2,094Grants to


Tree Frog (Leptopelis flavomaculatus)

Mohamed bin Zayed Species project number 0925565

Mohamed bin Zayed Species Conservation (Project No. 0925565) - Tree Frog - Awarded $5,000 on September 29, 2009


Leptopelis flavomaculatus (Günther, 1864) is a large East African lowland forest tree frog (Schiøtz, 1999). The dorsum is either uniform green with white heels or brown with a darker triangle with its blunt apex pointing forwards.  They have a dark bar between the eyes and a canthal line. Males have pectoral glands while juveniles often have yellow spots on the back and white heels. Leptopelis flavomaculatus is very similar to Leptopelis vermiculatus (Boulenger, 1909) but differs in lack of vivid green and vermiculated dorsum present in all juveniles and some males of Leptopelis vermiculatus. Again there is more webbing on the fourth toe of Leptopelis vermiculatus than Leptopelis flavomaculatus. The species is sporadically widely distributed from coastal Kenya southwards through Tanzania, Malawi, Zimbabwe and Mozambique (Schiøtz, 1999; Channing and Howell, 2006). Despite its wide distribution most of its breeding biology except the advertisement call is unknown (see Channing and Howell, 2006). In addition its conservation status in certain areas of its range is unknown and there are fears that such populations could be highly threatened. One such population is that of Kitobo forest, a poorly studied lowland arid land forest refuge. The present study therefore aimed to investigate this population by examining its population status, breeding habits and threats. The hypothesis was that the insularization of Kitobo forest could have negative effects on the long-term viability of Leptopelis flavomaculatus population.



Study area

Kitobo forest is a ground water forest located about 10 km South-East of Taveta town in the Taita-Taveta district, Coast Province, Kenya (Fig. 1). It is approximately 250 km inland from the Indian Ocean and on the extreme lowland North-East of the Tanzanian Eastern Arc Mountain block of North Pare Mountains near the Kenya-Tanzania border (point: S03o 26’ 45.7”; E 037o 36’ 39.7”). It covers an area of 160 ha at an altitude of ca. 730 m above sea level. It is largely an evergreen indigenous forest surrounded by arid lands of Acacia bushes. It owes its existence to the eruption on its edge of a large namely Njoro spring plus other small ones inside the forest originating from the volcanic Mt. Kilimanjaro to the north-east. The springs join and develop into a permanent stream that flow through the forest.  Bordering the forest on the southern and eastern sides are irrigation schemes that utilize water from these springs and other water canals growing rice, onions, maize, bananas, tomatoes, mangoes and citrus. Part of the north-eastern part of the forest is prone to frequent floods resulting from rains in Mt. Kilimanjaro.


Definition of the sampling habitats

1)      Forest

This is the evergreen forest within which occur natural swamps, springs, streams and ponds. Within the forest sampling was done in recently flooded and non-flooded areas to asses the effects of flooding on the frog abundance.

2)      Farms

These are the wetland habitats outside the forest including cultivated fields.


Frog sampling

Surveys were conducted from 4 to 15 December 2009 and again from 5 to 15 April 2010. Leptopelis flavomaculatus was sampled at night using Visual (VES) and acoustic encounter (AES) surveys as described by (Rödel and Ernst, 2004). The acoustic method was highly reliable because like many other species in the Genus Leptopelis, individuals are easily heard calling but difficult to locate (Channing and Howell, 2006). Each particular individual call was followed as much as possible to its source. Since only males call to advertise their position to potential mates and rivals, the results reflect this sexual bias as spotting females is merely by luck. During the sampling all potential amphibian micro-habitats inside and on the  forest edge were surveyed.

Selected individuals were kept as voucher specimens fixed in 10 % formalin after euthanasia. Tissues of selected specimens were preserved in absolute alcohol for the possibility of later molecular analyses. Tadpoles were fixed in 95 % ethanol and their labial tooth row formula (LTRF) later examined in the lab for specific species identification. All specimens collected are deposited at NMK. Colour photos of the species in wild habitat and their habitats were taken. GPS data were determined using a 12 Channel Garmin receiver.


Statistical analyses

One way analysis of variance (ANOVA) was used to test for the mean differences in individual abundance per sampling day across the three micro-habitat types (Flooded, non-flooded, farms habitats).  Data was analyzed with STATISTICA 6.0 software (StatSoft, 2001) at 5% significance level.



Species description, colour and size

The Yellow-spotted tree frog Leptopelis flavomaculatus from Kitobo forest is large in size. The body length of two males collected were 53 and 47 mm. Adults are brown in colour with a dark brown triangle in the centre with its apex at the back of the head. The juveniles are green with yellow spots.


