Chapter 3: Population genetic structure of the Arunachal
macaque, a specialist primate from the edge of the
Tibetan Plateau and the bonnet macaque, a generalist
primate from the plains of peninsular India
3.1 Introduction
The potential impact of current climate change on the distribution and conservation of
biodiversity is the subject of global concern. However, climatic changes and consequential
changes in species population biology have always been a recurrent phenomenon throughout the
history of our planet, the most recent being during the Quaternary period (Hewitt 2004). These
effects further varied with latitude and landscape features and yielded different patterns. In
temperate species, for instance, the effect of glacial periods has generally been associated with
shifts in species distribution and reduction in their range due to the unavailability of suitable
habitat, whereas warmer interglacial periods have typically been associated with expansions of
population range and size. In Europe, for example, the Pleistocene ice sheet covered the
northern part of the continent during major glaciations (Taberlet et al. 1998). As a result, many
species in Europe either went extinct, shifted their distribution ranges or moved further south to
warmer climatic refugia, as, for example, in the Balkans, and the Italian and Spanish peninsulae
and following subsequent warmer inter-glaciations, expanded northwards again, reoccupying
their old habitats (Hewitt 2000). In particular, the last glacial–interglacial transition coincided
with major extinction events in numerous species distributed in the northern hemisphere
(Nogués-Bravo et al. 2010, Lorenzen et al. 2011).
In contrast, although severely understudied, species from the sub-tropics and the tropics tend to
exhibit very different kinds of phylogeography, driven perhaps by other environmental factors in
addition to the Pleistocene glacial oscillations. In eastern Eurasia, for example, there was no
large-scale continental glaciation during the Quaternary period. Instead, this region is composed
of multiple, historically dynamic, tectonic areas that underwent notable landmass configuration
changes such as the uplift of the Tibetan Plateau before the Pleistocene epoch (Royden et al.
2008). This, effectively, produced a complex regional topography, consisting of a network of
high mountains and deep valleys, and, as a consequence, relatively small alpine glaciers, about
2,000 m above sea level, were formed in the plateau and along its edges (Xia 1997; Zhou et al.
2006). More complex geo-climatic effects than only historical glaciations are, therefore, expected
to have influenced biodiversity distribution in this region than, for example, in Europe and
North America.
Other potential abiotic factors that can influence species phylogeography include local
topography or geographical barriers while biotic factors such as socio-behaviour may work at
relatively smaller scales. Social behaviour, particularly, may influence life history, and, in turn,
may drive population genetic structure among group-living mammals. In such species, gene flow
may be reduced between social groups, leading to genetic differentiation and potential inbreeding
(Altmann et al. 1996; Storz 1999,). In contrast, sociality may itself be influenced by a range of
other factors including resource availability and inbreeding avoidance (Wrangham 1980; Clutton-
brock and Lukas 2012). Hence, the tempo and mode of population diversification under the
influence of all these factors differ between taxa and vary considerably across regions.
3.1.1 The Arunachal macaque, an “edge” primate of the Tibetan Plateau
The role of the factors, mentioned above, in driving biotic diversity remain underexplored in the
Tibetan Plateau, particularly on its southern edges, which constitutes one of the largest and
richest biodiversity hotspots in Asia, the Eastern Himalayas (Myers et al. 2000; Mittermeier et al.
2004). The high degree of biodiversity of this region (WWF/IUCN 1995; Mittermeier et al. 1997;
Olson and Dinerstein 1998) and more remarkably, a recent spurt in the discovery of many new
species of plants and animals (Gillson 2004; Thompson 2009) suggests that this region may be
characterised by a complex combination of specialised ecological niches and glacial refugia,
which may have faced long periods of isolation.
One of the endemic species of this biodiversity-rich region is the Arunachal macaque Macaca
munzala, described for the first time from the state of Arunachal Pradesh, northeastern India in
2005 (Sinha et al. 2005; see also Chapter 1). Geographically, western and central Arunachal
Pradesh is a part of the highly undulating southern edge of the Tibetan Plateau while the more
flat southern Arunachal Pradesh is a part of the great Brahmaputra river plains. Due to the
recent discovery of the species, our knowledge of its biology is fragmentary although it is
possibly one of the most endangered of all Indian primates (IUCN 2012). We, however, know
that it belongs to the sinica species-group of macaques and its ancestors possibly originated from
the eastern side of the river Brahmaputra approximately 3.2 mya or million years ago
(Chakraborty et al. 2007; Chapter 2). Although we have now established that, evolutionarily, the
closest relative of this large-bodied montane species is the smaller-bodied and generalist primate
from the plains of peninsular India, the bonnet macaque M. radiata and considering that their
ancestors must have once shared a common habitat, it is far from clear how or when such a shift
in their present habitats occurred.
It is also important to understand how the complex geographical features of Arunachal Pradesh,
combined with the behavioural and ecological traits of the macaques, might have shaped the
species‟ phylogeographic distribution. For instance, Pleistocene glacial fluctuations and/or
specific physical barriers have been described as important forces driving the population genetic
structure of many apes (Eriksson et al. 2004; Anthony et al. 2007; Arora et al. 2010) and a few
Old World monkeys (Shimada 2000; Marmi et al. 2004; Modolo et al. 2005; Smith and
McDonough 2005; Belay and Mori 2006). Have similar forces driven the evolutionary history of
the Arunachal macaque as well? Moreover, being an „edge‟ species that occurs in the Eastern
Himalayan ranges at the edge of the Tibetan Plateau, it would be illuminating to investigate
whether the phylogeographic patterns displayed by the species is different from that of the few
other species, not necessarily of primates, that occur on the Plateau itself and which have been
studied recently (Yang et al. 2009; Qu et al. 2010; Fan et al. 2011; Zhao et al. 2012). In summary,
therefore, one of the principal aims of this chapter was to explore the impacts of the Pleistocene
climate change and the local topography on the phylogeography of the Arunachal macaque.
In addition, the Arunachal macaque appears to be a typical member of the sinica species-group of
the genus in exhibiting a matrifocal society with tolerant social relationships among the adult
males and females within a troop (Sinha et al. 2013; see also Chapter 1). Each troop typically
consists of one or two adult males, several adult females and a variable number of subadults,
juveniles and infants of both sexes. Although we currently lack detailed knowledge of the
dispersal behaviour of the species, it is likely that they exhibit female philopatry and male-biased
gene flow, as is typical of Old World monkeys. Nevertheless, this prediction needs to be tested
and a comparison of genetic markers of different inheritance, maternal versus biparental, across
individuals could help us achieve this goal. In addition, such a comparison would underline the
effects of sex-biased dispersal, if any, against the background of large-scale effects such as those
wrought by historical climate change and regional topography.
It is noteworthy that such a study is of special relevance today from a conservation perspective
as the Eastern Himalayas faces unprecedented ongoing habitat destruction (Sahni 1979; Cincotta
et al. 2000). It is also crucial to model the potential impacts of future climate change (Barnett et
al. 2005; Hofreiter and Stewart 2009) on highly endangered subtropical montane species such as
the Arunachal macaque, which are endemic to this biodiversity-rich region.
3.1.2 The bonnet macaque, the generalist primate of the Indian peninsula
Habitat-specialist species generally inhabit highly homogeneous and less fluctuating
environments. In contrast, generalist species tend to have very large distributions across varieties
of habitats, and possibly have higher dispersal abilities. Unlike the Arunachal macaque, which is a
specialist species, habitat generalist species might be expected to exhibit simple patterns of
genetic isolation-by-distance even in a heterogeneous landscape (Slatkin 1993), perhaps only to
account for obvious dispersal barriers such as mountains or rivers. However, very few generalist
species have been studied in this context and the emergent pattern of genetic structure in these
cases have been found to be far from simple (Sacks et al. 2004). It is also possible that genetic
differentiation may vary at different scales; there may thus be a lack of any differentiation at the
larger scale while a finer-scale structure may be evident (Ansmann et al. 2012).
The bonnet macaque is an extreme generalist primate, which is endemic to peninsular India. Of
the five primate species found in southern India, the bonnet macaque is the most common and
extensively distributed across the peninsula (Krishnan 1972). They are also very good swimmers,
including females with infants, and thus rivers, which are small or intermittently dry in southern
India, may not deter their dispersal (Sinha 2001). Biogeographically, the bonnet macaque
distribution is restricted to the Deccan Plateau of peninsular India, which is a more or less a flat
triangular area that forms the southern part of India. The plateau is located between two
mountain ranges: the Western Ghats, which forms its western boundary, and the much lower
and fragmented Eastern Ghats to its east. The Western Ghats runs north to south and separates
the plateau from a narrow coastal plain (approximately 30 – 50 km wide) along the Arabian Sea.
It, however, has three major geographical breaks, the youngest (65–80 million years ago, mya)
and northernmost (16°N) Goa Gap (Storey 1995; Gunnell et al. 2003) and the two older (500
mya) gaps, the Palghat Gap (widest, 40 km at 11°N) and the Shencottah Gap (narrowest, 7.5 km
at 9°N) (Soman et al. 1990; Santosh et al. 1992). The Western Ghats mediates the rainfall regime
of peninsular India by intercepting the southwestern monsoon winds. The western slopes of
Western Ghats experience heavy annual rainfall while the eastern slopes are drier; rainfall also
decreases from south to north. The wide variation of rainfall patterns in the Western Ghats,
coupled with the region‟s complex geography, produces a great variety of vegetation types. These
include scrub forests in the low-lying rain-shadow areas and the plains, deciduous and tropical
rainforests up to about 1,500 meters, and a unique mosaic of montane forests and rolling
grasslands above 1,500 meters. Some high altitude habitats of the Western Ghats mountains
seem to be the only potential natural geographical barrier to the highly mobile and adaptive
bonnet macaques (Sinha 2001). In addition, two sides of the mountain also vary climatically. If
the high elevation of the mountain has actually been a source of vicariance, then populations
from either side of the mountains are expected to show genetic differentiation from one another.
Thus, the main aim of this section of the chapter is to investigate the role of different factors,
including climatic influences, landscape features and aspects of socio-behaviour in shaping the
population genetic structure of this generalist primate species.
3.2 Methods
3.2.1 Study area and sampling
3.2.1.1 The Arunachal macaque
We obtained dried skin samples of Arunachal macaque individuals, killed and preserved as
hunting trophies, from 14 villages across the districts of Tawang, Upper Subansiri and West
Siang in the state of Arunachal Pradesh (Fig. 3.1). While Tawang forms the western edge of the
state, both Upper Subansiri and West Siang are located in remote central Arunachal Pradesh.
These samples were preserved in 95% ethanol at ambient temperature till they were transported
to the laboratory. A single blood sample was also obtained from a captive individual in Typee,
Tawang District. We considered the samples to belong to three tentative populations – Tawang
(n = 5), Upper Subansiri (n = 12) and West Siang (n = 7) on the basis of their geographic origins
(Table 3.1).
For the purpose of phylogenetic comparisons, we collected five blood samples of individual
bonnet macaques M. radiata from wild-caught, captive individuals of known origin maintained in
government-regulated animal facilities in the country (Table 3.1).
Figure 3.1 Map of Arunachal Pradesh with locations of the sampling sites. The triangles correspond to sampling sites in West Siang, circles to those in Upper Subansiri and rectangles to Tawang sampling sites. Inset: Location of the study site at the edge of Tibetan Plateau.
3.2.1.2 The bonnet macaque
We initially divided the bonnet macaque distribution range into three tentative regions, namely,
Region 1 (north of the Western Ghats Mountains), Region 2 (western extent of the Western
Ghats, Nilgiri Hills and Kodaikanal Hills) and Region 3 (eastern extent of the Western Ghats and
the Deccan Plateau). There were a total of 24 sampling locations, comprising six in Region 1,
eight in Region 2 and ten in Region 3 (Fig. 3.2). Our data totally included 114 individuals, 26 of
which were blood samples and 88 faecal ones (Table 3.2). Blood samples were collected from
government-registered zoos and animal facilities by trained veterinarians. Only wild-caught
individuals of known origin were sampled. All the faecal samples were fresh, collected non-
invasively from the field and immediately stored in 90% ethyl alcohol for 24 hours. We then
dried the samples with silica gel and stored them at ambient temperature.
3.2.2 DNA extraction and PCR amplification
We extracted genomic DNA from blood samples using the QIAGEN DNeasy Blood & Tissue
Kit (Qiagen GMBH, Hamburg, Germany), following the manufacturer‟s protocols. For faecal
samples, we used QIAamp DNA Stool Mini Kit (Qiagen GMBH, Hamburg, Germany) and
Table 3.1 Arunachal and bonnet macaque samples used in the study and their sites of origin.
