Extent of genic variability among 22 tribal groups of South American Indians in terms of net codon differences per locus has been studied on the basis of determinations on ten genetic systems. The possibility of more than one (other than the north-south route, considered by others) migrational pattern is discussed in the light of the existing genetic variability among the Andean highland and the jungle populations. Genic similarities (as measured by gene identity) are related with geographic proximities which reflect the importance of random genetic drift in creating the present genic divergence among these population groups. Although the materials used in this paper form probably only a non-random sample of structural genome, intralocus variance (sampling) is found to contribute only a little in the variabilities of heterozygosity or genetic distance.