Content uploaded by Robin Kurian Abraham
Author content
All content in this area was uploaded by Robin Kurian Abraham on Jun 15, 2018
Content may be subject to copyright.
250
Accepted by M. Vences: 26 Apr. 2018; published: 15 Jun. 2018
ZOOTAXA
ISSN 1175-5326 (print edition)
ISSN
1175-5334
(online edition)
Copyright © 2018 Magnolia Press
Zootaxa 4434 (2): 250
–
264
http://www.mapress.com/j/zt/
Article
https://doi.org/10.11646/zootaxa.4434.2.2
http://zoobank.org/urn:lsid:zoobank.org:pub:6AD09FAF-DA28-43E0-9546-029E31C6D800
Elevational size variation and two new species of torrent frogs from
Peninsular Malaysia (Anura: Ranidae: Amolops Cope)
CHAN KIN ONN
1,5
, ROBIN KURIAN ABRAHAM
2
, JESSE L. GRISMER
3
& L. LEE GRISMER
4
1
Department of Biological Sciences, National University of Singapore, 14 Science Drive 4, Singapore 117543.
E-mail: chankinonn@gmail.com
2
Biodiversity Institute and Department of Ecology and Evolutionary Biology, University of Kansas, Lawrence, Kansas 66045, USA.
E-mail: robinabrahamf50@gmail.com
3
Department of Biological Sciences, University of Auburn, Alabama 36849, USA; La Kretz Center for Californian Conservation Sci-
ence, Institute of the Environment and Sustainability, University of California Los Angeles, Los Angeles, California 90095, USA
Email: jlg0067@auburn.edu
4
Herpetology Laboratory, Department of Biology, La Sierra University, 4500 Riverwalk Parkway, Riverside, California 92515, USA.
E-mail: lgrismer@lasierra.edu
5
Corresponding author
Abstract
Previously, only one species of torrent frog (Amolops larutensis) was thought to occur throughout Peninsular Malaysia.
However, genomic work has demonstrated that populations from eastern Peninsular Malaysia form two separate lineages
that are genetically distinct from A. larutensis that is now restricted to the western half of Peninsular Malaysia. This study
demonstrates that all three lineages can be morphologically distinguished from each other, thereby providing additional
support for the recognition of the eastern lineages as two distinct species. These lineages are described herein as Amolops
gerutu sp. nov. from the eastern states of Kelantan, Terengganu, and Pahang, and A. australis sp. nov. from the southern-
most state of Johor. In general, these two new species form a clade that is sister to A. larutensis and can be readily distin-
guished from it by having: (1) considerably denser and more pronounced dorsal tubercles, and (2) the posterodorsal sur-
face of thighs having dense, dark stippling as opposed to broad vermiculations. Although differences in other
morphometric characters were detected, their utility as diagnostic characters should be applied with caution due to the
large intraspecific variation that overlaps among different species in many of the characters we measured. As such, we
advocate for the use of tuberculation and pattern of the posterodorsal portion of the thighs as primary diagnostic charac-
ters. These characters can readily distinguish A. larutensis from the two new species. To differentiate A. australis sp. nov.
from A. gerutu sp. nov. and A. larutensis, body size can be a good diagnostic character as A. australis sp. nov. is signifi-
cantly smaller in both males (mean = 31.04 ± 1.59 mm) and females (mean = 46.48 ± 3.2 mm). Additionally, we show a
strong positive correlation between body size and elevation, with populations from montane forests (>900 m asl) being
considerably larger than populations at lower elevations.
Key words: Taxonomy, systematics, morphology, amphibian, cryptic species, body size
Introduction
Torrent frogs of the genus Amolops are represented by 53 species and are widely distributed across Asia from
northeastern India, southern China, and southwards throughout mainland Southeast Asia (Frost 2018). The bulk of
this group’s species diversity lies in the northern part of its distribution (southern China and northern Indochina),
with only one species, A. larutensis (Boulenger) occurring in the southern-most portion of its range in Peninsular
Malaysia (Chan et al. 2010). However, a comprehensive molecular study of numerous Amolops populations
throughout Peninsular Malaysia revealed hidden diversity within A. larutensis (Chan et al. 2017). Using a suite of
species delimitation and gene flow analyses based on morphology, mitochondrial and genome-wide single
nucleotide polymorphism (SNP) data, this species complex was shown to comprise three independently evolving
Zootaxa 4434 (2) © 2018 Magnolia Press
·
251
NEW TORRENT FROGS FROM PENINSULAR MALAYSIA
lineages composed of the true A. larutensis and two undescribed species (Fig. 1; Chan et al. 2017). In this study, we
performed additional morphological analyses using data generated from Chan et al. (2017) to substantiate the
recognition of those lineages as distinct species. Due to large variations in body size observed among populations
across different elevational gradients, we also performed statistical analyses to determine whether body size
correlates significantly with elevation.
FIGURE 1. Distribution of sampled populations and an ultrametric maximum-likelihood phylogeny inferred from 1,466 bp of
the 16S rRNA-encoding mitochondrial gene. All major nodes were highly supported with >90% bootstrap. Populations
represented by circles belong to the western clade, triangles represent populations from the eastern clade, and star denotes the
type locality of Amolops larutensis at Bukit Larut, Perak. The red box indicates a contact zone between the eastern and western
clades. This figure is adapted from Chan et al. (2017).
Materials and methods
Morphological data were obtained from 141 vouchered specimens from 20 different localities throughout
Peninsular Malaysia. These localities include the type locality of A. larutensis and other sites from all the major
mountain ranges (Fig. 1). For consistency, we follow the same naming convention as Chan et al. (2017) i.e.
Larutensis (populations from Sungai Sedim and Bukit Hijau, state of Kedah; Bukit Larut and Ulu Kenas, state of
Perak), W1 (Belum-Temenggor, state of Perak), W2 (Cameron Highlands, state of Pahang), W3 (Fraser’s Hill,
state of Pahang), and W4 (Genting Highlands and surrounding lowlands, states of Pahang and Selangor) for
CHAN ET AL.
