408 0. N. Rood—Force involved in Crookes’ Radiometer. 
in ordinary daylight, or even in feeble daylight, the repulsion 
was so strong as to cause the discs to assume a parallel posi- 
diverged, right and left, with sensibly equal velocities, an 
after some slight oscillation came to rest. Below are given the 
Aluminium. Glass. 
12°°97 a 
12°°55 8°°43 
aking the mean of the two experiments, and calling n 
gt e 
deviation of the aluminium 100, that of the glass is 65°7. The 
been obtained repeatedly without greater variation ; still, owing 
to slight defects of workmanship in the apparatus, I do not lay 
any great stress on them, further than to prove that both discs 
the light is somewhat less than that of the aluminium away 
from it. In another experiment with a similar apparatus, It 
was found that if the deviation of the aluminium was taken a8 
100, that of the glass was 74°8. 
The result of this experiment, is, as it seems to me, absolutely 
fatal to any theory which assumes that the repulsion 1 4 
Crookes’ radiometer is due to the direct impact of heat or light, 
and I think also it cannot well be explained by assuming the 
' existence of ordinary convection currents. 
pound microscope. It was soon ascertained 
exercised, apparently, an attraction on the suspe peat 
when seated nearly in its prolongation ; it followed an ae ‘ 
toward the observer sitting at a distance of little less a 
meter with considerable promptitude, being thus defl 
3 
S 
Fi 
e 
$ 
be 
. & 
