PLANT HYBRIDIZATION BEFORE MENDEL 181 



to say, in order to express the relationship correctly, that in the case of 

 very vigorous hybrids the weakness of the sexual parts brings out an 

 increased development of the vegetative growth, whereas it is not the 

 case with others which are too weak for such reaction [meaning crosses 

 between two distant species]." (6, p. 43.) 



Wichura concluded from his observations that hybrids were 

 intermediate in respect to the differing parental characters. Cases 

 of dominance do not seem to have come under his hand. 



"Among the numerous artificial and natural willow hybrids observed 

 by me," he says, "I have throughout verified but one apparent exception 

 to the principle of intermediateness. 



". . . Even the time of flowering of hybrids holds the mean between 

 the times of flowering of the two parents." {ib., p. 47.) 



"As rich in species as the genus of the willows is, and as numerous 

 combinations of hybrid fertilizations as it has to show, nevertheless I 

 have never yet verified anything of a preponderant influence in any one 

 of its species, but rather always found that their hybrids hold the mean 

 between the constant characters of the parents." {ib., p. 50.) 



"In hybrid fertilization, if unlike factors [Factoren] unite, there arises 

 an intermediate formation, etc." {ib., p. 86.) 



The latter passage appears to be the first occasion where the 

 term "factor" has been used in the literature of plant breeding, 

 although here the "factors" referred to are perhaps the parents 

 as a whole which participate in the cross, rather than the charac- 

 ter-forming units from those parents. 



His general conclusion is (ib., p. 46.) : 



"Constant characters, through which the parent, species are distin- 

 guished from one another ... go half over to the hybrid, so that it 

 holds the middle position between them." 



Two observations of Gartner's were verified by Wichura — the 

 identity of reciprocal crosses (pp. 51 and 186), and the fact of 

 "variation" in hybrid progeny. 



As to the question of the relative importance of the egg or the 

 pollen in the result of fertilization, Wichura says (p. 57) : 



"One sees the question is still far removed from having been brought 

 to light, but from Gartner's and my own observations it appears at 

 least determined, that the products of hybrid pollen in breeding are 

 more various than those of the pollen of true species." 



Regarding the generally observed identity of reciprocal crosses, 

 Wichura draws the following conclusion (p. 86) : 



"We have found that the products which come from reciprocal cross- 

 ing, unlike the well-known experiments made in the animal kingdom, 

 completely coincide with each other. 



"From this it must follow, however, with mathematical necessity, that 



