is therefore no improbability in this power having been specially acquired.  In several respects light seems to act on plants in nearly the same manner as it does on animals by means of the nervous system.* With seedlings the effect, as we have just seen, is transmitted from one part to another.  An animal may be excited to move by a very small amount of light; and it has been shown that a difference in the illumination of the two sides of the cotyledons of Phalaris, which could not be distinguished by the human eye, sufficed to cause them to bend.  It has also been shown that there is no close parallelism between the amount of light which acts on a plant and its degree of curvature; it was indeed hardly possible to perceive any difference in the curvature of some seedlings of Phalaris exposed to a light, which, though dim, was very much brighter than that to which others had been exposed.  The retina, after being stimulated by a bright light, feels the effect for some time; and Phalaris continued to bend for nearly half an hour towards the side which had been illuminated.  The retina cannot perceive a dim light after it has been exposed to a bright one; and plants which had been kept in the daylight during the previous day and morning, did not move so soon towards an obscure lateral light as did others which had been kept in complete darkness.

Even if light does act in such a manner on the growing parts of plants as always to excite in them a tendency to bend towards the more illuminated side—­a supposition contradicted by the foregoing experiments on seedlings and by all apheliotropic * Sachs has made some striking remarks to the same effect with respect to the various stimuli which excite movement in plants.  See his paper ’Ueber orthotrope und plagiotrope Pflanzentheile,’ ’Arb. des Bot.  Inst. in Würzburg,’ 1879, B. ii. p. 282. [page 488]

organs—­yet the tendency differs greatly in different species, and is variable in degree in the individuals of the same species, as may be seen in almost any pot of seedlings of a long cultivated plant.* There is therefore a basis for the modification of this tendency to almost any beneficial extent.  That it has been modified, we see in many cases:  thus, it is of more importance for insectivorous plants to place their leaves in the best position for catching insects than to turn their leaves to the light, and they have no such power.  If the stems of twining plants were to bend towards the light, they would often be drawn away from their supports; and as we have seen they do not thus bend.  As the stems of most other plants are heliotropic, we may feel almost sure that twining plants, which are distributed throughout the whole vascular series, have lost a power that their non-climbing progenitors possessed.  Moreover, with Ipomoea, and probably all other twiners, the stem of the young plant, before it begins to twine, is highly heliotropic, evidently in order to expose the cotyledons or the first true leaves fully to the light.  With the Ivy the stems of seedlings are moderately heliotropic, whilst those of the same plants when grown a little older