METHODS OF STUDYING VISION IN ANIMALS 59 



light, and is at the same time illuminated from above by the 

 beam of monochromatic light coming from the speculum mirror. 

 The energy of the whole effective stimulus may be taken, or 

 that of each component separately. 



The plaster of pans boxes are identical with the former in 

 size and shape. The whole upper end of the box, however, is 

 filled with plaster of paris, in accordance with the following 

 method: A good piece of plate glass is first selected. This is 

 then ground with emery of coarse to medium fineness until 

 the surface of the glass shows a uniform grinding. The large 

 end of the metal box is inverted upon this. Dental plaster of 

 paris, mixed with water until it begins to set, is poured into 

 the box and allowed to stand 24 hours. The box is then care- 

 fully lifted from the surface of the plate glass, and all excess 

 plaster of paris is removed with a damp cloth, great care being 

 taken not to mar the diffusing surface. After the plaster is 

 thoroughly dry it offers a splendid diffusing surface with a 

 viinimum of direct reflection. A small dust-proof cap should 

 be kept over these surfaces when the apparatus is not in use. 



The stimulus boxes have small metal guides, D and G, attached 

 to their base, supplied with clamp-screws, by means of which 

 the boxes are easily attached to the supporting rods, R' and 

 R^ passing through the cabinet. From the drawing it is clear 

 that these boxes are inclined at an angle of 45°, so that the 

 face of the diffusing surface is parallel to the face of the mirror. 

 Thus, each part of the monochromatic beam passes through 

 the same distance. 



If one contrasts the use made of this cabinet with that of 

 the stimulus carriage H, one sees that in the latter there are 

 only two mirrors and two stimulus boxes, and that consequently 

 the whole has to move from left to right, or vice versa, in order 

 to catch the green beam as it is sent to the left or to the right 

 by means of the reversing device. The cabinet, however, sup- 

 plied with three mirrors and three stimulus boxes, remains 

 stationary while the beams are reversed as before, either auto- 

 matically by the movement of block 46 or by means of a light 

 cord, R, directly connected with the reversing mechanism. 



When the cabinet is used the animal is placed in a separate 

 experiment box in front of it (see fig. 4 and p. 84). The cage 

 containing the animal must be movable, parallel to the front 



