Removal of corn (Zea mays L.) stover for biofuel production may affect crop yields by altering soil properties. A partial stover removal may be feasible, but information on appropriate rates of removal is unavailable. We assessed the short-term impacts of stover management on long-term no-till (NT) continuous corn grown on a Rayne silt loam (fine loamy, mixed, active, mesic Typic Hapludults) at Coshocton, Hoytville clay loam (fine, illitic, mesic Mollic Epiaqualfs) at Hoytville, and Celina silt loam (fine, mixed, active, mesic Aquic Hapludalfs) at South Charleston in Ohio, and predicted corn yield from soil properties using principal component analysis (PCA). The study was conducted in 2005 on the ongoing experiments started in May 2004 under 0 (T0), 25 (T25), 50 (T50), 75 (T75), 100 (T100), and 200 (T200)% of stover corresponding to 0, 1.25, 2.50, 3.75, 5.00, and 10.00 Mg ha−1 of stover, respectively. Stover removal promoted early emergence and rapid seedling growth (P < 0.01). Early-emerging plants grew taller than late-emerging plants up to about 50 d, and then the heights reversed at Coshocton and were comparable at other two sites. Stover management affected corn yield only at the Coshocton site where average grain and stover yields in the T200, T100, T75, and T50 (10.8 and 10.3 Mg ha−1) were higher than those in the T0 and T25 treatments (8.5 and 6.5 Mg ha−1) (P < 0.01), showing that stover removal at rates as low as 50% (2.5 Mg ha−1) decreased crop yields. Soil properties explained 71% of the variability in grain yield and 33% of the variability in stover yield for the Coshocton site. Seventeen months after the start of the experiment, effects of stover management on corn yield and soil properties were site-specific.