The debates concerning individual development go back 2,500 years to the time of Aristotle in the fourth century before the present era. During his investigations of the embryo and fetus in a wide variety of species, Aristotle opened up fertilized eggs at different stages of incubation and noted that new structures appeared during the course of incubation. He was the first to perceive the antithesis between epigenesis (novel structures emerge during the course of development) and preformation (development is the simple unfolding or growth of preexisting structures). All subsequent debates about the nature of the developmental process are founded to some extent on this dichotomy. I say ‘to some extent’ because when one surveys the history of embryological thought, as, for example, embodied in Joseph Needham’s (1959) marvelous work, A History of Embryology, there is a second debate of utmost importance that is really at the heart of all debates about the nature of the developmental process: what causes development? What causes development to happen?
By the late 1700s and early 1800s, the debate over preformation and epigenesis was resolved in favor of epigenesis. Before proceeding to a review of the debates about the causes of epigenetic development, it is informative to go a bit deeper into the notions of preformation and epigenesis.
Preformation: ovists and animalculists
There were two main versions of preformation. Since, according to this view, the organism was preformed in miniature from the outset, it was believed by some to lie dormant in the ovary of the female until development was started by fertilization. This view was held by the ovists. To other thinkers, the preformed organism resided in the semen of the male and development was unleashed through sexual union with the female. These were the animalculists.
Many of the preformationists, whether ovists or animalculists, tended to be of a religious persuasion. In that case they saw the whole of humankind having been originally stored in the ovaries of Eve if they were ovists or in the semen of Adam if they were animalculists. Based upon what was known about the population of the world in the 1700s, at the time of the height of the argument between the ovists and animalculists, Albrecht von Haller (1708-1777), the learned physiologist at the University of Gottingen, calculated that God, in the sixth day of his work, created and encased in the ovary of Eve 200,000 million fully formed human miniatures. Von Haller was a very committed ovist.
The sad fact about this controversy was that the very best evidence to date for epigenesis was at hand when von Haller made his pronouncement for preformation: “There is no coming into being! [Nulla est epigenesis.] No part of the animal body was made previous to another, and all were created simultaneously… All the parts were already present in a complete state, but hidden for a while from the human eye.” Given von Haller’s enormous scientific stature in the 1700s, we can only assume that he had an overriding mental set about the question of ontogenesis (development of the individual), and that set caused him to misinterpret evidence in a selective way. For example, the strongest evidence for the theory of encasement, as the theory of preformation was sometimes called, derived from Charles Bonnet’s observations, in 1745, of virgin plant lice, who, without the benefit of a male consort, reproduce parthenogenetically (i.e., by means of self-fertilization). Thus, one can imagine the ovist Bonnet’s excitement upon observing a virgin female plant louse give birth to ninety-five females in a 21-day period and, even more strikingly, observing these offspring themselves reproduce without male contact. Here was Eve incarnate among the plant lice!
Epigenesis: emergent nature of individual development
The empirical solution of the preformation-epigenesis controversy necessitated direct observation of the course of individual development, and not the outcome of parthenogenetic reproduction, as striking as that fact itself might be. Thus it was that one Caspar Friedrich Wolff (1733-1794), having examined the developmental anatomy and physiology of chick embryos at various times after incubation, provided the necessary direct evidence for the epigenetic or emergent aspect of individual development. According to Wolff’s observations, the different organic systems of the embryo are formed and completed successively: first, the nervous system; then the skin covering of the embryo; third, the vascular system; and finally, the intestinal canal. These observations not only eventually toppled the doctrine of preformation but also provided the basis for the foundation of the science of embryology, which took off in a very important way in the next 150 years.
Fortunately, the microscopes of the late 1800s were a significant improvement over those of the late 1600s, whose low power allowed considerable reign for the imagination. Figure 1 shows the drawing of a human sperm cell by Nicholas Hartsoeker in 1694. Needless to say, Hartsoeker was a convinced animalculist prior to looking into the microscope.
Nature versus nurture: the separation of heredity and environment as independent causal agents
The triumph of epigenesis over preformation eventually ushered in the era of experimental embryology, defined as the causal-analytic study of early structural development, which unhappily coincided with the explicit separation of the effects of heredity and environment in Francis Galton’s formulation of the nature-nurture dichotomy in the late 1800s.
