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 Dusois (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); Stéphane

GROSIEAN (Paris, France); Julio Mario Hoyos (Bogotä, Colombia): Thierry LODÉ (Angers,

France); Jon LOMAN (Lund, Sweden); Alain PAGANO (Angers, France).

TARIFF FOR 2003

INDIVIDUALS

[Regular 2003 subscription to Alytes (volume 21) + ISSCA + Ci alpes” = _ 46€/60$

Stndent 2003 subscription to Alytes s (volume? 21) + ISSCA + Circalytes a _23€/30$

Regular 200

Student 200

{volume 21) alone

any single issue

any double issue

any complete volume (4 issues)

complete set of volumes 1 to 20 (1982-2002)

Regular ive=year (2003-2007, volumes 21 to 25) individual subscription 10 Alytes +

ISSCA +

Back issues of Alyte:

Life individual sub: Ivtes from 2003 on 900 € / 1180 $

Life individual subscription to Alvtes + ISSCA + Circalytes from 2003 on 1100 € / 1440 $

Patron individual subscription to A/yres from 2003 on 1760 € / 2304 $ or more

Patron individual subscription to Alyres + ISSCA + Circalytes from 2003 on 2200 € / 2880 $ or more

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

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

INSTITUTIONS

Pre JS to Alytes (volume 21) + ISSCA + Ci alytes n 92€/120$

issues of Ale: : any ny Single issue 22€/28$

Back issues of Alyier any double issue 4€156S

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

f complete set volumes 1 to 20 1120 € / 1467$

Sreciar orrrr: five-year (2003-2007, volumes 21 to 23) subscription to Alrtes + ISSCA + Circalytes.

with complete set of back issues of A/ytes (1982-2002, volumes 1 to 20) 534€/700$

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

prices can be provided upon request by our Secretariat.

MODES OF PAYMENT

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

above).

In Euros, by direct postal transfer to our postal account:

“ISSCA”, Nr. 1

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. pr Sà, Associate Professor,

Department of” Biology, of Richmond, Richmond, VA 23173, USA (e-mail:

rdesa@richmond.edu: Fax: ( 3

398-91 L, Paris: if you use this

Source : MNHN, Paris

AIVTES

INTERNATIONAL JOURNAL OF BATRACHOLOGY

May 2003 Volume 20, N° 3-4

Alytes, 2003, 20 (3-4) : 93-131.

The depressor mandibulae muscle in Anura

Adriana MANZANOŸ*, Silvia MoRO** & Virginia ABDALA**

* Centro de Investigaciones y Transferencia de Tecnologia a la Producciôn - CONICET,

Dr. Matteri y España 3105, Diamante, Entre Rios, Argentina

#* Fac. de Ciencias Naturales e IML, UNT, e Instituto de Herpetologia,

Fundaciôn Miguel Lillo - CONICE

Miguel Lillo 251, 4000 S. M. de Tucumän, Argentina

It is well known that the diversity in anuran cranial structure is large,

observed in the variation of the bone structure and associated muscles. The

depressor mandibulae is the jaw muscle that opens the mouth and was

considered by many authors useful in delimiting anuran groups. However,

there is still much controversy on the value of the m. depressor mandibu-

among different families of anurans. Specimens including 60 genera of 17

families of Anura were dissected, using traditional techniques of macroanat-

. Fifteen morphological groups can be distinguished within the anuran

species analysed. The insertion point demonstrates little variation. The

overall pattern of the origin is also quite stable, except for Hemisus g.

guineensis, Arthroleptis and Breviceps poweri, which have an unusual

anterior branch of m. depressor mandibulae coming from the maxilla. The

great diversity in shape of the m. depressor mandibulae correlates with the

plethora of patterns already mentioned for cranial morphology in Anura.

However, it is difficult to assign a specific morphology to a given higher

taxon. We also found no evidence that variations in the m. depressor

mandibulae are associated with particular habits.

INTRODUCTION

In frogs, the depressor mandibulae is the jaw muscle that opens the mouth. It is derived

from the hyoid arch and innervated by the facial nerve. It has a variable origin and, in general,

it occupies the area between the otic branch of the squamosal, the anulus (ympanicus, and the

dorsal fascia.

Bibliotheque Centrale Museum

(LULU

3001 00165040 6 MNHN, Paris

94 ALYTES 20 (3-4)

This muscle is one of the most controversial ones in frogs, evidenced by much discussion

around its utility as a phylogenetic character (GRiFrITHS, 1963; LIMESES, 1965; STARRETT,

1968; BURTON, 1983a-b, 1986; LyNCH, 1993; BURTON & ZWEIFEL, 1995; LA MARCA, 1995;

MANZANO, 1996, 1997; Hoyos, 1999). Many of these authors (BURTON, 1983a-b; MYERS &

ForD, 1986; FORD, 1989; SAVAGE, 1987; FORD & CANNATELLA, 1993) considered that the m.

depressor mandibulae is useful to delimit anuran groups. Nevertheless, other authors affirmed

that the morphology of the m. depressor mandibulae does not form consistent patterns with

cladistic or phenetic hypotheses (LYNCH, 1993; Hoyos, 1999).

Reviewing pseudid and hylid myology it was found (AM) that the m. depressor mandi-

bulae presents a distinctive morphology in pseudids and phyllomedusine frogs. The amount

of contradictory discussion led us to undertake a morphological descriptive survey of the m.

depressor mandibulae among anurans. The goal of this project is to explore the morphological

diversity of depressor mandibulae muscle among sixteen families of Anura.

MATERIALS AND METHODS

Two hundred and twenty seven specimens belonging to seventeen families of Anura (60

genera) were dissected under a stereomicroscope (app. 1). The skin was removed from the

cervical and cranial regions (both sides) on each specimen, exposing the depressor mandibulae

and adjacent musculature. When the muscle was divided, we removed the layers from the most

superficial to the deepest. Each dissection was described in detail, focusing especially on the

sites of origin and insertion, general morphological characteristics, and the relation of the m.

depressor mandibulae with other structures (bones, blood vessels, nerves and other muscles).

In some cases (particularly in small animals), differential staining with iodine solution (BOCK

& SHEAR, 1972) was done. Bones and cartilages of some specimens were stained with alizarine

red S and Alcian blue 8GX, respectively, as described by WASsERSUG (1976). The terminology

used for bone follows TRUEB (1993) and muscle terminology follows DUELLMAN & TRUEB

(1985). AI the material is deposited in the herpetological collection of the Fundaciôn Miguel

Lillo (FML), the herpetological collection of the Centro de Investigaciones y Transferencia de

Tecnologia a la Produccién (DIAM), Universidad del Valle de Cali (Colombia) (UVC) and

the African collection of Dr. Raymond Laurent (RL).

We consider that there are three generalized origin points of the depressor mandibulae

muscle: anulus tympanicus, squamosal and/or dorsal facia. When the muscle is divided in two

branches, we define the anterior branch as that branch corresponding to the first branch

present in a cephalo-caudal axis. When the anterior branch is overlapped by a branch, we

define this condition as deep anterior branch, and superficial anterior branch respectively.

The posterior one is that branch present behind the anterior branch. When an extra branch is

ahead of our anterior branch, we define that using a name related with its origin point. When

the muscle is divided in an external branch and an internal branch, we define those as

superficial depressor mandibulae and deep depressor mandibulae respectively.

To improve our understanding we gathered in groups, species that share similar condi-

tions of depressor mandibulae muscle in a table (tab. 1), adding schematic drawings — with

special reference to the origin points and general morphology, since the insertion points are

Stable —, and used STARRETT”S (1968) nomenclature with our modifications. STARRETT (1968)

Source : MNHN, Paris

MANZANO, MORO & ABDALA 95

proposed a notation system for the different depressor mandibulae morphologies. We used this

notation for our patterns (tab. 1) but found that it was necessary to define new states that were

not considered by the Starrett system. For example, the morphology of Hemisus guineensis

guineensis needs new letters (mx) to describe the branch's origin on the maxilla. Where

amuscle division into branches is present, it is represented by a dash. There is a limitation of

this nomenclature to describe different branch shapes. For example, the same notation is used

to symbolize our groups XIIb and XIIc, despite having differences between them. For that

reason we used the subindex 1 and 2 respectively.

RESULTS

FAMILY DISCOGLOSSIDAE

Alytes obstetricans (fig. 1a).— It has a single depressor mandibulae muscle that originates from

the dorsal fascia (on the upper part of the m. dorsalis scapulae) and m. levator mandibulae

posterioris longus. The m. depressor mandibulae is elongated and approximately triangular,

and inserts on the articular process of the lower jaw.

FAMILY PIPIDAE

Hymenochirus boettgeri camerunensis. — It has a single depressor mandibulae muscle that

originates from the otic process of the squamosal and posterior edge of the anulus tympanicus.

It is approximately rectangular, short, bulky and wider at the origin. The m. depressor

mandibulae does not contact the m. dorsalis scapulae. It inserts on the articular process of the

lower jaw.

Xenopus laevis. — It has a single depressor mandibulae muscle that originates from the dorsal

fascia at the level of the inferior edge of m. /evator mandibulae posterioris longus, superior

edge of the dorsal branch of m. dorsalis scapulae, and lateral edge of the m. romboideus

anterior. The m. depressor mandibulae is elongated and approximately triangular, and inserts

on the articular process of lower jaw.

FAMILY BUFONIDAE

Bufo arenarum. — Xt has a single depressor mandibulae muscle that originates from the otic

process of the squamosal. It is bulky and approximately fusiform, wider at the origin and is

quite apart from the m. dorsalis scapulae. The m. depressor mandibulae inserts on the articular

process of the lower jaw.

Bufo granulosus major. — I has a single depressor mandibulae muscle, but with fibers partially

dividing at the origin, arranged bundle-like in different directions, imbricating with each

other. It is fusiform and narrow, and covers the m. cuculari: sternocleidomastoideus).

originates from the posterior edge of the anulus tympanicus and posterior end of the otic

process of the squamosal. The m. depressor mandibulae inserts on the articular process of the

lower jaw.

Bufo paracnemis. — I has a single depressor mandibulae muscle that originates from the otic

branch of the squamosal. It is an approximately rectangular and slightly bulky muscle, wider

Source : MNHN, Paris

96 ALYTES 20 (3-4)

Fig, 1. (a) Alytes obstetricans (FML 02782). Morphology corresponding to group I (tab. 1). Scale bar:

0.5 mm. - (b) Melanophryniscus rubriventris rubriventris (FML 02502). Morphology corresponding

to group Ib (tab. 1). Scale bar: 0.5 mm. — (c) Ceratophrys cramwelli (FML 04924). Morphology

corresponding to group Ile (tab. 1). Scale bar: 0.5 mm. - Abbreviations: at, amulus tympanicus: dm,

m. depressor mandibulae; dsc, m. dorsalis scapulae: sq, squamosal.

Source : MNHN, Paris

MANZANO, MORO & ABDALA 97

at the origin and does not cover the m. dorsalis scapulae. It inserts on the articular process of

the lower law.

Bufo spinulosus. — I has a single depressor mandibulae muscle that originates from the otic

process of the squamosal. It is bulky and fusiform, and is formed by bundles of fibers that split

easily. It partially covers the m. cucularis. The m. depressor mandibulae inserts on the articular

process of the lower law.

Melanophryniscus rubriventris rubriventris (fig. 1b). — It has a single depressor mandibulae

muscle that originates from the otic process of the squamosal. It is a fan-shaped and bulky

muscle, wide at the origin (the last characteristic only in some specimens) and does not contact

the m. dorsalis scapulae. The m. depressor mandibulae inserts on the articular process of the

lower jaw.

FAMILY LEPTODACTYLIDAE

Ceratophryinae

Ceratophrys cranwelli (fig 1e). — It has a single depressor mandibulae muscle that originates

from the otic process of the squamosal, covered by the osiculum, on the posterior border of

the anulus tympanicus and dorsal fascia (at the level of the m. dorsalis scapulae). I is an

irregular and thick muscle, wide at the origin and narrow at the insertion. It partially covers

the m. dorsalis scapulae and inserts on the articular process of the lower jaw.

Chacophrys pierotti (fig. 2a).— It has a single depressor mandibulae muscle that originates from

the otic process of the squamosal, the posterior border of the anulus tympanicus and dorsal

fascia (at the level of the m. dorsalis scapulae). It is an irregular muscle, wide at the origin and

narrow at the insertion. It partially covers the m. dorsalis scapulae and inserts on the articular

process of the lower jaw.

Lepidobatrachus llanensis (ig. 2b). — It has a single depressor mandibulae muscle, fusiform,

very thick and long. It originates on the posterior edge of the anulus tympanicus and otic

process of the squamosal, and inserts on the articular process of the lower jaw.

Hylodinae

Crossodactylus gaudichaudiüi (fig. 2e). — It has a thin, short, and fan-shaped depressor mandi-

bulae muscle divided into two branches joined at mid-muscle level. Anterior fibers originate

on the otic process of the squamosal, and the posterior fibers originate from the dorsal fascia

(at the level of the posteromesial edge of the m. dorsalis scapulae). The posterior branch is

longer than the anterior. It inserts on the articular process of the lower jaw by a short tendon.

Leptodactylinae

Leptodactylus bufonius (fig. 3a). — It has a wide and approximately triangular depressor

mandibulae muscle divided into two branches joined at mid-muscle level. The anterior fibers

originate from the otic process of the squamosal and posterior fibers from the dorsal fascia (at

the level of the mesial edge of the m. dorsali: apulae and m. latissimus dorsi). The m.

depressor mandibulae inserts on the articular process of the lower jaw.

Leptodactylus chaquensis (fig. 3b). — It has a single depressor mandibulae muscle with a subtle

division at its origin. It originates on the otic process of the squamosal and dorsal fascia (at

Source : MNHN, Paris

98 ALYTES 20 (3-4)

Fig. 2. - (a) Chacophrys pierotti (FML 01020). Morphology corresponding to group Ha (tab. 1). Scale bar:

0.5 mm. - (b) Lepidobatrachus llanensis (FML 01095). Morphology corresponding to group Ile (tab.

1). Scale bar: 0.5 mm. - (c) Crossodacrylus gaudichaudit (FML 03497). Morphology corresponding.

to group I (tab. 1). Scale bar: 0.5 mm. - Abbreviations: at, anulus (Ympanicus; dm, m. depressor

mandibulae; dma, m. depressor mandibulae anterior: dmp, m. depressor mandibulae posterior, sc, m.

dorsalis scapulue: sq, squamosal.

Source : MNHN, Paris

MANZANO, MORO & ABDALA 99

ci c2

=) Leptodaetyus bufonius (FML04642). Morphology corresponding 10 group III (tab. 1). Scale

5 mm. (b) Leptodactylus chaquensis (FML 05490). Morphology corresponding to group III

le bar: 0.5 mm. — (1) Pseudopaludicola boliviana (FML 04307). Morphology correspon-

ding to group Ve (tab. 1). Scale bar: 0.2 mm. — (c2) Detail of deep m. depressor mandibulae of P.

boliviana. Scale bar: 0.2 mm. — Abbreviations: at, anulus tympanicus: dm, m. depressor mandibulac:

dmd, deep m. depressor mandibulae; dms, superficial m. depressor mandibulae; ds@y m. dorsalis

scapulae; nVI, branch of the nerve VII; sq, squamosal.

\ *) Source : MNHN, Paris

100 ALYTES 20 (3-4)

the level of mesial edge of the m. dorsalis scapulae and m. latissimus dorsi), The m. depressor

mandibulae inserts on the articular process of the lower jaw.

Pleurodema borellii. — It has a depressor mandibulae muscle divided into two slips, the anterior

branch originates from the otic process of the squamosal. It is approximately triangular, wide

at the origin. The posterior branch originates from the dorsal fascia (at the level of mesial edge

of the m. dorsalis scapulae and m. latissimus dorsi). It is thicker and wider that the anterior

branch. Both branches insert on the articular process of the lower jaw.

Physalaemus biligonigerus. — It has a depressor mandibulae muscle divided into two branches

joined at mid-muscle level. The anterior branch is short and triangular and originates from

the otic process of the squamosal. The posterior branch is triangular and wide and originates

from the dorsal fascia (at the level of the middle region of the m. dorsalis scapulae). Both

branches are inserted on the articular process of the lower jaw.

Pseudopaludicola boliviana (fig. 30). — It has a depressor mandibulae muscle divided into two

slips. À large and fan-shaped superficial slip arises from the dorsal fascia (at the level of the

cervical muscles, partially covering the m. levator scapulae). This slip is very wide and thin at

the origin, and it extends extensively on the dorsal region of the trunk. It covers the posterior

part of the anulus tympanicus without adhering to it. A nerve V branch passes through the

superficial slip. An approximately rectangular, flat and small deep slip originates on the otic

process of the squamosal (with some anterior fibers contacting the posterior edge of the

anulus tympanicus). I contacts the m. cucularis at the origin, partially covering it. Both

branches of the m. depressor mandibulae insert on the articular process of the lower jaw.

Telmatobiinae

Alsodes sp. — It has a depressor mandibulae muscle divided into two slips. A thick and

rectangular anterior slip originates from the otic process of the squamosal, and a posterior

slip that originates from the dorsal fascia (at the level of the middle region of m. dorsalis

scapulae). The posterior slip is longer than the anterior. Both branches insert on the posterior

end of the articular process of the lower jaw.

Batrachyla sp. — It has a depressor mandibulae muscle divided into two slips. An approximately

rectangular, short, and thick anterior slip originates from the otic process of the squamosal.

The posterior slip, approximately triangular, long, and thin, originates from the dorsal fascia

(at the level of the middle region of the m. dors ‘apulae). Both branches insert on the

articular process of the lower jaw.

Hylorina sylvatica. — I has a depressor mandibulae muscle divided into two equally long slips.

The anterior slip is thick and rectangular and arises from the otic process of the squamos

The posterior one, triangular and thicker than the anterior, originates from the dorsal fa

(at the level of the middle region of m. dorsalis scapulae). Both branches insert on the articular

process of the lower jaw.

Telmatobius laticeps (fig. 4a). — It has a depressor mandibulae muscle divided into two slip:

The anterior slip is approximately rectangular, narrow, and thick. It originates on the otic

of the squamosal. The posterior slip, approximately triangular and wide at the origin,

ises from the dorsal fascia (at the level of the middle region of m. dorsalis scapulae). Both

branches insert on the articular process of the lower jaw.

Source : MNHN, Paris

MANZANO, MORO & ABDALA 101

Fig. 4. - (a) Telmatobius laticeps (FML 03960). Morphology corresponding to group X (tab. 1). Sacle bar

0.5 mm. - (b) Crinia georgiana (DIAM 015). Morphology corresponding to group Ile (tab. 1). Scale

bar: 0.5 mm. — (c) Limmodynastes letcheri (DIAM 009). Morphology corresponding to group Va

(tab. 1). Scale bar: 0.5 mm. - Abbreviations: at, anulus sympanicus; dm, m. depressor mandibulac:

dma, m. depressor mandibulae anterior: dmp, m. depressor mandibulae posterior: dse, m. dorsalis

Source : MNHN, Paris

102 ALYTES 20 (3-4)

Telmatobius scrocchit. — It has a depressor mandibulae muscle divided into two slips. The

anterior slip is approximately triangular, narrow, short and thick. It arises from the otic

process of the squamosal. The posterior slip originates from the dorsal fascia (at the level of

the middle region of the m. dorsalis scapulae and the most anterior fibers at the level of the

otoccipital). Itis triangular, longer, and twice as thick as the anterior slip. Both branches insert

on the articular process of the lower jaw.

Telmatobius oXycephalus. — It has a depressor mandibulae muscle divided into two slips. The

anterior slip, approximately triangular and narrow, originates from the otic process of the

squamosal. The posterior slip, triangular, wide, and longer than the anterior, originates from

the dorsal fascia (at the level of the middle region of the m. dorsalis scapulae). Both branches

insert on the articular process of the lower jaw.

FAMILY MYOBATRACHIDAE

Myobatrachinae

Crinia georgiana (fig. 4b). — It has a bulky and fusiform single depressor mandibulae muscle. It

arises from the otic process of the squamosal and the posterior edge of the anulus tympanicus,

partially covering it. The m. depressor mandibulae does not cover the m. dorsalis scapulae.

There are two kinds of fibers: anterior fibers parallel to the posterior edge of the anulus

tympanicus, and posterior fibers oriented perpendicular to anterior ones. The m. depressor

mandibulae inserts on the posterior edge of the lower jaw by a tendon.

Taudactylus diurnus. — I has a fan-shaped single depressor mandibulae muscle, wide at its

origin. Some anterior fibers originate from the otic process of the squamosal. Posterior fibers

arise from the dorsal fascia (at the level of the anteromesial edge of m. dorsalis scapulae). The

m. depressor mandibulae inserts on the articular process of the lower jaw by a short and thick

tendon.

Uperoleia aspera. — I has a long and bulky single depressor mandibulae muscle, partially

covering the tympanum. It originates on the dorsal fascia (at the level of the occipital and otic

process of the squamosal); the anterior fibers are partially divided originating on the poste-

rodorsal edge of the anulus tympanicus. The m. depressor mandibulae inserts on the articular

process of the lower jaw by a short tendon.

Uperoleia boreali Ithas a depressor mandibulae muscle divided into two slips. An anterior

slip originates on the posterior edge of the anulus tympanicus. This slip is flat and approxi-

mately rectangular, and is partially covered by the posterior slip. The posterior slip originates

on the external surface of the otic process of the squamosal and the dorsal fascia. It is

approximately triangular, wider at the origin and does not contact the m. dorsalis scapulae.

Both branches insert on the articular process of the lower jaw.

