Rev. sci. tech. Off. int. Epiz., 1993,12 (1), 51-71

The ecology of dogs and canine rabies: a selective review *

A.I. WANDELER **, H.C. MATTER ***, A. KAPPELER *** and A. BUDDE ***

Summary: Although dogs are the most widespread and abundant of all carnivores, the role of the dog in human cultures and its impact on the environment have rarely been studied. These subjects are reviewed in the context of canine rabies.

To understand the epizootiology of canine rabies, the ecology and population biology of the dog must be considered. Information on dog populations (in relation to different habitats, cultures, social strata of human populations and epizootiological situations) was collected in Nepal, Sri Lanka, Switzerland and Tunisia.

In Switzerland (and Western Europe in general), rabies is maintained and spread by red foxes. The low prevalence of rabies in dogs may be explained by restrictive practices of dog-keeping and high rates of vaccination. In the other areas examined, dogs are poorly supervised and their population densities are high enough to support rabies, although it is questionable whether canine rabies exists independently of a wildlife reservoir. Dog-keeping practices, high rates of exposure and various cultural factors may lead to a high human rabies mortality rate. Nevertheless, dogs in these areas remain sufficiently accessible for vaccination and well-executed control programmes could prove successful.

KEYWORDS: Anthropology - Disease control - Dog - Nepal - Population biology - Rabies - Sri Lanka - Tunisia - Vaccination campaign - Zoonosis.

I N T R O D U C T I O N

Tremendous progress has been made in recent decades in the understanding of the biology of the rabies virus and the pathogenesis, prevention and epidemiology of the disease (9). Dreams of controlling rabies in wildlife through vaccination have become a reality. Molecular biology has led us to the threshold of a new era in disease prevention. But the number of human rabies deaths in the world has not diminished accordingly. The majority of these fatalities occur in developing countries as a result of virus transmission by dog bites.

* This paper is dedicated to the memory of the late Dr A. Kuruneru, former Chief of Veterinary Public Health in Colombo, Sri Lanka.

** Agriculture Canada, Animal Diseases Research Institute, Station H, P.O. Box 11300, Nepean, Ontario K2H 8P9, Canada.

*** Swiss Rabies Centre, Institute of Veterinary Virology, Länggass-strasse 122, CH-3012 Bern, Switzerland.

52

ECOLOGY OF THE MOST COMMON CARNIVORE SPECIES

Estimates of the world-wide dog population can rise as high as 500 million or more. No other carnivore species occurs so widely and with such a high average population density, and few other carnivore species reach peak densities similar to those observed for dogs. Domestic dogs are present on every continent and practically every island settled by humans. Only in a small minority of cultures are dogs completely rejected. The dog:human ratio most commonly lies between 1:10 and 1:6, but considerable variation exists (88,102,113,120,126). Recent investigations have often found this ratio to be considerably higher than assumed by local authorities (72, 111). Dogs may even outnumber humans in some areas, as in the Athapaskan villages described by Savishinsky (129). In 1967, the 75 inhabitants of one settlement kept 224 dogs, primarily for pulling sledges.

In urban areas of eastern North America , Beck (16, 17, 19) and Daniels (55) estimated a street dog population density of approximately 150 individuals per km 2 . Much higher densities of poorly supervised dogs were observed in suburban and rural settlements in south-western Sri Lanka, where estimates of up to 3,000 dogs per k m 2

were made.

The abundance of the dog population is dependent on the habitat , especially the availability of resources such as food, water and shelter. Access to these resources depends on settlement patterns, rubbish and waste disposal, rules for keeping animals and other cultural practices. For an understanding of the populat ion biology of the species, it is important to keep in mind that individual dogs may have different ownership status, different degrees of restriction on their movement, social interaction and reproduction, and different levels of dependence on human care (19,76,77,81,148, 156,158). A dog owner may feel responsible for certain aspects of the behaviour and well-being of the animal, while other dogs are ownerless. The laws of most nations oblige people to keep owned dogs under control and to care for them. The responsible dog owner both provides and limits access to resources. In addit ion, the owner theoretically restricts movement, social interaction and reproduction. However, the reality is often different. The meaning of ownership varies in different cultural and social settings (150). In many parts of the world, a large proportion of animals receive little or no supervision. Movement is not restricted, or restricted only for part of the day. Along with food received from their owners, dogs may have access to waste, refuse and other food sources. Reproduction is not under control, but litters are born and raised under the protection of a household. Dogs which do not have individual owners or a referral household may still be accepted by the neighbourhood as belonging to the community. These animals are "community-owned" (155) or, to use a bet ter term, "ne ighbourhood" dogs (158). Members of the ne ighbourhood assume occasional responsibility for these dogs, when it comes to protecting them from dog catchers or bringing them to anti-rabies vaccination clinics. H u m a n care for family and neighbourhood dogs practically eliminates interspecific competition for resources and greatly reduces intraspecific competition.

Dogs without owners, a referral household or social bonds to humans are described as "feral". As Daniels and Bekoff (60) explain, feralisation is not a reversal of domestication, but a process by which individual animals become desocialised from humans or, because born in the wild, never become socialised. Such animals may still depend heavily on resources provided by man. Neighbourhood and feral dogs often

53

scavenge for food in dumps, markets , s laughter areas , roadside areas and temples . Abandoned structures, parked cars and natural shelters provide refuge (16,17,18,19, 55,56,58,59,62).

DOGS AND ANTHROPOLOGISTS

Wolves were among the first animals to be domesticated by man, some 12,000 or more years ago (50, 63, 144). Mutual benefits and to lerance must have been so important (66,76,101,124,127) that the descendant of the wolf, the dog species, rapidly spread to all areas colonised by humans. The useful attributes of dogs include the ability to hunt cooperatively, to guard and defend people, livestock and other property, to pull vehicles and carry goods, to remove refuse and human faeces from the environment, to serve as objects of bar te r and a mea t source, and to act as social par tners and companions. Clutton-Brock (50) hypothesised that the change of weaponry observed during the mesolithic period reflected new hunting methods, which also favoured dogs as hunt ing par tners . Indeed, the anthropological l i terature contains numerous indications that dogs may contribute to the success of a hunter (84,116,161; for older references see 78, 100). However , few investigators present evidence that the accomplishments of a hunter using dogs are greater than those of one without dogs (151). A closer look often reveals that dogs may not be particularly efficient for hunting (85,112,162), or that they can be used only for certain types of game (36,43,161). The widespread use of dogs among northern circumpolar peoples as draught animals to pull sledges and travois (83) has attracted some attention (11, 85,116,117,121,129,142) . This is an example of the cultural and economic value of dogs changing with different circumstances. The practice developed relatively late in the American Arctic: dogs became important in transportation only after native peoples became heavily involved in fur trading (98,129). The popularity of dogs for traction is now declining with the greater use of snow-mobiles (98,116).

Dogs would not have such widespread distribution if their only value was as hunting aids or to pull vehicles. Dog functions are mentioned in numerous studies of cultural anthropology. The vast majority of references relate to the consideration of dogs as mythical or supernatural beings. A comprehensive review of the literature on dogs in African cultures was published by Frank (78), and Latocha (100) produced a similar review on dogs in native South America. Both publications make it clear that attitudes and cultural values regarding dogs vary enormously. In a specific cultural setting, there may be multiple reasons for keeping dogs. Harner (87), for instance, mentions that the Jivaro Indians keep at least one or two dogs in every household, to serve as watchdogs against surprise attacks, to protect garden crops from rodents, for hunting and also as companions which receive care and protection. Among the Jivaro, dogs are also given hallucinogenic drugs to help them obtain supernatural power. There exist only a few other, more detailed accounts of dog/human relationships (e.g. 104,112,129).

However, the reasons for the association of people with dogs are often not so obvious. More than the literature suggests, dogs may be kept or tolerated as social companions and pets; as something to care for, something to make one feel safe; as a social partner of inferior rank or simply as a focus of attention. The psychological importance of owning a pet is well documented for a wide variety of social settings in industrialised nations (53, 73, 91,94). To own an animal companion might also be more important in non-Western societies than recognised to date (104,112,129; see also 78,100). Savishinsky (129) points

54

out that the dogs kept by an Athapaskan band play an important part in social and psychological processes, especially among children. But this seems to be a troublesome concept for most investigators. One gets the impression that anthropologists have certain difficulties in dealing with the omnipresence of dogs and their possible role as pets. An interesting example is the analysis by Crocker of the keeping of animals by the Bororo in central Brazil (54). Crocker attributes pet qualities to the macaws and parrots tamed by the Indians, but not to dogs. He suggests that a distinction should be made between commensal association per se and the fostering of creatures as pets. Crocker admits that dogs have certain attributes of pets in that they are given names and are never eaten. Only the owner knows which of the dozens of dogs lying about the village are actually his, while taking a minimal interest in them. The animals are so unimportant as not to be included in the disposal of property following the death of the owner. The situation described by Crocker obtains in many rural areas of Latin America, Africa and Asia.

