ArticlePDF Available

Anuran habitat selection and tem¬poral partitioning in a montane and submontane rainforest in Southwestern Cameroon – first results

Anuran habitat selection and temporal partitioning in Southwestern Cameroon – first results
SALAMANDRA, Rheinbach, 40(3/4), 2004
Salamandra, Rheinbach, 31.12.2004, 40(3/4): 239-260.
© 2004 Deutsche Gesellschaft für Herpetologie und Terrarienkunde e.V. (DGHT)
Anuran habitat selection and temporal partitioning in
a montane and submontane rainforest in Southwestern
Cameroon – first results
Spatial and temporal partitioning of an anuran assemblage in a montane and in a submontane habitat
at Mt. Nlonako, Southwestern Cameroon, were studied by visual encounter surveys along
transects. Qualitative and quantitative data on anuran community structure at two different altitudes
are provided as are data on habitat selection (creek vs. forest habitats), seasonal patterns (beginning
of dry season vs. beginning of rainy season), activity patterns (diurnal vs. nocturnal) and substrate
preference. The communities showed high species richness with a general low abundance per
species with the exception of Phrynobatrachus auritus which represented 36% of all specimens;
this dominance was especially pronounced in the rainy season. The high altitude study site
assemblage was characterized by montane species. Most specimens were found at night perching
on vegetation along the sides of creeks. A cluster analysis involving the twenty-one most abundant
species resulted in one cluster representing all species but Bufo latifrons thus showing an
unexpected proximity in the ecological variables studied.
Keywords: Anura: Amphibia; Mt. Nlonako; Southwestern Cameroon; species diversity;
spatial partitioning; temporal partitioning; perching sites.
1 Introduction
Although the anuran fauna of Cameroon has been studied periodically by a number
of authors (MERTENS 1938, 1939, 1940, 1968; MONARD 1951; PERRET & MERTENS 1957;
PERRET 1959, 1960a, 1961, 1966; EISENTRAUT 1963, 1973; PERRET & AMIET 1971; AMIET
1971, 1978, 1989; BÖHME 1975; JOGER 1982; GARTSHORE 1986; BÖHME & SCHNEIDER 1987;
LAWSON 1993; HERRMANN et al. 2000) and detailed studies on the biology of specific
anuran species exist (PERRET 1962, 1977, AMIET 1977, 1980), only two studies have
focused on the ecology and spatial partitioning of an anuran assemblage (SANDERSON
1936, AMIET 1975).
Anurans have been identified as sensitive bio-indicators which can provide
important data critical for the evaluation of the condition of their habitats (SCHIØTZ
1999, RÖDEL 2000).
Anuran population declines are recognized in many areas worldwide (ALFORD &
RICHARDS 1999, HOULAHAN et al. 2000). To determine the cause of such declines long
term studies of anuran species assemblages are necessary (PECHMANN et al. 1991,
BLAUSTEIN et al. 1994). Few such studies have been carried out (JAEGER 1980, PECHMANN
et al. 1991) and are entirely lacking for most of tropical Africa and especially for
Here we describe anuran assemblages at a montane and submontane rainforest in
Southwestern Cameroon and provide first qualitative and quantitative data on habitat
selection and temporal partitioning within these assemblages. Our aim is to shed light
on these anuran communities and their ecology and provide a basis for future studies,
which may assist in detecting long-term trends in amphibian population fluctuations.
SALAMANDRA, Rheinbach, 40(3/4), 2004
2 Materials and methods
2.1 Study area
All field work was carried out at Mt. Nlonako which is of volcanic origin, rising up
to 1825 m above sea level and is situated in Southwest Cameroon (Fig. 1a). From 21
to 26 November 2000 and 31 March to 5 April 2001 transects (4°51’N, 9°54’E, 450 m
altitude) were sampled in the vicinity of the village Ekomtolo at the southwestern foot
of Mt. Nlonako. From 24 - 27 March 2001 transects at the higher altitude of 1140 m
(4°55’N, 9°59’E) were sampled in the vicinity of the village Nguengue on the
northeastern flanks of the mountain (Fig. 1b).
The temperature during the time period of 21 November 2000 to 30 December
2001 varied from a minimum of 16.4 °C to a maximum of 29.5 °C at the Ekomtolo
study site and 12.9 °C to 23.6 °C at the Nguengue study site respectively. The relative
humidity during that time period was between 48.7 % and 100 % at Ekomtolo and
23.5 % and 100 % at Nguengue. For climatic details during the study periods refer to
Figures 2 and 3. All temperature and relative humidity data were recorded with Onset
Hobo® 08 t/rh data loggers with measuring intervals set to one hour. Temperature and
relative humidity data were processed with the Onset BoxCar® version 4.2 program.
Precipitation occurs throughout the year with a maximum during the peak rainy season
from August to October. The study period from the end of November 2000 coincided
with the beginning of the dry season which lasts from December to February. The
studies during March/April 2001 were carried out when the first heavy rain showers
were falling after a long dry period.
Fig. 1. Study sites (a) in Cameroon and (b) at
Mt Nlonako. Square: Ekomtolo; dot:
Nguengue; stippled areas: rural communities.
Untersuchungsgebiete (a) in Kamerun und (b)
am Mt Nlonako. Quadrat: Ekomtolo; Punkt:
Nguengue; schattiert: Siedlungen.
a) b)
Anuran habitat selection and temporal partitioning in Southwestern Cameroon – first results
SALAMANDRA, Rheinbach, 40(3/4), 2004
The vegetation in areas below an altitude of 800 m consists of typical lowland
rainforest followed by submontane rainforest up to an altitude of 1200 m. Above
1200 m montane rainforest and cloud forest exists at Mt. Nlonako. At the higher
altitudes primary rain forest dominates whereas the lower altitudes are dominated by
Fig. 2. Temperature
and relative humidity
at the Ekomtolo study
site (a) from Novem-
ber 2000 to December
2001, (b) during the
study period 21 to 26
November 2000 and
(c) during the study
period 31 March to 5
April 2001. Black:
temperature; grey:
relative humidity.
Temperatur und
relative Luftfeuchtig-
keit im Untersu-
chungsgebiet bei
Ekomtolo (a) von
November 2000 bis
Dezember 2001, (b)
während des Untersu-
chungszeitraums 21.-
26. November 2000
und (c) während des
raums 31. März bis
05. April 2001.
Schwarz: Temperatur;
grau: relative Luft-
SALAMANDRA, Rheinbach, 40(3/4), 2004
secondary forest interspersed with primary rainforest patches. Floristically Mt. Nlona-
ko belongs to the Guinea Congo region.
The nomenclature follows FROST (2002) with the exception of Arthroleptis poeci-
2.2 Transects
Visual encounter surveys along transects were carried out at the Ekomtolo site at
around 450 m altitude from 21-26 November 2000 and 31 March to 5 April 2001 and
at the Nguengue site at around 1140 m altitude from 24-27 March 2001. Each transect
had a length of 400 m. Transects were located either along streams or in the forest
occasionally crossing small streams (Fig. 4 and 5). Transects were sampled during the
day (n = 20) and during the night (n = 24, for details see Tab. 1). All transects were
Fig. 3. Temperature
and relative humidity
at the Nguengue study
site, a) from Novem-
ber 2000 to December
2001 and b) during
the study period 24 to
27 March 2001.
Black: temperature;
grey: relative humidi-
Temperatur und
relative Luftfeuchtig-
keit im Untersu-
chungsgebiet bei
Nguengue, a) von
November 2000 bis
Dezember 2001 und
b) während des
zeitraums 24.-27.
März 2001. Schwarz:
Temperatur; grau:
relative Luftfeuchtig-
Anuran habitat selection and temporal partitioning in Southwestern Cameroon – first results
SALAMANDRA, Rheinbach, 40(3/4), 2004
searched unidirectionally. The search direction within transects was alternated each
24 hrs period. Transects were sampled with two persons each searching one side of the
transect. Sides searched were rotated nightly between persons to minimize sampling
bias. All anurans sighted within five meters to either side of the transect were captured,
identified and either released at the site of capture or preserved in 70 % ethanol and
later deposited as voucher specimens at the Zoologisches Forschungsinstitut und
Museum Alexander Koenig, Bonn (ZFMK) (Appendix 1).
The following three habitat types were recognized in the stream transects: (a)
dominated by sandy banks, middle density of stream bank vegetation, slow water flow,
(b) very dense stream bank vegetation encroaching upon the stream area, very slow
water flow, (c) shallow stream bed, highly structured by rocks of varying size, highly
variable, sometimes fast water flow.
The forest transects were characterized by dense vegetation, clearings due to tree
fall and areas with an abundance of old fallen logs as well as areas of new growth.
Perching height and substrate was recorded for each individual anuran. Table 2
shows the various categories for vegetation and substrate.
Nearest neighbor cluster analysis was performed using the Statgraphics® Plus
version 5.0 software package.
