MATING SYSTEMS (Concluded)

Polyandry is also a well-established mating system in a few human societies. It occurs in Tibet. Sons inherit the family's land and livestock, and these resources typically aren't divided. The sons, as a group, marry an unrelated female (often a young teen). The males then cooperate to maintain the farm. This system results in a large number of single women in Tibetan society. Some of them do have children, but their average fecundity rate is only about a third of the average fecundity of a married woman. The reason for this system seems to be control of resources. There is only a limited number of areas where agriculture is possible (the primary crop is barley), and all these areas are already owned. There are major population consequences; reduced fertility of nonmarried females has resulted in a stationary population size.

Summary

In general, the tendency towards polygyny or polyandry can be affected by differences in female spacing patterns, temporal variability in their receptivity, and in maternal and paternal care. If paternal care is more extensive, polyandry is possible (e.g., phalaropes). If maternal care is more extensive, polygyny is possible. The degree of polygamy then depends on the operational sex ratio (availability of unmated males in polyandrous systems and unmated females in polygynous systems). This framework works reasonably well for birds and mammals but not for fishes.

In most fishes there is no parental care. In the minority of fishes with care (guarding and fanning of eggs), female care tends to occur only when there is internal fertilization and male care is correlated with external fertilization (see Alcock for additional discussion). In male care systems, however, polygyny is prevalent -- males guard nests where females come and lay the eggs. Because a male can guard multiple clutches and male nests are limited resources for females, polygyny occurs. This apparently is an unlikely scenario for birds and mammals because the quality of parental care probably declines considerably as the number of offspring increases, i.e., care is "divisible" in fishes without decreasing probability of offspring survival but not in birds or mammals.

Alcock concludes the chapter on mating systems with a discussion of the variable mating system of the dunnock. Females establish breeding territories and then males overlap their territories on those of one or (rarely) more females. Occasionally a male can defend a territory which encompasses the territories of two females, resulting in polygyny. Sometimes a male is unable to maintain exclusive use of a single female's territory, particularly if her territory is large. If a second male arrives, the female actively solicits copulations with him as well as with her first mate; males that mate with a particular female will feed her offspring. On such territories the mating system is polyandrous. Occasionally, females on two neighboring territories successfully solicit copulations from the neighboring males, resulting in polygynandry. Such variability emphasizes that mating systems are the outcome of strategies and counter-strategies of individuals rather than invariable characteristics of populations or species.

Extrapair copulations (EPCís)

In many apparently monogamous species, both the male and female may seek and achieve copulations outside the pairbond, i.e., extrapair copulations (EPCís). For males such may increase his genetic contribution to the next generation without increasing his paternal effort. There are several potential advantages to females: females may be guarding against infertility of their mates, increasing the genetic variability of their offspring, or copulating with males that are genetically superior to their mates (Allcock provides examples that fit each of the above possibilities).

In the monogamous swallow, Moller (1988. Nature 332:640-642 and 1990. Anim. Behav. 39:458-465) found that (1) males with the longest tails return first and compete most extensively for nest sites, (2) females prefer to pair with the longest tails, and (3) females that arenít paired with long-tailed males will have EPCís with them. These results suggest that females unable to pair with the most preferred males may still have them father part of the clutch. However, Moller found no relationship between tail length of males and their provisioning of offspring.

Alcock discusses extrapair copulations in detail.

COOPERATIVE BEHAVIORS

Before discussing various examples of cooperation, it is important to remember that animals in groups are not necessarily cooperative , even when they associate with kin. Even though many social vertebrates do associate with kin, these associations may be more competitively than cooperatively based. An extreme example is that of infanticide in black-tailed prairie dogs, which we studied at the beginning of the semester (Hoogland, J. 1985. Science 230:1037-1040). In the prairie dog, lactating females are highly infanticidal, often killing (and sometimes eating) the offspring of close kin. Hoogland found 73 cases of infanticide. Infanticide occurs before and after weaning, usually in burrows, and infanticidal females are almost always genetically related to the mothers. Presumably, the lactating females enhance their own fitness by reducing the number of competitors for their young in the natal coterie, where their daughters typically remain throughout their lives. Incoming males also are infanticidal on juveniles; the advantage of infanticide to males seems to be due to an increased probability of successful weaning when fewer yearlings reside in the coteries.

