LIFE AND DEATH. 289 



than penetrate into the cellulose walls and diffuse 

 a poison into the adjacent cells, being- utterly 

 incapable of directly facing, or mingling with the 

 living protoplasm of such cells, whereas the proto- 

 plasm of another organism e.g. Rhizobium 

 will penetrate directly into the cells, live in them 

 for weeks or months without injury nay even 

 with advantage to their life. And hundreds of 

 similar cases can be selected. 



We may, therefore, conclude that Variation 

 depends fundamentally on alterations in the 

 structure or mode of building up and disin- 

 tegration of the protoplasmic molecular unit, 

 brought about either by direct modifying action 

 of the inorganic environment nutrition, tem- 

 perature, oxygen supply, light, etc., etc. or by 

 the mingling with it of other protoplasm, the 

 molecules of which since they have already a 

 slightly different composition, configuration, mode 

 of breaking down and building up, etc., affect its 

 molecules by supplying them with altered nutri- 

 tive atom-complexes, by competing with them for 

 oxygen, etc., etc. Once these molecules are 

 affected, we must assume that long sequences of 

 other chemical and molecular changes will be also 

 modified ; and although we have no conception 

 of how these changes bring about changes in 

 form, that they do so is only a conclusion of the 

 same order as that which we hold regarding the 

 much simpler changes concerned in the formation 

 of crystals. 



That such variations may be of every degree as 



T 



