Evolution can occur through random processes or through selection

In western New York State there is a herd of white-tailed deer that looks very different from most white-tailed deer. Many of the deer living on the 4,300-ha Seneca Army Depot do not have typical reddish brown coats. Rather, their coats are white (Figure 7.4). The white coat phenotype is due to a rare mutation that occurs throughout the range of white-tailed deer. Because a white coat can frequently make a deer more visible to predators, it does not provide any fitness benefits and we would not expect it to persist. So why is there a high frequency of white deer on the Seneca Army Depot? It turns out that when the army depot was built in 1941, the 4,300-ha area was fenced and several dozen deer were trapped inside. A few years later, 2 white deer were observed. Given that white deer were such an unusual sight, authorities at the depot banned hunters from killing the deer with the white phenotype. Over time, the deer population grew larger and the white phenotype became more frequent. Today, of the nearly 800 deer living on the property, about 200 are white.

Figure 7.4 A mutant white-tailed deer. At the Seneca Army Depot in western New York State, a mutation for white hair appeared in the 1940s. Since that time, the white phenotype was protected while the normal brown phenotype was hunted. Over the subsequent 70 years, the white phenotype has come to compose about 25 percent of the population.

Photo by Syracuse Newspapers/Dick Blume/The Image Works.

The story of the white deer demonstrates how evolution often happens through multiple processes. Random events, such as mutations, may confer no fitness advantage when they first appear. Such is the case for most deer populations in which the white mutation occasionally arises. However, if selection for the mutant phenotype begins to occur, as happened at the Seneca Army Depot with reduced hunting pressure, the mutant can become more frequent in the population.

Evolution Through Random Processes

As we saw with the mutation that causes white deer, random processes can facilitate evolutionary change in a population. In addition to mutation, random processes include genetic drift, bottleneck effects, and founder effects.

Mutation

We have seen that mutation is one of the two primary ways that genetic variation arises in a population. Because genes often code for functions that are vital to performance and fitness, mutations that negatively impact these functions are not favored by selection. However, a small fraction of mutations can be beneficial. For example, Figure 7.5 illustrates what happened in the deer herd of the Seneca Army Depot. In a herd of deer, a mutation for the white coat appeared, which added genetic variation to the population. After the mutation appeared and became protected, the frequency of white deer increased.

Figure 7.5 Evolution by mutation. Mutations, such as white coat in white-tailed deer, can arise in populations. If the mutation confers a fitness benefit, the mutation can increase in frequency in the population over multiple generations.

Mutation rates vary a great deal in different groups of organisms, but among genes that are expressed and can be observed as altered phenotypes mutation rates range from 1 in 100 to 1 in 1,000,000 per gene per generation. The more genes that a species carries, the higher the probability that at least one gene will experience a mutation. Similarly, the larger the size of the population, the higher the probability that an individual in the population will carry a mutation.

Genetic Drift

Genetic drift, another random process, occurs when genetic variation is lost because of random variation in mating, mortality, fecundity, or inheritance. Genetic drift is more common in small populations because random events can have a disproportionately large effect on the frequencies of genes in the population. But how does one determine whether an evolved phenotype is the outcome of drift versus some other process, such as natural selection? Research on the Mexican cavefish (Astyanax mexicanus) provides an answer.

The Mexican cavefish is a species composed of some populations that live in cave streams and other populations that live in surface streams. Although the populations can interbreed, they look quite different. As is the case with many cave-adapted animals, the cave populations have very reduced eyes and reduced pigmentation (Figure 7.6a). However, populations living in surface streams have normal eyes and dark pigmentation. To determine whether these changes were due to natural selection or to genetic drift, researchers raised individuals from the cave population, individuals from the surface population, and hybrid offspring made by interbreeding the cave and surface populations. The researchers then examined regions of the fish DNA that coded for eye size and pigmentation, which could contain one or more genes. In 2007, they reported that the 12 regions of DNA that coded for large eyes in the surface population all coded for small eyes in the cave population. The results are shown in Figure 7.6b. This suggests that natural selection favored all of the eye genes to evolve in a similar direction to produce smaller eyes. In contrast, when they examined the 13 regions of DNA that coded for pigmentation, they found that 5 of the regions coded for increased pigmentation in the cave populations and 8 of the regions coded for decreased pigmentation, as shown in Figure 7.6c. The lack of a consistent pattern among the 13 regions of DNA suggests that natural selection was not involved. Instead, the differences in pigmentation in the cavefish populations were likely produced by genetic drift. Given that small populations tend to experience more genetic drift than large populations, it may be the case that the cave population was initially very small.

