2. Adaptation as a hypothesis of evolutionary history
3. Molecular population genetics of adaptation
4. Adaptation and selfish genetic elements
Darwin’s theory of natural selection explains how genetically variable populations gradually accumulate traits that enhance an organism’s ability to survive and to reproduce. Calling a particular character an adaptation denotes the hypothesis that the character arose gradually by natural selection for a particular biological role, which is called the character’s function. Any hypothesis of character adaptation is therefore a historical explanation that must specify the particular population, the interval of evolutionary time, the geographic conditions in which the relevant evolution occurred, and the nature of character variation that was sorted by natural selection. Empirical rejection of the hypothesis of character adaptation suggests the alternative hypotheses of exaptation (a character co-opted by natural selection for a biological role not associated with the character’s origin), nonaptation (a character not discriminated from alternatives by natural selection), or disaptation (a character disfavored by selection relative to alternative forms). I illustrate the contrast between adaptationist and anti-Darwinian theories of character origination using a longstanding debate concerning evolution of mimicry of wing patterns among butterfly species. I describe adaptation as a molecular population-genetic process using as an example the medical syndrome of sickle-cell anemia in African populations; depending upon its genetic and environmental contexts, hemoglobin S may constitute an exaptation, a nonaptation, a disaptation, or a component of an adaptive complex of epistatically interacting genes. Evolutionary developmental modularity and phenotypic accommodation may enhance the role of phenotypically discontinuous changes in evolution by natural selection. Selfish genetic elements likely underlie most organismal characters that arise as disaptations and nonetheless persist despite natural selection against them. Suppression of selfish genetic elements is potentially a major source of evolution by natural selection. The explicitly historical approach to adaptation illustrated here contrasts strongly with a now largely discredited analogistic approach used in older ecological literature.
adaptation (as a process). Evolution of a population by natural selection in which hereditary variants most favorable to organismal survival and reproduction are accumulated and less advantageous forms discarded; includes character adaptation and exaptation.
balanced polymorphism. Occurrence in a population of a selective equilibrium at which two or more different allelic forms of a gene each have frequencies exceeding 0.05.
character adaptation. A character that evolved gradually by natural selection for a particular biological role through which organisms possessing the character have a higher average rate of survival and reproduction than do organisms having contrasting conditions that have occurred in a population’s evolutionary history; adaptation in this usage contrasts with disaptation, exaptation, and nonaptation.
developmental constraint. A bias in the morphological forms that a population can express caused by the mechanisms and limitations of organismal growth and morphogenesis.
disaptation. A character that decreases its possessors’ average rate of survival and reproduction relative to contrasting conditions evident in a population’s evolutionary history; a primary disaptation is disadvantageous within the populational context in which it first appears; a secondary disaptation acquires a selective liability not present at its origin as a consequence of environmental change or an altered genetic context.
exaptation. Co-option of a character by natural selection for a biological role other than one through which the character was constructed by natural selection.
function. The biological role through which an adaptive character was constructed by natural selection.
gradualism. Accumulation of individually small quantitative changes in a population leads to qualitative change; contrasts with saltation, in which a single genetic change induces a large qualitative change in phenotype.
mimicry. Evolution by natural selection in which a character is favored because it closely resembles one present in a different species; the species whose character is copied by a “mimic” is called the “model.”
modularity. Evolution of developmental constraints by which one of two or more alternative, qualitatively different suites of characters can be activated by particular genetic or environmental cues.
nonaptation. A character not selectively distinguishable from contrasting conditions present in the evolutionary history of a population.
saltation. Evolution of a large, qualitative change in phenotype in a single mutational step; contrasts with gradualism.
selfish genetic element. Genes that spread at a cost to the organism; stretches of DNA that act narrowly to advance their own proliferation or expression and typically cause negative effects on nonlinked genes in the same organism (modified from Burt and Trivers, 2006).