Francis Galton’s influential legacy
Francis Galton (1822-1911) was a second cousin of Charles Darwin and a great admirer of Darwin’s concept of natural selection as a major force in evolution. Galton studied humans and advocated selective breeding or non-breeding among certain groups as a way of, respectively, hastening intellectual and moral evolution and saving humankind from degeneracy. Galton coined the term eugenics, and its practice in human populations eventually resulted from his theories, among others. He advocated positive eugenics, which encouraged people of presumed higher moral and intellectual standing to have larger families. (Negative eugenics, which he did not explicitly advocate, resulted in sterilization laws in some countries, including the United States, so that people judged unfit would have fewer children.)
Figure 1. Drawing of the contents of a human sperm cell by the preformationist Nicholas Hartsoeker in 1694.
Galton failed completely to realize that valued human traits are a result of various complicated kinds of interactions between the developing human organism and its social, nutritional, educational, and other rearing circumstances. If, as Galton found, men of distinction typically came from the upper or upper-middle social classes of 19th-century England, this condition was not only a result of selective breeding among ‘higher’ types of intelligent and moral people, but was also due in part to the rearing circumstances into which their progeny were born. This point of view is not always appreciated even today; that is, the inevitable correlation of social class with educational, nutritional, and other advantages (or disadvantages) in producing the mature organism. Negative eugenics was practiced in some European countries (e.g., Sweden, Switzerland) and in some states in the USA for much of the twentieth century.
Galton’s dubious intellectual legacy was the sharp distinction between nature and nurture as separate, independent causes of development, although he said in very contemporary terms, “The interaction of nature and circumstance is very close, and it is impossible to separate them with precision” (Galton, 1907, p. 131). While it sounds as if Galton opts for the interpenetration of nature and nurture in the life of every person, in fact he means that the discrimination of the separate causal effects of nature and nurture is difficult only at the borders or frontiers of their interaction. Thus, he wrote:
Nurture acts before birth, during every stage of embryonic and pre-embryonic existence, causing the potential faculties at the time of birth to be in some degree the effect of nurture. We need not, however, be hypercritical about distinction; we know that the bulk of the respective provinces of nature and nurture are totally different, although the frontier between them may be uncertain, and we are perfectly justified in attempting to appraise their relative importance.
Since we still retain, albeit unknowingly, many of Galton’s beliefs about nature and nurture, it is useful to examine his assumptions more closely. He believed that nature, at birth, offered a potential for development, but that this potential (or reaction range, as it is sometimes called) was rather circumscribed and very persistent. In 1875, he wrote: “When nature and nurture compete for supremacy on equal terms … the former proves the stronger. It is needless to insist that neither is self-sufficient; the highest natural endowments may be starved by defective nurture, while no carefulness of nurture can overcome the evil tendencies of an intrinsically bad physique, weak brain, or brutal disposition.” One of the implications of this view was, as Galton wrote in 1892: “The Negro now born in the United States has much the same natural faculties as his distant cousin who is born in Africa; the effect of his transplantation being ineffective in changing his nature.” The conceptual error here is not merely that Galton is using his upper-middle class English or European values to view the potential accomplishments of another race, but it is rather that he has no factual knowledge of the width of the reaction range of African blacks – he assumes it not only to be inferior, but to be narrow and thus without the potential to change its phenotypic expression.
This kind of assumption is open to factual inquiry and measurement. It requires just the kind of natural experiment that Galton would have marveled at, and perhaps even enjoyed, given its simple elegance, namely, the careful monitoring and measurement of presumptively in-built traits within generations in races that have migrated to such different habitats, sub-cultures, or cultures that their epigenetic potential would be allowed to express itself in previously untapped ways. Thus, we can draw a line of increasing adult stature as Oriental groups migrate to the United States and substantially change their diet. More importantly we can measure the increase in IQ of blacks (within as well as between generations) as they move from the rural southern United States to the urban northeast, and its further increase the longer they remain in the urban northeast (Otto Klineberg’s topic, Negro Intelligence and Selective Migration, published in 1935). The same is true for lower-class whites coming from the rural south to the urban northeast. Galton’s concept of’like begets like,’ whether applied to upper-class Englishmen or poor blacks and whites, requires that their rearing circumstances and opportunities remain the same.
Galton’s dubious intellectual legacy is notoriously long-lived, no matter how many times the nature-nurture controversy has been claimed to be dead and buried. An analysis of psychology text topics reveals the heartiness of Galton’s dichotomous ideas up to the late 20th century (Johnston, 1987).
