Adaptation

. Michael R. Rose and George V. Lauder (eds). Academic Press, San Diego. 1996. Pp. 511. Price £24.95, paperback. ISBN 0 12 596421 8.

Adaptation harks back 18 years to Gould and Lewontin's 1979 paper ‘The spandrels of San Marco and the Panglossian paradigm’, a citation classic which tore at the heart of adaptive story-telling in evolution. According to Gould and Lewontin, many adaptive explanations had been devised post hoc and depended on the tautological assumption that adaptation was the only explanation possible. ‘Spandrels’ are attractive architectural features in San Marco Church, Venice, which were by-products of placing a round dome onto a square building, equivalent in evolutionary terms to a structural necessity, rather than having been designed especially for decoration. Pangloss was Voltaire's caricature of a philosopher for whom ‘all is for the best in this best possible of all worlds’; the implication being that the adaptationist programme was equally illogical.

By its publication, this lengthy book (511 pages of 9 pt type) shows just how successful the 18-year-old ‘spandrels’ critique has been. The introduction is even called ‘post-spandrel adaptationism’ (Rose & Lauder). The adaptationist and gene selectionist ‘right wing’ (so-to-speak) of evolutionary biology has conceded ground to the ‘left's’ non-adaptive alternatives. Sadly, after a debate such as this, all that remains of a battle between provocative extremes is the bourgeois middle ground. And that is what this book is: very solid, and very balanced, with each point of view treated carefully. There are a few omissions. ‘Gene selectionism’ in sociobiology is not treated, perhaps because the editors think the subject is dead. Maybe this is as well. There is plenty of philosophy in the book already: each author, in a careful, balanced way, defines exactly what he (no ‘she's here, except one co-author) means by ‘adaptation’. I would like to have read a discussion of whether speciation is adaptive, another omission. Perhaps, we know too little about the subject for its inclusion, even though the topic dates from the days of Darwin.

Although the even, careful tone can sometimes make for boring reading, this is a very useful book with some excellent high points. Amundson, writing history of science from the unlikely location of Hilo, Hawaii, produces a beautifully fresh history of adaptative explanation. I liked Kirkpatrick's population genetic caveats about adaptation, and I intend to use them in teaching undergraduates. But Kirkpatrick's final statement ‘Perhaps it is serendipity that allows adaptation to flourish in our world’ is extraordinary, and atypically unbalanced for this book. It's not at all surprising that optimization will be achieved in particular fitness dimensions, while at the same time, organisms will rarely approach a global fitness optimum (Seger & Stubblefield).

Reznick & Travis summarize their own elegant work on quantitative genetics of adaptation, though I was dissatisfied with their attempts to generalize to other areas. Their discussion of mimicry, my speciality, is particularly flawed. Reznick & Travis state that warning colour was discovered before mimicry, as seems, in retrospect, logical. However, the idea of warning colour was actually suggested by Wallace eight years AFTER publication of the theory of mimicry by Bates in 1862. Reznick and Travis also ignore the fact that recent experts like J.R.G. Turner and H.F. Nijhout have returned close to the Punnett and Goldschmidt view that major gene mutation does a lot (though not all) of the work involved in mimetic adaptation; this was not mentioned, and indeed does not sit easily with the multi-locus generalization of adaptive evolution espoused by Reznick & Travis. Nor do the authors discuss recent mapping studies of quantitative trait loci, which show that many quantitative adaptations may actually involve relatively few genes of major effect.

The chapters by Novacek and Vermeij provide well-written, insightful (to this neontologist) summaries of controversies in macro-evolution, particularly on whether clade- and species-level adaptation is possible. Hudson reviews the impressive evidence for adaptation at the molecular level, mostly undreamed of 18 years ago, but many features of which are still unclear. Summaries of phenotypic manipulations (Sinervo & Basolo), comparative methods based on phylogenies (Larson & Losos) and genomic parasites (Hurst) were also very useful. Other useful, careful, balanced, etc, chapters had traits that mildly annoyed me: Lauder (too much on Dennett), Rose (too many ‘Alice in Wonderland’ metaphors), Wade (interdemic selection too narrowly defined, and too many own papers cited — 32!).