Definitions

By-product beneficence occurs when the selfish behavior of one individual indirectly benefits another individual. For example, if one individual in a hunting group makes a kill, others may also feed at its kill and thus benefit from its behavior. Alarm calls by an individual that cause the predator to leave the area but also alert others would be another example.

In pseudoreciprocity an original recipient of aid later behaves in a way that benefits both itself and the original donor of aid. Helpers-at-the-nest in the Green Woodhoopoe (Phoeniculus purpureus) sometimes are unrelated to the parents and nestlings. Nestlings form social bonds with individuals who feed them and sometimes join the helper once it becomes a breeder. In turn, they may be helpers at the original helperís nest. This behavior seems like reciprocity (see below), but helping seems to be an avenue to establishing alliances and thus eventually breeding.

In reciprocal altruism or reciprocity, an individual aids another at some present cost to itself, and in the future this benefit is reciprocated. Donation of blood meals by vampire bats to group members that have not obtained meals is a well-documented case of reciprocity.

Vampire bats will die if they don't recieve blood meals on two successive nights. On any one night 7-30% fail to obtain a blood meal. Roosting groups of vampire bats generally number 8-12 females with their dependent young. Males leave their natal groups and form dominance hierarchies with the dominant male roosting closest to the females. Females do switch roosting trees and groups from time to time. Females reciprocally share blood often forming alliances with nonrelatives as well as relatives (see handout; Wilkinson, G. S. 1990. Food sharing in vampire bats. Sci. Amer. Feb. 1990:76-82).

In simultaneous cooperation 2 or more individuals donate and may receive benefits at the same time. Older savannah baboon (Papio spp.) males that no longer rank near the top of breeding male dominance hierarchies sometimes solicit aid of males of similar status in challenging males who are consorting with estrus females (see Alcock). Either the solicitor or the joiner may end up consorting with the female if they are successful (Bercovitch, P. B. 1988. Anim. Behav. 36:1198-1209). This form of cooperation also probably applies to coalitions of unrelated male lions (see below).

In cooperative breeding systems nonbreeding individuals aid in raising young that are not their own. Examples include canid helper-at-the-den and avian helper-at-the-nest systems. One of the best documented examples is the Florida Scrub Jay. In communal breeding systems two or more individuals of the same sex contribute to production and raising of the young within a nest. The Groove-billed Ani is a well-studied example. Plural breeding systems occur when groups defend territories but pairs within the groups nest separately.

PRISONERíS DILEMMA

Models based on game theory suggest that a strategy of helping those who help you and refusing to help those who have refused to help you in the past can be evolutionarily stable. These models are variations on the Prisonerís Dilemma gambling game. In its original version there are 2 players and each has 2 cards, labeled COOPERATE (C) and DEFECT (D). Simultaneously, each player plays one of the 2 cards. When you play, you receive the highest payoff if you play D and the other individual plays C, but the next highest payoff is if both play C (see handout from lab earlier in the semester). (Note that one condition of the payoff matrix is that the average payoff of III and IV must be less than that for I.)

If the game is played only once the stable strategy is to defect. However, if it is played repeatedly, various forms of cooperation may evolve and a cooperative strategy may be highly rewarding. Robert Axelrod (e.g. Axelrod, R. and W. D. Hamilton. 1981. Science. 211:1390-1396) asked experts in game theory to submit strategies and then played all of them against each other, using the payoff matrix in Figure C on the handout. The strategy that did best overall was submitted by Anatol Rapoport. He submitted a strategy called ìTit-for-Tatî which begins by cooperating on the first move and thereafter simply copies the previous move of the other player. ìTit-for-Tatî is ìniceî-- it is never the first to defect. It is also ìforgivingî-- it has a short (1 interaction) memory, i.e., it doesnít hold a grudge. Although nobody submitted ìTit for Two Tatsî (allowing 2 defections before retaliation), it would have won Axelrodís first tournament. In a second round, Tit-for-Tat won again. Tit for Two Tats was submitted in the second round but didnít fare so well-- some new nasty strategies did better. Consequently, niceness and forgiveness can pay in group-living organisms, but their effectiveness can vary depending on strategies of other interacting individuals.

