1044 
THE PHARMACEUTICAL JOURNAL AND TRANSACTIONS. 
[June 28, 1873. 
to saturate an amount of material to supply a light for 
nine hours. 
It has often been thought that of a necessity these very 
volatile liquids must be much more liable to explosion 
than those of higher boiling-points; such, however, is not 
the case, not but that it is possible to cause them to ex¬ 
plode, because the vapour of any combustible hydrocarbon, 
when mixed with that proportion of air which contains 
just sufficient oxygen to burn up the hydrogen and carbon, 
will burn throughout at once, producing thereby sudden 
expansion, which is nothing more than explosion; but the 
conditions necessary in order to obtain this result are more 
difficult to bring about in the case of these heavy vapours of 
very volatile bodies, as, just above the explosive point, the 
air and vapour simply inflame and burn, while just below 
it inflammation will not occur at all, from the presence of 
too small a quantity of vapour. This is contrary to the 
case of hydrogen gas, which explodes with ease when 
mixed with air in very variable proportions. The fear has 
been expressed that possibly, even after the complete 
mixture of such a heavy vapour and air, separation other 
than condensation to the liquid state might occur, that is 
to say, that a layer of heavy vapour might lie at the 
bottom and a layer of the lighter air at the top ; but the 
law of the diffusion of gases, which states that gases and 
vapours, whatever their differences in density, will become 
and remain completely mixed, invariably prevents such 
separation, just as in the atmosphere the heavier oxygen 
is not found at the bottom, and the lighter nitrogen at the 
top, but the two are found together in practically the 
same proportions from wherever a sample of air is ob¬ 
tained. 
If comparisons be instituted between gases made in the 
ordinary manner and these air-gases, they should be made 
under the same circumstances, and if this is done it will 
be found that coal-gas itself suffers, either on exposure to 
a .low temperature or by long keeping—not in any way a 
surprising result, since it contains the vapours of benzole, 
toluole, etc., which are capable of being condensed to 
liquids. In this respect rich gases always suffer the most; 
thus, on continued exposure to a temperature of 0° C., or 
32° F., gas made from cannel coal loses much more of its 
illuminating power than gas made from ordinary coal. 
The employment of a combustible gas as the vehicle for 
holding in solution, and carrying, vapours of volatile 
liquids, appears at first sight to be a more rational, and, 
probably, a more successful method of applying the prin¬ 
ciple of carburation, for in the former cases not only has a 
sufficiency of vapour to be retained to confer luminosity 
upon the flame, but a further quantity must be present 
for the production of the flame itself, or we may express 
this difference by saying that in the one case the combus¬ 
tible gas is already manufactured, and requires only to be 
endowed with luminous properties by the hydrocarbon va¬ 
pour ; in the other the combustible gas has itself to be 
manufactured, and then, also, to acquire luminosity from the 
same hydrocarbon vapour, and therefore it will be perfectly 
■clear that a non-combustible gas will require to hold a 
larger quantity of vapour in solution than a combustible 
gas, in order to attain the same amount of lighting power. 
This statement does not, of course, deny the possibility of 
a non-combustible gas, such as air, acquiring and holding 
a sufficient amount of vapour, but all other things being 
equal, a combustible gas has a distinct advantage in this 
respect, and probably such a gas might be carburetted 
with a hydrocarbon liquid, having a lower boiling-point, and 
hence less volatile than an incombustible gas, such as air. 
This advantage is also increased when air is employed as the 
vehicle, from the presence in it of one-fifth of its volume of 
oxygen which burns up—without other effect than increas¬ 
ing the heat of the flame—a proportionate quantity of 
vapour. The nature of the combustible gas used is not of 
any very great importance, and hence cheapness and eco¬ 
nomy in production become the chief guides for its manu¬ 
facture. The possibility, without any great amount of 
trouble, of obtaining a quantity of hydrogen gas from 
water has always been a favourite theme with inventors, 
and many schemes for its practical utilization have been 
introduced ; but one hardly expected to recognize in a 
“ new gas ” our old friend this water-gas once more ; such 
is, however, the case, and a patent has lately been ob¬ 
tained for the carburation of this gas with some light 
hydrocarbon liquid. The plan is being carried out upon a 
small scale, and will doubtless very soon be tried upon a 
more extended one. The amount of attention that has 
been at various times bestowed upon this subject may be 
gathered from the number of patents obtained for the pro¬ 
duction of this so-called water-gas. The first appears to 
have been granted to Michael Donovan in 1830, a second 
to G-eorge Lowe in 1831, a third to Gilbert Saunders in 
1833, and two others to Mr. Floret and Jean Baptiste 
Molerat about the same time, and many others, all differ¬ 
ing by some slight modifications, but all dependent upon 
the decomposition of steam by red-hot coal, coke, or metal. 
This decomposition takes place with very great ease when¬ 
ever steam is passed over such heated material. Hydro¬ 
gen, carbonic oxide, and carbonic acid being produced, 
probably two different actions take place, as expressed by 
the following equations :— 
h 2 o + c=h 2 + co 
2H 2 0 + C = 2H 2 -f C0 2 
The amounts of hydrogen, carbonic oxide, and carbonic 
acid obtained are variable under different circumstances. 
An analysis of the gas obtained by passing steam over 
red-hot charcoal for some hours gave as a mean— 
Carbonic acid. 20 per cent. 
Carbonic oxide . 20 „ 
Hydrogen . 60 „ 
100 
Any form of carbon is capable of effecting this decompo¬ 
sition. If a metal, such as iron, be present, it adds to the 
quantity of hydrogen obtained according to the following 
action :— 
4H 2 0-f 3Fe = Fe 8 0 4 + H 8 
This mixture of gases is inflammable, and burns with a 
non-luminous but hot flame, the heat of which would be 
increased if the carbonic acid were previously removed, 
the flame of hydrogen gas having a temperature of 
3776° F. This heat of the flame is an undoubted advan¬ 
tage when the gas is carburetted, as the particles of car¬ 
bon are thereby heated to whiteness ; but, on the other 
hand, the presence in the gas of a quantity of carbonic 
oxide is as great a disadvantage, on account of its inju¬ 
rious and exceedingly harmful character. Carbonic acid 
is an injurious gas, but its action upon the system is not 
that of an active poison, but rather that of preventing the 
necessary amount of oxygen from entering the lungs, and 
hence death from its effect will be death from suffocation. 
On the other hand, carbonic oxide is an active poison, and 
recovery from its effect is much more doubtful; indeed, 
it is calculated that 2 per cent, in any atmosphere would 
prove fatal, a result that should cause great caution in the 
use of a gas containing it in any quantity. 
In this recent process it is proposed to use retorts of the 
ordinary description, set, as usual, in benches of five, 
seven, or nine. All, except the upper one of each set, are 
filled with some carbonaceous substance, usually coke, 
together with some scrap-iron, or iron chairs, and heated 
up to a comparatively high temperature; steam—under 
pressure, generated from an ordinary boiler, and super¬ 
heated by its passage through pipes set in and heated by 
the same furnace as the retorts—i3 introduced into the 
back of each retort, and has to traverse its whole length, 
passing through the red-hot coke and issuing from the 
front. During this passage the above decompositions 
take place. The upper retort is of a larger size, and is 
also filled with coke, but no steam is introduced into it; 
the gases formed and the remaining undecomposed steam 
from the other retorts passing through it together, this 
