EVOLUTIONARY LOSS OF LARVAL FEEDING: DEVELOPMENT, FORM AND FUNCTION IN A FACULTATIVELY FEEDING LARVA, BRISASTER...

Species with large eggs and nonfeeding larvae have evolved many times from ancestors with smaller eggs and feeding larvae in numerous groups of aquatic invertebrates and amphibians. This change in reproductive allocation and larval form is often accompanied by dramatic changes in development. Little is known of this transformation because the intermediate form (a facultatively feeding larva) is rare. Knowledge of facultatively feeding larvae may help explain the conditions under which nonfeeding larvae evolve. Two hypotheses concerning the evolutionary loss of larval feeding are as follows: (1) large eggs evolve before modifications in larval development, and (2) the intermediate form (facultatively feeding larva) is evolutionarily short-lived. I show that larvae of a heart urchin, Brisaster latifrons, are capable of feeding but do not require food to complete larval development. Food for larvae appears to have little effect on larval growth and development. The development, form, and suspension feeding mechanism of these larvae are similar to those of obligate-feeding larvae of other echinoids. Feeding rates of Brisaster larvae are similar to cooccurring, obligate-feeding echinoid larvae but are low relative to the large size of Brisaster larvae. The comparison shows that in Brisaster large egg size, independence from larval food, and relatively low feeding rate have evolved before the heterochronies and modified developmental mechanisms common in nonfeeding echinoid larvae. If it is general, the result suggests that hypotheses concerning the origin of nonfeeding larval development should be based on ecological factors that affect natural selection for large eggs, rather than on the evolution of heterochronies and developmental novelties in particular clades. I also discuss alternative hypotheses concerning the evolutionary persistence of facultative larval feeding as a reproductive strategy. These hypotheses could be tested against a phylogenetic hypothesis.