48 
T. TADOKORO. 
From the nature of the methods followed, it should be remembered here 
that pentosan and gummy substance are identical or at least very much 
resemble each other in nature and that the pectin compounds, at least in 
part, must also be the compounds of pentosan. Hence in discussing the 
results, we take the pentosan only into account, neglecting the other entirely. 
From the results obtained we notice the following : 
1. Pentosan suffers most in retting process. 
2. A small quantity of cellulose is also lost, after the middle period 
of retting. 
3. Pectin is not completely lost. This is apparently iu contradiction to 
the results of Omelianski, but our experience tends to show that 
Omelianski used too small a quantity of the sample, making his 
results rather doubtful. 
4. Glucose and tannin are completely lost in the retting. 
Pentosan-free fiber consists of cutin, cellulose and lignin. Attempt was 
made to determine the three components separately by König’s method, but 
unfortunately it was found impossible to oxdize lignin with hydrogen 
peroxide, hence the attempt was given up. Concerning the nature of the 
changes in pentosan -free fiber we must therefore depend upon the results of 
microchemical observations. 
To know whether the öhanges above mentioned take place in the bark 
portion or woody portion of the stem, the following experiments were per- 
formed. The stem was carefully divided from the cambium layer into two 
parts, i. e., bark and woody portions and weighed. The results are as follows : 
Table 
Bark portion. 
Woody portion. 
Stem taken. 
gm. 
% 
gm. 
% 
Control stem 
16.3876 3.6663 
35.5 
6.6613 
64.5 
Retted stem 
13.2562 3.7647 
28.4 
9.4915 
71.6 
In the retted stem the proportion of the bark portion is far less than in 
