AND CHLORINE UNDER THE INFLUENCE OF LIGHT. 
101 
be intermediate between that of the fresh gas and that when the light is kept on 
continuously while fresh gas is added. 
The effect of the added uninsolated gas thus persists for some time and makes 
itself evident by the prolongation of the acceleration j^ei'iod. This is shown still 
more clearly on curves for the velocity of the action. In fig. 15 the curves I, II, III 
are the velocity curves corresponding to the curves in fig. 14. It is at once obvious 
from these curves that after the admission of uninsolated gas the attainment of the 
maximum action is slower than when action has begun after all the gas has been 
uniformly insolated. The fresh gas thus apparently exercises a retarding effect as 
long as it has not been brought to the condition of the rest of the gas. Mere contact 
with the insolated gas diminishes this retarding effect, since the maximum is reached 
sooner after the gases have stood for some time in the dark. This is shown by 
curve III in fig. 15 being intermediate in character between curves I and IL 
The effect of admission of uninsolated gas increases with the quantity of fresh gas 
added. The velocity curves change from the type (1) to the type (2) as the quantity 
of fresh gas is increased. 
§ 4. The Effect of Previous Insolation of the Separate Gases. 
It has already been stated that Draper found that previous exposure of chlorine 
to light produced an increase in the rate of action, when the chlorine was mixed with 
hydrogen and exposed to light. 
This conclusion was contradicted by Bunsen and Roscoe, but the explanation of 
this contradiction is that the chlorine in Bunsen and Roscoe’s experiment was 
bubbled through water, after it had been illuminated, and this destroys the effect. 
To investigate this effect two inverted burettes were placed side by side, the 
open ends dipj)ed under water in a trough and the burettes were filled with water. 
Chlorine was slowly passed into the burettes, one of which was darkened by enclosing 
it in a metal tube. The tube had a slit through which the volume of the gas could 
be read. By passing the chlorine slowly into the tubes, the water in them becomes 
almost saturated, so that solution of chlorine in the water does not take place to any 
appreciable extent during an experiment. After the chlorine in the tube exposed to 
diffuse daylight had stood for two or three minutes, this tube was also darkened and 
approximately the same quantity of hydrogen was introduced into the two burettes, 
so that in each tube there was an approximately molecular mixture of the two gases. 
The tubes were then left for some time to give the gases opportunity for thoroughly 
mixing, and then the two tubes were simultaneously exposed to the light. Combi¬ 
nation took place, the hydrochloric acid formed being absorbed by the water, wliicli 
rose in the burettes, and thus afforded a means of measuring the velocity of the 
combination. In every case the tube which originally had been exposed to light 
with chlorine in it showed a greater velocity of action at the beginning. In some 
