18 BOTANICAL GAZETTE [JANUARY | 
did not extend above the gash there was a typical curve between 
the gash and the tip. The gashes were made deep enough to 
cut off the axial cylinder more or less completely, though as this — 
could not be seen at the time of cutting, in many of the roots a | 
small part of the axial strand must have been intact. 4 
TABLE TI. 
ROOTS GASHED THROUGH AXIAL CYLINDER. 
Branded on opposite sides ar gpg cer ne but not 
Date 4 
Distance of gash from tip) Roots used |Roots curved| Roots used |Rootscurved 
By 
1896 
Now: 7 2.5-3 ™™ s 3 
10 2 4 2 
II 2 6 3 
12 8 6 8 I 
* ig 0? 
17 17 7 
23 5 2 
26 3-4 8 3 
30 I 
Dec. 1 ; : 
3-4 7 3 
3 3-4 8 3 
3-4 8 I 
1897 
Mar. 18 - oe: 8 6 
108 49 
Table I shows the different lots in a series of experiments in 
which the gash was made through the axial cylinder and the 
roots were branded on the opposite side of the tip. In the 
second column the blank Spaces mean that no record was made 
of the distance from the tip at which the gashes were made. In 
the fourth column the numbers show how many of the roots curved 
away from the brand both above and below the gash. Many 
more of them showed a traumatropic curve between the gash 
and the tip, and this was especially true of those in which the 
gash was made 3-4™™ from the tip. It may be objected that 
there are not enough control roots, but many of these lots were 
