ISSCA

International Society for the Study

and Conservation of Amphibians

(International Society of Batrachology)

SEAT

Reptiles et Amphibiens, Département de Systématique et Evolution, Muséum national d'Histoire

naturelle, CP 30, 25 rue Cuvier, 75005 Paris, France. — Tel: (33).(0)1.40.79.34.87. — Fax:

(33).(0)1.40.79.34.88. - E-mail: ohler@mnhn.fr.

BOARD

President: Jon LOMAN (Lund, Sweden).

General Secretary: Annemarie OHLER (Paris, France).

Treasurer: Stéphane GROSIEAN (Paris, France).

Councillors: Lauren E. BROWN (Normal, USA); Rafael O. DE SA (Richmond, USA); Rémi DUGUET

(Saint-André, France); JIANG Jianping (Chengdu, China): Esteban O. LAvILLA (San Miguel de

Tucumän, Argentina); Edgar Len (Lawrence, USA): Thierry LODÉ (Angers, France): Stefan

LôTTERs (Amsterdam, The Netherlands); Alain PAGANO (Angers, France); Heike PRÔHL (Hanno-

ver, Germany) ; Miguel VENCES (Braunschweig, Germany).

TARIFF FOR 2007

INDIVIDUALS

[Regular 2007 subscription to Apres (volume 25) + ISSCA + Circalytes 50€ orS

Student 2007 subscription to Apres (volume 25) + ISSCA + Circalytes 25€0or$

SreciaLorrer: gift half-pricesubscription of one year for a colleagueof your choice: Half of price above

Regular 2007 subscription to Alyres (volume 23) alone 46€ 0r$

Student 2007 subscription to Alyres (volume 25) alone 23€or$

Back issues of Alyres: any single issue 15€or$

Back issues of Alyres: any double issue 25€or$

Back issues of Alytes: any complete volume (4 issues) 40€ or$

Back issues of Alptes: complete set of volumes 1 to 24 (1982-2006) 70€ or$

Regular five-year (2007-2011, volumes 25 to 29) individual subscription to Alyres +

SSCA + Circalytes 25€or$

Regular five-year (2007-2011, volumes 25 to 29) individual subscription to Alyres 200 € or $

SrectaL orrer: five-year (2007-2011, volumes 25 to 29) individual subscription to Alytes + ISSCA +

Circalytes, with complete set of back issues of Alytes (1982-2006, volumes 1 to 24) 800€ or $

Life individual subscription to A/ytes from 2007 on 1080€or$

Life individual subscription to A/ytes + ISSCA + Circalytes from 2007 on 1200€or$

Patron individual subscription to A/ytes from 2007 on 2160 € or $ or more

Patron individual subscription to Alytes + ISSCA + Circalytes from 2007 on 2400 € or $ or more

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

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

subscription.

INSTITUTIONS

2006 subscription to Apres (volume 25) + ISSCA + Circalyres 100 € or S]

2006 subscription to Alytes (volume 25) alone : 2€or$

Back issues of Alyres: any single issue 30€or$

Back issues of Apres: any double issue 50€ or$

Back issues of Aiyres: any complete volume (4 issues) 80€ or$

Back issues of Alyres: complete set of volumes 1 to 24 1440 € or$

SreciaL orrer: five-year (2007-2011, volumes 25 to 29) subscription to Alytes + ISSCA + Circalytes,

with complete set of back issues of Alytes (1982-2006, volumes 1 to 24) 1600 € or $

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

prices can be provided upon request by our Secretariat.

MODES OF PAYMENT

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

above).

— In Euros, by direct postal transfer to our postal account: “ISSCA”, Nr. 1-398-91 L, Paris. If youuse this

mode of payment, add 2.50 € to your payment for postal charges at our end.

— In US Dollars, by credit card or by cheque sent to: Bibliomania!, PO. Box 58355, Salt Lake City,

UT 84158, USA: phone/fax: +1-801-562-2660; e-mail: Breck@<Herplist.com. Source : MNAN, Paris

AINTES

INTERNATIONAL JOURNAL OF BATRACHOLOGY

May 2007 Volume 25, N° 1-2

Alytes, 2007, 25 (1-2): 1-37. 1

Zoogeography of the treefrogs

in Africa’s tropical forests

Arne SCHIOTZ

Humlehaven 2, 4571 Grevinge, Denmark

<arne@schiotz.dk>

The zoogeographical patterns of the forest faunas of treefrogs (Hype-

rolidae and genus Leptopelis) in tropical Africa are analysed. This group is

well-suited for such a work since especially the genus Hyperolius shows

great diversification and signs of recent speciations. For the purpose of this

analysis, the species are separated into sylvicolous (“high forest”) and

parasylvicolous (‘“farmbush”). The forests in this study are tentatively

divided into ten forest blocks based on differences in the treefrog fauna, and

such differences are analysed.

The sylvicolous species of the genus Hyperolius have a distribution

pattern with a total separation at species level between forest blocks which

may reflect the division of the forest belt through drier periods of late

Pleistocene into a number of isolated refugia where moist forest and its

fauna have persisted and may have resulted in allopatric speciation. The

genera Afrixalus and Leptopelis show less diversification since the sylvi-

colous species of these genera are separated into three regions, or groups

of forest blocks, with groups of species common for these regions, but with

no species common between these regions, and only slightly overlapping in

southern Nigeria. These regions are West Africa west of the Dahomey

Gap/Nigeria, Central Africa and the Eastern Forests.

The parasylvicolous species show a fundamentally different distribution

pattern. One group is found in West Africa and in Central Africa along the

Atlantic coast but not far inland. Another group is found in Central Africa,

including Cameroun and easternmost Nigeria and the forest block along the

Atlantic coast. Finally one group is widely distributed in the dry parts of the

Eastern Forests. The three groups are separated at species level, but the

former two have a considerable area of geographical overlap along the

Atlantic coast of Central Africa from Cross river southwards. The distribu-

tion of the parasylvicolous fauna does not show well-defined “faunal

breaks” similar to the sylvicolous fauna, but species diversification and

distinction between vicariant taxa seem to have taken place at different,

apparently “random” places, perhaps reflecting the original, linear distri- 2

bution of this fauna in a narrow forest edge towards the savanna. È

The ill-delimited orophile treefrog faunas are found in three areas with %

full separation at species level: Mount Cameroun and the Cameronese £

ridge; the mountains in Central Africa (Albertine Rift and adjacent È

highlands); and the Eastern Arc Mountains. 8

A recent proposal that the species structure in the genus Hyperolius ®

can be traced back to Cretaceous is discussed. It is argued that climatic os

changes in late Pleistocene ar ieng to explain the present species Ê

structure.

Source : MNHN, Paris

2 ALYTES 25 (1-2)

INTRODUCTION

It can be argued that the exploration of Amphibia in tropical Africa is still so fragmen-

tary that a comprehensive zoogeographical analysis is premature — or that attempts to

perform such an analysis, even on an incomplete background, are useful as an inspiration for

further studies. Such an attempt is presented here, based on many years of fieldwork in

tropical Africa, and on the taxonomical work of many authors (compilation in SCHIOTZ,

1999).

The aim of the present paper is to analyse the zoogeographical patterns of the forest

faunas of treefrogs in tropical Africa. For the analysis, these forest-related faunas are divided

into two groups relating to their habitat, generally sympatric but not syntopic, i.e. sylvicolous

(high forest) and parasylvicolous (farmbush/bushland) species.

Treefrogs are here defined as the group conventionally termed the family Hyperoliidae,

although recent studies (FRosr et al., 2006) indicate that the genus Leptopelis does not belong

in this group (see app. 1). The mainly Asian treefrog family Rhacophoridae has only one forest

related member in Africa, Chiromantis rufescens, but there is some doubt (RÔDEL, in litt.)

whether populations from western and central Africa are conspecific. This species is therefore

not treated further here.

The present attempt to base a zoogeographical treatment on ecological divisions is

inspired by studies from West Africa (Scmiorz, 1967) of the three clearly separate

lowland faunas, associated with savanna, high forest and farmbush respectively. Among

them, especially the latter is sympatric but rarely syntopic with the former two. It was shown

in the West African study that these three faunas show three clearly different distribution

patterns: the savanna-living species are generally widely distributed throughout the West

African savanna and further East, often stretching to western Ethiopia and Uganda, the high

forest species are clearly localised to “forest islands”, today partiy confluent, and the farm-

bush fauna has a wide distribution, with tendencies to subspecific or specific splitting up at

apparently “random” places throughout West Africa.

The paper is an attempt to carry out a similar analysis for the entire forest belt of tropical

Africa, and more specifically to address a number of questions, such as: (1) is there a basic

difference in the zoogeographical pattern of the two proposed “faunas”, sylvicolous and

parasylvicolous, making it relevant to analyse the two faunas separately?; (2) what are the

zoogeographical patterns for the two faunas?; (3) what is a possible explanation for these

patterns?

A problem with the present treatment based on an ecological division is that a direct

comparison with Poynton’s several papers (e.g. POYNTON’s 1999 benchmark paper) is diffi-

cult, since his approach, without a clear distinction between the faunas, tends to obscure part

of what I see as the distinctness of the biogeographical regions and blocks.

The African treefrogs are deemed suitable for such a study because: (1) their distribution

is reasonably well known on a continental scale: (2) the taxonomy of the forest-related species

is reasonably settled: (3) so are their habitat preferences; (4) they show signs of recent

speciation:; (5) they contain a suficiently large number of taxa for analysis.

Source : MNHN, Paris

SCHIOTZ 3

Fig. 1.— Division into forest blocks.

The dotted areas on the map represent for West- and Central Africa moist, evergreen forest (rainforest) as

climax vegetation. In Eastern Africa such areas represent rainforest in the Eastern Arc as well as

dry, semideciduous forest inhabited by the parasylvicolous fauna.

The numbers refer to forest blocks (see text): 1, Liberia; 2,: Gold Coast; 3, Trans-Volta-Togo;

4, Southern Nigeria; 5, Cross-Sanaga Coastal Forests; 6, Congolian Coastal Forests; 7, North-

western Congolian Forests; 8, Central Congolian Forests; 9, North-eastern Congolian Forests: 10,

Eastern Forests.

Our knowledge of the distribution of the Amphibia in the forest regions of Africa

spreads from very good (Cameroun, western Côte d'Ivoire, Eastern forests) to grossly

inadequate (much of the Congo basin).

The taxonomic knowledge of most treefrog genera is fair in the way that they have been

subjected to rather recent treatments over large areas - sometimes continent-wide — although

it can be argued that a thorough systematical treatment based on DNA, which will give us a

deeper understanding of the phylogenetic relationships, has just started, and has until now

been so sporadic that the conclusions drawn can be disputed.

The level of knowledge of the two faunas, parasylvicolous and sylvicolous, is very

different. The conspicuous and easily accessible parasylvicolous fauna tends to be well-

known, whereas the sylvicolous fauna is incompletely known until searched for by experi-

enced collectors, seeking unconventional and sometimes inaccessible places and finding the

Source : MNHN, Paris

4 ALYTES 25 (1-2)

Fig 2. - Transect lines used in tables 1-4.

few and scattered specimens guided by their voices. The rule is therefore that the sylvicolous

fauna tends to have been recently described and that new discoveries are mainly made in this

group.

Although most taxa treated here are known from a reasonable number of localities, the

exact borders of their distribution are often guesswork. For the parasylvicolous fauna, in

most cases we have a satisfactory density of records, and furthermore often sets of allopatrie

taxa — sometimes regarded as subspecies — replacing each other in a way we interpret as

vicariance, not only geographically but also ecologically. The sylvicolous fauna is more

difficult. A number of species were recorded only from part of the proposed ecoregion.

Sometimes it is because only this part has been thoroughly searched, sometimes because the

species may be confined to a wetter, “richer” core area. Both explanations seem valid for the

rich fauna of western Côte d'Ivoire, the part of the forest block with most rainfall, but also by

far the best explored forests in the Liberia block. The richness of the faunas of southern

Cameroun compared to areas to the East and South may mainly be due to the thorough

collecting effort here. À consequence of this uncertainty — not likely to be solved in the near

future — is that whereas the distinction between faunas of the forest blocks can be assessed, the

identification of the exact borders between them, especially in Central Africa, must be left

open.

Source : MNHN, Paris

SCHIOTZ 5

Table 1. - Distribution of sylvicolous taxa, from Sierra Leone to the Indian ocean, along transect W-X

(fig. 2).

The numbers refer to forest blocks (see text and fig. 1): 1, Liberia; 2,: Gold Coast; 3, Trans-Volta-Togo: D:

Dahomey Gap; 4, Southern Nigeria; 5, Cross-Sanaga Coastal Forests; 6, Congolian Coastal Forests: 7, North-western

Congolian Forests; 8, Central Congolian Forests; 9, North-eastern Congolian Forests; A, Arid Corridor: 10, Eastern

Forests. FN, Footnotes. Note that block 8 breaks the linear sequence. +++, widely distributed within the block; ++,

collection effort insufficient to show distribution: +, only known from very restricted range in spite of ample collection

Central: E, East; N, North, S, South; W, West. Abbreviations of generic names in this table and next ones: 4, Afrixalus; Ac,

Acanthixalus; Al, Alexteroon:; Ar, Arlequinus: C, Callixalus: Ch, Chlorolius; Cr; Cryptothylax; H, Hyperolius; K, Kassinaï

L, Leptopelis: O, Opisthothylax: P, Phivetimantis

Species name 1 2 3 D | 4 3 6 7. 8 9 | A | 10 | FN

H. chlorosteus +++

H. zonatus +++

H. wermuthi +

H. nienokouensis +.

K. lamottei +

Ac. sonjae #4 [4

L. macrotis +++ [+++

A. vibekensis + +

H. viridigulosus E.|+#+

H. laurenti +++

H. bobirensis ++

L. occidentalis +4 le 9 .

A. nigeriensis ++ [44e +++

H. torrentis ++

L. brevirostris #4 [4 N

©. immaculatus ++ [++ N ++

L. boulengeri 9 [Hell 0 ns cet

AL jvnx

H. bopeleti SIN 3

A. schneideri ?

Ar. krebsi dE

L. modestus LEE 7

Ch. kochleri tel N

H. acutirostris #4 N

L. rufus pla

H. endjami a+ [4e

AL hypsiphonus HN PURE nr

Al obstetricans +++ 44e

L. omissus 4 le |

L. crvstallinoron | ++

L. zebra N | +

1. Records from western Nigeria refer either to L. occidentalis or to L. houlengeri (SCMOTZ. 1967).

2. Records from blocks 7-9 (LAURENT, 1973) are doubiful.

3. Erroneously termed a bushland species by SHOT (1999) (AMIE, in.lite.

4. Doubtful records further cast

Source : MNHN, Paris

6 ALYTES 25 (1-2)

Table 1 (continued).

Species name ll 2 3: D 4 5 6 7 8 9 A 10 | FN

H. mosaicus N

H. ghesquieri ++

L. 0. schiotzi +

L. ©. meridionalis +

A. equatorialis 444 [et

A. laevis 444 lee Le Les Le

H. ocellatus ++ le Le a la

Ac. spinosus 444 [ee Lee ++

L. millsoni 4e Lee ae Let a

L. c. calcaratus +++ [+++ +++

L. o. ocellatus ++ +++

À. leucostictus 9 le

C. pictus + 5

H. leleupi + 5

H. frontalis + 6

H. alticola + 6

L. kivuensis + 6

À. uluguruensis ne

L. parkeri He

L. barbouri +++

L. vermiculatus Hs

L. uluguruensis +

H. tannerorum N

ge) Ë

3 [19 ile [13

TOTAL 11 8 1 © leo 20 al æ@ lo 7

c to the ltombwe Plateau.

rit mountains.

+ Five species from the Albertine rifl mountains omitted (see text).

Extrapolation of the distribution brings an element of subjectivity into this study. I

believe this to be defensible especially in the many cases where vicariant taxa replace each

other. But there is a danger that preconceived ideas about the zoogeographical units filter into

published range maps, thus making appear legitimate what is in fact only a working hypoth-

esis.

METHODS

Detailed taxonomic information on taxa are not provided here. Appendix 1 gives the

complete list of taxa mentioned in the text, with their authors and dates and their taxonomic

allocation (in families, genera, species and subspecies) according to FROST et al. (2006). Some

recent information can be found in ScHioTz (1999), RÜDEL et al. (2003), AMIET (2000, 2001,

2004b, 200$), Lôrrers & ScHmirz (2004), Duois et al. (2005), LÔTTERs et al. (2005) and

KÔHLER et al. (2006).

Source : MNHN, Paris

SCHIOTZ

Table 2. Distribution of sylvicolous taxa, from Cross river to Angola, along transect Y-Z (fig. 2). In

order to seek maximal resolution, in this table forest block 6 is subdivided after the lines shown on

fig. 8. The area between lines F and G is block 5, that between G and K is block 6. FN, footnotes.

For other abbreviations, see legend of tab. 1.

Species name H FN

AT. jynx F

Ar: krebsi +

4. schneideri 2

H. acutirostris ++ + 1

H. bopeleti S + 1

H. endjami + res ï

H. mosaicus + 2

LA. laevis ++ ++ 3

L. zebra eo) À

L. crystallinoron +

Ch. koehleri ++ + fe

L. brevirostris + ++ #+ :

O. immaculatus ++ ++ ++ “

L. milisoni #4 ++ #+ # 3

L. calcaratus + + ++ ++ 3

L. rufus ++ +#+ + # + 1

AL. obsterricans ++ ++ # #+ + 2

L. boulengeri + #4 #+ + 5

AL hypsiphonus H+ ++ #+ # # 2

L. omissus ++ ++ #+ # + 2

Ac. spinosus ++ ++ #+ + 3

LL. ocellarus E # + #+ 3

H. ocellatus ++ +6 + ++ + 3

TOTAL Je 19 12* 10 8

(9)

1. Confined to block 5 and (part of) block 6.

2. Also in south-eastern Cameroun, and presumably further east in block 7.

3. Also further east in the Congo Basin.

4. Distribution badly known. Found in south-eastern Cameroun, one record touches present area.

* BURGER et al. (in press) listed 30 treefrogs from south-western Gabon (sector 1-I in

including 7 or 9 unidentificd species, but presented no information about habitat preference.

resent paper),

The forest-related treefrogs in tropical Africa are divided into sylvicolous and parasyl-

vicolous species as further elaborated on pages 10-12 and in tables 1-4. A search for clusters in

their distribution was undertaken. The result is that the parasylvicolous fauna does not fall

into well-defined clusters, but the sylvicolous fauna does. Partly based on such clusters of

sylvicolous species and partly based on recognized zoogeographical entities characterized by

distribution of other animals (OLSON & DINERSTEIN, 1998), the forest was tentatively divided

into ten forest blocks (p. 12). An attempt was made to characterize these forest blocks by their

treefrog fauna and to assess the validity and significance of the blocks.

An attempt was made to make a hierarchical dendrogram illustrating the similarity

between the sylvicolous treefrog faunas in the forest blocks using the Quotient of Similarity

Source : MNHN, Paris

8 ALYTES 25 (1-2)

(QS) (SORENSEN, 1948): 2 X number of taxa common to both areas, divided by sum of totals

of taxa from both areas X 100 (see p. 20 and fig 6).

The statement of close affinity between habitat and fauna is based on numerous

observations of animals at the breeding site. Anecdotal evidence points at a less rigorous

affinity outside the breeding season, where, however, observations are scarce.

The distribution data are taken from many sources, mostly compiled in ScHiorz (1999).

WHY TREEFROGS?

It has been shown earlier in a study from West Africa (SCHOTZ, 1967) that the treefrogs

seem well suited for a zoogeographical analysis, in fact better than other Amphibia. Especially

the genus Hyperolius, which seems to be in a process of recent speciations, shows a finer

geographical separation at the species- or subspecies-level through West Africa than most or

all other genera. The genera Afrixalus and Leptopelis show less splitting-up, but still a

significant one, Genera in other families are less suited for such an analysis, sometimes

because their taxonomy is far less understood. For instance, the large genera Phrynobatrachus

and Arthroleptis (only surpassed by Hyperolius in number of species), suffer from an unsettled

taxonomy. To measure with confidence the level of taxonomic uncertainty is hardly possible,

and is not always apparent from published fauna-lists, but an indication of the uncertainty is

that J.-L. Amiet, a most meticulous worker in the best explored country in tropical Africa,

Cameroun, only lists 3 certain names in the genus Arthroleptis out of a total of 12 species

occurring in that country, and only 8 certain names out of 20 species of Phrynobatrachus

(AMHT, pers. comm.). Such reservations, even from a very well-explored area, make the

genera unsuited for continent-wise zoogeographical analyses. Amiet in comparison lists none

of the 24 Hyperolius and none of the 16 Leptopelis taxa as taxonomically unsettled.

THE TWO FAUNAS IN THE AFRO-TROPICAL FOREST

As generally acknowledged, two well-defined faunas of Amphibia co-exist in the forest

belt, sympatric but normally not syntopic. The first one is, at least in the breeding season and

on breeding sites, strictly confined to reasonably undisturbed closed-canopy forests and has

been termed the High Forest Fauna (ScHioTz, 1967, 1975, 1999). The other is confined to

disturbed, often heavily disturbed forest, cultivated clearings in forests, dry forests and

well-developed gallery forests in the humid savanna. The name Farmbush Fauna was used

in ScHiorz, (1967, 1975) for this fauna, changed to the Bushland Fauna in ScHiorz (1999).

The change, which may have been unfortunate, was undertaken because several colleagues,

mainly those not familiar with the tropical moist lowland forests of Africa, had difMiculties

with the distinction between the two faunas and in understanding the name farmbush which

Source : MNHN, Paris

SCHIOTZ

Table 3. - Distribution of parasylvicolous taxa, from Sierra Leone to Angola, along transect W-Y-Z (fig. 2). Capital

letters refer to “breaks” (see fig. 8) as distinet from the forest blocks of table 1. EW and EE delimit the

Dahomey Gap. FN, footnotes. For other abbreviati

ns, see legend of tab. 1.

Species name A B C D Ew E F

K LEN

FH. occidentalis FF

H. nimbae E

K. cochranae +

K. arboricola 4e Let

P. boulengeri W 444 [ae w

H. picturatus 444 fe Let

H. baumanni 4H

H.f. fusciventris ++ [ee

H. f. lamioensis ++

H.f. burtoni +++ ++

HS: ssp. LEE

L. hyloides ++ [ee Le Lee Lee 44

H. c. concolor 444 [ee lets Lee

:. ibadanensis +++

He. ssp. ++

Hs. ivorensis +++

Hs. sylvaticus ++

Hs. nigeriensis +44 [44e

A guttulatus ++ Le Le +4 [4

4. d. dorsalis ++ Lee Le Lee ++ [4er

4. de regularis

P. boulengeri E

H. tuberculatus

4. lacteus +

Hi r. riggenbachi +

Hr. hieroglyphicus +

H. camerunensis û

L. notatus ++

H. bolifambae 44

H kuligae +++

4. paradorsalis

H. dintelmanni

H. pardalis

L. aubryi

4. phantasticus

H. platyceps

H. adspersus +

P. leonardi

Cr greshofi

H. cinnamomeoventris

A. osorioi

+++

“+

44 [4

mess

#47

++ | ++

TOTAL 3191940 8 |17

8 |3(4)

1. Known only from a very small area near cutting line B.

2. The widely separated P. houlengeri East and West are here treated as separate taxa

Possibly Separate, vicariant species (RODEL & ERNST, 2003, RODEL et al., 2005).

4. Largely montane. endemic to the Cameronese ridus

$. Confined 10 a small area in Cameroun.

6. Doubtful Uganda record.

7. Distribution virtally unknown due 10 confusion with */. matt”.

+ BURGER et al. (in press) listed 30 treefrogs from south-western Gabon (sector

unidentified species. but presented no information about habitat prefere

121 in present paper). including 7 or 9

Source : MNHN, Paris

10 ALYTES 25 (1-2)

Table 4. — Distribution of parasylvicolous taxa, from Cross river to the Indian ocean, along transect Y-X

(fig. 2).

Only species ocurring east of Cross river are included. For species further west, see table 3. Numbers refer 10

forest blocks (fig. 1) since data are 100 incomplete to suggest dividing breaks. Block 6 is only touched in its northern part.

Blocks 7, 8 and 9 do not form a linear sequence. For other abbreviations, see legend of tab. 1.

Species name 5 6 7 8 9 A 10 FN

FH fasciventris FF ñ

H. concolor ++ ï

H. svlvaticus ++ ï

4. bopeleti +

A. lacteus + 2

Jr. riggenbachi + 2

Er. hieroglyphicus + 2

H. camerunensis #

H gutiularms ++ # *

4. dorsalis ++ ++ =

PP. boulengeri E +++ N

4. dintelmanni +

4. paradorsalis ++ ++ w

H. bolifambae ++ + w

Laubrvi ++ ++ +

H adspersus + #4 w 4

A. kuligae ++ ++ w

H. pardalis ++ +

P. leonardi s w w

H. platvceps ++ w #

L. notatus ++ ++ ++ ++

A. phantastic ++ ++

C. greshoffi +++ ++ ++

H mberculatus +4 ++ ++ ++ ++

H. cinnamomeoventris ++ ++ #+ ++

L christvi , #4 5

A. osorioi E ++ ++

K: mertensi + ##+

A. robustus + +

P verrucosus ++ ++

L. fisiensis #+

A huisebauui +

H. castaneus : 6

H. lateralis #4

L. mackayi E? 7

A Kivuensis ++ 8

H. puncticulatus ++

4, müchelli ++

L. flavomaculatus ++

IL rubrovermiculatus N

À. sylvatieus N

P heithae | À à

Li 7 E

orat 18 1 A ni 5 6

further west (table 3).

lemic 10 the Cameronese ridue.

Montane.

3. Oceurring further west and south (table 3)

4. Distribution badly known.

5. Cameronese populations may belong 10 a different taxon (AMIET. 2004u)

6. Montane from Albertine rifl mount:

7. Only known from western Kenya, possibly also in République Démocratyique du Congo.

$. Also dense savanna south of forest belt

9. Habitat preference badly understood.

Source : MNHN, Paris

SCHIOTZ Il

is meaningless in the savanna and dry forests where farmed or degraded land harbour the

same fauna as unfarmed land. To avoid such misunderstandings, I will instead use terms

which may give fewer associations, namely sy/vicolous for the High Forest Fauna, and

parasylvicolous for the Farmbush/Bushland fauna, names proposed by Amir (1989) — in

French as sy/vicole and parasylvicole — in addition to the term savanicolous for the savanna-

living forms.

It was shown for West Africa (SCHIGTZ, 1967) and for Eastern Africa (SCHIOTZ, 1976,

1981) that the zoogeographical patterns for the sylvicolous and the parasylvicolous faunas

differ profoundly from each other (and both differ profoundly from the pattern of the savanna

fauna), so that the most “precise”, and therefore most informative picture is obtained by

keeping these faunas separate. This is therefore also done in this study.

The 103 forest-related treefrog species — 115 taxa when considering recognized subspecies

— suited for analyses can be separated into 49 parasylvicolous and 54 sylvicolous species, or

59/56 when subspecies are considered, plus a few excluded species (app. 3). Since the

separation into the geographically largely overlapping sylvicolous and parasylvicolous faunas

can only be undertaken based on a thorough field-knowledge of the forms, there is a certain

element of subjectivity in this separation and it can hardly be expected that workers in

different parts of the forest belt agree completely on the distinction between sylvicolous and

parasylvicolous species. It might, however, be relevant to compare this distinction in two

comparable areas covered by different workers, namely the rather well-investigated West

Africa west of the Dahomey Gap (mainly Schiotz and Rôdel, compilation in SCHIOTZ, 1999

and RÔDEL, 2000) and the very well investigated Cameroun (mainly Mertens, Perret and

Amiet, compilation in PERRET, 1966 and AMIET, pers. comm.). West Africa has 14 species of

sylvicolous treefrogs, 13 parasylvicolous; Cameroun has 22 and 23 respectively. These ratios

could indicate that the distinction between the two faunas by the two set of workers is

congruent, supported by the fact that Amiet’s and my references to habitat type for the species

we both know from the field is identical.

Practical field experience is deemed necessary. The division in this paper is thus based on

the author having encountered 81 of the 115 treated taxa in the field, supplemented with

Amiet’s field experience of an impressive 46 taxa out of a total of 48 in Cameroun, of which

21 have not been encountered in the field by me (AMIET, 1986 and in litt.). Habitat preference

for the taxa encountered neither by Amiet nor by me is given by a number of other authors,

accepted here since there is no disagreement with these authors’ general allocation of species

to habitat. À compilation of data for distribution and habitat preference is found in SCHIOTZ

(1999), supplemented with data found in the description of more recently described species

(see DuBois et al., 2005). Only a recent paper by WiECZOREK et al. (2000) differs profoundly

from the present paper in its allocation of Hyperolius to habitat (discussed further p. 30). In

spite of his many years in Africa, Laurent, the authority on the African treefrogs, only rarely

mentioned habitat affinity in his papers.

