This etext was prepared by David Price, email ccx074@coventry.ac.uk from the 1906 John Murray edition.



THE MOVEMENTS AND HABITS OF CLIMBING PLANTS



PREFACE



This Essay first appeared in the ninth volume of the 'Journal of the Linnean Society,' published in 1865. It is here reproduced in a corrected and, I hope, clearer form, with some additional facts. The illustrations were drawn by my son, George Darwin. Fritz Muller, after the publication of my paper, sent to the Linnean Society (Journal, vol. ix., p. 344) some interesting observations on the climbing plants of South Brazil, to which I shall frequently refer. Recently two important memoirs, chiefly on the difference in growth between the upper and lower sides of tendrils, and on the mechanism of the movements of twining-plants, by Dr. Hugo de Vries, have appeared in the 'Arbeiten des Botanischen Instituts in Wurzburg,' Heft. iii., 1873. These memoirs ought to be carefully studied by every one interested in the subject, as I can here give only references to the more important points. This excellent observer, as well as Professor Sachs, {1} attributes all the movements of tendrils to rapid growth along one side; but, from reasons assigned towards the close of my fourth chapter, I cannot persuade myself that this holds good with respect to those due to a touch. In order that the reader may know what points have interested me most, I may call his attention to certain tendril-bearing plants; for instance, Bignonia capreolata, Cobaea, Echinocystis, and Hanburya, which display as beautiful adaptations as can be found in any part of the kingdom of nature. It is, also, an interesting fact that intermediate states between organs fitted for widely different functions, may be observed on the same individual plant of Corydalis claviculata and the common vine; and these cases illustrate in a striking manner the principle of the gradual evolution of species.



APPENDIX TO PREFACE (1882).



Since the publication of this Edition two papers by eminent botanists have appeared; Schwendener, 'Das Winden der Pflanzen' (Monatsberichte der Berliner Akademie, Dec. 1881), and J. Sachs, 'Notiz uber Schlingpflanzen' (Arbeiten des botanischen Instituts in Wurzburg, Bd. ii. p. 719, 1882). The view "that the capacity of revolving, on which most climbers depend, is inherent, though undeveloped, in almost every plant in the vegetable kingdom" ('Climbing Plants,' p. 205), has been confirmed by the observations on circumnutation since given in 'The Power of Movement in Plants.'



ERRATA.



On pp. 28, 32, 40, 53, statements are made with reference to the supposed acceleration of the revolving movement towards the light. It appears from the observations given in 'The Power of Movement in Plants,' p. 451, that these conclusions were drawn from insufficient observations, and are erroneous.



THE MOVEMENTS AND HABITS OF CLIMBING PLANTS.



CHAPTER I.—TWINING PLANTS.



Introductory remarks—Description of the twining of the Hop—Torsion of the stems—Nature of the revolving movement, and manner of ascent- -Stems not irritable—Rate of revolution in various plants—Thickness of the support round which plants can twine—Species which revolve in an anomalous manner.

I was led to this subject by an interesting, but short paper by Professor Asa Gray on the movements of the tendrils of some Cucurbitaceous plants. {2} My observations were more than half completed before I learnt that the surprising phenomenon of the spontaneous revolutions of the stems and tendrils of climbing plants had been long ago observed by Palm and by Hugo von Mohl, {3} and had subsequently been the subject of two memoirs by Dutrochet. {4} Nevertheless, I believe that my observations, founded on the examination of above a hundred widely distinct living species, contain sufficient novelty to justify me in publishing them.

Climbing plants may be divided into four classes. First, those which twine spirally round a support, and are not aided by any other movement. Secondly, those endowed with irritable organs, which when they touch any object clasp it; such organs consisting of modified leaves, branches, or flower-peduncles. But these two classes sometimes graduate to a certain extent into one another. Plants of the third class ascend merely by the aid of hooks; and those of the fourth by rootlets; but as in neither class do the plants exhibit any special movements, they present little interest, and generally when I speak of climbing plants I refer to the two first great classes.

TWINING PLANTS.

This is the largest subdivision, and is apparently the primordial and simplest condition of the class. My observations will be best given by taking a few special cases. When the shoot of a Hop (Humulus lupulus) rises from the ground, the two or three first-formed joints or internodes are straight and remain stationary; but the next- formed, whilst very young, may be seen to bend to one side and to travel slowly round towards all points of the compass, moving, like the hands of a watch, with the sun. The movement very soon acquires its full ordinary velocity. From seven observations made during August on shoots proceeding from a plant which had been cut down, and on another plant during April, the average rate during hot weather and during the day is 2 hrs. 8 m. for each revolution; and none of the revolutions varied much from this rate. The revolving movement continues as long as the plant continues to grow; but each separate internode, as it becomes old, ceases to move.

To ascertain more precisely what amount of movement each internode underwent, I kept a potted plant, during the night and day, in a well-warmed room to which I was confined by illness. A long shoot projected beyond the upper end of the supporting stick, and was steadily revolving. I then took a longer stick and tied up the shoot, so that only a very young internode, 1.75 of an inch in length, was left free. This was so nearly upright that its revolution could not be easily observed; but it certainly moved, and the side of the internode which was at one time convex became concave, which, as we shall hereafter see, is a sure sign of the revolving movement. I will assume that it made at least one revolution during the first twenty-four hours. Early the next morning its position was marked, and it made a second revolution in 9 hrs.; during the latter part of this revolution it moved much quicker, and the third circle was performed in the evening in a little over 3 hrs. As on the succeeding morning I found that the shoot revolved in 2 hrs. 45 m., it must have made during the night four revolutions, each at the average rate of a little over 3 hrs. I should add that the temperature of the room varied only a little. The shoot had now grown 3.5 inches in length, and carried at its extremity a young internode 1 inch in length, which showed slight changes in its curvature. The next or ninth revolution was effected in 2 hrs. 30 m. From this time forward, the revolutions were easily observed. The thirty-sixth revolution was performed at the usual rate; so was the last or thirty-seventh, but it was not completed; for the internode suddenly became upright, and after moving to the centre, remained motionless. I tied a weight to its upper end, so as to bow it slightly and thus detect any movement; but there was none. Some time before the last revolution was half performed, the lower part of the internode ceased to move.

A few more remarks will complete all that need be said about this internode. It moved during five days; but the more rapid movements, after the performance of the third revolution, lasted during three days and twenty hours. The regular revolutions, from the ninth to thirty-sixth inclusive, were effected at the average rate of 2 hrs. 31 m.; but the weather was cold, and this affected the temperature of the room, especially during the night, and consequently retarded the rate of movement a little. There was only one irregular movement, which consisted in the stem rapidly making, after an unusually slow revolution, only the segment of a circle. After the seventeenth revolution the internode had grown from 1.75 to 6 inches in length, and carried an internode 1.875 inch long, which was just perceptibly moving; and this carried a very minute ultimate internode. After the twenty-first revolution, the penultimate internode was 2.5 inches long, and probably revolved in a period of about three hours. At the twenty-seventh revolution the lower and still moving internode was 8.375, the penultimate 3.5, and the ultimate 2.5 inches in length; and the inclination of the whole shoot was such, that a circle 19 inches in diameter was swept by it. When the movement ceased, the lower internode was 9 inches, and the penultimate 6 inches in length; so that, from the twenty-seventh to thirty-seventh revolutions inclusive, three internodes were at the same time revolving.

The lower internode, when it ceased revolving, became upright and rigid; but as the whole shoot was left to grow unsupported, it became after a time bent into a nearly horizontal position, the uppermost and growing internodes still revolving at the extremity, but of course no longer round the old central point of the supporting stick. From the changed position of the centre of gravity of the extremity, as it revolved, a slight and slow swaying movement was given to the long horizontally projecting shoot; and this movement I at first thought was a spontaneous one. As the shoot grew, it hung down more and more, whilst the growing and revolving extremity turned itself up more and more.

With the Hop we have seen that three internodes were at the same time revolving; and this was the case with most of the plants observed by me. With all, if in full health, two internodes revolved; so that by the time the lower one ceased to revolve, the one above was in full action, with a terminal internode just commencing to move. With Hoya carnosa, on the other hand, a depending shoot, without any developed leaves, 32 inches in length, and consisting of seven internodes (a minute terminal one, an inch in length, being counted), continually, but slowly, swayed from side to side in a semicircular course, with the extreme internodes making complete revolutions. This swaying movement was certainly due to the movement of the lower internodes, which, however, had not force sufficient to swing the whole shoot round the central supporting stick. The case of another Asclepiadaceous plant, viz., Ceropegia Gardnerii, is worth briefly giving. I allowed the top to grow out almost horizontally to the length of 31 inches; this now consisted of three long internodes, terminated by two short ones. The whole revolved in a course opposed to the sun (the reverse of that of the Hop), at rates between 5 hrs. 15 m. and 6 hrs. 45 m. for each revolution. The extreme tip thus made a circle of above 5 feet (or 62 inches) in diameter and 16 feet in circumference, travelling at the rate of 32 or 33 inches per hour. The weather being hot, the plant was allowed to stand on my study- table; and it was an interesting spectacle to watch the long shoot sweeping this grand circle, night and day, in search of some object round which to twine.

