Differences in dispersal of an intertidal gastropod in two habitats: the need for and design of repeated experimental transplantation

Applying general ecological models to new habitats is made difficult where ecological processes vary from one habitat to another. To improve the generality of such models and the precision of predictions made from them, it is necessary to understand the processes that prevent generalizations. The behaviour of intertidal gastropods (Bembicium auratum) varies from rocky shores to mangrove forests near Sydney (Australia) and creates a barrier to generalization of models from one habitat to the other. Differences in behaviour could result from intrinsic differences in the populations of Bembicium occupying the different habitats or from responses to extrinsic cues that vary from habitat to habitat or from a combination of these factors. To test hypotheses from these three models, Bembicium were reciprocally transplanted between the habitats and their subsequent behaviour was compared to that of controls. Outcomes of the experiment varied considerably in space and time, but there was no evidence of intrinsic differences between populations of snails in the two habitats. Some results suggested that a combination of intrinsic and extrinsic factors causes the observed differences in behaviour. Most results were consistent with the model that differences in behaviour of snails on rocky shores and from that shown in mangrove forests are caused by differences between the two habitats and not the snails. The behaviour of individual Bembicium was extremely plastic and changed rapidly on arrival in a new habitat. Experiments of this type are an effective tool for investigating variation in behavioural processes. The study demonstrated that behavioural plasticity can act as a barrier to generalizations from some ecological models. (C) 1999 Elsevier Science B.V. All rights reserved.