Title: Chapter 11. Kin Selection and Social Behavior
1Chapter 11. Kin Selection and Social Behavior
- Interactions between individuals can have 4
possible outcomes in terms of fitness gains for
the participants.
2Kin Selection and Social Behavior
- Cooperation (mutualism) fitness gains for both
participants. - Altruism instigator pays fitness cost, recipient
benefits. - Selfishness instigator gains benefit, other
individual pays cost. - Spite both individuals suffer a fitness cost.
3Kin Selection and Social Behavior
- No clear cut cases of spite documented.
- Selfish and cooperative behaviors easily
explained by selection theory because they
benefit the instigator.
4The puzzle of altruism
- Altruism is the difficult one to explain because
the instigator pays a cost and another individual
benefits. - Hard to see how selection could favor an allele
that produces behavior benefiting another
individual at the expense of the individuals
bearing the allele.
5The puzzle of altruism
- For Darwin altruism presented a special
difficulty, which at first appears to me
insuperable, and actually fatal to my whole
theory. - Darwin suggested however that if a behavior
benefited relatives, it might be favored by
selection.
6The puzzle of altruism
- W.D. Hamilton (1964) developed a model that
showed an allele that favored altruistic behavior
could spread under certain conditions.
7Coefficient of relatedness
- Key parameter is the coefficient of relatedness
r. - r is the probability that the homologous alleles
in two individuals are identical by descent.
8Calculating r
- Need a pedigree to calculate r that includes both
the actor and recipient and that shows all
possible direct routes of connection between the
two. - Because parents contribute half their genes to
each offspring, the probability that genes are
identical by descent for each step is 50 or 0.5.
9Calculating r
- To calculate r one should trace each path between
the two individuals and count the number of steps
needed. Then for this path r 0.5 (number of
steps) - Thus, if two steps r for this path 0.5 (2)
0.25. - To calculate final value of r one adds together
the r values calculated from each path.
10(No Transcript)
11(No Transcript)
12(No Transcript)
13Hamiltons rule
- Given r the coefficient of relatedness between
the actor and the recipient, Hamiltons rule
states that an allele for altruistic behavior
will spread if - Br - C gt0
- Where B is benefit to recipient and C is the cost
to the actor. Unit of measurement for B and C is
surviving offspring.
14Hamiltons rule
- Altruistic behaviors are most likely to spread
when costs are low, benefits to recipient are
high, and the participants are closely related.
15Inclusive fitness
- Hamilton invented the idea of inclusive fitness.
Fitness can be divided into two components - Direct fitness results from personal reproduction
- Indirect fitness results from additional
reproduction by relatives, that is made possible
by an individuals actions.
16Kin selection
- Natural selection favoring the spread of alleles
that increase the indirect component of fitness
is called kin selection.
17Alarm calling in Beldings Ground Squirrels
- Giving alarm calls alerts other individuals but
may attract a predators attention. - Beldings Ground Squirrels give two different
calls depending on whether predator is a
predatory mammal (trill) or a hawk (whistle
Sherman 1985).
18Is alarm calling altruistic?
- Sherman and colleagues observed 256 natural
predator attacks. - In hawk attacks whistling squirrel is killed 2
of the time whereas non-whistling squirrels are
killed 28 of the time. - Calling squirrel appears to reduce its chance of
being killed.
19Beldings Ground Squirrels
- In predatory mammal attacks trilling squirrel is
killed 8 of the time and a non-trilling squirrel
is killed 4 of the time. - Calling squirrel thus appears to increase its
risk of predation. - Whistling appears to be selfish, but trilling
altruistic.
20Beldings Ground Squirrels
- Beldings Ground Squirrels breed in colonies in
Alpine meadows. - Males disperse, but female offspring tend to
remain and breed close by. Thus, females in
colony tend to be related.
21Beldings Ground Squirrels
- Sherman had marked animals and had pedigrees that
showed relatedness among study animals. - Analysis of who called showed that females were
much more likely to call than males.
22(No Transcript)
23Beldings Ground Squirrels
- In addition, females were more likely to call
when they had relatives within earshot.
24(No Transcript)
25Beldings Ground Squirrels
- Relatives also cooperated in behaviors besides
alarm calling. - Females were much more likely to join close
relatives in chasing away trespassing ground
squirrels than less closely related kin and
non-kin.
