Mongooses are found all over Africa and have managed to make a living in just about every terrestrial habitat. Most African mongooses are small carnivores that eat smaller vertebrates in forests or thickets, but several species have moved into open grasslands to eat insects. Eating insects in grasslands is associated with a suite of behavioral and social characteristics that are very attractive to modern biologists. These insect-eating species live their entire lives in well-organized groups composed of a dominant breeding pair, their young offspring, and an assortment of related and unrelated subordinates. The presence of other group members decreases the risk of predation while foraging and provides the breeding pair with assistance in raising offspring.

At least one species fails to conform to this social pattern for insect-eating mongooses. The yellow mongoose (Cynictis penicillata) lives in the veld and bush of southern Africa in a range that covers most of the Republic of South Africa, Botswana, and southern Namibia. Although yellow mongooses live in stable social groups at the den, they forage alone. At first glance, this appears to violate the established story for the evolution of sociality in mongooses. Their habits are particularly interesting because of their apparently close phylogenetic relationship to two of the more famous social mongooses, the meerkat (Suricata suricatta) and the dwarf mongoose (Helogale parvula).

Yellow mongooses belong to the taxonomic family Viverridae, which encompasses four subfamilies: the Malagasy mongooses (the Galidiinae) the civets and genets (the Viverrinae), the palm civets (the Paradoxurinae) and the mongooses (the Herpestinae). The common ancestor of the Viverridae is thought to have split from the common ancestor of the hyenas only about 25 million years ago (Ewer 1973). Viverrids are therefore relatively recent arrivals on the mammalian scene and they have since undergone an explosive adaptive radiation into 18 genera and 31 species (Corbet and Hill 1991).

The nomenclature and phylogenetic status of the Herpestinae have been revised often over the years. A variety of authors have attempted to construct phylogenies from morphological and behavioral characters, which usually results in two clusters of species (cf. studies digested in Taylor et al. (1991). The social species share characteristics of dentition, habitat, and behavior, so they form one cluster. By these criteria, yellow mongooses are placed in the other cluster with the solitary and semi-solitary species. This split is traditional and reliable, but it is not confirmed by phylogenies based on less evolutionarily flexible characters. Using biochemical traits, which are more reliable than physical or behavioral characters, Taylor et al. (1991) constructed a family tree that placed meerkats, dwarf mongooses, and yellow mongooses together in an unresolvable triad. To date this phylogeny has not been confirmed using selectively neutral genetic markers, which should be the most reliable of all.

The close evolutionary relationship between yellow mongooses and meerkats is interesting because of their different lifestyles, but it is even more interesting because the two species are sympatric throughout most of their range. Yellow mongooses and meerkats both forage in the same grassland habitats in southern Africa, but additionally they can be found denning in the same

tunnel systems (Lynch 1980). Unfortunately (but sensibly), few field studies have looked at behavioral or ecological relationships and interactions between yellow mongooses and meerkats. Anecdotally, Lynch (1980) has reported that yellow mongooses and meerkats ignored each other at the den, and his diet data indicate that they are probably not in intense competition for food.

The ecological circumstances thought to lead to group living in social mongooses such as meerkats relate to the costs and benefits of foraging in groups, but yellow mongooses do not forage in groups. It could be that this difference is an evolutionary artifact. Yellow mongooses may have followed the standard story for the evolution of sociality then lost the group-foraging characteristic. Alternatively, yellow mongooses may only have recently moved to grassland habitats, at which point they acquired the trait of denning in groups.

As in other social carnivores, sociality in mongooses is associated with a particular diet that both requires and facilitates group living. Large carnivores, such as wild dogs (Lycaon pictus), wolves, and lions, hunt large game that can only be efficiently subdued by a team of cooperating hunters. Except when they are young and vulnerable, these carnivores are themselves too big to be frequently taken as prey. Unlike their larger relatives, mongooses are quite small (700 to 900 grams) and, even as a team, cannot realistically capture a single large meal. Instead, most mongooses make a living by hunting small prey such as rodents and other vertebrates, which are accessible and abundant in African forests and bush.

A diet of small prey is associated with solitary living in the carnivores as a group (Gittleman 1989), perhaps in part because they require quiet hunting (Ewer 1973). One would expect that mongooses would hunt these mouthfuls under circumstances that minimize their own chances of being eaten by middle-sized predators such as snakes, raptors, and large lizards. Fortunately for the mongooses, small vertebrates are found in greater numbers in forests and bush areas, where a hunting mongoose has better cover from these predators, than in the insect-rich grasslands, where raptors in particular have an easy shot at a mongoose dinner (Macdonald and Nel 1986).

