262 



HOVEY JORDAN 



sities. These were placed within a small wooden reflector-box which 

 had a small hole in one end. The box was lined with tin for the pur- 

 pose of increasing the intensity of the illumination. It is to be re- 

 gretted that no means was available for testing the exact intensity of 

 the light; 3 but for the study of variations in the sensitivity in different 

 regions of the integument and of photodynamic responses, relative in- 

 tensities, such as are obtained from wattages of various magnitudes, 

 are sufficiently precise. The illumination of the fish, however, was 

 less intense than the indicated wattage of the lamp, because a certain 



Figure of the aquarium used in testing fishes for photosensitivity 



amount of light was cut off by the edges of the reflector-box and by the 

 water 4 between the fish and the light. 



These experiments were conducted in an aquarium which has already 

 been figured and described elsewhere (Jordan '17); the figure is repro- 



3 The candle power of mazda electric light bulbs is approximately 1 per watt; 

 but, as the candle power was not measured in the following experiments, wattages 

 instead of candle powers will be used to express the intensity of illumination. 



4 Temperature changes are negligible, since a thermometer held in the position 

 occupied by the fish was unaffected by the light. There was usually about 6 to 

 8 inches of water between the light, which was held against the glass of the aqua- 

 rium, and the fish. 



