72 CHEMISTRY: BAXTER, GROVER AND THORVALDSON 
In order to test the purity of the lead salt, metaUic electrodes were 
prepared from the purified material, and the ultraviolet spark spectrum 
of the metal was photographed by means of a Fery quartz spectrograph. 
The only lines which could be detected were those commonly ascribed 
to lead. In fact the original material was fairly pure, containing only 
traces of silver, copper, and iron, which rapidly disappeared in the 
course of the purification. Incidentally we found that 0.004% of copper 
in lead yields fairly strong copper lines and that 0.0001% of silver could 
readily be detected. Our material was further tested in a sensitive gold 
leaf electroscope, but showed no evidence of radioactivity. 
Several difficulties were met which deserve special attention. Lead 
bromide, when dissolved in water, hydrolyses to give an insoluble basic 
bromide. This phenomenon was first met with material which had been 
fused in an atmosphere containing a large proportion of hydrobromic 
acid gas, but it was also found that even unfused salt which has been 
crystallized from concentrated hydrobromic acid solution, and which 
probably contains a small amount of the latter substance, hydrolyses 
in the same way, and, apparently to about the same extent. Thus it 
was evident that at any rate the greater part of the hydrolysis occurred 
during the solution of the salt and not during the fusion. The evidence 
that no hydrolysis with loss of bromine occurred during the fusion is 
that the different samples of fused salt, in spite of the widely varied 
atmospheres in which they were fused, were essentially identical in com- 
position, and that the atomic weight of lead found from the bromide 
agrees very closely with that found later in this research by analysis of 
the chloride. 
In order to prevent the basic salt from forming during the solution 
of the bromide, a small quantity of nitric acid was added to the hot water 
in which the salt was dissolved. It was proved experimentally that the 
nitric acid added was far too small in amount to cause loss of either 
bromine or hydrobromic acid from the solution. 
A second difficulty was caused by the presence of a small amount of 
black insoluble material in the fused salt. Although puzzling at first, 
this was eventually found to consist partly of carbon, partly of silica. 
The greatest care in the purification of the material failed to eHminate the 
trouble entirely, although the proportion of insoluble residue was re- 
duced on an average to 0.004%. Error from this source was avoided 
by filtering the lead bromide solution through a weighed platinum- 
sponge crucible in each analysis before precipitation of the silver bromide, 
and weighing the residue. 