Limnodynastinae

Limnodynastes dumerili. — K has a depressor mandibulae muscle divided into two slips. An

anterior and approximately rectangular slip originates on the ventral edge of the anulus

tympanicus and otic process of the squamosal, partially covered by the posterior branch. The

Source : MNHN, Paris

MANZANO, MORO & ABDALA 103

posterior slip originates on the dorsal fascia, partially covering the m. dorsalis scapulae. K is

wide and approximately triangular, slightly curved at the origin, surrounding the middle

anterior edge of m. dorsalis scapulae. Both branches insert on the articular process of the

lower jaw.

Limnodynastes convexiusculus. — It has a depressor mandibulae muscle divided into two slips.

An anterior slip originates on the posteroventral edge of the anulus tympanicus. It is approxi-

mately quadrangular, very short, thin, and narrow. The posterior slip is narrow, elongated and

has a noticeable curve caudad. It originates on the dorsal fascia (at the level of the anterior

edge of the suprascapula, partially covering the m. dorsalis scapulae). Both branches insert on

the articular process of the lower jaw.

Limnodynastes dorsalis. — It has a depressor mandibulae muscle divided into two slips. An

anterior slip, very thin and short, originates on the posterior edge of the anulus tympanicus.

The posterior slip is fan-shaped and very wide that originates from the dorsal fascia (at the

level of the medial edge of m. dorsalis scapulae and the otic process of the squamosal). It

partially covers the anterior branch. Both branches insert on the articular process of the lower

jaw by a tendon.

Limnodynastes fletcheri (fig. 4c).— It has a depressor mandibulae muscle divided into two slips.

An anterior approximately triangular, very thin, and short slip originates on the posteroven-

tral edge of the anulus tympanicus. The posterior slip originates on the dorsal fascia (at the

level of the mesial edge of the m. dorsalis scapulae) and the otic process of the squamosal. It

is fan-shaped and wide at the origin, and partially covers the anterior branch. Both branches

insert on the articular process of the lower jaw.

Limnodynastes ornatus. — It has a depressor mandibulae muscle divided into two slips. The

anterior slip, approximately rectangular and very short, originates on the posterior edge of

the anulus tympanicus. The posterior slip is fan-shaped and wide at the origin and originates

from the dorsal fascia (at the level of the mesial edge of the m. dorsalis scapulae and the otic

process of the squamosal). It partially covers the anterior branch. Both branches insert on the

articular process of the lower jaw.

Notaden melanoscaphus (fig. Sa). It has a depressor mandibulae muscle divided into two slips.

The anterior slip, approximately rectangular, long and narrow, originates from the otic

process of the squamosal. The posterior slip originates on the dorsal fascia (at the level of the

origin of the m. dorsalis scapulae). is approximately rectangular and longer than the

anterior slip and is very narrow and curved, crossing over middle part of the m. cucularis.

Both branches insert on the articular process of the lower jaw.

Notaden nichollsi. — has a depressor mandibulae muscle divided into two slips. The anterior

slip, approximately fusiform and long, originates on the otic process of the squamosal. The

posterior slip originates on the dorsal fascia (at the level of the origin of the m. dorsalis

scapulae). His approximately rectangular and longer than the anterior slip, very narrow and

curved. Both branches insert on the articular process of the lower jaw.

FAMILY RHINODERMATIDAE

Rhinoderma darwinit (fig. 5b). — It has a single depressor mandibulae muscle, approximately

triangular and wide at the origin. It contacts the anterior border of the m. dorsalis scapulae.

Source : MNHN, Paris

104 ALYTES 20 (3-4)

Fig. 5. (a) Notaden melanoscaphus (FML 03783). Morphology corresponding to group IX (tab. 1).

bar: 0.5 mm. — (b) Rhinoderma darwinii (FML 03694). Morphology corresponding to group Vila

(tab. D). Scale bar: 0.5 mm. — (€) Gastrotheca christiani (FML 02117). Morphology corresponding to

group IV (tab. 1). Scale bar: 0.5 mm. - (d) Gastrotheca gracilis (FML 01769). Morphology

corresponding to group IV (tab. 1). Scale bar: 0.5 mm. - Abbreviati anulus tympanicus: dm, m.

depressor mandibulae dm, m. depressor mandibulue; ma, m. depressor mandibulae anterior; dmp, m.

ssor mandibulae posterior; se, m. dorsalis scapule: sq, squamosal.

Source : MNHN, Paris

MANZANO, MORO & ABDALA 105

It has anterior fibers originating on the posterior edge of the anulus tympanicus. The posterior

fibers originate from the posterior end of the otic process of the squamosal. The m. depressor

mandibulae inserts on the articular process of the lower jaw.

FAMILY HYLIDAE

Hemiphractinae

Gastrotheca christiani (fig. 5c). — It has a single depressor mandibulae muscle that originates

from the dorsal fascia, the otic process of the squamosal, and the posterior edge of the anulus

tympanicus. Approximately rectangular and wide at the origin, it contacts with the anterior

border of the m. dorsalis scapulae. The m. depressor mandibulae inserts on the articular

process of the lower jaw.

Gastrotheca chrysosticta. It has a single depressor mandibulae muscle that originates from the

dorsal fascia, the otic process of the squamosal, and the posterior edge of the anulus

tympanicus. Approximately rectangular, wide at the origin, it contacts the anterior border of

the m. dorsalis scapulae. The m. depressor mandibulae inserts on the articular process of the

lower jaw.

Gastrotheca gracilis (fig. 5d). It has a single depressor mandibulae muscle that originates from

the dorsal fascia, the otic process of the squamosal, and the posterior edge of the anulus

tympanicus. Approximately triangular and wide at the origin, the muscle contacts the anterior

border of the m. dorsalis scapulae. The m. depressor mandibulae inserts on the articular

process of the lower jaw.

Hylinae

Acris crepitans (fig. 6a). — It has a depressor mandibulae muscle divided into two slips. The

anterior slip, approximately triangular and elongated, originates on the otic process of the

squamosal, and partially covers the posterior edge of the anulus tympanicus. The posterior

slip, very narrow, thin, and curved, originates on the dorsal fa. (at the level of the middle

part of m. dorsalis scapulae). Both branches insert on the articular process of the lower jaw.

Argenteohyla siemersii. — I has a depressor mandibulae muscle divided into three slips. The

anterior slip, approximately rectangular and narrow, originates on the posterior edge of the

anulus tympanicus. The middle slip, approximately rectangular and wide at the origin (longer

than anterior slip), originates on the posterior edge of otic process of the squamosal. The

posterior slip, triangular and wide (longer than middle slip), originates on the dorsal

the level of the middle part of m. dorsalis scapulae). The three branches insert on the articular

process of the lower jaw.

Hyla boans (fig. 6b).— It has a depressor mandibulae muscle divided into two slips. The anterior

slip, approximately triangular (wider at the origin) originates on the posterior edge of the otic

process of the squamosal. The posterior slip, approximately triangular and wide at the ori:

(longer than anterior slip), originates on the dorsal fa: (at the level of the middle region of

‘apulae). Both branches insert on the articular process of the lower jaw.

m. dorsali:

Hyla andina.— has a single depressor mandibulae muscle that originates from the otic process

of the squamosal and the posterior edge of the anulus tympanicus. I is approximately

Source : MNHN, Paris

106 ALYTES 20 (3-4)

. Re

Fig. 6. — (a) Acris crepitans (RL 017838). Morphology corresponding to group IX (tab. 1).

0.3 mm. — (b) Hyla boans (FML 09543). Morphology corresponding to group X (tab. 1)

0.5. (c) Phrynohyas venulosa (ML 02303). Morphology corresponding to group XI (tab. 1)

bar:0.5mm. - Abbreviations: at, anulus ympanicus: pressor mandibulae; dma, m. depri

mandibulae anterior; mm, m. depressor mandibulae medialis; dmp, m. depressor mandibulae poste-

rior: dse, m. dorsalis scapulae: sq, squamosal.

Source : MNHN, Paris

MANZANO, MORO & ABDALA 107

triangular, wider at the origin and does not contact with m. dorsalis scapulae. The m. depressor

mandibulae inserts on the articular process of the lower jaw.

Phrynohvas venulosa (fig. 6c).— It has a depressor mandibulae muscle divided into three slips.

The anterior slip is short and approximately rectangular and originates on the ventral edge of

the anulus tympanicus. The middle slip is wide and partially covered by the anterior slip (the

otic process of the squamosal crossing over it). It originates on the epimysium of the lateral

edge of m. levator mandibulae posterioris longus. The posterior slip, long and approximately

triangular, arises from the dorsal fascia (at the level of the suprascapula) and contacts the m.

levator scapulae. The three branches insert on the articular process of the lower jaw by a thick,

wide and very short tendon.

Plectrohyla guatemalensis. — I has a depressor mandibulae muscle divided into two slips. The

anterior slip arises from the otic process of the squamosal. It is very thin, approximately

rectangular, and short. The posterior slip originates from the dorsal fascia (at the level of the

mesial edge of the suprascapula). It is very wide and approximately triangular. Both branches

insert on the articular process of the lower jaw.

Piychohyla ignicolor. — X has a depressor mandibulae muscle divided into two slips. The

anterior slip originates from the otic process of the squamosal. It is thick and approximately

rectangular. The posterior slip originates from the dorsal fascia (at the level of the edge of the

anterior half of the m. dorsalis scapulae). It is very thin and approximately rectangular. Both

branches insert on the articular process of the lower jaw by a short tendon.

Scinax fuscovarius (fig. 7a).— It has a fan-shaped single depressor mandibulae muscle (wide at

the origin) that originates from the dorsal fascia (at the level of the superior branch of m.

dorsalis scapulae), otic process of the squamosal, and the anterior fibers on the posterior edge

of anulus tympanicus. Y inserts on the articular process of the lower jaw.

Smilisca sila. — K has a depressor mandibulae muscle divided into two slips. The anterior slip

originates from the posterior end of the otic process of the squamosal. It is long, narrow, and

approximately rectangular. The posterior slip originates from the dorsal fascia (at the level of

the mesial half of the m. dorsalis scapulae). It is long, thin, approximately rectangular, and

curved. Both branches insert on the articular process of the lower jaw.

Phyllomedusinae

Agalychnis saltator (fig. 7b).— It has a depressor mandibulae muscle divided into two slips. The

anterior slip arises from the posteroventral edge of the anulus tympanicus. His approximately

rectangular, short, and thin. The posterior slip arises from the otic process of the squamosal.

Itis very thick, wide, and approximately rectangular. Both branches insert on the articular

process of the lower jaw.

Phyllomedusa boliviana. — I has a depressor mandibulae muscle divided into three slips. The

anterior slip originates from the posteroventral edge of the anulus tympanicus. Xt is approxi-

mately rectangular, short, and thin. The middle slip is approximately triangular (wider at the

origin). It arises from the otic process of the squamosal, and partially covers the posterior Slip,

which arises from the dorsal fascia (at the level of the anterior edge of the m. dors

scapulae). is long, curved, and approximately triangular; it covers the anterior edge of the

m. dorsalis scapulae. The three branches insert on the articular process of the lower jaw.

Source : MNHN, Paris

108

ALYTES 20 (3-4)

= (a) Scinax fuscovarius (FML 04635). Morphology corresponding to group IV (tab. 1). Scale bar:

mm. - (b) Agalychnis saltator (FML 09541). Morphology corresponding to group VII (tab. 1).

Scale bar: 0.5 mm. - (c) Cenrrolene grandisonae (FML 04980). Morphology corresponding to group

Vila (tab. 1). Scale bar: 0.5 mm, — (d) Cochranella ignora (U VC 12091). Morphology corresponding

to group Vila (tab. 1). Scale bar: 0.5 mm. - Abbreviations: at, anulus tympanicus; dm, m. depr

mandibulae; dma, m. depressor mandibulae anterior; dmp, m. depressor mandibulae posterior, ds

dorsalis scapulae: sq, squamosal.

Fig. 7.

Source : MNHN, Paris

MANZANO, MORO & ABDALA 109

Phyllomedusa hypocondrialis. — It has a depressor mandibulae muscle divided into three slips.

The anterior slip originates from the ventral edge of the anulus tympanicus. Itis approximately

rectangular, short, and thin. The middle slip originates from the otic process of the squamo-

sal. It is approximately rectangular, wide at the origin. The posterior slip originates from the

dorsal fascia (at the level of the anterior edge of m. dorsalis scapulae). It is long, thin,

approximately triangular, and curved. It covers the anterior edge of m. dorsalis scapulae. The

three branches insert on the articular process of the lower jaw.

Phyllomedusa sauvagii. — I has a depressor mandibulae muscle divided into three slips. The

anterior slip originates from the posteroventral edge of anulus tympanicus. It is approximately

rectangular and thin. The middle slip originates from the otic process of the squamosal.

It is approximately triangular (wide at the origin)and partially covers the posterior slip.

The posterior slip originates from the dorsal fascia (at the level of the middle region of the

m. dorsalis scapulae). I is long and approximately rectangular. It covers the anterior half

of the m. dorsalis scapulae. The three branches insert on the articular process of the lower

jaw.

FAMILY CENTROLENIDAE

Centrolene grandisonae (fig. 7c). — It has a single depressor mandibulae muscle. It originates on

the posterior edge of the anulus tympanicus and the otic process of the squamosal. It is short,

small and approximately triangular. It does not contact the m. dorsalis scapulae. The m.

depressor mandibulae inserts on the articular process of the lower jaw.

Cochranella ignota (fig. 7d).— It has a depressor mandibulae muscle divided into two slips. The

anterior slip arises from the posteroventral edge of the anulus tympanicus. It is approximately

rectangular, short, and thin. The posterior slip is approximately triangular and it originates on

the otic process of the squamosal. It does not contact the m. dorsalis scapulae. It inserts on the

articular process of the lower jaw.

FAMILY PSEUDIDAE

Lysapsus limellus (fig. 8a). — It has a single depressor mandibulae muscle. It originates on the

posterior edge of the anulus tympanicus, the otic process of the squamosal, and the dorsal

The m. depressor mandibulae is approximately triangular, wider at the origin, and

contacts the m. dorsalis scapulae. The m. depressor mandibulae inserts on the articular proces

of the lower jaw.

Pseudis minutus (fig. 8b). — It has a single depressor mandibulae muscle that originates on the

posterior edge of the anulus tympanicus, the otic process of the squamosal, and the dorsal

fascia. It is a very thick, wide and fan-shaped muscle that contacts the m. dorsalis scapulae.

The m. depressor mandibulae inserts on the articular process of the lower jaw.

fig. 8c). — It has a single depressor mandibulae muscle. It originates on the

the otic process of the squamosal, and the dorsal

e, which contacts the m. dorsalis scapulae. The m.

ular process of the lower jaw.

Pseudis paradox

posterior edge of the anulus 1ympanicu

fascia. It is a long, fan-shaped mus

depressor mandibulae inserts on the al

Source : MNHN, Paris

110 ALYTES 20 (3-4)

x >

UW,

Fig. 8. — (a) Lysapsus limellus (FML 00791). Morphology corresponding to group IV (tab. 1). Scale bar:

0.5 mm. - (b) Pseudis minutus (FML 03676). Morphology corresponding to group LV (tab. 1). Scale

bar: 0.5 mm. - (c) Pseudis paradoxus (FML 00936). Morphology corresponding to group LV (tab. 1)

Scale bar: 0.5 mm. - Abbreviations: at, anulus Lympanicus; dm, m. depressor mandibulue: ds, m

dorsalis scapulae; sq, squamosal.

Source : MNHN, Paris

MANZANO, MORO & ABDALA 111

FAMILY MICROHYLIDAE

Brevicipitinae

Breviceps poweri (fig. 9a). — It has a depressor mandibulae muscle divided into two slips. The

anterior slip originates on the posterior end of the maxilla by a short tendon. Itis thick, short,

and approximately fusiform. The posterior slip originates on the otic process of the squamo-

sal. This slip is long, thick, and approximately triangular. Both branches insert on the

articular process of the lower jaw.

Microhylinae

Dermatonotus muelleri (fig. 9b).— It has a depressor mandibulae muscle divided into two slips.

The anterior slip originates on the ventral edge of the anulus tympanicus. Itis short, thick, and

approximately rectangular. The posterior slip originates from the dorsal fascia (at the level of

the mesial edge of the m. dorsalis scapulae and the m. latissimus dorsi, almost reaching the

posterior edge of the eyes). It is long, thick, approximately triangular, and very wide at the

origin. Both branches insert on the articular process of the lower jaw.

Elachistoscleis bicolor (fig. 90). - It has a single depressor mandibulae muscle. It originates from

the ventral and anterior edge of the anulus tympanicus and the dorsal fascia (at the level of the

anterior edge of the m. dorsalis scapulae and the m. latissimus dorsi). The m. depressor

mandibulae is a very wide and fan-shaped muscle that partially covers the tympanum, the

jugal and the posterior edge of the maxilla. A branch of cranial nerve VII passes through it.

The muscle inserts on the articular process of the lower jaw.

Phrynomerinae

Phrynomantis bifasciatus. — It has a depressor mandibulae muscle divided into two slips. The

anterior slip originates on the lateral edge of the anulus tympanicus. I is short, thick, and

approximately rectangular. The posterior slip originates on the dorsal fascia, and covers the

mesial edge of the m. cucullaris and the anteromedial edge of the m. dorsalis scapulae. I is a

fan-shaped muscle with a longer posterodorsal projection. It surrounds the posterior edge of

the tympanum. A branch of cranial nerve VIT passes through it. Both branches insert on the

articular process of the lower jaw.

FAMILY DENDROBATIDAE

Colostethus subpunctatus.— It has a depressor mandibulae muscle divided into two slips. The

ia (at the level of the m. dorsalis scapulae and the

m. levator mandibulae posterioris longus). I is a wide and fan-shaped muscle. The deep slip

originates on the posterior edge of the anulus tympanicus and the otic process of the

squamosal. It is short, small, and rectangular. Both branches insert on the articular process of

the lower jaw.

Dendrobates auratus (fig. 10a).— It has à depressor mandibulae muscle partially divided into

three slips. The anterior slip originates from the dorsal fascia (at the level of the m. dorsalis

scapulae and the m. levator mandibulae posterioris longus). Iis approximately triangular and

Source : MNHN, Paris

112 ALYTES 20 (3-4)

Fig. 9. - (a) Breviceps poweri (FML 03165). Morphology corresponding to group VIe (tab. 1). Scale bar:

0.5 mm. - (b) Dermatonotus muelleri (FML 05365). Morphology corresponding to group XHb (tab.

1). Scale bar: 0.5 mm. - (c) Elachistocleis bicolor (FML 00251). Morphology corresponding to group

XLa (tab. 1). Scale bar: 0.5 mm. - Abbreviations: at, anulus (ympanicus; dm, m. depressor mandibu-

lac; dma, m. depressor mandibulae anterior; dmmx, m. depressor mandibulae maxilaris: dmp, m.

depressor mandibulae posterior; dse, m. dorsalis scapulae: by, branch of the nerve VII: sq, squamosal.

Source : MNHN, Paris

dsc dmsp dmsa

sq at dmsp dsc

dmd

di d2

Fig. 10. - (al) Dendrobates auratus (FML 01722). Morphology corresponding to group Vb (tab. 1). Scale

bar: 0.5 mm. - (a2) Detail of deep m. depressor mandibulae of D. auratus. Scale bar: 0.5 mm. - (bl)

Epipedobates pictus (FML 03516). Morphology corresponding to group Vd (tab. 1). Scale bar:

mm. — (b2) Detail of deep m. depressor mandibulae of E. pictus. Scale bar: 0.1 mm. — (c) Hem

guineensis guineensis (FML 01244). Morphology corresponding to group VIa (tab. 1). Scale bar: 0.5

mm. - (d1) Arthroleptis tuberosus (FML 00051). Morphology corresponding to group VIb (tab. 1).

Scale bar: 0.3 mm. - (d2) Detail of deep m. depressor mandibulae of A. tuber le bar: 0.3

mm. Abbreviations: at, anulus tympanicus: dma, m. depressor mandibulae anterior: dmd, deep m.

depressor mandibulae, dmmx, m. depressor mandibule maxilaris: dmp, m. depressor mandibulae

posterior; dmsa, m. depressor mandibulae superficial anterior: dmsp, m. depressor mandibulae

Superficial posterior: dsc, m. dorsalis scapulue: sq, squamosal.

Source : MNHN, Paris

114 ALYTES 20 (3-4)

wider at the origin. The posterior slip originates from the dorsal fascia (at the level of the m.

dorsalis scapulae). X is approximately rectangular and as long as the anterior slip. A deep slip

originates on the posterior edge of the anulus tympanicus and the otic process of the

squamosal. It is short, approximately rectangular, wide at the origin. The three branches

insert on the articular process of the lower jaw.

Epipedobates pictus (fig. 10b). — It has a depressor mandibulae muscle divided into two slips:

the superficial one is partially divided. The anterior superficial slip, which is short and

rectangular, originates on the posterior edge of the anulus tympanicus. The posterior super-

ficial slip originates from the dorsal fascia (at the level of the posterolateral region of the

anteromesial part of the m. dorsalis scapulae). À deep slip, narrow and rectangular, originates

on the otic process of the squamosal and the posteroventral edge of the anulus tympanicus.

The three branches insert on the articular process of the lower jaw, the deep slip mesial to

superficial slips.

FAMILY HEMISOTIDAE

Hemisus guineensis guineensis (fig. 10c). — It has a depressor mandibulae muscle divided into

three slips. The anterior slip, short and fusiform, originates on the alary process of the maxilla

by a tendon. The middle rectangular slip originates on the otic process of the squamosal. The

posterior slip originates on the dorsal fascia (at the level of the suprascapula, partially

covering the m. dorsalis scapulae and the m. latissimus dorsi). I is triangular and broad,

curving toward the posterior region of the body. The three branches insert on the posterior

end of the quadrate.