252
·
Zootaxa 4434 (2) © 2018 Magnolia Press
populations assigned to the western clade; E1 (Gunung Tebu and Lata Tembakah, state of Terengganu; Gunung
Stong, state of Kelantan; Sungai Lembing, state of Pahang) and E2 (Endau-Rompin, state of Johor) for populations
from the eastern clade (Fig. 1). Nine continuous morphological characters were measured following Chan et al.
(2016): snout-vent length (SVL), head length (HL), head width (HW), internarial distance (IND), snout length
(SNL), forearm length (FAL), femur length (FL), tibia length (TBL), and third-finger disc width (3FinDW). Details
on sampling localities and raw morphological measurements can be obtained from Table S1 in Chan et al. (2017).
Due to obvious sexual size dimorphism, adult male and female measurements were analyzed separately and
measurements were adjusted for allometric growth using the following equation: X
adj
=X-β(SVL-SVL
mean
), where
X
adj
=adjusted value; X=measured value; β=unstandardized regression coefficient for each population;
SVL=measured snout-vent length; SVL
mean
=overall average SVL of all samples (Lleonart et al. 2000). Adjusted
variables were then log-transformed prior to downstream analyses. We used principal components analysis (PCA)
to find the best low-dimensional representation of variation in the data to determine whether morphological
variation could form the basis of detectable group structure. Eigenvalues >1 were retained according to Kaiser’s
criterion (Kaiser 1960). The R package “hypervolume” (Blonder et al. 2014) was used to construct hypervolumes
using Gaussian kernel density estimation to estimate the probability density function of the retained principal
components.
An ANOVA was performed to evaluate whether the means of morphological characters differed significantly,
followed by a Tukey HSD test to determine specifically, which pair of character means differed after adjusting for
multiple testing. To examine patterns in body size variation, a Pearson’s correlation test was performed to examine
the relationship between SVL and elevation. Finally, we performed a T-test to determine if populations at montane
forests (>900 m) differ significantly in body size (SVL) from populations at lower elevations (<650 m). These
elevational thresholds were used because 900 m asl is the lowermost level of cloud base in Malaysia where hill
dipterocarp forests transition to montane forests (Symington 1943). All morphological analyses were performed
and visualized in R (R Core Team 2017). All type material are deposited at the La Sierra University Herpetological
Collection (LSUHC), La Sierra University, Riverside, California, USA.
FIGURE 2. Principal components scores of morphological variables visualized as hypervolumes constructed using kernel
density estimation. Geometry of hypervolumes correspond to a minimum convex hull (polytopes) that minimally encloses the
data. Axes show the first three principal components and their proportion of variance.
Results
To provide context, we recapitulate relevant results from Chan et al. (2017) that demonstrated the genetic
Zootaxa 4434 (2) © 2018 Magnolia Press
·
253
NEW TORRENT FROGS FROM PENINSULAR MALAYSIA
differentiation of the undescribed lineages. Phylogenetically, E1 + E2 formed a clade that is sister to Amolops
larutensis (Fig. 1). The split between A. larutensis and E1 + E2 were 14–16% divergent at the 16S mitochondrial
gene (uncorrected p-distance), while E1 and E2 were 7–8% divergent from each other [Fig. S2 in Chan et al.
(2017)]. Species delimitation methods based on mitochondrial and genome-wide SNPs inferred between 5–7
putative species. However, using a population genomics approach, gene flow was shown to occur among
populations from the western clade (Larutensis, W1, W2, W3, and W4) but populations from the eastern clade (E1
and E2) remained genetically isolated from each other and from any of the western clade populations [Fig. 4 in
Chan et al. (2017)]. Consequently, populations from the western clade were considered a single, cohesive
metapopulation lineage under the name A. larutensis, whereas the two populations from the eastern clade (E1 and
E2) were considered distinct, undescribed lineages.
FIGURE 3. Boxplots representing the log-transformed and body-size corrected morphological characters in males. Clusters
with the same letter code are not significantly different (Tukey HSD test at p<0.05). Blue=E2; purple=E1; orange=Amolops
larutensis.
CHAN ET AL.
254
·
Zootaxa 4434 (2) © 2018 Magnolia Press
FIGURE 4. Boxplots representing the log-transformed and body-size adjusted morphological characters in females. Clusters
with the same letter code are not significantly different (Tukey HSD test at p<0.05). Blue=E2; purple=E1; orange=Amolops
larutensis.
The E2 population is on average smaller than populations of A. larutensis and E1 for both males and females
(Table 1). For the male only dataset, the first three principal components had eigenvalues >1 and accounted for
69.6% of the total variation. The first principal component (PC1) accounted for 39.2% of the variation and was
most heavily loaded on the character TBL (Table 2). Along the PC1 axis, E1 and E2 showed complete separation
with no overlap, whereas E2 and A. larutensis were mostly separated with a slight overlap (Fig. 2). There were no
clear separations along PC2 (18%) and PC3 (12.4%). For the female dataset, the first three principal components
accounted for 82% of the total variation. The E2 population was markedly segregated from E1 and A. larutensis
along PC1 (50.5%), which was most heavily loaded on the characters IND, SNL, and FAL (Table 2). Similarly,
there were no clear separations along PC2 (20.9%) and PC3 (10.5%).
The ANOVA showed that the mean of each morphological character was significantly different except SNL in
males and TBL in females (Table 3). Subsequently, the Tukey HSD test (α = 0.05) revealed that E1 and E2 differed
significantly in all characters except male SNL and female TBL. Between A. larutensis and E1, males did not differ
significantly in SVL, IND, SNL, FAL, and Fin3DW, whereas females showed no significant differences in HW,
Zootaxa 4434 (2) © 2018 Magnolia Press
·
255
NEW TORRENT FROGS FROM PENINSULAR MALAYSIA
SNL, FAL, and FL. When A. larutensis was compared with E2, males were not significantly different in HL, HW,
SNL, FL, and Fin3DW, whereas females showed no significant differences in TBL (Table 3; Figs. 3, 4). The
Pearson’s correlation test showed a significant positive relationship between SVL and elevation and a student’s t-
test showed that populations from montane forests (>900 m asl) were significantly larger than populations from
lower elevations (<650 m asl) (Fig. 5).