COOPERATIVE BREEDING (ìHelpers-at-the-nest or denî)

Cooperative breeding (nonbreeding helpers-at-the-nest-or-den) is known to occur in over 150 bird species and several mammalian carnivores. Nonbreeding helpers are adults that are physiologically capable of breeding and thus seem to be acting altruistically.

Florida Scrub Jay

Glen Woolfenden and John Fitzpatrick have shown that the system of cooperation in Florida Scrub Jays is based on selfish (and inclusive fitness) interests of the helpers. To breed, females must find a territorial male outside the natal territory who is without a mate. Females often explore surrounding areas for an available mate while living on the natal territory. In contrast, subordinant males can breed on the natal territory of the father dies, or they can form a territory of their own within the natal territory if it is large enough. Consequently, males have a more direct interest in territory size and in raising younger siblings to aid in territorial defense. Males remain on the natal territory as helpers longer than do females, and a third of the males eventually breed on the natal territory. Florida Scrub Jays are long-lived and the habitat is fully saturated, i.e., all suitable habitat is occupied by territorial pairs or groups. Individuals have little chance of dispersing as juveniles and establishing a territory. Helping is a means of remaining on the natal territory until breeding is possible.

The amount of food a Florida Scrub Jay helper delivers to nestlings depends on its position within the social unit (influenced by sex and age), not on its relatedness to the nestlings. Ligon (1983. Am. Nat. 121:366-384) termed such behavior reciprocity because the helper may later gain its fitness through the aid of the nestlings it helps. If a male helps nestlings, those nestlings may later be instrumental in enlarging the territory so that he can form his own territory.

Itís also been found that adults with helpers have lower mortality rates than those without helpers, and Woolfenden has suggested that helpers live longer when accompanied by other helpers. Thus it benefits a female helper to aid in raising nestlings as she will more likely avoid predation if she associates with younger and more vulnerable flock members (recall the ìSelfish Herdî argument).

Green Woodhoopoe

J. David Ligon (1982. Sci. Amer. 1982. Am. Nat. 121:366-384) summarized his studies of the cooperative breeding Green Woodhoopoe (Phoeaiculus purpureus), an East African species. He described a variety of mutually beneficial interactions among birds that are typically not genetically related. He emphasized the concept of reciprocity (this probably is an example of pseudoreciprocity, however). In 120 flock-years of observations, he recorded only 5 pairs with no helpers; helpers are normally a prerequisite for obtaining and retaining breeding status. Thus the breeding pair is obligated to seek the aid of the helper. In turn, the helper benefits eventually because nestlings he or she feeds may later be allies to the helper when it attempts to breed. There is intense competition among helpers to feed nestlings, including food stealing, because only by feeding a nestling does a helper establish a social bond with it. Successful moves into nearby territories are by groups, not individuals.

General Interpretations

Koenig and Pitelka (1979. Ecological factors and kin selection in the evolution of cooperative breeding in birds, Chapt. 17 in R. D. Alexander and D. W. Tinkle. Eds. Natural selection and social behavior) attempted to formulate a general hypothesis for the evolution of helpers at the nests. Virtually all species with helpers occupy permanent territories, and in many the habitat appears fully saturated with territorial groups. If there is a major difference in suitability of optimal and marginal habitats, selection will act against individuals who disperse to marginal habitats and attempt to breed. Young will be better off associating with territorially established birds, usually (but not always) their parents. Reproductive success of the breeding pair may be enhanced by the presence of helpers but generally not to the extent that either parents or helpers would benefit genetically if the young could find a breeding opportunity.

