48 CRYSTALS AND LIVING UNITS 



temperatures in supersaturated solutions of sodium sulphate by the 

 access thereto of microscopic crystals of this salt, which exist 

 (almost always) as constituents of atmospheric dust. These 

 microscopic crystals, not much larger than the germs of microbes, 

 act much like germs, and lead to the separation of crystals from the 

 mother liquor. 



Ostwald showed that salol, a white crystalline substance which 

 melts at 39-5° C, may, when in this condition, be rubbed with a 

 hair, a thread of glass or of platinum, or any angular substance, 

 without inducing crystalHsation. It will remain in the fluid state 

 so long as it is not brought into contact with a crystal of salol ; but 

 directly it is touched with one of the previous objects after it has 

 been brought into contact with salol, crystallisation is at once 

 initiated in the fluid and spreads rapidly through its mass. The 

 inoculating object, however, may be easily sterilised by raising it 

 even momentarily to some point above 39"5° C. Very similar results 

 were obtained with supersaturated solutions of hyposulphite of 

 soda and chlorate of soda, which remained without change (sterile) 

 until they were inoculated with microscopic crystals of these sub- 

 stances, of extreme minuteness. 



" In face of such facts," says Errera, " it is natural to ask how the 

 first crystal of each of these substances is born." And the reply, 

 he says, must be, " by spontaneous generation." 



Some supersaturated solutions may (when protected from the 

 advent of all germs) be rendered fertile by allowing them to 

 evaporate and become more concentrated, others by lowering the 

 temperature of the solutions, and others stiU, even by shght 

 mechanical shocks when the supersaturation is strong. De Coppet 

 also made out the interesting fact that " other things being equal, 

 the first crystal forms more quickly in large than in small masses of 

 liquid," when dealt with in either of these ways. 



In illustration of the effects of temperature some important 

 investigations have been made with hitol by Tammann, of Dorpat. 

 " Its point of fusion is more elevated than that of salol : it is 96° C. 

 Melted at about 100°, and subsequently kept in little sealed tubes 

 and cooled, it remains liquid for a period more or less long. But 

 sooner or later, according to the temperature and the quantity of 

 liquid employed, centres of crystallisation gradually appear, whence 

 solidification is slowly diffused throughout the whole mass. For 

 the same volume of liquid the number of its centres of crystallisation 

 increases with the cooling, attains a maximum, and then decreases 



