The best-known theory for how the world became populated with a profuse assortment of disparate organisms is an extension of the concept of natural selection put forward by Charles Darwin and Alfred Russel Wallace in the mid-nineteenth century. The idea is straightforward: complexity is built up gradually by the differential reproductive success of members of a population that differ from one another in small ways, generated in a fashion disconnected from any prospective advantage. As Darwin himself put it in The Origin of Species "Natural selection can act only by the preservation and accumulation of infinitesimally small inherited modifications, each profitable to the preserved being" (Origin, 1st edition, 1859).
It's important to understand Darwin's objective in proposing this mechanism. There was no scientifically based scenario in his time for remolding material (e.g., an animal's body) other than continuous deformation. Perhaps if he were aware of such mechanisms (we now know many from newer understanding of the physics of "soft matter") he would have incorporated them into his theory. But Darwin also recognized another problem: the morphological novelties ("sports") that were sometimes observed among domesticated animals and plants were sporadic and followed no apparent rules in their generation: "He who believes that some ancient form was transformed suddenly through an internal force or tendency into, for instance, one furnished with wings, will be almost compelled to assume, in opposition to all analogy, that many individuals varied simultaneously" (Origin, sixth edition, 1872).
But when the phenomenon of discontinuous variation began to be appreciated during the 1890s, and particularly after Mendel's genetics, with its law-like regularities and the existence of cellular determinants of discrete phenotypic differences (which after all, were a key element of Mendel's observations), came to the attention of the biological community at the turn of the twentieth century, the systematic study of abrupt organismal change became a real possibility. A reaction then set in against Darwin's theory among "mutationists," who suggested that evolution could occur in sudden leaps, with no adaptive intermediates.
Although the mutationists endorsed selection as the means by which novelties could become propagated to subsequent generations, they sharply distinguished their view from the gradualism of Darwin. While this distinction may appear puzzling to a modern culture habituated by an unremitting 90-year campaign to identify "evolutionary theory" with "Darwinism," the two theories were debated vigorously at the time. Mutationism lost the battle, though not because it lacked scientific merit.
The mutationist program included efforts to understand how genes participated in developmental processes, with the objective of explaining biological form and function. William Bateson was interested in the basis of duplicated and repetitive structures like the symmetrical halves of animal bodies, teeth, digits, axial segments, the petals of flowers. Thomas Hunt Morgan was a mutationist and an embryologist before he converted to become a Darwinian and a transmission geneticist. Part of the reason mutationism did not prevail was the inability in the early twentieth to identify the actual molecules involved in developmental patterning and morphogenesis and the cellular responses to them. In contrast, genes - as chromosomal loci - were readily studied with the technology of the time, and could be tracked over generations of breeding experiments, where they could be associated, though not causally, with elements of the phenotype. In his Nobel Prize lecture Morgan stated:
There is no consensus of opinion amongst geneticists as to what the genes are - whether they are real or purely fictitious - because at the level at which the genetic experiments lie, it does not make the slightest difference whether the gene is a hypothetical unit, or whether the gene is a material particle.
The modern evolutionary synthesis (MS) was formulated by embracing this indifference to how genes and their products function in the generation of phenotypes. By making the (highly questionable) Darwinian assumption that every existing complex trait had arisen as the result of many gene changes of small effect (i.e., different from what Mendel studied), a mathematical theory - population genetics - was devised by R. A. Fisher and other MS founders in which evolutionary dynamics was identified exclusively with the further accretion of gene variants of small effect, and the dynamics of actually phenotypic change were ignored. Darwinism was thus salvaged from the messiness of "phenotype biology" by decreeing the irrelevance of developmental mechanisms to evolutionary ones and making gene frequency in populations the proper subject of evolutionary theory.
Contemporary MS adherents continue to uphold the view that as long as selection results in changed gene frequencies, the relation between the phenotype and genotype is irrelevant. Brian Charlesworth, a prominent theorist of this camp, has stated "our understanding of the molecular basis of development-however fascinating and important in revealing the hidden history of what has happened in evolution-sheds little light on what variation is potentially available for the use of selection" (Charlesworth, 2005).
