The present study analyzed the relationships of glycine (GLY)-immunoreactive (-IR) and unlabeled profiles in the primate spinal cord. Light microscopic analysis demonstrated GLY-IR profiles in laminae III-VII, with fewer labeled profiles in laminae I, II, VIII, IX and X. The dorsal part of the lateral funiculus and the dorsal funiculus contained few labeled axons, in contrast to all other areas of white matter, which were heavily labeled. At the electron microscopic level, GLY-IR terminals in monkeys contained mainly round, with occasional pleomorphic, clear vesicles; however, F-type GLY-IR terminals synapsing on motoneurons contained pleomorphic vesicles. This seems to be an important species difference because vesicles in GLY-IR terminals in rat and cat are predominantly oval or elliptical. GLY-IR terminals synapsed on unlabeled as well as GLY-IR cell bodies and dendrites. This is morphological evidence that GLY may be both an inhibitor (GLY-IR terminals synapse on and presumably inhibit non-GLY cell bodies and dendrites) and a disinhibitor (GLY-IR terminals synapse on and presumably inhibit other GLY elements) of spinal activity. Also noteworthy was the conspicuous absence of axoaxonic interactions involving GLY-IR terminals. A related finding was that GLY profiles were always postsynaptic, never presynaptic, to glomerular primary afferent terminals. The functional implications would seem to be that primary afferent input can activate the spinal GLY system but that there is little GLY presynaptic control of afferent input in monkeys. This is in contrast to rats and cats, in which axoaxonic interactions involving GLY-IR terminals have been observed and where it is common to find GLY-IR terminals presynaptic to glomerular primary afferent terminals.