( 418 ) 
catalytic influence of aluminium chloride, because the potential fall 
of the reaction : 
K.C1 -f A1 Cl, = RC1. A1C1, 
must necessarily be deducted from the potential fall of the entire 
reaction, so that one must still search for the cause of the velocity 
of these intermediate reactions. 
The chlorides which form these additive products with difficulty, 
or not at all, such as C Cl 4 , CH Cl, , S 2 Cl, , P CI, etc. are as a 
rule attacked with greater violence than those which form solid 
molecular compounds with the catalyst. 
As a matter of fact, I have come to the conclusion, and have 
stated so previously on account of the conversion of yellow phos¬ 
phorus into the red modification (1. c.) that the explanation of the 
catalytic action must be looked for in the dissociating influence 
which the catalyst exercises when nearing the molecules. 
Whatever we may think of the cause of the permanent coherence 
of the molecules, it is certain that there exists between the atoms a 
certain contrast, which cannot be quite effaced because the atoms 
cannot annihilate each other. 
This contrast appears to be dependent on circumstances; if these 
change, the contrast also changes and this causes a modification in 
the existing equilibrium. To these changes belongs the approach of 
other atoms or atom-groups. The action of positive catalysts may 
now be represented as being that of atoms or atom-groups which 
owing to their immediate proximity so modify the existing contrast 
that another equilibrium can be attained by the first molecule. 
In the reaction of Friedel and Crafts, I therefore, assume that 
in the first place the aluminium chloride loosens the cloride (perhaps 
also the double bonds in the benzine derivative, for it appears that 
AlClj exercises a condensing influence on unsaturated solutions, the 
least on benzene itself as was to be expected), and that the reaction 
can then take its course in the usual manner. 
There are in the literature many indications that aluminium chloride 
has a purily dissociating action : I will mention the following cases. 
Methyl (ethyl) perchloroformate is resolved into C0 2 and CC1 4 (C 3 C1,) 
(Hentschel J. pract. [2] 36 p. J00 and Muller Ann. 258 p. «D* 
Pentachloroethane is decomposed into HC1 and C,C1 4 Mounetrat Bull- 
L3J 19 p. 179), CH 2 C1. CH.Cl into HC1 and C,H, (Bull. [3] 19 P 445) ’ 
ropyl chloride yields HC1 and propylene (Mouneyrat Bull. [3] 2 
p. 616) etc. 
Sulphurylchloride is resolved into SO, and Cl, (Ruff Ber. I 902 