Among the various meanings given to the term adaptation in evolutionary ecology, synonymy with evolution by natural selection is probably the most common one. Darwin’s theory of natural selection explains how genetically variable populations accumulate traits that enhance an organism’s ability to survive and to reproduce by making resources more accessible (see chapter I.14). Less-favorable alternative traits decline in frequency and are lost from the population because their possessors lose the struggle for survival and reproduction. A population produces variant forms at random with respect to an organism’s needs, and natural selection retains only the advantageous forms.
Closely allied with Darwin’s theory of natural selection is his theory of gradual change. Darwin considered abrupt changes of organismal form or physiology likely to disrupt normal functioning and thereby to be discarded by natural selection. The favorable traits that natural selection accumulates across generations each contribute only small phenotypic effects in the traditional Darwinian hypothesis. Evolution of a qualitative change in organismal form, such as the origin of a new anatomical structure or color pattern, occurs gradually across many generations as natural selection increases the populational frequencies of many small component parts so that they come to reside in the same individuals. Natural selection acting on incremental variation thus provides Darwin’s major explanation for evolution of novel organismal forms.
To call a particular character an adaptation denotes the hypothesis that the character arose gradually by natural selection for a particular biological role, which is termed the character’s function (Gould and Vrba, 1980). Any hypothesis of adaptation is a historical explanation that must specify a particular population, interval of evolutionary time, and geographic conditions in which the relevant evolution occurred. Gould and Vrba (1980) use an extant population as the focal point for analysis and restrict the term adaptation to a character whose current utility matches the function for which the character arose by natural selection. They apply the contrasting term exaptation to a character co-opted by natural selection for a biological role not associated with the character’s origin. One need not restrict hypotheses of adaptation versus exaptation to extant populations, but the historical frame of reference must make explicit the temporal and spatial dimensions across which the relevant evolutionary processes occurred.
I illustrate the contrast between adaptationist and anti-Darwinian theories of character origination using a longstanding debate concerning evolution of mimicry of wing patterns among butterfly species. Many cases are documented in which two or more butterfly species share the same potential avian predators and also share closely matched patterns of warning coloration on their dorsal wing surfaces. Because an avian predator learns to associate specific warning coloration with distastefulness, a distasteful “model” species often evolves characteristic warning coloration. Other species that share the same potential predators as the model can gain a selective advantage by “mimicking” the warning coloration of the model species. In some cases, the mimic species is a desired prey item that tricks its potential predator by adopting warning coloration deceptively (called Batesian mimicry). If the mimic is distasteful, sharing of the same warning coloration among species provides mutual benefit (called Müllerian mimicry). In each case, evolution of warning coloration deters avian predators, all of which seek their prey visually and learn to associate particular wing patterns with distasteful prey. The adaptationist and anti-Darwinian explanations of Müllerian mimicry concur that natural selection for a shared warning pattern benefits members of each species because a predator needs to learn only one warning pattern to reduce mortality in each species.
In the late 1800s, the anti-Darwinian orthogenetic evolutionist Theodor Eimer used butterfly mimicry among other empirical examples to support an argument that natural selection cannot construct complex morphological characters by accumulating gradual changes. He argued that butterfly species have inherited from their common ancestor similar mechanics of wing development and shared biases in production of new patterns; genetic changes that introduce pigments onto a wing surface are therefore likely to produce similar geometric patterns in all species that share a particular set of developmental mechanisms. Natural selection acts to preserve shared warning coloration in multiple species, but the specific pattern is formally an exaptation; it is a consequence of developmental mechanics, not something evolved gradually by natural selection acting on randomly produced variation in pigmentation.
Ronald Fisher in 1930 used butterfly mimicry to support the opposite, adaptationist hypothesis: a mimic species gradually evolves a sequentially improved match to its model by natural selection acting on many genes whose variation exerts random and incremental effects on pigment deposition across the wing surface. The detailed matching of the model’s pattern and coloration by the mimic species therefore constitutes character adaptation.