Species, Population Location Sample Latitude Longitude Altitude (m)
Arunachal macaque, Tawang
Jang S1 27.58 91.98 -
Gronkhar S2 27.55 91.90 1997
Zemithang S3 27.72 91.73 -
Typee S6 27.11 92.57 -
Lomphu S7 27.71 91.72 -
Arunachal macaque, Upper Subansiri
Taksing S4 28.58 93.22 -
Ketenallah S9 28.21 93.32 1133
Yeaza S10 28.26 93.16 2100
Taksing
S11 28.58 93.72 2000
S12 28.58 93.72 2000
S13 28.58 93.72 2400
S15 28.58 93.72 2100
S16 28.58 93.72 2100
TCC Camp S17 28.58 93.72 2400
S18 28.58 93.72 2000
Arunachal macaque, West Siang
Lungte
S19 28.36 94.24 1480
S20 28.36 94.24 1480
S21 28.36 94.24 1480
Peidi S22 28.37 94.21 1640
Papikurung
S23 28.48 94.16 2051
S24 28.48 94.16 2051
S25 28.48 94.16 2051
S26 28.48 94.16 2051
Tato Gitu S40 28.51 94.42 1238
Bonnet macaque
Yehalanka 607 13.13 77.59 -
Nelamangala 326 13.50 77.23 -
Tumkur 30 13.34 77.10 -
Aurangabad Au1 19.89 75.32 -
Chamundi Hills BS1 12.30 76.65 -
Pimpri Pi3 18.62 73.80 -
Surat Su1 21.20 72.82 -
Thrissur TCR3 10.52 76.22 -
followed the manufacturer‟s protocol with a slight modification; we doubled the time for all
incubation and elution steps. We used extraction blanks as negative controls in downstream
polymerase chain reaction (PCR) amplifications.
3.2.2.1 The Arunachal macaque
In order to sample a non-coding region of the Arunachal macaque mitochondrial genome, we
amplified a 534bp-long D-loop (hyper-variable segment 1) region using the primer set from Li
and Zhang (2005). We conducted standard 35-cycle PCR to amplify the target regions following
Chakraborty et al. (2007; also described in Chapter 2).
We also downloaded sequences of Arunachal, bonnet, Assamese, Tibetan, stump-tailed
macaques and Hamadryas baboons from Genbank and used them in our molecular phylogenetic
analyses.
We amplified 22 fluorescently labelled microsatellite loci (DXS571, DXS6810, DXS8043,
DXS6799, D20S171, D4S243, D12S372, D8S1466, D9S934, D7S794, D10S611, D8S1106,
D15S823, D19S255, D2S146, D17S791, D18S869, D18S537, D6S2419, D16S403, D5S1457,
D10S179, D11S2002), already established for other macaque species (Rogers et al. 2005;
Kanthaswamy et al. 2006). PCR amplification of 35 cycles was conducted for upto five loci
simultaneously with combinations selected on the basis of fragment size, annealing temperature,
and the fluorescent dye set DS - 33 components used (dy6FAM, VIC, PET or NED). The PCR
products were resolved with an ABI 3130xl automated sequencer and analysed with
GeneMapper software (version 4.0; Applied Biosystems, Foster City, USA).
3.2.2.2 The bonnet macaque
In order to sample a non-coding region of the mitochondrial genome of bonnet macaque, we
amplified a 630 bp-long D-loop (hyper-variable segment 1) region using the primers, BF
(5' AACTGAACCCCTCATCACCA 3') and BR (5' GTAGCACTCTTGTGCGGGAT 3'),
designed by us. We conducted standard 35-cycle PCR to amplify the target region of the DNA
from blood samples but increased the number of cycles to 45 for the faecal samples. We used an
annealing temperature of 59.1°C for the primers.
Figure 3.2 Map of peninsular India with locations of the sampling sites of bonnet macaque individuals. The circles correspond to Region 1, triangles to Region 2 and the rectangles to Region 3 sites. Inset: Distribution range of bonnet macaques in peninsular India.
Table 3.2 Bonnet macaque samples used in the study and their sites of origin.
Region Population Sample Source of DNA
Latitude Longitude
Region 1
Surat
Surat1 Blood 21.23 72.90
Surat2 Blood 21.23 72.90
Surat3 Blood 21.23 72.90
Maharashtra1
Pimpri1 Blood 18.69 73.80
Pimpri2 Blood 18.69 73.80
Pimpri3 Blood 18.69 73.80
Pune1 Blood 18.52 73.86
Pune2 Blood 18.52 73.86
Maharashtra2
Aurangabad1 Blood 19.88 75.34
Ahmadnagar1 Blood 19.88 75.34
Ahmadnagar2 Blood 19.88 75.34
Ramling3 Faeces 18.17 76.05
Ramling2 Faeces 18.17 76.05
Ramling4 Faeces 18.17 76.05
Panchgani Panchgani1 Faeces 17.92 73.79
Maharashtra3
Elephanta Island1
Faeces 18.96 72.93
ElephantaIsland2 Faeces 18.96 72.93
ElephantaIsland3 Faeces 18.96 72.93
Mumbai1 Blood 18.98 72.84
Nasik Shaptashrung2 Faeces 20.33 73.88
Shaptashrung3 Faeces 20.33 73.88
Region 2 AndhraPradesh
Atmakur3 Faeces 15.88 78.58
Appanpalli2 Faeces 16.76 78.05
Appanpalli3 Faeces 16.76 78.05
Gooty2 Faeces 15.11 77.54
Pennakonda3 Faeces 14.13 77.60
Pennakonda1 Faeces 15.33 78.16
Pennakonda2 Faeces 15.37 78.90
Chandragodu1 Faeces 13.58 79.32
Chandragodu2 Faeces 13.58 79.32
Pataparu3 Faeces 15.33 78.16
Guddalur3 Faeces 15.37 78.90
Mahanandi4 Faeces 15.41 78.59
Kurnool1 Faeces 15.55 78.28
Kurnool2 Faeces 15.55 78.28
Kurnool3 Faeces 15.55 78.28
Kurnool5 Faeces 15.55 78.28
Kurnool6 Faeces 15.55 78.28
Kurnool7 Faeces 15.55 78.28
Kurnool8 Faeces 15.55 78.28
Karnataka1
Badami1 Faeces 15.92 75.68
Badami5 Faeces 15.92 75.68
Badami7 Faeces 15.92 75.68
Yelatti2 Faeces 16.50 75.29
Karnataka2 Aiyanur2 Faeces 13.92 75.57
Aiyanur1 Faeces 13.92 75.57
Karnataka3
Yehalanka607 Blood 13.12 77.59
Nelamangala690 Blood 13.10 77.39
Nelamangala326 Blood 13.10 77.39
Shivagange30 Blood 13.17 77.24
Yehalanka617 Blood 13.12 77.59
Yehalanka38 Blood 13.12 77.59
Yehalanka35 Blood 13.12 77.59
PRL201 Blood 13.02 77.57
PRL327 Blood 13.02 77.57
PRL55 Blood 13.02 77.57
GKVK Faeces 13.08 77.58
BangaloreUniv1 Faeces 12.94 77.50
Karnataka4
Bandipur1 Faeces 11.67 76.63
Bandipur2 Faeces 11.67 76.63
Bandipur4 Faeces 11.67 76.63
TamilNadu1
Thiruvannamalai2 Faeces 12.23 79.08
Thiruvannamalai3 Faeces 12.23 79.08
Thiruvannamalai1 Faeces 12.23 79.08
AnnaUniv2 Faeces 13.01 80.24
TamilNadu2 Kodaikkarai2 Faeces 10.37 79.78
Kodaikkarai1 Faeces 10.37 79.78
Suryakad1 Faeces 11.22 79.75
Suryakad2 Faeces 11.22 79.75
Kodaikkarai3 Faeces 10.37 79.78
Kodaikkarai4 Faeces 10.37 79.78
Pudukottai2 Faeces 10.38 78.83
TamilNadu3
Alagar Coil3 Faeces 10.07 78.21
Alagar Coil2 Faeces 10.07 78.21
AlagarCoil1 Faeces 10.07 78.21
Region 3
NilgiriHills
Nilgiri Hills4 Faeces 11.42 76.50
Nilgiri Hills5 Faeces 11.42 76.50
NilgiriHills1 Faeces 11.42 76.50
KodaiKanalHills Silver Cascade3 Faeces 10.24 77.51
Attakatti
Attakatti1 Faeces 10.44 76.99
Attakatti3 Faeces 10.44 76.99
Attakatti4 Faeces 10.44 76.99
ThenmalaShendurni
ThenmalaJ2 Faeces 8.95 77.07
ThenmalaJ1 Faeces 8.95 77.07
Thenmala2 Faeces 8.95 77.07
Shendurni2 Faeces 8.94 77.05
Shendurni3 Faeces 8.94 77.05
Shendurni1 Faeces 8.94 77.05
Kerala1
Calicut1 Faeces 11.26 75.78
Chelavoor3 Faeces 11.30 75.85
Chelavoor5 Faeces 11.30 75.85
Phookote1 Faeces 11.54 76.03
Phookote2 Faeces 11.54 76.03
Phookote3 Faeces 11.54 76.03
Phookote4 Faeces 11.54 76.03
Phookote5 Faeces 11.54 76.03
Kerala2
Trichur1 Blood 10.53 76.22
Trichur2 Blood 10.53 76.22
Trichur3 Blood 10.53 76.22
Trichur4 Blood 10.53 76.22
Walayar2 Faeces 10.83 76.85
Kerala3
Periyar2 Faeces 9.62 77.19
Tholupuzha1 Faeces 9.89 76.72
Tholupuzha2 Faeces 9.89 76.72
Tholupuzha4 Faeces 9.89 76.72
Tholupuzha4(2) Faeces 9.89 76.72
Kerala4
Elanjimel3 Faeces 9.29 76.57
Elanjimel2 Faeces 9.29 76.57
Elanjimel4 Faeces 9.29 76.57
Kollam2 Faeces 8.88 76.60
Kollam3 Faeces 8.88 76.60
Kerala5 Pechipara1 Faeces 8.47 77.29
Pechipara4 Faeces 8.47 77.29
KMTR KMTRKN Faeces 8.65 77.37
We next amplified 22 fluorescently labelled microsatellite loci, already established for other
macaque species (Rogers et al. 2005; Kanthaswamy et al. 2006). PCR amplification of 35 cycles
was conducted for upto five loci simultaneously with combinations selected on the basis of
fragment size, annealing temperature, and the components of the fluorescent dye set DS-33 used
(dy6FAM, VIC, PET or NED). The PCR products were resolved with an ABI 3130xl automated
sequencer and analysed with GeneMapper software (version 4.0; Applied Biosystems, Foster
City, USA). Not all of the loci gave satisfactory results for the faecal DNA. Hence, we filtered
the data further and retained only the ten loci (D17S791, DXS571, D7S794, D5S1457, D16S403,
D9S934, DXS6810, D8S1106, D12S372, D6S2419) that yielded consistent results. Each
individual sample was genotyped four times independently for each of these ten loci. We only
used those genotypes that were consistent at least for 75% of the replicates (Mondol et al. 2009).
3.2.3 Phylogenetic reconstruction
3.2.3.1 The Arunachal macaque
We conducted model selection tests on jModelTest 0.1 (Posada 2008), using the Akaike
information criterion to choose the most suitable substitution model for our data. These analyses
revealed HKY+Γ+I to be the most appropriate evolutionary model for our mtDNA data.
Phylogenetic analyses were conducted using the maximum likelihood (ML) and Bayesian
inference (BI) methods in PAUP* (Swofford 2002) and MrBayes, version 3.2 (Ronquist et al.
2012) respectively. The ML trees were reconstructed with 1000 bootstrap replicates. The BI
analyses were run for 107 generations for the mitochondrial data to ensure convergence. Samples
were collected every 1000 generations and 4 chains (1 cold and 3 heated) were used for the
Markov chain Monte Carlo (MCMC) procedure in all cases. The first 25% of the collected
posterior data were discarded to allow „burn-in‟ (Ronquist et al. 2012).
To infer the coalescence date for Arunachal macaque mtDNA haplotypes, we used a Bayesian
Markov chain Monte Carlo analysis as implemented in BEAST 1.7.2 (Drummond et al. 2012).
Based on the Akaike information criterion from jModelTest 0.1 (Posada 2008), we selected the
HKY+ Γ+I model. We used an uncorrelated relaxed log-normal clock (Drummond et al. 2006),
specifying a normal distribution with a mean HVS1 substitution rate of 0.1643 substitutions per
nucleotide per million year (Myr) for the mean rate prior. We chose this corrected HVS1
estimate (Soares et al. 2009) because it takes into account the effects of purifying selection on the
entire mtDNA molecule as well as saturation factors affecting the molecular rate decay described
in many studies (Ho et al. 2005; Endicott et al. 2009); it is therefore appropriate for population-
level analyses (Ho et al. 2008). The 95% confidence interval for the normal distribution spanned
HVS1 substitution rates obtained in other studies, from 0.07 to 0.25 substitutions/site/Myr
(Santos et al. 2005). Additionally, we used three divergence estimates as priors. The three
calibration points were the baboon-macaque divergence, approximately between 8.6 and 10.9
million years ago (mya) based on whole mtDNA (Raaum et al. 2005), the sinica-fascicularis species-
group divergence, at approximately 3.2 (SD 0.3) mya (Tosi et al. 2003) and the last common
ancestor of the sinica species-group, at approximately 1.7 mya (Tosi et al. 2003). For the baboon-
macaque divergence, we used a normal mean of 9.75 and SD of 0.42, thereby obtaining a broad
distribution with a 95% interval from 8.6 to 10.9 mya. For the sinica-fascicularis species-group
calibration, we used a normal mean of 3.2 and SD of 0.3, spanning a 95% interval from 2.3 to
3.9 mya. Finally, we used a normal mean of 1.7 and SD of 0.4 for the last calibration point, thus
spanning a wide timescale between 1.04 and 2.36 mya.
