ILMtf^Uifl^ mom ®lje ^. P- pa pbrary NORTH CAROLINA STATE UNIVERSITY LIBRARIES S01112583 L Date Due i^^^^-^ SEP 0 .31995" THIS BOOK IS DUE ON THE HATE INDICATED BELOW AND IS SUB- JECT TO AN OVERDUE FINE AS POSTED AT THE CIRCULATION DESK. HXCEPTION earlier if this It 150M/01 -92— 920179 Date due will be misRECAiign THE STATURAL HISTORY OF PLANTS W ^^ A-.i W\\\IA t^^ yi/L^K^ spicacity than all his predecessors and contem- ^^^tt ^^<-^-.,_ poraries, " to reduce to one simple universal prin- ,^;-— ^^v/^^ ^{{^^^^^ ciple all the multifarious phenomena of the glorious ^^4l/\ 1 vJ/^^ garden of the world," Goethe conceived the notion of a typical plant, an ideal, the realization of which is achieved in nature by means of a mani- fold variation of individual parts. This abstract notion of a plant's development with its six stages corresponding to "three wave-crests" or expan- sions (Leaf, Petal, Carpel) and "three wave- troughs" or contractions (Cotyledon, Sepal, Sta- men) is expressed graphically in figure 3. It still holds its ground at the present day under the name of Goethe's " Urpflanze," and the credit of its invention is entirely his. But it is not quite right to claim for Goethe, in addition, the title of founder of the doctrine of vegetable metamor- phosis; for in reality he only offered another inter- pretation and mode of representation of a pheno- menon already included by Linnaeus under the term metamorphosis. Linnaeus had instituted a comparison between the metamorphosis of plants and that of insects; in particular, he likened the calyx to the ruptured integument of a chrysalis and the internal parts of a flower to the perfect insect (Imago). He also made many different attempts to establish analogies between the development of plants and that of animals; and in so doing he opened up a wide field for the speculations of the "nature philosophers" in the earlier part of the nineteenth century. An extensive study of this subject now commenced; and writers on nature- philosophy worked indefatigably at the amplification and modification of this theme, first broached by Linnaeus. "A plant is a magnetic needle attracted towards the light from the earth into the air. It is a galvanic bubble, and, as such, is earth, water, and air. The plant- bubble possesses two opposite extremities, a single terrestrial end and a dual aerial end; and so plants must be looked upon as being organisms which manifest a Fig. 3.— Goethe's "Urpflanze.' THE STUDY OF PLANTS IN ANCIENT AND MODERN TIMES. 13 continual struggle to become earth on the one hand and air on the other, unmixed metal at one end, and dual air at the other. A plant is a radius, which becomes single towards the centre, whilst it divides or unfolds towards the periphery; it is not therefore an entire circle or sphere, but only a segment of one of those figures. The individual animal, on the contrary, constitutes of itself a sphere, and is there- fore equivalent to all plants put together. Animals are entire worlds, satellites or moons, which circle independently round the earth; whereas plants are only equal to a heavenly body in their totality. An animal is an infinitude of plants. A blossom which, when severed from the stem, preserves by its own movement the galvanic process or life, is an animal. An animal is a flower-bubble set free from the earth and living alone in air and water by virtue of its own motion." Page after page of the writings on Nature-philosophy of Oken (1810) and other contemporary naturalists is filled with interminable statements of the same kind. At the present day it seems scarcely credible that such propositions were then received with admiration as profound and ingenious utterances, and that they were even adopted as mottoes for botanical and geological treatises. For example, it is worthy of record that as late as the year 1843 the Austrian botanist Unger made use of the last of the flowers of rhetoric above quoted from Oken's Nature- philosophy as a motto for one of his first works on the history of development, the title of which is Plants at the Moment of their becoming Animals. The general divisions or systems of the vegetable kingdom which were evolved by adherents of the school of Nature-philosophy were, as may be imagined, just as absurd as the speculations on which they were based. In his Philosophical Systems of Plants Oken develops in the first place the idea that the vegetable kingdom is a single plant taken to pieces. Inasmuch as the ideal highest plant is composed of five organs, there must likewise be five classes: root-plants, stem-plants, leaf- plants, flower-plants, and fruit-plants. The world is fashioned out of the elements: earth, water, air, and fire. Hereupon is founded a classification of root-plants into earth-plants or lichens, water-plants or fungi, air-plants or mosses, and light-plants or ferns. Proceeding from the assumption that all the groups are parallel and that the principle of classification for each group is always given by the one preceding it, we have next, to take one instance, the second class — that of stem-plants — divided (in accordance with the subdivision of earth into earths, salts, bronzes, and ores) into earth-plants or grasses, salt-plants or lilies, bronze-plants or spices, and ore-plants or palms. SCIENTIFIC METHOD BASED ON THE HISTORY OF DEVELOPMENT. Though as we see the doctrine of metamorphosis, with its conception of a typical plant, degenerated thus into the most barren of fancies, still from it originated the line of research based on the history of development which has since borne fruit in every department of botany. Observers arrived at the conviction that every living plant undergoes a continuous transformation which follows a definite 14 THE STUDY OF PLANTS IN ANCIENT AND MODERN TIMES. course, and that accordingly every species is constructed on a plan fixed within general limits and exhibiting variation in externals only. These, it is true, are often more conspicuous at first sight than the direction and disposition of the parts which are really fundamental, and secure the stability of the entire structure. But in order to ascertain the plan of construction it was found necessary to go back to the very first visible appearance of each organ; to determine how the original rudi- ments of the embryo and the beginnings of roots, stems, leaves, and parts of the flower are formed, and to see what rudiments succeed in opening out, branching and dividing, and what remain behind to perish and be displaced by organs growing vigorously in close proximity to them. These researches into the course of development of the separate parts of flower- ing plants, and to a still greater extent the observations of the development of cryptogams or spore-plants (rendered possible by improvements in the construction of microscopes), led naturally to a study of the history of the elementary structures of which all plants are composed. Previously three kinds of elementary organs had been supposed to exist, utricles, vessels, and fibres. The observations of Brown and Mohl (1830-1840) resulted, however, in the identification of the cell as the common starting-point of all these elementary organs. This led to the further discoveries that protoplasm is the formative and living part of a cell, and that each cell is difierentiated into a protoplasmic cell-body and a cell-membrane. It followed that the envelope of the protoplasmic body, the cell-membrane, which had hitherto been considered the primary formation, was in reality a ^product of the protoplasm enveloped by it, and this discovery resulted in a complete revolution in the con- ception of cells generally. Further investigation led to the conclusion that the various modes of growth and multiplication depend on definite laws. That even in the mode of juxtaposition of daughter-cells arising in reproduction, a certain plan of construction may be distinguished in each species which must stand ultimately in some causal relation to the structural system of the whole plant. The progress achieved along these lines in the course of a few decades has been extraordinarily great, no doubt due to the peculiar fascination which the study of the life-histories and transformations of living organisms and the observation of mysterious processes invisible to the naked eye have had for the mind of the inquirer. In that group of plants which includes the forms classed together by the earlier botanists under the name of Cryptogamia an altogether new world was revealed. An undreamed-of variety was discovered to exist in the processes of propagation and rejuvenescence of these forms of plants by means of single cells or spores. Objects which, having regard to their external form, had been assigned to widely different groups, were found to be connected with one another as stages in the development of one and the same species; and one result of these discoveries was the establishment in this division of the vegetable kingdom of an entirely new system of classification based on life-histories. The systematic arrangement of Flowering-plants or Phanerogams also underwent essential alteration. The Linnsean system, founded on the numerical relations between the difterent parts of the flower, THE STUDY OF PLANTS IN ANCIENT AND MODERN TIMES. 15 had indeed already been displaced by another method of classification, that of the French observers Jussieu (1789) and De Candolle (1813), who framed systems said to be natural when contrasted with the artificial system of Linnaeus. At bottom, however, these classifications only differed from the Linngean in the fact that they multiplied and widened the grounds of division. The main division of Phanero- gamia into those which put forth one cotyledon (or seed-leaf) on germinating (Monocotyledones) and those whose seedlings bear two cotyledons (Dicotyledones) is the only one that could serve as a starting-point for a system based on the history of development; but when we come to the grouping of Dicotyledones into those destitute of corolla (Apetalas), those with the corolla composed of coherent petals (Monopetalse), and those with the corolla composed of distinct petals (Dialy- petalae), we have already to admit something forced, and a reliance on characteristics merely external. The system which is the outcome of the study of development starts with the idea that similarity between adult forms is not always decisive evidence of their belonging to the same group, and that the relationships of different plants is much more surely indicated by the fact of their exhibiting the same laws of growth and the same phenomena of reproduction. Plants exhibiting widely different external forms in the mature state are nevertheless to be looked upon as closely allied if they are constructed according to the same plan, and vice versd. There can be no question that a system based on these principles means a material advance. At the same time it cannot be overlooked that great difficulties are involved in hitting upon the right selection from among the number of phenomena observed in the course of a plant's development, and in determining which of these phenomena are to be referred to a mode of construction common to a number of plants, and therefore treated as fundamental properties, and which should be esteemed merely as outcomes of the conditions of life affecting the existence of the plant in question. OBJECTS OF BOTANICAL EESEAECH AT THE PRESENT DAY. Descriptive Botany only concerns itself with the configuration of a plant. Comparative Morphology endeavours to trace back to a single prototype the extremely various forms exhibited by mature plants. The history of development deals with the growth and differentiation of such forms. But all these paths of research shirk the problem of the biological significance of the different forms. The line of investigation starting from the conception of a plant's life as a series of physical and chemical processes, and which attempts to elucidate the configui-a- tion of a plant in the light of its environment, could not be developed with the slightest prospect of success until physics, chemistry, and other allied sciences had reached a high degree of perfection, and till botanists had become convinced that the phenomena of life are only to be fathomed by means of experiment. The earliest attempts to define the biological significance of the several parts of 16 THE STUDY OF PLANTS IN ANCIENT AND MODERN TIMES. a plant do, it is true, take one back as far as Aristotle and his school; but the ideas of vegetable life entertained at that time are scarcely more than fantastic dreams; and the recognition now accorded to them springs rather from a reverence for antiquity than from any intrinsic merit which they possessed. The first experi- mental investigations into the vital phenomena of plants were published by Stephen Hales in 1718; but it was not till a hundred years later that this kind of research really came into vogue. It brought with it the conception of a cell as a miniature chemical laboratory, and looked for mechanical interpretations of the phenomena of nutrition, sap-circulation, growth, movement — in short, all vital processes — and for some connection between these processes and the external form. Whereas, in the case of descriptive and speculative botany, and in the study of development, the entire plant was first taken into consideration, next its several parts, and lastly the cells and protoplasm; in the new department of inquiry, on the contrary, the complete histories of the ultimate organs were studied first of all, then the significance of the diflferent forms of the several members, and lastly the phenomena occasioned by the aggregate life of all the various kinds of animals and plants. Modem science, governed as it is by the desire to lay bare the causes of all phenomena, is no longer satisfied with knowledge concerning the existence of cells, the arrangement of the difierent forms of cell, the development of their contents, and the changes undergone by cell-membranes. At the present day we inquire what are the functions of the various bodies which are formed within the proto- plasm? Why is the cell-membrane thickened at a particular spot in a particular manner? What is the meaning of all the tubes and passages which exhibit such great diversity of size and shape? What part is played by the peculiar mouths of these channels, and why do they vary so greatly in shape and distribution in plants which are subject to different external conditions? We are no longer content to determine in what manner the rudimentary organ of a plant is produced, or how it expands in one case and frequently divides, or else is arrested in its growth and shrivels up; but we inquire the reason why one rudiment grows and develops whilst another is obliterated. For us no fact is without significance. Our curiosity extends to the shape, size, and direction of the roots; to the configuration, venation, and insertion of the leaves; to the structure and colour of the flowers; and to the form of the fruit and seeds; and we assume that even each thorn, prickle, or hair has a definite function to fulfil. But efforts are also made to explain the mutual relations of the different organs of a plant, and the relations between different species of plants which grow together. Lastly, this department of research (the rapid growth of which is due to Darwin) includes amongst its objects a solution of the problem of the ultimate grounds of morphological variety, the causes of which can only be sought for in a qualitative variation of protoplasm. Specific relationship is explained by attributing it to similarity in the constitution of the protoplasm of allied species, and the affinities exhibited by living and extinct plants are used as means of unfolding the hereditary connection between the THE STUDY OF PLANTS IN ANCIENT AND MODERN TIMES. 17 thousands of different sorts of forms, and of tracing the history of plants and vegetable life all over the earth. The various lines of botanical research described in the foregoing pages, with their particular problems and objects, have but slight connection one with another. They run side by side along separate paths, and it is only occasionally that a junction is apparent which establishes a communication between one path and another. The subject-matter, however, is always the same. Whether we have to do with the perfected form or with its growth, whether we try to interpret the processes of life or to trace the genealogy of the vegetable kingdom, we always start from the forms of plants; and the ultimate result is never anything more than a description of the varying impressions which we receive at different times from the objects observed, and which we endeavour to bring into mutual connection. All the different departments of botany are accordingly more or less limited to description; and even when we endeavour to resolve vital phenomena into mechanical processes we can only describe, and not really explain, what happens. The processes which we call life are movements. But the causes of those move- ments, so-called forces, are purely subjective ideas, and do not involve the concep- tion of any actual fact, so that our passion for causality is only ostensibly gratified by the help of mechanics. Du Bois Reymond is not far wrong when he follows out this train of thought to the conclusion (however paradoxical it may sound) that there is no essential difference between describing the trajectory (or particular kind of curve) in which a projectile moves on the one hand, and describing a beetle or the leaf of a tree on the other. But even though the ultimate sources of vital phenomena remain unrevealed, the desire to represent all processes as effects, and to demonstrate the causes of such effects — a desire which is at the very root of modern research — finds at least partial gratification in tracing a phenomenon back to its proximate cause. In the mere act of linking ascertained facts together, and in the creation of ideas involv- ing interdependence among the phenomena observed, there lies an irresistible charm which is a continual stimulus to fresh investigations. Even though we be sure that we shall never be able to fathom the truth completely, we shall still go on seeking to approach it. The more imaginative an investigator the more keenly is he goaded to discovery by this craving for an explanation of things and for a solution of the mute riddle which is presented to us by the forms of plants. It is impossible to overrate the value and eflficiency of the transcendent gift of imagination when applied to questions of Natural History. Thus when we inquire whether certain characters noted in a plant are hereditary, constant, and inalienable, or are only occasioned by local influences of climate or soil, and hence deduce whether the plant in question is to be looked upon as a species or a variety; when we conclude from the fact of a resemblance between the histories of the develop- ment of various species that they are related, and place them together in groups and series; when we unravel the genealogies of different plants by comparing forms still living with others that are extinct; when we try to represent clearly 18 THE STUDY OF PLANTS IN ANCIENT AND MODERN TIMES. the molecular structure of the cell -membrane by arguing from the phenomena manifested by that membrane; when we investigate the meaning of the peculiar thickenings and sculpturings of the walls of cells, or when we discover the strange forms of flowers and fruits to be mechanical contrivances adapted to the forms of certain animals, and judge the extent to which these contrivances are advan- tageous, or the reverse, to the plants — in all these and similar investigations imagination plays a predominant part. Experiment itself is really a result of the exercise of that faculty. Every experiment is a question addressed to nature. But each interrogation must be preceded by a conjecture as to the probable state of the case; and the object of the experiment is to decide which of the preliminary hypotheses is the right one, or at least which of them approaches nearest to the true solution. The fact that when the imagination has been allowed to soar unre- strained, or without the steadying ballast of actual observations, it has frequently led its followers into error, does not detract at all from its extreme value as an aid to research, notwithstanding the fact that it is responsible for the wonderful fantasies of nature-philosophy of which a few specimens have been given. Nor should we esteem it the less because enlargements of the field of observation and improvements in the instruments employed have again and again led to the sub- stitution of new ideas for those which careful observers and experimentalists had arrived at by collating the facts ascertained through their labours. For the same reasons it is unfair to regard with contempt the ideas of plant- life formed by our predecessors. It should never be forgotten how much smaller was the number of observations upon which botanists had to rely in former times, and how much less perfect were their instruments of research. Every one of our theories has its history. In the first place a few puzzling facts are observed, and gradually others come to be associated with them. A general survey of the phenomena in question suggests the existence of a definite uniformity underlying them; and attempts are made to grasp the nature of such uniformity and to define it in words. Whilst the question thus raised is in suspense, botanists strive with more or less success to answer it, until a master mind appears. He collates the observed facts, gathers from them the law of their harmony, generalizes it, and announces the solution of the enigma. But observations continue to multiply; scientific instruments become more delicate, and some of the newly-observed facts will not adapt themselves to the scheme of the earlier generalization. At first they are held to be exceptions to the rule. By degrees, however, these exceptions accumulate; the law has lost its universality and must undergo expansion, or else it has become quite obsolete and must be replaced by another. So it has been in all past times, and so will it be in the future. Only a narrow mind is capable of claiming infallibility and permanence for the ideas which the present age lays down as laws of nature. These remarks on the limitations of our knowledge of nature, the importance of imagination as an aid in research, and the variability of our theories are made with a view to moderate, on the one hand, the exuberant hopes raised by the belief THE STUDY OF PLANTS IN ANCIENT AND MODERN TIMES. 19 that the great questions connected with the phenomenon of life will be solved, and to correct, on the other, the habit of not appreciating impartially the various methods which have been and are still employed by different botanists. In our own time, adhering as we do to the principle of the division of labour, it has become almost universal for each investigator to advance only along a single, very narrow path. But owing to the fact that one-sidedness too often leads to self-conceit, the lines of study followed by others are not infrequently despised, just as overweening confidence in the infallibility of the discoveries of the present day leads to deprecia- tion of the labours of former times. For the building-up of the science of the Biology of Plants everything relating to the subject has its value, and is capable of being turned to account. Whether the materials are rough or elaborated, massive, fragmentary, or merely connective, howsoever and whensoever they have been acquired, they all are useful. The study of dried plants made by a student in a provincial museum, the discoveries of an amateur regarding the flora of a sequestered valley, the contributions of horticul- turalists on subjects of experiment, the facts gleaned by farmers and foresters in fields and woods, the disclosures which have been wrested from living plants in university laboratories, and the observations conducted in the greatest and best of all laboratories — that of Nature herself — all these results should be turned to account. Let us take for the motto of the following pages the text: "Prove all things; hold fast that which is good." THE LIVING PRINCIPLE IN PLANTS, 1. PROTOPLASTS CONSIDERED AS THE SEAT OF LIFE. Discovery of the Cell. — Discovery of Protoplasm. DISCOVERY OF THE CELL. What is life? This ever-interesting question has seemed to approach nearer solution on the occasion of every great scientific discovery. But never did the hope of being able to penetrate the great secret of life appear better founded than at the time when, among other memorable developments of science, it was discovered that objects could be rendered visible on an enlarged scale by the use of glass lenses, and the microscope was invented. These magnifying glasses were expected to yield, not only an insight into the minute structure of living beings which is invisible to the naked eye, but also revelations concerning the processes which constitute life in plants and animals. The first discoveries made with the microscope, between 1665 and 1700, produced a profound impression on the observers. The Dutch philosopher Swammerdam became almost insane at the marvels revealed by his lenses, and at last destroyed his notes, having come to the conclusion that it was sacrilege to unveil, and thereby profane, what was designed by the Creator to remain hidden from human ken. The observations of Leeuwenhoek (1632-1723) with magnifying glasses formed by melting fine glass threads in a lamp, were for a long time held to be delusions; and it was not till the English observer Robert Hooke had confirmed the fact of the existence of the minute organisms seen by Leeuwenhoek in infusions of pepper, and had exhibited them under his microscope in 1667 at a meeting of the Royal Society in London, that doubts as to their actual existence disappeared. Indeed a special document was then drawn up and signed by all those who were satisfied, on the evidence of their own eyesight, of the accu- racy of the observation; and this clearly shows how greatly people were impressed with the importance of these discoveries. Of the difierent forms of the tiny organisms, amounting to nearly four hundred, which were at that time distinguished, and all included under the name Infusoria, because first seen in infusions of pepper- corns, some only are at the present day reckoned as animals. In many cases it has been ascertained that they are the spores of plants, whilst others again belong to the boundary-land where the animal and vegetable kingdoms are merged. The presence or absence of movement used to be considered as the most decisive mark of the difierence between animals and plants, and, accordingly, all the minute 22 DISCOVERY OF THE CELL. beings which were seen bustling about in watery media were described and labelled as animals. No movement was found in the higher plants which were studied with the microscope about the same time by Dutch, Italian, and English observers; but, on the other hand, these investigations led to a recognition of the quite special peculiarities of such structures as leaves and stem, wood and pith. These parts of plants appeared under the microscope like honey-combs, which are built up of a ^""-i^S^zisJjii-iS^ Fig. 4.— Vegetable Cells (from Grew's Anatomy of Plants). Longitudinal section through a young apricot seed. * Transverse section of the petiole of the Wild Clary. » Transverse section of a pine branch. great number of cells, some empty and some full of honey. From this similarity the term "cell" arose, which later was to play so important a part in botany. In the drawings of parts of plants as seen under the microscope the resemblance to a honey-comb is very apparent; indeed, it is sometimes rather more striking than when seen in reality, as, for instance, is the case in the above reproduction of three engravings from Nehemiah Grew's fine work published in London, 1672. It was also noticed that, besides the structures which resembled honey-comb, there were little tubes and fibres which were distributed and aggregated in very various ways, and were bound up together into strands and membranes, and into pith and wood; further, all these things were seen to increase in size and number in the growing DISCOVERY OF THE CELL. 23 parts of plants. How growth and multiplication took place, and where exactly the seat of a plant's life lay, remained, of course, obscure. It was, however, natural to assume that the walls of these small cells constituted the essential part and living substance of plants, that they drew materials from the fluids which rose by suction in the tubes, and so increased in size and were renewed. It was as yet hardly suspected that the slimy substance which filled the cells of a plant, like honey in a honey-comb, was the basis of life. The observation made again and again at the beginning of the nineteenth century, that the cell-contents of certain algae are extruded in the form of globules of jelly, and that each globule moves independently and swims about in the water for a time, but then comes to rest and becomes the starting-point of a new alga, might undoubtedly have led to this conclusion. The accounts of these occurrences were, however, considered incredible by the majority of contemporary observers; and it was not till recently, when Unger established the phenomenon as an indubitable fact, that a proper estimation of its value was accorded. In the year 1826 this botanist investigated under the microscope a water-weed found at Ottakrinn, near Vienna, which had been described by systematic writers as an alga, and named Vaucheria clavata. To the naked eye it appears like a dense plexus of dark-green irregularly branched and matted filaments. These filaments, when magnified, are seen to be tubular cells which wither and die away at the base whilst growing at the apex, and developing sac-like branches laterally. (Fig. 25a.) The free ends of these tubes are blunt and rounded. The substance they contain is slimy, and, though itself colourless, is studded throughout with green granules; whilst near the blunt end of each filament these green particles are so closely packed that the entire contents of that part appear of a dark-green colour. Now, there comes a time in the life of every one of these filaments when its extremity swells and becomes more or less club-shaped. The moment this occurs, the dark-green contents withdraw somewhat from the extremity, leaving it hyaline and transparent. Almost simultaneously the contents of the swollen part of the tube nearest the apex become transparent, whilst further down the colour becomes very dark. (Figure 25a, a.) Twelve hours after the commencement of this change, that portion of the tube's contents which occupies the club-shaped end separates itself entirely from the rest. A little later, the cell-wall at the apex of the tube suddenly splits, the edges of the slit fold back, and the inclosed mass travels through the aperture (fig. c). This jelly-like ball, having a greater diameter than the hole, is at first strangulated as it struggles forward, so that it assumes the shape of an hour-glass and looks for an instant as if it would remain stuck fast. There now arises, however, in the entire mass of green jelly an abrupt movement of rotation combined with forward straining, and in another instant it has escaped through the narrow aperture and is swimming freely about in the surroundmg water (fig. d). The entire phenomenon of the escape of these bodies takes place between 8 and 9 a.m., and, in any one case, in less than two minutes. When free, each individual assumes the shape of a perfectly regular ellipsoid (fig. d), having 24) DISCOVERY OF THE CELL. one pole of a lighter green than the other; it moves always in the direction of the former, so that the lighter end may be properly designated the anterior. At first the ball rises to the surface of the water towards the light, but soon after it again sinks deep down, often turning suddenly half-way round and pursues for a time a horizontal course. In all these movements it avoids coming into collision with the stationary objects which lie in its path, and also carefully eludes all the creatures swimming about in the same water with it. The motion is efiected by short pro- cesses like lashes or "cilia," which protrude all round from the enveloping pellicle of the jelly-like body and are in active vibration. With the help of these cilia, which occasion by their action little eddies in the water, the whole ball of green jelly moves in any given direction with considerable rapidity. But at the same time as it pushes forward, the ellipsoid turns on its longer axis, so that the resultant motion is obviously that of a screw. It is worthy of note that this rotation is invariably from east to west, that is, in the direction opposed to that of the earth. The rate of progress is always about the same: a layer of water of not quite two centimetres (1'76 cm.) is traversed in one minute. Now and then, it is true, the swimming ellipsoid allows itself a short rest; but it begins again almost immediately, rising and sinking, and resumes its movements of rotation and vibration. Two hours after its escape the movements become perceptibly feebler, and the pauses, during which there is only rotation and no forward motion of the body, become both longer and more frequent. At length the swimmer attains permanent rest. He lands on some place or other, preferably on the shady side of any object that may be floating or stationary in the water. The axial rotation ceases, the cilia stop their lashing motion and are withdrawn into the substance of the body, and the whole organism, hitherto ellip- soidal and lighter at its anterior end, becomes spherical and of a uniform dark- green colour. So long as it is in motion the gelatinous body has no definite wall. Its outermost layer is, no doubt, denser than the rest; but no distinct boundary is to be recognized, and we cannot properly speak of a special enveloping coat. No sooner, however, is the ball stranded, no sooner has its movement ceased and its shape become spherical, than a substance is secreted at its periphery; and this substance, even at the moment of secretion, takes the form of a firm, colourless, and transparent membrane. Twenty-six hours afterwards, very short branched tubes begin to push out from the interior, and these become organs of attachment. In the opposite direction the cell stretches into a long tube which divides into branches and floats on the water. After fourteen days the free ends of this tube and of its branches swell once more and become club-shaped; a portion of their slimy contents is, as before, separated from the rest and liberated as a motile body, and the whole performance described above is repeated. *KK«W«EtWttH: DISCOVERY OF PROTOPLASM. 25 DISCOVERY OF PROTOPLASM. The study of Vaucheria led, then, to the discovery that there are plants which, in the course of their development, pass through a motile stage, propelling them- selves about the water as tiny balls of jelly with ciliary processes, and giving exactly the same impression as infusoria. Hand in hand with this discovery went the further observation that a portion of the plastic cell-contents in all plants lies, like a lining, in contact with the inner face of the cell- walls, so that we find that these latter, at a certain stage of maturity, are made up of two layers lying close Fig. 6.— Protoplasm Inclosed in Cells. Protoplasm in cells of Orobanche. a Streaming protoplasm in cells of Vallisneria. » Streaming protoplasm in cells of Elodea. together, the outer one firm and the inner soft. The name of " primordial utricle" was given to this inner layer. On further investigation it turned out that this primordial utricle belongs to a body of gelatinous, slimy consistency which lives in the cell-cavity like a mussel or a snail in its shell. At first it is shapeless and fills the whole cavity with what appears to be a homogeneous mass; but later on it is difierentiated into a number of easily- recognizable parts — i.e. into the above- mentioned lining towards the inner surface of the cell-membrane, and into folds, strands, threads, and plates stretching across the interior of the cell. (See fig. 5.) Mohl of Tubingen, the discoverer of these facts, applied in 1846 tho name of proto- plasm to the substance of which the cell-contents are composed. It is possible for protoplasm, under certain conditions, to exist for a time without any special protective envelope; but, as a general rule, it secretes at once a firm, 26 DISCOVERY OF PROTOPLASM. continuous coat, and, so to speak, builds itself a little chamber wherein to live. We may therefore distinguish naked protoplasm from that kind which inhabits the interior of a cell of its own creation, and compare the former to a shell-less snail, and the latter to a snail that constructs the house in which its life is spent. Still better may we compare the firm and solid cell-membrane with which the protoplasm clothes itself to a protective coat, a garment fitted to the body; and, following out this analogy, the protoplasm must be designated the living entity in the cell, and the secreted envelope must be considered as merely the skin of the cell. Conse- quently, although this cell-wall was the part which was first revealed by magni- fying glasses, and was called a cell on account of its form, this is not the essential formative element, which has the power of nourishing and reproducing itself. It is the body within the cell, the slimy, colourless protoplasm in full activity within the surrounding membrane made by itself, which must be taken to be the essential part of the cell and the basis of life. The term cell had become so naturalized in the science that protoplasm which had escaped from a cell-cavity was also called a cell, and the unfortunate name of " naked cell " was brought into use to designate it. More recently many of these older designations have been abandoned as unsuitable. We now include under the term "protoplasts" all these individual organisms, consisting of protoplasm, which occupy little chambers made by themselves, living either alone like hermits or side by side in sociable alliance in more or less extensive structures, able under certain circumstances to leave their domiciles, laying aside their envelopes and swimming about as naked globules. Only when the protoplasts live in innumerable little cavities congregated close together in colonies, and when these cavities are bounded by even walls and are for the most part uniformly developed in all directions, does the part of a plant com- posed of them look under the microscope like a honey-comb, and each cavity like a cell. But even in these cases of external similarity there is the essential diflference that in a honey-comb each of the walls separating individual cells is common to both the adjacent spaces, and, accordingly, the cells of the comb are like excavations in a continuous matrix; whereas, in sections of cellular plants, every cell possesses its own particular and independent wall, so that in them every partition-wall between neighbouring cavities is composed, properly speaking, of two layers (fig. 6). These two layers are scarcely distinguishable in the case of delicate cell-membranes newly secreted by the protoplasts. Later on, however, they are always to be made out clearly (fig. 62). Frequently the layers separate one from another at certain spots, and thus channels are formed between the cells (fig. 6 ^); these are called " inter- cellular spaces." One often sees cells, too, whose entire surfaces are, as it were, glued together with a kind of cement, and then this substance which is stored between the two layers is called "intercellular substance" (fig. 6^). By loosening the intercellular substance, where present, by mechanical or chemi- cal means, we can easily separate adjacent cells from one another; the two layers of the partitioning cell- walls come asunder, and then each separate cell exhibits a DISCOVERY OF PROTOPLASM. 27 complete envelope. The individual cell-cavities are often elongated and shaped like either rigid or flexible tubes; or the wall of such a cavity may become very thick and encroach to such an extent on the cavity that the latter is scarcely recognizable. Cells of this kind look like fibres and threads, groups of them look like bundles and strands, and do not resemble even remotely the cells of a honey-comb. The term " cellular " is hence no longer suitable in the case of these structures. The expression " cellular tissue " is calculated also to occasion a wrong idea of the grouping and connection of the single cell-cavities. By a tissue one would surely understand a collection of thread-like elements so arranged that some of the threads run parallel to one another in one direction, whilst similar threads crossing Fig. 6.— Cell-chambers. Showing Intercellular Spaces ( i and « ) and " Intercellular Substance " (S) in the Partition-walls of the Chambers. the first at right angles are interwoven with them. In such a tissue, as of woven silk or the web of a spider, the threads are held together by intertwining ; but this is by no means the case with the collections of cells which have been called cell- tissues. Even where the parts of a so-called tissue of cells are tubular, thread-like, or fibrous, they lie side by side and are joined as it were by a cement, but are never crossed' or twisted together like the threads in a woven fabric. Again, cells have been compared to the bricks of a building, but this analogy is not exact. The process of formation of a cubical crystal from a solution of common salt may perhaps be compared to the piling up of bricks; but when a leaf grows the process is not for one layer of cells to be superimposed from the outside upon another previously deposited. The development of new cells proceeds in the inside of exist- ing cells and ensues from the activity of the protoplasts inclosed within the cell- walls; and these protoplasts not only provide the building materials, but are them- selves the builders. It is in this very fact indeed that we grasp the sole distinction between organic and inorganic structures, and on this account especially the above analogy is inadmissible and should be avoided. Cells and cell-aggregates may be conceived most clearly by considering their analogy to the shells of living creatures, as we have already done more than once in the foregoing pages. Protoplasts are either solitary, inhabiting isolated cell-cavities; or else they live in associated groups, the cells being crowded close together in great numbers and firmly attached to one another — each cavity being inhabited by one such protoplast. When the latter is the case, division of labour usually takes place 28 SWIMMING AND CREEPING PROTOPLASTS. in a plant, so that, as in every other community, some of the members undertake one function, some another. The older cells in these plants often lose their living protoplasts, and then, for the most part, serve as an uninhabited foundation to the entire edifice, which may thus be penetrated by air and water channels. The proto- plasts have meanwhile erected new stories for themselves and their posterity on the old deserted foundations, and are pursuing their indefatigable labours in the little chambers of these upper stories. This work of the living protoplasts consists in absorbing nutriment, increasing their own substance, maturing ofispring, searching for the places which offer most favourable conditions with a view to an eventual transmigration and to colonization by their families; and lastly, securing the region where all these tasks are performed against injurious external influences. The sequence of these labours is always governed by conditions of time and place. Many of them are only to be observed with difficulty in their actual performance and are first recognized in their perfected products, while others are attended by very striking phenomena and are easily followed in their progress. 2. MOVEMENTS OF PROTOPLASTS. Swimming and creeping protoplasts. — Movements of protoplasm in cell-cavities. — Movements of Volvocinese, Diatomaceae, Osciilariae, and Bacteria. SWIMMING AND CEEEPING PEOTOPLASTS. Among the most striking phenomena observed in connection with living proto- plasts are, without question, the temporary locomotion of the protoplast as a whole and the displacement and investment of its several particles. The freest motion is of course exhibited by protoplasts which are not inclosed in cell-cavities, but have forsaken their dwelling and are wandering about in liquid media. Their number, as well as the variety of their forms, is extremely great. These naked protoplasts are evolved by several thousands of kinds of cryptogamic plants, at the moment of sexual or asexual reproduction in these plants. The escape from the enveloping cell- wall alone takes place in countless difierent ways, though the process, as a whole, is conducted in the manner already described in the case of Vaucheria clavata. Sometimes a single comparatively large protoplast glides out of the opened cell by itself; at other times, before the cell opens the protoplasmic body divides into several parts — often into a great number — and then a whole swarm of protoplasts struggle out. These swarming protoplasts differ considerably in form. Usually their outline is almost ellipsoidal or oval ; but pear-shaped, top-shaped, and spindle-shaped forms also occur. Often the body of the protoplast is spirally twisted like a corkscrew, and has in addition one end spatulate or clavate. Thread-like processes, definite in number and dimensions and arranged variously, according to the kind of protoplast. SWIMMING AND CREEPING PROTOPLASTS. 29 project from the surface of its body. In some instances the whole surface is thickly- covered with short cilia, as in Vaucheria (fig. 7 ^); in others the cilia form a close ring behind the conical or beak-like end of the pear-shaped body, as in (Edogonium (fig. 7 2); and in others again, one or two pairs of long and infinitesimally thin threads, like the antennae of a butterfly, proceed from some spot, generally the narrow end (fig. 7 ^ and 7 *). Many forms are provided with a single long lash or flagellum at one extremity (fig. 7^), and yet others are spirally wound and are beset with cilia, thus presenting a bristly or hirsute appearance (fig. 7 ^^). These ciliary processes have a combined lashing and rotatory motion, and by their means the protoplasts swim about in water. In many cases, however, swim- Fig. 7.— swimming Protoplasm. 1 Vaucheria; « CEdogonium; « Draparnaldia ; * Coleochcete; s and ? Botrydium; « Ulothrix; 8 Fucus; • Funaria; 10 Sphagnum; " Adiantum. ming is hardly an appropriate expression; certainly not if one associates the term with the idea of fishes swimming with fins. In point of fact there is, associated with progression in a particular direction, a continuous rotation of the protoplast round its longer axis, and on this account its motion may be compared to that of a rifle-bullet, since in both cases the movement of translation takes place in the direction of the axis round which the whole body spins. The movement in question is not unlike the boring of one body inside another; according to this, the soft protoplasts bore through the yielding water, and by this action make onward progress. The microscope magnifies not only the moving body, but also the path traversed; and when one contemplates a protoplast in motion, magnified, say, three hundred times, its speed appears to be three hundred times as fast as it really is. As a matter of fact, the motion of protoplasts is rather slow. The swarm-spores of Vauche7'ia, described above, which traverse a distance of 17 millimeters in a minute are amongst the fastest. The majority accomplish an advance of not more than 5 m.m., and many only 1 m.m. per minute. 30 SWIMMING AND CREEPING PROTOPLASTS. As was mentioned in the description of Vaucheria the locomotion of ciliated protoplasts lasts for a comparatively brief period. It gives the impression of being a journey with a purpose: a search, as it were, for favourable spots for settle- ment and further development; or else a hunt after other protoplasts moving about in the same liquid. Green protoplasts always begin by seeking the light, but after a time they swim back into the shadier depths. Many of these, especially the larger ones, avoid coming into collision, and are careful to give each other a wide berth. If numbers are crowded together in a confined space, and two collide or their cilia come into contact, the motion ceases for an instant, but in a few seconds they free themselves and retire in opposite directions. Contrasting with these unsociable protoplasts are others, which have a ten- dency to seek each other out and to unite; and protoplasm acts in many cases on protoplasm of identical or similar quality, perceptibly attracting it and deter- mining the direction of its motion. It is very curious to watch the tiny pear- shaped whirling protoplasts of Draparnaldia, Ulothrix, Botrydium, and many others, as they steer towards one another and, upon their ciliated ends coming into contact, turn over and lay themselves side by side (fig. 7^); or, to see one pursued and seized by another, the foreparts of their bodies brought into lateral contact, and, finally, the two, after swimming about paired for a few minutes, fusing together into a single oval or spherical protoplast (fig. 7 ^). Even the minute fusiform protoplasts which are moved by cilia proceeding from the sides of their bodies (fig. 7^), as well as the spirally -coiled forms (figs. 79-10-") endeavour to unite with some other protoplast. They always move towards larger protoplasmic bodies at rest, cling to them closely, and at last coalesce with them into single masses (fig. 7 ^). As a rule no striking change is to be perceived in the inside of motile proto- plasmic bodies during the rotatory and progressive motion caused by their cilia; and the granules and chlorophyll-corpuscles dotted about in the body of the protoplast seem to remain, throughout the period of locomotion, almost unchanged as regards both position and shape. It is only in the vicinity of certain little spaces, called "vacuoles," in the substance of the protoplasm, that changes in many instances are observed, which indicate that, during the motion of the whole apparently rigid mass, slight displacements may also occur in the interior, some- what in the same way as, when a man walks, the heart inside his body is not still (relatively to the body), but continues to pulsate and cause the blood to circulate. The changes observed in vacuoles have, moreover, been described as pulsations, because they are accomplished rhythmically and manifest themselves as alternate expansions and contractions of the vacant space. In each of the motile protoplasts of Ulothrix (fig. 8) there is found, near the conical end, which is furnished with four cilia, a vacuole which contracts in from 12 to 15 seconds, and dilates again in the succeeding 12 or 15 seconds. In the swarm-spores of Chlamydomonas and those of Draparnaldia two such vacuoles may be observed close together, whose rhythmic action is alternate, so that the SWIMMING AND CREEPING PROTOPLASTS. 31 systole (contraction) of the one always takes place synchronously with the diastole (expansion) of the other. The contraction often continues until the cavity entirely disappears. It must depend, as also does the expansion, on a displacement of that part of the protoplasm which immediately surrounds the vacuole. But such a motion as this in the protoplasmic substance, even if only visible in a small part of the whole body, can scarcely be without its effect on other more distant parts; and it may, therefore, be concluded that the interior of a protoplast, endowed with ciliary motion, rotatory and progressive, does not remain quite at rest relatively, as seems on cursory inspection to be the case. Protoplasts whose motion is effected by means of cilia have no more need of their vibratile organs when once they have reached their destination. The cilia. Fig. 8.— Pulsating Vacuoles in the Protoplasm of the large Swarm-spores of Ulothrix. whether numerous or solitary, whether short or long, first of all become stationary and then suddenly disappear. Either they are drawn in or else they deliquesce into the surrounding liquid. Whether the motile protoplasts have come to rest because they have reached a suitable place for further development, as happens in Vaucheria, or because they have united, like with like, into a single mass, the form taken by the resulting non-motile body is always spherical. The final act is the development around itself of an investing cell-membrane, so that its soft and slimy substance may be protected by a firm covering from external influences. Essentially difierent from the motion just described is that of certain proto- plasts which are unprovided with cilia, but perpetually change their outlines, thrusting out considerable portions of their gelatinous bodies in one direction or another, and at the same time drawing in other parts. At one moment they appear irregularly angular, shortly afterwards stellate; then, again, they elongate, become fusiform, and gradually almost round (fig. 9). The protruded parts are sometimes delicate, tapering off into mere threads; sometimes they are com- paratively thick, and have almost the appearance of arms and feet in relation to the principal mass. The motion is not in this case like boring, but is best described as creeping. As one or a pair of foot-like appendages is thrown out 32 MOVEMENTS OF PROTOPLASM IN CELL-CAVITIES. in one direction, others on the opposite side are retracted, and the protoplast as a whole glides over the intervening space like a snail without its shell. The analogy is all the more exact since the protoplast, as it glides onward, leaves a slimy trail in its wake, so that the latter is marked by a streak resembling the track of a snail. When two or more of these creeping protoplasts, or plasmodia, meet, they merge into one another, flowing together somewhat in the same way as two oil-drops on water coalesce into one — leaving no distinguishable boundaries between the united bodies. Thus, slimy lumps of protoplasm, which may attain to the dimensions of a closed or open hand, result from the coalescence of great numbers of minute protoplasts. And it is a very remarkable fact that these Plasmodia can themselves change their form, putting out lobes and threads, and rMK_ Fig. 9.— Creeping Protoplasm. creeping about in the same way as the single protoplasts from whose fusion they have arisen. Creeping masses of jelly sometimes move in the direction of incident light; at other times they avoid light and hide in obscure places, wriggling through the interstices of heaps of bark or into the hollows of rotten trunks; or they may creep up the stems of plants, or glide over the brown earth in a viscous condition. On these occasions they resolve themselves not infrequently into bands, cords, and threads, which surround fixed objects, divide, and combine again, forming a net- work of meshes, or else perhaps frothy lumps like cuckoo-spit. If foreign bodies of small size are enmeshed by the viscous threads of the reticulum, they may be drawn along by the protoplasm as it creeps; and if they contain nutritive material, they may be eaten up and absorbed. Plasmodia are, for the most part, colourless, but some are brightly tinted; in particular may be mentioned the best-known of all plasmoid fungi, the so-called "Flowers of Tan" (Fuligo varians), which are yellow, and Lycogala Epidendron, which comes out on old stumps of pines, and is vermilion in colour. MOVEMENTS OF PROTOPLASM IN CELL-CAVITIES. In the case of a protoplast which is not naked, but clothed with an attached ceU-membrane, the movements are limited to the space included by the membrane, that is to say to the cell-cavity. Until the protoplasmic cell-body is differentiated into distinct individual portions no very lively motion can in general take place LQ the coated protoplast; though it is not to be assumed that it abides completely MOVEMENTS OF PROTOPLASM IN CELL-CAVITIES. 33 at rest at any time, except perhaps during periods of drought in summer and of frost in winter, and in seeds during their time of quiescence. This applies par- ticularly to immature cells. In them the protoplast forms a solid body whose substance entirely fills the cell-cavity. The young cell, however, grows up quickly, its cavity is enlarged, and the space, hitherto filled by the protoplast, becomes two or three times as large as before. But the increase of volume on the part of the protoplast itself does not keep pace with the enlargement of its habitation. It is true that it continues to cling closely to the inner face of the cell- wall, thus forming the primordial utricle; but the more central part of its body relaxes, and in it are formed vacant spaces, the vacuoles above mentioned, wherein collects a watery fluid known as the "cell-sap." The portions of protoplasm which lie between the vacuoles resolve themselves gradually into thin partitions bounding them; and lastly, these partitions split up into bands, bridles, and threads, which stretch across the cell-cavity from one side of the primordial utricle to the other, and are woven together here and there where they intersect. With these protoplasmic strands we have already become acquainted. But the protoplasm in the interior of a growing cell, whilst relaxing and breaking up, also becomes motile if the liquid attams a certain temperature, and then the appearance presented is like that of a lump of wax melting under the action of heat. These movements may be observed very clearly under the micro- scope in the case of large cells with thin and very transparent cell-membranes, especially when the colourless, translucent, and gelatinous substance of the proto- plasm— not always sharply defined in contour — happens to be studded with minute dark granules, the so-called "microsomata." These granules are driven backwards and forwards with the stream, like particles of mud in turbid water, and their motion reveals that of the protoplasm wherein they are embedded. Seeing particles gliding in all directions through the cell-cavity, arranged irregularly in chains, rows, and clusters in the protoplasmic strands, we are justified in concluding that this motion takes place in the substance of the strands itself. The movement, moreover, is not confined to isolated strands, but occurs in all. Granular currents flow hither and thither, now uniting, now again dividing. They often run in opposite directions even when only a trifling distance apart; sometimes two chains are drifted in this way when actually close together in the same band of proto- plasm. The streams pour along the primordial utricle and whilst there divide into a number of arms, meeting and stemming one another and forming little eddies; then they are gathered together again and turn into another strand of the more central protoplasm. The individual granules in the currents are seen to move with unequal rapidity according to their sizes; the smaller particles progress faster than ihe larger, and the larger are often overtaken by the less, and when this happens the result often is that the entire stream stops. If so, however, the crowded particles are suddenly rolled forward again at a swifter pace, like bits of stone in the bed of a river as it passes from a level valley into a gorge. The course of the streaming protoplasm remains throughout sharply marked ofi" from the watery sap Vol. I. 3 34 MOVEMENTS OF PROTOPLASM IN CELL-CAVITIES. in the vacuoles, and none of the granules ever pass over into the cell-sap from the protoplasm. Larger bodies, such as the round grains of green colouring-matter or chlorophyll, are in many instances not carried forward, but remain stationary, the protoplasmic stream gliding over them without altering them in any way. Further, the outer- most layer of the protoplast, contiguous with the cell-membrane, is not in visible motion in most vegetable cells. On the other hand, occasionally the entire pro- toplast undoubtedly acquires a movement of rotation, and then the larger bodies imbedded in its substance, i.e. chlorophyll corpuscles, are driven along like drift- wood in a mountain torrent (fig. 5^ and 5^). On these occasions a wonderful circulation and undulation of the entire mass takes place: chlorophyll grains are whirled along one after the other at varying speeds as if trying to overtake one another; and yet another structure, the cell-nucleus presently to be discussed, is dragged along, being unable to withstand the pressure, and, following the various displacements of the net-work of protoplasmic strands in which it is involved, is at one moment pulled alongside of the cell-wall, at another again is taken in tow by a rope of central protoplasm and hauled transversely across the interior of the cell (fig. 5^). When the rate of the current itself is estimated by the pace at which the gran- ules are driven along, results which vary considerably are obtained, depending chiefly on a qualitative difference in the protoplasm, but secondarily also on temperature and other external conditions. A rise in temperature up to a certain point as a general rule accelerates the rate of the stream. Particles of protoplasm in particularly rapid motion pass over 10 m.m. in a minute; others in the same time traverse from 1 to 2 m.m.; and some, in still less haste, advance only about a hundredth part of a millimeter. Larger bodies, especially the bigger chlorophyll grains, move slowest of all. So it is often hours before chlorophyll grains lying near one side of a cell are pushed through the protoplasm over to the other side, a distance only equal to a small fraction of a millimeter. The minute granules, as well as the larger grains of chlorophyll and the cell- nucleus, are entirely surrounded by protoplasm ; and the protoplasm, whether in the form of bands or threads, whether a peripheral lining or an indefinite mass, must be conceived as always composed of two layers, the outer "ectoplasm" being tougher and denser than the inner "endoplasm," which is softer and somewhat fluid. The former is homogeneous and non-granular, so that it is the more transparent and has the effect of a skin clothing the inner, softer layer, which is granular and turbid. It would be incorrect, however, to think of this as a very strongly-marked contrast, sufficient to mark off* one layer clearly from the other. In reality there are no such sharp boundaries, and the tougher ectoplasm passes gradually into the softer and more mobile endoplasm. Of course the granules and corpuscles which one sees drifting in streaming protoplasm are situated within the more yielding endoplasm. It is true, minute particles often appear to glide from one side to the other upon a delicate protoplasmic strand as if it were a tight-rope; but on closer MOVEMENTS OF PROTOPLASM IN CELL-CAVITIES. 