Mechanism of the Movements, and Nature of the Motor Impulse.—Whatever may be the means of movement, the exterior tentacles, considering their delicacy, are inflected with much force. A bristle, held so that a length of 1 inch projected from a handle, yielded when I tried to lift with it an inflected tentacle, which was somewhat thinner than the bristle. The amount or extent, also, of the movement is great. Fully expanded tentacles in becoming inflected sweep through an angle of 180o; and if they are beforehand reflexed, as often occurs, the angle is considerably greater. It is probably the superficial cells at the bending place which chiefly or exclusively contract; for the interior cells have very delicate walls, and are so few in number that they could hardly cause a tentacle to bend with precision to a definite point. Though I carefully looked, I could never detect any wrinkling of the surface at the bending place, even in the case of a tentacle abnormally curved into a complete circle, under circumstances hereafter to be mentioned.

All the cells are not acted on, though the motor impulse passes through them. When the gland of one of the long exterior tentacles is excited, the upper cells are not in the least affected; about halfway down there is a slight bending, but the chief movement is confined to a short space near the base; and no part of the inner tentacles bends except the basal portion. With respect to the blade of the leaf, the motor impulse may be transmitted through many cells, from the centre to the circumference, without their being in the least affected, or they may be strongly acted on and the blade greatly inflected. In the latter case the movement seems to depend partly on the strength of the stimulus, and partly on [page 255] its nature, as when leaves are immersed in certain fluids.

The power of movement which various plants possess, when irritated, has been attributed by high authorities to the rapid passage of fluid out of certain cells, which, from their previous state of tension, immediately contract.* Whether or not this is the primary cause of such movements, fluid must pass out of closed cells when they contract or are pressed together in one direction, unless they at the same time expand in some other direction. For instance, fluid can be seen to ooze from the surface of any young and vigorous shoot if slowly bent into a semi-circle. In the case of Drosera there is certainly much movement of the fluid throughout the tentacles whilst they are undergoing inflection. Many leaves can be found in which the purple fluid within the cells is of an equally dark tint on the upper and lower sides of the tentacles, extending also downwards on both sides to equally near their bases. If the tentacles of such a leaf are excited into movement, it will generally be found after some hours that the cells on the concave side are much paler than they were before, or are quite colourless, those on the convex side having become much darker. In two instances, after particles of hair had been placed on glands, and when in the course of 1 hr. 10 m. the tentacles were incurved halfway towards the centre of the leaf, this change of colour in the two sides was conspicuously plain. In another case, after a bit of meat had been placed on a gland, the purple colour was observed at intervals to be slowly travelling from the upper to the lower part, down the convex side of

* Sachs, 'Trait de Bot.' 3rd edit. 1874, p. 1038. This view was, I believe, first suggested by Lamarck.

Sachs, ibid. p. 919. [page 256]

the bending tentacle. But it does not follow from these observations that the cells on the convex side become filled with more fluid during the act of inflection than they contained before; for fluid may all the time be passing into the disc or into the glands which then secrete freely.

The bending of the tentacles, when leaves are immersed in a dense fluid, and their subsequent re-expansion in a less dense fluid, show that the passage of fluid from or into the cells can cause movements like the natural ones. But the inflection thus caused is often irregular; the exterior tentacles being sometimes spirally curved. Other unnatural movements are likewise caused by the application of dense fluids, as in the case of drops of syrup placed on the backs of leaves and tentacles. Such movements may be compared with the contortions which many vegetable tissues undergo when subjected to exosmose. It is therefore doubtful whether they throw any light on the natural movements.

If we admit that the outward passage of fluid is the cause of the bending of the tentacles, we must suppose that the cells, before the act of inflection, are in a high state of tension, and that they are elastic to an extraordinary degree; for otherwise their contraction could not cause the tentacles often to sweep through an angle of above 180o. Prof. Cohn, in his interesting paper* on the movements of the stamens of certain Compositae, states that these organs, when dead, are as elastic as threads of india-rubber, and are then only half as long as they were when alive. He believes that the living protoplasm

* 'Abhand. der Schles. Gesell. fr vaterl. Cultur,' 1861, Heft i. An excellent abstract of this paper is given in the 'Annals and Mag. of Nat. Hist.' 3rd series, 1863, vol. xi. pp. 188-197. [page 257]

within their cells is ordinarily in a state of expansion, but is paralysed by irritation, or may be said to suffer temporary death; the elasticity of the cell-walls then coming into play, and causing the contraction of the stamens. Now the cells on the upper or concave side of the bending part of the tentacles of Drosera do not appear to be in a state of tension, nor to be highly elastic; for when a leaf is suddenly killed, or dies slowly, it is not the upper but the lower sides of the tentacles which contract from elasticity. We may, therefore, conclude that their movements cannot be accounted for by the inherent elasticity of certain cells, opposed as long as they are alive and not irritated by the expanded state of their contents.

A somewhat different view has been advanced by other physiologists—namely that the protoplasm, when irritated, contracts like the soft sarcode of the muscles of animals. In Drosera the fluid within the cells of the tentacles at the bending place appears under the microscope thin and homogeneous, and after aggregation consists of small, soft masses of matter, undergoing incessant changes of form and floating in almost colourless fluid. These masses are completely redissolved when the tentacles re-expand. Now it seems scarcely possible that such matter should have any direct mechanical power; but if through some molecular change it were to occupy less space than it did before, no doubt the cell-walls would close up and contract. But in this case it might be expected that the walls would exhibit wrinkles, and none could ever be seen. Moreover, the contents of all the cells seem to be of exactly the same nature, both before and after aggregation; and yet only a few of the basal cells contract, the rest of the tentacle remaining straight.

A third view maintained by some physiologists, [page 258] though rejected by most others, is that the whole cell, including the walls, actively contracts. If the walls are composed solely of non-nitrogenous cellulose, this view is highly improbable; but it can hardly be doubted that they must be permeated by proteid matter, at least whilst they are growing. Nor does there seem any inherent improbability in the cell-walls of Drosera contracting, considering their high state of organisation; as shown in the case of the glands by their power of absorption and secretion, and by being exquisitely sensitive so as to be affected by the pressure of the most minute particles. The cell-walls of the pedicels also allow various impulses to pass through them, inducing movement, increased secretion and aggregation. On the whole the belief that the walls of certain cells contract, some of their contained fluid being at the same time forced outwards, perhaps accords best with the observed facts. If this view is rejected, the next most probable one is that the fluid contents of the cells shrink, owing to a change in their molecular state, with the consequent closing in of the walls. Anyhow, the movement can hardly be attributed to the elasticity of the walls, together with a previous state of tension.

With respect to the nature of the motor impulse which is transmitted from the glands down the pedicels and across the disc, it seems not improbable that it is closely allied to that influence which causes the protoplasm within the cells of the glands and tentacles to aggregate. We have seen that both forces originate in and proceed from the glands within a few seconds of the same time, and are excited by the same causes. The aggregation of the protoplasm lasts almost as long as the tentacles remain inflected, even though this be for more than a week; but the [page 259] protoplasm is redissolved at the bending place shortly before the tentacles re-expand, showing that the exciting cause of the aggregating process has then quite ceased. Exposure to carbonic acid causes both the latter process and the motor impulse to travel very slowly down the tentacles. We know that the aggregating process is delayed in passing through the cell- walls, and we have good reason to believe that this holds good with the motor impulse; for we can thus understand the different rates of its transmission in a longitudinal and transverse line across the disc. Under a high power the first sign of aggregation is the appearance of a cloud, and soon afterwards of extremely fine granules, in the homogeneous purple fluid within the cells; and this apparently is due to the union of molecules of protoplasm. Now it does not seem an improbable view that the same tendency—namely for the molecules to approach each other—should be communicated to the inner surfaces of the cell-walls which are in contact with the protoplasm; and if so, their molecules would approach each other, and the cell-wall would contract.

To this view it may with truth be objected that when leaves are immersed in various strong solutions, or are subjected to a heat of above 130o Fahr. (54o.4 Cent.), aggregation ensues, but there is no movement. Again, various acids and some other fluids cause rapid movement, but no aggregation, or only of an abnormal nature, or only after a long interval of time; but as most of these fluids are more or less injurious, they may check or prevent the aggregating process by injuring or killing the protoplasm. There is another and more important difference in the two processes: when the glands on the disc are excited, they transmit some influence up the surrounding [page 260] tentacles, which acts on the cells at the bending place, but does not induce aggregation until it has reached the glands; these then send back some other influence, causing the protoplasm to aggregate, first in the upper and then in the lower cells.

