Gas-Burners , and on the Illuminating Power of the Gases. 11 
ford himself ; and he is abundantly borne out in it by the observation of 
others. 
We may refer, for example, to some experiments on the illuminating power 
of oil and coal gas, contained in an elaborate and interesting paper by Dr Fyfe 
of this city, ( Edin . Phil. Journ. xi. 370.) They are the last that have been pub- 
lished, and almost the only ones of which the particulars are faithfully record- 
ed. There are five double observations, made on different days, with the 
same burners, and the same gases. The burners employed were two of the 
Edinburgh Coal-Gas Company, with 10 and 14 holes; one of the Glasgow 
Company with 10 holes ; and a 10-holed and 14-lioled oil-gas burner. The 
expressions procured for the light of a given quantity of gas, say a cubic foot, 
in relation to that of a (short-six) candle, making its light equal to 10, and 
neglecting decimals, are the following. For the 
Edin. 10 holed coal-gas burner, 38 in one, and 40 in the other experiment. 
14 
66 ... 
... 43 
Glasg. 
10 
72 ... 
... 59 
Edin. 
10 
... oil -gas burner, 68 ... 
... 42 
14 
78 ... 
... 56 
The first experiment is good. But, with that single exception, it seems quite 
impossible to strike an average between observations so very discordant ; and 
there cannot be a doubt, that the discrepancy must have arisen in a great 
measure from the impossibility of making the standard light burn uniformly. 
It is not a little extraordinary that all, or almost all who have tried this me- 
thod of determining the relative light of the gases, have shewn so little regard 
to the caution given by Count Rumford. 
The standard we have invariably used for a train of comparative experi- 
ments, was a gas-jet of a certain length. — In order to preserve its length uni- 
form, we had a gasometer constructed according to the principle described by 
Biot, in his Traite de Physique , and originally conceived by Girard, for sup- 
plying a uniform current of oil to the wick of Argand’s lamps. By means of 
the principle alluded to, water is made to drop in a steady stream from an up- 
per vessel into a lower one containing the gas. 
In its simple state this apparatus was not quite fit for our purpose, because 
the concussion caused in the lower vessel by the dropping of the water, and con- 
veyed along the exit-tube, produced a jumping or flickering flame, which ren- 
dered it impossible to compare the shadows with nicety. But the inconve- 
nience was remedied in the following manner. In the original instrument, the 
end of the tube from which the water drops, is bent a little upwards, to prevent 
the gas from ascending into the water-vessel. In our apparatus it was straight, 
and terminated near the bottom of a little cup, from the middle of the side of 
which a tube, somewhat wider than the other, proceeded downward, to open 
near the bottom of the gasometer. By this contrivance, after the water rose 
to the level of the lateral hole in the cup, all that entered flowed gently down 
to the bottom. The gasometer held a cubic foot and a half ; and the whole 
apparatus was so accurately made, that a gas-jet of three or four inches burnt 
from beginning to end without varying above a 20th part of an inch in its 
length. In order to keep this flame steady, and detect any accidental varia- 
