232 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



of nitrogen for plants was the fixed nitrogen supplied through the roots 

 from the soil. Plants were shown to be incapable of utilizing any of the 

 enormous supplies with which they were surrounded in the atmosphere. 

 This conclusion was confirmed by the researches of Lawes, Gilbert, and 

 Pugh * at Kothamsted, and for the vast majority of plants it is held to be 

 true to-day. It was utterly at variance, however, with what was known 

 concerning the behaviour of legumes in the field, and it was sought to 

 reconcile the facts by the suggestion that there was some peculiarity 

 inherent in the legumes themselves. It was not until long after, in 

 1880, that Hellriegel t announced that these plants unquestionably 

 obtained their nitrogen from the atmosphere, and in 1888 he and 

 \Yilfartht showed that leguminous plants were capable of reaching 

 their full development when they were grown in a soil free from nitrogen, 

 but only when they possessed nodules upon their roots. In sterilized soils 

 nodules were not formed and the plants died of nitrogen starvation. 

 Thus the presence of nodules was shown to be coincident with the fixation 

 of nitrogen from the air. 



The nodules had been noted long before. Malpighi ^ is the first, 

 apparently, to mention them, and he looked upon them as root galls. 

 Others who described them considered them normal structures, while yet 

 others regarded them as a diseased condition of the root. Woronin ! 

 first described their structure in detail, and stated that he had found them 

 to contain small organized bodies, probably bacteria, an announcement 

 that was contradicted by other observers, who looked upon the bacteria- 

 like structures as proteid bodies, but confirmed by Eriksson and others 

 very soon after. 



Legumes are seldom without the nodules on their roots either in this 

 country or abroad ; indeed, the only leguminous plant on which nodules 

 have never been found is Gleditschia triacanthos. Frank showed that the 

 formation of nodules could be prevented by sterilizing the soil in which 

 the plants were grown, thus proving that the nodules owe their origin to 

 microbes living in the soil. This investigator adopted other views later, but 

 Marshall Ward inferred that the nodules " so common on the roots of 

 the bean are due to the action of the fungus the very minute germs of 

 which are all but universally present in the soil," basing his inference 

 partly on the facts that the " tubercles are all but invariably developed 

 within a month when the beans are germinated in sand or soil not 

 previously heated " ; (2) they are not developed in media sterilized by 

 heat ; (3) the roots may be infected by means of pieces of the old nodules 

 placed among the root hairs ; and (4) the infecting hypha may frequently 

 be discovered entering the cortex by means of the root hairs. This view 

 ol t he origin of the nodules is now generally held. 



As we have stated Hellriegel and Wilfarth showed that plants in 

 sterilized soil, lacking nodules, died of nitrogen starvation, but when the 



A wry complete and interesting account of these researches is contained in The 

 Book of Hi>- Rothanuted Experiment*) by A. D. Hall, M.A., 1006. 

 t Taatbl 59. V< rmmml. heat. Naturf. u. Aerste, Berlin, vii. p. 290. 

 t ZeUmhr. Par. hrut. Hubmz. Tndast., Nov. 1888. 

 55 Ojvra Omnia, AnaUnn. Plantamm, 1C>87, vol. ii. p. 126. 



Mi'm. Acad. Imp. Sri. Saint l'ttersb. vii. ser. 10 (1886), No. 6. 

 ' I'liil. Trau Hoy. Soe. Lond. Ser. B. 178 (1887), p. 556. 



