RECENT INVESTIGATIONS. 205 



the various species. Thus, flower fragrance is a distinguishing mark for 

 bees, and perhaps the most important one that the flower possesses. In 

 the light of this fact, it is readily understood how bees can remember for 

 days or even weeks an odor to which they have been trained for a short time 

 only. 



It seems significant of the psychic processes in these highly organized 

 insects, which are able to perceive flower odors so quickly and to make use 

 of them, that they refuse almost completely to respond to training experi- 

 ments with foul-smelling substances, and that certain other odorous sub- 

 stances give only poor training results. There seems no other explanation 

 of this than the assumption that there are odors which had been without 

 significance for bees through countless generations and to which they 

 habituated themselves not to respond. In like manner they readily learn 

 to distinguish forms which suggest those of flowers, while training them to 

 geometric figures is completely unsuccessful. 



In their entire behavior bees show in their sense of smell an agreement 

 with that of man which is as far-reaching as it is unexpected. All of the 

 32 odors to which they were trained are effective for them, as for us. All 

 the substances tested that are inodorous for us are equally so for them. Sub- 

 stances with strong odor for us are likewise strong for bees and the reverse, 

 just as those that are similar for the one are alike for the other. Thus, it 

 is possible to evaluate the biological significance of flower odors with far 

 greater certainty than heretofore. 



Bombylius and the colors of flowers. — Knoll has presented the 

 results of his investigations of the behavior of Bombylius and Macroglossa 

 in two comprehensive monographs, which must be ranked with those 

 of Frisch with respect to the detail and accuracy of the treatment. The 

 first deals primarily with the response of Bombylius fuliginosus to the flowers 

 of Muscari, under the following headings: (1) main objects of the experi- 

 ments; (2) experiments, (a) choice and preparation of the experimental 

 area, (b) determination of the optical attraction of flowers, (c) studies of 

 the chemical attraction of flowers, (d) experiments with Frisch's methods, 

 (e) near-by attraction of Muscari; (3) constancy of Bombylius in its visits 

 to certain flowers and the behavior of other species of the genus to the same 

 flowers; (4) general considerations on the color sense of Bombylius and that 

 of the honey-bee. The experimental results showed that the flowers of 

 Muscari attracted this fly at a distance by color and form and that fragrance 

 had no part in this. It was also found that Bombylius possessed the ability, 

 though probably in smaller degree than in the honey-bee, to associate the 

 color of a flower and the presence in it of an easily accessible store of nectar. 

 Further, since this fly visits the brightest or pure white flowers as well as 

 the less bright ones of the blue group, but ignores the deep yellow, which 

 are also very bright, it seems that the attraction is determined by the quality 

 of the color rather than by its intensity. It may be assumed that the 

 color vision of insects and of vertebrates has been developed independently 

 on the basis of the independent evolution of the sense organs and the related 

 nerve systems. A consideration of the structure of the sensory apparatus 

 and the central organ of the two groups confirms us in the assumption that 



