1. The response to static tilts was studied in peripheral otolith neurons in the barbiturate-anesthetized squirrel monkey (Saimiri sciureus). Each unit was characterized by a functional polarization vector, which defines the axis of greatest sensitivity. A circumstantial criterion was used to assign units to the inferior (IN) or superior (SN) vestibular nerves. The former neurons should innervate the sacculus, the latter mainly the utriculus. Confirming pasting experiments, the polarization vectors for SN units lay near the plane of the utricular macula, those for IN units near the plane of the saccular macula. The polarization vectors for IN units were compared in two groups of animals. In one group, the vestibular nerve was intact; in the other, the superior nerve was sectioned. No differences were found and this, together with other observations, demonstrate that the sacculas in mammals functions mainly (if not solely) as an equilibrium organ. 2. The resting discharge of otolith neurons averages some 60 spikes/s, the sensitivity some 30-40 spikes/s-g. IN units tend to have slightly lower sensitivities than do SN units. IN units with upwardly directed (+Z) vectors have substantially higher resting discharges than do units with downwardly directed (-Z) vectors. The +Z units are also characterized by more linear force-response relations. 3. There is a strong positive relation between the resting discharge and sensitivity of units characterized by regular steady-state discharge patterns. A weaker, but statistically significant, relation is demonstrable for irregular units. It is suggested that the relation seen in regular units is the result of the neurons differing from one anothrer in terms of a receptor bias, a transduction gain, or both. Only the mechanism based on transduction gain is thought to be operative among the population of irregular units. 4. Centrifugal-force trapezoids were used to study the response adaptation to prolonged stimulation. Adaptation was more conspicious in irregular units and was characterized by perstimulus response declines and poststimulus secondary responses. For regular units, adaptive properties were similar during excitatory and inhibitory responses. For irregular units, response declines were larger during excitatory stimuli, secondary responses larger following inhibitory stimuli. Typically, response declines were most rapid at the start of the force plateau. A few units, all of them irregular, exhibited a delayed adaptation with response declines beginning only after a constant force had been maintained for 10-20 s. 5. Excitatory responses of regular units are almost always larger than inhibitory responses. This is so during both the dynamic and static portions of force trapezoids. A similar asymmetry is seen in the dynamic response of irregular units; static response asymmetries of the latter units are more variable.