the reaction, or else depend on the nature of these substances 
themselves. 
It strikes me that the number of different syntheses made since 
the discovery of the catalytic action of aluminium chloride is. large 
enough to enable us to explain the cause of this different conduct 
by a somewhat systematic consideration. 
It must be well remembered that aluminium chloride can only 
then exert its power when it is capable of rendering the chloride 
(or anhydride) active; that is to say it must be present in the mass 
either in a free or loosely-combined state. 
This mass contains besides the catalyzer (and eventually some 
diluent, such as CS,): A the chloride (or anhydride), B the benzene 
derivative, C the formed product. If now we disregard the above- 
mentioned secondary decomposition phenomena the following cases 
may occur.: 
I. The aluminium chloride combines with zone of these substances 
or the compounds are completely dissociated at the reaction-tempe- 
rature. 
We are then dealing with the catalytic action in the truest form. 
A trace of aluminium chloride will suffice to convert unlimited quan- 
tities of A and B into C. This is for instance the case in the chlo- 
rination (bromination) of benzene at the ordinary temperature. If the 
substances used have been carefully dried more than 1 kilo of chloro- 
(bromo) benzene may be prepared with the aid of 0.5 gram of 
aluminium chloride without a visible diminution of the quantity of 
the catalyzer. When preparing diphenylmethane from benzyl chloride 
and benzene we can also work with very small quantities of the 
catalyzer if the strongly diluted benzyl chloride is poured into a 
large excess of benzene and the reaction-mass is from time to time 
replaced by new benzene; yet the decomposition of the benzyl 
chloride by the catalyzer cannot be entirely prevented (Recueil 
XXIII p. 98). 
II. The aluminium chloride combines with the chloride (A) to a more 
or less strong additive product. When these compounds are very 
stable, the reaction may not take place at all: the phosphorus oxy- 
chloride combines with strong evolution of heat with a mol. of 
aluminium chloride (Cassetmann, Ann. 98 p. 220), and this product 
is not attacked by benzene or toluene. In the other cases the reaction 
proceeds, however, very favourably. As aluminium chloride also 
combines with the formed product (C) only one mol. of the catalyzer 
is required for equimolecular quantities of the components. 
Here we must still distinguish between the following categories; 
