Ecology and Demography of the Critically Endangered Kandian Torrent Toad Adenomus Kandianus: a Long-Lost Endemic Species of Sri Lanka

Ecology and demography of the Critically Endangered Kandian torrent toad Adenomus kandianus: a long-lost endemic species of Sri Lanka

S URANJAN K ARUNARATHNA,SUJAN H ENKANATHTHEGEDARA,DINESH G ABADAGE M ADHAVA B OTEJUE,MAJINTHA M ADAWALA and T HILINA D. SURASINGHE

Abstract The tropical island nation of Sri Lanka is a bio- Keywords Adenomus kandianus, Amphibia, Bufonidae, diversity hotspot with a high diversity and endemism of am- conservation, Critically Endangered, montane streams, phibians. The endemic, stream-dwelling Kandian torrent Sri Lanka, tropical rainforests toad Adenomus kandianus is Critically Endangered and was considered to be extinct until its rediscovery in . The species is now known from two localities in tropical Introduction montane forests. We conducted a -year study using transect surveys and opportunistic excursions to assess habitat he South Asian tropical island of Sri Lanka is rich in associations, demographics and abundance of A. kandianus Tamphibian diversity (Meegaskumbura et al., ). in and around Pidurutalagala Conservation Forest. We re- Of the country’s  described amphibian species  corded a mean of . post-metamorphs per year, with a (c. %) are endemic and . % are restricted to rainforests  density of ,  individual per  m , with occurrence within (Surasinghe, ; Wickramasinghe et al., ). Sri Lanka’s  a narrow extent (c. . km ) of the stream channel. amphibians are threatened by deforestation, environmental Behaviour and microhabitat selection varied depending on pollution and road traffic (Pethiyagoda et al., ; sex and stage of maturity. The species preferred moderately Karunarathna et al., ). These anthropogenic stressors have sized montane streams with rocky substrates and woody contributed to the extinction of  amphibian species, and de- debris, colder temperatures, and closed-canopy, intact ri- clining populations of nearly half of the extant species (MOE, parian forests. We noted size-based reversed sexual di- ). morphism and a strong ontogenetic relationship between The stream-dwelling toad genus Adenomus contains two snout–vent length and body weight. Anthropogenic activ- species, Kelaart’storrenttoadA. kelaartii and the Kandian ities such as intensive crop farming deterred the species; torrent toad A. kandianus (Meegaskumbura et al., a). The proximity to croplands had a negative influence on abun- latter was considered to be extinct for  years (Stuart et al., dance. We recommend re-delineation of the boundary of ) until its rediscovery at Peak Wilderness Sanctuary in Pidurutalagala Conservation Forest to incorporate the  (Wickramasinghe et al., ). Currently, this species is toad’s habitat into the core of the reserve and thus limit known from only two localities in the Central Highlands the impacts of human activities. Conservation and manage- (. , m; Gabadage et al., ) and it is categorized as ment actions such as ex-situ breeding, population monitor- Critically Endangered on the IUCN Red List (IUCN SSC ing, and restoration of degraded habitats could also Amphibian Specialist Group, ). The species’ molecular contribute towards the persistence of this toad. Our findings phylogeny,bioacoustics,osteology,andlarvalandadultmorph- provide useful insights into ecological research on and con- ology have been studied (Meegaskumbura et al., a,b). servation of range-restricted aquatic amphibians. In the literature on A. kandianus, details about the species’ natural history, habitat associations, behaviour and population status are limited. This paucity of ecological in- formation impedes conservation action. We therefore stud- SURANJAN KARUNARATHNA Nature Explorations & Education Team, Soysapura Flats, Moratuwa, Sri Lanka ied the population abundance, behaviour, body size variation with sex and sexual maturity, and microhabitat as- SUJAN HENKANATHTHEGEDARA Department of Biological & Environmental Sciences, Longwood University, Virginia, USA sociations and environmental correlations of A. kandianus.