Dichotomous thinking about individual development in early experimental embryology
In the late 1800s and early 1900s, the main procedure of experimental embryology, as a means of implementing a causal analysis of individual development, was to perturb normal development by deleting cells or moving cells to different places in the embryo. Almost without exception, when normal cellular arrangements were changed developmental outcomes were altered, giving very strong empirical support to the notion that cell-cell or cell-environment interactions are at the heart of individual development: interactions of one sort or another make development happen (i.e., make development take one path rather than another path).
This major conceptual advance was only incompletely realized because of the erroneous interpretation of one of the earliest experiments in the new experimental embryology. In 1888, Wilhelm Roux (1850-1924), one of the founders of experimental embryology, used a hot needle to kill one of the two existing cells after the first cleavage stage in a frog’s egg and observed the development of the surviving cell. The prevalent theory of heredity at the time held that one-half of the heredity determinants would be in each cell after the first cleavage, and, indeed, as called for by the theory, a roughly half embryo resulted from Roux’s experiment.
However, when Hans Driesch (1867-1941), another of the founders of experimental embryology, performed a variation of Roux’s experiment by separating the two cells after cleavage by shaking them completely loose from one another, he observed an entire embryo develop from the single cells. Eventually, Roux accepted that the second, dying cell in his experiment interfered with the development of the healthy cell, thus giving rise to the half-embryo under his conditions.
Before he accepted that, however, Roux had begun theorizing on the basis of his half-embryo results and came up with a causal dichotomy that continues to haunt embryology to the present day: self-differentiation versus dependent differentiation. These two terms were coined by Roux as a consequence of his half-embryo experiment, which he believed erroneously to be an outcome of self-differentiation, implying an independent or non-interactive outcome, in contrast to dependent differentiation where the interactive component between cells or groups of cells was necessary to, and brought about, the specific outcome. The concept of self-differentiation is akin to the concept of the innate when the term is applied to an outcome of development, as in the innate (hereditary) – acquired (learned) dichotomy that is prevalent in much of psychological theorizing.
Roux, himself, gave up the self- and dependent-differentiation dichotomy as he came to accept Driesch’s procedure as being a more appropriate way to study the two post-cleavage cells. Unfortunately, Roux’s concepts lived on in experimental embryology in disguised form as mosaic development versus regulative development. In the latter, the embryo or its cells are seen as developing in relation to the milieu (environment), whereas the former is understood as a rigid and narrow outcome fostered by self-differentiation or self-determination, as if development were non-interactive. Here is the way the American embryologist W. K. Brooks (1902, pp. 490-491) expressed concern about the notion of self-differentiation:
A thoughtful and distinguished naturalist tells us that while the differentiation of the cells which arise from the egg is sometimes inherent in the egg, and sometimes induced by the conditions of development, it is more commonly mixed; but may it not be the mind of the embryologist, and not the material world, that is mixed? Science does not deal in compromises, but in discoveries. When we say the development of the egg is inherent, must we not also say what are the relations with reference to which it is inherent?
This insight that developmental causality is relational (interactive or coactive) has eluded us to the present time, as evidenced in the various causal dichotomies extant in the developmental-psychological literature of today: nature-nurture, innate-acquired, maturation- experience, development-evolution, and so forth. We need to move beyond these dichotomies to understand individual development correctly.
