The spatial learning abilities of young, middle-age, and senescent rats were investigated in two experiments using several versions of the Morris water maze task. In Experiment 1, Long-Evans hooded rats were trained to find a submerged escape platform hidden within the water maze. During this phase of testing, aged rats exhibited acquisition deficits compared with either young or middle-age subjects. With continued training, however, all age groups eventually achieved comparable asymptotic levels of performance. Subsequent testing in Experiment 1 revealed that following original training, aged rats were not impaired in learning a novel escape location or in their ability to locate a visible, cued escape platform. In an attempt to identify the basis of the age-related impairments observed in Experiment 1, naive young and aged rats in Experiment 2 were initially tested for their ability to locate a cued escape platform in the water maze. During this phase of testing, the escape latencies of both young and aged rats rapidly decreased to equivalent asymptotic levels. Subsequent analyses revealed that following cue training, young subjects exhibit a significant spatial bias for the region of the testing apparatus where the platform was positioned during training. In contrast, aged rats showed no spatial bias. Training was continued in Experiment 2 using a novel submerged platform location for each subject. During these place training trials, the escape latencies of senescent rats were longer than those of young subjects. These impairments were also accompanied by a lack of spatial bias among aged rats relative to young control subjects.(ABSTRACT TRUNCATED AT 250 WORDS)