Animal Play: Evolutionary, Comparative and Ecological Perspectives edited by Marc Bekoff and John A. Byers Tim Caro tmcaro@ucdavis.edu Trends in Ecology & Evolution 1999, 14:122-123 Dept of Wildlife, Fish and Conservation Biology and Center for Population Biology, University of California, Davis, CA 95616, USA

Cambridge University Press, 1998. £55.00/$ 80.00 hbk, £19.95/$32.95 pbk (xvi + 274 pages)

ISBN 0 521 58383 7 / ISBN 0 521 58656 9

The central issue in the study of animal play is its function. We still have very little idea of why animals play after 20 years of research and, almost out of frustration, research on play has moved into realms of mechanism, neurobiology and even philosophy. The 12 chapters in this book reflect the diversity of all these interests.

Chapters 1 and 2, by Burghardt, and Heinrich and Smolker, give anecdotal accounts of play in a turtle (Trionyx triunguis) and in ravens (Corvus corax). These chapters are not only persuasive in showing that taxa other than mammals play, but also raise issues as to how play is defined. Currently, most of us define play as behaviour that appears in young homeotherms that has no obvious function, even though we know this definition is paradoxical, weak, and may subsume a number of causal or functional components ¹. Hall's third chapter discusses a somewhat tangential topic of object play in adult animals. The following chapter is a descriptive cross-taxonomic comparison of play in Macropoidean marsupials by Watson; qualitative arguments are used to suggest that locomotor play might serve an antipredator function. Next is a chapter on intentionality in play by Bekoff and Allen drawing heavily on Bekoff's coyote (Canis latrans) data set collected 20 years ago.

Now come three chapters on play fighting. First, based on its structure, Pellis and Pellis discuss the idea that rodent play fighting is not practice for adult fighting. For example, in serious fighting the attacker shows simultaneous attack and defense, whereas in play these are sequential; in serious fighting the defender vigorously defends itself, whereas in play it shows ineffective defense allowing for some attacks. Play fighting, they surmise, could be a means of reducing fear or re-appraising social relations. Next, in the only quantitative chapter in the book, Miller and Byers discuss sparring as play in pronghorn antelope (Antilocapra americana) males looking at partner preferences, and the relationship between dominance and play. In complete contrast to the proceeding chapter, they conclude that sparring is a way in which males improve fighting skills. Finally, Biben speculates that play fighting in squirrel monkeys (Saimiri sciureus) is a means of socialization in which individuals experience what it is like to be dominant and subordinate in social interactions. Play fighting, she argues, is unlikely to practise motor skills because biting and wounding do not occur.

In Chapter 9, Thomson suggests that play is a means of self assessment. For instance, as success on a given test in play improves above a threshold, individuals should switch to a more challenging test. There is virtually no data regarding this issue. Chapter 10 by Byers is a rehash of his paper with Walker ² showing that the time course of play in house mice (Mus domesticus), Norway rats (Rattus norvegicus) and domestic cats (Felis catus) follows the time course of final synapse formation in Purkinje dendrites in the cerebellum, and the developmental window in which muscle fibre type differentiation occurs. Play seems to occur during a sensitive period when the cerebellum can be modified by experience and postnatal muscle fibres can be altered. Chapter 11 by Siviy is a straightforward review of neuronal processes underlying play in rats. The book ends with a personal account of Brown's experiences in clinical medicine as they relate to play in humans.

This book depressed me because some of the most prominent names in play research are here but their labs are producing new material slowly and no new generation of active researchers is coming forward to replace them. Many of the chapters are little more than synopses of old work. Moreover, the standard of evidence in the majority of chapters is weak: armchair speculation exceeds quantitative data by a huge margin (the first P value was on p. 127). The field seems to be stalling because we have no clear positive evidence for the function of play and this impedes movement in other areas. In addition, we have no honed methodology for depriving animals of play without depriving them of other variables. Furthermore, the number of species for which there are quantitative data on play in natural settings is extremely limited. As a result, we can use experimental data only sparingly and comparative evidence only qualitatively to investigate the function of play. Again and again, we fall back on arguments based on the design of play. Even Byers and Walker's exciting attempt to match play to neuronal development comes down to this. If this book is 'destined to become the benchmark volume in this subject for many years to come', as the blurb on the cover tells us, then the subject will continue to move at a snail's pace. In short, we need a new infusion of ideas, methods and students to provide relief to the flat field of play.



References

[1] Martin P. and Caro T.M. (1985)
Adv. Stud. Behav., 15:59–103.

[2] Byers J.A. and Walker C. (1995)
Am. Nat., 146:25–40. [Cited by]