BIOL 4160 Natural Selection

BIOL 4160

Evolution

Phil Ganter

301 Harned Hall

963-5782

A favorite fruit, not yet ripe

Natural Selection as an Evolutionary Mechanism

Email me

Link to a list of Specific Objectives for lectures

Back to:

`Academic``Page`	`Tennessee State` `Home page`
`BIOL4160` `Page`	`Ganter``home page`

Natural Selection
Levels of Selection
Adaptation

Natural Selection

Natural selection is a mechanism (process, means) of evolution

Requirements for evolution through the mechanism of natural selection

Variation among members of the population

Some variants must be more successful in producing viable, fertile offspring

The characteristic of an organism that makes it more successful must be heritable, i. e. it must be possible to transmit the mechanism of success to one's offspring

Natural selection, narrowly defined, is the part of the process by which some organisms, because of their particular phenotype, are more successful than others

Success may involve many aspects of an organism's life history

Greater success may be achieved through optimizing viability, rate of development, body size, or fecundity (this is not an exhaustive list)

The phenotypic trait or combination of traits by which the greater success is achieved is an Adaptation

The verb adapt is used to describe evolution that increases the fitness of the population (populations adapt but not individuals, which can be confusing as this is not so when speaking in general)

What do we mean when we say (as biologists often do) that an organism is well-adapted for a particular habitat?

Complex adaptations perform complex functions and much theoretical and experimental work has gone into demonstrating that random mutation + natural selection is able, usually through small steps, to produce a complex adaptation by modification of simpler precursors

evolution of eye from photoreceptive patch of skin

evolution of bacterial flagelllar motor from precursor proteins

Thus, natural selection is sufficient to explain complex characters

Fitness is the per capita rate of increase of some biological unit (we will see that this can be a locus, an individual, a deme, or even a species)

fitness is the measure of success of a biological unit so it can be used to predict the effect of natural selection

so, it is reasonable to refer to the fittest individual in a population or the fittest allele at a locus or the fittest deme among all demes in a metapopulation

how it is measured is complex and will be discussed more later

anything that effects the ability of an organism to survive, develop, grow, and reproduce affects fitness, so it is a difficult measurement to do in the best of circumstances

to really measure fitness, it is necessary to measure it over at least two generations

often, we measure only a component, something that affects some aspect of fitness, as a substitute for measuring actual fitness

so, it is reasonable to refer to the fittest individual in a population

populations can have an average fitness (calculated from the fitnesses of the members of the population)

Does natural selection have a goal?

We often speak as though it does ("In order to expand it's distribution into colder water, the northern pike had to adapt its blood by lowering its freezing point.")

The statement above is teleological in that it gives the northern pike the goal of expanding its distribution

complex adaptations arise because of opportunity, not because of intent

Selection has a direction but not a goal

Goals are in the future and the future cannot be predicted

A trait may be selected to become larger for 100 generations because, in each generation, those with the larger phenotype are more successful

In the 101th generation, the environment has changed and those with smaller phenotypes are favored and selection reverses direction and will continue in the new direction as long as smaller phenotypes are favored (and as long as variation in size for this trait exists)

This is not goal oriented behavior as it is not the next generation that sets selection in the current generation

Natural Selection and (versus?) Sexual Selection

Selection based on increased mating succes (or increased gamete success) is Sexual Selection

Can take many forms (contests among members of a sex for mates, attempts to overcome resistance to mating by members of the opposite sex, sperm competition, infanticide of rivals offspring)

by one sex (often, but not always, the female) for characteristics of the other sex is called sexual selection

A male's reproductive success is enhanced by having the most extreme phenotype of the characteristic used by the female to choose a mate

As only those mates chosen will mate and successfully reproduce, female choice will a favor males with the extreme form of the trait upon which selection is made

Is this a kind of natural selection or not?

The book says yes and I am not so sure

They are similar in that some phenotypes are more successful than others

They differ in that females make the choice in sexual selection and, if anything can be said to make the choice in natural selection then an organism's environment "makes" it in natural selection

Does the distinction matter?

