62 PHYSIOLOGY OF NUTRITION. 



amount of Nitrogen in the seeds used with that in the plants ob- 

 tained, we shall find, provided the experiments have been carefully 

 carried out, that the differences are at most such as are due to 

 small errors of experiment. The plants are not in a position, 

 under the conditions specified, to utilise the free Nitrogen of the 

 atmosphere for the formation of proteids. 2 



Similar experiments are to be made if we desire to satisfy our- 

 selves that seedlings vegetating in darkness neither take up free 

 Nitrogen from the air nor suffer loss of Nitrogen. We have only 

 to take care that the seedlings are not exposed to rays of light. 3 



1 On Nitrogen determinations according to the methods of Dumas and Kjel- 

 dahl, with the mode of preparing the necessary normal acids, etc., see Konig, 

 Arileitung zur Untertuchung landwirthschaftl. wiclitiger Stoffe, 1891, pp. 150 and 

 682, and Fresenius, Quantitative Analysis. 



2 This fact has been particularly determined by Boussingault (see Compt. 

 rend., T. 39, p. 601). 



3 I have collected together my own results and those of other observers on 

 this subject in my Veryleichenie Physiologie des Keimungsprozesses der Samen, 

 1830. 



20. Bacterium Radicicola and the Papilionaceae. 



The question of the significance of free Nitrogen as food for 

 plants has frequently been experimentally investigated. Till 

 recently, bearing in mind the results of Boussingault's researches 

 and those of other investigators, scientists in general held the 

 view that plants were unable to employ elementary Nitrogen in 

 the process of nutrition. The conditions under which Boussin- 

 gault's experiments were made actually excluded the possibility, 

 as we now know, of free Nitrogen being worked up in the cells. 

 Under other conditions, however, the process can take place, and 

 the great honour of having determined this with certainty be- 

 longs especially to Hellriegel. 1 



To obtain general information we make the following experi- 

 ment, which, however, can only be conducted during the warmer 

 part of the year. 



A number of cylinders 24 cm. in height and 14 cm. in diameter, 

 capable of holding about 4-5 kg. of sand,* are covered at the 

 bottom to a depth of 3 cm. with washed and ignited fragments of 



* In my experience good results are also obtained with cylinders holding only 

 2| kg. of sand. 



