Notes on the diet and habitat selection of the Sri Lankan Leopard Panthera pardus kotiya (Mammalia: Felidae) in the central highlands of Sri Lanka
Andrew M. Kittle 1, Anjali C. Watson 2, P.H.S. Chanaka Kumara 3, S.D. Kashmi C. Sandanayake 4, H.K. Nimalka Sanjeewani 5 & T. Saminda P. Fernando 6
1,2,3,4,5,6 The Leopard Project, Wilderness & Wildlife Conservation Trust, 130 Reid Avenue, Colombo 4, Sri Lanka
1 Department of Integrative Biology, University of Guelph, Guelph, Ontario, Canada
3,4,5 Department of Natural Resources, Faculty of Applied Sciences, Sabaragamuwa University of Sri Lanka, Belihul Oya, Sri Lanka
6 Department of Zoology, University of Colombo, Colombo, Sri Lanka
1 firstname.lastname@example.org (corresponding author), 2 email@example.com, 3 firstname.lastname@example.org, 4 email@example.com, 5 firstname.lastname@example.org, 6 email@example.com
Abstract: The endangered Sri Lankan Leopard Panthera pardus kotiya occupies the islandŐs highly fragmented central hills where data on its feeding ecology and habitat use is largely absent. This studyŐs objective was to investigate diet and resource selection of leopards here with a focus on the extent of potential interactions with humans in this heavily populated, largely unprotected landscape. Fecal sample analysis was undertaken to investigate diet and sign index counts and selectivity index analysis to determine habitat and landscape features important to fine scale leopard utilization. Results indicated that leopards in the central hills hunt a wide range of prey (at least 10 genera), including larger species where available (e.g., Sambar Rusa unicolor) and smaller, more specialized prey (e.g., Porcupine Hystrix indica) where necessary. No domestic species were recorded in scat analysis (N=35) despite the availability of dogs Canis familiaris, suggesting such predation may be atypical in Sri Lanka. Leopards use a range of landscapes within the region including established and regenerating forests, plantation lands (e.g., pine, eucalyptus, tea), and areas in close proximity to human settlement. At a fine scale, areas of dense undergrowth including tall grasslands were preferred to more open forest, as well as to Pine Pinus caribaea monocultures. Avoidance of humans may be influencing these patterns. This study has important implications as researchers and managers necessarily expand beyond focusing on protected areas toward integrated, landscape-level conservation strategies.
Keywords: Anthropogenic disturbance, diet, habitat use, Panthera pardus kotiya, unprotected areas.
The Sri Lankan Leopard Panthera pardus kotiya is an Endangered (Kittle & Watson 2008) endemic sub-species and the islandŐs apex predator (Miththapala et al. 1996; Uphyrkina et al. 2001). It has evolved without intra-guild competition at least since Sri Lanka split off from the Indian subcontinent ~5,000–10,000 ybp (Deraniyagala 1992; Yokoyama et al. 2000), a unique scenario which may mean that this top carnivore is of keystone importance to the system (Paine 1966).
Data on the island-wide distribution and relative abundance of leopards is becoming increasingly refined (Kittle & Watson 2008; Kittle et al. 2012) and it is now acknowledged that these versatile predators occupy unprotected as well as protected areas (Watson & Kittle 2004). Between these extremes are semi-protected areas, small forest reserves and sanctuaries demarcated on paper but not actively patrolled and routinely utilized by people. Wildlife populations in unprotected and semi-protected areas are particularly susceptible to heightened levels of human-wildlife conflict.
Sri Lanka is a small (65,610km²) mostly rural (74–78 %) nation with substantial spatial overlap between humans and wildlife (Dept of Census and Statistics 2012). In the central hills, high human population density (301–450 /km², Dept of Census and Statistics 2012), heterogeneous land cover and a lack of protected areas makes understanding leopard diet and resource selection vital. Minimal diet analysis has been conducted here (Ranawana et al. 1998) and leopard diet outside protected areas remains unknown. Furthermore, how leopards use this heavily fragmented habitat mosaic has not been investigated. This data deficiency has been recognized as a barrier to designing conservation and management action plans to ensure the leopardŐs continued survival in the wild (Samarakoon 1999). A minimum of 66 leopards were killed island-wide from 2002–2012, mostly caught in snares set for other species (Kittle & Watson, unpublished data). True mortality figures are unknown but if leopards are preying on domestic species outside protected areas, human-leopard conflict is likely to increase.