The question of why dogs are kept or tolerated becomes even more relevant when one considers the costs involved; for the benefits of dog-keeping may turn into disadvantages. The impact of owned dogs on wildlife was probably, of minor concern to colonists, for instance, who introduced dogs into new geographical locations; however, the ecological consequences of the introduction are becoming increasingly evident (65; see also 12,118, 122,135). Dogs probably played a significant role in the extinction of native wildlife when they were introduced onto islands previously free of vertebrate predators (13,89,95,103). Other costs related to the presence of dogs were most prominently formulated by Feldmann (70; see also 18, 44, 71). Dogs prey not only on wildlife (12), but also on livestock (52); they bite and occasionally kill humans, especially children (8 ,20,41,57, 110,128,130,141,159). Originally, they may have helped to offset some of the dangers of living in fixed settlements by removing human faeces and recycling refuse (67); however, in a modern urban environment dogs are often a nuisance and a source of pollution. Cohen (51) suggests that the domestication of animals must have had various health consequences for human communities. High densities of human populations and accompanying dogs must provide ample opportunity for transmissible parasites to evolve into these epidemiological niches. Indeed, dogs are associated with more than fifty zoonotic disease agents (1,18,90,132,152,153,154) of which echinococcosis/hydatidosis, toxocarosis and rabies are the most prominent. Human atti tudes towards dogs, dog-keeping practices and other aspects of human behaviour influence disease transmission (98,114,132). A well-documented example is echinococcosis in Kenya, where the highest known incidence of human hydatid disease is found among the Turkana (79,80,105,106, 107,108,115). The Turkana keep more dogs and have more intimate contact with them than other tribes in the same area (108). Two main factors may account for a high prevalence of Echinococcus granulosus worms in dogs. Al though the prevalence of Echinococcus cysts in ruminants is low, transmission of the worms may be increased by the fact that dogs have unlimited access to offal and forage on the numerous livestock carcasses which become available during periodic droughts. Dogs also scavenge on human corpses, which are neither buried nor cremated. The fact that humans are active intermediate hosts (105) must influence the evolution of parasite adaptation.

DOG POPULATIONS: TWO CASE STUDIES

As Beck (16) points out, dogs lend themselves perfectly to field studies and testing methods recommended for game biology (154,158).

55

Dogs in Tunisia: a long-term investigation

The human-dog relationship has a long tradition in North Africa (7 ,78,101) . It is interesting to note that Islam obviously does not discourage people from keeping dogs in these countries (5, 6, 24, 25, 33,130), contrary to other parts of the Islamic world (29,120,133).

In 1982, a National Programme for Rabies Control was adopted in Tunisia. Annual mass vaccination campaigns were organised: first in a limited area in the north-east and later covering the whole country. These activities resulted in a marked regression of the incidence of rabies . However , the goal of total eradicat ion of the disease was not achieved (26,27,28,86). In order to collect the data needed for population management and disease control, dog populations in rural and urban environments were studied between 1986 and 1990 (111, 156,157). Five rural and eight urban or suburban study sites were selected in nor thern and central Tunisia. In general, urban and suburban study sites were highly popula ted , predominant ly residential areas which were economically depressed. The methods applied included cap ture-mark- recapture techniques, quest ionnaire surveys, individual marking, focal animal and ad libitum observations (3) and radio-telemetry.

A surprising uniformity was observed in the s t ructure and dynamics of the populations under study; however, dog densities differed markedly between rural and urban/suburban zones. Between 7 and 30 dogs per k m 2 were counted in rural zones, where more than 80% of all households kept one or more dogs (3.0-5.5 inhabitants per dog). In urban and suburban zones, dog populat ion density varied between 700 and 1,000 dogs per km 2 . The percentage of households keeping dogs was less than 20% and sometimes even less than 10% (16 to 46 inhabi tants per dog). The percentage of ownerless dogs was de te rmined in four study sites. In three areas - one rural , one suburban and one urban - the percentage ranged from 7.0% to 8 .3%. In another suburban zone, ownerless dogs accounted for 14.3% of the total dog population, due to the proximity of this area to a large dump site on the outskirts of a provincial capital. However, most dogs in Tunisia are owned. The proportion of ownerless dogs exceeds 10% only in zones where an accumulat ion of resources is available which at tracts unsupervised and ownerless dogs from less favourable areas. These findings are consistent with results from other countries (40,88,109,113).

The mean age in owned dog populations varied between 1.8 and 3.3 years: 12% to 23% were less than one year of age. High turnover rates were observed, with 2 3 % , 30% and 40% of the owned dog population changing within six months in rural, suburban and urban areas respectively. One- and two-year-old bitches had the highest reproductive rate of all age classes.

However, only 20% of live-born pups survived the first three months of life in a rural area. In spite of the high mortal i ty ra te (many owners killed undesired pups) the population was self-sustaining and showed a high reproductive potential. The mean age of well supervised dogs in the Uni ted States of America and in the canton of Bern, Switzerland, is approximately 4.5 years (16,110,113,131). In Tunisia, the situation is closer to that of Mexico City (126), where only 12% of dogs are older than five years. Interestingly, dogs in south-western Sri Lanka also reach a relatively high age. The slower population turnover in these areas is probably linked to frequent reproductive failures (partly caused by Brucella canis). This may be taken as an indication that life expectancy increases when production of viable offspring decreases.

56

The percentage of owned dogs allowed to roam freely is twice as high in rural areas as in urban and suburban areas of Tunisia. However , as a consequence of high population densities, numbers of free-roaming owned dogs in urban and suburban zones were 10- to 100-fold higher than in rural areas. Non-restricted dogs occasionally or regularly moved away from their owners to satisfy fundamental needs such as food, shelter and mates. It is on these occasions that owned dogs frequently come into contact with ownerless dogs. Observations on almost 250 owned dogs in one rural study site revealed that male dogs left their referral household significantly more often and covered greater distances than bitches. However, maximal distances of eight kilometres between localisation site and household were observed for dogs of both sexes, although in most cases (57.7%) dogs were observed at less than 500 metres from their household.

In 1986 and 1987, over a per iod of 22 months , 34 ownerless dogs (17 males and 17 females) were studied in a rural area in north-western Tunisia. At the time of first observation, thirty of the dogs were adult and four juvenile. Eleven had been born to ownerless bitches, while another seven had been chased away or abandoned by their original owners. The origin of sixteen dogs was unknown. Fourteen litters (59 pups) from the observed ownerless bitches were localised. Mortality was relatively high early in life: eleven pups were known to have been killed by residents, five died from natural causes and twenty-one disappeared. Twenty pups were adopted by residents; however, these also suffered extremely high mortality. Lastly, only two pups born in the wild survived in the wild and joined a pack of dogs. The overall rearing success of ownerless bitches was insufficient to compensate for the losses of this population segment.

The social organisation of urban and rural dog populations has been described and discussed on several occasions (21,32, 52,55,56, 61,74,76,77,134) . Beck (16,17) and Daniels and Bekoff (61) found a stronger tendency for dogs to avoid each other than to form groups. There were more single dogs, fewer pairs and slightly more large groups than expected when comparing the observed frequency distribution of groups to a 0-truncated Poisson distribution. Font (74) criticised this statistical approach as an inappropriate means of analysing the social tendencies of free-roaming dogs. In fact, pack formation has been described both in urban (52, 74,77) and, more frequently, in rural environments (13,41, 95,102,118,122,135). The traditional definition of a group is based on temporal, spatial and behavioural components (125), and therefore tends to be impractical in field work (55). In the Tunisian study, grouping was described by spatial distribution only. This allowed measurement of social tendencies by calculating for each dog the overall relative frequency with which it was observed together with other dogs, and the individual relative frequency of observation together with any specific dog within the study area.

