without mothers or peers. Davenport and others (1973) and Fritz (1986) reported on problems encountered in the resocialization of chimpanzees that had been maintained without appropriate social conspecific interaction. The manifestations of inadequate early rearing include a broad range of species-inappropriate behaviors, such as the inability to cope with stress as evidenced by self-biting; lack of appropriate breeding and parental skills; rocking, eye-poking, and other stereotypic behaviors; coprophagy; and inability to interact appropriately in social situations. Many of these behaviors are refractory to change and persist for life; at best remediation programs are labor intensive and expensive. Clearly, the goal of housing nonhuman primates is to avoid the development of behavior problems through the careful planning, execution, and assessment of an institutional strategy, or plan, for ensuring the psychological well-being of the animals.
Although social housing is a critical component of psychological well-being, careful consideration is required in developing the procedures to achieve this objective. Because of the xenophobic reactions of many primate species, which can result in severe aggression to strangers, attempts to pair monkeys or create social groups must be handled with care (Clarke and others 1995). A number of different strategies—which vary according to the species, age, sex, and social experience—are possible for forming pairs or groups of primates (Coe 1991; Cooper and others 1997; Crockett and others 1997; Fritz 1986, 1989, 1994; Reinhardt 1988, 1989a, 1991a; Vermeer 1997). It should be remembered, however, that many species of primates express social dominance, and fighting between animals can occur (Bayne and others 1995). Although unfamiliar rhesus monkeys can be introduced to one another and become compatible pairs (Reinhardt 1989a), it is not so easy to introduce a rhesus monkey into a previously established group, because of the likelihood of a severe group attack. Rhesus groups are best formed with total strangers so that individuals are protected by the ''organized chaos" of the group (Bernstein 1964). Other species, with different social dynamics, such as capuchins, present different challenges (Fragaszy and others 1994).
As with all close human-nonhuman primate interactions, personnel safety should receive the utmost consideration when forming pairs or groups of primates. The ability to separate incompatible animals should be well planned before new introductions.
Although a social living situation is important, there can be practical and scientific reasons for using individual housing, such as research protocols, medical conditions, the possibility of disease transmission, hyperaggressiveness, and hypersubmissiveness. When experimental protocols require individual housing, nonhuman primates should, whenever it is possible, have visual, auditory, or olfactory contact with each other. Animals that have been individually housed, even for long periods, have been successfully resocialized when efforts have been made to find compatible companions.