FITNESS AND ADAPTATION.

General observation: most organisms show "good fit" to their environment. Desert

plants (desiccation resistant), marine organisms (fusiform body), cryptic insects All
generally can be placed in the context of the environment presenting a problem and
the organism, through evolutionary changes, providing a solution to the problem
posed by the environment. The apparent "good fit" between organism and
environment suggests that this state increases the survivorship and/or reproduction of
the organism.

This may lead us to identify certain problems posed by the environment and ask how
well does the organism solve the problem Leads to a design/engineering approach to
defining survival ability, design fitness or engineering fitness. We do this by
comparing form and function; often there is an "appealing" match or sense of "fit"-
ness. Is this what we mean by fitness

We then ask if the design/solution is the optimal solution suggesting an absolute

engineering fitness scale where the trait/attribute is assessed with respect to
an ideal solution to the problem. Alternatively, comparing different available
solutions may be all that matters: a relative engineering fitness

This can be difficult because we could almost always "design" a system or part of an
organism with a slight improvement over what it has achieved by evolution. Does the
presence of a design that "appears" to have room for improvement mean that it is not
"fit" Need to be aware of the design constraints of organisms before making
useful/contributory statement about level of engineering fitness.

Darwinian fitness is assessed not by looking at the "fit" between form and function,
but by attempting to determine the contribution to future generations. What
about grandchildless mutants in Drosophila: have normal number of offspring, but
these offspring are sterile. How many generations ahead do we need to look? Since
we can never really look ahead can we really identify fit individuals/genotypes?

Another old problem: is Darwinian fitness a tautology? (a statement that is true by

virtue of its logical form; needless repetition of an idea), e.g. my father is a man;
Brown students are smart; the water is wet. Natural selection was characterized by
Huxley as "the survival of the fittest". The allele with high fitness will increase;
allele A has high fitness; thus allele A will increase in frequency. The premise
provides for no alternative outcome, therefore it is a useless theory. Survival of the
fittest; who are the fittest? those that survive. Thus it reduces to survival of the
survivors. Characterized as such this tell us nothing and cannot be predictive?.

How do we get out of this? Assign the predictive aspect to the term fitness.

Distinguish between realized vs. expected fitness. A population passes through a
generation, look at frequencies of genotype before and after transition, define fitnesses
by changes in frequencies. This approach to fitness can lead us into the tautology trap:
fitness defined by those that survive better (post hoc), but the change may not have
been due to natural selection (e.g. drift). With true Darwinian fitnesses we would
be able to predict which genotype would survive better (fitness values assigned to
genotype "before" generation transition). The "fitter" genotype would have
an inherent tendency to out reproduce competing genotypes.

G. C. Williams in Adaptation and Natural Selection, 1966, Princeton U. Press, states

that the problem "is not survival as such, but design for survival" (pg. 159). Thus
overall "fitness" is a combination of Darwinian and engineering fitnesses.

How do we measure fitness? Bacterial chemostats, one genotype increases at the

expense of the other. Fine if we believe that environments are ~constant over time
(maybe they are in the intestine) and that alleles with different fitnesses can be
considered outside the context of the entire genotype. Probably unrealistic for many
haploids (and most diploids).

In diploids a bit more difficult: is it the phenotype, genotype or allele that determines

fitness. Well it is all: selection acts on phenotype which is produced by a
genotype in a given environment leading to the change in allele frequencies which
in the next generation gives different genotype frequencies. Problem: neither genotype
or phenotype are inherited intact (phenotypic "inheritance" depends on heritability of
trait VG/VP )

One approach is to determine fitness of alleles e.g. "average excess" of allele:

p' = (p2 wAA + pqwAa) / (p2 wAA + 2pqwAa + q2waa) = [p (pwAA + qwAa)] / (p2 wAA +

2pqwAa + q2waa)

where pwAA+ qwAa = wA = fitness of A allele. Thus p' = p

Think about it: this is the frequency with which the A allele pairs with itself (the A
allele) resulting in fitness wAA plus the frequency with which the A allele pairs with
the a allele resulting in fitness wAa relative to (i.e., divided by) the average fitness in
the population (wbar). This is a very nice formula since the ratio w A/wbar determines
whether p will increase or decrease in the next generation.

Fitness (selective) values of genotypes are a similar thing since they average the
interaction effects of the AA genotype over all other genotypes at all other loci in the
genome for a given environment: average fitness of AA when coupled with the B
locus: (wAA = p2wAABB + 2pqwAABb + q2wAAbb ) assuming p=f(B) and q=f(b).

Note: selection will continue until: wA = wbar (i.e., the ratio is = 1.0 and p' = p(1), no
change). Thus average fitness in the population will be maximal but this does not
mean the fitness of the population is maximal.

Need to distinguish between individual fitness and population fitness. wbar is a

description of the average fitness of all individuals in the population. Does this
determine the fitness of the population? If we think of the population as the unit of
"survival" a population with high fitness is one that has a high tendency to out
reproduce other populations. This is largely a demographic question since drought,
predators, parasites, etc. could alter a populations' probability of survival irrespective
of the number of individuals it is producing each generation.

A highly fit population is one that has a high reproductive output or is unlikely to go
extinct. These qualities can be unrelated to the average fitness in the population
(although they can be related).

Consider wasps living in nest holes. Population is limited by a fixed number of nest

holes. Now a mutation occurs that doubles the egg output of individuals carrying the
mutation. Those individuals with mutation are more fit, but the population will not
increase due to limitation of number of nest holes.

