METHODS OF STUDYING VISION IN ANIMALS 35 



light. Their chief advantage over solutions for visual tests 

 is portability. 



At present the following are the most satisfactory colored 

 glasses of which we have knowledge, together with sources of 

 supply : 



E. Grosse, Berlin, N. W. 52, Paulstrasse 5, supplies flashed 

 glasses which yield red, yellow, green, blue, and violet light. 

 These glasses are, on the whole, the most satisfactory of the 

 flashed glasses which we have been able to obtain. Only the 

 red, however, is monochromatic. By using combinations of 

 the blue and violet glasses, it is possible to get a stimulus which 

 ■contains no red or yellow. If monochromatic light is desired 

 these glasses, with the exception of the ruby, are of no value. 



Chance Brothers and Company, Birmingham, England, manu- 

 facture colored glasses which are used for signals, but examin- 

 ation indicates that except in combination they have no special 

 value for our purposes. 



The Central Scientific Company, 14-28 Michigan street, 

 Chicago, Illinois, sells (a) plates of pot glass, and (b) plates of 

 flashed glass. The latter they furnish in seven colors. Plates 

 10 cm. square, 11 cents each. These glasses, according to our 

 spectroscopic examinations are not as satisfactory as those 

 supplied by E. Grosse. 



Schott and Genn, Jena, Germany, manufacture a high grade 

 of colored glass (spectro-optical). much more expensive and 

 also more satisfactory than the flashed or pot glasses of com- 

 merce. Their latest series consists of six colored glasses: "dark, 

 medium, and light yellow, blue filter, green glass, and red filter." 

 These may be obtained in dift'erent thicknesses, and, according 

 to catalogue statements, in pieces not exceeding 5 cm. square. 

 The price of pieces approximately 4 cm. X 4 cm. is about $1.25 

 each. 



Glass filters greatly reduce the intensity of a chromatic stim- 

 ulus and it is therefore necessary to employ with them an intense 

 source of light. They are of extremely dift'erent values with 

 respect to dift'erent wave-lengths. 



For red one or two thicknesses of flashed ruby glass is ex- 

 cellent. 



For yellow no colored glass, or combination of glasses, is 

 satisfactory. 



