Insects and Flowers 571 



the Umbellifcrac and the numerous Compositx'. In the umbels and flower- 

 heads, often rather inconspicuous but nearly always well provided with 

 nectar, the sweet drink is easily got at even by short-tongued insects, so 

 that some of the species have a surprising host of visitors. For example, 

 Robertson found 275 different insect species visiting Pastinaca saliva (an 

 umbellifer with exposed nectar) in the neighborhood of Carlinville, 111.; 

 238 visiting Cicuta maciilata, and 191 visiting Siiim ciciilajolium; observing 

 some of the composites, more specialized, Robertson noted 146 insect species 

 at goldenrod {Solidago canadensis) in eleven days during August, September, 

 and October, and 100 at Aster paniculatus in four days in October. 



Of course not all the insect visitors to a flower are cross-pollinating agents; 

 some are deliberate thieves, some may or may not help in cross-pollination, 

 and some are reliable, although, of course, unwitting, pollinators. As an 

 interesting test of the proportion of actual pollinators to the whole number 

 of insect visitors may be taken Robertson's observations on the milkweed 

 (Asclepias) and its visitors (see account of the conditions in Asclepias on 

 P- 573)- Of 115 insect species which visited flowers of Asclepias verticillata 

 (Carlinville, 111.) in fifteen days, representatives of 58 of these actually got 

 poUinia (pollen-masses) attached to themselves; while of 80 species visit- 

 ing A . incarnata in twenty-four days, 63 carried off pollinia. I do not know 

 of any other records which show the proportion of actual pollination to 

 total number of visitors, but it is highly desirable that such observations 

 be made for other flowers. Asclepias obviously offers a particularly favor- 

 able opportunity for such tests (on account of the conspicuousness of the 

 pollinia), but an ingenious observer will be able to study the matter success- 

 fully with other plants. 



With the flowers of tubular corolla the pollinating insects are of course 

 neither so many nor do they represent such varied insect groups. The 

 long-tongued bees and flies can get nectar from a flower-cup not too deep, 

 but in the deeper cups the moths and butterflies are the only insects which 

 can reach the nectar. The common jimson-weed. Datura stramonium, is, as 

 Stevens says, an excellent illustration of this. "The corolla is about five centi- 

 meters long, and the cavity of the tube is nearly closed at about the middle 

 of its length by the insertion of the filaments there. When the flower opens 

 in the evening it emits a strong musky odor, and a large drop of nectar is 

 already present in the bottom of the tube; so that large sphinx-moths, leav- 

 ing the places of seclusion occupied by them during the day, are attracted 

 by the strong odor and white color of the flowers. 



"Flying swiftly from flower to flower, the moth thrusts its long proboscis 

 to the bottom of the tube and secures the nectar; and while it is tarrying 

 briefly at each flower, keeping itself poised by the swift vibration of its wings, 

 it is pretty certain to touch with its proboscis both anthers and stigmas, 



