INTRODUCTION

Current socio-ecological theory assumes that social systems are the result of adaptations, and thus subject to evolutionary modification by selection (Crook & Gartlan, 1966; Emlen & Oring, 1977; Wrangham, 1987; Ims, 1988). Social systems are not direct targets of selection, however, because they represent the outcome of individual behavioral interactions and strategies (Hinde, 1976). The underlying behavior of individuals towards conspecifics is thought to be largely shaped by ecological factors, such as the distribution of risks and resources in the environment (Wrangham, 1980; van Schaik & van Hooff, 1983; Terborgh & Janson, 1986; van Schaik, 1989), as well as the resulting social boundary conditions (Janson, 1986; van Schaik, 1996). Behavior constitutes, therefore, the crucial interface between individuals and their environment (Terborgh, 1992). Individuals decide at the behavioral level whether they lead a solitary life or form permanent groups, and which group size and composition is optimal under a given set of ecological conditions. An animal's behavioral decisions at this level are therefore a target of selection where convergences among distantly related taxa as a result of similar selection pressures can be expected.

Among mammals, primates exhibit an extraordinary level of diversity in social systems (Smuts et al., 1987), rivaled perhaps only by that found among marsupials and carnivores (Gittleman, 1989; Strahan, 1995). They display all fundamental types of mammalian grouping patterns (Clutton-Brock, 1989) and show a variety of bonding patterns, especially in gregarious taxa (van Schaik, 1989). Thus, the evolution of convergences is not principally constrained by primate-specific traits.

For the purpose of an analysis of convergence in social systems, the living primates can be divided into four major groups with potentially important variation in taxonomic diversity.