If we take hold of a growing sapling, we can of course bend it to all sides in succession, so as to make the tip describe a circle, like that performed by the summit of a spontaneously revolving plant. By this movement the sapling is not in the least twisted round its own axis. I mention this because if a black point be painted on the bark, on the side which is uppermost when the sapling is bent towards the holder's body, as the circle is described, the black point gradually turns round and sinks to the lower side, and comes up again when the circle is completed; and this gives the false appearance of twisting, which, in the case of spontaneously revolving plants, deceived me for a time. The appearance is the more deceitful because the axes of nearly all twining-plants are really twisted; and they are twisted in the same direction with the spontaneous revolving movement. To give an instance, the internode of the Hop of which the history has been recorded, was at first, as could be seen by the ridges on its surface, not in the least twisted; but when, after the 37th revolution, it had grown 9 inches long, and its revolving movement had ceased, it had become twisted three times round its own axis, in the line of the course of the sun; on the other hand, the common Convolvulus, which revolves in an opposite course to the Hop, becomes twisted in an opposite direction.

Hence it is not surprising that Hugo von Mohl (p. 105, 108, &c.) thought that the twisting of the axis caused the revolving movement; but it is not possible that the twisting of the axis of the Hop three times should have caused thirty-seven revolutions. Moreover, the revolving movement commenced in the young internode before any twisting of its axis could be detected. The internodes of a young Siphomeris and Lecontea revolved during several days, but became twisted only once round their own axes. The best evidence, however, that the twisting does not cause the revolving movement is afforded by many leaf-climbing and tendril-bearing plants (as Pisum sativum, Echinocystis lobata, Bignonia capreolata, Eccremocarpus scaber, and with the leaf-climbers, Solanum jasminoides and various species of Clematis), of which the internodes are not twisted, but which, as we shall hereafter see, regularly perform revolving movements like those of true twining-plants. Moreover, according to Palm (pp. 30, 95) and Mohl (p. 149), and Leon, {5} internodes may occasionally, and even not very rarely, be found which are twisted in an opposite direction to the other internodes on the same plant, and to the course of their revolutions; and this, according to Leon (p. 356), is the case with all the internodes of a certain variety of Phaseolus multiflorus. Internodes which have become twisted round their own axes, if they have not ceased to revolve, are still capable of twining round a support, as I have several times observed.

Mohl has remarked (p. 111) that when a stem twines round a smooth cylindrical stick, it does not become twisted. {6} Accordingly I allowed kidney-beans to run up stretched string, and up smooth rods of iron and glass, one-third of an inch in diameter, and they became twisted only in that degree which follows as a mechanical necessity from the spiral winding. The stems, on the other hand, which had ascended ordinary rough sticks were all more or less and generally much twisted. The influence of the roughness of the support in causing axial twisting was well seen in the stems which had twined up the glass rods; for these rods were fixed into split sticks below, and were secured above to cross sticks, and the stems in passing these places became much twisted. As soon as the stems which had ascended the iron rods reached the summit and became free, they also became twisted; and this apparently occurred more quickly during windy than during calm weather. Several other facts could be given, showing that the axial twisting stands in some relation to inequalities in the support, and likewise to the shoot revolving freely without any support. Many plants, which are not twiners, become in some degree twisted round their own axes; {7} but this occurs so much more generally and strongly with twining-plants than with other plants, that there must be some connexion between the capacity for twining and axial twisting. The stem probably gains rigidity by being twisted (on the same principle that a much twisted rope is stiffer than a slackly twisted one), and is thus indirectly benefited so as to be enabled to pass over inequalities in its spiral ascent, and to carry its own weight when allowed to revolve freely. {8}

I have alluded to the twisting which necessarily follows on mechanical principles from the spiral ascent of a stem, namely, one twist for each spire completed. This was well shown by painting straight lines on living stems, and then allowing them to twine; but, as I shall have to recur to this subject under Tendrils, it may be here passed over.

The revolving movement of a twining plant has been compared with that of the tip of a sapling, moved round and round by the hand held some way down the stem; but there is one important difference. The upper part of the sapling when thus moved remains straight; but with twining plants every part of the revolving shoot has its own separate and independent movement. This is easily proved; for when the lower half or two-thirds of a long revolving shoot is tied to a stick, the upper free part continues steadily revolving. Even if the whole shoot, except an inch or two of the extremity, be tied up, this part, as I have seen in the case of the Hop, Ceropegia, Convolvulus, &c., goes on revolving, but much more slowly; for the internodes, until they have grown to some little length, always move slowly. If we look to the one, two, or several internodes of a revolving shoot, they will be all seen to be more or less bowed, either during the whole or during a large part of each revolution. Now if a coloured streak be painted (this was done with a large number of twining plants) along, we will say, the convex surface, the streak will after a time (depending on the rate of revolution) be found to be running laterally along one side of the bow, then along the concave side, then laterally on the opposite side, and, lastly, again on the originally convex surface. This clearly proves that during the revolving movement the internodes become bowed in every direction. The movement is, in fact, a continuous self-bowing of the whole shoot, successively directed to all points of the compass; and has been well designated by Sachs as a revolving nutation.

As this movement is rather difficult to understand, it will be well to give an illustration. Take a sapling and bend it to the south, and paint a black line on the convex surface; let the sapling spring up and bend it to the east, and the black line will be seen to run along the lateral face fronting the north; bend it to the north, the black line will be on the concave surface; bend it to the west, the line will again be on the lateral face; and when again bent to the south, the line will be on the original convex surface. Now, instead of bending the sapling, let us suppose that the cells along its northern surface from the base to the tip were to grow much more rapidly than on the three other sides, the whole shoot would then necessarily be bowed to the south; and let the longitudinal growing surface creep round the shoot, deserting by slow degrees the northern side and encroaching on the western side, and so round by the south, by the east, again to the north. In this case the shoot would remain always bowed with the painted line appearing on the several above specified surfaces, and with the point of the shoot successively directed to each point of the compass. In fact, we should have the exact kind of movement performed by the revolving shoots of twining plants. {9}

It must not be supposed that the revolving movement is as regular as that given in the above illustration; in very many cases the tip describes an ellipse, even a very narrow ellipse. To recur once again to our illustration, if we suppose only the northern and southern surfaces of the sapling alternately to grow rapidly, the summit would describe a simple arc; if the growth first travelled a very little to the western face, and during the return a very little to the eastern face, a narrow ellipse would be described; and the sapling would be straight as it passed to and fro through the intermediate space; and a complete straightening of the shoot may often be observed in revolving plants. The movement is frequently such that three of the sides of the shoot seem to be growing in due order more rapidly than the remaining side; so that a semi-circle instead of a circle is described, the shoot becoming straight and upright during half of its course.

When a revolving shoot consists of several internodes, the lower ones bend together at the same rate, but one or two of the terminal ones bend at a slower rate; hence, though at times all the internodes are in the same direction, at other times the shoot is rendered slightly serpentine. The rate of revolution of the whole shoot, if judged by the movement of the extreme tip, is thus at times accelerated or retarded. One other point must be noticed. Authors have observed that the end of the shoot in many twining plants is completely hooked; this is very general, for instance, with the Asclepiadaceae. The hooked tip, in all the cases observed by me, viz, in Ceropegia, Sphaerostemma, Clerodendron, Wistaria, Stephania, Akebia, and Siphomeris, has exactly the same kind of movement as the other internodes; for a line painted on the convex surface first becomes lateral and then concave; but, owing to the youth of these terminal internodes, the reversal of the hook is a slower process than that of the revolving movement. {10} This strongly marked tendency in the young, terminal and flexible internodes, to bend in a greater degree or more abruptly than the other internodes, is of service to the plant; for not only does the hook thus formed sometimes serve to catch a support, but (and this seems to be much more important) it causes the extremity of the shoot to embrace the support much more closely than it could otherwise have done, and thus aids in preventing the stem from being blown away during windy weather, as I have many times observed. In Lonicera brachypoda the hook only straightens itself periodically, and never becomes reversed. I will not assert that the tips of all twining plants when hooked, either reverse themselves or become periodically straight, in the manner just described; for the hooked form may in some cases be permanent, and be due to the manner of growth of the species, as with the tips of the shoots of the common vine, and more plainly with those of Cissus discolor—plants which are not spiral twiners.

The first purpose of the spontaneous revolving movement, or, more strictly speaking, of the continuous bowing movement directed successively to all points of the compass, is, as Mohl has remarked, to favour the shoot finding a support. This is admirably effected by the revolutions carried on night and day, a wider and wider circle being swept as the shoot increases in length. This movement likewise explains how the plants twine; for when a revolving shoot meets with a support, its motion is necessarily arrested at the point of contact, but the free projecting part goes on revolving. As this continues, higher and higher points are brought into contact with the support and are arrested; and so onwards to the extremity; and thus the shoot winds round its support. When the shoot follows the sun in its revolving course, it winds round the support from right to left, the support being supposed to stand in front of the beholder; when the shoot revolves in an opposite direction, the line of winding is reversed. As each internode loses from age its power of revolving, it likewise loses its power of spirally twining. If a man swings a rope round his head, and the end hits a stick, it will coil round the stick according to the direction of the swinging movement; so it is with a twining plant, a line of growth travelling round the free part of the shoot causing it to bend towards the opposite side, and this replaces the momentum of the free end of the rope.

