MR. W. CROOKES OK REPULSION RESULTING FROM RADIATION. 
285 
fly carries sufficient driving power to enable it easily to pass the dead centre opposite 
the candle. Therefore, as soon as the candle has shone on the aluminium radiometer 
long enough to warm the vanes a little, rotation readily continues. 
279. The action of the pith radiometer is similar to the aluminium. On refer- 
ring to Table XII., par. 237, it is observed that white pith is repelled with a force of 
17'7 by rays which are not cut off by water ; that is, by dark heat rays which have 
escaped the absorbing action of the side of the glass tube in which the experiment 
was tried. The rays quenched by the pith will reappear as heat, and the slice of pith . 
being of a sensible thickness, and almost a non-conductor, nearly all the molecular 
pressure will act on the flrst surface. The action of the candle on the pith radio- 
meter will therefore be similar to that on the aluminium, represented in fig. 14, 
except that the dissipation of pressure from the back surface of the pith will be almost 
nil. The pith, moreover, being sensitive to the heat rays, and being a non-conductor, 
may be expected to move quicker than the aluminium, which requires time to get 
warm throughout. This is found to be the case. 
280. The aluminium vanes blacked on both sides, as might be expected, act like the 
plain aluminium vanes, but in a much stronger degree. 
281. The agreement between theory and observation, so far, seemed exact. In the 
endeavour to clear up the discrepancies I had so frequently met with, I now tried 
numerous experiments with dark heat applied in various ways to these five radio- 
meters. 
282. In par. 245 it is shown that, when I wanted to try the effect of dark heat on 
a radiometer, I covered it with a hot glass shade, a convenient means of warming 
the bulb uniformly by heat rays which would not pass through the glass in a radiant 
form, and thus avoiding the interference of other rays on the material composing 
the fly. On trying this experiment with the five sloping-vaned radiometers, the 
results proved very contradictory (273). Generally the effect of the hot shade was to 
produce negative rotation. Sometimes, however, the movement was positive, and 
occasionally no rotation at all could be obtained. The vanes would go round once 
or twice, then quickly stop, and reverse their movement, as if under the influence of 
two nearly equal opposing forces, sometimes obeying one and sometimes the other. 
Thinking that the hot shade might have become electrified on handling, I heated it 
in hot water, and inverted it dripping wet over the radiometers, still with the same 
contradictory result. 
283. To immerse the radiometers in water of about 70° C. appeared a better means of 
heating the bulbs uniformly on all sides, without producing electrical disturbance. 
Heated in this manner, the thick mica radiometer moved slowly in the negative 
direction, and after a short time came to rest. 
The radiometer was now removed from the hot water, and allowed to cool. Positive 
rotation commenced, and after continuing for about a minute it stopped, and then 
gave one or two turns negatively. 
