The theory of evolution by natural selection explains the changes that take place in populations of organisms as being the result of changes in the relative frequencies of different variants in the population. If there is no variation within a species for some trait, there can be no evolution. Moreover, that variation must be influenced by genetic differences. If differences are not heritable, they cannot evolve, because the reproductive advantage of a variant will not carry across generational lines. It is crucial to understand that the mutational processes that generate variation within the genome act at random, but that the selective process that sorts out the advantageous and disadvantageous variants is not random.

The ability to study evolution at the level of DNA and proteins has transformed our understanding of the evolutionary process. Before we had the ability to study evolution at the molecular level, there was no inkling that much of evolution was in fact a result of genetic drift and not natural selection. A great deal of molecular evolution seems to be the replacement of one protein sequence by another one of equivalent function. Among the evidence for the prevalence of neutral evolution is that the number of amino acid differences between two different species in some molecule—for example, hemoglobin—is directly proportional to the number of generations since their divergence from a common ancestor in the evolutionary past. We would not expect such a “molecular clock” with a constant rate of change to exist if the selection of differences were dependent on particular changes in the environment.

So much sequence evolution is neutral that there is no simple relation between the amount of change in a gene’s DNA sequence and the amount of change, if any, in the encoded protein’s function. Some protein functions can change through a single amino acid substitution, whereas others require a suite of substitutions brought about through cumulative selection. Such multistep, adaptive walks may follow different paths even when the conditions of natural selection are the same. This is because the paths available to any population at any given moment depend on the chance occurrence of mutations that may not arise in the same order in different populations. Furthermore, the previous steps taken may affect whether a new mutation is favored, disfavored, or neutral.

Before the advent of molecular genetics, it was not possible to know whether independent evolutionary events might have given rise to the same adaptation multiple times. By pinpointing the genes and exact mutations involved in changes in function, we now appreciate that evolution can and does repeat itself by acting on the same genes to produce similar results in independent cases. For example, changes to the same genes are responsible for independently arising cases of melanism and albinism in some vertebrates or for the loss of pelvic spines in different stickleback-fish populations. Evolution may repeat itself by altering the very same nucleotide in the case of independently arising sickle-cell mutations that lead to adaptive resistance to malaria.

An important constraint on the evolution of coding sequences is the potentially harmful side effects of mutations. If a protein serves multiple functions in different tissues, as is the case for many genes involved in the regulation of developmental processes, mutations in coding sequences may affect all functions and decrease fitness. The potential pleiotropic effects of coding mutations can be circumvented by mutations in noncoding regulatory sequences. Mutations in these sequences may selectively change gene expression in only one tissue or body part and not others. The evolution of cis-acting regulatory sequences is central to the evolution of morphological traits and the expression of toolkit genes that control development.

New protein functions often arise through the duplication of genes and subsequent mutation. New DNA may arise by duplication of the entire genome (polyploidy), a frequent occurrence in plants, or by various mechanisms that produce duplicates of individual genes or sets of genes. The fate of duplicate genes depends a great deal on the nature of mutations acquired after duplication. Possible fates are the inactivation of one duplicate, the splitting of function between two duplicates, or the gain of new functions.

Overall, genetic evolution is subject to historical contingency and chance, but it is constrained by the necessity of organisms to survive and reproduce in a constantly changing world. “The fittest” is a conditional status, subject to change as the planet and habitats change.

791