1921] 
MATSUMOTO— SPECIALIZATION IN RHIZOCTONIA 40 
water. After a treatment of 4 hours the bag was transferred 
for a few minutes to a jar containing sterile water and then 
again rinsed in a second jar. The contents of each bag were 
carefully emptied into a sterile Petri dish. "The seeds were then 
transferred to Petri dishes containing potato decoction agar. 
All these treatments were made in a transfer room in which all 
dust was thoroughly precipitated by steam. 
The dishes were then incubated at about 25? C. After germi- 
nation took place each one of the seeds free from contamination 
was transferred to a large sterile test-tube (1 inch in diameter, 
. 8 inches in height) containing damp filter-paper at the bottom. 
The basal part of the tube was then covered with black paper 
to prevent exposure to sunlight. All the cultures were then 
placed in a greenhouse, and after a few days, when the plants 
had grown to 2 inches, inoculation with the different strains 
was made. 
The first appearance of the disease was generally observed 
after a few days, showing a remarkable discoloration in the part 
attacked. The first sign of the disease in the plants inoculated 
with P1 and H usually appeared at least 1 or 2 days earlier than 
that in the plants inoculated with P4 or Bl. In any case, no 
disease was produced by B3. The strain P7 is of low virulence, 
and if it may attack the plants the first symptom appears later, 
usually after 2 or 3 weeks. In general, the fungi attacked the 
plants both on stems and roots, but a direct attack of leaves was 
rather infrequent. 
From the microscopical investigations it appears that the 
hyphae may enter the plant directly through the surface and in 
no case through natural openings (fig. 5a and b). Of course, it 
should be remembered that the stomatal condition is greatly 
modified by environment, consequently the evidence obtained 
from the saturated conditions of the cultures cannot be readily 
applied to natural conditions. 
Attention would then naturally turn to the question whether 
the penetration of the epidermal cell wall is effected in a purely 
mechanical way or by some special secretion of the fungous 
hyphae. To solve this question a few drops of mycelial extrac- 
tion of P1 were put on the upper surface of a pea leaf which was 
placed in a sterilized Petri dish. For comparison another leaf 