Using the birth-death prior for tree-branching rates, we carried out three runs for 106 generations
with parameter sampling every 1000 generations. Tracer 1.4.1 (Rambaut and Drummond 2005)
was then used to examine whether the 10% burn-in period and effective sample sizes were
adequate. All the runs were combined in LogCombiner 1.4.8 (Drummond et al. 2012), and the
resulting tree visualised and edited using Figtree 1.2 (Rambaut 2008).
3.2.3.1 The bonnet macaque
We conducted model selection tests on jModelTest 0.1 (Posada 2008), using the Akaike
information criterion to choose the most suitable substitution model for our data. These analyses
revealed TVM+Γ+I to be the most appropriate evolutionary model for our mtDNA data.
Phylogenetic analyses were conducted using the Bayesian inference (BI) methods in Beast,
version 1.7.2 (Drummond et al. 2012). The BI analyses were conducted exactly as described in
Chapter 3. To determine the coalescence dates of the bonnet macaque mtDNA haplotype clades,
we used a Bayesian Markov chain Monte Carlo (MCMC) analysis as implemented in Beast,
version 1.7.2 (Drummond et al. 2012). We included the collapsed haplotypes from bonnet
macaque (98 individuals) as well as from chimpanzee Pan troglodytes (3), rhesus macaque Macaca
mulatta (1), long-tailed macaque M. fascicularis (1), hamadryas baboon Papio hamadryas (1), gelada
Theropithecus gelada (1) and grivet Chlorocebus aethiops (1) as outgroups. Based on the Akaike
information criterion from jModelTest 0.1 (Posada 2008), we selected the HKY+ Γ+I model.
We used an uncorrelated relaxed log-normal clock (Drummond et al. 2006), specifying a normal
distribution with a mean HVS1 substitution rate of 0.1643 substitutions per nucleotide per
million year (Myr) for the mean rate prior (see section 3.2.3.1). Additionally, we used two
divergence estimates as priors. The two calibration points were the hominoid-cercopithecoid
divergence, approximately 35 million years ago (mya), based on genomic data (Steiper et al. 2004)
and the baboon-macaque divergence, approximately between 8.6 and 10.9 million years ago
(mya), based on whole mtDNA (Raaum et al. 2005). Other priors and MCMC conditions used in
this analysis were exactly as described in section 3.2.3.1.
3.2.4 Genetic structure analysis
3.2.4.1 The Arunachal macaque
First, we calculated modified moment estimates of F-statistics using an analysis-of-molecular-
variance (AMOVA) approach with Arlequin, version 3.5 (Excoffier et al. 2005) to investigate the
genetic differentiation between the three study sites for both the data sets.
For haplotypic data (mtDNA), Arlequin estimates ΦST using information from both the allelic
content and frequency of haplotypes (Excoffier et al. 1992). For genotypic data (microsatellites),
with an unknown gametic phase, as is the case for most natural populations, the AMOVA is
based on F-statistics. The algorithm partitions the total genetic variance into covariance
components, due to inter-individual and inter-population differences, following a hierarchical
analysis of variance (Weir 1996). The covariance components were then used to compute
fixation indices (Wright 1965). The values span between 0 and 1. Values in the range of 0 – 0.05
indicate “little” genetic differentiation; 0.05 – 0.15, “moderate” differentiation; 0.15 – 0.25,
“great” differentiation; and values above 0.25, “very great” genetic differentiation (Wright 1978;
Hartl and Clark 1997). The significance of the fixation indices was tested using a non-parametric
permutation test (Excoffier et al. 1992) with 10000 permutations. The tested populations were
considered to be genetically differentiated if inter-population genetic variation was found to be
higher than that within the populations.
Geographic patterns of genetic structure can often entail complex combinations of clines,
clusters and patterns of isolation by distance (François and Durand 2010), and multiple analysis
methods can provide complimentary information regarding such patterns (Balkenhol et al. 2009).
Thus, as another means of examining patterns of genetic structure, we employed a spatial
Bayesian clustering method using the program BAPS 5 (Corander et al. 2008). For mtDNA data,
we performed a mixture analysis using the „clustering of linked molecular data‟ method. The
analysis consisted of five iterations of each value of K max (the maximum number of
populations) for the range K max = 1 – 20. This step determines the optimum number of
genetic clusters in the sample based on the partition with the maximum likelihood [L(K)] and
highest posterior probability (p), and then assigns each individual to a cluster. For microsatellite
data, we first performed a mixture analysis using the „clustering of groups of individuals‟ model
and with similar number of iterations as above. The model showed limited power to differentiate
the populations for microsatellite data.
To further test the population structure with more complex models, we employed the model-
based algorithm in the program Structure, version 2.3.3 (Pritchard et al. 2000; Falush et al. 2003,
2007) which, given the number of clusters (K), estimates allele frequencies in each cluster and
the population membership of each individual. We initially tested two-ancestry models with both
„no-admixture‟ and „admixture‟ scenarios to estimate admixture proportions for every individual.
As these models showed limited assignment power, the Locprior model, which allows for the
use of sample group information in the clustering process and which is capable of detecting
structures at lower levels of divergence or with less data, was then implemented (Hubisz et al.
2009). It is noteworthy that this model does not find structure when none is present and is able
to ignore the sample group information when the ancestries of individuals do not correlate with
their sampling locations (Hubisz et al. 2009; Arora et al. 2010). Each model considered K
between one and six and all the models were tested with 100000 Markov Chain Monte Carlo
(MCMC) simulations (20000 burn-in) with each simulation repeated ten times. We used the
online version 6.0.1 of Structure Harvester (Earl 2011) to plot the maximal values of ln P(D), the
posterior probability of the data for a given K and ΔK based on the rate of change in the log
probability of data between successive K-values (Evanno et al. 2005), and identify K.
3.2.4.2 The bonnet macaque
We calculated modified moment estimates of F-statistics using an analysis-of-molecular-variance
(AMOVA) approach with Arlequin, version 3.5 (Excoffier et al. 2005) to investigate the genetic
differentiation between the three putative regions for both the data sets.
To further test the population structure with the microsatellite data, we employed the model-
based algorithm in the program Structure, version 2.3.3 (Pritchard et al. 2000; Falush et al. 2003,
2007).
Both the above analyses exactly followed the procedures described in the Section 3.2.4.1.
3.2.5 Gene diversity in Arunachal macaque populations
We calculated gene diversity, the expected heterozygosity under random mating, for both the
datasets. For microsatellite data, gene diversity can be defined as the probability that two
randomly chosen haplotypes are different in the sample. It, basically, is equivalent to the average
of expected heterozygosities over all the loci. For mtDNA, gene diversity, also known as
nucleotide diversity for sequence data, is the probability that two randomly chosen homologous
nucleotide sites are different. It is a good measure of genetic diversity and is relatively little
affected by sample size (Chikhi and Bruford 2005). For microsatellites, gene diversity can vary
enormously even for large sample sizes, but this effect can be reduced when many loci are used
such as in this study (22 loci), or when the two alleles have similar frequencies (Chikhi and
Bruford 2005).
3.2.6 Gene flow between Arunachal macaque populations
We estimated population pair-wise estimates of FST, which provides some insight into the degree
to which the populations are historically connected (Slatkin 1991; Holsinger and Mason-Gamer
1996). They, by themselves, however, do not allow us to determine whether the connections
between populations are actually results of long-term gene flow between them or rather that of
recent common ancestry (Nielsen and Wakeley 2001). But we tried to differentiate between these
two scenarios with the help of phylogenetic reconstruction of the studied populations. We
employed Arlequin, version 3.5 (Excoffier et al. 2005) to compute pairwise FST for all pairs of
populations.
The null distribution of pairwise FST values under the hypothesis of no difference between the
populations was obtained by permuting haplotypes between them. The P-value of the test was
the proportion of permutations leading to a FST value larger or equal to the observed one.
3.2.7 Genetic variability and differentiation in bonnet macaque populations
The level of genetic differentiation between groups was determined by calculating pairwise FST
(Weir and Cockerham 1984) and testing for significance using 1000 permutations. Genetic
variation within groups was estimated by calculating the mean number of alleles (NA), allelic
range, inbreeding coefficient (FIS) and the observed and unbiased expected heterozygosity (Ho
and He) using Arlequin, version 3.5 (Excoffier et al. 2005).
3.3 Results
3.3.1 The Arunachal macaque
3.3.1.1 Phylogenetic reconstruction and divergence times
We generated both the maximum likelihood (ML) and Bayesian inference (BI) phylogenetic
trees. We only show the BI cladogram here as both of them are well supported and show similar
topologies (Fig. 3.3). Most of the nodes have bootstrap values more than 80% and none lower
than 50%. The eastern Assamese and Tibetan macaque individuals form a monophyletic out-
group to the Arunachal and bonnet macaque clades, as was demonstrated in Chapter 2 (see also
Chakraborty et al. 2007). The Arunachal macaque individuals, however, show extensive
diversification across sampling sites. Individuals across the three sampling locations form three
distinct monophyletic clades according to their geographic origins.
Figure 3.3 Bayesian phylogenetic tree of the Arunachal macaque mtDNA haplotypes. The posterior probabilities and maximum likelihood values are above or below the tree branches. The differently coloured bars at the branch tips correspond to the species and population locations in Figure 1.
Surprisingly though, while the Tawang and Upper Subansiri groups share a most recent common
ancestor, the West Siang group seemed to have diverged from a common ancestor with bonnet
macaque, which itself is monophyletic in nature. None of these groups, however, are reciprocally
monophyletic, thus attesting to an intriguing phylogeographic pattern and resulting complex
phylogenetic relationship between the two species.
According to our estimates, the putative ancestor of the sinica species-group began to diversify
during the transition period between the Pliocene and Pleistocene epochs, approximately 2.2
(95% Highest Posterior Density or HPD 1.65 – 2.75) mya. Next, the diversification of the
Arunachal macaque-bonnet macaque ancestor into two stocks appears to have occurred during
the early Pleistocene, approximately 1.61 (95% HPD 1.14 – 2.12) mya. It was soon followed by
the separation of the West Siang ancestors from that of the bonnet macaque, approximately 1.32
(95% HPD 0.87 – 1.81) mya. Finally, the ancestors of the present-day Tawang and Upper
Subansiri populations diverged approximately 0.80 (95% HPD 0.5 – 1.16) mya during the middle
Pleistocene period.
3.3.1.2 Population genetic structure
The AMOVA yielded results that were clear but contradictory in nature for both the markers. A
considerable 64.09% of the total genetic variation in mtDNA was partitioned among the studied
populations than within (35.91%) each of them. As a result, the fixation index (ΦST = 0.64, P <<
0.05) was quite high. Thus, the mtDNA analysis exhibited a signature of population
differentiation between the Arunachal macaque populations (Table 3.3).
Table 3.3 Analysis of molecular variance (AMOVA) of the Arunachal macaque populations showing strong population differentiation for mitochondrial DNA but a lack of differentiation for nuclear DNA. Both the fixation indices are significant (P << 0.05).
Source of Variation Percentage of Variation
(mtDNA)*
Percentage of Variation
(microsatellites)#
Among populations 64.09 2.56
Within populations 35.91 97.45
Fixation index 0.64 0.03
* Populations: Tawang, Upper Subansiri and West Siang #Populations: Tawang-West Siang and Upper Subansiri
In contrast, microsatellite data analysis revealed an immense 95.98% of the total genetic variation
to be partitioned within the study sites in comparison to a mere 4.02% amongst them.
Consequently, the fixation index was low (FST = 0.04, P << 0.05), suggesting that 4% of the
observed genetic diversity alone was accounted for by genetic differences among the three
tentative geographical populations; these populations, thus, appear to be more genetically similar
to one another than different, at least, for the nuclear markers.
