STUDIES ON OAT BREEDING. 139 



2. Pure line selections. The second and third Hnes of work 

 have to do with the production of new strains or varieties which 

 will yield better under our conditions. In the pure line selec- 

 tions, individual plants are first selected from the general plots. 

 The following year the grain from each plant is planted in a 

 separate row and each row harvested separately. The next 

 year the grain from the most promising rows is planted in small 

 plots and in the third year this grain, all of which originated 

 from a single plant, is planted in a larger plot and tested under 

 field conditions. In this way it is hoped to obtain some strains 

 which will do better, under our conditions, than any of the 

 existing varieties. 



3. Hybridisation. The third Hne of work includes the cross- 

 ing of varieties or strains, each of which are superior in some 

 characters, and then to isolate from the progeny new varieties 

 combining the desirable qualities of both parents. 



Aside from the more immediate practical objects in view, as 

 noted above, this work is planned so as to furnish scientific data 

 on general questions of heredity. 



The work with oats was begun in the season of 1910. A 

 considerable amount of data on these several lines of work are 

 already in hand. It is expected that the present bulletin, which 

 deals with the results of the variety tests for the past four years, 

 will serve as an introduction to a series of papers dealing with 

 problems of breeding and inheritance in oats. In this paper we 

 shall describe ( i ) the cultural and experimental methods used 

 in these variety tests, (2) the seed used and the source from 

 which it was obtained, and (3) some of the results of these 

 four years tests. 



Methods. 



size and location of plots. 



For the years 1910, 1911 and 1912 the variety tests were car- 

 ried out in one-tenth acre plots, usually allowing only one plot 

 to each variety. In the season of 191 3 a different method was 

 adopted. On the basis of a series of careful experiments at the 

 Rothamstead Experiment Station in England, Mercer and Half 



'fiercer, W. B. and Hall, A. D. The Experimental Error of Field 

 Trials. Jour, of Agr. Science, Vol. IV, Pt. II, pp. 107-132, 191 1. 



