Theory of the Flicker Photometer. 319 



This setting £01* no colour difference should always be 

 determined. 



It is our belief that a number o£ anomalous results which 

 have been obtained and reported in the use of the flicker 

 photometer are to be traced to asymmetric conditions such 

 as those just discussed. We may note among these : — 



First : Comparison of the readings of flicker and equality 

 of brightness photometers. Measurements of a colour 

 difference have been made first with an equality photometer 

 and then with a substituted flicker photometer, with sur- 

 prising differences. Such readings should always be com- 

 pared against the reading of the same instruments with no 

 colour difference present. 



Second : Comparison of the readings of a flicker photo- 

 meter which exposes each field equally with ones in which 

 one colour is exposed longer than the other. Very large 

 differences between such photometers have been reported, 

 contrary to our findings reported above. Here, again, the 

 no-colour-difference setting should be determined for each 

 instrument and taken as the starting point. 



Third : Comparison of the areas of spectral luminosity 

 curves obtained by the flicker and equality of brightness 

 methods. These have been frequently found to differ, 

 although the shapes of the curves are. under certain con- 

 ditions, the same. These differences may, in part, be real, 

 but the curves, as obtained by the senior author *, were 

 determined under conditions which would readily permit 

 the occurrence of dissymmetries of the kind which have 

 been discussed, and whose effect would be to cause area- 

 differences of just the kind found. The conclusions drawn 

 as to the generality of these area-differences may therefore 

 be modified. In the experimental determination of the 

 equality of the sum of the parts to the whole, with the 

 flicker photometer t, which was the only case where the exact 

 measurement of areas was of vital significance, most elaborate 

 precautions were taken to obtain a photometric arrangement 

 free from any mechanical irregularities, the importance of 

 this being recognized from general considerations. 



We have dwelt at considerable length on these effects of 

 accidental dissymmetry because it is important that their 

 nature be understood and the necessary precautions taken 



* M Spectral Luminosity Curves obtained by the Equality of Bright- 

 ness Photometer and the Flicker Photometer under Similar Conditions " 

 Ives, Phil. Mag. July 1912, p. 149. 



t " The Addition of Luminosities of Different Colour," Ives, Phil. 

 Mag. Dec. 1912, p. 845. 



