American Experiments 231 



placed in the ground together with the seed. The plants 

 thus inoculated developed normally^ and produced an 

 abundance of tubercles. 



This experience demonstrated the need of soil-inocu- 

 lation for soybeans. Many cases are reported in experi- 

 ment station literature in which thefee inoculations gave 

 positive results. For instance, in the experiments of 

 the New Jersey Station, on light sandy soils at Ham- 

 monton, when cowpeas and soybeans were planted in 

 the same ground, the former grew luxuriantly and 

 gathered nitrogen from the air by means of their numer- 

 ous nodules, while the soybeans remained small and 

 yellow and produced no tubercles. It was not until the 

 introduction of some soil from a field where these plants 

 had been grown successfully for several years that the 

 soybeans developed properly and grew as luxuriantly 

 as did the cowpeas. 



Similar observations were made time and again in 

 the case of alfalfa. Soils to which this crop was new 

 usually required inoculation, even though they had 

 successfully produced red or crimson clover. It appeared, 

 as in the case of soybeans, that the bacteria capable of 

 producing nodules on alfalfa were absent, as a rule, 

 from soils in which this crop had not been raised before. 

 This observation has led to the rather common practice 

 of scattering old alfalfa soil on new fields where this 

 plant was to be established. It was found subsequently 

 that the bacteria causing nodule-formation on alfalfa 

 were seemingly identical with those producing tubercles 

 on sweet clover. Inoculation was therefore superfluous 

 on soils to which sweet clover was native, and, more- 



