This section is from the book "The Gardener's Monthly And Horticulturist V23", by Thomas Meehan. See also: Four-Season Harvest: Organic Vegetables from Your Home Garden All Year Long.
Of the numerous problems of vegetable life which yet remain unsolved is the nature of the power the roots of plants seem to possess of taking up just those elements required to make the plant, and no other. The chemical constituents of no two plants are exactly alike - in their proportions, at least Plants of two species may grow in soil containing the same amount ot lime or potash, and yet the one species may show double the quantity in its ash of the other. The only explanation our text-books give is that the feeding roots have the power of selection. But facts sometimes appear which show that this power of selection in roots cannot be taken as an illustration of "free will," but is governed by some other power in the plant. The latest proof of this is furnished by some experiments detailed in the London Gardener's Weekly Magazine. In the spring of 1878 he grafted some tomatoes on potato stalks. The potato stems were not allowed to make any potato leaves and the tomato was prevented from sending any roots into the soil; but the tomato bore the usual tomato fruit, while beneath the soil potato tubers were borne as in the usual way.
Now the elements drawn from the earth are differently compounded in the tomato fruit and the potato apples or berries; in the tomato leaves and stems from what they are in the potato. It is not reasonable to suppose that the roots take in just what a potato root ought to take in, and that this is sent up to the tomato, which then takes what it wants and sends the rest back again; but we must conclude that the leaves have sent up to them through the potato stems to the tomato only just what they ask for, and this, of course implies that the power of selection is rather in the leaves than in the roots. A fact of additional interest in this experiment is in the production of tubers by the potato, though it had no potato leaves on the plant. It would be well worth while to pursue such investigations further, and ascertain whether the tomato leaves had any influence in changing the character of the potatoes; but this could only be tested by planting these tubers and waiting for another year. - Independent.