3 Results
A total of 46 species with 456 individuals were recorded. Nine species were found only
at the Nguengue transects, 28 species only at the Ekomtolo transects. Nine species
occurred at both sites. 54 % (n = 25) of the species were represented by less than four
specimens each. Examples are Acanthixalus spinosus (BUCHHOLZ & PETERS, 1875),
Leptodactylodon boulengeri NIEDEN, 1910, Opisthothylax immaculatus (BOULENGER,
1903), Petropedetes parkeri AMIET, 1983, Werneria mertensiana (AMIET, 1976) and
Xenopus fraseri BOULENGER, 1905. Phrynobatrachus auritus BOULENGER, 1900 (Fig. 6)
was the most abundant anuran species in the study with 165 specimens (36 % of all
specimens) followed by Amnirana asperrima PERRET, 1977 (8 %, Fig. 7), Phrynodon
cf. sandersoni PARKER, 1935 (7 %), Dimorphognathus africanus (HALLOWELL, 1858
“1857”) (5 %) and Amnirana amnicola PERRET, 1977 (5 %). Figure 8 shows all species
and provides numbers as well as localities, whereas Figure 9 shows the abundance of
species and families. Ranidae clearly dominate the anuran assemblage with 69 % of
all specimens. Relative to species diversity, no family exhibited a pronounced high
diversity. Pipidae and Rhacophoridae were represented by only one species each.
study site transect n November 2000 March/April 2001
day night day night
Ekomtolo forest 16 4 2 4 6
Ekomtolo creek 14 4 2 2 6
Nguengue forest 7 0 0 3 4
Nguengue creek 7 0 0 3 4
Tab. 1. Number of sampled transects and their distribution over research sites, periods, habitats
and time.
Anzahl der untersuchten Transekte und deren Verteilung auf Untersuchungsgebiete, -zeiträume,
-habitate und -zeiten.
SALAMANDRA, Rheinbach, 40(3/4), 2004
Fig. 4. Section of the
stream transect at the
Ekomtolo study site.
Photo: H.-W.
Ausschnitt des
Bachtransekts im
bei Ekomtolo.
Fig. 5. Section of the
forest transect at the
Ekomtolo study site.
Photo: H.-W. HERR-
Ausschnitt des
Waldtransekts im
bei Ekomtolo.
Anuran habitat selection and temporal partitioning in Southwestern Cameroon – first results
SALAMANDRA, Rheinbach, 40(3/4), 2004
3.1 Seasonal partitioning
Only the Ekomtolo site was visited during two different seasonal periods, namely
November 2000 which represents the beginning of the dry season and March/April
2001 which represents the end of the dry season/beginning of the rainy season.
Variation in species composition is shown in Figure 10. From a total of 37 species
found at Ekomtolo, 16 species were exclusively found during the March/April period.
Species found with fewer than four specimens (and thus not represented in Fig. 10) are
Afrixalus paradorsalis PERRET, 1960b, Hyperolius cf. sylvaticus SCHIØTZ, 1967, H.
kuligae MERTENS, 1940, Leptopelis omissus AMIET, 1992 “1991” (Fig. 11), L. rufus
REICHENOW, 1874, Opisthothylax immaculatus, Cardioglossa elegans BOULENGER, 1906,
Petropedetes cf. newtoni (BOCAGE, 1895), Phrynobatrachus batesi (BOULENGER, 1906),
Phrynodon cf. sandersoni, Nectophryne afra BUCHHOLZ & PETERS, 1875 (Fig. 12), N.
batesii BOULENGER, 1913, Xenopus fraseri. Only three species, Acanthixalus spinosus,
Cardioglossa gracilis BOULENGER, 1900 and Astylosternus fallax AMIET, 1978 “1977”,
were exclusively found in November. All these species were represented by only one
specimen. The abundance of individuals within most species encountered during both
study periods varied little with the exception of Phrynobatrachus auritus (beginning/
Fig. 6. Phrynobatra-
chus auritus male.
Photo: H.-W.
Fig. 7. Amnirana
asperrima male.
Photo: H.-W.
Amnirana asperri-
SALAMANDRA, Rheinbach, 40(3/4), 2004
Fig. 8. List of all species found ordered by habitat and study site (adapted after AMIET 1975). Ek:
Ekomtolo, Ngu: Nguengue.
Liste aller gefundenen Arten geordnet nach Habitat und Untersuchungsgebiet (verändert nach
AMIET 1975). Ek: Ekomtolo, Ngu: Nguengue.
Anuran habitat selection and temporal partitioning in Southwestern Cameroon – first results
SALAMANDRA, Rheinbach, 40(3/4), 2004
Fig. 9. Abundance of species and families. Parentheses: number of species/number of specimens.
Abundanz der Arten und Familien. Klammern: Anzahl der Arten/Anzahl der Individuen.
Fig. 10. Comparison of the number of specimens of species (n = 4 specimens) found at Ekomtolo
for November 2000 (black) and March/April 2001 (white). The dotted line marks equal specimen
numbers during both periods.
Vergleich der Individuenzahlen der Arten (n = 4 Individuen), die bei Ekomtolo im November 2000
(schwarz) und März/April 2001 (weiß) gefunden wurden. Die gestrichelte Linie markiert gleiche
Anzahl von Individuen während beider Zeiträume.
SALAMANDRA, Rheinbach, 40(3/4), 2004
end of dry season: 37/127), Dimorphognathus africanus (3/19), Hyperolius ocellatus
GÜNTHER, 1858 (10/2) (Fig. 13) and Arthroleptis adelphus PERRET, 1966 (6/1).
3.2 Altitudinal partitioning
Of the nine species found exclusively in Nguengue, Phrynobatrachus cricogaster
PERRET, 1957 was represented by more than three specimens (Fig. 14). Phrynodon cf.
sandersoni was mostly found in Nguenge (n = 29) although one specimen was found
in Ekomtolo. Other species found in small numbers only in Nguengue are Astyloster-
nus montanus AMIET, 1978 “1977”, A. perreti AMIET, 1978 “1977”, Bufo tuberosus
GÜNTHER, 1858, Cardioglossa venusta AMIET, 1972, Leptodactylodon boulengeri,
Petropedetes parkeri, Werneria mertensiana, Wolterstorffina parvipalmata (WERNER,
1898). Vice versa, a number of species mostly found in Ekomtolo were found with only
one specimen in Nguengue. Examples are Phrynobatrachus auritus, Dimorphogna-
thus africanus, Amnirana amnicola and Leptopelis calcaratus (BOULENGER, 1906).
Fig. 11. Leptopelis omissus male. Photo: H.-W.
Leptopelis omissus-Männchen.
Fig. 12. Nectophryne afra. Photo: H.-W. HERR-
Fig. 13. Hyperolius ocellatus female. a) dorsal; b) ventral. Photo: H.-W. HERRMANN
Hyperolius ocellatus-Weibchen. a) dorsal; b) ventral.
a) b)
Anuran habitat selection and temporal partitioning in Southwestern Cameroon – first results
SALAMANDRA, Rheinbach, 40(3/4), 2004
3.3 Daily partitioning
From a total of 44 transects searched, 20 were worked during the day and 24 during
the night; this may result in a sampling bias towards nocturnality (Table 1). Bufo
tuberosus, Leptodactylodon ovatus ANDERSSON, 1903 and Xenopus fraseri were only
found diurnally, as well as two of the three specimens of Werneria mertensiana
(Fig. 15). Specimens of L. mertensi and X. fraseri were hiding under stones and under
leaf litter respectively on a sandy bank. Many more species were exclusively seen at
night such as Hyperolius ocellatus, Chiromantis rufescens (GÜNTHER, 1869), Astylo-
sternus diadematus WERNER, 1898 as well as all members of the genus Leptopelis with
the exception of L. brevirostris (WERNER, 1898) of which only three of a total of seven
specimens were found active nocturnally. Similar ratios of diurnal versus nocturnal
activity could be found in Arthroleptis variabilis MATSCHIE, 1893 (10/5) and Scoto-
bleps gabonicus BOULENGER, 1900 (4/7) (Fig. 16). 13 more species with a total of more
than three specimens were found active during the day and at night but showed a much
stronger tendency towards nocturnality. For example, only five individuals of Phry-
nobatrachus auritus were active during the day whereas 155 specimens (97 %) showed
nocturnal activity. A similar pattern was found for Amnirana asperrima with 97 % of
all individuals observed at night.
3.4 Habitat partitioning
17 species (37 %) were found in both habitats, along the stream and in the forest
(Fig. 8). Both species of Nectophryne, N. afra and N. batesii, were found only in the
forest (Fig. 17). Most specimens and all species of Arthroleptis were found in the forest
Fig. 14. Comparison of the number of specimens of species (n = 4 specimens) found at Ekomtolo
450 m altitude (white) and Nguengue 1140 m altitude (black) in March/April 2001. The area
between dotted lines marks equal specimen numbers at both altitudes.
Vergleich der Individuenzahlen der Arten (n = 4 Individuen), die bei Ekomtolo 450 m Höhe (weiß)
und Nguengue 1140 m Höhe (schwarz) gefunden wurden. Der Bereich zwischen den gestrichelten
Linien markiert gleiche Individuenanzahl in beiden Höhen.
