During this same time the basal part of the hook (and subsequently of the loop) curved itself slowly upwards; and this must have been wholly due to apogeotropism in opposition to hyponasty.  The loop was then fastened upside down, so that its basal half would be simultaneously acted on by hyponasty (if present) and by apogeotropism; and now it curved itself so greatly upwards in the course of only 4 h. that there could hardly be a doubt that both forces were acting [page 274] together.  At the same time the loop became open and was thus reconverted into a hook, and this apparently was effected by the geotropic movement of the apex in opposition to epinasty.  In the case of Ampelopsis hederacea, weight plays, as far as we could judge, a more important part in the hooking of the tip.

In order to ascertain whether the shoots of A. tricuspidata in straightening themselves under the combined action of hyponasty and apogeotropism moved in a simple straight course, or whether they circumnutated, glass filaments were fixed to the crowns of four hooked tips standing in their natural position; and the movements of the filaments were traced on a vertical glass.  All four tracings resembled each other in a general manner; but we will give only one (see Fig. 122, p. 273).  The filament rose at first, which shows that the hook was straightening itself; it then zigzagged, moving a little to the left between 9.25 A.M. and 9 P.M.  From this latter hour on the 13th to 10.50 A.M. on the following morning (14th) the hook continued to straighten itself, and then zigzagged a short distance to the right.  But from 1 P.M. to 10.40 P.M. on the 14th the movement

Fig. 123.  Smithia Pfundii:  hyponastic movement of the curved summit of a stem, whilst straightening itself, traced from 9 A.M.  July 10th to 3 P.M. 13th.  Apex 9 ½ inches from the vertical glass.  Diagram reduced to one-fifth of original scale.  Plant illuminated through skylight; temp. 17 1/2o — 19o C. [page 275]

was reversed and the shoot became more hooked.  During the night, after 10.40 P.M. to 8.15 A.M. on the 15th, the hook again opened or straightened itself.  By this time the glass filament had become so highly inclined that its movements could no longer be traced with accuracy; and by 1.30 P.M. on this same day, the crown of the former arch or hook had become perfectly straight and vertical.  There can therefore be no doubt that the straightening of the hooked shoot of this plant is effected by the circumnutation of the arched portion—­that is, by growth alternating between the upper and lower surface, but preponderant on the lower surface, with some little lateral movement.