Evolution, in the modern sense, refers to changes in the genetic composition of populations over time and is the result of natural selection and/or genetic drift acting on population variation. In this Darwinian paradigm, species may change or split into more than one species (speciation). All extant species are descendants of a common ancestor (descent with modification). Definition of the term in this framework encompasses the gene-frequency changes of population microevolution, anagenetic changes within a lineage, cladogenesis, and the appearance of evolutionary novelties. The present diversity of organisms was produced by change within species (anagenesis) and splitting of species (cladogenesis) through geological time, in contrast to the explanations offered by separate creation and Lamarckian transformism. The former denies both the mutability and common ancestry of species, and the latter invokes change within species but denies the splitting and common ancestry of species. Historically, the term evolution was also used to describe development of the embryo and in the theory of embryonic recapitulation.
Evolution is a process that occurs in populations rather than in individuals: The genetic composition of a population changes, but that of the individual does not. Evolution occurs in populations by changes in the frequencies of alleles and genotypes. The changes at the genetic level are observed in the phenotypes of individuals (e.g., protein structure, individual morphology, behavior).
Ernst Mayr identifies five major postulates that form the foundation of the Darwinian paradigm: (1) organisms change through time, (2) all organisms have a common ancestor, (3) species multiply by splitting or budding, (4) gradualism, and (5) natural selection. It should be pointed out that the Darwinian concept of gradualism is not that evolutionary change is necessarily slow and steady. Rather, it is that evolution does not occur by saltations, the results of macromutations that render offspring reproductively isolated from the parental generation, sensu the writings of Richard Goldschmidt.
Evidence for Evolution
Matt Ridley points out that two fundamental questions need to be answered to demonstrate evolution. Do populations and species change through time? And do living organisms share a common ancestor? The evidence used to answer these questions in the affirmative comes from a number of sources, including observation of evolution on a small scale, the presence of extensive variation among populations and subspecies, homology, adaptive radiation, and the fossil record.
Evolution “on a small scale,” over a number of generations rather than over millions of years, has been achieved under domestic and experimental conditions and observed in the wild. The origin and development of domestic breeds of animals and plants has been a topic of interest to anthropologists and clearly demonstrates that species change. In On the Origin of Species (1859), Charles Darwin discusses the production of domestic breeds as the result of “accumulative selection” by humans and points out that breeders and horticulturalists intentionally modify breeds within a human lifetime. He particularly concentrates on the origin of pigeon breeds. Darwin also addresses “unconscious selection,” in which a breed is altered simply by breeding the “best” individuals (for example, the English pointer dog). In both techniques, the breeder is allowing only those individuals with extreme values of a particular character to produce offspring during each consecutive generation (the “traditional breeder’s approach”). This shifts the mean value for the selected characteristic during successive generations in the descendants: The breeder is applying directional selection to the particular trait of interest (the somewhat unfortunate term “artificial selection” is often applied to this methodology, which could be misconstrued). Darwin expands upon this discussion in The Variation of Animals and Plants Under Domestication (1883) and states that domestication is a gigantic experiment in selection. Modern discussions of the principles and results of plant and animal breeding are couched in terms of genetics.
In an experimental context, both the traditional breeders’ approach and natural selection in a controlled environment have been employed to demonstrate change in populations. Well-known examples of the former have focused on disruptive selection on the number of sternopleural chaetae (stout hairs) in the fly, Drosophila melanogaster. These experiments not only sought to demonstrate that change in mean population chaetal number would result from differential reproduction but also tested whether disruptive selection could overwhelm the results of gene flow—a central issue in sympatric and parapatric speciation models. The general approach was to allow flies with low chaetal number and flies with high chaetal number to breed, but not allow breeding by those with intermediate numbers. The results consistently showed shifts in mean population numbers over time. The same approach was employed in a behavioral context by J. C. Tryon, a pioneer of behavioral genetics, in his famous experiment on maze-running ability in rats. Later applications include the work of Rodriquez-Ramilo and associates on the effects of inbreeding and selection on D. melanogaster.
In experiments that employ natural selection in a controlled environment, subjects are exposed to a predetermined environmental regimen for one or more generations, but the experimenter does not directly determine which individuals breed. D. N. Reznick and associates used this approach in field experiments in which guppies were moved from areas where they were subject to predation by cichlid fishes to sites where they were not; the descendants of guppies from sites with less predation matured later and at a larger size than descendants of those from high predation sites. The approach is particularly valuable for the study of niche dimensions, range limits, and character displacement.
