Mr Stevenson’s Hxperiments on Lights in Rapid Motion. 273 
ing a much more brilliant appearance, by the compensating 
influence of the bright flashes, which he expected would pro- 
duce impulses sufficiently powerful and durable to make the 
deficiency of light in the dark spaces almost imperceptible. 
The mean effect of the whole series of changes would, he ima- 
gined, be thus greatly superior to that which can be obtained 
from the same quantity of light equally distributed, as in fixed 
lights, over the whole horizon. Now this expectation, if it be 
considered solely in reference to the physical distribution of 
the light, involves various difficulties. The quantity of light 
subjected to instrumental action is the same whether we em- 
ploy the refracting zones at present used in fixed dioptric 
lights, or attempt to obtain continuity of effect by the rapid 
revolution of lenses; and the only difference in the action of 
these two arrangements is this, that while the zones distribute 
the light equally over the whole horizon, or rather do not in- 
terfere with its natural distribution, the effect of the proposed 
method is to collect the light into pencils, which are made to 
revolve with such rapidity, that the impression from each 
pencil succeeds the preceding one in time to prevent a sensible 
occurrence of darkness, To expect that the mean effect of 
the light, so applied, should be greater than when it is left to 
its natural horizontal divergence, certainly appears at first to 
involve something approaching to a contradiction of physical 
laws. In both cases the same quantity of light is acted upon 
by the instrument, and in either case any one observer will 
receive an impression similar and equal to that received by 
any other stationed at a different part of the horizon ; so that 
unless we imagine that there is some loss of light peculiar to 
one of the methods, we are, in the physical view of the ques- 
tion, shut up to the conclusion, that the impressions received 
by each class of observers must be of equal intensity. In other 
words, the same quantity of light is by both methods employed 
to convey a continuous impression to the senses of specta- 
tors in every direction, and in both methods equality of dis- 
tribution is effected, since it does not at all consist with our 
hypothesis that any one observer in the same class should re- 
ceive more or less than his equal share of the light. Then, as 
to the probability of the loss of light, it seems natural to ex- 
