1. We have studied the responses of neurons in the lateral and descending vestibular nuclei of decerebrate cats to stimulation of neck receptors, produced by rotating the body in vertical planes with the head stationary. The responses to such neck stimulation were compared with the responses to vestibular stimulation produced by whole-body tilt, described in the preceding paper. 2. After determining the optimal vertical plane of neck rotation (response vector orientation), the dynamics of the neck response were studied over a frequency range of 0.02-1 Hz. The majority of the neurons were excited by neck rotations that brought the chin toward the ipsilateral side; most neurons responded better to roll than to pitch rotations. The typical neck response showed a low-frequency phase lead of 30 degrees, increasing to 60 degrees at higher frequencies, and a gain that increased about threefold per decade. 3. Neck input was found in about one-half of the vestibular-responsive neurons tested with vertical rotations. The presence of a neck response was correlated with the predominant vestibular input to these neurons; neck input was most prevalent on neurons with vestibular vector orientations near roll and receiving convergent vestibular input, either input from both ipsilateral vertical semicircular canals, or from canals plus the otolith organs. 4. Neurons with both vestibular and neck responses tend to have the respective orientation vectors pointing in opposite directions, i.e., a head tilt that produces an excitatory vestibular response would produce an inhibitory neck response. In addition, the gain components of these responses were similar. These results suggest that during head movements on a stationary body, these opposing neck and vestibular inputs will cancel each other. 5. Cancellation was observed in 12 out of 27 neurons tested with head rotation in the mid-frequency range. For most of the remaining neurons, the response to such a combined stimulus was greatly attenuated: the vestibular and neck interaction was largely antagonistic. 6. Neck response dynamics were similar to those of the vestibular input in many neurons, permitting cancellation to take place over a wide range of stimulus frequencies. Another pattern of interaction, observed in some neurons with canal input, produced responses to head rotation that had a relatively constant gain and remained in phase with position over the entire frequency range; such neurons possibly code head position in space.