A field trial was conducted for 6 yr (1998 through 2003) at Scottsbluff, NE, to measure weed shifts following multiple applications of two rates of glyphosate or alternating glyphosate with nonglyphosate treatments in continuous corn or in a crop rotation of corn, sugarbeet, and spring wheat with all three crops resistant to glyphosate. After 6 yr, plant densities of common lambsquarters, redroot pigweed, hairy nightshade, and common purslane increased in the crop-rotation treatment compared with continuous corn. There were four weed control subplot treatments consisting of two in-crop applications of glyphosate at 0.4 or 0.8 kg ae/ha each spring, alternating two applications of glyphosate at 0.8 kg/ha one year with a nonglyphosate treatment the next year, or a nonglyphosate treatment each year. The composition of the weed population averaged across all four treatments shifted from kochia and wild proso millet to predominately common lambsquarters. After 3 yr of using glyphosate at 0.4 kg/ha twice each year, common lambsquarters density increased compared with that in the 0.8 kg/ha rate of glyphosate or alternating glyphosate treatments. By the sixth year, the density of common lambsquarters in the glyphosate at 0.4 kg/ha treatment had increased to the extent that corn grain yield was reduced 43% compared with corn grain yield in the 0.8 kg/ha glyphosate treatment. Using glyphosate at either rate for 6 yr decreased the densities of kochia, wild proso millet, and longspine sandbur, did not alter densities of redroot pigweed and green foxtail, and increased the density of hairy nightshade. In the low-rate treatment of glyphosate, the number of common lambsquarters seeds in the seed bank were 134 seeds/kg soil in 1998, declined to 15 seeds/kg by 2002, but began to increase in 2003 as the densities of plants not controlled by glyphosate increased.