Although it is widely assumed that genetically diverse parents facilitate the creation of superior progeny, few studies have examined the relationship between parental genetic distance and the creation of successful cultivars. In theory, matings of distantly-related parents will produce a greater number of transgressive segregates than matings of closely-related parents. However, for many crops, yield improvements have come from matings of closely-related genotypes. The objective of this study was to determine the relationship between parental genetic distance and development of successful cotton (Gossypium hirsutum L.) cultivars. Pedigrees of cultivars that occupied more than 1% of the total U.S. plantings for at least 1 yr from 1987 to 1996 were examined, and the genetic relatedness of the parents in the final cross was estimated as coefficient of parentage. A total of 60 successful cultivars was identified. The majority of these cultivars was derived from two-way crosses (60%), whereas the remainder were the result of reselections within cultivars or germplasm lines (25%), or complex crosses (15%). The average coefficient of parentage between parents used in the final cross was 0.29. This value was greater than was expected with random pairing of parents (coefficient of parentage = 0.09). In general, the genetic distance for parents of successful cultivars reflected the level of diversity within the regionally-adapted cultivars. Although many germplasm lines were available to breeders, use of diverse germplasm appeared to be restricted to a few agronomically-suitable lines. Of the 668 cotton germplasm lines registered in Crop Science during 1972 to 1996, only four (0.03%) appeared in the pedigrees of successful cultivars. The high frequency of successful cultivars derived from reselections, and the occurrence of several widely-grown cultivars that were developed from closely-related parents, indicated that genetically diverse parents have not been imperative to cotton cultivar improvement in recent years.