It should be noted that the habitat affinity, even of the rarer taxa of which only few

s normally based on many records of voices where the

s calling from an unusual place. Exceptions from

d rather than downplayed by our collecting

species has its own ecological requirements and

specimens have been collected

collectors especi:

the habitat affinity

methods. Since it is to be expected that eve

Source : MNHN, Paris

12 ALYTES 25 (1-2)

preferences it is not surprising that some species do fall outside this rigorous division,

sometimes perhaps because of a broader requirement, sometimes perhaps by requirements

not expressed by the sylvicolous/parasylvicolous division. Such species are dealt with in

appendix 3.

Amiet has in his later papers (first AMIET, 2001) introduced the term phonocénose for the

“sound-scape” in which the calling males are found, “des ensembles d’espèces qui exercent

leur activité vocale... pendant la même période de l’année et dans des sites présentant des

caractéristiques écologiques similaires”. He uses this concept in addition to, not replacing, his

distinction between sylvicolous and parasylvicolous species. This term is perhaps a more

precise definition of what I have called “faunas”, for strictly speaking we know very little of

the affinity to the vegetation outside the breeding season and breeding sites. Very few

specimens are collected outside the breeding season and breeding sites, and there is some

anecdotal evidence that some migration through “alien” vegetation occurs, at least for

savanicolous and parasylvicolous species (ScHIOTz & DAELE, 2003: 143; AMIET, in litt.). On

the other hand, the term phonocénose may be too narrow for the present purpose since, at

least within the sylvicolous species, some are connected with small streams, others with small

stagnant swamps, etc. They may not belong to the same phonocénose, but are here regarded

as belonging to the same fauna. Also the timing of breeding activity in the rainy season may

separate species belonging to the same fauna into different phonocénoses.

Poynton, in several papers, expressed some reservation as to the link between Amphibia

distribution and vegetation (e.g., POYNTON, 1962: 34: “large-scale faunal patterning which is

determined mainly by the vegetation patterning should be treated with a great deal of

caution”). My approach, however, is to regard affinity to the vegetation types as the funda-

mental basis for the distribution of the species. We do not know specifically which factors are

decisive, but field observation (not only of frogs) demonstrates clearly this dependence, and it

is extremely rare to find specimens in “the wrong vegetation”. Microclimate, so dramatically

different especially between savanna and closed canopy forest, may be one factor. This could

explain POYNTON’s (2000b) observation that species which are forest-limited in lowlands,

sometimes occur in open formations at high altitude.

The smallest recognised systematical unit, subspecies, is used in this paper. It is notewor-

thy that no cases are known of different subspecies within a species occupying different

habitats. Habitat affinity seems to be a fundamental species character. The two faunas,

sylvicolous and parasylvicolous, are thus taxonomically separate at species level, and are

approximately of similar magnitude of species diversity in any given area.

FOREST BLOCKS

A basis for the present study is the forest ecoregions proposed by the World Wildlife

Fund (WWF) in their “Global 200” study (OLSON & DINERSTEIN, 1998). WW] ï

ecoregions is a consensus result of wide consultations with workers familiar with diff

animal and plant groups. An ecoregion is defined by WWF as “a geographically distinct

assemblage of natural communities that share a large majority of their species, ecological

dynamics and environmental conditions”. This division differs somewhat from that proposed

Source : MNHN, Paris

SCHIOTZ 13

by POYNTON (1999), whose definition of regions is more pragmatic: “a region is... an area

covered by a perceived set of species ranges”. Another definition —even more pragmatic is

that of NoBLe (1924: 152) for his biogeographical regions: “convenient areas for distribution-

al discussion”.

Since several of WWESs ecoregions invite to further subdivision when looking at the

fauna of treefrogs, several of them has been subdivided in this paper into forest blocks based

on a perceived set of species ranges. In this paper, the term forest block is used for all the units

studied here. Unfortunately, the terminology for blocks used here is not directly comparable

with the divisions in POYNTON (1999) (app. 2).

THE SYLVICOLOUS FAUNA

DISTRIBUTION

In this study, ten forest blocks (fig. 1) are tentatively recognised as a basis for discussion of the

distribution of the sylvicolous fauna. They are a result of my analyses of the distribution of

the sylvicolous fauna. The parasylvicolous fauna has not contributed to their characteriza-

tion. The validity and significance of these blocks are described and discussed in this and the

following chapter and tables 1-2.

Block 1. Liberia Block

Distribution. —- From western Guinea, western Senegal and Sierra Leone to V-baole which is

a tongue of savanna stretching almost to the coast along Bandama river in eastern Côte

d'Ivoire.

Knowledge. — Well explored (LAURENT, 1958, GUIBÉ & LAMOTTE, 1958; RÔDEL, 2000;

ScHiorz, 1967).

Description. — Forest degraded and fragmented. The westernmost part (western half of Sierra

Leone, western Guinea and Senegal) seems to harbour a dry part of the forest with only one

sylvicolous species, Hyperolius chlorosteus, recorded. The parasylvicolous fauna it this part is

distinctive (p. 26).

Endemics. — Hyperolius chlorosteus; *H. nienokouensis: H. wermuthi: H. zonatus: *Kassina

lamottei. Species marked with * are only recorded from the central part of the region.

Block 2. Gold Coast Bloc

Distribution. — Eastern part of Côte d'Ivoire from V-baole eastwards to Volta river in eastern

Ghana.

Knowledge. — Rather well explored (SCHiOTZ, 1967).

Source : MNHN, Paris

14 ALYTES 25 (1-2)

Description. - Forest much degraded and fragmented.

Endemics. - Hyperolius bobirensis; H. laurenti, H. viridigulosus. The latter occurs also in the

easternmost part of block 1, east of river Sassandra.

Block 3. Trans Volta-Togo

Distribution. — À narrow peninsula of forest-clad hills and low mountains in Ghana east of

Volta river and in western Togo.

Knowledge. — Not well explored (ScHiorz, 1967; RôDEL & AGYEI, 2003).

Endemics. - Two endemic treefrogs, the parasylvicolous Hyperolius baumanni and the sylvi-

colous A. torrentis.

Block 4. Southern Nigeria

Distribution. —- From the Dahomey Gap (or Benin Gap) to Cross river.

Knowledge. - The parasylvicolous fauna is well explored, the sylvicolous fauna that may exist

in the few and scattered remaining forests is almost unexplored (SCHiOTZ, 1967).

Description. — Southern Nigeria from the border with Benin to Cross river is today a

forest-savanna mosaic so heavily influenced by man that closed-canopy forests are few and

scattered. These forests have not been the subject of qualified collection, so the sylvicolous

fauna, if still existing, is almost unknown. Collections indicate one, possibly two sylvicolous

species belonging to western blocks (4frixalus nigeriensis; possibly Leptopelis occidentalis), a

few to the eastern (Ophistothylax immaculatus; Leptopelis brevirostris; possibly L. boulengeri).

Block 5. Cross-Sanaga Coastal Forests

Distribution. - From Cross river in eastern Nigeria to Sanaga river in Cameroun, covering the

coastal lowlands and the forested parts of the Cameronese ridge.

Knowledge. — Very well explored (PERRET, 1966; AMIET, 1986). Unpublished fauna list

distributed privately by Amiet in 2004 (AMIET, pers. comm.).

Description. — The lowlands are today a mosaic of humid forest, degraded forest and

farmland. Mount Cameroun and the Cameronese ridge have a number of species often

regarded as montane but, according to Amiet (in litt.), only Afrixalus lacteus is strictly so.

Endemics. -“*Montane” species, endemic to the ridge, a: perolius riggenbachi (found also

at low altitude on the Benue plains), Afrixalus lacteus and some populations of Leptopelis

modestus. In lowlands, Arlequinus krebsi, AleXteroon jynx and the enigmatic Afrixalus schnei-

deri.

Block 6. Congolian Coastal Forests

Distribution. — Stretching from Sanaga river in Cameroun in a belt along the Atlantic coast to

Angola. South of Cameroun, the vegetation is a mosaic of forests and savannas. The eastern

Source : MNHN, Paris

SCHIOTZ 15

limit towards block 7 is not well defined. The southern limit seems to coincide with the border

Angola-République Démocratique du Congo although one, probably two parasylvicolous

species, Afrixalus dorsalis and probably Hyperolius adspersus, are found as far south as coastal

central Angola.

Knowledge. - Northern part (Cameroun) is very well explored (see under block 5). Scattered

collections from further south: mount Alén, Equatorial Guinea (Lasso et al., 2002; Riva,

1994), central Gabon (FRETEY et al., 1998, 2001), south-western Gabon (BURGER et al., 2004),

south-western République du Congo (LARGEN & DOWSETT-LEMAIRE, 1991) and Mayombe,

coastal République Démocratique du Congo (LAURENT, 1943, 1972, 1976, 1982).

Description. — The northern, well explored part (Cameroun) has one endemic species,

Hyperolius dintelmanni, and shares the following with block 5: H. acutirostris, H. bopeleti and

Chlorolius koehleri. Hyperolius endjami and Leptopelis rufus are endemic for block 5 and 6. In

north-western Gabon, Leptopelis crystallinoron is found. Several species are common for

block 5 and 6 and furthermore occurring to the east (table 2). Otherwise this region is

characterised by the occurrence of a number of West African parasylvicolous species (p. 26).

No sylvicolous species is known south of the border Angola-République Démocratique du

Congo.

Block 7. North-western Congolian Forests

Distribution. — Preliminary delimitation are Ubangi and Congo rivers and to the west the

Congolian Coastal Forests.

Knowledge. - The part of the forested Congo cuvette situated in south-eastern Cameroun has

been explored by Amiet but, according to him (in litt.), needs further research. The remaining

area is virtually unexplored. The entire Congo basin (blocks 7, 8, 9, part of 6) has been

covered by papers by LAURENT: Hyperolius (1943), Leptopelis (1972b), Cryptothylax, Kassina

and Phlyctimantis (1976), Afrixalus (1982). Outside of the three old national parks these

papers, however, are based on scattered and limited material. Of these parks only part of one,

Pare National des Virunga, is within the scope of the present paper.

Endemics. - No endemic treefrogs are known for this block.

Note. - The Congo-Lualaba and Ubangi rivers are regarded by WWF (OLSON & DINERSTEIN

1998) as borders between ecoregions — here listed as 7, 8 and 9. It may seem doubtful if rivers

in the wettest forest or swamp forest along them really are major zoogeographical borders for

Amphibia.

Block 8. Central Congolian Forests

Distribution. — The forested parts of the Congo basin, south of the Congo-Lualaba

river.

Knowledge. - The least known of the ecoregions. Only few, scattered collections (SCHIOTZ,

2006).

Source : MNHN, Paris

16 ALYTES 25 (1-2)

Endemics. — Cryptothylax minutus, Kassina mertensi, Hyperolius ghesquieri, H. robustus and

H. sankuruensis are recorded only from this region, but since they all are known only from a

single or very few localities, their real distribution is unknown.

Block 9. North-eastern Congolian Forests

Distribution. — East of Ubangi river, north and east of Congo-Lualaba river, here taken also

to include forests at medium altitude in Uganda and westernmost Kenya. In WWF's Global

200 (OLsoN & DINERSTEIN, 1998), the eastern border is set at the Albertine rift.

Knowledge. — Virunga National Parc (mainly savanna-forest mosaic) is well explored (LAU-

RENT, 1950, 19724), but in spite of large collections the sylvicolous lowland fauna is virtually

unknown. There are lowland collections mainly from the Ituri area (NOBLE, 1924; names

updated by POYNTON, 1998) and from montane western Uganda (DREWES & VINDUM, 1994),

and collections from forests at low and medium altitude in Uganda and western Kenya

(ScHorz, 1975; ScHick et al., 2005). Otherwise very few collections exist. The suggested

border areas between this block and blocks 7 and 8 are quite unexplored.

Endemics. - From lowland, the imperfectly known Hyperolius ferrugineus and H. langi

(appendix 3). Afrixalus leucostictus may also be endemic: one doubtful record exists from

block 8 (LAURENT, 1982), this specimen has been examined, and I find it impossible to decide

whether it is À. equatorialis or À. leucostictus. Endemic are also the parasylvicolous Leptopelis

Jiziensis, L. mackayi, Hyperolius lateralis and H. hutsebauti.

The Albertine Rift mountains and highlands harbour a number of montane forest-

related treefrogs, all endemic to these mountains, several of them very poorly known and with

no ecological information: Callixalus pictus, Hyperolius frontalis, H. alticola, H. castaneus,

*H. chrysogaster, *H. diaphanus, H. leleupi, *H. leucotaenius, *H. xenorhinus, *Leptopelis

fenestratus and L. kivuensis. Those with an * are omitted in this study because of lack of

information about habitat.

Block 10. Eastern Forests

Distribution. — The forest block, as here understood, consists of small patches of moist,

closed-canopy forests in eastern Tanzania and Malawi (Eastern Arc mountains and outlying

mountain blocks to the south) with a mixture of sylvicolous and parasylvicolous species, and

of areas of dry, semi-deciduous forests in the lowland to the east and south-east of the Eastern

Arc with a mixture of parasylvicolous and savanicolous species. The dry forests form a mosaic

with Miombo Woodland from coastal Kenya to north-eastern South Africa. This latter area

has been termed eastern lowlands (SCH1OTZ, 1976; POYNTON, 1995, 1999). The southern

border of this block is here set at approximately 22S, which seems to represent the southern

limit of the parasylvicolous tropical fauna of treefrogs (Lepropelis flavomaculatus, Hyperolius

puncticulatus and H. mitchelli. Other parasylvicolous species occurring further south, in

South Africa, are not treated in this study.

The eastern forests are separated from the forests in Central and West Africa by an

impressive gap of dry savanna, approximately 500 km wide or more, termed “the arid

corridor” by POYNTON (1995).

Source : MNHN, Paris

SCHIOTZ 17

Knowledge. — Well explored (ScHioTz, 1975; POYNTON, 2000b).

Endemics. - The entire sylvicolous and parasylvicolous fauna in this block is endemic.

Description. — It is important to note that the large forest block “Eastern Forests” as here

understood differs profoundly from the other forest blocks: whereas these other blocks have

closed canopy forest (‘“rainforest”) as their climax vegetation and have been almost totally

forest clad until the influence of man, the climax vegetation in the eastern forests outside of

the Eastern Arc is either dry, semi-deciduous forest or forest-savanna mosaic. Only on the

Eastern Arc and a few areas to the south of them is the climax vegetation moist evergreen

forest with a sylvicolous fauna. The distinction between forest types versus the distinction

between a montane and a lowland fauna is discussed on p. 19.

DISCUSSION

), subdivided in table

Tables 1 and 2 summarize the distribution along transect lines (fi

2asin fig. 8.

The sylvicolous fauna shows a complete separation at species level between the

three major regions with the Dahomey Gap or southern Nigeria as one divide, the Arid

Corridor as the other. These regions are thus: (1) West Africa west of the Dahomey Gap:

(2) Central Africa from southern Nigeria eastwards; (3) the forests of the Eastern Arc

mountains. These regions can be subdivided into a number of forest blocks as indicated above

Gig. D.

The West African forest blocks 1 to 3 are reasonably well explored, well defined and well

delimited. They differ most conspicuously in the genus Æyperolius for which each block has its

distinct fauna of sylvicolous species. Three of the endemic species in block 1 are only known

from the central part of the block, western Côte d'Ivoire (Kassina lamottei, Hyperolius

wermuthi and H. nienokouen, Fhis central part of the block is the best investigated, which

might explain why these secretive species were found there, or the species may be confined to

the wettest, central part of the block, as suggested by RÔDEL (in litt.).

The blocks west of the Dahomey Gap (blocks 1-3) are completely separate at species level

from block 5 and eastwards, but the badly explored - and much degraded - southern Nigeria

{block 4) seems to represent a transition, with the occurrence of the western Afrixalus

nigeriensis and the eastern Opisthothylax immaculatus and L. brevirostris. Either the western

L. occidentalis or the eastern L. boulengeri were collected there (ScHiorz, 1967).

Blocks 5 and 6 have today a confluent forest cover, but with signs (patches of savanna) of

having formerly been divided by a savanna tongue along the Sanaga river (AMIET, 1987). In a

study of some typical Cameronese amphibia - not including any treefrogs —, AMIET (1987) has

demonstrated the importance of the Sanaga river as a border between several species or

subspecies pairs, such as Leprodactylodon a. albiventris versus L. a. bueanus, Leptodactylodon

ovatus vs. L. ventrimarmoratus, Astylosternus diadematus vs. A. batesi, and Cardioglossa

nigromaculata vs. C. gratiosa. In general the difference in the treefrog fauna in Cameroun

north and south of the Sanaga river is not conspicuous (AMIET, in litt.). Arleguinus krebsi and

Alexteroon jvnx are endemic to the northern block, Hyperolius mosaicus occurs only south of

Source : MNHN, Paris

18 ALYTES 25 (1-2)

the river whereas Hyperolius acutirostris and H. endjami are common to block 5 and the

northern, Cameronese part of block 6. Leptopelis rufus seems endemic to the entire blocks 5

and 6.

The fauna of block 6 is not very distinct. Almost all the sylvicolous treefrogs known from

this block are found also in block 5 to the north and/or in blocks 7-8 further east (table 2). The

part of block 6 situated south of the Cameronese border seems poorer in species (both

sylvicolous and parasylvicolous) than the Cameronese part of block 6. It cannot be said

whether this is due to the fauna becoming poorer south of Cameroun -— as is the case when

moving from Cameroun to the west — or because there has been far less collecting effort in this

area.

Block 7, south-eastern Cameroun and north-western Congo, has no sylvicolous species

distinct from block 6. This is probably not due to lack of collecting, for although much of

region 7 is virtually unexplored, its north-western corner (south-eastern Cameroun) is well

explored. Of block 7, only the north-western corner, in Cameroun, is explored, of block 9 only

the north-eastern, leaving a virtually unexplored gap of more than 1000 km, and we comple-

tely lack collections showing whether the suggested borders between the blocks — or ecore-

gions in WWF's terms — 7, 8 and 9 (Congo and Ubangi rivers) are well-defined faunal breaks,

or whether the fauna changes gradually.

Block 8 is the least explored of the recognized blocks, since only scattered, rather

unsystematic collections have been made. Attempt by the author to collect in this block in

1975 and 2005 met only with limited success and could confirm an impression of a species-

poor fauna. The five apparent endemic species for this region are so badly known that nothing

can be said of their real distribution.

Block 9 has apparently only one endemic sylvicolous lowland species (4frixalus leucos-

tictus albeit with one doubtful locality in block 8), in addition to a number of species confined

to highlands at the Albertine rift. In the lowlands outside these mountains the numbers of

sylvicolous species are low (5) which seems difficult to explain except as a collection artefact,

remarkable in view of the large collections made in Parc National des Virunga, also in its

low-lying, partly forested parts (LAURENT, 1950, 1972). Large collections by non-specialists

do, however, not necessarily mean that all sp are ultimately found. Only 3 sylvicolous

Leptopelis and no sylvicolous lowland Hyperolius reported from this area would point

towards grave under-collecting.

Forest block 10, Eastern Forests, has a fauna, both sylvicolous and parasylvicolous,

completely separate at species level from the fauna of the remaining blocks. The sylvicolous

fauna is almost exclusively found in the forests of the Eastern Arc where local topography has

ensured sufficient rainfall to maintain a closed canopy forest also through drier periods.

Considering the minute size of these forests, probably also in the past, the fauna is quite rich

The forests are today split into a number of isolated units the best explored of which are

eastern Usambara, western Usambara, Uluguru and Udzungwa. The distribution pattern

between these isolated forest “islands” seems very similar to the pattern of the two major West

African regions, | and 2, namely that the genera Leptopelis and Afrixalus have species in

common for these “islands”, whereas the two known sylvicolous Hyperolius sem endemic to

one of the islands. The distributions of these sylvicolous Æyperolius are, however, not well

are only known from the type locality — and they may be more widely

known — both speci

Source : MNHN, Paris

SCHIOTZ 19

distributed, so no attempt was made to divide the Eastern Arc Forests into separate forest

blocks on this slender evidence. The dry forests in the lowlands, including the coastal forests,

harbour a parasylvicolous fauna (p. 26).

The fauna and flora in the Eastern Arc Forests is conventionally termed a montane fauna

and flora (e.g., LoverT, 1988, 1990; POyNTON, 1990, 1999, 2000; PoynTon et al., 2006) as

different from the lowland fauna, the forest-related fauna in the eastern lowlands. I have in the

present study preferred to distinguish between a sylvicolous and a parasylvicolous fauna

rather than between a montane versus lowland fauna. Although we do not know the causal

connections between altitude and/or vegetation and the distribution of animals, I do find

such a distinction more relevant for a study of the treefrogs since the known altitudinal ranges

(app. 4) do not seem to support a distinction based on altitude. The sylvicolous fauna is found

as far down as the closed canopy forest (to 200 m on eastern slopes of the Usambaras),

whereas the parasylvicolous fauna is found as far up as the right habitat (degraded forest and

farmland) is found in the Eastern Arc mountains as well as in dry forest in the lowland. The

closed canopy forest seems to harbour the sylvicolous fauna of treefrogs regardless of its

altitude, and the degraded forests harbour the parasylvicolous fauna regardless of its altitude.

Some species (Hyperolius tannerorum, H. kihangensis and Phlyctimantis keithae) were only

recorded from localities higher than 1400 m, but since they are only known from one or a few

localities it is unclear whether they are confined to such altitude. I do not see a dramatic

deviation from common thinking by viewing the distribution of the treefrogs from a

sylvicolous/parasylvicolous perspective rather that one determined by altitude, remembering

Hengeveld’s dictum, cited in POYNTON & BROADLEY (1991): “[biogeographical] classifications

are not right or wrong, only useful or not”.

The general picture for the sylvicolous treefrogs in tropical Africa is that the genus

Hyperolius shows the maximal species diversification (fig. 3) with distinct species in each forest

block. This is clearly the case in Western Africa and along the Atlantic Coast (blocks 1-5 and

the northern part of 6), whereas the data for Central Africa (blocks 7-9) are far too incomplete

to draw any conclusion. In fact, the only sylvicolous Hyperolius we know from these regions

are known from single localities. In the well-explored Eastern Forests, the only known

sylvicolous Æyperolius, H. tannerorum and H. kihangensis, seem to be confined to one forest

each (western Usambaras and Udzungwas), although they may have been overlooked

elsewhere.

Hyperolius ocellatus seems to be the only sylvicolous Æyperolius breaking this pattern,

being found in blocks 5, 6, 7, 8 and 9. It may be significant that this i ecies being both

sylvicolous and parasylvicolous (AMIET, 1986, in litt.), although predominantly sylvicolous.

The genera Leptopelis and Afrixalus are more widely distributed (fig. 4), in most cases so

that one set of species is found in West Africa (block 1-3), another in Central Africa (5-9) and

yet another in the Eastern Forests. Alexteroon may belong to this group, although the

ibution of its three members outside of Cameroun is badly known.

dis

Finally, there is a group of sylvicolous amphibia with a wide distribution in blocks 1 to

9, namely, with our present understanding of their taxonomy, some Bufo and Prychadena, and

the only sylvicolous rhacophorid, Chiromantis rufescens. No sylvicolous Hyperoliidae or

Leptopelis has such a wide distribution (fig. 5).

Source : MNHN, Paris

20 ALYTES 25 (1-2)

H. chlorosteus

H. laurenti

H. torrentis

H. acutirostris

H. tannerorum

6. H. kihangensis

Fig. 3.— Maximal species diversification in sylvicolous Amphibia. This pattern is shown by sylvicolous

Hyperolius. Our knowledge of the sylvicolous Hyperolius in the Congo basin is not suflicient to:

indicate distribution here.

The genus Kassina has only one sylvicolous member, . lamottei from western Côte

d'Ivoire, Phlyctimantis and Cryptothylax none. The genus Acanthixalus is so cumbersome to

collect that little can be said about its real distribution. One species is known from two

localities in West Africa, the other species from a few widely scattered localities in Central

Africa. Opisthothylax consists of one sylvicolous species in Central Africa.

An attempt is made to make a hierarchical dendrogram illustrating the similarity

between the sylvicolous treefrog faunas in the forest blocks (fig. 6) through the Quotient of

Similarity (QS) (p. 7). The QS should in the present case be taken with great reservation since

several blocks are badly explored and/or the number of taxa is so low that the discovery of a

single or a few taxa may change its value considerably. Therefore blocks 3 and 4 have been

omitted and 7, 8 and 9 combined. Block 10 has been divided into three units, Usambaras

(10a), Ulugurus (10b) and Udzungwas (10c). The difference between these reflects the

occurrence of the two sylvicolous Æyperolius, Which may or may not be endemic to Usam-

baras and Udzungwas respectively. An attempt is also made to show similarity between block

5 and those parts of blocks 6 and 7 situated within the well-explored Cameroun in order to

analyse similarities between areas with comparable exploration.

Source : MNHN, Paris

L. macrotis

L. calcaratus

V2 +rus

C_] L. parkeri

Fig. 4. - Medium species diversification in sylvicolous Amphibia, This pattern is shown by sylvicolous

Lepropelis and Afrixalus

Chiromantis rufescens

Fig. 5. — Minimal species diversification in forest Amphibia. This pattern is not shown by any Hypero-

liidae or Leptopelis.

Source : MNHN, Paris

22 ALYTES 25 (1-2)

D F2 1. Liberia Block

2. Gold Coast Block

* so 5. Cross-Sanaga

L___ 6. Congolian Coastal

7-9: Congo Basin

91 40a: Usambara

+— 10b: Uluguru

—— 10c: Udzungwa

5 6n 7nw

5 = 82 64

6n = 53

10a 10b 10c

10a - 91 83

10b - 91

Fig, 6. - Quotient of Similarity between sylvicolous Amphibia in forest blocks. See text for explanation.

*, QS between 5+6 versus 7-9.

The two tables show in greater details the QS between selected areas.

There are no common taxa between the three “ecoregions”, blocks 1-2, 5-9 and 10. When

a possible connection between 1-2 and 5-9 nevertheless is indicated in fig. 6, is it because it

could be argued that these regions show similarity in their faunas through species pairs,

probably closely related. The West African Leptopelis macrotis, L. occidentalis, Afrixalus

vibekensis and Acanthixalus sonjae may thus have “sister species” in Central Africa, namely L.

rufus or L. millsoni, L. boulengeri, Afrixalus laevis and Acanthixalus spinosus. An additional

argument is also that several members of other families, such as Chiromantis rufescens, several

Bufo, etc, are in common between these ecoregions. Block 10 has only one species, Afrixalus

uluguruensis showing similarity to a Central African form, A. laevis. The four Leptopelis

seem dissimilar to any of their Central African counterparts. It can therefore be argued that

the sylvicolous fauna of West and Central Africa may have a common root, whereas the

fauna of the Eastern Arc Forests has developed independently of the rest of Africa’s forest

treefrogs.

Source : MNHN, Paris

SCHIOTZ 23

ZA Afrixalus dorsalis: D

KK Hyperoïius tuberculatus

Leptopelis flavomaculatus 2

ai p.

ruse,

Fig. 7. — The three typical distribution patterns found among the parasylvicolous treefrogs: a West

African distribution (Afrixalus dorsalis, the species which is found further to the South), a Central

African distribution (Hyperolius tuberculatus) and an East African distribution (Leptopelis flavo-

maculaus).

THE PARASYLVICOLOUS FAUNA

DISTRIBUTION

The distribution of the parasylvicolous fauna shows a pattern fundamentally different

from that of the sylvicolous fauna. The overall picture is of three separate regions where,

however, the borders between two of them differ from those of the sylvicolous faunas

(fig. 7):

(1) A western fauna found throughout West Africa and further south in a belt along the

Atlantic coast of Central Africa (blocks 5 and 6). Afrixalus dorsalis and probably Hyperolius

adspersus are found as far south as coastal central Angola, the other species taper out further

north along the coast (table 3 and fig. 8). It is noteworthy that none of these West African

species are found further inland in Central Africa into block 7, in spite of lack of obvious

phys or vegetation barriers.

Source : MNHN, Paris

Fig 8. — Boundaries in the distribution of parasylvicolous treefrogs.

The lines A-G are based on a dense network of collecting localities and in most cases on the

occurrence of vicariant taxa on either side of the line. In contrast, the credibility of the boundaries H-K

is basically different since collecting there was so sporadie that arbitrary lines had to be drawn based on

the scattered collections, and placed roughly between such collecting localities and between the refuges

proposed in fig. 9. Available data do not permit to conclude that the boundaries between several pairs of

vicariant taxa coincide exactly, even when this map implies so.

A, Eastern limit for Hyperolius occidentalis, and western limit for H. picturatus, H. concolor and

Afrixalus d. dorsalis.

B, Boundary H. f. fusciventris vs. H. f. lamtoensis,

C, Eastern limit for H. f. burtoni, and boundary H.

D, Boundary H. picturatus vs. H. baumanni.

E, Dahomey Gap, where w and e represent the western and eastern border of the savanna

gap:boundary A. s. srlvaticus VS. H. $. nigeriensis, H. €. concolor vs. H. €. ibadanensis, and castern

boundary for K. arboricola.

F, Cross river: boundary Æ. c. ibadanensis vs. H. c. H. f burtoni Vs.H. f.

ssp., and western limit for H. bolifambae, H. ocellatus, A. paradorsalis and H. kuligae.

G, Sanaga river: southern limit for H. fusciventris, H. aticus and H. concolor, north-western limit

for H.pardalis, H. platyceps and H. phantasticus, northern limit for Cryptothylax greshoffii, and south-

eastern limit for Æ. camerunensis.

