Monday, 25 June 2012

Asiatic Wild Dogs

So you know I didn't post for a while - well this is something to show that I wasn't entirely idle during that time. I have been taking courses in animal behaviour and physiology, as well as doing an internship with my vet.

My first assignment for my behaviour course was to study the behaviour of an animal in captivity, and of course I chose a relative to the dog. It is a very limited study, but as a fist attempt at a bit of scientific writing I am quite pleased with it. So I am posting the whole thing here, perhaps it will be of some interest.

Pack cohesion behaviour in the Dhole (Cuon alpinus) in captivity

The dhole (Cuon alpinus), or Asiatic Wild Dog, is the only extant species of the Cuon genus of the Canidae family. Widespread in Central, Southern and Eastern Asia in prehistoric times, it is currently found mainly on the Indian subcontinent and in smaller numbers across South-East Asia (Durbin et al. 2004). The species is found in a range of habitats from tropical forest to high mountainous regions (Maisch 2010). However, the dhole is now on the IUCN Red List of Threatened Species, because loss of habitat and prey animals to human activity, and persecution by humans in the past (IUCN 2011, Fox 1984).
The dwindling numbers of dholes in the last century, and the fact that dholes are very wary of humans, means that the study of their behaviour in the wild has been difficult (Johnsingh 1982, Cohen 1982, Fox 1984). Early problems with breeding in captivity has also limited the number of observations of dhole behaviour in captivity (Cohen 1982, Maisch 2010). However, in the last decade the numbers of captive dholes has increased, as have studies of their behaviour (Maisch 2010).
Dholes are medium-sized canids (12-20kg) with a red or brown, thick coat and a darker bushy tail (Durbin et al. 2004). Taxonomically the dhole has been placed together with the African wild dog (Lycaon pictus) and the South American bush dog (Speothos venaticus) due to the lesser number of post-carnassial molars found in these species, distinguishing them from the genus Canis  (including the wolf and the domestic dog) (Davidar 1975, Durbin et al. 2004). This classification has been questioned, however, attributing these similarities in dentition to convergent evolution due to diet (Durbin et al. 2004). Recent genetic studies confirm that the dhole is in fact more closely related to the genus Canis than previously believed (Durbin et al. 2004, Graphodatsky et al. 2008, Zhang and Chen 2010).
The dhole is a highly social animal, living in packs of up to 40 animals, although the average pack size is between 8-12 (Davidar 1975, Johnsingh 1982, Fox 1984). Dholes are highly co-operative, undertaking both hunting and breeding as a group (Kleiman 1967, Venkatamaran et al. 1995, Fox 1984). They use a wide range of vocalizations for individual recognition and pack co-ordination (Johnsingh 1982, Volodin et al. 2001, Durbin et al. 2004, Volodina et al. 2006).
The dhole displays a notably wide range of behaviours relating to pack cohesion and hierarchy (Cohen 1982, Johnsingh 1982). From early studies onward the predominance of socio-positive and submissive behaviours in establishing and maintaining dhole pack hierarchy has been noted (Kleiman 1967, Davidar 1975, Cohen 1982, Johnsingh 1982, Fox 1984). This has been more recently contrasted to the more fractious, dominance driven dynamics of the wolf pack (Durbin et al. 2004, Maisch 2010).
As the hegemony of interpreting the behaviour of the domestic dog (Canis familiaris) as an evolution from the rigid pack structure of the wolf (Canis lupus) is being reappraised (Koler-Matznick 2002, Bradshaw 2011), studies of the behaviour of the dhole are ever more pertinent to an understanding of the evolution of the Canidae. Bradshaw has suggested that the behaviour of the common ancestor of both the domestic dog and the wolf would have been different to that which has evolved in the current living wolf, which is not easily domesticated (Bradshaw 2011). Indeed, the dhole has been mentioned along other species of wild dogs as a more likely behavioural analogue for an early ancestor of the Canis genus (Koler-Matznick 2002). An ancestor animal with a predisposition to cementing social relationships with submissive or playful interaction would have not only been more easily domesticated, but would explain the relative placidity and lack of aggression in the domestic dog as opposed to the wolf.
The aim of this study was to obeserve the social interaction of dholes in captivity, to test the hypothesis that this interaction relies heavily on socio-positive and co-operative behaviour rather than socio-negative and antagonistic behaviour.

