Reactions of I so pods to Light 249 



hereafter referred to by numbers, i to 6. This served as a means 

 for estimating the relative positions of the animals at any given 

 instant. The records were made by counting and recording the 

 number of individuals in each section of the tank at certain inter- 

 vals during the experiment. The enumerations were made by 

 observing from above, care being taken to prevent, or reduce to a 

 minimum, the possible reflection of light from above, so that the 

 records were obtained without in any way interfering with the 

 course of the experiment. For convenience and safety in handling 

 the glass tank (Fig. i, IT) it was placed within a larger wooden 

 tank [OT) with glass ends 63.5 cm. long and 30 cm. broad. 

 Both tanks were filled with water to a depth of 3 cm. 



During experiments with horizontal illumination there was used 

 as a heat screen {HS)3. rectangular glass jar, 31 cm. long, 20 cm. 

 high and 8 cm. from front to back, filled with filtered water. 



Difi^erent sources of illumination (L) were used: a 6-glower, 

 220-volt, Nernstlampof 772 c. p., and for lower intensities, either 

 a 19 c.p., an 8 c.p., a 5 c.p., or an 0.8 c.p. incandescent lamp. 

 Variation of the distance between the lamp and the tank was also 

 used as a means of regulating the intensity of illumination. Much 

 of the time while experimenting with horizontal illumination two 

 lamps of the same intensity were placed at opposite ends of the 

 tank. By a switch device, one light could be turned off and the 

 other on, thereby reversing the direction of illumination without 

 disturbing the animals or interrupting the observations. Only 

 one 6-glower Nernst lamp was available however, so that it had to 

 be shifted when a change in the direction of the light was desired. 

 Extraneous light was carefully excluded. The lamp used was 

 placed inside a lamp container [LC) made from a piece of black- 

 ened sheet-iron bent so as to form a rectangular box with open 

 ends. One of these open ends was kept covered with black cloth; 

 the other, which was directed toward the tank, was fitted with 

 an opaque screen (S) that had in its center a diaphragm of ad- 

 justable size. When the lamp container was placed at some dis- 

 tance from the tank, the rays of light passing to the tank were con- 

 fined within a hollow blackened half-cylinder (HC) thus pre- 

 venting the escape of light into the room. 



