Creosotebush (Larrea tridentata (DC) Cov.) dominates much of the formerly productive rangelands of the southwest. A successful and practical method of eradicating this shrub, followed by revegetation of the area by desirable forage plants, would greatly improve these creosotebush-infested areas for livestock use. The purpose of this factorially designed experiment was to study the effects of seeding, furrowing, brush removal, and rabbit exclusion on three creosotebush-infested sites on the Jornada Experimental Range, Las Cruces, New Mexico. Brush removal consistently decreased the canopy cover of creosotebush and total shrubs, regardless of other treatments. The brush removal alone or the brush removal x rabbit exclusion treatment significantly increased the grass species at the |Ragged and Parker Tank sites. The rabbit exclusion fence was present in the treatment applications that significantly increased the grass cover at the Dona Ana site. This treatment was significant in increasing the grass cover due to its soil retaining and soil depositional effects on this site with the steeper slope and the more severe erosion and runoff condition.