Demonstration of the occurrence of evolution in the wild often falls within the discipline termed ecological genetics, a combination of naturalistic observation, experimentation in the field, and the use of laboratory genetics. Ecological genetics seeks to establish that change occurs within populations and species, to demonstrate that there is a hereditary basis to phenotypic variation, and to identify the agent of change. The work of J. H. Gerould on color polymorphism in caterpillars is cited as an early example.
A common methodology employs the following components: (1) observations and collections made to establish that distinct phenotypes (generally of a trait that manifests clear polymorphism) are found in the wild, (2) establishing that there is a change in the proportions of the phenotypes over time or geographic distance, (3) determination that there is a genetic basis to the various phenotypes, and (4) identification of the agent that is responsible for altering the frequencies of phenotypes and genotypes. The latter step is often the most difficult, because the occurrence of a feature may be influenced by more than one process. In Ecological Genetics, Edmund B.
Ford synthesizes the extensive literature on shell polymorphism in Cepaea nemoralis, chromosome polymorphisms in Drosophila, mimetic polymorphism in butterflies such as Papilio dardanus, and industrial melanism.
Industrial melanism is generally cited in discussions of evolution demonstrated on a small scale. A classic example has been that of wing and body color polymorphism in the English peppered moth, Biston betularia. Michael Majerus states that melanism occurs whenever there is a pervasive darkening of the ground color or patterning of an organism. The term industrial melanism applies to darkening as a result of industrial pollution of the environment. The first-known melanic specimen of the peppered moth was captured in 1848, in Manchester, and melanics rapidly increased to 98% to 99% of the population by 1895 to 1898. The books of Ford, Bernard Kettlewell, and Majerus summarize literature on the phenomenon dating back to 1864. It is often held that the origin of the melanic carbonaria was a single point mutation in the region around Manchester, with subsequent migration to other areas, but Majerus indicates that there may have been more than one mutational event. Early explanation of the occurrence of melanism invoked Lamarckian transformism. A mutation pressure explanation was advanced by J. W. H. Harrison in 1928. He argued that melanism was induced in adults as a consequence of larvae feeding on leaves contaminated with mutagens that increased the mutation rate of genes associated with melanin production. However, it has never been confirmed that the influence of environmental stimuli during development can induce melanism in adult Lepidoptera.
The Darwinian natural selection explanation replaced those of the inheritance of acquired characteristics and mutation pressure. The primary explanation of the geographic spread, and numerical increase of carbonaria became that of selective advantage of the melanic form over that of the lighter typica in areas of industrial air pollution, because of camouflage from insectivorous birds. The occurrence of industrial melanism in Lepidoptera gave rise to a series of elegant studies of Biston betularia, summarized by Ford, Kettlewell, and Majerus, and Kettlewell’s work had a significant impact on the acceptance of natural selection as an agent responsible for evolution. His investigations included a controlled experiment on bird predation on moths in an aviary to construct quantitative degrees of camouflage effectiveness, field experiments in polluted (Christopher Cadbury Bird Reserve, Birmingham) and unpolluted (Deanend Wood, Dorset) localities, and surveys of peppered-moth frequencies throughout Britain. The fieldwork included mark-recapture, studies to determine whether moths released in early morning were still at the release point later in the day, and (with Niko Tinbergen) filming of actual bird predation on moths. Kettlewell concluded that differential selective predation by birds was responsible for differences in the relative frequencies of typica and carbonaria in areas that had become polluted.
It has been realized that the classic exposition of natural selection on Biston betularia is too simple, in part because camouflage and predation do not seem to be important in maintaining melanism in some other taxa. A variety of workers have studied industrial melanism in the ensuing decades since Kettlewell’s landmark work. The term industrial melanism is now applied to taxa such as the two-spot ladybird and the pale brindled beauty moth, without the implication that predation is responsible for maintaining the polymorphism. Kettlewell himself pointed out that melanism affects many aspects of an organism’s life, including behavior, sexual selection, temperature tolerance, and physiology of pigment production; natural selection could potentially affect the moth population through any of these features, among others. Publications by Bruce R. Grant and Majerus review many of the major criticisms, including that large numbers of released moths produced artificially high prey densities and may have attracted birds; moths settled on tree trunks, but it is unclear whether this is a natural hiding place for B. betularia; moths released during daylight hours do not behave normally; a combination of lab-raised and wild-caught moths were used; there are differences between human and bird vision; and there was no evidence that typicals and melanics prefer to rest on different backgrounds—contra Kettlewell’s barrel experiments. However, these authors conclude that such shortcomings do not jeopardize the principal conclusions that the intensity of bird predation varies according to habitat and that industrial melanism in the peppered moth is a valid example of natural selection and evolution in the wild. There was a solid correlation between pollution levels and the frequency of the melanic form. Kettlewell compared the relative success of the morphs on the same parts of trees in different areas (not different parts of trees in the same area), and he showed that moths that were most conspicuous to humans were also eaten first by birds. Kettlewell’s conclusions are bolstered by recent studies of predation on male mosquitofish by Lisa Horth.
