The selective mu-opioid agonist, D-Ala2,N-Me-Phe4,Gly5-ol-enkephalin (DAMGO), or the selective A1-adenosine agonist N6-cyclopentyladenosine (CPA), when coinjected intradermally with prostaglandin E2 (PGE2), dose- dependently inhibited PGE2-induced mechanical hyperalgesia in the rat hindpaw, as determined by the Randall-Selitto paw-withdrawal test. Repeated (hourly x 3) intradermal injections of DAMGO or CPA produced tolerance to the antinociceptive effect of a fourth injection 1 hr later. Furthermore, repeated (hourly x 3) intradermal injections of DAMGO produced cross-tolerance to the antinociceptive effect of CPA, and repeated (hourly x 3) intradermal injection of CPA produced cross- tolerance to the antinociceptive effect of DAMGO. The demonstration of the bidirectional cross-tolerance between the peripheral antinociceptive effects of DAMGO and CPA supports the hypothesis that both these agents produced antinociception by acting on the same cell, presumably the primary afferent nociceptor, and that the development of tolerance involves changes downstream to activation of mu-opioid and A1- adenosine receptors. The opioid antagonist naloxone, which had no effect on paw-withdrawal threshold in normal paws, produced withdrawal threshold in normal paws, produced withdrawal hyperalgesia in DAMGO- tolerant paws. Furthermore, naloxone elicited a cross-withdrawal hyperalgesia response in CPA-tolerant paws. Similarly, the A1-adenosine antagonist 1,3-dipropyl-8-(2-amino-4- chlorophenyl)-xanthine (PACPX), which had no effect on paw-withdrawal threshold in normal paws, elicited a withdrawal hyperalgesia response in CPA-tolerant paws and cross-withdrawal hyperalgesia responses in DAMGO-tolerant paws. These cross-dependence and cross-withdrawal responses suggest that the development of dependence to mu-opioid and A1-adenosine agonists involves changes in the same second messenger system downstream to both mu-opioid and A1-adenosine receptor activation.