Consider prey switching in birds: mutation increases the fecundity of a moth. More
caterpillars present, birds switch their search image and prey more heavily on the
abundant caterpillars: reduces population size clearly the population is less fit.

These examples have assumed that population size is some measure of average

fitness May not be the case depending on the mode of selection: Soft versus hard
selection. Soft selection is like grading on a curve: there are always a certain number
of As (or F's). Regardless of the absolute fitness of individuals or the population the
"better" ones will survive to fill those spots. Hard selection assumes
some threshold level of "fitness" that could be set by the environment. If all
individuals are below this level, population crashes; if all are above, population
grows. See figures 7.4, 7.5, pgs. 166 - 167) .

FITNESS AND ADAPTATION II

Adaptation is a central issue or concept in evolution, but one must be very specific
when defining or deciding that one is actually "looking at" an adaptation or that
something is adapted. The issue revolves around the general belief that the
environment presents problems for the organism and that adaptations provide
solutions to these problems.

First: clarify some terminology. Adaptation come from ad (to, towards) and aptus (a

fit). But it is important to distinguish different uses of the word "adaptation" in the
biological sciences. An adaptation in physiology is a change in response to a certain
problem: you heat up and respond by taking off your jacket
(a behavioral "adaptation" to an environmental problem); you continue to heat up and
respond by sweating (a physiological response to an environmental problem).

In an evolutionary context: also a change in response to a certain problem. This time

the change is genetic, is achieved by the process of natural selection and takes place
over a period of time considerably longer than the physiological time scale. But note:
the physiological response itself could be an "adaptation" in the evolutionary sense: it
can be (is) adaptive (genetically) to adapt physiologically or behaviorally.

The word Adaptation is both a state of being (phenotypic trait or character) and it

is a process by which such traits come to be called "adaptations".

Implicit in the term adaptation is the belief that an adaptation serves some function or
purpose. Dispersal and reproduction are the function or purpose of an apple, and
apples are an adaptation apple trees use to achieve reproduction. This assumes natural
selection lead to the apple as the agent of dispersal and reproduction. Avoiding
predation is the function or purpose of leaf-like coloration in katydids and preying
mantids and their coloration is an adaptation these insects use to avoid predation.

As argued by G. C. Williams it is important to distinguish adaptations from

"effects": an effect of being an apple is to provide food for insect larvae or humans;
food is an effect of an apple's phenotype (good resources); apple farming is
and effect of apples' good taste and nutritional value (apples did not evolve to solve
the problem of providing work for apple farmers); the cryptic coloration of a katydid
is not for the purpose of demonstrating adaptation in evolution lectures; demonstration
is an effect of the striking morphology.

To reiterate: we identify traits as adaptations only when they evolved for the solutions

of a specific problem (function/purpose).

Selection is myopic: evolutionary trends are clear and presumably adaptive; some of
these lead to intensification other trends lead to diminution of characters. Examples:
Elaborate secondary sexual characteristics: increased horns, weapons in males
displaying for females; winglessness in insects: fleas adapted to reaching scalp/skin in
hairy animals; eye loss and reduced pigmentation in cave organisms: increase fitness
by not shunting energy into useless organs/tissues. All of these trends are adaptive but
adaptation is not seeing the "goal" of larger antlers or no eyes or smaller wings.

Now we have a problem of identifying adaptations in this context. Many traits

evolved under one selective regime and are now being used under a very different
selective regime. The current function may not reflect the context in which a trait
evolved. We have to be able to distinguish current utility from historical origin.
Some traits may have evolved in one context but later such a trait may be co-opted for
use in a different role. One term used to refer to such traits is Preadaptation. Some
evolutionary biologists dislike this term (some get nauseous when they hear it!)
because the term implies that an adaptive trend was anticipating some future need. We
know that evolution is "blind", "shortsighted" and can't foresee or anticipate new
selective regimes. Key point is that a trait's function can change faster than its form.

S.J. Gould and E. Vrba (1982, Paleobiology vol 8 pg 4-15) have suggested a different

term: exaptation to stress the cooptedness of traits.

Three examples: the evolution of bone tissue is believed to have proceeded under
selection for a tissue that stores inorganic ions (e.g. phosphate ions). The ions need to
be stored and released depending on the physiological demands of the body. The
tissue best at doing this became rigid and could be coopted as a structural member.
Thus organisms with "bone" as a structural tissue entered a new "adaptive zone" and
adapted for various functions. Skull sutures in mammals appear as an adaptation for
birth since they allow the skull to deform when passing through the birth canal (a tight
squeeze). But reptiles and birds have them and they hatch out of eggs. Sutures evolved
in one context (allow for growth of brain, head) but are an exaptation for birth in
mammals (they do allow for the head to change shape during birth which is
adaptive). Isolating mechanisms that prevent gene flow between incipient species.
These evolved in allopatry prior to any exposure to the sister taxon; isolating
mechanisms may undergo subsequent adaptive changes after being challenged by the
related species. In all these cases the historical origin is quite distinct from
the current utility.

Exaptation also allows for the evolution of traits that originally had no "adaptive"
function, but later get coopted for a function.

The Adaptationist Program as it has been called by Gould and Lewontin (Section
reading) seeks to find adaptive explanations for every characteristic of the organism.
Some things are not "for" the "purpose" or "role" they seem to be filling (e.g.
"Spandrels" [you must understand the analogy of spandrels!]; read the paper!) some
things are just nonadaptive and can be distinguished from maladaptive (former =
neutral; latter = bad). Think of your chin; it's not "for" something, it is there due to
differential growth rates of two growth fields (dentary and maxillary) of your skull. Its
just there. The striking pattern of white triangles on the Conus shell: looks like it is
"for" something but they live under the sand and mud and are not visible. Could be
due to chemistry of shell deposition or might have been "useful" in the shell's
ancestor.