H, Boundary Cameroun-Rio Muni: boundary A. d. dorsalis vs. A. d. regularis, and Phlyctimantis

Leonardi vs. P boulengeri.

1, Central Gabon.

J, Boundary Gabon-République du Congo: southern limit for A. gurtulatus, L. notatus and H.

pardalis.

K. Boundary République Démocratique du Congo-Angola: southern known limit for L. aubryi,

phantasticus, H. adspersus ©), P Leonardi and €. greshofii: northern limit for A. osorioi along coast.

L, Southern limit for A. dorsalis regularis and probably H. adspersus.

M. Boundary A. mitchelli vs H. rubrovermiculatus.

and Kassina cochranae vs. K. arboricola.

sylvaticus ivorensis vs. H. s ic

Source : MNHN, Paris

SCHIOTZ 25

(2) A Central African fauna, from Cross river eastwards, different from the western fauna

at species level. In the western part of the range, in the Cross-Sanaga forest and Congolian

Coastal Forest ecosystem, this fauna is overlapping with the western fauna.

(3) A distinct fauna in the Eastern Forests showing a total separation at species level from

the remaining forests.

There is a considerable splitting up into subspecies, especially in the western fauna.

DISCUSSION

A pattern similar to the splitting up of the sylvicolous Hyperoliidae into distinct insular

faunas in discrete forest blocks is not seen among the parasylvicolous species. Instead, there is

a considerable splitting up at places that appear “random” in the way that a clustering in

groups of taxa sharing the same distribution seems to be lacking (fig. 8). Thi especially

apparent in the well-explored West Africa-Cameroun region and at the subspecific level in the

most “versatile” genus, Hyperolius. In ScHioTz (1967), this “random” splitting up in West

Africa was explained by the natural habitat, namely open, dry forest, being originally only

found in a narrow belt, a forest edge at the well-defined forest-savanna border — plus in gallery

forests stretching as tongues up into the savanna. Only when man disrupted the moist forest

belt through farming and felling and thus created a network of open habitats, did large

areas become suitable for the parasylvicolous fauna. Such an original, “linear” or “one-

dimensional” distribution might facilitate taxonomic splitting up at rather insignificant and

perhaps temporary barriers.

The westernmost part of West Africa (western Sierra Leone and adjacent parts of

Guinea and Senegal) seems distinguished by the presence of the endemic Hyperolius occiden-

talis, as well as by the absence of the otherwise abundant and widespread species Hyperolius

concolor, H. picturatus and Afrixalus dorsalis.

Itis noteworthy that the two most distinct gaps for the sylvicolous faunas in West Africa

are either not recognisable for the parasylvicolous fauna (V-baole) or rather unimportant

(Dahomey Gap).

The Cross river is the western limit for several central African species, whereas a majority

of the parasylvicolous species from West Africa continues along the Atlantic coast at least

into Cameroun. The parasylvicolous fauna of the Cross-Sanaga and the Congolian Coastal

Forests thus consists of two distinct fauna elements, a Central African mixed with a distinct

West African element where no less than 5 out of the 8 parasylvicolous species of treefrogs

occurring in Ghana are also found in remote Cameroun, including 4 out of $ Hyperolius, a

number tapering out when moving south from Cameroun along the coast. None of them are

found inland in Central Africa. In comparison, none of the sylvicolous treefrogs from Ghana

occurs in Cameroun.

uted in the Congo Basin but

not possible with our present

Another element of parasylvicolous species is widely

vicolous fauna of central Afr

a closer analysis of the par

knowledge.

Source : MNHN, Paris

26 ALYTES 25 (1-2)

The Eastern Forest has a pattern of distribution different from the western forest blocks.

The parasylvicolous fauna is widely distributed on suitable localities in the eastern lowlands

as well as on the Eastern Arc. Contrary to the proposed “linear” or “one-dimensional”

original distribution in West Africa, the original (i.e., before man transformed the landscapes)

distribution of the parasylvicolous species in the eastern forest has been “two-dimensional”,

not confined to a narrow zone between moist forest and savanna, since the climax vegetation

in the eastern forests has been large areas with a dry forest/savanna mosaic, habitable

throughout by the parasylvicolous fauna. Of the parasylvicolous species in the eastern forests,

only the species pair Hyperolius mitchelli vs. H. rubrovermiculatus (found respectively south

and north of the Kenya-Tanzanian border) shows a pattern similar to that of West Africa with

vicariant taxa replacing each other. Significantly this is where this eastern forest becomes a

narrow “corridor” near its northern border, and a division there could be explained in the

same way as the distribution pattern in West Africa (fig. 8).

POYNTON (1990) discussed the distinctness of the East African lowland fauna (below

300 m) and reached the conclusion that this fauna is not well defined or well delimited, but

that nevertheless the lowland fauna has “a homogeneity which accords with the idea of a

discrete ‘fauna”* (p. 291). If his analysis (e.g., his figure 4) had included all the parasylvicolous

frogs and had separated them from the savanicolous group, a much clearer pattern would have

emerged with a distinct fauna found in the eastern lowlands, but generally penetrating higher

up than the 300 m limit, into the parasylvicolous habitats in the Eastern Arc forests,

apparently as high up as such habitats are found or explored (app. 4). Members of this fauna

are here found to above 1000 m, sympatric with the sylvicolous fauna. West of the Eastern Arc

mountains, the parasylvicolous fauna of block 10 seems to be absent, perhaps because of

absence of the relevant habitats throughout the Arid Corridor.

POYNTON (1999: 485) found “endemism relatively low” in the East African lowlands. I

claim, however, that endemism in the forest block here termed eastern forests is 100 %, not

only for treefrogs, but also for other Amphibia. This apparently dramatic discrepancy is partly

because Poynton limited his study to lowlands below 300 m, although a significant part of the

fauna is found higher up, partly because I restrict my study to the parasylvicolous and

sylvicolous faunas, omitting the numerous savanicolous species, many of which are found

further inland.

THE MONTANE FAUNAS

Several families of Anura have a strong element of orophile, forest-related species

and genera. In many cases the tadpoles are adapted to swift-flowing streams, and the

species could be termed rheophile rather than orophile. No rheophile treefrogs are

known, and an orophile element in this group is ill-defined and seems weak. A number of

species are only or predominantly found above 1000-1500 m and are often regarded as

montane, but it can be argued that there are few real montane species among the treefrogs

since most species can be found as far down as suitable vegetation is found, to 100-200 m.

Therefore no attempt has been made to distinguish the montane species as a special category

in this study.

Source : MNHN, Paris

SCHIOTZ 27

The “montane” species in the tropical forest belt are found in three areas, namely mount

Cameroun and the Cameronese ridge, the Albertine rift mountains including the Itombwe

plateau, and finally the Eastern Arc mountains. The “montane” treefrog faunas in these three

areas are separate from each other at species level. The best developed orophile element is

found in the Albertine rift montane areas with a number of species known only from high

altitude (see block 9, p. 16). In the Eastern Arc mountains, a few species are known only from

altitude above 1000 m but they are all species known from so few localities that a realistic

assessment of their altitudinal distribution is uncertain (app. 4, discussion p. 19). In the

Cameronese ridge only one treefrog, Afrixalus lacteus, is confined to high altitude, the

remaining treefrogs from this area can occasionally be found at low altitude (AMIET, in litt.).

The well-investigated mount Nimba (1750 m) in West Africa has no orophile treefrogs.

CONCLUSION

THE HISTORY OF THE SYLVICOLOUS AND PARASYLVICOLOUS FAUNAS

In ScHiotz (1967), it was postulated that the distribution pattern of the West African

sylvicolous treefrogs could be explained fully from the vicissitudes in late Pleistocene, during

which the last arid period has divided the forest into isolated islands for sufficiently long

periods to allow the most versatile genus, Æyperolius, to develop discrete species, while the

more conservative genera Afrixalus and Leptopelis did not show a splitting up on either side

of V-baole, but a distinct break at the species level at the Dahomey Gap. The most conserva-

tive genera Chiromantis, Bufo and Ptychadena show only limited division between West Africa

and Central Africa. There are thus three “levels” of species diversification for the sylvicole

Amphibia (fig. 3-5).

The relevant factors seem to be climatic fluctuations resulting in vegetational fluctuations

during the latter part of Pleistocene, over approximately the last 40,000 years (TINSLEY et al.,

1996; MALEY, 1996): a cold and arid period from 40,000 to 12,000 years before present (ybp)

resulted in the lowland forest retracting considerably, to be partly replaced by savanna. Most

extreme aridity occurred from about 25,000 to about 12,000 ybp, during the glacial maxima in

the northern hemisphere. There is, however, extensive evidence that lowland rainforest

persisted in a number of refugia determined by local conditions of altitude, precipitation and

soil type. The most common state in the upper Pleistocene would be a relatively fragmented

forest mass, but with larger areas of forest than those postulated for the most arid phase.

The end of the last glaciation was characterised by a rapid rise in temperature approxi-

mately 12,000 years ago. The lowland tropical forest expanded, reaching a maximum circa

7,000 ybp, a maximum which, however, did not unite the Central African forests with those of

the Eastern Arc.

It is suggested (MaLEY, 1996) that in the last major arid phase (circa 18.000 ybp) forest

refugia persisted as two or three areas in West Africa, one in the Cross-Sanaga forest, four

refugia in the Congolian Coastal forest, one large refugium along Congo river, another along

the Albertine rift and finally a number of small refugia in the Eastern Arc mountains (fig. 9).

CoLyn et al. (1991) suggested a more complicated pattern for the Congo Basin.

Source : MNHN, Paris

28 ALYTES 25 (1-2)

Fig 9. - Proposed dry-period forest refugia. Redrawn after MALEY (1996), with a small refugium added at

mount Doudou, Gabon, as proposed by Soser (1994). Also added are the Eastern Forests.

Based on the proposed forest refugia, one can explain the distinctness of blocks 1 and 2,

and 5 and the northern part of 6, whereas the southern part of 6, south of Cameroun, and of

7,8 and 9 are not well collected enough to allow drawing conclusions. Therefore the four or

five refugia suggested (MALEY, 1996) from Equatorial Guinea (Rio Muni) to Mayombe

cannot with our present knowledge be substantiated through the distribution of the treefrogs,

although we have collections from localities situated at or very close to all of them.

The three ecoregions in the Congo Basin seem, with our present knowledge, to have a

very similar lowland fauna. Whether the small differences between the three blocks point at

three discrete regions with the Congo and Ubangi rivers as borders, or gradual changes in the

huge Congo basin is not known. Three well-defined regions with clear borders, as suggested

by WWE divisions into ecoregions, would seem unlikely since drier periods in Pleistocene

probably never split the Congo Basin forest up into three refugia separated by the river; a

situation of Congo basin refugia as indicated in fig. 9 would seem rather more likely.

Data for primates supported by plants (COLYN et al., 1991) seem to point at faunal breaks

in the Congo basin originating from several nuclear areas on both sides of the Congo-Lualaba

river, suggesting that, during the last arid period. populations survived in several refugia

Source : MNHN, Paris

SCHIOTZ 29

within the Congo basin. Zones of intergradation strongly suggest that dispersal operated

from these areas and not only from the Albertine rift and from the West Central (Cameroun-

Gabon) refugia.

Block 10 shows no sign of having been connected with the rest of the blocks, with no

species in common.

It is difficult to say whether the occurrence of several small genera of Hyperoliidae in

Cameroun is caused by a refugium there having persisted even longer back in time, or by the

forests there having been consistently wetter and thus more favourable for frogs and their

diversification, or on the perseverance of herpetological scrutiny.

The forest in the Congo basin is frequently described in the literature as species-poor, a

poverty explained by it being a young forest as shown by it standing on Kalahari sand. Thus

FAIRBRIDGE (1968) wrote: “almost the whole Congo basin was invaded by dune sands from

the Kalahari during the period 50.000-10.000 B.P.”, and many other authors supported this

idea. WHITE (1993) rejected this theory and suggested it was based on confusion between

“Kalahari sands” and the “Kalahari geological system”. According to White, the sand dunes

and formations are nowhere approaching the equator, and the forest in the central Congo

basin is not particularly young.

The theory that refugia in the lowland forests has acted as evolutionary centres was

challenged by FieLpsA & LoverT (1997). Based on birds and some plant families in forest

biota, where the authors distinguished between old and young species, they did not find strong

evidence for the importance of Pleistocene refugia as centres for allopatric speciation. They

saw the refugia rather as “museums”, maintaining old species, and demonstrated that “the

proliferation of young species takes place mainly at the periphery of the main rainforest

blocks”. The data for sylvicolous treefrogs, meagre when they are compared to birds and

plants, might add arguments to this discussion. For most genera of Amphibia — among the

treefrogs Leptopelis and Afrixalus — the refugia can be seen as museums, carrying sets of

species through periods of forest division, but for Hyperolius the rule seems to be that each

major, well-investigated forest block which has harboured a refugium has developed its own

set of sylvicolous Æyperolius. An explanation could be that the genus Hyperolius is more

“versatile” so that the rather brief periods of isolation of the refugia have been sufficient for

these refugia to act as evolutionary centres. Hyperolius would thus according to Fjeldsä &

Lovett's term be composed of “young species”. The considerable subspecific splitting up of

the parasylvicolous treefrogs compared to the sylvicolous may suggest the parasylvicolous

taxa to be younger than the sylvicolous. The diversification of the parasylvicolous species,

apparently unrelated to former forest refugia, seems to have taken place at “the periphery of

the forest bel”, namely along the original forest-savanna border (present paper, p. 15). Data

for diversity of treefrogs (ScHiorz, 1999: fig.2) using WorldMap as basis for analysis,

apparently support the view of Fjeldsä & Lovett by showing a peak of diversity in the

periphery of the forest belt, most conspicuous in the well-investigated West Africa. A closer

analysis of the data for treefrogs shows, however, that this is solely caused by an increased

habitat diversity in the forest-savanna transition zone within the rather coarse 110 X 110 km

grids used by WorldMap, where the sylvicolous, parasylvicolous and savanicolous faunas are

superimposed. None of these three faunas becomes more diverse here, understood as richer in

are present within a grid, as different from the

species or subspecies, but all three faunas

Source : MNHN, Paris

30 ALYTES 25 (1-2)

condition further away from the transition zone. This comment is not relevant to Fjeldsä &

Lovett’s own conclusions since they deal only with forest-related species.

In strong contrast to the explanation presented here of the present distribution pattern

being explicable by climate fluctuations in late Pleistocene and Hyperolius being a genus of

young species, WIECZOREK et al. (2000) postulated a far longer history for the diversification

in the genus Æyperolius. Their thesis is based on mtDNA sequence data, habitat preference

and current distribution. By referring each node in their dendrogram, based on mtDNA (p.

1237), to a distinct geological, climatic and vegetation change — without making it clear why

each step on their dendrogram necessarily corresponds with one step on their list of palaeoen-

vironmental changes — they claim that they can trace the species diversification in the genus

Hyperolius back to Cretaceous and Palaeocene. An assessment of their theory is difficult since

both their distribution maps and their references to habitat preference are unconventional.

The authors thus work with habitat preferences savanna, forest and generalists. The latter

term is not explained. If the term generalist refers to what is here termed parasylvicolous I do

question the correctness. The parasylvicolous fauna does not seem more generalised in its

habitat choice than the sylvicolous and savanicolous, and the 6 species referred to that

category in their paper do not clarify matters since it refers to 2 montane sylvicolous species,

2 parasylvicolous and 2 savanicolous, whereas all their “forest species” are parasylvicolous in

my terminology. Rather than giving distribution maps based on collected specimens, their

distribution is “extrapolated taking into account. habitat utilisation”. But since their

allocation to habitat utilisation is most unconventional - and unexplained in the text - so are

the resulting distribution maps.

A theory of theirs is that the genus Æyperolius originated as generalists and became

increasingly specialised in terms of habitat. Strangely their only supporting argument,

Hyperolius cinnamomeoventris, which is both the basal lineage on the dendrogram and the

only suspected generalist in the genus (apparently being both parasylvicolous and savanico-

lous, unless it is a composite of several species), is termed in their paper as a forest species and

would thus seem to contradict their thesis. Unfortunately no sylvicolous species from lowland

forests is included in their study. Bringing species diversification in Hyperolius back to

Cretaceous would make this genus the oldest African frog genus together with Xenopus. Not

until Miocene do we have records of other families than Pipidae (DUELLMAN, 1993). The latest

diversification in their dendrogram on p. 1237 is during “Pleiocene” (Pliocene or Pleistocene?)

but seems only to have influenced speciation in the superspecies Hyperolius viridiflavus, a

group of very closely related savanicolous taxa

1 doubt whether the present species structure in Æyperolius has its rooting back in

Cretaceous, 70 million years ago. 1 feel that Hyperolius shows many signs of being a very new

genus, more in the process of rapid, recent speciation than most or all other African genera of

Amphibia. I feel that the species pattern in the genus Æyperolius can be fully explained as the

result of climatie and vegetation changes during late Pleistocene, a period of approximately

35.000 years.

Source : MNHN, Paris

SCHIOTZ 31

ACKNOWLEDGEMENTS

The Carlsberg Foundation has through half a century generously supported my African fieldwork,

the necessary basis for the present study, and has more specifically covered the cost of the publication of

the present paper.

My colleagues J.-L. Amiet and J.C. Poynton as well as three reviewers, among them M.-O. Rôdel, are

thanked for their critical and constructive comments. Birgitte Rubæk, Zoological Museum in Copenha-

gen, has kindly produced the maps. J.C. Poynton and Anne Marie Warfield have revised the language at

two stages of the manuscript.

REFERENCES

Amir, J-L., 1986. - La batrachofaune sylvicole d’un secteur forestier du Cameroun: la région de

Yaoundé. Mém. Mus. nat. Hist. nat., (A), 132: 29-42.

= 1987. — Aires disjointes et taxons vicariants chez les Anoures du Cameroun: implications paléocli-

matiques. Alytes, 6 (3-4): 99-115.

- 1989. — Quelques aspects de la biologie des Amphibiens Anoures du Cameroun. Ann. biol., 28 (2):

73-136.

000. — Les Alexteroon du Cameroun. Alytes, 17 (3-4): 125-164.

2001. - Un nouveau Lepropelis de la zone forestière camerounaise. A/ytes, 19 (1): 29-44.

2004a. — À propos de deux Lepiopelis nouveaux pour la faune du Cameroun. Alytes, 21 (3-4):

111-170.

- 2004. — Une nouvelle espèce d'Ayperolius du Cameroun. Rev. suisse Zool., MU: 567-583

Les Hyperolius camerounais du groupe d'A. nasutus (Amphibia, Anura, Hyperoliidae). Rev.

e Zool., 112: 271-310.

. BRANCH, W. R. & CHANNING, A., 2004. - Amphibians and Reptiles of Monts Doudou,

Gabon. Calif. Acad. Sci. Mem., 28: 145-186.

M., PAUWELS, O. S. G., BRANCH, W. R., Tomt, E., YOGA, J.-A. & MikoL0, E.-N., 2006. - An

assessment of the amphibian fauna of the Gamba Complex of Protected Areas, Southwestern

Gabon. /n: A. ALONSO et al. (ed), Gamba, Gabon: biodiversity of an African equatorial forest, Bull

biol. Soc. Washington, 12: 297-307.

CoLv, M., GAUTIER-HION, À. & VERHEYEN, W., 1991. - A re-appraisal of palaoenvironmental history

in Central Africa. J Biogeogr., 18: 403-407.

DiamonD, A. W. & HAMILTON, À. C., 1980. — The distribution of forest passerine birds and Quarternary

climate changes in tropical Africa. J. Zool., London, 191: 379-402.

R. C. & VinDuM, JL V., 1994. — Amphibians of the Impenetrable Forest, SW Uganda. J af:

Zool., 108: 55-70.

Dunois, A., CROMBIE, R. 1. & GLAW

infrageneric taxonomic additions (1981-2002). Alvres, 23 (1

DusLimaN, WE. 1993. - Amphibians in Africa and South America: evolutionary history and ecological

comparaisons, : P. GoLDBLATT (ed.), Biological relationships benveen Africa and South America,

Yale, U.P.: 200-243

FaiRBkIDGE, R. W., 1968. — Terraces, fluvial-environmental controls. Zu: R. W. FAIRBRIDGE (ed.),

Enclypedia of geomorphologr, 3 124-1138.

FDsA, JL. & Lovert, J. 1997. - Geographical patterns of old and young species in Afri

the significance of specific montane areas as evolutionary centres, Biol. Conserv, 6: 325-346.

Freriy, T. & BLANC, C. P., 2000. - Liste des Amphibiens d'Afrique centrale. Dossiers de l'ADIE, 2: 1-40.

OÛI. — Inventaire systématique des Amphibiens Anoures du centre du Gabon. Bull. Soc. zoo.

France, 126 (4): 375-390.

2005. — Amphibia Mundi, 1.2. Recent amphibians: generic and

n forest biota:

Source : MNHN, Paris

32 ALYTES 25 (1-2)

FRETEY, T. & DE

112 (3): 171-184.

FROST, D. R., GRANT, T., FAIVOVICH, J., BAZIN, R. H., HaAS, A., HaDDAD, C. F. B., DE SÀ, R. O.,

CHANNING, A., WILKINSON, M., DONNELLAN, S. C., RAXWORTHY, C. J., CAMPBELL, J. A., BLOTTO,

M. 1998. - Amphibiens Anoures de la forêt des Abeilles (Gabon). Z afr. Zool.,

C. 2006. - The amphibian tree of life. Bull. amer. Mus. Nat. Hist., 297: 1-370.

Ë, J. & LAMOTTE, M., 1958. - Réserve naturelle du Mt. Nimba; Batraciens. Mém. IFAN, 53: 241-273.

KÔHLER, J, BWONG, B. À., SCHICK, S., Ver, M. & LÔTTERS, S., 2006. — À new species of arboreal

Leptopelis from the forests of western Kenya. Herp. J., 16: 183-189.

LarGEN, M. & Dowserr-LemaiRE, E, 1991. - Amphibians from the Kouilou River Basin, Rép. du

Congo. Tauraco Research Rep., 41: 45-168.

Lasso, C. A. RiAL, A. L., CASTROVIEO, J. & RIVA, L. DE LA, 2002. - Herpetofauna del Parque Nacional de

Monte Alén, Rio Muni. Graellsia, 88 (2): 21-34.

LAURENT, R. E, 1943. - Les Hyperolius du Musée du Congo. Ann. Mus. Congo belge, (1), 4: 61-140.

1950. - Genres Afrixalus et Hyperolius. Expl. du Pare nat. Albert, Miss. de Witte, 64: 1-120.

1958. - La réserve du Mt. Nimba. Les Rainettes du genre Hyperolius. Mém IFAN, 53: 275-299.

19724. - Amphibiens. Expl. du Pare Nat. des Virunga, 22:1-125.

19726. - Le genre Leptopelis au Zaïre. Ann. Mus. r. Afr: centrale, 202: 1-62

1976. — Les genres Cryptothylax, Phlyctimantis et Kassina au Zaïre. Ann. Mus. r. Afr. centrale, 213:

1-67.

1982. — Le genre Afrixalus en Afrique centrale. Ann. Mus. r. Afr: centrale, 235: 1-58.

, Rôbez, M.-O. & BURGER, M., 2005. - À new large tree frog from north-western Gabon.

: 149-15

Lürres, $. & Scumrrz, A., 2004. À new species of tree frog from the Bankossi Mountains, south-west

Cameroun. Bonner zool. Beiträge, 52: 149-154.

LoveripGr, A. 1937. - Scientific results of an expedition Lo rain forest regions in eastern Africa. IX.

Zoogeography and itinerary. Bull. Mus. comp. Zool., 79: 479-541

Lover, J. C. 1988. - Endemism and afinities of the Tanzanian montane forest flora. Monogr. Syst. Bot.

Missouri Bot. Gard., 25: 591-598

—— 1990. - Classification and status of the moist forests of Tanzania. Mit. Inst. allg Bot. Hamburg, 23a:

,, 1996. - The African rain forest: main characteristics of changes in vegetation and climate from

the Upper Cretaceous to the Quarternary. Proc: r Soc. Edinburgh, 104B: 31-73.

None, G. K., 1924. - Contributions to the herpetology of the Belgian Congo. Bull. amer. Mus. nat. Hist.,

49: 147-347.

OLsoN, D. M & DinersTEIN, E., 1998. - The Global 200: a represc

Earth's most biologically valuable ecoregions. Consers. Biol., 12: 502-515.

1966. - Les Amphibiens du Cameroun. Zoo!. Jb. (Syst. ), 8: 289-464.

d'Hyperoliidae restés en question. Bull. Soc. neuchät. Sci. nat., M:

ntation approach to conserving the

Poynron, I. C., 1962. - Geographical and ecological determinants of distribution patterns. Ann. Cape

Prov. Mus. nat. Hist., 2: 32:

—— 1990. - Composition and subtraction patterns of the East African lowland Amphibian fauna. /n: G:

PETERs & R. HUTTERER (ed.), Vertebrates in the tropics: 285-296.

—— 1995. The “arid corridor” distribution in Afrik

Madoqua, 19: 45-48.

= 1998. — Introduction. Gladwyn Kingsley Noble and the study of African amphibians. nr: K. P.

ScHmbr & G. K. NoBLe, Contributions to the herpetology of the Belgian Congo, Reprint edition,

Soc. Study Amph. Rept.. Mhaca, N. Y.

= 1999, — Distribution of Amphibians in Sub-Saharan Africa, Madagascar and the Seychelles. r: W.

E. DUELLMAN (ed.), Patterns of distribution of Amphibia: a global perspective: 483-533.

0004. — Evidence for an Afrotemperate amphibian fauna. 4fr JZ Herp., 49 (1): 33-41

2000b. - Amphibians of Coastal Forests. /r: N. D. BuRGESS & G. P. CLARKE (ed.), Coastal forests of

Eastern Africa, TUCN.

POYNTON, JC. & BROADLEY, D. G., 1991. Amphibia Zambesiaca. 5. Zoog

32: 221-277

among Amphibia.

raphy. Ann. Natal. Mus..

Source : MNHN, Paris

SCHIOTZ 33

POYNTON, J. C., LOADER, S. P., SHERRATT, E. & CLARKE, B. P., 2006. - Amphibian diversity in an East

African biodiversity hotspot: altitudinal and latitudinal patterns. Biodivers. & Conserv.: 1-15.

Riva, I. DE LA, 1994. — Anfibios anuros del Parque Nacional de Monte Alén, Rio Muni, Guinea

Ecuatoreal. Rev. esp. Herp., 8: 123-139.

RÔDEL, M.-O., 2000. — Amphibians of the West African savanna. Chimaira:1-332.

2003. - The amphibians of Marahoué and Mt. Péko National Parks, Ivory Coast. Herpetozoa, 16:

23-39.

RôDeL, M.-O., KosuCH, J., VerrH, M. & ERNST, R. 2003. — First record of the genus Acanthixalus from

the Upper Guinean Rain Forest, West Africa, with the description of a new species. J Herp., 37:

43-52.

RôDeL, M.-O. & AGyYrt, A. C., 2003. — Amphibians of the Togo-Volta highlands, eastern Ghana.

Salamandra, 39 (3-4): 207-234.

RôDeL, M.-O. & BRANCH, W., 2002. - Herpetological survey of the Haute Dodo and Cavally forests,

western Ivory Coast. Salamandra, 38 (4): 245-268.

RÔDEL, M.-O. & ERNST, R., 2001. - Redescription of the tadpole of Phlyctimantis boulengeri. Alytes, 18:

178-186.

RÔDeL, M.-O., Gr, M., AGYEL, A. C., LEACHÉ, A. D., Diaz, R. E., FuyITA, M. K. & ERNST, R.., 2005. —

The amphibians of the forested parts of south-western Ghana. Salamandra, 41: 107-127.

4 GRAFE, T. U., RUDOLF, V. H. W. & ERNST, R., 2002. — A review of West African Spotted

ina. Copeia, 2002: 800-814.

Scick, S, Verrn, M. & LÔTTERs, S., 2005. — Distribution patterns of amphibians from the Kakamega

Forest, Kenys. 4fi: J. Herp., 54 (2): 185-190.

ScmioTz, A., 1967. - The treefrogs of West Africa. Spolia zool. Mus.haun., 25: 1-346.

- 1975, - The treefrogs of Eastern Africa. Steenstrupia: 1-2

= 1976. — Zoogeographical patterns in the distribution of East African treefrogs. Zoo! aff. 11 (2):

335-338.

= 1981. - The Amphibia of the forested basement hills of Tanzania: a biogeographical indicator group.

Afr. J. Ecol., 19: 205-207.

= 1999. — Trcefrogs of Africa. Chimaira: 1-350.

- 2006. - Notes on the genus Hyperolius in central République Démocratique du Congo. A/yies, 24

ELE, P. V., 2003. - Notes on the treefrogs of North-Western Province, Zambia. Alytes,

20 (3-4): 137-149.

M.S. M. 1994. - Refuge Begonias. Studies in Begoniacea, Univ. of Wageningen, 5.

T. 1948. — A method of establishing groups of equal amplitude in plant sociology based on

Similarity of contents and its application to analyses of the vegetation of a Danish

commons. Kgl. Danske Vid. Selskab., Biol. S)

Tinsurv, R. €. LouMoNT, C. & Komrt, H_ R., 1996. - Gcographical distribution and ecology. Ju: The

Biology of Xenopus, Symposia zoo. Soc. London, 68: 35-60.

