Content uploaded by Kanishka Ukuwela
Author content
All content in this area was uploaded by Kanishka Ukuwela on Mar 21, 2020
Content may be subject to copyright.
DOI: 10.30906/1026-2296-2020-27-1-33-40
NEW LOCALITIES, DISTRIBUTION AND HABITAT MODELING
OF THE CRITICALLY ENDANGERED SRI LANKAN FROG Nannophrys marmorata
Kanishka D. B. Ukuwela,1Imesh Nuwan Bandara,2H. K. Sameera De Zoysa,3
U. A. L. Deshan Rupasinghe,1Rajnish P. G. Vandercone1
Submitted April 15, 2018
Nannophrys marmorata is a critically endangered species of frog known only from a few locations in the
Knuckles Conservation Forest (KCF) in the Knuckles mountain range of Sri Lanka. Here, we report new localities
outside its known range in the KCF and examine the distribution of N. marmorata providing the most extensive
distribution map for this species. Furthermore, using the distribution information, we model the fundamental eco-
logical niche of this species through Ecological Niche Modeling (ENM) to predict suitable habitats and fine-scale
distribution patterns. ENM predicted eastern slopes of the Knuckles mountain range in central Sri Lanka as areas
with high habitat suitability (suitability >75%) for N. marmorata. Predicted Extent of Occurrence for N. marmo-
rata was 831 km2and predicted Area of Occupancy was 275 km2. We discuss these findings in light of conserva-
tion for this species threatened with extinction.
Keywords: area of occupancy; conservation; DNA barcoding; ecological niche modeling; extent of occurrence;
Maxent; VRR sanctuary.
INTRODUCTION
Amphibians are the most threatened group of verte-
brates in the world with nearly 41% of the global diver-
sity at risk of extinction (Hoffmann et al., 2010). Sri
Lanka is an amphibian hotspot (Meegaskumbura et al.,
2002) which is home to 119 species with 105 species be-
ing endemic to the island (Wickramasinghe et al., 2015;
Senevirathne et al., 2018). Of this, 65% of the diversity is
threatened with extinction, while approximately 34% are
categorized as critically endangered and 23% categorized
as endangered species (Manamendra-Arachchi and Mee-
gaskumbura, 2012). Unfortunately, 18 of the 119 species
have already become extinct (Wickramasinghe et al.,
2012, 2013a, 2013b). Thus, conservation measures are
urgently necessary to stem the rate of amphibian popula-
tion declines of many threatened species. Fundamental to
the conservation of amphibians is an understanding of
fine-scale distribution patterns and identification of iso-
lated populations (Marsh and Trenham, 2001). However,
determining the fine scale distribution and identification
of isolated populations of species is time consuming and
costly; hence computational methods such ecological
niche modeling (ENM) have proved to be useful alterna-
tives in determining the potential areas of occupancy of
species (Peterson et al., 2000).
Nannophrys is a genus of frogs endemic to Sri Lanka,
which consists of four species (Fernando et al., 2007). All
species of the genus are threatened with extinction while
a single species (N. guentheri) is already extinct (Mana-
mendra-Arachchi and Meegaskumbura, 2012). Members
of this genus are closely associated with wet rock sur-
faces and boulders and are characterized by flattened
bodies that enable them to live under boulders and in rock
crevices and thus have earned the name “Rock Frogs”
(Dutta and Manamendra-Arachchi, 1996; Senevirathne
and Meegaskumbura, 2015). The Marbled or Kirthi-
singhe’s Rock Frog, Nannophrys marmorata is a criti-
1026-2296/2020/2701-0033 © 2020 Folium Publishing Company
Russian Journal of Herpetology Vol. 27, No. 1, 2020, pp. 33 – 40
1Department of Biological Sciences, Faculty of Applied Sciences,
Rajarata University of Sri Lanka, Mihintale 50300, Sri Lanka;
e-mails: kanishkauku@gmail.com; deshan1st@gmail.com;
vandercone@gmail.com
2Ellangàwa Unity Care for Community and Nature, Hapugoda, Am-
bathenna, Sri Lanka; e-mail: imeshnu1@gmail.com
3Department of Bioprocess Technology, Faculty of Technology,
Rajarata University of Sri Lanka, Mihintale 50300, Sri Lanka;
e-mail: dezoysahks@yahoo.com
cally endangered species of frog known only from a few
locations in the Knuckles Conservation Forest (KCF) of
the Knuckles mountain range. Here, we report new local-
ities and examine the distribution of N. marmorata. Fur-
ther, we model the fundamental ecological niche of
N. marmorata using published and new locality data to
unravel potential areas of occupancy and fine scale distri-
bution patterns of this critically endangered frog.
