISCUSSION

Les deux spécimens mâles étudiés possèdent des glandes pectorales couvertes d’épines. A

haute altitude dans l'Himalaya et l'Asie centrale, parmi les Amphibiens Anoures seuls les

Ranidae du genre Paa (tribu Paini; voir DuBois, 1992) et les Megophryidae des genres

Oreolalax et Scutiger possèdent ce caractère (DuBois & OuLER, 1998; FE1, 1999). Une esquisse

de phylogénie des Megophryidae, s'appuyant sur 54 caractères de morphologie externe

(DELORME & DuBoIs, en préparation), suggère que ces deux derniers genres constituent un

clade distinct au sein de cette famille, pour lequel nous proposons d'employer le nom tribal

Oreolalagini Tian & Hu, 1985 (voir Dupois, 1987h). L'espèce nouvelle appartient sans

conteste à ce dernier taxon, comme en témoignent par exemple ses glandes axillaires, ainsi que

l'ensemble de sa morphologie (voir Duois, 2001). Pour déterminer à quel groupe au sein des

Source : MNHN, Paris

148 ALYTES 19 (2-4)

genres Oreolalax et Scutiger l'espèce nouvelle doit être attribuée, celle-ci a été comparée à

divers spécimens de ces deux genres (voir app. 1).

Huit caractères de morphologie externe différents dans les deux genres excluent l’espèce

nouvelle du genre Oreolalax: (1) étalement de la callosité nuptiale du doigt I jusqu’à la

première articulation; (2) écart important entre les talons quand les cuisses sont disposées à

angle droit par rapport au corps; (3) glandes axillaires présentes sur les côtés de la poitrine;

(4) absence d’anneau tympanique; (5) replis supratympaniques en forme de glandes paro-

toïdes; (6) petites épines noires, très kératinisées, sur différentes parties du corps (sauf sur les

coussinets nuptiaux) chez les mâles adultes; (7) absence de dents maxillaires; (8) absence de

glandes fémorales visibles.

Selon Fer (1999: 339-341), les deux sous-genres du genre Scutiger se distinguent princi-

palement par leurs types de caractères sexuels secondaires mâles: tandis que chez les espèces

du sous-genre Scutiger des callosités nuptiales sont présentes sur les trois premiers doigts,

dans le sous-genre Aelurophryne celles-ci ne sont présentes que sur les deux premiers doigts.

Pour ce caractère, Scutiger bhutanensis correspond donc à ce dernier sous-genre. FEI (1999:

340-341) reconnaît trois groupes d'espèces dans celui-ci. Selon cet auteur, le groupe de

Scutiger gongshanensis contient la seule espèce chinoise de ce sous-genre dont les mâles

possèdent des sacs vocaux; il faut y ajouter Scutiger adungensis du Myanmar, dont les mâles

ont de tels sacs (DuBois, 1979: 637). Le groupe de Scutiger mammatus se caractérise par une

palmure grande ou réduite, une absence de sacs vocaux et l'absence d’épines sous les bras,

comme chez l’espèce nouvelle; mais les espèces de ce groupe ne possèdent qu’une paire de

plaques d’épines sur la poitrine au lieu de deux comme chez S. bhutanensis. Enfin le groupe de

Scutiger glandulatus réunit des espèces avec ou sans palmure, sans sacs vocaux et avec deux

paires de plaques d’épines sur la poitrine, mais possédant des épines sous les bras. Toutefois ce

dernier caractère peut être lié à l’état physiologique des mâles, n'étant pleinement développé

que lors de la saison reproductive (Liu, 1936), et il n’est pas certain que les deux spécimens ici

étudiés étaient au summum de la période reproductive, avec des caractères sexuels mâles

pleinement développés. Toutefois, la présence d’épines bien noires sur la poitrine et les deux

premiers doigts des deux spécimens ici décrits suggère qu'ils n’en étaient pas éloignés, et que

les mâles de cette espèce ne présentent jamais d’épines nuptiales sur le troisième doigt et sous

les bras. Nous proposons donc de rapporter provisoirement cette espèce au groupe de Scutiger

glandulatus. Au de celui-ci, S. bhutanensis se distingue de Scutiger jiulongensis par la

grande taille des épines sur sa poitrine, et de Scutiger glandulatus par sa palmure étendue et ses

verrues aplaties sur le dos. Scutiger tuberculatus semble l'espèce la plus proche de S. bhuta-

nensis, dont elle se distingue notamment par la présence d’épines sous ses bras et sa grande

taille (SVL des mâles adultes de 68,0 à 76,0 mm, moyenne 72 mm; FEI, communication

personnelle).

L'esquisse de phylogénie des Megophryidae évoquée ci-dessus suggère un paraphylé-

tisme du sous-genre Scutiger par rapport au sous-genre Aelurophryne (fig. 5). Dans le

sous-genre Scutiger tel que défini actuellement, plusieurs états de caractères sont inconstants,

comme la présence d'épines sur le dos ou sur les bras, l'élargissement des bras des mâles par

rapport à ceux des femelles et la position des glandes pectorales. Dans le passé, d’autres

auteurs ont fait état d’une gradation dans la présence ou non de bourgeons de dents

maxillaires au sein de ce groupe (MYERS & LEVITON, 1962: INGER, 1966; ANONYME, 1977;

Source : MNHN, Paris

DELORME & DUBOIS 149

Dusois, 1980, 1987a), ou ont retrouvé un paraphylétisme de celui-ci sur la base de caractères

ostéologiques (FU et al., 1997; Fu & MurPrxy, 1997). Le sous-genre Aelurophryne constitue-

rait un groupe mieux défini par la présence des callosités nuptiales sur le doigt I et II

uniquement. Toutefois ces données ne sont encore que préliminaires et les relations phylogé-

nétiques au sein des Megophryidae ne peuvent être tenues pour fermement établies. Pour

l'instant, il nous paraît préférable de conserver les deux sous-genres, après les avoir redéfinis

de manière qu'ils restent homogènes quant à la forme et la distribution des callosités nuptiales

des mâles reproducteurs, caractère particulièrement frappant dans ce groupe. Pour ces

raisons, nous proposons de nouvelles diagnoses pour les deux sous-genres de Scutiger.

La même analyse (fig. 5) suggère également un paraphylétisme du groupe de Scutiger

glandulatus par rapport au groupe de Scutiger mammatus, un résultat qui devra être confirmé

(ou infirmé) par une analyse phylogénétique de l’ensemble des espèces de Scutiger. Pour

l'instant nous conservons ces deux groupes d'espèces tels qu’ils ont été définis et diagnosés par

Fe (1999).

Sous-genre Scutiger Theobald, 1868

Espèce-type. — Bombinator sikimmensis Blyth, 1854, par monotypie.

Diagnose. — Ce sous-genre se distingue du sous-genre Aelurophryne par la combinaison

suivante de caractères: (1) absence de dents maxillaires ou présence de bourgeons; (2) absence

ou présence de tubercules sous-articulaires sur quelques doigts; (3) plaques pectorales des

mâles de taille équivalente, ou presque, aux glandes axillaires; (4) présence de callosités

nuptiales sur le doigt III des mâles; (5) épines fines et marrons sur les callosités nuptiales des

mâles; (6) élargissement ou non des avant-bras des mâles; (7) présence ou non d’aspérités

(selon la définition de LYNCH & DUELLMAN, 1997) sur le bord des lèvres inférieures et sur le

dos des mâles; (8) absence ou présence d’épines éparses sur les bras des mâles.

Groupes d'espèces. — Fe1 (1999: 339-340) a défini dans ce sous-genre trois groupes d'espèces.

Cette taxinomie provisoire peut être utilisée comme hypothèse de travail (voir DuBois,

1999: 82-84). Toutefois, afin de se conformer à la règle de priorité, qui s’applique aussi aux

noms de taxons de rang infragénérique et supraspécifique tels que les groupes d’espèces

(voir Dugois & OHLER, 1999: 137), deux de ces groupes doivent être renommés, comme

ci-dessous.

Espèces incluses. — Neuf espèces, réparties comme suit dans trois groupes d'espèces: (1) groupe

de Scutiger boulengeri: Scutiger (Scutiger) boulengeri (Bedriaga, 1898); Scutiger (Scutiger )

liupanensis Huang, 1985; Scutiger (Scutiger) ningshanensis Fang, 1985: (2) groupe de Scutiger

chintingensis: Scutiger (Scutiger) chintingensis Liu & Hu, 1960; Scutiger (Scutiger ) pingwue)

sis Liu & Tian, 1978; (3) groupe de Scutige utiger (Scutiger) maculatus (Liu,

1950); Scutiger (Scutiger) nepalensis Dubois, 1974; Scutiger (Scutiger) nyingchiensis Fei,

1977; Scutiger ( Scutiger) sikimmensis (Blyth, 1854).

Source : MNHN, Paris

150 ALYTES 19 (2-4)

Megophrys montana

Leptobrachium (Leptobrachium) hasseltii

Leptolalax gracilis

Leptobrachella mjobergi

Oreolalax (Oreolalax)schmidti

Oreolalax (Oreolalax) popei

Oreolalax (Aelurolalax)weigoldi

Oreolalax (Oreolalax) pingii

Scutiger (Scutiger) boulengeri

Scutiger (Scutiger) nepalensis

Scutiger (Scutiger) sikimmensis

Scutiger (Scutiger) nyingchiensis

Scutiger (Aelurophryne) bhutanensis

Scutiger (4elurophryne) mammatus

Scutiger (Aelurophryne) glandulatus

Scaphiopus couchii

Pelodytes punctatus

Fig. 5. - Consensus de 12 arbres équiparcimonieux de 124 pas (DELORME & DuBoIs, en préparation) au

sein de 17 espèces de Megophryidae, basé sur 54 caractères informatifs tirés de la morphologie

externe. Indice de Bremer sur les branches: CI = 0,581; RI = 0,713. Liste des caractères utilisés: (1)

largeur de la tête (aussi large que longue | plus longue que large ! plus large que longue); (2) projection

du museau en avant de la mâchoire inférieure (absence ! présence); (3) forme de la mâchoire

inférieure (ronde ou presque ronde | semi-circulaire avec base large ! avec base allongée l avec base

ronde mais extrémité rétrécie ): (4) position des narines par rapport au bout du museau et aux yeux (à

mi-distance entre les yeux et le bout du museau / plus proche du bout du museau que des yeux): (5) pente

du museau {profil aigu l obtus / rond); (6) forme du bout du museau (bien rond | avec des arêtes ou des

projections dermales): (7) glandes sur les paupières (absence | présence): (8) forme du canthus

rostralis (très obtus avec une région loréale concave | peu distinct avec une région loréale plutôt verticale

L'indistinct avec une région loréale convexe ): (9) écart entre les narines (largement inférieur à l'espace

interorbital | aussi grand que ou légèrement supérieur à l'espace interobital); (10) tympan (absence |

présence distincte ou cachée sous la peau); (11) forme du repli tympanique (fin / en forme de glande

parotoïde ); (12) glande à l'extrémité du repli tympanique (absence | présence sous forme d'une glande

rictale bien ronde | présence d'une glande en contact avec le repli supratympanique ): (13) tache claire à

l'extrémité du museau {absence | présence ); (14) aspérités ou épines sur la lèvre supérieure (absence

1 présence): aspect des épines l'ensemble du corps, coussinets nuptiaux exclus (très fines et

brunes l très kératinisées et noires l très »s et noires, de très grande taille}: (16) aspérités ou

épines sur la lèvre inférieure {absence / présence ); (17) disposition des aspérités sur la lèvre inférieure

(non regroupées | regroupées en plaques); (18) dents maxillaires (absence | présence}; (19) dents

vomériennes (absence | présence ); (20) liberté de la langue (soudée, ou libre sur 14 maximum de sa

longueur | libre sur 1/4 à 113 de sa longueur ! libre sur plus de 213 de sa longueur }; (21) crêtes près des

choanes {absence | présence de crêtes horizontales liant les choanes | présence de crêtes retombant en

arrière de la bouche ); (22) fentes de sacs vocaux sur les du plancher buccal {absence | présence};

(23) position et largeur des fentes des sacs vocaux (petite ouverture, près de la commissure des lèvres

Source : MNHN, Paris

DELORME & DUBOIS 151

Sous-genre Aelurophryne Boulenger, 1919

Espèce-type. - Bufo mammatus Günther, 1896, par monotypie.

Diagnose. - Ce sous-genre se distingue du sous-genre Scutiger par la combinaison suivante de

caractères: (1) absence de dents maxillaires et de bourgeons; (2) présence de tubercules

sous-articulaires sur tous les doigts; (3) plaques pectorales des mâles légèrement ou bien plus

grandes que les glandes axillaires; (4) absence de callosités nuptiales sur le doigt III des mâles;

(5) épines noires de grande taille, très kératinisées, sur les callosités nuptiales des mâles; (6)

élargissement des avant-bras des mâles; (7) présence d’aspérités sur le bord des lèvres

inférieures et sur le dos des mâles; (8) absence d’épines éparses sur les bras des mâles.

Groupes d'espèces. — FE1 (1999: 340) a défini dans ce sous-genre trois groupes d'espèces, qui

peuvent ici aussi être employés au titre de taxinomie provisoire de travail.

Espèces incluses. - Huit espèces, réparties comme suit dans trois groupes d'espèces: (1) groupe

de Scutiger glandulatus: Scutiger (Aelurophryne) bhutanensis sp. nov; Scutiger (Aelu-

l'ouverture moyenne étendue légèrement vers l'avant de la bouche | grande ouverture, très étendue vers l'avant

de la bouche : (24) verrues dorsales (absence | présence ; (25) taille des verrues dorsales (homogènes | très

diverses; (26) aspérités sur le dos {absence | présence); (27) épines dorsales (absence | présence d'épines

fines et brunes, bien définies, sur le bas du dos | présence d'épines fines et brunes, bien définies, sur l'ensemble

du dos): (28) épines cloacales (absence | présence); (29) glandes latérales (absence | présence); (30)

ion des glandes axillaires (sur le flanc, à l'insertion du bras | sur le côté de la poitrine l étendue entre

l'insertion du bras et le côté de la poitrine ); (31) taille des glandes axillaires (aussi large que le doigt II | deux

fois plus large que le doigt 11): (32) épines sur les glandes axillaires (absence / présence); (33) glandes

pectorales (absence | présence ): (34) taille des glandes pectorales (de taille similaire aux glandes axillaires

1 beaucoup plus grandes que les glandes axillaires); (35) écart entre les talons quand les membres

postérieurs sont placés à angle droit avec le corps (talons très écartés l talons se touchant ou se recouvrant

très légèrement | talons se recouvrant très largement ); (36) forme du tubercule palmaire externe (allongé et

diflus î rond et de la taille du tubercule palmaire interne); (37) tubercule palmaire médian (absence |

présence ): (88) tubercules subarticulaires {absence / présence sur quelques doigts / présence sur l'ens

des doigts): (39) crêtes sur les doigts (absence | présence }: (40) différence de quantité de tubercule:

subarticulaires entre pieds et mains (pas de différence | présence sur les pieds mais pas sur les mains |

présence sur les mains et en moindre importance sur les pieds ): (41) verrues rondes et plates sur les bras et

les mains {absence | présence}; (42) coussinets nuptiaux sur les doigts I et II des mâles (absence /

présence }; (43) étalement du coussinet nuptial sur le doigt [ (étendu en arrière de la deuxième phalange du

doigt 1 jusqu'à l'articulation entre le métacarpe et la première phalange | étendu en arrière de la deuxième

Phalange du doigt L jusqu'à l'articulation du poignet, et même un peu au-delà ): (44) épines des cou

nuptiaux (très fines, à peine visibles | fines, en groupes bruns | trè.

coussinets nuptiaux sur le doigt III des mâles (absence | présence.

bras (absence | présence ): (47) épines éparses sur les bras (absence / présence ): (48) élargissement des bras

des par rapport aux femelles (absence | présence }: (49) extension de la palmure (orteils libres ou

légère palmure au niveau des premières phalanges | palmure large, étendue à toutes les phalanges | palmure

étendue en diagonale jusqu'à la deuxième phalange }; (50) épines sur les pieds (absence / présence }: (51)

glandes fémorales (non individualisées sur la cuisse ! individualisées ); (52) coloration des membres (en

bandes | diffuse ): (53) coloration de l'anus (absence ! présence ): (54) longueur museau-anus (inféri

25 mm l'entre 25 er 60 mm | supérieure à 60 mm).

Source : MNHN, Paris

152 ALYTES 19 (2-4)

rophryne) glandulatus (Liu, 1950); Scutiger (Aelurophryne) jiulongensis Fei, Ye & Jiang, 1999;

Scutiger (Aelurophryne) tuberculatus Liu & Fei, 1979; (2) groupe de Scutiger gongshanensis:

Scutiger (Aelurophryne) adungensis Dubois, 1979; Scutiger (Aelurophryne) gongshanensis

Yang & Su, 1979; (3) groupe de Scutiger mammatus: Scutiger (Aelurophryne) mammatus

(Günther, 1896); Scuriger (Aelurophryne) muliensis Fei & Ye, 1986.

REMERCIEMENTS

Nous remercions Rainer Günther (Berlin) de nous avoir transmis ces deux spécimens pour étude, et

les lecteurs pour leurs suggestions sur le manuscrit original. Cet article est la publication N° 48 du PPF

“Faune et flore du sud-est flore du sud-est asiatique” (N° 47, voir Dunois & OnLer, 2001).

RÉFÉRENCES BIBLIOGRAPHIQUES

ANONYME [Herpetological Department, Sichuan Biological Institute], 1977. — A survey of amphibians in

Xizang (Tibet). Acta zool. sinica, 23 (1): 54-63, pl. 1-2. [En Chinois].

BARONI URBANI, C., STEMMLER, O., WITTMER, W. & WÜRMLI, M., 1973. - Zoologische Expedition des

Naturhistorischen Museums Basel in das Kônigreich Bhutan. Verhandl. naturforsch. Ges. Basel, 83

(2): 319-336.

Dusois, A., 1979. - Une espèce nouvelle de Scutiger (Amphibiens, Anoures) du nord de la Birmanie. Rev.

suisse zool., 86: 631-640.

= 1980. - Notes sur la systématique et la répartition des Amphibiens Anoures de Chine et des régions

avoisinantes IV. Classification générique et subgénérique des Pelobatidae Megophryinae. Bull. Soc.

linn. Lyon, 49: 469-482.

1987a. — Miscellanea taxinomica batrachologica (1). Alytes, 5 (1-2): 7-95.

1987. - Miscellanea nomenclatorica batrachologica (XIV). Alyres, “1986”, 5 (4): 173-174.

1992. - Notes sur la classification des Ranidae (Amphibiens, Anoures). Bull. Soc. linn. Lyon, 61 (10):

305-352.

--- 1999. - Miscellanea nomenclatorica batrachologica. 19. Notes on the nomenclature of Ranidae and

related groups. Alytes, 17 (1-2): 81-100.

= 2001. - The taxonomic status of Scuriger mokokchungensis Das & Chanda, 2000, Amphibia-Reptilia,

sous presse.

Dusois, À. & OuLer, A., 1998. — A new species of Leptobrachium (Vibrissaphora) from northern

Vietnam, with a review of the taxonomy of the genus Leprobrachium (Pelobatidae, Megophryi-

nac). Dumerilia, 4 (1): 1-32.

1999. - Asian and Oriental toads of the Bufo melanostictus, Bufo scaber and Bufo stejnegeri groups

(Amphibia, Anur: list of available and valid names and redescription of some name-bearing

types. J South Asian nat. Hist., 4 (2): 133-180.

2001. A new genus for an aquatic ranid (Amphibia, Anura) from Sri Lanka. Alyres, 19 (2-4): 81-106.

Fet, L. (ed.), 1999. — Arlas of amphibians of China. Zhengzhou (China), Henan Press of Science and

Technology: [i-üi] + 1-432. [In Chinese]

Fu, L, YE, C. & HUANG, Y., 1991. — Key 10 Chinese Amphibia. Chongqing, Editions of Sciences and

Techniques: [i-iv] + 1-2 + 1-364. [In Chinese].

