Density and fecundity data for plants of A. fecunda in 

 removal and control plots are presented in Table 4. In general, 

 these data show the same trends as found in the long-term 

 monitoring ^ study: a drastic reduction in number and percent of 

 plants fruiting and an increase in the average number of fruits 

 per inflorescence and per fruiting plant in 1988 (see above for 

 discussion) . Density of A. fecunda plants at Charleys Gulch 

 showed no increase in 1988 in either the control plots or the 

 removal plots. At Birch Creek, while control plots showed no 

 appreciable increase in density of A. fecunda plants in 1988, 

 there was a more than three-fold increase in the removal plots. 

 The density of A. fecunda in individual removal and control plots 

 in 1987 and 1988 at Birch Creek are presented in Table 5. 

 Between 1987 and 1988, density decreased slightly in all but one 

 of the control plots, while density increased in all but one of 

 the knapweed removal plots. Our notes from 1987 indicate that 

 there were large numbers of seedlings in plots 8, 9, and 10. 

 Many of these survived in the removal plots (plots 8 and 9) , but 

 few survived in the control plot (plot 10) . These results 

 suggests that A. fecunda recruitment may be curtailed by the 

 presence of knapweed. 



Soil Crust Ecology Study 



In this study, the null hypothesis is that within the three 

 belt transects Arabis fecunda plants are distributed at random, 

 i.e., without respect to the presence of soil crusts. The 

 results of a chi-square analysis of the data are presented in 

 Table 6. The null hypothesis was strongly rejected in all three 

 cases, indicating that the distribution of A. fecunda is not 

 random. The data show that A. fecunda is associated more often 

 with soil crusts than bare soil within the transects. There are 

 two possible explanations for these results: (1) A. fecunda is 

 able to establish and survive better in soil crusts, and/or (2) 

 A. fecunda is able to establish and survive with equal success in 

 bare soil and soil crusts; however, perturbations caused by 

 livestock destroy not only the A. fecunda plants growing in bare 

 soil and soil crust, but also the soil crust itself. This would 

 result in an increase in the amount of bare soil without A. 

 fecunda . These two explanations are not mutually exclusive. 



If the first explanation is correct, A. fecunda may be 

 dependent on the presence of soil crusts to maintain viable 

 population levels within these transects. Grazing by livestock 

 has been shown to reduce the cover of soil crusts (St. Clair et 

 al. 1984), thus reducing the availability of microsites which 

 are important for seedling establishment and survival. If the 

 second explanation is correct, livestock grazing is destroying 

 A. fecunda plants in these transects regardless of whether they 

 are rooted in bare soil or soil crusts. Under either 

 interpretation, our results indicate that livestock grazing is 

 detrimental to populations of A. fecunda on the steep, highly 

 erodible slopes in these transects. 