The Re-expansion of the Tentacles.—This movement is always slow and gradual. When the centre of the leaf is excited, or a leaf is immersed in a proper solution, all the tentacles bend directly towards the centre, and afterwards directly back from it. But when the point of excitement is on one side of the disc, the surrounding tentacles bend towards it, and therefore obliquely with respect to their normal direction; when they afterwards re-expand, they bend obliquely back, so as to recover their original positions. The tentacles farthest from an excited point, wherever that may be, are the last and the least affected, and probably in consequence of this they are the first to re-expand. The bent portion of a closely inflected tentacle is in a state of active contraction, as shown by the following experiment. Meat was placed on a leaf, and after the tentacles were closely inflected and had quite ceased to move, narrow strips of the disc, with a few of the outer tentacles attached to it, were cut off and laid on one side under the microscope. After several failures, I succeeded in cutting off the convex surface of the bent portion of a tentacle. Movement immediately recommenced, and the already greatly bent portion went on bending until it formed a perfect circle; the straight distal portion of the tentacle passing on one side of the strip. The convex surface must therefore have previously been in a state of tension, sufficient to counter-balance that of the concave surface, which, when free, curled into a complete ring.

The tentacles of an expanded and unexcited leaf [page 261] are moderately rigid and elastic; if bent by a needle, the upper end yields more easily than the basal and thicker part, which alone is capable of becoming inflected. The rigidity of this basal part seems due to the tension of the outer surface balancing a state of active and persistent contraction of the cells of the inner surface. I believe that this is the case, because, when a leaf is dipped into boiling water, the tentacles suddenly become reflexed, and this apparently indicates that the tension of the outer surface is mechanical, whilst that of the inner surface is vital, and is instantly destroyed by the boiling water. We can thus also understand why the tentacles as they grow old and feeble slowly become much reflexed. If a leaf with its tentacles closely inflected is dipped into boiling water, these rise up a little, but by no means fully re-expand. This may be owing to the heat quickly destroying the tension and elasticity of the cells of the convex surface; but I can hardly believe that their tension, at any one time, would suffice to carry back the tentacles to their original position, often through an angle of above 180o. It is more probable that fluid, which we know travels along the tentacles during the act of inflection, is slowly re-attracted into the cells of the convex surface, their tension being thus gradually and continually increased.

A recapitulation of the chief facts and discussions in this chapter will be given at the close of the next chapter. [page 262]



CHAPTER XI.

RECAPITULATION OF THE CHIEF OBSERVATIONS ON DROSERA ROTUNDIFOLIA.

As summaries have been given to most of the chapters, it will be sufficient here to recapitulate, as briefly as I can, the chief points. In the first chapter a preliminary sketch was given of the structure of the leaves, and of the manner in which they capture insects. This is effected by drops of extremely viscid fluid surrounding the glands and by the inward movement of the tentacles. As the plants gain most of their nutriment by this means, their roots are very poorly developed; and they often grow in places where hardly any other plant except mosses can exist. The glands have the power of absorption, besides that of secretion. They are extremely sensitive to various stimulants, namely repeated touches, the pressure of minute particles, the absorption of animal matter and of various fluids, heat, and galvanic action. A tentacle with a bit of raw meat on the gland has been seen to begin bending in 10 s., to be strongly incurved in 5 m., and to reach the centre of the leaf in half an hour. The blade of the leaf often becomes so much inflected that it forms a cup, enclosing any object placed on it.

A gland, when excited, not only sends some influence down its own tentacle, causing it to bend, but likewise to the surrounding tentacles, which become incurved; so that the bending place can be acted on by an impulse received from opposite directions, [page 263] namely from the gland on the summit of the same tentacle, and from one or more glands of the neighbouring tentacles. Tentacles, when inflected, re-expand after a time, and during this process the glands secrete less copiously, or become dry. As soon as they begin to secrete again, the tentacles are ready to re-act; and this may be repeated at least three, probably many more times.

It was shown in the second chapter that animal substances placed on the discs cause much more prompt and energetic inflection than do inorganic bodies of the same size, or mere mechanical irritation; but there is a still more marked difference in the greater length of time during which the tentacles remain inflected over bodies yielding soluble and nutritious matter, than over those which do not yield such matter. Extremely minute particles of glass, cinders, hair, thread, precipitated chalk, &c., when placed on the glands of the outer tentacles, cause them to bend. A particle, unless it sinks through the secretion and actually touches the surface of the gland with some one point, does not produce any effect. A little bit of thin human hair 8/1000 of an inch (.203 mm.) in length, and weighing only 1/78740 of a grain (.000822 mg.), though largely supported by the dense secretion, suffices to induce movement. It is not probable that the pressure in this case could have amounted to that from the millionth of a grain. Even smaller particles cause a slight movement, as could be seen through a lens. Larger particles than those of which the measurements have been given cause no sensation when placed on the tongue, one of the most sensitive parts of the human body.

Movement ensues if a gland is momentarily touched three or four times; but if touched only once or twice, [page 264] though with considerable force and with a hard object, the tentacle does not bend. The plant is thus saved from much useless movement, as during a high wind the glands can hardly escape being occasionally brushed by the leaves of surrounding plants. Though insensible to a single touch, they are exquisitely sensitive, as just stated, to the slightest pressure if prolonged for a few seconds; and this capacity is manifestly of service to the plant in capturing small insects. Even gnats, if they rest on the glands with their delicate feet, are quickly and securely embraced. The glands are insensible to the weight and repeated blows of drops of heavy rain, and the plants are thus likewise saved from much useless movement.

The description of the movements of the tentacles was interrupted in the third chapter for the sake of describing the process of aggregation. This process always commences in the cells of the glands, the contents of which first become cloudy; and this has been observed within 10 s. after a gland has been excited. Granules just resolvable under a very high power soon appear, sometimes within a minute, in the cells beneath the glands; and these then aggregate into minute spheres. The process afterwards travels down the tentacles, being arrested for a short time at each transverse partition. The small spheres coalesce into larger spheres, or into oval, club-headed, thread- or necklace-like, or otherwise shaped masses of protoplasm, which, suspended in almost colourless fluid, exhibit incessant spontaneous changes of form. These frequently coalesce and again separate. If a gland has been powerfully excited, all the cells down to the base of the tentacle are affected. In cells, especially if filled with dark red fluid, the first step in the [page 265] process often is the formation of a dark red, bag-like mass of protoplasm, which afterwards divides and undergoes the usual repeated changes of form. Before any aggregation has been excited, a sheet of colourless protoplasm, including granules (the primordial utricle of Mohl), flows round the walls of the cells; and this becomes more distinct after the contents have been partially aggregated into spheres or bag-like masses. But after a time the granules are drawn towards the central masses and unite with them; and then the circulating sheet can no longer be distinguished, but there is still a current of transparent fluid within the cells.

Aggregation is excited by almost all the stimulants which induce movement; such as the glands being touched two or three times, the pressure of minute inorganic particles, the absorption of various fluids, even long immersion in distilled water, exosmose, and heat. Of the many stimulants tried, carbonate of ammonia is the most energetic and acts the quickest: a dose of 1/134400 of a grain (.00048 mg.) given to a single gland suffices to cause in one hour well-marked aggregation in the upper cells of the tentacle. The process goes on only as long as the protoplasm is in a living, vigorous, and oxygenated condition.

The result is in all respects exactly the same, whether a gland has been excited directly, or has received an influence from other and distant glands. But there is one important difference: when the central glands are irritated, they transmit centrifugally an influence up the pedicels of the exterior tentacles to their glands; but the actual process of aggregation travels centripetally, from the glands of the exterior tentacles down their pedicels. The exciting influence, therefore, which is transmitted from [page 266] one part of the leaf to another must be different from that which actually induces aggregation. The process does not depend on the glands secreting more copiously than they did before; and is independent of the inflection of the tentacles. It continues as long as the tentacles remain inflected, and as soon as these are fully re-expanded, the little masses of protoplasm are all redissolved; the cells becoming filled with homogeneous purple fluid, as they were before the leaf was excited.

As the process of aggregation can be excited by a few touches, or by the pressure of insoluble particles, it is evidently independent of the absorption of any matter, and must be of a molecular nature. Even when caused by the absorption of the carbonate or other salt of ammonia, or an infusion of meat, the process seems to be of exactly the same nature. The protoplasmic fluid must, therefore, be in a singularly unstable condition, to be acted on by such slight and varied causes. Physiologists believe that when a nerve is touched, and it transmits an influence to other parts of the nervous system, a molecular change is induced in it, though not visible to us. Therefore it is a very interesting spectacle to watch the effects on the cells of a gland, of the pressure of a bit of hair, weighing only 1/78700 of a grain and largely supported by the dense secretion, for this excessively slight pressure soon causes a visible change in the protoplasm, which change is transmitted down the whole length of the tentacle, giving it at last a mottled appearance, distinguishable even by the naked eye.

In the fourth chapter it was shown that leaves placed for a short time in water at a temperature of 110o Fahr. (43o.3 Cent.) become somewhat inflected; they are thus also rendered more sensitive to the action [page 267] of meat than they were before. If exposed to a temperature of between 115o and 125o(46o.1-51o.6 Cent.), they are quickly inflected, and their protoplasm undergoes aggregation; when afterwards placed in cold water, they re-expand. Exposed to 130o (54o.4 Cent.), no inflection immediately occurs, but the leaves are only temporarily paralysed, for on being left in cold water, they often become inflected and afterwards re-expand. In one leaf thus treated, I distinctly saw the protoplasm in movement. In other leaves, treated in the same manner, and then immersed in a solution of carbonate of ammonia, strong aggregation ensued. Leaves placed in cold water, after an exposure to so high a temperature as 145o (62o.7 Cent.), sometimes become slightly, though slowly, inflected; and afterwards have the contents of their cells strongly aggregated by carbonate of ammonia. But the duration of the immersion is an important element, for if left in water at 145o (62o.7 Cent.), or only at 140o (60o Cent.), until it becomes cool, they are killed, and the contents of the glands are rendered white and opaque. This latter result seems to be due to the coagulation of the albumen, and was almost always caused by even a short exposure to 150o (65o.5 Cent.); but different leaves, and even the separate cells in the same tentacle, differ considerably in their power of resisting heat. Unless the heat has been sufficient to coagulate the albumen, carbonate of ammonia subsequently induces aggregation.

