Using the retrograde horseradish peroxidase technique, we have examined the distribution of labeled thalamic-, spinal- and cerebellar-projecting neurons in the trigeminal sensory nuclei of the cat. Injections into the nucleus ventralis posterior of the thalamus resulted in labeling of neurons in lamina I (subnucleus zonalis), the deeper part of lamina IV (the subnucleus magnocellularis) of the nucleus caudalis and in lamina V (the lateral extension of the nucleus medullae oblongatae centralis) on the contralateral side. A very large number of labeled small neurons were observed mainly in the caudal part of the nucleus interpolaris and in the ventral division of the principal sensory nucleus on the contralateral side and in the dorsal division of the principal sensory nucleus on the ipsilateral side. Injections into the known projection areas of the cerebellar cortex labeled mainly ipsilaterally the trigeminocerebellar neurons in a restricted ventrolateral area of lamina IV of the nucleus caudalis at its rostral level and in lamina V. Many labeled neurons were also observed in the nucleus interpolaris. Although the distribution overlapped with that of the trigeminothalamic neurons, the greatest majority were concentrated in its rostral part where the trigeminothalamic neurons were very small in number. In addition, labeled neurons were observed in the rostral part of the nucleus oralis and the ventralmost part of the ventral division of the principal sensory nucleus. No labeled neurons were observed in the dorsal division of the principal sensory nucleus and the mesencephalic nucleus. The trigeminospinal neurons were labeled mainly ipsilaterally following injections into the upper cervical cord. They were located in laminae I and III, the deeper part of lamina IV of the nucleus caudalis and in lamina V. Only scattered labeled neurons were found in the nucleus interpolaris. The number of labeled neurons increased in the nucleus oralis at the level of the superior olive. They tended to be distributed around or dorsal to the group of the trigeminothalamic neurons at the caudal part of the principal sensory nucleus. No neurons of the principal sensory nucleus appeared to project to the spinal cord. Based on the large size and location, the trigeminospinal neurons could be differentiated from the other projection neurons in the nucleus oralis. The present study demonstrates that the trigeminal sensory nuclei are composed of groups of neurons with different projections, since the main aggregations are localized at different levels. However, it should be examined whether the neuronal groups, which are labeled from the different structures in similar locations, are composed of individual neurons projecting to more than one of these structures.