Geographic variation in intraspecific phenotypes and proportions of genotypes provides evidence for evolution. If a species is studied in the wild, it will generally be found that there are phenotypic variants in different geographic areas. Often, these variants are designated as subspecies of the species; variation is often so extensive that small differences in samples of populations have been used to assign subspecific rank, and some have argued that the subspecies category has been overutilized. Examples of variation within the primates include the geographic variants found within sub-Saharan vervet monkeys (Cercopithecus aethiops), South American white-fronted capuchins (Cebus albifrons), tufted capuchins (Cebus apella) and common squirrel monkeys (Saimiri sciureus), and the northern greater galago (Otolemur garnetti). Interpopulation variability is often readily interpreted in terms of environmental factors to which each population has adapted, as in the cases of Peromyscus mouse populations to substrate color and cyanogenic clover (Trifolium repens) to probability of frost. The latter also illustrates that interpopulation variation can be distributed in the form of a cline, a gradual change in a phenotypic character or genotypic along a geographic transect. In humans, there is a cline in the frequency of B type blood across Eurasia. A cline may be formed by interbreeding between formerly isolated populations or by geographic variation in selection pressures on a character. Descent from a common ancestor and subsequent local evolution are the most logical and parsimonious explanations for geographical variants. Special creation and transformism would have to explain each local variant as having a separate origin. Explanation by Lamarckian transformism would be particularly hard to envision, as it would necessitate separate origin by spontaneous generation for each population, followed by the inheritance of acquired characteristics that would make populations separated by hundreds of miles sufficiently different to be distinguished only at the subspecies level.
Powerful evidence from geographic variation comes from ring species, in which there is an almost continuous set of intermediates (cline) between two sympatric species, and the intermediates are arranged in a ring. The set of intermediates between the lesser black-backed gull (Lams fuscus) and herring gull ( L. argentatus) is distributed around the North Pole. As one moves along the ring, there is only one species present in any local area. However, there are reproductively and ecologically separate species where the two end points of the ring meet in Europe. Mayr terms this “speciation by distance.” Ring species demonstrate that intraspecific variation can be great enough to make two species and that there is a continuum from interindividual to interspecific variation. Recent research on greenish warblers (Phylloscopus trochiloides) by Darren E. Irwin and colleagues, using amplified fragment length poly““““““““““““““““““““““““““““morphism markers, provides evidence that specia-tion by distance can produce distinct species despite gene flow along the ring.
Homology is the possession by two or more species of a trait derived, with or without modification, from a common ancestor. Gunter P. Wagner terms it the central concept of comparative morphology. It is also robust evidence for evolution, for structures are homologous because of common descent, not because of close similarity in function. Wagner explains that homologous structures (homologs) are those that have been acquired only once in the history of a group of organisms possessing the character; those features of a development system that cause restriction in phenotypic consequences of genetic variation are important in determining that features are homologs. Well-known examples are the pentadactyl hand of a primate and the pentadactyl wing of a bat: They both possess five digits because of descent from a common ancestor, not because of close similarity in function.
Wagner also points out that the study of homologs is the basis for the reconstruction of phylogenetic history by morphological methods. This has historically been true in the study of adaptive radiations before the advent of molecular phylogenetic approaches. An adaptive radiation occurs when a single ancestral species diversifies into a large number of descendant species that occupy a variety of ecological niches and differ in traits used to exploit those differing niches. Articles in Molecular Evolution and Adaptive Radiation (1997), edited by Thomas J. Givinish and Kenneth J. Sytsma, give alternate definitions. Well-known examples include the radiations of placental and marsupial mammals, Hawaiian silver-sword plants, African cichlid fishes, and Caribbean Anolis lizards. All 14 species of Galapagos finches share finch homologies and are members of the single subfamily Emberizinae of the finch family Fringillidae. In the even more diverse radiation of Hawaiian honeycreepers, all species share finch homologies. Although beak size is a continuously varying character governed by both genetic and environmental factors, Galapagos finch and Hawaiian honeycreeper species manifest beaks adapted to particular foraging strategies and diverse niches.
The evidence for evolution that is most often cited is from the fossil record. Four aspects of the fossil record are particularly important: Many species are extinct; fossil and extant forms in the same geographic region are related to each other (law of succession); the history of the Earth has been one of environmental change; and species have changed through time and there are transitional forms in the fossil record. The latter aspect of the fossil record is one that receives particular attention, in terms of both describing forms that are transitional between major taxa and in documenting evolution within lineages and clades.
