c°

HARVARD UNIVERSITY

ASIATIC

HEMATOLOGICAL

Li

ARVARD

VERSITY

RESEARCH

.

*5

■ft

* •

VOLUME 5

1993

Asiatic Herpetological Research

Editor

Ermi Zhao

Chengdu Institute of Biology, Academia Sinica, Chengdu, Sichuan, China

Associate Editors

J. Robert Macey

Department of Biology, Washington University, St.

Louis, Missouri, USA

Theodore J. Papenfuss

Museum of Vertebrate Zoology, University of

California, Berkeley, California, USA

Editorial Board

KRAIG ADLER

Cornell University, Ithaca, New York, USA

NATALIA B. ANANJEVA

Zoological Institute, St. Petersburg, Russia

STEVEN C. ANDERSON

University of the Pacific, Stockton, California, USA

AARON BAUER

Villanova University, Villanova, Pennsylvania, USA

LEO BORKIN

Zoological Institute, St. Petersburg, Russia

BIHUI CHEN

Anhui Normal University, Wuhu, Anhui, China

ILYA DAREVSKY

Zoological Institute, St. Petersburg, Russia

INDRANEIL DAS

Madras Crocodile Bank, Vadanemmeli Perur, Madras, India

WILLIAM E. DUELLMAN

University of Kansas, Lawrence, Kansas, USA

HAJIME FUKADA

Sennyuji Sannaicho, Higashiyamaku, Kyoto, Japan

CARL GANS

University of Michigan, Ann Arbor, Michigan, USA

HUI-QING GU

Hangzhou Teacher's College, Hangzhou, Zhejiang, China

ROBERT F. INGER

Field Museum, Chicago, Illinois, USA

MAHMOUD LATIFI

Institut d'Etat des serums et vaccins Razi, Teheran, Iran

KUANGYANG LUE

National Taiwan Normal University, Taipei, Taiwan, China

RONALD MARLOW

University of Nevada, Las Vegas, Nevada, USA

ROBERT W. MURPHY

Royal Ontario Museum, Toronto, Ontario, Canada

GOREN NILSON

University of Goteborg, Goteborg, Sweden

HIDETOSHI OTA

Department of Biology, University of the Ryukyus, Nishihara,

Okinawa, Japan

JIONG-HUA PAN

South China Normal University, Guangzhou, Guangdong,

China

YUN-XU TONG

Lanzhou University, Lanzhou, Gansu, China

KE-MING XU

Liaoning Normal University, Dalian, Liaoning, China

YU-HUA YANG

Sichuan University, Chengdu, Sichuan, China

KEN-TANG ZHAO

Suzhou Railway Teacher's College, Suzhou, Jiangsu, China

Asiatic Herpetological Research is published by the Asiatic Herpetological Research Society (AHRS) and the

Chinese Society for the Study of Amphibians and Reptiles (CSSAR) at the Museum of Vertebrate Zoology, University

of California. The editors encourage authors from all countries to submit articles concerning but not limited to Asian

herpetology.

Authors should consult Guidelines for Manuscript Preparation and Submission at the end of this issue.

All correspondence outside of the of China and requests for subscription should be sent to AHR, Museum of Vertebrate

Zoology, University of California, Berkeley, California, USA 94720. All correspondence within China should be sent

to Ermi Zhao, Editor, Chengdu Institute of Biology, P.O. Box 416,Chengdu, Sichuan Province, China.

Subscription and membership are $25 per year ($45 for libraries). Postage outside of the USA and China, please

add $5 per issue for surface mail or $10 per issue for air mail. Make checks or money orders payable in US currency to

AHRS. If you do not have access to US currency, please notify us, and we will make other arrangements.

Asiatic Herpetological Research Volume 5 succeeds Volume 4 published in 1992, Volume 3 published in 1990

and Chinese Herpetological Research Volume 2, which was published at the Museum of Vertebrate Zoology,

1988-1989 as the journal for the Chinese Society for the Study of Amphibians and Reptiles. Volume 2 succeeded

Chinese Herpetological Research 1987, published for the Chengdu Institute of Biology by the Chongqing Branch

Scientific and Technological Literature Press, Chongqing, Sichuan, China. Acta Herpetologica Sinica ceased

publication in June, 1988.

Cover: Eublepharus turkmenicus from vicinity of Temen Spring, 2.5 km west of Danata (39° 07' N 55° 08' E),

Krasnovodsk Region, Turkmenistan. Photo by J. Robert Macey.

December 1993

Asiatic Herpetological Research

Vol. 5, pp. l-To]

i U

Coluber atayevi Sp. Nov. (Ophidia, Colubridae) from the Kopet-Dag

Mountains of Turkmenistan

BORIS S. TUNIYEV1 AND SAHAT M. SHAMMAKOV2

Caucasian Slate Biosphere Reserve, Sochi, Russia

^Institute of Zoology, Turkmenistan Academy of Sciences, Azadi Street 6, 744000 Ashgabat,

Turkmenistan

Abstract. -An examination snakes, formerly considered to be Coluber najadum, from the Kopet-Dag

Mountains of Turkmenistan, leads us to believe that this population represents an undescribed species. We

here describe this population as Coluber atayevi.

Key words: Reptilia, Ophidia, Colubridae, Coluber, Turkmenistan, systematics.

Introduction

During field work in the Kopet-Dag

Mountains, Turkmenistan, we observed

and analyzed many individuals of Coluber

najadum under natural conditions.

Comparison of these individuals with

Caucasian material led to the conclusion

that the Kopet-Dag snakes belong to a

previously unrecognized species. This

conclusion is supported by significant

morphological divergence from the other

representatives of the najadum-rubriceps

complex.

Methods

We analyzed 10 specimens of Coluber

najadum (Eichwald) from various areas of

the Caucasian Isthmus and 10 specimens

from the Kopet-Dag belonging to the new

taxon. Morphometric data were compared

with the available information in the

literature about Coluber najadum in its

natural habitat throughout its range

(Terentjev and Chernov, 1949; Bannikov et

al., 1977) and for separate regions

(Ananjeva and Orlov, 1977;

Muzskheleshvili, 1970).

The following features and indices have

been used: 1) L. = length of body, mm; 2)

L. cd. = length of tail, mm; 3) Sq.=

number of scales around body; 4) Ventr.=

number of ventral scutes; 5) S. cd.=

number of subcaudal scutes; 6) Lab.=

number of upper labials; 7) Sublab.=

number and size of sublabials; 8) form of

mandibular scutes; 9) shape of the head;

10) distribution of scutes on throat; 11)

L/L. cd.= body length/tail length; 12) Pr.

oc.= number of preorbital scales; 13) Post.

oc.= number of postorbital scales; 14)

Temp.= number of temporal scales; 15)

A.= form of anal scutes.

For numerical features and indices, we

have calculated the mean (x), mean error

(m), mean square deviation (S2), using the

formulas for small samples (Lakin, 1980)

When describing the biotopes, we

determined the plant species according to

Nitikin and Geldykhanov (1988); the

general vegetation type follows Korovin

(1934), with some corrections.

History of the Study of ''Coluber

najadum'" in the Kopet-Dag

Zamenis dahli Fitzinger was first

mentioned from the environs of Sulukli

Spring and the Kuchan road by Varentzov

(1894). Nikolskij (1905, 1916) observed

this species in the vicinity of Ashkhabad.

On the basis of these records, Coluber

najadum (Eichwald) was included in the list

of the Turkmenistan herpetofauna

(Chernov, 1934; Terentjev and Chernov,

1949; Bogdanov, 1962). All previous

collections from the territory of

Turkmenistan and the neighboring parts of

Iran have been analyzed by Ananjeva and

Orlov (1977). They added the localities of

Vol. 5, p. 2

Asiatic Herpetological Research

December 1993

TABLE 1 . Morphometry of Coluber atayevi paratypes in the collection of the Caucasian

Reserve, Sochi, Russia (see text for abbreviations).

No. L. L.cd. Sq. L./L.cd Ventr. S.cd. Lab. Preoc. Postoc. Temp. Sublab.

Firuza settlement and, tentatively, Dzhebel

Station to the distribution of Coluber

najadum. The last locality was doubtful

(Shcherbak and Golubev, 1981; Ataev

1985). Rustamov and Shammakov (1979)

and Shcherbak and Golubev (1981)

mentioned it from Dushak Mountain.

Recent localities include the Babazon region

in the Kopet-Dag Reserve (Shcherbak, et

al., 1986), Saivan and Imarat villages and

Kara-Kala settlement (Ataev, et al., 1991).

During almost a century, less than 20

representatives of this taxon have been

recorded, half of them in recent years

(Ataev etal., 1991).

The small amount of preserved material,

part of which had been lost (Ananjeva and

Orlov, 1977), dissociation of time of

collection and place of storage of

specimens, led to the opinion that Coluber

najadum was the taxon distributed in the

Kopet-Dag. This form was included in the

nominate form, and had never even been

considered as a separate subspecies

(Bannikov et al., 1977). It is interesting to

note that Bannikov et al. (1977) included

C. n. rubriceps Mertens (now recognized

as a distinct species [Engelman et al., 1986;

Rehak, 1986; Ananjeva et al., 1988]) in the

synonymy of Coluber najadum.

Coluber atayevi

Shammakov, sp. nov.

Tuniyev and

Zamenis dahli: Varentzov, 1894:27;

Nikolskij, 1905:233; 1916:92.

Coluber najadum: Chernov, 1934:273;

Terentjev and Chernov, 1949:240-241

(part); Bogdanov, 1962:167; Bannikov et

al., 1977:262-263 (part); Ananjeva and

Orlov, 1977:14-16; Rustamov and

Shammakov, 1979:144; Shcherbak and

Golubev, 1981:70-72; Ataev, 1985:242-

243; Shcherbak et al., 1986:98-100; Latifi,

1991:102-103.

We name the new species in honor of the

famous Turkmen herpetologist, Chary

Ataevich Ataev, who studies reptiles of the

mountains of Turkmenistan.

Holotype: Collection of the Caucasian

Reserve, Sochi, Russia, No. 420, adult

male, environs of Saivan Village, Saivan-

Nokhur Plateau, western Kopet-Dag,

Bakharden Region, Turkmenistan, 12 May

1991, collected by B. S. Tuniyev (Fig. 1).

Paratypes: Twenty two specimens.

Collection of the Caucasian Reserve,

Sochi, Nos. 421-429, 4 adults and 5

juveniles, same data as holotype, collected

by C. A. Ataev and B. S. Tuniyev;

California Academy of Sciences (CAS)

Nos. 182948-182950, same locality as

holotype, May 1990; CAS 185185-

185194, 5 adults and 13 juveniles, Elev.

1200-1300 m, 38° 30 N, 56° 47' E, 2 km

SE (airline) of Saivan, Ashgabad Region

Turkmenistan, 21 May 1992, collected by

B. S.. Tuniyev, S. M. Shammakov, N. B.

Ananjeva, T. J. Papenfuss and R. Macey

(Plate 1).

Tuniyev and Shammakow

Asiatic Herpetological Research

Adult Coluber atayevi.

Type locality of Coiluber atayevi, environs of Saivan Village, Saivan-Nokhur Plateau,

western Kopet-Dag, Bakharden Region, Turkmenistan.

December 1993

Asiatic Herpetological Research

Vol. 5, p. 3

FIG. 1 . Holotype of Coluber atayevi sp. nov. (Collection of the Caucasian Reserve no. 420).

Description of holotype: Snout-

vent length 500 mm, tail 172 mm; head

length 19.6 mm, head width 8.4 mm, head

height 6.2 mm. Head smoothly rounded,

narrow, covered with large regular scutes;

8 upper labials, 5th upper labial touches

lower postorbital and large lower temporal

with its extended upper posterior side; 9

lower labials, 6th largest, last two pairs

almost covered by upper labials; a single

preorbital on either side, 2 small scutes

below; posterior pair of upper temporals

slightly larger than anterior one; seen from

above, rostral extends slightly between

internasals. Narrow genial scutes in

contact along mental groove, no space

between posterior genials.

Nineteen scale rows at midbody; 206

abdominal scutes; 97 pairs of subcaudals;

anal divided. Scales bordering abdominal

scutes of same size as other lateral scales;

body scales smooth, rhombic; ridge on

lateral aspect of abdominal scutes indistinct,

almost absent.

Coloration in preservative:

Dorsum gray, venter grayish; 5 large dark

ocelli bordered with light circles on sides of

neck; lateral row of small black dots ending

abruptly on anterior third of body. Eye

outlined with white lines extending

anteriorly and posteriorly; lower white

stripe extends over upper labials; a narrow

black streak running posteriorly and down

from eye, situated on 5th upper labial and

slightly touching 6th.

Description of paratypes: Counts

and measurements of the paratypes in the

collection of the Caucasian Reserve, Sochi,

Russia are given in Table 1 .

Diagnosis: Comparatively small

snake (Fig. 2), smaller than C. n. najadum,

C. n. dahli, and C. r. rubriceps in

dimensions. It is comparable in size with

the European subspecies, C. rubriceps

thracius. In contrast to C. najadum, whose

tail constitutes 1/3 of its total length, C.

atayevi has a comparatively short tail,

Vol. 5, p. 4

Asiatic Herpetological Research

December 1993

-•iSfe, - '

FIG. 2. Representatives of the " najadum-rubriceps" -complex. Left- Largest specimen of Coluber atayevi

sp. nov. (Collection of the Caucasian Reserve no. 421). Right- Medium-sized specimen of Coluber

najadum (Sochi environs, Maly Akhun, Collection of the Caucasian Reserve no. 94).

approximately 1/4 of total length, also

characteristic of C. rubriceps.

Habitus and elements of coloration of C.

atayevi are intermediate between C .

najadum and C. rubriceps; its narrow,

sharp, and flat head with the rostrum

beveled downward is closer to the C.

rubriceps head shape than to that of C.

najadum, with its comparatively wide,

rounded, and high head, where the upper

and lower surfaces of head are parallel.

Color pattern of C. atayevi resembles that

of C. najadum, but brown colors prevail

instead of olive-green ones. Lateral

abdominal ridges are practically absent in

C. atayevi, in contrast to the above-

mentioned species, and consequently the

body is round in cross-section and not

rectangular as in C. najadum and C .

rubriceps.

Genial scutes of C. atayevi contact one

another along the mental groove, rarely

having a few isolated granules between the

posterior pair, whereas between the widely

separated posterior genials of C. najadum

there are always 2-4 rows of well-

developed scales (Fig. 3). Posterior upper

labials of C. atayevi are weakly

December 1993

Asiatic Herpetological Research

Vol. 5, p. 5

FIG. 3. Distribution of head scales on

representatives of the Coluber najadum-Coluber

rubriceps-complex: a, b, c- a young specimen of C.

najadum; d, e, f- a young specimen of C. atayevi

sp. nov.

distinguished from the throat scales,

whereas all upper labials of C. najadum are

strongly pronounced.

Geographic distribution: The range

of Coluber atayevi includes the western and

central Kopet-Dag, from the surroundings

of the Kara-Kala settlement in the west to

the Sulukli Spring in the east. This is an

upland species, associated with such

vegetation types as "prashiblyak," "broad-

leaved forest" (Kamelin, 1970) and

"phrygana," and in the western part of the

uplands, where these plant associations

occur at lower elevations, individuals of C.

atayevi are found at elevations of 400-1600

meters (Shcherbak et al., 1991). At the

eastern end of its range (Dushak

Mountain), the snake has been found at

2000 meters elevation (Shcherbak et al.,

1986).

Biotopes: According to our

observations, Coluber atayevi is found on

the highest parts of the Saivano-Nokhur

Plateau and on the crests of mountains at

900-1400 meters elevation. The most

typical biotopes of the species are ecotones

of mesophilous derivatives of deciduous

forest and meadow-steppe coenosis along

the edges of small ravines having deposits

of limestone and argillaceous slates.

Indicators of the deciduous forest are

isolated old trees of Oriental plane (Platanus

orientalis) and English walnut (Juglans

regia). Forest plots of "prashiblyak" are

typified by Aceretum fruticans and Acer

turcomanicum, a subdominant role played

by Crataegus turcomanica, Lonicera

floribunda, Prunus cerasifera, Cotoneaster

nummularioides, Cotoneaster ovatus, and

Rubus anatolicus. In the herb layer, there

are such species as Alliaria petiolata,

Lamium album, Geranium pussilum, Arum

juquemontii, and Allium paradoxum.

Rocky-shrubby vegetation is usual for

the ecotone of forest ravines on the rocky

and scree slopes, with isolated trees of

Celtis caucasica and many different shrubs

and semi-shrubs: Colutea gracillis,

Ephedra equisitina, Thelycrania meyeri,

Rhamnus coriacea, Hymenocrater

bituminosus, and Artemisia turcomanica.

Representatives of Coluber atayevi are

found in Cousinia smirnovii associations

and Astragalus piletoclados groups

(phrygana type) in the immediate vicinity of

the forest and shrubby communities with

steppe wedges, mainly in overgrazed

places. The region is comparatively well-

watered, because almost every ravine has a

spring or stream.

The vegetation of the eastern border of

the distribution of C. atayevi is described

by Korovin (1934): "Dushak mountain is

an isolated massif, formed by light

limestone. Its steep slopes serve as a home

for typical mountainous xerophytes. Here

we find juniper, both isolated trees and

groups of them. A number of semishrubs

and xerophytic herbs form flora of these

mountains." Later, the author mentions the

domination of shrubs of Astragalus

piletoclados, as well as groups of different

species of Acantholimon and here very

common gray cushion-like groups of

Onobrychis comma. Korovin concludes

Vol. 5, p. 6

Asiatic Herpetological Research

December 1993

that the cushion-like xerophytes of the

Kopet-Dag are better developed on the tops

of mountains (about 2000 meters

elevation). This is higher than the steppe

zone, so phrygana belongs to high-altitude

vegetation.

To our regret, absence of data about the

biotopes of species from the other places

gives us no opportunity to characterize the

Cenozoic ties of the species throughout the

whole area.

Population density: Coluber atayevi

is the most numerous snake species on the

Saivano-Nokhur Plateau. Six specimens

were found during a three-hour excursion

in the vicinity of Saivan village in May,

1990; 12 specimens were noticed during

the same period of time in May, 1991. The

largest number of snakes (as many as 5

specimens per 300 meters) was among the

shrubs of rocky-scree plots of ravines.

Isolated specimens of snakes were met in

stoneless places. The fact that this species

is rather common for the western Kopet-

Dag is proved by the data of Ataev et al.

(1991). All other authors mention only

isolated findings, reckoning it among the

rarer species of the Kopet-Dag.

Apparently, the sporadic distribution of the

species and the considerable altitudes at

which its habitat occurs, are the reasons

why it is rarely met. It is not excluded that

its population density in the eastern part of

its range is significantly lower than in its

western part.

Seasonal and daily activity:

Presumably, the species' activity begins in

the middle of April, considering

temperature conditions of this mountain

zone (Babaev, et al., 1982) and preferable

temperatures of daily activity, noted in

May, 1990-1991. Coluber atayevi is a

diurnal species with two-peak activity in

May; the morning peak (9:00-11:00) is

strongly pronounced and the evening one is

feebly marked. The snakes are active in

sunny, windless weather. We have never

seen them when it rains, there are strong

winds, or heavy overcast. In bad weather

the snakes are absent not only on the

surface, but from under the plates of slate,

where they are usually met in sunny

weather.

Diet: Lizards in the habitat of Coluber

atayevi are Ablepharus pannonicus, Stellio

caucasicus, and Pseudopus apodus.

Cyrtopodion caspius and Eremias strauchi

are common, though not so numerous,

while Mabuya aurata and Eumeces

taeniolatus are rare. Considering the small

size of the head and body of Coluber

atayevi, the bulk of its diet must be formed

only of Stellio caucasicus, young

specimens of Eremias strauchi, and

possibly Cyrtopodion caspius. Evidently,

C. atayevi is a saurophage exclusively,

since the common and abundant Microtus

socialis and Saxetania cultricolis (a

micromammal and an orthopteran insect,

respectively) are too large to be eaten by

this snake.

Syntopical species of snakes:

Subdominants of Coluber atayevi are C.

nummifer, C. ravergieri and Vipera

lebetina; Typhlops vermicularis is common;

Agkistrodon halys caucasicus and Natrix

tesselata are rare. At the borders of the

species' biotopes Eirenis meda,

Psammophis lineolatum, Naja oxiana, and

Eryx miliaris are met. These species are

more characteristic for smaller hypsometric

marks.

Discussion

Coluber atayevi is most closely related to

C. najadum and C. rubriceps, possessing

features of both species; in habitus (head

shape, in particular) it resembles C .

rubriceps, but it is similar to C. najadum in

color pattern. We should note that these

features are characteristic for juvenile

specimens as well as adults; in other

words, we cannot distinguish ancestral

features that would allow us to unite C.

atayevi with either of the species mentioned

(Table 2, Fig. 3).

Judging from the contemporary

distribution of the three species, we

propose that the center of the complex is the

Eastern Mediterranean region, so-called dry

land of Asia Minor or Balkan-Caucasian

December 1993

Asiatic Herpetological Research

Vol. 5, p. 7

Vol. 5, p. 8

Asiatic Herpetological Research

December 1993

dry land. The ancestral form could have

been widespread in the Upper Miocene-

Pliocene over subtropical semiarid regions

of Asia Minor united with the Central Asian

Massif, Caucasian Island and the Balkan

Mountains in the Miocene. Mixed savanna-

hylile landscape, combining alternation of

open spaces with subtropical forests were

characteristic for the northern parts of the

Iranian Plateau in that period. Fossil

mammals and birds testify to this fact

(Vereshchagin, 1959). In Vereshchagin's

opinion, similarity of the Upper Miocene

fauna unites Central Asia, the Caucasus,

the Crimea, and the Balkan Mountains.

Analysis of fossil tortoises testifies to the

presence of a bridge between the Balkan-

Caucasian dry-land and an Iranian-

Pakistanian Peninsula (Chkhikvadze,

1991).

Alpine orthogenesis that has led to the

formation of mountain relief over the whole

area of study, from the Balkan Mountains

to the Kopet-Dag, evidently promoted the

breakup of the original area of the ancestral

form of the najadum-rubriceps complex. A

number of populations along the sea coasts

of that time (Mediterranean, Black, and

Caspian Seas) could already have been

isolated in the Pleistocene. The event has

been connected with aridization of the

continental interior and the formation of

new ecological conditions. These

conditions have given impetus to the

development of the highland xerophytic

vegetation of Iran and the formation of the

fauna of arid mountains. It should be noted

that the vegetation of the central and

western Kopet-Dag has more features in

common with the vegetation of southern

and western territories than with the other

mountain massifs of Central Asia. Korovin

(1934) emphasizes that the Kopet-Dag

forests are the connecting link with

Mediterranean macchias. The provenance

of highland xerophytes (phrygana), as well

as forest vegetation, Korovin stated to be

Persia and Armenia.

Pleistocene cataclysms, connected with

the glaciation in the Caucasus, Asia Minor,

and the Iranian Plateau, as well as the

changes in the basins of the Black and

Caspian seas, could have favored the

secondary overlapping of the diverged C.

najadum and C. rubriceps. This sympatry,

which can be seen today in some areas,

served to reinforce the species divergence.

In the east, representatives of this complex

could survive only in relict populations on

the northern (Caspian) slope of the Alborz

Range and, in isolated form, in the

derivatives of forest coenosis in the western

Kopet-Dag that were never subjected to

glaciation (Gvozdetzkij and Mikhailov,

1987). In the Iranian part of the Turkmen-

Khorasan mountains there are islands of

gyrkana forests to the east of Astrabad,

reaching the longitude of Gombedezh-

Kabus. Some islands of oak forests reach

Bodgenurd in Khorasan (Menitskij, 1984).

Anderson (1968) wrote about possible

preservation of lizard refuges since the

Pleistocene in the Kopet-Dag mountains.

Increasing climatic aridization in the

Holocene can be the cause of disappearance

of C. atayevi in the foothills and severe

restriction of its area in the middle-altitude

and high-altitude parts of the western and

central Kopet-Dag and, possibly, to the

breakup of the area into several local

refuges. The "primitive" morphological

features were preserved in the absence of

contacts with closely related forms. In any

case, only further study of species

variability in specific localities of the

Kopet-Dag can throw light on this

question.

Literature Cited

ANANJEVA, N. B., L. YA. BORKIN, I. S.

DAREVSKIJ, AND N. L. ORLOV. 1988.

Pyatiyazychnyj slovar' nazvanij zhivotnykh.

Amphibii i reptilii [Five-language dictionary of

animal names. Amphibians and reptiles].

Russian Language Publishing House, Moscow.

290 pp.

ANANJEVA, N. B. AND N. L. ORLOV. 1977. O

nakhodkakh olivkovogo poloza Coluber

najadum (Eichw.) v Jugo-Zapadnoj Turkmenii

[About the Coluber najadum finds in

southeastern Turkmenia]. Transactions of the

Zoological Institute of the Academy of

Sciences, USSR 74:14-16.

December 1993

Asiatic Herpetological Research

Vol. 5, p. 9

ANDERSON, S. C. 1968. Zoogeographic analysis

of the lizard fauna of Iran. Pp. 305-37 1 . In The

Cambridge History of Iran. Vol. 1. The land of

Iran. 784 pp.

ARNOLD, E. N. AND J. A. BURTON. 1978. A

field guide to the reptiles and amphibians of

Britain and Europe. Collins, London. 272 pp.

ATAEV, CH. 1985. Presmykayushchiesya gor

Turkmenistana [Reptiles of the Turkmenistan

mountains]. Ylym publishing House,

Ashkhabad. 344 pp.

ATAEV, CH., YU. KHOMUSTENKO, AND S.

SHAMMAKOV. 1991. Novye dannye o

rasprostranenii i chislennosti nekotorykh

redkikh vidov zmej v Yugo-Zapadnom Kopet-

Dage [New data concerning distribution and

quantity of some rare snake species in the

southwestern Kopet-Dag]. Herpetological

Investigations, Leningrad 1:51-53.

BABAEV, A. G. AND KH. D. DURDYEV. 1982.

Physico-geographicheskaya khakteristika

Zapadnogo Kopetdaga. Priroda Zapadnogo

Kopetdaga [Physico-geographic characteristics of

the western Kopet-Dag. Nature of the western

Kopet-Dag]. Ylym Publishing House,

Askhabad.

BANNIKOV, A. G., I. S. DAREVSKIJ, V. G.

ISHENKO, A. K. RUSTAMOV, AND N.N.

SHCHERBAK. 1977. Opredelitel'

zemnovodnykh i presmykayushchikhsya fauny

SSSR [Identification guide to the amphibians

and reptiles of the USSR fauna].

Prosveshchenie Publishing House, Moscow.

414pp.

BOGDANOV, O. P. 1962. Presmykayushchiesya

Turkmenii [Reptiles of Turkmenia]. Academy

of Sciences Publishing House, Ashkhabad.

167pp.

CHERNOV, S. A. 1934. Presmykayushchiesya

Turkmenii [Reptiles of Turkmenia].

Transactions of the Council for Productive

Forces Study. Turkmen Series. Issue 6.

Academy of Sciences of the USSR Publishing

House.

CHKHIKVADZE, V. M. 1991. Cherepakhi

kainozoya SSSR [Tortoises of the Cainozoic

Period found on the territory of the USSR].

Abstract of thesis. Tbilisi. 60 pp.

ENGELMAN W.-E., J. FRITZSCHE, R. GUNTHER,

AND F. J. OBST. 1986. Lurche und Kriechtiere

Europas. Ferd. Enke Verlag, Stuttgart. 420 pp.

GVOZDETSKIJ, N. A., AND N. I. MIKHAJLOV.

1987. Fizicheskaya geographiya SSSR.

Aziatskaya chast' [Physical geography of the

USSR. The Asian part]. Vysshaya Shkola

Publishing House, Moscow.

KAMELIN, R. V. 1970. Botaniko-

geographicheskie osobennosli flory Sovetskogo

Kopetdaga [Botanical and geographic features of

the Soviet Kopet-Dag flora]. Botanical

Magazine 55(10).

KOROVIN, E. N. 1934. Rastitel'nost' Srednej Azii

i Yuzhnogo Kazakhstana [Vegetation of Central

Asia and southern Kazakhstan]. Association of

the State Publishing Houses, Moskow-

Tashkent. 443 pp.

LAKIN, G. F. 1980. Biometriya [Biometry].

Vysshaya Shkola Publishing House, Moscow.

290 pp.

MENITSKIJ, YU. L. 1984. Duby Azii [Oaks of

Asia]. Nauka Publishing House, Leningrad.

MUSKHELISHVILI, T. A. 1970.

Presmykayushchiesya Vostochnoj Gruzii

[Reptiles of southern Georgia]. Metsniereba,

Tbilisi.

NIKITIN, V. V. AND A. M. GELDYKHANOV.

1988. Opredelitel' rastenij Turkmenistana

[Identification guide to Turkmenian plants].

Nauka Publishing House, Leningrad. 660 pp.

N1KOLSKIJ, A. M. 1905. Presmykayushchiesya i

zemnovodnye Rossijskoj imperii (Herpetologica

rossica) [Reptiles and amphibians of the

Russian Empire (Herpetologica rossica)]. Zap.

Imp. Acad, nauk [Transactions of the Imperial

Academy of Sciences] 1 7(1): 1-518.

NIKOLSKIJ, A. M. 1916. Presmykayuschiesya

(Reptilia) [Reptiles (Reptilia)]. Fauna Rossii i

sopredel'nykh stran [Fauna of Russia and

neighboring countries]. Academy of Sciences

Publishing House, Petrograd 2:1-350.

RUSTAMOV, A. G. AND S. M. SHAMMAKOV.

1979. Redkie i ischezayushchie vidy reptilij

Turkmenistana [Rare and disappearing reptile

species of Turkmenistan]. Okhrana prirody

Turkmenistana [Turkmenistan Nature

Preservation]. Ylym Publishing House,

Ashkhabad: 139- 146.

Vol. 5, p. 10

Asiatic Herpetological Research

December 1993

SHCHERBAK, N. N. AND M. L. GOLUBEV. 1981.

Novye nakhodki zemnovodnykh i

presmykayushchikhsya v Sredney Azii i

Kazakhstane [New finds of reptiles and

amphibians in Central Asia and Kazakhstan].

Vestn. zool. [Zoological Bulletin] 1981(1):70-

72.

SHCHERBAK, N. N., YU. D. KHOMUSTENKO AND

M. L. GOLUBEV. 1986. Zemnovodnye i

presmykayushchiesya Kopetdagskogo

gosudarstvennogo zapovednika i prilezhashchikh

territorij [Amphibians and reptiles of the Kopet-

Dag State Reserve and the neighboring

territories]. Priroda Tsentral'nogo Kopetdaga

[Nature of the central Kopet-Dag]. Ylym

Publishing House, Ashkhabad:98-100.

TERENTJEV, P. V. AND S. A. CHERNOV. 1949.

Opredelitel' presmykayushchikaya i

zemnovodnykh [Identification guide to reptiles

and amphibians]. Soviet Sciences Publishing

House, Moscow. 315.

VARENTSOV, P. A. 1894. Nablyudenija nad

pozvonochnymi i spiski zhyvotnykh,

najdennych v 1890-1892 gg. v Zakaspijskoj

oblasti [Observations of the vertebrates and list

of animals found in 1890-1892 in the

Transcaspian Region]. Prilozhenie k obzoru

Zakaspijskoj oblasti za 1892 g. Fauna

Zakspijskoj oblasti [Appendix to the review of

the Transcaspian Region in 1892. The

Transcaspian Region fauna], Ashkhabad: 3-38.

VERESHCHAGIN, K. K. 1959. Mlekopitayushchie

Kavkaza [The mammals of the Caucasus].

Academy of Sciences of the USSR Publishing

House, Moscow and Leningrad. 246pp.

I December l'W

Asiatic Herpetological Research

Vol. 5, pp. 11-13]

On the Status of Rhacophorus prasinatus Mou, Risch, and Lue

(Anura: Rhacophoridae)

WEN-HAO Chou

Department of Zoology, National Museum of Natural Science, Taichung, Taiwan 40419,

Republic of China

Key Words: Amphibia, Anura, Rhacophoridae, Polypedates prasinatus, Rhacophorus smaragdinus, Taiwan,

New Combination.

FIG. 1. The species of Polypedates and Rhacophorus on Taiwan: P. megacehalus (upper left), R.

taipeianus (upper right), R. moltrechti (lower left), P . prasinatus (lower right).

The emerald tree frog, Rhacophorus

prasinatus Mou, Risch and Lue, 1983 (=/?.

smaragdinus Lue and Mou, 1983), has had

an uncertain taxonomic status. When first

described, it was thought to be closely

related to R. chenfui, currently placed in

Polypedates (Jiang, Hu and Zhao, 1987),

on morphological grounds and to P.

leucomystax on account of genetic

similarity. Having examined the external

physical structure, muscular structure and

stained skeletons of specimens from the

type locality, I conclude that this species

should be classified in Polypedates in

accordance with the definitions and criteria

set forth by Liem (1970) and Jian et

al.( 1987) to distinguish the genera

Rhacophorus and Polypedates. The

following are the key features observed:

Dermal fold along outer edges of

forearm and above anus in rows of

Vol. 5, p. 12

Asiatic Herpetological Research

December 1993

tubercles, tarsal fold not evident or absent.

Fingers half webbed; web of all toes except

the fourth extending beyond subarticular

tubercles; fourth toe webbed to middle

subarticular tubercle or somewhat beyond.

Dorsal view of intercalary cartilage heart-

shaped. Esophageal and lateral processes

of laryngeal apparatus present. M.

extensor radialis accessorius lateralis

moderately large, originating along lateral

side of humerus and inserting on the disto-

dorsal end of radio-ulna; M. extensor

brevis superficialis of the first digit present.

Vomerine teeth present. Parieto-squamosal

arch of the fronto-parietal bone absent.

Vertebral column procoelous.

Polypedates prasinatus (Mou, Risch

and Lue 1983) comb. nov.

Rhacophorus prasinatus Mou, Risch and

Lue, 1983;

Rhacophorus smaragdinus Lue and

Mou, 1983

The two junior synonyms share the same

holotype and bear the same publication

year. However, the Rhacophorus

prasinatus has priority. It was published

on 30 December 1983 in Alytes 2(4) which

was mailed on that date, as indicated in the

journal. Although R. smaragdinus has the

same publication date in the Journal of

Taiwan Museum 36(2), I have learned from

the editor, S. F. Hung, that the mailing date

of this issue was 17 January 1984. This

supports a previous arbitrary adoption of

the first junior synonym (Frost, 1985).

In general, Polypedates and

Rhacophorus species can be distinguished

from the other treefrogs by the Y-shaped

terminal phalanx and absence of anterior

horns of the hyoid (Liem, 1970). The

species in Taiwan (Fig. 1) can be

distinguished by the following key:

Key To Species Of Polypedates And Rhacophorus Of Taiwan

la. Dermal fold along the forearm absent or in rows of tubercles; tarsal fold not evident or

absent (Polypedates)

2a. Skin shagreened, ground color green above; supratympanic fold yellowish brown

anteriorly P . prasinatus

2b. Skin smooth, ground color light brown or brown, usually with dark spots or stripes

P. megacehalus

lb. Dermal fold along the forearm and tarsus present (Rhacophorus)

3a. Skin smooth, with black spots or blotches on flanks and inner side

of thighs R. moltrechti

3b. Skin shagreened to granulated, with fine dark dots on inner side

of thighs R. taipeianus

These two genera in Taiwan also exhibit

different reproductive modes. Both

Polypedates species usually deposit foamy

masses suspended on low tree branches

overhanging pools or on cistern walls

above the water. Egg masses become light

brown when the outer foamy substance

dries. In contrast, Rhacophorus species

deposit foamy egg masses near puddles in

holes burrowed by the males or beneath

soil or fallen leaves.

Specimens examined: NMNS 01455-

01459, 18 specimens.

Acknowledgments

I thank C. K. Starr, G. F. Wu and S. S.

Lin for helpful comments and suggestions

on the manuscript. I am grateful also to A.

Dubois, editor of Alytes, and S. F. Hung

of the Taiwan Museum for providing dates

of publications relevant to priority.

December 1993

Asiatic Herpetological Research

Vol. 5, p. 13

References

FROST, D. R. 1985. Amphibian species of the

world. Association of Systematics Collections.

Lawrence, Kansas. 732 pp.

JIANG, S., HU, S. AND E. ZHAO. 1987. The

approach of the phylogenetic relationship and

the supraspecific classification of 14 Chinese

species of treefrogs (Rhacophoridae). Acta

Herpetologica Sinica 1987, 6(l):27-42. (In

Chinese).

LIEM, S. S. 1970. The morphology, systematics,

and evolution of the Old World treefrogs

(Rhacophoridae and Hyperoliidae). Fieldiana:

Zoology 57:1-145.

LUE, K. Y. AND Y. P. MOU. 1983. Rhacophorus

smaragdinus (Anura: Rhacophoridae) a new

rhacophorid tree frog from Taiwan. Journal of

Taiwan Museum. 36(2):15-22.

MOU, Y. P., RISCH, J. P. AND K. Y. LUE. 1983.

Rhacophorus prasinatus, a new tree frog from

Taiwan, China (Amphibia, Anura,

Rhacophoridae). Alyles 2(4): 154-162.

Vol. 5, pp. 14-30

Asiatic Herpetological Research

December 1993

Studies on Pakistan Reptiles. Pt. 3. Calotes versicolor

WALTER AUFFENBERG1 AND HAFIZUR REHMAN2

h008 NW 67 th Area. Gainesville, Fl., 32606, USA.

2 Zoology Survey Department, Karachi 1 , Pakistan

Abstract: -Variation in the scutelation and color of Calotes versicolor populations in Pakistan are

analyzed, leading to the recognition of a new subspecies (C v. nigrigularis) from the front ranges of the

Himalayan Mountain complex in Afghanistan, Pakistan and India. Several variant populations of the same

species in other parts of its range are noted, but not given taxonomic recognition at this time.

Key words: Reptilia; Sauria; Lacertilia; Agamidae; Calotes

Introduction

Calotes versicolor , a large, common,

widespread and showy lizard, was

described early in the history of reptilian

study in the Indian subcontinent (Daudin

1802, as Agama versicolor, type loc.

"India", restricted to near Pondicherry,

India by Kuhl 1820). Variation in color

and scalation was also documented early,

resulting in the description of several

species (now synonyms, see Smith 1935

for review) and races. No subspecies are

recognized at the present time, in spite of

obvious geographic variation and a wide

ecologic and geographic range.). The

current study of color and scalation

supports the contention of earlier workers

that morphologically distinct populations

with circumscribed geographic boundaries

exist. The latest morphological study is by

Tiwari and Aurofilio (1990), though it is

restricted to populations from Tamil Nadu,

India.

During the collection of new material for

a future major publication on the

herpetology of Pakistan, populations of

Calotes versicolor from the mountains of

northern Pakistan were noted as being

distinctly different from those in other parts

of the country. This discovery suggested

an analysis of geographic and sexual

variation in several scute and color

characters, similar to our earlier study of

Pakistan Echis carinatus populations

(Auffenberg and Rehman 1991). The

following is the result of this analysis.

Methods

This study is based on slightly more than

500 specimens located in the 14 institutions

listed below. The museum source of those

specimens specifically referred to are

identified by the abbreviations given.

American Museum of Natural History,

New York (AMNH); Natural History

Museum, London (BMNH); Bombay

Natural History Society (BNHS);

California Academy of Sciences, San

Francisco (CAS); Field Museum of Natural

History, Chicago (FMNH); Museum of

Comparative Zoology, Havard University

(MCZ); Pakistan Museum of Natural

History, Islamabad (PMNH); Senckenberg

Museum, Frankfurt (SMF); Florida

Museum of Natural History, University of

Florida (FMNH/UF); University of

Michigan Museum of Zoology, University

of Michigan (UMMZ); National Museum of

Natural History, Washington (USNM);

Zoological Survey Department, Karachi

(ZSD); Zoological Survey of India,

Calcutta (ZSI) and Alexander Koenig

Museum, Bonn (ZFMK).

All drawings were done by the senior

author.

Figure 1 shows the localities from which

specimens were examined. Appendix 1

provides data on museum holdings of

specimens examined from these geographic

locations. No specimens with either

general, questionable, or erroneous locality

December 1993

Asiatic Herpetological Research

Vol. 5, p. 15

FIG. 1. Localities in Pakistan and adjacent

countries from which we examined specimens

(dots). Pakistan localities for which we know

material is available in museums, but which we

have not seen are indicated by circles. The area

represented in Fig. 9 is shown with a rectangle.

data have been included. All data used in

the analyses were obtained by us - none

were drawn from the literature.

The following characters were tabulated

for all the specimens listed above: 1)

number of subdigital laminae under toe IV,

2) number of scale rows at midbody, 3)

snout-vent length (SVL), 4) degree of

mucronation of dorsal scales (0 none or

very weak, 1 moderate mucronation, 2

strong mucronation; Figure 2), 5) angle of

posterior edge of dorsal scale rows (Fig.

3), 6) number of gular scales from just

behind mental to a level even with the

middle of the eye, 7) number of scales in

the dorsal crest that are higher than the

length of their base, 8) color pattern of

chin, 9) color pattern of belly, 10) color

pattern of dorsal body surface, 11) degree

of darkening of postorbital stripe (0 none, 1

moderately dark, 2 very noticeable, see

Figs.9, 12), and sex.

Figure 4 shows the locations and general

size of sample areas chosen. The

geographic limits of the samples were

FIG. 2. Side of body of Calotes versicolor,

showing method of determining the angle of the

dorsal scale rows.

selected principally on the basis of sample

size (museum material available), but in

some cases partly on environmental

differences between closely approximated

geographic areas (i.e., elevation, major

habitat, etc.). Each of these sample areas

served as the basis for all calculations and

evaluations, so that all specimens available

from each area were considered as

constituting the same sample for

computational purposes.

Results

Only one species of Calotes - C .

versicolor - has been identified by us in

Pakistan, though two others were

previously listed or implied as occurring

there. Murray 1886 reports Calotes

grandisquamis (Gunther 1875, a valid

species from southern India) from Sindh

Province (Karachi and Jerruck), Pakistan.

There are no substantiating specimens and

none of the several herpetologists who have

worked in the Karachi area for extensive

periods since have ever found this species.

It is distinctly different from C. versicolor

in it's decidedly green body color, the

significantly lower number of transverse

midbody scale rows (27-35), and in the

presence of a short, oblique fold

(sometimes called a pit) in front of the

shoulder.

Vol. 5, p. 16

Asiatic Herpetological Research

December 1993

FIG. 3. Geographic variation in degree of lateral body scale mucronation. A, Calotes v. versicolor,

FMNH/UF 70516, adult male, Karachi, Karachi Dist., Sindh Prov., Pakistan..

The second species is Calotes jerdoni

(Gunther 1870, type loc. Khasi Hills,

Assam, India. It is represented in the

BMNH by two preserved specimens cited

by Boulenger 1885), said to have been

taken in Afghanistan. This locality

suggests that the species should also be

found in the intervening Pakistan area. The

Afghanistan data are obviously incorrect, as

has already by suggested by Smith (1935).

Calotes jerdoni. We have examined the

specimen in question and confirm it

belongs to this species, which is easily

distinguished from C. versicolor on the

basis of its bright green color and the

parallel rows of enlarged and keeled scales

on top of the head, and in lacking the

characteristics pair (usually) of enlarged

spines above the tympanum.

We believe that Calotes versicolor is the

only species of the genus in Pakistan.

Between populations of this species in and

beyond the borders of Pakistan we are able

to demonstrate significant clinal variation

(north/south, east/west) beyond that

ascribable to race. Such clines occur in at

least four scale characters. Sexual variation

is demonstrated in color and adult size.

Within Pakistan boundaries, geographic

variation suggests the recognition of two

races of Calotes versicolor, one of which is

new. It is described below. Additionally,

the populations found essentially east of

India are distinguished on the basis of color

and scale characters. However, in this

paper we do not recognize them as separate

nomenclatorial entities. The solution to the

question of their status must await the

availability of additional, fresher material.

Individual And Geographic Variation

Here we discuss the variations in color,

proportion, and scutelation which are

correlated with geographic locality, sex,

environment, or ontogeny.

Clinal Variation

The term cline has been used to express

a condition in which the values of a variable

character form a slope or gradient over a

geographic area. With increasing

knowledge of variation systematists have

come to recognize different types of clines.

Some are related to gradual changes in

environment (including climate) and others

are not. The change in character state over

distance have varying slopes when the

values of the characters being examined are

plotted against distance. Two extreme

types of clines are recognizable - narrow

(or steep) and broad (or low) slopes. The

former is represented by a character-

gradient which significantly changes its

slope in a step-like fashion, with separate

subspecies corresponding to two more or

less level character values (flat or slightly

December 1993

Asiatic Herpetological Research

Vol. 5, p. 17

Will

FIG. 4. Location, size, and approximate area of

samples used in this study

sloping) on either side of a uniting steeper

slope - the zone of character intergradation.

The broad cline is one which does not

show any steepening of the character

gradient in a particular place, but is

represented by a continuous slope with no

obvious interruptions. There are, of course,

all gradations of slope between these two

extreme types. Both narrow and broad

clines, as well as intermediate types, exist

within the pattern of character variation in

the species Calotes versicolor.

Number of midbody scale rows (Fig. 5).

— The mean number of midbody scale

rows illustrate the narrow type of clinal

character change in which there is a distinct

and rapid change (= steep slope) from one

rather uniform area to another. In Calotes

versicolor a steep north-south cline exists

between the northern Himalayan Mountain

and Indo-Gangetic Plain populations. The

mean dorsal scale rows of all southern

(plains) populations (India and Pakistan

combined) is 43.0 + 1.0 (OR sample means

41.0-44.1); for all northern (upland)

populations (India, Pakistan, Afghanistan,

Nepal) the mean is 45.9 +1.5 (OR sample

means 43.0-49.2). The difference in mean

number of dorsal scale rows between

northern and southern populations is

FIG. 5. Geographic variation in mean number of

dorsal scale rows of Calotes versicolor samples

studied. Lines (isophenes) enclose samples of

similar value (see text).

highly significant (/ 5.83, df 13, p

<0.001), with the distinct change in slope

of the character gradient occurring along the

frontal hills of the northern and

northwestern mountains of Pakistan.

Similar north-south clines can be

demonstrated within this species in other

characters as well (see below). However,

there is no significant east-west change in

the mean character state value in either the

plains or the mountain populations, in spite

of the fact that other character states do

show important east-west changes (see

below). Thus the east-west axis of the

pattern of character change is independent

of the north-south axis. Within southern,

plains populations, several samples have

graphically different values for this

character than all the surrounding ones

(Quetta is higher, Rajasthan and the

Mekkran Coast are lower). However, in

every case, the samples from these sites are

small and the differences are not statistically

significant.

Subdigital lamellae under 4th Toe (Fig.

6). — In this character the pattern of

geographic variation is more complex.

There is no clear north-south trend. While

Vol. 5, p. 18

Asiatic Herpetological Research

December 1993

FIG. 6. Geographic variation in the mean number

of lamellae under the 4th toe.

equally high values occur in the upland,

northwestern part of the species range, the

same isophene sweeps down to sea level in

the Calcutta area, terminating in the high

value for Chittagong (though the latter is

statistically insignificant). This tongue of

higher values separates the Myanmar-

Malaysian populations in the east from the

Indo-Pakistan ones in the west. Within

Pakistan there is no clear evidence of the

even clinal changes witnessed in the mean

number of dorsal scale rows (the difference

in means between eastern and western

Mekkran populations is based on small

samples and is statistically insignificant).

Thus the overall pattern of character change

is one of an east-west component in which

geographically intermediate populations

(West Bengal) have distinctly higher mean

values than populations to both the east and

west. At the same time, the "isolated"

eastern section includes a population in

Thailand in which the mean value (low) is

statistically distinct from its neighbors (with

Arakan t 3.23, p <0.01; Rangoon t 2.70, p

0.02; Penang t 3.56, p 0.001). Near the

western edge of the species range there is

an apparent north-south trend, in which the

former area (Jalalabad, Peshawar and

Taxila combined) has a statistically

significant higher mean value than the south

FIG. 7. Geographic variation in the mean number

of gular scales.

(Baluchistan) ones (t 3.15 p 0.01). This

pattern is, however, confused by

populations from Azad Kashmir and Swat

District, which are more like one another in

having statistically similar low mean values

than either is to populations from the

geographically intermediate area of

Manshera District.

Number of Gular Scales (Fig. 7). — In

this character there is a distinct two-way,

east- west cline which proceeds from lowest

mean values in peninsular India to higher

values to the west (reaching maximum

values in those samples in the arid

mountains at the eastern edge of the Iranian

Plateau) and the east (highest in Chittagong

and Mandalay). There is no evidence for a

north-south cline anywhere. The

Chittagong sample is again distinctive

(though sample size is small and the values

are not significantly different from

neighboring ones).

Angle of Dorsal Scale Rows (Fig. 8).

— This character illustrates still another

clinal pattern- essentially northwest to

southeast. Highest mean values (111-119)

are found in the former and lowest in the

latter (90-98). There are, however, some

exceptional points outside this general

December 1993

Asiatic Herpetological Research

Vol. 5, p. 19

FIG. 8. Geographic variation in the mean angle of

the dorsal scale rows.

trend, particularly within Pakistan. Thus

the values for the Kirthar, Dadu, Rajasthan

samples are significantly lower than all the

surrounding ones (t values 2.56 to 4.60, p

0.02 to 0.001). The differences between

the the Khuzdar-Quetta samples and

Khuzdar-Panjgur samples are not

significant. In the eastern sector, the

Penang sample has a significantly higher

mean angle of the dorsal scale rows (with

Rangoon t 3.06 p 0.01; with Thailand t

4.20, p <0.001). The Madras sample is

also significantly different from that from

Bombay (r 4.01 p <0.001).

Head and Body Length (SVL). —In a

short discussion of SVL, Smith (1935)

provides data suggesting that the peninsular

Indian populations are larger than those of

the Indo-Chinese region. Our data confirm

this statement, but the larger number of

specimens available to us allow a finer-

grained breakdown of the size and

geographic representation of our sample

areas. We find, first of all, that the

geographic trends in SVL of males and

females parallel one another. Thus,only the

males are discussed here (the females

follow identical patterns). Adult males

from peninsular India and the Indus Valley

are significantly larger than those of all

surrounding samples. The mean SVL for

the (combined) Indus Valley-peninsular

India sample is 94.3 mm. This differs

significantly from the combined sample of

mountainous Pakistan (Student's t 7.25, p

< 0.001, and is significantly different from

a combined sample from mountainous

India-Nepal (mean 82.1, r 3.32, p < 0.01).

The sample from Calcutta has a smaller

SVL, but the difference in means is not

significant atp < 0.05. However, there is a

very significant difference (t 5.6, p <

0.001) between the Calcutta mean (92.2

mm) and that of the combined Myanmar

sample (81.2 mm). Samples from Thailand

and Malaysia have still lower values, but

the means are not significantly different

from that of Myanmar at p < 0.05. The

calculated differences in the means of all of

these samples suggest a broad bi-directional

clinal in which the central part of the

species range has the highest mean values,

with gradually lower ones in all directions,

rather than in only one. Analysis of

additional samples from peninsular India

would undoubtedly clarify the shape of this

cline better than we are able to do on the

basis of our present material.

What is obvious here is that each of the

scale characters analyzed from the

standpoint of geographic variation is

represented by a different pattern of

variation. Thus each of these characters

show a pattern of geographic variation that

is independent of one another. The patterns

undoubtedly reflect the complexity of

selective factors acting through the clinally

changing physical and/or biotic

environments found throughout the species

range.

Sexual Variation

As stated above, within all samples

examined, males attain a greater SVL than

females (though statistically not significant

in one, see Table 1). Overall SVL range of

all mature males is 70-138, mean 99.3 +

17.2, females 64-121, mean 80.5 + 15.7;

Student's t test for difference in means =

5.6, df 264, p < 0.001 . Our analysis also

shows that this dimorphism is

geographically variable (Table 1), with the

Vol. 5, p. 20

Asiatic Herpetological Research

December 1993

TABLE 1. Geographic variation in SVL of adult Calotes versicolor.

Sample Area

O.R

Mean

/Test

Probability

Mts. Pakistan

Males

Females

Penin. India

Males

Females

Myanmar

Males

Females

Thailand

Males

Females

1.93

4.20

2.14

2.05

not sig.

< 0.001

<0.05

0.05

strongest divergence in the peninsular

Indian sample.

In a recent morphological study of Tamil

Nadu, India samples, Tiwari and Aurofilio

(1990) report no sexual difference in

scalation. While we find this to be true of

almost all of the scale characters we

studied, we do find significantly different

means in the number of midbody scale

rows, with females having a higher number

than males by a factor of from 9.7 to 1 1.7

percent, (six of our largest samples were

analyzed; mean in males 41.8 + 9.7 to

46.7 +11.1; females 46.7 + 3.1 to 48.1 +

9.9; t 3.5 to 19.5, p < 0.05 to < 0.01.

Many workers have described the

difference in color and pattern of adult male

and female Calotes versicolor. In general,

the male is lighter, with no, or 4-8 very dim

crossbars on the dorsal part of the body.

The ground color is usually some shade of

tan in the Indian subcontinent and more

grayish in the eastern sectors. Adult

females are darker, the ground color being

brownish to grayish, with the same

number, but more obvious narrow brown

to black crossbars (or remnants of them).

There is often a faint to quite obvious

lighter dorso-lateral stripe on each side of

the body. These are missing in adult

males. Females usually have a series of

circumorbital radiating darker bars which

are usually lacking in males. Females lack

a dark ventral partial collar at the base of the

neck, which is characteristic of the adult

males of some populations. Finally, adult

males have a remarkable change in color

and pattern during the breeding season,

which is absent in the females.

Ontogenetic Variation

Ontogenetic variation is noted in color

and pattern. The belly of juveniles (< 75

mm SVL) possess 5 to 7, dim,

longitudinal, grayish stripes or dashes,

each one scale wide. These disappear with

age, though they are usually represented in

the adult by traces of the median member.

In neonates the chin is white to dirty gray,

laterally streaked in the sub-infralabial area

with a series of medium gray to black

diagonal stripes, which become dimmer

with age. In one-third- to one-half- grown

individuals the chin is additionally suffused

with light pink. A series of gray to black

December 1993

Asiatic Herpetological Research

Vol. 5, p. 21

J ' • • i • a '• • ; •

\

71'

74°

FIG. 9. The geographic distribution of Calotes versicolor nigrigularis (solid dots) is restricted to elevations

between 300 and 1800 m. The stippled zone is < 300 m, and the cross-hatched one is > 1800 m.

Localities of specimens morphologically intermediate between this race and Calotes v. versicolor arc

indicated as half-circles..

radiating circumorbital bars are almost

always evident, which also fade with age.

Taxonomioc Considerations

locality Pondicherry, India.

Calotes versicolor (Daudin), Jerdon

1853:470.

This report recognizes a subspecies if 75

percent of the available individuals from a

geographic area can be correctly assigned to

that provenance on the basis of one or more

characters. Analysis of the data suggests

that Calotes versicolor is divisible into at

least two subspecies. Geographic

discontinuities in the diagnostic characters

are the basis for the generalized racial

distributions shown in Figure 9.

Calotes versicolor versicolor (Daudin)

Agama versicolor Daudin 1802:395.

Type locality "India".

Agama tiedmanni Kuhl 1820:109. Type

? Calotes viridis Gray 1846:648. Type

locality Madras. (Type specimen lost).

A subspecies of C. versicolor distributed

from Sri Lanka north through most of

peninsular India and Pakistan, west to the

Kabul Valley in southeastern Afghanistan,

northeast to Hainan Island, China and

southeast to Sumatra, Indonesia; replaced

in the northern mountains of Pakistan and

adjacent Afghanistan and India by C. v.

nigrigularis (nov. ssp, described below).

Other undescribed races probably replace

this plains form in northeastern India, and

lowland areas of Myanmar, Thailand,

Malaysia and Sumatra.

Vol. 5, p. 22

Asiatic Herpetological Research

December 1993

- -

' ■ ■

v^v

--r ■ - v~\ , ,

FIG. 10. Chin and throat color pattern in adult Calotes v. versicolor. A, FMNH/UF 7051 1, adult male,

Karachi, Karachi Dist., Sindh Prov., Pakistan. B, FMNH/UF 19952, adult male, 8 mi. W. Madras, Tamil

Nadu State, India.