FIGURE 5. Top: scatterplots of snout-vent length (SVL) vs. elevation with the corresponding regression line and 95%
confidence interval shaded in gray. Inset values represent the correlation coefficient (R) and p-value for the Pearson’s
correlation test. Bottom: boxplots of SVL vs. elevation binned into high (>900 m) and low (<650 m) elevation categories.
Blue=E2; purple=E1; orange=Amolops larutensis.
CHAN ET AL.
256
·
Zootaxa 4434 (2) © 2018 Magnolia Press
In concert with results from Chan et al. (2017), this study demonstrates that A. larutensis, E1, and E2 are not
only genetically distinct but can also be morphologically distinguished from each other. These lines of evidence
provide sufficient support for the recognition of E1 and E2 as separate species that we describe below.
TABLE 1. Summary of morphometric measurements for males and females of Amolops larutensis, A. gerutu sp. nov.,
and A. australis sp. nov.
Systematics
Amolops gerutu sp. nov.
Tuberculated Torrent Frog
Figs. 6, 7B, D
Amolops larutensis, Sumarli, Grismer, Anuar, Muin & Quah, 2015, pp 4,9,12.
Holotype. LSUHC 11178, adult female, collected 2 September 2012 by Lee Grismer, Chan Kin Onn, Alexandra
Sumarli, Evan Quah, Mohd. Abdul Muin, Shahrul Anuar, Ariel Loredo, and Anthony Cobos from Gunung Tebu,
Terengganu, Malaysia (05°35′35.47″ N, 102°36′45.63″ E; 1039 m asl).
Paratopotypes. LSUHC 11179, adult male and 11176–77, 11187–88, adult females bear the same collection
data as the holotype.
Description of holotype (Fig. 6). Adult female; habitus moderately robust; head slightly longer than wide;
snout sharply angular, obtusely pointed in dorsal profile, slightly projecting beyond lower jaw in lateral profile;
nostrils located laterally, closer to tip of snout than to eye; canthus rostralis sharp, constricted anteriorly; lores
vertical, distinctly concave; eye diameter less than snout length; pineal body visible, located level to the anterior
margin of orbits; tympanum distinct, tympanic rim slightly elevated; vomerine teeth well-developed, on two
oblique, oval ridges almost contacting each other; tongue cordiform.
Forelimbs moderately robust; relative length of fingers, II=IV<I<III; fingers without web; tips of fingers
expanded into large discs bearing circummarginal and transverse dorsal and ventral grooves, discs of third and
fourth fingers distinctly larger than discs of first and second fingers; fringe of skin along lateral edges of all fingers;
subarticular tubercles slightly raised, proximal subarticular tubercle on fourth finger oval, others round; inner
metacarpal tubercle elongate, medial and outer metacarpal tubercle oval, in contact, outer slightly smaller than
Male Female
A. larutensis
n=43
A. gerutu
n=7
A. australis
n=8
A. larutensis
n=32
A. gerutu
n=29
A. australis
n=6
SVL 31.5–40.4
(35.95 ± 2.2)
34–37.2
(35.67 ± 1.25)
28.7–32.7
(31.04 ± 1.59)
46.4–66
(57.62 ± 4.86)
45–58.2
(53.27 ± 0.47)
45.8–47
(46.48 ± 3.2)
HL 11.77–15.1
(13.45 ± 0.85)
13.2–14.2
(13.77 ± 0.34)
11.5–12.6
(12.16 ± 0.38)
17.3–25
(21.38 ± 1.87)
17.2–21.4
(19.57 ± 0.51)
16.5–17.9
(17.15 ± 1.08)
HW 10.7–13.7
(12.1 ± 0.81)
11.9–12.8
(12.43 ± 0.3)
10.1–11.3
(10.76 ± 0.44)
16.2–24.4
(19.78 ± 2.03)
16.6–20.3
(18.17 ± 0.33)
15.3–16.2
(15.65 ± 0.99)
IND 3.3–4.5
(3.97 ± 0.29)
3.8–4.1
(3.93 ± 0.11)
3.3–3.8
(3.54 ± 0.14)
5–7.1
(5.92 ± 0.54)
4.9–6.4
(5.7 ± 0.2)
4.8–5.3
(5 ± 0.36)
SNL 5–6.3
(5.59 ± 0.33)
5.3–6.1
(5.63 ± 0.28)
4.8–5.5
(5.14 ± 0.22)
7.2–10
(8.74 ± 0.68)
7.3–9.1
(8.21 ± 0.32)
7.0–7.8
(7.37 ± 0.41)
FAL 7.3–10.9
(8.94 ± 0.88)
7.9–9.3
(8.86 ± 0.49)
6.4–7.1
(6.81 ± 0.29)
10.5–15.7
(13.17 ± 1.35)
10.7–13.1
(12.04 ± 0.63)
8.6–10.3
(9.55 ± 0.72)
FL 17.7–23.3
(20.31 ± 1.47)
20.3–22.3
(20.9 ± 0.73)
15.4–19.5
(17.53 ± 1.18)
26.1–37.7
(31.47 ± 2.96)
27.1–33
(29.6 ± 1.09)
22.9–25.6
(24.33 ± 1.39)
TBL 19–24.1
(21.3 ± 1.44)
20.7–23
(21.99 ± 0.91)
17.3–19.5
(18.11 ± 0.82)
26.5–38.6
(33.08 ± 3.01)
28.6–34.4
(31.62 ± 0.9)
25–27.2
(25.87 ± 1.41)
Fin3DW 1.7–2.6
(2.11 ± 0.23)
1.9–2.5
(2.26 ± 0.19)
1.7–2.1
(1.81 ± 0.14)
2.6–4.2
(3.42 ± 0.35)
3.1–4.1
(3.48 ± 0.25)
2.6–3.3
(2.9 ± 0.24)
Zootaxa 4434 (2) © 2018 Magnolia Press
·
257
NEW TORRENT FROGS FROM PENINSULAR MALAYSIA
medial; one supernumerary tubercle posterior to proximal subarticular tubercle on second, third and fourth fingers,
that on second finger barely visible.
