Oecologia (1993) 96:331 338 Oecologia Springer-Verlag 1993 Fitness consequences of egg-size variation in the lesser T.D. Williams 1, D.B. Lank 2, F. Cooke 2, R.F. RockwelP 1 Department of Animal & Plant Sciences, PO Box 601, University of Sheffield, Sheffield,S10 2UQ, UK 2 Department of Biology, Queen's University, Kingston, Ontario, Canada, K7L 3N6 3 Department of Ornithology, American Museum of Natural History, New York, 10024, USA Received: 28 June 1993 / Accepted: 16 August 1993 snow goose Abstract. We investigated the relationship between egg- size variation and (a) egg hatching success, (b) chick survival to fiedging and recruitment, and (c) adult female survival, over 12 years in the lesser snow goose (Anser caerulescens caerulescens). By comparing the means and variances of egg size for successful and unsuccessful eggs, our aim was to assess the relative fitness of eggs of different sizes and to determine the type of selection operating on egg size in this species. As both egg size and reproductive success vary with age in the lesser snow goose we controlled for the effects of female age. Egg-size variation is very marked in this population, varying by up to 52% for eggs hatching successfully. However, there was no relationship between egg size and post-hatching survival of goslings to fledging or recruitment, either within or between broods, pooling across years. Egg size varied significantly between successful and unsuccessful clutches in only 2 of 33 individual year comparisons. First-laid eggs surviving to onset of incubation, and eggs hatching successfully, were on average larger than unsuc- cessful eggs, but this was probably due to the confound- ing effects of female age-specific and sequence-specific egg survival. Variance of egg size differed significantly between successful and unsuccessful eggs in only 3 of 24, and 0 of 21, individual year comparisons for pre- and post-hatching survival respectively. We therefore found little evidence for a relationship between egg-size varia- tion and offspring fitness, or for strong directional, nor- malising or diversifying selection operating on egg size, in the lesser snow goose. In addition, there was only weak support for the hypothesis that egg-size variation is maintained by temporal variation in selection pressure (sensu Ankney and Bisset 1973). It is likely that egg-size variation represents the pleiotropic expression of alleles affecting more general physiological or metabolic processes. While this does not rule out the existence of alleles with more direct effects on egg size we suggest that their contribution to heritable egg size is small. Correspondence to. T.D. Williams Key words: Egg size - Fitness consequences - Offspring survival - Selection - Lesser snow goose Egg-size is thought to be highly heritable in birds (Ojanen et al. 1979; Boag and van Noordwijk 1987; Lessells et al. 1989; Larsson and Forslund 1992) and it is generally assumed that characters with high heritabilities have no, or only weak, directional or normalising selection acting on them (Fisher 1958; Falconer 1981; Mousseau and Roff 1987), i.e. such characters should not be closely related to fitness. Despite this, many studies have demon- strated a positive relationship between egg size and off- spring survival in birds (Parsons 1970; O'Connor 1979; Thomas 1983; Magrath 1992) and this relationship is often considered to be widespread (Clutton-Brock 1991; Amundsen and Stokland 1990; although numerous stud- ies have failed to find such a relationship, e.g., Leblanc 1987; Arcese and Smith 1988; Galbraith 1988). A major criticism of many of these studies is that they fail to control for variables which correlate with both egg size and reproductive success, such as parental age or "qual- ity" (Mueller 1990; Reid and Boersma 1990; Magrath 1992). Recent studies which have controlled for such correlated variables have often failed to find a residual effect of egg size on offspring fitness (Amundsen and Stokland 1990; Reid and Boersma 1990; Sydeman and Emslie 1992). Another criticism is that few studies have considered chick survival up to recruitment, which is the ultimate measure of fitness (Price and Boag 1987). If large eggs are associated with benefits in terms of increased chick growth or survival, and therefore with increased inclusive fitness of parents, there should be strong directional selection for increased egg size within populations. Nevertheless, genetic variation could persist in the face of strong selection in various ways and recent- ly several different models have been proposed to explain this apparent contradiction (Price et al. 1988; Cooke et al. 1990; Price and Schluter 1991). Firstly, selection may