JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. British Ecological Society is collaborating with JSTOR to digitize, preserve and extend access to Journal of Animal Ecology. Summary 1. Prey-size selectivity of predators can play an important role in determining the predators' impact on their prey. In marine systems, a pattern of preference for small-sized molluscan prey is widespread among crustacean predators, even though predators are often able to consume prey over a wider size range. 2. In laboratory tests the blue crab Callinectes sapidus showed preference for smaller individuals of the hard clam Mercenaria mercenaria. This was true for crabs starved for different durations of time, prior to experiment. Hungrier crabs tended to be less selective than the less hungry ones, although not significantly so. 3. In order to determine whether such selectivity is hard-wired (i.e. engrained) or can be modified through experience, adult blue crabs were tested for preference between two different sizes of hard clams after a phase of conditioning on different combinations of live and sham (i.e. empty valves glued together and deployed in living position) clams of the two sizes. 4. In the conditioning trial, crabs consumed more sham clams of both sizes than live clams, although sham clams did not yield any energy return. The greater consumption of sham clams by blue crabs might be explained by the lower cost of crushing dead clam shells, as dead shells seemed to be more brittle than the live ones. 5. Conditioning significantly affected size selectivity in the subsequent test trial. Crabs assigned to different conditioning treatments consumed significantly different proportions of large clams in the test trial. In particular, crabs that had consumed greater proportions of large (sham) clams during the conditioning trial also ate significantly greater proportions of large (live) clams in the test trial than those which had consumed greater proportions of small (sham) clams during conditioning. 6. These results indicate that the blue crabs' preference for small-sized clams does not result from a fixed decision rule and suggest that crabs might modify their preferences through experience. Furthermore, the relative strength of shells seemed more important than profitability ratios in determining patterns of prey-size consumption. 7. An understanding of what perceptual and cognitive constraints underlie feeding preferences of 'keystone' marine predators could help in making predictions about the impact of such predators on prey communities.