to become deflected considerably beneath the horizon.  A tentacle in the second row from the margin was selected for observation, and was found to be moving outwards at a rate of 1/500 of an inch in 20 m., or 1/100 of inch in 1 h. 40 m.; but as it likewise moved from side to side to an extent of above 1/500 of inch, the movement was probably one of modified circumnutation.  A tentacle on an old leaf was next observed in the same manner.  In 15 m. after being placed under the microscope it had moved about 1/1000 of an inch.  During the next 7 ½ h. it was looked at repeatedly, and during this whole time it moved only another 1/1000 of an inch; and this small movement may have been due to the settling of the damp sand (on which the plant rested), though the sand had been firmly pressed down.  We may therefore conclude that the tentacles when old do not circumnutate; yet this tentacle was so sensitive, that in 23 seconds after its gland had been merely touched with a bit of raw meat, it began to curl inwards.  This fact is of some importance, as it apparently shows that the inflection of the tentacles from the stimulus of absorbed animal matter (and no doubt from that of contact with any object) is not due to modified circumnutation.

(15.) Dionoea muscipula (Droseraceae).—­It should be premised that the leaves at an early stage of their development have the two lobes pressed closely together.  These are at first directed back towards the centre of the plant; but they gradually rise up and soon stand at right angles to the petiole, and ultimately in nearly a straight line with it.  A young leaf, which with the [page 240] petiole was only 1.2 inch in length, had a filament fixed externally along the midrib of the still closed lobes, which projected at right angles to the petiole.  In the evening this leaf completed an ellipse in the course of 2 h.  On the following day (Sept. 25th) its movements were traced during 22 h.; and we see in Fig. 106 that it moved in the same general direction, due to the straightening of the leaf, but in an extremely zigzag line.  This line represents several drawn-out or modified ellipses.  There can therefore be no doubt that this young leaf circumnutated.

Fig. 106.  Dionaea muscipula:  circumnutation of a young and expanding leaf, traced on a horizontal glass in darkness, from noon Sept. 24th to 10 A.M. 25th.  Apex of leaf 13 ½ inches from the glass, so tracing considerably magnified.

A rather old, horizontally extended leaf, with a filament attached along the under side of the midrib, was next observed during 7 h.  It hardly moved, but when one of its sensitive hairs was touched, the blades closed, though not very quickly.  A new dot was now made on the glass, but in the course of 14 h. 20 m. there was hardly any change in the position of the filament.  We may therefore infer that an old and only moderately sensitive leaf does not circumnutate plainly; but we shall soon see that it by no means follows that such a leaf is absolutely motionless.  We may further infer that the stimulus from a touch does not re-excite plain circumnutation.