10 PLANT BREEDING. 



formation, nor even in keepinj? np and improving old forms. They 

 have liere a most legitimate field; unless they develop tliis feature, 

 they remain only seed merchants, and can hardly hope to gain that 

 position in the minds of purchasers which they should hold to have 

 the most profitable seed trade. 



Plant breeders need not work along nan-ow lines nor follow set 

 rules. There are many ways of conforming to tlie broadly applicable 

 principles of producing, finding or discerning, fixing, and multiplying 

 desirable variations. The devices of each plant breeder will solve 

 the problem for his environment and for the objects he seeks. Plans 

 outlined to suit the needs of one may not wholly meet the require- 

 ments of another situated differently. 



The great system of American experiment stations has fairly begun 

 the work of variety formation in important plants. We can, there- 

 fore, hope for the subject to be gradually placed upon a thoroughly 

 scientific basis. 'J'here is room and need in this line of work for the 

 best energies of all the agencies, amateurs, seed growers, seed firms, 

 and experiment stations, including that largest one, the national 

 Department of Agriculture. 



The writer is under obligation to Mr. Andrew Boss, assistant in 

 agriculture in the University of Minnesota, and to gthers who have 

 assisted in the experiments (as yet not all reported in station bulle- 

 tins) upon which this paper is in part based; also to student assist- 

 ants who have aided in preparing the illustrations. 



GENERAL OBSERVATIONS ON PLANT BREEDING. 

 RELATION OF PLANT BREEDING TO WEALTH. 



The economic results from plant improvement are already enormous 

 in the aggregate, and the possibilities for the future are so great as to 

 be truly dazzling. Circular No. 11, recently issued by the Division of 

 Statistics of the U. S. Department of Agriculture, for example, esti- 

 mates the world's wheat product for 1899 at over 2,500,000,()(){) Imshels. 

 Assuming the yield to be 20 bushels per acre, this required 125,000,000 

 aci-es of land. In ten years the Minnesota station, by careful breed- 

 ing, produced a new variety of wheat, which yielded nearly 25 per 

 cent more grain on the university farm than its parent varietj', which 

 was the best variety generally grown in the State. The following 

 table (Bulletin No. 62, Minnesota State station) shows comparative 

 yields of the new and the parent variety on several experimental 

 farms. In the last two columns appear the average results obtained 

 in five successive seasons at the university farm, near Minneapolis. 

 It should be observed that this new variety has not proven so much 

 superior in yield to the parent variety on other experimental farms a 



