A total of 1037 individuals living in five different sociogeographic regions of Brazil were studied in relation to 12 short tandem repeat polymorphisms. The objective was to assess the degree of European, African, and Amerindian contributions to their gene pools. Although most of the genetic variability was found within regions, significant differences were also observed between regions. The estimated relative proportions of the above-indicated continental contributions showed intermediate values between those obtained with uniparental (mtDNA, Y-chromosome) data, and a higher percentage of European heritage as compared to previous autosome results. A north-south trend of increasing European contribution was also found, in agreement with the history of the Brazilian population.

New frequencies for the beta-globin gene cluster haplotypes are presented for the Aché (N = 82 individuals), Guarani (N = 76), and Kaingang (N = 54), three Native South American populations that live in an area between parallels 20 degrees S and 30 degrees S not covered by previous studies at this locus. The haplotype frequencies obtained for the three populations are within the interval observed for 28 other Native American populations. The Aché show much less haplotypes (five) than the other two populations (9-10), the haplotype prevalences being more similar to those of the Guarani than to the Kaingang. The Native American total heterozygosity was about half (0.41) that obtained for the African populations (0.71), but was not much different from those obtained for other continents. A geographical pattern was disclosed in South America by mapping the frequencies of the most common haplotype (haplotype 2), and by means of spatial correlation analysis. The analysis of molecular variance (AMOVA) and pairwise F(ST) data suggest three distinct sectors for the genetic landscape of Native South America: the Andes, the Center/Southeast region, and the Amazon.

To investigate the population structure and variation in Gran Chaco's Amerindian population, data from 15 short tandem repeats (STRs) were determined in 128 individuals from three tribes of the Argentinean part of this region. STR genotypic differences, structure analysis, and multidimensional plot for the D(A) distances indicated that (1) Wichí from the Chaco Province are genetically distinct from the other populations, but still preserve a fair amount of genetic similarity with Wichí from Formosa; (2) the Toba populations studied are genetically indistinguishable; and (3) Toba subjects from Formosa are similar to the Pilagá of the same linguistic group (Guaykurú) and to the Wichí from Formosa who speak a Mataco language. This similarity could be due to their past mobility and the custom of absorbing females taken as prisoners from groups raided by them. Language, geography, and genetics seem to play similar roles in determining the population structure of these groups. Analyses of molecular variance and G(ST)' values calculated considering three South American regions indicated that the Argentinean Chaco is genetically homogeneous; addition of the Ayoreo Amerindians of the Paraguayan Chaco, however, led to diversity values that are not much different from those of South Amerindians in general. The present data contribute to efforts that try to understand in what way groups with diverse sociocultural settings (tribal, agricultural, and industrial) differ in genetic structure.

The new data provided additional important information about the biological history of people from a remote South American region, indicating how gene diversity analyses can be used to increase understanding of human microevolutionary processes.