ArticlePDF Available

The Taxonomy, Distribution, and Conservation Status of the Slender Loris (Primates, Lorisidae: Loris) in Sri Lanka

Authors:
  • Land Owners Restore Rainforests in Sri Lanka

Abstract and Figures

External body dimensions and proportions, skull morphology, coat coloration, vocalizations, and genetics have contributed to an increase in the number of diagnosable species among nocturnal primates. Two species of slender loris are currently recognized for Sri Lanka: the red slender loris Loris tardigradus (Linnaeus, 1758), endemic to the wet zone and montane areas; and the grey slender loris Loris lydekkerianus Cabrera, 1908, which is widespread and also occurs in India. The red slender loris has two subspecies, namely the western red slender loris Loris tardigradus tardigradus (Linnaeus, 1758) and the Horton Plains slender loris Loris tardigradus nycticeboides Hill, 1942. Loris t. tardigradus is found in the lowland wet zone and L. t. nycticeboides is restricted to the montane region of south-central Sri Lanka. Two subspecies are also ascribed to Loris lydekkerianus in Sri Lanka, namely the northern Ceylon slender loris Loris lydekkerianus nordicus Hill, 1933, and the highland slender loris Loris lydekkerianus grandis Hill and Phillips, 1932. Loris l. nordicus is found in the dry zone, and L. l. grandis is restricted to the sub-montane region of Kandy and Matale. Another two subspecies are known from southern India. We examined specimens (live and museum) from all climate/vegetation zones in Sri Lanka, for facial and pelage features, external body morphology, and skull morphology, and concluded that there are at least two species and at least six subspecies, though we suspect that some, or all, of these subspecies may be distinct species. Names are available for four of these taxa, and here we describe two new subspecies.
Content may be subject to copyright.
83
Primate Conservation 2017 (31): 83-106
The Taxonomy, Distribution, and Conservation Status of the Slender
Loris (Primates, Lorisidae: Loris) in Sri Lanka
Saman N. Gamage1, Colin P. Groves2†, Fais M. M. T. Marikar3, Craig S. Turner4,
Kalinga U. K. G. Padmalal5 and Sarath W. Kotagama1
1Department of Zoology, Faculty of Science, University of Colombo, Colombo, Sri Lanka
2School of Archaeology and Anthropology, Australian National University, Canberra, Australia
3Staff Development Centre, Faculty of Medicine, General Sir John Kotelawela Defence University, Ratmalana, Sri Lanka
4Zoological Society of London, Regent’s Park, London, UK
5Department of Zoology, The Open University of Sri Lanka, Nawala, Nugegoda, Sri Lanka
Abstract: External body dimensions and proportions, skull morphology, coat coloration, vocalizations, and genetics have con-
tributed to an increase in the number of diagnosable species among nocturnal primates. Two species of slender loris are currently
recognized for Sri Lanka: the red slender loris Loris tardigradus (Linnaeus, 1758), endemic to the wet zone and montane areas;
and the grey slender loris Loris lydekkerianus Cabrera, 1908, which is widespread and also occurs in India. The red slender loris
has two subspecies, namely the western red slender loris Loris tardigradus tardigradus (Linnaeus, 1758) and the Horton Plains
slender loris Loris tardigradus nycticeboides Hill, 1942. Loris t. tardigradus is found in the lowland wet zone and L. t. nyctice-
boides is restricted to the montane region of south-central Sri Lanka. Two subspecies are also ascribed to Loris lydekkerianus in
Sri Lanka, namely the northern Ceylon slender loris Loris lydekkerianus nordicus Hill, 1933, and the highland slender loris Loris
lydekkerianus grandis Hill and Phillips, 1932. Loris l. nordicus is found in the dry zone, and L. l. grandis is restricted to the sub-
montane region of Kandy and Matale. Another two subspecies are known from southern India. We examined specimens (live
and museum) from all climate/vegetation zones in Sri Lanka, for facial and pelage features, external body morphology, and skull
morphology, and concluded that there are at least two species and at least six subspecies, though we suspect that some, or all, of
these subspecies may be distinct species. Names are available for four of these taxa, and here we describe two new subspecies.
Key words: taxonomy, Loris lydekkerianus, Loris tardigradus, morphology, slender loris, Sri Lanka
Introduction
The slender lorises (Suborder Strepsirrhini, Family Lori-
sidae, Loris É. Geoffroy Saint-Hilaire, 1796) are small noc-
turnal primates found only in India and Sri Lanka (Groves
1998, 2001; Nekaris and Jayawardene 2004). Groves (1998)
recognized two species of slender loris for Sri Lanka: the red
slender loris Loris tardigradus (Linnaeus, 1758) and the grey
slender loris Loris lydekkerianus Cabrera, 1908. The former
is endemic to Sri Lanka, whereas the latter species is also
found in southern India.
The oceanic island of Sri Lanka is 65,610 km2, and is sep-
arated from India by the 19-km-wide Palk Strait (Wijesinghe
et al. 1993). Sri Lanka, along with the Western Ghats of India,
is remarkable for its biodiversity, and is one of the world’s
Biodiversity Hotspots following the analyses and parameters
of Mittermeier et al. (2004). Our knowledge of its biodiver-
sity is, however, still highly dependent on surveys completed
a century or more ago (Pethiyagoda 2005). Recent taxonomic
studies are scarce for small mammals and primates (Weera-
koon and Goonatilake 2006) and the inventory is evidently
far from complete. This is highlighted by the descriptions
of a new species of mouse deer Moschiola kathygre Groves
and Meijaard, 2005, a new species of shrew Suncus montanus
†Colin P. Groves 1942-2017
Gamage et al.
84
Meegaskumbura and Schneider, 2008, and a new species
of Golden palm civet Paradoxurus stenocephalus Groves,
Rajapaksha and Manemandra-Arachchi, 2009, over the last
seven years.
The slender loris was rst described as Lemur tardigradus
Linnaeus, 1758, based on an illustration in Seba (1735), prob-
ably depicting a red slender loris from Sri Lanka (Thomas,
1908). Geoffroy Saint-Hilaire (1796), under the impression
that Linnaeus had described a slow loris, described the slen-
der loris as a new genus and species, Loris gracilis. The
generic name was conserved by the International Commission
on Zoological Nomenclature (1999). Until the 20th century,
Linnaeus's name was taken by almost all authors to be based
on a slow loris (genus Nycticebus). It was Stone and Rehn
(1902) who argued that Linnaeus’s name tardigradus actu-
ally referred to a slender loris, and Thomas (1908) concurred,
tightening and extending aspects of their argument. Gentry
et al. (1998) later designated a lectotype in the Stockholm
Museum of Natural History.
Fischer (1804) recognized É. Geoffroy Saint-Hilaire’s
(1796) Loris gracilis as a reddish species (“schlanker Loris”),
from Ceylon (Sri Lanka), and in addition described Loris cey-
lonicus (“ceylonischer Loris”) as a yellowish-brown species.
From the descriptions, it is difcult to determine whether
these do or do not refer to the same taxon, and in the absence
of type specimens it is impossible to say with certainty. Pro-
visionally both names may be placed as junior synonyms of
Linnaeus’s name. The measurements given by Fischer (1804:
163−166) are apparently within the range of what is here con-
sidered to be Loris tardigradus tardigradus (Linnaeus, 1758):
Loris gracilis has a greatest skull length from the tip of the
snout to the convexity of the occiput of 48 mm, while in Loris
ceylonicus this measurement is 50 mm.
Lesson (1840) renamed the reddish slender loris from
“l’île de Ceylan” as Arachnocebus lori, but apparently for the
rst time described one of the blackish forms (likewise from
“l’île de Ceylan”) under the name Bradylemur tardigradus
var. c—that is to say, he failed to recognize its afnity to the
reddish slender loris and referred it to the genus Bradylemur,
which he had erected for slow lorises (now Nycticebus).
From 1840 to 1905, no new species or subspecies of loris
were added to the Sri Lankan loris fauna. Then Lydekker
(1905), evidently unaware of Stone and Rehn’s (1902) paper,
and taking two mounted specimens from Madras (now Chen-
nai) as “typical” for Loris gracilis, described “the Ceylon
Loris” as Loris gracilis zeylanicus on the evidence of another
mounted specimen (this is BM 1904.10.12.3, with no precise
locality other than “Ceylon”, according to Jenkins, 1987).
Although not strictly a homonym of Fischer’s Loris ceyloni-
cus, Recommendation 58A of the Code states that “An author
should not base a new species-group name on a personal or
geographical name if another name derived from the same
word or from words of the same meaning (even if differ-
ently formed) is in use in the same or an allied or associated
genus…” Notwithstanding, the name has not been used for
over a century, and so ranks as a nomen oblitum.
In 1932, a new subspecies of slender loris, the highland
slender loris Loris tardigradus grandis, was described by Hill
and Phillips (1932) from Gammaduwa in the Knuckles Range,
and the following year Hill (1933) described another subspe-
cies from the northern dry zone, the northern Ceylon slender
loris Loris tardigradus nordicus. In 1937, Mr. A. C. Nol-
thenius caught a pair of loris on his estate below the Horton
Plains (1522−1826 m), and kept them in captivity for several
years in Colombo (Nicholls 1939). They were described as a
further new subspecies by Hill (1942), the Horton Plains slen-
der loris Loris tardigradus nycticeboides. Two further sub-
species had meanwhile been described from southern India,
the Mysore slender loris Loris tardigradus lydekkerianus
Cabrera, 1908, from Madras, and the Malabar slender loris
Loris tardigradus malabaricus Wroughton, 1917, from Kutta,
South Coorg (more fully described in Hill 1953).
The classication of slender lorises has been debated for
many decades (Hill 1953; Phillips 1980; Groves 1998, 2001;
Brandon-Jones et al. 2003; Roos 2003; Nekaris and Jayawar-
dene 2004; Nekaris et al. 2006), yet we are still essentially
using Hill (1953) for the taxonomy, with a slight revision by
Groves (1998). According to Hill (1953), there is a single
species of slender loris, with six subspecies: four in Sri Lanka
and two in India. On the basis of museum specimens, Groves
(1998) recognized two distinct species: Loris tardigradus
(Linnaeus, 1758), monotypic and restricted to the wet zone
of southwestern Sri Lanka, and Loris lydekkerianus Cabrera,
1908, in the rest of the range of the genus in both Sri Lanka and
India. These changes were corroborated by Coultas (2002)
and Nekaris and Jayawardene (2004) on the basis of behav-
ioral and morphological evidence from wild populations, and
were further supported by phylogenetic analyses and stud-
ies of museum specimens by Roos (2003) and Nekaris et al.
(2006). Groves (1998) placed the taxon nycticeboides in L.
lydekkerianus as a subspecies, but Nekaris and Jayawardene
(2004) transferred it to L. tardigradus, while Yapa and Ratna-
vira (2013) suggested that it might be a distinct species.
Valid taxonomy is essential for species conservation.
Biodiversity assessment—the taxonomy, biogeography and
conservation status of a region’s fauna and ora—is vital for
investment in and implementation of conservation measures
(Mace 2004). Conservation management demands reliable
data to verify the distribution boundaries of taxa for their
identication as spatial conservation units. To fulll these
concerns, we reviewed the taxonomy of the Sri Lankan slen-
der lorises, following a meeting of experts and interested
parties in 2009 that included representatives of the following
organizations: The Open University of Sri Lanka, University
of Colombo, Ministry of Environment and Natural Resources,
Department of Wildlife Conservation, Forest Department,
National Science Foundation, National Zoological Garden,
University of Peradeniya and IUCN Sri Lanka.
In our study, we examined the following alternative
hypotheses regarding the taxonomy of the slender loris: (1)
that a two-species classication—L. tardigradus, either
monotypic or with nycticeboides as a subspecies, and L.
Slender lorises of Sri Lanka
85
Table 1. Number of samples represented in seven groups of Sri Lankan lorises (see Table 2).
Sample origin (n = sample size) Group 1 Group 2 Group 3 Group 4 Group 5 Group 6 Group 7
Live lorises captured from wild (n = 82) 10 28 7 2 12 5 18
Live lorises observed from the wild (n = 13) 2 3 2 1 1 1 3
Photograph from the live lorises (n = 9) 1 2 1 1 1 1 2
Fresh specimens (dead specimens) (n = 4) 1 - - 1 - - 2
Mounted skins with skulls (n = 20) 4 2 1 5 8
Skin only (n = 5) - - - 1 2 1 1
Skulls only (n = 29) 5 6 - - 1 13 4
Complete mounted skeleton (n = 1) 1 - - - - - -
Figure 1. Variation of circumocular patch shapes in different slender loris groups in Sri Lanka: a. Northwestern group (marquise -elliptical with pointed ends), b.
Southwestern group (ovoid/rounded), c. Uva group (ovoid/rounded), d. Highland grey (pear), e. Montane group (broad pear), and f. Northern grey group (teardrop).
Gamage et al.
86
lydekkerianus, with subspecies grandis and nordicus and per-
haps nycticeboides—adequately describes the taxonomy of
the slender loris in Sri Lanka; and (2) that a single-species
classication, with four subspecies, adequately describes the
taxonomy of the slender loris in Sri Lanka.
Methods
Examination of live animals and photographs
Ninety-ve live lorises were examined, 82 of which were
captured and measured in the wild, and 13 were closely exam-
ined in the eld, but not captured, by the rst author. A further
four lorises were received as dead specimens. Clear photo-
graphs of nine live slender lorises contributed further infor-
mation (see Table 1). These samples were from different cli-
matic regions: (1) wet zone (annual rainfall >2500 mm), (2)
intermediate zone (annual rainfall, 2500−1900 mm), and (3)
dry zone (annual rainfall <1900 mm), at different elevations
(1) low elevation (0−400 m asl), (2) mid elevation (400−1200
m asl), and (3) high elevation (>1200 m asl), in Sri Lanka.
Captures were done by hand by trained eld researchers;
two to four at a time. After taking measurements, the lorises
were released where they were captured. The research was
carried out under the Department of Wildlife Conservation
Sri Lanka permit number WL/3/2/1/9, and guided and super-
vised by the National Research Committee of the Department
of Wildlife Conservation. Measurements taken in the eld
followed the guidelines of Groves (2003). Facial and pelage
features were examined of all of the 82 live animals captured.
There are wet and dry seasons in Sri Lanka. The rst
author has been studying wild lorises since 2002, and has
never observed seasonal variation pelage. Aging is the only
factor affecting pelage variation, and only facial and pelage
features of mature animals were used for analysis. Facial
marks and terms used in the study are shown in Figures 1 and
2 and in the conservation database for lorises (Schulze et al.
2003). Nine measurements were taken from each live animal
to assess phenetic variation among samples. Measurements
were taken to the nearest 0.01 mm using digital calipers. The
measurements were: upper arm length (UAL), forearm length
(FAL), thigh length (TL), leg length (LL), knee length (KL),
maximum head length (MHL), head breadth (HB), maximum
breadth over postorbital bars (MBOP) and ear length (EL).
