Goals: the goals of this lecture are to introduce more aspects about natural selection, including how the fitness of a trait may depend on how common it is, and on the other traits in an organism, and how the evolution of different traits of an organism may be related.
Related Textbook Material: Freeman and Herron (2001) Chapter 7
Lab Manual Questions over this material are in Lab Manual Chapter VIII
We have seen how natural selection affects traits that are controlled by many different genes (click here if you want to review natural selection on quantitative or polygenic traits) rather than just a single gene. We also started to look at some situations in which the fitness of traits is more complicated than what we assumed when we first modeled natural selection. We saw that when the environment changes, that the fitness of traits can change (for example, when the climate where Darwin's Finches live fluctuates between wet years and dry years, the fitnesses of different bill sizes also changes.) We also saw that a trait may have high fitness in one context but low fitness in another context (for example, brown one year old male Darwin's Finches have high survival, but black one year old male Darwin's Finches are more likely to attract mates and reproduce.) These can lead to traits that retain genetic variation (rather than the loss of variation typical for natural selection, as described by Fisher's Fundamental Theorem.)
In this lecture, we will continue to examine ways in which fitness can change over time, and we will look at situations in which the evolution of one trait in an organism depends on other traits of that organism. We will consider how these situations relate to questions such as maintenance of genetic variation and traits that eventual evolve in species.
The fitness of a trait can change over time when something called frequency dependent selection occurs. Frequency dependent selection occurs when the fitness of a trait depends on how common individuals with that trait are in a population. There are two forms of frequency dependent selection:
Positive Frequency Dependent Selection occurs when a trait has higher fitness when it is common than when it is rare. For example, bright warning (aposematic) coloration in a poisonous species is likely to have higher fitness when it is common because if most individuals are brightly colored, then predators will have learned to avoid brightly colored individuals and not attack them.
Negative Frequency Dependent Selection occurs when a trait has higher fitness when it is rare than when it is common. Negative frequency dependent selection will turn out to have some important implications for evolution of other traits, so we will consider a couple of examples.
Consider what will happen to eye color over time in such a fruit fly population. If white eyes start out rare, females will mate preferentially with males with white eyes, so white-eyed males reproduce most and white eyes will become common. Once they become common, females no longer prefer to mate with white-eyed males, and mate with red-eyed males instead. Now red eyes have higher fitness and red eyes become common again.
This example illustrates something that is generally true of negative
frequency dependent selection: since rare traits have high fitness,
and will become common again, negative frequency dependent selection
maintains genetic variation in populations
Pathogens will have high fitness if they can successfully attack their hosts, and evade their hosts' immune system defenses. We would expect pathogens to evolve through natural selection to be able to attack whatever the most common immune defense in the population is, since this is the environment in which they will most commonly be occurring. The result of this is what causes negative frequency dependent selection in the host: common forms of the host immune system will have low fitness because the pathogens will have evolved to be able to attack them, but rare forms of the host immune system will have higher fitness because they can fight off the pathogens.
Again, consider what happens over time. A rare form of the immune system that can avoid being attacked by some disease will have high fitness and as a result will become common over time, while the common forms, having lower fitness because they are more susceptible to disease, will become rare. Once the initially rare forms become common, the pathogens will evolve to be able to attack them, and now they will have lower fitness and become rarer. Again, frequencies of traits will tend to fluctuate over time, and genetic variation in these traits that are subject to negative frequency dependent selection will be maintained.
Now we will consider how the evolution of different traits may depend on one another. We have been considering traits one at a time -- what is the fitness of black color versus gray in moths, for example. Clearly, organisms consist of many different traits. We will look at two reasons that it is important to consider the many traits of an organism: first, the development of one trait in an organism may necessarily depend on the development of other traits, and second, the fitness of a trait in an organism may depend on what other traits it has.
Correlated characters are different aspects of an organism whose degrees of development depend on one another. For example, in Darwin's finches, the size of the bill and the overall size of the body are related -- birds with large bodies also develop large bills, birds with small bodies develop small bills.
Characters may be correlated because they are controlled by some of the same genes. For example, there may be genes that affect body size in general, and that result in larger bills, bodies, legs, etc. because they affect all of these at once. Characters may also be correlated because their embryonic development is related. For example, salamander feet develop through early embryonic stages with fewer than 5 toes, and with webbing on the feet; later they develop 5 unwebbed toes. Some species, however, have fewer toes and webbed feet. They apparently stop their development at an earlier stage than other species. Toe number and webbing are correlated with each other in salamanders because they are correlated during their development.