All the authors, except Palm and Mohl, who have discussed the spiral twining of plants, maintain that such plants have a natural tendency to grow spirally. Mohl believes (p. 112) that twining stems have a dull kind of irritability, so that they bend towards any object which they touch; but this is denied by Palm. Even before reading Mohl's interesting treatise, this view seemed to me so probable that I tested it in every way that I could, but always with a negative result. I rubbed many shoots much harder than is necessary to excite movement in any tendril or in the foot-stalk of any leaf climber, but without any effect. I then tied a light forked twig to a shoot of a Hop, a Ceropegia, Sphaerostemma, and Adhatoda, so that the fork pressed on one side alone of the shoot and revolved with it; I purposely selected some very slow revolvers, as it seemed most likely that these would profit most from possessing irritability; but in no case was any effect produced. {11} Moreover, when a shoot winds round a support, the winding movement is always slower, as we shall immediately see, than whilst it revolves freely and touches nothing. Hence I conclude that twining stems are not irritable; and indeed it is not probable that they should be so, as nature always economizes her means, and irritability would have been superfluous. Nevertheless I do not wish to assert that they are never irritable; for the growing axis of the leaf-climbing, but not spirally twining, Lophospermum scandens is, certainly irritable; but this case gives me confidence that ordinary twiners do not possess any such quality, for directly after putting a stick to the Lophopermum, I saw that it behaved differently from a true twiner or any other leaf-climber. {12}

The belief that twiners have a natural tendency to grow spirally, probably arose from their assuming a spiral form when wound round a support, and from the extremity, even whilst remaining free, sometimes assuming this form. The free internodes of vigorously growing plants, when they cease to revolve, become straight, and show no tendency to be spiral; but when a shoot has nearly ceased to grow, or when the plant is unhealthy, the extremity does occasionally become spiral. I have seen this in a remarkable manner with the ends of the shoots of the Stauntonia and of the allied Akebia, which became wound up into a close spire, just like a tendril; and this was apt to occur after some small, ill-formed leaves had perished. The explanation, I believe, is, that in such cases the lower parts of the terminal internodes very gradually and successively lose their power of movement, whilst the portions just above move onwards and in their turn become motionless; and this ends in forming an irregular spire.

When a revolving shoot strikes a stick, it winds round it rather more slowly than it revolves. For instance, a shoot of the Ceropegia, revolved in 6 hrs., but took 9 hrs. 30 m. to make one complete spire round a stick; Aristolochia gigas revolved in about 5 hrs., but took 9 hrs. 15 m. to complete its spire. This, I presume, is due to the continued disturbance of the impelling force by the arrestment of the movement at successive points; and we shall hereafter see that even shaking a plant retards the revolving movement. The terminal internodes of a long, much-inclined, revolving shoot of the Ceropegia, after they had wound round a stick, always slipped up it, so as to render the spire more open than it was at first; and this was probably in part due to the force which caused the revolutions, being now almost freed from the constraint of gravity and allowed to act freely. With the Wistaria, on the other hand, a long horizontal shoot wound itself at first into a very close spire, which remained unchanged; but subsequently, as the shoot twined spirally up its support, it made a much more open spire. With all the many plants which were allowed freely to ascend a support, the terminal internodes made at first a close spire; and this, during windy weather, served to keep the shoots in close contact with their support; but as the penultimate internodes grew in length, they pushed themselves up for a considerable space (ascertained by coloured marks on the shoot and on the support) round the stick, and the spire became more open. {13}

It follows from this latter fact that the position occupied by each leaf with respect to the support depends on the growth of the internodes after they have become spirally wound round it. I mention this on account of an observation by Palm (p. 34), who states that the opposite leaves of the Hop always stand in a row, exactly over one another, on the same side of the supporting stick, whatever its thickness may be. My sons visited a hop-field for me, and reported that though they generally found the points of insertion of the leaves standing over each other for a space of two or three feet in height, yet this never occurred up the whole length of the pole; the points of insertion forming, as might have been expected, an irregular spire. Any irregularity in the pole entirely destroyed the regularity of position of the leaves. From casual inspection, it appeared to me that the opposite leaves of Thunbergia alata were arranged in lines up the sticks round which they had twined; accordingly, I raised a dozen plants, and gave them sticks of various thicknesses, as well as string, to twine round; and in this case one alone out of the dozen had its leaves arranged in a perpendicular line: I conclude, therefore, Palm's statement is not quite accurate.

The leaves of different twining-plants are arranged on the stem (before it has twined) alternately, or oppositely, or in a spire. In the latter case the line of insertion of the leaves and the course of the revolutions coincide. This fact has been well shown by Dutrochet, {14} who found different individuals of Solanum dulcamara twining in opposite directions, and these had their leaves in each case spirally arranged in the same direction. A dense whorl of many leaves would apparently be incommodious for a twining plant, and some authors assert that none have their leaves thus arranged; but a twining Siphomeris has whorls of three leaves.

If a stick which has arrested a revolving shoot, but has not as yet been encircled, be suddenly taken away, the shoot generally springs forward, showing that it was pressing with some force against the stick. After a shoot has wound round a stick, if this be withdrawn, it retains for a time its spiral form; it then straightens itself, and again commences to revolve. The long, much-inclined shoot of the Ceropegia previously alluded to offered some curious peculiarities. The lower and older internodes, which continued to revolve, were incapable, on repeated trials, of twining round a thin stick; showing that, although the power of movement was retained, this was not sufficient to enable the plant to twine. I then moved the stick to a greater distance, so that it was struck by a point 2.5 inches from the extremity of the penultimate internode; and it was then neatly encircled by this part of the penultimate and by the ultimate internode. After leaving the spirally wound shoot for eleven hours, I quietly withdrew the stick, and in the course of the day the curled portion straightened itself and recommenced revolving; but the lower and not curled portion of the penultimate internode did not move, a sort of hinge separating the moving and the motionless part of the same internode. After a few days, however, I found that this lower part had likewise recovered its revolving power. These several facts show that the power of movement is not immediately lost in the arrested portion of a revolving shoot; and that after being temporarily lost it can be recovered. When a shoot has remained for a considerable time round a support, it permanently retains its spiral form even when the support is removed.

When a tall stick was placed so as to arrest the lower and rigid internodes of the Ceropegia, at the distance at first of 15 and then of 21 inches from the centre of revolution, the straight shoot slowly and gradually slid up the stick, so as to become more and more highly inclined, but did not pass over the summit. Then, after an interval sufficient to have allowed of a semi-revolution, the shoot suddenly bounded from the stick and fell over to the opposite side or point of the compass, and reassumed its previous slight inclination. It now recommenced revolving in its usual course, so that after a semi- revolution it again came into contact with the stick, again slid up it, and again bounded from it and fell over to the opposite side. This movement of the shoot had a very odd appearance, as if it were disgusted with its failure but was resolved to try again. We shall, I think, understand this movement by considering the former illustration of the sapling, in which the growing surface was supposed to creep round from the northern by the western to the southern face; and thence back again by the eastern to the northern face, successively bowing the sapling in all directions. Now with the Ceropegia, the stick being placed to the south of the shoot and in contact with it, as soon as the circulatory growth reached the western surface, no effect would be produced, except that the shoot would be pressed firmly against the stick. But as soon as growth on the southern surface began, the shoot would be slowly dragged with a sliding movement up the stick; and then, as soon as the eastern growth commenced, the shoot would be drawn from the stick, and its weight coinciding with the effects of the changed surface of growth, would cause it suddenly to fall to the opposite side, reassuming its previous slight inclination; and the ordinary revolving movement would then go on as before. I have described this curious case with some care, because it first led me to understand the order in which, as I then thought, the surfaces contracted; but in which, as we now know from Sachs and II. de Vries, they grow for a time rapidly, thus causing the shoot to bow towards the opposite side.

The view just given further explains, as I believe, a fact observed by Mohl (p. 135), namely, that a revolving shoot, though it will twine round an object as thin as a thread, cannot do so round a thick support. I placed some long revolving shoots of a Wistaria close to a post between 5 and 6 inches in diameter, but, though aided by me in many ways, they could not wind round it. This apparently was due to the flexure of the shoot, whilst winding round an object so gently curved as this post, not being sufficient to hold the shoot to its place when the growing surface crept round to the opposite surface of the shoot; so that it was withdrawn at each revolution from its support.

When a free shoot has grown far beyond its support, it sinks downwards from its weight, as already explained in the case of the Hop, with the revolving extremity turned upwards. If the support be not lofty, the shoot falls to the ground, and resting there, the extremity rises up. Sometimes several shoots, when flexible, twine together into a cable, and thus support one another. Single thin depending shoots, such as those of the Sollya Drummondii, will turn abruptly backwards and wind up on themselves. The greater number of the depending shoots, however, of one twining plant, the Hibbertia dentata, showed but little tendency to turn upwards. In other cases, as with the Cryptostegia grandiflora, several internodes which were at first flexible and revolved, if they did not succeed in twining round a support, become quite rigid, and supporting themselves upright, carried on their summits the younger revolving internodes.

Here will be a convenient place to give a Table showing the direction and rate of movement of several twining plants, with a few appended remarks. These plants are arranged according to Lindley's 'Vegetable Kingdom' of 1853; and they have been selected from all parts of the series so as to show that all kinds behave in a nearly uniform manner. {15}

The Rate of Revolution of various Twining Plants.

(ACOTYLEDONS.)

Lygodium scandens (Polypodiaceae) moves against the sun.