26(No Transcript)
27Beldings Ground Squirrels
- Overall, data show that altruistic behavior is
not randomly directed. It is focused on close
relatives and should result in indirect fitness
gains.
28Helping behavior in birdsWhite-fronted
Bee-eaters
- In a large number of birds young that are old
enough to breed on their own instead help their
parents rear siblings. - Helpers assist in nest building, nest defense and
food delivery.
29Helping behavior in birdsWhite-fronted
Bee-eaters
- Helping usually occurs in species where breeding
opportunities are limited territories or nest
sites are hard to acquire. - Young make the best of a bad job by remaining
home to assist their parents.
30Helping behavior in birdsWhite-fronted
Bee-eaters
- Steve Emlen et al. studied white-fronted
bee-eaters intensively in Kenya. - Nest in colonies of 40-450 individuals. Groups
of relatives (clans) defend feeding territories
in vicinity of colony.
31Helping behavior in birdsWhite-fronted
Bee-eaters
- First year birds that opt to help can choose
among many relatives when deciding whom to help. - Bee-eaters conform to predictions of Hamiltons
rule.
32- Coefficient of relatedness determines whether a
bee-eater helps or not. - Also, bee-eaters choose to help their closest
relatives.
33- Nonbreeders in clan that are not relatives (birds
that have paired with members of the clan) are
not related to offspring being reared and are
much less likely to help than relatives.
34(No Transcript)
35- Assistance of helpers is of enormous benefit to
parents. More than 50 of bee-eater young starve
before leaving the nest. - On average, presence of each helper increases
number of offspring successfully reared to
fledging by 0.47. Thus, there is a clear
inclusive fitness benefit.
36(No Transcript)
37Kin selection and cannibalism in tadpoles
- Spadefoot toad tadpoles come in two morphs.
- Typical morph is omnivorous mainly eats decaying
plant material. - Cannibalistic morph has bigger jaws and catches
prey including other spadefoot tadpoles.
38Kin selection and cannibalism in tadpoles
- Pfennig (1999) tested whether cannibals
discriminate between kin and non-kin. - Placed 28 cannibalistic tadpoles in individual
containers. Added two omnivorous tadpoles
(tadpole had never seen before) to each
container. One was a sibling, the other non-kin.
39Kin selection and cannibalism in tadpoles
- Pfenning waited until cannibal ate one tadpole,
then determined which had been eaten. - Found that kin were significantly less likely to
be eaten. Only 6 of 28 kin were eaten, but 22 of
28 non-kin.
40(No Transcript)
41Kin selection and cannibalism in tadpoles
- Pfennig also studied tiger salamanders whose
tadpoles also develop into cannibalistic morphs. - Kept 18 cannibals in separate enclosures in
natural pond. To each enclosure added 6 siblings
and 18 non-kin typical morph tadpoles.
42Kin selection and cannibalism in tadpoles
- Some cannibals discriminated between kin and
non-kin. Others did not. - Degree of relatedness to siblings 1/2
43Kin selection and cannibalism in tadpoles
- Thus, by Hamiltons rule discrimination in favor
of kin favored if B(r) - C gt 0 - Benefit estimated by counting number of siblings
that survived. Siblings of discriminating
cannibals twice as likely to survive as siblings
of non-discriminating cannibals.
44Kin selection and cannibalism in tadpoles
- Benefit thus approximately 2.
- Cost assessed by evaluating effect of not eating
siblings by comparing growth of discriminating
and non-discriminating cannibals. No difference
in growth rates. Cost then estimated as close to
0.
45(No Transcript)
46Kin selection and cannibalism in tadpoles
- By Hamiltons rule discrimination should be
favored because 2(1/2) - 0 1 which is gt0.
47Altruistic sperm in wood mice
- Moore et al. have demonstrated altruistic
behavior by sperm of European wood mice. - Females highly promiscuous. Males have large
testes and engage in intense sperm competition
with other males.
48Altruistic sperm in wood mice
- Wood mice sperm have hooks on their heads. And
connect together to form long trains of sperm
that can include thousands of sperm. - Swimming together sperm travel twice as fast as
if they swam separately.
49(No Transcript)
50Altruistic sperm in wood mice
- To fertilize egg, train must break up.
- To break up train many sperm have to undergo
acrosome reaction releasing enzymes that usually
help fertilize an egg.
