In previous studies polygene mapping of a backcross population derived from haploid potato (Solanum tuberosum) and a diploid wild species (Solanum berthaultiij showed at least eight quantitative trait loci (QTLs) associated with tuber dormancy. The same population was mapped for abscisic acid (ABA) content in tubers so that any QTLs identified could be compared with those detected previously. At least three distinct loci on three chromosomes (2, 4, and 7) were associated with variation in ABA content. One of the QTLs was detected only as a main (single locus) effect, and two QTLs were found through two-locus interaction analysis (epistasis). Interaction between QTLs at markers TG234 (chromosome 2) and TG755 (chromosome 4) explained 20% of total phenotypic variante for this trait. The interaction closely resembled one previously detected for dormancy, suggesting an association between high ABA content and long tuber dormancy. Although relationships between ABA leve1 and dormancy could be demonstrated through polygene mapping, there was no indication of a relationship between these traits when they were subjected to a conventional correlation test. This illustrates the usefulness of polygene mapping as a tool to identify possible associations between hormone levels and plant development.A well-tested approach for understanding growth and development is the analysis of differences between two phenotypes that contain different alleles of a single major gene. For example, analyses of dwarf and wild-type siblings have revealed hormonal differences associated with the mutant genotype (Reid and Howell, 1995). Although this approach has been productive, a limitation is that the control of many developmental processes is polygenic, i.e. inheritance is quantitative rather than qualitative. Thus, although it is interesting to know what physiological effect was produced by a dwarf mutation, it might be much more informative to learn about the physiological effects pro- duced by the polygenes that control plant height. In the past the study of quantitative inheritance was so difficult that it yielded only very generalized information, and little could be learned that would elucidate the physiology of growth and development. The situation has changed dramatically with the availability of new methods for mapping polygenes (Tanksley et al., 1989;Tanksley, 1993;Young, 1993). The new mapping methods rely on DNA-based genetic markers, such as RFLP markers. Through such mapping, a number of QTLs may be found that are associated with the control of a trait such as plant height. A QTL denotes a region of chromosome linked to the marker gene that has a significant effect on the quantitative trait (Tanksley, 1993).With the availability of this tool for mapping quantitative traits, polygenes can be used to study the physiology of growth and development in a manner analogous to, but more powerful than, the way single genes have been used in the past. Once the QTLs have been detected that are associated with the control of a given stage of...