Predetermined and probabilistic epigenesis
At the root of the problem of understanding individual development is the failure to truly integrate biology into developmental psychology in a way that does empirical justice to both fields. The evolutionary psychologists, for example, are still operating in terms of Galton’s legacy, as witnessed by the following quotations. They start off seemingly on the right foot, as we saw in Galton’s introductory remarks about nature and nurture: “The cognitive architecture, like all aspects of the phenotype from molars to memory circuits, is the joint product of genes and environment… EPs [evolutionary psychologists] do not assume that genes play a more important role in development than the environment does, or that ‘innate factors’ are more important than ‘learning.’ Instead, EPs reject these dichotomies as ill-conceived” (Cosmides & Tooby, 1997, p. 17). However, several pages later, when they get down to specifics, the nature-nurture dichotomy nonetheless emerges: “To learn, there must be some mechanism that causes this to occur. Since learning cannot occur in the absence of a mechanism that causes it, the mechanism that causes it must itself be unlearned – must be innate” (Cosmides & Tooby, 1997, p. 19). Since one must certainly credit these authors (as well as others who write in the same vein) with the knowledge that development is not preformative but epigenetic, in 1970, extending Needham’s (1959, p. 213, note 1) earlier usage, I employed the term ‘predetermined epigenesis’ to capture the developmental conception of the innate that is embodied in the above quotation. (Cosmides and Tooby do not stand alone; other evolutionary theorists such as the ethologist Konrad Lorenz (1903-1986) posited an ‘innate schoolmarm’ to explain the development of species- specific learning abilities.) The predetermined epigenesis of development takes this form:
Predetermined Epigenesis Unidirectional Structure – Function Development
experience (e.g. species-specific learning abilities)
In contrast to predetermined epigenesis, I put forward the concept of probabilistic epigenesis:
Probabilistic Epigenesis Bidirectional Structure – Function Development
In this view, prior experience, function, or activity would be necessary for the development of species-specific learning abilities. Epigenesis is probabilistic because there is some inevitable slippage in the very large number of reciprocal coactions that participate in the developmental process, thereby rendering outcomes probable rather than certain.
By way of defining the terms and their relationships, as it applies to the nervous system, structural maturation refers to neurophysiological and neuroanatomical development, principally the structure and function of nerve cells and their synaptic interconnections. The unidirectional structure-function view assumes that genetic activity gives rise to structural maturation that then leads to function in a non-reciprocal fashion, whereas the bidirectional view holds that there are reciprocal influences among genetic activity, structural maturation, and function. In the unidirectional view, the activity of genes and the maturational process are pictured as relatively encapsulated or insulated, so that they are uninfluenced by feedback from the maturation process or function, whereas the bidirectional view assumes that genetic activity and maturation are affected by function, activity, or experience. The bidirectional or probabilistic view applied to the usual unidirectional formula calls for arrows going back to genetic activity to indicate feedback serving as signals for the turning on and turning off of genetic activity.
The usual view in the central dogma of molecular biology calls for genetic activity to be regulated by the genetic system itself in a strictly feed-forward manner, as in the unidirectional formula of DNA ^ RNA ^ Protein above. Thus, the central dogma is a version of predetermined epigenesis. Note that genetic activity is involved in both predetermined and probabilistic epigenesis. Thus, what distinguishes the two conceptions is not genes versus environment, as in the age-old nature-nurture dichotomy, but rather the unidirectional (strictly feed-forward or -upward influences) versus the bidirectional nature of the coactions across all levels of analysis. There is now evidence for all of the coactions depicted in the probabilistic conception, including those at the genetic level of analysis (Gottlieb, 1998). Given that genes, however remotely, are necessarily involved in all outcomes of development, it is dismaying to see that that fact is not universally recognized, but rather is seen as some outdated relict of hereditarianism: “… although genetic effects of various kinds have been conclusively demonstrated, hereditarian research has not produced conclusive demonstrations of genetic inheritance of complex behaviors … The behaviorists’ approach … should be – and generally is – to accept a genetic basis only if research designed to identify effects of social or other environmental variables does not reveal any effects” (Reese, 2001, p. 18). This is a particularly blatant example of either/or dichotomous causality: develop mental outcomes are caused either by genes or by environment. Given the recent date of the quotation, this is evidence that the nature-nurture dichotomy is not dead and, if it is buried, it has been buried alive.
Figure 2. Probabilistic-epigenetic framework: depiction of the completely bidirectional and coactional nature of genetic, neural, behavioral, and environmental influences over the course of individual development.
From central dogma of molecular biology to probabilistic epigenesis
In addition to describing the various ramifications ofthe nature-nurture dichotomy, the other purpose of this entry is to place genes and genetic activity firmly within a developmental-physiological framework, one in which genes not only affect each other and mRNA (messenger RNA that mediates between DNA and protein), but are affected by activities at other levels of the system, up to and including the external environment. This developmental system of bidirectional, coactional influences is captured schematically in Figure 2. In contrast to the unidirectional and encapsulated genetic predeterminism of the central dogma, a probabilistic view of epigenesis holds that the sequence and outcomes of development are probabilistically determined by the critical operation of various endogenous and exogenous stimulative events (Gottlieb, 1997).