Overall relative frequencies varied between 0% and 86% (mean: 62%; s tandard deviation: 15%). One male dog, which had been abandoned by its owner near a dump, was never observed in contact with other dogs. For five months (until death), this dog did not move from the site of abandonment and limited its activity range to a minimal area of 0.06 km 2 as estimated by the convex polygon method. In the same study area, the home range sizes of twenty-one ownerless dogs, calculated on the basis of more than 4,300 localisations, averaged 1.9 km 2 , ranging from 0.2-8.5 km 2 . Similar behaviour of recently abandoned dogs was observed by Daniels and Bekoff in New Mexico (61, 62). A very low contact rate was also determined for a pregnant female. Before and after parturition, this bitch avoided any contact with other dogs except her own pups. The existence of four stable non-oestrous groups was demons t ra ted by tabulat ing the individual relative frequency of grouping for all dogs in the study area. Group size

57

varied mainly due to the death of group members and, in one case, due to integration of two wild-born pups (group A: 5-8 members ; B: 4-6 members ; C: 2-4 members ; D: 2 members) . At least two factors seemed to contribute to group formation in the study area: genetic relatedness and familiarity between non-related individuals (111). Of the observed non-oestrous groups, 80% were formed exclusively by ownerless dogs. However, the situation is quite different in oestrous groups. U p to 55% of observed oestrous groups consisted of bo th owned and ownerless dogs. Ownerless males attempted to mate with owned bitches in heat and owned males joined oestrous groups with ownerless oestrous females. This regular contact between ownerless dogs and occasionally or permanently free-roaming owned dogs may be of special importance in the transmission of rabies and other pathogens affecting dogs and man (33,90).

Lalitpur : an example of operational research

Human rabies is not uncommon in Nepal, especially in the Kathmandu Valley and in the Tera i areas (92). Dogs are the main source of disease. Access to p roper post­exposure treatment is far from optimal. In the 1980s, there were fewer than 500 medical doctors employed in government services in Nepal; only a small fraction were serving in rural areas, whereas 95% of Nepalis live in such areas (137). Local authori t ies are greatly in need of a programme of mass vaccination for dogs, in order to reduce the number of human exposures to rabid dog bites. In 1989, the French non-profit organisation Vétérinaires sans frontières sponsored a dog vaccination campaign in the city of Lalitpur (Patan) in Nepal. The World Health Organisation (WHO) supported the collecting of information on the dog populat ion size in the project area and the vaccination coverage achieved.

Lali tpur is a tradit ional city, highly structured, both socially and architecturally, which is situated on the southern banks of the Bagmati River, in the immediate vicinity of Kathmandu. Plans called for vaccination of a minimum of 75% of the dog population of the city against rabies within one month. To achieve this goal, a strategy of temporary "neighbourhood clinics" was adopted, as opposed to door-to-door vaccination. Five teams each operated a vaccination clinic every morning for three hours. The location was changed daily. Each dog presented at a clinic was vaccinated, regis tered, and marked with a plastic collar, colour-specific for a given clinic. Within 19 working days, vaccination clinics had operated at a total of 110 locations, covering almost the entire area. After this regular campaign, two mobile teams vaccinated unmarked dogs in those areas of town where the percentage was below the desired target (D .D. Joshi, unpublished findings). The number of dogs collared during vaccinations and the propor t ion of collared dogs registered during re-observat ion surveys were used to estimate the total population size and the percentage of vaccinated dogs. At the end of the regular vaccination campaign, a total of 6,886 dogs had been vaccinated. Each of the five teams had treated between 8 and 195 dogs per morning (average: 72). The daily total for all five clinics varied between 141 and 614 (average: 362). Another 1,090 dogs were vaccinated by the two mobile teams during the follow-up campaign. An additional 520 were treated in the Lalitpur Veterinary Hospital over the whole period of the initial and follow-up campaigns. Taking into account that some dogs had lost their collars between the regular campaign and the follow-up, an estimated total of 8,200 dogs were vaccinated.

After the regular campaign, 1,135 dogs were observed in the course of eighteen re-observation surveys: 753 were wearing collars. With a total of 6,886 dogs vaccinated and marked, the total dog population of Lalitpur was estimated to be 10,375, using Bailey's

58

direct sampling method (45). Collar loss rates were estimated to be about 0.75% per day on the basis of information collected by "photographic recapture" (154) in a limited number of areas. If these losses are taken into account, an additional 14 of the 1,135 dogs can be assumed to have been vaccinated, reducing the populat ion est imate to 10,186. The method of direct observation tends to underestimate numbers of puppies. With a correction factor of 5%, a conservative estimate for the total dog population is 10,700. Frith (unpublished findings) found a ratio of people to dogs of 9.42:1, based on quest ionnaire survey data. Estimating the human populat ion as 117,875 (which is probably too high), Frith calculated the dog population to be 12,513 animals. With a corrected estimate of the human population, this figure is reduced to 11,054 dogs, a figure which is relatively close to the estimate based on mark-recapture data. From direct observation, the overall vaccination coverage after the initial campaign was estimated to be 66.3%. If a correction is made for collar losses, a coverage of 67.6% is calculated. Vaccination coverage achieved within specific control areas varied considerably (between 46% and 89%) . The best coverage was found in the "Old Town", where the density of operating points for vaccination centres was highest. Finally, the follow-up campaign increased vaccination coverage to approximately 75%. This relatively easy accessibility of Kathmandu dogs for vaccination was predicted by Bögel and Joshi (39). The same seems to be true in many parts of the world (23,28, 29, 30, 31, 48,49, 99). Similar vaccination coverages were observed in south-western Sri Lanka and Tunisia but, in both areas, high percentages were reached only in close vicinity to temporary vaccination clinics (see also 42).

RABIES IN WESTERN EUROPE AND SRI LANKA: A COMPARISON

Fox rabies in Western Eu rope has been described and analysed by numerous authors (e.g. 35, 139). For rabies virus to survive, it is essential tha t the virus be transmitted by an infected fox, during the short period of virus excretion, to enough other susceptible individuals. The rate of infective contacts is dependent on population density; rabies transmission ceases when popula t ion density drops below a certain level. In summer and autumn, only a small proportion of juvenile foxes are found to be rabid; juveniles attain the same level of infection as adults only during winter. This difference must be due to the higher probabil i ty of adults exposing themselves to infectious bites when they attack disoriented rabid foxes intruding into their territory, whereas juveniles avoid outsiders. The slow spread of a fox rabies epizootic suggests that the disease is usually t ransmit ted by an infected animal to the occupants of a neighbouring territory and only rarely over larger distances. This is also suggested by the observation of radio-tagged infected foxes dying from rabies within their original home range or not very far from it (4). Only occasionally is rabies introduced into a new area during dispersal movements of an infected subadult animal. The progress of epidemic waves is halted by zones of low fox density (139) and areas where more than 60% of the fox population has been immunised (through oral vaccination with baits containing attenuated rabies virus) (140). Where the fox population has been reduced below a certain level (by rabies itself and/or fox control), rabies disappears not only in foxes, but also in all other terrestrial mammal species (but not in bats). Even though the disease can be eliminated under these circumstances, a number of questions still

59

remain with regard to fox rabies (149). In Western Europe, only 0.1-5% of the rabies cases r epor ted annually are in dogs (139). T h r e e factors may account for the low prevalence of rabies in dogs:

- most dogs are restricted in their movements - dogs are kept indoors or in enclosures and are kept on leashes when outside - vaccination of dogs is strongly recommended or even compulsory.

It may also be that virus strains adapted to wild species are not very well suited for propagation within dog populations. There is no recent evidence that fox rabies gives rise to epizootics in urban stray dog populations.

The situation is different in large parts of Asia, Africa and Latin America, where rabies virus circulates in the dog populat ion, accounting for 95% or more of all diagnosed rabies cases (10, 34, 46, 47 ,143 ,145 ,154) . Despi te the easy access to dog populations, little is known about the epizootiology of rabies. Most accounts give the impression that dog rabies is highly enzootic with only mode ra t e fluctuations in prevalence (22,69,82,97,136,138,147).

Sri Lanka can serve as an example for pointing out some of these problems. In Sri Lanka, rabies is endemic in all provinces and dog population densities are extremely high. The human set t lement pa t te rn allows the existence of a cont inuous dog populat ion along the coast and into the areas of cultivated paddies . Most dogs are owned, and between 65% and 85% are accessible for parenteral rabies immunisation. Many dogs living with families are adopted or are tolerated strays. When interviewed, a respondent may decide arbitrarily to declare such an animal as owned or unowned. Less than one-third of all dogs are permanent ly confined to gardens or houses. Approximately one-third of the owned dogs are always free. Most owned dogs may move freely in or out of the houses of their owners.

