as connected with the Theory of Subsiztutions. 355 
atom, and chlorohydric acid form. It is probable, indeed, that 
the two oxygen atoms in peroxide of hydrogen are related to their 
hydrogen atoms with different degrees of affinity, and that one of 
them is retained far more loosely than the other. But this would 
correspond with our ideas of oxygenized water and not peroxide 
of hydrogen, and leads us to the conclusion that the solution em- 
ployed in this memoir is strictly a solution of chlorine in water. 
XI. The decomposition of chlorine water, when placed in the 
sunbeam, does not begin at once, but a certain space of time inter- 
venes, during which the chlorine is undergoing its specific change. 
I need quote no further instance of the truth of this than the 
experiment given in support of the second fact. 'Thisis the same 
phenomenon which takes place when chlorine and hydrogen are 
exposed together; they do not begin to unite at once, but a cer- 
tain space of time elapses, during which the preliminary tithoni- 
zation is taking place; and when that is over union begins.* 
On the Relations of Chlorine and Hydrogen. 
We have thus traced the cause of the decomposition of water, 
in the case before us, to a change impressed upon the chlorine by 
exposure to the rays of the sun. In this decomposition three el- 
ementary bodies are involved—chlorine, oxygen, and hydrogen. 
We can therefore reduce the problem under discussion to sim- 
ple conditions, and study the relations of each of these substances 
to each other and to the solar rays successively. 
When a mixture of oxygen and hydrogen gases, in the propor- 
tion to form water, is exposed to the most brilliant radiation con- 
verged upon it by convex lenses, union does not ensue ; the rea- 
son being, as I have formerly shown, that those gases are per- 
fectly transparent to the rays, and do not possess either real or 
ideal coloration. For the same cause, water exposed alone for 
any length of time to the sun, or to the influence of a large con- 
vex lens, does not decompose. It is transparent, and cannot ab- 
sorb any of the rays. 
But, as is well known, a mixture of chlorine and hydrogen 
unites, under the same circumstances, with anexplosion. I have 
formerly proved that this depends on the absorption of the indigo 
rays. For in the indigo space the action goes on with the greatest 
activity. 
* Philosophical Magazine, July, 1844. 