FAMILY ARTHROLEPTIDAE

Arthroleptinae

Arthroleptis tuberosus (fig. 10d). — It has a depressor mandibulae muscle divided into three

slips. The anterior superficial slip originates on the posterior edge of the maxilla. It is long,

very thin, and approximately fusiform. The posterior superficial slip is fan-shaped and wide at

its origin on the dorsal fascia. It partially covers the m. dorsalis scapulae and m. lai

dorsi, and completely covers the deep slip. The deep slip, approximately rectangular and

slightly wider at its origin, originates on the otic process of the squamosal. It partially covers

the posterior border of the tympanum. The three branches insert on the articular process of

the lower jaw.

Arthroleptis stenodactylus stenodactylus. — I has a depressor mandibulae muscle divided into

three slips. The anterior superficial slip originates on the posterior edge of the maxilla. It is

long, very thin, and approximately fusiform. The posterior superficial slip, fan-

wide at its origin on the dorsal fascia, covers the m. dorsalis scapulae and m. ai

The deep slip, approximately rectangular, originates on the otic process of the oui | ni

partially covers the posterior border of the tympanum. The three branches insert on the

articular process of the lower jaw.

Source : MNHN, Paris

MANZANO, MORO & ABDALA 115

Schoutedenella sp. — It has a depressor mandibulae muscle divided into two slips. The anterior

slip, short and rectangular, is wide at the origin and originates on the posteroventral edge of

the anulus tympanicus. The posterior slip, fan-shaped and wider at the origin, originates on the

dorsal fascia (at the level of the m. dorsalis scapulae). Its posterior part partially covers the

suprascapula. Both branches insert on the articular process of the lower jaw.

Schoutedenella sylvatica. — It has a depressor mandibulae muscle divided into two slips. The

anterior slip, short and rectangular, is wide at the origin and originates on the posteroventral

edge of the anulus tympanicus. The posterior slip originates on the dorsal fascia (at the level of

the m. dorsalis scapulae). It is fan-shaped and wider at the origin with a posterior part that

partially covers the suprascapula. Both branches insert on the articular process of the lower

jaw.

Cardioglossa sp. (fig. 11a). — It has a depressor mandibulae muscle divided into two slips. The

anterior slip is short, rectangular, and thick, originating from the otic branch of the squamo-

sal at the posteroventral edge of the anulus tympanicus. The posterior slip is fan-shaped and

wider at the origin, originating on the dorsal fascia (at the level of the m. dorsalis scapulae and

the m. rhomboideus anterior). A nerve V branch passes through its fibers at mid-level. This slip

partially covers the anterior slip. Both branches insert on the articular process of the lower

jaw.

FAMILY RANIDAE

Dicroglossinae

Conraua crassipes. — It has a depressor mandibulae muscle divided into two slips. The anterior

slip, short and approximately rectangular, originates on the inferior edge of the anulus

tympanicus. The posterior fan-shaped slip, wider at the origin, originates on the dorsal fascia

(at the level of the dorsal part of the m. dorsalis scapulae). Both branches insert on the

articular process of the lower jaw by a short tendon.

Hoplobatrachus occipitalis. — It has a single depressor mandibulae muscle. It originates on the

inferior edge of the anulus tympanicus and the dorsal fascia (at the level of the anterodorsal

part of the m. dorsalis scapulae, covering most of it). The depressor mandibulae is a fan-shaped

and very wide muscle. It inserts on the articular process of the lower jaw by a short tendon.

Ptychadeninae

Piychadena mascareniensis hylaea. — X has a depressor mandibulae muscle divided into two

slips. The anterior slip, short and approximately rectangular, originates on the posteroventral

edge of the anulus tympanicus. The posterior slip, thin and approximately triangular, is wider

at the origin and originates on the dorsal fascia (at the level of the m. dorsalis scapulae). Both

branches insert on the articular process of the lower jaw.

Pyxicephalinae

Aubria subsigillata (fig. 11b).— It has a depressor mandibulae muscle divided into two slips. The

anterior slip, short, thick, and approximately rectangular, originates on the posteroventral

edge of the anulus 1ympanicus. The posterior fan-shaped slipis flat and wider at the origin, and

Source : MNHN, Paris

116 ALYTES 20 (3-4)

Fig. 11. = (a) Curdioglossa cyaneospila (FML 06477). Morphology corresponding to group XV (tab. 1).

igillata (FML03154). Morphology corresponding to group XIIb

(tab. D) ar: 0.5 mm. (c) Rana angolensis (FML 03188). Morphology corresponding to group

Va (tab. ). Scale bar: 0.5 mm. - Abbreviations: at, anulus tympanicus; dma, m. depressor mandibulae

anterior, dmp, m. depressor mandibulae posterior: dse, m. dorsalis scapulae; sq, squamosal

Source : MNHN, Paris

MANZANO, MORO & ABDALA 117

originates on the dorsal fascia. It partially covers the m. dorsalis scapulae and m. latissimus

dorsi. Both branches insert on the articular process of the lower jaw.

Raninae

Rana albolabris. — It has a depressor mandibulae muscle divided into two slips. The anterior

slip, short and approximately triangular, originates on the posteroventral edge of the anulus

tympanicus. The posterior slip originates on the dorsal fascia covering the m. dorsalis

scapulae. Both branches insert on the articular process of the lower jaw.

Rana angolensis (fig. 110). — It has a depressor mandibulae muscle divided into two slips. The

anterior slip, short and approximately rectangular, originates on the posteroventral edge of

the anulus tympanicus. The posterior slip originates on the dorsal fascia (at the level of the m.

dorsalis scapulae). Xt is a fan-shaped slip, wider at the origin and partially covers the anterior

slip. Both branches insert on the articular process of the lower jaw.

Rana lepus. — It has a depressor mandibulae muscle divided into two slips. The anterior slip

originates on the inferior edge of the anulus tympanicus. It is short thick and approximately

rectangular. The posterior slip originates on the dorsal fascia (at the level of the m. dorsalis

scapulae) and is fan-shaped and wider at the origin, with the fibers partially divided. Both

branches insert on the articular process of the lower jaw.

FAMILY PETROPEDETIDAE

Phrynobatrachus acutirostris (fig. 12a).-— It has a depressor mandibulae muscle divided into two

slips. The anterior slip, short and approximately rectangular, originates on the posteroventral

edge of the anulus tympanicus. The posterior slip originates on the dorsal fascia (at the level of

the m. dorsalis scapulae). It is approximately triangular, flat, thin and wider at the origin. Both

branches insert on the articular pro of the lower jaw.

Phrynobatrachus calcaratus.— Y has a depressor mandibulae muscle divided into two slips. The

anterior slip, short and approximately rectangular, originates on the posteroventral edge of

the anulus tympanicus. The posterior slip originates on the dorsal fascia (at the level of the m.

dorsalis scapulae). It is approximately triangular, flat, thin and wider at the origin. A nerve V

branch passes through its fibers at mid-level. Both branches insert on the articular process of

the lower jaw.

FAMILY HYPEROLHDAE

Hyperoliünae

Afrixalus orophilus (fig. 12b). — It has a depressor mandibulae muscle divided into two

branches. The anterior slip, approximately rectangular and narrow, originates on the antero-

ventral edge of the anulus tympanicus. The posterior slip originates on the dorsal fascia (at the

level of the m. levator mandibulae posterioris longus), and the posterior edge of the anulus

Lympanicus. Y is approximately triangular and wider at the origin. Both slips partially

surround the tympanum. Both branches insert on the articular process of the lower jaw by a

short and thin tendon.

Source : MNHN, Paris

118

Fig

ALYTES 20 (3-4)

12. (a) Phrynobatrachus acutirostris (DIAM 006). Morphology corresponding to group Va (tab. 1).

Scale bar: 0.5 mm. - (b) Afrixalus orophilus (FML 03229). Morphology corresponding to group XIV

(©) Rhinophrynus dorsalis (FML 01720). Morphology corresponding 10

5 mm. - Abbreviations: at, aulus tympanicus: dma, m. depressor

dmp, m. depressor mandibulae posterior; dse, m. dorsalis scapulue: sq, squamo-

(tab. 1). Scale bar:

group IX (tab. 1).

mandibulae anterio

sal.

Source : MNHN, Paris

MANZANO, MORO & ABDALA 119

Chrysobatrachus cupreonitens. — It has a depressor mandibulae muscle divided into two slips.

The anterior slip, approximately rectangular, originates on the posteroventral edge of the

anulus tympanicus. The posterior slip originates on the dorsal fascia (at the level of the m.

dorsalis scapulae). It is fan-shaped, wider and very thin at the origin, and thicker and narrower

at the insertion point. The posterior slip partially covers the anterior slip. Both branches insert

on the articular process of the lower jaw.

Kassininae

Kassina senegalensis angeli. — It has a depressor mandibulae muscle divided into two slips. The

anterior slip, approximately rectangular and short, originates on the posteroventral edge of

the anulus tympanicus. The posterior slip originates on the dorsal fascia (at the level of the m.

dorsalis scapulae). It is triangular, wider at the origin with a nerve V branch passing through

it. This slip partially covers the anterior slip. Both branches insert on the articular process of

the lower jaw by a short tendon.

Leptopelinae

Leptopelis notatus.—1t has a depressor mandibulae muscle divided into two slips. The anterior

slip, short, bulky and approximately fusiform, originates on the posteroventral edge of the

anulus tympanicus. The posterior slip originates on the dorsal fascia (at the level of the m.

dorsalis scapulae). It is triangular, very wide at the origin and partially covers the anterior slip.

Both branches insert on the articular process of the lower jaw.

FAMILY RHINOPHRYNIDAE

Rhinophrynus dorsalis (fig. 120). — It has a depressor mandibulae muscle divided into two slips.

The anterior slip, approximately triangular and elongated, originates on the otic process of

the squamosal. The posterior slip, very narrow, thin, and curved, originates on the dorsal

fascia (at the level of the middle part of m. dorsalis scapulae). Both branches insert on the

articular process of the lower jaw.

DISCUSSION

There is a considerable variation in the morphology of the m. depressor mandibulae.

Nevertheless, fifteen morphological groups can be distinguished within the anuran species

analysed (tab. 1). Although many studies have shown that the variation of the jaw muscles

involves either their origin or insertion (ECKER, 1889; BEDDARD, 1908, 1911; EDGEWORTH,

1935; BaLDAUr, 1955), in these fifteen groups, the insertion points of the m. depressor

mandibulae is relatively invariant (see BAUER, 1997). The area of the origin on the posterior

region of the head is also stable (allowing the muscle to open the mouth), except in Hemisus,

Arthroleptis and Breviceps which have an extra anterior branch coming from the maxilla.

Most variation involves muscle shape and often demonstrates intra- or intergeneric variation.

We found the following variation in the taxa analysed.

Source : MNHN, Paris

120

ALYTES 20 (3-4)

Table 1.- Schematic representation of the different conditions of the depressor mandibulae muscle

in anurans, with notation following STARRETT (1968) and modifications in this work. Each

group is identified by a Roman number (I to XV). Different morphotypes in a same group

are identified by a letter (a, b, c or d). Superficial and deep slips of the depressor

mandibulae muscle, when present, are identified by a subindex (1 and 2, respectively).

Group. Scheme Taxa STARRETT, Present

1968 work

Xenopus Dfsq Dfa

Abytes DFa DFa

Il ( >

a |Chacophrys DFSQAT DFSQ

Bufo arenarum; B. paracnemis;

B. spinulosus | sQ

D |Rhinophrynus; Melanophryniscus sQ sQ

B. g. major, Lepidobatrachus; Crinia _— Sqat

Ceratophrys DFSQAT SQat

€ |Hymenochirus sQ SQat

Leptodactylus bufonius; L. chaquensis;

Physalaemus biligonigerus;

Crossodactylus; Taudactylus | -—— DFsq

IV 5 L

Pseudidae; Scimax | Dfsqat

Gastrotheca DFsq DFsqat

Hyla andina DF Dfsqat

| = al

Limnodynastes fletcheri; L. dorsalis;

L. ornatus; Phrynobatrachus;

a2 | Rana angolensis; Uperoleia borealis | -— DFSQ-ats

2 bl

= b2 | Dendrobates DFSQat | DF-DF-sqats

Source : MNHN, Paris

Table 1.- (Continuation)

Group Scheme Taxa STARRETT, Present

1968 work

(cont.)

j Colostethus; Limnodynastes dumerili —— DF-sqat

e2 | Pseudopaludicola boliviana DFSQAT | DF-sqats

\r di

ee d2 | Epipedobates 2e DF-at-squ

vi | Ë

a |Hemisus g. guineensis | -— DF-sq-mx

= bl

Arthroleptis tuberosus; À. s.

b2 |stenodactylus | DF-squ-mx

== ae Es

Centrolenidae Dfa Sqat

a |Rhinoderma sQ Sqat

= b |Agalychnis SQ-at

VII

Phyllomedusa sQ DF-SQ-at

Argenteohyla | -— DF-SQ-at

Acris; Notaden; Pychohyla; Smilisca dfSQ

Rhinophrynus dfSQ

Source : MNHN, Paris

122 ALYTES 20 (3-4)

Table 1.- (Continuation)

Group Scheme Taxa STARRETT, | Present

1968 work

Hyla boans; Pleurodema borelli;

Telmatobiinae | DF-sq

Plectrohyla DF DF-sq

à Nr

Phrynohyas | DF-DF-at |

2 a |Elachistocleis; Hoplobatrachus | -—- DFat

Aubria; Chrysobatrachus; Conraua;

Dermatonotus; Leptopelis;

Phrynobatrachus; Phrynomantis;

Ptychadena; Rana albolabris; R. lepus:

Schoutedenella DF-atl

b |Kassina DF AT DF-atl

SE € |Limnodynastes convexiusculus _ DF-at2

XII

Æ, Uperoleia aspera = DFSQat

XIV

É Afrixalus orophilus | DFsq-at

L | 2

Cardioglossa UE DF-sqat

Pipids and discoglossids have a single m. depressor mandibulae. STARRETT (1968) found

that the apparent single muscle in Xenopus is a tripartite muscle, not very differentiated, but

with three different origins (anulus tympanicus, dorsal fascia and squamosal). The only origin

of the muscle on the dorsal fa in our observations, with no apparent divisions, lead us to

consider the m. depressor mandibulae morphology in Xenopus as a single one. Our descrip-

tions of the Alytes m. depressor mandibulae agree with STARRETT's. However, in her descrip-

tions, Alytes and Xenopus have different morphological patterns. In our case both taxa belong

to the same group I.

Source : MNHN, Paris

MANZANO, MORO & ABDALA 123

Interestingly, both of these groups traditionally considered basal related to the Neoba-

trachia (FORD & CANNATELLA, 1993; CANNATELLA, 1999; MAGLIA et al., 2001; Haas, 2001)

have a simple morphology of the m. depressor mandibulae, with an origin on the dorsal fascia.

GRIFFITHS (1954) considered this state the primitive morphology of the muscle; but STARRETT

(1968) did not agree with his hypothesis. Hymenochirus has a bulky m. depressor mandibulae

like bufonids. This morphology was observed by STARRETT (1968) in some isolated species of

different families. This situation is described by her also in Rhinophrynus. We could not find

any fibers arising from the crista parotica in Rhinophrynus dorsalis as STARRETT (1968)

described; we observed that all the muscle origins are at the otic ramus of the squamosal.

The Bufonidae analysed have a single bulky muscle similar to Lepidobatrachus llanensis,

Ceratophrys cranwelli and Chacophrys pierotti (Leptodactylidae, Ceratophryinae) (group IL,

tab. 1). Ceratophryines have been considered alternatively related to bufonids (LIMESES, 1965),

as a subfamily of the Leptodactylidae (Ceratophryinae) (LYNCH, 1973; HEYER, 1975; DUEL-

LMAN & TRUEB, 1985) or a separate family (REIG, 1972). The particular morphology of the

depressor mandibulae muscle in the Bufonidae (also considered by STARRETT, 1968, as a single

slip with a non-fascial origin), seems to be a conservative condition and could be considered

a primary homology shared by the Ceratophryinae-Bufonidae groups.

We found three more morphological patterns (groups III, V and X) for the rest of

leptodactylids analysed. Only the species included in our group III have the pattern consid-

ered by STARRETT (1968) as the common type of the family, which includes origin from the

squamosal. However, the inclusion of Crossodactylus in this group is at odds with her results.

PALAVECINO (2000), analysing four of our species (Physalaemus biligonigerus, Leptodactylus

chaquensis, L. bufonius and Pleurodema borelli), found very different patterns. AI the species,

except P horellii, are in our group II. The most striking difference is that in her results the m.

depressor mandibulae origin never includes the squamosal (as it is in STARRETT, 1968, and in

our results). This difference is probably related with the age (stage) of the specimens analysed.

Another difference involves the anulus tympanicus that in our results in no case has any m.

depressor mandibulae fiber attached.

HEYER (1975) found six character states of the m. depressor mandibulae morphology in

the leptodactylid frogs he analysed. Among the genera analysed here, ten were also observed

by him. Some of his results are coincident with ours. Crossodactylus, Leptodactylus and

Physalaemus share the same pattern, described as B by HEYER (1975), and are together in our

group III. In the other cases there are no coincidences at all.

Among hylids, we found seven different m. depressor mandibulae morphologies (tab. 1).

Within the genus Phyllomedusa we considered only those species with grasping feet. AII these

species have the same morphological type of the depressor mandibulae muscle with three

branches (“partes: pars timpanica, pars Squamosal and pars scapularis”) (group VIIL, tab.1),

similar to Phyllomedusa atelopoides (MANZANO, 1997), P perinesos (DUELLMAN, 1973), P

tomopterna and P. tarsius (CANNATELLA, 1980). The Phyllomedusa buckleyi group lacks the

posterior branch of the depressor mandibulae muscle (“pars scapularis”) as does Agalvchnis

(CANNATELLA, 1980). This particular condition was previously mentioned as one of the

synapomorphies of the buckleyi group, suggesting that the genus Phy/lomedusa is an unnatu-

ral assemblage of species (CANNATELLA, 1980). Moreover, some authors considered that the

buckleyi group should be accorded generic status (DUELLMAN, 1968, 1969), representing a

Source : MNHN, Paris

124 ALYTES 20 (3-4)

separate group closer to Agalychnis than to Phyllomedusa. Al the species of this genus

analysed by STARRETT (1968), including P boliviana, have the condition SQ. We found that P.

boliviana and the other two species of this genus have the condition DF-SQ-at. The morphol-

ogy of the m. depressor mandibulae in Agalychnis saltator (group VITb) is similar to that of

centrolenids described here (group VIIa), a condition also observed by RUEDA-ALMONCAID

(1994) in Centrolene geckoideum. However, there is a big variation within the Centrolenidae,

exhibiting a depressor mandibulae muscle composed of two or three branches and different

designs (MANZANO, 2000), so the character is of limited usefulness in the comparison of both

groups. STARRETT (1968) considered a single slip lying over the squamosal as the pattern of

adult centrolenids, but her results disagree with MANZANO (2000) and ours.

Within the Hylinae, a large morphological diversity makes it impossible to assign them a

single defined pattern. For example, Acris crepitans, Ptychohyla ignicolor and Smilisca sila

have a morphotype similar to that of Notaden (a Myobatrachidae) and Rhinophrynus dorsalis

(a Rhinophrynidae) (group IX, tab. 1). Our results disagree with STARRETT’S (1968). She

found that Acris and Prychohyla have a DF condition instead of df-SQ as in our specimens. In

her results Smilisca has the condition DFsq, instead of df-SQ as in ours. Hyla andina and

Scinax fuscovarius have a morphotype similar to that found in the Pseudidae and Gastrotheca

(a Hemiphractinae hylid) (group IV, tab. 1). In our results, Gastrotheca has the condition

DFsqat. STARRETT (1968) found that the only hylid presenting an origin including the anulus

tympanicus is Cryptobatrachus fuhrmanni. The morphology of the depressor mandibulae

muscle in Hyla boans and Plectrohyla guatemalensis is similar to the Telmatobinae (Lepto-

dactylidae) analysed and Pleurodema borellit (another leptodactylid) (group X, tab. 1).

Argenteohyla siemersii has a morphology similar to that of the Phyllomedusa species analysed

(group VIID). Phrynohyas venulosa has a morphology that differs from other groups and also

from the rest of the hylids (group XI). Adding to this, STARRETT (1968) found individual

variation for the m. depressor mandibulae morphology in a series of Hyla crucifer from the

same locality. No intraspecific variation was found in Hyla in this study.

The pseudids analysed (group IV, tab. 1) have a very characteristic morphological pattern

of the m. depressor mandibulae, that is also present in Scinax fuscovarius, Gastrotheca and

Hyla andina. This group includes species with the condition DFsqat. STARRETT (1968) found

the condition DFa for Pseudis paradoxa, again at odds with our results. It is interesting that

Da SiLva (1999) hypothesized in his phylogenetic analysis a relationship between Pseudidae

and Hylinae.

Another example of intrageneric variation is found in the family Myobatrachidae.

Within this family, Limnodynastes fletcheri, L. dorsalis and L. ornatus (Limnodynastinae) and

Uperoleia borealis (Myoba nae) have similar morphologies (group Va, tab. 1) that differ

from the rest of the myobatrachids previously analysed (Limnodynastes convexiusculus, group

XIIe; Crinia georgiana, group lic; and Taudactylus diurnus, group HD). Uperoleia aspera,

group XIV (Myobatrachinae) also presents a distinctive morphotype, not shared by other

myobatrachids (groups Va and Ve, tab. 1). Notaden (Limnodynastinae) (group IX) does not

share the myobatrachid morphology of the depressor mandibulae muscle.

STARRETT (1968) described a depressor mandibulae muscle represented by three branches

for the Dendrobatidae: two superficial, originating on the dorsal fascia and the anulus

Lympanicus, and a deep branch originating on the squamosal (based on

Epipedobates pictus).