MATERIAL AND METHODS
New localities and distribution. We examined the
distribution of N. marmorata through field surveys con-
ducted in the Knuckles mountain range and the adjacent
areas since 2006. The species was identified using the de-
scriptions available in Dutta and Manamendraarachchi
(1996) and Fernando et al. (2007). Whenever a positively
identified N. marmorata was observed, the coordinates
of the location were recorded using a Garmin etrex GPS
(Garmin International, USA) and the habitat characteris-
tics were noted. We then mapped the distribution of
N. marmorata using these locations and locations taken
from published literature (Clarke, 1983; Dutta and Mana-
mendra-Arachchi, 1996; de Silva et al., 2005; Fernando
et al., 2007).
During the course of the survey, several frogs belong-
ing to the genus Nannophrys were observed on wet rock
surfaces on the northern portion of the Victoria-Randeni-
gala-Rantambe (VRR) Sanctuary on the 19 April 2008
and the 14 August 2008. Though they generally resem-
bled N. marmorata, their coloration was somewhat dif-
ferent to the typical coloration of N. marmorata seen in
the Knuckles mountain range. We revisited the site on the
13 February 2016 and observed a total of 32 frogs in four
different nearby locations. Four specimens were photo-
graphed in several different angles and a total of 12 mor-
phometric measurements were taken using a Mitutoyo
digital caliper (Mitutoyo Corp, Kanogawa, Japan) and re-
leased back to the same site. The 12 morphometric mea-
surements were taken following Fernando et al. (2007):
snout to vent length (SVL), head length (HL), head width
(HW), head depth (HD), eye diameter (ED), inter orbital
length (IO), tympanum width (TW), upper arm (UA);
lower arm (LA), femur length (FL), tibia length (TL), tar-
sus length (TAL).
To confirm the identity of the frogs through DNA
barcoding, 0.5 mm parts of the fingertip was clipped
from two individual frogs and were preserved in 70%
ethanol solution. No specimens were collected and pre-
served during the study. For DNA barcoding, we PCR
amplified and sequenced an approximately 550 bp length
region of the small subunit of the 16S rRNA gene. Whole
genomic DNA was extracted from the two fingertips us-
ing Promega Wizard®genomic DNA purification kit
(Promega Corporation, Madison, Wisconsin, USA) fol-
lowing manufacturer’s protocols. 16S ribosomal RNA
gene region was amplified using the primers 16SA-L
(CGCCTGTTTATCAAAAACAT and 16SB-H (CCGGT
CTGAACTCAGA TCACGT) (Vences et al., 2005). The
PCR amplification employed 35 cycles with an annealing
temperature of 55°C (Vences et al., 2005) following stan-
dard PCR protocols with Promega PCR master mix (Pro-
mega Corporation, Madison, Wisconsin, USA). The PCR
products were purified and sequenced in both directions
at the Institute of Biochemistry, Molecular Biology and
Biotechnology, University of Colombo, Sri Lanka
(IBMBB). Consensus sequences from forward and re-
verse reads were assembled in GENEIOUS PRO 5.6
software (Drummond et al., 2009). The newly generated
DNA sequences have been submitted to the genbank un-
der the accession numbers MN022541-MN022542. We
downloaded several DNA sequences belonging to N. cey-
lonensis and N. marmorata from the Genbank (www.
