39 
The Localisation of Perception. 
further investigation. It is clear for instance that if the cotyledon 
(by which the seedling is supported) is vertical, and if the tip of the 
cotyledon is directed upwards, there is no reason why the hypocotyl 
should curve since the cotyledon is in the position of equilibrium— 
but there was occasional curvature under these conditions. It is 
possible that this was due to the cotyledons not being strictly 
vertical, for it is conceivable that a deviation of 0-5° acting as a 
•continuous stimulus might produce an effect. It is noteworthy that 
when special care was taken to make the cotyledons vertical, the 
curvature of the hypocotyls was small. Thus in one experiment 
seven Sorghum seedlings only gave an average deviation from the 
vertical of about 30° in two days, but when the same seedlings were 
placed horizontal, they curved through 180 n in one day. 
Some cases similar to the above have been observed by 
Massart. 1 Thus he found that a shoot of Lysiinachia, supported by 
the tip, curves past the vertical in away which suggests localisation 
of gravi-perception in the apical region. 
Some interesting observations were made by Kohl 2 and Miehe 3 
on the apo-geotropic movements of Tradescantia which take place 
at the nodes by means of pulvini, as occurs in the haulms of grasses. 
Kohl found that a joint does not bend geotropically if the joint 
next above it is removed. According to Miehe this is not the whole 
truth. Geotropism really depends on the bud in the upper joint. 
To prevent curvature in the joint below it is not necessary to 
remove the bud, the same effect is produced by maltreating it, e.g., 
by chilling or by checking its growth by means of gypsum. How 
far the geotropism of the lower joint is regulated by the position of 
the upper one is not clear. It is probably safer at present to 
confine ourselves to the statement that the geotropism of a joint 
•depends on the bud in the joint above functioning normally. 
The method used in my experiments on Setaria, &c. has also 
been employed with roots. After fruitless efforts to induce a seed¬ 
ling pea to support itself by the apex of the root, a plan suggested 
by the Cambridge Scientific Instrument Company was adopted 4 in 
which the weight of the cotyledons was supported by an arrange- 
1 Mem. Acad. Belg., LXII., 1902. 
2 Bot. Zeitung, 1900. 
3 Pringsheim’s Jahrb., 1902. 
4 F. Darwin in Linn. Soc. Journal, XXXV. My results are con¬ 
firmed by Massart (Mem.-Couronnes R. Acad. Belg., LXII., 
1902) who succeeded in supporting seedlings of Ipomcea tricolor 
by the root-tip and obtained curvatures far past the vertical. 
See fig. 10, p. 22, fig. 6, p. 31. 
