Bifurcation of subcortical afferents to visual areas 17, 18, and 19 in the cat cortex.

We have examined the pattern of axon bifurcation in the thalamic and claustral afferents to visual areas 17, 18, and 19 in the adult cat neocortex. This was achieved by injecting two fluorescent retrograde tracers, fast blue and diamidino yellow, in retinotopically corresponding regions of two of these three cortical areas. The pattern of single- and double-labelled cells was then examined in subcortical structures and the presence of double-labelled cells was interpreted as indicating that these neurons send bifurcating axons to the two injected areas. The size of the cortical region surrounding the injection site where each fluorescent dye is taken up was studied by making side-by-side injections of the two tracers in area 17 and examining the size and the separation of the two groups of labelled cells in the lateral geniculate nucleus (LGN). From these experiments we conclude that the uptake region is smaller than 1 mm and is included in the region of dense coloring surrounding the track of the injection needle. Injections were made in cortical regions which were in retinotopic correspondence as determined by electrophysiological recording. The double-labelled neurons were always found in the zone of overlap of the two populations of colored cells and no double-labelled neurons were found when there was no overlap between these populations. This indicates that the bifurcating axons send branches to strictly retinotopically corresponding regions in the two cortical areas. After injections in areas 18 and 19, numerous double-labelled cells were observed in laminae C of the LGN, in the medial interlaminar nucleus (MIN), the posterior nucleus (PN), and the lateral part of the lateral posterior nucleus (LP), in the retinorecipient zone of the pulvinar (RRZ-Pul), the intralaminar nuclei (ILN), and the claustrum. The proportions of double-labelled cells with respect to the total number of labelled neurons were computed in the region of overlap of the two populations of labelled cells. These percentages ranged between 5 and 20% and were highest in the C laminae of the LGN, the intralaminar nuclei, and the claustrum. After injection of areas 17 and 18, similar proportions of double-labelled cells were observed in the same structures, as well as in the A laminae of the LGN. Here again, the intralaminar nuclei and the claustrum tended to have slightly higher (20-30%) proportions of double-labelled cells. When the nonadjacent areas 17 and 19 were injected, doubled-labelled neurons were also observed in all these structures, except the A laminae of the LGN.(ABSTRACT TRUNCATED AT 400 WORDS)