The majority of animal rabies cases are observed in dogs. But, as in so many other countries with "endemic" dog rabies, there are also a few confirmed wild animal cases. These may be dog rabies offshoots, but could also be the tip of an iceberg of wild carnivore rabies. Sri Lanka has a rich wild carnivore fauna. Jackals (Canis aureus), civet cats (Viverricula indica) and mongooses (Herpestes edwardsi and H. fuscus) occur at densities potentially high enough to support independent cycles of epizootic rabies. The question is not unimportant; dog rabies eradication is an easier goal if the occurrence of the disease in dogs is independent of wildlife rabies.

In the late 1970s and 1980s, the Veter inary Public Hea l th Services in Colombo conducted campaigns in tended to control rabies (96). The strategy was to remove unsupervised dogs and vaccinate supervised owned animals. Teams of dog catchers worked with great skill, but their effect on the dog popula t ion density remained negligible. Less then 5% of the populat ion was removed every year. Such losses are rapidly replaced by offspring. Dogs were immunised by well-trained vaccinators at temporary vaccination clinics. The locations of t emporary clinics were posted and announced by loudspeaker just prior to the campaign. Between 40% and 85% of the dogs in the immediate vicinity of a clinic were brought for immunisation. The spacing of vaccination clinics was such that not more than about 30% of all dogs in an area received vaccine during a part icular campaign. This relatively low immunisat ion coverage cannot prevent a relatively high prevalence of dog rabies and, as a result, frequent human exposures.

60

HUMAN RABIES IN SRI LANKA

Approximately 35,000 persons world-wide die from dog-transmitted rabies each year, and approximately 3.5 million people receive post-exposure treatment annually, mostly after dog bites (37,38). Almost all human rabies deaths and the vast majority of treated bite exposures occur in developing countries (2). South-East Asia is an area of extraordinarily high density of human rabies occurrence (154) and Sri Lanka provides a typical example.

This island nation has approximately 16 million inhabitants. Every year, between 10,000 and 14,000 persons begin post-exposure treatment in many of the approximately 400 hospitals in the country. The annual death toll is between 100 and 400 registered cases. This may be due, in part, to an unusually high rate of exposure to biting rabid animals. But even if this assumption is correct, it does not fully explain the high number of rabies casualties. In view of the high efficacy of modern post-exposure treatment, nearly all human rabies cases must be considered as failures of the medical system; the correct treatment was not applied, or not applied in time. The appropriate treatment may not be universally available (spatially, temporally, socially or economically), or the appropriate treatment is not in compliance with traditional (religious) beliefs. It is also possible that the necessity of the appropriate treatment is not recognised because other treatments are considered equivalent or superior, or because the disease entity is not recognised (e.g. rabies and tetanus are considered identical). Popular or traditional disease classification affects the selection of medical services (for a West African example and short literature review, see 75).

Information has been collected in several areas of Buddhist Sinhalese Sri Lanka on the following:

- frequency of exposure

- the nature of post-exposure treatment

- the nature of the treatment undergone by people who subsequently died of rabies

- the level of knowledge of the disease among the population.

Those interviewed were asked if they or any member of their household had been bitten by a dog in the previous twelve months and, if so, what kind of t reatment had been applied. Depending on the area, between 1% and 12% of the populat ion had experienced dog bites. Some of those bit ten were t rea ted for the injury only, while others received prophylactic anti-rabies treatment. In seeking t reatment to heal dog bites and provide protection against rabies, a bitten Sinhala patient may choose from among folk medicine, exorcism, Ayurvedic medicine and Western medicine (14,68, 93, 123,146,160,163) . Dog bites are often t rea ted with herbal medications alone. For severe injuries, most Sinhala pat ients visit several specialists, including a Western physician and the Sinhala rabies healer . The Western physician in Sri Lanka has a choice between European tissue culture vaccine and locally-produced goat brain vaccine. The lat ter is of quest ionable potency and safety. Tissue culture vaccine is available only to those patients able to afford it (for a short time, tissue culture vaccine was also provided through the public health system). Since treatment with goat brain vaccine is painful and of uncertain success, many Sinhala pat ients prefer to use traditional methods of care alone.

61

The concepts and practices governing health care in Sinhalese Sri Lanka continue past traditions and include the Western medicine introduced by the British (146). When resorting to curers, Sri Lankan patients may apply strategies similar to those of patients in southern India (15). Many Sinhalese resist the idea that diseases are caused by germs. The agent which a rabid animal transmits is considered to be a toxin, and the disease is diagnosed and t rea ted as an intoxication similar to a snake bite (64). Ayurvedic concepts of humora l disorders provide the fundamental principles for t radi t ional Sinhalese t rea tments (119), although they are not free from shamanistic or magical aspects, which are at least partially in accordance with Buddhist religion (93). In all traditional treatments recorded in this study, offerings to a supernatural power, spells of exorcism and/or protective amulets formed an essential part. These transcendental and spiritual aspects are completely lacking in the post-exposure treatments which a patient receives in a government hospital. Traditional medical systems are more holistic than Western medicine and may therefore be psychologically and culturally more acceptable.

CONCLUSIONS AND REMARKS ON DOG RABIES CONTROL

Dogs are kept and/or tolerated in very large numbers in most human societies. This abundance cannot be explained by the economic usefulness of these animals, which is somewhat limited. The tolerance granted to dogs must find an explanation in processes of socialisation and psychology. Cultural practices determine the level of supervision of the social interactions of dogs and access to resources (food, water, shelter and mates). It is assumed that high-density dog popula t ions permit the occurrence of enzootic canine rabies; but this is not very well documented. There are grounds for suspecting that the disease in dogs may not always exist independently of wildlife rabies. However, there is no doubt that rabid dogs are the major source of human infection. The widespread occurrence of human rabies is due not only to the frequency of exposure, but also to the failure to apply proper treatment after bites from rabid animals. In some areas, it might be more cost-effective to bring dog rabies under control than to treat cases of human exposure. Eradication programmes must take into account the biology of the target species. If the target species is the domestic dog, cultural constraints imposed by the human population must also be considered (150).

The W H O (155) and the World-wide Society for the Protection of Animals (158) have issued comprehensive guidelines for dog rabies control and dog popula t ion management. Both documents give detailed guidance on the planning and management of control programmes, legislation and techniques for the execution of local programmes. The contents of these documents cannot be reiterated here; however, a few details can be mentioned.

Taking the cost and benefits of a campaign into consideration, it can be suggested that the goal of such a campaign should be disease eradication, rather than a temporary reduction of the incidence rate. The most economical means for achieving this goal is through mass vaccination. There is a need for comprehensive national plans, rather than temporary, local measures . These plans must identify a goal and take account of national structures and resources. Effective intersectoral cooperation is also necessary. In the past, international organisations (WHO, the Food and Agriculture Organisation

62

of the United Nations [FAO] and the Office International des Epizooties [OIE]) have provided training courses in programme planning and management, and logistics. It can be expected that these programmes will continue in the future, and they could be usefully supplemented by training administrators and organisers in population biology and epidemiology, to increase understanding of the consequences of these campaigns. Pilot projects may help in assessing the following:

- accessibility of dogs - possible means of cooperating with local residents - avenues through which information and education may be provided.

Plans for large-scale operat ions , vaccination strategies and logistic aspects of programmes can then be adjusted to findings in the pilot phase. It is also suggested that some operational research be conducted in future programmes, in order to monitor campaign efficiency.

A C K N O W L E D G E M E N T S

The authors wish to thank K.M. Charl ton, M. Eaglesome and two anonymous reviewers for help and excellent advice.

* * *

ÉCOLOGIE DES CHIENS ET DE LA RAGE CANINE : REVUE SÉLECTIVE. -A.I. Wandeler, H.C. Matter, A. Kappeler et A. Budde.

Résumé : Bien que les chiens soient les plus répandus et les plus nombreux de tous les carnivores, leur rôle culturel et leur impact sur l'environnement ont rarement été étudiés. Ces sujets sont abordés ici, dans le contexte de la rage canine.

Pour comprendre l'épizootiologie de la rage canine, il faut prendre en compte l'écologie et la biologie des populations canines. Des informations sur ces populations ont été recueillies au Népal, à Sri Lanka, en Suisse et en Tunisie et analysées par rapport aux différences d'habitat, de culture et de classe sociale des populations humaines ainsi qu'aux différentes situations épizootiologiques.