One of the presumed problems with using the fossil record as evidence for evolution is that there are gaps between fossil forms and a smooth transition is not always documented. Darwin listed this as the first difficulty of the theory of descent with modification in On the Origin of Species (1859) and sought to answer the objection by pointing out the imperfect and intermittent nature of preservation of “parent-forms” and intermediate links. This explanation has been used subsequently on many occasions, and workers have sought to chronicle the resolution of the fossil and archeological records.
A particularly well-known fossil history is that of horses. Since the early work of Kovalevsky, many eminent paleontologists (including the rivals O. C. Marsh and Edward Drinker Cope) have studied horse evolution. The writings of George Gaylord Simpson have made the evolution of the horses one of the best-known, and debated, examples of long-term evolution. Simpson’s work has sometimes been characterized as advocating that the morphological trends in dental morphology, crown size, body size, skull size, preoptic dominance, brain size, and limb ratios were the result of phyletic evolution within a direct phylogenetic lineage, that is, a single line of gradual transformation from Eocene Hyracotherium to modern Equus. However, in The Major Features of Evolution (1953), Simpson points out that horse evolution is characterized by repeated and complex splitting (he specifically states that it is “bush-like”). He notes that there were no trends that continued throughout the history of the family in any line.
Figure 5a -Trends in the evolution of horses. Simpson termed the early Miocene “the great transformation” of horse anatomy correlated with the spread of grassland. 5A presents the evolution of the skull, showing the increase in size and changes in cranial proportions including lengthening of the preorbital region; 5B depicts the trend toward increased brain size and complexity; 5C shows the trend toward increased crown height in upper molar teeth; 5D illustrates the molarization of premolars in the Eocene; 5E further depicts evolution of horse cheekteeth from brachydont, browsing teeth to hypsodont, grazing teeth for transverse shearing of abrasive grasses; 5F provides selected stages in the evolution of the forefoot, with a reduction of metapodials from 4 in Hyracotherium (eohippus) to 3 in Mesohippus and eventually to 1 in “Pliohippus” and Equus. Simpson viewed each column as a distinct mechanical type.
Contrary to creationist views, horse fossils chronicle large-scale evolutionary change even if the various genera were separated from one another by tens of millions of years. This is true regardless of whether horse evolution is interpreted in the context of anagenesis or as an example of “bushy,” cladogenetic evolution. Controversies regarding whether evolution has occurred within a lineage or clade by phyletic evolution or punctuated equilibrium are debates regarding the tempo and pattern of morphological change, not whether change has occurred—both patterns document evolution. The same is true of contrasting interpretations of hominin evolution. However, creationist tracts sometimes cite sources emphasizing rectangular patterns of evolution to argue that evolution does not occur because of a purported lack of transitional forms; they demand evidence of phyletic evolution to support evolution as a fact.
Robert L. Carroll summarizes several well-documented cases of phyletic, directional evolution in late Cenozoic mammals in Patterns and Processes of Vertebrate Evolution (1997), including increase in length of limb bones and in the number of osteoderms in the giant armadillo (Homesina) lineage from Florida, morphology of the upper and lower molars in the sagebrush vole (Laqurus curtatus) over 287,000 years, and change in size and thickness of enamel in the molars of the water vole (Arvicola cantiana) over 300,000 years.
Examples of phyletic evolution have been chronicled for the primate fossil record, including within the Eocene Cantius lineage by W. C. Clyde and P. D. Gingerich and the Eocene Tetonius-Pseudotetonius lineage by Kenneth Rose and Thomas Bown. Meave Leakey discusses phyletic evolution in the Theropithecus oswaldi baboon lineage, from the upper Burgi Member Koobi Fora Formation to upper Bed II of Olduvai Gorge, in characters such as increase in complexity of molars and their size compared with those of anterior teeth, decrease in size of canines and in the sectorial cusp of P3, development of a “reversed cusp of Spee” on cheekteeth, increase in size of the glenoid process, and increase in estimated body weight. She interprets the evidence as representing an unbranched, evolving lineage that can be divided into the three subspecies: T. oswaldi darti (earliest in time and smallest in size), T. o. oswaldi (intermediate) and T. o. leakeyi (latest and largest). Gerald G. Eck also interprets it as a phyletically evolving lineage but prefers to assign the earliest material, from Hadar and Makapansgat, to the species T. darti rather than using subspecific designation.
The recording of the tempo and pattern of change in individual characters is the methodology used in the studies cited above. It is preferable to plotting the range of species to document the pattern of evolution. The results produced by the latter often depend on the degree of variability in morphological parameters of each named species or genus, as well as on whether species are recognized on the basis of meristic characters or on those that vary continuously. Carroll states that the character-based approach often shows that evolution occurs in a mosaic manner: that different characters, or the same character at different stratigraphic levels, will show different tempos and patterns of evolution—an observation also discussed by Simpson in reference to horse evolution.
