Restriction fragment length polymorphisms (RFLPs) have been proposed as molecular markers for characterizing the genetic diversity in maize (Zea mays L.). The objectives of this study were to evaluate the usefulness of RFLP data for (i) elucidating heterotic patterns among maize inbreds and (ii) assessing genetic similarity among related and unrelated lines. Thirty-two maize inbreds from the U.S. Corn Belt were analyzed for RFLPs with two restriction enzymes and 83 DNA probes distributed over the maize genome. Eighty-two probes detected polymorphisms with at least one enzyme. On average, 4.3 variants were found per probe-enzyme combination across all 32 inbreds. Genetic distances among lines, estimated from RFLP data as Rogers' distance (RD), revealed considerable diversity among lines from Iowa Stiff Stalk Synthetic (BSSS), Reid Yellow Dent (RYD), and Lancaster Sure Crop (LSC). Lines from different heterotic groups had a slightly greater RD mean than unrelated lines from the same heterotic group, yet differences were small when compared with the wide range of RDs for individual lines combinations within each group. RDs between related lines agreed well with expectations based on coancestry coefficients determined from pedigree data with few exceptions. Principal component analyses of RFLP data resulted in a separate grouping of lines from BSSS/RYD and LSC. Dispersion of lines of miscellaneous origins was generally consistent with expectations based on known breeding behavior and pedigrees. Results from this study suggest that RFLP data can be used for assigning inbreds into heterotic groups and quantifying genetic similarity between related lines, but it seems that a large number of probe-enzyme combinations are required to obtain reliable estimates of genetic distance.