FU, JL, LarRoP, À. & Murrny, R., 1997. - Phylogeny of the genus Seuriger (Amphibia: Megophryidae):

a re-evaluation. Asiat. herp. Res., 7: 32-37.

Fu, J. & MurPuy, R., 1997. - Phylogeny of Chinese Oreolalax and the use of functional outgroups to

select among multiple equally parsimonious trees. Asiat. herp. Res, T: 38-43.

GLAW, FE, KÔHLER, J, HOFRICHTER, R. & DUBOIS, A., 1998b. — Systematik der Amphibien: Liste der

rezenten Familien, Gattungen und Arten. /n: R. HOFRICHTER (ed.), Amphibien, Augsburg, Natur-

buch Verlag: 252-258.

Source : MNHN, Paris

DELORME & DUBoIs 153

INGER, R. F,, 1966. - The systematics and zoogeography of the Amphibia of Borneo. Fieldiana: Zool., 52:

1-402.

Liv, C.-C., 1936. - Secondary sex characters of Chinese frogs and toads. Zool. Ser, Field Mus. nat. Hist.,

22 (2: 115-126.

---- 1950. - Amphibians of western China. Fieldiana: Zool. Mem., 2: 1-400, pl. 1-10.

LyNCH, J. D., & DUELLMAN, W. E., 1997. — Frogs of the genus Eleutherodactylus in western Ecuador:

systematics, ecology, and biogeography. Spec. Publ. Unix. Kansas nat. Hist. Mus., 23: 24-28.

Myers, GS. & LEVITON, A. E., 1962. - Generic classification of the high-altitude pelobatid toads of Asia

(Scutiger, Aelurophryne, and Oreolalax). Copeia, 1962: 287-291.

OuLer, À & Dugois, A, 1992. - The holotype of Megalophrys weigoldi Vogt, 1924 (Amphibia: Anura,

Pelobatidae). J Herp., 26: 245-249,

Sworrorp, D. L., 1993. - PAUP: Phylogenetic Analysis Using Parsimony, version 3.1.1. Computer

program distributed by the Illinois Natural History Survey, Champaign, Illinois.

YANG, D.-T., Su, C.-Y. & Li, S.-M., 1979. - New species and new subspecies of amphibians and reptiles

from Gaoligong Shan, Yunnan. Acta zootaxonomica sinica, 4 (2): 185-188.

YE, C., Fer, L, & Hu, S., 1993. - Rare and economic amphibians of China. Chengdu, Sichuan Publishing

House of Science and Technology: [iii] + 1-2 + 1-2 + 1-7 + 412. [In Chinese].

APPENDICE 1

AUTRES SPÉCIMENS EXAMINÉS

Leptobrachella mjobergi Smith, 1925. - MALAISIE: Sarawak: 7 division (FMNH 222727, 222751,

222763-73, 222806).

Leptobrachium hasseltii Tschudi, 1838. — PHILIPPIN

(FMNH 51027-29).

Leptolalax gracilis (Günther, 1872). - MALAISIE: Sabah: Kina Balu (MNHN 1898.284-287); Sarawak: 7

division (FMNH 5641, 9170, 12542, 13819, 222663 687).

Megophrys montana Kuhl & Van Hasselt, 1822. - INDO: fava(MNHN 1211); PHILIPPINES: Mindanao

(MNHN 4537); Palawan (MNHN 1881.102-103, 1889.350-352).

Oreolalax (Aelurolalax) weigoldi (Vogt, 1924). - Données tirées de OnLer & Dugois (1992).

Oreolalax (Oreolalax) pingi (Liu, 1943). — CHixE: Sichuan (FMNH 232922-29).

Oreolalax (Oreolalax) popei (Liu, 1947). - Car ichuan (FMNH 232953).

Oreolalax (Oreolalax ) schmidti (Liu, 1947). - CHine: Sichuan (FMNH 232938-944); Emei Shan (MNHN

1987.3854).

Pelodytes punctatus (Daudin, 1802). - FRANCE: Indre (MNHN 6465-6468, 1980.1785-1787).

Scaphiopus couchii Baird, 1854. - MEXIQUE (MNHN 1897.398): Baja California (MNHN 1984.148-1 50);

Coahuila: au nord de Torreon (MNHN 2152).

Scutiger (Aelurophryne) glandulatus (Liu, 1950). - Données tirées de Liu (1950).

Scutiger (Aclurophryne) mammatus (Günther, 1896). - CNE: Sichuan: Hlagong, près de Yajiang, à

l’ouest de Kangding (MNHN 1987.3852-3853).

Scutiger (Scutiger) boulengeri (Bedriaga, 1898). - NépaL: Muktinath (MNHN 1977.1146-1155).

Scutiger (Scutiger) nepalensis Dubois, 1974. - NépaL: Khaptar, 2950 m, entre Doti et Chainpur (MNHN

1974.1095-1098, 1989.3361-3362).

Scutiger (Scutiger) nyingchiensis Fei, 1977. — INDE: Jammu & Kashmir: Shukdhari, 2920-2940 m,

au-dessus de Sonamarg (MNHN 1977.1070-1128).

Scutiger (Scutiger) sikimmensis (Blyth, 1854). - NéPAL: Pungotanga (MNHN 1977.1199-1233); Salpa

Pokhari (MNHN 1977.1263-1283).

Mindanao (FMNH 14893, 50919-23); Palawan

Corresponding editor: Franco ANDREONE.

Source : MNHN, Paris

Alytes, 2001, 19 (2-4): 154-172.

Anâlisis morfométrico de Pleurodema thaul

(Lesson, 1826) (Anura, Leptodactylidae)

y algunas consideraciones acerca

de su morfologia esternal

Sergio D. Rosser*, Néstor G. Basso** & Rubén J. LOMBARDO***

* Instituto de Limnologia “Dr. Raül A. Ringuelet” (ILPLA), Casilla de Correo 712, 1900 La Plata, Argentina

** Centro Nacional Patagénico (CENPAT), Blvd. Brown s/n., 9120 Puerto Madryn, Argentina

*** Laboratorio de Limnologia, Departamento de Ciencias Biolôgi

Facultad de Ciencias Exa y Naturales, Universidad de Buenos Aires,

Ciudad Universitaria, 1428 Buenos Aires, Argentina

Morphometry and anatomical variability of Pleurodema thaul popula-

tions from Argentina and Chile were analyzed. Multivariate statistical tests

were applied to 16 morphometric variables. Results indicate that the

studied populations of P. thaul cannot be distinguished in different mor-

photypes, as previously proposed. Skeletal morphology of the sternum

{xiphisternum) reveals that the Argentine populations of P. thaul have an

incised sternum, as the sternum observed in chilean specimens. Our results

do not allow supporting the separation of P. thaul in different specific

entities.

INTRODUCCION

Pleurodema Tschudi, 1838 ha sido considerado el género mäs primitivo de los leptodac-

tilidos de la subfamilia Leptodactylinae debido a caracteristicas plesiomorfas presentes tanto

en la morfologia externa de adultos y larvas como en la osteologia (LYNCH, 1971). Las

especies que integran este género poseen un tamaño mediano (30-55 mm entre hocico y

cloaca), häbitos terrestres o cavadores y, la mayoria de ellas, un par de gländulas lumbares

bien desarrolladas. Osteolégicamente, se caracterizan por la presencia de un amplio estilo

esternal 6seo, fontanela frontoparietal grande y por la ausencia de osificacién cuadrado-

yugal en el arco maxilar (LYNCH, 1971; Cet, 1980).

Actualmente, en el género Pleurodema se reconocen 12 especies (FROST, 1985) y persisten

algunas controversias taxonémicas. Un ejemplo lo constituye Pleurodema borellit (Peracca,

1895), considerada por PARKER (1927) dentro de la sinonimia de Pleurodema cinerea Cope,

1877, pero sustentada como especie distinta por GALLARDO (1968) sobre la base de diferencias

de tamaño, coloraciôn y morfologia del esternén. Autores tales como BARRIO & RINALDI DE

CHiEri (1970) y Mc ListER et al. (1991) se inclinan a favor de tal sinonimia, mientras que

DUELLMAN & VELOSO (1977) y Cet (1980), entre otros, reconocen a P. borellit como especie

välida y distinta de P cinerea.

Source : MNHN, Paris

Rosser, BASSO & LOMBARDO 155

Por otro lado, Pleurodema thaul (Lesson, 1826), distribuida en el sur de Argentina y a lo

largo de Chile, sinonimizada tiempo atrâs con las poblaciones uruguayas de Pleurodema

bibroni (ver C1, 1962; Donoso-BaRRoS, 1969), es actualmente considerada una especie välida

de elevado polimorfismo (Crr & CaPuRRO, 1957; Cer & EspiNA AGUILERA, 1957; CEr, 1958,

1962). Sin embargo, debido a la amplia variabilidad morfolégica, observada fundamental-

mente en poblaciones chilenas, han existido dudas sobre su estatus taxonémico. DUELLMAN &

VELOSs0 (1977) señalan la posibilidad de que bajo esta entidad puedan reconocerse mâs de una

especie.

Pleurodema thaul se distribuye en los bosques cordilleranos patagénicos de la Argentina

desde el alto valle del rio Neuquén, en la provincia de Neuquén, hasta la regién de los lagos

Menéndez y Futalaufquen, en la provincia de Chubut. En Chile se extiende desde Antofa-

gasta y los rios Copiapé y Huasco, al norte, hasta la regiôn de Aisén al sur (Cet, 1962; VELOSO

& NAvaARRO, 1988), ocupando ambientes tan variados como son las regiones desérticas del

norte, los bosques de la regiôn central, la selva valdiviana, las zonas cordilleranas (hasta los

1500 m) y las orillas rocosas del mar (Cet, 1958).

Ce (1958, 1962) considera que la amplia variabilidad geogräfica de P thaul en caracteres

morfolégicos, fisiolégicos y etolégicos tales como la longitud total, el tamaño de las gländulas

lumbares, la coloracién, las secreciones cutäneas y el nümero de relieves glandulares, repre-

sentaria un polimorfismo, debido, en parte, a que algunos de estos caracteres presentan

evidentes variaciones clinales a lo largo de Chile. Ademäs, segün Ce1 (1960), las poblaciones

chilenas del norte (Copiapé, Huasco) y del sur (Llanquihue, Aisén), fenotipicamente diferen-

tes, estarian conectadas genéticamente a través de las poblaciones intermedias y no constitui-

rian especies o subespecies distintas. Las poblaciones argentinas de P thaul, escasamente

estudiadas, serian para Cet (1960) comparables a las de la regiôn valdiviana de Chile.

DUELLMAN & VELOSO (1977) describen variacién entre poblaciones de P thaul en

caracteres morfolégicos, en el modo de deposicién de los huevos, el tipo de amplexo y la

morfologia del esternôn. Basändose en esta variabilidad y en las diferencias poblacionales

observadas por VELOSO et al. (1973) a nivel cariolôgico, proponen la existencia de tres

morfotipos dentro de P thaul. Segün DUELLMAN & VELOSO (1977), estos morfotipos, identi-

ficados sobre la base de su distribuciôn geogräfica como Argentina Sur, Chile Sur y Chile

Central, deberian nominarse como especies distintas. Sin embargo, estos autores se abstienen

de su reconocimiento taxonémico a causa del incompleto conocimiento de los limites

distribucionales y de la variabilidad intrapoblacional. El morfotipo referido a Argentina Sur

coincide con toda la extensién de la especie en Argentina; Chile Sur abarca la provincia de

Llanquihue, la Isla de Chiloé y la regién de Aisén; y Chile Central se extiende desde La Serena

hasta Concepcién. Las poblaciones chilenas distribuidas entre Concepciôn y Valdivia

podrian estar integradas por representantes de los morfotipos chilenos central y sur o, incluso,

constituir un taxén distinto (DUELLMAN & VELOSO, 1977).

Con el objeto de analizar la variabilidad de las diferentes poblaciones actualmente

referidas a Pleurodema thaul, identificar caracteres que permitan establecer diferencias entre

las poblaciones y aportar mayor informaciôn para dilucidar su posicién taxonémica, se Ilevé

a cabo un anälisis morfométrico y se realizaron observaciones comparativas de su osteologia.

Source : MNHN, Paris

156 ALYTES 19 (2-4)

MATERIALES Y MÉTODOS

Se estudiaron 301 ejemplares adultos pertenecientes a Pleurodema thaul que forman

parte de las colecciones herpetolôgicas del Museo Argentino de Ciencias Naturales “Bernar-

dino Rivadavia” (MACN), del Field Museum of Natural History (FMNH), del Museo de La

Plata (MLP) y del Instituto de Limnologia “Dr. Raül A. Ringuelet” (ILPLA).

Sobre cada ejemplar se registraron 16 variables morfométricas exosomäticas: (1) longi-

tud hocico-cloaca (LHC); (2) ancho de la cabeza (AC); (3) longitud de la cabeza (LC); (4)

longitud de la tibia (LT); (5) longitud del pie (LP); (6) longitud del fémur (LF); (7) distancia

internarial (DIN); (8) distancia interorbital (DIO); (9) longitud de la gländula lumbar (LGL);

(10) ancho de la gländula lumbar (AGL); (11) distancia del ojo a la narina (ON); (12)

distancia del ojo al hocico (OH); (13) longitud del pärpado superior (PAR): (14) longitud del

ojo (OJO); (15) longitud del tubéreulo metacarpal interno (MCI); (16) longitud del tubérculo

metatarsal interno (MTI).

Las medidas fueron tomadas con calibre de escala Vernier con una precisién de 0,02 mm.

Se tuvieron en consideracién aquellos especimenes cuya longitud hocico-cloaca superara los

21 mm, para asegurar el estudio con ejemplares que hayan alcanzado la madurez sexual (Cet,

1962). Ademäs, para cada uno de los especimenes se determiné el sexo sobre la base de la

presencia o ausencia de saco vocal.

El material proviene de 52 localidades de Argentina y Chile y abarca la mayor parte de la

distribuciôn geogräfica de P thaul. Los especimenes estudiados en cada una de las localidades

se encuentran en el ap.1. Para tener en cuenta los tres morfotipos propuestos por DUELLMAN

& VELOSO (1977), el material se organiz6, de acuerdo a su procedencia, de la siguiente manera:

() Argentina Sur: incluye las 36 localidades estudiadas de Argentina (7 = 198); (2) Chile

Central: abarca Coquimbo, Zapallar, Viña del Mar, Valparaiso, Santiago y Concepciôn (7 =

40); (3) Chile Sur: abarca Lago Todos Los Santos, Llanquihue, Correntoso, Chamiza, Puerto

Montt e Isla de Chiloé (7 = 43). Las localidades de Vegas Blancas, Cordillera de Nahuelbuta,

Lago Villarrica y Valdivia pertenecen a la regién chilena intermedia a Chile Sur y Chile

Central (n = 20).

ANAÂLISIS MORFOMÉTRICO

Los valores de las 16 variables morfométricas obtenidos de 301 especimenes fueron

analizados utilizando técnicas estadisticas multivariadas: anälisis de componentes prince

les (ACP), anälisis multivariado de la varianza (MANOVA), anälisis discriminante y anälisis

de agrupamientos.

de componentes principales se Ilevaron a cabo utilizando el programa

NTSYS-pc versiôn 1.8 (ROHLF, 1993), mientras que el MANOVA, el anä discriminante, el

é de agrupamientos y las pruebas de normalidad y homogeneidad de varianza-

covarianza se realizaron usando el programa STATISTICA versiôn 5.1 (Srarsorr, 1996).

Source : MNHN, Paris

Rosser, BASso & LOMBARDO 157

Andlisis de componentes principales

Se Ilevaron a cabo, por sexo, tres anälisis de componentes principales para establecer si

las variables morfométricas permiten ordenar a los especimenes de P thaul en relaciôn con su

distribuciôn geogräfica. En cada uno de estos anälisis se extrajeron los tres primeros compo-

nentes principales. El primer ACP se desarrollé a partir de una matriz de varianza-covarianza

de los datos transformados a logaritmo natural (ACP de varianza-covarianza); esta transfor-

maciôn se [levé a cabo con el fin de homogeneizar la magnitud de las variables morfométricas.

En un anälisis de este tipo los individuos se ordenarän en funciôn de su tamaño y forma

(REYMENT et al., 1984).

El segundo anälisis de componentes principales se Ilev6 a cabo a partir de una matriz de

correlaciôn obtenida de la estandarizaciôn de las 16 variables morfométricas (ACP de

correlacién). A través de la estandarizaciôn todas las variables contribuyen equitativamente

en el anälisis, permitiendo analizar los cambios en las proporciones de las variables morfo-

métricas. Finalmente, el tercer ACP se [levé a cabo siguiendo el método de Burnaby (ACP de

Burnaby) segün lo indicado por RoHLr (1993) para remover el efecto del tamaño en la

ordenacién de los especimenes.

Anälisis multivariado de la varia discriminante

za (MANOVA) y andlisi:

El MANOVA se realizé para poner a prueba la hipôtesis de existencia de los grupos

Argentina Sur, Chile Sur y Chile Central dentro del material estudiado. El anälisis diserimi-

nante (“forward stepwise”) permitié seleccionar las variables morfométricas que mejor

discriminan entre los conjuntos mencionados. Se utilizaron las variables morfométricas

transformadas a logaritmo natural y los supuestos estadisticos de normalidad y homocedacia

se probaron mediante el test de Kolmogorov-Smirnov y el test multivariado Box M de

homogeneidad de varianza-covarianza, respectivamente.

Andlisis de agrupamientos

Se realizaron, por sexo, dos anälisis de agrupamientos para estudiar si la similitud

morfométrica de los especimenes permite establecer agrupaciones relacionadas con la distri-

buciôn geogräfica. En el primer anälisis se utilizaron las 16 variables morfométricas estanda-

rizadas y como medida de similitud se empleé el coeficiente de correlaciôn de Pearson. El

segundo anälisis de agrupamientos se realizé con las 16 variables morfométricas transforma-

das a logaritmo natural y se utilizé como medida de similitud la distancia euclideana. En

todos los casos se empleé la técnica de ligamiento promedio de la media aritmética no

ponderada (UPGMA) (CRisc1 & LÔPEZ ARMENGOL, 1983).

ANÂLIS

[S DEL ESQUELETO

La morfologia del esqueleto se analizé mediante preparados obtenidos segün el método

de doble tincién y transparentado descripto por TAYLOR & VAN DYKE (1985). El procedi-

miento se aplicé a un total de 19 especimenes (17 de Argentina y 2 de Chile) e incluye la tinciôn

del tejido cartilaginoso por medio de una solucién de azul Alcian, la tinciôn del tejido 6seo

Source : MNHN, Paris

158 ALYTES 19 (2-4)

utilizando una solucién de rojo de alizarina y la diafanizacién del tejido muscular por medio

de una solucién de KOH o tripsina. La duracién de cada paso y las concentraciones de las

soluciones se ajustaron segün el estado en que se encontraba el material. Para una mejor

visualizaciôn del esqueleto, la conservaciôn final de los ejemplares se realizé en glicerina al

100 %.

RESULTADOS

ANÂLISIS MORFOMÉTRICO

Los estadisticos descriptivos de las 16 variables morfométricas estudiadas en Pleurodema

thaul se encuentran detallados por sexo en la tab. 1. Ademäs, se presentan los resultados de las

pruebas de igualdad de medias entre sexos, realizadas, para cada variable, mediante el test de

Student. Se observa que a excepcién de la variable longitud del ojo, las demäs variables

morfométricas muestran diferencias significativas entre los sexos (P < 0,05). Por otro lado, la

media de las variables es mayor en las hembras que en los machos, salvo en el caso de longitud

del tubéreulo metacarpal interno.

