We next tried how small a beam of light would act; as this bears on light serving as a guide to seedlings whilst they emerge through fissured or encumbered ground.  A pot with seedlings of Phalaris was covered

* Strasburger says (’Wirkung des Lichtes auf Schwärmsporen,’ 1878, p. 52), that the spores of Haematococcus moved to a light which only just sufficed to allow middle-sized type to be read. [page 456]

by a tin-vessel, having on one side a circular hole 1.23 mm. in diameter (i.e. a little less than the 1/20th of an inch); and the box was placed in front of a paraffin lamp and on another occasion in front of a window; and both times the seedlings were manifestly bent after a few hours towards the little hole.

A more severe trial was now made; little tubes of very thin glass, closed at their upper ends and coated with black varnish, were slipped over the cotyledons of Phalaris (which had germinated in darkness) and just fitted them.  Narrow stripes of the varnish had been previously scraped off one side, through which alone light could enter; and their dimensions were afterwards measured under the microscope.  As a control experiment, similar unvarnished and transparent tubes were tried, and they did not prevent the cotyledons bending towards the light.  Two cotyledons were placed before a south-west window, one of which was illuminated by a stripe in the varnish, only .004 inch (0.1 mm.) in breadth and .016 inch (0.4 mm.) in length; and the other by a stripe .008 inch in breadth and .06 inch in length.  The seedlings were examined after an exposure of 7 h. 40 m., and were found to be manifestly bowed towards the light.  Some other cotyledons were at the same time treated similarly, excepting that the little stripes were directed not to the sky, but in such a manner that they received only the diffused light from the room; and these cotyledons did not become at all bowed.  Seven other cotyledons were illuminated through narrow, but comparatively long, cleared stripes in the varnish—­namely, in breadth between .01 and .026 inch, and in length between .15 and .3 inch; and these all became bowed to the side, by which light entered through the stripes, whether these were directed towards the sky or to one side of [page 457] the room.  That light passing through a hole only .004 inch in breadth by .016 in length, should induce curvature, seems to us a surprising fact.

Before we knew how extremely sensitive the cotyledons of Phalaris were to light, we endeavoured to trace their circumnutation in darkness by the aid of a small wax taper, held for a minute or two at each observation in nearly the same position, a little on the left side in front of the vertical glass on which the tracing was made.  The seedlings were thus observed seventeen times in the course of the day, at intervals of from half to three-quarters of an hour; and late in the evening we were surprised to find that all the 29 cotyledons were greatly curved and pointed towards