H. M. June 18, 1st circle was made in 6 0 18, 2nd 6 15 (late in evening) 19, 3rd 5 32 (very hot day) 19, 4th 5 0 (very hot day) 20, 5th 6 0

Lygodium articulatum moves against the sun.

H. M. July 19, 1st circle was made in 16 30 (shoot very young) 20, 2nd 15 0 21, 3rd 8 0 22, 4th 10 30

(MONOCOTYLEDONS.)

Ruscus androgynus (Liliaceae), placed in the hot-house, moves against the sun.

H. M. May 24, 1st circle was made in 6 14 (shoot very young) 25, 2nd 2 21 25, 3rd 3 37 25, 4th 3 22 26, 5th 2 50 27, 6th 3 52 27, 7th 4 11

Asparagus (unnamed species from Kew) (Liliaceae) moves against the sun, placed in hothouse.

H. M. Dec. 26, 1st circle was made in 5 0 27, 2nd 5 40

Tamus communis (Dioscoreaceae). A young shoot from a tuber in a pot placed in the greenhouse: follows the sun.

H. M. July, 7, 1st circle was made in 3 10 7, 2nd 2 38 8, 3rd 3 5 8, 4th 2 56 8, 5th 2 30 8, 6th 2 30

Lapagerea rosea (Philesiaceae), in greenhouse, follows the sun.

H. M. March 9, 1st circle was made in 26 15 (shoot young) 10, semicircle 8 15 11, 2nd circle 11 0 12, 3rd 15 30 13, 4th 14 15 16, 5th 8 40 when placed in the hothouse; but the next day the shoot remained stationary.

Roxburghia viridiflora (Roxburghiaceae) moves against the sun; it completed a circle in about 24 hours.

(DICOTYLEDONS.)

Humulus Lupulus (Urticaceae) follows the sun. The plant was kept in a room during warm weather.

H. M. April 9, 2 circles were made in 4 16 Aug. 13, 3rd circle was 2 0 14, 4th 2 20 14, 5th 2 16 14, 6th 2 2 14, 7th 2 0 14, 8th 2 4

With the Hop a semicircle was performed, in travelling from the light, in 1 hr. 33 m.; in travelling to the light, in 1 hr. 13 m.; difference of rate, 20 m.

Akebia quinata (Lardizabalaceae), placed in hothouse, moves against the sun.

H. M. March 17, 1st circle was made in 4 0 (shoot young) 18, 2nd 1 40 18, 3rd 1 30 19, 4th 1 45

Stauntonia latifolia (Lardizabalaceae), placed in hothouse, moves against the sun.

H. M. March 28, 1st circle was made in 3 30 29, 2nd 3 45

Sphaerostemma marmoratum (Schizandraceae) follows the sun.

H. M. August 5th, 1st circle was made in about 24 0 5th, 2nd circle was made in 18 30

Stephania rotunda (Menispermaceae) moves against the sun

H. M. May 27, 1st circle was made in 5 5 30, 2nd 7 6 June 2, 3rd 5 15 3, 4th 6 28

Thryallis brachystachys (Malpighiaceae) moves against the sun: one shoot made a circle in 12 hrs., and another in 10 hrs. 30 m.; but the next day, which was much colder, the first shoot took 10 hrs. to perform only a semicircle.

Hibbertia dentata (Dilleniaceae), placed in the hothouse, followed the sun, and made (May 18th) a circle in 7 hrs. 20 m.; on the 19th, reversed its course, and moved against the sun, and made a circle in 7 hrs.; on the 20th, moved against the sun one-third of a circle, and then stood still; on the 26th, followed the sun for two-thirds of a circle, and then returned to its starting-point, taking for this double course 11 hrs. 46 m.

Sollya Drummondii (Pittosporaceae) moves against the sun kept in greenhouse.

H. M. April 4, 1st circle was made in 4 25 5, 2nd 8 0 (very cold day) 6, 3rd 6 25 7, 4th 7 5

Polygonum dumetorum (Polygonaceae). This case is taken from Dutrochet (p. 299), as I observed, no allied plant: follows the sun. Three shoots, cut off a plant, and placed in water made circles in 3 hrs. 10 m., 5 hrs. 20 m., and 7 hrs. 15 m.

Wistaria Chinensis (Leguminosae), in greenhouse, moves against the sun.

H. M. May 13, 1st circle was made in 3 5 13, 2nd 3 20 16, 3rd 2 5 24, 4th 3 21 25, 5th 2 37 25, 6th 2 35

Phaseolus vulgaris (Leguminosae), in greenhouse, moves against the sun.

H. M. May, 1st circle was made in 2 0 2nd 1 55 3rd 1 55

Dipladenia urophylla (Apocynaceae) moves against the sun.

H. M. April 18, 1st circle was made in 8 0 19, 2nd 9 15 30, 3rd 9 40

Dipladenia crassinoda moves against the sun.

H. M. May 16, 1st circle was made in 9 5 July 20, 2nd 8 0 21, 3rd 8 5

Ceropegia Gardnerii (Asclepiadaceae) moves against the sun.

H. M. Shoot very young, 2 inches } in length } 1st circle was performed in 7 55 Shoot still young 2nd 7 0 Long shoot 3rd 6 33 Long shoot 4th 5 15 Long shoot 5th 6 45

Stephanotis floribunda (Asclepiadaceae) moves against the sun and made a circle in 6 hrs. 40 m., a second circle in about 9 hrs.

Hoya carnosa (Asclepiadaceae) made several circles in from 16 hrs. to 22 hrs. or 24 hrs.

Ipomaea purpurea (Convolvulaceae) moves against the sun. Plant placed in room with lateral light.

{Semicircle, from the light in 1st circle was made in 2 hrs. 42 m. { 1 hr. 14 m., to the light { 1 hr. 28 m.: difference 14 m.

{Semicircle, from the light in 2nd circle was made in 2 hrs. 47 m. { 1 hr. 17 m., to the light 1 hr. { 30 m.: difference 13 m.

Ipomaea jucunda (Convolvulaceae) moves against the sun, placed in my study, with windows facing the north-east. Weather hot.

{Semicircle, from the light in 1st circle was made in 5 hrs. 30 m. { 4 hrs. 30 m., to the light 1 hr. { 0 m.: difference 3 hrs. 30 m.

2nd circle was made in 5 hrs. {Semicircle, from the light in 20 m. (Late in afternoon: { 3 hrs. 50 m., to the light 1 hr. circle completed at 6 hrs. 40 m. { 30 m.: difference 2 hrs. 20 m. P.M.)

We have here a remarkable instance of the power of light in retarding and hastening the revolving movement. (See ERRATA.)

Convolvulus sepium (large-flowered cultivated var.) moves against the sun. Two circles, were made each in 1 hr. 42 m.: difference in semicircle from and to the light 14 m.

Rivea tiliaefolia (Convolvulaceae) moves against the sun, made four revolutions in 9 hrs.; so that, on an average, each was performed in 2 hrs. 15 m.

Plumbago rosea (Plumbaginaceae) follows the sun. The shoot did not begin to revolve until nearly a yard in height; it then made a fine circle in 10 hrs. 45 m. During the next few days it continued to move, but irregularly. On August 15th the shoot followed, during a period of 10 hrs. 40 m., a long and deeply zigzag course and then made a broad ellipse. The figure apparently represented three ellipses, each of which averaged 3 hrs. 38 m. for its completion.

Jasminum pauciflorum, Bentham (Jasminaceae), moves against the sun. A circle was made in 7 hrs. 15 m., and a second rather more quickly.

Clerodendrum Thomsonii (Verbenaceae) follows the sun.

H. M. April 12, 1st circle was made in 5 45 (shoot very young) 14, 2nd 3 30 {(directly after the 18, a semicircle 5 0 { plant was shaken { on being moved) 19, 3rd circle 3 0 20, 4th 4 20

Tecoma jasminoides (Bignoniaceae) moves against the sun.

H. M. March 17, 1st circle was made in 6 30 19, 2nd 7 0 22, 3rd 8 30 (very cold day) 24, 4th 6 45

Thunbergia alata (Acanthaceae) moves against sun.

H. M. April 14, 1st circle was made in 3 20 18, 2nd 2 50 18, 3rd 2 55 18, 4th 3 55 (late in afternoon)

Adhadota cydonaefolia (Acanthaceae) follows the sun. A young shoot made a semicircle in 24 hrs.; subsequently it made a circle in between 40 hrs. and 48 hrs. Another shoot, however, made a circle in 26 hrs. 30 m.

Mikania scandens (Compositae) moves against the sun.

H. M. March 14, 1st circle was made in 3 10 15, 2nd 3 0 16, 3rd 3 0 17, 4th 3 33 April 7, 5th 2 50 7, 6th 2 40 {This circle was made { after a copious water- { ing with cold water at { 47 degrees Fahr.

Combretum argenteum (Combretaceae) moves against the sun. Kept in hothouse.

H. M. {Early in morning, when Jan. 24, 1st circle was made in 2 55 { the temperature of the { house had fallen a { little.

24, 2 circles each at an } average of } 2 20 25, 4th circle was made in 2 25

Combretum purpureum revolves not quite so quickly as C. argenteum.

Loasa aurantiaca (Loasaceae). Revolutions variable in their course: a plant which moved against the sun.