Standard measuring points are shown in the conservation
database for lorises (Schulze et al. 2003).
Examination of museum specimens
Forty-four specimens from natural history museum col-
lections were examined: National Museum of Sri Lanka,
Colombo (NHMC) (skin + skull = 4; skin only = 2), Univer-
sity of Colombo Zoology Museum (UOCSL) (skull only = 1;
complete mounted skeleton = 1), British Museum (Natural
History), London (BMNH) (skin + skull = 6; skin only = 2;
skull only = 3), Field Museum of Natural History Chicago
(FMNH) (skin + skull = 8; skull only = 1), Royal College of
Surgeons of London (RCSL) (skulls only = 15, but two of them
belonging to skins in the NHMC), and National Museum of
Scotland (NMS) (skin only = 1). We also examined 11 slen-
der lorises captured live that were euthanized (Department of
Wildlife Conservation Sri Lanka permit number WL/3/2/1/9),
using standard protocols under the supervision of a qualied
veterinarian and strictly following the American Society of
Primatologists’ principles for the ethical treatment of pri-
mates and the International Primatological Society’s inter-
national guidelines for the acquisition, care and breeding of
nonhuman primates. Their skins and skulls are preserved as
voucher specimens. Three road kills, one electrocuted speci-
men, and seven skulls were received from various parts of the
country during the study period. These and the 11 live speci-
mens that were euthanized will be deposited in the National
Museum of Sri Lanka, Colombo (NHMC) under the Slender
Loris Conservation Project (SLCP) collection.
Pelage characters and facial mask differences were exam-
ined from the 23 skins deposited in the various museums, and
from the three road kills, the electrocuted individual and the
11 euthanized specimens. Major fur characters, color marks
and terms used in the study are shown in Figures 1 and 2, and
in the conservation database for lorises (Schulze et al. 2003).
The sample included the type specimens for the described
subspecies Loris gracilis zeylanicus, Loris tardigradus nyc-
ticeboides, L. t. nordicus and L. t. grandis.
Eleven cranial and mandibular measurements were taken
(standard measuring points are shown in Schulze et al. 2003).
Measurements were taken to the nearest 0.01 mm with digital
calipers, as follows: greatest length of skull (GLS), length of
nasal (LON), biorbital breadth (BB), zygomatic breath (ZB),
breadth of braincase (BOB), mastoid breadth (MB), palate
length (PL), condylobasal length (CBL), condylo-canine
length (CCL), alveolar length of maxillary toothrow (ALMT),
mandible length (ML).
Analysis – facial and pelage features
Recent studies suggest that vision may play a greater
role in the lives of nocturnal primates than was originally
supposed (Bearder et al. 2006). Lorisiforms have monochro-
macy resulting from the loss of a functional SWS1 opsin (Tan
Figure 2. Some facial features used for the analysis.
Slender lorises of Sri Lanka
87
et al. 2005); therefore, white and strongly contrasting colours
and patterns are easily distinguishable, aiding these animals
in identifying potential mates and conspecics. Variation in
the facial mask is especially useful in distinguishing between
species (Nekaris and Jaffe 2007; Nekaris and Munds 2010;
Munds et al. 2013), and such contrasting patterns are seen
in the facial masks of slender lorises. Thus live animals and
skins were grouped according to facial/pelage features and
area of origin. A guide was created for future researchers to
replicate or expand upon this study (Figs. 1 and 2).
Twenty-three facial and pelage features were used for
this study, as follows: overall shape of the circumocular patch
(Fig. 1); shape of the top and bottom of the circumocular
patch (patch top is rounded, pointed distinct or pointed dif-
fuse, patch bottom is broad, pointed toward muzzle, narrow
rounded or extends toward zygomatic arch); width of the
median facial strip, white rim around the circumocular
patches (prominent, thin or absent), presence of dorsal frost-
ing, yellow pigmentation on muzzle and ears, and base colour
of ventral hair. Facial and pelage features of only mature ani-
mals were used for analysis; 106 loris specimens were used
(Table 2). Initially, cluster analysis was carried out for all
specimens, using 23 features (variables), by the complete
linkage method by Euclidean Distance, in Minitab 16. Then
the specimens were grouped by area of origin and again the
complete linkage method by Euclidean Distance, given in
Minitab 16, was used to create a dendrogram. Standardized
Canonical Discriminant Function Coefcients for each of
the variables (23 facial/pelage variables) was calculated by
using General Discriminant Analysis (GDA) as given in STA-
TISTICA 10. Variables with "0.000" values of Standardized
Canonical Discriminant Function Coefcients were omitted
from the subsequent analysis. Discriminant Function Analy-
sis (DFA) was undertaken with STATISTICA 10 to assess the
signicance of this clustering pattern. A subsequent Princi-
pal Components Analysis was carried out for the groups with
small sample size (n<13) using Minitab 16.
Analysis – external body measurements
Nine body measurements were used for each live animal,
in order to analyze the morphometric variation among sam-
ples. Males and females were combined because limited
samples were available from each sex. Most external mea-
surements given by Kar Gupta (2013) of 12 females and 22
males from an Indian site differ little between sexes (with the
exception of body weight), and initial inspection of our own
metric data did not reveal consistent differences between the
sexes. All these measurements have been widely used to ana-
lyze morphometric variation in Loris (Hill and Phillips 1932;
Hill 1942).
Two separate analyses were done to test the two different
models, as described above. For the rst model, live speci-
mens were grouped based on distribution and the identica-
tion characters given by Hill (1953) and Phillips (1980). For
the second model, they were grouped based on similarities of
facial and pelage features.
Initially, a one-way ANOVA was used to determine any
signicant differences between the groups for each measure-
ment, using Minitab16. Principal Components Analysis
(PCA) and subsequently Discriminant Function Analysis
(DFA) were undertaken to assess patterns in the data, where
evident, and whether this pattern was signicant; and further,
these groups were compared using box plots with STATIS-
TICA 10.
Analysis – skull measurements (cranial and mandibular)
Eleven cranial and mandibular measurements were taken
(see above). Only adult specimens (based on the degree of
fusion of skull sutures, especially the basilar suture) were
used in the analyses. All these measurements have been
widely used to analyze cranial and mandibular variation in
Loris (Hill and Phillips 1932; Hill 1942; Groves 1998).
Two separate analyses were again done to test the two
different models. For the rst model, skulls were grouped
based on names given by specimen collectors (when avail-
able) or area of origin; for the second model, skulls were
grouped based on similarities of facial/pelage features (where
an associated skin was available), and skulls without skins
were grouped based on area of origin.
Again, a one-way ANOVA was used to determine any
signicant differences between the groups for each skull
measurement and subsequent Principal Components Analysis
(PCA) were done using Minitab16. Discriminant Function
Analysis (DFA) using STATISTICA 10 was undertaken to
assess patterns in the data, where evident, and whether this
pattern was signicant; and further, these groups were com-
pared using box plots using STATISTICA 10. The sizes of the
groups in DFA must be greater than the number of variables
on which the analysis is based, to avoid the almost certain risk
of a spurious positive separation (Mitteroecker and Bookstein
2011). Accordingly, we reduced the skull measurements for
the DFA to 7, and ran a series of analyses on the basis of those
groups of size >7 (namely, groups 1, 2, 5, 6 and 7), enter-
ing the two smaller groups as ungrouped. To avoid having
to visualise plots of more than two dimensions, we ran three
different analyses, entering the ve basic groups in different
batches of three.
Geographical distribution
In all, 154 sites were surveyed in the wet and intermedi-
ate zones and part of the dry zone in Sri Lanka, using the
occupancy monitoring techniques of Mackenzie et al. (2003).
Another 38 sites were surveyed opportunistically for lorises
using a broad reconnaissance survey technique, which had
been employed for a previous island-wide study of slender
lorises by Nekaris and Jayawardene (2004); in this method
observers followed pre-existing trails and did both repeat and
one-off transect surveys (White and Edwards 2000; Nekaris
and Jayawardene 2004). GPS locations were recorded for all
loris observations, and ArcGIS® and ArcMap™ version 10
was used to create the map.
Gamage et al.
88
Figure 3. PC1 vs. PC2 factor scores graph for all (L. tardigradus = 34, L. lydekkerianus = 24) phenotypes captured (adult) during the study (male and female lumped)
for nine measured variables (UAL, FAL, TL, LL, KL, MHL, HB, MBOP and EL); PC1 accounts for 62.6 % of the variance, PC2 for 15.1%, and PC3 (not shown) for
6.5%. PC 1 is largely a size factor mainly dependent on KL, LL and FAL, which are negatively weighted. PC 2 mainly dependent on EL, which is positively weighted:
group 1 – Loris lydekkerianus group and 2 Loris tardigradus group. Specimens were separated based on the area of origin (group 1 – Wet zone lorises and group
2 – dry and intermediate zone lorises). Separation between the two groups is nearly, but not fully, complete.
Figure 4. PC1 vs. PC2 factor scores graph (47 loris skulls) for 11 measured variables (GSL, LON, BB, ZB, BOB, MB, PL, CBL, CCL, ALMT and ML); PC1 accounts
for 68.0% of the variance, PC2 for 7.6%, and PC3 (not shown) for 4.9%. Skulls were separated into two groups [group 1 - Loris tardigradus (n = 23) and group
2 - Loris lydekkerianus (n = 34)] based on the name given by the specimen collector (if available) or, if not, according to the area of origin and distribution given by
Hill (1953) and Phillips (1980). Note that the separation of the two groups is incomplete.
Conservation status
We followed the 2001 IUCN Red List Categories and
Criteria version 3.1 (IUCN 2014) to evaluate the conserva-
tion status.
Results – Phenotypic Study
One hundred and sixty-two specimens (live wild-caught
lorises = 82; live lorises observed in the wild = 13; photo-
graphs of live lorises = 9; fresh dead specimens = 4; mounted
skins with skulls = 20; skin only = 5; skulls only = 28; and
one complete mounted skeleton) were examined (Table 1).
Test of the current two-species classication – external body
measurements
One-way ANOVA showed that the Loris lydekkerianus
group (n = 34) was signicantly larger than the Loris tardi-
gradus group (n = 24) for nine body variables—UAL (F =
11.2, p = 0.001); FAL (F = 14.6, p<0.000); LL (F = 8.1, p =
0.006); KL (F = 19.3, p<0.000); MHL (F = 13.4, p = 0.001);
HB (F = 7.5, p = 0.008); MBOP (F = 30.2, p<0.000)] and EL
(F = 39.5, p<0.000). The PC1 vs. PC2 graph also showed a
clear separation of these two groups (Fig. 3). The two speci-
mens of Montane slender loris (Loris tardigradus / lydekkeri-
anus nycticeboides) from Nuwara Eliya and one specimen of
Slender lorises of Sri Lanka
89
Figure 3. PC1 vs. PC2 factor scores graph for all (L. tardigradus = 34, L. lydekkerianus = 24) phenotypes captured (adult) during the study (male and female lumped)
for nine measured variables (UAL, FAL, TL, LL, KL, MHL, HB, MBOP and EL); PC1 accounts for 62.6 % of the variance, PC2 for 15.1%, and PC3 (not shown) for
6.5%. PC 1 is largely a size factor mainly dependent on KL, LL and FAL, which are negatively weighted. PC 2 mainly dependent on EL, which is positively weighted:
group 1 – Loris lydekkerianus group and 2 Loris tardigradus group. Specimens were separated based on the area of origin (group 1 – Wet zone lorises and group
2 – dry and intermediate zone lorises). Separation between the two groups is nearly, but not fully, complete.
Figure 5. PC1 vs. PC2 factor scores graph for all (n = 58) phenotypes captured (adult) during the study (male and female lumped) for nine measured variables (UAL,
FAL, TL, LL, KL, MHL, HB, MBOP and EL); PC1 accounts 61.0 % of the variance, PC2 for 16.7%, and PC3 (not shown) for 6.5%. Specimens were separated
based on the area of origin and identication given by Hill (1953) and Phillips (1980). Groups: 1 – L. t. tardigradus, 2 – L. t. nycticeboides, 3 – L. l. grandis, 4 – L.
l. nordicus and 5 –unidentied specimens. Only Group 1 is fully separated from the other groups.
Figure 6. PC1 vs. PC2 factor scores graph (47 loris skulls) for 11 measured variables (GSL, LON, BB, ZB, BOB, MB, PL, CBL, CCL, ALMT and ML); PC1 accounts
for 67.6% of the variance, PC2 for 7.0%, and PC3 (not shown) for 5.8%. Skulls were separated into groups based on the name given by the specimen collector (if
available) or if not according to the area of origin and distribution given by Hill (1953) and Philips (1980). Groups: 1 – L. t. tardigradus, 2 – L. t. nycticeboides, 3 – L.
l. grandis, 4 – L. l. nordicus, and 5 – unidentied specimens. The groups are mostly incompletely separated from each other.
Highland grey loris (Loris lydekkerianus group) from Knuck-
les were clustered between the two groups.
Test of the current two-species classication skull
measurements
The Loris lydekkerianus group was signicantly larger
than the Loris tardigradus group for 10 skull variables—MHL
(F = 7.6, p = 0.008), LON (F = 8.5, p = 0.007), BB (F = 16.2,
p<0.000), ZB (F = 30.4, p<0.000), BOB (F = 20.1, p<0.000),
MB (F = 25.9, p <0.000), PL (F = 8.9, p =0.005), CBL (F =
9.2, p = 0.004), CCL (F = 13.3, p = 0.001), and ALMT (F =
8.5, p = 0.006). The PC1 vs. PC2 graph showed separation
of these two groups, but again the two specimens of montane
slender loris (nycticeboides) were clustered with the Loris
lydekkerianus group, while one male specimen from Mir-
igama fell between the two groups (Fig. 4).
Test of the current four-subspecies classication external
body measurements
One-way ANOVA showed that L. l. nordicus was sig-
nicantly larger than the other three described subspecies
(L. t. tardigradus, L. t. nycticeboides and L. l. grandis) and
unidentied specimens for nine body variables—UAL (F =
16.52, p<0.001); FAL (F = 18.14, p<0.001); TL (F = 5.35, p
= 0.023); LL (F = 7.78, p = 0.007); KL (F = 16.33, p<0.001);
MHL (F = 4.59, p = 0.037); HB (F = 4.48, p =0 .038); MBOP
(F = 19.57, p<0.001), and EL (F = 33.9, p<0.001). The other
groups did not show any signicant differences (p>0.05) from
each other. The PC1 vs. PC2 graph showed a clear separation
of L. t. tardigradus and L. l. nordicus from other groups; how-
ever, a single female specimen of L. l. grandis from Knuckles
was clustered with the L. l. nordicus group (Fig. 5).
Gamage et al.
