Kin Selection and Social Behavior Chapter 11 1

Types of social interactions among members of the same species (Table 11. 1) • The actor in any social interaction affects the recipient of the action as well as himself. The costs and benefits of interactions are measured in units of surviving offspring (fitness). Actor benefits Actor is harmed Recipient benefits Cooperative Altruistic Recipient is harmed Selfish Spiteful 2

Kin selection and altruistic behavior • For Darwin, the apparent existence of altruism presented a “special difficulty, which at first appeared to me insuperable, and actually fatal to my whole theory. ” • However, he also suggested a solution — selection might favor traits that decreased the fitness of the actor if they increased the reproductive success of close relatives • This form of selection, which takes into account the fitness benefits to relatives is kin selection 3

Hamilton’s Rule • William Hamilton (1964) developed a genetic model showing that an altruistic allele could increase in frequency if the following condition is satisfied: Br - C > 0 Where B is the benefit to the recipient and C is the cost to the actor, both being measured in units of surviving offspring, and r is the coefficient of relatedness (or relationship) between actor and recipient 4

Inclusive fitness • Hamilton introduced the concept of inclusive fitness, which includes direct fitness + indirect fitness • Direct fitness is personal reproduction • Indirect fitness is the additional reproduction of relatives that is made possible by an individual’s actions 5

The coefficient of relatedness, r • The proportion of alleles in two individuals that are identical by descent (ibd) • The coefficient of relatedness of full sibs is r = 0. 5 • To see why this is so, we can use the following example, in which we give each of the four alleles at a locus in the two parents a unique label 6

Proportion of alleles shared ibd by full sibs • P: • O: A 1 A 2 x A 3 A 4 Sib 1 Sib 2 A 1 A 3 A 2 A 3 A 1 A 4 A 2 A 4 A 1 A 3 1. 0 0. 5 0 A 2 A 3 0. 5 1. 0 0 0. 5 A 1 A 4 0. 5 0 1. 0 0. 5 A 2 A 4 0 0. 5 1. 0 • The average proportion of alleles shared ibd by pairs of full sibs is 0. 5 7

Some coefficients of relatedness • • • Parent to offspring, r = 1/2 Full sibs, r = 1/2 Half sibs, r = 1/4 First cousins, r = 1/8 Grandparent to grandchild, r = 1/4 Aunt or uncle to niece or nephew, r = 1/4 8

Altruistic behavior in Belding’s ground squirrels • Breed in colonies • Male offspring disperse far from native burrow • Female offspring tend to remain and breed close by. Therefore, females in proximity tend to be closely related • Squirrels give alarm calls when predators are spotted (different calls for mammalian predators vs. birds of prey) • Is alarm calling altruistic and can it be understood as a result of kin selection? 9

In ground squirrels most alarm calling is done by females (Sherman 1977) (Fig. 11. 2 b) • Based on 102 encounters with predatory mammals • Blue line is expected frequency if each type of individual called in proportion to the number of times it was present when a predator appeared • Mortality is 8% for calling individual vs. 4% for noncallers when predator is a mammal 10

Female ground squirrels are more likely to give alarm calls when close kin are nearby (Sherman 1977) (Fig. 11. 3) • Based on 119 encounters with predatory mammals • Blue line is expected frequency if each type of pairing produced calls in proportion to the number of times it was present when a predator appeared 11

Nest helping in white-fronted bee-eaters • In white-fronted bee-eaters (and some other birds where breeding opportunities are extremely restricted), young adults often forego their own reproduction to help at the nests of other individuals. • This is clearly altruistic. Evidence suggests that nest helping can be explained by kin selection 12

White-fronted bee-eater (Merops bullockoides) 13

In bee-eaters, helpers assist close relatives (Emlen and Wrege 1988) (Fig. a) b) Among non-breeders, those born in a clan are much more likely to be nest helpers than those who enter a clan from outside and are unrelated to offspring being raised in that season Nest helping is disproportionately directed toward close relatives 14

Nest helpers increase the number of young birds fledged (Emlen and Wrege 1991) (Fig. 11. 7) • A group size of 2 represents a pair without helpers. The average number of young fledged = 0. 51 for pairs • Each additional helper at the nest increases the number of young fledged by 0. 47 on average 15

Kin-selected discrimination in cannibalistic spadefoot toad tadpoles (Pfennig 1999) (Fig. 11. 8 a, b) • • Tadpoles develop into typical morphs that eat mostly decaying plant matter or into carnivores that eat other tadpoles. Carnivores are more likely to eat non-sibs than sibs when given a choice between one of each kind 16

Kin selection can explain the presence of discrimination between sibs and non-sibs in cannibalistic tiger salamander larvae (Pfennig et al. 1999) (Fig. 11. 8 c) Benefit: B ≈ 2 Cost: C ≈ 0 Experiment consisted of placing 1 predatory morph + 6 sibling typical morphs + 18 non-sibling typical morphs in each of 18 enclosures in natural pond. 17

Coots can avoid parasitic altruism (helping non-kin) (Lyon 2003) (Fig. 11. 10 c, d) • • • If selection can favor helping kin, it should also favor avoiding sacrifices for non-kin Coots that accept eggs from other birds lose 1 offspring of their own for each parasitic offspring Coots that reject parasitic eggs have the same number of offspring as unparasitized birds (dashed line) 18

