MR. W. CROOKES ON REPULSION RESULTING FROM RADIATION. 
113 
(which I will call positive). The lower half of the bulb grasped in the hand causes 
pressure to be exerted on the vanes from below, and the resulting rotation is 
negative. 
435. Ether dropped on the top of the bulb to chill it causes rapid negative rotation. 
If the radiometer is floated in a vessel of ice-cold water, and exposed to the air of a 
warm room above, it rotates rapidly in the positive direction, acting as a heat-engine, 
and continuing so to act until the rotating fly has equalised the temperature of the 
upper and lower portions of the bulb. 
By reversing the cycle of operations—by floating the radiometer in hot water, and 
cooling the upper portion of the bulb—the fly instantly revolves in the negative 
direction. 4 ' 
436. The turbine fly mounted vertically works equally as well horizontally. In this 
form it is supported on a double pointed needle, working horizontally in glass caps. A 
little care is required in balancing the vanes, otherwise one side tends to fall to the 
lowest point, and there is a difficulty in first starting the movement. This windmill 
form is very sensitive to radiation falling on it parallel with its axis of rotation, and it 
has the peculiarity of continuing its movement in any position. When once started 
it may be turned round, upside down, or on either side, without interfering with the 
rotation of the fly so long as the sloping vanes catch the light. 
If mica vanes are used in the turbine radiometer they should be oval, the arms of 
the support passing through the minor axis, so that when viewed from the direction 
in which the light should fall on them their appearance will be circular. 
437. The cause of the movement of the radiometer being pressure between the 
driving surface and the glass case of the instrument, it would follow that, other things 
being equal, the fly should revolve faster in a small bulb than in a large one.t This 
Fig. 21. 
cannot well be tested with two different radiometers, as the weight of the fly and 
other essential points would not be the same in each, but I have constructed a double 
radiometer which shows this fact in a very satisfactory manner. It consists of two 
* Radiometers working in tliis manner were exhibited, with appropriate descriptions, at the soiree of 
the Royal Society, April, 5, 1876. 
f Proc. Roy. Soc., November 16, 1876. 
MDCCCLXXIX. Q 