FIGURE 6. Holotype (LSUHC 11178) of Amolops gerutu from Gunung Tebu, Terengganu.
Hind limbs robust; tips of toes expanded into round discs bearing circummarginal and transverse dorsal and
ventral grooves, width of fourth toe disc less than width of third finger disc; toes fully webbed; wide, movable flap
of skin on inner side of first toe and a narrower flap of skin on outer side of fifth toe from the level of the proximal
subarticular tubercle to toe disc; subarticular tubercles distinct, oval; inner metatarsal tubercle elongate, outer
round, approximately one third size of inner.
Skin on dorsum and dorsal surfaces of forelimbs densely covered with irregularly-sized, raised tubercles;
tubercles on dorsolateral region slightly enlarged, elongated, ridge-like; tubercles on flanks larger than those on
dorsum; tubercles on dorsal surfaces of hind limbs sparser; supratympanic fold absent; skin on gular region and chest
smooth, belly slightly granular; anterior half of ventral surface of thigh smooth, posterior half granular; large pectoral
glands present.
Measurements of holotype (mm): SVL=58.2; HL=21.4; HW=20.2; IND=6.2; SNL=9.1; FAL=13; FL=31.3;
TBL=33.6; Fin3DW=3.8. Additional measurements: horizontal diameter of orbit=6.6; interorbital distance=5.3;
largest diameter of tympanum=2.6; distance between anterior margin of tympanum and posterior margin of orbit=2.7
Diagnosis. Amolops gerutu sp. nov. is placed in the genus Amolops based on its overall morphological
similarity with A. larutensis and phylogenetic placement as its sister lineage (Chan et al. 2017). It can be
distinguished from other congeners by the following combination of characters: adult females large (SVL 45.0–
58.2 mm), males moderate in size (SVL 34.0–37.2 mm); dorsum densely covered with irregularly-sized, raised
tubercles; tubercles on dorsolateral region slightly enlarged, elongated, ridge-like; tubercles on flanks larger than
CHAN ET AL.
258
·
Zootaxa 4434 (2) © 2018 Magnolia Press
those on dorsum; tubercles on dorsal surfaces of hind limbs sparser; posterodorsal surface of thighs with dense,
dark stippling.
FIGURE 7. Top row: (A) female Amolops larutensis from Fraser’s Hill, Pahang; (B) female A. gerutu from Chemerong,
Pahang; Middle row: (C) male A. larutensis from Fraser’s Hill; (D) male A. gerutu from Sekayu, Terengganu; Bottom row: (E)
color-pattern comparisons of the posterodorsal portion of the thigh that diffentiates A. larutensis from (F) A. gerutu and (G) A.
australis.
Coloration in preservative. Dorsum dark-brown, overlain with irregular light-gray patches; flanks yellowish
white with dark-brown blotches; lips yellowish white, mottled with dark-brown; dorsal surface of fore and hind
limbs light-brown with distinct, wide, dark-brown crossbars; dorsal surface of third and fourth fingers brown with
dark-brown crossbars, second finger yellowish white, mottled with dark-brown, first finger yellowish white with
minute brown speckling; posterodorsal surface of thighs with dense, dark stippling; toe webbing light-brown with
Zootaxa 4434 (2) © 2018 Magnolia Press
·
259
NEW TORRENT FROGS FROM PENINSULAR MALAYSIA
fine, dense, dark-brown veins; venter yellowish white; ventral surface of hands speckled with dark-brown; dark-
brown stripe on outer side of forearm; gular with irregular, dark-brown streaks; anterior half of ventral surface of
thighs orange, posterior half dark-brown with light-colored spots; ventral surface of feet and outer half of tarsus
dark-brown; pectoral glands orangish.
Va ri at io n . Males differ from females in being smaller; having paired vocal sacs; distinct nuptial pads on base
of first finger; lacking humeral glands; and having indistinct pectoral glands.
Comparisons. Because the new species forms a clade with A. larutensis and is substantially distant genetically
from all other congeners, we restrict our comparisons to A. larutensis and the other undescribed species. Amolops
gerutu sp. nov. differs from A. larutensis by having denser and more pronounced dorsal tubercles and
posterodorsal side of thighs having dense, dark stippling as opposed to vermiculations (Fig. 7E, F). After correcting
for body size, males of A. gerutu sp. nov. have larger heads (HL and HW) and longer hind limbs (FL and TBL; Fig.
3), whereas females have smaller SVL, HL, and larger IND, TBL, and Fin3DW (Table 3; Fig. 4). See the
Discussion section for the use of continuous morphological characters to diagnose this species.
TABLE 2 . Summary of PCA results for male and female morphological datasets. SD = standard deviation, prop. var =
proportion of variation; cum. var = cumulative variance, EV = eigenvalues. Other abbreviations relate to morphological
characters that are defined in materials and methods.
