364 
DR. P. E. SHAW ON THE NEWTONIAN CONSTANT OF 
silver tube, C t, inside AB, has a rosette hole (see fig. 6) cut in it at the window, W. 
Fig. 9. Shows the final form as used with rigid 
suspension. (No wrappings shown.) 
( T X o full size.) 
By opening the rosette carefully it is 
possible to see mirror, S 1? whilst exposing 
little of the inside of the tube. Thus mirror, 
Si should be as far as possible screened from 
any radiometric action and from electro¬ 
static charges on window, W. The actual 
small masses, m, m, are (in the form shown 
in the figure) made of chain, 30 cm. long, 
and are hung from the beam ends by very 
fine wire. 
The materials used throughout the 
system are wholly copper or wholly silver 
(both having very low permeability). The 
metal in each case was supplied by Messrs. 
Johnson, Matthey, in the form of wires 
and foils. They stated that the copper 
was electrically refined, prepared from 
precipitate. If any traces of foreign 
matter exist it would be minute particles 
of gold and silver. The silver should be 
absolutely “chemically pure.” The torsion 
fibre (usually quartz, sometimes phosphor- 
bronze) is 480 mm. long, soldered to the 
beam system below and to the torsion-head 
above by Margot’s solder. # 
5. The Optical Arrangement. — The 
window glass, W, which is of selected 
plate glass, is sealed to the window flange 
by white wax. The window glass has two 
holes drilled in it by means of which a 
mirror, S 2 , can be firmly fixed. This mirror 
which is adjustable in elevation shows the 
azimuth of the whole vacuum tube, and 
therefore that of the torsion-head, whereas 
mirror, Si, shows the azimuth of the torsion 
beam. Thus the difference between the 
two telescope readings given by these two 
mirrors shows at any time the exact amount 
of torsion on the fibre with elimination of any error due to movement of tube or telescope. 
* See “ Sealing Metals,” by P. E. Shaw, ‘Proc. Phys. Soc.,’ Feb., 1912. 
