On the assumption that by the use of the hot-plate procedure the antagonist properties of narcotic analgesics could be detected, the effect of morphine, pentazocine, nalorphine and naloxone were investigated. The latency of paw-licking and jumping-off were determined and compared. The agonist, morphine, at doses of 0.025, 0.05 and 0.1 mmole/kg injected IP significantly increased paw-lick and jump-off latency above that seen in saline controls. The mixed agonist-antagonist, pentazocine, at doses of 0.048, 0.096 and 0.192 mmole/kg and nalorphine, an antagonist with some agonist activity, at doses of 0.032, 0.064 and 0.128 mmole/kg significantly increased the latency of paw-licking, but did not significantly change the jump-off latency. At a dose of 0.016 mmole/kg naloxone treated mice jumped from the hot-plate significantly sooner than controls but no effects of naloxone on paw-licking latency were observed. These results suggest that agonist properties are involved in the paw-lick response and that antagonistic properties determine jumping-off behavior.