193 



Tlie geologists contend that some rock inclusions are in 

 the subcooled state and that they have been in that condi- 

 tion for millions of years. If this could be established as 

 a fact it would serve the purpose of the longest imaginable 

 experiment to check the theory according to which crystal- 

 lization always occurs if enough time is allowed for it ; and 

 it would speak against this theory. 



Among observations carried on over definite periods of 

 time, let us mention that of de Coppet (1907) who main- 

 tained 16 soliTtions of sodium sulfate in the supersaturated 

 state, far below the precipitation point, for 33 years. 



Moran (1925) kept an egg yolk subcooled for a week at 

 -11° (that is, more than 10 degrees below the freezing- 

 point), while, in almost all the other cases observed, sub- 

 cooling of but a few degrees lasted only several hours. 



These few facts, and many others, seem to indicate that 

 the fundamental theories proposed for explaining the ac- 

 tion of the time factor in initiating crystallization are none 

 too reliable. More investigations on the relation between 

 time and temperature and on the influence of the nature 

 of the substance on that relation could pave the way to a 

 deeper understanding of the problem. 



4. Mechanical Disturbance. The types of disturbance 

 usually mentioned as inducing crystallization are : shocks 

 on the container of the subcooled liquid, vibrations of the 

 building, the room or the table where the experiments are 

 made, and shaking or stirring the liquid. The mechanism 

 of action of the disturbance consists evidently in giving to 

 the system which is in an instable equilibrium, the last im- 

 pulse necessary to destroy that equilibrium. It is thought 

 that this is done by some local compression on the path of 

 the vibratory waves created by the disturbance. 



Despretz, already in 1837, pointed out the frequent in- 

 efficacy of repeated shaking. Several investigators after 

 him reported frequent negative results in attempts to 

 induce crystallization by shocks or by stirring. 



De Coppet (1907) could not obtain the congelation of 

 salol by vigorous and repeated stirring, at a temperature 