SALAMANDRA, Rheinbach, 40(3/4), 2004
(partially: A. variabilis (73 %); exclusively: A. adelphus, A. adolfifriederici NIEDEN,
1911 “1910”, A. poecilonotus). Within Astylosternus only A. montanus was found
exclusively in the forest. Astylosternus diadematus, A. fallax and A. perreti were all
found only in the stream transects. Other species found only along the stream were
Conraua crassipes (BUCHHOLZ & PETERS, 1875), Chiromantis rufescens, Hyperolius
ocellatus and Petropedetes cameronensis REICHENOW, 1874. 20 specimens (96 %) of all
Dimorphognathus africanus were found along the stream transects. The genus Amni-
rana shows remarkably different habitat preferences with A. asperrima being found
almost exclusively along the streams (97 %) and A. amnicola predominately in the
Fig. 15. Relative abundance of species (n = 4 specimens) found at night (black) and/or during the
day (white).
Relative Abundanz der Arten (n = 4 Individuen), die nachts (schwarz) und/oder tags (weiß)
gefunden wurden.
right page: Tab. 2. Perching sites. Numbers in site categories indicate number of specimens found.
Numbers in parentheses indicate specimens found in day retreats. Bold: species with more than
50 % of the specimens found on the ground in day retreats. Underlined: species with at least 80 %
of the specimens found on the ground in day retreats.
rechte Seite: Sitzwarten. Zahlen entsprechen der Anzahl der gefunden Individuen. Zahlen in
Klammern entsprechen der Anzahl der Individuen, die in Tagverstecken gefunden wurden. Fett:
Arten mit mehr als 50 % der Individuen auf dem Boden in Tagverstecken. Unterstrichen: Arten
mit mindestens 80 % der Individuen auf dem Boden in Tagverstecken.
Anuran habitat selection and temporal partitioning in Southwestern Cameroon – first results
SALAMANDRA, Rheinbach, 40(3/4), 2004
species total vegetation substrate
leaf branch root logs leaf rocks soil water
Acanthixalus spinosus 11
Afrixalus paradorsalis 22
Hyperolius ocellatus 12 5 5 2
Hyperolius cf. sylvaticus 321
Hyperolius kuligae 11
Leptopelis boulengeri 44
Leptopelis brevirostris 752
Leptopelis calcaratus 624
Leptopelis cf. millsoni 22
Leptopelis omissus 211
Leptopelis rufus 11
Opisthothylax immaculatus 22
Chiromantis rufescens 615
Arthroleptis adelphus 75 1(1) 1(1)
Arthroleptis adolfifriederici 11
Arthroleptis poecilonotus 541
Arthroleptis variabilis 15 6 1(1) 5(5) 2(2) 1(1)
Cardioglossa elegans 11
Cardioglossa gracilis 11
Cardioglossa leucomystax 16 7 4 1(1) 2(2) 2(1)
Cardioglossa venusta 11
Amnirana amnicola 22 10 4 4(3) 31(1)
Amnirana asperrima 38 10 10 7 1 4 6(1)
Conraua crassipes 10 2 1 1 3 3(1)
Dimorphognathus africanus 23 19 1(1) 3(2)
Petropedetes cameronensis 10 8 1(1) 1(1)
Petropedetes cf. newtoni 11
Petropedetes parkeri 11
Phrynobatrachus auritus 165 148 10 1(1) 4(3) 2(1)
Phrynobatrachus batesi 22
Phrynobatrachus cricogaster 12 9 1(1) 2(1)
Phrynodon cf. sandersoni 30 28 1(1) 1(1)
Astylosternus diadematus 41 3
Astylosternus fallax 11
Astylosternus montanus 211
Astylosternus perreti 11
Leptodactylodon boulengeri 11
Leptodactylodon ovatus 32(2) 1(1)
Scotobleps gabonicus 11 1(1) 1(1) 11(1) 61(1)
Bufo latifrons 51 1(1) 1(1) 2(2)
Bufo tuberosus 11(1)
Nectophryne afra 22
Nectophryne batesii 88
Werneria mertensiana 31 1(1) 1(1)
Wolterstorffina parvipalmata 33
Xenopus fraseri 11(1)
SALAMANDRA, Rheinbach, 40(3/4), 2004
forest (59 %). Similarly, Phrynobatrachus auritus was found predominately in forest
habitat (61 %). Another species of Phrynobatrachus, P. cricogaster was also found in
both habitats with a trend towards the forest (67 %, 8 out of 12 specimens).
Remarkably, Leptopelis brevirostris was the only species within the genus found only
along forest transects. Cardioglossa leucomystax (BOULENGER, 1903) (Fig. 18) is an
example of a species which occurs in both habitats but is predominant in streams.
3.5 Perching site partitioning
21 of all 46 species (46 %) were found only on plants (Tab. 2); the treefrogs
Chiromantis rufescens, Leptopelis boulengeri (WERNER, 1898) (Fig. 19), L. breviros-
tris, L. calcaratus and the dwarf toad Nectophryne batesii for example. On the
contrary, ten species (22 %) were encountered only on the ground (6 of these species
were represented by only one specimen). 15 species (32 %) were found on the ground
and in vegetation. Nine of these 15 species showed a strong link between their
microhabitat (ground/vegetation) and their temporal activity pattern (diurnal/noctur-
nal). In Arthroleptis adelphus, A. variabilis, Cardioglossa leucomystax, Petropedetes
cameronensis and Phrynodon cf. sandersoni all specimens encountered during the
day were found in retreats such as crevices, under rocks or logs or hidden in leaf litter.
With the exception of Arthroleptis variabilis more than half of the specimens of these
species were found active (not in retreats) during the night. This suggests that in most
if not all of these five species specimens found during the day were encountered in
their day retreats. In Dimorphognathus africanus, Amnirana amnicola, Phrynobatra-
chus auritus and P. cricogaster a similar pattern is observed, though not as substan-
tiated, 50-75 % of the specimens found during the day were in retreats.
Leptodactylodon ovatus, Bufo latifrons BOULENGER, 1900, B. tuberosus, Werneria
mertensiana and Xenopus fraseri were also predominately found in retreats during the
Fig. 16. Scotobleps
gabonicus male.
Photo: H.-W. HERR-
Scotobleps gaboni-
Anuran habitat selection and temporal partitioning in Southwestern Cameroon – first results
SALAMANDRA, Rheinbach, 40(3/4), 2004
day, however, with the distinction that there were no majorities of specimens found
active during the night.
All but two Hyperolius ocellatus were found at night in vegetation. The two
specimens on the ground were found in amplexus.
Of all species found in vegetation only Leptopelis boulengeri, L. cf. millsoni
(BOULENGER, 1895) and L. rufus were exclusively found perching on branches. All other
species were found perching on leaves as well or showed a marked preference for the
latter. Ten species showed a remarkably narrow substrate preference, found only on
In contrast Amnirana asperrima and Conraua crassipes were the only species
found on a variety of different substrates in the vegetation as well as on the ground.
3.6 Ecological relationships between species
A cluster analysis with the variables altitude (Ekomtolo/Nguengue), habitat (creek/
forest), activity pattern (nocturnal/diurnal) and perching site distribution (vegetation/
ground) (for data matrix see Appendix 2) was conducted (Fig. 20). Species with more
than 50 % of specimens found during the day in retreats on the ground were treated
as operational units only found in vegetation. Thereby we tried to avoid bias resulting
from possible retreats being scored as perching sites. Only species with four or more
specimens were included in the analysis. A second such cluster analysis excluding
specimens found in Nguengue was performed to check for cluster robustness by
avoiding bias from unequal study periods and unequal numbers of transects searched.
The analysis resulted in one large cluster representing all species except Bufo
latifrons. This dominant pattern was supported in the second analysis for specimens
Fig. 17. Relative abundance of species (n = 4 specimens) found along the creek (black) and/or
the forest (white) transects.
Relative Abundanz der Arten (n = 4 Individuen), die entlang von Bach- (schwarz) und/oder
Waldtransekten (weiß) gefunden wurden.
SALAMANDRA, Rheinbach, 40(3/4), 2004
from Ekomtolo only. Within the cluster of the remaining species, two well defined
groups were determined: one consisting of the species Leptopelis brevirostris and
Arthroleptis variabilis and one represented by Arthroleptis adelphus and A. poecilo-
notus. The former is supported by the Ekomtolo-specimens-alone analysis. A less
defined subgroup consisting of Hyperolius ocellatus, Chiromantis rufescens, Lepto-
pelis boulengeri, Cardioglossa leucomystax, Dimorphognathus africanus, Petro-
pedetes cameronensis, Leptopelis calcaratus, Phrynobatrachus auritus, Amnirana
asperrima, Conraua crassipes and Astylosternus diadematus exists within the main
4 Discussion
The high species diversity, 46 species from seven different families, found in two
localities at Mt. Nlonako while working only 24 transects is characteristic for ancient
tropical rainforests. The Southwest-Cameroonian mountain chain called “Dorsale
Camerounaise”, beginning with Mt. Cameroon in the South and ending with Tchabal
Mbabo in the North, of which Mt. Nlonako is part, is possibly the area with the highest
anuran species diversity in Africa (POYNTON, 1999). Another typical pattern for rainfor-
ests is that more than half of all species are represented by very low numbers of
specimens. The high species diversity found is possibly also based on the location of
the transects in an ecotone area with secondary and primary forest patches. For Mt.