Wurre, F., 1993. - Refuge theory, ice-age aridity and the history of tropical biotas: an essay in plant

geography. Fragm. For. Gcobor., Suppl 2: 385-409.

WisczoREk, À. M. DREWES R. C. & CHANNING, À., 2000. - Biogeography and evolutionary history of

Hyperolius species: application of molecular phylogeny. Z Biogeogr.. 27: 1231-1243.

Corresponding editor: Alain Duois.

Source : MNHN, Paris

34 ALYTES 25 (1-2)

APPENDIKX |

LIST OF TAXA MENTIONED IN THIS PAPER,

ACCORDING TO THE FAMILIAL AND GENERIC TAXONOMY OF FROST ET AL. (2006)

Genus Species and subspecies

Family HYPEROLIDAE Laurent, 1943

FAcanthixalus Laurent, 1944 sonjae Rüdel, Kosuch, Veith & Ernst, 200

2. Peters, 1875)

CAfrixalus Laurent, 1944 dorsalis dorsalis (Peters, 1873); dorsalis regularis Laurent, 1951:

equatorialis (Laurent, 1941) filvovittatus (Cope, 1861); lacteus Perret,

1976: laevis (Ah, 1930); leucostictus Laurent, 1950; nigeriensis

Schiotz, 1963; osorioi (Ferreira, 1906); paradorsalis Perret, 1960:

schmeideri (Boetger, 1889); sylvaticus Schiotz, 1974; uluguruensis

(Barbour & Loveridge, 1948); vibekensis Schiotz, 1967

pinosus (Buchholz &

Alexteroon Perret, 1988 hypsiphonus Amiet, 2000; jynx Amiet, 2000; obstetricans (AI, 1931)

Arlequinus Perret, 1988 krebsi (Mertens, 1938)

Callixalus Laurent, 1950 jpictus Laurent, 1950

Chlorolius Perret, 1988 koehleri (Mertens, 1940) un

Cryptothylax Laurent & Combaz, 1950 |greshoffi (Schilthuis, 1889); minutus Laurent, 1976

Hyperolius Rapp, 1842 acutirostris Buchholz & Peters, 1875: adspersus Peters, 1877; alticola

Ahl, 1931; baumanni Ahl, 1931: bobirensis Schiotz, 1967: bolifambae

(Mertens, 1938); hopeleti Amiet, 1980: brachiofasciatus Ahl, 1931;

camerunensis Amiet, 2004; castaneus Ahl, 1931; chlorosteus

(Boulenger, 1915); chrysogaster Laurent, 1950; cinnamomeoventris

Bocage, 1866: concolor concolor (Hallowell, 1844); concolor

ibadanensis Schiotz, 1967; diaphanus Laurent, 1972; dintelmanni

Lütters & Schmitz, 2004; endjami Amiet, 1980: ferrugineus Laurent

1943; frontalis Laurent, 1950; fusciventris burtoni (Boulenge:

fusciventris fusciventris Peters, 1876: fusciventris lamtoensis Schioi

1967; ghesquieri Laurent, 1943; guttulatus Günther, 1858; hutsebauti

Laurent, 1956; inornatus Laurent, 1943; kihangensis Schiotz &

Westergaard, 1999; kivuensis Ahl, 1931: kuligae Mertens, 1940: langi

Noble, 1924; Jateralis Laurent, 1940; laurenti Schiotz, 1967: leleupi

Laurent, 1951: feucotaenius Laurent, 1950; mitchelli Loveridge, 1953:

mosaicus Perret, 1959; nasutus Günther, 1865; nienokouensis (Rüdel,

1998); mimbae Laurent, 1958: occidentalis Schiotz, 1967; ocellatus

Günther, 1858: pardalis Laurent, 1948; phantasticus (Boulenger, 1899);

picturatus Peters, 1875; platyceps (Boulenger, 1900): puncticulatus

(Pieffer, 1893); riggenbachi hieroglyphicus Al, 1931 riggenbachi

riggenbachi (Nieden, 1910} robustus Laurent, 1979; rubrovermiculatus

Schiotz, 1975: sankuruensis Laurent, 1979: spinigularis Stevens, 1971:

sylvatieus ivorensis Schiotz, 1967; sylvaticus nigeriensis Schiotz, 1967:

sylvaticus s chiotz, 1967; rannerorum Schiotz, 1982; torrentis

Schiotz, 1967: muberculatus (Mocquard, 1897); viridiflavus (Duméril &

Bibron, 1841); viridigulosus Schiotz, 1967: wermuthi Laurent, 1961:

xenorhinus Laurent, 1972: zonatus Laurent, 1958

Kassina Girard, 1853 arboricola Perret, 1985: cochranae (Lo

1940); lamottei Schiotz, 1967; maculosa (

Laurent, 1952: schioetzi Rüdel, Gra

immaculatus (Bou

rent & Combaz, 1950 |boulengeri Perret, 1986: keithae Schiotz, 1975: leonardi (Boulenger,

1906); verrucosus (Boulenger, 1912)

Phlyctimantis L

Source : MNHN, Paris

SCHIOTZ 35

APPENDIX | (continued)

Genus Species and subspecies

Family ARTHROLEPTIDAE Mivart, 1869

Arthroleptis Smith, 1849

Asiylosternus Werner, 1898 batesi (Boulenger, 1900); diadematus Werner, 1898

Cardioglossa Boulenger, 1900 gratiosa Amiet, 1972: migromaculata Nieden, 1908

Leptodactylodon Andersson, 1903 albiventris albiventris (Boulenger, 1905). albiventris bueanus Amiet,

1981: ovarus Andersson, 1903; ventrimarmoratus (Boulenger, 1904)

Lepropelis Günther, 1859 aubryi (Duméril, 1856); barbouri Ah, 1929: boulengeri (Werner,

1898): brevirostris (Werner, 1898); calcaratus calcaratus (Boulenger,

1906), calcaratus meridionalis Laurent, 1973; christyi (Boulenger,

1912); crystallinoron Lëtters, Rëdel & Burger, 2005; fenestratus

Laurent, 1972; fiziensis Laurent, 1973; flavomaculatus (Günther, 1864).

hyloides (Boulenger, 1906); kivuensis Ahl, 1929; mackayi Këhler,

Bwong, Schick, Veith & Lôtters, 2006; macrotis Schiotz, 1967; millsoni

(Boulenger, 1895): modestus (Werner, 1898); notatus (Peters, 1875);

occidentalis Schiotz, 1967: ocellatus ocellatus (Mocquard, 1902);

ocellatus schiotzi Laurent, 1973; omissus Amiet, 1992: parkeri Barbour

& Loveridge, 1928: rufus Reichenow, 1874: wluguruensis Barbour &

Loveridge, 1928: vermiculatus (Boulenger, 1909); zebra Amiet, 2001

Family BUFONIDAE Gray, 1828

Bufo Laurenti, 1768 =

Family PIPIDAE Gray, 1828

Nenopus Wagler, 1827 =

Family PrYCHAD

Piychadena Boulenger, 1917 =

Family PHRYNOBATRACHIDAE Laurent, 1941

Phrynobatrachus Günther, 1862 =

Family RHACOPHORIDAE Hoffman, 1932 (1858)

Chiromantis Peters, 1854 rufescens (Günther, 1869)

ENIDAE Dubois, 1987

APPENDIX 2

COMPARISON OF TERMINOLOGY BETWEEN THIS PAPER,

WWE's GLOBAL 200 (OLSON & DINERSTEIN, 1998) AND POYNTON (1999)

Present treatment Global 200 POYNTON, 1999

[7 Western province of West Equatorial

Region

Clock 1) + Gold Coast (block 2)

‘rans-Volta -Togo (block 3)

Southern Nigeria (block 4)

s (block 5)

er part of Congolian Central, East and

anaga Coastal For

Coastal Forests South Province

Souher of Coi Sal | of West Equatorial Region

| n Coastal Forests (lock 6) | Seuhern part of on ar Go A ta

orest

North-western Congolian Foresis (block 7)| Western Congo Basin Moist

Forests Not congruent with

Central Congolian Forests (block 8) __ | Centrai Congo Basin Moist Forests subdivisions of WWF

North-eastern Congo Basin

solian Forests (block 9) Moist Forests

North-eastern Cor

Easter province of Intertropical Montane

Region + East Coast Forests

+ Forested parts of eastern Coast Lowlands

Eastern Are Montane For

Coastal Forests

Eastern Forests (block 10)

Source : MNHN, Paris

36 ALYTES 25 (1-2)

APPENDIX 3

NOTES ON PROBLEMATIC SPECIES

A number of treefrogs, especially from Central Africa, have unknown habitat preferences and are

therefore omitted here, They are: Cryptothylax minutus, Hyperolius brachiofasciatus, H. chrysogaster, H.

diaphanus, H. ferrugineus, H. inornatus, H. leucotaenius, H. sankuruensis, H. xenorhinus and Leptopelis

fenestratus.

A group of treefrogs seems to prefer habitats which are intermediary between sylvicolous and

parasylvicolous, and reference to cither of these categories can be disputed. It is my impression that such

species do not generally have broader habitat requirements, being more generalised than others, but are

rather “in between”, preferring dense vegetation in parasylvicolous habitats or open vegetation in

sylvicolous habitats. Hyperolius splvaticus, Kassina cochranae, K. arboricola and Phlyctimantis boulengeri

were with reservations referred to as sylvicolous by Scmiorz (1967). Here, partly based on the extensive

field knowledge of RGDEL (in litt.), they are termed parasylvicolous although found in denser vegetation

in this habitat type. Similarly, AMIET (1986) has listed Acanthixalus spinosus, Afrixalus paradorsalis,

Hyperolius kuligae, H. ocellatus and H. platyceps as occasionally occurring in both habitats. The preferred

habitat is used here.

The so-called Afrixalus fulvovittatus group (Scmorz, 1999) (striped Afrixalus) has an unsettled and

confused taxonomy, probably consisting of several species, some savanicolous, others parasylvicolous.

The parasylvicolous form in western West Africa (4. fulvovittatus sensu stricto) seems well defined in

westernmost Africa, but RÔDEL et al. (200$) mentioned the possible occurrence of this species in

south-western Ghana. RôDEL & PICKERSGILL (in litt.), however, raised doubts about the identity of the

eastern samples. À photo by Lea ent to me by Pickersgill (from Ankassa, Ghana) is of 4. dorsalis with

a middorsal stripe. In Cameroun and Central Africa the situation is unsettled. This complex is disregar-

ded here.

Afrixalus schneideri is an enigmatie species with a characteristic morphology which could indicate

that it is the Cameronese representative of a group of large sylvicolous Afrixalus including À. equatorialis,

A. leucosticrus and A. nigeriensis, but only one specimen is known, from a very well explored part of

Cameroun.

Hyperolius spinigularis has an enigmatie habitat preference, being apparently found both in dense

forest and in very open farmland, but so inconspicuous and therefore easily overlooked that distribution

cannot be given. Most records are above 1000 m. The species is omitted in this study.

ence and taxonomy are badly understood espe-

1999, as A.

Hyperolius langi is omitted since its habitat pr

cially in relation to H. kuligae and H. platyceps. Doubtful records from Uganda (Scrie

kuligae) may be H. langi

Hyperolius cinnamomeoventris is the only Hyperolius assumed to be both parasylvicolous and

savanicolous. Only parasylvicolous populations are dealt with here, as this name might possibly cover

several species.

Spotted Kassina from forest regions form a parasylvicolous group with an unsettled taxonomy. In

West Africa, RôDEL et al. (2002) revised the group, and two members, X. arhoricola and K. cochranae are

included in this study. The third West African species, X. schioetzi, is omitted because of uncertainty

about its habitat affinity. Further east the situation is unsettled. In the well investigated Cameroun, several

taxa seem to be confused under the name Æ. maculosa or K. decorata. Further east, available material is

insuficient. The spotted forest kassinas from Central Africa are therefore omitted.

Leptopelis modestus has a doubtful taxonomy. Possibly four taxa are involved: from Cameroun

possibly two taxa with slightly different morphologies, from montane forests (eastern Nigeria and western

Cameroun) and lowland forests (the latter harbouring the type locality), furthermore recorded from

eastern Congo and from western Kenya, the latter population having recently been described as L

Source : MNHN, Paris

SCHIOTZ 37

mackayi. Only the Cameronese populations of L. modestus and L. mackayi are treated here, the eastern

Congo populations omitted.

Phlyctimantis boulengeri oceurs in the Liberia block and in the westernmost Gold Coast block, and

then, after a hiatus of 1000 km, in easternmost Nigeria and in Cameroun. Being a conspicuous species, it

was probably not overlooked in between. There is suspicion that the two populations, “2 boulengeri W°

and “P. boulengeri E” are two different taxa (RôDEL & ErNsT 2001: ScHi6TZ, 1999), therefore the two

populations are treated here as separate taxa.

APPENDIX 4

ALTITUDINAL RANGES OF SPECIES IN BLOCK 10

Data were kindly provided by POyNTON (in litt.) and supplemented with my own data.

200 m is the lowest investigated altitude with closed canopy moist forest (east slopes of Eastern

Usambaras).

PARASYLVICOLOUS SPECIES.

Hyperolius mitchelli: 10-1050 m.

Hyperolius puncticulatus: 10-2100 m.

Lepropelis flavomaculatus : 10-1370 m.

Phlyctimantis keithae: 1840-1950 m.

SYLVICOLOUS SPECIES

Afrixalus uluguruensis : 750-1740 m.

Hyperolius kihangensis: 1410 m.

Hyperolius tannerorum: 1740 m.

Lepiopelis barbouri: 1000-1740 m.

Leptopelis parkeri: 410-1740 m.

Leptopelis uluguruensis: 200-1250 m.

Leptopelis vermiculatus: 200-1410 m.

Source : MNHN, Paris

Alytes, 2007, 25 (1-2): 38-44.

Defensive behaviour with stiff-legaged

posture in the Brazilian tree toads

Dendrophryniscus brevipollicatus

and D. leucomvystax (Anura, Bufonidae)

Jaime BERTOLUCI*, Ricardo Augusto BRASSALOTI*,

Henrique Oliveira SAWAKUCHI*, José Wagner RIBEIRO Jr.*

& Germano WOEHL Jr.**

* Departamento de Ciëncias Biolôgicas, Escola Superior de Agricultura “Luiz de Queiroz”!

Universidade de Säo Paulo, Av. Pädua Dias 11, Caixa Postal 9, 13418-900, Piracicaba, SP, Brazil

<bertoluc@esalq.usp.br>

** Instituto Rä-bugio para a Conservaçäo da Biodiversidade, Rua Antonio Cunha

160, Sala 25, Baependi, 89256-140, Jaraguä do Sul, SC, Brazil

We observed death feigning with stiff-legged posture in several individ-

uals of the Brazilian tree toads Dendrophryniscus brevipollicatus and D.

leucomystax. This behaviour was formerly described for three other phy-

logenetically unrelated Brazilian leaf-litter frogs. Besides their cryptic

colouration, tonic immobility during this posture strongly enhances the

resemblance of these anurans to fallen leaves on the forest floor. This

behaviour seems to have evolved independently in different lineages as a

very effective defensive mechanism against visually oriented predators,

especially vertebrates.

INTRODUCTION

Death feigning, or tonic immobility or thanatosis, as an anti-predator behaviour arose

independently in several animal lineages. In neotropical frogs, death feigning was already

reported in species belonging to the families Bufonidae (HADDAD & SAZIMA, 1992; ABBADIÉ-

BisoGNo et al., 2001; RUSSELL, 2002; TOLEDO, 2004; Vaz-SiLva & FROTA, 2004), Cycloram-

phidae (SAZIMA, 1978), Dendrobatidae (VAZ-SiL va & FROTA, 2004), Hylidae (SAZIMA, 1974;

DUELLMAN & TRUEB, 1994; AZEVEDO-RAMOS, 1995; VRCIBRADIC & VAN SLUYS, 2000; GOMES

et al., 2002), Microhylidae (SAZIMA, 1978; VAZ-SILVA & FROTA, 2004) and Rhinodermatidae

(PouGH et al., 2004). Death feigning with stif- d posture, however, was described only for

three cryptically coloured leaf-litter frogs: the microhylid Stercocyclops parkeri (Wetistein,

1934), the cycloramphid Proceratophrys appendiculata (Günther, 1873) (SAZIMA, 1978) and

the leptodactylid Scyrhrophrys sawayae (Cochran, 1953) (GARCIA, 1999).

Source : MNHN, Paris

BERTOLUCI, BRASSALOTI, SAWAKUCHI, RIBEIRO & WOEHL 39

Here we describe death feigning with stiff-legged posture in the Brazilian tree toads

Dendrophryniscus brevipollicatus Jiménez de la Espada, 1871 and D. leucomystax Izecksohn,

1968 (Bufonidae) from three localities in the states of Säo Paulo and Santa Catarina,

southeastern and southern Brazil.

MATERIAL AND METHODS

Dendrophryniscus brevipollicatus is a small toad (snout-vent length [SVL] & 14.5 mm,

% 19.3 mm; HEYER et al., 1990) known from the coastal ranges of southeastern Brazil (HEYER

et al., 1990; IZECKSOHN & CARVALHO-E-SILVA, 2001). This species is easily diagnosed by its

small size, a pointed-mucronate dorsal snout shape, and reduced thumbs (HEYER et al., 1990).

According to HEYER et al. (1990), it is apparently active on the forest floor during the day,

resting on leaves above the ground at night. It is strongly associated to terrestrial bromeliads,

where their eggs are laid and tadpoles develop (CARVALHO, 1949; LuTz, 1954; PEIXOTO,

1995).

Dendrophryniscus leucomystax superficially differs from D. brevipollicatus by the pres-

ence of a white stripe extending from snout tip to base of arms, but several other differences

distinguish these two similar toadlets (IZECKSOHN, 1968). The species is known from the

lowland coastal forests of southeastern and southern Brazil (IZECKSOHN, 1968; WoEHL &

WoEnL., 2006). It is a small toad (SVL & 21.0 mm, ® 25.0 mm; IZECKSOHN, 1968) that is found

in secondary and primary forests on leaves near the ground, including terrestrial bromeliads,

and on the forest floor (IZECKSOHN, 1968; pers. obs.). Eggs are laid in temporary puddles on

the forest floor, where tadpoles develop and attain metamorphosis (IZECKSOHN & CRUZ,

1972).

One individual of each species from localities in the state of Säo Paulo was collected,

killed in a CO, artificial atmosphere, preserved in 10 % formalin and housed in the herpeto-

logical collection of the Laboratério de Zoologia de Vertebrados, Departamento de Ciéncias

Biolégicas, Escola Superior de Agricultura “Luiz de Queiroz”, Universidade de Säo Paulo,

Brazil (field numbers IC 012 and CB 002). Voucher specimens (colour transparencies) from

the state of Santa Catarina are housed in the herpetological collection of the Kansas

University, Lawrence, USA (KU-CT 11954-11957).

SruDY SITES

Observations were made in three sites belonging to the Atlantic Rainforest Morphocli-

matic Domain (AB'SABER, 1977), southern and southeastern Brazil, as follows.

(1) The Parque Estadual Carlos Botelho (PECB) is a 37793 ha reserve of well-preserved

Atlantic rainforest located in the municipality of Sete Barras, state of Säo Paulo, southeastern

Brazil (24°00°-24°1 47°45"-48°10°W). Altitudes vary from 30 to 1003 m (DomiIN &

SILVA, 1988). In the area where our observations were made, climate is Cfb of Koeppen, with

the mean temperature of the warmest month not superior to 22°C (SETZER, 1946).

Source : MNHN, Paris

40 ALYTES 25 (1-2)

(2) The Parque Estadual da Iha do Cardoso (PEIC) is a 22,500 ha island located in

municipality of Cananéia, state of Säo Paulo, southeastern Brazil (25°03°-25°18'S, 47°53°-

48°05°W). For the period 1990-1991, the minimum daily temperatures averaged 19°C and the

maximum daily temperatures averaged 27°C, and the annual rainfall varied between 1800 and

2000 mm (MELO & MANTOVANI, 1994). Altitudes vary from sea level to 800 m. Our observa-

tions were made almost at sea level in a well-preserved dense Restinga.

(3) The third site is a well-preserved dense Restinga fragment located in Quati, munici-

pality of Guaramirim, state of Santa Catarina, southern Brazil (26°26’S, 48°57°W). Obser-

vations were made at 10m above sea level. Mean annual rainfall between 2002 and 2005 was

1900 mm. Climate is subtropical, with mean annual air temperature around 20°C (ANONY-

MOUS, 1997).

RESULTS

DENDROPHRYNISCUS BREVIPOLLICATUS

On October 8‘, 2005, we observed death feigning in two individuals of D. brevipollicatus

in PECB. At 12.11 h, one individual (SVL 17.5 mm) defended itself stifing its legs in response

to hand capture. The same behaviour was observed in another individual (13.7 mm) at

17.05 h (fig. la). We then gently rolled the toadlet around its longitudinal axis, simulating a

bird beak with a finger, and observed that it remained immobile in its rigid posture (fig. 1b).

At the same site, on June 10°", 2006, four individuals displayed stifl-legged defensive

behaviour. At 12.10 h one individual (SVL 15.4 mm) after disturbation maintained the rigid

posture for almost five minutes. At 12.47 h one individual (16.8 mm) defended itself in

response to searcher moving. At 14.42 h another individual (16.5 mm) feigned death several

times following manipulation. At 21.10 h a female apparently bearing eggs (SVL 21.0 mm)

adopted the stiff-legged posture.

DENDROPHRYNISCUS LEUCOMYSTAX

On July 9, 2002, around 15.00 h, death-feigning was displayed by an adult D. leucomys-

tax (SVL 16 mm) in Guaramirim Restinga fragment. When the animal was put upside down

to be photographed, it remained in thanatosis posture (fig. 1e). On July 27", 2002, between

14.00 and 17.00 h, at the same site, another individual (SVL 14 mm) was found on the

leaf-litter and med the typical stifl-legged defensive posture when touched by the sear-

cher, remain mmobile for almost 1 minute (fig. Id). Snout-vent length for the individuals

Guveniles and ‘adult males and females) sampled in this fragment varied from 14 to 22 mm

{mean 17 + 2 mm; n = 29).

On September 4", 2005, death-feigning was displayed during the day (12.11 h) by an

adult D. leucomystax in a Restinga area of the PEIC. After hand capture for identification, the

Load was put upside down to take photographs of its ventral surface. It remained motionless

Source : MNHN, Paris

BERTOLUCI, BRASSALOTI, SAWAKUCHI, RIBEIRO & WOEHL 41

Fig. 1. - Death feigning in the Brazilian tree toads Dendrophryniscus brevipollicatus (a-b) and D.

leucomystax (c-e).

in a death feigning-like posture with eyes opened. We then caught the toad again and it leaped

and landed on a leaf of a terrestrial bromeliad with the legs stretched backwards, remaining

motionless and with eyes opened. The toad was photographed but not collected. On Novem-

ber 61", 2005, at the same site, death-feigning by another adult D. leucomystax was observed

during the day (17.39 h). After hand capture, the toad leaped and landed on a tree trunk,

where it assumed the stifl-legged posture (fig. le). Two additional observations of this

behaviour were made in 2006, on January 10° (09.42 h) and February 4 (15.33 h).

Source : MNHN, Paris

42 ALYTES 25 (1-2)

DISCUSSION

Besides their cryptic colouration, the stifl-legged posture adopted by these toadlets

greatly enhances their resemblance to fallen leaves both in colour and shape. As pointed out

by SAZIMA (1978) for Proceratophrys appendiculata and Stereocyclops parkeri, tonie immobil-

ity in these cases may cause an animal to be confused with a casually dislodged leaf. This is

suggested by the result of our simple experiment of rolling the animal around its axis. This

defensive strategy may protect the individual against visually oriented predators like birds,

snakes and mammals that actively search for their prey in the leaf-litter. The maintenance of

the rigid posture for rather long periods — almost five minutes in Dendrophryniscus brevipol-

licatus and up to 10 minutes in Scythrophrys sawayae (GARCIA, 1999) — suggests that this

behaviour was selected against visually oriented predators. Similar predatory pressures on the

forest floor possibly resulted in the convergent behavioural patterns observed in several

unrelated frogs (SAZIMA, 1978). The instinctiveness of this behaviour is suggested by its being

triggered and maintained in different species even in unexpected situations, like while sub-

merged (GARCIA, 1999) or vertically positioned on a tree trunk, or at night (this paper).

The related Dendrophryniscus minutus from the Amazonian region of South America

also displays death feigning, but in a very different way. When captured, this toad flips over on

its back and remains motionless in dorsal recumbency with the legs held upward. This posture

makes evident the bright orange path on its venter and its orange palms and soles, thus

consisting in a defensive behaviour associated to aposematism (RUSSEL, 2002). Venter, palms

and soles of D. brevipollicatus and D. leucomystax (fig. 1b-c) and of Scythrophrys sawayae

(GarciA, 1999) do not have such warning colours, so these species did not evolve an

aposematic behaviour similar to D. minutus. Thanatosis in dorsal recumbency, however,

seems to be a common (probably primitive) defense strategy, since it was displayed by almost

all individuals in different species and lineages that display stifl-legged posture.

RÉSUMÉ

Une attitude de mort feinte, avec les pattes postérieures raides, a été observ

plusieurs individus des Bufonidés brésiliens Dendrophryniscus brevipollicatus et D. leucomy

pour trois autres anoures brésiliens non directement apparentés. Associée à leur coloration

cryptique, cette immobilité tonique renforce la ressemblance de ces anoures avec des feuill

mortes tombées sur la litière forestière. Ce comportement semble avoir évolué d'une manière

indépendante dans des lignées différentes comme un mécanisme défensif très efficace

contre les prédateurs, notamment des Vertébrés qui effectuent une recherche visuelle des

proies.

Source : MNHN, Paris

BERTOLUCI, BRASSALOTI, SAWAKUCHI, RIBEIRO & WOEHL 43

ACKNOWLEDGEMENTS

We thank FAPESP for financial support (processes 04/15938-5 and 99/09635-0), Instituto Florestal

(SP) for permitting access 10 Parque Estadual da Ilha do Cardoso and to Parque Estadual Carlos

Botelho, and IBAMA for collection license (number 430/05).

LITERATURE CITED

ANonymous [Governo do Estado de Santa Catarina], 1997. - Diagnéstico geral das bacias hidrogräficas de

Santa Catarina. Florianépolis, Secretaria de Estado do Desenvolvimento Urbano e Meio

Ambiente.

ABBaDié-BisoGNo, K., Ouiver-Lôrez, L. & RaMirez-BAUTISTA, A., 2001. — Bufo occidentalis (Pine

Toad). Death feigning. Herp. Rev, 32: 247.

AwSamer, A. N. 1977. — Os dominios morfo-climäticos na América do Sul. Primeira aproximaçäo.

Geomorfologia, 52: 1-22.

Azevrpo-Ramos, C., 1995. - Defense behaviors of the Neotropical treefrog Hyla geographica (Aura,

hylidae). Rev. brasil. Biol., 55: 45-47.

CaRvaLHo, A. L., 1949. - Notas sôbre os häbitos de Dendrophryniscus brevipollicatus Espada (Amphibia,

Anura). Rer. brasil. Biol, 9: 223-227.

DoMiNGurs, E. N. & SILVA, D. À. 1988. - Geomorfologia do Parque Estadual de Carlos Botelho (SP).

Bol. téen. Inst. Flor., 42: 71-105.

DueLLMAN, W. E. & TRUE, L., 1994. - Biology of amphibians. Second edition. Baltimore, John Hopkins

University Press: i-xxi + 1-670.

GarciA, PC.A.; 1999. — Scpthrophrys sawayae. (NCN). Defensive behavior. Herp. Rev., 30:

ER. BeviER, C. R. & NAS, C. À. 2002. - Environmental and physiological facto!

antipredator behavior in Scinax hiemalis (Anura: Hylidae). Copeia, 2002: 994-1005.

Happap, C. F. B. & Sazima, LL, 1992. - Anfibios anuros da Serra do Japi. fr: L. P. C. MORELATO (ed.),

Histôria natural da Serra do Japi: ecologia e preservaçäo de uma drea florestal no sudeste do Brasil,

Campinas, Editora da UNICAMP/FAPESP: 188-211.

Hever, W.R., RaND, A. $., Cruz, C. À. G,, PEIXOTO, O. L. & NELSON, C. E., 1990. - Frogs of Boracéia.

influence

Arg 31-410.

IzECKSOHN, E., 1968. — Nova espécie de Dendrophryniscus do Estado do Rio de Janeiro (Amphibia,

Salientia). Rev. brasil. Biol., 28: 357-362.

IzEcksouN, E. & Cruz, C. A. G. 1972. - Notas sobre os girinos de Dendrophryniscus leucomystax

Izecksohn e D. brevipollicatus Espada (Amphibia, Anura, Bufonidae). Arg. Univ. fed. rural R. Jan.

2: 69.

IZFCKSONN, E. & CARVALHO-E-SILVA, S. P., 2001. — Anfibios do municipio do Rio de Janeiro. Rio de

Janeiro, Editora UFR 1-148.