En Suisse (et plus généralement en Europe occidentale), la rage est entretenue et propagée par les renards roux. La faible prévalence de la rage chez le chien peut s'expliquer par les restrictions liées à la possession de chiens domestiques et par un taux de vaccination élevé. Dans les autres régions étudiées, les chiens sont mal surveillés et les populations sont suffisamment nombreuses pour entretenir la maladie, bien qu'il soit permis de se demander si la rage canine peut exister indépendamment d'un réservoir dans la faune sauvage. Le mode de surveillance des chiens, les taux de contamination élevés et différents facteurs culturels peuvent entraîner un nombre important de cas de rage humaine. Toutefois, dans ces régions, les chiens restent suffisamment accessibles à la vaccination et les programmes de contrôle pourraient réussir s'ils sont correctement appliqués.

63

MOTS-CLÉS : Anthropologie - Biologie des populations - Campagne de vaccination - Chien - Népal - Prophylaxie - Rage - Sri Lanka - Tunisie -Zoonose.

*

ECOLOGÍA DE LOS PERROS Y DE LA RABIA CANINA: REVISTA SELECTIVA. -A.I. Wandeler, H.C. Matter, A. Kappeler y A. Budde.

Resumen: Aunque los perros son los carnívoros más extendidos y más numerosos, raros son los estudios sobre su papel en la cultura humana y sobre su impacto en el medio ambiente. Los autores abarcan estos temas en el contexto de la rabia canina.

Para comprender la epizootiología de la rabia canina es necesario tener en cuenta la ecología y la biología de las poblaciones caninas. Una serie de datos sobre las poblaciones caninas, reunidos en el Nepal, Sri Lanka, Suiza y Túnez han sido analizados en relación con los distintos hábitats, culturas, estratos sociales de las poblaciones humanas y con las diversas situaciones epizootiológicas existentes.

En Suiza (y Europa occidental en general), la rabia se mantiene y propaga a través de los zorros rojos. La escasa prevalencia de la rabia en los perros se puede explicar por las restricciones impuestas al vagabundeo de los perros y por la alta tasa de vacunación. En las demás regiones estudiadas, los perros son apenas objeto de vigilancia y sus densidades de población son lo suficientemente altas como para permitir que se mantenga la rabia, aunque cabe preguntarse si la rabia canina puede existir independientemente de un reservorio en la fauna salvaje. La ausencia de vigilancia de los perros, las altas tasas de exposición y diversos factores culturales pueden dar lugar a un alto índice de mortalidad por rabia en el hombre. No obstante, la posibilidad de acceso a los perros en esas regiones es suficiente para su vacunación, y programas de control correctamente aplicados podrían dar buenos resultados.

PALABRAS CLAVE: Antropología - Biología de las poblaciones -Campaña de vacunación - Control de enfermedades - Nepal - Perro - Rabia -Sri Lanka - Túnez - Zoonosis.

* * *

REFERENCES

1. ACHA P.N. & SzYFRES B. (1980). - Zoonoses and communicable diseases common to man and animals. PAHO, Washington D.C., 700 pp.

2. ACHA P.N. & ARAMBULO P.V. (1985). - Rabies in the tropics: history and current status. In Rabies in the tropics (E. Kuwert, C. Mérieux, H. Koprowski & K. Bögel, eds). Springer Verlag, Berlin, 343-359.

3. ALTMANN J. (1974). - Observational study of behaviour: sampling methods. Behaviour, 49,227-267.

64

4. ANDRAL L., ARTOIS M., AUBERT M.F .A. & BLANCOU J. (1982). - Radiopistage de renards enragés. Comp. Immun. Microbiol. infect. Dis., 5,285-291.

5. ARTOIS M. & BEN OSMAN F. (1986). - Le chien et la rage en Tunisie. Anthropozool., 4,33-37.

6. ARTOIS M., BEN OSMAN F., KILANI M. & WANDELER A . I . (1986). - New contribution to the knowledge of the ecology of stray dogs in Tunisia (abstract). Comp. Immun. Microbiol. infect. Dis., 9,4-5.

7. AuLAGNiER S. (1990). - Zoogéographie et statut des carnivores sauvages du Maroc. Vie et Milieu, 40,150-155.

8. AVNER J.R. & BAKER M.D. (1991). - Dog bites in urban children. Pediatrics, 88,55-57.

9. BAER G.M. (ed.) (1991). - The natural history of rabies, 2nd Ed. CRC Press, Boca Raton, Florida, 620 pp.

10. BAER G.M. & WANDELER A . I . (1987). - Virus infections of dogs: rabies virus. In Virus infections of carnivores (M.J. Appel, ed.). Elsevier, Amsterdam, 167-182.

11. BALIKCI A . (1970). - The Netsilik Eskimo. Natural History Press, Garden City, New York, 264 pp.

12. BARICK F .B. (1969). - Deer predation in North Carolina and other southeastern States. In Symposium on white-tailed deer in the southern forest habitat. Nacogdoches, Texas, 25-31.

13. BARNETT B .D . & RUDD R .L. (1983). - Feral dogs of the Galapagos islands: impact and control. Int. J. Stud. Anim. Prob., 4,44-58.

14. BASHAM A .L. (1976). - The practice of medicine in ancient and medieval India. In Asian medical systems: a comparative study (C. Leslie, ed.). Univ. of California Press, Berkeley, California, 18-43.

15. BEALS A . R . (1976). - Strategies of resort to curers in south India. In Asian medical systems: a comparative study (C. Leslie, ed.). Univ. of California Press, Berkeley, California, 184-200.

16. BECK A .M. (1973). - The ecology of stray dogs: a study of free-ranging urban animals. York Press, Baltimore, Maryland, 98 pp.

17. BECK A .M. (1975). - The ecology of "feral" and free-roving dogs in Baltimore. In The wild canids, their systematics, behavioral ecology and evolution (M .W. Fox, ed.). Van Nostrand Reinhold, New York, 380-390.

18. BECK A .M. (1975). - The public health implications of urban dogs. Am. J. Pubi. Hlth, 65,1315-1318.

19. BECK A .M. (1980). - Ecological aspects of urban stray dogs. Compend. cont. Educ, 2,721-724.

20. BECK A .M. (1981). - The epidemiology of animal bite. Compend. cont. Educ. pract. Vet., 3,254-258.

21. BEKOFF M. (1979). - Scent-marking by free-ranging domestic dogs: olfactory and visual components. Biol. Behav., 4,123-139.

22. BELCHER D .W., WURAPA F.K. & ATUORA D.O.C. (1976). - Endemic rabies in Ghana: epidemiology and control measures. Am. J. trop. Med. Hyg., 25,724-729.

23. BELOTTO A.J. (1988). - Organization of mass vaccination for dog rabies in Brazil. Rev. infect. Dis., 10, S693-S696.

65

24. BEN OSMAN F. (1965). - Considérations épidémiologiques sur l'hydatidose animale en Tunisie. Arch. Inst. Pasteur, Tunis, 409-418.

25. BEN OSMAN F. (1985). - Le chien errant en Tunisie. Rev. Ecol. (Terre et Vie), 40,197-201.

26. BEN OSMAN F. (1987). - Programme national de lutte contre la rage en Tunisie (1982-1987). Maghreb vét.,3,5-12.

27. BEN OSMAN F. & HADDAD N. (1985). - Les animaux vecteurs de rage : rôle des espèces domestiques et rage canine. In Pasteur et la rage. Infos tech. Serv. vét., 92-95,89S-95S.

28. BEN OSMAN F. & HADDAD N. (1988). - Experience in field rabies control programs. Rev. infect. Dis., 10, S703-S706.

29. BERAN G.W. (1982). - Ecology of dogs in the central Philippines in relation to rabies control efforts. Comp. Immun. Microbiol. infect. Dis., 5,265-270.

30. BERAN G.W. (1985). - Ecology of dogs in developing countries in relation to rabies control programs. In Rabies in the tropics (E. Kuwert, C. Mérieux, H. Koprowski & K. Bögel, eds). Springer Verlag, Berlin, 691-697.

31. BERAN G.W., NOCETE A.P., ELVINA O. , GREGORIO S.B., MORENO R.R., NAKAO J.C., BURCHETT G.A., CANIZARES H.L. & MACASAET F.F. (1972). - Epidemiological and control studies on rabies in the Philippines. Southeast Asian J. trop. Med. publ. Hlth, 3,433-445.