In the fifth chapter, the results of placing drops of various nitrogenous and non-nitrogenous organic fluids on the discs of leaves were given, and it was shown that they detect with almost unerring certainty the presence of nitrogen. A decoction of green peas or of fresh cabbage-leaves acts almost as powerfully as an infusion of raw meat; whereas an infusion of cabbage- [page 268] leaves made by keeping them for a long time in merely warm water is far less efficient. A decoction of grass-leaves is less powerful than one of green peas or cabbage-leaves.

These results led me to inquire whether Drosera possessed the power of dissolving solid animal matter. The experiments proving that the leaves are capable of true digestion, and that the glands absorb the digested matter, are given in detail in the sixth chapter. These are, perhaps, the most interesting of all my observations on Drosera, as no such power was before distinctly known to exist in the vegetable kingdom. It is likewise an interesting fact that the glands of the disc, when irritated, should transmit some influence to the glands of the exterior tentacles, causing them to secrete more copiously and the secretion to become acid, as if they had been directly excited by an object placed on them. The gastric juice of animals contains, as is well known, an acid and a ferment, both of which are indispensable for digestion, and so it is with the secretion of Drosera. When the stomach of an animal is mechanically irritated, it secretes an acid, and when particles of glass or other such objects were placed on the glands of Drosera, the secretion, and that of the surrounding and untouched glands, was increased in quantity and became acid. But, according to Schiff, the stomach of an animal does not secrete its proper ferment, pepsin, until certain substances, which he calls peptogenes, are absorbed; and it appears from my experiments that some matter must be absorbed by the glands of Drosera before they secrete their proper ferment. That the secretion does contain a ferment which acts only in the presence of an acid on solid animal matter, was clearly proved by adding minute doses of [page 269] an alkali, which entirely arrested the process of digestion, this immediately recommencing as soon as the alkali was neutralised by a little weak hydrochloric acid. From trials made with a large number of substances, it was found that those which the secretion of Drosera dissolves completely, or partially, or not at all, are acted on in exactly the same manner by gastric juice. We may, therefore, conclude that the ferment of Drosera is closely analogous to, or identical with, the pepsin of animals.

The substances which are digested by Drosera act on the leaves very differently. Some cause much more energetic and rapid inflection of the tentacles, and keep them inflected for a much longer time, than do others. We are thus led to believe that the former are more nutritious than the latter, as is known to be the case with some of these same substances when given to animals; for instance, meat in comparison with gelatine. As cartilage is so tough a substance and is so little acted on by water, its prompt dissolution by the secretion of Drosera, and subsequent absorption is, perhaps, one of the most striking cases. But it is not really more remarkable than the digestion of meat, which is dissolved by this secretion in the same manner and by the same stages as by gastric juice. The secretion dissolves bone, and even the enamel of teeth, but this is simply due to the large quantity of acid secreted, owing, apparently, to the desire of the plant for phosphorus. In the case of bone, the ferment does not come into play until all the phosphate of lime has been decomposed and free acid is present, and then the fibrous basis is quickly dissolved. Lastly, the secretion attacks and dissolves matter out of living seeds, which it sometimes kills, or injures, as shown by the diseased state [page 270] of the seedlings. It also absorbs matter from pollen, and from fragments of leaves.

The seventh chapter was devoted to the action of the salts of ammonia. These all cause the tentacles, and often the blade of the leaf, to be inflected, and the protoplasm to be aggregated. They act with very different power; the citrate being the least powerful, and the phosphate, owing, no doubt, to the presence of phosphorus and nitrogen, by far the most powerful. But the relative efficiency of only three salts of ammonia was carefully determined, namely the carbonate, nitrate, and phosphate. The experiments were made by placing half-minims (.0296 ml.) of solutions of different strengths on the discs of the leaves,—by applying a minute drop (about the 1/20 of a minim, or .00296 ml.) for a few seconds to three or four glands,—and by the immersion of whole leaves in a measured quantity. In relation to these experiments it was necessary first to ascertain the effects of distilled water, and it was found, as described in detail, that the more sensitive leaves are affected by it, but only in a slight degree.

A solution of the carbonate is absorbed by the roots and induces aggregation in their cells, but does not affect the leaves. The vapour is absorbed by the glands, and causes inflection as well as aggregation. A drop of a solution containing 1/960 of a grain (.0675 mg.) is the least quantity which, when placed on the glands of the disc, excites the exterior tentacles to bend inwards. But a minute drop, containing 1/14400 of a grain (.00445 mg.), if applied for a few seconds to the secretion surrounding a gland, causes the inflection of the same tentacle. When a highly sensitive leaf is immersed in a solution, and there is ample time for absorption, the 1/268800 of a grain [page 271] (.00024 mg.) is sufficient to excite a single tentacle into movement.

The nitrate of ammonia induces aggregation of the protoplasm much less quickly than the carbonate, but is more potent in causing inflection. A drop containing 1/2400 of a grain (.027 mg.) placed on the disc acts powerfully on all the exterior tentacles, which have not themselves received any of the solution; whereas a drop with 1/2800 of a grain caused only a few of these tentacles to bend, but affected rather more plainly the blade. A minute drop applied as before, and containing 1/28800 of a grain (.0025 mg.), caused the tentacle bearing this gland to bend. By the immersion of whole leaves, it was proved that the absorption by a single gland of 1/691200 of a grain (.0000937 mg.) was sufficient to set the same tentacle into movement.

The phosphate of ammonia is much more powerful than the nitrate. A drop containing 1/3840 of a grain (.0169 mg.) placed on the disc of a sensitive leaf causes most of the exterior tentacles to be inflected, as well as the blade of the leaf. A minute drop containing 1/153600 of a grain (.000423 mg.), applied for a few seconds to a gland, acts, as shown by the movement of the tentacle. When a leaf is immersed in thirty minims (1.7748 ml.) of a solution of one part by weight of the salt to 21,875,000 of water, the absorption by a gland of only the 1/19760000 of a grain (.00000328 mg.), that is, about the one-twenty-millionth of a grain, is sufficient to cause the tentacle bearing this gland to bend to the centre of the leaf. In this experiment, owing to the presence of the water of crystallisation, less than the one-thirty-millionth of a grain of the efficient elements could have been absorbed. There is nothing remarkable in such minute quantities being absorbed by the glands, [page 272] for all physiologists admit that the salts of ammonia, which must be brought in still smaller quantity by a single shower of rain to the roots, are absorbed by them. Nor is it surprising that Drosera should be enabled to profit by the absorption of these salts, for yeast and other low fungoid forms flourish in solutions of ammonia, if the other necessary elements are present. But it is an astonishing fact, on which I will not here again enlarge, that so inconceivably minute a quantity as the one-twenty-millionth of a grain of phosphate of ammonia should induce some change in a gland of Drosera, sufficient to cause a motor impulse to be sent down the whole length of the tentacle; this impulse exciting movement often through an angle of above 180o. I know not whether to be most astonished at this fact, or that the pressure of a minute bit of hair, supported by the dense secretion, should quickly cause conspicuous movement. Moreover, this extreme sensitiveness, exceeding that of the most delicate part of the human body, as well as the power of transmitting various impulses from one part of the leaf to another, have been acquired without the intervention of any nervous system.

As few plants are at present known to possess glands specially adapted for absorption, it seemed worth while to try the effects on Drosera of various other salts, besides those of ammonia, and of various acids. Their action, as described in the eighth chapter, does not correspond at all strictly with their chemical affinities, as inferred from the classification commonly followed. The nature of the base is far more influential than that of the acid; and this is known to hold good with animals. For instance, nine salts of sodium all caused well-marked inflection, and none of them were poisonous in small doses; whereas seven of the nine corre- [page 273] sponding salts of potassium produced no effect, two causing slight inflection. Small doses, moreover, of some of the latter salts were poisonous. The salts of sodium and potassium, when injected into the veins of animals, likewise differ widely in their action. The so-called earthy salts produce hardly any effect on Drosera. On the other hand, most of the metallic salts cause rapid and strong inflection, and are highly poisonous; but there are some odd exceptions to this rule; thus chloride of lead and zinc, as well as two salts of barium, did not cause inflection, and were not poisonous.

