120 Sprat t . — The Morphology of the Root Tubercles of 
subsequent treatment with xylol. Microtomed sections were almost exclu- 
sively employed, as the details could not be satisfactorily determined with 
certainty from the freehand sections which have been recommended by 
some writers. A large variety of stains were used, the principal being 
Flemming’s triple, Heidenhain’s iron haematoxylin, Ehrlich’s haematoxylin, 
Loffler’s blue, methyl green and fuchsin, methyl violet and fuchsin, carbol 
fuchsin, and Kiskalt’s amyl gram. The last named proved the most satis- 
factory, since it is a good nuclear stain and also one by means of which 
Pseudomonas radicicola can be differentiated. 
In order to ascertain whether these root tubercles were connected with 
the presence of organisms possessing the power of assimilating atmospheric 
nitrogen, an attempt was made to isolate any such Bacteria from their 
internal tissues. For this purpose a medium of the following composition 
was utilized : 
i grm. saccharose, 
o«5 grm. acid potassium phosphate, 
o-o2 grm. magnesium sulphate, 
ioo c.c. distilled water. 
Into each of three Erlenmeyer flasks, 300 c.c. capacity, 50 c.c. of such 
a medium was placed, and then submitted to a temperature of 140° C. 
and a pressure of two and a half atmospheres for ten minutes in an auto- 
clave. They were then neutralized with sodium hydrate and allowed 
to cool. Some of the tubercles were removed from the roots of both Alnus 
and Elaeagnus , thoroughly washed, and then placed in a sterilizing fluid 
composed of : 
2-5 c.c. concentrated hydrochloric acid, 
1 grm. mercuric chloride, 
500 c.c. distilled water. 
They were allowed to remain in this solution for two minutes, when 
they were removed with sterile forceps and washed in distilled water con- 
tained in a sterile vessel. These nodules were then crushed with previously 
sterilized instruments, and the Alnus nodules were put into one of the above 
flasks, the Elaeagnus into another, and the third was left untouched. All 
were then incubated at a temperature of 25 0 C. 
After two days the solutions into which the nodules had been placed 
were quite cloudy, whilst the other one had remained unchanged. A drop 
of the cloudy liquid was placed on a slide, air-dried, then stained with carbol 
fuchsin, and examined microscopically. In both cases it was a pure culture 
of rod-shaped organisms, many of which contained two or three small round 
densely staining bodies, and were apparently very characteristic Pseudomonas 
radicicola (PI. XIV, Fig. 9). Another slide was prepared, and aniline gentian 
violet employed as a staining reagent, and this was found to be immediately 