For Acorn Woodpeckers and some other species, Stacy and Ligon (1987. Am. Nat. 13:654-676) have argued that variability in territory quality is important, and it may be better to remain on a high quality natal territory (e.g., one with storage sites for acorns, rather than dispersing to a low quality territory. Consequently, cooperative breeding would result from advantages of philopatry (staying at home), rather than from habitat saturation and imposed philopatry (see handout).

Jan Komdeur's studies of the Seychelles Warbler (Acrocephalus sechellensis) show that territory quality affects decisions of young to remain on their natal territories as helpers or to disperse to vacant territories and become breeders. Young that delay reproduction on high quality territories and eventually breed on high quality territories have higher lifetime reproductive success than young that immediately start breeding on poor quality territories. Experimental removals (trapped birds were transplanted to another island where the species recently had become extinct) showed that one helper greatly improved the reproductive success of parents but two or more helpers could actually reduce reproductive success, particularly when one or more helpers were females because all females attempt to incubate, often simultaneously (Komdeur , J. 1994. Behav. Ecol. Sociobiol. 34:175-186.).

White-fronted Bee-eater

Returning to the White-fronted Bee-eater: bee-eaters nest colonially but feed on individual feeding territories that may be several kilometers away from the colony (highly unusual for colonially breeding birds). Within the colony males typically settle near their natal nests, attracting mates from elsewhere in the colony. Settlement patterns result in a male-based extended family system ("clans") within the colony. If a pair fails to start a reproductive attempt or fails during its reproductive attempt, the male will then aid his closest relatives. Although pairs typically form for life, females typically do not help within the "adopted" breeding clan until they have raised recruits to the clan.

Particularly in dry years an unaided pair generally cannot bring back enough food for its brood. As shown in the video, even with the aid of several helpers the pair may not be able to adequately provision more than one of the nestlings in the brood. In general, it appears that cooperative breeding systems occur when unaided pairs have poor reproductive success and helpers enhance the success of the breeding pair. Interestingly, male white-fronted bee-eaters frequently interfere with the reproductive attempts of their sons, and when sons do fail they aid their fathers. Steve Emlen's calculations actually suggest that a grown offspring helping to raise full sibs will accrue 0.47 offspring equivalents on average and unaided breeders will produce 0.51 offspring (on average), thus sons don't lose much by abandoning their reproductive attempts and helping their parents. Harrassing less related males wouldn't aid in recruitment of helpers because less related males would return and help at their own natal nests once their reproductive attempts have failed (Emlen, S. T., and P. H. Wrege. 1992. Nature 356:331-333).

Dwarf mongoose (remainder of lecture not covered in 1996)

Dwarf mongooses (Helogale parvula) are small social carnivores that breed communally. They live in stable packs (averaging 9 individuals) which consist of a dominant male and female, subordinate adult males and females and yearlings born the previous breeding season. Social rank among pack members is strongly age dependent. Males are more likely to disperse than females and sometimes disperse in groups, but subordinants of both sexes may remain in their natal packs and eventually breed there or disperse to nearby packs in search of breeding opportunities.

Subordinate pack members of both sexes help to raise the pack's young. Subordinates groom, feed, carry, and guard young at rates similar to those of parents. Subadult males provide vigilance for the group: when a group leaves the termite mound where it has roosted for the night, one member will stay behind and maintain a watch for predators, facing away from the pack as it moves away from the mound. When the pack approaches another termite mound another individual will run ahead and establish the next watch. The rearguard will then run and join the pack; guards are at greater mortality risk to avian predators such as goshawks, than the other pack members. Vigilant guards give predator alarm calls for the protection of the foraging group. Though subordinate male vigilance is extremely efficient, up to 93% of raptor attacks are aborted after an alarm is given, these males suffer extremely high mortality compared to others in the group.

The number of helpers id positively correlated with reproductive success. Unaided pairs are rare and almost never successfully raise young. Average relatedness between subordinants and dominants is high (r=0.33), and subordinants accrue indirect fitness benefits by helping.

