Cortical projections to the dorsal column nuclei (DCN) of cats arise from layer V pyramids throughout the sensorimotor cortex and also from the second somatosensory cortical area (SII). In this study, the relative contribution from various cortical areas to different portions within the DCN-complex has been investigated in cats with restricted cortical lesions and using the Fink-Heimer technique.

The sensorimotor cortex (areas 4, 3, 1, and 2) projects throughout most of the rostro-caudal extent of the gracile nucleus. Densest projections arise from the upper bank of the cruciate sulcus containing mainly the hindlimb region (area 4), from the medial wall of the posterior sigmoid gyrus (mainly area 3a) and terminate in the rostral as well as in the clusters regions of the gracile nucleus. The contribution of cortico-gracile projections from the hindlimb regions of areas 3b, 1 and 2 are much less conspicuous and terminate mainly in the rostral region of the nucleus. The nucleus z, the medullary relay for Ia fibers from the hindlimb, receives exceedingly sparse cortical projections. Dense projections to the cuneate n. arise from the cortical forelimb representation in the posterior sigmoid gyrus (areas 3a and 4) and laterally in the upper bank of the cruciate sulcus (area 4). The distribution pattern of cortical projections within the cuneate nucleus is however different from that within the gracile n. since (a) none of the lesioned cortical areas in the forelimb representation project to the clusters region, and (b) cortical projections terminate most densely in the base of the cuneate n., a medullary relay for Ia fibers from the forelimb. Sparse projections to the rostral region of the cuneate n. arise uniformly from all cytoarchitectonic areas of the forelimb representation. Cortical projections from SII are sparse and confined to the base and rostral region of the cuneate n.

The results provide further evidence for the origin of descending projections upon the DCN mainly from cortical regions related to movement control. Further, it is suggested that cortico-DCN control is biased toward nuclear subdivisions related to cutaneous afferents from the hindlimb within the gracile n.-nucleus Z complex but toward nuclear subdivisions related to deep and muscle afferents from the forelimb within the cuneate nucleus.