66 AMERICAN BREEDERS ASSOCIATION . 



basis either directly or indirectly, but many elementary forms have a 

 wide range of fluctuating variability and there is in such cases the 

 capacity for great improvement within the elementary species. In the 

 improvement of strains by the selection of the best fluctuating vari- 

 ants, however, unceasing attention is necessary: for the possibility of 

 fixing an extreme of the fluctuating kind lacks experimental support. 

 No matter how long the selection of the same extreme is carried on, 

 the instant the selection ceases to counteract the regression, a deteri- 

 oration begins, and in the course of time there is a complete return to 

 the norm of the particular species or strain in question. When a new 

 elementary species originates by mutation, a new norm is produced 

 and if this is a higher norm with respect to the aggregate of economic 

 features than that possessed by the parental strain, a distinct and 

 permanent advance has been achieved. A new range of fluctuating 

 variability has also been produced and if this range is greater than 

 that of the parent, new possibilities of improvement within the ele- 

 mentary species is provided for. 



It has been found that mutants arise in a species with something 

 like a constant frequency and that the economical way to secure 

 mutants is to raise enormous numbers of seedlings among which 

 search for them. It is confidently expected that at some 



t distant time, ways will be found to increase the relative number 

 mutants, or even to influence the direction of mutation. A first 

 important step has been made in this direction by Dr. D. T. Mac- 

 Dougal, the Director of the Department of Botanical Eesearch of the 

 Carnegie Institution at Washington, who has apparently induced mu- 

 tations in several species by injecting into the developing ovaries vari- 

 ous solutions of mineral and organic substances, but aside from the 

 suggestion offered by these interesting experiments, we seem to be at 

 present entirely at the mercy of nature. However, nature treats us 

 well if we stand equipped with as complete understanding of her ways 

 as science provides, ready to take advantage of and preserve every 

 advance she makes. 



In summing up, it may be said that the value of the mutation theory 

 in its bearing upon practical breeding lies in the better appreciation 

 it gives of the significance of the phenomena with which the breeder 

 must deal. New methods and the extension of old methods to new 

 material are indicated, the characteristic features of the method being 

 the production of enormous numbers from which to select, the com- 

 plete isolation of each individual whose characters suggest the possi- 

 bility that it may be the starting point of a new strain, the complete 

 control of the fertilization processes, and the rearing of the offspring 

 of the guarded plants under conditions that will allow all distinguish- 

 ing characteristics to reach a normal development. 



But even if no changes in method were suggested if old methods 

 adopted empirically should be found to harmonize perfectly with the 

 theory so that no improvement could be made the value of the 

 mutation theory is great in the feeling it gives the breeder that he 