H. M. June 20, 1st circle was made in 2 37 20, 2nd 2 13 20, 3rd 4 0 21, 4th 2 35 22, 5th 3 26 23, 6th 3 5

Another plant which followed the sun in its revolutions.

H. M. July 11, 1st circle was made in 1 51 } 11, 2nd 1 46 } Very hot day. 11, 3rd 1 41 } 11, 4th 1 48 } 12, 5th 2 35 }

Scyphanthus elegans (Loasaceae) follows the sun.

H. M. June 13, 1st circle was made in 1 45 13, 2nd 1 17 14, 3rd 1 36 14, 4th 1 59 14, 5th 2 3

Siphomeris or Lecontea (unnamed sp.) (Cinchonaceae) follows the sun.

H. M. {(shoot extremely May 25, semicircle was made in 10 27 { young) 26, 1st circle 10 15 (shoot still young) 30, 2nd 8 55 June 2, 3rd 8 11 6, 4th 6 8 { Taken from the 8, 5th 7 20 { hothouse, and 9, 6th 8 36 { placed in a room { in my house.

Manettia bicolor (Cinchonaceae), young plant, follows the sun.

H. M. July 7, 1st circle was made in 6 18 8, 2nd 6 53 9, 3rd 6 30

Lonicera brachypoda (Caprifoliaceae) follows the sun, kept in a warm room in the house.

H. M. April, 1st circle was made in 9 10 (about) {(a distinct shoot, very April, 2nd circle was made in 12 20 { young, on same plant) 3rd 7 30 {In this latter circle, { the semicircle from { the light took 5 hrs. 4th 8 0 { 23 m., and to the { light 2 hrs. 37 min.: { difference 2 hrs 46m.

Aristolochia gigas (Aristolochiaceae) moves against the sun.

H. M. July 22, 1st circle was made in 8 0 (rather young shoot) 23, 2nd 7 15 24, 3rd 5 0 (about)

In the foregoing Table, which includes twining plants belonging to widely different orders, we see that the rate at which growth travels or circulates round the axis (on which the revolving movement depends), differs much. As long as a plant remains under the same conditions, the rate is often remarkably uniform, as with the Hop, Mikania, Phaseolus, &c. The Scyphanthus made one revolution in 1 hr. 17 m., and this is the quickest rate observed by me; but we shall hereafter see a tendril-bearing Passiflora revolving more rapidly. A shoot of the Akebia quinata made a revolution in 1 hr. 30 m., and three revolutions at the average rate of 1 hr. 38 m.; a Convolvulus made two revolutions at the average of 1 hr. 42 m., and Phaseolus vulgaris three at the average of 1 hr. 57 m. On the other hand, some plants take 24 hrs. for a single revolution, and the Adhadota sometimes required 48 hrs.; yet this latter plant is an efficient twiner. Species of the same genus move at different rates. The rate does not seem governed by the thickness of the shoots: those of the Sollya are as thin and flexible as string, but move more slowly than the thick and fleshy shoots of the Ruscus, which seem little fitted for movement of any kind. The shoots of the Wistaria, which become woody, move faster than those of the herbaceous Ipomoea or Thunbergia.

We know that the internodes, whilst still very young, do not acquire their proper rate of movement; hence the several shoots on the same plant may sometimes be seen revolving at different rates. The two or three, or even more, internodes which are first formed above the cotyledons, or above the root-stock of a perennial plant, do not move; they can support themselves, and nothing superfluous is granted.

A greater number of twiners revolve in a course opposed to that of the sun, or to the hands of a watch, than in the reversed course, and, consequently, the majority, as is well known, ascend their supports from left to right. Occasionally, though rarely, plants of the same order twine in opposite directions, of which Mohl (p. 125) gives a case in the Leguminosae, and we have in the table another in the Acanthaceae. I have seen no instance of two species of the same genus twining in opposite directions, and such cases must be rare; but Fritz Muller {16} states that although Mikania scandens twines, as I have described, from left to right, another species in South Brazil twines in an opposite direction. It would have been an anomalous circumstance if no such cases had occurred, for different individuals of the same species, namely, of Solanum dulcamara (Dutrochet, tom. xix. p. 299), revolve and twine in two directions: this plant, however; is a most feeble twiner. Loasa aurantiaca (Leon, p. 351) offers a much more curious case: I raised seventeen plants: of these eight revolved in opposition to the sun and ascended from left to right; five followed the sun and ascended from right to left; and four revolved and twined first in one direction, and then reversed their course, {17} the petioles of the opposite leaves affording a point d'appui for the reversal of the spire. One of these four plants made seven spiral turns from right to left, and five turns from left to right. Another plant in the same family, the Scyphanthus elegans, habitually twines in this same manner. I raised many plants of it, and the stems of all took one turn, or occasionally two or even three turns in one direction, and then, ascending for a short space straight, reversed their course and took one or two turns in an opposite direction. The reversal of the curvature occurred at any point in the stem, even in the middle of an internode. Had I not seen this case, I should have thought its occurrence most improbable. It would be hardly possible with any plant which ascended above a few feet in height, or which lived in an exposed situation; for the stem could be pulled away easily from its support, with but little unwinding; nor could it have adhered at all, had not the internodes soon become moderately rigid. With leaf- climbers, as we shall soon see, analogous cases frequently occur; but these present no difficulty, as the stem is secured by the clasping petioles.

In the many other revolving and twining plants observed by me, I never but twice saw the movement reversed; once, and only for a short space, in Ipomoea jucunda; but frequently with Hibbertia dentata. This plant at first perplexed me much, for I continually observed its long and flexible shoots, evidently well fitted for twining, make a whole, or half, or quarter circle in one direction and then in an opposite direction; consequently, when I placed the shoots near thin or thick sticks, or perpendicularly stretched string, they seemed as if constantly trying to ascend, but always failed. I then surrounded the plant with a mass of branched twigs; the shoots ascended, and passed through them, but several came out laterally, and their depending extremities seldom turned upwards as is usual with twining plants. Finally, I surrounded a second plant with many thin upright sticks, and placed it near the first one with twigs; and now both had got what they liked, for they twined up the parallel sticks, sometimes winding round one and sometimes round several; and the shoots travelled laterally from one to the other pot; but as the plants grew older, some of the shoots twined regularly up thin upright sticks. Though the revolving movement was sometimes in one direction and sometimes in the other, the twining was invariably from left to right; {18} so that the more potent or persistent movement of revolution must have been in opposition to the course of the sun. It would appear that this Hibbertia is adapted both to ascend by twining, and to ramble laterally through the thick Australian scrub.

I have described the above case in some detail, because, as far as I have seen, it is rare to find any special adaptations with twining plants, in which respect they differ much from the more highly organized tendril-bearers. The Solanum dulcamara, as we shall presently see, can twine only round stems which are both thin and flexible. Most twining plants are adapted to ascend supports of moderate though of different thicknesses. Our English twiners, as far as I have seen, never twine round trees, excepting the honeysuckle (Lonicera periclymenum), which I have observed twining up a young beech-tree nearly 4.5 inches in diameter. Mohl (p. 134) found that the Phaseolus multiflorus and Ipomoea purpurea could not, when placed in a room with the light entering on one side, twine round sticks between 3 and 4 inches in diameter; for this interfered, in a manner presently to be explained, with the revolving movement. In the open air, however, the Phaseolus twined round a support of the above thickness, but failed in twining round one 9 inches in diameter. Nevertheless, some twiners of the warmer temperate regions can manage this latter degree of thickness; for I hear from Dr. Hooker that at Kew the Ruscus androgynus has ascended a column 9 inches in diameter; and although a Wistaria grown by me in a small pot tried in vain for weeks to get round a post between 5 and 6 inches in thickness, yet at Kew a plant ascended a trunk above 6 inches in diameter. The tropical twiners, on the other hand, can ascend thicker trees; I hear from Drs. Thomson and Hooker that this is the case with the Butea parviflora, one of the Menispermaceae, and with some Dalbergias and other Leguminosae. {19} This power would be necessary for any species which had to ascend by twining the large trees of a tropical forest; otherwise they would hardly ever be able to reach the light. In our temperate countries it would be injurious to the twining plants which die down every year if they were enabled to twine round trunks of trees, for they could not grow tall enough in a single season to reach the summit and gain the light.

By what means certain twining plants are adapted to ascend only thin stems, whilst others can twine round thicker ones, I do not know. It appeared to me probable that twining plants with very long revolving shoots would be able to ascend thick supports; accordingly I placed Ceropegia Gardnerii near a post 6 inches in diameter, but the shoots entirely failed to wind round it; their great length and power of movement merely aid them in finding a distant stem round which to twine. The Sphaerostemma marmoratum is a vigorous tropical twiner; and as it is a very slow revolver, I thought that this latter circumstance might help it in ascending a thick support; but though it was able to wind round a 6-inch post, it could do this only on the same level or plane, and did not form a spire and thus ascend.