Source : MNHN, Paris

MANZANO, MORO & ABDALA 125

The condition that we observed in Colosthetus subpunctatus (group Ve, tab. 1) was also

described by MYErs et al. (1991) for Aromobates and by La MARCA (1995) for Mannophryne.

In this case, the m. depressor mandibulae has two branches. Thus, there are at least two

character states for the family with variation in the superficial branch. The particular

disposition of the branches in relation to associated skull elements is common to most

dendrobatids, which is why we insist on the existence of a characteristic morphotype for the

family (groups Vb, Ve and Vd, tab. 1). In fact, Myers & FORD (1986), FORD (1989) and FORD

& CANNATELLA (1993) considered the arrangement of the superficial slip of the m. depressor

mandibulae as a synapomorphy of the Dendrobatidae.

The condition of the depressor mandibulae muscle in Dermatonotus and Elachistocleis,

group XII (Microhylinae) is similar to that described by BURTON (1983) for other microhy-

lids. Three origin points of the muscle are present in all microhylids analysed by him, except

for the Albericus and Choerophryne data of Burton. Breviceps poweri (group Vic) has a

morphology of the m. depressor mandibulae shared by Arthroleptis (group VIb) and Hemisus

(group VIa). The morphotype of Microhylinae and Phrynomerinae depressor mandibulae is

shared by some Ranidae (group XI1b), also mentioned by Hoyos (1999), including a nerve V

branch that runs through the superficial slip of the muscle in most of the microhylids. This

branch is also present in a leptodactylid, Pseudopaludicola boliviana (group Ve) and in an

arthroleptid, Cardioglossa (group XV). STARRETT (1968) considered the pattern for the species

of ranids as DFAT and DFSQAT. We define also two morphotypes: DFat and DF-at, but

neither of them includes a squamosal origin. Hoyos (1999) found a third branch arising from

the squamosal. Although we could not find this branch, we agree with him with respect to

sampling matters since we did not analyse all the same species.

The variation in our analysis far exceeds that described by STARRETT (1968). In fact, we

found only weak coincidences among her and our patterns (bufonids and some leptodacty-

lids).

This leads us to believe that despite me great number of specimens analysed by various

authors (GAuPP, 1896; BIGALKE. ,1963; LIMESES, 1965, 1969; STARRETT,

1968; MCDIARMID, 1971; HEYE MERSON, 1976; CANNATELLA, 1980; TYLER & DAVIES,

1980; Davies & BURTON, 1982; BURTON, 1983a-b; LYNCH, 1986, 1993; CANNATELLA & TRUEB,

1988; Myers et al., 1991: WaKkE, 1993; RUEDA-ALMONACID, 1994; BURTON & ZWE 1995;

La MaRCA, 1995; MANZANO, 1997, 2000; Hoyos, 1999), examination of additional specimens

will doubtless uncover additional variation. Our frequent disagreement with STARRETT

(1968), HeyeR (1975), Hoyos (1999) or PALAVECINO (2000) constitutes the rule rather than the

exception. We think that this situation expresses the true characteristic of this muscle: its

extreme variability. We agree with LYNCH’s (1973) conclusion that this muscle is of little use to

analyse the relationships of eleutherodactyline frogs, and further extend his conclusion to the

majority of the taxa observed by us.

The great diversity of the m. depressor mandibulae correlates with the plethora of

patterns mentioned by TRUEB (1993) for cranial morphology in Anura.

We find no evidence that variations in the m. depressor mandibulae are associated with

particular habits. For example, Hemisus g. guineenseis (group VIa, tab. 1) has a distinct

morphology of the depressor mandibulae, formed by three well-differentiated slips with a

particular anterior branch originating on the alary process of the maxilla by a tendon. This

Source : MNHN, Paris

126 ALYTES 20 (3-4)

morphology can be interpreted as a muscle modification in relation to the animals excavating

habits, and the variation in the origin of this branch can be explained according to WakE’s

(1993) interpretations. However, the presence of a similar morphotype in Arthroleptis, a

non-excavating anuran, would falsify the hypothesis that intends to explain these morpholo-

gies as a function of a particular habit.

Nevertheless, there is a structural base of the muscle determined by its function as a

mouth-opener. As the m. depressor mandibulae opens the mouth, it must originate and

insert in areas where this function is possible. The variations in the muscle shape, possibly

determined by epigenetic factors, would be related to the switch from aquatic to terrestrial life

habits. The remarkable consistency of this muscle in saurians and ophidians (Moro & Abdala,

personal communication) allows us an epigenetic interpretation for the anatomic variability

in anurans. However, the epigenetic action has constraints. For example: Gastrotheca gracilis,

G. chrysosticta and G. christiani have different development, direct and with tadpole larvae,

needing very different environments to develop. However, and despite of variable epigenetic

pressures on the different species mentioned, the m. depressor mandibulae morphology is

exactly the same in all of them. Changes during early development of the jaw muscles with

little or no effect on the adult morphology have been described previously for direct devel-

oping frogs (Eleutherodactylus coqui, by HANKEN et al., 1997).

Likewise, the presence of the same depressor mandibulae muscle structure in anurans with

different diets (STARRETT, 1968) does not agree with EMERSON’s (1985) idea about m. depressor

mandibulae shape being directly related to feeding habits, which limits its action as an

epigenetic factor.

CONCLUSIONS

The morphology of the m. depressor mandibulae is too variable to be of much use in

determining relationships or indicating habitat specializations. This variability seems to

represent the most remarkable characteristic of this muscle.

In spite of the extreme variability of this muscle, we recognize some morphotypes that

are useful to characterize some groups. They allow the use of the m. depressor mandibulae

pattern as one character to be used in phylogenetic hypotheses including those groups.

ACKNOWLEDGEMENTS

We are grateful to Ronald Heyer, Tom Burton, Marissa Fabrezi and Greg Ellis for reviewing the

manuscript and the language. Marvalee Wake made valuable comments. Raymond Laurent allowed us to

use his African collection, and M. J. Tyler generously made some specimens available. Thanks to J. Calvo

for his assistance with the table. This work was partially supported by CONICET.

Source : MNHN, Paris

MANZANO, MORO & ABDALA 127

LITERATURE CITED

BALDAUF, R. J., 1955. - Contributions to the cranial morphology of Bufo w. woodhousei Girard. Texas J.

Sci., 7 (3): 275-311.

BAUER, W. J., 1997. — A contribution to the morphology of visceral jaw opening muscle of urodeles

(Amphibia: Caudata). J Morph., 233: 17-97.

BeDDaRD, F. E., 1908. - On the musculature and other points in the anatomy of the engystomatid frog

Breviceps verrucosus. Proc. zool. Soc. London, 1908: 11-41.

me 1911. - Contributions to the anatomy of the Anura. Proc. zool. Soc. London, 1911: 393- 412.

BIGALKE, R., 1926. - Zur Myologie der Erdkrôte (Bufo vulgaris, Laurenti). In: BERGMANN & SPRINGER,

Zeitschrift für Anatomie und Entwicklungsgeschichte, München & Berlin: 236-353.

Bock, W. & SHEAR, C. H., 1972. — A staining method for gross dissection of vertebrate muscles. Anar.

Anz., 130: 222-227.

BURTON, T., 19834. - The musculature of the Papuan frog Phrynomantis stictogaster (Anura: Microhy-

lidae). J Morph., 175: 307-324.

= 1983b. — The phylogeny of the Papuan subfamily Asterophryinae ( Anura: Microhylidae). PhD Thesis,

Adelaide, University of Adelaide: 1-283.

Sn 1986. — A reassessment of the Papuan subfamily Asterophryninae (Anura: Microhylidae). Rec. $.

Aust. Mus., 19 (19): 405-450.

BURTON, T. & ZWEIFEL, R., 1995. — À new genus of genyophrynine microhylid frogs from New Guinea.

Am. Mus. Novit., 3129: 1-7.

CANNATELLA, D. C., 1980. - Two new species of Cenrrolenella from Bolivia (Anura: Centrolenidae). Proc.

biol. Soc. Washington, 93 (3): 714-724.

_—— 1999. — Architecture. Cranial and axial musculoskeleton. /n: R. MCDiarMiD & R. ALTIG (ed.),

Tadpoles - The biology of anuran larvae, Chicago & London, The University of Chicago Press:

52-91.

CaNNATELLA, D. C. & TRuëR, L., 1988. — Evolution of pipoid frogs: morphology and phylogenctic

relationships of Pseudohymenochirus. JL Herp., 22: 439-456.

Da SILVA, H., 1999, — Phylogenetic relatior of the family Hylidae with emphasis on the relationships

within the subfamily Hylinae (Amphibia: Anura). PhD Thesis, Lawrence, The University of

Kansas: 1-175.

Davis, M. & BURTON, T., 1982. - Osteology and myology of the gastric brooding frog Rheobatrachus

silus Liem (Anura; Leptodactylidae). Ausr. J. Zool., 30: 503-521.

DuELLMAN, W., 1968. - The genera of phyllomedusine frogs (Anura: Hylidae). Univ. Kansas Publ. Mus.

nat. Hist., 18: 1-10.

= 1969. — Phyllomedusa buckleyi Boulenger: Variation, distribution and synonymy. Herpetologica, 25

—… 1973. - Description of new hylid frogs from Colombia and Ecuador. Herpetologica, 29 (3)

DUELLMAN, W. E. & TRurS, L., 1985. - Biology of amphibians. New York, MeGraw-Hill, “198

1-670.

r, A., 1889. - The anatomy of the frog. Oxford, Clarendon Press: 1-450.

WortH, F.H., 1935.- The cranial muscles of vertebrates. Cambridge, Cambridge Univ. Press: i-viii

+ 1-493.

EMERsoN, S., 1976. - À preliminary report on the superficial throat musculature of the Microhylidae and

its possible role in tongue action. Copeia, 1976: 546-551.

= 1985. - Skull shape in frogs - correlations with diet. Herperologica, 41: 171-188.

Forp, L. S., 1989. — The phylogenetic position of poison-dart frogs (Dendrobatidae ): reassessment of the

ncobatrachian phylogeny with commentary on complex character systems. PhD Thesis, Lawrence,

The University of Kansas: i-xi + 1-307.

Forp, LS. & CaNNATELLA, D., 1993. - The major clades of frogs. Herp. Mon., 7: 94-117.

Gaurr, E., 1896. - 4. Ecker's und R. Wiedersheim's Anatomie des Frosches. Vol. 2. Braunschweg

Friedrich Vieweg & Sohn: 1-961.

Source : MNHN, Paris

128 ALYTES 20 (3-4)

GrirrrrHs, L., 1954. - On the “otic element” in Amphibia, Salientia. Proc. zool. Soc. London, 124: 35-50.

_ 1963. - The phylogeny of the Salientia. Biol. Rev., 38: 241-292.

Haas, À., 2001. - Mandibular arch musculature of anuran tadpoles, with comments on homologies of

amphibian jaw muscles. J Morph., 247 (1): 1-33.

HaNkEN, J., KLYMKkowskY, M., ALLEY, K. & JENNINGS, D, 1997. — Jaw muscle development as evidence

for embryonic repatterning in direct-developing frogs. Proc: r. Soc. London, (B), 264: 1349-1354.

Hever, W. R.,1975. — À preliminary analysis of the intergeneric relationships of the frog family

Leptodactylidae. Smithson, Contrib, Zool., 199: 1-55.

Hovos, J. M., 1999. - The musculus depressor mandibulae complex in frogs, with special remarks on

ranids. Amphibia-Reptilia, 20: 265-278.

La Marca, E., 1995. - Biological and systematic synopsis of a genus of frogs from northern mountains

of South America (Anura: Dendrobatidae: Mannophryne). Bull. Maryland herp. Soc., 31 (2): 40-77.

Limeses, C., 1965. — La musculatura mandibular en los ceratofrinidos y formas afines (Anura: Cera-

tophrynidae). Physis, 25 (70): 41-58.

= 1969. — Lepidobatrachus Budgett (Anura, Ceratophrynidae). Nota miolôgica complementaria.

Physis, 28 (76): 127-134.

Lynou, J. D. 1973. - The transition from archaic to advanced frogs. Jr: J. L. ViaL (ed.), Evolutionary

biology of the anurans: contemporary research on major problems, Columbia, University of Mis-

souri Press: 133-182.

ee 1986. - The definition of the Middle American clade of Eleutherodactylus based on jaw musculature

(Amphibia: Leptodactylidae). Herpetologica, 42 (2): 248-258.

a 1993. — The value of them. depressor mandibulae in phylogenetic hypotheses for Eleutherodactylus

and its allies (Amphibia: Leptodactylidae). Herpetologica, 49 (1): 32-41.

MaGLia, À., PUGENER, À. & TRUEB, L., 2001. - Comparative development of anurans: using phylogeny

to understand ontogeny. Am. Zool., 41 (3): 538-551.

MANZANO, A. S., 1996. — Andlisis miolégico de la familia Pseudidae (Amphibia: Anura). PhD Thesis,

Tucumän, Universidad Nacional de Tucumän: 1-207.

+ 1997. — Estudio comparativo de la miologia de la cintura pectoral de algunos phyllomedusinos

(Anura: Hylidae). Boll. Mus. reg, Sci. nat. Torino, 15: 255-277.

2000. - Miologia pectoral de algunos Centrolenidae (Amphibia: Anura). Cuad. Herp., 14(1): 27-45.

McDiarMi, R. W., 1971. - Comparative morphology and evolution of frogs of the neotropical genera

Atelopus, Dendrophryniscus, Melanophryniscus, and Oreophrynella. Sci. Bull. nat. Hist. Mus. Los

Angeles Co. 12: 1-66.

Myers, C. W. & FORD, L.S., 1986. - On Aropophrynus, a recently described frog wrongly assigned to the

Dendrobatidae, Am. Mus. Novit., 2843: 1-

Myers, C. W., PAOLILLO, À. & DaLY, J., 1991. - Discovery of a defensively malodorous and nocturnal

frog in the family Dendrobatidae: phylogenctic significance of a new genus and species from the

Venezuelan Andes. Am. Mus. Novit., 3002: 1-33.

PALAVECINO, P., 2000. - Desarrollo de la musculatura mandibular e hioidea en Leptodactylinae del noroeste

argentino. PhD Thesis, Tucumän, Argentina, Universidad Nacional de Tucumän: i-vii + 1-138.

ReG, O., 1972. - Macrogenioglottus and the South American bufonid toads. /n: F. BLAIR (ed.), Evolution

in the genus Bufo, Austin, University of Texas Press: 14-36.

Ru, O. & Limrses, C. E., 1963. — Un nuevo género de anuros ceratofrinidos del distrito chaqueño.

Physis, 24 (67): 113-128.

RUHDA-ALMONACID, J. V., 1994. - Estudio anatémico y relaciones sistemäticas de Centrolene geckoideum

(Salientia: Anura: Centrolenidae). Trianca, 5: 133-187.

, 1987. - Systematies and distribution of the Mexic ican rainfrogs of the

utherodactylus golimeri group (Amphibia: Leptodactylidae). Fieldiana: Zoo, 3%: 1-51.

STARRETT, P., 1968. - The phylogenetic significance of the jaw musculature in anuran amphibians. PhD

of Michigan:

Patterns of the cranial diversity among the Lissamphibia. Jr: J. HANKEN & B. K. HAL

kull, vol. 2, Patterns of structural and systematic diversity, Chicago, University of

go Pre 43

TyueR, M. & Davies, M. 1980. - The status of Kankanophryne Heyer & Liem (Anura: Leptodactylidae).

Trans. r. Soc. S. Aust., 104: 17-20.

SAvaAG

Source : MNHN, Paris

MANZANO, MORO & ABDALA

129

Wake, M., 1993. - The skull as a locomotor organ. /n: J. HANKEN & B. K. HALL (ed.), The skull, vol. 3,

Functional and evolutionary mechanisms, Chicago, University of Chicago Press: 197-240.

WASsERSUG, R., 1976. — A procedure for differential staining of cartilage and bone in whole formalin-

fixed vertebrates. Srain Technol., 51 (2): 131-134.

APPENDIX 1

MATERIAL ANALYSED

FAMILY DISCOGLOSSIDAE

Alytes obstetricans (1).- FML 02782.

FAMILY PIPIDAE

Dactylethrinae

Xenopus laevis (2). - FML 03162.

Hymenochirus boettgeri camerunensis (1). - FML 03113.

FAMILY BUFONIDAE

Bufo arenarum (7). - FML 00540, 04722.

Bufo granulosus major (5). - FML. 01052, 04224.

Bufo paracnemis (2). - FML 02168, 00364.

Bufo spinulosus (4). - FML 00759.

Melanophryniscus rubriventris rubriventris (8). - FML 02502, 02116.

FAMILY LEPTODACTYLIDAE

Ceratophryinae

Ceratophrys cramvelli (3). - FML 0492.

Chacophrys pierottii (2). - FML 01020.

Lepidobatrachus llanensis (3). - FML 01095, 05290, 08902.

Hylodinae

Crossodactylus gaudichaudit (1). - FML 03497.

Leptodactylinae

Leptodactylus bufonius (5). FML 04642, 05364.

Leptodactylus chaquensis (4). - FML 05490, 05111.

Physalaemus biligonigerus (10). - FML 05288, 05986.

Pleurodema borellit (6). - ML 01377, 02694.

Pseudopaludicola boliviana (7). - FML 04307, 04309.

Telmatobiinae

Alsodes sp. (2). - FML 04435.

Batrachyla sp. (1). - FML 03713.

Hylorina sylvatica (1). - FML 03910.

Telmatobius laticeps (3). - FML 03960.

Telmatobius oxycephalus (3). - FML 02529.

Telmatobius scrocchit (3). - FML 05772.

FAMILY MYOBATRACHIDAE

Myobatrachinae

Crinia georgiana (1). - DIAM 015.

Taudacrylus diurnus (1). - FML 03774.

Uperoleia aspera (1). - DIAM 017.

Uperokeia borealis (D. - DIAM 016.

Limnodynastinae

Linmodynastes dumerili (1). DIAM 011.

Limmodynastes convexiusculus (D. DIAM 012.

Source : MNHN, Paris

130 ALYTES 20 (3-4)

Limnodynastes dorsalis (1). - DIAM 008.

Limnodynastes fletcheri (1). - DIAM 009.

Limmodynastes ornatus (1). - DIAM 010.

Notaden melanoscaphus (16). - FML 03783, DIAM 013.

Notaden nichollsi (1).- DIAM 014.

FAMILY RHINODERMATIDAE

Rhinoderma darwini (4).— FML 03694.

FAMILY HYLIDAE

Hemiphractinae

Gastrotheca gracilis (6). - FML 01769, 02209.

Gastrotheca christiani (2 adults and 3 juveniles). - FML 02117, 02918.

Gastrotheca chrysosticta (2). - FML 02759.

Hylinae

Acris crepitans (1).—FML 09478.

Argenteohyla siemersii (4). - FML 00056.

Hyla andina (3). FML 04758, 00698.

Hyla boans (1).-RL 105039, FML 09543.

Phrynohyas venulosa (3). FML 02303, 02712.

Plectrohyla guatemalensis (1). RL 64105.

Prychohyla ignicolor (1).- RL 137222.

Scinax fuscovarius (4). — FML 04635, 02716, 02754.

Smilisca sila (1). RL 918887.

Phyllomedusinae

Agalychnis saltator (1).- FML 09541.

Phyllomedusa boliviana (4). FML 01345, 02706.

Phyllomedusa hypocondrialis (4). FML 01483, 04286.

Phyllomedusa sauvagii (4). FML 04899.

FAMILY CENTROLENIDAE

Centrolene grandisonae (1). - FML 04980.

Cochranella ignota (3). - FML 04965, UVC 12091.

FAMILY PSEUDIDAE

Lysapsus limellus (3). FML 04341, 00791.

Pseudis minutus (3). - FML 03676.

Pseudis paradoxus (6). - FML 00708, 04661, 00936.

FAMILY MICROHYLIDAE

Brevicipitinac

Breviceps poweri (1). - FML 03165.

Microhylinae

Dermatonotus muelleri (5). - FML 05365, 01074.

Elachistocleis bicolor (7). - FML 00251, 00151, DIAM 029.

Phrynomerinae

Phrynomantis bifasciatus (1). FML 03029.

FAMILY DENDROBATIDAE

Colostethus subpunctatus (1). FML 02598.

Dendrobates auratus (1). -FML 01722.

Epipedobates pictus (1). FML 03516.

FAMILY HEMISOTIDAF

Hemisus guineensis guineensis (3). - FML 01244.

Source : MNHN, Paris

MANZANO, MORO & ABDALA

FAMILY ARTHROLEPTIDAE

Arthroleptinae

Arthroleptis stenodactylus stenodactylus (1). - FML 00066.

Arthroleptis tuberosus (2). - FML 00051.

Cardioglossa cyaneospila (1). - FML 06477.

Schoutedenella sp. (1). - RL Lote 24.

Schoutedenella sylvatica (1).- RL Lote 231.

FAMILY RANIDAE

Dicroglossinae

Conraua crassipes (2). - FML 03068.

Hoplobatrachus occipitalis (1). - No data.

Ptychadeninae

Ptychadena mascareniensis hylaea (3). - FML 01238.

Pyxicephalinae

Aubria subsigillata (1). — FML 03154.

Raninae

Rana albolabris (2). - Lote 231.

Rana angolensis (3). - FML 03188.

Rana lepus (1). -RL Lote 231.

FAMILY PETROPEDETIDAE

Phrynobatrachus calcaratus (2). - RL Lote 231.

Phrynobatrachus acutirostris (1) - DIAM 006.

FAMILY HYPEROLIIDAE

Hyperoliinac

Afrixalus orophilus (4). - FML 03229, 03230.

Chrysobatrachus cupreonitens (3). FML 02081.

Kassininae

Kassina senegalensis angeli (2). - FML 02076.

Leptopelinae

Leptopelis notatus (2). -FML 03205.

FAMILY RHINOPHRYNIDAE

Rhinophrynus dorsalis (1). -FML 01720.

