9g Hagedoorii. 



from the Andes, only a rosespecialist would care to use a newly im- 

 ported siugleflowered wild rose from Manchuria for crossbreeding- pur- 

 poses. Very often this fact, the use of new species for crossl)reeding: 

 work, in order to produce novelties, has been interpreted by admiring: 

 botanists as an example of the wonderful knowledge, the sure intuition 

 of the plantbreeders. In reality, the plantbreeder is quite in the dark 

 himself. All he counts on, is the fact, that speciescrosses will almost 

 always produce the most astonishing- variability in all directions, and 

 great variability is what he is anxious to see. Who could predict, that 

 the glaring brickorange colour of the African marigold, Dimorphotheca 

 aurantiaca coukl be changed to a dozen new and pleasing hues by 

 crossing it to an insignificant yellow marigold. Calendula pluviatilis? 

 And yet Haage and Schmidt obtained creams, and blues, and pale 

 lilacs and pinks from this cross. Surely, nobody could imagine that 

 from a cross between two very similar Argemone species, double flowers 

 and fimbriation and polycephaly would crop up in the second gene- 

 ration, as we saw at de Vilmorin's. 



Speciescrosses produce novelties, new characters. This truth is 

 well recognized by the Horticulturists, and begins to be recognized by 

 Botanists. Geneticians generally are only just beginning to realize, 

 that wholly new characters can be due to crossing. The reason for 

 this lies in the fact, that the determinant conception of heredity has 

 taken such a hold upon the minds of the majority. Crossing 

 evidently recombines genes, and if we conceive of genes as of deter- 

 minants for characters, we must conclude, that as the result of crossing, 

 characters can be variously combined, but not created de novo. It 

 is evident, that crossbreeding can recombine characters which are 

 each determined by a certain gene, if by the cross these genes, which 

 until now had been present in different forms, could be brought together 

 in one organism. 



It is very clear now, that there is no such direct relationship between 

 genes and qualities. We have to conceive of the qualities of every 

 organism as the result of the organism's development, and we know, 

 that this development is a result of the cooperation of a host of deve- 

 lopmental factors, some environmental and some inherited. In coope- 

 ration with a given set of othei- factors in the development, a gene 

 will always produce the same effect. But no gene can possibly have 

 any effect in itself, or can produce a certain influence, independent 

 from the nature of the individual in which it occurs. 