90
Figure 7. Dendrogram of complete linkage and Euclidean distance for the 23 features (facial and pelage) examined in 97 slender loris specimens, and photographs
of nine live slender lorises were clearly seperated into two clusters: the wet zone cluster (Loris tardigradus) and dry zone cluster (Loris lydekkerianus). Subgroups
can be designated as: 1 – Northwestern (Gampaha and Kurunegala); 2 – Southwestern (Colombo, Kalutara, Rathnapura, Kegalla, Galle, and Matara), 3 – Rakwana
(Deniyaya-Rakwana mountain range), 4 – Montane (Nuwara Eliya and Badulla), 5 – Highland grey (Matale, Kandy and Kurunegala), 6 – Uva (Badulla, Ratnapura,
Monaragala and Ampara) and 7 – Northern grey (Anuradhapura, Polonnaruwa, Puttlum, Mannar, Vauniya and Trincomale). Separation between the groups is
complete, although groups 3 and 4 consist of a limited number of specimens (group 3 = 6, and group 4 = 7).
Figure 8: Discriminant Function Analysis, based on 13 facial and pelage features of 106 slender loris specimens (Function 1 vs Function 2 graph); Function 1 accounts
48.2% of the variance and Function 2 for 33.5% of the variance: Group 1 - Northwestern, Group 2 - Southwestern, Group 5 - Highland gray, Group 7 - Northern grey,
and ungrouped [combined group 3 (Rakwana) + group 4 (Montane) + group 6 (Uva)]. Groups 1 and 2 are completely separated from each other and from groups 5 and
7, which however overlap slightly with each other.
Test of the current four-subspecies classication – skull
measurements
One-way ANOVA showed that L. l. nordicus was signi-
cantly larger than the three other lorises (L. t. tardigradus, L. t.
nycticeboides and L. l. grandis) for 10 skull variables—MHL
(F = 23.35, p<0.001), LON (F = 18.7, p = 0.003), BB (F =
17.87, p<0.001), ZB (F = 50.91, p<0.001), BOB (F = 41.78,
p<0.001), MB (F = 61.75, p<0.001), PL (F = 29.95, p<0.001),
CBL (F = 11.53, p = 0.001), CCL (F=21.12, p<0.001) and
ALMT (F = 14.55, p<0.001). No signicant differences
(p>0.05) were observed between the other groups. The PC1
vs. PC2 graph did not show any clear separation (Fig. 6).
Proposed classication based on facial and pelage features
We observed several differences in the facial and pelage
features (Fig. 1) of live individuals encountered in the eld.
A cluster analysis using the complete linkage method divided
the 106 loris specimens into seven clusters, which are sorted
Slender lorises of Sri Lanka
91
Figure 9: PC1 vs. PC2 factor scores graph for adult wet zone phenotypes (n=34) examined during the study (male and female lumped) for nine measured variables
(UAL, FAL, TL, LL, KL, MHL, HB, MBOP and EL); PC1 accounts 56.2 % of the variance, PC2 for 18.6%, and PC3 (not shown) for 7.6%. Groups: 1 - Northwestern,
2 - Southwestern, 3 - Rakwana, and 4 - Montane. The four groups are completely separated from each other, although note that sample sizes of groups 3 and 4 are
small.
Figure 10. PC1 vs. PC2 factor scores graph for adult dry and intermediate zone phenotypes (n=24) examined during the study (male and female lumped) for nine
measured variables (UAL, FAL, TL, LL, KL, MHL, HB, MBOP and EL); PC1 accounts 60.4 % of the variance, PC2 for 13.6%, and PC3 (not shown) for 9.0%.
Groups: 5 - Highland grey, 6 - Uva, and 7 - Northern grey. The three groups are completely separated from each other.
largely, but not 100%, according to areas of distribution (Fig.
7), as follows:
Group 1: Northwestern (Gampaha, Kurunegala)
Group 2: Southwestern (Colombo, Kalutara, Ratnapura,
Kegalla, Galle, Matara)
Group 3: Rakwana (Deniyaya-Rakwana range)
Group 4: Montane (upper montane region of Nuwara
Eliya, Badulla)
Group 5: Highland grey (Matale, Kandy, Kurunegala)
Group 6: Uva (Badulla, Ratnapura, Monaragala, Ampara)
Group 7: Northern grey (Anuradhapura, Polonnaruwa,
Puttlum, Mannar, Vauniya, Trincomale)
Frequencies of facial and pelage features in the seven
groups are given in Table 2. Thus, the classication proposed
is based on these facial pelage features and distribution. The
members of group 1 (n = 14) clustered with 61% similarity, of
group 2 (n = 29) with 61% similarity, group 3 (n = 8) at 80%,
group 4 (n = 6) at 80%, group 5 (n = 16) at 61%, group 6 (n =
8) at 69% and group 7 (n = 22) at 66% similarity. All of the
wet zone lorises (groups 1, 2, 3 and 4) (n = 57) were clustered
at 36% similarity level, and the dry zone and intermediate
zone lorises (groups 5, 6 and 7) (n = 49) at 44% similarity.
One individual from Dambulla was clustered with group 6
rather than group 7 where it theoretically should belong (58%
Gamage et al.
92
Slender lorises of Sri Lanka
93
Figure 11. Dorsal views of skulls from the different loris groups: a. Northwestern [group 1 - ♂ Henarathgoda-Gampaha (30 m asl)], b. Southwestern [group 2 -
♂ Kottawa-Galle (100 m asl)], c. Rakwana [group 3 - ♂ Morningside-Singharaja (1100 m asl)], d. Montane [group 4 - ♀ below Horton Plains, holotype of L. t.
nycticeboides (1520–1824 m asl)]; e. Highland grey [group 5 - [Gammaduwa-Knuckles holotype of L. t. grandis (850 m asl)], f. Highland grey [group 5 -
Redbana-Knuckles (650 m asl)]; g. Uva [group 6 - ♂ Nilgala-Monaragala (200 m asl)], and h. Northern grey [group 7 - ♀ Talawa-Anuradhapura (100 m asl)].
Figure 12. Discriminant Analysis, based on seven skull variables (GLS, LON, BB, MB, CBL, ALMT and ML): group 1 - Northwestern, group 2 - Southwestern and
ungrouped [group 3 (Rakwana) + group 4 (Montane)]: a = Montane ♀, b = Montane ♂, c = Rakwana ♀ and d = Rakwana ♂. Function 1 accounts for 77.8% of the
variance and mainly contrasts biorbital breadth (BB), and condylobasal length (CBL). Function 2 accounts for 22.2% of the variance and contrasts biorbital breadth
(BB), mastoid breadth (MB) and mandible length (ML). Groups 1 and 2 are completely separated from each other.
Gamage et al.
94
Figure 13. Discriminant Analysis, based on seven skull variables (BB, ZB, BOB, MB, CCL, ALMT and ML): Groups: 5 - Highland grey, 6 - Uva and 7 - Northern
grey. Function 1 accounts for 82.8% of the variance and mainly contrasts biorbital breadth (BB), zygomatic breath (ZB), alveolar length of maxillary toothrow
(ALMT, breadth of braincase (BOB), condylo-canine length (CCL) and greatest length of skull (GSL); Function 2 accounts for 17.2% of the variance and contrasts
mastoid breadth (MB), biorbital breadth (BB) and mandible length (ML). The three groups are completely separated from each other.
similarity), and two individuals from Redbana-Knuckles
were clustered with group 6 rather than group 5 at 58% and
66% similarity respectively.
General Discriminant Analysis (GDA) showed that only
17 of 23 variables were supported to create a function, and
these variables were used for subsequent Discriminant Anal-
ysis. Thirteen variables were selected for the Discriminant
Analysis. Only four groups [northwestern (group 1), south-
western (group 2), highland grey (group 5) and northern grey
(group 7)] had large enough sample sizes to form the basis
for the analyses; thus the other three groups [group 3 (Rak-
wana), group 4 (montane) and group 6 (Uva)] were entered
as ungrouped. The DF1 largely reects the median facial
strip width and overall shape of the circumocular patch; DF2
reects the median facial strip width, dorsal frosting, white
rim around circumocular patch, and overall shape of the cir-
cumocular patch. Figure 8 shows DFA plots for 13 facial and
pelage features. Separations were complete, except that one
specimen of group 7 from Dambulla (Matale) was classed
with group 5. The other specimens, entered as ungrouped,
classed separately in the basic four groups. Northwestern
(group 1), southwestern (group 2), highland grey (group 5)
and northern grey (group 7) were all signicantly separated
from each other. Thus DFA supported the cluster analysis
divisions (P<0.001). PCA results for the ungrouped speci-
mens showed a clear separation only of group 3 (Rakwana),
group 4 (montane) and group 6 (Uva) from each other.
Test of the Proposed Classication Using External Body
Measurements
We tested the possibility of the proposed classication
based on facial and pelage features, using external body mea-
surements. External body morphology of slender lorises show
minor sexual dimorphism (Kar Gupta 2013). but we did not
separate sexes in our analyses due to the limited number of
samples.
External body measurements of wet zone lorises (Loris
tardigradus)
The lorises of the Southwestern group (group 2) were
signicantly larger (p<0.05) than those of the Northwestern
group (group 1) for eight measured variables—UAL (F = 23.1,
p<0.001); FAL (F = 60.1, p<0.001); TL (F = 50.5, p<0.001);
LL (F = 33.3, p<0.001); KL (F = 38.8, p<0.001); MHL (F =
39.1, p<0.001); HB (F = 21.4, p<0.001); MBOP (F = 11.8, p
= 0.002). The PCA results for wet zone lorises showed (Fig.
9) that the three groups [Northwestern (group 1), Southwest-
ern (group 2) and Montane (group 4)] were clearly separate
from each other, while the Rakwana (group 3) was clustered
between the three. We then performed a Discriminant Analy-
sis on the basis of seven body variables. Only the Northwest-
ern (group 1) and Southwestern (group 2) groups had a large
enough sample size to form a basis for the analysis; thus the
other two groups [Rakwana (group 3) and Montane (group 4)]
were entered as ungrouped. The three groups (Northwestern,
Southwestern and ungrouped) were signicantly (p<0.001)
distinct; DF1 accounted for 70.6% of the total variance, and
DF2 accounted for 29.4%.
External body measurements of dry and intermediate zone
lorises (Loris lydekkerianus)
The animals of the Northern grey group (group 7) were
signicantly larger (p<0.05) than any other group found in
the dry and intermediate zones for seven measured vari-
ables—UAL (F = 37.6, p = 0.003); FAL (F = 40.5, p<0.001);
TL (F = 22.5, p<0.001); LL (F = 12.0, p = 0.002); KL (F =
11.4, p<0.001); HB (F = 12.3, p = 0.002); MBOP (F = 13.3, p
= 0.002). The PCA results for the dry and intermediate zone
lorises showed that the three groups [Highland grey (group
Slender lorises of Sri Lanka
95
Figure 14. Distributions of the different slender loris groups in Sri Lanka.
Gamage et al.
96
Table 3: Summary of the species and subspecies classication.
Group Species
classication
Subspecies
classication
Common name Climatic zone and
distribution
Coat color and identication No. of
individuals
examined
Group 1 Loris tardigradus L. tardigradus
parvus
Northwestern red
slender loris
Northwestern wet
zone and intermediate
zone. Gampaha and
Kurunegala districts
Dark golden brown. Marquise-
shaped, broad circumocular
patch
24
Group 2 Loris tardigradus L. tardigradus
tardigradus
Southwestern red
slender loris
Southwestern wet zone.
Colombo, Kalutara,
Ratnapura, Kegalla,
Galle and Matara
districts
Varies from yellow brown
to dark brown or grayish.
Circumocular patches are
ovoid in shape and much
narrowed at the bottom
41
Group 3 Loris tardigradus L. tardigradus ?Rakwana slender
loris
Southwestern wet zone.
High altitude (>800
m asl) of Deniyaya-
Rakwana mountain
range.
Dark grey brown or brownish
black with heavily frosted.
Circumocular patches are pear
shaped and white rim around
circumocular patch is absent.
10
Group 4 Loris tardigradus L. tardigradus
nycticeboides
Montane slender
loris
Wet zone, found 1700-
1800 m asl of Nuwara
Eliya
Dark brown or brownish black.
Long, thick, woolly coat.
Circumocular patches are pear
shaped and white rim around
circumocular is prominent.
7
Group 5 Loris
lydekkerianus
L. lydekkerianus
grandis
Highland grey
slender loris
Wet zone and
intermediate zone.
Kandy and Matale
districts
Grey or grey-brown with
heavily frosted. Circumocular
patches are broad, pear
shaped and white rim around
circumocular is prominent.
22
Group 6 Loris
lydekkerianus
L. lydekkerianus
uva
Uva red slender
loris
Southeastern dry and
Intermediate zone.
Badulla, Monaragala,
Ratnapura and Ampara
districts
Yellow brown or reddish
brown with heavily frosted.
Circumocular patch is
relatively narrow and
rounded. White rim around
circumocular is prominent.
21
Group 7 Loris
lydekkerianus
L. lydekkerianus
nordicus
Northern grey
slender loris
Northern dry zone.
Anuradhapura,
Polonnaruwa,
Kurunegala, Puttalam,
Vouniya, Trincomale and
Matale districts
Grey or grey-brown with
heavily frosted coat. Broad
median facial stripe, and the
circumocular patches are
elongated with a tear-drop
shape.
38
5), Uva (group 6) and Northern grey (group 7)] were clearly
separated from each other (Fig. 10).
Test of the Proposed Classication Using Skull Morphology
The dorsal views of skulls from the seven different
loris groups are shown in Figure 11. Slender loris skulls are
slightly sexual dimorphic (Groves 1998), but the data had to
be lumped due to the limited number of specimens.
Skull measurements of wet zone lorises (Loris tardigradus)
One-way ANOVA shows the Northwestern group (group
1) was signicantly smaller than the Southwestern group
(group 2) (p<0.05) for seven skull variables—GLS (F = 8.1,
p = 0.011); BB (F = 15.6, p<0.001); ZB (F = 6.9, p = 0.017);
PL (F = 12.4, p = 0.002); CBL (F = 13.2, p = 0.002); CCL (F =
12.3, p = 0.003); ML (F = 11.3, p = 0.004). In order to inves-
tigate these differences further, and nd whether the groups
[northwestern (group 1) and southwestern (group 2)] sepa-
rate absolutely, we ran Discriminant Analyses while entering
the two available specimens of group 3 (Rakwana) and the
two available of group 4 (montane) as ungrouped. The three
groups: northwestern (group 1), southwestern (group 2) and
ungrouped were signicantly (p<0.001) distinct from each
other (Fig. 12). DF1 largely reects condylo-canine length
and biorbital breadth; DF2 reects biorbital breadth, mastoid
breadth and mandible length. The two montane specimens
(group 4) fall close to each other, well away from any of the
wet zone specimens on which the dispersion is based; we take
this as strong evidence that this group, for which the name
nycticeboides is available, constitutes a distinct taxon. On the
other hand, the two specimens from Rakwana (group 3) fall
far from each other; one is on the edge of group 2 (southwest-
ern) and the other one is on the edge of group 1 (northwestern).