51Altruistic sperm in wood mice
- Releasing these enzymes before reaching an egg
means these sperm cannot fertilize the egg.
These sperm sacrifice themselves. - Because other sperm carry half of the same
alleles, sacrifice makes sense in terms of kin
selection.
52Discrimination against non-kin eggs by coots
- Important to avoid paying costs on behalf of
non-kin. - Lyon (2003) studied defense against nest
parasitism in American coots. - Coots often lay eggs in other coots nests in
hopes of having them reared.
53Discrimination against non-kin eggs by coots
- Accepting parasitic eggs is costly because half
of all chicks starve and same number reared in
parasitized and non-parasitized nests. - Thus, host parent loses one offspring for every
successful parasite.
54Discrimination against non-kin eggs by coots
- Because of high cost of being parasitized and
lack of benefit (assuming parasites are non-kin)
Hamiltons rule predicts coots should
discriminate against parasitic eggs. - Coot eggs very variable in appearance. If 2 eggs
laid within 24 hours Lyon knew one was a parasite.
55Discrimination against non-kin eggs by coots
- Among 133 hosts 43 rejected one or more
parasitic eggs. Rejected eggs differed from
hosts eggs significantly more than did accepted
eggs.
56(No Transcript)
57Discrimination against non-kin eggs by coots
- Females who accepted eggs laid one fewer egg of
their own for each parasitic egg they accepted.
Average total clutch (including parasites) 8
eggs,
58(No Transcript)
59Discrimination against non-kin eggs by coots
- Females who rejected eggs laid an average of 8 of
their own eggs even though they waited to finish
laying before disposing off eggs they were
rejecting. Coots can count! - By counting eggs and rejecting extras that do not
look right coots prevent themselves from being
parasitized.
60(No Transcript)
61The greenbeard effect
- Sometimes altruistic alleles help different
alleles inadvertently when they help kin.
However, behavior is still favored because it
assists identical alleles half of the time.
62The greenbeard effect
- If alleles could recognize which individuals
carried other copies of them then they could
selectively act altruistically towards those
individuals. - Dawkins (1976) called this the greenbeard effect.
63The greenbeard effect
- Dawkins imagined an allele that caused its
carriers to grow green beards, to recognize green
beards in others and act altruistically towards
them. - Hard to imagine in wild because single allele
must cause three different effects.
64The greenbeard effect
- However, Quellar et al. (2003) have described
greenbeard effect in slime molds. - Slime molds live in soil. Germinate from spores
and spend most of life as independent,
single-celled amoebae.
65Slime molds
- When food scarce, individuals signal each other
chemically and aggregate together to form a
slug-like mass. - Slug travels some distance, then transforms into
a tall, thin stalk with fruiting body on top. - Cells in fruiting body form spores which disperse
and begin cycle again.
66(No Transcript)
67Slime molds
- Cells in stalk (20 of the individuals) sacrifice
themselves.
68Slime molds
- Quellar et al. studied wild-type allele csA
- Allele codes for protein on cell surface of
amoeba and that protein sticks to same protein on
other amoebae. Allele thus codes for both trait
and recognition (adhesion).
69Slime molds
- Remaining greenbeard trait is discriminating
altruism. - Quellar et al. mixed wild-type amoebae and
amoebae carrying a knocked-out version of the csA
allele and grew them on agar plates. - Starved amoebae to induce slime molds to form
fruiting bodies.
70Slime molds
- Quellar et al. found that wild-type cells were
disproportionately represented in the stalk
(suckers!) and knock-out type in the fruiting
body. - Wild-type apparently ended up in stalk because
they stuck together better.
71(No Transcript)
72Slime molds
- Situation reversed when slime molds grown on
soil, their natural environment. - More difficult for amoebae to stream on soil and
wild-type can stick together and pull each other
along.
73Slime molds
- Wild-type cells disproportionately represented in
fruiting body as well as stalk. - Less adhesive knockout cells tend to get left out
of aggregations altogether.
74(No Transcript)
75Slime molds
- Thus, in natural conditions wild-type allele of
csA makes its carriers altruistic towards other
wild-type cells. - Kin selection thus works at level of individual
alleles, not just individual organisms.
76Evolution of Eusociality
- Eusociality (true sociality).