In spite of this danger, there are good livings to be made in grasslands, and accordingly the adaptive radiation of the mongoose family has produced several species that do forage in open grassland or semi-wooded grassland. Some mongooses find protection by hunting when their predators are handicapped by darkness or cold. The (mostly) solitary white-tailed mongoose, found in the north eastern African plains, hunts small vertebrates and insects in scrubby grassland, but it is nocturnal and thus somewhat protected from predators by cover of night (Waser and Waser 1985). To exploit grassland insects fully, however, a forager must be out when insects are active, and this means facing the dangers of grasslands during the day.

Although an insect diet invites predation, it also provides the key to avoiding predation by permitting several animals to forage in close proximity. Ground-dwelling insects are apparently not very responsive to noise, so the presence of one foraging mongoose nearby is unlikely to affect the success of another (Ewer 1973). In addition, insects are superabundant during certain wet months, so a given area of habitat can support several resident animals. Much like a flock of birds or herd of antelope, each group-foraging mongoose gains the benefit of additional animals to watch for predators, as well as a diluted chance of being the one animal in the group that gets killed should a predator have a chance to attack.

The first groups to form in this way were probably family groups, either mothers who permitted their daughters to stay with them in their territories, or sisters who continued to share a territory after their mother's demise (Waser and Waser 1985). Related white-tailed mongooses who share territories also den together, and presumably any group that forages together will den also den together. At some point males may also have been tolerated in the group. This process of aggregation is the most interesting part of the evolution of sociality, but it is a process that leaves no historical record.

In summary, sociality in small carnivores is favored when a diet of abundant and dispersed insect prey leads to high predation pressure that can be mitigated by group vigilance. Yellow mongooses, as stated earlier, forage alone for insects in open grassland during the day, but return to a common den to rest as a group. Could yellow mongooses have developed a group-foraging social system through offspring retention and predator avoidance, but then have lost the group foraging part while retaining their habit of denning in groups?

If the trait "group-foraging-for-insects" were developed but then lost, one would expect that the dentition of yellow mongooses would show evidence of an insectivorous ancestry and therefore would be adapted for crushing insects (i.e., blunter molars and canines). Instead their teeth are designed for shearing flesh, which is the standard ancestral carnivore type (Ewer 1973). This discordance is hardly a reason to discard the theory that yellow mongooses were once group foragers, but it is not encouraging.

Denning as a group has advantages of its own, however, that may have lead to sociality at the den without requiring sociality at the table, so to speak. In harsh habitats, a mother, even with the assistance of her mate, may not be able to rear a litter successfully. If offspring from previous litters are permitted to stick around after becoming independent, they are available to help their parents rear the next litter by guarding the new litter against predators, or by provisioning them. The rules that govern which animals should help, how much they should help, and when they should stop helping and leave the group to seek their reproductive fortunes elsewhere, are too complex to be done justice here. Suffice it to say that the more closely related a helper is to the new litter, the more it should help. At some point, however, no amount of this indirect reproductive success will substitute for a chance actually to breed, and if these chances are not forthcoming in the current group, the animal should leave for more reproductively favorable pastures. Those interested in kin selection, dispersal, and enlightened self-interest in mongooses should consult Creel and Waser (1994), Keane et al (1994) or a host of others.

Distinguishing between these two paths to partial sociality will require more information on the rearing of young, foraging, and predation pressure while foraging and at the den. To date, there is simply not enough information to explain why yellow mongooses exhibit social denning, but not social foraging. What is known about the behavior and habits of the yellow mongoose is interesting in itself and in comparison to the meerkat.

Yellow mongoose life centers around the group den, which may be in one of several suitable sites contained in the group's defended territory. Dens are useful for avoiding predators and extremes of temperature. A yellow mongoose's day begins about half an hour before dawn when it leaves the den for a quick trip to the colony midden to relieve itself. It then retires back to the burrow until half and hour after dawn, when the whole colony emerges and spends up to one and a half hours sunbathing. Yellow mongooses expose their bellies to the sun by lying with fur erected, or by standing or sitting with the belly exposed. During this morning time at the den, some animals may do a little burrowing, groom themselves, or engage in mutual grooming with another yellow mongoose (Cavallini 1993b, Early 1981). The main activity, however, is warming up in the sun, so most social interactions occur in the evening (Cavallini 1993b).