Skull measurements of dry and intermediate zone lorises
(Loris lydekkerianus)
One-way ANOVA shows the northern grey group (group
7) was signicantly larger (p<0.01) than Uva (group 6) for 11
skull variables—GLS (F = 15.5, p = 0.001); LON (F = 16.6,
Slender lorises of Sri Lanka
97
p = 0.001); BB (F = 17.0, p = 0.001); ZB (F = 66.5, p<0.001);
BOB (F = 67.6, p<0.001); MB (F = 44.6, p<0.001); PL (F =
22.1, p<0.001); CBL (F = 13.2, p = 0.002); CCL (F = 29.0,
p<0.001); ALMT (F = 14.3, p = 0.002); ML (F = 7.6, p =
0.014)], and highland grey (group 5) for six variables [namely,
GLS (F = 5.0, p = 0.042); LON (F = 5.2, p = 0.039); ZB (F =
10.7, p = 0.006); BOB (F = 12.9, p = 0.004); MB (F = 20.8,
p<0.001); PL (F = 7.0, p<0.019)]. Furthermore, the highland
grey (group 5) was larger than Uva (group 6) for ve skull
variables—BB (F = 6.1, p = 0.027); ZB (F = 4.6, p = 0.002);
BOB (F = 8.5, p = 0.011); CBL (F = 5.7, p = 0.032); CCL (F =
13.4, p = 0.003); ALMT (F = 14.3, p = 0.002). The PCA result
for dry and intermediate zone lorises showed that the three
groups [highland grey (group 5), Uva (group 6) and north-
ern grey (group 7)] were clearly separate from each other.
Discriminant Analysis on the basis of seven body variables
showed the three groups [highland grey (group 5), Uva (group
6) and northern grey (group 7)] were signicantly (p<0.001)
distinct from each other (Fig. 13). DF1 largely reects con-
dylo-canine length and biorbital breadth; DF2 reects bior-
bital breadth, mastoid breadth and mandible length.
Discussion
Initial analysis of the external body morphology conrms
the existence of two major groups in Sri Lanka, namely, wet
zone lorises and dry/intermediate zone lorises. This supports
the current two species classication of Loris tardigradus
(wet zone) and Loris lydekkerianus (dry/intermediate zone)
of Groves (1998) and Brandon-Jones et al. (2003). The posi-
tion of the taxon nycticeboides is equivocal—a subspecies of
L. lydekkerianus in Groves (1998) but of L. tardigradus in
Nekaris and Jayewardene (2004)—and it is interesting that
in the present study three specimens of the montane slender
loris (group 4), potential representatives of nycticeboides,
were clustered between the two groups. Separate analysis
within these two groups (wet zone and dry/intermediate zone)
conrmed the existence of several taxa (which we here rank
as subspecies) within both of them: in the wet zone groups,
northwestern (group 1), southwestern (group 2) and montane
(group 4), and in the dry/intermediate zone groups, the high-
land grey (group 5), the Uva (group 6), and the northern grey
(group 7).
Our analysis of the skull morphology also conrms the
existence of these two major groups, again tending to sup-
port the two species classication; but again the two avail-
able skulls of the montane slender loris (nycticeboides) were
clustered between the two groups. Separate analysis within
these two groups also conrmed the existence of several
potential taxa (presumed subspecies) within the two major
groups. Overall, skull morphology shows that the north-
western (group 1), southwestern (group 2), montane (group
4), highland grey (group 5), Uva (group 6) and northern grey
(group 7) lorises are quite distinct from each other, although
groups 7 and 5 may interbreed in the vicinity of Kurunegala
and Dambulla; while group 3 (Rakwana) is possibly not a
homogeneous group craniometrically, in that one specimen
falls within group 2, the other within group 6.
Both the cluster analysis based on facial/pelage features
and the Discriminant analysis of skull measurements concur
in showing that Sri Lankan lorises can be categorized into six
groups: 1-northwestern; 2-southwestern; 4-montane; 5-high-
land grey; 6-Uva; and 7-northern grey. Further, our results
of the facial and the pelage analysis of Sri Lankan slender
lorises shows some consistency with the previous studies of
slow lorises by Nekaris and Jaffe (2007), Nekaris and Munds
(2010), and Munds et al. (2013).
Analysis of mitochondrial DNA has started, and will be
reported in a future publication when further data have been
collected. Generally, the members of the northern grey group
(group 7) are large and distinctive in both body and skull pro-
portions and facial features. This group is distributed in the
northern dry zone up to the Jaffna peninsula; its southwestern
boundary is the Deduru Oya, whereas the southeastern bound-
ary is not clear. All specimens of group 7, both museum spec-
imens and living animals, can be readily differentiated from
the other 6 groups.
Members of the northwestern group (group 1) are smaller,
and have unique facial features and distinctive body and skull
proportions when compared to other groups. Based on our
eld observations, the distribution of this group is concen-
trated around Gampaha and Kurunegala, with its southern
boundary at the Kelani Ganga River; although two specimens
(skulls with skins) labeled simply as Colombo (BMNH.1937-
7-2.7 and CONHM 7H) are identical with the northwestern
group (group 1), our eld observations are unable to detect
any northwestern (group 1) loris from the Colombo district.
The northern and eastern boundaries of the northwestern
group are still not clear. The pelage colour of the northwest-
ern (group 1) shows some similarity to that of the southwest-
ern (group 2), but they differ in size, especially FAL and TL,
and in the shape of the circumocular patch (marquise shaped,
broader at the bottom, and extended toward the muzzle in the
northwestern group); and all specimens, both museum speci-
mens and living animals, can be readily differentiated.
The southwestern group (group 2) lorises are medium-
sized, with distinctive body proportions when compared to
the northwestern (group 1), montane (group 4), highland
grey (group 5), Uva (group 6) and northern grey (group 7)
groups. Considering the skull variables, the group is well dis-
criminated from most other groups; only the separation from
the Rakwana group is not signicant, although the discrimi-
nant analysis of external body morphology does distinguish
them, and similarly in facial/pelage features the southwest-
ern (group 2) and Rakwana (group 3) lorises are signicantly
(p<0.001) distinct from each other. The coat is hardly frosted
in the southwestern (group 2), but dorsally strongly frosted in
the Rakwana group (group 3).
In the montane group (group 4), facial/pelage features
are noticeably different from all other loris groups, and the
external body proportions are strongly different from those
of the other wet zone lorises, while their very long, dense fur
Gamage et al.
98
is unique to this group and easily distinguishes it from any
other loris taxon in Sri Lanka. Furthermore, the two avail-
able skulls [the holotype of L. l. nycticeboides Hill, 1942, and
the specimen from Conical Hill, Nuwara Eliya] are notice-
ably different from all wet zone groups. The distribution of
this group is apparently conned to the upper montane region
(>1600 m asl) of the Central Highlands. Initial mitochondrial
DNA study (based on cytochrome oxidase I) results shows
the montane (group 4) is genetically close to the southwestern
(group 2) (Gamage 2015).
The Rakwana group (group 3) is a unique taxon restricted
to high altitudes (over 700 m asl) in the Rakwana-Deniyaya
mountain range. The facial/pelage features are signicantly
(p<0.001) different from all other loris groups, the notable
unique characters being the broad, pear-shaped, dark circum-
ocular patches, which are extended up to the crown, and no
white rim is visible around the eyes. The single male skull
clearly separates from the skulls of all other groups; yet the
skull of a subadult female from Gongala was clustered with
the southwestern (group 2). The Rakwana group is parapat-
ric to several other groups. Along its western, southern and
northern boundary, it meets the southwestern (group 2), and
on its eastern boundary it meets the Uva (group 6) (Fig. 14).
This group needs further study with more material.
In its external morphology and facial features the highland
grey (group 5) is clearly different from northern grey (group
7), Uva (group 6), northwestern (group 1) and the southwest-
ern (group2), although in its external body morphology the
separations from the Rakwana (group 3) and montane (group
4) groups are not quite so clear. The prominent white rim
around the circumocular patch is present in both the highland
grey (group 5) and montane (group 4) groups, but not in the
Rakwana (group 3). It is markedly different in body and skull
proportions, especially from the Uva (group 6), with which
it shares a high degree of frosting; the circumocular patch is
always pear-shaped compared to being rounded in the Uva
(group 6). The highland grey (group 5) is restricted to high
altitudes (>400 m asl) in the Knuckles Range and the Kandy
region (especially the wet part of the central highlands), while
its sister group, the Uva (group 6), is widely distributed in the
intermediate and dry zone areas at lower elevations (<500 m
asl) in the Uva basin (northern and eastern ank of the Central
Highlands up to the eastern dry zone) (Fig. 14).
Our examination of type specimens shows that the type
of nycticeboides belongs to the montane group (group 4), the
type of nordicus belongs to the northern grey group (group
7), and the highland grey group (group 5) includes the type of
grandis. The type of zeylanicus probably belongs to south-
western group (group 2).
We maintain that there is good evidence for the existence
of at least two different species, of which one must at pres-
ent be regarded as conspecic with the Indian lorises, while
the existence of at least three, probably four, further species
is probable, but more evidence is needed before they can be
fully diagnosed and dened.
Taxonomic Conclusions
Here, we distinguish provisionally two species of slender
loris in Sri Lanka with six subspecies (see Table 3). As noted
in the introduction, the names tardigradus, gracilis, ceyloni-
cus, lori and zeylanicus are based on small red lorises, and are
likely to represent one of our groups. Although Linnaeus used
depictions of specimens in Albertus Seba’s (1734) collection
for his description of what he called Lemur tardigradus, it
is hardly possible to tell which group of the red loris [north-
western (group 1), southwestern (group 2) or Uva (group 6)]
it actually belongs to; and the type locality was described
simply as Ceylon. The type specimen nominated by Gentry
et al. (1998) does not resolve this question: as illustrated in
<http://linnaeus.nrm.se/zool/mamm/images/M532011.jpg>,
it is a faded specimen preserved in alcohol and little trace of
the original colour or pattern remains.
Lesson’s (1840) Arachnocebus lori also has Seba as the
primary reference, so is an objective synonym of Linnaeus's
tardigradus. Lesson’s other name, Bradylemur tardigradus,
is a secondary homonym of tardigradus Linnaeus, and is in
any case a strange mixture of slow and slender lorises, only
his Variety C being a Sri Lankan slender loris (one of the black
ones), but it is hard to tell which to group it might pertain.
Remaining are the two species described by Fischer
(1804). His Loris gracilis is evidently just a renaming of Lin-
naeus's tardigradus, but his Loris ceylonicus is meant to be
something different. He described it as "yellowish brown",
which could correspond, like Linnaeus’s tardigradus, to
either the northwestern group (group 1), or the southwest-
ern group (group 2) or the Uva group (group 6), but it seems
impossible to say what group is actually involved, and as this
name has not been used for more than two hundred years and
no type specimen is known to be available, here we propose
arbitrarily that the name Loris ceylonicus is a synonym of
tardigradus. Hill and Phillips (1932) wrote “Hitherto, how-
ever, all the lorises from Ceylon that appear to have been at
all thoroughly examined have come from within a radius of
30 or 40 miles of Colombo”(Hill and Phillips 1932, p.109).
Two groups described here—the northwestern (group 1) and
southwestern (group 2) groups—are found in the western
region, within a radius of 30 or 40 miles of Colombo. Of
these, the southwestern group (group 2) is widely distributed
throughout the wet zone proper, and may thus be considered
more likely to represent tardigradus Linnaeus, 1758.
Our taxonomic arrangement of the slender lorises of Sri
Lanka is as follows.
Loris tardigradus tardigradus (Linnaeus, 1758)
Southwestern group (group 2)
Type: Gentry et al. (1998) designated a Swedish Museum
of Natural History, Stockholm, specimen (NRM 532011) as
Slender lorises of Sri Lanka
99
lectotype (see <http://linnaeus.nrm.se/zool/mamm/images/
M532011.jpg>).
Type locality: Ceylon.
Diagnosis: Loris tardigradus has distinctive facial features:
the circumocular patches are ovoid in shape and much nar-
rowed at the bottom and in the middle; the patches are brown,
dark brown or chestnut brown. The median facial strip is nar-
rower than in any other Sri Lankan loris group; with strong
yellow pigmentation on the muzzle, hands, feet, ears and eye-
lids in both sexes. Preocular hair varies from brown to dusky
white to silvery grey. Coat colour on the body varies from
yellowish brown to dark brown or grey brown. The fur is
between wavy and curly; and longer than in Loris tardigradus
parvus n. ssp. (see below). The pelage of the ventral surface
is generally yellowish, creamy or dusky whitish. Throat hair
is generally supercially yellow/cream, with bases yellow-
ish, light grey, grey or black. A dark dorsal stripe is visible.
Skull size is similar to the montane (4), highland grey (5) and
Uva (6) groups, and much larger than in the northwestern (1)
group. Ridges on the skull are moderately developed, includ-
ing the temporal ridges and the curved ridge on the occiput.
The dorsal surface of the skull is not smooth, with a mod-
erately developed wing-shaped mastoid; skull length ranges
from 48.7 mm to 51.1 mm. The body is much larger and more
elongated than in the northwestern (1) and Uva (6) groups.
Description: A few specimens from Kalutara have broadly
pear-shaped circumocular patches. Animals found in Galle
and Matara have a much more brightly coloured belly and
the hair bases are dark grey. Those found in Ratnapura, Hini-
duma, Akuressa and Deniyaya have a dusky white belly and
the hair base colour is black.
Distribution: The distribution is shown in Figure 14. This
species is found throughout the wet part of the southwestern
region, from Colombo, Kalutara, Ratnapura, Kegalla, Galle,
Matara and possibly the wetter part of Hambantota.
Common name: Southwestern red slender loris
The following specimens have been examined:
Skin plus skull: NHMC 7I [adult ♂ Colombo]; FMNH 95027
[adult Maharagama], SLCP 2015.09 [adult ♀ Madakada-
Kaluthara], SLCP 2015.15 [adult ♂ Kottawa-Galle], SLCP
2012.02 [adult ♀ Kudawa-Sinharaja].
Skull only: SLCP 2012.03 [adult sex? Oliyagan-Matara],
SLCP 2012.11 [adult sex? Maharagama], SLCP 2012.07
[adult ♂ Kottawa-Galle], SLCP 2012.01 [adult sex? Mass-
mulla]; RCSL. A112.53 [adult Polgahawela], UOCSL
99.7.31 [adult sex? Kaduwela-Colombo].