In the end, perhaps, the only test of all this soul-searching will be in the application of theories of adaptation to alternative, non-biological systems like computer programs, economics, and robots (Frank). While it is tempting to sneer at catch-all complexity theory, if ideas about adaptation cannot be generalized to such systems, they fail an important test of their right to be called science. Besides, general adaptation theory may be much more useful than its biological subsidiary. Recently, the organizing committee of the major UK population genetics group e-mailed computer ‘virus’-infested documents to its entire membership. These crude computer viruses are quite bad enough; we should be thankful they do not (yet) evolve, but are merely designed by pranksters. When self-adjusting, reproducing computer programs become common, we are going to have real problems! I look forward with interest to the day when biologists who have worked in muddy ponds or obscure branches of population genetic theory are invited by governments and corporations to help design defences against generalized evolutionary threats.

Shapes of Time — The Evolution of Growth and Development. Kenneth J. McNamara. The Johns Hopkins University Press, Baltimore. 1997. Pp. 342. Price £29.00, hardback. ISBN 0 8018 5571 3.

I began this book with keen anticipation, having long been interested in the links between development and evolution, and knowing that McNamara has discussed this extensively from a palaeontological perspective. But I soon began to worry, and finished alarmed that a book so full of dubious assertions, errors and speculations had been published by a reputable press.

Let's cut straight to the main problem. Chapter 12 concludes: ‘Evolution is not only about genetics and natural selection. Just as crucial are the changes in timing and rate of development, with the three, genetics, heterochrony, and natural selection, forming an interdependent evolutionary triumvirate’. Earlier we read: ‘Evolution of new structure is not a function of profound genetic mutations — we are dealing here with changes in the time that particular structures start to grow/under the influence of changes in the timing of expression of certain proteins’.

It is widely accepted that the developmental basis of many evolutionary changes has been heterochrony (Haeckel coined this term to denote alterations in developmental timing that could distort the recapitulatory sequence he regarded as a key feature of embryonic sequences. Since then, the term has come to mean any timing changes in development, starting, stopping or rate, that lead to changes in the end result). How does heterochrony occur? Most researchers would see heterochrony as the result of changes in the timing of genetically regulated processes. Indeed, if these changes are to be evolutionary, they must be heritable, and therefore based on genetic differences. Yet McNamara continually plays down the role of genes. The best documented example of heterochrony in action is the ‘heterochronic genes’ that have produced the variable vulval morphologies in nematodes: we see here how an extra set of cell divisions can form a larger structure, or delay in an inductive event a different structure.

McNamara, despite an otherwise amazingly eclectic range of examples, fails to discuss this one, and generally gives the impression that heterochrony is a process separate from what genes do.

Since the days of Goldschmidt's ‘hopeful monster’, few have thought that large, sudden mutations are important in evolution. When McNamara writes of ‘profound genetic mutations’, he is decrying a view of evolution held by no-one, if by this phrase he means Goldschmidt-style mutations. However, he fails to point out that ‘changes in the timing of expression of certain proteins’ involve genetic changes, their ‘profundity’ being somewhat dependant on their results.

Further, he seems to regard heterochrony as the only way in which new morphologies (and behaviours) can appear. In chapter after chapter, he recklessly interprets every imaginable evolutionary development heterochronically. Haeckel, however, had another term, ‘heterotopy’ – the change in position of a structure over evolutionary time, and modern writers have described a wide range of ways in which developmental changes other than timing can lead to new forms.

The book is not all bad. It is written in a lively, personal style (aimed, possibly, at the general readers who learn their biology from Gould or Dawkins) and is best when McNamara is enthusing over fossils. The chapter heading quotations are nice: I enjoyed being reminded of Aldous Huxley's neoteny novel After Many a Summer although my copy keeps the title at that, rather than completing the Tennyson quotation. The illustrations are workmanlike, but too few; many of McNamara's descriptions cry out for a helpful illustration. There are numerous fascinating examples: did you know of the Drosophila species with 4 cm long sperm? But then there are the errors: W.S. Haldane? Shea's anti-human-neoteny paper ‘The case for human neoteny re-considered’ has the ‘re-considered’ omitted; ‘plethodontid’ is used to mean ‘urodele’; von Baer could hardly have used embryological concepts in the quest for ancestral relationships, since he was an anti-evolutionist; a poor section on vertebrate development (Chapter 8) has the skeleton developing from endoderm and ectoderm; the Irish elk is said to have horns.