The mixture model for the linked sequence data in BAPS predicted population structuring within
Arunachal macaque samples (Fig. 3.4). The three sampling locations largely exhibited their
distinctness with the formation of three separate clusters – Tawang (Red), Upper Subansiri
(Green) and West Siang (Blue). However, while all the individuals from Tawang clearly adhered
to their geographic fidelity, it is not exactly the case for a few individuals from other two
locations. For example, individuals S19, S21 and S40 were sampled from West Siang but
according to their genetic assignment, were associated with the Upper Subansiri (S40) and an
unknown (Pink, S19 and S21) populations. Similarly, two individuals (Yellow, S15 and S17),
sampled from Upper Subansiri, were found to be of unknown origin. Four of these „migrant‟
individuals are males (S15, S19, S21 and S40) while only one is female (S17), according to the
molecular sexing that we independently carried out (data not shown).
Figure 3.4 Genetic structure of the Arunachal macaque populations based on mitochondrial DNA. A Bayesian clustering method (BAPS 5) was used to infer the clusters.
Our analysis of the microsatellite data with BAPS failed to predict any population structure,
which may have been the result of extremely low level of population differentiation. Similarly, in
Structure too, the admixture model did not yield a clear genetic population structure for the
samples, giving further support to a lack of strong population structure in the sampled Arunachal
macaque populations. The Locprior model, however, predicted a putative two (K = 2)
population genetic structure, though the signal continued to be weak (Log likelihood, Ln‟(K) =
−94.93). In accordance with this prediction, individuals from Tawang and West Siang
populations clustered together as one population while individuals from the Upper Subansiri
population maintained their distinctness (Fig.3.5).
Figure 3.5 Genetic structure of Arunachal macaque populations based on nuclear (microsatellite) markers. We used Locprior model (Structure 2.3.3) to infer the population structure.
At this stage, following the Structure results, we reconsidered our AMOVA analysis for the
microsatellite data, putting the Tawang and West Siang individuals together. As a result, the
percentage of total genetic variation between the tested populations further reduced to a scant
2.56% (Table 3.3).
3.3.1.3 Gene diversity
The values of the mtDNA nucleotide diversity of the Arunachal macaque populations seemed to
vary from moderate to low, with values 0.06 (West Siang), 0.03 (Upper Subansiri) and 0.01
(Tawang), when compared to other primate species where they have been usually observed to be
low due to long-term population contractions (Chikhi and Bruford 2005; Modolo et al. 2005;
Smith and McDonough 2005). The microsatellite gene diversity, however, was higher than those
of mtDNA, as has been found for many cercopithecine primates (Chikhi and Bruford 2005).
3.3.1.4 Gene flow
Finally, the population pairwise FST for mtDNA showed significant (P << 0.05) differentiation
between the three sampling sites (Table 3.4A). Interestingly, the Tawang and Upper Subansiri
groups, which share the most recent common ancestor, also showed the highest FST value (0.7, P
= 0.001). In contrast, the West Siang population, which is phylogenetically paraphyletic to both
the other groups, shared the same FST values with both of them (0.62, P << 0.05). Thus, the
obtained FST values appear to have originated from a gene flow between the populations than
from their recent shared ancestry. They also suggest that the West Siang population might have
been historically more connected independently to the Tawang and Upper Subansiri populations
than the two with one another after the two latter populations had separated out.
Table 3.4 Population pairwise FST values between the Arunachal macaque populations. (A) Pairwise values for mitochondrial DNA, (B) Pairwise values for nuclear markers. A
Pairwise FST Tawang Upper Subansiri West Siang
Tawang
Upper Subansiri 0.70 (P << 0.05)
West Siang 0.62 (P << 0.05) 0.62 (P << 0.05)
B
Pairwise FST Tawang-West Siang Upper Subansiri
Tawang-West Siang
Upper Subansiri 0.03 (P = 0.04)
The Bayesian clustering analysis of the microsatellite data (Table 3.3) shows a significant FST
value (P = 0.04) only when the Tawang – West Siang group is compared to the Upper Subansiri
group (population pairwise FST = 0.03; Table 3.4B). This result, therefore, clearly establishes, on
the one hand, enhanced levels of gene flow between the sampling locations than that revealed by
the mtDNA analysis, possibly as a result of male migration (Chikhi and Bruford 2005). On the
other hand, it supports the continued relative isolation of the Upper Subansiri group from those
of the other two locations notwithstanding the apparently enhanced levels of gene flow between
them.
3.3.2 The bonnet macaque
3.3.2.1 Phylogenetic reconstruction
We generated a phylogenetic tree for the mitochondrial (mtDNA) haplotypes from 98
individuals (Fig. 3.6) distributed across 22 sampling sites in peninsular India (Fig. 3.2). The tree
shows a monophyletic bonnet macaque clade with a median coalescent date of 1.73 mya (95%
Highest Posterior Density or HPD 0.56 – 5.39 mya) agreeing broadly with the estimate from
Chapter 3. The phylogenetic tree further illustrates the presence of two broad haplogroups.
Haplogroup A consists of individuals from Regions 1 and 2 while Haplogroup B consists mostly
of individuals from Region 3. It should be noted that a few individuals, sampled from Region 1
(an adult female, Surat population; an individual of unknown age and sex, Maharashtra 2
population) and Region 2 [an adult female, Karnataka 3 population; an adult male, Karnataka 4
population) showed phylogenetic allegiance with the Haplogroup B. Similarly, six individuals
from Region 3 (two adult females, Kerala 2 population; three individuals of unknown age and
sex, Nilgiri Hills population; an individual of unknown age and sex, Kodaikanal population)
appeared to be phylogenetically related to Haplogroup A. The coalescent node between these
two haplogroups, however, has a very low posterior probability value (< 0.5), clearly indicating a
rather weak divergence between the groups. Furthermore, the two subspecies of the bonnet
macaque, as currently recognised on the basis of morphological characteristics (reviewed in Sinha
2001), were not observed to be reciprocally monophyletic and should, therefore, be
reconsidered.
Figure 3.6 Bayesian phylogenetic tree of the bonnet macaque mtDNA haplotypes. The posterior probabilities are from 0.5 to 0.75 (open circles) or greater than 0.75 (filled circles). The geographical origin of the individuals is indicated next to the branch tips: Region 1 (red), Region 2 (yellow) and Region 3 (blue).
3.3.2.2 Genetic differentiation between regions
Previous studies have suggested that mountains can impede mammalian gene flow, consequently
increasing genetic differentiation between populations separated by these ranges (Pérez-Espona
et al. 2008; Zalewski et al. 2009). The effectiveness of mountains as a physical barrier to primate
dispersal is, however, not well known. The high-elevation, wet montane forests of the Western
Ghats in southwestern India could potentially act as a barrier to bonnet macaque dispersal as
populations are not usually known from these high-elevation areas (Sinha 2001). To investigate
this hypothesis, we examined genetic differentiation between the three sampling regions
(Regions 1, 2 and 3), described in the Methods, using mitochondrial and microsatellite data.
The results of the AMOVA for both the markers clearly ruled out any evidence of strong
population differentiation (Table 3.5). A mere 0.06% of the total genetic variation in mtDNA
was partitioned among the study regions than within each of the 22 sampled populations
(94.89%). As a result, the fixation index (ΦST = 0.05, p << 0.05) was low but nevertheless
statistically significant. The microsatellite data indicated a relatively higher level of genetic
differentiation. An immense 92.48% of the total genetic variation was observed to be partitioned
within each of the 24 sampled populations in comparison to a mere 0.53% between the three
regions (Table 3.5). Consequently, the fixation index was low (FST = 0.08, p << 0.05), suggesting
that 8% of the observed molecular variance is accounted for by genetic differences among the
three tentative geographical regions; these regions, therefore, appear to be more genetically
similar to one another than different, for both sets of markers.
Table 3.5 Analysis of molecular variance (AMOVA) of the bonnet macaque populations showing weak structure for both mitochondrial DNA and nuclear DNA. Both the fixation indices are significant (P << 0.05)
Source of Variation Percentage of Variation
(mtDNA)
Percentage of Variation
(microsatellites)
Among regions 0.06 0.53
Among populations within regions 5.18 6.99
Within populations 94.89 92.48
FST 0.05 0.08
3.3.2.3 Population genetic structure
Model-based Bayesian clustering analysis with microsatellite DNA also failed to yield any
population structure, providing further support to a lack of genetic differentiation in the species,
at least at the regional-level (Fig. 3.7). The Locprior model, however, predicted a putative two-
population (K = 2) genetic structure, though the signal continued to be weak (Log likelihood,
LnP(K) = −3939.83). The closest alternative was single ancestry for the sampled individuals
(LnP(K) = −3969.99). Although two clusters were identified by the Locprior model, many (70)
individuals showed only a weak genetic assignment (Q = 0.2 – 0.8); 20 individuals were strongly
assigned (Q > 0.8) to Cluster 1 and eight to Cluster 2.
Figure 3.7 Genetic structure of the bonnet macaque populations based on nuclear (microsatellite) markers. We used Locprior model (Structure 2.3.3) to infer the population structure.
The majority of the individuals that exhibited strong assignment to Cluster 1 (henceforth
referred to as the Eastern Group, n = 20) belonged to the eastern, hot and dry areas of the
Deccan Plateau with the exception of a few individuals that were sampled from the more
southern, colder Nilgiri and Kodaikanal Hills (Fig. 3.8). The individuals that were strongly
assigned to Cluster 2 (henceforth, Northern Group, n = 8) were mostly from the northern low-
elevation stretches of the Western Ghats (more undulating, less warmer and more humid
regions) and the adjacent northward transition zone of the Deccan Plateau. The weakly assigned
animals in this cluster hailed from across the peninsula, suggesting a considerable degree of
admixture between the two groups (henceforth considered as loosely belonging to a Mixed
Group, n = 70; Fig. 3.8).
Figure 3.8 Sampling locations of bonnet macaque individuals that were genetically assigned, based on microsatellites, to the Northern (triangles), Eastern (circles) and Mixed (crosses) Groups (see text for details). Note that each sampling site could include more than one individual belonging to the same or different groups. The Western Ghats region has been shown in grey.
The genetic differentiation between the Northern and Eastern Groups, as measured by
microsatellite-based FST, showed that 18% of the molecular variance was explained by variation
among the groups (FST = 0.18, P = 0.001). The mtDNA- based differentiation between these
two groups was also highly significant at ΦST = 0.08 (P < 0.05) (Table 3.6). This is a considerable
increase in population genetic structure from the earlier hypothetical structure based on the three
regions hypothesised above (Table 3.2). Note that the FST (derived from biparentally inherited
microsatellites) continues to be greater than the corresponding ΦST (derived from maternally
inherited mtDNA). This is an extremely unusual finding, as bonnet macaques are generally
known to be a typically female-bonded cercopithecine species in which females are philopatric
and usually stay back in their natal troops all their lives (reviewed in Sinha 2001, but see Sinha et
al. 2005 and Mukhopadhyay and Sinha 2010). Attention must also be drawn to the fact that the
subsequent analysis led to a considerable increase in FST values (from 0.08 to 0.18; Table 3.6),
indicating relatively weaker male migration between the Northern and Eastern Groups as
compared to that between the regions hypothesised earlier.
Table 3.6 Nuclear microsatellite FST and mitochondrial ΦST values for the bonnet macaque populations. The regions and groups have been explained in the text.
FST ΦST
Between the three regions# 0.08** 0.05**
Between the two groups† 0.18** 0.08*
** P < 0.001, * P < 0.05
3.3.4 Genetic variability and differentiation
We further analysed the sampled bonnet macaque population data to examine differences in
genetic variability and relative patterns of differentiation only between individuals that belonged
to either the Northern or the Eastern Group while reducing „noise‟ due to admixture (after Sacks
et al. 2004 and Ansmann et al. 2011). Patterns of gene flow were also investigated between the
three groups, with individuals of admixed genetic background forming the third Mixed Group,
as described above.
A pairwise comparison of the FST values revealed significant genetic differentiation between all
the pairs of groups, with the FST values being highest between the Northern and Eastern Groups
(FST = 0.12; Table 3.7), in agreement with the results of the earlier Bayesian cluster analysis.
Pairwise comparisons of the mitochondrial DNA data found ΦST values that were generally
lower than their corresponding FST values and significant only between the Eastern and Mixed
Groups (ΦST = 0.02; Table 3.7). Again, a comparison of these values suggest relatively greater
gene flow through females between the groups than through the males, as may be expected for a
cercopithecine primate species.
Table 3.7 Pairwise nuclear microsatellite FST (above diagonal) and mitochondrial ΦST values (below diagonal) between the bonnet macaque groups.
Group Northern Eastern Mixed
Northern 0.18** 0.04*
Eastern 0.02 0.03**
Mixed 0.002 0.02**
** P < 0.001, * P < 0.05
Analysis of the microsatellite-based genetic variability indicated that individuals of the Northern
Group had the lowest mean number of alleles across loci (NA = 3.9) and allelic range (15.4), as
compared to the Eastern (NA = 10.3, allelic range = 43.6) and Mixed (NA = 12.8, allelic range
= 45.8) Groups, which were more comparable to one another. None of the loci showed any
significant difference between the observed (Ho) and unbiased expected (He) heterozygosity
values, the population average values being shown in Table 3.8. The FIS values in both the
Northern and the Mixed Groups were not significant, thus ruling out any significant inbreeding
in these groups. In the Eastern Group, however, the FIS value was significantly different from
zero, indicating significant levels of inbreeding within this group (Table 3.8).
