40 
Francis Darwin. 
ment of levers allowing movement in all directions. The experi¬ 
ment was started by fixing the tip of the root in a horizontal tube,, 
and it was found that in successful experiments geotropism expressed 
itself in an obvious tendency to continuous curvature. There are 
however a number of difficulties connected with the method. The 
root-tip is constantly slipping out of the tube so that a large pro¬ 
portion of the experiments failed for this reason. Again the root 
has to be kept damp by water dripping on it, and these causes 
seriously depressed the rate of curvature. Lastly since roots are 
easily affected by injury to their tips, it is conceivable that the 
curvature is traumatic in origin. But I have given what still seem 
to me good reasons against such a supposition. 1 In the same place 
(ibid, p. 272) I have put forward, in regard to roots, a difficulty 
identical in principle with Miehe’s criticism of my experiments 
on Sorghum and Setnria; it will therefore be discussed in relation 
to those plants. Let us assume for the moment that sensitiveness- 
to gravitation is not confined to the cotyledon of a Sorghum seed¬ 
ling, but that it resides in the hypocotyl. If a Sorghum is 
supported by pushing the cotyledon into a horizontal tube, the 
hypocotyl being by hypothesis apogeotropic will curve upwards 
as shown in the figure. The whole hypocotyl, however, is not 
vertical, for the part between C x and B (the base of the cotyledon) 
is oblique, and will therefore be gravitationally stimulated. If BC X 
curves upwards into position BC 2 ,the region above C 2 will be carried 
beyond the vertical. It might be expected that this would be 
corrected by a reversal of the curvature at C». But the region of 
greatest growth in these grass-seedlings is close to the base of the 
cotyledon, and it is conceivable that by the time the plant assumes 
1 F. Darwin, loc. cit., p. 274. 