32. BERMAN M. & DUNBAR I. (1983). - The social behaviour of free-ranging suburban dogs. Appl. Anim. Ethol., 10,5-17.

33. BERNARD J., B E N OSMAN F. & JUMINER B. (1967). - Enquêtes sur les helminthes parasites du chien (CFL) à Tunis-Ville. Arch. Inst. Pasteur, Tunis, 44,1-89.

34. BHATIA R., BHARDWAJ M. & SEHGAL S. (1988). - Canine rabies in and around Delhi: a 16 year study. J. commun. Dis., 20,104-110.

35. BLANCOU J. (1988). - Ecology and epidemiology of fox rabies. Rev. infect. Dis., 10, S309-S311.

36. BLURTON-JONES N. & KONNER M.J. (1976). - !Kung knowledge of animal behavior: or the proper study of mankind is animals. In Kalahari hunter-gatherers (R.B. Lee & I. De Vore, eds). Harvard Univ. Press, Cambridge, Massachusetts, 325-348.

37. BÖGEL K. & MOTSCHWILLER E. (1986). - Incidence of rabies and post-exposure treatment in developing countries. Bull. Wld Hlth Org., 64,883-887.

38. BÖGEL K. & MESLIN F. (1990). - Economics of human and canine rabies elimination: guidelines for programme orientation. Bull. Wld Hlth Org., 68,281-291.

39. BÖGEL K. & JOSHI D.D. (1990). - Accessibility of dog populations for rabies control in Kathmandu valley, Nepal. Bull. Wld Hlth Org., 68,611-617.

40. BOITANI L. & FABBRI M.L. (1982). - Censimento dei cani in Italia con particolare riguardo al fenomeno del randagismo. Ric. Biol. Selvaggina, 73,1-51.

41. BORCHELT P.L., LOCKWOOD R., BECK A.M. & Voith V.L. (1983). - Attacks by packs of dogs involving predation on human beings. Publ. Hlth Rep., 98,57-66.

42. BROOKS R. (1990). - Survey of the dog population of Zimbabwe and its level of vaccination. Vet. Rec., 127, 592-596.

43. BULMER R. (1968). - The strategies of hunting in New Guinea. Oceania, 38,302-318.

44. CARDING A.H. (1969). - The significance and dynamics of stray dog populations with special reference to the U.K. and Japan. J. small Anim. Pract., 10,419-446.

66

45. CAUGHLEY G. (1977). - Analysis of vertebrate populations. John Wiley, Chichester, United Kingdom, 234 pp.

46. CHADLI A., BAHMANYAR M. & CHAABOUNI A. (1982). - Epidémiologie de la rage en Tunisie. Etude comparative des résultats des 28 dernières années. Arch. Inst. Pasteur, Tunis, 59,5-21.

47. CHADLI A. & BENLASFAR Z. (1985). - Epidémiologie de la rage en Tunisie. Analyse des résultats des 30 dernières années. In Rabies in the tropics (E. Kuwert, C. Mérieux, H. Koprowski & K. Bögel, eds). Springer Verlag, Berlin, 363-369.

48. CHOMEL B., CHAPPUIS G., BULLON F., CARDENAS E., D E BEUBLAIN T.D., LOMBARD M. & GiAMBRUNO E. (1988). - Mass vaccination campaign against rabies: are dogs correctly protected? The Peruvian experience. Rev. infect. Dis., 10 (Suppl. 4), S697-S702.

49. CIFUENTES E.E. (1988). - Program for the elimination of urban rabies in Latin America. Rev. infect. Dis., 10, S689-S692.

50. CLUTTON-BROCK J. (1984). - Dog. In Evolution of domesticated animals (I.L. Mason, ed.). Longman, London & New York, 198-211.

51. COHEN M.N. (1989). - Health and the rise of civilization. Yale Univ. Press, New Haven, Connecticut, 285 pp.

52. COMAN B.J. & ROBINSON J.L. (1989). - Some aspects of stray dog behaviour in an urban fringe area. Aust. vet. J., 66,30-32.

53. CORSON S.A., CORSON E.O'L., GWYNNE P.H. & ARNOLD L.E. (1977). - Pet dogs as nonverbal communication links in hospital psychiatry. Comp. Psych., 18,61-72.

54. CROCKER J.C. (1977). - My brother the parrot. In The social use of metaphor (J.D. Sapir & J.C. Crocker, eds). Univ. of Pennsylvania Press, Philadelphia, Pennsylvania, 164-192.

55. DANIELS T.J. (1983). - The social organization of free-ranging urban dogs. I. Non-estrous social behavior. Appl. Anim. Ethol., 10,341-363.

56. DANIELS T.J. (1983). - The social organization of free-ranging urban dogs. II. Estrous groups and the mating system. Appl. Anim. Ethol., 10,365-373.

57. DANIELS T.J. (1986). - A study of dog bites on the Navajo reservation. Publ. Hlth Rep., 101,50-59.

58. DANIELS T.J. (1987). - Conspecific scavenging by a young domestic dog. J. Mammal., 68,416-418.

59. DANIELS T.J. (1988). - Down in the dumps: where abandoned domestic dogs must turn wild to survive. Natural History, 4/88,8-12.

60. DANIELS T.J. & BEKOFF M. (1989). - Feralization: the making of wild domestic animals. Behav. Process., 19,79-94.

61. DANIELS T.J. & BEKOFF M. (1989). - Population and social biology of free-ranging dogs, Canis familiaris. J. Mammal., 70,754-762.

62. DANIELS T.J. & BEKOFF M. (1989). - Spatial and temporal resource use by feral and abandoned dogs. Ethology, 181,300-312.

63. DAVIS S.J.M. & VALLA F.R. (1978). - Evidence for domestication of the dog 12,000 years ago in the Natufian of Israel. Nature, 276,608-610.

64. D E SILVA A. & URAGODA C.G. (1983). - Traditional methods of snake-bite treatment in Sri Lanka. Ceylon med. J., 28,170-174.

67

65. DIAMOND J.M. (1984). - Historic extinctions: a Rosetta stone for understanding prehistoric extinctions. In Quaternary extinctions: a prehistoric revolution (P.S. Martin & R.G. Klein, eds). Univ. of Arizona Press, Tucson, Arizona, 824-862.

66. DOWNS J.F. (1960). - Domestication: an examination of the changing social relationships between man and animals. Kroeber anthropol. Soc. Papers, 22,18-67.

67. DUNN F .L. (1972). - Intestinal parasitism in Malay aborigines (Orang Asli). Bull. Wld Hlth Org., 46,99-113.

68. DUNN F .L. (1976). - Traditional Asian medicine as adaptive systems. In Asian medical systems: a comparative study (C. Leslie, ed.). Univ. of California Press, Berkeley, California, 133-158.

69. FEKADU M. (1982). - Rabies in Ethiopia. Am. J. Epidemiol, 115,266-273.

70. FELDMANN B.M. (1974). - The problem of urban dogs. Science, 185,903.

71. FELDMANN B.M. & CARDING T.H. (1973). - Free-roaming urban pets. Hlth Serv. Rep., 88,956-962.

72. FISHBEIN D.B., FRONTINI M.G., DOBBINS J.G., COLLINS E.F., HUERTA G.Q., RODRIGUEZ J.D.G., WOO-MING B., RAMOS J.G., BELOTTO A.J., TORRES J.M.B., YENNE K.M., LINHART S.B. & BAER G.M. (1992). - Prevention of canine rabies in rural Mexico: an epidemiologic study of vaccination campaigns. Am. J. trop. Med. Hyg.,41,317-327.

73. FOGLE B. (ed.) (1981). - Interrelations between people and pets. Thomas, Springfield, Illinois, 370 pp.

74. FONT E. (1987). - Spacing and social organization: urban stray dogs revisited. Appl. Anim. Behaviour Sci., 17,319-328.

75. Fosu G.B. (1981). - Disease classification in rural Ghana: framework and implications for health behaviour. Soc. Sci. Med., 15B, 471-482.

76. Fox M.W. (1978). - The dog: its domestication and behavior. Garland Publishing Inc., New York, 296 pp.

77. Fox M.W., BECK A.M. & BLACKMAN E. (1975). - Behavior and ecology of a small group of urban dogs. Appl. Anim. Ethol., 1,119-137.

78. FRANK B. (1965). - Die Rolle des Hundes in afrikanischen Kulturen. Franz Steiner Verlag, Wiesbaden, Federal Republic of Germany, 256 pp.

