Late instar larvae of the army cutworm, Euxoa auxiliaris (Grote), infested replicated test plots of 43 wheat varieties in Hays, Kansas in early April, 2004. Five varieties, Trego, 2147, Betty, Thunderbolt, and Jagger, were selected for sampling cutworm defoliation and larval densities. Damage was lowest in Trego (12–41 damaged tillers/m of row), and highest in Betty (128–285 damaged tillers/m). Plots of Betty averaged 105 cutworms/m², significantly more than plots of other varieties where densities ranged from 13 to 42 cutworms/m². Plots of Thunderbolt sustained significantly more damage than plots of 2137, Trego or Jagger, but did not have significantly more larvae. The small size of the test plots appeared to facilitate aggregation of cutworms in plots of the preferred variety Betty. Grain yields of all varieties in 2004 were average for the region and the final grain yield of Betty plots averaged 20% less than that of Thunderbolt and Trego plots. This compares to an average of 10% lower yield for Betty compared to these cultivars in years without cutworm damage. Late instar cutworms offered equal amounts of Trego and Betty in Petri dishes in the laboratory consumed 10 times the leaf area of the latter variety over a 16 h period, supporting the inference that varietal differences in damage and cutworm densities reflected larval feeding preferences.