The probabilistic-epigenetic framework presented in Figure 2 not only is based on what we now know about mechanisms of individual development at all levels of analysis, but also derives from our understanding of evolution and natural selection. As everyone knows, natural selection serves as a filter and preserves reproductively successful phenotypes. These successful phenotypes are a product of individual development, and thus are a consequence of the adaptability of the organism to its developmental conditions. Therefore, natural selection has preserved (favored) organisms that are adaptably responsive to their developmental conditions, both behaviorally and physiologically. As noted above, genes assist in the making of protein; they do not predetermine or make finished traits. Thus, organisms with the same genes can develop very different phenotypes under different ontogenetic conditions, as witness the two extreme variants of a single parasitic wasp species shown in Figure 3 and identical twins reared apart in the human species (Fig. 4).
Figure 3. Two very different morphological outcomes of development in the minute parasitic wasp. The outcomes depend on the host (butterfly or alder fly) in which the eggs were laid. The insects are of the same species of parasitic wasp (Trichogramma semblidis).
Since the probabilistic-epigenetic view presented in Figure 2 does not portray enough detail at the level of genetic activity, it is useful to flesh that out in comparison to the previously mentioned central dogma of molecular biology.
As shown in Figure 5, the original central dogma
Protein, and was silent about any other flows of ‘information’ (as Francis Crick wrote in 1958). Later,
information transfer), Crick (1970) did not claim to have predicted that phenomenon, but, rather, that the original formulation did not expressly forbid it. At the bottom of Figure 5, probabilistic epigenesis, being inherently bidirectional in the horizontal and vertical levels (Fig. 2), has information flowing not only from
translation (protein altering the structure of RNA), but there are other influences of protein on RNA activity (not its structure) that would support such a directional flow. For example, a process known as phosphorylation can modify proteins such that they activate (or
when activated, trigger rapid association of mRNA
transcribed by DNA, they do not necessarily become immediately active but require a further signal to do so. The consequences of phosphorylation could provide
Protein). A process like this appears to be involved in the expression of ‘fragile X mental retardation protein’ under normal conditions and proves disastrous to neural and psychological development when it does not occur. The label of’fragile-X mental retardation protein’ makes it sound as if there is a gene (or genes) that produces a protein that predisposes to mental retardation whereas, in actual fact, it is this protein that is missing (absent) in the brain of fragile X mental retardates, and thus represents a failure of gene (or mRNA) expression rather than a positive genetic contribution to mental retardation. The same is likely true for other ‘genetic’ disorders, whether mental or physical: these most often represent biochemical deficiencies of one sort or another due to the lack of expression of the requisite genes and mRNAs to produce the appropriate proteins necessary for normal development. Thus, the search for ‘candidate genes’ in psychiatric or other disorders is most often a search for genes that are not being expressed, not for genes that are being expressed and causing the disorder.
Figure 4. Remarkable illustration of the enormous phenotypic variation that can result when monozygotic (single egg) identical twins are reared apart in very different family environments from birth.
So-called cystic fibrosis genes and manic-depression genes, among others, are in this category. The instances that I know of in which the presence of genes causes a problem are Edward’s syndrome and trisomy 21 (Down’s syndrome), wherein the presence of an extra, otherwise normal, chromosome 18 and 21, respectively, causes problems because the genetic system is adapted for two, not three, chromosomes at each location. In some cases, it is of course possible that the expression of mutated genes can be involved in a disorder, but, in my opinion, it is most often the lack of expression of normal genes that is the culprit. Most mutations impair fitness. In one of the very rare cases of benefit, in sickle-cell anemia (a defect in red blood cells), the bearer is made resistant to the malaria parasite. Amplifying the left side of the bottom of Figure 5, it is known that gene expression is affected by events in the cytoplasm of the cell, which is the immediate environment of the nucleus and mitochondria of the cell wherein DNA resides, and by hormones that enter the cell and its nucleus. This feed-downward effect can be visualized thusly:
According to this view, different proteins are formed depending on the particular factors influencing gene expression. Concerning the effect of psychological functioning on gene expression, we have the evidence of decreased interleukin 2 receptor mRNA, an immune system response, in medical students taking academic examinations (Glaser et al., 1990). More recently, in an elegant study that traverses all levels from psychological functioning to neural activity to neural structure to gene expression, Cirelli, Pompeiano, & Tononi (1996) showed that genetic activity in certain areas of the brain is higher during waking than in sleeping in rats. In this case, the stimulation of gene expression was influenced by the hormone norepinephrine flowing from locus coeruleus neurons that fire at very low levels during sleep, and at high levels during waking and when triggered by salient environmental events. Norepinephrine modifies neural activity and excitability, as well as the expression of certain genes. So, in this case, we have evidence for the interconnectedness of events relating the external environment and psychological functioning to genetic expression by a specifiable hormone emanating from the activity of a specific neural structure whose functioning waxes and wanes in relation to the psychological state of the organism.
