tle this question; among them the following: A forked branch of lilac (Fig. 1) was so disposed that the one branch was immersed in water, while the other was exposed to the ordinary atmospheric conditions. The superficies of foliage was the same on both branches. The transpiration from the surface of the leaves on the latter branch was the same as under normal circumstances, and after the lapse of two weeks the foliage was as fresh as at the commencement, showing that the submerged leaves were fully able to replace the roots in one of their functions. In an experiment with a beet in which one-half of the leaves were in water and one-half in the air, communication being maintained by means of the root, the free portion of the leaves wilted in the course of a day, the neck of the root apparently not offering a sufficient means of communication with the submerged leaves A grapevine shoot half plunged in water (Fig. 2) maintained a normal evaporation in the free foliage, and remained fresh for over a month. An oleander shoot under similar conditions maintained its normal appearance for four months. With the artichoke it was found necessary that the surface of the leaves beneath the water should be four times that of the leaves above.
Closely bordering on this question is another which has excited much dispute, viz., the ability of leaves to draw water from the surrounding air or by immersion, after having suffered losses by transpiration. Prof. Boussingault's numerous experiments show that leaves, after having been exposed to influences causing a rapid evaporation, are able to absorb water rapidly on immersion, and even from an atmosphere saturated with aqueous vapor. There is, however, in both cases no absorption unless the leaves have lost a portion of their water of constitution, i. e., that which is essential to their normal existence. Thus, a wilted branch of periwinkle weighing 4 grammes, after remaining in an atmosphere saturated with aqueous vapor for a day and a half, weighed 4∙2 grammes, and after twelve hours' immersion in water 9∙4 grammes.
Fig. 3.—A, drop of solution; B, watch-glass.
The last function of leaves studied by Prof. Boussingault is their ability to absorb solutions of mineral matter, i. e., perform another of the ordinary duties of the roots. For this purpose a solution of gypsum containing 21000 of solid matter was used. Drops of this solution were placed on the leaves of a great variety of plants—under conditions favoring absorption, as in the experiments just described—and protected
