This study demonstrates that the relative performance of male and female rats on a Morris water maze task changes when the task parameters are varied. In three separate experiments, male and female rats were tested on a different variation of the Morris water maze. In all cases, on each of 10 days of testing, rats were given an initial trial in which the escape platform was randomly placed in a new position. A second trial was given one 1 h later. When the release position did not change between daily trials, no gender differences were observed. When the release position was changed between the initial and subsequent trial, females, but not males, showed reduced retention of the platform location on the second trial. This implies a male superiority for the task. However, a third manipulation of the task parameters demonstrated that females were as accurate and efficient as males at finding and remembering the platform location, even when released from a new position, as long as major landmark cues in the room remained constant. This study supports the hypothesis that male and female rats use different types of spatial cues when solving maze tasks, and stresses the importance of separating the effects of task variables from possible endogenous sender differences in ability.