FIFTEENTH ANNUAL YEAR BOOK— PART VIIT. 575 



for the waspling (Oxybelus, r). Such a wonderful variety of habits as 

 we find among the wasps, and such a diversity of meats as are used by 

 their young! Yet the parent wasps all find a common meeting place on 

 the parsnip umbel, where they may satisfy their longing for sweets. The 

 parsnip is a cafe at which the ants also are most welcome, though their 

 limitations exclude them from more stylish places. Here we found an 

 ichneumon fly whose young is a parasite feeding upon a living insect 

 grub. Ten kinds of flies are fed here. There is a small butterfly, the 

 larger members of the tribe being where the picking is better. Here 

 are also five beetles, some of which, the lady beetles, are after an insect 

 diet. 



The third class of flowers, which we have been considering, falls into 

 two natural subdivisions; those which make an appeal to some of the 

 larger insects, honey bees at least, as well as to insects which are re- 

 stricted to exposed nectar, and those having as their attendants only 

 the smallest insects, such as the little mining bees, the ants and the 

 small flies. I think that the former secrete more nectar, and that more 

 is required to attract a bee than to attract a wasp, moth or fly. A hint 

 that such is the case comes from the study of buckwheat, which is vis- 

 ited by the bees only during the early forenoon hours, but by these other 

 forms all day, even after the bulk of the nectar has been removed. It 

 is also noticeable that generally those flowers of the subclass which the 

 bees visit are in more conspicuous clusters, while the others are more 

 scattered. Doubtless the massing of color has not a little to do with 

 their attraction for bees. 



To the first subdivision belong along with our parsnip, the basswood, 

 the elder, the buckwheat, the heartsease, the dandelion, the goldenrod. 

 and most of the blossoms of our fruit trees and plants. In the second, 

 those that scarcely make an appeal to honey bees, we find yellow hop 

 clover, purslain, wild lettuce, buckhorn, mustards, yarrow and others. 



As yet our work has given us no experimental knowledge of the role 

 of wild bees in pollination. There can be no question, though, as to 

 the importance of insects, especially bees, in the pollination of flowers, 

 and there is no set of correlations in nature nicer than that between 

 the varying types of flowers and the varying types of insects. 



I shall give an illustration to show how amply able are the bees to 

 take care of the details of pollination. On July 14, 1914, I kept under 

 close observation for a day a clump of white sweet clover of about three 

 by twenty feet and about three feet high. There were about 4,000 spikes 

 of 20 flowers each, or 80,000 flowers in the patch. About 55 honey bees, 

 on an average, put in nine hours work here. Each bee visits 40 flowers 

 per minute. Hence there were, during the day, 55x9x40x60 or 1,188,000 

 visits of bees to these flowers — an average of fifteen to a flower. Of wild 

 insects at least five more visited each flower, giving each twenty insect 

 visits in all. This ought to be amply sufficient to pollinate every flower 

 in the clump, and that such visits are necessary, was shown by cover- 

 ing a few of the buds with muslin to exclude insects, and noticing that 

 these buds failed to mature seeds. 