C. v. vesicolor has the following suite of

characters which distinguishes it from C. v.

nigrigulariss: larger adult size, mean angle

of dorsal scale rows 90 - 105 , mean

number two distinct postocular stripes

(absent in largest males, Fig. 10A), dorsal

body pattern usually indistinct, tending to

uniform tan during most of the year,

becoming pink to reddish in males during

the breeding season; 5-7 crossbars may be

present (particularly in juveniles and adult

females), each 1-2 scales long at the

vertebral line; gulars either uniformly light-

colored or marked with narrow, diagonal,

faint, dusky or sometimes black stripes;

scales of the throat and pre-shoulder areas

vary from the same color as the gulars to

having dark brown or black bases. Large

males from southern India often have a

ventrally located black, partial collar (Fig.

11); the belly is always uniformly

yellowish- to dirty-white.

Several months before the breeding

season, the color and pattern of adult males

change. At this time the lateral and dorsal

surfaces of the head, neck, and shoulders,

and the sides of the body all become

suffused with yellow, pink, orange, or

even dull red (depending on geographic

location, and age of the individual). The

throat and chest change to orange or red

with black mottling (seasonal adult color

changes in Indian subcontinent populations

also described by Murray 1886, Smith

1935, and Minton 1966); the tail and limbs

become black.

Holotype — Presumably in the Paris

Museum, but now lost. The type locality

had been simply stated as "India", but this

was later restricted to Pondicherry, India by

Kuhl (1820).

Exemplary Material Examined. — The

specimen materials examined by us that

best fit the type description are listed

below. We do not include any specimens

here from the eastern Himalaya Mountains

and from West Bengal eastward, as we

believe that those populations will

eventually be recognized as representing

one or more races distinct from the

nominate form and that one described

below: FMNH/UF 19886, 19955-9,

Kanheri Caves, nr. Borivli, Maharashtra

State, India; FMNH/UF 70535-7, Khadiji

Falls, Dadu Dist., Sindh Prov., Pakistan;

FMNH/UF 19949-53, 8 mi. W. Madras,

Tamil Nadu State, India;FMNH/UF 79087,

79099, Sujabad, Deri Ghazi Khan Dist.,

Punjab Prov., Pakistan; FMNH/UF 78932,

Sonmiani, Las Bela Dist., Baluchistan

Prov., Pakistan; AMNH 39377-8, 5 mi. E.

Kalka and AMNH 39382, nr. Kalka,

Amballa Dist., Punjab State, India; CAS

94337-8, 3 mi. SE Sirohi, Rajasthan State,

India; MCZ 55502-3, Baroda, Gujarat

December 1993

Asiatic Herpetological Research

Vol. 5, p. 23

B

FIG. 11. Chin and throat color pattern in Calotes v. nigrigularis. A, FMNH/UF 79470, adult male, nr.

Chergal, Manshera Dist., NWFP, Pakistan. B, FMNH/UF, juvenile male, Miandam, Swat Dist., NWFP,

Pakistan. C, FMNH/UF 70503, adult female, Abbottabad, Abbottabad Dist., NWFP, Pakistan.

State, India; FMNH/UF 19884-5, New

Delhi, India; FMNH/UF 78420, Multan,

Multan Dist., Punjab Prov., Pakistan; ZSI

20798, 20800, Pali, Pali Dist., Rajasthan

State, India; ZSI 1383, 13486, Ajmer,

Rajasthan State, India; ZSI 20796,

Jodhpur, Rajasthan State, India; BNHS

325-6, Wanothi, Kutch Dist., Gujarat

State, India; BNHS 318 Bhavnagar, Rajkot

Dist., Gujarat State, India; SMF 70074

Amritsar, Punjab State, India; SMF 61925,

Bangalore, Karnataka State, India; SMF

55444, Meerut, Uttar Pradesh State, India;

UMMZ 172083-90, Bhubaneswar, Orissa

State, India; and BMNH 1923-3-445,

Mirpur Sakro, Thatta Dist., Sindh State,

Pakistan.

Scute lation Characteristics. — Overall

ranges and means for the scale characters of

the samples of Calotes v. versicolor

examined are as follows (Details regarding

geographic variation in these parameters are

found in the diagnosis above, in Figures 5-

8, and in the text discussions regarding

them): Upper head scales unequal, smooth

to feebly keeled; two well -separated spines

on each side of the back of the head above

the tympanum; canthus and superciliary

ridge sharp; 11-13 (mean 12.0) infra- and

11-15 (mean 12.3) supralabials; dorsal

scales large, distinctly keeled, all pointing

backwards and upwards, larger than the

ventral scales, which are always strongly

keeled and mucronate, in 35-52 scale rows

at midbody (mean 44.5); subdigital laminae

of 4th toe 20 - 27 (mean 23.1); gular scales

behind mental to middle of eye 7 to 15

(mean 11.2).

SVL and Color Variation. — Juveniles

have a dorsal color partem like that of adult

females, except that the ground color is

grayish, rather than the usual brownish,

Vol. 5, p. 24

Asiatic Herpetological Research

December 1993

and the dorso-lateral stripes are usually

dirty white, rather than yellow. The chin

and throat are the same color and pattern as

in adult females, except that the lateral black

diagonal stripes are usually better defined

(Fig. 11).

Nomenclature. — The type localities of

both C. vultuosa Harlan (1825) and C.

gigas Blyth (1853) (synonyms of Calotes

versicolor) are Calcutta, West Bengal,

India. Our studies show that C. versicolor

from West Bengal exhibit a high level of

character variation. They are excluded

from our synonomy of C. v. versicolor on

the basis that we cannot confidently place

them in any named valid race at the present

time, as they are intermediate between

surrounding populations in many respects.

Murray (1886) reported Calotes viridis

Gray (1846) from Upper Sindh,

Baluchistan, Punjab, southern India and the

Deccan Plateau (this reference not included

in C. versicolor synonomy given by Smith

1935). Murray's identification of some

material from southern Pakistan as C.

viridis is clearly incorrect, for no specimens

referable to this name have been found

there by any of the several thorough

herpetologists who have worked

extensively in Sindh Province since that

time. The species has been considered a

probable synonym of C. versicolor; the

species type locality is Madras; the type

specimen is lost.

Because the original type locality of

Calotes versicolor was imprecise,

("India"), Kuhl (1820) re-designated it as

Pondicherry, India. It then follows that the

peninsular Indian and Indus Valley (sensu

latu) populations are to be given the name

Calotes versicolor versicolor. In addition,

we note an apparently distinct population of

C. versicolor which occurs from Thailand,

Myanmar, Assam, Sikkim, Darjeeling, and

Nepal. However, we believe that

taxonomic recognition of this population is

not currently warranted until sample sizes

are increased and fresher material becomes

available for study. Specimens from this

area differ from C. v. versicolor in having

more mucronate dorsal scales, a lower

number of scales under the fourth toe, a

higher number of gular scales, wider body

crossbands, in having many adults and

subadults (in addition to in juveniles) with

dusky longitudinal stripes on the belly, and

the adults having a smaller SVL.

Specimens we recognize as intermediate

between these populations and those typical

of Calotes v. versicolor occur in parts of

Nepal and West Bengal.

Geographic and Vertical Range. —

Calotes v versicolor does not occur above

2000 m in the Indian Ghats (this study).

Populations from the Himalaya Mountains

(racially not yet defined), from Garwhal,

India east through Sikkim and Bhutan are

found to 2500 m elevation (this study), but

to only 1030 m in Indochina (Smith 1935).

Calotes v. versicolor (sensu latu) is

distributed from the drier, more open

forests of Sumatra and the Malay Peninsula

north to near Hong Kong and Hainan, west

through the mainland to southeastern

Afghanistan, and eastern Iran, including the

Andaman Islands and Sri Lanka.

Additional study will undoubtedly lead to

the recognition of additional races in the

eastern parts of the range as here defined.

If so, the nominate form C. v. versicolor

will undoubtedly become restricted to those

populations living in the lowlands of the

Indo-Pakistan subcontinent.

Calotes versicolor nigrigularis ssp. nov.

Holotype.— FMNH/UF 79470 (Figs.

11, 12), adult male, on shrub on rocky

hillside, Shargal, 20 km S Balakot,

Manshera Dist., Northwest Frontier

Province (lat. 34.3° N, long. 73.4° E),

Pakistan. Pakistan Museum Natural

Science field crew, June 15, 1990.

Paratypes (N 16, all from Pakistan). —

AZAD KASHMIR PROVINCE:

FMNH/UF 79049, Gulpur; Kotli Dist.,

FMNH/UF 79396, 81165, Red Fort,

Muzaffarabad, Muzaffarabad Dist.;

FMNH/UF 79472, Chalpani,

Muzaffarabad Dist.; FMNH/UF 79494,

Panyola, Poonch Dist.; FMNH/UF 79495,

December 1993

Asiatic Herpetological Research

Vol. 5, p. 25

g£^K£ tk*

U. n

B

FIG. 12. Side views of head of adult Calotes versicolor. A, C. v. nigrigularis, FMNH/U 79470, adult

male, Chergal, Manshera Dist., NWFP, Pakistan.. B, Calotes v. versicolor, FMNH/UF 78926, adult

male, Mach, Quetta Dist., Baluchistan, Pakistan. C, C. v. versicolor, FMNH/UF 70516, adult male,

Karachi, Karachi Dist., Sindh Prov., Pakistan.

Seri, Muzaffarabad Dist.; FMNH/UF

79601, Chela, Muzaffarabad Dist.;

NORTHWEST FRONTIER PROVINCE:

FMNH/UF 78944, Charsadda, Mandan

Dist.; FMNH/UF 79229, Bahrain, Swat

Dist.; FMNH/UF 79326, Miandam, Swat

Dist.; FMNH/UF 79471 Khakai, Manshera

Dist.; FMNH/UF 81133, 11 km W.

Hungu, Togh Serai, Kohat Dist.;

FMNH/UF 82243, Temargarh, Dir Dist.;

FMNH/UF 70503, Abbottabad,

Abbottabad Dist.; FMNH/UF 79462

Dharial, 4 km SW Balakot. PUNJAB

PROVINCE: FMNH/UF 81136,

Company Bagh, N. Tret, Rawalpindi Dist.;

FMNH/UF 82242, 1.9 km SE Kohala,

Rawalpindi Dist.

Other exemplary material (all juveniles,

or in poor condition): AZAD KASHMIR:

FMNH/UF 82244, Pottri, nr. Bhimber,

Mirpur Dist.; PUNJAB PROVINCE: ZSD

1231, Ghora Gali, Rawalpindi Dist.;

FMNH/UF 82656, 1 1 km S Kohala,

Rawalpindi Dist.; NORTHWEST

FRONTIER PROV.: FMNH/UF 79138,

Miandam, Swat Dist.; BNHS 313, Drosh,

Chitral Dist.; FMNH/UF 82821, 1.1 km

SW Garh Habibullah, Manshera Dist.;

FMNH/UF 82822, 0.6 km SW Garh

Habibullah, Manshera Dist.; BNHS 341,

Parachinar, Kuram Dist.; FMNH/UF

81218, 13 km NE Abbottabad, Manshera

Dist.; FMNH/UF 81093, 2 km W Hungu,

Kohat Dist.; and FMNH/UF 82077, 5.8

km NW Khaki, Manshera Dist.

Diagnosis. — Conspecific with Calotes

versicolor on the basis of it's short head,

the scales on the sides of the body pointing

upwards and backwards, and that it lacks a

fold or pit in front of the shoulder. It

differs from the nominate race in having

more strongly keeled (and usually more

mucronate) body scales, more transverse

scale rows at midbody, generally more

Vol. 5, p. 26

Asiatic Herpetological Research

December 1993

median gular scales from the tip of the jaw

behind the mental to a level perpendicular to

the middle of the eyes, fewer enlarged

vertebral scales composing the in the

nucho-dorsal crest, and in the adult state it

lacks dark postocular stripes. During the

breeding season the skin over the posterior

part of the lower jaw in adult males (only)

is jet black, except for a longitudinal

median ventral band, which varies with

season from pink to scarlet. The most

vivid red color is found in the largest males

during July and August. Each black gular

patch extends (during the breeding season)

posteriorly along the side of the neck,

thence dorso- posteriorly at an upward angle

to the vertebral line, including the entire

shoulder region (Figs. 10, 12A). At the

same time the entire head and dorsal neck

surface are pinkish-red. In some

individuals, both the red and black pattern

may disappear at death.

Adult males of this race lack the greatly

swollen jaw muscle mass of the nominate

form, resulting in a head that in top view

has more parallel posterior borders behind

the eyes than that of C. v. versicolor

(where these edges are clearly divergent).

In many individuals the toes are shorter

than in those of the nominate population

and the brachium is usually as long as the

antibrachium; in C. v. versicolor the

antibrachium is often shorter.

Description of the Holotype. — Length of

the head 1.48 times its width; snout broad,

a little longer than the orbit; top of head

from side slightly convex, slightly concave

from the front; upper head scales unequal,

smooth, to faintly keeled or tuberculate;

canthus rostralis and supraciliary ridge

sharp; two thin, spinous scales above the

tympanum, the anterior one smallest,

separated from the tympanum by about 5

scale rows; 13 supra- and infralabials; body

somewhat compressed laterally; dorsal

scales medium in size, distinctly keeled,

most being mucronate, pointing backwards

and upwards, larger than the ventrals,

which are more strongly keeled and

mucronate; 51 scales round the middle of

the body. No gular pouch; gular scales like

those of the ventrals, but larger. Nuchal

and dorsal crests developed, composed

anteriorly of lanceolate spines, gradually

decreasing in size to the base of the tail.

Limbs moderate; fingers 3 and 4 almost

equal in length; toe 4 longer than 3. Tail

rounded, covered with more or less equal-

sized, strongly keeled, mucronate scales.

The measurements (in mm) are as

follows: total length 339; SVL 94; tail

length 245; body length (axilla-groin) 47;

greatest head length (snout tip to posterior

extent of lower jaw) 34; greatest head width

(across most posterior part of lower jaw)

23; greatest head height (just behind

posterior edge of eye) 18.5; height of ear

opening 3.1; length of brachium (axilla to

elbow) 15; length of antibrachium (elbow

to wrist) 13.6; posterior limb when

extending anteriorly nearly reaches

posterior edge of eye.

The dorsal ground color is more or less

grayish-tan (pinkish-tan in life) over the

posterior 2/3 of the body. From the level

of the posteriorly extended elbows to and

onto the base of the tail. The body is very

indistinctly marked with 3 slightly darker

cross bars. Anteriorly it is almost

completely black, being lightest along the

vertebral line. Ventrally it is dark gray

from near a line connecting the anterior

edges of the shoulder posteriorly to just

before the insertion of the hind limbs,

where the color changes abruptly to grayish

cream. The hind limbs and tail base are

more or less uniform above and below,

matching the colors of adjacent body

surfaces. On the dorsal caudal surface,

from about the level of the posteriorly

adpressed knee to slightly beyond the claw

tips of the hind foot, faint darker cross bars

can be discerned, fading posteriorly as the

tail becomes suffused with dark gray from

its middle to the tip. The front limbs from

shoulder to claw tips are uniform grayish

black. The sides of the neck are black,

continuing anteriorly to the black color of

the limbs and the sides of the body. The

most intense black on the entire individual

occurs from the anterior lateral surface of

the neck anteriorly onto the jaws and gular

region. This black jaw marking is

distinctly set off from the lighter color of

December 1993

Asiatic Herpetological Research

Vol. 5, p. 27

FIG. 13. General body color pattern of Cables v. nigrigularis; A, adult female; B, adult male in breeding

coloration.

the head and midgular areas, forming a

more or less arrow-shaped black patch on

each side (Fig. 10, 12A). Between the two

black jaw patches is a dark pink (scarlet in

life) median longitudinal band, beginning in

the postmental region and extending onto

the gular fold. The top of the head is

medium gray, slightly mottled with

grayish-tan. Laterally and posteriorly this

fades into a pinkish-tan which covers all of

the temporal areas and extends posteriorly

in a V-shaped mark to above the shoulders.

The eyelids above and below are light gray,

with a nearly black spot in the anterior

corner and a larger one in the posterior

comer. Both the supra- and infralabials are

light grayish-tan with faint grayish stripes

radiating from the orbit. The lightest part

of the body is in the area of the mental and

surrounding shields. The most striking

part of the entire color scheme is the black

and scarlet gular pattern.

Sex and Color Variation.. — Adult

females lack distinct metachroic color

changes (Fig. 13). The dominant dorsal

color is gray to grayish-brown, usually

with a narrow white to yellow dorsolateral

stripe (sometimes represented by a series of

dashes) from the neck to above the hind

limb insertion. All these markings are

variable in intensity and completeness.

Along the vertebral area between the light

stripes are 5 to 6 darker brown to black

blotches or crossbands, 4 to 5 scales long,

which in the largest females fade into the

ground color. The ground color of the

gular region is white or pale gray to pink

(latter during the breeding season only).

There are no large jet-black gular patches as

found in adult males, though the base of the

scales in this area may be dark gray (Fig.

11). Frequently the gular area is also

marked with 5 to 7 more or less distinct

black lines or dashes running postero-

medially from the infralabials toward the

midline.

The reddish throat of adult males is first

evident in individuals about 50 mm SVL

(FMNH/UF 82656), i.e., at the end of the

first year of life. The ventral surface of

neonates of both sexes (mean ca 37 mm

SVL), through nearly the entire first year is

uniform dirty white. The smallest male

with well defined black jowls with a red

median area has a SVL of 76 mm

(FMNH/UF 82623).

Distribution. — This subspecies is

restricted to the foothills and outliers of the

Himalaya Mountains, from the Jhellum and

Neelam River Valleys of Azad Kashmir,

Pakistan, west to the Hindu Kush

Mountains and foothills bordering the

Kabul River Valley in southeastern

Afghanistan, south to include the Safed

Koh Range on the Pakistan-Afghanistan

border (Fig. 1). It may extend further

south to Waziristan or even Quetta, but this

will only be proven with fresh material (see

below). Within this area the race is

apparently restricted to subtropical chir pine

(Pinus roxburghii) and oak (Quercus

Vol. 5, p. 28

Asiatic Herpetological Research

December 1993

incana) forests, which are found at

elevations between 1000 and 3000 m,

depending on exposure and slope

conditions.

Individual character states of the plains

race C. v. versicolor extend into the

foothills along several of the larger rivers.

Such changes coincide with the general

floral change from the plains into the

foothills. The Indus Valley (which further

specimens may prove completely divides

C. v. nigrigularis into isolated eastern and

western populations) is an example (see

remarks of intergradation under C. v.

versicolor). Such intergrade populations

nearly bisect the mountain range of C. v.

nigrigularis in the Kabul River Valley near

Peshawar (specimens in ZSDP, uncat.).

We have not yet found any intermediate

populations in the valleys of the Neelam or

Jhellum Rivers. However, intergrades do

occur near the foot of the Himalayan front

range (Taxila and Islamabad, PMNH

uncat.). The fact that specimens somewhat

intermediate between the two races have

been found as far south as Khuzdar,

Baluchistan (BMNH H 1964-276-8)

suggests that C. v. nigrigularis may

eventually be found throughout the Quetta

area and the Central Brahui Range as well,

though no fresh specimens are available for

study at this time.

The only other intensive study of

geographic variation of character states in

Pakistan with which these results can be

compared is our earlier study of Echis

carinatus (Auffenberg and Rehman 1991).

Like Calotes versicolor, this species is

found over virtually all of Pakistan except

the higher mountains. Calotes versicolor

does not, however, occur in the sandy

deserts of northwest Pakistan.

In the Echis study we analyzed 12

characters. While each of these

demonstrate a unique pattern of geographic

variation, several features common to most

of them stand out. These comprise what

we believe to be five major centers of

adaptive speciation in Echis carinatus -

Transcaspia, Iranian Plateau, Astola Island,

Indo-Gangetic Plain, Himalayan foothills,

and the Cholistan-Thar Desert. Of these,

the first three are essentially extralimital

from the standpoint of the current study.

The remaining three areas (Himalayan

foothills, Indo-Gangetic Plain and

Cholistan-Thar Desert) are also

recognizable on the basis of distinctive

character states, or combination thereof, in

the characteristics of Calotes versicolor

populations studied in this report. Thus,

the Himalaya foothills populations are

distinguishable from those of the Indo-

Gangetic Plain and Cholistan-Thar Desert

on the basis of several significantly

different scale characters as well as a

strikingly different metachroic color and

pattern change in the adult males during the

breeding season. Likewise, the Desert

populations are different at statistically

significant levels from those of all the

surrounding Indo-Gangetic Plains

populations in regard to certain scale

characters. Though the the recommended

nomenclatorial designation for these Echis

and Calotes populations is different in each

case, the correspondence of similar

geographic patterns of variation is certainly

important from the standpoints of both

zoogeography and speciation in the

subcontinent.

We have not found any evidence for the

curious mosaic of mean character states

found in the Indus Delta region, as we did

for Echis carinatus. The reason may be

related to the fact that Calotes versicolor is

often found in riverain forests, so that river

and channel changes may be less important

as an isolating mechanism in this species

than in Echis carinatus.

Acknowledgments

We particularly thank the United States

Fish and Wildlife Service (Washington),

the Deutscher Akademischer

Austauschdienst (Bonn, Germany), and the

Office of Sponsored Research , University

of Florida for providing funds to conduct

this study. To all curators and collection

managers of the institutions listed above,

we extend our sincere thanks for the many

ways in which they have contributed to the

success of this project. Finally we wish to

December 1993

Asiatic Herpetological Research

Vol. 5, p. 29

acknowledge the support offered by our

respective institutions.

Literature Cited

AUFFENBERG, W. AND H. REHMAN.

1991. Studies on Pakistan reptiles. Pt. 1.

The genus Echis (Viperidae). Bulletin of the

Florida Museum, Natural History 35(5):263-

314.

BOULENGER, G. A. 1885. Catalog of the

Lizards in the British Museum of Natural

History. British Museum, London. 497 pp.

KUHL, H. 1820. Beitrage zur Zoologei und

vergleichenden Anatomic Frankfurt a. Mein.

1 14 pp.

MURRAY, J. A. 1886. The Reptiles of Sind;

A Systematic Account. Richardson and Co,

London. 92 pp.

SMITH, M. A. 1935. Fauna of British India,

Vol. 2, Reptilia and Amphibia. Taylor and

Francis, London. 440 pp.

TIWARI, M. AND AUROFILIO (sic). 1990.

Biology of the Indian garden lizard, Calotes

versicolor (Daudin). Part I: Morphometries.

Hamadryad 15(l):30-33.

APPENDIX 1

Localities (to district only) from which specimens

were examined, the museum collections in which

they are found, and the number studied (in

parentheses).

AFGHANISTAN: BMNH (3); Jalalabad CAS (3).

BANGLADESH: Chittagong MCZ (1).

INDIA: Assam State: Chabus AMNH (1),

Goalpara Dist., Raimona FMNH (3); Behar

State: Benares BMNH (8), Patna BMNH (1);

Gujarat State: Baroda MCZ (2); Bhaunagar

BNHS (1), Hingolgadh BNHS (2) Rajkot BNHS

(1), Kutch BNHS (2); Himachal Pradesh

State: Amballa MCZ (2); Kulu Valley MCZ (2);

Jammu-Kashmir State: Jammu BMNH (1);

Karnataka State: Bangalore SMF (1);

Maharashtra State: AMNH (1), Bombay

FMNH/UF (7); Orissa State: Bhubaneswar

UMMZ (13); Punjab State: BMNH (1),

Amritsar SMF (1), Amballa Dist. nr. Kalka

AMNH (3); Rajasthan State: Ajmer ZSI (2),

Bikaner ZSD (1), Pali ZSI (1), Jodhpur BNHS (1)

CAS (2), SMF (1), ZSI (3) Ml Abu/Abu Rd.

CAS (1), AMNH (1); Nagaur ZSI (2), Jaipur ZSI

(4); Tamil Nadu State: Madurai FMNH/UF (7);

Uttar Pradesh State: nr, Chalu BMNH (1),

Delhi FMNH/UF (2), Kanpur AMNH (1);

Pitharagah (Kumaon) BMNH (1), Meerut SMF (1),

Mussoorie ZSI (1); West Bengal State:

Calcutta UMMZ (1), FMNH (2), MCZ (6),

Darjeeling MCZ (1), Kalimpong Dist. Tarkhala

MCZ (1).

HONGKONG: BMNH (3).

MY ANMAR ("Burma"): Rangoon FMNH/UF

(27); Arakan FMNH/UF (4); Mandalay FMNH/UF

(2); "at Chinese border" BMNH (1).

MALAYSIA: Penang State: Penang

FMNH/UF (8).

NEPAL: BMNH (10), Katmandu SMF (1),

Swayabonath SMF (1), Lapha Kamali Valley

BMNH (1), Rasna Dist. BMNH (1), Maewa-Khola

BMNH(12).

PAKISTAN: Azad Kashmir Prov.:

Muzzafarabad Dist. SMF (1), FMNH/UF (5); Kotli

Dist. FMNH/UF (2); Poonch Dist. FMNH/UF (1);

Mirpur Dist. FMNH/UF (2); Baluchistan

Prov.: Kalat Dist., AMNH (1), BMNH (2), ZSI

(1); Las Bela Dist. AMNH (4), FMNH/UF (2);

Panjgur Dist., MCZ (1); Quetta Dist. FMNH/UF

(1); Khuzdar Dist. BMNH (3); Waziristan Dist.

BNHS (1); Northwest Frontier: Chitral Dist.

BNHS (1); Dir Dist. FMNH/UF (1); Abbottabad

Dist. ZSD (1), FMNH/UF (1); Kohat Dist.

FMNH/UF (1); Kuram Dist. BNHS (1); Manshera

Dist. ZSD (2), FMNH/UF (7);Peshawar Dist.

BNHS (1), ZSDP (3), Swat Dist. FMNH/UF 6);

Punjab Prov.: Dera Ghazi Khan Dist.

FMNH/UF (2); Lahore Dist. ZSI (1); Kohat Dist.

FMNH/UF (1); Multan Dist. ZSDM (4),

FMNH/UF (1); Chakwal Dist. ZSI (2); Bahawalpur

Dist. ZSDM (3); Rawalpindi Dist. ZSD (1), CAS

(1), FMNH/UF (3); Sindh Prov.: FMNH (1);

Badin Dist. ZSD (1); Dadu Dist. AMNH (1), ZSD

(1), FMNH/UF (4); Hyderabad Dist. AMNH (7),

BMNH (1), ZSD (1); Karachi Dist. AMNH (4),

BMNH (4), CAS (7), FMNH (4), UMMZ (4),

SDSNH (1), FMNH/UF (91), ZSD (30), ZSI (2);

Thatta Dist. AMNH (7), UMMZ (1), ZSD (6),

FMNH/UF (2); Thar Parkar Dist. BNHS (1), ZSD

(2).

REPUBLIC OF THE MALDIVE ISLANDS:

Addu Atoll BMNH (2), Baras Isl. BMNH (1);

Hulalay Isl. Bmnh (3); RAF BAse BMNH (1).

Vol. 5, p. 30 Asiatic Herpetological Research December 1993

SIKKIM: BMNH (1); Mangpu FMNH (34),

Teesta Valley MCZ (2).

THAILAND: Bangkok Prov.: FMNH/UF (2);

Chiang Mae Prov.: FMNH/UF (5); Mae

Hong Son Prov.: FMNH/UF (1); Udon

Thani Prov.: FMNH/UF (1); Yala Prov.:

FMNH/UF (1); Phattabung Prov.: FMNH/UF

(1).

I December 1993

Asiatic Herpetological Research

Vol. 5, pp. 3 1-44 J

Holocene anurans from Caucasus

ZBYNEK rocek

Department of Paleontology, Geological Institute, Czechoslovak Academy of Sciences, Rozvojova 135, CS-165

00 Prague, Czech Republic

Abstract. -Holocene deposits of the Kudaro I Cave from the vicinity of Ertso Lake (South Ossetia, NW

Caucasus) yielded, among others, rather numerous disarticulated anuran bones. Examination of this sample

revealed that majority of this material belongs to the genus Bufo and to the family Ranidae. This generally

corresponds to the composition of the contemporary anuran fauna of that region.

Key Words: Anura, Holocene, Caucasus, osteology.

Introduction

The material described in the present

paper was recovered from the deposits of

the cave Kudaro I. The cave is on the slope

of Mt. Chasavalskaya (1600 m altitude),

valley of Dzhordzhori River, in the vicinity

of Ertso Lake, about 90 km NE from the

town Kutaisi, South Ossetia, near

Rachinsky ridge in the NW Caucasus

(approx. 42° N, 43° W; see also Lyubin

1980a). The deposits are of the Holocene

age (Lyubin 1980b).

First description of the anuran and reptile

material from this cave was published by

Darevsky (1980). His taxonomic

assignments generally agree with those in

the present paper. The material consists of

isolated bones; anuran bones described in

this paper are deposited in the Zoological

Institute of the USSR Academy of

Sciences, St. Petersburg, under collection

numbers ZIL/EL/1 - ZIL/EL/185.

Anatomical terminology mostly follows

that of Bolkay (1919) and Gaupp (1896).

It should be noted that most of elements in

the sample are postcranial bones and only

few cranial bones are present. Fragmentary

material bearing no diagnostic characters

was excluded from the account below.

Systematic Paleontology

Bufo sp.

Material: Left scapula, ZIL/EL/7 (Fig.

7B). Probably also ZIL/EL/1 40 (Fig. 7 A).

Description: The margin of the cavitas

glenoidalis is elevated. Although both the

pars acromialis and proc. glenoidalis are

broken off it is obvious that there was a

deep incision between them. There is a

moderately prominent and rather pointed

outgrowth on the margo anterior. Both

scapulae are comparatively big elements

corresponding by their size to the below

described humeri and ilia.

Material: Humeri ZIL/EL/21 (Fig. 2B),

ZIL/EL/25, ZIL/EL/55,

ZIL/EL/79, ZIL/EL/1 12 (Fig. 2A),

ZIL/EL/132 (Fig. 2C), ZIL/EL/139,

ZIL/EL/153, ZIL/EL/173.

These specimens (except for ZIL/EL/153

that includes also the most distal section of

the crista ventralis) are preserved only as

distal parts of the humerus. All of them are

characteristic by conspicuous degree of

development of the epicondylus medialis,

the distal margin of which extends almost

to the level of the distal margin of the caput

humeri. Hence, there is a distinct notch

between the both structures that can

continue onto the dorsal surface of the

distal section of the bone. On the ventral

surface of the medial epicondylus one can

discern an indistinct ridge running onto its

distal surface. The lateral epicondylus is

developed in lesser degree, extending

laterally from the outline marked by the

crista lateralis in some specimens (see Fig.

1993 by Asiatic Herpetological Research

Vol. 5, p. 32

Asiatic Herpetological Research

December 1993

2A, B). The lateral surface of the caput

humeri is worn away in larger specimens,

so the ball is not complete. The whole

distal end of the bone is bent ventrally; this

is correspondingly reflected on its dorsal

surface which is convex along its

longitudinal axis. Some variation may be

observed concerning the extent of the

medial and lateral cristae which might be,

however, assigned to secondary sex

differences. This might be suggested also

by small specimen ZIL/EL/55 which may

represent an immature individual, and in

which both cristae are lacking. On the

other hand, all specimens have their crista

medialis directed laterally, so its ventral

surface meets the medial surface of the

diaphysis almost perpendicularly. In

ZIL/EL/25 and ZIL/EL/132 the margins of

both cristae are rather undulated and

thickened.

Discussion: These humeri (preserved

only as distal sections) are morphologically

closest to those of Bufo. In large with the

determination of size-corresponding ilia.

However, some (esp. smaller) humeri may

fall into the variation range of the ranids,

but the latter assignment lacks reliable

foundation if only distal part of the bone is

at the disposal. Discoglossids may be

excluded because their medial and lateral

cristae are confluent with the diaphysis,

with no distinct border. The only fossil

anuran that is of similar size as ZIL/EL/1 12

is Latonia seyfredi v. Mayer (=

Discoglossus giganteus Wettstein-

Westerheimb). However, its morphology

and stratigraphic range are different (see

e.g. Mlynarski, 1976, pi. 1/4).

Material: Radioulnae ZIL/EL/43,

ZIL/EL/62 (Fig. 8 A), ZIL/EL/1 06 (Fig.

8B), ZIL/EL/1 16, ZIL/EL/120 (Fig. 8C),

ZIL/EL/1 62.

Description: The margin of the

olecranon rimming the articular cavity is

formed either by calcified cartilage or

ossified tissue lacking periostal bone. The

border between the smooth periostal bone

and rough surface rimming (and also

covering) the articular cavity is distinctive.

It seems that this most proximal part of the

olecranon may be abraded in large

specimens (e.g., ZIL/EL/62).

On the inner surface of the bone, close to

the point where the margins of the

articulation cavity of the olecranon and

capitulum meet with one another, is a small

but deep pit. Similar pit is lacking or not so

deep in ranids, but regularly present in

Bufo. It serves as a muscle insertion area

and in some specimens may be doubled.

The posterior margin of the bone (i.e., of

its ulnar part) bears an indistinct crista in

some specimens.

Material: Ilia ZIL/EL/41, ZIL/EL/66,

ZIL/EL/75, ZIL/EL/94 (Fig. 4A),

ZIL/EL/97, ZIL/EL/99 (Fig. 4B),

ZIL/EL/1 24, ZIL/EL/1 29.

Description: The ala ossis ilii, if

compared with the posterior part of the

bone, are stout (ZIL/EL/94) or rather

slender (ZIL/EL/99). Their dorsal margin

is rounded, only in the mid-part it becomes

an edge distinctly bent medially along its

whole extent. In its anterior part, the ala is

compressed dorsoventrally, ellipsoid in

cross-section. On the medial surface of the

ala, approximately at the level of the highest

point of their arch, there is an orifice of the

narrow horizontal canal coming onto the

bone surface from the posterior. There is

certain variation in the location and

morphology of this canal - it may continue

as a groove for a short distance

anteriorwards, and the orifice may be

located on the bottom of a horizontal

depression developed below the above-

mentioned edge. The depression may

terminate anteriorly on the dorsal surface of

the ala or, in some specimens (esp. smaller

ones), the orifice is located posterior to the

depression. The torus superior is

extensive, with two to three tubercles

continuing (except the most anterior one) as

a short and low ridges laterally. The

anterior-most tubercle continues as a short

rounded ridge anteroventrally, onto the

medial surface of the bone. ZIL/EL/1 24

(and some other ones) is much smaller but

except for size its morphology corresponds

in all principal features to that described

above.

December 1993

Asiatic Herpetological Research

Vol. 5, p. 33

«&&

CM

EM

FIG.1. A- Rana sp. (ZIL/EL/148). Left humerus in ventral, dorsal and lateral views (from the left to

right). B-fianasp. (ZIL/EL/151). Left humenis in ventral, dorsal and lateral view (from the left to right).

C- Rana sp. (ZIL/EL/74). Ventral (left) and dorsal (right) view of the proximal section of theright

humerus (drawing reversed for comparison). Bar equals 1mm. Abbreviations: C. A. - crista adventiva;

C.L. - crista lateralis; C. M. - crista medialis; C. V. - crista ventralis; E. M. - epicondylus medialis; F. D. -

fossula dividens; P. L. -processus lingualis.

Discussion: The shape of the ilium

corresponds to that in contemporary Bufo

bufo and B. viridis. The only difference is

that in both latter forms the longitudinal

depression on the medial surface of the ala

is developed in much lesser degree (due to

lesser extent of the edge). It should also be

noted that the orifice of the mentioned canal

on the medial surface of the bone displays

certain variation in contemporary forms (the

orifice may be doubled, and the differences

in this respect may be found also between

the left and right ilium of a single

individual). The same seems to hold for

fossil material. Size differences may be

ascribed either to interspecific variation or

to differences between both sexes (the latter

may reach quite a high degree in some

contemporary representatives of the genus

Bufo.

Vol. 5, p. 34

Asiatic Herpetological Research

December 1993

B

D

FIG. 2. A- Bufo sp. Right humerus (ZIL/EL/1 12). B- Bufo sp. Right humerus (ZIL/EL/21). C- Bufo

sp. Left humerus, drawing reversed for comparison (ZIL/EL/1 32). D- Ranidae indet. Right humerus

(ZIL/EL/150). E- Ranidae indet. Right humerus (ZIL/EL/1 10). Bar is 1 mm.

Rana sp.

Material: Humeri ZIL/EL/56, ZIL/EL/74

(Fig. 1C), ZIL/EL/83, ZIL/EL/128,

ZIL/EL/1 48 (Fig. 1A), ZIL/EL/1 51 (Fig.

IB), ZIL/EL/1 79.

Description: The crista ventralis and

crista ad ventiva delimit a wide, shallow and

rather longitudinal depression for muscle

insertion. The latter crista may be

developed to various degree, whereas the

crista ventralis is well developed in nearly

all individuals, with a distinctive lingual

process (only in ZIL/EL/128 this process is

poorly developed, and the crista ventralis

continues distally as a gradually lowering

ridge). The crista ventralis has a hollow

inside its free margin; consequently, it is

thinner along its attachment to the

diaphysis. This is manifested externally by

grooves along the insertion of the crista, on

:ither side. The crista lateralis and medialis

are directed dorsally, forming thus a

longitudinal groove on the dorsal surface of

the bone. The proc. lingularis and the

outgrowth produced by the crista adventiva

may form together a roof over the fossula

dividens; this nearly results in a canal. The

caput humeri is well prominent ventrally

(clearly seen in lateral aspect). Although

the distal part of the bone is straight, the

crista medialis and lateralis make it

seemingly "S" shaped. Lateral epicondylus

is entirely absent. Other features may be

seen in Fig. IB.

Some variation may be observed, mainly

in the degree of development of the lingual

process and in the extent of the medial and

lateral cristae, as well as of the crista

adventiva.

Anatomical notes: As may be deduced

from the condition in Rana esculenta

(Gaupp, 1896) the depression between the

crista ventralis and adventiva could serve as

an area of insertion for three heads of the

December 1993

Asiatic Herpetological Research

Vol. 5, p. 35

FIG. 3. A- Ranidae indet. Right ilium in lateral view (ZIL/EL/87). B- Ranidae indet. Left ilium in lateral

view (ZIL/EL/1). C- Ranidae indet. Lett ilium in lateral view (ZIL/EL/14). D- Anura indet. Left ilium in

lateral view (ZIL/EL/3). B, C, and D reversed for comparison. Bar equals 1mm. Abbreviations: A. -

acetabulum; A. O. I. - ala ossis ilii; C. O. I. - crista ossis ilii; P. A. - pars ascendens; P. C. -

parscylindriformis; P. D. - pars descendens; T. S.- tuber superius.

m. pectoralis (portio epicoracoidea,

sternalis and abdominalis), whereas the

proximal part of the crista itself (its edge)

could serve for attachment of two heads of

the m. deltoideus (pars clavicularis and

scapularis). The third head of the

deltoideus muscle (pars episternalis) is

fixed to the ventral ridge of the medial

epicondylus. The fossula dividens

probably served for tendon of the m.

coracoradialis. The groove between both

the crista medialis and lateralis served no

doubt for insertion of the caput profundum

of the m. anconeus. The medial crista

serves in anurans for attachment of the m.

flexor carpi radialis and the lateral crista for

the m. extensor carpi radialis. The former

has its special function in amplexus.

Consequently, the crista medialis is usually

better developed in males, and the degree of

its development is considered secondary

sex character.

Taxonomic notes: Humeri recalling

those described above may be found in

some individuals of contemporary "brown"

frogs, i.e. of Rana temporaria R. arvalis,

R. dalmatina, R. latastei, and R .

macrocnemis. I found morphology closely

resembling that in ZIL/EL/1 51 (Fig. IB) in

contemporary Rana arvalis (DP FNSP

5830) from S Bohemia (Czechoslovakia),

in R. arvalis wolterstorfii (DP FNSP 6264)

from Soroksar (Hungary), and in R.

latastei (DP FNSP 6419) from Como

(Italy). In other individuals, the cristae and

the lingual process were developed to

various degree, similar to the condition in

the described fossil material. In all cases

these humeri belonged to males. Hence, it

Vol. 5, p. 36

Asiatic Herpetological Research

December 1993

FIG. 4. A- Bufo sp. Right ilium in lateral view (ZIL/EL/94). B-Bufo sp. Left ilium in lateral view

(ZIL/EL/99). C- Ranidae indet. Left ilium in lateral view (ZIL/EL/68). D- Ranidae indet. Right ilium in

lateral view (ZIL/EL/1 11). B and C reversed for comparison. Bar equals 1 mm.

may be concluded that the described

characters on the humerus may be ascribed

to sexual dimorphism, but they are not

present in all males. In any case, relations

to contemporary "brown" ranids seems to

be beyond any doubt. It is quite possible

that the above described humeri assigned to

Rana sp. and humeri (and other elements)

described below as Ranidae indet. might

belong to a single form.

Material:

(Fig. 7E).

Ranidae indet.

Right coracoid, ZIL/EL/54

Description: The intumescentia

glenoidalis is circular, with distinct but

small fovea acetabuli where the ligament of

the humerus is inserted. The fovea is

surrounded by marginal part for the

cartilago paraglenoidalis that connects this

bone with the proc. glenoidalis scapulae.

The pars epicoracoidealis is nearly regularly

arch-like, exceeding by its antero-posterior

diameter the lateral part of the bone.

Material: Humeri ZIL/EL/1 10 (Fig. 2E),

ZIL/EL/1 50 (Fig. 2D), ZIL/EL/1 68.

Description: The crista ventralis humeri

in ZIL/EL/1 10 (and in ZIL/EL/1 68 that is

similar) is prominent, gradually lowering

distally. Part of its margin is tongue-like

bent medially. Only within the proximal

section of the crista there is a groove along

the medial surface of its basis. The crista

medialis is well developed, but only in the

distal third of the bone. The lateral crista is

developed in lesser degree. The medial

December 1993

Asiatic Herpetological Research

Vol. 5, p. 37

FIG. 5. Ranidae indet. Vertebrae in ventral view. A- Sacral vertebra (ZIL/EL/10). B- V6 (ZIL/EL/70).

C- V2 (ZIL/EL/8). Bar equals 1 mm. Fig.6 Ranidae indet. Praesacral vertebrae in dorsal (above) and

ventral (below) views. A- V5 (ZIL/EL/17). B- V2 (ZIL/EL/125). Bar equals 1 mm.

Vol. 5, p. 38

Asiatic Herpetological Research

December 1993

epicondylus is well developed, the lateral

one is rudimentary. On the medial surface

of the diaphysis there is low but discernible

crista. Specimen ZIL/EL/150 has its

ventral crista well developed, with rounded

and almost straight margin. It is thin along

its attachment to the diaphysis. This,

together with the fact that the diaphysis is

oval in cross-section, caused that there is a

groove along the basis of the crista,

especially well developed on its medial

surface. Neither medial nor lateral crista

are developed in this specimen, and the

lateral epicondylus is absent, too. In spite

of these differences, both humeri may be

assigned to the Ranidae.

Material: Ilia ZIL/EL/1 (Fig. 3B),

ZIL/EL/2, ZIL/EL/14 (Fig. 3C),

ZIL/EL/18, ZIL/EL/19, ZIL/EL/28,

ZIL/EL/29, ZIL/EL/49, ZIL/EL/68 (Fig.

4C), ZIL/EL/87 (Fig. 3 A), ZIL/EL/1 11

(Fig. 4D), ZIL/EL/135, ZIL/EL/164,

ZIL/EL/177. Specimens ZIL/EL/2,

ZIL/EL/19, ZIL/EL/29, ZIL/EL/135 and

ZIL/EL/177 are similar to ZIL/EL/1 1 1 (Fig.

4D) in that the tuber superius is continuous

with the crista ilii.

Description: The crista ossis ilii

(vexillum of some authors) and the pars

cylindriformis can be well distinguished in

medial view, whereas only with some

difficulties in lateral view. The crista does

not reach up to the anterior end of the ala.

It is directed dorsally in its section adjacent

to the tuber superius, and bent

dorsomedially in its anterior portion.

Consequently, it forms wide groove on the

medial surface of the ala, roofed dorsally

by the edge of the crista, and ventrally

delimited by the pars cylindriformis. The

tuber superius is prominent above the level

of the crista (but not in ZIL/EL/1 11),

however, its margin corresponds by its

thickness to the edge of the latter. Only

posteriorly the tuber has a rounded margin,

declined rather laterally. In specimen

ZIL/EL/68 the tuber is prominent but not

extensive; it has conspicuous ridge running

down obliquely from its top. There is a

distinct depression between the tuber and

pars cylindriformis. The depression

extends to the dorsal margin of the ala,

separating thus the tuber from the crista.

The acetabulum is with acute and prominent

margins, considerably lifted above the pars

descendens, but rather slanting down

towards the pars ascendens. However,

even here the margin of the acetabulum is

represented by a distinct ridge. There is a

considerable notch between the tuber

superius and the dorsal margin of the pars

ascendens. The dorsal margin of the pars

ascendens continues anteroventrally onto

the medial surface of the bone as a rounded

ridge that disappears after a short distance.

ZIL/EL/1 (Fig. 3B) is essentially the same

but the crista is much lower than is the

dorsoventral diameter of the pars

cylindriformis, in whole its extent anterior

to the tuber superius. Anteriorwards it is

even getting lower, so its transition into the

dorsal margin of the pars cylindriformis is

indistinct. The tuber is prominent and

extensive anteroposteriorly.

Anatomical notes: The lateral surface of

the crista ossis ilii is an insertion area for

the m. iliacus externus, the other end of

which is fixed by a tendon to the proximal

part of the femur (Gaupp, 1896, figs 104,

105). The inner surface of the crista is

occupied by the m. coccygeo-iliacus that

runs to the urostyle. The iliacus externus

muscle is, together with the iliacus

internus, one of the most robust pelvic

muscles in ranids and perhaps it plays an

important role in jumping, despite of the

fact that its tendon is fixed close to the

proximal end of the femur. The tuber

superius serves for attachment of the m.

glutaeus magnus, m. ilio-fibularis, and m.

ilio-femoralis. The first is the most robust

muscle of the dorsal side of the thigh, and

together with other heads of the m. triceps

femoris it stretches the knee joint. All

muscles that are inserted on the tuber

superius are important for locomotion.

Discussion: Contemporary European

ranids mostly have the torus superior ilii

continuous with the crista ilii, regardless if

they belong to the esculenta or temporaria

complexes. However, certain variation

exists in this respect, so one can find

individuals with prominent torus also in

those forms in which it is continuous with

December 1993

Asiatic Herpetological Research

Vol. 5, p. 39

B

FIG. 6. Ranidae indet. Prae sacral vertebrae in dorsal (above) and ventral (below) views.

(ZIL/EL/17). B- V2(ZIL/EL125). Bar equals 1 mm.

A- V5

the crista in most individuals. This is why

more precise assignment is difficult.

Material: Praesacral vertebrae V2 -

ZIL/EL/8 (Fig. 5C) and ZIL/EL/125 (Fig.

6B); V3 - ZIL/EL/9; V5 - ZIL/EL/17 (Fig.

6A); V6 - ZIL/EL/70 (Fig. 5B).

Description: The centra are procoelous,

dorsoventrally compressed. In V2, the

diapophyses are distinctly inclined

anteriorly and slightly also ventrally; they

are oval in cross-section due to moderate

dorsoventral compression. ZIL/EL/125 is

similar in its preserved characters but

differs in having the postzygapophyses

more robust, and the posterior convexity of

the centrum more prominent (see Fig. 6B).

Besides that, the anterior-posterior distance

between the prae- and postzygapophyses is

greater than in ZIL/EL/8 because the former

processes extend anteriorly beyond the

level of the centrum. The neural arches of

ZIL/EL/125 (they are not preserved in

ZIL/EL/8) produce distinct proc. spinosus

which is, as usually in V2 of ranids,

directed posteriorly. Its dorsal surface is

flat, only anteriorly there is a narrow and

low ridge paralleled by a depression on

either side. V3 is represented by ZIL/EL/9

which is preserved only as fragment

lacking the centrum, but its diapophysis

with some rugosity in the middle of its

length, as well as an extent of its neural

canal and shape of its praezygapophysis,

suggest its relations to the ranids.

ZIL/EL/17 is V5; it has its proc. spinosus

directed dorsally (again, as usual in ranids).

Perhaps it might be associated with

ZIL/EL/70 (see below), judging by the

shape of the centrum in ventral view (also

in this specimen the posterior convexity is

divided by a slot, though visible only in

posterior aspect). Peculiar feature of this

specimen is the ventral edge of its anterior

concavity which runs out anteriorly in the

mid-line (see Fig. 6A). It is difficult to say

whether this is of some taxonomic

Vol. 5, p. 40

Asiatic Herpetological Research

December 1993

FIG. 7. A- cf. Bufo sp., left scapula in lateral view (ZIL/EL/140). B- Bufo sp., left scapula in lateral view

(ZIL/EL/7). C- Anura indet., left praearticular in dorsal view (ZIL/EL/12). D- Anura indet., parasphenoid

(ZIL/EL/185). E-Ranidae indet., right coracoid in ventral view (ZIL/EL/54). F- Anura indet., right

pterygoid (ZIL/EL/141). Abbreviationsx. gl. - cavitas glenoidalis; i. gl. - intumescentiaglenoidalis; m.a. -

margo anterior; p. a. pars acromialis; p. gl. - proc. glenoidalis. Bar equals 1 mm.

importance. V6 is represented by

ZIL/EL/70. Its diapophyses are rounded in

cross-section, and are of the same diameter

both proximally and distally. They are

horizontal, not inclined dorsally. The

posterior convexity is clearly divided

vertically by a slot which is better

developed than in ZIL/EL/17. Signs of

such slots may be observed in

corresponding vertebrae of some

individuals of the contemporary Ranidae

(e.g., Rana esculenta). All the described

praesacral vertebrae have in common a

distinct indentation along the posterior edge

December 1993

Asiatic Herpetological Research

Vol. 5, p. 41

of the neural arches; this is interrupted only

in the mid-line where a distinct ridge runs

down from the proc. spinosus.

All the above features suggest that

vertebrae of the Ranidae should be

concerned. Bufonids are excluded mainly

because of the morphology of their proc.

spinosus and because their neural canal is

less spacious.

Material:

(Fig. 5A).

Sacral vertebra ZIL/EL/10

Description: The centrum is

dorsoventrally compressed, bicondylar

posteriorly, both condyli being separated

by a comparatively wide notch. The

anterior side of the centrum is not preserved

but comparison with contemporary ranids

suggests that there could be a mineralized

intervertebral disc. The diapophyses are

inclined dorsally, and are distinctly

compressed dorsoventrally. The

articulation surface of the

praezygapophyses is, in correspondence

with the inclination of the diapophyses,

facing dorsomedially.

Anura indet.

Material: Parasphenoid, ZIL/EL/185

(Fig. 7D).

Description: The shape and proportions

of the bone may be seen from Fig. 7D.

Among the characters that should be

mentioned are the proc. posterior which is

well developed, narrow proximal parts

(compared with the distal sections) of the

lateral processes and of the pars medialis,

and distinct indentations on the transition

between the pars medialis and lateral

processes caused by a low ridge on either

side; similar ridge continues on both sides

from the lateral edge of the proc. posterior

onto the surface of the proc. lateralis where

it disappears.

Material: Left praearticular, ZIL/EL/12

(Fig. 7C).

Description: The proc. coronoideus is

well developed, nearly vertical in position.

The sulcus pro cart. Meckeli is, especially

in the posterior moiety of the bone, only

moderately developed. These characters do

not allow precise assignment.

Material: Right pterygoid, ZIL/EL/141

(Fig. 7F).

Description: The inner surface (margo

orbitalis) of the ramus maxillaris does not

bear any crista and the ramus itself is

almost straight. The distal section of the

ramus posterior is broken off, so the real

proportions of the bone are difficult to

reconstruct.

Material: Ilium, ZIL/EL/3 (Fig. 3D).

Description: Although this ilium is

preserved only as a small section, important

diagnostic characters are preserved. The

crista ossis ilii is well developed, and may

be distinguished both in medial and lateral

view. In contrast to ranids, the torus

superior is not developed, and the anterior

margin of the pars descendens is directed

posteroventrally instead of ventrally or even

anteroventrally (see Bohme 1977, fig. 9).

Material: Praesacral vertebra (most

probably V5 or V6), ZIL/EL/1 1.

Description: Only the centrum and bases

of the left transverse process incl. adjacent

praezygapophysis are preserved.

However, one can conclude after the

declination of the proximal part of the

transverse process that V5 or V6 should be

concerned. The centrum is dorsoventrally

compressed and procoelous, though its

posterior side is also slightly concave. As

its surface does not display spongious

structure (indicating a crack) it can be

admitted that there could be a free

intervertebral disc that in living animal

adhered the posterior end of the centrum.

The ventral surface of the centrum is almost

at the same level as the proximal section of

the transverse processes, and the centrum

itself is short antero-posteriorly. The

praezygapophysis is comparatively small

and located far laterally (its distance from

the lateral edge of the proximal concavity of

the centrum is about half the diameter of

Vol. 5, p. 42

Asiatic Herpetological Research

December 1993

-OL

D

Fig. 8. Radioulnae in medial view. A- Bufo sp. (ZIL/EL/62). B-Bufo sp. (ZIL/EL/106). C- Bufo sp.

(ZIL/EL/120). D- Anura indet. (ZIL/EL/36). E- Anura indet. (ZIL/EL/46). F- Anura indet. (ZIL/EL/72).

D-F reversed for comparison. Arrows indecate border between periost and that part of the bone without

periostal surface. Abbreviations: CAP. - capitulumradioulnae; OL. - olecranon; cr. - crista on the ulnar

margin. Line equals 1 mm.

this concavity). The neural arches are

comparatively thin, and the neural canal

was obviously quite spacious.

Material: Urostyle, ZIL/EL/88.

Description: This element fits

morphologically into the variation range of

contemporary Ranidae. Both in Ranidae

and Bufonidae this range is rather broad

which precludes precise assignment of the

specimen.

Material: Radioulnae ZIL/EL/36 (Fig.

8D), ZIL/EL/46 (Fig. 8E), ZIL/EL/72 (Fig.

8F), ZIL/EL/73, ZIL/EL/1 14, ZIL/EL/152.

Description: These radioulnae are

medium to small sized (compared with

those identified as Bufo). A conspicuous

character is that most of them are laterally

compressed in their columnar section. This

results in that their anterior and posterior

margins run out in a distinct ridge. The

smallest specimen (ZIL/EL/72), however,

has its margins rounded. These radioulnae

might be ascribed to the Ranidae, however,

lack of diagnostic characters of these

elements casts some doubts on this

assignment.

Accompanying Vertebrate Fauna in the

Sample

From Kudaro I Cave, Tsepkin (1980)

gave a list of accompanying fishes,

Darevsky (1980) mentioned one lizard

genus (Lacerta sp.), Burchak-Abramovich

(1980) gave a list of birds, Gadzhiev

(1980) bats, Gromov & Fokanov (1980)

December 1993

Asiatic Herpetological Research

Vol. 5, p. 43

rodents, and Vereshchagin & Baryshnikov

(1980) large mammals. In the sample that

was placed at my disposal there were,

besides frogs, also two different forms of

birds, and following mammals

(determination by Dr. Ivan Horacek,

Department of Zoology, Charles

University, Prague): Talpa cf. caeca

Prometheomys schaposchnikovi, Arvicola

cf. terresths, Microtus cf. gud, Microtus

("Pitymys") cf. majori, and cf. Lupus.

References

BANNIKOV, A. G., I. S. DAREVSKY, V. G.

ISHCHENKO, A. K. RUSTAMOV, AND N. N.

SHCHERBAK. 1977. [Key to determination of

amphibians and reptiles of the fauna of the

USSR]. Prosveshchenie, Moscow. (In

Russian).

BOHME, G. 1977. Zur Bestimmung quarterer

Anuren Europas an Hand von Skelettelementen.

Wiss. Zeitschr. Humb. Univ. Berlin, math.-

nat., 26:283-300.

Conclusions

Determination of the material revealed

that its substantial part belongs to the genus

Bufo and to the family Ranidae. Minor part

(represented by fragmentary or less

numerous elements) could be determined

only as Anura indet. Precise determination

was impossible because of supposed

individual and developmental variation.

Nevertheless, results of this determination

show that generic composition of the

Holocene anuran fauna in this region was

basically the same as contemporary one.

The genus Bufo in the corresponding

altitudes of Caucasus is nowadays

represented by. Bufo verrucosissimus, and

B. viridis, genus Rana by R. macrocnemis,

and possibly also by R.

ridibun da(Bannikov et al., 1977;

Kuznetsov, 1974; Tuniyev, 1990).

