Intrafascicular multiunit activity and impulses in single mechanoreceptive afferents were recorded from the human lingual nerve with permucosally inserted tungsten microelectrodes. Nylon filaments and blunt glass probes were used for mechanical stimulation of the mucosa of the dorsal surface of the tongue. The innervation territories of nine nerve fascicles were mapped during multiunit recordings. All fascicle fields included the tip of the tongue, suggesting a particularly high innervation density for this area. Thirty-three single mechanoreceptive afferents were isolated and studied. Of these afferents, 22 were characterized by very small mucosal receptive fields (range: 1-19.6 mm2; geometric mean: 2.4 mm2) and responded to extremely low mechanical forces (force threshold range: 0.03-2 mN; geometric mean: 0.15 mN). As such, it was concluded that these "superficial" units terminated near the surface of the tongue. The remaining 11 units responded to probing of large areas of the tongue (> 200 mm2) and exhibited high force thresholds (> or = 4 mN). It was concluded that these "deep" units terminated in the muscle mass of the tongue. Fourteen of the superficial units were classified as rapidly adapting and resembled the fast-adapting type I afferents described for the glabrous skin of the human hand. The rapidly adapting units responded both during the application and removal of, but not during maintenance of, the mechanical stimuli on the receptive field. Two types of slowly adapting responses were observed. One type (characteristic of only 2 units) was characterized by a pronounced sensitivity to force change during the application and removal of the mechanical stimuli and an irregular static discharge during maintenance of the stimulus on the receptive field. In contrast, the other six units exhibited a weak sensitivity to force change, a highly regular static discharge, and spontaneous activity. As such, these two types of slowly adapting units resembled the slowly adapting I and II afferents, respectively, described for the hand. All 11 deep units were slowly adapting, and 7 were, in addition, spontaneously active. The units were not equally sensitive to the application and removal of the mechanical stimuli, suggesting at least two different modes of termination in tongue muscle. The deep units reliably encoded information about tongue movements in the absence of direct contact with the receptive field. In contrast, the superficial units responded vigorously when the tongue was moved to bring the receptive field into physical contact with other intraoral structures.