266 THE ROYAL SOCIETY OF CANADA 



Young was able to show that the product of the degree of super- 

 cooling and shock (measured in some mechanical units) was roughly 

 a rectangular hyperbola, and that, therefore, for a small degree of 

 super-cooling or of supersaturation, the necessary mechanical shock 

 must be very great. Since the shock need be given at one point only, 

 it would seem that this might be tested by means of minute amounts 

 of explosives. 



However the metastable theory is, for many systems a 

 very satisfactory one. For supersaturated solutions of sodium 

 sulphate or super-cooled salol I have never succeeded in causing 

 crystallization by mechanical shock, by an electric spark, by a small 

 arc playing below the surface, by radium in solution, by radium A, 

 B and C deposited on a needle and held near the surface, or by cathode 

 rays. As shown by Violette and Gernez, crystallization in a sodium 

 sulphate solution is caused by the particles in the air. If the air be 

 "sterilized" by bubbling through water, or drawn through cotton 

 wool, or through a tube heated above the melting point of sodium 

 sulphate decahydrate, crystallization never results. DeCoppet has 

 preserved tubes for thirty years, and in only a few cases has crystal- 

 lization occurred. 



That crystallization is brought about by a crystal, and stops when 

 the crystals are removed, can be shown with supersaturated sodium 

 thiosulphate. A hair which has touched a crystal induces crystal 

 formation, and as these adhere to the hair they may be removed and 

 crystallization stopped (Ostwald). I have tfrequently been able to 

 perform this experiment with sodium sulphate. A crystal placed in 

 the bottom of a supersaturated solution of sodium chlorate pre- 

 cipitates the solute and leaves a supersaturated solution in the upper 

 portion of the vessel. 



The analogy to the phenomena of bacteriology is so striking that 

 Gernez, Pasteur's assistant, was thus led to a study of supersaturated 

 solutions; and the technique developed in these investigations is so 

 similar that it is convenient to use the terms "infection," "steriliza- 

 tion," etc. 



It is worth while here mentioning a phenomenon similar in many 

 respects to that met with in cultures containing spores, that is, the 

 spontaneous crystallization of many solutions, which have not been 

 properly ''sterilized." 



I may cite one example from many: 



Eleven tubes containing salol were fastened to a header in order 

 to be exhausted. These were sterilized by heating to 78° (M.P. salol 

 48°) for 2 hours, and left over night. 



