> 



8 



three components (p<0.001), while interaction effects were not 

 significant (p>0.130). 



Discussion 



Our results clearly demonstrate a strong positive 

 association between the presence of cryptogamic soil crust and 

 the occurrence of Arabis fecunda plants. There are two possible 

 explanations for this phenomenon: (1) A^ fecunda is able to 

 establish and/or survive better in soil crust at the three sites 

 and (2) A^ fecunda establishes and survives with equal success in 

 bare soil and in soil crust, but disturbances destroy soil crust 

 and the plants growing in it, resulting in an increase in the 

 amount of bare soil without A_^ fecunda in proportion to the 

 amount of crust with plants. These two hypotheses are not 

 mutually exclusive. However, the association between crust and 

 A. fecunda was as strong at Lime Gulch as at the Birch Creek or 

 Charley's Gulch, even though Lime Gulch former had much lower 

 levels of disturbance. Thus, it is unlikely that disturbance can 

 be an important cause of the A^ fecunda -crust correlation. 



At both Charley's Gulch and Lime Gulch North there was no 

 evidence of differential survival of a life stage class between 

 soil crust and bare soil. In these transects it appears that 

 seedling establishment and survival of older plants are favored 



