When the leaves go to sleep, each leaflet twists half round, so as to present its edge to the zenith, and comes into close contact with its fellow. The pinnae also approach each other closely, so that the four terminal leaflets come together. The large basal leaflets (with the little rudimentary ones in contact with them) move inwards and forwards, so as to embrace the outside of the united terminal leaflets, and thus all eight leaflets [page 381] (the rudimentary ones included) form together a single vertical packet. The two pinnae at the same time that they approach each other sink downwards, and thus instead of extending horizontally in the same line with the main petiole, as during the day, they depend at night at about 45o, or even at a greater angle, beneath the horizon. The movement of the main petiole seems to be variable; we have seen it in the evening 27o lower than during the day; but sometimes in nearly the same position. Nevertheless, a sinking movement in the evening and a rising one during the night is probably the normal course, for this was well-marked in the petiole of the first-formed true leaf.

The circumnutation of the main petiole of a young leaf was traced during 2 3/4 days, and was considerable in extent, but less complex than that of M. pudica. The movement was much more lateral than is usual with circumnutating leaves, and this was the sole peculiarity which it presented. The apex of one of the terminal leaflets was seen under the microscope to travel 1/50 of an inch in 3 minutes.

Mimosa marginata.—The opposite leaflets rise up and approach each other at night, but do not come into close contact, except in the case of very young leaflets on vigorous shoots. Full-grown leaflets circumnutate during the day slowly and on a small scale.

Schrankia uncinata (Tribe 20).—A leaf consists of two or three pairs of pinnae, each bearing many small leaflets. These, when the plant is asleep, are directed forwards and become imbricated. The angle between the two terminal pinnae was diminished at night, in one case by 15o; and they sank almost vertically downwards. The hinder pairs of pinnae likewise sink downwards, but do not converge, that is, move towards the apex of the leaf. The main petiole does not become depressed, at least during the evening. In this latter respect, as well as in the sinking of the pinnae, there is a marked difference between the nyctitropic movements of the present plant and of Mimosa pudica. It should, however, be added that our specimen was not in a very vigorous condition. The pinnae of Schrankia aculeata also sink at night.

Acacia Farnesiana (Tribe 22).—The different appearance presented by a bush of this plant when asleep and awake is wonderful. The same leaf in the two states is shown in the following figure (Fig. 160). The leaflets move towards the apex of the pinna and become imbricated, and the pinnae then look like bits of dangling string. The following remarks and measurements [page 382] do not fully apply to the small leaf here figured. The pinnae move forwards and at the same time sink downwards, whilst the main petiole rises considerably. With respect to the degree of movement: the two terminal pinnae of one specimen formed together an angle of 100o during the day, and at night of only 38o, so each had moved 31o forwards. The penultimate pinnae during the day formed together an angle of 180o, that is, they stood in a straight line opposite one another, and at night each had moved 65o forwards. The basal pair of pinnae were directed

Fig. 160. Acacia Farnesiana: A, leaf during the day; B, the same leaf at night.

during the day, each about 21o backwards, and at night 38o forwards, so each had moved 59o forwards. But the pinnae at the same time sink greatly, and sometimes hang almost perpendicularly downwards. The main petiole, on the other hand, rises much: by 8.30 P.M. one stood 34o higher than at noon, and by 6.40 A.M. on the following morning it was still higher by 10o; shortly after this hour the diurnal sinking movement commenced. The course of a nearly full-grown leaf was traced during 14 h.; it was strongly zigzag, and apparently [page 383] represented five ellipses, with their longer axes differently directed.

Albizzia lophantha (Tribe 23).—The leaflets at night come into contact with one another, and are directed towards the apex of the pinna. The pinnae approach one another, but remain in the same plane as during the day; and in this respect they differ much from those of the above Schrankia and Acacia. The main petiole rises but little. The first-formed leaf above the cotyledons bore 11 leaflets on each side, and these slept like those on the subsequently formed leaves; but the petiole of this first leaf was curved downwards during the day and at night straightened itself, so that the chord of its arc then stood 16o higher than in the day-time.

Melaleuca ericaefolia (Myrtaceae).—According to Bouch ('Bot. Zeit.,' 1874, p. 359) the leaves sleep at night, in nearly the same manner as those of certain species of Pimelia.

Oenothera mollissima (Onagrarieae).—According to Linnaeus ('Somnus Plantarum'), the leaves rise up vertically at night.

Passiflora gracilis (Passifloracae).—The young leaves sleep by their blades hanging vertically downwards, and the whole length of the petiole then becomes somewhat curved downwards. Externally no trace of a pulvinus can be seen. The petiole of the uppermost leaf on a young shoot stood at 10.45 A.M. at 33o above the horizon; and at 10.30 P.M., when the blade was vertically dependent, at only 15o, so the petiole had fallen 18o. That of the next older leaf fell only 7o. From some unknown cause the leaves do not always sleep properly. The stem of a plant, which had stood for some time before a north-east window, was secured to a stick at the base of a young leaf, the blade of which was inclined at 40o below the horizon. From its position the leaf had to be viewed obliquely, consequently the vertically ascending and descending movements appeared when traced oblique. On the first day (Oct. 12th) the leaf descended in a zigzag line until late in the evening; and by 8.15 A.M. on the 13th had risen to nearly the same level as on the previous morning. A new tracing was now begun (Fig. 161). The leaf continued to rise until 8.50 A.M., then moved a little to the right, and afterwards descended. Between 11 A.M. and 5 P.M. it circumnutated, and after the latter hour the great nocturnal fall commenced. At 7.15 P.M. it depended vertically. The dotted line ought to have been prolonged much lower down in the figure. By 6.50 A.M. on the following morning (14th) the [page 384] leaf had risen greatly, and continued to rise till 7.50 A.M., after which hour it redescended. It should be observed that the lines traced on this second morning would have coincided with and confused those previously traced, had not the pot been slided a very little to the left. In the evening (14th) a mark was placed behind the filament attached to the apex of the leaf, and its movement was carefully traced from 5 P.M. to 10.15 P.M.

Fig. 161. Passiflora gracilis: circumnutation and nyctitropic movement of leaf, traced on vertical glass, from 8.20 A.M. Oct. 13th to 10 A.M. 14th. Figure reduced to two-thirds of original scale.

Between 5 and 7.15 P.M. the leaf descended in a straight line, and at the latter hour it appeared vertically dependent. But between 7.15 and 10.15 P.M. the line consisted of a succession of steps, the cause of which we could not understand; it was, however, manifest that the movement was no longer a simple descending one.

Siegesbeckia orientalis (Compositae).—Some seedlings were raised in the middle of winter and kept in the hot-house; they flowered, but did not grow well, and their leaves never showed any signs of sleep. The leaves on other seedlings raised in May were horizontal at noon (June 22nd), and depended at a consi- [page 385] derable angle beneath the horizon at 10 P.M. In the case of four youngish leaves which were from 2 to 2 inches in length, these angles were found to be 50o, 56o, 60o, and 65o. At the end of August when the plants had grown to a height of 10 to 11 inches, the younger leaves were so much curved downwards at night that they might truly be said to be asleep. This is one

Fig. 162. Nicotiana glauca: shoots with leaves expanded during the day, and asleep at night. Figures copied from photographs, and reduced.

of the species which must be well illuminated during the day in order to sleep, for on two occasions when plants were kept all day in a room with north-east windows, the leaves did not sleep at night. The same cause probably accounts for the leaves on our seedlings raised in the dead of the winter not sleeping. Professor Pfeffer informs us that the leaves of another species (S. Jorullensis ?) hang vertically down at night. [page 386]

Ipomoea caerulea and purpurea (Convolvulaceae).—The leaves on very young plants, a foot or two in height, are depressed at night to between 68o and 80o beneath the horizon; and some hang quite vertically downwards. On the following morning they again rise into a horizontal position. The petioles become at night downwardly curved, either through their entire length or in the upper part alone; and this apparently causes the depression of the blade. It seems necessary that the leaves should be well illuminated during the day in order to sleep, for those which stood on the back of a plant before a north-east window did not sleep.

Nicotiana tabacum (var. Virginian) and glauca (Solaneae).—The young leaves of both these species sleep by bending vertically upwards. Figures of two shoots of N. glauca, awake and asleep (Fig. 162), are given on p. 385: one of the shoots, from which the photographs were taken, was accidentally bent to one side.

Fig. 163. Nicotiana tabacum: circumnutation and nyctitropic movement of a leaf (5 inches in length), traced on a vertical glass, from 3 P.M. July 10th to 8.10 A.M. 13th. Apex of leaf 4 inches from glass. Temp. 17 1/2o - 18 1/2o C. Figure reduced to one-half original scale.

At the base of the petiole of N. tabacum, on the outside, there is a mass of cells, which are rather smaller than elsewhere, and [page 387] have their longer axes differently directed from the cells of the parenchyma, and may therefore be considered as forming a sort of pulvinus. A young plant of N. tabacum was selected, and the circumnutation of the fifth leaf above the cotyledons was observed during three days. On the first morning (July 10th) the leaf fell from 9 to 10 A.M., which is its normal course, but rose during the remainder of the day; and this no doubt was due to its being illuminated exclusively from above; for properly the evening rise does not commence until 3 or 4 P.M. In the figure as given on p. 386 (Fig. 163) the first dot was made at 3 P.M.; and the tracing was continued for the following 65 h. When the leaf pointed to the dot next above that marked 3 P.M. it stood horizontally. The tracing is remarkable only from its simplicity and the straightness of the lines. The leaf each day described a single great ellipse; for it should be observed that the ascending and descending lines do not coincide. On the evening of the 11th the leaf did not descend quite so low as usual, and it now zigzagged a little. The diurnal sinking movement had already commenced each morning by 7 A.M. The broken lines at the top of the figure, representing the nocturnal vertical position of the leaf, ought to be prolonged much higher up.