ncbi.nlm.nih.gov/genbank). Nannophrys marmorata ref-
erence sequence available in the Genbank has been col-
lected from Laggala, Matale in the northern KCF (Vences
et al., 2000). All the sequences were aligned using the
GENEIOUS PRO 5.6 software (Drummond et al., 2009)
and then manually edited and refined by eye. A neigh-
bor-joining tree was constructed to examine the identity
of the two DNA sequences in MEGAversion 7.0 (Kumar
et al., 2016). Hoplobatrachus rugulosus was used as an
outgroup since it has been shown that Nannophrys is sis-
ter to both Euphlyctis-Hoplobatrachus clade (Pyron and
Wiens, 2011). The branch support for the Neighbor-Join-
ing tree was assessed by implementing 200 bootstrap
pseudoreplicates. Average pairwise genetic distances
were calculated in MEGA version 7.0 using K2P substi-
tution model.
Ecological Niche modeling (ENM). ENM was done
using coordinates of the occurrence localities of N. mar-
morata obtained from literature (n= 7) (Dutta and Mana-
mendra-Arachchi, 1996) and from our own observations
(n= 4) with environmental variable layers. We obtained
environmental variable layers (n= 55) for South Asian
region from BIOCLIM (http://www.worldclim.org/bio-
clim), a portal that provides free climate data. The data
layers were cropped to a smaller size to only include Sri
Lanka (9.9433 – 5.86810° N 79.31250 – 82.2285° E)
from its larger tile size that included the Indian subconti-
nent. All the environmental variable layers were re-
sampled into 1 km2resolution using ArcMap 10.3 soft-
ware (ESRI, 2014). ENM was done using maximum-en-
tropy techniques implemented in the software Maxent
(Phillips et al., 2006) with 15 replicates and 5000 itera-
34 Kanishka D. B. Ukuwela et al.
tions while keeping all other parameters default. The out-
put map was processed and interpreted using ArcMap
10.3 software (ESRI, 2014). The habitat suitability layer
properties of the predicted map were classified into five
classes (100 – 75%, 75 – 50%, 50 – 30%, 30 – 15%, 15 –
0%) using ArcMap 10.3.
We calculated the predicted Extent of Occurrence
(EOO) for N. marmorata using the predicted distribution
map by drawing a polygon connecting all the outer grids
which has over 75% habitat suitability and calculating
the area within the polygon using ArcMap 10.3 software.
We calculated the predicted Area of Occupancy (AOO)
by counting the number of 1 km2grids over 75% habitat
suitability using ArcMap 10.3.
RESULTS
New localities. The specimens observed in the VRR
sanctuary were assigned to the genus Nannophrys by the
following combination of characters: dorso-ventrally
flattened body; horizontal pupil; blunt digit tips; partially
webbed toes; well developed supra-tympanic fold; bound
outer metatarsals; fourth toe webbing between penulti-
mate subarticular tubercle and antepenultimate subarticu-
lar tubercle; presence of scattered white-tipped tubercles
on the upper surface of head, shoulders and limbs; pres-
ence of an elongate and a compressed inner metatarsal tu-
bercle (Clarke, 1983; Dutta and Manamendra-Arachchi,
1996) (Fig. 1). The specimens from the VRR sanctuary
were distinguished from N. ceylonensis and N. guentheri
by the smoothly rounded snout when viewed laterally
and the presence of scattered white tipped tubercles on
head and by the presence of marbled markings on legs
(Fig. 1A, B). They were differentiated from N. naeyakai
by the presence of closely placed distal subarticular and
penultimate subarticular tubercles on the 4th toe (vs. well
separated in N. naeyakai) tubercle (Clarke, 1983; Dutta
and Manamendra-Arachchi, 1996) (Fig. 1C).
Description of the specimens. Dorsum heavily tu-
berculated in males and subadults but smooth in the two
New Localities and Habitat Modeling of Nannophrys marmorata 35
Fig. 1. Nannophrys marmorata observed in the VRR Sanctuary, Sri Lanka: A, male; B, female; C, ventral aspect of the left foot showing the
closely placed distal subarticular and penultimate subarticular tubercles on the 4th toe.
females observed, venter smooth in all the observed indi-
viduals. Warty pustules present on the sides of the belly
in all. Dorsum yellow ochre to light brown with oval dark
brown spots in subadults and males and it was brown
with yellow marbled pattern on the female. Limbs mar-
bled, belly whitish yellow in all specimens. Snout to vent
length ranged from 25.32 to 34.89 mm (Table 1).
