This project addressed the need for economical technology for the conversion of lignocellulosic biomass to fuels, specifically the conversion of pretreated hardwood to ethanol. The technology developed is a set of strains of the bacterium Thermoanaerobacterium saccharolyticum and an associated fermentation process for pretreated hardwood. Tools for genetic engineering and analysis of the organism were developed, including a markerless mutation method, a complete genome sequence and a set of gene expression profiles that show the activity of its genes under a variety of conditions relevant to lignocellulose conversion. Improved strains were generated by selection and genetic engineering to be able to produce higher amounts of ethanol (up to 70 g/L) and to be able to better tolerate inhibitory compounds from pretreated hardwood. Analysis of these strains has generated useful insight into the genetic basis for desired properties of biofuel producing organisms. Fermentation conditions were tested and optimized to achieve ethanol production targets established in the original project proposal. The approach proposed was to add cellulase enzymes to the fermentation, a method called Simultaneous Saccharification and Fermentation (SSF). We had reason to think SSF would be an efficient approach because the optimal temperature and pH for the enzymes and bacterium are very close. Unfortunately, we discovered that commercially available cellulases are inactivated in thermophilic SSF by a combination of low redox potential and ethanol. Despite this, progress was made against the fermentation targets using bacterial cellulases. Thermoanaerobacterium saccharolyticum may still prove to be a commercially viable technology should cellulase enzyme issues be addressed. Moreover, the organism was demonstrated to produce ethanol at approximately theoretical yield from oligomeric hemicellulose extracts, an ability that may prove to be uniquely valuable in pretreatment configurations in which cellulose and hemicellulose are separated.