PATTERN-DISCRIMINATION IN VERTEBRATES 



337 



enough for extensive work, although the gratings were satisfactory 

 for Problems 1 and 3. I should recommend, therefore, that the 

 investigator provide a second set of gratings, of say 1-100 inch 

 width for use on animals whose vision is poorer than that of the 

 chick. The extra set of gratings can be cemented into their 

 own rings, and would thus be made quickly interchangeable with 

 the finer set. 



For testing the threshold for contrast, the writer has had two 

 special Lummer-Brodhun cubes of 6 cm. square face, constructed 

 as shown in Fig. 5. These cubes are obtainable from Schmidt 



FIG. 5. Special Lummer-Brodhun cube showing pattern. 



& Haensch at about 490 marks the pair. About two months' 

 time is required for manufacture. The pattern etched on the 

 diagonal face is three bands, each 1 cm. wide and 8.42 cm. long, 

 the clear portion lying between any two of the etched bands 

 being of the same size. The cubes, C 1 and O, are set a little 

 back of the windows, W 1 and W 2 , of the Yerkes box, in which 

 the animal is placed. L 1 , L 2 , L 3 and L l represent the stimulus- 

 lamps, and Sc 1 , Sc 2 , Sc 3 and Sc 4 , opal-glass diaphragms, indicated 

 in the sketch but not lettered, are used to protect the cubes and 

 the windows of the Yerkes box from reflected light. D 1 , D 2 , D 3 , 

 and D 4 represent sectored discs. D 1 and D 2 as a pair are inter- 

 changeable with D 3 and D 4 as a pair. The portions of the fields 

 illuminated by each lamp are equalized in brightness by placing 



