THE FOOD OF PLANTS. 65 



soil, in which according to experience peas grow well. To pre- 

 pare the extract we shake up 8 gr. of the soil with 100 c.c. of 

 water, and allow to stand till the greater part of the sand and 

 clay has settled, which often takes several hours. It must be 

 observed that the quantity of combined Nitrogen in such an ex- 

 tract is so slight that it may be neglected in our investigation. 

 Finally to the sand in a few cylinders is added soil-extract (again 

 25 c.c.), which has previously been sterilised by protracted boil- 

 ing. 



The culture vessels are placed in a room with a south aspect, 

 the windows of which are kept closed. It is still better to con- 

 duct the cultures in a special small plant house which is not 

 employed for the cultivation of other plants. 



The experiments, if carefully made, show that all the seedlings 

 at first grow normally. After three to four weeks, however, they 

 manifestly get into a state of hunger. The plants in the cylin- 

 ders without soil-extract and with sterilised extract do not re- 

 cover, but henceforth gain a bare existence. Their dry weight at 

 the end of the experiments, say after three to three and a half 

 months, is very slight, and Nitrogen estimations show that they 

 contain little or no more Nitrogen than the seeds sown.* The 

 hunger stage is soon passed by the plants provided with unsteril- 

 ised soil-extract. They thrive vigorously, develop ripe fruits and 

 seeds (this takes place with peas in closed spaces from which in- 

 sects are excluded), and produce large quantities of proteid. 



Peas, and the Papilionaceae in general, have therefore the ex- 

 ceedingly remarkable power of producing proteid in large quanti- 

 ties, although nitrogenous compounds are absent from the soil in 

 which they are growing. In the Gramineae, as also in many other 

 plants, this capacity is entirely wanting; they thrive normally 

 only when nutriment in the form of nitrogenous compounds stands 

 at their disposal. The Papilionaceae must be in a position to 

 utilise free atmospheric Nitrogen for proteid formation, and this, 

 as we shall see, can only come about through the intervention of 

 bacteria. 3 



If shortly before the time of blooming, or during this period, we 

 root up a pea plant (the observations may be made also with other 

 Papilionaceae) from the fertile soil in which it is growing, we shall 



* Any increase in Nitrogen which may appear is due to the sand not being 

 completely devoid of Nitrogen, and to the fact that the leave, as is known, can 

 absorb some ammonia from the air. 



P.P. F 



