114 
MR P. V. SEVAN ON THE COMBINATION OF HYDROGEN 
compared with h, the period of induction would be correspondingly short. In other 
words, the duration of the induction period will depend on the stability of the 
intermediate compound. 
If, then, there is an intermediate compound formed, a period of induction is to he 
expected. This phenomeijon is observed in many cases, and especially in those which 
require for their finite velocity the presence of a catalyser. For example, the 
hydrogen and chlorine action. The action of bromine on fatty acids.^ The reduction 
of metallic oxides by hydrogen and carbon monoxide.! In these actions the evidence 
for the period of induction is comjfiete, and it cannot be regarded as due to 
disturbances arising from secondary causes unconnected with the action. Secondary 
causes may produce acceleration or retardation of an action, but the period of 
induction proper must be regarded as a characteristic of the action itself This is 
in opposition to the view of Van’t Hoff. He says : “ An acceleration of this kind 
is incompatible with the principles laid down in this book” (‘Studies in Chemical 
Dynamics ’), and he goes on to examine certain actions, and concludes that the 
accelerations observed are to be explained by secondary reactions. The reactions, 
he considers, are the change of rhombic into monoclinic sulphur, the formation ol 
water from hydrogen and oxygen, and the polymerization of cyanic acid. The first 
and third of these actions are probably simple combinations, so that no acceleration 
would be expected other than that due to the action of the monoclinic siilj^hur in 
the one case and the cyamelide in the other. In the case of the formation of water 
the experimental evidence is not sufficient to justify the conclusion that there is no 
period of induction, for this period may be short and its effect observable only in the 
very early stages of the action, which for this particular action have not been observed. 
Tlie very definite and constant character of the induction period in the hydi’ogen 
and chlorine action, which is one of the few actions whose early stages can be 
observed accurately, is sufficient evidence of the essential character of this period, 
and we cannot agree with Van’t Hoff in his conclusion that the existence of the 
induction period is merely “ an indication that some necessary precaution has been 
omitted.”! 
We come now to consider in more detail the hydrogen and chlorine action. The 
rate of progress of this action depends largely on the presence of water vapour. 
The drier the gases are the slower the action is, so that when the gases are dried 
with great care tliey do not combine completely after three or four days’ exposure 
to liright sunlight. It seems, therefore, reasonalde to suppose that under ordinary 
conditions the water vapour plays an essential part in the combination, and that 
without any water vapour present the action would not take place with measurable 
velocity. The fact that illumination of the chlorine alone before mixing with 
* Urech, ‘ Berichte,’ vol. 13, p. 539, 1880. 
t Wright, Luff and Rennie, ‘Journal of Chem. Soc.,’ 1879, p. 495. 
1 ‘Studies in Chemical Dynamics,’ p. 98. 
