ME. W. CEOOKES ON EEPULSION EESULTING EEOM EADIATION. 
381 
such action as would show that it followed the law of inverse squares (109). There 
were, however, many reasons why this might not have come out with the apparatus 
then used ; the glass torsion-thread might have been too stiff, or the source of light too 
near ; the pith surfaces were white instead of black, and the vacuum was by no means 
so good as I have subsequently been able to obtain. The experiment described in 
par. 129, where the angle formed by the arm carrying the black and white disks was 
found to vary as the light approached or receded, appeared to me likely to afford valu- 
able information on this point ; and 1 accordingly fitted up more delicate apparatus on 
the same principle. 
133. I wished to suspend the arm carrying the blackened pith in such a manner that 
it should move to the very slightest force, and still return accurately to zero when the 
force ceased to act on it. The principle of Professor Zollnee’s horizontal pendulum * 
seemed well adapted for this ; and I accordingly fitted up an apparatus shown in the 
annexed figure. 
A tube ( a b) about an inch in diameter has two narrower tubes (c d) blown on to it 
Kg. 2. 
near one end, so that they shall be at right angles to the large tube, but not quite in 
the same straight line, the upper tube ( c ) being about a quarter of an inch nearer the 
end a of the wide 'tube. In the wide tube is a straw beam, carrying at the a end a 
disk of lampblacked pith, and at the other end a silvered glass mirror. At e is a plug 
of glass, firmly fixed in the tube c, and carrying a very fine glass thread. In the tube d 
is another similar thread of glass, having at the end a weight made of glass tube and 
mercury. The two threads are firmly fastened to the straw beam, behind the mirror, in 
such a manner that the upper thread in c holds the beam a quarter of an inch nearer 
the pith end than the lower thread in d holds it, as shown in the enlarged view. By 
adjusting the tension on the glass fibres, the beam can be kept in a horizontal position 
* PoG-Gr. Ann. 1873, vol. cl. pp. 131, 134. 
