The aim of this study was to evaluate the efficacy, optimal dose, and optimal time-window of gacyclidine, a novel N-methyl-D-aspartate (NMDA) receptor antagonist, in terms of its functional, histopathological, and electrophysiological effects after experimental spinal cord injury. The spinal cord of rats was damaged by a photochemical method and the animals were treated by saline or gacyclidine at doses of 1, 2.5, or 5 mg/kg 10 min after injury or gacyclidine 1 mg/kg 10, 30, 60, and 120 min after injury. The time-course of the motor score (walking and inclined-plane stability) was evaluated until day 18, and somatosensory evoked potentials were determined on day 18. The animals were then sacrificed, and the cross-sectional area of the spinal cord (at the epicenter of the injury, above and below the injury) was measured. Walking recovery was better in most of the groups treated after injury than in the untreated injured animals. Motor performances were related to preservation of a larger undamaged area of spinal cord at the level of the injury and, interestingly, with prevention of extension of the anatomical lesion above the level of the injury. Somatosensory evoked potential amplitudes were often higher in treated groups. These results confirm that gacyclidine induces dose-dependent and time-dependent attenuation of spinal cord damage after an experimental vascular lesion. Although all three doses induced neuroprotective effects, recovery was greater and very homogeneous in the group treated with 1 mg/kg. Moreover, recovery was slightly better and more homogeneous within the groups treated 10 and 30 min after injury compared to the other groups. It appears that, according to the existing evidence, NMDA antagonists are an essential component in the elaboration of a neuroprotective strategy after spinal cord trauma.