METHODS OF STUDYING VISION IN ANIMALS 83 



trum appears vertically above the other. The slit in front of 

 the eyepiece is then adjusted so as to admit the whole of the 

 variable beam. Immediately below it with no line of demar- 

 cation one finds a band of monochromatic light of exactly the 

 same wave-length, coming from the standard source. The 

 latter in this work is adjusted in intensity so that the standard 

 beam is always fainter than the comparison. The comparison 

 beam is then decreased in intensity by the use of the analyzing 

 Nichol until the two are of the same brightness. The angle 

 of the analyzer is then read, which gives after proper translation 

 the intensity of the variable beam in terms of the standard.' 



This method, while in general use, is subject to all the varia- 

 tions present in the photometry of white light. X standardized 

 procedure, accordingly, should be adopted. Numerous deter- 

 minations should be made and with a dark-adapted eye. Since 

 they do not have to be made frequently, it seems desirable 

 to allow at least 15-20 minutes for adaptation to darkness before 

 taking readings. 



The light chosen as a standard should be a well seasoned 

 miniature tungsten with approximately straight filament (the 

 bulb should be marked with fiducial lines and the lamp should 

 be burned in a position constant with relation to the instrument) 

 of about i^ c. p. at 2 volts pressure. It should be burned in 

 series with a rheostat and one two-volt storage battery. A 

 volt meter reading 0-6 in 1-25 volt divisions should be placed 

 in the circuit. Switchboard type C voltmeter, supplied by Jas. 

 G. Biddle, Philadelphia, for $15, will suffice, but a precision 

 voltmeter of the well known Westinghouse type is better. 



Flicker photometer. — In making color tests upon animals, 

 the student of behavior is not especially interested in obtaining 

 color stimuli which offer "equality of brightness" to the human 

 eye. This relation may or may not have bearing upon the 

 animal work. It is always interesting, however, to test the 

 animal under such conditions. The flicker photometer presents 

 probably the most uniform way of obtaining between dis- 

 parate colors a relation which is considered by some to be that 

 of equality. For a thorough experimental treatment of this 



^ For further description of the instrument, see P. G. Nutting, Bull. Bureau of 

 Standards, 1906, vol. 2, p. 317. 