131

Corresponding editor: W. Ronald HEYER.

Source : MNHN, Paris

Alvtes, 2003, 20 (3-4): 132-136.

The tadpole

of Phrynobatrachus mababiensis

FitzSimons, 1932

(Anura, Ranidae, Petropedetinae)

Rafael O. DE SA* & Alan CHANNING**

rsity of Richmond, Richmond, VA, USA

@richmond.edu>

* Depariment of Biology, Uni

** Department of Zoology, University of the Western Cape,

Private Bag X17, Bellville 7535, South Africa

<achanning@uwc.ac.za>

The tadpole of the poor known pudde frog Phrymobatrachus maba-

biensis FitzSimons, 1932 from Eastern Africa is described and illustrated.

INTRODUCTION

The puddle frog genus Phrynobatrachus Günther, 1862 comprises about 64 currently

recognized species (FRosT, 1985). Of these, only the tadpoles of P. natalensis (Smith, 1849)

(POWER, 1927; CHANNING, 2001), P guineensis Guibé & Lamotte, 1961 (RÔDEL, 1998) and P

alticola Guibé & Lamotte, 1961 (RôDEL & ERNST, 2002) have been described. Phrynobatra-

chus mababiensis FitzSimons, 1932 (Dwarf Puddle frog, WAGER, 1986; Mababe River frog,

FRANK & RAMUS, 1996) is a small frog that usually calls from low in thick vegetation on

flooded terrains close to the water. Very little has been published about the biology of this

species. PASSMORE & CARRUTHERS (1979) reported the advertisement call of P mababiensis,

and WAGER (1986) provided a few comments about the tadpole (see discussion below). Herein

we describe the tadpole of Phrynobatrachus mababiensis.

MATERIAL AND METHODS

Tadpoles of Phrynobatrachus mababiensis (34 individuals) were collected at Kibebe

Farm (08°29°0.5"S, 3 ), Iringa, Tanzania, by the authors on 8 February, 2000.

Specimens were fixed in 10 % formalin (commercial grade) at the time of collecting them. One

tadpole was at developmental stage 30 (GosNER, 1960), whereas the remaining tadpoles were

at developmental stage 26 or earlier. These specimens were deposited at the National Museum

of Natural History, Smithsonian Institution, Washington, USA (USNM 539462-82).

Source : MNHN, Paris

DE SÂ & CHANNING 133

Species identification was based on comparisons of our material with specimens collect-

ed by A. Channing on 12 January, 1986, at Katima Mulilo, Caprivi Region, Namibia

(17°38°00"S, 24°11"00”°E), in a shallow, muddy pool with grasses. The tadpoles collected in

Iringa, Tanzania, are identical to those from Katima Mulilo, Namibia. Of the three tadpoles

collected at Katima Mulilo, one was preserved whereas the other two were raised to juveniles

for identification purposes, and they correspond to P mababiensis. These specimens

were deposited at the Transvaal Museum, Pretoria, South Africa (TM 83618). The Katima

Mulilo specimens are the closest material available to the type locality; Katima Mulilo is

about 120 km north of the type locality of P mababiensis in the Mababe Depression, NW

Botswana.

Specimens were staged according to GOsNER (1960). The labial tooth row formula is

given according to ALTIG (1970). Terminology of measurements taken follows ALTIG &

McDiarMiD (1999). Measurements (in millimeters) were made using a Mitutoyo digital

calliper and are based on specimens (7 = 19) at Gosner stage 26 (USNM 539462-539480).

Means and standard deviations are given in the description (see tab. 1). The tadpole illustra-

tion is based on specimen USNM 539481 (Gosner stage 30), the most advanced stage

available in our sample.

RESULTS AND DISCUSSION

TADPOLE DESCRIPTION

Tadpoles of Phrynobatrachus mababiensis have a depressed and elliptical body (fig. 1). In

dorsal and lateral views the snout is rounded; in lateral view the snout slopes gradually

anteriorly toward the oral disc. The eyes arge and lateral. The external nares are located

half way between the eyes and the tip of the snout. The narial aperture is small, rounded, and

laterodorsally positioned. Tail fins are low; dorsal and ventral fins almost parallel the tail

musculature and are of approximately equal height. The dorsal fin originates at the tail-body

junction and the ventral fin originates at the posterior ventral terminus of the body. Tail fins

slope gradually posteriorly to a narrowly rounded tip. The tail musculature extends to the tip

of the tail. The spiracle is sinistral with a midlateral opening directed dorsally. The vent tube

and aperture is dextrally placed relative to the ventral fin. Measurements of the illustrated

tadpole (USNM 539481) at Gosner stage 30 are: TL17.2; BL 5.9; MTH 2.2; TMW 1.0; E0.8;

IOD 2.6. Measurements and summary statistics of additional 19 tadpoles at developmental

stage 26 are given in table 1.

The oral dise is anteroventrally positioned, emarginate, and has a uniserial row of conical

papillae with rounded tips (fig. 2). The row of marginal papillae has à large dorsal gap

occupying most of the upper labium. In addition, two pairs of long papillae project from the

lower labium, posterior to the marginal papillae. Th re about 2-3 longer than the marginal

papillae. À few submarginal papillae are found laterally on the upper labium. The labial tooth

row formula is 4(2-4)/4(1). Upper and lower jaw sheaths are wide, pigmented for about one

third of their width, and have widely serrated edges.

Source : MNHN, Paris

134

ALYTES 20 (3-4)

Fig. 1. - Tadpole of Phrynobatrachus mababiensis FitzSimons, 1932, stage 30, USNM 539481. Bar:

5.0 mm.

Table 1. - Measurements (in mm) of 19 tadpoles of Phrynobatrachus mababiensis FitzSimons,

1932. TL, total length; BL, body length; MTH, maximum tail height; TMW, tail muscle

width: E, eye diameter; IOD, interorbital distance.

USNM number TL BL MTH TMW E I0D

539462 17.1 6.0 24 0.9 0.8 27

539463 18.2 6.8 32 13 09 29

539464 16.2 57 34 1.0 10 25

539465 15.9 59 2.9 1.0 10 25

539466 16.0 6.1 32 11 10 29

539467 163 59 2.6 0.9 08 27

539468 164 6.1 2.8 1.0 1.0 2.6

539469 16.3 5.8 34 11 0.8 24

539470 154 6.2 2.8 11 1.0 27

539471 162 5.8 3.6 11 08 23

539472 16.8 5.9 3.0 11 10 24

539473 174 5.9 3.5 12 0.9 AT:

539474 16.4 5.8 3.6 11 0.9 27

539475 16.3 5.6 32 12 0.9 24

539476 16.6 5.7 29) 0.9 0.8 25

539477 16.0 5.8 32 1.0 0.8 27

539478 15.1 5.5 28 11 0.8 2.8

539479 16.2 5.6 2.9 12 0.9 25

539480 16.6 5.8 2.7 10 0.9 2.2

Mean 16.40 5.88 3.06 1.07 0.89 2.59

Standard deviation 0.68 0.30 0.34 0.10 0.06 0.19

Source : MNHN, Paris

DE SA & CHANNING 135

nm à

ROLL

Ones EU

Fig. 2. - Oral disk of Phrynobatrachus mababiensis FitzSimons, 1932, stage 30, USNM 539481. Bar:

1.0 mm.

In preservative, specimens are dark brown. The fins and tail musculature are speckled

with dark melanophores. These melanophores are more dense on the dorsal and ventral edges

of the tail musculature; melanophores also are dense along and immediately below the main

axis of the tail musculature. Visually, this accumulation of melanophores outlines a distinct,

pale, whitish band that extends along the epaxial muscles for about two thirds of the tail. A

second, but shorter and less distinct, band is present on the hypaxial muscles. The posterior

third of the tail musculature is homogeneously dark and the myotomes are poorly defined.

Melanophores are abundant on the dorsal and lateral surfaces of the body where they are

homogeneously distributed; ventrally they are present only on the anterior half of the body.

TADPOLE COMPARISONS

WAGER (1986) provided greatly oversimplified descriptions, including outline illustra-

tions, of tadpoles of Phrynobatrachus mababiensis and P. natalensis. These descriptions do

not agree with the previously reported tadpole of P natalensis (POWER, 1927) nor with the

present description of P mababiensis.

WAGER (1986) reported the oral disk of P mababiensis as having a labial tooth row

formula of 1/2 and possessing a double row of marginal papillae in the lower labium, with the

outer row consisting of elongated papillae. CHANNING (2001) included Wager‘

is section of tadpoles. A row of elongated papillae on the lower labium for early stages (total

length 6.0 mm) of P. natalensis Was reported by POWER (1927). It is possible that WAGER

(1986) may have misidentified the larvae; alternatively, a row of long papillae on the lower row

may be present in early stages of P. mababiensis, however we have not seen it.

Source : MNHN, Paris

136 ALYTES 20 (3-4)

In contrast to those of P mababiensis, the tadpoles of Phrynobatrachus guineensis and P

alticola (RÔDEL, 1998; RôDEL & ERNST, 2002) have morphological characteristics and modi-

fications typical of phytotelmic anuran larvae.

ACKNOWLEDGMENTS

Funds for RdS were provided by the Faculty Research Committee, University of Richmond, Grant

# F99513. We are thankful to the Tanzanian Commission for Science of Technology (COSTECH) for

scientific research permits (numbers 99-55-NA-99-40 and 99-287-NA-99-66, to AC and RdS respec-

tively). We thank Jenny Channing for her companionship and help during the fieldwork. This manuscript

benefited from the comments of two anonymous reviewers.

LITERATURE CITED

AuniG, R. 1970. - A key to the tadpoles of the continental United States and Canada. Herpetologica, 26:

180-207.

ALnG, R. & MCDiarMiD, R. W., 1999. — Body plan: development and morphology. Jr: R. W. McDiaR-

MID & R. ALTIG (ed.), Tadpoles: the biology of anuran larvae, Chicago, Univ. Chicago Press: 24-51.

CHANNING, À. 2001. - Amphibians of central and southern Africa. Yhaca, Cornell Univ. Pres 470.

FRANK, N. & RAMUS, E., 1996. — À complete guide to scientific and common names of reptiles and

amphibians of the world. Pottsville, NG Publ., Inc: 1-377.

Frost, D. R., (ed.), 1985. — Amphibian species of the world: a taxonomic and geographical reference.

Lawrence, Allen Press & The Association of Systematics Collections: i-v + 1-732.

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

identification. Herpetologica, 16: 183-190.

PASSMORE, N. I. & CARRUTHERS, V. C., 1979. — South African frogs. Johannesburg, Witwatersrand Univ.

Press: i-xviii + 1-270.

Power, J. H., 1927. - Notes on the habits and life histories of South African Anura with descriptions of

the tadpoles. Trans. r Soc. South Africa, 14: 237-247.

Rôez, M.-O., 1998. — A reproductive mode so far unknown in African ranids: Phrynobatrachus

guineensis Guibé & Lamotte, 1961 breeds in tree holes (Anura: Ranidae). Herpetozoa, 11 (1-2):

19-26.

Rôbez, M.-O. & ERNST, R., 2002. - A new reproductive mode for the genus Phrynobatrachus Günther,

1862: Phrynobatrachus alticola Guibé & Lamotte, 1961 has non-feeding, non-hatching tadpoles. /

Her 6 (1): 121-125.

WAGER, V. A., 1986. — Frogs of South Africa: their fascinating life stories. Craighall, Delta Books: 1-183.

Corresponding editor: W. Ronald HEYER.

Source : MNHN, Paris

Alytes, 2003, 20 (3-4): 137-149. 137

Notes on the treefrogs (Hyperoliidae)

of North-Western province, Zambia

Arne SCHIOTZ* & Paul VAN DAELE**

* Humlehaven 2, Atterup, 4571 Grevinge, Denmark

<schioetz@worldonline.dk>

#* Blommekens 3, 9900 Eeklo, Belgium

<pvdaele@zamnet.zm>

This paper is based on a collection of treefrogs (Hyperolidae) from

northern Mwinilunga district, North-Western Zambia. Hyperolius cinnamo-

Hyperolius kivuensis and the doubtful

Leptopelis parbocagii are additions for the district. Hyperolius nasutus

and Hyperolius benguellensis are established as separate species based on

distinct structures of their calls. Hyperolius bocagei is established as a

synonym of a member of the Hyperolius viridiflavus superspecies.

INTRODUCTION

Northern Mwinilunga district in the North-Western province of Zambia stretch sa

peninsula between Angola and République Démocratique du Congo, countries where biolo-

gical fieldwork is difficult. This part of Zambia has therefore received considerable attention

from herpetologists, and the fauna of Amphibia is reasonably well known (for treefrogs, see:

Scmiorz, 1975, 1999; POYNTON & BROADLEY, 1987; BROADLEY, 1991).

The present note is based on a collection made by the authors during a brief stay in the

district, 11 days in November 1999. The following localities were visited: (1) 2-5.11.1999,

Hillwood farm (11°15S, 24°18E); (2) 5-6.11.1999, Zambesi rapids (11°08'S, 24°08'E); (3)

6-9.11.1999, Jimbe river (10°57'S, 24°07°E); (4) 9-11.11.1999, Kachifiwiru (10°57'S, 24°06'E).

Although before the onset of the heavy rains, it was pos

probably all the species that have been reported from the area, with two additions to the

district and one to the country. This note thus covers all species of Hyperoliidae which have

been recorded from the Mwinilunga district. Only information regarded as new is given in this

paper. POYNTON & BROADLEY (1987) and ScHiorz (1975, 1999) gave a more general tre

ment. The preserved material is deposited in the Zoological Museum, Copenhagen (ZMUC).

sible to secure or observe

Source : MNHN, Paris

138 ALYTES 20 (3-4)

TAXONOMY

Afrixalus wittei (Laurent, 1941)

Comments. — Abundant and conspicuous in savanna localities. AIl collected specimens have

an identical pattern.

Material. — Fishpond near Jimbe river: ZMUC R.077939-40, R.077999 (2 6,1 ©); Hillwood:

ZMUC R.076676-77, R.076696, R.077926-38 (16 4).

Hyperolius nasutus Günther, 1864

Comments. - See remarks under Hyperolius benguellensis. Sixteen of the specimens (ZMUC

R.77957-72) were collected when calling and could therefore be distinguished with absolute

certainty from A. benguellensis by the voice. The remaining material is from the same ponds,

where no H. benguellensis were heard.

Voice. — A brief scream, apparently similar to the voice of this species elsewhere in its vast

range (fig. 1).

Material. — “Paul's Fishpond”, Hillwood: ZMUC R.077957-83 (25 &,2 ©).

Hyperolius benguellensis (Bocage, 1893)

The possible occurrence of a species very similar to, but distinct from, Hypero-

lius nasutus in southern Africa has long been discussed. ScHi6TZ (1975) recognised with doubt

the species Hyperolius granulatus (Boulenger, 1901) and so did, with similar doubt, POYNTON

& BROADLEY (1987) although they used the older name /. benguellensis, whereas SCHIOTZ

(1999) reluctlantly lumped H. nasutus with H. benguellensis.

During our field studies, we observed two structurally different voices in the study area.

The two types of calls were not heard from the same breeding localities, but from localities

only a few hundred meters apart.

We believe that the two call types represent two different species and that our separation

of the material between the species Æ. benguellensis and H. nasutus is correct, also for the

specimens not heard calling when collected since we spent several nights collecting in the

localities and listened especially for aberrant voices. Many voices were heard, but none

belonging to H. nasutus at a H. benguellensis locality or vice versa.

POYNTON & BROADLEY’S (1987) meticulous discussion of the Æ. benguellensis-nasutus

group is based on the assumption that the paravertebral lines are the key characters for H.

benguellensis. However, of our material of this species, out of 41 specimens only 6 show

paravertebral lines after preservation. We are therefore not confident that POYNTON &

BROADLEY"’s (1987) distinction between the two species is congruent with ours.

If samples are separated according to the voices it is possible to differentiate the two

species on external morphology. In mixed, preserved samples we believe that not all specimens

Source : MNHN, Paris

ScHioTz & VAN DAELE 139

1.0

s nasutus, Hillwood. The horizontal lines are 1KHz apart, the marks on the

- apart.

Fig. 1. Voice of Hyperoli

horizontal axis are 0.1 s

can be identified with certainty. The extent of webbing seems to be the best character,

although the webbing of Æ. nasutus varies so much over its range that this character must be

treated with caution and may be valid only in this part of Africa. The dorsal skin may be more

coarsely granulated in H. benguellensis, but the difference is not great (see tab. 1).

Table 1. - Comparison between Hyperolius nasutus Günther, 1864 and Hyperolius benguellensis

(Bocage, 1893).

Hyperolius nasutus Hyperolius benguellensis

Characters in life

Translucent green Often darker green

No paravertebral lines Sometimes paravertebral lines

Sometimes fine middorsal line No middorsal line

Voice a scream Voice a brief rattle

Characters after preservation

Ground colour lighter Ground colour darker

Nares less protruding Nares more protruding

More webbing Less webbing

Dorsum finely granulated Dorsum more coarsely granulated

Source : MNHN, Paris

140 ALYTES 20 (3-4)

WILSON (in press) discovered a degeneration in the tympanic apparatus in A. benguellen-

sis, a character not found in 1. nasutus. She kindly examined our material of both species and

found the difference in tympanum consistent with our separation.

In spite of the degeneration of the tympanic apparatus, the presence of a voice seems to

imply that A. benguellensis is not deaf. HETHERINGTON & LINDQUIST (1999) pointed at

alternative hearing mechanisms.

H. benguellensis is similar in colour pattern to Hyperolius viridis Schiotz, 1975 from

south-western Tanzania. H. viridis is, however, a larger frog (male snout-vent length 22-26 mm

vs. 17-22 mm in H. benguellensis) and especially a much more massive, broader frog, similar in

body proportions to the smaller Hyperolius pusillus (Cope, 1862).

Colour in life. — Translucent green, sometimes darker, more “dense” green. Many specimens

have a pair of light dorsolateral lines, and an additional, more diffuse pair of paravertebral

lines (fig. 3a). Other specimens lack the paravertebral lines, and some also lack the dorsolat-

eral lines and have diffuse dark spots on dorsum (fig. 3b). The latter two morphs seem

inseparable in pattern from Æ. nasutus, although they are sometimes somewhat darker.

Breeding.— The eggs have a white and a dark greenish pole.

Voice. — A brief rattle, acoustically quite distinct from the voice of H. nasutus. The sonogram

(ig. 2) shows a brief series of rather indistinct figures at 4000-4200 Hz. The voice illustrated

in ScHiorz (1975: fig. 96) from Kabwe (Zambia) as that of A. nasutus is in fact that of A.

benguellensis.

Fig. 2. Voice of Hyperolius benguellensis, Hillwood.

Source : MNHN, Paris

ScHiorz & VAN DAELE 141

Fig. 3. — (a-b) Hyperolius benguellensis, Hillwood: (a) specimen with paravertebral lines: (b) spotted

morph. (c-d) Æyperolius major, Kachifwiru: (c) phase 2: (d) phase 1. (e) Hyperolius parallelus

alborufus, ® , Hillwood. (1) Leptopelis cynnamomeus, 8 , Zambesi rapids. () Leptopelis parbocagit

; Zambesi rapids

Source : MNHN, Paris

142 ALYTES 20 (3-4)

Only few calling populations of H. nasutus and H. benguellensis were heard, so no

conclusions in relation to habitat preference could be drawn. The two species were not heard

at the same localities. While H. nasutus was heard in old, partly overgrown fishponds with

much low vegetation. H. benguellensis was taken in more open water holes on the grass-

covered plains at Hillwood. These last localities would seem to be similar to localities with /1.

nasutus from other places in Africa.

Of the present material, 10 males were taken when calling (ZMUC R.076709-18).

Material. — Hillwood: ZMUC R.076709-49 (35 4,6 ®).

Hyperolius quinquevittatus Bocage, 1866

Comments. — No breeding activity was observed. The specimens were taken by chance far

from water. Some of the specimens have a pattern which differs somewhat from that hitherto

recorded for the species: in life dorsum brown with three conspicuous golden-green bands (i.e.

a middorsal and two lateral bands), delimited with dark brown lines. Although it is in

principle the same pattern as e.g. that shown in ScHiorz (1999: fig. 175), the general

impression is that of a dark frog with 3 light stripes.

Material. — Kachifwiru: ZMUC R.771016-19 (4 6); Hillwood: AMUC R.076707,

R.771104-05 (2 ©, 1 juvenile); Zambesi rapids: ZMUC R.076750 (1 juvenile).

Hyperolius kivuensis Ahl, 1931

Comments. — Only a single male collected.

Material. — Kachifwiru: ZMUC R.0771015 (1 4).

Hyperolius parallelus alborufus Laurent 1964

Comments. — À member of the Hyperolius viridiflavus superspecies. The species structure

within this group is unsettled and disputed. The present form is conventionally included in the

species Hyperolius marmoratus Rapp 1842, whereas ScHioTz (1971, 1975) has argued for its

inclusion in the species Hyperolius parallelus Günther, 1859.

The type series from Cazombo (Angola) has black dots round the anus and on the tarsi.

Such spots are absent in the present sample. The form alborufus is very similar to other forms

in the very variable complex from this part of Africa, and a detailed study of the variation

would seem rewarding.

The present collection shows great uniformity in pattern as all specimens of the female

phase have the pattern shown in figure 3e. It is remarkable that a sample collected by Ronalda

Keith, also from the Mwinilunga district, shows much variation in pattern (see SCHiOTZ, 1971:

fig. 13, 1975: fig. 181, 1999: fig. 474a-e). This may be due to selective collection by Keith. Our

sample was collected without any bias as to pattern. Also the sample reported in BROADLEY

(1991) shows little variation (Broadley, in lit.).

