Depths at Which Plants Absorb Water and Nutrients. 95 



metal hoops fastened with small bolts. When filled with firmly com- 

 pacted soil, the smooth overlapping edges fit tightly, while at the end of the 

 experiment the whole core of soil could easily be exposed for examination by 

 simply removing the bottom of the containers and the hoops (plate 9, a and b). 

 Plants grown for shorter periods were in containers of the same design, but 

 with a diameter of 1 foot and a depth of either 2 or 2.5 feet. In this experi- 

 ment 16 containers were used, but the plants in one were accidentally de- 

 stroyed. 



On May 19 to 21 two long trenches were dug to a depth of 2 to 3.5 feet, 

 the rich, alluvial, silt-loam soil being kept separate from the somewhat more 

 clayey subsoil. The containers were then placed in a row in the trenches 

 about 6 inches apart, the lower end of each container resting evenly upon 

 its inverted bottom. The soil which had been removed from the trench was 

 next mixed very thoroughly by shoveUng it back and forth on an improvised 

 platform. It was in excellent tilth, with a water-content of about 30 per cent. 

 The subsoil, similarly treated on a separate platform, had approximately the 

 same water-content. After mixing, the subsoil and surface soil were slowly 

 poured into the containers and continuously tamped, so that thej'' were com- 

 pacted to a degree not unhke that of the natural soil in the adjoining plats, 

 where roots were repeatedly excavated. When filled, the soil in each con- 

 tainer was again at approximately the same level it had formerly occupied. 

 In filhng the containers, duphcate samples were taken at the several levels 

 (mostly at 6-inch intervals) for water-content. Moreover, at the time of 

 filling, wax seals were inserted in about one-half of the containers between the 

 several layers of soil at intervals of 0.5 foot, beginning at 6 inches depth. 

 Finally, each container was fitted with a sloping wooden roof, the soil having 

 been mounded up so as to fit snugly under the roof. Thus no rain could enter 

 except through a central sHt three-fourths inch wide and with a length equal 

 to the diameter of the container, 1 or 1.5 feet respectively. Through this 

 opening barley was planted thickly at a depth of 1.5 inches on May 19, but 

 in only the odd-numbered containers, the ones with even numbers being 

 used as checks. The trench about the containers was refilled, the soil thor- 

 oughly compacted and ridged up in such a manner that drainage was away 

 from the disturbed area, the surface-water being carried away through 

 shallow ditches about 2 feet distant on either side of the original trench. 

 Grass was allowed to grow on this disturbed area which adjoined the crop 

 plats. At four periods during June and July, when the crop was in various 

 stages of development, certain containers were removed, and the root dis- 

 tribution and the water-content determined. The size and arrangement 

 of containers and the position of the seals, as well as the relative heights and 

 root development of the several plant-groups at the time of examination, 

 are shown in figure 40. 



On May 25 to 27, when the barley was about an inch high, the plants 

 were thinned to 20 and 25 in each of the small and large containers respectively. 

 By June 4 they had reached a height of 4 inches and practically all had 3 or 4 

 leaves. The much more rapid development of this late-planted crop than 

 that sowed April 9 (p. 41) is clearly correlated with environmental conditions 

 which have already been discussed (c/. p. 44 and fig. 20). On June 15 the 

 plants were 6 to 8 inches high; some had 2 or 3 tillers. At this time they were 



