260 BOTANICAL GAZETTE [OCTOBER 
condition of the several experiments. In some experiments the 
weights were light, scarcely exceeding the weight of the plants; in 
other experiments they were equal to many times the weight of 
the plants themselves. Great care was taken to see that the stems 
were not bent by the weights. 
MECHANICAL TESTS 
At the conclusion of each experiment, all the plants were meas- 
ured in the same manner as at the beginning, and the stems taken 
12 : Hic. 1 
to the laboratory for mechanical tests and anatomical study. The 
_ compressed part of each experimental plant, as well as the corre- 
sponding part of each control plant, was tested for rigidity, as 
shown by its resistance to bending, and for resistance to crushing. 
The resistance to bending was determined by a simple piece of 
apparatus, which is represented by the diagram (fig. 1): The 
basal end of the stem to be tested was clamped securely at 4, the ” 
cord was looped over the stem at b, always at a fixed distance 
from a; the weight w was then slowly released and the amount 