Eusociality: the ultimate in reproductive altruism • Characteristics of eusociality 1. Overlap in generations between parents and offspring 2. Cooperative brood care 3. Specialized castes of non-reproductive individuals • Insects (termites, hymenoptera), snapping shrimp, naked mole rats 19

Haplodiploidy and eusociality in hymenoptera (bees, wasps, ants) • Males are haploid (develop from unfertilized eggs) and females are diploid • Hamilton (1972) proposed that haplodiploidy predisposes hymenoptera to eusociality because females are more closely related to one another (r = 3/4) than they are to their own offspring (r = 1/2) • Females may maximize inclusive fitness by being sterile workers and helping to produce reproductive sisters (rather than by being reproductives themselves) 20

Proportion of alleles shared ibd by sisters in a haplodiploid species • P: • O: A 1 A 2 x A 3 Sister 1 Sister 2 A 1 A 3 A 2 A 3 A 1 A 3 1. 0 0. 5 A 2 A 3 0. 5 1. 0 • The average proportion of alleles shared ibd by pairs of full sibs is 0. 75 21

Is haplodiploidy the explanation of eusociality in hymenoptera? • Probably not – The preceding analysis assumed only 1 male fertilizes a queen — this is not true in honeybees, for example – In some species, colonies may be founded by more than 1 queen – Many eusocial non-hymenoptera are diploid (e. g. , termites) – Many hymenoptera are not eusocial (eusociality may have three independent origins associated with nest-building and the need to supply larvae with food) • Haplodiploidy may facilitate the evolution of eusociality but a more important factor may be the need for help in rearing young 22

• Families that include eusocial species are indicated in boldface type Sociality and nesting behavior in hymenoptera (Hunt 1999) (Fig. 11. 13 23

Naked mole rats • All young in a colony produced by a single queen and 2 – 3 reproductive males • Not haplodiploid, but colony members are highly inbred (average r = 0. 81) • 85% of matings are between full-sibs or parents and offspring • Queens use physical dominance to coerce help from less closely related individuals 24

Naked mole rat queens preferentially shove nonrelatives (Reeve and Sherman 1991) (Fig. 11. 16) 25

Parent – offspring conflict • Parental care is a special case of providing fitness benefits for close relatives • The offspring is the fitness of the parent (this means that benefits and costs of parental care both accrue to the parent) • In species that provide extensive parental care, the fitness benefit to the parent of providing additional care to current offspring needs to be weighed against the fitness cost of that additional care in terms of lost future offspring 26

Parent – offspring conflict occurs because parents and offspring value the costs of parental care differently (Trivers 1985) (Fig 11. 18) • • As offspring grow, the benefit/cost ratio for the parent declines. Benefit (B) is measured in terms of increased survival of offspring receiving parental care; cost (C) is measured in terms of lost future offspring due to continued parental care. From parent’s point of view, it should stop giving parental care when B/C declines to 1. But, the offspring devalues the cost to the parent by 1/2 because lost future full-sibs are related to the offspring by r = 1/2. Therefore, the offspring wants parental care to continue until the B/C ratio for the parent is 1/2 - fig. (a) [(or 1/4 if future offspring are half-sibs - fig. (b)] 27

Harassment in white-fronted bee-eaters can also be analyzed in the context of parent-offspring conflict and kin selection • Fathers occasionally coerce sons into helping to raise their siblings by harassing sons who are trying to raise their own young • Sons are as closely related to their full sibs as they are to their own offspring (r = 1/2 in both cases) • Furthermore the average number of offspring in nests without helpers is 0. 51, whereas every helper increases the number of surviving nestlings by 0. 47 on average • Therefore, the direct fitness lost by a son who is coerced into helping his parents is balanced by the increase in indirect fitness that results from helping his parents (provided that the son would not have had helpers) 28

Bee-eaters recruit helpers who are younger and closely related (Emlen and Wrege 1992) (Fig. 11. 19 b) 29

Reciprocal altruism • Reciprocation is offered to explain altruism between unrelated individuals • The necessary conditions for reciprocal altruism to evolve are: – The fitness cost to the actor must be ≤ the fitness benefit to the recipient – Non-reciprocators must be punished in some way (otherwise alleles that caused “cheating” would displace alleles for altruism) • Conditions that favor the evolution of reciprocal altruism are: – Stable social groups (so that individuals are involved in repeated interactions with one another) – Lots of opportunities for altruistic interactions during an individual’s lifetime – Good memory – Symmetry of interactions between potential altruists 30

Blood-sharing in vampire bats — an example of reciprocal altruism? (Wilkinson 1984) — 1 • Reciprocal altruism has been difficult to document • In vampire bats (Desmodus rotundus) the basic social unit is 8 – 12 females and their dependent offspring that frequently roost together • Many individuals preferentially associate with one another when roosting • The altruistic act is sharing blood meals by regurgitation • 33% of young and 7% of adults fail to get a blood meal on any given night • Bats are likely to starve to death if they go three consecutive nights without a meal 31

Blood-sharing in vampire bats — an example of reciprocal altruism? (Wilkinson 1984) — 2 • Bats are more likely to share blood with relatives and with associates. Both effects (relationship and association) were statistically significant. (These data do not include regurgitation by mother to child because that is parental care. ) • Blood sharing was not random in a group of captive individuals. Bats were more likely to receive blood from and individual that they had fed previously. 32

Association, relatedness and altruism in vampire bats (Wilkinson 1984) 33