Our research objective was to investigate leopard diet and habitat use at two semi- to un-protected study sites in the central hills where leopards have been previously documented (Kittle et al. 2012).
The central highlands occupy a ~2400km² section of Sri LankaŐs south-central interior (Fig. 1), encompassing a vast expanse of often-rugged terrain, which varies greatly in elevation, topography and climate.
The Dunumadalawa Forest Reserve is a secondary growth mid-country wet zone forest situated at the northern end of Sri LankaŐs Central massif (Figs. 1, 2a). This was an active estate until the early 1900s when it was released from plantation use (tea, coffee and cocoa) and allowed to naturally re-generate, with some active reforestation of native plant species since 2000. It now protects the watershed of two reservoirs providing water for Kandy, a town of ~100 000 within the limits of which the reserve is located. Dunumadalawa is characterized by high canopy mixed forest which includes Albizia sp. and jak Artocarpus heterophyllus remaining from the estate days. Large-leaf Mahogany Sweitinia macrophylla, Ceylon Almond Canarium zeylanicum and various Ficus species are also conspicuous. The understory is dominated by unattended tea, coffee and cocoa plants grown wild in sections and the ironwood tree Mesua ferrea in others. Patches of grassland, composed mostly of Imperata cylindrica are scattered throughout the reserve. This grass can grow up to 3m tall and form dense monocultures. The abundant jak appears to provide an important continuous food source for potential leopard prey species including Toque Macaques Macaca sinica aurifrons, Barking Deer Muntiakus muntjak, Wild Boar Sus scrofa and Porcupines Hystrix indica. The reserve includes a small (<40ha) Caribbean Pine Pinus caribaea plantation at its highest point in the southeast. Anthropogenic disturbance is acute in the surrounding area which is characterized by a mosaic of land-use types including sub-montane forest, tea estates, pine and eucalyptus plantations, open scrub, home gardens and villages. The reserve falls under the purview of the Kandy Municipal Council and is extensively utilized by a number of surrounding villages for firewood and fruit collection.
The Agrapatana Arboreatum is in the heart of the tea industryŐs vast plantation lands near the southern extent of the central hills (Figs. 1, 2b; Image 1). A regenerating forest, this was part of a much larger tea estate until the 1990s. It adjoins the Agra-Bopats Forest Reserve, a thickly vegetated 9,800ha sub-montane, montane and cloud forest bordered by tea estates on three sides and connected to Horton Plains National Park (3160ha) to the southeast. Sambar Deer Rusa unicolor, Wild Boar Sus Scrofa, Toque Macaque Macaca sinica, Purple-faced Langur Trachypithecus vetulus and Porcupine all reside within this forest reserve. The two study areas differ in a number of important aspects (Table 1).
Diet: Leopard scat was collected opportunistically and during regular sign index surveys (Rabinowitz 1997) conducted between 2003 and 2009. These were air dried and stored in re-sealable plastic bags. To differentiate leopard from Fishing Cat Prionailurus viverrinus scat only samples with bolus width >2.5cm were retained (Henschel & Ray 2003). Samples were washed and sieved to eliminate unwanted particles, and bones, quills, nails and hooves were manually separated and identified. Remaining hairs were mixed and oven dried at 600C for 24hr. Dried samples were placed in a shallow dish, a point grid overlaid and 50 hairs randomly sampled (Ciucci et al. 2004). Microscopic analysis followed Amerasinghe et al. (1990).
Habitat selection: We investigated fine scale leopard habitat selection at Dunumadallawa by comparing habitat variables at used and available locations (Table 2). Used locations were defined by the presence of leopard spoor (scat, pugmarks, scrapes) on reserve trails detected during regular sign index surveys between 2003 and 2009. Available locations were determined from a systematic random sample conducted in November 2009, with points taken every 100m along the same reserve trails. Four trails (1300–2000 m) that traverse the forest reserve were utilized for the use/availability analysis.
We used JacobŐs selectivity index to determine habitat selection where data was categorical (Fig. 3):
D =(r – p) /(r + p - 2rp)
where r is the proportion of used locations of a given habitat/landscape class and p is the proportional availability of that habitat/landscape class (Jacobs 1974). Values of D range from -1 indicating maximum avoidance to 1 indicating maximum selection, with 0 value indicating use in proportion to availability (Jacobs 1974). This selectivity index minimizes bias from small sample sizes (e.g., rare food items or habitat proportions <10%; Hayward & Kerley 2005). Continuous data was not normally distributed but had equal variance (LeveneŐs test P>0.05) so Mann-Whitney Test was used for analysis (Zar 1999).
