Several investigations have shown that the N-methyl-D-aspartate (NMDA) antagonist MK-801 interferes both with learning and a candidate neural mechanism for learning, the long-term potentiation (LTP) phenomenon. The low doses of MK-801 reported to block learning, however, may not be sufficient to block LTP. The present experiments examined the effects of 4 doses (0.05, 0.10, 0.50 and 1.00 mg/kg) of MK-801 on LTP of the perforant path-granule cell population excitatory postsynaptic potential (EPSP) and spike in unanesthetized rabbits. MK-801 did not significantly affect the threshold intensity for LTP of the population EPSP but the 3 highest doses did increase the threshold for LTP of the population spike. The 3 highest doses also reduced the peak magnitude and the duration of LTP of the spike to less than 24 h. The 0.05 m/kg dose did not affect the threshold or peak magnitude of spike LTP, but did decrease the decay time constant to 10.4 days, compared to 20.8 days in control rabbits. Only the 1.00 mg/kg dose reduced the magnitude of LTP of the EPSP. It was not possible to determine if MK-801 altered the longer-lasting component of LTP of the EPSP as it never persisted for longer than 24 h. These studies demonstrate that MK-801 disrupts LTP at doses also known to interfere with learning.