Distribution. Besides the type locality, Amolops gerutu sp. nov. has been documented from a number of other
Male PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9
SD 1.88 1.27 1.06 0.90 0.76 0.70 0.67 0.52 0.38
Prop. var 0.39 0.18 0.12 0.09 0.06 0.05 0.05 0.03 0.02
Cum. var 0.39 0.57 0.70 0.79 0.85 0.90 0.95 0.98 1.00
EV 3.53 1.62 1.11 0.81 0.58 0.48 0.45 0.27 0.14
SVL -0.25 0.37 -0.52 0.12 -0.03 0.72 -0.05 -0.01 -0.05
HL -0.37 -0.24 0.11 0.27 -0.47 0.06 0.68 -0.15 -0.04
HW -0.33 -0.44 0.07 -0.12 0.53 0.24 0.17 0.50 0.24
IND -0.31 0.32 0.05 0.52 0.58 -0.30 0.10 -0.30 -0.04
SNL -0.27 -0.43 -0.14 0.52 -0.25 -0.11 -0.61 0.11 0.01
FAL -0.36 0.33 -0.29 -0.19 -0.18 -0.49 0.08 0.55 -0.23
FL -0.36 0.08 0.60 -0.20 -0.01 0.24 -0.25 -0.03 -0.59
TBL -0.43 0.26 0.23 -0.27 -0.20 -0.05 -0.21 -0.19 0.71
Fin3DW -0.28 -0.37 -0.44 -0.47 0.14 -0.16 -0.04 -0.53 -0.19
Female
SD 2.13 1.37 0.97 0.77 0.54 0.53 0.43 0.42 0.31
Prop. var 0.51 0.21 0.11 0.07 0.03 0.03 0.02 0.02 0.01
Cum. var 0.51 0.71 0.82 0.89 0.92 0.95 0.97 0.99 1.00
EV 4.55 1.88 0.95 0.60 0.29 0.28 0.18 0.18 0.09
SVL -0.27 -0.04 -0.66 0.61 -0.26 0.20 -0.05 0.09 0.03
HL 0.32 -0.25 -0.50 -0.36 -0.30 -0.47 0.28 -0.12 -0.23
HW -0.16 -0.56 -0.32 -0.34 0.54 0.22 -0.11 0.16 0.23
IND 0.42 -0.08 0.08 0.02 -0.11 0.48 0.41 0.61 -0.18
SNL 0.45 0.05 -0.05 0.00 -0.23 0.00 -0.11 -0.02 0.85
FAL 0.43 0.06 -0.12 0.01 0.05 -0.01 -0.81 0.19 -0.32
FL 0.38 -0.28 0.05 0.23 0.12 0.41 0.08 -0.71 -0.16
TBL 0.09 -0.58 0.32 0.52 0.08 -0.49 0.01 0.20 0.03
Fin3DW -0.28 -0.44 0.27 -0.25 -0.68 0.25 -0.27 -0.04 -0.03
CHAN ET AL.
260
·
Zootaxa 4434 (2) © 2018 Magnolia Press
localities east of the Titiwangsa mountain range including Gunung Stong Forest Reserve, in the state of Kelantan;
Lata Tembakah, Lata Belatan, and Sekayu Recreational Forest in the state of Terengganu (Dring 1979; Sumarli et
al. 2015); and Sungai Lembing, Sungai Pandan Waterfall, and Chemerong Amenity Forest in the state of Pahang.
At Gunung Stong, A. gerutu sp. nov. occurs in syntopy with A. larutensis (Fig. 1).
Natural history. Like most congeners, Amolops gerutu sp. nov. is a strict torrent specialist that only occurs
within or along torrential zones of rocky streams from lowland to montane forests. During the day, frogs dwell in
rock cracks and sheltered areas among boulder stacks along streams and are rarely seen out in the open. They can
be seen in abundance at night, most frequently on boulders by splash zones and occasionally on adjacent low
vegetation. When disturbed, frogs dive into the rapids and float downstream. Like other congeners, tadpoles of this
species are gastromyzophorous (Pham et al. 2015) and can be seen clinging onto boulders in the splash zone. On
such boulders, tadpoles are usually observed above or just below the water line.
Etymology. The specific epithet “gerutu” (English pronunciation “gir-roo-too”) refers to the Malay word of
the same construct, meaning “tubercle”, in reference to the pronounced dorsal tubercles that are diagnostic of this
species.
Amolops australis sp. nov.
Southern Torrent Frog
Fig. 8
Amolops larutensis, Ahmad, Senawi & Lim 2004, p 26; Belabut & Hashim, 2005, p 200; Wood, Grismer, Youmans, Nasir,
Ahmad & Senawi, 2008, p 118; Grismer & Pan, 2008, p. 277 (in part); Shahriza, Ibrahim, Anuar & Muin, 2012, p 558,
561.
Staurois larutensis, Belabut & Hashim, 2004, pp. 67, 69.
Holotype. LSUHC 7665, adult female, collected on 27 August 2005 by L. Grismer, P. L. Wood, J. L. Grismer, T.
M. Youmans, N. Nasir, J. Senawi, and N. Ahmad from Peta, Endau-Rompin National Park (2°26'20.09" N,
103°16'22.29"; 103 m asl).
Paratypes. LSUHC 7672–73, 7686, 8097–8101, adult males, and 7671, 7684, 7687, 8096, adult females, with
the same collection data as the holotype.
Description of holotype (Fig. 8). Adult female; habitus robust; head slightly wider than long; snout sharply
angular, obtusely acuminate in dorsal profile, projecting significantly beyond lower jaw in lateral profile; nostrils
located laterally, closer to tip of snout than to eye; canthus rostralis sharp, constricted anteriorly; lores vertical,
concave; eye diameter lesser than snout length; pineal body visible, located level to the anterior margin of orbits;
tympanum distinct, tympanic rim slightly elevated; vomerine teeth well developed, on two oblique, oval ridges
almost contacting each other; tongue cordiform.
Forelimbs moderately robust; relative length of fingers, II<I<IV<III; fingers without webbing; tips of all
fingers expanded into large discs bearing circummarginal and transverse dorsal and ventral grooves, discs of third
and fourth fingers distinctly larger than discs of first and second fingers; fringe of skin along lateral edges of all
fingers; subarticular tubercles slightly raised, rounded; inner metacarpal tubercle elongate, medial and outer
metacarpal tubercle oval, in contact, outer slightly smaller than medial; one supernumerary tubercle posterior to
proximal subarticular tubercle on second, third and fourth fingers, that on second finger barely visible.
Hind limbs robust; tips of all toes expanded into round discs bearing circummarginal and transverse dorsal and
ventral grooves, width of fourth toe disc less than width of third finger disc; full web on all toes; wide, movable
flap of skin on preaxial side of first toe and a narrower flap of skin on postaxial side of fifth toe from the level of the
proximal subarticular tubercle to toe disc; subarticular tubercles distinct, oval; inner metatarsal tubercle very
elongate, outer round, approximately one third size of inner.