Besides, there occurs sporadically also

Pelodytes caucasicus in South Ossetia,

however, presence of this genus in the

fossil material could not be confirmed.

Acknowledgments

I am grateful to Professor I. S. Darevsky

(Zoological Institute, St. Petersberg) for the

loan of the fossil material for study, and to

Dr. I. Horacek (Department of Zoology,

Charles University, Prague) for the

determination of accompanying

micromammalian fauna. Thanks are due

also to Dr. B. Sanchiz (Museum of Natural

History, Madrid) for his valuable

suggestions.

BOLKAY, S. 1919. Osnove uporedne osteologije

anurskih batrahija. Glasnik Zemaljskog muzeja

Bosni i Hercegovini, (1919):277-357.

BURCHAK-ABRAMOVICH, N. I. 1980. [Remains

of birds from the Kudaro I Cave]. Pp. 98-110.

In Ivanova, I. K. and A. G. Tchemyakovsky

(eds) Kudarskye peshchernye paleoliticheskie

stoyanki v Yugo-Ossetii. Nauka, Moscow. (In

Russian).

DAREVSKY, I. S. 1980. [Amphibians and reptiles

from Kudaro I cave]. Pp. 125-127. In Ivanova,

I. K. and A. G. Tchernyakhovsky (eds)

Kudarskye peshchernye paleoliticheskie stoyanki

v Yugo- Osetii. Nauka, Moscow. (In

Russian).

GADZHIEV, D. V. 1980. [Remains of bats

(Chiroptera) from the Kudaro I Cave]. Pp. 11-

124. In Ivanova, I. K. and A. G.

Tchernyakhovsky (eds) Kudarskye peshchernye

paleoliticheskie stoyanki v Yugo-Ossetii.

Nauka, Moscow. (In Russian).

GAUPP, E. 1896. Anatomie des Frosches. Lehre

vom Skelet und vom Muskelsystem. Fridrich

Vieweg und Sohn, Braunschweig.

GROMOV, I. M. AND V. A. FOKANOV. 1980.

[On remains of Late- Pleistocene rodents from

the Kudaro I Cave].Pp. 79-89. In Ivanova, I. K.

and A. G. Tchernyakhovsky (eds) Kudarskye

peshchernye paleoliticheskie stoyanki v Yugo-

Ossetii. Nauka, Moscow. (In Russian).

KUZNETSOV, B. A. 1974. [Key to determination

of vertebrate animals of the fauna of the USSR.

I. Cyclostomata, fishes, amphibians and

reptiles]. Prosveshchenie, Moscow. (In

Russian).

LYUBIN, V. P. 1980a [Geographical position of

cave settlements of Yugo-Ossetia]. Pp. 6-12.

Vol. 5, p. 44

Asiatic Herpetological Research

December 1993

In Ivanova, I. K. and A. G. Tchemyakhovsky

(eds) Kudarskye peshchernye paleoliticheskie

stoyanki v Yugo-Ossetii. Nauka, Moscow. (In

Russian).

LYUBIN, V. P. 1980b. [Geological-stratigraphic

conditions of paleolithic deposition in eastern

gallery of the Kudaro I Cave]. Pp. 13-32. In

Ivanova, I. K. and A. G. Tchemyakhovsky (eds)

Kudarskye peshchemye paleoliticheskie stoyanki

v Yugo- Ossetii. Nauka, Moscow. (In

Russian).

MLYNARSKI.M. 1976. Discoglossus

giganteusWellslein- Westerheimb, 1955

(Discoglossidae, Anura) from the Miocene of

Przeworno in Silesia (Poland). Acta Zool.

Cracov. 21:1.12.

TSEPKIN, E. A. 1980. [Remains of fishes from

the Kudaro I Cave]. Pp. 90-97 In Ivanova, I. K.

and A. G. Tchemyakhovsky (eds) Kudarskye

peschemye paleoliticheskie stoyanki v Yugo-

Ossetii. Nauka, Moscow. (In Russian).

TUNIYEV, B. S. 1990. On the independence of the

Colchis Center of amphibian and reptile

speciation. Asiatic Herpetological Research

3:67-84.

VERESHCHAGIN, N. K. AND G. F.

BARYSHNIKOV. 1980. [Remains of mammals

in eastern gallery of the Kudaro I Cave

(excavations made by V. P. Lyubin in 1957-

1958)]. Pp. 51-62 In Ivanova, I. K. and A. G.

Tchemyakhovsky (eds) Kudarskye peschemye

paleoliticheskie stoyanki v Yugo-Ossetii, pp.

51-62. (In Russian).

I December 1993

Asiatic Herpetological Research

Vol. 5, pp. 45-50

Karyotype, C-Band and Ag-Nors Study of Three Stink Frogs

Gang Wei1, ning Xu1, dejun Li1, Guanfu wu2 and xiquan Song3

'Department of Biology, Zunyi Medical College, 563003 Guizhou, China

2Chengdu Institute of Biology, Academia Sinica, 610015 Sichuan, China

■* Department of Biology, Zunyi Teachers College, 563002 Guizhou. China

Abstract.-The karyotypes.C-bands and Ag-NORs of Rana kuangwuensis, R.andersonii and R.margaratae

were analyzed. Intra- and interspecific chromosome variations, including centromeric type and C-banding

patterns, were detected. It was assumed that the Guizhou Plateau was the distributional center of the origi-

nal place of the group.

Key Words: Amphibia, Ranidae, Rana kuangwuensis, Rana andersonii, Rana margaratae, China, kary-

otype, C-band, Ag-NORs.

U«M«K"*»

lYltfXXftlAAAA

•t}f!Mtt(,t,«tt|,|'*M

FIG. 1. a: Karyotype of Rana andersonii. b: showing C-bands. c: showing Ag-NORs.

Introduction

The group of stink frogs which have a

special stink smell from the skin consists of

nine species, i.e. Rana andersonii, R. an-

lungensis, R. grahami, R. kuangwuensis,

R. lungshengensis, R. margaratae, R.

schmackeri, R. tiannanensis and R.

wuchuanensis. They are considered to be

phylogenetically close because of similar

morphological characters in adults and tad-

poles. Among them, the karyotype and C-

bands of R. grahami from Kunming, Yun-

nan has been studied by Li (1982). In ad-

dition, the karyotype, C-bands and Ag-

NORs of R. margaratae from Emei Moun-

tain, Sichuan have been analyzed by Wang

(1983) and Wu (1990). In the present pa-

per, the karyotypes, C-bands and Ag-

NORs of R. kuangwuensis, R. andersonii

and R. margaratae were analyzed.

Methods

Two females and one male R. andersonii

were captured at Qianxi, Guizhou

Province, (27°20' N,106°16' E). One fe-

male and three male R. kuangwuensis were

captured at Nanjiang, Sichuan Province

(32°30' N, 106°40' E) and one female and

1993 by Asiatic Herpetological Research

Vol. 5, p. 46

Asiatic Herpetological Research

December 1993

^{((IK11'1"1""1"

at

XilYA

if H h }] ,| M .1

•»

I! f.

v r vtffVWM*

ik

mMm^mw^m

A A

RG.2. a: Karyotype of Rana kuangwuensis. b: showing C-bands. c: showing Ag-NORs.

one male R. margaratae from Zunyi (27°40'

N, 106°50' E) were captured. Karyotypes,

C-bands and Ag-NORs preparations were

made after Wei et al. (1990).

Results

Figures 1, 2, and 3 depict the

karyotypes, C-bands and Ag-NORs of/?.

andersonii, R. kuangwuensis and R.

margaratae. For the measurment of the

karyotypes see table 1 . The diploid number

of the species are all the same, 2n=26,

comprising two groups.

The large chromosome group includes

chromosome Nos. 1-5, with a relative

length (R.L.) larger than 9%. With regard

to the arm ratio (A.R.), chromosome Nos.

1 and 5 are metacentric in all three species.

No. 2 is metacentric in R. margaratae and

R. andersonii, but submetacentric in R.

kuangwuensis. No.3 is submetacentric in

R. margaratae and R. kuangwuensis , but

metacentric in R. andersonii. No.4 is sub-

metacentric in R. andersonii and R. mar-

garatae, but metacentric in R. kuangwuen-

sis.

The small chromosome group comprises

chromosome Nos. 6-13, with a R.L. less

than 7%. Nos.6, 8, 10, 12 and 13 are

metacentric, No.7 is submetacentric in all

the three species. Nos. 9 and 1 1 are sub-

metacentric in R. kuangwuensis and R. an-

dersonii but metacentric in R. margaratae.

Secondary constrictions are observed in the

long arms of No. 10 of R. margaratae (only

one homologous) and R. andersonii but not

observed in R. kuangwuensis.

Treatment of the chromosome of the

three species according to the C-banding

method shows that each species has a

centromeric C-band on each chromosome.

For interstitial C-band, it is quite different

among the species. There is only an

interstitial C-band in lOq in R. andersonii.

And there is an interstitial C-band in 2p

(stained weakly), 3q and 4q (only one

homologous) in R. margaratae. But there

are much more interstitial C-bands in R.

kuangwuensis than in the other two

species. R. margaratae and R. andersonii

have not any telomeric C-band. But R.

kuangwuensis has some telomeric C-

bands.

December 1993

Asiatic Herpetological Research

Vol. 3, p. 47

Knuunxu"

lIAXtlAXfti

|^ KK]J ]i ,! iiwii it u •« ii

|5 Hi M'* II » *i I

FIG. 3. a: Karyotype of Rana margaratae. b: showing C-bands. c: showing Ag-NORs.

Specific staining of the NORs with silver

(Ag) confirms that the regions of NORs in

all the three species are the same, in the

long arms of chromosome No. 10. But the

relation between the NORs and the con-

strictions is quite different. The regions of

the one pair of NORs in R. andersonii cor-

respond to the regions of the secondary

constritions. R. margaratae has only one

NORs in the chromosome where the sec-

ondary constriction locates, and no NORs

is observed in the other homologous which

has no secondary constriction. While for

R. kuangwuensis, no secondary constric-

tion is observed in the regions of the

NORs.

Discussion

Comparing the karyotypes between the

three species, we detect some interspecific

variations. The secondary constriction is

not detected in R. kuangwuensis. The

chromosomes consist of 8 metacentric and

5 submetacentric pairs in R. kuangwuensis

and R. margaratae, while 9 metacentric and

4 submetacentric in R. andersonii.

C-banding patterns of all 13 pairs of R.

margaratae are similar to those of R. ander-

sonii, except for the variant bands on

chromosomes 2, 3, 4, and 10. While C-

banding pattern of R. kuangwuensis is

quite different from those of the other two

species, for R. kuangwuensis has more in-

terstitial C-bands and telomeric C-bands as

well. And the relation between the NORs

and the secondary constrictions is quite dif-

ferent among the three species.

This study also indicates that

intraspecific chromosome variations exist in

R. margaratae from different distributional

areas. We observe 8 metacentric and 5

submetacentric pairs in the present study,

as opposed to 12 metacentric and 1

submetacentric pairs from Emei Mountain,

Sichuan (Wang et al., 1983) and 9

metacentric and 4 submetacentric pairs also

from Emei Mountain (Wu 1990).

The C-banding pattern of chromosomes

1-13 of R. margaratae from Zunyi is com-

pared with those from Emei Mountain.

There is a centromeric C-band in each

chromosome and a terminal C-band at each

chromosome terminus, and an interstitial C-

band in the long arm of No. 3 from Emei

Vol. 5 p. 48

Asiatic Herpetological Research December 1993

TABLE 1 . Karyotypic data for Rana margaratae, R. andersonii, and R. kuangwuensis.

Mountain (Wang et al , 1983). Besides

those above, there is an interstitial C-band

in the aero long arm of No.7, and even het-

erogeneity observed in No.9. There is an

interstitial C-band in the middle of the long

arms of both homologues of No.9 in fe-

male, while only one homologous of No.9

is observed having an interstitial C-band,

the other homologue has not an interstitial

C-band in the middle of the long arm, but

has an interstitial C-band near the terminus

of the long arm (Wu 1990). We also ob-

served indeed C-band heterogeneity of

chromosome No.9 from Emei Mountain.

Yet in our present study, we do not detect

C-band heterogeneity of chromosome No.9

from Zunyi. And we detect other interstitial

C-band (2p, 4q) but no telomeric C-band

has been observed.

In the early stage of the karyotypic

evolution, a karyotype had generally more

metacentric chromosomes. With the

development, the karyotype differenciated

in the direction of having more

submetacentric or telocentric chromosomes

(Li, 1985). Generally speaking,

karyotypes with more telomeric and less

interstitial C-band are more original.

Between the two distributional areas of

R. margaratae, the specimens from Emei

Mountain has 12 (Wang 1983) or 10 (Wu,

1990) metacentric, and has more telomeric

and less interstitial C-bands, and that from

Zunyi has 8 metacentric and has less

telomeric and more interstitial C-bands. On

the view of point above, R. margaratae from

Emei Mountain is more original than that

from Guizhou.

The stink frog group is composed of 9

species. The distributions of them are as

follows:

R. andersonii: upper Burma to Yunnan,

Guizhou, Guangxi, Hainan

December 1993

Asiatic Herpetological Research

Vol. 5, p. 49

R. anlungensis: Guizhou (Anlung

County)

R. grahami: Sichuan, Guizhou, Yunnan

R. kuangwuensis: Sichuan (Nanjiang

County)

R. lungshengensis: Guizhou, Guangxi,

Hunan

R. margaratae: Gansu, Sichuan,

Guizhou

R. schmackeri: Henan, Gansu,

Sichuan, Guizhou, Hubei, Anhui, Jiangsu,

Zhejiang, Jiangxi,Hubei, Guangdong

R. tiannanensis: Yunnan, Hainan

R. wuchuanensis: Guizhou (Wuchuan,

Libo)

From the discription above, it could be

found that the distributional areas of some

species are very limited, only one or two

counties. So they are very rare and pre-

cious wildlife. And it could also be found

that the stink frog group is distributed

mainly in the south of China, and most of

them (7 species) are found on the Guizhou

Plateau. So, the Guizhou Plateau might be

the distributional center of the group.

There were another two species of stink

frogs. Their karyotypes and C-banding

patterns were published. They are R. gra-

hami and R. schmackeri. Both the species

have 10 metacentric and 3 submetacentric

pairs in their karyotypes, and both species

have one telomeric C-band, but the former

has 5 and the latter has 4 interstitial C-

bands.

Among the 5 species published their

karyotypes and C-banding patterns, R.

margaratae from Emei Mountain has most

metacentric pairs and most telomeric C-

bands. Although it does not have less in-

terstitial C-bands, it could still be consid-

ered as the most original in the viewpoint of

cytogenetics. Considering that R. mar-

garatae from Guizhou is more evoluted than

that from Emei Mountain, it might be as-

sumed that the stink frog group originated

in Emei Mountain and its adjacent plateau.

Acknowledgments

We are most grateful to Ms. Wei Wu for

providing us with a pre-publication

manuscript and Ms. Xiaomong Zheng for

help in collecting specimens and

experimental work. This research was

supported by the Guizhou Provincial

Committee of Science and Technology.

Literature Cited

FROST, D. R. 1985. Amphibian species of the

world. Pp. 480-520. The Association of Sys-

tematics, Lawrence, Kansas.

IKEBE, C, A. KEN-ICHI and K. SEI-ICHI.

1987. Karyotype analysis of two Japanese

salamanders, Hynobius nebulosus (Schlegel) and

Hynobius dunni Tago, by means of C-banding.

Zoological Science 4:621-626.

LI, GUOZHENG. 1985. Chromosome and its

research method. Science Press, Beijing Pp.l-

261. (In Chinese).

LI, S., Y. WANG, C, LI, R. WANG and G. LIU.

1982. An investigation for the karyotypic and

C-banding pattern on the two anuran Amphibia.

Acta Genetica Sinica (6):473-478. (In Chinese).

SCHMID, M. 1978. Chromosome banding in

Amphibia. Constitutive heterochromatin and

nucleolus organizer regions in Ranidae,

Microhylidae and Rhacophorodae. Chromosoma

68:131-148.

WANG, Z., X, WANG, and W. CHEN. 1983. A

comparative study on constitutive hete-

rochromatin and nucleolus organizing regions

(NORs) of three species of the genus Rana.

Acta Herpetologica Sinica 2(4): 1-6. (In

Chinese).

WEI, G., F. CHEN, and N. XU. 1990. An

investigation of the karyotypic C-banding and

Ag-NORs pattern on Rana chensinensis.

Hereditas (Beijing) 12(l):24-26. (In Chinese).

WEI, G., N. XU, and D. LI. 1990. An

investigation of the karyotype, C-banding, Ag-

NORs pattern on Rana schmackeri.

Cytogenetics of Amphibians and Reptiles,

Advances in Life Sciences. 147-152. Birkhauser

Vol. 5, p. 50 Asiatic Herpetological Research December 1993

Verlag (publisher) 270 pp. and F. H. Pough (Eds.), Biology of the Rep-

tilia, Vol. 12, Physiological Ecology.

Huey, R. B. 1982. Temperature, physiology, and Academic Press, New York,

the ecology of reptiles. Pp. 25-91. In C. Gans

I December 1993

Asiatic Herpetological Research

Vol. 5, pp. 5 1-58 1

The Variegated Toad Agama in Djungar Gate (Eastern Kazakstan) with

Notes on Certain Systematic Problems of Phrynocephalus versicolor Str.

(Reptilia: Agamidae)

M. L. GOLUBEV

Institute of Zoology. Academy of Sciences, Kiev, Ukraine

Abstract.-The distribution of Phrynocephalus versicolor in Djungar (=Junggar) Gate (Eastern Kazakstan)

was investigated. The characteristics of these lizards are: absence of red axillary spots and the presence of

red-orange subcaudal coloration. The taxonomic status of this population and of subspecies of P. versicolor

is discussed. The variegated toad agama is presumed to be a "composed" species.

A^y Words: Reptilia, Sauria, Agamidae, Phrynocephalus versicolor, Kazakstan, China, Djungar Gate,

systematics, distribution.

75

U r u w q i

FIG. 1. The distribution of Phrynocephalus versicolor hispida (I), P. gutlatus salenskyi (II), P. g.

alpherakii (III), and P. v. doriai (IV) in Eastern Kazakhstan and Chinese Djungaria.

Introduction

There are few reliable literature citations

concerning the distribution of the variegated

toad agama, P. versicolor Str. from the

eastern part of the Balkhash-Alkol

Depression. Only Paraskiv (1956) reported

finding the toad agama "near Lake

Zhalanashkol". He also indicated that the

distribution of this lizard is a natural

continuation of the the Djungar Gate in the

Alakol Depression, "somewhere along the

northern Lake Alakol shore" (Paraskiv,

1956). According to Kubykin (1975), one

specimen of this lizard was captured by him

on Sredniy Island (Lake Alakol).

However, this specimen is not mentioned

in the collection list of the Institute of

Zoology of the Kazakh Academy of

Sciences (Brushko and Kubykin, 1988).

Semenov (1986) andSemenov et al. (1987)

refered to toad agamas collected "from

Alakol Hollow" in the Zoological Museum

of Moscow State University. All other

references to the distribution of this lizard

cite the above references.

Vol. 5, p. 52

Asiatic Herpetological Research

December 1993

FIG. 2. Habitat of Phrynocephalus versicolor in Djungar Gale near Druzhba Railway Station: crushed-

stony and gravel semidesert covered with boyalych (Sal sola orbuscula).

Methods

We studied the distribution of the

variegated toad agama in June 1991 during

investigations of the Djungar Gate territory

(Fig. 1).

Results and Discussion

The Djungar Gate is a relatively narrow

pass between the Balkhash-Alakol

Depression and Chinese Djungaria. This

pass is oriented from the northwest to the

southeast and rising in elevation towards

the southeast. The pass enters China near

the Druzhba ("Friendship") railway station

along the Lankol Valley. The Djungar Gate

Valley is formed by the broad alluvial plain

and gentle foothills of Maily Ridge and the

Djavlau Mountains to the northeast and the

more abrupt upthrust of the Djungar Alatau

to the southwest. The surface of the valley

alluvial plains varied with the degree of

slope and ranged from boulders, rubble.

gravels and fine gravels on a loess base

with the finer sorted materials deposited

farther from the mountains slopes.

Southeast of Lake Zhalanashkol the valley

floor is a broad alkaline plain (20-25 km.)

with subsurface water.

The dominate plant, Salsola orbuscula,

(common Russian name = boyalych, also

known in the United States as Russian

thistle) is found on the lower slopes and

alluvial plains of the valley. It is more

widely distributed on the northeastern

slopes but it is sometimes replaced by

saxaul (Haloxylon sp). Wormwood

(Artemisia sp.) is dominate among the

grasses and nearly the only plant on flats

without shrubs.

From the Lankol Valley the toad agamas

are distributed along both the northeastern

and southwestern slopes above the Djungar

Gate. Along the southwestern alluvial plain

the toad agama is distributed 15-20 km

December 1993

Asiatic Herpetological Research

Vol. 5, p. 53

from the Druzhba station to the northwest.

The lizard occurs along the foothills of the

Maily Ridge and the Djavlau Mountains for

55-60 km where the northern limit of its

distribution coincides with the border of the

boyalych dominate, gently sloping alluvial

plain composed of rock rubble and gravels.

Further north, on the steeper slopes of the

alluvial plain composed of larger rock

rubble and dominated by wormwood, the

sunwatcher (P. helioscopus) is found. It is

possible that P. versicolor occurs much

further to the northeast into the Alakol

Depression along the foothills, however

field work in this area is difficult because of

the presence of military installations along

the border. A small isolated area with

conditions which would make good habitat

for this species is found along the railway

tracks between the Zhalanashkol Station

and the 19th Station near the mouth of the

Irgaty River.

Toad agamas are found under single

bushes in small groups composed of 1-2

males and 2-5 females of different ages. In

addition, groups of up to 10 subadults were

observed. The density of the lizards is

variable, with higher densities in gravelly

areas with boyalych (Fig. 2) as well as in

areas of colonies of the great gerbil

(Rhombomys opimus ) that have excavated

through the darker colored gravels and rock

rubble and where the lighter loes makes up

the predominant coloration of the surface.

It is interesting to note that the lizards

inhabiting the gravel plains in the Djungar

Gate have retained the sand burrowing

behavior, involving rapid lateral

movements of the body, observed in

populations inhabiting sandy areas.

Pregnant P. versicolor as well as females

of other Phrynocephalus species assume

the "copulation avoidance" posture when

pursued by males (Polynova, 1982; 1989;

Rogovin, 1991; and our observations of P.

strauchi in the Fergan Valley). To assume

the "copulation avoidance" posture the

female turns onto her back as the male

approaches and maintains this position

while he is nearby.

Currently P. versicolor is considered to

be a polymorphic species and it is

interesting to determine the subspecific

position of the form inhabiting the Djungar

Gate.

For a long time it was assumed that in

eastern Kazakhstan this toad agama was

found in three isolated populations: the

Zaissan Depression, Alakol Depression and

the Hi River Depression (Paraskiv, 1956;

Bannikov et al., 1977). Peters (1984)

considered the Zaissan Depression and Hi

River Depression lizards to be two seperate

species: P. salenskyi Bedr. and P.

alpheraki Bedr. Three years later a new

subspecies, P. versicolor paraskiwi

(Semonov et al., 1987), was described

from the Hi River Depression. These

authors speculated that the two Chinese

Djungar Depression subspecies, P. v.

hispida Bedr. and P. v. doriai Bedr. were

conspecifics. However, becuase of a

shortage of material, they did not determine

the taxonomic status of the Alakol

Depression variegated toad agama. Soon

after the most significant attempt to analyze

the intraspecific variaton of P. versicolor

was undertaken (Semenov and Shenbrot,

1989).

Semenov and Shenbrot (1989) examined

675 specimens: 580 from Mongolia and

Tuva, 65 from the Hi River Depression, 19

from Chinese Kuldja (now Yining,

Xinjiang, China), 11 from the Alakol

Depression, but no specimens from the

remainder of the range of this species in

China. The authors, using discriminant

analysis techniques, felt their material was

adequate to discuss all known subspecies

of the variegated toad agama.

Semenov and Shenbrot (1989) indicated

that P. v. paraskiwi was detached from the

main group. Also, P. v. doriai from

Kuldja and P. v. kulagini from western

Mongolia were resurrected. The Alakol

variegated toad agama was singularly

attributed to P. v. doriai. These authors

were unable to distinguish the lizards from

Mongolian Djungaria from "typical" P. v.

hispida, however they did not indicate

which P. v. hispida they considered

typical. They rejected their original view

on the close relationship of the Djungar

forms. Phrynocephalus v. hispida was

recognized as identical to the nominative

form.

Discriminat analysis also has shown that

the presence of red axillary spots are useful

characteristics for seperating the closely

related pairs of subspecies, P. v. versicolor

- P. v. kulagini and P. v. paraskiwi - P.

v. doriai. The remaining characteristics

were found not to be useful for this

purpose. This was already noted in earlier

research (Nikolsky, 1915; Leroy, 1940;

Terentjev and Chernov, 1949).

However, a number of questions remain

unanswered. Has it been demonstrated that

axillary spots are absent in P. v. doriai and

in the Alakol toad agama? Is it apropriate to

include in the nominative subspecies,

charaterised by the presence of axillary

spots, the form P. v. hispida in which the

axillary spots may be present or absent

(Bedriaga, 1909)? If so, then what are the

reasons for seperating into distinct taxa P.

v. kulagini , which lacks axillary spots and

P. v. paraskiwi which has the axillary

spots? The latter form should be excluded

from further discussion since it has been

shown (Golubev, 1989) that it was

erroneously described and should be

attributed to P. gutatus alpherakii.

Discriminant analysis did not clarify the

relationship between P. v. doriai and P. v.

hispida.

It is clear (Table 1) that such

characteristics as the relative size of the

head and supraocular scales can not be used

unless they are standardized. Dorsal and

ventral coloration vary widely and are

effected in life by such physiological

considerations as body temperature and

ambient light and in preserved specimens

by the manner of preservation. Body

length to tail length, when expressed in

ratios (Bedriaga, 1909) and repeated

measurements of type specimens from the

Zoological Institute of the Russian

Academy of Sciences, St. Petersberg (ZIN)

did not confirm the differences noted by

Bedriaga (P. v. hispida ZIN 6637 females:

0.78-0.81; males 0.66-0.68; P. v. doriai

ZIN 5549, 8160 females 0.71-0.74; males

0.64-0.72). The ratios could be confirmed

by using liner dimension L. and L. cd. but

this was not done. Two other

characteristics, number of scales along and

across the top of the head are known to

vary widely among populations. Only

presence or absence of axillary spots

remains as a useful character for seperating

subspecies. However, we have no

information concerning this character in P.

v. doriai. Bedriaga (1909) used material

that had been in preservative for more than

December 1993

Asiatic Herpetological Research

Vol. 5, p. 55

FIG. 3. Dorsal view of Phrynocephalus versicolor from Djungar Gate (Djungar Railway Station).

10 years and these spots might have

disappeared during this time. Also there

are no detailed data on the distribution of

this character in lizards from eastern

Djungaria. It is known only that such spots

are present in the Mongolian part of the

range of the toad agama (Semenov and

Shenbrot, 1989). However, the nearly

isolated Mongolian Djungaria (Barun-

Khuray Depression) differs from Chinese

Djungaria in several geogrphic parameters

such as altitude.

The type specimens of P. v. hispida are

dated 1879 (ZIN 6637 and 6638) and 1880

(ZIN 6639). Nikolay M. Przewalsky's

Third Central Asian Expedition (First

Tibetan Expedition) took place at this time

(Dubrovin, 1980). Przewalsky left the city

of Zaissan on March 21 (April 4 by the

modern calendar) and reached Ulungur

Lake (Ulungur Hu) on March 31 (April 12

by the modern calendar). He followed the

Urungu River (Ulungur He) and its

tributary the Bulugun to the Barun-Khuray

Depression and crossed it from north to

south. He then crossed the Baytik Shan

Ridge and the plains of south eastern

Djungaria. On May 18 (May 30 by the

modern calendar) he reached Barkul and

did not return to Djungaria during this

expedition (Przewalsky, 1883). In

Przewalsky's journal the toad agamas

{Phrynocephalus sp.) are casually

mentioned for the middle and lower reaches

of the river. Thus, it is possible to draw

two conclusions: (1) the P. v. hispida type

specimens may have been collected in

different parts of Djungaria and thus

include different forms, (2) the date of

collection for the ZIN 6639 sample is

incorrect.

There is an overlap between both

subspecies of toad agamas from the

Djungar Gate region in body proportions

(females 0.6-1.06; males 0.64-0.75),

number of scales across (19-29; not

counting the supraocular scales) and along

(8-14) the top of the head. The scales on

the thigh are smooth and dorsal and ventral

coloration are highly variable (Figs. 3, 4,

5). In the material we collected, a slight

shift in these character's values toward

Vol. 5, p. 56

Asiatic Herpetological Research

December 1993

FIG. 4. Ventral view of Phrynocephalus versicolor from Djungar Gale (Djungar Railway Station).

hispida can be noted. It is important to note

that the subcaudal surface in living lizards

of both sexes is a red-orange color, that

with age looses its lustre and disappears

(Fig. 4). We also discovered remnants of

the red-orange coloration in one of 13

specimens from "Alakol Depression"

(Zoological Museum of Moscow State

University, MSU R7779). Recently we

examined specimens from China with the

same subcaudal coloration in the California

Academy of Sciences collected in the

central (northeast of the city of Karamay)

and southeastern (east from the city of

Urumqi) Djungaria. In all specimens

examined there are dark transverse bars on

the ventral tail surface and this agrees with

Bedriaga (1909) but not with Semenov

(1986).

From the above data, it follows that

variegated toad agamas inhabiting the

Alakol Depression and Djungar Gate to

southern Djungaria (>500 km) differ from

all other forms of P. versicolor in the

absence of axillary spots and brightly

colored subcaudal surface. Does this

indicate the existence of a new subspecies?

There are also red tailed toad agamas in

northern Djungaria and the Zaissan

Depression variously described as P.

guttatus, P. salenskyi and P. versicolor.

In the Zaissan Depression this lizard is

mostly sand-dwelling and similar in habits

to P. guttatus and P. frontalis (Golubev,

1989), while in northern Djungaria and in

some places in the Zaisson Depression they

are found in more stabilized soils. From

the Alakol Depression and the Djungar

Gate, where this species is found, there are

over 300 km of continuous habitat without

noticeable barriers into central Djungaria

where P. versicolor is found. This may

represent a cline with a gradual transition

from one form to the other.

In September 1991 repeated copulations

between a male P. salenskyi (Zaissan

Depression) and a female P. versicolor

(Djungar Gate) were observed in the Kiev

Zoo terrarium. If precopulation barriers

exist, they apparently can be broken in

terrarium conditions. Both Przewalsky

(1883) and Potanin (1948), when traveling

December 1993

Asiatic Herpetological Research

Vol. 5, p. 57

along the lower reaches of the Urungu

River and the southern shore of Ulungur

Lake (type locality of P. salenskyi ), noted

the variable coloration of toad agamas. We

discovered fragments of a light longitudinal

caudal stripe (a characteristic of P. g.

salenskyi) in some specimens of P.

versicolor from Djungar Gate (Fig. 5).

Red axillary spots are present not only in

the Alashan variegated toad agama (species

type locality) but in lizards inhabiting the

area south of Beishan Ridge in the Gashun

Goby. However, here P. versicolor are

connected by coloration and pattern

transitions with P. axillaris Blanf.

Thus, the question of the taxonomic

status of the variegated toad agamas from

Djungaria and Alakol Depression should

again be considered open as does the

question of the position of P. v. doriai.

The Kuldja and western Djungarian

populations are seperated by the Tianshan

Mountains. There are reasons to believe

that the Kuldja P. v. doriai is actually an

ecological race of the Hi P. g. alpherakii

while the Ebinurian P. v. doriai is

assignable to the acutirostris group (which

also might be no more than one of the color

variants of the axillaris- guttatus complex.

In summary, it appears that only two

subspecies of the variegated toad agama can

be recognized. P. v. kulagini inhabits

southern Tuva and northwestern Mongolia

and forms a narrow zone of intergradation

with the nominative subspecies P. v.

versicolor. However, there are doubts that

the axillary red spots constitute a

characteristic which allows one to delineate

populations specifically on the level of

geographaic race, i.e. subspecies. It is

possible that P. versicolor consists of

isolated, genetically differentiated color

morphs associated with stabilized soils.

Taxonomic seperation of these variants

should occur only after a detailed study of

the entire group. Bedriaga (1909)

recognized five subspecies and considered

P. versicolor to be a species composed of

many highly variable populations.

Bedriaga expressed concern that his

taxonomic arragement of these subspecies

did not represent a natural assemblage.

FIG. 5. A specimen of Phrynocephalus versicolor

from Djungar Gate (Djungar Railway Stauon)with

light longitudinal caudal stripe.

Further he stated "[only when we are more

familiar with the fauna of Central Asia, will

it be clear whether I have exaggerated

distinguishing characters]". Leroy (1940)

proposed that the species P. versicolor be

abolished. Leroy's point of view may be

closest to the truth.

Acknowledgments

The author is most grateful for the loan

Vol. 5, p. 58

Asiatic Herpetological Research

December 1993

of Phrynocephalus collections from M. E.

A. Dunajev and Dr. V. F. Orlova of the

Zoological Museum of the Moscow State

University, Russia; Mrs. L. Jogansen and

Prof. I. S. Darevsky of the Zoological

Institute of the Russian Academy of

Sciences, St. Petersberg; and Mr. J. V.

Vindum and Dr. A. E. Leviton of the

California Academy of Sciences, San

Francisco, California, U.S.A. Also, I

would like to thank Mr. V. B. Zaykovsky,

Mr. G. Makhnin, and Mr. G. I.

Zveryansky of the Railway Antiplague

Service, Alma-Ata, Kazakhstan for the

friendly help during the 1991 field season.

Literature Cited

NIKOLSKY, A. M. 1915. [Fauna of Russia and

adjacent countries. Reptiles. Vol. I. Chelonia

and Sauria]. Petrograd, Imperial Academy of

Sciences. 532 pp. (In Russian).

PARASKIV, K. P. 1956. [Reptiles of

Kazakhstan]. Kazakh Academy of Sciences

Press, Alma-Ata. 228 pp. (In Russian).

PETERS, G. 1984. Die Krotenkopfagamen

Zentralasiens (Agamidae: Phrynocephalus).

Mitteilungen aus dem Zoologischen Museum in

Berlin. Akademie Verlag, Berlin 60(1): 23-67.

POLYNOVA, G. V. 1982. [Demonstrative

behavior of Phrynocephalus mystaceus].

Zoological Journal, Moscow 61(5):734-741.

(In Russian).

BANNIKOV, A. G., I. S. DAREVSKY, V. G.

ISCHENKO, A. K. RUSTAMOV, AND N. N.

SCHERBAK. 1977. [Field guide of the USSR

amphibians and reptiles]. Moscow,

Prosveschenje Publishing House, Moscow.

369 pp. (In Russian).

BEDRIAGA, Y. 1909. Amphibian und Reptilien.

Pp. 73-502. In Wissenschaftliche Resultate der

Reisen N. M. Prezewalskijs durch Zentralasien.

Zoologische Teil. Band 3. Part 1. Lacertilia.

St. Petersburg. (In Russian/German).

Brushko, Z. K. and R. A. Kubykin. 1988.

[Catalogue of the herpetological collection of

the Institute of Zoology of the Kazakh Academy

of Sciences]. Kazakh Academy of Sciences

Press, Alma-Ata. 42 pp. (In Russian).

DUBROVIN, N. F. 1980. [Nikolay Mikhailovich

Przewalsky. Biographical essay]. St.

Petersburg, Military Printing House. 602 pp.

(In Russian).

GOLUBEV, M. L. 1989. [Phrynocephalus guttatus

(Gmel.) or Ph. versicolor Str. (Reptilia,

Agamidae): which Phrynocephalus species

occurs in Kazakhstan?]. Zoological News, Kiev

5:38-46. (In Russian).

KUBYKIN, R. A. 1975. [Ecological-faunistical

survey of reptiles of the Lake Alakol islands].

Transaction of the Kazakh Academy of Sciences.

Biological Series 3:10-16. (In Russian).

LEROY, P. 1940. Les Phrynocephales de

Mongolie et du N.-W. Chinois. Peking Natural

M, i,,,, n,,n..i.., i/i/")\.iir» iah

History Bulletin 14(2): 139-146.

POLYNOVA, G. V. 1989. [The new data on the

functional role of the posture "copulatory

avoidance" in the genus Phrynocephalus]. Pp.

200-210. In Questions in herpetology.

Abstracts of the Report at the Seventh All

Union Conference of Herpetology, Naukova

Dumka, Kiev. (In Russian).

POTANIN, G. N. 1948. [Field trips to Mongolia].

Geographgiz Publishing, Moscow. 480 pp. (In

Russian).

PRZEWALSKY, N. M. 1883. [The trip from

Zaissan diroughout Hami to Tibet and the upper

reaches of the Yellow River]. Balashev's Press,

St. Petersburg. 473 pp. (In Russian).

ROGOVIN, D. V. 1991. [Social behavior of

Phrynocephalus helioscopus and Ph. reticulatus

(Reptilia, Agamidae) and their relationships in

joint settlements]. Zoological Journal, Moscow

7(X3):61-72. (In Russian).

SEMENOV, D. V. 1986. [Materialen zur

Variabitat und innerartlichen systematik der

Buntkrotenkopfagame (Phrynocephalus

versicolor Sir.) in der Mongolei], Pp. 157-173.

In Herpetologissche Untersuchungen in der

Mongolischen volksrepublik. Moscow,

Academy Press. (In Russian; German

summary).

SEMENOV, D. V., Z. K. BRUSHKO, R. A.

KUBYKIN, G. I. SHENBROT. 1987.

[Taxonomic position and protected status of the

Round-headed Lizard (Reptilia, Agamidae) in the

territory of the USSR]. Zoological Journal,

Moscow 68(12):79-87. (In Russian)

December 1993

Asiatic Herpetological Research

Vol. 5, pp. 59-64

Allozyme Variation and Genetic Relationships within the Phrynocephalus

guttatus Species Group (Sauria: Agamidae) in the Former USSR.

Sergei mezhziierin and Michael l. Golubev

Institute of Zoology, Academy of Sciences, Kiev, Ukraine

Abstract. -An electrophoretic analysis of several populations of Phrynocephalus guttatus s. lato. shows

that there are two groups with a remarkable level of genetic differentiation. There is an eastern Palearctic P.

versicolor from southern Mongolia, and a western Palearctic P. guttatus s. str. which includes: g.

guttatus, g. kushackewitschii g. alpherakii, g. salenskyi, g. moltschanovii, guttatus ssp. from northern

Turkmenia and versicolor hispida from Djungar Gate. There are no objective criteria for subspecific

separation by biochemical genetic markers.

Key words: Reptilia, Sauria, Agamidae, Phrynocephalus, electrophoresis, systematics.

40

FIG. 1. Distribution of the Phrynocephalus guttatus species group in the former USSR. la- P. g. guttatus;

lb- P. g. moltschanovi; II- P. g. kushackewitschii; III- P. g. alpherakii; IV- P. g. salenskyi; V- P.

versicolor hispida; VI-P. guttatus ssp. The numbering of the populations is as given in Table 1.

Introduction

The agamid genus Phrynocephalus

includes some polytypic species groups.

One of the most complicated species

complexes is Phrynocephalus guttatus s.

lato. Representatives of this species group

are widely distributed in Middle and Central

Asia from the northern Caucasus to China.

The systematics of this species group is

highly controversial and needs revision.

There are some alternative viewpoints on

the status and systematic relationships of its

representatives. The classical viewpoint of

Terentjev and Chernov (1949) recognized

only two species: P. versicolor and P.

guttatus (P. g. guttatus and P. g.

kushackewitschii). A new concept was

developed during the last decade by

Semenov and Shenbrot (Semenov and

Shenbrot, 1982; Shenbrot and Semenov,

1987; Semenov et. al., 1987). According

to this concept, the guttatus-group consists

of four species: P. guttatus (Gmel), P.

moltschanovi Nik., P. melanurus Eichw.

(=P. salenskyi Bedr.) and P. versicolor

Str. The last form includes the nominal

subspecies (China: Alashan to Djungaria),

P. v. kulagini (Tuva, Russia; western

Mongolia) and the western Palearctic

subspecies, P. v. paraskii Semenov,

Brushko, Kubykin et Shenbrot. Golubev

(1989) lowered the status of "salenskyi" to

subspecific level, united P. v. paraskii with

P. guttatus alpherakii Bedr. and included

"moltschanovi" only as a color variation of

P. g. guttatus.

Vol. 3, p. 60

Asiatic Herpetological Research

December 1993

TABLE 1 . Localities, sample sizes and taxa of Phrynocephalus guttatus S. lato. populations

collected and investigated in this study.

N

Taxa

Locality

No.

1 P. g. guttatus

NORTHERN TRANSCAUCASUS REGION

Daghestan: Tersky Sands near Chervlenny Buruny

Russia: Stavropol Dist., Tersky Sands: Roshchino

Chencheno-Ingushety: near Starogladkovsky

N. Daghestan: sands on the right bank of Kuma River

2

3

1

A high level of morphological variation

on the one hand, and caryological

conservatism on the other, doesn't allow

one to decide problems of systematic status

and specific identity of representatives of P.

guttatus s. lato. Therefore, in order to

decide controversial systematic problems of

this agamid group, we used biochemical

genetic markers.

Methods

Electrophoretic analysis was carried out

on geographic forms of P. guttatus s. lato

from different geographic regions (Fig. 1,

Table 1). The geographic form from

northern Turkmenia was excluded from P.

g. guttatus on the basis of the red spots on

the arm pits, a very rare characteristic in P.

guttatus. We also studied two well

differentiated species, P. strauchi Nik. and

P. helioscopus saidalievi Sattorov, both

from the Fergan Valley, as an external

control for genetic differentiation.

Each adult specimen was processed in

the laboratory for blood and muscle

samples and immediately studied by

standard vertical acrylamide

electrophoresis. Homogenates obtained

from muscle, crushed in distilled water

with 5 per cent sucrose, were processed for

the following enzymes and proteins (Table

2).

Isozymes were numbered in order of

decreasing mobility from the most anodal

one. Allozymes were designated

numerically according to their mobility,

relative to the most frequent allele (100),

faster mobility (>100), slower mobility

(<100). The genetic divergence between

populations and divergence time were

estimated with indices of standard genetic

distances by formulas proposed by Nei

(1975).

December 1993

Asiatic Herpetological Research

Vol. 3, p. 61

Note: TEB- Tris-EDTA-NA2 boric acid Ph 8.5 (Peacock et al., 1965). TG- Tris-glycin, disc-

electrophoresis (Davis, 1964)

Results

Allozyme variation. — Four of 18 loci

analyzed (Es-3, Pt-2, Pt-3, IdhS) were

monomorphic and fixed for the same allele

in all populations and species considered.

Fourteen loci were polymorphic within or

between population and their allelic

frequencies are given in Table 3. Expected

genotypes distributions were in equilibrium

according to Hardy-Weinberg formula in

the investigated populations at all loci

observed. Exceptions were obtained only

in "salenskyi" and "kushackewitschii" at

the Ldh-B locus (Table 4). In our opinion,

absence of heterozygote genotypes can be

explained by introgression of Ldh-B 90

from "salenskyi" to "kushackewitschii" or

vice versa.

Levels of genetic variation are given in

Table 3 (Mean proportion of heterozygosity

observed (H obs.) and expected (H exp.)}.

H obs. ranged from 0 in "salenskyi" to

0.05 in "kushackewitschii" with a mean of

0.04. This meaning of heterozygosity is

near the level usual for Reptilia (Nevo,

1984).

Genetic divergence. — Only two loci

(Me-S, Mdh-S) display fixation or

predominance of alternative alleles between

P. versicolor (Mongolia) and the P.

guttatus group. All representatives of

Western palearctic P. guttatus have

common gene pools of the loci considered.

Phrynocephalus. v. hispida is an exception

and has alternative allelic fixation at Es-2

(Table 3). Genetic distances among P.

guttatus forms are rather low and range

from 0 to 0.05 i.e. on interspecific level of

differentiation.

There are four loci which display

alternative fixation between P. strauchi and

P. guttatus. It shows a clear intraspecific

level of genetic divergence. The largest

genetic distance is between P. h. saidalievi

and P. guttatus species group (D=0.832).

Differences between these species include 8

loci which display alternative allelic fixation

(Pgm, Mdh-S; Aat-S, Me-S, Alb; Pt. 1;

Ldh-B, Es-2).

From the allelic frequencies at 18 loci

tested, we calculated Nei's genetic distance

and constructed a matrix of genetic

distances (Table. 5). A UPGMA

phenogram was calculated on the basis of

this matrix. This reflects the relationships

Vol. 3. p. 62

Asiatic Herpetological Research

December 1993

TABLE 3. Allelic frequencies.

s-Me

-100

90

91

98

100

102

105

108

1.00

0.20

0.53

0.14

0.11

0.83

0.13

0.04

1.00

0.50

0.40

0.10

1.00

Pgm

95

100

102

105

1.00

0.97

0.03

1.00 1.00

1.00

0.17

0.83

1.00

6-pgdh

Es-2

80

85

88

90

98

100

103

108

110

112

1.00

0.03

0.01

0.96

0.04

0.96

0.08

0.04

1.00 0.88 1.00

0.08

0.32

0.60

0.05

0.87

0.08

0.03

0.94

0.03

1.00

Under the electrophoretic condition used, the following loci are monomorphic: s-Idh, Es-3, Pt-2,

Pt-3. See table 1 for population numbers.

between the P. guttatus s. lato

representatives and the other two species

that we studied (Fig. 2).

Discussion

Two distinctive gene pools.

differentiated from one another only by two

diagnostic loci, were found between

representatives of P. guttatus s. lato and P.

versicolor (southern Mongolia). The

genetic differentiation corresponds to the

division of eastern Palearctic (P. versicolor

from southern Mongolia) and Western

December 1993

Asiatic Herpetological Research

Vol. 3, p. 63

TABLE 4. Distribution of genotypes at the Ldh-A locus in populations of different geographiac

forms of toad agamas of the Phrynocephalus guttatus group.

Note: O- observed distribution; E- expected distribution; *p < 0.05; **p < 0.001

TABLE 5. Matrix of genetic distances (D, Nei, 1975) among the taxa P. guttatus s. lato, P.

strauchi, and P. helioscopus saidalievi.

Palearctic forms (P. guttatus) which

diverged around 500,000 years ago (Late

Pleistocene).

P. guttatus consists of conspecific

forms, diverse morphologically, but

conservative on the molecular level. In this

species, the more differentiated form is P.

v. hispida. This is supported by the

fixation of Es-2 (94) which is absolutely

absent in P. v. versicolor.

The level of genetic differentiation of P.

guttatus s. lato from P. strauchi and P.

helioscopus saidalievi is higher and

corresponds to good species which

diverged about 1,500,000-2,000,000 years

ago, i.e. in Middle or Early Pleistocene.

On the basis of the data we obtained, our

main conclusion is that P. guttatus s. lato.

consists of two groups with a remarkable

level of genetic differentiation. There is an

eastern Palearctic P. versicolor from

southern Mongolia, and a western

Palearctic P. guttatus s. str.* which

includes: g. guttatus, g. kushackewitschii

g. alpherakii, g. salenskyi, g.

moltschanovii, guttatus ssp. from northern

Turkmenia and versicolor hispida from

Djungar Gate. There are no objective

criteria for subspecific separation by

biochemical genetic markers.

*This abbreviation, which as generally

known means "sensu stricto", was

erroneously deciphered as "s. Strauch"

(Mezhzherin and Golubev, 1992). This

somewhat distorted the intended meaning.

Literature Cited

DAVIS, B. J. 1964. Disc-electrophoresis. 2.

Method and application to human serum

Vol. 3. p. 64

Asiatic Herpetological Research

December 1993

-/h

-//-

■ P. g guttatus

*• P. g. moltschanovi

V-P.g.spp.

1 — P. v. paraskivi

P. g. salenskyi

P. v. hispida

P. g. kushackewitschii - I

P. g. kushackewitschii - II

P. v. versicolor

P. strauchi

P. helioscopus saidalievi

-//-

0.8 0.7 0.6 0.3 0.2

D ( Nei, 1975)

0.1

0

FIG. 2. UPGMA Phenogram of relationships among Phrynocephalus guttatus s. lato.

proteins. Annals of the New York Academy of

Sciences 121:404-408.

GOLUBEV, M. L. 1989. [Phrynocephalus guttatus

(Gmel. ) or Ph. versicolor Str. ( Reptilia,

Agamidae): which Phrynocephalus species

occurs in Kazakhstan?]. Zoological News, Kiev

(5):38-46. (In Russian).

MEZHZHERIN, S. V. AND M. L. GOLUBEV. 1992.

Allozyme variation and genetic relationships

among Phrynocephalus guttatus species group

(Agamidae) of the former USSR fauna. 1st

Asian Herpetological Meeting, 15-20 July,

1992, Huangshan, China. Abstract Book:52.

NEVO E., A. BEILES AND R. BEN-SHLOMO. 1984.

The evolutionary significance of genetic

diversity: ecological, demographic and life

history correlates. Lecture Notes in

Biomathematics 53:13-213.

PEACOCK, A. C, S. L. BUNTING AND K. G.

QUINN. 1965. Serum protein electrophoresis

in acrylamide gel: patterns from normal human

subjects. Science 147:1451-1455.

SEMENOV, D. V. AND G. I. SHENBROT. 1982.

[On species independence of Phrynocephalus

moltschanovi (Reptilia, Agamidae)].

Zoological Journal, Moscow 61(8):1 194-1204.

(In Russian).

SEMENOV, D. V. , Z. K. BRUSHKO, R. A.

KUBYKIN AND G. I. SHENBROT. 1987.

[Taxonomic position and protective status of the

Round-headed Lizard (Reptilia, Agamidae) in the

territory of the USSR]. Zoological Journal,

Moscow 68(1 2):79-87. (In Russian) .

SHENBROT, G. I. AND D. V. SEMENOV. 1987.

[Present distribution and taxonomy of

Phrynocephalus guttatus (Reptilia, Agamidae)].

Zoological Journal, Moscow 66(2):259-272.

(In Russian).

TERENTYEV, P. V. AND S. A. CHERNOV. 1949.

[Guide to reptiles and amphibians], Soviet

Sciences Publishing, Moscow. 315 pp. (In

Russian).

December 1993

Asiatic Herpetological Research

Vol. 5

pp. 65-73 1

Geographic Variation and Diversity in Three Species of Phrynocephalus

the Tengger Desert, Western China

YlTEZHAO WANG1 AND HUIZHAO WANG2

'Chengdu Institute of Biology, Accidentia Sinica, Chengdu, Sichuan, China

^ Chengdu Library. Academia Sinica, Chengdu, Sichuan, China

Abstract. -Univariate and multivariate statistical analysis of 5 merisuc characters, 4 metric characters and

8 ratio characters recorded for 9 samples of lizards were used to assess non-geographic variation and

geographic variation in the southeast area of the Tengger Desert. A cluster analysis indicated that there were

three major groups which represented Phrynocephalus versicolor, P. pnewalskii, and P. frontalis, based on

morphological characters, respectively. The cluster analysis and canonical analysis showed that among

samples, phenetic similarity was not always predicted by geographic proximity. Dispersal and divergence

of these species of Phrynocephalus with the relationships of paleogeography and paleoclimatology are

discussed. It is evident that the Yellow River fails to cause geographic isolation.

Key words: Reptilia, Sauria, Agamidae, Phrynocephalus, China, biogeography.

Introduction

Strauch (1876) described four new

species of Phrynocephalus from the Altan

Desert (including the Tengger Desert) and

the Mu Us Desert (Ordos). They were P.

pnewalskii, P. versicolor, P. affinis, and

P. frontalis. He pointed out that P. affinis

was very similar to P. pnewalskii. Pope

(1935), with out any discussion,

considered P. affinis to be synonymous

with P. pnewalskii. Leroy (1939), who

studied the geographic variation and

distribution of P. pnewalskii, P. frontalis,

and P. vlangalii, failed to recognize P.

frontalis and P. versicolor, which were

quite different on morphological characters.

Therefore, his distribution of P. frontalis

included P. versicolor . Zhao (1979)

deduced that the distribution of

Phrynocephalus in China almost reached

the shore of the Bohai Sea and in the Altan

Desert (including the Tengger Desert) no

distribution for P. frontalis was mentioned.

In this paper we discuss (1) the

geographic variations and distributions of

P. pnewalskii, P. frontalis, and P.

versicolor, and (2) the dispersal and

divergence tracts of the three species in the

Tengger Desert.

Materials and Methods

Study Areas. — The Tengger Desert is

situated in western Nei Mongol

Autonomous Region, China. The east edge

reaches the Helan Mountains, and the

southern edge, the Yellow River. In the

west and north, the study area connects

with the Badain Jaran Desert and the Ulan

Behou Desert, respectively. The climate in

these areas is arid-continental. The

topography is moving sand dunes and

gobi. In this area the vegetation is scarce.

Ammopiptanthus sp., Potaninia sp.,

Teraena sp., and Caragana sp. are present.

Populations of P. pnewalskii, P. frontalis,

and P. versicolor were sampled from nine

localities in the southeast area of the

Tengger Desert from the south and north

shore of the Yellow River, through

southeast Tengger Desert to Hala Woo

Valley of the Helan Mountains, and in the

west at Xial Hong Shan (Fig. 1 and Table

1).

Methods. — Five meristic characters,

four metric characters and eight ratio

characters were chosen for study. These

characters are shown in Table 2.

Statistical analysis. — For analysis of

geographic variation, lizards were grouped

into nine samples (Fig. 1 and Table 1)

within which gene flow was assumed to

occur freely. Sexual variation in metric

characters was examined in sample 1

(representing P. pnewalskii), and samples

2 and 3 (representing P. frontalis). We

Vol. 3, p. 66

Asiatic Herpetological Research

December 1993

TABLE 1 . Locality data for Phrynocephalus from the Tengger Desert, western China used in this study.

100 km

l

TENGGER

DESERT

FIG. 1. Locations in the Tengger Desert for the nine samples of Phrynocephalus used in the study of

geographic variation (see Table 1 for exact localities).

determined that individuals of the three

species are reproductively mature at an SVL

of 45 mm or larger.

The ANOVA test was used to test for

differences between the sexes in samples 1,

2, and 3. Standard univariate statistics

were calculated for these three samples and

to examine geographic variation in single

meristic characters for each sample. For all

December 1993

Asiatic Herpetological Research

Vol. 3, p. 67

TABLE 2. Characters used in analysis of variation

in three species of Phrynocephalus.

A. Meristic Characters

TABLE 3. Sexual variation in three samples of

Phrynocephalus (sample number as in Fig. 1 and

Table 1. Levels of significance are indicated.

(*=p<0.05: **=p<0.01)

1. NSPL No. of supralabials

2. NIFL No. of infralabials

3. NSDT No. of subdigital lamellae on longest toe

4. NSNT Number of scales between nostrils

5. NDVT No. of ventral dark bands on tail

B. Metric Characters

6. SVL Snout-vent length

7. TL Tail Length

8. LFL Foreleg length (including fingers)

9. LHL Hindleg length (including longest toe)

C. Ratio characters

* Head length (rostral up to and including

parietales)/Head width (front of ears).

** Length from nostril to eye/Length between

nostrils.

samples, meristic characters were calculated

and standard values from the matrix of

intersample phenotypic distances were

clustered with the unweighted furthest-

neighbor method using arithmetic average.

To overcome some disadvantages of the

clustering, the multivariate analysis of

variance (MANOVA) and canonical

analysis were used to provide weighted

combinations of characters to analyze

variations of the nine samples. A set of

canonical was calculated and the mean

values of each sample were plotted on the

first two axes. Additionally, the relative

contribution of each character in Table 4 to

each of these two axes was calculated,

which collectively accounted for over 86%

of the total variation. All calculations were

performed on an IBM-PC/XT computer at

the Chengdu Library, Academia Sinica by

means of FORTRAN Program II.

Results

Sexual variation (non-geographic

variation). — Results for intersex

comparisons (adults only) in three samples

are summarized in Table 3. In ANOVA the

sexes from sample 1 differed significantly

in SVL. In sample 2, the metric character

LHL showed significant sexual

dimorphism . In sample 3, three metric

characters, SVL, LFL, and LHL showed

significant sexual dimorphism.