35 study it is apparent that the granules which seem to be travelling on the proto- plasmic thread are covered by a delicate and transparent protoplasmic pellicle. Thus, these granules imbedded in the substance of protoplasts have no independent motion, but are pushed along by the spreading protoplasm. Each stream of protoplasm is shut off from its environment and limited by a layer tougher than the rest. But this does not prevent the currents, with their crowds of drifting granules, from changing their direction. In fact we have only to follow for a short time the course of one such granular stream to remark a continuous series of changes: a current from being in a straight line bends suddenly to one side, it broadens and contracts again, now it runs close alongside another channel, now breaks away once more, divides into two little arms, and loses itself finally in the primordial utricle. On the other hand, fresh folds start from the primordial utricle, stretch and grow until they have pushed across the cell-cavity to the other side in the form of bands, or the protoplasm may be drawn out into threads, which elongate until they encounter other similar strings and form a junction with them. The same processes then that are observed in free creeping protoplasts take place to some extent here. Imagine a protoplast captured whilst on its travels — creeping along the level ground — and imprisoned in a completely closed vessel; it would spread itself out over the inner surface of the vessel, would branch and creep about and have just the same appearance as the protoplasts, just described, which inhabit cell-cavities from their earliest youth. This is but the converse of the power possessed by a protoplast set free from its cell, which enables it to move, stretch out, and draw in its various parts, and so to effect locomotion. Another motion, differing from the creeping, gliding, and streaming action of protoplasts, manifests itself in the so-called swarming of granules contained in the protoplasm. It may be best observed in the cells of the genera Penium and Closterium, both of which are shown in figure 25a, i, k, though the same phenomenon is to be seen in many allied forms, living in lakes and ponds either singly or congregated in colonies, and remarkable for their bright green colour. The above-mentioned genus Closteriurti includes delicate unicellular forms having a curved or scimitar shape unusual in plants, whence one of its species, in which the semi-lunar form is most striking, has been named Closterium lunula. The cell-membrane in all these little water-plants is clear and quite transparent. The greater part of the cell-contents consists of a dark -green chlorophyll body longitudinally grooved; but the protoplasm which is visible in the two sharply tapering ends of the cell-cavity is colourless, and embedded within it is a swarm of microsomata. These granules or microsomata appear to be in a most curious state of motion so long as the protoplast lives. They are to be seen plainly within the limits of the tiny cavity, jumping up and down, whirling, dancing, and rushing about without really changing their position. One is reminded of the apparently purposeless journeyings to and fro within reach of their homes of ants or bees, and the movement has been called not inaptly 36 MOVEMENTS OF PROTOPLASM IN CELL-CAVITIES. "swarming." It is difficult to imagine the kind of motion possessed by the protoplasm in which these swarming microsomata are embedded; but however closely it is confined, there must be continual rapid displacements in its substance, which is very fluid, and it may be assumed that here again it is not so much the tiny grains that bestir themselves as the protoplasm which holds them. Probably the protoplasmic matter spreads and stretches out and rotates, and individual granules are carried about by it. This, of course, does not exclude the possibility of the granules possessing a vibratory motion of their own within the mass of protoplasm. Similar, but not identical, is the swarming movement of protoplasm observed in cells of the Water-net {Hydrodictyon utriculatwni), and in several other plants allied to it. Hydrodictyon looks like a net in the form of a sac, and composed of green threads. The meshes of this net, which are generally hexagonal, consist, however, not of filaments but of slender cylindrical cells joined together by threes at their extremities, somewhat in the same way as are the leaden frames of the little hexagonal panes of glass in gothic windows. The protoplasmic body of one of these cells in due time breaks up into a great multitude (7000-20,000) of tiny clots, which begin to move and swarm within the cell-cavity in what appears to be a disordered medley. In half an hour, however, the excited mass is again restored to rest: the minute particles take form and arrange themselves in definite order, each having two others at either extremity, making an angle of 120° with it; and, lastly, all unite to form a single tiny net having exactly the same shape as the one whose component cell constituted the arena of this process of construc- tion. The miniature water-net so formed then slips out of the cell, the latter opening for the purpose, and in from three to four weeks it grows to the same size as the parent plant. In the above we have an instance of a protoplast producing a whole colony of cells, which are obliged to leave their home for want of space. In cases previously considered we have found the protoplast stretching and elongating in all directions, drawing itself out into bridles and spreading as a delicate lining to walls, and so endeavouring generally to expand and present the greatest surface possible. Again, we have seen it wandering freely, creeping, swim.ming, and rotating, and by this method also covering as much space as it can. But, con- versely, there is a time when a protoplast tends to the other extreme; the expanded mass of its body gathers itself together again, contracts more and more, and at length becomes a resting sphere, that is to say, it assumes the con- figuration which exposes the least surface to the environment. This process exhibits itself with particular clearness within the cell-cavities of the green algae known by the name of Spirogyra, a species of which is represented, magnified three hundred times, in figure 25a, I. In this alga the protoplasm in each mature cell-cavity forms, as a general rule, a very deli- cate parietal lining wherein green chlorophyll bodies are embedded, arranged in a spiral band. All of a sudden, however, this lining strips itself off the inner MOVEMENTS OF SIMPLE ORGANISMS. 37 face of the cell-wall and shrinks together so as in a short time to present the appearance of a sphere occupying the middle of the cell-cavity. Again, just as this contraction is an instance of a special form of protoplasmic motion, so also the further change which the contracted protoplast in a cell of Spirogyra under- goes is reducible to displacements in its substance, and must be mentioned as a special kind of protoplasmic movement. For the conglomerated protoplast remains but a short time in the middle of the cell-cavity. It leans almost immediately to one side, thrusting itself into a protuberance of the cell-mem- brane, which is concurrently developed, and which, when further developed, forms a passage leading over into another cell-cavity. Its body becomes longer and narrower, and at last slips through the passage into the next cavity, where a second protoplast awaits it; and the two then unite, fusing together into one mass. It is not premature to remark that all these displacements and invest- ments of the protoplasmic substance in cells of Spirogyra, including the pheno- mena of contraction, as well as those of pushing forward, escape, and coalescence, are not produced as the results of a shock, impulse, or stimulus from without, but are to be looked upon as movements proper to the protoplasm, and resulting from causes inherent in the protoplasm. MOVEMENTS OF VOLVOCINE^, DIATOMACE^, OSCILLARI^ AND BACTEEIA. Very remarkable is the movement of those wonderful organisms which are comprised under the name of Volvocineae. One species, Volvox globator, was known to so ancient an observer as Leeuwenhoek; but he, and after him Linnaeus, took it to be an animal on account of its extraordinary power of locomotion, and it was named the "globe-animalcule." A Vol vox-sphere consists of a large number of green protoplasts living together as a family and arranged with great regularity within their common envelope. They appear to be disposed radially, and to be linked together and held firm by a net-work of tough threads, their poles being directed towards the centre and the periphery of the sphere respectively. From the peripheral extremity, which in each protoplast is marked out by a bright red spot, proceed a pair of cilia, and these protrude through the soft gelatinous envelope of the whole sphere, and move rhythmically in the surrounding water. A Volvox-globe rolls along in the water propelled by regular strokes, like a boat manned by a number of oarsmen, as soon as the protoplasts, which form the crew of this strange vessel, begin to manipulate their propellers. The effect is exceed- ingly graceful, and has justly filled observers of all periods with astonishment; indeed no one seeing for the first time a Volvox-sphere rolling along can fail to be impressed and delighted. Another plant allied to the foregoing, the so-called "red-snow," has always excited wonder in no less degree from the remarkable phenomena of motion which it exhibits, but also because of its characteristic occurrence in situations where one 38 MOVEMENTS OF SIMPLE ORGANISMS. might suppose all vital functions would be extinguished. It was in the year 1760 that De Saussure first noticed that the snowfields on the mountains of Savoy were tinged with red, and described the phenomenon as "red-snow." Once on the look-out for it, people found this red-snow on the Alps of Switzerland, Tyrol, and the district of Salzburg, on the Pyrenees, the Carpathians, and the northern parts of the Ural Mountains, in arctic Scandinavia, and on the Sierra Nevada in California. But red- snow has been seen on the most magnificent scale in Greenland. When Captain John Ross in 1818 sailed round Cape York on his voyage of discovery to Arctic America, he noticed that all the snow patches lying in the gorges and gullies of the cliffs on the coast were coloured bright crimson; and the appearance was so start- ling that Ross named that rocky sea-shore the "Crimson Cliffs." On the occasion of later expeditions to the arctic regions, red-snow was observed off' the north coast of Spitzbergen, and in Russian Lapland and Eastern Siberia, but never in such sur- prising luxuriance as on the Crimson Cliff's of Greenland. If a snow-field coloured by red-snow is examined near at hand it is found that only the most superficial layer, about 50 millimeters in depth, is tinged. It is also present in the greatest quantities in places where the snow has been temporarily melted by the heat of summer, particularly therefore in depressions, whether big or little, and towards the edges of the snow-field, where the so-called snow-dust or Cryoconite extends regularly in the form of dark, graphitic smeary streaks. Exam- ined under the microscope, the matter which causes the redness of the snow appears as a number of spherical cells having a rather substantial colourless cell- membrane and protoplasmic contents permeated by chlorophyll. The green colour of the chlorophyll is, however, so disguised by a blood-red pigment that it is only possible to detect it when the latter has been extracted, or in cases where it is limited to a few definite spots in the cell. These spherical cells do not move, and so long as the snow is frozen they show no sign of life. But as soon as the heat of the summer months melts the snow, these cells acquire vitality, visibly increasing in size and preparing for division and multiplication the moment they have attained a certain volume. The growth, so far as it depends on nutrition, takes place at the expense of carbon dioxide absorbed by the melted snow from the atmosphere and of the inorganic and organic constituent parts of the dust. We shall frequently have occasion to return to this dust, but at present it is only neces- sary to observe, for the comprehension of the drawing of red-snow as seen under the microscope (figure 25a, e-K), that in the Alps, amongst the organic materials which constitute the dust, pollen-grains of conifers occur with great frequency, especially those of the fir, arolla, and mountain pine. These pollen-grains have been swept up into the high Alps by storms, and are already partially decayed. In all the material that I investigated I found the red-snow cells mixed with pollen-grains of the above-mentioned conifers. The pollen-grains are oval in cross- section, of a dirty yellow colour, and swollen laterally into two hemispherical wings, as is shown in figure 25a, e-Ti. As has been stated, the red cells are nourished by the constituent elements of MOVEMENTS OF SIMPLE ORGANISMS. 39 the dust, which are dissolved in the melted snow. They grow and at last divide so as to form daughter-cells, usually four in number but often six or eight and less frequently two only (figure 25a, /, g). As soon as the division is accom- plished, the daughter-cells, so produced, free themselves, assume an oval shape, and display at their narrower extremity two rotating cilia by means of which they move about in snow-water with considerable vivacity. The interstices of the still unmelted, but now granular, snow, are filled with water from the melted parts, and through these the red cells swim away and are thus diffused over the snow-field. At the moment of escape and first assumption of movement the cell-body appears to be uninclosed. But it soon clothes itself with an extremely delicate, though clearly discernible skin, which, curiously enough, does not lie close to the proto- plasm, which is withdrawn slightly and inclosed as in a distended sac (see figure 25a, e). Only in front, where the two cilia carry on their whirling motion, does the skin lie close to the body of the cell; and it must be presumed that the cilia, which are simply extensions of the protoplasmic substance, are projected through the envelope. The swarm-spores afford an example of an unusual type of protoplasts, namely of those that move about singly in the water by means of cilia and at the same time carry their self-made cell-membranes with them. How long the motile stage lasts under natural conditions has not been deter- mined for certain. On the mountains of central and southern Europe, where hot days are followed, even in the height of summer, by bitterly cold nights, causing the melted snow which has not run off to freeze again in the depressions of the snow, the movement no doubt is often interrupted. On the other hand, in high latitudes, where the summer sun does not set for weeks together, such interruption would be exceptional. In any case, however, the locomotion of the red cells with their hyaline cell-membranes is not limited to so short a period as is that of naked ciliated protoplasts. Moreover they have the power of nutrition and growth like the red resting-cells from which they originate, and they have been observed, in a culture, to increase in size fourfold within two days. When at last they come to rest they draw in their cilia, assume a spherical shape, thicken their cell-membrane, which now once more lies close to the protoplasmic body, and divide anew into two, four, or eight cells (figure 25a, f,g). The fusion of the protoplasts of the red cells in pairs, and their sexual propagation, which has been observed in addition to the above-described asexual multiplication, will be the subject of discussion later on. At present we need only add with reference to this remarkable plant that it was named Sijhcerella nivalis by the botanist Sommerfelt, and that not only in mode of life, but also in form and colour, it most closely resembles a kind of blood-red alga, which makes its appearance in Central Europe in little hollows temporarily filled with rain-water in flat rocks and slabs of stone, and also inside receptacles exposed to the open. This alga has received the name of Sphcerella phivicdis, and also that of Hcematococcus pluvialis. Lastly, we have to consider the mysterious movements exhibited by many Diatomacese, and by the filamentous species of Zonotrichia, Oscillaria, and 40 MOVEMENTS OF SBIPLE ORGANISMS. Beggiatoa. As regards the Diatoms, some of them are firmly attached to a support, and are not generally capable of locomotion; but others are almost in- cessantly in motion, and these little unicellular organisms steer themselves about with great precision near the bottom of the pools of water in which they live. Their cell-membrane is transformed into a siliceous coat, and this coat, which is hyaline and transparent, but very hard, consists of two halves shutting together like the valves of a mussel. The entire cell thus coated has the form of a gondola or little boat, with a keel either straight or curved (Pleurosigma, Pinmdaria, Navicula), and is provided with various bands, ribs, and sculpturings on its siliceous walls. Driven by inherent forces, these little protected cruisers pursue their way at the bottom of the water or over objects which happen to be in the water. They either glide evenly over the substratum, or else proceed by fits and starts at rather long intervals, and apparently with difficulty. For some time they may hold a straight course, but not infrequently they deviate side- ways without apparent cause, and after deviating return again. They double round projecting objects or push them out of the way with one of their hard points, which are often thickened into nodules, and cause the obstructing objects to slip by alongside the keel of the little vessel. Yet no paddles or cilia are to be seen projecting from it, as in the case already described of Volvocineae; nor does the siliceous coat exhibit any sort of motile processes whereto the move- ments might be attributed. But the strong analogy between the structure of these Diatomacese and that of mussels seems to justify the assumption that the two siliceous valves, which are fast shut during the period of rest of the Diatoms in question, move a little apart, so that the protoplast living within can push out one edge of its body and creep along over the substratum by means of it. The movements of the filaments of Beggiatoa, Oscillaria, and Zonotrichia are explained in a similar manner. These filaments are made up of a number of short cylindrical or discoid cells, and are attached by one end, but with the other execute most striking movements. They stretch themselves and then contract again, coil up and straighten out like snakes, and, most characteristic of all, make periodic oscillations in the water. The belief is that the mechanism of this motion is similar to that of the preceding, that infinitesimally fine fila- ments of protoplasm inserted spirally penetrate the cell-walls, and that these act like the propeller of a ship. On looking back over the multifarious examples of movement that have been described, the conviction that the capacity for motion is inherent in all living protoplasts is difficult to resist. In many cases, of course, the displacement and replacement of the substance no doubt takes place so slowly that it is scarcely possible to express its amount numerically. Movement may even entirely cease for a time; but, as necessity arises, and under favourable external circumstances, the protoplasmic mass always becomes mobile again — the direction of its motion being determined by inherent forces. There is still much to learn, no doubt, con- cerning the objects and significance of the different movements of protoplasm; CELL CONTENTS. 41 but in this connection we are justified in assuming that all these movements have to do with the maintenance and multiplication of the protoplasts. For instance, amongst the objects of the various movements are the search for food, the elimination of useless material, the production of offspring, the discovery of the rays of sunlight necessary to the existence of chlorophyll-bodies and of suitable spots to colonize. This conception has been hrought out frequently in the course of the foregoing description, and will again engage our attention in succeeding pages. 3. SECRETIONS AND CONSTRUCTIVE ACTIVITY OF PROTOPLASTS. Cell-sap. — Cell-nucleus. — Chlorophyll-bodies. — Starch. — Crystals. — Construction of the Cell-wall and Establishment of Communication between Neighbouring Cell-cavities. CELL-SAP.— CELL-NUCLEUS.— CHLOROPHYLL-BODIES.— STAECH.— CRYSTALS. In addition to the powers which the living protoplast possesses of shifting its parts, of expanding and contracting, of dividing and of fusing like with like, it has also the properties of adapting different parts of its body to particular functions, of building up various chemical compounds, and of separating them out when necessary. As the protoplast stretches and expands, spaces and depressions arise within it, and these form ultimately, when the protoplast is limited to a peripheral layer lining the walls of the cavity, a single central vacuole. In the spaces there is secreted, in the first instance, the cell-sap, a watery fluid containing a variety of substances either suspended or in solution, of which the chief are sugar, acids, and colouring matters. Moreover, in the interior of the protoplasm itself, structures with quite different forms occur, and are easily recog- nizable by their contours; these are the cell-nucleus, chlorophyll-bodies, and starch- grains. The principal feature of the cell-nucleus is that, although the substance of which it is composed is only slightly different from the general protoplasm of the cell, yet it is always clearly marked oflf from the protoplasm. In the un- developed protoplast the nucleus is usually situated in the middle, but in mature protoplasts it is either pressed against one wall of the cell or suspended in a sort of pocket of protoplasmic filaments in the interior (fig. 5 ^ and 5 ^ ). It may be pushed along by the streaming protoplasm and dragged into the middle of the cell, and in that case its shape is sometimes altered and it becomes for a time somewhat elongated and ffattened. The nuclear substance, which, as has been already mentioned, differs but little from ordinary protoplasm, is colourless, and studded with microsomata, and is liable to internal displacements similar to those of the entire cell-body. When a protoplast divides, the nucleus plays a very 42 THE CELL-WALL. important part in the process, and it will be necessary later on to discuss its significance in this connection. The chlorophyll-bodies, mentioned already more than once incidentally, are green corpuscles, roundish, ellipsoidal, or lenticular in shape, and grouped in a great variety of ways (figure 25a, i, k, I, m, p). They are produced generally in great numbers by the protoplast in special sac-like excavations in its body but nowhere except where they are necessary, that is, in those cells wherein the transmutation of inorganic food-stufis into organic matter takes place. This transformation, so important to the existence of the organic world, will be con- sidered in detail later on. Chlorophyll-corpuscles are not, as regards their material basis, essentially different from the substance of the protoplasm in which they are formed, and in which they remain embedded for life, but their green colour distinguishes them very clearly from their environment. This greenness is due to a colouring matter stored in the protoplasmic substance of the corpuscle; and our ideas of plant-life are so intimately associated with this remarkable pigment, that a plant that is not green seems to us to be almost an anomaly. Besides the nucleus and the chlorophyll-bodies or corpuscles, protoplasts pro- duce starch-grains, aleurone-grains, crystals of oxalate of lime, and drops of oil, all of which will be dealt with presently in their proper place. They are evolved in accordance with the requirements of the moment and with the position held in the edifice of the plant by the cells concerned. Moreover, the walls of the cells them- selves are the work of the protoplasts, and it is not a mere phrase, but a literal fact, that the protoplasts build their abodes themselves, divide and adapt the interiors according to their requirements, store up necessary supplies within them, and, most important of all, provide the wherewithal needful for nutrition, for maintenance, and for reproduction. CONSTRUCTION OF THE CELL- WALL AND ESTABLISHMENT OF CONNECTIONS BETWEEN NEIGHBOUEING CELL-CAVITIES. Of all these performances, the construction of the cell- wall shows the greatest variety from the nature of the case. For the envelope with which each individual protoplast surrounds itself serves at once as a protection for the delicate protoplasm, and as a firm support for structural additions; and, at the same time, it must not impede the reciprocal action betw^een the protoplasts and the external world, or the intercourse between those living in adjoining cavities. These cell-walls are accord- ingly very wonderful structures, and we shall often have occasion to discuss them, especially with reference to the significance of variations in their structure in particular cases. At present it is sufficient to remark that the original envelope which is secreted from the body of a protoplast and which appears at first as a delicate skin, is made of a substance composed of carbon, hydrogen, and oxygen, belonging to the class of carbohj^drates. The name of cell-membrane, usually applied to the original envelope formed by THE CELL-WALL. 43 the cell-body, is one quite suitable for the purpose. But this earliest covering under- goes many modifications. The protoplast is able to store up in it suberin, lio-nin, silica, and water in greater or smaller quantities, and by this means it either makes the envelope more flexible than it was in the first instance, or else hard and stifi", converting it into a shell-like case. Even the shape is seldom preserved as it was originally. The solitary protoplast surrounded by its cell-membrane is gener- ally in the form of a roundish ball, and its envelope, which is closely adherent, exhibits a corresponding configuration. Young cells, aggregated together, have outlines too which remind one of crystalline forms, such as dodecahedra, cubes, and short six-sided prisms. But when a protoplast has produced its first delicate covering it does not come to rest, but goes on working at the membrane, distending and thickening it, transforming a cavity which was originally spherical or cubical into one of cylindrical, fibrous, or tabular shape,' and strengthening its walls with pilasters, borders, ridges, hooks, bands, and panels of various kinds. Where a number of protoplasts work gregariously at one many-chambered edifice, cells of most diverse forms are produced in close proximity to one another. These varieties are, however, never without method and design, but are invariably such as to adequately equip each cell for the position it holds and for the particular task allotted to it in the general domestic economy. The volume attained by cell-cavities in consequence of the expansion of their walls varies within very wide limits. The smallest cells have a diameter of only one micro-millimeter, i.e. the thousandth part of a millimeter; others, as for example yeast-cells, measure perhaps two or three hundredths of a millimeter; and yet others have outlines perceptible to the naked eye and have a volume amounting to one cubic millimeter. Tubular and fibrous cells often stretch longitudinally to such an extraordinary extent that some with a diameter of scarcely the hun- dredth part of a millimeter reach a length of one, two, or even as many as five centimeters. An instance may be seen in the filaments of Vaucheria clavata (figure 25a, a-d), and again in the fibrous cells from which our linen and cotton fabrics are manufactured. The enlargement of a cell-cavity, or, in other words, the growth in area of its walls, ensues in consequence of the intercalation of fresh particles between those which, by their mutual coherence, form the delicate skin of the protoplast — the earliest stage of the cell-wall. When these intercalated particles are situ- ated in the same plane as are those already deposited, the cell-wall resulting from this method of construction will increase in area without adding to its thickness. But when once the cells are full-sized, the constructive activity of the protoplasts has to be directed in many cases to the strengthening and thick- ening of their walls, so that later on they may be able to perform special duties. From the appearance of this thickening one would judge that a number of layers were deposited on the thin original wall according to requirement, and in many instances no doubt the process corresponds to this appearance; but, as a rule, the thickness of the wall is increased by intercalation, on the part of the protoplasts, of 44 THE CELL-WALL. additional material between the original particles, a process which has been termed " intussusception." The appearance of stratification in thickened cell-walls is naturally most strik- ing where substances of different kinds have been deposited alternately in the different parts of the wall, and when successive layers take up unequal quantities of water. The thickening may at length result in such an extreme restriction of the cell-cavity that its diameter is less than that of the inclosing wall. Sometimes nothing remains of the cavity but a narrow passage, and then the cells are like solid fibres. Formerly they would not have been classed with cells at all, but would have been distinguished under the name of fibres, from the forms resembling honey-comb cells. The protoplasts in these contracted cells languish and often die, especially when the walls of the self-made prison are greatly thickened and do not allow of intercourse with the world outside. But generally a protoplast takes care, in constructing its dwelling, not to close itself in entirely, nor to cut itself off permanently from the outer world. It either makes from the very beginning little windows in the walls of its house, leaving them quite open or closed only by thin, easily-permeable, membranes; or else, after constructing a completely closed enve- lope, it redissolves a piece of it, thus making an aperture through which in due time it is able to effect its escape. The scope of this work does not admit of an exhaustive treatment of the formative power possessed by protoplasts needful for these results; it will be sufficient to give a general description of some of the more important processes which have for their object the establishment of a connection between adjacent cell-cavities and of communication with the external world. The new particles of material, or cellulose, which are to strengthen the delicate original cell-membrane, are in many instances not deposited or intercalated evenly over the entire surface of the protoplast. Little isolated spots are left unaltered, and these may be compared in a way to the small glazed windows in a living-room, or cabin port-holes closed by thin panes of glass. The part of the thickened wall which immediately surrounds the little window, and which so to speak constitutes its frame, has, besides, often a very characteristic structure, being elevated so as to form first a ring - like border, and eventually a hood, arching over the window and perforated in the middle (see fig. 10 ^). A comparison of this structure, arched over the thin spots in a cell-wall, to the iris spread in front of the crystalline lens in an eye would be still more appropriate. A similar annular border projects likewise from the window -frame on the other side, facing a neighbouring cell-cavity, so that the window appears symmetrically vaulted on both sides by mouldings with round central apertures (fig. 10 ^). Supposing someone wanted to pass from one cell-cavity to the other he would have in the first place to go through the hole in the moulding on his side. He would then find himself in a roomy space, which we will call the vestibule, and would next have to break through the little window, which is somewhat thickened in the middle, but elsewhere is as soft and thin as possible. On the further side THE CELL- WALL 45 again would be a vestibule, and it would not be until he had emerged from this through the aperture in the second moulding that he would reach the interior of the adjoining cell. Seen from in front, the outline of one of these windows, or rather the outline of the common floor of the vestibules, appears as a circle, whilst the aperture or opening in the moulding — which is exactly in the centre of this circle — is seen as a bright dot or pit encompassed by the circle which defines the limits of the vestibule. Hence these curiously protected window structures are named bordered pits. They are shown in fig. 10 ^ and 10 ^, and are to be seen in great perfection in the wood-cells of pines and firs. Whenever bordered pits are formed, the thickening of the cell -membrane is comparatively slight; the frame of the window in the cell-wall is never more than 0 ©1 ©1 ©' -^>-^ Fig 10.— Connecting Passages between adjacent Cell-cavities. Bordered pits. 2, Section of a bordered pit. 8_ Mode of connection of adjacent cells in the bundle-sheath *, Sieve-tubes. s_ Group of cells from seed of Nux-vomica, the protoplasts of adjoining cell-cavities protoplasmic filaments. of Scolojiendri connected by five times as thick as the window-pane itself. In other cases, however, the cell-wall becomes twenty or thirty times as thick as it was at first, and the interior of the cell is thereby seriously diminished in size. But even if, little by little, the cell- wall augments in thickness a hundredfold, any spot where thickening has not taken place from the first, and where, accordingly, a little depression occurs, is not subsequently covered with cellulose, but is carefully kept open by the protoplast as it builds. A greatly thickened wall of this kind resembles a fortification provided here and there with deep, narrow loopholes. Where two cells thus provided adjoin one another, the windows in the one occur, normally, exactly opposite those of its neiglibour, and the result is the formation of canals, very long relatively, which penetrate through the two adjacent cell-walls and connect the neighbouring cell- cavities together (fig. 10^). A canal of this kind is still closed, it is true, in the middle by the original cell-membrane as though by a lock-gate; but this slight obstruction may be removed later by solution, and the contiguous cells have then perfectly open connection through the canal. Very frequently provision is made in the very first rudiments of a cell-mem- 46 THE CELL-WALL. brane, destined to constitute a partition-wall, for open communications such as the above. For segments of the wall of various sizes are made from the beginning with sieve-like perforations, as is shown in fig. 10 *, which represents diagrammatically portions of tubular cells called " sieve-tubes," The pores are crowded close together on the perforated areas of the walls of the sieve-tubes, and their dimensions are relatively broad and short. Thus, when two neighbouring protoplasts reach out to one another through these pores, that is to say, when there is continuity of the protoplasm of the two cell-cavities, the connecting filaments, which pass through the pores and which fill them completely, are short and thick and have the appearance of pegs or stoppers. But in many cases the pores through which adjoining cell-cavities communicate are drawn out to a great length, forming infinitesimally slender passages. Thej^ are situated close together in great numbers and penetrate transversely through the thick cell- walls (fig. 10^). Neighbouring protoplasts may be brought equally well into mutual connection by means of these canals, or perhaps it would be better to say that their connection may be equally well maintained. For it is very probably the case that in the first rudimentary partition- wall, which is produced between the products of division of a protoplast, minute spots remain open and are occupied by connecting threads common to both halves of the protoplasm as they draw apart. Then in proportion as the partition-wall between the two protoplasts, produced by the division, becomes thicker, the openings take the form of fine canals, and the con- necting filaments are modified into long and exceedingly fine threads which fill the canals. These protoplasmic threads pierce through the thickened cell-wall in the same way as a dozen telegraph-wires might be drawn through a partition from one room into another. Often a number of protoplasts living side by side and one above the other are linked together by filaments of this kind, which radiate in all directions. This species of connection, of which an intelligible idea is given by fig. 10^, escaped the notice of observers in former times owing to the extraordinary minute- ness of the canals, and delicacy of the protoplasmic filaments. Another method of communication between protoplasts in adjoining cells has, on the other hand, been long known and often described, its phenomena being very striking and visible when onl}'- slightly magnified. The connection referred to is that which is afforded by the formation of so-called "vessels." By vessels the older botanists understood tubes or utricles, arising from the dissolution of the partition-walls between a series of cells. Either the partition-walls in a rectilineal row of cells vanish, in which case long straight tubes are produced; or portions of the walls of cells arranged at different angles to one. another are dissolved, and then tubes are formed having an irregular course, and sometimes branching or even uniting, so as to make a net-work. In instances of the first kind the lateral walls of the series of cells which are to lose their transverse partitions are previously thickened and made stiff by the proto- plasts, which also provide them with various mouldings and panellings, and above all with bordered pits. This task accomplished, the protoplasts forsake the tubes, whose TRANSMISSION OF STIMULI. 47 function thenceforth it is to serve as passages for air and water; thus the con- tinued presence of the protoplasts is no longer advantageous. On the other hand, in the second class of vessels the lateral walls of the cells, which have coalesced to form them, exhibit no thickening, but are soft and delicate, and resemble flexible tubing. These tubes, moreover, are not deserted by their protoplasts; but, after the coalescence of a number of cells into a smgle duct has taken place, the protoplasts in the cells are themselves merged together, and the entire tube is then occupied by an uninterrupted mass of protoplasm, which generally persists as a lining to the wall. As the initiation and construction of cell-walls are the work of the living proto- plast, so also is their removal. The home it has made for itself the protoplast can also demolish — either partially or completely. But this demolition is preluded by the importation of particles of water into the portions of the wall which are to be destroyed. The introduction of water brings the wall into a gelatinous condition ; the cohesion of its constituent particles is loosened, little by little, and at lengtli completely abolished. 4. COMMUNICATION OF PROTOPLASTS WITH ONE ANOTHER AND WITH THE OUTER WORLD. The transmission of stimuli and the specific constitution of protoplasm. — Vital Force, Instinct, and Sensation. THE TRANSMISSION OF STIMULI AND THE SPECIFIC CONSTITUTION OF PROTOPLASM. As has been already intimated, the breaking down of mdividual cell-walls and the formation of the various pits, sieve-pores and fine canals in thickened mem- branes, m the manner described in preceding pages, are processes of great import- ance to the life of protoplasts. In the first place, many of the resulting structures are the means of preserving the possibility of intercourse with the outside world. In a space inclosed by evenly thickened walls, the absorption of air, water, and other raw materials from the environment would be very difficult if not impossible; the protoplast inside would soon lack the provisions needful for further development, and would at last die of starvation, drought, and suffocation. But the little win- dows, whether open or closed by thin permeable membranes, enable it to supply itself with all necessaries of life. Another advantage is derived, in the case of many of these structures, inasmuch as the protoplasts on occasion escape through the open doors and settle down in some other part of the cell-colony, where they are able again to make themselves useful. Lastly, one of the most important benefits of all is due to the fact that mutual intercourse between protoplasts, living together as a commonwealth, is rendered possible by the canals which join them together. And 48 TRANSMISSION OF STIMULI. such an intercourse must of necessity be presumed to exist. When one considers the unanimous co-operation of protoplasts living together as a colony, and observes how neighbouring individuals, though produced from one and the same mother-cell, yet exercise different functions according to their position; and, further, how uni- versally there is the division of labour most conducive to the well-being of the whole community, it is not easy to deny to a society, which works so harmoniously, the possession of unity of organization. The individual members of the colonj^ must have community of feeling and a mutual understanding, and stimuli must be pro- pagated from one part to another. No more obvious explanation offers than that the protoplasmic filaments, which run like telegraph-wires through the narrow pores and canals in the cell- walls (see fig. 10^), serve to propagate and transmit stimuli from one piotoplast to another. These threads of protoplasm may indeed be likened to nerves which convey impulses determining definite actions from cell to cell. Imagination takes us further still, and raises the cell-nucleus to the position of the dominant organ of the cell-body For the nucleus not only determines the activity of the individual protoplast within its own cavity, but continues in sympathetic communion with its neighbour by means of all the threads and liga- ments which converge upon it. This last idea in particular derives support from indications that the filaments uniting neighbouring protoplasts have their origin in specific transformations in the substance of the nucleus itself. When a proto- plast living in a cell-cavity is about to divide into two, the process resulting in division is as follows: — The nucleus places itself in the middle of its cell, and at first characteristic lines and streaks appear in. its substance, making it look like a ball made up of threads and little rods pressed together. These threads gradu- ally arrange themselves in positions corresponding to the meridian lines upon a globe; but, at the place where on a globe the equator would lie, there then occurs suddenly a cleavage of the nucleus — a partition- wall of cellulose is interposed in the gap, and from a single cell we now have produced a pair of cells. In this way, from the nucleus, and from the protoplast of which the nucleus is the centre, two protoplasts have been produced, each having a nucleus of its own, and they thenceforth live side by side, each in its own chamber. It has been proved that in this process of division the substance of the nucleus is not completely sundered by the partition as it grows, but that, as we have already mentioned, minute pores are kept open in the cellulose wall, and that the pair of protoplasts continue joined together by threads running through these pores. When we realize that every plant was once only a single minute lump of protoplasm, inasmuch as the biggest tree, like the smallest moss, has its origin in the protoplasm of an egg-cell or a spore; and when we consider how, by growth and repeated bipartition, thousands of cells are evolved, step by step, from a single one, whilst their protoplastic bodies still remain united by fine filaments, we arrive of necessity at the conclusion that the whole mass of protoplasm, living in all the myriads of cells whose aggregation constitutes a tree, really is, and TRANSMISSION OF STIMULI. 49 continues to be, a single individual, whose parts are only separated by perforated sieve-like partitions. Every member of this community occupies a particular compartment or cavity, and is governed by a central organ, the cell-nucleus; but being linked to its fellows by connecting threads of protoplasm, a mutual under- standing is thus established among them. The physical basis of such an understanding may in this manner be represented with tolerable certainty. But it is extremely difficult to throw light upon the process of this mutual intelligence, the actual method whereby the cell-nuclei not only govern within their own narrow spheres, but also co-operate harmoniously for the good of the whole. And yet the problem involved in this unanimity of action, with a view to a systematic development of the plant in its entirety, is of such extreme importance that we cannot evade it even if, in the endeavour to solve it, we have to move altogether in the region of hypothesis. In every attempt at explanation of the kind we must, at all events, bear in mind that the agreement in question, as well as the processes which take place in pursuance of this agreement, such as the nutrition, growth, and the organization of the entire plant, are reducible to the subtlest atomic agencies in the living protoplasm. They may be resolved into the motion of minute particles, into attractions and repulsions, oscillations and vibrations of atoms, and into re-arrange- ments of the atomic groups called molecules. Again, these movements are the result of the action of forces, especially of gravity, light, and heat. As regards gravity and light, experiment shows, however, that, when acting on living proto- plasm, they give rise to varying effects even under the same conditions; and this fact, which will be discussed frequently later on, indicates that these forces are at any rate only to be conceived as stimulative and not coercive, and that they have no power to determine the kind of form. It is characteristic of the processes set up by gravity and light, especially when they take place in the continuous protoplasm of a great cell-community, that the coarser movements visible to the naked eye are often manifested in members comparatively remote from the part immediately affected by the stimulus. We cannot well represent this to ourselves except by supposing that the stimulus, which is the cause of the movement, is propagated through the threads of protoplasm from atom to atom, and from nucleus to nucleus. But the great puzzle lies, as already remarked, in the circum- stance that the atomic and molecular disturbances occasioned by such stimuli and transmitted through the connecting filaments are not only different in the proto- plasm of different kinds of plants, but even in the same plant they are of such a nature, according to the temporary requirement, that each one of the aggregated protoplasts in a community of cells undertakes the particular avocation which is most useful to the whole, the effect of this joint labour conveying the impression of the presence of a single governing power of definite design and of methodical action. That a stimulus causes different occurrences in different species of plants, and, more especially, that cell-communities arising from different egg-cells develop into 50 TRANSMISSION OF STIMULI. different forms, though under identical conditions and subjected to the same stimuU, are phenomena which have parallels in the inanimate world. A different sound is produced by striking the key of a piano which is connected to an A-string from that resulting from the transmission of a similar impulse to an F-string; and the difference depends on a difference of structure and an inequality of tension in the strings. Again, solutions of the sulphate and of the hyposulphite of sodium in similar glass vessels are indistinguishable at sight, both being colourless and transparent. These solutions will preserve their liquid condition when cooled down gradually to below freezing-point if they are kept absolutely still; but the moment the vessels are touched and a vibration thereby transmitted to the contents, they freeze. Crystals are formed in the apparently identical liquids, but crystals of different kinds, Glauber's salts in the one case, hyposulphite of sodium in the other. The variety of form depends simply on the sort of atoms, and on their number and mode of grouping. In a similar manner must be explained the variety of forms in many plant- species developed under the same conditions and affected by the same stimuli. Dozens of kinds of unicellular Desmids and Diatoms are often developed at the same time in a single drop of water in close proximity to one another. Although the protoplasm in the spores of these different species is absolutely identical to our vision, aided by the best microscopes, yet the mature cells exhibit a multiplicity of form which is quite astonishing to the observer on first inspection. One cell is semi-lunar, another cylindrical, a third stellate, a fourth lozenge-shaped, and a fifth acicular. In one specimen the cell-membrane is smooth, in another it is beaded; some are provided with siliceous coats, whilst others have flexible envelopes. The same thing holds good with respect to the vegetable structures, which are composed of myriads of cells, and develop into huge shrubs or tall trees. The protoplasm in the egg-cell of an oleander is produced close to that of a poplar on the same river-bank, and under exactly the same external conditions. The cells divide, and partition-walls are introduced in the proper direction in either case, according to a plan of structure which is adhered to with marvellous precision by the protoplasts engaged in the work of construction. In each species, stem, branches, foliage, and blossoms have invariably a particular form and arrangement, have the same colour and smell, and contain the same substances. How utterly different are the mature leaf, the opened flower, and ripe fruit of the oleander from the corresponding parts of a poplar. Yet both were nourished by the same earth, were surrounded by the same atmosphere, and encountered the same rays of sun- shine. We cannot otherwise explain it than by the supposition that, in a case like this, the difference of form in the perfected state is based upon a difference in the self-developing protoplasm, and that the atoms and molecules of this proto- plasm, which appears to us to be uniform, vary in kind, number, and grouping in the two species of plants. Consequently, we must assume that every vegetable organism, every species of plant that appears invariably in the same external form when mature, and develops according to an invariable plan, has a protoplasm i VITAL FORCE, INSTINCT, AND SENSATION. 51 of its own of a certain specific constitution. And, further, we must assume that this specific protoplasmic constitution is transmitted from one generation to another, so that the protoplasm of the oleander, for example, had exactly the same constitu- tion thousands of years ago as it has to-day. Lastly, we must assume that each special kind of protoplasm has the power to reproduce its like, ever anew, from the raw materials occurring in its environment. VITAL FORCE, INSTINCT, AND SENSATION. The phenomena observed in living protoplasm, as it grows and takes definite form, cannot in their entirety be explained by the assumption of a specific con- stitution of protoplasm for every distinct kind of plant; though this hypothesis will again prove very useful when we inquire into the origin of new species. What it does not account for is the appropriate manner in which various functions are distributed amongst the protoplasts of a cell-community; nor does it explain the purposeful sequence of different operations in the same protoplasm without any change in the external stimuli, the thorough use made of external advan- tages, the resistance to injurious influences, the avoidance or encompassing of insuperable obstacles, the punctuality with which all the functions are performed, the periodicity which occurs with the greatest regularity under constant condi- tions of the environment, nor, above all, the fact that the power of discharging all the operations requisite for growth, nutrition, renovation, and multiplication is liable to be lost. We call the loss of this power the death of the protoplasm. It ensues upon assaults from without if they succeed in destroying the molecular structure so entirely as to render reconstruction impossible; but, furthermore, death may take place without external cause. If cells of the blood-red alga, previously mentioned as allied to the red-snow, are collected from hollows in stones, casually full of rain-water, and are kept dry for weeks and then again moistened, the water is found to have a very power- ful eftect. The protoplasm becomes mobile, and swarm-spores are formed which put forth vibratile cilia, propel themselves about for a short time in the water, and then settle down in some favoured spot, draw in their cilia, come to rest and divide, producing ofispring which again are motile. This alga may be kept dry for months, nay even over a year, and still its cells exhibit the movements above described when put into water. But if a mass of it is preserved under these same conditions for many years and then moistened, the little cells will, it is true, take up additional water, but motile cells are no longer formed. The cells do not move, nor grow, nor divide, but gradually become discoloured; are first disintegrated and then dissolved. We say then that in them life could no longer be recalled, and we describe them as dead. The same thing is observed in great cell-communities. The seeds of many species of plants preserve the capacity for germination for an incredibly long period, especially when kept in a dry place. If after ten years such seeds are transferred into 52 VITAL FORCE, INSTINCT, AND SENSATION. moist earth, the protoplasm in the majority of cases begins to bestir itself and to move, and the embryo grows out into a seedling. After twenty years, perhaps, only about five per cent of the seeds preserved would germinate. The rest are not stimulated by damp earth to further development; their protoplasm no longer possesses the power of augmenting its volume by absorption of matter from the environment, or of developing a definite form, but is disintegrated by the influx of air and water and breaks up into simpler compounds. After thirty years hardly one of the seeds would sprout. Yet all these seeds were kept throughout the time at one place and under precisely the same external conditions; nor can the slightest change in their appearance be detected. Gardeners express the fact by saying that the capacity for germination becomes extinct in from twenty to thirty years. But what kind of a force is this which may perish without a physical change of the substance concerned affording the basis of the extinction ? In former times a special force was assumed, the force of life. More recently, when many phenomena of plant life had been successfully reduced to simple chemical and mechanical processes, this vital force was derided and eflfaced from the list of natural agencies. But by what name shall we now designate that force in nature which is liable to perish whilst the protoplasm suffers no physical alteration and in the absence of any extrinsic cause; and which yet, so long as it is not extinct, causes the protoplasm to move, to inclose itself, to assimilate certain kinds of fresh matter coming within the sphere of its activity and to reject others, and which, when in full action, makes the protoplasm adapt its movements under external stimulation to existing conditions in the manner which is most expedient? This force in nature is not electricity nor magnetism; it is not identical with any other natural force, for it manifests a series of characteristic effects which differ from those of all other forms of energy. Therefore, I do not hesitate again to designate as vital force this natural agency, not to be identified with any other, whose immediate instrument is the protoplasm, and whose peculiar effects we call life. The atoms and molecules of protoplasm only fulfil the functions which constitute life so long as they are swayed by this vital force. If its dominion ceases, they yield to the operations of other forces. The recognition of a special natural force of this kind is not inconsistent with the fact that living bodies may at the same time be subject to other natural forces. Many phenomena of plant life may, as has been already frequently remarked, be conceived as simple chemical and mechanical processes, without the introduction of a special vital force; but the effects of these other forces are observed in lifeless bodies as well, and indeed act upon them in a precisely similar manner, and this cannot be said of the force of life. Were we to designate as instinctive those actions of the vital force which are manifested by movements purposely adapted in some manner advantageous to the whole organism, nothing could be urged against it. For what is instinct but an unconscious and purposeful action on the part of a living organism ? Plants, then, possess instinct. We have instances of its operation in every swarm-spore VITAL FORCE, INSTINCT, AND SENSATION. 53 in search of the best place to settle in, and in every pollen-tube as it grows down through the entrance to an ovary and applies itself to one definite spot of an ovule, never failing in its object. The water-crowfoot, in deep water, fashions its leaves with finely divided tips, large air-passages, and no stomata; whilst, growing above the surface of the water, its leaves have broad lobes, contracted intercellular spaces and numerous stomata. Linaria Cymbalaria (see fig. 11) raises its flower-stalks from the stone wall over which it creeps towards the light, but as soon as fertilization has taken place, these same stalks, in that very place and amidst unchanged external conditions, curve in the opposite direction, so as Fig. 11. — Linaria Cymbalaria dropping its Seeds into Clefts in tlie Rocks. to deposit their seeds in a dark crevice. The flower-stalk of Vallisneria twists itself tightly into a screw and draws the flowers, which previously it had borne upon the surface of the water, down to the bottom when their stigmas have been covered with pollen-dust at the surface. These are all cases of unconscious action for a definite object, that is to say, they are the result of instinct. If, however, we attribute instinct to living plants, it is but a step further to consider them as endowed with sensation also. Feeling in animals is the con- comitant of a condition of disturbance in nerves and brain caused by a stimulus, which acts on the organs of sense, and is conveyed by nerves to the central organ. The transmission of the stimulus and the excited state of the brain and nerves are only molecular movements of the nervous substance, or, let us say, of the protoplasm, for nerve-fibres and nerve-cells are simply protoplasm developed in a particular manner. But the state induced by the stimulation of protoplasm, which is what we call sensation, cannot be essentially different in vegetable protoplasm from what it is in animal protoplasm, since the protoplasm itself, the physical basis of life in both plant and animal, is not different. In isolated plant-cells, indeed, it may amount to such a concentration of the condition of stimulation as to be called sensation, for the cell-nucleus is to all appearance 54 VITAL FORCE, INSTINCT, AND SENSATION. a central organ in relation to the protoplast that lives in a solitary cell. It is not of course to be supposed that within a whole plant-structure, that is in the community of live protoplasts which constitutes an individual plant, such a con- centration of stimulation could occur as is the case with individual animals which have nerve- fibres all converging into the brain; but between the sensation of animals without nerves and that of plants no essential difference can exist. Hence we infer that there is no barrier between plants and animals. The attempt to establish a boundary-line where the realm of plants ceases and the animal world begins is a vain one. If we naturalists, all the same, agree to separate plants and animals, we do so only because experience shows that a division of labour conduces to a speedier attainment of our object. On the intermediate ground where animals and plants meet, zoologists and botanists encounter one another, not, however, as hostile rivals with a view to exclusive possession of the field, but as colleagues with a common interest in the adminis- tration and cultivation of this jointly tenanted region. ABSOKPTION OF NUTEIMENT. 1. INTRODUCTION. Classification of plants with refei-ence to nutrition. — Theory of food-absorption. CLASSIFICATION OF PLANTS WITH REFERENCE TO NUTRITION. The object of a plant's vital energy, next in importance to the resistance of such influences as are likely to bring about the death of the protoplasm, is growth, i.e. the addition of substance to its body, or, in other words, the absorption of nutriment. A living plant, whether consisting of a single cell or of a vast community of cells, takes up food from its environment in quantities varying according to the needs of the moment. But its method of action — how it sets about acquiring possession of this raw material, how it manages to incorporate the substances absorbed from with- out, how it contrives to retain only such part as is useful to it, and to reject and get rid of, like ballast, what does not subserve its own growth — is infinitely varied. This variety in the processes of food-absorption corresponds, on the one hand, to differences in the habitat of plants, and, on the other, to the requirements of particu- lar species, which requirements in their turn depend upon a specific constitution of the protoplasm in each species concerned. The diflference must be very great between this process as manifested in plants which are immersed in water during their whole lives and the same as it occurs in plants which live in desert sands and are not supplied with water for months together. And again, absorption in those fungi which grow luxuriantly on damp timber in the deep obscurity of a mine must take place very differently from the corresponding process in the delicate alpine plants which on our mountain slopes are exposed periodically to the most intense sunlight, and then, for weeks at a time, are wreathed in sombre mists. So, also, the reciprocal action between plants and their environment must have a character of its own in the case of parasitic growths which absorb their food from other living organisms, and in those remarkable plants, too, which catch and devour small insects, and in such minute organisms as yeast, the vinegar ferment, and others, which play so important a part in our daily life, and lastly, in the gigantic trees which form our forests. To acquire a general notion of these forms, with reference to their varieties as regards nutrition, it is best to classify them in the first place in groups according to their habitat, viz.: into water-plants or hydrophytes, stone-plants or lithophytes, land-plants, and epiphytes. But here again it is necessary to remark tliat no sharp 56 CLASSIFICATION OF PLANTS WITH REFERENCE TO NUTRITION. line of demarcation exists between these groups; all are connected by numerous intermediate links, and there are forms which belong to one group at one stage of development and to another at another stage. The distinctive property of aquatic plants is that they derive their nourishment either entirely or principally from the surrounding water. Some preserve their freedom, floating or swimming about in the liquid medium; but the majority are fixed somewhere under the water by special organs of attachment. Many plants that are rooted in the mud at the bottom of pools are able to derive their food from the water when it is high, and when it is low, from the atmosphere as well: such amphibious organisms form a transitional group between water-plants and land- plants. The number of lithophytes is comparatively very small. They include those lichens and mosses which cling in immediate contact to the surface of stones and derive their food in a fluid state direct from the atmosphere. All lithophytes are so constituted that they can, without injury, dry up and suspend their vitality for a time when there is a failure of atmospheric precipitation lasting over a long period or when the air itself is very dry. But not every plant which grows upon rocks is to be regarded as a lithophyte in the narrower acceptation of the term. Those that are rooted in earth in the cracks and crevices of the rock must be classed amongst land-plants. To this class indeed more than half the plants now in existence belong. Though surrounded by air as regards a part of their structure they have another part sunk in the soil, and from the soil they take up water and inorganic compounds in aqueous solution. Plants which grow attached to other plants or to animals are called epiphytes. The majority of plants are during the period of food-absorption connected with the foster-earth and are not capable of locomotion. The plant being fixed to one spot must therefore sooner or later exhaust the ground in its neighbourhood, and must require a further supply of nutritive substances. The parts specially devoted to food-absorption often lengthen out in these circumstances beyond the im- poverished region, and thus endeavour to bring areas more and more distant within the range of absorption. Many plants possess the faculty, to which reference has already been made, of alluring animals and of killing and sucking their juices. Not only amongst saprophytes and parasites, but also amongst aquatic plants, instances occur in which certain movements are performed involving the whole body of the organism, with a view to promoting the absorption of nutriment. Particularly striking in this respect are many plasmoid fungi (which we may well refer to here, not on this account alone, but also for the additional reason that they take in nourishment without the intervention of a cell-membrane). The naked protoplasm in these cases, which include in particular the class of Amoebae, crawls in its search for food over the nourishing substratum, and derives from it immediately the materials needful for growth. Loose bodies are liable to be seized by the radiating processes of the proto- plasm, which then closes round them and drains them completely of their juices (see fig. 9, the last figure to the right). These bodies encompassed by the protoplasm, if small, are drawn inwards from the periphery and are regularly digested in the THEORY OF FOOD-ABSORPTION. 