22 
species varying very slightly would require difficult bacili. One 
investigator who was working at this particular subject, has indeed 
succeeded in showing that in many cases a particular leguminous 
plant thrives much better when it is supplied with its particular 
bacilli than when it is supplied with the bacilli of another species. 
Thus three robinia plants were grown simultaneously; No. 1 was 
not allowed any bacilli at all, No. 2 was allowed the bacilli from 
pea-tubercules, and No. 8 the bacilli from robinia, and as a conse- 
quence Nos. 1 and 2in 34 months were mere bare stalks. No. 3 
was a fine well grown plant. Three pea-plants were treated in the 
same way; No. | was supplied with the bacilli from pex-tubercules, 
No. 2 with bacilli from lupin, and No. 3 with hacilli from robinia. 
In about eight weeks No. 1 was a thriving plant, No. 3 a poor 
withered up specimen, and No. 2 rather intermediate. But I don’t 
think this proves that even these three are distinct species of bacilli. 
Klein has worked at the nodules in the lupin and has found that the 
bacilli are of two kinds, one about twice as numerous as any other: 
He says ‘‘ the examination of the nodules was carried out towards 
the end of last year; the nodules were very abundant on the roots 
obtained ; some were of a round shape, others oval, some were flat, 
others more of spheroidal shape ; all were more or less eccentrically 
attached to the roots, in fact, as-is well known, were local eccentric 
thickenings of the roots themselves. A transverse section through 
a root nodule, shews underneath the brown covering a yellowish 
brownish mass representing the main cellular tissue of the nodular 
excrescence, and gradually passing into the cellular and vascular 
tissues which form the central or axial portion of the root itself.’’ ° 
‘* Examined under a microscope; the yellowish substance is seen to 
contain large numbers of cylindrical bacilli with rounded ends, and 
amongst these some which are short oval, and others which are 
dumb-bells of spherical or short oval corpuscles, but the cylindrical 
bacilli are barely predominating in numbers” The bacilli are 
small and rather difficult to see. I saw one or two preparations of 
them in the Bacteriological Department at the meeting of the 
British Medical Association held at Bristol in the beginning of last 
August. They seemed rather smaller than the bacillus of ‘‘ tuber- 
culosis”’ and very difficult to observe in a section of the nodule. 
They are also very slow in staining. The course of proceedings in 
the formation of the nodule is as follows :—A bacillus at some stage 
of its life penetrates the point of one of the minute hairs of the root 
of the plant, and spreads by going through the whole length of the 
hair into the root itself; its presence there causes the exuberant 
growth of cells which constitute the nodule, whilst colonies of the 
bacillis grow in its deeper parts, surrounded by some layers of these 
cells. The bacilli apparently benefit by the jucies of the plant, 
whilst the plant benefits by the nitrogen, which in some way, it is 
