The following paper was read by H. F. Roberts: 



METHODS OF CEREAL BREEDING IN KANSAS 



By H. F. Roberts, Botanist at the State College, Manhattan, Kansas, 



The botanical and chemical departments of the Kansas State Agricultural 

 College, at Manhattan, Kansas, have in charge, conjointly, the work in plant 

 breeding at the Kansas Experiment Station. Hitherto our efforts have been 

 concentrated upon the improvement of wheat and corn. During the present 

 season, however, the breeding experiments have been extended to cover rye, 

 oats, barley, kafir corn, soy beans and cow peas. But since the investigations 

 have thus far been confined exclusively to wheat and corn, it is to a discussion 

 of our work with these cereals that this paper will chiefly confine itself. 



Whether the cereal wheat has arisen from a few or from many wild 

 forms, it is certainly true that the wheat of to-day consists of a. very great 

 number of sub-varieties. A visit to almost any average wheat field reveals the 

 presence of several such varieties, which, despite the various manifest differ- 

 ences in the character and quality of their seed, in the quantity of yield, in 

 the vegetative characters of the plants, etc., are allowed to grow undisturbed 

 together, in the end are harvested together, and the heterogeneous mixture of 

 seeds is used again as a basis for the next year's sowing. Now and then 

 one finds an occasional wheatgrower who endeavors by careful selection to 

 breed his wheat up to a uniform type. Such instances, however, are not 

 numerous, nor is it even the case that a large number of wheat raisers practice 

 even the simplest form of seed selection. Indeed, I have found it to be the 

 case in many instances, that farmers will systematically retain only the lightest 

 and least serviceable seed for planting, under the impression that it will yield 

 just as well as the larger, plumper and heavier seed which they dispose of in 

 the market. 



Our experiments in the breeding of wheat began in the fall of 1898. Since 

 that time we have originated in the neighborhood of 150 crossbred strains of 

 winter wheat. Of these, 36 are now growing in our experimental plots at Man- 

 hattan, the remainder having been gradually eliminated through a rigid process 

 of selection, and through extinction in the face of trying climatic conditions. 

 As the result of these crosses we have secured enough varying lines to enable 

 us to begin the process of breeding by selection. 



That portion of the State of Kansas in which the experiment station is 

 located lies in what is known as the red winter wheat region, and our 

 problems in wheat breeding have largely been connected with the improvement 

 of wheat locally grown and especially adapted to that district. The western 