Living animals: Madakada-Kalutara [adult ♀ = 1; juve-
niles = 2], Beraliya-Kalutara [adult ♂ = 1], Yagirala-Kalutara
[juvenile ♀ = 1], Kottawa-Galle [adult ♂ = 3, juveniles =
1], Hiyara-Galle [adult = 1], Massmulla-Matara [adult ♂
= 1, juveniles = 2], Rammale Kanda-Matara [adult ♂ = 1],
Diyadawa-Matara [adult ♂ = 2], Velihena-Matara [adult ♀ =
2], Kekanadura-Matara [adult ♀ = 1], Oliyagan-Matara [adult
♂ = 2, ♀ = 2, juvenile = 1], Kondagulankanda-Matara [adult ♀
= 1], Kalubovitiyana-Matara [juvenile = 1], Kanneliya-Galle
[sub-adult = 1], Kudawa-Sinharaja [adult = 1, = 1],
Madampa-Rathnapura [adult ♂ = 1, ♀ = 1], Delwala-Rathna-
pura [adult ♀ = 1], Gilimalee-Rathnapura [adult ♂ = 1], Peak
Wilderness-Rathnapura [adult ♀ = 1].
Conservation status: Our study conrmed that this subspe-
cies is found only in primary and secondary forest in the
southwest wet zone. All of 48 sites where it is known to occur
are fragmented. The Extent of Occurrence (EOO) is 4,800
km2 and the Area of Occupancy (AOO) is 751 km2. Further-
more, the estimated population is <2,500 mature individuals.
Thus, this subspecies is assessed as Endangered [B1b(i, ii and
iii), C1].
Loris tardigradus parvus new subspecies
Northwestern group (group 1)
This group has unique facial features, body morphology and
skull morphology. Further research on this new subspecies
is needed. It seems sharply distinct from the southwestern
Loris t. tardigradus, and we predict that further work will
show that it is a distinct species, but the present sample size
is insufcient.
Type: Adult female skin, skull and tissues (in alcohol), SLCP
2015.13, to be deposited in the Natural History Museum of
Colombo (NHMC). Collected by the Slender Loris Conserva-
tion Project (SLCP), 11 February 2010.
Type locality: Mirigama, Gampaha District, Western Prov-
ince, Sri Lanka (07°15.813'N, 80°08.405'E).
Diagnosis: This subspecies has a unique marquise-shaped,
broad circumocular patch, which is much broader at the
bottom compared to Loris tardigradus tardigradus, and is
extended toward the muzzle. The median facial strip is rela-
tively broad. There is yellow pigmentation on the muzzle,
hands, feet, ears, and eyelids, rather slight in males and more
marked in females. Pre-ocular hair is light brown in males,
creamy or silvery in females. Dark, golden-brown pelage
with short fur. Males have yellowish hair on the ventrum,
females also have yellowish hair but a little darker; throat hair
is supercially creamy, the base colour is yellow; the hind
limbs are more strongly coloured, both ventrally and dorsally
(see Fig. 15). This is the smallest form of loris; body weight
is 123−170 g. Head and body length (<200.5 mm), upper
arm length (<55.5 mm) and thigh length (66.6 mm) are much
shorter than in any other loris. Temporal ridges are only
moderately developed, and the dorsal surface of the skull is
smooth and rounded, with a moderately developed mastoid.
Distribution: The distribution is shown in Figure 14. Live
animals were observed in Gampaha and Kurunegala dis-
tricts. Three museum specimens (BMNH 1937.7.2.7, BMNH
10.5.19, and NHMC 7H) labeled “Colombo” (listed below)
were classied in this group.
Common name: Northwestern red slender loris
The following specimens have been examined:
Skin plus skull: BMNH 1937.7.2.7 [adult ♀ Colombo],
NHMC 7B [adult ♂ Henarthgoda-Gampaha]; 7d [adult
Gamage et al.
100
Henarthgoda-Gampaha], 7H [adult Colombo]; SLCP
2015.06 [adult ♀ Mirigama-Gampaha], 2015.17 [adult
Mirigama-Gampaha].
Skull only: BMNH 10.5.19 [adult sex? Colombo]; FMNH
92861 [adult ♀ Gampaha]; RCSL OH/69 [adult ♀ Ceylon];
SLCP 2015.00 [adult ♂ Mirigama-Gampaha], 2015.20 [adult
♂ Dunagaha-Gampaha].
Living animals: Mirigama [adult ♂ = 2, ♀ = 1, juveniles =
2]; Pilikuththuwa- Gampaha [adult = 1]; Horagolla-Gam-
paha [adult ♀ = 1, juvenile ♀ = 1].
Etymology: The species epithet is in reference to its small
size. The word parvus is Latin for little.
Conservation status: Inhabits the Northwestern part of the
wet zone region. Unfortunately, nearly 96% of the natural
vegetation has already been destroyed in this region. Available
natural vegetation is <9km2, which is severely fragmented,
and deforestation continues apace (Jayasuriya et al. 2006).
The Extent of Occurrence is 140 km2 and Area of Occupancy
is 9 km2. Further, the Sri Lanka 2011−2030 National Physi-
cal Plan has many of its remaining habitats included in the
Western Metro Region (Sri Lanka, Ministry of Construction
2012). The estimated population is <250 mature individuals
and no subpopulation is estimated to contain more than 50
mature individuals. Thus, this subspecies is assessed as Criti-
cally Endangered [B2a,b(i, ii & iii), C2a(i)].
Loris tardigradus nycticeboides Hill, 1942
Montane group (group 4)
Facial and pelage features, external body morphology and
skull morphology are unique to this loris, genetic data (based
only on the CO1 region) place it very close to L. tardigradus
(see Gamage 2015). Yapa and Ratnavira (2013) have argued
that it could represent a distinct species, and we have much
sympathy with this view, but further research is needed.
Type: Adult female, skin and skull, British Museum (Natural
History) BMNH 45.3.Plains. Collected by Mr. A. C. Nolthe-
nius in 1937.
Paratype: Adult male, skin only, National Museum of Scot-
land NMS 1946.4.
Type locality: Below Horton plains, Sri Lanka.
Diagnosis: A long, thick, woolly coat is a diagnostic character
for this species; the colour is brown, dark brown or brownish
black; frosting is observed on wild animals. The circumocular
patch is pear shaped, black or brownish black in colour, and
wider above in both sexes; a whitish rim is visible all around,
in both sexes. The median facial strip is wide; yellow pig-
mentation on the muzzle, hands, feet, ears and eyelids is very
reduced; the hands, feet and muzzle are esh colour. Muscu-
lar ridges are moderately developed, including the temporal
ridges and the curved ridge on the occipital; dorsal surface
of the skull is relatively curved; mastoid is poorly developed.
Description: Body weight ranges from 140−220 g, with head-
body length 202−220 mm; maximum head length 52.0−57.0
mm.
Distribution: The distribution is shown in Figure 14. Accord-
ing to eld observations, this species is found at elevations
of 1700−2100 m asl around Nuwara Eliya. The slender loris
found at a similar elevation in the Peak Wilderness is much
closer to Loris tardigradus tardigradus (group 2).
Common name: Montane slender loris, Horton Plains slen-
der loris
Notes: The rst author (SNG) has been conducting a long-
term study (10 years) on the ecology of the montane slender
loris since 2006, but has yet to hear any vocalizations.
The following specimens have been examined:
Skin plus skull: BMNH 45.3 [adult ♀ below Horton Plains,
holotype of nycticeboides Hill, 1942]. SLCP 2015.18 [adult
♂ Conical Hill-Nuwara Eliya]
Skin only: NMS 1946.4 [adult ♂ below Horton Plains, para-
type of nycticeboides Hill, 1942].
Live animals: Conical-Nuwara Eliya [adult = 1, = 1],
Kikiliamana Nuwara Eliya [adult = 1, ♀ = 1], Horton Plains
National Park [♀ = 1], Hakgala-Nuwara Eliya [2].
Conservation status: A recent occupancy modeling study
identied L. t. nycticeboides as closely associated with eleva-
tion, canopy height and canopy connectivity; the best habitat
is montane evergreen forest at elevations of 1600 m to 2100
m asl, with a tall canopy (height >4 m) and good canopy con-
nectivity (Gamage et al. 2015). Forest dieback, uncontrolled
Figure 15. Northwestern red slender loris, Loris tardigradus parvus, female
holotype from Mirigama, Gampaha District, Sri Lanka (07°15.813'N,
80°08.405'E).
Slender lorises of Sri Lanka
101
rewood extraction, and encroachment are causing a continu-
ous reduction of its habitat. The Extent of Occurrence is 76
km2, and the Area of Occupancy is 28 km2. The estimated
population is <100 mature individuals and no subpopulation
is estimated to contain more than 50 mature individuals. Thus,
this subspecies is assessed as Critically Endangered [B1b(i, ii
and iii), C2a(i)].
Loris lydekkerianus nordicus Hill, 1933
Northern grey group (group 7)
Here, we provisionally retain the northern grey group (7) as
a subspecies of Loris lydekkerianus, described from the dry
country of south-eastern India. We lack morphological and
genetic data from Indian slender lorises, and direct compari-
sons are necessary to test this allocation.
Type: Sub-adult female skin and skull, British Museum (Nat-
ural History) BMNH 35.4.1.1.
Type locality: Thalawa, Anuradhapura, Sri Lanka.
Diagnosis: Loris lydekkerianus nordicus has a distinctive
broad median facial stripe, and the circumocular patches
are elongated with a tear-drop shape; the patches are grey or
grey brown, and a white rim is visible around them. Grey
or grey-brown coat with short, thin wavy fur. The hairs of
the ventral surface are white, with hair bases also white or
light grey, never becoming dark. The throat hair is white, and
the bases are also white. Yellow pigmentation on hands, feet,
ears and eyelids present in both sexes. The ears are large,
length 23−29 mm. Size averages larger than any other taxon.
All muscular ridges of the skull are well developed, especially
the temporal ridges and the superior nuchal line (curved ridge
on the occiput), which forms a small raised crest or ridge.
The occiput is not rounded (nearly at in dorsal view) and
the dorsal surface of the skull is at with a prominent wing-
shaped mastoid process.
Description: Animals in the northwestern region of the range
show less yellow pigmentation than elsewhere; especially
the muzzle is hardly pigmented. Females are more heavily
frosted than males. Well-built body, weight 230−293 g, head
body length 214−240 mm, and head length 52.0−58.6 mm .
Skull length ranges from 49.0−54.8 mm.
Distribution: The distribution is shown in Figure 14. This
species is found throughout the northern dry zone; the south
western boundary is Deduru Oya (river) and the southeastern
boundary is Maduru Oya National Park. This loris has never
been observed in the southeastern dry zone or wet zone areas,
although its range appears to overlap with the Highland grey
group (5) in Matale and Kurunegala.
Common name: Northern grey slender loris
The following specimens have been examined:
Skin plus skull: BMNH 35.4.1.1 [sub-adult Thalawa-North
Central Province, holotype of nordicus Hill, 1933], 1966.3916
[adult ♂ Wilachchia-Anuradhapura], 15.3.1.16 [adult ♀Anu-
radhapura)]; FMNH 95028 [adult ? Chawakacheri-Jaffna],
95029 [adult ♂ Chawakacheri-Jaffna], and 95030 [adult ?
Chawakacheri-Jaffna]; SLCP 2015.16 [adult ♀ near Thalawa-
Anuradhapura]; University of Rjarata Sri Lanka, Zoology
Museum 2008.04 [adult ♀ Mihintale-Anuradhapura].
Skin only: BMNH 15.3.1.14 [adult ♂Anuradhapura].
Skull only: BMNH 1966.3916 [adult ♂Wilachchia-Anurad-
hapura], 15.3.1.15 [♂ Anuradhapura]; SLCP 2015.05 [adult
♂ Kebethigollawa-Anuradhapura], SLCP 2015.10 [adult ?
Dambulla].
Living animals: Thalawa-Anuradapura [adult ♀ = 1], Mihin-
talee-Anuradhapura [adult ♀ = 3, adult ♂ = 1, Juveniles =
2], Kebethigollawa-Anuradhapura [♂ adult = 1, sub-adult =
1], Anawilundawa-Puttlum [adult = 3, = 2], Dambulla-
Matale [adult ♀ = 1], Wilpattu National Park [adult ♀ = 1],
Galenbindunuwewa-Anuradapura [adult = 1], Mannar [♂
adult = 1], Polonnaruwa [adult ♂ = 1, ♀ = 2] and Trincomale
[adult ♂ = 1].
Conservation status: This subspecies inhabits the northern
dry zone of Sri Lanka. In contrast to the wet zone lorises,
this subspecies has a considerable amount of forest left in its
range; but these forests are highly fragmented and continually
being lost because of deforestation for large-scale agriculture
and infrastructure development (Jayasuriya et al. 2006). The
Extent of Occurrence is 17,400 km2, and the Area of Occu-
pancy is 3000 km2. The estimated population is <8500 mature
individuals but no subpopulation is estimated to contain
more than 1000 mature individuals. Thus, this subspecies is
assessed as Vulnerable [B1ab(i, ii and iii), C1,2a(i)].
Loris lydekkerianus grandis Hill and Phillips, 1932
Highland grey group (group 5)
Facial and pelage features, external body morphology and
skull morphology is unique and well differentiated from Loris
lydekkerianus nordicus. This form may turn out to be a dis-
tinct species.
Type: Sub-adult female skin and skull, British Museum (Nat-
ural History) BMNH 32.6.17.1.
Type locality: Gammaduwa, Knuckles, Sri Lanka.
Diagnosis: This taxon has a distinct, much darker, pear-
shaped circumocular patch, and the rim around the patch is
white and prominent. The median facial strip is relatively
wide. The face and hind limbs are more whitish than Loris
tardigradus tardigradus (group 2). General character is a
grey or grey-brown with thick, relatively long fur; females
are more heavily frosted than males, as in the Rakwana group
(3). The coat contains a mixture of wavy and woolly hair. The
ventrum is supercially white or dusky white in females, and
slightly more yellowish brown in males, whereas hair bases
are black in both sexes. The throat hair is supercially white
and the hair bases are dark grey or black. Both sexes have
yellow pigmentation on the muzzle, hands, feet, ears and
eyelids. The ears are relatively short and the preocular hair
is grey. The body is relatively stout, with muscular ridges
moderately developed, including the temporal ridges and the
curved ridge on the occiput. The occiput is less rounded in
dorsal view and the dorsal surface of the skull is relatively
Gamage et al.
102
curved with moderate development of the mastoid; skull
length ranges 49.1−51.1 mm.
Description: Those found at higher altitudes are stouter and
less frosted than those at lower altitudes. Those from high
elevations (>900 m asl) in Knuckles have a very low degree
of frosting and relatively stout body and short limbs com-
pared to those found at lower elevations. Some individuals
from Kurunegala and Dambulla show intermediate characters
between this species and L. l. nordicus. Body weight ranges
from 176−216 g, head-body length 208−220 mm, and maxi-
mum head length 53.9−55.3 mm.
