VALUE OF THE INTRODUCTION. 35 



the kernel, but considerably below the best in this respect, the Sixty- 

 Day and the Tartarian being usually the best. In average weight 

 per bushel it was good and stood rather high in weight of 1,000 

 kernels. 



On the whole, the tests just mentioned indicate that the Swedish 

 Select oat is usually high in weight per bushel and very high in abso- 

 lute weight of kernels. At the same time, the percentage of meat in 

 the kernel is found to be higher than might be expected, always above 

 the average, and sometimes unusually high. The combination of 

 these qualities therefore indicates that the feeding value from these 

 standpoints is much above the average. 



CHEMICAL ANALYSES. 



For comparison, many chemical analyses of this oat and of others 

 have been made by Dr. J. S. Chamberlain, formerly of the Bureau of 

 Chemistry of this Department, in connection with laboratory work 

 in cooperation with this Bureau. These analyses show no extreme 

 quality in the Swedish Select, but indicate that it is a variety of good 

 feeding value, possessing a protein content above the average and a 

 narrow nutritive ratio; that is, the ratio of protein to carbohydrates 

 and fats is rather high. There is an unusually important feature of 

 these analyses, however, in the fact that the average results with 128 

 samples of domestic-grown Swedish Select oats show a considerable 

 increase in protein content over the original introduced seed. This 

 is very gratifying, in view of the claim sometimes made that intro- 

 duced crops, although an improvement at first, tend to deteriorate 

 afterwards. 



VALUE OF THE INTRODUCTION. 



As previously mentioned, the Swedish Select oat has received the 

 greatest attention in Wisconisn, though also grown to a large extent 

 in South Dakota and Montana. It has been shown that the 10-year 

 average yield of this oat at the Wisconsin Agricultural Experiment 

 Station was 8£ bushels greater than the average of all other varieties. 

 Also, calculating the averages of the highest yielding varieties for 

 several different periods, in which all varieties were grown every year, 

 there is a constant difference in yield of about 12 bushels per acre in 

 favor of the Swedish Select in comparison with the next highest variety. 

 Taking the mean of these two determinations, or, rather, reducing 

 the latter one (for it is the most accurate), we can probably safely 

 assume a constant average increase in yield of at least 10 bushels per 

 acre for the Swedish Select over that of other varieties in these tests. 



As all varieties were grown under the same conditions it is reason- 



aSee "The Feeding Value of Cereals," Bulletin 120, Bureau of Chemistry, U. S. 

 Dept. of Agriculture, 1909, pp. 18-25. 

 182 