It is unfortunate that this book has appeared at the same time (and at a similar price) as Gerhart and Kirschner's beautifully produced Cells, embryos and evolution. Buy it, not this.

Selection Indices and Prediction of Genetic Merit in Animal Breeding. N.D. Cameron. CAB International, Wallingford. 1997. Pp.208. Price £22.50, paperback. ISBN 0 85199 169.

Selection theory, on which all organized animal breeding programs are now securely based, has its roots in the work of J.L. Lush, his colleagues and graduate students at Iowa State University in the late 1940s and early 50s. In essence, the theory is a wholly statistical one. The fundamental effects it deals with are all the results of individual genes coding for specific proteins. Until the DNA revolution of the 1980s, these individual gene effects were, in general, beyond our reach. Selection theory was therefore built on models which describe the aggregate effects of many genes. Parallels exist in other fields, for example in physics, where the aggregate effects of large numbers of individually unpredictable atoms can be described, analysed and forecast with considerable accuracy.

Selection theory has proved remarkably robust. Many experiments and analyses of field data have confirmed beyond reasonable doubt that it works in practice. The building and testing of this secure body of theory has also involved a great deal of intellectual energy from many fine scientists in the elaboration and exploration of algebraic models, and of their numerical consequences in computer simulation studies. Harnessing of the theory into practical use has required a great deal of computational ingenuity to cope with massive data volumes and complex structures of modern cattle, pig, sheep, poultry and fish populations.

All of this collective and successful effort has been accompanied by parallel growth in undergraduate and post-graduate training programmes. Given the expanding scale of both instruction and application, it is surprising that the field has produced so few authoritative textbooks along the way — a rough assessment would indicate perhaps one per decade. Neil Cameron's book is therefore doubly welcome — for its intrinsic merits and because it adds a solid contribution to a thinly populated field.

His title is well chosen and accurately reflects the book's content. It indicates that he is not attempting to chase every aspect of animal breeding theory, but is clearly focused on the central issues that go to make an effective selection programme. The choice of depth over breadth is clearly a deliberate one. On occasion, I think it is overdone. For example, (page 90) he gives references, but says ‘there is no need for a detailed derivation of economic values within this text’. This is such a centrally important issue, both for its impact on theoretical selection structures and its self-evident importance in practice, that it merits a chapter on its own. A suggestion for the next edition? Likewise (p.122) a central issue in best linear unbiased prediction (BLUP) models is dealt with by saying ‘the subject of fixed and random effects is discussed by Searle, 1971’. Even within the terms declared, more is required. That being said, this book is very comprehensive on its chosen ground. The first ten of its fourteen chapters give the most complete treatment of selection index theory to be found. The algebra is thorough, though generally developed in 2-trait examples where each step can be followed. The graphics are good and the text clear and unlaboured.

In the final chapters, some additional topics are touched on. These are clearly intended simply as introductions to the subjects and serve this purpose well. Chapter 11 deals with BLUP, chapter 12 with multivariate selection, chapter 13 with single genes and marker-assisted selection, and chapter 14 with binary traits. Each would justify a book on its own, and indeed the CAB publication by R.A. Mrode (Linear Models for the Prediction of Animal Breeding Values, 1996) is an excellent companion volume. A brief, perhaps too brief, appendix on matrix algebra is provided. Throughout, many worked examples are given, together with thinking problems (with answers).

This book clearly grew out of, and is intended for, a graduate level course in animal breeding theory. It serves that purpose admirably well. In the hands of a lecturer who knew their stuff, I would consider it both necessary and sufficient. For all those professionals involved in research, and in the planning and management of modern breeding programs, it will be a useful source and reference book. Neil Cameron has done a good job.