Animals and Study Area
The animals studied were in captivity in Howletts Wild Animal Park in Kent, United Kingdom. The group studied consisted of twelve females, ranging in age from one to eight years, including two sibling yearlings. All dholes had been bred at Howletts.
The animals were housed in a roughly square enclosure measuring approximately 50m x 50m. The enclosure consisted of level grassy terrain, enriched with a number of trees and bushes, a small pond, two wooden platforms, a group of concrete pipes, and a wooden shelter. Within the enclosure there was a smaller fenced-off area of 20m x 20m, to which the doles had free access via a small opening. See Figure 1.

Figure 1: Dhole enclosure

The animals were observed from the accessible northwest and northeast sides of the enclosure, primarily from the small viewing area at the western corner.
Individual dholes are notoriously difficult to distinguish (Fox 1984, Johnsingh 1982), and the animals were thus studied as a group. Thirty minutes of initial ad libitum observation was undertaken between 3.30pm and 4.00pm on 21 November 2011, noting the full range of behaviours seen. The weather was misty but dry, and visibility was fairly good on this day.
An ethogram was devised following these observations in conjunction with previous studies of dhole behaviour (see Appendix I: Ethogram). From of this ethogram, eleven categories of behaviour were chosen for the main scan sampling study. All data gathering on vocalizations was abandoned, since it was not possible to ascertain how many animals were producing sounds at any one time. In addition, categories that were deemed less defined and/or more difficult to observe reliably, such as sniffing and moving away from conspecifics were dropped. During the main study it also became clear that the category of approaching a conspecific was too difficult to observe in a number of individuals at the same time, and it was effectively abandoned.
Data was then gathered for this selection of behaviours using the scan sampling method, recording the number of animals performing each behaviour at 5 minute intervals. The data gathering session lasted two hours between 1.00pm and 3.00pm on the 24 November 2011. The weather went from overcast to sunny spells, with good visibility.
There are several limitations to this study, notably the fact that the group of dholes consisted exclusively of females. An all-male group of a similar size exists at Howletts, however, so there is scope for a comparative study across the sexes.
The visibility of the female dhole group was sometimes partial, due to the amount of vegetation in the enclosure, and the limited area and elevation from which observations could be made. Also, the observer was clearly visible to the dholes, and they were aware of her presence, acknowledging it with curiosity and sometimes fear and aggression. These facts, as well as the relatively small enclosure available to the captive dholes, who have a range in the wild of 40km squared (Johnsingh 1982), has to be kept in mind when considering the results of this study.
In addition, the preliminary observations were made on a day on which the dholes had been fed in the morning. The dholes are fed every other day, and the scan sampling was thus undertaken on a day when the dholes were not fed. It has to be assumed that this will have had some impact on the animals’ behaviour, although the keepers at Howletts suggested that their behaviour was fairly uniform over the feeding and non-feeding days (pers. comm.).

The scan sample data can be found in Appendix II. The results are represented graphically in Chart 1. Note that behaviours which were not observed at all during the scan sampling were excluded from the chart.

It has to be noted that from minute 75 onwards, the active period, at several sample points not all dholes were visible to the observer.
During the first hour of observation the group was resting and immobile the vast majority of the time. There was one case of an animal defecating away from but near its sleeping site, and there were some movement to rearrange resting positions in the other animals. On the whole, the animals remained in the same places, in two groups of three, one group of two and four solitary animals. The group of two appeared to be the two yearlings, judging by their size, and were partially hidden in the wooden shelter.
Two of the four solitary animals appeared to have look-out or guarding roles. Guarding in the dhole pack has indeed been reported by previous observers (Johnsingh 1982, Fox 1984). They were placed on the extremes of the area occupied by the resting pack, and showed more movement and alertness than the other animals. However, as they were also intermittently resting they were not counted as guarding on the scan sampling data chart. The resting positions of the animals from 0-60 minutes are recorded in  Figure 2.

