In the preface to the first volume of this series we set out our aims, which were to encourage fresh perspectives in ethology and provide a forum for new ideas. We still feel that in the perfectly proper search for high stan­ dards of evidence, methodology has tended to remain the master rather than the servant of most aspects of ethological work. It is easy for us all to forget that the kinds of data we collect are largely determined by the kinds of questions we ask. Even an ethologist with the professed goal of providing a straightforward account of behavior must incorporate into his or her descriptions a great many assumptions about the organization of that be­ havior. Inevitably some facets of what went on will have been selected at the expense of others. This is sometimes done, for example, in the service of a theory that the fundamental unit for description is the fixed action pattern. Our point is not that constraints on the collection of data are bad but that the theory which gives rise to the selection of evidence should not be neglected. In the first volume, the choice of topics and authors was based upon our views about the exciting or developing issues in ethology. This volume represents a more opportunistic approach: the articles were selected from among the many offered to us as best conforming to our aims. Neverthe­ less, certain themes do emerge.