Genetics, Epigenetics, and Flexibility: A Reply to Crozier Mary Jane West-Eberhard The American Naturalist, Vol. 139, No. 1 (Jan., 1992), 224-226. STOR ? Stable URL: http://links.jstor.org/sici?sici=0003-0147%28199201%29139%3Al%3C224%3AGEAFAR%3E2.0.CO%3B2-I The American Naturalist is currently published by The University of Chicago Press. Your use of the JSTOR archive indicates your acceptance of JSTOR' s Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/ucpress.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is an independent not-for-profit organization dedicated to creating and preserving a digital archive of scholarly journals. For more information regarding JSTOR, please contact support@jstor.org. http://www.j stor.org/ Thu May 6 17:19:16 2004 Vol. 139, No. 1 The American Naturalist January 1992 GENETICS, EPIGENETICS, AND FLEXIBILITY: A REPLY TO CROZIER Crozier (1992) criticizes me for "two major errors" (p. 219). One is to have distinguished between "naturalists" and "geneticists" since, as he points out, the two approaches can be pursued by the same person. This does not seem worth a quibble, except to insist on the point behind the distinction: Hamilton's rule has been used in two different ways, one natural historical (as a decision rule tested by field studies) and the other genetic (as a predictor of allele- frequency change). Crozier confirms this point when he says (p. 220) that "Hamil- ton's rule is a predictor of allele-frequency change and not a 'model of gene expression' except in terms of studying the results of such change" (italics mine). Studying the results of such [evolutionary] change?phenotypes, and the condi- tions of their expression?could serve as a definition of what naturalists (who may also be geneticists) do. I did not refer to either approach as "wise" or "foolish" (Crozier 1992, p. 219). The other criticism has to do with my portrayal of genetic models. It is certainly correct and important to point out that not all genetic models of altruism (traits costly in terms of the performer's fitness) specify or assume obligatory expression of alleles for altruism. Crozier cites many models postulating alleles with condi- tional expression or incomplete penetrance, and, although this is not mentioned by Crozier, I also discussed them (West-Eberhard 1987) in order to show why they are not necessary to explain the evolution of a population capable of condi- tional worker behavior. I regret the misleading statement (1988, p. 128) where I mentioned "a gene for worker altruism (that is, one whose bearers are obligate helpers)" as if that were the only kind treated by genetic models. But I wish to emphasize that this does not affect my argument that all models depicting genes for altruism, including those with conditional expression, may be irrelevant to the evolutionary establishment of conditional worker behavior in a population. If the epigenetic hypothesis for the evolution of the worker phenotype is correct, a gene for altruism need not be involved. Instead, helping behavior and sterility can occur as a side effect of selection for "selfish" traits like aggressiveness and group living whose mean benefits more than compensate their mean costs (includ- ing the cost of a certain frequency of sterility). Kin selection would further reward helping behavior in groups of kin but is not necessary to explain the spread of the underlying genes. I worry that Crozier's lengthy defense of quantitative theories in general will give the mistaken impression that I disparaged them in general. The epigenetic hypothesis (West-Eberhard 1987) challenges only one kind of genetic model (that Am. Nat. 1992. Vol. 139, pp. 224-226. Copyright is not claimed for this article. NOTES AND COMMENTS 225 treating alleles for altruism) in only one context (the origin of social insect work- ers), and the challenge applies whether the model is verbal or quantitative. Crozier's final paragraph, regarding "source laws" and "consequence laws" (from Sober 1984), points the way toward understanding how comparative study and phenotype-transaction hypotheses relate to genetic models of evolution. As Crozier says, genetic models describe results of selection (consequence laws) for a particular kind of trait. Phenotype transition models like the one I proposed attempt to reconstruct the phenotypic changes that might have led (under selec- tion in certain contexts) to the evolution of a novel trait. Such hypotheses, and the "source laws" they may illuminate for natural selection, make it possible to identify the kinds of genetic models (consequence laws) appropriate to describe a particular change. Accordingly, the "epigenetic" transition hypothesis suggests that ordinary models of selection without genes for altruism are sufficient to explain the evolution of worker behavior beginning with an ancestral workerless population. That is, a genetic model is inappropriate when it proves inconsistent with the source law best supported by comparative study (which is necessarily qualitative). (Neither phenotype-transition