Table 3.8 Measures of nuclear genetic diversity within the bonnet macaque groups. N, sample size; NA, mean number of alleles across all loci; Ho, observed heterozygosity; He, unbiased expected heterozygosity; FIS, inbreeding coefficient; ns = not significant; * P = 0.03. Values in parentheses are standard deviations.
Group N NA Allelic range Ho He FIS
Northern 8 3.9 (1.2) 15.4 (6.87) 0.51 (0.26) 0.62 (0.16) 0.08*
Eastern 20 10.3 (2.45) 43.6 (17.35) 0.69 (0.2) 0.89 (0.05) 0.01ns
Mixed 70 12.8 (4.37) 45.8 (20.58) 0.77 (0.11) 0.89 (0.09) 0.01ns
3.4 Discussion
3.4.1 The Arunachal macaque
Like several other species from the Tibetan plateau that have been studied (Yang et al. 2009; Qu
et al. 2010; Fan et al. 2011; Zhao et al. 2012), the Arunachal macaque exhibits significant
population structure across its distribution range. But, unlike most of the other species, which
have been analysed using only mitochondrial DNA, the population structure of this macaque,
when analysed using both mitochondrial and nuclear markers, exhibits effects of multiple forces
that have resulted in complex phylogeographic patterns.
3.4.1.1 Pleistocene glaciations and population divergence
The Tibetan Plateau is recognised to have experienced at least four major glaciation events
during the Quaternary period – the Xixiabangma (Early Pleistocene), Naynayxungla (Middle
Pleistocene), Guxiang (late Middle Pleistocene) and the last glaciation, including two glacial
stages, the latter of which corresponds to the last glacial maximum (LGM), c. 20 thousand years
ago (Shi 2002; Gibbard and Head 2009). Molecular dating analysis suggests that the common
ancestor of the extant populations of Arunachal – bonnet macaques harks back to the beginning
of the Pleistocene period (1.61 mya, 95% HPD 1.14 – 2.12). This date, more or less, overlaps
with „the most recent common ancestor‟ (TMRCA) of other species of birds and mammals from
the Tibetan Plateau (Yang et al. 2009; Qu et al., 2010). This time period is marked, geologically,
as the beginning of the Quaternary glacial period in China and adjacent areas such as the Tibetan
Plateau (Zhan et al. 2011).
The next divergence event followed soon (1.32 mya, 95% HPD 0.87 – 1.81) and appears to have
resulted in the founding of the West Siang population and the ancestors of the present-day
bonnet macaques. This date coincides with another cold stage in the Xixiabangma glaciations
(Zheng et al. 2002); it was one of coldest recent climatic periods in the region, possibly creating
many refugia in the southern and eastern boundaries of the Tibetan Plateau (Zhan et al. 2011). In
the Eastern Himalayas, however, glaciations were restricted to the relatively high altitudes and
did not affect the lower slopes or valleys (Zhou et al. 2006) in these ranges. Palaeo-climatic and
palynological studies from this region reveal a vegetational shift over the Pleistocene glacial
cycles (Kou et al. 2006; Yu et al. 2007); during the glacial periods, cool-temperate vegetation,
such as shrub-lands, expanded to the lower elevations and contracted to the high elevations
during the warmer and wetter interglacial periods. Such an elevation-dependent niche separation
may have been responsible for diversification at the population-level in species that inhabit this
region (Qu et al. 2011). In accordance with such a pattern, we postulate that a lower-elevation
group of macaques, adapted to a grassland habitat, may have found itself isolated from its high-
altitude cousins during these climatic and resultant vegetational shifts, and may have eventually
served as the ancestors of present-day bonnet macaques.
We could date the last intraspecific diversification event to the Middle Pleistocene (0.80 mya,
95% HPD 0.5 – 1.16). Expectedly, this period too recorded another glaciation event, the
Naynayxungla glaciation between 0.78 and 0.5 mya (Zheng et al. 2002). This was considered to
be the most extensive glaciation with many large ice caps, glacier complexes and great valley
glaciers, covering a total area more than 500 thousand km2 across the Tibetan Plateau.
3.4.1.2 Historical factors: Effect of Pleistocene climate change and topography
It is now well known from many recent studies that Quaternary glaciations fragmented large
populations in the Tibetan Plateau to many small refugial relict populations over the long-term,
leading to distinct population structures (Qu et al. 2010, 2011; Zhan et al. 2011; Zhao et al.
2012). The Arunachal macaque populations too show a considerable amount of structure for
mitochondrial DNA with the Tawang, Upper Subansiri and West Siang populations being largely
isolated from one another. An intriguing feature that the macaque populations display, however,
is a moderate to low nucleotide diversity that reduces from the east to the west, unlike the other
species in this region, virtually all of which have fairly low nucleotide diversities (Qu et al. 2010,
2011; Zhan et al. 2011). Such low nucleotide diversity is also characteristic of other macaque
species known to be affected by Pleistocene glaciations across the world (Chikhi and Bruford
2005; Modolo et al. 2005) although rhesus macaques, nevertheless, show a similar gradient in
south Asia (Smith and McDonough 2005). The directional decrease in nucleotide diversity
shown by the Arunachal macaque supports the hypothesis of the origin of the species in the east
of the Indian subcontinent, as was suggested by our phylogenetic analysis in Chapter 2 (see also
Chakraborty et al. 2007). In this scenario, as in the case of the eastward dispersal of modern
humans from Africa (Cann et al. 1987; Vigilant et al. 1991), founder effects would be expected to
lead to reduced genetic heterogeneity in the Tawang group relative to that in West Siang, exactly
as we find in our study. In other words, the ancestors of the Tawang-Upper Subansiri population
may have originated in the Upper Subansiri region approximately 1.6 mya, and later diversified to
give rise to the Tawang population about 0.80 mya.
A second implication of our analysis is that while most of the species that have low nucleotide
diversity are distributed further north in the Tibetan Plateau, sometimes extending into the
northern hemisphere (Qu et al. 2010, 2011; Zhan et al. 2011) the Arunachal macaque is restricted
to the southern edge of the Tibetan Plateau, the southernmost border of the Pleistocene glaciers.
The eastern and southeastern parts of the Tibetan Plateau have long been considered to harbour
several glacial refugia during the Pleistocene (Zhan et al. 2011). It is thus expected that the
Arunachal macaque would be least affected by the Quaternary climatic shifts, as compared to
other more temperate species. The Pleistocene glaciations on the Tibetan Plateau are also known
to be punctuated by four to five interstadials or inter-glaciations (Zheng et al. 2002). We thus
hypothesise that although the ancestral Arunachal macaque populations became periodically
isolated and small in size during glaciations, they may have intermittently increased in size during
the interstadials when the climate became warmer and wetter, and consequently, more conducive
to population expansion. The only other species from the Tibetan Plateau that shows a similar
level of gene diversity is a small passerine bird, the black redstart Phoenicurus ochruros, with an
elevation range from 2000 to 3500 m on the southeast and northwest edges of the plateau (Qu et
al. 2010).
An examination of the pairwise FST values of the Arunachal macaque populations for
mitochondrial DNA also makes evident that among the three studied populations, the Upper
Subansiri population is the most isolated one. Geographically, the Upper Subansiri and West
Siang populations are closer to one another than West Siang to Tawang. The West Siang
population, nevertheless, displays comparative FST values with both the other populations. The
Upper Subansiri population, in turn, reveals a higher FST value with the Tawang population, with
which it shares a common ancestry. These patterns suggest the possible existence of a corridor
between West Siang and Tawang while there may be a physical barrier that segregates the Upper
Subansiri population from the other two and which may have arisen after the Tawang and Upper
Subansiri populations separated from one another, approximately 0.8 mya. The present-day
Arunachal macaque populations are mostly distributed in the high-altitude areas of the Eastern
Himalayas, with lofty ranges separated by deep valleys. Such a rugged topography may make
migration difficult, as is also evident from the deep isolation shown by the local human
populations of the region from human populations of the rest of India (Cordaux et al. 2004).
The typical topography of the region may have thus also served as a barrier for the macaque
populations in recent times.
These patterns are further reinforced by our analysis of the biparental microsatellite markers.
Both AMOVA and Bayesian clustering analyses suggest that the Tawang and West Siang groups
genetically belong to a single population while the Upper Subansiri population appears to be
distinct. Its relatively low pairwise FST value with the other conjoined population, as revealed by
the microsatellite analysis in contrast to that with the mitochondrial DNA, nevertheless, hints at
a low level of genetic connectivity between them.
3.4.1.3 Socio-behavioural factors: Effects of female philopatry and male migration
Despite the geographical proximity of the study populations, there is absolutely no
mitochondrial haplotype shared among them, a pattern typical of other cercopithecine primates
(reviewed in Chikhi and Bruford 2005). This, however, contrasts with the situation presented by
patrilocal chimpanzees and bonobos (Eriksson et al. 2006; Langergraber et al. 2007) in which
mtDNA sharing is extensive, such gene flow being promoted by dispersing females typical of the
species. It is noteworthy that the Bayesian cluster analysis of the mtDNA revealed the presence
of five Arunachal macaque individuals, two sampled from Upper Subansiri and three from West
Siang, but which were genetically dissimilar from their parent populations. While four of these
individuals, including a female, were assigned to an unknown population, which was not
sampled, the fifth individual, a male, was a potential migrant from Upper Subansiri to West
Siang. The geographic fidelity of the maternally inherited mitochondrial DNA, the majority of
the potential migrants being male, and the high nuclear DNA nucleotide diversity accompanied
by a reduction in population structure for nuclear markers suggest female philopatry and male-
mediated gene flow in this species, features again characteristic of cercopithecine primates in
general (Chikhi and Bruford 2005). Along with these four lines of evidence, the relatively higher
FST values for the mtDNA data than that reflected in the microsatellite analysis further confirms
the importance of the typical cercopithecine social structure in determining the patterns of
genetic diversity in the study species. The typical female philopatry observed in cercopithecine
primates including the Arunachal macaque is expected to generate higher FST values in mtDNA
because firstly, mtDNA has a lower effective population size, which increases the chances of
random fixation of alleles and secondly, female philopatry effectively reduces mtDNA gene flow
between populations (Chikhi and Bruford 2005).
3.4.2 The bonnet macaque
3.4.2.1 Region-level population genetic structure
The lack of a physical barrier and high dispersal ability can be expected to give rise to low levels
of population structure in generalist species (Bohonak 1999). The Deccan Plateau of peninsular
India is a vast plain that does not seem to offer much physical impediment to the movement of a
generalist species like the bonnet macaque, which is also known to be extremely adaptable to
human presence (Sinha 2001 and references therein). Given the known distribution of bonnet
macaques in a great variety of ecological habitats, it was expected a priori that the species would
lack any significant population genetic structure. Commensurate with this expectation, the
phylogenetic tree of the bonnet macaque individuals exemplifies a lack of strong geographic
fidelity across sampling locations. The two resultant haplogroups, A and B, appear to have
weakly diverged from their most recent common ancestor, as evidenced by the poor posterior
probability that characterises their phylogenetic node. Nonetheless, the geographical origins of
the two haplogroups roughly suggest the presence of the Western Ghats as a potential divide,
albeit a weak one.
Subsequent analysis of population differentiation on the basis of both mitochondrial and nuclear
DNA, however, failed to provide any significant evidence for the Western Ghats having played a
role in impeding dispersal of the species across its range. For both sets of markers, the greatest
amount of variation was partitioned within the sampled populations not among the three pre-
defined regions. This lack of any considerable structure, which was further supported by the
subsequent Bayesian cluster analysis, may be due not only to the generalist life-history traits of
the species that allow it to adapt to different habitats but also because the Western Ghats is not a
continuous mountain range. The presence of gaps like Shencottah, Palghat and Goa may have
facilitated gene flow across the Ghats thus obliterating any differentiation, however weak, that
may have been established when the ancestors of the species originally colonised the peninsula.
Thus, the bonnet macaque populations of today appear not to be genetically differentiated at the
large regional-scale and the Western Ghats does not seem to be a significant physical barrier to
dispersal by members of this ecologically adaptable primate species.
