along the lower side of a part, which is thus caused to bend downwards; and hyponasty is used for the reversed process, by which the part is made to bend upwards.  These actions come into play so frequently that the use of the above two terms is highly convenient.  The movements thus induced result from a modified form of circumnutation; for, as we shall immediately see, an organ under the influence of epinasty does not generally move in a straight line downwards, or under that of hyponasty upwards, but oscillates up and down with some lateral movement:  it moves, however, in a preponderant manner in one direction.  This shows that there is some growth on all sides of the part, but more on the upper side in the case of epinasty, and more on the lower side in that of hyponasty, than on the other sides.  At the same time there may be in addition, as De Vries insists, increased growth on one side due to geotropism, and on another side due to heliotropism; and thus the effects of epinasty or of hyponasty may be either increased or lessened.

He who likes, may speak of ordinary circumnutation as being combined with epinasty, hyponasty, the effects of gravitation, light, etc.; but it seems to us, from reasons hereafter to be given, to be more correct to say that circumnutation is modified by these several agencies.  We will therefore speak of circumnutation, which is always in progress, as modified by epinasty, hyponasty, geotropism, or other agencies, whether internal or external.

[One of the commonest and simplest cases of epinasty is that offered by leaves, which at an early age are crowded together round the buds, and diverge as they grow older.  Sachs first remarked that this was due to increased growth along the upper side of the petiole and blade; and De Vries has now shown in more detail that the movement is thus caused, aided slightly by [page 269] the weight of the leaf, and resisted as he believes by apogeotropism, at least after the leaf has somewhat diverged.  In our observations on the circumnutation of leaves, some were selected which were rather too young, so that they continued to diverge or sink downwards whilst their movements were being traced.  This may be seen in the diagrams (Figs. 98 and 112, pp. 232 and 249) representing the circumnutation of the young leaves of Acanthus mollis and Pelargonium zonale.  Similar cases were observed with Drosera.  The movements of a young leaf, only 3/4 inch in length, of Petunia violacea were traced during four days, and offers a better instance (Fig. 111, p. 248) as it diverged during the whole of this time in a curiously zigzag line with some of the angles sharply acute, and during the latter days plainly circumnutated.  Some young leaves of about the same age on a plant of this Petunia, which had been laid horizontally, and on another plant which was left upright, both being kept in complete darkness, diverged in the same manner for 48 h., and apparently were not affected by apogeotropism; though their stems were in a state of high tension, for when freed from the sticks to which they had been tied, they instantly curled upwards.