604 EXPERTMEISTT STATION RECORD. 



the results useless for a study of inlieritance and varial)ility ; but 

 how important for this jDurpose might be a study of the reversions 

 and tlie causes which lead to them. 



Certain supj^osed limitations in plant breeding are so often men- 

 tioned that they have become almost axiomatic. Among these are 

 the incompatibility of earliness and prolificacy, of large size and 

 number of jxirts, and the correlation of other qualities. There is 

 opioortunity for further observation on these points, and if this 

 limitation is found to prevail, the physiological phenomena which 

 determine this antagonism remain to be worked out. Moreover, 

 the recent work of Bateson has cast a doubt on the limitations of 

 selection as a means of fixing a type which is w^orthy of further 

 investigation. 



The study of mutations, as described by De Vries, seems a fruitful 

 Jield for investigation. The old taxonomic idea of the fixity of species 

 aj^peai's to be doomed to give way to a diiferent concept. MacDougal 

 has recently shown that plants can be so ])rofoundly aifected by chem- 

 icals as to lose their character sufficiently to result in the i^roduction of 

 new forms. If this should be true it may result in a new method of 

 plant breeding, and at all events it calls attention to the possil)le influ- 

 ence of certain environmental conditions. Fertilizers and soil condi- 

 tions are known to have an effect on the composition of the plant as 

 well as on the quality. The turnip has been found to be quite con- 

 stantly influenced in composition in proportion to the deficiency of the 

 soil in phosphoric acid, and in the case of cereals an attemi^t has been 

 made to work out a method for ascertaining the fertilizer require- 

 ments of the soil on the basis of the composition of the crop grown. 



The necessity of taking these matters into account when plants are 

 being bred for increased protein, and of fully checking or controlling 

 all conditions other than individuality which may cause changes, will 

 be apparent. In the case of leguminous plants bred for increased 

 protein content the possible effect of nitrogen assimilation as a dis- 

 turbing factor should l)e borne in mind. It has been shown that nitro- 

 gen assimilation may have the effect of increasing the nitrogen content 

 of the plant or its seed, and it has also been found that plants appar- 

 ently vary considerably in the energy with which they assimilate 

 atmospheric nitrogen. 



The conditions surrounding this activity and the factors which 

 influence it are little understood, but thes6 differences in conditions 

 or capacity of the plant may nullify the results of breeding work 

 which does not take account of them. The obvious remedy would 

 seem to be the treatment of the soil in which these experiments are 

 made in such Avay as to eliminate nitrogen assimilation and its attend- 

 ant influences. 



