PURSLANE FAMILY 79 



blossom to plunder it of pollen and nectar, but on 

 account of the closed stigmas they cannot fertilize 

 the ovaries of the pistil, either with the pollen of 

 the same or that from another flower. On the 

 second day the filaments have bent outward in 

 such a way as to press the anthers against the 

 petals, and thus to keep them away from the stig- 

 mas which have now opened. If at this time the 

 flower is visited by a bee more or less covered 

 with pollen from another plant, some of the 

 pollen-grains will pretty surely be brushed upon 

 the stigmatic surface, and in consequence cross- 

 fertilization will result. The statement has fre- 

 quently been made that the closing of the petals 

 must bring about self-fertilization, but this in 

 general is not the case. Mr. Charles Robertson, 

 who has studied with his usual care the pollina- 

 tion of the Virginia Spring Beauty, has said : 

 " If self-fertilization by closing of the floAver 

 occurs, it is after the anthers have been exposed 

 to insects for two days and the stigma for one, 

 but many flowers which I marked exposed their 

 stigmas again on the third day, showing that 

 fertilization of any kind had failed on the day 

 before." My own observations on the Carolina 

 Spring Beauty indicate a similar behavior of the 

 flowers of this species. 



The blossoms of the Spring Beauty are visited 

 by an extraordinary number and variety of in- 