Multivariate analysis. — Fig. 3 presents a

distance phenogram, based on 5 meristic

characters, clustering 9 OTUs which

correspond to the group sample localities

(Fig. 1 and Table 1) used in this study.

Vol. 3, p. 68

Asiatic Herpetological Research

December 1993

-5-4-3-2-1012345

FIG. 2. Projections on tlie first two canonical

vectors of centroids representing the nine samples

of Phrynocephalus. The number of centroid

corresponds to the localities in Fig. 1.

The cophenetic correlation of the

phenogram with the distance matrix was

0.912. The first major dichotomy separates

sample 9 from all others, and the second

major dichotomy groups samples 1 and 7.

The third major dichotomy groups samples

2, 3, 4, 5, 6, and 8. Most of the clusters

group samples that do not have any

geographic affinities or relationships. For

example, samples 3 and 9 in fairly close

geographic proximity (about 10 km

separate each other) were separated by

different dichotomies in the phenogram,

conversely, samples 2, 8, and 3 clustered

with one another, despite being separated

by over 150 km.

Univariate analysis (geographic-

variation). — Standard statistics are

presented for two meristic characters. Fig.

4 and Fig. 5 depicts geographic variation in

NSPL and NIFL counts for nine samples

(males only). The mean NSPL counts of

samples 1 and 7 were 17.50 and 16.00,

respectively. For samples 2, 3, 4, 5, 6,

and 8, the mean NSPL counts were

between 13.00 and 14.90, and for sample 9

was 15.00 (Fig. 4). Fig. 5 shows NIFL

counts for all nine samples. Samples 1 and

7 show higher NIFL counts, 17.10 and

5-

4"

453826179

Sample

FIG. 3. Distance phenogram resulting from the

cluster analysis with five meristic characters of nine

samples (each considered an OTU) of

Phrynocephalus. The cophenetic correlation

coefficient was 0.912.

16.00 respectively, than those of samples 2

(15.00), 3 (14.80), 4 (14.10), 5 (14.00), 6

(13.20), and 8 (14.00), as well as 9

(14.00).

In the MANOVA, the overall tests of the

hypothesis of no effect due to geography

were rejected (P < 0.001) for Wilks'

Criterion, and Roy's Maximum Root

Criterion. The first two canonical vectors

extracted from the variance-covariance

matrix accounted for 77.58 % and 8.67 %

of the total variation. The samples are

plotted along these two vectors in Fig. 2

and Table 4 summarizes the percent

influence of each character to each of the

two vectors.

Three major groups were discernible in

Fig. 2, the most strongly differentiated of

which separated primarily along vector I

and to some extant along vector II. The

second group in clustering (samples 1 and

7) was separated along vector I, which also

separated by high NSPL and NIFL counts

in Figs. 4 and 5, and form a distinct

phenetic cluster in Fig. 3. The first group

December 1993

Asiatic Herpetological Research

Vol. 3, p. 69

TABLE 4. Variable coefficients for canonical variates I and II and estimated % influence of each vector for

nine samples of Phrynocephalus.

Vector I (77.58%)

Vector II (8.67%)

Character

Variable

coefficients

% influence

Variable

coefficients

% influence

NSPL

0.332

8.040

0.715

4.430

NIFL

NSDT

0.355

0.871

9.780

58.862

0.715

-0.450

4.430

1.763

NSNT

NDVT

0.035

-0.093

0.097

0.007

-0.245

-0.036

0.011

0.001

SVLTTL

SVL/LFL

SVL/LHL

LNE/LN

LFL/LHL

-0.008

0.026

-0.004

0.032

-0.001

0.000

0.001

0.000

0.001

0.000

-0.004

-0.043

0.005

-0.010

0.001

0.000

HL/HW

TL/LFL

TL/LHL

0.003

0.007

0.023

0.000

0.009

-0.003

0.031

0.000

TABLE 5. Characters for three species of population-groups of Phrynocephalus.

Character

SVL

TL

LFL

LHL

NSNT

NSPL

NIFL

NDVT

NSDT

P. frontalis

Range (Mean)

P. przewalskii Range

(Mean)

P. versicolor

Range (Mean)

45.8-57.2(51.5)

50.7-71.4(61.6)

46.1-55.6(49.5)

58.4-82.1 (72.4)

76.1-101.5(86.3)

53.9-69.4 (60.8)

23.3-29.5 (26.1)

27.4-35.9 (31.4)

22.7-26.9 (24.4)

37.1-48.7(43.1)

43.7-59.6(51.1)

36.5-43.4 (39.1)

2-4(3.2)

2-4 (2.9)

2-4 (2.0)

11-17 (14.9

15-20(17.6)

13-17(15.0)

12-17(15.1)

14-20(17.1)

12-16(14.0)

2-5 (2.8)

1-4(2.2)

0-3 (2.0)

Color on ventral surface of tail tip

Color on armpits

24-30 (26.4)

27-34 (29.9)

21-26(22.0)

dark

grey-white

Color on back of body

No. of teeth on maxillary

No. of teeth on dentary

2-5 pairs of black spots

irregular spots

longitudinal line

2-4 transverse black-

reddish bands

11

10

11

12

11

Ishium

Cartilage

Posterior 1/2 is cartilage

Meckel's cartilage

Covered by splenial Covered by splenial

Native not covered by

splenial

in clustering (sample 9) was separated

along vector II and vector I. The third

group, represented by samples 2, 3, 4, 5,

6, and 8 in clustering, was plotted along the

centroid in Fig. 2.

Characters having the greatest influence

to vector I and vector II are NSDT, NSPL,

and NIFL, respectively.

Only males were used in univariate and

multivariate analysis for geographic

variation.

Vol. 3, p. 70

Asiatic Herpetological Research

December 1993

12 3 4 5 6 7 8 9

Localily

FIG. 4. Diagram depicting geographic variation

among nine samples of Phrynocephalus in

supralabial number. Verticle line represents sample

mean; open and closed bars represent range and one

standard deviation, respectively.

Discussion

According to the morphological

characters, populations of all nine samples

can be divided into three major groups as

shown in Fig. 2, 3, and Table 5. The first

group (sample 9), a second group (samples

1 and 7), and a third group (samples 2, 3,

4, 5, 6, and 8). These are very similar to

the descriptions by Strauch (1876) of P.

versicolor, P. przewalskii and P. frontalis,

respectively.

It is apparent that P. frontalis is

distributed in the Tengger Desert. This was

not mentioned by authors who had studied

the distribution of P. frontalis before.

Strauch (1876) had pointed out that the

color patterns of Phrynocephalus were very

variable. In our study, we found that the

color on the armpits of the group of P.

frontalis and P. versicolor in the Tengger

Desert was grayish-white. This was

different from P. frontalis in the Mu Us

Desert which has reddish on the armpits,

and P. versicolor with reddish-blue on the

armpits at Anxi and Wuwui in Gansu

Province. In 142 specimens of P. frontalis

in the Tengger Desert, only 4 specimens

had reddish on the armpits. Therefore, we

suggest that the color on armpits could not

be a character for taxonomy in the genus

Phrynocephalus.

In the group of P. frontalis, sample 2 is

!s 2(H

ca

= 15i

10

* 1

1 1

8

ill

Locality

FIG. 5. Diagram depicting geographic variation

among nine samples of Phrynocephalus in

infralabial number. Verticle line represents sample

mean; open and closed bars represent range and one

standard deviation, respectively.

located on the south shore of the Yellow

River, but samples 3, 4, 5, 6, and 8 are

located on the north side of the Yellow

River. They are very similar in

morphological characters and they belong

to the same species, P. frontalis. It is

evident that the Yellow River does not

function to geographically isolate P.

frontalis. According to information on

paleogeography and paleoclimatology, in

Late Tertiary the Tengger Desert began to

form . The area was very dry and cool.

The Yellow River began to form in the

Pleistocene with the uplift of the Qinghai-

Xizang Plateau (Li, 1984). This means that

the Tengger Desert was forming earlier than

the Yellow River. Additionally, the Yellow

River in this area had shifted its route

several times from west to east during the

past. We suggest the Phrynocephalus

might have invaded and dispersed into the

Tengger Desert before the formation of the

Yellow River and that the river fails to be a

geographic barrier for these lizards.

Although the group of P. przewalskii

located on the north side of the Yellow

River is the nearest neighbor of the group

of P. frontalis, they are quite different from

each other on morphological characters. It

is believed that this is the result of

interspecific isolation.

According to the study of morphological

and skeletal characters of P. versicolor, P.

przewalskii and P. frontalis, P. versicolor

was quite different from the other two

December 1993

Asiatic Herpetological Research

Vol. 3, p. 71

60°

80°

100°

120°

r-

i

v

<m

S y

Si:

MONGOLIA

"T

yyj-:-

I.

I

"V,

""*/ '*\ ,,'V-

1 P. versicolor

P. przewalskii

P. frontalis

1200 km

_1

140° 50o

\

/■' 40°

30c

20 c

FIG. 6. Geographic distribution of P. versicolor, P. przewalskii, and P. frontalis.

(Table 5; Wang, 1987), and showed the

characters replacement. For example, the

pro-half part of the ischiopubis in P.

versicolor is ossified and the post-half part

still is cartilage. The ischiopubis in P.

przewalskii and P. frontalis, conversely, is

all cartilage. Additionally, the Meckle's

cartilage is not covered by the splenial in P.

versicolor but it is covered by the splenial

in P. przewalskii and P. frontalis.

Electrophoresis also showed that P.

versicolor is different from P. przewalskii

and P. frontalis, the latter two species being

similar (R. Macey pers. comm.).

Therefore we suggest that P. versicolor

was probably derived earlier from the

ancestral stock than P. przewalskii and P.

frontalis.

The distribution of P. versicolor is from

central Asia through western China to the

central parts of Mongolia (Boblov, 1986)

and Nei Mongol (Zhao, 1978; 1975).

Phrynocephalus przewalskii occurs from

Zhang Ye, Gansu Province through the

Tengger Desert to the west edge of the

Helan Mountains. Phrynocephalus

frontalis is found from Zhang Ye (Yao,

1983) through the Tengger desert to the Mu

Us Desert (Ordos) (Schmidt, 1927) and

forward to the south in the east part of

Gansu Province. There is also an isolated

population at Xun Yang in Shaanxi

Province (Song, 1987) (Fig. 6).

Phrynocephalus is distributed to the east to

about 1 12°E but does not reach the shore of

the Bohai Sea as Zhao (1979) stated.

In the Early Tertiary, there was an arid-

subtropical continental climate belt from

central Asia through the southern edge of

the depression areas in the Tian Shan

Mountains and the north edge of the Tarim

Depression, the north part of the Qinghai-

Xizang "initial plateau", the Qaidam

Depression, depression areas in the Qilian-

Qinling Mountains, and the south edge of

depression areas in the Helan-Liupan

Vol. 3, p. 72

Asiatic Herpetological Research

December 1993

Mountains to the central part of China (Li,

1984). In some areas, desert had been

forming and the ancestral stock of the

Agamidae might have followed this belt to

invade the areas of the south side of the

depression areas in the Helan-Liupan

Mountains from central Asia through

western China. As evidence, some fossils

of Agamidae were discovered from the

Eocene of western and central China (Li,

1984). We suggest that P. versicolor might

have derived from the ancestral stock

during that time. After the Late Tertiary

and Early Quaternary, with the uplift of the

Qinghai-Xizang Plateau, this arid-

continental climate belt shifted forward to

northern regions, including the Tengger

Desert (Li, 1984). The ancestral stock of

Phrynocephalus and P. versicolor might

have followed the shift of this belt again to

disperse into northern regions, including

the Tengger Desert and P. przewalskii and

P. frontalis derived from ancestral stock in

the Tengger Desert during that time. Then

P. frontalis dispersed further east to the Mu

Us Desert (Ordos) and south part of the

Qinling Mountains. According to the

distribution of P. frontalis, it is obvious

that it had a wider range in the past.

Phrynocephalus przewalskii and P.

frontalis show sexual dimorphism.

According to observations in the breeding

season on mating behavior, males run after

females quickly for mating (Song et al.

1987). We deduced that the individual

males with long legs would have more

successful opportunities for mating with

females than males with shorter legs. We

suggest that the sexual dimorphism

between males and females was the result

of sexual selection.

Acknowledgments

We thank Prof. Ermi Zhao, Prof.

Yaoming Jiang, Prof. Liang Fei and Mr.

Jarging Pam for critical review of the

manuscript. We also thank Dr. Theodore

J. Papenfuss, Mr. Robert Macey and Mr.

Keller Autumn for their support of this

work. Also, we want to thank Dr. Erhu

Sun for helping us translate Russian papers

into Chinese.

Literature Cited

BEDRIAGA, J. V. 1907. Wissenschaftliche

Resultate der von N. M. Przewalski nach

central-Asien Unternommenen Reisen. Annals

of Zoology. Museum of the Imperial Academy

of St. Petersbourg. Zoologisher vol. Ill pp.

179-278. (In Russian with German translation).

BOBLOV, V. V. 1986. [Zoogeographical analysis

of herpetofauna in Mongolia]. Symposium of

Herpetological Research. USSR Academy of

Sciences, Moscow. Pp. 85-119. (In Russian).

LEROY, P. 1939. Phrynocephalus de Mongolie et

du N-W Chinois. Peking Natural History

Bulletin 14(2):139-145.

LI, Y. 1984. [The Tertiary system of China].

Science Press, Beijing, pp. 278-338. (In

Chinese).

NIKOLSKI A. M. 1915. Fauna of Russia and

adjacent countries. Reptiles, Vol. 1 Chelonia

and Sauria. Museum of Imperial Academy of

Sciences. St. Petersbourg. pp. 93-153.

(Translated from Russian, 1963, by Israel

Program for Scientific Translations.

Jerusalem).

POPE, C. H. 1935. The reptiles of China.

Natural History of Central Asia X. American

Museum of Natural History, New York. 604

pp.

SICHUAN INSTITUTE OF BIOLOGY. 1979. [A key

to Chinese reptiles]. Science Press, Beijing.

27 pp. (In Chinese).

SCHMIDT, K. P. 1927. Notes on Chinese

reptiles. Bulletin of the American Museum of

Natural History 44:483-485.

STRAUCH, A. 1876. [Mongolia and the Tangut

Region]. Opisan. Przesmykajuschtsch.

Semnowodn., Przewalskago Mongol. Strana

Tangut. Vol. 1. Part III. pp. 3-26. (In

Russian).

SONG, M. 1987. [Survey of the reptiles of

southern Shaanxi]. Acta Herpetologica Sinica

6(l):59-64. (In Chinese).

SONG, M. 1987. [The herpetofauna of Shaanxi

Province]. Acta Herpetologica Sinica 6(4):63-

73. (In Chinese).

December 1993

Asiatic Herpetological Research

Vol. 3, p. 73

SONG, A., L. CHEN, AND Q. CHEN. 1987.

[Studies on the breeding habits of

Phrynocephalus przewalskii]. Acta

Herpetologica Sinica 2(3):63-73. (In Chinese).

WANG, Y. 1987. [The skeletal anatomy of four

species oiPhrynocephalus, with a discussion on

their interspecific relationships]. Acta

Herpetologica Sinica 6(4):27-34. (In Chinese).

YAO, C. 1983. [Lizards of Gansu Province].

Acta Herpetologica Sinica 2(3):66-67. (In

Chinese).

ZHAO, K. 1978. [An investigation of amphibians

and reptiles in Nei Menggu]. Acta Nei Menggu

University 2:65-69. (In Chinese).

ZHAO, K. 1979. [A survey of the classification

and distribution of the Toad-headed Agamids

(Phrynocephalus) in China]. Acta Nei Menggu

University 2:111-121. (In Chinese).

ZHAO, K. 1985. [An investigation on the lizards

of Xinjiang Uygur Autonomous Region]. Acta

Herpetologica Sinica 4(l):25-29. (In Chinese).

Vol. 5, pp. 74-84

Asiatic Herpetological Research

December 1993

Sympatric Amphibians of the Yew-box Grove, Caucasian State Biosphere

Reserve, Sochi, Russia

BORIS. S. TUNIYEV AND SVETLANA. Yll. BEREGOVAYA

Caucasian State Biosphere Resen>e, Sochi, Russia

Abstract. -The Yew-box Grove of the Caucasian State Biosphere Reserve is home to seven species of

amphibian. These species occur in a wide range of aquatic environments. The species composition,

physical characteristics and history of each aquatic site was evaluated. The reproductive biology and food

habits of each species was studied. These amphibians divide their niches on daily activity, seasonal

activity, breeding site, microhabitat and food habits. The highest amphibian diversity and species overlap

occurs in the most stable aquatic environments.

Key Words: Amphibia, Russia, Caucasus, ecology.

Introduction

It is important when studying the

influence of environmental factors on life

history characteristics to distinguish those

factors that are significant and those that are

part of the "neutral background"

(Monchadsky, 1958). The study of

environmental influences is accomplished

on sympatric species, usually closely

related species (Orr and Maple, 1978;

Ananjeva, 1981) at the population level

(Pianka, 1973; Schoener, 1974; Schoener,

1977; Lyapkov and Severtsev, 1981). This

is the study of the ecological niche (Pianka,

et al., 1979.

In the former Soviet Union one of the

areas of highest amphibian diversity is

found in the western Caucuses. The Yew-

box Grove of the Caucasian State

Biosphere Reserve is inhabited by eight

amphibian species: Triturus vulgaris lantzi,

T. cristatus karelini, T. vittatus ophryticus,

P elodyte s caucasicus, Bufo

verrucosissimus, Hyla arborea

schelkownikowi, Rana ridibunda, and R.

macrocnemis.

Methods

Field studies were conducted from 1980-

1982 in the Yew-box Grove (approximate

area 302 ha) in the Caucasian State

Biosphere Reserve and on adjacent land.

Transect routes and study sites were

selected on the basis of local forest

typography (Gulisashvili, et al., 1975).

Observations on these study sites along the

FIG. 1. Study sites of sympatric amphibians in the

Yew-box Grove, Caucasian State Biosphere

Reserve. 1- Spring 118; 2- Opolznevaya Ravine:

3- Labirintovaya Ravine; 4- Glubokaya Ravine; 5-

Khosta River; 6- Samshit Pond; 7- Pond on

Malaya Khosta River.

transects were made throughout the year

(Fig. 1). The intensity of calls was

recorded in spring (2-3 times per week),

summer (1-2 times per month) and fall (2-3

times per week). Over 200 individuals

were recorded.

December 1993

Asiatic Herpetological Research

Vol. 5 p. 75

iy y yi yn yui ix x xi xn

Note :

- river Khosta

- - spring 118

- ravine Labirintovaya

- ravine Glubokaya

FIG. 2. Mean monthly temperatures of perennial

bodies of water in the Yew-box Grove (1 980-1982).

2C

10

0

91

95

i n II iy y yi yn yn ix x xi xn

ioo

humidity

FIG. 3. Climatram of the Yew-box Grove (1982).

The location, weather conditions, air and

body temperature, and behavior of each

specimen was recorded. During 1982,

detailed microclimate records were made in

at the depth of 5 cm

at the depth of 1C cm

at the depth of 15 cm

■ Bt the depth of c'O cm

FIG. 4. Winter soil temperature in the Yew-Box

Grove (1982-1983).

the Yew-box Grove (Fig. 2). Air

temperature and humidity above the surface

were recorded weekly by a thermograph

(M-16AN) and a hydrograph (M-16AN)

placed in a meteorological kiosk (Fig. 3).

Soil temperatures were recorded at 5, 10,

15 and 20 cm depths (Fig. 4).

Water samples were taken periodically

for chemical analysis by established

methods (Anonymous, 1978). Ambient

light was measured with a light meter and

converted into percent relative illumination.

Climagrams were made (Formozov, 1934).

Species composition of each biotope was

calculated using conventional methods

(Kashkarov, 1927; Dinesman and

Kaletskaya, 1952). Food habits were

studied using non-lethal methods

(Verzhutsky and Zhuravlev, 1977).

Results

Description of Study Sites

Study Site 1. Spring 118. This site

contains a small perennial stream running

Vol. 5 p. 76

Asiatic Herpetological Research

December 1993

TABLE 1 . Some hydrochemical indices of the study sites.

through a sub-tropical, mixed broad-leafed

forest (Fagus orientalis, Taxus baccata,

Carpinus betulus) with an evergreen

understory (Buxus colchicus, Ilex

colchicus, Laurocerasus officinalis) and

lianas (Hedera colchica, Smilax excelssior).

Relative illumination is 1-2%. The stream

flows over a bed composed of clay and

sandstone. The stream flow derives from

runoff and a sub-surface aquifer. The

water chemistry of the spring water was

hydrocarbonic-calcic with moderate

mineralization, and moderately hard (Table

1). Hydrogen ion concentration is neutral

to slightly basic, pH ranges from 6.89-

8.50. The concentration of nitrogen and

nitrates (0.97-2.67 mg/1) is higher than in

other waterways. This is a result of the

subterranean flow. Ammonium

concentration is normally low and increases

during flash floods (up to 0.45 mg/1).

Nitrites are also found in low

concentrations except during flash floods

(up to 0.08 mg/1). Phosphorus

concentration is considerably higher than in

other waterways (up to 0.16 mg/1). Rana

macrocnemis, R. ridibunda, Bufo

verrucosissimus, Hyla arborea

schelkownikowi , Pelodytes caucasicus,

and Triturus vittatus ophryticus are found at

this study site and the latter two species

breed there (Fig. 5, Table 2).

Study Site 2. Opolznevaya Ravine. A

small intermittent waterway flows through

an eroded ravine through carbonic rock and

clay. The vegetational community is

analogous to Study Site 1, but box yews

and beeches are dominate. Relative

illumination is 2%. Stream flow is derived

from runoff and aquifer. The water

chemistry of the spring water was

hydrocarbonic-calcic with moderate

mineralization, and soft. Hydrogen ion

concentration is neutral to slightly basic,

pH ranges from 8.0-8.27. No amphibians

were observed at this study site.

Study Site 3 . Labirintovaya Ravine. A

small intermittent, vernal-autumnal stream

flows through an eroded, karst gorge with

steep walls in a box yew forest. Relative

illumination is 2%. Stream flow is derived

from runoff and subsurface flow. During

low waters periods the stream falls into a

number of stagnant pools. The water

chemistry of the spring water was

hydrocarbonic-calcic with moderate

mineralization, and soft. Hydrogen ion

concentration is neutral to slightly basic,

pH ranges from 7.3-8.45. Water content

of nitrogen compounds is low but increases

during flash floods. R. ridibunda is found

here at this study site and P. caucasicus

reproduces here.

Study Site 4. Glubokaya Ravine. This

site is a small pond in a limestone gorge

with steep walls and surrounded by a box

yew forest. Relative illumination is 2-3%.

The pond is fed by a small, relatively

constant spring flowing from Karst.

December 1993

Asiatic Herpetological Research

Vol. 5 p. 77

FIG. 5. Distribution of sympatric amphibians at Spring 118 in the Yew Box Grove. 1- Triturus vittatus

ophryticus; 2- Pelodytes caucasicus; 3- Rana ridibunda; 4- Hyla arborea schelkownikowi; 5- Rana

macrocnemis; 6- Bufo verrucosissimus.

Spring temperature ranges from 11-16° C.

The water chemistry of the spring water

was hydrocarbonic-calcic with moderate

mineralization, and moderately hard (2.42 -

3.6 mg-eq/1). Ammonium concentration is

low (maximum = 0.04 mg/1) and in winter

drops to 0. Nitrates are highest during

summer low water period (up to 0.37

mg/1). Phosphorus concentration is very

low (>0.01 mg/1). Bufo verrucosissimus,

R. macrocnemis, P. caucasicus, R.

ridibunda and T. v. ophryticus are found

living at this study site and the last three

species breed here.

Study Site 5. Khosta River. This site is

a small mountain stream 21 km long that

drains a watershed of approximately 96

km2. The stream flows through a canyon

formed in Cretaceous limestone at an

average of 5 m-Vsec. Water flow is derived

from runoff and springs arising in karst

rock formations. Vegetation is a broad-

leafed, subtropical Colchis type forest.

Relative illumination is as high as 100%.

The water chemistry of this stream is

hydrocarbonic-calcic with moderate

mineralization and basic pH (7.7-8.5).

Ammonium concentration is not high and

varies from 0 to 0.07 mg/1, but during

floods it can reach 2.32 mg/1. This stream

is subject to occasional flooding.

Concentration of nitrates reaches a

maximum during floods of 1.44 mg/1.

Dissolved oxygen is 10-15mg/l and

carbonic acid is 10mg/l.

The Nizhe-Khostinsky Spring, flowing

out of a karst formation, contributes 1-1.5

m-Vsec of flow at 11-13° C to the Khosta

River. In low water periods the Khosta

River above the spring nearly dries up and

its temperature ranges from 0.6 (the river

freezes) - 26° C. During low waters on the

Khosta River the Nizhe-Khostinsky Spring

provides a relatively stable flow and

temperature regime. This spring serves as

a barrier for dispersal of amphibians at this

study site. Rana ridibunda and Hyla

arborea schelkownikowi live and reproduce

above the spring. Pelodytes caucasicus and

Bufo verrucoisissimus live and breed

below the spring.

Study Site 6. Samshit Pond. This site

is a small pond with flowing water in a

stand of hornbeam {Carpinus betulus) in

the broad-leafed forest. Relative

NOTE: X- reproduce, O- inhabit

illumination is 100%. The pond flow is

feed by run off and flow from the aquifer.

The water chemistry of this pond is

hydrocarbonic-calcic with moderate

mineralization, moderately hard and

alkaline. Hyla arborea schelkownikowi,

Triturus vulgaris lantzi, Triturus vittatus

ophryticus, Rana ridibunda, and R.

macrocnemis, are found living at this study

site and the last three species breed here.

Study Site 7. Pond on the Malaya

Khosta River. This site is a small stagnant

pond located in the flood plain forest. The

pond is filled by runoff and flow from the

aquifer. Relative illumination is 50%.

Bufo verrucosissimus, T. vulgaris lantzi,

T v. ophryticus ,H. a. schelkownikowi

and R. macrocnemis, are found living at

this study site and the last two species

breed here.

Species Accounts

Triturus cristatus karelini. This newt is

an extremely rare and declining species

along the Caucasian Black Sea coast. It

was observed only once, at Study Site 1, in

the box yew forest.

Triturus vulgaris lantzi. This species

exclusively inhabits stagnant ponds and

ponds with flowing water, in well

illuminated stands of hornbeam in broad-

leafed forests, and adjacent areas. These

newts are found in the ponds beginning in

early March. Reproductive activity begins

when water temperature reaches 10° C

(usually from mid-March to early April.

Females lay their eggs in shallow,

thoroughly warmed waters at a depth of 5

cm and remain in the pond until the end of

June. The newts over-winter in forest leaf

litter and underground (Table 3).

Triturus vittatus ophryticus. This

species is found in both well illuminated

broad-leafed flood plain forests and thick

yew-box groves. It appears in bodies of

water from the end of November through

January. Reproduction occurs from

January until the middle of April at water

temperatures of 7-9° C. Females lay their

eggs at depths of 5-10 cm. The adults stay

in the water until the end of May. The

newts over-winter in forest leaf litter.

Bufo verrucosissimus. This toad is

found throughout the Yew-box Grove with

the exception of the steeper parts of the

Khosta Canyon. Reproduction takes place

in well illuminated running water in the

Khosta River from February until May at

water temperatures from 9.5-16° C. Eggs

are deposited at a depth of 20-70 cm in

strings through vegetation and other

underwater objects. These toad over-

winter in forest leaf litter beginning in

December.

Pelodytes caucasicus. This species

inhabits back-water vegetation communities

with flowing water. Reproduction lasts

from the end of May until the end of

October at water temperatures of 13-16° C.

Females lay their eggs at a depth of 10-20

cm. These anurans over-winter in forest

leaf litter.

December 1993

Asiatic Herpetological Research

Vol. 5 p. 79

I—

U

o.

Vol. 5 p. 80

Asiatic Herpetological Research

December 1993

TABLE 4. Comparison of development periods of Triturus vittatus ophryticus and Pelodytes caucasicus.

MONTHS:

* Note: Shaded area represents the periods of mutual number limitation

Hyla arborea schelkownikowi. This

species is found in open, well illuminated

ecotones throughout the grove.

Reproduction take place from March until

October in warm (>11" C), stagnant

waters. Eggs are laid at a depth of 10-12

cm. These frogs over-winter in the forest

leaf litter.

Rana ridibunda. This frog is numerous

in Khostinsky Canyon in open, well

illuminated areas in the water-box yew

forest ecotone. Reproduction last from

January until March at water temperatures

from 5-9° C. Frogs over-winter at the

bottom of stagnant bodies of water.

Rana macrocnemis. This frog is found

in low numbers in all areas of the grove

except in the rocks. Reproduction takes

place from February to March in warm (4-

9° C), shallow water. Eggs are laid at 0-5

cm depth. During cool winters this frog

over-winters in the forest leaf litter and

during warm winter it remains abroad.

Discussion

During the summer all of the species of

amphibians in the Caucasian Biosphere

Reserve are broadly sympatric. However,

during reproduction and winter retreat there

is habitat segregation (Table 3). The

highest level of overlap occurs in the

summer in the flood plain forest where all

of the above are found. The lowest level of

overlap occurs in the box yew-cherry laurel

stands with only B. verrucosissimus, and

R. macrocnemis present. The box yew-

cherry laurel stands are the most ancient

forest type preserved in the Yew-box

Grove. This ancient forest is dominated by

box yew trees ranging from 500 to 2000

years old and has been virtually unchanged

in appearance during that time. It is

interesting that this ancient forest is

inhabited by the indigenous Caucasian

species B. verrucosissimus, and R .

macrocnemis.

The hornbeam tree is a pioneering

species that invades disturbed areas such as

those that have burned or been logged. It

also grows in barren areas. The diversity

of amphibian species in the hornbeam

forest is caused by a number of factors: 1)

secondary character of hornbeam forest; 2)

relatively higher illumination (compared to

the box yew forest; 3) presence of suitable

water conditions for reproduction. When

the hornbeam trees are young and the

habitat is still open, such species as R.

ridibunda, T. v. lantzi, and H. a.

schelkownikowi are found. Later, when

the trees become mature and the forest more

closed and less illuminated, B .

verrucosissimus, R. macrocnemis and P.

caucasicus become established. This

environment supports the highest level of

sympatry of amphibian species in the Yew-

box Grove. In box yew stands T. v. lantzi

and H. a. schelkownikowi, the most

illumination tolerant species, are not

represented. These two species appear in

December 1993

Asiatic Herpetological Research

Vol. 5 p. 81

TABLE 5. Daily activity of amphibians of the Yew-box Grove.

the ecotone adjoining the southwest

boundary of the grove (Table 3).

It is interesting to note that B .

verrucosissimus and R. macroenemis

have very specific breeding requirements in

terms of the aquatic environment required,

but they occur over a large area during the

terrestrial stages of their life history. They

can be termed stenotopic or very restricted

in terms of the reproductive requirements

(B. verrucosissimus lays its eggs in rivers

and R. macroenemis in small pools) and

eurytopic in terms of their general

distribution (Dazho, 1975). These two

species are autochthones or indigenous to

the broken country Caucasian region and

reproduce in rapid mountain stream and

ephemeral pools. The majority of lakes in

this region are of recent origin and formed

by glaciation, karst or from landslides. As

this range of aquatic environments became

available at the end of the last glacial

period, amphibians successfully colonized

those environments which met their

reproductive requirements (Monchadsky,

1958).

Bufo verrucosissimus lays its eggs in

strings and wraps them around aquatic

vegetation and other anchored objects in the

water. This allows the toad to lay its eggs

in fast flowing mountains streams which

are generally unsuitable for other species.

It lays its eggs at depths of 20 to 70 cm

thus providing some protection from flash

Hoods though many eggs perish in such

floods. Pelodytes caucasicus and R.

ridibunda are found sympatric with B.

verrucosissimus. Rana ridibunda lays its

eggs in the shallow, slow-moving sections

of the river or in pools formed by floods.

Reproduction in R. ridibunda is limited to

the short period of winter low water and

lasts from the end of January to the end of

March.

Pelodytes caucasicus is isolated from

other breeding anurans temporally. It

attaches its eggs to thin roots (2-10 cm)

beginning in mid-June, when other species

have finished breeding. Typically breeding

sites are in the backwaters of streams,

under vegetation canopies where

temperatures are moderate. In small

streams R. ridibunda and T. v. ophryticus

are sympatric with P. caucasicus .

At Glubokaya Ravine, Study Site 4,

there is a high level of sympatry among the

amphibian species. There is significant

temporal segregation. Adult T. v.

ophryticus remain in the water from the end

of November to the end of May. Larval

Vol. 5 p. 82

Asiatic Herpetological Research

December 1993

TABLE 6. Faeces compostion of sympatric amphibian species of the Yew-box Grove

SPECIES:

development occurs from March until

August (Table 4). Pelodytes caucasicus

can be heard calling from May until the

middle of October.

The adults of T. v. ophryticus and P.

caucasicus prey upon each others larval

stages. When post metamorphic T. v.

ophryticus begin leaving the water in late

July they are prey upon by adult P.

caucasicus . In December when adult

newts enter the water to breed they capture

the smaller sizes of P. caucasicus tadpoles.

During these periods of intense competition

and predation both species adopt several

strategies for preying upon the other for

avoiding predation (Smith 1981).

December 1993

Asiatic Herpetological Research

Vol. 5 p. 83

TABLE 7. Size limits of feeding objects of sypatric amphibians in the Yew-box Grove

SIZE OF FEEDING OBJECTS ( in mm)

The highest population densities of T.

v. ophryticus, T. v. lantzi and T. c. karelini

occur in small forest lakes. Much lower

population densities are found in mountain

streams. Recently formed lakes are

apparently the most suitable habitat for

these species of newts. As lakes mature,

sediments accumulate and they become less

suitable habitat and populations decline and

are preserved at low levels in nearby

streams.

The highest level of sympatry occurs in

lakes during breeding season (Table 3).

There is, however, very little competition

because of temporal and microhabitat

segregation for egg deposition and

deposition sites.

Lakes and deep pits (Study Site 7) are

breeding sites for H. a. schelkownikowi

where it is spatially segregated from other

amphibians but overlaps temporally with

P. caucasicus (Table 3) and activity

patterns (Table 5).

In the Yew-box Grove P. caucasicus

and H. a. schelkownikowi are allopatric at

breeding sites. On the Caucasian Black Sea

Coast they are sympatric at breeding sites.

The interrelationships of these populations

in the zones of sympatry have not been

studied.

On the Caucasian Black Sea Coast

winters are mild with abundant

precipitation. Most amphibians remain

active and abroad through the winter.

Though, on the occasional cold days they

become torpid. The exceptions are adult P.

caucasicus and H. a. schelkownikowi.

During some cold winters when night

temperatures fall to -10 to -12° C all

amphibian species enter hibernation.

Bufo verrucosissimus, P. caucasicus,

H. a. schelkownikowi. and T. v. lantzi

pass hibernation hidden in the soil and leaf

litter. Rana macrocnemis hibernates in the

soil, leaf litter and in the water. Winter soil

temperatures at 20 cm depth ranges from

3.5 - 7.2" C. In the spring the water

warms more quickly than the soil and this

may explain why the those amphibians

hibernating in the water breed earlier than

those hibernating in the soil.

The mechanism of niche isolation of

symbiotopic species includes layering, or

the formation of adaptive groups

(Dinesman 1948a) and differences in daily

and seasonal activity. For example, at

Study Site 1, T. v. ophryticus occupies the

deepest aquatic level, R. ridibunda and P.

caucasicus are found in the lower

intermediate levels, R. macrocnemis and B.

verrucosissimus occupy the higher

intermediate levels and H. a.

schelkownikowi is found at the shallowest

level and onto land (Fig. 4). Some of the

pattern of species distribution can be

explained by varying tolerance of

desiccation among the different species

(Dinesman, 1948b). Those species that are

found at the same level are active at

different times of the day (Table 5).

The amphibians of the Yew-box Grove

can be divided into three groups based on

their food habits: 1) feeding on

hydrobionts (newts), 2) feeding on arboreal

invertebrates (tree frogs), 3) feeding on

Vol. 5 p. 84

Asiatic Herpetological Research

December 1993

terrestrial invertebrates (all other species).

Rana ridibunda feeds on both aquatic and

terrestrial invertebrates (Table 6, 7).

In a given area species diversity is

dependent upon niche separation (Pianka,

1981). The seven amphibian species

studied here are characterized by biotopic

(including breeding site choice, hibernation

sites and summer activity ranges), seasonal

activity period, daily activity period and

food habits isolation. In the Yew-box

Grove the highest level of species diversity

is achieved in the most stable aquatic

environments. Water temperature during

the breeding season, level of illumination

and water chemistry appear to be an

important characteristics.

Acknowledgments

We wish to express our gratitude to N.

B. Ananjeva for valuable consultations and

to I. V. Marchukaitis for preparing the

illustrations.

Literature Cited

ANANJEVA, N. B. 1981. [On the studies of

sympatric species using reptiles as an example]..

Nauka Publishing House, Leningrad. Pp. 15-

26. (In Russian).

ANONYMOUS. 1978. [Unified methods of water

analysis in the USSRL Leningrad,

Gidrometizdat Publishing, Leningrad. (In

Russian).

DAZHO, P. 1975. [Foundations of ecology!

Progress Publishing House, Moscow. 408 pp.

(In Russian).

DINESMAN, L. G. 1948a. [On the question of

ecological differentiation of amphibian species].

Bull. MSNT 3(6):47-50. (In Russian).

DINESMAN, L. G. 1948b. [Adaptation of

amphibians to varying humidity]. Zoological

Journal 27(3):231-240. (In Russian).

DINESMAN, L. G. AND M. L. KALETSKAYA.

1952. [Methods of amphibian and reptile

counts! Pp. 329-341. In Methods of counting

and geographical distribution of terrestrial

vertebrates. Moscow.

FORMOZOV, A. N. 1934. [Special type of

climagrams for ecological studies]. Scientific

Transactions of Moscow University 2:271-274.

(In Russian).

GULISASHVILI, V. Z., L. B. MAHATADZE, AND

L. I. PRILIPKO. 1975. [Vegetation of the

Caucasus]. Nauka Publishing House, Moscow.

227 pp. (In Russian).

KASHKAROV, D. N. 1927. [Methods of

qualitative studies of vertebrate and analysis].

Tashkent. 23 pp. (In Russian).

LYAPKOV, S. M. AND A. S. SEVERTSEV. 1981.

[Mechanism of co-existence of two species of

Far-Eastern Anura]. Zoological Journal 3:398-

408. (In Russian).

MONCHADSKY, A. S. 1958. [On classification of

factors of environment]. Zoological Journal

37(5):680-691. (In Russian).

ORR, L. P. AND W. T. MAPLE. 1978.

Competition avoidance mechanisms in

salamander larvae of the genus Desmognathus.

Copeia 1978(4):679-685.

PIANKA, E. R. 1973. The structure of lizard

communities. Annual Review of Ecology and

Systematics 4:53-74.

PIANKA, E. R. 1981 (Evolutionary ecology!

Mir Publishing House, Moscow. (In Russian).

PIANKA, E. R., R. B. HUEY, AND L. R.

LAWLOR. 1979. Niche segregation in desert

lizards. Pp. 67-115. In Analysis of ecological

systems. Columbus, Ohio State University

Press.

SCHOENER, T. W. 1974. Resource partitioning

in ecological communities. Science 185:27-38.

SCHOENER, T. W. 1977. Competition and the

niche. Pp. 35-136. In C. Cans and D. W.

Tinkle (eds.). Biology of the Reptilia, Vo. 7,

Ecology and Behaviour. Academic Press, New

York.

SMITH, J. M. 1981. [Evolution of behaviour].

Mir Publishing House, Moscow. (In Russian).

VERZHUTSKY, B. N. AND B. E. ZHURAVLEV.

1977. [Sparing method of studies of reptile

trophic spectrum]. Pp. 58-59 In The Problems

of Herpetology. Fourth USSR Herpetological

Conference. Science Press, Leningrad.

(Abstr.XIn Russian).

December 1993

Asiatic Herpetological Research

Vol. 5, pp. 85-95 1

Notes on a Collection of Squamate Reptiles from Eastern Mindanao,

Philippine Islands Part 1: Lacertilia

Brian E. Smith

Museum of Natural History. Louisiana Stale University, Baton Rouge, Louisiana 70803-3216. Present Address:

Department of Biology. University of Texas at Arlington Box 19498 Arlington, Texas 76019-0498

Abstract. -In 1982, I spent six months collecting the herpetofauna of several areas in eastern Mindanao,

Philippine Islands. I present species accounts for all lizards collected at these sites, and draw conclusions

about the biodiversity of second-growth and primary forest habitats. I found second-growth habitats to be

depauparate compared to primary forest habitats. Species which may depend on primary forest habitat and

also species apparently restricted to such habitats are detailed. The importance of primary forest reserves,

selective logging, and a mosaic of successional habitats within primary forest reserves is discussed.

Key words: Reptilia, Squamata, Lacertilia, Philippines, taxonomy, ecology, biodiversity

Introduction

Little is known of the ecology,

distribution, and life history of Philippine

lizards. Taylor (1922a, b, c; 1923; 1925)

investigated the systematics and

zoogeography of the Philippine

herpetofauna and provided some ecological

information. Brown and Alcala (1961,

1964, 1978, 1980) undertook ecological

and systematic studies on some islands,

Alcala (1986) recently reviewed the

herpetofauna, and Auffenberg (1988) has

detailed the ecology of the Gray's monitor,

Varanus olivaceus. However, many

islands remain poorly known. This and a

subsequent paper provide information on

the natural history of squamate reptiles

collected in eastern Mindanao, Philippine

Islands, during a collecting trip to the area

made in 1982. I also comment on

squamate assemblages occurring in primary

and second-growth dipterocarp forest in an

effort to pinpoint species of special concern

should primary forest continue to be lost to

logging and agricultural practices.

Site Description Vegetation. — The

vegetation of the Philippines was described

by Brown (1919) and Dickerson (1928)

and a detailed study of southeast Asian rain

forests was published by Richards (1952).

Trees of the family Dipterocarpaceae

dominate the rain forests of the Philippines.

For the purposes of my study, primary

dipterocarp forest is defined as dipterocarp

forest that has apparently never been

logged. Early second-growth dipterocarp

forest is defined as dipterocarp forest

selectively logged one or two years

previously. It has abundant ground cover

of rattan, herbaceous vines, shrubs, and

lianas. A few small (<2 m) trees were

present, and the ground was littered with

fallen logs. Selective logging was practiced

at my study sites, and some small

dipterocarps were left standing, though not

enough to cast appreciable shade. Late

second-growth dipterocarp forest is defined

as forest selectively logged three or more

years ago. It usually had a closed canopy

composed of young fast-growing trees

(mostly Trema species) about 10-15 m tall.

Small dipterocarps were present, and the

herbaceous understory was extremely

dense. The vegetation of my study areas is

discussed in more detail in Smith (1985).

Site 1. — This site was located in the

coastal mountains (Diuata Range) of east-

central Mindanao, 55 km south, 20 km

west of Bislig Bay and 10 km southeast of

Mt. Agtuuganon (Fig. 1). Areas from 500-

800m elevation were sampled. Vegetation

of this area was early second-growth and

primary dipterocarp forest. Slopes were

generally very steep (Fig. 2, 3, 4). This

site was sampled from April 6 to August

15, 1982.

Site 2. — This site was also in the Diuata

Range, 33 km south and 7 km west of

Bislig Bay (Fig. 1). Vegetation at this site

consisted entirely of late second-growth

Vol. 5 p. 86

Asiatic Herpetological Research

December 1993

M-f 12^° 125°

East Longitude

2#

FIGURE 1. Map of Mindanao, Philippine Islands,

showing locations of collecting sites 1 and 2

(triangle) and site 3 (circle).

dipterocarp forest. Slopes were moderately

inclined to steep. The site was 400 m in

elevation, and was sampled from August

21 to September 4, 1982.

Site 3. — I also visited Mount Talomo, a

2693 m peak located 30 km west of Davao

City in the Mount Apo Range of southern

Mindanao (Fig. 1). I sampled areas around

the Philippine Eagle Captive Breeding

Project (PECBP) field station (about 1000

m elevation) from September 8-13, 1982.

Workers at the field station made incidental

collections at this site from April to

September. According to residents,

logging in this area was discontinued

sometime in the mid- to late-1960's. The

area was selectively logged and appears to

be more similar to typical pristine forest

than many other logged areas I visited.

Human disturbance is considerable on the

slopes of Mount Talomo. Farms extend up

the slopes from Davao Gulf to about 900 m

elevation. Coconuts, bananas, pineapples,

coffee, and various fruits and vegetables

are grown. Areas sampled range from 600-

1050 m. Slopes are gently to steeply

inclined.

Climate. — Rainfall was heavy at all

sites. At sites 1 and 2, there is no marked

dry season. The wetter season usually

occurs from November to March, with the

heaviest rains in December and January

(Census Office of the Philippine Islands

1920; Dickerson, 1928; Willmott et al,

1981). Annual rainfall in Surigao (the

nearest weather station to sites 1 and 2) is

3647 mm, with 2360 mm falling from

November to March, and 1 191 mm of this

amount in December and January alone

(Willmott et al., 1981). Surigao is in the

lowlands, and rainfall at my collecting sites

may have been higher. Site 3 also does not

have a marked wet or dry season (Census

Office of the Philippine Islands, 1920;

Dickerson, 1928). Annual rainfall in

Davao (at the base of the Mount Apo range)

is 1942 mm (Willmott et al., 1981). This

figure undoubtably increases with

elevation. During some times of the year,

the PECBP station may be shrouded in

clouds for weeks at a time. Temperature

was relatively constant at all sites. At site 1

it usually ranged from 20-25 C under the

canopy, and from 20-30 C in the open.

The minimum temperature recorded was 18

C, the maximum was 34 C. Temperatures

at site 2 were similar. Temperature

readings were not available from site 3, but

it was slightly cooler because of the higher

elevation. Humidity at all sites varied from

79-100%. Occasionally typhoons strike

Mindanao, but they generally lack the

severity of those hitting more northerly

islands (Census Office of the Philippine

Islands, 1920, Dickerson, 1928). In

March of 1982 a typhoon struck Bislig Bay

near sites 1 and 2. The only effect at the

study sites was moderate, steady rain for

several days.

Methods

Specimens were collected using drift

fences (Gibbons and Semlitsch, 1981) and

by hand. Drift fences were 0.5 m high and

18 m long and constructed as detailed in

Gibbons and Semlitsch (1981). Pit-cans

and funnel traps were placed at either end

of the fence. Pit-cans seemed effective at

capturing all terrestrial lizards except

Varanus spp. Funnel traps were primarily

useful in capturing snakes. Drift fences

appeared to adequately sample reptiles

moving on the soil surface. To sample

December 1993

Asiatic Herpetological Research

Vol. 5 p. 87

FIGURE 2. Primary forest at the edge of a road cut at site 1. The ridge top is about 900m elevation.

arboreal fauna, I examined epiphytic ferns

and trees felled during logging. These

efforts were largely unsuccessful and

arboreal species are under-represented in

the collection. Data on macrohabitat

(primary, early, or late second-growth

dipterocarp forest), microhabitat (fossorial,

terrestrial, or arboreal), elevation, date, and

time of day were recorded for each

specimen collected. Standard

measurements including snout-vent (SVL),

tail (TL), and total length (TTL) were taken

in the field on freshly killed specimens.

General weather conditions were noted

daily. Specimens were dissected in the

laboratory to determine sex, stomach

contents, and reproductive condition of

females. Food contents were identified

usually to family for the arthropod prey of

these lizards. Standard scale counts were

also taken, but no deviation from

previously published data was noted, and

scale count data are not reported herein.

Species Accounts

Family Gekkonidae

Cyrtodactylus agusanensis: — Specimens of

this species were collected in all habitats at

sites 1 and 2. This species' morphology

suggests nocturnal habits (vertically

elliptical pupils), but half the specimens

were captured in drift fences during the

day. Specimens captured at night were

taken on bushes and logs, 1-3 m above

ground. My observations do not agree

with Alcala (1986), who states that this

species is found in swamps and along

rivers. My specimens were all taken far

from such habitats. I also did not find this

species to be particularly rare, as did Alcala

Vol. 5 p.

Asiatic Herpetological Research

December 1993

(1986). Females taken June 14 and 23 had

one large egg (15.7-18.2 mm long) in each

oviduct. Small males were captured April 2

and 9 and a small female July 31. One

juvenile was captured at site 2 on August

26. Stomach contents included insects of

several families and a shed skin, prohahly

of C. agusanensis.

Specimens examined: LSUMZ 41601-

41609, 41640.

Family Agamidae

From their morphology, all agamids

collected on the study sites appear to be

arboreal, although most specimens, except

Draco species, were caught on or near the

ground.

Calotes cristatellus: — A female was

collected at about 550 m in early second-

growth forest at site 1. A male specimen

lacks additional data. This species is rare,

transient in the habitats sampled, or mostly

arboreal and hence under-represented in the

collection. It has morphology indicative of

a highly arboreal lifestyle. The female,

collected on July 1, had one large egg (38.4

and 35.2 mm) in each oviduct. Stomach

contents included lepidopteran larvae,

unidentified insects, and a snail.

Specimens examined: LSUMZ 41737,

41738.

Draco mindanensis: — Taylor (1922a)

collected only two specimens of this

species. They were taken at 1 100 ft. at the

base of Malindang Mountain, northwestern

Mindanao. Inger (1983) examined

taxonomic characters in nine specimens but

gave no ecological data for them. My

specimens were taken at 650 m at site 1 in

primary forest. The species may be

confined to primary forest. Draco

mindanensis is diurnal and arboreal. A

female contained two oviducal eggs 17.5

and 18.4 mm long. The date of capture of

this specimen is unknown. Stomach

contents consisted of several families of

insects. This species is apparently not an

ant-feeding specialist like its congener, D.

volans (see below). D. mindanensis is

reported from Catagan and Malindang

Mountain in northwestern Mindanao

(Taylor, 1922a) and the Diuata Mountains

in the province of Davao del Norte, east-

central Mindanao (this study).

Specimens examined: LSUMZ 41678-

41680.

Draco volans. — This species is very

common in early second-growth forest and

is probably the most conspicuous lizard

species in this or any other habitat.

Individuals are commonly seen running

along branches, displaying, and gliding.

They are diurnal and exclusively arboreal.

This species was never seen in primary

forest. A female containing two eggs (14.7

and 13.9 mm long) was collected on July

20. Taylor (1922a) stated, and my data

confirm, that this species feeds exclusively

on ants.

Specimens examined: LSUMZ 41741-

41748.

Gonyocephalus semperi. — Although

arboreal by morphology, five specimens

were captured on the ground in drift fences.

One was caught by hand 1 m above the

ground on a large tree. This species is

diurnal, as is its' congener G. godeffroyi in

the Solomon Islands (McCoy 1980). Four

were captured in primary forest, two in late

second-growth forest. Due to the species'

arboreal habits, it is highly likely that G.

semperi does not occur in highly disturbed

areas largely lacking trees. McCoy (1980)

found that G. godeffroyi also avoids open

areas in the Solomons. One G. semperi

exhibited aggression and grunted when

handled. The single female captured June

16 contained three developing eggs (5.8,

6.6, and 7.3 mm in length). Stomachs

examined contained the remains of insects

of the families Chilopoda, Coleoptera,

Hymenoptera, Orthoptera, and larval

Lepidoptera.

Specimens examined: LSUMZ 41730-

41735.

Hydrosaurus pustulosus. — This is a

juvenile specimen collected by a native on

December 1993

Asiatic Herpetological Research

Vol. 5 p. 89

Figure 3. Early second-growth forest at site 1. This site had been selectively logged two or three years

prior to this photograph. Trees in the middle backgound were deliberately left standing during the selective

logging procedure.

June 19. Its' stomach was empty. It is

said by Alcala (1986) to be omnivorous,

which generally agrees with observations

of captive specimens at the Dallas Zoo.

Auffenberg (1988) states that adult H.

pustulosus are entirely folivorous in the

wild. Captive specimens usually lay 6-8

eggs measuring roughly 50 mm in length

about once a year (Mitchell 1985). This

species is said to be common in the

Philippines near unpolluted mountain

streams (Alcala 1986). It has also been

observed around coastal fishing villages,

utilizing as vertical perches the stilts or

piers that support homes over water (L. A.

Mitchell, personal observation).

Specimen examined: LSUMZ 41739.

Family Scincidae

In the Philippines, skinks far exceed the

other lizard families in number of species,

abundance, and probably in the variety of

niches they occupy. They are often the

most abundant lizards in all habitats

sampled, with the exception of early

second-growth forest, where Draco volans

is more abundant. The leaf litter

herpetofauna is dominated by species of the

family Scincidae.

Brachymeles gracilis hilong: —

Specimens were captured only in late

second-growth forest (site 2, 400 m) and at

Mount Talomo (site 3). Brown and Alcala

(1980) stated that this fossorial species is

found under leaves, duff, rotting logs, and

Vol. 5 p. 90

Asiatic Herpetological Research

December 1993

Figure 4. Small permanent stream in primary forest at site 1. Small streams such as this one were

common in areas of primary forest at all sites.

in loose soil, usually only in primary forest

from 50-1000 m elevation. This species is

apparently rare or absent in early second-

growth forest. Stomach contents indicate

that this species is a generalized insectivore,

however, part of a skink tail (Mabuya or

Sphenomorphus species) was found in one

stomach.

Specimens examined:

41725.

LSUMZ 41719-

Brachymeles schadenbergi oriental is. —

This fossorial species was often caught in

drift fences after long, steady rains. It is

found under logs and in leaf litter and loose

soil in primary and second-growth forests

at elevations from 50- 10(H) m (Brown and

Alcala, 1980). A female collected July 5

had three developing embryos 18.1, 15.9,

and 15.0 mm in her oviduct. Brown and

Alcala (1980) stated that this subspecies is

ovoviviparous and usually has 2 or 3

young. This species is also a generalized

insectivore. In addition, a lizard tail

(probably Brachymeles species) was found

in one specimen's stomach.

Specimens examined: LSUMZ 41726-

41729.

Lamprolepis smaragdina philippinica. —

This was a very common arboreal species

in cultivated areas and villages, and a single

specimen was taken in a coconut plantation

on Mount Talomo between 700 and 800 m

elevation. In contrast to Brown and Alcala

(1980) and Alcala (1986), I never observed

this species in primary or second-growth

forests. My study sites did not encompass

December 1993

Asiatic Herpetological Research

Vol. 5 p. 91

agricultural areas or villages.

Specimen examined: LSUMZ 4 1 648.

Lipinia semperi. — Taylor (1922a) noted

that this species is commonly found in old

tree stumps and hollow trees. Alcala

(1986) states that it is arboreal and rare.

My specimen was taken in daylight from an

epiphytic fern 4 m above ground in primary

forest at 800 m elevation.

Specimen examined: LSUMZ 41740.

Mabuya multicarinata multicarinata. —

This is a very active and abundant skink

that favors open areas. It is diurnal and

terrestrial. All but two specimens were

caught in early second-growth forest. One

specimen was taken in late second-growth

forest. An additional specimen was

observed in primary forest, but this was

within 100 m of a logging road and it may

have been transient. This species seems to

be a lizard of open areas, and it may have

originally occupied natural treefall gaps in

the forest. With extensive clearing of areas

for logging and cultivation, M. m.

multicarinata has probably increased

substantially in numbers. Of eleven

females collected throughout this study,

only two failed to contain well-developed

eggs. The rest had 2 (6 specimens) or 3 (3

specimens) large oviducal eggs 11.0-15.5

mm long. Juveniles 29-41 mm SVL were

captured on June 24 and August 1,4, 12,

and 26. Another juvenile was taken from

the stomach of a snake (Cyclocorus

nuchalis taylori) that was collected July 1.

Stomachs examined contained insects of

many taxa, and I consider this species a

generalized insectivore.

Specimens examined: LSUMZ 41610-

41639.

Sphenomorphus Species

Lizards of the genus Sphenomorphus

dominate the leaf litter herpetofauna of the

primary forest in the areas I investigated,

and they are sometimes conspicuous in

secondary growth as well.

Sphenomorphus exceeds all other genera at

my sites in number of species and

individuals. Its' species are mostly diurnal

and terrestrial.

Sphenomorphus acutus: — Brown and

Alcala (1980) state that this species is

strictly arboreal, which may account for the

paucity of specimens collected. However,

three specimens were collected in drift

fences during daylight, so they are at least

occasionally found on the ground. This

agrees with Alcala's (1986) observations.

Stomach contents included arachnids and

orthopterans.