Most of the acids which were tried, though much diluted (one part to 437 of water), and given in small doses, acted powerfully on Drosera; nineteen, out of the twenty-four, causing the tentacles to be more or less inflected. Most of them, even the organic acids, are poisonous, often highly so; and this is remarkable, as the juices of so many plants contain acids. Benzoic acid, which is innocuous to animals, seems to be as poisonous to Drosera as hydrocyanic. On the other hand, hydrochloric acid is not poisonous either to animals or to Drosera, and induces only a moderate amount of inflection. Many acids excite the glands to secrete an extraordinary quantity of mucus; and the protoplasm within their cells seems to be often killed, as may be inferred from the surrounding fluid soon becoming pink. It is strange that allied acids act very differently: formic acid induces very slight inflection, and is not poisonous; whereas acetic acid of the same strength acts most powerfully and is poisonous. Lactic acid is also poisonous, but causes inflection only after a considerable lapse of time. Malic acid acts slightly, whereas citric and tartaric acids produce no effect. [page 274]

In the ninth chapter the effects of the absorption of various alkaloids and certain other substances were described. Although some of these are poisonous, yet as several, which act powerfully on the nervous system of animals, produce no effect on Drosera, we may infer that the extreme sensibility of the glands, and their power of transmitting an influence to other parts of the leaf, causing movement, or modified secretion, or aggregation, does not depend on the presence of a diffused element, allied to nerve-tissue. One of the most remarkable facts is that long immersion in the poison of the cobra-snake does not in the least check, but rather stimulates, the spontaneous movements of the protoplasm in the cells of the tentacles. Solutions of various salts and acids behave very differently in delaying or in quite arresting the subsequent action of a solution of phosphate of ammonia. Camphor dissolved in water acts as a stimulant, as do small doses of certain essential oils, for they cause rapid and strong inflection. Alcohol is not a stimulant. The vapours of camphor, alcohol, chloroform, sulphuric and nitric ether, are poisonous in moderately large doses, but in small doses serve as narcotics or, anaesthetics, greatly delaying the subsequent action of meat. But some of these vapours also act as stimulants, exciting rapid, almost spasmodic movements in the tentacles. Carbonic acid is likewise a narcotic, and retards the aggregation of the protoplasm when carbonate of ammonia is subsequently given. The first access of air to plants which have been immersed in this gas sometimes acts as a stimulant and induces movement. But, as before remarked, a special pharmacopoeia would be necessary to describe the diversified effects of various substances on the leaves of Drosera.

In the tenth chapter it was shown that the sensitive- [page 275] ness of the leaves appears to be wholly confined to the glands and to the immediately underlying cells. It was further shown that the motor impulse and other forces or influences, proceeding from the glands when excited, pass through the cellular tissue, and not along the fibro-vascular bundles. A gland sends its motor impulse with great rapidity down the pedicel of the same tentacle to the basal part which alone bends. The impulse, then passing onwards, spreads on all sides to the surrounding tentacles, first affecting those which stand nearest and then those farther off. But by being thus spread out, and from the cells of the disc not being so much elongated as those of the tentacles, it loses force, and here travels much more slowly than down the pedicels. Owing also to the direction and form of the cells, it passes with greater ease and celerity in a longitudinal than in a transverse line across the disc. The impulse proceeding from the glands of the extreme marginal tentacles does not seem to have force enough to affect the adjoining tentacles; and this may be in part due to their length. The impulse from the glands of the next few inner rows spreads chiefly to the tentacles on each side and towards the centre of the leaf; but that proceeding from the glands of the shorter tentacles on the disc radiates almost equally on all sides.

When a gland is strongly excited by the quantity or quality of the substance placed on it, the motor impulse travels farther than from one slightly excited; and if several glands are simultaneously excited, the impulses from all unite and spread still farther. As soon as a gland is excited, it discharges an impulse which extends to a considerable distance; but afterwards, whilst the gland is secreting and absorbing, the impulse suffices only to keep the same tentacle [page 276] inflected; though the inflection may last for many days.

If the bending place of a tentacle receives an impulse from its own gland, the movement is always towards the centre of the leaf; and so it is with all the tentacles, when their glands are excited by immersion in a proper fluid. The short ones in the middle part of the disc must be excepted, as these do not bend at all when thus excited. On the other hand, when the motor impulse comes from one side of the disc, the surrounding tentacles, including the short ones in the middle of the disc, all bend with precision towards the point of excitement, wherever this may be seated. This is in every way a remarkable phenomenon; for the leaf falsely appears as if endowed with the senses of an animal. It is all the more remarkable, as when the motor impulse strikes the base of a tentacle obliquely with respect to its flattened surface, the contraction of the cells must be confined to one, two, or a very few rows at one end. And different sides of the surrounding tentacles must be acted on, in order that all should bend with precision to the point of excitement.

The motor impulse, as it spreads from one or more glands across the disc, enters the bases of the surrounding tentacles, and immediately acts on the bending place. It does not in the first place proceed up the tentacles to the glands, exciting them to reflect back an impulse to their bases. Nevertheless, some influence is sent up to the glands, as their secretion is soon increased and rendered acid; and then the glands, being thus excited, send back some other influence (not dependent on increased secretion, nor on the inflection of the tentacles), causing the protoplasm to aggregate in cell beneath cell. This may [page 277] be called a reflex action, though probably very different from that proceeding from the nerve-ganglion of an animal; and it is the only known case of reflex action in the vegetable kingdom.

About the mechanism of the movements and the nature of the motor impulse we know very little. During the act of inflection fluid certainly travels from one part to another of the tentacles. But the hypothesis which agrees best with the observed facts is that the motor impulse is allied in nature to the aggregating process; and that this causes the molecules of the cell-walls to approach each other, in the same manner as do the molecules of the protoplasm within the cells; so that the cell-walls contract. But some strong objections may be urged against this view. The re-expansion of the tentacles is largely due to the elasticity of their outer cells, which comes into play as soon as those on the inner side cease contracting with prepotent force; but we have reason to suspect that fluid is continually and slowly attracted into the outer cells during the act of re-expansion, thus increasing their tension.

I have now given a brief recapitulation of the chief points observed by me, with respect to the structure, movements, constitution, and habits of Drosera rotundifolia; and we see how little has been made out in comparison with what remains unexplained and unknown. [page 278]



CHAPTER XII.

ON THE STRUCTURE AND MOVEMENTS OF SOME OTHER SPECIES OF DROSERA.

Drosera anglica—Drosera intermedia—Drosera capensis—Drosera spathulata—Drosera filiformis—Drosera binata—Concluding remarks.

I EXAMINED six other species of Drosera, some of them inhabitants of distant countries, chiefly for the sake of ascertaining whether they caught insects. This seemed the more necessary as the leaves of some of the species differ to an extraordinary degree in shape from the rounded ones of Drosera rotundifolia. In functional powers, however, they differ very little.

[Drosera anglica (Hudson).*—The leaves of this species, which was sent to me from Ireland, are much elongated, and gradually widen from the footstalk to the bluntly pointed apex. They stand almost erect, and their blades sometimes exceed 1 inch in length, whilst their breadth is only the 1/5 of an inch. The glands of all the tentacles have the same structure, so that the extreme marginal ones do not differ from the others, as in the case of Drosera rotundifolia. When they are irritated by being roughly touched, or by the pressure of minute inorganic particles, or by contact with animal matter, or by the absorption of carbonate of ammonia, the tentacles become inflected; the basal portion being the chief seat of movement. Cutting or pricking the blade of the leaf did not excite any movement. They frequently capture insects, and the glands of the inflected tentacles pour forth much acid secretion. Bits of roast meat were placed on some glands, and the tentacles began to move in 1 m. or

* Mrs. Treat has given an excellent account in 'The American Naturalist,' December 1873, p. 705, of Drosera longifolia (which is a synonym in part of Drosera anglica), of Drosera rotundifolia and filiformis. [page 279]

1 m. 30 s.; and in 1 hr. 10 m. reached the centre. Two bits of boiled cork, one of boiled thread, and two of coal-cinders taken from the fire, were placed, by the aid of an instrument which had been immersed in boiling water, on five glands; these superfluous precautions having been taken on account of M. Ziegler's statements. One of the particles of cinder caused some inflection in 8 hrs. 45 m., as did after 23 hrs. the other particle of cinder, the bit of thread, and both bits of cork. Three glands were touched half a dozen times with a needle; one of the tentacles became well inflected in 17 m., and re-expanded after 24 hrs.; the two others never moved. The homogeneous fluid within the cells of the tentacles undergoes aggregation after these have become inflected; especially if given a solution of carbonate of ammonia; and I observed the usual movements in the masses of protoplasm. In one case, aggregation ensued in 1 hr. 10 m. after a tentacle had carried a bit of meat to the centre. From these facts it is clear that the tentacles of Drosera anglica behave like those of Drosera rotundifolia.

If an insect is placed on the central glands, or has been naturally caught there, the apex of the leaf curls inwards. For instance, dead flies were placed on three leaves near their bases, and after 24 hrs. the previously straight apices were curled completely over, so as to embrace and conceal the flies; they had therefore moved through an angle of 180o. After three days the apex of one leaf, together with the tentacles, began to re-expand. But as far as I have seen— and I made many trials—the sides of the leaf are never inflected, and this is the one functional difference between this species and Drosera rotundifolia.