Figure 5. Different views of influences on genetic activity in the central dogma and probabilistic epigenesis. The filled arrows indicate documented sources of influence, while the open arrow from Protein back to RNA remains a theoretical possibility in probabilistic epigenesis and is prohibited in the central dogma (as are Protein ^ Protein influences). Protein ^ Protein influences occur (1) when prions transfer their abnormal conformation to other proteins and (2) when, during normal development, proteins activate or inactivate other proteins as in the phosphorylation example described in the text. The filled arrows from Protein to RNA represent the activation of mRNA by protein as a consequence of, for example, phosphorylation, and the reshuffling of the RNA transcript by a specialized group of proteins called spliceosomes (‘alternative splicing’). DNA ^ DNA influences are termed ‘epistatic,’ referring to the modification of gene expression depending on the genetic background in which they are located. In the central dogma, genetic activity is dictated solely by genes (DNA ^ DNA), whereas in probabilistic epigenesis internal and external environmental events activate genetic expression through proteins (Protein ^ DNA), hormones, and other influences. To keep the diagram manageable, the fact that behavior and the external environment exert their effects on DNA through internal mediators (proteins, hormones, etc.) is not shown; nor is it shown that the protein products of some genes regulate the expression of other genes. (Further discussion in text.)
Table 1. Developmental-behavioral evolutionary pathway.
I: Change in behavior
II: Change in morphology
III: Change in gene frequencies
First stage in evolutionary pathway: change in ontogenetic development results in novel behavioral shift, which encourages new environmental relationships.
Second stage in evolutionary change: new environmental relationships bring out latent (already existing epigenetic) possibilities for morphological-physiological change.
Third stage of evolutionary change resulting from long-term geographic or behavioral isolation (separate breeding populations). It is important to observe that evolution has already occurred phenotypically before stage III is reached.
Role of ontogenetic development in evolution
Though not a debate about the nature of ontogenetic development or the epigenetic process as such, the role of development in evolution takes two very different forms. In its most conventional form, a change in genes (via mutation, sexual recombination, or genetic drift) brings about an enduring change in development that results in the appearance of different somatic, behavioral, and psychological features. That is the standard sequence of events in bringing about evolution in what is called the ‘Modern synthesis’ in biology. A change in genes results in a change in development in this scenario. Since evolution need not occur in only one mode, in another, more recent, scenario, the first stage in the evolutionary pathway is a change in ontogenetic development that results in a novel behavioral outcome. This novel behavior encourages new organism-environment relationships. In the second stage, the new environmental relationships bring out latent possibilities for somatic-physiological change without a change in existing genes. The new environmental relationships activate previously quiescent genes that are correlated with a novel epigenetic process, which results in new anatomical and/or physiological arrangements. This evolutionary scenario is based on two facts: firstly, the empirical fact that specific kinds of changes in species-typical development result in the appearance of behavioral novelties (e.g., increased exploratory behavior, changes in learning ability or preferences, enhanced coping with stress), and, secondly, there is a relatively great store of typically unexpressed genetic (and, therefore, epigenetic) potential that can be accessed by changing developmental conditions.
As long as the changed developmental circumstances prevail, in generation after generation, the novel behavior will persist without any necessary change in genes. Now, eventually, long-term geographic or behavioral isolation (separate breeding populations) may result in a change in gene frequencies in the new population, but the changes in behavior and morphology will already have occurred before the change in genes. No one is denying that genetic mutations, recombination, or drift can bring about evolution; the point is that those are not the only routes to evolutionary change. The three-stage developmental-behavioral evolutionary scenario is shown in Table 1.
That a developmental change in behavior can result in incipient speciation and in genetic change has recently been demonstrated in the apple maggot fly, Rhagoletis pomonella. The original native (USA) host for the female apple maggot fly’s egg laying was the hawthorn, a spring-flowering tree or shrub. Domestic apples were introduced into the USA in the 17th century. Haws and apples occur in the same locale. The first reported infestation of apple trees by apple maggot flies was in the 1860s. There are now two variants of R. pomonella, one of which mates and lays its eggs on apples and the other of which mates and lays its eggs on haws (Table 2). The life cycles of the two variants are now desynchronized because apples mature earlier than haws. Incipient speciation has been maintained by a transgenerational behavior induced by early exposure learning: an olfactory acceptance of apples for courting, mating, and ovipositing based on the host in which the fly developed (Bush & Smith, 1998).
