30 BOTANICAL GAZETTE [july 



cause large variations in the intensity. In the third place, unless 

 an automatically adjusted arc is available, it is necessary to adjust 

 the carbons by hand every 5 to 15 minutes, and, when an 8 hour 

 illumination is desired, this entails considerable inconvenience. 

 The experiments were begun with this light, however, and the 

 results on rhizoidal phototropisms in monochromatic lights made 

 equal in intensity were obtained with it. 



The spectroscopic analysis of the light passing the screens deter- 

 mines definitely the quality of light entering each box. It is at 

 once evident that the quantity or intensity of light behind filters 

 placed at equal distances from the source varies, both because the 

 intensity of light transmitted by the different screens is different, 

 and because the different colors are not radiated by the arc with 

 equal intensity. This being the case, differences in results obtained 

 behind the screens could not be attributed to differences in the 

 quality of the light stimulus alone. 



One of the simplest means of comparing the radiant energy of 

 colored lights is by the use of the thermopile and sensitive galva- 



nometer. 



energy 



red 



heat rays, or the shorter actinic rays of the spectrum. 



m 



elimi 



filters employed 



from the arc at which each 



mi 



should be placed in order that the intensity 



each case. These distances were those at which the deflections of 



the galvanometer were equal when the thermopile was exposed to 



filter 



necessary 



thermopile 



le energy of the red and of the 

 therefore, an accurate measure 



amount of light acting behind each color screen. The difference 

 between heat and light is only a matter of wave length. The 

 thermopile measures light in terms of the electric current produced 



temperature of the exposed 



energy 