Variation and the Processes That Affect It
Genetic variation in characters is the raw material for evolution. As pointed out by Ronald Fisher in the fundamental theorem of natural selection and supported by subsequent experimentation by Francisco Ayala, evolution can occur only in populations that manifest genetic variation. Population geneticists have therefore spent considerable amounts of effort to measure variation. Since the pioneering work of Richard Lewontin and J. L. Hubby, gel electrophoresis has often been the technique employed, and estimates of genetic variation have been made by electrophoresis of proteins and both mitochondrial and nuclear DNA. Mean heterozygosity and the percentage of polymorphic loci are then calculated. Although John H. Gillespie argues that electrophoretic studies of enzymes may overestimate the amount of protein variation (because these proteins may be more polymorphic than are “typical” protein loci), the basic conclusion has been that natural populations are very variable. Jeffrey B. Mitton, in Selection in Natural Populations (1997), states that 33% to 50% of enzyme loci are polymorphic and the average individual is polymorphic for 4% to 15% of its genes, but there is considerable variety in the percentage of genes that are lymorphic, from 0% in the modern cheetah and northern elephant seal to 100% in the mussel Modiolus auriculatus. Studies of plants and of Peromyscus mice suggest that the amount of genetic variation increases with the size of the geographic range.
High levels of variation counter the lassical model,” associated with Herman J.Muller, which held that populations in the wild would manifest little genetic variation: The model stulated that one allele of each gene would unction best in each deme and would therefore be favored by natural selection. Almost all members of the population would be ozygous for this “wild-type” allele. If a superior new mutation arises, it eventually becomes the new ild type. A population would evolve, but there would not be much genetic variation at any one time in the population. Mutations would be deleterious or have adaptive value, but would not be neutral in their effects.
With the discovery of large amounts of genetic variation in populations, two major competing models (balance and neutral mutation) sought to explain the maintenance of variation. The balance model, associated with Theodosius Dobzhansky, observes that many individuals are heterozygous at many loci: No single allele is best, and balancing selection would prevent a single allele from reaching very high frequency. Evolution would occur by gradual shifts in the frequency of alleles. Examples of heterozygote advantage, in which the heterozygote is fitter than is either homozygote (as in sickle-cell trait), bolster this argument. The neutral-mutation hypothesis (termed by Mark Ridley, the “purely neutral” theory) states that much, if not most, variation is selectively neutral because different genotypes are physiologically equivalent. Evolution is not driven so much by natural selection acting on alternate phenotypes/genotypes, but by genetic drift. The purely neutral hypothesis predicts that larger populations will contain more genetic diversity than smaller ones, a prediction that appears to be contradicted by evidence.
Five processes influence the amount of variation in a population: mutation, recombination, gene flow, genetic drift, and natural selection. Mutation is any heritable change in the DNA, including both point mutations (for example, alleles that produce the varieties of hemoglobin) and chromosomal mutations (for example, polyploidy, polysomy). Recombination (crossing over), a process by which segments of homologous chromosomes are exchanged during Prophase I of meiosis, produces new combinations of alleles on chromosomes. Gene flow is the incorporation of alleles into the gene pool of a population from one or more other populations. The archetypal view is that it homogenizes genetic composition if it is the only operating factor; that is, the amount of gene flow between local populations influences the degree to which each deme is an independent evolutionary unit, for if there is little gene flow, then each deme evolves independently from others. Gene flow would not only deter genetic drift but also potentially constrain the adaptive divergence of populations living in different environments. This issue is important because models of sympatric and parapatric speciation assume that if selection is strong enough, it can overwhelm the effects of gene flow and adaptive divergence and speciation can occur. Recent work by Andrew P. Hendry and Eric B. Taylor on lake and stream populations of the threespine stickleback fishes (Gasterosteus aculeatus) suggests that gene flow may constrain adaptive divergence, but to varying degrees, and that different traits do not respond in the same way to the same amount of gene flow.
Genetic drift (the “Sewall Wright effect”) is the random fluctuation of allele frequencies in effectively small populations: populations in which the number of individuals that actually breed is small. The basic premise is that small samples are often not representative of the range of genetic variation within a population and that the frequency of an allele in a small sample is unlikely to be equal to its frequency in the entire gene pool of a large population. Genetic drift is sometimes subdivided into continuous drift, intermittent drift, and the founder principle. The founder principle refers to situations in which populations are begun by a small number of colonists, which carry a fraction of the variability of the parental population, and allele frequencies differ from those of the parental population. When a population is subject to drift, the frequency of an allele will randomly change from one generation to the next until the allele is either fixed or lost (see Figure 7). It, therefore, reduces the amount of genetic variation within a population. Genetic drift has been produced in experimental populations of Drosophila, extensively treated mathematically, and modeled by computer simulation. It has been discussed in the microevolution of aboriginal Australians, the high incidence of hereditary eye disease and clinodactyly in the human island population of Tristan da Cunha, Ellis-van Creveld dwarfism in eastern Pennsylvanian Amish, and oculocutaneous albinism Type 2 in the Navajo.
