1. We have recently demonstrated that reversible, cooling-induced inactivation of the face motor cortex results in a severe impairment in the ability of monkeys (Macaca fascicularis) to perform a tongue-protrusion task but produces only relatively minor effects on the performance of a biting task by the same monkeys. To establish a neuronal correlate for these different behavioral relations, the present study has detailed the afferent input and intracortical microstimulation (ICMS)-defined output features of a population of face motor cortical neurons, and in a subsequent study we have documented the activities of the same population of neurons during the performance of the tongue-protrusion and biting tasks. 2. Of the 231 single neurons recorded within the face motor cortex, 163 were located at sites from which ICMS (less than or equal to 20 microA) could evoke tongue movements (i.e., "tongue-MI" sites) at the lowest threshold for eliciting orofacial movements. The remainder were located at sites from which ICMS evoked jaw movements ("jaw-MI" sites), face movements ("face-MI" sites), or at a few sites, tongue movements and, at the same threshold intensity, either a jaw movement or a facial movement. 3. We confirmed the general organizational features of the face motor cortex that have been defined in previous studies, but we documented in detail the organizational features for tongue-MI. Thus we found that tongue movements were well represented, whereas jaw-closing movements were poorly represented; the representations for face, jaw, and tongue movements were overlapped; the same ICMS-evoked tongue movement could be multiply represented within tongue-MI; tongue-MI was characterized by a prominent input from superficial mechanosensory afferents, whereas there was little evidence for deep input; a close spatial match was found between ICMS-defined motor output and somatosensory afferent input for tongue-MI. 4. A variety of tongue movements could be evoked by ICMS at tongue-MI sites and were categorized into protrusion, retrusion, laterally directed, and other types of tongue movement. Low-threshold (i.e., less than or equal to 5 microA) ICMS-defined tongue-MI sites, which were considered to represent "efferent zones" projecting relatively directly to motoneurons, were reconstructed three dimensionally to provide insights into the spatial organization of tongue-MI. Examples of each of the four low-threshold efferent-zone categories were usually found throughout the ICMS-defined tongue-MI without any apparent preferential distribution. Furthermore, different low-threshold efferent-zone categories had close spatial relationships to each other in cortex.(ABSTRACT TRUNCATED AT 400 WORDS)