57 interior. Such parts of foreign bodies as are not serviceable for nutrition are sub- sequently eliminated or are left behind by the protoplast as it creeps onward. But this method of food-absorption is limited to amoeboid forms belonging to the boundary-land of animal and vegetable life. The movements of other naked proto- plasts, such as those which are carried about in the water by vibratile cilia, have nothing to do with the search for food or with its absorption, but are connected rather with the processes of distribution and propagation. THEORY OF FOOD-ABSORPTION. In the case of protoplasts inclosed in cell-membranes the food necessary for nourishment must always pass through the cell-membrane and peripheral proto- plasmic layer (ectoplasm) into the interior of the protoplasmic bodies. And so, conversely, such of the substances absorbed as are of no use in the construction of the organism or for any other purpose, must be sepai^ated and passed out through these envelopes. The cell-membranes of those protoplasts which are employed in absorbing food must accordingly have a special structure: the ultimate particles must be so arranged as to allow of the passage of nutritious material inwards, and of rejected matter outwards, without prejudice to their own stability. The passages in cell-walls used for this purpose are very minute, much smaller at all events than the pore-canals described above as being occupied by fine protoplasmic filaments; the dimensions are in fact so trifling as to be invisible even with the best microscopes. Still we are forced to conclude that they exist by a posteriori reasoning from a series of phenomena, and to assume that the cell- membrane, like almost every other kind of body, consists not of continuous matter, but of minute particles, which are termed atoms, and are separated from one another by infinitesimally small spaces. Various processes and appearances have also led physicists and chemists to the conclusion that these atoms are not aggre- gated in disorder, but are always combined together in groups of two or more, even in the case where all the atoms in a body are of the same kind, i.e. are the same element. If a body contains different elements they are not mixed together indiscriminately, but are grouped in conformity to a definite law: every group includes atoms of all the different elements concerned, arranged in a certain in- variable manner, not only as regards number, but also as regards relative position. Groups of atoms of this kind are called " molecules," and the spaces between them are supposed to be larger than those between single atoms. Further, it is not improbable that the molecules themselves form groups, each group consisting of molecules conglomerated in a definite manner, and that the passages separating these molecular groups are larger again than those separating the single molecules within each group. These groups of molecules have been called "micellae" or Tagmata, and they also are supposed to be aggregated together in definite order. According to this theory the cell -membrane is analogous to a sieve, the pores of which are grouped in a definite manner, the broadest perforations being between 58 THEORY OF FOOD-ABSORPTION. the micellae or groups of molecules, narrower apertures between the molecules or groups of atoms in each micella, and lastly the finest pores between the atoms themselves in each molecule. These interspaces are liable to contraction and expansion, for the union of the molecules is affected by two forces, one of which manifests itself as a mutual attraction between atoms and atomic groups, whilst the other tends to drive atoms and molecules asunder. Of these forces the former, i.e. the attractive force existing in all material particles, is called chemical affinity when it causes atoms of different kinds to unite to form a molecule; and it is called cohesion when applied to the mutual attraction of similar molecules, and adhesion where it holds together masses of molecular groups with their surfaces in contact. The action of heat is opposed to this attractive force, which is only effective at infinitesimal distances. Bodies are all caused to expand by heat, their atoms, mole- cules, and micellae being forced apart. Heat is believed to be a vibratory motion of these ultimate particles, and it is supposed that the greater the vibrations the greater is the separation of atoms and atomic groups, the interspaces expanding and the heated body increasing consequently in volume. As is well known, the atoms and molecules may be forced so far apart by increase of temperature that cohesion is entirely overcome, and solids are converted, first into liquids and at last into gases. The interspaces or passages between the molecules and molecular groups com- posing a cell-membrane are penetrable by molecules of other substances, provided always, firstly, that the admitted molecules are not larger than the passages; and secondly, that there exists between the molecules of the cell-wall and those of the penetrating body that sort of attractive force which has been designated chemical affinity. Both premises are satisfied in the case of aqueous molecules, and experi- ment proves that they are admitted into the inter-molecular spaces of a cell- membrane with great ease and readiness. The cell-membrane saturates itself with water, or, to use the technical phrase, it has the tendency and ability to "imbibe" water. The force of attraction between molecules of a cell-membrane and water- molecules is indeed so intense that the cohesion of the molecules in the membrane is partially neutralized, and the imbibed water causes them to move apart. In consequence of this, the cell-membrane swells up and its dimensions are increased. It is also supposed that the micellae of a cell-membrane attract and admit water- molecules to such an extent as to surround themselves with watery envelopes. Such a condition would no doubt be nothing but beneficial, promoting, as it would, the interchange of materials through the cell-membrane, and the mixing of fluid substances situated on either side of the porous membrane. At all events this mixing process must ensue in the interspaces of the cell-membrane; and, in the particular case out of which this discussion has arisen, viz. food-absorption, the interacting substances are, on the one hand, the compounds in the soil outside the cell-membrane, and, on the other, the organic compounds under the control of the live protoplast within the cell-membrane. Both the outgoing and the in- coming substances must be soluble in water, and must, therefore, have an attraction THEORY OF FOOD-ABSORPTION. 59 for water. But the power of a substance in aqueous solution, whether without or within the cell-membrane, to permeate the saturated pores, and to mix thoroughly there, certainly depends also on the degree of chemical affinity and of adhesion existing between the molecules and micellae of the cell-membrane on the one hand, and these infiltrating substances on the other. A very complex interaction of forces takes place which we cannot here investigate any further, as it would take us much too far afield. Returning to the explanation of food -absorption, attention must be drawn to the fact that the mixing or diffusion which takes place through the cell-membrane differs from the free diffusion which would occur if the cell-membrane were not present. Experiment has proved that if one side of a cell-membrane is steeped in a saline solution and the other in an equal volume of pure water, the number of saline particles which pass through into the water are many fewer than the number of water-particles which pass into the solution of salt; and, moreover, if an organic compound, such as albumen or dextrin, is on one side, and water on the other, water transfuses to the organic compound, whereas no trace of the albumen or dextrin (as the case may be) passes through to the water. Now this phenomenon, which is called "osmosis" (" endosmosis and exosmosis"), is of great importance for the conception we have to form of food-absorption. It is clear that, whilst water and substances dissolved in water are brought under the control of the protoplast within a cell through the cell-membrane, as a consequence of the action of albuminous and other compounds constituting the body of the protoplast, and of the salts dissolved in the so-called cell-sap in the vacuoles, there is no necessity for any part of the cell-content to pass out through the cell-membrane. Thus the protoplasm is able to exercise an absorptive action on aqueous solutions outside the cell-membrane, and to continue to absorb until the cell is filled. Indeed, the chemical affinity for water possessed by the substances in a cell may occasion so great an absorption of water that, in consequence, the volume of the cell is enlarged and the cell-membrane is subjected to pressure from within. The cell- membrane is able to yield to this pressure to the extent permitted by its elasticity; but excessive stretching of the cell-membrane is at length counteracted by cohesion, and thus a condition is attained in which the cell-contents and the cell-membrane are subjected to mutual pressure, a state which is called " turgid ity." The process just described, of the absorption of water in large quantities into the precincts of the protoplasm without any simultaneous transmission of matter to the outside, is certainly in no respect an exchange. But it obviouslj'' does not exclude the possibility of a real exchange taking place between substances on either side of a cell-membrane, i.e. between solutions in the soil and those in the cell- sap contained in lacunae of the protoplasm. Certain phenomena in fact put it beyond doubt that on occasion a real exchange of this kind does occur. But it is complicated by the circumstance that substances in process of being exchanged have to pass not only through the cell-membrane but also through the primordial utricle; and the primordial utricle consists of molecules of a kind other thiui €0 NUTRIENT GASES. those of tlie cell-wall, having different chemical affinities, and these molecules again are differently grouped; nor are the passages for aqueous solutions the same. All this cannot but have an important bearing on the permeating capacity of the substances that are being interchanged. Although all these ideas concerning the molecular structure of cell-membranes and of protoplasm, concerning the intermixture and exchange of materials and the absorption on the part of cells and their swelling up, have only the value of theories, still we have good ground for assuming that they are fairly near the truth. They give us, at all events, an intelligible representation of the inter- action which takes place between living protoplasts, with their need for food, and the environment, which supplies the nutriment. 2. ABSORPTION OF INORGANIC SUBSTANCES. Nutrient Gases. — Nuti'ient Salts. — Absorjjtion of Nutrient Salts by Water-plants, Stone-plants, and Laud-plants. — Relations between the position of Foliage-leaves and Absorption-roots. NUTRIENT GASES. One of the most important sources of the nourishment of plants is carbonic acid. The living protoplasts appropriate it from water and from air, in the latter case chiefly by attracting the carbon-dioxide.^ This gas penetrates a cell- wall satur- ated with water more readily than the other constituent gases of the atmosphere (nitrogen and oxygen). In the wall it is converted into carbonic acid, and it then passes on into the cell-sap contained in the cavities of the protoplast. Apart from the effects of temperature and atmospheric pressure, the quantity of carbonic acid absorbed is chiefly determined by the requirements of the cells whose nourish- ment is in question. These requirements, however, vary considerably according to the specific constitution of the protoplasm and with the time of day. During daylight the need of carbon is very great in all green plants. As soon as the carbonic acid reaches the cell-sap it is decomposed and reduced by the action of sunlight, and from it are formed compounds known as carbo-hydrates. The oxygen thus set free is, however, removed from the cell precincts, and expelled into the surrounding air or water. In this way the gas when barely absorbed is withdrawn, as such, from the cell-sap, the carbon alone being retained and the ■oxygen eliminated, and a renewed attraction of carbon-dioxide from the sur- rounding medium ensues. The fresh supply again is immediately worked up in the green chlorophyll-bodies, so that there is a constant influx of carbon-dioxide, and therefore indirectly of carbonic acid, from the environment into the interior of green cells to the part where its consumption takes place. Were it possible to see 1 The atmosphere contains free carbon-dioxide and not carbonic acid. But carbonic acid is formed when the dioxide is absorbed into water. NUTRIENT GASES. 61 the molecules of carbon-dioxide in the air, we should observe how much faster they are impelled towards the leaves and other green parts of plants, where the intense craving for carbon is localized, than are the other constituent particles of the air. This impulsion and influx lasts so long as the green cells are under the influence of daylight. The first thing in the morning when the first ray of sunshine falls upon a plant the protoplasts begin work in their little laboratories decomposing carbonic acid, and producing from it sugar, starch, and other similar organic compounds. And it is not till the sun sets that this work is suspended, and the influx of carbon- dioxide stopped till the following morning. The green plants that spend all their lives under water are supplied with car- bonic acid by the water surrounding their cells, which always contains some of that material. In the case of unicellular plants of this class, absorption of carbonic acid takes place through the whole surface of the cell-membrane. Multicellular plants, with their cells arranged in filaments or plates, only take in carbonic acid through those parts of the walls of their cells which are in immediate contact with the water. This applies also to submerged plants composed of several layers of cells and of considerable dimensions. Thus, in plants of this kind, the cells in contact with the water constitute the skin. They are always pressed closely together and squeezed flat, are not thickened on the side exposed to the water, and are united everywhere edge to edge leaving no gaps. But in the interior of these water-plants large lacunae and cavities are formed from earliest youth, owing to the detachment of single rows of cells, and the spaces so formed are filled with a quantity of nitrogen, oxygen, and carbon-dioxide, that is to say, with a gaseous mixture not essentially different from atmospheric air. Although this organiza- tion may have as its primary object the reduction of the plant's weight as a whole, it cannot be without a further importance inasmuch as carbonic acid can be taken up from the air-spaces into adjacent cells. But there is no doubt that, even in this case (of water-plants provided with large internal air-cavities), the chief absorption of carbonic acid is through the epidermis, or more precisely through those walls of the epidermal cells which are in immediate contact with the water. The carbonic acid taken up by cells, wholly or partially immersed in water, is either contained as such dissolved in the watery medium, or occurs in com- bination with calcium as bicarbonate of lime. Part of the carbonic acid in this bicarbonate in aqueous solution is susceptible of being withdrawn by water-plants, mono-carbonate of lime, which is insoluble in water, being then precipitated on the cell-wall through which the rest of the carbonic acid has passed into the cell-interior. Accordingly, a large number of water-plants are found incrusted with lime in both fresh and salt water. We shall return to this important pheno- menon when we treat of the influence of living plants on that part of the environ- ment which comes within their sphere of action for purposes of nutrition. Lithophytes obtain carbonic acid from the moisture deposited upon them from the aqueous vapour in the atmosphere, and attract carbon-dioxide direct from the <52 NUTRIENT GASES. air around them. The chief members of this class are those mosses, liverworts, and lichens which, though clinging to dry rocks, behave just like water-plants as regards the absorption of carbonic acid. There is no reason to think that these plants absorb carbonic acid in dry weather; for under the influence of dry air they lose water fast, and meanwhile receive no compensation from the rock to which they are attached, and in a short time they become so dry that they crumble into powder when rubbed between the fingers. Vitality is suspended for a time, and it is out of the question that there should be any absorption of carbon-dioxide from the atmosphere under such circumstances. But the moment the plant is moistened by rain or dew, the cell-walls directly exposed to the air become saturated, and are enabled to admit water into the interior. Then the lithophytes suck up water very fast; the dry, apparently dead, incrustations swell up again, and, together with the rain and dew, carbonic acid is absorbed, it being contained in all depositions of atmospheric moisture. A tumescent moss tuft can, in addi- tion, absorb carbon-dioxide direct from the atmosphere through its saturated superficial cells; but the quantity of carbonic acid thus acquired by a plant is in any case only secondary. Many mosses, as for example the widely-distributed Grim- mia apocarpa, are also able to live just as well under water as in air; nor is any alteration of their leaves necessary in either condition, nor any special contrivance for the absorption of carbonic acid and water. These substances reach the interior by similar passage through cell-walls of identical construction, whether the Griminia spends its life attached to submerged rocks or in the open air at the top of a mountain; whence we may infer that there is a greater resemblance between lithophytes and water-plants as regards nutrition than between litho- phytes and land-plants. Land-plants satisfy their need of carbon almost exclusively by withdrawing the dioxide from atmospheric air. For the purpose of this direct appropriation, specially adapted structures are found in them. Seeing that these plants are not able to endure periodic desiccation in times of drought, as lithophytes are, it is necessary for them to be secured against excessive loss of water. Accord- ingly, the cell-walls in immediate contact with the aii', that is to say, the outer walls of the epidermis, ai-e thickened by a layer (cuticle) which is impermeable by air or water, and, in general, they are so organized that water cannot readily escape from the interior of the cells. Obviously, however, a cell- wall which opposes a strong resistance to the extravasation of water will not give easy admittance to an influx either, and the conditions for the passage of gases through a cell-membrane, thickened and cuticularized in this way, would be far from favourable. As a matter of fact many of the constituent gases of the atmosphere permeate these thickened walls of the epidermal cells only with great difficulty, and others not at all. Carbon-dioxide alone has the power of penetrating, but even in the case of this gas the quantity is not always sufficient to satisfy the demand. To ensure that so important a form of plant-food should reach in proper amount those cells lying under the epidermis, which are occupied by protoplasts engaged in the regu- NUTRIENT GASES. 63 lation of nutrition, there is an adaptation of structure of the following nature. Among the firmly connected epidermal cells with their thickened outer walls al- most impervious to air, other cells are interspersed at intervals. They are always in pairs, are generally rather smaller than the rest, and have a little cleft open between them. Inasmuch as these apertures (stomata) always exist where passages and canals, the so-called intercellular spaces, have arisen from the separation of indi^'idual cells of the sub-epidermal tissues, each stoma constitutes the mouth of a system of channels ramifying between the thin-walled cells of the interior. The components of the atmosphere, especially carbon-dioxide, are able to reach these internal passages through the stomata, and in them they travel to the chlorophyll- containing cells. Through the thin, saturated walls of these cells they are able to penetrate with ease, and so they reach the living protoplasts, with their equipment of chlorophyll, whose daily work it is, as already mentioned, to decompose — under the transforming power of light — the carbonic acid as it reaches the chlorophyll- bodies, to work up the carbon and expel by the same path as they entered not only the oxygen but also all other aerial constituents which may have penetrated and for the moment find no employment. These ventilation-canals, with stomata as orifices at the epidermis, have other uses besides the importation of carbon-dioxide (and therefore of carbonic acid) and the exportation of oxygen. For the same pores, passages, and lacunae, as serve for the influx and exit of carbon-dioxide and oxygen respectively, are the channels of a plant's respiration. Moreover, they play a very important part also in the escape of aqueous vapour, the process known as "transpiration;" and as the variety in their structure is to be interpreted chiefly as an adaptation to the different condi- tions under which transpiration occurs, it cannot be profitably discussed until we treat of that process. Those saprophytes and parasites which contain no chlorophyll or pi-actically none, do not absorb any free carbon-dioxide from the atmosphere, but supply them- selves with carbon from the organic compounds in the nutrient substratum on which they grow. But saprophytes and parasites, abundantly furnished with chlorophyll, doubtless do attract free carbon-dioxide in addition. They may do so either after the manner of water-plants and lithophytes, as is the case with Euglenoe, and with mosses growing on the dung of mammalia; or else after the manner of land-plants, as instances of which the cow-wheat, yellow-rattle, and eye-bright may be quoted. It is a very remarkable fact that no plant is known which takes up carbon- dioxide or carbonic acid from the earth. One might expect that the roots of land- plants at any rate, ramifying as they do in a stratum of earth saturated with water containing carbonic acid in solution, would suck up to some extent so important a food, and that it would be from them conducted to the green -foliage leaves. But so far as experiments have gone, they indicate that this is not the case. Equally curious is the circumstance that nitrogen, which is an indispensable constituent of protoplasm, and therefore a very important means of subsistence, is 64 NUTRIENT GASES. not absorbed from the surrounding air, although, as is well known, the atmosphere contains nitrogen to the amount of 79 per cent of its volume. There can be no doubt that though nitrogen permeates the cell-walls of an air-encompassed plant much less readily and quickly than carbon-dioxide, yet it is carried from the atmos- phere into the ventilation-spaces of green foliage-leaves, and further through the thin cell-walls into the laboratories of the protoplasts, where one would expect it to be worked up in the same way as carbonic acid. The most careful experiments have determined, however, that it is not turned to account in this form by the proto- plasts, but that on the contrary it is given back unused to the air, and only such nitrogen as reaches the interior of plants in combination with other substances is of any service there. The principal sources of the nitrogen required by plants are nitrates and ammoniacal compounds absorbed from the ground; but nitric acid and ammonia themselves, of which there are traces in the atmosphere and in water, must not be overlooked. The quantity of nitric acid m air is, it is true, even less than that of carbon-dioxide; but just as the small amount of carbon-dioxide can be absorbed from the air with highly productive results, so may also the still smaller proportion of nitric acid be turned to account. The sources of nitric acid are dead organic bodies as they decompose and become oxidized. In many ways the process of formation of nitric acid from decaying bodies may take place so as to produce ammonia in the first place and from it nitric acid. It would seem possible, though it is an unproved assumption, that in places where dead bodies of plants and animals vegetable mould, manure, and such things are undergoing oxidation, that is to say, in woods and fields, the small quantities of nitric acid that are given off are imme- diately taken up by the plants growing there. It must be borne in mind that plants behave with reference to what is necessary or useful to them like a chancellor of the exchequer preparing his budget; they take these things where they find them. The question has been raised, too, as to the source from which the first plants that appeared on the earth were able to obtain nitric acid. We are obliged to assume that, at that time before the existence of nitx^ogenous organisms to supply nitric acid by oxidation of their dead bodies, all nitric acid, and therefore all tlie nitrogen used in the nourishment of plants, was generated by thunder-storms. We know that nitric acid is formed in the air on occasion of electric discharges and is deposited on the earth together with rain and dew. This source of nitric acid is not yet exhausted, and even at the present day it no doubt plays the same part as in the ages long past at the commencement of all vegetable life. If nitric acid is used by protoplasts, in the building up of the highly important albuminous compounds, it is broken up in a manner similar to the decomposition of carbonic acid to form carbohydrates, that is to say, oxygen is separated out. In this case, however, sunlight and, therefore, chlorophyll are not immediately con- cerned. Moreover, the oxygen that is set free is not eliminated, but is used in the manufacture of other compounds in process of formation in the plant, probably in that of vegetable acids. NUTRIENT GASES. 65 Ammonia behaves in relation to plants just in the same way as carbon-dioxide and nitric acid. It is disengaged from dead decomposing organic bodies, and is found in traces, either alone or with equally minute quantities of carbon-dioxide and cai-bonic and nitric acids in the air, in atmospheric deposits, and in all water wherein animals and plants reproduce their kind, the old individuals dying and making way for the young. Water-plants are all limited to this source for acquisi- tion of nitrogen. As regard lithophytes, it stands to reason that they must derive their nitrogen from the ammonia contamed in the air, in atmospheric deposits, and from nitric acid. Whence otherwise could a crustaceous lichen attached to a quartz rock on a mountain supply itself with the nitrogen essential for the growth of its protoplasm? Moreover, some of the larger lithophytes, especially mosses, seem to be capable of absorbing ammonia direct from the air. An observation made in the Tyrolese Alps has some bearing on this question: — The ridges of the Hammerspitze, a peak rising to 2600 meters between the Stubaithal and the Gschnitzthal, is, in favourable weather in the summer, the resting-place of hun- dreds of sheep, and is consequently covered with an entire crust of the excrements of these animals. A highly offensive and pungent smell of ammonia is evolved, and renders a prolonged stay on this spot anything but pleasant, notwithstanding the beauty of the view. Now, it is worthy of note that the mosses, which are produced in abundance on the rocks above this richly-manured ground, but are not them- selves actually amongst the sheep-droppings, exhibit a luxuriance unparalleled on any of the neighbouring summits belonging to the same formation but unfre- quented by sheep. The gaily-coloured green carpet extends as far as the ammo- niacal odour is perceptible, and it is natural to suppose that this luxuriant growth is stimulated by the absorption of ammonia direct from the air. Land-plants also can take up ammonia from the air. It has been shown that the glandular hairs of many plants, for instance those on the leaves of Pelargonium and of the Chinese Primrose, have the power of absorbing traces of ammonia, and of sucking up carbonate and nitrate of ammonia in water with rapidity. When we consider that a single one of these primroses (Primula sinensis) possesses two and a half millions of absorbent glandular hairs so placed as to be able to take up the ammonia brought to the plant by rain, we are unable to look upon this process as of altogether trifling importance. It is highly probable that almost all ammonia, after its formation from decaying substances in the ground, is at once absorbed by the plants growing in the immediate neighbourhood, and that the relatively small quantity of ammonia in the upper atmospheric strata is referrible to this cause. The splendid luxuriance of the pelargoniums, thickly studded with glandular hairs, which one sees in front of cottage windows in mountain villages where a dung heap is close by, and in the windows of stables, frequently excites admiration and surprise. Whether it is due to the fact that in these situations there is the possi- bility of absorbing an unusually large quantity of ammonia is a question which we will leave undecided. 66 NUTRIENT SALTS. NUTRIENT SALTS. If wood, leaves, seeds, or any other parts of plants are subjected to a hioh temperature with free access of air, the first changes that occur are in the com- pounds of nitrogen and of carbon contained in the heated matter. They turn black, are charred and burnt, and ultimately the products of combustion pass into the atmosphere in gaseous condition. The incombustible part which remains behind is called the " ash." The quantity of this ash, as well as its composition, varies very much in different species of plants, and even in different parts of the same plant. Generally the weight of ash is only one or two per cent of the entire weight of the plant in a dry state before burning. The greatest relative proportion of ash is that which is obtained from the combustion of those hydrophytes which live in the sea; and next in quantity is the ash of the family of Oraches which abound on salt-steppes. On the other hand, the smallest quantity is that afforded by fungi and mosses, by Sphagnum in particular, and with these must be mentioned the tropical orchids living on the barks of trees. Seeds and wood yield relatively much less ash than leaves. But, as above remarked, some ash is formed upon the combustion of any part of a plant or even of a single cell, and this residue of ash sometimes allows of our recognizing exactly the size, form, and outline of the cells. The universal distribution of ash-forming constituents permits us to conclude with certainty that they do not exist fortuitously in plants, but are essential to them. That these constituents are indispensable may also be proved directly. If an attempt is made to nourish a plant on filtered air and distilled water exclusively, the plant soon dies; but if a small quantity of the constituents of its ash are added to the distilled water in which the roots are immersed, the plant grows visibly in the solution, and develops leaves and flowers and even seeds capable of germination. Experiments of this kind with cultures have been the means of almost com- pletely establishing the division between those constituents which are indispensable for all plants, and those which are only necessary under certain conditions and to particular species, or, still less, only beneficial. Those elements must be regarded as essential, which are used by plants for the process of construction, and enter into the composition of the protoplasm or of the cell-membrane — such, for instance as are essential constituents of proteid substances, or are in some way necessary to the formation of these products. Amongst these must be included sulphur, phosphorus, potassium, calcium, and magnesium. Some plants, especially those that live in the sea, require sodium, iodine and chlorine, and, for green plants, iron is necessary. Silicon is also very important for most plants in helping them to flourish in the wild state. Most of these elements are taken into a plant, in the corrse of nutrition, in a condition of extreme oxidation, that is to say in combina- tion with a quantity of oxygen; in fact, as a general rule, they are absorbed in the form of salts, and we may for the sake of brevity include all the mineral food- stuffs under the name of nutrient salts or food-salts. NUTRIENT SALTS. 67 It is obvious that food-salts can only pass through cell-membranes and reach the interior of a plant in a state of solution. On this account the soluble sul- phates, phosphates, nitrates and chlorides of calcium, magnesium, potassium and iron, may pre-eminently be called food-salts. Whether an essential element is absorbed by a plant in the form of one of these compounds or another appears to be unimportant; phosphorus, for example, may be proffered by the soil in the form either of potassium phosphate or of sodium phosphate, with like results. As regards the importance of sulphur to plants, it is at any rate established that it is necessary for the production of proteid substances. Phosphorus appears to be indispensable in the transformation of certain compounds of nitrogen. Potassium is supposed to play a part in the formation of starch. Calcium is introduced into plants in combination with sulphuric acid as calcium sulphate. This salt is decom- posed, the lime combining with oxalic acid to form insoluble calcium oxalate, and the sulphur going to form the sulphuric acid which is used in the construction of albuminous substances or proteids. Lime is therefore important, inasmuch as it is a medium of transport for sulphur. Iron certainly participates in the forma- tion of chlorophyll, even if it does not enter into its composition, as was formerly supposed. For, it has been proved, by means of artificial cultures, that plants reared in solutions free from iron were white instead of green, and died at last; whereas, after the addition of a small quantity of a soluble iron salt, such plants became green in a very short time, and were able to continue their development. The utility of most of these elements does not therefore appear to consist necessarily in their entering into the composition of organic compounds, but in the promotion and regulation of the constructive and destructive chemical processes. Silicic acid, which occurs so plentifully in the ash of many plants as to con- stitute often more than 50 per cent, has a different function. If the minute unicellular water-plants known as Diatoms are incinerated, or if stems of Equisetum, Juniper-needles, or leaves of grasses, &c., are subjected to a red heat, white skeletons remain behind which consist almost entirely of silicic acid, and exhibit not only the forms of the cells, but even the finest sculpturing of the cell- walls. In par- ticular, the stiff hairs on the leaves of grasses are preserved, and better still the cell-membranes of diatoms. The latter present very beautiful forms with their outlines quite distinct, and many structural properties of the cell-membranes, especially their moulding, striation, and the dots and other excrescences are to be seen much more clearly after than before ignition, when the transparency was less owing to the protoplast occupying the interior of each cell. In order to describe exactly the very varied form of Diatomacese, specimens are carefully and thor- oughly ignited, and the descriptions and illustrations of these microscopic plants are for the most part made from siliceous skeletons prepared in this way. These skeletons show clearly that silicic acid occurs only in the cell-membrane, and plays no part as constituent of any chemical compound in the protoplasm; nor does it appear to be instrumental in the formation of any such compound. The molecules of silicic acid are so closely packed and so evenly distributed amongst the mole- 68 NUTRIENT SALTS. cules of cellulose that, even after the removal of the latter, the entire structure is preserved in outline and in detail. They form, therefore, a regular coat of mail which may be looked upon as a means of protection against certain injurious ex- ternal influences. For a large number of plants living in the sea, sodium, iodine, and bromine also are of especial importance as food-stufFs. How far fluorine, manganese, lithium, and various other metals, which have been detected in the ash of some plants, are of use is not determined, for our knowledge is particularly incomplete with respect to the various uses subserved in nutrition and growth by the different mineral food-stuffs. It is worthy of note that alumina, which is so widely distributed and easily accessible to plants, is only very rarely absorbed. The ash of Lycopodium is the only kind in which this substance has been identified with certainty in any considerable quantities. Lastly, amongst the sources of elements contained in the food-salts, we must consider the solid crust of the earth. But it is only in the case of comparatively few vegetable organisms that this earth-crust forms the immediate foster-soil. The majority derive the salts that nourish them from the products of the weather- ing of rocks, from refuse and the decaying remains of dead animals and plants, which, in decomposing, give back their mineral substances to the ground, from underground waters that filter through fissures in rocks and through the interstices of sandy or clayey soils soaking with lye, the adjacent parts of the earth's crust., and, lastly, from the water of springs, streams, ponds, and lakes, which have come to the surface holding salts in solution, as also from sea-water with its rich supply of salts. The very salts that are needed by most plants are amongst the most widely distributed on the earth's surface. The sulphates of calcium and of magnesium, for example, and salts of iron, potassium, &c., are found almost everywhere in the earth, and in water, whether subterranean or superficial. At the same time it is very striking that these mineral food-salts are not introduced into plants by any means in proportion to the quantity in which they are contained in the soil, but that, on the contrary, plants possess the power of selecting from the abundance of provisions at their disposal only those that are good for them and in such quantity as is serviceable. This selective capacity of plants is manifested m many ways, and we will now briefly consider some of the most important of them. In the first place we have the fact that plants reared close together in the same soil or medium may yet exhibit an altogether dififerent composition of ash. This is particularly striking in water and bog-plants, which, though rooted in close proximity and immersed in the same water, show very considerable differences in respect of mineral food absorbed. The result, for instance, of testing specimens of the Water-soldier (Stratiotes aloides), the White Water-lily (Nymphcea alba), a species of Stone- wort {Chara foetida), and the Reed (Phragmites communis), all growing close together in a swamp, was as follows as regarded the potash, soda, lime, and silicic acid, held by them respectively: — NUTRIENT SALTS. 69 Water-soldier. Water-lily. Potash, Soda, Lime, Silicic Acid,. 30-82 2-7 10-7 1-8 14-4 29-66 18-9 0-5 0-2 0-1 54-8 0-3 8-6 0-4 5-9 71-5 The other constituents of the ash of these plants, in particular iron oxide, mag- nesia, and phosphoric and sulphuric acids, exhibited less marked differences; but the inequality in the amounts of potash, soda, lime and silicic acid are so great, as only to be explicable on the assumption of a power of selection on the part of these plants. Various species of brown and red sea- weeds, which had been attached to the same rock and developed in the same sea- water, showed similar variations in the composition of their ash. On the mountains of serpentine rock near Gurhof, in Lower Austria, specimens of Biscutella Icevigata and Dorycnium decumibens were collected from plants growing together, and one above the other, upon a declivity which they clothed. Their roots, interlaced here and there, were fixed in the same ground, and drew nourishment from the same store. The following table gives the composition of the ash in these two species: — Biscutella Icevigata. Dorycnium decumbens. Biscutella l£evigata. Dorycnium decumbens. Potash, Lime . 9-6 14-7 280 7-8 16-7 20-9 19-6 2-8 Silicic Acid, Suliihur, 13-0 5-2 15-9 5-4 6-3 1-6 22-3 9-7 Lon Oxide, Carbonic Acid, The differences here seem to be not so great as in the case of the water-plants previously given, but they are sufficient to prevent our regarding them as merely the result of chance. If, on the other hand, we compare the composition of the ash of different specimens of the same species, which have been reared on similar soils, but at great distances from one another, the discrepancies are comparatively slight. Foliage from beech-trees growing on the limestone mountains near Regensburg yielded an ash practically identical with that obtained from leaves of beeches on the Bakonyer-Wald hills in Hungary. The ash of different individuals of a single species even exhibits the same constitution, in the main, when those individual plants have obtained their nutriment from soils differing greatly in chemical composition. Only in cases where the quantity of a substance in one soil is more abundant than in the other there is generally a greater or less amount of it to be found in the ash. That under these circumstances certain substances may replace one another is not improbable. But such substitution must be confined to those nearly allied com- pounds whose molecules are capable of being used indifferently by the formative 70 NUTRIENT SALTS. protoplasm in construction, and in the storage of materials. The annexed- table, which gives side by side analyses of the ash of branches of the Yew (Taxus baccata) with their leaves attached, illustrates the replacement of calcium by magnesium : — Ash from branches and leaves of the Yew from Serpentine. Limestone. Gneiss. Silicic Acid, Sulphuric Acid, 3-8 1-9 8-3 3-6 1-6 5-5 37 1-9 4-2 0-6 27-6 24-4 Iron Oxide 2-1 1-7 16-1 „„„ 36-1 } .. - Lini6, .. . ... Magnesia, Potash, Carbonic Acid, Traces of Manganese, Chlorine, &o 22-7 \ 2^'S 29-6 14-1 5-1 r^-^ 21-8 23-1 Totals, 99-6 98-5 98-7 The Yew occurs in Central Europe on very various mountain formations, chiefly on limestone, but not infrequently on gneiss, and occasionally on serpentine rocks. On comparing the quantities of calcium and of magnesium in the ash of yews, grown on lime and on gneiss respectively, with those yielded in the case of serpentine for- mation, we find that magnesia preponderates considerably in weight over lime in a yew from serpentine rocks (which are in the main a compound of magnesia and silicic acid), whilst the proportion between these two salts is reversed in a yew grown upon limestone. The obvious inference from the table is that, in plants from a serpentine ground, lime is to a great extent replaced by magnesia. This is fur- ther supported by the circumstance that if lime and magnesia are counted together the resulting numbers are very near one another, namely 41-2 per cent of the ash for limestone, 38-8 per cent for serpentine rock, and 36"3 per cent for gneiss. But all these phenomena observed in connection with the selection of food-salts are not nearly so surprising as the fact that plants are also capable of singling out from an abundance of other matter particular substances, which are of impor- tance to them, even from a soil containing them in barely perceptible quantities, and of concentrating them to a certain extent. As has been shown above, nearly a third of the ash of the white water-lily is composed of common salt. One might, therefore, suppose that the water in which water-lilies flourish contains a particu- larly large quantity of common salt. But nothing of the kind is the case. The bog water which bathed the stem and leaves of this specimen only contained 0*335 per cent of common salt, and the mud through which the roots straggled contained only 0*010 per cent. No less astonishing is it to find Diatomaceae, with cell-membranes, as above mentioned, sheathed in silicic acid, existing in water which contains no trace of silicic acid. Above the Arzler Alp, in the Solstein chain near Innsbruck, there is a spring of cold water which falls in little cascades between blocks of rock. The NUTRIENT SALTS. 71 water of this spring is hard, and it deposits lime at a little distance from the source. Exactly at the spot where it wells out of a fissure in the rock its bed is entirely- filled by a dark-brown flocculent mass which consists of millions of cells of the beautiful Odontidium hiemale, a species of diatom with siliceous coating. These cells are ranged together in long rows, and are present in numbers and luxuriance such as are scarcely ever to be observed in other situations. Yet the spring water flowing round contains so little silicic acid that no trace of this substance could be discovered in the residue from the evaporation of 10 litres. An instance similar to this of silicic acid, is affbrded by the iodine in the sea. Most of the sea-wracks inhabiting the North Sea contain iodine, many indeed in considerable quantity, and yet we have not hitherto succeeded in detecting iodine in the water of the North Sea. Similar phenomena, sometimes quite bafiling explana- tion, are exhibited by land-plants. The clefts in the rocks of quartziferous slate in the Central Alps are, in many places, overgrown by saxifrages (Saxifraga Sturmiana and Saxifraga oppositifolia) with leaves aggregated together in closely-crowded rosettes, which are conspicuous from afar, owing to their pale colouring. On closer inspection one finds that the apices and edges of these rosulate leaves are covered with little incrustations of carbonate of lime, a substance which will be frequently referred to in connection with its importance to plants. But one seeks in vain for any lime compound in the earth which fills the clefts, and the only traces of lime contained in the adjacent rock itself are those occurring in the little scales of mica scattered about, and these are not readily decomposable. Yet the lime incrusting the saxifrage leaves can only be derived from the underlying rock, just as in former instances the silicic acid in the cell-membranes of diatoms must be secreted from the spring described, the iodine in sea-weeds from the sea, and the common salt in water-lilies from the pond where they grow, although in each case the substance concerned is only to be found, if at all, in scarcely ponder- able traces in the soil or liquid serving as medium. Facts of this kind have a special interest, because they prove that plants have the power of appropriating a substance, if it is important to them, even when it is only present in extremely minute quantities. Where a plant is surrounded by liquid, we can well imagine that fresh portions of the medium are constantly coming into contact with its surface; for, even in water apparently still, compensating currents are con- tinually being caused by changes of temperature. Thus, in the course of a day, thousands of litres of sea-water may flow over a sea-weed with a surface of one square meter, and, even if only a small portion of the substance, traces of which we are supposing to exist in the water, is wrested from each litre, still, the absorbing plant might collect quite a profitable quantity in a number of days. The volume of water flowing over a plant situated in the source of a spring is still greater, and it is readily conceivable that even the most minute trace of silicic acid may become of account in course of time. There is more difficulty in understanding how plants with roots in the earth set about utilizing substances contained in the soil in scarcely appreciable quantities. These plants 72 NUTRIENT SALTS. must at all events come into contact with as great a mass of nutrient soil as possible, and this is effected by means of a widely-ramifying system of roots; and, in addition, they must assist in making available desirable matter in the soil by the elimination from themselves of certain substances. In order to explain the remarkable power that plants possess of exercising a choice in the absorption of certain food-stuffs from amongst the whole number presented to them, we must in the first place assume a special structure to exist in the cells which are in immediate contact with the nutrient medium. To reach the interior of a cell, the salts must pass through the cell-membrane and the so-called ectoplasm. We may look upon these walls, that are to be pene- trated, as filters, or, to abide by our previous simile, as sieves, which allow only certain kinds of molecules to pass and arrest others. Moreover, just as the structure of a sieve, especially the size and shape of its pores, has its eflfect in the separation of the particles of the matter sifted, so also may the structure of a cell-wall have a discriminating influence in the absorption of food-salts. It may be supposed that the cell-wall in one species of plant acts as a sieve capable of letting through molecules of potash but none of alumina, whilst the cell-wall in a second species allows molecules of alumina to pass as well, but is impervious to those of chloride of sodium. This hypothesis would also explain why the absorp- tion of food-stuffs by plants generally takes place through cell-walls, and why absorption into the organs concerned by means of open tubes, which would be at all events a much simpler method, is not preferred. It is, however, necessary to investigate first the nature of the force which causes molecules of the various salts to move from the soil to the cell-membranes, which we suppose to be like sieves, and through them into the interior of a plant. A force acting in this sense from without is inconceivable, and we must therefore look for the motive stimulus in the plant itself. As has been already stated in connection with the absorption of carbonic acid, it is believed that the cause of this movement is the disturbance of the molecular equilibrium in the growing vegetable organism. If at one spot in the protoplasm * of a cell a particular substance is altered, and, let us say, converted into an insoluble compound, the previous grouping of molecules appears to be altered, or in other words, the molecular equilibrium is disturbed. To restore equilibrium, there must be a re-introduction of molecules of the material that has been removed ; and the attraction of them from the quarter where they occur in a fluid, that is ; to say in a mobile condition, is the more energetic. Supposing, for instance, gypsum (i.e. sulphate of lime) is being decomposed within a cell, and the lime combines with the oxalic acid (set free in the same cell) to form insoluble oxalate j of lime, whilst the sulphur combines v/ith other elements to form insoluble i albuminoids, this use of the gypsum occasions a violent attraction of that sub- , stance from the environment, or, to put it another way, it causes a movement of i gypsum towards the place of consumption. If this latter place is a cell in imme- diate contact with the nutrient substratum, the absorption of the substance NUTRIENT SALTS. 73 attracted is direct; but if the cell in which the material is used up is separated from the substratum by intervening cells, the attraction must act through all those cells upon it. The substance consumed must be taken in the first place from the cell adjoining the consuming cell on the side towards the periphery; this cell again must take it from its neighbour, which is still nearer the periphery, and so on until the external cells themselves exercise their influence upon the nutrient sub- stratum. Thus, one may regard the growing cells in which substances are used up, as centres of attraction with respect to those substances. This also explains why it is that the influx of food-salts takes place only so long as the plant is grow- ing; and we see, too, that the direction of the current must vary according to the position of the growing cells, and according to the degree of their constructive activity. But that one plant prefers one substance and another another — that one species attracts iodine, a second sodium, and a third iron — can only be interpreted as a result of the specific constitution of the protoplasm. The protoplasm of a growing cell which contains no iodine does not require that substance either, for the pro- cesses of transmutation and storage. A protoplast of this kind will not therefore be a centre of attraction for iodine, but will draw from the environment with great force substances which are its essential constituents. Having gained this conception of the absorption and selection of food-salts, we are able to imagine the possibility of a substance being sought after by one species whilst acting as poison on another. Iodine itself exercises a prejudicial eflect on many plants, even when present in very small quantities. Cell-membranes in immediate contact with a medium containing iodine are modified as regards their structure by the iodine: their pores are enlarged, lose their value as orifices adapted to the admit- tance of certain food-salts in limited quantities, and they no longer prevent the influx of injurious substances. Ultimately they die, and by so doing the entire plant sufiers. On the other hand, plants to which iodine is an indispensable constituent are not hurt in any way by the presence of small quantities of this substance in the nutrient medium : their cell-membranes are neither paralysed nor destroyed, and suction is able to take place through them in a perfectly normal manner. But we must in this case specially emphasize the condition of the amount being small, for a larger quantity of this substance is positively injurious even to plants which require iodine. The general rule for a great number of plants is that they thrive best when the food-salts necessary to them are supplied in very dilute solutions. An increase in the quantity of the salts administered not only fails to promote development, but, on the contrary, arrests it. This is the result even if the salts are such as are absolutely necessary in small quantities to the plants in question. A very minute amount of an iron salt is indispensable to all green plants ; but, if a certain measure is exceeded, iron salts have a destructive effect on the cell-membranes and protoplasm, and cause the plant to die. But at what point the boundary lies between salubrious effects and the reverse, where the beneficial action of particular 74 NUTRIENT SALTS. substances ceases and detrimental action begins, is not known more precisely than has been stated. We only know that different plants behave very differently in this respect. Suppose, for example, that we scatter wood-ash over a field which is overgrown by grasses, mosses, and various herbs and shrubs. The result is that the mosses die; in the case of the grasses growth is somewhat increased; whilst some of the herbs and shrubs, notably polygonaceous and cruciferous plants, exhibit a strik- ingly luxuriant growth. If we scatter gypsum instead, the development of clover is enhanced, and, on the other hand, there are certain ferns and grasses that die earlier when gypsum is supplied, or, at least, are considerably stunted in their growth. The fact that certain plants predominate on calcareous and others on siliceous ground has been the subject of very thorough investigation; and these researches were regarded as justifying the assumption that particular species require a more or less considerable quantity of lime for food, whilst others require similarly silicic acid. Hereupon was founded a division of plants into those which required and were tolerant of lime, and into such as required and tolerated silica. The explana- tion given of these facts does not seem, however, to be satisfactory, at any rate in the case of siliceous plants. It is much more probable that the so-called silica- loving plants are produced on ground composed of quartz, granite, or slate, not by reason of the abundance of silicic acid, but because of the absence of lime in any large quantity, such as would be liable to injure plants of the kind; for only traces of lime are found, and its presence to this extent is absolutely necessary for every plant. This is not of course inconsistent with the fact that individual species require larger quantities of particular food-salts and only flourish luxuriantly when these nutritive salts are not meted out too sparingly. In the case of oraches, thrifts, wormwood species, and cruciferous plants, alkalies, in comparatively large quantities, are necessary for hardy development. The proper habitat for these plants, therefore, is on soils which contain an abundance of easily soluble alkaline compounds, in places where the ground is regularly saturated by saline solutions, and where crystals of salt effloresce on the drying surface. Such places are the sea-shore, the salt steppes, and the neighbourhood of salt-mines. The above plants not only flourish in these localities in great abundance and perfection, but they supplant all other species on which the excessive provision of soluble alkaline salts is not beneficial. If the seeds of such plants happen to fall upon the salt ground they germinate, but only drag out a miserable existence for a short time, and in the end are crowded out by the luxuriant oraches and crucifers. Plants which only flourish abundantly on soils rich in alkaline salts are called halophytes. The same name has also been applied to plants which only thrive in sea- water. Most of the species used by us as edible vegetables, as, for instance, cabbages, turnips, cress, &c., are really descended from halophytes, and accordingly require a soil that contains a comparatively rich supply of alkalies. An oppor- tunity will occur, later on, of returning to the question as to how far agriculture has gained by all these discoveries, and of considering what processes, based upon ABSORPTION OF FOOD-SALTS BY WATER-PLANTS. 75 the results of scientific research, have been introduced into practice. Amongst these processes may be mentioned the rotation of crops, the artificial application of manure to exhausted land, and the restitution of the mineral food-salts which the particular plants last cultivated have withdrawn from the land under tillage. ABSORPTION OF FOOD-SALTS BY WATER-PLANTS. It is usual to designate all plants that grow in water as hydrophytes or water- plants. But in their narrower sense these names are only applicable to those plants which, during their entire lives, vegetate under w^ater and derive their nutriment, especially carbonic acid, direct from the water. A number of plants have widely ramifying roots fixed in the earth at the bottom of water, and the lower parts of their stems, either temporarily or throughout life, immersed in water, whilst the upper parts of their stems and their upper leaves are exposed to the air and take carbonic acid direct from the atmosphere, and these should be regarded as marsh- plants and classed with land-plants so far as regards food-absorption. Reeds and rushes, water-fennel and water-plantain, the yellow water-lily, even the amphibious Polygonum and the white water-lily, are marsh-plants and not true hydrophytes. It is characteristic of all these marsh-plants, that if they are entirely submerged for any length of time they die, whereas they are not injured if the water's level at the place where they grow sinks so as to expose the lower portions of the stem. In places formerly submerged, but from which, in course of time, the water has retreated, so that they have been turned into meadows, one may come across not only clumps of reeds and rushes but even yellow and white water-lilies, flourishing perfectly on the moist earth. Water-plants, or hydrophytes in the proper acceptation of the term, perish if they are kept for a length of time out of their proper medium and exposed to the air. In most of them death ensues quickly, for their delicate cell-membranes are not able to prevent the exhalation of water from the interior of their cells; and, there being no provision for a replacement of the evaporated fluid, the whole plant dries up. If one supplies aquatic plants, thus desiccated, with water, though it is indeed absorbed it no longer has the power of reviving them. Those hydrophytes which occur in the sea, near the shore, are able to stand exposure to the air for a comparatively long time, and they are regularly sub- ject to it during ebb-tide. Sea-wracks which at high-tide were floating in the water are then seen lying on the dry rocks or sand of the shore. But the mem- branes of the cells forming the outermost layer in all these sea- wracks is very thick. They retain water staunchly and prevent the plants from drying up, at least until high-tide occurs again, when they are once more submerged. Amphibious plants in which the lower leaves are like those of aquatics and the upper like those of land-plants so far as desiccation is concerned {e.g. several kinds of pond-weed — Potamogeton heterophyllus and P. nutans — and a few white-flowered Ranunculi — Ranunculus aquatilis and R. hololeucus), exhibit a transition stage from 76 ABSORPTION OF FOOD-SALTS BY WATER-PLANTS. aquatic plants to land-plants. When the water sinks and they are finallj' left lying exposed on the mud or wet sand, to which they appear to be firmly attached by their abundant roots, it is only the previously submerged leaves that dry up. That part of the foliage which floated on the surface and was consequently always in contact with the air continues to thrive, and any fresh leaves that may be developed adapt themselves completely to the new environment. Similar behaviour is ob- served in many of the plants which float freely on the surface of water. Such, for instance, is the case with some species of duckweed (Lemna tninor and L. polyrrhiza), with Azolla, Pontederia and Pistia; they do not die when the water sinks, leaving them stranded, but absorb food-stufls from the wet earth through their roots, and in this condition are not to be distinguished from land-plants. Hydrophytes in the narrow sense, i.e. plants which are entirely submerged and die if they are surrounded by air instead of water for any length of time, are for the most part fixed to some support beneath the water. In many cases the characteristic method of reproduction consists in the separation of special cells, which then swim about for a time in the water. Sooner or later, however, they re-attach themselves to some seemingly suitable spot, and the further phases of their development are again stationary. Comparatively few permanently submerged species are freely suspended in the liquid medium in every stage of development. Such free plants are liable to be shifted by currents in the water, but the extent of their displacement is never very great, owing to the fact that submerged species of this kind occur almost exclusively in still water. As instances may be mentioned the ivy-leaved duckweed (Lemna trisulca), the water-violet (Hottonia 2^CLl'^stris), the various species of horn wort {GeratophylluTn), in all of which roots are absent; and in addition amongst the lower or cryptoganiic plants liiccia Jiiiitans, and many of the Desmidiacese, Spirogyras and Nostocineae. Some of these aquatic plants periodically rest on the bottom of the pond or lake in which they live. An example is afforded by the remarkable plant known as the water-soldier (Stratiotes aloides), w^iich, as is indicated by its Latin name, is not unlike an aloe in appearance. During the winter, this plant rests at the bottom of the pond it inhabits. As April draws near, the individual plants rise almost to the surface and remain floating there, producing fresh sword-shaped leaves and bunches of roots which arise from the abbreviated axis, and finally flowers which, when the summer is at its height, float upon the surface. When the time of flowering is over, the plant sinks again to mature its fruit and seeds, and develop buds for the production of young daughter-plants. Towards the end of August, it rises for the second time in one year. The young plants that have meantime grown up resemble their parent completely, except that their size is smaller. They grow at the end of long stalks springing from amongst the whorled leaves, and the stately mother-plant is now surrounded by them like a hen by her chickens. During the autumn, the shoots connecting the daughter-plants with their parent rot away, and, thus isolated, each little rosette, as well as the mother-plant, sinks once more to the bottom of the pond and there hibernates. ABSORPTION OF FOOD-SALTS BY WATER-PLANTS. 