It is a tubular prolongation of a single cell of the external covering of the young root, usually about 1 to 3 mm. in length, and 0.01 to 0.10 mm. in diameter. In special cases the root-hairs of some water plants may reach 5 to 18 mm. in length, but of course I am referring to the ordinary land plants of agriculture and forestry. This tubular prolongation is closed and rounded off at the distal free end, and opens at the proximal end into the cell of which it is a protrusion. The whole structure is bounded by an extremely delicate and elastic wall of cellulose, which Frank says is of special composition, almost too thin to measure in many cases, but often somewhere near 0.005 to 0.001 mm. in thickness. This thin membrane is remarkably permeable by water, or dilute solutions, as is shown by the rapidity with which a root-hair collapses if exposed to evaporation, or with which dense solutions abstract water from it, or with which solutions may be seen to penetrate it under the microscope. Overlying the thin cell-wall proper, on the outside, is a thin gelatinous layer, a product of alteration of the outermost lamellæ of the former. Closely lining the proper cell-wall on the inside, is an extremely thin layer of living protoplasm, and somewhere in this protoplasm is a distinct cell-nucleus. The interior of the tube is filled with cell-sap, and it is the osmotic pressure of this cell-sap which keeps the whole living instrument tense and rigid, and the thin protoplasmic film close pressed against the cellulose cell-wall. Nothing whatever can pass into the cell-sap, or out from it, without traversing both the lining of living protoplasm and the cell-wall. If we gently pull a living root, of wheat, pea, mustard, etc., from a normal soil, we find particles of soil so closely adherent to the root-hairs that they cannot all be washed off without tearing the hairs: the root-hairs establish relations of contact with these particles, so close that they are cemented to the solid surfaces by means of the gelatinous layer already referred to. This peculiarity has the following consequences. In the first place, the enormous holdfast, ensured by the millions of points of adherence, enables the plant to withstand even powerful lever actions from above, and provides fixed points against which the root-tips can work as they drive deeper into the soil. In the second place, the intimate contact of the root-hairs and particles of soil, ensures that the films of water held by surface-action on the soil-particles and root-hairs shall be in continuity with the water saturating the cell-walls of the latter, and therefore with the protoplasm and cell-sap in their interior. The importance of this at periods when the soil is "dry" will be obvious, when we reflect that no soil is ever naturally so dry that surface-films of water are absent from the particles.