Spring wheat (Triticum aestivum L.) production in the northern Great Plains generally utilizes conventional tillage systems. A 12-yr study evaluated the effects of tillage system [conventional-till (CT), minimum-till (MT), and no-till (NT)], N fertilizer rate (0, 22, and 45 kg N ha(-1)), and cultivar (Butte86 and Stoa) on spring wheat grain yields in a dryland spring wheat-fallow rotation (SW-F). Butte86 yields with CT exceeded NT yields in five out of 12 years with 0 and 22 kg N ha(-1) applied, and four years with 45 kg N ha(-1) applied. Stoa yields with CT exceeded NT yields in three out of 12 years with no N applied, four years with 22 kg N ha(-1) applied, and only one year with 45 kg N ha(-1) applied. Yields with NT exceeded those with CT in one year. Most years, yields with MT equaled those with CT. Responses to N tended to be greatest in years when spring soil NO3-N was lowest. Positive yield responses to N fertilization with CT occurred in three years with Butte86 and two years with Stoa; with MT, four years with Butte86 and two years with Stoa; and with NT, five years with Butte86 and three years with Stoa. Cultivars were not consistent in their response to tillage and N fertilization. These results indicate that farmers in the northern Great Plains can successfully produce spring wheat in a SW-F system using MT and NT systems, but yields may be slightly reduced when compared with CT systems some years.