that there may be genetic differentiation for demographic 

 parameters among populations of A^ fecunda. 



In many instances populations of Arabis fecunda at Birch 

 Creek and Vipond Park had a different response than the other two 

 populations to similar weather. In spite of this, our results 

 suggest that winter precipitation is an important determinant of 

 survivorship and recruitment. Low mortality often occurred in 

 years with wet winters, and low recruitment often occurred 

 following dry winters. The importance of winter precipitation 

 could be in providing a protective blanket of snow during the 

 coldest months or in providing ample early-season moisture. The 

 importance of winter precipitation is probably at least partially 

 responsible for the strong negative correlation between mortality 

 and recruitment. There was no correlation between fecundity and 

 recruitment or between fecundity and mortality (N=15, r<0.08). 

 Higher fecundity was associated with above normal fall and winter 

 temperatures and above normal fall precipitation. Since even in 

 a poor year a reproductive plant produces an average of at least 

 120 seeds, it seems unlikely that year-to-year fluctuations in 

 fecundity would have an important effect on population growth. 

 Thus, it appears that winter precipitation is the most important 

 weather variable affecting A^ fecunda population growth. 



MANAGEMENT IMPLICATIONS 



A recent study of Arabis fecunda using isozyme markers 

 suggests that there is little genetic differentiation among 

 populations (Leeper et al. 1993) . However, isozyme studies 

 detect differentiation of neutral genotypes that occurs over 

 relatively long periods of time. Genetic distance based on 

 isozyme markers is generally related to length of time since 

 population separation and degree of gene flow. Significant 

 selection-driven differentiation can occur without 

 differentiation among neutral alleles. Our study suggests the 

 existence of ecotypic differentiation among A^ fecunda 

 populations. If differentiation does exist, a conservation 

 strategy for this rare state endemic must strive to protect the 

 genetic variation present among the many populations. 



The Birch Creek and Vipond Park populations of Arabis 

 fecunda demonstrated relatively large fluctuations in population 

 size compared to those at Charleys Gulch and Lime Gulch. For 

 example, population growth decreased in all populations in 1991; 

 however, -the decrease was less than 5% at Charleys Gulch and Lime 

 Gulch but was 50-90% at Birch Creek and Vipond Park (Fig. 2) . 

 Populations subject to large fluctuations are generally thought 

 to be more prone to extinction (Diamond 1984, Pimm et al. 1988). 

 The Birch Creek and Vipond Park populations are large 

 (Schassberger 19??) , but smaller populations with the same 

 demographic behavior may be at risk if prolonged stress occurs. 



15 