La variaciôn geogräfica de las poblaciones de P thaul en cuanto a longitud hocico-cloaca

y longitud de la glândula lumbar se visualiza en la fig. 1, en donde se representan, por sexo, la

media, el desvio estandar y el rango de estas variables. Las localidades se encuentran

ordenadas de norte a sur tanto en Argentina como en Chile.

Andlisis de componentes principales

El anälisis de componentes principales de varianza-covarianza Ilevado a cabo con las

hembras explica, a través de los tres primeros componentes principales, el 88,05 % de la

variabilidad. Las variables que mäs contribuyen a describir esta variabilidad morfométrica

son: longitud de la gländula lumbar, ancho de la gländula lumbar, longitud del tubérculo

metatarsal interno, longitud del tubéreulo metacarpal interno y longitud hocico-cloaca. En la

tab. 2 se encuentran los autovalores y autovectores obtenidos. Las ordenaciones resultantes de

graficar los especimenes hembra sobre los componentes 1 vs. 2, 1 vs. 3 y 2 vs. 3se presentan en

la fig. 2. Se observa una ünica nube de puntos dentro de la cual no es posible encontrar ningün

tipo de ordenacién de los especimenes relacionada con su distribuciôn geogräfica. Por otra

parte, existe una amplia superposicién de los conjuntos formados por los ejemplares perte-

necientes à Argentina Sur, Chile Central y Chile Sur. Elmayor nûmero de especimenes de Chile

Sur se ubica a altos valores del primer y segundo componentes, mientras que la mayor

cantidad del material proveniente de Chile Central se ubica a valores bajos de estos

componentes. Sin embargo, no es posible establecer diferencias morfométricas entre estos

grupos debido a su amplia superposicién. Los ejemplares de la regién chilena intermedia

(Vegas Blancas, Cordillera de Nahuelbuta y Valdivia) se agrupan tanto con los ejemplares de

Chile Central como con los de Chile Sur. Los ejemplares de Argentina se encuentran dispersos

en la nube de puntos y superpuestos con los ejemplares chilenos, a lo largo de los tres

componentes.

Source : MNHN, Paris

RossEeT, BASsO & LOMBARDO

159

Tabla 1. — Estadisticos descriptivos de las variables morfométricas estudiadas en

Pleurodema thaul por sexo (H, hembras, n = 172; M, machos, n = 129). C.V.,

coeficiente de variacién.

Variable Sexo Media Dee Minimo Mäximo CV. Test

H 3882 736 Z2I/4 5425 019 6.9

De À HAN CNE M 3432 427 2580 4744 012 P<0.05

Le nes H 1304 232 692 1822 018 446

M 1201 144 892 1665 012 P<0.05

us H 1139 199 603 1550 O17 3.93

M 1055 1.57 746 1995 015 P<0.05

Longitud de la tibia H 1698 300 1038 2230 018 4.97

M 1549 186 10.90 2206 0.12 P<0.05

Ts H 2755 519 1572 3776 O19 4.79

M 2508 315 1810 3310 013 P<0.05

RO H 1553 316 868 2348 020 4.84

M 1398 206 986 1955 015 P<0.05

RON ELUIT H 268 042 165 390 016 681

M 240 024 176 300 010 P<0.05

Distancia interorbital Fe sos qe 745 848 026, 6.59

M 534 053 435 786 010 P<0.05

! H 564 142 255 940 025 602

Longitud dela gléndula lumbar | NE 476 097 295 784 020 P<0.05

j H 305 (080 119 540 026 5.84

Ancho dela gländulalumbar | Nf 55 060 140 430 024 P<0.05

ler A H 300 048 182 460 016 595

M 272 030 205 360 0.11 P<0.05

ee H 542 090 325 740 017 5.86

M 489 055 365 680 O11 P<0.05

; H 512 088 297 715 017 3.18

Longitud del pârpado superior |A Ug4 O61 370 705 013 P<0.05

PAL H 392 065 229 540 017 181

M 379 049 262 514 013 P>0.05

Longitud del tubérculo H 214 0.50 120 395 023 3.52

metacarpal interno M 232 036 140 310 016 P<0.05

Lonsitid AEtiuséreuts H 202 046 090 305 023 5.68

metatarsal interno M 175 036 070 290 021 P<0.05

Source : MNHN, Paris

HEMBRAS MACHOS

LONGITUD HOGICO-CLOACA

LONGITUD DE LA GLANDULA LUMBAR

6) Valdivia; (7) Llanquihue; (8)

() Puerto Mont; (10) Isla de Chiloé: (11) Aluminé; (12) Lago Tromen: (13) Lago

Curruhue: (14) Junin de los Andes: (15) Lago Läcar: (16) San Martin de los Andes: (17) Pichi Traful:

(18) Rio Pichi Traful: (19) Isla Victoria; (20) Puerto Blest (21) Lago Frias: (22) Lago Nahuel Huapi:

(23) Monte Tronador: (24) Rio Manso Superior: (25) Lago Hess: (26) El Bolsén: (27) Lago Puelo:

(28) Cerro Puntudo: (29) Lago Verde: (30) Lago Futalaufquen.

Source : MNHN, Paris

Rosser, Basso & LOMBARDO 161

Tabla 2. — Resultados de los tres anälisis de componentes principales realizados con

las hembras. Se muestran los autovectores, los autovalores y el porcentaje de

varianza explicada.

ACP varianza-covarianza ACP correlaciôn ACP Burnaby

C1 c2 c3 ci c2 c3 ci c2 c3

LHC 0.196 0.002 0.023 | 0977 0025 0068 | 0.836 0376 0.303

Variable

AC 0.181 (0006 0.029 | 0.963 0019 0.105 | 0.131 (0.049 —0.185

LC 0.175 (0016 0.024 | 0944 0.045 0.103 | 0.072 -0.069 —0.250

LT 0.179 (0.021 0.023 | 0.967 -0.086 0.070 |-0.164 -0.097 —0.257

LP 0.187 0.025 0.026 | 0.950 -0.102 0.077 |-1.316 0.209 0.055

LF 0.191 0019 0.035 | 0.895 0.051 0.083 | 0.071 -1.322 0.038

DIN 0.136 -0.010 0.006 | 0.835 0.194 -0.041 | 0.027 0.049 —0.033

DIO 0.156 0.008 0.000 | 0.940 0.007 -0.062 | -0.007 0.062 —0.079

LGL 0.227 -0.085 -0.039 | 0.824 0.378 —0.112 | 0.240 0.146 —0.662

AGL 0225 -0.130 -0.017| 0793 0.490 -0.127 | 0.175 0.035 —0.153

ON 0.132 0014 0.015 | 0817 -0050 0.022 | -0.016 -0.019 -0.036

OH 0.157 0012 0.025 | 0912 -0015 0.134 | 0.021 -0.066 —0.004

PAR 0.160 0.023 0.006 | 0.908 -0.160 0.036 |-0.029 0.090 —0.008

010 0146 0017 0.036 | 0849 -0.105 0284 |-0.018 0.072 0.004

MCI 0185 04063 -0.105 | 0.793 -0276 -0479 | -0.034 0.007 -0.040

MTI 0208 0.051 -0.050 | 0.866 -0.141 -0.264 | -0.006 -0.019 -0.019

F'Autovalor | 0.517 0034 0022 | 12725 0579 0478 | 2.577 1997 O.731

Varianza | 7946 522 337 | 7053 362 299 | 3339 2588 948

Acumulado | 79.46 84.68 88.05 | 79.53 83.5 86.14 | 33.39 5927 68.75

El ACP de varianza-covarianza Ilevado a cabo con los machos explica el 75,91 % de la

variabilidad morfométrica a través de los tres componentes obtenidos. Al igual que en el

ané de las hembras las variables que mäs contribuyen a explicar la variabilidad son: ancho

de la gländula lumbar, longitud de la gländula lumbar, longitud del tubéreulo metatarsal

interno y longitud del tubérculo metacarpal interno (tab. 3). Las ordenaciones de los ejem-

plares que se obtienen a través de los componentes principales, comparadas con las ordena-

ciones del anälisis anterior, muestran una mayor superposiciéôn de los conjuntos formados por

los ejemplares pertenecientes a Argentina Sur, Chile Central y Chile Sur; por lo tanto, no es

posible establecer diferencias morfométricas entre estos grupos.

Los ejemplares de Argentina pertenecientes a una misma localidad estän ampliamente

distribuidos en la nube de puntos y superpuestos con ejemplares de otras localidades, de

manera que no es posible establecer diferencias morfométricas entre las poblaciones argenti-

nas de P. thaul. Estas ordenaciones se obtienen tanto al considerar a las hembras como a los

machos.

En el anälisis de componentes principales de correlaciôn realizado con las hembras los

tres componentes principales extraidos explican el 86,14 % de la variabilidad morfométrica.

Las variables que mäs contribuyen a estos componentes son: longitud hocico-cloaca, longitud

Source : MNHN, Paris

COMPONENTE 2

COMPONENTE 3

Fig. 2. - Ordenaciones de los especimenes hembra de Pleurodema thaul ba:

036

024

016

070

066

062

058

078

074

070

066

062

COMPONENTE 1

ër

COMPONENTE 1

016

020 024

COMPONENTE 2

028

032 036

adas en los tres primeros

‘entral, (D) Chile Sur, (@)

componentes principales del ACP de varianza-covarianza. (M) Chile C

Argentina Sur, (+) localidades de la regiôn intermedia a Chile Central y Chile Sur.

Source : MNHN, Paris

RossET, Basso & LOMBARDO 163

Tabla 3. — Resultados de los tres anälisis de componentes principales realizados con

los machos. Se muestran los autovectores, los autovalores y el porcentaje de

varianza explicada.

ACP varianza-covarianza ACP correlaciôn ACP Burmnaby

C1 C2 C3 C1 C2 C3 C1 C2 C3

LHC 0.111 0.016 0.018 0.937 0.032 0.116 0.910 0.184 0.149

AC 0.106 0.020 0.017 0.920 0.021 0.026 0.132 -0.228 —0.102

Variable

LC 0.114 0.016 0.014 0.813 -0.043 0.148 0.419 -0.339 0.167

LT 0.102 0.023 0.023 0.881 0.060 —0.089 —0.178 -0.056 —0.836

LP 0.100 0.038 0.032 0.855 0.192 —0.186 -1.109 0.677 0.170

LF 0.115 0.020 0.024 0.781 -0.057 -0.308 0.771 -1.036 0.175

DIN 0.066 0.021 0.028 0.706 0.204 —0.058 -0.025 -0.016 0.004

DIO 0.074 0.011 0.012 0.812 0.020 0.139 0.068 -0.034 -0.056

LGL 0.149 —0.102 0.025 0.664 0.580 0.169 0.137 -0.286 —0.049

AGL 0.174 0.136 -0.037 0.646 0.579 0.327 0.157 -0.155 —0.062

ON 0.073 0.013 0.013 0.715 —0.115 —-0.269 —0015 -O0.011 —0.052

OH 0.081 0.010 0.005 0.709 -0.183 -0.378 —-0.058 0.081 0.010

PAR 0.089 0.024 0.010 0.746 0.250 0.418 0.115 0.042 0.021

OJO 0.082 0.042 0.037 0.694 0.445 0.293 0.025 0.071 —0.016

MCI 0.116 0.048 0.005 0.725 0.199 -0.097 —0.030 -0.024 —0.035

MTI 0.149 0.065 0.130 0.665 -0.025 -0.232 0.005 -0.031 —0.004

Autovalor | 0.195 0.042 0.024 9.540 1.108 0.870 2.945 1.858 0.833

Varianza 56.66 12.18 7.07 59.62 6.92 5.44 35.93 22.66 10.16

Acumulado | 56.66 68.84 75.91 59.62 66.55 71.99 35.93 58.59 68.75

de la tibia, ancho de la gländula lumbar, longitud de la gländula lumbar y longitud del

tubérculo metacarpal interno (tab. 2). A partir de las ordenaciones de este anälisis, al igual que

en el ACP de varianza-covarianza, se observa una ünica nube de puntos dentro de la cual no

es posible encontrar una ordenacién relacionada con la distribuciôn geogräfica. Se observa

una amplia superposiciôn de los ejemplares pertenecientes a Argentina Sur, Chile Central y

Chile Sur y por lo tanto, no es posible establecer diferencias morfométricas entre ellos. La

mayor parte de los ejemplares de la regiôn intermedia (Vegas Blancas, Cordillera de Nahuel-

buta y Valdivia) se superponen con los de Chile Central. En la fig. 3 se representan los

componentes 2 vs. 3 de este anälisis.

En el ACP de correlaciôn realizado con los machos los tres primeros componentes

principales extraidos explican el 71,98 % de la variabilidad. Los resultados son similares al

anälisis realizado con las hembras en cuanto a las variables de mayor peso en los componentes

(tab. 3) y en cuanto a la considerable superposiciôn entre los grupos (fig. 3). En este caso, los

ejemplares de Valdivia, Vegas Blancas, Cordillera de Nahuelbuta y Lago Villarri encuen-

tran superpuestos tanto con los especimenes de Chile Central como con los de Chile Sur.

El anälisis de componentes principales siguiendo el método de Burnaby explica,

mediante los tres componentes extraidos, un porcentaje de varianza del 68,75 % tanto si el

Source : MNHN, Paris

164 ALYTES 19 (2-4)

anälisis incluye a las hembras como a los machos. En el ACP de Burnaby realizado con las

hembras las variables de mayor peso son: longitud del pie, longitud hocico-cloaca, longitud de

la gländula lumbar, longitud del fémur y longitud de la tibia (tab. 2) mientras que en el ACP

de Burnaby realizado con los machos las variables son: longitud del pie, longitud hocico-

cloaca, longitud del fémur, longitud de la cabeza y longitud de la tibia (tab. 3). Las ordena-

ciones que resultan de graficar los componentes 1 vs. 2 de los ACP de Burnaby de hembras y

de machos se representan en la fig. 4. En ninguno de estos anälisis es posible establecer

diferencias morfométricas entre los grupos Argentina Sur, Chile Central y Chile Sur. La

superposiciôn de las nubes de puntos correspondientes a estos grupos es mucho mayor que en

las ordenaciones obtenidas por ACP de varianza-covarianza y ACP de correlacién. Ademäs,

no es posible encontrar ninguna otra ordenacién de los ejemplares que se relacione con su

distribucién geogräfica.

MANOVA y anälisis discriminante

Se puso a prueba mediante MANOVA la hipétesis que considera a los grupos Argentina

Sur, Chile Central y Chile Sur como entidades diferentes. Utilizando los datos morfométricos

de los machos, el MANOVA Ilevado a cabo detecté diferencias significativas entre los grupos

(Wilks’ à = 0.6507, P < 0.05). Las variables que mejor discriminan entre Argentina Sur, Chile

Central y Chile Sur son longitud de la cabeza, ancho de la cabeza, longitud del tubéreulo

metacarpal interno y distancia del ojo al hocico. Por otro lado, utilizando una funcién

discriminante basada en estas variables, los porcentajes de especimenes correctamente clasi-

ficados fueron 61 % para Argentina Sur, 67% para Chile Central y 69 % para Chile Sur. Con

las hembras se obtuvieron resultados similares.

Andlisis de agrupamientos

Los anälisis de agrupamientos realizados, tanto con las hembras como con los machos,

no permitieron la agrupacién de los especimenes segün un patrôn de distribuciôn geogräfica.

Los diversos agrupamientos formados contienen representantes de localidades muy variadas,

provenientes tanto de Argentina Sur como de Chile Central o Chile Sur. En todos los casos, los

especimenes procedentes de Argentina se agrupan con especimenes de Chile. Con respecto a

las localidades argentinas, tampoco es posible encontrar agrupaciones que se relacionen con

su distribuciôn geogräfica; especimenes del norte de Argentina (Caviahue, Lago Tromen) se

agrupan con aquellos del centro (Puerto Blest) y sur (Lago Futalaufquen). Los resultados del

anälisis de agrupamientos empleando el coeficiente de correlaciôn de Pearson se presentan,

para los machos, en la fig. 5.

ANÂLISIS DEL ESQUELETO

Las observaciones realizadas sobre los distintos constituyentes del esqueleto de Pleuro-

dema thaul no permitieron encontrar diferencias morfolégicas de relevancia en el material

estudiado. Los datos obtenidos mäs importantes se refieren a la morfologia de la porciôn

posterior del esternôn (xiphisternôn), de estructura cartilaginosa.

Source : MNHN, Paris

ROSSET, BASsO & LOMBARDO

165

06 —

04 a

COMPONENTE 3 (hembras)

5 24 22 00 02 04

COMPONENTE 2 (hembras)

COMPONENTE 3 (machos)

08 -06 -04 -02 00 02 04

COMPONENTE 2 (machos)

Fig. 3. Ordenaciôn de especimenes de Pleurodema thaul base

da en los componentes principales

ACP de correlaciôn para hembras y machos. (M) Chile Central, (0) Chile Sur, (@) Argentina Sur, (+)

localidades de la regiôn intermedia a Chile Central y Chile Sur.

Source : MNHN, Paris

166

COMPONENTE 2 (hembras)

COMPONENTE 2 (machos)

ALYTES 19 (2-4)

-02

-08

02 04

COMPONENTE 1 (hembras)

-06 00

COMPONENTE 1 (machos)

Fig 4. - Ordenaciôn de especimenes de Pleurodema thaul resultante de los componentes principales 1 y 2

del ACP de Burnaby para hembras

localidades de la regiôn intermedia a Chile Central y Chile Sur.

machos. (M) Chile Central, (5) Chile Sur, (@) Argentina Sur, (+)

Source : MNHN, Paris

Fig.

RossET, BASsSO & LOMBARDO 167

>> >>> >

>> >>

)

>>>

o own 00

& ou © 0

>> >>» >>> EEE > >> > 2H > PEHPHEPT PHP PP > >

>> >>> > >>> >

00 02 04 06 m 10 12

Distancia de Ligamiento

5. Fenograma obtenido del anälisis de agrupamientos de los machos, utilizando el coeficiente

de correlaciôn de Pearson y la técnica de ligamiento promedio de la media aritmética no

ponderada. C. Chile Central, S. Chile Sur; À. Argentina Sur; **, localidades chilenas interme-

dias.

Source : MNHN, Paris

168 ALYTES 19 (2-4)

Fig. 6. Esternôn de Pleurodema thaul (ILPLA A.490, macho), Lago Futalaufquen, Chubut, Argentina.

En la totalidad de los especimenes examinados (9 hembras y 10 machos), pudo visuali-

zarse claramente la presencia de un xiphisternôn inciso posteriormente, tanto para los

especimenes de Chile como para los de Argentina (fig. 6).

DISCUSION Y CONCLUSIONES

A través de anälisis de componentes principales, utilizando 16 variables morfométricas,

se obtuvieron ordenaciones de especimenes de Pleurodema thaul basadas en la variabilidad de

tamaño y forma (ACP de varianza-covarianza), en las proporciones entre las variables (ACP

de correlaciôn) y en la variabilidad en forma (ACP de Burnaby).

En los tres casos, estas ordenaciones constituyen un ünico conjunto de ejemplares dentro

del cual no es posible encontrar diferencias morfométricas entre las poblaciones estudiadas

que se correspondan con la distribucién geogräfica. En la mayor parte de los casos, la

distancia morfométrica entre ejemplares pertenecientes a una misma localidad es mayor que

aquella entre ejemplares pertenecientes a distintas localidades, lo cual no permite establecer

diferencias en la morfometria de las poblaciones.

Por otro lado, tampoco es posible establecer diferencias morfométricas entre las pobla-

ciones asignadas a Argentina Sur, Chile Central ÿ Chile Sur, que resultan de agrupar la

localidades de colecta del material teniendo en cuenta las consideraciones taxon6m

realizadas por DUELLMAN & VELOSO (1977).

