358 THE SURVIVAL OF THE UNLIKE. [XXII. 



merit are more numerous in all plants now than thej* 

 were a century or even a human generation ago, it 

 must follow that new varieties have been appearing 

 all these years which were good enough to obtain 

 the confidence of all careful growers. In two papers 

 which I have presented to this society,* I have shown, 

 I think, that varieties do not wear out ; but all 

 plants which are habitually propagated by seeds, as 

 garden vegetables and flowers, tend constantly to 

 change or differ from their parents, and finally to 

 pass so far away from them that they receive new 

 names ; and plants which are propagated from cut- 

 tings of abnormally developed parts, as the potato, 

 constantly tend to deteriorate unless grown and 

 selected under the very best conditions ; but all 

 plants propagated from normal or un variable parts, 

 as by ordinary cuttings, cions, and layers, remain 

 substantially the same from century to century, as 

 is the actual case with several prominent orchard 

 fruits. If the orchard fruits do not run out, there- 

 fore, the only reason why the varieties should change 

 is because better ones appear and drive them out ; 

 and inasmuch as it is a matter of common knowledge 

 that change does take place, it follows that profitable 

 novelties have appeared. 



Up to this time, therefore, novelties, or at least 

 many of them, have paid. Is there any reason for 

 supposing that they will not pay equally well in the 

 future ? Or, to raise my original question : Is prof- 

 itable variation no longer possible ? This question 

 is not new, and there is no special reason for asking 



* Reprinted in Egs«y XXTV. 