In the 1980s, John Turner reported detailed genetic analyses of Müllerian mimicry among South American species of Heliconius butterflies to reconstruct the genetic histories of evolution of their mimicry patterns (figure 1). He concluded that genetic changes of major phenotypic effect were important for producing close matches in pigmentation pattern among geographically codistributed butterfly species and that subsequent improvement of the matching occurred by accumulating multiple genetic changes of smaller phenotypic effect. This interpretation supports Eimer’s general hypothesis that shared developmental constraints explain evolution of shared patterns and that mimicry evolves by exaptation; only the detailed fine-tuning of the matched patterns, as explained by Turner, constitutes character adaptation as argued by Fisher.
Each specific case of butterfly mimicry involves separate evolutionary histories of at least two species, and hypotheses of character adaptation versus exaptation therefore must be tested separately for each case. The prevailing pattern reported by John Turner for Heliconius populations might or might not prevail in other groups. In each separate test of a hypothesis of adaptation, one seeks evidence capable of rejecting the claim that a hypothetically adaptive character arose gradually through accumulation of many genetic variants, each of which gave its possessors a higher net rate of converting resources into survival, growth, and/or reproduction (= “Darwinian fitness,” see chapter I.14) than did the alternatives with which it formed population-level polymorphisms.
I emphasize the importance of historical precision in formulating and testing hypotheses of adaptation because careless uses of adaptation have elicited condemnation of adaptationist studies. I agree with the critics that one must resist an analogistic tradition in which one equates as similar or equivalent the character variation and selection pressures described for distantly related species. For example, claims in sociobiological literature that one can use behavioral ecological studies of “helpers at the nest” in a bird species to explain analogous behaviors in human families must be rejected as having no historical equivalence. Evolution by natural selection depends as critically on the specific character variation produced in a population and the genetic structure of that variation as it does on environmental conditions. The kinds of phenotypic variation produced independently in different species are comparable only to the extent that homologous developmental mechanisms channel the morphological expression to a few major alternative forms in each case, as appears to occur in wing patterns of Heliconius butterflies.
Historical hypotheses of adaptation can be categorized as microevolutionary or macroevolutionary depending on the investigator’s vantage point with respect to the historical process being studied (Rose and Lauder, 1996). A microevolutionary study measures dynamics of populational polymorphisms on a generational time scale (see chapter I.17). At this scale, an investigator must distinguish natural selection per se from the genetic response of a population to natural selection. Because the relationship between genotype and phenotype is complicated by genetic dominance and epistasis and phenotypic plasticity, natural selection on phenotypic variation does not guarantee a particular genetic response of the population to selection. Agricultural geneticists are well aware that selecting for a favorable characteristic in a crop species does not always cause a corresponding genetic improvement of the population in the following generation. A macroevolutionary study, by contrast, begins with the knowledge that a particular evolutionary change, as inferred by phylogenetic analysis, has occurred (see chapter I.16). The unanswered question is whether the organismal variation and environmental contexts of a particular character transition are compatible with a specific selective explanation. For example, one can reject the macroevolutionary hypothesis that bird feathers evolved by natural selection for utility in flight because the fossil record shows that evolution of feathers preceded evolution of flight in birds. The utility of feathers for flight in living birds is therefore an exaptation, although details of the size and shape of wing feathers in particular species might constitute adaptations for flight evolved more recently in the species’ evolutionary histories.
Figure 1. Evolution of shared antipredatory warning patterns by multiple geographic races of Heliconius melpomene (left) and H. erato (right) as interpreted by John R. G. Turner. Numbered areas on each map (top) indicate geographic distributions of corresponding numbered wing patterns below the map. Four inferred ancestral wing patterns (A–D) also are shown for each species. Tree diagrams show genetic substitutions of major effect that transform one color pattern to another by adding or subtracting large areas of pigmentation. Shared developmental constraints by these species likely underlie their parallel, saltational origin of the same major patterns; subsequent fine-tuning of the match between patterns of geographically codistributed races occurs by polygenic changes compatible with an interpretation of gradual adaptive evolution. (After Turner, J.R.G. 1981. Adaptation and evolution in Heliconius: A defense of neoDarwinism. Annual Review of Ecology and Systematics 12: 99–121. © 1981 by Annual Reviews, www.annualreviews.org. Used with permission.)
