Spatial relationships between the terminations of somatic sensory and motor pathways in the rostral brainstem of cats and monkeys. I. Ascending somatic sensory inputs to lateral diencephalon

Abstract

Projections to the lateral diencephalon from the dorsal column nuclei (DCN), lateral cervical n. (LCN), and spinal cord (ST) in cats and monkeys, and from the spinal portion of the trigeminal n. (sTN) in the cat were compared using a double‐orthograde labeling strategy. This strategy combines autoradiographic and degeneration tracing methods in the same animal and permits direct comparisons of the terminal labeling patterns of two different pathways in each experiment. The results suggest that the major part of the lateral diencephalon which receives input from the somatic sensory pathways in both the cat and the monkey is arranged in a core‐shell fashion. The core consists of the group of nuclei which together constitute the ventrobasal complex (VB). The shell consists of a group of nuclei which together tend to surround VB nearly completely. This group includes the posterior group (PO), the ventral posteroinferior n. (VPI), and the border region between VB and the ventrolateral n. (VB‐VL). In addition to the core and shell regions, two other regions in the lateral diencephalon receive input from the somatic sensory pathways. These regions are the ventromedial part of the magnocellular portion of the medial geniculate n. (MGNm) and caudomedial portion of the zona incerta (ZI). The cytoarchitectural and hodological patterns of the core region differ from those of the shell region. In both the monkey and the cat, the core region (VB) has a relatively homogeneous cytoarchitectural appearance and is filled by dense inputs from DCN, LCN, and sTN in the cat and from DCN, LCN, and ST (and probably from sTN) in the monkey. Direct comparisons of the terminals of fibers from different pathways demonstrate that although there is some convergence on the same neurons within VB, the major tendency is for each of the inputs to form its densest terminations on different neurons. This partial segregation manifests itself in two ways. First, each pathway has its own preferred territory within VB where its terminations are the densest. Second, the terminal fields of the inputs usually have a clustered appearance which is characterized by dense patches of terminals separated by regions in which the terminations appear quite sparse. The dense patches from different pathways do not occur in relation to the same groups of neurons. In contrast, most portions of the shell region have a lower cell density than that of the core and a heterogeneous cytoarchitectonic appearance which can often be described as transitional in character between its neighboring areas. In both species, different parts of the shell region receive sparse and scattered input from those pathways which project densely and precisely to areas immediately adjacent to that part of the shell. Very few of the terminals of these different inputs appear to converge on the same groups of neurons. The two otehr recipient targets of somatic senory input (i.e., MGNm, ZI) each has its own characteristic connective pattern that differs from taht of either the core or the shell region. The connective patterns in the cat and monkey are quite similar. The mian differences are in the projections of parts of the ST ans LCN pathways. The nature of these differences suggest taht it might be useful from a functional perspective to consider the LCN and ST pathways together as part of the same spinal system, rather than as separate functional entities. The LCN pathyway could then be viewed as having perhaps been dervied from different parts of a single population of diencephalic‐projecting neurons in the spinal cord of the two species. When these anatomical results are considered together with the available electrophysiological evidence, it appears that the response properties and functions of some portions of the somatic senory regions within the diencephalon can be generally predicted from knowledge of the particular pathways whose axons terminate within these regions. Such predictions can be made, however, only when the input pathways have markedly different functions (e.g., vestibular, auditory, cutaneous). At present, more precise kinds of predicitions are precluded by the similarity that exists between the functional properties of many of the units in teh DCN, sTN, LCN, and ST pathways, and the luck of knowledge of the sorting processes which occur as fibers in each of these pathways diverge to terminate in different parts of the brain.