Figure 2: Resting positions of Dholes

At the interval 65-70 minutes into the study, one of the yearlings emerged from the shelter. After a few minutes of stretching and looking around it rushed towards Group 1, three adults, making the characteristic repeated yipping sounds of the dholes. It engaged in begging behaviour towards one of the larger animals. This raised the whole group into intense activity, and multiple cases of begging behaviour took place. At any time in this interval there were three or four groups of two or three animals engaged in begging behaviour directed towards one individual in each group.
From this time and for the rest of the study, 70-120 minutes, the dholes remained relatively active. Apart from multiple instances of begging, guarding and patrolling recorded, and the frequent reorganization of small groups of sitting and lying dholes, the animals engaged in some brief chase games and play fighting. During this period of activity the dholes frequently vocalized, making a repeated yipping sound. It appeared to be made by several if not all individuals, at intervals of a few seconds, lasting from ten seconds to several minutes at a time.
There was an instance of communal defecation. Initially three to four individuals defecated on the same spot, in the interval 80-85 minutes into the study. The place was later repeatedly visited, sniffed and defecated on by other individuals throughout the rest of the observation period.
Apart from one staring match observed during the preliminary study, no fights or threat behaviour was observed among the dholes, although some begging behaviour was met with brief inhibited biting. In contrast, the dholes growled at the observer on a couple of occasions. This was usually performed by a single animal, in one case reminiscent to the “grumble on hind legs” decribed by Volodin et al. 2005 (See ethogram in Appendix II). Towards the end of the observation time, between 110-115 minutes, almost the whole group approached the fence in front of the observer, apparently led by a one of the larger animals. The group was yipping excitedly and two of the larger animals reared up, growled and bit at the fence. After a few minutes the group seemed to lose interest and carried on with their other activities.

All behaviour observed was of a neutral or socio-positive kind. Two characteristics of the behavioural data gathered stand out:
1) Approach and interaction between dholes was mainly performed through begging behaviour, which, as there was no food present at the day of the main observation, must be interpreted as a submissive greeting and social bonding ritual (Maisch 2005). In the active period, begging was frequent (6 out of 11 sample points) and undertaken by a significant proportion (10-25%) of the individuals observed. In addition to this some play was observed, although not sampled.
Also, the over-marking of feaces by the group was observed, effectively producing a communal latrine. It has been suggested that this behaviour serves an intra-group communicative function, rather than being a territorial marking (Cohen 1982, Johnsingh 1982, Durbin et al. 2004). However, it has to be noted that the latrine was placed on the edge of the dhole enclosure, bordering that of the African wild dogs (see Figure 1 and 2). There may therefore also have been a territorial element to this behaviour in this case.
2) Throughout the observation period a significant proportion of the dholes were engaged in behaviours undertaken in a group, either passively sitting or lying together, or actively interacting in the begging ritual. On average, these behaviours together accounted for 55% of individuals at any one time. If patrolling is included as a group behaviour, the average rises to 62% of individuals at any one time.
The yipping noise made by many dholes while patrolling has been interpreted as a way of coordinating the group’s movement (Johnsingh 1982, Fox 1884, Durbin et al. 2004, Volodina et al. 2006, Maisch 2010), suggesting that patrolling can be seen as a group behaviour, although the data gathered by this study is not sufficient to distinguish between group and solitary patrolling.
It was also notable that throughout the initial period of inactivity, the dholes were resting in well-defined groups, as seen in Figure 2. It was clear that Group 1 (see Figure 3.), which included the largest animals, as well as Group 2, were composed of higher ranking animals. These two groups occupied the highest points in the enclosure, the two wooden platforms. Indeed, these animals seemed to be among the ones mostly at the receiving end of begging behaviour, although as mentioned, the reliable distinction of individuals was not possible. The smaller group of two yearlings that were hiding in the shelter, on the other hand, appeared to be the instigators of much begging.

Figure 3: Group 1 consisting of three large adult dholes on Platform 1.

