180 PRINCIPLES AND CONCLUSIONS. 



received 38 visits, 25 by bees, the inverted beaker 17, and the four "odor" 

 glasses but one, giving a ratio of 55 : 1 in favor of color. Even when the 

 artificial primrose was placed 5 meters from the flower-bed, it received 13 

 visits from bees to none for the more accessible "odor" glasses. To deter- 

 mine the effect of a difference in form the artificial primrose was re- 

 placed by a peony of the same material and color, which was placed 7 

 meters from the bed of primroses. One meter from the latter stood four 

 glasses, one filled with honey, the other three faintly perfumed with jas- 

 mine, mignonette, and violet respectively. However, the latter obtained 

 no visits in contrast to 41 for the imitation flower. 



Twelve meters from a bush of Doronicum caucasicum, a yellow composite 

 actively sought by various insects, was located an artificial flower-bed 

 composed of the primrose and peony already used and an orange-red and a 

 dark-red poppy, all of which were a meter apart. As an additional attrac- 

 tion, artificial cornflowers and snowballs were placed in the center of the 

 square. The yellow peony received 31 visits, the orange-red poppy 9, 

 the dark-red one 3, the central group 7, and the primrose but 1, showing 

 a clear relation between conspicuousness and attractiveness. A further 

 test of the effect of odor was made by filling an inverted beaker with fresh 

 primroses, so that they were plainly visible, while a bell-glass was likewise 

 filled with several hundred fragrant flowers of the same species, but these 

 rendered invisible by means of dark paper. In spite of the advantages 

 given the latter, it received but 3 visits to 35 for the former. 



When 5 funnels were filled with fragrant flowers of the peony and sur- 

 rounded with paper, and placed above the bush, now past blooming, they 

 received but 3 visits in competition with a bell-glass in which several 

 flowers were visible, this yielding 40 visits. To test the effect of the glass 

 itself, the same bell-glass full of peonies was placed 15 meters away, while 

 a small empty one was put but a meter distant. The latter received no 

 visits to 5 for the former. In the next trial two artificial flowers were made 

 of bright-yellow and rose-red tissue paper, provided with dark-green below 

 to furnish a contrast, and attached to stakes 2 meters high. Within a 

 period of 2 hours 17 visitors went to one and 16 to the other, the total of 33 

 including 12 bees. When a yellow, a dark-yellow, and a purple-red cloth 

 flower were employed, the yellow obtained 21 visits and the red 9, during 

 3 hours. In the next case, 3 flowers of Papaver orientalis were placed in a 

 funnel so that the odor could escape without their being seen, while to the 

 single normal flower left on the plant was added a larger artificial one of 

 cloth, in order to determine whether conspicuous color, color and odor, 

 or the latter alone would be most effective. The results showed 81 visits 

 to the imitation, chiefly by Osmia and Apis, 56 to the normal flower, and 

 none to the funnels. In the control the last normal flower was added to 

 those in the funnel and this placed 5 meters distant against the wind. In 

 spite of this it received no notice from the bees, which meanwhile made 

 43 visits to the imitation. 



Experiments with flowers of dull color but nectar-bearing showed that 

 the color was often attractive, sometimes more attractive than the odor 

 of nectar or honey. Studies were also made of flowers with dull color and 

 marked perfume, such as Reseda, Dahlia, etc. When stems of the former 