No, if those male characteristics chosen by females are chosen because they increase the average fitness of offspring

If the characteristic chosen by the females does not increase the average fitness of offspring, then sexual selection and natural selection are operating in opposite directions and should be kept separate

Evidence has accumulated that the exaggerated male traits, while themselves often bearing a cost to the males (outside of mating), can act as signals of the health of the males

This means that males with a successful genotype will be able to develop the costly, exaggerated trait fully and less successful males will not be able to bear the costs as well

Females that choose males with the exaggerated phenotype are therefore not choosing just the exaggerated phenotype but, instead, are choosing to mate with males who demonstrate their success through the fully developed exaggerated phenotype

By doing so, she can increase the average fitness of her offspring, both males and females

However, this is not the only possible scenario and, until we are very sure that sexual selection always increases the average fitness of offspring, it is best to keep them separate

Natural Selection and Genetic Drift

Since both of these evolutionary mechanisms can fix and allele in a population, how is one to tell which is responsible without measuring the fitness of various alleles (often a practical impossibility)?

Since chance is involved, repetition is the key to telling them apart

In experimental situations, one may set up an experiment multiple times and, if selection favors and allele, it should almost always become fixed (I say almost because experimental populations are often small and chance may overcome selection in a few populations)

In nature, we often do not have multiple similar situations from which to make the judgment but, if the conditions favoring one allele over another are constant (so that the direction of selection is constant), then observing the increase of the allele over many generations is a form of repetition

Genetic drift should reverse direction far more often and one can infer selection from the constant direction of change of the allele frequency

Does constant directional change in an allele's frequency guarantee that selection is operating?

If the direction of change is constant enough to reasonably eliminate genetic drift, the answer is yes but it does not mean that selection is favoring the allele you are observing

Hitchhiking (the change in frequency of one allele because it is linked to another allele that is favored by selection) may be occurring, so selection is happening but perhaps not to the allele you are observing

Hitchhiking is more common in asexual species and it prevalent in haploid asexual species (think of bacteria without any parasexuality) but can also be important for very closely linked loci

The book has a good example where one might observe that red balls are under selection when it is actually size of the ball that is under selection. Because size of the ball is linked to color, color appears to be changing due to selection but it is not.

It is also important to realize that natural selection, because it rests on a chance event (mutation) and the history of mutations that have occurred in the population, will not always find an "ideal" phenotype

natural selection will favor the best available phenotype

if a more fit phenotype exists in another population or is only a theoretical possibility unachievable with the genetic variation available in the population, selection will not produce the Panglossian (ask me for the definition of this) ideal

Trade-Offs and Optima

Does selection always produce extreme phenotypes (will selection for larger flies produce flies the size of volkswagens)?

Natural selection depends on the fitness of various phenotypes and extremes will be favored only when they actually have the highest fitness

Often a change in a phenotype that makes it more successful with respect to one aspect of fitness will have opposite effects on other aspects

if females choose for longer tails, longer tailed males mate more successfully, so, with respect to mating success longer tails are favored

if longer tails make flying more difficult then longer tailed males may burn more energy and be more likely to be eaten by predators than short-tailed males and selection favors shorter tails

When different components of fitness are affected in opposite ways, we describe this situation as a trade-off in the sense that fitness gained by more mating is traded off against greater risk of predation and greater cost of flying in males with long tails

What makes fitness hard to measure is that many aspects of an organisms life will affect fitness

This fact about fitness is also an advantage because fitness gathers the effects of lots of different processes into a single, overall measurement and we are not stuck with trying to "compare apples to oranges"

The best solution to a problem is the optimal solution

If no trade-offs apply, then the optimum will be the extreme form of a phenotype that is possible in a population (given the genetic variation present)

If trade-offs apply, then the optimum phenotype will be somewhere between the extreme phenotypes possible in a population

Levels of Section

Selection is a subtle idea and we have already seen how apparent conflicts might arise under the general heading of "selection"
- natural and sexual selection might produce such a conflict
- trade-offs result from conflicting effects of a phenotype on different components of fitness
There is another facet to selection from which conflict might arise
- What is favored by selection?
  - Is it the allele that produces the fitter phenotype?
  - Is it the individuals that have copies of that allele?
  - Is it the population that has individuals with copies of that allele?
  - Is it the species with populations with individuals with copies of that allele?
- The above questions refer to the Levels of Selection discussed below

Genic Selection

Some find it hard to separate the individual from its genes but they are not the same thing

Average fitness is often used in models to predict the fitness of an allele

Because genes at different loci interact (epistasis) an allele that increases fitness in one combination of genes might reduce it in another

To see the poltential conflict between gene and individual, imagine a partially recessive lethal (lethal when homozygous) gene that is beneficial when in the homozygous condition (an extreme case of Heterosis).