Lurz, B., 1954. - Anfibios anuros do Distrito Federal. Mem. Inst. Osw. Cruz, 52: 15: 8.

MeLo, M. R. EF. & Manrovant, W., 1994. — Composiçäo floristica e estrutura fitossociolégica da mata

atlântica de encosta na [ha do Cardoso (Cananéia, Brasil). Bol. Inst. Bot., 9: 107-158.

Paxoro, O. L.. 1995. - Associaçäo de anuros à bromeliäceas na Mata Atlântica. Revista da Univ. rural,

17: 75-83.

PouGH, F. H., ANDREWS, R. M., CADLE, L E., CRUMP, M. L., SAVITZKY, A. H. & WELLs, K. D., 2004. —

Herpetology. Third edition. Upper Sade River, Prentice-Hall: i-ix + 1-726

M..J. 2002. - Dendrophryniscus minutus (Amazon Toadlet). Defensive behavior. Herp. R

302.

Sazima, L., 1974. — Experimental predation on the leaf-frog Phrllomedusa rohdei by the water snake

Liophis miliaris. J. Herp., 8: 376-377.

RUSS

Source : MNHN, Paris

44 ALYTES 25 (1-2)

—— 1978. - Convergent defensive behavior of two leaf-litter frogs of southeastern Brazil. Biotropica, 10:

158.

SETZER, J., 1946. - Contribuiçäo para o estudo do clima do Estado de Säo Paulo. Säo Paulo. Escolas

Profissionais Salesianas.

ToLepo, L.F., 2004. — Bufo cf. crucifer (Sapo Cururu). Defensive behavior. Herpetol. Rev., 35: 370-371.

VaAz-SiLva, W. & FROTA, J. G., 2004. — Bufo marinus (Marine Toad). Defensive behavior. Herpetol. Re.

35: 371.

VRCIBRADIC, D. & VAN SLUYS, M., 2000. — Hyla alvarengai (NCN). Death feigning and size at maturity.

Herpetol. Rev., 31: 40-41.

WoruL, G. JR, & WornL, E. N., 2006. - Dendrophryniscus leucomystax. Geographic distribution.

Herpetol. Rev., 37: 231.

Corresponding editors: Franco ANDREONE & Thierry LoDé.

Source : MNHN, Paris

Alytes, 2007, 25 (1-2): 45-54. 45

The tadpole of Eupsophus queulensis

(Anura, Cycloramphidae)

Diana R. CARDENAS-RoyAs*, Alberto VELOSO** & Rafael O. DE SÂA***

* Instituto de Zoologfa,

Universidad Austral de Chile, Casilla 567, Valdivia, Chile <diandrobates@gmail.com>

** Departamento de Ciencias Ecolôgicas, Facultad de Ciencias, Universidad de Chile,

Casilla 653, Santiago, Chile

*** Department of Biology, University of Richmond, Richmond, 23173 Virginia, USA

The genus Eupsophus is characterized by endotrophic larvae that

inhabit water-filled cavities. The larvae feed from large volk reserves during

metamorphosis. The external morphology has been described for four of the

ten known species and data on larval chondrocranial morphology are

available only for one. We describe the external morphology, oral disc and

chondrocranial anatomy of the larvae of Eupsophus queulensis. The

characteristics of E. queulensis tadpoles are compared with those of other

Eupsophus larvae.

INTRODUCTION

The genus of 10 recognized species, including the rece:

Eupsophus queulensis (VELOS0 et al., 200$). They are distributed throughout the temperate

forests of southern Chile and Argentina. The larvae of only four species of Eupsophus have

been described so far: E. roseus (FORMAS & PUGIN, 1978), E. emiliopugini (FORMAS, 1989),

E. calcaratus (FORMAS, 1989b; VERA CANDIOTI et al., 2005) and £. vertebralis (FORMAS, 1992).

These larvae have been classified as endotrophic-nidicolous (ALTIG & JOHNSTON, 1989). They

are characterized by having few denticles, scarce pigmentation, and internal organs visible

through the unpigmented ventral skin.

Herein, we describe and illustrate the external morphology of the tadpole of Eupsophus

queulensis and its chondrocranial anatomy, comparing the characteristics of this species with

those previously described for the genus.

MATERIALS AND METHOD

Alberto Veloso and Klaus Busse collected a clutch of eggs, along with a guarding adult

female, from a water-filled cavity at the type locality, Reserva Nacional Los Queules (35°5!

Source : MNHN, Paris

46 ALYTES 25 (1-2)

72°41°W), Maule Region, Chile, on November 2003. The female was included among the type

series in the description of the species (VELOSO et al., 2005). The eggs were brought and

hatched in the laboratory. Four larvae developed and they were used in this description.

Specimens were fixed in 95 ethanol and subsequently preserved in 10 % commercial grade

formalin. Tadpoles are in Gosner stage 34 (GOSNER, 1960).

Measurements, terminology and labial tooth row formula follow ALTIG & MCDIARMID

(1999). Measurements were taken using an eye micrometer under a Wild M3C Leica stereo-

microscope. The description of larval external morphology is based on the examination of

four larvae (Cäârdenas-Rojas Field Number DRCR 062). After examination of external

morphology, two larvae were prepared for scanning electron analysis. The other two

larvae were deposited at the herpetological collection of the Instituto de Zoologia Universi-

dad Austral de Chile (IZUA), Valdivia, Chile. Tadpole illustration is based on individual

IZUA 3708. The other specimen was used for analysis and description of chondrocranial

anatomy (IZUA 3709); the specimen was cleared and double-stained for bone and cartilage

using a modified technique of DINGERKUS & UHLER (1977). Chondrocranial terminology

follows Haas (1995, 2003) and LARsON & DE SÀ (1998). Illustrations of larval morphology

and chondrocranial anatomy were made using a Wild M3C Leica stereomicroscope with a

camera lucida attachment.

RESULTS

DESCRIPTION OF TADPOLE

Tadpoles of Eupsophus queulensis have an elliptical body (fig. 1). The snout is rounded in

dorsal and lateral views; the dorsal contour of the body gradually curved from mid-body to

snout. The nostrils are situated midway between the tip of snout and the anterior border of

eyes. The eyes are large, they are laterally positioned and directed anterodorsally; they are

separated by a distance 1.9 X the eye diameter. The spiracular tube is absent; the spiracular

opening is small, lateral and sinistral. The length of the tail is nearly twice the body length, the

dorsal and ventral fins are well developed and almost parallel to the tail musculature; the

dorsal fin originates on the posterior third of the body and the ventral fin originates posterior

to the vent tube. The caudal musculature is moderately robust with poorly defined myotomes;

the tail tipis rounded. The vent tube and opening are medial. The color of preserved specimens

is yellowish white whereas the abdomen and fins are transparent and the internal organs are

visible. The dorsal surface area is light brownish with scarce and minute melanophores.

The oral disc is positioned ventrally, it is not emarginated and it p a single row of

marginal papillae with a wide rostral gap (about 1/3 of the anterior labium); submarginal

papillae are absent (fig. 2). The labial tooth row formula is 2(2)/2, the posterior labial tooth

rows are about equal in length. The upper and lower jaw sheaths are pigmented for about

one-third of their width and have serrated edges.

Source : MNHN, Paris

CARDENAS-ROIJAS, VELOSO & O. DE SÂ 47

Fig. 1. - Lateral view of Eupsophus queulensis, IZUA 3708, Gosner stage 34 tadpole. Bar: 5.0 mm.

int HERNEE NL

ALES

RNA A PR RAT

Fig. 2. - Oral disc of Eupsophus queulensis, IZUA 3708, Gosner Stage 34 tadpole. Bar: 0.5 mm

Source : MNHN, Paris

48 ALYTES 25 (1-2)

— Suprarostral cartilages of Eupsophus queulensis, IZUA 3709, Gosner stage 34 tadpole. Bar:

0.5 mm.

Measurements in millimeters of tadpole IZUA 3708, following ALTIG & MCDIARMID'S

(1999) methodology, are as follows: body length 6.3; body height 3.7; body width 4.2; tail

length 12.0; maximum tail height 4.5; tail muscle height 2.2; tail muscle width 1.9: interorbital

distance 2.9; internarial distance 1.6; oral disc length 1.9.

CHONDROCRANIUM

The paired suprarostral cartilages support the upper horny beak. Each suprarostral

consists of a central body and a lateral wing. The bodies of the suprarostrals are vertically

rectangular and they are largely fused to the wings (fig. 3). The bodies of the suprarostrals are

connected ventromedially via a narrow cartilage. The lateral wings of the suprarostrals are

triangular-shaped; the width of each suprarostral’s wing is about twice that of the body: the

wings have a short but clearly visible processus (p.) posterior dorsalis. The suprarostral bodies

and the medial margins of the suprarostrals’ wings articulate with the tip of the trabecular

horns.

The cornua trabeculae are short, less than 20 % of the total chondrocranial length. They

project anteriorly from the planum trabeculare anticum, diverging from each other and

forming a wide and shallow “U”. The cornua trabeculae are almost uniform in width

throughout their length and they curve strongly ventrally to articulate with the suprarostrals

(fig. 4A, C). The p. lateralis trabeculae is poorly developed and almost indistinguishable: it

connects laterally the cornua trabeculae With the p. quadratoethmoidalis via the ligamentum

quadratoethmoidale. The posterior confluence of the cornua trabeculae is continuous with the

planum trabeculare anticum and the developing planum ethmoidale, the later begins to

form the anterior wall of the braincase at this stage. The rectum nasi has not yet begun to

develop.

Atstage 34, the cranial floor is not yet fully chondrified, a thin cartilage closes the fénestra

basicranialis. Two sets of openings are found on the cranial floor. The anterior pair, the

all and almost not visible, whereas the posterior pair, the

s well defined (fig. 4B).

foramina craniopalatina,

foramina carotica primaria,

The cartilago orbitalis forms the lateral walls of the braincase. However, these cartilages

do not connect to the otic capsule yet: consequently, the dorsal and anterior margins of the

Source : MNHN, Paris

CARDENAS-ROJAS, VELOSO & ©. DE SÂ 49

Fig. 4, — Chondrocranium and hyobranchial apparatus of Eupsophus queulensis, IZUA 3709, Gosner

stage 34 tadpole. (A) dorsal, (B) ventral, and (C) lateral views; (D) ventral view of hyobranchial

apparatus. Bar: 1.0 mm. Cb I-IV, ceratobranchial cartilages 1-IV: ch, ceratohyal: cqa, commissura

quadratocranialis anterior: et, cornua trabeculae: ets, commissurae terminales: fah, facies articularis

hyalis: ®b, fenestra basicranialis: Fep, foramen caroticum primarium: 3, foramen jugulare: fo, foramen

opticumr: loc, foramen oculomotoriunr: fov, fenestra ovalis: fpi, foramen perilymphaticum inferior: hp,

hypobranchial plate: hy, hyoquadrate process: i, infrarostral cartilage; m, Meckel’s cartilage: oc,

otic capsule: pa, processus ascendens; pab, processus anterior branchialis: pah, processus anterior

hyalis: pal, processus anterolateralis hyalis: pag, pars articularis quadrati: pm, processus muscularis:

ppd, processus posterior dorsalis: pph, processus posterior hyalis: pr, pars reuniens: pu, processus

urobranchialis; sa, Suprarostral ala:; se, suprarostral corp picula; so, subocular bar of palato-

quadrate: ts, tecum synoticum: tm, taenia tecti marginalis

Source : MNHN, Paris

50 ALYTES 25 (1-2)

foramen (f.) prooticum are open (fig. 4C). The f opticum and the f. oculomotorium perforate

the cartilago orbitalis, the later being the smaller of the two and found posterior and ventrally

to the former.

The fenestra frontoparietalis is not yet complete. Laterally, we found very thin raeniae

tecti marginales, and posteriorly the rectum sinoticum is found as a narrow cartilage connec-

ting the otic capsules (fig. 4A).

The otic capsulae are about 33 % of the total chondrocranial length and 38 % of the total

chondrocranial width. The ventrolateral surface of the otic capsules bears a large fenestra

ovalis. À larval crista parotica and p. oticum are absent. The arcus occipitalis extends ventrally

from the posteromedial margin of the otic capsule, fusing with the planum basale and giving

rise to the occipital condyles. The f perilymphaticum inferior is found lateral to the f jugulare

on the ventromedial margin of the otic capsule (fig. 4B).

The cartilago palatoquadrati has a constant width throughout its length. It has two

attachments to the braincase, anteriorly the commissura quadratocranialis anterior and pos-

teriorly the p. ascendens. The commissura quadratocranialis anterior extends between the

cartilago palatoquadrati, at a level just posterior to the pars articularis quadrati, and the floor

of the neurocranium (fig. 4A-B). The anterior margin of the commissura quadratocranialis

anterior bears a triangular p. quadratoethmoïidalis, which serves as the point of attachment for

the ligamentum quadratoethmoidale. À very short, almost indistinct, and blunt p. antorbitalis

is present. The posterior curvature of the cartilago palatoquadrati extends slightly beyond the

level of attachment of the p. ascendens to the braincase. The lateral and posterior margins of

the palatoquadrate curve slightly dorsally, giving the palatoquadrate a concave appearance in

dorsal view. The p. ascendens is a narrow and rod-like cartilage connecting the posteromedial

margin of the cartilago palatoquadrati to the cartilago orbitalis. The p. ascendens attaches just

posterior to the oculomotor foramen, i.e., SoKOL’s (1981) intermediate condition. The

p. ascendens is almost perpendicular (85-90 angle) to the main axis of the chondrocranium

(fig. 4A-C).

Anteriorly, the cartilago palatoquadrati has two processes, the p. muscularis quadrati and

the pars articularis quadrati. The p. muscularis is broad, flat, and extends dorsally from the

lateral margin of the cartilago palatoquadrati posterior to the pars articularis quadrati. The p.

muscularis is visible in lateral view. The dorsal edge of the p. muscularis inclines medially; a

sura quadratoorbitalis is absent. Immediately below the p. muscularis and ventrally on

the cartilago palatoquadrati there is a notch, the facies articularis h) which serves as the

point of articulation of the ceratohyal with the cartilago palatoquadrati. On the articular

surface is the hyoquadrate process which in lateral view is a small and sub-triangular process.

The pars articularis quadrati is slightly angled medially and articulates broadly with the

cartilago Meckeli.

Meckel's cartilages, together with the infrarostral cartilages, form the mandible during

larval stages. Meckel's cartilage is stout and has a sigmoid-shape: its anteromedial margin is

concave wher( s anterolateral margin is convex. Laterally, the cartilago Meckeli articulates

with the pars articularis quadrati via the p. retroarticularis. This process curves ventrally

beneath the pars articularis quadrati. Anteromedially, the cartilago Meckeli has two processes,

the p. dorsomedialis and the p. ventromedialis. The posterodorsal portion of the infrarostral

cartilages articulates with Meckel’s cartilages between these two proce:

es. The infrarostral

Source : MNHN, Paris

CARDENAS-ROIJAS, VELOSO & O. DE SÂ 51

cartilages provide support for the lower keratinized beak. Each infrarostral is rectangular in

shape and outwardly rounded over their entire anterior margin.

HYOBRANCHIAL APPARATUS

The ceratohyalia are medially wide and flat cartilages; laterally, they are twisted dorsally

to articulate with the cartilago palatoquadrati at the facies articularis hyalis. Each ceratohyal

bears two processes on the anterior margin, a triangular p. anterior hyalis and a rounded and

small p. anterolateralis hyalis. The ceratohyalia also have a well-developed p. posterior hyalis.

Medially, the ceratohyalia are connected by a pars reuniens, which is continuous with the

copula posterior. Posteroventrally, the copula posterior has a small p. urobranchialis. The

copula posterior is continuous with the hypobranchial plates. The hypobranchial plates do not

contact each other medially; they are continuous with the ceratobranchials (Cb). Posteriorly,

the hypobranchial plates diverge and their posterior edges form an inverted V (fig. 4D).

The branchial baskets consist of four ceratobranchials (Cb 1-Cb IV). The p. branchialis is

absent. The ceratobranchials are distally continuous via the commissurae terminales, except

between Cb I and Cb IT. Proximally, ceratobranchial I is attached to the hypobranchial plate

by a wide strip of cartilage; Cb I bears a wide and flattened p. anterior branchialis on its

anterior margin. Ceratobranchials IE, III and IV are fused to the hypobranchial plate, the

former two via a narrow bar of cartilage and Cb IV via a wider cartilaginous connection. Only

ceratobranchial I and II bear poorly developed spiculae, near their point of attachment to the

hypobranchial plate.

DISCUSSION

The tadpoles of Eupsophus queulensis are very similar in morphology and color patterns

to those of E. roseus, E. emiliopugini, E. calcaratus and E. vertebralis (FORMAS & PUGIN, 1978;

FoRMAS, 1989a-b, 1992; VERA CANDIOTI et al., 2005). They have elliptical body, approximately

two times longer than higher; their tail length is almost twice the body length. Eyes are

positioned dorsolaterally. The creamy white color and scant pigmentation are common traits

of endotroph nidicolous tadpoles (THIBAUDEAU & ALTIG, 1999).

Eupsophus tadpoles differ in a few characteristics. The spiracular tubes of Eupsophus

emiliopugini and E. vertebrali: ï and ventrolaterally located (FORMAS, 19894,

1992). Eupsophus queulensis, seus (FORMAS & PUGiN, 1978) and E. calcaratus (FORMAS,

1989b) share a small and sinistral spiracular opening without spiracular tube. A single,

sinistral, and ventral spiracular tube with a visible opening has been previously reported for

E. calcaratus (VERA CANDIOTT et al., 2005). Tadpoles of £. calcaratus (EZUA 2896-2897, 2957)

examined herein do not have spiracular tube. The lack of a spiracular tube is a modification

for tadpoles with nidicolous life history (THIBAUDEAU & ALTIG, 1999).

The oral dise of £. queulensis and all described Eupsophus larvae does not differ from that

of typical exotrophic larvae. It is ventral and has à single row of marginal papillae with a

Source : MNHN, Paris

52 ALYTES 25 (1-2)

distinct rostral gap. The reported labial tooth row formulae range from 2/2 in E. roseus

(FORMAS & PUGIN, 1978), E. emiliopugini (FORMAS, 1989a) and E. calcaratus (FORMAS, 1989b)

to 2(2)/2 in E. queulensis (this work) and 2(2/2(1) in E. vertebralis (FORMAS, 1992). The

presence of a vestigial third posterior row in one individual and the following intraspecific

variation was reported for larvae of £. calcaratus: 2(2)/2, 2(2)/2(1) (VERA CANDIOTI et al.,

2005).

Only the chondrocranium of Eupsophus calcaratus had been described so far (VERA

Canpiori et al., 2005). Herein we compare the main differences between Æ. queulensis

(stage 34) and E. calcaratus (stage 31); characteristics for E. calcaratus are given in parenthe-

sis. Eupsophus queulensis has: body and wings of the suprarostrals largely fused (joined by a

cartilaginous dorsal bridge); the trabecular horns are of uniform width throughout their

length (anteriorly narrow); the trochlear foramen is absent (present); the p. ascendens of the

palatoquadrate has an intermediate attachment to the braincase (low attachment); the

infrarostrals are connected medially through connective tissue (infrarostrals fused into a

single element); commissurae terminales are absent between Cb I and Cb II (present); and

spiculae absent in Cb IT and Cb IV (present in Cb II).

Based on adult morphology, karyotypes, call characteristics, and a recent molecular

phylogeny, the ten species currently included in the genus Eupsophus have been separated into

two species groups, the roseus and the vertebralis groups (FORMAS 1980, 1991, 1993; FORMAS

et al., 1983, 1992; FERNANDEZ DE LA REGUERA, 1987; PENNA & VELOSO, 1990; ORTIz &

IBARRA-VIDAL, 1992; PENNA & Sois, 1995; NÜREZ et al., 1999; NÜREZ, 2003). The available

chondrocranial data are restricted to two species of the roseus group. Given the lack of

baseline tadpole data, it is not yet possible to determine if larval characteristics (external and

internal) can provide additional support to differentiate the two groups.

RESÜMEN

El género Eupsophus se caracteriza por poseer larvas endotréficas que habitan cavidades

con agua. Durante la metamorfosis las larvas se sustentan de sus grandes reservas vitelinas. La

morfologia externa ha sido descripta para cuatro de las diez especies conocidas en el género:

caracteristicas de la anatomia condrocraneal fueron reportadas sélo para una especie. En este

trabajo se describen la morfologia larval externa, el disco oral larval y la anatomia condro-

craneal de Eupsophus queulensis. Las cteristicas de la larva de esta especie se comparan

con las de otras as en el género Eupsophus.

ACKNOWLEDGEMENTS

This work was done with the support of Project D2004-01 and travel grant MECESUP

UCO214 and AUSOI 11, Direcciôn de Postgrado and Direcciôn de Investigaciôn y Desarrollo

Universidad Austral de Chile to DRCR. The research was partially funded through NSF

Source : MNHN, Paris

CARDENAS-ROJAS, VELOSO & O. DE SÂ 53

Award 0342918 to RdS and WRH. We thank two positive and anonymous reviewers from the

Journal of Herpetology for their suggestions; that journal rejected the manuscript for being a

“simple description”.

LITERATURE CITED

ALTIG, R. & JOHNSTON, G. F., 1989. - Guilds of anuran larvae: relationships among development modes,

morphologies and habitats. Herpetological Monographs, 3: 81-109

AuriG, R. & MCDiarMID, R. W., 1999. - Body plan: development and morphology. fr: R. W. McDraR-

MD & R. ALTIG (ed.), Tadpoles, Chicago & London, University of Chicago Press: 24-51.

DinGerkus, G. & UHLer, L., 1977. - Enzyme clearing of alcian blue stained whole small vertebrates for

demonstration of cartilage. Stain Technology, 52: 229-232.

FERNANDEZ DE LA REGUERA, P. A., 1987. — Identifying species in the Chilean frogs by principal

components analysis. Herpetologica, 43: 173-177.

ForMas, J. R., 1980. - The chromosomes of Æ. calcaratus and the karyological evolution of the genus

Eupsophus (Anura: Leptodactylidac). Experientia, 36: 1163-1164.

—— 19894. — A new species of Eupsophus (Amphibia: Anura: Leptodactylidae) from southern Chile.

Proceedings of the biological Society of Washington, 102: 568-576.

1989. — The tadpole of Eupsophus calcaratus in southern Chile. Journal of Herpetology, 23: 195-197.

1991. — The karyotypes of the Chilean frogs Eupsophus emiliopugini and E. vertebralis (Amphibia:

Anura: Leptodactylidae). Proceedings of the biological Sociery of Washington, 104: 7-11.

1992, — The tadpole of Eupsophus vertebralis (Anura: Leptodactylidue). Herpetologica, 48: 115-119.

1993. — Allozymic and morphological differentiation between two South American frogs, genus

Eupsophus (E. vertebralis and E. emiliopugini). Comparative Biochemistry & Physiology, (B), 106:

FoRMaS, J. R., LACRAMPE, S. & BRIEVA, L., 1992. — Allozymic and morphological differentiation among

three South American frogs, genus Eupsophus (E. roseus, E. insularis and E. contulmoensis).

Comparative Biochemistry & Physiology, (B), 102: 57-60.

FoRMAS, JL. R. & PUGIN, E., 1978. - Tadpoles of Eupsophus roseus and Bufo variegatus (Amphibia, Anura)

in southern Chile. Journal of Herpetology, 12: 243-246.

FORMAS, J. R., Vera, IL. & LACRAMPE, S., 1983. — Allozymic and morphological differentiation in the

South American frogs, genus Eupsophus. Comparative Biochemistry & Physiology, (B), 75: 475-478.

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

identification. Herpetologica, 16: 183-190.

Haas, A., 1995. - Cranial features of dendrobatid larvae (Amphibia: Anura: Dendrobatidae). Journal of

Morphology, 224: 241-264.

Phylogeny of frogs as inferred from primarily larval characters (Amphibia: Anura).

sties, 19: 23-89.

* P. M. & DE SA. R. O., 1998. - Chondrocranial morphology of Leptodactylus larvae (Leptodac-

tylidae: Leptodactylinae): its utility in phylogenetic reconstruction. Journal of Morphology, 238:

287-30:

Rez, J. JL. 2003. — Taxonomia y sistemätica de las ranas del género Eupsophus (Leptodactylidae).

Unplub. PhD Thesis, Universidad Austral de Chile, Valdivia, Chile.

NUREZ, J. J., ZARRAGA, A. M. & FORMAS, J. R., 1999. - New molecular and morphometric evidence for

the validation of Eupsophus calcaratus and E. roseus (Anura: Leptodactylidae). Studies on Neo-

tropical Fauna and Environment, 4: 15

Ortiz, J. C. & IBaRRA-VIDAL, H., 1992. - Una nueva especie de Leptodactylidae (Eupsophus) de la

cordillera de Nahuelbuta (Chile). Acta =oolôgica lilloana, 41: 75-79.

PENNA, M. & SoLis, R.. 1995. - Influence of burrow acoustics on sound reception by frogs Eupsophus

(Leptodactylidae). Animal Behaviour, 54: 1-9.

PENNA, M. & VELOs0, A., 1990, - Vocal diversity in frogs of the South American temperate forest. Journal

of Herperology, 24: 23-33.

NÜ

Source : MNHN, Paris

54 ALYTES 25 (1-2)

Sokoz, O. M., 1981. - The larval chondrocranium of Pelodytes punctatus, with a review of tadpole

chondrocrania. Journal of Morphology, 169: 161-183.

THIBAUDEAU, G. & ALTIG, R., 1999. - Endothrophic anurans: development and evolution. /n: R. W.

McDiarmiD & R. ALTIG (ed.), Tadpoles, Chicago & London, University of Chicago Press:

170-188.

VeLoso, A., Ceuiz-Diez, J. L., GUERRERO, P. C, MENDEZ, M. A., ITURRA, P. & SIMONETTI, J., 2005. —

Description of a new Æupsophus species (Amphibia, Leptodactylidae) from the remnants of

Maulino forest, central Chile. Herpetological Journal, 15: 159-165.

Vera Canniori, M. F., UBepa, C. & LaviLca, E. O., 2005. - Morphology and metamorphosis of

Eupsophus calcaratus tadpoles (Anura: Leptodactylidae). Journal of Morphology, 262: 161-177.

Corresponding editor: Alain DuBois.

Source : MNHN, Paris

Alytes, 2007, 25 (1-2): 55-74. 55

New synonyms in specific names

of frogs (Raninae)

from the border regions between China,

Laos and Vietnam'

Annemarie OHLER

Vertébrés: Reptiles et Amphibiens, USM 0602 Taxonomie & Collections,

Département Systématique & Evolution, Muséum national d'Histoire naturelle,

25 rue Cuvier, CP 30, 75005 Paris, France

<ohler@mnhn.fr>

Species pairs described on either side of the Chinese/Vietnamese-Lao

border are compared and the following new synonyms are proposed: Rana

daorumis a subjective junior synonym of Amolops mengyangensis and the

valid name of the species is Amolops mengvangensis; Rana hmongorum

is a subjective junior synonym of Rana jingdongensis and the valid name of

the species is Odorrana jingdongensis; Rana megatympanum and Rana

heatwolei are subjective junior synonyms of Rana tiannanensis and the

valid name of the species is Odorrana tiannanensis; Rana nigrolineata is

a subjective junior synonym of Rana lateralis and the valid name of the

species is Pelophylax lateralis; Rana nigrotympanica is a subjective

junior synonym of Rana cubitalis and the valid name of the species is

Svlvirana cubitalis; Rana bannanica is a subjective junior synonym of

Rana milleti and the valid name of the species is Sylvirana milleti.

Evidence from morphological and morphometrical study of type material

and topotypical material is given. Attention is drawn to the fact that

taxonomy is an international endeavour, and that describing new taxa often

requires knowing the fauna of neighbouring countries.

INTRODUCTION

Preliminary to any systematic work we need to define the samples we have to study. Some

groups may be of small size, so there is no major problem to compare all their members.

Others may count numerous species and taxonomic delimitation may be very poor. In some

cases also allocation of an observed taxon to a particular group will be difficult, and in the end

no relevant comparison will be done

1. This paper is issued from à communication entitled “Transborder systematics: a critical evaluation of

xphy and conservation biology at the crossroads between Vietnam, Laos and China”, presented

during the 13° Ordinary General Meeting of the Societas Europaca Herpetologica (SEH) in Bonn

(Germany) on 30 September 2005.

Source : MNHN, Paris

56 ALYTES 25 (1-2)

Another fact is that in recent years many new species have been described and the

discovery of many more is expected (KÜHLER et al., 2005), which suggests to the taxonomist

that what he has in his hands could be probably new. And it can, but sometimes it won't.

Thirty percent of the amphibian species are poorly known: only the original description and

a few old records are available and no proper figure or photograph have been published.

Many authors focus on the southern limit of Oriental region, known as the Wallace line

(OosrERZEE, 1997), but the north-eastern limit of this region is still more problematic and

quite unique among realms, Over a distance of several thousands kilometers, no geographic

limits, such as deserts or high mountains, exist between the Palearctic realm and the Oriental

realm, so that an intergradation zone of roughly 1500 km has developed. In fact in a large

zone, including central and southern China, northern Vietnam and small parts of northern

Laos, north-eastern India, northern Thailand and northern Myanmar, members of the two

faunas co-exist in the same habitats (Æ/yla sensu stricto, Rana sensu stricto are found along

with Xenophrys, Odorrana, etc.), and members of both faunas might have their major centre

of diversification in this region (e.g., in Hyla and Xenophr.

