Before Anglo-Saxon settlement, the tallgrass prairie biome of North America was thought to contain some of the largest soil organic carbon (SOC) stocks in the continental U.S. However, the practices of cultivation and row crop agriculture, which now occur across >95% of this region, have dramatically changed the plant communities and C and N inputs to the soil. Recent changes in agricultural economics, coupled with growing interest in C sequestration, have led to grassland restoration in some areas. The objective of this study was to determine how quickly grassland soils recover following cessation of cultivation. We employed a chronosequence approach with paired land uses (continuous cultivated, different aged restored prairies, and native prairie) at 4 sites spread across the tallgrass prairie biome (Texas, Kansas, Iowa, and Minnesota). To better quantify the current and/or historic C stocks in native, uncultivated tallgrass prairie, we also sampled 44 remnant prairies across the region. One meter deep soil cores (n=4 per land use per site) for C & N pool size determinations and one composite surface soil sample for microbial analyses were collected in 2004 and 2005. Averaged across sites, SOC and total N pools from continuous cultivation were 26.8% and 21.8% (respectively) lower than native prairie; however, significant variation exists across land uses at individual sites. Grassland restoration has increased both C and N pools, but not in a consistent manner across time or space. Interestingly, despite altering nutrient pools, land use played little to no role in determining bacterial community structure, as determined by T-RFLP methodology. Instead, bacterial communities closely tracked latitudinal trends in mean annual temperature across the region. These results suggest that the factors controlling grassland restoration following cessation of cultivation differ for soil nutrient pools and bacterial communities.