206 
is almost entirely a local contact one, the growth is “ 
localized in the region directly stimulated. 
The removal of the epidermis from the concave surface acts 
as a stimulus to growth, and causes the formation of a corky 
layer over the part injured. If the injury is made along the 
whole concave surface, growth takes place along the whole 
length of the hook. In the following example, the hooks all 
developed leafy branches at the first joint. 
oe pad ra 
After 1 week. 295.975 [830.980]; 110-170 [110.165]; "80.140 [85-185] 
y 6 weeks. 420.465 [385 . 395]; 255.390 [180.300]; 125.235 [105 . 185] 
The breaking strain for the injured stimulated hooks averaged 
8 to 10 kilo’s, for the untouched ones 4 to 6 kilo’s. 
In 4rtabotrys, the development of leafy branches by the 
hooks, does not retard the growth of the latter, but instead, 
tends to hasten and increase it. In Roucheria, the excessive 
development of a permanently attached hook may tend to 
diminish the growth of the distal portion of the branch, from 
which the hook arises, causing the sessile hook to resemble 
the pedunculated branching one of Artabotrys. Tn Uncaria, however, 
in which the hook also is sessile, its attachment and growth 
do not seem in any way except favourably, and indirectly by the 
support afforded, to affect the growth of the distal portion of 
the stem. 
The following measurements, shewing the results of different 
stimuli on the different regions of the hook, were all taken 
after 4 weeks. 
Artabotrys Blumei. _ 
ae _p s—S A-? 
Hook bound round fairly cio with string 248 318 ; 990 3635 108 . 150 
Loose loop of twine 0.25 grms on “3” 973 .332; 150.278; 92. 158 
Tightened loop around “1” 340.445; 195.255; 80.150 
Loose loop 0.4 grm on “3” 190.980; 115.205; 95. 185 
Loop of Cn wire over “2” and with 5 grms attached 342.410; 235.385; 110. 175 
Slight injury between “1” and “2” 995 . 355; 170.295; 90. 145 
Epidermis is removed from concave surface of “2” 960. 345; 250.390; 105. 170 
Untouched hooks 930.290; 130.245; 80-140 
The above data emphasize the fact, that, when the hook is 
subjected to a marked strain, the growth in thickness is distri- 