The border between China on one hand, and Laos and Vietnam on the other, has been

used as a limit for biogeographical investigations (INGER, 1999; ZHAO, 1999), although it

does not correspond to any natural limit. In many studies by Chinese authors, insufficient

comparison with the Indochinese species has been done due to work in isolated conditions, so

that synonyms have been created, some of which still need to be ified. In recent times,

taxonomic works have been carried out in Vietnam without comparison to the Chinese fauna

(BAIN et al., 2003), thus again creating new names for already named species.

Another problem of the work of BAIN et al. (2003) is that, although they provided

comparisons to many of the “stream frogs” from the Oriental region, they did not refer to the

voucher specimens they studied, nor gave the geographical origin of these specimens. In

consequence, verification of specific allocation of these species cannot be carried out and

distinctive characters might be useless or uninformative in some c: below for Rana

hmongorum). Use of such specimens for further analyses, such as phylogenetic hypotheses and

classificatory proposals based on them (FRosr et al., 2006), introduces important doubts

about the results obtained.

The aim of this study is to confront information in some pairs of nominal species of the

Oriental region in order to understand taxonomic allocation of specimens to biological

species and to have additional data on their distribution. In many cases this provides more

data on biology and conservation of the species, as often one of the two species is better

known. In this first clarification (to be followed by others), pairs of names available for six

ranine frogs of the genera Amolops, Odorrana, Pelophylax and Sylrirana ill be discussed.

MAT

IAL AND METHODS

MATERIAL

See appendix 1.

Source : MNHN, Paris

OHLER 57

CLASSIFICATION

FRosr et al. (2006) published an important and necessary work in presenting a tree of all

amphibians. Nevertheless the sampling is not equally distributed and the Oriental ranid frogs

are particularly poorly represented as compared to the diversity of lineages. In conclusion,

FRosr et al. (2006) decided to lump large groups of frogs (e.g., Huia Yang, 1991) but kept

other lineages apart without any evidence beside original definition (e.g., Pterorana Kiyasetuo

& Khare, 1986). For this reason, I here still keep the groups proposed as subgenera by DUBoIs

(1992), but use them as genera. Odorrana Fei, Ye & Huang, 1990 (type species by original

designation Rana margaretae Liu, 1950) is applied here to a group of frogs with its centre in

China and northern Vietnam and northern Laos. These frogs are stream-living species with

long legs, but shorter than in Amolops Cope, 1865. In particular they can be distinguished by

the shape of their toe pad grooves which always show a little terminal gap in Odorrana (OHLER

& Duois, 1989). Their tadpoles have neither abdominal suckers nor glands on head and

back; they show 4 rows of keratodonts on lower lip. Huia as defined by YANG (19914) is a

group of frogs with tadpoles bearing abdominal suckers and with adults having longer legs

than those of Amolops. It was originally erected for species from Borneo, Java, Thailand and

Yunnan. YANG (1991b) added Rana nasica Boulenger, 1903 to this genus on the basis of

tadpoles allocated to this species, but identity of these tadpoles was supported by morpholog-

ical evidence only. Thus the taxonomic allocation of this species must be studied again on

properly identified specimens and Rana nasica cannot be used as voucher for Huia as done in

FRosr et al. (2006). Allocation of recently described frogs to Huia must be re-evaluated in

further works as it was done on adult morphology only, and as taxonomic conclusions were

not based on type species but on species that might be phylogenetically closer to other clades

than to that containing Rana cavitympanum Boulenger, 1893 (type species of Huia Yang, 1991

by original designation).

METHODS

For most of the name-bearing types of the Oriental region, detailed description and

photographs of the specimens are available at MNHN. These descriptions will be published

elsewhere, in particular in a revision of Sylvirana (OHLER & DUBOïS, in preparation). For

recent names, good original descriptions are available. Morphological comparison of type

material, topotypical material and specimens from various localities in the Oriental region

was carried out based on voucher specimens or descriptions. Morphometric measurements of

basic characters such as snout-vent length (SVL), head width (HW), head length (HL), tibia

length (TL) and tympanum diameter (TY D) were made if necessary and are compared with

published measurements. The following terms are used in the synonymies below (DUBOIS,

2006): Aypodigm (SIMPSON, 1940) for a set of specimens used by an author to establish and

describe a new sp onomatophore (SIMPSON, 1940) for the name-bearing type of a taxon:

onymotope (DuBois, 20054-b) for the type locality of a species.

Contrary to the methods available to point to differences, there is no methodol

for the definition of synonyms. À synonym clearly cannot be defined as “identical”, as

within a biological species different kinds of morphological variation (sexual dimorphism,

Source : MNHN, Paris

58 ALYTES 25 (1-2)

ontogenetical variation, life stages) occur, caused either by genetic of by epigenetic factors, or

both. To establish synonymy, we need information on the intraspecific variation in order to

show that the synonymous taxon falls within the range defined for the other taxon. But often

the data available on various species are very poor (about 30 % of “data deficient” species

in the Global Amphibian Assessment; STUART et al., 2004). Also as in an example below,

males of one taxon and females of another taxon might be available and thus no direct

comparison possible. In such a case, sexual dimorphism must be evaluated by comparison

with phylogenetically close taxa.

A very particular problem in anurans is intraspecific variation in size. DuBois (1976)

proposed a ratio of extreme values (RE) which divides the size of the largest known specimen

of a species by the smallest known of this species. If applied to species recognized on

biological criteria as the treefrogs of the genus Hyperolius Rapp, 1842 described by SCHIOTZ

(1999), RE of adult male frogs varies from 1.04 to 1.56. In Afrixalus Laurent, 1944 the value

of RE is between 1.11 and 1.42. A value of 1.5 means that the smallest male has a SVL of two

thirds of that of the largest known male. The sole presence of size variation cannot be used as

a taxonomic argument. It may be due to variation in presence of age classes in a particular

population or to variation in external factors such as predation or harvesting. On the other

hand, size, independent of its origin, is linked to allometric variation in different body parts

such as tibia length, head shape, etc., and might lead to differences in relative size of various

body parts. If the sample is large enough, detailed analysis on growth can be made especially

using comparison of slopes of allometric growth curves. But on small samples, which are

often the samples we have to deal with in taxonomy, systematic studies can have only poor

arguments. Nowadays, size variation alone is not considered a sufficient argument for defining

a species. Like any other character, only its correlation to other characters will be considered

as sufficient support to indicate genetic isolation of a group of frog to others such groups, and

its meaning will be different in situations of sympatry, parapatry or allopatry.

ABBREVIATIONS

Collections

BMNH.- The Natural History Museum, London, United Kingdom.

CB. - Chengdu Institute of Biology, Chengdu, China.

FMNH. - Field Museum of Natural History, Chicago, USA.

KIZ. - Kunming Institute of Zoology, Kunming, China

MAS. - Malcolm A. Smith collection.

MNHN, - Muséum national d'Histoire naturelle, Paris, France.

MSNG. - Museo di Storia Naturale di Genova, Genova, Htaly.

ROM. - Royal Ontario Museum, Toronto, Canada

ZSI. - Zoological Survey of India, Kolkota, India.

Abbreviations used in text and tables

8. Adult male.

2. Adult female

alt. - Altitude above sea level (in metres).

n.— Sample size.

Source : MNHN, Paris

OHLER 59

HL. - Head length.

HW. - Head width.

SVL. - Snout-vent length.

TL. - Tibia length.

TYD.- Tympanum diameter.

TAXONOMIC DISCUSSIONS

Amolops mengyangensis Wu & Tian, 1995

Amolops mengyangensis Wu & Tian, 1995: 50. - HypopiGM: 3 4. — ONOMATOPHORE: Holotype,

by original designation, CIB 579034, 4, SVL 38.7 mm (personal observation). — ONYMOTOPE:

Mengyang (22°04N, 100°53E; alt. 680 m), Xishuanbanna Daizu Zizhizhou, Yunnan Sheng,

China.

Rana daorum Bain, Lathrop, Murphy, Orlov & Ho, 2003: 38. - HyPobiGM: 48 d, 10 ©, 1 subadult ©.

OxoMaroPHoRE: Holotype, by original designation, ROM 26381, ?, SVL 55.7 mm. - ONYMOTOPE:

Approximately 5 km NW of Sa Pa Village, near O Qui Ho Pass (22°20°09"N, 103°50°14"E; alt.

1400 m), Lao Cai Province, Vietnam. — SraTus: New synonym.

Amolops chunganensis (Pope, 1929) was reported to occur in southern Yunnan (Meng-

yang) by YANG (19914). In 1995, Wu & TAN described a new species, Amolops mengyang-

ensis, from specimens previously reported as Amolops chunganensis. OHLER et al. (2000)

reported the presence of Amolops chunganensis in northern Vietnam. These specimens

(MNHN 1999.5799-5813) are the same biological species as those described as Rana daorum

by BAIN et al. (2003). Comparison of the hypodigm of Amolops mengyangensis (fig. la) with

topotypes of Rana daorum (fig. 2a), as well as re-evaluation of the characters indicated by

BAIN et al. (2003), in particular absence of vomerine teeth, small tympanum/eye ratio and

morphometric data (tab. 1), support the synonymy of the two nominal species.

These frogs can be assumed to be a different taxon from chunganensis as proposed by Wu

& Tia (1995) and Bain et al. (2003), but should be placed in the genus Amolops, because of

the presence of a continuous fold on the ventrolateral border of their toe pads and their long

legs, similar to the species Amolops monticola (Anderson, 1871) and Amolops chunganensis.

These species of Amolops have dorsolateral folds, white upper lip and toe pads that are

rounded but not transversally enlarged as in Amolops formosus (Günther, 1876) or allied

species, nor pointed as in Odorrana (OHLER & Dumois, 1989).

Amolops chunganensis was described from Fujian and has been reported from south-

eastern and western China (FE1, 1999). It should be removed from the species lists of

amphibians of Vietnam and Yunnan as the vouchers are specifically identical with Amolops

mengyangensis.

Source : MNHN, Paris

60 ALYTES 25 (1-2)

Fig. 1. — (a) Amolops mengyangensis Wu & Tian, 1995. Holotype, CIB 579034, 4, SVL 38.4 mm;

Mengyang, Yunnan, China, in dorsal view. (b) Odorrana jingdongensis Fei, Ye & Li

2001.

Topotype of Rana hmongorum Bain, Lathrop, Murphy, Orlov & Ho, 2003, MNHN 1999.5772, 6,

SVL 64.0 mm; Sa Pa, Vietnam, specimen in life in dorsal view.

Table 1. - Measurements of Amolops mengyangensis Wu & Tian, 1995, including type

specimens of Amolops mengvangensis Wu & Tian, 1995, type specimens and

topotypes of Rana daorum Bain, Lathrop, Murphy, Orlov & Ho, 2003

(minimum-maximum) are given in millimetres. Measurements are 0

Measurements

ginal except for

those of the type specimens of Rana daorum which are from BAIN et al. (2003); note

that their measurements of head length (in italics) have been taken on radiographs and

cannot be compared to standard measures.

Amolops mengyangensis | Rana daorum

Measurement |

Yunnan | Vietnam

ei - _ | = ce

Holotype | Paraypes | Holowype | Paratypes Ÿ Paratypes

n=l | n=2 | n=1 n=7 | n=8

svi 38.7 383-385 | 557 348-381 53.3-57.6

HW 123 123-127 | 188 11-130 15.6-17.6

ni 144 134-142 | 254 16.5-21.0 16.7-19.4

TL 23,1 | 215-237 | 33.6 | 19.1-23.5 32.7-364

opotypes

n=17

364-404

114-127

12.7-144

20.7-25.1

Source : MNHN, Paris

OHLER 6l

Von D

pe 77

Fig. 2. (a) Amolops mengrangensis Wu & Tian, 1995. Topotype of Rana duorum Bain, Lathrop, Murphy.

Orlov & Ho, 2003, MNHN 1999.5813, &, SVL 37.5 mm: Sa Pa, Vietnam: specimen in life at night

(b) Odorrana tiannanensis Yang & Li, 1980. MNHN 20040408, ?, SVL 104.1 mm: Nathen,

Phongsaly Province, Laos, specimen in life.

Source : MNHN, Paris

62 ALYTES 25 (1-2)

Odorrana jingdongensis Fei, Ye & Li, 2001

Odorrana jingdongensis Fei, Ye & Li, 2001: 110. - HyroniGm: 3 6, 11 9, tadpoles. — ONOMATOPHORE:

Holotype, by original description, CIB 581505, ?, SVL 90.4 mm. - Onymororr: Xinminxiang

GÆ4SN, 100°75E; alt. 1480 m), Jingdong Xian, Yunnan Sheng, China.

Rana hmongorum Bain, Lathrop, Murphy, Orlov & Ho, 2003: 40. — HypoDiGM: 15 4, 24 9, 5 young. —

OxowaTorHoRE: Holotype, by original designation, ROM 26376, ?, SVL 86.8 mm.- ONYMOTOPE:

Approximately 5 km NW of Sa Pa Village, near O Qui Ho Pass (22°20'09°N, 103°50"14”E; alt.

1400 m), Lao Cai Province, Vietnam. - STATUS: New synonym.

Until 2001, the name Odorrana andersonii was used for a group of sibling species of frogs

occurring over a large range in southern China and the northern Oriental region. Rana

andersonii Boulenger, 1882 had been first described as Polypedates yunnanensis Anderson,

1879 from Hotha valley in Yunnan. As this name was then a secondary homonym in the genus

Rana Linnaeus, 1758, BOULENGER (1882) coined a replacement name. The only extant syntype

(BMNH 1947.2.2.60) is a subadult male of 50.0 mm in SVL. Fer et al. (2001) revised the frogs

of the andersoni group and described two new species from China, thus limiting Odorrana

andersoni to north-western Yunnan (Longling and Tengchong Counties).

Without reference to the works of the Chinese authors and without comparison with the

syntypes of Polypedates yunnanensis, BAIN et al. (2003) described Rana hmongorum from Sa

Pa based on specimens similar to those called Rana andersonii by BOURRET (1942) and OHLER

et al. (2000) (fig. 1b). BAIN et al. (2003) gave differences to specimens allocated to Rana

andersonii but without indicating origin of these specimens or collection references. As they

did study neither type specimens nor topotypical specimens of the existing or newly coined

species group names, their conclusions are not convincing, and as we do not have any

information on the voucher specimens referred by them to Rana andersonii, their comparisons

are useless as several new taxa have been recognized in this group recently. Comparisons are

here provided for morphometric data (tab. 2) concerning the purported diagnostic characters

of Odorrana jingdongensis and Rana hmongorum. In these species, the discs of fingers are

moderately expanded, indicated as 1.4 times the width of third fingers in the description of

Odorrana jingdongensis, and as less than twice the base of phalanges in the description of

Rana hmongorum. Al fingers have lateroventral grooves joining terminally very closely in

Rana hmongorum. For both species terminal phalanges are described as T-shaped. The animal

pole of eggs is described as unpigmented in Odorrana jingdongensis and as white in

Rana hmongorum. Geographical distribution is continuous as the onymotopes of Odorrana

jingdongensis and Rana hmongorum are in the same mountain range. On the basis of

morphological character study, Rana hmongorum should be considered a junior subjective

synonym of Odorrana jingdongensis, with the valid name Odorrana jingdongensis.

Odorrana tiannanensis (Yang & Li, 1980)

Rana tiannanensis Yang & Li, 1980: 261

designation, KIZ 77.1.01

(2293/°N, 103°59'E), Yunnan S

HyrobiG: 2 8,4 9 .— ONOMATOPHORE: Holotype, by orig

mm. — ONYMOTOPE: Dawei Shan (alt. 1200 m), Hekou Xian

heng, China

Source : MNHN, Paris

OHLER 63

Table 2. - Mcasurements of Odorrana jingdongensis Fei, Ye & Li, 2001, including type

specimens of Odorrana jingdongensis Fei, Ye & Li, 2001, type specimens and

topotypes of Rana hmongorum Bain, Lathrop, Murphy, Orlov & Ho, 2003.

Measurements (minimum-maximum) are given in millimetres. Measurements are

original except for those of the type specimens of Odorrana jingdongensis which are

from FEï, YE & Li (2001) and those of the type specimens of Rana hmongorum which

are from BAIN et al. (2003); note that measurements of head length by the latter (in

italics) have been taken on radiographs and cannot be compared to standard measures.

The maximum value for TL in @ paratypes of Rana hmongorum given as 51.2 mm (in

bold) is probably in fact close to 41.2 mm.

Odorrana jingdongensis Rana hmongorum

Measurement

Yunnan Vietnam

Holotype and Holotype and

Paratypes paratypes | Topotypes Topotypes Paratypes paratypes

4 ? Gi) 9 ?

n=s n=19 n=8 n=2 n=12 n=10

svL 620-815 | 647-1080 | 529-640 750-777 | 547-653 | 743-868

HW 20.5-266 | 206364 | 195-223 273276 | 189211 | 292-313

HL 213-294 | 235384 | 211-233 287296 | 207304 | 370-423

TL 377-477 | 393-619 | 355-307 457-477 | 334512 | 43.1-492

ONOMATOrHORE: Holotype, by original designation, ROM 39684, ?, SVL 93.6 mm. - ONYMOTOPE:

Khe Moi River (18°5630°N, 104°48°35"E), approximately 24 km west of Con Cuong village (by

road), Con Cuong District, Nghe An Province, Vietnam. — STATUS: New synonym.

Rana hearwoleï Stuart & Bain, 2005: 487. - HyPobiGM: 6 , 5 ?.— ONOMATOPHORE: Holotype, by original

designation, FMNH 258134, 4, SVL 57.3 mm. - Onvmororr: Tributary of Nam Ou River

(22°0538"N, 102°12°50"E), Phou Dendin National Biodiversity Conservation Area, Phongsaly

District, Phongsaly Province, Laos. - STATUS: New synonÿm.

Rana tiannanensis was described by YANG & Li (1980) from Hekou, southern Yunnan. It

is still known only from a few places close to the Lao-Vietnamese border (YANG, 19914; FEt,

1999). In 2003, Bain et al. described a frog from northern Vietnam, Rana megatympanum,

that they considered close to Rana tiannanensis but being nevertheless a distinct species

because of absence of a vertical lip stripe (present in R. riannanensis), absence of dorsolateral

folds in males and dises of toes larger than those of fingers. STUART & BAIN (2005) described

a species close to R. megatympanum from northern Laos, Rana heanwolei, which can be

distinguished from the Vietnamese form by the presence of spinules on back and venter and

the presence of small dark brown spots.

New material available from northern Laos (Phongsaly Province) adjacent to the Chi-

nese locality of Rana tiannanensis and study of descriptions in the Chinese literature (FEI,

1999) show that Rana tiannanensis (fig. 2b) cannot be distinguished by any of these characters

Source : MNHN, Paris

64 ALYTES 25 (1-2)

Table 3. — Measurements of type specimens of Odorrana tiannanensis (Yang & Li, 1980).

including type specimens of Rana tiannanensis Yang & Li, 1980 and of Rana

megatympanum Bain, Lathrop, Murphy, Orlov & Ho, 2003, and other specimens

allocated to this taxon. Measurements (minimum-maximum) are given in millimetres.

Measurements are original except for those of the type specimens of Rana

tiannanensis Which are from YANG & Li (1980), and those of the type specimens of

Rana megatympanum which are from BAIN et al. (2003); note that measurements of

head length by the latter (in italics) have been taken on radiographs and cannot be

compared to standard measures.

Rana tiannanensis Rana megatympanum

Measurement

Yunnan Laos Vietnam

Holotype Holotype

and paratype | Paratypes Paratypes _ |'and paratypes

C 4 g C

n=2 n=4 n= n-4 n=10

SVL 523-535 | 905-1076 | 45.9-56.7 100.2 486-552 | 936-1053

HW 19.1-19.3 33.0-34.5 16.0-20.2 36.0 18.0-19.1 34.1-35.7

HL 20.5-20.8 | 360-386 | 180-222 377 246-271 | 41.3-47.6

TL 33.6-34.0 63.2-67.5 31.0-35.1 65.7 32.3-33.1 55.8-67.7

TYD 5.1-5.3 5.35-5.8 64-79 6.0 4.3-5.1 4.6-5.9

from Rana megatympanum or Rana heatwolei. Differences are due to incomplete descriptions

based on specimens of a single sex and reproductive stage. Striped lips are present in females

of specimens allocated to Rana tiannanensis (FEI, 1999) as they are in specimens allocated to

Rana megatympanum. The sexual dimorphism in tympanum size can be confirmed for

populations allocated to both species (tab. 3). Ranges of adult size of males and females

correspond. Comparison to the newly described Rana heanwolei Stuart & Bain, 200$ shows

that Lao specimens from Nam Ou valley (onymotope of R. hearwolei) share distinct small

spots with this new taxon but do not have spinules on body. In conclusion and reconsidering

character distribution, Rana megatympanum and Rana heatwolei should be treated as junior

subjective synonyms of Rana tiannanensis, With the valid name Odorrana tiannanensis.

Pelophylax lateralis (Boulenger, 1887)

Rana lateralis Boulenger, 1887: 483. — HyPoDyGM: 1 4. - ONOMATOrHORE: Holotype, by monotypy,

MSNG 29324, €, SVL 53 mm. - ONYMoToPE: Kokarit [*Kaw-ka-riet"] (98° 14'E, 16933 N), east of

Moulmein, Karen State, Myanmar

original designation as “type”. CIB 571085, 4, SVL 50.5 mm. - OnymororE: Mengyang (21°37°N,

101°22'E; alt. 680 m). Xishuanbanna Daizu Zizhizhou, Yunnan Province, China. — Srarus: New

synonym.

Source : MNHN, Paris

OHLER 65

Table 4. — Measurements of Pelophylax lateralis (Boulenger, 1887), including holotype of

Rana lateralis Boulenger, 1887, type specimens of Rana nigrolineata Liu & Hu, 1959

and other specimens allocated to this taxon. Measurements (minimum-maximum) are

given in millimetres. Measurements are original except for those of the holotype of

Rana lateralis Which are from BOULENGER (1920) and those of the type specimens of

Rana nigrolineata which are from Liu & Hu (1959).

Rana lateralis

Measurement

Myanmar Laos

Holotype

) 4 g

n=1 n=9 n=1

SVL 47 518 42.5-46.9 488

Hw 17 176 15.7-17.1 16.8

HL 16 189 16.4-18.1 178

TL 23 24.1 217-236 -3

Rana lateralis Rana nigrolineata

Measurement à

Thailand Yunnan

Holotype

and paratypes Paratypes

ê $ ê 9

n=3 n=3 n=10 n=8

SVL 52.4-55.0 58.8-67.5 43.0-52.5 51.0-61.3

Hw 18.5-22.0 20.9-22.3 145-184 18.0-21.8

HL 19.2-20.0 21.8-23.0 16.0-20.4 18.9-22.0

TL 23.6-24.7 26.9-313 22.5-28.4 26.8-33.0

This frog was originally described from Myanmar. It is currently known to occur in

Cambodia, Laos, Thailand and Vietnam. In southern Yunnan, a species of Pelophylax, Rana

nigrolineata, With transversal dorsal stripes, was described by Liu & HU (1959). I had the

opportunity to study the holotype of Rana nigrolineata and material from near the onymo-

tope of Rana lateralis in Myanmar (fig. 3), which prove very similar in general coloration

pattern, skin structure and webbing. The holotype of Rana lateralis has no oblique dark

bands on back according to its original description (BOULENGER, 1887), but oblique granular

folds (BOULENGER, 1912). Other specimens from Cambodia (unpublished data), Laos

(MNHN 1997.4097-4088), Myanmar (BMNH 1893.10.9.16, MNHN 1893.0456) and

Thailand (MNHN field numbers 17543-17544, Y.0041-0044) show those stripes more or less

distinctly either as granular folds or as dark stripes. In the diagnosis of Rana nigrolineata,

presence of oblique stripes in all specimens was mentioned (LIU & Hu, 1959: 530), as well as

external voc: of Rana lateralis Were said to be internal in the holotype,

and (TAYLOR, 1962), and appear as external foldings on sides

present in s

Source : MNHN, Paris

ALYTES 25 (1-2)

Source : MNHN, Paris

OHLER 67

of the throat in specimens from Laos and Thailand. As internal openings of vocal sacs are

present in the holotype, the absence of external sacs may be due to age or reproductive stage

of the specimen which is either a subadult or a non-reproductive male. In samples from a

single locality, coloration and presence of oblique bands is very variable and cannot be used

as a discriminating character. I could measure the holotype of Rana nigrolineata and

specimens from Laos, Myanmar and Thailand. Comparison of these frogs with those from

Yunnan (Liu & Hu, 1959) shows minor variation in size (SVL) ($ RE = 1.29; ® RE = 1.39)

(tab. 4) and measurements are largely overlapping between all samples. Considering morpho-

logical similarity, colour pattern and morphometrics, Rana nigrolineata should be considered

a junior subjective synonym of Rana lateralis.

Sylvirana cubitalis (Smith, 1917)

Rana cubitalis Smith, 1917: 277. - HYPODIGM: 2 6. — ONOMATOPHORE: Holotype, by original designation

as “type”, BMNH 1947.2.2.35 [ex BMNH 1921.4.1.228, ex MAS 1106], d, SVL 66.8 mm (given as

68 mm in SmiTH, 1917). -ONYMororE: Doi Nga Chang (alt. 490 m [1600 ft.]), [northern Siam; precise

locality unknown because name has probably changed], Thailand. - COMMENTS: According to SurrH

(1930: 104), the male described and measured by BOULENGER (1920: 138-139) is not the holotype.

Holotype, by original designation, CIB 571162, ?, SVL 61 mm. Description and figures of holotype:

Liu & HU, 1959: 518, fig. 4. 1961:201, .— ONYMoTOPE: Mengyang (21°37°N, 101°22E),

Xishuanbanna Daizu Zichizhou, Yunnan Sheng, China. - SraTus: New synonym.

Sylvirana cubitalis, first described by SMrrH (1917) from Myanmar (fig. 4), is a frog

species with a large distribution in northern Thailand, The males are easily recognized

because they have a prominent glandular patch of nuptial spines on the forearm unique

among ranine frogs. For thirty years, the Philippine species Rana varians Boulenger, 1894 was

considered to occur in southern China. On morphological data, particularly on the difference

in tadpole mouth structure, Duois (1992) distinguished the Chinese form as Rana nigrotym-

panica (fig. 5a). This species was only known from three female specimens (Liu & HU, 1961:

YANG, 19914). Description of holotypes of these two nominal species will be published

elsewhere (OHLER & DuBois, in prep.). STUART et al. (2006) published a redescription of this

frog from northern Laos as Rana nigrotympanica and compared it briefly with Rana cubitalis.

In the specimens they studied, vocal pouches are absent.

Recent collections in northern Laos adjacent to the onymotope of Rana nigrotympanica

included a frog species whose females fit with the Chinese form in having a distinct dark

triangle at the tympanic region, but the males have the sexual characters of Rana cubitalis

(fig. 6). Morphometrical data on Lao specimens and types are given in table 5 and support

similarity of the frogs from different populations. Though there is variation in size,

the observed values for RE are not particularly high (4 RE = 1.30; ? RE = 1.32). As

Fig.

— Pelophylax lateralis (Boulenger. 1887). Specimens in dorsal view. (a) Left, dark coloured

specimen with oblique bands, MNHN field number Y.0041, 4, SVL 53.2 mm. Right,

t coloured specimen without bands, MNHN field number 1 2. SVL 58.8 mm. Both

S nens from Phitsanuluk Province, Thailand. (b) Holotype of Rana nigrolineara Liu & Hu.

1959, CIB 571085, 4, SVL 50.5 mm: Mengyang. Yunnan, China

Source : MNHN, Paris

68 ALYTES 25 (1-2)

Fig. 4. - Sylrirana cubitalis (Smith, 1917). Holotype of Rana cubitalis Smith, 1917, BMNH 1947.2.2.35,

8, SVL 66.8 mm; Doi Nga Chang, Thailand. (a) Dorsal view. (b) Ventral view of throat and right

forcarm showing glandular pad.

Fig. 5. — (a) Sylvirana cubitalis (Smith, 1917). Holotype of Rana {Sylvirana) nigrotympanica Dubois

1992, CIB 571162, ®, SVL 61 mm; Mengyang, Yunnan, China: dorsal view. (b) Sy/virana milleti

(Smith, 1921). Syntype of Rana milleti Smith, 1921, BMNH 1947.2.1.45 [ex MAS 2602], 4.

SVL 36.2: Da Lat. Lam Dong Province, Vietnam: dorsal view.

Source : MNHN, Paris

OHLER 69

Fig. 6.— (a) Sylvirana cubitalis (Smith, 1917). MNHN 2005.0224, & , SVL 52.4 mm; Long Nai, Phongs:

Province, Laos; specimen in life, (b) Sy/virana milleti (Smith, 1921). MNHN 2004.0370, &,

40.3; Long Mai Khao, Phongsaly Province, Laos: specimen in life.

morphological data are available only on small samples of this species (21 4, 10 9).

conclusions on variation must be prudent. In specimens collected in Phongsaly Province

(MNHN 2005.0224-0230), adult males have more or less distinct foldings on sides of throat

but no extended vocal pouch. In the male specimens described by SruART et al. (2006), vocal

pouches were absent. SmiTrH (1917) described these pouches as external, whereas BOULENGER

(1920: 139), Bourker (1942: 317) and TayLoR (1962:426) mentioned internal vocal sacs.

