ON THE COMEESTION OF GAS FOR ECONOMIC PURPOSES. 
169 
For 12-candle gas 
35 
14 
?5 
35 
16 
55 
55 
18 
35 
0‘44< inch. 
0-48 
0-55 
0-60 
5J 
3) 
All these Argands have the holes the 0‘06 of an inch diameter, and the pressure is only 
0’07 of an inch instead of 0‘17, as with the old Sugg of 0'04 diameter. Above 18 
candles the bat’s-wing is the best burner for educing the light, and it should be regulated 
from 4'5 feet to 4 feet, according to the richness of the gas. And now, before I leave 
this part of the subject, I will show' you some of the contrivances which have been pro¬ 
posed for increasing the illuminating power of a poor gas. 
You have already seen that the single jet gives proportionably less light than the 
double jet or fishtail, and this is because of the larger surface of the flame exposed to 
oxidation. In this experiment, wdien I bring the jets together, you will notice how the 
light is at once increased, the proportion of increase being shown in the diagram. 
Helative Illuminating JPoiver of Jets separate and together. 
Size of Jet, Pressure, Relative value per F'oot Gas. 
Inch. Inch. Separate. Together. 
0067 . . . 0-24 ... 100 ... 164 
0-083 . . . 0-20 ... 100 ... 190 
0-100 . . . 0-12 . . . 100 . . . 184 
But the pressure may be such as to spread out the flame too much, and then it is over¬ 
oxidated. To check this there are the contrivances of Hart, Williamson, and others, 
which are fishtail burners attached to a box stuffed wdth wmol, or having a small aperture 
within, as compared Avith the aperture without. This offers resistance to the flow of 
the gas, and by making it tail a little it thickens the flame and brightens the light; but 
the same effect would also be produced by altering the tap, provided the tap is placed, 
as it always should be, at a distance of about 18 inches from the burner: in fact, if it is 
nearer than this, as is generally the case, there is no space or chamber for the equaliza¬ 
tion of the pressure, and the gas ahvays burns at a disadvantage. 
Again, there are contrivances on the outside of the burners—as caps, and rings, and 
thickenings of the top of the jet—whereby the flow of air to the gas is checked and 
oxidation diminished. 
Even with the Argand burner, if the gas is over-oxidated, as by burning it with too 
large an inner aperture, or w-ith too high a chimney, or at too small a rate, the light is 
improved by checking the draught of air; and this may be done, as you see, by putting 
a cap of wire gauze over the chimney. In fact, the whole of these contrivances have for 
their object such an adaptation of the gas to the air, or the air to the gas, as that the 
flame is just short of smoking. Under these circumstances, the solid particles remain as 
long as possible in an ignited state, and yet at last they are perfectly consumed. 
And now I am anxious to draw your attention to the effect of rarefying the atmo¬ 
sphere, for it has been noticed that the intensity of a flame is much less at high altitudes 
than at low. This was particularly observed by Dr. Frankland and Professor Tyndall 
in the autumn of 1859, when they were making experiments on the combustion of 
candles at the top of Mont Blanc. They noticed that although the candles burnt at the 
same rate as they did in the valley of Chamounix, yet the flames Avere blue, and large, 
and feeble. Dr. Frankland AA^as so much struck with the phenomenon that he afterwards 
made it the subject of careful investigation. He found, indeed, that a gas-flame, like 
that of a candle, gave less and less light with the rarefaction of the air in Avhich it was 
burning; and his results show that the loss of light is about o'l per cent, for every inch 
of diminished mercurial pressure, up to a rarefaction of 14 inches. If, for example, the 
light of a flame be equal to 100 at 30 inches of the barometer, it is but 94-9 at 29 inches, 
and 89-8 at 28 inches; and so on up to 14 inches, when it is only 18-4 per cent, of the 
original light. Fortunately, in our photometrical inquiries the loss of light is equally 
great with the gas and the standard, or the variations of atmospheric pressure from day 
to da)’-, or even from hour to hour, would show a marked difference in the value of the 
light. As it is a variation of 3 inches of the barometer must cause a difference of more 
than 15 per cent.; and it is not improbable that this may have something to do with 
visible variations in the light of the public lamps. Certain it is that the same gas in 
places at different altitudes will have A-ery different values. The gas, for example, which 