Natural selection was defined by Darwin as the preservation of favorable individual differences and variations and the destruction of those that are disadvantageous. Selection acts directly on the phenotypes of organisms and indirectly on their genotypes. In a genetic sense, alleles that promote greater ability to survive and reproduce in a particular environment (higher fitness) will leave more offspring than do less fit genotypes. The difference in frequency between various genotypes and phenotypes is a result of differences in ecology and not because of chance. Selection can occur at any stage of the life span, but it has maximal evolutionary impact when it acts on those age classes with the highest reproductive values. There are several basic premises: Phenotypic variation exists between individuals in a species; species have more offspring than can survive and reproduce; individuals with different phenotypes vary in their ability to survive and reproduce in a particular environment; and a portion of the ability to survive and reproduce is hereditary. While selection can cause evolution to occur in a population, a population may be at equilibrium as a result of natural selection and other processes.
It has been posited that selection can occur at any level of organization that fulfills the above premises: individual genes or larger parts of thegenome (for example, meiotic drive in Drosophila and other taxa), individual organisms that have different genotypes and phenotypes, groups of organisms (discussion generally focused on kin selection), and entire species (species selection, a tenant of punctuated equilibrium and a proposed mechanism to explain evolutionary trends without invoking phyletic evolution). Most of the naturalistic, experimental, and theoretical work centers on selection at the level of the individual. A number of well-known naturalistic cases are cited: melanism in peppered moths, survival of English sparrows during the winter of 1898, shell color polymorphism in British populations of European landsnails, water snakes on islands in western Lake Erie, heavy metal tolerance in the grass Agrostis tenuis on mine tailings, stolon length in populations of Agrostis stolonifera that grow in different environments, Batesian mimicry in Lepidoptera, and immunity to malaria in human populations with sickle-cell trait.
Neutral Theory and the Molecular Clock
While it is clear that the above five processes affect the amount of variation in populations, there is considerable debate over which mechanisms are most important in maintaining and changing the amount of genetic diversity. Natural selection and adaptation have received considerable attention, as has the purely neutral theory of molecular evolution (termed the “neoclassical theory” by Lewontin). In this hypothesis, most evolutionary changes in the DNA do not result from the action of natural selection on variation. Rather, the mechanism of change is the random fixation of neutral or almost neutral mutants by the action of genetic drift. The probability of fixation of a neutral allele in a finite population is equal to the initial gene frequency (p); the rate of gene substitution is equal to the mutation rate per locus. Most intraspecific protein and DNA polymorphisms are selectively neutral, and polymorphic alleles are maintained by a balance between origination by mutation or introduction into a population by gene flow and elimination by random extinction. Therefore, most molecular polymorphisms are not maintained by selection. It is predicted that genes that evolve rapidly will manifest high degrees of intraspecific variability.
Masatoshi Nei, in Molecular Evolutionary Genetics (1987), summarizes major points in the neutralist theory: many mutations at the nucleotide level are apparently neutral or nearly neutral; only a small proportion of mutations are advantageous; natural selection is a process that preserves advantageous mutations and eliminates disadvantageous ones; new mutations spread by genetic drift or by natural selection, but a large proportion of new mutations are eliminated by chance; and populations do not always have the genetic variation for new adaptation. In the neutralist paradigm, natural selection is invoked to explain the loss of disadvantageous mutations, but genetic drift to account for fixation of mutations. This differs from the selectionist approach, which invokes the effects of natural selection to explain both fixation and loss.
Related to the idea of selective neutrality is the molecular clock hypothesis, in which the rate of nucleotide or amino acid substitution is constant per site per year. The clock predicts a stochastically, not absolutely, constant rate of change. In his writings, Moto Kimura assumes that the rate of neutral mutation per year is almost constant among different organisms with very different generation spans, a conjecture that appears to depend on whether mutations are synonymous (silent) or nonsynonymous (replacement). The application of the molecular clock achieved notoriety by the work of Vincent Sarich, in which he concluded that the earliest possible date of divergence of the African apes and hominins was slightly more than 4 million years ago. This suggestion strongly contradicted the dominant paradigm of the period, which placed the divergence in the Miocene with the identification of Ramapithecus as a hominin, and Sarich made the extreme statement that morphology was an unreliable indicator upon which to base estimates of dates of divergence. “Universal” rate calibrations have been suggested for mitochondrial DNA, albumin immunological distances, codon substitutions for cytochrome, myoglobin, alpha and beta hemoglobin, fibrinopeptides A and B and insulin, and the 16S rRNA gene.
