Like many other complex biological phenomena, pain is starting to be studied at the level of the gene. Advances in molecular biological technology have allowed the cloning, mapping, and sequencing of genes, and also the ability to disrupt their function entirely (i.e. via transgenic knockouts). With these new tools at hand, pain researchers have begun in earnest the task of defining (a) which of the 70,000-150,000 mammalian genes are involved in the mediation of pain, and (b) which of the pain-relevant genes are polymorphic, contributing to both natural variation in responses and pathology. Although there are only a few known examples in which single gene mutations in humans are associated with pain conditions (e.g. an inherited form of migraine and congenital insensitivity to pain), it is likely that others will be identified. Concurrently, a variety of genes have been implicated in both the transmission and control of "pain" messages in animals. The present review summarizes current progress to these ends, focusing on both transgenic (gene-->behavior) and classical genetic (behavior-->gene) approaches in both humans and laboratory mice.