Hypocotyl.—­A seed lying on damp sand was firmly fixed by two crossed wires and by its own growing radicle.  The cotyledons were still enclosed within the seed-coats; and the short hypocotyl, between the summit of the radicle and the cotyledons, was as yet only slightly arched.  A filament (.85 of inch in length) was attached at an angle of 35o above the horizon to the side of the arch adjoining the cotyledons.  This part would ultimately form the upper end of the hypocotyl, after it had grown straight and vertical.  Had the seed been properly planted, the hypocotyl at this stage of growth would have been deeply buried beneath the surface.  The course followed by the bead of the filament is shown in Fig. 28.  The chief lines of movement from left to right in the figure were parallel to the plane of the two united cotyledons and of the flattened seed; and this movement would aid in dragging them out of the seed-coats, which are held down by a special structure hereafter to be described.  The movement at right angles to the above lines was due to the arched hypocotyl becoming more arched as it increased in height.  The foregoing observations apply to the leg of the arch next to the cotyledons, but [page 41] the other leg adjoining the radicle likewise circumnutated at an equally early age.

The movement of the same hypocotyl after it had become straight and vertical, but with the cotyledons only partially expanded, is shown in Fig. 29.  The course pursued during 12 h. apparently represents four and a half ellipses or ovals, with the longer axis of the first at nearly right angles to that of the others.  The longer axes of all were oblique to a line joining the opposite cotyledons.  The actual extreme distance from side to side over which the summit of the tall hypocotyl passed in the course of 12 h. was .28 of an inch.  The original figure was traced on a large scale, and from the obliquity of the line of view the outer parts of the diagram are much exaggerated.

Cotyledons.—­On two occasions the movements of the cotyledons were traced on a vertical glass, and as the ascending and descending lines did not quite coincide, very narrow ellipses were formed; they therefore circumnutated.  Whilst young they rise vertically up at night, but their tips always remain reflexed; on the following morning they sink down again.  With a seedling kept in complete darkness they moved in the same manner, for they sank from 8.45 A.M. to 4.30 P.M.; they then began to rise and remained close together until 10 P.M., when they were last observed.  At 7 A.M. on the following morning they were as much expanded as at any hour on the previous day.  The cotyledons of another young seedling, exposed to the light, were fully open for the first time on a certain day, but were found completely closed at 7 A.M. on the following morning.  They soon began to expand again, and continued doing so till about 5 P.M.; they then began to rise, and by 10.30 P.M. stood vertically and were almost closed.  At 7