Source : MNHN, Paris

ROSSET, BASSO & LOMBARDO 169

A través de MANOVA y anälisis discriminante, se puso a prueba la hipôtesis de que los

morfotipos referidos a Argentina Sur, Chile Central y Chile Sur representan entidades

morfométricamente distintas. Los resultados de estos anälisis muestran que no es posible

diferenciar a través de la morfometria estas tres agrupaciones geogräficas de P thaul. Aunque

con MANOVA se detectan diferencias estadisticamente significativas, la correcta clasificaciôn

de los especimenes en estos grupos involucra un porcentaje de error muy alto (31-39 %) como

para establecer diferencias morfométricas significativas entre ellos.

Los anälisis de agrupamientos presentan resultados similares en cuanto a que no se ha

podido establecer una correspondencia entre los grupos obtenidos del anälisis y la distribu-

ciôn geogräfica.

En este trabajo no ha sido posible el estudio de especimenes de P thaul provenientes de

las poblaciones del norte de Chile (Antofagasta, Rio Copiap6, Rio Huasco). DUELLMAN &

VELOSsO (1977) señalan la existencia de una poblacién en el desierto de Pajonales (al norte de

la provincia de Coquimbo) que podria representar un tax6n distinto. NORTHLAND et al. (1996)

muestran que las poblaciones de Antofagasta son morfométricamente mäs similares a las de

la zona central de Chile que a las de Copiapé, geogräficamente mâs cercanas.

DUuELLMAN & VELOSO (1977) utilizan la ausencia de incisin esternal como el caracter

mäs importante para diferenciar las poblaciones distribuidas en Argentina de las poblaciones

chilenas de 2 thaul. Estos autores atribuyen a las poblaciones argentinas un xiphisternôn

redondeado posteriormente, mientras que las poblaciones atribuidas a Chile presentan el

xiphisternén hendido. A través del estudio detallado de la morfologia del esternén se observé

que todos los especimenes estudiados, tanto de Argentina como de Chile, poseen el xiphis-

ternén hendido.

Los resultados morfométricos obtenidos en este trabajo, junto con los datos provenientes

de la morfologia del esternén, no permiten identificar caracteres para establecer diferencias

entre las poblaciones estudiadas de P. thaul.

RESUMEN

Se analiza la variabilidad morfométrica y anatémica de poblaciones argentinas y chile-

nas del Pleurodema thaul. Se aplicaron métodos estadisticos multivariados sobre un total de

16 variables morfométricas. Los resultados indican que las poblaciones estudiadas de P thaul

no pueden ser diferenciadas en morfotipos distintos, segün ha sido propuesto en trabajos

previos. El estudio morfolégico del esqueleto revela que el esternén (xiphisternôn) de las

poblaciones argentinas de P thaul presenta una incisiôn en su porcién posterior, simil la

observada en las poblaciones chilenas. Estos resultados no permiten sustentar la propuesta de

separar a P. thaul en entidades especific: i

Source : MNHN, Paris

170 ALYTES 19 (2-4)

AGRADECIMIENTOS

Este trabajo fue parcialmente financiado por el subsidio PIP 0788 del Consejo Nacional de

Investigaciones Cientificas y Técnicas de Argentina, otorgado a G. R. SPINELLI y N. G. BASS, y por el

PICT 01-03698 de la Agencia Nacional de Promociôn Cientifica de Argentina otorgado a N. G. BASS.

Agradecemos a G. CaRR1ZO por facilitarnos lugar de trabajo y por el préstamo de especimenes a su cargo

en el Museo Argentino de Ciencias Naturales “Bernardino Rivadavia”. H. K. VORIS y A. RESETAR n0S

facilitaron espacio y especimenes depositados en el Field Museum of Natural History.

LITERATURA CITADA

BARRIO, À. & RINALDI DE CHIERI, P., 1970. - Estudios citogenéticos sobre el género Pleurodema y sus

consecuencias evolutivas (Amphibia, Anura, Leptodactylidae). Physis, 30 (80): 309-319.

Ce, J. M. 1958. - Polimorfismo y distribuciôn geogräfica en poblaciones chilenas de Pleurodema bibroni

Tschudi. /nv. Zool. Chilenas, 4: 300-327.

_—— 1960. — La batracofauna chilena: muestra de procesos evolutivos. Actas y Trabajos del Primer

Congreso Sudamericano de Zoologia, La Plata, Argentina, 4 (9): 183-187.

1962. — Batracios de Chile. Santiago, Ediciones de la Universidad de Chile: 1-128.

1980. - Amphibians of Argentina. Monit. zool. ital., (n.s.), Mon. 2: 1- 609.

Ce, J. M. & CaPuRRO, L.S., 1957. — La distribucién de los patrones de coloraciôn en Pleurodema bibroni

en relaciôn con la distribuciôn geogräfica y el häbitat. /nv. Zool. Chilenas, 3: 156-161.

Cr, J. M. & EsPINA AGUILERA, S., 1957. — La vibraciôn sexual preventiva (“warning vibration”) en

Pleurodema chilenas. Inv. Zool. Chilenas, 4: 15-21.

CRisct, J. V. & LOPEZ ARMENGOL, M. F., 1983. — Introducciôn a la teoria y präctica de la taxonomia

numérica, Organizaciôn de los Estados Americanos, 26: 1-128.

DoNoso-BARROS, R., 1969. - Posiciôn nomenclatural de un leptodactylido uruguayo (Amphibia: Anura).

Bol. Soc. Biol. Concepciôn, 41: 161-162.

DueLLMAN, W. E. & VELoso, A. M., 1977. - Phylogeny of Pleurodema (Anura: Leptodactylidae): a

biogeographie model. Occ. Pap. Mus. nat. Hist. Univ. Kansas, 64: 1-46.

FRosT, D. R. (ed.), 1985. - Amphibian species of the world. Lawrence, Allen Press & Assoc. Syst. Coll.:

i-iv] + iv + 1-732.

GaLLARDo, J. M., 1968. — Sobre la validez de algunas especies argentinas de Pleurodema (Anura,

Leptodactylidae). Physis, 28 (76): 135-144.

LyncH, J D., 1971. - Evolutionary relationships, osteology, and zoogeography of leptodactyloid frogs

Misc. Publ., Mus. nat. Hist., Univ. Kansas, 53: 1-238.

Mc Lister, J. D., LOUGHEED, S. C. & BOGART, J. P., 1991. - Electrophoretic and vocalization compari-

sons among three leptodactylid frogs (Pleurodema spp.) from northwestern Argentina. Can. J

Zool., 69: 2397-2403.

NORTHLAND, L., CAPETILLO, J. & NUNEZ, H., 1996. - Caracterizaciôn morfolôgica de Pleurodema thaul de

poblaciones de Antofagasta (Amphibia: Leptodactylidae). 1 Congr Latinoamer. Herp., Santiago,

Chile: 241.

PARKER, H. W., 1927. — A revision of the frogs of the genera Pseudopaludicola, Physalaemus and

Pleurodema. Ann. Mag. nat. Hist., 20 (60): 450-478.

REYMENT, R. A., BLACKITH, R. E. & CAMPBELL, N. AÀ., 1984. — Multivariate morphometrics. Second

edition. London, Academic Press: 1-233.

ROHLE, F. J., 1993. NTSYS-pe. Numerical taxonomy and multivariate analysis system, Version 1.8. New

York, Exeter Software, Programa y documentaciôn.

Srarsorr, 1996. — Sratistica for Windows, Release 5.1. Tulsa, OK, USA, Computer program manual,

StatSoft, Inc.

TAYLOR, W. R. & Van DYyKE, G. C., 1985. - Revised procedures for staining and clearing small fishes and

other vertebrates for bone and cartilage study. Crbium, 9 (2): 107-119.

Source : MNHN, Paris

ROSsET, BASSO & LOMBARDO 171

VELOSO, À, GALLEGUILLOS, R. & Diaz, N., 1973. — Karyotypic analysis of allopatric populations of

Pleurodema thaul (Lesson) Amphibia, Leptodactylidac. Caryologia, 26 (1): 69-16.

VeLoso, A. & Navarro, J., 1988, — Lista sistemätica y distribuciôn geogräfica de anfbios y reptiles de

Chile. Boll. Mus. reg. Sci. nat. Torino. 6: 481-530.

APÉNDICE I

Material examinado

Los especimenes resaltados en negritas corresponden a ejemplares diafanizados y teñi-

dos para estudio de su esqueleto.

® y 3 d, MACN 28703, 28704-6, 32117, 32118, 32120; (10) Lago Läcar, 4 9, MACN (ex

CENAD) 1878-81; (13) Rio Pichi Traful, 3 ? y3 &, MACN (ex CENAI) 901, 903-7; (14) Lago

Espejo, 2 4, MACN (ex CENAI) 1052-3; (15) Isla Victoria, 4 ? y4 4, MACN (ex CENAI)

4090, 4094, 4097, 4099, 4107, MACN 9092, MLP A.532-3; (16) Puerto Blest, 17 ? y9 4,

MACN (ex CENAÏ) 1527.1-3, 1527.5, 1527.7, 1527.8, 1527.9, 1527.11, 1527.13-16, 1527.19-

20, 1956-8, 1960-3, 1967-8, 3312, MACN 31552, MLP A.425; (17) Lago Fri Sy4d,

MACN (ex CENAI) 7151-6; (18) Lago Perito Moreno, 2 ® y 1 8, MACN (ex CENAI)

2671-3; (19) Lago Nahuel Huapi, 4 ® y 2 &, MACN (ex CENAI) 1611, 2305-7, MLP

A.1079-80; (20) San Carlos de Bariloche, 2 4, MACN 28209-10; (21) Cerro Otto, 2 ?,MACN

11180, MLP A.421; (22) Monte Tronador, 3 ®, MACN (ex CENAI) 915-6, MACN 9651; (23)

Cerro Catedral, 1 9 y 1 &, MACN (ex CENAI) 8801-2; (24) Cerro Challhuaco, 1 ® y1 4,

ILPLA A.481, 488; (25) Rio Manso Superior, 4 ® y 4 4, MACN (ex CENAI) 917-23, 925;

(ex CENAI) 1097-102, 1104, 1107.6, 1107.12; (28) El Foyel, 2 ?, MACN 11435-6; (29) Rio

26144-50; (32) Cerro Puntudo, 7 ® y 7 4, MACN (ex CENAI) 8805-6, 8807, 8808-9, 8811-6,

CENAI) 7171-2, ILPLA A.479, 482, 484-6, 490; (36) Lago Situaciôn, 2 ?, MACN 29826-7.

(6) Concepcién, 4 ® y9 4, MACN (ex CENAIT) 4428-32, FMNH 214120, 214122, 214137-8,

Source : MNHN, Paris

172 ALYTES 19 (2-4)

1912-5; (8) Cordillera de Nahuelbuta, 1 9 y 1 4, MACN (ex CENAI) 1906-7; (9) Lago

212612, 212775, 212779, 212782, 212787-8, 212792; (13) Correntoso, 1 ® y 1 &, MACN (ex

CENAÏD) 1923-4; (14) Chamiza, 3 4, MACN (ex CENAI) 1951, 1952, 1953; (15) Puerto

Montt, 4 9 y4 8, MLP A.1989, 1991, 1994-5, 2000-1, 2004-5; (16) Isla de Chiloé, 4 ® y3 4,

FMNH 212596-7, 212599-602, 212613.

Corresponding editor: Alain DuBois.

Source : MNHN, Paris

Alytes, 2001, 19 (2-4): 173-182. 173

Microscopia electrônica de barrido

del aparato bucal y cavidad bucofaringea

de la larva de Atelognathus nitoi

(Anura, Leptodactulidae)

Dinorah D. ECHEVERRIA*, Carmen A. ÜBEDA** & Néstor G. BAssO***

* Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires,

Departamento de Ciencias Biolôgicas, Laboratorio de Vertebrados,

1428 Buenos Aires, Argentina

** Centro Regional Bariloche, Universidad Nacional del Comahue,

Unidad Postal Universidad, R 8400 FRF, Provincia de Rio Negro, Argentina

*#* Departamento Zoologia Vertebrados, Museo de La Plata,

Paseo del Bosque, 1900 La Plata, Argentina

Larvae of Atelognathus nitoi were collected in the Province of Rio

Negro (Argentina). The microanatomy of the oral apparatus and of the

buccopharyngeal cavity of these tadpoles is described with scanning elec-

tron microscope and observed with stereoscopic microscope. The buccal

apparatus has short labial teeth formed by a short base, a short neck, and

a multicusped paddle with 10-12 short denticles. The horny beak is wider

than high and the teeth are unicusped. The bucopharingeal cavity has two

pairs of infralabial papillae and four lingual papillae, the median ridge is

triangular, the lateral ridges are simple, the ventral velum and the glandular

zone are present at the posterior side. The intestinal contents consist of

cysts and scales of algae (Chrysophyta, Chrysophycea), frustula of diatoms

and fungi.

INTRODUCCION

Atelognathus nitoi(Barrio, 1973) es un leptodactilido microendémico de la Laguna Verde

del Cerro Challhuaco (Parque Nacional Nahuel Huapi), ubicado en el noroeste de la

patagonia argentina. De las ocho especies de Arelognathus reconocidas, hasta el presente sélo

fueron descriptas las larvas de Atelognathus patagonicus (Gallardo, 1962) (Cet, 1965), Atelo-

gnathus reverberii (Ceï, 1969) (Cet, 1969) y Arelognathus nitoi (Basso & UBEDA, 1997), y la

microanatomia de la cavidad bucofaringea de las das primeras species fue descripta por

WaASsERSUG & HEYER (1988). La larva de À. nitoi, recientemente descripta (BASSO & UBEDA,

1997), se desarrolla en la citada laguna y en ocasiones en ambientes temporarios cercanos. En

sus primeras semanas de vida, los renacuajos habitan las playas de la laguna y muestran un

häbito nadador. À medida que crecen se internan en la parte mäs profunda de la laguna,

Source : MNHN, Paris

174 ALYTES 19 (2-4)

Tabla 1. - Larvas de Atelognathus nitoi: material estudiado del Parque Nacional Nahuel

Huapi, Argentina.

N° ejemplar| 113 114 115 181 189 190 214 218

Estado | 31 26. | 31 26 | 27 | 7 | 37

Localidad sin sin sin Los

manteniendo el häbito nadador y frecuentando el fondo (Ügepa et al., 1999). Durante el

periodo premetamérfico presentan un morfotipo adaptativo léntico-béntico (Basso &

ÜBEDA, 1997).

El propôsito de este trabajo es describir el aparato bucal y la cavidad bucofaringea de 4.

nitoi mediante observaciones con microscopio electrénico de barrido, discutir sus relaciones

ecomorfolôgicas y comparar con las otras especies conocidas de Atelognathus y con géneros

afines.

MATERIAL Y MÉTODOS

Las larvas estudiadas provienen de la Laguna Verde (localidad tipo de la especie) y de

ambientes acuäticos temporarios vecinos, ubicados en un bosque de lenga (Nothofagus

pumilio) entre 1350 y 1550 m s.n.m, en el Cerro Challhuaco, Parque Nacional Nahuel Huapi,

provincia de Rio Negro, Argentina. Las larvas fueron capturardas con redes de mano y fijadas

in toto en formol 10 %. Siete ejemplares (N°113-115, 181, 189-190 y 214, Colecciôn Larvas

DDE, para Microscopio electrénico de barrido, tab. 1) corresponden a los estadios 26 al 31 de

la tabla de desarrollo de GOsNER (1960). Se disecé un ejemplar en estadio 37 (ejemplar N°218,

lote 206, Comahue, tab. 1) para comparar el desarrollo de la papilaciôn bucofaringea con los

estadios menores.

Las observaciones en microscopio electrénico de barrido se realizaron tratando al

material segün las técnicas propuestas por WaASsERSUG (1980) y ECHEVERRIA (1995), efectuän-

dose una deshidratacién räpida en alcohol etilico (80 %, 8 h;96 %, 2 h; 100 %, 1 h),en alcohol

etilico-acetona (3:1, 1 h; 1:1, 1 h; 1:3, 10h) y en acetona pura, como mäximo dos horas antes

de proceder a la desecaciôn final. Se efectué el punto critico de desecacién en un secador al

vacio Balzers 030. La metalizacién se realizé con oro en un metalizador ION Sputtering

Balzers CPD 040. Las observaciones y fotografias se realizaron en un microscopio electrénico

de barrido Philips 505. El punto critico de desecaciôn se efectué en un secador al vacio Balzers

CPD 030. La metalizaciôn se efectu6 con oro en un metalizador Balzers SCD 040. Las

descripciones de las formaciones côrneas del aparato bucal y cavidad bucofaringea se

realizaron segün la terminologia propuesta por DEUNFF & BEAUMONT (1959), VIERTEL (1982)

y McDiarmiD & ALTIG (1999). La formula dentaria se expresé siguiendo la propuesta de

Dugois (1995).

Source : MNHN, Paris

ECHEVERRA, ÜBEDA & Basso 175

El contenido intestinal de las larvas fue examinado en diferentes estadios de desarrollo

(26, 27 y 31) a fin de relevar en forma cualitativa los diferentes items alimentarios y sus

caracteristicas ecolégicas.

Las ilustraciones fueron realizadas con un microscopio estereoscépico Nikon SMZ 10

TD, provisto de câmara de dibujo.

RESULTADOS

APARATO BUCAL

El disco oral es emarginado y con un amplio claro rostral. Estä limitado lateralmente por

una hilera de papilas marginales cénicas y simples y por papilas submarginales similares, bien

desarrolladas, en las regiones supraangular e infraangular (fig. 1a-b). A nivel del cartilago de

Meckel pueden observarse 2 papilas angulares intramarginales bajas (fig. 1c). Las papilas

mentales son cénicas, alargadas, con el extremo agudo, extendiéndose en una hilera simple

(fig. Id).

Los queratodontes forman una hilera continua y homogénea; el nümero de dientes en un

ejemplar correspondiente al estadio 31 es de aproximadamente 5 a 7 en 100 um. Estän

formados por una base, un cuello y una espätula con 10 a 12 denticulos cortos. La longitud

total de un diente labial es 25 a 30 4m, con un ancho mäximo de la espätula de 15 um (fig. le).

Los dientes del pico cérneo estän organizados en empalizada, con una densidad de 5 dientes

en 100 ym; presentan un largo total de aproximadamente 25 a 30 um y un ancho mäximo de

20 um (fig. 1f).

CAVIDAD BUCOFARINGEA

Piso

La regiôn prelingual presenta cuatro papilas infralabiales, dos ventrales y dos laterales.

Sobre el esbozo lingual se desarrollan cuatro papilas linguales simples; las papilas centrales

son mäs largas y mâs cercanas al borde lingual anterior que las laterales (fig. 2a). La arena del

piso de la cavidad oral estä limitada lateral y posteriormente por papilas periféricas bien

desarrolladas. En el estadio 31 las papilas son mäs abundantes que en los estadios 26 y 27 (fig.

2b).

son alargadas y estän ubicadas transversalmente con res]

interna de la prehendidura estä precedida por proye:

Éstas presentan los bordes rugosos y se destacan del resto de las papilas por su mayor

desarrollo. Cercanas al borde anterior de la hendidura se observan 1 a 3 papilas bajas (fig. 2b).

El velo ventral tiene 4 a 5 proyecciones marginales pronunciadas y espaciadas à ambos lados

de la escotadura mediana. Las proyecciones que forman la escotadura son pronunciadas y le

Source : MNHN, Paris

176 ALYTES 19 (2-4)

Fig, LL — Aparato bucal de Atelognathus mto (stadio 31) (a) Vista general. Escala: 1 mm. (b) Detalle de

las papilas del disco oral. Escala: 0,1 mm. (c) Detalle de la

disco oral. Escala: 0,1 mm. (d) Papilas mentales del disco oral. À, papilas mental

) Dientes labiales en la hilera A-2. Escala: 10 um. (1) Pico côrneo. Escala: 10 um.

