Through the generation of a large number of expressed sequence tags ongoing efforts will likely identify most genes in switchgrass and also will uncover a large segment of the molecular diversity in this species that is likely to include allelic variation in genes contributing to biomass production and other traits relating to its properties as a feedstock.  The EST data will also enable basic genetic and physiological studies focused on feedstock improvement by providing an inventory of genes that can be studied collectively or individually. In a relatively low input production system on marginal agricultural land, one approach to improving biomass yield in switchgrass is through release of improved cultivars.  Exploitation of large sequence data sets will hasten development of DNA marker-based molecular breeding approaches that will speed the process of cultivar development by allowing artificial selection at the seedling stage.  One way in which switchgrass EST sequences are being used is to help create a genetic map.  Availability of such a map based on molecular markers is fundamental for advanced breeding strategies.  Using a tetraploid, F1 mapping population, we are in the process of analyzing linkage and genome organization relative to other grasses in this large C4 representative of the Paniceae.  Based on segregation of multiallelic markers, direct genetic evidence supporting a high degree of preferential pairing has been obtained. This implies that the homeologous genomes are distinct and do not readily recombine.