Naltrexone methylbromide and naloxone methylbromide, quaternary derivatives of naltrexone and naloxone respectively, are assumed to act peripherally. Both compounds reversed the intestinal stimulating effect of morphine in the dog. Naltrexone methylbromide 5 mg/kg s.c. blocked morphine-induced intestinal spike potentials for 50 min while intravenous doses caused antagonism for only 25 min. The antagonism by the s.c. route approximated that produced by naltrexone 0.2 mg/kg s.c. In morphine-dependent dogs, naltrexone methylbromide did not appear to antagonize morphine centrally in doses ranging from 0.25 to 50 mg/kg s.c. since it did not induce behavioral signs of narcotic withdrawal. Similarly, i.v. naloxone methylbromide was also able to reverse morphine-induced intestinal spike potential in dogs but the protection lasted only 25 min. In rats, naltrexone methylbromide 10 and 30 mg/kg i.p. neither reversed morphine block of PGF2 alpha-induced diarrhea nor antinociception. This suggests a lack of CNS narcotic antagonism in both test. In mice, naltrexone methylbromide, 60-720 mg/kg orally and 3-140 mg/kg i.p. failed to block morphine inhibition of prostaglandin F2 alpha-induced diarrhea. Paradoxically, in this species, 30 mg/kg s.c. of naltrexone methylbromide appeared to cross the blood-brain barrier since this dose reversed morphine-induced antinociception. In conclusion, naltrexone methylbromide effectively antagonizes the acute gut stimulating effect, but not the chronic behavioral effect of morphine administration in dogs. Based upon the antinociception test, naltrexone methylbromide does not cross the blood-brain barrier in rats but may in mice. Morphine inhibits prostaglandin F2 alpha-induced diarrhea by a central mechanism in rodents.