594 MR WILLIAM SWAN ON THE GRADUAL PRODUCTION OF 
the intensity of the impression is precisely the same in both cases. But at the 
velocity of 40 revolutions in a second, there are twice as many flashes in a given 
time as there are at the first velocity; and if the brightness of the flashes was 
the same as at the first velocity, the brightness of the impression produced by 
them would be doubled. Since, therefore, the impression, instead of being doubly 
bright, remains the same as at first, each flash at 40 revolutions in a second must 
only be half as bright as at 20 revolutions in a second. In like manner, by sup- 
posing the velocity increased to 80 revolutions in a second, it might be shewn 
that the brightness of the flashes is again halved. But the effect of doubling the 
velocity is to halve the duration of the flashes, therefore the brightness of the flashes 
is proportional to their duration. This law of vision may be thus stated: When 
light of a given intensity acts on the eye for a short space of time, the apparent 
brightness of the luminous impression on the retina is exactly proportional to the 
time during which the light continues to act. From the velocities of the discs, 
and the angles of the sectors used in the experiments, it will be seen that this 
law is true for impressions lasting from -1., to of a second; and it will 
18432 “" 120 
presently be shewn to be true for impressions of longer duration. 
V. Observations of the apparent Brightness of Luminous Impressions of short duration. 
In almost all the experiments hitherto described, the phenomena of vision 
which have been investigated have been derived from the observation of the 
aggregate effect of luminous impressions succeeding each other so rapidly as to 
produce a continuous impression on the eye. It is obvious, that such expe- 
riments afford no information regarding the absolute brightness of the sepa- 
rate impressions which are thus blended together. I adopted the indirect 
mode I have now described of ascertaining the connexion between the duration 
and apparent brightness of luminous impressions, from an apprehension of the 
difficulty of comparing the brightness of a constant light with that of an isolated 
flash. But repeated trials satisfied me that my fears were groundless; and the 
succeeding experiments prove that, with a little practice, the eye is perfectly capa- 
ble of making this comparison. Such experiments cannot, however, be long con- 
tinued without fatiguing the eye, and a considerable effort of attention is re- 
quired for their successful performance. 
In order to find the intensity of separate impressions of short duration, I 
used a disc of wood two feet in diameter, revolving once in a second; so that a 
sector, whose arc had a known ratio to the circumference of the disc, passed at 
each revolution before the aperture in one of the screens of the selaometer. In 
this manner, a series of perfectly isolated impressions was obtained; and the 
intensity of each could be compared with that of a light seen by continuous 
vision in the manner already described. The different sectors were cut in paste- 
board, and placed over an aperture in the disc. The following experiments were 
made :— 