Specimens examined: LSUMZ 41713-

41715.

Sphenomorphus coxi coxi. — This was

the most common ground-dwelling lizard in

the primary forest and is the most

conspicuous member of the leaf-litter

herpetofauna in this habitat. It is fairly

common in second-growth forest as well,

but it is not seen in the open as often as

Mabuya multicarinata multicarinata.

Although unquantified, I believe that these

observations represent a real difference,

and that M. m. multicarinata replaces 5. c.

coxi as the dominant skink in second-

growth habitats. Gravid females were

collected April 27 (1 egg, 5.2 mm long),

July 3 (2 eggs, 8.5 and 9.2 mm), and

August 5 (2 eggs, 14.7 and 15.2 mm).

One juvenile was collected May 2 1 and two

on May 24. Four others were collected

August 9, 13 (2), and 28. These juveniles

measured 34-45 mm SVL. One small

specimen identified as a male measured 45

mm SVL. These data suggest two hatching

seasons during the six months of this

study, one in May and the other in August.

Apparently, reproductive maturity is

reached at approximately 45 mm SVL.

Stomachs examined contained insects of

many taxa. I consider this species to be a

generalized insectivore.

Specimens examined: LSUMZ 41649-

41677.

Sphenomorphus decipiens. — Although

Alcala (1986) states that this species is rare,

I found it to be a fairly common diurnal

Vol. 5 p. 92

Asiatic Herpetological Research

December 1993

member of the leaf litter herpetofauna in

primary forest. It was never found in

second-growth forest. A juvenile (25 mm

SVL) was collected June 23. Stomachs

contained larval lepidopterans and other

insects. Specimens examined: LSUMZ

41704-41711.

Sphenomorphus fasciatus. — Brown and

Alcala (1980) noted that this species was a

common terrestrial skink. During my

study, I collected only one specimen at site

1. Since my capture techniques seemed

especially efficient in sampling the

terrestrial herpetofauna, I conclude that S.

fasciatus was rare at my collecting sites.

The stomach of this specimen contained

insects of several taxa.

were either captured in drift fence traps or

found under cover. At night, individuals

were found actively foraging. This species

is apparently secretive during daylight.

Specimens frequently grunted during

capture and attempted to bite. All

specimens were caught in primary forest

except one caught in late second-growth

forest. This species is apparently absent

from highly disturbed areas. Gravid

females were caught March 16 (5 eggs,

3.5-6.3 mm) and June 14 (5 eggs, 12.3-

13.9 mm).

Specimens examined: LSUMZ 41641-

41647.

Discussion

Specimen examined: LSUMZ 41716.

Sphenomorphus steerei. — This species

is common in primary forest, and is

occasionally found in second-growth

forest. In my study, it was more common

above 650 m elevation, although Brown

and Alcala (1980) stated that it occurs from

sea level to 2000 m. Stomachs contained

insects of several different taxa.

Specimens examined: LSUMZ 41697-

41703,41712.

Sphenomorphus variegatus. — This

species is a common diurnal leaf-litter lizard

of the primary forest. It was never found

in early second-growth forest, but two

specimens were taken in late second-

growth forest. S. variegatus appeared to be

completely absent from highly disturbed

areas. Females taken April 1 1 and May 24

each contained two eggs (11.0 and 10.3

mm; and 6.5 and 6.1 mm, respectively).

Juveniles (SVL 24-39 mm) were captured

May 23, June 29, August 9, and August

26. Stomachs contained insects of many

different taxa.

Specimens examined: LSUMZ 41681-

41696.

Tropidophorus partelloi. — This species

was found to be active during the day and

night. Specimens collected during the day

The continuing destruction of tropical

rain forest worldwide makes it imperative

that species of special concern are identified

in these habitats. In the section below, I

will attempt to pinpoint species which could

be adversely affected by continuing

deforestation, where my data are adequate

to do so. I make special reference to the

lizards of the forest floor, since I feel that

these lizards were the most accurately

sampled species with the techniques that I

used. These species are all skinks:

Mabuya multicarinata multicarinata,

Sphenomorphus coxi coxi, S. decipiens, S.

steerei, S. variegatus, and Tropidophorus

partelloi. Sphenomorphus fasciatus is not

considered, since only one specimen of this

species was collected.

Primary Forest

This was the richest habitat sampled,

containing all six of the skinks mentioned

above. Of these six, five seem to be

regular inhabitants of the primary forest. In

addition, there were two common arboreal

lizards: The gecko Cyrtodactylus

agusanensis and the agamid

Gonyocephalus semperi. The most

commonly collected terrestrial lizards listed

in decreasing order of abundance were:

Sphenomorphus coxi coxi, S. variegatus,

S. steerei, S. decipiens, and Tropidophorus

partelloi. Species which were sparsely

collected in primary forest include the

December 1993

Asiatic Herpetological Research

Vol. 5 p. 93

arboreal lizards Calotes cristate/Ins, Draco

mindanensis, Lipinia semperi, and

Sphenomorphus acutus, and the terrestrial

lizard Mabuya multicarinata multicarinata. I

do not consider M. m. multicarinata to be a

regular inhabitant of primary forest.

Early Second-Growth Forest

This was the most depauperate habitat

sampled. Two arboreal lizards were

common: Cyrtodactylus agusanensis and

Draco volans, the latter species being the

most common lizard observed in any

habitat. The common terrestrial lizards

were: Sphenomorphus coxi coxi and the

extremely abundant Mabuya multicarinata

multicarinata. This habitat contained only

three of the six terrestrial skinks of interest.

The skink Sphenomorphus steerei was only

represented by two individuals collected in

this habitat, and I consider it to be rare in

early second-growth habitats. M. m.

multicarinata was virtually absent in

primary forest, but it was found to be twice

as common in second-growth habitats as

the next most common lizard, S. c. coxi. I

spent approximately two man-months

apiece working in primary forest and early

second-growth forest at site 1. I have

assumed that the time spent in these two

habitat types was equal and have calculated

the Brillouin diversity index (Krebs 1989)

for each habitat using the absolute number

of specimens collected in each habitat type

of the six terrestrial skinks of interest noted

above. Indices of 0.610 and 0.309 are

calculated for primary forest and early

second-growth forest habitat types,

respectively. Although not an exact

measurement, this rough estimate of

diversity points out the major difference in

biodiversity between these two habitat

types as regards terrestrial skinks.

Late Second-Growth Forest

I spent very little time in this habitat

type, yet the species collected here provide

valuable insights into possible lizard

population successional patterns. The

arboreal agamid Gonyocephalus semperi,

absent in early second-growth forest, was

found in late second-growth forest.

whereas Draco volans, common in early

second-growth forest, was absent from late

second-growth forest. The terrestrial

skinks Sphenomorphus variegatus and

Tropidophorus partelloi, both absent from

early second-growth forest and common in

primary forest, were collected in late

second-growth forest.

Species of Special Concern

It is clear that less complex habitats will

support a less complex community of

plants and animals, by definition.

Although my data are sparse, it is obvious

that the biodiversity of terrestrial skinks

(and other lizards) is far different after

logging operations alter the forest structure.

The clearing of primary forest creates

habitat for open habitat specialists such as

Mabuya multicarinata multicarinata and

Draco volans, while decreasing habitat for

the species which seem to be confined to

the primary forest. Especially notable is die

lack of Sphenomorphus variegatus, which

is common in primary forest but absent in

early second-growth forest. Other species

which may be adversely affected by

logging activities include the skinks

Tropidophorus partelloi, Sphenomorphus

steerei, and 5. decipiens; and the agamids

Draco mindanensis and Gonyocephalus

semperi. Except for D. mindanensis, all

these species were collected in late second-

growth and primary forest. D.

mindanensis was only found in primary

forest. Sphenomorphus coxi coxi was the

only lizard commonly found in all habitats,

although it was less frequently seen in the

open where it was syntopic with Mabuya

multicarinata multicarinata.

Conclusions

The collection of primary forest species

in areas of late second-growth forest points

towards the possibility of a sustained

harvest of the primary forest for lumber.

However, there is absolutely no data on the

periodicity of such a harvest, nor any

precise information on changes in

biodiversity through a successional series.

It is imperative that primary forest be

conserved as a refuge for certain species

Vol. 5 p. 94

Asiatic Herpetological Research

December 1993

which may only occur there, such as Draco

mindanensis. Possibly, a system of

reserves with rotating areas of selective

logging may help to conserve Philippine

lizard species. However, at the present

time research towards such an end is

entirely lacking, and the conversion of large

parts of the Philippines towards a much

impoverished lizard fauna continues

unabated.

Acknowledgments

This paper is part of a thesis submitted to

the Louisiana State University Graduate

School in partial fulfillment of the

requirements for the degree of Master of

Science. Dr. D. A. Rossman supported

and guided all phases of this work. Dr. J.

V. Remsen, Dr. J. W. Fleeger, and Dr. W.

J. Harman also assisted in innumerable

ways. Dr. R. S. Kennedy of the Philippine

Eagle Conservation Project initially

suggested this study and was instrumental

in obtaining funding as were Dr. J. P.

CTNeill and Dr. D. A. Rossman. The

Philippine Paper Company provided the

study site, storage space, and

transportation. A collecting permit was

graciously provided by the government of

the Philippine Islands. Mario Caleda,

Hector Miranda, Kevin Moore, Bill

Wischusen, Mark Witmer, and especially

Susing Babao, all of the Philippine Eagle

Conservation Project, assisted in obtaining

specimens. Robin Lawson, Bill

Sanderson, and Randy Vaeth assisted in

identification of specimens and food

remains. Jim Murphy and staff at the

Dallas Zoo Herpetology Department

provided an intellectually stimulating

environment in which to finish final editing

of this paper. Ardell Mitchell shared his

observations on both captive and wild-

caught Hydrosaurus pustulosus. Jim

Murphy, Dr. E. D. Brodie, Jr., and Dr. J.

A. Campbell provided helpful comments on

the final draft.

Literature Cited

ALCALA, A. C. 1986. Guide to Philippine Flora

and Fauna, Vol.X: Amphibians and Reptiles.

Natural Resources Management Center,

Ministry of Natural Resources and University of

the Philippines, Manila. 195pp.

AUFFENBERG, W. 1988. Gray's Monitor Lizard.

University of Florida Press, Gainesville,

Florida. 419pp.

BROWN, W. C., AND A. C. ALCALA. 1961.

Populations of amphibians and reptiles in the

submontane and montane forests of Cuernos de

Negros, Philippine Islands. Ecology 42:628-

636.

BROWN, W. C., AND A. C. ALCALA. 1964.

Relationship of the herpetofauna of the non-

dipterocarp communities to that of the

dipterocarp forest of southern Negros Island,

Philippines. Senckenbergiana Biologia 45:591-

611.

BROWN, W. C., AND A. C. ALCALA. 1978.

Philippine Lizards of the Family Gekkonidae.

Silliman University Natural Sciences

Monograph Series Number 1. Silliman

University, Dumaguete City, Philippines.

146pp.

BROWN, W. C, AND A. C. ALCALA. 1980.

Philippine Lizards of the Family Scincidae.

Silliman University Natural Sciences

Monograph Series Number 2. Silliman

University, Dumaguete City, Philippines.

264pp.

BROWN, W. H. 1919. Vegetation of Philippine

Mountains. Manila Bureau of Printing, Manila.

434pp.

CENSUS OFFICE OF THE PHILIPPINE ISLANDS.

1920. Census of the Philippine Islands, 1918.

Vol. I, Part 1. Geography, History, and

Climatology. Manila Bureau of Printing,

Manila. 630pp.

DICKERSON, R. E., ED. 1928. Distribution of

Life in the Philippines. Monograph of the

Bureau of Science, Manila, Number 21. 322pp.

GIBBONS, J. W., AND R. D. SEMLITSCH. 1981.

Terrestrial drift fences with pitfall traps: An

effective technique for quantitative sampling of

animal populations. Brimleyana7:l-16.

INGER, R. F. 1983. Morphological and ecological

variation in the flying lizards (genus Draco).

Fieldiana: Zoology, New Series 18:1-35.

KREBS, C. J. 1989. Ecological Methodology.

December 1993

Asiatic Herpetological Research

Vol. 5 p. 95

Harper & Row, New York. 654pp.

MCCOY, M. 1980. Reptiles of the Solomon

Islands. Wau Ecology Institute, Handbook

Number 7. Wau, Papua New Guinea. 80pp.

MITCHELL, L. A. 1985. Comments on the

maintenance and reproduction of Hydro saurus

pustulosus at the Dallas Zoo. Pp. 185-186. In

S. McKeown, F. Caporaso, and K. H. Peterson

(eds.). Proceedings of the Ninth International

Herpetological Symposium on Captive

Propagation and Husbandry. International

Herpetological Symposium, San Diego,

California. 265 pp.

TAYLOR, E. H. 1922A. The Lizards of the

Philippine Islands. Manila Bureau of Printing,

Manila. 269pp.

TAYLOR, E. H. 1922B. Additions to the

herpetological fauna of the Philippine Islands, I.

Philippine Journal of Science 21:161-206.

TAYLOR, E. H. 1922C. Additions to the

herpetological fauna of the Philippine Islands,

II. Philippine Journal of Science 21:257-303.

TAYLOR, E. H. 1923. Additions to the

herpetological fauna of the Philippine Islands,

III. Philippine Journal of Science 22:515-555.

RICHARDS, P. W. 1952. The Tropical Rain

Forest: An Ecological Study, tlniversity

Press, Cambridge. 450 pp.

TAYLOR, E. H. 1925. Additions to the

herpetological fauna of the Philippines, IV.

Philippine Journal of Science 26:97-111.

SMITH, B. E. 1985. A systematic survey of the

squamate reptiles of eastern Mindanao,

Philippine Islands, with notes on

ecomorphology. M. S. Thesis, Louisiana State

University. 78pp.

WILLMOTT, C. J., J. R. MATHER, AND C. M.

ROWE. 1981. Average monthly and annual

surface air temperature and precipitation data for

the world. Part 1. The eastern hemisphere.

Publications in Climatology 34:1-395.

Vol. 5, pp. 96-102

Asiatic Herpetological Research

December 1993

Notes on a Collection of Squamate Reptiles from Eastern Mindanao,

Philippine Islands Part 2: Serpentes

Brian E. Smith

Museum of Natural History. Louisiana State University, Baton Rouge. Louisiana 70803-3216. Present Address:

Department of Biology. University of Texas at Arlington. Uta Box 19498, Arlington. Texas 76019-0498

Abstract. -A systematic collection of the herpetofauna occurring in the Diuata Range of eastern

Mindanao, Philippine Islands, with incidental collections from the Mount Apo Range, was made from

April-September 1982. This paper completes taxonomic and ecological notes that were taken on the

squamates of this region. Although conclusions about the effect of habitat alteration on snake populations

are necessarily tentative due to sampling difficulties, some comments on apparent and potential shifts in

population size and habitat use of the more common snake species are made.

Key words: Reptilia, Squamata, Serpentes, Philippines, taxonomy, ecology

Introduction

Except for reviews by Taylor (1922) and

Leviton (1959) and biogeographic works

by Brown and Alcala (1970), Leviton

(1963a, 1970) and Wiister and Thorpe

(1989, 1990, 1991a), very little is known

about the ecology and distribution of

Philippine snakes. This paper describes

ecological and distributional data taken on

the snakes of an area in eastern Mindanao,

Philippine Islands, that I visited from April-

September 1982. An earlier paper on the

lizards of this area describes the sites and

methodologies I used during this study.

Scalation data taken on snakes included

dorsal scale rows at ten ventral scutes

posterior to the head, mid-body, and ten

ventral scutes anterior to the anal plate;

ventral scute counts; and subcaudal scale

counts. This data is given only when it

adds to data previously published.

Species Accounts

Family Colubridae

Ahaetulla prasina preocularis. — Two

specimens were taken, one lacking specific

data that was probably collected at or above

the upper elevational limit of 800 m given

by Leviton (1967). This species has a

morphology typical of arboreal snakes, but

is frequently taken on the ground, where it

may descend to forage on Mabuya skinks

(Leviton 1967). If Mabuya is a favored

food item, then it is highly likely that A.

prasina favors open areas, as does Mabuya.

A female was taken on April 6 on a road in

early second-growth habitat near site 1.

Five oviducal eggs (22.8-31.1 mm in

length) were found in this specimen.

Specimens examined: LSUMZ 41804-

41805.

Boiga cynodon. — This snake is arboreal

by morphology, but two specimens were

captured in drift fences and one found dead

on a road at site 1 . They were all collected

or killed at night, and were found in both

early second-growth and primary forest

habitats. Prior studies give no elevational

information (Alcala, 1986; Leviton, 1968a;

Smith, 1943; Taylor, 1922); the present

specimens were found at 450-650 m. This

species is aggressive and has very large

palatine teeth which are quite effective in

defense. Taylor (1922) stated that this

species is rare in the Philippines, but it was

the most commonly captured Boiga species

at my study sites. As Leviton (1968a)

noted, B. cynodon has an arboreal

morphology and diet (birds and bird eggs),

but all specimens were captured on the

ground.

Specimens examined: LSUMZ 41814-

41816.

Boiga dendrophila latifasciata. — This

specimen was taken on a road at night at

about 500 m elevation in early second-

growth forest. Previous reports indicate

1993 by Asiatic Herpetological Research

December 1993

Asiatic Herpetological Research

Vol. 5 p. 97

that this species is a common inhabitant of

lowland swampy areas (Taylor, 1922;

1965; Tweedie, 1983), but the present

specimen was found in hilly country far

from such habitat. The stomach of this

specimen contained bird feathers. Previous

studies found this species to also eat bats

and lizards (Alcala, 1986; Leviton, 1968a;

Taylor, 1922).

Specimen examined: LSUMZ 41812.

Calamaria gervaisi. — This burrowing

species was common in primary forest, but

was rarely found in early second-growth

forest. One specimen was collected 20 cm

underground when digging a pit-can hole.

Four adults were taken in a single pit-can

during four days in April. The significance

of this aggregration is unknown. A female

collected April 10 contained one egg 3.4

mm in length. Segmented worms were

found in the stomachs of two specimens.

Specimens examined: LSUMZ 41769-

41778.

Cyclocorus nuchalis taylori:. — One

specimen was found in leaf litter in the

primary forest, the other was caught in a

drift fence in early second-growth forest.

One specimen had eaten a juvenile Mabuya

multicarinata. Leviton (1965) found

specimens of the skink genera Mabuya and

Sphenomorphus in stomachs of the C.

nuchalis he examined.

Specimens examined: LSUMZ 41796-

41797.

Dendrelaphis caudolineatus terrificus. —

Leviton (1968b) gave an elevational range

of 0-35 m for this species, but I collected

specimens of this snake from 100-1000 m.

Also, I did not note any particular

association with water, as did Alcala

(1986). Although thought to be primarily

arboreal (Leviton, 1968b), all the

specimens I collected were taken on the

ground. One specimen had eaten a

terrestrial skink, Sphenomorphus coxi. A

specimen examined by Leviton (1968b)

also contained a terrestrial skink, Mabuya

species. One specimen that I collected was

taken in grassland in an extremely large ca.

100 square km) clear-cut area. This species

is arboreal in its morphology, but these data

indicate common use of the terrestrial

microhabitat.

Specimens examined: LSUMZ 41798-

41800.

Dryphiops philippina. — This specimen

was collected by a native near Mount

Talomo and is the first specimen of this

species collected on Mindanao. The ventral

scute count of this specimen is 172,

increasing the known variation of 177-188

reported by Leviton (1964a). Previous

specimens were taken near sea level

(Leviton, 1964a), but this specimen may

have been collected as high as ca. 1000 m.

This species was previously known only

from Luzon, Negros, and Sibuyan (Leviton

1964a). Subsequent investigations should

discover it on other large islands in the

Philippine archipelago.

Specimen examined: LSUMZ 41790.

Elaphe erythrura erythrura. — This

species is a common ground-dwelling

diurnal inhabitant of all forest situations. It

feeds on lizards, birds, and mammals

(Alcala, 1986; Leviton, 1977; Taylor,

1922). My specimens commonly had

mammals or mammal remains in their

stomachs, and I consider this species to be

a typical, heavy-bodied, mammal-eating

constrictor that opportunistically takes other

prey. It is also a common food item of the

Philippine Eagle (Pithecophaga jefferyi)

and the Philippine Serpent-Eagle (Spilornis

holospilus). Alcala (1986) reports an

altitudinal range to 500 m, but I took

specimens up to 1065 m. I collected young

snakes August 1 and September 8 (two

snakes). These specimens measured 465,

383, and 409 SVL, respectively. The two

largest of this group were identified as

young males. Leviton (1977) stated that

year-old young probably measure 400 mm

SVL.

Specimens examined: LSUMZ 41807-

41811.

Vol. 5 p. 98

Asiatic Herpetological Research

December 1993

Oligodon maculatus. — This is only the

fifth known specimen of this species.

Taylor (1922) took two specimens beneath

sod and trash piles at Bunawan, Agusan, in

the Agusan Valley of Mindanao. A third

specimen was taken in northern Surigao

Province (Taylor, 1925). A fourth was

collected on Mount Todaya in the Mount

Apo Range (Leviton, 1962). My specimen

was taken during daylight in the forest floor

litter on Mount Talomo, Mount Apo Range,

at about 1000 m elevation. Alcala (1986)

reports two specimens from 400 and 850 m

elevation, but gives no further information.

Including my data, an altitudinal range of

400-1000 m is indicated. This species is

known only from eastern Mindanao.

Leviton (1962) stated that this species has

17 dorsal scale rows throughout; my

specimen reduces to 15 in the posterior

third of the body. Leviton also noted that

the loreal may be present or absent; in my

specimen it is absent. This specimen also

differs from the ones analyzed by Leviton

in having one instead of two preoculars.

My specimen also has fewer scutes than

those examined by Leviton, increasing the

known variability in ventral scute counts to

156-164. The specimen I collected is male.

The stomach was empty.

Specimen examined: LSUMZ 41806.

Oxyrhabdium modestum. — T h e

specimen taken was a gravid female with

eight eggs measuring 18.9-26.1 mm in

length. This specimen was collected on a

road at about 400 m elevation in late

second-growth forest. There is no further

data to add to that given by Leviton

(1964b).

Specimen examined: LSUMZ 41803.

Psammodynastes pulverulentus. — This

is a common and aggressive rear-fanged

colubrid of the forest floor litter. It is

diurnal and was taken in all habitat types.

Juveniles 190 and 197 mm SVL were

collected May 3 and May 23 at site 1. I

found insects, a lizard tail, and a snake

{Calamaria gervaisi) in the stomachs of the

specimens I examined. This species is

primarily known as a lizard feeder, but

frogs and snakes are also taken (Greene,

1989; Leviton, 1983). Greene (1989) did

not report insects as a food item in the

specimens he examined.

Specimens examined: LSUMZ 41779-

41789.

Rhabdophis auriculata auriculata. — This

species is a common diurnal member of the

leaf-litter herpetofauna in all habitat types

investigated. I did not find it to be

associated with water, as reported by Alcala

(1986). These are small inoffensive snakes

which usually attempt to conceal

themselves rather than flee or bite when

captured. I collected gravid females June

17 (three eggs, 9.9-10.5 mm in length) and

30 (one egg, 9.0 mm), and July 8 (three

eggs, 5.3-7.6 mm) and 23 (two eggs, 9.0-

9.2 mm). Juveniles 135-235 mm SVL

were collected April 7, 11, and 12, May 18

and 23, June 14, and July 8. The gradually

increasing SVL of these specimens indicate

that they could be the members of a single

cohort. Leviton (1970) indicated two

hatching seasons in the Mount Apo Range,

during June- July and October-November.

My observations within the Diuata Range

indicate that the hatching season here could

occur as early as March, but data are

scanty. Specimens had eaten frogs and

frog eggs, as also reported by Leviton

(1970).

Specimens examined: LSUMZ 41749-

41768.

Stegonotus muelleri. — This is the eighth

known specimen of this species, and only

the second taken on Mindanao (Leviton

1959). Virtually no ecological data are

available. This specimen was found dead

on a logging road in early second-growth

forest at about 590 m elevation. Leviton

(1959) found three adult Rana limnocharis

in the stomach of a large adult; the present

specimen's stomach was empty. The

specimen I collected is male. Its ventral

scute count is higher than that given for

males by Leviton (1959), thereby

increasing the range of this measurement to

217-236.

December 1993

Asiatic Herpetological Research

Vol. 5 p. 99

Specimen examined: LSUMZ 41 802.

Tropidonophis dendrophiops

dendrophiops. — Specimens were collected

during the day in all habitats, but always

near swift-flowing streams. The altitudinal

range given by Alcala (1986) extends to

700 m; one individual I collected was taken

at 900 m. Juveniles were collected April 18

(161 mm SVL), May 21 (242 mm SVL),

and June 24 (308 mm SVL) at site 1.

These could be members of a single cohort

born in March or April. One specimen was

collected in a pit-can eating a frog. Taylor

(1922) also reported frogs as common food

items of this species. Alcala (1986) refers

to this species as Natrix dendrophiops.

Malnate and Underwood (1988) have

recently assigned this species to the genus

Tropidonophis

Specimens examined: LSUMZ 41791-

41795.

Family Elapidae

Maticora intestinalis philippina. — One

specimen was taken in a pit-can at site 2 in

late second-growth forest at 400 m

elevation. This snake is thought to be rare

in the Philippines (Alcala, 1986; Taylor,

1922). Leviton (1963b) gives no

ecological information, and Taylor (1922)

describes an apparent anti-predator display

during which specimens exhibit aimless

thrashing motions, similar to displays

described for other New and Old World

coral snakes (Greene, 1973). Despite

Alcala's (1986) statement that this species

is found in arboreal ferns as well as under

rotten logs, I consider this species to be a

typical semi-fossorial coral snake. This

agrees with Taylor's (1922) observations.

There are no data on food habits of this

species; my specimen's stomach was

empty.

Specimen examined: LSUMZ 41817.

Naja samarensis. — This is an alert

diurnal species. It is quite common around

habitations and early second-growth

habitats. Contrary to Alcala (1986) and

Taylor (1922), this species was never

found in primary forest, and residents told

me that they never saw individuals of this

species in the forest. It is likely that the

conversion of much of the Philippines into

altered habitat has resulted in an increased

abundance of this highly venomous snake.

I saw or collected this species from sea

level to 1000 m elevation. Despite its

highly toxic venom (Minton 1967), the

local residents believed that this snake

brought luck, and individuals found under

and around houses were invariably left

alive. I actively sought out reports of

envenomation resulting from bites of N.

samarensis, but received no such reports,

and was told by locals that bites of this

species were seldom problematical. This is

in contrast to conclusions reached by Reyes

and Lamana (1955). This is a spitting

cobra, and there is at least one report of an

accurate strike in the eyes resulting in a

great deal of pain but no serious after-

effects (Van Wallach, personal

communication). I handled many

specimens, but never saw one exude any

appreciable quantity of venom. This

species is known to eat frogs, snakes

(Calamaria gervaisi), and rodents (Gressitt,

1937; Leviton, 1964c; Taylor, 1922). A

specimen I examined contained a small

Bufo marinus (SVL 95mm), and Van

Wallach reports (personal communication)

only B. marinus in the stomachs of N.

samarensis he collected in rice paddies on

Mindanao near the city of Surigao. N.

samarensis itself is eaten by the Philippine

Eagle {Pirhecophagas jefferyi) and the

Philippine Serpent-Eagle {Spilornis

holospilus) Two juveniles were taken with

an adult male from a hole in the ground at

1000 m on Mount Talomo. These

specimens were brought to me by a local

resident. It was not possible to ascertain

whether there was any significance to this

aggregation, but other cobras are known to

guard both eggs and young (Campbell and

Quinn, 1975; Tryon, 1979; Tweedie,

1983). Wiister and Thorpe (1991b) have

recently elevated Naja naja samarensis to

full species status, and I use this new

species designation in this paper.

Specimens examined: LSUMZ 41819-

41824.

Vol. 5 p. 100

Asiatic Herpetological Research

December 1993

Discussion

Unlike lizards, snakes are rarely taken in

such numbers that it is possible to gauge

their relative abundance in any given

habitat. Therefore, ecological conclusions

based on studies such as this one are few

and tentative. Most of these conclusions

have already been reached in individual

species accounts.

The pace of destruction of primary forest

in the Philippines has created a great deal of

second-growth habitats of various types.

At least one snake, Naja samarensis, has

probably benefitted from this trend. Taylor

(1922) observed this species in primary

forest, but neither I nor any other biologists

that I worked with, nor any local residents,

ever told me of seeing this species in the

primary forest. I consider it a common

snake of second-growth and agricultural

areas. Other species, such as Elaphe

erythrura erythrura and Rhabdophis

auriculata auriculata appear to be common

in all habitats, although they have not

undergone an apparent shift in habitat

preference similar to N. samarensis. The

only other very abundant snake, Calamaria

gervaisi, may be adversely affected by

deforestation, in that it was common in

primary forest but not in second-growth

habitats. Logging causes both soil

compaction from the movement of heavy

equipment and soil drying from increased

insolation. It seems likely that these

changes would adversely affect a

burrowing animal such as C. gervaisi.

Snake ecology suffers from the typical

problems inherent to studies of higher level

predators. Since they are frequently at the

top of the food chain, snakes never seem to

be very abundant. In addition, they are

secretive by nature, making collection and

observation difficult. If we are to

understand the effect of habitat loss on

snake populations, it will be necessary to

use novel approaches which are uniquely

designed to overcome these specific

problems.

Acknowledgments

This is a portion of a thesis submitted to

the graduate school of Louisiana State

University in partial fulfillment of

requirements for the Master of Science

degree. I thank my major advisor. Dr. D.

A. Rossman, and my committee members,

Drs. J. V. Remsen, J. W. Fleeger, and W.

J. Harman for their comments on my

thesis. Drs. R. S. Kennedy and J. P.

O'Neill helped to arrange funding for my

trip. Members of the Philippine Eagle

Conservation Project and local residents

assisted in the collections. The Dallas Zoo

provided an intellectually stimulating

environment in which to finish final editing

of this paper. Comments by Jim Murphy,

E. D. Brodie, Jr., and J. A. Campbell

improved the final draft.

Literature Cited

ALCALA, A. C. 1986. Guide to Philippine Flora

and Fauna, Vol. X: Amphibians and Reptiles.

Natural Resources Management Center,

Ministry of Natural Resources and University of

the Philippines, Manila. 195pp.

BROWN, W. C, AND A. C. ALCALA. 1970. The

zoogeography of the herpetofauna of the

Philippine Islands, a fringing archipelago.

Proceedings of the California Academy of

Sciences, Fourth Series 38:105-130.

Campbell, j. a., and h. r. quinn. 1975.

Reproduction in a pair of Asiatic cobras, Naja

naja (Serpentes, Elapidae). Journal of

Herpetology 9:229-233.

GREENE, H. W. 1973. Defensive tail display by

snakes and amphisbaenians. Journal of

Herpetology 7:143-161.

GREENE, H. W. 1989. Defensive behavior and

feeding biology of the Asian mock viper,

Psammodynastes pulverulentus (Colubridae), a

specialized predator on scincid lizards. Chinese

Herpetological Research 2:21-32.

GRESSITT, J. L. 1937. Note on a Philippine

cobra. Copeia 1937:73.

LEVITON, A. E. 1959. Systematics and

zoogeography of Philippinesnakes. Ph. D.

Dissertation, Stanford University. 865pp.

December 1993

Asiatic Herpetological Research

Vol. 5 p. 101

LEVITON, A. E. 1962. Contribution to a review

of Philippine snakes, I: The snakes of the

genus Oligodon. Philippine Journal of Science

91:459-484.

LEVITON, A. E. 1977. Contributions to a review

of Philippine snakes, XIII: The snakes of the

genus Elaphe. Philippine Journal of Science

106:99-119.

LEVITON, A. E. 1963a. Remarks on the

zoogeography of Philippine terrestrial snakes.

Proceedings of the California Academy of

Sciences, Fourth Series 31:369-416.

LEVITON, A. E. 1963b. Contributions to a

review of Philippine snakes, III: The genera

Maticora and Calliophis. Philippine Journal of

Science 92:523-550.

LEVITON, A. E. 1964a. Contributions to a review

of Philippine snakes, IV: The genera

Chrysopelea and Dryophiops. Philippine

Journal of Science 93:131-145.

LEVITON, A. E. 1964b. Contributions to a

review of Philippine snakes, VI: The snakes of

the genus Oxyrhabdium. Philippine Journal of

Science 93:407-422.

LEVITON, A. E. 1964c. Contributions to a review

of Philippine snakes, VII: The snakes of the

genera Naja and Ophiophagus. Philippine

Journal of Science 93:531-550.

LEVITON, A. E. 1965. Contributions to a review

of Philippine snakes, IX: The snakes of the

genus Cyclocorus. Philippine Journal of

Science 94:519-533.

LEVITON, A. E. 1967. Contributions to a review

of Philippine snakes, X: The snakes of the

genus Ahaetulla. Philippine Journal of Science

96:73-90.

LEVITON, A. E. 1968a. Contributions to a review

of Philippine snakes, XI: The snakes of the

genus Boiga. Philippine Journal of Science

97:291-314.

LEVITON, A. E. 1968b. Contributions to a

review of Philippine snakes, XII: The

Philippine snakes of the genus Dendrelaphis

(Serpentes: Colubridae). Philippine Journal of

Science 97:371-396.

LEVITON, A. E. 1970. Description of a new

subspecies of Rhabdophis auriculata in the

Philippines, with comments on the

zoogeography of Mindanao Island. Proceedings

of the California Academy of Sciences, Fourth

Series 38:347-362.

LEVITON, A. E. 1983. Contributions to a review

of Philippine snakes, XIV: The snakes of the

genera Xenopeltis, Zaocys, Psammodynastes,

and Myersophis. Philippine Journal of Science

112:195-223.

MALNATE, E. V., AND G. UNDERWOOD. 1988.

Australasian natricine snakes of the genus

Tropidonophis. Proceedings of the Academy of

Natural Sciences of Philadelphia 140:59-201.

MINTON, S. A., JR. 1967. Paraspecific protection

by elapid and sea snake antivenins. Toxicon

5:47-55.

REYES, A. C, AND C. LAMANA. 1955.

Snakebite mortality in the Philippines.

Philippine Journal of Science 84:189-194.

SMITH, M. A. 1943. The Fauna of British India:

Reptilia and Amphibia. Volume III: Serpentes.

Taylor and Francis, London. 583pp.

TAYLOR, E. H. 1922. The Snakes of the

Philippine Islands. Manila Bureau of Printing,

Manila. 312pp.

TAYLOR, E. H. 1925. Additions to the

herpetological fauna of the Philippines, IV.

Philippine Journal of Science 26:97-111.

TAYLOR, E. H. 1965. The serpents of Thailand

and adjacent waters. University of Kansas

Science Bulletin 45:609-1096.

TRYON, B. W. 1979. Reproduction in captive

forest cobras, Naja melanoleuca (Serpentes,

Elapidae). Journal of Herpetology 13:499-504.

TWEEDIE, M. W. F. 1983. The Snakes of

Malaya. Singapore National Printers, Ltd.

Singapore. 167pp.

WUSTER, W„ AND R. S. THORPE. 1989.

Population affinities of the asiatic cobra (Naja

naja) species complex in south-east Asia:

Reliability and random resampling. Biological

Journal of the Linnean Society 36:391-409.

424pp.

WUSTER, W., AND R. S. THORPE. 1990.

Systematics and biogeography of the Asiatic

Vol. 5 p. 102

Asiatic Herpetological Research

December 1993

cobra (Naja naja) species complex in the

Philippine Islands. Pp. 333-344. In G. Peters

and R. Hutterer (eds.), Vertebrates in the

Tropics. Museum Alexander Koenig, Bonn.

WiJSTER, W., AND R. S. THORPE. 1991b.

Systematics of Asiatic cobras. SSAR/HL

Annual Meeting. University Park,

Pennsylvania. [Abstr].

WUSTER, W., AND R. S. THORPE. 1991a.

Asiatic cobras: Systematics and snakebite.

Experientia 47:205-209.

December 1993

Asiatic Herpetological Research

Vol. 5, pp. 103-104

First Records for Ophisaurus hard and Python molurus bivittatus from

Jiangxi Province, China

Changfu zhong

Department of Biology. Jiangxi Medical College, Nanchang, Jiangxi. China.

Abstract.- Ophisaurus harti and Python molurus bivittatus are reported for the first time from Jiangxi

Province China. The measurements, characteristics, and distributions of these two species and subspecies

are described in detail.

Key Words: Reptilia, Lacertilia, Anguidae, Ophisaurus harti, Serpentes, Boidae, Python molurus

bivittatus, China, distribution

Ophisaurus harti Boulenger (Fig. 1)

On October 22, 1980 a specimen of O.

harti was caught by Weitao Ji and Songlin

Cheng of the Wuyi Shan Natural Reserve at

an altitude of 900 meters on Mt. Wuyi,

Yanshan County, Jiangxi Province, China.

The specimen is kept in the Department of

Biology, Jiangxi Medical College,

Nanchang.

Measurements of specimen, in mm.

Specimen Number 600

-t-

>

Characteristics. — Body cylindrical and

no vestiges of limbs externally. Head with

large symmetrical shields; two shields in a

line between the nasal and the azygous

prefrontal; ear-opening minute, smaller than

the nostril. Dorsal scales keeled, in 16

longitudinal rows and 99 transverse series

(counted in the length of the lateral fold);

ventrals smooth, in 10 longitudinal series.

The tail is long and fragile, and regenerates

quickly. The regenerated tail is shorter than

the original one, and the regenerated scales

are smaller than the original ones. Brown

above, with 21 transverse blue marking;

under parts whitish.

Distribution. — Vietnam; China:

Sichuan, Yunnan, Guizhou, Anhui,

Jiangsu, Zhejiang, Jiangxi (Mt. Wuyi in

FIG. 1. Ophisaurus harti.

Jiangsu, Zhejiang, Jiangxi (Mt. Wuyi in

Yanshan County), Hunan, Fujian, Taiwan,

Guangxi.

Python molurus bivittatus Schlegel

(Fig. 2)

A piece of skin of P. molurus bivittatus

was collected in 1965 from the people of

Mt. Daji in Quannan County by the author,

and one living specimen was caught in a

mountain stream in the countryside of the

city of Longnan County by a fisherman

with a fishing net. The specimen was

purchased by Chunhuo Teng of Nanchang

People's Park in 1979.

Measurements of specimen, in mm.

Specimen Number 1001

Vol. 5 p. 104

Asiatic Herpetological Research

December 1993

FIG. 2. Python molurus bivittatus.

Characteristics. — Size large, with

vestiges of hind limbs externally. Head

distinct from neck, with large symmetrical

shields; rostral with a deep pit on either

side; two internasals ; two pairs of

prefrontals, the anterior pair is longer than

the posterior one; frontal a little larger than

the supraocular, divided longitudinally;

parietal, loreal and temporal regions

covered with irregular scales; supralabials

13, the first two deeply pitted, 6th and 7th

separated from the eye by suboculars.

Scales smooth, in 50 rows on neck, 75

rows on midbody and 38 rows before the

vent; anal entire. Tail rather short.

Ventrals and subcaudals were not counted.

Light yellowish above, with a dorsal series

of large, more or less subquadrangular dark

gray, black-edged spots; flanks with

smaller, rounded or irregularly-shaped

spots of the same color. A lance-shaped

mark on the top of the head and the neck;

yellowish below, with a border of dark

spots on the outermost row of the scales;

tail below marbled with yellow and black.

Distribution.— Asia, Indo- Australian

Region; China: Yunnan, Jiangxi (Dajishan

in Quannan County, Longnan County),

Fujian, Guangdong, Hainan, Guangxi.

References

GRESSIT, J. L. 1941. Amphibians and reptiles

from southeastern China. Philippine Science

Journal 75(l):29-58.

POPE, C. H. 1935. The Reptiles of China. Vol.

10 of Natural History of Cenual Asia , New

York. 604 pp.

SMITH, M. A. 1935. The fauna of British India,

including Ceylon, and Burma. Reptilia and

Amphibia. Vol. II, Sauria. Taylor and Francis,

London. 440 pp.

SMITH, M. A. 1943. The fauna of British India,

including Ceylon, and Burma. Reptilia and

Amphibia. Vol. Ill, Serpentes. Taylor and

Francis, London. 583 pp.

TIAN, W. S. and Y. M. JIANG (eds.). 1986.

[Handbook for identification of Chinese

amphibians and reptiles]. Science Press,

Beijing. 164pp. (In Chinese).

ZHONG, C. 1984. [A new record of Jiangxi

Province]. Acta Herpetologica Sinica 3(2):20.

(In Chinese, with English abstract).

ZHONG, C. F. 1986. [Preliminary of survey of

reptiles in the Jinggangshan Natural Reserve].

Jiangxi University Journal (Natural Science)

10(2):71-74. (In Chinese, with English

abstract).

ZHONG, C. F. 1990. [Survey of reptiles in Ruijin

County, Jiangxi Province]. Pp. 236-238 In

Ermi Zhao (ed.) From Water Onto Land. China

Forestry Press, Beijing. (In Chinese, with

English abstract).

ZHONG, C. F. and G. F. WU. 1981. [A

preliminary list and its geographical distribution

of reptiles of Jiangxi Province]. Acta

Herpetologica Sinica (old series) 5(16):99-110.

(In Chinese, with English abstract).

December 1993

Asiatic Herpetological Research

Vol. 5, pp. 105-108

Karyotype Information on some Toad Agamas of the Phrynocephalus

guttatus Species Group (Sauria, Agamidae) of the former USSR.

VALANTINA V. MANILO AND MICHAEL L. GOLUBEV

Institute of Zoology, Academy of Sciences, Kiev, Ukraine

Abstract. -Karyotypes of several toad agamas of the Phrynocephalus guttatus species group (sensu law)

were investigated in specimens from a variety of localities of the former USSR. Differences in the

diakinetic stage of meiosis have been observed, permitting distinctions among three groups of species. The

forms guttatus, moltschanovi, kushackewitschii, and alpherakii comprise Group I; P. guttatus salenskyi

represents the second group; and P. versicolor hispida represents Group III.

Key words: Reptilia, Sauria, Agamidae, Phrynocephalus guttatus, Kazakhstan, Middle Asia, Precaucasus,

karyology.

40 .

FIG. 1. Scheme of distribution of forms of P. guttatus species group of the former USSR fauna: la- P. g.

guttatus; lb- P. g. moltschanovi; II- P. g. kushackewitschii; III- P. g. alpherakii; IV- P. g. salenskyi;

V- P. versicolor hispida; VI- P. guttatus spp. (the numbering of populations is in accordance with the data

in table 1).

Introduction

The first and only extensive karyological

investigation of the agamid lizard genus

Phrynocephalus Kaup is the work of

Sokolovsky (Sokolovsky, 1974; 1977).

Karyotype characteristics permitted the

recognition of five groups. The "guttatus"

group included two species, P. guttatus

(Gmel.) and P. versicolor Str. These

species have a diploid number of 46, all

chromosomes are telocentric. The

karyotypes could be divided into 12 pairs

of macrochromosomes and 1 1 pairs of

microchromosomes. Approximately 50%

of the metaphase plates in P. guttatus

contained satellite chromosomes on the first

pair of chromosomes, but these were never

observed in the P. versicolor karyotype.

The specimens examined came from

Daghestan (P. guttatus) and Central Gobi,

Mongolia (P. versicolor) and were believed

to represent the nominative forms of both

species.

The systematics of the P. guttatus group

based on external morphological

characteristics is extremely complicated and

remains unclear. At various times the

forms alpherakii Bedr., moltschanovi Nik.,

kushackewitschii Bedr., salenskyi Bedr.,

etc. have either been included in P.

guttatus, sensu stricto or treated as related

species. The forms bogdanowi Bedr.,

hispida Bedr., and paraskiwi Semenov et

al. have been assigned to P. versicolor.

(Bedriaga, 1909; Nikolsky, 1915;

Terentjev and Chernov, 1949; Peters,

Vol. 5 p. 106

Asiatic Herpetological Research

December 1993

m

\

i

• w •

i H HI lif) fl« AH Hi M

fin no no ~» >• -- -- -*

st

V

i ^

*%

\v - ■

FIG. 2 The karyogramme of Phrynocephalus

guttatus salenskyi.

1984; Semenov et al. 1987). Golubev

(1989) suggested that P. guttatus and P.

versicolor from Kazakhstan are

conspecific. Karyotype details of the forms

listed above have never before been

examined. The purpose of this study is to

determine whether karyotype information

will aid in our understanding of the

systematics and evolution of

Phrynocephalus.

Methods

Between 1989-1991 we collected

specimens of nearly all listed forms of both

species of Phrynocephalus inhabiting the

territory of the former USSR with the

exception of P. v. bogdanowi from the

extreme south of Tuva (Central Asia) and

P. guttatus ssp. from Turkmenistan (Fig. 1

and Table 1). Chromosome samples were

prepared from cellular suspension of bone

marrow, blood, and testis. We used a

smear method and a method known as

"digging out" in conformity with

procedures described by Ford and

Hamerton (1956) and McGregor and

Varley (1986) as partially modified by

Manilo (1986). Chromosomal staining was

FIG 3. Diakinetic stage of meiosis of five forms of

P. guttatus species group, s. lato: I- all elements

are ring- or stick-shaped; (guttatus, moltschanovi,

kushackewitschii, alpherakii); II- one or two

elements are cross-shaped (salenskyi); III- two and

more (up to four) elements are cross-shaped

(hispida).

performed by Giemsa stain (2% solution)

in 0.01 M sodium-phosphate buffer (pH

6.8) for 20-30 minutes. After washing in

distilled water, the preparations were

passed through alcohols and xylols

(orthoxylol) and subsequently embedded in

Canadian Balsam. In excess of 30

metaphase plates from each form were

investigated using a Biolam 1-212

microscope.

Metaphase plates of spermatogonial

division, spermatocyte I (diakinesis) and

spermatocyte II (metaphase II) bivalents

were investigated in testis preparations.

Chromosome morphology is described

according to the classification proposed by

Levari etal. (1964).

Results and Discussion

Our data support the findings of

Sokolovsky (1974, 1977). The diploid

number is uniformly 46 and the

Fundamental Number (NF) is 46. In

several forms {guttatus, moltschanovi,

kushackewitschii) we noted satellite

chromosomes on several plates; whereas in

other forms (alpherakii, salenskyi) we saw

no evidence of satellites. The revelation of

this structure largely depends on the degree

of spiralization. It is possible that satellites

will be found in the latter forms with

further investigation and more extensive

material.

December 1993

Asiatic Herpetological Research

Vol. 5 p. 107

TABLE 1 . Localities, sample sizes, and taxa of Phrynocephalus guttatus s. lato populations collected and

investigated in this study (numbering of populations is given in accordance with the data shown in Fig. 1.

Sokolovsky described all chromosomes

as telocentric. We cannot confirm this with

confidence. Second arms are clearly visible

on metaphasic plates with premetaphasic

(elongated) chromosomes on several pairs

of large elements. Such chromosomes

could be aero- or even subtelocentric. The

karyotype of salenskyi is an example (Fig.

2). This characteristic is not peculiar to any

one form or group of forms of the guttatus

group and cannot be used to distinguish a

subordinate group.

We also observed distinct peculiarities of

chromosome morphology in meiosis in the

diakinetic stage. The chromosome

bivalents of the various taxa differ in the

number of ring-shaped and cross-shaped

bivalents. Based on this difference in

pairing, it is possible that the taxa of the

toad agamas of the "guttatus" group might

be grouped in the following way;

I: guttatus*, moltschanovi,

kushackewitschii, alpherakii all diakinetic

bivalents are ring or stick-shaped.

II: salenskyi one or two of the elements

are cross-shaped (Fig. 3), the remainder as

in Group I.

Ill: hispida from two to four cross-shaped

elements (Fig. 3), the remainder as in

Group I.

This grouping by cross-shaped elements

in diakinesis is a continuum. In this

system, P. v. hispida is closer to P. g.

salenskyi from the Zaissan Depression

[sometimes attributed to P. versicolor

(Paraskiv, 1953; Bannikov et al., 1977)]

than to other forms of P. guttatus sensu

stricto. However, it may be important that

the toad agama from the Zaissan

Depression occupies an intermediate

position between Groups I and III. It is

interesting to note the absence of

chromosomal differences in the I-st. group,

while its members, as mentioned above, are

attributed by a number of authors to

different species.

The data may be interpreted to suggest

uniformity in the species of the guttatus

group from Kazakhstan, Middle Asia, and

the Precaucasus (Golubev, 1989) as well as

a close relationship between salenskyi from

Zaissan Depression and hispida from

Djungar Gate and northern Djungaria

(Golubev, 1992).

* On several testis preparations of the

nominative form we observed a picture

similar to that of the preparations in Group

II.

Literature Cited

BANNIKOV, A. G., I. S. DAREVSKY, V. G.

ISCHENKO, A. K. RUSTAMOV AND N. N.

Vol. 5 p. 108

Asiatic Herpetological Research

December 1993

SHCHERBAK. 1977. [Field guide of the USSR

amphibians and reptiles]. Prosveschenje

Publishing House, Moscow. 369 pp. (in

Russian).

BEDRIAGA, YA. 1909. Amphibien und Reptilien.

Pp. 73-502. In: Wissenschaftliche Resultate

der Reisen N. M. Przewalskijs durch

Zentralasien. Zoologische Teil. Band 3. Part

1. Lacertilia. Sankt-Petersbourgh. (In

Russian/German).

FORD, C. E., AND J. L. HAMERTON. 1956. A

colchicine hipotonic citrate squash sequence for

mammal's chromosomes. Staining Technology

31:247-251.

GOLUBEV, M. L. 1989. [Phrynocephalus

guttatus (Gmel.) or P. versicolor Str. (Reptilia,

Agamidae): which Phrynocephalus species

occurs in Kazakhstan?]. Zoological News, Kiev

(5):38-46. (In Russian).

GOLUBEV, M. L. 1992. [Variegated toad agama

Phrynocephalus versicolor (Reptilia: Agamidae)

of the Djungar Gate (Eastern Kazakhstan) with

notes on systematics of the species].

Zoological News, Kiev no. 2:31-38. (In

Russian).

LEVAN, A., K. FREDGA, AND A. A. SANDBERG.

1964. Nomenclature for centromeric position

on chromosomes. Hereditas 52:201-220.

MACGREGOR, H. C., AND J. M. VARLEY. 1986.

[Working with animal chromosomes]. Mir

Publishing House, Moscow. 272 pp. (In

Russian).

MANILO, V. V. 1986. [Karyotypes of gecko

genera Alsophylax and Crossobamon].

Zoological News, Kiev (5):46-54. (In Russian).

NIKOLSKY, A. M. 1915. [Fauna of Russia and

adjacent countries. Reptiles. Vol.1. Chelonia

and Sauria]. Imperial Academy of Sciences,

Petrograd. 532 pp. (In Russian).

PARASKIV, K. P. 1956. [Reptiles of

Kazakhstan]. Kazakh Academy of Sciences

Press, Alma Ata. 228 pp. (In Russian).

PETERS, G. 1984. Die krotenkopfagamen

Zentralasiens (Agamidae: Phrynocephalus).

Mitteilungen aus dem Zoologischen Museum in

Berlin. Akademie Verlag, Berlin 60(l):23-67.

SEMENOV, D. V., Z. K. BRUSHKO, R. A.

KUBYKIN, AND G. I. SHENBROT. 1987.

[Taxonomic position and protective status of the

round-headed lizard (Reptilia, Agamidae) in the

territory of the USSR], Zoological Journal,

Moscow 68(12);79-87. (In Russian).

SOKOLOVSKY, V. V. 1974. [A comparative

karyological study of lizards of the family

Agamidae. 1. Chromosome complements of 8

species of the genus Phrynocephalus (Reptilia,

Agamidae)]. Cytology, Moscow 16(7):920-

925. (In Russian).

SOLOLOVSKY, V. V. 1977. [Systematic relations

in the family Agamidae based on karyological

data]. P. 195 In: Questions of herpetology

abstracts of the report at the Fourth All Union

Conference of Herpetology, Nauka, Leningrad.

(In Russian).

TERENTYEV, P. V. AND S. A. CHERNOV. 1949.

[Guide to reptiles and amphibians]. Soviet

Sciences Publishing, Moscow. 315 pp. (In

Russian).

December 1993

Asiatic Herpetological Research

Vol. 5, pp. 109-111

A Karyosystematic Study of the Plate Tailed Geckos of the

Genus Teratoscincus (Sauria, Gekkonidae)

VALENTINA v. manilo

The Schmalhausen Institute of Zoology. Academy of Science of the Ukraine, Kiev, Ukraine

Abstract. -Karyotypes of two subspecies of Teratoscincus scincus are described (T. s. scincus and T. s.

rustamowi ). Both have 2n=36, and 46 arms in the karyotype (N. F. =46). A minor difference in

centromere position in two of the smaller chromosome pairs was noted, and may characterize the respective

subspecies sampled, or may reflect intra-population variation or even error in preparation. The karyotypes

differ from an earlier published description for T. scincus, in which de Smet (1981) reported 2n=34, N. F.

=42. Whether this represents intraspecific variation, error in preparation or interpretation, or a suggestion

that the name T. scincus is being applied to more than a single species will only be resolved with further

systematic study of this gekkonid lizard.

Key words: Reptilia, Sauria, Gekkonidae, Teratoscincus scincus, Kazakhstan, Tadjikistan, karyology.

TABLE 1. Karyotypic data for Teratoscincus scincus. Legends: M- macrochromosome, m-

microchromosome, v- metacentric, sT- subtelocentric, a (A)- acrocentric, NF- basic number.

Species, subspecies

Chromosomal formula

2n

NF

Author

T. scincus

T. s. scincus

T. s. rustamowi

4sT+4v+26A

24M(6sT+ 1 8 A)+ 1 2m(4v+8a)

24M(6sT+ 1 8 A+ 1 2m(4v+8a)

Introduction

The Central Asian Gekkonid genus

Teratoscincus is comprised of four

recognized species (Szczerhak and

Golubev, 1986). Karyotype data are

available only for the species Teratoscincus

scincus (de Smet, 1981). T. scincus is

presently divided into three subspecies: T

s. scincus; T. s. rustamowi; and T s.

keyserlingii. This paper provides

karyotypic descriptions of two races of T.

scincus.

Methods

A total of seven females and four males

representing four populations from

Turkmenistan (20 km north of Bami

station; 50 km north of Bakhardok; 45 km

north of Ashkabad; and the vicinity of

Gyaurs) were studied. Also, a Kazakhstan

population (the Chimkent Region, Syutkent

Settlement) was sampled, as were three

additional males of T. s. rustamowi from

Tadjikistan (Leninabad Region, in the

vicinity of Yakkatarak Settlement).

Chromosomal samples were prepared

from cellular suspensions of bone marrow,

blood and testis by the smear method and

by using the method of "digging out" as

described in Ford and Hamerton (1956)

and McGregor and Varley (1986), as

modified in part by Manilo (1986).

Cellular mitotic activity was increased by

injections of phytohemagglutinine solution

(0.02 ml/g body mass) and chorionic

gonadotrophin (50 units/g body mass).

Chromosome preparations were stained

with Giemsa (2% solution) in a 0.01 M

sodium-phosphate buffer (pH 6.8) for 20-

30 minutes. After washing in distilled

water, the preparations were passed

through alcohols and xylols (ortho-xylol)

and subsequently embedded in Canadian

balsam.

An NU-2 microscope with a 100x10

magnification was used for microscopy and

photomicrography. Chromosomes are

described using the centromeric position to

define morphology following the

Vol. 5 p. 110

Asiatic Herpetological Research

December 1993

AO »fl />fl °* ,Ml "M

ri^

IS ^1 *• •• •• ,0

< «

a •

An oi *• •*_ *» "^

A #j ~ a «* •»• »— ~ ~

AA *»*v **• *-«• - - ** -

|!

i i <

D

FIG. 1. Teratoscincus scincus scincus. a- mitotic

metaphase of a dividing cell of bone marrow; b-

bivalents of diakinesis; c, d- karyotype of female

and male, respectively; e- idiogram of the

karyotype.

classification suggested by Levan et al.

(1964).

Karyotype descriptions

Teratoscincus scincus scincus (Schlegel,

1858).

Type locality: The Hi River in

"Turkestan"

The diploid chromosome consists of 36

chromosomes. The karyotype is

provisionally divisible into 24

macrochromosomes (M) and 12

microchromosomes (m), but there is no

sharp demarcation between M and m.

Chromosomes decrease in size gradually

from largest (pair 1) to smallest (pair 18).