Distribution: The distribution is shown in Figure 14. This
species is found in the Knuckles Range and wet and interme-
diate parts of Kandy and Matale districts (200−1300 m asl).
If this taxon really does overlap with L. l .nordicus in Matale
and Kurunegala, as noted above under the heading of that
taxon, then of course the two cannot be regarded as conspe-
cic under any species concept; but this needs to be veried.
Common name: Highland grey slender loris
Notes: The skin from Opalgala, NHMC 7xA, is more brown-
ish-toned than other specimens seen, and so bears a super-
cial resemblance to Loris tardigradus nycticeboides.
The following specimens have been seen:
Skin plus skull: BMNH 32.6.17.1 [sub-adult ♀ Gammad-
uwa-Knuckles, type of grandis Hill and Phillips, 1932];
FMNH 99479 [adult ♂ Digana-Kandy], 95204 [adult Pin-
deniya-Central Province], 95025 [adult ♂ Peradeniya-Central
Province], 95026 [adult ♂ Peradeniya -Central Province];
SLCP 2015.16 [adult Pitawala-Knuckles], SLCP 2015.19
[adult ♀ Redbana-Knuckles].
Skull only: RCSL L2/OH 68 [Adult ♂ Opalgalla-Knuckles].
Skin only: NHMC 7xB [adult ♂ Mausakanda, Gammaduwa-
Knuckles], 7xA [adult Opalgalla Gammaduwa-Knuckles;
skin of RCSL L2/OH 68].
Living animals: Mausakanda-Knuckles [juvenile ♀],
Pitawala-Knuckles [adult ♂ = 1, ♀ = 1], Mahalakotuwa-
Knuckles [Juvenile ♂ = 1], Narangamuwa-Knuckles [adult
♀ = 1], Thangappuwa-Knuckles [adult ♂ = 1], Redbana-
Knuckles [Adult ♀ = 2], Udwattakele-Kandy [adult = 1],
Elahera-Matale [adult ♀ = 1], Dagawila-Kandy [sub-adult ♀
= 1], Wattegama-Kandy [adult ♀ = 1].
Conservation status: This subspecies inhabits lowland wet
evergreen forest and mid-elevation evergreen forests in the
Kandy and Matale districts. These forests are highly frag-
mented and continuously being lost due to encroachment,
rewood extraction and expansion of plantation industry
(Jayasuriya et al. 2006). In the Knuckles range, cardamom
cultivation degrades its habitat (S. N. Gamage pers. obs.).
The Extent of Occurrence is 1,750 km2, and the Area of Occu-
pancy is 520 km2. The estimated population is <1,200 mature
individuals. Thus, this subspecies is assessed as Endangered
[B1ab(i, ii and iii), C1].
Loris lydekkerianus uva new subspecies
Uva group (group 6)
Facial/pelage features, external body morphology and skull
morphology, all based on limited material, show apparently
signicant differences between the Uva group and other Sri
Lankan lorises. The genetic data (based only on the CO1
region) place it very close to L. lydekkerianus grandis (see
Gamage 2015). The Uva group can be treated provisionally
as a subspecies of Loris lydekkerianus, but it may turn out to
be specically distinct.
Type: Adult male skin, skull and tissues (in alcohol), SLCP
2015.12, to be deposited in the Natural History Museum of
Colombo (NHMC). Collected by the Slender Loris Conser-
vation Project on 1 May 2012.
Type locality: Nilgala, Monaragala District, Uva Province,
Sri Lanka (07°10.727'N, 81°17.477'E).
Diagnosis: The circumocular patch is relatively narrow,
rounded and light brownish. The pelage is yellow brown/red-
dish brown with a little grey tone; golden frosting is present.
The ventrum is more whitish than in Loris t. tardigradus or
L. t. parvus n. ssp.; hair-base color of the belly is grey, never
becoming black. The throat hair is creamy on both the shaft
and base. White frosting is present on the hind limbs. The
median facial stripe is wider than in Loris t. tardigradus or L.
t. parvus n. ssp. There is much yellow pigmentation on the
muzzle, hands, feet, ears and eyelids (see Fig. 16). Muscular
ridges of the skull are moderately developed; the occiput is
more rounded in dorsal view; and the dorsal surface of the
skull is rounded; a moderately developed mastoid. Relatively
short body and limbs; head-body length is 208.25±7.14 mm
and upper arm length is 54.30±1.25 mm.
Distribution: The distribution is shown in Figure 14. It is
found mainly in the Uva Basin (southeastern dry and interme-
diate zones); Badulla, Monaragala and Ampara districts.
Common name: Uva red slender loris.
Figure 16. Uva red slender loris Loris lydekkerianus uva, male holotype male
from Nilgala, Monaragala District, Sri Lanka (07°10.727'N, 81°17.477'E).
Slender lorises of Sri Lanka
103
The following specimens have been examined:
Skin plus skull: SLCP 2012.12 [adult ♂ Nilgala-Monaragala].
Skin only: BMNH 54.349 [♂ Namunukula-Badulla].
Skull only: RCSL - A112.528 [adult ♀ Monaragala-Uva],
A112.523 [adult ♂ Monaragala-Uva], A112.5292 [adult
Monaragala-Uva], A112.522 [adult Monaragala-Uva],
A112.524 [adult ♂ Monaragala-Uva], A112.525 [juvenile
♂ Monaragala-Uva], A112.526 [adult ♂ Monaragala-Uva],
A112.529 [adult ♀ Monaragala-Uva], A112.5294 [adult
Monaragala-Uva], A112.5296 [adult Monaragala-Uva],
A112.5291 [adult ♀ Monaragala-Uva], A112.5298 [adult
Monaragala-Uva], SLCP 2015.21 [adult Velioya -Balan-
goda]Living animals: Lahugala-Ampara [adult ♂ = 1], Nil-
gala-Monaragala [adult = 1], Rawana Ella-Badulla [adult
♂ = 1], Veelioya-Monaragala [adult ♂ = 1], Passara-Badulla
[adult ♂ = 1], Belihuloya-Balangoda [adult ♂ = 1; adult ♀ =
1].
Etymology: The species epithet is in reference to its distri-
butional range in the Uva Basin; Uva is a historical name for
the south eastern dry and intermediate zone of Sri Lanka.
Conservation status: The Extent of Occurrence is 7,520 km2,
and the area of occupancy is 2,600 km2. This subspecies is
very rare, being most common in undisturbed moist mixed
evergreen forests. These forests are highly fragmented and
continually being lost because of deforestation for large-scale
agriculture and infrastructure development (Jayasuriya et al.,
2006). We assess the subspecies as Vulnerable [B1ab(i, ii and
iii)].
Rakwana group (group 3)
Facial and pelage features, external body morphology and
skull morphology show some differences between the Rak-
wana group (3) and Loris tardigradus but our sample size is
not enough for any meaningful analysis. Until further evi-
dence becomes available, it is impossible to consider propos-
ing this as a further subspecies at this stage. We note, how-
ever, that “a very rich, rusty-coloured loris has been seen near
Morningside near the Sinharadja Forest Reserve… a new sub-
species or species?” (Yapa and Ratnavira 2013: 151).
Description: The limited material indicates a dark grey brown
or brownish black coat with long dense fur. Both sexes show
frosting, but females are more heavily frosted. Pelage of the
ventrum is cream or dusky-white with hair bases black. The
throat hair is white and bases are black. Dorsally very dark,
hence dorsal stripe is not visible. Circumocular patches are
pear shaped with black or brownish black color; wider above
in both sexes and extended up to crown; no rim is visible
Figure 17. The Uva red slender loris Loris lydekkerianus uva, left, and the Northwestern red slender loris, Loris tardigradus parvus, right. Illustrations by Stephen
D. Nash.
Gamage et al.
104
around the circumocular patch. Both sexes have very little
yellow pigmentation on the muzzle, hands, feet, ears and eye-
lids. Pre-ocular hair is dark-brown or nearly black. Muscu-
lar ridges are poorly developed, especially temporal ridges;
the dorsal surface of the skull is smooth and rounded. The
superior nuchal lines are not detectable or poorly developed;
the occiput is rounded in dorsal view; poorly developed mas-
toid in female (sub-adult), moderately developed in male.
Distribution: Distribution is shown in Figure 14. This form
is found at high elevations (800−1300 m asl) of Rakwana-
Deniyaya mountain range. Its western, northern and southern
boundary meets group 2 (Loris tardigradus tardigradus) and
its eastern boundary meets group 6 (Loris lydekkerianus uva
new subspecies).
The following specimens have been examined:
Skin plus skull: SLCP 2015.04 [sub-adult ♀ Gongala-Rath-
napura], 2015.14 [adult ♂ Morningside-Sinharaja].
Living animals: Gongala-Rathnapura [adult ♀ = 1], Morn-
ingside-Sinharaja [adult ♂ = 1, ♀ = 1, Juveniles = 1], Ensal-
watta-Estate Sinharaja division [adult ♂ = 1, Juvenile ♀ = 1],
Handapanella-Rathnapura [adult = 1], Deniyaya [adult ♂ = 1,
sub-adult ♀ = 1].
Future studies
We have retained the division of Sri Lankan slender
lorises into two species. Further surveys and DNA samples
will help to fully clarify the afnities of the Sri Lankan loris
taxa to each other and to the Indian lorises. The latter have
recently been briey surveyed by Kumara et al. (2013). In
Sri Lanka, the afnities of L. cf. tardigradus nycticeboides
are unclear; we have provisionally placed it in L. tardigradus,
following Nekaris and Jayewardene (2004). Groves (1998,
2001) placed it in L. lydekkerianus, while Yapa and Ratna-
vira (2013) have suggested that it may be specically distinct.
The latter authors have also claimed they found “a totally
isolated subpopulation of L. lydekkerianus […] deep within
tardigradus territory, in the Deniyaya area” (Yapa 2013: 146).
We intend to follow up on some of these questions in a future
publication.
Acknowledgments
We thank the Director General of the Department of
Wildlife Conservation (DWC), its Research Committee, the
Deputy Director of Research, and DWC staff for their sup-
port and providing necessary permission for the collection
of voucher specimens (permit no: WL/3/2/1/9). We extend
our thanks to the Conservator General of Forest and the Con-
servator of Forests (Research and Education), and the Forest
Department staff for their support and providing necessary
permission. Thanks are also due to the Secretary to the Minis-
try of Environment, and members of the Loris Steering Com-
mittee for their advice on this work. Special thanks to Helga
Schulze for providing valuable literature and comments. We
also thank C. P. Rathnayaka for his contribution to the statisti-
cal analysis, C. J. Hettiarachchi for his contribution to GIS
work, and A. Wadugodapitiya for his contribution to mea-
suring the skulls at RCSL. We thank W. K. D. D. Liyanage
for valuable comments on the manuscript. We give special
acknowledgment to the dedicated and professional Slender
Loris Conservation Project eld team: C. A. Mahanayakage,
G. D. C. Pushpa-Kumara, V. Sumanasekara, G. Rathnayake,
D. R. Vidan¬apathirana, S. Ariyarathna, V. Rathnayake, D.
Samarasingha, D. De Silva, T. Senevirathna, R. Peries, U.
Wickramasingha, N. Goonasekara, L. Rathnayake, S. Jayal-
ath, P. Yahanpath and M. A. C. Maheshani. The authors are
grateful to Mendis Wickramasinghe and Anura Edirisinghe
for providing some site information. We also wish to thank
Ravindra Wickramasingha (National Museum, Colombo) for
preserving and mounting some voucher specimens.
We are greatly indebted to the curators and staff of the
National Museum of Sri Lanka, Colombo; University of
Colombo Zoology Museum; Natural History Museum of
London; Field Museum of Natural History, Chicago; Royal
College of Surgeons, London; and the National Museum of
Scotland for their support. The programme is funded through
the EDGE of Existence Programme of the Zoological Soci-
ety of London and the Synchronicity Foundation—we are
extremely grateful for their support. Furthermore, the rst
author (SNG) gratefully acknowledges the Conservation
Leadership Programme (CLP) and the People’s Trust for
Endangered Species (PTES), UK, for providing equipment.
The authors also acknowledge the staff of the University of
Colombo, the Open University of Sri Lanka, and the Field
Ornithology Group of Sri Lanka for their kind support during
the study. All work carried out falls under permits and ethics
approvals issued to and held by Dr. U. K. G. K. Padmalal and
the Open University of Sri Lanka by the Department of Wild-
life Conservation. We followed the American Society of Pri-
matologists’ principles for the ethical treatment of primates
and the International Primatological Society (IPS) interna-
tional guidelines for the acquisition, care and breeding of
nonhuman primates. We extend our gratitude to anonymous
reviewers and the editor, who graciously reviewed this manu-
script and provided helpful suggestions for its improvement.
Photo Credits: C. A. Mahanayakage, D. R. Vidanapathirana,
S. N. Gamage, G. D. C. Pushpa-Kumara, R. Peries, and A.
Wadugodapitiya.
Literature Cited
Bearder, S. K, K. A. I. Nekaris and D. J. Curtis. 2006. A re-
evaluation of the role of vision in the activity and com-
munication of nocturnal primates. Folia Primatol. 77:
50–71.
Brandon-Jones, D., A. A. Eudey, T. Geissmann, C. P. Groves,
D.J. Melnick, J. C. Morales, M. Shekelle and C.-B.
Slender lorises of Sri Lanka
105
Stewart. 2003. Asian primates classication. Int. J. Pri-
matol. 25: 97–164.
Cabrera, A. 1908. Sobre los loris, y en especial sobre la forma
lipina. Bol. R. Soc. Esp. Hist. Nat. 8: 135–139.
Coultas, D. S. 2002. Bioacoustic Analysis of the Loud Call
of Two Species of Slender Loris (Loris tardigradus and
Loris lydekkerianus nordicus) from Sri Lanka. Master’s
thesis, Oxford Brookes University, Oxford
Fischer, G. 1804. Anatomie der Maki. Andreäischen Buch-
handlung, Frankfurt.
Gamage, S. N. 2015. Revision of Taxonomy, Distribution,
Abundance and Habitat Preference of Slender Loris (Pri-
mates, Lorisidae: Loris) in Sri Lanka. PhD thesis, Uni-
versity of Colombo, Colombo, Sri Lanka.
Gamage, S. N., C. J. Hettiarachchi, C. A. Mahanayakage,
U. K. G. K. Padmalal and S. W. Kotagama. 2015. Fac-
tors inuencing site occupancy of montane slender loris
(Loris tardigradus nycticeboides) in Sri Lanka. Wild-
lanka 3(2): 67−73.
Gentry, A., C. P. Groves and P. D. Jenkins, 1998. Comment on
the proposed conservation of the name Loris E. Geoffroy
Saint-Hilaire, 1796 (Mammalia, Primates). Bull. Zool.
