MODERN PLANT-BREEDING METHODS 703 



more particularly by De Vries, Johannsen, and Nilsson, with 

 results which throw a great deal of light upon this process of 

 selection. It is impossible to discuss these researches in detail 

 in a brief manner, but the outcome of them may be shortly 

 stated. Each cultural variety of any crop, though pure to the 

 casual observer, now appears to be composed of a number of 

 definite types distinct from one another in minor characters, and 

 the most that can be effected by selection is to pick out the best 

 of these types. 



Limits are set at once to the results which may be obtained 

 by selection, and as the production of radically fresh types by 

 mutation is not practicable at present, a halt will have to be 

 called in the process of plant improvement once the best types 

 have been isolated. 



A second method employed by breeders is hybridisation, and 

 the first year of the new century saw the rediscovery of Mendel's 

 researches, and with that the proper understanding of this 

 complex subject. The phenomena of hybridisation had been 

 investigated with great patience by a number of scientific 

 workers, such as Kolreuter, Gartner, Naudin, Godron, and 

 Rimpau, but it appeared impossible to draw any conclusions of 

 general application from their researches. To a great extent 

 this was due to the complexity of the cases which they examined 

 and to the fact that they were concerned with such problems as 

 the conversion of one species into another by repeated artificial 

 fertilisation, or with the part played by sex in determining the 

 form of the hybrids. 



The economic breeder with nothing to guide his work 

 trusted to chance. He recognised the fact that the offspring 

 of his hybrids showed great " variability " and amongst these 

 " variations " he hoped to find some showing improvements on 

 the parents. Once found, the new types had to be fixed. This 

 meant as a rule years of selection and possibly failure at the 

 end. So uncertain were the results that comparatively few 

 seedsmen have devoted much attention to the subject, and much 

 of the improvement we see in our garden plants we owe to the 

 patience of the amateur who has made some class of plants an 

 object of special study. Further, this difficulty of fixing the 

 new types has led largely to their vegetative propagation, and 

 it is now amongst plants which can be readily multipled in this 

 way that we find most hybrids. 