Molecular diagnosis. Neighbor-Joining tree placed
the two sequences generated from the samples collected
from the VRR Sanctuary with N. marmorata reference
sequence from the Genbank (AF215394) in a strongly
supported (bootstrap support >70) reciprocally mono-
phyletic group (Fig. 2). There was 0.4% average genetic
distance between the two sequences generated from the
samples collected from VRR Sanctuary and N. marmora-
ta sequence from the Genbank in the 16S rRNA gene
while it was 9.3% between the former and N. ceylonensis.
Description of the new locations and habitat. Indi-
viduals of N. marmorata were found on exposed wet rock
surfaces, roadside drains, rock crevices, under boulders
and along streams between Udadumbara and Gurulupo-
tha in three locations along the Kandy – Mahiyangana –
Padiyathalawa highway (coordinates 7°19¢56¢¢ N
80°54¢10¢¢ E – 7°19¢40¢¢ N 80°53¢41¢¢ E) (Fig. 3). The el-
evation of these locations ranged between 801 – 834 m
a.s.l. The surrounding habitat consisted of an ecotone be-
tween tropical dry-mixed evergreen forests and
submontane forests.
Distribution. The type locality of N. marmorata is
Maussakanda, Gammaduwa (7°34¢N 80°42¢E) at an ele-
vation of 915 m in the northwestern area of the Knuckles
Mountain range, which lies outside the KCF (Kirti-
singhe, 1946). Clark (1983) reported it from Laggala
(7°33¢N 80°44¢E) at an elevation of 1145 m. In addition
to these locations, Dutta and Manamendra-Arachchi
(1996) reported it from Gonewela (near Pallegama) in the
eastern slopes of the Knuckles mountain range. At an ele-
vation of 200 m a.s.l., Gonewela seems to be the lowest
elevation this species has been recorded so far. Fer-
nando et al. (2007) reported it near Rathna Ella falls
(7°22¢35¢¢ N 80°55¢06¢¢ E, elevation: 572 m a.s.l.) in the
Kandy district, which also lies outside the KCF. Further,
the species has been reported from Pitawala pathana
(7°32¢51¢¢ N 80°45¢18¢¢ E, elevation: 858 m a.s.l., near
Lakegala (7°26¢52¢¢ N 80°49¢23¢¢ E, elevation: 688 m
a.s.l., Emmadawa (7°30¢50¢¢ N 80°49¢38¢¢ E, elevation:
478 m a.s.l. and Kahatagolla (7°18¢19¢¢ N 80°52¢08¢¢ E,
elevation: 918 m a.s.l. (de Silva et al. 2005). More re-
cently the species was reported from an unspecified loca-
tion (thus not shown in map) in the Moragahakanda res-
ervoir construction site (Pollonnaruwa district, North
Central Province) (IUCN, 2007). These locations suggest
that the species is distributed between an elevation range
of 200 – 1200 m a.s.l. Observations made by us and pre-
vious authorities indicate that the frogs are mostly seen
on wet rock surfaces along streams. In Maussakanda,
Gammaduwa, we found the specimens along a stream
that flows through a tea estate and in Pitawala pathana the
frogs occur on wet rocks, under boulders and crevices
along a seasonal stream surrounded by grassland. Habi-
tats the specimens reported by us and other studies in-
clude tropical submontane forests, pathana grasslands
and tropical moist evergreen forests. Extent of occur-
36 Kanishka D. B. Ukuwela et al.
TABLE 1. Morphometric Measurements (mm) and the Sex of
Nannophrys marmorata Specimens Observed from the VRR Sanctu-
ary, Sri Lanka
Morphometric
measurement Specimen 1 Specimen 2 Specimen 3 Specimen 4
SVL 34.9 26.5 28.2 25.3
HL 13.3 11.3 11.8 10.1
HW 14.2 11.6 13.0 10.8
HD 7.6 5.2 4.4 4.2
ED 4.3 3.7 3.8 3.7
IO 4.0 2.9 3.0 2.4
TW 3.4 2.0 2.4 1.9
UA 6.3 4.8 6.1 4.2
LA 6.9 4.3 5.1 3.5
FL 13.0 11.3 11.5 10.2
TL 13.8 11.3 12.1 11.9
TAL 11.0 7.8 8.1 7.2
Sex Female Male Male Male
Fig. 2. Neighbor-Joining tree of 16S rRNA gene sequences showing
the affinities of the two Nannophrys specimens collected from VRR
Sanctuary. The outgroup, Hoplobatrachus rugulosus is not shown.
