Social organization

The female-centered or matriarchal nature of elephant society was recognized by some astute naturalists of the nineteenth century, in spite of numerous references in the colonial hunting literature to "master bulls" and their likes leading elephant groups. The description of an elephant hunt in 1863 by Victor Brooke and Douglas Hamilton in southern India's Biligirirangans, for instance, alludes to a male-led elephant society. A few years later, however, G. P. Sanderson (1878, p. 46) was to record, from observations in the same region, that an elephant "herd is invariably led by a female, never a male." He considered this a "necessity for the convenience of the mothers of the herd regulating its movement... as they must accommodate the length and time of their marches, and the localities in which they rest or feed at different hours, to the requirements of their young ones; consequently the guidance of a tusker would not suit them." Further, he clearly recognized that "each herd of elephants is a family in which the animals are nearly allied to each other." Even earlier, in 1856, the hunter James Chapman had noted in his diary (published in 1971) that the southern African elephants that they are "gregarious and run in herds which separate into families and meet again . . . the males are found separate from the female during most of the year" (quoted in Spinage 1994, pp. 129130).

A clearer picture of elephant social organization began to emerge from the early scientific studies in East Africa and Sri Lanka. From the population structures of culled elephants in Uganda and Kenya, Richard Laws and his associates recognized that elephants moved in social groups that represent family units comprising closely related individuals. They even constructed kinship diagrams for several cropped herds based on age-sex composition and occurrence of placental scars in adult cows.

Iain Douglas-Hamilton's pioneering study of elephant social organization and behavior at Lake Manyara National Park was the first indication, based on following identified individuals and groups over 4 years, that a stable family unit was the core of elephant society. He defined this family unit as comprising several adult cows and their offspring, including daughters of all ages and sons of prepubertal age. The average size of a family unit was 10 at Manyara. Further, he described a higher level of organization—the kinship group—in which two or more family units had close ties and were often seen together. The formidable matriarch Boadicea's social groups exemplified this situation. While this matriarch led her own family unit (or extended family) of 22 elephants, including 6 breeding cows, she also seemed to be the leader of a larger kinship group that included two other family units, a total of about 40 ele phants during the first year. He also observed the process of splitting of Boadi-cea's family during the third year, when one subgroup of an adult cow, her young mature daughter, and their three offspring increasingly began to move independently.

The high degree of stability in the family units or extended families at Manyara was not seen in other parts of Africa, such as Sengwa and Amboseli, during subsequent studies. There, the only stable unit seemed much smaller, consisting of only one adult cow and her immature offspring. The longest-running study of elephants anywhere, beginning in 1972 at Amboseli by Cynthia Moss, has given us the most detailed picture of the social relationships of females and their families. Moss visualizes elephant social organization as radiating from the family unit "through a multi-tiered network of relationships encompassing the whole population" (1988, p. 125). Several levels of organization emerge along this axis from the basic unit to the population (fig. 4.20).

Moss illustrates this with the example of the EB family, whose matriarch is Echo, popularized through a television film. In early 1983, the Amboseli

Figure 4.20

Levels of social organization in African elephants at Amboseli, Kenya, depicted through a multitier network of relationships for the adult female Echo. (From Moss and Poole 1983. Reproduced with the permission of Blackwell Publishers Ltd., U.K.)

Figure 4.20

Levels of social organization in African elephants at Amboseli, Kenya, depicted through a multitier network of relationships for the adult female Echo. (From Moss and Poole 1983. Reproduced with the permission of Blackwell Publishers Ltd., U.K.)

elephant population stood at 615 elephants—451 elephants in 48 family groups and 164 adult bulls. Echo's mother-offspring unit included three offspring, while her family unit was larger at 10 individuals, including three mother-offspring units. The EB family had been seen in association with each of the other 47 family units in the population, although infrequently with the majority of them. With 19 of these family units, the EB family was seen for more than 15% of the total sightings. Of these, associations with 18 units occurred on 15%-29% of occasions, while with 1 unit, the EA family, this was a significantly higher 53% of sightings. Apart from the spatial closeness of these two families, these elephants also greeted each other with an intensity quite different from greetings with other units. When the two families met after a separation of several hours, their greeting ceremony was accompanied by deep rumbling, trumpeting, screaming, and loud ear flapping. The EA and EB family units thus comprised a bond group, similar to the kin group of Douglas-Hamilton, but without necessarily implying close relatedness between the associating families.

During the dry season at Amboseli, the elephant families were found clustered at several different regions of the park. The families within any single congregation seemed to form a higher level of organization—the clan. This could have been a simple geographical separation of families when dry season home ranges shrank. But the families of a clan have been observed showing aggression toward families of other clans if these intruded into their dry season range. This suggests a biological significance in that matriarchs of higher social rank may defend scarce resources successfully. On the other hand, the families within each clan lived amicably, although they did not necessarily greet each other or form close bonds. Echo was part of a clan that included nine family units. About seven clans utilized Amboseli entirely or partly during the dry season.