Evolution (Oxford Reader Series). Mark Ridley (ed.). Oxford University Press, Oxford. 1997. Pp. 430. Price £10.99, paperback. ISBN 0 19 289287 8.

The Oxford Reader Series ‘represents a unique, interdisciplinary resource for students, teachers and the general reader, offering authoritative collections of primary and secondary sources on core issues and concepts’, according to the publisher's blurb. Mark Ridley's Evolution, stands alone as the only science-based volume among the sixteen titles advertised as published or forthcoming.

Ridley has gathered together a wonderful pot-pourri of original papers or sections from books organized, with some logic, into ten sections starting with Darwin and the Modern Synthesis and ending on a more philosophical and ethical note. Each section begins with a brief but excellent introduction to set the scene and to put the various contributions into context, historical or otherwise, and then the great and the good of evolutionary biology have their say in their own words. Here are extracts from one of Darwin's unpublished works on species, Fisher's treatise on the nature of adaptation, Cain's classic on the perfection of animals, Mayr's analysis of species concepts, Conway Morris’ description of the Burgess Shale faunas, Dobzhansky's exhortation that ‘nothing in biology makes sense except in the light of evolution’, Raff's recent account of the co-option of eye structures and genes, and Kreitman's views on the neutral theory, to give but a hint of the treats in store. With due regard to impartiality, Ridley even included one of his own papers. The book contains, in total, sixty four extracts and ends with a selected bibliographic section for readers who wish to probe more widely and a potted (one to three lines) biography of each author. The extracts have been skilfully edited with deleted material indicated by [....] so that the reader gets the main gist of the arguments but without becoming bogged down in unnecessary detail or side issues. This series is after all aimed at, among others, ‘the general reader’.

While perusing these extracts I was reminded somewhat of the crammer series’ so often displayed in book shops around the time when school students take their examinations. All one needs to know about Thomas Hardy's Far from the Madding Crowd, Dylan Thomas' Under Milk Wood or Shakespeare's Richard II to pass an exam, and in under 30 pages! The seed of the comparison was sown by the slight guilt I felt in reading here the essence of seminal papers on evolution I should have read in full much earlier in my academic career. I'm sure I will not be alone in this respect but better late, and in edited form, than never at all. One could, of course, argue about which extracts should have be chosen for inclusion but, on the whole, I think Ridley has picked a pretty balanced bunch.

Does Evolution succeed in its aim of bringing the primary and secondary literature within the grasp of ‘the general reader’? In my opinion, a ‘general reader’ would require a reasonable background in genetics and evolution to make much of this book, despite Ridley's valiant use of explanatory footnotes. To be told that an ‘allelomorph’ is now referred to as an ‘allele’ is unhelpful unless one is familiar with the latter term. It is an impossible task to get it right and Ridley admits he has had to guess the level of knowledge of the reader. For the ‘general reader’, therefore, this book would be relatively hard going in my estimation. However, it serves as a superb source of information for students (in the widest sense) of evolution from, say, second year undergraduates upwards. Those who have experienced a little of evolutionary biology will not be troubled by the terms and will, I hope, be highly stimulated by the contents. The juxtaposition of the language, styles and ages of the various extracts provides a powerful sense of history in the making – the evolution of evolutionary thought.

In our Department at York we still stubbornly believe in the value of weekly tutorials for first and second year undergraduates (and the students on the whole agree with us). As intimated above, for students in their second year Ridley's Evolution will provide a rich and stimulating seam of tutorial material, which is why, as book review editor, I decided to evaluate this one myself.

Genetic Engineering with PCR. Robert M. Horton and Robert C. Tait (eds). Horizon Scientific Press, Wymondham. 1998. Pp. 219. Price £34.99, paperback. ISBN 1 898486 12 3.

Practical Applications of Plant Molecular Biology. R. J. Henry. Chapman and Hall, London. 1997. Pp. 258. Price £24.99, paperback. ISBN 0 412 73220 3.