272 
MR. W. CROOKES OK REPULSION RESULTING FROM RADIATION. 
repelled. A spirit flame was now brought up about six inches from the opposite side 
of the radiometer, nothing intervening. The rotation gradually slackened, and then 
stopped. On bringing the spirit flame nearer, the heat overbalanced the candle-light 
and the chromic oxide was repelled, whilst on removing the spirit flame further off 
the light overcame the heat, and the selenium was repelled. The balance between 
the heat from the spirit lamp and the light from the candles was perfect, the fly 
obeying the stronger force as delicately as if it were a magnetic needle under the 
influence of two opposite currents. 
257. The very strong action exerted by solution of copper sulphate on the rays 
which are usually most powerful in causing repulsion in a vacuum, has already been 
mentioned (139, 153, 241). The radiometer was exposed to the light of the same two 
candles, a cell containing a saturated solution of copper sulphate, 5 millims. thick, 
being interposed ; rotation in this case was also produced, the selenium being repelled 
with a force only a little less than it would have been behind water or alum. 
A colourless gas flame from a Bunsen burner was coloured intensely green by thallium, 
and the radiometer brought near it. To the eye, by this light, the chromic oxide looked 
nearly white, and the selenium black. The rotation due to the repulsion of the 
chromic oxide was, however, apparently as strong as when the non-luminous flame was 
used. This is a proof that the train of argument I have employed on former 
occasions is correct — viz., that certain substances have an opposite absorptive action 
on rays of dark heat to what they have on light, and that an optically white body may 
be thermically black, and vice versd (170, 171, 172, 246). Here, for instance, chromic 
oxide is optically white and thermically black, whilst scarlet selenium is optically black 
and thermically white. 
258. The experiments with the Bunsen flame show that the behaviour of scarlet 
selenium in a vacuum is closely allied to the variations in the electrical resistance of 
crystalline selenium when exposed to similar flames. During some researches under- 
taken to determine whether the change in the electrical resistance of selenium is due 
to radiant heat, light, or chemical action, Professor W. G. Adams* tried the effect 
of exposing the selenium to a Bunsen burner alone, in its ordinary state, and when it 
is rendered luminous by stopping the air holes. I quote the following passages from 
his printed paper : — 
“ Exposure to the ordinary Bunsen flame for several seconds only caused a slight 
deflection of about 1 0 divisions of the scale. 
“ On making the flame luminous, the needle was suddenly deflected off the scale 
with great rapidity. 
“With the 10 shunt to the galvanometer, there was no deflection on exposure to 
the ordinary Bunsen flame ; but with the luminous flame there was a sudden 
deflection, which increased to 250 divisions of the scale in a few seconds. 
* “ On the Action of Light on Selenium.” Proc. Roy. Soc., June 17, 1875, vol xxiii. p. 535. 
