74 
mi. V. V. BEVAN ON THE COMBINATION OF HYDROGEN 
H 3 and CI 2 molecules into atoms, which, giving rise to a larger number of systems in 
the gas mixture, causes an expansion. But the immediate effect of such a dissociation 
would not be to increase the volume at constant pressure. For the volume at 
constant pressure is proportional to the sum of the kinetic energies of all the 
molecules; and by a mere dissociation this total kinetic energy cannot be incieased, 
so that the volume would at any rate not increase. In fact, we have 
pv = ^SmnC^, 
where ii is the number of molecules per unit volume of molecular weight ni, and C is 
their mean velocity. Then, in dissociation, tvin is unaltered, and in the most 
favourable case, where no energy is needed for the dissociation, G is unalteied. But 
now the total niimher of molecular systems is increased and the average kinetic 
energy of the molecules is diminished, as in place of certain molecules m with velocity 
C we have molecules with velocity C. The effect of this diminution of average 
kinetic energy is a fall in temperature. We have yet to consider whether energy is 
obtained from the liglit. Bunsen and Roscoe* have shown that the photochemical 
extinction—the light absorbed associated with the chemical action—is very consider¬ 
ably less than the optical extinction due to chlorine alone, and the amount of energy 
absorbed by chlorine alone from light is exceedingly small and not capable of 
producing an expansion at all comparable with that occurring with the mixture of 
gases. (See Pringsheim, he. cit., p. 413, and below, p. 90.) The energy absorbed 
from the light is thus very much too small in quantity to have any appreciable effect 
in the initial expansion. The effect of dissociation would, therefore, be a fall in 
temperature of the mixture ; and an expansion would only follow owing to heat 
being supplied from the walls of the containing vessel. Experiments described below 
(p. 88 ), however, show that associated with the expansion there is a rise in tempe¬ 
rature ; and that for very short illuminations, where there is not time for equalization 
of temperature between the gas and the vessel containing it, the rise in temperature 
is proportional to the amount of hydrochloric acid formed; and in all cases the 
expansion is fully accounted for by the rise 111 temperature obseived. XJndei 
favourable conditions the ratio of expansion to the subsequent contraction, measuiing 
tlie amount of hydrochloric acid formed, may be as much as 7 01 8 , so that in 
Puingsheim’s experiments, where the expansions were 2 millims. on his scale, the 
hydrochloric acid formed may have escaped detection. PringsheimI further con¬ 
cluded that the expansion due to a given quantity of light was independent of the 
amount of hydrochloric acid formed—^independent of the state of the induction. This 
conclusion, however, is not borne out by more careful experiments—in fact, a corre¬ 
sponding phenomenon to the period of induction appears in the case of the initial 
expansion (see below, p. 83). 
* ‘ Phil. Trans.,’ “ Photochemical Researches,” Part III. 
t PiUNGSllEiM, lor. rit., p. 413. 