As ferns differ so much in structure from phanerogamic plants, it may be worth while here to show that twining ferns do not differ in their habits from other twining plants. In Lygodium articulatum the two internodes of the stem (properly the rachis) which are first formed above the root-stock do not move; the third from the ground revolves, but at first very slowly. This species is a slow revolver: but L. scandens made five revolutions, each at the average rate of 5 hrs. 45 m.; and this represents fairly well the usual rate, taking quick and slow movers, amongst phanerogamic plants. The rate was accelerated by increased temperature. At each stage of growth only the two upper internodes revolved. A line painted along the convex surface of a revolving internode becomes first lateral, then concave, then lateral and ultimately again convex. Neither the internodes nor the petioles are irritable when rubbed. The movement is in the usual direction, namely, in opposition to the course of the sun; and when the stem twines round a thin stick, it becomes twisted on its own axis in the same direction. After the young internodes have twined round a stick, their continued growth causes them to slip a little upwards. If the stick be soon removed, they straighten themselves, and recommence revolving. The extremities of the depending shoots turn upwards, and twine on themselves. In all these respects we have complete identity with twining phanerogamic plants; and the above enumeration may serve as a summary of the leading characteristics of all twining plants.

The power of revolving depends on the general health and vigour of the plant, as has been laboriously shown by Palm. But the movement of each separate internode is so independent of the others, that cutting off an upper one does not affect the revolutions of a lower one. When, however, Dutrochet cut off two whole shoots of the Hop, and placed them in water, the movement was greatly retarded; for one revolved in 20 hrs. and the other in 23 hrs., whereas they ought to have revolved in between 2 hrs. and 2 hrs. 30 m. Shoots of the Kidney-bean, cut off and placed in water, were similarly retarded, but in a less degree. I have repeatedly observed that carrying a plant from the greenhouse to my room, or from one part to another of the greenhouse, always stopped the movement for a time; hence I conclude that plants in a state of nature and growing in exposed situations, would not make their revolutions during very stormy weather. A decrease in temperature always caused a considerable retardation in the rate of revolution; but Dutrochet (tom. xvii. pp. 994, 996) has given such precise observations on this head with respect to the common pea that I need say nothing more. When twining plants are placed near a window in a room, the light in some cases has a remarkable power (as was likewise observed by Dutrochet, p. 998, with the pea) on the revolving movement, but this differs in degree with different plants; thus Ipomoea jucunda made a complete circle in 5 hrs. 30 m.; the semicircle from the light taking 4 hrs. 80 m., and that towards the light only 1 hr. Lonicera brachypoda revolved, in a reversed direction to the Ipomoea, in 8 hrs.; the semicircle from the light taking 5 hrs. 23 m., and that to the light only 2 hrs. 37 m. From the rate of revolution in all the plants observed by me, being nearly the same during the night and the day, I infer that the action of the light is confined to retarding one semicircle and accelerating the other, so as not to modify greatly the rate of the whole revolution. This action of the light is remarkable, when we reflect how little the leaves are developed on the young and thin revolving internodes. It is all the more remarkable, as botanists believe (Mohl, p. 119) that twining plants are but little sensitive to the action of light.

I will conclude my account of twining plants by giving a few miscellaneous and curious cases. With most twining plants all the branches, however many there may be, go on revolving together; but, according to Mohl (p. 4), only the lateral branches of Tamus elephantipes twine, and not the main stem. On the other hand, with a climbing species of Asparagus, the leading shoot alone, and not the branches, revolved and twined; but it should be stated that the plant was not growing vigorously. My plants of Combretum argenteum and C. purpureum made numerous short healthy shoots; but they showed no signs of revolving, and I could not conceive how these plants could be climbers; but at last C. argenteum put forth from the lower part of one of its main branches a thin shoot, 5 or 6 feet in length, differing greatly in appearance from the previous shoots, owing to its leaves being little developed, and this shoot revolved vigorously and twined. So that this plant produces shoots of two kinds. With Periploca Graeca (Palm, p. 43) the uppermost shoots alone twine. Polygonum convolvulus twines only during the middle of the summer (Palm, p. 43, 94); and plants growing vigorously in the autumn show no inclination to climb. The majority of Asclepiadaceae are twiners; but Asclepias nigra only "in fertiliori solo incipit scandere subvolubili caule" (Willdenow, quoted and confirmed by Palm, p. 41). Asclepias vincetoxicum does not regularly twine, but occasionally does so (Palm, p. 42; Mohl, p. 112) when growing under certain conditions. So it is with two species of Ceropegia, as I hear from Prof. Harvey, for these plants in their native dry South African home generally grow erect, from 6 inches to 2 feet in height,—a very few taller specimens showing some inclination to curve; but when cultivated near Dublin, they regularly twined up sticks 5 or 6 feet in height. Most Convolvulaceae are excellent twiners; but in South Africa Ipomoea argyraeoides almost always grows erect and compact, from about 12 to 18 inches in height, one specimen alone in Prof. Harvey's collection showing an evident disposition to twine. On the other hand, seedlings raised near Dublin twined up sticks above 8 feet in height. These facts are remarkable; for there can hardly be a doubt that in the dryer provinces of South Africa these plants have propagated themselves for thousands of generations in an erect condition; and yet they have retained during this whole period the innate power of spontaneously revolving and twining, whenever their shoots become elongated under proper conditions of life. Most of the species of Phaseolus are twiners; but certain varieties of the P. multiflorus produce (Leon, p. 681) two kinds of shoots, some upright and thick, and others thin and twining. I have seen striking instances of this curious case of variability in "Fulmer's dwarf forcing-bean," which occasionally produced a single long twining shoot.

Solanum dulcamara is one of the feeblest and poorest of twiners: it may often be seen growing as an upright bush, and when growing in the midst of a thicket merely scrambles up between the branches without twining; but when, according to Dutrochet (tom. xix. p. 299), it grows near a thin and flexible support, such as the stem of a nettle, it twines round it. I placed sticks round several plants, and vertically stretched strings close to others, and the strings alone were ascended by twining. The stem twines indifferently to the right or left. Some others species of Solanum, and of another genus, viz. Habrothamnus, belonging to the same family, are described in horticultural works as twining plants, but they seem to possess this faculty in a very feeble degree. We may suspect that the species of these two genera have as yet only partially acquired the habit of twining. On the other hand with Tecoma radicans, a member of a family abounding with twiners and tendril-bearers, but which climbs, like the ivy, by the aid of rootlets, we may suspect that a former habit of twining has been lost, for the stem exhibited slight irregular movements which could hardly be accounted for by changes in the action of the light. There is no difficulty in understanding how a spirally twining plant could graduate into a simple root-climber; for the young internodes of Bignonia Tweedyana and of Hoya carnosa revolve and twine, but likewise emit rootlets which adhere to any fitting surface, so that the loss of twining would be no great disadvantage and in some respects an advantage to these species, as they would then ascend their supports in a more direct line. {20}



CHAPTER II.—LEAF-CLIMBERS.



Plants which climb by the aid of spontaneously revolving and sensitive petioles—Clematis—Tropaeolum—Maurandia, flower-peduncles moving spontaneously and sensitive to a touch—Rhodochiton— Lophospermum—internodes sensitive—Solanum, thickening of the clasped petioles—Fumaria—Adlumia—Plants which climb by the aid of their produced midribs—Gloriosa—Flagellaria—Nepenthes—Summary on leaf-climbers.

We now come to our second class of climbing plants, namely, those which ascend by the aid of irritable or sensitive organs. For convenience' sake the plants in this class have been grouped under two sub-divisions, namely, leaf-climbers, or those which retain their leaves in a functional condition, and tendril-bearers. But these sub-divisions graduate into each other, as we shall see under Corydalis and the Gloriosa lily.

It has long been observed that several plants climb by the aid of their leaves, either by their petioles (foot-stalks) or by their produced midribs; but beyond this simple fact they have not been described. Palm and Mohl class these plants with those which bear tendrils; but as a leaf is generally a defined object, the present classification, though artificial, has at least some advantages. Leaf-climbers are, moreover, intermediate in many respects between twiners and tendril-bearers. Eight species of Clematis and seven of Tropaeolum were observed, in order to see what amount of difference in the manner of climbing existed within the same genus; and the differences are considerable.

CLEMATIS.—C. glandulosa.—The thin upper internodes revolve, moving against the course of the sun, precisely like those of a true twiner, at an average rate, judging from three revolutions, of 3 hrs. 48 m. The leading shoot immediately twined round a stick placed near it; but, after making an open spire of only one turn and a half, it ascended for a short space straight, and then reversed its course and wound two turns in an opposite direction. This was rendered possible by the straight piece between the opposed spires having become rigid. The simple, broad, ovate leaves of this tropical species, with their short thick petioles, seem but ill-fitted for any movement; and whilst twining up a vertical stick, no use is made of them. Nevertheless, if the footstalk of a young leaf be rubbed with a thin twig a few times on any side, it will in the course of a few hours bend to that side; afterwards becoming straight again. The under side seemed to be the most sensitive; but the sensitiveness or irritability is slight compared to that which we shall meet with in some of the following species; thus, a loop of string, weighing 1.64 grain (106.2 mg.) and hanging for some days on a young footstalk, produced a scarcely perceptible effect. A sketch is here given of two young leaves which had naturally caught hold of two thin branches. A forked twig placed so as to press lightly on the under side of a young footstalk caused it, in 12 hrs., to bend greatly, and ultimately to such an extent that the leaf passed to the opposite side of the stem; the forked stick having been removed, the leaf slowly recovered its former position.

The young leaves spontaneously and gradually change their position: when first developed the petioles are upturned and parallel to the stem; they then slowly bend downwards, remaining for a short time at right angles to the stem, and then become so much arched downwards that the blade of the leaf points to the ground with its tip curled inwards, so that the whole petiole and leaf together form a hook. They are thus enabled to catch hold of any twig with which they may be brought into contact by the revolving movement of the internodes. If this does not happen, they retain their hooked shape for a considerable time, and then bending upwards reassume their original upturned position, which is preserved ever afterwards. The petioles which have clasped any object soon become much thickened and strengthened, as may be seen in the drawing.