Nlonako and adjacent areas data suggests that the total species count will be
considerably higher than that for the transects sampled in this study (H.-W. HERRMANN
et al., unpublished data).
The very high abundance of Phrynobatrachus auritus was somewhat suprising. As
the study period was very short, this pattern could possibly be explained by a
seasonality of the species which may have been triggered by the first heavy rain
showers after the dry season. This view is supported by the disproportion of specimens
found at the beginning (November 2000) and the end of the dry season (March/April
2001). This observation is in disagreement with MERTENS (1938) and LARGEN & DOWSETT-
LEMAIRE (1991) who found reproductive behavior in P. auritus at the end of the rainy
season (November) and during the dry season (January). A similar reproductive pattern
was found for Dimorphognathus africanus. In both species this pattern may be related
to reproductive activity synchronized by the first rains at the beginning of the rainy
season. 16 of all the species found in Ekomtolo were found only at the end of the dry/
rainy season transition. This could indicate a seasonal reproductive period.
The altitudinal pattern observed with some species found only in the Nguengue
transects is based on the high abundance of montane species such as Phrynobatrachus
cricogaster, Arthroleptis adolfifriederici, Astylosternus montanus, A. perreti, Cardio-
glossa venusta, Phrynodon cf. sandersoni, Werneria mertensiana and Wolterstorffina
parvipalmata (SCHMITZ, 1998, GARTSHORE, 1986, PERRET 1966, 1977, AMIET, 1975). Other
species such as Bufo tuberosus, Phrynobatrachus auritus, Dimorphognathus africa-
nus and Amnirana amnicola show a more submontaneous or lowland distribution and
may reach their upper limits around the Nguengue transects as only one specimen per
species was found at this higher altitude (1140 m).
The majority of all species showed a marked nocturnal activity. The few specimens
that were found during the day were often in places which suggested day retreats. Such
specimens can be considered to have been found “by accident” – not active – thereby
negating the conclusion of diurnality (see Tab. 2). In cases where more than half of
Anuran habitat selection and temporal partitioning in Southwestern Cameroon – first results
SALAMANDRA, Rheinbach, 40(3/4), 2004
such “diurnal” specimens were found in day retreats the diurnal data was excluded
from later analysis, i. e. the cluster analysis. A number of species such as Bufo latifrons,
B. tuberosus, Leptodactylodon ovatus and Xenopus fraseri were found exclusively or
mainly during the day, however, always in retreats. Due to the lack of data on nocturnal
activity in these species, no conclusions can be drawn relative to activity patterns.
From the “forest species”, i. e. species found exclusively or mostly within the forest
transects (species of the genera Nectophryne and Arthroleptis as well as Leptopelis
brevirostris and Astylosternus montanus), all but Astylosternus montanus are inde-
pendent from open water (streams or ponds) for their reproduction (MERTENS 1938,
AMIET & PERRET 1969, SCHEEL 1970, AMIET 1975, 1989, 1991, BARBAULT & TREFOUT
RODRIGUEZ 1979, FISCHER & HINKEL 1992); their eggs develop without a free larval stage.
Astylosternus montanus deposits its eggs in streams or temporal water bodies. The eggs
develop into tadpoles (AMIET 1977, 1989) which GARTSHORE (1986) found in February.
Two specimens of A. montanus were found in our forest transect at the end of March.
This suggests that the period of reproduction had already passed and may be at the
beginning of the dry season with larvae found later during that season. Scotobleps
gabonicus was found mainly in the stream transects with more specimens in November
than in March/April. This may indicate a reproductive activity peak at the beginning
of the dry season which is supported by the observations by LARGEN & DOWSETT-LEMAIRE
(1991) and LAWSON (1993) who observed calling males in November. Dimorphogna-
thus africanus, another species found almost entirely in the stream habitat, was found
much more often in March-April. Whether this is related to reproduction remains
unknown as information on the reproductive biology of D. africanus is widely
lacking. The increased occurrence of Leptopelis boulengeri at the stream in March/
April suggests reproduction at the beginning of the rainy season.
That most species were found primarily active on vegetation is expected for
treefrogs such as species of Acanthixalus, Afrixalus, Hyperolius, Leptopelis, Opistho-
thylax and Chiromantis. Of the species found in vegetation and on the ground a large
number occupied day retreats thus suggesting that the species are mainly active on
vegetation but retreat to the ground during the daily period of inactivity. After data
correction for such phenomena (see Tab. 2) the large majority of all species were found
on vegetation. Only Conraua crassipes and Amnirana asperrima show a broader
acceptance of perching sites on different structures in the vegetation and different
substrates on the ground. Six species were found exclusively active on the ground (not
in retreats) of which all were represented by only one specimen. Scotobleps gabonicus
was the only species with all eleven specimens encountered on the ground, four of
which were found in day retreats, indicating a nocturnal terrestrial activity.
The 21 species in the cluster analysis result in a cluster with 20 species except for
Bufo latifrons (Fig. 20). This is explained by the combination of diurnal activity, the
forest habitat and frequent occurrence on the ground by B. latifrons. Only two sub-
clusters of the main cluster formed by the remaining species showed a partial
distinctiveness, namely Leptopelis brevirostris and Arthroleptis variabilis as well as
A. adelphus and A. poecilonotus. These two groups together build another sub-cluster.
The whole sub-group with all four species is characterized by a forest habitat which
is directly linked to their reproductive mode. The two species subgroups are differen-
tiated mostly by a nocturnal activity in the A. adelphus-poecilonotus group and
predominately diurnal activity in the A. variabilis-L. brevirostris-group. Overall all
species included in the cluster analysis, except B. latifrons, show a remarkable
similarity within the ecological variables analyzed.
SALAMANDRA, Rheinbach, 40(3/4), 2004
We thank H. NOBIS, ALSCO Berufsbekleidungs-Service GmbH, for his generous financial support
through the sponsoring of the Cologne Zoo’s Cameroon Conservation Research Projects. The
Cameroon Ministry of Scientific and Technical Research (MINREST) granted research permits
and the Ministry of Environment and Forestry (MINEF) issued research, collecting and export
permits. We are especially indebted to Dr. TANCHOU and Dr. R. M. NEMBA (MINREST) and to
Dr. D. KOULAGNA (MINEF). In Cameroon, C. WILD (CRES Cameroon, Zoological Society of San
Diego) supported us with logistics. Many thanks to all our local assistants in Nyasoso, Nguengue
and Ekomtolo. Prof. W. BÖHME kindly verified the determination of the specimens and helped with
identification in difficult cases. P.A. HERRMANN, Dr. L. BECK and Dr. K. SOBJINSKI critically
reviewed the manuscript.
Habitatwahl und zeitliche Aktivitätsmuster von Anuren in einem Montan- und
Submontan-Regenwald im südwestlichen Kamerun – erste Ergebnisse
In Rahmen einer Untersuchung zur Anurenökologie im Südwesten Kameruns wurden vom 21.
bis 26. November 2000 (Beginn der Trockenzeit) sowie vom 31. März bis 5. April 2001 (Beginn
der Regenzeit) „visual encounter surveys“ entlang von Transekten in der Umgebung von Ekomtolo
(4°51’N, 9°54’E, 450 m ü.NN) am Mt. Nlonako durchgeführt. Vom 24. bis 27. März 2001 wurden
Transekte in der Umgebung von Nguengue (4°55’N, 9°59’E) in 1140 m Höhe bearbeitet. Bei der
qualitativen und quantitativen Arten-Erfassung wurden neben der Höhen- und der saisonalen
Verteilung auch Habitatwahl (Bach-/Waldtransekte), Aktivitätsmuster (tag-/nachtaktiv) und Sub-
stratpräferenz für Sitzwarten untersucht.
Die 456 nachgewiesenen Anurenindividuen verteilten sich dabei auf insgesamt 46 Arten, von
denen 54 % (n = 25) mit weniger als vier Individuen vertreten waren. Dominierende Art war
Phrynobatrachus auritus (36 %), gefolgt von Amnirana asperrima (8 %) und Phrynodon cf.
sandersoni (7 %), Dimorphognathus africanus (5 %) sowie Amnirana amnicola (5 %). Ranidae
machten 69 % aller Funde aus.