79. FRENCH C.M., NELSON G.S. & WOOD M. (1982). - Hydatid disease in the Turkana district of Kenya. I. The background to the problem, with hypotheses to account for the remarkably high prevalence of the disease in man. Ann. trop. Med. Parasitol., 76,425-437.

80. FRENCH CM. & NELSON G.S. (1982). - Hydatid disease in the Turkana district of Kenya. II. A study in medical geography. Ann. trop. Med. Parasitol., 76,439-457.

81. GIFFROY J.M. (1987). - Communication et structure sociale chez le chien. Revue Méd. vét., 138,361-369.

82. GLOSSER J.W., HUTCHINSON L.R., RICH A.B., HUFFAKER R.H. & PARKER R .L. (1970). -Rabies in El Paso, Texas, before and after institution of a new rabies control program. J. Am. vet. med. Assoc., 157,820-825.

83. GRABURN N.H.H. & STRONG B.S. (1973). - Circumpolar peoples: an anthropological perspective. Goodyear, Pacific Palisades, California, 236 pp.

84. GRIFFIN P.B. (1984). - Forager resource and land use in the humid tropics: the Agta of northeastern Luzon, the Philippines. In Past and present in hunter gatherer studies (C. Schrire, ed.). Academic Press, Orlando, Florida, 95-121.

68

85. GuBSER N.J. (1965). - Nunamiut Eskimos - hunters of caribou. Yale Univ. Press, New Haven, Connecticut, 384 pp.

86. HADDAD N., BLANCOU J., GRITLI A. & BEN OSMAN F. (1987). - Etude de la réponse immunitaire des chiens tunisiens à la vaccination antirabique. Maghreb vét, 3,61-64.

87. HARNER M.J. (1972). - The Jivaro: people of the sacred waterfalls. Natural History Press, Garden City, New York, 233 pp.

88. HEUSSNER J.C., FLOWERS A.I., WILLIAMS J.D. & SILVY N.J. (1978). - Estimation of dog and cat populations in an urban area. Anim. Regul. Studies, 1,203-212.

89. HOLDGATE M. W . & WACE N.M. (1961). - The influence of man on the floras and faunas of southern islands. Polar Rec., 10,473-493.

90. JAIEM A.A. (1982). - Contribution à l 'étude épidémiologique de l'echinococcose hydatique dans la région de Sousse. Thesis, Sidi Thabet, Tunisia, 67 pp.

91. JONES B.A. (1981). - The psychology of the human/companion animal bond; an annotated bibliography. Center for the Interaction of Animals and Society, School of Veterinary Medicine, Univ. of Pennsylvania, Philadelphia, Pennsylvania, 42 pp.

92. JOSHI D.D., GAUTAM A., PARAJULI M.B. & JOSHI A.B. (1991). - Human hydrophobia death cases recorded in Kathmandu. J. Nepal. med. Assoc., 31,135-139.

93. KAPFERER B. (1983). - A celebration of demons. Exorcism and the aesthetics of healing in Sri Lanka. Indiana Univ. Press, Bloomington, Indiana, 293 pp.

94. KATCHER A.H. & FRIEDMAN E. (1980). - Potential health value of pet ownership. Compend. cont. Educ. pract. Vet., 2,117-121.

95. KRUUK H. & SNELL H. (1981). - Prey selection by feral dogs from a population of marine iguanas (Amblyrhynchus cristatus). J. appl. Ecol., 18,197-204.

96. KURUNERU A. (1985). - Experience with rabies control in Sri Lanka. In Rabies in the tropics (E. Kuwert, C. Mérieux, H. Koprowski & K. Bögel, eds). Springer Verlag, Berlin, 726-736.

97. KUWERT E., MERIEUX C , KOPROWSKI H. & BOGEL K. (eds) (1985). - Rabies in the tropics. Springer Verlag, Berlin, 786 pp.

98. LANTIS M. (1980). - Changes in the Alaskan Eskimo relation of man to dog and their effect on two human diseases. Arctic Anthropol., 17,2-25.

99. LARGHI O .P . , ARROSI J.C., NAKAJATA A.J. & VILLA-NOVA A. (1988). - Control of urban rabies. In Rabies (J.B. Campbell & K.M. Charlton, eds). Kluwer Academic Publishers, Boston, Dordrecht & London, 407-422.

100. LATOCHA H. (1982). - Die Rolle des Hundes bei Südamerikanischen Indianern. Renner, Hohenschäftlarn, Federal Republic of Germany, 557 pp.

101. LAWRENCE B. (1967). - Early domestic dogs. Z. Säugetierk., 32,44-59.

102. LEHNER P.N., MCCLUGGAGE C , MITCHELL D.R. & NEIL D.H. (1983). - Selected parameters of the Fort Collins, Colorado, dog population, 1979-80. Appl. Anim. Ethol., 10,19-25.

103. LEWIN R. (1978). - Galapagos: the endangered islands. New Scientist, 20 July, 168-172.

104. LUOMALA K. (1960). - The native dog in the Polynesian system of values. In Culture in history (S. Diamond, ed.). Columbia Univ. Press, New York, 190-240.

105. MACPHERSON C.N.L. (1983). - An active intermediate host role for man in the lifecycle of Echinococcus granulosus in the Turkana district of Kenya. Am. J. trop. Med. Hyg., 32,397-404.

69

106. MACPHERSON C . N . L . ( 1 9 8 5 ) . - Epidemiology of hydatid disease in Kenya: a study of domestic intermediate hosts in Masailand. Trans. R. Soc. trop. Med. Hyg., 79 ,209-217.

107. MACPHERSON C.N .L. , KARSTAD L., STEVENSON P. & ARUNDEL J.H. ( 1 9 8 3 ) . - Hydatid disease in the Turkana district of Kenya. III. The significance of wild animals in the transmission of Echinococcus granulosus, with particular reference to Turkana and Masailand in Kenya. Ann. trop. Med. Parasitol., 77, 61-73 .

108. MACPHERSON C .N .L. , FRENCH C.M., STEVENSON P., KARSTAD L. & ARUNDEL J.H. ( 1 9 8 5 ) . - Hydatid disease in the Turkana district of Kenya. IV. The prevalence of Echinococcus granulosus infection in dogs and observations on the role of the dog in the lifestyle of the Turkana. Ann. trop. Med. Parasitol., 79, 5 1 - 6 1 .

109. MARX M.B. & FURCOLOW M.L. ( 1 9 6 9 ) . - What is the dog population? Arch. environ. Hlth, 19 ,217-219.

110. MATTER H . C . ( 1 9 8 5 ) . - Analyse einer Hundepopulation im Kanton Bern mit Hilfe eines Fragebogens. MSc Thesis, Univ. of Bern, Switzerland, 9 2 pp.

111. M A T T E R H . C . ( 1 9 8 9 ) . - P o p u l a t i o n s b i o l o g i s c h e U n t e r s u c h u n g e n an Hundepopulationen in Tunesien. PhD Thesis, Univ. of Bern, Switzerland, 9 7 pp.

112. MEGGITT M.J. ( 1 9 6 5 ) - The association between Australian aborigines and dingoes. In Man, culture, and animals (A. Leeds & A.P. Vayda, eds). American Association for the Advancement of Science, Washington D.C, 7-26.

113. NASSAR R. & MOSIER J.E. ( 1 9 8 0 ) . - Canine population dynamics: a study of the Manhattan, Kansas, canine population. Am. J. vet. Res., 41 ,1798-1803 .

114. NELSON G.S. ( 1 9 7 2 ) . - Human behaviour in the transmission of parasitic diseases. In Behavioural aspects of parasite transmission (E.U. Canning & C.A. Wright, eds). Academic Press, London, 109-122 .

115. NELSON G.S. & RAUSCH R.L. ( 1 9 6 3 ) . - Echinococcus infections in man and animals in Kenya. Ann. trop. Med. Parasitai, 57 ,136-149.

116. NELSON R.K. ( 1 9 7 3 ) . - Hunters of the Northern Forest. Univ. of Chicago Press, Chicago, Illinois, 3 3 9 pp.

117. NELSON R.K. ( 1 9 8 3 ) . - Make prayers to the raven. Univ. of Chicago Press, Chicago, Illinois, 2 9 2 pp.

118. NESBITT W.H. ( 1 9 7 5 ) . - Ecology of a feral dog pack on a wildlife refuge. In The wild canids, their systematics, behavioral ecology and evolution (M.W. Fox, ed.). Van Nostrand Reinhold, New York, 391-396 .