Species abundance and habitat characteristics

Leptopelis flavomaculatus were found mainly in sites with dense Aframomum and Phoenix species dominated vegetation types inside the forest. These were mainly confined to sites with natural springs. In these micro-habitats individuals were found or heard calling from one meter and above from the ground. In most of the time individuals could be heard calling within scattered points but hardly spotted. In the same sites occurred other tree frogs such as Hyperolius puncticulatus (Pfeffer, 1893), Hyperolius glandicolor (Peters, 1879), Hyperolius tuberilinguis Smith, 1849, as well as terrestrial frogs like Ptychadena mascareniensis (Duméril & Bibron, 1841), and Phrynobatrachus acridoides (Cope, 1867).

In total 73 frogs were counted within the forest habitats (flooded and non-flooded)  and only three within farms. Those in farms were only in abandoned farms overgrown with reeds. During the December sampling when there was moderate rainfall and less flooding 19, 3 and 1 individuals were counted in the forest flooded, non-flooded and farm habitats respectively. In April, after heavy rainfall and more flooding 16, 35 and 2 individuals were counted within the same habitats respectively. In April, after 8 sampling nights in non-flooded and flooded and 7 sampling in farms; the mean number of individuals counted per sample was significantly high (ANOVA; F2, 20 = 8.44, n = 3, P = 0.002) in the forest non-flooded habitats than either flooded or farmlands. Inside the forest Leptopelis flavomaculatus was mostly associated with the Aframomum and Phoenix plant species and mostly within springs. However, these kinds of microhabitats make a very small proportion of the total habitat in forest.


Breeding habits

Leptopelis flavomaculatus is nocturnal and start calling as early as 7 pm. The males mostly call from dense and high vegetation making them hard to observe. The call is a soft drawn-out cry consisting of closely spaced harmonics. It was within these Aframomum and Phoenix plant species where water puddles potential for breeding existed. A single juvenile was collected within a recently flooded Aframomum zone.  For the first time the three tadpoles collected within one particular spring water puddles inside the turned to be of this species. Otherwise all the other tadpoles collected were of Hemisus marmoratus (Peters, 1854). The labial tooth row formula (LTRF) of these tadpoles is 4(2-4)/3.




Considering the April data that allowed comparison of non-flooded, flooded, and farm habitats sampling, the abundance of Leptopelis flavomaculatus individuals decreased (35:16:2) from the forest non-flooded through flooded to farm habitats. During the first December survey rainfall was not heavy and there had been one flooding. Most of the sites harboring this frog were within this flood zone in the forest. Following very heavy rainfall around March just prior to the April second survey there were subsequent flooding that destroyed all the previous preferred micro-habitats resulting to drying of Aframomum species through siltation and stagnation.  Most of the species were now scattered in new sites within the forest the major one being those with Phoenix (palm trees) springs (Aframomum 10: Phoenix 29). This clearly shows the negative effects of the flooding through destruction of species micro-habitats.




From the results almost all the Leptopelis flavomaculatus individuals occurred in the forest. The species was found mainly on wetland vegetation (Aframomum and Phoenix species) with those most preferred being within springs. These microhabitats make a very small proportion of the general habitat of forest. The major threat to this tree frog population is the frequent flooding that causes population fluctuation. The flooding effect exacerbates the threat of Leptopelis flavomaculatus habitat population isolation.  Kitobo forest is like an island in a sea of arid lands surrounded by intensive furrow irrigation schemes. Studies have observed that patch size, isolation and disturbance have effects on biodiversity. Patch isolation is a measure of habitat in the landscape surrounding the patch (Fahrig, 2003). As landscapes become increasingly fragmented, animal populations in the remaining habitat patches are isolated. Animals with patchy distribution or those, which utilize a range of microhabitats, are especially vulnerable to losses in mosaics of habitats. This is true of Leptopelis flavomaculatus which seem to be only restricted to forest habitat and few to none outside the forest in Kitobo forest.  Insularization has both ecological and genetic implications and reduced genetic variability has long been recognized as a feature of small isolated populations due to inbreeding, genetic drift, ‘the bottleneck’ and ‘founder effects’ (Spellerberg, 1991). This ultimately leads to extinction due continued reduction of natural habitat and dispersal (Marsh and Pearmanm, 1997; DeMaynadier and Hunter, 1998). The isolation of amphibian populations in fragmented habitats is because of their low vagility, high philopatry (strong site fidelity) and limited dispersal capabilities (Wind, 2001; Marsh and Pearman, 1997; Gardner, 2001; Bell & Donnelly 2006). In case of Leptopelis flavomaculatus in Kitobo forest, the fact is that outside the forest there is no suitable dispersal habitat as it is purely an arid land.

