The influence of learning on the development of tolerance to the analgesic effect of morphine in rats was examined employing the hot plate procedure. A tested-reinforced (Tr) group and its yoked-control, a tested-non-reinforced (Tnr) group, received identical exposure to the testing procedure; the Tr group was reinforced daily for its behavior on the heated plate whereas the Tnr group was reinforced only on the last day of the experiment. Paired statistical comparisons between these two groups on the last day of the experiment revealed that: 1. premorphine control reaction times on the heated plate were significantly lower in Tr than in Tnr animals; and 2. post-morphine increases in reaction time did not differ between Tr and Tnr animals. It was concluded that whereas some learning does occur in this testing procedure, learning does not influence the behavioral tolerance to morphine which develops in this analgesiometric method. An hypothesis which accommodates this behavioral tolerance and a mechanistic scheme is offered.