16 der in Ekomtolo nachgewiesenen Arten wurden ausschließlich zu Beginn der Regenzeit
gefunden; ein mögliches Indiz für eine saisonale Reproduktionsabhängigkeit. Bei Arten, die in
beiden Untersuchungszeiträumen nachgewiesen wurden, zeigten nur Phrynobatrachus auritus
(Beginn der Trocken-/Regenzeit: 37/127), Dimorphognathus africanus (3/19), Hyperolius ocel-
latus (10/2) und Arthroleptis adelphus (6/1) einen deutlichen Unterschied in der Individuenanzahl.
Fig. 18. Cardioglossa leucomystax female.
Cardioglossa leucomystax-Weibchen.
Fig. 19. Leptopelis boulengeri male. Photo: H.-
Leptopelis boulengeri-Männchen.
Anuran habitat selection and temporal partitioning in Southwestern Cameroon – first results
SALAMANDRA, Rheinbach, 40(3/4), 2004
Einige Arten, wie zum Beispiel Phrynobatrachus cricogaster (Nguengue/Ekomtolo: 12/0) und
Phrynodon cf. sandersoni (29/1), wurden ausschließlich beziehungsweise fast ausschließlich bei
Nguengue gefunden und können als Montanarten bezeichnet werden. Andere Arten wurden fast
ausschließlich bei Ekomtolo gefunden und zeigen damit eine eher submontane Verbreitung;
Beispiele sind Phrynobatrachus auritus (1/164) und Dimorphognathus africanus (1/22).
Die Mehrheit der nachgewiesenen Arten war nachtaktiv (z. B. Hyperolius ocellatus, Chiro-
mantis rufescens, Astylosternus diadematus und alle Leptopelis-Arten mit Ausnahme von L.
brevirostris). Bei einer Anzahl von Arten lagen sowohl Nacht- als auch Tagfunde vor. Am Tag
gefundene Individuen wurden allerdings häufig in Verstecken gefunden und dann als „nicht
tagaktiv“ interpretiert.
Viele Arten, die ausschließlich oder überwiegend im Waldhabitat nachgewiesen wurden, haben
wahrscheinlich eine gewässerunabhängige Reproduktion ohne ein freies Larvenstadium (z. B.
Arten der Gattung Arthroleptis, Nectophryne und Leptopelis brevirostris). Zwei Astylosternus
montanus, eine Art, die in Bächen oder temporären Gewässern ablaicht, wurden Ende März im
Waldtransekt gefundenen. Dies kann ein Hinweis dafür sein, dass die Reproduktionsphase bereits
beendet war und wahrscheinlich am Anfang der Trockenzeit liegt. Ein verstärktes Auftreten von
Leptopelis boulengeri im Bachtransekt im März/April lässt eine Reproduktion zu Beginn der
Regenzeit vermuten.
Unabhängig vom Habitat wurde Vegetation in den meisten Fällen als Sitzwarte bevorzugt. Von
den zehn ausschließlich am Boden gefundenen Arten (22 %) sind sechs nur mit einem Individuum
vertreten. Hinzu kommt, dass bei neun der 15 Arten, die sowohl am Boden als auch in der
Vegetation gefunden wurden, eine enge Korrelation zwischen Substratwahl und Aktivitätsmuster
vorlag. So befanden sich alle am Tage gefundenen Individuen von Arthroleptis adelphus, A.
variabilis, Cardioglossa leucomystax, Petropedetes cameronensis und Phrynodon cf. sandersoni
in Verstecken, wie zum Beispiel in Spalten, unter Steinen oder im Laub. Amnirana asperrima und
Conraua crassipes zeigten keine ausgeprägte Substratpräferenz. Eine „nearest neighbor“-Cluster-
analyse mit den Variablen Höhenverteilung, Habitat, Aktivitätsmuster und Substrat ergab eine
einzige Artengruppe mit Ausnahme von Bufo latifrons. Dies spricht für eine unerwartet große
ökologische Ähnlichkeit der Arten.
Schlagwörter: Anura: Amphibia; Mt. Nlonako; Südwest-Kamerun; Artendiversität; räumliche
Einnischung; zeitliche Aktivitätsmuster; Sitzwarten.
Fig. 20. Nearest neighbor cluster analysis in-
cluding all species with n = 4 specimens for the
variables low/high altitude, creek/forest habitat,
nocturnal/diurnal activity and vegetation/
ground perching sites (for data matrix see
Appendix 2). Vegetation/ground perching sites
data corrected for day retreats (see text).
„Nearest neighbor“-Clusteranalyse mit allen
Arten mit n = 4 Individuen und den Variablen
niedrige/hohe Höhenverteilung, Bach-/Wald-
habitat, Nacht-/Tagaktivität und Vegetation-/
Bodensubstratsitzwarten (für die Datenmatrix
siehe Appendix 2). Vegetation-/Bodensub-
stratsitzwarten-Werte korrigiert für Tagver-
stecke (siehe Text).
SALAMANDRA, Rheinbach, 40(3/4), 2004
ALFORD, R.A. & S.J. RICHARDS (1999): Global amphibian declines: A problem in applied ecology.
– Annu. Rev. Ecol. Syst., 30: 133-165.
AMIET, J.-L. (1971): Les Batraciens orophiles du Cameroun. – Ann. Fac. Sci. Cameroun, Yaoundé,
5: 57-81.
— (1975): Ecologie et distribution des amphibiens anoures de la région de Nkongsamba
(Cameroun). – Ann. Fac. Sci. Yaoundé, Cameroun, 20: 33-107.
— (1977): Les Astylosternus du Cameroun (Amphibia, Anura, Astylosterninae). – Ann. Fac. Sci.
Yaoundé, Cameroun, 23-24: 99-227.
— (1978): Liste provisoire des amphibiens anoures du Cameroun. – Laboratoire Zool. Fac. Sci.
Yaoundé, Cameroun: 1-21.
— (1980): Révision du genre Leptodactylodon Andersson (Amphibia, Anura, Astylosterninae).
– Ann. Fac. Sci. Yaoundé, Cameroun, 27: 69-224.
— (1989): Quelques aspects de la biologie des Amphibiens Anoures du Cameroun. – Ann. Biol.
28: 73-136.
— (1991): Un Leptopelis méconnu de la faune forestière camerounaise (Amphibia, Anura,
Hyperoliidae). – Alytes, 9: 89-102.
— & J.-L. PERRET (1969): Contributions à la faune de la région de Yaoundé (Cameroun). II. –
Amphibiens, Anoures. – Ann. Fac. Sci. Yaoundé, Cameroun, 3: 117-137.
BARBAULT, R. & M. TREFAUT RODRIGUES (1979): Observations sur la reproduction et la dynamique
des populations de quelques anoures tropicaux III. – Arthroleptis poecilonotus. – Trop. Ecol.
20: 64-77.
BLAUSTEIN, A.R., D.B.WAKE & W.P. SOUSA (1994): Amphibian declines: Judging stability,
persistence, and susceptibility of populations to local and global extinctions. – Conservation
Biology, 17: 60-71.
BÖHME, W. (1975): Zur Herpetofaunistik Kameruns, mit Beschreibung eines neuen Scinciden. –
Bonn. zool. Beitr., 26: 2-48.
— & B. SCHNEIDER (1987): Zur Herpetofaunistik Kameruns (III) mit Beschreibung einer neuen
Cardioglossa (Anura: Arthroleptidae). – Bonn. zool. Beitr., 38: 241-263.
EISENTRAUT, M. (1963): Die Wirbeltiere des Kamerungebirges. – Hamburg (Parey Verlag), 353
— (1973): Die Wirbeltierfauna von Fernando Po und Westkamerun. – Bonn. zool. Monogr., 3:1-
FISCHER E. & H. HINKEL (1992): Natur Ruandas, La Nature du Rwanda; Einführung in die Flora
und Fauna Ruandas, Apercu sur la flore et faune rwandaises. – Mainz (Materialien zur
Partnerschaft Rheinland-Pfalz/Ruanda), 437 S.
FROST, D.R. (2002): Amphibian Species of the World: an online reference. V2.21 (15 July 2002).
– Electronic database available at
GARTSHORE, M.E. (1986): The status of the montane herpetofauna of the Cameroon highlands. –
pp. 1-263 in STUART, S.N. (ed.): Conservation of Cameroon montane forests. – Cambridge
(International Council for Bird Preservation, 1986 (i-iii)).
HERRMANN, H.-W., W. BÖHME & P. HERRMANN (2000): The ALSCO Cameroon expedition 1998:
The sampling of a mountain rainforest. – Proceedings of the 4th International Symposium of
Isolated Vertebrate Communities in the Tropics. – Bonn. zool. Monogr., 46: 95-103 .
HOULAHAN, J.E., C.S. FINDLAY, B.R. SCHMIDT, A.H. MEYER & S.L. KUZMIN (2000): Quantitative
evidence for global amphibian population declines. – Nature, 404: 752-755.
Anuran habitat selection and temporal partitioning in Southwestern Cameroon – first results
SALAMANDRA, Rheinbach, 40(3/4), 2004
JAEGER, R.G. (1980): Density-dependent and density-independent causes of extinction of a
salamander population. – Evoloution, 34: 617-621.