Diet: Of the 42 scat samples collected, 35 produced results: 13 from Hantane and 22 from Agrapatana (Table 3). Porcupine was the most frequently represented species found in scat at Dunumadalawa, followed by Barking Deer, whereas Sambar was the most common species in leopard scat in Agrapatana (Table 3). Domestic species were not represented in samples from either location (Table 3).
Habitat selection: Thirty-nine used locations were compared to 68 available locations within Dunumadalawa. Leopards utilized the entire 5km² study area but showed selection for grasslands (JacobŐs Index = 0.9963) and avoidance of the Pinus plantation (JacobŐs Index = -0.9867, Fig. 2a). Areas with dense understory (JacobŐs Index = 0.8158) and no canopy cover (JacobŐs Index = 0.7143) were selected, whereas areas with sparse canopy cover (0–25 %) were avoided (JacobŐs Index = -0.6277). Areas far from permanent water (JacobŐs Index = 0.5790) were preferred to areas close to permanent water (JacobŐs Index = -0.5014, Fig. 2b).
Neither slope (Zw=0.537, p= 0.5907) or elevation (Zw=1.405, p= 0.1611) were significant factors influencing fine scale habitat use, however leopards did prefer areas with lower tree density (Zw=-2.443, p=.0147) and larger trees (Zw =-1.919, p= 0.0554).
Diet: Sample size is too small to definitively characterize leopard diet in these areas (Mukherjee et al. 1994; Trites & Joy 2005), but given the complete dearth of dietary data from un-protected areas in the central highlands, even this preliminary information is valuable. The dietary breadth observed is wide, including 10 genera ranging in size from 112g (Palm Squirrel; Phillips 1980) to 130kg (Sambar hind; Phillips 1984). This catholic diet is consistent with leopard feeding ecology throughout its range reflecting opportunistic predation and the leopardŐs adaptive ability to broaden its diet beyond the preferred range of medium-sized ungulates, 23–56 kg (Stander et al. 1997; Hayward et al. 2006).
The Indian Porcupine was represented most frequently in scat (61.5%) in the small, isolated forest reserve despite it being absent from HaywardŐs (2006) exhaustive analysis of leopard prey preferences (111 prey species). Porcupines are potentially dangerous prey, but the prey suite available to leopards in Dunumadalawa is relatively limited (Sanjeewani 2010), so the likelihood of risking injury to secure adequate resources is increased (Brown & Kotler 2007). Alternately, leopards here may have adapted to hunting porcupines with minimum risk or a single leopard may be responsible for the majority of the scat samples collected here, although three individuals have been observed in the reserve (Kittle et al. 2012). Prey specialization by individuals within a population is not uncommon across taxa (Arajo et al. 2011) including among apex terrestrial carnivores (Ross et al. 1997; Elbroch & Wittmer 2013). Establishing whether observations are the reflection of a population-level adaptation to the suite of resources available or a case of individual specialization requires additional sampling. The implications of either can be profound (Bolnick et al. 2003).
In Agrapatana Sambar was most frequently represented in scat samples (59.1%), consistent with probable prey availability, given that the study area is contiguous with Horton Plains National Park (HPNP) where sambar density is estimated to be 66.5/km² (Rajapakse 2003). Previous scat analysis from HPNP found sambar in 75.8% of samples (N=22) (Ranawana et al. 1998). That fewer samples contained Sambar here than in HPNP, and more contained Black-naped Hare and Purple-faced Langurs (Table 3; 13.6% compared to 6.8% and 3.4% in HPNP, Ranawana et al. 1998) hints at variation in prey availability between the core of a protected area such as HPNP and an adjoining yet peripheral area such as Agrapatana. More samples are needed to validate this hypothesis.
Scat samples included no domestic species at either site despite dogs being common at both. This suggests low human-leopard conflict in these regions despite close proximity to human populations, but given the low sample size, interpretations must be made with caution. In Dunumadalawa dogs have been predated in the past, however, these incidents may be uncommon. Questionnaire surveys here show fishing cats responsible for some predation events blamed on leopards (Kittle & Watson, unpublished data). That predation on domestic animals appears atypical is relevant from a management perspective since predators viewed as a menace tend to be increasingly persecuted (Cavalcanti et al. 2010).
