58 IMPBOVING THE QUALITY OF WHEAT. 



When the crop was ripe in 1903 each plant was harvested sepa- 

 rately, and all of those resulting from spikes which the previous year 

 had shown a proteid nitrogen content of more than 4 per cent or less 

 than 2 per cent were analyzed, as were also a certain number resulting 

 from spikes of intermediate values. 



The good kernels on each plant were counted and weighed, thus 

 giving a record of the yield of each plant. From these data the 

 average weight of the kernels per plant was calculated. The specific 

 gravity was not determined and consequently the average volume of 

 the kernels on each plant was not calculated, as was done the previous 

 year. 



In Table 8 the plants harvested in 1903 are arranged in classes of 

 1 to 2 per cent proteid nitrogen, 2 to 2.5 per cent, 2.5 to 3 per cent, 

 3 to 3.5 per cent, 3.5 to 4 per cent, 4 to 4.5 per cent, and over 4.5 per 

 cent. The number and weight of the kernels on each plant are stated, 

 as is also the average weight of each kernel. The number of grams 

 of proteid nitrogen in all the kernels of the plant is shown, and also 

 the number of grams of proteid nitrogen in the average kernel on each 

 plant. Table 9 shows the average for each class. 



These results, so far as they cover the same ground as those of the 

 previous year, have the same significance. They show a quite uniform 

 although slight decrease in the weight of the average kernel accom- 

 panying an increase in the percentage of proteid nitrogen, and a very 

 marked increase in the number of grams of proteid nitrogen in the 

 average kernel. Especially marked is the increase in the amount of 

 proteid nitrogen in the average kernel, amounting to 28 per cent of 

 the weight of the kernel for every 1 per cent increase in the content 

 of proteid nitrogen. 



One column of this table, not contained in that compiled from 

 results of the previous year, shows the number of grams of proteid 

 nitrogen contained in all of the kernels on the plant; or, in other 

 words, the proteid nitrogen production of the plant. This appears, 

 on the whole, to increase with the percentage of proteid nitrogen, 

 although the results are not sufficiently consistent to permit of an 

 unqualified statement to that effect. The uneven stand of the plants, 

 before referred to, doubtless accounts for these inconsistent results. 



Two other columns contain data not obtained in 1902. The first 

 of these shows the number of kernels per plant, which apparently 

 decreases slightly as the percentage of proteid nitrogen increases, but 

 this can not be stated unqualifiedly. The next column shows the 

 weight of kernels per plant, or the yield per plant, which likewise 

 seems to decrease slightly with an increase in the percentage of pro- 

 teid nitrogen. 



