MR. W. CROOKES ON REPULSION RESULTING FROM RADIATION. 
bulb and fly are at tbe same temperature the movement stops; and as soon as the 
bulb and fly are cooling together, the fly being necessarily a little warmer than the 
bulb (306), the rotation is positive. 
A hot iron ring applied equatorially to the bulb, the screens being in position B, gave 
strong negative rotation, changing to positive on cooling. A hot ring applied to the 
top of the bulb gave very slight negative rotation, also changing to positive on cooling 
(298 to 301). 
397. The screens in the above-described radiometer were found to produce the greatest 
sensitiveness to radiation when in position B (fig. 5). Another instrument was there¬ 
fore made, in which the indications thus given were followed out, the screens being- 
placed in a different position, so as better to obstruct the molecular reaction between 
the vanes and the glass bulb. Fig. 6, A, shows the arrangement. The bright 
Elevation. 
Fig’. G. 
aluminium vanes, a a, are connected by thin arms, and are pivoted on a glass cap and 
needle point in the usual way. The mica screens, b b, are supported on independent 
arms, so that by tapping they can be put in any relative position in respect to the 
vanes a a. The screens are so fixed to the arms that they will pass between the vanes 
and the glass bulb, and be at right angles to them, as shown in the elevation. In 
position A, when exposed to light, or when heated with a lamp, no movement was 
produced. In position B strong rotation was given in the direction of the arrow, and 
when the screen was adjusted to shade oft' the other side, as in position C, equally 
strong rotation was produced in the opposite direction. 
On heating the bulb with the screens in position B or C, strong negative rotation 
was caused, reversing to positive on cooling. 
398. In a preliminary notice"' sent to the Royal Society, November 16, 1876, when 
discussing the action of light upon the cup-shaped vanes of a radiometer, I advanced 
the hypothesis that some of the phenomena might be explained on the assumption 
that the molecular pressure acted chiefly in a direction normal to the surface of the 
vanes; and concluded that “ it would not be difficult to test this view experimentally, 
by placing a small mica screen in the focus of a concave cup, where the moleciflar force 
should be concentrated.” 
* Proc. Roy. Soc., No, 175, 1876, vol xxv. p. 304. 