Habitat selection: Leopards in the Dunumadalawa Forest Reserve may be selecting fine scale habitats offering improved hunting efficiency. Leopards and lions select areas of high prey vulnerability not high prey density at the fine scale (Hopcraft et al. 2005; Balme et al. 2007; Davidson et al. 2012). As ambush predators, leopards require at least 20cm of cover for concealment (Bothma & LeRiche 1984), so tall grasses (>30cm) and thick vegetation are sufficient. However, in South Africa hunting leopards select intermediate cover, presumably because dense cover obscures their perception of prey (Balme et al. 2007). Alternatively then, where anthropogenic disturbance is high such as in isolated forests within densely populated urban areas, dense undergrowth and long grass might provide leopards concealment from people, a potential risk in this environment. Avoidance of permanent water may also indicate avoidance of people as the reserveŐs two large permanent water sources are regularly monitored by Water Department personnel. However, leopards did not avoid other high human-use areas (e.g., border villages). The reserve is replete with seasonal streams and pools which allow leopards to effectively exploit areas far from permanent sources.
In summary, this research provides valuable insight into the diet and habitat selection of an endangered apex predator, in an area of its range with little existing information and where it is afforded limited protection. Avoidance of humans in these heavily disturbed landscapes may be an important consideration for leopards when making space-use decisions, as it is in parts of their African range where bush meat is an important commodity and humans and leopards engage in a form of exploitative competition (Henschel et al. 2011). Camera trap images revealed leopards and humans engaged in temporal niche partitioning in both sites, with humans detected in the day and leopards at night at the same location (Kittle et al. 2012). Temporal niche partitioning is a mechanism used to explain carnivore co-existence in Africa (Hayward & Slotow 2009) and might similarly be important in South Asia where top carnivores necessarily share space with humans (Athreya et al. 2013).
As human populations continue to increase and protected areas become more and more isolated (DeFries et al. 2005), the effective conservation of biodiversity becomes increasingly reliant on processes outside protected areas (Hansen & DeFries 2007). These areas tend towards habitat mosaics with forested areas heavily fragmented and surrounded by a patchwork of alternate land use types. Large, wide-ranging species such as top carnivores nevertheless often persist (Lyra-Jorge et al. 2008) and can be adept at exploiting semi and unprotected areas even where human density is high (Athreya at al. 2013). We are increasingly recognizing the value of these sub-optimal, shared habitats in terms of connectivity and as supplements to remaining forest fragments (Caryl et al. 2012), as well as appreciating the important role played by predators in maintaining biodiversity within such fragments (Terborgh et al. 2001). As such, the need to increase our understanding of how apex predators utilize these landscape mosaics is of paramount importance.
Amerasinghe, F.P., U.B. Ekanayake & R.D.A. Burge (1990). Food habits of the leopard (Pantherapardusfusca) in Sri Lanka. Ceylon Journal of Science 21: 17–24.
Arajo, M.S., D.L. Bolnick & C.A. Layman (2011). The ecological causes of individual specialization. Ecology Letters 14: 948–958; http://dx.doi.org/10.1111/j.1461-0248.2011.01662.x
Athreya, V., M. Odden, J.D.C. Linnell, J. Krishaswamy & U. Karanth (2013). Big cats in our backyards: Persistence of large carnivores in a human dominated landscape in India. PlosOne 8(3): e57872; http://dx.doi.org/10.1371/journal.pone.0057872
Balme, G., L. Hunter & R. Slotow (2007). Feeding habitat selection by hunting leopards Panthera pardus in a woodland savanna: prey catchability versus abundance. Animal Behavior 74: 589-596; http://dx.doi.org/10.1016/j.anbehav.2006.12.014
Bolnick, D.I., R. Svanbck, J.A. Fordyce, L.H. Yang, J.M. Davis, C.D. Hulsey & M.L. Forister (2003). The ecology of individuals: Incidence and implications of individual specialization. American Naturalist 161(1): 1-–28; http://dx.doi.org/10.1086/343878
Bothma, J.D.P. & E.A.N. Le Riche (1984). Aspects of the ecology and the behavior of the Leopard (Panthera pardus) in the Kalahari desert. Koedoe, Supplement 27: 259–279.
Brown, J.S. & B.P. Kotler (2007). Foraging and the ecology of fear, pp. 437–480. In: Stephens, D.W., J.S. Brown & R.C. Ydenberg (eds.). Foraging behavior and ecology. University of Chicago Press, Chicago, xv+608pp.