Floods have been identified as one of the causes of global amphibian population decline (Gardner, 2001). Studies have found varying direct and indirect effects of floods on amphibian communities (Cogalniceanu and Miaud, 2003, Maltchick et al., 2007; Moreira et al., 2008; Furlani et al., 2009). The direct effects include affecting the species breeding phenology while indirectly they destroy the breeding sites and all are species specific resulting in either population decline, fluctuation, increase or no change.   With reference to Leptopelis flavomaculatus in Kitobo forest this seem to result to the tree frog population abundance and distribution to fluctuate and may possibly in the long run cause decline or disappearance in certain sites. Flooding result to destruction of breeding sites (water puddles) through siltation or even washing away of eggs and tadpoles to unsuitable location outside the forest resulting to their death.

Details of breeding in Leptopelis flavomaculatus are unknown (Channing and Howell, 2006) except its advertisement call.  The results of this study present the first record and analysis of its tadpoles labial tooth row formula of 4(2-4)/3. Despite the frogs being hard calling, collection of tadpoles on the same sites yielded those of different species. This possibly shows that the species breeding is affected by environmental factors like floods among others. Future studies should therefore focus on intensively examining the species breeding behaviour to assess its population viability in the forest.



Many thanks go to Mr. Nusu Mwawasi the Kitobo forest Guard for his assistant in the field. Kenya Forest Service issued a permit to work in the forest. Thanks also go to the local community for permission to sample in their farmlands. Am very grateful to my field assistants from the National Museums of Kenya, Joash Nyamache and Vincent Muchai. This work was undertaken with the support of The Mohamed Bin Zayed Species Conservation Fund, project number 0925565.



Bell, K.E. and Donnelly, M.A. 2006. Influence of forest fragmentation on community structure of frogs and lizards in north-eastern Costa Rica. Conservation Biology,  20:1750-1760.

Channing,  A.  and Howell, K. M.  2006. Amphibians of East Africa. Cornell University Press and Edition Chimaira Press, Ithaca and Frankfurt.

Cogalniceanu, D.  and Miaud, C. 2003. Population age structure and growth in four syntopic amphibian species inhabiting a large river floodplain. Canadian Journal of  Zoology,  81(6):1096–1106  


DeMaynadier, P.G. and Hunter, M.L. 1998. Effects of silviculture edges on the distribution and abundance of amphibians in Maine. Conservation Biology. 12(2): 342-352.

Fahrig, L. 2003. Effects of habitat fragmentation on Biodiversity. Annual Reviews of Ecolology and  Systematics 34: 487-515.

Furlani, D., Fecetola, G.F., Colombo, G., Ugurlucan, M. and Bernardi, F.D. 2009.  Deforestation and the structure of frog communities in the Humedale Terraba-Sierpe, Costa Rica.  Zoological Sciences, 23:197-202.

Gardner, T. 2001. Declining amphibian populations: a global phenomenon in conservation biology. Animal Biodiversity and Conservation, 24(2):25-35.

Maltchik, L., Peixoto, C.D., Sternet, C., Moreira, L.F.B., and Machado, I.F. 2007. Dynamics of the terrestrial amphibian assemblage in a flooded riparian forest fragment in a Neotropical region in the south of Brazil. Brazilian Journal of Biology, 68(4): 763-769.

Marsh, D.M. and Pearman, P.B.  1997. Effects of habitat fragmentation on the abundance of two species of Leptocyclid frogs in an Andean montane forest. Conservation Biology 11(6): 132-138.

Moreira, L.F.B., Machado, I.F., Lace, A.R.G.M. and Maltchik, L. 2008. Anuran amphibians dynamics in an intermittent pond in Southern Brazil. Acta Limnology Brasil, 20(3):205-212.

Pearman, P.B. 1997. Correlates of amphibian diversity in an altered landscape of Amazonian Ecuador.  Conservation Biology 11(5): 1211-1225.

Rödel, M-O. and Ernst, R. 2004. Measuring and monitoring amphibian diversity in tropical forests. I. An evaluation of methods with recommendations for standardization. Ecotropica 10:1-14.

Schiøtz, A. 1999. Tree Frogs of Africa. Edition Chimaira, Frankfurt am Main.

Spellerberg, I.F. 1991. Monitoring Ecological Change. Cambridge University Press

Statsoft, 2001. STATISTICA: Data analysis software system, Version 6.0. StatSoft, Oklahoma.

Veith, M., Lötters, S., Andreone, F. and Rödel, M-O. 2004. Measuring and monitoring amphibians diversity in tropical forests. II. Estimating species richness from standardized transect censing. Ecotropica 10:85-99.

Wind, E. 2000. Effects of habitat fragmentation on amphibians: What do we know and where do we go from here? Pp. 885-893. In, L. M. Darling, editor. 2000. Proceedings of a conference on the Biology and Management of Species and Habitats at Risk, Kamloops, B. C., 15-19 Feb., Volume Two. B. C. Ministry of Environment, Lands and Parks, Victoria, B. C. and University College of the Cariboo, Kamloops, B.C. 520 pp.

Project 0925565 location - Kenya, Africa