The effects of prenatal cocaine and protein malnutrition were examined on acquisition of the radial arm maze in adult male Sprague-Dawley rats whose mothers were provided with a 6% casein, a 25% casein or a standard chow diet and cocaine (30mg/kg) or saline injections beginning 5 weeks prior to mating and continuing to parturition. Rats were tested using an 8-arm radial maze with 4 baited arms and were required to collect all 4 food pellets within 5 min to complete a trial. Subjects were tested for 1 trial/day until they met criterion for successful acquisition of the task. Criterion was attained when the rat collected 3 out of the 4 food pellets within their first 4 arm entries within a trial (while still completing the trial) with this level of performance being maintained for 3 consecutive trials. The results showed dissociation between the effects of prenatal protein malnutrition and prenatal cocaine. Prenatally protein malnourished adult male rats required a greater number of trials to criterion, made more reference memory (but not working memory) errors, and required a longer time to complete each trial when compared with control males. However, rats with prenatal cocaine exposure showed no significant impairments in the radial arm maze. These results contrast with our previous findings using the Morris maze task in which adult male subjects exhibited impaired acquisition following prenatal cocaine while there were no effects following prenatal protein malnutrition. Thus, the radial arm maze and water maze procedures appear to engage different processes that are differentially sensitive to the prenatal insults.