The conversion of the cellulose to monomeric sugar units is affected by lignin content and composition. Lignin shields the cellulose from the cellulolytic enzymes, and binds the enzymes irreversibly, reducing net enzyme activity. Other factors affecting the efficiency of the conversion are the soluble carbohydrate content and biomass yield. Sorghum offers advantages that make it an attractive biomass crop. The production of cellulosic ethanol can be improved significantly through the use of brown midrib (bmr) mutants, which have changes in cell wall composition. Twenty-eight bmr mutants are known, but the allelic relationships were never fully established, nor were the underlying genes identified. Through a combination of genetic and chemical approaches we established that there are four independent bmr loci. We have characterized the changes in lignin composition in each of the mutant, to help identify the genes underlying the mutations, and to study the effect of the changes in the conversion properties of the stover. Two bmr mutants increase the yield of fermentable sugars obtained from enzymatic saccharification of stover by 30-60%. We have previously shown that the mutants allelic to bmr12 have a defective COMT gene. Based on a combination of chemical data and genetic mapping we have identified the bmr6 gene. The knowledge of the basis of the traits enables expedited development of novel germplasm for bioenergy production. We are developing brown midrib sweet sorghum as a dual-source feedstock for ethanol production: the accumulation of soluble sugars in the stem can be used directly for fermentation, and the remaining stover can be used for the production of cellulosic ethanol. We are currently studying the effect of pyramiding the non-allelic bmr genes on plant fitness and cell wall composition, and we are using a population of recombinant inbred lines to map the quantitatively inherited sweet sorghum trait.