Notion of optimality needs to be considered in a historical context: again, consider

current utility vs. historical origin (see figure 13.5, pg. 351).

Another important point contra the Adaptationist Program is that some traits may not
be capable of achieving "maximal adaptedness" selection acts on the entire phenotype
(debatable sensu units of selection; later) and phenotypes are compromises. Orians'
central place foraging model: bird sits in middle of territory, may fly a certain route
depending on availability and size of food items. Could design and optimal foraging
strategy BUT, when the birds leaves nest, young are available for predation. Foraging
strategy may not be the best foraging strategy, but the best compromise given
predation risk. Green sea turtle: excellent swimmer; terrible digger (not designed for
it) but must use flippers to dig hole for laying eggs. Flippers are not "optimal" for
digging, but they work.

Phenotypes as compromises underscores the importance of constraints. Evolution of

one trait can be constrained due to correlation among traits: selection for body
weight in broiler chickens: get more fat with it; selection for increase milk yield in
cows: more milk (with higher water content); selection for yield in soybeans: get less
protein per bean; selection for nicotine content in tobacco: tar content increases. These
are artificial selection examples but can apply to natural selection as well.

Constraints can be phylogenetic or developmental. (see figures 13.7, 13.8, pgs, 356-

7; figure 13.11, pg. 361). It might be adaptive for certain mammals to be able to
breath under water, but their phylogenetic history and developmental program
constrains them from evolving gills. They "solve" this "problem" by tolerating high
levels of lactic acid in their blood (and other physiological adaptations of the diving
response).

An informative means of analyzing adaptations is through the comparative

approach. Wise to put adaptations in a phylogenetic context example: rhinoceros
horns. Experiments need to be done to determine whether each unique pattern is
uniquely adaptive or simply neutral variations about a general adaptive theme
(difficult experiments! can you design some?)

In seeking adaptive explanations for phenomena we should seek parsimonious

adaptive explanations: Flying fish goes out of water; how does it get back down.
Physics offers a parsimonious explanation; we could be adaptationist about it and say
the fish evolved the means of returning to water. Problem is not how it comes down,
but why it takes so long to do so. Gliding must be a result of natural selection. Point
is: use the most efficient means to explain the existence of a trait. G. C. Williams:
Parsimony demands that we recognize adaptation at the level necessitated by the facts
and no higher.

LEVELS OF SELECTION

For natural selection to proceed there must be heritable variation in phenotypes and
the variation in phenotype must be associated with differential survival and/or
reproduction, i.e., there must be differential fitness. By inference then, any
entities exhibiting heritable variation in rates of reproduction can evolve. We need
not restrict our thinking to "individuals" in "populations" in the traditional senses of
these words.

Nature is organized in a hierarchical fashion. In terms of entities that can be

heritable we can consider genes, chromosomes, genomes, individuals, groups,
demes, populations, species, etc. Each of these entities meets the requirements of
units that can be acted upon by selection. At which level(s) does selection act?
Answer: all of them. What then is the important unit of selection? Answer: it depends.
First, some historical context. Serious consideration of a unit of selection other than
the individual was advanced by V. C. Wynne-Edwards (1962, Animal Dispersion in
Relation to Social Behavior). Populations have their own rates of origination and
extinction and selection could thus operate at the level of the group. Idea based on
observation that many species tend to curb their reproductive rate/output when
population densities are high. This behavior would favor groups that exhibited the
behavior and select against those that did not; i.e., there would be group selection.

G. C. Williams responded to this idea with Adaptation and Natural Selection (1966)

arguing that this behavior would be less fit than a cheating behavior where
individuals did not reduce their reproductive output at times of high density/low food
availability. In general selection at the level of the individual would be much stronger
than selection at the level of groups. In keeping with Williams' claim that one should
always seek the simplest explanation for selective/adaptive explanations, individual
selection is usually sufficient to account for patterns.

Group selectionist thinking leads to statements such as "good for the species" when it
is entirely likely that it may be good for the individual as well: reduced reproductive
effort in times of low food may increase an individual's reproductive output at a later
date.

Examples of selection at different hierarchical levels: Genic selection is selection at

the gene level; best example is meiotic drive (segregation distortion) where one
gametic type (often one chromosomal type) is transmitted into the gamete pool (or
next generation) in excess (or deficiency). The T locus in mice: affects tail length but
also viability. TT homozygotes have normal long tails; Tt heterozygotes have short
tails and transmit ~ 90% of the t allele to their sperm; tt homozygotes are sterile.
Meiotic drive will increase frequency of t allele to point where that become frequent
enough to occur as tt homozygotes with appreciable frequency, whereupon selection
works against t alleles. Opposite Selection at two levels selection for at the level of
the gene; against at the level of the genotype (organism).

In this system the balance of opposing selection coefficients at different levels should
give an equilibrium allele frequency of 0.7 for f(t) allele (using data not provided
here). In nature f(t) = 0.36. Discrepancy due to small local groups and drift. Some
local breeding groups (2 - 4 individuals) fixed for the t allele and since tt is sterile,
these demes go extinct reducing the f(t). Thus we have selection at three levels:
genic, individual and intergroup all contributing to the maintenance of the t/T
polymorphism.

