SELECTION WITHOUT ARTIFICIAL CROSSING. 



33 



give rise to inferior plants in the progeny of what was originally a 

 high-class individual. 



While experimental evidence for the above statements is not as 

 plentiful as it ought to be, Doctor Nilsson has done so much work 

 along this line that the propositions enunciated may be considered 

 practically estabhshed. Professor Johannsen, of Copenhagen, has 

 done a great deal of work of the same kind with exactly the same re- 

 sults. At the 1906 Genetic Conference in London he said, '^In a 

 population containing only one single type the selection of fluctua- 

 tions has no action at all." Johannsen has several races of beans 

 which he has grown pure for several years and which are fully ho- 

 mozygote. He has fully tested the effect of selection on certain 

 seed characters of these beans. Speaking of the results,^ he says, 

 ^'Selection for weight, for absolute length, or relative breadth has 

 had absolutely no observable influence on these characters." Johann- 

 sen has obtained similar results with barley. 



After what we have seen to be true in vegetatively propagated 

 races it should not be surprising that similar results occur in self- 

 fertilized races, for although such races go through the form of re- 

 combining the characters when they produce seed, the two mem- 

 bers of each pair of characters being exactly alike, we get no new 

 combinations, so that reproduction by seed in completely homozy- 

 gous strains differs little, if at all, in its results from vegetative propa- 

 gation. 



The fact that in a field of a self -fertilized crop a very large major- 

 ity of the plants are perfectly fixed in their hereditary characters 

 and will reproduce themselves with almost absolute fidelity from 

 seed has led a good many biologists to consider every one of these 

 plants which differs from its neighbors in any way to be what they 

 call ''elementary species." They overlook the fact that these 

 forms are fixed simply because they are homozygote in all their 

 characters and would behave in exactly the same manner whether 

 the evolutionary changes that produced them are either very slow 

 and gradual or occur suddenly at long intervals. 



The problem, then, in selecting self-fertilized plants is to find the 

 best individuals and propagate from them. There are two ways 

 of selecting such plants, which give somewhat different results. 

 One of these we may term "mass selection," the other "individual 

 selection." 



MASS SELECTION. 



Mass selection is that form of selection in which a number of supe- 

 rior plants or parts of plants are chosen but their seed is not kept 



« Zeitschrift fiir Induktive Abstammungs- und Vererbungslehre. September, 

 1908, p. 2. 

 165 



