202 



grow and multiply. They are so small that they are not visible 

 to the naked eye ; fifteen thousand of them attached end to end 

 would not extend more than one inch. These bacteria may live 

 in the soil, and when they come in contact with a legume root 

 they make their way. into it and there begin to multiply. In a 

 few days the root develops a swelling, which is a nodule, near 

 the point where the bacteria entered. In the mature nodule are 

 millions of these bacteria. 



LEGUMINOUS PLANTS WITH NODULES EN.XBLED TO USE FREE 

 NITROGEN FROM THE AIR. 



Chemists state that four-fifths of the air is nitrogen — an un- 

 limited supply — but the plants that are raised on our farms can- 

 not use this nitrogen because it is a gas and is not available to 



them. It has been observed, how- 

 ever, that when nodules develop on 



the roots of a leguminous plant, 



that plant is supplied with nitrogen 



which comes from the air. The 



bacteria that produce the nodules 



seem to have the peculiar ability to 



use nitrogen from the air and in 



some way to supply the leguminous 



plant with it. It is not known how 



the bacteria in the nodules of the 



leguminous plant get nitrogen 



from the air, but it is known that 



a leguminous plant with ])lenty of 



nodules on the roots accumulates 



a relatively large amount of nitro- 

 gen inside its tissues, and that a 



certain part of this nitrogen comes 



from the air. 



.■I mount of nitro^^oi taken from 

 the air by a Ici^Hiiiinoiis crop. 



Since it is well known that 

 legumes use nitrogen from the air, 

 farmers are naturally interested to 

 know the amount of nitrogen that 

 may be taken from the air by a,,, .,., ,, , , ,. , ,i , , 



- -'I' Id. .\2. — JiOol of ( lUKidii /}( III p< <i, 



leguminous crop. 'Ihis is very ,^Uou-i,uj noduhs. Natural si:c. 

 difficult to decide. Many experi- 

 ments have been made in order to 



determine this, but such experiments have been ])erformed under 

 special conditions. 'J'he results obtained, therefore, must not be 