Figure 7.6 Evolution by genetic drift. (a) Mexican cavefish populations that live in surface streams have large eyes and dark pigmentation, whereas populations that live in caves have small eyes and reduced pigmentation. (b) When researchers compared how different regions of DNA that code for eye size changed between surface and cave populations, they found that all 12 regions coded for smaller eyes in the cave population than in the surface population. Because all 12 regions changed in the same direction, this suggests that natural selection selected for smaller eyes. (c) When researchers looked at changes in regions of DNA that code for pigmentation, they found 5 regions coded for increased pigmentation while 8 regions coded for decreased pigmentation. Because the 13 regions did not change in the same direction, this suggests that changes in pigmentation were due to genetic drift.

Data from M. Protas et al., Regressive evolution in the Mexican cave tetra, Astyanax mexicanus, Current Biology 17 (2007): 452–454. Photo courtesy of Dr. Richard Borowsky.

Bottleneck Effects

A reduction in genetic variation can also occur because of a severe reduction in population size, known as the bottleneck effect. When a population experiences a large reduction in the number of individuals, the survivors carry only a fraction of the genetic diversity that was present in the original, larger population. Moreover, after being reduced to a small population by the bottleneck effect, the small population can then experience genetic drift. Population reductions can happen from natural causes, for example a drought that reduces the abundance of food, or anthropogenic causes, such as loss of habitat due to the construction of homes or factories.

One example of a bottleneck effect is the greater prairie chicken (Tympanuchus cupido), which is a grassland bird that historically lived throughout much of the middle United States, including Minnesota, Kansas, Nebraska, and Illinois. While prairie chickens have remained abundant in many states, the population in Illinois declined from approximately 12 million in 1860 to only 72 birds by 1990, as illustrated in Figure 7.7a. To determine whether this dramatic decline in population size was associated with a decline in genetic diversity, researchers collected DNA samples from museum specimens of prairie chickens from the 1930s, when the population was 25,000, and from the 1960s, when the population was 2,000. They defined this time period from the 1930s to the 1960s as “pre-bottleneck.” They also compared the genetic diversity of the Illinois birds before and after the bottleneck to the genetic diversity in the present-day populations of prairie chickens in Minnesota, Kansas, and Nebraska. In all cases, they examined the number of alleles that a population contained for each of six different genes. As shown in Figure 7.7b, the large populations of the neighboring states and the large historic population in Illinois all contained a high number of alleles. The current population in Illinois, however, has a lower number of alleles, which reflects the genetic bottleneck effect. The state of Illinois has since purchased more prairie habitat and introduced hundreds of prairie chickens from neighboring states to help bolster the Illinois population and increase its genetic diversity.

Figure 7.7 Evolution by the bottleneck effect. (a) The greater prairie chicken population in Illinois declined from approximately 12 million in the 1860s to 72 birds in 1990. (b) Averaged across six different genes, the mean number of alleles is high for birds in neighboring states that still have large populations and for the larger, historic population in Illinois that existed in the 1930s and 1960s. In the current population in Illinois, however, the small population is experiencing a population bottleneck and has a lower mean number of alleles. Error bars are standard errors.

Data from J. L. Bouzat et al., The ghost of genetic diversity past: Historical DNA analysis of the greater prairie chicken, American Naturalist 152 (1998): 1–6.

The bottleneck effect is of particular interest because the subsequent reduction in genetic diversity may prevent the population from adapting to future environmental changes. This is especially true for organisms that face deadly pathogens. An inability to evolve against new strains of a pathogen could lead to the extinction of the host organism. For example, the African cheetah (Acinonyx jubatus) experienced a population bottleneck approximately 10,000 years ago. Although the cause of that bottleneck is unknown, the current population contains very little genetic variation. This low genetic variation makes cheetahs more vulnerable to pathogens, including a deadly pathogen that causes the disorder known as AA amyloidosis and kills up to 70 percent of cheetahs held in captivity.