Values at the nodes indicate bootstrap support. Scale is in number of
substitutions per site.
rence with new locality data is 255.65 km2while the
AOO is 11 km2.
Ecological Niche modeling. ENM mostly predicted
eastern slopes of the Knuckles mountain range as areas
with high habitat suitability (suitability >75%) for
N. marmorata (Fig. 4). Predicted EOO for N. marmorata
was 831 km2and predicted AOO was 275 km2. The high
Area Under the Curve value (0.995, SD = ±0.004) sug-
gested good model performance and high accuracy of the
prediction. The environmental variable layer “Mean Di-
urnal Range (Mean of monthly (maximum temperature –
minimum temperature))” had the most useful informa-
tion for the model by itself as it had the highest gain when
used in isolation. Similarly, “Mean Diurnal Range (Mean
of monthly (maximum temperature — minimum temper-
ature))” had the most information for the model that is
not present in the other environmental variables as the
gain decreased the most when omitted.
DISCUSSION
Our morphological comparisons and molecular diag-
nosis strongly confirmed that the individuals of Nanno-
phrys observed in VRR Sanctuary are conspecific with
N. marmorata. The morphological characters and color-
ation of the specimens from the new locality reported
here were generally in agreement with that for the previ-
ously reported specimens from the northern Knuckles
mountain range (Clarke, 1983; Dutta and Manamendra-
Arachchi, 1996). Though, the dorsal color was lighter in
a few specimens observed, the two large females ob-
served had the marbled coloration, which is the typical
dorsal coloration for the species (Dutta and Manamen-
dra-Arachchi, 1996).
The presence of N. marmorata in the VRR Sanctuary
reported here makes it the fourth (other three locations
Gammaduwa, Rathna Ella and Moragahakanda) popula-
tion of this species known outside the Knuckles Conser-
vation Forest. Our findings and published data indicate
New Localities and Habitat Modeling of Nannophrys marmorata 37
Fig. 3. Distribution of Nannophrys marmorata, filled squares represent known locations of N. marmorata from the northern Knuckles mountain
range and the filled circles depict new localities. Dark green area represents the Knuckles Conservation Forest (KCF). Dark line illustrates the dis-
trict boundary that separates the Matale and Kandy districts.
that N. marmorata is mostly scattered in the Knuckles
mountain range and in a few adjacent areas. The novel
population reported here lies within the already known
elevation range of the species, which is between 200 –
1200 m a.s.l. (Dutta and Manamendra-Arachchi, 1996).
The Rathna Ella location reported by Fernando et al.
(2007) was not clearly depicted in their distribution map
and hence the significance of this finding was not high-
lighted then. The latter location, lies just outside the
northeastern border of the VRR Sanctuary in the Kandy
district (L. J. Mendis Wickramasinghe, personal commu-
nication) and is closer to new locations reported here (a
direct distance of 5 km). Moragahakanda is the northern
most location reported for this species so far. Though this
location is in the Northcentral Province, it actually lies
approximately 16 km north of Pitawala Pathana.
The new population reported here was observed in
the northern portion of the VRR Sanctuary, which is the
largest sanctuary in Sri Lanka. The northern portion of
the VRR sanctuary comprises regions of higher eleva-
tions (800 – 1000 m a.s.l.) with lower temperatures and
consists of ecotone vegetation of dry-mixed evergreen
forests and submontane forests. The rest of the sanctuary
consists of mid-elevational (300 – 600 m a.s.l.) grass-
lands, dry-mixed evergreen forests and anthropogenic
habitats. Thus it is not known whether the rest of the
sanctuary supports suitable habitats for N. marmorata
other than that of the area reported here. However, the
possible existence of N. marmorata in these areas cannot
be ruled out without a thorough survey of the sanctuary.
