Dr Fyfe on the Illuminating Power of Coal-Gas, §c. 225 
light was as 2.81, to that of the candle as 1; condensation by 
chlorine 15. Gas D. the light was 2.27, chlorine test 12, 
and as 2.81 : 2.27:: 11: 8.02 
andeas,. b> )°ais ys 1-+ 8.00, 
which is a very close approximation. 
Two gases were next contrasted with each other, being 
consumed with fish-tail burners. By the shadow the light for 
equal consumpt was 1 to .827, by the chlorine, 14: 12, and as 
14:12::1:.857. In another trial with the same burners, but 
with gases prepared at another time, the average of numer- 
ous trials by the photometric process, gave the result as 1 to 
.945; condensation by chlorine was 12.5 and 11.5, and as 
GV :s: 92. 
With jets and with other gases, the results were by the 
shadow 1 to 1.185, and by chlorine 11 to14, and 11:14::1: 
1.272. Here the approximation is not so close as in some of 
the others. 
The chlorine test was then tried with a gas, the illuminating 
power of which was inferior to that of the preceding. The 
trial by the shadow was made at different distances, to secure 
accuracy. By the one the result was as 1 to 1.347, by the 
other to 1.338, average 1 to 1.342. The condensation by 
chlorine was 10 and 14, which very nearly coincides with the 
others. 
The results above stated, very nearly agree with each other. 
In one trial, however, [ found that they did not come so close. 
By the shadow they were 1 to 1.33, by the chlorine 11 to 17, 
now as 11:17::1: 1.54. 
In this instance the discordance may, I think, be accounted 
for. It is well known that when the illuminating power of 
a gas is high, as when it is prepared by the decomposition of 
oil, it requires a burner with smaller apertures than those used 
for common coal-gas, otherwise it is not consumed to advan- 
tage. Now, in the experiment last recorded, in which the 
condensation by chlorine amounted to 17, a coal-gas jet was 
used, by which the gas would not give the same amount of 
light that it would have given, had a burner with smaller 
apertures been employed. Hence the illuminating power in- 
dicated by the shadow dves not come up to what most: likely 
