I found that some plants would go undetected for one to 

 several years but reappear in subsequent years. These 

 "underground" plants may have produced small leaves that had 

 senesced and disappeared by early July; however, my observations 

 in May and June suggest that most of them produced no vegetation 

 on the years in question. The presence of underground plants can 

 be inferred by comparing transect maps from the full sequence of 

 years. Underground plants were placed in the small size class 

 for analysis. The proportion of underground plants ranged from 

 1-23% with a mean of 10% in 1987-91. Plants have "disappeared" 

 for as many as five years before reappearing. However, in 1986- 

 92 at the two sites, 71% of the underground plants reappeared 

 after one year, and 88% reappeared after two years. As a result, 

 ca. 10% of the plants were undetected in the first and last of 

 years of the study, 3% were undetected in second and second from 

 last years, and ca. 1% were undetected on other years. Thus, I 

 have chosen to eliminate the first and last years of the study 

 from demographic analysis. 



On years when fruit production was adequate, I collected 50 

 randomly selected mature fruits from at least 25 plants. I 

 opened the pods, counted the intact seeds, and recorded whether 

 the fruit contained weevil larvae. 



Data Analysis 



Population growth for year t is the percent change in the 

 size of the sample population between year t-1 and year t. It is 

 calculated by PG = N^ - N^ ,/N^.,. Mortality rate is the ratio of 

 the number of plants dying between years t-1 and t to the number 

 surviving in the same period. Recruitment rate is defined as the 

 ratio of new plants observed in year t to the number of plants 

 surviving from year t-1 to year t. 



I compared survival rates of* uneven-age cohorts present at 

 the start of the study between the sites using the nonparametric 

 logrank test (Pyke and Thompson 1986, Hutchings et al. 1991). 

 Survivorship curves were constructed following methods outlined 

 in Hutchings et al. (1991). 



Fecundity is the number of fruits per plant. The effects of 

 site and year on fecundity were analyzed using analysis of 

 variance (ANOVA) . The interaction term was not significant 

 (£=0.7, P=0.56) and was deleted from the final model. The 

 dependent variable was log-transformed before analysis to conform 

 to the assumptions of the test. 



I used correlation analysis to explore relationships between 

 climate variables and Astragalus scaphoides mortality, 

 recruitment and proportion of reproductive plants at Sheep Corral 

 Gulch. Climate data are from Dillon, the closest recording 

 station. Mean monthly deviation from the 30-year normal for 



