Snow .— The Conduction of Geo tropic Excitation in Roots. 47 
is difficult to know just how much overlap to allow in order to ensure the 
best connexion of the conducting tissues. In all, 
20 roots were used for this experiment. Of these, 
after twenty-four Sours, 12 remained straight, 2 bent 
well up, 4 slightly up, and 2 slightly down. But 
since such roots, even if kept vertical, tend to bend 
slightly towards the side on which the tip is placed (cf. Experiment 5 above), 
such results are insufficient, even after allowance has been made for the 
slight sensitivity of the decapitated stump. 
Section 3. The Paths of Conduction. 
If excitations leading to tropic responses can be conducted through 
gelatine, this would seem, as pointed out by Paal ( 1918 ), to exclude any 
theory by which they consist in the passing on of any kind of induced 
protoplasmic polarity. Such a theory has been put forward by Fitting ( 1907 ) 
chiefly to explain certain experiments which seemed to show that in the 
cotyledon of Arena , after incisions had been made more than half-way 
through from opposite sides, photo-excitation could still pass down by 
a sinuous path from the tip to the responding region and there bring about 
phototropic response. 
A simpler view of the relation between conduction and response would 
be that curvature in a certain direction is due not to any polarized nature 
of the conducted excitation, but to the fact that unequal intensities of exci¬ 
tation are conducted along the two sides of the organ, and so set up differ¬ 
ences in rate of growth in the responding zone. These unequal intensities 
of excitation would in turn have been due to the unequal effects of the 
stimulus on the two sides of the perceptive region. 
If we adopt this view, it will clearly be necessary to suppose also that 
the excitation is conducted only in straight lines: for otherwise the differ¬ 
ence set up between the two sides of the perceptive region would be 
obliterated before the excitation could reach the responding region. Thus 
any evidence of conduction along a sinuous path would seem to make it 
necessary to reject any explanation along these lines, and to adopt one 
similar to Fitting’s: for no other would seem possible. 
Some, however, of his results with Arena have been doubted by later 
investigators (cf. Paal, 1918 , p. 444 seq.). 
In the case of roots, also, Pollock ( 1900 ) and Fitting ( 1907 , p. 231 seq.) 
have carried out experiments suggesting that traumatic excitation could be 
conducted tortuously around incisions. But apparently nothing was done 
to exclude conduction by diffusion straight across the cuts instead of round 
their margins. The question, therefore, seemed to need re-investigating. 
The case of geotropism was taken first for investigation, as the traumatic 
curvatures obtained appeared to be generally less vigorous. 
