In Experiment 1 female rats were given one trial per day for 8 days in a T-maze, and their initial direction of choice (left/right) was noted. Vaginal smears were also obtained daily. After this some animals were adrenalectomized and given Ringer's solution; others were adrenalectomized and given hydrocortisone replacement; a third group was sham adrenalectomized, and a fourth group was an intact control. A week after surgery the animals were again tested for 8 days in the T-maze and vaginal smears were obtained. In Experiment 2 rats were subjected to the same surgical treatments as described above and were then tested for 8 days in the T-maze. In Expt. 1 there was no direction bias among the four groups prior to surgery. However, after surgery the Adrenalectomy+Ringer's group showed a significant increase in their rightward choices in the T-maze. This was also found in Expt. 2. Both adrenalectomized groups in Expt. 1 had a significant reduction in the duration of the estrus phase of their cycle. We conclude that corticosterone can affect lateral preference in a T-maze through a mechanism independent of the hormonal changes involved in the estrus cycle.