* FIELD CROPS. 327 



The uveriigc of 26 samples of 182 grown in 1897 was a grain weight 

 of 52.2 nig. and a nitrogen content of 1.535 per cent, while the average 

 of all other samples grown that year was 61.3 mg. and 1.622 per cent. 



The results would seem to justify the conclusion that the correlation 

 ))etween grain weight and nitrogen content can be disturbed by sys- 

 tematic selection, and can not, therefore, be of paramount importance 

 in the improvement of plants. Correlations were found between the 

 length of head and the grain weight (the latter increasing with the 

 former), and between length of head and nitrogen content (the latter 

 likewise increasing with the former). But these correlations are still 

 less lixed than that of the grain weight and the nitrogen content. The 

 length of head and the number of kernels in the head, on the other 

 hand, stand in a much closer relation than the preceding factors, 

 although in the author's opinion a perfect correlation does not exist 

 even here. 



The theory of the incompatibility of valuable qualities in cultivated 

 plants, especially emphasized by Schindler and von Proskowetz, accord- 

 ing to the author, has but slight scientific importance, while, viewed 

 from a practical standpoint, it is only an expression of the fact that 

 certain ends aimed at in the improvement of crops are more difficult 

 to reach than others. Practical experiences must be carefully scru- 

 tinized before they are used for the support of the principle of hered- 

 it}'^ and variability. In the point under discussion one must be par- 

 ticularly on the lookout for complications brought about by external 

 conditions, the confounding of which with correlation phenomena may 

 cause considerable confusion. — f. w. woll. 



Some investigations of the relation of the size of grain to the 

 nitrogen content of -wheat and peas, W. Johannsex and F. Weis 

 ( r/V/.s-.sZv. Landhr. PI mi f earl, 5 {1899). pp. 91-100).~T\iq investiga- 

 tions were conducted with 5 difterent kinds of wheat, viz, Sonne Extra 

 Squarehead, Bahlson Squarehead, Urtoba Peters, Urtoba Metz, and 

 Light-colored East Prussian wheat. The results obtained show that 

 the different kinds of wheat grown in Denmark agree with the two- 

 rowed barley in the relation Ijetween grain weight and percentage of 

 nitrogen; as a general rule the percentage of nitrogen is increased 

 with increasing grain weight, but there are many exceptions to the 

 rule, so that this can not be used as an absolute starting point in 

 studies of plant improvement. 



In a supplement to this paper. W. Johannsen shows that this state- 

 ment also holds good in case of Victoria peas. Ten large peas had an 

 average weight of 0.450 gram, and an average nitrogen content of 

 3.35 per cent, against a grain weight of 0.226 gram and a nitrogen 

 content of 3.17 per cent for 10 small peas. The average of 20 deter- 

 minations was: Grain weight, 0.338 gram; nitrogen content, 3.26 per 

 cent. — F. w. WOLL. 



8873— No. 1 3 



