12 

 higher in treatment plots. Reproductive rates as measured by the 

 number of new plants per reproductive individual were also higher 

 in treatment plots at both sites for the first two years after 

 removal of C^_ maculosa , and recruitment rate over the period of 

 the study was significantly higher at both sites. With the 

 exception of the slight tendency for small plants to survive 

 better in treatment plots at Charley's Gulch, there is no 

 evidence that plants in any life-stage class performed better in 

 plots from which C^. maculosa had been removed. 



The presence of Centaurea maculosa could adversely affect 

 Arabis fecunda recruitment by (1) reducing fecundity and/or (2) 

 curtailing establishment of seedlings. We detected only a small 

 increase in A_j_ fecunda fecundity in the treatment plots during 

 the first two years following removal of Cj. maculosa . Although 

 this difference could be important, it is more likely that the 

 observed increase in A^. fecunda recruitment is due to enhanced 

 seedling establishment. Establishment of A^. fecunda seedlings 

 could have been inhibited by competition with C_j_ maculosa , but 

 this seems unlikely since C\_ maculosa did not interfere with 

 larger plants. Allelopathy could also explain our results, and 

 extracts of cnicin from Centaurea maculosa do inhibit germination 

 of grasses and conifers (Kelsey and Locken 1987) . However, 

 Kelsey and Bedunah (1989) concluded that concentrations of cnicin 

 high enough to cause interference are rarely present in natural 

 situations. Thus, it seems most likely that C\. maculosa reduces 



