D-2 



and recorded the total number of flowers, the number that had been 

 predated and the number that had matured fruit. I recognized mature fruit 

 by the presence of a large, swollen ovary. I collected this same 

 information for 25 randomly selected controls (unbagged plants) growing in 

 the same area. Inflorescences of both the bagged plants and the controls 

 were placed in paper bags to allow complete ripening of fruit and an 

 estimation of seed production. I dissected and examined flowers at 

 different stages after anthesis in order to determine the relative 

 ripening times of anthers and stigmata. 



On numerous occasions during the course of the study I observed 

 caterpillars (presumably lepidopteran larvae) feeding on the flowers of 

 Spalding's catchfly. These 2 cm-long larvae enter the base of a flower 

 and consume the ovary and other flower parts and then move to another 

 flower on the same inflorescence. I believe that these larvae are 

 responsible for most or all of the flcwer predation recorded in this 

 study. I attempted to remove all of these predators from the experimental 

 plants at the beginning of the study. 



KESLTLirS AND DISUSSICW 



Presumably due to the inclement weather, I observed only three 

 pollination episodes during the three days of the study. On these 

 occasions I observed bumblebees ( Bombus sp.) sequentially enter the 

 f lowers of at least three plants of Spalding's catchfly. I did not 

 observe any other flying insects visiting the flcwers. I collected one of 

 these bees and returned it to the laboratory. Pollen collected from the 

 body of this bee matched well with pollen taken from herbarium specimens 

 of Spalding's catchfly. These resiiLts suggest that bumblebees can be 

 effective pollinators of Spalding's catchfly. Further studies during 

 periods of good weather are needed to confirm and extend these results. 



Examination of numerous f lowers indicates that Spalding's catchfly is 

 protandrous. Anthers mature and dehisce pollen first. After which the 

 styles expand in length, and the stigmas spread apart and beccme 

 receptive. 



The nylon mesh bags of seven of the 21 ejqjerimental plants were chewed 

 open during the course of the study. I found the exoskeleton of a 

 grasshopper in one of the opened bags, and I suspect that they were 

 responsible for the damage. Since the inflorescences of these plants were 

 open to pollinators for part of the ejqseriment, they have not been 

 included in the following analyses. 



Fruit production data for the remaining 14 ej^)erimental plants and the 

 25 controls are presented in Table 1. There were a total of 161 flowers 

 on the 25 control plants. Of these, 40 (25%) were lost to predation. Of 

 the remaining 121 flowers, 91 (75%) produced mature fruit. There were a 

 total of 77 flowers produced by the experimental plants. Of these, 3 (4%) 

 were lost to predation. Of the remaining 74 flowers, 12 (17%) developed 

 mature fruits. Only one or two of the "mature" fruits collected from the 

 experimental plants actually contained ripened seed after they had been 

 stored in a paper bag for 1 month. Mature fruits from control plants 

 contained copious seed. These results suggest that Spalding's catchfly is 