Mirabilis longiflora and jalapa (Nyctagineae).—The first pair of leaves above the cotyledons, produced by seedlings of both these species, were considerably divergent during the day, and at night stood up vertically in close contact with one another. The two upper leaves on an older seedling were almost horizontal by day, and at night stood up vertically, but were not in close contact, owing to the resistance offered by the central bud.

Polygonum aviculare (Polygoneae).—Professor Batalin informs us that the young leaves rise up vertically at night. This is likewise the case, according to Linnaeus, with several species of Amaranthus (Amaranthaceae); and we observed a sleep movement of this kind in one member of the genus. Again, with Chenopodium album (Chenopodieae), the upper young leaves of some seedlings, about 4 inches in height, were horizontal or sub-horizontal during the day, and at 10 P.M. on March 7th were quite, or almost quite, vertical. Other seedlings raised in the greenhouse during the winter (Jan. 28th) were observed day and night, and no difference could be perceived in the position of their leaves. According to Bouch ('Bot. Zeitung,' 1874, p. 359) the leaves of Pimelia linoides and spectabilis (Thymeleae) sleep at night. [page 388]

Euphorbia jacquiniaeflora (Euphorbiaceae).—Mr. Lynch called our attention to the fact that the young leaves of this plant sleep by depending vertically. The third leaf from the summit (March 11th) was inclined during the day 30o beneath the horizon, and at night hung vertically down, as did some of the still younger leaves. It rose up to its former level on the following morning. The fourth and fifth leaves from the summit stood horizontally during the day, and sank down at night only 38o. The sixth leaf did not sensibly alter its position. The sinking movement is due to the downward curvature of the petiole, no part of which exhibits any structure like that of a pulvinus. Early on the morning of June 7th a filament was fixed longitudinally to a young leaf (the third from the summit, and 2 5/8 inches in length), and its movements were traced on a vertical glass during 72 h., the plant being illuminated from above through a skylight. Each day the leaf fell in a nearly straight line from 7 A.M. to 5 P.M., after which hour it was so much inclined downwards that the movement could no longer be traced; and during the latter part of each night, or early in the morning, the leaf rose. It therefore circumnutated in a very simple manner, making a single large ellipse every 24 h., for the ascending and descending lines did not coincide. On each successive morning it stood at a less height than on the previous one, and this was probably due partly to the increasing age of the leaf, and partly to the illumination being insufficient; for although the leaves are very slightly heliotropic, yet, according to Mr. Lynch's and our own observations, their inclination during the day is determined by the intensity of the light. On the third day, by which time the extent of the descending movement had much decreased, the line traced was plainly much more zigzag than on any previous day, and it appeared as if some of its powers of movement were thus expended. At 10 P.M. on June 7th, when the leaf depended vertically, its movements were observed by a mark being placed behind it, and the end of the attached filament was seen to oscillate slowly and slightly from side to side, as well as upwards and downwards.

Phyllanthus Niruri (Euphorbiaceae).—The leaflets of this plant sleep, as described by Pfeffer,* in a remarkable manner, apparently like those of Cassia, for they sink downwards at night and twist round, so that their lower surfaces are turned

* 'Die Period. Beweg.,' p. 159. [page 389]

outwards. They are furnished as might have been expected from this complex kind of movement, with a pulvinus.

GYMNOSPERMS.

Pinus Nordmanniana (Coniferae).—M. Chatin states* that the leaves, which are horizontal during the day, rise up at night, so as to assume a position almost perpendicular to the branch from which they arise; we presume that he here refers to a horizontal branch. He adds: "En mme temps, ce mouvement d'rection est accompang d'un mouvement de torsion imprim la partie basilaire de la feuille, et pouvant souvent parcourir un arc de 90 degrs." As the lower surfaces of the leaves are white, whilst the upper are dark green, the tree presents a widely different appearance by day and night. The leaves on a small tree in a pot did not exhibit with us any nyctitropic movements. We have seen in a former chapter that the leaves of Pinus pinaster and Austriaca are continually circumnutating.

MONOCOTYLEDONS.

Thalia dealbata (Cannaceae).—the leaves of this plant sleep by turning vertically upwards; they are furnished with a well-developed pulvinus. It is the only instance known to us of a very large leaf sleeping. The blade of a young leaf, which was as yet only 13 1/4 inches in length and 6 in breadth, formed at noon an angle with its tall petiole of 121o, and at night stood vertically in a line with it, and so had risen 59o. The actual distance travelled by the apex (as measured by an orthogonic tracing) of another large leaf, between 7.30 A.M. and 10 P.M., was 10 inches. The circumnutation of two young and dwarfed leaves, arising amongst the taller leaves at the base of the plant, was traced on a vertical glass during two days. On the first day the apex of one, and on the second day the apex of the other leaf, described between 6.40 A.M. and 4 P.M. two ellipses, the longer axes of which were extended in very different directions from the lines representing the great diurnal sinking and nocturnal rising movement.

Maranta arundinacea (Cannaceae).—The blades of the leaves, which are furnished with a pulvinus, stand horizontally during

* 'Comptes Rendus,' Jan. 1876, p. 171. [page 390]

the day or between 10o and 20o above the horizon, and at night vertically upwards. They therefore rise between 70o and 90o at night. The plant was placed at noon in the dark in the hot-house, and on the following day the movements of the leaves were traced. Between 8.40 and 10.30 A.M. they rose, and then fell greatly till 1.37 P.M. But by 3 P.M. they had again risen a little, and continued to rise during the rest of the afternoon and night; on the following morning they stood at the same level as on the previous day. Darkness, therefore, during a day and a half does not interfere with the periodicity of their movements. On a warm but stormy evening, the plant whilst being brought into the house, had its leaves violently shaken, and at night not one went to sleep. On the next morning the plant was taken back to the hot-house, and again at night the leaves did not sleep; but on the ensuing night they rose in the usual manner between 70o and 80o. This fact is analogous with what we have observed with climbing plants, namely, that much agitation checks for a time their power of circumnutation; but the effect in this instance was much more strongly marked and prolonged.

Colocasia antiquorum (Caladium esculentum, Hort.) (Aroideae).—The leaves of this plant sleep by their blades sinking in the evening, so as to stand highly inclined, or even quite vertically with their tips pointing to the ground. They are not provided with a pulvinus. The blade of one stood at noon 1 degree beneath the horizon; at 4.20 P.M., 20o; at 6 P.M. 43o; at 7.20 P.M., 69o; and at 8.30 P.M., 68o; so it had now begun to rise; at 10.15 P.M. it stood at 65o, and on the following early morning at 11o beneath the horizon. The circumnutation of another young leaf (with its petiole only 3 1/4 inches, and the blade 4 inches in length), was traced on a vertical glass during 48 h.; it was dimly illuminated through a skylight, and this seemed to disturb the proper periodicity of its movements. Nevertheless, the leaf fell greatly during both afternoons, till either 7.10 P.M. or 9 P.M., when it rose a little and moved laterally. By an early hour on both mornings, it had assumed its diurnal position. The well-marked lateral movement for a short time in the early part of the night, was the only interesting fact which it presented, as this caused the ascending and descending lines not to coincide, in accordance with the general rule with circumnutating organs. The movements of the leaves of this plant are thus of the most simple kind; and the tracing is not worth giving. We have seen that in another genus of the Aroideae, namely, Pistia, the leaves [page 391] rise so much at night that they may almost be said to sleep.

Strephium floribundum* (Gramineae).—The oval leaves are provided with a pulvinus, and are extended horizontally or declined a little beneath the horizon during the day. Those on the upright culms simply rise up vertically at night, so that their tips are directed towards the zenith. (Fig. 164.)

Fig. 164. Strephium floribundum: culms with leaves during the day, and when asleep at night. Figures reduced.

Horizontally extended leaves arising from much inclined or almost horizontal culms, move at night so that their tips point towards the apex of the culm, with one lateral margin directed towards the zenith; and in order to assume this position the leaves have to twist on their own axes through an angle of nearly 90o. Thus the surface of the blade always stands vertically, whatever may be the position of the midrib or of the leaf as a whole.

The circumnutation of a young leaf (2.3 inches in length) was traced during 48 h. (Fig. 165). The movement was remarkably simple; the leaf descended from before 6.40 A.M. until 2 or 2.50 P.M., and then rose so as to stand vertically at about 6 P.M., descending again late in the night or in the very early morning.

* A. Brongniart first observed that the leaves of this plant and of Marsilea sleep: see 'Bull. de la Soc. Bot. de France,' tom. vii. 1860, p. 470. [page 392]

On the second day the descending line zigzagged slightly. As usual, the ascending and descending lines did not coincide. On another occasion, when the temperature was a little higher, viz., 24o - 26 1/2o C., a leaf was observed 17 times between 8.50 A.M. and 12.16 P.M.; it changed its course by as much as a rectangle six times in this interval of 3 h. 26 m., and described two irregular triangles and a half. The leaf, therefore, on this occasion circumnutated rapidly and in a complex manner.

Fig. 165. Strephium floribundum: circumnutation and nyctitropic movement of a leaf, traced from 9 A.M. June 26th to 8.45 A.M. 27th; filament fixed along the midrib. Apex of leaf 8 1/4 inches from the vertical glass; plant illuminated from above. Temp. 23 1/2o - 24 1/2o C.

ACOTYLEDONS.

