Distinctioi BcIikjccii Spi'cics and I ariclics. 583 



We will now discuss the principle illustrated Ijy this 

 instance from a more general jjoint of \ iew. In the 

 literature of the subject we frecjuently iind the o])iniun 

 that forms which are nuitually fertile and i)r(j(luce a 

 normal harvest of seed, giving fertile hybrids, are to be 

 regarded as varieties of one and the same species. F(jrms, 

 howe\er, the union of which is followed by a diminutitjn 

 in fertility and the hybrids of wdiich are less fertile than 

 tlie species crossed, are regarded by the maj(jrity ui the 

 investigators as specifically distinct. These generaliza- 

 tions have ser\-ed as criteria of relationshi]) from the 

 time of KoLREL'TER and Gartner up to the present; and 

 Darwin himself relies on them in considerations of this 

 kind.^ Based as they were on extensive experience and 

 on a profound systematic knowledge, they constitute prin- 

 ciples which bid fair to become universally recognized. 

 For these reasons they deserve to be placed in the fore- 

 ground as a convenient point of departure for our dis- 

 cussion ; and our object will be not to find out their weak 

 points or to replace them by others, but rather to give 

 them the more definite forms reqtiired Iw' our i)resent 

 knowledge of hybridization. 



Therefore wq will start from the oft-cited proposi- 

 tion that varieties are only small species.- This means 

 that the difference betw^een species and \'arieties is not 

 of a fundamental nature but rather of a gradual or e\en 

 a conventional knid. Moreover we will start from the 

 conception enunciated in the first part of this volume, 

 according to which the forms wdiich compose the collec- 

 tive species are mainly of tw^o kinds (p. 64) : Honioiwni- 



See FocKE. Die Pflanzcnmiscldingc , pp. 436, 446-502, etc. 

 "Sec Vol. T. ]). 171 and above p. 5R0. 