77 Altogether the number of submerged plants which live suspended in water is very small. As has been said before, by far the greater number are attached some- where. Seed-bearing plants or Phanerogamia., such as Vallisneria, Ouvirandra, Myriophyllum, Najas, Zannichellia, Ruppia, Zostera, Elodea, Hydrilla, and several species of Potamogeton (P. pectinatus, P. pusillus, P. lucens, P. densus, P. crispjus); as also Cryptogams, such as the various species of Isoetes and Pilularia and sub- merged mosses, are fastened in the mud under water by means of attachment-roots or of rhizoids, whilst the almost illimitable host of brown and red sea-weeds are fixed by special cells or groups of cells, which are often root-like in appearance. The sea-weeds choose rocks and stones, by preference, for their support, but they also make use of animals and plants. The shells of mussels and snails are often completely overgrown by brown and red sea-weeds. Larger kinds of Fucaceae, especially the species of Sargassum and Cystosira, which form regular submarine forests, bear upon their branches numerous other small epiphytes, chiefly Florideoe, and these again are themselves covered by minute Diatomaceae. Many of the huge and lofty brown sea-weeds which raise themselves from the bottom of the sea, remind one forcibly of tropical trees covered with Orchideae and Bi'omeliaceae, whilst the latter are themselves overgrown by Mosses and Lichens. These epiphytes are for the most part, however, neither parasitic nor saprophytic. In general hydrophytes attached by means of single cells or groups of cells derive no nutriment, i.e. no food-salts, from the support they rest upon. When loosened from the substratum they continue to live in the water for a long time; they increase in size, and if they come into contact with a solid body are apt to attach themselves to it. In this connection it is well worthy of remark that certain Crustacea have their carapaces entirely covered by hydrophytes of this kind, and that it takes a very short time for the plants to establish themselves upon them. For instance, some species of crabs, such as Maja verrucosa, Pisa tetraodon and P. armata, Inachus scorpioides and Stenorrhyncus longirostris, cut off bits of Wracks, Floridese, Ulvae, &c., with their claws, and place them on the top of their carapaces, securing them on peculiar spiky or hooked hairs. The fragments grow firmly to the crabs' chitinous coats, and far from being harmful to the animals are, on the contrary, an important means of protection. The crabs in question escape pursuit in con- sequence of this disguise, and it is to be observed that each species chooses the very material which makes it most unrecognizable to plant upon the exterior of its body: those species which live chiefly in regions where Cj^stosiras are indigenous deck themselves in Cystosiras, whilst those which inhabit the same places as Ulva?, carry Ulvse on their backs. This phenomenon has for us a special interest in that it shows that the w^ater- plants we are discussing draw no food-salts from their place of attachment, and that accordingly the chemical composition of the support is a matter of utter indiflerence to all these Fucaceae, Florideae, Ulvae, &c. There is no doubt that food-salts are absorbed by these hydrophytes from the surrounding water through their whole surface. Accordingly the structure of their peripheral cells is much simpler than is the case in land-plants. In the latter very 78 ABSORPTION OF FOOD-SALTS BY WATER-PLANTS. complicated adaptations are necessary for the extraction of food-salts from the earth. In particular, the portions which are exposed to the air above ground exhibit a number of special structures connected with this extraction. These structures (cuticle, stomata, &c.) are superfluous in the case of aquatic plants, for there is with them no necessity for raising and conducting food-salts into the parts where they can be used up. Moreover the absorption of nutritious matter is much simpler, inasmuch as it is not necessary for the absorbent parts to search for a perpetual source of the requisite substances. The roots of land-plants have often to range over a wide area in order to find sufficient nourishment in the earth, and frequently they have then to liberate it, i.e. bring it into a state of solution. This is not the case with water-plants. They are completely surrounded by a medium which is itself to a large extent a solution of food-salts, and no sooner are substances withdrawn by the absorbent cells from the layers of water immediately bounding them than those substances are again supplied from the more remote environ- ment. Constant compensating currents occur in water, and there is, therefore, scarcely an aquatic plant towards which there is not a perpetual flow of the food- salts it requires in a form suitable for absorption. In connection with this kind of food-absorption there is also the fact that the parts by which hydrophytes attach themselves to a support are relatively small in area. Fucoids, as large as hazel trees in height and girth, are fixed to submerged rocks by groups of cells perhaps only 1 cm. in diameter. The quantity of food-salts absorbed by hydrophytes is very considerable com- pared with the amounts absorbed by other plants. As has been mentioned before, soda and iodine play a very important part in the thousands of different varieties which live in the sea. If Floridese are transferred from the sea into pure distilled water, common salt and other saline compounds dififuse out of the interior of the cells through the cell-membranes into the fresh water around. The red colouring matter of these Floridese also passes through the cell-walls into the water, proving that the molecular structure of the membrane is adapted to the agency of salt water in the osmotic processes of food-absorption. Plants living in fresh, or in brackish water, likewise absorb relatively large quantities of food-salts; and this accounts for the fact that water which is very poorly provided with nutriment of the kind contains only very few vegetable species. One would expect that exceedingly abundant vegetation would be evolved in running water, provided the latter contained food-salts in solution, however small they might be in quantity. For, in such a situation, it is not necessary to wait for the salts withdrawn by the plants from their immediate environment to be restored by the slow processes of mixture and equilibration; the water which has been drained of nutriment is replaced the next moment by other water bearing fresh food-salts. Experience shows, however, that flowing water is not so favourable to the develop- ment of hydrophytes as is the still water of pools, ponds, and lakes. This may partly depend on the fact that running water is always poorer in food-salts, and ABSORPTION OF FOOD-SALTS BY LITHOPHYTES. 79 partly also on the circumstance that mechanical difficulties are opposed to the takinf up of saline molecules from water in rapid motion. There are only a few plants that are able to absorb under these conditions, and these choose, by preference, the very spots where they are most exposed to the dash of the water. Thus, certain Nostocinese (Zonotrichia, Scytonema) are to be found constantly in waterfalls at the parts where the most violent fall occurs. Lemanea, Hydrurus, and many mosses and liverworts, grow by preference in the foaming cascades of rapid torrents. Amongst flowering plants we only know of the Podostemacese as choosing a habitat of this kind. Podostemacese are exceedingly curious little plants, which at first glance one would take for mosses or liverworts without roots. Some of them, e.g. the Brazilian species of the genus Lophogyne and the various species of Temiohi growing in Ceylon, exhibit no differentiation into stem and leaves, but are only represented by green fissured and indented lobes attached to stones. They belong without exception to the tropical zone, and occur there in the beds of streams, attached to rocks, over which the foaming water rushes. ABSORPTION OF FOOD-SALTS BY LITHOPHYTES. Nothing would seem more natural, as to the absorption of mineral salts by lithophytes, than that the stone which constitutes their support should yield the salts, and that the attached plants should suck them up; but, generally speaking, the case is not so simple. There are mosses and lichens which cling to the surfaces of rocks on mountain tops. These rocks are sometimes composed of perfectly pure quartz, and yet the plants in question contain very little silica; they contain, on the other hand, a number of substances entirely wanting in the composition of the underlying rock, and which could not, therefore, have been derived from that source. For many of these lithophytes the rock is, in the main, only a substratum for attachment, and in no way a nutrient soil; just as, in the case of many aquatic plants, the stones to which they cling by their discs of attachment are anything but sources of nourishment. From what source, then, do stone-plants of this kind derive the food-salts which are wanting in their substratum ? It may sound paradoxical, but it is nevertheless the fact, that they obtain those salts from the air through the medium of atmospheric precipitation. Rain and snow not only absorb carbon dioxide, sulphuric acid, and ammonia — which occur in air universally, although in extremely minute quantities — but they also collect, as they fall, floating particles of dust. The opinion is widely entertained that although the atmosphere is full of dust in the neighbourhood of cities and human settlements generally, where the soil is laid bare and ploughed up, and roads and paths have been made for purposes of traffic, and perhaps also over steppes and deserts where large areas of ground are destitute of vegetation, yet that there is no dust in the air over land remote from places of that kind or in the air of marshes, lakes, or seas. This notion has certainly some warrant if we regard as dust only the coarser particles which are raised from loose earth and 80 ABSORPTION OF FOOD-SALTS BY LITHOPHYTES. whirled into the air by the wind. Moreover, the quality of the dust will no doubt be charactei'istically affected by the vicinity of areas of industry. One has only to look at the sooty leaves and branches of trees in parks near manufactories to convince oneself of the reality of this influence. But it would be quite erroneous to suppose that the air in regions far from land that has been cultivated or otherwise opened up is free from dust. It contains dust everywhere. There is dust in the air of the extensive ice-fields of arctic regions and of high mountain glaciers, and there is dust in the air of great forests and over the boundless sea. If the rays of the setting sun fall obliquely through a gap between two peaks in a wood-clad mountain valley, sun-motes may be seen floating up and down and in circles, just as they do in a room when the last rays before sunset fall through the window. These motes are of course not usually visible, and they are moreover much smaller than the particles of dust which are raised by the wind from roads and then again deposited. Now, when rain falls, it takes the sun-motes from the air and brings them down to earth, and the air is thus washed to a certain degree of purity. This Happens still more completely in the event of snow. The latter acts not unlike a mass of gelatine used to purify cloudy liquids, its effect being to drag down with it all the particles to which the turbidity is due, leaving the upper part of the liquid quite clear. Similarly, falling snow-flakes filter the air; and, mixed with fallen snow, there are accordingly innumerable particles of dust. If afterwards the snow gradually melts, it dissolves some of the dust, which then drains away into chinks and depressions; but a portion remains behind undissolved. This portion is gradually consolidated, and then appears lying on the parts of the snow that are still unmelted in the form of dark patches, streaks, and bands; often also it forms a smeary graphitic covering so widely spreading over the last remnants of melting snow that the latter resemble lumps of mud rather than snow. Accord- ingly we find it everywhere — in regions cultivated and uncultivated, in tilled lowlands and on high grassy plains above forest limits, where no tilled land is to be seen in any direction, and lastly in arctic regions in the middle of glaciers several miles across. All this snow dust is not invariably deposited as a result of the filtering of the air by falling snow-flakes; an additional supply is brought by the winds which blow across the snow-fields. It is not of rare occurrence in the Alps for snow- fields to exhibit suddenly, after violent storms, an orange-red coloration. On closer inspection one finds that the surface of the snow is strewn with a layer of powder, infinitesimally fine and for the most part brick-red, which has been brought by the gales. Investigation of this " meteoric dust " shows that it is composed chiefly of minute fragments of ferruginous quartz, felspar, and various other minerals. Mixed with these there are, however, sometimes remnants of organic bodies, such as bits of dead insects, siliceous skeletons of diatoms, spores, pollen-grains, tiny fragments of stems, leaves, and fruits, and the like. Once, after a south wind had prevailed for several days, the snow-fields of the Solstein range near Innsbruck were covered, at a heiMit of from two to three thousand meters above the sea-level. ABSORPTION OF FOOD-SALTS BY LITHOPHYTES. 81 with millions of a species of Micrococcus, which lent a rosy hue to vast expanses of snow. Most of the dust in the atmosphere originates, doubtless, from our earth. The air that blows in waves over the earth can carry along with it not only dead and detached portions of plants, but also loose particles of rock, sand, earth, and dried mud. If one draws one's palm across the weather side of a dry rock composed of dolomitic limestone, gneiss, trachyte, or mica-schist, the surface of the stone always feels dusty, and the slightest movement of the hand is sufficient to detach a number of particles which were already separate from the rock and only held in loose con- nection with it. This dust is liable to be detached and carried away by any strong gust of wind. Larger and heavier particles are not, it is true, lifted much above the ground; they are rolled and pounded along and thereby reduced to a still finer powder. This finer dust may then be scattered afar by gales blowing horizontally, or even ascend into higher atmospheric strata. The finest dust in particular, how- ever, is carried up into the higher layers of the air by the currents which ascend from the earth in calm weather; and this applies not only to the tropics but to the temperate zones as well, and even to the frigid regions of the arctic zone. When, therefore, this dust is brought back by rain or snow from the upper aerial strata to the earth, it but completes a circuit. Indeed it is highly probable that the particles of dust restored to earth by means of atmospheric deposits recommence their aerial travels as soon as they are thoroughly dry again, and that there is thus a circulation of dust analogous to that of water. There is of course no inconsistency in the fact that meteoric dust, which is often drifted along in surprisingly large quantities, may originate quite suddenly during volcanic eruptions; nay, it is even possible that cosmic dust reaches our atmosphere and thence falls to the earth. Chemical investigation of aerial dust has, no doubt, yielded in most cases only sulphuric and phosphoric acids, lime, mag- nesia, oxide of iron, alumina, silica, and traces of potash and soda, that is to say, the most widely distributed constituents of the solid crust of our earth; but cobalt and copper have also been found in it, over and over again, and it has hence been inferred that the dust in these cases was of cosmic origin. In relation to the question which we have here to answer the above is, after all, almost a matter of indifference. The only important facts are that dust in a state of extremely fine division is blown about in the air, that this dust contains the salts required by plants for their food, that it is carried for the most part mechanically by drops of water and flakes of snow, condensed in the atmosphere, and is partially dissolved, that the atmospheric deposits supply lithophytic plants wath a sufficient quantity of nutrient salts, and that the aqueous solution so supplied is rapidly absorbed by the whole surface of the plants in question. We must not omit to mention here that the demand of lithophytes for mineral food-salts is not very great. In particular the protonemse and even the leafy shoots of Grimmice, Bhacomitrice, Andreoeaceoe and other rock mosses, and the Collemacece and most crustaceous lichens only contain very minute quantities of these substances. Water containing Vol. I. ^1 g 82 ABSORPTION OF FOOD-SALTS BY LAND-PLANTS. the usual mineral salts in about such proportion as is necessary for the cultivation of cereals in fields has actually an injurious effect on these lithoph3'tes and soon kills them. At the end of this section we shall consider what happens to dust which is brought to earth from the air by rain and snow but is not dissolved, and the important part it plays in clothing the naked ground and in changes of vegetation. Here, however, it must be noted that most lithophytes are true dust-catchers, that is to say, they are able to retain, mechanically, dust conveyed to them by wind, rain, and snow, and to use it in later stages of development by extracting nutriment from it. Many mosses are completely lithophytic in early stages of development whilst later they figure as land-plants. AESORPTION OF FOOD-SALTS BY LAND -PLANTS. In no class of plants is the absorption of mineral food-salts accomplished in so complicated a manner as in land -plants. Moreover, this absorption is by no means uniform in different forms of plants, and we must beware of generalizing with regard to processes which have only been traced and studied in isolated groups — perhaps only in the commonly distributed cultivated plants. On the other hand, with a view to synoptical representation, it is not desirable to enter into too great detail or to attempt to describe all the various differences minutely. At the outset, it is difficult to give an accurate account of the soil which constitutes the source of nutriment in the case of land -plants. From the dark graphitic mass composed of sun-motes, which is deposited in the place of a melted layer of snow, to coarse gravel, there is an unbroken chain of transition stages; loam, sand and gravel are only specially-marked members of this chain. Again, just as earth varies in respect of the size of its component parts, so also it varies in the mineral salts it contains, in the amount of admixture of decaying vegetable and animal remains, in the nature of the union of its constituents, and in its capacity to absorb, to retain, or to yield up water. Compare the sand composed of quartz on the bank of a mountain stream with that of calcareous origin which is found impregnated with salt on the sea-shore, or with the sand at the foot of mountains of trachyte, which has an efflorescence of soda-salts. Or compare the granite bed of a desert, bare of soil, with the loam on the granitic plateaus of northern regions where there is an intermixture of the remains of a vegetation for centuries active. How great is the difference in each case! But whatever the kind of earth, it is only of value as a source of nutriment for a plant when the interstices of its various particles are filled with watery fluid for the time during which the plant is engaged in the construction of organic substances. But how is the earth supplied with water? " Das hat iiicht East bei Tag und Nacht, 1st stets auf Wauderschaft bedacht." ABSORPTION OF FOOD-SALTS BY LAND-PLANTS. 83 Streams fall into lakes, rivers into the sea, and hence the water ascends into the atmosphere in the form of vapour, and returns once more to earth as snow, rain, and dew. Through porous earth it percolates until it has filled all the interspaces. If its further descent be impeded by impervious strata, it spreads literally as sub- terranean water, or else comes up at some special spot as a spring. Earth which is richly endowed with decaying vegetable remains is able to absorb vapour in addition from the atmosphere. When this occurs, carbonic and nitric acids are always absorbed along with the aqueous vapour. These are contained, as has been mentioned before, in atmospheric deposits, and another source of these acids is afforded by the decay of dead parts of plants. Water precipitated from the atmosphere, and con- taining carbonic and nitric acids, is able by their means to decompose the compounds in all the rocks which come in its way as it percolates through the ground, especially when its action is long continued. The siliceous compounds or so-called silicates — felspars, mica, hornblende, and augite in particular — and quartz, the anhydride of silicic acid, which form the preponderant mass of the rocks of the solid crust of our earth, either contain a great quantity of silica, alumina, and alkalies, or if they are relatively poor in silica they may be rich in iron. The former are found chiefly in granite, gneiss, mica-schist, and argillaceous slate; the latter preponderate in serpentine, syenite, melaphyr, dolerite, trachyte and basalt. First the felspars are decomposed by the acid water. Their alkalies combine with the carbonic and nitric acids forming soluble salts, and the alumina and silica remain behind as clay. Iron is also converted into soluble salts. The most difiicult substances to decompose are the mica and quartz, and it is on that account that they so often appear in the form of glittering scales and angular nodules mixed with the clay produced from the decomposition of felspar. But, ultimately, even they are unable to withstand the continuous action of the acidulated water. The result of these chemical changes is an earth, which, according to the nature of the parent rock, contains a preponderating amount of clay, of quartzose sand or of mica, which is coloured in various ways by iron compounds. Of substances useful to plants these earths yield generally on analysis the following: potash, soda, lime, magnesia, alumina, ferrous and ferric oxides, manganese, chlorine, sulphuric acid, phosphoric acid, silica, and carbonic acid, sometimes one sometimes another in greater proportion relatively, and traces of many substances often so slight as hardly to be detected. It is true that limestone and dolomite, which, next to the above-mentioned rocks, enter most largely into the composition of the solid crust of the eartli, consist chiefly of carbonate of lime and magnesium carbonate respectiveh^; but wherever they occur in extensive strata and piles, they always contain in addition an admixture of alumina, silicic acid, ferrous oxide, manganese, traces of alkalies in combination with phosphoric and sulphuric acids, &c. Of the carbonates of lime and magnesia a great part is gradually dissolved and carried away upon the invasion of water containing carbonic and nitric acids, and a proportion also of the substances mixed with them, as above mentioned, is lixiviated. What remains 84 ABSORPTION OF FOOD-SALTS BY LAND-PLANTS. bfhin.J then consists of an argillaceous, loamy mass, variously coloured by iron and very similar in appearance to the clay formed from the decomposition of felspar. According to the quantity of the substances mixed with the carbonate of lime in the rock, the loamy earth formed from limestone is either abundant or only in restricted layers, bands and pockets lying on, or intercalated within, the unde- composed debris of the stone. Chemical analysis has resulted in the discovery that there are, as a rule, in loamy earth of this kind the same ingredients avail- able for plants as have been identified in earth produced from silicates; and we are led to believe that earths, collected in widely different places and covering rocks of most various kinds, are much more uniform qualitatively than has been supposed. Only, the relative proportions of the substances forming the mixture are usually ditierent. Silica and the alkalies are less conspicuous in earth derived from limestone, and carbonate of lime in that which is formed from silicates. This difference is particularly striking in instances where the rock consisted almost entirely either of quartz and mica or of nearly pure carbonates of lime and magnesium. In these cases the earth formed is not argillaceous, but of loose consistence, very abundant, and composed, according to the kind of rock, of quartzose sand and mica scales or calcareous and dolomitic sand. The conversion of rocks into earths by the action of water from the atmosphere containing carbonic and nitric acids is, besides, materially modified by the disrup- tions which ensue from changes of temperature, more particularly by the freezing of water within the pores of rocks. It is also affected, though more remotely, by the mechanical action of water and air in motion, and, lastly, by the plants them- selves, which penetrate with their roots into the narrowest crevices and mingle their dead remains with the portions of the rock that are decomposed, broken up, or abraded by chemical and mechanical agencies. The substance produced from a rock in the manner explained is called earth-mould, or simply earth. The matter resulting from the decomposition of plants and animals is designated by the term " humus." Earth which includes an abundance of decomposed fragments of plants, i.e. has a large admixture of humus, is called vegetable mould. Every kind of earth, but especially earth rich in humus and clay, has the power of retaining gases, and especially water and salts. When water containing salts in solution is poured over a layer of dry vegetable mould, it percolates into the spaces between the particles of earth, and speedily drives out of them the air which has but slight adhesion, and which then ascends in bubbles. It is not till all the inter- spaces are full of water, whilst a fresh supply is constantly maintained from above, that any of the liquid oozes out from beneath the stratum of earth. The water remaining in the interstices is held there by adhesion to the particles of earth, and we must conceive each of these particles as surrounded by an adherent film of water. The inorganic salts, infiltrating with the water, are held with still greater energy. The water which trickles from the bottom of the earth always contains a much smaller proportion of salts in solution than that which was poured on above, whence we conclude that the latter are in part absorbed by the earth. ABSORPTION OF FOOD-SALTS BY LAND-PLANTS. 85 The salts are to be regarded as forming an extremely delicate coating round minute particles of earth where they are forcibly retained. If a plant rooted in the earth is to take in these salts it has to overcome the force by which their molecules are detained. This is effected, however, by means of a very powerful attraction exerted by the protoplasts of the plant as they grow, carry on the work of construction, and use up material. What actually happens is an energetic suction by the cells that are in close contact with particles of earth. This suction depends, however, upon the chemical affinity between the substances in the interior of the cells and the salts adhering to the earth-particles, as well as upon the consumption of food-salts for the manufacture of organic compounds within the green cells. It is supposed that whenever salts are abstracted from soil-particles by suction, a restitution of like salts immediately takes place, particles still unresolved in the immediate neighbourhood being dissolved, and a fresh influx taking place from the environment. Consequently the concentration of the solution retained by the earth is always approximately the same, or, at any rate, equilibrium is very quickly restored. One advantage of this is that the cells in immediate contact with particles of earth, and their adherent liquid, can only meet with a saline solution of constant weak concentration, and are therefore secure from injury such as would result in the case of most plants, from contact with a very concentrated solution. In other words, the absorptive power of earth acts as a regulator of the process of absorption of food-salts by plants, and is the means of keeping the saline solution in the earth always at the degree of strength best suited to the plants concerned. Naturally, the passage of salts from the earth to the interior of a plant is dependent on the aid of water containing both the substances composing cell- contents and the food-salts in solution. The cell-membranes, through which absorption takes place, are saturated with this solution. The aqueous films adhering to the particles of earth, the water saturating the cell-membrane, and the liquid inside the cells are really in unbroken connection, and along this continuous water- way the passage of salt molecules in and out can take place easily. The absorption of food-salts directly from the earth by green cells occurs very rarely. The protonema of Polytrichum, which spreads its threads over loamy earth and wraps it in a delicate green felt, and that of the famous Cavern Moss (Schis- tostega), whose long tubular lower cells penetrate the earth in the recesses of caves, do undoubtedly suck up their necessary food-salts by means of cells containing chlorophyll. A drawing of the latter is given in figure 25a, p. The majority of land-plants have, however, special absorptive cells for the taking-up of salts in solution. These cells are imbedded amongst or lodged upon the earth-particles, and are usually in intimate connection with portions of them. Any part of a plant that penetrates into the earth or lies upon it, may, if it performs the function of absorption, be equipped with cells of the kind. Plagiothecium nekeroideum, a delicate moss belonging to the flora of Germany, and growing on earth under overhanging rocks, where it is not exposed to rain, and therefore cannot receive any food-salts through that agency, develops absorption-cells on the apices 86 ABSORPTION OF FOOD-SALTS BY LAND-PLANTS. of its green leaflets. So also does Leucobryum javense, a species native to Java. Several delicate ferns of the family of the Eymenophyllacem exhibit them on their subterranean stems. Many liverworts and the prothalli of ferns bear them on the under surfaces of their flat thalli which lie outspread on damp earth. But most commonly of all are they to be found close behind the growing tips of roots. Their form does not vary very much. On the roots of plants fringing the sources of cold mountain-springs, as on those of many marsh-plants in low-lying land, they are in the form of comparatively large, oblong, flattened, closely united cells, with thin walls and colourless contents. In some conifers, whilst having in the main the siiape just described, they differ in that they are arched outwards so as to form papillfe; but in most other phanerogams the external cell-wall projects outwards, and the whole absorptive cell develops into a slender tube, set perpendicularly to the longitudinal axis of the root (fig. 12*). Seen with the naked eye, or but slightly magnified, these delicate tubes look like fine hairs, and have received the name of "root-hairs." The end of a root often appeal's to be covered with velvety pile, and the absorptive cells are then very closely packed; more than four hundred per square millimeter have been occa- sionally counted. In other cases, however, there are hardly more than ten on a square miUimeter. When in such small numbers they are usually elongated and clearly visible to the naked eye. Their length, for the most part, varies from the fraction of a millimeter to three millimeters, and their thickness between 0008 m.m. and 014i m.m. It is only exceptionally that one meets with plants, rooted in mud, possessing root-hairs 5 m.m. or more in length. The absorptive cells of phanero- gams are almost always simple epidermal cells of the particular part of the plant that bears them, and are not partitioned by any transverse walls. In mosses and fern prothalli, on the other hand, the absorption-cells are generally segmented by transverse septa and are usually greatly elongated. In those liverworts which belong to the genus Marchantia they form a thick felt on the under side of the leaf-like plant, or rather, on such part of it as is turned away from the light, and some of these tangled rhizoids attain a length of nearly 2 cm. The stems of many mosses also are wrapped in a regular felt. This property is rendered very striking in the species of Barbula, Dicranum, and Mnium, and especially in such forms as have bright green leaves, by the reddish-brown colour of the cells in question. Sometimes the long capillary cells of which the felt is composed are twisted together spirally like the strands of a rope. A good instance of this is Polytri- chavi. These fine, hair-like, segmented and branched structures, found on mosses, variously matted and intertwisted, are called rhizoids. But only those cells which come into contact with the earth-particles are truly absorbent. The rest do not serve to imbibe from the ground, but to conduct the aqueous solution of food-salts, after it has been taken up by the absorptive cells, to the stem and to the leaves. The tubular cells resulting from the development of a root's epidermis are placed, as before observed, at right angles to its longitudinal axis. They only grow, how- ever, in earth that is very damp, and even then their course is not always a straight ABSORPTION OF FOOD-SALTS BY LAND-PLANTS. 87 line, for as a rule they describe a spiral as they elongate. Their movement seems as though it were for discovering the most favourable parts of the earth for absorp- tion and attachment. In this manner they penetrate into the interspaces in the earth which are filled with air and water. They also have the power of thrusting aside minute particles of earth, especially if the latter consists of loose sand or mud. If they strike perpendicularly a solid immovable bit of earth, they bend aside and grow round it with their surfaces closely adpressed to that of the obstacle until they reach the opposite point on the other side, when they once more resiime their original direction (fig. 12^). When they encounter large grains of earth they -■a^r. Fig. 12.— Absorptive Cells on Root of Penstemon. •Seedling with the long absorptive cells of its root ("root-hairs") with sand attached. * The same seedling; the sand removed by washing, s Root-tip with absorptive cells ; x 10. * Absorptive cells with adherent particles of earth, s Section through the root-tip ; x 60. sometimes stop and swell up to the shape of a club. The club divides into two or more arms, which grasp and cling to the granule like the fingers of a hand. Many fragments of earth remain thus in the grasp of finger-like processes, whilst others are held fast in the knots and spirals of corkscrew-shaped root-hairs which are often found tangled together. But the retention of most of the earth-particles which adhere to a plant, including fragments of lime, quartz, mica, felspar, &c., as well as plant-residues, is due to the fact that the outermost layer of the absorptive cells is sticky, it being altered into a swollen gelatinous mass which envelops the particles. When this sticky layer becomes dry it contracts and stiffens, and the granules partially imbedded in it are thereby cemented so tightly to the absorptive cells that even violent shaking will not dislodge them. In the case of most seedlings, and in that of grasses, the absorptive cells which proceed from the roots and which are especially numerous in the latter, are generally thickly covered with particles of earth (see fig. 12*). If such a root is pulled out of sandy soil it appears to be completely encased in a regular cylinder of sand (fig. 12 1). A root of Clusia alba, taken from coarse gravel, had its root-hairs so tightly 88 ABSORPTION OF FOOD-SALTS BY LAND-PLANTS. adherent to bits of gravel that several little stones, weighing I'S grms., were found clinging to it when it was lifted. The gelatinous mass, resulting from the swelling- up of the extei'nal coat of the cell, does not in any way hinder absorption or the passage of food-salts in solution. Nor does the inner coat, the thickness of which varies between 00006 m.m. and 0"01 m.m., constitute any impediment to imbibition. In addition to the absorption of nutritive salts by root-hairs, there is also, in many cases, an interchange of materials; that is to say, not only do substances infiltrate from the earth into the absorption-cells, and 5,0 onward into the tissues of a plant, but others pass out of the plant through the absorptive cells into the earth. Amongst these eKminated substances, carbonic acid, in particular, plays an important part. A portion of the earth-particles adhering to root-hairs is decomposed by it, and food-salts in immediate proximity to those cells are hereby rendered available and pass into the plant by the shortest way. Having now seen that land-plants take in food-salts by means of special absorptive cells, it is natural to find that each of these plants develops its absorption-cells, projects them, and sets them to work at a place where there is a source of nutritive matter. The parts that bear absorptive cells will accord- ingly grow where there are food-salts and water, which is so necessary for their absorption. The Marchantias and fern prothalli spread themselves flat upon the ground, moulding themselves to its contour. From their under-surfaces they send down rhizoids with absorptive cells into the interstices of the soil. Roots provided with root-hairs behave similarly. If a foliage-leaf of the Pepper-plant or of a Begonia be cut up, and the pieces laid flat on damp earth, roots are formed from them in a very short time. The roots on each piece of leaf proceed from veins near the edge, which is turned away from the incident light, and grow vertically downwards into the ground. It is matter of common knowledge that roots which arise upon subterranean parts of stems, like those formed on parts above-ground, grow downward with a force not to be accounted for by their weight alone. This phenomenon, which is called positive geotropism, is looked upon as an efiect of gravitation. The idea is that an impetus to growth is given by gravity to the root-tip, and that a trans- mission of this stimulus ensues to the zone behind the tip where the growth of the root takes place. It is noteworthy that if bits of willow twigs are inserted upside down in the earth, or in damp moss, the roots formed from them, chiefly on the shady side, after bursting through the bark, grow downwards in the moist ground, pushing aside with considerable force the grains of earth which they encounter. The appearance of a willow branch thus reversed in the ground is all the more curious inasmuch as the shoots, which are developed simultaneously with roots from the leaf -buds, do not grow in the general direction of the buds and branches, but turn away immediately and bend upw^ards. Thus the direction of growth of roots and shoots produced on willow-cuttings remains always the same, whether the base or the top of the twig used as a cutting is inserted in the earth. A similar phenomenon is observed if the leafy rootless shoot of a succulent herb (e.g. Sedum reflexuTn) is cut ABSORPTION OF FOOD-SALTS BY LAND-PLANTS. 89 off and suspended in the air by a string. Whether it hangs with the apex upper- most, i.e. in the position in which it grew naturally, or with the apex towards the ground, it always, in a short space of time, produces roots which spring from the axis between the fleshy foliage-leaves and bending sharply grow to the earth. Thus in the former case their direction is contrary to the apex of the shoot; in the latter, curiously enough, it is in the same direction. If the height at which the shoot is suspended is only 2 cm. above the earth, the roots growing towards the ground develop their root-hairs 2 cm. from their place of origin. But if the shoot is at a distance of 10 cm., the roots only develop their root-hairs when they have attained a length of 10 cm. The rule is, therefore, for the roots to grow until they reach the nutrient soil without developing absorption-cells, and only to provide themselves with them when they are in the earth. It is to be observed that these roots are produced on the suspended shoot at places where, under normal conditions (i.e., if the shoot were not cut off" and hung up), no roots would be developed. Subject to abnormal conditions and liable to starvation, the plant sends out these roots for self-preservation. Phenomena of this kind force one to conclude that a plant discerns places which ofler a supply of nutriment, and then throws out anchors for safety to those places. This power of detection may, undoubtedly, be explained by the influence which conditions of moisture, in addition to the action of gravitation, have on the dii-ection taken by growing roots. The root-hairs can only obtain food-salts when the ground is thoroughly moist; and whenever roots, or rather their branches, have to choose between two regions, one of which is dry and the other wet, they invariably turn towards the latter. If seeds of the garden-cress are placed on the face of a wall of clay which is kept moist, the rootlets, after bursting out of the seeds, grow at first downwards, but later they enter the wall in a lateral direction. The longitudinal growth of the roots is greater on the dry side than on the wet side, and this results in a bending of the whole towards the source of moisture, in this instance the damp wall. It has been established that the tip of a rootlet is very sensitive to the presence of moisture in the environment. Where there is a moist stratum on one side and a dry stratum on the other, a root-tip receives a stimulus from the unequal conditions in respect of moisture; the stimulus is propagated to the growing part of the root, which lies behind the tip, and the result is a curvature of the root towards the moist side. Thus, the presence of absorbable nutriment, or rather of moisture, in the ground explains the divergence of roots from the direction prescribed by gravity. The extent to which the direction taken by roots in their search for food is dependent upon the presence of that food, and the fact that roots grow towards places that aflbrd supplies of nutritious material, are strikingly exhibited, also, by epiphytes growing on the bark of trees, such as tropical orchids and Bromeliaceoe ; and again by plants parasitic on the branches of trees, of which tlie Mistletoe and other members of the Loranthacece afford examples. Although the absorption of food by these plants will not be thoroughly discussed till a later 90 ABSORPTION OF FOOD-SALTS BY LAXD-PLANTS. stage, this is the proper place to mention the fact that in them positive geotropism appears to be completely neutralized. The growing rootlets which spring from the seed, and the absorptive cells produced from minute tubercles, grow upwards if placed on the under surface of a branch, horizontally if placed on the side, and downwards if on the upper surface. Thus, whatever the direction, they grow towards the moist bark which affords them nourishment. Positive geotropism seems to be quite abolished also in those marsh-plants which live under water. When, for instance, the seed of the Water-chestnut (Trapa natans) germinates under water in a pond, the main root emerges first from the little aperture of the nut and begins by growing upwards. Soon the smaller scale-like cotyledon is put forth, whilst the other, which is much larger, remains within the nut. The whole plant so far is standing on its head, as it were, and is growing upwards with its principal root directed towards the surface of the water. Gradually the leafy stem emerges from the bud between the two coty- ledons, and likewise curves upwards and grows towards the surface, whilst an abundance of secondary roots is developed at the same time from the main root. Their function is to absorb nutritive substances from the water around, now that the materials for growth stored in the seed are exhausted. Finding an aqueous solution of food-salts everywhere these roots grow in all directions, upwards, dowuAvards, or horizontally to right or left, forwards or backwards, only they carefully avoid touching one another or interfering with each other's sphere of absorption. It is not till much later that the main root changes the direction of its apex and bends downward. New roots are then produced from the stem; but this subject has no further bearing on the problems at present before us. The movements of roots, as they grow in earth, suggest that they are seeking for nutriment. The root-tip traces, as it progresses, a spiral course, and this revolving motion has been compared to a constant palpitation or feeling. Spots in the earth which are found to be unfavourable to progression are avoided with care. If the root sustains injury, a stimulus is immediately transmitted to the growing part, and the root bends away from the quarter where the wound was inflicted. When the exploring root-tip comes near a spot where water occurs with food-salts in solution, it at once turns in that direction, and, when it reaches the place, develops such absorptive cells as are adapted to the circum- stances. As has been mentioned before, the roots of most land-plants bear root-hairs on a comparatively restricted zone behind the growing point (see fig. 12^), and these hairs have only an ephemeral existence. As the root grows and elongates, new hairs arise (always at the same distance behind the tip), whilst the older ones collapse, turn brown, and perish. In ground which contains on every side food-salts in quantities adequate to the demand, and sufficient water to act as solvent and as medium for the transmission of the salts, the absorptive cells are rarely tubular, but exhibit themselves, as already described, in the form of flat cells destitute of outward curvature. This is the case, for instance, with those Alpine plants which grow in ABSORPTION OF FOOD-SALTS BY LAND-PLANTS. 91 ever-moist hollows and depressions in proximity to springs {e.g. Saxifraga ccizoides and many others). But wherever the substances to be absorbed are not so easily obtained, the surfaces of the absorptive cells are increased by means of a protrusion of the outer cell-wall, the whole cell being converted into a tube. These tubular absorptive cells are most elongated in mossy forests, where rather large gaps occur not infrequently in the soil. When a root in the course of growth reaches one of these lacunae, filled with moist air, its root-hairs often lengthen out to an extraordi- nary extent, and sometimes attain to twice the length of those which are in compact soil. The absorptive cells on the roots of the Water-hemlock (Gicuta virosa) and the Sweet Flag (Acorus Calamus) do not project at all if the earth in which they grow is muddy; whilst, if the earth is only slightly damp, and an increase of surface is therefore advantageous, the absorptive cells become tubular. Plants which grow in ground liable to periodic drought, and which at these times must secure all the moisture retained by the earth to save their aerial portions from death by desiccation, endeavour to obtain as great an area of absorption as possible by the development of long tubular cells. The fact must not be overlooked, however, that the form and development of absorptive cells depend partly on the quantity of water that is given off from the aerial parts of the plant, that is to say, by the transpiration of the foliage-leaves. Plants which lose a great deal of water in this way must provide for abundant resti- tution. They must absorb from as large an area as possible, and enlarge their absorp- tive surfaces adequately by pushing out the cells into long tubes. For this reason all plants with very thin, delicate, expanded foliage-leaves, which transpire readily and abundantly, have numerous long tubular root-hairs. Examples are afforded by Viola hiflora and the various species of Impatiens. On the other hand, plants with stiff, leathery leaves, being protected by a thick epidermis from excessive transpira- tion, as, for instance, the Date-palm, exhibit flat, non-protuberant absorptive cells, because there is a very limited amount of evaporation from these plants, and the quantity of water to be absorbed to replace what is lost is therefore small. The same thing holds in the case of evergreen Conifers, in which, owing to the structure of the stiff needles and to the peculiar formation of the wood, water is conducted very slowly from the roots to the transpiring green organs. It has been ascertained that they exhale from six to ten times less vapour than do ashes, birches, maples, and other flat-leaved trees growing on the same ground. We shall presently return to the question of the substitution for absorptive cells in many coniferous and angiospermic trees and in evergreen Daphnacece, EricacecB, Pyrolacea^, Epacridece, &c., of the mycelium of fungi, and shall treat also of the importance of the form of the absorptive cells, and of the roots which bear them, in relation to the mechanism of striking root in the ground. 92 RELATIONS OF FOLIAGE-LEAVES TO ABSORBENT ROOTS. EELATIOXS OF THE POSITION OF FOLIAGE-LEAVES TO THAT OF ABSORBENT ROOTS Anyone who has ever taken refuge from a sudden shower under a tree will remember that the canopy of foliage afforded protection for a considerable time, and that the ground underneath was either not wet at all, or only slightly so. No doubt some of the rain flows down the bark of the trunk, and in many species, as, for instance, the Yew and the Plane-tree, the volume of water conducted down the trunk is considerable; but in the case of most trees the rain-water which reaches the earth in this manner is not abundant, and in comparison with that which drips from the peripheral parts of the foliage its quantity is negligeable. This phenome- non is dependent upon the position of the foliage-leaves relatively to the horizon. In almost all our foliage-trees — in limes and birches, apple and pear trees, planes and maples, ashes, horse-chestnuts, poplars, and alders — these organs slope out- wards, and are so placed one above the other that rain falling upon a leaf on one of the highest branches flows along the slanting surface to the apex, collects there in drops, and then falls on to a lower leaf whose surface is also inclined outwards. Here it coalesces with the water fallen directly upon this leaf; and so it goes from one tier to another, lower and lower, and at the same time further and further from the axis, till a number of little cascades are formed all round the tree. From the under and outermost leaves of the entire mass of foliage the water falls in great drops to the ground, and after every shower of rain the dry area at the foot of the tree is surrounded by a circular zone of very wet earth. It is only necessary to dig at these places to convince one's self that the tree's absorptive roots penetrate the earth precisely to the wet zone. When a tree is j^oung, its roots lie in a small circle, and the crown too is not extensive, so that the damp zone is proportionately restricted. But as the latter is enlarged there is a corresponding elongation of the roots in their search for moisture, and thus roots and foliage progress ^mri passu in peripheral increase. It seems not improbable that the custom amongst gardeners and foresters of trimming the foliage and roots of trees when the latter are transplanted is to be attributed to the phenomenon above described. For the rule is observed that the branches of the trunk and those of the root must be about equally shortened, and accordingly the suction - roots, as they develop, reach the zone of drip of the growing crown. A similar method of carrying off water is to be observed in coniferous trees. Take, for example, the Common Pine. The lateral branches are horizontal near the main trunk; the secondary branches curve upwards like bows. The needles | near the tip of each of the latter slant obliquely upwards from the axis, whilst j the older needles, situated on the under side of the part of the branch which is \ almost horizontal and at some distance from its extremity, are directed obliquely i downwards and outwards. Rain-drops striking the upturned needles glide down | them to the bark of the branch in question, and thence to other needles whose ! RELATIONS OF FOLIAGE-LEAVES TO ABSORBENT ROOTS. 93 inclination is downwards and outwards. On their apices great drops are gradually formed, which finally detach themselves and fall on to the mass of needles be- longing to a lower branch. Thus transmitted, the rain-water travels through the foliage lower and lower and at the same time further from the axis. This is also the case with larches. The drops of rain which fall upon the erect needles of the tufted " short branches " collect and gradually descend to the needles of the drooping " long branches " on lower boughs. Large drops are always to be seen on their drooping apices, whence they drip to the earth. Owing to the pyramidal form of larches, and to the circumstance that the long shoots on each branch are terminal, almost all the water which falls upon one of these trees reaches the long shoots hanging down from the lowest branches, which discharge most of all. Although larches with their tender needles do not look at all as though they would be any protection against rain, the ground underneath them keeps dry nevertheless, the principal part of the water falling upon them being conducted to the periphery. Indeed, the larch belongs to the number of trees which conduct almost all the rain that falls upon them to a certain distance from the axis where the absorbent roots lie, and only allow a little to trickle down the bark of the main trunk. Many shrubs and perennial herbs also transmit the water, which falls on their upturned laminae, to parts of the ground where their absorbent roots are embedded; or, rather, the roots send forth their branches bearing absorptive cells to the area which is kept moist by drippings from the leaves. Particularly striking in this respect are the species of the two genera of Aroids Colocasia and Caladium. A specimen of the latter is figured below (fig. 13^). If one digs about individuals of this genus cultivated on open ground, one invariably finds that the tips of the lateral roots, which proceed in a horizontal direction from the bulbous root-stock, are buried under the point of the great leaves which slope obliquely outwards. We must not omit to mention, in addition, that the stalks of leaves which conduct the rain centrifugally are not channelled on the upper surface; they are round, and comparable to wires supporting at their upper extremities the laminae in an outward and downward direction. As instances we may quote the Horse-chestnut, Maple, and Lime, and many shrubby, suflfruticose, and herbaceous plants, such as Sparm.annia, Spiraea, Aruncus, and Corydalis, and also climbing and trailing plants (e.g. Menispermum, Banisteria, Aristolochia, Hoya, Zanonia, and Tropceolum). Whenever a system of grooves is developed on the surface of an outward sloping leaf, the channels run along the veins and terminate at the apex of the leaf, or at the apices of the leaf's lobes, and invariably cause the water to travel, not to the basal part, but to a spot on the margin whence j it will detach itself in the form of a drop, and fall upon the leaves situated I immediately below and at a greater distance from the axis. A striking contrast to these trees and shrubs, climbing and trailing plants, and sufFruticose and herbaceous species, with their absorptive roots lying in one plane, and usually spreading at but little depth, is afforded by plants which possess 94 RELATIONS OF FOLIAGE-LEAVES TO ABSORBENT ROOTS. bulbs or short root-stocks with deep-reaching suction-roots, and those which have tap-roots descending vertically in continuation of the main stem, and whose second- ary roots are short and travel only a little distance from their places of origin. This other extreme in root-structure, which is represented in fig. 13-, has its counterpart above-ground in the form and direction of the laminae upon which the rain falls. In all these plants the surfaces of the leaves are not directed Fig. 13.— Centrifugal and Centripetal Transmission of Water. 1 By a Caladitim. 2 By a Rhubarb plant. outwards, but slope obliquely towards the central axis. Their upper sides, more- over, are concave and exhibit a system of grooves, which conveys the water collected by the leaf towards the stem, and therefore also, towards the tap-root and suction- roots. The leaves of bulbous plants, such as the Hyacinth and Tulip, all stand up obliquely, and their upper surfaces are concave and often deeply channelled. Along the grooves the rain flows centripetally downwards, and so directly reaches the part of the earth where the bulbs and suction-roots, which proceed in a tuft from underneath the bulbs, are situated. The young leaves of Cannaceae and of the Lily -of -the -valley are coiled up like a trumpet; and rain, falling from above upon the expanded portion, is led along the coiled surface, describing a helix as RELATIONS OF FOLIAGE-LEAVES TO ABSORBENT ROOTS. 95 it goes, to the earth in the neighbourhood of the absorptive roots, whieli proceed from the short root-stock. When the leaves of plants furnished with tap-roots are arranged in whorls, and are without internodes, and the rosette rests upon the ground, as is the case in the Mandrake, the Dandelion, and several species of Plantain (Mandragora officinalis, Taraxacum officinale, Plantago media), there are always one or more main grooves on the upper surfaces of the leaves, and the leaves have always such form and position as compel the rain which falls upon them to flow centripetally, i.e. towards the tap-root growing vertically beneath the centre. Plants with petiolate leaves, which conduct rain centri- petally, always have on the upper side of each leaf -stalk an obvious groove, the depth of which is frequently increased by the development of green or (in many cases) membranous ridges on the two lateral edges. Grooves of this kind are to be seen particularly well on the petioles of the radical leaves of the Rhubarb (see fig. 13 - ), Beet-root, Funkias, and most Violets. Far more complicated in structure than the radical leaves just described, are cauline leaves. Leaves proceeding from the stem high above the ground, and forming receptacles for rain-water, like those of the Rhubarb, are best fitted to preserve their proper direction when they have no stalks and the base fits directly on to the stem or passes into it. Cup-shaped laminae, if borne on long erect petioles, necessitate a great expenditure on supporting-cells, and they are, tliere- fore, on the whole, rare. Of the plants we know, only certain Stork 's-bills, Felargonium zonale, P. heterogamum, &c., afibrd examples of cup-shaped, cauline leaves of the kind, borne on long, rigid petioles. In most cases, therefore, cauline leaves which conduct water centripetally are either sessile or very shortly petiolate, have their bases close to the stem, and even extend their edges down it more or less in the form of wings and ridges, or surround it in the form of collars, lobes, and auricles, as in the case of so-called amplexicaul leaves. When the leaves are in pairs opposite one another and the alternate pairs at right angles, an arrangement known as decussate, the surplus water is usually conveyed through two grooves, which run down the intervening piece of stem from one pair of leaves to the next. Each of these grooves begins in an indenta- tion between the margins of the bases of a pair of leaves, and terminates above the midrib of one of the leaves belonging to the next pair. Now, water trickling down such a groove falls precisely on that part of a lower leaf where the rain retained by the surface of that leaf is collected; and so the stream of water becomes more and more copious as it approaches the ground. These grooves may be seen in many species of ringent Lahiatoi, Scrophulariacece, Primidaceca, Gentianacece, Rubiacece, and Willow-herbs; the best-marked instances are found in the Knotty Fig-wort (Scrophularia nodosa), the Yellow-rattle (Rhinanthus), the meadow-gentians {Gentiana germxinica, Rhcetica, &c.), and the Centaury (Erythrcea). The grooves always posF,ess the property of being wetted by water, whereas the ungrooved parts of the same stem are not wetted. Sometimes the grooves are fringed with hairs which absorb the water like the threads of a yb RELATIONS OF FOLIAGE-LEAVES TO ABSORBENT ROOTS. wick. By means of both contrivances advantage is ensured in that the water only oozes quite gradually down the moistened grooves, or else is conducted by the hairy fringes to the base of the stem, and does not rebound at any spot in the form of drops. Irregularly bounding drops would be liable to fall on the ground at spots where no absorptive organs awaited them. In cases where foliage-leaves, adapted to a centripetal conduction of rain, are arranged upon a spiral line down the stem, instead of in pairs opposite one another, the water leaks away along the spiral from one leaf to the next, and finally to the bottom. Then, again, there are often grooves in the stem along which the water trickles, as, for instance, in the Common Whortleberry ( Vaccinium Myrtillus). The erect leaves of this plant conduct the drops as they fall to the branches, which are deeply furrowed. The water travels through the furrows into those of lower branches, and finally along those of the main stem of the whole bush down to the earth. In Veratrum album each of the concave cauline leaves has, on the upper surface, a number of deep longitudinal grooves, which all discharge together at the base of the leaf. The water collected there at length overflows and runs down the round stem in no particular channel. The descent of rain-water along a spiral line may be very clearly traced in many plants of the Thistle tribe. If tiny shot-grains are substituted for rain- drops in a stifi'-leaved plant, the course designed for the drops in that particular species may be followed with ease. When strewn on a mature plant of the Safllower {Carthamus tinctorius) or of Alfredia cernua (fig. 14^), the grains of shot roll down the somewhat channelled surface of the highest cauline leaf, which stands up obliquely, and dash against the stem. The latter is half encom- passed by the leaf-base, and the shot then roll over one of the basal lobes of the leaf and travel out of the range of that leaf, falling on to the middle of the one next below. For the amplexicaul foliar bases are so placed that each leaf has one of its basal lobes above a concave part of the next lower leaf. In precisely the same way the shot descend from the second leaf to the third, and so on until they reach the earth quite close to the stem. The descent reminds one of the game in which a little ball is made to roll along a spiral groove on to a board furnished with numbered holes. Rain-drops falling upon thistle-like plants of this kind naturally follow the same course as the shot. Only, the additional fact must be taken into account that not only the highest but all the leaves are adapted as receptacles for the rain as it falls, and that consequently the drops falling from leaf to leaf are augmented by new tributaries, and become greater and greater as they descend. A somewhat different method of water-conduction from that which occurs in the Safflower and in the nodding Alfredia is observed in the Milk Thistle (Silyhum Marianum), in the Cotton Thistle (Onopordon), and in the Mullein (Verbascum phlomoides). The upper leaves, which have two semi-amplexicaul lobes, are as nearly erect as those of the Safflower and the nodding Alfredia, and lead the rain off in exactly the same way. But the leaves in the middle RELATIONS OF FOLIAGE-LEAVES TO ABSORBENT ROOTS. 97 of the stem are only erect for about three-quarters of their length; the upper- most third, including the apex, is bent obliquely outwards and downwards. Drops of rain falling on this upper third of a leaf would flow in a centrifugal direction, and do, as a matter of fact, drip down from the apex. Now the leaves in all fj ^' Fig. 14.— Irrigation of Kain-water. In Alfredia cermia. 2 in a Mullein ( Verbascum phlomoides). these plants are shorter the higher their position upon the stem, so that the total contour of the plant may be described as a slender pyramid. In consequence of this, water dropping from the outward-bent and drooping apices of superior leaves IS arrested by that part of an inferior leaf which shelves towards the stem, and is thereby conducted centripetally. Thus all the rain-water received by a plant of this kind at last reaches the immediate neighbourhood of the tap-root, and is 98 RELATIONS OF FOLIAGE-LEAVES TO ABSORBENT ROOTS. a source of nutriment to the absorption - roots which proceed from it. In the Milk Thistle (Silybum Marianum) the margins of the cauline leaves are very much waved, and, in consequence of this undulation, three or four depressions exist on each side, through which part of the rain, when there is a heavy downpour, flows off sideways. But even this water, falling laterally, drops upon parts of lower leaves, which conduct centripetally, and so coalesces with the streamlets otherwise produced. It is very rare for plants which convey water centripetally to have their leaves arranged in two rows. The most striking example of this class is the Japanese Tricyrtis pilosa. Its leaves are situated on the fully-developed stem very regularly, one above the other, in two series. Each leaf has two lobes embracing the stem, but the base is fixed somewhat obliquely, so that one of the lobes is fixed higher than the other. Moreover, the higher lobe is closely adpressed to the stem, whilst the lower forms a channel which discharges exactly above the concave surface of the next lower leaf belonging to the other side. When rain falls on this plant, the water, collected by one leaf, flows through the broad exit-channel on to the leaf below on the other side. Thence a somewhat augmented stream falls upon a leaf of the first series, and so on, a peculiar cascade resulting, which falls in a zigzag, from leaf to leaf, until it reaches the bottom, close to the stem. It would, however, be wrong to suppose that the above explanation sets forth the only significance to be assigned to the various arrangements described. To many plants it is a matter of indifference in what direction rain-water falls from the leaves. Such, for instance, is the case with all marsh - plants with roots buried in mud under water, inasmuch as the rain, as it drops, only goes into the water in the pond or marsh, and could not be conveyed to a definite spot for the sake of the absorbent roots. In the Water-plantain, the Flowering-rush, and the Arrow-head {Alisma, Butomus, Sagittaria), accordingly, no relationship between the form and direction of the leaves and the position of the absorbent roots is to be discovered. On the other hand, in arundinaceous plants (Arundo, Phragmites, Phalaris) an arrangement has been hit upon which is obviously designed to prevent rain- water from collecting between the haulm and the leaf. As is the general rule with grasses, so also in the above-named kinds of reeds, the stem or haulm is furnished with nodes, and from each node proceeds a leaf the lower part of which encases the haulm in the form of a tube or sheath, whilst the upper part is expanded and presents a flat, strap-shaped or concave surface, standing well away from the stem. The leaves may be folded round the haulm like banners. At the place where the sheath passes into the part of the leaf which stands away from the axis at an obtuse angle, one observes on the edge of the leaf close to the angle, two distinct depressions which represent conduits and convey part of the rain from the lamina. There is also a very neat contrivance here in the form of an erect dry membrane which acts as a dam, the so-called " ligule." This membrane, inserted upon the leaf -sheath, is, like the sheath, in contact with the haulm. When rain- i SAPROPHYTES AND THEIR RELATION TO DECAYING BODIES. 