Chromosome pairs 4, 7, and 9 appear

subtelocentric; pairs 14 and 15 are

metacentric; and the remaining pairs are

FIG. 2. Teratoscincus scincus rustamowi. a-

mitotic metaphase of a dividing blood cell; b, c-

male karyotype; d- idiogram of the karyotype.

acrocentric. The chromosomal formula

could be stated: 2n=24 M (6 sT+18A) +

12m (4v+8a) = 36. The "fundamental

number" (N. F.) is 46. Sex chromosomes

are not evident. Male and female

karyotypes do not appear to differ in

chromosome number or morphology. (Fig.

1).

In male meiosis, the number of bivalents

at diakinesis is 18. The bivalents which

correspond to macrochromosomes have a

ring-like shape; the smaller bivalents

(microchromosomes) have a rod-like shape

(Fig. lb).

Teratoscincus scincus rustamowi

(Szczerbak, 1979)

Type locality: Fergan Valley in the sands in

the vicinity of Cokand and Kairakkum.

As in T. s. scincus the diploid number is

December 1993

Asiatic Herpetological Research

Vol. 5 p. Ill

36 with a somewhat arbitrary division

between 24 macrochromosomes and 12

microchromosomes; no obvious sex

chromosome heteromorphism; and

chromosome pairs 4, 7, and 9 are

subtelocentric. One possible difference

noted between the subspecies, however, is

that pairs 13 and 15 (instead of 14 and 15)

appear to be metacentric. Meiotic material

was not studied (Fig. 2).

Comparative analysis of karyotypes of the

genus

The karyotype of T. scincus was first

described by de Smet (1981). He reported

2n=34 with the following formula:

4sT+4V+26A. The total number of arms

(Fundamental Number, or N. F.) was 42.

Subsequently Manilo and Pisanets (1984)

obtained a different result (2n=36). This

stimulated us to re-examine the group in a

more detailed manner (Table 1). Our

conclusion is that two of the three

subspecies have very similar karyotypes

(the karyotype of T. s. keyserlingii remains

unknown). The present studies with

relatively small sample sizes, do not permit

us to determine whether the difference we

describe between the two subspecies

represents individual variation or a real

difference.

Because de Smet did not indicate the

locality for his specimens, we cannot

determine whether his description of 2n=34

and our description of 2n=36 represent a

difference in diploid number among

populations; variation within populations;

or problems in preparation and description.

Clearly this group of geckos should be

studied more extensively.

Literature Cited

DE SMET, W. H. O. 1981. Description of the

orsein stained karyotypes of 136 lizard species

(Lacertilia, Reptilia) belonging to the families

Teiidae, Scincidae, Lacertidae, Cordylidae and

Varanidae (Austarchoglossa. Acta Zoologica et

pathologica, Antverpiensia 76:407-420.

FORD, C. E. AND J. L. HAMERTON. 1956. A

colchicine, hypotonic citrate squash sequence for

mammalian chromosomes. Stain Technology

31:247-251.

LEV AN, A., K. FREDGA, AND A. A. SANDBERG.

1964. Nomenclature for centromeric position

on chromosomes. Hereditas 52:201-202.

MACGREGOR, H. C. AND J. M. VARLEY. 1986.

Working with animal chromosomes. Mir

Publishing House, Moscow. 272 pp. (In

Russian).

MANILO, V. V. 1986. Karyotypes of geckos of

the genera Alsophylax and Crossobamon.

Vestnik Zoologii 5:46-54. (In Russian).

MANILO, V. V. AND YE. M. PISANETS. 1984.

Karyotype of the plate-tailed gecko

(Teratoscincus scincus) from the Turkmenia

territory. Vestnik Zoologii 5:83-84. (In

Russian).

SZCZERBAK, N. N. 1979. A new subspecies of

plate-tailed gecko (Teratoscincus scincus

rustamowi ssp. n., Sauria, Reptilia) from

Uzbekistan and systematics of the species.

Protection of Turkmenistan Nature 1979:129-

138.

SZCZERBAK, N. N. AND M. L. GOLUBEV. 1986.

Geckos of the USSR fauna and of adjacent

countries. Naukova dumka, Kiev. 231pp. (In

Russian)

Vol. 5, pp. 112-116

Asiatic Herpetological Research

December 1993

Resting Metabolic Rate in Three Age-groups of Alligator sinensis

Pei-chao Wang and Jiang-hua Zhang

Department of Biology, East China Normal University, Shanghai 200062, People's Republic of China

Abstract. -The resting metabolic rate in three age-groups of Alligator sinensis is influenced the

temperature, season, body weight and other factors. Among these factors, the effect of body weight,

temperature, season, which are greater than that of others. The relationship between the body weight and

the resting metabolism is not consistent with the "third-quarter" surface area law. In the equation M=aW^,

the value of "a" ranges from 0.009 to 0.028, and the value of "b" ranges from 0.522 to 0.591, so that b is

approximately equal to 2/3 in empirical equation.

Key Words: Reptilia, Crocodilia, Alligatoridae, Alligator sinensis, China, metabolism.

Introduction

Alligator sinensis is a kind of special and

precious reptile. It is classed as a first

grade protected form of wildlife in China.

Much research on the adaptability of

Alligator sinensis have been done in the

fields of morphology, distribution,

reproduction and so on (Chen, 1985; Chen

and Wang, 1984). Zhang (1986; 1989)

reported information about infant Alligator

sinensis whose weights ranged from 35 to

50 grams. Because energy metabolism is a

very important criterion for the adaptability

of Alligator sinensis, the purpose of our

work was to explore the regularity of daily

gain in total energy from diet, with relation

to digestibility and energy allocation and to

explore the resting metabolic rate of

Alligator sinensis. This paper deals with

the resting metabolic rate of three age

groups in Alligator sinensis.

Materials and Methods

The Alligator sinensis that we used were

provided by the Shanghai Zoo. The total

number of tested animals was 13. Five of

them were bom in 1980, and their average

body weight was 2.84 ± 0.053 (M±SD) kg

at the beginning of the experiment in 1986

and 3.126±0.053 kg at the end of 1987.

Four of them were born in 1981, and their

average body weight was 1.5410.59 kg at

the beginning of the experiment in 1986

and 2.13910.857 kg at the end of the

experiment in 1987. Four of them were

born 1982, and their average body weight

0 12 3 k 5 6

Dody weitht(Ke)

FIG. 1. Relationship between RMR and body

weight in Alligator sinensis. Open circle- May

(20°C); solid triangle- July (28°C); x- September

(24°C); *- October (20°C); solid circle- January

(12'C).

was 0.90010.284 kg at the beginning of

the experiment and 1.23110.536 kg at the

end of the experiment in 1987.

The experiment began in May 1986 and

ended in January 1987. These animals

were reared in the three ponds according

their age differences. These ponds were

simple artificial environments and the size

of each pond was 3x4 square meters.

December 1993

Asiatic Herpetological Research

Vol. 5 p. 113

Table 1 . The linear regression equations and the approximate surface area equations of Alligator sinensis.

TABLE 2. T-test of regressional coefficient.

Months

May

July

September October

January

T=C

t-value

Note: all t values are over t o.ooidf6=5.96 and t 0.ooidf5=6.86.

In this paper the measure of the resting

metabolic rate is in ml CVkg-'hr1 or ml

02/W° 56hr-!. The closed-system

respirator meter of Wang et al. (1980) was

used to measure the oxygen consumption

of Alligator sinensis under two different

temperatures. The first, 25°C, is the

contrast temperature that comes from the

adaptive temperature of Alligator

mississippiensis reported by Coulson and

Coulson (1986). The other is the seasonal

temperature that is derived from the average

temperature of each month in the last five

years in Shanghai (Table 1). The ingestive

food behavior in the animals was fasted to

avoid its effect on metabolism during the

measuring of oxygen consumption.

Results and Discussions

The relationship between the body weight

and the resting metabolic rate (RMR)

The relationship between the body

weight and the resting metabolic rate of

Alligator sinensis is summarized in Fig. 1.

The resting metabolic rate to unit weight

declines with the raise of individual weight

in each month or under each temperature.

That is, there is a negative correlation

between the body weight and the weight-

specific resting metabolic rate of Alligator

sinensis that corresponds with the surface

area law. The further analysis of the

correlation between the weight and the

10.28 16.19 17.54 10.51 11.10 8.80 10.31

1.07 1.15 1.24 0.71 0.79 0.62 0.75

100 100 100 100 100 100 100

mlO^g"1!!-1

Mean 8.52

S. E. 0.77

A group % 100

Born 1980 ml 02/W°-56rr l

Mean 13.26 16.39 27.98 28.87 17.44 18.40 14.12 16.84

S. E. 0.39 0.46 1.85 0.66 0.32 0.37 0.34 0.91

% 100 100 100 100 100 100 100 100

5.44

0.36

100

8.76

0.09

100

mlO^g-1!!"1

Mean 10.98 13.89 21.08 21.85 13.09 13.84 9.87 12.19 6.49

S. E. 0.64 0.72 2.35 2.76 1.33 1.45 0.79 1.51 0.66

B group % 128.9 135.1 130.2 124.6 124.5 124.7 100.8 118.2 119.3

Born 1981 ml O2/^-56!!'1

Mean 12.61 16.55 26.41 27.23 17.73 18.72 13.53 16.55 8.63

S. E. 0.53 0.35 0.42 0.52 0.13 0.23 0.22 0.62 0.67

% 95.1 101.0 94.5 94.3 101.7 101.7 95.8 98.3 98.5

mlO^g"1!!-1

Mean 13.87 18.62 25.08 27.42 15.18 16.52 11.61 14.52 8.47

S. E. 1.16 1.42 1.84 2.33 0.76 1.46 0.73 1.14 0.79

C group % 162.8 181.1 154.9 156.3 144.4 148.8 131.9 140.8 155.7

Born 1982 ml O^0-56^1

Mean 12.28 15.96 26.79 29.19 17.00 18.31 12.87 15.88 8.87

S. E. 0.50 0.49 1.44 0.80 0.61 0.04 0.76 0.14 0.20

% 92.6 97.4 95.8 101.1 97.5 99.5 91.2 94.3 101.3

resting metabolic rate of Alligator sinensis

begin by converting or "transforming"

observed values to their logarithms and the

linear regression equation and the

approximate surface area equation (Table 1)

which are formed on the base of their

logarithms according to the methods of

Avery (1979). In equation M=aW"b from

Table 1, the value "a" ranges from 0.009 to

0.028, the value "b" ranges from 0.522 to

0.591.

The significance of coefficients on the

linear regression equations in Table 1 are

also examined through the T-test (T=b/sb)

and the results are shown in the Table 2.

All values of "t" are larger than

to.ooidf5=6.86 and to.ooidf5=5.96 (Table 2),

so the values of p are less than 0.001. We

may be to deduce that the body weight has

a great effect on the metabolic rate, and has

a similar effect in other crocodilians

(Coulson and Hernandez, 1983).

December 1993

Asiatic Herpetological Research

Vol. 5 p. 115

30

25

4=

s 15

7

Months

FIG. 2. Seasonal influence on the RMR of

Alligator sinensis maintained at 25°C. Solid circle-

1980 age group; x- 1981 age group; open circle-

1982 age group.

Effects of temperature and season

Table 3 shows that the resting metabolic

rate of Alligator sinensis affected by the

temperature and season. In the cool season

or under the low temperature, the resting

metabolic rate is at a lower level, and vice

versa. This is similar to that of the other

reptiles (Coulson and Coulson, 1986;

Huggins et al., 1971; Wang and Lu, 1986;

Wang and Xu, 1987; Wang et al., 1983,

1988). This metabolic character is due to

the result of acclimatization of seasonal

temperature rhythm in evolution of animals.

We further analyze the relationship

between RMR (resting metabolic rate) and

temperature as well as season. We used

0.56 power of body weight to adjust all

values of observing, so that the effect of

body weight in RMR is eliminated. The

results are summarized in Table 3. The

values of ml O2/W0-56hr1 from Table 3

show the temperature and season effect on

RMR. The levels of RMR are higher under

high temperatures than low, and it is a

similar state that there is a higher level of

RMR during hot seasons than during cool

seasons.

From Fig. 2, is is further shown that the

RMR changes with seasons under the same

temperature of 25°C. In July, the RMR is

the highest level; in September, the RMR

becomes lower and it is continues to fall in

October. This suggests that the energy

consumption is relevant to the seasonal

change which also corresponds with the

rules of the energy consumption and

requirements of Alligator sinensis. Among

the growth months of Alligator sinensis as

in July, there is a high water temperature,

and there is intensive metabolism and rapid

growth in Alligator sinensis. The data

below support this statement. The group

born in 1980 ingested daily 172.02±77.65

(M±SD) g fresh fish in June,

221.35±35.58 g fresh fish in July,

54.73±41.14 g fresh fish in September.

The daily body weight growth of the group

was also rapid, such as 10.3 g in June,

44.6 g in July and 15.8 g in September. In

May, Alligator sinensis had just awaked

from hibernation, when the RMR was

lower. In October, Alligator sinensis will

stop the feeding when the RMR declined to

a low level, and there were some changes

in their physiological attributes for the

coming hibernation stage. It is indicated

that Alligator sinensis has a series of

adaptive strategies for the seasonal

changes.

Tlie relationship between age and RMR

The relationships between age and RMR

in three age groups are shown in Table 3.

There are two kinds of data on the RMR.

The first RMR in Table 3 is influenced by

body weight, and does not eliminate the

effect of body weight. It is expressed in ml

02/kg1hr1. The second RMR eliminates

the effect of body weight by expressing

RMR in ml O2/W056hr1. A comparison

on the level of both kinds of RMR in three

age groups is based on 100 in RMR of age

groups in 1980. The results of comparison

suggest that the first kind of RMR falls as

the age of groups increases. The second

kind of RMR slightly falls in younger

groups, except in a few months.

Acknowledgments

This study was supported by the

Scientific Fund of the State Educational

Committee, China, East China Normal

Vol. 5 p. 116

Asiatic Herpetological Research

December 1993

University and Shanghai Zoo. We thank

Professor Ruyong Sun, Prof. Wenji

Huang, Prof. Ermi Zhao for helpful

discusssions and encouragement. We

would also like to thank Dr. Ted Joanen,

Dr. E. Norber Smith, and Dr. Shoji

Tokunaga for providing some literature.

Literature Cited

AVERY, R. A. 1979. Lizards-- A study in

thermoregulation. Printed by Lidio Ltd., East

Rilbride Scotland.

CHEN, B. H. AND C. L. WANG. 1984. Artificial

reproduction of Alligator sinensis. Acta

Herpetologica Sinica (new ser.) 3(2):49-54. (In

Chinese, with English abstract).

CHEN, B. H. 1985. Relationship between the

metabolic rate and seasonal changes in activity

of Alligator sinensis. Acta Herpetologica

Sinica (new ser.) 4(3):173-176. (In Chinese,

with English abstract).

COULSON, R. A. AND T. D. COULSON. 1986.

Effect of temperature on the rates of digestion,

amino acid absorption and assimilation in the

Alligator. Comparative Biochemistry and

Physiology 83A(l):585-588.

COULSON, R. A. AND T. HERNANDEZ. 1983.

Alligator metabolism. Comparative

Biochemistry and Physiology 74B(1): 1-182.

HUGGINS, S. E., H. E. HOFF AND M. E.

VALENTINUZZI. 1971. Oxygen consumption

of small caimans under basal conditions.

Physiological Zoology 44:40-47.

WANG, P. C, S. ZHAO, H. J. LU, L. B. ZHU,

AND Z. X. CHI. 1980. A simple closed-system

respirometer for measuring the oxygen

consumption of the terrestrial vertebrate.

Journal of East China Normal University

(Natural Science) 2:126-131. (In Chinese, with

English abstract).

WANG, P. C, K. C. QIAN, H. J. LU, L. B. ZHU,

AND S. ZHAO. 1983. Studies on physiological

ecology of Pallas' pit-viper. Acta Herpetologica

Sinica (new ser.) 2(1): 19-32. (In Chinese, with

English abstract).

WANG, P. C. AND H. J. LU 1986. Heat

metabolism and Uiermoregulation of Cope's rat

snake. Acta Herpetologica Sinica (new ser.)

5(1): 10- 16. (In Chinese,with English abstract).

WANG, P. C, AND H. F. XU. 1987. Influence of

ambient temperature on body temperature and

heat energy metabolism of Takydromus

septentrionalis. Acta Herpetologica Sinica (new

ser.) 6(2):10-15. (In Chinese, with English

abstract).

WANG, P. C, H. F. XU, W. MA, AND X. Jl.

1988. The influence of ambient temperature on

the body temperature and energy metabolism in

Chinemys reevesii. Acta Herpetologica Sinica

(new ser.) 7(2):122-127. (In Chinese, with

English abstract).

December 1993

Asiatic Herpetological Research

Vol. 5, pp. 117-126

Effects of Chinese Snake Venoms on Blood Coagulation, Purified

Coagulation Factors and Synthetic Chromogenic Substrates

YUN ZHANG1-2, YULIANG XlONG2 AND CASSIAN BON1

'Unite des Venins, Unite associee Instilul Pasteur/INSERM 285, 25, rue du Dr. Roiix, 75724 Paris Cedex 15

France

* Kunming Institute of Zoology, Academia Sinica, Kunming 650107, Yunnan, China

Abstract. -We examined the action of venoms from common Chinese Crotalidae and Elapidae snakes on

blood coagulation mechanisms. Procoagulant effects were observed with venoms from Agkistrodon acutus,

Trimeresurus stejnegeri, Ophiophagus luinnah and Bungarus fasciatus, the latter two only in the presence of

Ca2+. After treatment with a serine protease inhibitor (phenylmethanesulfonyl fluoride, PMSF),

Agkistrodon acutus venom lost its ability to clot purified fibrinogen but retained its capacity to clot human

plasma in the absence of Ca2+. An anticoagulant action was obtained with venoms from Trimeresurus

mucrosquamatus, Agkistrodon halys and Naja naja atra. This action was abolished after treatment with a

specific inhibitor of PLA2 activity (p-bromophenacyl bromide, BrPBr), revealing a procoagulant action with

high concentrations of treated venoms in the cases of Trimeresurus mucrosquamatus and Agkistrodon halys.

The effects of these venoms on hemostasis have been further characterized by measuring their phospholipase

A2 activity, their ability to hydrolyze synthetic chromogenic substrates and to activate purified blood

coagulation factors (prodirombin, factor X, protein C and plasminogen). These venoms showed an

amidolytic activity which was mainly due to serine proteases (90 to 95% of inhibition with PMSF).

Combining the observations obtained with human plasma and purified blood coagulation factors, we

concluded Uiat: i) six of the eight tested Chinese venoms {i.e.: Ophiophagus hannah, Bungarus fasciatus,

Agkistrodon acutus, Trimeresurus mucrosquamatus, Trimeresurus stejnegeri and Naja naja atra) contain

components which activate factor X in a Ca2+-dependent manner; ii) three venoms (Agkistrodon acutus,

Agkistrodon halys and Trimeresurus stejnegeri) contain prothrombin activators; iii) Ophiophagus hannah

venom has a weak protein C activating activity; and iv) Trimeresurus stejnegeri venom possesses

plasminogen activating activity. In addition, several of these venoms have previously been shown to

contain thrombin-like and fibrinogenolytic enzymes, anticoagulant phospholipases A2 (PLA2s) and/or non

enzymatic anticoagulant components.

Key Words: Snake venoms, blood coagulation, purified blood coagulation factors, chromogenic substrates.

Introduction

Snake venoms are known to be a rich

source of hydrolytic enzymes, mainly

proteases and phospholipases A 2 and of

non-enzymatic proteins, which induce

disorders of blood coagulation, hemorrhage

and shock (Pirkle and Markland, 1988;

Ouyang and Teng, 1972; Teng and

Seegers, 1981). Many proteases acting on

different steps of the blood coagulation

cascade have been purified from snake

venoms. They cleave blood coagulation

factors, either in a specific or in a non-

specific manner, and cause acceleration or

retardation of blood coagulation (Pirkle and

Markland, 1988). Some of these

proteases, such as thrombin-like enzymes

(Stocker and Meier, 1988) or protein C

activators (Kisiel et al., 1987), are serine

proteases which may be rapidly and

irreversibly inactivated by alkylation with

PMSF. Other procoagulant or

anticoagulant proteases, like factor X or

prothrombin activators from Bothrops atrox

venom (Hofmann and Bon, 1987a; 1987b)

or from Echis carinatus venom (Morita and

Iwanaga, 1978) are insensitive to PMSF

and have been postulated to be

metalloenzymes. PLA2s have also been

recognized for their anticoagulant activity,

which has been attributed to their ability to

antagonize the procoagulant action of

negatively charged phospholipids (Ouyang

et al., 1978).

In order to better understand the

pathophysiological action of snake venoms

on haemostasis, and to examine the

potential use of their procoagulant or

anticoagulant components as

pharmacological tools, we examined the

Vol. 5 p. 118

Asiatic Herpetological Research

December 1993

effects of various snake venoms on blood

coagulation mechanisms in vitro, using

human plasma, purified blood coagulation

factors (fibrinogen, prothrombin, factor X,

protein C and plasminogen), and synthetic

chromogenic substrates. We examined in

detail venoms from the eight most common

venomous snakes in China; four belonging

to the Elapidae family {Ophiophagus

hannah, Naja naja atra, Bungarus fasciatus

and Bungarus multicinctus) and the other

four to the Crotalidae family (Trimeresurus

mucrosquamatus, Trimeresurus stejnegeri,

Agkistrodon halys and Agkistrodon

acutus).

Methods

Venoms were supplied by the Kunming

Institute of Zoology (Academia Sinica,

China). The venoms were collected from

snakes living in the southern provinces of

China and stored desiccated. They were

dissolved in 50 mM Tris-HCl buffer, pH

7.8, at a concentration of 1 mg-ml"1 and

were used immediately.

Bovine factor X, human prothrombin

and human Glu-plasminogen were obtained

from Sigma (St. Louis, MO, USA).

Human protein C was obtained from

Diagnostica Stago (Asnieres, France).

Human fibrinogen (grade L) from Kabi

Vitrum (Stockholm, Sweden) was treated

with diisopropylfluorophosphate according

to the instructions of the manufacturer, in

order to irreversibly inactivate traces of

thrombin or other blood coagulation

factors. Platelet-poor human plasma was

the supernatant of human blood mixed with

1/10 volume of 3.8% sodium citrate and

centrifuged at 3000 rpm for 15 min. Pools

of normal citrated plasma obtained from 5-

10 healthy donors were stored at -20°C.

Chromogenic substrates H-D-Phe-Pip-

Arg-pNA (S-2238), H-D-Val-Leu-Lys-

pNA (S-2251), H-D-Val-Leu-Arg-pNA (S-

2266), H-D-Pro-Phe-Arg-pNA (S-2302)

and Bz-Ile-Glu-Gly-Arg-pNA (S-2222)

were obtained from Kabi Vitrum

^Stockholm, Sweden) and the chromogenic

substrate H-D-Lys(Cbo)-Pro-Arg-pNA

365-25) was from Diagnostica Stago

(Asnieres, France).

Phenylmethanesulfonyl fluoride (PMSF)

and /?-bromophenacyl bromide (BrPBr)

were purchased from Sigma (St. Louis,

MO, USA). All other reagents were of the

highest purity available.

Chemical modifications

Inactivation of serine proteases by

PMSF was performed in 50 mM Tris-HCl,

pH 7.8. Venom samples (2 mg-ml"1) were

incubated at 37°C for two hours with 5 mM

PMSF (stock solution: 0.1 M in

dimethylsulfoxid). Inactivation of PLA2s

was carried out in the same buffer by

incubating the venom (1 mg-ml"1) at 37°C

for one hour with 2 mM BrPBr (stock

solution: 0.1 M in acetone). Treated

venom samples were then dialyzed for 4 to

8 hours against large volumes of the same

buffer.

Chromogenic assays

Amidolytic activity was measured with a

Kontron spectrophotometer in 1 cm path-

length plastic cuvettes. Assays were

performed in 500 ml of 50 mM Tris-HCl,

pH 7.8, containing the appropriate

chromogenic substrate (0.2 mM). The

reactions was initiated by addition of the

sample to be tested (5 mg-ml"1 to 100

mg-ml"1, final concentrations) and the

formation of p-nitroanilide was monitored

at 405 nm. The amount of substrate

hydrolyzed was calculated using a molar

extinction coefficient of 10,000 M^-cm"1

for free p-nitroanilide.

Determination ofPLA2 activity

PLA2 activity was determined by the

titrimetric method described by Desnuelle et

al. (1955), according to the procedure of

Radvanyi and Bon (1982).

Effects of the venoms on blood coagulation

Citrated platelet-poor human plasma

(200 ml) was incubated at 37°C for 1 min,

then a 20 ml aliquot of diluted venom

samples was added and clotting time was

recorded. In some cases, 5 ml of CaCl2

December 1993

Asiatic Herpetological Research

Vol. 5 p. 119

6-

5""

4 -

B. multicinctus

i i i r"

0-1 1 1 1— — r

.001.01 .1 1 10 100

.001 .01

10 100

.001 .01

10 100

.001 .01

10 100

Venom concentration ((ig/ml)

Venom concentration (|!g/ml)

FIGURE 1. Effects of Elapidcie snake venoms on blood coagulation. Citrated platelet-poor human plasma

(200 ml) was incubated at 37°C for 1 min; dilutions of the sample to be tested (20 ml) were then added

simultaneously with 5 ml of 0.45 M CaCl2 (10 mM final concentration); native venom ( 9), PMSF-

treated venom (A) orp-bromophenacyl bromide-treated venom ( I). Values are the mean of triplicates.

Vol. 5 p. 120

Asiatic Herpetological Research

December 1993

TABLE 1. Phospholipase A2 activity of the venoms from the common Chinese venomous snakes

(unmolmiir'mg'1).

native venom

treated venoms

Agkistrodon acutus

Agkistrodon halys

Trimeresurus mucrosquamatiis

Trimeresurus stejnegeri

Ophiophagus harmah

Bungarus fascia tus

Bungarus multicinctus

Naja naja

The phospholipase A2 activity of the venoms were determined as described by Radvanyi and Bon (1982)

using egg lecithin solubilized by sodium cholate. Venoms were treated by p-bromophenacyl bromide as

indicated in the method then dialyzed to remove the excess of reagent. Each value is the mean of three

determinations ± standard deviation.

(10 mM final concentration) were added

simultaneously with the venom samples.

Thromhin-like activity was determined by

measuring the clotting time of purified

human fibrinogen (0.5%) in 50 mM Tris-

HCl, pH 7.8, containing 0.1 M NaCl.

Fibrinogen (200 ml) was incubated for 2

min at 37 °C before addition of 20 ml of

diluted venom samples.

Activation of blood coagulation factors by

snake venoms

Activation of prothrombin and factor X

was performed as described by Hofmann

and Bon (1987a; 1987b). Briefly, purified

human prothrombin (50 mg-ml"1) was

incubated at 37°C in 50 mM Tris-HCl, pH

7.8, containing 0.1 M NaCl and different

concentrations of the samples to be tested.

Aliquots (50 ml) were removed at various

times and their amidolytic activity was

tested in 500 ml of the same buffer

containing S-2238 (0.2 mM). Purified

bovine factor X (25 mg-ml"1) was

incubated at 37°C in 50 mM Tris-HCl, pH

7.8, containing 0.1 M NaCl, 10 mM CaCl,

and different concentrations of the samples

to be tested. Aliquots (50 ml) were

removed at various times and their

amidolytic activity was immediately

assayed in 500 ml of the same buffer

containing S-2222 (0.2 mM).

Protein C activation was assayed

according to the method of Orthner et al.

(1988), with minor modifications. Human

protein C (5 mg-ml"1) was incubated at

37°C in 50 mM Tris-HCl, pH 7.8,

containing 1 mg-ml"1 polyethylene glycol,

5 mM EDTA and dilutions of the samples

to be tested. At various times, aliquots (50

ml) were taken to measure the amidolytic

activity of activated protein C, in 500 ml of

the same buffer containing CBS65-25 (0.3

mM).

Plasminogen activation assay

Human Glu-plasminogen (100 mg-ml"1)

was incubated at 37°C in 200 ml of 50 mM

Tris-HCl, pH 7.8, containing 0.1 M NaCl,

0.01% Tween-80, and different

concentrations of the samples to be tested.

Aliquots (50 ml) were taken at various

times and assayed for plasmin activity.

They were introduced into a plastic cuvette

containing 450 ml of the same buffer

December 1993

Asiatic Herpetological Research

Vol. 5 p. 121

TABLE 2. Amidolytic activity (nmol-mirr'-mg"1) of venoms from Chinese snakes.

The amidolytic activities of each venom were determined as described in Methods, with the indicated

substrates. Venoms were treated by PMSF as indicated in Methods, then dialyzed to remove the excess of

reagent. Indicated values are the means of three determinations (standard errors were less than 10%).

supplemented with S-2251 (0.3 mM) and

the formation of p-nitroanilide was

monitored at 405 nm.

Results

Procoagulant and anticoagulant properties

of the venoms

The procoagulant and anticoagulant

actions of the venoms from the eight most

common venomous snakes in China

(Ophiophagus hannah, Naja naja atra,

Bungarus fasciatus, Bungarus multicinctus,

Trimeresurus stejnegeri, Trimeresurus

mucrosquamatus, Agkistrodon halys and

Agkistrodon acutus) were examined with

human plasma in the presence and in the

absence of calcium ions. Each venom was

tested in its native form, after treatment

with a specific and irreversible inhibitor of

serine proteases (PMSF), or after treatment

with a specific and irreversible inhibitor of

PLA2s (BrPBr) (Volwerk et al., 1974).

As indicated in Figure 1, Bungarus

multicinctus venom did not modify blood

coagulation in vitro. Venoms from

Bungarus fasciatus and from Ophiophagus

hannah showed a procoagulant action,

dependent on the presence of Ca2+. They

were unable to clot purified fibrinogen

(result not shown), indicating the absence

of thrombin-like enzymes. Their

procoagulant effect might therefore result

either from an ability to convert

prothrombin into thrombin in a calcium-

dependent manner, or more probably, from

a direct or indirect activation of factor X

into factor Xa. The fact that a treatment of

these venoms with PMSF significantly but

not completely reduced their procoagulant

action (Figure 1) suggests that this effect is

due, at least in part, to serine protease(s).

The venom from Naja naja atra was

characterized by a pronounced

anticoagulant action (Figure 1). It did not

prevent clotting of purified human

Vol. 5 p. 122

Asiatic Herpetological Research

December 1993

fibrinogen in the presence of thrombin

(result not shown), indicating that the

anticoagulant action is not due to

fibrinogenolysis. The anticoagulant effect

of Naja naja atra venom was completely

and irreversibly prevented by treatment

with BrPBr (Figure 1) which inhibited

PLA2 activity of the venom (Table 1). This

activity is therefore due to one or several

anticoagulant PLA2s, as reported in the case

of many other snake venoms.

The effects of venoms from Chinese

Crotalidae snakes on blood coagulation

mechanisms (Figure 2) appeared much

more complex than those observed with

Elapidae venoms (Figure 1). The venoms

from Agkistrodon acutus and from

Trimeresurus stejnegeri were characterized

by a procoagulant action, observed both in

the presence and in the absence of Ca2+

(Figure 2). These venoms also clotted

purified fibrinogen in a calcium-

independent manner (result not shown),

indicating that they contain potent

thrombin-like enzymes, in agreement with

the observations reported by Ouyang e t al.

(1971) and by Liu and Xiong (1990).

Treatment of Agkistrodon acutus venom

with PMSF completely abolished its ability

to clot purified fibrinogen (result not

shown) and strongly reduced its

procoagulant action (Figure 2), as expected

since this effect is mainly due to thrombin-

like serine proteases. However, PMSF-

treated Agkistrodon acutus venom showed

a complex action, an anticoagulant activity

at 10 mg-ml"1, and a clotting activity at

higher concentrations (Figure 2),

suggesting that it contains other

procoagulant components, in addition to

thrombin-like enzymes. The anticoagulant

effect of PMSF-treated Agkistrodon acutus

venom is consistent with the presence of an

anticoagulant protein of 26 kD, which is

devoid of enzymatic activity and which

prevents the formation of thrombin by

binding to the prothrombin activation

complex (Teng and Seegers, 1981). In

addition to this non-enzymatic component,

Agkistrodon acutus venom might contain

anticoagulant PLA2s, since treatment with

BrPBr (Table 1) significantly increased its

procoagulant effect (Figure 2).

Agkistrodon halys and Trimeresurus

mucrosquamatus venoms were

characterized by an anticoagulant action

which was completely abolished after

treatment with BrPBr, but not with PMSF

(Figure 2). This indicates that they contain

potent anticoagulant PLA2s, as previously

reported for Trimeresurus mucrosquamatus

(Ouyang et al., 1978) and for Agkistrodon

halys (Chen et al., 1987). In fact,

treatment of Agkistrodon halys and

Trimeresurus mucrosquamatus venoms

with BrPBr suppressed the anticoagulant

activity of these venoms and revealed a

weak procoagulant activity (Figure 2),

indicating the presence of procoagulant

components. Further, this procoagulant

action was observed only in the presence of

Ca2+, suggesting that the procoagulant

components are able to convert

prothrombin into thrombin or to activate

factor X. High concentrations (100 mg-ml"

') of Agkistrodon halys venom were able to

clot purified fibrinogen (result not shown).

This may be explained by the presence of a

thrombin-like enzyme, which has been

purified (Guan et al., 1988) but the level of

this enzyme in the venom is low and its

activity is weak.

Action of the venoms on purified blood

coagulation factors

We determined me amidolytic activity of

the venoms on a number of chromogenic

substrates, classically used for assaying

blood coagulation factors: thrombin (S-

2238), factor Xa (S-2222), activated

protein C (CBS65-25), kallikrein (S-2302

and S-2266) and plasmin (S-2251).

Assays were performed with native,

PMSF-treated and BrPBr-treated venoms

(Table 2). The venoms from Elapidae

snakes did not present detectable activities,

with the exception of the venom from

Ophiophagus hannah which hydrolysed

several substrates (Table 2). The venoms

from Crotalidae snakes exhibited significant

activities towards most chromogenic

substrates, but with important species

differences: the venoms from

Trimeresurus snakes presented much

higher activities than those of Agkistrodon

snakes and the venom from Agkistrodon

December 1993

Asiatic Herpetological Research

Vol. 5 p. 123

c

E

c

o

.001 .01

1 00

.001 .01

1 00

10 i

.001 .01

100

,001 .01

1 00

Venom concentration ()jg/ml)

Venom concentration ((ig/ml)

Figure 2. Effects of Crotalidae snake venoms on blood coagulation. Citrated platelet-poor human plasma (200

ml) was incubated at 37°C for 1 min then the sample to be tested (20 ml) was added simultaneously with

calcium (5 ml of 0.45 M CaCl2', 10 mM final concentration; closed symbols) or without calcium (open

symbols and dashed lines). Native venom (9; O); PMSF-treated venom (A; /\or p-bromophenacyl

bromide-treated venom ( I: D). The values are the mean of triplicates, standard errors being 10% of the values.

Vol. 5 p. 124

Asiatic Herpetological Research

December 1993

TABLE 3: Activation of factor X, prothrombin, protein C and plasminogen by the venoms from

common Chinese snakes.

The activation of the indicated blood coagulation factors was determined as described in Materials and

Methods by measuring the amidolytic activity of the factors after activation. Each value is the mean of

at least three independent determination, standard errors being less than ±10%. ND means not determined.

The results were expressed as (lie A O.D./min at 405 nm acquired in analysed solution divide incubation

time and divide venom concentration in activation solution.

acutus was 10 times more active on

substrate S-2238 than that from

Agkistrodon halys; in contrast substrate S-

2251 was hydrolysed by the venom of

Agkistrodon halys but not by the venom of

Agkistrodon acutus. These results are in

agreement with the general concept that the

proteolytic activities of the venoms from

Elapidae snakes are much lower than those

of Crotalidae snakes.

The procoagulant action of the various

venompurified bovine factor X, human

prothrombin, protein C and plasminogen,

and measuring their amidolytic activity after

activation (Table 3). Except for

Agkistrodon halys and Bun gar us

multicinctus, all venoms were able to

activate factor X, those from Ophiophagus

hannah and Bungarus fasciatus being far

more active than the others. Furthermore

the ability of Bungarus fasciatus venom to

activate factor X was markedly reduced

after treatment with PMSF (Table 3),

suggesting that the venom components

responsible for this activity may be serine

proteases. It should however be noticed

that neither Ophiophagus hannah nor

Bungarus fasciatus venoms showed any

activity on prothrombin or plasminogen,

emphasizing their rather specific action on

factor X.

On the other hand, three venoms

(Agkistrodon acutus, Agkistrodon halys

and Trimeresurus stejnegeri) among the

eight which have been tested, possessed

componentss was further examined in

detail, using which converted prothrombin

into thrombin. In all cases, this activity

was insensitive to PMSF, suggesting that it

is not due to serine proteases. Table 3 also

shows that Ophiophagus hannah venom

was able to activate protein C with a low

activity compared to that observed in the

venom of Agkistrodon contortrix contortrix

(Kisiel et al., 1987). Interestingly, the

venom from Trimeresurus stejnegeri was

characterized by the capacity to activate

plasminogen in vitro (Table 3). This action

December 1993

Asiatic Herpetological Research

Vol. 5 p. 125

appeared to be due to (a) serine protease(s),

since it was completely abolished by

PMSF.

We observed no correlation between the

amidolytic activity of snake venoms,

measured with the chromogenic thrombin

substrate (S-2238) and their thrombin-like

activity as determined by their ability to clot

fibrinogen (result not shown). Similarly,

there was no correlation between activation

of prothrombin (Table 3) and the amidolytic

activity measured with factor Xa substrate

S-2222 (Table 2). These results emphasize

the differences which exist between the

substrate specificity of human coagulation

factors and snake venom activators, and the

existence in snake venoms of proteases

which are able to hydrolyse chromogenic

substrates without possessing the capacity

to activate the corresponding blood

coagulation factors.

Discussion

In the present study, we found that,

except for Bungarus multicinctus, the

venoms from common Chinese venomous

snakes (Ophiophagus hannah, Naja naja

atra, Bungarus fasciatus, Trimeresurus

stejnegeri, Trimeresurus mucrosquamatus,

Agkistrodon acutus and Agkistrodon halys)

possessed procoagulant and/or

anticoagulant activities. An in vitro

analysis of these venoms indicated that their

action on blood coagulation results from the

combined effects of several procoagulant

and anticoagulant components. In

particular, comparing the effects of native

venom with those of venom in which PLA2

activity has been blocked revealed the

presence of anticoagulant PLA2s in

Agkistrodon halys venom, similar to that

descrided by Chen et al. (1987), which

masked the effect of procoagulant

component(s). We also showed that the

procoagulant action of Agkistrodon acutus

venoms does not result only from the

previously described thrombin-like enzyme

(Ouyang et al., 1971), but also from at least

two other components, a PMSF-insensitive

prothrombin activator and a calcium-

dependent factor X activator. This

illustrates the complexity of action of the

venoms on blood coagulation mechanisms.

We demonstrated the presence of

prothrombin activators in Agkistrodon

acutus, Agkistrodon halys and

Trimeresurus stejnegeri venoms, as well as

of factor X activators in Agkistrodon

acutus, Trimeresurus mucrosquamatus,

Trimeresurus stejnegeri, Ophiophagus

hannah, Bungarus fasciatus and Naja naja

atra venoms. These studies further

indicated that prothrombin and factor X

activators from Crotalidae (Agkistrodon

acutus, Agkistrodon halys, Trimeresurus

mucrosquamatus and Trimeresurus

stejnegeri) venoms are PMSF-insensitive,

and Ca2+-dependent in the case of factor X

activators. These activators might be

similar to those of Bothops atrox and

Vipera russelli venoms (Kisiel et al., 1976;

Hoffman and Bon, 1987a; 1987b).

Interestingly, factor X activators found in

Elapidae venoms (Ophiophagus hannah and

Bungarus fasciatus) were inactivated after

u-eatment with PMSF, suggesting that they

might be serine proteases.

We also showed the existence of a

protein C activator in the venom of

Ophiophagus hannah, although its activity

was low compared to that found in the

venom of Agkistrodon contortrix contortrix

(Kisiel et al., 1987; Orthneret al., 1988).

These studies also revealed the first

evidence of the existence of a plasminogen

activator. Trimeresurus stejnegeri contains

a PMSF-sensitive plasminogen activator.

Its biochemical structure and mechanism of

action are currently under investigation.

Acknowledgments

This research was supported in part by

funds from the Direction des Recherches et

Etudes Techniques. Zhang Yun is the

recipient of a fellowship from the

Fondation pour la Recherche Medicale.

References

CHEN, Y. C, J. M. MARAGANORE, I. REARDON,

AND R. L. HEINRIKSON. 1987.

Characterization of the structure and function of

Vol. 5 p. 126

Asiatic Herpetological Research

December 1993

three phospholipases A2 from the venom of

Agkistrodon halys pallas.. Toxicon 25:401-

409.

DESNUELLE, P., J. M. CONSTANTIN. AND J.

BALDY. 1955. Potentiometric assay of

pancreatic lipase activity. Bulletin tie la Societd

de Chimie Biologique 37:285-290.

GUAN, L. F., X. ZHANG, AND C. W. CHI. 1988.

Differences in fibrin polymerization and

fibrinopeptide A and B release induced by human

thrombin and by the thrombin-like enzyme from

the venom of Agkistrodon halys pallas. Pp. 93-

106. In: Pirkle, H. and F. S.' Markland (eds.),

Hemostasis and Animal Venoms. Marcel

Dekker, New York.

HOFMANN, H., AND C. BON. 1987a. Blood

coagulation induced by the venom of Bothrops

atrox. I. Identification, purification and

properties of a prothrombin activator.

Biochemistry 26:772-780.

HOFMANN, H., AND C. BON. 1987b. Blood

coagulation induced by die venom of Bothrops

atrox. II. Identification, purification and

properties of two factor X activators.

Biochemistry 26:780-787.

Journal of Biochemistry (Japan) 83:559-570.

(In Japanese).

ORTHNER, C. L., P. BHATTACHARYA, AND D. K.

STRICKLAND. 1988. Protein C activator from

the venom of Agkistrodon contort rix contortrix.

Biochemistry 27:2558-2564.

OUYANG, C, J. S. HONG, AND C. M. TENG.

1971. Purification and properties of the

thrombin-like principle of Agkistrodon acutus

venom and its comparison with bovine

thrombin. Thrombosis et Diathesis

Haemorrhagia 26:224-234.

OUYANG, C, AND C. T. TENG. 1972.

Purification and properties of the anticoagulant

principle of Agkistrodon acutus venom.

Biochimica Biophysica Acta 278:155-162.

OUYANG, C, C. M. TENG, Y. C. CHEN, AND S.

C.LIN. 1978. Purification and characterization

of the anticoagulant principle of Trimeresurus

mucrosquamatus venom. Biochimica

Biophysica Acta 541:394-407.

PIRKLE, H., AND F. S. MARKLAND. 1988.

Hemostasis and Animal Venoms. Marcel

Dekker, New York. 486 pp.

KISIEL, W., M. A. HERMODSON, AND E. W.

DAVIE. 1976. Factor X activating enzyme

from Russell's viper venom: isolation and

characterization. Biochemistry 15:4901-4906.

KISIEL, W., S. KONDO, K. J. SMITH, B. A.

MCMULLEN, AND L. F. SMITH. 1987.

Characterization of a protein C activator from

Agkistrodon contortrix contortrix venom.

Journal of Biological Chemistry 262:12607-

12613.

RADVANYI, F. R., AND C. BON. 1982. Catalytic

activity and reactivity with /?-bromophenacyl

bromide of the phospholipase subunit of

crotoxin. Journal of Biological Chemistry

257:12616-12623.

STOCKER, K. F., AND J. MEIER. 1988.

Thrombin-like snake venom enzymes. Pp. 67-

84. In: Pirkle, H. and F. S. Markland (eds),

Hemostasis and Animal Venoms. Marcel

Dekker, New York.

LIU N. K., AND Y. L. XIONG. 1990. Purification

and characterization of thrombin-like enzymes

from the venom of Trimeresurus stejnegeri.

Zoological Research (China) 11:234-240. (In

Chinese).

MORITA, T., AND S. IWANAGA. 1978.

Purification and properties of die prothrombin

activator from the venom of Echis carinatus.

TENG, C. M., AND W. H. SEEGERS. 1981.

Agkistrodon acutus snake venom inhibits

prothrombinase complex formation.

Thrombosis Research 23:255-263.

VOLWERK, J. J., W. A. PIETERSON, AND G. H.

DE HAAS. 1974. Histidine at the active site of

phospholipase. Biochemistry 13:1446-1454.

December 1993

Asiatic Herpetological Research

Vol. 5, pp. 127-136J

Herpetogeographical Map of Turkmenistan

SAHAT SCHAMMAKOV, CHARI. ATAEV, AND ELDAR. A. RUSTAMOV

Institute of Zoology, Turkmenistan Academy of Sciences, Azadi Street 6, 744000 Ashgabad, Turkmenistan

Abstract. -The herpetological map presented in this paper shows the distribution and abundance of the

reptiles of Turkmenistan. The country is divided into 17 complexes and the 84 species and subspecies

found in Turkmenistan are listed as occurring in mountains, plains, or both.

Key words: Reptilia, Turkmenistan, biogeography, distribution.

Introduction

In the mid-1960's biogeography entered

a new state of development with the

practice of ecosystems mapping

(Chyel'tsov-Bebutov, 1963, 1964, 1970,

1976). We do not here discuss the

principles of the preparation and

classification of geographical maps which

depict animal population areas. We can

only note that they make-up the series of

sections included in many integrated

regional atlases. Special surveys

(Chel'tsov and Chibisova, 1976;

Chel'tsov-Bebutov et al., 1972) have dealt

with them as well. Nevertheless, the above

mentioned maps were prepared for birds

and mammals. Until now there have been

no geographical maps (as geographical

science visualized these) which present the

quantitative proportions of reptiles in the

total animal kingdom of any region

(Chel'tsov-Bebutov and Chibisova, 1976).

The three authors of this article (Ataev,

Rustamov, and Shammakov, 1989) created

a color version of the Herpetogeographical

Map of Turkmenistan in 1989. It was

presented in 1989 at the All-Union Seminar

dealing with the animal kingdom

registration and cadastre (in Ufa), the

Zoological Section of the Moscow

Naturalists Society (in Moscow) and the

Vll-th All-Union Herpetological

Conference (in Kiev). This article presents

a black and white version of the map,

giving no consideration to color qualitative

background, on the scale 1:2,000,000, to

be included into the Turkmen SSR

Geographical Atlas (Fig. 1).

Field data, gathered throughout the

whole Turkmenistan during 1960-1985

(Schammakov, 1981; Ataev, 1985), served

as the main sources for this map. Other

data were obtained from literature

(Rustamov, 1966, 1981; Ataev, 1975;

Rustamov and Schammakov, 1982; Ataev,

Rustamov and Schammakov, 1985;

Rustamov and Shcherbak, 1986; Makeev et

al., 1988). Topographic maps on the scale

of 1:1,000,000 and 1:500,000 were used

as the cartographic basis.

The taxonomic generalization level of the

topological contours shown on the map

were dependent on both its scale and an

analysis of data gathered by Ataev and

Schammakov, unfortunately, apart, not in

assemblage, with the zoogeographical

survey of the country by Rustamov. A

further point: the whole complex of a

habitat and the animal population, which it

supports, was taken as a unit undergone to

zoogeographical mapping (Chel'tsov-

Bebutov, 1963, 1964, 1976). We tried to

single out the larger habitats at a level of an

ecosystem (landscape or land system,

according to Christian, 1975), not of the

land unit, which is in close correlation with

both the chosen scale and the content of the

rest of the maps belonging to the Nature

Division of the Atlas. The map scale

provoked the necessity to single out such

complexes of the reptiles population

territorial aggregation, which should be

grouped into a definite unity with regard to

both common conditions of the habitats (the

integral components of which are those

aggregations) and the dominant species

prevailing in number. A total of 17

complexes as such were revealed. Thus,

Vol. 5, p. 128

Asiatic Herpetological Research

December 1993

the map was build up on consideration of

the habitats of reptiles and their species

composition and density. Any territorial

differentiation not proved by distinctions in

the reptile population was not, as a rule,

taken into account.

The reptile fauna of the Turkmenistan

(Table 1) includes 78 species (84

subspecies) which belong to 2 orders and

14 families. The fauna consists of 3

species (3 subspecies) of tortoises, 47 (51)

lizards, and 28 (30) snakes. The

information on the reptiles species and

population quantities distributed through

every complex is placed in a special table

that is not given in the atlas, as well as the

Table 1, because of the lack of space. One

needs this table because the map contours

contain no concrete figures on the general

density and species number of reptiles.

The reptile populations are characterized

only according to their appropriate

abundance levels. This is quite enough for

examining the general content of the map.

Nevertheless, we provide herpetologists

using this map with more concrete figures

(Table 2). Reptile distributions, their

abundance, and correlation are dependent

upon habitats diversity as well as the fauna

richness and specific ecologico-

geographical peculiarities (Rustamov,

1966, 1981; Ataev, 1975; Rustamov and

Schammakov, 1982). This, in turn, forms

the physiognomy of the 17 territorial

herpetological complexes.

To optimize the reptiles population

characteristics, the map legend was made

up of 2 parts: the table (placed at the

Supplement) and the text. In addition, the

insets give information on the fauna

composition and contain the out-scale signs

characterizing the loci of the habitats. The

tables series are arranged according to the

principle that permitted us to depict the

territorial structure of herpetological

complexes, although the map scale and

content give no possibility to illustrate the

morphological specification of the habitats

occupied by these complexes. For

example, the table-legend horizontal

columns present the main groups of the

territorial herpetological complexes revealed

on the basis of common ecosystems

availability within the compared habitats.

Those (groups) are: plain-desert (4

habitats), flood-plain valley (7), piedmont

semi-desert (3), and mountain-arid (3).

The vertical columns present the territorial

units obtained as a result of geographical

regionalization that, in our case, merely

ground the boundaries of the

herpetocomplexes. Such units of the

regionalization scheme (zoogeographical

regions) within Turkmenistan include: 1

area, 1 sub-area, 3 provinces, 4 districts, 6

regions and 10 sections (Rustamov and

Scherbak, 1986).

The text of the legend gives the reptile

population characteristics for every habitat

gone into either complex. In front of the

latter's name there is a circle under the

correspondent number, the color map has

qualitative background representing the

complex. The latter's name is followed by

the species number and the animals total

density index (individuals per ha). The text

of the legend is reduced in this article as the

abundance indices are brought out in the

special table (see Table 2).

Further reptile population characteristics

for every complex are presented with

species numeration of a fixed sequence:

first species which use large areas are

listed, then the stenotypic ones, which are

confined to individual, smaller habitats

within a contour. For example, clay

surface, solonchaks, construction sites,

etc., which are evidently differentiated due

to their decreased sizes. The species names

are arranged one after another according to

decreasing population number within the

habitat, of which a brief description is

given immediately prior to the species

enumeration (see the text of the legend).

The dominant species are followed by (1),

the codominant ones by (2), and the minor

species by (3). The dominant species are

defined by us as those whose number is

over 10 per hectare, codominant species

from two to nine per hectare, and minor

species only one per hectare.

Thus, the map shows the herpetological

territorial complexes differentiated

December 1993

Asiatic Herpetological Research

Vol. 5, p. 129

according to their species composition, total

abundance and dominance levels (with

regard to the species number) as well as

principle features of the territory's

morphology and its ecosystems structure,

including the pattern of soils and vegetation

cover.

Mapping had proved to be the most

effective means to manifest and analyze the

reptiles population richness throughout the

country. The present map can serve as the

data source to evaluate the actual situation

with the Turkmenistan reptile resources, or

to elaborate the practical measures on

resource use and conservation. The map

can be a help to anybody who will create

new, more detailed, large-scaled

herpetological maps of either Turkmenistan

or any other country.

TABLE 1. Reptiles of Turkmenistan. Su- USSR Red Data Book; T- Turkman SSR Red Data Book.

Mountains Plains

Mountains

& Plains

Order Testudines

Emys orbicularis (Linnaeus, 1758)

Mauremys caspica (Gmelin, 1774)

Agrionemys horsfieldi (Gray, 1844)

Order Squamata

Suborder Sauria

+

Phrynocephalus helioscopus helioscopusPaWas, 1771

P. interscapularis Lichtenstein, 1858

P. maculatus Anderson, 1872 (Su, T)

P. mystaceus mystaceus Pallas, 1776

P. raddei raddei Boettger, 1888

P. r. boettgeri Bedriaga, 1905

P. reticulatus reticulatus Eichwald, 1831

P. r. bannikovi Darevsky, Rustamov et Schammakov, 1976

P. rossikowi rossikowi Nikolsky, 1899(Su, T)

P. r. schammakowi Szczerbak et Golubev 1979, (Su, T)

Stellio caucasius caucasius +

Stellio c. triannulatus Ananjeva et Ataev, 1984

S. chernovi (Ananjeva, Peters et Rzepakovsky, 1981) +

S. erythrogaster Nikolsky, 1896 +

S. lehmanni Strauch, 1896 +

Trapelus sanguinolentus aralensis (Lichtenstein, 1823)

Pseitdopus apodus apodus Pallas, 1775 +

Alsophylax laevis Nikolsky, 1907 (Su, T)

A. loricatus szczerbaki Golubev et Sattorov, 1979 (Su, T)

A. pipiens (Pallas, 1814) (T)

Bunopus tuberculatus Blanford, 1874 (Su, T) +

Crossobamon eversnumni (Wiegmann, 1834)

Cyrtopodion caspius caspius Eichwald, 1831

C. fedtschenkoi (Strauch, 1 887) +

C. longipes microlepis Lantz, 1918 (Su, T) +

C. russowi (Strauch, 1887)

C. spinicauda (Strauch, 1887) (Su, T) +

C. turcmenicus (Szczerbak, 1978) (Su, T) +

Eublepharis turcmenicus Darevsky, 1977 (Su, T) +

Teratoscincus scincus scincus Schlegel, 1858

+

Vol. 5, p. 130 Asiatic Herpetological Research December 1993

Eremias arguta uzbekistanica Chernov, 1934 (T) - + -

E. grammica (Lichtenstein, 1823) - + -

E. intermedia (Strauch, 1876) - + -

K lineolata (Nikolsky, 1896) - +

E. nigrocellata Nikolsky, 1896 (T) - +

E. persica Blanford, 1874 - - +

E. regeli Bedriaga, 1905 (T) - - +

E. scripta scripta Strauch, 1867 - + -

E. strauch i kopetdaghica Szczerhak, 197 1 + -

E. velox velox Pallas, 1771 - - +

Lacerta raddei raddei Boettger, 1892 (T) +

L. strigata Eichwald, 1 83 1 - - +

Mesalina guttulata wotsonana Stoliczka, 1872 - + -

Ablepharus deserti Strauch, 1868 - +

A. pannonicus (Lichtenstein, 1 823) - - +

Chalcides ocellatas ocellatus Forskal, 1775 (Su, T) + - -

Eumeces schneideri princeps Eichwald, 1839 - - +

E. taeniolatus taeniolatus Blyth, 1854 - - +

Mabuya aurata septemtaeniata Reuss, 1 834 +

Ophiomorus chernovi Anderson et Leviton 1966 (Su, T) +

Varanus griseus caspius Eichwald, 1831 (Su, T) - - +

Suhorder Serpentes

Eryx elegans (Gray, 1849) (Su, T) +

E. miliaris miliaris Pallas, 1773 - + -

E. tataricus specious Tsarevsky, 1915 (T) + -

Boiga trigonatum melanocephalia Annandale, 1904 (Su, T) - - +

Coluber caspius Gmelin, 1789 (T) - - +

C. karelini karelini Brandt, 1838 - +

C. najadum najadum Eichwald, 1831 (T) + -

C. ravergieri Menetries, 1 832 - - +

C rhodorhachis rhodorhachis (Jan, 1865) - - +

C. r. ladacensis (Anderson, 1871) - - +

Eirenis medus (Gernov, 1949) +

Elaphe dione (Pallas, 1773) - +

E. quatuorlineata sauromates Pallas, 1814 (T) - + -

Lycodon striatus bicolor Nikolsky, 1903 (Su, T) + -

Lythorhynchus ridgewayi Boulenger, 1887 (Su, T) - - +

Natrix natrix persa Pallas, 1814 - + -

N. tessellata (Laurenti, 1768) - - +

Oligodon taeniolatus (Jordan, 1853) (Su, T) +

Psammophis lineolatum (Brandt, 1838) - +

P. schokari schokari Forskal, 1775 +

Pseudocyclophis persicus persicus Anderson, 1872 +

Ptyas mucosus nigric ans Cernov, 1949 (Su, T) - - +

Spalerosophis diadema schiraziana Jan, 1865 - - +

Telescopus rhynopoma (Blanford, 1874) (Su, T) + -

Agkistrodon halys caraganus Eichwald, 1 83 1 (T) - +

A. h. caucasicus Nikolsky, 1916 (T) +

Naja oxiana (Eichwald, 1831) (Su, T) - - +

Typhlops vennicularis Menem, 1820 + -

Echis multisquamatus Cherlin, 1981 - - +

Wiper a lebetina turanica Cernov, 1940 - - +

December 1993

Asiatic Herpetological Research

Vol. 5, p. 131

TABLE 2. Abundance and proportions of ecologico-systematic groups within the territorial complexes of

Turkmenistan. 1*- species number; 2*- individuals per hectare.