I illustrate adaptation as a process with a strong empirical example of evolution by natural selection in the microevolutionary mode. Adaptation of human populations to resist malarial infection is perhaps the best case study in terms of documenting evolutionary change at the both the molecular genetic and phenotypic levels and in measuring critical environmental variables. My discussion draws on Templeton’s (2006) synthetic analysis of relevant medical and epidemiological data with comments on the respective roles of character adaptation versus exaptation.
The medical syndrome of sickle-cell anemia in central African populations is perhaps the best-known case of evolution by natural selection at the molecular population-genetic level. Epidemic malaria was likely established in Central Africa as a consequence of agricultural practices introduced there about 2000 years ago (Templeton, 2006). The gene whose variation illustrates selectively guided change is the gene encoding β-hemoglobin. The most common and inferred ancestral allelic form of β-hemoglobin possesses as its sixth amino acid glutamic acid and is called “hemoglobin A.” A single mutational change to hemoglobin A substitutes valine at this position to produce an alternative allele called “hemoglobin S.” The S allele could have arisen from A more than once by mutation in different human populations. Hemoglobin S has a genetically dominant phenotype of malarial resistance. Hemoglobin S molecules form sickle-shaped aggregations within an erythrocyte under low-oxygen conditions; erythrocytes distorted by these aggregations are promptly destroyed by the spleen. Infection of an erythrocyte by a malarial parasite deprives the cell of oxygen, causing sickling; the spleen then destroys the infected erythrocyte and its malarial parasite before the parasite has completed its life cycle. A person heterozygous for the A and S alleles thereby gains resistance to malaria from hemoglobin S, and hemoglobin A has a genetically dominant phenotype for normal respiration under most environmental conditions. An individual homozygous for the S allele suffers the severe respiratory disability called “sickle-cell anemia.” Severe anemia is therefore a recessive phenotype of hemoglobin S.
Before the introduction of epidemic malaria into Africa, the S allele would have been kept rare by natural selection because the SS homozygous individuals have greatly diminished chances of surviving to adulthood. The allele is preferentially removed from the gene pool by natural selection when it occurs in the SS genotype. If mating is random with respect to genotypic variation at the β-hemoglobin locus, the rare S allele occurs almost exclusively in heterozygous AS genotypes, and its selective consequences depend mainly on its phenotypic consequences in the AS genotype (a consequence of Hardy-Weinberg equilibrium; see chapter I.15).
In a malarial environment, selection favors AS individuals, thereby increasing the frequency of the S allele. Occurrence of the S allele at frequencies exceeding rarity is strongly geographically coincident with epidemic malaria (figure 2). An increase in frequency of S leads to more common occurrence of the selectively disfavored SS genotype producing a “balanced polymorphism” in which selection maintains both alleles in the population and moves allelic frequencies toward selective-equilibrium frequencies. The relative Darwinian fitnesses of the three genotypes in a malarial environment are AA (0.9), AS (1.0), and SS (0.3); the genotype with highest fitness is usually denoted 1.0 so that relative fitnesses of the other genotypes are expressed as a fraction of the optimal one. At selective equilibrium, the expected frequencies of the alleles in the population are A (0.89) and S (0.11). Because epidemic malaria in central Africa is a relatively recent introduction, many of these populations have not attained selective equilibrium for the hemoglobin ß-polymorphism.
