450 



SCIENCE. 



[N. S. Vol. XXII. No. 563. 



practical man himself; yet a knowledge of 

 their results may be indispensable to the 

 expert whom he consults about the char- 

 acter or management of his land, however 

 trivial and workaday the actual question 

 may seem. 



In any agricultural experiment station 

 worthy of its name, place should be found 

 for investigations of this latter class; un- 

 fortunately many such institutions are un- 

 der the necessity of showing 'results' which 

 immediately appeal to the practical man 

 and may be taken to justify the expendi- 

 ture of public money; so that it is only 

 by side issues, as it were, and by degrees, 

 as the general public can be brought to 

 trust its scientific men, that such work 

 will be undertaken. 



It is not my purpose, however, to deal 

 to-day with this form of abstract research. 

 I rather propose to point out certain lines 

 of work in agricultural science which are 

 now being pursued with increasing vigor, 

 and which, from the very outset, promise 

 to have considerable applications in prac- 

 tical life. 



It is in the domain of agricultural bac- 

 teriology that perhaps the greatest prog- 

 ress has been recently made, in the main 

 progress in connection with that perennial 

 problem— the sources of the nitrogen of 

 vegetation. From the very beginnings of 

 agricultural chemistry, which we may very 

 well date from the publication of De Saus- 

 sure's 'Recherches Chimiques sur la Vege- 

 tation' in 1804, discussion has raged round 

 this point. Liebig, in his famous report 

 to this association in 1842, regarded the 

 atmosphere as the source of the nitrogen 

 contained in the plant; but in the long 

 controversy that followed, the view finally 

 prevailed that the plant was only able to 

 utilize already combined nitrogen in the 

 soil, so conclusive seemed the experiments 

 conducted by Boussingault and by Lawes, 

 Gilbert, and Pugh at Rothamsted, But a 



fresh turn was given to the whole question 

 by the discovery made by Hellriegel and 

 Wilfarth in 1887 that the leguminous 

 plants in virtue of the bacteria living 

 symbiotically in the nodules on their roots 

 were able to fix atmospheric nitrogen. 

 From that time research has been directed 

 towards the problem of utilizing and ren- 

 dering more effective this particular Bac- 

 terium radicicola. Widely distributed as 

 it is in the soil, it is yet not universally 

 present ; heaths and peaty soils, for ex- 

 ample, that have never been under culti- 

 vation frequently lack it entirely; conse- 

 quently, it is impossible to obtain a satis- 

 factory growth of leguminous crops, upon 

 which in many cases the possibility of suc- 

 cessful reclamation is based, until this class 

 of land has been inoculated with the ap- 

 propriate organism. 



Again, although but one species of bac- 

 terium seems to exist, yet several investi- 

 gators have found that by its continued 

 existence in symbiosis with particular host 

 plants it has acquired a certain amount of 

 racial adaptation, so that, for example, 

 clover will flourish best and assimilate the 

 most nitrogen if it be inoculated with the 

 organism from a previous growth of clover, 

 and not from a pea or a bean plant. 



The conclusion naturally follows that 

 it may be necessary to inoculate each 

 leguminous crop with its appropriate or- 

 ganism in order to secure a maximum 

 yield. The first practical efforts in this 

 direction did not, however, meet with much 

 success : the cultivations used for inocula- 

 tion were weak, and, when sown with the 

 seed, in many cases died before infection 

 took place. Even when the formation of 

 nodules followed, yet the assimilation of 

 nitrogen was not great. The question in 

 fact turns upon the degree of 'virulence' 

 possessed by the subcultures used for in- 

 oculation. It is well known with other 

 bacteria how their specific actions may be- 