- Many eusocial insects (bees, ants, termites) do
not reproduce. Instead they act as helpers at
parents nests for their entire life. This is an
extreme type of altruism.
77Evolution of Eusociality
- Eusociality describes social systems with three
characteristics - Overlap in generations between parents and
offspring. - Cooperative brood care.
- Specialist castes of non-reproductive individuals.
78Haplodiploidy and eusocial Hymenoptera
- One idea advanced to explain eusociality is the
unusual genetic system (Haplodiploidy) of the
Hymenoptera (ants, wasps, bees, etc.). - Males are haploid and females diploid.
- Males develop from unfertilized eggs and females
from fertilized eggs.
79Haplodiploidy and eusocial Hymenoptera
- Daughters receive all of their fathers genes and
half of their mothers genes. Thus, daughters
share ¾ of their genes. - This suggests females would be better off if they
favored the production of reproductive sisters
rather than their own offspring.
80Haplodiploidy and eusocial Hymenoptera
- Queens are equally related to all offspring and
so should prefer a 11 ratio of sons to daughters
among reproductives. - Females workers however should prefer a 13 ratio
of brothers to sisters among reproductives.
81Haplodiploidy and eusocial Hymenoptera
- It has been shown in wood ants that queens
produce equal numbers of male and female eggs,
but the hatching ratio is heavily female biased.
Workers apparently selectively destroy male eggs.
82Haplodiploidy and eusocial Hymenoptera
- Haplodiploidy appears to influence worker
behavior, but consensus today is that it does not
explain evolution of eusocial behavior in
Hymenoptera. - There are several reasons why.
83Haplodiploidy and eusociality
- First, haplodiploid explanation assumes all
workers have the same father. However, honeybee
queens mate with more than 17 males on average. - As a result relatedness between worker honeybees
often below 1/3.
84Haplodiploidy and eusociality
- Second, in many species, more than one female
founds a nest. In this case workers may be
completely unrelated.
85Haplodiploidy and eusociality
- Third, many eusocial species are not haploid
(e.g. termites) and many haplodiploid species are
not eusocial.
86Haplodiploidy and eusociality
- Phylogenetic analysis of Hymenoptera by Hunt
(1999) emphasizes that eusociality relatively
rare even though haplodiploidy occurs in all
groups. - Eusociality occurs in only a few families which
are scattered around the tree, which suggests
eusociality has evolved independently multiple
times.
87(No Transcript)
88(No Transcript)
89Haplodiploidy and eusociality
- Hunt also points out that eusociality has only
evolved in groups that build complex nests, and
care for young for a long time. - Association between nest building, long term care
and eusociality suggests main driving force for
eusociality is ecological not genetic.
90Haplodiploidy and eusociality
- Nest building and need to supply offspring with a
steady stream of food make it impossible or very
difficult for a female to breed alone. - Also, if predation rates are high, solitary
breeding individuals may not live long enough to
raise their young.
91Facultative strategies in paper wasps.
- Paper wasps (Polistes) are not sterile (unlike
ant and bee workers). Females can nest with
other females or establish their own nest. - Nonacs and Reeve (1995) found in Polistes
dominulus that females follow one of three
strategies.
92Facultative strategies in paper wasps.
- Initiate own nest
- Join nest as a helper
- Wait for a nest to become available
93Facultative strategies in paper wasps.
- Individuals founding their own nest are very
likely to fail because adult mortality is high
and nests with multiple foundresses can keep the
nest going. - However in multifoundress nests there may be
frequent conflict. The nests that did best were
those where one female was markedly bigger than
the others, which reduced fighting.
94(No Transcript)
95Facultative strategies in paper wasps.
- The sit-and-wait strategy also can pay off
because females often can adopt an orphaned nest
or take one over late in the season.
96Facultative strategies in paper wasps.
- Overall, in paper wasps an individuals decision
is affected by her relative size, relatedness to
other females, and availability of unoccupied
nests.
97Naked Mole-rats
- Naked mole-rats are highly unusual mammals.
- They are nearly hairless and ectothermic. They
are eusocial and, like termites, can digest
cellulose with the help of bacteria in their gut.
98Naked Mole Rats
Fig 51.33
99Naked Mole-rats
- The behavior of naked mole-rats is similar to
that of colonial insects. - There is a single reproductive female (queen) and
1-3 reproductive males. The remaining
individuals act as workers. They dig tunnels to
find food, defend the tunnel system from other
mole-rats, and tend the young.