The cause of the original shift from hawthorns to apples as the host species for egg laying can only be speculated upon. Perhaps the hawthorn hosts became overburdened with infestations or, for other reasons, died out in a part of their range, bringing about a shift to apples in a small segment of the ancestral hawthorn population that did not have such well-developed olfactory sensitivity or an olfactory aversion to apples. This latter supposition is supported by behavioral tests, in which the apple variant accepts both apples and haws as hosts, whereas in the haw variant only a small percentage will accept apples and most show a strong preference for haws. As indicated by single host acceptance tests, the apple-reared flies show a greater percentage of egg-laying behavior on the apple host than do the hawthorn-reared flies. Thus, the familiarity-inducing rearing experience (exposure learning) makes the apple-reared flies more accepting of the apple host, although they still have a preference for the hawthorn host.
Given the ecological circumstances, the increased likelihood of acceptance of the apple host, even in the face of a preference for hawthorn, would perpetuate the transgenerational courting, mating, and laying of eggs in apple orchards. Apple maggot flies hatch out at the base of the tree in which their mother had laid their egg the previous summer. While becoming sexually mature, even though they have wandered tens or hundreds of yards, they are still in the vicinity of the apple orchard, if not still in the orchard. The scent of the apples attracts them, and the early rearing experience having rendered the apple scent acceptable, the cycle renews itself, because of the high probability that the early maturing apple maggot fly will encounter the odor of apples rather than hawthorns (see Table 2). In support of incipient speciation, the two variants are now genetically somewhat distinct and do not interbreed freely in nature, although they are morphologically the same and remain interfertile.
In contrast to the transgenerational behavioral scenario being put forward here, conventional evolutionary biological thinking would hold that “most likely some mutations in genes coding for larval/pupal development and adult emergence” brought about the original divergence and maintain the difference in the two populations ( Prokopy, personal communication, August 2000). Although we cannot know with certainty, present evidence (below) would suggest a genetic mutation was not necessary. This is not a behavior versus genes argument; the transgenerational behavioral initiation requires genetic compatibility, otherwise it would not work. The question is whether the original interaction (switch to the apple host) required a genetic mutation or not. The developmental timing change in the life histories of the two forms (Table 2) has resulted in correlated genetic changes in the two populations. That finding is consonant with the evolutionary model presented here (i.e., gene frequencies change some time after the behavioral switch).
From the present point of view, another significant feature of the findings is that, when immature hawthorn flies (pupae) are subjected to the pre-wintering environment of the apple flies (pupae), those that survive have a genetic make-up that is similar to the apple flies, signifying that environmental selection is acting on already-existing developmental-genetic variation. Most importantly, this result shows that there is still sufficient individual developmental-genetic variation in the hawthorn population, even at this late date, to support a transgenerational behavioral initiation of the switch from hawthorns to apples without the necessity of a genetic mutation.
To summarize, a developmental-behavioral change involving the apple maggot fly’s choice of oviposition site puts it in a situation where it must be able to withstand certain pre-wintering low temperatures for given periods of time, and that differ between the apple and hawthorn forms (Table 2). This situation sets up the natural selection scenario that brings about changes in gene frequencies that are correlated with the pre-wintering temperature regimen. The change in egg-laying behavior leads the way to genetic change in the population, the genetic change thus being a consequence of the change in behavior.
After hundreds of years of debate, epigenesis triumphed over preformation. Thus, the nature of the process of individual development was finally understood to be of an emergent character, wherein new structures and functions appear during the maturation of the organism. The next debates concerned the sources of these new structures and functions, and these were partitioned into nature (heredity or genes) and nurture (environment or learning). Recently, as probabilistic epigenesis has more or less triumphed over predetermined epigenesis, the cause of development is now understood to be relational (coactive), in which genetics, neurology, behavior, and environmental influences are all seen as essential and as acting in concert to bring about developmental outcomes, whether physical or psychological. Finally, ontogenetic development, particularly changes in behavioral development, can have a role in initiating evolution prior to genetic changes in the population.