The evolutionary ecology of offspring size in marine invertebrates.

Intraspecific variation in offspring size is of fundamental ecological and evolutionary importance. The level of provisioning an organism receives from its mother can have far reaching consequences for subsequent survival and performance. In marine systems, the traditional focus was on the remarkable variation in offspring size among species but there is increasing focus on variation in offspring size within species. Here we review the incidence and consequences of intraspecific offspring-size variation for marine invertebrates. Offspring size is remarkably variable within and among marine invertebrate populations. We examined patterns of variation in offspring size within populations using a meta-analysis of the available data for 102 species across 7 phyla. The average coefficient of variation in offspring size within populations is 9%, while some groups (e.g., direct developers) showed much more variation (15%), reflecting a fourfold difference between the largest and smallest offspring in any population. Offspring-size variation can have for reaching consequences. Offspring size affects every stage of a marine invertebrate's life history, even in species in which maternal provisioning accounts for only a small proportion of larval nutrition (i.e., planktotrophs). In species with external fertilization, larger eggs are larger targets for sperm and as such, the sperm environment may select for different egg sizes although debate continues over the evolutionary importance of such effects. Offspring size affects the planktonic period in many species with planktotrophic and lecithotrophic development, but we found that this effect is not universal. Indeed, much of the evidence for the effects of offspring size on the planktonic period is limited to the echinoids and in this group and other taxa there is variable evidence, suggesting further work is necessary. Post-metamorphic effects of offspring size were strong in species with non-feeding larvae and direct development: bigger offspring generally have higher post-metamorphic survival, higher growth rates and sometimes greater fecundity. Although there is limited evidence for the mechanisms underlying these effects, the size of post-metamorphic feeding structures and resistance to low-food availability appear to be good candidates. There was limited evidence to assess the effects of offspring size on post-metamorphic performance in planktotrophs but surprisingly, initial indications suggest that such effects do exist and in the same direction as for species with other developmental modes. Overall, we suggest that for direct developers and species with non-feeding larvae, the post-metamorphic effects of offspring size will be greatest source of selection. Offspring-size variation can arise through a variety of sources, both within and among populations. Stress, maternal size and nutrition, and habitat quality all appear to be major factors affecting the size of offspring, but more work on sources of variation is necessary. While theoretical considerations of offspring size can now account for variation in offspring size among mothers, they struggle to account for within-brood variation. We suggest alternative approaches such as game theoretic models that may be useful for reconciling within-clutch variation. While some of the first theoretical considerations of offspring size were based on marine invertebrates, many of the assumptions of these models have not been tested, and we highlight some of the important gaps in understanding offspring-size effects. We also discuss the advantages of using offspring size as a proxy for maternal investment and review the evidence used to justify this step. Overall, offspring size is likely to be an important source of variation in the recruitment of marine invertebrates. The quality of offspring entering a population could be as important as the quantity and further work on the ecological role of offspring size is necessary. From an evolutionary standpoint, theoretical models that consider every life-history stage, together with the collection of more data on the relationship between offspring size and performance at each stage, should bring us closer to understanding the evolution of such a wide array of offspring sizes and developmental modes among species.