At around nine o'clock in the morning, the mongooses set off separately to forage. Yellow mongooses spend about a third of their active hours foraging and travel through about a third of their entire home range in a day (Cavallini 1993b). The territory of a colony is usually around 80 ha and animals move about 3 km in a day when foraging. If threatened, a yellow mongoose may bolt as much as 150 meters to a burrow or low hiding spot (Macdonald and Nel 1986), although no predation events have been described in the field.

Having fasted all night, a mongoose is quite hungry in the morning when it begins foraging. Earle (1981) reports that his study animals took smaller and less preferred prey at the start of a day's foraging, but that they became more selective as the day went on. The preferred food in the diet is termites, but vertebrates, beetles, carrion, etc. also regularly appear in the diet (Earle 1981, Herzig-Straschil 1977).

Because yellow mongooses forage alone and have retained the ancestral carnivore dentition, it is usually assumed that they eat more vertebrates than group-foraging mongooses. Diet data have been collected at a number of sites, (see Figure 2) but only one study (Cavallini and Nel 1995) has directly addressed and confirmed this hypothesis. Diet studies have turned up a wide range of diet compositions in various locations. Table 2 presents comparable studies at different locations for inspection; these studies confirm that yellow mongooses are indeed opportunistic forages that adjust their diets to location and season. More studies are needed at more diverse sites before the relationship between diet and habitat can be resolved for yellow mongooses. Only one study, by Avenant and Nel (1992), has examined prey abundance as well as diet composition. Without prey abundance information from other sites, it is impossible to say whether a diet at any particular location reflects prey incidence or predator preference.

Many authors have reported that yellow mongooses will return to the den at midday if the weather is warm. Cavallini's (1993b) radio-collared animals spent up to between one and two hours in the den at midday when the air temperature exceeded 24C. After this siesta, the animals set out foraging again and returned around six o'clock According to Cavallini's studies, the mongooses began returning to the den at three o'clock in the afternoon.

If one or more females in the colony have dependent offspring, this schedule is different. A female subordinate will stay behind with the young while the mother forages, and another female will relieve the first baby-sitter after a few hours. Males assist with the provisioning of the young once they are weaned. It is not clear how these duties are assigned, or how guards are relieved of their duties. I have not found any studies that examine parental behavior in yellow mongooses in detail and this is certainly an area ripe for observational studies.

During the first few months of life the young are dominant to all group members other than their parents, although their status slips at ten months when they become young adults and a new litter is born (Earle 1981). Weaning occurs at six to eight weeks, after which the young are guarded and provisioned at the den until they are ready to forage themselves (Rasa et al. 1992). I know of no studies of whom the offspring forage with when they start, or whether provisioning involves the interesting shifts in parental feeding behavior described in meerkats by Ewer (1963)

This schedule is a generalization among several studies conducted at different sites. Cavallini worked in the West Coast National Park, on the Langsbaan peninsula, while Earle worked on an island in the Vaalsdam reservoir in the Vaal river in central South Africa. Yellow mongooses at these two sites differ in diet composition and possibly in territorial structure.

Cavallini (1993a) has reported that male and female Langsbaan yellow mongooses have separate territorial systems such that the two sexes have overlapping home ranges. Groups in this area consist of about five males and females denning together. Males have ranges that are about four times the size of females' ranges. There are more small vertebrates in the diet here than elsewhere, and it is conceivable that a diet rich in vertebrates would require looser groupings of animals. In absence of more data on prey density, diet and habitat, the potential effect of diet on groupings and territoriality is unclear.

In the Vaalsdam, the mongooses live in colonies about eight individuals, with a dominant breeding pair, their young offspring, and assorted subordinate adults. The diet at this site is mainly termites with an assortment of other insects (mainly beetles) and small mammals (Earle 1981, Lynch 1980). In contrast, at Langsbaan, although diet again contains mainly termites, it also contains nearly 30% small mammals (Avenant and Nel 1992).

