Primates maintain a variety of social relationships and these can have fitness consequences. Research has established that different types of social relationships are unpinned by different or interacting hormonal systems, for example, the neuropeptide oxytocin influences social bonding, the steroid hormone testosterone influences dominance relationships, and paternal care is characterized by high oxytocin and low testosterone. Although the oxytocinergic system influences social bonding, it can support different types of social bonds in different species, whether pair bonds, parent-offspring bonds or friendships. It seems that selection processes shape social and mating systems and their interactions with neuroendocrine pathways. Within species, there are individual differences in the development of the neuroendocrine system: the social environment individuals are exposed to during ontogeny alters their neuroendocrine and socio-cognitive development, and later, their social interactions as adults. Within individuals, neuroendocrine systems can also have short-term effects, impacting on social interactions, such as those during hunting, intergroup encounters or food sharing, or the likelihood of cooperating, winning or losing. To understand these highly dynamic processes, extending research beyond animals in laboratory settings to wild animals living within their natural social and ecological setting may bring insights that are otherwise unreachable. Field endocrinology with neuropeptides is still emerging. We review the current status of this research, informed by laboratory studies, and identify questions particularly suited to future field studies. We focus on primate social relationships, specifically social bonds (mother-offspring, father-offspring, cooperative breeders, pair bonds and adult platonic friendships), dominance, cooperation and in-group/out-group relationships, and examine evidence with respect to the ‘tend and defend’ hypothesis.