RESUME. 93 



420 visits to 845 for the latter. In the case of Aconitum they obtained 

 154 visits to 116 for the normal flowers, though they were more numerous 

 in the installation. This was likewise true of Chamaenerium, the numbers 

 being 111 to 103, though on the basis of the number of each the relation 

 is 44 : 103. The painted flowers of Rubus deliciosus were visited as frequently 

 as the natural ones, while in the case of Frasera the visits were a half and in 

 that in Delphinium a fifth as many as for the natural. 



The outstanding visitor to the painted flowers was Bombus juxtus, which 

 made more than half the total number of visits. However, in two experi- 

 ments it was exceeded by the honey-bee, which made 27 such visits to its 12. 

 The 27 visits were in contrast to 67 for the normal flowers, thus showing 

 clearly how much more natural the painted flowers appeared than the paper 

 ones, since Apis did not visit a single one of the latter. Of the four colors 

 most employed, blue obtained a distinct preference, the visits being, blue 98, 

 green 84, yellow 55, and red 43. 



The painted flowers were approximately five times as successful in 

 attracting bees as the paper ones. This was due to their identity with the 

 natural ones in everything but color, and indicates that the artificial flowers 

 were distinguished as such chiefly by small differences in form and by the 

 texture, thus demonstrating an acute perception of form and texture by 

 Bombus and A pis especially. The significance of this is further discussed 

 in the final r6sume\ 



Honey and odor. — In the majority of the species where it was employed, 

 honey actually decreased the number of visits, often tenfold, while in a few 

 it had no discernible effect. In no case did it increase the number of visits 

 to natural flowers, and it had practically no effect upon those to artificial 

 ones. The significant result of these experiments was to confirm the evi- 

 dence of the exceedingly poor sense of smell in bees for honey. Often they 

 would not discover the drop at a distance of a few millimeters and in some 

 cases they even became entangled in it without recognizing it. This may 

 have been due to the odor of the flower itself, but as it was true of all 

 the flowers employed, it suggests that the odor of nectar plays a very small 

 part in the attraction exerted by many species. 



The addition of perfumes derived from other flowers or of other odorous 

 substances decreased the total of visits 50 per cent in comparison with 

 normal flowers. Even when honey was also added, a similar relation 

 obtained, though it is certain if the bees had been trained to come for 

 honey on flowers with a certain perfume that the latter would have seemed 

 attractive. With odor, as with color, form, and texture, habit is the con- 

 trolling factor, and this probably explains why the various perfumes ap- 

 peared to repel. They disturbed the habitual response of the bee, and 

 this was true whether perfumes, essences, or odors unpleasant to us were 

 employed. The results certainly do not strengthen the role of odor in 

 attraction, though undoubtedly it often plays a part in forming the habit 

 of guidance. 