Skin on dorsum and dorsal surfaces of forelimbs densely covered with mostly circular but also a few
irregularly-sized, raised tubercles; tubercles on dorsolateral region slightly enlarged, elongated, ridge-like;
tubercles on flanks larger than those on dorsum; tubercles almost absent on dorsal surfaces of hind limbs;
supratympanic fold absent; skin on gular and chest smooth, belly slightly granular; anterior half of ventral surface
of thigh smooth, posterior half granular.
Zootaxa 4434 (2) © 2018 Magnolia Press
·
261
NEW TORRENT FROGS FROM PENINSULAR MALAYSIA
FIGURE 8. Holotype (LSUHC 7665) of Amolops australis from Endau-Rompin (Peta), Pahang.
Measurements of holotype (mm): SVL=46.3; HL=16.5; HW=15.8; IND=4.8; SNL=7.4; FAL=9.8; FL=25.6;
TBL=26.7; Fin3DW=3.0. Additional measurements: horizontal diameter of orbit=6.4; interorbital distance=4.0;
largest diameter of tympanum=2.0; distance between anterior margin of tympanum and posterior margin of
orbit=1.9
Diagnosis. Amolops australis sp. nov. is placed in the genus Amolops based on its overall morphological
similarity with A. gerutu sp. nov. and phylogenetic placement as its sister lineage (Fig. 1). It can be distinguished
from other congeners by the following combination of characters: adult females moderate in size (SVL 45.8–47.0
mm), males smaller (SVL 28.7–32.7 mm); skin on dorsum and dorsal surfaces of forelimbs densely covered with
mostly circular but also a few irregularly-sized, raised tubercles; tubercles on dorsolateral region slightly enlarged,
elongated, ridge-like; tubercles on flanks larger than those on dorsum; tubercles almost absent on dorsal surfaces of
hind limbs; posterodorsal surface of thighs with dense, dark stippling.
Coloration in preservative. Dorsum light-gray, interspersed with irregular dark-brown spots and blotches;
flanks faded white with small brown blotches; lips off-white; dorsal surface of fore and hind limbs light-gray with
distinct, wide, dark-brown crossbars; dorsal surface of third and fourth fingers brown with dark-brown crossbars,
second finger white, mottled with dark-brown, first finger white with minute brown speckling; posterodorsal
surface of thighs with dense, dark stippling; toe webbing light-brown with fine, dense, dark-brown veins; venter
orangish white; ventral surface of hands speckled with dark-brown; dark-brown stripe on outer side of forearm;
CHAN ET AL.
262
·
Zootaxa 4434 (2) © 2018 Magnolia Press
gular region with irregular, dark-brown streaks; anterior half of ventral surface of thighs pale and faded-orange,
posterior half dark-brown with light spots; ventral surface of feet and outer half of tarsus dark-brown; pectoral
glands orangish white.
TABLE 3. Results of the ANOVA on overall mean differences and post-hoc Tukey HSD test to compare all possible
pairs of means. P-values of males and females are separated by a forward slash.
Va ri at io n . Males differ from females in being smaller; having paired vocal sacs; distinct nuptial pads on base
of first finger; lacking humeral glands; and having indistinct pectoral glands.
Comparisons. Because the new species forms a clade with A. larutensis and is substantially distant from all
other congeners, we restrict our comparisons to A. larutensis and the other undescribed species. Amolops australis
sp. nov. differs from A. larutensis by having denser and more pronounced dorsal tubercles and posterodorsal side
of thighs with dense, dark stippling as opposed to vermiculations (Fig. 7E, G). Males and females of A. australis
sp. nov. are noticeably smaller in comparison to A. larutensis and A. gerutu sp. nov. (Table 1; Figs. 3, 4).
Specifically, males of A. australis sp. nov. have smaller SVL, IND, FAL, and TBL when compared with those of A.
larutensis and smaller SVL, HL, HW, IND, FAL, FL, TBL, and 3FinDW compared to A. gerutu sp. nov. (Table 2;
Fig. 3). For females, A. australis sp. nov. have smaller SVL, HW, and 3FinDW, but larger HL, IND, SNL, FAL,
and FL when compared with A. larutensis and A. gerutu sp. nov. (Table 3; Fig. 4). See the Discussion section for
the use of continuous morphological characters to diagnose this species.
Distribution. Amolops australis sp. nov. is only known from the southern state of Johor where it has been
confirmed to occur in Endau-Rompin National Park and Bantang River Amenity Forest. It is presumed to occur
more widely in suitable habitats in the surrounding southern region of Peninsular Malaysia.
Natural history. The natural history of this species is similar to that of Amolops gerutu sp. nov. and A.
larutensis. No information is available for tadpoles.
Etymology. The specific epithet is derived from the Latin word “australis”, meaning “southern” in English,
and is applied in reference to the distribution of this species in southern Peninsular Malaysia that also represents the
southern-most distributional limit of the entire genus.
Discussion
Although significant interspecies differences were detected in numerous continuous morphological characters,
their utility as diagnostic characters should be applied with caution. This is due to the large intraspecific variation
that overlaps among different species in many of the characters we measured, especially for Amolops larutensis
and A. gerutu sp. nov. As such, we advocate for the use of tuberculation and pattern of the posterodorsal portion of
the thighs as primary diagnostic characters (Fig. 6). These characters are effective at distinguishing A. larutensis
from the two new species. To differentiate A. australis sp. nov. from A. gerutu sp. nov. and A. larutensis, body size
can be a good diagnostic character as A. australis sp. nov. is significantly smaller in both males (mean = 31.04 ±
1.59) and females (mean = 46.48 ± 3.2; Table 1).