Pregnant and "pseudopregnant" subordinant females will nurse the young, but subordinate females rarely successfully bear young. Ovulation is apparently induced by copulation and even multiple copulations may fail to induce ovulation and result in pregnancy of subordinate females. Occasionally subordinate females who give birth synchronously with the alpha female are able to have some reproductive success. Subordinant males may have some reproductive success either if subordinate mates are successful or if they sire one or more of the dominant female's litter--the alpha female typically is in estrous for about a week and is only guarded intensely by the alpha male for about a day in mid-estrus--24% of young were sired by subordinant males and subordinant females accounted for 15% of the young in one study of 45 offspring.

Within a pack, reproduction is incompletely monopolized by the dominant pack member of each sex. During mating periods females typically come into estrus synchronously and most sexually mature pack members of both sexes mate. Normally only the oldest (alpha) female becomes pregnant. Subordinant females are usually reproductively suppressed by a rank-dependent dominance effect on endocrine function and reproductive behavior--subordinant females have lower baseline levels of estrogen, lower mating rates and lower peak levels of estrogen. The dominant female does not interfere with matings by subordinant females and there is no increase in aggression among females during mating periods, however.

Subordinant males have androgen levels that are indistinguishable from those of dominant males and are reproductively suppressed by aggression of the dominant male, particularly when they try to mate with the dominant female at her peak of estrus. During the peak of estrus the dominant male remains close to the dominant female and aggressively prevents subordinant males from mating.

Group living appears to be a necessity for dwarf mongooses because predation rates on mongooses are high and are negatively correlated with pack size. Groups show dominance structure which strongly influences reproductive success. Subordinants benefit through increased survival and some chance of reproductive success; dominants benefit through increased survival and high reproductive success.

MALE COALITIONS

In several bird and mammal species, males form coalitions, sometimes sharing estrus females. In some species males in trios (2 males, 1 female) have lower success/male than males in pairs (e.g., Galapagos Hawk) but have higher survivorship because they are better able to defend high quality territories.

In other species success/male is actually enhanced. For example, alpha Acorn Woodpecker males in a population in New Mexico permit copulations between the beta male and the single breeding female. There is more than a doubling of reproductive success/group when three or more woodpeckers are present. In this population the beta and alpha males are generally unrelated.

In other bird species in which males form coalitions, it has been shown that kin selection probably is not the key to cooperation and reciprocation. Rather, cooperation seems to be selfishly motivated and each participant stands to gain. Because of dominance and subordinant relationships, some participants may gain more than others, but the subordinates enjoy higher success with the group than they would if they adopted another strategy.

Lions

Reference: Packer, C. and A. E. Pusey. 1982. Nature 296:751-753.

Lion prides are stable groups composed of 2-18 adult females, their dependent offspring and a coalition of 1-7 adult males. Earlier studies assumed that male coalitions are always composed of relatives. Yet, many coalitions are now known to contain non-relatives. Between the time males leave their natal pride (when 2 to 4 years-old) and win a pride several years later, mortality rates are high -- initial companions may be lost and new ones found.

Within coalitions, males compete for estrus females, and the winner guards the female for several days. Males donít tend to share females, and the little ìsharingî that does occur is actually less frequent between relatives than between non-relatives. Alcock discusses the results of Packer et al.'s (1991. Nature 351:562-565) studies of kinship and reproductive success. The largest coalitions contain only related males (large groups of males that dispersed together from their natal pride and had high survivorship before acquiring a pride). In these groups there is high variance among males in mating success; apparently subordinant males have higher inclusive fitness by aiding the reproduction of related males than they would by dissociating from their original coalition and associating with unrelated males.

Single males and males in groups of two may form coalitions with one or two unrelated males during their nomadic phase. Forming such alliances seems essential for successful take-over of a pride. When males form coalitions with unrelated males they preferentially associate with males of similar age and condition and thus later have roughly equal reproductive success.

Back to lecture notes index.

Back home.