38 DOMINION EXPERIMENTAL FARMS 



respects as the work has advanced. In recent years, no prom- 

 ising kinds of spring wheat and scarcely any of six-row barley 

 have been obtained; but of two-row barley and oats some 

 interesting new sorts have been imported which are likely to 

 prove of value in Canada. 



While the foreign varieties hitherto tested have very often 

 been found somewhat unsuited to Canadian conditions, several 

 of them have been utilized as parent sorts for cross-breeding 

 purposes and have in this way contributed largely to the pro- 

 gress of the work. 



(3) Selection. 



While the progress which may be looked for by mere 

 selection has undoubtedly been seriously overestimated in some 

 quarters, it is, nevertheless, a valuable branch of experimental 

 work and one from which good results are being obtained, not 

 only in regard to the general appearance of the crops, but in 

 more important respects also. 



As most commercial grains are lacking in uniformity and, 

 sometimes, are strikingly complex mixtures, considerable 

 attention has been paid in the Cereal Division to the selection of 

 the best types from such of the commercial varieties as appeared 

 most valuable. 



Among the first sorts operated upon were Red Fife wheat, 

 Mensury barley and Banner oats. From each of these, several 

 good strains were isolated, the best of which were retained and 

 propagated. Two of these deserve special mention. Red Fife 

 H has not only proved of the highest quality, but has shown, 

 in some seasons, a slight advantage in earliness of ripening. 

 After several years' trial at Ottawa and at the branch Experi- 

 mental Farms, it has been used to supplant the ordinar}^ Red Fife. 

 It has also been widely distributed to farmers by means of the 

 annual distribution of seed grain from Ottawa. From Mensury 

 barley, an improved strain, now named Manchurian, has been 

 isolated. This has given very good results for several years 

 at Ottawa and elsewhere and is now displacing the parent 

 sort, both on the Experimental Farms and in the annual dis- 

 tribution. 



The method of selection employed is that which is now 

 generally recognized as the best, and which has been found to 

 give the most satisfactory results whenever fairly tried. At 

 the commencement, a number of plants or heads which appear 

 to be of the most desirable type are chosen. The seed from 

 each of these is sown separately and the groups of plants pro- 

 duced are careftBly studied. The best groups are retained, the 

 seed of each being kept separate. Study of this seed may 

 lead to the rejection of some lots. Those that remain are sown 