JOGER, U. (1982): Zur Herpetofaunistik Kameruns (II). – Bonn. zool. Beitr., 33: 313-342.
LARGEN, M.J. & F. DOWSETT-LEMAIRE (1991): Amphibians (Anura) from the Kouilou River basin,
République du Congo. – Tauraco Research Report, 4: 145-168.
LAWSON, D.P. (1993): The reptiles and amphibians of Korup National Park Project, Cameroon.
– Herp. Nat. Hist., 1: 27-90.
MERTENS, R. (1938): Herpetologische Ergebnisse einer Reise nach Kamerun. – Abh. Senck. naturf.
Ges., Frankfurt/Main, 442: 1-52.
— (1939): Über das Höhenvorkommen der Froschlurche am Großem Kamerun-Berge. – Abh.
Ber. Mus. Naturk. Vorgesch. Magdeburg, 3: 121-128.
— (1940): Amphibien aus Kamerun, gesammelt von M. KÖHLER und Dr. H. GRAF. – Sencken-
bergiana, 22: 103-135.
— (1968): Zur Kenntnis der Herpetofauna von Kamerun und Fernando Poo. – Bonn. zool. Beitr.
19: 69-84.
MONARD, A. (1951): Résultats de la mission zoologique Suisse au Cameroun. – Mém. Inst. fr. Afr.
noire, Dakar, sér. Sci. nat., 1: 123-170.
Declining amphibian populations: The problem of separating human impacts from natural
fluctuations. – Science, 253: 892-895.
PERRET, J.-L. (1959): Études herpétologiques africaines. – Bull. Soc. neuchâtel. Sci. Nat., 82: 247-
— (1960a): Études herpétologiques africaines II. – Bull. Soc. neuchâtel. Sci. Nat., 83: 93-100.
— (1960b): Batraciens du genre Afrixalus du Cameroun. – Rev. Zool. Bot. Afr., 61: 366-377.
— (1961): Études herpétologiques africaines III. – Bull. Soc. neuchâtel. Sci. Nat., 84: 130-138.
— (1962): La biologie d’Acanthixalus spinosus (Amphibia Salientia). – Rech. Ètudes Cameroun,
1: 90-101.
— (1966): Les Amphibiens du Cameroun. – Zool. Jb. Syst., 8: 289-464.
— (1977): Les Hylarana (Amphibiens, Ranidés) du Cameroun. – Rev. Suisse Zool., 84: 841-868.
— & J.-L. AMIET (1971): Remarques sur les Bufo (Amphibiens, Anoures) du Cameroun. – Ann.
Fac. Sci. Yaoundé, Cameroun, 5: 47-55.
— & R. MERTENS (1957): Étude d’une collection herpétologique faite au Cameroun de 1952 à 1955.
– Bull. Inst. fond. Afr. noire,19, Dakar, sér. A: 548-601.
POYNTON, J.C. (1999): Distribution of amphibians in sub-saharan Africa, Madagascar, and
Seychelles. – pp. 483-539 in DUELLMAN, W.E. (ed.): Patterns of distribution of amphibians –
a global perspective. Baltimore (John Hopkins University Press).
RÖDEL, M.O. (2000): Herpetofauna of West Africa Vol. I. Amphibians of the West African
savanna. – Frankfurt/Main (Edition Chimaira), 332 S.
SANDERSON, I.T. (1936): The amphibians of the Mamfe Division, Cameroons – ecology of the frogs.
– Proc. Zool. Soc. Lond., 1936: 165-208.
SCHEEL, J.J. (1970): Notes on the biology of the African Tree-toad, Nectophryne afra BUCHHOLZ
& PETERS, 1875, (Bufonidae, Anura) from Fernando Poo. – Rev. Zool. Bot. Afr., 81: 225-
SCHIØTZ, A. (1999): Treefrogs of Africa. – Frankfurt/Main (Edition Chimaira), 350 S.
SCHMITZ, A. (1998): Systematik und Zoogeographie der Herpetofauna einer montanen Regenwald-
region in Kamerun. – Unpublished M.Sc. thesis, University Bonn. 261 S.
SALAMANDRA, Rheinbach, 40(3/4), 2004
Appendix 1. ZFMK numbers of specimens collected
Species are listed in alphabetical order. Numbers after a slash indicate number of specimens under
one collection number.
Acanthixalus spinosus (1 specimen): 78313. Amnirana amnicola (15 specimens): 78314,
78316-7, 78319-20, 78322-29, 78331-2. A. asperrima (21 specimens): 78321, 78333-45, 78346/
2, 78348-52. Arthroleptis adelphus (6 specimens): 78417-21, 78657. A. adolfifriederici
(1 specimen): 78405. A. poecilonotus (5 specimens): 78424-5, 78427, 78431-2. A. variabilis
(15 specimens): 78406-15, 78434, 78436, 78654-6. Astylosternus diadematus (4 specimens):
78303-6. A. fallax (1 specimen): 78310. A. montanus (1 specimen): 78308. A. perreti (1 specimen):
78309. Bufo latifrons (3 specimens): 78221-3. Cardioglossa elegans (1 specimen): 78256. C.
gracilis (1 specimen): 78255. C. leucomystax (12 specimens): 78257-9, 78261-2, 78264-70. C.
venusta (1 specimen): 78253. Chiromantis rufescens (6 specimens): 78245-50. Conraua
crassipes (7 specimens) : 78273-8, 78280. Dimorphognathus africanus (22 specimens): 78433,
78435, 78437/2, 78439-42, 78444-52, 78454-58. Hyperolius cf. sylvaticus nigeriensis
(3 specimens): 78381-3. H. kuligae (1 specimen): 78367. H. ocellatus ocellatus (12 specimens):
78368-76, 78378-80. Leptodactylodon boulengeri (1 specimen): 78287. L. ovatus orientalis
(2 specimens): 78284, 78423. Leptopelis boulengeri (3 specimens): 78399-401. L. brevirostris
(6 specimens): 78392-6, 78398. L. calcaratus (6 specimens): 78386-91. L. cf. millsoni
(2 specimens): 78403-4. L. omissus (2 specimens): 78311-2. L. rufus (1 specimen): 78402.
Nectophryne afra (2 specimens): 78229-30. N. batesii (6 specimens): 78231-3, 78235, 78237-8.
Opisthothylax immaculatus (2 specimens): 78384-5. Petropedetes cameronensis (8 specimens):
78355-9, 78361-3. P. cf. newtoni (1 specimen): 78354. P. parkeri (1 specimen): 78366.
Phrynobatrachus auritus (133 specimens): 78286, 78351-7, 78426, 78505, 78507-24, 78527-30,
78538-59, 78561-84, 78586-601, 78603-25, 78627-41, 78653. P. batesi (2 specimens): 78503-
4. P. cricogaster (11 specimens): 78460-7, 78469-71. Phrynodon cf. sandersoni (27 specimens):
78473-93, 78495-9, 78501. Scotobleps gabonicus (11 specimens): 78288-94, 78296, 78299-301.
Werneria mertensiana (3 specimens): 78243, 78251, 78253. Wolterstorffina parvipalmata
(3 specimens): 78240-2). Xenopus fraseri (1 specimen): 78220.
Appendix 2. Data matrix for cluster analysis
Vegetation/ground perching sites data corrected for day retreats (see text).
Manuscript received: 27 May 2003
Authors: MIRCO PLATH & MARKUS S OLBACH, Philipps-Universität Marburg, Fachbereich Biologie,
Fachgebiet Systematische Zoologie, Karl-von-Frisch-Straße, D-35032 Marburg/Lahn, Germany,
E-Mail:; HANS-WERNER HERRMANN, Center for Reproduction of Endangered
Species (C.R.E.S.), Zoological Society of San Diego, P.O. Box 120551, San Diego, CA 92112-
0551, U.S.A., E-Mail:
... Kupe, Mt. Oku, and Tchabal Mbabo (Amiet 1975;Hofer et al. 1999Hofer et al. , 2000Plath et al. 2004;Herrmann et al. 2005aHerrmann et al. , 2005bHerrmann et al. 2006;Gonwouo et al. 2007;Ineich et al. 2015;Hirschfeld et al. 2016). Through this work, elevational distribution and biogeographic patterns have emerged, though information for many regions is incomplete or absent. ...
... Inventories of the herpetofauna of Mt. Nlonako, including sites ranging from 400-1,700 m elevation, resulted in the documentation of 93 amphibian and 89 reptile species, making it one of the most species-rich mountains in tropical Africa (Amiet 1975;Plath et al. 2004;Herrmann et al. 2005aHerrmann et al. , 2005b. Several distinctive anuran faunas converge in this region (Amiet 1975), partially explaining the high species richness. ...
