986 
percentage composition a straight line.’ If the specific gravity, of 
the AgN0 3 present in these rhombohedral crystals, is calcnlated on 
the supposition of the additive nature of this property, the value 
obtained is 4.19, a value which is different from that of the ordinary. 
rhombic form and hence it points to the existence of a labile rhomb- 
ohedral AgN0 3 in the mixed crystals. Besides these rhombohedral 
crystals, he obtained also some crystals very rich in AgN0 3 and 
showing the ordinary rhpmbic fofm of AgN0 3 . Stil! NaN0 3 cannot 
release the state of superSaturation of silver nitrate. 
From Küster’s [Zeit. Phys. Chem. 13 , 445 (1894) ; 17 , 357 (1895)] 
work we know that benzoic acid can form solid Solutions with 
salicylic acid. Yet either of the acids cannot release the supersaturation 
of the other. 
So Ostwald’s view [TJeber Katalyse — Deutsche Naturforscher- 
versammlung zu Hamburg (1901)] that substances which can form 
solid Solutions, can release the state of supersaturation of the other, 
is not applicable in the cases of, benzoic and salicylic acids. 
The catalysis of the release in supersaturated Systems can be 
fairly explained from Ostwald’s law of succësSive reactions. Like 
all examples of catalytic reactions in these cases also we notice 
the characteristic disproportion between the quantity of the acting 
substance and the amount of the substance changed by its inflüence. 
By a partiele of dust far below the limit of what is ponderable it 
is possible to cause the relief of snpersaturation of an indefinitely 
large amount, of supersaturated soiution. Ostwald (loc. ci'tj has 
shown that the smallest partiele capable of release has an approximate 
weight of 10 11 gram. 
In all these cases we have the formation of a System the stability 
of which is not the greatest possible under the existing conditions 
of temperature and pressure. There are on the contrary other more 
stable condilions, which are characterised by the fact that in them 
a new phase, that is, a physically different component with other 
properties makes its appearance. In the case of a supersaturated 
solution of copper sulphate, it is the solid pentahydrate. As a rule 
such a new phase does not appear of its own accord and the System 
behaves as if it were in a state of equilibrium and hence stable 
supersaturated Solutions can be kept unchanged for years; but if a 
small quantity of the absent phase comes in contact with the meta- 
stable System, the action at once sets in and the new phase increases 
until equilibrium is established. Thus we see that the essential con- 
ditiën is the presence of a metastable System which only passes 
to the more stable state by its inhérent forces, after the way 