DINESH GABADAGE and MADHAVA BOTEJUE Biodiversity Conservation Society, Our findings will help inform future conservation action for Nugegoda, Sri Lanka this Critically Endangered species and provide a foundation

MAJINTHA MADAWALA No. 18 Ratmal Mawatha, Sirimal Uyana, Ratmalana, for future research. Sri Lanka

THILINA D. SURASINGHE (Corresponding author) Department of Biological Sciences, Bridgewater State University, Bridgewater, MA 02325, USA Study area E-mail [email protected] Received  February . Revision requested  May . The study area is located partly in the buffer zone of Accepted  May . First published online  October . Pidurutalagala (Mount Pedro) Conservation Forest in

Oryx, 2017, 51(4), 619–626 © 2016 Fauna & Flora International doi:10.1017/S0030605316000594 Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.76, on 26 Sep 2021 at 16:33:49, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605316000594 620 S. Karunarathna et al.

cascades using hand ( ×  cm) and seine nets ( ×  m). For each individual found we recorded sex (based on the presence or absence of nuptial pads and gular sacs), snout–vent length (using a digital vernier caliper), stage of maturity (based on snout–vent length: juvenile #  mm, adult $  mm), and blotted wet weight (using a Taylor spring scale). Measuring blotted wet weight eliminates excess water on skin surface. All captured individuals were released to the wild without harm at the point of capture.

Environmental variables Where toads were found we recorded the following environmental variables: stream width, canopy cover (using a canopy densiometer), water and air temperature (using a standard mercury thermometer), relative humidity and light intensity (using a Digitech QM -in- multifunction environment meter), and pH (using aDigitech QM– pH meter). Wemeasured the length of boulders at the intermediate axis, and the FIG. 1 The location of the study site in Pidurutalagala above-water height of emergent boulders, pool depth, and Conservation Forest in the Central Highlands of Sri Lanka. the Euclidean distance from each toad to the nearest cultivated land. We obtained rainfall data from the nearest weather station (National Meteorological Department, Sri Sri Lanka’s Central Highlands (Fig. ), one of the two local- Lanka). Additionally, we observed the surrounding habitat ities from where A. kandianus is known (Gabadage et al., and recorded any evidence of anthropogenic disturbance. ). Although the forest is under governmental jurisdic- tion, its buffer zone is imperilled by a multitude of anthropogenic disturbances, the most detrimental being Behavioural observations We noted the behaviour of each clear-cutting (Plate ). It is one of the largest montane pro- toad at the time of observation, as well their microhabitat tected areas in Sri Lanka (,. ha) and is also a UNESCO and substrate use. World Heritage Site (UNESCO, ). The mean annual precipitation is . , mm, and mean annual temperature Statistical analysis To test for differences in relative is  °C (Gabadage et al., ). abundance among age–sex categories (male, female, juveniles) we ran a one-way ANCOVA (analysis of Methods covariance) in which abundance was considered to be the response variable, the age–sex category was the treatment, Survey design We conducted -week-long field excursions and rainfall and sampling dates were covariates. Pairwise annually during – and twice in . Surveys were differences in abundance of age–sex groups were tested conducted in the morning (.–.) and evening using a Tukey HSD (honest significant difference) test. To (.–.) to account for both diurnal and nocturnal test for differences in body size among age–sex categories activities. Across the entire survey period we surveyed we used one-way ANOVA (analysis of variance) in which both during (April, May, August) and outside (October, the age–sex category was considered to be the treatment December) the monsoon season. In addition to transect and snout–vent length and blotted wet weight were surveys we made opportunistic observations. Statistical response variables. A post-hoc Tukey test was performed analyses were based only on the transect surveys. We to discern pairwise differences among age–sex groups for placed   ×  m belt transects along stream habitats, blotted wet weight and snout–vent length. We constructed which covered stream channels and adjacent woodlands, a linear regression model to test for a significant relationship with a minimum distance of  m between neighbouring between the snout–vent length and blotted wet weight. We  transects. The four-person survey team walked along the used a χ test of independence to detect any influence of sex  transects, actively searching and visually scanning for on behavioural acts or habitat use. Based on a χ test for amphibians. Active searching included examining leaf goodness of fit, we tested if the behavioural acts and habitat litter, superficial roots and tree trunks, and understorey use were distributed evenly among individuals. To test for vegetation, lifting movable rocks and woody debris, and the importance of various environmental variables on the searching the stream channel (riffles and pools) and abundance of toads, we constructed a stepwise multiple