Nomencl. 55: 118–119.
Geoffroy Saint-Hilaire, É. 1796. Mémoire sur les rapports
naturels des Makis Lemur, L. et description d’une espèce
nouvelle de mammifère. Magasin Encyclopédique, An. 2,
1: 20–50.
Groves, C. P. 1998. Systematics of tarsiers and Lorises. Pri-
mates 39: 13–27.
Groves, C. P. 2001. Primate Taxonomy. Smithsonian Institu-
tion Press, Washington, DC.
Groves, C. P. 2003. Morphology, morphometrics and taxon-
omy. In: Field and Laboratory Methods in Primatology:
A Practical Guide, J. M. Setchell and D. J. Curtis (eds.),
pp.140–157. Cambridge University Press, Cambridge,
UK.
Groves, C. P. and P. Grubb. 2011. Ungulate Taxonomy. Johns
Hopkins University Press, Baltimore, MD.
Groves, C. P. and E. Meijaard. 2005. Interspecic variation
in Moschiola, the Indian chevrotain. Rafes Bull. Zool.
Suppl. (12): 413–421.
Groves, C. P., C. Rajapaksha and K. Manamendra-Arachchi.
2009. The taxonomy of the endemic golden palm civet of
Sri Lanka. Zool. J. Linn. Soc. 155: 238–251.
Hill, W. C. O. 1933. A monograph on the genus Loris with
an account of the external, cranial and dental characters
of the genus; a revision of the known forms; and the
description of a new form from Northern Ceylon. Ceylon
J. Sci. (Spolia Zeylanica) XVIII(2): 89–132.
Hill, W. C. O. 1942. The slender loris of the Horton Plains,
Ceylon, Loris tardigradus nycticeboides subsp. nov. J.
Bombay Nat. Hist. Soc. 43: 73–78.
Hill. W. C. O. 1953. Primates. Comparative Anatomy and
Taxonomy. I. Strepsirrhini. Edinburgh University Press,
Edinburgh.
Hill, W. C. O. and W. W. A. Phillips. 1932. A new race of
slender loris from the highlands of Ceylon. Ceylon J. Sci.
(Spolia Zeylanica) XVII(2): 109–122.
International Commission on Zoological Nomenclature. 1999.
Opinion 1922: Loris É. Geoffroy Saint-Hilaire, 1796
(Mammalia, Primates): conserved, and correction made
to the entry for Choloepus Illiger, 1811 (Xenarthra) on
the Ofcial List. Bull. Zool. Nomencl. 56: 101–103.
IUCN. (2013) IUCN Red List of Threatened Species. Version
2013.1. Available: http://www.iucnredlist.org (Date of
access on 22 June 2013).
Jayasuriya, A. H. M., D. Kitchener and C. Biradar. 2006.
Portfolio of Strategic Conservation Sites / Protected Area
Gap Analysis in Sri Lanka. Ministry of Environment and
Natural Resource, Sri Lanka. 284 pp.
Jenkins, P. D. 1987. Catalogue of Primates in the British
Museum (Natural History) and elsewhere in the British
Isles. Part IV: Suborder Strepsirrhini, including the Sub-
fossil Madagascar Lemurs and Family Tarsiidae. British
Museum (Natural History). London.
Kar Gupta, K. 2013. Slender Loris. In: Mammals of South
Asia., A. J. T. Johnsingh and N. Manjrekar (eds.), pp.94–
105. Universities Press, Hyderabad.
Kumara, H. N., M. Singh, M. Irfan-Ullah and S. Kumar. 2013.
Status, distribution, and conservation of slender lorises
in India. In: Leaping Ahead: Advances in Prosimian
Biology, J. Masters, M. Gamba and F. Génin, (eds.), pp.
343–352. Springer, New York.
Lesson, R. P. 1840. Species des Mammifères bimanes et qua-
drumanes; suivi d’un mémoire sur les Oryctéropes. J-B.
Baillière, Paris.
Linnaeus, C. 1758. Systema Naturae. 10th edition. L. Salvius,
Stockholm.
Lydekker, R. 1905. On two lorises. Proc. Zool. Soc. Lond.
(1904): 345–346.
Mace, G. M. 2004. The role of taxonomy in species conserva-
tion. Phil. Trans. Roy. Soc. B 359: 711–719.
MacKenzie, D. I. 2003. Assessing site occupancy modelling
as a tool for monitoring Mahoenui giant weta popula-
tions. DOC Science Internal Series 145. Department of
Conservation, Wellington, New Zealand.
Meegaskumbura, S. and C. J. Schneider. 2008. A taxonomic
evaluation of the shrew Suncus montanus (Soricidae:
Crocidurinae) of Sri Lanka and India. Ceylon J. Sci. (Biol.
Sci.) 37(2): 129–136.
Mittermeier, R. A., P. Robles Gil, M. Hoffmann, J. Pilgrim,
T. Brooks, C. G. Mittermeier, J. Lamoreux and G. A. B.
da Fonseca (eds.). 2004. Hotspots Revisited: Earth’s Bio-
logically Richest and Most Threatened Terrestrial Ecore-
gions. CEMEX, Mexico City.
Mitteroecker, P. and F. Bookstein. 2011. Linear discrimina-
tion, ordination, and the visualisation of selection gradi-
ents in modern morphometrics. Evol. Biol. 38: 100–114.
Munds, R. A., K. A. I. Nekaris and S. M. Ford. 2013. Tax-
onomy of the Bornean slow loris, with new species
Gamage et al.
106
Nycticebus kayan (Primates, Lorisidae). Am. J. Primatol.
75: 46–56. DOI:10.1002/ajp.22071.
Nekaris, K. A. I. and S. Jaffe. 2007. Unexpected diversity of
slow lorises (Nycticebus spp.) within the Javan pet trade:
implications for slow loris taxonomy. Contrib. Zool.
76(3):187−196.
Nekaris, K. A. I. and R. A. Munds. 2010. Using facial mark-
ings to unmask diversity: the slow lorises (Primates:
Lorisidae: Nycticebus) of Indonesia. In: The Primates of
Indonesia, S. Gursky and J. Supriatna (eds.), pp.383−396.
Springer, New York.
Nekaris, K. A. I. and J. Jayawardene. 2004. Survey of slender
loris (Primates, Lorisidae Grey, 1821: Loris tardigradus
Linnaeus, 1758 and Loris lydekkerianus Cabrera, 1908)
in Sri Lanka. J. Zool., Lond. 262: 327–338.
Nekaris, K. A. I., C. Roos, E. R. Pimley and H. Schulze. 2006.
Diversity slowly coming to light: reconsidering the tax-
onomy of pottos and lorises. Int. J. Primatol. 27 (suppl.
1): 441. [Abstract]
Nicholls, L. 1939. Period of gestation in Loris. Nature 143:
246.
Pethiyagoda, R. 2005. Exploring Sri Lanka’s biodiversity.
Rafes Bull. Zool. Supplement 12: 1–4.
Phillips, W. W. A. 1980. Manual of the Mammals of Sri Lanka.
3 Volumes. 2nd edition. Wildlife and Nature Protection
Society (WNPS) of Sri Lanka, Colombo.
Roos, C. 2003. Molekulare Phylogenie der Halbaffen, Sch-
lankaffen, und Gibbons. PhD thesis, Fakultät Wissen-
schaftszentrum Weihenstephan für Ernährung, Landnut-
zung, Umwelt der Technischen Universität München,
München.
Seba, A. 1735. Locupletissimi rerum naturalium accu-
rata descriptio et iconibus articiosissimis expressio
per universam physices historiam. Jansson-Waesberg,
Amsterdam.
Schultze, H., H. Preuschoft, C. P. Groves et al. 2003. Conser-
vation database for lorises (Loris, Nycticebus) and pottos
(Arctocebus, Perodicticus), prosimian primates. Stan-
dardized body measurement and description of lorises
and pottos for taxonomic purposes. Website: <http://
www.loris-conservation.org/database/population_data-
base/measure_index.html>. Accessed: 12 May 2017.
Sri Lanka, Ministry of Construction. 2012. Sri Lanka
2011–2030 National Physical Plan and Project Propos-
als. National Physical Planning Department, Ministry
of Construction, Engineering Services, Housing and
Common Amenities, Sethsiripaya, Battaramulla. 3 March
2012. URL:<http://www.acesl.org/download/conference/
tp_Lakshman%20Jayasekera%20and%20Veranjan%20
Kurukulasuriya.pdf>. Accessed 20 May 2017.
Stone, W. and J. A. G. Rehn. 1902. A collection of mammals
from Sumatra, with a review of the genera Nycticebus
and Tragulus. Proc. Acad. Nat. Sci. Philadelphia 54:
127–142.
Surridge, A. K., D. Osorio and N. I. Mundy. 2003. Evolution
and selection of trichromatic vision in primates. Trends
Ecol. Evol. 18: 198–205.
Tan, Y., A. D. Yoder, N. Yamashita and W. Le. 2005. Evi-
dence from opsin genes rejects nocturnality in ancestral
primates. Proc. Natl. Acad. Sci. USA 102:14712–14716.
Thomas, O. 1908. The nomenclature of certain lorises. Ann.
Mag. Nat. Hist., ser. 8, 1: 467−469.
Wadugodapitiya, A. 2011. Morphometric Variation in Slender
Lorises (Genus: Loris). MSc dissertation, Royal Veteri-
nary College, University of London, UK.
Weerakoon, D. K. and W. L. D. P. T. S. D. A. Goonatilake.
2006. Taxonomic Status of the Mammals of Sri Lanka.
The Fauna of Sri Lanka: Status of Taxonomy, Research
and Conservation. Edited by C. N. B. Bambaradeniya.
IUCN, Sri Lanka and Government of Sri Lanka.
White, L. and A. Edwards. 2000. Methods for assessing the
status of animal populations. In: Conservation Research
in the African Rainforests: A Technical Handbook, L.
White and A. Edwards (eds.), pp.191–201. Wildlife Con-
servation Society, New York.
Wijesinghe, L. C. A. D. S., I. A. U. N. Gunatilieke, S. D. G.
Jayawardena, S. W. Kotagama and C. V. S. Gunatilleke.
1993. Biological Conservation in Sri Lanka: A National
Status Report. IUCN Sri Lanka. 100pp.
Wroughton, R. C. 1917. The slender loris of Malabar. J.
Bombay Nat. Hist. Soc. 25: 45.
Yapa, A. and G. Ratnavira. 2013. Sri Lanka’s terrestrial and
marine mammals. In: The Mammals of Sri Lanka, A.Yapu
and G. Ratnavira, pp.126-−151. Field Ornithology Group
of Sri Lanka, Colombo.
Authors’ addresses:
Saman N. Gamage, Department of Zoology, Faculty of Sci-
ence, University of Colombo, Colombo 3, Sri Lanka, and
Land Owners Restore Rainforests in Sri Lanka, A30, Mad-
dumagewatta, Gangodawila, Nugegoda, Sri Lanka; Colin
P. Groves, School of Archaeology and Anthropology, Aus-
tralian National University, Canberra, Australia; Fais M. M.
T. Marikar, Staff Development Centre, Faculty of Medicine,
General Sir John Kotelawela Defence University, Ratmalana,
Sri Lanka; Craig S. Turner, Zoological Society of London,
Regent’s Park, London, NW1 4RY, UK; Kalinga U. K. G.
Padmalal, Department of Zoology, The Open University
of Sri Lanka, Nawala, Nugegoda, Sri Lanka; and Sarath W.
Kotagama, Department of Zoology, Faculty of Science, Uni-
versity of Colombo, Colombo 3, Sri Lanka. E-mail of rst
author: <samangam2004@yahoo.com>.
Received for publication:14 December 2016
Revised: 12 May 2017
Printed: 28 December 2017
... A comprehensive overview of the taxonomy of slender lorises, particularly those from Sri Lanka, is provided by Gamage et al. (2017). Osman Hill (1953) recognized six slender loris taxa of just one species, Loris tardigradus-two in India and four in Sri Lanka-but they are now grouped into two species, the grey slender loris with four subspecies (L. ...
... tardigradus tardigradus and L. t. nycticeboides) (Table 1). Gamage et al. (2017) added two subspecies from Sri Lanka (L. t. parvus and L. l. uva) noting the possibility of a third (a red slender loris L. tardigradus from Rakwana). ...
... Using mostly genetic data, Gamage et al. for the recognition of L. t. nycticeboides or L. l. uva. Gamage et al. (2017) published their taxonomic views on the grey and red slender lorises of Sri Lanka based on skull morphology, facemasks and pelage characteristics (grey slender lorises from India were not included in their study). The tion status and potential management implications of the two species and their putative subspecies in Sri Lanka, six being deemed valid by Gamage et al. (2017). ...
Article
Full-text available
Traditionally, two species of slender loris have been recognized, viz. the red slender loris Loris tardigradus, with two subspecies from Sri Lanka and the grey slender loris L. lydekkerianus, with four subspecies from Sri Lanka and India. In 2017, two more subspecies were described from Sri Lanka (L. t. parvus and L. l. uva) on the basis of morphological data, a conclusion purportedly supported by mtDNA sequence data. We conducted phylogenetic analyses of available cytochrome c oxidase subunit 1 (co1) sequences from 25 individuals of both the Sri Lankan slender loris species and found: (1) there is reasonable support for a two-species arrangement; (2) there is only a small amount of genetic differentiation within species; and (3) there is no support for the recognition of distinct, monophyletic taxa (subspecies or otherwise) that are genetically differentiated from other such taxa within each species. The management and practical conservation of the two species of slender lorises requires on-the-ground efforts in both India and Sri Lanka, and the availability of data on the distribution, morphology and genetics of the two species of slender loris needs to be taken into account when planning and executing conservation measures.
... The slender lorises (Suborder: Strepsirrhini, Family: Lorisidae, Genus: Loris É. (Geoffroy Saint-Hilaire, 1796) are small nocturnal primates inhabiting India and Sri Lanka (Groves, 1998(Groves, , 2001Nekaris and Jayawardene, 2004;Gamage et al., 2017). Based on museum specimens, Groves (1998) recognised 2 species of slender loris within Sri Lanka: Loris tardigradus (Linnaeus, 1758) and Loris lydekkerianus (Cabrera, 1908). ...
... Based on museum specimens, Groves (1998) recognised 2 species of slender loris within Sri Lanka: Loris tardigradus (Linnaeus, 1758) and Loris lydekkerianus (Cabrera, 1908). A Sri Lankan islandwide study conducted by Gamage et al. (2017) further confirmed that the classification of these 2 species needed to be reviewed. The classification of slender lorises in Sri Lanka suggested by Gamage et al. (2017) is as follows: the southwestern red slender loris Loris tardigradus tardigradus (Linnaeus, 1758), the montane slender loris Lori stardigradus nycticeboides, the northwestern red slender loris Loris tardigradus parvus (Gamage et al., 2017), the highland grey slender loris Loris lydekkerianus grandis (Hill,1953), northern grey slender loris Loris lydekkerianus nordicus (Hill, 1953), and the Uva red slender loris Loris lydekkerianus uva (Gamage,et al., 2017). ...