individuals homozygous for that gene bear the cost

individuals heterozygous at that locus reap the benefit

Alleles become common when their average fitness is greater than the average fitness of other alleles in the population

Extreme examples of genic selection are called Selfish Genes

Selfish genes are those that may improve their fitness but decrease an individual's fitness

Jumping genes - transposable genetic elements that insert and excise themselves into/out of the genome at radom (or at least randomly within some parameters) that can interrupt normal gene function and harm the individual in which the insertion occurs

Meiotic Drive - some alleles are more likely to make it into gametes than other alleles, a situation called meiotic drive because the allele that gets into more than its share of gametes will eliminate other alleles from the population, if all other things are unaffected

the t-allele (t for tailless) in mice can be found in over 90% of sperm produced by males heterozygous for the t-locus (hence it seems to be able to drive itself to fixation)

Fixation is prevented by the inviability of mice with the t-allele

tt mice often fail to survive or, if they do, are always sterile

the overall success rate of the t-allele depends on how likely it is to get into a heterozygous male rather than a homozygous male

as the t-allele become more common, heterozygotes become less common until a balance is achieved (another trade-off!)

Meiotic drive is known in Drosophila (segregation distorter locus, killer X and Y chromosomes) and fungi (spore killer loci)

Genetic imprinting is a form of selfish gene interaction

imprinting means that alleles from one parent are expressed but not the allele received from the other parent

usually because the unused allele is methylated (methyl groups added to cytosines that are directly followed by guanines) so that the gene can't be transcribed

conflict arises when mother donates resources to embryo

paternal genes are tied to the particular embryo they are in if there are multiple matings (other or subsequent embryos may have other fathers)

maternal genes are found in all mom's embryos (of course)

paternal genes favor maximizing care of particular embryo they are in (to, perhaps, the detriment of subsequent embryos)

maternal genes do not favor embryo they are in

this asymmetry in interest is the basis for conflict and imprinting the mechanism for one allele gaining advantage over another depending on which parent it comes from

Selfish genetic interactions are common

selection among mitochondria within cell lineages (yeast)

male sterility due to mitochondrial type

Lots of other selfish genetic elements

Individual Selection

Individual selection is what we have focused upon up until now

Individual selection depends on the fitness of individuals

Fitness of an individual depends on how many copies of their genes they manage to get into the next generation

Depends on living long enough, reproducing (mating, number of offspring, etc.) and also how soon one mates (generation time)

Kin-selection and relatives

Kin selection is a form of individual selection but is unusual in that it considers inheritance of an organism's genes through the reproduction of its relatives

Relatives are likely to carry copies of an organism's genes (these copies are said to the Identical-by-Descent or IBD)

How likely depends on how closely the organisms are related (measured by the Coefficient of Relatedness [r])

Siblings have, on average, half of one another's genes (r = 0.5) and most organisms share half of their genes with each parent (r = 0.5)

Inclusive fitness adds up the number of copies of an organism's genes that are in the next generation, which includes those transmitted directly by that organism and those from the reproduction of relatives

So, an organism might give up some direct reproduction if, by doing so, it was able to increase a close relative's reproductive success (discounted by the coefficient of relatedness)

Social insects - worker caste (females) do not reproduce but they help their sister (in some cases, mother) maximize her reproductive output

Haplodiploidy (helps but not necessary)

Group (sometimes called Interdemic) Selection

This is section that benefits the success of a group (often a small subset, called a deme, of the larger population or species - members of demes do not have to be closely related)

Here, it is groups that "reproduce" in that they give rise to other groups

Groups are selected if they produce more groups in the next generation that other groups do (just like individual selection but at the next organizational level up)

Those traits that enhance group success are then group adaptations

Group selection has been criticized because traits that benefit the group may often not benefit the individual

Altruistic traits - traits (often behaviors) that help others and, at the same time, harm the individual with the trait

Many (G. Williams) argued that group selection is unlikely

Groups giving rise to groups every generation is not a common population structure (now known to be more common than once thought)

Within any group, an allele that enabled cheating (not providing the group benefit) would enable individuals with that trait to gain the group benefit but not pay the cost

Within the group, the cheaters have the selective advantage and should become the most common phenotype

Cheating means that groups will not retain altruistic traits

Problem for critics was to explain why seemingly altruistic traits, well documented from nature, persisted

First crack in the wall of the "anti-group selectionists" was M. Wade's demonstration of group selection in laboratory populations of Tribolium - the common flour beetle that infests stored grain and can get into flour as well - (described in the text)