p-value (ANOVA) Adjusted p-value (Tukey test)
A. gerutu vs. A. australis A. larutensis vs. A. gerutu A. larutensis vs. A. australis
SVL 0.000 / 0.000 0.000 / 0.000 0.961 / 0.000 0.000 / 0.000
HL 0.001 / 0.000 0.001 / 0.00.0 0.005 / 0.039 0.256 / 0.000
HW 0.003 / 0.000 0.014 / 0.000 0.002 / 0.685 0.986 / 0.000
IND 0.000 / 0.000 0.001 / 0.000 0.961 / 0.028 0.000 / 0.000
SNL 0.363 / 0.000 0.329 / 0.000 0.590 / 0.942 0.647 / 0.000
FAL 0.000 / 0.000 0.000 / 0.000 0.807 / 0.447 0.000 / 0.000
FL 0.003 / 0.000 0.006 / 0.000 0.005 / 0.832 0.679 / 0.000
TBL 0.000 / 0.062 0.000 / 0.681 0.002 / 0.049 0.000 / 0.852
Fin3DW 0.008 / 0.000 0.006 / 0.000 0.068 / 0.000 0.136 / 0.000
Zootaxa 4434 (2) © 2018 Magnolia Press
·
263
NEW TORRENT FROGS FROM PENINSULAR MALAYSIA
Our results also showed a strong positive correlation between body size and elevation with high elevation
populations (>900 m) being considerably larger than populations from lower elevations (<650 m). This pattern is
best exemplified in A. larutensis from Fraser’s Hill, Cameron Highlands, and Genting Highlands where
populations from both low and high elevations were sampled. At Fraser’s Hill, males and females averaged 33.9
(N=9) and 47.5 mm (N=2) respectively at 270 m asl (Sungai Chiling) vs. 38.5 (N=6) and 62.0 mm (N=8) at 1000 m
asl (Jeriau Falls). At Cameron Highlands, males averaged 35.4 mm (N=2) at 440 m asl (Lata Iskandar) as opposed
to 39.5 mm (N=3) at 1400 m asl (Robinson Falls), whereas at Genting Highlands, the average size of males and
females was 36.5 (N=6) and 53.6 mm (N=5) respectively at 265 m asl (Ulu Gombak, Sungai Sendat, Ulu Yam) vs.
38.8 mm (N=3) and 60.5 mm (N=4) at 980 m asl (Gotong Jaya). These results are interesting in that low and high
elevation populations were shown to exchange genes freely (Chan et al. 2017), indicating that elevational
differences in body size could be purely driven by epigenetic responses to variations in climatic regimes as
opposed to genetic makeup. However, larger sample sizes from more localities are needed to ascertain conclusively
if this trend has ecological underpinnings as opposed to being an artifact of inadequate sampling.
The east-west boundary that divides Amolops larutensis from A. gerutu sp. nov. and A. australis sp. nov.
corresponds to the lowland regions east of the Titiwangsa Range. This east-west divide has also been documented
in other vertebrate fauna (e.g. Grismer et al. 2008, 2013, 2014; Wood et al. 2008; Chan et al. 2014; Davis et al.
2016), indicating that the Titiwangsa Range serves as a biogeographic barrier that played an important role in the
diversification of multiple unrelated taxa. At Gunung Stong on the northeastern margin of the Titiwangsa Range, A.
larutensis and A. gerutu sp. nov. occur in syntopy and one hybrid individual has been documented (Chan et al.
2017). However, the low occurrence of hybrids, lack of gene flow among all other individuals and populations of
both species, coupled with the high degree of genetic divergence, suggests that these species maintain reproductive
isolation, and that infrequent hybridization events do not affect the integrity of species boundaries. The occurrence
of A. gerutu sp. nov. on Gunung Stong is the only documented locality of this species on the Titiwangsa range and
alludes to the possibility of other contact zones along areas where the habitat of these two species are contiguous or
occur in proximity.
Amolops larutensis sensu lato has also been documented from extreme southern Thailand (Chan-ard 2003) and
based on the configuration of the Bintang and Titiwangsa mountain ranges that extend into southern Thailand,
populations from this region most likely belong to A. larutensis sensu stricto. However, due to the proximity of
Gunung Stong to the Thai border, it is possible that the distribution range of A. gerutu sp. nov. also extends into
extreme southeastern Thailand and that some if not all populations in that region may belong to that species.
Acknowledgements
We thank Evan Quah, Alex Sumarli, Mohd. Abdul Muin, Shahrul Anuar, Ariel Loredo, and Anthony Cobos for
field assistance; Kelvin Lim from the Lee Kong Chian Museum of Natural History for specimen loans; and Luke
Welton from the University of Kansas Biodiversity Institute for photographic and curatorial assistance.
Literature cited
Belabut, D. & Hashim R. (2004) Herpetofauna of the western region of Endau-Rompin, Johore, Peninsular Malaysia.
Malaysian Journal of Science, 23, 65–72.
Belabut, D. & Hashim, R. (2005) Amphibians of the Lubuk Tapah and adjacent areas in southwestern Endau-Rompin National
Park, Johor, Malaysia. In: Mohamad, H. & Zakaria-Ismail, M. (Eds.), The Forests and Biodiversity of Selai Endau-
Rompin. University of Malaya, Kuala Lumpur, pp. 218.
Blonder, B., Lamanna, C., Violle, C. & Enquist, B.J. (2014) The n-dimensional hypervolume. Global Ecology and
Biogeography, 23, 595–609.
https://doi.org/10.1111/geb.12146
Chan-ard, T. (2003) A Photographic Guide to Amphibians in Thailand. Darnsutha Press Co., Ltd., Bangkok, 175 pp.
Chan, K.O., Alexander, A.M., Grismer, L.L., Su, Y.-C., Grismer, J.L., Quah, E.S.H. & Brown, R.M. (2017) Species
delimitation with gene flow: a methodological comparison and population genomics approach to elucidate cryptic species
boundaries in Malaysian Torrent Frogs. Molecular Ecology, 26, 5435–5450.
https://doi.org/10.1111/mec.14296
CHAN ET AL.