Source : MNHN, Paris

ECHEVERRA, Ügepa & BASsO 177

Fig. 2.- Cavidad bucofaringea de Atelognathus nitoi. (a) Piso, area prelingual. I, papilas infralabiales: L,

papilas linguales. Estadio 27. Escala: 0,1 mm. (b) Region posterior del piso. H, hendidura bucal

derecha: P, papilas periféricas: $, tres papilas en el borde anterior de H. Estadio 31. Escala: 1 mm. (c)

Techo. C, coana; LP, pliegue lateral; M, pliègue mediano; N, papila postnasal. Estadio 26. Escala:

0,1 mm. (d) Regiôn glandular posterior del techo. Estadio 31. Escala: 0,1 mm.

confieren a ésta forma de “V”. El borde del velo y sus proyecciones presentan fosetas

glandulares.

Cabe destacar que en un ejemplar en estadio 37 (de 73 mm de longitud total), las papilas

infralabiales laterales presentaron bordes irregulares y las papilas linguales centrales se

hallaron ramificadas.

Techo

Las coanas son de forma eliptica, convergentes anteromedialmente, protruyentes hacia

la cavidad bucofaringea. En el ärea posnarial, en estadios 26 y 27, se halla un par de papilas

postnasales (fig. 2c). Los pliegues laterales son proyecciones simples, bien desarrolladas, con

Source : MNHN, Paris

178 ALYTES 19 (2-4)

A B

Fig. 3. - Morfologia de la cavidad bucofaringea de la larva de Arelognathus nitoi (estadio 31):

(A) piso; (B) techo. C, coana; D, papilas prehendidura; E, escotadura mediana; F, proyecciôn

marginal; G, regiôn glandular; H, hendidura bucal; 1, papila infralabial; L, papilas linguales; LP,

pliegue lateral; M, pliegue mediano: N, papila postnasal; P, papilas periféricas del ârea del piso de la

cavidad bucofaringea; T, papilas del margen del ärea del techo de la cavidad bucofaringea; V, velo.

forma de lämina comprimida en sentido anteroposterior; su borde ventral es irregular y el

dorsal liso. El pliegue mediano tiene forma triangular, paredes lisas y bordes irregulares. Las

papilas marginales de la arena del techo de la boca son escasas (de 5 a 6 pares), conicas y altas;

en la regiôn posterior se hallan numerosas püstulas. El ärea glandular estä muy desarrollada

se dispone en forma de banda en U abierta en los extremos:; las fosetas glandulares presentan

cular o irregular (fig. 2d). En el ejemplar 218, se hallaron dos pares de papilas

postnasales de diferente tamaño, siendo el par mäs externo de tamaño muy reducido.

En la figura 3 se reünen las caracteristicas tipicas de la cavidad bucofaringea de A. nitoi

para el estadio 31.

forma ci

CONTENIDO INTESTINAL

s correspondientes a los estadios 26 y 27 (ejemplares 181 y 189),

amas de algas (Chrysophyta,

um en los quistes y entre 10 a

En dos ejempl

provenientes de la Laguna Verde, se hallaron quistes y

Chrysophycea) cuyo tamaño de particula oscilé entre 3 a

Source : MNHN, Paris

ECHEVERRA, ÜBEDA & BASSO 179

12 um en las espiculas. Se hallaron früstulos de diatomeas (Bacillariophyceae, Diploneis sp.,

Navicula sp, Cymbella sp.) de largo mäâximo hasta 60 um. Se observé, ademäs, la presencia de

hifas de hongos acuäticos (ejemplares 115, 181 y 189).

DISCUSIÔN Y CONCLUSIONES

La larva de Atelognathus nitoi presenta una distribucién de hileras de queratodontes que

responde a la férmula 1:<1>/<1>:2, tal como fuera observado por Basso & ÜgrpA (1997).

Este tipo de distribucién de los denticulos cérneos es muy frecuente entre las larvas de aguas

quietas (ALTIG & JOHNSTON, 1989) y es compartida por la mayoria de los leptodactilidos

telmatobiinos patagénicos: A/sodes barrioi, A. gargola, A. monticola, Atelognathus patagoni-

cus, À. reverberii, Batrachyla antartandica, B. leptopus, B. taeniata, Caudiverbera caudiverbera

y Hylorina sylvatica (FORMAS & PUGIN, 1978a-b; LAviLLA, 1988), con la excepciôn del género

Eupsophus, que presenta una reduccién en el nümero de hileras de queratodontes asociada a

una alimentaciôn endotrôfica (FORMAS & PUGIN, 1978a-b; FORMAS, 1989u-b, 1992). Los

queratodontes son relativamente cortos en relacién con los dientes de otros leptodactilidos de

ambientes lénticos conocidos, como los observados en Odontophrynus americanus y Lepto-

dactylus ocellatus (ECHEVERRIA & MONTANELLI, 1992; ECHEVERRIA, 1995).

En cuanto a la papilacién del disco, 4. nitoi presenta escasas papilas suprangulares e

infrangulares con una distribuciôn irregular, y papilas mentales. Las papilas marginales

mentales son alargadas y se disponen en una hilera limitando al disco por su parte posterior,

confirmando lo indicado por LAvILLA (1988) para el género Atelognathus. El disco oral de À.

nitoi se distingue del de A/sodes gargola por presentar una hilera de papilas mentales

marginales, a diferencia del género A/sodes, que tiene dos hileras mentales, una marginal y

otra intramarginal, cada una con distinta morfologia (ECHEVERRiA et al., 2001). En la zona

angular intramarginal de 4. nitoi se observan dos papilas poco desarrolladas ubicadas a nivel

del cartilago de Meckel. Éstas son las ünicas papilas en la zona angular que se hallan en

posiciôn intramarginal, aisladas y ubicadas mäs internamente que las restantes papilas

submarginales.

La cavidad bucofaringea presenta caract icas anatémicas comunes con otros lepto-

dactilidos. Atelognathus nitoi comparte con las larvas de los leptodactilidos de los géneros

Alsodes, Atelognathus, Batrachyla, Caudiverbera, Hylorina, Pleurodema y Odontophrynus

(BRIEVA VASQUEZ, 1988; WASSERSUG & HEYER, 1988) la presencia de cuatro papilas linguales

y cuatro papilas infralabiales. Cabe destacar que las caracteristicas atribuidas a Eupsophus

roseus por WASSERSUG & HEYER (1988) no son consideradas en este trabajo debido a la

identidad dudosa del espécimen KU 162057, inferida a partir de las notas de campo sobre el

häbitat, la coloraciôn y el tamaño del ejemplar. La morfologia de la cavidad bucofaringea de

Atelognathus nitoi coincide con la observada en À. patagonicus y À. reverberii (Was: G&

HEYER, 1988) en cuanto a la presencia de cuatro papilas infralabiales, cuatro papilas linguales,

un pliegue mediano triangular y pliegues laterales simples y bien desarrollados. Comparte con

Leptodactylus ocellatus euatro papilas linguales ÿ la forma y textura de los bordes dorsal y

ventral del pliegue lateral (ECHEVERRIA, 1995).

Source : MNHN, Paris

180 ALYTES 19 (2-4)

Los telmatobinos patagénicos que poseen la misma férmula dentaria que 4. nitoi pueden

hallarse en ambientes lôticos, lénticos, temporarios o permanentes. Sus caracteristicas eco-

morfolégicas también son variadas, observandose formas benténicas y necténicas (CEt,

1980).

Al analizar el contenido intestinal de À. nitoi, se hallé un rango muy amplio de items

alimentarios (algas y hongos) con tamaños de particula comprendidos entre 3 y 60 um. De la

observaciôn con microscopio electrénico de barrido se amplian los items alimentarios

previamente reportados por ÜBEDA et al. (1999), confirmändose la predominancia de algas

plancténicas y perifiticas en el contenido intestinal de 4. nitoi.

La morfologia del aparato oral y de la cavidad bucofaringea observada en Atelognathus

nitoi presenta caracteristicas coincidentes con las reportadas para otras especies del género

Atelognathus. Los caracteres compartidos con otros géneros de telmatobinos patagénicos, de

häbitos diferentes, deben atribuirse fundamentalmente a una restricciôn filogenética dada por

ancestralidad comün, prevaleciendo sobre las adaptaciones ecolégicas convergentes.

RESUMEN

Se describe la microanatomia del aparato bucal y la cavidad bucofaringea de la larva de

Atelognathus nitoi observada con microscopio electrénico de barrido y microscopio estereos-

cépico sobre especimenes colectados en la provincia de Rio Negro (Argentina). El aparato

bucal presenta dientes labiales relativamente cortos formados por una base y cuello cortos y

una espätula multicuspidada, con 10 a 12 denticulos cortos. El pico cérneo es mäs ancho que

alto y los rostrodontos estän organizados en empalizada, son unicuspidados con el extremo

liso y agudo. La cavidad bucofaringea tiene dos pares de papilas infralabiales, cuatro papilas

linguales, el pliegue mediano es triangular, los pliegues laterales son proyecciones simples,

velo ventral y ärea glandular en la parte posterior. En el contenido intestinal se encontraron

principalmente quistes y escamas de algas (Chrysophyta, Chrysophysea), früstulos de diato-

meas e hifas de hongos.

AGRADECIMIENTOS

A la Administraciôn de Parques Nacionales de Argentina por autorizar la realizaciôn de trabajos de

campo en el Parque y Reserva Nacional Nahuel Huapi. A V. Conforti por asesorarnos en la determina-

ciôn de las algas del contenido del tubo digestivo. À L. Trueb por permitirnos el acceso a las notas de

campo relacionadas con el ejemplar KU 162057. À D. Jiménez del Servicio de Microscopia Electrénica

de Barrido dependiente del Instituto de Investigaciones Cientificas y Técnicas de las Fuerzas Armadas

(CITEFA). Este estudio fue financiado en parte con un subsidio para la investigaciôn otorgado por la

Agencia Nacional de Promociôn Cientifica y Tecnolôgica (ANPCYT) de Argentina (BID 1201 /OC-AR,

PICT 01-03698).

Source : MNHN, Paris

ECHEVERRA, ÜBEDA & Basso 181

LITERATURA CITADA

ALTIG, R. & JoHNsToN, G. F,, 1989. — Guilds of anuran larvae: relationships among developmental

modes, morphologies, and habitats. Herp. Mon., 3: 81-109.

Basso, N. G. & Usepa, C. A., 1997. — The tadpole of Atelognathus nitoi (Leptodactylidae, Telmatobii-

nae). Alytes, 15 (3): 105-136.

BARRIO, À. 1973. — Una nueva especie de Telmatobius (Anura, Leptodactylidae) procedente del dominio

austral cordillerano argentino. Physis, (C), 32 (84): 207-213.

BRiEVA VASQUEz, L. M., 1988. - Andlisis morfolégico comparativo de las larvas de las especies del género

Batrachyla (Anura: Leptodactylidae). Tesis Magister, Universidad Austral de Chile, Valdivia:

1-116.

Cri, J. M., 1965. - The tadpole of Batrachophrynus patagonicus Gallardo. Herpetologica, 20 (4): 242-245.

PRES 1969. - The Patagonian telmatobiid fauna of the volcanic Somuncura plateau of Argentina. J Herp.,

3(1-2): 1-18.

---— 1980. - Amphibians of Argentina. Monit. zool. ital., (n. s.), Mon. 2: [i-xii] + 1-609.

DEUNFF, J. & BEAUMONT, A. 1959. - Histogénèse des dents et du bec cornés chez les larves de Discoglossus

pictus Otth. Cr Soc. Biol., 153: 1162-1164.

Dugois, A., 1995. - Keratodont formulae in anuran tadpoles: proposals for a standardization. J. zool.

Syst. Evol. Res., 33 (1): i-xv.

EcHeveria, D. D., 1995. - Microscopia electrénica de barrido del aparato bucal y de la cavidad oral de

la larva de Leptodactylus ocellatus (Linnaeus, 1758) (Anura, Leptodactylidae). A/vres, 12 (4):

159-168.

ECHEVERRIA, D. D. & MONTANELLI, S. B., 1995. - Acerca del aparato bucal y de las férmulas dentarias de

Odontophrynus americanus (Duméril y Bibron, 1841) (Anura, Leptodactylidac). Physis, (B),

“1992”, 50 (118-119): 37-43. )

ECHEVERRIA, D. D., ALONSO, C. A. PILLADO, M.S., & ÜBrDa, C. A. 2001. - Microanatomia del aparato

bucal y cavidad bucofaringea de la larva de Alsodes gargola Gallardo, 1970 (Leptodactylidae, Tel-

matobiinae). Alytes, 18 (3-4): 187-200.

FoRMAS, J. R., 19894. — À new species of Eupsophus (Amphibia: Anura: Leptodactylidae) from Southern

Chile. Proc. biol. Soc. Wash., 102 (3): 568-576.

1989b. — The tadpole of Eupsophus calcaratus in Southern Chile. J Herp., 23 (2): 195-197.

1992. — The tadpole of Eupsophus vertebralis (Anura: Leptodactylidae). Herpetologica, 48 (1):

115-119.

ForMas, J. R. & PuGin, E., 1978a. - Tadpoles of Hylorina sylvatica, Eupsophus vittatus, and Bufo

rubropunctatus in southern Chile. Herpetologica, 34 (4): 355-358.

a — 1978b. - Tadpoles of Eupsophus roseus and Bufo variegatus (Amphibia, Anura) in southern Chile. J

Herp., 12 (2): 243-246.

GALLARDO, J. M. 1962. - Los géneros Telmatobius y Batrachophrynus (Amphibia, Leptodactylidae) en la

Argentina, Neorropica, 8 (26): 45-58.

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

identification. Herpetologica, 16: 183-190.

larval characters. Occ. Pap. Mus. nat. Hist. Univ. Kansas, 124: 1-19.

McDiarMiD, R. W. & ALTIG, R., 1999. - Tudpoles. The biology of anuran larvae. Chicago & London, The

University of Chicago Press: [i-xiv] + 1-444.

UneDa, C., ZAGARESE, H., Diaz, M. & PEDROZO, F., 1999. — First steps towards the conservation of the

microendemic Patagonian frog Atelognathus nitoi. Oryx, 33 (1): 59-66.

Vire, B. 1982. - The oral cavities of central European anuran larvae (Amphibia). Morphology.

ontogenesis and generic diagnosis. Amphibia-Reptilia, 4: 327-360.

Source : MNHN, Paris

182 ALYTES 19 (2-4)

WassersuG, R. J, 1980. — Internal oral features of larvae from eight anuran families: functional,

systematic, evolutionary and ecological considerations. Unix. Kansas Mus. nat. Hist. mise. Publ.,

68: 1-146.

WassESUG, R. J. & HEYER, W. R., 1988. — A survey of internal oral features of leptodactyloid larvae

(Amphibia: Anura). Smithsonian Contr. Zool., 457: 1-99.

Corresponding editor: Esteban O. LAVILLA.

Source : MNHN, Paris

Alytes, 2001, 19 (2-4): 183-195. 183

Cuidado parental y comportamiento

de cardumen de larvas

en Leptodactylus insularum

nura, Leptodactylidae)

Maria Laura PONSSA

Instituto de Herpetologia, Fundaciôn Miguel Lillo,

Facultad de Ciencias Naturales e Instituto Miguel Lillo,

Miguel Lillo 251, T4000JFE, San Miguel de Tucumän, Argentina

<mlponssa@arnet.com.ar>

Parental care of a similar nature has been noted in three species of

Leptodactylus from two groups (ocellatus group: L. ocellatus and L.

insularum; melanonotus group: L. validus). In these species, the female

remains with the foam nest throughout development and then accompanies

a school of tadpoles until metamorphosis. In addition, in L. insularum,

tadpoles form densely packed schools of hundreds or thousands of individ-

uals reminiscent of many Bufo species. In a two month study in Gamboa,

Panamä, 1 examined whether parental care in L. insularum involved

aggressive behavior and whether schooling behavior of tadpoles was stimu-

lated by high densities and predators. Adult behavior was monitored dai

nocturnally and diurnally, throughout the study. 1 performed two expe:

ments with tadpoles. First, 1 tested whether tadpoles at different densities

had different likelihood to form schools. Second, tadpoles were tested at

two densities, with and without predators to see their schooling response. I

found evidence of parental care, noting several schools of tadpoles with

attendant adults. Furthermore, aggressive behavior by adults, including

vocalizations, in the presence of tadpoles was noted on three occasions. In

tadpoles the formation of schools was dependent on density. In addition

insect predators acted as a stimulus to school formation in experiments at

two tadpole densities.

INTRODUCCION

El término “cuidado parental” fue introducido por TRIvERS (1972), quien lo defini

como “cualquier inversién por parte del progenitor en una prole determinada, que incre-

menta la oportunidad de supervivencia de la prole (y por lo tanto el éxito reproductivo) al

costo de la capacidad del progenitor para invertir en otra prole”, Desde entonces los patrones

de comportamiento asociados a este término fueron definidos y analizados por numerosos

autores quienes consideraron la relacién entre fertilizaciôn externa e interna-cuidado por

macho y/o hembra, y sus ventajas y desventajas para los padres y las crias (SMrrH, 1977;

Source : MNHN, Paris

184 ALYTES 19 (2-4)

WELLS, 1977; Gross & SHINE, 1981; WITTENBERGER, 1981; SIMON, 1983; TOWNSEND et al.,

1984; Gross & SARGENT, 1985).

Existen pocos estudios sobre el cuidado parental en anfibios en comparaciôn a aves y

mamiferos. Este comportamiento se presenta en los tres érdenes de anfibios. Aunque sélo fue

reportado para el 6 % de las especies de anuros, se distribuye en diversas familias (CRUMP,

1996). CruMP (1996) distingue seis modos de cuidado parental, de los cuales las especies del

genero Leptodactylus representarian ejemplos de cuidado de huevos y cuidado de las larvas.

Algunas funciones propuestas para el cuidado parental de huevos y larvas de anuros son:

proteccién contra patégenos (especialmente hongos) y predadores; aireacién de los huevos

acuâticos; hidratacién de los huevos terrestres; prevenciôn de anormalidades en el desarrollo

y de canibalismo (SIMON, 1983; CRuMP, 1996). Se postula que el beneficio del cuidado

parental es incrementar la supervivencia de la prole. Entre los costos para el vigilante se

considera el aumento de la vulnerabilidad a la predacién, menor calidad y cantidad del

alimento ingerido, y reduccién de las oportunidades para aparearse (SIMON, 1983; TOWNSEND

et al.,1984; CRUMP, 1996).

Una caracteristica comün de los miembros del genero Leptodactylus es la oviposicién en

una masa de espuma ya sea en la superficie del agua, o en câmaras incubatrices. Existen

reportes de cuidado parental en especies del grupo ocellatus del genero Leptodactylus (HEYER,

1969): L. chaquensis (DE ALMEIDA PRADO et al., 2000), L. ocellatus (VAZ-FERREIRA &

GEHRAU, 1975) y L. insularum (citada como L. bolivianus en WeLLs & BARD, 1988). En los dos

ültimos casos el comportamiento de cuidado parental es similar. La hembra permanece junto

al nido de espuma durante el desarrollo de los huevos y luego con el cardumen de larvas, dia

y noche hasta la metamorfosis. En L. ocellatus, VAZ-FERREIRA & GEHRAU (1975) registraron

sonidos producidos por las larvas, que explicarian la posibilidad que tienen los renacuajos de

detectar la presencia del grupo y orientarse hacia él, y podrian servir a la madre para guiarse

hacia el cardumen aûn estando a varios metros de distancia. Estos autores también registra-

ron ataques de la madre y otros adultos a potenciales predadores en el ârea donde estaban los

nidos o los renacuajos. WELLS & BARD (1988) determinaron en L. insularum un despliegue de

movimientos que representaria una inusual y compleja forma de comunicacién (fisica,

quimica, o ambas) que permitiria a la hembra guiar a sus larvas. Movimientos de “bombeo”

similares a los reportados para L. insularum (WELLS & BARD, 1988), también fueron obser-

vados en especies del grupo melanonotus: L. validus (DOWNIE, 1996), L. leptodactyloides

(Cocrorr & MORALES, comunicacién personal) y en L. podicipinus (MARTINS, 1996). Otros

miembros del genero con cuidado parental son L. colombiensis (ESTRADA, comunicaciôn

personal), L. fallax (LESCURE, 1979; LESCURE & LETELLIER, 1983) y L. fuscus (LESCURE, 1973).

