1896 - 97 . ] Mr W. W. J. Nicol on Supersaturation. 
479 
of the stable form, a change creeps over the Avhole of the prisms, 
which begin to grow again at the edges, and ultimately eat up the 
plate variety. 
Finally, I am indebted to Ostwald’s paper for yet another 
example under this head, but among inorganic compounds, — 
sodium chlorate. This salt, he states, forms supersaturated solu- 
tions, and a drop evaporating under the microscope is seen to 
deposit rhombic plates, which are succeeded by cubes, which 
slowly eat up the former crystals. Ostwald assumes that the first 
formed crystals are hydrated : had he realised that this is not the 
case, he would, I believe, have arrived at the explanation of super- 
saturation I have given above. 
The point here raised by Ostwald requires examination : — Is 
there any reason to suspect that the allotropic forms I have 
observed in the case of compounds usually crystallising without 
water consist of hydrates ? I have no hesitation in answering this 
question in the negative, for the following reasons : — All the 
unstable modifications of hydrated salts have been found to con- 
tain less water than the stable forms, and further, several of the 
modifications described by me as occurring with supersaturated 
solutions have been obtained by Lehmann by simple fusion of the 
dry substance ; and finally, the very examples cited by Ostwald in 
support of his assumption that the modification of sodium chlorate 
is hydrated are arguments against it, for the hydrated forms of 
sodium chloride and bromide are formed only at low temperatures, 
and follow, not precede, the separation of the anhydrous form as 
the temperature falls, that is, as the solution becomes more con- 
centrated ; and the observation of Lb wig, which requires confirma- 
tion, that sodium bromate crystallises at - 4° C. with one molecule 
of water, is also an argument against the probable existence of a 
hydrate of the chlorate when it is remembered that no less than 
five hydrates of the iodate are said to exist. 
In the foregoing, mention has been made of different crystalline 
forms only, but it is clear that what has been said applies equally 
to crystalline and amorphous forms. In cases where supersatura- 
tion is caused by the existence of an unstable amorphous form, and 
the temperature is above the critical point, it is most probable that 
there will be infinite solubility, which leads us at once to super- 
