The Limits to Reproductive Output: Offspring Size Versus Number in the Sand Lizard (Lacerta agilis)

Abstract

Natural selection does not always (or indeed, usually) produce phenotypes, because of a variety of phenomena including trade-offs (conflicting optimal phenotypes for different functions) and mechanistic constraints (e.g., physical limitations, lack of additive genetic variance, genetic correlations between the sexes). The importance of trade-offs and constraints has been widely appreciated (e.g., Gould and Lewinton 1979), but the ways in which such factors influence life-history traits remain obscure. One vigorous debate in reptilian biology concerns the determinants of reproductive output. In many reptile species, clutch sizes and offspring sizes are related to each other and to maternal body size. The causal basis for these relationships remains controversial, with two main approaches evident in the literature. One view stresses energy allocation, such that reproductive output (e.g., relative clutch mass [RCM] or reproductive effort [RE]) is determined by the optimal allocation of resources among maintenance, storage, growth, and reproduction (Tinkle and Hadley 1973, 1975; Congdon et al. 1982). Under this approach, the upper limit to reproductive investment is set by energy availability and, hence, is likely to vary in response to proximate factors (e.g., Ballinger 1983; James and Whitford 1994). An alternative approach suggests that physical constraints, such as the space available to hold the clutch in the female's abdominal cavity, determine the upper limit for reproductive investment (Vitt and Congdon 1978; Shine 1988, 1992). The central disagreement between the two approaches centers on which factor determines the upper limit to reproductive output: resources or physical constraints. The best evidence that such an upper limit does exist is the widespread occurrence of trade-offs between offspring size and clutch size: that is, larger clutches consist of smaller offspring (Elgar and Heaphy 1989; Madsen and Shine 1992, 1996; King 1993). The existence of such a trade-off implies that there is a fixed upper limit to investment. Interspecific comparisons suggest that RCMs are correlated with body shape, in keeping with the notion of physical constraints (if all females fill themselves with eggs, body shape should constrain reproductive output) (Shine 1992). However, the volume constraint hypothesis is difficult to reconcile with the high intrapopulational variance often observed in RCMs or the