INVESTIGATIONS ON THE PRODUCTION OF PLANT FOOD. 183 



proof has ever been given. The fact that land laid down to grass 

 tends to gain nitrogen has been adduced in proof, but it is not really 

 satisfactory. 



Table showing Gain in Nitrogen in Land laid down to Grass 

 in 1856 and mown annually. f.othamsted. 





! 1856 



1879 



1888 



1912 



Fer cent, of Nitrogen in top 9 inches . . 



1 152 



•205 



1 



•235 



•338 



On such land the ammonia and nitrates are assimilated more 

 rapidly and completely than in arable land, and there is therefore less 

 liability to loss of nitrogen ; further, leguminous herbage almost 

 invariably grows for part if not for the whole of the year, making 

 additions to the nitrogen supply. There is nothing to preclude the 

 action of the free-living nitrogen fixers, but nothing to prove it. 



These accumulated nitrogen compounds, whether built up by 

 organisms associated with the Leguminosae or by the free-living 

 forms, all break down by the processes just indicated and pass into 

 nitrates with or without loss of gaseous nitrogen, according to the 

 conditions. 



From the standpoint of plant nutrition we may look upon the 

 formation of nitrates in the soil as being the most important of all 

 the processes, and for long it was regarded as probably the only one 

 with which the agricultural chemist need concern himself. 



But recently another possibility has been opened up. Not long 

 ago the animal physiologists found that a mixture of fats, carbo- 

 hydrates, proteins, and certain mineral substances, which ever since 

 the beginnings of animal physiology had been looked upon as a satis- 

 factory diet for any animal whatsoever, is really not sufficient. 

 Hopkins found that animals could not build up their tissues when 

 fed with certain pure proteins, notwithstanding the ample nitrogen 

 supply. Certain molecular groupings are needed, and in practice 

 these can only be obtained in a mixed diet. 



Now it is very attractive to apply this to plants. We all know 

 how the gardener hankers after a mixed diet for his plants, and more- 

 over we all know that the opinions of a good practical man are entitled 

 to respectful consideration. It would be a simple matter to draw 

 a parallel between the plant and the animal, and to suppose that the 

 plant requires the mixture in order to ensure the supply of all necessary 

 molecular groups. 



There is, however, no experimental basis for such a view. Plants 

 have considerably greater power than animals of building up the 

 necessary proteins, and they can be, and at Rothamsted they regularly 

 are, grown to perfection in water cultures containing no other nitrogen 

 compound but sodium nitrate. We can find no evidence whatsoever 

 that it is necessary to add any other nitrogen compound. But although 

 not necessary it might be advantageous to the plant to receive some of 



