592 MR WILLIAM SWAN ON THE GRADUAL PRODUCTION OF 
Number of Flashes in Brightness of 
a Second. Impression. 

In these experiments it is assumed, that when the light is withdrawn from the 
screen, so as to diminish the intensity of its illumination, the brightness of a flash 
of short duration will be diminished in the same ratio. This will only be true, 
provided lights of different intensity produce impressions of proportional inten- 
sity in equal times; but if it be afterwards proved, independently of these obser- 
vations, that such is really the case, the conclusion which has now been drawn 
from the experiments will be perfectly correct. In order to avoid this assumption, 
the brightness of the impressions produced by the revolution of the different discs 
was next compared with the impression of uninterrupted light. 
1. A disc, with a single sector of 30°, was placed in the selaometer, with a gas- 
flame at a constant distance of 6 inches from its screen. In order to render the 
illumination of the screens equal, the other light had to be placed at a distance of 
59 inches from the second screen. When the disc revolved 20 times in a second, 
the latter light was gradually withdrawn until the apertures in the screens again 
appeared equally bright. The distance of that light was now found to be 22-2 inches. 
This experiment, eight times repeated, gave the following mean values :— 
d, = 578; d, = 22-21; 6, = 006773. 
where 6, denotes the brightness of the impression produced during the revolution 
of the disc. In like manner, a disc, with two sectors of 30°, treated in precisely 
the same manner, gave the following mean values of eight trials :— 
d, = 561; d, = 15-22; 6, = 0-1859. 
6, denoting the brightness of the impression produced by the revolution of this 
disc. From this experiment, the ratio of the brightness of the impressions pro- 
duced by the revolution of the two discs, or os will be found to be 2°006. 
if 
2. A disc, with three sectors of 30°, was compared with direct light in the 
same manner. The mean of eight experiments gave 
d, = 561; d, = 12-11; 6,= 0-2164. 
From which the ratio of the brightness of the impression produced by means 
of this disc, with three sectors, to that produced by means of the disc with a single 
(Dees 
sector, or 9 is 3°169. 