Would we expect to detect meiotic drive systems in natural populations? If a new

mutation arose that introduced a bias in the transmission of the chromosome on which
it was located, then it would sweep to fixation and the locus would be homozygous for
the "drive" allele. Meiotic drive can only be detected in heterozygous state, so the
drive system would disappear when the drive allele went to fixation. There will be a
window of time where the allele is increasing in frequency, but this could be short-
lived. If the drive allele reduced viability in the homozygous state (as the T locus
example), then variation can be maintained and the drive system would persist longer,
making it more likely to be detected.

Another case where genic selection may act: sex ratios. Why should the sex ratio be
1:1 in most diploid species? Assume a sex ratio of 40% males and 60% females.
Males in this case are in limited supply. Any gene leading to the production of more
males (a allele results in more males the A allele at a sex determination locus) will be
favored until the frequency of males is >50%. Sex ratios tend to stabilize at 50:50 (R.
A. Fisher, 1930).

Kin selection and altruistic behavior many species of animals that live in groups
give warning calls which alert other individuals about predators, etc. How could this
behavior evolve when making the call alerts the predator to the callers location and
increases the possibility of the caller becoming prey. Mammals that nurse their young:
major energy investment for the mother may be thought to reduce her fitness, but
make obvious sense sine the individuals that benefit are close relatives (offspring). Put
in fitness terms: how could a trait evolve that lowers individual fitness?

Key point is the term individual. If the ones that benefit from the behavior
are related, the loss in individual fitness may be regained in inclusive fitness, i.e.,
individual fitness plus fitnesses of relatives. W. D. Hamilton argued that an altruistic
trait could evolve if cost to altruist/benefit to recipient < genetic relatedness (C/B < r)
where r is an estimate of the probability of the donor and recipient having allele
identical by descent. For example parent - offspring have r=0.5; siblings have r=0.5;
grandparent-grandchild have r=0.25.

Idea of inclusive fitness implies that one's fitness is determined by one's own life time
reproductive output and the reproductive output of relatives, scaled by their degree of
relatedness (r). In the warning call example, if calling out to warn about the arrival of
a hawk killed you but saved three reproductively active siblings, it would have been
worth it. If it only saved two siblings, it probability wasn't worth it. Obviously one
cannot tabulate the payoff of event x, y, or z. The point is that the notion of inclusive
fitness provides a fitness context where altruistic behavior could evolve even when it
appears to decrease individual fitness. Two modes of selection: individual
selection opposes altruism; selection among kin groups favors altruism.

Classic examples: helpers at the nest in birds. Young offspring remain at the nest to
help their parents produce more siblings in subsequent years/seasons. Helpers may
contribute more to their own fitness by aiding in the production of siblings than by
trying to reproduce themselves and failing due to lack of experience or availability of
nest sites. Sterile workers in hymenoptera (ants, bees, wasps): males are haploid
(develop from unfertilized egg) so sisters have r=0.75 because male contributes the
same allele (relatedness between sibs for paternal genes = 1.0; relatedness between
sibs for maternal genes = 0.5; among diploid female workers this averages out to
r=0.75). A female worker does more to propagate her own genes by staying in the nest
and aiding in the production of sisters (r=0.75) than by going off and producing her
own daughters (r=0.5). Used as an explanation for the evolution of sociality
(e.g., colonies) in hymenoptera.

Group selection = variation in the rate of increase or extinction among groups as a

function of their genetic composition. Again consider how an altruistic trait could
increase in frequency. Differential rates of extinction: allele A confers altruistic
behavior; at a selective disadvantage to allele a within the group. Should lead to the
reduction of allele A. But groups with high frequency of A may be less likely to go
extinct (due to better exploitation of resources). Over all groups with high frequency
of A persist and f(A) increases.

Differential productivity: similar to model above but altruistic trait

affects reproductive output of group. Selection against A allele within groups (selfish
types have higher short term fitness) but groups with high frequency of A exploit
resources more prudently and actually produce more offspring over the long term:
f(A) increases. Model this as follows: assume a haploid trait with A=altruist,
a=selfish; p=f(A). In each population p decreases within a population through time
due to selfish individuals out competing altruistic individuals. But in all populations
as a whole the altruist gene increases over time. Below, the average f(A) across all
populations is 0.5 at the start:
Clearly, if this system were to continue for many generations, the frequency of the
altruist gene would decrease within each population. But under conditions where the
selection favoring selfish genes was weak and the group selection increasing the
probability of staying extant (or the growth rate of the population) was strong, and
altruist allele might be preserved. Because the conditions are so restrictive, group
selection is presumed to be a rare phenomenon.

Group selection often involves plausible models but require that interdeme (group)
selection be strong. Would have to be very strong to overcome selection among
individuals within populations. Other complicating factors: turnover rate of
individuals is faster than of populations/groups; fixation of less fit allele is unstable
to invasion by new mutant allele or "selfish" allele introduced by gene flow. New
research on multilevel selection suggests that there should not be the necessary
association between altruism and "sacrifice" or genetic "suicide". Cooperation among
individuals can actually result in higher group fitness without the assumed loss of
individual fitness (see a meeting review in Science (9 August 1996) vol 273:739-740).
D. S. Wilson makes the analogy between the optimal clutch size argument of D. Lack
and the optimal group of Wynne-Edwards. With too many eggs in a clutch an
individual may die trying to support them all, so some intermediate clutch size is
"optimal" (see Life History Lecture). Optimal groups may evolve intermediate density
by the same trade-off mechanism.

A further problem for group selection: with localize population structure, there can be
considerable inbreeding which increases relatedness (r). Thus inter "group"
selection that gives the appearance of evolution of altruistic traits may be mediated
by kin selection due to the high relatedness among individuals.