Marsilea quadrifoliata (Marsileaceae).—The shape of a leaf, expanded horizontally during the day, is shown at A (Fig. 166). Each leaflet is provided with a well-developed pulvinus. When the leaves sleep, the two terminal leaflets rise up, twist half round and come into contact with one another (B), and are afterwards embraced by the two lower leaflets (C); so that the four leaflets with their lower surfaces turned outwards form a vertical packet. The curvature of the summit of the petiole of the leaf figured asleep, is merely accidental. The plant was brought into a room, where the temperature was only a little above 60o F., and the movement of one of the leaflets (the petiole having been secured) was traced [page 393] during 24 h. (Fig. 167). The leaf fell from the early morning till 1.50 P.M., and then rose till 6 P.M., when it was asleep. A

Fig. 166. Marsilea quadrifoliata: A, leaf during the day, seen from vertically above; B, leaf beginning to go to sleep, seen laterally; C, the same asleep. Figures reduced to one-half of natural scale.

vertically dependent glass filament was now fixed to one of the terminal and inner leaflets; and part of the tracing in Fig. 167, after 6 P.M., shows that it continued to sink, making one zigzag, until 10.40 P.M. At 6.45 A.M. on the following morning, the leaf was awaking, and the filament pointed above the vertical glass,

Fig. 167. Marsilea quadrifoliata: circumnutation and nyctitropic movement of leaflet traced on vertical glass, during nearly 24 h. Figure reduced to two-thirds of original scale. Plant kept at rather too low a temperature.

but by 8.25 A.M. it occupied the position shown in the figure. The diagram differs greatly in appearance from most of those previously given; and this is due to the leaflet twisting and moving laterally as it approaches and comes into contact with [page 394] its fellow. The movement of another leaflet, when asleep, was traced between 6 P.M. and 10.35 P.M., and it clearly circumnutated, for it continued for two hours to sink, then rose, and then sank still lower than it was at 6 P.M. It may be seen in the preceding figure (167) that the leaflet, when the plant was subjected to a rather low temperature in the house, descended and ascended during the middle of the day in a somewhat zigzag line; but when kept in the hot-house from 9 A.M. to 3 P.M. at a high but varying temperature (viz., between 72o and 83o F.) a leaflet (with the petiole secured) circumnutated rapidly, for it made three large vertical ellipses in the course of the six hours. According to Brongniart, Marsilea pubescens sleeps like the present species. These plants are the sole cryptogamic ones known to sleep.]

Summary and Concluding Remarks on the Nyctitropic or Sleep-movements of Leaves.—That these movements are in some manner of high importance to the plants which exhibit them, few will dispute who have observed how complex they sometimes are. Thus with Cassia, the leaflets which are horizontal during the day not only bend at night vertically downwards with the terminal pair directed considerably backwards, but they also rotate on their own axes, so that their lower surfaces are turned outwards. The terminal leaflet of Melilotus likewise rotates, by which movement one of its lateral edges is directed upwards, and at the same time it moves either to the left or to the right, until its upper surface comes into contact with that of the lateral leaflet on the same side, which has likewise rotated on its own axis. With Arachis, all four leaflets form together during the night a single vertical packet; and to the effect this the two anterior leaflets have to move upwards and the two posterior ones forwards, besides all twisting on their own axes. In the genus Sida the leaves of some species move at night through an angle of 90o upwards, and of others [page 395] through the same angle downwards. We have seen a similar difference in the nyctitropic movements of the cotyledons in the genus Oxalis. In Lupinus, again, the leaflets move either upwards or downwards; and in some species, for instance L. luteus, those on one side of the star-shaped leaf move up, and those on the opposite side move down; the intermediate ones rotating on their axes; and by these varied movements, the whole leaf forms at night a vertical star instead of a horizontal one, as during the day. Some leaves and leaflets, besides moving either upwards or downwards, become more or less folded at night, as in Bauhinia and in some species of Oxalis. The positions, indeed, which leaves occupy when asleep are almost infinitely diversified; they may point either vertically upwards or downwards, or, in the case of leaflets, towards the apex or towards the base of the leaf, or in any intermediate position. They often rotate at least as much as 90o on their own axes. The leaves which arise from upright and from horizontal or much inclined branches on the same plant, move in some few cases in a different manner, as with Porlieria and Strephium. The whole appearance of many plants is wonderfully changed at night, as may be seen with Oxalis, and still more plainly with Mimosa. A bush of Acacia Farnesiana appears at night as if covered with little dangling bits of string instead of leaves. Excluding a few genera not seen by ourselves, about which we are in doubt, and excluding a few others the leaflets of which rotate at night, and do not rise or sink much, there are 37 genera in which the leaves or leaflets rise, often moving at the same time towards the apex or towards the base of the leaf, and 32 genera in which they sink at night.

The nyctitropic movements of leaves, leaflets, and [page 396] petioles are effected in two different ways; firstly, by alternately increased growth on their opposite sides, preceded by increased turgescence of the cells; and secondly by means of a pulvinus or aggregate of small cells, generally destitute of chlorophyll, which become alternately more turgescent on nearly opposite sides; and this turgescence is not followed by growth except during the early age of the plant. A pulvinus seems to be formed (as formerly shown) by a group of cells ceasing to grow at a very early age, and therefore does not differ essentially from the surrounding tissues. The cotyledons of some species of Trifolium are provided with a pulvinus, and others are destitute of one, and so it is with the leaves in the genus Sida. We see also in this same genus gradations in the state of the development of the pulvinus; and in Nicotiana we have what may probably be considered as the commencing development of one. The nature of the movement is closely similar, whether a pulvinus is absent or present, as is evident from many of the diagrams given in this chapter. It deserves notice that when a pulvinus is present, the ascending and descending lines hardly ever coincide, so that ellipses are habitually described by the leaves thus provided, whether they are young or so old as to have quite ceased growing. This fact of ellipses being described, shows that the alternately increased turgescence of the cells does not occur on exactly opposite sides of the pulvinus, any more than the increased growth which causes the movements of leaves not furnished with pulvini. When a pulvinus is present, the nyctitropic movements are continued for a very much longer period than when such do not exist. This has been amply proved in the case of cotyledons, and Pfeffer has given observations to the same effect with respect [page 379[97]] to leaves. We have seen that a leaf of Mimosa pudica continued to move in the ordinary manner, though somewhat more simply, until it withered and died. It may be added that some leaflets of Trifolium pratense were pinned open during 10 days, and on the first evening after being released they rose up and slept in the usual manner. Besides the long continuance of the movements when effected by the aid of a pulvinus (and this appears to be the final cause of its development), a twisting movement at night, as Pfeffer has remarked, is almost confined to leaves thus provided.

It is a very general rule that the first true leaf, though it may differ somewhat in shape from the leaves on the mature plant, yet sleeps like them; and this occurs quite independently of the fact whether or not the cotyledons themselves sleep, or whether they sleep in the same manner. But with Phaseolus Roxburghii the first unifoliate leaves rise at night almost sufficiently to be said to sleep, whilst the leaflets of the secondary trifoliate leaves sink vertically at night. On young plants of Sida rhombaefolia, only a few inches in height, the leaves did not sleep, though on rather older plants they rose up vertically at night. On the other hand, the leaves on very young plants of Cytisus fragrans slept in a conspicuous manner, whilst on old and vigorous bushes kept in the greenhouse, the leaves did not exhibit any plain nyctitropic movement. In the genus Lotus the basal stipule-like leaflets rise up vertically at night, and are provided with pulvini.

As already remarked, when leaves or leaflets change their position greatly at night and by complicated movements, it can hardly be doubted that these must be in some manner beneficial to the plant. If so, we [page 398] must extend the same conclusion to a large number of sleeping plants; for the most complicated and the simplest nyctitropic movements are connected together by the finest gradations. But owing to the causes specified in the beginning of this chapter, it is impossible in some few cases to determine whether or not certain movements should be called nyctitropic. Generally, the position which the leaves occupy at night indicates with sufficient clearness, that the benefit thus derived, is the protection of their upper surfaces from radiation into the open sky, and in many cases the mutual protection of all the parts from cold by their being brought into close approximation. It should be remembered that it was proved in the last chapter, that leaves compelled to remain extended horizontally at night, suffered much more from radiation than those which were allowed to assume their normal vertical position.

The fact of the leaves of several plants not sleeping unless they have been well illuminated during the day, made us for a time doubt whether the protection of their upper surfaces from radiation was in all cases the final cause of their well-pronounced nyctitropic movements. But we have no reason to suppose that the illumination from the open sky, during even the most clouded day, is insufficient for this purpose; and we should bear in mind that leaves which are shaded from being seated low down on the plant, and which sometimes do not sleep, are likewise protected at night from full radiation. Nevertheless, we do not wish to deny that there may exist cases in which leaves change their position considerably at night, without their deriving any benefit from such movements.

Although with sleeping plants the blades almost [page 399] always assume at night a vertical, or nearly vertical position, it is a point of complete indifference whether the apex, or the base, or one of the lateral edges, is directed to the zenith. It is a rule of wide generality, that whenever there is any difference in the degree of exposure to radiation between the upper and the lower surfaces of leaves and leaflets, it is the upper which is the least exposed, as may be seen in Lotus, Cytisus, Trifolium, and other genera. In several species of Lupinus the leaflets do not, and apparently from their structure cannot, place themselves vertically at night, and consequently their upper surfaces, though highly inclined, are more exposed than the lower; and here we have an exception to our rule. But in other species of this genus the leaflets succeed in placing themselves vertically; this, however, is effected by a very unusual movement, namely, by the leaflets on the opposite sides of the same leaf moving in opposite directions.

It is again a very common rule that when leaflets come into close contact with one another, they do so by their upper surfaces, which are thus best protected. In some cases this may be the direct result of their rising vertically; but it is obviously for the protection of the upper surfaces that the leaflets of Cassia rotate in so wonderful a manner whilst sinking downwards; and that the terminal leaflet of Melilotus rotates and moves to one side until it meets the lateral leaflet on the same side. When opposite leaves or leaflets sink vertically down without any twisting, their lower surfaces approach each other and sometimes come into contact; but this is the direct and inevitable result of their position. With many species of Oxalis the lower surfaces of the adjoining leaflets are pressed together, and are thus better protected [page 400] than the upper surfaces; but this depends merely on each leaflet becoming folded at night so as to be able to sink vertically downwards. The torsion or rotation of leaves and leaflets, which occurs in so many cases, apparently always serves to bring their upper surfaces into close approximation with one another, or with other parts of the plant, for their mutual protection. We see this best in such cases as those of Arachis, Mimosa albida, and Marsilea, in which all the leaflets form together at night a single vertical packet. If with Mimosa pudica the opposite leaflets had merely moved upwards, their upper surfaces would have come into contact and been well protected; but as it is, they all successively move towards the apex of the leaf; and thus not only their upper surfaces are protected, but the successive pairs become imbricated and mutually protect one another as well as the petioles. This imbrication of the leaflets of sleeping plants is a common phenomenon.