Source : MNHN, Paris

ScHoTz & VAN DAELE 143

Breeding activities had hardly started during our visit, and only a few males were calling

from the ponds. There was, however, quite a number of males calling widely scattered in the

rather dense Miombo woodland in the area, apparently while migrating towards water. The

voice emitted here was the coarse “initial sound” and only a few times was the melodic

breeding call heard. Males and females collected there and kept in plastic bags did not

produce eggs during the night, something that would always happen if collected on the

breeding site.

Colour in life. In life, all phase F specimens had a bright red vermiculation on a light, greyish

background. Females had a bright red ventrum and conspicuously blue subdermal lateral

band (for explanation of this term, see ScHiorz, 1999: 199). The males of phase F were

similar, but with a less conspicuous subdermal lateral band. Males of phase J were brown with

a darker hour-glass pattern.

Material. — Fishpond near Jimbe river: ZMUC R.771000-03, R.771045-47 (6 4, 1 ®);

Hillwood: ZMUC R.076683-95 (10 4,3 9).

Hyperolius major Laurent, 1957

Comments. Described as a subspecies of Hyperolius platyceps (Boulenger 1900) but regarded

as a full species by ScHioTz (1975), mainly because the status of the name A. platyceps was

unclear. H. platyceps has later been defined (AMIET, 1978; Scmiorz, 1999). In morphology and

pattern, the present taxon is very similar to A. platyceps from north-western Central Africa.

Itis here regarded as a full species because the voice differs from that of H. platyceps.

Colour in life. — Two phases. Phase 1: dorsum light brown with a darker hour-glass pattern;

very dark loreal area, continuing behind the eye; the light pattern forming a triangle on the

snout (fig. 3d). Phase 2 (presumed female phase): dorsum uniform brown with light canthal

and dorsolateral stripe, also extending over upper eyelid (fig. 3c). Both phases: arms and legs

dark brown, in some specimens with conspicuous small light spots. Many specimens have

similar light spots in the dark pattern. Ventrum and throat orange. Ventral side of limbs

darker orange. The single female collected is of phase 2, brown with a lighter brown lateral

stripe.

Breeding. — Found exclusively in the dense gallery forests, thus being a “farmbush form” or

“bushland form” according to the terminology of Schiotz, collected together with Lepropelis

cynnamomeus (Bocage, 1893).

The eggs have a white and a black pole.

Voice. — An initial coarse creak, followed by a series of hard, unmelodic clacks in rapid

succession, almost 4frixalus-like (fig. 4). Such a vocal structure, with elements in a measured

rhythm, is quite unusual in the genus, found in Hyperolius guttulatus Günther, 1859, Hypero-

lius tuberculatus Mocquard, 1897, Hyperolius pseudargus Schiotz & Westergaard, 1999 and a

few others. The sonogram of #. major in SCHiOTZ (1999: fig. 400) is of the initial creak.

Material. - Kachifwiru: ZMUC R.771025-37 (12 6, 1 ®); Jimbe river: ZMUC R.771038-44

Gé).

Source : MNHN, Paris

144 ALYTES 20 (3-4)

Fig. 4. — Voice of Hyperolius major, Kachifwiru.

Hyperolius kachalolae Schiotz, 1975

Comments. — The present samples differ somewhat from the small sample from the type

locality. After preservation they have a more baggy, often pigmented gular sac, and in life the

sac is yellow to black, not turquoise as in the types. Nevertheless we are convinced that the

specimens listed here are conspecific with A. kachalolae. The voice is the same, and the thin

red canthal- and dorsolateral lines seem diagnostic.

One of our samples (from granite outcrops near Zambesi rapids) differed in life

somewhat from the other samples, mainly in the absence in most specimens of red canthal and

dorsolateral lines, and with a large proportion of the males being very dark and with a black

gular sac. We were in doubt as to whether two species were involved, but after preservation is

it impossible to distinguish that sample from the rest as especially the lighter specimens show

red canthal and dorsolateral lines, sometimes very faint; also the voice is identical to that of

the other samples of H. kachalolae.

POYNTON & BROADLEY (1987) expressed doubt as to whether Æ. kachalolae is different

from H yperolius bocagei Steindachner, 1867. The senior author has examined the holotype of

H. bocagei (Naturhistorisches Museum Wien, NMW 14846), a surprisingly well-preserved,

large female (35.1 mm) with a well-developed gular fold and extensive webbing. These

and the white, “chalky” dorsal surface:

characters, make it certain that it is a member of the

Hyperolius viridiflavus superspecies uggested in ScHioTrz (1999). Neither the imprecise

type locality (Angola) nor the pattern of the type (chalky white, as normal in the dry season)

permit reference to a specific subspecies in the complicated AH. viridiflavus-marmoratus-

parallelus group, so no nomenclatorial change based on priority seems necessary.

Source : MNHN, Paris

SCHIOGTZ & VAN DAELE 145

H. bocagei has been recorded from a number of localities in Angola, southern R. D.

Congo and northern Zambia, and H. kachalolae so far only from the type locality. Since H.

Kachalolae is abundant and very conspicuous in the Mwinilunga district — in fact the

dominant Hyperolius -, it seems unlikely that it has not been collected in neighbouring

Angola and southern R. D. Congo, nor being found in the material referred to as H. bocagei

in POYNTON & BROADLEY (1987). According to the description especially the last could well be

H. kachalolae. There are some samples in museums of members of the Æ. viridiflavus group

identified as /. bocagei by R. F. Laurent (see ScHiorz, 1999: 188).

In conclusion, Hyperolius bocagei is a synonym of one of the members of the H.

ridiflavus superspecies whereas we believe that several later records of H. bocagei in the

literature refer to H. kachalolae.

Colour in life. — Males: dorsum green to straw-coloured to black with red fingers and toes.

Some populations have a conspicuous red canthal line, continuing behind the eye, often as

spots. Many specimens in other populations lack the red lines in life. Gular sac in males from

yellowish to black, the latter especially in the black-backed specimens.

Females bright orange (our observations) or tomato red (Channing & Drewes, in lit.).

Some females have a feebly developed transversal gular fold.

Material. — Fishpond, Jimbe river: ZMUC R0.77948-52, R.771048-67 (15 4, 10 & ); Zambesi

rapids: ZMUC R.077953-57, R.077968-078003 (39 &, 1 ©); Kachifwiru: ZMUC

R.771020-24 (5 8); Hillwood: ZMUC R.076678-79, R.076697-706 (11 4,1 ?).

Hyperolius cinnamomeoventris Bocage, 1866

Comments. — First record from Zambia, but not unexpected in view of its range. No calling

heard.

Material. — Hillwood: ZMUC R.771012-14 (2 4,1 ©).

Kassina senegalensis Duméril & Bibron, 1841

Comments. — AI collected specimens belong to the strange spotted, rather large morph (see

photo in ScHiorz (1999: fig 508), except one with an almost unbroken dorsal line.

Material. — Zambesi rapids: ZMUC R.77986-91 (4 4,2 8 );Hillwood: ZMUC R.771009-11 (3

4); Jimbe river: ZMUC R.076665 (1 4).

Kassina kuvangensis (Monard, 1937)

Comments. — The characteristic voice, quite different from that of K. senegalensis, was heard

at an overgrown fishpond at Hillwood. No specimens could be collected. We did not hear the

two species of Kussina from the same locality.

Source : MNHN, Paris

146 ALYTES 20 (3-4)

Kassinula wittei Laurent, 1940

Comments. — This tiny frog was quite common on the soggy, grass-covered rocks near the

Zambezi rapids, but it was difficult to track down.

Material. — Granite flats, Zambesi rapids: ZMUC R.77984-85 (2 d).

Leptopelis eynnamomeus (Bocage, 1893)

Comments. — This species was suggested to be one of the “savanna screamers” in SCHIOTZ

(1999), a somewhat flippant term for a group consisting of vicariating savanna forms with a

similar morphology and a very characteristic voice, but rather different patterns. We believe

that some members of this group (Leptopelis concolor Ahl, 1929; Leptopelis argenteus (Pfeffer,

1893); probably Leptopelis broadleyi Poynton, 1985) are closely related, probably at the

subspecies level, but cannot maintain L. cynnamomeus as a member of this group. It is not a

Ssavanna form, but a bushland form. It has feebly developed pectoral glands (absent in the

other forms), and the voice is audibly different from them.

Colour in life. - Golden brown with diffuse, darker transverse bands and a dark interorbital

streak. Loreal area dark (fig. 39).

Voice. — Calling from leaves, often several meters up, in dense riverine (gallery) forests. Voice

a scream followed by 2-3 quiet clacks (“yün-clack-clack”), sometimes clacks alone. See

sonogram in SCHIOTZ (1999: fig. 681).

Note. — In the distribution map of “the savanna screamers” in ScHiorz (1999: fig. 672), the

symbols for L. argenteus and L. broadleyi have unfortunately been switched.

Material. - Kachifwiru: ZMUC R.77992-97 (5 4,1 ® ); Zambesi rapids: ZMUC R.771004-08

( 4); Jimbe River: ZMUC R.076666-75 (8 S, 2 juveniles).

Leptopelis parbocagii Poynton & Broadley, 1987

Comments. POYNTON & BROADLEY (1987) established the species L. parbocagii with hesita-

tion, since the difference from L. bocagii (Gunther 1864) seems very small. Their key character

is the broader head of L. parbocagii, expressed as the ratio interorbital distance vs. nostril-

tympanum being greater than 36 %. We are indebted to J. C. Poynton for havingexamined our

material and confirmed that it falls well within the definition of parbocagii (? 37.5%, &

39.1-47.6 %), but we agree with Poynton & Broadley that L. parbocagii may not be a species

distinct from L. bocagit.

L. parbocagiü is so far unrecorded from this area. L. bocagii is recorded from Isombo

Stream (POYNTON & BROADLEY, 1987; BROADLEY, 1991) and Hillwood, (BROADLEY, 1991).

We have examined the material reported in BROAD 1991 (8 specimens from Hillwood, 1

from Isombo Stream). They seem to fall within the definition of L. parbocagii, with an

interorbital distance vs. nostril-tympanum of 36,8-50,9 %.

Source : MNHN, Paris

ScHioTz & VAN DAELE 147

1.0

Fig. 5. - Mating call of Leptopelis parbocagii, Zambesi rapids.

All our specimens were taken in the savanna. The males from Jimbe River were calling in

a cultivated field close to the gallery forest. Some males were calling from the ground, others

from a low height in bushes.

Colour in life. - Dorsum grey to brown, often with a darker, almost black blotch on the

dorsum (fig. 3g).

Voice. — A deep, atonal “waab”, possibly indistinguishable from the voice of L. bocagii (fig. 5).

As in several other Leptopelis, another call (territorial?) is sometimes heard, namely a long

succession of clacks, almost inaudible at first but growing in intensity until the call ends with

the usual clack. One such recorded call (fig. 6) has a total duration of 8 seconds.

Material. — Near Jimbe River: ZMUC R.77942-44 (5 S ); Hillwood: ZMUC R.77945 (1 ®);

Kachifwiru: ZMUC R.77946-47 (2 4).

ACKNOWLEDGEMENTS

The senior author is most indebted to the Carlsberg Foundation for its grant, making the tour

possible. We are indebted to Robstein L. Chidavaenzi and D. G. Broadley for giving us the opportunity

to examine their material of Leptopelis parbocagit and to F. Tiedemann for enabling us to examine the

type of Ayperolius bocagei. We are furthermore indebted to our colleagues D. G. Broadley, A. Channing,

R. C. Drewes, R. Stjernstedt, J.C. Poynton and Lindsay G. Wilson for comments and inspiring

discussions.

Source : MNHN, Paris

L ur h

M AI PT A A LS UT UT)

ŒRLITPPENSE | DLRLLELL TT =

PT T T EE AE DR T T

1.0 2.0

(t-£) 07 SALATV

Fig. 6 - Presumed territorial call of L. parbocagü, Kachifwiru. Only the last 2.3 seconds of the call with a total duration of 8 seconds in shown. The

call is the succession of figures with growing intensity seen between 1 and 2 kHz.

Source : MNHN, Paris

ScHioTz & VAN DAELE 149

LITERATURE CITED

Aer, J.-L. 1978. À propos d'Hyperolius platyceps, H. kuligae et H. adametzi. Ann. Fac. Sci. Yaounde,

25: 221-256.

BROADLEY, D. G., 1991. - The herpetofauna of Northern Mwinilunga District, North-Western Zambia.

Arnoldia Zimbabwe, 9 (37): 519-538.

HETHERINGTON, T. E. & LINDQUIST, E. D. 1999. - Lung-base hearing in an “earless” anuran amphibian.

JL comp. Physiol., (A), 184: 395-401.

Poywron, J. C. & BROADLEY, D. G., 1987. - Amphibia Zambesiaca 3. Rhacophoridae and Hyperoliidae.

Ann. Natal Mus., 28 (1): 161-229.

Scmorz, À., 1971. The superspecies Hyperolius viridiflavus. Vidensk. meddr. Dansk. naturh. Foren., 134:

21-76.

1975. The treefrogs of Eastern Africa. Sicenstrupiar 1-23

1999. — Treefrogs of Africa. Chimaira: 1-350.

WiLsoN, L., in press. - Discovery of a novel character involving the tympanic apparatus in five species of

Hyperolius.

Corresponding editor: Alain DUBoIs.

Source : MNHN, Paris

Alytes, 2003, 20 (3-4): 150-160.

Diving behaviour of the Andean

frog Hyla labialis

Horst LüDDECKE* & Manuel Hernando BERNAL**

+ Universidad de los Andes, Departamento de Ciencias Biolôgicas,

A.A. 4976, Bogotä, Colombia

<holuddec@uniandes.edu.co>

++ Universidad del Tolima, Departamento de Biologia,

A.A. 546, Ibagué, Colombia

<mhbernal@utolima.ut.edu.co>

We describe the behaviour of Hyla labialis during descent, stationary

phase and ascent of voluntary dives in a laboratory tank and in natural

ponds. We compare dive times and activity of submerged frogs assigned to

six categories established according to developmental stage, gender and

reproductive state. In general, the descent took 1.6 % of the total dive time

and contained 25 % of the total number of moves per dive, the stationary

phase lasted 60.1 % of the total dive time and contained 12.5 % of the

moves, and the ascent lasted 38.4 % of the total dive time, with 62.5 % of

the moves. The longer the dive, the more moves did a frog make. Among

adults, neïither gender nor reproductive state was associated with

rences in dive time or activity. Juvenile frogs had significantly shorter dive

times than adults. The average dive time was significantly longer in a natural

pond than in a laboratory tank. Individuals with longer dive times in the

laboratory also had longer dive times in the field. Short dive times of

juvenile frogs may be associated with their transition from an aquatic to a

terrestrial habitat. Reproductive activity was not associated with a pro-

longed dive time and therefore no particular breath holding capacities seem

to be needed for egg laving. À submerged clasping male was not in any

conflict situation between sustaining reproductive behaviour and physiolo-

gical limitations to hold his breath.

INTRODUCTION

The diving behaviour of frogs has received little attention. Large inter- and intraspecific

variations in dive time may indicate that the decision to immerse and to surface depends on an

individual’ s changing physiological and behavioural state, as well as on life-history aspects

and particular external situations (HUTCHISON et al., 1976; PANDIAN & MARIAN, 1985; WEST

& VAN VLier, 1992). Individuals of many frog species spawn in water at a depth that requires

complete submergence of the amplectant pair (ErBL-ElBESFELDT, 1956; DUELLMAN & TRUE

1985) and therefore interruption of breathing (ZUG, 1993). A clasping male, carried under

water by the female and perhaps depending on her decision to surface, may enter a conflict

between his first in priority activity, mating, and his second in priority activity, breathing,

Source : MNHN, Paris

LÜDDECKE & BERNAL 151

similar to that of the male newt (HALLIDAY & SWEATMAN, 1976). Escaping from predators to

an underwater shelter (SCHNEIDER, 1967), as well as diving behaviour involved in foraging

activity (GANS, 1969) may also turn into a conflict situation. Thus, dives may be variable in

duration because the animal’s change in priority depends on several factors.

Reproductively active Hyla labialis stay in or near the ponds used for spawning, and dive

when disturbed. Non-reproductive adults as well as juveniles live an almost entirely terrestrial

life away from water bodies (LÜDDECKE, 1995) and cannot avoid predators by diving.

Therefore, except when being reproductively active and spawning, individuals should show

little tendency to dive. To determine whether the tendency to dive, diving behaviour and dive

time vary according to changes in habitat use and reproductive activity, we performed diving

trials with recently metamorphosed juveniles and adult H. labialis encountered during and

between breeding seasons.

MATERIALS AND METHODS

STUDY ANIMALS

AI experimental frogs came from 3500 m altitude in the Parque Nacional Natural

Chingaza (4°42°N, 73°48°W). Details concerning the area and the collection technique were

given elsewhere (LÜDDECKE, 1997). Between August 1996 and May 1997, frogs spent a week or

less the laboratory. Frogs to be tested in the laboratory were selected in the field according to

criteria related to their developmental stage and reproductive state. Snout-vent length was

measured with a calliper to the nearest millimetre and body mass was determined with a

laboratory balance to the nearest 0.1 g. We established six frog categories: (J) recently

metamorphosed juveniles (7 = 19); (AM) reproductively active males encountered at the

breeding pond and recognised by large and dark thumb pads, dark gular sac, heavy skin

secretion and strong odor (7 = 15); (IM) reproductively inactive males, encountered far from

the breeding pond, with small and pale thumb pads and pail throat (n = 15); (GF) gravid

females, with a body condition index of 80 or more (LÜDDE 1995), oviducts or eggs often

visible through the ventral skin (7 = 15); (SF) spent females, with a body condition index of 70

or less (n = 15); and (P) amplectant pairs encountered in the breeding pond (x = 5). AM and

GF were usually found resting at the pond edge or in the water, whereas J, IM and SF were

found away from the pond.

EXPERIMENTAL PROCEDURE

Dive time measurements and behavioural observations were conducted in a glass aquar-

ium 40 x 20 x 80 em deep which had an internal jacket of plastic mesh 1 X 1 em. It was filled

with unaerated water to 4 em below the rim. Three controlled water temperatures were used:

8, 18 and 28°C, which are within the range of water temperatures in breeding ponds of /.

labialis (LübpeckE, 1995). The experimental temperature to be used was randomly deter-

mined for each of the three days (capture day and the following two days) when dive tests were

Source : MNHN, Paris

152 ALYTES 20 (3-4)

made. For a sequence of three days, each water temperature was used once. The assignment of

frogs to temperatures was random. Preliminary tests had revealed no significant differences in

dive time between frogs used on different days after capture.

In each dive test, a single individual was placed in a plastic tube at the experimental

temperature during 30 min prior to a diving event, thus its body temperature equalled

the experimental temperature when it was pushed from the tube into the tank at an angle of

about 20 degrees to the water surface. This instant marked the beginning of a diving event. A

diving event ended when the frog had returned to the water surface for breathing. Each

individual was submitted at approximately 30-min intervals to three dive tests at the same

temperature, carried out on the same evening between 18.00 and 21.00 h local time under

artificial light.

Members of amplectant pairs were separated and male and female were first tested

independently in the way described above, but only once. They were then allowed to clasp

again, and the amplectant pair was tested once. However, in order to avoid accidental

separation, the pair was not introduced into the tube and pushed into the water, but placed

gently at the water surface of the tank from where it initiated the dive spontaneously.

Before being released at the capture site in the field, each unmated frog was submitted to

one dive test in a natural pond. We placed it on a moss cushion at the pond edge, from where

it would either jump into the water and dive immediately, or after being touched lightly on its

back. Dive time and water temperature at the pond bottom were recorded.

BEHAVIOURAL OBSERVATIONS

Diving frogs were observed continuously from the start to the end of each diving

event and their behaviour recorded directly on a record sheet. According to preliminary

observations, we distinguished three diving phases: descent, stationary phase and ascent.

To describe diving behaviour we took into account the mode of locomotion, body postures,

changes in head, body and leg position, and diving depth. As a measure of a submerged

frog's activity we counted its moves. À move could be a small change in body posture, in the

position of a leg, or a swimming stroke. In each phase we counted all occurrences. Total time

of each diving event, and the duration of each diving phase, were determined with a stop

watch.

STATISTICAL ANALYSIS

Dive time values of zero (frog did not dive) were excluded from analysis. AI dive time

values were log(x + 1) transformed in order to s: criteria for applying parametric tests

(ZaR, 1996). ANOVAS were used for comparisons between frog categories. Simple and

multiple regressions were used to test for associations between temperature, dive time and

moves. The comparison of field and laboratory dive times was made with a paired r test. Data

were processed using SYSTAT 5.2.1 for the Macintosh (SYSTAT Statis nston, IL,

Systat Ine.). À significance level of P < 0.05 was applied in all tests.

Source : MNHN, Paris

LÜDDECKE & BERNAL 153

RESULTS

DIVING BEHAVIOUR

Most frogs dove in all three tests, except for one female and three juveniles. Another two

juveniles refused to dive in two of the three tests, and four females, four males and two

juveniles refused to dive in one test. Dive time was highly variable, the average for all events of

all frog categories except amplectant pairs was 5.04 + 6.06 min (range: 0.05-46.15 min, n =

211), during which a frog made on average 46.6 + 25.4 moves (range: 1-160 moves, 7 = 199)

at an average rate of 14.6 + 10.9 (range: 1.5-82.7) moves per minute. Frogs dove to the tank

bottom in 161 of 211 tests; in 50 tests they came to rest on the tank wall at various depths. At

28°C, we found no dive time difference between 29 tests where frogs came to rest on the tank

wall, and 40 tests where they dove to the bottom (ANOVA; I! event: P = 0.19, n = 24; 24

event: P = 0.69, n = 24; 3% event, P = 0.08, n = 21). The diving behaviour of amplectant pairs

did not differ from that of unmated frogs. The clasping male seemed to play a passive role

during the entire dive. None of the pairs spawned in the dive tank.

