The effects of blockade of opioid and dopamine receptors on relapse to heroin-seeking induced by footshock stress and re-exposure to heroin were examined in a reinstatement procedure. Male rats were trained to self-administer heroin (100 micrograms/kg per infusion, IV; four 3-h sessions/day for 8-11 consecutive days). Extinction sessions were given for 5-7 days during which saline was substituted for heroin. In nine groups, the effects on relapse induced by footshock (10 min, 0.5 mA, 0.5 s on with a mean off period of 40 s), heroin priming (0.25 mg/kg), and saline priming were studied after pretreatment with either naltrexone (1 or 10 mg/kg, SC), the D1-like receptor antagonist SCH 23390 (0.05 or 0.1 mg/kg, IP), the D2-like receptor antagonist raclopride (0.25 or 0.5 mg/kg, IP), the mixed dopamine antagonist flupenthixol decanoate (3 or 6 mg/kg, IM), or IP injection of saline (control condition). Naltrexone, flupenthixol, raclopride, and the highest dose of SCH 23,390 attenuated heroin-induced relapse: only the mixed DA receptor antagonist, flupenthixol, attenuated foot-shock-induced relapse. These results, and those from microdialysis showing that heroin elicits greater locomotor activity and DA release in the nucleus accumbens than footshock, suggest that the neurochemical events underlying stress- and heroin-induced relapse are not identical.