It's hard to credit this view when it is recognized that many complex phenotypes are in fact not the result of the accumulation of gene changes of small effect, and that the developmental mechanisms that produce such forms are often highly sensitive to the genetic context and the external environment. This opens up the possibility that new phenotypes may arise in subpopulations of organisms with no precipitating genetic changes. These forms, moreover, may serve as "adaptations" insofar as the altered organisms find them useful, despite their having been no adaptive intermediates on the way to producing them. "Genetic assimilation," evolution of genetic pathways that support the development of and stabilize the novel phenotype, may follow.
For example, at the end of the nineteenth century the mutationist William Bateson hypothesized that the formation of somites, the array of tissue blocks that give rise to the segmented backbone of vertebrates, is driven by an oscillatory effect in the embryo's axial tissue. Bateson's "vibratory theory" was a favorite subject of ridicule by MS adherents during the next century. However it turned out to be correct, as first shown by Olivier Pourquié and coworkers (Palmerim et al., 1997). Biochemical oscillations are sensitive to temperature and other environmental parameters. Perhaps that's why the number of neck vertebrae in one strain of mice can be changed by transferring them into the uterine environment of a strain with a different number (McLaren and Michie, 1958), or why the number of vertebrae in snakes can depend on egg incubation temperature (Osgood, 1978). A slightly different ratio of the oscillator components could change a segmented organism into a non-segmented one, or vice-versa.
Another example, straight from the pages of this week's Science magazine, shows that the number of digits in a tetrapod's limb (and in fact the presence of digits at all) is not something that was arrived at gradually over the course of vertebrate evolution, but which changes abruptly and dramatically with small variations in limb bud size and shape, and subtle differences in the balance of regulatory factors (Sheth et al., 2012). This is because the "core mechanism" of limb skeleton formation is a Turing-type reaction-diffusion process, which like the oscillation underlying somite formation can produce dramatically different morphological outcomes despite the associated genes only changing slightly (or not at all). The fact that the same, or marginally different, genotypes can be consistent with drastically different phenotypes undermines the strict genetic determinism that is a central tenet of the MS.
In his book River Out of Eden, the MS advocate Richard Dawkins states,
Evolution...must be gradual when it is being used to explain the coming into existence of complicated, apparently designed objects, like eyes [or, presumably, segments, or hands]. For if it is not gradual in these cases, it ceases to have any explanatory power at all. Without gradualness in these cases, we are back to miracle, which is simply a synonym for the total absence of explanation.
To anyone unfamiliar with the physics-based processes at work in the embryo, the rapid emergence of complex forms must indeed appear miraculous. But twentieth and twenty-first century science has taught us that complex forms also appear non-miraculously in the non-living world. Of course the matter that generates them does not contain genes nor do the forms compete for survival.
References
Charlesworth, B., 2005. On the origins of novelty and variation (Review of "The plausibility of life: resolving Darwin's dilemma" by Marc W. Kirschner and John C. Gerhart). Science 310, 1619-1620.
Dawkins, R., 1995. River out of eden: a Darwinian view of life. Basic Books, New York, NY.
McLaren, A., Michie, D., 1958. An effect of the uterine environment upon skeletal morphology in the mouse. Nature 181, 1147-1148.
Osgood, D. W., 1978. Effects of temperature on the development of meristic characters in Natrix fasciata. Copeia 1, 33-47.
Palmeirim, I., Henrique, D., Ish-Horowicz, D., Pourquié, O., 1997. Avian hairy gene expression identifies a molecular clock linked to vertebrate segmentation and somitogenesis. Cell 91, 639-48.
Sheth, R., Marcon, L., Bastida, M. F., Junco, M., Quintana, L., Dahn, R., Kmita, M., Sharpe, J., Ros, M. A., 2012. Hox genes regulate digit patterning by controlling the wavelength of a Turing-type mechanism. Science 338, 1476-80.