[15] ‘Arbeiten Bot. Institut, Würzburg,’ Heft iv. 1874, pp. 605–631. Geotropism, as was shown in the last chapter, excites the primary radicle to bend downwards with very little force, quite insufficient to penetrate the ground. Such penetration is effected by the pointed apex (protected by the root-cap) being pressed down by the longitudinal expansion or growth of the terminal rigid portion, aided by its transverse expansion, both of which forces act powerfully. It is, however, indispensable that the seeds should be at first held down in some manner. When they lie on the bare surface they are held down by the attachment of the root-hairs to any adjoining objects; and this apparently is effected by the conversion of their outer surfaces into a cement. But many seeds get covered up by various accidents, or they fall into crevices or holes. With some seeds their own weight suffices. The circumnutating movement of the terminal growing part both of the primary and secondary radicles is so feeble that it can aid them very little in penetrating the ground, excepting when the superficial layer is very soft and damp. But it must aid them materially when they happen to break obliquely into cracks, or into burrows made by earth-worms or larvae. This movement, moreover, combined with the sensitiveness of the tip to contact, can hardly fail to be of the highest importance; for as the tip is always endeavouring to bend to all sides it will press on all sides, and will thus be able to discriminate between the harder and softer adjoining surfaces, in the same manner as it discriminated between the attached squares of card-like and thin paper. Consequently it will tend to bend from the harder soil, and will thus follow the lines of least resistance. So it will be if it meets with a stone or the root of another plant in the soil, as must incessantly occur. If the tip were not sensitive, and if it did not excite the upper part of the root to bend away, whenever it encountered at right angles some obstacle in the ground, it would be liable to be doubled up into a contorted mass. But we have seen with radicles growing down inclined plates of glass, that as soon as the tip merely touched a slip of wood cemented across the plate, the whole terminal growing part curved away, so that the tip soon stood at right angles to its former direction; and thus it would be with an obstacle encountered in the ground, as far as the pressure of the surrounding soil would permit. We can also understand why thick and strong radicles, like those of Æsculus, should be endowed with less sensitiveness than more delicate ones; for the former would be able by the force of their growth to overcome any slight obstacle.