ON THE GASES EVOLVED FROM IRON FURNACES. 149 
in the eudiometer above that in the trough, so that the difference may be 
brought into the calculation. By reading in this manner the error is avoided; 
which otherwise would result from heating the gases by the hand in adjust- 
ing the outer and inner levels, and it also enables us to record the results 
without touching the apparatus, which thus preserves a constant temperature. 
It is quite necessary, in estimating the volumes of the gases, to use the sub- 
stances for absorption in a bulk as small as possible, and in a form which 
may easily and completely be removed from the tubes, so that the gases may 
neither be rendered impure by air introduced, nor their reading rendered 
erroneous by some of the absorbing substance adhering to the sides of the 
tube. This is best effected by casting the materials into the form of bullets, 
by means of a common bullet-mould, into which a thin piano wire hasbeen 
previously introduced. If there are to be two determinations of carbonic 
acid, the one before the combustion of the gas, the other after, it is neces- 
sary to transfer the gas from one eudiometer to another, after the first deter- 
mination, in order to avoid the chance of error which might result from potash 
adhering to the side of the tube during the first absorption ; and for this 
purpose it is obviously of little consequence whether the whole or only a part 
of the volume of the gas be transferred. The adhesion of air to the piano 
wire is so insignificant, that it might be completely neglected ; but to avoid 
error, it is better to amalgamate the outer surface of the iron wire; this may 
be done by rubbing it with an amalgam of potassium and mercury, without 
destroying its tenacity. Rusty iron wire must not on any account be em- 
ployed, and equal care must be taken to keep its inferior end under the mer- 
cury during the absorption; for if it be exposed to the air, an endosmose 
and exosmose is effected to such an extent, as in certain cases to endanger 
the value of the analysis. 
In order to estimate olefiant gas and the hydrocarbons accompanying it, 
we have invented a very simple and efficacious method, which may be use- 
fully employed in the analysis of coal-gas. A little bullet is prepared out of 
the same materials as those used for making the negative element of the coal 
battery. For this purpose a bullet-mould, supplied with a platinum-wire 
having a bent end, is filled with a pounded mixture of two parts of coke and 
one part of coal, and is then heated before the blowpipe flame. The ball 
made in this way is afterwards dipped into a concentrated solution of sugar, 
and heated very strongly in the open reducing flame of the blowpipe ; it is 
now ready, and must be preserved for use carefully protected from moisture. 
This lump of charcoal, about the size of a small pistol-bullet, is capable of 
absorbing into its pores 0°5 gramme of sulphuric acid without appearing wet 
on the surface, and it can be introduced into and withdrawn from the eudio- 
meter without moistening it to any appreciable extent. For the purpose of 
experiment, it is made to absorb a mixture consisting of one part of anhy- 
drous and two parts of concentrated hydrated sulphuric acid. The proof 
that the acid contained in the bullet has been sufficient for the absorption of 
the olefiant gas, is the emission of white fumes in the air after its withdrawal 
from the mixture of gases, which of course must be quite dry. As the an- 
hydrous sulphuric acid emits vapour possessing considerable tension, and is 
never obtained free from sulphurous acid, and as the latter gas is also formed 
by the action of sulphuric acid on the hydrocarbons, an augmentation of the 
volume of the mixture is thus produced. ‘To remove both these sources of 
error, after the conclusion of the above experiment, a little dry ball made of 
gypsum and peroxide of lead is introduced into the eudiometer. This has 
the double effect of removing both, for while the peroxide of lead absorbs 