Clematis montana.—The long, thin petioles of the leaves, whilst young, are sensitive, and when lightly rubbed bend to the rubbed side, subsequently becoming straight. They are far more sensitive than the petioles of C. glandulosa; for a loop of thread weighing a quarter of a grain (16.2 mg.) caused them to bend; a loop weighing only one-eighth of a grain (8.1 mg.) sometimes acted and sometimes did not act. The sensitiveness extends from the blade of the leaf to the stem. I may here state that I ascertained in all cases the weights of the string and thread used by carefully weighing 50 inches in a chemical balance, and then cutting off measured lengths. The main petiole carries three leaflets; but their short, sub-petioles are not sensitive. A young, inclined shoot (the plant being in the greenhouse) made a large circle opposed to the course of the sun in 4 hrs. 20 m., but the next day, being very cold, the time was 5 hrs. 10 m. A stick placed near a revolving stem was soon struck by the petioles which stand out at right angles, and the revolving movement was thus arrested. The petioles then began, being excited by the contact, to slowly wind round the stick. When the stick was thin, a petiole sometimes wound twice round it. The opposite leaf was in no way affected. The attitude assumed by the stem after the petiole had clasped the stick, was that of a man standing by a column, who throws his arm horizontally round it. With respect to the stem's power of twining, some remarks will be made under C. calycina.

Clematis Sieboldi.—A shoot made three revolutions against the sun at an average rate of 3 hrs. 11 m. The power of twining is like that of the last species. Its leaves are nearly similar in structure and in function, excepting that the sub-petioles of the lateral and terminal leaflets are sensitive. A loop of thread, weighing one-eighth of a grain, acted on the main petiole, but not until two or three days had elapsed. The leaves have the remarkable habit of spontaneously revolving, generally in vertical ellipses, in the same manner, but in a less degree, as will be described under C. microphylla.

Clematis calycina.—The young shoots are thin and flexible: one revolved, describing a broad oval, in 5 hrs. 30 m., and another in 6 hrs. 12 m. They followed the course of the sun; but the course, if observed long enough, would probably be found to vary in this species, as well as in all the others of the genus. It is a rather better twiner than the two last species: the stem sometimes made two spiral turns round a thin stick, if free from twigs; it then ran straight up for a space, and reversing its course took one or two turns in an opposite direction. This reversal of the spire occurred in all the foregoing species. The leaves are so small compared with those of most of the other species, that the petioles at first seem ill-adapted for clasping. Nevertheless, the main service of the revolving movement is to bring them into contact with surrounding objects, which are slowly but securely seized. The young petioles, which alone are sensitive, have their ends bowed a little downwards, so as to be in a slight degree hooked; ultimately the whole leaf, if it catches nothing, becomes level. I gently rubbed with a thin twig the lower surfaces of two young petioles; and in 2 hrs. 30 m. they were slightly curved downwards; in 5 hrs., after being rubbed, the end of one was bent completely back, parallel to the basal portion; in 4 hrs. subsequently it became nearly straight again. To show how sensitive the young petioles are, I may mention that I just touched the under sides of two with a little water-colour, which when dry formed an excessively thin and minute crust; but this sufficed in 24 hrs. to cause both to bend downwards. Whilst the plant is young, each leaf consists of three divided leaflets, which barely have distinct petioles, and these are not sensitive; but when the plant is well grown, the petioles of the two lateral and terminal leaflets are of considerable length, and become sensitive so as to be capable of clasping an object in any direction.

When a petiole has clasped a twig, it undergoes some remarkable changes, which may be observed with the other species, but in a less strongly marked manner, and will here be described once for all. The clasped petiole in the course of two or three days swells greatly, and ultimately becomes nearly twice as thick as the opposite one which has clasped nothing. When thin transverse slices of the two are placed under the microscope their difference is conspicuous: the side of the petiole which has been in contact with the support, is formed of a layer of colourless cells with their longer axes directed from the centre, and these are very much larger than the corresponding cells in the opposite or unchanged petiole; the central cells, also, are in some degree enlarged, and the whole is much indurated. The exterior surface generally becomes bright red. But a far greater change takes place in the nature of the tissues than that which is visible: the petiole of the unclasped leaf is flexible and can be snapped easily, whereas the clasped one acquires an extraordinary degree of toughness and rigidity, so that considerable force is required to pull it into pieces. With this change, great durability is probably acquired; at least this is the case with the clasped petioles of Clematis vitalba. The meaning of these changes is obvious, namely, that the petioles may firmly and durably support the stem.

Clematis microphylla, var. leptophylla.—The long and thin internodes of this Australian species revolve sometimes in one direction and sometimes in an opposite one, describing long, narrow, irregular ellipses or large circles. Four revolutions were completed within five minutes of the same average rate of 1 hr. 51 m.; so that this species moves more quickly than the others of the genus. The shoots, when placed near a vertical stick, either twine round it, or clasp it with the basal portions of their petioles. The leaves whilst young are nearly of the same shape as those of C. viticella, and act in the same manner like a hook, as will be described under that species. But the leaflets are more divided, and each segment whilst young terminates in a hardish point, which is much curved downwards and inwards; so that the whole leaf readily catches hold of any neighbouring object. The petioles of the young terminal leaflets are acted on by loops of thread weighing 0.125th and even 0.0625th of a grain. The basal portion of the main petiole is much less sensitive, but will clasp a stick against which it presses.

The leaves, whilst young, are continually and spontaneously moving slowly. A bell-glass was placed over a shoot secured to a stick, and the movements of the leaves were traced on it during several days. A very irregular line was generally formed; but one day, in the course of eight hours and three quarters, the figure clearly represented three and a half irregular ellipses, the most perfect one of which was completed in 2 hrs. 35 m. The two opposite leaves moved independently of each other. This movement of the leaves would aid that of the internodes in bringing the petioles into contact with surrounding objects. I discovered this movement too late to be enabled to observe it in the other species; but from analogy I can hardly doubt that the leaves of at least C. viticella, C. flammula, and C. vitalba move spontaneously; and, judging from C Sieboldi, this probably is the case with C. montana and C. calycina. I ascertained that the simple leaves of C. glandulosa exhibited no spontaneous revolving movement.

Clematis viticella, var. venosa.—In this and the two following species the power of spirally twining is completely lost, and this seems due to the lessened flexibility of the internodes and to the interference caused by the large size of the leaves. But the revolving movement, though restricted, is not lost. In our present species a young internode, placed in front of a window, made three narrow ellipses, transversely to the direction of the light, at an average rate of 2 hrs. 40 m. When placed so that the movements were to and from the light, the rate was greatly accelerated in one half of the course, and retarded in the other, as with twining plants. The ellipses were small; the longer diameter, described by the apex of a shoot bearing a pair of not expanded leaves, was only 4.625 inches, and that by the apex of the penultimate internode only 1.125 inch. At the most favourable period of growth each leaf would hardly be carried to and fro by the movement of the internodes more than two or three inches, but, as above stated, it is probable that the leaves themselves move spontaneously. The movement of the whole shoot by the wind and by its rapid growth, would probably be almost equally efficient as these spontaneous movements, in bringing the petioles into contact with surrounding objects.

The leaves are of large size. Each bears three pairs of lateral leaflets and a terminal one, all supported on rather long sub- petioles. The main petiole bends a little angularly downwards at each point where a pair of leaflets arises (see fig. 2), and the petiole of the terminal leaflet is bent downwards at right angles; hence the whole petiole, with its rectangularly bent extremity, acts as a hook. This hook, the lateral petioles being directed a little upwards; forms an excellent grappling apparatus, by which the leaves readily become entangled with surrounding objects. If they catch nothing, the whole petiole ultimately grows straight. The main petiole, the sub-petioles, and the three branches into which each basi-lateral sub-petiole is generally subdivided, are all sensitive. The basal portion of the main petiole, between the stem and the first pair of leaflets, is less sensitive than the remainder; it will, however, clasp a stick with which it is left in contact. The inferior surface of the rectangularly bent terminal portion (carrying the terminal leaflet), which forms the inner side of the end of the hook, is the most sensitive part; and this portion is manifestly best adapted to catch a distant support. To show the difference in sensibility, I gently placed loops of string of the same weight (in one instance weighing only 0.82 of a grain or 53.14 mg.) on the several lateral sub-petioles and on the terminal one; in a few hours the latter was bent, but after 24 hrs. no effect was produced on the other sub-petioles. Again, a terminal sub-petiole placed in contact with a thin stick became sensibly curved in 45 m., and in 1 hr. 10m. moved through ninety degrees; whilst a lateral sub-petiole did not become sensibly curved until 3 hrs. 30 m. had elapsed. In all cases, if the sticks are taken away, the petioles continue to move during many hours afterwards; so they do after a slight rubbing; but they become straight again, after about a day's interval, that is if the flexure has not been very great or long continued.

