This study investigated the hypothesis that for a subpopulation of horse mares continuation of estrous cycles during the nonbreeding season may be attributed to continued stimulatory glutamatergic activity on GnRH-secreting neurons. The gonadotropin response to the glutamatergic agonist N-methyl-DL-aspartic acid (NMA) was compared in cycling and anestrous mares during the nonbreeding season. It was anticipated that the gonadotropin response to NMA in cycling mares would be attenuated, compared with that of anestrous mares. The experiment used 16 anestrous mares and 15 mares that cycled during the nonbreeding season. The effect of NMA on prolactin secretion was also evaluated. In addition, the seasonal rhythm of prolactin secretion was compared in anestrous and cycling mares during October-April. In cycling mares, the response to NMA was dependent on the stage of the cycle, and a significantly (p < 0.05) larger proportion responded during the luteal phase (6 of 8), compared with the follicular phase (1 of 7 mares). The proportion of anestrous mares that responded to NMA was similar to that of cycling mares during the luteal phase, but larger than during the follicular phase. In anestrous and cycling mares, NMA suppressed prolactin secretion, and in both groups prolactin secretion decreased during the nonbreeding season. Thus, we conclude that differences in reproductive activity in mares during the nonbreeding season are unlikely to reflect a change in glutamatergic activity.