Gas-burners, and on the Illuminating Power of the Gases. 35 
solves itself into two : the first regards their relative economy ; 
the second their comparative utility. 
1. Before we can determine their relative economy, it is re- 
quisite to settle their average quality. Taking their specific 
gravity as the ground of comparison, we apprehend that, in 
small towns, where the cannel coal can be had at a low price, 
coal gas companies may be able to manufacture a gas of the 
density of 700. In larger cities, such as Glasgow arid Edin- 
burgh, where coal of every kind is dearer, and the cannel coal 
cannot easily be procured in sufficient quantity, the average 
specific gravity of the gas will not exceed 600. And, in such a 
town as London, where the cannel coal can scarcely be procured 
at all, the average specific gravity will not exceed 450. 
The average specific gravity of oil-gas should eventually be 
the same every where. It is difficult to ascertain what the average 
is at present, as made by large establish men fg ; but there is no sub- 
stantial cause why it should fall short of 920. We have assigned 
strong reasons, however, for believing that it must be soon improved 
considerably. This improvement, indeed, may be no great gain ; 
for the question will then occur, whether it can be effected 
without diminishing the quantity of gas in the same proportion 
with its increase in quality. It is generally supposed, that an 
improvement in the quality of oil-gas is necessarily attended by 
a loss in quantity ; but, so far as can be discovered, this idea 
rests on experiments performed by operatives only, whose au- 
thority we are satisfied, from repeated observation, can by no 
means be relied on. If charcoal is left in the retorts at the end 
of each charge, it is clear that the gas may be improved by the 
addition of all this charcoal, without any diminution in quantity ; 
for, if it be added to the light carburetted hydrogen, which gives 
little light, so as to convert it into the olefiant gas, which is 
powerfully illuminating, the change, it is well known, will take 
place without any alteration in volume. On the other hand, if 
good oil-gas be exposed to a high temperature, it is partly de- 
composed, and deposites some of its charcoal. Part of the ole^ 
fiant gas becomes light carburetted hydrogen, and without any 
increase in volume ; for the volume is not increased unless it is 
resolved into charcoal and hydrogen. Hence a bad gas may be 
made from oil, which shall not exceed in quantity the good gas 
of Taylor and Martineau. And, in point of fact, we have seve- 
