100 THE CHBMISTKY OF THE FARM 



If the crops on this experimental rotation should be 

 permanently maintained in quantity, of which at pre- 

 sent we cannot be certain, we must conclude that 

 these 32 lbs. of nitrogen, together with the unknown 

 additional quantity lost as nitrate by drainage, have 

 been annually derived from the atmosphere, partly as 

 rain, but mostly by direct absorption by the crops or 

 soil. 



The loss or gain of nitrogen in the soil is, to a con- 

 siderable extent, under the farmer's control. Much 

 may be done to improve the land without the purchase 

 of nitrogenous manures. Attention should be given to 

 three points : (1) the diminution of the losses occurring 

 during the making of farmyard manure, and (2) during 

 autumn and winter drainage, and (3) the turning to 

 full account the power of leguminous crops to obtain 

 nitrogen from the atmosphere. The best treatment of 

 farmyard manure has been already considered (p. 51). 

 "We will now say a few words on the remaining points. 



The losses to the soil by autumn and winter drainage 

 may generally be prevented by skilful cropping. In the 

 case of a bare fallow it has been found advantageous 

 to sow mustard early in August, and plough the crop in 

 in October before wheat sowing ; the nitrogen of the 

 nitrates is thus converted into vegetable matter, and 

 preserved from loss by drainage. 



An ordinary rotation supplies in part the conditions 

 needed for the preservation of nitrates. "When clover 

 is sown among barley, and is left in possession of the 

 land after harvest, the protection against loss of 

 nitrates is complete. When, however, wheat is fol- 

 lowed by barley or turnips, the land is left unprotected 

 after the wheat harvest, and will lose large quantities 