3.4.2.2 Fine-scale population genetic structure
Population genetic structure at a fine level is likely to be ubiquitous across mammalian species;
even marine mammals with their high dispersal ability and apparent lack of any barrier in their
habitats display population structure at a very fine scale (Ansmann et al. 2011). Terrestrial
primates, however mobile they may be, may be expected to have lesser capacities of dispersal
than do marine mammals. Hence, a fine-scale genetic structure could be expected even when
there does not seem to be any at a larger scale. Most of the previous studies on phylogeography
of primates have typically explained the effects of landscape on gene flow in terms of dispersal
barriers (Eriksson et al. 2004; Arora et al. 2011). Beyond this generality, little is known about the
relationship between movements of individual animals and the emergent genetic structure of
populations over different scales as there have been very few studies on generalist species until
very recently (Ansmann et al. 2012; Kobblemüller et al. 2012). Although such populations might
be expected to exhibit simple patterns of genetic isolation-by-distance (Slatkin 1993), it is also
possible that more complex patterns of genetic structure arise due to heterogeneous patterns of
habitat selection by individuals. For example, Sacks et al. (2004) showed that the population
genetic structure of the coyote, a continuously distributed generalist species in North America,
corresponded to habitat-specific breaks, a pattern expected if coyotes tended to disperse
preferentially to habitat similar to their natal ones.
In our study bonnet macaque populations too, we have discovered a fine-scale population
genetic structure. While most of the sampled individuals appear to be of admixed origin, we
found a significant number of individuals that were more clearly differentiated and belonged to
two ancestral groups, generally distributed across the Indian peninsula. The members of the
Northern Group largely belong to the north Western Ghats moist deciduous forests ecoregion
(Champion and Seth 1968), which represents a swath of lowland moist deciduous forests around
the montane rainforests in the northern section of the Western Ghats Mountains. The vegetation
is influenced by the south-western monsoons and consists of moist, teak-bearing forests, moist
mixed deciduous forest without teak, and secondary moist mixed deciduous forests. Patches of
evergreen forests extend into these moist deciduous forests, with many rainforest species
typically occurring in them (Champion and Seth 1968).
The Eastern Group falls within the south Deccan Plateau dry deciduous forests ecoregion with a
few individuals being found in the Deccan thorn scrub forests ecoregion. Both these areas
receive very low annual rainfall and are extremely dry in nature. The dry deciduous forests of this
region are flanked by the moist deciduous forests along the lower elevations and foothills of the
Western Ghats to the west and by thorn scrub to the east. Therefore, this ecoregion probably
represents a transition zone between the moister western vegetation and the drier vegetation to
the east (Champion and Seth 1968). It is well known that the Indian subcontinent experienced
major environmental fluctuations as a result of Pleistocene climate change even when it was not
under glaciation. One of the major factors in this historical climate change was the fluctuation in
monsoon strength. The change is monsoon patterns is believed to have affected the demography
of many species in peninsular India, including humans (Petraglia et al. 2012). Since the last
interglaciation, the two major fluctuations in the monsoons that altered vegetation patterns in
peninsular India occurred during the marine isotope stages (MIS) 6 and 4, approximately 130 and
70 thousand years ago, respectively (Petraglia et al. 2012). Both these dry spells transformed the
then wet tropical forests into dry scrublands across the peninsula. Such a major change in habitat
conditions can potentially profoundly influence the distribution of species, acting mainly through
population fragmentation. A subsequent secondary contact during later, more conducive,
environments, however, may give rise to a continuous distribution, often observed in more
recent times.
3.4.2.3 Mito-nuclear discordance and introgression in bonnet macaque populations
We sampled bonnet macaque populations across the entire distribution range of the species in
peninsular India. Despite the geographical distance between them, many of the mtDNA
haplotypes are shared across populations. Such a pattern of mtDNA haplotype sharing may be
expected because the mitochondrial genome is haploid and uniparentally inherited in most
animals, and therefore, has a four-fold smaller effective population size than does the
biparentally inherited, diploid nuclear markers (Hudson and Turelli 2003; Zink and
Barrowclough 2008). This implies that mtDNA will complete the process of lineage sorting,
where ancestral polymorphisms are lost over time, faster than will nuclear markers, as this rate is
inversely proportional to the effective population size (Funk and Omland 2003). As a result,
there can be mismatch between the sorting of these two kinds of markers by chance alone;
mtDNA may thus show phylogeographic patterns that do not agree with the divergence patterns
obtained using nuclear markers. This is a case of mito-nuclear discordance with incomplete
lineage sorting as its underlying mechanism. Mito-nuclear discordance may also, however, arise
due to sex-biased dispersal (Rheindt and Edwards 2011) with female dispersal promoting the
dispersal of mtDNA even in the absence of concordant movement of nuclear DNA (Funk and
Omland 2003).
Distinguishing between the different mechanisms underlying discordance can often be a difficult
exercise (McKay and Zink 2010). One important distinction, however, is that discordance that
arises from incomplete lineage sorting is not expected to leave any predictable biogeographic
pattern as it occurs by chance (Funk and Omland 2003). In the bonnet macaque populations that
we studied, the Northern Group was observed to include individuals of Haplogroup A alone.
The majority (85%) of the individuals of the Eastern Group also belong to this haplogroup. The
Mixed Group, in contrast, consists of individuals from both the haplogroups, thus showing a
clear biogeographic pattern in the mito-nuclear discordance. In such cases, where there are
strong geographic inconsistencies between patterns in mtDNA and nuclear DNA, incomplete
lineage sorting can usually be ruled out (Toews and Brelsford 2012).
This type of discordance, referred to, more generally, as biogeographic discordance, can result
from clines in mtDNA being displaced from nuclear DNA in both their locations and/or their
width. In bonnet macaques, we hypothesise that the Haplogroup A may have swept over the
majority of the erstwhile distribution of Haplogroup B, which now exists only in the
southwestern corner of the Western Ghats. This kind of biogeographic discordance, in which
one group extensively replaces the mtDNA of the other group, is also called „mitochondrial
capture‟ (Good et al. 2008). According to Toews and Brelsford (2012), two general situations can
lead to biogeographic discordance between mtDNA and nuclear DNA: in situ or following
isolation and hybridisation/introgression. In other words, the discordance is mediated by
primary or secondary contact, respectively. Most of the taxa that display such patterns are usually
groups that were isolated for long periods of time and are either currently in secondary contact
or have experienced contact at some point in their recent past (Toews and Brelsford 2012).
During those periods of isolation, it is assumed that divergent groups accumulated mutations in
both their mitochondrial and nuclear genomes, which increased to high frequency via selection,
drift or some combination of the two (Hudson and Turelli 2003). Upon secondary contact, these
groups formed hybrid or introgression zones, interbreeding to varying extents, and mito-nuclear
discordance was promoted by divergent patterns of gene flow, including sex-biased dispersal,
between the two genomes.
3.4.2.4 Patterns of gene flow and female-mediated dispersal
Geographic fidelity of mtDNA and a reduced population structure for nuclear DNA suggest
female philopatry and male-mediated gene flow in cercopithecine primates (Chikhi and Bruford
2005); this is in contrast to patrilocal chimpanzees, bonobos and humans (Eriksson et al. 2006;
Langergraber et al. 2007) wherein males are philopatric while females mediate gene flow. Being a
cercopithecine primate, the bonnet macaque too is expected to show female philopatry and
male-biased gene flow. But the “mitochondrial capture” that we witness across populations
suggests that such a pattern is possible only if a considerable amount of gene flow is mediated by
females as they are the only sex in primates that can transfer mtDNA to the next generation.
This fact is further supported by the population pairwise FST values that are significant and
higher for nuclear markers than for mtDNA. Female dispersal is also evident in the mtDNA
phylogenetic tree where a few known females, sampled from Regions 1 and 2 exhibited
phylogenetic allegiance with the Haplogroup B. Similarly, two females from Region 3 appear to
be phylogenetically related to Haplogroup A. Thus, bonnet macaque seems to represent a rather
unusual case of a female-bonded cercopithecine species in which females are also responsible for
gene flow. In the absence of male-specific genetic markers such as the Y-chromosome,
however, we cannot investigate the degree to which male-mediated gene flow compares to gene
flow mediated exclusively by females.
It must be pointed out here that, over the years, behavioural studies on wild bonnet macaques
have increasingly accumulated evidence for female movement in this species. Ali (1981), for
example, first reported two events of female movement in the southern, pale-bellied subspecies
of the bonnet macaque (Macaca radiata diluta) in the Kalakad-Mundanthurai Tiger Reserve and
extensively discussed the possibility that a quest for greater mate choice may have driven such
movement. Singh et al. (2006) also reported a case of female transfer into a particular troop of
the northern subspecies (M. r. radiata) in the Indira Gandhi Wildlife Sanctuary of the Annamalai
Hills of Tamil Nadu state. More systematic dispersal of adult and juvenile female bonnet
macaques, either singly or in groups, have been reported by Sinha and his colleagues in at least
two study populations, in the GKVK Campus of the University of Agricultural Sciences close to
Bangalore (Subramaniam and Sinha 1999) and more extensively in the Bandipur National Park –
Mudumalai Wildlife Sanctuary in Karnataka and Tamil Nadu (Sinha et al. 2003, 2005;
Mukhopadhyay and Sinha 2010), respectively. In the latter population consisting typically of
multimale-multifemale social groups, certain ecological factors appear to have led, in recent
years, to increasing competition among the resident females within these groups, causing small
groups of females to emigrate out and form small, stable associations, which are then taken over
by a single male to give rise to a novel form of social organisation, characterised by unimale-
multifemale troops. Such troops have also been reported from other populations, though at
lower frequencies (Sinha et al. 2003). There is now clear evidence that the females in these
groups suffer from increased aggression from the resident male and also have limited mate-
choice opportunities (Sinha et al. 2005; Chatterjee, submitted). This also seems to drive the
movement of females out of these unimale groups and a marked preference for them to join
multimale troops (Sinha et al. 2005; Mukhopadhyay and Sinha 2010). In summary, intra-sexual
competition for resources across and within social groups, and better mating and mate choice
opportunities seem to constitute the principal motivations driving extensive female movement in
this species. Taken together, there is now convincing on-ground evidence that female dispersal
may be widespread in different populations of bonnet macaques across their distribution range.
It has also been suggested that habitual female dispersal may occurs in species where the
breeding tenure of individual males or of all male kin commonly exceeds the age at which most
females are ready to breed, causing females to leave their natal groups to avoid the risk of
inbreeding with their close relatives and to locate unrelated mating partners (Berger 1986;
Clutton-Brock 1989; Harcourt and Stewart 2007). The species, which seem to support this cause
are usually long-lived and the average tenure of breeding males commonly exceeds the average
age of females at first breeding (Clutton-Brock 1989; Clutton-Brock and Lukas 2012).
Interestingly, in some of the studied groups of bonnet macaques, several males have been found
to stay back in their natal groups, sometime even gaining the position of the most dominant male
(Sinha 2001, pers. comm.). Consequently, we would expect females to disperse from those
groups as otherwise, inbreeding could ensue. If this were a widespread phenomenon, we would
further expect inbreeding to be significant at the population-level in the absence of considerable
female migration. Intriguingly, we found the inbreeding coefficient in both the Northern and
Mixed Groups not to be significantly high. The Eastern Group showed a significantly high value
but that could be due to the skew in the number of individuals sampled at some locations. It
should be noted, however, that there can still be alternative mechanisms wherein females and
males may both be philopatric and yet avoid inbreeding. Hence, avoidance of inbreeding may
not completely explain the incentive behind female dispersal. For example, in several species,
such as capuchin monkeys, killer whales and banded mongooses, females commonly remain in
their natal group despite the presence of their father or brothers and either breed with more
distantly resident males or with males from other groups (Baird 2000; Perry et al. 2008; Nichols
et al. 2010).
Finally, it may be important to explore to what extent the relative importance of different
ecological and evolutionary mechanisms that affect dispersal and philopatry depend on the scale
of analysis. For example, while the avoidance of close inbreeding may play an important role in
promoting dispersal from the natal group, there is little evidence that it has important effects on
dispersal distance or on the frequency of dispersal between sub-populations or demes, which is
more commonly affected by ecological parameters (Clutton-Brock and Lukas 2012). Likewise,
we could hypothesise that, in bonnet macaques, the species-level effect of female gene flow has
been brought about by the cumulative effects of much smaller-scale, yet significantly widespread,
female dispersal, possibly driven by a combination of ecological factors, such as seasonality in
the abundance and distribution of resources, and social drivers such as intra-group agonistic
relationships or mate choice.
3.5 Conclusions
The distribution patterns and population structure of the Arunachal macaque appear to have
been uniquely influenced by the location of the species on the southern edge of the Tibetan
Plateau, the extremely rugged topography of the region, Pleistocene climatic oscillations and by
the socio-behavioural patterns of male dispersal and female philopatry, typical of cercopithecine
primates. Our results also support an early to middle Pleistocene radiation of the species to the
southern edge of the Tibetan Plateau, characterised by groups of females strongly separated by
harsh climate, geographical barriers and phylogenetically constrained socio-behavioural patterns.