Research at Vaalsdam has also turned up interesting observations about scent marking, status, and sex. As a group, the Viverridae are smelly animals with facial and anal glands that produce sebaceous secretions of varying degrees of pungency. In the dwarf mongoose, the facial and anal secretions convey different types of information. The anal secretion provides cues to individual identities, whereas the facial scent conveys information about the excitatory state of the animal (Rasa 1973). Yellow mongooses use these glands to mark various sites and objects (Wenhold and Rasa 1994). Anal marks are left by pressing the scent sacs on the ground and anal scent also ends up on feces as they are deposited on the colony's midden. Yellow mongooses pick up the generalized scent of the colony by wiping their flanks against these middens. This sideswiping behavior is then used again to transfer the colony scent from the coat to another object. Anal marks are deposited throughout the colony's territory, but a yellow mongoose will show specific patterns of marking that depend on its status in the group (Wenhold and Rasa 1994). All animals in the group will anal mark near the den and the dominant male and female do most of their marking within the territory. This suggests that marking within the territory simply indicates proprietorship.

The situation at the territory boundary is quite different. All animals leave marks at the boundary, and it is an exciting place even for juveniles, since very young animals leave many cheek marks. Subordinate males indicate group ownership of the territory by depositing the fecal-derived group scent by sideswiping. This is sensible since subordinate males appear to be responsible for territory defense, and thus have a stake in indicating territory ownership.

Scent marking at the territory boundary serves another purpose. Both subordinate males and subordinate females leave their personal anal marks at the boundary, but the dominant animals rarely mark extensively at the border. Wenhold and Rasa suggest that anal marking by subordinate animals advertises the age, sex, and reproductive status of the subordinates to other subordinates in adjacent territories. Subordinate females frequently cross territory boundaries during the breeding season (in August through December) to mate, and these anal marks may serve to indicate who is available in the territory for these extramural affairs. In addition, since subordinate males and females are most likely permanently to disperse into neighboring groups in search of breeding opportunities, individual scents placed at boundaries may serve to introduce potential immigrants to the neighboring group.

Subordinate males and females see out clandestine matings because the dominant pair will not permit them to breed within the group. In contrast to other species of social mongooses, yellow mongoose dominants do not, however, suppress breeding between males and females from different groups. This contrast may be due to a critical difference in breeding biology.

Yellow mongoose females have two estrus cycles in each breeding season, which gives each female the potential to produce two small litters of one to three offspring each. The second litter is environmentally suppressed if conditions are harsh and food is scarce. Rasa et al (1992) suggest that two smaller litters in rapid succession may be favored in times of food shortage if breeding cannot be suppressed. If absence of breeding suppression is considered the ancestral trait in yellow mongooses, then multiple small litters within a breeding season may be the only way to ensure that there will be enough food available for the young at any one time during the breeding season.

In the social mongooses, and indeed in many social carnivores, young animals frequently leave the groups in which they were born to join another group. In general, these dispersing animals are seeking breeding opportunities that may be closed to them in their current groups. Rood (1990) has proposed that there are five ways in which a group-living subordinate can attain breeding status. If the animal chooses to remain in its natal group, it can simply wait around to become a breeder, or it can evict the same-sex dominant. It can join another group as a subordinate and wait for a breeding position to become vacant. It can emigrate and forcibly evict a breeding member of another pack. Lastly, it can join other transient mongooses and form a new pack.

From the point of view of resident animals, these immigrants threaten to bump them from their place in the breeding queue, or even to usurp the breeding positions immediately by taking over. Since there are separate breeding queues for males and females, resident males should put up the greatest resistance to immigrant males, and resident females should resist immigrating females. This is in fact the case in meerkats (Doolan and Macdonald 1996), but in yellow mongooses the situation is more complex. Intruding males are chased away by resident subordinate males as one would expect, but intruding females are tolerated and the resident females apparently do not attempt to repel them (Wehnold and Rasa 1994).

During estrus, these intruding females will mate with resident males, then immediately leave the territory. Because these intruding females do not join the group, they do not endanger the reproductive success of the dominant pair. Because they do not threaten the resident females, they are not driven away. Unfortunately, we do not have enough information in dispersal an juvenile survival in yellow mongooses to assess the impact of these permissive behaviors on group structure and breeding opportunities. The comparable studies of meerkats are just coming out.

With recent intensive study, the social behavior and group structure of meerkats has been found to resemble that of dwarf mongooses far more than was initially expected. This raises the interesting question of which evolutionary path has lead to group living in yellow mongooses as compared with meerkats. Did sociality arise once in the ancestral lineage, only to be lost in yellow mongooses, or did sociality arise in two different ways in each species because of different selective pressures and environmental opportunities? Will the parental behavior of yellow mongooses be more like that of meerkats, or more like that of solitary mongooses? Does social organization track diet closely, such that the oddities of male and female territoriality in the West Coast National Park are explained by the higher abundance of rodents? Or is the yellow mongoose simply an opportunistic forager whose social organization is unstable?

