240 PRINCIPLES AND CONCLUSIONS. 



dained imitations placed in the midst of dahlias, but visited artefacts 

 of Petunia and Hieracium. Moreover, it proved possible to train bees to 

 come to artificial flowers and ignore the dahlias. 



Reeker (1898) found that artificial flowers were visited so regularly and 

 frequently when placed at 6 to 12 meters from the natural ones that he 

 found it undesirable to carry his experiments further, and Schroder (1901) 

 induced Syritta to visit them almost as abundantly as the normal by the 

 addition of honey. In an admirable series of experiments that rivaled those 

 of Plateau, Andreae (1903) obtained numerous visits to artificial flowers, 

 in one instance 81 visits being made to the latter and but 56 to the normal. 

 In the competition between color and odor as represented by artificial 

 flowers on one hand and masked natural ones on the other, the results were 

 uniformly in favor of the former, as shown by such ratios as 55 : 1, 41 : 0, 31 : 1, 

 and 28:3. Wery's results (1904) were equally conclusive, the artificial 

 flowers often equaling or exceeding the normal ones in attractiveness and 

 regularly surpassing honey in this respect. They attracted honey-bees 

 readily and to the same degree as normal flowers placed in a globe. Lovell 

 (1912) employed dry yellow immortelles in competition with honey, and 

 obtained three times as many visits to them as to the latter. Allard (1911) 

 found that the placing of cloth petals over the natural ones of cotton usually 

 decreased inspections, but a crepe-paper blossom of the proper color received 

 3 inspections to 8 for the normal. Detto (1905) also found that replacing 

 the corolla with one of yellow paper stopped visits, which began again as 

 soon as the petals were replaced. On the other hand, flowers with a ring 

 of colored paper were readily visited. 



In the present experiments artificial flowers were visited about an eighth 

 as much as natural ones. In more than half the installations they were 

 completely ignored or visited but little. In several instances the total 

 number of visits to imitations was large, though the ratio rarely reached 

 one-half. The highest relative number of visits was in the case of a Frasera 

 supplied with Campanula petals, namely, 56:97, but this was doubtless 

 owing to the naturalness of the living tissue. The fact that Bombus went 

 readily to paper Mentzelias before the natural ones were open appears to 

 prove that the paper flowers were not repellent in themselves, but that the 

 habit response to the natural flowers was a very definite one, as would be 

 expected. While our results are more nearly in accord with those of Plateau 

 and Forel, there is sufficient variation in them with respect to the various 

 species of flowers and insects, as indeed there is in those of these two investi- 

 gators, to make it unnecessary to question the observations of Andreae, 

 Wery, and others. It has been repeatedly demonstrated that insects are not 

 repelled by paper or cloth flowers or by composite imitations, and it is no 

 longer possible to uphold Plateau's contention that insects do not go to arti- 

 ficial blossoms because odor is more powerful than color in attraction. In the 

 major number of investigations they have been found to visit such flowers 

 abundantly and repeatedly, and to prefer them all but exclusively to honey 

 or odor alone, as Andreae and Wery in particular have shown. 



Painted flowers. — While painted flowers are artificial in a sense, this 

 is true to a much smaller degree than of paper or cloth ones, as is shown 

 by the relative number of visits. They have been modified as to color 