Hemoglobin S arose by mutation before the establishment of epidemic malaria in central Africa; in the malarial environment, it was co-opted for fitness consequences in AS genotypes that are incidental to the mutational origin of hemoglobin S. Hemoglobin S is thus an exaptation in the evolutionary history of central African populations. The balanced polymorphism of alleles A and S constitutes a population-level adaptation. Eradication of malaria would convert the polymorphism from adaptation to what Baum and Larson (1991) call a secondary disaptation; an environmental change causes a character that formerly had a selective advantage over its evolutionary antecedent (allele A close to fixation in this example) to one that is selectively disfavored relative to that condition. A polymorphism for the A and S alleles at the malarial selective-equilibrium frequencies is selectively disadvantageous relative to a population fixed for the A allele in an environment from which malaria has been eradicated; when S occurs at a frequency of ~0.1, approximately 1% of the polymorphic population suffers severe anemia (SS individuals), and the S allele no longer confers a selective advantage in AS genotypes. As the frequency of S drops to very low values in a nonmalarial environment, the S allele occurs strictly in heterozygous individuals, and presence of hemoglobin S ceases to generate natural selection. Alternative forms that have no selective consequences constitute nonaptation (a selectively “neutral” character lacking current utility; Vrba and Gould, 1986).
Figure 2. Evolution by natural selection at the β-hemoglobin locus in human populations of Africa. Polymorphisms for the hemoglobin C (circle and arrows) and hemoglobin S (crosshatching) forms of β-hemoglobin are associated geographically with occurrence of epidemic malaria (Plasmodium falciparum, gray shading). Hemoglobin A is the most common allelic form in all areas. The frequency of allele C is highest (~0.10) in the circled area and declines gradually with distance outside the circle through the region marked by arrows. Frequencies of C and S alleles are inversely correlated in West Africa because as C becomes more common, natural selection favors C and disfavors S. Where C is absent or rare, natural selection moves S toward an equilibrium frequency of approximately 0.11. Depending on its environmental and populationgenetic contexts, hemoglobin S can be an exaptation, nonaptation, or disaptation for survival to adulthood.
I illustrate the importance of geographic variation in adaptive evolution by extending this example from central African populations to western African ones (figure 2). A third form of β-hemoglobin allelic to hemoglobin A and hemoglobin S occurs in some African populations (Templeton, 2006): the hemoglobin C allele, derived by a single mutation from hemoglobin A, differs from both hemoglobin A and hemoglobin S in having lysine at the sixth amino-acid position. Allele C has a genetically recessive phenotype of malarial resistance and is associated with severe anemia only when heterozygous with allele S. In a malarial environment, the relative Darwinian fitnesses of the six genotypes formed by the various combinations of the A, C, and S alleles are AA (0.7), AC (0.7), AS (0.8), CC (1.0), CS (0.5), and SS (0.2). The C allele likely arose from A in Africa before the introduction of epidemic malaria, when the A allele was close to fixation and S was rare. Under these conditions, the C allele would occur entirely in AC genotypes and would be a nonaptation relative to the A allele. On malarial introduction, selection increased the frequency of S through the higher viability of AS genotypes over AA and AC genotypes. As the frequency of the S allele increases toward the selective equilibrium frequencies (A = 0.89, S = 0.11), the C allele would appear in CS genotypes as well as AC genotypes, and selection would decrease its frequency by disfavoring CS individuals. Selection acts to keep the C allele rare under these conditions, in which C is a disaptation relative to A. Although the CC genotype is the most favorable possible condition, the CC genotype is too rare under conditions of random mating to contribute a selective advantage.
Because alleles A and C are selectively equivalent in a nonmalarial environment, random genetic drift (see chapter I.15) might increase the frequency of C in a local population to a frequency high enough that CC genotypes occur regularly. Introduction of epidemic malaria into such a population would act very differently than it does in the population described in the preceding paragraph. Over many generations, the net effect of natural selection would be to increase frequency of the C allele toward fixation and to decrease frequencies of A and S ultimately to zero. Allele C constitutes an exaptation for malarial resistance in some western African populations (figure 2) in which allele C had drifted to sufficiently high frequencies before epidemic malaria that CC individuals were produced and subject to selective retention in a malarial environment. The contrasting selective consequences of the A, C, and S alleles of β-hemoglobin under slightly different starting frequencies of C and S in African malarial environments show that adaptive evolution depends critically on particular historical conditions.