PLATE 1 (a) Adenomus kandianus underwater, camouflaged against the rocky streambed, (b) male (on top) and female Adenomus kandianus in amplexus, (c) and (d) plantation agriculture and deforestation around Pidurutalagala Conservation Forest, and (e) the stream segment at Pidurutalagala Conservation Forest, one of two known localities of Adenomus kandianus.

 regression model (a mixed-model approach) in which all the per  m . Abundance differed significantly among the environmental variables were considered to be predictor age–sex groups (F = .,P, .); effects of rainfall variables and the abundance of toads was the response (F = .,P, .) and date surveyed (F = ., variable. Statistical analyses were conducted using various P , .) were significant covariates. The abundance of  packages (ANCOVA: aov; linear regression: lm;andχ test: males was significantly higher than that of females (mean chisq.test)inRv... (R Development Core Team, ). difference = .,P, .). The differential mean abundance between males and juveniles (mean difference = ., P . .) and between females and juveniles was not signifi- Results cant (mean difference = .,P. .). The population was    The transects covered a total area of . km , in which we male-skewed (male : female = : ) throughout the study,     recorded A. kandianus only in a  ×  m(. km ) except in ( : ). stretch of a medium-sized fast-flowing perennial montane stream. The toads were found only along the stream chan- Behaviour and microhabitat use nel, including undercut banks, and never in the woodlands. Our opportunistic surveys of other neighbouring streams, Behavioural acts we observed included perching on wetlands and forests did not reveal any persistent popula- boulders, mating in a variety of microhabitats (boulders, tions of the species. submerged in water), swimming, and seeking refuge (Table ). Age–sex category had a significant influence on  behavioural acts (χ = .,P, .). Behavioural acts Total abundance  were not observed uniformly among toads (χ = ., We recorded a total of  individuals, with a mean of . P , .). Perching was the most commonly observed be- males,  females and . juveniles per year (Table ). The havioural act among males, followed by mating. Among fe- overall population density was . ± SD . individuals males, mating was the most dominant behavioural act

TABLE 1 Mean number of females, males and juveniles of Adenomus kandianus recorded per transect in Pidurutalagala Conservation Forest, Sri Lanka (Fig. ), in each of  sampling years (–).

Sampling year No. of females ± SD No. of males ± SD No. of juveniles ± SD Total ± SD 2011 3 ± 0.45 8 ± 0.49 3 ± 0.45 14 ± 0.39 2012 29 ± 3.78 47 ± 4.49 12 ± 0.51 88 ± 7.21 2013 6 ± 0.51 2 ± 0.35 7 ± 0.52 15 ± 0 2014 14 ± 0.43 41 ± 1.69 5 ± 0.34 60 ± 2.01 Mean per year 13 ± 10.07 24.5 ± 19.73 6.75 ± 3.34 44.25 ± 28.64

