On a method of studying Geotwpism . 
67 
the stalks, but by using a pair of crossed pins at each point 
which it was desired to confine. In this way every stalk 
was attached at both ends and in the middle. The boards 
were then placed in a tin box containing damp sand. One 
board was placed horizontally, and the others at angles of 
6o°, — in one case with the apical end of the stalk upwards, 
in the other downwards. The three sets may be distinguished 
as Above , Below , and Horizontal. The box was placed in 
a damp chamber at a constant temperature of 25 0 C, for 
two hours. The stalks were then released and placed in 
water for an hour, during which time the curvatures materially 
increased 1 . The form assumed by each was recorded by 
tracing 2 the curvature on paper. The amount of curvature 
was measured by taking, from the tracings, the angle between 
the older and younger parts of the stalks. This could be 
done by drawing tangents to the curves, and was found 
more satisfactory than estimating the radius of curvature in 
each case. 
The following table gives the results of experiments, made 
in June 1887, on 148 plantain stalks. The whole series of 
angles is given, in order that the great amount of inequality 
in the results may be seen. 
Above: 39”, 27°, 3 **°. 33 °, 37 °, 32°. 3 °°, 35 °, 57 °, 4 i°, 28°, 32°, 
43 °) 46°, 35 °. 30 °. 27°, 44 °> 3 °°, 35 °. 62°, 90°, 34°, 77 °, 
39°. 82°, 9°°> 39°. 8i°, 30°, 47 0 , 30°, 33 0 , 31 0 , 6o°, 29°, 
45 } 54} 63°, 41°, 40°, 28°, 52 0 , 56 °, 33 °. i 5 °> 4^°, 33 °. 
56°, 15 0 . Average of 50 angles=42-8. 
Below : 29°, 48°, 26°, 6o°, 53°, 0°, 39°, 37°, 4 4°. 34 °, 24°, 58°, 69°, 
43°, 3i°, 44 °, x8°. 41°, 44°, 26°, 33°, 5 c°, 83°, 34°, 42°, 48°, 
53 °. 59 °, 24°, 59 °, 56 °, 47 °, 4 °°, 3 «°, 84°, 45 °, 55 °, 4 °°, 26°, 
52°, 28°, 21°, 31°, 8o°, 32°, 45°, 31°. 
Average of 46 s angles = 43-1. 
1 The stalks were placed on their sides in a flat-bottomed vessel, so that the 
increase of curvature was due to after-effect, not to a continuance of geotropic 
stimulation. 
2 The form was traced with a fine paint-brush, by which means a more trust- 
worthy representation of the curvature can be made than with a pencil. 
3 Omitting one which remained straight. 
F % 