The nyctitropic movement of the blade is generally effected by the curvature of the uppermost part of the petiole, which has often been modified into a pulvinus; or the whole petiole, when short, may be thus modified. But the blade itself sometimes curves or moves, of which fact Bauhinia offers a striking instance, as the two halves rise up and come into close contact at night. Or the blade and the upper part of the petiole may both move. Moreover, the petiole as a whole commonly either rises or sinks at night. This movement is sometimes large: thus the petioles of Cassia pubescens stand only a little above the horizon during the day, and at night rise up almost, or quite, perpendicularly. The petioles of the younger leaves of Desmodium gyrans also rise up vertically at night. On the other hand, with Amphi- [page 401] carpaea, the petioles of some leaves sank down as much as 57o at night; with Arachis they sank 39o, and then stood at right angles to the stem. Generally, when the rising or sinking of several petioles on the same plant was measured, the amount differed greatly. This is largely determined by the age of the leaf: for instance, the petiole of a moderately old leaf of Desmodium gyrans rose only 46o, whilst the young ones rose up vertically; that of a young leaf of Cassia floribunda rose 41o, whilst that of an older leaf rose only 12o. It is a more singular fact that the age of the plant sometimes influences greatly the amount of movement; thus with some young seedlings of a Bauhinia the petioles rose at night 30o and 34o, whereas those on these same plants, when grown to a height of 2 or 3 feet, hardly moved at all. The position of the leaves on the plant as determined by the light, seems also to influence the amount of movement of the petiole; for no other cause was apparent why the petioles of some leaves of Melilotus officinalis rose as much as 59o, and others only 7o and 9o at night.

In the case of many plants, the petioles move at night in one direction and the leaflets in a directly opposite one. Thus, in three genera of Phaseoleae the leaflets moved vertically downwards at night, and the petioles rose in two of them, whilst in the third they sank. Species in the same genus often differ widely in the movements of their petioles. Even on the same plant of Lupinus pubescens some of the petioles rose 30o, others only 6o, and others sank 4o at night. The leaflets of Cassia Barclayana moved so little at night that they could not be said to sleep, yet the petioles of some young leaves rose as much as 34o. These several facts apparently indicate that the movements [page 402] of the petioles are not performed for any special purpose; though a conclusion of this kind is generally rash. When the leaflets sink vertically down at night and the petioles rise, as often occurs, it is certain that the upward movement of the latter does not aid the leaflets in placing themselves in their proper position at night, for they have to move through a greater angular space than would otherwise have been necessary.

Notwithstanding what has just been said, it may be strongly suspected that in some cases the rising of the petioles, when considerable, does beneficially serve the plant by greatly reducing the surface exposed to radiation at night. If the reader will compare the two drawings (Fig. 155, p. 371) of Cassia pubescens, copied from photographs, he will see that the diameter of the plant at night is about one-third of what it is by day, and therefore the surface exposed to radiation is nearly nine times less. A similar conclusion may be deduced from the drawings (Fig. 149, p. 358) of a branch awake and asleep of Desmodium gyrans. So it was in a very striking manner with young plants of Bauhinia, and with Oxalis Ortegesii.

We are led to an analogous conclusion with respect to the movements of the secondary petioles of certain pinnate leaves. The pinnae of Mimosa pudica converge at night; and thus the imbricated and closed leaflets on each separate pinna are all brought close together into a single bundle, and mutually protect one another, with a somewhat smaller surface exposed to radiation. With Albizzia lophantha the pinnae close together in the same manner. Although the pinnae of Acacia Farnesiana do not converge much, they sink downwards. Those of Neptunia oleracea likewise [page 403] move downwards, as well as backwards, towards the base of the leaf, whilst the main petiole rises. With Schrankia, again, the pinnae are depressed at night. Now in these three latter cases, though the pinnae do not mutually protect one another at night, yet after having sunk down they expose, as does a dependent sleeping leaf, much less surface to the zenith and to radiation than if they had remained horizontal.

Any one who had never observed continuously a sleeping plant, would naturally suppose that the leaves moved only in the evening when going to sleep, and in the morning when awaking; but he would be quite mistaken, for we have found no exception to the rule that leaves which sleep continue to move during the whole twenty-four hours; they move, however, more quickly when going to sleep and when awaking than at other times. That they are not stationary during the day is shown by all the diagrams given, and by the many more which were traced. It is troublesome to observe the movements of leaves in the middle of the night, but this was done in a few cases; and tracings were made during the early part of the night of the movements in the case of Oxalis, Amphicarpaea, two species of Erythrina, a Cassia, Passiflora, Euphorbia and Marsilea; and the leaves after they had gone to sleep, were found to be in constant movement. When, however, opposite leaflets come into close contact with one another or with the stem at night, they are, as we believe, mechanically prevented from moving, but this point was not sufficiently investigated.

When the movements of sleeping leaves are traced during twenty-four hours, the ascending and descending lines do not coincide, except occasionally and by accident for a short space; so that with many plants a [page 404] single large ellipse is described during each twenty-four hours. Such ellipses are generally narrow and vertically directed, for the amount of lateral movement is small. That there is some lateral movement is shown by the ascending and descending lines not coinciding, and occasionally, as with Desmodium gyrans and Thalia dealbata, it was strongly marked. In the case of Melilotus the ellipses described by the terminal leaflet during the day are laterally extended, instead of vertically, as is usual; and this fact evidently stands in relation with the terminal leaflet moving laterally when it goes to sleep. With the majority of sleeping plants the leaves oscillate more than once up and down in the twenty-four hours; so that frequently two ellipses, one of moderate size, and one of very large size which includes the nocturnal movement, are described within the twenty-four hours. For instance, a leaf which stands vertically up during the night will sink in the morning, then rise considerably, again sink in the afternoon, and in the evening reascend and assume its vertical nocturnal position. It will thus describe, in the course of the twenty-four hours, two ellipses of unequal sizes. Other plants describe within the same time, three, four, or five ellipses. Occasionally the longer axes of the several ellipses extend in different directions, of which Acacia Farnesiana offered a good instance. The following cases will give an idea of the rate of movement: Oxalis acetosella completed two ellipses at the rate of 1 h. 25 m. for each; Marsilea quadrifoliata, at the rate of 2 h.; Trifolium subterraneum, one in 3 h. 30 m.; and Arachis hypogaea, in 4 h. 50 m. But the number of ellipses described within a given time depends largely on the state of the plant and on the conditions to which it is exposed. It often happens that a single ellipse may be described during one [page 405] day, and two on the next. Erythrina corallodendron made four ellipses on the first day of observation and only a single one on the third, apparently owing to having been kept not sufficiently illuminated and perhaps not warm enough. But there seems likewise to be an innate tendency in different species of the same genus to make a different number of ellipses in the twenty-four hours: the leaflets of Trifolium repens made only one; those of T. resupinatum two, and those of T. subterraneum three in this time. Again, the leaflets of Oxalis Plumierii made a single ellipse; those of O. bupleurifolia, two; those of O. Valdiviana, two or three; and those of O. acetosella, at least five in the twenty-four hours.

The line followed by the apex of a leaf or leaflet, whilst describing one or more ellipses during the day, is often zigzag, either throughout its whole course or only during the morning or evening: Robinia offered an instance of zigzagging confined to the morning, and a similar movement in the evening is shown in the diagram (Fig. 126) given under Sida. The amount of the zigzag movement depends largely on the plant being placed under highly favourable conditions. But even under such favourable conditions, if the dots which mark the position of the apex are made at considerable intervals of time, and the dots are then joined, the course pursued will still appear comparatively simple, although the number of the ellipses will be increased; but if dots are made every two or three minutes and these are joined, the result often is that all the lines are strongly zigzag, many small loops, triangles, and other figures being also formed. This fact is shown in two parts of the diagram (Fig. 150) of the movements of Desmodium gyrans. Strephium floribundum, observed under a high temperature, [page 406] made several little triangles at the rate of 43 m. for each. Mimosa pudica, similarly observed, described three little ellipses in 67 m.; and the apex of a leaflet crossed 1/500 of an inch in a second, or 0.12 inch in a minute. The leaflets of Averrhoa made a countless number of little oscillations when the temperature was high and the sun shining. The zigzag movement may in all cases be considered as an attempt to form small loops, which are drawn out by a prevailing movement in some one direction. The rapid gyrations of the little lateral leaflets of Desmodium belong to the same class of movements, somewhat exaggerated in rapidity and amplitude. The jerking movements, with a small advance and still smaller retreat, apparently not exactly in the same line, of the hypocotyl of the cabbage and of the leaves of Dionaea, as seen under the microscope, all probably come under this same head. We may suspect that we here see the energy which is freed during the incessant chemical changes in progress in the tissues, converted into motion. Finally, it should be noted that leaflets and probably some leaves, whilst describing their ellipses, often rotate slightly on their axes; so that the plane of the leaf is directed first to one and then to another side. This was plainly seen to be the case with the large terminal leaflets of Desmodium, Erythrina and Amphicarpaea, and is probably common to all leaflets provided with a pulvinus.