Later we will consider species selection. Some lineages have more species than
others, but are these lineages more fit? Is this simply a pattern (more species) or is it
really a different process? Is it simple like bacteria in chemostats: a higher birth/death
ratio; some lineages seem to speciate faster than their members go extinct? Is this
mediated at the level of the species, or can we explain it (as G. C. Williams might
like) at the level of individuals within populations?

Richard Dawkins likes to couch this discussion in terms of replicators and vehicles.

Replicators are any entities of which copies are made; selection will favor replicators
with the highest replication rate. Vehicles are survival machines: organisms are
vehicles for replicators and selection will favor vehicles that are better at propagating
the replicators that reside within them. There is a hierarchy of both replicators and
vehicles. The key issues are that 1) the "unit" of selection is one that is
potentially immortal: organisms die, but their genes could be passed on indefinitely.
The heritability of a gene is greater than that of a chromosome is > that of a cell >
organism > and so on. But , because of linkage we should not think of
individual genes as the units; it is the stretch of chromosome upon which selection can
select, given certain rates of recombination. Issue 2) is that selection
acts on phenotypes that are the product of the replicators, not on the replicators
themselves, but the vehicles have lower heritability and immortality than replicators.
What then is the unit of selection?? All of them, just of different strengths and effects
at different levels.

EVOLUTION OF BEHAVIOR

One general view in the study of the evolution of behavior is that behaviors can have
a genetic basis. This is not to say that all behaviors are genetically based; indeed
many behaviors are entirely culturally transmitted or learned and may have little to do
with genetics (why are you sitting in the same seat?). For genetically influenced
behaviors we can treat them as we would treat any other genetically controlled trait of
an organism: 1) if there are genetically based differences in a behavior, and 2) these
differences affect fitness then, 3) behaviors can evolve by natural selection.

Two examples of genetically based behaviors: cricket song. Different species of

crickets have different calling songs with different characteristics, e.g., inter chirp
interval, pulse repetition rate, etc. Hybrids between closely related species often
exhibit songs with intermediate characteristics (pulse repetition will be
intermediate, inter pulse interval will be intermediate, etc.) a hypothetical example
with time on horizontal axis and each chirp = a group of vertical lines:
Another example (on a larger phylogenetic scale) is head scratching with the hind leg
in amniotes (reptiles, birds, mammals; those with an amniotic sac). Most reach the
hind leg over the fore limb to scratch the head; that birds and mammals do it suggests
that this behavior has a genetically programmed basis and has been inherited through
much of higher vertebrate evolution.

Behavior is usually dissected into two components for analysis: Proximate

causes/questions in which one asks how the behavior is performed and ultimate
causes/questions in which one asks why the behavior is performed. Tinbergen has
identified four questions to pose when analyzing a behavior 1) what is the cause, 2)
what is the development (ontogeny), 3) what is the current function 4) what is the
phylogenetic history. A strict course on evolution focuses more on the latter two
questions (recall adaptation/preadaptation/exaptation discussion and the identification
of current utility vs. historical origin).

Herring gulls breed is large colonies on the ground and defend territories. Two
separate calls used for 1) advertising nest site ("choking" call) and 2) as a territorial
claim (the "oblique pose" and "long call"). The Kittiwake also breeds in colonies but
nests on vertical cliffs and its nest pad is its territory and breeding site. In this species
only one behavior serves both functions: "choking" behavior is both defensive and
part of mate recognition/pair formation. This is seen as an adaptive behavioral shift
wit respect to the nest location (steep cliff).

There are many behaviors that at first appearance do not seem "adaptive". Infanticide
in lions was first viewed as "aberrant" behavior by abnormal individuals because it
was not "good for the species" (male lions displace other males from groups of
females and their offspring, and frequently kill the cubs). It is true that killing infants
is not, in the short term, an effective means of increasing population numbers of a
species. BUT, we now know (post W.D. Hamilton's 1963, 1964 papers on inclusive
fitness and kin selection and G. C. Williams book on Adaptation and Natural
Selection) that the more appropriate way to address such problems is to think about
them in the context of whether the behavior is good for the individual.

In analyzing infanticide from the perspective of gene thinking it is 1) not adaptive for
a male lion to invest reproductive effort in an individual with whom he shares no
genes and 2) once the infant is killed it is advantageous for the female to come into
estrous and have more offspring with the new male (this will increase her
reproductive output over leaving with the displaced male, and not benefiting from
other advantages of group living: foraging, avoiding predation on young). Given the
situation for both male and female, the observed behaviors make sense in terms
of propagating ones genes.
The role of the gene (or genes!) as the unit that is relevant in the evolution play an
important part in two influential books in the mid 1970s Sociobiology by E. O.
Wilson, and The Selfish Gene by Richard Dawkins. To grossly oversimplify one of
their main messages: "an organism is just DNAs way of making more DNA"

If we take the case of bird migration we want to know how the bird navigates to the
breeding location (solar and magnetic cues during flight), how the bird knows when to
begin migration (internal clocks and changes in day length [physiological changes]).
There is usually a high cost associated with migration so we also want to
know why birds do it since many die in the process (more time for feeding, more
available food). Individuals that do migrate must leave more offspring than those that
do not - again gene thinking helps account for why the behavior exists

Population genetic approaches to the evolution of traits rarely tell us why a phenotype
affects fitness in a particular way; the models usually look at whether fitness
increases or not. The optimality approach to the analysis of behavior attempts to
builds models where different behaviors are treated as the traits and asks which one of
these behaviors might evolve. The approach generally ignores the mechanics of
underlying genetic basis of the behavior (i.e., its mendelian and transmission
genetics). Optimal models assume there is a genetic basis and treat each behavior as
a haploid (asexual) trait that is inherited intact.