Drosera intermedia (Hayne).—This species is quite as common in some parts of England as Drosera rotundifolia. It differs from Drosera anglica, as far as the leaves are concerned, only in their smaller size, and in their tips being generally a little reflexed. They capture a large number of insects. The tentacles are excited into movement by all the causes above specified; and aggregation ensues, with movement of the protoplasmic masses. I have seen, through a lens, a tentacle beginning to bend in less than a minute after a particle of raw meat had been placed on the gland. The apex of the leaf curls over an exciting object as in the case of Drosera anglica. Acid secretion is copiously poured over captured insects. A leaf which had embraced a fly with all its tentacles re-expanded after nearly three days.

Drosera capensis.—This species, a native of the Cape of Good Hope, was sent to me by Dr. Hooker. The leaves are elongated, slightly concave along the middle and taper towards the apex, [page 280] which is bluntly pointed and reflexed. They rise from an almost woody axis, and their greatest peculiarity consists in their foliaceous green footstalks, which are almost as broad and even longer than the gland-bearing blade. This species, therefore, probably draws more nourishment from the air, and less from captured insects, than the other species of the genus. Nevertheless, the tentacles are crowded together on the disc, and are extremely numerous; those on the margins being much longer than the central ones. All the glands have the same form; their secretion is extremely viscid and acid.

The specimen which I examined had only just recovered from a weak state of health. This may account for the tentacles moving very slowly when particles of meat were placed on the glands, and perhaps for my never succeeding in causing any movement by repeatedly touching them with a needle. But with all the species of the genus this latter stimulus is the least effective of any. Particles of glass, cork, and coal-cinders, were placed on the glands of six tentacles; and one alone moved after an interval of 2 hrs. 30 m. Nevertheless, two glands were extremely sensitive to very small doses of the nitrate of ammonia, namely to about 1/20 of a minim of a solution (one part to 5250 of water), containing only 1/115200 of a grain (.000562 mg.) of the salt. Fragments of flies were placed on two leaves near their tips, which became incurved in 15 hrs. A fly was also placed in the middle of the leaf; in a few hours the tentacles on each side embraced it, and in 8 hrs. the whole leaf directly beneath the fly was a little bent transversely. By the next morning, after 23 hrs., the leaf was curled so completely over that the apex rested on the upper end of the footstalk. In no case did the sides of the leaves become inflected. A crushed fly was placed on the foliaceous footstalk, but produced no effect.

Drosera spathulata (sent to me by Dr. Hooker).—I made only a few observations on this Australian species, which has long, narrow leaves, gradually widening towards their tips. The glands of the extreme marginal tentacles are elongated and differ from the others, as in the case of Drosera rotundifolia. A fly was placed on a leaf, and in 18 hrs. it was embraced by the adjoining tentacles. Gum-water dropped on several leaves produced no effect. A fragment of a leaf was immersed in a few drops of a solution of one part of carbonate of ammonia to 146 of water; all the glands were instantly blackened; the process of aggregation could be seen travelling rapidly down the cells of the tentacles; and the granules of protoplasm soon united into spheres and variously shaped masses, which displayed the usual move- [page 281] ments. Half a minim of a solution of one part of nitrate of ammonia to 146 of water was next placed on the centre of a leaf; after 6 hrs. some marginal tentacles on both sides were inflected, and after 9 hrs. they met in the centre. The lateral edges of the leaf also became incurved, so that it formed a half-cylinder; but the apex of the leaf in none of my few trials was inflected. The above dose of the nitrate (viz. 1/320 of a grain, or .202 mg.) was too powerful, for in the course of 23 hrs. the leaf died.

Drosera filiformis.—This North American species grows in such abundance in parts of New Jersey as almost to cover the ground. It catches, according to Mrs. Treat,* an extraordinary number of small and large insects, even great flies of the genus Asilus, moths, and butterflies. The specimen which I examined, sent me by Dr. Hooker, had thread-like leaves, from 6 to 12 inches in length, with the upper surface convex and the lower flat and slightly channelled. The whole convex surface, down to the roots—for there is no distinct footstalk—is covered with short gland-bearing tentacles, those on the margins being the longest and reflexed. Bits of meat placed on the glands of some tentacles caused them to be slightly inflected in 20 m.; but the plant was not in a vigorous state. After 6 hrs. they moved through an angle of 90o, and in 24 hrs. reached the centre. The surrounding tentacles by this time began to curve inwards. Ultimately a large drop of extremely viscid, slightly acid secretion was poured over the meat from the united glands. Several other glands were touched with a little saliva, and the tentacles became incurved in under 1 hr., and re-expanded after 18 hrs. Particles of glass, cork, cinders, thread, and gold-leaf, were placed on numerous glands on two leaves; in about 1 hr. four tentacles became curved, and four others after an additional interval of 2 hrs. 30 m. I never once succeeded in causing any movement by repeatedly touching the glands with a needle; and Mrs. Treat made similar trials for me with no success. Small flies were placed on several leaves near their tips, but the thread-like blade became only on one occasion very slightly bent, directly beneath the insect. Perhaps this indicates that the blades of vigorous plants would bend over captured insects, and Dr. Canby informs me that this is the case; but the movement cannot be strongly pronounced, as it was not observed by Mrs. Treat.

Drosera binata (or dichotoma).—I am much indebted to Lady

* 'American Naturalist,' December 1873, page 705. [page 282]

Dorothy Nevill for a fine plant of this almost gigantic Australian species, which differs in some interesting points from those previously described. In this specimen the rush-like footstalks of the leaves were 20 inches in length. The blade bifurcates at its junction with the footstalk, and twice or thrice afterwards, curling about in an irregular manner. It is narrow, being only 3/20 of an inch in breadth. One blade was 7 1/2 inches long, so that the entire leaf, including the footstalk, was above 27 inches in length. Both surfaces are slightly hollowed out. The upper surface is covered with tentacles arranged in alternate rows; those in the middle being short and crowded together, those towards the margins longer, even twice or thrice as long as the blade is broad. The glands of the exterior tentacles are of a much darker red than those of the central ones. The pedicels of all are green. The apex of the blade is attenuated, and bears very long tentacles. Mr. Copland informs me that the leaves of a plant which he kept for some years were generally covered with captured insects before they withered.

The leaves do not differ in essential points of structure or of function from those of the previously described species. Bits of meat or a little saliva placed on the glands of the exterior tentacles caused well-marked movement in 3 m., and particles of glass acted in 4 m. The tentacles with the latter particles re-expanded after 22 hrs. A piece of leaf immersed in a few drops of a solution of one part of carbonate of ammonia to 437 of water had all the glands blackened and all the tentacles inflected in 5 m. A bit of raw meat, placed on several glands in the medial furrow, was well clasped in 2 hrs. 10 m. by the marginal tentacles on both sides. Bits of roast meat and small flies did not act quite so quickly; and albumen and fibrin still less quickly. One of the bits of meat excited so much secretion (which is always acid) that it flowed some way down the medial furrow, causing the inflection of the tentacles on both sides as far as it extended. Particles of glass placed on the glands in the medial furrow did not stimulate them sufficiently for any motor impulse to be sent to the outer tentacles. In no case was the blade of the leaf, even the attenuated apex, at all inflected.

On both the upper and lower surface of the blade there are numerous minute, almost sessile glands, consisting of four, eight, or twelve cells. On the lower surface they are pale purple, on the upper greenish. Nearly similar organs occur on the foot-stalks, but they are smaller and often in a shrivelled condition. The minute glands on the blade can absorb rapidly: thus, a piece of leaf was immersed in a solution of one part of carbonate [page 283] of ammonia to 218 of water (1 gr. to 2 oz.), and in 5 m. they were all so much darkened as to be almost black, with their contents aggregated. They do not, as far as I could observe, secrete spontaneously; but in between 2 and 3 hrs. after a leaf had been rubbed with a bit of raw meat moistened with saliva, they seemed to be secreting freely; and this conclusion was afterwards supported by other appearances. They are, therefore, homologous with the sessile glands hereafter to be described on the leaves of Dionaea and Drosophyllum. In this latter genus they are associated, as in the present case, with glands which secrete spontaneously, that is, without being excited.

Drosera binata presents another and more remarkable peculiarity, namely, the presence of a few tentacles on the backs of the leaves, near their margins. These are perfect in structure; spiral vessels run up their pedicels; their glands are surrounded by drops of viscid secretion, and they have the power of absorbing. This latter fact was shown by the glands immediately becoming black, and the protoplasm aggregated, when a leaf was placed in a little solution of one part of carbonate of ammonia to 437 of water. These dorsal tentacles are short, not being nearly so long as the marginal ones on the upper surface; some of them are so short as almost to graduate into the minute sessile glands. Their presence, number, and size, vary on different leaves, and they are arranged rather irregularly. On the back of one leaf I counted as many as twenty-one along one side.