Founder Effects

The founder effect occurs when a small number of individuals leave a large population to colonize a new area and bring with them only a small amount of genetic variation. Following the founding by this small population, genetic drift can cause additional reductions in genetic variation. The genetic variation remains low until enough time has passed to accumulate new mutations. The water hyacinth (Eichhornia crassipes), which has been introduced by humans to many parts of the world, provides an example of the founder effect. Water hyacinth is an aquatic plant that is native to South America. During the past 150 years, the plant has been either accidentally or intentionally introduced to many other parts of the world. Once introduced, it grows and spreads quite rapidly, dominating shallow-water areas and displacing native plants. Today, water hyacinth has become one of the most invasive plants in the world.

Because most water hyacinth introductions are thought to have happened with few individuals, researchers wondered if the plant would show indications of founder effects in those parts of the world where it was not native. They sampled 1,140 plants from around the world and determined their genotypes. In 2010, they reported a single widespread genotype occurred in 71 percent of the sampled plants and this genotype dominated 75 percent of all populations that were outside the plant’s native range, as shown in Figure 7.8. Moreover, 80 percent of all populations outside the native range were composed of a single genotype, whereas populations in the native range of South America had up to five different genotypes. This pattern suggests that there were few founders in the invaded regions of the world and that they contained a small proportion of the genetic diversity of native populations in South America.

Figure 7.8 Evolution by the founder effect. Water hyacinth is an aquatic plant that was native to South America, where many different genotypes exist, as indicated by the different colored dots. Introductions around the world are thought to have occurred with low numbers of founders. Today, most populations existing outside of South America are represented by a single genotype.

Data from Y.-Y Zhang, D.-Y. Zhang, and S. C. H. Barrett, Genetic uniformity characterizes the invasive spread of water hyacinth (Eichhornia crassipes), a clonal aquatic plant, Molecular Ecology 19 (2010): 1774–1786. Photo by Eastcott-Momatiuk/The Image Works.

Evolution Through Selection

The nonrandom process of selection also plays a substantial role in evolution. Selection is the process by which certain phenotypes are favored to survive and reproduce over other phenotypes. As we saw in the story of the medium ground finches at the beginning of this chapter, selection is a powerful force that can change the phenotypes (and therefore the gene frequencies) of a population in a relatively short period of time. Depending on how the environment varies over time and space, selection can influence the distribution of traits in a population in three ways: stabilizing selection, directional selection, and disruptive selection.

Stabilizing Selection

When individuals with intermediate phenotypes have higher survival and reproductive success than those with extreme phenotypes, we call it stabilizing selection. As shown in Figure 7.9a, stabilizing selection begins with a relatively wide distribution of phenotypes, as illustrated by the orange line. After stabilizing selects for parents who possess intermediate phenotypes, their progeny have a more narrow distribution of phenotypes, as illustrated by the blue line. In doing so, it performs genetic housekeeping for a population, sweeping away harmful genetic variation. An example of stabilizing selection can be seen in selection for body mass in a species of bird from South Africa called the sociable weaver (Philetairus socius). Over an 8-year period, researchers marked nearly 1,000 adult birds and examined how a bird’s mass was related to its subsequent survival. As you can see in Figure 7.9b, the mass of the adults in the study follows a normal distribution with a mean of approximately 29 g. The researchers then asked how well birds of different mass survive. When mass was graphed against survival, as shown in Figure 7.9c, they found that the smallest and largest birds survived poorly, whereas birds of intermediate mass survived the best. That is, selection favored the intermediate phenotype. When the environment of a population is relatively unchanging, stabilizing selection is the dominant type of selection. Because the average phenotype does not change, little evolutionary change takes place.