TABLE 2 Percentage of female, juvenile and male individuals of  we observed  amplexed pairs in a single in-stream Adenomus kandianus recorded mating, perching, seeking refuge, pool along with  advertising males, the only breeding con- and diving and swimming during surveys in Pidurutalagala gregation we recorded throughout our survey. Amplexus  Conservation Forest (Fig. ). was observed in a variety of microhabitats: underwater, on Behaviour Females (%) Juveniles (%) Males (%) emergent parts of moss-covered boulders, and in the leaf lit- Mating 82.69 0 40.57 ter on moss-covered boulders. We found no egg masses dur-  Perching 11.54 3 44.33 ing our survey. Amplexus lasted for hours on average but Seeking refuge 0 97 0 some amplexed pairs remained in position at the same loca- Diving & swimming 5.77 0 15.10 tion for  hours. On each of five different occasions we observed five amplexed pairs swimming continuously for . minutes (Plate ). On  occasions we observed non-mating (Table ). In a number of instances juveniles emerged from toads swimming as a group. Individuals remained under-  undercut banks at dusk (.–.) and perched at the water for a mean of minutes. stream edge in groups of –. Neither males nor females were observed in refuge. The juveniles had only partially de- veloped webbing in their feet and thus were weak swimmers. Body size variation Adults had complete webbing and were strong swimmers, Both snout–vent length and blotted wet weight differed sig- and swam perpendicular to the flow. nificantly among age–sex categories (Table ). Females were  The toads were found in five microhabitats: ( ) on moss- significantly larger than males and juveniles in terms of both covered boulders (most of the time, underneath leaf litter on snout–vent length and blotted wet weight. Males were sig-   the boulders), ( ) in pools with decaying organic matter, ( ) nificantly larger than juveniles in terms of snout–vent  – in pools with a sandy substrate, ( ) in riffles with a rocky length. There was a strong linear length–weight relationship –  cobble pebble substrate, and ( ) in undercut banks, among among all age–sex categories (linear regression model,   leaf litter and roots of riparian vegetation (Table ). Toads R = .; F = .;P, .; Fig. ). did not occur uniformly across all microhabitats  (χ = .,P, .); microhabitat use was dependent on – χ   ,   the age sex category ( = . ,P . ). Both males Environmental variables and females were found mostly on moss-covered boulders (Table ). A substantial proportion of adults were also The environmental variables indicated an association be- found in in-stream pools. A few adults were observed active- tween A. kandianus and microhabitats with higher canopy ly swimming across fast-flowing riffles. Neither adults nor cover (. %), cooler air temperatures (, °C), lower juveniles were found in the riparian zone or in woodlands. light intensity (,  lux) and high humidity (. %) Juveniles were found almost exclusively at the edge of the (Table ). The low-order montane stream in which we stream channel, within the leaf litter or finely branched found the species was cool (°C), fast-flowing and medium- roots of the riparian vegetation in undercut banks. We did sized ( m in width), with multiple pools and riffles and not observe any juveniles cohabiting with adults in the same near-neutral pH (.). The stream channel contained microhabitat. boulders of various sizes (– m in length), which were con- Advertisement calls and mating (amplexus) were re- tinually splashed by the turbulent flow. The stream substrate corded only during April–May in each year. On a given comprised cobbles and pebbles, leaf litter and sand. Given day, calling started at . and continued until .. the reduced flow, pool habitats had accumulated silt, sedi- Toads were calling from mid-stream, while perching on ment and organic matter (coarse woody debris).The best boulders. Precipitation was associated with increased inci- predictors of the abundance of A. kandianus were the dence of observed mating; for example, on a rainy day in Euclidean distance to the nearest cropland, and the ambient

TABLE 3 Percentage of female, juvenile and male individuals of Adenomus kandianus occupying various microhabitats in Pidurutalagala Conservation Forest (Fig. ).

Microhabitat Females (%) Juveniles (%) Males (%) Moss-covered boulders in stream channel 77.35 0 72.34 Pools with a decaying organic-matter substrate 20.15 0 11.27 Pools with a sandy substrate 0 0 13.19 Riffles with rocky-cobble-pebble substrate 2.5 0 3.2 Undercut banks with leaf litter & complex root structure of riparian vegetation 0 100 0

TABLE 4 Differences in snout–vent length and blotted wet weight among females, males and juveniles of Adenomus kandianus, with ANOVA and Tukey statistics.

Female Male Juvenile F statistics (ANOVA) Tukey statistics Blotted wet weight ± SD (g) 3.51 ± 0.11 2.41 ± 0.14 1.76 ± 0.11 387.25* Female–Male 1.10* Female–Juvenile 1.75* Male–Juvenile 0.65* Snout–vent length ± SD (mm) 41.16 ± 0.53 30.78 ± 0.47 19.62 ± 0.56 3,172.61* Female–Male 10.37* Female–Juvenile 21.54* Male–Juvenile 11.16*

*P , .

TABLE 5 Mean values of environmental variables measured when individuals of Adenomus kandianus were found during surveys in Pidurutalagala Conservation Forest (Fig. ), with partial correlation coefficients between the abundance of toads and environmental variables extracted from a stepwise multiple regression model.