Systematic groups and abundance

Testudines

Sauria

Scrpentes

Total

Complexes

%

1

%

(•) Alaophylax laevia

(\ Alaophylax laricatua

^^ Bunopua tuberculatum

^-p. Cyrtopodion longapee

UphiOinorua chernovi

iremiaa regeli ,r).arguta

K.nigrocellata ;

£ryx tataricua

tAj Chalcidea ocellatua

I J Lacerta raddei

V

Cyrtopodion turcmenicua 30 Lacerta atrigata

L^ Coluber najadutn

Phrynocephalua roaaicorf'i

Te lea c opus rhynopo.na

total reptile density (sped esAa

1/ / X <50 \/'//X50-'00 \fv^W0-!50\

tortoises

150-200

preval

species

/r^Sp-prevajent

^^^ < species

0> secondary

-pedes

mountain

sped es

plain

species

is

ESS pn«kfF

1 if *rdp

FIG. 1. Herpetological map of Turkmenistan. The species composition of various habitats within each of

the 17 complexes is listed below. We define dominant species (1) as those that number over 10 per hectare,

co-dominant species (2) as those that number from 2-9 per hectare, and minor species (3) as those that

number one or less per hectare.

Vol. 5, p. 132 Asiatic Herpetological Research December 1993

1. South-Ustjurt Complex

Various types of northwestern Turkmenistan deserts — Cyrtopodion caspius (1), Eremias intermedia (2),

Trapelus sanguinolenta (2), Agrionemys horsfieldi (2), Eryx miliaris (2), Psammophis lineolatum (2),

Coluber karelini (3), Spalerosophis diadema (3). Clay, crushed-stone and solonchak habitats —

Phrynocephalus helioscopus (2). Sandy and clay — Varanus griseus (3), Naja oxiana (3), Boiga trigonatum

(3), Agkistrodon halys (3). Clay — Cyrtopodion russowi (2). Sandy — Phrynocephalus inter scapularis (1),

Teratoscincus scincus (1), Crossobamon eversmanni (2), Eremias grammica (2), Eremias scripta (2),

Phrynocephalus mystaceus (2).

2. Caspian Complex

Various types of eastern Caspian deserts — Cyrtopodion caspius(2), Eryx miliaris (2), Coluber karelini (3),

Eremias intermedia (2), Eremias velox (2), Trapelus sanguinolentus (2), Agrionemys horsfieldi (2), £c/;/.v

multisquamatus (2), Psammophis lineolatum (2), £c/!/5 multisquamatus (2), Sandy, clay and solonchak

habitats — Eremias lineolata (2). Clay, crushed stone and solonchak — Phrynocephalus helioscopus (2).

Sandy, clay and construction sites — Eumeces schneideri (2), Coluber rhodorhachis (3), Varanus griseus (3),

Mi/a oxiana (3), Boiga trigonatum (3). Sandy and clay — Agkistrodon halys (3). Sandy and crushed stone —

Phrynocephalus reticulums (1). Sandy and on construction sites — Coluber ravergieri (3). Clay and on

construction sites — Mabuya aurata (2). Sandy — Phrynocephalus interscapularis (1), Teratoscincus scincus

(1), Crossobamon eversmanni (3), Eremias grammica (3), Eremias scripta (3), Phrynocephalus mystaceus

(3). Clay — Elaphe quatuorlineata (3). Solonchak — Lythorhynchus ridgewayi (3). By water bodies — Natnx

tessellata (3).

i. Karakum Complex

Various types of Karakum deserts — Cyrtopodion russowi (2), Cyrtopodion caspius (2), Agrionemys

horsfieldi (2),Eryx miliaris (2),Coluber karelini (3), Eremias grammica (2),Trapelus sanguinolentus (2),

Psammophis lineolatum (2), Eremias velox (2), Echis multisquamatus (2), Ecto multisquamatus (2),

Spalerosophis diadema (3). Sandy, clay and solonchak habitats — Phrynocephalus raddei (2), Eremias

grammica (2). Clay, crushed stone and solonchak — Phrynocephalus helioscopus (2). Sandy, clay and on

construction sites — Eumeces schneideri (2), Coluber rhodorhachis (3), Varanus griseus (3), Naja oxiana (3),

Bo/^a trigonatum (3). Sandy, less common crushed stone — Phrynocephalus reticulums (1). Sandy, clay,

less common crushed stone — Mesalina guttulata (2). Sandy and on construction sites — Coluber ravergieri

(3). Clay and crushed stone — Lythorhynchus ridgewayi (3). Clay and on construction sites — Mabuya

aurata (2). Sandy — Phiynocephalus interscapularis (I), Teratoscincus scincus (1), Crossobamon eversmanni

(2), Eremias grammica (2), Eremias scripta (2), Phrynocephalus mystaceus (2), Vipera lebetina (2).

4. Sundukli Complex

Various types of Sundukli massif deserts — Cyrtopodion caspius (I), Eremias intermedia (2),Trapelus

sanguinolentus (2),Agrionemys horsfieldi (2),Eryx miliaris (2), Psammophis lineolatum (2),Coluber

karelini (3), Spalerosophis diadem (3). Sandy, clay and solonchak habitats — Phrynocephalus raddei

(2), Eremias lineolata (2),Eremias velox (2),Cyrtopodion russowi (2),Echis multisquamatus (2). Clay,

crushed-stone and solonchak) — Phiynocephalus helioscopus (2). Sandy and clay — Varanus griseus (?>),Naja

oxiana (3), Boiga trigonatumO). Sandy — Phrynocephalus interscapularis (1), Teratoscincus scincus (1),

Crossobamon eversmanni (2), Eremias grammica (2), Eremias scripta (2), Phrynocephalus mystaceus (2).

Crushed stone — Cyrtopodion fedtschenkoi (2).

5. Sarykamysh Complex

Various habitats of the Sarykamysh Depression — Coluber karelini (3), Eremias velox (2), Trapelus

sanguinolentus (2), Agrionemys horsfieldi (2), Cyrtopodion caspius (2), Psammophis lineolatum (2), Eryx

miliaris (2), Spalerosophis diadema (3), Varanus griseusO). Sandy, clay and solonchak habitats — Eremias

grammica (2). Sandy, clay, solonchak and on construction sites — Cyrtopodion russowi (2). Clay, crushed-

stone and solonchak — Phiynocephalus helioscopus (2). Sandy, clay and solonchak — Eremias lineolata (2).

Sandy, clay and on irrigated lands) — Agkistrodon halys (3). Sandy — Phrynocephalus interscapularis (1),

Teratoscincus scincus (1), Crossobamon eversmanni (2), Eremias grammica (2), Phrynocephalus mystaceus

December 1993 Asiatic Herpetological Research Vol. 5, p. 133

(2). Solonchak habitats, along collectors and canals, in settlements — Elaphe dione (3). On irrigated lands

and water bodies; — Natrix tessellata (2).

6. Uzboi Complex

Various habitats of Western Uzboi Valley — Coluber karelini (3), Cyrtopodion caspius (2), Agrionemys

horsfieldi (2), Trapelus sanguinolentus (2), Eremias velox (2), Echis multisquamatus (2), Eryx miliahs (2),

Psammophis lineolatum (2), Spalerosophis diadema (3), Coluber rhodorhachis (3), Varanus griseus (3),

Naja oxianaO). Sandy, clay, solonchak and on construction sites — Cyrtopodion russowi (2). Sandy, clay,

solonchak and crushed-stone — Eremias intermedia (2). Flood-plain, clay and on construction sites —

Mabuya aurata (2). Clay, crushed-stone and solonchak — Phrynocephalus helioscopus (2). Sandy, clay and

solonchak — Phrynocephalus raddei (2), Eremias lineolata (2). Flood plain — Emys orbicularis (2). Sandy —

Phrynocephalus interscapularis (1), Teratoscincus scincus (1), Crossobamon eversmanni (2), Eremias

scripta (2), Eremias grammica (2), Phrynocephalus mystaceus (2).

7. Atrek-Sumbar Complex

Various habitats of Atrek and Lower Sumbar valleys — Trapelus sanguinolentus (2), Eremias velox (2)

Agrionemys horsfieldi (2), Echis multisquamatus (2), Eumeces schneideri (2), Cyrtopodion caspius (2),

Coluber karelini (3), Elaphe dione (3), Varanus griseus (3),Psammophis lineolatum (2), Spalerosophis

diadema (3), Boiga trigonatum (3). Flood plains and irrigated lands — Natrix natrix (2), Natrix tessellata (2),

Ablepharus pannonicus (2), Coluber caspius (3). Sandy, clay, crushed stone, and solonchak habitats —

Eremias intermedia (2), Eryx miliahs (2). Clay, crushed-stone and solonchak — Phrynocephalus raddei (2).

Clay, crushed-stone and solonchak — Phrynocephalus helioscopus (2). By water bodies — Emys orbicularis

(2), Mauremys caspica (2). Sandy — Teratoscincus scincus (1), Crossobamon eversmanni (2).

8, Tedzhen-Hauzkhan Complex

Various habitats of Tedzhen Valley and Hauzkhan Massif — Eremias velox (2), Natrix tessellata (2),

Trapelus sanguinolentus (2), Agrionemys horsfieldi (2), Echis multisquamatus (2), Eumeces schneideri (2),

Mabuya aurata (2), Coluber karelini 0),Coluber rhodorhachis (3), Naja oxiana (3), Varanus griseus (3),

Coluber ravergieri (3), Cyrtopodion caspius (2), Psammophis lineolatum (2), Boiga trigonatum (3), Vipera

lebetina (2), Eumeces taeniolatus (3), Spalerosophis diadema (3), Eryx miliahs (2), Eremias intermedia (2).

Sandy, clay and solonchak habitats — Phrynocephalus raddei (2), Eremias lineolata (2). Clay, crushed-stone

and solonchak— Phrynocephalus helioscopus (2). Sandy and clay— Mesalina guttulata (3). Sandy—

Phrynocephalus interscapularis(l), Phrynocephalus mystaceus (2), Eremias grammica (2). Clay —

Lytorhynchus hdgewayiO).

9. Murgab Complex

Various habitats of the Murgab Valley— Eremias velox (2), Trapelus sanguinolentus (2), Agrionemys

horsfieldi (2), Cyrtopodion caspius (2), Echis multisquamatus (2), Eumeces schneideri (2), Psammophis

lineolatum (2), Vipera lebetina (2), Mabuya aurata (2), Coluber karelini (3), Coluber rhodorhachis (3),

Varanus griseus (3), Naja oxiana (3), Spalerosophis diadema (3), ftya5 mucosus (3), flo/ga trigonatum (3).

On flood-plains and irrigated lands— Ablepharus deserti (1), Ablepharus pannonicus (2), Atom* tessellata

(2), Eumeces taeniolatus (3). Sandy, clay, crushed-stone and solonchak habitats— Eremias intermedia (2),

Eryx miliahs (2). Sandy, clay, solonchak habitats and on construction sites— Cyrtopodion russowi (2).

Sandy, clay and solonchak habitats— Phrynocephalus raddei (2), Eremias lineolata (2). Sandy—

Phrynocephalus interscapularis (1), Teratoscincus scincus (I), Crossobamon eversmanni (2). Clay—

Lythorhynchus hdgewayi (3).

/0. A/nu Darya Complex

Various habitats of the Amu Darya Valley— Ablepharus deserti (1), Eremias velox (2), Natrix tessellata (2),

Trapelus sanguinolentus (2), Agnortemy.? horsfieldi (2), Cyrtopodion caspius (2), £cto multisquamatus

(2), Psammophis lineolatum (2), V/pera lebetina (2), Eumeces schneideri (2), Mabuya aurata (2), Coluber

karelini (3), Spalerosophis diadema (3), Varanus griseus (3), My'a ox/ana (3), Coluber ravergieh,0), Elaphe

dione (3), flo/ga trigonatum (3) Agkistrodon halys (3), Eumeces taeniolatus (3). Sandy, clay, crushed-stone

and solonchak habitats— £ryx miliahs (2), Eremias grammica (2). Clay, crushed stone and solonchak—

Vol. 5, p. 134 Asiatic Herpetological Research December 1993

Phrynocephalus helioscopus (2). Sandy, clay and crushed stone — Phrynocephalus raddei(2), Eremias

lineolata(2). Sandy and crushed stone — Phrynocephalus reticulatus(l). Clay and crushed stone —

Lythorhynchus ridgewayi (3). Construction sites — Cyrtopodion fedtschenkoi (2).

77. Kopetdag Piedmont Anthropogenic Complex

Various habitats of Kopetdag piedmont oases — Mabuya aurata (1), Cyrtopodion caspius (1), Eremias velox

(2), NatrLx tessellata (2), Trapelus sanguinolentus (2), Agrionemys horsfiekii (2), Coluber rhodorhachis (3),

Coluber ravergieri.O), Naja oxiana (3). In flood-plains of shallow rivers and on construction sites — Eryx

miliaris (2), Echis multisquamatus (2), Spalerosophis diadema (3). On flood plains — Eremias lineolata (2),

Lythorhynchus ridgewayi (3).

12. Kopetdag Piedmont Complex

Various habitats of the Kopetdag piedmont plain — Cyrtopodion caspius (1), Phrynocephalus raddei (2),

Phrynocephalus helioscopus (2), Eremias intermedia (2), Eremias lineolata (2), Eremias velox (2),

Cyrtopodion russowi (2), Trapelus sanguinolentus (2), Agrionemys horsfieldi (2), Eryx miliaris (2), Echis

multisquamatus (2), Psammophis lineolatum (2), Coluber karelini (3), Spalerosophis diadema (3), Varanus

griseus (3), Naja oxiana (3). Clay and crushed-stone habitats — Mabuya aurata (1), Lythorhynchus

ridgewayi (3). Sandy and clay — Mesalina guttulata (2), Boiga trigonalum (3). Clay — Pseudocyclophis

persicus (3), Eirenis medus (3). Sandy — Phrynocephalus interscapularis (1), Teratoscincus scincus (1),

Crossobamon eversmanni (2), Eremias grammica (2), Eremias scripta (2), Phrynocephalus mystaceus (2).

7j. Kugitang Piedmont Complex

Various habitats of Kugitang piedmont plain — Cyrtopodion caspius (1), Phrynocephalus raddei (2), Eremias

intermedia (2), Phrynocephalus helioscopus (2), Eremias lineolata (2), Eremias velox (2), Trapelus

sanguinolentus (2), Agrionemys horsfieldi (2), £nu miliaris (2), £c/n.t multisquamatus (2), Psammophis

lineolatum (2), Coluber karelini (3), Spalerosophis diadema (3). Clay and crushed-stone habitats —

Cyrtopodion fedtschenkoi (2), Lythorhynchus ridgewayi (3). Sandy and clay — Varamw griseus (3), TVa/a

oxiana (3). Boiga trigonalum (3). Sandy — Phrynocephalus interscapularis (1), Phrynocephalus mystaceus

(2) Eremias grammica (2), Eremias scripta (2).

74. Badghyz-Karabil Complex

Various habitats of the Badghyz and Karabil hills — Agrionemys horsfieldi (1), Mabuya aurata (2),

Cyrtopodion caspius (2), Trapelus sanguinolentus (2), Eremias velox (2), Ablepharus pannonicus (2),

,Sre//io erythrogaster (2), Eumeces taeniolatus (2), Eumeces schneideri (2), Pseudophus apodus (3), Vipera

lebetina (3), /Va/a oxiana (3), Psammophis lineolatum (3), Varanus griseus (3), Spalerosophis diadema (3),

Coluber ravergieri (3), Coluber rhodorhachis (3). On slopes covered by stones and mud-streams — Stellio

caucasius (2), Typhlops vermicularis (3). On bare rocks-outcrops — Lycodon striatus (3). On stone

surfaces — Pseudocyclophis persicus (3). On slopes covered by loess and stones — Psammophis schokari (3).

Food-plains and mud-streams — Eremias persica (2). Mud-streams — Oligodon taeniolatus (2). Flood

plains — Natrix tessellata (3).

75. Balkhan Complex

Various habitats of the Great and Small Balkhan Mountains — Cyrtopodion caspius (1), Stellio caucasius

(2), Trapelus sanguinolentus (2), Eremias velox (2), Agrionemys horsfieldi (2), Coluber rhodorhachis (3),

iVa/fl oxiana (3), Ablepharus pannonicus (3). Piedmonts and inter-ridge hills — Va/xmi/.? griseus (3). Mud

streams, undulated surfaces, piedmonts and inter-ridge hills — Psammophis lineolatum (3). Inter-ridge hills,

piedmonts and construction sites — Spalerosophis diadema (3).

76. Kopetdag Mountain Complex

Various habitats of the Kopetdag Mountains — Stellio caucasius (2), Ablepharus pannonicus (2),

Agr/onmy.s horsfieldi (2), Coluber rhodorhachis (3), Vipera lebetina (3), /Va/a oxiana (3), Cyrtopodion

caspius (1), Trapelus sanguinolentus (2), Coluber ravergieri (3), Pseudopus apodus (3), Mabuya aurata (1),

Eremias strauchi (2), Eremias velox (2), Typhlops vermicularis (2), Eumeces schneideri (2), Eumeces

December 1993

Asiatic Herpetological Research

Vol. 5, p. 135

FIG. 2. Supplement to herpetological map of Turkmenistan.

taeniolatus (2), Cyrtopodion spintccuida (3), Pseudocyclophis persicus (3). Steppe-like, stone and inter-ridge

hills — Coluber caspius (3). Stone inter-ridge hills and piedmonts — Lycodon striatus (3), Psammophis

schokari (3). Steppe-like inter-ridge hills and piedmonts — Eirenis niedus (3). Inter-ridge hills, piedmonts,

and on construction sites — Psammophis lineolatum (3), Spalerosophis diadema (3). Stone and inter-ridge

hills — Oligodon taeniolatus (3), Eublepharis turcmenicus (3). Piedmonts — Varanus gnseus (3). Along

shallow rivers and other water bodies — Natri.x tessellata (2). Juniper stands — Eryx elegans (3), Agkistrodon

halys (3).

17. Kugitang Mountain Complex

Various habitats of the Kugitang Mountains — Cyrtopodion fedtschenkoi (1), Agrionemys horsfieldi (2),

Vipera lebetina (3), Coluber ravergieri (3), Stellio lehmanni (2), Eremias velox (2), Naja oxiana (3),

Ablepharus pannonicus (2), Coluber rhodorhachis (3), Eumeces schneideri (2). Inter-ridge hills, piedmonts,

and on construction sites — Trapelus sanguinolenlus (2). Piedmonts, stone habitats, and on construction

sites — Cyrtopodion caspius (2). Stone habitats and juniper stands — Stellio chernovi (3). Piedmonts, stone

and inter-ridge hills — Spalerosophis diadema (3). Inter-ridge hills and stone surfaces — Pseudopus apodus

(3), Typhlops vermicularis (3). Stone surfaces and piedmonts — Psammophis lineolatum (3). Piedmonts —

Lytorhynchus ridgewayi (3). Along shallow rivers and other water bodies — Nairix tessellata (2).

Literature Cited

House, Ashkhabad. 343pp.

ATAEV, CH. A. 1975. Geograficheskii obzor

herpetofauni gor Turkemenistana.

(Geographical overview on the Turkmenistan

mountains herpetofauna). Izvestiya Akad. nauk

TSSR, ser. biol. nauk, N 3, pp.75-80.

ATAEV, CH. A. 1985

gor Turkmenistana.

Presmykayushchiesya

(Reptiles of the

Turkmenistan mountains). Ylym Publishing

ATAEV, CH. A, A. K. RUSTAMOV, AND

S. M. SCHAMMAKOV. 1985.

Presmykayushchiesya (Reptiles). Pp. 208-270.

In Krasnaya kniga Turkmenskoi SSR (Turkmen

SSR Red Data Book). Turkmenistan

Publishing House, Ashkhabad.

ATAEV, CH. A., E. A. RUSTAMOV, AND

S. M. SCHAMMAKOV. 1989. Opit

Vol. 5, p. 136

Asiatic Herpetological Research

December 1993

gerpetogeograficheskogo kartografirovaniya

Turkmenskoi SSR (Experience of

herpetogeographical mapping of Turkmen

SSR). Pp. 14-15. In Voprosy gerpetologii

(Problems of herpetology - 7th All-Union

Herpetological Conference). Naukova dumka

Publishing House, Kiev.

CHEL'TSOV-BEBUTOV, A. M. 1963.

Voprosi melkomasshtabnogo

zoogeograficheskogo kartografirovaniya na

primere karty Kustanaiskoi oblasti (On small-

scale zoogeographical mapping with special

reference to Kustanai region). Pp. 124-126. In

Tezisi dokladov po voprosam zoologicheskoi

kartografii (Theses of reports read at the seminar

on problems of zoological mapping). Moscow.

Chel'tsov-Bebutov, A. M. 1964. Nekotorie

voprosi zoogeograficheskogo kartografirovaniya

na primere karti Kustanaiskoi oblasti (Some

problems of zoogeographical mapping with

special reference to Kustanai region). Pp. 3-24.

In Biogeograficheskie ocherki Kustanaiskoi

oblasti (Biogeographical essays of Kustanai

region). Moscow.

Chel'tsov-Bebutov, A. M. 1970.

Zoogeograficheskoye kartografirovanie i

landshaftovedeniye (Zoogeographical mapping

and land system discipline). Pp. 49-94. In

Landshafmyi sbornik (Articles on land systems).

Moscow.

Chel'tsov-Bubutov, A. M. 1976.

Zoogeograficheskoye kartografirovanie:

osnovnie printsipi i polozheniya

(Zoogeographical mapping: main principles and

theses). Vestnik MGU, series geographical

2:50-56.

Chel'tsov-Bebutov, A. M., A. K. Danilenko, and

V. V. Chibisova. 1972. Karti zhivotnogo mira

v Sovetskih regional'nih atlasah (Animal

kingdom maps in Soviet Regional Atlases). In

Metody sozdaniya kompleksnih regional'nih

atlasov SSSR. Kati prirodi. (Methods of

creation of Regional Integrated Atlases of

USSR. The Environmental maps). Moscow.

Chel'tsov-Bebutov, A.M., and V. V. Chibisova.

1976. Zhivotnii mir i yego resursi (Animal

kingdom and its resources). Pp. 326-341. In

Kompleksnie regional'nie atlasi (Integrated

Regional Atlases). Moscow.

Christian, C. S. 1975. The concept of land units

and land systems. Humid Tropics 20:74-81 .

Makeev, V. M., A. T. Bozhanskii, S. V.

Kudryavtsev, V. I. Frolov, and Yu. D.

Khomustenko. 1988. Nekotorie resul'tati

gerpetologicheskogo obsledovaniya Vostochnoi

Turkmenii (Some results of herpetological

examination of the Eastern Turkmenistan). Pp.

127-143. In Redkie i maloizuchennie zhivotnie

Turkmenistana (Rare and poor-studied animals

of the Turkmenistan). Ylym Publishing House,

Ashkhabad.

Rustamov, A. K. 1966. Kratki obzor gerpetofauni

Turkmenii i eyo zoogeograficheskie osobennosti

(Brief overview of the Turkmenistan

herpetofauna and its zoogeographical

peculiarities). Pp. 158-168. In Pozvonochnie

zhivotnie Srednei Asii (Middle Asian

vertebrates). Tashkent.

Rustamov, A.,K. 1981. Opit otsenki vidovogo

yendimizma gerpetofauni Irana, Afganistana i

Srednei Azii (Experimental estimation of

specific endemism of herpetofauna of Iran,

Afganistan and Middle Asia). Pp. 118-119. In

Voprosi gerpetologii (Problems of herpetology),

Leningrad.

Rustamov, A. K., and S. M. Schammakov. 1982.

On the herpetofauna of Turkmenistan.

Vertebrata Hungarica, Budapest 21:215-226.

Rustamov, A. K., and N. N. Shcherbak. 1986.

Cierpetogeograficheskoye raionirovaniye Srednei

Azii (Herpetogeographical regionalization of

Middle Asia). Izvestiya Akad. nauk SSSR, ser.

biol., pp. 13-20.

Schammakov, S. M. 1981. Premykayushchiesya

ravninnogo Turkmenistana (Reptiles of the flat

Turkmenistan). Ylym Publishing House,

Ashkhabad. 310 pp.

December 1993

Asiatic Herpetological Research

Vol. 5, pp. 137-142

A Karyosystematic Study of the Genus Bombina from China

(Amphibia: Discoglossidae)

WAN-ZHAO LIU AND DA-TONG YANG

Department of Vertebrate Zoology. Kunming Institute of Zoology, Academia Sinica,

Kunming. Yunnan. China

Abstract. -Chromosome number, morphology and positions of Ag-NORs were determined for four

Chinese species of Bombina. Chromosome numbers are: B. orientalis (2n=24, NF=48), B. maxima

(2n=28, NF=56), B. microdeladigitora (2n=28, NF=56). The Ag-NORs of B. orientalis are located on the

long arm of the 7th chromosome pair, where as those of the latter three are on the short arm of the 1 1th

pair. The subdivision of Bombina into two subgenera is supported by the karyology. A close relation

between Bombina and Discoglossus is suggested.

Key words: Anura, Discoglossidae, Bombina; karyotypes, Ag-NORs, China.

TABLE 1. Species, localities, and number of individuals used in this karyological study.

Species

Locality

No. of Individuals

B. (G.) fortinuptialis

B. (G) maxima

B. (G) microdeladigitora

B. (B.) orientalis

Jinxiu, Guangxi

Dayao, Yunnan

Jingdon, Yunnan

Qindao, Shandong

Introduction

There are five genera in the family

Discoglossidae. The systematics of this

family have long been under discussion.

The systematic position of the genus

Bombina within this family is the most

problematical. A variety of studies dealing

with this genus have been presented during

the past years (summarized by Lang, 1988;

1989a; 1989b). However, the systematics

of Bombina is still quite confusing. The

relationships within Bombina have not been

fully worked out.

Only six species belong to the genus

Bombina. All are distributed in Eurasia.

Karyological data are known for

B. bombina (2n=24, NF=48), B. variegata

(2n=24, NF=48) [Morescalchi, 1973], B.

orientalis (2n=24, NF=48) [Jiang et al.,

1984], B. maxima (2n=28, NF=56) [Zhao,

1986, no photographs presented]. No

chromosome banding data are available for

Bombina.

Careful morphological analysis and

banding of the chromosomes may yield

useful information not only on the

phylogeny of the genus itself, but also on

the relationships between Bombina and

other genera of the family Discoglossidae.

The purpose of this study is to analyze the

karyotypes and Ag-NORs of four species

in the genus from China Liu and Hu,

1961). The results, when compared with

known karyological data of B. bombina, B.

variegata and other genera of

Discoglossidae, should be helpful in

understanding the taxonomy and phylogeny

of Bombina.

Materials and Methods

The specimens used in this investigation

are listed in Table 1. The toads were

collected by the authors at time of the year

when both sexes are active for mating.

The specimens were kept at room

temperature (15-20°C) until the time of

investigation.

To block mitosis at metaphase, we used

a freshly prepared colchicine solution of

0.05%, and injected intraperitonealy l/20ml

of this solution per gram of body weight.

The animals were sacrificed 20-24 hrs later,

the spleen and small intestine were

Vol. 5 p. 138

Asiatic Herpetological Research

December 1993

TABLE 2. Observational results of the diploid chromosome number of four species of Bombina from

China.

B. (G.)fortinuplialis

percentage (%)

113

1 12 99 1

0.88 10.6 87.6 0.88

B. (G.) maxima

percentage (%)

198

3

1.5

3

1.5

14

7

176

88.9

B. (G.) microdeladigitora

percentage (%)

238

2 4 22 209 1

0.8 1.7 9.2 87.8 0.4

B. (G.) orientalis

percentage (%)

116 2 8 104

1.7 6.9 89.5

2

1.7

removed, and the intestine was opened with

a pair of fine scissors to expose the inner

epithelial surface. The exposed epithelial

surface was washed with a 0.64% NaCl

solution for several minutes in order to

remove all mucous and debris. The tissue

was cut into small pieces and placed in a

Petri-dish.

We added 8-10 ml of 0.4% KC1 into the

dish, suspended vigorously with a Pasteur

pipette. The hypotonic treatment lasted 30-

40 minutes. It was centrifuged at 1000 rpm

for five minutes, and then the hypotonic

solution was removed. The tissue was

fixed with 8- 10ml of freshly prepared

solution of 3:l(v/v) absolute methanol-

glacial acetic acid for three times, with a

total time of 60 minutes. The samples will

keep indefinitely in the fixative if stored at

1-4°C.

We prepared slides by transferring 3-4

pieces of the fixed tissue onto a dry, warm

(about 50°C) slide. We then added 5-10

drops of 60% acetic acid and siphoned the

solution up and down until the solution

evaporated completely. Slides were stained

in 10% Giemsa (pH 6.8) for 10 minutes.

Staining of nucleolus organizer regions

(NORs) followed the methods of Howell et

al. (1980).

A total of 675 mitotic chromosome

spreads were observed. Ten selected

metaphase plates for each species were

photographed, enlarged, and measured.

The chromosome nomenclature used is mat

suggested by Levan et al. (1964). For the

convenience of comparison, the

chromosomes are defined as being large (A

group), medium (B group) and small (C

group) according to their relative lengths.

Large chromosomes have a value of 100

units or more, small chromosomes have a

value of 40 or fewer units. Chromosomes

whose length falls between 40-99 units are

considered to be medium.

Results

The observed diploid chromosome

numbers are presented in Table 2.

Measurements of metaphase chromosomes

of four Chinese species of Bombina are

shown in Table 3.

It is obvious that, the karyotypes of

maxima, microdeladigitora and

fortimiptialis are equal to each other. They

have 2n=28, NF=56, composed of 6 pairs

of large homologous, one pair of medium-

small chromosomes and seven pairs of

small homologous; all the chromosomes are

metacentric (m), except for 6th, 7th, and

9th pair, which are submetacentric (sm). A

weak secondary constriction is observed on

the short arm of 1 1th pair. It appears in

about 10% of the cases. The Ag-NORs are

observed on short arm of 11th pair and

December 1993

Asiatic Herpetological Research

Vol. 5 p. 139

TABLE 3. Measurements of metapha.se chromosomes of four Chinese

B. fortinuptialis

species of Bombina Mean ± SE

B. maxima

relative length= (chromosome length/total of haploid chromosome length) x 1000

ratio= long arm/short arm

coincide with the position of secondary

constriction. B. orientalis has 2n=24,

NF=48, consisting of 6 pairs of large

homologues, one pair of medium-large

chromosomes and 5 pairs of smaller

homologues. All the chromosomes are m,

except for the 6th pair, which is sm. A

clear secondary constriction is found on the

long arm of 6th pair, and a weak one is

observed on the short arm of the 8th pair,

the latter appears in a case of 5%. Ag-

NORs were only observed on the long arm

of 7th pair. No heteromophic

chromosomes were found. The karyotypes

Vol. 5 p. 140

Asiatic Herpetological Research

December 1993

i; ii ir is ii ii /

\1\

1

iwam ^

A* »K •* ** M ** *'

V?

10 n

I IS II II II II

!! if If 11 if ii

:#» •»••»»- • <

II

J5ji_

;; ;; ji :•* n a

tt ** •• «« »»

> /•

FIG. 1. Karyotypes and Ag NO3 stained karyotypes of Bombina from China. Arrows show Ag-NORs. A:

B. fortinuuptialis; B: B. maxima; C: S. microdeladigitora; D: £. orientalis.

are presented in Fig. 1.

Discussion

Now, karyotypes are known for all the

recognized species of Bombina. We

compare the karyotypes and Ag-NORs of

them in Table 4.

All the chromosomes of Bombina have

median or submedian centromeres. In the

discoglossids, Alytes are rich in

acrocentrics, and with some

microchromosomes (2n=38, NF=64-72),

Discoglossus have 2n=28, NF=54, with

one pair of telocentrics (Morescalchi,

1973). So, from the karyololgical point of

view, Bombina is the most highly

differentiated.

Within Bombina, two different kinds of

karyotypes exist. The differences between

the two are mainly as follows: 1). The

morphology of 7th pair are quite different.

The 7th pair of maxima, microdeladigitora

and fortinuptialis are medium-small and s,

where as those of bombina, variegata and

orientalis are medium-large, m, with a clear

secondary constriction on the long arm. 2).

The number of smaller homologues is

different. The former three have 7 pairs of

small homologues, but the latter three have

only 5 small homologues. Tian and Hu

(1985) suggested a subdivision of Bombina

into two subgenera, the subgenus Bombina

containing the Palaearctic bombina,

variegata and orientalis, and the Oriental

Glandula containing maxima,

microdeladigitora and fortinuptialis. Our

December 1993

Asiatic Herpetological Research

Vol. 5 p. 141

TABLE 4. Comparison of karyotypes and Ag-NORs in Bombina.

Species

NF Chromosome

formula

Secondary

centremere

Locality of

A-NORs

Data

B. (G.) fort in uptialis

28 56 22m+6sm

No. 11

Present study

B. (G.) maxima

28 56 22m+6sm

No. 11

Present study

B. (G. ) microdeladigitora

28 56 22m+6sm

No. 11

Present study

B. (G.) orienlalis

24 48 22m+2sm

Nos. 7,8

No. 7

Present study

B. (G. ) bombina

24 48

unknown

Nos. 7,8

unknown

Morescalchi 1973

B. (G.) variegata

24 48

unknown

Nos. 7,8

unknown

Morescalchi 1973

karyological evidence support this

subdivision.

In B. maxima, microdeladigitora and

fortirtuptialis, the karyotypes are practically

equal to each other which indicates that

these three species may have diverged

recently. In the group consisting of B.

bombina, variegata and orientalis, bombina

and variegata are equal to one another

(Morescalchi, 1973), but orientalis has

some differences from them. The 8th and

12th pairs are m in orientalis, but st in

bombina and variegata. Thus the two

European species are more closely related,

which is congruent with immunological

evidence of Maxon (1979) and Maxson and

Szymura(1984).

Morescalchi et al. (1977) could not

resolve relationships within Discoglossidae

because the karyotypes of Discoglossus,

Alytes and Bombina are so different from

each other. Our investigation indicated that

Discoglossus and the Glandnla-group of

Bombina have the same diploid

chromosome number (Lanza et al., 1975;

1976). We suggest that those two genera

may be related. The secondary constriction

is the only "marker" currently available for

analysis in most anuran karyosystematics

studies. However, in the present study, we

found that the secondary constriction is

abrupt and the position of it is quite

variable. The Ag-NORs are stable and

clear and may be a more useful tool than the

place of the secondary constriction in some

cases. The mechanisms of karyotype

evolution of Bombina may take place

through unequal translocation. This is still

an open question. A further chromosome

banding study is necessary.

References

JIANG, S., WEN, C, SHEN, C. and MEN, Y.

1984. Preliminary observations on karyotypes

of Bombina orientalis. Acta Herpetologica

Sinica 3(l):25-27. (In Chinese).

LANG, M. 1988. Notes on the genus Bombina

Oken (Anura: Bombinatoridae). I. Recognized

species, distribution, characteristics and use in

laboratory. British Herpetological Society

Bulletin No. 26:6-13.

LANG, M. 1989a. Notes on the genus Bombina

Oken (Anura: Bombinatoridae). II. Life history

aspect. British Herpetological Society Bulletin

No. 27:13-17.

LANG, M. 1989b. Notes on the genus Bombina

Oken (Anura: Bombinatoridae). III. Anatomy,

systematics, hybridization, fossil record and

biogeography. British Herpetological Society

Bulletin No. 28:43-49.

LANZA, B„ J. M. CEI AND E. G. CRESPO.

1975. Immunological evidence for the specific

status of Discoglossus pictus Otth. 1837 and D.

sardus Tschudi 1837, with notes on the families

Discoglossidae Giinther, 1858 and Bombinidae

Fitzinger 1826 (Amphibia: Salientia).

Monitore Zoologico Italiano (N.S.) 10:153-162.

Vol. 5 p. 142

Asiatic Herpetological Research

December 1993

LANZA, B., J. M. CEI and E. G. CRESPO.

1976. Further immunological evidence for die

validity of die family Bombinidae (Amphibia:

Salientia). Monitore Zoologico Italaliano

(N.S.) 10:311-314.

LEVAN, A., K. FREDGA, AND A. A. SANDBERG.

1964. Nomenclature for centromeric position

on chromosomes. Hereditas 52:201-220.

LIU, C. C. and S. Q. HU. 1961. Tail-less

amphibians of China. Science Press, Beijing.

1-358 pp.

MAXSON, L. R. and J. M. SZYMURA. 1984.

Relationships among Discoglossid frogs: An

albumin perspective. Amphibia-Reptilia 5:245-

252.

MAXSON, L. R. 1979. Quantitative

immunological studies of the albumins of

several species of fire-bellied toads, genus

Bombina.. Comparative Biochemestry and

Physiology 63B:517-519.

MICHALOWSKI, J. 1961 Studies on species

characters in Bombina variegata (L.) and B.

bombina (L.): I. Applying the L:T indicator to

the classifying purposes. Acta Zoologica

Cracoviensia 6(3):51-59.

MORESCALCHI, A. 1973. Amphibia. Pp. 233-

348 In A. B. chiarelli and C. Capanna (eds).

Cytotaxonomy and vertebrate evolution.

Academic Press, New York.

MORESCALCHI, A. 1977. Phylogenetic species

of karyological evidence. Pp. 149-167. In

Hecht, M. K., P. C. Gody and B. M. Hecht

(eds.), Major patterns in vertebrate evolution.

Plenum Press, New York.

MORESCALCHI, A. , E. OLMO and V.

STINGO. 1977. Trends of karyological

evolution in pelobatid frogs. Experientia

3:1577-1578.

TIAN, W. S. AND Q. X. HU. 1985. Taxonomical

studies on the primajtive anurans of the

Hengduan Mountains, with descriptions of a

new subfamily and subdivision of Bombina.

Acta Herpetologica Sinica4(3):219-224.

ZHAO Y. F. 1986. Studies of the karyotype of

Bombina maxima. Acta Herpetologica Sinica

5(3):227-228. (In Chinese).

December 1993

Asiatic Herpetological Research

Vol. 5, pp. 143-146

A Study on the Purification and Pharmacological Properties of Two

Neurotoxins from the Venom of the King Cobra (Ophiophagus Hannah)

JIANX1NG SONG, YULIANG XlONG, WANYU WANG, AND XlAOCHUN PU

Toxinologica! Laboratory, Kunming Institute of Zoology. Academia Sinica, Kunming, Yunnan, China

Abstract- Using Sephadex G-50, CM-Sephadex C-25 and CM-cellulose 52 columns, two neurotoxins of

Ophiophagus hannah were purified to be homogeneous on acidic PAGE which contain 53 and 73 amino

acid residues respectively. The two neurotoxins were used as a substitute for morphine. Using a morphine

withdrawal jumping model, the results demonstrated that the effects of the two neurotoxins when

administered by injecting are very significant (P<0.01), and when administered orally are significant

(P<0.05).

Key words: King Cobra, Ophiophagus hannah, neurotoxins, morphine addiction, naloxone jumping model.

Introduction

The King Cobra (Ophiophagus hannah)

is the largest poisonous snake in the world

(Tu, 1977). It is extensively distributed in

southern China, India, Thailand and other

Asian countries. Joubert (1973) reported

on the purification and sequence

determination of two toxins from the

venom of King Cobras grown in Thailand.

Shun (1981) reported on the purification of

four postsynaptically acting toxins from the

venom of the King Cobra in Guangxi,

China. Our research showed that among

four toxins, there is a postsynaptically

acting toxin containing only 63 amino acid

residues. We determined the complete

amino acid sequence of neurotoxin, which

contain 73 amino acid residues. This

neurotoxin's sequence is analogous to that

of the neurotoxins determined by Joubert

(1973), but its C-terminal four amino acid

residues are very similar to that of

bungartoxin in hydrophobicity (Lin and

Wang, 1984). Now, the sequence of the

neurotoxin containing 53 amino acid

residues is in the process of being

determined.

Xiong and Wang (1987) reported the

clinical observation results of using the

neurotoxin from cobra venom to achieve

better analgesic effect on moiphine addicted

patients. The possibility to use the

neurotoxins from snake venom as a

substitute for morphine was also first

reported by Xiong (1990). The mechanism

has been discussed in the other papers. In

this paper, the research concentrated on the

purifying of two neurotoxins from the

venom of King Cobras from Guangxi,

China and using them as a substitute for

morphine during tests on the mice-jumping

model.

Methods

Venom of King Cobras was purchased

from Guangxi Province. Male and female

mice were provided by the feed lot in our

institute. Naloxone and morphine were

purchased from Qinghai Medical Factory.

CM-Sephadex C-25, Sephadex G-50, and

CM-Cellulose 52 were provided by our

pharmacy. Other chemicals of A. R. grade

were produced in China.

1. The purification of two neurotoxins

from the venom of King Cobras: The dry

venom powder was dissolved in the buffer

solution (pH 5.8, 0.05 M HAc-NaAc

buffer), then the solution was centrifuged

to discard the insoluble material. The

supernatant was loaded on a Sephadex G-

50 column (25 x 200 cm). The same buffer

was used to elute the column. The

fractions which contained only low

molecular weight components were

collected (The process was directed by

acidic PAGE).

The collected material was desalted and

concentrated. It was then loaded on a CM-

Sephadex C-25 column (4x80 cm). The

column was eluted first by the equilibrium

buffer, then was eluted by the buffer

1993 by Asiatic Herpetological Research

Vol. 5 p. 144

Asiatic Herpetological Research

December 1993

{0<f27b5"432Li

FIG. 1. The Acidic PAGE of the different fractions

after separation with a Sephadex G-50 column (2.5

x 200 cm). 1: Crude venom; 2-11: Different

fraction after separation with a Sephadex G- 50

column. Fractions 7-11 were collected together for

the further isolation.

containing NaCl gradient (00.4) and finally

was eluted by the buffer containing NaCI

gradient (0.4-0.8). The two neurotoxins

were eluted during the first gradient

process.

The neurotoxins from CM- Sephadex C-

25 still showed two other minor bands on

acidic PAGE. So each of them was further

purified on the CM- Cellulose 52 Column.

The equilibrium buffer is 0.05M, pH 5.8

HAc-NaAc buffer. The NaCI gradient is

0-0.5M in the same buffer.

2. Determination of purities: Acidic

PAGE was used. The separating gel was

15 %, and the spacal gel was 2.5 %. The

gel was stained by R- 250 dissolved in 375

ml ethanol, 125 ml water and 5 ml

methanol.

3. Determination of amino acid

composition: The samples were analyzed

with a Hitachi Model 835-50 High Speed

Amino Acid Analyzer.

4. Determination of amino acid

sequence of the neurotoxin: The method

was in accordance with the method of Lin

and Wang (1984).

FIG. 2. The acidic PAGE of the neurotoxins from

CM-Sephadex C-25 CM-Cellulose columns. 1.

Crude venom; 2. The fraction containing

neurotoxins from Sephadex G-50; 3. Neurotoxin

1; 4. Neurotoxin 2 from CM Sephadex C-25; 5.

Neurotoxin 2 further purified on CM cellulose 52.

5. Determination of LD 50: Mice were

used as the material. The method was

according to Gu (1965).

6. Morphine withdrawal jumping

model: Mice were divided into random

groups, with each group consisting of 10

mice, half male and half female, with body

weights of 20±2g. The groups were

treated by injecting S. V. with morphine,

for four days, three times at a dose of 10

mg/kg, three times at a dose of 20 mg/kg

and 12 times at a dose of 30 mg/kg. After

morphine was given the last time, the

neurotoxins were administered either orally

or by injecting. The control group was

treated with physiological saline of 0.2 ml

per mouse. Three hours later, naloxone

was administered S. V. to different groups

at a dose of 15 ml/kg. Then the jumping

numbers of the mice during a 30 minute

period were recorded. The data were

analyzed with statistical methods.

Results

The lyophilized venom powder was

dissolved in the buffer (0.05M, pH 5.8.

December 1993

Asiatic Herpetological Research

Vol. 5 p. 145

TABLE 1. The amino acid compositions of neurotoxins 1 and 2.

HAc-Ac buffer), loaded on the Sephadex

G-50 column, then eluted with the same

buffer. The fractions containing

neurotoxins were put together, m desalted,

concentrated, then loaded on the CM-

sephadex C-25 column, which was eluted

first with equilibrium buffer, then with

different NaCI gradient. After the CM-

Sephadex C-25 column, the two

neurotoxins were further purified on a CM-

Cellulose column.

The crude venom was isolated first on a

Sephadex G-50 column, then CM-

Sephadex C-25, and finally, a CM-

Cellulose column. Figure 1 and 2 shows

the acidic PAGE of different fractions.

The amino acid sequence of neurotoxin 1

was determined by the Immobilized Phase

Edman Method as: 1 Thr • Lys • Cys • Tyr

• Val • Thr • Pro • Asp • Val • Lys • Ser •

Glu • Thr • Cys • Pro • Ala • Gly • Gin •

Asp • Leu • Cys • Tyr • Thr • Glu • Thr •

Trp • Cys • Val • Ala • Trp • Cys • Thr • Val

• Arg • Gal • Lys • Arg • Val • Ser • Leu •

Thr • Cys • Aal • Ala • He • Cys • Pro • He •

Val • Pro • Pro • Lys • Val • Ser • He • Lys •

Cys • Cys • Ser • Thr • Asp • Aal • Cys •

Gly • Pro • Phe

Asn • Val • Arg

Pro • Thr • Trp • Pro

The toxicity of neurotoxins 1 and 2 was

determined by the following method: The

mice, with body weight of 18-20 g, were

divided at random into groups of 5 mice.

The different doses of the neurotoxins were

given to the groups S. V. The death

numbers of mice within 24 hours were

recorded and the LD 50 was determined by

modified a Ginsberg method. For

neurotoxin I the LD 50=0.21±0.013. For

neurotoxin 2 the LD 50=0.24±0.009

For the morphine withdrawal jumping

model, the effects of purified neurotoxins

on mice jumping numbers and the results of

statistical analysis are shown in Table 2.

The results demonstrated that the two

neurotoxins display very significant effects

(P<0.01) as a substitute for morphine, by

injecting S. V. and significant effects

(P<0.05) when administered orally.

Discussion

The origination and evolutionary

relationships of neurotoxins and

Vol. 5 p. 146

Asiatic Herpetological Research

December 1993

phospholipase A2 from snake venom are

still disputed problems. Some experts

thought that the original molecule is the

postsynaptical toxins containing only 60-61

amino acid residues. By increasing the

cycles in the molecules, the short-chain

neurotoxins evolved into long-chain

neurotoxin containing 7374 amino acid

residues, then evolved into phospholipase

A2. On the other hand, some experts

thought Phospholipase A2 was the original

molecule which diverged into neurotoxins

(including postsynaptical and presynaptical

toxin), proteinase inhibitor and a new

neurotoxin (dendrotoxin). In King Cobra

venom, both short and long chain

neurotoxins were isolated. Especially, a

postsynaptical toxin containing 53 amino

acid residues was purified. So this

neurotoxin is very important to understand

the evolutionary relationship of the

neurotoxin. Now, we are focused on the

determination of its sequences and its

conformational properties in the solution by

means of 2D- NMR method (mainly by

different correlated spectrum and NOSY

spectrum).

By using the withdrawal jumping

model, two neurotoxins display significant

effects on substituting for morphine. The

clinical observation in Kunming also

demonstrated remarkable effects. These

facts suggest that these two neurotoxins

from the King Cobra venom may be a

better medicine to be used as a substitute

for morphine. Now, experiments have

been accomplished to understand its

mechanism.

References

GU, H. Y. 1965. Pharmacology (Vol. 1)

edited by C. S. Zheng. People Health

Publishing House. 393 pp. (In Chinese)

JOUBERT, E. J. 1973. The amino acid

sequences of two toxins from Opliiophagus

hannah venom. Biochemica et Biophysica

317:85-93.

LIN, N. Q. AND W. Y. WANG. 1984.

[The amino acid sequence of the neurotoxin

from the venom of King Cobra in Guangxi

Province, China]. Acta Biochemica et

Biophysica Sinica 16(6):592-596. (In Chinese).

SHUN, X. 1981. [The Study on the

purification of four neurotoxins from the venom

of king cobra (Ophiophagus hannah)].

Zoological Research 2(l):260-270. (In

Chinese).

TU, A. T. 1977. Venoms: chemistry and

molecular biology. New York.

XIONG, Y. L. 1990. The study on given up

addiction by using snake venom.

ABSTRACTS of Second International Congress

of Ethnobiology.

XIONG, Y. L. AND W. Y. WANG. 1987.

[The exploiting on snakes resources.]

Symposium on Exploiting Yunnan Biological

Resource. Yunnan People's Publishing House.

(In Chinese).

I December 1993

Asiatic Herpetological Research

Vol. 5, pp. 147-165

The Ecology of the Caucasian Salamander (Mertensiella caucasica Waga) in

a Local Population

DAVID N. TARKHNISHVILI1 AND IRINA. A. SERBINOVA2

'Institute of Zoology, Georgian Academy of Sciences, Tbilisi, Georgia

^Moscow Zoo, 123242 Moscow, Russia

Abstract. -The different aspects of ecology of Mertensiella caucasica Waga, 1876 were investigated in a

local population from Borjomi Canyon (central Georgia) for five years (1985-1990). The aspects of the

species' life cycle were more precisely determined. The main fecundity is about 16.9 eggs per female.

There are about 2 years in a period from egg deposition (June to first half of July) to the end of

metamorphosis in nature. Animals have spent most of the time in shelters after metamorphosis. They

appear on the ground surface at night during the breeding period. Commonly the adults don't retreat to a

great distance from population localities. Localities are situated in comparatively small plots (100-300m)

along the streams. Estimation of adult animal number showed that the population consists of 1189

specimens (1989). Annual adult survival is higher than known values of most amphibians (approaches

0.77). Larval survival is 0.27-0.32 in the second year of life. The characteristics of demography

(especially, low renewal rates) and spatial restriction in localities depends mostly on subUe constitution of

the species (which is a result of allometric growth specifics). The small recent geographical range of M.

caucasica is explained as a result of morphological and ecological peculiarities. General morphological

constitution limits adaptive possibilities of any particular representative of die European salamander tribe.

This is an explanation of quite high ecological similarity of M. caucasica and Chioglossa lusitanica.

Key Words: Amphibia, Caudata, Salamandridae, Mertensiella caucasica, Caucasus Mountains, Georgia,

population ecology.

Introduction

Natural populations are the single way of

species existence. Autoecological research

doesn't allow a complete understanding of

the life of a species in nature. That is why

there must be information of life cycles,

geographical range, population size,

number dynamics, etc. On the other hand it

is hard to explain ecological aspects of the

species existence without any information

of their habitat preferences, feeding habits,

breeding sites, etc.

By analyzing connections between

species population ecology and autecology,

as well as morphology and geographical

distribution, the most complete notion can

be formed. Investigations on some

amphibian species biology have allowed

scientists to elaborate complex works

connected with different aspects of their life

history. A wonderful example is Bell's

works on the Smooth Newt (Bell and

Lawton, 1975; Bell, 1977).

There aren't many data of regular

stationary investigations about the ecology

of the rare or narrow-ranged species. Long

term research of such species enlarges the

knowledge of the biology of wide

taxonomic groups. Moreover, these

investigations may be useful to find out

ways of rare species preservation.

A local population of the endemic

salamander, (Mertensiella caucasica), from

the western Caucasus of Georgia has been

investigated for a five year period (1985-

1990). This work gives additional

information about the life history of this

species.

The geographical distribution of the

Caucasian Salamander was mainly

established in the beginning of this century.

Information was summarized by Nikolsky

(1913). Later investigations commonly

took place in earlier reported localities or in

adjacent areas. Some new localities for

salamanders were found by Bakradze and

Tartarashvili (pers. comm.). The real

geographic range of M. caucasica was

established. The Caucasian Salamander is

Vol. 5 p. 148

Asiatic Herpetological Research

December 1993

FIG. 1. Distribution of Mertensiella caucasica.

distributed in external spurs of the

Trialetian Mountain Range. Probably it is

the result of historical changes in the Kura

River bed (Fig. 1). Populations are mainly

distributed in the forest belt, but in some

places they can be found close to subalpine

meadows. Humidity in the species'

locations reaches 1000 mm or more per

year (another narrow-ranged representative

of the salamander tribe, Chioglossa

lusitanica, has similar requirements of

humidity).

In the most dry part of the range of M.

caucasica, the eastern one, salamanders live

only in coniferous forest. When humidity

reaches 1200 mm/year in the middle area,

they can also be found in subalpine

meadows. Salamanders are distributed in

deciduous forest only close to the Black

Sea coast, where humidity is very high

(2000-2400 mm/year; Fig. 1). The high

dependence of the animal on humidity does

not itself limit the species distribution, but

determines sensitivity of specimens to other

environmental factors. It is very interesting

that the rheophilous species Ranodon

sibiricus, more restricted to water habitats

than M. caucasica, is geographical limited

by coniferous forests like M. caucasica in

eastern localities (Paraskiv, 1953). Local

populations, distributed along tributaries of

the Chorokh and Kura rivers (in upper

flow), are formed by salamanders within its

area. Width of streams in salamander plots

is not more that 1-1.5 m in spring and

because of stepped disposition of streams,

they run slowly in some places. There are

many slowly draining pools about 20-30

cm in depth with a lot of shelters. The

bottoms of streams and pools are covered

with stones, and there is a lot of non-

decayed organic matter. Stepped

disposition of streams is formed by stoned

conglomerations and fallen logs.

Apparently, mountain ranges between

stream canyons don't allow wide

salamander migration and local populations

are comparatively isolated. There is no

evidence that direct migrations of animals

occurs during their life cycle. Individuals

are found a maximum of 200-300 m

distance from streams.

December 1993

Asiatic Herpetological Research

Vol. 5 p. 149

yyj — stream bed and branches

-*S — fallen trees and piles of stones

O — large pools where larvae were found

\ — steep banks where shelters are located

FIG. 2. Schematic diagram of the study site for Mertensiella caucaska.

Tfie Study Area

The studied population inhabits

coniferous forest ecosystems along the

second range tributary of the Kura River in

Borjomi Canyon (eastern part of the

species' range) (Fig. 1). The plant

association is formed by Taxus baccate,

Picea orientalis and deciduous spots. The

size of the inhabited location is a bit more

than 200 m, and it is situated between 1000

and 1300 m altitude, about 2 km from the

stream mouth. Slopes are precipitous, built

by corrosion of underground tree roots or

relatively gradual, partially covered by

pteridium, Matteuccia struthiopteris, from

the adjoining stream banks. There are

some stone conglomerations and fallen

trees in the study area, shown on the map

(Fig. 2). Air temperature is close to stream

water temperature (13-15°C in summer) in

shelters formed by stones and logs.

Dynamics of air temperature in Borjomi

Canyon in May to July, 1989 is shown in

Fig. 3. Quiet pools and shelters are relative

rare, slopes are steeper and stream flow is

faster at upper and lower localities. Density

of salamanders here falls rapidly as well as

away from the stream banks in these

places.

Methods

The main quantitative data were obtained

during excursions with a lantern after

sunset along the study area. The location

of each adult animal was mapped, substrate

type and distance from stream bank (more

or less than 50 cm) was recorded. Adult

animals were marked individually by toe-

clipping. Combinations from clipped digits

in hind-limbs (not more than 2 in 1 foot)

responds to individual number of animals

from 1 to 99. Zero-1 clipped digit in the

front leg mean number of hundred. Marks

of salamanders recaptured in the next year

were renewed. Data of capture-recapture

were statistically counted as in Kaughley

(1977). Substrates of animals caught were

subdivided in 6 types: shallow water; sand

and pebbles above water shore; wet stones;

wet ground; moss or lichens; dry ground

and stones. These types were ranked

according to their humidity. Basic

investigations were conducted on 8-10 and

21-23 June, 1986, 24-28 June, and 5-7.