99 water flows down to this place it is stemmed by the membrane, as by a dam, and diverted right and left into the two grooves. In this way water is prevented from accumulating between the leaf -sheath and haulm, where it might do damage. In many reeds the contrivances for irrigation are even more complete than this. Sometimes hairs depend from the margin of the membrane in the direction of the grooves and, like a wick, lead the water in the proper direction. An opportunity will occur later on of showing how the conduction of rain to particular spots has an important bearing on the phenomenon of absorption by aerial parts of plants: and also in the regulation of transpiration; and how, by means of the apparatus for water-irrigation, not only absorptive cells at the extremities of roots in the earth, but special organs on the foliage-leaves as well, are often supplied with water. 3. ABSORPTION OF ORGANIC MATTER FROM DECAYING PLANTS AND ANIMALS. Saprophytes and their relation to decaying bodies. — Saprophytes in water, on the bark of trees, and on rocks. — Saprophytes in the humus of woods, meadows, and moors. — Special relations between Saprophytes and the nutrient substratum. — Plants with traps or pitfalls for animals. — Insecti- vorous plants which perform movements for the capture of prey. — Insectivorous plants with adhesive apparatus. SAPROPHYTES AND THEIR RELATION TO DECAYING BODIES. Whenever plants which take up organic compounds formed in the process of decay are the subject of discussion, the first examples that occur to everyone are members of the great family of Fungi, specimens of which make their appearance wherever dead animals or plants are undergoing decomposition. We recall the moulds, Plasmodia, pufF-balls, and mushrooms, which grow from dead organic bodies, and are associated with the unpleasant mouldy and cadaverous smell always perceptible in their neighbourhood. Many of these organisms do, in fact, belong to the class of Saprophytes. Indeed, one group of them is itself the cause of the chemical decomposition of dead plants and animals called decay. Their elongated thin-walled cells, the so-called " hyphte", thread themselves through dead bodies, and unite to form strands, bundles, net- works, and membranes, the whole constituting a structure to which the term "mycelium" is applied. These mycelia are often to be seen, with the naked eye, covering large areas. For instance, in damp cellars, mines, and railway-tunnels, any old rotten wood-work is clothed with delicate, whitish reticula and membranes. The heaps of grape-skins, stalks, and other refuse piled up in the open air by the side of vineyards after a vintage, are usually so completely overgrown by mycelia 100 SAPROPHYTES AND THEIR RELATION TO DECAYING BODIES. that their colour is quite altered. The so-called "mushroom-spawn", used in the cultivation of mushrooms, is also nothing but a mycelium, which entirely invests the manure employed in the cultivation of that fungus, and gives it a white mottled appearance. In addition to Fungi, however, a number of Mosses, Liverworts, Ferns, Lycopods, and Phanerogams take up organic compounds from the products of decay to serve as their food. In deciding whether a plant takes up only the mineral substances rendered soluble by the decomposition of the soil, or only organic substances disengaged by the decay of dead plants and animals, we depend generally on the condition and appearance of the nutrient substratum, and, in particular, on its composition, i.e. whether it is exclusively or predominantly organic. But such observations give a very uncertain indication. For, on the one hand, it is possible for plants rooted in a substratum of decaying matter to take nothing but mineral salts (i.e. inorganic compounds) from it; and, on the other hand, it frequently happens that sand or clay, apparently uncontaminated with organic matter, is saturated by water which oozes from a layer of humus in the vicinity, and brings with it organic compounds in solution. The following facts are instructive with reference to the former of these two phenomena. Maize, barley, and other cereals may be reared in fluids, so prepared as to contain a small quantity of mineral food-salts dissolved in distilled water (12 mg. potassium phosphate, 12 mg. sodium phosphate, 27 mg. calcium chloride, 40 mg. potassium chloride, 20 mg. magnesium sulphate, 10 mg. ammonium sulphate, and a few drops of iron chloride in a litre of distilled water), all organic compounds being carefully excluded. When the plants germinate, they develop roots which descend in the liquid and absorb from it mineral salts according to their requirements. They produce stems, leaves, flowers, and, ulti- mately, seeds capable of germination. Other plants of maize or barley reared simultaneously in richly-manured ground develop likewise leaves, flowers, and fruit. Moreover, analysis of the ash in both cases reveals the fact that the plants which took their nutriment from the manure contain the same salts as those reared in the made-up solution of salts free from organic compounds. Hence, the conclusion may be drawn that a plant of this kind is capable of obtaining an adequate supply of food-salts equally well, either from earth free from humus and manure, or from humus or manure themselves. The experiment further shows that, in the latter case, organic compounds need not necessarily be absorbed, in addition to the mineral constituents of humus or manure which are disengaged during decomposition. We must next refer to a fact in connection with the second point above men- tioned, viz. that plants rooted in sand or loam devoid of humus may yet have organic compounds brought to them by water filtering through a stratum of humus near at hand. The fact in question is, that the very water which one would least expect should contain organic compounds, that, for instance, of cold mountain streams, does very generally include traces of such compounds. On looking through analyses of mineral springs, one finds for the most part, amongst their constituents, SAPROPHYTES AND THEIR RELATION TO DECAYING BODIES. 101 combustible bodies arising from the dissolution of organic matter. Even the acid formerly designated by Berzelius by the name of "spring-acid", is doubtless a pro- duct of the decay of fragments of plants in the place where the water of the spring collects. So also is humic acid, a compound produced by decay. The nature of this acid is not yet, it is true, thoroughly known, and it may be a mixture of several acids. We know, however, that it is easily soluble in water, and that it forms soluble compounds with alkalies. Brooks running through woods or meadows, small mountain lakes adjoining peat-beds, and pools in actual peat, consist of water, brown in colour, which gives an acid reaction, and contains invariably organic substances in solution. The following observations are of great interest in connection with this subject. In the salt-mine at Hallstatt (Upper Austria) one of the galleries, which is hewn through rock and contains no wood-work of aiiy kind, exhibited (spread out upon its smooth limestone roof) the mycelium of a fungus (an Omphalia), which certainly required organic nutriment. There were no decaying animal or vegetable remains anywhere in the gallery, and the mycelium derived nourishment solely from water oozing from above through a few narrow cracks in the stone whereby the surface of the latter was kept moist. This water came from a meadow lying high above the mine. Between the two was a thick stratum of limestone with a deep layer of earth resting upon it. The water was clear and colourless, and contained a certain amount of lime, but no perceptible trace of organic substances. Yet this water must have brought organic matter from the meadow above into the mine, and the minute quantity so introduced sufficed to enable the fungus mycelium to grow luxuriantly. In the Volderthal, near Hall, in Tyrol, there is a spring of cold clear water rising out of slate at a height of 1000 metres above the sea-level, which is filled at its source with a dark thick felt. The felt may be lifted out in pieces the size of one's hand, and it is the mycelium of a fungus, probably a Peziza. It clings to slabs of slate, between which the water trickles abundantly, and its nutriment can only be derived from this water. There are pine- woods and meadoM^s in the neighbourhood, but no greater amount of vegetation, humus, or rotten timber than is found near other springs. These instances satisfactorily prove that even the clearest mountain springs contain organic substances in quantities sufficient, however minute, to nourish fungi. When the origin of springs is taken into account, this result is not really surprising. They are fed by deposition from the atmosphere. The water thus deposited percolates into the ground, passing, in the first place, through a layer of earth-mould which is covered by vegetation, and contains more or less humus in its upper strata. A small quantity of the products of decay is inevitably absorbed, and even if they are partially withdrawn again in lower strata of the earth, traces are still retained by the water in its descent to greater depths, and re-ascent to the surface in the form of springs. The characteristics of the great veins of water which ascend in this way are no doubt common to the smaller veins which originate 102 SAPROPHYTES AND THEIR RELATION TO DECAYING BODIES. in the vegetable mould saturated by snow and rain on the ground of forests or in the humus covering meadows, and which percolate through into the sand or loam beneath. Plants whose roots ramify in this deeper layer of earth derive thence the organic compounds conveyed by the water, and have the additional advantage of being able to satisfy at the same time their requirements as regards mineral sub- stances. This circumstance is of importance not only to flowering-plants but also to many fungi, as, for instance, to all species of Phallus, they having need of a great deal of lime. An explanation is thus afforded of the fact, formerly difficult to understand, that in forests and meadows not only the upper black or brown humus layer, but also the underlying yellow loam, or pale sand, neither of which latter contains any humus, has mycelia of fungi running through it in every direction, and weaving their threads over little fragments of rock. Indeed, it sometimes happens that the lower layer of earth is more abundantly penetrated with plexuses of hyphse than is the upper layer, consisting of vegetable mould. The greatest number of saprophytes is to be found therefore at places where the humus layer is not too thick and loam or sand occurs at no great depth; but where decaying vegetable remains are piled metres high, as on moors, for example, instead of fungi being produced in extraordinary abundance, as one might expect, only a few occur. Pure peat is by no means a favourable soil for fungi, a circumstance which may be partly due to the antiseptic action of certain compounds developed in it. It follows from the foregoing observations that a sure conclusion as to the nature of plants rooted in a particular substratum cannot possibly be derived from the mere appearance of the substratum. Moreover, the conditions necessary for the growth of plants requiring organic products of decay as nutriment appear to be of much wider occurrence than one would suppose upon a cursory observation of the conditions existing in fields and forests, or, if one considers exclusively instances of cultivated plants reared on arable land, which is manured and constantly turned over. The great variety of plants produced on a limited area is also now intelligible. From the same soil some absorb organic compounds, others mineral substances only; whilst others again take some organic and some mineral food- salts. The determining factor is not the amount of a given substance present in the substratum, but rather the special needs of each species, and ultimately the specific constitution of the protoplasm in each one of the plants which thus, side by side, nourish themselves in totally different ways. If, then, neither the appearance of the ground nor its richness in respect of humus affords any certain indication as to whether a particular plant lives on organic products of decay or not, the question may perhaps be solved by the fact of the plant's containing or not containing green chlorophyll-corpuscles. We may take it as proved by many results of investigation, that the decomposition of the carbon-dioxide absorbed by a plant from the air, and the formation of the organic compounds of carbon, hydrogen, and oxygen known as carbohydrates (which play so important a part in vegetable economy), only take place in organs possessing the green pigment known as chlorophyll. We shall return to a discussion of these SAPROPHYTES AND THEIR RELATION TO DECAYING BODIES. 103 processes in detail later on, but the fact must be taken into consideration here. One would suppose, accordingly, that plants able to obtain ready-made organic compounds from a nutrient substratum could spare themselves the trouble of building them up, so that the presence of chlorophyll would be superfluous. This conjecture is in fact supported by the absence of chlorophyll in fungi, which are typical instances of saprophytes. But, on the other hand, some plants appear to negative this assumption, or at any rate to deprive it of general application. In mountain districts, where cattle continually pass to and from the meadows and alps, one notices on their halting grounds, and along their tracks, moss of a con- spicuous green colour growing on circumscribed spots. On closer examination we find that we have here an example of the remarkable group of the Splachnacece, and that it has selected the cow-dung to be its nutrient substratum. Each growth of emerald green, Splachnum ampullaceum, is strictly limited to the area of a lump of dung; no trace of it is to be seen elsewhere. All the stages of development of this moss follow one another upon the same substratum. First of all the lumps of dirt which are kept moist by rain or by standing water, become enveloped in a web of protonemae, and their surfaces acquire thereby a characteristic greenish lustre. Later, hundreds of little green stems, thickly clothed with leaves, emerge, and the spore-cases, which resemble tiny antique jars, and are amongst the prettiest exhibited by the world of mosses, become visible as well. Just as Splachnum ampuUaceum is produced on the dung of cattle, so is Tetraplodon angustatus on that of carnivorous animals, and there can be no doubt that these, and in general all Splachnacece, are true saprophytes, A similar remark holds with regard to the green Euglence which escape from Hormidium-cells, and fill the foul- smelling liquor in dung-pits and puddles near cattle-stalls in mountain villages, and which multiply to such an extent that in a few days the liquid changes colour from brown to green. Thus plants do exist containing chlorophyll although absorbing from the substratum organic compounds alone, and containing it, indeed, in such quantities that its presence cannot be looked upon as accidental. It follows, firstly, that absence of chlorophyll is not the distinguishing mark of saprophytic plants; and, secondly, that the organic nutriment of the plants above mentioned cannot be used forthwith unaltered in the building up and extension of their structures, but, like inorganic material, must undergo various changes, that is, must be to a certain extent digested before being used for construction. The probability is that green saprophytes take carbon from their substratum in a form unfitted for the manu- facture of cellulose and other carbohydrates. Saprophytes that are not green must obtain carbon from the substratum in the form of a compound, the direct absorption of which could be dispensed with if chlorophyll were present; but it does not necessarily follow that all the organic compounds absorbed by non-green saprophytes are capable of immediate service as materials for construction without any preliminary alteration. Impartial consideration of the above facts forces us to conclude that there is no 104 SAPROPHYTES IN WATER, ON THE BARK OF TREES, AND OX ROCKS. well-marked boundary line between plants which absorb organic compounds and those which absorb inorganic compounds from their respective substrata; and that there undoubtedly exist plants capable of taking up both kinds of material at the same time. This conviction is strengthened still further by the circumstance, which has been repeatedly confirmed by experiment, that plants susceptible of being successfully reared in artificial solutions of mineral salts — to the exclusion of organic compounds — do not entirely reject organic compounds when the latter are tendered to them, but unquestionably assimilate some of them (urea, uric acid, glycocoU, &c.) and work them up into constituents of their own frames. But, in spite of the impossibility of drawing a sharp line of demarcation between the two groups, it is convenient to treat of the absorption of organic compounds separately, because this division of the subject affords the best opportunity of inspecting in detail, and of surveying generally, the conditions of food-absorption, the comprehension of which is otherwise difficult. In oi'der to determine in each individual case whether a given plant lives either exclusively or principally upon organic food, derived from decaying animal or vegetable remains, reliance must be placed on expei'iments with cultures; and, in the absence of better vantage-ground, the results of the rougher experiments made by gardeners should not be neglected, always providing that they are accepted subject to possible correction by subsequent exact experiment. SAPKOPHYTES IN WATER, ON THE BARK OF TREES, AND ON ROCKS. Of the special cases of absorption of organic compounds from decaying bodies, we have first of all to consider those occurring amongst water-plants. In the sea, wherever thei-e is an abundance of animal and vegetable life there is also plenty of refuse, for there death and decay hold a rich harvest. The quantity of organic matter dissolved in the water is naturally greater in these places than where vegetation and animal life are less conspicuous. There is a much more varied flora and fauna to be met with in the sea near its coasts, especially in shallow inlets, than at a greater distance from the shore ; and the number of dead organisms is also greater near the coast. A mass of organic remains is throw^n up by the tide, and by waves in stormy weather. This mass rots during the ebb. Part of it is dragged out to sea again by the next high tide, and then flung up once more; so that the beach is always strewn with dead remains, and the sea near the shore contains more products of decomposition than in the open. In the immediate neighbourhood of seaports, moreover, or wherever people live, the volume of refuse is considerably increased, and the water in harbours and stagnant inlets behind breakwaters, and at the mouths of canals and sewers, contains such a large quantity of organic refuse in a state of decomposition that its presence is revealed by the odour emitted. Now it is just at these places that an abundant vegetation of hydrophytes is developed. Not only the bottom of shallows, but stones, stakes, quays, buoys, and even the keels and planks of boats long anchored SAPROPHYTES IN WATER, ON THE BARK OF TREES, AND ON ROCKS. 105 in harbour, are overgrown by Ulvoe, wracks, filamentous algae, and Floridece. Not a few, as, for instance, the so-called sea-lettuce (Ulva lactuca), several species of Gelidiwm, Bangia, and Ceramiwm, and the great Cystosira harhata, thrive best and in greatest abundance in polluted water of the kind; and there can be no doubt that this is to be accounted for by the presence of a greater quantity of organic compounds in that water. It is not only in contaminated sea- water, but also in other collections of water which contain products of putrefaction in solution, that we find a characteristic vegetation. We have already alluded to the presence of Euglenoe in the liquor of manure-pits. They occur also at the foot of shady walls, in dirty back streets in towns, in the puddles, and on ground which is saturated with urine and impurities of every kind. These places are the home of a number of other minute plants, which stain the polluted ground after rain with the gayest colours. There, side by side with black patches of Oscillaria antliaria and verdigris-coloured films of Oscillaria tenuis, are blood-red patches of Palmella cruenta, and brick-red patches of Chroococcus cinnamomeus. Equally characteristic is the vegetation which covers the earth at the mouths of drains, and is bathed by the trickling sewage. Large areas here are overgrown by the green JformidiuTn murale, which weaves itself over the mire, and by the dark, actively-oscillating Oscillaria liraosa; and, above all, the curious Beggiatoa versatilis makes itself conspicuous, sending out from a whitish gelatinous ground mass long oscillating filaments, which emerge after sundown, and next day split up into innumerable little bacteria-rods. The red-snow alga, too (represented in fig. 25a), lives at the expense of the pollen- grains, bodies of insects, and other decaying matter blown on to snow-fields; whilst the nearly allied blood-red alga (Hcematococcus pluvialis or Sphccrella pluvialis) lives in the water in hollow stones where all sorts of animal and vegetable remains collect. Leaves blown into deep pools, and lying rotting at the bottom, are everywhere overgrown by green CEdogonium, by Pleurococcua angulosus, and by the amethyst-coloured Protococcus roseo-persicinus. The bottoms of ditches on peat- bogs, which are full of brownish water containing an abundance of compounds of humic acid in solution, are covered with this amethyst Protococcus, whilst a profusion of small filamentous algae, Oscillariae and so fortli {Bulbochccte parvula, Schizochlamys gelatinosa, Sphoerozosma vertebrata, Micro- cystis ichthyloba, &c.), as well as a group of dusky mosses (Hypnum giganteum, H. sarmentosum, H. cordifolium), all have their home exclusively in still water richly supplied with organic compounds. When we include also the curious mould-like Saprolegnice produced on dead bodies floating in water — Saprolegnia ferax and Achlya prolifera on flies and fishes — some idea is obtained of the great variety of saprophytes living in fresh water, as well as of those inhabiting the sea. A much more agreeable and attractive picture than that of these aquatic sapro- phytes is afforded by plants whose sole habitat is the bark of trees. The dead [ bark does not constitute the nutrient base of all the plants which grow from trunks and branches, or climb up them in the form of clinging and twining lianas. 106 SAPROPHYTES IN WxlTER, OX THE BARK OF TREES, AND ON ROCKS. Often the trees only serve as supports, by means of wliich the plants in question raise themselves out of darkness into light. Such food -salts as they require they take, not from their support, but from the earth, into which they send absorptive roots. As years go by, a quantity of inorganic dust collects in the forks of branches and in the little rents and fissures in the bark of old trees, and this dust gets mixed with crumbled particles of bark. The clefts, therefore, are more or less full of vegetable mould, and this forms an excellent foster-soil for a large number of plants. But it is not necessarily the case that all plants rooting in this mould take up organic compounds from it. Thus, one finds not infrequently in the angles of bifurcation of the trunks of old limes and other trees, little gooseberry and elder bushes, and bitter-sweet plants, which have germinated there from fruits brought by black-birds, thrushes, and other frugivora. These shrubs, in the forks of limes and poplars hardly take any organic compounds from the mould in which they are rooted, but confine themselves to the absorption of such mineral salts as they may require. But, with the exception of instances of that kind, the great majority of plants, nestling in the mould in crevices of bark, do take nutriment from this their substratum in the form of organic compounds. In cold regions the plants living in the mould of bark are for the most part mosses and liverworts. They cover trunks and branches of old ashes, poplars, and oaks, with a thick green mantle, and grow especially on the weather-side of the trees. In the tropics, on the other hand, the fissured bark of trees is a rallying ground not only for delicate mosses and moss-like Lycopodia, but also for a whole host of ferns and vivid flowering plants. The number of small ferns which develop and unroll their fronds from chinks in the bark of trees is so great that old trunks appear wrapped in a regular foliage of fem-fronds. Of Phanerogams, in particular, the Aroidece, Orchidacece, Bromeliacece, Dorstenice Begoniacece, and even Cactacece (species of the genera Cereus and Rhipsalis) bury their roots in the mould of bark. It is to be remarked that the rosettes of Bromeliacece ornament chiefly the forks of trunks, whilst Dorstenice, Orchidece, and the various species of Rhipsalis grow on the upper side of branches that ramify horizontally; whilst, lastly, Aroidece and Begonice take root, for the most part, on the surfaces of huge erect trunks. Besides the mould collected in crevices and fissures of bark, the bark itself, that is, the cortical layer, dead but not yet crumbled and mouldered into dust, forms a nutrient substratum for a whole series of plants of most various affinity. Many fungi and lichens penetrate deeply the compact bark, and their hyphal filaments ramify between its dead cells. Other plants, instead of piercing through the substance of the bark, lay themselves flat upon its surface, and grow to it so firmly that if one tries to lift them away from the substratum, either part of the latter breaks oflf, or the adnata cell-strata are rent, but there is no separation of the one from the other. If a tuft of moss (e.g. Orthotrichum fallax, 0. tenellum, or 0. 'pollens), growing on bark, or a liverwort {e.g. Fridlania dilatata) closely adherent to a similar basis, is forcibly removed, little fragments of the bark may be SAPROPHYTES IN WATER, ON THE BARK OF TREES, AND ON ROCKS. 107 always seen torn off with the rhizoids at the places where they issue from the stemlets. The same thing occurs in the case of the roots of tropical orchids growing to the tree-trunks which constitute their habitat. The majority of these tree-orchids nestle, no doubt, in mould-filled crevices of the bark, and' nourish them- Fig. 15.-Aerial Koots of a Tropical Orchid {Sarcanthus rostratus) assuniin^' the form of stiaps.. ielves, besides, by means of special aerial roots which hang down in white ropes ^nd threads, like a mane, from the places where the plants are situated upon the rees, and which will presently be described in detail. But a small section develops trap-shaped roots as well, which adhere firmly to the bark with their flat surfaces, hi*? phenomenon is most strikingly exhibited by the splendid PhalcBnopsi^ 108 SAPROPHYTES IN WATER, OX THE BARK OF TREES, AND ON ROCKS. Schilleriana, a native of the Philippine Islands; its roots are rigid, compressed, and about 1 cm. in breadth; the surface turned away from the trunk is slightly- convex, and has a granular structure and metallic glitter like a lizard's or chame- leon's tail. The surface towards the trunk is flat and without metallic glitter, and upon it, close behind the growing point, there is a whitish fur consisting of short, thickly packed, absorptive cells. When the tip of one of these roots comes into contact with the bark it grows so firmly to the substratum by means of the absorption-cells, that it is easier to detach superficial bits of the bark itself than the root. The latter, once fixed, flattens out still more and becomes strap-shaped, whilst creeping outgrowths proceed from it, forming strips which may ultimately attain a length of li metres. The sight of a trunk covered with these long metallic bands is one that never fails to excite wonder even in the midst of the world of orchids, wherein, as is well known, there is much to marvel at. In other species of tropical orchids, e.g. in Sarcantlais rostratus (fig. 15), the roots are not flat from the beginning, but become so when they come into con- tact with the bark. A root is often to be seen which arises as a cylindrical cord from the axis, then lays itself upon the bark in the form of a band, and further on lifts itself once more, resuming at the same time the rope form, as is shown in the illustration. Here also complete coalescence takes place between the bands and the bark, and the union is extremely close. Similar conditions have been observed to hold in many Aroidece living on the bark of trees. The plants in question lie with their stems, leaves, and roots flat against the trunks, so that they suggest a covering of drapery. Taking, for instance, the Marcgravice {Marcgravia paradoxa, M. umhellata), one might at flrst sight suppose that they adhere to the bark not only by the roots, but also by the large discoid leaves, which are arranged in two rows. A very remarkable fact also, in connection with these plants, is that they only grow on very smooth and firm bark. When transferred to a soft substratum, such as mould or moss, they languish, because their roots are unable to enter into close union with a support of such loose texture. This is also true of most tropical orchids living on bark. When their seeds are transferred to loose earth devoid of humus, they do indeed germinate, but then perish, whereas when sown on the bark of a tree, they not only germinate, but grow up with ease into hardy plants. Where steep rocks occur near clumps of trees it is not uncommon for the same species of plants to grow on both. Allusion is not here made to kinds which, like ivy, have their roots in the earth at the foot of rocks and trees, and creep up the one or the other indiflferently, using both merely for support and not as sources of nutriment, and clinging to them by means of special attachment-roots. The remark is applicable also to plants which live on the products of the decay of organic bodies, for example many tropical Orchidece, Dorstenice, Begonice, and Ferns; and in cooler parts a number of Mosses and Liverworts. It is not difficult to explain this phenomenon in the case of species which derive their food from vegetable mould. The crannied wall of rock is, in a certain way, analogous to the rugged bark of a tree. The holes in the rock are filled in course of time with black vegetable mould, SAPROPHYTES IN THE HUMUS OF WOODS, MEADOWS, AND MOORS. 109 md plants with foliage, flowers, and fruit of a form adaptable to cracks and holes ire able to establish themselves in the mould there, just as well as in that collected n crevices of bark. In one respect, indeed, they are even more favourably situated. For the humus in bark gets quite dry in long periods of drought, because no water s yielded to the bark by the wood of a tree, even though the latter be abundantly supplied with sap; whereas, in the case of rocks the probability is, the clefts being lery deep, that even when the top layers of humus filling them yield up their water ;o the air, a certain restitution of moisture takes place from the deeper parts, which ire never quite dry. Moreover, plants growing in the mould of rock crevices are ible to send their roots down to much deeper strata than is possible in the case of Dark. This is another reason why deep cracks in rocks, filled with humus, exhibit I richer flora, as a rule, than do the much shallower crevices in the bark of trees, ilthough, as has been said before, the two habitats have many plants ia common. It is more difficult to explain how it happens that plants which derive their sustenance, not from the mould in crevices, but from the substance of the bark tself, and which lie flat against its surface, are also found adhering to walls of •ock. As an example take Frullania tamarisci, a Liverwort with small brown jifurcating stems, which bear double rows of leaves and are of dendritic appearance, rhis plant grows equally well on the bark of pines or on the face of adjacent gneiss •ocks. At first sight it would seem scarcely possible that a plant of this kind, slinging to the unfissured surface of rock, should be in a position to obtain organic jompounds from its substratum. This is nevertheless the case. Closer inspection 'eveals the fact that the Liverwort does not adhere to blank rock, but to a part "ormerly clothed by rock-lichens. This inconspicuous incrustation of dead lichens s a complete substitute for the superficial layer of bark, and it is into it that the Frullania tamarisci sinks its roots. Another way by which food is supplied to Dlants adherent, like the above, to vertical and unfissured rocks will be discussed ater on. SAPROPHYTES IN THE HUMUS OF WOODS, MEADOWS, AND MOORS. Damp shady woods, especially pine woods, are particularly well furnished with saprophytes. Here again we find representatives of the same families as choose the Dark of trees for their habitat. On the ground of woods, the most characteristic "orms are mosses, fungi, lycopods, ferns, aroids, and orchids. The dark -brown lumus, produced from dropped and decaying needles, is first of all covered by a •ich carpet of mosses, such as the widely distributed Hylocomium splendens, Eypnum triquetrum, and Hypnum Crista -castrensis. The mouldered dust of lead trees has a clothing of Tetraphis pellucida and of Webera nutans, and lecaying trunks are overgrown by the cushions of species of Dicranum (Dicranum >copa7'ium, D. congestum, Dicranodontium longirostre), pale feathery mosses Hypnum uncinatum and H. reptile) and various liverworts. Everywhere above ihe soft, ever-moist carpet of moss rise green fronds belonging to broad-leaved ferns. 110 SAPROPHYTES IX THE HUMUS OF WOODS, MEADOWS, AND MOORS Woods are also the special abode of fungi, and the damp ground is covered towards autumn by innumerable quantities of their curious fructifications. Dropped needles and cones, leaves and sticks strewn upon the ground, fallen trunks, and even the dark amorphous dust arising from the mouldering of these bodies and of the numerous roots ramifying in the ground, appear to be perforated by and wrapped in the protoplasmic threads of plasmoid fungi, or similarly invested by a plexus of filaments, the so-called mycelia of other forms of fungi. Amongst the scaly fragments of bark, peeling from the trees, they appear in the form of slimy strings, or as a dark trellis and net-work, inserted between the bark and wood of the rotting tree; on the stripped white trunk they are in dark zigzag lines like those of forked lightning; and between, the white mycelia of huge toadstools and tremellas are woven in all directions. Here and there large areas of the brown decaying soil are flecked and speckled by these mycelia, and even the dead stems of the mosses on the ground are festooned with white fleece, and wrapped round by hyphse. It is worth while to glance too at the reciprocal relations of these w^oodland plants. We find mosses, lycopods, and various ferns and phanerogams living upon the fallen twigs and needles, and on the mouldering roots of pines and fir- trees. The dead remains of those plants aflbrd sustenance to the fungi, which lift their fructification above the bed of moss. In their turn the rotting fructifications of the larger fungi form a nutrient substratum for smaller fungi, which cover the decajang caps and stalks wdth a dark-green velvet. Lastly, these little fungi, too, fall a prey to corrupting bacteria, and are resolved into the same simple inorganic compounds as were absorbed from the air and earth, in the first instance, by the pines and fir-trees. In the depths of forests there is going on, for the most part unseen by us, a mysterious stir and strife, accompanied by an uninterrupted process of exchange between the living and the dead, and a marvellous transformation of those very substances whose secret we have only partially succeeded in solving. The results of cultivation have proved that in the group of flowering-plants belonging to the woodlands of Central and Northern Europe, which derive sus- tenance partially or entirely from the organic compounds afforded by the humus, are to be included, amongst others, the various species of coral-wort (Dentaria bulbifera, D. digitata, D. enneaphyllos), Circcea alpina, Galium rotundifolium, and Linnoea horealis, and above all a large number of orchids. Of these, Dentaria prefers mould produced from the beech leaves, and Circcea, Galium, and Linnoia | appertain to the mould of pine- woods. Of the orchids some are provided with green leaves, as, for instance, the delicate little Listera cordata, Goodyera re2')ens . remarkable for its villous petals, and the various species of Cephalanthera, Ein- j pactis, and Platanthera; others, such as Limodorum abortivum, the bird's-nest j orchis, the coral-root, and Epipogium aphyllum have none. Limodorum abortivum ■ belongs rather to the warmer districts of Central Europe. It has fleshy root- i fibx-es, twisted and twined into an inextricable ball, and a slender steel-blue stem, | over half a metre in height, bearing a lax spike of fairly large flowers, which \ SAPROPHYTES IN THE HUMUS OF WOODS, MEADOWS, AND MOORS. Ill subsequently become paler in colour. The bird's-nest orchis {Neottia Nidus-avis) is of wide distribution both in forests of pines and in those composed of angio- spermous trees. Its stem and flowers are of a light-brown colour, unusual in plants, but somewhat like that of oak-wood. The flowers have no scent, and the numerous roots, issuing from the subterranean part of the stem and imbedded in humus, remind one in form and colour of earth-worms, and together constitute a strange tangled mass as large as a fist. The latter has been thought to resemble a bird's nest, and to this is due the name of the plant. The coral-root (Corallorhiza innata), unlike the bird's-nest orchis, has no root at all; but, on the other hand, the sub- terranean portion of the stem, the so-called rhizome, possesses a distant resemblance to the root -tangle of Neottia. Pale -brownish branches of this rhizome, which bifurcate repeatedly at their obtuse and whitish extremities, looking as if they had been subjected to pressure for a time, and all the short lobe-shaped branchlets thereby spread out into one plane, lie closely crowded together, sometimes crossing one another, and so form a body which vividly recalls the appearance of a piece of coral. This underground coral-like stem-structure develops each year pale greenish shoots which rise above the ground and bear small flowers speckled with yellow, white, and violet, and exhaling a scent of vanilla; later, green fruits of a com- paratively large size develop, turning brown when they ripen. The fourth mentioned of these pale wood-orchids, the Epijwgium ajphyllum, is at once the rarest and most curious of them all. Like the coral-root it has no true roots. Its rhizome so closely resembles the latter's that it is easy to mistake the one for the other; but they may be distinguished by the fact that in the case of Epijpogium the rhizome sends out long filiform shoots, which swell up like tubers at their tips, and may be regarded as subterranean runners. The swollen extremity becomes the point of origin of a new coral-like structure, which develops at about the distance of a span from the old one; whilst the latter, usually exhausted after flowering, gradually perishes. This coral -like stem lives of course underground, and is not visible till one lifts away the moss from the mould on the ground. It is often completely imbedded in sandy loam, lying immediately beneath the black mould. Many years frequently go by without the Epipogium producing flowers. The plant meanwhile lives entirely underground. In the course of a summer in which it has not flowered, anyone not having previous exact knowledge of its where- abouts might pass by without dreaming that the bed of moss and humus on his path concealed this strange growth. The flowering stems which at length emerge, when there is a warm summer, are right above the place where they branch ofl" from the subterranean rhizome. They are thickened in a fusiform manner, and have, for the most part on one side, a reddish or purplish tinge. Everything connected with them is tense, smooth, full of sap, and almost opalescent. The few flowers that are borne by the stem are comparatively large, and emit a strong perfume resembling that of the Brazilian genus of orchids Stanhopea. The colour- ing, too, a dull yellowish white with touches of pale red and violet, reminds one of these tropical orchids. 112 SAPROPHYTES IN THE HUMUS OF WOODS, MEADOWS, AND MOORS. The sight of the pale-coloured plants lifting their heads, at flowering time, from the tumid carpet of moss has all the stranger effect because, as a rule, no other flowering plants are visible in any direction. The flowers are suspended by delicate drooping pedicels, and owing to their peculiar colour, fleshy consistence, and form — the erect concave petal like a Phrygian cap or helmet, and the others stretched out like prehensile limbs — remind one of the opalescent medusae which float on the blue sea waves. The propriety of the analogy is enhanced by the fact that the form and colour of other saprophytes produced near Epii^ogiwni in woods have a striking resemblance to the animals and wracks which inhabit the sea-bottom. The fungi, known by the name of club-tops, much-branched, flesh- coloured, yellow or white Clavarice, which often adorn whole tracts of ground in a w^ood, imitate the structure of corals; Hydneoi are like sea-urchins, and Geaster like a star-fish, whilst the various species of Tremella, Exidia, and Guepinia, which are flesh-pink, orange, or brownish in colour, and the white translucent Tremellodon gelatinosum, resemble gelatinous sponges. The small stiff" toad-stools (Marasniius), which raise their slender stalks on fallen pine-needles, remind one of the rigid Acetabularide. Other toad-stools, with flat or convex caps exhibiting concentric bands and stripes, such as the different species of Graterellus, have an appearance similar to the salt-water alga known by the name of Padina. Dark species of Geoglossum imitate the brown Fucoidece; and one may fancy the red warts of Lycogala Epidendron, a plasmoid fungus inhabiting the rotten wood of dead weather-beaten trees, to be red sea-anemones with their tentacles drawn in, clinging to gray rocks. However far-fetched this comparison between the two localities may seem at first sight, everyone who has had an opportunity of thoroughly observing the characteristic forms of vegetable and animal life in woods, and at the bottom of the sea, will inevitably be convinced of its accuracy. Meadow-land, rich in humus, is much more sparsely occupied by saprophj^tes than the soil of woods. There is no lack of the strange forms of toad-stools and pufF-balls, whose fructifications often spring up in thousands, especially in the autumn, in company with the meadow-saffron; but in numbers they are not to be compared with those which occur in the mould of woods. Amongst ferns and phanerogams, the following species are dependent upon the organic compounds arising from the decomposition of the humus: Moonwort {Botrychium Lunaria), numerous orchids, blue and violet-flowered gentians, the famous Arnica, Poly- galacese, and more especially several grasses, chiefly the Matweed (Nardus strida) which, when once it has struck root in the humus, extends in dense masses over large areas. Several plants, too, adorning alpine pastures, and belonging for the most part to the same families as the species mentioned above, are to be regarded as humus-plants. Such are the Alpine Club-moss (Lycopodium alpinum), the dark-flowered Nigritella nigra, and several other sub-alpine orchids; a number of small, sometimes tiny, gentians (Gentiana nivalis, G. pi'ostrata, G. glacialis, G. nana, Lomatogonium Carinthiacum), Valeriana celtica, the Scottish asphodel (Tojieldia borealis) of the north, a few grasses, sedges, and rushes (e.g. Agrostis RELATIONS OF SAPROPHYTES TO THEIR NUTRIENT SUBSTRATUM. 113 alpina, Carex Gurvula, Juncus trijidus), various anemones, campions, umbelliferous plants, violets and campanulas {e.g. Anemone alpina, Silene Pumilio, Meum Mutellina, Viola alpina, Campanula alpina) and several mosses {e.g. Dicranum elongatum and Polytrichum strictum) which clothe the humus on stretches of turf and in inclosures. Many of the plants also that are native on the black graphitic soil in hollows of high mountain ridges take up organic food from their substratum. These include Meesia alpina and various other mosses produced exclusively in places of the kind; and, above all, numerous Primulacese and Gentianeae {Primula glutinosa, Soldanella pusilla, Gentiana Bavarica). It seems, moreover, to be by no means a matter of indifference to these plants at what temperature, and in what state of the air, in respect of moisture, the decomposition of humus takes place. If species which grow abundantly in these localities are dug up and transferred, together with the black earth in which their roots are imbedded, into a garden, and are there cultivated in such a way that the external conditions are as nearly as possible those of the original habitat; or if young plants are reared from seed in the same black humus-filled earth, they thrive only for a short time, soon begin to fade, and within the space of a year are dead; whereas, alpine plants belonging to the same altitude above the sea, but rooted in loamy or sandy earth, flourish excellently in gardens as well. Various moor-plants {e.g. Lycopodium inundatilm, Eriophorum vagin- atum, Trientalis Europcea) only live a short time in a garden even though the clods of peat, in which their roots are imbedded, are transplanted with them. This fact can scarcely be explained except by supposing that the organic compounds, produced by the decay of vegetable remains on alpine heights and moors, are essentially different from those evolved by similar matter under the changed conditions of temperature and moisture occurring in a garden at a lower level. Gardeners say that the peat and black graphitic soil from the slopes of snowy mountains turn sour in gardens, and they may be to this extent right, that in all probability the humic acids produced under altered circumstances are different. SPECIAL EELATIONS OF SAPEOPHYTES TO THEIE NUTEIENT SUBSTEATUM. In the plants under discussion, the cells which absorb organic compounds are, taken all in all, very similar to those which absorb mineral food-salts. Where there is no cell-membrane, as in the case of Plasmodia and Euglense, the food diffuses through the so-called ectoplasm, or outer layer of the protoplasm, into the interior of the cell. Saprophytic marine and fresh-water algae are able to absorb the products of decay in the water around by means of their superficial layers of cells. The mycelia of fungi have the power of taking in nourishment with special rapidity. Each hypha, or more accurately, each long, delicate-walled cell of a mycelium is, to a certain extent, an absorptive cell; its entire surface is capable of exercising the function of suction and of withdrawing from the environment, along with water, the very substances which are needed. The coral-like underground stem of 114 RELATIONS OF SAPROPHYTES TO THEIR NUTRIENT SUBSTRATUM. Epijpogium aphylluvi, as well as that of the " Coral-root", which is entirely destitute of roots, develop fascicles of absorptive cells on their ramifications, and on special little swellings; and the white subterranean stem structures of Bartsia alpina are also provided with long absorptive cells. The white, fusiform, tuberously thickened, underground stems of the Alpine Enchanter's Nightshade {Gircoia aljnna) exhibit no roots during autumn and winter, nor until such time as new leafy stems sprout from them and lift themselves into the daylight; they only have scattered club- shaped absorptive cells. Yet it is inconceivable that the few absorptive cells meet the entire requirements of these plants at the season of the development of their aerial stems. Food is absorbed in these cases also by the epidermal cells of the entire tuber, underground stem, or coral-like rhizome, as the case may be. The epidermal cells of these subterranean caulomes which lie immediately in contact with the black mould or humus on the ground of forests, have such thin and tender walls that they are quite as well adapted to the absorption of nutriment as are the projecting absorptive cells; indeed the club-shaped absorptive cells on the small tubers of Enchanter's Nightshade exhibit somewhat thicker walls than those forming the general epidermis of these tubers. We may compare food-absorption as performed by these coral-like and tuberous structures, imbedded in decaying plant residues, with the action of tape-worms in process of sucking in through their entire epidermis the fluid filling the intestines they inhabit. The epidermal cells of the thick tortuous root- fibres of Neottia Nidus-avis are all capable of absorbing nutriment, though they do not project as tubes, but are tabular, and have their outer walls, which are in immediate contact with the nutrient soil, only slightly arched outwards (see fig. 16 ^). The green leafy orchids rooted in the vegetable mould of woods and meadows are, on the contrary, furnished with very long tubular absorption cells; and these cells do not wither and collapse forthwith when the root elongates, but long retain their vigour and activity. Whereas in the case of land plants adapted to mineral food-salts, the tubular absorption cells ("root-hairs") are limited to a narrow zone behind the growing point of the root and always die comparatively soon; in the case of orchids, having cylindrical roots imbedded in vegetable mould, these structures appear to be beset from end to end with long scattered tubular absorption cells, which are retained even through the drought of summer or the frost of winter right into the next period of vegetative activity; and these cells occur most abundantly in parts of the ground where there happens to be a bed of humus or mouldering remains particularly amenable to their purpose. Similar relations are found to exist in the case of the dichotomously-branched roots of the Club-moss. They are twisted in spirals and bore into the vegetable mould like corkscrews, and their absorption cells form in some places regular tassels, which are completely cemented over with fine black mould. The roots of grasses which, like the Mat-grass, live on the decomposition-products of vegetable mould, are also distinguished by strikingly long absorption cells, which grow in black or brown humus and there undergo the strangest bends and contortions. When, for instance, a fragment of a dead root or RELATION OF SAPROPHYTES TO THEIR NUTRIENT SUBSTRATUM. 115 underground stem, peculiarly suitable for absorption, is encountered, it is regularly embraced by the suction cells, and as great an absorbent surface as possible is thus brought into contact with the nutritious fragment. Indeed, the development of suction cells on the roots of many gentians (viz. Gentiana ciliata, G. germanica, G. Austriaca, and G. Rhmtica) is confined to the parts of the root-branches, which, in the course of their passage through the vegetable mould, have come into contact with a particularly nutritious portion of it. Wherever there is contact, the root is thickened, and absorption cells project unilaterally from the epidermis and grow into the decaying fragment of wood or bark which is to be drained of its nutrient Fig. 16.— Transverse section through absorption-roots of Saprophytes. 1 Gentiana Bhcetica. a The Bird's Nest Orchis (Neottia Nidus-avis). material (see fig. 16^). Roots of this kind remind one of the root-structures of parasites which are furnished with so-called "haustoria", and which will be discussed more in detail in subsequent pages. But they are different in that tliey absorb food not from living but from decaying parts of the nutrient substratum. Most plants that grow on the vegetable mould of alpine meadows, and the black earth deposited by snow-drifts in mountainous regions, develop flat instead of tubular epidermal cells as suction cells, and in this resemble marsh-plants. In many of these cases the roots are so abundantly and minutely ramified that they form a plexus investing the humus. This is likewise true of the absorptive cells on the rhizoids of mosses. Plants which lie flat against the bark of trees and have no connection with the ground, so that they are unable to derive nutriment from it, have a very peculiar method of maintaining themselves. Their roots, rhizoids, or hyphge, as the case may be, either grow straight into the bark or are merely adnate to its surface. In the latter case they are exposed on one side to the open air, and form more or less projecting lines and ridges ramifying in all directions, often constituting a regular trellis- work cemented to the bark. Sometimes, too, they are represented 116 RELATIONS OF SAPROPHYTES TO THEIR NUTRIENT SUBSTRATUM. by thicker ropes or bands which run longitudinally down or encircle the trunk. These structures certainly serve as instruments of attachment, but at the same time they also absorb nutriment from the substratum, the decaying bark upon which the plant is epiphytic. In periods of drought the absorption of food by plants of this kind is, in general, interrupted and suspended. But when the rainy season commences and there is a long duration of wet weather, water trickling over the surface of boughs and trunks washes the bark, cleanses it as it were, and, falling lower and lower, brings down not only tiny loosened particles of bark but mineral and organic dust which has been blown into it by the wind; it dissolves all the soluble matter it finds on its way, and so reaches the roots, rhizoids, and hyphse which adhere to the bark, in the form of a solution of mineral and organic compounds, chiefly the latter. The trickling water is in some measure stopped by the projecting ridges of these adnate structures; here and there also it deposits particles mechanically suspended in it, and so it conveys to these curious epiphytes the requisite nourishment. In the same way, no doubt, epiphytes which grow upon other epiphytes are nourished. In more inclement regions, the green bark, stem, and, less frequently, the green leaves of the mistletoe are found to be beset by irosses and lichens; and, in the tropics it is a common phenomenon for mosses, liverworts, and even small kinds of Bromeliacese to settle on the green and still living leaves of Bromeliaceoe, Orchideae, and Loi'anthacese, although they are certainly not properly parasitic, and only use their absorption cells for the purpose of clinging to the thick epidermis of the living leaves or stems which support them. The principal part of the liquid substances absorbed by these plants is conveyed to them by the rain-water that washes over the substratum. The species of plants also which have been mentioned as sometimes growing on smooth vertical faces of rock, though the bark of trees is their usual habitat, are able to obtain their food-materials in a similar way. If the summit of a cliff is covered by a continuous carpet of plants, or if ledges and terraces projecting somewhat from its face support sods of grass, tufts of moss, and various small kinds of bushes, it must inevitably happen when there is an abundant fall of rain that the water flowing down the declivity conveys with it organic compounds in solution. First the sods of grass and moss on the ledges and on the top of the cliff are wetted, then the humus, which is their substratum, becomes saturated, and such part of the water as cannot be retained by this humus, or does not percolate into the cracks and crevices of the rock, trickles down from the ledges and moistens the face of the rock as it soaks down to the bottom. A rocky declivity is thus washed in the same way as is the bark of trees, and small fragments of organic and inorganic bodies must of necessity be rinsed out and carried down by the trickling water, and then again be deposited in heaps where projecting obstacles are encountered. It is just in the tracks along which the water flows down steep rocks of the kind that the plants of which we have made mention are situated. Associated with the above are generally a number of other plants, for the most RELATIONS OF SAPROPHYTES TO THEIR NUTRIENT SUBSTRATUM. 117 part microscopic, all of which cannot be classed as saprophytes, but which, in order to be able to thrive in the tracks of trickling water, must have the capacity of surviving desiccation for weeks, and even months, on the barren rock after having been previously supplied with copious moisture for a time. In the case of lichen- growths in particular these are very favourite sites; and when the lichens cover a large area they attract one's attention from afar. In limestone ranges, the light-gray rock of steep declivities, interrupted by ledges covered with grass and low brushwood, is extensively coloured by dark vertical bands and streaks, and the effect is the same as if a dye had flowed from the ledges over the face of the rock. These dark streaks indicate the course of the water which oozes from the humus and renders possible the existence of numberless minute plants on the precipitous face, in particular several dark crustaceous lichens {Acaros2^ora glaucocarpa, Aspicilia fiavida, Lecidea fuscoruhens, Opegrapha lithyrga, &c.). The quantity of organic compounds brought down in solution by the water which filters from the layers of humus on rocky ledges, and that which trickles down the bark of trees, is, however, very small. Still, it is amply sufficient to meet the requirements of the plants occurring at the spots in question. The claims made by them upon their nutrient source are very moderate. We may here recall the instances previously mentioned of mycelia of fungi which have been found satisfied with the scarcely perceptible quantities of organic compounds in water filtering into the shaft of a mine, and in the pure water of a mountain spring respectively. To these instances must here be added the production of mycelia in the wooden pipes through which the clear water of mountain springs is con- veyed. After these pipes, which are made from the trunks of pines, have been used as conduits for years, and their inner layers of wood have long since been washed out, the mycelium of the fungus Lenzites sepiaria is not infrequently developed within them, and in such luxuriance, indeed, that it forms great j^ellowish- gray flocculent masses, which issue from the pipe's inner surface, and float in the stream of running water. In time these flocculent masses increase in the clear spring-water to such a degree that the pipes become completely blocked, and the flow of water is arrested. And yet the water conducted through the pipes is so pure, where it enters into and issues from them, that the residue obtained by the evaporation of hundreds of litres afforded no trace of any organic matter. Seeing that most saprophytes absorb only such a comparatively small amount of organic matter, one is all the more surprised to notice that a large number of them fall suddenly, at certain times, into the opposite extreme. People speak of things rapidly produced in abundance as " mushroom-growths ", and as " shooting up like fungi ". The fructifications of many fungi are in fact developed with a rapidity which borders on the miraculous. The various species of Coprinus living on dung produce their long-stalked, cap-shaped fructifications during the night, and by the evening of the next day the caps have already fallen to pieces, and are in a state of decomposition, and nothing is to be seen in their place but a black deliquescent mass like a blot of ink. The weight of this fructification, thus matured within 118 RELATIONS OF SAPROPHYTES TO THEIR NUTRIENT SUBSTRATUM. twenty-four hours, is certainly many times as great as that of the entire rayceHum which produced it; and it is quite incomprehensible how this mycelium, which for weeks only achieves a moderate development, and adds but little to its dimensions, is in a position suddenly, and in so short a time, to supply the amount of water and organic compounds requisite for the building up of the fructification. Epijpogiwm aphyllum exhibits a similar property. After producing nothing for two years excepting a few branches on its subterranean stem, it develops all at once and in a very short space of time fleshy stems with large flowers, and one asks with astonishment how the relatively small coral-shaped stock sets about obtaining the quantity of nutrient materials necessary for the construction of these flowering stems. We are here confronted again with the great mystery of periodicity, the solution of which we must for the present forego. Saprophytes are much more fastidious as regards the quality of their nutriment than one might expect. It is true that certain fungi are produced wherever there are plants in a state of decomposition, and to them it is quite indifferent whether the mouldered dust, which serves as a nutrient soil for their mycelia, has arisen from one species or another. Also in the case of orchids imbedded in vegetable mould, and in that of most of the mosses and liverworts adherent to the barks of trees, it is, as a rule, of no consequence whether the tree constituting the substratum is a conifer or a dicotyledon. But a large number of species are associated with the decaying remains of particular plants or animals only. For example, certain small species of Marasmius, belonging to the group of the Agarici, occur only on moulder- ing pine-needles; another small fungus, Antennatula jnnophila, is found exclusively on fallen needles of the Silver Fir; Hypoderma Lauri, which resembles small black type on rotting laurel leaves, and the tiny Septoria Menyanthis on leaves of the Bog -bean {Menyanthes trifoliata) lying under water in a state of decay. The cinnamon -coloured receptacles of Lenzites sep>iaria only grow from prostrate trunks of conifers, and the black fuliginous fructifications of Bulgaria polyrtiorplia only on those of oaks. A small discoid fungus named Poronia punctata, white with black spots on the top, is only found on cow-dung; another fungus, Gymnoascus uncinatus on that of mice, and Ctenomyces serratus on decaying goose feathers. That many mosses are also very fastidious in the selection of their substratum has already been intimated. Just as in the Alps Splachnum ampullaceum is onl}' found growing on the putrefying dung of cattle, so in arctic regions the splendid, large-fruited Splachnum luteum and >S'. riibrum occur exclusively on that of rein- deer. Tetraplodon urceolatus is met with on mountains always with decaying excrements of chamois, goats, or sheep for a substratum, whilst Tetraplodon angustatus chooses the excrements of carnivorous animals, and Tayloria serrata is only seen near cow-chalets on decomposing human faeces. The circumstances of the occurrence of another moss belonging to the Splachnaceae, i.e. Tayloria Rudoljiana is also very interesting. It grows usually on the branches of old trees, especially maples in sub-alpine regions, and one is tempted to believe that in respect of its nutrient substratum it is an exception to the rule of the rest of the PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. 119 Splachnacese. But on closer examination there is convincing evidence that tliis moss also lives only on animal dung undergoing putrefaction. For remains of broken mouse and bird bones are invariably to be discovered in the substratum, and there can be no doubt that the Tayloria chooses for its site boughs of old trees upon which birds of prey have dropped their excrements. Of the mosses living on the bark itself, one instance is also worth mentioning. Whereas in the case of most species of the genus Dicranwm, the mouldering residues of conifers constitute the favourite substratum; there is one species, viz. Dicranum Sauteri, which is found only on the bark of the beech. The weather-worn bark of this tree is seen, in sub-alpine districts, covered with the most brilliant emerald-green films of the above-named moss; whilst on adjacent pines and fir-trees no trace of it can be found. PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. A number of plants exhibit contrivances which obviously have for their object the capture and retention of such small creatures as may fly or creep on to their leaves; and it has been ascertained by searching experiments that the majority of these plants use the animals they capture, in one way or another, as sources of nutriment. For the most part the animals that are caught are insects, and hence the term "insectivorous plants" has been applied to the class in question. The flesh of the insect being the part of it principally serviceable for food, the name "carnivorous" or "flesh-eating", or better, perhaps, "flesh-consuming" plants has also been used; and seeing that the most important part of the whole process is really the digestion, or taking in of organic compounds from the captured animals after they are dead, we might call those plants which are furnished with organs for the absorption of the dissolved flesh of animals ensnared by them, " flesh-digest- ing" plants as well. As will appear from the following discussion of the subject, no one of these names completely covers the wonderful phenomena in question, and it is scarcely possible to find a short and not too cumbrous expression which shall henceforward exclude all misconceptions. In round numbers we may estimate the plants which capture animals and demolish them for food at five hundred. Within this comparatively small range, however, the variety of the mechanism for seizure and absorption of nutritive matter is so great that in order to give a general picture of them it is necessary to classify them into several sections and groups. In the first section we have a series of plant-forms wherein chambers are developed, which admit of the entrance of small animals, but not of their escape. The organs of capture and digestion of the plants belonging to this section exhibit no external movements of any kind, and are thereby diflferentiated from the forms belonging to the second section, which perform definite movements, in response to a stimulus caused by the contact of the animals, with the object of covering the prey with as great a quantity of digestive fluid as possible. Lastly, there is a third section wherein the individual forms are 120 PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIISIALS. neither provided with pitfalls nor capable of performing special movements, but have leaves converted into lime-twigs and on them animals stick and are also digested. The first and most extensive group included in the first section is that of Utricularise or Bladderworts. Their capturing apparatus consists of little bladders with orifices closed in each case by a valve, which permits objects to penetrate into Fig. 17.