Thus, even the sleeping formation of the dhole pack seems to be related to the social relationships within the group, and must be seen as a behaviour that positively aids social cohesion. 

The prolonged period of inactivity at the beginning of the observed period may be explained by two factors. 1) The dholes had no food in the enclosure, the presence of which had been a reason for some activity seen during preliminary observations. 2) Dholes are often, although not exclusively, crepuscular (Johnsingh 1982, Durbin et al. 2004), and the period of activity seemed to correlate with approaching dusk. Indeed the preliminary observations, in which the dholes were more active, were undertaken later in the day. 
The data gathered in this brief study does seem to indicate a mainly socio-positive and co-operative behavioural pattern in the dhole pack. This result correlates with earlier observations of the pack-cohesion of the dhole (Kleiman 1967, Cohen 1982, Johnsingh 1982, Fox 1984). The predominance of submissive displays and play behaviour in the Cuon alpinus is particularly interesting in the context of the Canidae family and the evolution of the Canis genus. The majority of canids are far less socially co-operative (Kleiman 1967, Fox 1976, Bradshaw 2011) than the Canis genus and its precursors. In addition, the placidity and predisposition to submissive social bonding of the domestic dog contrasts with the dominance displays of the current wolf. The similarities between the socio-cohesive behaviour of the dhole and the domestic dog (Davidar 1975) suggests and interesting avenue of investigation of the Cuon alpinus as a more apt behavioural analogue of the ancestors of the Canis genus, and precursors to the domestic dog.


Bradshaw, J. (2011) In Defense of the Dog: Why dogs need our understanding London: Penguin.
Cohen, J. A. (1982) ‘A Note on the Behaviour of captive dholes (Cuon alpinus).’ Journal of the Bombay Natural History Society, 62: 146-148.
Durbin, L.S., Venkataraman, A., Hedges, S., and Duckworth, W. (2004) ‘Dhole.’ In Sillero-Zubiri, C., Hoffmann, M., and Macdonald, D. W. (eds.) Canids: Foxes, Wolves, Jackals and Dogs - 2004 Status Survey and Conservation Action Plan. IUCN/SSC Canid Specialist Group, 210-219. Available at: (Accessed 17 November 2011).
Davidar, E. R. C. (1975) ‘Ecology and the behaviour of the Dhole or Indian Wild Dog (Cuon alpinus).’ In Fox, M. W. (ed.) The Wild Canids: Their Systematics, Behavioural Ecology and Evolution Wenatchee: Dogwise Publishing.
Fox, M. W. (1984) The Whistling Hunters: Field Studies of the Asiatic Wild Dog (Cuon alpinus) Albany: Universiy of New York Press.
Graphodatsky, A. S., Perelman, P. L., Sokolovskaya, N. V., Beklemisheva, V. R., et al. (2008) ‘Phylogenomics of the dog and fox family (Canidae, Carnivora) revealed by chromosome painting.Chromosome Research 16(1), 129-143.
Karanth, K. U., and Sunquist, M. E. (1995) ‘Prey selection by tiger, leopard and dhole in tropical forests.’ Journal of Animal Ecology 64(4), 439-450.
Karanth, K. U., and Sunquist, M. E. (2000) ‘Behavioural correlates of predation by tiger (Panthera tigris), leopard (Panthera pardus) and dhole (Cuon alpinus) in Nagarahole, India.’ Journal of Zoology 250(2), 255-265.
Kleiman, D.G. (1967) ‘Some Aspects of Social Behavior in the Canidae.’  American Zoologist 7(2), 365-372.
Koler-Matznick, J. (2002) ‘The origins of the dog revisited.’ Antrhozoos 15(2), 98-118.
Johnsingh, A. J. T. (1982) ‘Reproductive and social behaviour of the Dhole, Cuon Alpinus (Canidae).’ Journal of Zoology, 198, 443-463.
Maisch, H. (2005) Ist das Fortpflanzungssystem bei Rothunden Cuon alpinus die Ursache oder eine Konsequenz des Rudellebens? Dissertation at Fachbereich Biologie/Chemie der Universität Osnabrück. Available at: (Accessed: 17 November 2011).
Maisch, H. (2010) ‘The influence of husbandry and pack management on Dhole Cuon alpinus reproduction.’ International Zoo Yearbook 44, 149-164.
Venkataraman, A. B. Arumugam, R., and Sukumar, R., (1995) ‘The foraging ecology of dhole (Cuon alpinus) in Mudumalai Sanctuary, southern India.’ Journal of Zoology 237(4), 543-561.
Volodin, I. A., Volodina, E. V., and Isaeva, I. V. (2001) ‘Vocal repertoire in the dhole Cuon alpinus (Carnivora, Canidae) in captivity.’ Entomological Review 81(1), S161-S166.
Volodina, E. V., Volodin, I. A., Isaeva, I. V. and Unck, C. (2006) ‘Biphonation may function to enhance individual recognition in the dhole, Cuon alpinus.’ Ethology 112(8), 815-825.
Wang, S. W. and Macdonald, D. W. (2009) ‘Feeding habits and niche partitioning in a predator guild composed of tigers, leopards and dholes in a temperate ecosystem in central Bhutan.’ Journal of Zoology 277(4), 275-283.
Zhang, H. & Chen, L. (2010) ‘The complete mitochondrial genome of dhole Cuon alpinus: phylogenetic analysis and dating evolutionary divergence within Canidae.’ Molecular Biology Reports, 38(3), 1651-1660.

