366 THE AMERICAN NATURALIST. (VoL. XXXII. 
(6) The author washed out the contents of tubercles in water, 
injected this into growing stems or roots of seedlings, and after a 
lapse of from four to six weeks washed their surface thoroughly with 
sterile water, bruised them in a sterile porcelain mortar, mixed thor- 
oughly with sterile earth, planted therein the sterilized seeds, and 
obtained root tubercles in three weeks on the roots of almost every 
plant. This would indicate that, if the germs are normally present 
in the stems and other non-tuberculous parts of leguminous plants, 
the bruised tissues added to sterile earth ought to infect the roots of 
seedlings grown therein. Consequently, roots, stems, leaves, and 
leafstalks of Phaseolus vulgaris, P. multiflorus, Vicia sativa, V. faba, 
Pisum sativum, Ervum lens, and Lupinus albus were tested in this 
way after cutting away the superficial portions with sterile knives, 
but in no case did any root tubercles appear. 
(c) After isolating the root tubercle organism from four plants 
(Phaseolus multiflorus, Pisum sativum, Lupinus albus, and Vicia faba) 
and determining that the cultures were able to produce root tuber- 
cles on the specified plants, and would remain alive for a long time 
when injected into parts above ground, attempts were made to isolate 
these organisms from other parts of the same plants, using four dif- 
ferent media, v7z.: (1) sterilized hydrant water; (2) water, meat 
extract, peptone, and sugar in the following proportions, — 100, 0.5, 
0.5, 3-0; (3) decoction of the plant mixed with washings of earth; 
(4) decoction of the plant, asparagin, and sugar in the following pro- 
portions, — 100, 0.25, 0.5. These nutrient solutions were sterilized 
in- small Erlenmeyer flasks. The inoculations were made in a room 
rendered as free as possible of floating germs by the introduction of 
vapor of water. The selected, above-ground tissues were washed. 
for a long time and very carefully in sterile water, cut into small 
pieces with flamed shears, and put carefully into the flasks, part of 
which were exposed to the air and the rest subjected to an atmos- 
phere from which the oxygen was removed as completely as possi- 
ble. Bacterial growths appeared in only a few of the flasks, and 
none of these produced any root tubercles when added to the sterile 
earth in which the seeds were grown. This experiment was repeated 
on roots free from tubercles with the same negative result. 
(d) When sterilized seeds had sufficiently germinated, the roots 
were put through a tiny opening in the bottom of a glass pot, and then 
the bottom of the pot was filled with a mixture of sterilized gypsum 
and water, so that the middle part of the roots was cemented fast. The 
part of the root above the partition of gypsum was then covered 
