PLANT BREEDING. I29 



As suggested at the beginning of this discussion, plant breed- 

 ing in its relation to pomology has as yet been largely fortuitous. 

 Little study of fundamental laws has been made. Thousands 

 of crosses have been made and hundreds of thousands of seed- 

 lings have been produced, but the work has been largely without 

 definite ideals in view, and without a view of probable means 

 of reaching ah ideal. In the judgment of the writer, the prob- 

 lems of propagation, environment, and individual variation are 

 of quite as much importance, and are certainly as little under- 

 stood, as are the obscure problems of cytological variations and 

 combinations. 



Many years ago Thomas Andrew Knight popularized the 

 method of root grafting, and the question of the mutual influ- 

 ence of cion and stock has long been a fertile one for discussion. 

 Nevertheless little accurate work has been done in studying the 

 problems thus involved. 



It is known, in a general way, that certain chemicals have 

 specific effects upon the color, composition or other character- 

 istics of fruits, but accurate data in this direction are scarce. 

 The fact of individuality in fruit plans is recognized, but its 

 importance as a factor in the development of a type has been 

 almost wholly overlooked. 



The fact of the existence of graft hybrids is freely maintained, 

 but the principles involved in the production of such forms 

 remain a closed book. 



In the past most discussions of pomological problems have 

 been empirical. There are certain principles underlying the 

 subject, however, which, in common with the improvement of 

 plants in general, are fundamental and far reaching in their 

 importance. It is to this class of problems, more scientific but 

 not less practical in their nature, that pomologists and plant 

 breeders alike are devoting thought and study at the present 

 time. The solution of some of these problems, and the classi- 

 fication of knowledge concerning the subject, is necessary in 

 order to raise pomology to the rank of a distinct science. 