Full-text available
We performed surveys at several lower elevation sites surrounding Mt. Kupe, a mountain at the southern edge of the Cameroonian Highlands. This work resulted in the sampling of 48 species, including 38 amphibian and 10 reptile species. By combining our data with prior survey results from higher elevation zones, we produce a checklist of 108 species for the greater Mt. Kupe region including 72 frog species, 21 lizard species, and 15 species of snakes. Our work adds 30 species of frogs at lower elevations, many of which are associated with breeding in pools or ponds that are absent from the slopes of Mt. Kupe. We provide taxonomic accounts, including museum specimen data and associated molecular data, for all species encountered. Finally, we compare the levels of biodiversity of Mt. Kupe to other regions, discuss biogeographic ties to other montane systems, and note current conservation threats.
... The species is known from lowland localities, but also reaches altitudes above 1000 m (e.g. Mertens 1965;Plath et al. 2004). Lawson (1993) mentions a specimen from 1769 m from a locality in the Rumpi Hills, but remarked that it has been assigned to this species with reservation. ...
... Assignments before official naming of species: Euskirchen et al. (1999): referred to as P. parkeri Herrmann et al. (2000): referred to as P. parkeri Plath et al. (2004): referred to as P. parkeri Herrmann et al. (2005): referred to as P. parkeri Petropedetes perreti Amiet, 1973 Assignments before official naming of species, problematic assignment: Perret & Mertens (1957): as P. palmipes Lawson (1993): P. perreti from Rumpi Hills, locality unsure Appendix II. Gazeteer ...
We review and summarize the present knowledge on Central African Petropedetes, discuss their taxonomy and describe three new species. We synonymise one name: P. newtonii is a junior synonym of P. johnstoni. Frogs from the African mainland, namely Cameroon and eastern Nigeria, assigned to “P. newtonii” in more recent literature, were misidentified. They are herein described anew as Petropedetes vulpiae sp. nov. Two further new taxa are morphologically similar to other described species. Petropedetes euskircheni sp. nov. is morphologically close to P. parkeri but differs foremost in the smaller size of the femoral glands in adult males. P. juliawurstnerae sp. nov. is morphologically similar to P. perreti and P. cameronensis. It can be distinguished from P. perreti by its less developed webbing and from P. cameronensis by the size of its tympanum and the occurrence of a tympanic papilla in males. A key to adult males of the currently known Central African Petropedetes species is given and an outlook on potential further new taxa is provided.
... Amphibian inventories in African tropical rainforest habitats are a first step. Quantitative ecological studies have only recently begun (PLATH 2003, SOLBACH 2003, PLATH et al. 2004. Ultimately, in-depth species studies would complete the picture. ...
... Transects of several hundred meters in length were surveyed along a creek between Ekomtolo and Badjong for several months during the dry and the rainy season. For details see PLATH 2003, SOLBACH 2003, and PLATH et al. 2004. ...
... Sierra Leone, through Guinea and Liberia to western Ivory Coast and isolated populations in western Ghana (Barbour & Loveridge 1927, 1930Guibé & Lamotte 1958;Lamotte 1971;Hughes 1988;Rödel & Branch 2002;Rödel 2003;Hillers et al. 2008); C. derooi occurring in southern-central and south-eastern Ghana and adjacent Togo (Schiøtz 1964;Hulselmans 1972;Segniagbeto et al. 2007Segniagbeto et al. , 2017Hillers et al. 2009;; C. crassipes (Buchholz & Peters, 1875) with records from eastern Nigeria through Cameroon into the Democratic Republic of Congo and Gabon (Perret 1966, Amiet 1990, Frétey & Blanc 2000, Lötters et al. 2001, Burger et al. 2004; C. goliath from south-western Cameroon into Equatorial Guinea (Perret 1966, Gartshore 1984, Frétey & Blanc 2000, Schäfer et al. 2019; C. robusta Nieden, 1908 confined to mountains of western Cameroon and adjacent south-eastern Nigeria (Perret 1966, Gartshore 1984, Lawson 1993); and C. beccarii (Boulenger, 1911) known from mountain streams in Ethiopia and Eritrea (Largen & Spawls 2010). In western Cameroon, C. goliath, C. robusta, and C. crassipes can occur in sympatry or even syntopy (Gartshore 1984, Lawson 1993, Plath et al. 2004, Herrmann et al. 2005), but all other species have distinct non-overlapping ranges. ...
Full-text available
Forty-nine years after the last description of a slippery frog, we describe a seventh species of the genus Conraua. The new Conraua is endemic to the Atewa Range Forest Reserve, central Ghana, and is described based on genetic, bioacoustics, and morphological evidence. Recent molecular phylogenetic and species delimitation analyses support this population as distinct from nominotypical C. derooi in eastern Ghana and adjacent Togo. The new species is sister to C. derooi, from which it differs ~4% in the DNA sequence for mitochondrial ribosomal 16S. Genetic divergences in 16S to other species of Conraua range from 4–12%. The new species is distinguished morphologically from its congeners, including C. derooi, by the combination of the following characters: medium body size, robust limbs, lateral dermal fringing along edges of fingers, cream ventral color with brown mottling, the presence of a lateral line system, indistinct tympanum, the presence of inner, outer, and middle palmar tubercles, and two subarticular tubercles on fingers III and IV. We compare the advertisement calls of the new species with the calls from C. derooi and find that they differ by duration, frequency modulation, and dominant frequency. We discuss two potential drivers of speciation between C. derooi and the new species, including river barriers and fragmentation of previously more widespread forests in West Africa. Finally, we highlight the importance of the Atewa Range Forest Reserve as a critical conservation area within the Upper Guinean biodiversity hotspot.
... Concerning the family Bufonidae, parental care for both tadpoles and eggs has been reported only for Nectophryne afra, Dendrophryniscus brevipollicatus and Frostius pernambucensis, (Scheel 1970;Crump 1996;Dias et al. 2016;Malagoli et al. 2017). All three are phytothelmatous species, and while N. afra occurs in the rainforests of Midwest Africa (Plath et al. 2004;Pauwels and Rödel 2007), D. brevipollicatus and F. pernambucensis are endemic to the Atlantic Rainforest domain in Brazil (Dias et al. 2016;Malagoli et al. 2017). Within the bufonids that exhibit parental care only to eggs, Nectoprhyne malcomi and Oreoprhrynella nigra, the males often perform parental egg care while attracting more females to breed, forming communal nests (Grandison 1978;Mcdiarmid and Gorzula 1989), where both species display a terrestrial reproductive mode and direct development, while Altiphrynoides malcomi produces endotrophic tadpoles that develop out of the water. ...
Parental care is a behavioural repertoire exhibited by parents to their offspring to ensure their survival. In anurans, parental care may be optional, maternal, paternal, alloparental or biparental, towards eggs, larvae and/or juveniles. Frostius pernambucensis is a bufonid species exclusively endemic to the Brazilian Atlantic Rainforests of northeastern Brazil, occurring in the states of Paraiba, Pernambuco, Alagoas and Bahia. This study describes the behavioural parental care repertoire displayed by F. pernambucensis at the Dois Irmãos State Park (Parque Estadual Dois Irmãos – PEDI), the species type locality, in the municipality of Recife, Pernambuco. Fieldwork was carried out between June 2015 and March 2017, in which parental care was observed for three F. pernambucensis males. In total, 45 behavioural acts were recorded (n = 526), distributed in six subcategories: agonistic behaviour (n = 228; 43.35%), vigilance (n = 92; 17.49%), offspring manipulation (n = 78; 14.82%), dislocation (n = 68; 12.93%), vocalisation (n = 51; 9.70%), and maintenance behaviours (n = 9; 1.71%). Descriptive parental care records for Brazilian amphibians are considered rare, emphasising the importance of the present study, which provides relevant information about this poorly known taxon.
... This Central African treefrog has a characteristically short and vertically truncated snout, a tooth-like process at the centre of the mandible, and a relatively small tympanum (Amiet 2012). Leptopelis brevirostris is thought to reproduce by direct development (Perret 1966;Amiet and Schiøtz 1974), because females deposit large eggs in the soil of the forest floor, far from bodies of water (Schiøtz 1999;Plath et al. 2004;, and unlike most species in the genus, L. brevirostris eats terrestrial gastropods (Schiøtz 1999). Several morphological features vary in this species, including the pectoral glands in males (present or absent), dorsal colouration (uniformly green, brown, beige, grey, or a combination of these forming an irregular pattern), the shape (rounded or elliptical) and orientation (oblique or vertical) of the tympanum, the distinctiveness of the tympanic annulus, and the extent that the tympanum is obscured by granulations of the surrounding integument (Boulenger 1906;Schiøtz 1999;Amiet 2012). ...
