202 PRINCIPLES AND CONCLUSIONS. 



conclusions as to the ability of bees to distinguish colors as such. It is 

 now impossible to deal with his experiments and criticisms in detail, and 

 it must suffice to give the summaries of the two papers that treat this 

 question directly. These will serve also to reveal the attitude of the author, 

 who goes far beyond Plateau in rejecting unqualifiedly the results of those 

 who do not agree with him. 



"It has been demonstrated that the earlier conclusions of Lubbock and Forel, as well 

 as the later ones of Frisch, in accordance with which it is possible to train bees to discrimi- 

 nate a particular color, are incorrect. If the colors are presented under conditions other- 

 wise identical, the bees find it wholly impossible to habituate themselves to a certain color 

 and to respond to it. The errors of the earlier investigators are in part to be ascribed to 

 the fact that they either ignored the contributing factors emphasized by me or failed to 

 take them sufficiently into account. 



" Up to the present time not a single fact has been adduced that makes even probable the 

 assumption that bees possess a color sense comparable with ours. On the contrary, my 

 earlier researches with spectral and other rays as well as the later ones with colored papers 

 refute this opinion conclusively. In virtue of the results contributed by me, which are 

 readily demonstrable, the view of Sprengel as to the significance of flower color in insect 

 attraction can no longer be maintained." (1913: 105.) 



"It has been demonstrated that every training experiment with bees, which has been 

 thought to prove their color sense, affords complete support to my researches which show 

 their total color-blindness. Moreover, Frisch's record shows, in entire agreement with 

 mine, that bees supposedly trained to blue or yellow were unable to distinguish the two 

 colors, but much more frequently confused one with the other or with gray. 



"The inadequacy of the method of training experiments is forcibly exhibited by these 

 new observations and measurements. Furthermore, the objections recently raised from 

 the zoological side to my investigations are vitiated by the results given here. Finally, 

 even Frisch himself has removed the last prop from the concept of color sense in the bees 

 by his training experiments and the Freiburg results." (1918:365.) 



Frisch (1919:122) has made an effective rejoinder to Hess's statements, 

 as is indicated by the following excerpts : 



"Hess begins his real objections with the surprising assertion that my record shows 

 that the bees were unable to distinguish blue and yellow from gray, and also blue from 

 yellow. I wish to ask him how he can advance such a statement when he must have seen 

 from my record that the bees trained to yellow discerned the latter and discriminated 

 between it and all the shades of gray in all seven experiments in which a yellow paper was 

 exposed in the gray series; that further, the bees trained to blue selected this color and 

 distinguished it from the gray shades in the entire 15 experiments with blue similarly exposed; 

 that bees trained to yellow sought this in overwhelming majority in competition with blue 

 and purple in all eight experiments, and finally that bees trained to blue chose blue and 

 purple just as decisively in 25 of the 26 tests with the complete color series, the exception 

 being readily explained by the circumstances (p. 126). 



"I have shown that bees trained to blue for two days were able to distinguish with 

 certainty a blue paper, which, according to Hess, they see as a colorless gray of a particular 

 shade, from gray papers of any shade. In order to remove the objection that they directed 

 themselves to the blue paper by virtue of an odor imperceptible to us, all the papers were 

 covered by a glass plate. When the mass of bees above the blue paper were brought upon 

 a gray one by moving the glass plate, the mass dissolved within ^ to J^ minute and a new 

 one formed upon the blue" (p. 135). 



Seasonal change in the response to honey. — Zander (1913:711) 

 found that, while honey exposed in summer stood day after day without 

 visits from honey-bees, in agreement with the results of Forel, it was sought 

 by hosts of them in September. This led him to make similar tests at inter- 

 vals of about two weeks from the end of April to the beginning of October. 