Nannophrys marmorata has been occasionally found
on exposed habitats such as man made boulders
(L. J. Mendis Wickramasinghe, personal communica-
tion). However, the species has generally been observed
on wet rock surfaces with seepage, in crevices and under
boulders in rocky hill streams of submontane forests,
grasslands and tropical moist evergreen forests in the
Knuckles mountain range (Clarke, 1983; Dutta and Ma-
namendra-Arachchi, 1996). The presence of these frogs
adjacent to highly disturbed places such as highways and
tea estates may indicate the ability of this frog to tolerate
and survive in disturbed habitats up to a certain extent.
Ecological niche modeling suggested highly suitable
areas for N. marmorata in the eastern slopes of the
Knuckles mountain range. However, apart from the few
isolated populations, populations in the intervening re-
gions are yet to be found. Ecological niche modeling fur-
ther indicated that the environmental variable layer
“Mean Diurnal Range (Mean of monthly (maximum
temperature – minimum temperature))” had the most
useful information for the model by itself and had the
most information for the model that is absent in the other
environmental variables. Both adult frogs and semi-ter-
restrial tadpoles of N. marmorata live on wet rock sur-
faces (Dutta and Manamendra-Arachchi, 1996; Senevi-
rathne and Meegaskumbura, 2015). High temperatures in
the surrounding environment may increase the drying of
wet rock surfaces that the adults and tadpoles dwell on. It
is well documented that increasing temperatures nega-
tively affect the survival of all stages of amphibians
(Blaustein et al., 2010; Li et al., 2013). Thus, prolonged
droughts in combination with high temperatures will
most likely reduce the most suitable habitats for N. mar-
morata in the future. This could be the reason that the
“Mean Diurnal Range (Mean of monthly (maximum
temperature – minimum temperature))” was the environ-
mental variable layer that contained the most useful in-
formation for the model by itself and had the most infor-
mation for the model that is absent in the other environ-
mental variables.
According to the IUCN, the EOO of N. marmorata is
81.73 km2and the AOO is <10 km2and hence has been
placed in the “Critically Endangered” category (criteria
B1ab) (Manamendra-Arachchi et al., 2008). The EOO
with new locality data has increased to 255 km2and it has
further increased to 831 km2according to the predictions
through ENM. Since the species was placed in the
“Critically Endangered” category based solely on EOO
(EOO <100 km2) and AOO (AOO <10 km2) (IUCN,
2012), the species can be now removed from the category
ass EOO and AOO are higher. The AOO from the predic-
tion of ENM has also increased to an area of 275 km2.
However, it is necessary to confirm the occurrence of
38 Kanishka D. B. Ukuwela et al.
Fig. 4. Suitable habitats for N. marmorata predicted through ecologi-
cal niche modeling in MAXENT. Model predicted most of the habitats
on the eastern slopes of the Knuckles mountain range as highly suitable
habitats (suitability >75%) for N. marmorata. Small black dots indicate
occurrence localities of N. marmorata, which were used in ENM.
N. marmorata in the predicted area before this can be re-
vised. Most importantly, the discovery of this critically
endangered restricted range species of frog outside its
previously known range highlights the importance of
taxon specific biodiversity surveys of vertebrate groups
in Sri Lanka. Hence, future surveys may even shed light
on species that are considered to be extinct such as
N. guentheri similar to the other studies that have redis-
covered other species of amphibians that were thought
to be extinct (Meegaskumbura et al., 2012; Wickra-
masinghe et al., 2012; Wickramasinghe et al., 2013a,
2013b).