These dorsal tentacles differ in one important respect from those on the upper surface, namely, in not possessing any power of movement, in whatever manner they may be stimulated. Thus, portions of four leaves were placed at different times in solutions of carbonate of ammonia (one part to 437 or 218 of water), and all the tentacles on the upper surface soon became closely inflected; but the dorsal ones did not move, though the leaves were left in the solution for many hours, and though their glands from their blackened colour had obviously absorbed some of the salt. Rather young leaves should be selected for such trials, for the dorsal tentacles, as they grow old and begin to wither, often spontaneously incline towards the middle of the leaf. If these tentacles had possessed the power of movement, they would not have been thus rendered more serviceable to the plant; for they are not long enough to bend round the margin of the leaf so as to reach an insect caught on the upper surface, Nor would it have been of any use if these tentacles could have [page 284] moved towards the middle of the lower surface, for there are no viscid glands there by which insects can be caught. Although they have no power of movement, they are probably of some use by absorbing animal matter from any minute insect which may be caught by them, and by absorbing ammonia from the rain-water. But their varying presence and size, and their irregular position, indicate that they are not of much service, and that they are tending towards abortion. In a future chapter we shall see that Drosophyllum, with its elongated leaves, probably represents the condition of an early progenitor of the genus Drosera; and none of the tentacles of Drosophyllum, neither those on the upper nor lower surface of the leaves, are capable of movement when excited, though they capture numerous insects, which serve as nutriment. Therefore it seems that Drosera binata has retained remnants of certain ancestral characters—namely a few motionless tentacles on the backs of the leaves, and fairly well developed sessile glands—which have been lost by most or all of the other species of the genus.]

Concluding Remarks.—From what we have now seen, there can be little doubt that most or probably all the species of Drosera are adapted for catching insects by nearly the same means. Besides the two Australian species above described, it is said* that two other species from this country, namely Drosera pallida and Drosera sulphurea, "close their leaves upon insects with "great rapidity: and the same phenomenon is mani-"fested by an Indian species, D. lunata, and by several "of those of the Cape of Good Hope, especially by "D. trinervis." Another Australian species, Drosera heterophylla (made by Lindley into a distinct genus, Sondera) is remarkable from its peculiarly shaped leaves, but I know nothing of its power of catching insects, for I have seen only dried specimens. The leaves form minute flattened cups, with the footstalks attached not to one margin, but to the bottom. The

* 'Gardener's Chronicle,' 1874, p. 209. [page 285]

inner surface and the edges of the cups are studded with tentacles, which include fibro-vascular bundles, rather different from those seen by me in any other species; for some of the vessels are barred and punctured, instead of being spiral. The glands secrete copiously, judging from the quantity of dried secretion adhering to them. [page 286]



CHAPTER XIII.

DIONAEA MUSCIPULA.

Structure of the leaves—Sensitiveness of the filaments—Rapid movement of the lobes caused by irritation of the filaments—Glands, their power of secretion—Slow movement caused by the absorption of animal matter—Evidence of absorption from the aggregated condition of the glands—Digestive power of the secretion—Action of chloroform, ether, and hydrocyanic acid- -The manner in which insects are captured—Use of the marginal spikes—Kinds of insects captured—The transmission of the motor impulse and mechanism of the movements— Re-expansion of the lobes.

THIS plant, commonly called Venus' fly-trap, from the rapidity and force of its movements, is one of the most wonderful in the world.* It is a member of the small family of the Droseraceae, and is found only in the eastern part of North Carolina, growing in damp situations. The roots are small; those of a moderately fine plant which I examined consisted of two branches about 1 inch in length, springing from a bulbous enlargement. They probably serve, as in the case of Drosera, solely for the absorption of water; for a gardener, who has been very successful in the cultivation of this plant, grows it, like an epiphytic orchid, in well-drained damp moss without any soil. The form of the bilobed leaf, with its foliaceous footstalk, is shown in the accompanying drawing (fig. 12).

* Dr. Hooker, in his address to the British Association at Belfast, 1874, has given so full an historical account of the observations which have been published on the habits of this plant, that it would be superfluous on my part to repeat them.

'Gardener's Chronicle,' 1874, p. 464. [page 287]

The two lobes stand at rather less than a right angle to each other. Three minute pointed processes or filaments, placed triangularly, project from the upper surfaces of both; but I have seen two leaves with four filaments on each side, and another with only two. These filaments are remarkable from their extreme sensitiveness to a touch, as shown not by their own movement, but by that of the lobes. The margins of the leaf are prolonged into sharp rigid projections which I will call spikes, into each of which a bundle

FIG. 12. (Dionaea muscipula.) Leaf viewed laterally in its expanded state.

of spiral vessels enters. The spikes stand in such a position that, when the lobes close, they inter-lock like the teeth of a rat-trap. The midrib of the leaf, on the lower side, is strongly developed and prominent.

The upper surface of the leaf is thickly covered, excepting towards the margins, with minute glands of a reddish or purplish colour, the rest of the leaf being green. There are no glands on the spikes, or on the foliaceous footstalk, The glands are formed of from [page 288] twenty to thirty polygonal cells, filled with purple fluid. Their upper surface is convex. They stand on very short pedicels, into which spiral vessels do not enter, in which respect they differ from the tentacles of Drosera. They secrete, but only when excited by the absorption of certain matters; and they have the power of absorption. Minute projections, formed of eight divergent arms of a reddish-brown or orange colour, and appearing under the microscope like elegant little flowers, are scattered in considerable numbers over the foot-stalk, the backs of the leaves, and the spikes, with a few on the upper surface of the lobes. These octofid projections are no doubt homologous with the papillae on the leaves of Drosera rotundifolia. There are also a few very minute, simple, pointed hairs, about 7/12000 (.0148 mm.) of an inch in length on the backs of the leaves.

The sensitive filaments are formed of several rows of elongated cells, filled with purplish fluid. They are a little above the 1/20 of an inch in length; are thin and delicate, and taper to a point. I examined the bases of several, making sections of them, but no trace of the entrance of any vessel could be seen. The apex is sometimes bifid or even trifid, owing to a slight separation between the terminal pointed cells. Towards the base there is constriction, formed of broader cells, beneath which there is an articulation, supported on an enlarged base, consisting of differently shaped polygonal cells. As the filaments project at right angles to the surface of the leaf, they would have been liable to be broken whenever the lobes closed together, had it not been for the articulation which allows them to bend flat down.

These filaments, from their tips to their bases, are exquisitely sensitive to a momentary touch. It is scarcely [page 289] possible to touch them ever so lightly or quickly with any hard object without causing the lobes to close. A piece of very delicate human hair, 2 1/2 inches in length, held dangling over a filament, and swayed to and fro so as to touch it, did not excite any movement. But when a rather thick cotton thread of the same length was similarly swayed, the lobes closed. Pinches of fine wheaten flour, dropped from a height, produced no effect. The above-mentioned hair was then fixed into a handle, and cut off so that 1 inch projected; this length being sufficiently rigid to support itself in a nearly horizontal line. The extremity was then brought by a slow movement laterally into contact with the tip of a filament, and the leaf instantly closed. On another occasion two or three touches of the same kind were necessary before any movement ensued. When we consider how flexible a fine hair is, we may form some idea how slight must be the touch given by the extremity of a piece, 1 inch in length, moved slowly.

Although these filaments are so sensitive to a momentary and delicate touch, they are far less sensitive than the glands of Drosera to prolonged pressure. Several times I succeeded in placing on the tip of a filament, by the aid of a needle moved with extreme slowness, bits of rather thick human hair, and these did not excite movement, although they were more than ten times as long as those which caused the tentacles of Drosera to bend; and although in this latter case they were largely supported by the dense secretion. On the other hand, the glands of Drosera may be struck with a needle or any hard object, once, twice, or even thrice, with considerable force, and no movement ensues. This singular difference in the nature of the sensitiveness of the filaments of Dionaea and of [page 290] the glands of Drosera evidently stands in relation to the habits of the two plants. If a minute insect alights with its delicate feet on the glands of Drosera, it is caught by the viscid secretion, and the slight, though prolonged pressure, gives notice of the presence of prey, which is secured by the slow bending of the tentacles. On the other hand, the sensitive filaments of Dionaea are not viscid, and the capture of insects can be assured only by their sensitiveness to a momentary touch, followed by the rapid closure of the lobes.

As just stated, the filaments are not glandular, and do not secrete. Nor have they the power of absorption, as may be inferred from drops of a solution of carbonate of ammonia (one part to 146 of water), placed on two filaments, not producing any effect on the contents of their cells, nor causing the lobes to close, When, however, a small portion of a leaf with an attached filament was cut off and immersed in the same solution, the fluid within the basal cells became almost instantly aggregated into purplish or colourless, irregularly shaped masses of matter. The process of aggregation gradually travelled up the filaments from cell to cell to their extremities, that is in a reverse course to what occurs in the tentacles of Drosera when their glands have been excited. Several other filaments were cut off close to their bases, and left for 1 hr. 30 m. in a weaker solution of one part of the carbonate to 218 of water, and this caused aggregation in all the cells, commencing as before at the bases of the filaments.