August, 1987, 3-5 and 21-24 July, 1988,

16 June- 12 July, 1989, 2-9 July, 1990.

We had 337 contacts with males and 202

with females (including specimens found

two or more time). Recording of larvae

was conducted during night excursions.

We caught females from nature in the

reproductive period and obtained eggs

using a hormonal stimulation method

(Gontcharov et al., 1989) to study some

ecological and morphological features of

early development. Eggs were incubated in

Weiss bowls in dechloronated water at a

temperature of 14°C as well as in aquaria at

Vol. 5 p. 150

Asiatic Herpetological Research

December 1993

Date

FIG. 3. Air temperature at the Mertensiella caucasica study site during the period of reproductive activity.

Stippled bars represent periods of rainfall.

40 -i

i- 30-

X>

S 20-

3

Z

10-

Males

Females

12 3 4

YV~ i-2

Substrate

5 6

FIG. 4. Location of salamanders by substrate type.

1- shallow water; 2- wet sand and pebbles; 3- wet

stones; 4- moist ground; 5- moss or lichens; 6- dry

stones. Solid bars represent males. Stippled bars

represent females.

a temperature varying from 5-22°C. Before

completion of metamorphosis, larvae were

kept in 20 liter aquaria, where water was

changed every third day. Food consisted

of crustaceans (Daphnia, Cyclops), Tubifex

and Chironomid larvae.

Morphological studies were conducted

on larvae, juvenile and adult animals with a

binocular magnifies and calipers. Their

snout-vent length (L), head length (Lc),

and tail length (Led) were measured with a

precision of 0.1 mm. The coloration

patterns of some animals was also

recorded.

Results

The niche of larvae and adult specimens.

Salamanders don't have an even

distribution within the study site.

Preference to every substrate depends on

the amount of moisture of each particular

substrate. Frequency of captures decreases

with distance from water or potential

shelters. The number of animals captured

out of shelters depends on time and season.

Most adult specimens were recorded close

to the stream (less than 50 cm from the

water shore): 60±4% of males, 62±5% of

females. In contrast to data on the

ecologically similar species, Chioglossa

lusitanica (Arntzen, 1981), there is no

difference between male and female

attachment to water in M. caucasica.

Animals commonly may be found on wet

sand or stones at the water shore, and they

avoid dry soil and stones. The distribution

of substrate type of captured animals for the

December 1993

Asiatic Herpetological Research

Vol. 5 p. 151

3

z

20

10-

Females

^KS^Ll

••*-*..

tC»

10

20

30 1

10

Date

FIG. 5. The seasonal dynamics of salamander occurrence.

period from 1988-1989 is presented in Fig.

4. It is obvious that females avoid open

rocky plots, fallen leaves or moss cover.

According to our observations, adult

salamanders spend only a small part of their

life on the surface of the ground. Even

during the active period, only a small part

of the population leaves their shelters at

night. Apparently, salamanders spent the

rest of the time in shelters, where they live

and feed. Seasonal dynamics and diurnal

activity are reported on a number of

captures during excursions in a 4 year

period (Fig. 5). The number of animals

found above ground decreased in July.

The decrease in capture was especially

sharp in the 5-15 July period. Even after

rains, individuals could hardly be found in

the second half of July (Fig. 5). Earlier

seasonal activity of males is noted for other

flowing-water (Arntzen, 1981) and

standing-water (Golubev, 1981; Beneski et

al., 1986) tailed amphibians. According to

phenological data the large number of

salamanders found in June is connected

with their reproductive activity.

The beginning of the reproductive period

is determined by air and water temperature.

The active season begins when minimal

night temperature is about 15°C (Fig. 3).

After mating and egg deposition, activity

decreases. Apparently, the decreasing

number of specimens found in July does

not depend on regular migrations.

Observations of adjacent parts of the stream

100

E

3

z

50

Females

23

24 1 22 23 24

Hour of occurence

FIG. 6. Number of salamanders observed by hour

of day.

did not have any result in late July and

August. The decrease in animal numbers

either depends on the dispersion of

individuals in the forest (or along stream

banks, as proposed by Arntzen, 1981, for

C. lusitanica), or more probably, they are

in shelters most of the time because of

greater abundance of their food, such as

Gammaridae and Lumbricidae, there.

The main nocturnal period of activity of

M. caucasica is between 2200 and 0100.

No active salamander has been found

before 2130, and active animals have been

rather rare before 2230. The number of

active animals decreased after 0130-0200.

The peak of activity was observed at about

2300 (Fig. 6).

Other nocturnal tailed amphibians also

have a short active period and the time of

activity is species specific (Semlitch and

Peachmann, 1985). In the summer,

Vol. 5 p. 152

Asiatic Herpetological Research

December 1993

D

T3

FIG. 7. Consecutive stages of courtship and

amplexux in Mertensiella caucasica.

Clutch Number

FIG. 8. Inter and intra-clutch variability in

fertilized eggs of Mertensiella caucasica.

salamander larvae can be found in some

parts of the stream bed in small pools with

slowly drying water. Outside of the

shelters, they are mainly in shallow water

with a depth of less than 5 cm. Generally,

the number of larvae doesn't exceed 10

individuals (maximum 14) in a pool.

Comparatively large premetamorphosed

larvae can be found in the stream. They

move actively along the stream. Like

adults, larvae spend most of the time in

shelters. The first individuals can be seen

in open water in late May. Larvae leave

their shelters at twilight, when absolute

sunlight is about 10 Lk. There are no

nitrates in the water composition in

salamander breeding and developing sites.

The pH reaches 7.8-8.3 and the hardness

of the water is 0.6-2.8 mg/equivalent 1.

Juveniles rarely leave their shelters during

the period from the end of metamorphosis

until first breeding.

The diet of larvae, juveniles, and adults

reflects their biotopical preferences and

does not show specialization in any

invertebrate group (Kuzmin, 1992). Both

terrestrial and aquatic organisms are found

in the diet of adult salamanders

(Ekvtimishvili, 1948; Kuzmin, 1992).

Only terrestrial organisms are found in the

diet of juveniles (Kuzmin, 1992).

Space used by salamanders.

The requirements of salamanders to

environmental characters limit the space

they use even within a local habitat. Thus,

salamander distribution isn't homogeneous

December 1993

Asiatic Herpetological Research

Vol. 5 p. 153

in a given locality. Obviously, animals

prefer places with plenty of shelters. The

mean number of salamanders captured

during night excursions per 10 m along the

stream bank in 1986, 1987, and 1989 was

9.2 ±1.86 males and 5.26+1.13 females.

According to the "mean crowding" index of

Lloyd (1967) (m=m+2^ -1) the mean value

is respectively 21.8™ and 12.78. The

highest density was observed in places

where there were logs and wooden blocks,

combined with stone conglomerations, and

a lot of small pools and shelters under tree

roots. The total number of adults recorded

for the 1986-1990 period, including

recaptures was 455. The place where each

specimen was found in 1989 is noted in

Fig. 2. This data could give a notion of

real space distribution of animals. Note

that larvae can be found outside of the local

population habitat significantly more often

than adults. This is connected with the fact

that some of the larvae leave the shelters

and continue development in the lower

parts of the stream (see below).

Apparently, this process does not disturb

normal metamorphosis and juvenile animals

return to the population locality.

Life Cycle.

The salamander breeding period in

investigated habitats occurs from the

second half of June until early July. Most

of the females found in July are ready for

egg deposition. The large oocytes can be

observed through the transparent ventral

skin. There are well distinguished mating

corns at the adult male shoulders.

Amplexing animals were found on the

ground close to shelters. Cyren (191 1) and

Obst and Rotter (1962) described normal

sexual behavior of salamanders in water in

natural and laboratory conditions. We

observed normal sexual behavior twice: on

28 June, 1988 and 4 July, 1990. In the

first case it took place about 2 m from the

stream bank in a conglomeration of tree

roots. In the second case it happened close

to water, at the entrance of a rock chink.

We don't exclude the possibility of normal

copulation in water. For example, mating

of C. lusitanica may take place both in

streams and on the shore. The consequent

states of courtship and amplexus are shown

in Fig. 7, b-f. According to our

observations, sexual behavior of M.

caucasica is similar to that of Salamandra

salarnandra (Joly, 1966). The corn on the

dorsal side of male tails has no special role

in courtship and amplexus. We have also

observed an attempt of copulation (Fig. 7h)

on 28 June, 1989 and 2 males in an

amplexus pose on 8 June, 1986.

Apparently, we have observed, in the latter

case, rival combats, described for S.

salamandra by Kiistle (1986), but in this

work the behavioral display is not the

same.

Copulated females have a slightly

opened cloaca. There are more than three

days between copulation and egg

deposition. Each female deposits from 1 1

to 24 eggs (N=9, M=16.9, a=3.9). Inter-

and intra-clutch variability of fertilized eggs

sizes is shown in Fig. 8. Darevsky and

Polozhikhina (1966) found that the sizes of

90 eggs found in nature ranged from 5.0-

5.6 mm. Females deposit separate eggs,

sticking them to the substrate in shaded

places. Activity of animals gradually

decreases after the completion of the

reproductive period.

Egg development takes 45 days until

hatching in aquaria, where the average

temperature is 14.8°C. When the

temperature changes from 6° to 26°

(M=16.5, o=4.7) development is extended

to about 48-5 1 days. We can expect similar

developmental rates in nature, when the

temperature of the water is about 14-15° in

July to August. The hatching of most part

of the generation takes place not earlier than

late August. Larvae found in June can be

divided into 3 groups on the basis of snout-

vent length: the I group- L=14.6-19.5 mm

(M=106-180 mg); the II group- L=23.7-

27.5 mm (M=330-664 mg); the III group-

L=29.0-35.4 mm (M=632-1400 mg). By

virtue of larval size distribution, Freytag

(1954) as well as Koroljov (1986)

concluded a 3 year period of larval

development in the Caucasian Salamander.

Kuzmin (1992) established no annual ring

in the I and II larvae group hip bones and

only one annual ring in the III group larvae

Vol. 5 p. 154

Asiatic Herpetological Research

December 1993

YEAR 4

YEAR

active

HHH hibernation

loo! egg deposition

YEAR 2

| embryonal development

j^s hatching

ci=i^ metamorphosis

FIG. 9. The reproductive cycle of Mertensiella caucasica. A- underground parts of the stream bed; B- open

water; C- surface and temporary shelters on die ground; D- constant shelters on the ground.

hip bones. On the basis of this information

Kuzmin (1992) supposed that larvae of I

and II size groups had most probably

hatched in the given year. Nevertheless,

analysis of time of reproduction and

embryonic development during the year

opposes Kuzmin's opinion.

ApparenUy, salamander larvae remain in

shelters after hatching and go out when

water temperature increases al least to 13°C

(at the beginning of the next summer).

Lack of annual ring on hip bones can be

explained by incomplete development of

hind legs just after the hatch. Larvae

growth is delayed by autumn temperature

decreasing. Thus, animals of the I and II

size groups have a previous year hatch and

represent a single generation.

Size differences within a generation are

formed by prolonged breeding time (not

only between breeding locations [Mertens,

1968], but within populations, too) and/or

by variation of individual growth rates.

Larvae have developed during the warm

period of the second year. After

hibernation, they have a metamorphosis in

July-August of the 3rd year. Their

development from fertilized egg to

completed metamorphosis takes about two

years in nature.

Salamanders have a concealed life during

the period after metamorphosis and before

maturity. According to Kuzmin (1992), 3-

December 1993

Asiatic Herpetological Research

Vol. 5 p. 155

5 annual rings can be observed in hip bones

of adults. Therefore, salamanders can first

breed in the 3rd year after metamorphosis.

The total life cycle of the Caucasian

Salamander from egg to egg is about 4

years (Fig. 9).

Growth and Development

Embryogenesis of M. caucasica is

similar to other large-size egg amphibian

development (Fig. 10 a-c). Analyzing

experimental observations, hatching takes

place when total length is 17-20 mm

(L=10.5-11.5 mm). When hatching starts,

a larvae has well developed external gills

and a tail fin. Sometimes the rudiments of

3 toes can be distinguished on the hind

legs. Pigmentation is formed by a couple

of faded pigmented stripes. There are rare

individual melanophores on the surface of

the stripes. A line of small circular non-

pigmented patches lays along each stripe

(Fig. 10 d). The gut is filled with yolk.

The total length of the smallest larva caught

in nature was at least 25 mm (commonly,

L=15 mm, minimum = 14.6 mm). Larvae

found in nature already have no yolk in the

gut. Their external gills are smaller than

those of animals that have not yet hatched.

There are 4-5 toes on the hind legs and

more pigment cells (Fig. 10 e).

In laboratory conditions, when

temperature is 14.8°C, yolk disappears

from the frontal part of the gut 16 days after

hatching at a length of 13.8114 mm. The

first larvae with a snout-vent length of 15-

16 mm can be found in streams in early

June, but most of them appear in July. The

small larvae, which have over wintered,

appear in stream pools in small, probably

sibling groups. The largest group (8

larvae) was found on July 5, 1988.

Individual sizes in that group provides

some information about intra-clutch larval

size variation: when L=17.35±0.48,

min=16.0, max = 19.2, coefficient of

variation approaches 7.8, L

total=29.25±0.60, min=27, max=31.3,

CV=5.8%. Summer growth of first

hibernated animals (in 1985) is shown in

the histograms of June and August larval

size distribution (Fig. 11). Mean total

FIG. 10. Embryo and larval development of

Mertensiella caucasica.

length increases 6.27 mm for 70 days and

the specific growth approaches 0.24%/day.

At the same time homogeneity of generation

increases: CV=23% in June and becomes

9.9% in August. This process is probably

caused by more rapid growth and/or

comparatively high mortality of the small

larvae.

Morphological changes, connected with

metamorphosis (i.e. yellow coloration of

unpigmented spots, decreasing of gill size,

reduction of tail fin- Fig. 10 f) began in

animals with at least 30 mm snout-vent

length. They approach that size in the 3rd

year of larval development. Comparing

sizes of the 2nd and 3rd year larvae, the

specific body growth rate is 0.12%/day in

the period between August and June of the

next year. Commonly, metamorphosis

takes place at a snout- vent length of 30-35

Vol. 5 p. 156

Asiatic Herpetological Research

December 1993

10

1 5-

June

Males

ru

o-1

10

kLi

xi

—I T"

30 40 50

60 70

£

s

Z

August

r^M ■■

~i 1 1 1 1 t~

30 40 50

1 1 1

60 70

L (mm)

FIG. 11. Size distribution of Mertensiella

caucasica larvae at the study site in 1985.

mm. Maximal larval size approaches 35.4

mm (L total=70.3 mm) in the population

studied. Snout-vent length of

metamorphosed animals varies within the

limits of 32.9-42.4 mm (body mass,

M=694-1592 mg, N=ll). At the same

time 2 larvae, with L=43.8 and 44.6 mm,

are in a collection from the surroundings of

Batumi (Kuzmin, pers. comm.). Animals

of all three size groups can be found in

streams even in May (Korolyov, 1986).

Peculiarities of natural growth of M.

caucasica are similar to C. lusitanica.

Although this species passes

metamorphosis at smaller sizes (i.e. L=24-

25 mm) their linear growth for two summer

months approaches 0.29%/day and

0. 10%/day for the rest of the year (Arntzen,

1981) and it is very similar to the analogous

index of M. caucasica. The slow growth of

salamander larvae is mainly the result of

low water temperature in streams. The

specific total length growth rate of a single

animal (from 27.9 to 56.7 mm) was

0.54%/day, and total length increased from

37.0 to 59.2 mm was 0.42%/day under

laboratory conditions, at 23-25°.

The snout-vent length of adults varies

insignificantly. Data on animals measured

X>

£2

N=10

x = 71.1

6 = 8.4

SE = 2.65

Females

50

I H

60

L (mm)

1 1 m i

70

80

FIG. 12. Size distribution of adult Mertensiella

caucasica in the study area. L- snout-vent length.

in the studied population are shown in Fig.

12. There aren't considerable intersexual

differences in sizes and general body

proportions, but apparently females can

begin breeding at a smaller body size.

Other authors (Cyren, 1911; Knoblauch,

1905; Nesterov, 1911) reported that mean

L in males approached 68.9 mm (N=7) and

females, 63.5 mm (N=ll). The animal

body length of an outlying population

(Goderdzi Mountain Pass) varies between

68-77 mm in males and 56-73 mm in

females. However, our population does

not show any specifics in adult animal size

distribution.

We will briefly discuss morphological

changes in the period from die beginning of

active feeding to the end of metamorphosis.

When larvae begin to feed, melanophores

gradually disperse from the lateral sides,

filling the ventral surface of the larval body.

Even the size II larval group have only a

narrow non-pigmented stripe remaining on

the ventral side. All lower surface is filled

by pigmented cells and non-pigmented

patches remain only on the lateral sides of

the size III larval group. These patches are

used as a substrate of xanthophores and

iridiophors, forming yellow spots later on

(Tarkhnishvili and Tartarashvili, 1987).

The intensity of basic coloration is

correlated with the size of the animal that

has already started metamorphosis. The

animals with a large size at the beginning of

metamorphosis have a dark-brown (not as

dark as in spotted salamander) coloration

with bright and comparatively large yellow

spots. Smaller size larvae do not have such

December 1993

Asiatic Herpetological Research

Vol. 5 p. 157

an intense basic coloration and spot pattern

is more or less reduced (spots are smaller

and/or poorly expressed). The intensity of

salamander pigmentation, like other tailed

amphibians, may vary depending on the

light intensity at the larval location

(Fernandez and Collins, 1988).

Ground coloration of adults varies from

reddish-brown (similar to Chioglossa

lusitanica or some M. luschani subspecies

(Winter et al., 1987) to dark brown. The

spotted pattern may be expressed in a

different degree to full reduction (especially

in light colored specimens) (Fig. 13).

Poorly pigmented animals with

comparatively reduced spots (described by

Tartarashvili and Bakradze (1989) as the

subspecies M. c. djanashvilii- Fig. 13 a)

predominate in some populations from the

surroundings of Batumi, at the Black Sea

coast. Nevertheless, dark colored animals,

with well developed spots (Fig. 13 e, f)

predominate in the population from the

subalpic zone (Mountain Pass Goderdzi, in

Bakradze's collection). Salamanders with

an intermediate intensity of coloration are

more abundant in our studied population,

but there are some specimens with less or

more reduced spot pattern. There is also a

female, colored as the form described by

Tartarashvili and Bakradze. Probable, the

specific coloration of adults is connected

with the character of larval development,

which depends on the special climatic

conditions of each habitat. That is why

many light-colored animals occur in the

warm sea cost habitat and dark colored

ones are found at high altitudes.

Populations from Borjomi Canyon are in an

intermediate place. Or course, we don't

exclude the possibility of inheritable fixing

of one or another coloration type in

different populations.

The Caucasian salamander is included

with the Luschan Salamander, M. luschani,

in the same genus because of the tail corn,

the secondary sexual character of males

(Ozeti, 1967). This character appears in

males with a length of at least 130 mm and

it seems to be of no functional importance

as some investigators have proposed, for

example Cyren, (191 1).

FIG. 13. Color variation in Menensiella caucasica.

The main changes of general body

proportion occur during ontogenetic

development. First, relative tail length

increases after hatch. Mertens (1968)

reported this for larvae with a total length of

more than 45 mm. Based on our data,

comparatively rapid tail growth begins at

the earliest stages of development and

extends to the adult stage. On the other

hand, comparative length of head decreases

(Fig. 14). The changes of general

proportions have different intensity in

different developmental stages.

Allometrical dependence of head and body

length on the total length of the I and II size

group larvae is described by equations:

Lcd=0.35L126

Lv=1.71L°-26

The coefficients of allometric equations

for the III size group larvae is different:

Lcd=0.08L17'

Lc=1.5L°-52

Vol. 5 p. 158

Asiatic Herpetological Research

December 1993

2.0

1.5

Led 10

0.5

0J

■ Spotted Salamander

D Caucasian Salamander

10

_L

Lc

Stage of Ontogenesis

Fig. 14. Changes in body length proportions in Mertensiella caucasica (open square) and Salamandra

salamandra (solid square) during ontogenesis. 1- larvae with total length less than 35 mm; 2- larvae with

total length greater than 35 mm; 3- yearlings; 4- juveniles; 5- adults, solid line- L/Lccj; broken line- 1/LC.

The coefficients of static allometrical

equations for recently metamorphosed

animals are:

Lcd=0.23L142

Lc=1.04L062

Hence, the most rapid comparative

increasing of tail length is during late larval

development. The comparative decreasing

of head is the most rapid in the I and II size

groups, but later this process isn't so clear.

Nevertheless, it takes part in

metamorphosis.

Plenty of eco-morphological features

separate M. caucasica from the other

representatives of European salamander

tribe, depended on subtilization (Ozeti,

1967). The latter is a base determined

ecological similarity between M. caucasica

and C. lusitanica (Borja-Sanchiz and

Mlinarsky, 1979). Perhaps this is a reason

of similar breeding ways of these species

differing from other European salamanders.

The changes of general proportion of M.

caucasica and S. salamandra, which are

shown in Fig. 14, allow a comparison of

these species. The general trends of body

proportion changes are common in the two

species (as in most Tetrapoda): the

comparative length of head (L/Lc)

decreases and that of tail (L/Lcd) increases.

But in these tendencies, both are quite rapid

in M. caucasica. In S. salamandra the

changes are gradual and moreover, tail

length growth is poorly distinguished (Fig.

14).

Population Number Dynamics and

Regulation.

The analysis of our 1986-1990 capture-

recapture data allow us to study population

number, number dynamics and

demographic peculiarities. We obtained

quite full information in 1989. In that year

the captured animal number, in relation to

real population number, was comparatively

high. The general picture of results is

shown in Table 1. We had 532 contacts

with animals. We met an animal twice in

the same night only on 7 occasions. They

were on the surface, essentially not

moving. In all cases animals were captured

again in the same plot, not more than an

hour later. The 50 animals of 68 recaptured

December 1993

Asiatic Herpetological Research

Vol. 5 p. 159

TABLE 1. Mark-recapiure results for Mertensiella caucasica in 1989.

Note: i, j- the number of census; nj- the number of specimens examined in i-tli census; Rj- the number of

recaptured and escaped specimens; X4_ [jj- the number of animals recaptured in the 4-th census that had been

marked during the 3-rd census: rj- the total number of recaptured specimens that had been marked in the i-th

census; Nj- the estimate of population size in the moment i, using the lolly-Seber method; SEj- standard

error (lolly-Seber); Pj^ j- the probability of the individual remaining in the active part of the population

between the i-th and j-th census; Aj_ j- the number of specimens supplemented the active part of the

population between the i-th and j-th census.

in a year of marking were caught in the

same or adjacent plot (the place of 28

captured animals was not recorded). On

the basis of these data, we conclude that

these salamanders have a low moving

ability and a short period of nocturnal

activity. Only an approximate

representation of activity dynamics can be

given by captured animal number, though

in some cases, data of captured animals are

Vol. 5 p. 160

Asiatic Herpetological Research

December 1993

used as an index of number (Bozhanski and

Semenov, 1982). We used the Jolly-Seber

method (see Caughley, 1977) to estimate

the real number of the local population and

its dynamics during the breeding period in

1989. The total number of breeding

animals was 1187 and the percentage of

males was 58±0.001. The greatest number

of active animals was in the end of June.

Females appeared a bit later than males.

Unfortunately, the data of 1986-1988

don't allow us to correctly estimate the

number of salamanders in those years. It

varied from 10-20 to 460 individuals, when

the errors exceeded mean values, i.e.

significantly lower than real quantity. The

highest value (460) was recorded when the

Schumacher method was used on 1990

data. The values recorded for the 1986-

1988 data didn't exceed 200. Thus, the

annual number value significantly increases

when the research period is prolonged and

captured specimen number increases. That

is because only a small part of the

population left their shelters, even in the

highest activity period in late June to early

July.

The estimate of the C. lusitanica

population (Arntzen, 1981) is different

because of the permanent migration of part

of the population. The number of two local

populations of this species is 1236 and

1324 respectively. This is similar to our

information about M. caucasica, moreover,

the study sites have a size similar to ours.

Usually, the number of animals in the

widely distributed in Europe S. salamandra

populations can exceed some thousand

specimens (Klewen, 1986). Their

populations are spread over several

hectares, and animals can be found far from

the breeding sites.

The capture of salamanders marked in

previous years gives some information

about mortality rates of adults. Ninety

eight males and 45 females were marked in

June, 1989. There were 38.5%

(OM=7.8%) recaptured males of the total of

39 found in July 1989, and 28.9%

(OM=7.3%) recaptured females of the 38

total found. The 31.8% (OM=5.7%) of

males and 17% (OM=5.9%) of females

captured in July 1990 were marked in June

1989. Hence, survival rates of the period

from July, 1989 to July, 1990 is

Pm=31.8/38. 5=0.83 for males and

Pm= 17/28.9=0.59 for females. The part of

all marked adults was 33.8±9.3% in July,

1989 and 26.2±4.2% in July, 1990.

Hence, the annual survival approached

0.77. It should be noted that only 8

individuals (9.0±2.0%) of all 89 marked in

1986-1988 were found again in 1989. This

is quite a high number because 89

individuals aren't more that 10% of the

adult population.

Unfortunately, there is a lack of

information about renewal rates of tailed

amphibian populations. Ignoring age

structure, the annual mean survival of

Ambystoma maculatum approaches 0.72

for males and 0.60 for females (the mean

male number is 641 and capturing of males

is a bit more often (Husting, 1965). These

data are quite similar to ours. On the other

hand, annual survival of the Smooth Newt

is only 0.45 for males and 0.55 for females

in England (Bell, 1977). There are

considerable low survival rates of

Notophtalmus viridescens and some Anura

when higher mortality of males is observed

(Ischenko, 1989). Klewen's (1986)

quantitative data for S. salamandra show

that annual survival varies between 0.55-

0.81 (mean 0.66 in four years of

investigation. A higher female survival

was recorded for the genus Desmognathus

(Husting, 1965).

The Caucasian Salamander has a

comparatively low population renewal rate,

when mortality is low. Perhaps, this kind

of population dynamics is typical for

populations with low total number and high

male survival. Organ (see Husting, 1965)

mentioned that a higher male survival was a

result of significant energy expenses of

females during breeding.

We can only indirectly estimate

salamander mortality before mating. On the

basis of adult female number and mean

fecundity, we estimate that there were

8000-9000 eggs deposited in the study site.

December 1993

Asiatic Herpetological Research

Vol. 5 p. 161

TABLE 2. Data on mark-recapture of Meriensiella caucasica from 1986 to 1988.

Note: nj- number of individuals examined in the i-th census; Rj- number of recaptured and escaped

individuals; m;- total number of individuals captured in the i-th census that were marked in the same year

(but on another day).

Nevertheless, significantly few larvae could

be found in the stream when salamander

density was the highest.

Korolyov (1986) found only 90 larvae in

1984 in the whole area of the stream

described here. Bozhansky and Semenov

(1982) counted 1-33 larvae in every 700 m

of flow in August. We found 1 16 larvae in

the study site in June, 1985, of which 91

had already gone through hibernation. In

the beginning of July, 1990, 74 larvae of

the I and II size groups (hibernated once)

were counted in the same plot. The number

of second year larvae is quite constant in

different years. Although it was

considerably lower that the total number of

eggs deposited, it did not have an influence

on the real larval mortality rates in the first

year. More probably, larvae were carried

by flow along the stream and were

distributed more uniformly than adults and

eggs.

On the base of the 2nd and 3rd year

larvae proportion at the site, we were able

to judge larval mortality from egg

deposition to the second year. The twice

hibernated were 27.8% (1985) to 31.9%

(1990) of the total number of once

hibernated specimens. Apparently these

values express the real survival rates of a

year (within twelve months).

The stable larval density and ratio of the

second and third year animals at the locality

is a result of the stability of the adult

salamander population number, and

moreover, conversely to stagnant water

amphibians, quite constant developmental

conditions. The later is a reason of

Caucasian Salamander number dynamics

peculiarities. The basic reasons of natural

salamander mortality are not completely

clear. Perhaps, egg and larval mortality

caused by ponds drying up is not as

important for M. caucasica as it is for many

amphibians (for example, Ambystoma

maculatum, Albersetal., 1987). There are

practically no predatory insect larvae

dangerous to salamander larvae in the

stream. Young trout (Salmotrutta labrax),

which are possible predators, are also very

rare. Perhaps the main reasons of larval

mortality are over wintering and larval

diversion into the stream flow. Grass

Snakes (Natrix natrix) could of course

cause great damage during metamorphosis.

There were from one to five just

metamorphosed specimens in the stomachs

of the five Grass Snakes captured in the

salamander locality. We did not find any

adult salamanders in snake stomachs. It is

unlikely the low vulnerability of adults is

connected with autotomy ability or

coloration. The juvenile animals do not

have the same coloration as adults and they

are able to autotomise also (Golubev,

1981). The reason is rather adult animal

size and antipredator behavior of this

species described by Brodie et al. (1984).

Discussion

According to the view of adaptionists,

the peculiarities of the morphology of the

Caucasian Salamander mainly are a result

of general body constitution. The

vulnerability of this species results in the

high requirements to environmental

Vol. 5 p. 162

Asiatic Herpetological Research

December 1993

condition, especially temperature and

humidity. The comparative large body

surface reduces homeostatic ability.

Another main feature of M. caucasica as

well as of C. lusitanica delimiting these

species from all other European

salamanders is breeding by egg deposition,

when fecundity is comparatively low.

The biogeographical and ecological

characteristics of M. caucasica can be

explained by consequences of its

morphological type. On the basis of

paleontological date, Mertensiella aff.

caucasica was distributed sympatrically

with the Spotted Salamander in an area

extending to central Europe in the Pliocene.

The reduction of the range of Mertensiella

was the result of the last glacial periods

(Borja-Sanchiz and Mlinarsky, 1979).

Nevertheless, the present range of S.

salamandra is quite wide (Thorn, 1968),

while the range of M. caucasica, like other

representatives of the tribe, M. luschani and

C. lusitanica, is comparatively narrow and

in areas with mild climate. It is unlikely

that the Spotted Salamander can affect the

geographic range of these species.

Although members of the genus

Mertensiella are allopatric to S. salamandra,

C. lusitanica has a wide sympatric zone

with this species (Bas Lopez, 1984). The

absence of S. salamandra in the Caucasus,

including the Great Caucasus, obviously

depends on historical reasons. The

geographic range of subtle species is

limited most of all by climatic factors.

Wolterstorff et al. (1936) mentioned that

the range of M. caucasica has not changed

considerably since the Eocene. Perhaps the

low homeostatic ability of adults limits their

migratory possibilities.

As we already noted, the captures of

animals far from their population locality

were very rare. The salamanders don't

penetrate the comparatively distant

mountain systems like the Great Caucasus.

They also don't occur in comparatively dry

localities along the Trialeti Mountain Range

in the East (Fig. 1) where there is no relief

limit. This is one of the reasons for the

restricted salamander distribution. On the

basis of different research (Obst and Rotter,

1962; Tartarashvili, pers. comm.) we

conclude that the area of salamander

localities of high altitude and on the Black

Sea coast isn't larger than ours.

Chioglossa lusitanica localities have a

similar distribution (Arntzen, 1981). On

the other hand, a small population area

might depend on attachment to the breeding

sites (stream plots suitable for egg

deposition and larval development).

Spotted salamander populations are always

distributed in significantly wider areas

(Klewen, 1985).

As a result of the small area of the

localities and the lack of breeding sites, the

population number is limited at a

comparatively low level, about 1000

individuals, when the sex ratio is close to

equal. This amount is enough to maintain

the populations demographically and

genetically (Lande and Barrowclough,

1989). The potential population growth

rate is also limited by comparatively low

fecundity. Nevertheless, the breeding sites

are used rather efficiently. Temporary

ponds vulnerable to periodic natural

disturbances aren't used for egg deposition.

Thus, the population renewal possibilities

of M. caucasica are different from stagnant-

water amphibians. A significant part of the

latter species are not able to breed

efficiently because many of the breeding

ponds within localities are destroyed during

egg and larval development. Apparently

the egg deposition of M. caucasica is done

only in places suitable for further

development. This type of reproduction is

correlated with the high stability of

population number though the resilience to

habitat transformation is low.

This described model of population

dynamics practically excludes the number

of outbursts caused by climatic

perturbations which could stimulate

considerable migrations. Since, settling of

investigated species is determined by the

low tolerance of adults, the small area of

population localities and breeding sites, and

the low fecundity. Hence the main reason

for the narrow geographical range of M.

caucasica are its morphological features and

the stable type of population cycle.

December 1993

Asiatic Herpetological Research

Vol. 5 p. 163

In analyzing morphological reasons of

Caucasian Salamander ecological specifics

in comparison to related species, we can

not accept as a main character only body

proportions. Ecological particularities of all

the European salamander tribe mainly

depend on comparatively large egg

formation (and probably skin structure).

These characters limit European

salamanders to breed only in flowing water

in comparatively wet places.

Although the European salamander

adaptive type allows comparatively wide

interspecies variability in some features, for

example coloration patterns and degree of

subtilisation, the general morphological

constitution restricts adaptive ability of

particular representatives of the group. In a

sense M. caucasica is a morpho-ecological

equivalent of C. lusitanica. These species

have similar life cycle, population spatial

structure and number dynamics, climatic

and biotopic preferences, etc. The central

adaptive possibility and the feature

determined place in the group can be

distinguished among a lot of morphological

characteristics. There are some other

features which separate these species and

reveal the independent origin of both of

them. This is, for example, tail corn in

males. Nevertheless, this structure does

not take place among main ecological

features of species and reflects only the

complexity of the phylogenetical ways.

Color patterns have rarely been used in

phylogenetical speculations, but this

characteristic is a favorable object of

adaptationists. The presence of light-

colored specimens in some M. caucasica

populations, their predominance in other

populations of this species and in M.

luschani and, finally, fully reduced of

spotted specimens in C. lusitanica are not

connected with variability of the plant cover

in localities and do not affect their

ecological preferences. We can't speculate

about the adaptive meaning of coloration in

this case. Coloration is closely related to

the climate type of localities. Apparently,

we could consider this characteristic as a

fixed non-adaptive reaction to temperature

and humidity changes.

In conclusion, we would like to give an

opinion on an interesting detail connected

with the distribution of M. caucasica.

Three anuran species, the Colchic Toad

{Bufo verrucosissimus), the Caucasian

Parsley frog (Pelodytes caucasicus), and

the Asia Minor Frog (Rana macrocnemis)

live sympatrically with the Caucasian

Salamander. Rana macrocnemis is

distributed all over the Caucasus. Bufo

verrucosissimus and Pelodytes caucasicus

like M. caucasica do not penetrate the

eastern part of the Trialeti Mountains

because of lack of humidity. However,

they are distributed in some locations in the

Great Caucasus. The northern parts of the

Trialetian and Adjaro-Imeretian mountains

are more that 50 km from the southern parts

of the Great Caucasus in Central Georgia.

There are no suitable localities for forest

amphibians between these mountain

systems. The comparatively small

transitional zone could have been crossed

many times by B. verrucosissimus and P.

caucasicus after the Great Caucasus system

was formed. If we take into consideration

the comparatively high fecundity (about

500 eggs per year for P. caucasicus and

10,000 eggs per year for B .

verrucosissimus) and the temporal

variability of the breeding sites, a few

climatically favorable seasons could cause a

great increase in population number and as

a final result, a massive migration.

According to the peculiarities of M.

caucasica population dynamics, we can not

expect any similar process. Hence, the lack

of M. caucasica in the Great Caucasus has

historical rather than autecological reasons.

Acknowledgments

The authors wish to express their

gratitude to Dr. S. L. Kuzmin for his great

help in providing material and literature.

We also thank Mr. R. V. Tartarashvili for

interesting information and material. Dr. J.

Ilieva helped us greatly to prepare the

English text of this article.

Literature Cited

ALBERS, P. H. AND R. M. PROUTY 1987.

Survival of spotted salamanders eggs in

Vol. 5 p. 164

Asiatic Herpetological Research

December 1993

temporary woodland ponds of coastal Maryland.

Environmental Pollution 46(1):45-61.

ARNTZEN, J. W. 1981. Ecological observation on

Chioglossa lusitanica (Caucata, Salamandridae).

Amphia Reptilia 3/4:187-203.

BAS LOPEZ, S. 1984. Biogeographia de los

amfibies y reptiles de Galicia, un ensago de

sintesis. Amphibia Reptilia 5:289-310.

BELL, G. A. C. 1977. The life of the smooth

newt (Triturus vulgaris) after metamorphosis.

Ecological Monographs 47(3):279-299.

BELL, G. A. C. AND J. H. LAWTON. 1975. The

ecology of the eggs and larvae of the smooth

newt, Triturus vulgaris (Linn.). Journal of

Animal Ecology 44(2):393-424.

BENESKI, J. T., E. J. ZABISKO, AND J. H.

LARSEN. 1986. Demography and migratory

patterns of the eastern long-toed salamander,

Ambystoma macrodactylum columbianum.

Copeia 1986(2):398-408. '

BOZHANSKI, A. T. AND D. V. SEMONOV. 1982.

[The data on the biology of the Caucasian

salamander (Mertensiella caucasica)].

Zoologichesky Zhurnal 61(8): 1 1 18- 1 129. (In

Russian).

BORJA-SANCHIZ, F. DE AND M. MLYNARSKI.

1979. Pliocene salamandids (Amphibia,

Caudata) from Poland. Acta Zoologica Cracov

24(4): 175-188.

FERNANDEZ, P. J. AND J. P. COLLINS. 1988.

Effect of environment and ontogeny on color

pattern variation in Arizona tiger salamanders

(Ambystoma tigrinum nebulosum Hallowell).

Copeia 1988(4):928-938.

FREYTAG, G. E. 1954. Der Kaukasus-

Salamander, ein seltener Terrarien pflegling.

Die Aquarien und Terrarien Jahrbuch:l 15-1 19.

GOLUBEV, N. S. 1981. [The peculiarities of the

tail autotomy and regeneration of the Caucasian

salamander]. Voprosy gerpetologii. Nauka,

Leningrad 5:40. (In Russian).

GONTCHAROV, B. F., O. I. SHUBRAVY, I. A.

SERBINOVA, AND V. K. UTESHEV. 1989. The

USSR program for breeding amphibians,

including rare and endangered species.

International Zoo Yearbook 28:10-21.

HUSTING, E. L. 1965. Survival and breeding

structure in a population of Ambystoma

maculatum. Copeia 1965(3):352-362.

ISCHENKO, V. G. 1989. [The reproductive tactics

and demography of amphibian population].

Itogi nauki i tekhniki. Zoologiya

pozvonochnykh. V. 17. Problemy

populyatsionnoy ekologii zemnovodnykh i

presmykayushchikhsya. Moscow, VINITI:5-51.

(In Russian).

JOLY, J. 1966. Sur l'ethologie sexuelle de

Salamandra salamandra (L.). Z. Tierpsychol.

23(l):8-27.

BRODIE, E. D., R. A. NUSSBAUM, AND M. Dl

GIOVANI. 1984. Antipredator adaptations of

Asian salamanders (Salamandridae).

Herpetologica 40(l):56-68.

CYREN, O. 1911. Beitrage zur Kenntnis der

Kaukasichen Feuer-salamanders, Salamandra

caucasica (Waga), seiner Lebensweis und

Fortpflanzung. Bericht der Senckenbergischen

Naturforschended Gesellschaft, Frankfurt a/M

42:175-189.

DAREVSKY, I. S. AND V. F. POLOZHIKHINA.

1966. [On the reproductive biology of the

Caucasian salamander- Mertensiella caucasica

(Waga)]. Zoologichesky Zhurnal 45(3):465-

466. (In Russian).

EKVTIMISHVILI, Z. 1948. [Feeding of the

Caucasian salamander]. Trudy Zoologitcheskogo

Instituta AN GSSR 8:239-245. (In Georgian).

KASTLE, W. 1986. Rival combaLs in Salamandra

salamandra. Pp. 525-528 In Z. Rocek (ed.),

Studies in Herpetology. Prague.

KAUGHLEY, G. 1977. Analysis of vertebrate

populations. A. Wiley- Interscience

Publications., London, New York, Sydney,

Toronto. 361 pp.

KLEWEN, R. 1986. Population ecology of

Salamandra salamandra terrestris in an isolated

habitat. Pp. 395-398 In Z. Rocek (ed.),

Studies in Herpetology. Prague.

KOROLJOV, A. V. 1986. Some data on the larvae

of Mertensiella caucasica. Pp. 281-283 In Z.

Rocek (ed.), Studies in Herpetology. Prague.

KUZMIN, S. L. 1992. Feeding ecology of the

Caucasian Salamander (Mertensiella caucasica)

with comments on life history. Asiatic

December 1993

Asiatic Herpetological Research

Vol. 5 p. 165

Herpetological Research 4:123-131.

LANDE, R. AND G. F. BARROWCLOUGH. 1989.

The effective numbeity of population and

genetic variability and their application in the

population monitoring. Pp. 117-157 In M. E.

Sould (ed.) Viable populations for conservation.

Mir, Moscow. (Russian translation).

OZETI, N. 1967. The morphology of the

salamander Mertensiella luschani (Steindachner)

and the relationships of Mertensiella and

Salamandra. Copeia 1967(2):287-298.

PARASKIV, K. P. 1953. [Siberian salamander

(Ranodon sibiricus)]. Izvestiya AN KazSSR,

Biologicheskaya seriya 1:47-56. (In Russian).

KNOBLAUCH, A. 1905. Der Kaukasische

Feuersalamander, Salamandra caucasica (Waga).

Berichte der Senckenbergischen

Naturforschenden Gesellschaft in Frankfurt am

Main 36:89-110.

LLOYD, M. 1967. Mean crowding. Journal of

Animal Ecology 36(1): 1-30.

MERTENS, R. 1968. Bemerkungen zur

"Normalentwicklung" des Kjaukasus-

Salamanders. Salamandra 4:44^45.

NESTEROV, P. V. 1911. Salamandra caucasica

Waga. Izvestiya Kavkazskogo Muzuya 5:319-

327. (In Russian).

NIKOLSKY, A. M. 1913. Herpetologia Caucasica.

Kantselariya Namestnika E. I. V. na Kavkaze,

Tiflis. 272 pp. (In Russian).

OBST, F. J. AND J. ROTTER. 1962. Notizen zu

Mertensiella caucasica (Waga) 1876. Aquarien

und Terrarien Zeitschrift 15(2):50-52; (3):84-86.

SEMLITCH, R. D. AND H. K. PEACHMANN.

1985. Diel pattern of migratory activity for

several species of pond-breeding salamanders.

Copeia 1985(1):86-91.

TARKHNISH VILI, D. N. AND R. V.

TARTARASHVILI. 1987. [On the development

of spotted pigmentation in some

Salamandridae]. Zoologichesky Zhurnal

66(9):1328-1338. (In Russian).

TARTARASHVILI, R. V. AND M. L. BAKRADZE.

1989. [The new subspecies of the Caucasian

Salamander]. Soobshcheniya AN GSSR

133(1):177-179. (In Russian).

THORN, R. 1968. Les salamandres d'Europe,

d'Asie et d'Afrique du Nord. Editions Paul

Lechevarier, Paris. 376 pp.

WOLTERSTORFF, W., L. A. LANTZ, AND W.

HERRE. 1936. Ein Import des kaukasischen

Salamanders (Mertenjsiella caucasica).

Zoologischer Anzeiger 116(1/2):1-13.

December l'W Asiatic Herpetological Research Vol. 3, pp. 166-169

Guidelines for Manuscript Preparation and Submission

Summary

Manuscripts must:

1) be written in English.

2) be of letter quality (typewritten on bond paper).

3) include camera ready figures (if any).

4) include complete and accurate literature citations.

5) include complete and accurate localities with latitude and longitude.

6) include a camera ready map illustrating regions discussed (when applicable).

Manuscripts failing to meet these criteria will be returned for correction.

Purpose and Content

Asiatic Herpetological Research publishes articles concerning but not limited to Asian

herpetology. The editors encourage publications from all countries in an attempt to create

an open forum for the discussion of Asian herpetological research.

Articles should be in standard scientific format and style. The following sections should

be included:

Title. The title should reflect die general content of the article in as few words as possible.

A poor tide may cause readers to read no further.

Names and Addresses. The names and addresses of all authors must be complete enough

to allow postal correspondence.

Abstract. The abstract should briefly summarize the nature of the research, its results, and

the main conclusions. Abstracts should be less than 300 words.

Key Words. Key words provide an index for the filing of articles. Key words provide the

following information (when applicable): 1) Taxonomy (e.g. Reptilia, Squamata,

Gekkonidae, Gekko gecko ). 2) Geography (e.g. China, Thailand). 3) Subject (e.g.

taxonomic validity, ecology, biogeography). The order of taxonomy, geography, and

subject should be observed.

Text. Manuscripts must be in English and spelling must be correct and consistent. Use

Webster's New International Dictionary for reference. For clarity, use active voice

whenever possible. For example, the following sentences in active voice are preferable to

those in passive voice.

Active voice:

"Lizards were extremely common on the site." and "The three snakes examined were female."

Passive voice:

"Lizards were observed to be extremely common on the site." and "Three snakes were examined

and were found to be female."

Abbreviation. Do not abbreviate unless the full phrase has already appeared. Scientific

names may be abbreviated only if they have appeared fully in the same paragraph. Never

begin a sentence with an abbreviation of a scientific name.

1993 by Asiatic Herpetological Research

December 1993 Asiatic Herpetological Research Vol. 3. p. 167

Standard Format

Manuscripts following standard format should include introduction, methods, results,

and discussion sections. While other formats are acceptable, the editors encourage the use

of standard format.

Introduction. The introduction typically states the significance of the topic and reviews

prior research.

Methods. This section should clearly state where, when, and how research was carried

out. Include sample sizes. Protocols designed by other investigators must be properly

cited. Research materials and their manufacturers should be listed. The reader must be

able to replicate the methods of the author(s).

Results. This section states the results and their significance to the investigation. Figures

and tables may be used to clarify, but not to replace, results statements in the text.

Statistics should be used when applicable. Large amounts of data should be avoided, or

included as an appendix at the end of the article.

Discussion. The discussion is a synthesis of the introduction and the results. No new

information should be discussed unless it was presented in the results section. New

findings should be discussed in relation to prior research. The author(s) should feel free to

present several possible interpretations of the results. The editors particularly encourage

suggestions of future research in Asian herpetology.

Section Headings

Articles will be published using three section heading styles. Level 1: text is bold and

centered Level 2: text is italic and centered. Level 3: text is italic and at the beginning of

the first paragraph. Authors should take this style into account when writing manuscripts.

Statistics

Statistics must be accompanied by sample sizes, significance levels, and the names of

any tests. Investigators should pay careful attention to independence and applicability of

tests, and randomness of samples. One of the most frequent examples of nonindependence

is the use of multiple, paired t-tests instead of analysis of variance (ANOVA). In general,

multiple tests on the same data set are not valid. Descriptive statistics are in

many cases more appropriate than inferential statistics.

References

Accurate and standard references are a crucial part of any article. This is especially

important when dealing with publications from many different countries. The reader must

be able to precisely identify any literature cited. References in the text must be checked for

consistency with references in the literature cited section. All references cited in the

text must be in the literature cited section. The literature cited section may

not contain any references not mentioned in the text. Articles containing

inaccurate or inconsistent literature citations will be returned for correction.

References In Text

1) References to articles by one or two authors must include both surnames in the order

they appear in the original publication. References to articles by more than two

authors must include the first author's surname, followed by "et al."

2) The year of article follows the authors, separated only by a space.

3) References with the same author and year are distinguished by the lower case characters

"a, b, c, ..."

Vol. 3, p. 168 Asiatic Herpetological Research December 1993

4) References cited in text are listed in alphabetical order by first author.

For example, "My results also incorporate literature records (Marx et al. 1982; Marx and Rabb

1972; Mertens 1930; Pope 1929; Wall 1909, 1910a, 1910b, 1910c).

References In Literature Cited

1) References must include all authors, in the order that they appear in the original

publication; "et al." is never used in a literature cited section.

2) The first author is listed surname first, initial(s) last. All other authors are listed

initial(s) first, surname last.

3) References with the same author and year are distinguished by the lower case

characters, "a, b, c, ..."

4) References cited are listed in alphabetical order by first author.

5) Names of journals are not abbreviated.

See below for examples:

Journal article.

Dial, B. E. 1987. Energetics and performance during nest emergence and the hatchling frenzy in

loggerhead sea turtles (Caretta caretta ). Herpetologica 43(3):307-315.

Journal article from a journal that uses year instead of volume.

Gatten, R. E. Jr. 1974. Effect of nutritional state on the preferred body temperatures of turtles.

Copeia 1974(4):912-917.

Journal article, title translated, article not in English.

Ananjeva, N. B. 1986. [On the validity of Megalochilus mystaceus (Pallas, 1776)]. Proceedings

of the Zoological Institute, Leningrad 157:4-13. (In Russian).

Note that for Acta Herpetologica Sinica, the year must precede the volume number. This

is to distinguish between the old and new series, and between 1987, Vol. 6 numbers 1-4

and 1988, Vol. 6 numbers 1-2.

Cai, M., J. Zhang, and D. Lin. 1985. [Preliminary observation on the embryonic development of

Hynobius chinensis Guenther]. Acta Herpetologica Sinica 1985, 4(2):177-180. (In Chinese).

Book.

Pratt, A. E. 1892. To the snows of Tibet through China. Longmans, Green, and Co., London.

268 pp.

Article in book.

Huey, R. B. 1982. Temperature, physiology, and the ecology of reptiles. Pp. 25-91. In C. Gans

and F. H. Pough (eds.). Biology of the Reptilia, Vol. 12, Physiological Ecology. Academic

Press, New York.

Government publication.

United States Environmental Data Service. 1968. Climatic Atlas of the United States.

Environmental Data Service, Washington, D. C.

Abstract of oral presentation

Arnold, S. J. 1982. Are scale counts used in snake systematics heritable? SSAR/HL Annual

Meeting. Raleigh, North Carolina. [Abstr].

Thesis or dissertation.

Moody, S. 1980. Phylogeneuc and historical biogeographical relationships of the genera in the

Agamidae (Reptilia: Lacertilia). Ph.D. Thesis. University of Michigan. 373 pp.

Anonymous, undated.

Anonymous. Undated. Turpan brochure. Promotion Department of the National Tourism

Administration of the People's Republic of China, China Travel and Tourism Press, Turpan,

Xinjiang Uygur Autonomous Region, China.

December 1993 Asiatic Herpetological Research Vol. 3, p. 169

Figures and Tables

Figures and tables should be referenced in order in the text Each table should be

typewritten, double spaced on a separate sheet. See below for instructions for figures.

Plates

All figure plates submitted must be of publication quality, and should ideally be camera

ready. All text in figures must be of typeset quality. Times Roman typeface is

preferred.

If typeset quality lettering is not possible for the author(s), Asiatic Herpetological

Research will accept figure plates without lettering. The following instructions must be

followed precisely:

1) Carefully label figures in pencil, on the back, or attach a photocopy or an additional

sheet with instructions.

2) Do not submit figures with poor type or handwriting on the face of the figure.

Substandard figures will be returned for correction. In order to avoid wasted effort, please

follow the above instructions carefully.

Figure Legends

Figure legends should be typed on a separate sheet. Legends should explain the figure

without reference to the text. A figure and legend should make sense if separated from the

rest of the article. For example:

FIG. 2. Lateral view of live Psammodynastes pulverulentus holding a prey lizard (Anolis

carolinensis ). Note buccal tissue surrounding the enlarged anterior maxillary and dentary teeth of the

snake.

Copyright

Asiatic Herpetological Research reserves the copyrights to all material published therein,

except that excluded by permission of the editors. Any material under a prior copyright

submitted to Asiatic Herpetological Research must be accompanied by the written consent

of the copyright holder.

Submission of Manuscripts

Authors should submit letter quality, double spaced, single-sided manuscripts both in

English and in the original language on 21.5 x 28 cm (8.5 x 1 1 inch) white bond paper.

If possible, include a computer diskette containing the manuscript. Macintosh diskettes

with Mac write. Write Now, Microsoft Word, or text files, or MS/PC DOS diskettes with

Word Perfect, Wordstar, Microsoft Word, RTF, or ASCII files are preferable.

Computerized manuscripts should include italic, bold, and centered text only. Additional

formatting is not necessary or desirable.

Manuscripts will be reviewed. The editors will attempt to choose reviewers whose

research knowledge most closely matches the content of the manuscript.

Asiatic Herpetological Research requests $25 US per printed page from authors with

funds available. Please indicate if funds are available.

ISSN 1051-3825

CONTENTS

TUNIYEV, BORIS S., AND SAHAT M. SHAMMAKOV. Coluber atayevi Sp. Nov. (Ophidia,

Colubridae) from the Kopet-Dag Mountains of Turkmenistan 1

CHOU, WEN-HAO. On the Status of Rhacophorus prasinatus Mou, Risch, and Lue

(Anura: Rhacophoridae) 11

AUFFENBERG, WALTER, AND HAFIZUR REHMAN. Studies on Pakistan Reptiles. Pt. 3.

Calotes versicolor 14

ROCEK, ZBYNEK. Holocene anurans from Caucasus 31

Wei, Gang, Ning xu, dejun Li, Guanfu wu and xiquan Song. Karyotype, C-

Band and Ag-Nors Study of Three Stink Frogs 45

GOLUBEV, M. L. The Variegated Toad Agama in Djungar Gate (Eastern Kazakstan) with

Notes on Certain Systematic Problems of Phrynocephalus versicolor Str. (Reptilia:

Agamidae) 51

Mezhzherin, Sergei and Michael L. Golubev. Allozyme Variation and Genetic

Relationships within the Phrynocephalus guttatus Species Group (Sauria: Agamidae)

in the Former USSR .' 59

WANG YUEZHAO AND HurZHAO WANG. Geographic Variation and Diversity in Three

Species of Phrynocephalus in the Tengger Desert, Western China 65

TUNIYEV, BORIS. S. AND SVETLANA. YU. BEREGOVAYA. Sympatric Amphibians of the

Yew-box Grove, Caucasian State Biosphere Reserve, Sochi, Russia 74

SMITH, BRIAN E. Notes on a Collection of Squamate Reptiles from Eastern Mindanao,

Philippine Islands Part 1 : Lacertilia 85

SMITH, BRIAN E. Notes on a Collection of Squamate Reptiles from Eastern Mindanao,

Philippine Islands Part 2: Serpentes 96

ZHONG, CHANGFU. First Records for Ophisaurus harti and Python molurus bivittatus

from Jiangxi Province, China 103

MANILO, VALANTINA V. AND MICHAEL L. GOLUBEV. Karyotype Information on some

Toad Agamas of the Phrynocephalus guttatus Species Group (Sauria, Agamidae) of

the former USSR ' 105

MANILO, VALENTINA V. A Karyosystematic Study of the Plate Tailed Geckos of the

Genus Teratoscincus (Sauria, Gekkonidae) 109

WANG, PEI-CHAO AND JlANG-HUA ZHANG. Resting Metabolic Rate in Three Age-groups

of Alligator sinensis 112

ZHANG, YUN, YULIANG XlONG AND CASSIAN BON. Effects of Chinese Snake Venoms

on Blood Coaguladon, Purified Coagulation Factors and Synthetic Chromogenic

Substrates 117

SCHAMMAKOV, SAHAT, CHARI. ATAEV, AND ELDAR. A. RlJSTAMOV.

Herpetogeographical Map of Turkmenistan 1 27

LIU, WAN-ZHAO, AND DA-TONG YANG. A Karyosystematic Study of the Genus

Bombina from China (Amphibia: Discoglossidae) 137

SONG, JIANXING, YULIANG XlONG, WANYU WANG, AND XlAOCHUN PIT. A Study on

the Purification and Pharmacological Properties of Two Neurotoxins from the

Venom of the King Cobra (Ophiophagus hannah) 143

TARKHNISHVILI, DAVID N., AND IRINA. A. SERBINOVA. The Ecology of the Caucasian

Salamander {Mertensiella caucasica Waga) in a Local Population 1 47

GUIDELINES FOR M ANUSCRIPT PREPARATION AND Si IBMISSION 1 66

Harvard MCZ Library

3 2044 066 300 419