250 BRADLEY M. PATTEN 



plus. The sum of the plus and minus deflections, divided by the 

 total number of trails, would give the average angular deflection 

 toward the less effective color. The fact that certain of the larvae 

 turned toward the color which in general acted more strongly, I 

 believe to be due to the same asymmetry of responsiveness which, 

 in my own experiments, caused the deflections toward the stronger 

 of two white lights. The lack of balance of the larva in such a 

 case would more than offset the difference in the colors, and the 

 less effective color acting on the more sensitive side would pro- 

 duce a reflection toward the more effective color. In a large 

 number of trails, however, this unstable asymmetry in sensi- 

 tiveness would tend to balance out, and the average deflection 

 would be a reasonably accurate expression of the relative stimu- 

 lating powers of the two colors. In the case of green opposed to 

 blue, the average angular deflection was about 24° away from 

 the green toward the blue. This angle of orientation could be 

 attained by a turning either to the right or left and is represented 

 by the heavy lines A and A' in the figure. As Gross has pointed 

 out, this deflection is due to the qualitative difference between 

 green and blue lights, for by actual physical measurement the 

 intensities of the lights are the same.*^ Comparing the angle of 

 deflection under green and blue lights with that under unequal 

 white lights, the angle of response corresponds to that produced 

 when the difference in the intensity of the white lights is 56 per 

 cent (fig. 15). In other words, green light is about twice as effec- 

 tive as a stimulus for the blowfly larva, as blue light. The values 

 in the other color combinations could be estimated in the same 

 way, but without access to the original records such computations 

 would be only approximate and have therefore been omitted. 

 There appears to be no other way in which the relative effective- 

 ness of different pairs of colors can be estimated quantitatively. 

 The method is presented more in the hope that it may prove of 

 value in future work, than for the significance of the approx- 

 imate relationship pointed out in this special case. 



« Gross also showed that with opposed lights of the same color and intensity, 

 the average trail lay perpendicular to the line connecting the sources of light, 

 thus establishing his zero point for the deflections obtained under different colors. 