With respect to the periodicity of the movements of sleeping leaves, Pfeffer* has so clearly shown that this depends on the daily alternations of light and darkness, that nothing farther need be said on this

* 'Die Periodischen Bewegungen der Blattorgane,' 1875, p. 30, et passim. [page 407]

head. But we may recall the behaviour of Mimosa in the North, where the sun does not set, and the complete inversion of the daily movements by artificial light and darkness. It has also been shown by us, that although leaves subjected to darkness for a moderately long time continue to circumnutate, yet the periodicity of their movements is soon greatly disturbed, or quite annulled. The presence of light or its absence cannot be supposed to be the direct cause of the movements, for these are wonderfully diversified even with the leaflets of the same leaf, although all have of course been similarly exposed. The movements depend on innate causes, and are of an adaptive nature. The alternations of light and darkness merely give notice to the leaves that the period has arrived for them to move in a certain manner. We may infer from the fact of several plants (Tropaeolum, Lupinus, etc.) not sleeping unless they have been well illuminated during the day, that it is not the actual decrease of light in the evening, but the contrast between the amount at this hour and during the early part of the day, which excites the leaves to modify their ordinary mode of circumnutation.

As the leaves of most plants assume their proper diurnal position in the morning, although light be excluded, and as the leaves of some plants continue to move in the normal manner in darkness during at least a whole day, we may conclude that the periodicity of their movements is to a certain extent inherited.* The strength of such inheritance differs

* Pfeffer denies such inheritance; he attributes ('Die Period. Bewegungen,' pp. 30-56) the periodicity when prolonged for a day or two in darkness, to "Nachwirkung," or the after-effects of light and darkness. But we are unable to follow his train of reasoning. There does not seem to be any more reason for [[page 408]] attributing such movements to this cause than, for instance, the inherited habit of winter and summer wheat to grow best at different seasons; for this habit is lost after a few years, like the movements of leaves in darkness after a few days. No doubt some effect must be produced on the seeds by the long-continued cultivation of the parent-plants under different climates, but no one probably would call this the "Nachwirkung" of the climates. [page 408] much in different species, and seems never to be rigid; for plants have been introduced from all parts of the world into our gardens and greenhouses; and if their movements had been at all strictly fixed in relation to the alternations of day and night, they would have slept in this country at very different hours, which is not the case. Moreover, it has been observed that sleeping plants in their native homes change their times of sleep with the changing seasons.*

We may now turn to the systematic list. This contains the names of all the sleeping plants known to us, though the list undoubtedly is very imperfect. It may be premised that, as a general rule, all the species in the same genus sleep in nearly the same manner. But there are some exceptions; in several large genera including many sleeping species (for instance, Oxalis), some do not sleep. One species of Melilotus sleeps like a Trifolium, and therefore very differently from its congeners; so does one species of Cassia. In the genus Sida, the leaves either rise or fall at night; and with Lupinus they sleep in three different methods. Returning to the list, the first point which strikes us, is that there are many more genera amongst the Leguminosae (and in almost every one of the Leguminous tribes) than in all the other families put together; and we are tempted to connect this fact with the great

* Pfeffer, ibid., p. 46. [page 409]

mobility of the stems and leaves in this family, as shown by the large number of climbing species which it contains. Next to the Leguminosae come the Malvaceae, together with some closely allied families. But by far the most important point in the list, is that we meet with sleeping plants in 28 families, in all the great divisions of the Phanerogamic series, and in one Cryptogam. Now, although it is probable that with the Leguminosae the tendency to sleep may have been inherited from one or a few progenitors, and possibly so in the cohorts of the Malvales and Chenopodiales, yet it is manifest that the tendency must have been acquired by the several genera in the other families, quite independently of one another. Hence the question naturally arises, how has this been possible? and the answer, we cannot doubt is that leaves owe their nyctitropic movements to their habit of circumnutating,—a habit common to all plants, and everywhere ready for any beneficial development or modification.

It has been shown in the previous chapters that the leaves and cotyledons of all plants are continually moving up and down, generally to a slight but sometimes to a considerable extent, and that they describe either one or several ellipses in the course of twenty-four hours; they are also so far affected by the alternations of day and night that they generally, or at least often, move periodically to a small extent; and here we have a basis for the development of the greater nyctitropic movements. That the movements of leaves and cotyledons which do not sleep come within the class of circumnutating movements cannot be doubted, for they are closely similar to those of hypocotyls, epicotyls, the stems of mature plants, and of various other organs. Now, if we take the simplest [page 410] case of a sleeping leaf, we see that it makes a single ellipse in the twenty-four hours, which resembles one described by a non-sleeping leaf in every respect, except that it is much larger. In both cases the course pursued is often zigzag. As all non-sleeping leaves are incessantly circumnutating, we must conclude that a part at least of the upward and downward movement of one that sleeps, is due to ordinary circumnutation; and it seems altogether gratuitous to rank the remainder of the movement under a wholly different head. With a multitude of climbing plants the ellipses which they describe have been greatly increased for another purpose, namely, catching hold of a support. With these climbing plants, the various circumnutating organs have been so far modified in relation to light that, differently from all ordinary plants, they do not bend towards it. with sleeping plants the rate and amplitude of the movements of the leaves have been so far modified in relation to light, that they move in a certain direction with the waning light of the evening and with the increasing light of the morning more rapidly, and to a greater extent, than at other hours.

But the leaves and cotyledons of many non-sleeping plants move in a much more complex manner than in the cases just alluded to, for they describe two, three, or more ellipses in the course of a day. Now, if a plant of this kind were converted into one that slept, one side of one of the several ellipses which each leaf daily describes, would have to be greatly increased in length in the evening, until the leaf stood vertically, when it would go on circumnutating about the same spot. On the following morning, the side of another ellipse would have to be similarly increased in length so as to bring the leaf back again into its diurnal position, when it would again circumnutate [page 411] until the evening. If the reader will look, for instance, at the diagram (Fig. 142, p. 351), representing the nyctitropic movements of the terminal leaflet of Trifolium subterraneum, remembering that the curved broken lines at the top ought to be prolonged much higher up, he will see that the great rise in the evening and the great fall in the morning together form a large ellipse like one of those described during the daytime, differing only in size. Or, he may look at the diagram (Fig. 103, p. 236) of the 3 ellipses described in the course of 6 h. 35 m. by a leaf of Lupinus speciosus, which is one of the species in this genus that does not sleep; and he will see that by merely prolonging upwards the line which was already rising late in the evening, and bringing it down again next morning, the diagram would represent the movements of a sleeping plant.

With those sleeping plants which describe several ellipses in the daytime, and which travel in a strongly zigzag line, often making in their course minute loops, triangles, etc., if as soon as one of the ellipses begins in the evening to be greatly increased in size, dots are made every 2 or 3 minutes and these are joined, the line then described is almost strictly rectilinear, in strong contrast with the lines made during the daytime. This was observed with Desmodium gyrans and Mimosa pudica. With this latter plant, moreover, the pinnae converge in the evening by a steady movement, whereas during the day they are continually converging and diverging to a slight extent. In all such cases it was scarcely possible to observe the difference in the movement during the day and evening, without being convinced that in the evening the plant saves the expenditure of force by not moving laterally, and that its whole energy is now expended [page 412] in gaining quickly its proper nocturnal position by a direct course. In several other cases, for instance, when a leaf after describing during the day one or more fairly regular ellipses, zigzags much in the evening, it appears as if energy was being expended, so that the great evening rise or fall might coincide with the period of the day proper for this movement.

The most complex of all the movements performed by sleeping plants, is that when leaves or leaflets, after describing in the daytime several vertically directed ellipses, rotate greatly on their axes in the evening, by which twisting movement they occupy a wholly different position at night to what they do during the day. For instance, the terminal leaflets of Cassia not only move vertically downwards in the evening, but twist round, so that their lower surfaces face outwards. Such movements are wholly, or almost wholly, confined to leaflets provided with a pulvinus. But this torsion is not a new kind of movement introduced solely for the purpose of sleep; for it has been shown that some leaflets whilst describing their ordinary ellipses during the daytime rotate slightly, causing their blades to face first to one side and then to another. Although we can see how the slight periodical movements of leaves in a vertical plane could be easily converted into the greater yet simple nyctitropic movements, we do not at present know by what graduated steps the more complex movements, effected by the torsion of the pulvini, have been acquired. A probable explanation could be given in each case only after a close investigation of the movements in all the allied forms.

From the facts and considerations now advanced we may conclude that nyctitropism, or the sleep of leaves [page 413] and cotyledons, is merely a modification of their ordinary circumnutating movement, regulated in its period and amplitude by the alternations of light and darkness. The object gained is the protection of the upper surfaces of the leaves from radiation at night, often combined with the mutual protection of the several parts by their close approximation. In such cases as those of the leaflets of Cassia—of the terminal leaflets of Melilotus—of all the leaflets of Arachis, Marsilea, etc.—we have ordinary circumnutation modified to the extreme extent known to us in any of the several great classes of modified circumnutation. On this view of the origin of nyctitropism we can understand how it is that a few plants, widely distributed throughout the Vascular series, have been able to acquire the habit of placing the blades of their leaves vertically at night, that is, of sleeping,—a fact otherwise inexplicable.

The leaves of some plants move during the day in a manner, which has improperly been called diurnal sleep; for when the sun shines brightly on them, they direct their edges towards it. To such cases we shall recur in the following chapter on Heliotropism. It has been shown that the leaflets of one form of Porlieria hygrometrica keep closed during the day, as long as the plant is scantily supplied with water, in the same manner as when asleep; and this apparently serves to check evaporation. There is only one other analogous case known to us, namely, that of certain Gramineae, which fold inwards the sides of their narrow leaves, when these are exposed to the sun and to a dry atmosphere, as described by Duval-Jouve.* We have also observed the same phenomenon in Elymus arenareus.

