je5 Arend. L. Hagedoorn and A. C. Hagedoorn. 



fore are not all pure for the same genes, such as populations of 

 maize or of beets. 



Still less effect has selection in populations of organisms, which are 

 allogamous and not pure genotypically, such as populations of dioecious 

 plants or of the higher animals, and here again the effect of selection 

 will be very much greater if coupled with inbreeding than without it. 



Finally, selection can have no effect whatever in populations of 

 plants or animals, having all the same genotype, whether they belong 

 to the same clone or to tlie same pure sexually propagated type, 

 such as in pure lines of beans or wheats, but also in inbred families 

 derived from one plant, pure for all its genes, even if not self- 

 fertilized, in pure clones of Parauieecium or of Potatoes, in purebred 

 homozygous strains of animals (if such exist). 



But we have to reckon with a certain group of Geneticians, 

 chiefly Zoologists, who maintain that there exists a certain kind of 

 continuous variation which is not caused by genotypical variation, 

 not by presence and absence of genes, but which is nevertheless in- 

 heritable, in so far, that continued selection of extreme variates shifts 

 the mean in the selected direction. Their conception is essentially 

 Lamarckian, though they would be possibly the first to repudiate 

 this. Let us see on what they found their doctrine, and whether 

 the facts they have assembled can not be explained in a normal way. 

 In the first place, it must be noted, that, for as far as we know, 

 no Botanists hold this belief today. At the time of pubhcation of 

 his book ,,Die Mutationstheorie", de Vries emphatically did, but 

 this is more than ten years ago. 



Castle is one of the chief spokesmen of this group. He starts 

 from the fact that "unit-characters vary", as they undoubtedly do. 

 He wants to know, whether it is possible by selection to change "a 

 unit-character". 



Now, here it is necessary again to ask : what is a unit-character ? 

 If a unit-character is a "character determined by one single genetic 

 factor" or rather, if a pair of unit-characters differ in that the 

 organism presenting one has in its germ one gene which the other 

 had not, than the answer must be affirmative, for in as far as other 

 genes, for which the group of organisms may be impure, influence 

 these same characters, selection may have effect. But from his 

 writings it is evident that Castle, when he asks the question: can 

 selection modify a unit-character, really means: "Can selection modify 

 a genetic factor?" 



