136 BACTERIA IN RELATION TO PLANT DISEASES. 
In 1907, Gino de Rossi contributed a paper- on ‘‘The Micro-organism Which Causes the Root- 
nodules of the Leguminosae,” taking a view quite different from the ordinary one. 
_ Heused a large number of plants of Vicia faba (fully 60). From each plant, except the youngest 
which had no nodules, 12 or 15 nodules were removed from different parts of the roots, and repeatedly 
washed in tap-water. 
They were then placed in a stoppered vessel of sterilized distilled water. The water was changed 
twenty times in an hour, and each time the vessel was shaken vigorously. Finally, without other 
attempt at surface sterilization the nodules were placed by means of sterile forceps in a Petri-dish 
containing.a few pieces of filter paper, the whole having been previously sterilized in the dry oven. 
All liquids, instruments, and vessels used in the further handling of the nodules were very 
carefully sterilized by heat. 
After every trace of water was removed by the filter paper, the nodules were removed to another 
Petri-dish where they were sectioned, and a small portion of the interior removed with a sharp needle, 
care being used not to touch the surface layers. This fragment was crushed out in 2 to 3 cc. of water 
in a stoppered vessel. ‘Three or four such vessels were prepared from each plant. Microscopic 
preparations and cultures were made from this material. 
The following solid culture-media were used: 
(a) Simple ro per cent gelatin, with beef-extract (1 per cent), pepton (1 per cent) and sodium chloride (0.5 per 
cent), the reaction of which was mildly alkaline or mildly acid (natural acidity). 
(6) Ten per cent gelatin prepared with Vicia faba extract (1 part leaves cooked in 8 to 10 parts of tap-water and 
filtered), 1 to 2 per cent cane-sugar or glucose, 1 per cent pepton and 0.5 per cent sodium chloride, with a 
mildly acid or alkaline reaction. 
(c) Simple agar (1.5 per cent) or agar with Vicia faba extract. Poured plates were made. In some experiments 
all these media were used, in some only one. In every case, however, the somewhat acid gelatin containing 
Vicia faba extract was used. 
De Rossi examined the contents of the nodules in hanging drops and in stained preparations, 
He found the contents of very young nodules to consist of non-motile rodlets, constant in breadth 
but not in length, and easily stained with aniline dyes. He did not find motile bacilli, either large 
or small. 
In older stages of the nodules a series of transitions was observed, the rodlets showing first one 
end slightly swollen, then a slight dichotomy at the end, and finally the typical X and Y shapes, 
which, contrary to current statements, were constant in form and dimensions. 
In this phase the central part of the nodule contains an enormous number of the branched forms; 
only in a few cases were straight rods seen. De Rossi considers these bacteroids as a stage in the 
development of the organism, not as degenerate forms. 
He observed the development of what appeared to be vacuoles in the bacteroids, and states that 
this is a constant feature, belonging to a phase in the development of the organism. The bacteroids 
at this stage are somewhat swollen. 
In no case did he see any very small infecting bacilli or the development of such bacilli into 
bacteroids. 
Cultures, made on the solid media by flooding the surface with a dilution from nodules full of 
bacteroids, developed a small number of tiny, whitish colonies. ‘These appeared in 2 or 3 days, at 
a temperature of 16° C. As development on the plates proceeded a mixture of very different colonies 
was often apparent. 
By selecting plates with one dominant form, pure cultures of four sorts of Schizomycetes were 
made. ‘This variety of forms together with the presence of numerous non-germinating bacteroids in 
the surface of the gelatin (as determined by a microscopic examination) suggested to de Rossi that 
the bacteroids were unable to grow on artificial media, and that the colonies present belonged to 
other organisms which had penetrated the nodule and multiplied there sparingly [they may have 
come from the surface], but which had nothing to do with its production. He further supposes that 
these non-infectious intruders have been commonly mistaken for the root-nodule organism. One 
white form was non-liquefying and motile by means of a polar flagellum. 
To prove these hypotheses he isolated the bacteroids from the surface of such cultures in the 
following way: 
After 12 to 15 days in the thermostat (probably at 15° to 20° C.), the surface of plates, which had 
yielded a few colonies only, was moistened, and scrapings from the apparently sterile parts between 
colonies were used for inoculating various media. The actual presence of numerous bacteroids on 
this surface was demonstrated by microscopic examination: The results were all negative. The media 
used were gelatin with mineral salts or Vicia faba extract, beet roots, and raw and cooked potato. 
Cultures were also attempted in pure nitrogen without positive result. 
Inoculations were then made on plants of Vicia faba using: (1) Scrapings containing the bac- 
teroids which would not grow on his media; (2) the fourth or fifth sub-cultures of the colonies which 
