We studied the projections to the caudate nuclei, thalami and red nuclei from the remaining sensorimotor cortex in adult cats that had sustained a unilateral frontal cortex resection prenatally or neonatally. Four cats had the lesion at age E 50-55 and six animals sustained the ablation at age P 8-14 (seven cats were intact controls). All cats grew to young adulthood and then received injections of tritiated leucine-proline in the remaining sensorimotor cortex. Injection sites and axon terminal fields were reconstructed using autoradiography-processed tissue. In all cats the label filled a similar extent of the right pericruciate cortex. Terminal field densities in the subcortical nuclei were estimated using computer-based video software. Three medial-lateral sectors at five coronal levels were examined in the caudate nucleus. Three nuclear groups were analyzed in the thalamus (intralaminary, ventralis lateralis and ventrobasal complex). For the red nucleus, the four quadrants were examined at four coronal levels. The main goal of the study was to assess possible changes in the cortical innervation of the nuclei ipsilateral to the lesion. Therefore, the mean particle counts per nucleus (and per area or sector of nuclei) and per animal group were used to calculate percentage values for the decussated (crossed, or contralateral to the injection site) as a function of the non-decussated (uncrossed, or ipsilateral to the injection site) innervation. The percentage values for the crossed projections were: (a) for the entire caudate nucleus, 61.3% for the intact. 56.7% for the fetal-lesioned and 42.7% for the neonatal-lesioned cats, with no statistical differences between groups; (b) for the thalamus the proportion of crossed projections was minimal fluctuating between a low 0.06-0.16% for the nucleus ventralis lateralis and a high of 2.01-3.46% for the intralaminary nuclei, with the highest values belonging to the lesioned groups but with no significant differences between groups: (c) for the entire red nucleus, 1.98%, 12.74% (P < 0.05) and 6.76% for the intact, fetal- and neonatal-lesioned cats respectively. In the lesioned cats, the topography of the distribution of the axon terminals was bilaterally the same as in the controls. In conclusion, only the red nucleus of the frontal-lesioned cats showed an increased crossed innervation from the remaining sensorimotor cortex but this was relatively weak and statistically significant only for the fetal-lesioned animals. These results as well as the literature suggest that: (a) the crossed corticorubral projections in fetal cats may represent true reinnervation (i.e., newly originated, no preexisting terminals); (b) the relative paucity of the crossed projections in the present cats as compared to the extensive reorganization of subcortical terminals seen after cerebral hemispherectomy (our original postnatal lesion model) may be due to the much smaller size of the present cortical lesion which presumably induced only a limited amount of subcortical nuclear deafferentation.