442 Proceedings of Royal Society of Edinburgh [july 16 , 
of a stimulus of given physical intensity depends upon the time 
during which it acts. But stimuli which individually are unable to 
produce a muscular contraction may be summed, if they are thrown 
in at a sufficiently short interval; and the same is true of stimulation 
of the retina, at least in this sense, that stimuli which act for 
too short a time to produce a sensation when isolated, may do so 
if allowed to follow each other rapidly, without diminishing the 
length of each. This can only happen, however, when each 
stimulus produces some impression, which, though not amounting to 
a sensation, is a step on the way to one. If a limit can be reached 
where a stimulus acts for so short a time that it produces no change 
which helps towards a complete impression, it is clear that at and 
below this limit Talbot’s law cannot be true. If only every second 
flash were effective, we should expect the intensity of the resultant 
sensation to be diminished by half ; if only every fourth flash were 
effective, by three-fourths, and so on. If all the flashes were 
effective, but in a diminished degree, the resultant impression would 
be feebler to a corresponding extent. If none of the flashes were 
effective, there could, of course, be no sensation, because the sum of 
a series of zeros could not be finite. 
Method of the Investigation . — My plan was to obtain fusion of a 
series of very short flashes, and then to abruptly diminish the length 
of each, while increasing the number in the same proportion, the 
interval between two successive flashes being very great compared 
with the time of each. It was to be observed whether any change 
took place in the intensity. 
In a room about 10 metres long in the Physical Laboratory of 
the University of Edinburgh, which could be made almost 
absolutely dark, were placed two plane mirrors, one of which could 
be rotated, while the other was fixed. The latter was a piece of 
ordinary looking-glass, set at a height of 1 metre from the floor, 
and with its plane perpendicular to the length of the room and to 
the horizon. The other mirror was 25 mm. long by 12 mm. broad, 
and was firmly fixed in a strong brass frame, which was attached to 
an axle gearing into a train of wheelwork, by means of which it 
could be rotated at a very rapid rate round a horizontal axis whose 
direction made an angle of about 45° with the plane of the fixed 
mirror. One of the mirrors was placed at each end of the room. 
