the action of apogeotropism was quite eliminated.  Nevertheless, they did not rise nearly so much at night, as when subjected to apogeotropism.  Is it not possible, or even probable, that leaves and cotyledons, which have moved upwards in the evening through the action of apogeotropism during countless generations, may inherit a tendency to this movement?  We have seen that the hypocotyls of several Leguminous plants have from a remote period inherited a tendency to arch themselves; and we know that the sleep-movements of leaves are to a certain extent inherited, independently of the alternations of light and darkness.

In our observations on the circumnutation of those cotyledons and leaves which do not sleep at night, we met with hardly any distinct cases of their sinking a little in the evening, and rising again in the morning,—­that is, of movements the reverse of those just discussed.  We have no doubt that such cases occur, inasmuch as the leaves of many plants sleep by sinking vertically downwards.  How to account for the few cases which were observed must be left doubtful.  The young leaves of Cannabis sativa sink at night between 30o and 40o beneath the horizon; and Kraus attributes this to epinasty in conjunction with the absorption of water.  Whenever epinastic growth is vigorous, it might conquer diaheliotropism in the evening, at which time it would be of no importance to the plant to keep its leaves horizontal.  The cotyledons of Anoda Wrightii, of one variety of Gossypium, and of several species of Ipomoea, remain horizontal in the evening whilst they are very young; as they grow a little older they curve a little downwards, and when large and heavy sink so much that they come under our definition of sleep.  In the case of [page 445] the Anoda and of some species of Ipomoea, it was proved that the downward movement did not depend on the weight of the cotyledons; but from the fact of the movement being so much more strongly pronounced after the cotyledons have grown large and heavy, we may suspect that their weight aboriginally played some part in determining that the modification of the circumnutating movement should be in a downward direction.

The so-called Diurnal Sleep of Leaves, Or Paraheliotropism.—­This is another class of movements, dependent on the action of light, which supports to some extent the belief that the movements above described are only indirectly due to its action.  We refer to the movements of leaves and cotyledons which when moderately illuminated are diaheliotropic; but which change their positions and present their edges to the light, when the sun shines brightly on them.  These movements have sometimes been called diurnal sleep, but they differ wholly with respect to the object gained from those properly called nyctitropic; and in some cases the position occupied during the day is the reverse of that during the night.