Partial correlation Variable Mean ± SD coefficients Canopy cover (%) 68.38 ± 1.41 0.71 Cloud cover (%) 37.03 ± 1.48 0.505 Air temperature (°C) 25.36 ± 0.14 −0.98* Water temperature (°C) 17.16 ± 0.13 −0.75 Monthly rainfall (mm) 305.41 ± 17.35 0.83 Relative humidity (%) 72.00 ± 1.50 0.87 ± FIG. 2 Length–weight relationship for adult females, adult males Water pH 6.14 0.03 0.39 and juveniles of Adenomus kandianus in Pidurutalagala Illuminance (lux) 32.00 ± 2.83 0.45 Conservation Forest, Sri Lanka, based on a linear regression of Stream width (m) 6.08 ± 0.54 0.29  snout–vent length (SVL) against blotted wet weight (R = ., Euclidean distance to 57.23 ± 14.48 0.99* F = .,P=.). nearest cropland Size of in-stream 4.5 ± 3.5 0.75

 boulders (m) temperature (stepwise multiple regression, R = ., Height of submerged 1.58 ± 1.1 0.56 F = .,P, .); the former was the most influential boulders (m) variable (model coefficients: distance to croplands = ., *P , . ambient temperature = .). Partial correlation analyses revealed that rainfall, relative humidity, canopy cover, water temperature and size of boulders had the highest in- and were attached to rocky substrates by their oral discs. Their fluence on the abundance of toads (Table ). bodies were dark-brown to black in coloration.

Observations of tadpoles Discussion

We observed a mean of only  tadpoles per year in in-stream In our surveys in both wet and dry seasons over  years we pools. All tadpoles were found completely submerged in water recorded only c.  post-metamorphs in any given year. The

 overall density was low (,  individual per  m ) and the high humidity implies that this species is susceptible to des- area of occupancy did not extend beyond a stream corridor iccation. Medium-sized, moist boulders are likely to be pre-  of . km . Similar demographic characteristics have been ferred sites for advertisement calls. During surveys we reported for other threatened, narrow-ranging tropical am- observed a few potential predators of A. kandianus: phibians (Sodhi et al., ). We found neither eggs nor Tickell’s blue flycatcher Cyornis tickelliae, Sri Lanka whist- gravid females, and only a small number of juveniles and ling thrush Myophonus blighi, Indian blackbird Turdus tadpoles. Although no egg masses were observed at Peak simillimus,SriLankahillmynaGracula ptilogenys and river Wilderness (the only other known habitat of A. kandianus), otter Lutra lutra. Prolonged amplexus may predispose A. kan- gravid females were recorded there (Meegaskumbura et al., dianus to predation where leaf litter on boulders provides b). Unlike adults, tadpoles and juveniles are weak swim- concealment. mers and may not withstand the heightened discharge after Species detectability may have played a significant role in torrential precipitation, and therefore they may have been our study. Precipitation and seasonality influenced abun- washed downstream (Meegaskumbura et al., b). This dance and breeding of A. kandianus, and may also have in- speculation was substantiated by our opportunistic observa- fluenced the species’ detectability and site occupancy. We tions of juveniles in lower altitude paddy fields after heavy are unaware of any refugia used by toads when they are in- rainfall. Adenomus kandianus is a specialist toad adapted to active. It is possible that these toads use subterranean bur- montane, undisturbed forested streams, and therefore tad- rows or deep crevices in the stream bed that were poles and juveniles transported into anthropogenic environ- undetectable by our surveys. ments are unlikely to survive. As a protective measure (to prevent downstream drift, predation, ambient exposure and radiation damage) adults may oviposit in protected, Threats and conservation deep interstices, which are undetectable by surveys (Refsnider & Janzen, ). Low abundance of larval and ju- In contrast to our findings, surveys at Peak Wilderness re- venile stages may indicate lower recruitment rates and lower vealed that females occurred outside the stream channel survival of early age classes. During surveys we found (Meegaskumbura et al., a,b). It was implied that the spe- dragonfly larvae in the stream; these macroinvertebrates cies may not be rare, at least at the Peak Wilderness locality, could predate on early life stages of A. kandianus, leading given en masse reproduction and high tadpole densities. to higher mortality among tadpoles and juveniles (Skelly, There has been relatively less habitat destruction at the ). Peak Wilderness locality compared to Pidurutalagala There was a strong positive correlation between snout– Conservation Forest. We recorded the presence of multiple vent length and blotted wet weight, indicating an ontogenic commercial plantations (e.g. banana, tea, cardamom, to- relationship. Distinct size segregation among various age– mato, cabbage, beans) during our study. In addition, clear- sex categories indicates that body dimensions are sexually cut forestlands, gem and graphite mining, heavy extraction dimorphic features for this species. Larger body size of fe- of non-timber forest products, and canopy dieback were males compared to males (reversed size-based sexual di- also observed in close proximity to our study site. These an- morphism) has been observed in many anurans thropogenic stressors may have impaired the reproductive (Woolbright, ). Among basal vertebrates, large females fitness of A. kandianus, and precluded local movements in may produce a greater number of eggs, reproducing early in upland habitats. the season, which may ensure early metamorphosis of large Ongoing deforestation and agricultural expansion may juveniles; larger body size confers higher fecundity, timely lead to higher rates of soil erosion; subsequent siltation reproduction and a high juvenile survival rate, and thus and sedimentation in stream channels could smother eggs higher reproductive fitness overall (Tejedo, ). and destroy larval feeding grounds (Richter et al., ). Microhabitat use and behaviour differed substantially We observed the use of agrochemicals on nearby croplands, between adults and juveniles: adults were mostly observed which can enter the stream habitat through surface run-off mating or perching, whereas juveniles were seeking refuge. and result in developmental abnormalities, growth retard- Niche partitioning within conspecifics, such as microhabitat ation, delayed metamorphosis and increased egg and larval and resource specialization, can mitigate offspring–adult mortality (Sampath et al., ). We also noted that some competition (Meegaskumbura et al., a). A closed can- invasive plant species had colonized the riparian forests opy, high humidity, low temperatures, intermediate-sized and adjoining woodlands. Anthropogenic disturbance boulders, and absence of anthropogenic disturbances were such as forest clearing and crop farming may have facilitated important environmental conditions for A. kandianus. the establishment of these species (Didham et al., ). These ecological niche dimensions are peculiar to many Our extensive survey confirmed that at Pidurutalagala rainforest amphibians (Pethiyagoda et al., ). The de- Conservation Forest A. kandianus is restricted to a narrow  pendence on well-shaded canopy, colder temperatures and stretch (. km ) of a perennial mountain stream.

