\VE have ah'eady shown 1 that at least certaiu of tile Leguminosae have a distinct effect upon the process of nitrification in the soils in which they grow. Their beneficial influence on nitrification may be connected with the symbiotic relation they bear to nitrogen fixing bacteria. It is conceivable that the root tuber des with their abuudaut supply of nitrogenous matter may by their deea), even during the life of the plant, so increase the supply of easily decomposable nitrogeuous matter in the soil as to account for tile increased nitrifying power of legume bearing soils. If this be the explanation for the phenomenon it marks a sharp line in the 1-elation of legumes loward the soil ill distinction from non-legllmes.
Experiments that we are Collducting with cereals aml with other non-legumes indicate that these crops differ one frolu another in their relation to tile occurreuee of nitrates in the soil on which they gl-O~. The nitrate content of the soil is largely dependent on the kind of crop grown. The extensive literature ~Journal of Industrial and Engineering Chemistry (igio), ii, 313-3J5. NOTE.--The Franklin Institute is not r~ponsible for the Statements and opinions advan~d by contr butors to the Journal. VOL. CLXXI, NO. to2I--1 /~-'~'y'~.
LYON ANn ]{[ZZIΒ£LL~ bearing on the nitrate content of soils contains a number of instances illustrating this relationship between the soil and the crop growing on it, but in no case has any particular significance been attached to it. Deherain 2 found a greater loss of nitrates from the soils growing peas and clover than from those with grain crops in the drainage of the soil tanks at Grignon. He accounts for this through the inability of the legumes to take up nitrates as readily as do the graminaceous plants.
Webbel a took soils from tiehls in which the character of the soil was similar, but in which the crops were different. These were placed in lysimeters. While in the lysimeters no crops were grown in the soils, but they were allowed to leach ; the drainage being collected and analyzed.
Soil was taken from fallow land, wheat land and alfalfa land. The soils were kept in lysimeters from Nov., 19o 4 to Oct., I9o6. The nitrates expressed in grammes HNO~ per square metre surface amounted to 4 Β° Gin. from the fallow land and 48 Gin. from the wheat land in the lysimeters 25 era. deep aml 66 Gin. from the fallow land, 77 Gin. from the wheat land and 78 Gin. from alfalfa land in lysime~ers 5o cm. deep. With Ebermeyer lysimeters, which are funnels placed under a column of soil in the field by hollowing out the soil under the column, they also obtained more nitrates in the drainage from wheat soil than from fallow land, both where the column of soil was 25 cm. deep and where it was 5o cm. deep.
King and }Vhitson 4 report an experiment in which one of the two contiguous plats of oats bad a crop entirely removed on June 2o, on which date determinations of nitrates were made in the soil of both plats to a depth of four feet. The crop on the harvested plat was weighed and dry mattdr determined. At the end of nineteen days nitrates were again determined in both plats and the remaining oats were harvested, weighed and dry matter determined. No determinations of nitrogen were made in the crop, but the gain in nitrogen for the nineteen days was calculated from the dry weight on the basis of average analyses ~Ann. Agron. (i891), xvii, 49-82.