114 MR. W. CROOKES ON REPULSION RESULTING- FROM RADIATION. 
bulbs, one large and the other small, blown together so as to have a wide passage 
between them, as shown in fig. 21. In the centre of each bulb is a cup, held in its 
place by a glass rod, and in the bulb is a small four-armed fly with roasted mica vanes 
blacked on one side. The fly can be balanced on either cup, so as to admit of experi¬ 
ments being tried in position A or position B. In the larger bulb there is about half 
an inch between the vanes and the glass, whilst in the smaller bulb there is a space of 
a quarter of an inch. When exposed to the same source of fight under identical 
circumstances, the mean of several experiments shows that in the small bulb (position 
B) the fly rotates about 50 per cent, faster than it does in the large bulb (position A). 
438. [The following experiments show very clearly the influence of an alteration of 
density in the residual gas on the movement of a radiometer. 
A small sensitive radiometer was fitted in a bulb 1\ inch diameter, and a large bulb 
3 inches diameter was connected to the lower limb of the radiometer by a narrow glass 
tube. The whole was then exhausted to about 35 M, and sealed off. 
The large bulb was then placed in an air bath so arranged that it could be raised or 
lowered in temperature without altering the temperature of the radiometer, except by 
convection or conduction through the connecting glass tube. A candle was placed 
3 inches from the radiometer, and observations of speed were taken whilst the tem¬ 
perature of the large bulb was raised or lowered. The mean of several observations 
showed that the speed of the radiometer was always reduced by heating the supple¬ 
mentary bulb, the expansion of the residual gas contained in it causing increased 
pressure in the radiometer bulb. On cooling the bulb, the speed of the radiometer 
increased. Taking the speed at the lowest temperatures at 10 a minute, the variations 
are as follows :— 
Temperature. Revolutions per Minute. 
0 ° 10 - 
15° 10- 
200° 6-9 
300° 5- 
W. C., November 21, 1878.] 
439. Advantage has been taken of the fact, elicited in the experiment described in 
par. 437, in the construction of radiometers having flies of a spiral form.* By cutting 
a thin disk of aluminium into the form of a spiral, then drawing it out corkscrew' 
fashion, and suspending it on a needle point in a tube in the usual way, the spiral 
rotates very quickly on exposure to fight after proper exhaustion. The upper surface 
of the spiral is blacked, and it is kept of as large a diameter as will conveniently go 
in the tube. In this form, as in the turbine form of radiometer, the black surface is 
always exposed to the incident fight instead of being alternately in fight and in 
darkness. The driving surface is greater in this form than in the usual kind of fly, 
* Proc. Roy. Soc., November 16, 1876. 