Fluctuating abundance implies instability of the population. we recommend could be adapted for other threatened aquatic Relatively small, isolated populations are susceptible to both species, especially those of tropical montane streams, where demographic and environmental stochasticity, and human there is a need for science-based conservation. disturbances can cause further decline and result in an extinction vortex (Sodhi et al., ). Although ecological niche modelling predicted a potential distribution range of Acknowledgements     c. km for the species, only km of suitable habitat We express our sincere gratitude to Anslem de Silva, Thasun was available within the predicted range (Meegaskumbura Amarasinghe, Kelum Manamendra-Arachchi, Mendis  et al., b). Thus, it is possible to infer a population decline Wickramasinghe, the Forest Department, the Department of .  of % within the past decade. Like many amphibians Wildlife Conservation, the National Museum of Sri Lanka A. kandianus faces a high risk of extinction because of eco- (Sanuja Kasthuriarachchi, Nanda Wickramasinghe, Lankani physiological constraints, high habitat specialization, poor Somaratne, Manori Nandasena, Chandrika Munasinghe and dispersal capacity and high site fidelity (Blaustein et al., Rasika Dasanayake), and the Young Zoologists’ Association  ’ ). Our findings support the species current IUCN cat- for help during this study. egorization as Critically Endangered. Despite the narrow area of occupancy, no national-level actions have been im- plemented to protect this toad. Our study area falls partly Author contributions within the buffer zone of Pidurutalagala Conservation Forest, where multiple uses and human activities are permit- SKandSHdesignedthefieldsamplingprotocols.SK,DG,MB ted. We recommend that this vital stream habitat and the and MM conducted the fieldwork and data entry. SH and TS surrounding woodlands be incorporated into the core habitat conducted the statistical analyses and SH, TS, and SK wrote the of the reserve, thus shifting the buffer zone to a lower altitude. article. SH and TS produced the maps, figures and tables. All It is crucial to protect both the immediate riparian zone authors contributed equally in finalizing the manuscript. and the surrounding woodlands to sustain watershed-scale ecological processes such as provision of allochthonous ma- References terials (Olson et al., ). Delineating a core and a terres-

