Book review in four dimensions Genetic, epigenetic, behavioral, and symbolic variation in the history of Eva Jablonka and Marion J. Lamb The MIT Press. Cambridge, Massachusetts, USA. 2005. 472 pp. $34.95. ISBN: 0-262-10107-6 (hardcover). Reviewed by Mike Benton University of Bristol, Bristol, United Kingdom. E-mail: [email protected].

In Evolution in four dimensions: genetic, epi- with genes for loss of cross veins; the heat impinge on evolution through assimilation. genetic, behavioral, and symbolic variation in shocking was irrelevant. Genetic assimila- Their take on the cream-drinking tits is that the , Eva Jablonka and Marion J. tion was under a new name. learned behavior can become innate if it is Lamb make a concerted effort to expand on Studies of behavior also brought chal- selectively important. If survival through evolutionary theory. The book is one of the lenges. Another study done in Britain in the winter depends on cream drinking, tits best written that I have read in years: clear, the 1940s suggested Lamarckism. Great tits that learn fast would survive best. It would succinct, thoroughly signposted, and with and blue tits “learned” to pick open the foil then be the propensity to learn fast that is ample summaries and reviews of difficult tops of milk bottles and drink the cream selected for. Over time, the authors suggest, areas. The illustrations by Anna Zeligowski on top. The habit spread throughout the what is learned is learned faster and faster are excellent. The care in writing and the country and passed from generation to gen- through the generations until it becomes attractive format should not, however, eration, so it was, in a loose sense, inherited. encoded in the genotype. But how? No one mask the fact that the authors are attempt- Some analysts have since characterized such has yet shown that assimilation happens — ing something revolutionary — a complete behavioral transmission as a kind of evolu- how can it be distinguished from selection rethinking of evolutionary theory. tion, where the behavioral trait is termed a for learning fast or having no cross veins? Lamarckism is alive and well — or is it? meme. In memetics, transmission of such This is where the authors’ thesis seems to Jean-Baptiste Lamarck (1744–1829) is behavior is Lamarckian: it was acquired by fall down. Their case studies can be explained remembered for only a small part of his one or more birds during their lifetimes, and by simple . Behavior is part writings on evolution, what we now call the the meme was then passed down. of the , as are developmental inheritance of acquired characters. Lamarck As Jablonka and Lamb point out, the con- pathways and some fundamental aspects of believed that giraffes strove to reach higher cepts of memes and memetics are nonsense, the human ability to learn language. So the leaves, resulting in elongated necks over time. bearing no relation to genes and . “four dimensions” become one again. Modern Darwinian evolutionary theory does Genes are real entities, copied and transmit- Other authors have tried to expand on not allow for such transmission of acquired ted through generations. The meme for bot- evolution before, yet their efforts have rarely characters. Increasing numbers of observa- tle opening by tits is a chance observation; met with enthusiasm for long. The late Steve tions and intuitions, however, seem to point its bounds are determined by the observer. Gould sought to expand on evolution by to something Lamarckian in evolution. There is no concrete entity, nothing to be adding new hierarchical levels, especially in Jablonka and Lamb begin by discussing copied and transmitted. Milk bottles with the realm of . I believed him genetic assimilation, proposed in 1942 by foil tops are rare in Britain now, but some for a while because I, like Gould, am a pale- British geneticist and embryologist C.H. birds still filch the cream. Has the meme ontologist, and it was encouraging to believe Waddington. Waddington treated the pupae survived, or have birds reinvented the behav- that the long time scales offered uniquely of fruit with heat shock, causing many ior from time to time? These are interesting by the fossil record allowed us to detect new of the flies to have abnormal wings lacking questions, but wrapping the behavior up as kinds of evolutionary phenomena. But the cross veins. Waddington bred these flies, a meme is neither here nor there. critics were right; in theory, species selection repeating the treatments for generations, Jablonka and Lamb, however, hang onto could happen, but an example has yet to be and eventually found that flies without Waddington’s concept of genetic assimi- convincingly demonstrated. Likewise, I have cross veins emerged from pupae that had lation and extend it to develop ideas of to conclude that though the experiments are not been heat treated. He then argued that “behavioral assimilation” and “symbolic fascinating and the case studies intriguing, an acquired character had been assimilated assimilation.” The four dimensions in the none of them indicates that natural selection into the genetic code of the flies, thereby title of the book refer to evolution as quad- has to admit epigenetic, behavioral, and sym- breaking the neo-Darwinian rule that there ripartite: genetic, epigenetic (developmen- bolic inheritance as separate, let alone quasi- can be no information transmission from tal), behavioral, and symbolic (linguistic). equal, processes. I loved the book, admire phenotype to genotype. Strict Darwinians Jablonka and Lamb argue that develop- the authors intensely, and wish I could agree argued that Waddington had selected flies mental, behavioral, and linguistic attributes with their thesis.

The Journal of Clinical Investigation http://www.jci.org Volume 115 Number 11 November 2005 2961