While Gould and Lewontin (and many others) have criticized optimal models, the
builders of optimal model (e.g., John Maynard-Smith, Univ. Sussex) argue that the
models do not assume that the organisms are optimal (because there are constraints on
evolution of traits), but by treating the problem as an optimality issue, it can tell you
what kinds of behaviors might evolve.

Two general type of optimal models: frequency independent models are designed

independent of what other strategies are doing, and seek to define the conditions
which might influence behavior (recall the "optimal foraging" model we described in
the adaptation lecture where a bird assess, quality, availability, distance to food items,
etc.).

Frequency dependent models are ones where the strategy of one type depends on the
strategies and frequencies of other types in the population. The general approach is to
look for Evolutionary Stable Strategies (ESS) = a strategy that, if adopted by all,
cannot be "invaded" by a mutant strategy. Here a strategy = the behavior of an
individual in a certain situation. These types of model apply nicely to ritualized
behaviors, distinct display behaviors which take the place of aggressive interactions.
Maynard-Smith's approach involves:
H = hawk who fights until the opponent retreats or will continue fighting injured with
cost C

D = dove who displays but will retreat if the opponent escalates

V = payoff of winning an encounter

C = cost of losing an encounter

These values can be put into a payoff matrix:

In encounter with:
H D
Payoff to: H 1/2 (V-C) V
D 0 V/2

H:H interaction = 1/2(V-C) because each individual hawk will win half of the time
and lose half of the time. In the D:D interaction each will win half of the time and
retreat half of the time (retreat with no cost). Which strategy is an ESS? Answer by
asking if a strategy can invade. Can H invade a population of D's?: Is payoff (D
against D) > payoff (H against D)? i.e., is V/2 > V? Answer = NO, so H can invade a
population of D's.

Is H an ESS? Is payoff (H against H) > payoff (D against H)? i.e., is 1/2(V-C) > 0?
Answer: it depends on the values of V and C: if V > C then payoff to H will
be positive and H is an ESS; if V < C then payoff to H will be negative and neither D
nor H will be favored (H will always invade a population of D's until H's become so
frequent that they encounter each other frequently. D can invade a population of H's
because H's tend to damage each other too much. In fact a population of all H's with
V<C would go extinct. Thus which behavior evolves depends on the nature of the
interactions.

One can imagine many other games and payoff matrices that could be built to model
other behaviors. The point of all this is to imagine the following: some species have
ritualized displays that appear "civil" in an anthropomorphic sense. Have these
behaviors evolved through a stage where hawks killed each other (C was high) to their
current state where the cost C to engaging in a behavior is considerably less? This
question could be addressed by comparing the behaviors of related species and
applying the game theory approach.
LIFE HISTORY EVOLUTION

For each organism a story can be told about how it makes a living. Most of the traits
that are part of this story have genetic bases and they contribute to fitness in some
way. Usually in the context of life history strategies fitness is discussed in the
context of different growth rates. New mutations that alter the growth rate-related
traits of organisms should lead to the evolution of new life histories.

A few stories: the agave plant lives for many/several years in the harsh desert
environment and when sufficient rain occurs it sends up a tall reproductive structure,
flowers, seets seed and dies. The codfish after reaching sexual maturity may produce
a million eggs in a reproduction, and do so over several seasons; species
of salmon forego reproduction until a late age, swim up current in certain
rivers/streams, sink all of their digestive tract and much of their muscle mass into eggs
in one reproductive effort, and die. Marine invertebrates: viviparous species have 20-
100 offspring, primitive brooders have 100-1000 eggs/offspring, species with no
parental care have 1000-500,000,000 eggs (Aplesia, sea slug). How these different
species achieve these feats is one question (a physiological one). Why they do it is a
very different one and difficult to answer. We address this question by analyzing life
history strategies.|

Best done by considering components of life history 1) survivorship and

2) reproduction. Two further questions: 1) what is the best relative allocation of
resources to each of these components, 2) what is the best timing of reproduction?

G. C. Williams suggested that there is a trade-off between survival and reproduction:

if one puts more resources into survival then there are fewer resources remaining for
reproduction. Conversely, reproduction is costly and will reduce survivorship in
subsequent years and reduce the future reproductive output.

The bigger they get the more eggs they can produce. This presents a dilemma: when
should they reproduce?; now so they don't miss out?, later so they can produce more?
but what if they get killed? if they are going to die anyway why not now? how many
eggs?, etc., etc. These are the questions facing the existential poker player. One
starts with a given number of chips (~ egg yolk), how does one play one's cards? how
does one invest one's chips (~ resources during growth and development)? when does
one decide to "fold" (wait until next year to reproduce when the conditions might be
better, new hand ~ random set of new environments)? what cards do the other players
hold (~ unknown variables of the biotic environment ~ competitive abilities of the
other "players").

All of these questions assume that the strategies have some high fitness solution. Lets
consider growth rate as one measure of a given genotype's (~ individual) fitness.

time 0 1 2... t
population size (at time t) N(0) N(1) N(2) N(t)
growth parameter   

N(1) = N(0) =>  = N(1)/N(0); similarly N(t) = N(0)t => t = N(t)/N(0)

This can be converted to an instantaneous growth rate using an exponential: N(t) =

N(0)ert where r is comparable to  . The point is that different genotypes could
have different  's and hence different fitnesses; selection = the difference between
genotypes in their growth rates.