— Bladderworts. In the foreground Utricularia Grafiana; in the background Utricularia minor. the cavity of the bladder, but not to issue out of it. The Utricularise are rootless plants which live suspended in water, and, according to the season of the year, either sink down to the bottom or ascend to just below the surface. Upon the approach of winter, when animal life is gradually disappearing in the chilled and freezing upper layers of water, the leaves at the extremities of the floating stems are enlarged and form spherical winter buds; the older parts of the stems together with the leaves die, their cavities hitherto occupied by air are filled with water, and they sink to the bottom drawing down with them the winter buds. After the winter these buds elongate, detach themselves from the old stems and ascend near the surface, where innumerable little aquatic animals are swimming to and fro, and there develop two rows of lateral branches in rapid succession. Either all of these are thickly covered with leaves which are divided into thread-like, repeatedly PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. 121 bifurcating, segments, or else only half of them are thus clothed with leaves whilst the other half bear the before-mentioned bladders. The former is the case in Utricularia minor, the plant represented in the background of the figure on p. 120; and the latter in Utricularia Grafiana, which is drawn in the foreground. In instances of the former kind obliquely ellipsoidal bladders are to be seen on short stalks on the principal segments of the leaves, usually quite near their angles of bifurcation. In the smaller species, such as Utricularia minor, they have a diameter of about 2 mm. In individuals of the latter kind the bladders have longer stalks, and are about 5 mm. in diameter. They are always pale-green and partially trans- parent. Each bladder is somewhat flattened at the sides and exhibits a markedly convex dorsal surface and slightly curved lateral surface. An orifice, whose border is fringed with peculiar stiff" tapering bristles, leads into the interior of each of these stalked bladders. The aperture has four rounded angles and is framed as it were, by a pair of lips. The under lip is strong- ly thickened, and is furnished with a solid cushion projec- ^ /I 1 I \ \ ^l ting into the inte- / i \\ \ \\ rior of the bladder. From the upper lip Fig. IS.— Traps nf Ut cglecla. Sa thin trans- i Ablacldermagnifled(x4). 2 Section of a bladder, s Absorption-cells on the internal surface of the bladder (x 250). parent, obliquely- placed valve (see fig. 18^), the free edge of which rests upon the inner surface of the cushion before referred to, and closes the entire orifice. This valve is very elastic and yields easily to any pressure from outside. A tiny animal is able, by pressing against it, to force a way without difficulty from the nether lip into the interior of the bladder, and to slip in through the opening thus made. But as soon as the animal has got inside, and ceases to press upon the valve, its elasticity brings it back upon the under lip again. It cannot be opened by pressure from within; for, resting as it does upon the projecting cushion, it is impossible for th^ little prisoner to force it over the latter in an outward direction. The whole apparatus forms a trap for small aquatic animals, they being able, as before observed, to slip into the bladder but not to get out again. Most animals that enter make, it is true, eflEbrts to escape, but they are all in vain. Many perish in a short time — about twenty-four hours — others live from two to three, or, in some cases, even as much as six days. But in the end they must suffer death by suffocation or starvation, and they then decay, and the products of their decomposi- tion are sucked in by special absorption cells developed within the bladder. These absorption cells (see fig. 18^) are linear-oblong and somewhat like little rods in shape, and they line the whole internal surface of the cavity of the bladder. They are arranged in fours, each group of four forming a cross and being united by a 122 PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS common basal cell. The basal cells themselves are intercalated amongst the cells lining the bladder. The organic substances from the decaying bodies of captured animals are sucked up by these stellate groups of cells, and from them pass into the basal cells, and later, into the other adjacent cells of the bladder and those of the plant at large. The majority of the animals caught by the bladders are crustaceans. It is principally larvae and adult individuals of small species of Cypris, DajyJinia, and Cyclops that fall into the trap; but larvse of gnats, and various other small insects, little worms, and infusoria, are also not infrequently met with imprisoned in the bladders. The number of animals captured is comparatively large. In single bladders the remnants of no less than twenty-four small crustaceans have been observed. The prey secured by Utricularia minor (fig. 17), which lives in little pools of still water in peat-bogs, is very abundant. The North American Utricularia clandestinoj seems also to use its capturing apparatus with great success. What it is that induces the animals to press upon the valves and so fall into the trap is not fully explained. We may suppose that they expect to find food in the bladder-cavity, or that they hope it will afibrd a shelter where they can rest for a time and be protected from their pursuers. The last suggestion is especially supported by the circumstance that the approach to the valve-covered orifice of the bladder is guarded against the intrusion of larger animals by stiff sharp bristles which stick out from it (fig. 18 ^ ). Only very small animals, which can easily slip in between the relatively large bristles, reach the inside of the bladder, whilst larger creatures, which would injure the whole apparatus, are prevented from coming near it. Thus, the most probable explanation is that lesser animals pursued by greater take refuge in the hiding-places behind the bristles, and so fall into the trap. Another very striking fact is that the bladders of Utricularige, living in still water, look delusively like certain Ostracoda, especially species of the genus Bajjhnia. The bladder itself resembles the shell-covered body in size and form, and the bristles the antennae and swimmerets of one of these crustaceans. Whether there is any significance in this curious similarity of outward appearance must be left undecided. The majority of Utricularioe live in pools of water beside foot-tracks on moors and in the little collections of water between clumps of reeds in peat-bogs; and these are precisely the haunts of the little creatures that are to fall into the traps. Every handful of water that one scoops up contains hundreds of midge -larvse, water-fleas, Ostracoda, and one-eyed Cyclops, which rush about promiscuously, pursuing and seizing one another. One species of these plants lives in the moun- tains of Brazil in the rain-filled receptacles of Tillandsia plants. The Tillandsia is allied to the pine-apple, and has rosettes of concave leaves, the latter resting one upon the other in such a way as to form a niche or cavity in front of each leaf which fills with rain like a cistern. Many different kinds of small animals are always swimming about in these little cisterns, and almost every one of the latter PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. 123 is the sphere of activity of an individual JJtricularia nelumbifolia. This plant is remarkable also from the fact that long runners are thrown out from its stems, which grow across, in wide arches, from its cistern to a neighbouring Tillandsia, where it selects one of the reservoirs in the rosettes as a new site and dips down into the water — a fantastic method of propagation of which we shall speak again on another occasion. A few Utricularise do not live in water at all, but grow amongst mosses, liver- worts, and lycopods, in the vegetable mould filling the clefts and crevices of rocks, and the bark-fissures of old trees. Of this habit, for example, is the pretty Brazilian JJtricularia r)iontana, which, in spite of the difference of its habitat, is provided with an apparatus for capturing animals agreeing in all essential respects with the description already given. The bladders used by these plants for pur- poses of prey are produced on subterranean filiform stems which thread their way in the vegetable mould and wefts of decaying moss-stems, and here and there swell into tubers. The bladders are hyaline and transparent, and are filled with watery liquid, sometimes also with air. They are only 1 millimeter in diameter, but are present in large numbers. The entrance into these bladders is much more con- cealed than in the species that live in water. The dorsal surface of the bladder being still more strongly curved, the position of the orifice is altered so as to be quite close to the little stalk of the bladder. In addition, the oi^ifice is, as it were, roofed over, and thereby protected against the possibility of being stopped up by particles of earth, and the passage leading to it is very narrow. That, in spite of the difficulty of entrance, a number of minute animals do seek a hiding-place here is proved by the circumstance that, besides various infusoria, rhizopoda, and creatures of that kind inhabiting damp earth, species of Acarus and larvse of other animals have been found, both dead and alive, in the bladders. With this first group of insectivorous plants, wherein the capturing apparatus includes a valve to prevent the egress of such animals as fall into the trap, is associated in the first section a second group, viz. that of the ascidia-bearing or pitcher-plants, in which the foliage - leaves are converted into pitfalls, and the escape of the captured prey prevented by a number of points lining the inner wall of the cavity, and directed from the aperture towards the closed bottom. There is an extraordinary variety in the form of the pitfalls. Sometimes they are tubular, utricular, or funnel-shaped cavities, sometimes mug or pitcher-shaped, or urceolate; in some cases these are straight, in others bowed like sickles, or spirally twisted. They always arise from the part of the petiole upon which the lamina immediately rests. The lamina is always relatively small, being represented in the majority of the traps by a scale or lobe, and it only appears to be an appendage of the large expanded and hollowed -out petiole. In many pitcher- plants the little lamina looks like a lid placed over the orifice to the pitfall, as, for instance, is shown in the illustration (fig. 21 *), whilst in others {Nepenthes ampidlaria and N. vittata) it has the form of a handle or stalk, and serves as a place for animals visiting the pitchers to alight upon. 124 PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. In each pitfall there are always three kinds of contrivance to be distinguished: first, a device for the allurement of animals; secondly, an arrangement for entrap- ping the prey enticed, which at the same time prevents individuals once imprisoned from returning and escaping through the entrance hole; and thirdly, a structure for causing the decay or dissolution of the dead animals at the bottom of the pit- falls, and for rendering possible the absorption of the products of decomposition as nutriment. The means of allurement are similar to those which cause the visits ■of small creatures to flowers, that is to say, principally honey and bright and varied coloration, whereby the nectar -secreting spots are recognized from afar, especially by flying insects. The escape of animals when they have once entered the cavity of a petiole is prevented, as has been already mentioned, by a fringe of sharp hairs pointed downwards, or by various spinous structures on the inner Fig. 19.— Spinous Structures in tlie Pitfalls of Carnivorous Plants. 1 Genlisea; a piece of the tube seen from inside. 2 Heliarnphora nutans; spines on tlie walls of pitfalls, s Snrracenia purpurea; a piece of the lining of the pitcher near tlie orifice seen from insitle. * Sarracenia purpurea; longitudinal section through the membrane covered with spinous bristles in the lower part of the pitcher. ^ Nepenlhcs hybrida; fringe of spines at the orifice of the pitcher, i, 2, 4^ s greatly magnified ; s sliglitly magnified. surface of the cavity. The decomposition and dissolution of the prey are effected by fluids secreted by special cells at the bottom of the utricles and pitchers. But although in respect of the consecutive and co-ordinate operation of these three kinds of contrivance, all ascidia- bearing and pitcher -plants resemble one another, there are considerable individual divergences as to structure and function that it is well worth while to study in some detail the most noticeable of them. One of the most noteworthy is the genus Genlisea, which is nearly related to Utriculariace« in the structure of its flowers and fruit. It is composed of a dozen species growing in water and marshy places. Of these one is a native of tropical and southern Africa, whilst others are found in Brazil and the West Indies. In addition to ordinary leaves, which in them are spatulate, most of the Genliseae possess leaf-structures metamorphosed so as to constitute pitfalls. Each pitfall consists of a long, narrow, cylindrical utricle, which at its blind end is enlarged into a bladder, whilst at the narrow orifice at the opposite end are placed two peculiar ribbon-shaped processes twisted spirally. The orifice of the utricle is set with very small sharp teeth bent inwards; and the tubular part of the utricle has its inner surface lined throughout with innumerable little bristles, which arise from rows PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. 125 of cells forming inwardly projecting ridges, and have their sharply-pointed tips directed downwards (see fig. 19 i). Amongst these needles are also found, scattered over the whole internal surface, roundish wart-like glands or papillae, composed of four or eight cells. The bottom of the bladder-like cavity in which the utricle terminates is destitute of bristles, and provided only with glands arranged in rows. Small worms, mites, and other segmented animals which enter through the orifice of the utricle can easily reach the enlarged base. But as soon as tliey try to com- Fig 20 — Sarracenia pxirpui mence the return journey they are opposed by the points of a thousand bristles^ Thus caught they die, and the products arising from the decay of their bodies are absorbed by the glands situated, as above mentioned, at the bottom of the bladder and on the walls of the utricle. As types of a second series of carnivorous plants belonging to the group of pitcher-plants may be taken Heliamphora nutans, a native of moorlands on the mountains of Roraima, on the borders of British Guiana, and Sarracenia purpurea (see fig. 20), which is widely distributed in the marshes of eastern North America from Hudson's Bay to Florida. In both instances the leaves metamorphosed into ascidia are arranged in rosettes, rest their bases on damp earth and thence curve upwards. They are somewhat inflated, like bladders, at about their middle, but contract again at the orifice where they pass into the relatively small laminjB. The latter are threaded by red streaks like blood-vessels, have the form of valves,. 126 PLAXTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. and turn their concave surfaces towards falling rain. They serve, moreover, at least in Sarracenia purpurea, to catch the drops of rain, which then flow down into the bottom of the ascidia and fill them more or less with water. There is very little evaporation from the hollow pitchers; and even when there has been no rain for a week, one always finds some of the previously-collected water at the bottom. The inner surface of a pitcher is lined by cells arranged like the scales of enamel on a pike's back (see fig. 19-). The internally-projecting wall of each of these scales is transformed into a stiff decurved point, and the lower the position of the cells the longer do the points become. The shell-like lamina again, above the contracted orifice, bears glandular hairs which exude honey, so that the parts surrounding the aperture are covered by a thin film of sweet juice. Many animals are attracted by this honey. Some are winged and alight from flying; others, being wingless, make use of a peculiar ridge, which projects on the concave side of the utricle, to help them to creep up the latter. If these honey- eaters happen to travel away from the lamina to that part of the pitcher which is lined with the smooth and slippery decurved cells, they are as good as lost. They slip down over the brink, every attempt to climb up again being rendered futile by the downwardly-pointing needles which clothe the lower part of the wall; and ultimately they fall into the water collected at the bottom, where they are drowned and their bodies putrefy. The products of decay are absorbed as nutriment by the epidermal cells in this region. The number of animals meeting with this fate is often so great that an offensive odour, arising from the decaying bodies, is emitted by the utricles and is noticeable at a considerable distance. In the wild state, the ascidiform utricles are often half-full of drowned animals and it is stated that in these circumstances birds also put in an appearance and pick some of the dead remains out of the utricles. Whether the liquid filling the bottom of the pitchers consists simply of rain- water, or whether the latter is modified by a secretion originating in the gland- like groups of cells there (see fig. 28^), is still uncertain. A centipede over 4 centimeters long having fallen into a utricle of Sarracenia 2^urpurea in the night was found only half immersed in the water. The upper half of the creature projected above the liquid, and made violent eflbrts to escape; but the lower part had, after a few hours, not only become motionless but had turned white from the effect of the surrounding liquid; it appeared to be macerated, and exhibited alterations which are not produced in so short a time in centipedes immersed in ordinary rain-water. When a number of captured animals are undergoing putre- faction at the same time in a pitfall, the liquid turns brown and has the appearance of manure-liquor. There is a great difference between the utricles of Sarracenia purpurea and the apparatus adapted to the capture of prey in the plants of which we have chosen as examples, Sarracenia variolaris, a native of the marshes of Alabama, Florida, and Carolina, and the Darlingtonia Californica, found growing at a height of from 300 to 1000 meters above the sea on Calif ornian uplands from the borders of PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. 127 Oregon to Mount Shasta. In both of these the liquid with an acid reaction, which fills the bottom of each utricle, is certainly only secreted by the cells in the interior of the cavity itself, and it is quite impossible that a single drop of the rain or dew deposited upon the plant should reach the interior of the cavity. The hollow petiole is in both plants, above mentioned, utricular or tubular, and only slightly Fig. 21.— Ascidia-bearing and Pitcher-plauts. 1 Sarracenia variolaris. ^ Darlingtonia Californica. » Sarraceuia laciniata. * Nepenthes villosa, reduced to one-half natural size. enlarged towards the top. The dorsal side of each leaf is, however, at its upper end hollowed out like a cowl or a helmet, and forms a cupola as is shown in fig. 21 - and 21 2. The orifice or entrance into the utricle is consequently covered over and is reduced to a slit or hole under the hood. The lamina is trans- formed into a lobe, which in Sarracenia variolaris is small and roofs over the orifice of the utricle, and in Darlingtonia is shaped like the tail of a fish, and hangs down in front of the aperture. The lower part of the utricle is of a uniform green colour, but the upper part {i.e. the cupola and lobe-like appendage) has red ribs and veins, and here and there is quite purple. Between the veins the 128 PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. leaf is thin, translucent, and pale-green or whitish; and these clear translucent patches, framed by purple or green ribs, look as if they were little windows, especially when seen from within the utricle. The mixture of green, red, and white gives the upper parts of the leaves such a gay appearance that, from a distance, they might be mistaken for flowers. Insects are doubtless attracted by these bright colours, and both round the orifice, and on the inner surface of the cupola, they find exudations of honey which they suck or lick up with avidity. In Sarracenia variolaris, honey is to be seen besides, on the edge of a broad free border which is decurrent along the utricle, and extends from the ground to the orifice. This border forms a favourite path for wingless insects, especially ants, which are particularly eager in their quest for honey. For them it is a sure way to destruction, for when they, gradually following the honey-baited pathway, arrive at the orifice to the utricle and pass through it, they inevitably get upon the smooth decurved points of the epidermal cells, constructed just like those in Sarracenia purpurea, and then, unable to stop themselves, slip down to the bottom of the pitcher. When small winged insects alight from flying and fall down the slide into the interior, they make use of their wings in the hope of saving themselves, but they never succeed in finding the aperture by which they entered, as it slants downwards and is situated in shadow. They invariably try to escape through the cupola, mistaking the thin portions, through which the light penetrates into the interior, for gaps permitting egress. But just as flies in rooms dash against the windows hoping to pass through them into the open air, so the small insects in the utricles of Sarracenia variolaris and Darlingtonia Galifornica knock against these windowed cupolas, in their desire to save themselves by flying through. They always fall down again to the bottom of the utricle as though into a cistern. If they are immersed in the liquid there secreted, or only in partial contact with it, they are stupefied, but not immediately killed. They often live incarcerated for two days, and it would therefore be erroneous to suppose that the fluid in the pitchers acts on the prey as a deadly poison. But it assists the decay and dissolution of the captives as they die of starvation and suflfocation, and, as in the case of the utricle-plants previously described, a brown liquor of very unpleasant odour is produced, and there is a residue of solid pieces of skeleton difficult to decompose, such as the wing-cases, claws, and thoraces of various beetles, lice, ants, and other small insects which have shared the same unlucky fate. The number of animals captured is very considerable. The pitchers of Sarracenia variolaris, which attain to a length of 30 cm., are usually found, when growing in their natural habitat, filled to a height of from 8 to 10 cm. with animal remains, and even a heap 15 cm. high has been observed. We must here remark that in the ascidia of Sarracenia variolaris, wingless insects, which creep about the earth, are found to predominate, whilst in Darlingtonia, on the contrary, most of the insects are winged. The cause of this is easily understood. The former plant has honey exuding on the flap or ridge running down from the orifice to the PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. 129 ground, and many wingless insects are thus induced to climb up the alluring path and to enter the cavity of the pitcher. Darlingtonia, on the other hand, is destitute of honey on its decurrent ridge, and only provides the sweet meal at the top in the vicinity of the orifice, where it is available for flying insects, which, as a rule, only visit nectar-secreting flowers. The purplish-red scale, shaped like a fish's tail, and hung out like the sign-board of an inn in front of the entrance to the pitcher, constitutes an instrument for the attraction, from afar, of these winged creatures, which are endowed with a vivid sense of colour; and, as experience shows, it does not fail in its object. What significance is to be attributed to the spiral torsion of Darlingtonia leaves (see fig. 21 ^) it is difficult to say. Perhaps the escape of animals once imprisoned in the depths of a pitfall is hereby rendered more remote. It would at all events be much more difficult for an insect trying to escape by the use of its wings to ascend a canal which, in addition to being lined with decurved points, was spirally wound, than a similar canal, straight and widened towards the top. We must not omit to mention that a few flies and a small moth have selected as their ordinary habitat the pitchers of both the plants just described, in spite of their being so fatal to most insects. The grubs of a blow-fly (Sarcophaga Sarracenice), in particular, live in large numbers amidst the heaps of decaying insect bodies at the bottom of the pitchers, and are there nourished just as are the grubs of allied species in the rotten flesh of birds and mammals. When mature, the grubs quit the environment of dead remains, passing through holes which they bore in the side wall of the pitcher, and turn into chrysalises in the earth. But the fly itself can without danger pass in and out of the pitfalls, which are so perilous in the case of other insects, and it is enabled to do this by means of the special structure of its feet. On the last joint of each foot it has a long claw and sole-like attachment-lobe, and it is able to push these appendages between the sharp, slippery, decurved hairs lining the inner surface of the pitcher, and so to hook itself to the deeper strata of the wall. This apparatus may be likened to the grapple-like climbing irons of Tyrolese mountaineers, and, thus armed, the fly is in a position to ascend the inner wall of a pitcher unscaleable by other insects. The case of the small moth Xanthoptera sertiicrocea is similar. The tibi^ of this insect are armed with long, sharp spurs, one pair on each of the two middle legs, and two pairs on each of the two hindermost legs; and, by the help of these spurs it likewise is able to tread uninjured over the dangerous surface of the wall. Its caterpillars, too, cover the sharp slippery hairs with a web, and so render them harmless. The presence of these animals in the death-traps of Sarracenias is of special interest, inasmuch as it shows that the animals which perish at the bottom of the pitchers are not exactly digested. If maggoty flesh enters the stomach of a carnivorous animal, not only the flesh itself but the maggots as well (which, indeed, immediately die on reaching the stomach) are speedily dissolved by the action of the gastric juice. Such is also the case with several animal-capturing plants to be described in the next pages. But the fluid secreted in the pitchers of Darlingtonia Vol. I. ^ *= "^ 9 130 PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. and Sarracenia variolaris cannot exercise this digestive action, for if it did the maggots in the heap of rotting insects could not remain alive and well. Its action is limited to the promotion of decay and the formation of a foul liquor, in other words, a liquid manure, which is absorbed as nutriment by the epidermal cells at the bottom of the pitchers. Another series of pitcher-plants comprises forms in which the petioles are converted into symmetrical sacs with apertures at the top, and the laminae spread out over them like lids for protection. Most frequently the pitfalls in plants of this kind are shaped like pitchers, jars, urns, cups, or funnels; and the lid over the orifice of each cavity is, for the most part, so placed as to prevent rain-drops from falling in, but not to hinder in any way the entrance of animals. In this series are included, firstly, a few species of Sarracenia, viz. Sarracenia Drummondii and S. undulata, next, the Australian Cephalotus follicularis, and lastly, the numerous species of the genus Nejpenthes, which are designated by gardeners by the name of "pitcher-plants" in the narrow sense. The leaves in both the Sarracenias just named are heteromorphic. Some of them have acute linear-lanceolate petioles of a uniform green colour, and not hollowed out; and it is only in the case of from three to five leaves in each individual plant that the petioles are transformed into tubes with inf undibulif orm enlargements at the top. The rim round the mouth of the funnel is somewhat swollen and doubled down externally; but above the orifice the lamina is arched so as to form a cover to the pitcher. The margin of the leaf of Sarracenia laciniata, which is shown in fig. 21 ^, is crinkled and sinuously folded. The cover and also the upper funnel-shaped enlargement of the pitcher are very conspicuous on account of the contrast of the colours displayed upon them. The green of the lower part of the pitcher gets paler and paler above, and merges into a pure white, whilst dark-red veins stand out from the green and white ground tints, having the efifect of a net-work of blood- vessels. At the mouth of the pitcher, and on the under side of the lid, honey is secreted in such abundance that little drops of it are not infrequently to be seen on the swollen rim, and some oozes down into the infundibuliform portion of the pitcher. But at the very spots where the honey occurs there are also innumerable smooth conical cells with their solid apices directed downwards; and these cells become longer the lower their position in the pitcher. When insects, attracted by the gaj^'-coloured lid, and lured on by the honey, come to the mouth of the pitcher and tread upon the parts covered with the sharp slippery papillae, they are drawn into the depths as though by an invisible power. After they have once alighted on the perilous area, every movement and every effort to climb up against the points causes them to slide further and further down towards the bottom of the pitcher, where they are hopelessly lost, being killed within a short time and ultimately decomposed. An instance of an exactly similar kind is aflforded by Cephalotus follicularis, which has long been known as a plant native on moorlands in eastern Australia, [t is allied to saxifrages and currants, and is represented on a scale of half the PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. 131 natural size in fig. 22. This Cephalotus also has two kinds of leaves, which are closely crowded in a rosette round the erect flower-stalk. Only the lower leaves of the rosette are transformed into traps for animals, and these are pre-eminently adapted for wingless creatures creeping upon the earth. The tankard-shaped traps all rest on the damp earth, and are furnished externally with borders or winged ridges, which facilitate the ascent of crawling animals to the mouth of the tankard. Flying insects are of course not excluded, and here again they are made aware from afar of the feast of honey provided by the presence of bright colours. The half-open lid is very prettily adorned with white patches and brilliant purple veins, and at a distance is readily mistaken for a flower. When small animals, whether with or without wings, approach to take the honey, they are so eager in their search that they get upon the inner surface of the mouth of the tankard-pitcher, which, though fluted, is also very smooth and slippery, and thence they easily slide into the interior of the cavity. The pitchers being half-full of liquid, most of the un- lucky creatures die there in a short time by drowning. But even if this were not the case, they would never succeed in working their way up to the light of day. For every animal that wishes to save itself from a Cejyhalotus pitcher has three obstacles to overcome : first, a circular ridge projecting mside the pitcher; sec- ondly, a bit of wall thickly covered with little papillae, sharp, ridged, and pointed downward, the whole being comparable to a flax-comb; and, lastly, on the involute rim round the mouth of the pitcher, another fringe composed of hooked, decurved spines which bristle like an im- penetrable row of bayonets in front of such animals as may have surmounted the other difficulties. The abundance of the booty found at the bottom of Cepha- lotus pitchers shows how efficiently these contrivances serve to prevent escape. Ants, for instance, sacrifice themselves recklessly in their pursuit of honey, and one often finds great numbers of them drowned in the liquid in the pitchers. The prey is not in this case converted into a putrid liquor, but is partially dissolved by a secretion having an acid reaction. This secretion is separated out by special Fig. 22.— Cephalotus follicularis. 132 PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. Fig. 23.— Young Nepenthes plants. glandular cells situated on the lining of the pitcher; and the whole process, wherein they are concerned, corresponds to that which obtains in the pitchers of Nei^enthes, and which will be more thoroughly discussed in the case of these latter plants. The species of the genus Nepenthes, of which we know at the present time thirty-six, are all confined to the tropics. Their area of distribution extends from New Caledonia and New Guinea over tropical Australia to the Seychelles Islands and Madagascar, and over the Sunda Islands, the Philippines, Ceylon, Bengal, and Cochin-China. They only flourish on marshy ground on the margin of small collections of water in damp primeval forests. There the seeds germinate in shallow water. The young plants (see fig. 23), which spring from the boggy ground, have their leaves ar- ranged in rosettes just like those of Sarracenias (see fig. 20). They are, too, so nearly identical in form with the latter that anyone seeing a young Nepenthes plant for the first time, and not knowing the history of its development, would take it for a Sarracenia. The leaves, succeeding the cotyledons and forming a circle above them, rest their lower portions upon the mud, but their upper parts are curved upwards, and each carries at its extremity a scale resembling a cock's comb, which is, strict speaking, the lamina. This scale roofs over a slit-like aperture, the entrance to a cavity within the swollen petiole. In addition a green lobe with a few coarse projecting points is to be seen on either side of the orifice. Altogether diflferent from the rosettes of 3"oung Nepenthes plants are the foliar structures clothing the stems which subsequently arise from the rosettes (see fig. 24). In these leaves the lower part of the petiole is winged and flat, has a linear or lanceolate outline, and resembles the leaf -blade of Dracaena; its functions, too, are those of a green lamina. This expanded section of the leaf-stalk passes next into a part which is terete and coiled like a snake, and acts as a tendril. Every stem or branch belonging to a plant, whether living or dead, with which this part of the petiole comes into contact, is seized and encircled by it; and the third portion of the petiole, i,e. the pitcher, being situated at the extremity of this clasping portion, IS thus slung upon the branch of some other plant growing at the edge of a pool of water. Meanwhile the Nepenthes plant rises higher and higher above the wet soil where its seeds germinated and the young rosette rested, becomes entangled with the ramifications of the underwood and with prostrate branches of trees of the primeval forest; in a word, with everything available as a support, and so not infrequently climbs, as a true liane, to the tops of trees of moderate height. The pitcher must be looked upon as an excavated portion of the petiole, and PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. 133 134 PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. what appears to be the lid of the pitcher is the lamina, as it is in Cephalotiis and the Sarracenias. In this case also the lamina seems to be but little developed in comparison with the wonderfully metamorphosed petiole. In the majority of the species of Nepenthefi, the mature pitchers are from 10 cm. to 15 cm. in height. In the graceful Nepenthes ampullaria they are only from 4 cm. to 6 cm. high; but, on the other hand, in the species indigenous to the primeval forests of Borneo they reach a height of 30 cm. or even more. The pitchers of Nepenthes Rajah have a height of 50 cm., and their orifices are 10 cm. in diameter, whilst below the orifice they expand to 16 cm.; so that if a pigeon were to fly into a pitcher of this kind it would be completely hidden in it. Immature pitchers are still closed by their covers. Often they are hairy outside; and, according to the colour and lustre of the hairs, they may be rusty in tone or glittering like gold; not rarely they look as if they were powdered with flour (e.g. N. albo-marginata), and sometimes are even snow-white. Subsequently the lid is raised, and the downy coat disappears either partially or entirely. Having thus become glabrous, the pitchers display a yellowish- green ground colour, for the most part flecked and veined with purple; and many are of a bluish, violet, or rose tint near the orifice, or dark-red as though saturated with blood. The lid is similarly gaily coloured; and the variety of the tints is increased by the fact that a pale-blue zone is visible in the interior, beneath the swollen involute rim of the opening, which is itself brownish, yellowish, or orange- red. Gaily-coloured pitchers of this kind look at a distance just like flowers, and remind one, in particular, of the most brilliant floral forms of the liane-like Aristolochias indigenous to tropical forests. This fact is the more noteworthy, because the genus NepentJtes is closely allied to the genus Aristolochia in respect of systematic relations. The bright pitchers of Nepenthes, visible from afai*, are sought, just as flowers are, by insects, and probably by other winged creatures as well; and this occurs all the more because there is a copious secretion of honey by the epidermal cells upon the under surface of the lid, and on the rim round the mouth of each pitcher. The swollen and often delicately-fluted rim, in particular, drips and glitters with the sugary juice; and it would be permissible in this connection to speak of a honeyed mouth and sweet lips in the most literal sense of the words. Animals which suck honey from the lips of Nepenthes pitchers wander, as they do so, only too readily upon the interior surface of the orifice. But the inner face is smooth and precipitous, and rendered so slippery by a bluish coating of wax that not a few of the alighted guests slip down to the bottom of the pitcher and fall into the liquid there collected. Many of them perish in a short time; others try to save themselves by climbing up the internal face of the pitcher, but they always slip again on the polished, wax-coated zone, and tumble back once more to the bottom. In large pitchers the involute rim of the aperture is in addition armed with sharp teeth, which are pointed downwards and bristle in front of such of the unlucky victims in the pitfall as try to emerge (see fig. 19^). In a number of species (iY. Raffiesiana, N. echinostoma, N. Rajah, N Edwardsiana, and N. Veitchii, all PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. 135 natives of Borneo) this fringe of sharp teeth looks like the set of teeth of a beast of prey; and in Nepenthes villosa, of which a pitcher is represented in fig. 21 *, a double row of bigger and smaller teeth directed towards the bottom of the pitcher is developed, and renders the escape of prey, once caught in the trap, impossible. Most of the creatures that fall into the pitchers are, however, speedily drowned in the large quantity of liquid at the bottom. For a third part or even a half of the cavity is filled with liquid. This liquid originates from special gland-cells on the inner surface of the pitcher, consists mainly of water, and so long as there are no animals in the pitfall, gives only a very weak acid reaction. But as soon as the body of an animal reaches the bottom, more fluid is secreted. This has a distinctly acid taste, possesses the power of dissolving albuminous substances, such as flesh and coagulated blood, and corresponds, not only in respect of this action but also in chemical composition, to the gastric juice. For, in addition to organic acids (malic, citric, and formic acids), an organic body like pepsin has been detected in it, and nitrogenous organic compounds have been brought into solution in it artificially as well. If the liquid from a Nepenthes pitcher, which has not yet captured any animal, is poured into a glass vessel containing a small piece of meat, the flesh is at first but little afifected ; but, if a few drops of formic acid are added, the flesh is dissolved and undergoes the very same changes as it does in the stomach of a mammal. The process going on in the pitchers of Nepenthes when animals fall into them is therefore not only analogous to digestion, but may be properly designated digestion. The digested portions of the bodies are afterwards absorbed by special cells at the bottom, and on the lower parts of the lining wall of the Nepenthes pitchers. Another series of plants was at one time regarded as belonging to our present section of carnivorous plants. These include forms possessing subterranean stem structures, bearing hollow, scale-like leaves, or leaves so arranged that chink-like spaces exist between them. Into these chambers or spaces it was supposed that minute animals, Infusoria, Rhizopods, Aphides, and the like found their way, and that here they met their death, their bodies being digested through the agency of peculiar glands which line the walls of these chinks and spaces. Though this view of the carnivorous function of these subterranean organs has failed to become established on a solid basis of fact, the plants in question are of considerable interest and may be conveniently treated here. One of the most remarkable of the plants belonging to this group is the Tooth- wort {Lathrcea Squamaria), of which we shall repeatedly have occasion to speak. It is nearly allied to the Yellow-Rattle and Cow-wheat, but it is destitute of chlorophyll, and lives underground, parasitic on the roots of arborescent Angio- sperms, except during a brief period annually when it sends up above-ground a few short shoots covered with flowers. The subterranean stems are white, have a fleshy, solid, and elastic appearance, and are covered throughout their entire length with thick squamous leaves placed closely one above the other (see fig. 25 ^ and fig. 37). In colour and consistence these leaves are like the stem; in outline they 136 PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. are broadly cordate, and they give the impression of being mounted fairly and squarely upon the stem by means of the highly swollen and notched basal portion. But it is only necessary to detach one of the scales from the stem to convince one's self that this is not the case, and that the part taken at first sight to be the underside or back of the leaf is only a portion of the superior surface. In reality each of these thick squamiform leaves is rolled back, and in it the following parts may be distinguished: first, the place of insertion on the stem (fig. 25^), which is relatively small; secondly, the portion taken on cursory examination to be the whole upper surface of the leaf, and consisting of an obliquely ascending blade limited by a sharp border; next, starting from this sharp border, the part which, owing to its being suddenly bent down at an acute angle and falling away steeply, is usually taken for the dorsal or inferior surface of the leaf, but which belongs, in point of fact, to the front of the lamina; fourthly, the free extremity of the leaf in the form of an involute limb; and fifthly, the true dorsal part, which is very small relatively and is not visible until the involute tip is removed. Owing to the involution of the apex, a canal or rather a recess is formed and runs across beneath the leaf, close under the place where the latter is joined to the stem (see fig. 25 "). From five to thirteen (usually ten) chambers open into these recesses through a series of little holes. They are excavations in the thickness of the scales and are probably, in this form at any rate, unique in the realm of plants. These extraordi- nary chambers must be described as deep excavations in the foliar substance proceeding from the back of the leaf. With a view to elucidating their function in relation to the life of the plant, their structure must be more particularly considered. The chambers radiate as it were from the orifice at the base of the leaf. Though closely adjoining one another, they are not in lateral connection by means of pass- ages or canals. Their walls are irregular and undulating (see fig. 25 ^), and are characterized by the peculiar structures which are borne on the lining — raised up above the ordinary epidermal cells and projecting into the cavity. These structures, of two sorts, are shown in fig. 25 *, under a considerable magnification. One sort, and these are by far the more numerous, are of the nature of short capitate hairs. The head is formed of a pair of cells, and they are supported on a short cylindrical cell which serves as a stalk. The other sort is sparsely scattered amongst these capitate hairs. They are oval in outline and but slightly raised above the ordinary epidermal cells. Each consists of a tabular cell upon which rests a slightly convex • cushion composed of not more than four cells all lying in the same plane. One such sessile gland is shown in the centre of fig. 25 *. In this case the cushion consists of three cells. A further peculiarity has been observed in these sessile glands. The summit of each is marked by a tiny pore (not shown in the figure), an actual hole in the wall at the geometrical centre of the convex surface. In the wall of the chamber, just below the lining epidermis, run the vascular strands (fig. 25 ^). The vessels of which they are composed form a considerable plexus or net-work in this region. Now it is known that the ground in which PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. 137 Lathrcea passes its existence is often drenched through with moisture in the immediate neighbourhood of the plant. A water-excreting function has long been attributed to the chambered leaves of the rhizomes. That such is the case has been demonstrated by forcing water under pressure through the cut ends of the rhizomes, when streams gush forth from the basal orifices of the leaves. In this Fig. 25.— Glandular structures in the Toothwort, Bartsia, and Butterwort. iPiece of an underground leaf-shoot of the Toothwort. ^Longitudinal section through the same; x2. sLongitudinal section through one of these underground leaves; x60. *Piece of the wall of a cavity; x200. sSubterranean bud of Lartnia; natural size. 6 Cross-section tlirough part of this bud; x60. 'The margin of a bud-scale in section; x200. spiece of the epidermis of a leaf of Butterwort; x ISO. 9 Transverse section through the leaf of a Butterwort (Pinguicula alpiiia); x 50. 10 Transverse section through Butterwort leaf; natural size. instance it is uncertain whether the stalked or the cushion glands assist in this excretion, though from the minute details of their structure it would seem probable that it is the latter. On any other hypothesis it is difficult to understand the meaning of the pore on the summit. The matter has, however, been placed beyond doubt by experiments on other allied plants, as, for instance, the Lousewort (Pedicu- laris imlustris), in which the glands are more easily kept under observation. We have apparently in these gland-bearing chambers of Lathrcea a water-excreting mechanism for the elimination of the surplus moisture, which in most plants is transpired or evaporated into the air. Lathroaa being almost wholly subterranean 138 PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. is unable to do this, as the air in the chinks and crannies of its matrix of soil is generally saturated. The water is therefore excreted in liquid form by a special mechanism. This view of the function of the scales is confirmed b}^ reference to other allied types with subterranean scales. An instance in point is afforded by Bartsia aljyina. This remarkable plant is distributed in the Arctic region and amongst the high mountain flora throughout almost the whole of Europe, and is very striking owing to the colour of its foliage being a mixture of black, violet, and green. The flower, too, is of a sombre dark- violet hue, and the entire plant, by reason of this peculiar colouring, gives a truly funereal impression. We may remark incidentally that the name Bartsia was chosen by Linnaeus for this sad-hued plant as an expression of his own grief at the death of the zealous naturalist and physician, Bartsch, who was his intimate friend, and who succumbed at a comparatively early age to the climate of Guiana. Damp black earth in the neighbourhood of springs constitutes the favourite habitat of these plants. Upon digging in summer time down to their roots, one sees that a few suckers proceed from them, and fasten upon the sedges and other plants growing in the vicinity; but one also discovers subterranean shoots having " root-hairs " developed near the nodes, at which are inserted the paired white scales ; and these " root-hairs " have the function of absorption-cells. To- wards the autumn, oval buds, likewise subterranean, are matured, in form not unlike horse-chestnut buds (see fig. 25 ^), and composed of etiolated scales arranged in four rows and overlapping one another like tiles, so that only the back of the upper part of each scale is visible, the lower part being covered by the scale next beneath it. On the visible part of each scale's convex under surface three sharply projecting ribs are noticeable near the middle, whilst the two margins are rolled back so as to form a recess in each case. But, as may be seen in the cross-section of a Bartsia bud (see fig. 25 ^), one pair of scales lies over the next higher pair in such a way as to convert the recesses into ducts. Owing to this construction the interior of the bud is perforated by twice as many ducts as there are covered leaf-scales, and the orifices of each pair of ducts occur at the spots where the evolute margins of one scale begin to be covered by the middle of the next lower scale. On one wall of the ducts, i.e. in the recesses, structures like those which occur in the cavities of Lathrcea are developed, i.e. stalked glands, each composed of two cells borne upon a basal cell; secondly, pairs of hemispherical domed cells; and, lastly, ordinary flat epidermal cells (see fig. 25 ''). There can be little doubt that the whole apparatus acts in the same way as in Lathrcea, and is adapted to the excretion of water. The cavities and spaces between the scales of the buds serve the same purpose as the chambers in the leaves of Lathrcea, viz., that of aflfording cover to the delicate excretory glands and of px-otecting them from immediate contact with the soil. Mechanisms of this sort are not restricted to subterranean organs, but are found likewise on the aerial leaves of many plants. Indeed such arrangements, supple- menting ordinary transpiration, are common, especially amongst tropical plants. PLANTS WITH TRAPS AND PITFALLS TO ENSNARE ANIMALS. 13 Root with sucliers; natural size. 2 Piece of a root with sucker in section; x35. constricted near their insertion, and the strangulated portion often gives the impression of being a pedicel upon which the knob is seated. This knob is ditFerentiated into a central core and a multicellular, cortical coat enveloping it. The cellular coat rests upon the root of the host attacked, and does not merely adhere to one limited spot, but spreads itself out over the root like a plastic mass, and forms a cushion surrounding about a fourth or fifth part of the circum- ference (see fig. 36^) without, however, penetrating into the substance of the root. There are in the core two strands or bundles of vessels, and between them small cells arranged in rows, from which absorption-cells arise at the spot where the sucker first applied itself to the nutrient root. These absorption-cells grow out beyond the rind-like envelope round the core, perforate the cortex of the host, penetrate into the wood at the centre of the invaded root, and there diverge like the hairs of a dry paint-brush. The suckers of the green-leaved Rhinanthaceae are on the whole similarly constructed; only they are relatively smaller and more delicate, being sometimes almost translucent, and they are either not at all or only slightly constricted at the Vol. I. "^ 12 178 CLIMBING PARASITES. GREEN-LEAVED PARASITES. TOOTH WORT. base. Whereas in ThesiuTn they never issue otherwise than laterally from the ramifications of the roots, in Rhinanthacese they are often terminal. A differentia- tion into core and rind-like envelope is never clearly marked; a vascular bundle runs through the middle of the sucker and is surrounded by thick-walled cells. The absorbent cells are, moreover, shorter than in the Santalaceee. The individual genera of the Rhinanthace^e exhibit amongst themselves only very slight differences in respect of their suckers. On the roots of Eyebright (Euphrasia), the haustoria are tiny roundish nodules which rest upon the host's root without encompassing it. The absorption-cells are very short, and only just penetrate into the host. The vascular bundle is either entirely wanting within the sucker, or its place is taken by a single, comparatively large vessel. On the roots of the Yellow-rattle (Rhinanthus) the suckers are spherical and of considerable size (up to 3 mm. in diameter); their margins are swollen and often encompass more than half the circumference of the roots attacked. The absorbent cells are short but very numerous. In the Cow-wheat {MelainpyruTn) the suckers resemble those of the Yellow-rattle in size and shape and in the shortness of the absorption-cells; but in the former the margins of the suckers not only embrace the roots of the host, but cling to them in such a way as to penetrate their substance and form circular grooves upon them. All the Rhinanthaceae mentioned are herbaceous annuals. Their suckers are few in number, and therefore easily escape observation. By the time these plants ripen their seeds any piece of a root that has been attacked has for the most part already turned brown and been killed, and is in a state of decay. But shortly afterwards the parasite itself withers. The comparatively large seeds, well- furnished with reserve-material for the nourishment of the embryo, fall out of the dry capsules, and generally reach the ground at no great distance from the mother- plant and germinate there. In the autumn, close to Cow-wheat plants, which are still green but have already let fall the seeds from their lowest capsules, individual examples of those seeds may be seen already sprouting in the damp moss and mould on the ground of woods. If they fall to earth not very far from the parent-plant, the seedlings may happen to attack the host which has already had one of the branches of its root sucked and killed by the latter in the previous summer. Nearly all these annual green-leaved parasites make their appearance in num- bers close together. If, for instance, a species of Cow-wheat has taken up its quarters in a particular part of a wood, there are always collections of hundreds and thousands of specimens to be found together. The small-flowered Yellow- rattle often grows so abundantly in damp meadows that one might suppose it to have been sown by the bushel. The large-flowered, hairy Yellow-rattle is similarly exuberant in ploughed fields, and the Eyebright, with its large number of species, is produced in such abundance in mountainous districts that, at the season when its little milk-white flowers are open, regular milky ways seem to stretch across the green meadow^s. Millions of them are situated together rooted in the grass-covered ground, and one would suppose that in course of time the growth of CLIMBING PARASITES. GREEN-LEAVED PARASITES. TOOTHWORT. 179 grass at such places would be injured. This conclusion appears to be supported by the assertion of the country folk that after the season when the Eyebright is in full bloom, the cows yield less milk, a fact which explains the German name of " Milchdieb " (milk-thief) popularly given to the plant. The diminution in the quantity of milk yielded is, however, certainly connected with other circumstances. It depends especially upon the universal abatement of the growth of grasses in early autumn and the consequent curtailment of the food afforded by the pastures. The injury done by the Eyebright to its hosts by the withdrawal of nutriment and destruction of rootlets cannot be very considerable, for the appearance of the grasses and other host-plants, which are affected, is not noticeably different from that of the plants of the same kind which escape invasion. The same statement is true in the case of the various species of Lousewort {Pedicular is), almost all of which are meadow-plants; that is to saj^ they are present in great abundance in upland and alpine pastures without apparently injuring the species growing in their company and used by them as hosts. Unlike the species of Cow-wheat, Yellow-rattle, and Eyebright, however, nearly all the Louseworts are perennial, and accordingly differ from them also in the construction of their suckers. There is, it is true, no difference in shape between the suckers of the Cow-wheat and those of Pedicidaris, but they are dissimilar in respect of size and place of origin. The suckers of the perennial Louseworts are barely more than half the size, and are only developed near the attenuated extremity of a rootlet. They are very few in number; each of the long, thick, fleshy rootlets, proceeding from the base of the stem usually produces a single sucker only which settles upon the root of a suitable host-plant in the same way as the suckers of Cow-wheat. By the time that the parasite's fruit ripens, the piece of root which has been invaded has usually already turned brown and fallen into decay. Now in the case of Cow-wheat it may undoubtedly be immaterial whether the piece of root attacked by it is living or not when its fruit is ripening, inasmuch as its own annual root rots as soon as the seeds have been produced from the flowers above ground. But with Pedicularis it is different. The perennial roots of this plant require a host to nourish them next year, and when the piece of a host's root which has been attacked and sucked as a nutrient substratum one year dies, the sucker belonging to the root parasitic upon it is no longer in a position to fulfil its function by continuing to absorb fresh juices. Suckers thus reduced to a state of quiescence soon perish, and only leave little scars to indicate the places where they existed. The perennial root of the Pedicularis has now to seek a new source of nutriment, and this is effected by the elongation of its tip, which continues to grow until it reaches the living root of another plant suitable as host, whereupon it develops a fresh sucker upon that root. This elongation doubtless requires a large quantity of plastic materials; but these are found stored in abundance in the older parts of the parasitic root. These circumstances explain, at anyrate in part, the characteristic structure and disproportionate length of the roots of Pedicularis. From all round the short erect 180 CLIMBING PARASITES. GREEN-LEAVED PARASITES. TOOTHWORT. root-stock, which is generally only from i cm. to 2 cm. long, issue fleshy rootlets of the thickness of a quill, but, in many species, as long and thick as a little finger. These rootlets are abundantly supplied with starch, and, in course of time, elongate till they measure 20 cm. They radiate in all directions in the black soil of the meadow, wherein are buried the root-systems of grasses, sedges, and various other plants, and fasten on to suitable hosts by means of one or two suckers yearly, and repeating this process until at length their tips travel into earth devoid of roots, where no more prey is to be found, and there growth ceases. This explains also why these long Pedicularis-roots never descend vertically in the earth, but remain only in the upper strata of soil on a meadow, where a number of other roots are interwoven together, and where it is most likely that the tapering growing-point will meet with the root of some new host or other. The Alpine Bartsia (Bartsia alpina), one of the perennial Rhinanthacese prevailing in the arctic regions as well as in mountainous parts of Europe on damp, marshy, grass-covered spots, is distinguished by the sombre dusky violet colouring of its leaves, and has already been noticed amongst carnivorous plants. On the secondary roots are suckers exactly like those of the Yellow-rattle (Rhinanthus), and by means of these organs it clings to the fibrous roots of sedges and grasses, and sucks their juices. The long, subterranean, runner-like stems, which are covered with small, whitish scales, also bear, however, elongated absorption-cells (root-hairs), which are distinctly differentiated, and take up nutriment from the vegetable mould around. This Bartsia is, therefore, half-parasitic and half-saprophytic, and it is not improbable that many other perennial Rhinanthacese behave in the same wa,y. The species of Pedicularis which constitute the most extensive group of perennial green-leaved and parasitic Rhinanthacese are, it is true, destitute of tubular absorption-cells (root-hairs) whether on the subterranean stem-structures or on the root-tip, with the exception of those which develop in the middle of the suckers. But the construction of the epidermal cells on the roots, and the circum- stance that these epidermal cells are always in intimate connection with dark particles of humus, would favour the idea that these plants are capable of taking up organic compounds from the mould of meadows in addition to the food acquired by means of suckers from their hosts. This supposition is further supported by the fact that I succeeded in rearing a species belonging to the Rhinanthaceae, namely. Odontites lutea, from a soil composed of a mixture of sand and humus, in which no other plants were rooted, so that the possibility of a withdrawal of nutritive matter from hosts was excluded. It is true that the plants thus reared remained comparatively small and poor, and only developed few flowers and fruits. But at anyrate they may be considered to prove that plants exist, which, though normally parasitic, are yet on occasion able to subsist in vegetable mould without the assistance of hosts. The third series of parasitic flowering-plants is very restricted, contrasting in this respect with the second series, composed of the numerous green-leaved Santalacese and Rhinanthaceae. The species belonging to it differ from those of the CLIMBING PARASITES. GREEN-LEAVED PARASITES. TOOTHWORT. 181 second series chiefly in their lack of chlorophyll. They all live underground on the roots of trees and shrubs, develop deep down in the earth a number of flowerless perennial shoots thickly covered with scales, and, in addition, push up annually into the light temporary axes bearing flowers, which ripen their fruits and die after the fall of the seed. As the best known representative of this series, we may take the Toothwort (Lathrcea Squamaria), which is represented in fig. 37, and has been already Fig. 37.