METHODS OF STUDYING VISION IN ANIMALS 33 



not meet the fundamental requirements, controllability and 

 describability. 



Although for quantitative experiments the reflection method, 

 in its available forms, is quite unsatisfactory, it is not to be 

 discarded lightly, for it offers important conditions for the pre- 

 liminary investigation of the nature of an animal's color vision 

 which no other method furnishes. Above all, the naturalness 

 of the stimulus, and of the conditions under which it may be 

 applied, would seem to be important. We wish, therefore, to 

 recommend the employment of papers, cloths, and oil pigments 

 under suitable conditions, and with recognition of their limi- 

 tations. No experimenter can reasonably hope to gain adequate 

 knowledge of the visual capacity of an animal by the use of this 

 method alone, although by means of it he may obtain knowledge 

 which will enable him to formulate his problems and advanta- 

 geously apply other methods to their solution. 



h. The transmission method. — In this rubric we have tried to 

 include all suitable means of obtaining chromatic stimuli by 

 selective absorption and transmission. Among the most im- 

 portant of these means are glasses and gelatines ("dry filters") 

 and solutions ("wet filters"). Before taking up these several 

 types of absorption media, we may enumerate the chief merits 

 and defects of the transmission method, in contrast with the 

 reflection and dispersion methods. 



Merits: Availability in many forms; cheapness, in compar- 

 ison with all forms of apparatus for the dispersion method ; 

 convenience and simplicity of apparatus (this applies especially 

 to glasses and gelatines — the so-called dry filters — it does not 

 hold to the same extent of w^et filters) ; the possibility of use 

 in daylight and sunlight ; ease of changing quality and intensity 

 of stimulus independently, within certain limits ; reasonable 

 ease and accuracy of measuring wave-length and intensity of 

 stimulus. (In the last two features the transmission method 

 is infinitely superior to the reflection method.) 



Defects : Inconstancy of qualitative values of commercial 

 colored glasses and gelatines (this does not apply to solutions ) ; 

 more or less rapid fading (this renders unsatisfactory expensive 

 sets of glasses) ; unnaturalness of the stimulus, as compared 

 with colored papers, or cloths or other surfaces, viewed in sun- 

 light (it is possible to use color filters in sunlight, but their 



