Efferent projections from the paraflocculus of the tree shrew (Tupaia glis) were studied utilizing the Fink and Heimer ('67) method. Cerebellar corticonuclear fibers of both dorsal (Dpf) and ventral (Vpf) divisions of the paraflocculus terminate in the lateral cerebellar nucelus (NL) and in the posterior interposed nucleus (NIP). These fibers are ipsilateral, topographically organized and arranged into at least two zones. Following injury to either the Dpf or Vpf, degenerated axons are found in lateral and caudal regions of the NIP respectively. Consequently, these two portions of the paraflocculus have relatively exclusive terminal fields with overlap only at the periphery. Preterminal debris is seen in basically similar areas of the NL (caudolateral, caudal, ventral) after damage to either the Dpf or Vpf. This observation leads to the conclusion that the terminal fields for these areas of parafloccular cortex are largely coextensive in the NL. In addition to the topographical representation of the Dpf and Vpf in both the NL and NIP, there is evidence that these corticonuclear fibers are also organized into the general zonal pattern hypothesized by Voogd ('69). Persistent and numerous degenerated axons from both the Dpf and Vpf end in lateral and caudal NIP, respectively, corroborating the presence of a relatively wide zone C2 in both divisions of the paraflocculus. The Dpf and Vpf also project into the NL to a terminal field that appears to consist of two portions. One part located in caudal, ventral and/or caudoventral areas of the NL and a second at slightly more rostral and rostrolateral areas. The presence of a cortical region which is affiliated with two areas of the NL substantiates not only the existence of zone D in the paraflocculus, but gives experimental evidence that it may consist of two parts as previously suggested (Voogd, '69). From the Dpf many fibers enter the NL while few are seen in this nucleus after damage to the Vpf. This suggests that zone D is wider in the Dpf and narrower in the Vpf. The results of this study do not support previous suggestions that the paraflocculus projects to either only the NL or to most of, or all of, the ipsilateral cerebellar nuclei. No projections to any contralateral cerebellar nuclei are seen. There is also no conclusive proof that the parafloccular cortex of Tupaia projects into the vestibular complex or to any other brainstem relay nuclei.