* 'Annal. des Sc. Nat. (Bot.),' 1875, tom. i. pp. 326-329. [page 414]

There is another movement, which since the time of Linnaeus has generally been called sleep, namely, that of the petals of the many flowers which close at night. These movements have been ably investigated by Pfeffer, who has shown (as was first observed by Hofmeister) that they are caused or regulated more by temperature than by the alternations of light and darkness. Although they cannot fail to protect the organs of reproduction from radiation at night, this does not seem to be their chief function, but rather the protection of the organs from cold winds, and especially from rain, during the day. the latter seems probable, as Kerner* has shown that a widely different kind of movement, namely, the bending down of the upper part of the peduncle, serves in many cases the same end. The closure of the flowers will also exclude nocturnal insects which may be ill-adapted for their fertilisation, and the well-adapted kinds at periods when the temperature is not favourable for fertilisation. Whether these movements of the petals consist, as is probable, of modified circumnutation we do not know.

Embryology of Leaves.—A few facts have been incidentally given in this chapter on what may be called the embryology of leaves. With most plants the first leaf which is developed after the cotyledons, resembles closely the leaves produced by the mature plant, but this is not always the case. the first leaves produced by some species of Drosera, for instance by D. Capensis, differ widely in shape from those borne by the mature plant, and resemble closely the leaves of D. rotundifolia, as was shown to us by Prof. Williamson of Manchester. The first true leaf of

* 'Die Schutzmittel des Pollens,' 1873, pp. 30-39. [page 415]

the gorse, or Ulex, is not narrow and spinose like the older leaves. On the other hand, with many Leguminous plants, for instance, Cassia, Acacia lophantha, etc., the first leaf has essentially the same character as the older leaves, excepting that it bears fewer leaflets. In Trifolium the first leaf generally bears only a single leaflet instead of three, and this differs somewhat in shape from the corresponding leaflet on the older leaves. Now, with Trifolium Pannonicum the first true leaf on some seedlings was unifoliate, and on others completely trifoliate; and between these two extreme states there were all sorts of gradations, some seedlings bearing a single leaflet more or less deeply notched on one or both sides, and some bearing a single additional and perfect lateral leaflet. Here, then, we have the rare opportunity of seeing a structure proper to a more advanced age, in the act of gradually encroaching on and replacing an earlier or embryological condition.

The genus Melilotus is closely allied to Trifolium, and the first leaf bears only a single leaflet, which at night rotates on its axis so as to present one lateral edge to the zenith. Hence it sleeps like the terminal leaflet of a mature plant, as was observed in 15 species, and wholly unlike the corresponding leaflet of Trifolium, which simply bends upwards. It is therefore a curious fact that in one of these 15 species, viz., M. Taurica (and in a lesser degree in two others), leaves arising from young shoots, produced on plants which had been cut down and kept in pots during the winter in the green-house, slept like the leaves of a Trifolium, whilst the leaves on the fully-grown branches on these same plants afterwards slept normally like those of a Melilotus. If young shoots rising from the ground may be considered as new individuals, partaking to a certain extent of the nature of seedlings, then the peculiar manner in which their leaves slept may be considered [page 416] as an embryological habit, probably the result of Melilotus being descended from some form which slept like a Trifolium. This view is partially supported by the leaves on old and young branches of another species, M. Messanensis (not included in the above 15 species), always sleeping like those of a Trifolium.

The first true leaf of Mimosa albida consists of a simple petiole, often bearing three pairs of leaflets, all of which are of nearly equal size and of the same shape: the second leaf differs widely from the first, and resembles that on a mature plant (see Fig. 159, p. 379), for it consists of two pinnae, each of which bears two pairs of leaflets, of which the inner basal one is very small. But at the base of each pinna there is a pair of minute points, evidently rudiments of leaflets, for they are of unequal sizes, like the two succeeding leaflets. These rudiments are in one sense embryological, for they exist only during the youth of the leaf, falling off and disappearing as soon as it is fully grown.

With Desmodium gyrans the two lateral leaflets are very much smaller than the corresponding leaflets in most of the species in this large genus; they vary also in position and size; one or both are sometimes absent; and they do not sleep like the fully-developed leaflets. They may therefore be considered as almost rudimentary; and in accordance with the general principles of embryology, they ought to be more constantly and fully developed on very young than on old plants. But this is not the case, for they were quite absent on some young seedlings, and did not appear until from 10 to 20 leaves had been formed. This fact leads to the suspicion that D. gyrans is descended through a unifoliate form (of which some exist) from a trifoliate species; and that the little lateral leaflets reappear through reversion. However this may be, [page 417] the interesting fact of the pulvini or organs of movement of these little leaflets, not having been reduced nearly so much as their blades—taking the large terminal leaflet as the standard of comparison—gives us probably the proximate cause of their extraordinary power of gyration. [page 418]

CHAPTER VIII.

MODIFIED CIRCUMNUTATION: MOVEMENTS EXCITED BY LIGHT.

Distinction between heliotropism and the effects of light on the periodicity of the movements of leaves—Heliotropic movements of Beta, Solanum, Zea, and Avena—Heliotropic movements towards an obscure light in Apios, Brassica, Phalaris, Tropaeolum, and Cassia—Apheliotropic movements of tendrils of Bignonia—Of flower-peduncles of Cyclamen—Burying of the pods—Heliotropism and apheliotropism modified forms of circumnutation— Steps by which one movement is converted into the other— Transversal-heliotropismus or diaheliotropism influenced by epinasty, the weight of the part and apogeotropism—Apogeotropism overcome during the middle of the day by diaheliotropism—Effects of the weight of the blades of cotyledons—So called diurnal sleep—Chlorophyll injured by intense light—Movements to avoid intense light

SACHS first clearly pointed out the important difference between the action of light in modifying the periodic movements of leaves, and in causing them to bend towards its source.* The latter, or heliotropic movements are determined by the direction of the light, whilst periodic movements are affected by changes in its intensity and not by its direction. The periodicity of the circumnutating movement often continues for some time in darkness, as we have seen in the last chapter; whilst heliotropic bending ceases very quickly when the light fails. Nevertheless, plants which have ceased through long-continued darkness to move periodically, if re-exposed to the light are still, according to Sachs, heliotropic.

Apheliotropism, or, as usually designated, negative

* 'Physiologie Veg.' (French Translation), 1868, pp. 42, 517, etc. [page 419]

heliotropism, implies that a plant, when unequally illuminated on the two sides, bends from the light, instead of, as in the last sub-class of cases, towards it; but apheliotropism is comparatively rare, at least in a well-marked degree. There is a third and large sub-class of cases, namely, those of "transversal-Heliotropismus" of Frank, which we will here call diaheliotropism. Parts of plants, under this influence, place themselves more or less transversely to the direction whence the light proceeds, and are thus fully illuminated. There is a fourth sub-class, as far as the final cause of the movement is concerned; for the leaves of some plants when exposed to an intense and injurious amount of light direct themselves, by rising or sinking or twisting, so as to be less intensely illuminated. Such movements have sometimes been called diurnal sleep. If thought advisable, they might be called paraheliotropic, and this term would correspond with our other terms.

It will be shown in the present chapter that all the movements included in these four sub-classes, consist of modified circumnutation. We do not pretend to say that if a part of a plant, whilst still growing, did not circumnutate—though such a supposition is most improbable—it could not bend towards the light; but, as a matter of fact, heliotropism seems always to consist of modified circumnutation. Any kind of movement in relation to light will obviously be much facilitated by each part circumnutating or bending successively in all directions, so that an already existing movement has only to be increased in some one direction, and to be lessened or stopped in the other directions, in order that it should become heliotropic, apheliotropic, etc., as the case may be. In the next chapter some observations on the sensitiveness of plants to light, their [page 420] rate of bending towards it, and the accuracy with which they point towards its source, etc., will be given. Afterwards it will be shown—and this seems to us a point of much interest—that sensitiveness to light is sometimes confined to a small part of the plant; and that this part when stimulated by light, transmits an influence to distant parts, exciting them to bend.

Heliotropism.—When a plant which is strongly heliotropic (and species differ much in this respect) is exposed to a bright lateral light, it bends quickly towards it, and the course pursued by the stem is quite or nearly straight. But if the light is much dimmed, or occasionally interrupted, or admitted in only a slightly oblique direction, the course pursued is more or less zigzag; and as we have seen and shall again see, such zigzag movement results from the elongation or drawing out of the ellipses, loops, etc., which the plant would have described, if it had been illuminated from above. On several occasions we were much struck with this fact, whilst observing the circumnutation of highly sensitive seedlings, which were unintentionally illuminated rather obliquely, or only at successive intervals of time.

Fig. 168. Beta vulgaris: circumnutation of hypocotyl, deflected by the light being slightly lateral, traced on a horizontal glass from 8.30 A.M. to 5.30 P.M. Direction of the lighted taper by which it was illuminated shown by a line joining the first and penultimate dots. Figure reduced to one-third of the original scale.

[For instance two young seedlings of Beta vulgaris were placed in the middle of a room with north-east windows, and were kept covered up, except during each observation which lasted for only a minute or two; but the result was that their hypocotyls bowed themselves to the side, whence some light occasionally entered, in lines which were [page 421] only slightly zigzag. Although not a single ellipse was even approximately formed, we inferred from the zigzag lines - and, as it proved, correctly— that their hypocotyls were circumnutating, for on the following day these same seedlings were placed in a completely darkened room, and were observed each time by the aid of a small wax taper held almost directly above them, and their movements were traced on a horizontal glass above; and now their hypocotyls clearly circumnutated (Fig. 168, and Fig. 39, formerly given, p. 52); yet they moved a short distance towards the side where the taper was held up. If we look at these diagrams, and suppose that the taper had been held more on one side, and that the hypocotyls, still circumnutating, had bent themselves within the same time much more towards the light, long zigzag lines would obviously have been the result.

Fig. 169. Avena sativa: heliotropic movement and circumnutation of sheath-like cotyledon (1 inch in height) traced on horizontal glass from 8 A.M. to 10.25 P.M. Oct. 16th.