Long immersion of the filaments in distilled water likewise causes aggregation. Nor is it rare to find the contents of a few of the terminal cells in a spontaneously aggregated condition. The aggregated [page 291] masses undergo incessant slow changes of form, uniting and again separating; and some of them apparently revolve round their own axes. A current of colourless granular protoplasm could also be seen travelling round the walls of the cells. This current ceases to be visible as soon as the contents are well aggregated; but it probably still continues, though no longer visible, owing to all the granules in the flowing layer having become united with the central masses. In all these respects the filaments of Dionaea behave exactly like the tentacles of Drosera.

Notwithstanding this similarity there is one remarkable difference. The tentacles of Drosera, after their glands have been repeatedly touched, or a particle of any kind has been placed on them, become inflected and strongly aggregated. No such effect is produced by touching the filaments of Dionaea; I compared, after an hour or two, some which had been touched and some which had not, and others after twenty-five hours, and there was no difference in the contents of the cells. The leaves were kept open all the time by clips; so that the filaments were not pressed against the opposite lobe.

Drops of water, or a thin broken stream, falling from a height on the filaments, did not cause the blades to close; though these filaments were afterwards proved to be highly sensitive. No doubt, as in the case of Drosera, the plant is indifferent to the heaviest shower of rain. Drops of a solution of a half an ounce of sugar to a fluid ounce of water were repeatedly allowed to fall from a height on the filaments, but produced no effect, unless they adhered to them. Again, I blew many times through a fine pointed tube with my utmost force against the filaments without any effect; such blowing being received [page 292] with as much indifference as no doubt is a heavy gale of wind. We thus see that the sensitiveness of the filaments is of a specialised nature, being related to a momentary touch rather than to prolonged pressure; and the touch must not be from fluids, such as air or water, but from some solid object.

Although drops of water and of a moderately strong solution of sugar, falling on the filaments, does not excite them, yet the immersion of a leaf in pure water sometimes caused the lobes to close. One leaf was left immersed for 1 hr. 10 m., and three other leaves for some minutes, in water at temperatures varying between 59o and 65o (15o to 18o.3 Cent.) without any effect. One, however, of these four leaves, on being gently withdrawn from the water, closed rather quickly. The three other leaves were proved to be in good condition, as they closed when their filaments were touched. Nevertheless two fresh leaves on being dipped into water at 75o and 62 1/2o (23o.8 and 16o.9 Cent.) instantly closed. These were then placed with their footstalks in water, and after 23 hrs. partially re-expanded; on touching their filaments one of them closed. This latter leaf after an additional 24 hrs. again re-expanded, and now, on the filaments of both leaves being touched, both closed. We thus see that a short immersion in water does not at all injure the leaves, but sometimes excites the lobes to close. The movement in the above cases was evidently not caused by the temperature of the water. It has been shown that long immersion causes the purple fluid within the cells of the sensitive filaments to become aggregated; and the tentacles of Drosera are acted on in the same manner by long immersion, often being somewhat inflected. In both cases the result is probably due to a slight degree of exosmose. [page 293]

I am confirmed in this belief by the effects of immersing a leaf of Dionaea in a moderately strong solution of sugar; the leaf having been previously left for 1 hr. 10 m. in water without any effect; for now the lobes closed rather quickly, the tips of the marginal spikes crossing in 2 m. 30 s., and the leaf being completely shut in 3 m. Three leaves were then immersed in a solution of half an ounce of sugar to a fluid ounce of water, and all three leaves closed quickly. As I was doubtful whether this was due to the cells on the upper surface of the lobes, or to the sensitive filaments, being acted on by exosmose, one leaf was first tried by pouring a little of the same solution in the furrow between the lobes over the midrib, which is the chief seat of movement. It was left there for some time, but no movement ensued. The whole upper surface of leaf was then painted (except close round the bases of the sensitive filaments, which I could not do without risk of touching them) with the same solution, but no effect was produced. So that the cells on the upper surface are not thus affected. But when, after many trials, I succeeded in getting a drop of the solution to cling to one of the filaments, the leaf quickly closed. Hence we may, I think, conclude that the solution causes fluid to pass out of the delicate cells of the filaments by exosmose; and that this sets up some molecular change in their contents, analogous to that which must be produced by a touch.

The immersion of leaves in a solution of sugar affects them for a much longer time than does an immersion in water, or a touch on the filaments; for in these latter cases the lobes begin to re-expand in less than a day. On the other hand, of the three leaves which were immersed for a short time in the solution, and were then washed by means of a syringe inserted [page 294] between the lobes, one re-expanded after two days; a second after seven days; and the third after nine days. The leaf which closed, owing to a drop of the solution having adhered to one of the filaments, opened after two days.

I was surprised to find on two occasions that the heat from the rays of the sun, concentrated by a lens on the bases of several filaments, so that they were scorched and discoloured, did not cause any movement; though the leaves were active, as they closed, though rather slowly, when a filament on the opposite side was touched. On a third trial, a fresh leaf closed after a time, though very slowly; the rate not being increased by one of the filaments, which had not been injured, being touched. After a day these three leaves opened, and were fairly sensitive when the uninjured filaments were touched. The sudden immersion of a leaf into boiling water does not cause it to close. Judging from the analogy of Drosera, the heat in these several cases was too great and too suddenly applied. The surface of the blade is very slightly sensitive; It may be freely and roughly handled, without any movement being caused. A leaf was scratched rather hard with a needle, but did not close; but when the triangular space between the three filaments on another leaf was similarly scratched, the lobes closed. They always closed when the blade or midrib was deeply pricked or cut. Inorganic bodies, even of large size, such as bits of stone, glass, &c.—or organic bodies not containing soluble nitrogenous matter, such as bits of wood, cork, moss,—or bodies containing soluble nitrogenous matter, if perfectly dry, such as bits of meat, albumen, gelatine, &c., may be long left (and many were tried) on the lobes, and no movement is excited. The result, however, is widely different, as we [page 295] shall presently see, if nitrogenous organic bodies which are at all damp, are left on the lobes; for these then close by a slow and gradual movement, very different from that caused by touching one of the sensitive filaments. The footstalk is not in the least sensitive; a pin may be driven through it, or it may be cut off, and no movement follows.

The upper surface of the lobes, as already stated, is thickly covered with small purplish, almost sessile glands. These have the power both of secretion and absorption; but unlike those of Drosera, they do not secrete until excited by the absorption of nitrogenous matter. No other excitement, as far as I have seen, produces this effect. Objects, such as bits of wood, cork, moss, paper, stone, or glass, may be left for a length of time on the surface of a leaf, and it remains quite dry. Nor does it make any difference if the lobes close over such objects. For instance, some little balls of blotting paper were placed on a leaf, and a filament was touched; and when after 24 hrs. the lobes began to re-open, the balls were removed by the aid of thin pincers, and were found perfectly dry. On the other hand, if a bit of damp meat or a crushed fly is placed on the surface of an expanded leaf, the glands after a time secrete freely. In one such case there was a little secretion directly beneath the meat in 4 hrs.; and after an additional 3 hrs. there was a considerable quantity both under and close round it. In another case, after 3 hrs. 40 m., the bit of meat was quite wet. But none of the glands secreted, excepting those which actually touched the meat or the secretion containing dissolved animal matter.

If, however, the lobes are made to close over a bit of meat or an insect, the result is different, for the glands over the whole surface of the leaf now secrete copiously. [page 296] As in this case the glands on both sides are pressed against the meat or insect, the secretion from the first is twice as great as when a bit of meat is laid on the surface of one lobe; and as the two lobes come into almost close contact, the secretion, containing dissolved animal matter, spreads by capillary attraction, causing fresh glands on both sides to begin secreting in a continually widening circle. The secretion is almost colourless, slightly mucilaginous, and, judging by the manner in which it coloured litmus paper, more strongly acid than that of Drosera. It is so copious that on one occasion, when a leaf was cut open, on which a small cube of albumen had been placed 45 hrs. before, drops rolled off the leaf. On another occasion, in which a leaf with an enclosed bit of roast meat spontaneously opened after eight days, there was so much secretion in the furrow over the midrib that it trickled down. A large crushed fly (Tipula) was placed on a leaf from which a small portion at the base of one lobe had previously been cut away, so that an opening was left; and through this, the secretion continued to run down the footstalk during nine days,—that is, for as long a time as it was observed. By forcing up one of the lobes, I was able to see some distance between them, and all the glands within sight were secreting freely.

We have seen that inorganic and non-nitrogenous objects placed on the leaves do not excite any movement; but nitrogenous bodies, if in the least degree damp, cause after several hours the lobes to close slowly. Thus bits of quite dry meat and gelatine were placed at opposite ends of the same leaf, and in the course of 24 hrs. excited neither secretion nor movement. They were then dipped in water, their surfaces dried on blotting paper, and replaced on the same [page 297] leaf, the plant being now covered with a bell-glass. After 24 hrs. the damp meat had excited some acid secretion, and the lobes at this end of the leaf were almost shut. At the other end, where the damp gelatine lay, the leaf was still quite open, nor had any secretion been excited; so that, as with Drosera, gelatine is not nearly so exciting a substance as meat. The secretion beneath the meat was tested by pushing a strip of litmus paper under it (the filaments not being touched), and this slight stimulus caused the leaf to shut. On the eleventh day it reopened; but the end where the gelatine lay, expanded several hours before the opposite end with the meat.

