Since their discovery in the 1980s microsatellite or simple sequence repeat (SSR) markers have been widely used in many species to generate relatively dense genetic maps or framework maps on which to anchor more abundant, but anonymous, markers such as amplified fragment length polymorphisms (AFLPs). They are typically highly polymorphic, robust, and often portable, particularly among different mapping populations or crosses and often to related species. They have been useful in species where low levels of genetic diversity limit the use of other markers. Cultivated cotton (Gossypium hirsutum L.) has a history of genetic bottlenecks that have considerably reduced its diversity, with the consequence that most molecular marker genetic linkage studies are done with inter-specific crosses. In this study we evaluated the potential for SSR markers to be used in marker-assisted selection (MAS) breeding in cotton by quantifying the level of polymorphism detected with a set of commercially available SSR markers between and within a collection of cotton cultivars being used in our breeding programs. Although the majority of these markers are polymorphic between the 2 tetraploid species of cotton, G. barbadense and G. hirsutum, they are not highly polymorphic (~5%) either among or within G. hirsutum cultivars. However, 6 of the 8 cultivars studied were found to be segregating for alleles of these SSR markers. This suggests that where polymorphisms exist, heterozygosity within cultivars is maintained by the breeding strategies adopted by many modern cotton breeders. Although SSRs clearly have utility in genetic studies using inter-specific crosses or in the introgression of wild germplasm, they will be more difficult to use for standard cotton breeding until greater numbers are available. The utility of some markers may be reduced in some breeding populations where heterozygosity remains in the parental material.