382 



EXPERIMENT STATION RECORD. 



tiou is o-iven as for tlu^ previous lot. The plants of the 3 seediii^s upon 

 analysis yielded as follows: 



AnaJijHhs of airial parls of plants. 



The nitrogen balance in grams for the whole series is as follows: 



NUroqen halaure. 



Peas. 



Mustard. 



I Grams 



Nitrogen in original soil j 3. 320 



Nitrogen m seed 1 0. 401 



Nitrogen in liarvest 0. 084 



Nitrogen m roots 0. 175 



Nitrogen in soil alter har- 1 



vest I 3.399 



!. 721 



Total gain in nitrogen. 



Grams. 



3. 320 ^ o 



0. 018 5 "■ 



0. 237 I 



0, 068 ( o 

 |.3. 



3. 269 J 



Buckwheat. 



Gra'ins. 

 3. 320 ^ 

 0.l)27r 

 0. 234 I 



"•"''2ln.C02 

 3. 326 I 



Oats. 



Grams. 

 3. 320 i 3 3gg 

 0. 048 J "*■ ^^ 

 0. 201 1 

 0. 226 [ 



Check 

 pot. 



- 4. 105 



3. 618 J 



Grams. 

 3.320 



3.374 



0. 054 



From the la.st table it is shown that there were gains made in every 

 case over the original. In the case of the check pot the gam is so 

 slight as to fiill within the limits of error. 



The first table shows that the peas alone were able to acquire the 

 nitrogen of the air, and the other 3 genera showed a decline in spite of 

 the increased amount of nitrogen in the soil. It goes to show that (1) 

 peas and other tubercle-forming plants occupy a position separate from 

 the non-tubercle-forming plants so f;iT as nitrogen assimilation is con- 

 cerned, and (2) the 3 non-leguminous plants were unable to take up the 

 nitrogen, which Avent to enrich the soil. 



From the plants of the second and third sowings it can not be claimed 

 the increased nitrogen was acquired through the plants, as they would 

 hardly be stunted in growth due to a deficiency of nitrogen while stor- 

 ing it up in the soil. There seems no other explanation than that the 

 enrichment of the soil takes place in the soil itself, as has been shown 

 by Winogradsky, lierthelot, and others, as due to the action of soil 

 bacteria. The nitrogen so stored up is not immediately available to the 

 plants, but is rendered so by nitriticatiou. 



The author.s' exi>eriments in 1891 tend to su})port this supposition. 



The formation of root tubercles among legumes, C. Nattdin 

 {Jour. Af/r. .J*r((f., '>s {1891), Rq.SD, pp. 153, 45-1). — ^The author mentions 

 the unsuccessful attempts to grow Lesptcdcza vi)rjaf((, a S])eciesof Cassia 

 from Senegal, Balsamocarpou brevifoUum, and other species of tropical 