The purely neutral and molecular clock hypotheses have received both support and criticism. In The Genetic Basis of Evolutionary Change (1974), Richard Lewontin extensively reviews the evidence for the maintenance of genetic variation available to that date, including the neutralist explanation and the suggestion that variation is maintained by frequency-dependent selection. More recent reviews include those by John C. Avise (Molecular Markers, Natural History, and Evolution, 1994), Gillespie (The Causes of Molecular Evolution, 1991), Roger Lewin ( Patterns in Evolution: The New Molecular View, 1997), Nei (Molecular Evolutionary Genetics, 1987), and Ridley (Evolution, 2004), as well as many professional papers. Discussion can be divided into that which deals with the relative importance of selective neutrality in evolution and that which is aimed at the assumptions and application of the molecular clock.
Because of the large amount of electrophoretic work that has been done on allozymes, examples of protein variation that can be shown to have important physiological effects and impact on fitness are used to counter the neutralist hypo-thesis. Different allozymes should work better in different environments: It should be possible to discover the kinetic and thermostability properties that make one variant work better than others in a particular environment and to find higher percentages in areas where it works better than other variants. There should also be increases and decreases in the relative percentages of allozymes as environments change. Studies that have sought to understand enzyme variation in an adaptive context include those of J. E. Graves and G. N. Somero on lactate dehydrogenase in four barracuda species, Ward Watt and colleagues on phosphoglucose isomerase in Colias butterflies, and D. A. Powers and P. M. Schulte on lactate dehydroge-nase-B in the fish Fundulus heteroclitus.
Figure 6 – Diagram of the evolution of characters in horses. Vertical lines are not proportional to the size of the changes, and the curves show rates, times, and directions of change in a relative way. Simpson’s commentary states that trends were often different in rate and direction in different lines and in the same lines at different times, different trends occurred at different times and rates within a single functional system, a character may not show any trend for a long period even though it may have manifested a trend earlier or later in the fossil record, and a character may change from one “stable adaptive level” to another by a unique sequence of step-like shifts.
Of particular interest are studies that have examined species on either side of the Isthmus of Panama, as the date of the origin of the isthmus is considered to be well established at about 3 to 3.1 million years ago. Results have been used to counter both the interpretation of allozyme variants as selectively neutral and the application of the molecular clock. Gillespie, citing the work of Graves and colleagues on pairs of fish species that are separated by the isthmus, argues that adaptive changes in the enzyme lactate dehydrogenase occur with only a 2-to-3-degree difference in average environmental temperature between the Atlantic and Pacific. Application of Atlantic and Pacific comparisons to the molecular clock has yielded contradictory results, at least in the study of mtDNA. Avise and associates find that there is a definite separation in the mtDNA phylogeny of the Atlantic and Pacific populations of the green turtle (Chelonia mydas) but that the extent of the sequence divergence is 10 times lower than expected if the mammal-derived evolutionary rate of 2% divergence per million years is applied (Allan Wilson argues that this rate should be applicable to all animals). Avise concludes that mtDNA evolution is slower in turtles than in mammals. In contrast, a study by E. Bermingham and H. A. Lessios finds that the rate of divergence per million years for sea urchins is close to that calculated for mammals. Further complicating the molecular clock debate, Nancy Knowlton examined mtDNA and allozymes for snapping shrimp and concluded that the age of the biogeographic vicariance event probably does not accurately document the point at which gene flow stopped between the Atlantic and Pacific sides of the isthmus.
A revision of the original selective neutrality hypothesis is the “nearly neutral theory,” associated with T. Ohta, in which there are small positive or negative selection coefficients associated with mutations. In this paradigm, nearly neutral mutations are affected more by drift when populations are small, but more by natural selection when populations are large. It may be argued that these effects of population size are true regardless of whether the mutation is nearly neutral or has a substantial positive or negative selection coefficient. Ridley extensively discusses this paradigm and perceives several advantages in comparison with the “purely neutral” hypothesis, including the prediction of a more erratic, less constant rate of evolution based on the influence of population size on whether slightly disadvantageous mutations will be affected by drift or selection.
