438 ‘Mr. Garrett and Dr. Willows : Chemical 
Here is a typical instance from Beattie’s papers, and one 
which we have had under investigation. An air-condenser 
is formed of zine plates; when iodine is placed in it no leak 
occurs beyond the small amount due to defective insulation 
of the supports, although the temperature is raised to 350°. 
All the iodine has long before this disappeared as vapour. 
Common salt alone in the condenser also produces no effect 
when heated to the same temperature; but if the salt is 
sprinkled with iodine and placed on the insulated plate, then 
a large leak of positive electricity takes place when the tem- 
perature is raised. If the mixture of salt and iodine is placed 
on the earthed plate, the sign of the discharge is changed, 
otherwise no difterence is noticed. Bromine water can replace 
the iodine without much affecting the result. 
Experience had shown that Beattie’s form of apparatus 
was troublesome to use and unsuitable for quantitative ex- 
periments. That finally used is shown in fig. 1. 
Fig. 1. 
A is a metal cylinder about 12 cms. long and from 4 to 
6 cms.in diameter. Caps of the same material partially close 
the ends. Cis a wire placed along the axis of A, it passes 
through a hole in a metal plate D and then through ebonite 
plugs ina tube J; the latter is to screen it from external 
disturbances. D serves the same purpose for all parts of the 
apparatus on the left. During the early part of the experi- 
ments the measuring instrument was a gold-leaf electroscope. 
This was connected to C and charged, the case of the in- 
strument, D and A all being earthed. Finally, two hundred 
small cells were obtained and an electrometer was used. 
One pole of the battery was connected to C by means of a 
paraffin switch, the other pole was connected to earth. Both 
al atti aie 
