Pe ee Ne ee oe fee eee a ee 
1900] MECHANISM OF ROOT CURVATURE a7 
and v» curving more strongly, but it was very noticeable that 
v pulled much farther away from a than « did, often forming’a 
half-circle or more, and doing this with a sudden jerk as soon as 
it was free to do so. In other cases x lay up close to a, as before 
separation. These two cases aré combined in column four, which 
shows the number, 55.8 per cent., in which 7 pulled farther away 
from @ than « did. In this column x may either have pulled 
away from @ or not. 
TABLE Al, 
MEDIAN SLICE OF CURVED ROOTS SPLIT INTO THREE STRIPS, CONVEX 
CORTEX, CONCAVE CORTEX, AND AXIAL CYLINDER. 
Duis. Haas ae v curves from | v and x curve | + curves from | + and v curve 
ste). | | branding..-|,, Reet Haede | Sane | ae «me ae 
Jan Li 2 I I 
14 2-4 6 4 rt Z I 
cas 3 Z ; ; 
ch Ort 5 4 I 
18 4 5 5 
19 2-4 6 3 3 
19 7 2 2 
20 2-5 2 2 
March 29 4-6 6 2 4 
4 7 7 
43 24 15 ? : 
55.8% 34.8% 4.7% 4.7% 
In the fifth column is shown the number of roots in which 
both x and », after separation from 4a, increased their curve in 
the direction of that of the whole root; 34.8 per cent. of the 
total did this, though only four different lots showed it at all. 
Only a very smal! number showed # pulling away from @ more 
than v, or and v standing at equal distances from a. These 
experiments show that, aside from the difference in tension 
between cortex and axial cylinder the stimulus may also cause 
such a tension within the cortex of either side that it tends to 
curve away from the brand when completely free from other 
tissues. The next table will show that there is no such tendency 
in roots not wounded at one side of the tip. 
