1 6 The Rural Library. 



similar species which have arisen quite independently in places remote from 

 each other. In the one case, the species have likely struggled with each other 

 until each one has attained a degree of divergence which allows it to persist ; 

 while in the other case there has been no struggle between the species, but simi- 

 lar conditions have brought about similar results. These similar species which 

 appear independently of each other in different places are called representative 

 species. Islands remote from each other, but similarly situated with refer- 

 ence to climate, very often contain representative species, and the same may 

 be said of "other regions much like each other, as eastern North America and 

 Japan. Now it follows that if representative species are less opposed than 

 others, they are more likely to hybridize with good results ; and this fact is 

 well illustrated in the Kieffer and allied pears, which appear to be hybrids 

 between representative species of Europe and Japan, and I am inclined to 

 think that the same may be found to be true of the common or European 

 apple and the wild crab of the Mississippi valley. Various crabs of the 

 Soulard type, which I once thought to constitute a distinct species, appear upon 

 further study to be hybrids. We will also recall that the hybrid grapes 

 which have so far proved most valuable, are those obtained by Rogers be- 

 tween the American Vitis Labrusca and the European wine grape ; and that 

 the attempts of Haskell and others to hybridize associated species of native 

 grapes have given at best only indifferent results. To these good results 

 from hybrids of fruit-trees and vines I shall revert presently. 



Another theoretical point which is borne out by practice is the conclusion 

 that, because of the great differences and lack of affinity between parents, 

 pronounced hybrid offspring is unstable. This is one of the greatest diffi- 

 culties in the way of the summary production of new varieties by means of 

 hybridization. It would appear also, that because of the unlikeness of 

 parents, hybrid offspring must be exceedingly variable, but as a matter of 

 fact, in many instances the parents are so pronouncedly different that the 

 hybrids represent a distinct type by themselves or else they approach very 

 nearly to the characters of one of the parents. There are, to be sure, many 

 instances of exceedingly variable hybrid offspring, but they are usually the 

 offspring of variable parents. In other words, variability in offspring ap- 

 pears to follow rather as a result of variability in parents than as a result 

 of mere unlikeness of character. But the instability of hybrid offspring 

 when propagated by seed is notorious. Wallace -writes that ' ' the effect of 

 occasional crosses often results in a great amount of variation, but it also 

 leads to instability of character ,_ and is therefore very little employed in the 

 production of fixed and well-marked races." 



I may remark again that, because of the unequal and unknown powers of 

 the parents, we can never predict what characters will appear in the hybrids. 