Most individuals (96 %) descended with powerful leg strokes while the arms were held

backwards against the flanks. Few frogs descended by letting themselves sink. Some frogs

initially sank and then swam toward the bottom. After making contact with the plastic mesh,

the frog generally moved a short distance on the substrate before becoming stationary, the

instant we took to be the end of the descent. On average, the descent took 0.08 + 0.11 min

(range: 0.03-0.96 min, n = 89) or 1.6 % of total average dive time, and a frog made on average

25 % of its moves (mean = 10.9 + 5.1, range: 1-28, n = 174).

During the stationary phase, individuals remained at the same depth, most took a sitting

or crouching posture and made occasional postural and limb movements and exceptionally

short-distance moves. On average, this phase occupied 2.8 + 2.9 min (range: 0.2-15.0 min, n

= 89) or 60.1 % of total average dive time, and contained 12.5 % of the moves (mean = 5.4 +

8.9, range: 1-67, n = 174).

The ascent began when the frog, still in a crouched posture, slightly lifted its head. The

most common locomotory pattern was walking up the plastic mesh like on a ladder. Most

individuals (93 %) ascended slowly by a sequence of alternating moves and halts. Some

individuals actively swam to the surface and only a few let themselves float upward. There

were also some cases of combining different locomotory patterns. The ascent ended when the

frog’s nostrils broke the water surface. On average, this phase took 1.8 + 2.0 min (range:

0.08-10.5, n = 71) or 38.4 % of total average dive time, and a frog made on average 62.5 % of

its moves (mean = 27.34 15.3, range: 1-144, n = 174).

COMPARISON OF SUCCESSIVE DIVING EVENTS OF AN INDIVIDUAL

All frog categories pooled, the first diving event (6.3 + 7.9 min) was significantly longer

than the second and the third (4.7 + 4.9 min, 4.9 + 5.4 min, respectively; repeated measures

ANOVA; df = 2, P = 0.036). The difference became more pronounced after excluding juvenile

frogs from the analysis (repeated measures ANOVA; df = 2, P = 0.019). The main reason was

Source : MNHN, Paris

154 ALYTES 20 (3-4)

that on average the first diving event of juvenile frogs was the shortest, whereas in the other

frog categories the first diving event was the longest (GF, SF, AM and IM pooled; 1‘'event: 7.2

+ 8.5 min; 2" event: 5.2 + 5.2 min; 3“! event: 5.4 + 5.7 min; for each category separately see

tab. 1). We found no differences between the three diving events of all frog categories pooled,

neither regarding the number of moves per dive (1** event: 50.9 + 29.4, n = 69; 2°% event: 43.7

+ 22.2, n = 67; 3" event: 44.9 + 23.5, n = 63; ANOVA\; df = 2, P = 0.207), nor the move rate

(event: 15.7 + 12.6, n = 69; 2" event: 14.9 + 11.7, n = 67: 3% event: 13.1 + 7.9, n = 63:

ANOVA:; df= 2, P = 0.395).

COMPARISON AMONG CATEGORIES AND GENDER

There were significant dive time differences among frog categories (ANOVAs: all events

pooled: df = 4, P = 0.0001; 1‘ event: df = 4, P = 0.029, n = 58; 2"! event: df = 4, P = 0.167,n

= 57; 3“ event: df=4, P =0.017, n = 54), due to juveniles having shorter dive times than adults.

After excluding juveniles from analysis, no differences remained between categories

(ANOVA; df = 3, P = 0.246). Thus, although gravid females and reproductively active males

tended to dive less time than spent females and reproductively inactive males in almost all

diving events at all temperatures, adult males and females had statistically indistinguishable

dive times (ANOVAs:; all events: df= 3, P = 0.724; 1‘ event: df= 3, P = 0.69, n = 58; 2" event:

df=3,P=0.29, n= 57; 3% event: df=3, P=0.31, n = 54). We found no differences among frog

categories regarding the number of moves during the dive (ANOVA; df = 4, 0.841), but

significant differences in move rate (ANOVA: df = 4, P = 0.015), because juveniles had higher

rates due to their shorter dive times (tab. 2).

Dive time of amplectant pairs (mean = 6.4 + 5.2 min, range: 2.4-15.3 min), only

measured at the intermediate temperature, did not differ significantly from that in any event of

all unmated adult frog categories at that temperature (ANOVAs; df = 4 and n = 43 in all cases:

I‘event: P = 0.903; 2" event: P = 0.366; 3“ event: P = 0.858). When clasped by a male, the

female tended to dive insignificantly longer than when diving by herself (mean = 5.4 + 3.2

min, range: 1.0-9.5 min; paired f test, 1 = 0.33, P = 0.759). The male of mated pairs, diving by

himself, had an insignificantly shorter dive time (mean = 1.7 + 0.5 min, range: 1.2-2.4 min)

than the female diving by herself (paired, two-tailed r test, 1 = 2.29, P = 0.084), and a

marginally significant shorter dive time than when being taken under water by the female he

was clasping (paired, two-tailed r test, 1 = 2.53, P = 0.064).

RELATIONSHIPS BETWEEN DIVE TIME, MOVES, AND TEMPERATURE

The longer the dive, the more moves did a frog make (fig. la). Multiple regression

analysis shows that the number of moves was related to dive time (Beta coefficient = 0.41, P

= 0.0001), but not to temperature (Beta coefficient = — 0.04, P = 0.388; fig. 2a). Since the

number of moves was similar at different temperatures, but average total dive time shortened

at higher temperatures (linear correlation coefficient, r = 0.46, P = 0.0001; fig. 2b), short dives

had very high move rates (fig. 1b). Multiple regression analysis indicates that both, move rate

and temperature (fig. 2c), significantly affect dive time (Beta coefficients; move rate: -0.515, P

= 0.0001, n = 199; temperature: — 0.346, P = 0.001). The positive relation between moves and

Source : MNHN, Paris

LÜDDECKE & BERNAL 155

Table 1. — Dive time comparison of 75 Hyla labialis belonging to five categories, cach individual

tested three times (events 1, 2, 3) at one of three temperatures. (J) juvenile frogs; (IM)

reproductively inactive males; (AM) reproductively active males; (SF) spent females; (GF)

gravid females. Sample size (in brackets) refers to number of individuals.

Temperature | Event J IM AM SF GF

1 | 51468(5) [1244136(5)| 7443.6(5) [140418.2(5)| 9.8448(S)

8°C 2 | 58448(4) | 71461(5) | 46+1.9(5) |13.2411.1(5)| 6.542.6(4)

3 | 224+41(4) | 52224(5) | 6643.6(5) |184+12.7()] 48422(5)

1 [214155 | 78452(5) | 98411.9(4) | 4342.1(4) | 37418(5

18°C 2 | 214#13(5) | 6825.1(5) | 3.542.8(5) | 45424(4) | 3.741.9(5)

3 | 37+31(5) | 69+4.9(5) | 52424(5) | 3.142.9(4) | 42+17(5)

1 | 134064) | 21417(9) | 27#1.1(5) | 49441(5) | 14407(5)

28°C 2 | 1741.7(5) | 1841.7(5) | 1941.6(5) | 2.541.3(5) | 1.8+04(4)

3 | 154#16(5) | 25+06(3) | 24+0.8(4 | 194+0.7(4) | 144+0.5(5)

1 |24236(12 | 84293(13) | 6746.9(13) [8.9412.0(12)| 4844.5(13)

Alltemperatures | 2 | 2.6424(12) | 5.94 5.0(13) | 3.34 24 (13) | 7.74 84 (12) | 3.942.6(13)

3 [28430 (12) | 5243.6(13) | 4.94 3.1 (13) [7.84 10.4 (12)| 3.6+2.2(13)

Table 2. — Dive time and activity comparisons of 75 Hyla labialis belonging to five categories,

including all three events and all three temperatures. (J) juvenile frogs; (IM) reproductively

inactive males; (AM) reproductively active males; (SF) spent females; (GF) gravid females.

Sample size refers to the number of dives.

Frog category | " | Divetime(min) | Number of moves Moves per minute

J 31 3.7+3.6 45.7 + 28.3 19.1 + 14.9

IM 43 6.0+6.3 46.8 + 28.1 12.7+9.1

AM 43 4.8+4.5 50.4 + 28.6 15.0+ 11.7

SF 39 7.749.8 45.7 +25.9 10.8+ 7.8

GF 43 41433 44.2 + 15.4 16.2 + 10.1

dive time is mainly due to the ascent, which contained 62.5 % of the moves, and less to the

descent and the stationary phase, which contained only 25 % and 12.5 %, respectively, of the

total number of moves.

FIELD OBSERVATIONS

None of the frogs tested in the field tried to escape by moving away from the pond. Al

frogs dove to the bottom of the pond and usually entered the soft layer of mud where they

were invisible to observers. They reappeared at the beginning of the ascent, when they lifted

their head out of the mud, usually close to where they had entered. They remained for up to

several minutes in this position, before they continued the ascent by leaving the mud and

Source : MNHN, Paris

156 ALYTES 20 (3-4)

Number of moves

æ 70

& 60

E 50

ro]

a 40h

8 F

L 304

à ;

Z 204

10 À

0 10 20 30 40 50

Dive time (min)

Fig. 1. - Relationships between dive time in the laboratory, number of moves during a dive, and rate of

moves, in Hyla labialis. AN events included.

swimming to the surface, or by first walking some distance on the mud surface and then slowly

climbing up aquatic plants.

There were no significant field dive time differences among frog categories (ANOVA: d/

—4, P = 0.069), even though juvenile frogs had rather short, and reproductively inactive males

had rather long dive times. Pooled dive times for all categories at corresponding temperatures

(using the laboratory record for low or intermediate temperatures closest to the temperature

of each frog diving in the field) were significantly longer in the field (8.9 + 6.6 min) than in the

laboratory (6.5 + 5.8 min; two-tailed paired r test; df= 37,1 = 3.27, P = 0.002). Individual field

and laboratory dive times were significantly and positively correlated for all individuals (linear

licient, r = 0.74, P = 0.0001, n = 37), and also within each of three (SF, AM,

J) of five categories (fig. 3)

correlation coe

Source : MNHN, Paris

Fig

Dive time (min) Number of moves

Moves per minute

a dive, and rate of moves, in Æ/vla labialis. A events included.

LÜDDECKE & BERNAL

SRESSe

8 3833

a œ

So ©

5 10 15 20 25

Temperature (°C)

157

2. Relationships between water temperature, dive time in the laboratory, number of moves during

Source : MNHN, Paris

158 ALYTES 20 (3-4)

Dive time in field (min)

+nu

5 À AM: P = 0.0055

X SF; P = 0.0009

CXca

0 5 10 15 20 25 30 35

Dive time in laboratory (min)

Fig. 3. - Relationship between dive times in the laboratory at 8 or 18°C, and in a natural pond at 9-16°C,

of 37 Hyla labialis belonging to five frog categories. Diagonal traced line indicates dive times of equal

duration. Continuous lines are regression lines for three frog categories.

DISCUSSION

sun basking on a moss cushion at the pond edge plunged into the water when

disturbed, diving was assumed to be a protective behaviour against terrestrial predators. In

this context, the function of the descent is to take the frog down fast and efficiently by actively

swimming towards the bottom. The relatively long lasting and passive stationary phase also

indicates that a frog dives to escape, particularly since in a natural pond the diver usually

penetrates into the mud layer at the bottom of the pond, where itis visually concealed. Finally,

lifting the head and looking upward before beginning to ascend, as well as the frequent halts

of the frog on its way up, could be interpreted as vigilance by visually ing the

surroundings for predators across the totally transparent water of these natural ponds.

According to our procedure, we assumed that the frogs would perceive the push into the water

as a predator attack, and expected dive time and behaviour to be related to the risk. If risk

were related to body size, then juvenile frogs should dive longest. This expectation was not

fulfilled. Alternativ probably ignores a predator’s persistance in Waiting for it

to surface, its antipredator Strategy may as well be to surface at an unpredictable moment

since a fr

Source : MNHN, Paris

LÜDDECKE & BERNAL 159

(Bab, 1983). Further, despite the difference in procedure, amplectant pairs showed the same

behavioural diving patterns and similar dive times as did unmated frogs, which suggests that

the predator-simulating push did not have much influence on dive time.

The high variability of dive times suggests that changes in behavioural priorities are

involved (HALLIDAY & SWEATMAN, 1976). For instance, juvenile Æ. labialis are in transition to

a terrestrial lifestyle, thus the lower tendency to dive and shorter dive times of metamor-

phosed frogs may be associated with avoidance of a high predator risk in the aquatic habitat

(WizBur & CoLLiINs, 1973; WAssERSUG & SPERRY, 1977; SKELLY, 1994). Reproductively active

frogs may terminate a dive in order to continue mate searching, which is done at the water

surface. The exceptionally short dive times of males of pairs, diving by themselves after having

been separated from the female, may reveal restlessness motivated by searching for the missing

mate.

The rather paused return to the surface suggests that 7. labialis begin to ascend long

before depleting their oxygen stores. Although an urge to breathe is unlikely to be second in

priority, physiological aspects may play a role in diving behaviour. For instance, the negative

correlation between underwater activity and dive time of H. labialis, also found in other

amphibians (HALLIDAY & Worsnop, 1977), is associated with temperature, and together they

may determine when threshold levels of blood oxygen concentration are reached (HUTCHISON

et al., 1976; BOUTILIER, 1990). Since unmated males had dive times similar to those of pairs,

we assume that a male clasped to a submerged female is not in any conflict between holding on

to her or breathing. Apparently, spawning behaviour does not rely on particular breath

holding capacities, because the time spent submerged for spawning seems to be very short in

most anuran species (see WEYGOLDT & PoTsCH, 1992). The tendency to dive may also be

associated with temperature preferences or individual experiences. Although //. labialis is

thermophilie (VaLDIviEsO & Tamsrrr, 1974; LÜDDECKE, 1995), the unfamiliar experience to

descend into a warm depth may be the reason for a reduced tendency to dive to the bottom of

the tank at high water temperatures, since in nature it would always find cooler temperatures

as it dives deeper. Finally, since an individual frog tended to make either longer or shorter

dives, irrespective of place and time of day, performance differences among individuals may

be due to inherent physiological or behavioural properties.

ACKNOWLEDGEMENTS

We thank an anonymous reviewer for suggestions to improve the manuscript. COLCIENCIAS

(grant 1204-05-197-94) and Universidad de los Andes gave financial support, The Colombian Ministry of

the Environment granted permission {o experiment with frogs from the National Park Chingaza

LITERATURE CITED

BaiRD, T. A., 1983. - Influence of social and predatory stimuli on the air-breathing behavior of the

African clawed frog, Xenopus laevis. Copeia, 1983: 411-420.

BouriLirr, R. G.. 1990. — Respiratory gas tensions in the environment. /n: R. G. BOUTILIER (ed.)

Vértebrate gas exchange from environment to cell, Vol. 6, Advances in environmental and comparative

pringer-Vertag: 1-13.

physiology, Berlin,

Source : MNHN, Paris

160 ALYTES 20 (3-4)

DUELLMAN, W. E. & TRUEB, L., 1985. — Biology of amphibians. New York, McGraw-Hill, 1986": i-xix +

1-670.

ErBL-EiBEsFELDT, L., 1956. — Vergleichende Verhaltensstudien an Anuren. 2. Zur Paarungsbiologie der

Gattungen Bufo, Hyla, Rana und Pelobates. Zool. Anzeiger, Suppl, 19: 315-323.

Gans, C., 1969. - Comments on inertial feeding. Copeia, 1969: 855-857.

HALLIDAY, T. R. & SWEATMAN, H. P. À. 1976. - To breathe or not to breathe; the newt's problem. Anim.

Behav., 24: 551-561.

HaLLIDAY, T. R. & Worsnop, A., 1977. - Correlation between activity and breathing rate in the smooth

newt, riturus vulgaris (Amphibia, Urodela, Salamandridae). J. Herp., 11: 244-246.

HurcuisoN, V. H., HAINES, H. B. & ENGBRETSON, G. A. 1976. - Aquatic life at high altitudes: respiratory

adaptations in the Lake Titicaca frog, Telmatobius culeus. Respir. Physiol., 27: 115-129.

LüDDEckE, H., 1995. - Intra- and interpopulational comparison of temperatures selected by Hyla labialis

(Anura). /n: G. A. LLORENTE, A. MONTORI, A. X. SANTOS & M. A. CARRETERO (ed.), Scientia

herpetologica, Barcelona: 192-196.

== 1997. — Field reproductive potential of tropical high mountain Hyla labialis females: direct and

indirect evidence from mark-recapture data. Amphibia-Reptilia, 18: 357-368.

PANDIAN, T. J. & MaRIAN, P., 1985. - Physiological correlates of surfacing behaviour. Effect of aquarium

depth on surfacing, growth and metamorphosis in Rana tigrina. Physiol. & Behav., 35: 867-872.

SCHNEIDER, H., 1967. - Rufe und Rufverhalten des Laubfrosches, Hyla arborea arborea (L.). Z. vergl.

Physiol., 57: 174-189.

SkeLLY, D. K., 1994. — Activity level and the susceptibility of anuran larvae to predation. Anim. Behar.,

47: 465-468.

VALDIVIESO, D. & TAMSITT, JL R., 1974. - Thermal relations of the Neotropical frog Hyla labialis (Anura:

Hylidae). Royal Ontario Mus. Life Sci. oce. Papers, 26: 1-10.

WassERSUG, R. J. & SPERRY, D. G., 1977. - The relationship of locomotion to differential predation on

Pseudacris triseriata (Anura: Hylidae). Ecology, 58: 830-839.

Wesr, N. H. & van Vuier, B. N., 1992. — Sensory mechanisms regulating the cardiovascular and

respiratory systems. /n: M. E. FeDER & W. W. BURGGREN (ed.), Environmental physiology of the

amphibians, Chicago, University of Chicago Press: 151-182.

WEYGOLDT, P. & PotscH, S., 1992. - Mating and oviposition in the hylodine frog Crossodactylus

gaudichaudii (Anura: Leptodactylidae). Amphibia-Reptilia, 13: 35-46.

Wizeur, H. M. & COLLINS, JP, 1973. - Ecological aspects of amphibian metamorphosis. Science, 182:

1305-1314.

Zar, J.H., 1996. - Biostatistical Analysis. 3 edition. Upper Saddle River, New Jersey, Prentice Hall: i-x

+ 1-662, app. 1-205, ans. 1-11, 11-19, i1-21

Zu, G. R., 1993. - Herpetology. An introductory biology of amphibians and reptiles. \* edition. San

Diego, Academic Press: i-xv + 1-527.

Corresponding editors: Thierry Lobé & Alain PAGANO.

Source : MNHN, Paris

Alytes, 2003, 20 (3-4): 161-168. 161

On distribution of and hybridisation

between the newts

Triturus vulgaris and T. montandoni

in western Ukraine

Spartak N. LITrvINCHUK* **, Leo J. BORKIN* & Juri M. ROSANOV**

* Department of Herpetology, Zoological Institute, Russian Academy of Sciences,

199034 St. Petersburg, Russia

** Institute of Cytology, Russian Academy of Sciences,

194064 St. Petersburg, Russia

A case of natural hybridisation between the newts Triturus vulgaris and

T. montandoni was recorded in the Maloye Opolie area, Lvov province,

western Ukraine, where the latter species is represented by a population

geographically isolated from the main range in the Carpathian mountains. À

female hybrid, probably of the first generation, was identified by means of

DNA flow cytometry. No hybrids were found in the Ukrainian Carpathians.

Syntopic occurrence of T. montandoni and T. vulgaris in western Ukraine

is discussed.

INTRODUCTION

The Montandon’s newt, Triturus montandoni (Boulenger, 1880), is endemic to the

Carpathian mountains. The species occurs in the Ukrainian Carpathians from 150 up to 2000

m above sea level (SZCZERBAK & SZCZERBAN, 1980). The range of 7 montandoni is wholly

surrounded by the range of T7: vulgaris. However, both ranges in fact, parapatric.

Nevertheless, the two species can coexist in the same water body. Such cases of syntopic

occurrence have been reported in Romania (FUHN, 1963; Sova, 1973; FUHN et al., 1976: our

data), Slovakia (GuL1CKA, 1954), Czech Republic (REHAK, 1993), Poland (SZYMURA, 1974:

Juszczyk, 1987) and in Ukraine (HORBULEWICZ, 1927; SZCZERBAK & SZCZERBAN, 1980;

fig.D).

Triturus montandoni is genetic

closely related to T vulgaris, and both species have

similar sexual behaviours (BELYAEV, 1979, 1981; PECIO & RaAFINSKI, 1985; RAFINSKI &

ARNTZEN, 1987; ARNTZEN & SPARREBOOM, 1989). Therefore, it is not surprising that in the

laboratory the two species can hybridise and that such cro: can provide fertile adult

offspring (WOLTERSTORFF, 1925; GEYER, 1953 — cited from FUHN, 1963; MACGREGOR et al.,

1990; COGALNICEANU, 1994; our data).

Source : MNHN, Paris

162 ALYTES 20 (3-4)

UKRAINE

Chernovtsy

Fig. 1. Distribution of Triturus montandoni (open circles) in western Ukraine. The localities shared with

T. vulgaris (semiopen circles) are: (1) Rakovets, 295-350 m:; (2) Stary Sambor, 709 m; (3) Truskavets,

300-360 m; (4) Skole, 430 m; (5) Voinilov, 350 m:; (6) Delyatin, 500 m; (7) Chernovtsy, 150 D m; (8)

Dolishny Shepot; (9) Delovoe, 365 m; (10) the Karpaty sanatorium, 165 m; (11) Poroshkovo: (12)

Domanintsy; (13) Kriva. Sources: (1), (3) & (13), our data; (2) HOFMANN (1908); (3) POLUSHINA &

KUSHNIRUK (1962); (4-8) & (11). BAK & SZCZERBAN (1980); (9) O. A. Lugovoy & VF

Pokinchereda (pers. comm.); (10) N. L. Orlov (pers. comm.; ZMM 1404 & 2975; ZISP 5025, 5026 &

5561); (12) KUSHNIRUK (1963). Solid circles are sporadic localities of 7: vulgaris in the Carpathians.