Another dimension to adaptive evolution is gene exchange among different geographic populations of a species. As western African populations evolve by selection to increase the frequency of the C allele and central African populations evolve toward selective-equilibrium frequencies of the A (0.89) and S (0.11) alleles of β-hemoglobin, gene exchange occurs by interbreeding among these populations. Because fixation of allele C is the superior adaptive condition with respect to these three alleles, preferential contribution of C alleles from the favored west African genotypes should enable all populations eventually to undergo adaptive evolution in the manner of the west African populations by fixing C and eliminating A and S.
Further analysis of geographic variation in evolution of malarial resistance by Templeton (2006) reveals several cases in which genetic epistasis between the sickle-cell polymorphism at β-hemoglobin and variation at other loci produces different kinds of adaptive evolution. In Greek and Arabian populations, hemoglobin S occurs against a genetic background in which a mutation at a genetically linked locus causes fetal hemoglobin to be expressed throughout adulthood (called “persistence of fetal hemoglobin”) rather than ceasing its expression after birth. This combination of alleles at two loci provides a simple example of a “co-adapted gene complex”; hemoglobin S provides malarial resistance, while persistence of fetal hemoglobin alleviates the severe anemia associated with SS homozygotes. High fitness depends on a particular combination of alleles at these two genes.
One expects evolution of many advantageous phenotypes to occur by natural selection increasing the frequencies of selectively favored alleles at many genes. A consequence of such selection is that new mutations arising in different individuals and at different times in a population’s history can be brought to high frequency independently by selection, thereby increasing their chances of occurring together in the same individuals by genetic recombination. If the combination of alleles thus achieved is favored by selection for the same biological role that brought the individual mutations to high frequency, selection gradually constructs an adaptive composite character. Such characters can become developmentally and genetically integrated into modules, whose expression or suppression during development can provide a store of potential exaptations, sometimes called a “toolkit” for constructing new organismal forms. A toolkit of adaptively evolved modules makes possible further evolution not confined to traditional Darwinian gradualism. A developmental module can be activated potentially by genetic or environmental factors or their interactions; “phenotypic accommodation” denotes a beneficial modification of organismal development made in response to a novel behavioral or environmental stimulus (West-Eberhard, 2005).
Hypotheses of developmental modularity and phenotypic accommodation share the expectation that evolutionary saltations are more likely than acknowledged by traditional Darwinism. A well-studied case of polymorphism for discrete developmental modules involves two constrasting feeding morphologies in tropical American fish of the genus Cichlasoma. The contrasting forms differ abruptly in the structures of jaws located in the pharynx and used to crush food (figure 3). The alternative states of the pharyngeal jaws are termed the “papilliform” morph versus the “molariform” morph; the molariform morph has hypertrophied skeletal and muscular components and greater ability to masticate hard food items, such as snails, which are often the less preferred food items. In some species, consuming snails early in life appears facultatively to trigger development of the molariform morph. The morphological contrast between these states in C. minckleyi of Mexico is so great that it has been called “intraspecific macroevolution.” Widespread occurrence of molariform and papilliform feeding morphs among cichlid species indicates that they likely represent alternative developmental modules first assembled in the ancient history of cichlid fishes. The evolutionary origin of the alternative morphs is a macroevolutionary question, testable using interspecific phylogenetic analyses of how one form was constructed, perhaps gradually, from the other one. Microevolutionary studies of the alternative conditions as polymorphisms within species involve saltational changes governed by developmental switches and a prominent role for character exaptation in the adaptive trophic evolution of polymorphic populations.
