MR. W. CROOKES ON REPULSION RESULTING FROM RADIATION. 
281 
269. The above experiment was repeated, the only change being that the plates 
were folded across their horizontal diagonal, the crease being in a line with the 
supporting arm. The movements to a candle and a hot shade were the same as in 
the last experiment (268), but considerably weaker. 
270. Flat plates were now used, attached to the arms at an angle of 45°, as shown 
in fig. 10. They were blacked on the insides, away from the bulb. Exposed to 
candle light, there was slight repulsion on each face, that on the bright surface being 
rather the stronger, but not enough to commence rotation. On starting rotation by 
shaking the bulb, it kept up very slowly in the direction of the arrows. A hot shade 
gave good rotation in the same direction. There was no reverse movement on cooling. 
271. As in the last experiment (270), the plates were mounted at an angle, but 
they were blacked on the outsides as shown at fig. 11. A candle gave rapid positive 
Fig. 11. 
rotation. A hot shade gave negative rotation with good reversal of movement on 
cooling. This form is the most sensitive yet experimented with. 
272. Radiometers made of silver flake mica being more sensitive than those made 
of pith or metal (238), it was deemed of interest to ascertain whether this extra 
sensitiveness would be retained when the vanes were mounted at an angle, as in the 
last experiment. This was found to be the case ; radiometers constructed with silver 
flake mica vanes set at an angle, as shown in fig. 11, and blacked on the outside, 
prove the most sensitive for light hitherto constructed. 
SPECIAL EXPERIMENTS WITH SLOPING-YANED RADIOMETERS. 
273. From the foregoing experiments it appears that the favourable presentation ol 
the surface of the vanes to the inside of the bulb, has more influence on the movement 
than has the colour of the surface. A series of experiments was instituted in the 
hope of clearing up many anomalous results which had attended the application of 
heat either by hot shades or by hot water to radiometers. The results hitherto 
obtained had been contradictory, and it was now seen that there might be an 
antagonistic action between the effect of shape and that of colour of surface, the two 
actions sometimes acting together and sometimes in opposition. 
Five radiometers were made exactly alike in size of bulb, shape of vanes, and 
degree of exhaustion, and only differing in the material of which the vanes were 
composed. No. 1 was made of mica, 0'003 inch in thickness.'"’ No. 2 of mica, 
* Measured by a Wbitwortb’s measuring machine. 
2 o 
MDCCCLXXVIII. 