Figure 7.9 Stabilizing selection. (a) Stabilizing selection favors intermediate phenotypes and selects against both extremes. (b) In the sociable weaver bird, body size has a normal distribution. (c) The bird experiences stabilizing selection on body size because birds with intermediate body sizes experience high survival while birds with low and high body sizes have low survival. This selection for intermediate phenotypes would cause the next generation of birds to have a narrower distribution of phenotypes.

Data from R. Covas et al., Stabilizing selection on body mass in the sociable weaver Philetairus sociu., Proceedings of the Royal Society of London Series B 269 (2002): 1905–1909.

Directional Selection

At the beginning of this chapter, we saw that beak size of the medium ground finch evolved to be larger during a drought when only the large seeds were available. This is an example of directional selection, which occurs when an extreme phenotype experiences higher fitness than the average phenotype of the population, as shown in Figure 7.10a. In the medium ground finch, for example, Peter and Rosemary Grant quantified the distribution of beak sizes in offspring hatched in 1976, which was just prior to a drought. As you can see in Figure 7.10b, the beak sizes of these offspring followed a normal distribution with a mean beak depth of 8.9 mm. When the drought came, although all seeds became less abundant, there were proportionately more large seeds remaining. The large seeds are harder to crack, so birds with deeper beaks were better able to feed on hard seeds and they had better survival. Because beak depth is a heritable trait, the offspring that were hatched in 1978 had deeper beaks, as shown in Figure 7.10c.

Figure 7.10 Directional selection. (a) Directional selection favors phenotypes of one extreme and selects against phenotypes of the other extreme. (b) Prior to a drought in 1976, the beak size in offspring of the medium ground finch had a mean depth of 8.9 mm, as indicated by the dashed red line. During the drought, when mostly large seeds were available, birds with larger beaks survived better. (c) Two years later, finch offspring had a mean beak depth of 9.7 mm, confirming that larger seeds caused directional selection for larger beaks.

Data from R. Grant and P. Grant, What Darwin’s finches can teach us about the evolutionary origin and regulation of biodiversity, BioScience 53 (2003): 965–975.

Disruptive Selection

Under some circumstances, we see another type of selection known as disruptive selection. In disruptive selection, individuals with extreme phenotypes at either end of the distribution can have higher fitness than individuals with intermediate phenotypes. Disruptive selection is illustrated in Figure 7.11a. For example, tadpoles of the Mexican spadefoot toad (Spea multiplicata) can express a range of possible phenotypes that are related to what they eat. At one extreme is the omnivorous phenotype that has small jaw muscles, numerous little teeth, and a long intestine that makes it well suited for feeding on detritus. At the other extreme is the carnivorous phenotype that has large jaw muscles, a notched mouthpart, and a short intestine that makes it well suited for feeding on freshwater shrimp and cannibalizing conspecifics. Intermediate phenotypes are not well suited to feed on either of the two food types. To test whether the tadpoles experienced disruptive selection, researchers collected more than 500 tadpoles from a desert pond, marked them to indicate their phenotype, and then returned them back to their pond. They sampled the pond again 8 days later to determine the survival of the three phenotypes. As you can see in Figure 7.11b, the omnivorous and carnivorous phenotypes survived relatively well, but the intermediate phenotypes—which had intermediate jaw muscles, tooth number, and intestine length—survived poorly. Because disruptive selection removes the intermediate phenotypes, it increases genetic and phenotypic variation within a population. In doing so, it creates a distribution of phenotypes with peaks toward both ends of the original distribution.

Figure 7.11 Disruptive selection. (a) Disruptive selection favors both extreme phenotypes and selects against intermediate phenotypes. (b) In the tadpoles of the spadefoot toad, an individual can have specialized mouthparts for carnivory and cannibalism (shown in this photo), omnivory, or an intermediate phenotype. (c) When more than 500 tadpoles of each phenotype were marked and released, about 10 percent were recovered. Of those recovered, the omnivores and carnivores experienced relatively high survival while the intermediate phenotypes experienced lower survival.

Data from R. A. Martin and D. W. Pfennig, Disruptive selection in natural populations: The roles of ecological specialization and resource competition, American Naturalist 174 (2009): 268–281. Photo by Thomas Wiewandt/wildhorizons.com.