The Genetical Theory of Natural Selection — A Complete Variorum Edition.

R. A. Fisher (edited by Henry Bennett). Oxford University Press, Oxford. 1999. Pp. 318. Price £25.00, hardback. ISBN 0 19 850440 3.

This is perhaps the most important book on evolutionary genetics ever written. Nearly every topic currently discussed is to be found in it, often dealt with in brief passages of condensed prose (it is gratifying to read that Fisher himself described Chapter 2 as ‘heavy’). The treatment is anchored in ecology, so that absolute rather than relative fitness is discussed, and the central theory is developed in spectacular and challenging directions. A new edition needs no justification.

At different times I have struggled with passages in the 1930 Oxford edition and the Dover edition of 1958, without taking note of the differences between them. In his foreword J. H. Bennett outlines the background to production of the first edition and the steps leading to the later one, which incorporated a variety of alterations. The variorum component identifies these for us. At first sight, at least, they reveal little of the author, given the developments in genetics and human affairs between the two editions. Many are grammatical improvements or replacement of archaic phraseology. Some of the larger ones are designed to make a passage clearer; for example, the discussion of genetic variance is extended as ‘Many readers may prefer a more explicit analytic treatment of the problem,…’. A new section on self-sterility alleles is included, ending with a swipe at Sewall Wright. Other changes which reflect contemporary issues are one minimizing the importance of drift in small populations, and another challenging the argument that mutational load caused by atomic radiation presents a danger to humans. Experimental results are included to support the chapter on the evolution of dominance.

Almost two-fifths of the book is concerned with humans and society. This section tends to be overlooked by geneticists interested in Malthusian parameter, fundamental theorem or runaway process, but has received adverse attention in other quarters. Fisher the eugenicist can be read as a proponent of obnoxious doctrines which have cast a shadow over the twentieth century. When he had finished his first draft, he worried that the publishers did not know the extent of the section on humankind. They were happy to print it, however, and Fisher was to be disappointed by the lack of interest shown by reviewers. The 1958 edition contains this section almost unchanged (formidable enough replaces hardy and war-like when discussing the Roman British, adequate replaces generous with respect to family allowances). The only significant insertion is a business-like statement of what is to be covered: the magnitude and heritability of variation in human fertility, evidence of association of fertility with class, and the theory of selective processes which could be responsible. ‘On this theory it may seem that its destructive consequences are not incapable of rational control.’ That statement, at least, is contentious. Fisher did not choose to review further research, by then more extensive and sometimes contradictory, and was evidently content with his treatment. Perhaps the choice of changes tells us more about the author than is at first apparent.

Illustrations of mimetic Lepidoptera from the original plate are reproduced on the dust jacket of this well-presented book, looking much fresher than they did in either of the preceding editions.