Interesting outcome of the experiments was that group selection can not only oppose natural selection but can also cooperate with natural selection to increase the evolutionary response to selection

D. S. Wilson then developed theoretical models that made group selection much more probable

Status of group selection still undecided

As the book says, most evolutionary biologists today do not believe that it is an important evolutionary force

However, most evolutionary biologists today study only individual or genic selection and this makes the idea's status a natural outcome of the fact that it is ignored

Species Selection

Species level selection is not an area with a firm theoretical underpinning

Textbook refers to differences in lineages in their rates of speciation (branching into separate species) and extinction

Sees this as analogous to birth and death rates (think of asexual lineages, not sexual species unless new lineages are always the product of allopolyploid hybridization)

The change in the lineages over time then becomes the macroevolutionary trends observed in the fossil record

analogous to the changes in phenotypic traits seen in evolving populations of individuals

Next level "groups of species selection" - community level selection??

selection of communities of species that replace other communities of species

communities are sets of ecologically related, not necessarily genetically related species

remember that demes do not have to be composed of related individuals

Adaptation

Are all traits observed in an organism the products of selection and, therefore, adaptations?
- No - adaptations must be demonstrated to be the products of selection
What else might account for a characteristic?
- Other processes might constrain a character
- "Neutral" differences, where there is no advantage to any of the phenotypes in the population
- Functional constraints (Spandrels of San Marco)
- Other characteristics limit the phenotypic possibilities of the character of interest
- Redness of the blood is text example
- Red is a characteristic of our blood but is not an adaptation
- It is a functional consequence of the use of iron to bind oxygen in the hemoglobin molecule
- Developmental constraints
- Variation in phenotypic traits may be constrained by the developmental process such that a potentially fitter phenotype is not a achievable
Preadaptation
- a characteristic that gains in adaptive value due to a change in the habitat (it is already there when it becomes advantageous) without changing its essential function
- Elongated feathers on the forelimb of the ancestors of birds may have evolved to help in gliding and became preadaptations for powered flight

Exaptation

a characteristic that serves two functions at once (the more recent function is referred to as the exaptation)

many pelagic birds use their wings to swim after prey underwater

wings developed as flight adaptations

when ancestors of pelagic birds took to feeding on marine prey, wings served a second function in swimming, the exaptation

the idea of a preadaptation is very close to an exaptation

How to tell that a trait is an adaptation

Experimental evidence that the particular phenotype increases fitness of individuals possessing it (or groups!) over other phenotypes available in the population is the soundest means of determining that a trait is an adaptation

Practical methods can infer that a characteristic is an adaptation when it is impossible or very difficult to demonstrate it experimentally

Similarity between phenotype and optimized human design (the camera and the camera eye) demonstrates functional significance of the phenotype

complexity in a trait is often a reasonable indication that the trait is an adaptation

Comparative Method

Uses the phylogeny of a group to search for evidence of adaptation

Traits the evolve more than once in multiple lineages under similar habitat conditions are likely to be adaptations (similar solutions to the same problem)

Traits that are present in the ancestor of all lineages currently possessing them may or may not be adaptations

Do adaptations mean that organisms make progress?

Have the phenotypic changes from our ancestors (fish) to man constituted "progress"

Progress is movement (in some sense) toward a goal

Evolution is not goal directed and so progress is not an evolutionary concept in that sense

Progress as improvement

If one defines improvement in an objective way (faster running, more efficient use of a material resource) then one may contrast some phenotypes as improvements in the defined sense

However, these improvements may not always be overall improvements in the fitness of the organism

More efficient wings that are less attractive to potential mates may not improve a male bird's fitness

If progress means "betterment" then one has to link it directly to fitness

since fitness depends on current circumstances, what is progress now (direction of selection) may be reversed tomorrow which is not usual for progress

All in all, progress is not a very useful evolutionary concept and is often misleading when it is used in evolution

Evolution and ethics

Thorny issue

Those who refer to nature to justify some human behavior often take a very selective view of what is "natural"

The simple case of "if it's natural, it's good" (Naturalistic Fallacy) is easy to disparage

Rape, murder, incest, war, theft, and numerous other activities universally condemned in humans all have been selected in some natural situations

The primacy of competition and predation, often cited as "nature's way" is only a partial truth at best

Cooperation, altruism, play, and sharing are all common species characteristics

Last updated March 24, 2010