These influences led to highly differentiated populations of the species but with dispersing males
exerting a homogenising effect on the nuclear gene pool. There is, however, a need to
independently analyse the demographic history of the populations to test if there were indeed
changes in historical population size correlated with Pleistocene climate change, as postulated
above. Our analysis of the demographic history of the study populations of the Arunachal
macaque is described in the following Chapter 4. Independent of such investigations, however,
there is also an urgent need to sample the species more intensively in order to determine precise
locations of the corridors and barriers between the study populations, and to detect other, yet
unknown, populations of the species, all of which have been predicted by the genetic analyses
described in this chapter. Such discoveries are likely to be of utmost importance while planning
the management of this endangered montane primate in the years to come.
On the other hand, from our analyses of the bonnet macaque data, we found scale to be an
important factor to interpret the population genetic structure and especially for a generalist
species like bonnet macaque with an extensive, continuous distribution. To begin with, while
both historical climate change and landscape features may not be important at one geographic
scale, surprisingly, these same factors may become critically important at a different, larger scale.
Similarly, while social organisation and behaviour may provide crucial insights into genetically
driven group structure and dynamics within particular populations or demes, these effects may
either not be apparent or may be overridden by other factors at a much higher level of
organisation. More specifically, we made a surprising discovery of a fine-scale population genetic
structure in bonnet macaques, driven considerably by female-mediated gene flow, a notable
exception in typical cercopithecine biology. The population genetic structure of this species also
appears to have been shaped, over the long-term, by the phenomenon of mito-nuclear
discordance, possibly brought about by secondary contact between two ancestrally separated
groups. If this is indeed true, we need to explore the demographic history of the species over its
distribution range in an effort to better understand the origin of the two ancestral populations
and the factors that may have mediated their subsequent contact.
3.6 References
Ali, R. (1981). The ecology and social behaviour of the Agastyamalai bonnet macaque. Ph. D. thesis, Bristol University, Bristol, UK.
Altmann, J., Alberts, S. C., Haines, S. A., Dubach, J., Muruthi, P., Coote, T., Geffen, E., Cheesman, D. J., Mututua, R. S., & Saiyalel, S. N. (1996). Behavior predicts genes structure in a wild primate group. Proceedings of the National Academy of Sciences, 93(12), 5797–5801.
Ansmann, I. C., Parra, G. J., Lanyon, J. M., & Seddon, J. M. (2012). Fine‐scale genetic population structure in a mobile marine mammal: inshore bottlenose dolphins in Moreton Bay, Australia. Molecular Ecology, 21(18), 4472–4485.
Anthony, N. M., Johnson-Bawe, M., Jeffery, K., Clifford, S. L., Abernethy, K. A., Tutin, C. E., Lahm, S. A., White, L. J. T., Utley, J. F., Wickings, E. J. (2007). The role of Pleistocene refugia and rivers in shaping gorilla genetic diversity in central Africa. Proceedings of the National Academy of Sciences, 104(51), 20432–20436.
Arora, N., Nater, A., Van Schaik, C. P., Willems, E. P., Van Noordwijk, M. A., Goossens, B., Morf, N., Bastian, M., Knott, C., & Morrogh-Bernard, H. (2010). Effects of Pleistocene glaciations and rivers on the population structure of Bornean orangutans (Pongo pygmaeus). Proceedings of the National Academy of Sciences, 107(50), 21376–21381.
Baird, R. W. (2000). The killer whale - foraging specializations and group hunting. In J. Mann, R.C. Connor, P. L. Tyack, H. Whitehead (Eds.), Cetacean societies: field studies of dolphins and whales. (pp. 127–153). University of Chicago Press.
Balkenhol, N., Gugerli, F., Cushman, S. A., Waits, L. P., Coulon, A., Arntzen, J. W., Holderegger, R., Wagner H. H. (2009). Identifying future research needs in landscape genetics: where to from here? Landscape Ecology, 24(4), 455–463.
Barnett, T. P., Adam, J. C., & Lettenmaier, D. P. (2005). Potential impacts of a warming climate on water availability in snow-dominated regions. Nature, 438(7066), 303–309.
Belay, G., & Mori, A. (2006). Intraspecific phylogeographic mitochondrial DNA (D-loop) variation of Gelada baboon, Theropithecus gelada, in Ethiopia. Biochemical Systematics and Ecology, 34(7), 554–561.
Berger, J. (1986). Wild Horses of the Great Basin: Social Competiton and Population Size. University of Chicago Press.
Bohonak, A. J. (1999). Dispersal, gene flow, and population structure. Quarterly Review of Biology, 74, 21–45.
Cann, R. L., Stoneking, M., & Wilson, A. C. (1987). Mitochondrial DNA and human evolution. Nature, 325(6099), 31–36.
Chakraborty, D., Ramakrishnan, U., Panor, J., Mishra, C., & Sinha, A. (2007). Phylogenetic relationships and morphometric affinities of the Arunachal macaque Macaca munzala, a newly described primate from Arunachal Pradesh, northeastern India. Molecular Phylogenetics and Evolution, 44(2), 838–849.
Champion, H. G., & Seth, S. K. (1968). A revised survey of the forest types of India (Vol. 1). Manager of Publications New Delhi, India.
Chikhi, L., & Bruford, M. (2005). 21 Mammalian Population Genetics and Genomics. In A. R. and J. M. Graves (Ed.), Mammalian Genomics (pp. 539–584). CAB International.
Cincotta, R. P., Wisnewski, J., & Engelman, R. (2000). Human population in the biodiversity hotspots. Nature, 404(6781), 990–992.
Clutton-Brock, T. H. (1989). Female transfer and inbreeding avoidance in social mammals. Nature, 337, 70–72.
Clutton‐Brock, T. H., & Lukas, D. (2011). The evolution of social philopatry and dispersal in female mammals. Molecular Ecology, 21(3), 472–492.
Corander, J., Marttinen, P., Sirén, J., & Tang, J. (2008). Enhanced Bayesian modelling in BAPS software for learning genetic structures of populations. BMC Bioinformatics, 9(1), 539.
Cordaux, R., Weiss, G., Saha, N., & Stoneking, M. (2004). The northeast Indian passageway: a barrier or corridor for human migrations? Molecular Biology and Evolution, 21(8), 1525–1533.
Drummond, A. J., Ho, S. Y. W., Phillips, M. J., & Rambaut, A. (2006). Relaxed phylogenetics and dating with confidence. PLoS biology, 4(5), e88.
Drummond, A. J., Suchard, M. A., Xie, D., & Rambaut, A. (2012). Bayesian phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and Evolution, 29(8), 1969–1973.
Earl, D. A. (2011). Structure harvester V 0.6.1. Retrieved August 5, 2011, from http://taylor0.biology.ucla.edu/structureHarvester/
Endicott, P., Ho, S. Y. W., Metspalu, M., & Stringer, C. (2009). Evaluating the mitochondrial timescale of human evolution. Trends in Ecology and Evolution, 24(9), 515–521.
Eriksson, J., Hohmann, G., Boesch, C., & Vigilant, L. (2004). Rivers influence the population genetic structure of bonobos (Pan paniscus). Molecular Ecology, 13(11), 3425–3435.
Eriksson, J., Siedel, H., Lukas, D., Kayser, M., Erler, A., Hashimoto, C., Hohmann, G., Boesch,
C., Vigilant, L. (2006). Y‐chromosome analysis confirms highly sex‐biased dispersal and suggests a low male effective population size in bonobos (Pan paniscus). Molecular Ecology, 15(4), 939–949.
Excoffier, L., Laval, G., & Schneider, S. (2005). Arlequin (version 3.0): an integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online, 1, 47.
Excoffier, L., Smouse, P. E., & Quattro, J. M. (1992). Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics, 131(2), 479–491.
Falush, D., Stephens, M., & Pritchard, J. K. (2003). Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies. Genetics, 164(4), 1567–1587.
Falush, D., Stephens, M., & Pritchard, J. K. (2007). Inference of population structure using multilocus genotype data: dominant markers and null alleles. Molecular Ecology Notes, 7(4), 574–578.
Fan, Z., Liu, S., Liu, Y., Zhang, X., & Yue, B. (2011). How Quaternary geologic and climatic events in the southeastern margin of the Tibetan Plateau influence the genetic structure of small mammals: inferences from phylogeography of two rodents, Neodon irene and Apodemus latronum. Genetica, 139(3), 339–51.
François, O., & Durand, E. (2010). Spatially explicit Bayesian clustering models in population genetics. Molecular Ecology Resources, 10(5), 773–784.
Funk, D. J., & Omland, K. E. (2003). Species-level paraphyly and polyphyly: frequency, causes, and consequences, with insights from animal mitochondrial DNA. Annual Review of Ecology, Evolution, and Systematics, 34, 397–423.
Gibbard, P., & Head, M. J. (2009). IUGS ratification of the Quaternary System/Period and the Pleistocene Series/Epoch with a base at 2.58 Ma. Quaternaire (Paris), 19(4), 411.
Gillson, A. N. (2004). Biodiversity assesment in the North Bank landscape, N. E. India: A preliminary survey. World Wildlife Fund for Nature, India.
Gunnell, Y., Gallagher, K., Carter, A., Widdowson, M., & Hurford, A. J. (2003). Denudation history of the continental margin of western peninsular India since the early Mesozoic–reconciling apatite fission-track data with geomorphology. Earth and Planetary Science Letters, 215(1), 187–201.
Harcourt, A. H., & Stewart, K. J. (2007). Gorilla society: conflict, compromise, and cooperation between the sexes. University of Chicago Press.
Hartl, D. L., & Clark, A. G. (1997). Principles of population genetics (Vol. 116). Sunderland, MA: Sinauer Associates.
Hewitt, G. (2000). The genetic legacy of the Quaternary ice ages. Nature, 405(6789), 907–913.
Hewitt, G. M. (2004). Genetic consequences of climatic oscillations in the Quaternary. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences, 359(1442), 183–195.
Ho, S. Y. W., Phillips, M. J., Cooper, A., & Drummond, A. J. (2005). Time dependency of molecular rate estimates and systematic overestimation of recent divergence times. Molecular Biology and Evolution, 22(7), 1561–1568.
Ho, S. Y. W., Saarma, U., Barnett, R., Haile, J., & Shapiro, B. (2008). The effect of inappropriate calibration: three case studies in molecular ecology. PLoS One, 3(2), e1615.
Hofreiter, M., & Stewart, J. (2009). Ecological change, range fluctuations and population dynamics during the Pleistocene. Current Biology, 19(14), R584–R594.
Holsinger, K. E., & Mason-Gamer, R. J. (1996). Hierarchical analysis of nucleotide diversity in geographically structured populations. Genetics, 142(2), 629–639.
Hubisz, M. J., Falush, D., Stephens, M., & Pritchard, J. K. (2009). Inferring weak population structure with the assistance of sample group information. Molecular Ecology Resources, 9(5), 1322–1332.
Hudson, R. R., & Turelli, M. (2003). Stochasticity overrules the “three‐times rule”: genetic drift, genetic draft, and coalescence times for nuclear loci versus mitochondrial DNA. Evolution, 57(1), 182–190.
IUCN. (2012). IUCN Red List of Threatened Species. Retrieved September 23, 2012, from www.iucnredlist.org
Kanthaswamy, S., Von Dollen, A., Kurushima, J. D., Alminas, O., Rogers, J., Ferguson, B., Lerche, N. W., Allen, P. C., & Smith, D. G. (2006). Microsatellite markers for standardized genetic management of captive colonies of rhesus macaques (Macaca mulatta). American Journal of Primatology, 68(1), 73–95.
Koblmüller, S., Wayne, R. K., & Leonard, J. A. (2012). Impact of Quaternary climatic changes and interspecific competition on the demographic history of a highly mobile generalist carnivore, the coyote. Biology Letters, 8(4), 644–647.
Kou, X. Y., Ferguson, D. K., Xu, J. X., Wang, Y. F., & Li, C. S. (2006). The reconstruction of paleovegetation and paleoclimate in the Late Pliocene of West Yunnan, China. Climatic Change, 77(3), 431–448.
Krishnan, M. (1972). An ecological survey of the large mammals of peninsular India. Journal of Bombay Natural History Society, 69, 322–349.
Langergraber, K. E., Siedel, H., Mitani, J. C., Wrangham, R. W., Reynolds, V., Hunt, K., & Vigilant, L. (2007). The genetic signature of sex-biased migration in patrilocal chimpanzees and humans. PLoS One, 2(10), e973.
Li, Q. Q., & Zhang, Y. P. (2005). Phylogenetic relationships of the macaques (Cercopithecidae: Macaca), inferred from mitochondrial DNA sequences. Biochemical Genetics, 43(7), 375–386.
Lorenzen, E. D., Nogués-Bravo, D., Orlando, L., Weinstock, J., Binladen, J., Marske, K. A., Ugan, A., Borregaard, M. K., Gilbert, M. T. P., & Nielson, R. (2011). Species-specific responses of Late Quaternary megafauna to climate and humans. Nature, 479(7373), 359–364.
Marmi, J., Bertranpetit, J., Terradas, J., Takenaka, O., & Domingo-Roura, X. (2004). Radiation and phylogeography in the Japanese macaque, Macaca fuscata. Molecular Phylogenetics and Evolution, 30(3), 676–685.