En los dos ültimos casos se determiné que la hembra permanece junto a la puesta de huevos.

En este trabajo se estudié el comportamiento de cuidado parental en L. insularum, bajo la

hipôtesis de que el cuidado parental en esta especie implica un comportamiento agresivo ante

posibles predadores.

La formacién de cardûmenes (grupos relativamente estacionarios de cientos o miles de

individuos), es una caracteristica distintiva de ciertas larvas de anuros. Existen numerosas

discusiones sobre si estos agrupamientos son équivalentes a los cardümenes de peces. Was-

SERSUG et al. (1981) distinguen entre los cardümenes de renacuajos y peces en que los primeros

tienden a ser relativamente estacionarios y las distancias entre vecinos son azarosas. Algunos

Source : MNHN, Paris

PONSSA 185

autores, en un intento de definir y caracterizar los cardümenes (de peces o larvas de anfibios),

han distinguido entre distintos tipos de agrupamientos de acuerdo a sus caracteristicas y al

estimulo que los desencadenan (DUELLMAN & LESCURE, 1973; WASSERSUG, 1973; BEISWEN-

GER, 1977; CALDWELL, 1989).

Ademäs de los ya mencionados L. insularum y L. ocellatus, existen numerosos ejemplos

de comportamiento de cardumen en larvas de anuros (WASSERSUG & HESSLER, 1971; BEISWEN-

GER, 1975, 1977, 1981; WaLDMAN, 1981; BREDEN et al., 1982; KEHR, 1994). BRANCH (1983)

estudi6 la relacién espacial, patrones de respiracién y comportamiento de alimentacién en

cardümenes de larvas de Phyllomedusa vaillanti; ALTIG & CHRISTENSEN (1981) estudiaron el

comportamiento de cardûmenes de Rana heckscheri. CALDWELL (1989) describié tres modos

de comportamiento de cardumen en larvas de Hyla geographica. Esta autora determiné que

las larvas de L. insularum y L. ocellatus forman cardûmenes moderadamente polarizados en

bordes sombreados de charcos o en el fondo de charcos, similar a los formados por las larvas

de Bufo. Se postula que la presencia de otras larvas estimula la formacién de estas agregacio-

nes en larvas de anuros (DUELLMAN & LESCURE, 1973; BEISWENGER, 1981). Algunos autores

han hipotetizado sobre las posibles funciones de termoregulaciôn y defensa de los cardüme-

nes (BREDER, 1967; O’HaRA & BLAUSTEIN, 1981; CALDWELL, 1989; DE Viro et al., 1999).

Como defensa, actuarian confundiendo al predador en la selecciôn de una presa, especial-

mente cuando los predadores son insectos acuâticos (KEHR, 1994): de esta manera las

agregaciones aumentarian la supervivencia de la puesta. En este estudio se plantearon los

objetivos de determinar si la formacién de cardümenes en L. insularum depende de la

densidad larval (‘“estimulo social”) y/o es estimulada por la presencia de predadores.

Existen distintas hipôtesis sobre el mecanismo y funciôn del cuidado parental en anfibios

y sobre el comportamiento de cardumen. Las hipôtesis planteadas en este trabajo son: (1) el

cuidado parental en L. insularum implica un comportamiento agresivo ante posibles preda-

dores; (2) la formacién de cardümenes de larvas en L. insularum depende de la densidad

larval; (3) la presencia de predadores acuâticos es un estimulo para la formaciôn de cardü-

menes de larvas en L. insularum. Para el estudio del comportamiento de cuidado parental se

realizaron observaciones en el campo, y las hipôtesis sobre cardumen de larvas se pusieron a

prueba de manera experimental.

MATERIALES Y MÉTODOS

SISTEMA DE ESTUDIO

El ärea de estudio fue Kent's Marsh, ubicado en Gamboa, Panamä (9°7°30"N; 79°42°O),

el cual es un charco temporario, de aproximadamente 430 X 45 m, cubierto con vegetaciôn de

tipo pastizal, la cual era cortada periédicamente. En la zona central del charco, donde se

registré la mayor diversidad de anuros, habia un mayor estancamiento de agua. Este charco

fue seleccionado porque habia mayor actividad de L. insularum que en otros cuerpos de agua

del äârea. Leptodactylus insularum es una especie de tamaño relativamente grande (largo

hocico-cloaca medio: hembra 86.16 mm, macho 85.6 mm), sin embargo dificil de loc.

que estä cripticamente coloreada y se confunde con la vegetaciôn caida circundante.

Source : MNHN, Paris

186 ALYTES 19 (2-4)

El estudio fue realizado durante la época Iluviosa, desde fines de junio a mediados de

septiembre de 1998.

En el ârea se determinaron cinco sectores de igual superficie, los cuales fueron monito-

reados durante el dia y la noche. El monitoreo se realizé simultäneamente mediante encuen-

tros visuales (CRuMP & Scorr, 1994), transectas acüsticas (ZIMMERMAN, 1994) y muestreo de

transectas (JAEGER, 1994). Durante el dia ademäs de los métodos anteriores se realizé el

reconocimiento de los sitios de puesta (Scorr & WoopwaRD, 1994). Los sitios donde estaban

los nidos fueron marcados con cintas de color atadas a la vegetaciôn circundante.

CUIDADO PARENTAL

Cuando un adulto estuvo presente junto al nido o cardumen, se determiné la reacciôn del

adulto cuando el investigador aproximaba la red de colectar larvas (considerado como

hipotético predador) al nido o cardumen. Se consideraron tres tipos de reacciones: escape;

agresién; no-agresiôn sin escape. La red fue colocada, en distintas ocasiones, junto al nido o

cardumen, a 50 cm, a 1 m de distancia. Las observaciones se realizaron en el campo.

COMPORTAMIENTO LARVAL

&La formaciôn de cardümenes depende de la densidad larval”

Para determinar si la formaciôn de cardmenes depende de la densidad larval, se colecté

un cardumen de larvas del ârea de estudio. Este cardumen se dividié en grupos de 200, 100, 50

y 25 larvas que se colocaron en recipientes idénticos, de material plästico, de 35 cm de

diâmetro, en los que se colocaron 1,5 1 de agua. Cuando se intenté colocar larvas a densidades

mäs bajas (10 larvas por litro de agua) se observé una alta tasa de mortalidad, lo que obligé

a interrumpir los experimentos reiteradas veces. Para descartar la posibilidad de que la

calidad del agua fuera responsable de la mortandad se colocaron las larvas en agua de Iluvia

y después en agua corriente estacionada en recipientes un tiempo suficiente para que se

evapore el cloro que pudiera tener, en ambos casos las larvas murieron. Finalmente, para los

experimentos se opt6 por utilizar agua del charco donde eran colectadas las larvas diluida con

agua corriente estacionada y no realizar tratamientos a tan bajas densidades.

Los experimentos se realizaron en laboratorio, y consistieron en determinar si habia

formacién de cardumen en los diferentes tratamientos. Se consideré como cardumen a los

agrupamientos de larvas (larvas contiguas unas a otras) y a estos se los consider en cinco

categorias: (0) 0 % de larvas agrupadas; (1) hasta 10 %; (2) hasta 25 %; (3) hasta 50 %; (4)

hasta 100 %. Durante una hora se registré cada 10 minutos presencia/ausencia y categoria de

agrupamiento en el instante en que se observaban los recipientes, y después una vez cada hora

por 8 horas mäs.

Los datos se analizaron con el test de Kruskal-Wallis, ya que no estaban distribuidos

normalmente y no presentaban varianzas homogéneas. Se realizé a posteriori un test de Dunn

para determinar entre que pares de tratamientos habia o no diferencias significativas.

Source : MNHN, Paris

PONSSA 187

<La formaciôn de cardumen es estimulada por la presencia de predador acuätico?

Para responder este punto se realizaron pruebas de palatabilidad a fin de seleccionar un

buen predador de larvas de anfbios. Estos experimentos consistieron en colocar a los

renacuajos en recipientes con potenciales predadores (coleépteros acuäticos y larvas acuâti-

cas de insectos) y seleccionar a los predadores que atacaban mâs râpido y comian mäs

räpidamente a las larvas de anuros. Los mejores predadores fueron larvas de Hydrophylidae

(Coleoptera) (longitud media: 27,36 mm), que eran comunes en el charco donde estaban los

renacuajos. Estas larvas atacaban râpidamente y sostenian con su aparato bucal al renacuajo

mientras lo ingerian en corto tiempo. Los tratamientos consistieron en recipientes idénticos,

con igual densidad de larvas de un mismo cardumen, con predador (una larva de Hydrophy-

lidae en cada tratamiento) y sin predador. Cada cinco minutos, durante dos horas, se

determiné si habia formacién de cardumen en cada tratamiento. Las categorias de agrupa-

mientos consideradas fueron: (0) sin agrupamiento: (1) < 5 larvas; (2) < 20 larvas; (3) < 50

larvas: (4) < 100 larvas; (5) < 200 larvas; (6) < 300 larvas. No hubo reposiciôn de los renacuajos

ingeridos por el predador, ya que en el tiempo que duraron los experimentos, esta predaciôn

no hizo variar considerablemente la cantidad de renacuajos.

Se realizaron dos experimentos independientes. En uno de ellos fueron colocadas 300

larvas por recipiente, y se hicieron tres réplicas por tratamiento (con predador y sin predador).

Estas larvas pertenecian a un cardumen que habia eclosionado cinco dias antes, y en el que se

contabilizaron 2014 larvas. En el otro experimento se colocaron 100 larvas por recipiente, se

realizaron dos réplicas por tratamiento (con predador y sin predador). En este cardumen se

contabilizaron 481 larvas, en estadios 29-37 de la tabla de GOsNER (1960). Como control se

empleé larvas de Physalaemus pustulosus, de estadios 29-31 de la tabla de GOSNER (1960),

cuyos nidos fueron colectados en la misma zona que los de L. insularum. El control fue

empleado para comparar el comportamiento de los renacuajos cuando forman cardumen de

cuando no lo hacen, ya que las larvas de P pustulosus colocadas a la misma densidad, y en

presencia de los mismos predadores nadaban independientemente, sin colocarse contiguas

unas de otras o demostrar algün tipo de agrupamiento.

Los datos se analizaron con un test de independencia (chi cuadrado) entre grupos (con

depredador / sin depredador) y respuestas (con agrupaciôn / sin agrupacién).

RESULTADOS

De los 31 dias que estuvo presente L. insularum, 25 estuvieron en la Zona central del ärea

de estudio, la cual, a diferencia de las otras zonas, tuvo agua estancada durante todo el

periodo de estudio. En ella se registré la mayor cantidad de especies de anuros (hasta 12

especies), y presentaba mayor heterogeneidad, en cuanto a la fisonomia de la vegetaciôn, que

el resto del ärea. Los nidos y cardümenes también se encontraron en esta zona. Se registraron

un total de 8 nidos y 7 cardümenes. Los nidos observados tenian forma de corona con un

orificio en el centro. Los nidos, al igual que los adultos, estaban generalmente en äreas con

vegetaciôn alta (vegetaciôn de altura menor o igual a 1.6 m), tipo pastizal. Fue muy comün

encontrar los nidos sucesivos en los mismos sitios en donde habian otros anteriormente.

Source : MNHN, Paris

188 ALYTES 19 (2-4)

CUIDADO PARENTAL

Junto a algunos cardümenes habia adultos, los cuales huian ante la presencia del

investigador (7 = 2) o demostraban un comportamiento agresivo (n = 3). El comportamiento

agresivo fue registrado en tres ocasiones, el 30 de Julio, el 12 y el 25 de Agosto de 1998. No se

pudo determinar el sexo de estos adultos, o si se trataba del mismo o diferentes individuos. El

comportamiento agresivo fue observado en dos ocasiones durante el dia y en una durante la

noche. En los tres casos este despliegue consistia en que cuando se simulaba colectar larvas

con una red (colocando la red sobre las larvas), el adulto saltaba instantäneamente sobre la

red, mordiendo, y en algunos casos emitiendo un grito o gruñido. El adulto luego saltaba lejos

de ella (menos de 1 m), quedando de espaldas al cardumen y a la red. Al colocar de nuevo la

red en contacto con el agua en la Zona donde estaban las larvas, se volteaba y volvia a saltar

sobre la red. No era necesario mover la red o agitar con ella el agua para que el adulto

reaccionara atacando inmediatamente, Este comportamiento se repitié sucesivamente hasta

por dos horas en cada adulto (7 = 20 a 25 en cada una de las tres ocasiones).

En dos de las tres ocasiones en que se registré el comportamiento agresivo, se colocé

luego la red a una mayor distancia (aproximadamente a 1 m de un individuo en una ocasiôn

y a 50 cm en otra ocasiôn). En estos casos el ataque no fue instantäneo, tardando de 20 a 40

segundos, pero hubo agresiôn en el 70 % de las veces en que se colocé la red a 50 cm (n = 20),

y en el 66.6 % cuando se la colocé a 1 m (7 = 15). Cuando la red fue colocada a aproximada-

mente 1 m, el adulto no la alcanzé de un solo salto, se acercé y cuando estuvo a la mitad de la

distancia, salt sobre ella. En esta misma ocasién se colocé la red a poca distancia del adulto

(menos de 50 cm), pero en una zona sin larvas y, aunque demoré mäs, reaccioné saltando

sobre la red y volviendo a la zona donde estaban las larvas en el 100 % de los casos (n = 15).

Cuando se le colocé la red a mayor distancia (a 1 m aproximadamente), en una zona sin larvas,

el adulto no reaccioné.

COMPORTAMIENTO LARVAL

«La formaciôn de cardümenes depende de la densidad larval?

Los resultados demuestran que a bajas densidades no hay tendencia a la formacién de

cardumen (4 = 23,018; P < 0,0001). La tendencia a formar cardumen mostré una diferencia

altamente significativa (P < 0, 0001) entre los tratamientos de 200 y 25 larvas; y una diferencia

significativa (P < 0, 05) entre las densidades de 100 y 25 larvas y las densidades de 50 y 25

larvas, mientras que no hubo diferencias significativas (P > 0, 05) en la tendencia a formar

cardumen entre los grupos de 200 y 100 larvas, 200 y 50 larvas, y 100 y 50 larvas. Por lo tanto

los resultados demuestran que el grupo de 25 larvas por tratamiento es el que no demuestra

una significativa tendencia a formar cardümenes (fig. 1).

«La formaciôn de cardumen es estimulada por la presencia del predador acuätico?

En experimentos de predacién realizados con 300 larvas por tratamiento, la tendencia a

formar cardumen fue mayor en presencia del predador que en ausencia del mismo (y? =

Source : MNHN, Paris

PONSSA 189

de observaciones

nûmero

Là

nümero de observaciones

200 100 50 25

densidades larvales

El con agrupamiento sin agrupamiento

Fig. 1. (a) Categorias de agrupamientos a diferentes densidades larvales en Leprodactylus insularum.

Categorias consideradas: (0) 0 % de larvas agrupadas: (1) hasta 10 %; (2) hasta 25 %;

(4) hasta 100 %. (b) Tendencia a formar cardümenes a diferentes densidades, considerando ünica-

mente la presencia (categoria de agrupamiento 1 a 4) o la ausencia (categoria de agrupamiento 0) de

agrupamiento.

Source : MNHN, Paris

190 ALYTES 19 (2-4)

nümero de observaciones

(Sin drocdor | 341 o | o

(Con deprodador | 5 | 13 | 15 9 | 8 | 15 10

categorias de agrupamientos

25.

Ê 15

Ë os

Sin prod 21 | |

1 Con deprechdor 2 6 4 | % 3

categorias de agrupamientos

Fig. 2. - (a) Nümero de observaciones de agrupamientos de larvas por categorias, utilizando 300 larvas

por tratamiento. (b) Nümero de observaciones de agrupamientos de larvas por categorias

empleando 100 larvas por tratamiento. Categorias consideradas: (0) sin agrupamiento: (1) < 5 larvas:

(2) < 20 larvas: (3) < 50 larvas: (4) < 100 larvas: (5) < 200 larvas: (6) < 300 larvas.

Source : MNHN, Paris

PONSSA 191

221.92; df = 1; P < 0.001) (fig. 2A). Lo mismo ocurrié en los experimentos realizados con 100

larvas por tratamiento (4? = 188.02; df =1; P < 0.001) (fig. 2B).

DISCUSION

CUIDADO PARENTAL

El cuidado parental en L. insularum ha sido reportado por WELLS & BARD (1988) en la

misma ärea en la que fue hecho este estudio (Gamboa, Panamä). Se pudo corroborar dicho

comportamiento aunque en algunos casos las larvas no se encontraban en compañia de un

adulto, posiblemente porque el descubrimiento de los nidos y los cardümenes implicaba

disturbio de la vegetacién circundante, y podria causar que los adultos escaparan. Estos

autores determinaron un despliegue de movimientos, por el cual el adulto se comunicaria con

las larvas y las guiaria a sectores del cuerpo de agua con suficiente profundidad para

completar el desarrollo. Este despliegue consiste en “bombeos” en el agua que comienzan

cuando la hembra arquea su espalda y eleva sus patas posteriores y cloaca sobre la superficie

del agua, luego su parte posterior desciende hasta el agua produciendo ondas en la superficie

que se dirigen hasta los renacuajos. DOWNIE (1996) observé “movimientos de bombeo”

similares en L. validus, aunque sus datos no demuestran que estos movimientos tengan la

funciôn de guiar a las larvas, posiblemente porque sus observaciones fueron hechas durante el

dia, y L. insularum fue observado guiando a las larvas principalmente durante la noche

(WELLS & BaRD, 1988). En este estudio no se observ este despliegue, ni tampoco el

desplazamiento de las larvas en direccién a la madre, como lo describen estos autores. La

observaciôn de que es la hembra quien realiza el cuidado estaria contradiciendo las hipétesis

de “la certeza paterna” y la del “orden de liberaciôn de los gametos” (GROSS & SHINE, 1981).

La primera hipôtesis propone que es mäs probable que las familias con fertilizaciôn externa

presenten cuidado paternal, porque la certeza de paternidad es mayor que en los casos con

fertilizaciôn interna. La segunda hipôtesis propone que el cuidado parental es el resultado de

la diferente oportunidad para abandonar a los cigotos, de manera que el sexo que desova

ültimo, deberia realizar el cuidado.

Los nidos que se encontraron estuvieron siempre en los mismos sitios, esto podria ser un

indicio de fidelidad al sitio. SEXTON (1962) reporté comportamiento de territorialidad en esta

especie; él observé que los machos cantan desde depresiones localizadas en el centro del nido

y los defienden de la invasiôn de otros machos. Las observaciones de WELLS & BARD (1988) y

VaIRA (1997) (quienes determinaron que las hembras realizan el cuidado parental) se contra-

dicen con las hechas previamente por TON (1962). Tal vez el macho y la hembra realizan el

cuidado de la puesta. A este tipo de cuidado se le da el nombre de “’anfisexual”, y ha sido

reportado para Cophixalus parkeri (SIMON, 1983), para algunos leptodactylidos, dendrobati-

dos y myobatrachidos (MCDiarMip, 1978). En este estudio no se pudo comprobar si era el

macho o la hembra el que realizaba el cuidado, ya que no se quisieron colectar los individuos,

para observar el comportamiento agresivo sin causar demasiado disturbio. Serian necesarias

observaciones que clarifiquen cual de los tres tipos de cuidado parental se presenta efectiva-

mente en esta especie.