Study of the type specimens (BMNH 1947.2.2.36 [ex MAS 1106] and 1921.4.1.229 [ex

MAS 2107)) confirms the vocal pouches being internal without morphological modifica-

tion on throat (fig. 4b). Thus the description of SMITH (1917) is wrong and does not

Source : MNHN, Paris

70 ALYTES 25 (1-2)

Table 5. - Measurements of Sy/virana cubitalis (Smith, 1917), including type specimens of

Rana cubitalis Smith, 1917, holotype of Rana (Sylvirana) nigrotympanica Dubois,

1992 and other specimens allocated to this taxon. Measurements (minimum-

maximum) are given in millimetres. Measurements are original except for those of the

Yunnan specimens of Rana nigrotympanica Which are from YANG (1991a).

Rana cubitalis

Measurement ,

Myanmar Thailand

Holotype Paratype

ê CJ 9

n=1 n= n=4

SVL 668 648 55.2-678 60.7-75.2

HW 208 183 15.6-20.9 19.1-22.1

HL 232 210 20.1-245 22.9-26.5

TL 391 374 29.5-40.7 35.5-42.6

EC CE |

Rana nigrotympanica

Measurement

Yunnan Laos

Holotype

8 8 8 g

n n=2 n=4 #=3

SVL 57.1 63.5-65 51.4-572 59.4-66.6

HW 18.2 19-19.5 147-168 18.6-19.8

HL 21.0 21.5-22 19.1-204 21.9-23.5

TL 35.3 37-39 29.8-34.0 34.0-39.1

provide a discriminant character as supposed by STUART et al. (2006). In conclusion, the

name Rana nigrotympanica refers to the female of Rana cubitalis, and the name Rana

nigrotympanica is a junior subjective synonym of Rana cubitalis, with Sylvirana cubitalis as its

valid name.

Sylvirana milleti (Smith, 1921)

Rana milleti Smith, 1921: 431. - HyPobiGM: 5 &, 5 ?.— ONOMATOPHORE: Syntypes, BMNH 1947.2.1.37-

46, 5 8,5 ?, SVL respectively 35.0-37.9 mm and 44.4-47.8 mm. - ONYmororr: Da Lat [Dalat]

(LIS58"N, 108°24'E), Langbian Plateau, Lam Dong Province, Vietnam

Rana bannanica Rao & Yang, 1997: 157. - HyPoiGn: 8 4. — ONOMATOPHORE: Holotype, by original

designation, KZ 94001, 4, SVL 43.0 mm. - Oxvmororr: Mo-han (21°10°N, 101*45'E: alt. 850 m),

China-Laos border, Xishuangbanna Daizu Zizhizhou, Yunnan Sheng, China. — Srarus: New

synonym

SmrrH (1921) described Rana milleri from the Lang Bian Plateau of central Vietnam

(fig. 5b). The species was known only from this region until the recent discovery of the species

Source : MNHN, Paris

OHLER 71

Table 6. - Measurements of Sylvirana milleti (Smith, 1921), including type specimens 0°

Rana milleti Smith, 1921 and of Rana bannanica Rao & Yang, 1997, and othe-

specimens allocated to this taxon. Measurements (minimum-maximum) are given in

millimetres. Measurements are original except for those of the type specimens of Rana

bannanica which are from RAO & YANG (1997).

Rana milleti

Measurement

Vietnam Thailand

Syntypes Syntypes

g O)

n=5 n=5 n=14 n=1

SVL 35.0-37.9 444-478 383-417 494

HW 123-12.8 144-155 113-12.9 157

Rana bannanica

Measurement

Yunnan Laos

Holotype Paratypes |

é ê

n=7 n=6

SvL 430 38.0-43.0 37.5-41.1 45.6

HW 140 12.0-14.0 120-135 147

ul 17.0 15.0-17.0 15.1-16.1 175

TL 22.0 20.0-23.0 18.7-22.9 23.9

from Thailand and Cambodia (CHUAYNKERN et al., 2004). In 1997, Rao & Y ANG described a

brown Rana similar to Rana nigrovittata from the China-Laos border.

When doing field work in Phongsaly province of Laos, near to the Chinese border, I

discovered a small brown frog (fig. 6b) which corresponds in morphological characters

closely to the description of Rana bannanica, but also to the type specimens and other

specimens studied of Rana milleti, These two forms are the only Sy/virana that share the

following characters: small body size, moderately large head, reduced webbing, dorsal skin

bearing glandular warts and dense horny spinules. Morphological analysis shows no major

differentiation but a slightly smaller size in the syntypes from Vietnam (tab. 6), but size

variation is of small or moderate amplitude (4 RE = 1.23; RE = 1.11). Smaller size is not

correlated to any other character of shape, morphology or coloration. Pending genetic data,

Rana bannanica should be regarded a junior subjective synonym of Rana milleti, with the

valid name Sylvirana milleti. The distribution area of the species is extended far north to

southern Yunnan. Rarity of records may be due to problems of taxonomic allocation (some

specimens could be regarded as juveniles of “nigrovittata”) but also due to ecological

demands. Specimens have been collected in marshes and ponds close to forests or in primary

Source : MNHN, Paris

72 ALYTES 25 (1-2)

forest. As pointed out by OHLER & DELORME (2006), this kind of habitats is particularly rare

in Asia as most of lowland forests and forests on non-sloping parts have disappeared to be

replaced by paddies.

ACKNOWLEDGEMENTS

For access provided to their collections, I give my thanks to the curators of the Natural History

Museum, of the Chengdu Institute of Biology and of the Field Museum of Natural History. The

Phongsaly Forest Conservation and Rural Development Program organized and supported fieldwork in

northern Laos and the PPF “État et structure phylogénétique de la biodiversité actuelle et fossile”

gave financial support. Yodchayi Chuernkern made specimens collected in Thailand available for this

study. I thank Prof. Fei Liang and Prof. Ye Changyuan as well as Prof. Jiang Jiangping for valuable

discussion of these matters on my visits to Chengdu.

LITERATURE CITED

BAIN, RH, LaTHROr, A., Murray, R.W., ORLOV, NL. & Ho, C.T., 2003. - Cryptic species of a cascade

frog from Southeast Asia: taxonomic revisions and descriptions of six new species. Am. Mus.

Novit., 417: 1-60.

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

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

— 1887. - An account of the Batrachians obtained in Burma by M.L. Fea, of the Genoa Civic Museum.

Ann, Mus. ci. Stor. nat. Genova, (2), 5: 418-424, pl. 3-5

Lx 1912. — A vertebrate fauna of the Malay Peninsula from the Isthmus of Kra to Singapore including the

adjacent islands. Reptilia and Batrachia. London, Taylor & Francis: i-xiii + 1-294, 1 pl.

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

Rana. Rec. Ind. Mus., 20: 1-126.

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

+ 1:547, 4 pl.

CHuaAYNKERN, Y., OnLER, À. INTHARA, C., KLUMTHONG, P. & Duroïs, A., 2004. - The recent distribu-

tion of Rana milleti Smith, 1921 in mainland Southeast Asia with the first record from Cambodia.

Nat, Hist. J. Chulalongkorn Univ., 4: 1-13.

Dumois, À., 1976. — Les Grenouilles du sous-genre Paa du Népal (famille Ranidae, genre Rand). Cahiers

népalais-Documents, Paris, CNRS, 6: i-vi + 1-275.

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

(10): 305-352.

2005a - Propositions pour incorporation des nomina de taxons de rang supérieur dans le Code

international de nomenclature =vologique. In: À. Dusois, O. PoNCY, V. MALÉCOT & N. LÉGER (ed.),

Comment nommer les taxons de rang supérieur en =oologie et en botanique?, Biosystema, 23: 73-96

me 2005b. - Proposed Rules for the incorporation of nomina of higher-ranked z0ological taxa in the

International Code of Zoological Nomenclature. 1. Some general questions, concepts and terms of

biological nomenclature. Zoosystema, 27 (2): 365-426.

ae 2006. - Proposed Rules for the incorporation of nomina of higher-ranked zoological taxa in the

International Code of Zoological Nomenclature. 2. The proposed Rules and their rationale.

Zoosystema, 28 (1): 165-258.

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

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

Source : MNHN, Paris

OHLER 73

Fer, L., YE, C.-Y. & Li, D., 2001. — Description of two new species of the genus Odorrana in China. Acta

zootaxonomica sinica, 26: 10-114. [In Chinese].

FRosT, D. R., GRANT, T., FAIVOVICH, J., BAZIN, R. H., HAAS, A., HaDDAD, C. F. B., DE SÀ, R. O..

CHANNING, A., WILKINSON, M., DONNELLAN, S. C., RAXWORTHY, C. J., CAMPBELL, J. A., BLOTTO,

B. L., MoLr, P., DREWES, R. C., NUSSBAUM, R. A., LYNCH, J. D., GREEN, D. M. & WHEELER, W.

C., 2006. - The amphibian tree of life. Bull. am. Mus. nat. Hist., 297: 1-370.

INGER, R. F, 1999. — Distribution of amphibians of southern Asia and adjacent islands. Jr: W. E.

DUELLMAN (ed.), Patterns of distribution of amphibians: a global perspective, Baltimore, John

Hopkins University Press: 456-482.

KÔHLER, J., VIFITES D. R., BONNET, R. M., GARCIA, F. H., GLAW, F., STEINKE, D. & VENCES, M., 2005.

— New amphibians and global conservation: a boost in species discoveries in a highly endangered

vertebrate group. Bioscience, 55: 693-696.

Liv, C.-C. & Hu, S.-C., 1959. - Preliminary report of Amphibia from southern Yunnan. Acta zoologica

sinica, Beijing, 11: 509-533, 5 pl. [In Chinese].

ut 1961. - The tailless amphibians of China. Beïjing, Science Press: [i

[In Chinese].

OuLe, A. & DELORME, M., 2006. — Well known does not mean well studied: morphological and

molecular evidence for existence of sibling species in the Javanese gliding frog Rhacophorus

reinwardtit (Amphibia, Anura). Comptes rendus Biologies, 329: 86-97.

OnLe, A. & Duois, A., 1989. - Démonstration de l’origine indépendante des ventouses digitales dans

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

419-422.

= in preparation. — Revision of the species of Sy/virana Dubois, 1992, with description of name-

bearing types.

OHLER, A., MARQUIS, O., SWAN, S. & GROSIEAN, S., 2000. - Amphibian biodiversity of Hoang Lien

Nature Reserve (Lao Cai Province, northern Vietnam) with description of two new species.

Herpetozoa, 13 (1-2): 71-87.

OosTERZEE, P. VAN, 1997. — Where worlds collide: the Wallace line. Ithaca, Cornell University: i-xii+1-234.

RA0, D.-Q. & YANG, D.-T., 1997. - The description of a new ranid species from Xishuangbanna — Rana

bannanica Sp. nov. Zool. Res, Kunming, 18: 157-161. [In Chinese].

ScHiorz, A., 1999. eefrogs of Africa. Chimaira, Frankfurt am Main, 1-350.

Simpson, G. G., 1940. - Types in modern taxonomy. American Journal of Science, 238: 413-431.

Suit, M. À., 1917. - Description of a new snake and a new frog from Siam. J nat. His. Siam, 2 (4):

276-278.

——— 1921. - New or little-known Reptiles and Batrachians from Southern Annam (Indo-China). Proc.

zool. Soc. London, 1921: 423-440.

Sruarr, B. L. & BAIN, R. H., 2005. - Three new species of spinule-bearing frogs allied to Rana

megatympanum Bain, Lathrop, Murphy, Orlov and Ho, 2003 from Laos and Vietnam. Herpetolo-

gica, 61: 478-49°

+ ixvi + 1-364, pl. 1-6 + 1-28.

F. & TAN F. L., 2006. - Record of the little-known Rana nigrotympanica

Ranidae) from northern Laos. Hamadryad, 30: 108-113.

R. W., 2004. — Status and trends of amphibian declines and extinctions worldwide.

306: 1783-1786.

— The amphibian fauna of Thailand. Univ. Kansas Sci. Bull, 63: 265-599.

Wu G.-F & TiaN W:8., 1995. - À new Amolops species from southern Yunnan. /n: E.-M. ZHAO (ed.),

Amphibian =oogeographic division of China, Herperological Series, 8, Sichuan Journal of Zoology,

Supplement: 50-52.

Yan, D. (ed.), 191a. — The Amphibia fauna of Yunnan. Kunming, China Forestry Publishing House:

fi-vii] + isiv + 1-259. [In Chinese].

ee 1991 28 February]. - Phylogenetic systematics of the Amolops group of ranid frogs of Southeastern

Asia and the Greater Sunda Islands. Fieldiana: Zoology, (ns), 63: i-iit + 1-42. [Publication date

according to cover of publication; not 18 February as stated in FRosr et al.

YANG, DT. & Li, S.-M., 1980. - À new species of the genus Rana from Yunnan. Zool. Res, Kunming,

:261-264, [In Chinese].

Source : MNHN, Paris

74 ALYTES 25 (1-2)

Zuao, E., 1999. - Distribution patterns of amphibians in temperate East Asia. /n: W. E. DUELLMAN (ed.),

Patterns of distribution of amphibians: a global perspective, Baltimore, John Hopkins University

Press: 421-443.

Corresponding editor: Alain DUnots.

APPENDIX |

SPECIMENS STUDIED

Amolops mengyangensis. - Caina: Yunnan Sheng, Mengyang, CIB 579034, CIB 579037, CIB 64.1.1901,

8. VirNamr: Lao Cai Province, Sa Pa, MNHN 1990.5709-5813, .

Odorrana jingdongensis. — VirrNAM: Lao Cai Province, Sa Pa, MNHN 1938.0054-0057; MNHN

1999.5762-5781, topotypes of Rana hmongorum, d , 2, juv.

Odorrana tiannanensis. — LOS: Phongsali Province, Nathen, MNHIN 20040408, 9; Houey Phihet,

MNHN 2005.0180-0188, d .

Pelophylax lareralis. - Cana: Yunnan Sheng, Mengyang, CIB 571085, holotype of Rana nigrolineata,

C1B 570392, paratype of Rana nigrolineata, ® . LAoS: Vientiane market: MNHN 1997.4079-4088, 6

2. Myanmar: Rangoon, BMNH 1893.10.9.16, ; MNHN 1893.0456, &. THAILAND: Phitsanulok

Province, Nam Pad District, MNHN field number Y.0041-0044, 4, 2.

Sylvirana cubitalis. - Laos: Phongsali Province, Long Mai Kao, MNHN 2005.0224-0227, 4, ?; Nathen,

2005.0230, 9. THAILAND: “Doi Nga Chang”, BMNH 1947.2.2.36, holotype of Rana cubitalis, 6

BMNH 1921.4.1.229, 4; FMNH 135341-2, 6, ®; Loci Province, Dansai District, FMNH 196016,

Sylvirana milleti. — Laos: Phongsali Province, Long Nai, MNHN 2004.0370-0376, 4, . THAILAND:

Khao Yai, MNHN 1987.3433-3444, &. Vienna: Langbian Plateau, BMNH 1947.2.1.37-46, syn-

types of Rana milleti, 8, $.

Source : MNHN, Paris

Alytes, 2007, 25 (1-2): 75-82. 75

The status of the nomen Rana (Paa)

dhakuriensis Ray, 1997

(Anura, Ranidae), and comments

on the Amphibia reported from

the Nanda Devi Biosphere Reserve

(Uttar Pradesh, India)

Alain DuBois* & K. P. DiNESH**

* Vertébrés: Reptiles & Amphibiens, USM 0602 Taxonomie & Collections,

Département de Systématique & Evolution, Muséum national d'Histoire naturelle,

25 rue Cuvier, 75005 Paris, France, <adubois@mnhn.fr>

** Zoological Survey of India, Western Ghats Regiona

Annie Hall Road, 673 002, Calicut, Kerala, India, <dinesh

Station,

@gmail.com>

The nomen Rana (Paa) dhakuriensis Ray, 1997 was published without

any character allowing to recognize the taxon and is therefore a nomen

nudum. The status of the Amphibia reported by Ray (1997) from the Nanda

Devi Biosphere Reserve, under this nomen and seven others, is briefly

discussed.

Dusois et al. (2005: 45) included the nomen Rana dhakuriensis Ray, 1997 in their list of

recent amphibian taxonomic additions, without having had the opportunity to see the

original publication where this nomen had been created. Examination of this paper shows

that this new nomen was proposed for a taxon that was not described or diagnosed in any way.

Here is an integral copy of the part of this text dealing with this purported new species:

“Morphologically differs from all other species known from the neighbouring areas. Cryptic

colouration perfectly camouflaged these individuals with the natural surroundings. Detailed

description will be published after thorough study of the material”. There is in this text no

“description or definition that states in words characters that are purported to differentiate

the taxon”, or mention of “a bibliographic reference to such a published statement”, so that

this nomen is à nomen nudum according to Article 13.1 of the Code (ANONYMOUS, 1999). One

could at first sight consider “cryptic coloration” as a “character” of the species, but it is not,

as would be mention of a colour (brown, green, etc.): itis just an interpretation of what in the

eyes of a human this frog looks like, but it does not describe it. We are not aware that any

“detailed description” of this taxon was published later on, so that this nomen has no status

in zoological nomenclature. The question may arise, however, for which taxon was this new

nomen coined, and in which synonymy, if any, should the latter be placed. As Pranjalendu

Source : MNHN, Paris

76 ALYTES 25 (1-2)

Ray does not seem to have commented again on the Nanda Devi amphibian fauna in

subsequent works, and as the specimens are presumably kept in the Dehra Dun station of the

Zoological Survey of India which we did not have the opportunity to visit, all we can do is to

start from the information provided in Ray (1997).

To the best of our knowledge, the Amphibia of the Nanda Devi Biosphere Reserve Park

(Uttar Pradesh, India; quite close to the western border of Nepal) had never been studied

before the work of Ray (1997), so that a priori one could consider plausible the discovery of

a new species in this area. But this possibility is slight, regarding the uncertainties of the

taxonomy of amphibians used in this publication, as shown below. Based on collections

including 13 adults, one juvenile and 349 tadpoles, RAY (1997) reported 8 species from this

reserve, under the following nomina: (1) Bufo himalayanus Günther, 1864; (2) Bufo melano-

stictus Schneider, 1799; (3) Megophrys sp., (4) Amolops sp.; (5) Rana (Paa) annandalii

Boulenger, 1920; (6) Rana ( Paa) blanfordii Boulenger, 1882; (7) Rana (Paa) sp.; (8) Rana

{Paa) dhakuriensis n. sp. We comment below on these reports following the generic taxonomy.

Bufo Laurenti, 1768

The presence of the two species of Bufo reported by Ray (1997) is likely in this area, as

both these species occur in Nepal (DuBois, 1976, 1980, 1981, 2000) and in the Indian western

Himalayas as far West as Jammu and Kashmir (DuBois & MARTENS, 1977; DuBoIs, 1980,

1981). The brief notes of Ray (1997) suggest that his identification of the specimens was

correct.

According to the proposals of Dugois (1988, 2004b), in zoology two species liable to

hybridize successfully, either in the field or in artificial conditions, and to provide genuine

adult hybrids (at least in some cases), should never be allocated to different genera (but may

be placed in different subgenera). Adopting this point of view, we cannot follow the sugges-

tion of FRosr et al. (2006) to dismantle the genus Bufo into several genera between which

some species are known to hybridize (BLAIR, 1972). For example, successful hybridization

until the adult stage is known to occur (at least in some cases, as “best result”) between species

belonging in Frosr’s et al. (2006) genera Anaxyrus Tschudi, 1845 and Bufo Laurenti, 1768

(e.g., Bufo bufo and Bufo woodhousii, see BLAIR, 1972: 420), Anaxyrus and Cranopsis Cope,

1875 (e.g., Bufo terrestris and Bufo valliceps, see BLAIR, 1941 and MOORE, 1955; or Bufo fowleri

and Bufo valliceps, see BLAIR in MOORE, 1955), Bufo and Pseudepidalea Frost et al., 2006 (e.g.,

Bufo bufo and Bufo viridis, see HEMMER & BÔHME, 1974), Epidalea Cope, 1864 and Pseudepi-

dalea (e.g., Bufo calamita and Bufo viridis, see FLINDT & HEMMER, 1967; HEMMER, 1973;

SCHLYTER et al., 1991). It also probably occurs between Bufo Laurenti, 1768 and Epidalea

Cope, 1864 (Bufo gargarizans [as asiaticus] and Bufo raddei, according to CHEN, 1940), if, as

suggested by the data of Srôcx et al. (2001), the species Bufo raddei belongs indeed in the

Epidalea group rather than in the Pseudepidalea group as tentatively proposed by Fosr et al.

(2006). According to the cladogram of Frosr et al. (2006: 218), following Dugois’s (2004b)

guidelines and in order not to recognize paraphyletic genera, placing Bufo bufo and Bufo

viridis in the same genus requires to include also in the latter not only all other species of

FRosr et al.’s (2006) genera Bufo and Pseudepidalea, but also all species of their genera

Amietophrynus, Anaxyrus, Chaunus, Cranopsis, Duttaphrynus, Epidalea, Mertenso-

Source : MNHN, Paris

Dugois & DINESH 77

phryne, Peltophryne and Vandijkophrynus. I is however possible, if one wishes to recognize

taxonomically these “subclades” as taxa, to give them the status of subgenera of the genus

Bufo. Under this arrangement, the two indian species mentioned above can be referred to the

subgenus Duttaphrynus Frost et al., 2006 (type-species by original désignation Bufo melanos-

tictus Schneider, 1799) and can therefore be known as Bufo (Duttaphrynus) himalayanus

Günther, 1864 and Bufo (Duttaphrynus) melanostictus Schneider, 1799. Another advantage

of this solution is that it does not require to change the well-known binomen of the latter

species (Bufo melanostictus), one of the most quoted nomina of amphibians in the world,

being one of the commonest species in Asia.

Xenophrys Günther, 1864

The use of the generic nomen Megophrys Kuhl & Van Hasselt, 1822 is now restricted to

frogs of the Sunda islands, and the related species of the Himalayas are now referred to the

genus Xenophrys Günther, 1864 (FRosr et al., 2006; DELORME et al., 2006). A single species of

this genus, Xerophrys parva (Boulenger, 1893), is known to occur in the western part of Nepal.

The westernmost locality for which referenced voucher specimens have been reported so far

(Dusois, 1974) is Ghasa (28°37°N, 83938'E; alt. 2050-2100 m). In ANDERS’s (2002: 167)

distribution map of the species, a dot indicates its presence in a locality of the extreme western

part of the country, close to the border of Uttar Pradesh, but as the book of SCHLEICH &

KÂSTLE (2002) provides no reference to collection numbers of voucher specimens examined,

and as the amphibian taxonomy used in this book is not reliable (as it contains gross

misidentifications of specimens; see DUBOIS, 2004a), the validity of this record is open to

question. Regarding the specimens from the Nanda Devi reported by Ray (1997: 110), they

are stated to consist in “one hundred forty tadpoles of different stages” and mention is made

of their “Funnel type mouth feeding from the surface material and tail flickering continuously

under water”, a brief description which clearly points, indeed, to a species of the genus

Xenophrys. Pending obtention of adults from this area, the most parsimonious attitude is to

refer these specimens to Xenophrys parva. This is a genuine addition to the amphibian fauna

of Uttar Pradesh, and indeed an important range extension to the West, not only for this

species, but also for the genus Xenophrys and for the subfamily Megophryinae as defined by

DELORME et al. (2006).

Amolops Cope, 1865

In contrast, the genus Amolops has long been known from the western Himalayas

(ACHARI & KRIPALANI, 1951: KRIPALANI, 1952). Itis represented there by at least two distinct

species, now known (DuBois, 1974, 1992, 2000) as Amolops formosus (Günther, 1875) and

Amolops marmoratus (Blyth. 1855). Both these species are present from eastern to western

Nepal (Dugois, 1974, 1976, 2000) and in the western Himalayas as far west as Himachal

Pradesh (Dugois, 1981). A third species of this genus, Amolops monticola (Anderson, 1871),

is known from eastern Nepal but has not yet been reported in this country west of the Arun

valley (DuBois, 1980). The material from Nanda Devi reported by Ray (1997) consists in 13

Source : MNHN, Paris

78 ALYTES 25 (1-2)

tadpoles with ventral abdominal sucker, which therefore most likely belong in this genus, but

specific allocation is unknown. In another work on the amphibians of the Dehra Dun district

(Uttar Pradesh), RAY (1992) described two new species: Amolops chakrataensis (apparently

very similar to Amolops monticola) and Amolops jaunsari (apparently quite similar to Amolops

marmoratus). Both these species were redescribed as new by Ray (1999), in a paper

erroneously considered to provide their original descriptions by Dugois et al. (2005). The

validity of both these species will have to be confirmed by comparative studies in the frame of

a comprehensive revision of the genus 4molops, which is badly needed to solve the problems

left aside or created by the work of YANG (1991) on this genus. Although tadpoles of these

frogs can rather easily be collected in large numbers, e.g. by drying up portions of the torrents

where they live, adults are usually nocturnal, secretive, and they rarely leave the torrent bed: to

find them in significant numbers one has to climb slowly within the bed of the torrent at night

with head lamps or torchs. They are therefore quite seldom collected during standard surveys

of amphibians which are often made mostly around villages, in open habitats like paddy fields,

and at day time. Thus, these frogs are poorly known, and herpetologists who incidentally

collect them may consider them as new without making appropriate comparisons. Combined

with the rather high intraspecific variability of several species of this genus, this probably

explains the existence of many synonyms for some of them, including A. formosus and A.

marmoratus (Duois, 1974, 2000), and ignoring some of these synonymies without providing

new data, as done by some recent authors (YANG, 1991; ANDERS, 2002: FRosr et al., 2006) is

not likely to help our understanding of the taxonomy of this genus (see appendix 1).

Chaparana Bourret, 1939

The most problematic aspect of the paper of Ray (1997) is the taxonomy used for the

frogs referred to the subgenus Rana ( Paa), including the so-called new species. This subgenus

is now included in the genus Chaparana Bourret, 1939 (OnLer & DuBois, 2006), and it has

been the matter of several important works since the monography of BOULENGER (1920)

which seems to have been used as the basic taxonomic reference for Ray’s (1997) work.

Neither species Chaparana annandalii (Boulenger, 1920) and Chaparana blanfordii (Bou-

lenger, 1882) can be present in the Nanda Devi region. These are East Himalayan species,

which both occur only east of the Arun valley in eastern Nepal, in north-eastern India, and,

only for the second species, Bhutan (DUBOIs, unpublished) and southern Xizang (Tibet) in

China (DuBois, 1976, 1979; GROSIEAN & DuBois, 2006).

RaY’s(1997) report of Rana blanfordii in the reserve was based on 7 “examples” (adults?)

and 73 tadpoles, but no diagnostic characters were mentioned allowing to support their

identification. The specimens from Mussoorie (now in Uttar Pradesh) and the Balaya valley

near Simla (now in Himachal Pradesh) referred by BOULE (1920: 84) to Rana blanfordii

were shown by DuBois (1975, 1976) to be members of the species Chaparana minica (Dubois,

could possibly be mistaken for Chaparana blanfordii (SVL & 36.0-40.5 mm, ? 41.0-48.0 mm:

Dugois, 1976). The occurrence of C. minica is likely in the Nanda Nevi region, since it was

reported both east and west of the Park (Dumois, 1976, 1992; TiLak & Ray, 1985, as Rana

{ Paa) tuberculata), bat it only occurs at rather low altitudes (1000-2440 m; DuBois, 1976), so

Source : MNHN, Paris

Dusois & DINESH 79

it may only be present in the lowest parts of the Park whose elevation range spreads from 1500

to 5600 m.

Ray’s (1997) report of Rana annandalii in the Nanda Devi was based on a single juvenile,

and its identification was stated to be “based on descriptive morphology described by

BOULENGER (1920)”, without further details. Chaparana annandalii, which has never been

reported from west of Nepal, is similar in size to C. blanfordii or a little larger (SVL G

Chaparana rarica (Dubois, Matsui & Ohler, 2001), which is of similar size (SVL 4 37.3-

being, the latter is known with certainty only from one locality in western Nepal, the lake Rara

(2993 /1°N, 82°05'E; alt. 2990 m). The second locality mentioned by ANDERS (2002: 285, 1052),

Gurja Ghat, is based on 7 specimens referred with doubts to this species by NANHOE &

OUBOTER (1987), the status of which is not clear. The discovery of this species in Uttar

Pradesh would be an interesting range extension and would add one species to the fauna of

India.

Beside the two species above, RAY (1997) reported two other samples of Paa from the

Nanda Devi reserve: 100 tadpoles as Rana ( Paa) sp., and 6 “examples” (adults?) as “Rana

(Paa) dhakuriensis”® (nomen nudum). Could these specimens represent still one or two other

species of Chaparana?