An important extension of the Darwinian evolutionary framework is to recognize semiautonomous selective processes occurring at the levels of genomic elements and species lineages in addition to the traditional level of varying organisms within populations. The abstract concepts of character sorting and selection have been expanded to encompass these levels (Gould, 2002; Vrba and Gould, 1986). At the genomic level, characteristic structures of retrotransposons include long terminal repeats and a coding region for reverse transcriptase, whose biological role is integral to the operation of the transposable element but whose origin cannot be explained as an organismal-level character adaptation. Unlike the genes encoding β-hemoglobin, mutational changes in transposable elements cannot be interpreted as having reached high evolutionary frequency to serve an organismal-level function, although their consequences can be co-opted as exaptations for organismal-level roles. Burt and Trivers (2006) use “selfish genetic element” to denote “the minority of genes that spread at a cost to the organism” and “stretches of DNA... that act narrowly to advance their own interests” and “typically cause negative effects on non-linked genes” in the same organism. For example, during spermiogenesis of male mice heterozygous for the t-allele, developing sperm containing the t-allele disable those containing the wild-type allele, permitting the t-allele to persist in populations despite selection against its detrimental effects on organismal phenotype, such as absence of a tail. Someone studying occurrence of the tailless phenotype in mice in a Darwinian context would conclude that this trait is a primary disaptation (Baum and Larson, 1991), a character disadvantageous relative to its ancestral alternative condition in the environmental context of its origin. Identification of primary disaptation using the adaptationist methodology described above would lead one to hypothesize association of such a phenotype with evolution of selfish genomic elements.
Figure 3. Cineradiographic tracing of the upper and lower pharyngeal jaws contacting to grind a snail in the molariform morph of Cichlasoma minckleyi. Arrows denote the directions of movement of the jaws as they make contact. This hypertrophied molariform morph and a much smaller papilliform morph represent alternative developmental modules in this and other species of Cichlasoma. The molariform morph can masticate hard food items, such as snails, not available to the papilliform morph. (After Liem, K. F., and L. S. Kaufman. 1984. Intraspecific macroevolution: Functional biology of the polymorphic cichlid species, Cichlasoma minckleyi. In A. A. Echelle and I. Kornfield, eds., Evolution of Fish Species Flocks. Orono: University of Maine Press, 196–217)
Selfish genomic elements that harm their “host” organism incur natural selection for their suppression by other genetic functions. One therefore expects to observe evolution of organismal-level adaptations whose function is to suppress selfish genomic elements. Given the phylogenetically widespread occurrence of the selfish genetic elements reviewed by Burt and Trivers (2006), mechanisms evolved to stabilize genomic structure and function probably constitute a large portion of the adaptive evolutionary diversity of life. The basic concepts of adaptation and exaptation and historical methods for testing specific hypotheses of adaptation as discussed above are directly applicable to studies of inherent conflicts between organismal character adaptation and the proliferative drives of selfish genomic elements.
Baum, D. A., and A. Larson. 1991. Adaptation reviewed: A phylogenetic methodology for studying character macro-evolution. Systematic Zoology 40: 1–18.
Burt, A., and R. Trivers. 2006. Genes in Conflict: The Biology of Selfish Genetic Elements. Cambridge, MA: Harvard University Press.
Gould, S. J. 2002. The Structure of Evolutionary Theory. Cambridge, MA: Harvard University Press.
Gould, S. J., and E. S. Vrba. 1982. Exaptation: A missing term in the science of form. Paleobiology 8: 4–15.
Rose, M. R., and G. V. Lauder. 1996. Adaptation. San Diego: Academic Press.
Templeton, A. R. 2006. Population Genetics and Micro-evolutionary Theory. Hoboken, NJ: John Wiley & Sons.
Vrba, E. S., and S. J. Gould. 1986. The hierarchical expansion of sorting and selection: Sorting and selection cannot be equated. Paleobiology 12: 217–228.
West-Eberhard, M. J. 2005. Phenotypic accommodation: Adaptive innovation due to developmental plasticity. Journal of Experimental Zoology 304B: 610–618.