We can describe the net reproductive rate (Ro) of a genotype or individual in terms

of its age specific survivorship and fecundity: x = age, lx = probability of surviving
from age 0 to age x, mx = number of offspring produces by an individual of age x.
Ro =  lx mx = expected total offspring per female during her lifetime. Now we can
Calculate an lx mx table, or life table.

Age lx mx #offspr lx ' mx' #offspr'

0 1 0 0 1 0 0
1 0.5 1 0.5 0.5 0 0
2 0.4 4 1.6 0.4 1 0.4
3 0.2 4 0.8 0.2 4 0.8
4 0.1 2 0.2 0.1 4 0.4
5 0 0 0 0 2 0
Ro = 3.1 Ro = 1.6
Note that the right hand column (a hypothetical new genotype called ') has the same
lx schedule and the same mx schedule except that it has been moved down one year
(start reproducing later). The genotype with the earlier age of first reproduction has
the higher net reproductive rate. This genotype will outcompete the other
(') and selection acting on such strategies will have the effect of reducing the age of
first reproduction because those genotype with thes trait (early reprod.) will have
higher growth rate = fitness. This approach (l x mx) helps us identify those components
of a life history that might affect fitness and hence be selected for/against.

Charnov and Schaefer proposed a simple model to examine whether annuals (species

living one year) or perennials (species living more than one year) should be
favored. P = number of progeny that survive to reproduce (including the parent in the
perennial case), B = number of seeds produced, S1 = survival during first year of
growth, S2 = survival during subsequent years (subscript a = annual, p = perennial).
for the annual: Pa = BaS1; for the perennial: Pp = BpS1+S2. Now we wont to
n=know when is Pa > Pp (i.e., when is the reproductive output of the annual greater
than that of the perennial?) Answer = when BaS1 > BpS1+S2. Rearrange to find that
the annual habit will be favored when Ba - Bp > S2/S1

In english this means that for the annual habit to be favored it has to have a
reproductive output greater than the perennial by the amount S 2/S1 (i.e., the ratio of
the "old" survivorship to the "young" survivorship. The result fits well with
observations about where annual and perennial plants are found: many desert plants
are annuals (low probability of survivorship), perennials common in tropics).

Other issues that might affect the evolution of life history traits: Certain environments
(arctic) dramatic changes in weather/seasons can cause lots of death in a genotype
independent manner (in an extreme case). Selection that goes on might be more
effective in the growth phase after population crash; this would select for genotypes
that have high rates of reproduction (r), so-called r-selected species/genotype/traits
(rapid development, early reproduction, small body size, semelparity [reproduce
once]). In more stable environments, where the population might be near its maximum
size, selection favors competitive ability to survive at the "carrying capacity" (= K)
of the population; such species/genotype/traits are said to be K-selected (slow
development, reduced resource requirement, delayed reproduction, large body
size, iteroparity [reproduce more than once]). This is best thought of a
a continuum since species are not either/or.

Another important way of thinking about life histories is bet hedging where variable
environments don't allow for the evolution of one particular strategy. Consider the
evolution of optimal clutch size ( how many eggs should I lay?). Again, the trade off
between survivorship and reproduction may determine one clutch size, but in
a variable environment, clutch sizes tend to be lower because there is an increased
risk of losing an average investment in reproduction. A slightly smaller clutch can
lead to a higher net reproductive rate (Ro)because it can allow for more reproductive
seasons than might be physiologically possible with larger clutch sizes. This argument
applies iteroparous organisms; the bet hedging in a semelparous orgnanism might
involve hedging on how long to wait before reproducing because semelparous
breeders will not have another clutch.

Other alternative strategies for dealing with spatial and temporal

heterogeneity Dormancy (wait out the bad times; seeds are good life history strategy
here; an associated strategy is responding to appropriate environmental cues for
growth). Dispersal (spreading the risk in unpredictable environments; fruits (carrying
seeds) again good examples; aphids, crickets respond to crowding by producing
wings and dispersing.

The life table approach shows us that individuals of different ages are not "worth" the
same in terms of reproductive potential. A newborn at time 0 when population size =
N(0) is 1/N(0) of the population. In the next time interval the population has grown so
a newborn is 1/N(0) of the population and relative to a newborn at time 0, this
newborn is "worth" = (1/N(0)1/ = 1. Similarly a newborn born at
time t is 1/N(0)t of the population and relative to a newborn at time 0 it is "worth" =
(1/N(0)1/t = t. Since the population is growing, a newborn born at a
later age is a smaller proportion of the population. This means that older
individuals contribute less to the makeup of the population. Also, older individuals
may be a smaller proportion of the pool of reproductive adults, so they contribute
even less to the population. The net effect is that the reproductive value of older
individuals is less than that of younger individuals. Reproductive value of an
individual is the contribution to future generations of an individual of age a. It can
be quantified as

Va = (a / la)  xlxmx Another way to think of this is the age distribution of test tubes
in a lab: more new one than old ones, associated with ageing is a finite probability of
being broken, thus a smaller proportion of all test tubes in the lab will be veery old.
These old one will not contribute much to "next generation"
Two points: 1) reproductive value may increase from birth to first reproduction (in
species that delay reproduction) because as one survives to this age, one increases the
probability that one will actually reproduce; 2) the drop in reproductive value with age
can account for the evolution of senescence (physiological decay with age).

Obviously genes for physiological decay would be selected against. But if ageing is
due to the expression of "decay" genes late in life, selection will not be able to act on
them because the genotypes which express this phenotype (decay) have very little
reproductive value at that age. Extreme example: a gene for death at age 80 could not
be selected out of the population unless individuals were reproductively
active past this age. If you stop reproducing before age 80 selection will not result in
differential survival of genotypes via phenotypic selection.

