1893.] NATURAL SCIENCES OF PHILADELPHIA. 295 



one stage of growth, or at one season, from another, I must con- 

 clude that it is fertilized before expansion, and that insects in no 

 way assist the operation. 



Early in October I noticed a plant, five feet in diameter, that 

 had evidently been blooming for months previously. It had many 

 hundreds of capsules, and there were no indications that a single 

 flower had failed to be fertile. With my own proposition before me 

 that perfect fertility indicates close fertilization, I was led to place 

 the plant under close observation many times a day for a week. 



During all this time I did not notice a single insect visitor. 

 Watching the plant a few times by night, no nocturnal visitors were 

 seen. At seasons more pregnant with insect life the plant would no 

 doubt be visited, for no flower is slighted in case of necessity by 

 honey or pollen gatherers. There were none at this time, but the 

 seeding was still going on, and continued till the first severe frost in 

 December. It can be surely stated as a matter of fact, that there 

 were no insect visitors during the last half of the season, and yet 

 the plant was abundantly and unfailingly fertile. 



An interesting result of my close observation was the discovery 

 of three great periods of rest and of growth-resumption during anthe- 

 sis. In the early morning of one day is seen the folded pink corolla 

 in the position of a small cone peeping through the green calyx. 

 Commencing watch at this stage, growth progresses till about noon, 

 when there is a rest till the next morning ; again there would be the 

 same period of advance to rest, and the third day again an advance, 

 followed by the withering of the floral organs. 



Returning to the first period, the unopened bud, dissected at 8 

 A. M., showed no indications of pollen having been discharged. 

 By 10 A. M., the flowers were still tightly closed, but the anther 

 sacs had become ruptured, and the pretty purple styles, nestling in 

 the center of the mass of stamens, were completely covered by pol- 

 len. By 11 A. M., the corolla had so far untwisted that a slight 

 glance of the stamens and styles could be had in some flowers. By 

 12, noon, the corolla had taken on a narrow campanulate form, as 

 wide at the bottom as at the top. By 1 P. M., it had again closed, 

 so that the flower was expanded but about an hour only during the 

 whole period of its existence. If it were desirable, therefore, for 

 any insect to aid it in fertilization it would have but an hour at 

 noon in which to work ; but as we have already seen, the pollen had 

 reached the stigmas a short time before. 



