ASPECTS OF AN OLD AGRICULTURAL QUESTION. 265 



Frank (25) has long distinguished himself as the chief 

 upholder of the hypothesis that the fixation occurs in the 

 leaves and other sub-aerial tissues of the plant itself ; but al- 

 though the admirably ingenious experiments of Kossowitsch 

 (26) did not settle the question whether the fixation occurs 

 in the subterranean parts, or in the superterranean foliar 

 region, they did render the latter view the more improbable 

 one. And many considerations prevent our accepting 

 Frank's hypothesis that the symbiotic organism so stimulates 

 the entire organisation of the infected plant that its assimi- 

 lating cells, in the organs exposed to light, force the nitrogen 

 straightway into combination. 



The structure of the nodules, as shown by the researches 

 of others as well as by my own analyses, certainly lends 

 support to the otherwise probable view that whether the 

 combination is effected by the bacteroids themselves, or by 

 the cells infected by them, the actual seat of the action is in 

 the nodules. Large quantities of carbo-hydrates are carried 

 to these cells, a complex and very regular system of vascular 

 bundles connects them with the mother-root, the infected 

 cells show all the well-known signs of intense physiological 

 activity, and it is known that the plant eventually drains 

 the nodules of materials, and leaves them exhausted. That 

 carbo-hydrates, water, and nitrogen come abundantly into 

 contact in these cells, there is no doubt, though we have 

 no measure of the quantities of materials passing to and 

 from the nodules. 



On the other hand, all attempts to demonstrate that the 

 bacteroids cultivated outside the plant can assimilate free 

 nitrogen have failed, unless we except the very doubtful 

 results of Beyerinck (27). It is true, we must not lay too 

 much stress on these negative results, because the cultiva- 

 tion of an organism so highly adapted as this must be to its 

 symbiotic life may well fail outside the cells of the host. 



Here, then, for the present the matter stands. The 

 infected leguminosse do bring about the fixation of free 

 nitrogen, but other plants do not ; the fixation occurs in the 

 underground parts of the plant, and not by physical or 

 other processes in the soil, and the seat of fixation is the 