Source : MNHN, Paris

192 ALYTES 19 (2-4)

De acuerdo a los resultados obtenidos, el cuidado parental en L. insularum implica

comportamiento agresivo. El ataque de los adultos hacia un hipotético predador soporta la

hipôtesis de que el comportamiento epimelético en esta especie tendria funciôn de defensa del

cardumen de larvas contra predadores. Esta misma funciôn fue determinada en L. ocellatus

(VAZ-FERREIRA & GEHRAU, 1975). En esta especie la hembra cuida al nido ubicändose en el

orificio central del mismo. Aunque los nidos de L. insularum observados presentaban un

orificio en el centro, nunca se vieron adultos instalados en él. Cuando los adultos estuvieron

con el cardumen, lo hicieron desde algün sustrato préximo, como vegetacién caida. WELLS &

BaARD (1988) observaron a una hembra de L. insularum ubicada en el centro del grupo de

renacuajos. A diferencia de L. insularum, en donde la hembra conduce a las larvas, en L.

ocellatus, aparentemente la hembra sigue a los renacuajos (VAZ-FERREIRA & GEHRAU, 1975).

Durante su vigilia, las hembras de L. ocellatus huyen al acercarse “una persona, dedos o

un päjaro”, o atacan saltando desde el orificio y mordiendo. Durante el salto emiten a veces

un grito de alarma (VAZ-FERREIRA & GEHRAU, 1975). De manera similar, el grito emitido por

L. insularum cuando saltaba sobre la red de colectar larvas, tendria funciôn agresiva contra el

predador de renacuajos. WELLS & BARD (1988) no observaron ataques agresivos en L.

insularum. VAIRA (1997) reporté ataques agresivos similares en hembras de L. insularum

(como L. bolivianus). Las hembras atacaban un palo colocado sobre sus cabezas, y luego una

bola de plästico atada a un palo a 20 cm de ellas, aunque no aclara si fueron colocadas en las

proximidades del cardumen. En el presente estudio los adultos se encontraron ocultos bajo la

vegetaciôn durante el dia, al igual que las hembras observadas por VAIRA (1997). Sin embargo

este autor registré ataques sélo durante la noche, y en este estudio se registraron durante el dia

en dos de las tres ocasiones en que fue observado. Las hembras de L. ocellatus en Uruguay

(VAZ- FERREIRA & GEHRAU, 1975) y Cordoba (VAIRA, 1997) también atacaron durante el dia.

Ataques similares también fueron registrados en el Ranidae africano Pyxicephalus adspersus:

en esta especie el macho acompaña al cardumen y ataca inclusive a grandes vertebrados

(BALINSKY & BALINSKY, 1954; ROSE, 1956; POYNTON, 1957). Ademäs de las especies de

Leptodactylus anteriormente mencionadas, L. ocellatus, L. validus y L. insularum, también

existen reportes de cuidado parental en L. chaquensis (DE ALMEIDA PRADO & UETANABARO,

2000), L. podicipinus (MARTINS, 1996), L. colombiensis (ESTRADA, comunicacién personal), L.

Jallax ÿ L. fuscus (LESCURE, 1973; 1979; 1983). Dada la falta de informacién sobre la

ocurrencia del cuidado parental en otras especies de Leptodactylus y sobre las relaciones

filogenéticas del género, por el momento no es posible establecer si el cuidado parental es una

sinapomorfia solamente de algunas especies del género, de todo el género o de un clado mäs

inclusivo. En un anälisis filogenético, elcuidado parental podria ser considerado como mäs de

ter segün lo realice el macho, la hembra o ambos; implique o no comportamiento

predadores; abarque el periodo de huevos y larvas, sélo el de huevos o sélo el

larvario.

COMPORTAMIENTO LARVAL

Efecto de la densidad larval en la formaciôn de cardumen

Los resultados obtenidos indican que a baja densidad, los renacuajos no tienden a

formar cardumen. Se postula que el “estimulo social”, es decir el estimulo generado por la

Source : MNHN, Paris

PONSSA 193

presencia de otras larvas, es una de las causantes de las agregaciones de larvas de anuros

(DuELLMAN & LESCURE, 1973; BEISWENGER, 1981). Posiblemente, como a bajas densidades la

probabilidad de encuentro entre larvas es menor, no actüe el estimulo social y por lo tanto no

haya formacién de cardumen. Por otro lado la mortalidad observada cuando se colocaron

larvas a muy bajas densidades (10 larvas por litro de agua), plantea el interrogante de si la

causa de la misma pudiera ser precisamente la baja densidad larval.

Efecto de la predaciôn en la formacién de cardumen

Los resultados de los experimentos de laboratorio demuestran que la presencia de

predadores estimula la formaciôn de cardimenes de larvas en L. insularum. En larvas de Hyla

regilla también se determiné un nivel mäs alto de agregaciôn en presencia de viboras que

actuaban como predadores (DE Viro et al., 1999). El häbito de formar agregaciones confiere

ventajas tanto a predadores como a presas (MA1OR, 1978). En las presas, incrementa su

habilidad para escapar al ataque del predador, y en este ültimo aumenta el éxito de captura.

De acuerdo con Ma1oR (1978), los predadores se orientan hacia las presas mediante el

estimulo visual. Durante el tiempo que requiere esta orientaciôn, el predador recibe la

informacién necesaria para predecir cuando y donde abrir la boca para un ataque seguro. Este

patrôn de acciôn fija puede ser critico, limitando la maniobrabilidad del predador durante los

milisegundos finales del ataque. Asi el predador podria no ser confundido por un cardumen

de presas, pero pierde el tiempo necesario para alinearse con una presa individual para un

ataque exitoso. Es en este contexto donde el comportamiento de formar cardümenes tiene

valor para la supervivencia de las presas. En el caso de larvas de anuros, KEHR (1994)

determiné que aunque los cardûmenes de larvas pueden producir “confusién” del predador

en la selecciôn de una presa, esta interacciôn se produciria especialmente cuando los preda-

dores son ciertos insectos acuâticos, como fue el caso de los experimentos realizados en este

estudio.

Otros factores, ademäs de la presencia de predadores y del estimulo “social”, han sido

propuestos para explicar las funciones y las causas de la formacién de cardümenes: estimulos

“ambientales”, como luz, temperatura, alimento, actuarian estimulando la formacién de

cardümenes (WASSERSUG & HESSLER, 1971; DUELLMAN & LESCURE, 1973; BEISWENGER, 1975,

1977, 1981; KATZ et al., 1981; O'Hara & BLAUSTEIN, 1981; CALDWELL, 1989).

CONCLUSIONES

Las ventajas y los costos del cuidado parental son mültiples, probablemente dependerän

no sélo de la especie que lo presente, sino del ambiente con el que se enfrenten los individuos

anuros. En base a los datos obtenidos en este estudio se puede concluir que en L. insularum el

cuidado parental implica un comportamiento agresivo y tendria una funcién de defensa

contra predadores posibles funciones a ser evaluadas. Los experimen-

tos realizados demuestran que la formaciôn de cardümenes responde al estimulo social (la

presencia de otras larvas, a densidades suficientemente altas) y a la presencia de predadores.

sto no descarta ot

Source : MNHN, Paris

194 ALYTES 19 (2-4)

AGRADECIMIENTOS

AL Dr. $. Rand por su gufa, sugerencias y comentarios sobre el trabajo. AI Dr. E. O. Lavilla y a 1. M.

Estrada por sus comentarios al manuscrito. À R. Cocrofi, por sus comentarios y a C. Molineri, por la

identificaciôn del predador acuâtico. Este trabajo se realizô gracias a una beca “Short-Term” de la

Smithsonian Tropical Research Institution (STRI).

LITERATURA CITADA

N, M. T., 1981. - Behavioral characteristics of the tadpoles of Rana heckscheri

J. Herp., 15 (2): 151-154.

BaLiNskY, B. I. & BALINSKY, J. B., 1954. - On the breeding habits of the South African bullfrog,

Pyxicephalus adspersus. S. African J. Sci., 51: 55-58.

BEISWENGER, R. E., 1975. - Structure and function in aggregations of tadpoles of the American toad,

Bufo americanus. Herpetologica, 31: 222-233.

ss 1977. - Diel patterns of aggregative behavior in tadpoles of Bufo americanus, in relation to light and

temperature. Ecology, 58: 98-108.

ai 1981. Predation by gray jays on aggregating tadpoles of the boreal toad (Bufo boreas). Copeia, 1981:

459-460.

BRANCH, L. C., 1983. - Social behavior of the tadpoles of Phyllomedusa vaillanti. Copeia, 1983: 420-428.

BREDEN, F., LUM, A. & WaASsERSUG, R., 1982. - Body size and orientation in aggregates of toad tadpoles

Bufo woodhousei. Copeia, 1982: 672-680.

BREDER, C. M., 1967. - On the survival value of fish schools. Zoologica, New York zool. Soc., 82 (4): 25-40.

ELL, J. P., 1989. - Structure and behavior of Hyla geographica tadpole schools, with comments on

ation of group behavior in tadpoles. Copeia, Le 938-950.

Cruwr, M., 1996. — Parental care among Amphibia. Jn: J. S. ROSENBLATT & C. T. SNOWDON (ed.),

Parental care: evolution, mechanism, and adaptive significance, Academic Press, 25: 109-144.

CrumP, M. L. & Scorr, N. J. Jr., 1994. — Visual encounter surveys. /n: W. R. HEYER, M. A. DONELLY, R.

W. McDrarMip, L. C. Hayeck & M.S. FOSTER (ed.), Measuring and monitoring biological diversity

— Standard methods for amphibians, Washington, Smithsonian Institution Press: 84-92.

DE ALMEIDA PRADO, C., UENTANABARO, M. & Lors, F. S., 2000. - Reproductive strategies of Leptodac-

1ylus chaquensis and L. podocipinus in the Pantanal, Brazil. J Herp., 34 (1): 135-139.

De Viro, J., CHivers, D. P., KIESECKER, J. M., BELDEN, L. K. & BLAUSTEI R., 1999. - Effects of snake

predation on aggregation and metamorphosis of pacific treefrog (Hyla regilla) larvae. J. Herp., 33

(3): 504-507.

Down, L R., 1996. - À new example of female parental behaviour in Leptodactylus validus, à Frog of the

leptodactylid “melanonotus” species group. Herp. J., 6: 32-34.

DuëLLMAN, W. E. & LEsCURE, J., 1973. - Life history and ecology of the hylid frog Ostcocephalus taurinus,

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

Goswer, K. L., 1960. - A simplified tale for staging anuran embryos and larvae with notes on identifica-

tion. Herpetologica, 16: 183-190.

Gross, M. R. & SARGE R. C., 1985. - The evolution of male and female parental care in fishes. Am.

Zool., 25: 807-

Gross, M. & SHiNE, R., 1981. — Parental care and mode of fertilization in ectothermic vertebrates.

Evolution, 35 (4): 775-793.

Hever, W. R., 1969. — The adaptative ecology of the species groups of the genus Leprodactylus

(Amphibia, Leptodactylidae). Evolution, 23: 421-428.

JABGER, R. G., 1994. - Transect sampling. /n: W. R. HEYER, M. A. DONELLY, R. W. MCDiaRMID, L. C.

Haveck & M.S. FOSTER (ed.), Measuring and monitoring biological diversity — Standard methods

for amphibians, Washington, Smithsonian Institution Press: 103-107.

KaATZ, L. M. Po’ & WASsERSUG, R. J., 1981. - Structure and mechanism of schooling in tadpoles of

the clawed frog, Xenopus lue Anim. Behav.. 29: 20-33.

Source : MNHN, Paris

PONSSA 195

Ke, A. L., 1994. - Patrones de dispersiôn espacio-temporales y su influencia en la biologia larval de

Bufo arenarum (Amphibia: Anura). Neotrépica, 40 (103-104): 35-40.

LESCURE, J,, 1973.- Contribution à l'étude des amphibiens de Guyane française. II. Leptodactylus fuscus

(Schneider). Observations écologiques et éthologiques. Ann. Mus. Hist. nat. Nice, 1: 91-100.

= 1979. - Etude taxonomique et éco-éthologique d’un amphibien des petites Antilles: Leptodactylus

fallax Müller, 1926 (Leptodactylidae). Bull. Mus. nat. Hist. nat. (4), 1 (3): 757-774.

LESCURE, J. & LETELLIER, F,, 1983. — Reproduction en captivité de Leptodactylus fallax Müller, 1926

(Amphibia, Leptodactylidae). Rer. fr: Aquariol, 10: 61-64.

McDiarMiD, R. W., 1978. - Evolution of parental care in frogs. /n: G. M. BURGHARDT & M. BEKOFF (ed.),

The development of behavior comparative and evolutionary aspects, New York, Garland: 127-147.

Mason, PE, 1978. Predator-prey interactions in two schooling fishes, Caranx ignobilis and Stolephorus

purpureus. Anim. Behav., 26: 760-777.

MARTIN, I. A., 1996. — Reproductive biology of Leptodactylus podicipinus (Cope, 1862) (Anura,

Leptodactylidae) in the northwest region of the state of So Paulo. Unpubl. Masters Thesis,

Universidade Estadual Paulista, Botucatu, SP, Brasil.

O'HarA, R. K. & BLAUSTEIN, A. R., 1981. — An investigation of sibling recognition in Rana cascadae

tadpoles. Anim. Behav., 29: 1121-1126.

Poyron, JL. C., 1957. - Bullfrog guardians. African Wildlife, 11: 80.

Rose, W., 1956. - Parental care in batrachians. African Wildlife, 10: 257.

Sexrow, O. J., 1962. - Apparent territorialism in Leptodactylus insularum Barbour. Herpetolôgica, 18 (3):

212-214.

Scorr, N. I. & WoopwaRD, B. D., 1994. — Surveys at breeding sites. Zn: W. R. HEYER, M. A. DONELLY, R.

W. McDiarmip, L. C. HayEck & M.S. FOSTER (ed.), Measuring and monitoring biological diversity

— Standard methods for amphibians, Washington, Smithsonian Institution Press: 118-125.

Simon, M. P. 1983. - The ecology of parental care in a terrestrial breeding frog from New Guinea, Behav.

Ecol. Sociobiol., 14: 61-67.

Surr, J. M., 1977. - Parental investment: a prospective analysis, Anim. Behav., 25: 1-9.

TownsenD, D. S., Srewarr, M. M. & PouGH, F H., 1984. - Male “mA care and its adaptive

significance in a neotropical frog. Anim. Behav., 32: 421-431.

TRIVERS, R. L., 1972. - Parental investment and sexual selection. /n: B. CAMPBELL (ed.), Sexual selection

and the descent of man, Chicago, Aldine Press: 136-179.

VaiRA, M., 1997. — Leptodactylus bolivianus (NCN). Behavior. Herp. Rev. 28 (4): 200.

VAZ-FERREIRA, R. & GEHRAU, A. 1975. - Comportamiento epimelético de la rana comün, Leptodactylus

ocellatus (L.) (Amphibia, Leptodactylidae). I. Atenciôn de la cria y actividades alimentarias

relacionadas. Physis, (B), 34 (88): 1-14.

WaLDMaN, B., 1981. - Sibling recognition in toad tadpoles: the role of experience. Z. Tierspsychol., 56:

341-358.

WASsERSUG, R. J., 1973. - Aspects of social behavior in anuran larvae. /n: J. L. ViaL (ed.), Evolutionary

biology of the anurans, Columbia, Missouri, University of Missouri Press: 273-297.

WassersUG, R. & HessLer, C. M. 1971. - Tadpole behavior: aggregation in larval Xenopus laevis. Anim.

Behav., 19: 386-389.

SSERSUG, R., LUM, A. M. & Po

Amphibia). Behav. d

WELLS, K. D., 1977.-The social behaviour of anuran amphibians. Anim. Behav., 25: 666-693.

K. D. & Bar», K. M. 1988. - Parental behavior of an aquatic-breeding tropical frog, Leptodac-

us bolivianus. J_ Herp., 22 (3): 361-364.

WITTENBERGER, J. F., 1981. — Animal social behavior. Boston, Duxbury Press.

ZIMMERMAN, B. L., 1994. - Audio strip transects. Zn: W. R. HEYER, M. A. DONELLY, R. W. MCDIARMID,

L. C. Hayeëck & M. S. FOSTER (ed.), Measuring and monitoring biological diversity — Standard

methods for amphibians, Washington, Smithsonian Institution Press: 92-97.

Corresponding editor: Karen R. Lirs.

ëL, M.J, 1981 Ananalyis of scioosrusture for tadpoles (nur:

Source : MNHN, Paris

Alytes, 2001, 19 (2-4): 196-204.

On the relevance of the character

‘absence of epicoracoid horns”’

in the systematics of anurans

Moises KAPLAN

Division of Reptiles and Amphibians, Museum of Zoology,

University of Michigan, Ann Arbor, Michigan 48109, USA.

<moiseska@umich.edu>

Several authors consider the character ‘absence of epi

a synapomorphy of the group that includes the anuran fa

tidae, Hyperoliidae, Microhylidae, Ranidae and Rhacophoridae (i.e., Ran:

dea). However, others have suggested that this condition is plesiomorphic

fie. epicoracoid horns are absent in several non-Ranoidea frogs) and that

Ranoidea frogs possess epicoracoid horns. The pectoral girdles of several

Ranoidea and non-Ranoidea frogs were sectioned histologically, and all had

epicoracoid horns. The presence of epicoracoid horns in Ranoïdea frogs

suggests the character “absence of epicoracoid horns” is not a synapomor-

phy of this group.

INTRODUCTION

Several authors (GRiFrITHS, 1963; KLUGE & FARRIS, 1969; LYNCH, 1973; DUELLMAN &

TRUEB, 1985) hypothesized that the families Dendrobatidae, Hyperoliidae, Microhylidae,

Ranidae and Rhacophoridae form a monophyletic group supported by the synapomorphy

“absence of epicoracoid horns” (i.e., posteriorly directed processes of the epicoracoid car-

tilages: GRiFFITHS, 1957, 1960a-b, 1963).

The hypothesis that the character “absence of epicoracoid horns” is a synapomorphy of

Ranoidea is questionable, first, because the absence of epicoracoid horns (MCDiARMID, 1971;

TRUEB, 1973; RUIZ-CARRANZA & HERNANDEZ-CAMACHO, 1976; GRANDISON, 1981: LYNCH &

RUIZ-CARRANZA, 1982; Myers & FORD, 1986; TYsON, 1987) in several non-Ranoïdea frogs

(ie. Atelopus, Atopophrynus, Brachycephalus, Dendrophryniscus, Didynamipus sjoestedti,

Melanophryniscus, Oreophrynella, Osornophryne, Rhinophrynus dorsalis) suggests this charac-

ter is plesiomorphic at the level of Ranoïdea, and second, because it has been suggested

(KAPLAN, 1994) that Ranoidea frogs have epicoracoid horns that are fused, indistinguishably,

to the sternum. However, the absence of epicoracoid horns in non-Ranoidea frogs and their

Source : MNHN, Paris

KAPLAN 197

presence in Ranoïidea frogs is controversial (KAPLAN, 1994; TysoN, 1987) and requires

confirmation through detailed (ï.e., histological) morphological examinations.

Herein, the girdles of several species of Ranoïidea and non-Ranoidea frogs are examined

through histological sectioning in order to assess the distribution of the character “absence of

epicoracoid horns” in frogs. The hypothesis that the character “absence of epicoracoid

horns” is a synapomorphy of Ranoidea is evaluated in light of the new morphological

observations.

