In southeastern Utah, Bromus tectorum occurs where Hilaria jamesii is dominant and rarely where Stipa hymenoides/S. comata dominate. To determine whether this distribution is due to soil characteristics or microhabitat, we transplanted H. jamesii soil to a Stipa site and vice versa during a severe drought (2001) and a wetter year (2002). Additionally, we planted B. tectorum under H. jamesii and Stipa canopies, with or without H. jamesii litter, and with or without herbivory. Bromus tectorum emergence and biomass in reciprocal transplants were similar at both sites; there were no site differences for all microhabitat treatments. Being under a plant canopy increased emergence in 2001 and decreased survival during 2002. Herbivory decreased emergence in 2001 and decreased survival during 2002. Litter increased emergence only under the canopy in 2001 but did not affect survival in 2002. Survival in 2001 was so low that biomass was unattainable; no microhabitat treatments affected biomass in 2002. We found that soil characteristics and microhabitat affected B. tectorum similarly in H. jamesii and Stipa patches, suggesting that these factors do not explain the association between B. tectorum and H. jamesii. However, these relationships may change during wet years when B. tectorum invasions most often occur. ?? 2006 Elsevier Ltd. All rights reserved.