SEXUAL SELECTION

So far all of our discussions of selection have been without much regard for the sex of
the individual under selection (evolution of the sex ratio to 1:1 can occur through
selection for alleles in either males or females that favor the production of the rare
sex). But just as with natural selection individuals may differ in their ability to
reproduce, in sexual selection there can be differential reproductive success among
individuals of the same sex (and species). In order to mate, males need to gain access
to females and vice versa, and not all individuals will be equally successful at this
task. If there are genetically based differences in the ability of one sex to insure
successful mating with the other sex, sexual selection will occur.

The patterns and processes of sexual selection are best understood in the context
of parental investment. Sexual selection occurs because there is a correlation
between the gender of an individual and its parental investment in each
offspring. Parental investment is the investment of resources that increases in the
probability offspring will survive (a benefit) while decreasing the parent's ability to
produce more offspring (a cost). By investing in the production of current offspring a
parent will reduce the likelihood that it will be able to invest in future offspring. Costs
can be the energetic demands of parenting, increased predation risk, etc.

Parental investment should be proportional to:

(Benefit to current offspring survival) / (Cost to future offspring survival)

One important component of parental investment is the investment in gametes. This

serves as one means of distinguishing the sexes. Females are the sex with a large
parental investment per gamete. Example are eggs which have the nutrients to
promote the development once fertilization occurs. Males have a smaller parental
investment per gamete; generally carries only genetic information.

Some birds: each egg = 15-20% of a females body weight. In males, millions of

sperm produced each ejaculation; total output per reproductive season < 5% of body
weight. (The difference is less pronounced, but still significant, in mammals). This
difference in investment in gametes has important consequences for how
these resources are invested and hence how selection might act differently in the two
sexes. In general, females should commit their eggs for reproduction prudently;
males need not be so cautious in their commitment of sperm. Put in evolutionary
terms, the reduction in fitness of a female that squanders an egg will be greater than
the reduction in fitness of a male that squanders a sperm (give or take a million).

Another important component of parental investment is parental care, care of the

zygote after fertilization. The amount of parental care is also frequently quite different
in males and females. In species with internal fertilization the female is "stuck" with
the egg after fertilization. In many species the female also cares for the young after
they hatch. Males are not anatomically tied to the egg and hence are sometimes
"freed" from this additional parental care.

Moreover, in species where multiple matings occur, a male cannot be certain that a

females offspring were fathered by him. Benefit of caring for unrelated offspring
is low, a male can increase fitness by avoiding parental care.

General result is that sex with higher parental investment = limiting resource.

Since this sex is usually females, this will lead to male-male competition for access
to mates (like resource competition in ecology), and female choice where females
choose among males so that they may prudently commit their higher parental
investment.

Put another way, males will increase fitness by increasing the number of

fertilizations they can perform, leads to Intrasexual selection (selection among
individuals of the same sex). Females will increase fitness by being choosy, leads
to Intersexual selection where the high-investment sex chooses among the low
investment sex.
As a consequence, these forms of sexual selection can lead to the evolution of traits
that better enable each sex to perform its "fitness-increasing" behavior. In
males: Horns, large body size, sperm competition, mating plugs. In females:
choosy behavior . The existence of choice in females can lead to traits in males that
tend to give individuals an advantage in attracting mates: Plumes, coloration. These
present a problem, however. Have these phenotypic characters evolved so that males
can out compete other males to gain access to females, OR, have they evolved so that
a male might win and the be in a position to be chosen by females? The relative
contributions of intra and intersexual selection in the evolution of some traits can be
difficult to distinguish.

Ecological contexts in which parental care might be given can influence the amount of
parental care invested in offspring.

For species in the lower left of the graph it doesn't pay either sex to invest in parental
care; species in the middle should exhibit sexual dimorphism (male showy, female
less so or cryptic to reduce predation risk); species at the upper right should both
exhibit parental care, monomorphic for plumage characteristics in birds

Evidence that the certainty an individual has regarding

his/her paternity/maternity influences the amount of parental care is available in
fishes and amphibians. The table presents the numbr of genera in each group (fish,
amphibians) that fall into the categories.

Sex: Male Parental Care Female Parental Care

Fertilization: Internal External Internal External
Group
Fishes 0 48 15 21
Amphibians 2 13 11 7

Internal fertilization: male cannot be sure that his sperm will fertilize the eggs in
the female; another male could come along and mate the female, displacing his
sperm parental care does not pay. With external fertilization a male can be quite
certain that some of his sperm will fertilize eggs so parental care does pay

Examples of sex role reversal: male has higher parental care, is the high investment
sex and should choose among females; females are the showy sex, males are the
cryptic sex (another bird example = phalaropes, female is showy). Difficulties of
determining what is being chosen: good genes?, good gifts? (wasp example) what is
it about a male that will bring high fitness to one's offspring?

Zahavi handicap model (a peacock's tail is a handicap because it may reduce fitness
by natural selection): females choose males with handicaps because males with
handicaps must have "good genes" in order to be exant while carrying the fitness-
reducing handicap. Such a system might evolve if the female's advantage by mating
with the "good genes" of this male outweighed the cost of her offspring having to
carry this handicap around.

Kirkpatric model: flashy trait in male and the preference for it in female will

become associated (in technical terms, the alleles for the male's flashy trait and the
alleles for females choosing this trait come into linkage disequilibrium). Runaway
sexual selection results which can be non adaptive; sexual and natural selection can
oppose one another.