Abstract
The cytoarchitecture of areas 5a and 5b of the cat's parietal cortex was re-examined and the afferent connections from the thalamus were investigated using the horseradish peroxidase (HRP) retrograde transport technique. Single or multiple small injections of the enzyme were made in different points of these areas in the rostral sectors of the lateral and middle suprasylvian gyri. The cytoarchitecture of the cortical region affected by the injections was carefully assessed in each case, and the labeled neurons found in the thalamus were plotted on projection drawings of each histological section. A prominent projection to area 5a arises from the posterior (Po) and ventral lateral (VL) complexes; less substantial projections originate in the ventral anterior nucleus (VA), the lateral intermediate complex (LI), and the central lateral nucleus (CL). Projections to area 5b (and to the laterally adjacent area suprasylviana anterior) mainly arise from LI, the dorsal part of VL, and the caudodorsal part of VA and CL; a moderate projection was also found from Po, the pulvinar, and the lateral dorsal complex. The main conclusions of this study are as follows. The shape and extent of areas 5a and 5b show notable variations when only their projection on the convoluted cortical surface is considered; however, they are relatively constant when plotted on unfolded cortical maps. The thalamic neurons projecting to areas 5a and 5b are organized according to a loose topographic plan, particularly noticeable in Po, VL and LI. In general, the rostral portion of this cortex (5a) receives projections from more ventral regions of the thalamus (mainly Po and VL), whereas the caudal part (5b) has connections from more dorsal regions (mainly LI and VA-VL). Moreover, the medial portions of these areas receive projections from lateral and ventral parts of the thalamic nuclei, whereas more dorsal and medial sectors of the thalamus project to the lateral portions of areas 5a and 5b. When labeled thalamic cell populations resulting from cases with single injections in neighboring cortical loci were compared, no abrupt changes of labeling were observed; rather, we generally observed gradual transitions and overlaps, even across nuclear boundaries. When only layers I and II of the cortex received the HRP, the number of labeled neurons and the intensity of their labeling decreased, their location in the thalamus was more restricted, and the mean size of the labeled cells was significantly smaller than that of the neurons labeled in the same regions after deep HRP injections.