The development of useful Simple Sequence Repeat (SSR) or microsatellites DNA markers Is dependent upon a source of SSRs that are frequent In the genome and maximally polymorphic. In our effort to develop a microsatellite map of soybean, numerous (AT)n, (CT)n, and (ATT)n, loci have been identified. It was the purpose of the work reported here to assess the level of length polymorphism associated with loci having these core motifs. A representative group of 40 (AT)n, 20 (CT)n, and 35 (ATT)n loci were characterized in ten soybean cultivars: Williams, Amsoy, Fiskeby V, Jackson, Tokyo, Minsoy, Noir, Archer, and isolines of Clark and Harosoy. In William soybean, the average number of repeats (n) of the three groups of was 24, 14, and 15, respectively. Gene diversity was calculated for each marker. The average gene diversity for the (AT)n, (CT)n, and (ATT)n microsatellite loci was 0.74, 0.29, and 0.50, respectively. The low level of length polymorphism of (CT)n SSRs makes them poor candidates for efficient use in cultivated soybean genetic materials. In contrast (AT)n microsatellites function as highly informative markers. However, difficulties in optimizing PCR amplification of (AT)n loci are frequent as are problems of multi-band products and 'stuttering'. Although (ATT)n SSRs are less polymorphic than those with AT core motifs, PCR amplification of ATT-repeats can usually be optimized with few difficulties. Based upon these findings we have focused our soybean microsatellite development work on (ATT)n and other trinucleotide SSRS.