Oct. IS, I9I9 Rate of Absorption of Soil Constituents 53 



at one time, but may be regarded as vitiating the data obtained from the 

 soils for certain other purposes. To obviate criticism on this point we 

 shall in the present paper confine our conclusions as to the soil to those 

 obtained from the data of the 191 7 experiment, although these are in 

 all general relations confirmed by the work of the preceding year. 



CHEMICAL TECHNIC IN THE EXPERIMENTAL WORK 



The water extractions of the soils and the analyses of the extracts were 

 made by the methods described by Stewart {8) ; the analyses of the plants 

 were made by the usual methods of treatment for nitrogen and plant 

 ashes, with proper precautions to prevent the loss of partially volatile 

 constituents. 



MECHANICAL DIVISION OF PLANTS 



Since the object of the investigation was to bring out certain general 

 relations between the type crop and the soil, it was not deemed desir- 

 able or profitable to make a complete separation of the plants based on 

 their more minute anatomical structure. The well-known and sub- 

 stantial differences in composition between the heads of grain crops and 

 the vegetative tissues seemed, however, to require the separation of 

 the plant into at least these two parts; furthermore the roots obviously 

 deserve separate consideration. The procedure actually followed was 

 to remove the entire plant with as much of the roots as possible. In 

 all cuttings after the heads had formed these were separated from the 

 vegetative portion, designated stems and leaves. The roots of all 

 plants were severed by a lateral cut through the base of the crown. 

 As separated, this portion of the plants therefore included a small por- 

 tion of the lower part of the crown. The roots so recovered included 

 only a small proportion of the finer rootlets. Most of these had broken 

 loose from the plants and were to be found throughout the soil in very 

 fine and fragile strands which it is hopeless to expect to remove in their 

 entirety. 



The results reported hereafter on the upper part of the plant make it 

 appear probable that a study of root composition would give important 

 data on the mechanism of plant absorption, but that such a study should 

 be based on very exact measurements of the quantities and composition 

 of the roots. For this purpose natural soils, even when the roots are 

 washed free, do not appear to offer the best medium for growing the 

 plants, inasmuch as the data would always be subject to the criticism 

 that the recovery was incomplete, or that the roots had sustained losses 

 of solutes from the washing or gains from adhering soil particles. On 

 this account we abandoned the idea of making a chemical study of the 

 roots collected in the plot experiments, but will present evidence from 

 other sources as to the effect of root composition on our conclusions. 



