The place-preference paradigm was evaluated as a measure of morphine's positive reinforcing properties. Previous place-preference studies obtained a morphine place preference of 26%-63%. In order to examine whether differences in procedure may account for this scatter, the present experiment investigated whether there is any difference in the absolute magnitude of preference when animals are conditioned on their non-preferred side of the box or when animals are randomly assigned to the side of conditioning. Furthermore, the number of conditioning days was extended with 3 intervening test days, and drug doses were doubled following each test day. The results showed no significant difference between conditioning animals on their non-preferred side or randomly assigning them to the side of conditioning. However, by extending the number of conditioning days, as well as by following the drug regimen used, the animals showed a greater magnitude of preference than that observed in previous studies. The implications of these findings for the usage of this paradigm as a measure of morphine's positive reinforcing properties are discussed.