Appendix I: Ethogram
Observation ad libitum of all-female group of twelve individuals aged 1-8 years, captive in Howletts Widlife Park, 3.30pm -4.00pm on 21 November 2011.
Behaviour observed by Volodin 2001, and Maisch 2005.

Behaviour selected to be studied by scan sampling in the main data gathering session of this study are indicated in bold.

Short squeaking noise.
Responses: yips / no response.
Longer growling noise.
Responses: quick retreat / no response.
Grumble on hind legs
Longer growling noise, standing on hind legs facing the “threat”.

Individual  sitting or standing near perimeter of enclosure, paying attention to the outside.
Moving at a moderate pace, without a specific goal. Alert and attentive to other dholes. Often accompanied by repeated “yip” vocalization.
Approaching other dhole
Walking or running towards another dhole and coming within 1 meter of other individual.
Moving away from other dhole
Purposefully walking or running away from other individual, increasing distance to over 5 feet. Not a chase game.
Sniffing ground
Stationary or walking, nose to the ground.
Sniffing conspecific
Stationary or walking, nose close to conspecific.
Sitting or lying down – solitary
Animal sitting or lying further than 1 meter away from other animal.

Positive social interactions:
Sitting or lying down – group
Animal sitting or lying within 1 meter of other animal.
Approach to conspecific, with head held low, ears lying back. Tail wagging and/or coiled on the side. Contact made with snout, often licking conspecific’s muzzle.
May be accompanied by pawing, circling of the conspecific, and whining.
Responses: snarling, inhibited muzzle biting, reciprocating, playing, giving food (if food is around), denying food, moving away, passive.
Chase game
Enticing chase by approaching and then quickly running away from conspecific, which follows. Roles may be swapped. May lead to play fight.
Play fight
Enticed by play bowing, nudging or muzzle nipping. Involves one individual playing submissive, head held low, rolling over, and other individual standing over or jumping over, nudging, bumping and playfully biting neck of “submissive” animal. Roles frequently reversed. May be combined with chase game.

Negative social interactions:
Staring match
Both parties head held low, ears back, eyes almost closed. Stare at each other and try to wait each other out. May be accompanied by growling.
Wrestling fight
Standing on hind legs, trying to press opponent to the ground. The loser runs away. Growling and “miii” sounds may be made.
No physical contact. Tail held horizontal, u-shaped or up. Hackles raised. Ears and eyes pointing at opponent.
Back may be arched, legs may be stiff and pushing into ground.

Appendix II: Scan Sample Data and Calculations

1 comment:

browndogcbr said...

Hi Y'all,

Just wanted to tell you the information you posted is fascinating.

Wishin' y'all a great week and a wonderful summer! Stay cool!

Y'all come by now,
Hawk aka BrownDog