By contrast with a multitude of laboratory studies on the social organization of fish, relatively little is known about the size, composition and dynamics of free-ranging fish shoals. We give an overview of the available information on fish shoals and assess to what degree the predictions made from laboratory studies are consistent with field data. The section on shoal choice behaviour in the laboratory is structured so that the evidence for different shoaling preferences is discussed in the context of their mechanisms and functions. Predictions based on experiments in captivity regarding preferences for conspecifics, individuals of similar body length and unparasitized fish were highly consistent with field observations on free-ranging shoals whereas preferences for familiar conspecifics and kin remain to be conclusively demonstrated in the field. In general, there is a shortage of studies in which shoaling preferences have been investigated both in the laboratory and the field, and field studies have so far been largely descriptive revealing little about the underlying mechanisms of observed patterns. Given the great importance of fish shoals both in fundamental and applied research, an advancement of our knowledge of their social organization should significantly contribute to a better understanding of a whole range of topics including reciprocal altruism, group-living and self-organization.