1897.] NEW YORK ACADEMY OF SCIENCES. 209 



flickering when rotated in pair-wise combination. Now we 

 would expect the differences in measured luminosity to be very 

 small if the violence of the flicker is dependent on difference in 

 luminosities. The luminosities of the disks were then measured 

 by the method of Equal Flickers, the average of five determina- 

 tions being taken for each disk, and the difference between the 

 measured luminosities of the disks of each pair was found in no 

 case to be over two per cent. The actual results were as follows: 



A red disk and a dark blue disk were found which gave very 

 little flickering when rotated in pair-wise combination. Their 

 measured difference in luminosity was 1.7 per cent. 



A yellow disk and a blue disk gave very little flicker when 

 rotated in pair-wise combination. Their measured difference in 

 luminosity was 1.6 per cent. 



A light green and a light blue disk gave less flicker than either 

 of the above combinations. The measured difference in lumi- 

 nosity was 1.2 per cent. 



A red and a yellow disk gave very little flicker. Their 

 measured difference in luminosity was 1.0 per cent. 



A very dark red and a very dark green disk gave very little 

 flicker. Their measured difference in luminosity was 1.1 per 

 cent. 



The above results show that where the flickering between 

 differently colored disks is very slight the measured diflference 

 in luminosity is small. Expressing the general result in a few 

 words one can say that, in the matter of flicker, under daylight 

 illumination differently colored disks behave in the same way 

 with each other as they do with gray disks. 



To still further test the matter a number of combinations 

 (See Figs. IV. to VI.) of colored disks, the luminosities of 

 which had been determined, were so made that on the inside two 

 colors of the same hue but of different luminosities were opposed, 

 while on the outside two colors of different hues but having a 

 less difference in their measured luminosities were opposed to 

 each other. Each combination was then placed on the axis of 

 the machine and rotated at such a speed that all flickering dis- 

 appeared. Then, as the speed was gradually diminished, the disk 

 was observed through a rectangular opening in black cardboard 

 and the relative violence of the flickering outside and inside 

 thus compared. The following is a record of the results obtained. 



A large green disk and a large blue disk which I will call 

 Green 1. and Blue 1. respectively (See Fig. IV.) were opposed 

 outside. Their measured difference in luminosity was 1.2 per 

 cent. On the inside a small green disk, Green 2, was opposed 



Transactions N. Y. Acad Sc, Vol. XVI.. Sig. 14, April 28, 1897. 



