ElvISHA MITCHEIvL SCIENTIFIC SOCIETY. 23 



in fairly large amounts. Nitrates are continually 

 formed according to Bertlielot and Andre (Storer, 

 Vol. 1, pp. 307-8) in certain parts of plants. Here 

 the plant cells promote oxidation in a manner anal- 

 ogous to that of the micro-organism. They prove 

 this by inserting- portions of the stems of the amaranth, 

 plant into washed and sterilized soil. After a time a 

 notable nitrate formation had taken place in the soil 

 containing the plant stems, while none was found in 

 other soils similarh" treated but containing none of the 

 amaranth. Small amounts of nitrates are formed from 

 the action of electricity upon the nitrogen and oxygen 

 of the air. Rain water collected immediately after a 

 thunder storm invariably contains a greater percent- 

 age of nitrates than at other times. There are also 

 various oxidation processes continually producing small 

 quantities of nitrates. The}^ can usually be detected 

 in certain metallic oxides as ferric-oxide and manganese 

 dioxide, though it has not yet been explained in what 

 way these substances aid in the formation. The ac- 

 tion of all these agencies is necessarily slow, and some 

 nitrogen is lost to the soil, being given ofF in the free 

 state. In other changes, too numerous to mention, 

 great quantities of nitrogen jQ^i'lj return to the air. 



There must exist somewhere in nature a means of 

 supph^ing this deficiency or the visible supply of avail- 

 able nitrog-en would annualW become less and less. 

 The experiments of Lawes and Gilbert and Pugh in 

 England and Boussingault in France, in which combined 

 nitrogen was excluded b}^ a series of wash bottles, were 

 long ago accepted as proving that a plant could not 

 utilize free nitrogen of the air. Recent investigations 

 show that this is not true of leguminous plants when 

 aided bv the action of certain soil organisms. It is be- 