Recent attention has also focused on microsatellite analysis rather than on the more traditional allozymes, mtDNA and RFLPs. Microsatellites are short, repeated sequences that are usually found in noncoding regions of nuclear DNA. They manifest a Mendelian pattern of inheritance, unlike mtDNA. Because of the large amount of variability in microsatellite arrays, researchers have used them in identification of parents/offspring and in forensic applications. It has also been suggested that they may show selective neutrality because their function is currently unknown. John Hawks and colleagues incorporate microsatellites in their discussion of population bottlenecks and Pleistocene hominin evolution: Variation in microsatellite base pairs should reflect population size expansions if the loci are selectively neutral. However, as the authors point out, if it can be demonstrated that selection affects microsatellites, then variability in these regions need not indicate population expansions.
Evolution and Physical Anthropology
The Darwinian paradigm of evolution provides a conceptual framework in which the fossil record, primate behavior/ecology, and human and nonhuman primate population biology can be interpreted. While it was possible for Linnaeus to assign formal taxonomic names to primates (for example, Cercopithecus aethiops) and Baron Cuvier to describe Adapis parisiensis from the Montmartre gypsum quarries prior to Darwinism, the evolutionary paradigm provides a unifying, explanatory framework for physical anthropology. In their recent edition of Principles of Human Evolution (2004), Roger Lewin and Robert A. Foley discuss the four types of biological explanation; two varieties that are relevant to our discussion are explanations about adaptive value and evolutionary history. The former inquiries about the function of a trait or behavior and the latter about evolutionary history; both are central to the investigations conducted by physical anthropologists. It is not adequate to simply describe a morphological feature or behavior. Rather, it is crucial to (1) understand how that feature or behavior allows the individual to successfully interact with its environment and (2) to understand the diversity of extinct and extant primates in the context of common descent.
Physical anthropology has borrowed from other disciplines of evolutionary biology and ecology in these endeavors and, increasingly, contributed to the formulation of evolutionary concepts with application beyond the primates. Examples of concepts and techniques adapted from other areas of evolutionary biology and ecology are too numerous to completely enumerate here. The study of variation was a major contribution of the New Synthesis of evolutionary biology and was emphasized by papers in the seminal Classification and Human Evolution (1963), edited by Sherwood L. Washburn; it has formed a major aspect of primate taxonomy in both neontological and paleontological settings. Paradigms of phyletic evolution and punctuated equilibrium have been used in the interpretation of the primate fossil record. Various species concepts, such as the biological species concept, the phylogenetic species concept, and the recognition species concept, have been applied to primates. Cladistic taxonomic philosophy and methodology, initially invented by the entomologist Willi Hennig, has been widely adopted and applied by physical anthropologists. The idea of the ring species (speciation by distance) was discussed by Yoel Rak to analyze hominin evolution in the Levant. It has been suggested that the phenomenon of nanism (dwarf forms), observed in island forms of ducks, hippopotami, elephants, and deer, may have occurred in hominin evolution, as documented by the recent discovery of Homo floresiensis. The ideas of sociobiology, employed outside of physical anthropology to explain social systems as diverse as eusocial insects and naked mole rats, are widely employed in modern primatology. And formulations associated with population genetics (such as the Hardy-Weinberg equilibrium and effective population size) are routinely used in the study of human and nonhuman primate population biology.
Physical anthropologists have been somewhat slower to apply their findings to larger evolutionary issues and concepts. This is unfortunate, because primate morphology, ecology, population biology, and behavior have been studied with an intensity that is unequalled in mammalogy. Long-term field studies, such as those of the Awash and Amboseli baboons, provide a wealth of detailed data that are of great benefit to the specialties of speciation research and behavioral ecology.
References:
- Avise, J. C. (1994). Molecular markers, natural history, and evolution. New York: Chapman & Hall.
- Carroll, R. L. (1997). Patterns and processes of vertebrate evolution. Cambridge: Cambridge University Press.
- Conner, J. K. (2003). Artificial selection: A powerful tool for ecologists. Ecology, 84, 1650-1660.
- Freeman, S., & Herron, J. C. (2001). Evolutionary analysis. (2nd ed.). Upper Saddle River, NJ: Prentice Hall.
- Gillespie, J. H. (1991). The causes of molecular evolution. New York: Oxford University Press.
- Givnish, T. J., & Sytsma K. J. (1997). Molecular evolution and adaptive radiation. Cambridge: Cambridge University Press.
- Kimura, M. (1983). The neutral theory of molecular evolution. Cambridge: Cambridge University Press.
- Majerus, M. E. N. (1998). Melanism: Evolution in action. Oxford: Oxford University Press.
- Mayr, E. (1991). One long argument: Charles Darwin and the genesis of modern evolutionary thought. Cambridge, MA: Harvard University Press.
- Ridley, M. (2004). Evolution (3rd ed.). Maiden, MA: Blackwell.
