Chap. VIII. OF FIRST CIJOSSES AND OF HYBRIDS. 243 



first been used as the ftithcr and tlien as the mot her, though 

 thev rarely ditFer in external characters, yet generally difier in 

 fertility in a small, and oeeasionally in a high degree. 



Several other singular rules could be given iVom Gartner ; 

 for instance, some species have a remarkable power of crossing 

 with other species ; other species of tlie same genus have a re- 

 markablt; power of impressing their likeness on their hybrid 

 ollspring ; but these two powers do not at all necessarily go 

 together. There are certain hybrids which, instead of having, 

 as is usual, an intermediate character between their two par- 

 (mts, always closely resemble one of them ; and such hybrids, 

 tliougli externally so like one of their pure parent-species, are 

 witli rare exceptions extremely sterile. So again among hy- 

 brids which arc usuall}' intermediate in structure between their 

 jiarents, exceptional and abnormal individuals sometimes are 

 l)orn, Avhich closely resemble one of their pure parents; and 

 these hybrids are almost always utterly sterile, even Avhen the 

 other hybrids raised from seed from tlie same capsule have a 

 considerable degree of fertility. These facts .show how com- 

 pletely fertility in the hybrid is independent of its external 

 lesemblance to either pure parent. 



Considering the several rules now given, Avhich govern the 

 fertility of first crosses and of hybrids, we see that when forms, 

 wliich must be considered as good and distinct species, are 

 united, tlieir fertility graduates from zero to perfect fertility. 

 <ir even to fertility iinder certain conditions in excess ; that 

 their fertility, besides being eminently susceptible to favorable 

 and imfavorable conditions is innately variable ; that it is by 

 no means always the same in degree in the first cross and in 

 th(> hybrids produced from this cross ; that the fertility of hy- 

 brids is not related to the degree in which the}' resemljle in 

 external appearance either parent ; and, lastly, that the facili- 

 ty of making a first cross between any two species is not al- 

 ways governed by their systematic aiHnity or degree of re- 

 semblance to each other. This latter statement is clearly 

 proved by the dillerencc in the result of reciprocal crosses be- 

 tween the same two species, for, according as the one species 

 or the other is used as the father or the mother, there is gen- 

 erally some difieriMice, and occasionally the widest possible dif- 

 ference, in the facility of effecting a union. Tlie hybrids, more- 

 over, produ(<'d from reciprocal crosses often dilTer in fertility. 



Now do these complex and singular rules indicate that .spe- 

 cies have b(>en endowed with sterility simply to prevent their 



