With the exception of the tectopulvinar pathway the subcortical afferents to pulvinar still remain an enigma. In view of the suggestions by earlier authors, based largely on observations of normal material, that the pulvinar receives input from adjacent thalamic sensory relay nuclei, medial dorsal nucleus 18, amygdala 14 and retina 5, we have attempted to identify subcortical structures efferent to pulvinar in rhesus monkey using horseradish peroxidase (HRP). A preliminary report of these findings has appeared elsewhere 17.

Four rhesus monkeys received unilateral injections of HRP into pulvinar subsequent to transection of the splenium of corpus callosum and visualization of this nucleus. In 5 additional monkeys bilateral stereotaxic injections were made using the atlas of Olszewski 13 for coordinates after first taking a zero reading from the habenular commissure subsequent to transection of the splenium. Injections were made using 10~ and 20~ o aqueous horseradish peroxidase (Sigma VI) via a 10# 1 Hamilton syringe in volumes of 0.3-0.6/zl. The aqueous solution also contained dissolved tritiated amino acids so that at a later time the efferents of pulvinar could be studied. The animals survived for 2-3 days and after being deeply anesthetized were perfused with a mixture of 0.5')~, paraformaldehyde and 2.5~; glutaraldehyde buffered with sodium cacodylate or phosphate buffer at pH 7.2 preceded by washing with a solution of 6~, i Dextran in 0.9.'o saline. The retinas as well as the upper 5-7 segments of cervical cord in addition to the brains were removed and the tissue was processed as previously described8, 16 for HRP. A series of every tenth section cut at 40/~ m in the stereotaxic plane or 80/~ m in the parasagittal plane in the case of the retinas was examined for the presence of HRP positive neurons under dark-and light-field illumination using an XY plotter. The cover slips were then removed and the previously examined sections were stained with cresyl violet in order to determine the exact location of the HRP labeled cells. The injection sites were reconstructed and indicated on representative thalamic diagrams based on the atlas of Olszewski 13. Of the 14 attempted injections 12 were identified. In 2 animals, ie 4 and 5 (Fig. 2) in which bilateral injections were attempted, no injection site could be identified in one of the hemispheres of each animal. The needle tract could be identified but insufficient HRP was deposited in pulvinar due to plugging of the needle and