14 

 populations will decrease and the risk of chance extinction 

 events will become more likely (Goodman 1987) . 



Herbicide application is currently the most commonly 

 employed method for controlling Centaurea maculosa infestations 

 (Harris and Cranston 1979, McKone et al. 1989); however, little 

 is known about specific effects on Arabis fecunda, and correct 

 rates of application are difficult to determine for novel 

 situations, such as the azonal A^. fecunda sites (Peter Rice, 

 pers. com.) . Removing C^ maculosa by hand would result in 

 serious damage to the fragile slope habitat. To date, biological 

 control vectors have been unable to reduce infestations of 

 Centaurea maculosa . Although introduced seed-head gall flies 

 ( Urophora af finis and Ul guadrifasciata ) can reduce seed 

 production from 50-75% (Story 1989), they have been unable to 

 reduce the density of C^ maculosa (Muller and Schroeder 1989) . 

 However, with the recent introduction of a root-boring moth 

 ( Aqapeta zoeqana ) , control of C^. maculosa should be possible 

 (Story 1989, Muller and Schroeder 1989). Since biological 

 control causes minimal disturbance and promises to have the 

 ability to halt population growth of C^. maculosa , it appears to 

 be the best option for protecting populations of Arabis fecunda 

 from extinction due to encroachment by Centaurea maculosa . 