Full-text available
The Central African treefrog Leptopelis brevirostris has a characteristically short and vertically truncated snout, a tooth-like process at the centre of the mandible, and unlike most species of Leptopelis, eats terrestrial gastropods. Two morphologically similar species (L. crystallinoron and L. brevipes) have been described in sympatry with L. brevirostris, each on the basis of a single female specimen and then provisionally synonymised with L. brevirostris by subsequent authors. Here we re-examine the holotypes of the three species and collect both genetic (mtDNA and 25 nuDNA loci) and morphological data across the range of L. brevirostris and its sister species L. notatus to test the hypothesis that L. brevirostris is comprised of more than one species. We found that the diagnostic characters for L. crystallinoron (lacking a visible tympanum and having larger vomerine teeth in contact) and L. brevipes (distinctive dorsal colouration) are also present in L. brevirostris and L. notatus. Furthermore, these characters vary among individuals that share the same mitochondrial haplotype and that occur within a single locality, including the type locality of L. crystallinoron. Although all samples from the type locality of L. crystallinoron form a distinct mitochondrial lineage relative to L. brevipes and L. brevirostris, our analysis of 25 nuclear loci does not support this relationship and instead suggests that this mitochondrial divergence reflects phylogeographic structure across the range of L. brevirostris, including samples from the type localities of L. crystallinoron and L. brevipes. By contrast, the mitochondrial divergence between L. notatus and L. brevirostris is strongly supported in the nuDNA dataset. Consequently, we recognise both L. brevipes and L. crystallinoron as synonyms of L. brevirostris. Given the extensive variation in the colour pattern, the tympanic membrane, and the size and shape of the vomerine teeth among L. brevirostris collected from one locality, variation in these traits should be interpreted with caution in taxonomic research on Leptopelis.
... Most species are found in Cameroon, with many species restricted to small altitudinal zones of mountainous regions in the west of the country. There, they inhabit a variety of forests and varying stream types (Mertens 1938;Amiet 1989;Oates et al. 2004;Plath et al. 2004;Herrmann et al. 2005). In contrast, the widespread A. batesi (Boulenger, 1900) and A. occidentalis Parker, 1931 prefer humid lowland forests, where their tadpoles are often found in small, shallow, slow flowing creeks (Amiet 1989;Rödel et al. 2012). ...
Full-text available
Herein, we describe the tadpoles of six Astylosternus species, A. fallax, A. cf. fallax, A. laurenti, A. montanus, A. perreti, A. ranoides, and Scotobleps gabonicus, and redescribe the tadpoles of A. batesi, A. diadematus, A. laticephalus, A. occidentalis, A. rheophilus, and Nyctibates corrugatus. All Astylosternus tadpoles are adapted to torrent currents and share a long, oval body, slightly flattened in lateral view, with very long muscular tails with narrow fins. The jaws are massive, serrated, and often show a tooth-like medial projection (fang) in the upper jaw. Body proportions of Astylosternus tadpoles are extremely similar. The best characters to distinguish species might be life coloration and potentially the shape of labial papillae. The tadpole of Scotobleps gabonicus is similar to Astyloster-nus and differs only slightly by a narrower body with a shorter and rounder head. The upper jaw of Scotobleps carries two or three lateral fangs instead of one medial one. The tadpole of Nyctibates corrugatus is easily distinguishable from the other two genera on the basis of their very long, eel-shaped body and tail and the bluish-black color.
Amphibienpopulationen sind weltweit bedroht. Für fundierte Entscheidungen im Naturschutz ist ein wissenschaftliches Hintergrundwissen notwendig. Eine wichtige Komponente ist die Verbreitung der Arten und die Gründe hierfür. Dies setzt auch Klarheit bezüglich des taxonomischen Status voraus. Vernachlässigte Regionen liegen meist in den Tropen. Um diese Lücke zu füllen, untersucht die vorliegende Arbeit die Makroökologie westafrikanischer Amphibien. Eine neue Art der Familie Phrynobatrachidae wird beschrieben. Anschließend werden die taxonomischen Unterschiede zweier Mitglieder der Familien der Hyperoliidae und der Arthroleptidae untersucht. Dies ebnet den Weg für die Makroökologie. Es wird analysiert, ob Westafrika eine einzigartige biogeographische Region ist. Die Untersuchung zeigen, dass Westafrika in der Tat einzigartig ist. Die Ähnlichkeiten innerhalb der Region sind größer als die innerhalb ähnlicher Habitate aus unterschiedlichen Regionen. Der Cross River ist die wichtigste Barriere. Mehrere geographische Zonierungen innerhalb Westafrikas werden entdeckt. Es wird untersucht, ob der Chytrid Pilz, Batrachochytrium dendrobatidis, (Bd), ein wichtiger Faktor beim Rückgang der Amphibien Populationen, in West Afrika vorkommt. Bisher gibt es keinen positiven Nachweis westlich Nigerias. Dennoch sagen Modelle geeignete Habitate für Bd vorher. Die wahrscheinlichste Erklärung ist, dass die Dahomey Gap als natürliche Barriere die Ausbreitung von Bd verhindert. Als letztes werden die Nischen westafrikanischer Amphibien eruiert. Für die meisten Arten werden Nischenmodelle berechnet. Dies bestätigt bereits bekannte Gebiete hoher Alpha Diversität und zeigt bisher unbekannte Gebiete auf. Des Weiteren werden Erklärungen für unterschiedliche Verbreitungsgebietsgrößen gesucht. Generell wird die Nischenbreite hierfür verantwortlich gemacht. Die vorliegenden Daten lassen allerdings auch den Schluss zu, dass das Ausbreitungsvermögen das beobachtete Muster ebenfalls erklären kann.
Torrent frogs of the genus Petropedetes Reichenow, 1874 as currently understood have a disjunct distribution with species endemic to West, Central or East Africa. We herein present a phylogenetic analysis including all but one of the currently described 12 species of the genus. Maximum Likelihood and Bayesian analyses of combined nuclear (rag1, SIA, BDNF) and mitochondrial (16S, 12S, cytb) genes of more than 3500 base pairs, revealed clades corresponding to the three sub-Saharan regions. Molecular results are confirmed by morphological differences. Surprisingly, the three geographic clades do not form a monophyletic group with respect to closely related families Pyxicephalidae and Conrauidae and therefore require taxonomic changes. We resurrect ArthroleptidesNieden, 1911 for the East African taxa. The Central African taxa remain in the genus Petropedetes. The West African members are placed in the new genus Odontobatrachus gen. nov. The phylogenetic position of the new genus remains incertae sedis as it was not assigned to any of the four families included in our analyses. Potential new species have been detected within all three major clades, pointing to a still not fully clarified diversity within African torrent frogs.
Full-text available
Extinctions are normal biological phenomena. Both mass extinctions in geological time and local extinctions in ecological time are well documented, but rates of extinction have increased in recent years—especially in vertebrates, including amphibians—as illustrated by recent reports of their population declines and range reductions. We suggest that long-term population data are necessary for rigorously evaluating the significance of the amphibian declines. Due to the physiological constraints, relatively low mobility, and site fidelity of amphibians, we suggest that many amphibian populations may be unable to recolonize areas after local extinction. Las extinciones son un fenómeno biológico normal. Extinciones en masa en una escala temporal geológica y extinciones locales en una escala temporal ecológica, están bien documentadas, pero en años recientes las tasas de exinción han incrementado, especialmente en vertebrados, incluyendo a los anfibios tal como ha sido ejemplificado en reportes recientes sobre la declinación de su población y la reducción de su área de distribución. Nosotros sugerimos que datos poblacionales a largo plozo son necesarios para evaluar rigurosamente la significancia de la declinación en anfibios. Nosotros sugerimos que muchas de las poblaciones de anfibios son incapaces de recolonizar áreas después de extinciones locales debido a las restricciones fisiológicas, la relativamente baja movilidad y la filopatría de los anfibios.
Full-text available
Declines and losses of amphibian populations are a global problem with complex local causes. These may include ultraviolet radiation, predation, habitat modification, environmental acidity and toxicants, diseases, changes in climate or weather patterns, and interactions among these factors. Understanding the extent of the problem and its nature requires an understanding of how local factors affect the dynamics of local populations. Hypotheses about population behavior must be tested against appropriate null hypotheses. We generated null hypotheses for the behavior of amphibian populations using a model, and we used them to test hypotheses about the behavior of 85 time series taken from the literature. Our results suggest that most amphibian populations should decrease more often than they increase, due to highly variable recruitment and less variable adult mortality. During the period covered by our data (1951-1997), more amphibian populations decreased than our model predicted. However, there was no indication that the proportion of populations decreasing changed over time. In addition, our review of the literature suggests that many if not most amphibians exist in metapopulations. Understanding the dynamics of amphibian populations will require an integration of studies on and within local populations and at the metapopulation level.
Full-text available
Reports of declining amphibian populations in many parts of the world are numerous, but supporting long-term census data are generally unavailable. Census data from 1979 to 1990 for three salamander species and one frog species at a breeding pond in South Carolina showed fluctuations of substantial magnitude in both the size of breeding populations and in recruitment of juveniles. Breeding population sizes exhibited no overall trend in three species and increased in the fourth. Recent droughts account satisfactorily for an increase in recruitment failures. These data illustrate that to distinguish between natural population fluctuations and declines with anthropogenic causes may require long-term studies.
Quelques aspects de la biologie des Amphibiens Anoures du Cameroun
— (1989): Quelques aspects de la biologie des Amphibiens Anoures du Cameroun. – Ann. Biol. 28: 73-136.