94 MASSACHUSETTS HORTICULTURAL SOCIETY. 



breeding plants by continuous so-called pedigree selection, was 

 built up, and the close of the 19th century found breeders here and 

 there throughout the world carrying on laborious methods of selec- 

 tion year after year, in the belief that they were obtaining gradually 

 better and better stock by continually selecting from the best. The 

 DeVriesian conception of new tyj^es being produced suddenly 

 would, if true, challenge the correctness of this policy. DeVries 

 would have us believe that in breeding wheat for instance, if we 

 take up the improvement of a certain so-called variety or race, all 

 we have to do is to cultivate a sufficient number of individuals to 

 obtain all of the t^-pes or mutations which the race has produced. 

 Then, by selecting out these different types and cultivating them, 

 we have exhausted the possibilities of improvement until further 

 mutations occur. Almost every race of wheat contains as we well 

 know, several different types. These t}^es usually reproduce true 

 to seed and they may easily be isolated and their comparative value 

 tested and determined. The breeding would, therefore, consist 

 in isolating the types, propagating them pure, and testing their 

 comparative value until the best type is secured. This, it will be 

 readily seen, is in marked contrast to the continuous methods of 

 selection, carried on year after year, to get a cumulative action of 

 the selection. 



Is there a cumulative action of selection ? This is the question 

 which all experimenters in the field of breeding are now asking. 

 It is probable that no definite answer can be given to this question 

 for many years, until the long time experiments necessary for proof 

 can be carried out. No doubt all of us have a deep seated convic- 

 tion that continued selection does result in improvement. I must 

 confess to such a conviction myself, and yet the careful examination 

 of all of the evidence now available, leaves one decidedly in doubt 

 and rather inclined to agree with DeVries. 



The sugar beet has probably been bred more carefully and for a 

 longer time than any other crop and we should find some evidence 

 of importance here. When Louis Vilmorin began to breed the sugar 

 beet, over sixty years ago, he found by chemical analysis that some 

 roots contained as high as 21 per cent of sugar. At that time sugar 

 analyses were laborious and only a few analyses were made. Since 

 that time the determination of the sugar content by polarization 