A second bit of roast meat, which appeared dry, though it had not been purposely dried, was left for 24 hrs. on a leaf, caused neither movement nor secretion. The plant in its pot was now covered with a bell-glass, and the meat absorbed some moisture from the air; this sufficed to excite acid secretion, and by the next morning the leaf was closely shut. A third bit of meat, dried so as to be quite brittle, was placed on a leaf under a bell-glass, and this also became in 24 hrs. slightly damp, and excited some acid secretion, but no movement.

A rather large piece of perfectly dry albumen was left at one end of a leaf for 24 hrs. without any effect. It was then soaked for a few minutes in water, rolled about on blotting paper, and replaced on the leaf; in 9 hrs. some slightly acid secretion was excited, and in 24 hrs. this end of the leaf was partially closed. The bit of albumen, which was now surrounded by much secretion, was gently removed, and although no filament was touched, the lobes closed. In this and the previous case, it appears that the absorption of animal matter by the glands renders [page 298] the surface of the leaf much more sensitive to a touch than it is in its ordinary state; and this is a curious fact. Two days afterwards the end of the leaf where nothing had been placed began to open, and on the third day was much more open than the opposite end where the albumen had lain.

Lastly, large drops of a solution of one part of carbonate of ammonia to 146 of water were placed on some leaves, but no immediate movement ensued. I did not then know of the slow movement caused by animal matter, otherwise I should have observed the leaves for a longer time, and they would probably have been found closed, though the solution (judging from Drosera) was, perhaps, too strong.

From the foregoing cases it is certain that bits of meat and albumen, if at all damp, excite not only the glands to secrete, but the lobes to close. This movement is widely different from the rapid closure caused by one of the filaments being touched. We shall see its importance when we treat of the manner in which insects are captured. There is a great contrast between Drosera and Dionaea in the effects produced by mechanical irritation on the one hand, and the absorption of animal matter on the other. Particles of glass placed on the glands of the exterior tentacles of Drosera excite movement within nearly the same time, as do particles of meat, the latter being rather the most efficient; but when the glands of the disc have bits of meat given them, they transmit a motor impulse to the exterior tentacles much more quickly than do these glands when bearing inorganic particles, or when irritated by repeated touches. On the other hand, with Dionaea, touching the filaments excites incomparably quicker movement than the absorption of animal matter by the glands. Nevertheless, in [page 299] certain cases, this latter stimulus is the more powerful of the two. On three occasions leaves were found which from some cause were torpid, so that their lobes closed only slightly, however much their filaments were irritated; but on inserting crushed insects between the lobes, they became in a day closely shut.

The facts just given plainly show that the glands have the power of absorption, for otherwise it is impossible that the leaves should be so differently affected by non-nitrogenous and nitrogenous bodies, and between these latter in a dry and damp condition. It is surprising how slightly damp a bit of meat or albumen need be in order to excite secretion and afterwards slow movement, and equally surprising how minute a quantity of animal matter, when absorbed, suffices to produce these two effects. It seems hardly credible, and yet it is certainly a fact, that a bit of hard-boiled white of egg, first thoroughly dried, then soaked for some minutes in water and rolled on blotting paper, should yield in a few hours enough animal matter to the glands to cause them to secrete, and afterwards the lobes to close. That the glands have the power of absorption is likewise shown by the very different lengths of time (as we shall presently see) during which the lobes remain closed over insects and other bodies yielding soluble nitrogenous matter, and over such as do not yield any. But there is direct evidence of absorption in the condition of the glands which have remained for some time in contact with animal matter. Thus bits of meat and crushed insects were several times placed on glands, and these were compared after some hours with other glands from distant parts of the same leaf. The latter showed not a trace of aggregation, whereas those which had been in contact with the animal matter were [page 300] well aggregated. Aggregation may be seen to occur very quickly if a piece of a leaf is immersed in a weak solution of carbonate of ammonia. Again, small cubes of albumen and gelatine were left for eight days on a leaf, which was then cut open. The whole surface was bathed with acid secretion, and every cell in the many glands which were examined had its contents aggregated in a beautiful manner into dark or pale purple, or colourless globular masses of protoplasm. These underwent incessant slow changes of forms; sometimes separating from one another and then reuniting, exactly as in the cells of Drosera. Boiling water makes the contents of the gland-cells white and opaque, but not so purely white and porcelain-like as in the case of Drosera. How living insects, when naturally caught, excite the glands to secrete so quickly as they do, I know not; but I suppose that the great pressure to which they are subjected forces a little excretion from either extremity of their bodies, and we have seen that an extremely small amount of nitrogenous matter is sufficient to excite the glands.

Before passing on to the subject of digestion, I may state that I endeavoured to discover, with no success, the functions of the minute octofid processes with which the leaves are studded. From facts hereafter to be given in the chapters on Aldrovanda and Utricularia, it seemed probable that they served to absorb decayed matter left by the captured insects; but their position on the backs of the leaves and on the footstalks rendered this almost impossible. Nevertheless, leaves were immersed in a solution of one part of urea to 437 of water, and after 24 hrs. the orange layer of protoplasm within the arms of these processes did not appear more aggregated than in other speci- [page 301] mens kept in water, I then tried suspending a leaf in a bottle over an excessively putrid infusion of raw meat, to see whether they absorbed the vapour, but their contents were not affected.

Digestive Power of the Secretion.*—When a leaf closes over any object, it may be said to form itself into a temporary stomach; and if the object yields ever so little animal matter, this serves, to use Schiff's expression, as a peptogene, and the glands on the surface pour forth their acid secretion, which acts like the gastric juice of animals. As so many experiments were tried on the digestive power of Drosera, only a few were made with Dionaea, but they were amply sufficient to prove that it digests, This plant, moreover, is not so well fitted as Drosera for observation, as the process goes on within the closed lobes. Insects, even beetles, after being subjected to the secretion for several days, are surprisingly softened, though their chitinous coats are not corroded,

[Experiment 1.—A cube of albumen of 1/10 of an inch (2.540 mm.) was placed at one end of a leaf, and at the other end an oblong piece of gelatine, 1/5 of an inch (5.08 mm.) long, and

* Dr. W.M. Canby, of Wilmington, to whom I am much indebted for information regarding Dionaea in its native home, has published in the 'Gardener's Monthly,' Philadelphia, August 1868, some interesting observations. He ascertained that the secretion digests animal matter, such as the contents of insects, bits of meat, &c.; and that the secretion is reabsorbed. He was also well aware that the lobes remain closed for a much longer time when in contact with animal matter than when made to shut by a mere touch, or over objects not yielding soluble nutriment; and that in these latter cases the glands do not secrete. The Rev. Dr. Curtis first observed ('Boston Journal Nat. Hist.' vol. i., p. 123) the secretion from the glands. I may here add that a gardener, Mr. Knight, is said (Kirby and Spencer's 'Introduction to Entomology,' 1818, vol. i., p. 295) to have found that a plant of the Dionaea, on the leaves of which "he laid fine filaments of raw beef, was much more luxuriant in its growth than others not so treated." [page 302]

1/10 broad; the leaf was then made to close. It was cut open after 45 hrs. The albumen was hard and compressed, with its angles only a little rounded; the gelatine was corroded into an oval form; and both were bathed in so much acid secretion that it dropped off the leaf. The digestive process apparently is rather slower than in Drosera, and this agrees with the length of time during which the leaves remain closed over digestible objects.

Experiment 2.—A bit of albumen 1/10 of an inch square, but only 1/20 in thickness, and a piece of gelatine of the same size as before, were placed on a leaf, which eight days afterwards was cut open. The surface was bathed with slightly adhesive, very acid secretion, and the glands were all in an aggregated condition. Not a vestige of the albumen or gelatine was left. Similarly sized pieces were placed at the same time on wet moss on the same pot, so that they were subjected to nearly similar conditions; after eight days these were brown, decayed, and matted with fibres of mould, but had not disappeared.

Experiment 3.—A piece of albumen 3/20 of an inch (3.81 mm.) long, and 1/20 broad and thick, and a piece of gelatine of the same size as before, were placed on another leaf, which was cut open after seven days; not a vestige of either substance was left, and only a moderate amount of secretion on the surface.

Experiment 4.—Pieces of albumen and gelatine, of the same size as in the last experiment, were placed on a leaf, which spontaneously opened after twelve days, and here again not a vestige of either was left, and only a little secretion at one end of the midrib.

Experiment 5.—Pieces of albumen and gelatine of the same size were placed on another leaf, which after twelve days was still firmly closed, but had begun to wither; it was cut open, and contained nothing except a vestige of brown matter where the albumen had lain.

Experiment 6.—A cube of albumen of 1/10 of an inch and a piece of gelatine of the same size as before were placed on a leaf, which opened spontaneously after thirteen days, The albumen, which was twice as thick as in the latter experiments, was too large; for the glands in contact with it were injured and were dropping off; a film also of albumen of a brown colour, matted with mould, was left. All the gelatine was absorbed, and there was only a little acid secretion left on the midrib.