The finding that serotonin (5-HT) treatments as short as 1.5 h in duration produce phase shifts in a circadian rhythm from the isolated eye of Aplysia suggested that release of 5-HT was part of an ocular entrainment pathway. Since light cycles entrain this rhythm, we compared phase shifting by 5-HT and by light. The similarity in the shapes of the phase-response curves for 5-HT and light pulses indicates that 5-HT treatments are capable of entraining the rhythm. Also, "skeleton" 5-HT treatments phase shift as well as continuous 5-HT treatments. However, 5-HT does not appear to mediate the phase shifts produced by light, since 1) treatments that should block transmitter release do not change the phase shifts produced by light pulses; 2) the response curves of 5-HT and light pulses are displaced by 12 h relative to one another on the phase axis of the response curve; and 3) light-induced phase shifts are apparent almost immediately, whereas 5-HT-induced phase shifts become evident only about 24 h after 5-HT treatment. The eye appears to contain two independent entrainment pathways, one for light and one utilizing 5-HT.