In 1939 an experiment was established on the Jornada Experimental Range to evaluate the effects of brush removal, rabbit exclusion, furrowing, and seeding in creosotebush. Sixteen plots (21.3 x 21.3 m) were established in four rows of four with a buffer zone of 7.6 m between plots. A barbed wire fence excluded cattle and poultry wire fencing excluded lagomorphs. The treatments were factorially applied at two levels, yielding a check plot, 4 single-factor plots, and 11 plots with varying degrees of interaction. In 1989 the plots were sampled for shrub canopy cover and grass basal area with randomly located, line-intercept transects. The noneffective furrowing and seedling treatments were excluded from analyses. Multivariate analysis of variance was conducted by plant groups (shrubs, subshrubs, and grasses) to determine the effect of shrub removal and lagomorph exclusion. Grass basal area was higher (P = 0.026) on cleared (4.8%) than on noncleared plots (4.0%). Grass basal area on lagomorph-excluded plots (4.3%) was higher (P = 0.018) than on plots grazed by lagomorphs (3.7%). Subshrub canopy cover was different (P = 0.029) on lagomorph-excluded and lagomorph-grazed plots with different species showing opposite responses. These results indicate that lagomorphs have a profound influence on the species composition of creosotebush vegetation.