MATERIALS AND METHODS

The medial portion of the ventral elements (omosternum, clavicles, procoracoids, epico-

racoids, coracoids, sternum) of the pectoral girdle of individuals at different developmental

stages (GOsNER, 1960: in parentheses) of the following species was sectioned (App. 1):

Ranidae: Rana blairi (44, adult); Dendrobatidae: Colostethus subpunctatus (44, 46, adult);

Microhylidae: Gastrophryne carolinensis (43, 44, adult); Hyperoliidae: Kassina senegalensis

(44, adult); Rhacophoridae: Rhacophorus moltrechti (44, adult): Bufonidae: Oreophrynella

quelchii (adult), Melanophryniscus stelzneri (adult) and Dendrophryniscus brevipollicatus

(adult); and Brachycephalidae: Brachycephalus ephippium (adult). Abbreviations for institu-

tions are as follow: ICN, Instituto de Ciencias Naturales, Universidad Nacional de Colombia;

UMMZ, University of Michigan Museum of Zoology.

The medial portion of the ventral elements of the pectoral girdles of adults and juveniles

was excised by cutting through the right and left procoracoid cartilages, clavicles, coracoid

bones, and around the omosternum and sternum, when present; the medial portion of the

ventral elements of the pectoral girdles was removed, decalcified (Cal-Ex IT, Fisher Scientific),

and sectioned transversely from the anterior tip of the omosternum to the posterior tip of the

sternum; Brachycephalus ephippium was sectioned from the tip of the snout to the posterior

part of the girdle. Tissues were embedded in paraffin (WESsNER, 1960), sectioned (15 m), and

stained with hematoxylin eosin. Histological and osteological terminologies follow those of

FawceTT (1986) and DE VILLIERS (1924), respectively.

Herein, 1 consider the epicoracoid horns to be parts of the epicoracoid cartilages that

extend posteriorly to the posteromedial part of the coracoid.

RESULTS

Descriptions of the zonosternal articulation (fig. 1) of non-Ranoïidea and Ranoiïdea

frogs.

Source : MNHN, Paris

198 ALYTES 19 (2-4)

Fig, 1. Ventral view of the pectoral girdle of Rana blairi, showing the interaction between the right and

left halves as the meet medially: cl, clavicle; co, coracoid: e, epicoracoid: o, omosternum; p,

procoracoid; s, sternum; zsa, zonosternal articulation. Grey, bone; clear, cartilage.

NON-RANOIDEA FROGS

In Dendrophryniscus brevipollicatus, Melanophrynisc eri and Oreophrynella quel-

chii, each epicoracoid cartilage bears a tapered, divergent cartilaginous process that extends

posteriorly from the coracoids (fig. 4A-C). In Brachycephalus ephippium, the posterior pro-

cesses of the epicoracoids are short, thin, not diverging from, and firmly attached (but not

indistinguishably fused) to, each other (fig. 4D).

In Oreophrynella quelchii, the m. sternoepicoracoideus does not insert on the posterior

terminus of the posterior processes of the epicoracoids. In Melanophryniscus stelzneri, a

laterally directed ligament inserts on the posterior terminus of each process: the m1. s

picoracoideus does not insert on the ligaments. In Dendrophryniscus brevipollicatus, à poste-

riorly directed ligament inserts on the posterior terminus of each process; these ligaments are

long and extend parallel to the lateral margins of the sternum; the mn. sternoepicoracoideus

inserts on the posterior ends of these ligaments. In Brachycephalus ephippium two short, wide

ternoe-

Source : MNHN, Paris

tions of the zonosternal articulation of the premetamorphic individuals (develop-

mental stage in parenthesis) of the following species: (A) Colostethus subpunctatus (46): (B) Gas-

trophryne carolinensis (43); (C) Kassina senegalensis (44); (D) Rana blairi (44); (E) Rhacophorus

moltrechti (44). 4, dorsal sternal blade: i, sternal isthmus; p, posterior process of the epicoracoids; v,

ventral sternal blade.

Fig. 2. - Transverse «

slips of them. sternoepicoracoideus insert on the posterior termini of the posterior processes

of the epicoracoids; no ligament mediates the insertion of the m. sternoepicoracoideus on the

posterior processes of the epicoracoids.

As seen in sectional view, in Dendrophryniscus brevipollicatus, Melanophryniseus stelzneri

and Oreophrynella quelchii, each lateral margin of the isthmus of the sternum bears a lateral

Source : MNHN, Paris

200 ALYTES 19 (2-4)

groove formed by dorsal and ventral blades (fig. 4A-C). In Brachycephalus ephippium, the

sternum is absent.

In Dendrophryniscus brevipollicatus, Melanophryniscus stelzneri and Oreophrynella quel-

chii, the posterior processes of the epicoracoids are housed in the sternal grooves (fig. 4A-C).

In D. brevipollicatus, M. stelzneri and O. quelchii, the anterior portion of the posterior

processes of the epicoracoids has its dorsal, medial and ventral surfaces separated from the

sternum by a gap. In D. brevipollicatus and M. stelzneri, a thick ligament surrounds and

attaches the lateral surfaces of the posterior processes of the epicoracoids to the dorsal and

ventral sternal blades; posteriorly, the posterior processes of the epicoracoids run parallel to

the lateral margins of the sternum for a short distance; in M. stelzneri, the posterior ends of

parts of the posterior processes of the epicoracoids are fused to the sternum.

RANOIDEA FROGS

In premetamorphic individuals of Colostethus subpunctatus, Gastrophryne carolinensis,

Kassina senegalensis, Rana blairi and Rhacophorus moltrechti, each epicoracoid cartilage

bears a blunt, tapered, ovoid (in transverse section), divergent cartilaginous process that

extends posteriorly from the coracoids (fig. 2A-E); the m. sternoepicoracoideus (ï.e., epicora-

coid horn muscle: GRiIFFITHS, 1963) does not insert on the posterior terminus of the process.

As seen in sectional view, each lateral margin of the isthmus of the sternum bears a lateral

groove formed by dorsal and ventral blades; the posterior processes of the epicoracoids are

housed in the sternal grooves and fused to the sternum.

In adult individuals of Colostethus subpunctatus, Rana blairi, and Rhacophorus mol-

trechti, the medial surfaces of the posterior processes of the epicoracoids are separated from

the sternum by a gap (fig. 3A-C). In R. blairi and R. moltrechti, the posterior tips of the

posterior processes of the epicoracoids are mostly eroded:; thus, the posterior parts of the

sternal grooves are empty or contain few fragments of the processes. In R. moltrechti, the

sternal isthmus is thinner (in transverse section) than the posterior processes of the epicora-

coids. In Gastrophryne carolinensis and Kassina senegalensis, the posterior processes of the

epicoracoids are indistinguishably fused to the sternum. The ventral sternal blade is absent in

Colostethus subpunctatus.

DISCUSSION

AL the species of frogs examined exhibit two processes of the epicoracoids that extend

posteriorly to the coracoids. In all Ranoïidea frogs examined, the posterior processes of the

epicoracoids are evident in premetamorphic stages; however, these processes in adults can be

partly eroded (e.g.. Rana blairi, Rhacophorus moltrechti) or indistinguishably fused to the

sternum (e.g.. Gastrophryne carolinensis and Kassina senegalensis). In all non-Ranoidea frogs

examined (i.e., Brachycephalus ephippium. Dendrophryniscus brevipollicatus, Melanophrynis-

Source : MNHN, Paris

KAPLAN 201

Fig. 3. -Transverse sections of the zonosternal articulation of adult individuals of the following speci

(A) Colostethus subpunctatus; (B) Rana blairi; (C) Rhacophorus moltrechti. d, dorsal sternal blade: i,

Sternal isthmus: f, fragment of one of the posterior processes of the epicoracoids: o, empty sternal

groove; p, posterior process of the epicoracoid; x, gap.

cus stelzneri, Oreophrynella quelchü), the posterior processes of the epicoracoids are evident in

adults.

The posterior processes of the epicoracoids of all the species here studied must be

considered epicoracoid horns because, like the horns of arciferal frogs (e.g., Ascaphus,

Bufo: GrirFirHs, 1963), they are posterior extensions of the epicoracoid cartilages and

show the same topographie relation to other body parts (e.g., they are housed in sternal

grooves).

This study challenges previous reports (MCDiARMID, 1971; TRUEB, 1973; TYsON, 1987)

that epicoracoid horns are absent in Brachycephalus, Dendrophryniscus, Melanophryniscus

and Oreophrynella. The presence of epicoracoid horns in these taxa suggests that all non-

Ranoidea frogs have epicoracoid horns (however, note that the presence of epicoracoid horns

in taxa such as Osornophryne, Geobatrachus, Atopophrynus, Didynamipus and Rhinophrynus

dorsalis still needs to be demonstrated), and therefore, that the character “absence of

epicoracoid horns” is not pleisiomorphie at the level of Ranoïdea. However, the character

“absence of epicoracoid horns” is invalid as a synapomorphy of Ranoïdea because, as this

study suggests, all Ranoidea frogs have epicoracoid horns (note that the monophyly of

Source : MNHN, Paris

202 ALYTES 19 (2-4)

Fig. 4. - Transverse sections of the zonosternal articulation of adult individuals of the following species:

(A) Dendrophryniscus brevipollicatus: (B) Melanophryniscus stelzneri: (C) Oreophrynella quelchit: (D)

Brachycephalus ephippium. d, dorsal sternal blade: i, sternal isthmus; p, posterior process of the

epicoracoid; v, ventral sternal blade.

Ranoidea is still supported by the characters “epicoracoid cartilages completely fused” and

“medial end of the coracoids wider than lateral end”: FORD & CANNATELLA, 1993).

I found considerable morphological variation in the epicoracoid horns of frogs: these can

be fused or free from the sternum, short or long, rounded or thread-like, extending parallel to

the lateral edge of the sternum or not. The systematic value of these characters, and the

independence from each other, is currently unknown.

RESUMEN

Varios autores consideran que el caracter “ausencia de cuernos epicoracoidales” es una

sinapomorfia del grupo (Ranoidea) que incluye à las familias Dendrobatidae, Hyperoliidae,

Microhylidae, Ranidae y Rhacophoridae. Sin embargo, se ha sugerido que esta condicién es

pleisiomorfica (esta presente en varios anuros que no son Ranoïdea) y que los anuros

Ranoiïdea tienen cuernos epicoracoidales. Las cinturas pectorales de varios anuros Ranoidea

y no Ranoïdea fueron seccionados histologicamente y se encontro que todos tienen cuernos

Source : MNHN, Paris

KAPLAN 203

epicoracoidales. La presencia de cuernos epicoracoidales en anuros Ranoïdea sugiere que el

caracter “ausencia de cuernos epicoracoidales” no es una sinapomorfia de este grupo.

ACKNOWLEDGMENTS

Lthank John D. Lynch and Arnold G. Kluge for helpful comments on the manuscript, Amold G.

Kluge and the late Pedro M. Ruiz for loans of material, Greg Schneider for help in processing the

specimens, John Wessman for providing histological facilities, John Megahan for helping with the

illustrations.

LITERATURE CITED

De Viurs, C. G. $., 1924. — On the anatomy of the breast-shoulder apparatus on Xenopus. Ann.

Transvaal Mus., 10: 197-211.

DueLLMaN, W. E, & TRUE L., 1985. - Biology of amphibians. New York, MeGraw-Hill, 1986”: i-xix +

1-670

Fawcerr, D. W, 1986. — Bloom and Fawcett; a textbook of histology. Philadelphia W. Saunders Com-

pany.

Forp, L. & CANNATELLA, D. C., 1993. - The major clades of frogs. Herp. Mon., 7: 94-117.

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

identification. Herpetologica, 16: 183-190.

GranisoN, A. G. C., 1981. - Morphology and phylogenetic position of the west African Didynamipus

sjoestedti Andersson, 1903 (Anura Bufonidac). Monit. zool. ital, (n. s), 15 (suppl): 187-215.

GrirriTHs, L., 1957. — The structure and development of the breast-shoulder apparatus in Amphibia.

Salientia. PhD Thesis, London, University of London: 1-339.

pus 1960. - The phylogeny of Sminthillus limbatus and the status of the Brachycephalidae (Amphibia:

Salientia). Proc. zool. Soc. London, 132: 457-487.

. 1960b. - The phylogenetic status of Sooglossinae. Ann. Mag. nar. Hist., (13), 2: 626-640.

Eee 1963. - The phylogeny of the Salientia. Biol. Rev, 38: 241-292.

KarLAN, M., 1994. — Analysis of some long-standing controversies concerning the pectoral girdle of

Atelopus (Bufonidac) using ontogenetic data. J. Herp., 28: 128-131.

KLUGE, A. G, & Fark, J.S., 1969. - Quantitative phyletics and the evolution of anurans. Syst. Zool., 18:

132

Lynn, J D. 1973. - The transition from archaic to advanced frogs. /n: J. L. ViaL (ed.), Evolutionary

biology of the anurans: contemporary research on major problems, Columbia, University of Mis-

souri Press: 133-182.

LyNCH, JD. & RUIZ-CARRANZA, P. M. 1982. - À new genus and species of poison-dart frogs (Amphibia:

Dendrobatidac) from the Andes of Northern Colombia. Proc. biol. Soc. Washington, 95: 557-562.

McDiarMiD, R. W. 1971. - Comparative morphology and evolution of frogs of the neotropical genera

Atelopus, Dendrophryniscus, Melanophryniseus, and Oreophrynella. Nat. Hist. Mus. Los Angeles

Co. Sci. Bull., 12: 1-66.

Myers, C. W. & Forp, L. S., 1986. - On Aropophrynus, a recently described frog wrongly assigned to

Dendrobatidae. Am. Mus. Novit.. 2843: 1-15.

RUIZ-CaRRANZA, P. M. & HERNANDEZ-CAMACHO, JL, 1976, - Osornophryne, genero nuevo de anfibios

bufonidos de Colombia y Ecuador. Caldasia, 11: 93-148.

Truës, L. 1973. - Bones, frogs, and evolution. rr: J. L. ViaL (ed.), Evolurionary biology of the anurans:

‘ch on major problems, Columbia, University of Missouri Press: 65-132.

contemporary res

Source : MNHN, Paris

204 ALYTES 19 (2-4)

Tysox, H., 1987. - The structure and development of the anurans breast-shoulder apparatus, forelimb, and

associated musculature. PhD Thesis, Edmonton, University of Alberta: 1-1275.

WESSNER, F. M., 1960. — General zoological microtechniques. Baltimore, Williams & Wilkins Co.

APPENDIX 1

SPECIMENS EXAMINED

Rana blairi (UMMZ 224284-5); Colostethus subpunctatus (ICN 15822, 23308; UMMZ 224288);

Gastrophryne carolinensis (UMMZ. 108418, 224286-7); Rhacophorus moltrechti (UMMZ 199779,

201298); Kassina senegalensis (UMMZ 151702, 210193); Oreophrynella quelchii (UMMZ 85141); Mela-

nophryniseus stelzneri (UMMZ 166804); Brachycephalus ephippium (UMMZ. 103568); Dendrophryniseus

brevipollicatus (UMMZ 204307).

Corresponding editor: W. Ronald HEYER.

ADTES

International Journal of batrachology

published by ISSCA

EDITORIAL BOARD

Chief Editor: Alain Dusoïs (Laboratoire des Reptiles et Amphibiens, Muséum national d'Histoire naturelle,

25 rue Cuvier, 75005 Paris, France; <dubois@mnhn.fr>).

Deputy Editor: Thierry Lové (Laboratoire d'Ecologie animale, Université d'Angers, 2 boulevard Lavoisier,

49045 Angers Cedex, France; <thierry.lode@univ-angers.fr>).

Conservation Editor: Stephen J. RicHARDS (Vertebrates Department, South Australian Museum, North Terrace,

Adelaide, S.A. 5000, Australia; <Richards.Steve@saugov.sa.govau>).

Editorial Board: Franco ANDREONE (Torino, Italy); Lauren E. BROWN (Normal, USA); Janalee P. CALDWELL

(Norman, USA); Ulisses CARAMASCHI (Rio de Janeiro, Brazil); Günter GOLLMANN (Wien, Austria); Heinz

GiLLirsCH (Wien, Austria); Tim HaLLIDAY (Milton Keynes, United Kingdom); W. Ronald HEÿER

(Washington, USA); Esteban O. LavILLA (Tucumän, Argentina); Karen R. Lies (Canton, USA); Masafumi

Marsut (Kyoto, Japan): Donald F, MCALPINE (Saint John, Canada); Alain PAGANO (Angers, France): John

C. PoynroN (London, England); Miguel VENCES (Konstanz, Germany).

Technical Editorial Team (Paris, France): Alain Dusois (texts); Roger BoUR (tables); Annemarie OuLEr (figures).

Book Review Editor: Annemarie OHLER (Paris, France).

SHORT GUIDE FOR AUTHORS

(for more detailed /nstructions to Authors, see Alytes, 1997, 14: 175-200)

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 colour or black and white photographs), showing beautiful or rare species, interesting behaviours, 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 tab. 4. Figures should not exceed 16 x

24 em. 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 (BOURRET, 1942; GRAF & POLLS PELAZ, 1989; INGER

et al., 1974). 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: i-x + 1-547,

pl. 1-4.

GRAF, J.-D. & PoLLs PELAZ, M. 1989. —Evolutionary genetics of the Rana esculenta complex. In: R. M. DAWLEY

& I. P. BOGART (ed.), Evolution and ecology of unisexual vertebrates, Albany, The New York State Museum:

289-302.

INGER, R. E,, Vorts, 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, anatomy, systematics, biogeography, evolution, genetics, anomalies or developmental

biology, or to Thierry LoDÉ (address above) if dealing with amphibian population genetics, ecology, ethology or

life history, or to Stephen J. RICHARDS (address above) if dealing with conservation biology, including declining

amphibian populations or pathology. 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 %4) 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 "4 high density 1.44 Mo floppy disk only).

Page charges are requested only from authors having institutional support for this purpose. The publication

of colour photographs is charged. For each published paper, 25 free reprints are offered by ISSCA 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 Dusors.

Numéro de Commission Paritaire: 64851.

©ISSCA 2001

Source : MNHN, Paris:

Alytes, 2001, 19 (2-4): 53-204.

Contents

Alain DuBois, Annemarie OHLER & S. D. Buu

A new genus and species of Ranidae (Amphibia, Anura)

ITOMSOUTDENES CNE Ho. 53-79

Alain DuBois & Annemarie OHLER

A new genus for an aquatic ranid (Amphibia, Anura) from Sri Lanka . 81-106

Miguel VENCES & Frank GLAW

Systematic review and molecular phylogenetic relationships

of the direct developing Malagasy anurans

of the Mantidactylus asper group (Amphibia, Mantellidae) .......... 107-139

Magali DELORME & Alain Dugois

Une nouvelle espèce de Scutiger du Bhutan,

et quelques remarques sur la classification subgénérique

du genre Scutiger (Megophryidae, Leptobrachiinae) 141-153

Sergio D. Rosser, Néstor G. Basso & Rubén J. LOMBARDO

Anälisis morfométrico de Pleurodema thaul (Lesson, 1826)

(Anura, Leptodactylidae) y algunas consideraciones acerca

JET MOTO IOBELCSICAIR LT dealer 154-172

Dinorah D. ECHEVERRIA, Carmen A. Up & Néstor G. Basso

Microscopia electrénica de barrido del aparato bucal

y cavidad bucofaringea de la larva de Atelognathus nitoi

CATUTASLEDTOd AE VHC) RE LRU RAR RSR Cent à concis 173-182

Maria Laura PONSSA

Cuidado parental y comportamiento de cardumen de larvas en

Leptodactylus insularum (Anura, Leptodactylidae) ................... 183-195

Moises KAPLAN

On the relevance of the character “absence of epicoracoid horns”

intbeSyStematICN CANUTEDS 2 V0 ee ac Meur LI nent 196-204

ANNOUNCEMENTS

Alytes is printed on acid-free paper.

Alytes is indexed in Biosis, Cambridge Scientific Abstracts, Current Awareness in Biological

Sciences, Pascal, Referativny Zhurnal and The Zoological Record.

Imprimerie F. Paillart, Abbeville, France.

Dépôt légal: 4° trimestre 2001.

Source : MNHN, Paris