LIFE, CARBON'S OUTSTANDING PROPERTY 73 



small crystals. If we dissolve salt or sugar in a little water, 

 filter and evaporate a part of the water, the salt or sugar will 

 crystallize again and again in ''purer" form, since traces 

 of other substances which may have been in it in small 

 amounts will stay in the solution. The cheaper grades of 

 salt or of sugar contain small amounts of other substances; 

 repeated crystallization is the widely used method to re- 

 move these other materials. 



If virus is a pure material in the same sense as sugar 

 or table salt, should it not be capable of crystallization, 

 even though it is alive? This is a thought of momentous 

 importance: a living entity capable of crystallization! We 

 have previously discussed the occurrence in plants and ani- 

 mals of crystalline elements which can be demonstrated by 

 x-rays. But now we are confronting a different problem; 

 we are not concerned with the question whether each in- 

 dividual virus may or may not be an invisibly small crystal. 

 The question before us is whether a multitude of viruses, 

 if present in bulk, can be induced to form crystals, like the 

 sugar which settles from an evaporating sugar solution. 



Certain technical difficulties must be overcome in order 

 to solve this problem by experimentation. The total 

 amount of virus which grows in animal or man when a 

 disease (like small-pox) develops, is quite small as compared 

 with the total weight of the body. Even at the height of 

 a fatal disease, the virus has not so multiplied that the body 

 substance is consumed by it. Crystallization is impossible 

 under such circumstances. Even sugar crystals will not 

 form if the sugar is mixed with a greater amount of other 

 material, as is done in sweetened food like cake or ice 

 cream. 



Before crystallization is possible the virus material must 

 be separated from the body materials with which it is mixed. 

 Such a separation is in most cases difficult, but it is rela- 

 tively easy in certain virus diseases of plants, of which the 



