92 CORA D. REEVES 



of light not entirely dependent upon the intensity of the 

 light. 



V. SUMMARY OF EXPERIMENTS 



A. Apparatus 



1. A T-shaped experiment aquarium of 

 galvanized iron supplied with running 

 filtered water was used, and in this the 

 fish remained not only during experi- 

 ments but in the intervals between 

 them. p. 9, figs. 1, 2. 



2. The aquarium was divided into com- 

 partments. The first of these con- 

 tained two stimulus plates of opal 

 ground glass, which could be exposed 

 to the fish to be immediately tested 

 by lifting a sliding door separating it 

 from a second compartment. The third 

 compartment contained fish not being 

 tested, p. 5, fig. 1. 



3. The partitions between the compart- 

 ments were movable, and by manipu- 

 lating them the fish could be shifted 

 from compartment to compartment 

 without causing fear reactions, p. 9, 

 figs. 1, 2, 6. 



4. The stimulus plates were illuminated 

 by lamps in fixed positions, contained 

 in a light-tight house, pp. 7, 10, fig. 2. 



5. The intensities of the illumination on 

 the plates were varied by means of 

 adjustable slits beneath the lamps, pp. 

 7, 12, fig. 2. 



6. Well defined areas of light of practi- 

 cally uniform distribution were secured 

 upon the stimulus plates by the use 

 of cylindrical lenses, pp. 7, 12, fig. 2. 



7. Red and blue filters were used to 

 secure light of restricted wave-lengths, 



