628 
MR. H. B. DIXON OX CONDITIONS OF CHEMICAL CHANGE 
taken by '78 volume. By an analysis of the residual gases the liberated hydrogen 
was directly estimated, and its volume and was found to be exactly 78 volume. 
Volume of residual gases = 1970 
,, On addition of oxygen = 51*12 
x „ After explosion = 40‘49 
„ After absorption = 21‘58 
CO = 18*91 
IB = | {10-63 - 1^1} 
= *78 
It was evident from these experiments that the results obtained with carbonic oxide 
and electrolytic gas were affected by the presence of the aqueous vapour. Since at the 
temperature reached in the explosion the excess of carbonic oxide reacts with the 
steam, forming carbonic acid and liberating hydrogen, the presence of aqueous vapour 
appears to increase the affinity of carbonic oxide for oxygen and diminish that of 
hydrogen. Such an effect is exactly the opposite of that produced by adding carbonic 
acid to the mixture before explosion. Bunsen- showed in his first paper that when an 
excess of hydrogen is exploded with oxygen in presence of some carbonic acid, some of 
the latter is reduced by the excess of heated hydrogen to carbonic oxide. Thus the 
previous addition of aqueous vapour, one of the products of the reaction, alters the 
apparent division of the oxygen, just as the previous addition of carbonic acid, the other 
product of the reaction, alters it in an opposite direction. This fact accounts in part 
for the discrepancies observed in comparing my experiments with Bunsen’s, for while 
my explosions were made at an initial temperature varying between 15° C. and 17° C., 
Bunsen made one set of explosions between 22° C. and 23° C., and the other set between 
2° C. and 5° 0., so that very different quantities of aqueous vapour were present in the 
different experiments. In order, therefore, to make the experiments a real test of the 
law of mass, it was necessary to perform the explosions with dry gases in a dry 
eudiometer. 
Before proceeding to repeat the experiments with dry carbonic oxide and electro¬ 
lytic gas, another attempt was made to determine whether, in a dry eudiometer, 
oxygen is completely burnt when exploded with a large excess of carbonic oxide. 
This experiment led to the important discovery that dry carbonic oxide and oxygen do 
not combine when submitted to the electric spark. A mixture containing 3 volumes 
of carbonic oxide to 1 of oxygen was brought over into the dried eudiometer, and a 
spark from a Leyden-jar was passed through it without causing explosion. A little 
more oxygen was added, and the spark again passed without result. 
A fresh charge of the carbonic oxide, prepared from recrystallized oxalic acid, was 
next brought into the dried eudiometer and mixed with an excess of oxygen. The 
following numbers taken from my laboratory note-book are the measurements made 
in this experiment with McLeod’s form of gas analysis apparatus. 
2nd contraction = 10‘63 
2nd absorption = 18-91 
