It is widely believed that opioid withdrawal symptoms contribute to relapse to opioid use, but relapse is highly probable in experienced users even after prolonged abstinence and during opioid maintenance therapy. We have found using an animal model of relapse, the reinstatement procedure, that the two events that reliably reinstate heroin-seeking behavior are reexposure to heroin, and brief exposure to footshock stress. Contrary to expectation, opioid antagonist-induced withdrawal does not reinstate heroin-seeking. We now report on reinstatement of heroin-seeking in rats trained to self-administer heroin and subsequently exposed to a maintenance dose of heroin via minipump and allowed to self-administer saline. With the minipump in, naloxone-induced withdrawal did not reinstate drug-seeking, a priming injection on heroin was only mildly effective, and footshock was highly effective. Twenty-four hours after removal of the minipump (spontaneous withdrawal), animals reinitiated heroin-seeking and, subsequently, both heroin and footshock reinstated heroin-seeking. In summary, brief exposure to stress reinstated heroin-seeking in both heroin-maintained and withdrawn animals. The heroin prime reliably reinstated drug- seeking only in the absence of the minipump; opioid “withdrawal,” as such, did not reinstate drug-seeking behavior. Naloxone given to heroin- maintained animals induced withdrawal symptoms, caused a mild depression in the levels of dopamine and its metabolites in the nucleus accumbens septi (NAS), but did not reinstate drug-seeking. Reinstatement of heroin-seeking during spontaneous withdrawal was not accompanied by reductions in basal dopamine and its metabolites in NAS.