Postnatal development and plasticity of corticocortical projections from area 17 to area 18 in the cat's visual cortex.

We used retrogradely transported fluorescent tracers to study the development of projections from area 17 to area 18 in normal and monocularly deprived kittens. In newborn animals, cells in area 17 that were labeled from small, discrete injections in area 18 were concentrated around the retinotopically corresponding zone, but distributed with lower density over a very wide surrounding area. Hence, the total convergence and divergence of the projection were initially enormous, but they decreased dramatically, mainly during the first postnatal month, through elimination of the sparse, widespread distribution of projections. Injections of two different tracers close together in area 18 produced very few double-labeled cells in area 17 at any age, implying that most individual axons arborize over very small territories even at birth. In normal kittens the peak density of association cells in the upper layers, corrected for the overall expansion of the cortex, doubled over the first postnatal month and then declined gradually over the following several months, presumably because of continuing selection and elimination. As shown in previous work (Price and Blakemore, 1985a), area 17 to 18 cells in newborn kittens were distributed in two continuous bands in supragranular and infragranular layers. During normal maturation, elimination of projections results in the formation of distinct clusters; these lie preferentially in the upper layers above patches of ipsilateral eye input to layer 4 (Price et al., 1994). Monocular deprivation, which causes the terminal patches representing the deprived eye to become much smaller than normal, did not stop the normal decrease in overall convergence/divergence or the appearance of clusters of association cells, but the clusters were distinctly larger than normal in both hemispheres. Monocular deprivation also prevented the normal reduction in density of association cells within clusters after 1 month of age. Comparison with results from binocularly deprived animals, where clusters also form but association cell density is low, suggests that the size of clusters and the density of association cells retained depend on the overall level of cortical activity.