Three other species of Chaparana have been reported so far from the regions neighbour-

ing the reserve: Chaparana vicina (Stoliczka, 1872), Chaparana polunini (Smith, 1951) and

Chaparana ercepeae (Dubois, 1974). Chaparana vicina is known from northern Pakistan,

Jammu & Kashmir and Himachal Pradesh (Dugois, 1980; GROSIEAN & DuBois, 2006), and its

finding in the Nanda Devi would be an important range extension to the East. Chaparana

polunini has so far been reported only from Nepal, from the East (west of the Arun valley) to

the extreme West of the country (Dumois, 1976), and from southern Xizang (China), at

altitudes between 2610 and 3990 m (Dugois, 1979). Its discovery in the Nanda Devi would be

a modest extension of its range to the West, but a new species record for India. Finally,

Chaparana ercepeae is known only from the extreme West of Nepal, between 2200 and 2650 m

(Dusois, 1976; Dusois & MaTsUI, unpublished), and its presence in the Nanda Devi is also

quite possible, but would also be a new record for India. Finally, ANDERS (2002: 275) also

reported another species, Chaparana liebigii (Günther, 1860) from extreme western Nepal, but

this record is highly open to question and might be based on a confusion with Chaparana

ercepeae. The westernmost locality known with certainty for C. liebigit and based on an

identified voucher is Lumsum (28°3/°N, 83°17°E : alt. 1980-2130 m) in central-western Nepal

(Duois, 1976: 259).

In conclusion, the region of the Nanda Nevi Reserve Biosphere, in medium and high

altitude just west of the occidental border of Nepal, certainly harbours frogs of the genus

Chaparana. The species C. minica, present both to the West and to the East of the reserve, is

most likely present in the latter. Three other species, C. ercepeae, C. polunini and

present in W n Nepal, could possibly occur there, whereas, given geographic distance, the

presence of C. vicina is more unlikely. At any rate, before describing a new species of

specimens collected there, including those

, should be done with reliably identified

rarica,

Chaparana from this area, careful comparison:

used to create the nomen “Rana dhakuriensis

Source : MNHN, Paris

80 ALYTES 25 (1-2)

specimens of these five species. Until such a work can be carried out, we suggest to place

provisionally the latter nomen nudum, with a query, in the synonymy of Chaparana minica,

together with the nomen Rana tuberculata Tilak & Ray, 1985 (see Dugois, 1992: 339).

LITERATURE CITED

ANONYMOUS [International Commission on Zoological Nomenclature], 1999. — International code of

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

ture: ixxix + 1-306.

ACHaRyI, M. N. & KRIPALANI, M. B., 1951. - On a collection of Reptilia and Batrachia from the Kangra

and Kulu valleys, western Himalayas. Rec. indian Mus., 49: 175-184.

AxDers, C., 2002. - Class Amphibia (Amphibians). fn: ScHLricH & KÂSTLE (2002): 133-348.

BLAIR, A. P., 1941. - Variation, isolating mechanisms, and hybridization in certain toads. Genetics, 26:

398-417.

BLAIR, W. E (ed), 1972. 11, Evidence from hybridization. Jn: W. F. BLAIR (ed.), Evolution in the genus

Bufo, Austin & London, Univ. Texas Press: 196-232 + 382-435.

BOULENGER, G. À., 1920. — À monograph of the South Asian, Papuan, Melanesian, and Australian frogs

of the genus Rana. Rec. india Mus., 20: 1-26.

Curs, C.-H., 1940. - Notes on a new hybrid toad (Bufo raddei ? X Bufo asiaticus 6). Chinese Journal of

experimental Biology, 1: 335-338.

DeLorme, M., Duuois, À. GROSEAN, S. & OnLer, A., 2006. - Une nouvelle ergotaxinomie des

Megophryidae (Amphibia, Anura). Alytes, 24 (1-4): 6-21.

Dusois, A., 1974. - Liste commentée d’Amphibiens récoltés au Népal. Bull. Mus. natn. Hist. nat., (3), 213

(Zool.143): 341-411

- 1975. - Un nouveau sous-genre (Paa) et trois nouvelles espèces du genre Rana. Remarques sur la

phylogénie des Ranidés (Amphibiens, Anoures). Bull. Mus. nan. Hist. nat.. (3), 324 (Zoo!. 231):

1093-1115

1976. - Les Grenouilles du sous-genre Pa du Népal (famille Ranidae, genre Rana). Cahiers népalais-

Documents, Paris, CNRS, 6: i-vi + 1-275

--- 1979. - Notes sur la systématique et la répartition des Amphibiens Anoures de Chine et des régions

avoisinantes. IL. Rana blanfordii Boulenger, 1882, Rana polunini Smith, 1951 et Rana vadongensis

Wu, 1977. Bull. mens. Soc. linn. Lyon., 48: 657-661

—— 1980. — L'influence de l’homme sur la répartition des Amphibiens dans l'Himalaya central et

occidental. C. r Soc. Biogéogr., 55: 155-178.

—— 1981. — Biogéographie des Amphibiens de l'Himalaya: état actuel des connaissances. /n: Paléos

graphie et biogéographie de l'Himalaya et du sous-continent indien. , Paris, CNRS: Cahiers népalais:

63-74.

= 1988. - The genus in zoology: a contribution to the theory of evolutionary systematies. Mém. Mus

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

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

(10): 305.

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

example of critical evaluation of biogeographical data. Zn: G. Mine & Y. ZHANG (ed.), Environ-

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

2004a. - Book review. Amphibians of Nepal: a few words of caution. Alytes, 21 (3-4): 174-180.

004. - Developmental pathway, speciation and supraspecific taxonomy in amphibians. 2, Develop-

mental pathway, hybridizability and generic taxonomy. Alries, 22 (1-2): 38-52

Dunois, À. CROMBIE, R L. & GLAW, F, 2005. - Amphibia Mundi. 12. Recent amphibians: generic and

infrageneric taxonomic additions (1981-2002). A/vres, 23 ( 69.

Dunois. À. & MARTENS, J., 1977. — Sur les Crapauds du groupe de Bufo viridis (Amphibiens, Anoures) de

l'Himalaya occidental (Cachemire et Ladakh). Bull. Soc. =ool. Fr, 102: 459-465

Source : MNHN, Paris

Dugois & DINESH 81

Dumois, A. & Marsut, M., 1983. - A new species of frog (genus Rana, subgenus Paa) from western Nepal

(Amphibia: Anoura). Copeia, 1983: 895-901.

FLINDT, & HEMMER, H., 1967. - Nachweis natürlicher Bastardierung von Bufo calamita und Bufo

viridis. Zool. Anz., 178 (5-6): 419-429.

FRosr, D. R., GRANT, T., FAIVOVICH, J., BAZIN, R. H., Haas, A., HADDAD, C. F. B., DE Sä, R. O.,

CHANNING, À., WILKINSON, M., DONNELLAN, S. C., RAXWORTHY, C. J., CAMPBELL, J. A., BLOTTO,

B. L., MOLER, P., DREWES, R. C., NUSSBAUM, R. A., LYNCH, J. D., GREEN, D. M. & WHEELER,

W. C., 2006. - The amphibian tree of life. Bull. amer. Mus. Nat. Hist., 297: 1-370.

GROSIEAN, S. & Dumois, A., 2006. — Description of advertisement calls of six species of the genus

Chaparana (Ranidae) from Nepal and India. A/ytes, 23 (3-4): 103-122.

HEMMER, H., 1973. - Die Bastardierung von Kreuzkrôte (Bufo calamita) und Wechselkrôte (Bufo viridis).

Salamandra, 9 (3-4): 118-136.

HEMMER, H. & BÔHME, W., 1974. - Nachweïs natürmicher Bastardierung der Erdkrôte (Bufo b. bufo) mit

der Wechselkrôte (Bufo v. viridis) in Theinland (Salientia, Bufonidae). Salamandra, 10 (3-4):

126-130.

KRIPALANI, M., 1952. - On Indian tadpoles with a suctorial disc. Rec. indian Mus., 50: 359-366, pl. 8.

Moore, J. À., 1955. — Abnormal combinations of nuclear and cytoplasmic systems in frogs and toads.

Ady. Genet. 7: 139-182.

NANHOE, L. M. R. & OUBo! E., 1987. — The distribution of Reptiles and Amphibians in the

Annapurna-Dhaulagiri region (Nepal). Zoologische Verhandelingen uitgeven door het Rijksmuseum

van natuurlijke Historie te Leiden, 240: 1-105.

A. & Dumois, A., 2006. — Phylogenetic relationships and generic taxonomy of the tribe Paini

(Amphibia, Anura, Ranidae, Dicroglossinae), with diagnoses of two new genera. Zoosystema, 28

(3): 769-784.

Ray, P., 1992. - Two new hill-stream Frogs of the genus Amolops Cope (Amphibia: Anura: Ranidae) from

Uttar Pradesh (India). Indian Journal of Forestry, 15 (4): 346-350.

a 1997. — Amphibia. Jn: J. R. B. ALFRED, (ed.), Fauna Nanda Devi Biosphere Reserve, Calcutta,

Zoological Survey of India, Fauna of Conservation Areas, 9: 109-114.

_— 1999. - Systematic studies on the amphibian fauna of the district Dehradun, Uttar Pradesh, India.

Memoirs of the zoological Survey of India, 18 (3): 1-102.

ScHLeICH, H. H. & KÂsre, W. (ed.), 2002. — Amphibians and Reptiles of Nepal. Koenigstein, A. R. G.

Gantner Verlag Kommanditgesellschaft: [i-ii] + i-x + 1-1201.

SCHLYTER, F., HÔGLUND, J. & STRÔMBERG, Hybridization and low numbers in isolated

populations of the natterjack, Bufo calamita, and the green toad, B. viridis, in southern Sweden:

possible conservation problems. Amphibia-Reptilia, 12: 267-281.

Srück, M., BRETSCHNEIDER, P. & GROSSE, W.-R., 2001. - The mating call and male release call of Bufo

raddei Strauch, 1876 with some phylogenetic implications, Russ. J. Herp.. 7 (3): 215-226.

TiLAK, R. & Ray, P. 1985. — Description of a new species of the sub-genus Rand {Paa) Dubois from

Chakrati Hills, district Dehra Dun, Uttar Pradesh, India (Ranidae, Anura). Zoo!. Anz., 215:

Yan D. 1991 (28 February]. — Phylogenetic systematics of the Amolops group of ranid frogs of

Southeastern Asia and the Greater Sunda Islands. Fieldiana: Zoology, (n.s), 63: iii + 1-42.

[Publication date according to cover of publication: not 18 February as stated in FROST et al. 2006:

253].

Corresponding editor: Annemarie OHLER

APPENDIX |

MISCELLANEOUS TAXONOMIC COMMENTS ON THE GENUS AMOLOPS

Some of the current taxonomic problems pointed out above in the genus Amolops simply

came from quick and careless reading of the works of colleagues, as exemplified in several

s below.

Source : MNHN, Paris

82 ALYTES 25 (1-2)

(1) YANG (1991: 16) wrote: “Dubois (1974) suggested 4. kaulbackï is probably a subspe-

cies of À. afghanus”, whereas DuBois (1974: 361) had suggested that 4. kaulbacki might be a

subspecies of 4. formosus.

(2) YANG (1991) completely ignored the synonymisation by DuBois (1974) of À. hima-

layanus With 4. formosus, although it was based on statistical comparisons of series of

specimens: YANG (1991), followed by FRosr et al. (2006: 367), recognized both species but did

not even compare them in their respective diagnoses.

(3) Quite similarly, FRosr et al. (2006: 252) wrote: “Dubois (2000: 331; 2004a: 176)

suggested, on the basis of the examination of the holotype, this taxon [Amolops nepalicus] is

synonymous with Amolops formosus”, whereas Dugois (2000: 333, 335) had considered 4.

nepalicus a synonym of À. marmoratus. Such gross misquoting testify to poor attention given

to the publications at stake.

(4) Frosr et al. (2006: 252) further complained that “[Dubois] did not provide any

discussion regarding the differences itemized in the original description or the diagnostic

differences noted by Yang (1991b)”. Beside the absence of meaning of “or” in this sentence

(as the original description of À. nepalicus was indeed in YANG, 1991), this statement is

misleading, as a single difference was stated by YANG (1991: 23) to distinguish À. nepalicus

from À. afghanus (now A. marmoratus): “A. nepalicus differs (...) from A. afghanus in having

the vomerine tooth groups oblique instead of transverse”, which, to experienced taxonomists

nowadays, is at least a doubtful criterion for species recognition. The description of 4.

nepalicus from a single adult male and 5 tadpoles (without any information on their locality

data and on the reasons for their allocation to the same species as the adult male), without any

morphological or statistical comparison with the many adults of 4. marmoratus available

from Nepal (e.g., 39 4,9 9,7 juveniles and 14 tadpoles listed in Dugois, 1974: 397-398), looks

more like a 19° century's typological description, ignoring intraspecific variability, than like à

recent taxonomic work, and it is surprising to see subsequent support for such a hasty work

(ANDERS, 2002; Frosr et al., 2006).

(5) Instead of “resurrecting” specific synonym nomina without any evidence (e.g., À.

himalayanus and A. nepalicus, but also Rana barmoachensis Khan & Tasnim, 1979, synony-

mized with R. hazarensis by Dusois, 1992), Frosr et al. (2006) could have cared for presenting

a consistent generic taxonomy of frogs of this complex. It is thus difficult to understand on

which basis they put some species in two different genera, although at least morphologically

they are very similar and appear very closely related: e.g., Rana monticola Anderson, 1871

(placed by them in Amolops Cope, 196$) and Rana archotaphus Inger & Chan-ard, 1997

(placed by them in Huia Yang, 1991, but which should probably rather be known as Amolops

archotaphus).

(6) As explained by Dumois (20044), the specimens referred by ANDERS (2002) to Amolops

monticola are in fact Amolops formosus, whereas they described their specimens of Amolops

monticola as “ Polypedates species, not identified”.

Source : MNHN, Paris

Alytes, 2007, 25 (1-2): 83-85. 83

Provenance de trois espèces

de Phrynobatrachus d’Afrique centrale

décrites par Ahl en 1925 (Anura, Ranidae)

Thierry FRÉTEY

Association RACINE, 2 rue de la Cité, 35360 Médréac, France

The precise origin of the specimens described by Ant (1925) as three

new species is shown to be Bouala in République centrafricaine.

AuL, en 1925, a décrit trois espèces nouvelles: Hylarthroleptis elberti (17 syntypes),

Pararthroleptis nanus (1 seul spécimen, holotype par monotypie) et Arthroleptis pygmaeus

(1 seul spécimen, holotype par monotypie) qui ont été transférées depuis dans le genre

Phrynobatrachus Günther, 1862 (DuBois, 1981).

Ces trois nouvelles espèces proviennent de “Buala am Uam, Neu Kamerun”. En 1985,

FR dans son ouvrage de référence, citait donc tout simplement comme localité-type :

“Buala am Uam, Neu Kamerun”.… du Cameroun. Depuis 2004, dans la version 3, et en 2006,

dans la version 4, de “Amphibian Species of the World” sur le site internet de l'American

Museum of Natural History [http//research.amnh.org/herpetology/amphibia/], Frost a

modifié cette localité-type en : “Buala am Uam, Neu Kamerun”.… du Tchad.

Sur une carte du Tchad, aucune localité Buala am Uam” ne figure. En revanche, il existe

une rivière “Ouham” qui coule au Tchad avant de se jeter dans le Chari. Or, cette rivière prend

sa source en République centrafricaine. Sur des cartes détaillées, nous avons pu retrouver la

localité dénommée “Bouala” (06°22°N, 15°37'E; altitude 998 m) en remontant le cours du

“Ouham”, à 45 km au nord de Bouar, en République centrafricaine et à moins de 60 km de la

amerounaise (fig. 1).

frontière

L'histoire du Cameroun, au début du siècle dernier, est en effet assez tourmentée: en

1911, ce pays, alors colonie allemande dénommée * Neu Kamerun”, engloba

l'Est, des régions du Gabon, du Congo, du Tchad et de la République centrafricaine actuels,

cédées par la France. Ce n'est qu'en 1919, après la guerre, que ces territoires ont réintégré

l'Afrique équatoriale française (fig. 2).

au Sud et à

La localité-type ou onymotope (Dumois, 200$) de Phrynobatrachus elberti, P. nanus et

P pygmaeus est donc en fait en République centrafricaine.

Source : MNHN, Paris

84 ALYTES, 25 (1-2)

WE 1GE 18€ 2€

‘ ï

Bouala

ouala,

+ Bouar

5£scnnE3>0û

République centrafricaine

Fig. 1.- Carte de localisation de Bouala, République centra

Source : MNHN, Paris

FRÉTEY 85

Tchad

Nigeria

Fig. 2. - Carte du “Neu Kamerun” entre 1911 et 1919. En trait plein : frontières actuelles des différents

pays. En trait pointillé : frontières du “Neu Kamerun” entre 1911 et 1919. Gé, Guinée équatoriale;

RCA, République centrafricaine.

RÉFÉRENCES BIBLIOGRAPHIQUES

Auz, E., 1925. - Ueber neue afrikanische Frôsche der Familie Ranidae. Sitzungsberichte der Gesellschaft

Naturforschender Freunde zu Berlin, 1923: 96-106

Dumois, A., 1981. - Liste des genres et sous-genres nominaux de Ranoïdea (Amphibiens, Anoures) du

monde, avec identification de leurs espèces-types: conséquences nomenclaturales. Monitore oolo-

gico italiano, (ns), 15, suppl: 225-284.

Le 2005. - Proposed Rules for the incorporation of nomina of higher-ranked zoological taxa in the

International Code of Zoological Nomenclature. 1. Some general questions, concepts and terms of

biological nomenclature. Zoosystema, 27 (2): 365-426.

Fosr, D. R. (ed.), 1985. - Amphibian species of the world. Lawrence, Allen Press & The Association of

Systematies Collections: [i-iv] + iv + 1-732.

Corresponding editor: Alain DuBois.

Source : MNHN, Paris

Alytes, 2007, 25 (1-2): 86-88.

The publication date of the book Key

to Chinese Amphibia by Fei, Ye & Huang

Fe Liang *, YE Changyuan * & HUANG Yongzhao **

+ Chengdu Institute of Biology, Chinese Academy of Sciences,

Chengdu 610041, China

** Chongging Nature Museum, Chongqing 400700, China

The book Key to Chinese Amphibia, by Fei Liang, Ye Changyuan and

Huang Yongzhao, has sometimes be quoted as having been published in

1991. As shown by an official certificate here reproduced, this is wrong, as

it was published in China in December 1990.

The book Key to Chinese Amphibia, by Fe Liang, YE Changyuan and HUANG Yongzhao

was a landmark in the history of Chinese batrachology. It was the first comprehensive list of

the amphibians of China published after the Systematic key to the Amphibians of China

published in 1977 by the Sichuan Institute of Biology (ANONYMOUS, 1977). It provided keys,

diagnoses and figures for the adults, larvae and eggs of 279 species, arranged in 58 genera and

subgenera and in 11 families and subfamilies, and it gave the distribution of these species in

the provinces of China. It contained the creation of 15 new nominal taxa: 6 species (Lepto-

lalax alpinus, Leptolalax liui, Leptolalax ventripunctatus, Megophrys glandulosa, Megophrys

mangshanensis, Oreolalax granulosus), 5 genera (Glandirana, Odorrana, Pseudorana, Rugosa,

Tigrina), 3 subgenera (Quadrana, Tenuirana, Unculuana) and 1 subfamily (Occidozyginae).

In the second, unnumbered, page of the book (p. [i]), the date of the book was given as

1.12.1990 (1° December 1990). The book was distributed first in China in December 1990,

then mailed free to various colleagues in the world, but was apparently received there only

several weeks after the beginning of the year 1991. Probably for this reason, ZHAO & ADLER

(1993: 367) credited this book with the publication year 1991 instead of 1990. This informa-

tion was repeated in several works, especially in FRosr et al. (2006: 253, 266), who furthermore

wrote that “this publication did not appear until at least March of 1991” (FRosr et al., 2006:

253).

This latter information is wrong, hown by the official certificate reproduced here in

figure 1, and which can be translated as follows:

“Certificate

We formally here certify that the book Key 10 Chinese Amphibia by Fei Liang, Ye

Changyuan and Huang Yongzhao was published in fact in December 1990 by the Chongqing

branch of the Science and Technology Literature Publishing House.

Source : MNHN, Paris

Fer, YE & HUANG 87

CHERE) — HART 1900 #F 12 JF HNERAT I, KF

DArviE

EU PRE BERDCRR EE NRRE ER SM”

DEEE re

À H 4% :

HAN DTA aeé.

2006 # 9 A 30 H

Fig. 1. - Official certificate attesting the publication date of the book Key 10 Chinese Amphibia, by FEI

Liang, YE Changyuan and HUANG Yonghao.

Wang Yifu, Editor responsible for the book Key 10 Chinese Amphibia in the Chongqing

Branch of the Science and Technology Literature Publishing House.

Huang Yongzhao, Professor at the Chongqing Nature Museum.

30 September 2006.”

The book should therefore be quoted as FEI, YE & HUANG (1990), and the new taxa

named therein are dated December 1990.

Source : MNHN, Paris

88 ALYTES 25 (1-2)

LITERATURE CITED

ANONYMOUS [Sichuan Institute of Biology], 1977. - Sysematic key to the Amphibians of China. Beijing,

Kexue Chupanche: [i-iv] + i-v + 1-93, pl. 1-17. [In Chinese].

Fe, L., YE, C. & HUANG, Y., 1990. — Key to Chinese Amphibia. Chongging, Editions of Sciences and

Techniques: [i-iv] + 1-2 + 1-364. [In Chinese].

FRosT, D. R., GRANT, T., FAIVOVICH, J., BAZIN, R. H., Haas, A., HaDDAD, C. F. B., DE SA, R. O.,

CHANNING, A., WILKINSON, M., DONNELLAN, S. C., RAXWORTHY, C. J., CAMPBELL, J. A., BLOTTO,

B. L., MOLER, P., DREWES, R. C., NUSSBAUM, R. A., LYNCH, J. D., GREEN, D. M. & WHEELER, W.

C., 2006. - The amphibian tree of life. Bull. amer. Mus. Nat. Hist., 297: 1-370.

ZH40, E. & ADi K., 1993. — Herpetology of China. Oxford, Ohio, USA, SSAR, 1-522 + [i-v], pl. 1-48

LS A

Corresponding editor: Alain DUBOIs.

Source : MNHN, Paris

AIN7TTES

International Journal of Batrachology

published by ISSCA

EDITORIAL BOARD

Chief Editor: Alain Dusois (Reptiles et Amphibiens, Département de Systématique & Evolution, Muséum

national d'Histoire naturelle, CP 30, 25 rue Cuvier, 75005 Paris, France; <adubois@mnhn.fr>).

Deputy Editor: Franco ANDREONE (Museo Regionale di Scienze Naturali, Via G. Giolitti 36, 10123 Torino, Italy:

<fandreone@libero.it>).

Alytes Editorial Board: Lauren E. BrowN (Normal, USA): Heinz GiLLirscH (Wien, Austria); Stéphane

GROSIEAN (Paris, France); W. Ronald HevER (Washington, USA); Esteban O. LAVILLA (Tucumän, Argen-

tina); Thierry Lopé (Angers, France); Masafumi Marsur (Kyoto, Japan); Annemarie OHLER (Paris,

France); Alain PAGANO (Angers, France); John C. Poynron (London, England); Mark-Oliver RÔDEL

(Würzburg, Germany): Miguel VENCES (Braunschweig, Germany).

Amphibia Mundi Editorial Board. Alain Dupois, Chief Editor (Paris, France): Ronald I. CRoMBIE (San

Francisco, USA); Stéphane GROSIEAN (Paris, France), W. Ronald HEYER (Washington, USA); JIANG

Jianping (Chengdu, China); Esteban O. LaviLLA (Tucumän, Argentina); Jean-Claude RAGE (Paris,

France): David B. Wake (Berkeley, USA).

Technical Editorial Team (Paris, France): Alain Dusors (texts); Roger BouR (tables); Annemarie Ouer (figures).

Book Review Editor: Annemarie OHLER (Paris, France).

SHORT GUIDE FOR AUTHORS

(for more detailed Instructions to Authors, see Alytes, 1997, 14: 175-200)

Alytes publishes original papers in English, French or Spanish, in any discipline dealing with amphibians.

Beside articles and notes reporting results of original research, consideration is given for publication to synthetic

review articles, book reviews, comments and replies, and to papers based upon original high quality illustrations

(such as colour or black and white photographs), showing beautiful or rare species, interesting béhaviours, etc.

The title should be followed by the name(s) and address(es) of the author(s). The text should be typewritten

or printed double-spaced on one side of the paper. The manuscript should be organized as follows: English

abstract, introduction, material and methods, results, discussion, conclusion, French or Spanish abstract,

acknowledgements, literature cited, appendix.

Figures and tables should be mentioned in the text as follows: fig. 4 or tab. 4. Figures should not exceed 16

* 24 cm. The size of the lettering should ensure its legibility after reduction. The legends of figures and tables

should be assembled on a separate sheet. Each figure should be numbered using a pencil.

References in the text are to be written in capital letters (BOURRET, 1942; GRAF & POLLS PELAZ, 1989; INGER

et al., 1974). References in the Literature Cited section should be presented as follows:

Bourker, R., 1942. - Les batraciens de l'Indochine. Hanoï, Institut Océanographique de lIndochine: x + 1-547,

pl. 14 J

Gr4r, J-D. & PoLLs PeLAz, M., 1989. — Evolutionary genetics of the Rana esculenta complex. In: R. M. DAWLEY

& JP BoGaRT (ed.), Évolution and ecology of unisexual vertebrates, Albany, The New York State Museum:

289-302.

INGER, RE, Voris, H. K. & Voris, H. H., 1974. - Genetic variation and population ecology of some Southeast

Asian frogs of the genera Bufo and Rana. Biochem. Genet, 12: 121-145.

Manuscripts should be submitted either as attached document by e-mail, or in paper form by mail but then

in triplicate, either to Alain DuBois (address above) if dealing with amphibian morphology, anatomy, systema-

tics, biogeography, evolution, genetics, genetics, anomalies or developmental biology, or to Franco ANDREONE

(address above) if dealing with amphibian population geneties, ecology, ethology, life history or conservation

biology, including declining amphibian populations or pathology. Acceptance for publication will be decided by

the editors following review by at least two referees.

After acceptance, a copy of the final manuscript should be sent to the Chief Editor, either as attachment by

e-mail, or by mail on a floppy disk (3 % or 5 4). We welcome the following formats of text processing: (1)

preferably, MS Word (1.1 t0 6.0, DOS or Windows), WordPerfect (4.1 to 5.1, DOS or Windows) or WordStar (3.3

to 7.0); (2) less preferably, formated DOS (ASCII) or DOS-formated MS Word for the Macintosh (on a 3 % high

density 1.44 Mo floppy disk only).

Page charges are requested only from authors having institutional support for this purpose. The publication

of colour photographs is charged. For each published paper, a free pdf or 25 free reprints are offered by ISSCA

to the author(s). Additional reprints may be purchased.

Published with the support of AALRAM

(Association des Amis du Laboratoire des Reptiles et Amphibiens

du Muséum National d'Histoire Naturelle, Paris, France).

Directeur de la Publication: Alain DuBois.

Numéro de Commission Paritaire: 64851.

Alytes, 2007, 25 (1-2): 1-88.

Contents

Arne SCHIOTZ

Zoogeography of the treefrogs in Africa’s tropical forests …

Jaime BerToLUCI, Ricardo Augusto BRASSALOTI, Henrique Oliveira SAWAKUCHI, José

Wagner RiBEIRO Jr. & Germano WOEHL Jr.

Defensive behaviour with stiff-legged posture in the Brazilian tree toads

Dendrophryniscus brevipollicatus and D. leucomystax (Anura, Bufonidae) .. 38-44

Diana R. CARDENAS-Ro1AS, Alberto VELOSO & Rafael O. DE SÂ

The tadpole of Eupsophus queulensis (Anura, Cycloramphidae) ........... 45-54

Annemarie OHLER

New synonyms in specific names of frogs (Raninae)

from the border regions between China, Laos and Vietnam ............... 55-74

Alain Dugois & K. P. DINESH

The status of the nomen Rana (Paa) dhakuriensis Ray, 1997

(Anura, Ranidae), and comments on the Amphibia reported from

the Nanda Devi Biosphere Reserve (Uttar Pradesh, India) ................ 75-82

Thierry FRÉTEY

Provenance de trois espèces de Phrynobatrachus d’Afrique centrale

décrites par Ahl en 1925 (Anura, Ranidae) ............................... 83-85

Fi Liang, YE Changyuan & HUANG Yongzhao

The publication date of the book

Key to Chinese Amphibia by Fei, Ye & Huang ............................ 86-88

ISSCA and Alytes are now accessible online at our website:

http:/www2.mnhn.fr/alytes/

Alytes is printed on acid-free paper.

Alytes is indexed in Biosis, Cambridge Scientific Abstracts, Current Awareness in Biological

Sciences, Pascal, Referativny Zhurnal and The Zoological Record.

Imprimerie F. Paillart, Abbeville, France.

Dépôt légal: 2° trimestre 2007.

Source : MNHN, Paris