The population of elephants at Amboseli could be further resolved into two subpopulations, one of which congregated around the swamps (the "central subpopulation"), while the other was a "peripheral subpopulation." The home ranges of clans overlap considerably. During the wet season, the elephants grouped into larger aggregations and ranged more widely, including over areas outside the park boundaries.

A similar basic social structure has been observed in Asian elephant populations in the dry forests of Sri Lanka and India, although this species has been studied in less detail. The early Sri Lankan studies, by observers like George McKay and Fred Kurt, while recognizing the existence of families of related individuals, used the term herd to refer loosely to any elephant group. They further differentiated these into "nursing units" and "juvenile-care units" with a certain degree of spatial separation. Distinct levels of social organization were not explicitly discerned (indeed, they may not exist here).

Charles Santiapillai and associates, also working in southeastern Sri Lanka, used the term group to refer to any aggregation of animals, but used herd to denote a more cohesive unit of adult cows and offspring. They opined that the larger groups were random associations of the herd units.

My observations in southern India suggested that elephant social groups could be categorized into families, bond groups, and perhaps even clans. The stable family units, however, were small and consisted of only one or two adult cows and their children (fig. 4.21). I therefore suggested that the term family is restricted to a single adult cow and offspring unit, while another term, joint family, be used to describe larger groups comprising two or more adult cows plus offspring even if these are relatively stable. The existence of bond groups was suspected, but was not based on quantitative data. Over a geographical area of about 1,000 km in the Biligirirangans, I also noted at least five distinct clusters of elephant groups during the dry season. Four of these congregations comprised between 50 and 125 elephants each and corresponded to the clans described at Amboseli. A fifth congregation in a river valley to the southwest of the study area was much larger, with over 200 elephants. This may have been a subpopulation comprised of two or more clans. Baskaran's detailed study of three adult females through telemetry in the Nilgiris also suggested the existence of clans of up to 65 elephants each with overlapping ranges.

Figure 4.21

A family of Asian elephants comprising the matriarch (extreme left) and her five offspring of various ages in the Biligirirangans, India.

Figure 4.21

A family of Asian elephants comprising the matriarch (extreme left) and her five offspring of various ages in the Biligirirangans, India.

The few observations in tropical moist forests hinted at a much simpler social organization among elephants in close-canopied forest (fig. 4.22). Confirmation that this is indeed so has come from several years of observations of the African forest elephant (L. a. cyclotis) by Andrea Turkalo at a small forest clearing, the Dzanga Bai, in the Central African Republic. Elephants can be seen in this clearing during the late afternoon and night practically every day. An identification file of these elephants has been maintained since 1990 at this site. The overwhelming majority (89%) of family groups that visit the site are units comprising only one adult cow with one, two, or occasionally three offspring. Interestingly, the next most frequent category (10%) has been of solitary females that have passed the reproductive age. Units of two or three adult females with their offspring are seen very rarely. Although there are indications of some form of association or bonding between family groups in a few cases, this is by no means common or evident from behavioral observations at the clearing, unlike the case of savanna elephants. Greeting ceremonies between families have never been observed. There is, however, intense socializing among elephants that gather at the clearing. Much of the social interactions observed here are related to dominance bouts such as at a water hole. Over 2,000 elephants, or a majority of the population in the contiguous Dzanga-Sangha Reserves, visit this clearing; indeed, they may visit several other clearings in the region. Most of the identified groups have been sighted only 1-5 times at the clearing over a 5-year period, indicating a high rate of turnover and perhaps no distinct pattern of movement.

Asian elephants in the rain forests of Peninsular Malaysia again do not seem to have any higher level of organization than the family or bond group. Mohamad Khan's few observations here point to smaller family group sizes than with the populations in the South Asian dry forests, although the groups seem larger than those encountered in the moist forests of Central Africa.

Social relationships of male elephants also can be expected to vary across habitats and populations. The dispersal of a young male from its natal family is related to the age of puberty. This also effectively reduces the chances of inbreeding, although some form of kin recognition through pheromones may also play a role, as does social separation. At Amboseli, the males were observed to become independent of their families at any time between the ages of 10 and 19 years, with an average age of 14 years. In southern India, I also observed most males leading an independent life between ages 10 and 15 years. The dispersal of a young bull may depend both on behavioral development of the bull and on the degree of intolerance shown by its mother or other adult cows in the family group. This is a subject that needs further study.

It is generally agreed that the separation of a male from its family is a gradual process and not an abrupt one. As a male grows older, it spends more and more time in all-male groups or even briefly associates with strange families. By about 14 or 15 years old, the young males at Amboseli spend over 80% of their time away from their families, a stage that can be considered as

being independent. Adult females have also been observed showing aggression toward pubertal males, which may hasten their decision to leave the family.

Once a bull disperses from its family, it may range solitarily or in the company of other bulls. Solitariness seems to be the rule in dense forest such as in the Dzanga-Sangha Reserves. Any association between bulls at the clearing lasts for only a few minutes or at most a day. Among Asian elephants in dry forest, the bulls again range mostly solitarily. Data on bull group frequency distributions from Sri Lanka and southern India show that sightings of solitary bulls are by far the most common.