The border of the main distribution of T vulgaris, beyond the mountains, is marked by a black line.

In the breeding period, male hybrids obtained from laboratory crosses have a combina-

tion of various features: from 7! montandoni — long tail filament (4 mm), well-developed

dorsolateral ridges, black feet, gray throat without spots and bright orange belly —, and from

T: vulgaris — dorsal crest (height 1.5 mm) and small black dots on the belly. Female hybrids

fer slightly from females of 7: montandoni, although hybrids have small black spots on the

In the field, several authors found newts with a pattern similar to that of the laboratory

hybrids and recognized them as natural hybrids. For instance, such newts were reported from

Source : MNHN, Paris

LITVINCHUK, BORKIN & ROSANOV 163

western Ukraine (fig. 1: localities 2, 3 and 12), Romania (FUHN, 1963; FUHN et al., 1976) and

Czech Republic (REHAK, 1993). However, such records are uncommon, and in Slovakia, for

instance, long term (16 years) observations on syntopic populations of T: montandoni and T.

vulgaris provided no hybrids (GULIÈKA, 1953).

An application of biochemical techniques could facilitate more reliable identification of

such presumptive hybrids, based on external characters only. We know only two cases where

interspecific hybridisation has been confirmed by allozyme analysis. In some syntopic locali-

ties in Poland, the incidence of hybrids, which were mainly recombinants, varied between four

and 60 % (PECIO & RAFINSKI, 1985; RAFINSKI, 1985). In a Czech locality where both species

coexisted, five specimens with intermediate characters were collected (KorLik et al., 1997).

Based on three loci, four animals were identified as hybrids: one individual, perhaps, was a

product of a backcross of the second generation and others were the offspring of more distant

crosses. No hybrids of the first generation were found.

Triturus montandoni is protected in many countries. The species is listed in the Red Data

Book of Ukraine although its local density can be quite high. Based on our data, T

montandoni is very common and obviously predominates over any other newt species in the

Ukrainian Carpathians. COGALNICEANU (1997) suggested that, although not sufficiently

documented, hybridisation with 7: vulgaris may contribute to the reduction of its range and

even pose a threat to its long term conservation.

The goal of our study was to investigate possible hybridisation between T montandoni

and T. vulgaris in western Ukraine.

MATERIALS AND METHODS

In 1989-1996 we searched for presumed hybrids between T° montandoni and T: vulgaris in

all districts of Zakarpatskaya Province, as well as in Turka, Sambor, Khyrov, Drogobych,

Nikolaev and Pustomyty districts of Lvov Province. In total we examined, mainly in the field,

the external characters of above one thousand individuals of T: montandoni and T. vulgaris

from western Ukraine including 567 individuals from the Maloye Opolie area (tab. 1). After

careful examination, the animals were usually released to the same water body where they had

been collected. Some animals, including 27 individuals from Maloye Opolie, were studied by

DNA flow cytometry.

The amount of DNA per nucleus (genome size) was determined in relative units as a ratio

of the fluorescence intensity of cells from an individual examined to that of reference cells.

The details of the technique have been published by VINOGRADOY et al. (1990). Peripheral

blood cells of Pleurodeles waltl were used as a standard.

Apart from our field study, we examined some museum collections. Museum abbrevia-

tions are: IZK, Institute of Zoology, Ukrainian Academy of Sciences, Kiev, Ukraine; ZISP,

Zoological Institute, Russian Academy of Sciences, St. Petersburg, Russia; ZMM, Zoological

Museum, Moscow State University, Moscow, Russ

Source : MNHN, Paris

164 ALYTES 20 (3-4)

Table 1. - Occurrence of four newt species along the Rakovets gradients from the forest edge to the

deep forest. Abbreviations: T. m., Triturus montandoni; T. v., T. vulgaris: T. a., T. alpestris:

T. c., T. cristatus.

600-700 m from the 2200-3000 m from

Waterbodies Forest edge

forest edge forest edge

Newt species Tm. Tv. Ta TelTm Tv Ta TelTm Tv Ta Tec

Date: 29.05.94 no data 123 3 19 3 no data

18.04.95 0 2 0 0 |186 0 10 10 no data

11.05.96 2 O0 OO |144 O0 23 2 |102 O0 35 O0

Table 2. - Genome size variation in Triturus vulgaris and T. montandoni.

Province n Mean Standard deviation Range

T. vulgaris

St. Petersburg 77 1.121 0.011 1.076-1.147

Zakarpatskaya 75 1.120 0.011 1.089-1.144

Lvov % 1.111 0.011 1.096-1.127 |

T. vulgaris x T. montandoni

Lvov 1 - _- 1.150

T. montandoni

Lvov 19 1.182 0.009 1.167-1.199

Zakarpatskaya 6 1176 0.011 1.158-1.189

RESULTS AND DISCUSSION

Having checked the zone of parapatry of T montandoni and T: vulgaris, we found that

some localities (fig. 1: 9 and 10) were inhabited by either of the two species only, although

previous authors listed both species there. No morphological hybrids were identified.

On June 9, 1990, a sole locality with syntopic T! montandoni and T vulgaris was recorded

in the surroundings of Truskavets, Lvov province, near the town's water reservoir (fig. 1:

locality 3). In a small round puddle (diameter 3 m, depth 15 cm) situated in an depression on

the ground road nearby the forest, six individuals of T! vulgaris (3 males and 3 females) and a

female of T: montandoni were found. Despite our attempts, no males of 7: montandoni were

observed there. Curiously, the only female of the latter species was transferred to St.

Source : MNHN, Paris

LITVINCHUK, BORKIN & ROSANOV 165

Petersburg and laid eggs successfully. Normal larvae hatched and metamorphosed. The

juveniles had unspotted bellies and were similar to those of 7: montandoni. Later these

animals died for various technical reasons. Unfortunately, we failed to identify the presump-

tive hybrid origin of these young individuals.

In 1994-1996, our special attention was focused on an isolated population of T: montan-

doni from the surroundings of the village Rakovets, Pustomyty district, Lvov province, which

is situated 30 km south of Lvov city (fig. 1: locality 1). This hilly area named Maloe Opolie is

recognized by geographers as an extension of the Bobrko-Stolskoye Kholmogorie which

belongs to the Podolskaya eminence. The area is partly covered by a small beech forest of the

so-called Carpathian type which is separated from the beech forests of the Carpathian

mountains by the valley of Dniester river over a distance above 50 air kilometers. This

isolation seemed to be associated with changes in the Carpathian forest limits in the Atlantic

epoch, i.e. 5-8 thousand years ago (MALINOVSKY, 1991).

This area is inhabited by all four newt species which are distributed in the Carpathian

mountains, i.e., T. vulgaris, T. cristatus, T. alpestris and T. montandoni. The two latter species

are obviously represented by populations geographically isolated from their main ranges in

the Carpathians. BAYGER (1937, 1959) was the first to publish local records of 7: montandoni

and T. alpestris. He listed 38 localities covering the Maloye Opolie area as a whole (BAYGER,

1938 — cited from GULIÈKA, 1954). Surprisingly, later in soviet time these Bayger”’s papers were

forgotten, and the locality labels of his Opolie specimens kept in the Ukrainian museums of

natural history in Kiev (e.g., a jar with the Rakovets sample of 7! montandoni, IZK 80) and in

Lvov (TARASHCHUK, 1959) were considered to be incorrect! However, relatively recently (in

1985), some localities mentioned by Bayger were repeatedly discovered by local researchers

(S.V. Shaytan, pers. com. — cited from TATARINOV, 1989; POLUSHINA et al., 1989).

In 1994-1996, we monitored distribution of both species in Maloye Opolie along a

gradient from the forest edge to deep forest which coincided with a gap arranged by forest

cutting and used by forest trucks. Three sites were observed.

(1) The first one was situated on the military site nearby the forest edge (295 m above sea

level). There were few water bodies (ditches, total length 10 m, width about 1 m, depth 0.5 m)

with dense water vegetation and mosses. These water bodies were visited in 1995 and 1996:

during our first visit only 7: vulgaris was found, whereas in the second visit we observed both

species and captured only 7: montandoni (tab. 1). The densities of the two species were quite

low.

(2) The second site was situated in the forest, 600-700 m from the edge, on a slope of a

small hill (350 m above sea level). Here in 1994, in wheel-tracks filled by water, we found four

newt species (tab. 1), with a predominance of 7: montandoni. The latter species was charac-

terized by high density: for instance, 123 individuals were collected in a wheel-track (length 25

m, width 0.6 m, depth 0.4 m). However, T. vulgaris was quite rare (two males and one female

only). In 1994, we also captured a female which had some external features similar to those of

T. montandoni, and black spots on the belly like in 7° vulgaris. The study by flow DNA

cytometry established that this female had a genome size (tab. 2) intermediate between those

of both species. Interestingly, the lab hybrids of the first generation, obtained by us in the cross

of a male T vulgaris and a female T! montandoni, had a similar genome size. Therefore, we

incline to identify this “morphological hybrid” female as a true first generation hybrid

Source : MNHN, Paris

166 ALYTES 20 (3-4)

between both species. The study of the second water body was continued in 1995 and 1996.

However, we failed to find either 7: vulgaris or new hybrids (tab. 1).

(3) The third site (350 m above sea level) was also situated in a wheel-track in deep forest,

2200-3000 m from the edge. We visited the water body in 1996 only and found only 7.

montandoni and T: alpestris. The former species was represented by some unusual individuals,

e.g. by a female with a shortened body or by other animals (1.5 % of the Maloye Opolie

sample) with partial or total lacks of black or orange pigmentation. However, such abnor-

malities in local animals seemed to have no association with any hybridisation between 7!

montandoni and T. vulgaris, because we found similar colour abnormalities in T: alpestris as

well, with an incidence up to 2.3 %. In contrast, we observed no abnormal individuals of these

species in the Carpathian mountains. We suggest that the appearance of such abnormalities

may be explained by peculiarities of the isolated newt populations of Maloye Opolie.

Consequently, despite the “enclave” status of 7: montandoni in the Maloye Opolie area,

the local contact zone between this species and 7! vulgaris seems to be very narrow. Only few

individuals of the latter species penetrated into the forest water bodies predominantly

inhabited by 7! montandoni. RAFINSKI (1985) reported that in Poland hybrid individuals were

more common in populations where T. vulgaris was more abundant rather than in populations

with predominating 7: montandoni. Our finding is in agreement with this observation because

the sole Maloye Opolie hybrid was found in the site with very sparse presence of T! vulgaris.

ACKNOWLEDGMENTS

We express our gratitude to Prof. N. N, Szezerbak (Kiev) for consultations and for providing permit

for collecting two samples of 7: montandoni for DNA cytometry. We thank O. A. Lugovoy and V. F.

Pokinchereda (Rakhov) for providing permit for work in the Carpathian Reserve and for their field

assistance. N. L. Orlov (St. Petersburg) shared with us an information on syntopie occurrence of the

newts. V. E Orlova (Moscow) offered the museum collection. Our special thanks to T. R. Halliday and an

anonymous referee who read an earlier version of the paper and made helpful comments. The St.

Petersburg Association of Scientists and Scholars provided facilities for preparation of the manuscripts.

S supported by grants of the International Science Foundation (Nr. R.60000 and R.60300)

and Russian Foundation of Basic Researches (Nr. 02-04-49631)

LITERATURE CITED

ARNTZEN, JW. & SparreBooM, M., 1989. — A phylogeny of the Old World newts, genus Zriturus:

biochemical and behavioural a. J Zool., Lond., 219: 645-664.

BAYGER, J. A. 1937. — Klucz do oznaczania plazôw i gadôw. Zeszyt II klucza do oznaczania zwierzgt

kregowych polski. Krakôw, Druk W. L. Anezyca i spôlki: 1-93.

= 1959. —"Z driejow poznania traski Karpackiej, Triturus montandoni (Boulenger). Wszechwiat,

78-79.

979. — [A comparative analysis of the sexual and territorial behaviour of 5 newts

of the genus Friturus (Urodela, Salamandridae)]. Larvijas PSR mugurkaulnieku uz

populaciju dinamika, Riga, P. Stuckas Latvijas Valsts Univ.: 29-49. [In Russian, with English

summary].

Source : MNHN, Paris

LITVINCHUK, BORKIN & ROSANOV 167

«8 1981. — [On application of some characters of sexual behavior to solution of problems of systematics

and evolution of animals (on an example of representatives of the genus Zriturus: Urodela,

Salamandroidea)]. Priroda i Muse, Riga, 1: 32-52. [In Russian].

COGALNICEANU, D., 1994. — Experimental hybridization within the Triturus vulgaris species-group

(Amphibia, Caudata). Rev. roum. Biol, (ser. Biol. anim.), Bucarest, 39: 145-150.

—— 1997. Triturus montandoni (Boulenger, 1880). /n: J.-P. GAsC, A. CABELA, J. CRNOBRNJA-Isailovic, D.

DOoLMEN, K. GROSSENBACI P. HAFFNER, J. LESCURE, H. MARTENS, J. P. MARTINEZ RICA, H.

MaURIN, M. E. OLIVERA, T. S. SOFIANIDOU, M. VErTH & A. ZUIDERWUK (ed.), Atlas of amphibians

and reptiles in Europe, Paris, Societas Europaca Herpetologica & Muséum national d'Histoire

naturelle: 84-85.

FUHN, L. E., 1963. - Beobachtungen über Amphibien- und Reptilien-Bastarde in freier Wildbahn. Acta

Soc. zool. Bohemosloven., 27: 70-73.

Fun, L. E., Sova, C. & Dimrrrescu, M., 1976. — Une population hybridogène Triturus v. vulgaris X T.

montandoni Boul. du lac Cräcurele (Mts. Nemira, départ. Bacäu). Muz. Stünf. Natur. Bacäu., Stud.

si Comun., “1975”: 225-236.

GuLièka, J., 1954. — K rozSireniu a ekolégii mloka karpatského (Triturus montandoni Boul.) na Slo-

vensku. Biologia, Casop. sloven. Akad. Vied. Bratislava, 9: 545-560.

HOFMANN, O., 1908. — Ueber Triton montandoni Blg. in Galizien. Lacerta, Beilage Wochenschr. Aquar.-

Terrarienk., 17: 65-66.

HorguLEwICZ, M. L., 1927. - Die Verbreitung der Bombinator- und Triton-Arten im Bereiche der Bezirke

Sambor, Drohobyez, Stryi (Kleinpolen). Bull. int. Acad. polon. sci. lettr, CI. sci, math. natur, (ser.

B: Sci. natur.), Cracovie, 1-2: 87-111.

JuszczyK, W., 1987. — Plazy i gady krajowe. 2. Plazy — Amphibia. 2% edition. Warszawa,

Panstwowe Wydawnictwo Naukowe: 1-384, pl. 1-58.

KorLik, P. SéLov, K., SLEcHTA, V. & ZAVADIL, V., 1997. — Allozyme evidence for hybridization between

Triturus vulgaris (Linnaeus, 1758) and Triturus montandoni (Boulenger, 1880) (Amphibia: Cau-

data) in the Czech Republic. Preliminary results. Z7: W. BôHME, W. BiscHorr & T. ZIEGLER (ed.),

Herpetologia Bonnensis: 195-198.

KusHNiRUK, V. À., 1963. - [On biology of the Carpathian newt Triturus montandoni Boulenger, 1880].

Zool. Zhurnal, Moscow, 22: 300-302. [In Russian].

MACGREGOR, H. C. K. & ARNTZEN J. W., 1990. - An integrative analysis of phylogenctic

relationships among newts of the genus Triturus (family Salamandridac), using comparative

biochemistry, cytogenetics and reproductive interactions. Z evol. Biol., 3: 329-373.

MALINOVSKY, A. K., 1991. -[Mountainous component of the Ukrainian Carpathian Flora], Kiev, Naukova

Dumka: 1-205. [In Russian].

Peci0, A. & RarINski, J., 1985. — Sexual behaviour of the Montandon’s newt, Triturus montandoni

(Boulenger) (Caudata: Salamandridae). Amphibia- Reptilia, 6: 11-22.

POLUSHINA, N. A., BODNAR, B. N. & MATKIVSKAYA, L. I. 1989. — [New data on distribution and the

number of amphibians listed in the Red Data Book in western Ukraine]. Zn: N. N. SZCZERBAK

(ed.), The problems of herpetology, Abstracts of the 7 USSR Herpetological Conference, Kiev,

Naukova Dumka: 199-200. [In Russian].

POLUSHINA, N. A. & KUSHNIRUK, V. O., 1962. - [Data on the batrachofauna of Lvov Province]. l

Lvov State Univ, (ser. Biol. 127-141. [In Ukrainian].

RAFINSKI, J., 1985. - Natural hybridization between Triturus vulgaris and T. montandoni. In: Third ord. gen

Meet. Soc. europ. Herpetol., Prague: 94-95. [Abstract].

RAFINSKI, J. & ARNTZEN, J. W., 1987. - Biochemical systematics of the Old World newts, genus Triturus:

allozyme data. Herpetologica, 43: 446-457.

Renk, L., 1993. - Colek karpatskÿ (Triturus montandoni) na Moravé a jeho kïiZeni s éolkem obecnÿm (7:

vulgaris). Akvarium Terrarium, Praha, 36 (3): 32-34.

Sova. €. 1973. - Contributions to the ecology of amphibians (ord. Caudata, genus Zriturus) from the

Sereth river basin. Univ. Bucarest, Fac. Biol., Stud. si Comur . Stünf. Natur. Bacäu, 6: 1-286.

S2CzERBAK, N. N. & SZCZERBAN, M. L., 1980. - [Amphibians and reptiles of the Ukrainian Carpathians].

Kiev, Naukova Dumka: 1-268. [In Russian].

SzymuRA, J. M., 1974. - A competitive situation in the larvae of four sympatric species of newts (Zriturus

cristatus, T. alpestris, T. montandoni and T. vulgaris) living in Poland. Acta Biol. Cracor., (ser.

Zool.), 17: 235-262.

Source : MNHN, Paris

168 ALYTES 20 (3-4)

TARASHCHUK, V.L., 1959. - [Amphibians and reptiles]. Fauna Ukrainy, Kiev, 7: 1-246. [In Ukrainian].

TaTarINov, K. A. 1989. - [Ecological plasticity of the Montandon’s and Alpine newts]. /m: N. N.

SzCzERBAK (ed.), The problems of herpetology, Abstracts of the 7!" USSR Herpetological Confe-

rence, Kiev, Naukova Dumka: 251-252. [In Russian].

VINOGRADOV, À. E., BORkIN, L. J., GÜNTHER, R. & ROSANOY J. M., 1990. - Genome elimination in

diploid and triploid Rana esculenta males: cytological evidence from DNA flow cytometry.

Genome, 33: 619-627.

WourersroRFr, W, 1925. - Katalog der Amphibien-Sammlung im Museum fur Natur- und Heimat-

kunde. Abh. Ber. Mus. Natur. Heimatk. naturwis. Verein Magdeburg, 4 (2): 155-234.

Corresponding editor: Tim HALLIDAY.

Source : MNHN, Paris

AINTTES

International Journal of Batrachology

published by ISSCA.

EDITORIAL BOARD

Chief Editor: Alain Dusois (Laboratoire des Reptiles et Amphibiens, Muséum national d'Histoire naturelle,

25 rue Cuvier, 75005 Paris, France; <dubois@mnhn.fr>).

Deputy Editor: Thierry LODÉ (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

GiLLrrsCH (Wien, Austria); Tim HALLIDAY (Milton Keynes, United Kingdom); W. Ronald HEYER

(Washington, USA): Esteban O. LaviLLA (Tucumän, Argentina); Masafumi Maïsui (Kyoto, Japan

Donald F. MCALPINE (Saint John, Canada); Alain PaGaNo (Angers, France); John C. POYNTON (London,

England): Miguel VENCES (Konstanz, Germany).

Technical Editorial Team (Paris, France): Alain Durois (texts); Roger BouR (tables): Annemarie OHLrr (figures).

Book Review Editor: Annemarie OHLER (Paris, France).

SHORT GUIDE FOR AUTHORS

(for more detailed Instructions 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 qualit:

{such as colour or black and white photographs), showing beautiful or rare species, interesting behaviours, etc.

The title should be followed by the names) 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 cm. The size of the lettering should ensure its legibility after reduction. The legends of figures and tables

should be assembled on à 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. Hanoï, Institut Océanographique de l’Indochine: i-x + 1-547,

GRAF, .-D. & Pos PELAZ, M., 1989. - Evolutionary genetics of the Rana esculenta complex. In: R. M. DAWLEY

& J. P. BoGarT (ed.), Evolution and ecology of unisexual vertebrates, Albany, The New York State Museum:

INGER, R. EF, Voris, H. K. & Voris, H. H., 1974. - Genetic variation and population ecology of some Southeast

Asian frogs of the genera Bufo and Rana. Biochem. Genet. V2: 121-145.

Manuseripts should be submitted in triplicate cither to Alain Dumois (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 1wo referces 4

If possible, after acceptance, a copy of the final manuscript

the Chief Editor. We welcome the following formats of text proc Word (L.1 10 6.0, DOS

KA. to 3.1, DOS or Windows) or WordStar (3.3 to 7.0 ii $ preferably, formated

DOS (ASCII) or DOS-formated M$ Word for the Macintosh (on a snsity 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 10 the

author(s). Additional reprints may be purchased.

5 1) should be sent to

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 DUroNs.

Numéro de Commission Paritaire: 64851.

Source : MNHN, Paris