The graduated difference in the extension of the sensitiveness in the petioles of the above-described species deserves notice. In C. montana it is confined to the main petiole, and has not spread to the sub-petioles of the three leaflets; so it is with young plants of C. calycina, but in older plants it spreads to the three sub-petioles. In C. viticella the sensitiveness has spread to the petioles of the seven leaflets, and to the subdivisions of the basi-lateral sub- petioles. But in this latter species it has diminished in the basal part of the main petiole, in which alone it resided in C. montana; whilst it has increased in the abruptly bent terminal portion.

Clematis flammula.—The rather thick, straight, and stiff shoots, whilst growing vigorously in the spring, make small oval revolutions, following the sun in their course. Four were made at an average rate of 3 hrs. 45 m. The longer axis of the oval, described by the extreme tip, was directed at right angles to the line joining the opposite leaves; its length was in one case only 1.375, and in another case 1.75 inch; so that the young leaves were moved a very short distance. The shoots of the same plant observed in midsummer, when growing not so quickly, did not revolve at all. I cut down another plant in the early summer, so that by August 1st it had formed new and moderately vigorous shoots; these, when observed under a bell-glass, were on some days quite stationary, and on other days moved to and fro only about the eighth of an inch. Consequently the revolving power is much enfeebled in this species, and under unfavourable circumstances is completely lost. The shoot must depend for coming into contact with surrounding objects on the probable, though not ascertained spontaneous movement of the leaves, on rapid growth, and on movement from the wind. Hence, perhaps, it is that the petioles have acquired a high degree of sensitiveness as a compensation for the little power of movement in the shoots.

The petioles are bowed downwards, and have the same general hook-like form as in C. viticella. The medial petiole and the lateral sub- petioles are sensitive, especially the much bent terminal portion. As the sensitiveness is here greater than in any other species of the genus observed by me, and is in itself remarkable, I will give fuller details. The petioles, when so young that they have not separated from one another, are not sensitive; when the lamina of a leaflet has grown to a quarter of an inch in length (that is, about one-sixth of its full size), the sensitiveness is highest; but at this period the petioles are relatively much more fully developed than are the blades of the leaves. Full-grown petioles are not in the least sensitive. A thin stick placed so as to press lightly against a petiole, having a leaflet a quarter of an inch in length, caused the petiole to bend in 3 hrs. 15 m. In another case a petiole curled completely round a stick in 12 hrs. These petioles were left curled for 24 hrs., and the sticks were then removed; but they never straightened themselves. I took a twig, thinner than the petiole itself, and with it lightly rubbed several petioles four times up and down; these in 1 hr. 45 m. became slightly curled; the curvature increased during some hours and then began to decrease, but after 25 hrs. from the time of rubbing a vestige of the curvature remained. Some other petioles similarly rubbed twice, that is, once up and once down, became perceptibly curved in about 2 hrs. 30 m., the terminal sub-petiole moving more than the lateral sub-petioles; they all became straight again in between 12 hrs. and 14 hrs. Lastly, a length of about one-eighth of an inch of a sub-petiole, was lightly rubbed with the same twig only once; it became slightly curved in 3 hrs., remaining so during 11 hrs., but by the next morning was quite straight.

The following observations are more precise. After trying heavier pieces of string and thread, I placed a loop of fine string, weighing 1.04 gr. (67.4 mg.) on a terminal sub-petiole: in 6 hrs. 40 m. a curvature could be seen; in 24 hrs. the petiole formed an open ring round the string; in 48 hrs. the ring had almost closed on the string, and in 72 hrs. seized it so firmly, that some force was necessary for its withdrawal. A loop weighing 0.52 of a grain (33.7 mg.) caused in 14 hrs. a lateral sub-petiole just perceptibly to curve, and in 24 hrs. it moved through ninety degrees. These observations were made during the summer: the following were made in the spring, when the petioles apparently are more sensitive:- A loop of thread, weighing one-eighth of a grain (8.1 mg.), produced no effect on the lateral sub-petioles, but placed on a terminal one, caused it, after 24 hrs., to curve moderately; the curvature, though the loop remained suspended, was after 48 hrs. diminished, but never disappeared; showing that the petiole had become partially accustomed to the insufficient stimulus. This experiment was twice repeated with nearly the same result. Lastly, a loop of thread, weighing only one-sixteenth of a grain (4.05 mg.) was twice gently placed by a forceps on a terminal sub-petiole (the plant being, of course, in a still and closed room), and this weight certainly caused a flexure, which very slowly increased until the petiole moved through nearly ninety degrees: beyond this it did not move; nor did the petiole, the loop remaining suspended, ever become perfectly straight again.

When we consider, on the one hand, the thickness and stiffness of the petioles, and, on the other hand, the thinness and softness of fine cotton thread, and what an extremely small weight one-sixteenth of a grain (4.05 mg.) is, these facts are remarkable. But I have reason to believe that even a less weight excites curvature when pressing over a broader surface than that acted on by a thread. Having noticed that the end of a suspended string which accidentally touched a petiole, caused it to bend, I took two pieces of thin twine, 10 inches in length (weighing 1.64 gr.), and, tying them to a stick, let them hang as nearly perpendicularly downwards as their thinness and flexuous form, after being stretched, would permit; I then quietly placed their ends so as just to rest on two petioles, and these certainly became curved in 36 hrs. One of the ends touched the angle between a terminal and lateral sub-petiole, and it was in 48 hours caught between them as by a forceps. In these cases the pressure, though spread over a wider surface than that touched by the cotton thread, must have been excessively slight.

Clematis vitalba.—The plants were in pots and not healthy, so that I dare not trust my observations, which indicate much similarity in habits with C. flammula. I mention this species only because I have seen many proofs that the petioles in a state of nature are excited to movement by very slight pressure. For instance, I have found them embracing thin withered blades of grass, the soft young leaves of a maple, and the flower-peduncles of the quaking-grass or Briza. The latter are about as thick as the hair of a man's beard, but they were completely surrounded and clasped. The petioles of a leaf, so young that none of the leaflets were expanded, had partially seized a twig. Those of almost all the old leaves, even when unattached to any object, are much convoluted; but this is owing to their having come, whilst young, into contact during several hours with some object subsequently removed. With none of the above-described species, cultivated in pots and carefully observed, was there any permanent bending of the petioles without the stimulus of contact. In winter, the blades of the leaves of C. vitalba drop off; but the petioles (as was observed by Mohl) remain attached to the branches, sometimes during two seasons; and, being convoluted, they curiously resemble true tendrils, such as those possessed by the allied genus Naravelia. The petioles which have clasped some object become much more stiff, hard, and polished than those which have failed in this their proper function.

TROPAEOLUM.—I observed T. tricolorum, T. azureum, T. pentaphyllum, T. peregrinum, T. elegans, T. tuberosum, and a dwarf variety of, as I believe, T. minus.

Tropaeolum tricolorum, var. grandiflorum.—The flexible shoots, which first rise from the tubers, are as thin as fine twine. One such shoot revolved in a course opposed to the sun, at an average rate, judging from three revolutions, of 1 hr. 23 m.; but no doubt the direction of the revolving movement is variable. When the plants have grown tall and are branched, all the many lateral shoots revolve. The stem, whilst young, twines regularly round a thin vertical stick, and in one case I counted eight spiral turns in the same direction; but when grown older, the stem often runs straight up for a space, and, being arrested by the clasping petioles, makes one or two spires in a reversed direction. Until the plant grows to a height of two or three feet, requiring about a month from the time when the first shoot appears above ground, no true leaves are produced, but, in their place, filaments coloured like the stem. The extremities of these filaments are pointed, a little flattened, and furrowed on the upper surface. They never become developed into leaves. As the plant grows in height new filaments are produced with slightly enlarged tips; then others, bearing on each side of the enlarged medial tip a rudimentary segment of a leaf; soon other segments appear, and at last a perfect leaf is formed, with seven deep segments. So that on the same plant we may see every step, from tendril-like clasping filaments to perfect leaves with clasping petioles. After the plant has grown to a considerable height, and is secured to its support by the petioles of the true leaves, the clasping filaments on the lower part of the stem wither and drop off; so that they perform only a temporary service.

These filaments or rudimentary leaves, as well as the petioles of the perfect leaves, whilst young, are highly sensitive on all sides to a touch. The slightest rub caused them to curve towards the rubbed side in about three minutes, and one bent itself into a ring in six minutes; they subsequently became straight. When, however, they have once completely clasped a stick, if this is removed, they do not straighten themselves. The most remarkable fact, and one which I have observed in no other species of the genus, is that the filaments and the petioles of the young leaves, if they catch no object, after standing for some days in their original position, spontaneously and slowly oscillate a little from side to side, and then move towards the stem and clasp it. They likewise often become, after a time, in some degree spirally contracted. They therefore fully deserve to be called tendrils, as they are used for climbing, are sensitive to a touch, move spontaneously, and ultimately contract into a spire, though an imperfect one. The present species would have been classed amongst the tendril-bearers, had not these characters been confined to early youth. During maturity it is a true leaf-climber.

Tropaeolum azureum.—An upper internode made four revolutions, following the sun, at an average rate of 1 hr. 47 m. The stem twined spirally round a support in the same irregular manner as that of the last species. Rudimentary leaves or filaments do not exist. The petioles of the young leaves are very sensitive: a single light rub with a twig caused one to move perceptibly in 5 m., and another in 6 m. The former became bent at right angles in 15 min., and became straight again in between 5 hrs. and 6 hrs. A loop of thread weighing 0.125th of a grain caused another petiole to curve.