EFFECT OF MINERAL MANURES 197 



crop from 14*5 to 19 '4 bushels per acre. This increase again 

 is chiefly found on the plots growing clover and beans, 

 and so receiving nitrogen collected from the air; the two 

 quarter plots which were fallowed after the barley, actually grew 

 till 1903 less than the corresponding unmanured plots. On 

 these latter plots the preceding growth of a comparatively 

 large crop of roots has removed so much nitrogen that the soil 

 is left poorer than on the wholly unmanured plot, which has 

 been taxed less severely, though both are alike in receiving no 

 supply of nitrogen during the whole course of the rotation. 

 From this we may conclude that, in the absence of nitrogen, 

 mineral manures are of no use to the barley crop, the magnitude 

 of which will depend on the amount of nitrogen available, 

 even when the mineral resources of the soil have been consider- 

 ably drawn upon. In other words, with barley on unmanured 

 land nitrogen starvation sets in long before the deficit in 

 minerals is felt, the reverse being the case with Swedes. 



Coming to the leguminous crop, the mineral manures have 

 a very powerful effect, although they are applied a year before 

 the clover is sown and two years before the crop is grown. 

 The increase brought about is from 9*9 to 42*3 cwt. of clover 

 hay, and in the case of beans, from 15 '8 bushels of corn and 

 87 cwt. of straw to 28*2 bushels of corn and 16 '9 cwt. of 

 straw. This illustrates well the generally accepted fact that 

 the leguminous plants are in the main independent of manurial 

 sources of nitrogen, which element they are able to draw from 

 the atmosphere, especially when they are provided with plenty 

 of mineral plant-food. 



In considering the wheat crop, it is necessary to distinguish 

 between the plots which have previously grown clover or beans 

 and those which have been fallowed, because in the former 

 case there has been such an accumulation of nitrogen in the 

 soil that the succeeding wheat crop is very much stimulated. 

 It will be seen that the crops on the fallowed portions 

 average about 25 '4 bushels per acre, as against 201 bushels 

 per acre on the corresponding unmanured plots, an increase 



