The scanning electron microscope (SEM) was used to compare erect and nonerect penile glandes of gonadally intact Norway rats (group I) and of castrated rats exposed to the following hormonal conditions: maintained with testosterone (T)-filled Silastic capsules implanted subcutaneously (group T); maintained with implants of estradiol (E) for 8 or 12 days (group E1); maintained with E implants for 10 weeks (group E10); maintained with E implants for 9 weeks, then injected daily with testosterone propionate for 8 or 17 days (group E/T). As in previous SEM depictions of nonerect glandes of intact rats, spines projected toward the base of the glans at a shallow angle from the sulci of deep epithelial folds. In contrast, the folds on erect glandes of groups I and T were completely smoothed, and the spines were themselves erected. The penile cup formed at the distal end of the erect glans also contained spines; these were centrifugally directed at the rim and centripetally directed on the inner surface of the cup. The glandes of group E1 males were similar to those in groups I and T, with the spines showing no disorientation. Males in group E10 lacked spines in the cup and along most of the shaft of the glans, but erection revealed many sharp spines just proximal to the cup and on its rim. In group E/T, no papillae were detectable on the nonerect glandes, but erection revealed many small rounded papillae on the shaft and within the cup. The erection of the spines that occurs on the shaft and in the cup of the erect glans may facilitate previously proposed functions of the spines, including vaginal and cervical stimulation and removal of the copulatory plug. Our perfusion method may also facilitate estimation of the number, size, distribution, and hormonal sensitivity of penile papillae.