Again, two seedlings of Solanum lycopersicum were illuminated from above, but accidentally a little more light entered on one than on any other side, and their hypocotyls became slightly bowed towards the brighter side; they moved in a zigzag line and described in their course two little triangles, as seen in Fig. 37 (p. 50), and in another tracing not given. The sheath-like cotyledons of Zea mays behaved, under nearly similar circumstances, in a nearly similar manner as described in our first chapter (p. 64), for they bowed themselves during the whole day towards one side, making, however, in their course some conspicuous flexures. Before we knew how greatly ordinary circumnutation was modified by a lateral light, some seedling oats, with rather old and therefore not highly sensitive cotyledons, were placed in front of a north-east window, towards which they bent all day in a strongly zigzag course. On the following day they continued to bend in the same direction (Fig. 169), but zigzagged much less. The sky, however, became between 12.40 and 2.35 P.M. [page 422] overcast with extraordinarily dark thunder-clouds, and it was interesting to note how plainly the cotyledons circumnutated during this interval.

The foregoing observations are of some value, from having been made when we were not attending to heliotropism; and they led us to experiment on several kinds of seedlings, by exposing them to a dim lateral light, so as to observe the gradations between ordinary circumnutation and heliotropism. Seedlings in pots were placed in front of, and about a yard from, a north-east window; on each side and over the pots black boards were placed; in the rear the pots were open to the diffused light of the room, which had a second north-east and a north-west window. By hanging up one or more blinds before the window where the seedlings stood, it was easy to dim the light, so that very little more entered on this side than on the opposite one, which received the diffused light of the room. Late in the evening the blinds were successively removed, and as the plants had been subjected during the day to a very obscure light, they continued to bend towards the window later in the evening than would otherwise have occurred. Most of the seedlings were selected because they were known to be highly sensitive to light, and some because they were but little sensitive, or had become so from having grown old. The movements were traced in the usual manner on a horizontal glass cover; a fine glass filament with little triangles of paper having been cemented in an upright position to the hypocotyls. Whenever the stem or hypocotyl became much bowed towards the light, the latter part of its course had to be traced on a vertical glass, parallel to the window, and at right angles to the horizontal glass cover. Fig. 170. Apios graveolens: heliotropic movement of hypocotyl (.45 of inch in height) towards a moderately bright lateral light, traced on a horizontal glass from 8.30 A.M. to 11.30 A.M. Sept. 18th. Figure reduced to one-third of original scale.

Apios graveolens.—The hypocotyl bends in a few hours rectan- [page 423] gularly towards a bright lateral light. In order to ascertain how straight a course it would pursue when fairly well illuminated on one side, seedlings were first placed before a south-west window on a cloudy and rainy morning; and the movement of two hypocotyls were traced for 3 h., during which time they became greatly bowed towards the light. One of these tracings is given on p. 422 (Fig. 170), and the course may be seen to be almost straight. But the amount of light on this occasion was superfluous, for two seedlings were placed before a north-east window, protected by an ordinary linen and two muslin blinds, yet their hypocotyls moved towards this rather dim light in only slightly zigzag lines; but after 4 P.M., as the light waned, the lines became distinctly zigzag. One of these seedlings, moreover, described in the afternoon an ellipse of considerable size, with its longer axis directed towards the window.

We now determined that the light should be made dim enough, so we began by exposing several seedlings before a north-east window, protected by one linen blind, three muslin blinds, and a towel. But so little light entered that a pencil cast no perceptible shadow on a white card, and the hypocotyls did not bend at all towards the window. During this time, from 8.15 to 10.50 A.M., the hypocotyls zigzagged or circumnutated near the same spot, as may be seen at A, in Fig. 171. The towel, therefore, was removed at 10.50 A.M., and replaced by two muslin blinds, and now the light passed through one ordinary linen and four muslin blinds. When a pencil was held upright on a card close to the seedlings, it cast a shadow (pointing from the window) which could only just be distinguished. Yet this very slight excess of light on one side sufficed to cause the hypocotyls of all the seedlings immediately to begin bending in zigzag lines towards the window. The course of one is shown at A (Fig. 171): after moving towards the window from 10.50 A.M. to 12.48 P.M. it bent from the window, and then returned in a nearly parallel line; that is, it almost completed between 12.48 and 2 P.M. a narrow ellipse. Late in the evening, as the light waned, the hypocotyl ceased to bend towards the window, and circumnutated on a small scale round the same spot; during the night it moved considerably backwards, that is, became more upright, through the action of apogeotropism. At B, we have a tracing of the movements of another seedling from the hour (10.50 A.M.) when the towel was removed; and it is in all essential respects [page 424] similar to the previous one. In these two cases there could be no doubt that the ordinary circumnutating movement of the hypocotyl was modified and rendered heliotropic.

Fig. 171. Apios graveolens: heliotropic movement and circumnutation of the hypocotyls of two seedlings towards a dim lateral light, traced on a horizontal glass during the day. The broken lines show their return nocturnal courses. Height of hypocotyl of A .5, and of B .55 inch. Figure reduced to one-half of original scale.

Brassica oleracea.—The hypocotyl of the cabbage, when not disturbed by a lateral light, circumnutates in a complicated [page 425] manner over nearly the same space, and a figure formerly given is here reproduced (Fig. 172). If the hypocotyl is exposed to a moderately strong lateral light it moves quickly towards this side, travelling in a straight, or nearly straight, line. But when the lateral light is very dim its course is extremely tortuous, and evidently consists of modified circumnutation. Seedlings were placed before a north-east window, protected by a linen and muslin blind and by a towel. The sky was cloudy, and whenever the clouds grew a little lighter an additional muslin blind was temporarily suspended. The light from the window was

Fig. 172. Brassica oleracea: ordinary circumnutating movement of the hypocotyl of a seedling plant.

thus so much obscured that, judging by the unassisted eye, the seedlings appeared to receive more light from the interior of the room than from the window; but this was not really the case, as was shown by a very faint shadow cast by a pencil on a card. Nevertheless, this extremely small excess of light on one side caused the hypocotyls, which in the morning had stood upright, to bend at right angles towards the window, so that in the evening (after 4.23 P.M.) their course had to be traced on a vertical glass parallel to the window. It should be stated that at 3.30 P.M., by which time the sky had become darker, the towel was removed and replaced by an additional muslin blind, which itself was removed at 4 P.M., the other two [page 426] blinds being left suspended. In Fig. 173 the course pursued, between 8.9 A.M. and 7.10 P.M., by one of the hypocotyls thus

Fig. 173. Brassica oleracea: heliotropic movement and circumnutation of a hypocotyl towards a very dim lateral light, traced during 11 hours, on a horizontal glass in the morning, and on a vertical glass in the evening. Figure reduced to one-third of the original scale.

exposed is shown. It may be observed that during the first 16 m. the hypocotyl moved obliquely from the light, and this, [page 427] no doubt, was due to its then circumnutating in this direction. Similar cases were repeatedly observed, and a dim light rarely or never produced any effect until from a quarter to three-quarters of an hour had elapsed. After 5.15 P.M., by which time the light had become obscure, the hypocotyl began to circumnutate about the same spot. The contrast between the two figures (172 and 173) would have been more striking, if they had been originally drawn on the same scale, and had been equally reduced. But the movements shown in Fig. 172 were at first more magnified, and have been reduced to only one-half of the original scale; whereas those in Fig. 173 were at first less magnified, and have been reduced to a one-third scale. A tracing made at the same time with the last of the movements of a second hypocotyl, presented a closely analogous appearance; but it did not bend quite so much towards the light, and it circumnutated rather more plainly.

Fig. 174. Phalaris Canariensis: heliotropic movement and circumnutation of a rather old cotyledon, towards a dull lateral light, traced on a horizontal glass from 8.15 A.M. Sept. 16th to 7.45 A.M. 17th. Figure reduced to one-third of original scale.

Phalaris Canariensis.—The sheath-like cotyledons of this monocotyledonous plant were selected for trial, because they are very sensitive to light and circumnutate well, as formerly shown (see Fig. 49, p. 63). Although we felt no doubt about the result, some seedlings were first placed before a south-west window on a moderately bright morning, and the movements of one were traced. As is so common, it moved [page 428] for the first 45 m. in a zigzag line; it then felt the full influence of the light, and travelled towards it for the next 2 h. 30 m. in an almost straight line. The tracing has not been given, as it was almost identical with that of Apios under similar circumstances (Fig. 170). By noon it had bowed itself to its full extent; it then circumnutated about the same spot and described two ellipses; by 5 P.M. it had retreated considerably from the light, through the action of apogeotropism. After some preliminary trials for ascertaining the right degree of obscurity, some seedlings were placed (Sept. 16th) before a north-east window, and light was admitted through an ordinary linen and three muslin blinds. A pencil held close by the pot now cast a very faint shadow on a white card, pointing from the window. In the evening, at 4.30 and again at 6 P.M., some of the blinds were removed. In Fig. 174 we see the course pursued under these circumstances by a rather old and not very sensitive cotyledon, 1.9 inch in height, which became much bowed, but was never rectangularly bent towards the light. From 11 A.M., when the sky became rather duller, until 6.30 P.M., the zigzagging was conspicuous, and evidently consisted of drawn-out ellipses. After 6.30 P.M. and during the night, it retreated in a crooked line from the window. Another and younger seedling moved during the same time much more quickly and to a much greater distance, in an only slightly zigzag line towards the light; by 11 A.M. it was bent almost rectangularly in this direction, and now circumnutated about the same place.

