No. 6. DEPARTMENT OF AGRICULTURE. 207 



gen direct from the air could survive such conditions. In this way 

 a strain of bacteria was developed which was, so to speak, "nitrogen- 

 hungry," and when again released in the soil they showed the re- 

 sult of this forced activity in using atmospheric nitrogen. This re- 

 sult was not arrived at in a single test nor even in a single season. 

 Continuous work was done in laboratory, greenhouses and at the 

 Government Experimental Farm at Arlington for three years. In 

 one season, something over 10,000 separate inoculation experiments 

 were conducted on small plots at the Experimental Farm, and the 

 practica' side of the question was continually kept in view. 



It was not considered sufficient that the bacteria should merely be 

 grown without deterioration: the constant aim was to increase by 

 artificial methods of cultivation their usefulness as "nitrogen-fixers" 

 and to adapt the cultures to actual use under field conditions. You 

 are all familiar with the marvels which horticulturists have pro- 

 duced by scientific plant-breeding and selection. Who can say that 

 similar results, at least, are not possible in dealing with microscopic 

 plants — for instance, the nodule bacteria of legumes? As a matter 

 of fact, the methods of growth previously outlined have been shown 

 by actual tests in the laboratory to produce bacteria of higher effi- 

 ciency n gathering nitrogen than the forms grown without regard 

 to the laws of plant breeding and selection, and their effect when in- 

 troduced iv the soil has in many cases verified the belief that these 

 highly-bred forms are actually given an advantage over the wild 

 forms existing in many soils. 



It was not only necessary to" properly grow the bacteria but they 

 must be distributed in a form which the farmer could handle for 

 himself. Hence, resulted the "dry culture." By growing the bac- 

 teria first in a liquid culture, soaking this up with absorbent cot- 

 ton and drying the cotton, it was found that bits of this dried cotton 

 would serve as "starters" for new cultures. The bacteria on the 

 cotton had only to be immersed in water and supplied with a few 

 simple nutrient salts in the solution to resume their interrupted 

 growth and thus soon fill the water with active bacteria. By sprink- 

 ling this liquid culture upon the seeds to be sown, the bacteria may 

 be carried into the soil and assist the plants in securing their nitrogen 

 supply; or, if the liquid be mixed with clean soil and this inoculated 

 soil spread over the prepared ground previous to seeding the same 

 result would be obtained. This method seemed sufficiently simple 

 to be handled by any person of average intelligence, and accordingly 

 plans were perfected by the spring of 1904 to test the method on a 

 large scale. During last spring, summer and fall about 12,500 pack- 

 ages were sent out, reaching farmers in nearly every state and ter- 

 ritory and supplying cultures for tests with all the common legumes. 

 The whole success or failure of the method, of course, depended upon 