— Toothwort (Lathrcea Squamaria) with suckers upon the roots of a Poplar. described on a previous occasion as an instance of a plant possessing in the seclusion of its curious hollowed scale-leaves a special mechanism for the elimi- nation of water from its system quite supplementary to the normal method of surface transpiration. Formerly, the Toothwort used to be included in the family of Broom-rapes (Orobancheas) on account of the structure of its capsules, but it is entirely different as regards the form of its seedling. For, whereas the seedling of a Broom-rape is a thread without any trace of cotyledons, as will be seen when we study its development and mode of attachment to the host in the next few pages, that of the Toothwort is clearly difl"erentiated into radicle, cotyledons, and rudimentary stem, corresponding in this respect entirely with the Rhinanthaceae. Moreover, the Toothwort resembles Rhinanthacese much more than Broom-rapes in the manner in which it attacks its hosts and withdraws nutriment from them. 182 CLIMBING PARASITES. GREEN-LEAVED PARASITES. TOOTH WORT. The seed of Lathroea germinates on damp earth. The young root of the seedhng grows at first at the expense of reserve material stored in the seed, penetrates vertically into the earth and sends out lateral branches, which, like the main root, follow a serpentine course and search in the loose damp earth for a suitable nutrient substratum. If one of these meets with a living root belonging to an ash, poplar, hornbeam, hazel, or other angiospermous tree, it fastens on to it at once and develops suckers at the points of contact; these suckers are at first shaped like spherical buttons, but soon acquire, as their size increases, the form of discs adherent to the host's root by the flattened side and with the convex hemispherical side turned towards the rootlet of the parasite. These discoid suckers cling to the root attacked by means of a viscid substance produced by the outermost layer of cells. As in the case of the parasites already described, a bundle of absorption-cells grows out of the core of each sucker into the root of the plant serving as host, and the tips of the absorbent cells reach to the wood of the root. The shoot extremity of the seedling, thus nourished by the juices of the host, now develops very quickly, elongating and producing thick, white, fleshy, scale-like leaves which overlap one another closely, the whole thus acquiring the appearance of an open fir-cone. The scaly stems also branch underground, and thus a curious structure is gradually produced, consisting of crossed and entangled cone-like shoots covered with white scales, and this structure fills entirely the nooks and corners between the woody roots on which it preys. Individual plants extending over a square meter and weighing 5 kilograms are by no means rare. Later on, inflorescences raise them- selves above the surface from the extremities of the scaly subterranean shoots. Their axes are at first curved like crooks, but straighten themselves out by the time the fruit ripens. Whereas the subterranean portions are white as ivory, the flowers and bracts pushed up above the earth are of a purplish tinge. The roots, which issued originally from the seedling, and their suckers have long since ceased to meet the requirements in respect of nourishment of so greatly augmented a structure, and therefore additional adventitious roots are produced every year, springing from the stem and growing towards living woody branches of the thickness of a finger, belonging to the root of the tree or shrub that serves as host. When there, they bifurcate, forming numerous thickish filiform arms, which lay themselves upon the bark of the nutrient root and weave a regular web over it. Sometimes two or three of these root-filaments of the parasite coalesce, forming tendrils, and the resemblance to a lace-work or braid is then all the more pronounced. Suckers such as have been described are developed by these root- filaments laterally, and more especially on the ends of the branches. Lathroea is interesting in so many different connections that we shall again return to this plant later on. As has been stated before, it affords a type of a series of parasites which resembles the species of Cassytka and Cuscuta in the absence of chlorophyll, Rhinanthacese in the shape and development of the seedling and the form of the suckers, and the Balanophorese, presently to be described, in being parasitic upon the roots of woody plants. Lathrwa Squamaria, the species repre- BROOM-RAPES, BALANOPHORE^, RAFFLESIACE^. 183 sented in fig. 37, is indigenous to Europe and Asia, its area of distribution extending from England eastwards to the Himalayas, and from Sweden southwards to Sicily. Two species are confined to the East, the Crimea and the Balkans, and another Toothwort (Lathrcea clandestina), distinguished by large flowers, but slightly raised above the earth, extends in western and southern Europe from Flanders over France to Spain and Italy. This last has the distinctive feature that the discoid suckers developed on its yellow roots, which latter are of the thickness of a quill, are as large as lentils and the biggest hitherto discovered on any plant. BEOOM-RAPES, BALANOPHORE^, RAFFLESIACE^. The fourth series of parasitic Phanerogamia is composed of plants destitute of chlorophyll, whose seed contains an amorphous embryo without cotyledons or radicle. The seed germinates on the earth, and the embryo grows as a filiform body into the ground and there fastens upon the root of a host-plant, penetrates into and coalesces with it in growth, forming a tuberous stock, from which, later on, flowering stems are projected above the earth. To this series belong the Broom-rapes or Orobancheae and the Balanophoreae. Of the genus Orobanche about 180 species are recognized, which, exhibiting great uniformity in floral structure and in their general development, can only be distinguished by minute characteristics. The flowering stem growing up from the subterranean tuber is, in all the species, rigid, erect, thick, fleshy, and covered at the top with dry scales. The open flowers, ringent in shape, are crowded together in a terminal spike, and often emit a strong scent like that of pinks or sometimes of violets. The colour of the flowers is in one group (Phelypcea) mostly blue or violet; in the rest it is waxen yellow, yellowish-brown, dark-brown, rose-red, flesh-tint, or whitish. Orobanche violacea and 0. lutea, both natives of Northern Africa, have stems which grov/ to a height of half a meter and become almost as thick as an arm. The best-known species is the Branched Broom-rape (Orobanche ramosa), which is parasitic on the roots of hemp and tobacco plants, and is very widely distributed. The greatest number of species belong to the East and to Southern Europe. The extreme north of America harbours one species which bears a single flower at the end of its stem. In all the species the stem projects only partially above the earth. The subterranean portion, adherent to the root of a host, is often greatly swollen and thickened above the place of attachment; in the case of Striga orobanchoides, which is prevalent in the Nile basin, it is irregularly lobed above the host's root. The root of the nutrient plant also is usually somewhat swollen wherever a parasitic Orobanche has settled upon it, and sometimes it exhibits an irregular outgrowth inclosing the spot whereto the Orobanche is adnate like a cup. Beyond the place of attachment of the parasite the root has often the appearance of having been bitten oflf, and this is owing to the fact that the particular piece of root has been killed and demolished by the attack of the parasite. From the base of the stem, near the point of adhesion to the host, spring short, thick, fleshy fibres, and 184 BROOM-RAPES, BALANOPHORE^, RAFFLESIACE^. one or other of these bends its tip towards the root of the foster-plant and clings to it. These fibres are, in many species, very numerous, and are interlaced and entangled so as to form a reticulate mass, which vividly recalls that of the Bird's- nest, and is an instance of the general resemblance existing between Orobanchese and the Orchidese destitute of green leaves {Neottia, Corallorhiza, Epipogum, Limodorum), which have already been discussed. The establishment of parasitic Orobanchese upon the roots of host-plants takes place in the following manner. The embryo imbedded in the small seed shows no trace of differentiation into root and stem, possesses no cotyledons, and indeed consists only of a group of cells; it is surrounded by other cells filled with reserve- nutriment. When this embryo grows forth from the seed, during which process it consumes the reserve-food, it exhibits no distinction between root, stem, and leaf, but is a spiral filament consisting of delicate cells. One extremity, the shoot end, of this filiform seedling, remains covered by the seed-coat, which looks like a dark cap (fig. 34^); the opposite extremity is the root. The seedling Broom-rape stretches downwards just as the Dodder (Cuscuta) extends upwards. In so doing the descending tip traces a spiral line, and so, as it were, seeks in the earth for the root of a plant suitable as host. If the search is fruitless, and if the reserve-material in the seed has meantime been altogether consumed, the seedling begins to wither and gradually shrivels, turns brown, and dries up. It lacks the power of nourishing itself by means of the surrounding earth. But, if the lower, foraging extremity of the seedling succeeds in finding a live root belonging to a plant able to serve as host, it not only adheres closely to it, but swells in such a way as to give the young plantlet a flask-shaped appearance (fig. 34^ and fig. 34 ^''). The upper end is still inclosed by the seed- coat, but in proportion as the lower part thickens, the upper shrivels till no trace of it is left. The thickened part, on the other hand, which has become attached to the root of the host, becomes nodulated and papillose. Some of the papillse develop into elongated conical pegs, and the young Broom-rape now rests upon the nutrient root in the shape of the head of a fighting-club (see fig. 34^^). At the place of attachment one of the conical pegs has meanwhile penetrated the cortex of the root, and there it continues to grow energetically, forcing the cortical tissue apart, until it reaches the wood. Vessels now arise in the body of the young club- like plant, and, passing through the middle of the plug, wedged in the nutrient root, are brought into connection with the vessels of the latter. At the point of union between host and parasite, a bud is formed, clothed with abundant scales, which may best be likened to the bulb of the Martagon Lily. Lastly, out of this bud grows a strong, thick stem, which breaks through the earth and lifts a spike of flowers into the sunlight. That portion of the Broom-rape which is buried in the root of the host-plant is so intimately associated with the separate parts of that root in the development of a tuber that it is usually difficult to determine which cells belong to the parasite and which to the host. The degree of union is such that one cannot even state with BROOM-RAPES, BALANOPHORE^, RAFFLESIACEiE. 185 certainty where the epidermis of the nutrient root ceases, and that of the Broom- rape begins. The latter looks as if it were a branch growing out of the root it preys upon, and this apparent fusion gave some colour to the view of the earlier botanists, who, ignorant of the life-history of these parasites, believed that they did not arise from seeds, but were pathological outgrowths of the roots, produced from their tainted juices; in other words, that they were "pseudomorphs" sprouting from diseased roots in the place of leafy branches. It is also deserving of mention that some of the thick, fleshy fibres issuing laterally from the nodulated seedlings curve towards the host's root, bury their tips in the cortex, and thenceforth behave exactly like the peg which was inserted at the point where the seedling first became attached. We must leave undecided the questions as to whether the other fibres, which terminate freely in the earth, are capable of taking up food-materials from that source, whether these fibres are only present in perennial species and become the starting-points of new individuals, and lastly, whether they should be looked upon as root-structures or as stem-structures. In addition, it is noteworthy that in many Orobanchege only those embryos continue to develop which meet with a plant suitable to be their host. Although it is not the case that every species of Orobanche adopts one particular species of plant as foster-parent, yet thus much is certain, that most of them only thrive on members of a limited circle of species; one lives exclusively on kinds of Wormwood, a second on species of Butter-bur, and a third on those of Germander. For example, Orobanche Teucrii prevails on Teucrium Chamcedrys, Teucrium mon- tanum, &c., the hosts being invariably species of the genus Teucrium. Suppose a hill thickly covered with plants comprising Teucrium montanum growing in company with thyme, rock-roses, globe-flowers, sedges, and grasses, but no great abundance of the Teucrium, a plant belonging to the species named occurring only here and there, and let Orobanche Teucrii have established itself at one particular spot, have attained to flowering and developed fruits, the tiny seeds of which have been shaken by the wind out of the ripe capsules. Owing to the exceptional minuteness and lightness of its seeds, every gust of wind will scatter them in innumerable quantities over the entire hillside and beyond it. The next step is germination. Filiform embryos emerge from the seeds, in the manner described above, and penetrate into the earth. Teucrium, montanum being only sparsely present on the hill in question, comparatively few seedlings will meet with the roots of that plant, whereas thousands will fall in with the roots of the thymes, rock- roses, globe-flowers, sedges, and grasses. But, curious to relate, only those seedlings of Orobanche Teucrii which come into contact with the roots of Teucrium montanum establish themselves firmly, penetrate into them, and continue their development; whilst the numerous individuals which touch the roots of the thyme and other plants perish. This phenomenon can scarcely be explained in any other way than by the supposition that the roots of Teucrium montanum alone, by virtue of their special structure and quality, aflbrd a suitable nutrient substratum, and therefore constitute centres of attraction for seedlings of Orobanche Teucrii; 186 BROOM-RAPES, BALANOPHORE^, RAFFLESIACE^. and that the roots of the thyme, rock-roses, and other plants growing upon the hill side by side with Teucriura montanum do not share this property. Whereas the Broom-rapes constitute a family of plants, the species of which, though very numerous, are so similar in the structure of flowers and fruit, in the history of their development and in the general impression they convey, that it is necessary to discover minute distinctive marks in order to be able to classify them with tolerable completeness, the Balanophoreoe, which, together with these Oro- banchese, belong to the fourth series of parasitic Phanerogams, are related to one another in a manner quite the reverse. Only forty species of them are known, but they are so various that, on the basis of the obvious differences, no less than fourteen genera have been distinguished, among which the forty species are fairly equally divided. In respect of distribution and occurrence they also contrast strikingly with both Broom-rapes and Rhinanthacece. The Orobanchese belong in particular to the Mediterranean flora, and to the East, and the Rhinanthacese, as has been already stated, adorn chiefly sunny pastures in arctic regions and in moun- tain districts of the northern hemisphere. Balanophorese, on the other hand, are only found within a belt encircling the Old and New Worlds, which stretches little beyond the equatorial zone to the north or south, and they almost all inhabit the dark bed of primeval forests, where they are parasitic on the roots of woody plants, beneath a covering of vegetable mould. The genus of Balanophorere named Langsdorffia is confined exclusively to tropical America. One of its species {Langsdorffia Moritziana) is found native in the damp forests of Venezuela and New Granada, where it is parasitic on the roots of palms and fig-trees; a second species {Langsdorffia rubiginosa) occurs in Guiana and Brazil in the region of the sources of the Orinoco, and a third, the most common of all {Langsdorffia hypoga^a) represented in fig. 38, has an area of distribution extending from Mexico to the south of Brazil. They all avoid the hottest districts, remaining rather in cool regions; indeed the species first named has been found at an elevation of from 2000 to 3000 meters. Unlike all the rest of the Balanophoreae, Langsdorffia exhibits a branched, cylindrical stock ascending from the place of attachment to the nutrient root, more or less felted externally, and before putting forth any flowers has a remote resemblance to a doe's antlers with their winter covering of downy skin. These stems are almost as thick as a little finger, have a fleshy consistence, and exhibit a clavate expansion at the base where they rest upon the root of the host. Many of those stems which bear the male flowers are 30 cm. long; those which bear the female flowers are usually somewhat shorter. They are all of a pale-yellowish colour; the thickly tomentose Langsdorffia rubiginosa looks as if it were covered with a yellowish velvet. At the extremity of each of the ramifications of the stem, which are often extremely short, having then the form of lobes or knobs, a bud is developed sooner or later in the lower cortical layer. This bud swells, bursts the outer layer of cortex, uplifts itself and grows out as an inflorescence between the four lobes formed by the cruciform rupture of the bark. The inflorescence is surrounded, like BROOM-RAPES, BALANOPHORE^, RAFFLESIACEyE. 187 the capituhim of a composite, by a whorl of imbricating scales, of which the lower are shorter and broader, and the upper longer, narrower, and pointed at the apex. These scales being stiff, somewhat shiny, and varying in colour from a waxen yellow to orange or red — in the case of Langsdorffia Moritziana brown-red, — the whole inflorescence has a vivid resemblance to certain immortelles, namely, the large species of Helichrysum occurring at the Cape. The inflorescences bearing male flowers are elongated and egg-shaped, those possessing only female flowers are shorter and capitulate. The seeds dropped from the nut -like fruits, which are pulpy internally, have no special integument. The embryo exhibits no trace of iu' 3t5 ^Lanj d ijjia hji jan, fiuiu Centi d Ameiicx cotyledons or radicle, but consists of an undifferentiated group of cells which may be likened to a tiny bulbil. Seeds of this kind germinate like those of Lathrcea, and upon meeting with the root of a tree or shrub suitable for prey, develop into larger tubercles and have a remarkable effect upon the substratum. The cortex of the host-root is destroyed at the place of adhesion of the tubercle, and its wood is laid open, lacerated, and unravelled. The woody bundles are diverted from their previous direction, ascend towards the parasitic tubercle, which meantime has grown into a full-sized tuber, and spread out like fans. The cells and vessels of the parasite penetrate between the ascending wood-fibres, and this results in the formation of a zone at the place of union of the parasite and root, where cells and vessels belonging to both inter- lace, traverse, and join one another, coalescing completely in exactly the same way as happens in the case of the species of Toothwort. A similar phenomenon occurs also when one of the wavy stems of Langsdorffia comes into contact with a root adapted to the purpose. The cortex of the root is demolished at the place of 188 BROOM-RAPES, BALANOPHORE^, RAFFLESIACE^. contact; the wood is exposed, split open, and unravelled, whilst the tissue of the parasitic stem fills up all the interspaces in the upeurved and sundered woody- bundles and fibres, and so intimate is the union thus effected that the stem of the Langsdorffia might be taken to be a branch of the root of the host-plant which sustains it. At the point of connection of an already adult Langsdorffia stem, the hypertrophy of the tissue is not very striking; but the base of each stem of an indi- vidual produced from a seed presents a highly swollen and clavate appearance. At first the parasite is only fastened by one side of this thickened base to the nutrient root, but later on it wraps both sides round the root, and rests upon the latter like a saddle on the back of a horse. Between the bundles of a Langsdorffia stem there are passages filled with a peculiar wax-like matter named balanophorin. The quantity of this substance is so great that if one end of a stem of Langsdorffia is lighted, it burns like a wax- taper, and in the region of the Bogota these Langsdorfiias are collected and sold under the name of "siejos", and are used for illuminating purposes on festive occasions. In New Granada they have also been employed in the making of candles; and, although this source of wax is not sufficiently abundant for us to be able to believe in its consumption and conversion on a large scale, the fact of its application in this manner shows that the parasite we are discussing must occur in great exuberance in many tracts of country in Central America. Much rarer than the parasitic Langsdorffias are the species belonging to the genus Scybalium. Like the former these are confined to the equatorial zone of America. Two species, viz. Scybalium Glaziovii and S. depressum, flourish in mountainous districts, one of them indeed occurring only on the mountains of New Granada; two other species (Scyhalium jamaicense and S. fungiforme) live in the woods and savannahs of lower-lying regions. The aspect of the last-named species when seen growing on the ground of a primeval forest, tempts one to suppose it to be a fungus, and it is easily understood why the first discoverer selected the term fungiforme to apply to it. Figure 39 ^, representing this rare and marvellous plant, is taken from the original specimens discovered in the year 1820 by Schott in the Sierra d'Estrella of Brazil, and brought thence by him to Vienna. We see that, in this case, instead of the elongated, wavy, branched stem characteristic of Langs- dorfiias, a lumpy, tuberous mass rests upon the root of the host-plant. This tuber is sometimes rounded and sometimes compressed and discoid; it is nodulated and often irregularly lobed also, and grows to the size of a fist. It is developed from a seed which, as is the case in all Balanophoreoe, is a cellular structure without integument containing an embryo destitute of cotyledons and radicle, and is best described as a minute tubercle. The embryo, after emerging from the seed and finding the living root of a woody plant, increases in volume, and, in the form of a little knob the size of a pea, exercises the same influence on the plant preyed upon as has been noted in the case of Langsdorffia. The root attacked is stripped of bark at the place where the tubercle is attached; the wood is then resolved into a fringe of fibres which stand straight up, and, diverging like the spokes of a fan, BROOM-RAPES, BALANOPHOREiE, RAFFLESIACE^. 189 distribute themselves in the tissue of the parasite, the latter having in the mean- time developed into a tuberous stock as large as a nut. These radiating bundles, issuing from the wood of the nutrient root, come then into such intimate connection with the vessels formed in the tuber of the parasite, that the one appears to be a continuation of the other. They are, besides, entangled together, and between them is intercalated a mass of small parenchymatous cells which also adheres to the yet unfrayed portion of the foster-root's wood, and coalesces with it. The tuberous body of the parasite, which in the first instance is only adnate to the host on one . ^^-^t^'likfVc. ■ Scybalium fungi forme, from Brazil. -Parasitic Balanophorese. ^Balanophora nUdenbrandtii, from the Comoro Islands. side, gradually encompasses it entirely, and the nutritive root then appears to perforate this irregular tuber. The inflorescences are produced direct from buds, which are formed under the bark at projecting spots of the brown tuberous stem, the cortex bursting open and allowing a thick flesh-coloured shoot, closely beset by ovoid pointed scales, to emerge and grow up into a form resembling a mortar-pestle. At the summit this shoot expands into a disc, and upon this are borne little capitu- late groups of flowers, which are inserted amongst innumerable quantities of scales and hairs. The pistillate and staminate flowers are separated in different inflo- rescences, whilst the entire structure has an undeniable resemblance when in bloom to the inflorescence of an artichoke gone to seed, and later on to a toad-stool. In the eastern hemisphere we find the various species of the genus Balanophora replacing the Langsdorffias and Scybalia. One of these, Balanophora Hilden- 190 BROOM-RAPES, BALANOPHORE^, RAFFLESIACE^. brandtii, which is represented on the left side of the figure 39, occurs in the Comoro Islands oflf the east coast of Africa; seven species inhabit the islands of Java, Ceylon, Borneo, Hong-Kong, and the Philippines, and three species the East Indies. Balanophora fungosa, first discovered by Forster, is parasitic on the roots of Eucalyptus and Ficus, and is indigenous to Australia and the New Hebrides. The more elevated regions of Java and the Himalaya abound especially in these singular organisms. Balanophora elongata is so prevalent in Java on mountains of between 2000 and 3000 metres, that it is collected in quantities for the sake of the wax-like matter obtained from it. In that island candles are made from Balanophoras as they are from Langsdorfiias in New Granada, or else rods of bamboo are smeared with the viscid substance, as they are then found to burn quite quietly and slowly. In the Himalaya, Balanophora dioica or B. p>olyandra are the commonest and most widely distributed species, and Balanophora involucrata is there met with upon the roots of oaks, maples, and araliads even at a height of from 2300 to 2500 metres above the sea-level. They possess in almost all cases very vivid and conspicuous colouring — deep-yellow, purple, red-brown or flesh-tint, thus resembling the Gastromycetes, Clavarieae, and Toad-stools, in whose company they grow, and with which they manifest an additional uniformity in being all of fleshy consistence and containing no trace of chlorophyll. At a certain distance, moreover, the inflorescences rising from the dark ground in a wood, have the appearance of fungi, and all the early observers describe these Balanophoreae with one accord as truly abnormal growths, viz. as fungi which by some marvellous accident bear flowers. They were also the object of the boldest speculations and most exuberant imagery on the part of the botanists belonging to the school of the " nature philosophers " of the first decades of this century. Even as late as the forties a famous German botanist says of them : " The}^ are in the position of a hiero- glyphic key between two worlds, which intercept and evade one another in an infinite variety of ways, like dreaming and waking moments", and the worth}' Junghuhn, who discovered several of these plants in Java, writes: "Those are words which we may hope will be rightly interpreted thousands of years hence. Their sublime truth aflected me deeply. There, flowerless and leafless, stood the mysterious plants which aflford an instance of the combination of special vessels in a stalk like that of Balanophorese with the fructification of imperfect Hypho- mycetes!" A young Balanophora not in flower is not unlike a Seybalium in appearance at the corresponding stage of its development. It consists of an irregular tuberous stem, which rests upon the creeping root of a tree or shrub. The exterior of this structure, which sometimes attains to the size of a man's head, is uneven, and in some cases convoluted like the human brain, or it may project in humps and knobs, or be divided into lobes or short branches like a coral-stem. The resemblance to the latter is heightened by the fact that the surface is covered by little papillce shaped like stars or forget-me-nots, which distinguish the genus Balanopltora from all allied genera. BROOM-RAPES, BALANOPHORE.E, RAFFLESIACEiE. 191 The seeds settle upon the roots of trees, develop into tuberous axes, and unite with the nutrient root in the same manner as the Balanophorese already described. Also the inception of the rudimentary inflorescence beneath the cortex of the tuber and its eruption are similarly accomplished. In this genus the cortical layer thus broken through and forced outward always forms a large cup-shaped or crateriform sheath with an irregularly-lobed margin surrounding the base of the inflorescence. Fi^'. iO.— Parasitic Lulauophuieie. phalloides, from Java. * Helosis gujaiiensis, from Mexico. The inflorescence itself is spadiciform, and is borne by a thick shaft beset with large squamous leaves. The spadices growing from a tuber-stock are, for the most part, only as long as a little finger, but occasionally they reach a height of 30 cm., as, for example, is the case in the Balanophora elongata of Java, which is parasitic on the roots of TJdbaudia. The species of the American genus Helosis, whereof the most common {Helosis gujanensis) is represented above, resemble those of the genus Balanophora in the shape of the inflorescence. There is, however, considerable difierence in the method adopted by these Helosis species of settling upon the roots of host-plants and in 192 BROOM-RAPES, BALANOPHORE.E, RAFFLESIACE^. the whole mode of growth. The phenomena of the swelling of the embryo into a tubercle after it has chanced upon a nutritive root, the destruction of the cortex, the exposure of the wood at that part of the root where the tubercle is adnate, and the derangement of the course of the woody bundles ensue, it is true, in the same manner as in the other Balanophorese ; but the frayed wood-bundles of the foster- root only form quite short lobules which penetrate but a short distance into the parasitic tuber-stock, whilst the vascular bundles, formed meantime in the latter, adhere to them in such a manner that they might be mistaken for direct continua- tions of them. When once the parasitic tubers have thus become adnate to a root, and by means of this union are provided with food, they grow round the nutrient roots in such a way that the latter appear to perforate or actually to issue from the tubers. They are always roundish, brown outside, and warty, but without scales, and they never produce inflorescences directly, but put forth in the first place several whitish or yellowish runners varying in thickness from a quill to a finger, which creep along horizontally under the ground, bifurcating, and becoming interlaced with other ramifications. At the places of contact they coalesce, and so occasionally form a net-work which is almost inextricably entangled with the root- system of the plant preyed upon. Whenever a runner of this kind comes into contact with a living root belonging to the host -plant, the surface of contact at once swells up. The part afiected is converted into a tuberous mass and becomes adnate to the root, the process being the same as occurs in the case of the tubercle pro- duced from seed, A net-work of runners thus connected with the root-system of the nutrient plant at several spots by means of tubers as large as peas might be compared to the reticulum woven by Lathrcea round the roots of its hosts; but, apart from the size, there is the essential difference that inflorescences are never pro- duced from the white threads of the ramifying and sucker-bearing roots of Lathrcea, whereas the runners of Helosis afford points of origin for new inflorescences. Warts are produced on the surfaces of the thicker cylindrical runners, and within these are developed the buds of the inflorescences. The outer coat of the warts is then rent open at the top and constitutes a little cup, out of which grows a naked, scale- less shaft terminated by an oval spadix. Seeing that the runners rest horizontally under the earth whilst the shafts ascend bolt upright from the ground, the latter are always at right angles to the runners, of which they are to be regarded as branches. The flowers are grouped in capitula, presenting in the spadix a dense mass. They are protected by peculiar bract-scales, each of which by itself is like a nail with a facetted head. These heads are in close contact with one another, so that the young inflorescence seems to be inclosed in a panelled coat of mail, and resembles to a certain extent a closed fir-cone. By degrees, however, these bract- scales detach themselves and fall off", and thus the flowers, till then roofed over by them, become visible. When the seeds are mature, the whole runner concerned in the production of the inflorescence, and usually also the tuber which served as the BROOM-RAPES, BALANOPHORE^, RAFFLESIACEiE. 193 starting-point of that runner, perishes, and another tuber belonging to the net-work above described, or rather the system of runners proceeding from it, becomes the basis for the development of new inflorescences. To this extent we may regard these Helosis species as perennial plants, whereas the majority of the other Balanophoreoe can lay no claim to this distinction, inasmuch as in their case the whole plant dies after it has flowered and ripened its seeds. The floral spadices in Helosis have a purple or blood-red colour, and in Brazil are called "Espigo de sangue". Only three species of Helosis have been discovered up to the present time, and those are distributed over equatorial America, in the Antilles, and from Mexico to Brazil. Nearly allied to Helosis is the genus Goryncea, which resembles it in having facetted bract-scales like nails and a cone-like inflorescence, but differs entirely in other respects in its mode of growth, especially in being without runners. Four species of this genus have been discovered in the Andes of South America, in Peru, Ecuador, and New Granada, where they are parasitic, like the rest of the Balano- phorese, upon the roots of trees. One of them, Goryncea Turdiei, is worthy of notice as living on the roots of Peruvian-bark trees, and is rendered conspicuous by its purple spadix, borne on a white shaft. Rhopalocnemis phalloides (see fig. 40^) is another root-parasite related to Helosis, and the single representative in Asia of these pre-eminently American groups. It is found preying upon the roots of fig-trees, oaks, and various lianes, in mountainous parts of Java and the eastern Himalayas, and is one of the biggest of all the Balanophoreae. The fleshy, yellowish or reddish-brown tuber-stock attains to the size of a man's head; the inflorescences, which burst from the protuberances of this lumpy mass and are from two to six in number, are over 30 cm. long and from 4 to 6 cm. thick. The protuberances are light-brown in colour, and resemble in form a cycad-cone, Rhopalocnemis, a drawing of which is given in fig. 40 ^ on a scale of one-half the natural size, is distinguished, like Goryncea, from Helosis by having no runners issuing from the tuberous axes. The Lophophytese are set apart as a further group of parasitic Balanophorese, and differ from all the groups hitherto described in having their flowers arranged in separate roundish capitula upon a fleshy rachis springing from the tuberous-stock. They, again, belong to Central America, and are divided into three genera (Lophophytum, Omhrophytum, and Lathrophytum) into particulars of which we cannot enter without exceeding our limits. Only the genus Lophophytum, which is in many respects different from other Balanophorese, and in particular has been more thoroughly studied with reference to its peculiar mode of connection with the host-plant, demands special consideration. The Lophophytum mirabile (see fig. 41 ^ ) found in the primeval forests of Brazil adhering to the roots of Mimoseae, to those of Inga-trees especially, occurs at some places in such profusion that areas of ground, occupied by Inga-roots, from twenty to thirty paces in circumference appear to be entirely overgrown by the parasite. Hundreds of tubers, some large, some small, rest upon the roots of the trees, covered by fallen leaves and a light Vol. I. 13 194 BROOM -RAPES, BALANOPHORE^, RAFFLESIACE^. stratum of vegetable mould. Most of them are the size of a fist, but a few are as big as a head, and then weigh 15 kilogr. and more. The tubercles formed directly by the germinating seeds which chance upon the roots are, by the time they attain to about the size of a pea, already in connection with the wood of the attacked root The cortex and a portion of the wood at the place where the parasite is adnate are absorbed by this root. The tissue of the small tuber-stock is squarely and firmly inserted into the superficial notch thus made in the root, and short, peg- shaped bundles, isolated by the loosening of the wood of the nutrient root, appear to grow into the substance of the parasite. As the tuber increases in size vascular bundles are developed in it also, and these grow towards the said bundles of the host and unite with them. No boundary can then any longer be certainly recognized between host and parasite, and the strangest fact of all is that we find, in these bundles, cells concerning which we are not able to decide, even by reference to their shape, whether they belong to the one or to the other. The cells which belong undoubtedly to the wood of the nutrient root have dotted walls; the bundles unquestionably developed in the parasitic tuber exhibit, on the other hand, cells with reticulate thickening, which, when slightly magnified, look as if they were transversely striated. Wherever these pitted and reticulate cells meet, cells are intercalated which do not altogether correspond either to the pitted variety belonging to the host or to the reticulate cells of the parasite, but display a form intermediate between the tw^o. Here and there, too, cell-groups belonging to the parasite are entirely buried in the wood of the foster-root in its growth, and in the older tubers the cellular elements of the two plants there bound together are so involved that it is, as has been stated, impossible to establish any line of demarca- tion between the two. By the time the tubers have reached the size of a fist their cortical layer is always solid, corkj?-, and areolated; each of the areas being more or less uniformly angled, as is shown in the illustration below. Some of the more protuberant portions elongate and grow out into short, thick stumps bearing scales all round, each of the little areas having a triangular-pointed scale situated in the middle of it. At this stage of development the entire Lophophytum plant has an extraordinary resemblance to the squamigerous rhizome of a fern, or to a dwarf cycad-tree, stripped of its green leaves; and this likeness is enhanced by the fact that the bark and scales of Lophophytum are dark-brown in colour. From the centre of each of these thick stumps, which often reach a height of 15 cm., there now arises a spadiciform inflorescence. At first it is so thickly covered with ovate lanceolate scales possessing dark-brown, quasi-horny tips, overlapping one another like tiles, that the spadix as a whole looks extremely like an erect cycad-cone. Imagine the surprise of a traveller, who chances upon a spot in the depths of a primeval forest where the ground is occupied by Lophophytum, upon seeing hundi-eds of these brown, scaly cones grow up suddenly, in the course of a night following some days of rain, from the subterranean roots of the trees. A day or two later, this gardea BROOM-RAPES, BALANOPHORE^, RAFFLESIACE^. 195 of Lophophyta presents an altogether different picture. The brown scales have detached themselves from the rachis, first those at the base of the cone, then also those on the upper parts. They fall off almost simultaneously, and with them the envelope which up to that time has concealed the flowers. The erect, fleshy, white, or reddish rachis bearing the flowers then becomes visible. The female flowers are Fig. 41. — Parasitic lialanophoreai. Lophophytum mirabile, from Brazil. 2 Sareophyte sanguinea, from the Cape of Good Hope. on the lower part, and arranged in spherical, deep yellow or orange-coloured capitula which are packed close together; the male flowers are situated above the lowermost third of the spadix, and are arranged in looser and less crowded capitula of a pale yellow colour. However striking the phenomenon presented by these flowering cones of Lophophytum mirabile, it is surpassed by another native of Brazilian forests, the Lophophytum Leandri. The colouring of the inflorescence in this species cannot 196 BROOM-RAPES, BALANOPHORE^, RAFFLESIACE^. be exceeded in vax-iety, its rachis being pale reddish-violet, the bract-scales gamboge, the ovaries yellowish, the styles red, and the stigmas white. It is not surprising that even in Brazil, where there is certainly no lack of curious plant- forms, they have attracted attention, and that they are used there, as is the case with all rare plants, for purposes of healing and magic. The tubers of Lopho- phytuTYi mirdbile, which have a disagreeable, bitter, resinous taste, and bear the popular name of " Fel de terra ", or earth-gall, are employed by quacks against jaundice, and a belief also prevails that by secretly eating the blossoms youths are enabled to win the aifection of the maidens they admire. The same may be said of Lophophytum Leandri, and, in addition, there is a tradition that the eating of it brings luck and agility in hunting, fishing, fighting, and dancing, and for this reason the Indian youth collect the plants secretly and eat them on particular days. Of the other parasitic Balanophoreae most nearly allied to LophopLytum we will hei-e only mention in passing the species of Ombrophytum, known in Peru by the name of "Mays del monte", which has a yellowish inflorescence over 30 cm. high, and from 6 to 7 cm thick, somewhat resembling a spike of maize, and lastly, the Lathrophytum Peckoltii of Brazil, to which a special interest attaches inasmuch as it is the sole instance of a flowering plant entirely destitute of all structures of the nature of leaves, with the exception of the stamens and ovaries. Langsdorffta, Scyhalium, LopJioplLytuni, and even Balanophora, Helosis, and Rhopalocnemis exhibit scales, which, though transformed in various ways, are yet always in point of position and form recognizable as leaves; but neither on the tuber, shaft, nor spadix of this Lathrophytum is any trace of a scale to be seen, nor even a swelling or rim that might be looked upon as a degenerate leaf. In comparison with equatorial America with its wealth of parasitic Balano- phoreae the corresponding zone of Africa must be called poor so far as these plants are concerned. Possibly further explorations may bring to light a few more of these wonderful vegetable parasites, but it is hardly to be expected that such a variety as is presented in Brazil, the Peruvian Andes, New Granada, and Bolivia will be found Only three Balanophoreae have been discovered in the Cape regions, where the flora is well known. One of these, which is represented on the right- hand side of fig, 41, bears the name of Sarcophyte sanguinea (i.e. blood-red flesh- plant), whilst the name of Icthyosoma (i.e. fish-carcase) has also been applied to it because it smells of rotten fish. These names imply that the plant resembles an animal rather than a vegetable organism. The host -plants adapted to this Sarcophyte are various Mimosese, especially Acacia caffra, Acacia capensis, &c. In the first place, as is the case with all Balanophoreae, small tubers are formed on the roots of the above-mentioned woody hosts, and enter into connection with the wood of the nutrient roots in the manner already described more than once. An inflorescence then emerges from a bud originating beneath the cortex of the tuber, and rapidly grows up from out of the cortex, which is rent and pushed up m the process. The axis of this inflorescence resolves itself into a number of thick, repeatedly ramifying, fleshy branches, diflering in this respect from every other Fiff. iZ~Ci/(uius Hyp, "■"""° "'";^;' *~~*—.v„„ .„ „,„„,, 198 BROOM-RAPES, BALANOPHOREvE, RAFFLESIACE^. example of the Balanophoreae. The flowers are arranged side by side on the branches, staminate flowers on one plant, and pistillate flowers on another, the latter always grouped in spherical capitula, as is shown in fig. 41 ^. Reddish- brown scale-like leaves are situated at the points of origin of the branches, and also at the base of the entire inflorescence. The general aspect is that of a bunch of verrucose grapes ascending from the root, or of the fruiting axis of Ricinus, and is very striking owing to the blood-red colouring of all the parts. As a final instance of the Balanophoreae we may take the genus Cynomoriumiiy which was so highly valued in olden times, and is the sole species belonging to this family of plants indigenous in the south of Europe. A drawing of it is given on the right-hand side of fig. 42. Whilst other Balanophoreae are parasitic on the roots of trees and lianes in the shade of lofty woods, this Cynomorium thrives most luxuriantly upon plants near the sea -coast, on the roots of Pistacias and Myrtles, and even on actual salt-loving maritime plants, the various Tamarisks, Salicorniae, Salsolaceae, and Oraches, which d-re sprinkled with foam whenever the breakers are high. The seed is like that of other Balanophorege and those of the Orobanche species, and germinates in the same way as they do. From the group of cells in the seed which represent the embryo, a filiform body emerges, and then grows downwards, its upper part remaining for some time in connection with the other cells in the seed, which are richly furnished with food -materials. The filiform embryo continues to grow deeper and deeper at the expense of this nutritive store, and as soon as it reaches a living root, swells into an oval or irregularly-lobed tubercle, which unites with the wood of the nutrient root in the manner already described. These tubercles swell, and from the summit of each a spadix is produced, as in Lophophytum, which is raised above the surface of the earth. The spadix is clothed with pointed scales, and is clearly differentiated into a lower stalk-like support, and a fleshy inflorescence resembling a cone. The small scales are separated from one another by the process of elongation of the spadix, and some fall off". Others of them, situated about the middle of the inflorescence, persist, however, until the time when the entire spadix dries up. The whole of the structure standing above the ground has a blood-red colour, and when it is injured a red fluid exudes, which was at one time supposed to be blood. At an age when the peculiar pro- perties of extraordinary plants were looked upon as an indication given by higher powers that they were to be used for curative purposes, it was believed that the spadices of Gynomorium, being blood-red in colour, and bleeding when wounded, had styptic properties. In those days they were even collected for the sake of this property, and sold in apothecaries' shops under the name of the Maltese fungus {Fungus melitensis). Various miraculous virtues were also attributed to this plant, and the demand for it was so great that it became a regular article of commerce, its main source being the Island of Malta, whence is derived the name above referred to. Of the Hydnorese, which are most properly included in the same series as BROOM-RAPES, BALANOPHORE^, RAFFLESIACE.E. 199 Balanophorese in consideration of their coalescence with the roots of their hosts, only three species are known. Two of them (Rydnora Africana and H. triceps) belong to South Africa, the third (Hydnora Americana = Prosopanche Burmeisteri) to South Brazil. The tuber is represented by a prismatic body with from four to six angles furnished with papillse along the edges. The flower -buds which burst from it have at first the form of spherical Gasteromycetes, but gradually elongate and assume the form of a large fig or upright club. This structure opens at the thickened upper extremity by three stout fleshy valves representing petals. At the base of this curious flower no appendage is to be seen that could be interpreted as a bract or leaf. The fleshy mass of flowers evolves a disagreeable putrid odour, and in this property the Hydnorese resemble the Rafllesias, which belong to the next group of parasitic Phanerogams. The fifth series of flowering parasites is composed of the Rafiiesiacese, plants connected with Balanophorese and Hydnorese by their general aspect, the absence of chlorophyll, and the undifferentiated embryo which consists merely of a group of cells. They used all to be classed together under the name of Rhizanthese; but the Rafflesiacese are now treated as a separate family on account of the characteristic structure of their flowers and fruit. The formation of these organs will again come up for discussion later on when we treat of the wonderful structure of the famous giant-flower Rajfflesia; at present we are only concerned with the relationship of the parasite to the food-providing host-plant. This is, if possible, even more remarkable than in the case of Balanophorese and Hydnorese. In the latter the union is effected within a structure like a tuber or a rhizome, the vessels and cells of the parasite coalescing with the exfoliated and disordered wood-cells belonging to the root or stem of the host-plant; whereas in Rafflesiacese the embryo, having penetrated beneath the cortex of the host, produces a more or less definite hollow cylinder which surrounds the wood of the host's root or stem (as the case may be), and constitutes a sort of vestment intercalated between the wood and the cortex of the host. There is no production of tuberous enlargements as in the Balanophorese. The stem or root attacked by the parasite only exhibits a moderate thickening at the place where the parasite dwells beneath the cortex, and the cortex itself is only destroyed at the spot where the embryo pierces through it, and where subsequently the flowers emerge. When roots constitute the substratum whereupon the parasite has established itself, they are always of a kind that run throughout upon the surface of the ground; when stems are chosen for attack, they are either the branches of trees or shrubs, shoots clothed with dead foliage belonging to dwarf suffruticose bushes, or else woody lianes of tropical forests. The seeds are con- veyed to the host-plants through the intervention of animals. Rafflesias are found in the haunts of elephants and along the tracks followed by those beasts. The Raffiesia-fruits are accordingly no doubt trampled upon and crushed, and the little seeds imbedded in the pulpy mass of the fruit thus have an opportunity of adhering to the elephants' feet. The seeds are afterwards rubbed oft by projecting roots at places more or less remote from the original locality, and if 200 BROOM-RAPES, BALANOPHOREiE, RAFFLESIACE^. the root upon which they are detained belongs to a Cissus plant, they germinate. On the other hand, such Rafflesiacese as occur on the woody branches of trees, shrubs, and undergrowths, or on lianes, develop succulent fruits, which are eaten by animals. Their seeds are protected by a horny coat, and preserve their power of germination unimpaired as they pass through the animals' alimentary canals and are deposited with the excrements on the stems of fresh host-plants; or the seeds may stick to some part of an animal that happens to rub against them, and be brushed off later on as being an uncomfortable appendage, and in this way also they may fall upon the stem of a host-plant. Those Rafflesiacese which occur in Venezuela on the woody lianes (Caulotretus), known by the name of "monkey -ladders", owe their dispersion for the most part probably to monkeys. Now, if a seed has been deposited in one way or another upon a woody root, creeping along the surface of the ground, or upon the stem of a woody plant, the filiform embryo emerging from the seed finds a suitable nutrient substratum present and it pierces the cortex of the root, and develops beneath it a tissue, which incloses the wood like a sheath. In Rafflesia and in the Pilostyles parasitic on the sufiruticose shrubs of Tragacanth (P. Haussknechtii, see fig. 43 ^ ), this tissue consists of rows of cells, which to the naked eye look like threads. Some are simple and greatly elongated, others branched, and they are united together to form a net-work, so closely resembling the mycelium of a fungus as to be readily mistaken for one. The most complete similarity to these vegetative bodies living beneath the cortex of a host-plant is exhibited by the mycelia of the toad-stools which spread themselves in the form of nets and webs between the wood and the cortex of old trunks of trees. The vegetative bodies of the other species of Pilostyles consist, in each case, of a tissue composed of many layers of cells forming a parenchyma imbedded between wood and cortex in the host-plant and including some vessels and rows of cells capable of being interpreted as vascular bundles. Only in rare instances does this tissue of the parasite form an unbroken hollow cylinder encompassing the wood of the host; usually the elements of the host's tissues penetrate into it and permeate and split up the cylindrical soma (vegetative body) in the form of bands, ribs, and fibres. Many elements of the tissues, which the imbedded parasite has displaced from the living wood, and carries, as it were, on its back, perish; but sometimes these discarded layers remain in connection with other living tissues and so preserve their own vitality and power of expansion, and develop layers of wood- cells covering the parasite. There is then a general confusion and entanglement, and it is difficult to say what part belongs to the parasite and what to the host. When the somatic tissue of the parasite has accomplished its connections with the host-plant in the manner just described, the latter is unable to rid itself of its occupant. A portion of the juices of the host-plant passes into the parasite's cells and the unwelcome guest augments in volume, and endeavours forthwith to reproduce and distribute its kind by the formation of fruit and seeds. For this purpose buds are developed at suitable spots in the reticular body of the parasite, each of which is manifested as a parenchyma of pulvinate appearance, and is BROOM-RAPES, BALANOPHORE^, RAFFLESIACE^. 201 termed a floral cushion. The cells in this cushion, however, now group them- selves in a definite way; ducts and vessels are produced, and, at the same time, a differentiation into axis and flowers is exhibited. These members continue their development, increase in size, and finally the enlarged bud breaks through the cortex of the host-plant under shelter of which it has been evolved. In the genus Gytinus alone do we find a stem richly furnished with leaves and bearing at the top a flattened symmetrical tuft of flowers (see fig. 42, left-hand side) developed from this bud; in the rest of the RafilesiacefB, the bud, which has Fig. 43.— Rafflesiacese parasitic on trunks and branches. I Pilostyles Haussknechtii. 2 Apodanthes Flacourtiana. » pUostyles Caulotreti. emerged from beneath the cortex of the host, is the flower-bud itself. The axis supporting the bud is extremely abbreviated and clothed merely by a few scales, and the flowers are sessile directly upon the root or stem of the host (see fig. 43). In the case of roots creeping upon the ground, the buds always emerge only on the side turned towards the light; on lianes, also, they are only formed on the side more exposed to light where subsequently the opened flowers are easily accessible to flying insects (see fig. 43^); on upright shrubs and under-shrubs, on the other hand, they burst forth on all sides upon the branches. Branches of this kind bearing ubiquitously extruded flowers of a parasite such as Apodanthes Flacourtiana (see fig. 432) look delusively like the Mezereon {Daphne Mezereum) when the latter is in bloom in the early spring before the development of foliage-leaves, its woody branches being similarly studded all round with flowers, which stand out horizontally 202 BROOM-RAPES, BALANOPHORE.i:, RAFFLESIACE^. from them; but, in the one case the flowers belong to a foreign parasite living under the cortex and have broken through it, whereas in Mezereon it is the flowers of the plant itself that have unfolded. In the case of Pilostyles Haussknechtii, which is parasitic on the low bushy tragacanth shrubs of the Persian plateaus, the buds are formed regularly on both sides of the leaf -bases of the host, so that at the insertion of every one of the older foliage-leaves, one finds a pair of buds, which subsequently expand into flowers (see fig. 43 ^). Fig. 44.— Parasitic Kafflesiacea {Brugmamia Zipellii) upon a Cia:,ua Throughout the species of Aj^odanthes and Pilostyles the flowers are small — about the size of elder, jasmine, or winter-green blossoms — and by no means conspicuous. But this is not the case in the genera Brugmansia and Bafflesia. The Brugmansias, indigenous to Borneo and Java, have very handsome flowers, as may be seen in the above drawing, which represents on the natural scale Brugmansia Zipellii parasitic upon the root of a Cissus. But in magnitude they are far surpassed by the flowers of the Rafflesise, one of which, viz.: Baffiesia Arnoldii, may be described as actually the largest flower in the world. When open it has a diameter of 1 meter, a dimension exceeding even that of the gigantic blooms of South American aristolochias. At the period of emergence of the buds of Bajfflesia Arnoldii from the roots of the vines which serve them as hosts, they BROOM-RAPES, BALANOPHORE^, RAFFLESIACEiE. 203 are only as large as a walnut and give scarcely any indication of their future inao-nitude; but they gradually increase in size, and before opening are curiously like a cabbage. Up to this time the bracts still inclose the flower proper, and !to them is due the above-mentioned resemblance. They now open back, and the flower, which, to the last, grows rapidly, unfolds and displays five immense lobes around a central bowl or cup-shaped portion. The form of the giant-flower when open is best likened to that of a forget-me-not blossom. The semicircular joutline of the lobes, at least, is similar, and the very short throat of the flower also jexhibits a distant resemblance. At the part where the bowl-shaped centre, which Fig. id.—Mafflesia Padina, parasitic on roots upon the surface of the ground. has the stamens and styles inserted in it, passes into the lobes there is a thick, fleshy ring like a corona. The upper surface of the lobes is covered with numbers of papillae. The lobes themselves, the hollow central bowl, and the ring, are all flesh}^, and the flower, as a whole, emits an unpleasant putrescent smell. This floral prodigy was first discovered in the year 1818 in the interior of Sumatra at Pulo Lebbas on the river Manna, where it occurs parasitic on the roots of wild vines in places where the ground is strewn with the dung of elephants. It has never yet been seen anywhere outside Sumatra. Four other Rafflesise have, jhowever, been discovered, but all in the islands of the Indian Ocean — Java, Borneo, and the Philippines. In mode of growth, as also in the form of the flowers, they resemble the species above described, but their flowers are rather smaller. Rafflesia Padma, which occurs in Java, and is represented in fig. 45, possesses flowers with a diameter of half a meter. The hollow, somewhat ventricose centre and the ring bordering the floral receptacle are in this Rafflesia of a dirty 204 MISTLETOES AND LORANTHUSES. blood-red, whilst the verrucose lobes have almost the colour of the human skin. The flowers are sessile upon roots which wind about upon the dark forest ground, and a cadaverous smell, anything but pleasant, issues from them. All these peculiarities explain the uncanny impression made by the organisms in question upon their original discoverers and upon all subsequent observers. Whilst the Rafflesiae, as well as the genera Brugma^isia and Sapi'ia, belong to the tropical and sub-tropical regions of Asia, and to the world of islands adjacent thereto on the south side, the genus ApodantJtes is confined to tropical America. Most of the species of Pilostyles also appertain to tropical America, especially to Brazil, Chili, Venezuela, and New Granada. One species alone — Pilostyles JEthio- \ pica — has been observed in the mountains of Angola, and another, as has been mentioned before, in Persia. The only European representative of the remarkable group of Rafflesiacese is Gytimis Hypocistus, represented on the left side of fig. 42, but its distribution is coincident with the entire range of the Mediterranean flora. The roots of cistus shrubs, plants which are characteristic of the vegetation belonging to the basin of I the Mediterranean, constitute the nutrient substratum in the case of Cytinus. It is ' especially where the layer of earth-mould is not deep, and consequently the roots of the shrubs in question are exposed, that Cytinus is met with growing in abundance amongst the under- wood of the cistus plants. The squamous leaves clothing the stem of this parasite being scarlet, and the plants not solitary but in large numbers, one sees here and there a flaming red colour glowing in the gaps in the cistus- groves, and one is thus from far oflf made aware of the presence of the parasite. The flowers themselves, which open between the red scale-like bracts, are yellow. The combination of colour thus afforded is a rare phenomenon in the vegetable ! world, and gives a very strange appearance to the plant. Besides the species of - Cytinus distributed over the area of the Mediterranean flora, there are two other \ species in Mexico, and one also at the Cape, which, although not parasitic on Cistus { shrubs but on other woody plants, especially Eriocephalus, yet do not difler from | Cytinus Hypocistus in floral structure or in mode of connection with their host. ' MISTLETOES AND LORANTHUSES. The sixth and last series of parasitic phanerogams includes epiphytes of bushy appearance with much bifurcated branches, green cortex, green leaves, and berries containing large seeds, which germinate whilst resting immediately upon the branches of such trees as are adapted to act as host-plants, and will surrender to the invader a portion of their nutriment. To this series belong a dozen different species of the genus Henslowia, belonging to the family of Santalacese, and indigenous to the South of Asia — chiefly the East Indian Archipelago — and, in addition, upwards of 300 species included in the family Loranthaceae. Amongst the latter, the plant that is best known and most widely distributed is the Euro- pean Mistletoe (Viscum album) represented in fig. 46, and as it is also fitted, in MISTLETOES AND LORANTHUSES. 205 respect of its life-history, to serve as type of the entire series, we will describe it first of all. As is well known, the Mistletoe is parasitic upon trees, and these may be either Angiosperms or Gymnosperms. Most frequently it establishes itself upon trees the branches of which are coated by a soft sappy cortex — an extremely delicate and tender cork-tissue in particular — as is the case with silver-firs, apple-trees, and poplars. The Mistletoe's favourite tree is certainly the Black Poplar {Populus I nigra). It flourishes with astonishing luxuriance on the branches of that tree, and wherever there is a small plantation of Black Poplars, the Mistletoe takes up its abode. Along the shores of the Baltic and by the Danube near Vienna — especially' in the celebrated Prater from which fig. 47 is taken, one finds, on many of the Black Poplars, tufts of Mistletoe measuring 4 meters in circumference, and with axes of a thickness of 5 cm. Birds use their most crowded branches, by preference, to nest in. In the forests of Karst, in Carniola, and in the Black Forest, where poplar trees play merely a subordinate part, whilst on the other hand, quantities of silver firs shade the ground, large numbers of these conifers have their tops covered with Mistletoe; and in the Rhine districts and the valley of the Inn in Tyrol, the same parasite occurs as a troublesome visitor upon apple-trees in the neighbourhood of the peasants' farms. In localities destitute of these three kinds of trees, which are pre-eminently the Mistletoe's favourite host-plants, it puts up with other trees, and is then usually found on whatever species happens to be the most common in each particular country. Thus, in the Black Pine district of the Wiener Wald, it occurs upon the Corsican Pine, whilst on the heaths of the sandy lowlands of the March, it settles upon the Scotch Pine. Much less frequently j it has been observed on walnut-trees, limes, elms, robinias, willows, ashes, white- thorns, pear-trees, medlars, damsons, almond-trees, and on the various species of Sorbus. Mistletoe has also been found by way of exception upon the oak and the maple, and upon old vines. On one occasion, in the district of Verona, it has been seen established upon the parasitic shrubs of Loranthus Eitropceus, that is to say, one member of the Loranthaceae was found parasitic upon another. The birch, the beech, and the plane, are avoided by the Mistletoe, a fact which no doubt depends upon the special structure of the cortex in those trees. The dissemination of the European Mistletoe is effected, as in all the other i Loranthaceee, through the agency of birds — thrushes in particular — which feed upon the berries and deposit the undigested seeds with their excrement upon the branches of trees. That a preliminary passage through the alimentary canal of birds is essential to the germination of these seeds is no doubt a delusion, this assumption of former times being easily refuted by the fact that one can readily induce the seeds of berries, taken fresh from a tree, and stuck into fissures in the bark of moderately suitable trees, to germinate; it is, however, true, that in nature, mistletoe-seeds are dispersed exclusively by birds in the manner above mentioned. To this method of dissemination must be attributed the phenomenon, which, at first 206 MISTLETOES AND LORANTHUSES. sight, is surprising, that Mistletoe-plants are rarely seated upon the upper surface of branches, but very frequently on the sides. For the dung of thrushes, which live upon Mistletoe-berries, is in the form of a semi-fluid, highly viscid mass, ductile like bird-lime; and, even when it is deposited upon the upper surface of slantiuo- branches, it immediately runs down the sides, sometimes extending in ropes 20 or 30 centimeters in length. Owing to the viscous mass thus following the law of gravity, the Mistletoe-seeds imbedded in it are conveyed to the sides, and even to the under surface of the bark, and there remain cemented. Fig. 4G.