Late in the evening of the 28th, some hours after the above observations were completed, the epicotyl had grown much straighter, for the upper part now formed a widely open angle with the lower part.  A filament was fixed to the upright basal part, higher up than before, close beneath the lowest scale-like process or homologue of a leaf; and its movement was traced [page 33] during 38 h. (Fig. 23).  We here again have plain evidence of continued circumnutation.  Had the bean been properly planted, the part of the epicotyl to which the filament was attached, the

Fig. 23.  Vicia faba:  circumnutation of the same epicotyl as in Fig. 22, a little more advanced in age, traced under similar conditions as before, from 8.40 A.M.  Dec. 28th, to 10.50 A.M. 30th.  Movement of bead here magnified 20 times.

movement of which is here shown, would probably have just risen above the surface of the ground.

Lathyrus nissolia (Leguminosae).—­This plant was selected for observation from being an abnormal form with grass-like leaves.

Fig. 24.  Lathyrus nissolia:  circumnutation of stem of young seedling, traced in darkness on a horizontal glass, from 6.45 A.M.  Nov. 22nd, to 7 A.M. 23rd.  Movement of end of leaf magnified about 12 times, here reduced to one-half of original scale.

The cotyledons are hypogean, and the epicotyl breaks through the ground in an arched form.  The movements of a stem, 1.2 inch in height, consisting of three internodes, the lower one almost wholly subterranean, and the upper one bearing a short, [page 34] narrow leaf, is shown during 24 h., in Fig. 24.  No glass filament was employed, but a mark was placed beneath the apex of the leaf.  The actual length of the longer of the two ellipses described by the stem was about .14 of an inch.  On the previous day the chief line of movement was nearly at right angles to that shown in the present figure, and it was more simple.

Cassia tora* (Leguminosae).—­A seedling was placed before a

Fig. 25.  Cassia tora:  conjoint circumnutation of cotyledons and hypocotyl, traced on vertical glass, from 7.10 A.M.  Sept. 25th to 7.30 A.M. 26th.  Figure here given reduced to one-half of original scale.

* Seeds of this plant, which grew near the sea-side, were sent to us by Fritz Müller from S. Brazil.  The seedlings did not flourish or flower well with us; they were sent to Kew, and were pronounced not to be distinguishable from C. tora. [page 35]

north-east window; it bent very little towards it, as the hypocotyl which was left free was rather old, and therefore not highly heliotropic.  A filament had been fixed to the midrib of one of the cotyledons, and the movement of the whole seedling was traced during two days.  The circumnutation of the hypocotyl is quite insignificant compared with that of the cotyledons.  These rise up vertically at night and come into close contact; so that they may be said to sleep.  This seedling was so old that a very small true leaf had been developed, which at night was completely hidden by the closed cotyledons.  On Sept. 24th, between 8 A.M. and 5 P.M., the cotyledons moved five times up and five times down; they therefore described five irregular ellipses in the course of the 9 h.  The great nocturnal rise commenced about 4.30 P.M.