- 
TRANSACTIONS OF SECTION K. 779 
the usual idea, would mean that the terminal walls are not sensitive. But the 
walls must be sensitive in some way, or the plant would not react to the gravi- 
stimulus, as it undoubtedly does. The only conclusion I can come to is that the 
position of the statoliths shown in fig. 3, in which they rest partly on the terminal 
wall and partly on the lateral (tangential) wall, must be capable of giving the 
combined stimulus,' as above suggested. 
Personally I do not attach great importance to the details of how the statoliths 
act on the different walls of the cells, although as part of the history of the inquiry 
I feel bound to discuss it. The broad fact that the statoliths rest on different parts 
of the cell-walls when the geotropic organ is placed at different angles with the 
vertical seems to me sufficient. The precise manner in which various reactions 
are associated with the position of the statoliths may be confessed to be for the 
present beyond our knowledge or powers of imagination, and such confession need 
not weaken the position of our theory. 
Finally, I desire to say a word on a subject having but a remote connection 
with my theme. There is at the present time a tendency to pay an increasing 
attention to what is known as rectipetality or autotropism—viz., the inherent 
capacity of rectilinear growth. In my Cardiff Address* to Section D I showed 
that rectipetality is really part of the phenomena of circumnutation. We must 
believe that rectipetality does not merely come into play in those comparatively 
crude experimental instances in which a geotropic curvature is flattened out by 
means of growth on the klinostat. We must believe that it also corrects curvatures 
which arise from the slight irregularity of normal every-day growth. This will imply 
that normal growth is built of a series of internal corrections; in other words, of 
circumnutation. The point I wish now to emphasise is that the stimuli, be they of 
geotropic or any other nature, should be conceived as acting not on a stationary 
but on a moving plant—acting, in fact, on the spontaneous correcting power, 
whether we call it rectipetality, autotropism, or circumnutation. It is impossible 
to say how this consideration might modify our speculations as to the manner of 
action of the gravistimulus. It is quite conceivable that it might not alter our 
theoretic views at all, but without more knowledge we cannot be certain. My 
only point at present is that if we are led into contradictions or confusion by 
attempts to analyse what goes on in the gravisensitive region according to the 
statolith theory, such a result must not be held to be fatal to the theory until we 
know more of the problem. 
In conclusion—and to clear our minds of the doubtful speculations in which 
I have entangled myself—I should like to reiterate my belief in the general, 
though not the universal, applicability of the statolith theory. I find it impossible 
to doubt that, in the case of the higher plants, sensitiveness to the pressure of 
heavy bodies will be found to be by far the most important, if not the exclusive, 
means by which gravity is perceived. We have seen that the stimulus must 
depend on weight; and since neither the theory of radial pressure nor Noll’s 
supposition of stimulation by small unknown bodies lends itself to experimental 
n uiry we are driven, as practical people, to test the views of Haberlandt and 
mec. 
I base my belief partly on what I have aiready said, namely, that geotropism, 
being an adaptive reflex action, must during its development have been correlated, 
by that mysterious bond which unites stimulus to reaction, with some change, by 
which in the natural course of events it is uniformly preceded, Now the most 
obvious change which precedes geotropism is the disturbance of the falling starch- 
grains, ‘This fact, together with what we know of the distribution of statoplasts, 
would almost force conviction on me. But this is not the whole of the evidence. 
We know from Némec’s researches that the protoplasm, in. the cells assumed 
1 In Noll’s diagram of the stimulation-areas ina diageotropic organ the obliquely 
placed areas seem to suggest a similarity to what is here given [see Noll (92, p. 29)]. 
But his stimulation areas in which only a single statolith occurs are not strictly 
comparable to cells containing numerous statoplasts., 
3 F, Darwin (91), 
