456 APPENDIX 
varnish between the lead and the stopper, which was of 
cork, no varnish coming between that and the neck. The 
electrometer was divided into thirty parts of 14 inch. 
The plate at first refused to go round, as mine had done 
before, the cushions being drawn together by the glass, to 
which they seemed to adhere, probably from their dampness. 
After some time, however, this went off, and in about ten 
minutes electricity was excited. 
The electrometer was then applied and went off at 7. 
2. Electricity was kept in the phial thirty seconds with- 
out any appreciable quantity being lost. 
3. A hole was struck through two cards by the discharg- 
ing wire. 
4. Much greater shocks were given to several people 
than any that could be given by my machine. 
5. The phenomenon of the floor-cloth proving a con- 
ductor was tried more fully than before. A wire (0, see 
figure) was passed through 
the phial (a), the two ends 
of which were taken hold 
of by two people (cc), who 
each took hold of another 
person (dd); the operator 
(e) then touched the phial 
with his discharging wire, 
and received the shock 
through both arms, as did 
(ec)and (dd). Sometimes, 
however, the others (d d) 
felt it only in the arm 
by which they held (¢c). 
The comparative force of 
the blows which each felt 
were difficult to ascertain, 
but we supposed that (cc) 
felt more than (dd), and probably the operator most of all. 
The chief reason that this machine worked better than 
mine seems to be that the bottle was coated with varnish 