... A Sri Lankan islandwide study conducted by Gamage et al. (2017) further confirmed that the classification of these 2 species needed to be reviewed. The classification of slender lorises in Sri Lanka suggested by Gamage et al. (2017) is as follows: the southwestern red slender loris Loris tardigradus tardigradus (Linnaeus, 1758), the montane slender loris Lori stardigradus nycticeboides, the northwestern red slender loris Loris tardigradus parvus (Gamage et al., 2017), the highland grey slender loris Loris lydekkerianus grandis (Hill,1953), northern grey slender loris Loris lydekkerianus nordicus (Hill, 1953), and the Uva red slender loris Loris lydekkerianus uva (Gamage,et al., 2017). Gamage et al. (2017) discussed a further subgroup of slender loris from the Rakwana mountain range, which they nominated as the Rakwana slender loris. ...
Article
Full-text available
There are 2 species of slender lorises found in Sri Lanka: the red slender loris, Loris tardigradus (Linnaeus, 1758), endemic to the wet zone and montane areas, and the grey slender loris, Loris lydekkerianus (Cabrera, 1908), which is widespread. In addition, currently there are 2 subspecies recognized for each of these 2 slender loris species of Sri Lanka. Yet their taxonomy, evolutionary history, and biogeography are still poorly understood, and there are some preliminary data suggesting the presence of other subspecies. In this study, we have analysed the partial mitochondrial cytochrome oxidase I gene sequence of 21 Loris specimens collected from 7 different regions in Sri Lanka. The phylogenetic tree reconstructed based on the neighbour joining method and Bayesian approach revealed 2 clades with high branch supports (>60%), consistent with the 2 slender loris species. Moreover, the samples within the Loris lydekkerianus clade were divided into 3 taxa and the samples within the Loris tardigradus clade were divided into 4 taxa, all consistent with the 7 sampling zones. The results of the study have provided the first DNA barcoding analysis on slender lorises in Sri Lanka to resolve phylogenetic relationships. Further studies on Loris are necessary for decoding their taxonomic issues in Sri Lanka.
... The ranges of some taxa, particularly into the zone C may be greater (or less) than indicated. Most recently, morphometric comparisons among loris populations point to two new subspecies ( Gamage et al., 2017 online). Phylogenetic studies, in particular, have uncovered at least five new species that formerly were thought of as subspecies ( Groves and Meijaard, 2005;Meegaskumbura et al., 2007;Meegaskumbura and Schneider, 2008;Groves et al., 2009;Dissanayake and Oschida, 2012). ...
... Sri Lanka Pigmy Chevrotain (or Mouse-deer) D3 23 ,1,2 Gamage et al., 2017). ...
Article
Full-text available
All mammals originated on the supercontinent of Pangaea in the Mesozoic era during the “Age of Reptiles.” However, the crown ancestors of contemporary mammals did not flourish until major environmental and biotic changes had occurred. An asteroid collided with earth at the end of the Cretaceous Period (the K-Pg boundary event) wiping out non-flying dinosaurs and primitive mammals. It was followed by large-scale volcanism, a spike in atmospheric oxygen and the proliferation of flowering plants. New niches became available for the ancestors of today’s mammals to fill. Evidence suggesting whether the ancestors of the Sri Lankan and Indian mammals originated on the tectonically marooned Indian plate before crashing into Asia or on the Laurasian supercontinent is inconclusive. Modern Sri Lankan mammals show their greatest affinity with those of southern India, and were more diverse in the Pleistocene when rhinoceros, hippopotamus, wild dogs, gaur and lions enriched the island’s landscapes. Native Sri Lankan land-based mammals are diversified into about 108 unique taxa (among 91 species and 53 genera), differentiated as phenotypic adaptations to sharply contrasting environments among seven major phyto-climatic zones. Endemic subspecies are distributed fairly equally across different phyto-climatic zones (n=24 to 29), except in the highlands where they are fewer (n=14) having evolved rapidly to species and genera among the insectivores and rodents whose reproductive rates are high. Conversely, greater numbers of endemic species (n=13) and genera (n=3) occur in the highlands than in the other zones (2-6 endemic species, no endemic genera). The prevalence of endemism is inversely related to body size or vagility. This suggests a greater probability of genetic exchange among distant populations of large bodied mobile mammals within Sri Lanka, as well as with Indian fauna during periods of land bridges in the Pleistocene. Most (8 of 13) endangered and critically endangered endemic mammals occur in the wet montane regions that offer the least Protected Areas (PAs). More than 85% of PAs occur in the extensive dry zone about half of which is not suited for water dependent mammals whose distributions are restricted to alluvial forests (less than 1% of PAs). Historically, the national cost of conservation has been low and therefore politically palatable. Current conservation urgently requires a major change in management policy combined with realistic investment to prevent extinctions of many endemic mammals and other unique Sri Lankan biota.
... The Critically Endangered subspecies of the purple-faced langur S. v. nestor has received greatest attention [Dela, 2004[Dela, , 2011Rudran, 2007;Parker et al., 2008;Nijman and Nekaris, 2010;Nekaris et al., 2013;Cabral et al., 2018a, b;Panagoda and Weerasinghe, 2019;Rudran et al., 2020;Wijethilaka et al., 2021]. The loris of this region [Nekaris et al., 2012] as well as in other areas [Nekaris and Jayewardene, 2004] and of the disturbed montane forests also have been reviewed [Gamage et al., 2014[Gamage et al., , 2017. The gray langur, whose distribution is confined to the Dry Zone, has been assessed regarding HMC issues [Unanthanna and Wickramasinghe, 2010;Dittus et al., 2019]. ...
Article
Many investigators of human-monkey competition (HMC) in Sri Lanka have revealed some common threads. Except at temple and protected sites, all monkeys were considered as household or agricultural pests wherever they shared space with humans. This included the widely distributed toque macaque (Macaca sinica), the grey langur (Semnopithecus priam thersites) of the Dry Zone, and the purple-faced langur (S. vetulus) of the southwestern and central rain forests where human densities and habitat fragmentation were greatest. People sharing space with monkeys resorted to various non-lethal methods to chase monkeys away from their properties and most preferred to have monkeys removed to protected areas; such translocations have been politically popular, though contrary to ecological principles. The main cause of HMC near primate habitats has been environmental conversion to agriculture, whereas in many towns the refuse generated in the wake of widespread growing tourism lured omnivorous macaques towards human habitation and stimulated macaque population growth. While most Sri Lankans share space with monkeys reluctantly, only a minority, flouting cultural restraints, want monkeys destroyed. Nonetheless, a major threat to primate conservation has been habitat loss and the killing of monkeys, especially in the densely populated southwestern area of the island where recent surveys showed that most macaques have been wiped out. Two subspecies, S. v. nestor of the rain forest lowlands and M. s. opisthomelas of the montane forests, are Critically Endangered. Sharing space with monkeys rests on public tolerance, understanding, and empathy with monkeys. Religious concepts venerating monkeys provide fertile ground for this. Our science-based educational documentaries (n > 35), among other efforts, also have contributed to these human sentiments in Sri Lanka and globally. The trends in HMC suggest that protected nature reserves for all wildlife are more secure for primate survival than ethnoprimatology by itself would be. Rudran [Folia Primatologica 2021, DOI: 10.1159/000517176] criticized our recent publication on HMC in Sri Lanka [Dittus et al., Folia Primatologica 2019, 90: 89-108]. We consider his comments as misconstruing efforts in primate conservation through denying the importance of traditional protected areas, overlooking our achievements in educating the public and reducing HMC, as well as misunderstanding the limits of marketing monkeys to tourists as a source of income to support conservation.
... Loris tardigradus (Linnaeus, 1758) (Strepsirrhini, Lorisiformes, Lorisidae), the red slender loris, is a nocturnal slow climber endemic to the humid and montane forests of southwestern Sri Lanka, whose movements depend a lot on the grip of the limbs [64,65]. Two species of Loris are found in Sri Lanka, Loris tardigradus (red slender loris) and Loris lydekkerianus (grey slender loris) which is larger and can also be found in Southern India [66]. As emphasised by Subramoniam's drawings [64], the locomotor and postural activity of Loris tardigradus involves frequent tensile loading of the limbs [67]. ...
Article
Full-text available
The vestibular system of the mammalian inner ear senses angular and linear velocity of the head and enables animals to maintain their balance. Vestibular anatomy has been studied extensively in order to link its structure to particular kinds of locomotion. Available evidence indicates that, in primates, slow-moving species show higher levels of vestibular variation than fast-moving taxa. We analysed intraspecific morphological variation and fluctuating asymmetry (FA) levels in the semicircular canal systems of six species of lorisiform primates: three slow-moving lorisids and three fast-moving galagids. Our results showed clear differences in levels of intraspecific variation between slow-moving and fast-moving taxa. Higher levels of variation were responsible for deviations from coplanarity for synergistic pairs of canals in slower taxa. Lorisids also presented higher levels of FA than galagids. FA is a better indicator of agility than intraspecific variation. These results suggest that in order to function efficiently in fast taxa, semicircular canal systems must develop as symmetrically as possible, and should minimise the deviation from coplanarity for synergistic pairs. Higher levels of variation and asymmetry in slow-moving taxa may be related to lower levels of stabilising selection on the vestibular system, linked to a lower demand for rapid postural changes.
... This leaves Sri Lanka with the endemic species L. tardigradus and two endemic subspecies of L. lydekkerianus, the latter with two additional subspecies in India (see also Brandon-Jones et al., 2004). Taxonomy of the Sri Lankan Slender Lorises remains in flux, some authors keeping L. l. nordicus (Dittus, 2013;Roos et al., 2014), some moving the subspecies nycticeboides from L. lydekkerianus to L. tardigradus (Nekaris & Jayawardene, 2004) and some splitting the taxa into additional subspecies, L. t. parvus and L. l. uva (Gamage et al., 2017). The genetic data presented by Pozzi et al. (2015), confirms the existence of two species of Slender Loris (L. ...
Article
Full-text available
Five primate species, representing three genera and 12 subspecies, occur in Sri Lanka. We conducted an island-wide questionnaire survey of primate presence/absence, based on a 5x5 km grid with three questionnaires administered to residents in each cell. Respondents were queried on the presence of Slender Lorises, Toque Macaques, Sri Lankan Sacred Langurs and Purple-faced Langurs in their neighbourhood. Results indicated that Slender Lorises and Toque Macaques were distributed over 88% and 90% of Sri Lanka, respectively, including the wet and dry zones, but with patchy wet-zone distributions. Sri Lankan Sacred Langurs were present over 86% of the island but absent from large parts of the wet zone. Purple-faced Langurs were distributed over 53% of Sri Lanka with a disjunct distribution consisting of a patchy dry-zone presence and a more uniform wet-zone distribution. The maps presented are the first based on a systematic island-wide survey. We discuss the implications of the observed distributions for primate taxonomy and conservation
Article
Full-text available
The Montane slender loris Loris tardigradus nycticeboides is an endangered and endemic primate species inhabiting the montane region of Sri Lanka. Recent studies suggested the L. t. nycticeboides should be elevated to Critically Endangered status. However, information available for the effective conservation of the L. t. nycticeboides is limited. This research was undertaken to clarify the above issues and to gather data on factors influencing the occurrence of L. t. nycticeboides using the technique of occupancy modeling. The results show that the probability of detection (p) and site occupancy (Ψ) of the L. tardigradus varied due to numerous factors such as altitude, canopy connectivity and canopy height. Most suitable habitats for L. t. nycticeboides occupancy was montane forests situated in 1600-2100m altitude with tall canopy (height >4m) and good canopy connectivity (canopy openings<25%). Focus-ing on reducing the negative effects of these factors from anthropogenic activities is essential to ensure that L. tardigradus conservation programs are effective.
Chapter
Full-text available
According to literature there are 91 species of indigenous mammals in Sri Lanka, of which 16 species are endemic to the island. Another 12 species have been introduced to Sri Lanka, of which four species, namely Bubalas bubblis, Equus caballus, Equus asinus and Rattus norvegicus have well established feral populations. In addition, 26 species of marine mammals have been recorded from the maritime zone of Sri Lanka. The paper highlights the issues and research gaps related to the taxonomy of mammals of Sri Lanka
Article
Full-text available
The Sri Lankan and Indian populations of the mountain shrew Suncus montanus have hitherto been recognized as a single species, due to their similarity in size and colour. Here we present mitochondrial DNA sequence data from the cytochrome-b and 16S genes that suggest these populations represent distinct species. Phylogenetic analyses further reveal that the Sri Lankan and Indian populations are not sister taxa: S. montanus sensu stricto from Sri Lanka is the sister species of S. murinus, while 'S. montanus' from India is the sister species of S. stoliczkanus. The uncorrected genetic distance for cytochrome-b between Sri Lankan and Indian 'S. montanus' is ca. 7.5%, which is substantially higher than genetic distances within other species of shrews, or within vertebrate species in general, again supporting their recognition as distinct species. The name Suncus niger is available for the Indian shrew hitherto referred to S. montanus. Suncus montanus and S. niger should now be considered as endemic to Sri Lanka and India, respectively.
Chapter
Two recognized subspecies of slender loris, Loris lydekkerianus lydekkerianus and L. l. malabaricus , are endemic to Peninsular India. L. l. lydekkerianus has been reported from the drier forests of the Eastern Ghats south of the Godavari River, while L. l. malabaricus has been found in the wet forests of the Western Ghats south of the river Tapti. On the basis on con firmed sightings of individuals, we present a review of current understanding of the distribution and conservation status of India's endemic slender loris subspecies. A possible third subspecies occupies a patchy distribution along the eastern foothills of the southern Western Ghats. To facilitate conservation planning, we modeled potential distributions for each recognized subspecies.
Article
Fieldworkers are often unaware of the value of making accurate records of their study animals' morphology. External morphology includes measurements, observations of glandular activity and detailed description of the pelage; internal morphology may include skull and postcranial measurements, recording of suture closure, epiphyseal fusion, dental eruption and wear, and observations on the gut. All these observations will yield information on taxonomy, age–sex class (Table 9.1), reproductive status, individual variation, growth, development, sexual dimorphism and so on. It may also be interesting to look at individuals' behaviour in the light of their external differences: does facial coloration correlate with behaviour in mandrills (Mandrillus sphinx), does flange development correlate with behaviour in orangutans (Pongo spp.), and so on?
Article
IN a recent communication1, Dr. W. C. O. Hill suggests from certain evidence that ``the gestation period [in Loris] is somewhere hi the neighbourhood of six months''. The following history of a pair of lorises will be of interest, and confirms Dr. Hill's opinion.