McKay, B. D., & Zink, R. M. (2010). The causes of mitochondrial DNA gene tree paraphyly in birds. Molecular Phylogenetics and Evolution, 54(2), 647–650.
Mittermeier, R. A., Myers, N., Mittermeier, C. G., & Robles Gil, P. (1999). Hotspots: Earth’s biologically richest and most endangered terrestrial ecoregions. CEMEX, SA, Agrupación Sierra Madre, SC.
Mittermeier, R. A., Robles Gil, P., & Mittermeier, C. G. (1997). Megadiversity. Mexico City (Mexico): CEMEX.
Modolo, L., Salzburger, W., & Martin, R. D. (2005). Phylogeography of Barbary macaques (Macaca sylvanus) and the origin of the Gibraltar colony. Proceedings of the National Academy of Sciences of the United States of America, 102(20), 7392–7397.
Mondol, S., Karanth, K. U., & Ramakrishnan, U. (2009). Why the Indian subcontinent holds the key to global tiger recovery. PLoS genetics, 5(8), e1000585.
Mukhopadhyay, K. & Sinha, A. (2010). Quo vadis? Dual-sex emigration in a female-bonded cercopithecine species, the bonnet macaque, in Bandipur National park, southern India. Poster at the Twenty-third Congress of the International Primatological Society, Kyoto, Japan.
Myers, N., Mittermeier, R. A., Mittermeier, C. G., Da Fonseca, G. A. B., & Kent, J. (2000). Biodiversity hotspots for conservation priorities. Nature, 403(6772), 853–858.
Nichols, H. J., Amos, W., Cant, M. A., Bell, M. B. V, & Hodge, S. J. (2010). Top males gain high reproductive success by guarding more successful females in a cooperatively breeding mongoose. Animal Behaviour, 80(4), 649–657.
Nielsen, R., & Wakeley, J. (2001). Distinguishing migration from isolation: a Markov chain Monte Carlo approach. Genetics, 158(2), 885–896.
Nogués‐Bravo, D., Ohlemüller, R., Batra, P., & Araújo, M. B. (2010). Climate predictors of Late Quaternary extinctions. Evolution, 64(8), 2442–2449.
Olson, D. M., & Dinerstein, E. (1998). The Global 200: a representation approach to conserving the Earth‟s most biologically valuable ecoregions. Conservation Biology, 12(3), 502–515.
Perry, S., Manson, J. H., Muniz, L., Gros-Louis, J., & Vigilant, L. (2008). Kin-biased social behaviour in wild adult female white-faced capuchins, Cebus capucinus. Animal Behaviour, 76(1), 187–199.
Petraglia, M. D., Ditchfield, P., Jones, S., Korisettar, R., & Pal, J. N. (2012). The Toba volcanic super-eruption, environmental change, and hominin occupation history in India over the last 140,000 years. Quaternary International, 258, 119–134.
Posada, D. (2008). jModelTest: phylogenetic model averaging. Molecular Biology and Evolution, 25(7), 1253–1256.
Pritchard, J. K., Stephens, M., & Donnelly, P. (2000). Inference of population structure using multilocus genotype data. Genetics, 155(2), 945–959.
Pérez‐Espona, S., Pérez ‐Barbería, F. J., McLeod, J. E., Jiggins, C. D., Gordon, I. J., & Pemberton, J. M. (2008). Landscape features affect gene flow of Scottish Highland red deer (Cervus elaphus). Molecular Ecology, 17(4), 981–996.
Qu, Y., Lei, F., Zhang, R., & Lu, X. (2010). Comparative phylogeography of five avian species:
implications for Pleistocene evolutionary history in the Qinghai‐Tibetan plateau. Molecular Ecology, 19(2), 338–351.
Raaum, R. L., Sterner, K. N., Noviello, C. M., Stewart, C. B., & Disotell, T. R. (2005). Catarrhine primate divergence dates estimated from complete mitochondrial genomes: concordance with fossil and nuclear DNA evidence. Journal of Human Evolution, 48(3), 237–257.
Rambaut, A. (2008). FigTree: Tree figure drawing tool, Version 1.2.2. University of Edinburgh, Edinburgh.
Rambaut, A., & Drummond, A. (2005). Tracer v1.3: MCMC trace analysis tool. University of Edinburgh, Edinburgh.
Rheindt, F. E., & Edwards, S. V. (2011). Genetic Introgression: An Integral but neglected component of speciation in birds. The Auk, 128(4), 620–632.
Rogers, J., Bergstrom, M., Garcia IV, R., Kaplan, J., Arya, A., Novakowski, L., Johnson, Z., Vinson, A., & Shelledy, W. (2005). A panel of 20 highly variable microsatellite polymorphisms in rhesus macaques (Macaca mulatta) selected for pedigree or population genetic analysis. American Journal of Primatology, 67(3), 377–383.
Ronquist, F., Teslenko, M., Van der Mark, P., Ayres, D. L., Darling, A., Höhna, S., Larget, B., Liu, L., Suchard, M. A., & Huelsenbeck, J. P. (2012). MrBayes 3.2: efficient Bayesian phylogenetic inference and model choice across a large model space. Systematic Biology, 61(3), 539–542.
Royden, L. H., Burchfiel, B. C., & Van der Hilst, R. D. (2008). The geological evolution of the Tibetan Plateau. Science, 321(5892), 1054–1058.
Sacks, B. N., Brown, S. K., & Ernest, H. B. (2004). Population structure of California coyotes
corresponds to habitat‐specific breaks and illuminates species history. Molecular Ecology, 13(5), 1265–1275.
Sahni, K. C. (1979). Endemic, relict, primitive and spectacular taxa in eastern Himalaya and strategies for their conservation. Indian Journal of Forestry, 2, 181–190.
Santos, C., Montiel, R., Sierra, B., Bettencourt, C., Fernandez, E., Alvarez, L., Lima, M., Abade, A., & Aluja, M. P. (2005). Understanding differences between phylogenetic and pedigree-derived mtDNA mutation rate: a model using families from the Azores Islands (Portugal). Molecular Biology and Evolution, 22(6), 1490–1505.
Santosh, M., Kagami, H., Yoshida, M., & Nanda-Kumar, V. (1992). Pan-African ch.arnockite formation in East Gondwana: Geochronologic (Sm-Ndand Rb-Sr) and petrogenetic constraints. Bulletin of Indian Geologists Association, 25, 1–10.
Shimada, M. K. (2000). Geographic distribution of mitochondrial DNA variations among grivet (Cercopithecus aethiops aethiops) populations in central Ethiopia. International Journal of Primatology, 21(1), 113–129.
Singh, M., Kumara, H. N., Kumar, M. A., Singh, M., & Cooper, M. (2006). Male influx, infanticide, and female transfer in Macaca radiata radiata. International Journal of Primatology, 27(2), 515–528.
Sinha A, Chakraborty D, Datta A, Gama N, Kumar, R S, Madhusudan M D, Mendiratta U, Ramakrishnan, U. & M. C. (2013). Arunachal macaque Macaca munzala. In A. J. T. J. and N. Manjrekar (Ed.), Mammals of South Asia: Ecology, Behaviour and Conservation (pp. 198–210). Universities Press, Hyderabad, India.
Sinha, A. (2001). The Monkey in the Towns Commons: A Natural History of the Indian Bonnet Macaque. National Institute of Advanced Studies, Bangalore, India.
Sinha, A., Datta, A., Madhusudan, M. D., & Mishra, C. (2005). Macaca munzala: A New Species from Western Arunachal Pradesh, Northeastern India. International Journal of Primatology, 26(4), 977–989.
Sinha, A., Mukhopadhyay, K., & Dutta Roy, A. (2003). Evolution of unimale social organisation in bonnet macaques. In J. Vanitharani, R. Annamalai, M. Narayanan (Eds.), Proceedings of the 28th Conference of the Ethological Society of India, Department of Zoology, Sarah Tucker College and Tamil Nadu Forest Department, Kalakad Mundanthurai Tiger Reserve, Tirunelveli (pp. 110–115).
Sinha, A., Mukhopadhyay, K., Datta-Roy, A., & Ram, S. (2005). Ecology proposes, behaviour disposes: Ecological variability in social organization and male behavioural strategies among wild bonnet macaques. Current Science, 89(7), 1166–1179.
Slatkin, M. (1991). Inbreeding coefficients and coalescence times. Genetical Research, 58(02), 167–175.
Slatkin, M. (1993). Isolation by distance in equilibrium and non-equilibrium populations. Evolution, 264–279.
Smith, D. G., & McDonough, J. (2005). Mitochondrial DNA variation in Chinese and Indian rhesus macaques (Macaca mulatta). American Journal of Primatology, 65(1), 1–25.
Soares, P., Ermini, L., Thomson, N., Mormina, M., Rito, T., Röhl, A., Salas, A., Oppenheimer, S., Macaulay, V., & Richards, M. B. (2009). Correcting for purifying selection: an improved human mitochondrial molecular clock. The American Journal of Human Genetics, 84(6), 740–759.
Soman, K., Thara, K., Arakelyants, M. & Golubyev, V. (1990). Mineral ages of pegmatites from the Palghat Gap region in Kerala and their tectonic significance. Journal of the Geological Society of India, 35, 82–86.
Storey, B. C. (1995). The role of mantle plumes in continental breakup: case histories from Gondwanaland. Nature, 377(6547), 301–308.
Storz, J. F. (1999). Genetic consequences of mammalian social structure. Journal of Mammalogy, 553–569.
Subramaniam, K. A., and Sinha, A. (1999). Social relationships influence decision-making during group fission in wild bonnet macaques. Advances in Ethology, 34, 142.
Taberlet, P., Fumagalli, L., Wust‐saucy, A. G., & Cosson, J. F. (2002). Comparative phylogeography and postglacial colonization routes in Europe. Molecular Ecology, 7(4), 453–464.
Thompson, C. (2009). The eastern Himalays: Where worlds collide. World Wildlife Fund for Nature, India and Nepal.
Toews, D. P. L., & Brelsford, A. (2012). The biogeography of mitochondrial and nuclear discordance in animals. Molecular Ecology.
Tosi, A. J., Morales, J. C., & Melnick, D. J. (2003). Paternal, maternal, and biparental molecular markers provide unique windows onto the evolutionary history of macaque monkeys. Evolution, 57(6), 1419–1435.
Weir, B. S. (1996). Genetic data analysis II: methods for discrete population genetic data. Sunderland, MA: Sinauer Associates.
Weir, B. S., & Cockerham, C. C. (1984). Estimating F-statistics for the analysis of population structure. Evolution, 38(6), 1358–1370.
Wrangham, R. W. (1980). An ecological model of female-bonded primate groups. Behaviour, 75(3-4), 262–300.
Wright, S. (1965). The interpretation of population structure by F-statistics with special regard to systems of mating. Evolution, 395–420.
Wright, S. (1978). Evolution and the genetics of populations. University of Chicago Press, Chicago.
Xia, Z. K. (1997). Environment in the Quarternary. Beijing University Press, Beijing.
Yang, S., Dong, H., & Lei, F. (2009). Phylogeography of regional fauna on the Tibetan Plateau: A review. Progress in Natural Science, 19(7), 789–799.
Yu, G., Gui, F., Shi, Y., & Zheng, Y. (2007). Late marine isotope stage 3 palaeoclimate for East Asia: a data–model comparison. Palaeogeography, Palaeoclimatology, Palaeoecology, 250(1), 167–183.
Zalewski, A., Piertney, S. B., Zalewska, H., & Lambin, X. (2009). Landscape barriers reduce gene flow in an invasive carnivore: geographical and local genetic structure of American mink in Scotland. Molecular Ecology, 18(8), 1601–1615.
Zhan, X., Zheng, Y., Wei, F., Bruford, M. W., & Jia, C. (2011). Molecular evidence for Pleistocene refugia at the eastern edge of the Tibetan Plateau. Molecular Ecology, 20(14), 3014–3026.
Zhao, N., Dai, C., Wang, W., Zhang, R., Qu, Y., Song, G., Chen, K., Yang, X., Zou, F., & Lei, F. (2012). Pleistocene climate changes shaped the divergence and demography of Asian populations of the great tit Parus major: evidence from phylogeographic analysis and ecological niche models. Journal of Avian Biology.
Zheng, B., Xu, Q., & Shen, Y. (2002). The relationship between climate change and Quaternary glacial cycles on the Qinghai–Tibetan Plateau: review and speculation. Quaternary International, 97, 93–101.
Zhou, S., Wang, X., Wang, J., & Xu, L. (2006). A preliminary study on timing of the oldest Pleistocene glaciation in Qinghai–Tibetan Plateau. Quaternary International, 154, 44–51.
Zink, R. M., & Barrowclough, G. F. (2008). Mitochondrial DNA under siege in avian phylogeography. Molecular Ecology, 17(9), 2107–2121.