Caryl, F.M., C.P. Quine & K.J. Park (2012). Martens in the matrix: the importance of non forested habitats for forest carnivores in fragmented landscapes. Journal of Mammalogy 93(2): 464–474; http://dx.doi.org/10.1644/11-MAMM-A-149.1
Cavalcanti, S.M.C., S. Marchini, A. Zimmermann, E.M. Gese & D.W. Macdonald (2010). Jaguars, livestock, and people in Brazil: realities and perceptions behind the conflict, pp. 383–402. In: Macdonald, D.W. & A.J. Loveridge (eds.). Biology and Conservation of Wild Felids. Oxford University Press, Oxford, xix+762pp.
Ciucci, P., E. Tosoni & L. Boitani (2004). Assessment of the point-frame method to quantify Wolf Canis lupus diet by scat analysis. Wildlife Biology 10(2): 149–153.
Davidson, Z., M. Valiex, A.J. Loveridge, J.E. Hunt, P.J. Johnson, H. Madzikanda & D.W. MacDonald (2012). Environmental determinants of habitat and kill site selection in a large carnivore: scale matters. Journal of Mammalogy 93(3): 677-685; http://dx.doi.org/10.1644/10-MAMM-A-424.1
DeFries, R.A. Hansen, A.C. Newton & M.C. Hansen (2005). Increasing isolation of protected areas in tropical forests over the past twenty years. Ecological Applications 15:19–26; http://dx.doi.org/10.1890/03-5258
Department of Census and Statistics (2012).Population Atlas of Sri Lanka. Ministry of Finance and Planning, Government of Sri Lanka, Colombo, Sri Lanka, 15pp.
Deraniyagala, S.U. (1992). The prehistory of Sri Lanka: an ecological perspective. Volume 8, Part 2, Archaeological Department, Colombo, Sri Lanka, 430pp.
Elbroch, M.L. & H.U. Wittmer (2013). The effects of puma prey selection and specialization on less abundant prey in Patagonia. Journal of Mammalogy 94(2): 259–268; http://dx.doi.org/10.1644/12-MAMM-A-041.1
Hansen, A.J. & R. DeFries (2007). Ecological mechanisms lining protected areas to surrounding lands. Ecological Applications 17: 974–988; http://dx.doi.org/10.1890/05-1098
Hayward, M.W. & G.I.H. Kerley (2005). Prey preferences of the Lion (Panthera leo). Journal of Zoology (London) 267: 309–322; http://dx.doi.org/10.1017/S0952836905007508
Hayward, M.W., P. Henschel, J. OŐBrien, M. Hofmeyr, G. Balme & G.I.H. Kerley (2006). Prey preferences of the Leopard (Panthera pardus). Journal of Zoology (London) 270: 298–313.
Hayward, M.W. & R. Slotow (2009). Temporal partitioning of activity in large African carnivores: tests of multiple hypotheses. South African Journal of Wildlife Research 39(2): 109–125; http://dx.doi.org/10.3957/056.039.0207
Henschel, P., L.T.B. Hunter, L. Coad, K.A. Abernethy & M. Mhlenberg (2011). Leopard prey choice in the Congo Basin rainforest suggests exploitative competition with human bushmeat hunters. Journal of Zoology (London) 285: 11–20.
Henschel, P. & J. Ray (2003). Leopards in African Rainforests: Survey and Monitoring Techniques. Wildlife Conservation SocietyŐs Global Carnivore Program, New York, USA, 50pp.
Hopcraft, G.C., A.R.E. Sinclair & C. Packer (2005). Planning for success: Serengeti lions seek prey accessibility rather than abundance. Journal of Animal Ecology 74: 559–566; http://dx.doi.org/10.1111/j.1365-2656.2005.00955.x
Jacobs, J. (1974). Quantitative measurement of food selection - a modification of the forage ratio and IvlevŐs electivity index. Oecologia 14: 413-417; http://dx.doi.org/10.1007/BF00384581
Kittle, A.M. & A.C. Watson (2008). Panthera pardus ssp. kotiya. In: IUCN 2012. 2012 IUCN Red List of Threatened Species <www.iucnredlist.org>. [Accessed on 09 March 2013]
Kittle, A.M., A.C. Watson, P.H.C. Kumara & H.K.N. Sanjeewani (2012).Notes on the status, distribution and abundance of the Sri Lankan Leopard in the central hills of Sri Lanka. CatNews 56: 28–31.
