Strio-pallidal projection in the monkey.

The corpus striatum in the mammalian brain appears to receive only two major groups of afferents: from the intralaminar nuclei of the thalamus and from most areas of the cerebral cortex. A common feature of both groups of afferents is their well-developed topographical organization. Rostral and caudal parts of the thalamus and cortex are related to the corresponding parts of the striatum (Powell and Cowan, 1956; Webster, 1961, 1965; Carman, Cowan, and Powell, 1963); in the medio-lateral dimension the organization is more complex, and does not conform, as might have been supposed, to the simple division of the striatum into a medial caudate nucleus and a laterally placed putamen. Instead, the available evidence suggests that the efferents from the greater part of the cerebral cortex and most of the intralaminar nuclei terminate on both sides of the internal capsule. Thus, for example, in the rabbit the cortex on the dorso-lateral surface of the hemisphere sends fibres both to the caudate and putamen (Carman et a!., 1963), and in the monkey, following a lesion in the centromedian nucleus degenerating fibres can be traced into both the caudate and putamen (Mehler, 1966). In view of this evidence it was of some interest to determine whether the functional organization imposed upon the architectonically homogeneous striatum by its afferents is reflected also in its efferent connexions. Our attention was drawn to this question several years ago when studying the thalamo-striate connexions (Powell and Cowan, 1956) because in most of the material used in that study discrete areas of intense gliosis were observed in different parts of the globus pallidus and substantia nigra with experiments involving varying portions of the caudate nucleus and putamen. The publication in 1960 and 1962 of papers by Voneida and by Szabo on the distribution of striatal efferents showed a close correspondence between the pattern of the projection of the striatum as determined by the Nauta method and our own findings on the distribution of gliosis in our earlier material and in several additional brains prepared since that time (Powell and Cowan, 1966). It was considered worthwhile, therefore, to put our observations on record not only because they provide independent confirmation of the organization of strio-pallidal and strio-nigral connexions, but also because they demonstrate the value of the study of the distribution of gliosis as an analytical method in fibre systems in which the direction of conduction is known. The organized nature of the afferent and efferent connexions of the striatum has assumed greater significance with the demonstration by Nauta and Mehler (1966) that the small-celled external segment of the globus pallidus sends fibres only to the subthalamic nucleus, whereas the large-celled inner segment is related to the thalamus and midbrain. Their additional finding that part of the projection from the internal segment is to the centromedian nucleus is of particular interest: first, because it may provide a 'circular pathway' from the striatum to the globus pallidus and thalamus, and back to the striatum; second, because of the substantial projection to the intralaminar nuclei of the thalamus (including the centromedian nucleus) from the cerebral cortex (Auer, 1956; Niimi, Kishi, Miki, and Fujita, 1963; Astruc, 1964; Petras, 1964; Mehler, 1966; Powell and Cowan, 1966). In the present paper, on the basis of our own and other workers' findings on the strio-pallidal projection, an attempt will be made to synthesize some of the recent anatomical work on the inter-relationships between the cerebral cortex, thalamus, and the striopallidum.