90 



TBOPISMS 



trail with the line connecting the two lights. In this way in each observa- 

 tion the average trail of thousands of individuals is measured. By using 

 one constant and one intermittent source of light and comparing the 

 results with those obtained by two constant lights we can test the validity 

 of the Bunsen-Roscoe law. 



The method of the experiments was as follows: abed (Fig. 31) is a 

 square dish of optical glass with blackened bottom and containing a 



A 



o 



B 



FIG. 31. Method for the proof of the validity of Bunsen-Roacoe law for the positively 

 heliotropic larvae of the barnacle. (After Loeb and Northrop.) 



layer of sea water. A and B are two lights, the intensity of which is 

 determined by a Lummer-Brodhun contrast photometer. In front of 

 each light is a screen with a round hole permitting a beam of light to 

 go to the dish. The lights and the dish abed are so adjusted that the 

 two beams of light striking the sides ab and be at right angles cross 

 each other in the middle of the dish. The light A is fixed while the 

 light B is movable on an optical bench. The experiment is made in 