100Naked Mole-rats
- Leading hypothesis for why naked mole-rats are
eusocial is inbreeding. - Average coefficient of relatedness is 0.81 and
about 85 of matings are between parents and
offspring or between full siblings.
101Naked Mole-rats
- Despite high level of relatedness conflicts still
occur because reproductive interests of workers
and reproductives are not identical.
102Naked Mole-rats
- Queens maintain control through physical
dominance. - Queen aggressively shoves workers who do not work
hard enough and shoves are mainly directly
towards less closely related individuals. - Workers double their work rate after being
shoved.
103Naked Mole-rats
- In addition to inbreeding, ecological factors
such as severely limited breeding opportunities
and group defense appear to contribute to
eusociality in naked mole-rats.
104Parent-offspring conflict.
- Parental care is an obvious form of altruism. In
many species parents invest huge quantities of
resources in their offspring. - Initially, parent and offspring agree that
investment in the offspring is worthwhile because
it enhances the offsprings prospects of survival
and reproduction.
105Parent-offspring conflict.
- However, a parent shares only 50 of its genes
with the offspring and is equally related to all
of its offspring, whereas offspring is 100
related to itself, but only shares 50 of genes
with its siblings. - As a result, at some point a parent will prefer
to reserve investment for future offspring rather
than investing in the current one, while the
current offspring will disagree. This leads to a
period of conflict called weaning.
106Parent-offspring conflict.
- The period of weaning conflict ends when both
offspring and parent agree that future investment
by the parent would be better directed at future
offspring. This is when the benefit to cost
ratio drops below ½.
107Fig 11.18
Figure shows B/C benefit to cost ratio of
investing in the current offspring. Benefit is
measured in benefit to current offspring and cost
is measured in reduction in future offspring.
108Parent-offspring conflict
- In instances where parents produce only half
siblings we should expect weaning conflict to
last longer because the current offspring is les
closely related to future offspring. - This has been confirmed in various field studies.
109(No Transcript)
110Siblicide
- In many species there is intense conflict between
siblings for food that may result in younger
weaker chicks starving to death. - In other species regardless of food supplies
first hatched offspring routinely kill their
siblings.
111Siblicide
- For example, in Black Eagles the first hatched
chick hatches several days before its sibling.
When the younger chick hatches its older sibling
attacks and kills it.
112Siblicide
- In species such as Black Eagles siblicide is
obligate in that the younger offspring is always
killed. Black Eagles are only capable of rearing
one young. - The most likely explanation for the later hatched
young is that for the parents it serves as an
insurance offspring in case the first offspring
fails to hatch or develop.
113Siblicide
- In other species such as Cattle Egrets there is
intense conflict that establishes a clear
age-based hierarchy in the brood that determines
how food is divided among the brood members. - In cattle egrets, younger chicks usually starve,
but if it is a good food year they often fledge.
114Siblicide
- Siblicide is thus facultative in cattle egrets
because restraint by the older chicks in not
killing the younger siblings can be rewarded in
good years. - In Black Eagles there is no prospect of two young
being reared, so the older chick ensures its own
survival by eliminating its sibling.
115Siblicide
- Siblicide shows that relatedness does not
necessarily lead to altruistic behavior. For
Cattle Egrets and Black Eagles selfishness is
better because the costs of altruism are too high.
116Reciprocal Altruism
- Some animals occasionally behave altruistically
towards non-relatives. - Such behavior is adaptive if the recipient is
likely to return the favor in the future.
117Reciprocal altruism
- Reciprocal altruism most likely in social animals
where individuals interact repeatedly because
they are long-lived and form groups, and also
when individuals have good memories.
118Reciprocal altruism in Vampire bats
- E.g. Vampire Bats. Feed on blood and share
communal roosts. - Bats may starve if they fail to feed several
nights in a row. - However, bats who have fed successfully often
regurgitate blood meals for unsuccessful bats.
119Reciprocal altruism in Vampire bats
- Cost of sharing some blood is relatively low for
donor bat but very valuable for recipient. - Research shows that Vampire bats share with
relatives, but also share with individuals who
have shared with them previously and with whom
they usually share a roost.
120Association is measure of how frequently two
individuals associate socially.
Regurgitators regurgitate to individuals
they associate with regularly.