135 
of the root-tip in relation to geotropism. 
fixed into a tube or other support, and since the cotyledons are of 
relatively great weight the difficulty appears insuperable without 
some special device. This has been supplied by Mr H. Darwin 
who devised a “ Root-Lever ” which has been made for me by the 
Scientific Instrument Company. It will be described in a fuller 
paper ; it will be sufficient for the reader to know that while 
the weight of the cotyledons is supported, they are free to move 
in any direction in obedience to the curvature of the root. The 
cotyledons being supported by the root-lever, the tip of the root is 
inserted into a fixed horizontal tube and the root is kept from 
withering by water slowly and gently dripping on it. The great 
difficulty is to keep the tip of the root from slipping out of the 
fixed tube. A great majority of the experiments failed from this 
cause, but this does not invalidate the theory, although it points 
to the need of improving the method. Omitting from consider- 
ation these cases and a few in which the roots did not curve 
geo tropically, my results point clearly to a strong tendency in the 
root to continue curving. In several cases the cotyledons travel 
through 180° and in one case through about 360°, so that the root 
looked as though a knot had been tied in it. 
In the case of peas the difficulty of keeping the root in place is 
less, the failures are fewer, and the tendency to continued curva- 
ture in the root is strongly marked. These experiments are still 
in progress and will be described in detail in a fuller publication. 
There too will be fully discussed the interpretation of the results 
which presents some difficulties depending on the distribution of 
growth in roots (as compared with stems) and on the existence of 
other forms of irritability in the root-tip. One consideration must 
not however be omitted. The tip of the root is sensitive to 
unequal pressure or unilateral injury, and it may be supposed that 
the continued curvature of the root is due to pressure exercised on 
the tip of the root by the fixed tube. But the results of a number 
of experiments in which a root was placed tip downwards in a 
vertical tube do not support this view. There seems no reason 
why a vertical tube should not be equally effective with a hori- 
zontal one in producing contact stimulation. But there was no 
tendency to continuous curvature when the tube was vertical. 