trial buffer around streams may negate the adverse effects of BLAUSTEIN, A.R., WAKE, D.B. & SOUSA,W.P.() Amphibian agriculture (Semlitsch & Bodie, ). Pidurutalagala is a declines: judging stability, persistence, and susceptibility of unique montane rainforest rich in endemic native biodiver- populations to local and global extinctions. Conservation Biology, , – sity and threatened by human activities (Werner, ), and . DIDHAM, R.K., TYLIANAKIS, J.M., HUTCHISON, M.A., EWERS, R.M. & we recommend that it be incorporated into Sri Lanka’s GEMMELL, N.J. () Are invasive species the drivers of ecological Central Highlands UNESCO World Heritage Site. Given change? Trends in Ecology & Evolution, , –. the presence of a Critically Endangered species, headwater DUDLEY,N.() Guidelines for Applying Protected Area streams and anthropogenic stressors, we recommend up- Management Categories. IUCN, Gland, Switzerland. grading the forest to a Strict Nature Reserve, where actions GABADAGE, D.E., DE SILVA, A., BOTEJUE, W.M.S., BAHIR, M.M., SURASINGHE, T.D., MADAWALA, M.B. et al. () On the discovery can be taken to restrict human impacts, implement continu- of second living population of Adenomus kandianus (Günther, ) ous monitoring and scientific research, and promote from Sri Lanka: with the bioecology, and detailed redescription to ecosystem-wide biodiversity conservation (Dudley, ). the species. Herpetotropicos, , –. IUCN SSC AMPHIBIAN SPECIALIST GROUP () Adenomus kandianus. The IUCN Red List of Threatened Species :e. Conclusion TA. Http://dx.doi.org/./IUCN.UK..RLTS. TA.en [accessed  May ]. Future studies should focus on enumerating the abundance KARUNARATHNA, D.M.S.S., HENKANATHTHEGEDARA, S.M., of tadpoles, juveniles and subadults, population viability AMARASINGHE, A.A.T. & DE SILVA,A.() Impact of vehicular analyses, and the reproductive biology of A. kandianus, so traffic on Herpetofaunal mortality in a savannah forest, eastern that scientifically robust inferences can be made about the Sri Lanka. Taprobanica, , –. species’ population dynamics. Given that A. kandianus is MEEGASKUMBURA, M., BOSSUYT, F., PETHIYAGODA, R., MANAMENDRA-ARACHCHI, K., BAHIR, M., MILINKOVITCH, M.C. now limited to two known populations, both small, we rec- &SCHNEIDER, C.J. () Sri Lanka: an amphibian hot spot. ommend a preliminary attempt at captive breeding and ex Science, , . situ conservation. Given the species’ cryptic nature and sea- MEEGASKUMBURA, M., SENEVIRATHNE, G., WIJAYATHILAKA, N., sonality we recommend that any future surveys use occu- JAYAWARDENA, B., BANDARA, C., MANAMENDRA-ARACHCHI,K.&  pancy modelling to account for imperfect detection. PETHIYAGODA,R.( a) The Sri Lankan torrent toads (Bufonidae: Adenominae: Adenomus): species boundaries assessed using Our research methods could serve as a model for multiple criteria. Zootaxa, , –. ecological studies on other less-studied, threatened, MEEGASKUMBURA, M., WIJAYATHILAKA, N., ABAYALATH,N.& range-restricted amphibians. The conservation actions SENEVIRATHNE,G.(b) Realities of rarity: climatically and