Lyra-Jorge, M.C., G. Ciocheti & V.R. Pivello (2008). Carnivore mammals in a fragmented landscape in northeast of So Paulo State, Brazil. Biodiversity and Conservation 17: 1573–1580; http://dx.doi.org/10.1007/s10531-008-9366-8
Miththapala, S., J. Seidensticker & S.J. OŐBrien (1996). Phylogeographic subspecies recognition in leopards (Panthera pardus): molecular genetic variation. Conservation Biology 4: 1115–1132.
Mukherjee, S., S.P. Goyal & R. Chellam (1994).Standardisation of scat analysistechniques for Leopard (Panthera pardus) in Gir National Park, Western India.Mammalia 58: 139–143; http://dx.doi.org/10.1515/mamm.19184.108.40.206
Paine, R.T. (1966). Food web complexity and species diversity. American Naturalist 100: 65–75; http://dx.doi.org/10.1086/282400
Phillips, W.W.A. (1980). Manual of the Mammals of Sri Lanka, Part II. Wildlife and Nature Protection Society of Sri Lanka, 2nd Edition, Colombo, xi+151pp.
Phillips, W.W.A. (1984). Manual of the Mammals of Sri Lanka, Part III. Wildlife and Nature Protection Society of Sri Lanka, 2nd Edition, Colombo, xiv+120pp.
Rabinowitz, A. (1997). Wildlife Field Research and Conservation Training Manual. Wildlife Conservation Society, New York, USA, 277pp.
Ranawana, K.B., C.N.B. Bambaradeniya, T.D. Bogahawatte & F.P. Amerasinge (1998). A preliminary survey of the food habits of the Sri Lanka Leopard (Panthera pardus fusca) in three montane wet zone forests of Sri Lanka. Ceylon Journal of Science 25: 65–69.
Rajapakse, I.K. (2003). Ecology of Sambar Deer (Cervus unicolor unicolor – Kerr, 1792) in relation to habitat requirements and predator pressure by the Leopard (Panthera pardus kotiya - Meyer, 1974) at the Horton Plains National Park of Sri Lanka. PhD Thesis. Open University of Sri Lanka, Colombo, Sri Lanka, vii+125pp.
Ross, P.I., M. Jalkotzy & M. Festa-Bianchet (1997). Cougar predation on bighorn sheep in southwestern Alberta during winter. Canadian Journal of Zoology 75: 771–775; http://dx.doi.org/10.1139/z97-098
Samarakoon, P.A.G.V. (1999). Conserving the leopard in Sri Lanka: the need for a management strategy. Loris 22: 14–16.
Sanjeewani, H.K.N. (2010). Determining the abundance of prey available to leopards in a mid-elevation wet zone forest reserve: A comparison of methods. BSc. Thesis. Department of Natural Resources, Faculty of Applied Sciences, Sabaragamuwa University of Sri Lanka, xiii+44pp.
Stander, P.E., P.J. Haden, ll. Kaqece & ll. Ghau (1997). The ecology of asociality in Namibian leopards. Journal of Zoology (London) 242: 343–364; http://dx.doi.org/10.1111/j.1469-7998.1997.tb05806.x
Terborgh, J., L. Lopez, P.V. Nuez, M. Rao, G. Shahabuddin, G. Orihuela, M. Riveros, R. Ascanio, G.H. Adler, T.D. Lambert & L. Balbas (2001). Ecological meltdown in predator-free forest fragments. Science 294: 1923–1926; http://dx.doi.org/10.1126/science.1064397
Trites, A.W. & R. Joy (2005). Dietary analysis from fecal samples: how many scats are enough? Journal of Mammalogy 86: 704–712; http://dx.doi.org/10.1644/1545-1542(2005)086%5B0704:DAFFSH%5D2.0.CO;2
Uphyrkina, O., W. Johnson, H. Quigly, D. Miquelle, L.L. Markar, M. Bush & S.J. OŐBrien (2001). Phylogenetics, genome diversity and origin of modern Leopard (Panthera pardus). Molecular Ecology 10: 2617-–2633; http://dx.doi.org/10.1046/j.0962-1083.2001.01350.x
Watson, A.C. & A.M. Kittle (2004). Distribution and status of the Sri Lankan leopard - a short report. Cat News 41: 12–15.
Yokoyama, Y., K. Lambeck, P. de Deckker, P. Johnson & L.K. Fifield (2000). Timing of the last glacial maximum from observed sea-level minima. Nature 406: 713–716; http://dx.doi.org/10.1038/35021035
Zar, J.H. (1999). Biostatistical Analysis - 4thEdition. Prentice Hall, New Jersey, USA, 944pp.