Oct. 20 2.50 P.M...25o above horizon. 
Oct. 20 4.20 P.M...22o above horizon. 
Oct. 20 5.20 P.M...15o above horizon. 
Oct. 20 10.40 P.M...8o above horizon. 
Oct. 21 8.40 A.M...28o above horizon. 
Oct. 21 11.15 A.M...35o above horizon. 
Oct. 21 9.11 P.M...10o below horizon.

The position of the two cotyledons was roughly sketched at various hours with the same general result.

In the following summer, the hypocotyl of a fourth seedling was secured to a little stick, and a glass filament with triangles of paper having been fixed to one of the cotyledons, its movements were traced on a vertical glass under a double skylight in the house.  The first dot was made at 4.20 P.M.  June 20th; and the cotyledon fell till 10.15 P.M. in a nearly straight line.  Just past midnight it was found a little lower and somewhat to one side.  By the early morning, at 3.45 A.M., it had risen greatly, but by 6.20 A.M. had fallen a little.  During the whole of this day (21st) it fell in a slightly zigzag line, but its normal course was disturbed by the want of sufficient illumination, for during the night it rose only a little, and travelled irregularly during the whole of the following day and night of June 22nd.  The ascending and descending lines traced during the three days did not coincide, so that the movement was one of circumnutation.  This seedling was then taken back to the hot-house, and after five days was inspected at 10 P.M., when the cotyledons were found hanging so nearly vertically down, that they might justly be said to have been asleep.  On the following morning they had resumed their usual horizontal position.

Oxalis rosea (Oxalideae).—­The hypocotyl was secured to a little stick, and an extremely thin glass filament, with two triangles of paper, was attached to one of the cotyledons, which was .15 inch in length.  In this and the following species the end of the petiole, where united to the blade, is developed into a pulvinus.  The apex of the cotyledon stood only 5 inches from the vertical glass, so that its movement was not greatly exaggerated as long as it remained nearly horizontal; but in the course of the day it both rose considerably above and fell beneath a horizontal position, and then of course the movement was much exaggerated. [page 24] In Fig. 13 its course is shown from 6.45 A.M. on June 17th, to 7.40 A.M. on the following morning; and we see that during the daytime, in the course of 11 h. 15 m., it travelled thrice down and twice up.  After 5.45 P.M. it moved rapidly downwards, and in an hour or two depended vertically; it thus remained all night asleep.  This position could not be represented on the vertical glass nor in the figure here given.  By 6.40 A.M. on the following morning (18th) both cotyledons had risen greatly, and they continued to rise until 8 A.M., when they stood almost horizontally.  Their movement was traced during the whole of this day and until the next morning; but a tracing is not given, as it was closely similar to Fig. 13, excepting that the lines were more zigzag.  The cotyledons moved 7 times, either upwards or downwards; and at about 4 P.M. the great nocturnal sinking movement commenced.