If instead of a piece of the stem, a piece of a root is removed, the results are as follows.[31] It should be remembered that the basal end of a root is the part nearer the stem, the apex is the part nearer the apex of the root. If pieces of the root of the poplar, _Populus dilatata_, are suspended vertically (Fig. 32, _D_) in a moist chamber, a covering of new cells, a callus, appears over the cut-ends. From the basal callus numerous leaf-shoots may develop. Pieces of large roots may produce over a hundred of these shoots from a single basal callus. In some cases adventitious shoots may also arise from the side of the root near the basal end. Roots develop from the callus over the apical end; less often from the sides near the end. If a similar piece of root is suspended with its apical end upward, the new shoots arise as before over the basal end, that is now turned downwards. The leaves of some plants, as has long been known, are able to produce new plants. The begonias are especially well suited for experiments of this kind. A piece of the stalk of a leaf suspended in a moist atmosphere produces roots near its base. In most cases the opposite end of the stalk, _i.e._ the end nearest the leaf, putrefies and slowly dies toward the base. Near the base there may arise, before the breaking down of the piece has reached this point, leaf-buds that arise just above the first-formed roots. When these new shoots have reached a certain size they may produce their own roots at or near the base. If, however, a portion of the leaf is left attached to the leaf-stalk (Fig. 35, _A_), new roots arise near the basal end of the stalk, and later shoots grow out near the point of union of the leaf and its stalk at the point where the veins of the leaf come off. These shoots produce roots of their own near the base, and roots may also appear on the part of the leaf-stalk near its union with the lamina. If a part of the mid-vein, or of any large vein of the leaf, is cut out, leaving a part of the lamina on each side (Fig. 35, _B_), and the piece is suspended vertically, roots appear on the basal end of the vein, and in the same region one or more shoots arise. Leaves of heterocentron with the stalk attached, if kept in diffuse light, produce roots along the stalk, especially near the basal end, but shoots do not appear, even after five months (Fig. 35, _C_).