1646 
can be explained, according to the investigations of WreLanD, in a 
perfectly analogous manner. 
Palladium is capable of withdrawing hydrogen from an alcohol 
and this absorbed and activated hydrogen is then attacked by the 
oxygen. 
In the phota-reaction the case is in reality still simpler and purer. 
With the co-operation of definite rays of light, the ketone is 
capable, if not of actually withdrawing the H, of modifying the same 
(in some cases perceptible by the yellow colour). 
Which practically will amount to the same thing, for as soon as we want to 
get a conception as to the change of the H, a form must be given to it in which 
either the position of the atoms or that of the electrons must be considered as 
modified. 
The photo-active H can now form water with the oxygen just 
like the Palladium-active hydrogen can do this. 
We have tried whether the element H, might also be rendered 
active by means of the system Ketone + light, but this did not succeed. 
From the experiments of Cramician and Sitper and also from our 
own we must, however, conclude that a large variety of hydrogen 
compounds is certainly capable of it. 
In consequence of our observations as to the activation of the 
alcoholic hydrogen the following question still arose, in connexion 
with the catalytic phenomena. 
We can conceive ethyl alcohol to be built up of ether and 
water and there are indications that in the system: 
(C,H,),0 + H,O = 2C,H,OH 
the alcohol at the ordinary temperature is more stable than the 
mixture of ether and water.’ 
If this is so, the formation of alcohol under the influence of defi- 
nite catalysts ought to be capable of demonstration. 
Now, we have first of all tried the effect of illumination on benzo- 
phenone in absolutely dry ether; no action had taken place in the 
time that it was converted with absolute alcohol, completely into 
pinacone. 
We remind the reader that water exerted a very retarding influence 
on the photo-reaction, 