264
·
Zootaxa 4434 (2) © 2018 Magnolia Press
Chan, K.O., Belabut, D. & Norhayati, A. (2010) A revised checklist of the amphibians of Peninsular Malaysia. Russian Journal
of Herpetology, 17, 202–206.
Chan, K.O., Grismer, L.L., Zachariah, A. & Brown, R.M. (2016) Polyphyly of Asian tree toads, genus Pedostibes Günther,
1876 (Anura: Bufonidae ), and the description of a new genus from Southeast Asia. PLoS ONE, 11 (1), e0145903.
https://doi.org/10.1371/journal.pone.0145903
Chan, K.O., Wood, P.L.Jr, Anuar, S., Muin, M.A., Quah, E.S.H., Sumarli, A.X.Y. & Grismer, L.L. (2014) A new species of
upland stream toad of the genus Ansonia Stoliczka, 1870 (Anura: Bufonidae) from northeastern Peninsular Malaysia.
Zootaxa, 3764 (4), 427–440.
https://doi.org/10.11646/zootaxa.3764.4.3
Davis, H.R., Grismer, L.L., Klabacka, R.L., Muin, M.A., Quah, E.S.H., Anuar, S., Wood, P.L.Jr. & Sites, J.W. (2016) The
phylogenetic relationships of a new stream toad of the genus Ansonia Stoliczka, 1870 (Anura: Bufonidae) from a montane
region in Peninsular Malaysia. Zootaxa, 4103 (2), 137–153.
https://doi.org/10.11646/zootaxa.4103.2.4
Dring, J.C.M. (1979) Amphibians and reptiles from northern Trengganu, Malaysia, with descriptions of two new geckos:
Cnemaspis and Cyrtodactylus. Bulletin of the British Museum of Natural History, Zoology, 34, 181–241.
Frost, D.R. (2018) Amphibian Species of the World: an Online Reference. Version 6.0. American Museum of Natural History,
New York, USA. Electronic Database accessible. Available from: http://research.amnh.org/herpetology/amphibia/
index.html (accessed 10 January 2018)
Grismer, L.L. & Pan, K.A. (2008) Diversity, endemism, and conservation of the amphibians and reptiles of southern Peninsular
Malaysia and its offshore islands. Herpetological Review, 39 (3), 1–12.
Grismer, L.L., Grismer, J.L., Wood, P.L.Jr. & Chan, K.O. (2008) The distribution, taxonomy, and redescription of the geckos
Cnemaspis affinis (Stoliczka 1887) and C. flavolineata (Nicholls 1949) with descriptions of a new montane species and
two new lowland, karst- dwelling species from Peninsular Malaysia. Zootaxa, 24, 1–24.
Grismer, L.L., Wood, P.L.Jr., Anuar, S., Muin, M.A., Quah, E.S.H., McGuire, J.A., Brown, R.M., Ngo, V.T., Hong Thai, P. &
Pham, H.T. (2013) Integrative taxonomy uncovers high levels of cryptic species diversity in Hemiphyllodactylus Bleeker,
1860 (Squamata: Gekkonidae) and the description of a new species from Peninsular Malaysia. Zoological Journal of the
Linnean Society, 169, 849–880.
https://doi.org/10.1111/zoj.12064
Grismer, L.L., Wood, P.L.Jr., Anuar, S., Quah, E.S.H., Muin, M.A., Mohamed, M., Chan, K.O., Sumarli, A.X., Loredo, A.I. &
Heinz, H.M. (2014) The phylogenetic relationships of three new species of the Cyrtodactylus pulchellus complex
(Squamata: Gekkonidae) from poorly explored regions in northeastern Peninsular Malaysia. Zootaxa, 3786 (3), 359–381.
https://doi.org/10.11646/zootaxa.3786.3.6
Kaiser, H.F. (1960) The application of electronic computers to factor analysis. Educational and Psychological Measurement,
20, 141–151.
https://doi.org/10.1177/001316446002000116
Lleonart, J., Salat, J. & Torres, G.J. (2000) Removing allometric effects of body size in morphological analysis. Journal of
Theoretical Biology, 205, 85–93.
https://doi.org/10.1006/jtbi.2000.2043
Norhayati, A., Senawi, J. & Lim, B.L. (2004) A photographic guide to amphibians of Endau-Rompin State Park, Pahang. The
Forestry Department of Peninsular Malaysia, Kuala Lumpur, 59 pp.
Pham, C.T., Dogra, A., Gawor, A., Rauhaus, A., Kloeble, G., Nguyen, T.Q. & Ziegler, T. (2015) First record of Amolops
cremnobatus from Thanh Hoa Province, Vietnam, including an extended tadpole description and the first larval staging for
Amolops. Salamandra, 51, 111–120.
R Core Team (2017) A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna,
Available from: http://www.R-project.org (accessed 16 May 2018)
Shahriza, I., Shahrul, A. & Muin, M.A. (2012) Herpetofauna of Peta area of Endau-Rompin National Park, Johor, Malaysia.
Pertanika Journal of Tropical Agricultural Science, 35 (3), 553–567.
Symington, C.F. (1943) Foresters’ manual of dipterocarps. Syonan-Hakubutukan, Kuala Lumpur, 244 pp.
Sumarli, A.X., Grismer, L.L., Anuar, S., Muin, M.A. & Quah, E.S.H. (2015) First report on the amphibians and reptiles of a
remote mountain, Gunung Tebu in northeastern Peninsular Malaysia. Checklist, 11, 1–32.
https://doi.org/10.15560/11.4.1679
Wood, P.L.Jr., Grismer, L.L., Youmans, T.M., Nasir, N., Ahmad, N. & Senawi, J. (2008) Additions to the herpetofauna of
Endau-Rompin, Johor, West Malaysia. Herpetological Review, 39 (1), 112–121.
Wood, P.L. Jr., Grismer, L.L., Ahmad, N. & Senawi, J. (2008) Two new species of torrent-dwelling toads Ansonia Stoliczka,
1870 (Anura: Bufonidae) from Peninsular Malaysia. Herpetologica, 64, 321–340.
https://doi.org/10.1655/07-065.1
