[The aim of this review is to present a coherent explanation of the ultimate causation and the evolution of primary social systems. The explanation is based on the assumption that social systems of all diurnal primates are derived from female defence polygyny. Group living is therefore regarded as the essential first step in the social evolution of diurnal primates. Primate social systems are generally classified into four main types, each of which can be contrasted with the others along the following lines: solitariness versus group living, monogamy versus polygyny and reproductive activity by only one male versus reproduction of several males per group. First of all we present some comparative evidence to support the hypothesis that monogamy in primates has evolved from polygynous groups, in two different ways. A commonly held view is that monogamy is most likely to develop where paternal care of the offspring is advantageous. This is almost certainly the case in the very small species of primates. In these monogamy is the consequence of twinning and the resultant need for active paternal care. It runs contrary to the evident tendency of these species to form larger groups in response to the pressure of predation. The latter is achieved by an extended group membership of grown-up offspring in a non-reproductive and even 'helping' role. In the larger species, however, monogamy is most probably the outcome of, on the one hand, a strong need to reduce competition for food favouring very small group sizes, and, on the other hand, of the virtual absence of predation permitting such a development. Secondly, we look critically at several hypothetical explanations for the existence of multi-male groups and one-male groups. We argue that the ultimate explanation for their existence is not to be found in differential male mortality, anti-predator defence, availability of food, and differential expulsion of expendable males. Instead, comparative evidence indicates that the distinction between a single-male or a multi-male system is ultimately dependent on the varying ability of a male to monopolize access to a breeding group of females. This ability in turn depends on group cohesiveness as determined by feeding strategies and predation pressure and is often, though not necessarily, related to the size of the group., The aim of this review is to present a coherent explanation of the ultimate causation and the evolution of primary social systems. The explanation is based on the assumption that social systems of all diurnal primates are derived from female defence polygyny. Group living is therefore regarded as the essential first step in the social evolution of diurnal primates. Primate social systems are generally classified into four main types, each of which can be contrasted with the others along the following lines: solitariness versus group living, monogamy versus polygyny and reproductive activity by only one male versus reproduction of several males per group. First of all we present some comparative evidence to support the hypothesis that monogamy in primates has evolved from polygynous groups, in two different ways. A commonly held view is that monogamy is most likely to develop where paternal care of the offspring is advantageous. This is almost certainly the case in the very small species of primates. In these monogamy is the consequence of twinning and the resultant need for active paternal care. It runs contrary to the evident tendency of these species to form larger groups in response to the pressure of predation. The latter is achieved by an extended group membership of grown-up offspring in a non-reproductive and even 'helping' role. In the larger species, however, monogamy is most probably the outcome of, on the one hand, a strong need to reduce competition for food favouring very small group sizes, and, on the other hand, of the virtual absence of predation permitting such a development. Secondly, we look critically at several hypothetical explanations for the existence of multi-male groups and one-male groups. We argue that the ultimate explanation for their existence is not to be found in differential male mortality, anti-predator defence, availability of food, and differential expulsion of expendable males. Instead, comparative evidence indicates that the distinction between a single-male or a multi-male system is ultimately dependent on the varying ability of a male to monopolize access to a breeding group of females. This ability in turn depends on group cohesiveness as determined by feeding strategies and predation pressure and is often, though not necessarily, related to the size of the group.]