CEREAL IMPROVEMENT SELECTION 189 



value, as, being made up of two or more types, it may be 

 adaptable to more varied conditions than a pure line 

 selection. 



189. Mass selection at Svalof. Even at Svalof, 

 mass selection was at first practiced wholly, and has 

 never been entirely abandoned. Several good new 

 varieties were produced by that method. A selection 

 for density of spike was made from the native Plumage 

 barley, with the idea that this quality is correlated with 

 strength of straw, which was afterward found not to be 

 true. However, there resulted from this work two new 

 barley varieties. One thousand selected spikes were 

 divided into two groups showing the two extremes of 

 density, and a third group representing the average 

 density of all. From the seed of the group containing 

 the most open spikes, density 40 to 41 (see 110), was 

 ultimately developed a sort which remained relatively 

 lax, and was called Clay barley. From seed of the other 

 extreme group, density 45 to 48, the variety Moss was 

 produced, which at the end of three years still possessed 

 very dense spikes. By the same method Princess barley, 

 Renodlad Squarehead wheat, and Probsteier awnless oat 

 were developed. 



190. Maintaining the purity of varieties. An im- 

 portant use of frequent mass selections is for maintaining 

 the purity of all varieties, including pedigreed sorts. In 

 the report on seed wheat by Cobb (1903, p. 48), previously 

 mentioned, it is stated that " no farmer should allow 

 himself to forget that all the various items of cost in con- 

 nection with grading and selecting seed are much reduced 

 by selecting from his crop, before it is harvested, certain 

 good portions to be taken off specially for seed purposes." 

 The use of a special seed plat on every farm is strongly 