Whatever the cause of correlated characters, the important point is that the fact that they are correlated means they necessarily develop together. As a result, evolution of one trait necessarily causes evolution in the other correlated trait. Consider Darwin's Finches again. You have seen that in dry years, large billed birds have higher fitness. The result is that during dry years bills evolve to become larger. During this time, body size also evolves to become larger because it is correlated with bill size. Large body size in itself does not have higher fitness than small body size; in fact, it may have lower fitness since large bodied birds require more nutrients and reproduce later. However, the high fitness of large bills results in the evolution of large bodies, too, despite the slightly lower fitness of large bodies. The important point here is that when characters are correlated, evolution of one character because of its strong effects on fitness can cause the evolution of other characters that are correlated with it, even if these other characters cause slightly lower fitness.
Now let's consider a different way in which different characters' evolution may depend on each other. The fitness of one trait may depend on the other traits present in the organism. For example, consider flowers that attract pollinators. Attracting more pollinators increases reproduction, and therefore increases fitness. Different kinds of pollinator are attracted by suites of different characteristics of the flowers. Honeybees, for example, are likely to be attracted to flowers that are blue or violet and that have petals that form a landing platform for the bee. In contrast, hummingbirds are likely to be attracted to flowers that are red and that have a long tubular shape into which the bird can insert its long bill. So we could ask what is the fitness of red color in a flower? It depends on another trait -- the shape of the flower. If the flower is long and tubular, red color would have high fitness since it would be successfully pollinated by hummingbirds. If the flower has a shape that would provide a landing platform for bees, it would have low fitness -- the shape is wrong for hummingbirds, but the color is wrong for attracting bees. The point, then, is that we can't simply say what fitness one trait has unless we also consider other traits.
A coadapted gene complex is a group of genetic traits which have high fitness when they occur together, but which without each other have low fitness. In the flower example given above, red color and a long tubular shape would be a coadapted gene complex.
Note that the concept of a coadapted gene complex is very different from that of correlated characters. Coadapted gene complexes are groups of different genes and they do NOT have to occur together ? they will only have high fitness if they occur together, but low fitness combinations of these genes can be produced. In contrast, remember that correlated characters were traits whose development depended on each other, so they DO have to occur together.
In a population, there may be just one coadapted gene complex, or there might be several different combinations of traits, each of which could have high fitness. This latter possibility gives rise to another concept: that of an adaptive landscape. An adaptive landscape is the description of the fitnesses of all possible combinations of different traits in a population. Adaptive landscapes are frequently represented graphically; fitness is plotted on a vertical axis and trait values for different genes are plotted on other axes. Combinations of traits that have high fitness thus appears as peaks, and combinations that have low fitness appear as valleys. Here is an example of an adaptive landscape:
This example shows two traits, and a situation in which there are two combinations of traits, the peaks in the graph, shown in purple, that have high fitness, while other combinations of the traits have low fitness. These peaks in the adaptive landscape can be called adaptive peaks; note that they are also combinations of different genetic traits that, together, have high fitness, so they are coadapted gene complexes. An adaptive peak and a coadaptive gene complex are thus basically the same thing.
Note that in this case the left hand peak is taller, meaning that it has higher fitness than the right hand peak. Situations in which there is more than one adaptive peak, with some higher and some lower, are interesting evolutionarily. We have come to expect that populations will evolve to have the highest fitness. However, in this case, suppose the average combination of traits in a population is on the slope to the lower adaptive peak. Individuals produced with trait combinations that are a little closer to the higher peak will actually be lower on the slope, or perhaps in the valley, and will have low fitness. The population thus tends to evolve toward the closest adaptive peak, even if it is not the peak with the highest fitness. Populations can be "stuck" on lower adaptive peaks because intermediate forms between the adaptive peaks have very low fitness, and natural selection can not result in evolution toward these lower fitness conditions.
Note that we have seen a situation somewhat like this before, when we looked at underdominance. The population would evolve until one allele OR the other was fixed, because the intermediate state (the heterozygote) had low fitness than either homozygote. The allele fixed tended to be the more common allele, and it was possible that the allele for the homozygote that had slightly lower fitness to be fixed because the low fitness heterozygotes died out, and the less common allele is more likely to occur in the heterozygotes.
At this point, you have seen several examples of how natural selection can be occur when we consider that fitness may depend on how common a trait is, and on other traits that exist in an organism. You should consider how these will affect how much genetic variation will exist in traits and whether traits will evolve so that all are perfectly adapted to the environment.
To test your understanding of these different forms of evolution, you should work the practice questions in your lab manual, Chapter IX, now.