ecologically restricted, critically endangered Kandian Torrent Toads SURASINGHE, T.D. () Conservation and distribution status of (Adenomus kandianus) breed en masse. PeerJ PrePrints, ,e. amphibian fauna in Sri Lanka. Biodiversity, , –. MOE (MINISTRY OF ENVIRONMENT)() The National Red List TEJEDO,M.() Effects of body size and timing of reproduction on  of Sri Lanka; Conservation Status of the Fauna and Flora. reproductive success in female natterjack toads (Bufo calamita). Biodiversity Secretariat, MinistryofEnvironment, Colombo, SriLanka. Journal of Zoology, , –. OLSON, D.H., ANDERSON, P.D., FRISSELL, C.A., WELSH JR, H.H. & UNESCO () World Heritage List. Http://whc.unesco.org/en/list/ BRADFORD, D.F. () Biodiversity management approaches for [accessed  March ]. stream–riparian areas: perspectives for Pacific Northwest headwater WERNER,W.() Human impact on natural environment in the forests, microclimates, and amphibians. Forest Ecology and Central Highlands of Sri Lanka. Universitas, , –. Management, , –. WICKRAMASINGHE, L.J.M., BANDARA, I.N., VIDANAPATHIRANA,D.R., PETHIYAGODA, R., MANAMENDRA-ARACHCHI, K., BAHIR, M.M. & TENNAKOON, K.H., SAMARAKOON,S.R.&WICKRAMASINGHE,N. MEEGASKUMBURA,M.() Sri Lankan amphibians: diversity, () Pseudophilautus dilmah, a new species of shrub frog uniqueness and conservation. In The Fauna of Sri Lanka: Status of (Amphibia: Anura: Rhacophoridae) from a threatened habitat Taxonomy, Research and Conservation (ed. C.N.B. Bambaradeniya), Loolkandura in Sri Lanka. Journal of Threatened Taxa, , –. pp. –. The World Conservation Union, Colombo, Sri Lanka & WICKRAMASINGHE, L.J.M., VIDANAPATHIRANA, D.R. & Government of Sri Lanka. WICKRAMASINGHE,N.() Back from the dead: the world’s rarest RDEVELOPMENT CORE TEAM () R: A Language and Environment toad Adenomus kandianus rediscovered in Sri Lanka. Zootaxa, , for Statistical Computing. R Foundation for Statistical Computing, –. Vienna, Austria. WOOLBRIGHT, L.L. () Sexual selection and size dimorphism in REFSNIDER, J.M. & JANZEN, F.J. () Putting eggs in one basket: anuran amphibia. The American Naturalist, , –. ecological and evolutionary hypotheses for variation in oviposition-site choice. Annual Review of Ecology, Evolution, and Systematics, , –. RICHTER, B.D., BRAUN, D.P., MENDELSON, M.A. & MASTER, L.L. Biographical sketches () Threats to imperiled freshwater fauna. Conservation Biology, , –. S URANJAN K ARUNARATHNA is a field biologist, and his research fo- SAMPATH, K., KENNEDY, I.J.J. & JAMES,R.() Pesticide impact on cuses on herpetofaunal ecology, taxonomy and conservation of threa- excretory physiology of the common frog, Rana tigrina (Daud) tened species. He is an active member of the IUCN Amphibian tadpoles. Bulletin of Environmental Contamination and Toxicology, Specialist Group. S UJAN H ENKANATHTHEGEDARA’s research fo- , –. cuses on the ecology, conservation and taxonomy of aquatic species. SEMLITSCH, R.D. & BODIE, J.R. () Biological criteria for buffer D INESH G ABADAGE is a field biologist. He studies the distribution zones around wetlands and riparian habitats for amphibians and of birds and mammals, and promotes wildlife conservation pro- reptiles. Conservation Biology, , –. grammes in Sri Lankan communities. M ADHAVA B OTEJUE is the SKELLY, D.K. () Activity level and the susceptibility of anuran Regional Vice Chairman (South Asia & Iran) of the IUCN Crocodile larvae to predation. Animal Behaviour, , –. Specialist Group. His research focuses on the ecology, distribution and SODHI, N.S., BICKFORD, D., DIESMOS, A.C., LEE, T.M., KOH, L.P., conservation of herpetofauna, avifauna and mammals. M AJINTHA BROOK, B.W. et al. () Measuring the meltdown: drivers of M ADAWALA is engaged in numerous habitat restoration and snake global amphibian extinction and decline. PLoS ONE, (), e. rescue programmes and is an active member of the IUCN Crocodile STUART, S.N., HOFFMANN, M., CHANSON, J.S., COX, N.A., BERRIDGE, Specialist Group. T HILINA S URASINGHE’s research interests include R.J., RAMANI,P.&YOUNG, B.E. (eds) () Threatened landscape-scale biodiversity conservation, freshwater ecology, and Amphibians of the World. Lynx Edicions, Barcelona, Spain. conservation of threatened species.