nor game-theoretic models should be seen as crude or facile versions of quantitative genetic models; each deals more gracefully than the others with a different aspect of evolution.) What about the subsequent elaboration of the worker phenotype? A "flat" view of the phenotype (disregarding mechanisms of development) might, again, suggest that models for conditionally expressed alleles would apply. Analysis of the phenotype including its development, however, indicates that several classes of traits are involved: those regulating expression (e.g., sensitivity to external cues and response thresholds), "specific modifiers" (sensu Turner 1976) of the worker phenotype (traits controlled by the regulatory apparatus and expressed in workers, but not queens), and "nonspecific modifiers" (expressed in both castes). Only one of these (the specific modifiers) might require a model of a conditionally expressed allele for altruism. Altruism alleles affecting regulation need not be conditionally expressed, and a general model for an evolutionary increase in worker altruism would have to encompass both conditionally and unconditionally expressed alleles. By now it should be clear that my use of "epigenetic" does not mean "nonge- netic." I have never argued that "evolution may proceed . . . without genetic innovation" (Crozier 1992, p. 221), nor have I departed from a genetic definition of evolution, made explicit (West-Eberhard 1988, p. 126) to avoid misinterpreta- tions of this kind. All of the evolved traits I discussed?for example, cyclic reproductive behavior, aggressiveness, and group life, and their pleiotropic effects?presumably reflect the genetic makeup of the individuals performing them. Indeed, it seems likely that even caste determination (by which some indi- viduals end up as helpers and others as egg-laying queens) depends to some degree on genotype (e.g., heritable differences in aggressiveness, hormone titers, etc.), especially in relatively simple societies lacking extensive manipulation of the brood (which can overwhelm heritable variation). In this respect I would criticize models, cited approvingly by Crozier, in which "queenness strikes at random" (p. 219) with respect to genetic makeup. By ignoring epigenesis?the 226 THE AMERICAN NATURALIST regulatory architecture of flexible traits?such models disregard the kinds of ge- netic variation that permit adaptive adjustment of individual decisions and caste ratios. For evidence of genotypic influence on determination of conditional alter- natives, see West-Eberhard (1989) and Hazel et al. (1990). I hope that readers shocked by Crozier's citation (p. 219) of my statement that "eusociality can originate 'without genetic innovation' " (freely interpreted by him to possibly mean evolution without genetic change) will examine the original text (West-Eberhard 1988, p. 127). It discusses environmental change as one conceivable way (in a list including evolved, genetic changes) that conditional phenotypes can be triggered for the first time in individuals genetically predis- posed to express them due to selection in other contexts. This is just old- fashioned preadaptation and genotype-environment interaction, seen at the dawn of an evolutionary transition. Until we can discuss the environment as building block and cue (not just an agent of selection) without being suspected of advocat- ing some weird form of "nongenetic evolution," we will not progress far toward understanding the evolution of plastic traits. ACKNOWLEDGMENTS R. Crozier kindly sent me two different versions of his critique, affording ample opportunity to reply. I benefited from comments on this subject by A. Bourke, W. D. Hamilton, N. Knowlton, E. G. Leigh, and especially W. G. Eberhard. LITERATURE CITED Crozier, R. H. 1992. The genetic evolution of flexible strategies. American Naturalist 139:218-223. Hazel, W. N., R. Smock, and M. D. Johnson. 1990. A polygenic model for the evolution and mainte- nance of conditional strategies. Proceedings of the Royal Society of London B, Biological Sciences 242:181-187. Sober, E. 1984. The nature of selection. MIT Press, Cambridge, Mass. Turner, J. R. G. 1976. Butterfly mimicry: genetical evolution of an adaptation. Evolutionary Biology 10:163-206. West-Eberhard, M. J. 1987. Flexible strategy and social evolution. Pages 35-51 in Y. Ito, J. L. Brown, and J. Kikkawa, eds. Animal societies: theories and facts. Japan Scientific Societies Press, Tokyo. . 1988. Phenotypic plasticity and "genetic" theories of insect sociality. Pages 123-133 in G. Greenberg and E. Tobach, eds. Evolution of social behavior and integrative levels. T. C. Schneirla Conference Series. Vol. 3. Erlbaum, Hillsdale, N.J. . 1989. Phenotypic plasticity and the origins of diversity. Annual Review of Ecology and Systematics 20:249-278. MARY JANE WEST-EBERHARD* SMITHSONIAN TROPICAL RESEARCH INSTITUTE Submitted August 28, 1990; Revised May 28, 1991; Accepted June 6, 1991 * Correspondence address: Escuela de Biologia, Universidad de Costa Rica. Associate Editor: Paul W. Sherman