SPECTRAL COLOR AND STIMULATION 507 



scarcely any other organisms. There was consequently no lack 

 of material for study. Moreover, these creatures respond very 

 strongly to light and orient remarkably precisely, but there is 

 one peculiar feature connected with their response that inter- 

 fered considerably with the work. 



This form is the first one that was studied and it was soon 

 learned that while in a series of successive readings extending 

 over a short period of time there was usually but little variation, 

 there was often marked variation if the series was long continued, 

 e.g., on July 27 in a series of readings the sector opening, with the 

 wave-length of the colored light 524 /x^t, was 48.5 and 46.3 in 

 two successive tests made at 8.37 and 8.39 a.m. respectively, 

 while in two tests made about an hour later (9.35 and 9.36) un- 

 der the same conditions of illumination the opening was 66.8 

 and 70.3 respectively. Similar incongruities were observed re- 

 peatedly in long continued series of tests. 



Now, these variations were no doubt due in part to slight move- 

 ment of the image of the luminous filament on the spectrometer 

 slit resulting in variation in the illumination in the spectral beam 

 as previously pointed out, but they were, I believe, more largely 

 caused by the following peculiarity in the response of these 

 organisms. 



In a horizontal beam of light consisting of parallel rays, the 

 colonies of Pandorina, like those of Volvox (Mast '07), do not 

 swim directly toward or from the source of light but deflect to 

 the left, and in a field of light consisting of two beams crossing 

 at right angles, they also tend to deflect to the left so that in 

 order to make them proceed on a line which bisects the angle 

 between the beams, it is necessary to make the light in the beam 

 to the right stronger than that in the beam to the left. This 

 fact would not interfere in the least with the process of ascer- 

 taining the relative stimulating efficiency of the spectral colors 

 by the method employed in this work if the tendency to deflect 

 were continuously the same, but unfortunately it is not. The 

 degree of deflection varies with the condition of the organisms 

 and with the intensity of the light. The more strongly positive 

 or negative the organisms are the less the deflection and conse- 