Bulls do associate in groups of two or occasionally more individuals (fig. 4.23). The largest bull group recorded by George McKay in Sri Lanka was seven animals. In southern India, I have observed associations of over two bulls only rarely in several hundred sightings of bull groups, with the largest group being a loosely spaced congregation of eight bulls along the banks of the Kabini Reservoir in the Nagarahole National Park. While there was no evidence for any special bonding between bulls seen in the forest during the day, some bulls associated in pairs for up to a month while raiding agricultural fields at night (see chapter 8).

Figure 4.23

An association of four Asian bull elephants at Nagarahole National Park, India. Such associations are usually quite temporary.

Figure 4.23

An association of four Asian bull elephants at Nagarahole National Park, India. Such associations are usually quite temporary.

The larger bull groups of over 10 individuals have been recorded exclusively in the African savannas (the largest bull group on record seems to be one of 144 individuals seen by Ian Parker at the Galana Ranch of Kenya). Here, too, the typical association involves only two or three bulls. The nature of the social relationship among associating bulls has naturally attracted the attention of elephant observers. Iain Douglas-Hamilton found the bull groups at Manyara to be very unstable. The longest association of two bulls was merely 14 consecutive days. Bull groups were of ever-changing composition with no evidence of any lasting ties.

The influx of elephants into the Serengeti National Park during the 1960s and, in particular, the destruction of larger trees in the Seronera area by bull groups attracted much attention. These bull associations were studied by H. Hendrichs (cited by Croze 1974a) and later by Harvey Croze. The Seronera bull groups typically had one bull distinctly larger and presumably older than the rest. A close examination of associations among 25 identified bulls showed that, in statistical terms, these were not different from chance associations except in two instances. One such pair spent 51 days moving together, fed "on the same bushes simultaneously, rolled in the same wallows, and even scratched themselves against one another" (Croze 1974a, p. 8). Such observations obviously lend credence to traditional descriptions of friendship bonds between bulls (in northeastern India, such associations, termed locally as mal-juria, are known from many anecdotal accounts). Eventually, these bulls went their own way.

In Amboseli's more sedentary elephant population, there are stronger hints of associations among bulls. Although the males here do not form stable groups, several pairs have been seen in association; over 30% of the time, each had been observed in the company of a male.

Irrespective of whether the adult bulls range solitarily or in association with other bulls, a complex pattern of changing dominance relationships exists among them. The social relationships depend on frequency of encounter (which in turn is related to habitat, elephant density, and ranging patterns), body size, age, and most important, the sexual state of the bulls. When a bull comes into a state of heightened sexual activity associated with musth, it moves solitarily in search of estrous cows or is seen in the company of female groups (chapter 3). The six oldest bulls at Amboseli were seen solitarily 29% of the time and in association with cow groups 66% of the time on average when in musth. When they were not in musth, the same bulls were seen solitarily 26% of the time and in the company of other bulls 68% of the time. While bulls that are larger or older are generally more dominant, the situation may be reversed when a smaller or younger bull comes into musth. A musth bull shows increased aggression toward other bulls. When bulls are not in musth, they may associate peaceably in bull areas. During such times, the less-aggressive sparring bouts may help decide dominance hierarchies (fig. 4.24). Joyce Poole has suggested that the relationships also may be based on relatedness among bulls. Even when bulls move solitarily, as in tropical moist forests, their visits to

Figure 4.24

Two Asian elephant bulls at Nagarahole National Park, India, lock tusks in a dominance contest. Such contests rarely escalate into a serious fight.

Figure 4.24

Two Asian elephant bulls at Nagarahole National Park, India, lock tusks in a dominance contest. Such contests rarely escalate into a serious fight.

clearings or mineral licks (such as at Dzanga) provide an opportunity to meet and socialize with other bulls.

Finally, we should take a brief look at adult bull associations with cows (this theme is the subject of chapter 3). Male-female associations in elephants are predominantly for the purpose of reproduction. Adult males can successfully mate when they are out of musth, but are far more likely to do so when in musth. At Amboseli, the six oldest bulls that spent two-thirds of their time with cow groups when in musth did so only 7% of the time with cows when out of musth. The available evidence suggests that bulls associate with family groups at random, depending on the presence of an estrous cow. It has been suggested that both cows and bulls may have their preferred partners. This is likely in a long-lived species for which individual recognition plays an important role in social relationships. It would take years of observation of a large number of identified elephants to confirm or reject this speculation.

Before concluding this discussion of male-female relationships by rejecting the "herd bull" concept, we must consider an interesting theoretical possibility. Using a plausible mathematical formulation, Richard Barnes has argued that, in regions of low elephant density, it would be advantageous for an adult bull to be attached to a female group for an extended period of time. Perhaps the last word has not yet been said on this subject.

Was this article helpful?

0 0

Post a comment