Acknowledgments. We thank the Department of Wildlife
Conservation and the Forest Department of Sri Lanka for the re-
search permits (WL/3/2/79/15 and R&E/RES/NFSRCM/
2016-02) and for the continuous support of our work. The
Mohamed Bin Zayed Species Conservation Fund is highly ap-
preciated for generously funding this study. We appreciate the
assistance given by Gayan Abeysinghe and Krishanthi Bandara
with laboratory work. We are thankful to Mendis Wickrama-
singhe, Anslem de Silva and Miguel Vences for sharing their in-
sights on the N. marmorata locations. Assistance rendered by
Ruwandi Ranasinghe of the IBMBB for DNA sequencing is
highly appreciated. We thank the editor Natalia Ananjeva and
the anonymous reviewer for their comments that improved the
manuscript. Finally, we appreciate the advice given by Senani
Karunaratne on ENM.
REFERENCES
Blaustein A. R., Walls S. C., Bancroft B. A., Lawler J. J.,
Searle C. L., and Gervasi S. S. (2010), “Direct and indi-
rect effects of climate change on amphibian populations,”
Diversity,2(2), 281 – 313.
Clarke B. T. (1983), “A morphological re-examination of the
frog genus Nannophrys (Anura: Ranidae) with comments
on its biology, distribution and relationships,” Zool. J. Linn.
Soc.,79(4), 377 – 398.
de Silva A., Goonewardene S., Drake J., and de Silva P.
(2005), “The Amphibian diversity in the Knuckles massif
with special reference to the relict species,” Lyriocephalus
(Special Issue), 6(1/2), 163 – 172.
Drummond A. J., Ashton B., Cheung M., Heled J.,
Kearse M., Moir R., Stones-Havas S., Thierer T., and
Wilson A. (2009), Geneious. Ver. 5.6, Biomatters Limited,
http://www.geneious.com
Dutta S. K., and Manamendra-Arachchi K. (1996), The Am-
phibian Fauna of Sri Lanka, Wildlife Heritage Trust of Sri
Lanka, Colombo, Sri Lanka.
ESRI (2014), ArcGIS Desktop: Release 10.3, Environmental
Systems Research Institute, Redlands, CA.
Fernando S. S., Wickramasinghe L. J. M., and Rodri-
go R. K. (2007), “A new species of endemic frog belonging
to genus Nannophrys Günther, 1869 (Anura: Dicrogloss-
inae) from Sri Lanka,” Zootaxa,1403(1), 55 – 68.
Hoffmann M., Hilton-Taylor C., Angulo A., Böhm M., et al.
(2010), “The Impact of Conservation on the Status of the
World’s Vertebrates,” Science,330(6010), 1503.
IUCN (2007l), Biodiversity Assessment of the Moragahakan-
da Agriculture Development Project. Draft Final Report,
IUCN, Colombo, Sri Lanka.
IUCN (2012), IUCN Red List Categories and Criteria. Ver.
3.1. Second edition, IUCN, Gland, Switzerland and Cam-
bridge, UK.
Kirtisinghe P. (1946), “The genus Nannophrys Gunther (Am-
phibia, Ranidae) with the description of a new species,”
Ceylon J. Sci. B,23(2), 105 – 107.
Kumar S., Stecher G., and Tamura K. (2016), “MEGA7:
Molecular Evolutionary Genetics Analysis Version 7.0 for
Bigger Datasets,” Mol. Biol. Evol.,33(7), 1870 – 1874.
Li Y., Cohen J. M., and Rohr J. R. (2013), “Review and syn-
thesis of the effects of climate change on amphibians,” In-
tegr. Zool.,8(2), 145 – 161.
Manamendra-Arachchi K., de Silva A., and Pethiyagoda R.
(2008), “Nannophrys marmorata,” The IUCN Red List of
Threatened Species,2008, e.T58391A11766060, http://dx.
doi.org/10.2305/IUCN.UK.2008.RLTS.T58391A11766060.
en (accessed October 10, 2017).
Manamendra-Arachchi K. and Meegaskumbura M. (2012),
“The Taxonomy and Conservation Status of Amphibians in
Sri Lanka,” in: D. K. Weerakoon, and S. Wijesundara
(eds.), The National Red List 2012 of Sri Lanka: Conserva-
tion Status of the Fauna and Flora, Ministry of Environ-
ment, Colombo, Sri Lanka, pp. 88 – 98.
Marsh D. M. and Trenham P. C. (2001), “Metapopulation
Dynamics and Amphibian Conservation,” Conserv. Biol.,
15(1), 40 – 49.
