54 Journal of Agricultural Research voi. xvni, no. 2 



PRESENTATION OF DATA 



As a matter of record the original data are given in terms of yields 

 and of the composition of the moisture-free material. (See Tables I 

 and II.) 



All figures are, however, recomputed in terms of the absolute amounts 

 of the constituents reported. These are expressed both as grams per 

 plant and as parts per million of soil. (See Tables III and IV.) 



Inasmuch as the succeeding discussion centers about the graphs, 

 presented in some detail, a word concerning these is perhaps necessary. 

 The 2-week interval between cuttings, used for the most part, appears 

 to have been sufficiently short to bring out the more important charac- 

 teristics of growth and absorption. It will be noted, however, that in 

 the 1 916 experiment the first cutting did not take place until six weeks 

 after planting. The fact that the plants did not appear above ground 

 for several weeks and the further fact that the earlier harvests of the 

 1 91 7 experiment showed very characteristic changes, indicate that the 

 use of a straight line to cover this period in the 191 6 graphs is highly 

 artificial. However, the method is consistent in that we have con- 

 nected only known points; and the facts as to absorption and growth 

 at this stage are sufficiently clarified by the 191 7 experiment. 



Since our data are derived from two relatively independent studies 

 of the growth of barley on two very different soils in two different calen- 

 dar years, it is obvious that differences may be expected to appear 

 which are equally assignable to one or another of the conditions of 

 growth such as soil, season, etc. Our present purpose is not to account 

 for such differences, but to call attention to certain similarities which 

 are rendered all the more striking in view of the great differences in the 

 yields actually obtained. 



