246 PROCEEDINGS OF SECTION A. 
places, so that the inside could be perfectly screened from external 
action. No difference of effect was at any time noticed upon 
removing or replacing these shutters, so that the screening from 
outside effects must have been sufficiently perfect. 
The induction from the charged parts of the apparatus upon 
the sulphur regarded as a conductor was shown in the course of 
the experiments to have considerable effect, the conductor induc- 
tion upon the insoluble parts being greater than upon the soluble, 
as their conductivity is greater. This source of error was guarded 
against in using the sulphur quadrants, as will be explained later, 
but in using the sulphur needle the disturbing effect was con- 
siderable. For if the semicircles were electrified oppositely, the 
insoluble end of the needle would have its halves oppositely 
charged by induction, the corresponding effect upon the soluble 
end being much weaker, owing to difference of specific resistance 
and specific inductive capacity, as has been stated. The effect of 
induction would therefore be to subject the insoluble end of the 
needle to two pulls in opposite directions. Any want of symmetry 
might cause one of these pulls to be greater than the other, and 
so a deflection might be produced due to induction alone. The 
reversibility of such an effect would depend largely upon the 
rapidity with which induction charges could distribute themselves 
on the needle, and other things that cannot be determined. As 
it is impossible to ensure perfect symmetry, some error is sure to 
occur due to this cause; accordingly the results obtained with 
the needle were unsatisfactory. Upon hanging the ordinary 
needle in position and charging the quadrants, connected as 
described, deflections of over 40° were obtained by induction 
effects, but as the needle was connected with the condenser, the 
case hardly corresponded to that of the sulphur needle. The 
needles were cast in an aluminium mould. To make this two 
pieces of aluminium plate 1 mm. in thickness were taken. From 
one of these a piece was cut out resembling in shape half of an 
electrometer needle. This plate was then sawn across so that it 
could be removed from the sulphur in two pieces. The two 
plates were then clamped together. 
In order to obtain the soluble half of the needle the mould was 
heated gradually on a brass plate with a gas flame till particles of 
sulphur dropped on it began to melt. The flame was then 
removed and sufficient sulphur placed in the mould to fill it to 
the level of the second plate. After the sulphur was melted and 
cooled the pieces of the upper plate were carefully removed. In 
order to remove the sulphur needle from the second plate, the 
latter was heated very gradually until the sulphur could just be 
pushed off by means of a glass rod. The half needle thus obtained 
was heated for some time in an air bath to a temperature of 108° 
C. and then allowed to cool very slowly. 