The answers to these questions will not be found close to home. Yellow mongooses are plentiful in southern Africa and adapt reasonably well to captivity (Earle 1981), but they are rarely studied outside their endemic range. There are no captive colonies in the U.S., and the only animals I have seen are in the Regent's Park Zoo in London. It is encouraging that the Republic of South Africa has recently become a more politically attractive place to work. This, and the increasing interest from sociobiologists in these appealing and interesting animals, bodes well for future field research.
 

LITERATURE CITED

Avenant, N.L. and J.A.J. Nel. 1992. Comparison of the diet of the yellow mongoose in a coastal and Karoo area. S. Afr. J. Wildl. Res. 22: 89-93.

Cavallini, P. 1993a. Spatial organization of the yellow mongoose Cynictis penicillata in a coastal area. Ethology, Ecology and Evolution 5: 501-509.

Cavallini, P. 1993b. Activity of the Yellow mongoose Cynictis penicillata in a coastal area. Z.Saugetierkunde 58: 281-285.

Cavallini, P and JAJ Nel. 1995. Comparative behaviour and ecology of two sympatric mongooses (Cynictis penicillata and Galerella pulverulenta). S. Afr. Tydskr. Dierk. (S. Afr. J. Sci.) 30(2): 46-49.

Corbet, G.B. and J.E. Hill. 1991. A World List of Mammalian Species. Oxford University Press, Oxford.

Creel, S.R. and P.M. Waser. 1994. Inclusive fitness and reproductive strategies in dwarf mongooses. Behav. Ecol. 5: 339-348.

Doolan, S.P and D.W. Macdonald. 1996. Dispersal and extra-territorial prospecting by slender-tailed meerkats (Suricata suricatta) in the south-western Kalahari. J. Zool. Lond. 240: 59-73.

Earle, R.A. 1981. Aspects of the social and feeding behaviour of the yellow mongoose Cynictis penicillata (G. Cuvier). Mammalia 45(2): 143-152.

Ewer, R.F. 1963. The behaviour of the meerkat, Suricata suricatta (Schreber). Z. Tierpsychol. 20: 570-607.

Ewer, R.F. 1973. The Carnivores. Cornell University Press, Ithaca.

Gittleman, J.L. 1989. Carnivore group living: Comparative trends. Ch. 7 in Carnivore Behavior, Ecology, and Evolution, J.L. Gittleman, ed.

Herzig-Straschil, B. 1977. Notes on the feeding habits of the yellow mongoose, Cynictis penicillata. Zoologica Africana 12(1): 225-256.

Lynch, C.D. 1980. Ecology of the Suricate, Suricata suricatta and the yellow mongoose Cynictis penicillata

with special reference to their reproduction. Mem. Nat. Mus. Bloemfontein 14: 1-145.

Keane, B., P.M. Waser, S.R. Creel, N.M. Creel, L.F. Elliot, and D.J. Minchella. 1994. Subordinate reproduction in dwarf mongooses. Anim. Behav. 47: 65-75.

Macdonald, J.T. and J.A.J. Nel. 1986. Comparative diets of sympatric small carnivores. S. Afr. J. Wildl. Res. 16(4): 115-121.

Rasa, O.A.E. 1973. Marking behavior and its social significance in the African dwarf mongoose, Helogale undulata rufula. Z. Tierpsychol. 32: 293-318.

Rasa. O.A.E., B.A. Wenhold, P. Howard, A. Marias, and J. Pallett. 1992. Reproduction in the yellow mongoose revisited. S.-Afr.Tydskr.Dierk 27(4): 192-195.

Rood, J.P. 1990. Group size, survival, reproduction, and routes to breeding in dwarf mongooses. Anim. Behav. 39: 566-572.

Taylor, P.J. and J. Meester. 1993. Cynictis penicillata. Mammalian Species 432: 1-7.

Taylor, P.J., G.K. Campbell, J.A.J Meester, and D. van Dyk,. 1991. A study of allozyme evolution in African mongooses (Viverridae: Herpestinae). Z. Saugetierkunde 56: 135-145.

Waser, P.M. and M.S. Waser. 1995. Ichneumia albicauda and the evolution of Viverrid gregariousness. Z. Tierpsychol. 68: 137-151.

Wenhold, B.A. and O.A.E. Rasa. 1994. Territorial marking in the Yellow mongoose Cynictis penicillata: Sexual advertisement for subordinates? Z. Saugetierkunde 59: 129-138.