Long-term tillage effects on soil texture, structure and organic matter were investigated using a 9 tillage system experiment located at the Upper Piedmont Research Station in Reidsville, North Carolina. The systems were initiated in 1984 and vary with respect to timing (spring and fall) and method of soil disturbance (no-till, chisel, moldboard plow and disk). The systems are replicated 4 times in a randomized complete block design. The plots are ~ 0.01 ha and were planted to continuous corn (Zea mays L.) for the first decade and subsequently to alternating corn and soybeans (Glycine max L.). The entire experimental site is mapped as Wedowee fine, kaolinitic, thermic Typic Kanhapludult. Soil samples were collected (0-7.5 cm and 7.5-15 cm) in the fall of 2003 and 2004. All samples were analyzed for particle size distribution, particulate and silt+clay sized C (total and MnO₄ oxidizable), particulate and silt+clay sized N, and macroaggregate stability. The highest and lowest values for all parameters (0-7.5 cm) were associated with the most contrasting systems: continuous no-till (NT) and 2 moldboard plow/disk (CT) systems. Total C and N varied between 14.9 and 5.3 and 1.33 and 0.43 gkg^-1, respectively. MnO₄ oxidizable C varied between 0.83 and 0.17gkg^-1. Particulate C and N varied between 8.9 and 1.48 and 0.74 and 0.07 gkg^-1, respectively. Macroaggregate stability varied between 65.7 and 14.6%. Sand content varied between 77.5 % (NT) and 61.7% (CT). Differences between parameters for the 7.5-15 cm samples were smaller and in some cases insignificant. Covariate analysis with respect to silt+clay content improved discrimination between tillage systems.