119. OBEYESEKERE G. ( 1 9 7 6 ) . - The impact of Ayurvedic ideas on the culture and the individual in Sri Lanka. In Asian medical systems: a comparative study (C. Leslie, ed.). Univ. of California Press, Berkeley, California, 201-226 .

120. OBOEGBULEM S.I. & NWAKONOBI I.E. ( 1989) . - Population density and ecology of dogs in Nigeria: a pilot study. Rev. sci. tech. Off. int. Epiz., 8 (3 ) , 733-745 .

121. OSWALT W.H. ( 1 9 6 7 ) . - Alaskan Eskimos. Chandler, New York, 2 9 7 pp.

122. OTTO H. (1909) . - Beobachtungen an verwilderten Haushunden. Zool. Beob., 50,176-178.

123. POLUNIN I. ( 1 9 7 6 ) . - Disease, morbidity, and mortality in China, India, and the Arab world. In Asian medical systems: a comparative study (C. Leslie, ed.). Univ. of California Press, Berkeley, California, 120-132 .

124. PRICE E . O . ( 1 9 8 4 ) . - Behavioral aspects of animal domestication. Quart. Rev. Biol., 59,1-32.

70

125. PuLLiAM H.R. & CARACO T. (1984). - Living in groups: is there an optimal group size? In Behavioural ecology: an evolutionary approach, 2nd Ed. (J.R. Krebs & N.B. Davis, eds). Blackwell Scientific Publications, Oxford, United Kingdom, 122-147.

126. RANGEL M.C.F., CARDENAS LARA J. & D E ALUJA A . S . (1981). - The canine population of Mexico City: an estimative study. Anim. Regul. Stud., 3,281-290.

127. REED C.A. (1984). - The beginnings of animal domestication. In Evolution of domesticated animals (I.L. Mason, ed.). Longman, London & New York, 1-6.

128. SACKS J.J., SATTIN R.W. & BONZO S.A. (1989). - Dog bite-related fatalities from 1979 through 1988. J. Am. med. Assoc., 262,1489-1492.

129. SAVISHINSKY J.S. (1975). - The dog and the hare: canine culture in an Athapaskan band. National Museum of Man, Canadian Ethnology Service, Ottawa, Canada, Mercury Series No. 27,462-515.

130. SCHNEIDER R. & BEN OSMAN F. (1964). - Note statistique sur les morsures par chiens errants dans la ville de Tunis. Arch. Inst. Pasteur, Tunis, 41,423-428.

131. SCHNURRENBERGER P.R., KANGILASKI E. & BASHE W.J. (1961). - Characteristics of a rural Ohio dog population. Vet. Med., 56,519-523.

132. SCHWABE C.W. (1969). - Veterinary medicine and human health, 2nd Ed. Williams & Wilkins, Baltimore, Maryland, 713 pp.

133. SCHWABE C.W. & ABOU DAOUD K. (1961). - Epidemiology of echinococcosis in the Middle East. I. Human infections in the Lebanon, 1949-1959. Am. J. trop. Med. Hyg., 10,374-381.

134. SCOTT J.P. & FULLER J.L. (1965). - Genetics and the social behavior of the dog. Univ. of Chicago Press, Chicago, Illinois, 468 pp.

135. SCOTT M.D. & CAUSEY K. (1973). - Ecology of feral dogs in Alabama. J. Wildl. Manag., 37,253-265.

136. SEKAR'M., APPAJI RAO V.N. & RAGHAVAN N. (1989). - Seasonal occurrence of rabies in man and animals. Cheiron, 18,5-9.

137. SHRESTHA R.M. & LEDIARD M. (1980). - Faith healers: a force for change. Educational Enterprises, Kathmandu, Nepal, 129 pp.

138. STANLEY M.J. (1990). - Rabies in Yemen Arab Republic, 1982 to 1986. Trop. Anim. Hlth Prod., 22,273-274.

139. STECK F. & WANDELER A. (1980). - The epidemiology of fox rabies in Europe. Epidemiol. Rev., 2,71-96.

140. STECK F., WANDELER A., BICHSEL P., CAPT S. & SCHNEIDER L. (1982). - Oral immunization of foxes against rabies: a field study. Zentbl. VetMed., B, 29,372-396.

141. SZPAKOWSKI N.M., BONNETT B.N. & MARTIN S.W. (1989). - An epidemiological investigation into the reported incidents of dog biting in the city of Guelph. Can. vet. J., 30,937-942.

142. TAYLOR J.G. (1974). - Labrador Eskimo settlements of the early contact period. National Museum of Man, Canadian Ethnology Service, Ottawa, Canada, Publications in Ethnology, No. 9,105 pp.

143. TSELIKAS E. (1985). - La rage en région méditerranéenne. In Pasteur et la rage. Infos tech. Serv. vét., 92-95,31S-36S.

144. TURNBULL F.F. & REED C.A. (1974). - The fauna from the terminal Pleistocene of Palegawra cave, a Zarzian occupation site in northeastern Iraq. Fieldiana Anthropology, 63, 81-146.

71

145. TURNER G.S. (1976). - A review of world epidemiology of rabies. Trans. R. Soc. trop. Med. Hyg., 70,175-178.

146. URAGODA C.G. (1987). - A history of medicine in Sri Lanka. Sri Lanka Medical Association, Colombo, Sri Lanka, 326 pp.

147. WALTNER-TOEWS D., MARYONO B.T., AKOSO S., WISYNU S. & UNRUH D.H.A. (1990). -An epidemic of canine rabies in central Java, Indonesia. Prevent. vet. Med., 8,295-303.

148. WANDELER A.I. (1985). - Ecological and epidemiological data requirements for the planning of dog rabies control. In Rabies in the tropics (E. Kuwert, C. Mérieux, H. Koprowski & K. Bögel, eds). Springer Verlag, Berlin, 657-661.

149. WANDELER A.I. (1991). - Carnivore rabies: ecological and evolutionary aspects. Hystrix, 3,121-135.

150. WANDELER A.I., B U D D E A., CAPT S., KAPPELER A. & MATTER H.C. (1988). - Dog ecology and dog rabies control. Rev. infect. Dis., 10, S684-S688.

151. WASHBURN S.L. & LANCASTER C.S. (1968). - The evolution of hunting. In Man the hunter (R.B. Lee & I. De Vore, eds). Aldine, Chicago, Illinois, 293-303.

152. WHO (1981). - FAO/UNEP/WHO Guidelines for surveillance, prevention and control of echinococcosis/hydatidosis. WHO, Geneva, 148 pp.

153. WHO (1981). - WHO/WSAVA Guidelines to reduce human health risks associated with animals in urban areas. WHO, Geneva, 74 pp.

154. WHO (1984). - Guidelines for dog rabies control. WHO, Geneva, 351 pp.

155. WHO (1984). - WHO Expert Committee on Rabies. Seventh Report. Tech. Rep. Ser., No. 709. WHO, Geneva, 104 pp.

156. WHO (1988). - Report of WHO Consultation on dog ecology studies related to rabies control. WHO, Geneva, 35 pp.

157. WHO (1990). - Second WHO Consultation on oral immunization of dogs against rabies. WHO, Geneva, 21 pp.

158. WHO/WSPA (1990). - Guidelines for dog population management. WHO, Geneva, 116 pp.

159. WINKLER W.G. (1977). - Human deaths induced by dog bites, United States, 1974-1975. Pubi. Hlth Rep., 92,425-429.

160. WIRZ P. (1941). - Exorzismus und Heilkunde auf Ceylon. Huber, Bern. (Translated 1954. - Exorcism and the art of healing in Ceylon. Brill, Leyden, the Netherlands, 275 pp.)

161. YELLEN J.E. & LEE R.B. (1976). - The Dobe-/Du/da environment: background to a hunting and gathering way of life. In Kalahari hunter-gatherers (R.B. Lee & I. De Vore, eds). Harvard Univ. Press, Cambridge, Massachusetts, 27-46.

162. YOST J.A. & KELLEY P.M. (1983). - Shotguns, blowguns, and spears: the analysis of technological efficiency. In Adaptive responses of native Amazonians (R.B. Hames & W.T. Vickers, eds). Academic Press, New York, 189-224.

163. ZIMMER H.R. (1948) - Hindu medicine. Johns Hopkins Univ. Press, Baltimore, Maryland, 201 pp.