Testing behavioural "preference" for suitable microhabitat

Patterns of distribution with respect to microhabitat do not necessarily reflect the "preferred" situation for a given species. Carefully controlled experiments are required to test the model that behaviour rather than processes of recruitment or mortality maintains the pattern. The model that behaviour could maintain associations between intertidal gastropods (Bembicium auratum) and microhabitat provided by oysters and rock pools was examined experimentally. All individuals transplanted to plots on bare rock dispersed, whereas up to 50% of control animals remained in situ after being moved within or between plots of oysters or rock pools. Disproportionately large numbers of animals recaptured after the experiment were found among oysters or in rock pools. The scope and limitations of conclusions from such experiments are discussed in addition to some alternative approaches to testing the behavioural "preference" model. (C) 1998 Elsevier Science B.V.