A moderately thick bristle (which on measurement was found rather flattened, being 0.33 mm. in one diameter, and 0.20 mm. in the other) was cut into lengths of about 1/20th of an inch.  These after being touched with thick gum-water, were placed on the tips of eleven radicles.  Three of them were affected; one being deflected in 8 h. 15 m. to an angle of about 90o from the perpendicular; a second to the same amount when looked at after 9 h.; but after 24 h. from the time of first attachment the deflection had decreased to only 19o; the third was only slightly deflected after 9 h., and the bit of bristle was then found not touching the apex; it was replaced, and after 15 additional hours the deflection amounted to 26o from the perpendicular.  The remaining eight radicles were not at all acted on by the bits of bristle, so that we here appear to have nearly reached the minimum [page 148] of size of an object which will act on the radicle of the bean.  But it is remarkable that when the bits of bristle did act, that they should have acted so quickly and efficiently.

As the apex of a radicle in penetrating the ground must be pressed on all sides, we wished to learn whether it could distinguish between harder or more resisting, and softer substances.  A square of the sanded paper, almost as stiff as card, and a square of extremely thin paper (too thin for writing on), of exactly the same size (about 1/20th of an inch), were fixed with shellac on opposite sides of the apices of 12 suspended radicles.  The sanded card was between 0.15 and 0.20 mm. (or between 0.0059 and 0.0079 of an inch), and the thin paper only 0.045 mm. (or 0.00176 of an inch) in thickness.  In 8 out of the 12 cases there could be no doubt that the radicle was deflected from the side to which the card-like paper was attached, and towards the opposite side, bearing the very thin paper.  This occurred in some instances in 9 h., but in others not until 24 h. had elapsed.  Moreover, some of the four failures can hardly be considered as really failures:  thus, in one of them, in which the radicle remained quite straight, the square of thin paper was found, when both were removed from the apex, to have been so thickly coated with shellac that it was almost as stiff as the card:  in the second case, the radicle was bent upwards into a semicircle, but the deflection was not directly from the side bearing the card, and this was explained by the two squares having become cemented laterally together, forming a sort of stiff gable, from which the radicle was deflected:  in the third case, the square of card had been fixed by mistake in front, and though there was deflection from it, this might have been due to Sachs’ curvature:  [page 149] in the fourth case alone no reason could be assigned why the radicle had not been at all deflected.  These experiments suffice to prove that the apex of the radicle possesses the extraordinary power of discriminating between thin card and very thin paper, and is deflected from the side pressed by the more resisting or harder substance.