378 THE SURVIVAL OF THE UNLIKE. [XXIV. 



of the variety, rather than to extend it to the phil- 

 osophical question as to whether varieties, like indi- 

 viduals, become old and die, or wear out. 



My proposition and the proof of it are simply 

 these : Running out is the disappearance of varietal 

 characteristics through change ; all plants vary or 

 change ; therefore varieties must tend to run out. 

 While there can be no doubt of this general fact or 

 law, there are still degrees of running out, because no 

 two plants vary in the same way or at the same 

 rate ; that is, as there are diverse kinds of variation, 

 so there must be diverse kinds of running out. The 

 causes of running out are, therefore, as numerous as 

 the causes of plant variation, and they include all 

 such considerations as the influences of soils, climates, 

 methods of cultivation, attacks of fungi and insects. 

 It is necessary, however, to distinguish between the 

 disappearance of varieties through natural change 

 and through mere fashion, for the latter often ban- 

 ishes varieties which are useful and well marked. 



We can divide variation into two general groups, 

 seed- variation and bud -variation. 



Seed -variation may be called a progressive ten- 

 dency, because the new forms or variations are gen- 

 erally markedly unlike their ancestors, and possess a 

 greater or less tendency to perpetuate themselves. 

 The seed grower is obliged to exercise constant vigi- 

 lance to keep his stock "true." He knows that, as 

 a rule, stock is more likely to remain true on poor 

 soils than on very rich ones, because on the latter 

 it tends to sport or "break" more. Dwarf peas 

 soon become half -dwarfs upon strong soils, and they 

 possess a tendency to perpetuate the new characteris- 



