MR. W. CROOKES ON REPULSION RESULTING FROM RADIATION. 
91 
the movement under the influence of radiation is similar to that observed when a 
mica screen was fixed on each side of the blacked disk, the rotation being slow, or 
altogether stopping if the candle is five or six inches off. 
When, by tapping, the disks are removed to about 1 millim. apart, as in position B, 
the sensitiveness is about at the maximum, and the rotation is in the negative direc¬ 
tion, the same as in the radiometer shown in fig’. 1. 
o 
The speed of rotation in the positions between B and C gradually diminishes as the 
plates approach the latter position ; and in position C, the plates being about 7 millims. 
apart, the sensitiveness has vanished, no rotation taking place even in a strong light. 
Between 7 millims. and 12 millims. apart there is an almost total absence of sensi¬ 
tiveness. Beyond 12 millims. apart the repulsion begins to act most on the black 
surface, causing positive rotation, as shown in position D. When the two arms are at 
right angles to each other, positive rotation is strong, and it increases in speed as the 
clear disk approaches the plain side of the blacked disk; the speed being at its 
maximum in position E. 
393. I am now able to decide between the two hypotheses advanced in par. 390. 
The result obtained when the screen is close to the black surface (fig. 4, A), in which 
the speed was considerably less than when they were 1 millim. apart, shows that the 
negative rotation obtained in these positions is not due to a molecular disturbance 
generated on the plain side of the blacked disk, and reacting directly between that and 
the glass case (the second hypothesis), but that it is caused by the warming up of the 
black surface by radiation falling direct on it through the clear screen, and the 
deflection backwards of the lines of molecular pressure thereby generated. At 1 millim. 
apart, sufficient of this pressure is deflected to produce good negative rotation; at less 
distances much of the force is taken up in molecular beating to and fro between the 
disks ; while at greater distances the negative rotation due to the reflected force 
is more or less counterbalanced by the positive rotation due to the molecular impacts 
exerted directly between the black surface and the glass bulb. 
The cause of the maximum positive rotation taking place when the clear screen 
is close to the plain side of the blacked disk is, that in this position the difference of 
temperature between the outer sides of the compound disk is greatest. 
394. The action of these radiometers is somewhat complicated, owing to the surfaces 
of the fixed disks being different in absorptive power. Another instrument was 
therefore made in which the vanes were of polished aluminium, perfectly flat and 
symmetrical to the bulb. The screens were of clear mica, and instead of being parallel 
to the vanes, as in the former instruments, they were at right angles to it, as in fig. 5. 
The aluminium plates are shown at a a ; the mica screens are at b c, they are fixed 
to their arms at a slight angle, so that in position A the screens touch the centre 
of the vanes at c, and project from them at right angles, whilst in position B the 
screens touch the vanes at b, near the outer edge, meeting them at an acute angle. 
The arms carrying the screens move easily on the centre cap, as in the previously- 
N 2 
