Soybean genotypes with high oleic acid and low linolenic acid contents are demanded as a source of raw material for the food industry. At the Soybean Breeding Program for Quality and Flavor (PQMS/BIOAGRO), there are accessions with modified fatty acid contents which are being incorporated to the program as new sources of variability. In the present work the genetic distances among seven soybean progenitors (BARC-12, B412113, C2022, CS-303, CS92 6602, N85-2124, N85-2176), which include new sources of modified fatty acid contents, were determined. The results should provide guidelines for prediction of the best combinations among the genotypes. The progenitors were evaluated in a completely randomized design with ten repetitions. The fatty acid composition was determined by gas chromatography from bulks of seeds in each repetition. The matrix of distances among the progenitors was obtained based on Mahalanobis generalized distances. The progenitors were grouped in a dendrogram for a better visualization of the distances. By the Tocher cluster method, four progenitor groups were formed: group 1: B412113, CS303; group 2: BARC-12, N85-2176; group 3: C2022, N85-2124; group 4: CS92 6602. The best predicted combinations were those obtained for the progenitors in different groups. Based on the performance of the genotypes (fatty acid content), the progenitors of group 2 are recognized as good sources for increasing oleic acid content and groups 2 and 3, for the reduction of linolenic acid content.