Tropaeolum majus.—Some very young seedlings, bearing only two leaves, and therefore not as yet arrived at the climbing stage of growth, were first tried before a north-east window without any blind. The epicotyls bowed themselves towards the light so rapidly that in little more than 3 h. their tips pointed rectangularly towards it. The lines traced were either nearly straight or slightly zigzag; and in this latter case we see that a trace of circumnutation was retained even under the influence of a moderately bright light. Twice whilst these epicotyls were bending towards the window, dots were made every 5 or 6 minutes, in order to detect any trace of lateral movement, but there was hardly any; and the lines formed by their junction were nearly straight, or only very slightly zigzag, as in the other parts of the figures. After the epicotyls had bowed themselves to the full extent towards the light, ellipses of considerable size were described in the usual manner. [page 429]

After having seen how the epicotyls moved towards a moderately bright light, seedlings were placed at 7.48 A.M. (Sept. 7th) before a north-east window, covered by a towel, and shortly afterwards by an ordinary linen blind, but the epicotyls still moved towards the window. At 9.13 A.M. two additional muslin blinds were suspended, so that the seedlings received very little more light from the window than from the interior of the room. The sky varied in brightness, and the seedlings occasionally

Fig. 175. Tropaeolum majus: heliotropic movement and circumnutation of the epicotyl of a young seedling towards a dull lateral light, traced on a horizontal glass from 7.48 A.M. to 10.40 P.M. Figure reduced to one-half of the original scale.

received for a short time less light from the window than from the opposite side (as ascertained by the shadow cast), and then one of the blinds was temporarily removed. In the evening the blinds were taken away, one by one. the course pursued by an epicotyl under these circumstances is shown in Fig. 175. During the whole day, until 6.45 P.M., it plainly bowed itself towards the light; and the tip moved over a considerable space. After 6.45 P.M. it moved backwards, or from the window, till [page 430] 10.40 P.M., when the last dot was made. Here, then, we have a distinct heliotropic movement, effected by means of six elongated figures (which if dots had been made every few minutes would have been more or less elliptic) directed towards the light, with the apex of each successive ellipse nearer to the window than the previous one. Now, if the light had been only a little brighter, the epicotyl would have bowed itself more to the light, as we may safely conclude from the previous trials; there would also have been less lateral movement, and the ellipses or other figures would have been drawn out into a strongly marked zigzag line, with probably one or two small loops still formed. If the light had been much brighter, we should have had a slightly zigzag line, or one quite straight, for there would have been more movement in the direction of the light, and much less from side to side.

Fig. 176. Tropaeolum majus: heliotropic movement and circumnutation of an old internode towards a lateral light, traced on a horizontal glass from 8 A.M. Nov. 2nd to 10.20 A.M. Nov. 4th. Broken lines show the nocturnal course.

Sachs states that the older internodes of this Tropaeolum are apheliotropic; we therefore placed a plant, 11 3/4 inches high, in a box, blackened within, but open on one side in front of a north-east window without any blind. A filament was fixed to the third internode from the summit on one plant, and to the fourth internode of another. These internodes were either not old enough, or the light was not sufficiently bright, to induce apheliotropism, for both plants bent slowly towards, instead of from the window during four days. The course, during two days of the first-mentioned internode, is given in Fig. 176; and we see that it either circumnutated on a small scale, or travelled in a zigzag line towards the light. We have thought this case of feeble heliotropism in one of the older internodes of a plant, [page 431] which, whilst young, is so extremely sensitive to light, worth giving.

Fig. 177. Cassia tora: heliotropic movement and circumnutation of a hypocotyl (1 inch in height) traced on a horizontal glass from 8 A.M. to 10.10 P.M. Oct. 7th. Also its circumnutation in darkness from 7 A.M. Oct. 8th to 7.45 A.M. Oct. 9th.

Cassia tora.—The cotyledons of this plant are extremely sensitive to light, whilst the hypocotyls are much less sensitive than those of most other seedlings, as we had often observed with surprise. It seemed therefore worth while to trace their movements. They were exposed to a lateral light before a north-east window, which was at first covered merely by a muslin blind, but as the sky grew brighter about 11 A.M., an additional linen blind was suspended. After 4 P.M. one blind and then the other was removed. The seedlings were protected on each side and above, but were open to the diffused light of the room in the rear. Upright filaments were fixed to the hypocotyls of two seedlings, which stood vertically in the morning. The accompanying figure (Fig. 177) shows the course pursued by one of them during two days; but it should be particularly noticed that during the second day the seedlings were kept in darkness, and they then circumnutated round nearly the same small space. On the first day (Oct. 7th) the hypocotyl moved from 8 A.M. to 12.23 P.M., toward the light in a zigzag line, then turned abruptly to the left and afterwards described a small ellipse. Another irregular [page 432] ellipse was completed between 3 P.M. and about 5.30 P.M., the hypocotyl still bending towards the light. The hypocotyl was straight and upright in the morning, but by 6 P.M. its upper half was bowed towards the light, so that the chord of the arc thus formed stood at an angle of 20o with the perpendicular. After 6 P.M. its course was reversed through the action of apogeotropism, and it continued to bend from the window during the night, as shown by the broken line. On the next day it was kept in the dark (excepting when each observation was made by the aid of a taper), and the course followed from 7 A.M. on the 8th to 7.45 A.M. on the 9th is here likewise shown. The difference between the two parts of the figure (177), namely that described during the daytime on the 7th, when exposed to a rather dim lateral light, and that on the 8th in darkness, is striking. The difference consists in the lines during the first day having been drawn out in the direction of the light. The movements of the other seedling, traced under the same circumstances, were closely similar.

Apheliotropism.—We succeeded in observing only two cases of apheliotropism, for these are somewhat rare; and the movements are generally so slow that they would have been very troublesome to trace.

Fig. 178. Bignonia capreolata: apheliotropic movement of a tendril, traced on a horizontal glass from 6.45 A.M. July 19th to 10 A.M. 20th. Movements as originally traced, little magnified, here reduced to two-thirds of the original scale.

Bignonia capreolata.—No organ of any plant, as far as we have seen, bends away so quickly from the light as do the tendrils of this Bignonia. They are also remarkable from circumnutating much less regularly than most other tendrils, often remaining stationary; they depend on apheliotropism for coming into [page 433] contact with the trunks of trees.* The stem of a young plant was tied to a stick at the base of a pair of fine tendrils, which projected almost vertically upwards; and it was placed in front of a north-east window, being protected on all other sides from the light. The first dot was made at 6.45 A.M., and by 7.35 A.M. both tendrils felt the full influence of the light, for they moved straight away from it until 9.20 A.M., when they circumnutated for a time, still moving, but only a little, from the light (see Fig. 178 of the left-hand tendril). After 3 P.M. they again moved rapidly away from the light in zigzag lines. By a late hour in the evening both had moved so far, that they pointed in a direct line from the light. During the night they returned a little in a nearly opposite direction. On the following morning they again moved from the light and converged, so that by the evening they had become interlocked, still pointing from the light. The right-hand tendril, whilst converging, zigzagged much more than the one figured. Both tracings showed that the apheliotropic movement was a modified form of circumnutation.

Cyclamen Persicum.—Whilst this plant is in flower the peduncles stand upright, but their uppermost part is hooked so that the flower itself hangs downwards. As soon as the pods begin to swell, the peduncles increase much in length and slowly curve downwards, but the short, upper, hooked part straightens itself. Ultimately the pods reach the ground, and if this is covered with moss or dead leaves, they bury themselves. We have often seen saucer-like depressions formed by the pods in damp sand or sawdust; and one pod (.3 of inch in diameter) buried itself in sawdust for three-quarters of its length.** We shall have occasion hereafter to consider the object gained by this burying process. The peduncles can change the direction of their curvature, for if a pot, with plants having their peduncles already bowed downwards, be placed horizontally, they slowly bend at right angles to their former direction towards the centre of the earth. We therefore at first attributed the movement to geotropism; but a pot which had lain horizontally with the pods

* 'The Movements and Habits of Climbing Plants,' 1875, p. 97.

** The peduncles of several other species of Cyclamen twist themselves into a spire, and according to Erasmus Darwin ('Botanic Garden,' Canto., iii. p. 126), the pods forcibly penetrate the earth. See also Grenier and Godron, 'Flore de France,' tom. ii. p. 459. [page 434]

all pointing to the ground, was reversed, being still kept horizontal, so that the pods now pointed directly upwards; it was then placed in a dark cupboard, but the pods still pointed upwards after four days and nights. The pot, in the same position, was next brought back into the light, and after two days there was some bending downwards of the peduncles, and on the fourth day two of them pointed to the centre of the earth, as did the others after an additional day or two. Another plant, in a pot which had always stood upright, was left in the dark cupboard for six days; it bore 3 peduncles, and only one became within this

Fig. 179. Cyclamen Persicum: downward apheliotropic movement of a flower-peduncle, greatly magnified (about 47 times?), traced on a horizontal glass from 1 P.M. Feb. 18th to 8 A.M. 21st.

time at all bowed downwards, and that doubtfully. The weight, therefore, of the pods is not the cause of the bending down. This pot was then brought back into the light, and after three days the peduncles were considerably bowed downwards. We are thus led to infer that the downward curvature is due to apheliotropism; though more trials ought to have been made.

In order to observe the nature of this movement, a peduncle bearing a large pod which had reached and rested on the ground, was lifted a little up and secured to a stick. A filament was fixed across the pod with a mark beneath, and its move- [page 435] ment, greatly magnified, was traced on a horizontal glass during 67 h. The plant was illuminated during the day from above. A copy of the tracing is given on p. 434 (Fig. 179); and there can be no doubt that the descending movement is one of modified circumnutation, but on an extremely small scale. The observation was repeated on another pod, which had partially buried itself in sawdust, and which was lifted up a quarter of an inch above the surface; it described three very small circles in 24 h. Considering the great length and thinness of the peduncles and the lightness of the pods, we may conclude that they would not be able to excavate saucer-like depressions in sand or sawdust, or bury themselves in moss, etc., unless they were aided by their continued rocking or circumnutating movement.]