282 BUNSEN ON THE CACODYL SERIES. 
which result, afford a striking example of accordance between 
the laws of organic and inorganic bodies. This harmony may 
not, perhaps, be without some influence in a science where 
every idea is supported by analogical reasoning. In conse- 
quence of the difficulties which attend the analyses of these sub- 
stances, it appears to me necessary that I should at first make 
some general observations on the nature of the analytical me- 
thods employed in the following researches. 
Nitric acid oxidizes arsenic only imperfectly, but it is per- 
formed, without loss, by means of a red heat. Neither oxide of 
copper, nor such bodies as give off oxygen when heated, can 
be employed as the substance for combustion, because in the 
first case, the separation afterwards of the arsenic and copper 
is attended with insuperable difficulties, and in the other it 
is impossible to avoid the most dangerous explosions. I have 
therefore employed a fused mixture of Glauber-salt, bisulphate 
of soda and powdered glass. The experiment even succeeds 
better by using oxide of zinc or oxide of nickel free from arsenic. 
In preparing the latter in the best form for analysis, the sulphate 
should be used, which loses its acid without melting at a white | 
heat, and leaves the oxide behind in the form of a very fine powder, 
which is much more bulky than the oxide of copper prepared 
in a similar way from the nitrate. When the nickel of com- 
merce is used, a current of sulphuretted hydrogen gas ought to 
be passed through the warm sulphuric acid solution at least eight 
or ten days, as it usually contains molybdenum, which is thrown 
down with great difficulty by this reagent. The estimation of 
the carbon and hydrogen cannot be accomplished at the same 
time by means of this substance, as it only perfectly oxidizes the 
arsenic. In order to avoid any loss, the combustion tube ought 
to extend some inches out of the furnace, and be connected with 
a Liebig’s condenser filled with water. The insupportable smell 
of alcarsin which is absorbed by the fluid in the condenser, 
quickly indicates an imperfect combustion of the arseni¢, which 
however seldom happens in a well-executed experiment. The 
contents of the combustion tubes are then dissolved in aqua 
regia by a gentle digestion, sulphuric acid added, and the whole 
evaporated to get rid of the nitric acid, again heated with water 
and filtered. The solution thus obtained contains the arsenic, 
which can be determined in the usual way by means of sulphu-— 
retted hydrogen. 