Meegaskumbura M., Bossuyt F., Pethiyagoda R., Mana-
mendra-Arachchi K., Bahir M., Milinkovitch M., and
Schneider C. (2002), “Sri Lanka: an amphibian hot spot,”
Science,298(5592), 379 – 379.
Meegaskumbura M., Manamendra-arachchi K., Bowat-
te G., and Meegaskumbura S. (2012), “Rediscovery of
Pseudophilautus semiruber, a diminutive shrub frog (Rha-
cophoridae: Pseudophilautus) from Sri Lanka,” Zootaxa,
3229(2012), 58 – 68.
Peterson A. T., Egbert S. L., Sanchez-Cordero V., and
Price K. P. (2000), “Geographic analysis of conservation
priorities using distributional modelling and complementar-
ity: endemic birds and mammals in Veracruz, Mexico,”
Biol. Conserv.,93(1), 85 – 94.
Phillips S. J., Anderson R. P., and Schapire R. E. (2006),
“Maximum entropy modeling of species geographic distri-
butions,” Ecol. Model.,190(3), 231 – 259.
Pyron R. A. and Wiens J. J. (2011), “A large-scale phylogeny
of Amphibia including over 2800 species, and a revised
classification of extant frogs, salamanders, and caecilians,”
Mol. Phylogen. Evol.,61(2), 543 – 583.
New Localities and Habitat Modeling of Nannophrys marmorata 39
Senevirathne G. and Meegaskumbura M. (2015), “Life
among crevices: osteology of Nannophrys marmorata
(Anura: Dicroglossidae),” Zootaxa,4032(2), 241 – 245.
Senevirathne G., Samarawikrama V. A. M. P. K, Wijayati-
laka N., Manamendra-Arachchi K., Bowatter G., Sama-
rawikrama D. R. N. S., and Meegaskumbura M. (2018),
“A new frog species from rapidly dwindling cloud forest
streams of Sri Lanka — Lankanectes pera (Anura, Nyctiba-
trachidae),” Zootaxa,4461(4), 519 – 538.
Vences M., Wanke S., Odierna G., Kosuch J., and Veith M.
(2000), “Molecular and karyological data on the south
Asian ranid genera Indirana, Nyctibatrachus and Nanno-
phrys (Anura: Ranidae),” Hamadryad,25(2), 75 – 82.
Vences M., Thomas M., van der Meijden A., Chiari Y., and
Vieites D. R. (2005), “Comparative performance of the 16S
rRNA gene in DNA barcoding of amphibians,” Front.
Zool.,2(1), 5.
Wickramasinghe L. J. M., Vidanapathirana D. R., and
Wickramasinghe N. (2012), “Back from the dead: The
world’s rarest toad Adenomus kandianus rediscovered in Sri
Lanka,” Zootaxa,3347(2012), 63 – 68.
Wickramasinghe L. J. M., VIdanapathirana D. R., Airya-
rathne S., Rajeev G., Chanaka A., Pastorini J., Chathu-
ranga G., and Wickramasinghe N. (2013a), “Lost and
found: One of the world’s most elusive amphibians, Pseu-
dophilautus stellatus (Kelaart 1853) rediscovered,” Zoo-
taxa,3620(1), 112 – 128.
Wickramasinghe L. J. M., VIdanapathirana D. R., Ra-
jeev G., and Wickramasinghe N. (2013b), “Rediscovery
of Pseudophilautus hypomelas (Günther, 1876) (Amphibia:
Anura: Rhacophoridae) from the Peak Wilderness, Sri
Lanka, a species thought to be extinct!” J. Threat. Taxa,
5(17), 5181 – 5193.
Wickramasinghe L. J. M., Bandara I. N., Vidanapathi-
rana D. R., Tennakoon K. H., Samarakoon S. R., and
Wickramasinghe N. (2015), “Pseudophilautus dilmah,a
new species of shrub frog (Amphibia: Anura: Rhacophori-
dae) from a threatened habitat Loolkandura in Sri Lanka,”
J. Threat. Taxa,7(5), 7089 – 711.
40 Kanishka D. B. Ukuwela et al.