— The Jiuropeau Mistletoe (Viscum album). It may be a long time before a seed of the kind germinates, especially if it does not become attached until the autumn. The embryo is completely surrounded in the seed by reserve food. It is club-shaped and comparatively large, and is dis- tinguished by the fact that the two oblong cotyledons, which are closely pressed together, but often somewhat wavy at the margins, are coloured dark green by chlorophyll, like the environing cellular mass filled with reserve materials. In the process of germination the axis of the embryo, especially the part lying beneath the cotyledons, and passing into the hemispherical radicle, lengthens out; the white seed-coat is pierced, and the radicle makes its appearance through the breach. Under all circumstances the emergent radicle is directed towards the bark of the branch to which the seed is adherent. This is the case even when the seed chances MISTLETOES AND LORANTHUSES. 207 to stick with the radicle of the seedling pointing away trom the bi-anch; the whole axis of the embryo curving towards the surface of the bark in a very striking manner. Thus the radicle always reaches the bark, and having done so it becomes adpressed and cemented to its surface, spreads itself out in the form of a doughy i mass, and so develops into a regular attachment-disc. From its centre a slender pro- cess now grows into the bark of the host-plant, piercing the latter and penetrating as far as the wood, but not growing into that tissue. This penetrating process has been termed a " sinker ", and must be looked upon as a specially modified root. w \i'- ¥1 XM^l^^ '^H. Fig. 47. — Bushes of Mistletoe upon tlie Black Poplar iu winter. The development of the first year ends with the formation of this sinker. When the winter is over, the branch, into which the sinker is inserted so as just to reach the wood with its point, grows in thickness, a new layer of wood-cells — a so-called annual ring — being superimposed upon the wood of the previous year. The increasing mass of wood first surrounds the tip of the sinker with wood-cells, then forms a rampart all round it, pushing the cortical tissue, wherein that organ has hitherto been wedged, in front of it in an outward direction, and in this way the sinker is at length fixed deep within the woody cylinder. The process of inclosure by the wood-layers, as they are built up, may be compared to the gradual surrounding of a stake on the sea-shore by the rising tide; the lowermost extremity is first immersed and then higher and higher parts until the whole is enveloped. The 208 MISTLETOES AND LORANTHUSES. sinker itself remains, strictly speaking, stationary ; it does not grow into the wood, but the wood overgrows it. But what happens in the following season when a fresh annual ring is once more added to the wood? If the sinker had entirely ceased growing it would of necessity be ultimately completely closed by the layers of wood, as they develop with ever-increasing energy and add to the thickness of the branch, and at last it would be quite buried. To prevent this result, which would be fatal to the Mistletoe, a zone of cells is provided near the base of the sinker, which zone, at the time when the rampart of wood is being raised, adds in an equal degree to its own height, and causes, of course, an elongation of the sinker in a peripheral direction. The length of the piece thus intercalated in the haus- torium is exactly equal to the thickness of the corresponding annual ring in the surrounding wood of the branch. Thus at length the Mistletoe-sinker is found imbedded in a number of annual rings, although it has not grown into the latter, but has been banked up by them year by year. That zone of the sinker which possesses the capacity for growth, and which is always to be sought, in accordance with what has been said above, at the outside limit of the wood of the branch, in the so-called " bast " layer situated on the inner face of the cortex, produces, in the second year after the adhesion of the Mistletoe- embryo, lateral ramifications which are called cortical roots. They are thick, cylindrical, or somewhat compressed filaments, and all run close together under the cortex in the bast layer of the invaded branch. These rootlets issuing from the sinkers pursue a course parallel to the longitudinal axis of the branch, whilst the sinkers themselves are at right angles to the axis (see fig. 48 ^). If a rootlet springs from the sinker in a direction transverse to the longitudinal axis it bends imme- diately afterwards so as to be parallel to the long axis, and adopts the same direction as the rest, or else it bifurcates just above its place of origin into two branches which separate suddenly, and in their further course follow the axis of the branch. Thus it comes to pass that all the rootlets of a Mistletoe run up and down in the infested branch of the host-plant in the form of thick green parallel strands, but that none of them ever encircle the branch in the form of an annulai coil. Each of these cortical roots may now develop from behind the growing-point new sinkers, which are formed in the same way as the first one above described as proceeding from the actual seedling. They, too, penetrate into the branch per- pendicularly to the axis, and as far as the solid wood are then encompassed by the growing mass of wood, but maintain the power of growth in the part close to their insertions, and in their growth keep pace with the thickening of the wood of the branch. The fact of the yearly recurrence of this formation of sinkers explains how it is that those situated nearest the growing-points of the cortical roots are the shortest, they being the youngest, whilst those which arise near the first sinker are the longest and oldest. It also accounts for the former being only inclosed by one annual ring of the host's wood, and the others being surrounded by an increasing number of rings the nearer they are to the spot where the Mistletoe-plant first struck root. MISTLETOES AND LORANTHUSES. 209 The root-system of the Mistletoe taken as a whole may be described as like a jaw-bone in shape, or, still better, a rake. The cross-beam of the rake corresponds to the cortical root, whilst the teeth are analogous to the sinkers; the cross-piece must be supposed to be parallel to the axis of the branch and lying under the bark, and the spokes must be thought of as perpendicular to the axis and driven into the wood. Whilst the roots of the Mistletoe-plant are spreading in the interior of the branch in the manner described, the stem is developed outside. At the time when the process, subsequently to be the first sinker, emerges from the attachment-disc of Fig. 48.— 1 Lomnthus Europceus, and 3 Mistletoe ( Viscum album)— hoi\\ parasitic ou brandies of trees, and seen In section. 2 A piece of the wood of a Fir-tree perforated by the sinkers of a Mistletoe. the embryo and pierces through the bark, the cotyledons are still covered by the white seed-coat, which rests upon them like a cap. But when once this first sinker is firmly fixed and in a position to take up nutritive juices from the wood of the host, the seed-coat is thrown off"; the apex of the stem, which is still very short, is raised; the cotyledons are detached, whilst close above them is produced a pair of green leaves. Thenceforward the development of the visible portion of the Mistletoe- plant outside the bark keeps pace with that of the roots underneath the cortex, and is moreover dependent upon the quantity of food taken up by the sinkers from the wood. Where there is an abundant supply of nutriment, as in the case of poplars, the growth of the Mistletoe is correspondingly exuberant; where the flow of juices is scarce, the parasite is stunted in its growth, and often develops only small yellowish sickly-looking tufts. If the foster-plant is of a lavish nature, adven- VOL. I. 1* 210 MISTLETOES AND LORANTHUSES. titious buds are produced regularly by the cortical roots to which the absorbed nutriment is first of all conveyed from the sinkers. These buds occur on the side of the rootlets nearest the exterior of the bark, and later they burst through the rind, and develop into new Mistletoe-plants. These outgrowths are analogous to the adventitious shoots produced from the subterranean roots of the Aspen, and this comparison is rendered all the more appropriate by the fact that the removal of the tuft of Mistletoe encourages the sprouting of adventitious root-buds just as in the case of the Aspen, the growth of s] loots from the roots is promoted by the felling of the trees to which those roots belong. If a large Mistletoe-bush, growing in solitude on a Black Poplar, is removed from the tree with the intention of freeing the latter from its parasite, the hopes entertained by the operator are disappointed; for, an outgrowth of shoots from the cortical roots ensues at a number of different spots, and in a few years' time the poplar in question is the prey of a dozen Mistletoe-bushes instead of one. Inasmuch as these bushes, produced from offshoots, are able, under favourable conditions, to send out fresh roots, and these again may develop shoots, a good host of the kind will at last have all its boughs from top to bottom overgrown by Mistletoes. In the Prater at Vienna there are poplars beset by at least thirty large Mistletoe- shrubs, and double that number of small ones, and if one catches sight of such a tree at some distance in winter-time when the branches have lost their leaves, one takes it to be a Mistletoe-tree, for almost the entire system of branches is mantled in a continuous tangle of evergreen bushes of Mistletoe, which are in a state of parasitism upon it. Sinkers of the Mistletoe, 10 cm. in length, and inclosed in forty annual rings, have been found in the wood of the Silver Fir, whence we may conclude that the Mistletoe may live for forty years. A greater age could scarcely be attained by one and the same bush of the parasite. If the Mistletoe dies, the rootlets and haustoria survive for a time, but at length moulder and fall to pieces, whilst the wood in which they were imbedded remains unaltered. The affected parts of the wood exhibit in that case numerous perforations, and look just like the wood of a target which has been fired at and struck by shot or small bullets (see fig. 48 ^). .A small plant belonging to the Loranthaceae and named Juniper-Mistletoe (Vis-' cum Oxycedri or Arceuihobium Oxycedri) occurs on the red-berried juniper bushes (Juniperus Oxycedrus) of the Mediterranean flora. It is very different from the common European Mistletoe, as is obvious at first sight, its foliage-leaves being reduced to little scales, which gives a characteristic jointed appearance to the rami- fications. A whole series of leafless forms allied to this species is found to exist in India, Japan, Java, Bourbon, Mexico, Brazil, and at the Cape. They are nearly all small bushes which project from the boughs of host-plants and sometimes clothe the latter so thickly that the boughs in question serving as nutrient substratum are entirely enshrouded by the parasitic growth. The Juniper-Mistletoe is only from 3 cm. to 5 cm. tall, and the branchlets are not woody, but soft and herbaceous; the fruits are blue oblong berries, almost destitute of succulence. The latter are MISTLETOES AND LORANTHUSES. 211 dispersed by birds like the berries of the common Mistletoe, and the way in which the pcirasite settles upon and clings to branches of the host-plant is the same as in that species. It also develops sinkers and cortical roots, but these root-structures are not by any means so regularly arranged as in ViscuTYi album, but form an inex- tricable web of strands and filaments pervading the internal layers of cortex, and resolving itself into finer and finer groups of cells, which end by looking not unlike a mycelium, and also remind one of the suction-apparatus possessed by Rafflesiacese. Such of these strands and cellular filaments as are imbedded in tlie wood of the juniper do undoubtedly play the part of suction-organs. They are present in large numbers, and some of them are occasionally encompassed by several annual rings. They possess no special zone of growth. The elongation necessary to prevent their being enveloped and overwhelmed by the wood, as it adds to its thickness, is effected by the division of individual cells and groups of cells. The outgrowth of shoots from the root is much more exuberant than in the common Mistletoe; but the death of the original plant takes place much earlier, and close to yellowish-green bushes of various degrees of smallness, one finds very regularly dead or dying shrublets already turned brown, all growing promiscuously over the somewhat swollen branches of the red-berried Juniper. The behaviour of Loranthus Europceus, which is parasitic on oaks and chestnuts in the east and south of Europe, is altogether unique. The mode of its attack upon the branches of oaks is, it is true, similar to that of the two other Loranthaceae just described. The yellow berries, which are grouped in graceful biseriate racemes, are eaten with avidity by thrushes in the autumn and winter, and the undigested seeds are deposited with the dung of those birds upon the branches of trees. The embryo, on emerofino: from the seed, bends towards the bark and sticks to it, at the bottom of little rifts and crevices, for the most part, by means of the radicle, which becomes an attachment-disc. A process now arises from the centre of the attachment-disc, and pierces through all the cortical layers of the oak-branch as far as to the zone of young wood, just as if it were a small nail driven in. This process increases in thickness at the expense of the nutriment it withdraws from the young wood, and from it are developed one, two, or three branches, which, however, invariably run downwards beneath the bark, that is to say, in the direction opposed to that of the stream of sap ascending in the oak-wood, and never produce the sinkers so characteristic of the Mistletoe. Each of these roots is shaped like a wedge, even from the rudimentary stage, and acts, too, in the manner of a wedge, penetrating between the yet soft and delicate cells of the cambium, which were formed in the spring at the periphery of the solid older wood of the previous year, and were destined to constitute a new annual ring, splitting and tearing in the process that cell-tissue. Such of these tender cells as lie outside the wedge die, those situated within become lignified and altered into solid wood, to which the wedge- shaped root firmly adheres. Beneath the apex of the wedge, the lignification of cambium cells naturally extends much further towards the exterior, because there it is not at all broken or dead. In front of the apex of the wedge, therefore, there 212 MISTLETOES AND LORANTHUSES. is, presently, solid resisting wood. The root being no longer able to split the tissue with its point, is stopped in its growth at this spot. But there is nothing to pre- vent its continuing to grow along a course somewhat nearer the periphery, and outside the limit of the new annual ring of solid wood, where a fresh development of soft and tender cells has taken place in the cambium, and this indeed actually happens. Thus, every addition to the length of the Loranthus-root, as it grows onward between the wood and the cortex of the oak -branch, is further removed from the axis of the branch; or, in other words, the surface of contact between root and wood has the conformation of a flight of stairs, of which the lowest step constitutes the base, and the uppermost the apex of the root (see fig. 48^). These steps are very small, their height varying from about 5 mm. to 7 mm., but they may be distinguished quite clearly in longitudinal sections, on account of the darker colour of these roots contrasting with the lighter oak-wood. Nutritive fluids are imbibed by the Loranthus-root from the wood of the oak at the surface of contact, and it is probable that this absorption takes place especially at the notches forming the steps. The root can only elongate, naturally, during the period when there is a young and fi-agile cell-layer suj^erimposed upon the solid wood, whence it follows that in Loranthus the continuation of the root's growth is more dependent upon a particular season and upon the annual progress of development of the host than is the ease with the Mistletoe. There may be some connection between this circum- stance and the fact that the Mistletoe possesses evergreen leaves, whilst Loranthus is green only in summer, acquiring fresh green foliage in the spring in the very same week as the oak does, and casting its leaves in the autumn simultaneously with the tree it infests. The stem which issues from the embryo of a Loranthus-seed grows away from the oak-branch into the open air, and develops with great rapidity at the expense of the nutriment absorbed from the host's wood, and conveyed to it by the root above described, into a dense, dichotomouslj^-branched bush. In summer it is not unlike a Mistletoe-bush, but in autumn, when it has cast its leaves, it acquires a totally different aspect owing to the dark-brown branches and the conspicuous yellow clusters of berries. Bushes of Loranthus grow to a greater size even than those of the Mistletoe; their stems attain not infrequently a thickness of 4 cm., and clothe themselves with a blackish, rugged bark, the older stems of this kind being then usually studded by an abundance of lichens. At the spots where stems of Loranthus spring from an oak-branch they are always surrounded by a great rampart of wood belonging to the oak, and the base of the stem is often fixed in a deep symmetrically-rounded bowl reminding one vividly of the similar structures out of which the stems of Balanophorese arise. But whereas in Balanophoreae this bowl-shaped rampart appertains to the parasite, in Loranthus it is formed from the wood of the host- plant, i.e. the oak. It must, in the case we are considering, be interpreted as an exuberant growth of wood-cells and compared to the hypertrophies called galls. \ \ GRAFTING AND BUDDING. 213 which will be treated of in detail in a subsequent part of this book. On old oaks in the east of Europe these growths round the bases of Loranthus-plants sometimes reach the size of a man's head. In the case of a bush of Loranthus nearly 100 years old, from the Ernstbrunner Wald, in Lower Austria, which had reached a height of 1"2 m. and a circumference of 5'5 m., the hypertrophy in question measured 70 cm. round. It is not only the base of a bush that is overgrown by wood-cells, but the older portions of the roots described above are frequently walled in and partially inclosed by the wood of the branch as it becomes thicker. They may often be seen fixed deep in the wood, yet still preserving their freshness and vitality, and this is to be explained by the fact that they retain connection with other parts of the roots by means of isolated ledges and bridges. Indeed an adventitious shoot may develop from a piece of a root thus deeply wedged in the wood of the oak, and this shoot then grows so outwards and breaks through all the layers lying above it and originates a young bush, which pushes roots under the host's bark and afterwards behaves in exactly the same manner as a plant produced from a seed cemented to the oak-branch. The Loranthus chosen here for description (L. Eurojpoeus) has only small inconspicuous yellowish flowers; on the other hand, under the tropical sun of Africa, Asia, and, above all, Central America, the parasitic species of this genus are amongst the most splendid-flowered of plants. There are species in the tropics — e.g. Loranthus formosus, L. grandiflorus, and L. Mutisii — whose flowers attain a diameter of 10, 15, or even 20 centimeters, and are besides clothed in the most gorgeous purple and orange colours. Many Loranthi are like small trees grafted upon other trees. The host-plants of these Loranthi are principally angiospermous trees; members of the genus have also repeatedly been met with parasitic upon one another — as, for instance, Loranthus buxifolius upon L. tetrandriis in Chili. The fact has been already mentioned that the European Mistletoe has been observed near Verona parasitic upon Loranthus. It is also worth noticing, in order to complete the account of the complex relationships between parasites, that one species of Viscum has been found in India parasitic upon another, viz.: — Viscum moniliforme on V. orientate. GEAFTING AND BUDDING. Parasitism of one woody plant upon another, such as occurs in the case of Loranthacese, calls to mind certain modes of organic union between woody plants that are artificially effected by gardeners. From ancient times gardeners have performed special operations which are known as processes of "ennobling", and consist in the transference of the branch or bud of one plant on to another plant as substratum, and the inducement of organic union between the two. The plant from which the branch or bud is taken is perhaps a valuable variety of fruit-tree, or a handsome specimen of an ornamental shrub, whilst for the purpose of a substratum a robustly -growing individual belonging to a wild species of shrub or tree is selected 214 GRAFTING AND BUDDING. as a rule, and constitutes the so-called wild "stock". The branch which yields tho bud for the operation or which is itself transferred in its entirety to the wild stock is named, in the terminology of horticulture, the noble "scion". The process of ennobling is effected either by grafting or by budding. In grafting the stem of the stock is cut off transversely, an excision is made at the periphery of the surface of the section and the scion is inserted in this opening, The scion must be previously trimmed to fit; in preparing it care must be taken that it bears a pair of healthy buds, and that the end to be inserted is cut so as to correspond to the form of the fissure made in the stock. In inserting it one must see that, as far as possible, the bark, bast, and wood of the one come into contact with the corresponding parts of the other. The wounds of the stock caused by the operation are then covered by a mass of putty, wax, or some other protective medium, and the chances are that the branch thus introduced will contract an organic union with the substratum, that nutritive matter will be supplied it by the substratum, and that new branches will sprout from its buds. In this case there- fore the nutriment taken from the ground by the stock passes into the grafted scion, and the scion, which develops branches from its buds, and ultimately may become a densely ramifying tree-top, behaves as a parasite, whilst the stock playg the part of host. It not infrequently happens that a substratum supporting at its summit the branches of a grafted scion develops subsequently branches of its own lower down as well, and the curious sight is then afforded of a tree or shrub bearing different foliage, flowers, and fruit on its inferior parts from those of its upper regions. If, for example, the stem of a Quince is used as substratum, and Medlar branches are grafted upon it, the result may be a bush or tree which exhibits below branches with the round leaves, rose-coloured flowers, and golden " pomes " of the Quince, and above branches with the oblong leaves, white flowers, and brown fruit of the Medlar, Gardeners, of course, do not willingly allow this to happen, but carefully remove the branches belonging to the stock in order that all the food materials may fall to the lot of the grafted plant, and the latter thrive as vigorously and luxuriantly as possible. The same result is obtained by budding as by grafting; but here a single bud of the scion, instead of an entire branch, is transferred to the stock. This is accom- plished in the following manner: — Two incisions at right angles forming a T, are made in a branch of not too great age belonging to the plant employed as substratum. These cuts are carried through the bark as far as the wood. The two lobes of bark, formed by the T-shaped incision, are then carefully raised from the wood, and the bud to be transplanted is pushed in under them. The bud which has previously been taken away from the scion must have retained in that process a portion of bark, and usually the bit of bark peeled off is given the shape of a little shield. This shield, carrying the bud that is to be transferred upon it, is now introduced between the two lobes above mentioned, and the lobes are folded over it Id such a manner as to allow the bud to project freely from the slit between the GRAFTING AND BUDDING, 215 lobes. Besides this, the whole is held together by a bandage, the shield in particular with its bud being pressed firmly on to the new substratum, and thereupon, as a rule, coalescence takes place at once, and the inserted bud grows out into a branch which stands in exactly the same relation to the stock as a Loranthus to the oak whereon it is parasitic. All the branches belonging to the substratum, that is to say, to the wild stock, may then be removed, leaving only the one branch that has sprung from the stranger-bud, the result being that all the juices absorbed from the ground by the substratum are concentrated in this branch and cause it to grow with the greatest exuberance. There is between this process of budding and the settling of a parasite a further resemblance in that shrubs and trees cannot all be made to unite at pleasure one with the other. A successful result of grafting or budding can only be counted upon when nearly allied species, belonging to the same genus or family, are employed for the purpose. Almonds, peaches, apricots, and plums can be grafted the one upon the other; so also can quinces, apples, pears, medlars, and white- thorns. But we must relegate to the realms of fiction such assertions as that peaches might be successfully grafted upon willow stocks, or that the Siberian Crab (Pyrus salicifolia) has sprung from the grafting of branches of the Pear upon the Willow and other tales of the sort. Whether it is possible by grafting or budding to produce new forms, or at least hybrids, is a question which will claim our attention in connection with the problem of the origin of new species. The only additional remark to be made here is that notwithstanding the undeniable simi- larity between grafted or budded plants and the parasitic Loranthaceas, a very essential difference exists in the circumstance that the latter develops roots which continue to grow year by year, and are always penetrating into new layers of the host's tissues, whereas this is never observed in the case of grafted or budded plants. When the branch of a Peach is grafted on an Almond-tree, there is, it is true, an organic union of the two at the place of contact, and the juices from the wood of the Almond stock are conducted direct into the grafted Peach-branch; but neither roots nor sinkers ever arise from the base of the adnate branch or penetrate into the stem of the Almond-tree. 216 IMPORTANCE OF WATER TO THE LIFE OF A PLANT. 5. ABSORPTION OF WATER. Importance of water to the life of a plant — Absorption of water by Lichens and Mosses, and by Epiphytes furnished with aerial roots — Absorption of rain and dew by foliage-leaves — Develop- ment of absorptive cells in special cavities and grooves in the leaves. IMPORTANCE OF WATER TO THE LIFE OF A PLANT. In the building up of the molecules of sugar, starch, cellulose, fats, and acids, of proteids, and, in short, of all the important substances of which a plant is composed, atoms of water have to be incorporated as constructive material, and without water no growth or addition to the mass of a plant whatsoever could take place. From this point of view water must be considered just as indispensable an item in the food of plants as the carbon-dioxide of the air. But water plays, in addition, another important part in plant-life. The mineral salts which serve to nourish hydro- phytes, land-plants, and lithophytes, as also the organic compounds which are the food of saprophytes and parasites, can only reach the interior of plants in the form of aqueous solutions. They can only pass through a cell-wall when it is saturated with water, and, having reached the interior of a plant, they can only be convej'ed to the places where they are worked up through the medium of water. In con- nection with the discharge of these functions in a living plant, water must be regarded as a dynamic agent. Just as a mill on a stream only works so long as its wheels are kept in motion by the water, and stops at once if the latter fails, or flows by in insufficient quantity, so the living plant, as it nourishes itself, grows and multiplies, needs a continuous and abundant supply of available water to render possible the performance of the complicated vital processes within it. This avail- able or organizing water is not in chemical combination like that which is present as food-material, and is, in general, not permanently retained. On the contrary, we must conceive it as perpetually streaming through the living plant. In the course of a summer, quantities of water, weighing many times as much as the plant itself, pass through it. The total amount of water in chemical combination in the organic compounds of a plant is very trifling compared with this, though it often happens that the weight of the latter in a particular plant is greater than that of all the other substances put together. Inasmuch as this water evaporates from plants in dry air, and that it may also easily be withdrawn by alcohol or other means, very simple experiments suffice to give an idea of the great bulk of free water in any plant. Berries, fleshy fungi, succulent leaves, and things of that kind, if left in alcohol, are reduced in a short time to barely half their size in the fresh state. The Nostocineae, which are gela- tinous when alive, and many fungi {e.g. Guepinia, Phallus, Spathularia, Dacryo- myces) shrivel up so stringently in drying, that a piece possessing an area of 1 square centimeter when fresh leaves only a dry crumbling mass covering scarcely 3 square millimeters. A Nostoc, which weighed 2-224 grms. in the fresh state only ABSORPTION OF WATER BY LICHENS AND MOSSES. 217 weighed 0*126 grm, after desiccation, so that when alive it must have contained 94 per cent, of water. Bog-moss, weighing 25*067 grms. before the abstraction of the water was reduced to 2*535 grms. afterwards, showing that the percentage of water was 90. Similar results are obtained in the cases of succulent leaves and stems of flowering plants, Cucurbita, and other fruits. The least proportion of water is contained by mature seeds, solid stony seed -coats, wood, and bark; but even in these an average proportion of 10 per cent of water has been detected. We shall not go wrong in assuming, on the evidence of the weights determined, that most parts of plants, when fresh, consist of dry substance only as regards a third, and as regards two-thirds, of water of imbibition, which passes over into the surrounding air in the form of vapour when desiccation takes place. From all this it follows that water is absolutely necessary to plants as food- material, that it is indispensable as a medium of transport of other substances, and that the demand for water on the part of all plants is very great. Further, we may infer that the importation and exportation of water must be regulated with exacti- tude if the nutrition is not to be disturbed and development hindered. Water-absorption is at its simplest in hydrophytes. In this case it coincides with the absorption of the rest of the food-materials, and there is therefore nothing material to add to the statements already made on that subject. As regards land-plants, lithophytes, and epiphytes, we may likewise refer to what has been already said in so far as these plants suck up water at the same time as food-salts, by means of absorption-cells, from the substratum to which they are attached, or the earth in which they are rooted; but to the extent that they take also water direct from the atmosphere, and have the power of absorbing that water immediately they require it, must be discussed in the following pages. ABSORPTION OF WATER BY LICHENS AND MOSSES, AND BY EPIPHYTES FURNISHED WITH AERIAL ROOTS. The plants which absorb water direct from the atmosphere may be classified in several groups with reference to the contrivances adapted to the purpose. Of all plants lichens are most dependent on atmospheric moisture. Many of them, especially the Old Man's Beard Lichens, which hang down from dried branches of trees, and the gelatinous, crustaceous, and fruticose lichens, which cling to dead wood, and on the surface of rocks and blocks of stone, do in fact derive their necessary supply of water entirely from the atmosphere, and that by absorbing it, not in a liquid but in a gaseous form. The latter circumstance is of the greatest importance to those species in particular which occur on receding rocks, or on the under face of overhanging slabs of stone. Rain and dew cannot reach such places directly, but only by some of the water trickling down from the wet top and sides of the rocks on to the receding wall, and this happens but seldom. Accordingly, lichens occurring in situations of the kind are entirely dependent upon the water contained in the air in the form of vapour. Lichens, however, are also, of all plants, 218 ABSORPTION OF WATER BY LICHENS AND MOSSES. the best adapted for the absorption of aqueous vapour from the air. If living lichens, which have become dry in the air, are left in a place saturated with mois- ture, they take up 35 per cent of water in two days, and as much as 56 per cent in six days. Water in the Kquid form is naturally absorbed much more rapidly still. When Gyrophoras, which project in the form of cups after a long continuance of dry weather, are moistened by a fall of rain, they swell up completely within ten minutes, and spread themselves flat upon the rocks, having in that short space of time absorbed 50 per cent of water. The sajdng, " Light come, light go," is no doubt true in these cases. When dry weather sets in, evaporation from the masses of lichens goes on at a pace corresponding to the previous absorption. In the Tundra, the lichens, which form a soft tumid carpet when moistened by rain, are liable to be so powerfully desiccated in the course of a few hours of sunshine, that they split and crackle under one's feet, so that every step is accompanied by a crunching noise. In the power of condensing and absorbing the aqueous vapour of the atmos- phere, lichens are most analogous to mosses and liverworts, and to those pre- eminently which live on the bark of dry branches of trees or on surfaces of rock, covering places of the kind with a carpet which is often enough interspersed and interwoven with lichens. Like the latter these mosses and liverworts are able to remain as though dead in a state of desiccation for weeks together, but as soon as rain or dew falls upon them they resume their vitality; and similarly if the air is so damp as to enable them to derive sufficient water of imbibition from that source. A specimen of Hypnum molluscum, a moss which covers blocks of limestone in the form of soft sods, was after a few rainless days detached from the dry rock and placed in a chamber saturated with vapour, and it was found that after two days it had absorbed w^ater from the air to the extent of 20 per cent, after six days 38 per cent, and after ten days 44 per cent. Many mosses condense and absorb water with the whole surfa'ces of their leaflets, others — as, for example, the gray rock- mosses clinging to slate formations (Rhacomitriae and Grimmise) — do so especially with the long hair-like cells at the apices of the leaflets, whilst others again only use the cells situated on the upper saucer-shaped or canaliculate leaf -surface. In some bearded mosses (Barhula aloides, B. rigida, and B. ambigua) chains of barrel-shaped cells occur closely packed together upon the upper surface of the leaf and at right angles to it, which to the naked eye have the appearance of a spongy dark -green pad. The terminal cells of these short moniliform chains have their upturned walls strongly thickened, but the other cells have very thin walls and take up water rapidly. It is the same with the various species of Polytrichum, which are provided on their upper leaf-surfaces with parallel longitudinal ridges likewise composed of thin-w\alled, highly-absorbent cells. The rhizoids also play an important part in these processes. These brown, elongated, thin-walled cells entirely clothe the moss stems, usually in the form of a dense felt, and often pro- ject from the under surface of the leaves, whilst in a few tropical species they make their appearance, strangely enough, in the form of little tufts at the apices of the ABSORPTION OF WATER BY LICHENS AND MOSSES. 2] 9 leaflets. In many instances this felt of rhizoids does not come into contact at all Avith the soil, rock, or bark (as the case may be), but is surrounded by air alone, and is able to condense or attract, to use a common expression, the aqueous vapour of the air like a piece of cloth or blotting-paper. In dry weather, it is true, mosses, like lichens, lose their water, but they part with it much more slowly than the latter. This is chiefly due to the fact that the moss-leaflets at the commencement of a drought wrinkle, curl up, become concave, and lay themselves one above the other, so that the water is retained at the bottom for a longer period. A very remarkable contrivance for the absorption of water from the atmos- phere is also exhibited by the white-leaved Fork-mosses (Leucobri/urn) and Bog- mosses (Sphagnaceffi). Although they possess chlorophyll, and assimilate under the \ .^-ipr^}Mrf[M >0i|^0p n ^t^^ ^m^^'^-'^ ^'^^^'i:!lJ^M^iiXi^>^^^^i^-^ Fig. 49.— Porous Cells. 1 Of the white-leaved Fork-moss {Leueobryum) ; x 550. 2 of the Bog-moss (Sphagnum); {Loelia gracilis) ; x 310. ! Of the root of an Orchid influence of sunlight, yet they look like parasitic and saprophytic plants destitute of chlorophyll. They are of a whitish colour and always grow in great cushion- like sods, so that the spots where they grow are deficient in verdure, and stand out conspicuously from their surroundings in consequence of their pale tint. Microscopic investigation at once explains this appearance. The cells containing chlorophyll and living active protoplasts are relatively small, and, as it were, wedged and hidden between other cells many times as great, which have entirely lost their protoplasm by the time they are mature, and then cause the paleness of colour appertaining to the plant as a whole. The walls of these large colourless cells are very thin, and in the Bog-mosses have spiral thickening-bands running round them, being thus secured against collapse. After remaining for a time in a dry environment they are full of air only; but the moment they are moistened they fill with water. If there were an actively absorbent protoplast at work in the interior, the water would be able to pass into the cell-cavity through this easily moistened wall, as in the case of other mosses, owing to the delicacy of the cell- membrane. But the air which fills the cells is not absorptive, and in the case of Leueobryum and Bog-mosses the water reaches the interior, not in consequence of 220 ABSORPTION OF WATER BY LICHENS AND MOSSES. a chemical affinity on the part of the cell-contents, but solely by capillary action. All the cell-walls are perforated and furnished with pores, and through these the water rushes into the interior with lightning rapidity. This extremely rapid influx of water into an air-filled cavity leads us necessarily to the conclusion that each cell has a number of pores in its walls, and that in proportion as water enters through one of the small apertures the air can escape equally fast through another. This is in fact the case. The large cells not only have pores on their external walls, but communicate one with another by similar lioles, and the water soaks in from the one side as it does into a bath-sponge, whilst the air is at the same time forced out on the other. This absorptive apparatus is exceptionally elegant in Leucobryum, which grows abundantly in many woods. In it, as is shown in the illustration above (fig. 49 ^), the adjacent prismatic cells communicate by highly symmetrical, circular gaps made in the middle of the partition-walls, whilst in the Bog-mosses (the various species of SphagnuTn), they are to be seen scattered here and there between the thickening bands on the cell- walls (see fig. 49^). Now these porous groups of cells possess not only the power of taking up water in the liquid state, but also that of condensing it when in the form of vapour. There is no need of any more proximate proof of the fact that the cells previously mentioned as containing chlorophyll, and lying imbedded between the large perforated cells, take up water supplied by the latter, or perhaps it is better to say that the large perforated cells suck in the water for the living green cells. We have only to ask why it is, then, that these small green cells do not absorb water themselves direct from the environment, as is done in the case of so many other mosses and liverworts. It is difficult to answer this quite satisfactorily, but thus much seems certain, that the large porous cells, when full of air, afibrd a means of protecting the small living cells from too excessive desiccation, and that they are in addition preservative of the chlorophyll in the small cells, a matter to which we shall return presently. A certain resemblance to these Leucobryums and Sphagnums, in respect of water- absorption, is exhibited by a few Aroidese, and more especially by a whole host of Orchidaceae. Of the 8000 different orchids hitherto discovered, a good proportion, it is true, are rooted in the earth. But more than half these wonderful plants flourish only on the bark of old trees, and most of them would quickly perish if they were detached from that substratum and planted with their roots buried in earth. A double function appertains to the roots of these Orchideee which inhabit trees. On the one hand they have to fix the entire orchid -plant to the bark, and, on the other, to supply it with nutriment. When the growing tip of an orchid's root comes into contact with a solid body, it adheres closely to it, flatucns out more or less, sometimes even becoming strap-shaped (see fig. 15), and develops papilli- form or tubular cells, which grow into organic union with the substratum, and might conveniently be termed clamp-cells. In many cases these cells creep over the bark, divide, interlace, and form regular wefts. The organic connection with the substratum is so intimate that an attempt to separate the two usually results ABSORPTION OF WATER BY EPIPHYTES. 221 in a detachment of the most superficial parts of the bark, but not of the tubular cells. Now, if a root, after having sent out cells of this kind which contract an organic union with the substratum, reaches into the open, beyond the limit of the Fig. 50.— Aerial Roots of an Orcliid epiphytic upon the bark of the branili of a tree. substratum, it immediately ceases to develop clamp-cells, loses its ligulate shape, and hangs down from the tree in the form of a sinuous white filament. A few root-fibres are as a rule sufficient to fix the plant to its substratum, the bark of the tree, and the rest of the roots put forth by the orchid grow from beginning to end, 222 ABSORPTION OF WATER BY EPIPHYTES. freely in the air. The^^ are not infrequently to be seen crowded together in great numbers at the base of the plant, fox-ming regular tassels suspended from the dark bark of the branches as may be seen in fig. 50, where an Oncidium is represented. Each of these aerial roots is invested externally by a white membranous or papery envelope, and it is the cells of this covering that own the resemblance, above referred to, to the cells of Leucohryuvi and Bog-mosses. Their walls are furnished with narrow, projecting spiral thickenings and therefore do not collapse, notwith- standing their delicacy or the circumstance of their inclosing at times an air-filled cavity; they are further abundantly perforated, two kinds of apertures indeed being found. The one variety arises in consequence of the tearing of the portions of the cell-wall situated between the rib-like projections and consisting of extremely thin and delicate membranes (see fig. 49^); the existence of the other variety is due to the detachment of the cells which protrude in the form of papillae, the result being, in this latter case, the formation of circular holes very similar to those already described as occurring in Leucohryum. The cells resembling papillae have the peculiarity that they roll off" when they get old in the form of spiral bands. The holes, of course, can only occur on the external walls of the outermost cells which border upon the open air, whilst in the interior the communication between the cells themselves is established by means of the rents previously referred to. The entire covering thus composed of perforated cells may be compared to an ordinary sponge, and, indeed, acts after the manner of a sponge. When it comes into contact with water in the liquid state, or more especially when it is moistened by atmospheric deposits, it imbibes instantaneously its fill of water. The deeper- lying living green cells of the root are then surrounded by a fluid envelope and are able to obtain from it as much water as they require. But these roots also possess the power of condensing the aqueous vapour contained in the air. They act upon the moist air in which they are immersed in exactly the same way as spongy platinum or any other porous body. If the aerial roots of OncidiuTn sphacelatiim are transferred from a chamber full of dry air to one full of moist air, they take up in 24 hours somewhat more than 8 per cent of their weight of water, those of Epidendron elongatum absorb 11 per cent, whilst in the case of many other tropical orchids the amount thus imbibed is doubtless much more considerable still. The power of condensing aqueous vapour, and other gases as well, is of the greatest importance to these plants. The tree-bark serving as their substratum, to which they are fastened merely by a few fibres, is anything but a permanent source of water. Such water as the bark does contain reaches it, not from the interior of the trunk and indirectly from the soil in which the trunk has its roots, but from the atmosphere; that is to say, from the very source whence the epiphytes upon the bark must also derive their supply. Now, when on the occa- sion of a long-enduring uniform aerial temperature, there is a failure of atmos- pheric deposits, which is a regularly recurring circumstance in the habitat of the orchids in question, the sole source of water left is the vapour in the air, and the ABSORPTION OF WATER BY EPIPHYTES. ' 223 only possible method of acquiring that vapour is the condensation of it by the porous tissue investing the roots. In the event of the air around the orchid-plant containing temporarily but very little moisture, the porous tissue dries up, it is true, very quickly; its cells fill with air and their function as condensers is interrupted. But these air-filled cellular layers then form a medium of protection against excessive evaporation from the deeper strata of the root's tissues, which might be very dangerous in the case of this kind of epiphyte. There is a wide-spread impression that the tropical orchids grow in a perpetually moist atmosphere in the dark shade of primeval forests, and this preconception is fostered by pictures of tropical orchids representing these plants as living in the most obscure depths of woods. In reality, however, the orchids of the tropics are children of light. They thrive best in sunny spots in open country. Those species in particular which have their aerial roots invested each by a thick, white, papery, porous covering belong to regions where a long period of drought occurs regularly every year, and where, in consequence, vegetative activity is subject to periodical interruption, as it is in the cold winter season of the more inclement zones. For epiphytes inhabiting these regions of the tropics a more expedient structure of root cannot easily be imagined. In the dry season the papery covering reinforces the safeguards against too profuse transpiration on the part of the living cells in the interior of the root, and in the wet season it provides for the continuous supply of the requisite quantity of water. In this sense the porous layer is to a certain extent a substitute for wet soil, or, in other words, the concealment of the living part of an aerial root in the saturated envelope is analogous to that of the root- fibres of land-plants in the damp earth. The manner in which the water reaches the inner cells of an aerial root from the saturated envelope is also quite characteristic. Under the porous tissue lies a layer composed of two kinds of cells of different sizes. The larger cells are elongated and have their external walls, which are adjacent to the porous tissue, thickened and hardly permeable by water. Between these lie smaller, thin-walled, succulent cells, which admit the water from the porous envelope, and should therefore be regarded as absorption-cells. It is also noteworthy that the porous, paper-like covering is discarded as soon as an aerial root is placed in earth. Most orchids with aerial roots perish, it is true, when they are treated like land-plants and planted in soil; but a few species, on occasion, bury their aerial roots spontaneously in the earth and push off" their envelopes, and then the imbedded parts exercise the same functions as in the case of land-plants. We have already mentioned that, in addition to thousands of orchids, several Aroidese exhibit the porous, papery covering on their aerial roots. But still more frequently the air-roots of Aroids, which live as epiphytes upon trees, are furnished with a dense fringe of so-called root-hairs in a broad zone behind the growing- point. The hairs project on all sides from the roots, which are surrounded by air; they are crowded very closely together and give the parts affected a velvety appearance. Besides several Aroidese, one of which (Philodendron Lindeni) is drawn on the left side of fig. 51, many other epiphytes, such as the South 224 ABSORPTION OF WATER BY EPIPHYTES. 7- 7 ...-^ belono-ino- to the Commelynacese, represented on the w w \ y IV il.-AH 1.1 K- 1. "ill, „..t I, ,1, , , on the let. Philodmdro,. Lmdtnl. <,n llie rlgln c««p.l.» 2»~"i«. coating on their aerial roots. The roots o£ the tree-ferns are short but spring in thou^atds from the thick stem, and are so closely packed that ''- ™ho e surfac clothed as it were by a woven mantle of rootlets. After some tone these aerua roots turn deep brown, whilst the hairs collapse and die, and both are converted ABSORPTION OF RAIN AND DEW BY THE FOLIAGE-LEAVES. 225 into a mouldering mass. But as soon as they perish other new air-roots, covered with golden-brown velvet, make their appearance and take their place. These aerial roots never reach the ground or adhere to any substratum, so that their hairs cannot contract an organic connection with a solid body. It is consequently also impossible in this case for the root-hairs to draw moisture from the soil in the capacity of absorption-cells. These root-hairs, however, are scarcely ever in a position to take up even the atmospheric deposits. The various species of Philodendron and the other epiphytes referred to, have large leaves which cover the air-roots hanging from the stem like umbrellas, and every tree-fern also bears at the top of its stem a tuft of great fronds, which prevents falling rain from wetting the aerial roots. Moreover, the very plants whose air-roots exhibit a velvety coating occur in woods where the tops of the trees arch over the ground in lofty domes, and form a sheltering roof against deposits from the atmosphere. On the other hand, the air within these forests is saturated with aqueous vapour, and it is certain that the velvety roots have the power of condensing vapour, and that the root-hairs instantly suck up the condensed water and convey it to the deeper-lying layers of cells. The truth of this has been established by the results of repeated experiments. Thus, air-roots of the tree-fern Todea harhata, after being transferred from moderately damp air into a chamber full of vapour, condensed and absorbed in the space of twenty-four hours water amounting to 6 '4 per cent of their weight. There is, therefore, no doubt that water may be acquired in this way also by plants, even though the instances may not be very numerous. All plants in which this kind of water-absorption has been hitherto observed grow in places where the air is very moist the whole year round, and where there is also no risk of the temperature falling below freezing- point. Under other conditions, especially in places where the air is periodically very dry, these plants would not be able to survive; for, although they possess organs for the condensation and absorption of water, they have no means of protec- tion against the desiccation of these organs. ABSOEPTION OF RAIN AND DEW BY THE FOLIAGE-LEAVES. The idea that plants absorb with their roots such water as they require is so intimately associated with our whole conception of plant-life, that this process is commonly adduced for the purpose of analogies of the most various kinds, and one looks upon the water-absorption effected by aerial roots in the manner just described really as a thing to be expected, notwithstanding the fact that in this case, as the above account shows, the phenomenon is not so simple as is usually supposed. We now turn to the consideration of land-plants. If the leaves of plants cultivated in pots become flaccid, water is poured as quickly as possible upon the dry soil with a view of supplying the roots which ramify in it with moisture. Nor does the result fail to be produced. In a short time the foliage becomes fresh and elastic again, the roots having discharged their function. Even in the open air, it is especially 226 ABSORPTION OF RAIN AND DEW BY THE FOLIAGE-LEAVES the soil in which the roots are imbedded that a gardener waters on dry days, although incidentally he may pour the water over the aerial parts of the plants. He sees, however, that the water which falls in the form of rain or dew upon the foliage and stems normally runs off them at once, or else collects in drops, which trickle down whenever the plant is shaken by the wind, and are sucked up by the thirsty ground. This phenomenon must be due to the possession by the leaves of special contrivances to prevent their being wetted. It does not in any case support the idea that foliage is as well adapted for the absorption of water as experience has proved subterranean roots to be. This train of thought, which forces itself upon every unbiassed observer of the processes as they take place in nature, is certainly warranted in the majority of cases. Each absorption-cell on the roots buried in the earth has an easily permeable membrane, and, as is well known, water passes from damp earth through the cell-membranes into the interior of a plant with great rapidity. The water in the interior of the plant would be equally easily withdrawn through these cell-membranes by dry surroundings, but, as it is, this scarcely ever happens, in consequence of the roots being situated underground. In the case of aerial parts, especially the foliage-leaves, the circumstances are quite different. The leaves have to yield up to the air a portion at least of the water conducted from the roots, because, as will be more thoroughly explained later on, it is only by means of this evaporation that the entire machinery in the interior of the plant can be kept in motion. But this evaporation must not go too far; it must be in proper relation to the absorption of water by the subterranean roots, and be regulated to that end if the plant is not to run the risk of drying up altogether at times — an occurrence which flowering plants are unable to survive, although the mosses described in former pages have that power. Accordingly, in the case of the foliage-leaves of flowering plants, evaporation is confined to certain cells and groups of cells, and these, in addition, have contrivances by means of which evaporation can be entirely stopped on occasion of great drought. It stands to reason that all contrivances which make it impossible for water to pass from the interior of the leaves through the walls of the superficial cells into the surrounding air also hinder the entrance of water into the leaves from the atmosphere. It would be altogether inconsistent with the system of arrangement of the sub- ject adopted in this book if we were to discuss here all the contrivances serving to regulate the exhalation of water by leaves, and we must, therefore, confine ourselves to referring, by way of introduction, quite briefly, to the following facts, namely, that those pores on the surface of leaves which are known by the name of stomata, and are used as doors of egress by the exhaled water, do not admit rain or dew, or in general, any water in the liquid state; that the so-called cuticle covering the exter- nal walls of the epidermal cells in leaves is an additional barrier to both egress and ingress of water; that when, in particular, this cuticle is furnished with a wax-like coating, water does not adhere to the surface of cells so protected; and, lastly, that atmospheric moisture can only penetrate into the interior of the plant at parts of the leaves where the waxen incrustations are absent, where water remains adherent ABSORPTION OF llAIN AND DEW BY THE FOLIAGE-LEAVES. 227 to the leaf -surfaces, and they are distinctly wetted. But even cells and groups of cells of this kind usually act but for a short time as absorption-cells, and only when the necessity and craving for water is very great, or when there is an opportunity of acquiring nitrogenous compounds at the same time as the water; and here, again, special contrivances are always present which regulate this kind of water- absorption, and render it impossible whenever it is not truly advantageous. At first one would suppose that amongst the cells composing the epidermis of foliage-leaves, those are best adapted to the absorption of water from the atmosphere which take the form of hairs. The superficial area being as great as possible, and the contained matter relatively little, one can scarcely in fact conceive a conforma- tion better suited to the purpose of water-absorption. As, moreover, the area of contact between the cells of the leaf and of a hair is small, there would afterwards be but very little evaporation through the surface of the hair of the water once sucked up by it and conducted into the interior of the leaf. In a word, these hairs on the surface of a leaf appear to be peculiarly adapted to the taking up of water, and not at all favourable to its exhalation. The hypothesis based on these observations is indeed entirely applicable to the case of hairs occurring on the leaflets of mosses, as has been already stated. But it does not hold in the case of the hair-like struc- tures which spring from the leaf -surfaces of flowering plants. These are frequently not wetted at all by water; rain and dew roll ofl'them in drops, and cannot, there- fore, be absorbed by them. This is true even of many soft trichomes (hair-structures) which form investments upon leaves, and which seem to be more than any fitted for the absorption of water. For instance, experiments upon the woolly leaves of the Great Mullein {Verhascum Thapsus) have shown that they neither condense aqueous vapour nor take up water in liquid drops. Small importance must be attributed to the thickness of the cuticle, for sometimes it is the very cells which are equipped with a cuticle of considerable stoutness that are adapted to admit water, under certain circumstances, through their walls. On the other hand, much depends upon the presence of wax in the cuticle and upon the contents of the cells; that is to say, upon whether those contents in particular have a strong or weak affinity for water. If the cells of the hairs are full of air they are not adapted to the absorption of water. If a hair is septate, i.e. consists of a simple series of cells, only the undermost or else only the uppermost cells of the series absorb water. Instances wherein it has been observed that the lowest cells alone in hairs of the kind become absorption- cells are afforded by the Alfredia, represented in fig. 14, by Salvia argentea, and several other steppe-plants. The same statement is made concerning the widely- distributed Stellaria media, the common Chickweed. This last has hairs on the intemodes of the stem, running down in ridges from node to node. Usually only one side of the stem exhibits a ridge of hairs of the kind, and the ridge always terminates at the thickened node, whence springs a pair of opposite leaves. The stalks of these leaves are somewhat hollowed out and have their edges beset with hairs like lashes. The hairy ridges on the segments of the stem are readily wetted 228 ABSORPTION OF RAIN AND DEW BY THE FOLIAGE-LEAVES. by rain and retain a considerable quantity of water. The water that they cannot hold they conduct downwards to the ciliate axils of the next lower pair of leaves, where it is drawn through the lash-like hairs in due course and collected into a ring of water surrounding the node (see fig. 52^). If this accumulation of water becomes so voluminous and heavy that it cannot any longer be retained by the fringe of lashes, the surplus glides on to the unilateral ridge of hairs on the adjacent internode down to the pair of leaves below. Accordingly, after a shower every node from which leaves arise is seen to be inclosed in a water-bath, and the hairy Fig. 52. — Hairs and Leaves which retaiu Dew and Rain. Dwarf Gentian {Gentiana acaulis). * Lady's Mantle {Alchemilla vulgaris). 3 Chickweed (Stellaria media). ridges also are so soaked with water that they look like edgings of glass. All the individual cells in each of the hairs are full of protoplasm and cell-sap, but only the lowest, which are very short, really act as absorption -cells. When these cells become at all relaxed in dry air, the fact is indicated by the appearance on the external cell- wall of fine striae (see fig. 53 ^ and 53 2). The protoplasts inhabiting them attract water, and after being relaxed in the manner referred to the cells regain their turgidity on being wetted, whilst the fine wrinkles on the outer membrane are in consequence immediately smoothed out. Although the upper cells of the hair possess a less thick cuticle, they, on the other hand, seem not to absorb any water, but to serve rather to conduct it by their surfaces. This case is, as we have said, comparatively rare, and the corresponding absorp- ABSORPTION OF RAIN AND DEW BY THE FOLIAGE-LEAVES. 229 tion of water is not veiy considerable. But it often happens that the uppermost cells of a septate hair are developed into absorption-cells. The terminal cell is then usually spherical or ellipsoidal and larger than the rest, or else this cell is divided into two, four, or a greater number of cells, which together form a little head, whilst the lower cells constitute a stalk supporting it (see fig. 53 ^ and 53 *). In botanical terminology structures of this kind are named capitate or glandular hairs. The protoplasm in the cells of the head is, for the most part, of a dark colour, and the Fig. 53.— iHairs from stem of SfeZtona media; xllO. 2 Lowest cells of the same hairs ; x200. s Capitate hairs ot Centaurea Balsamita; xl50. * Capitate hairs of Prfarfiromurn Zmdmn. ; x 150. cell-membranes are readily permeable by water, which is attracted with great energy by the cell-contents. The cell-membrane is often very thick, it is true, but as soon as water comes into contact with it the outer layer is discarded, the inner layers swell up and the water passes through these swollen layers into the interior of the cell. This happens, for instance, in many pelargoniums and geraniums, wherein the capitate cells go through a process of excoriation on every occasion of the imbibition of water (see fig. 53"^). In other plants the walls of the capitate cells are everywhere thin, and not only do the cell-contents consist of a viscid gum- like mass, but the external surface of the wall is also covered by a layer of viscid excretion. In many cases the viscid matter excreted by the glands spreads over the entire surface of the leaf, so that the latter feels sticky and looks as if it were 230 ABSORPTIOX-CELLS ON LEAVES. coated with varnish. Many plants which have their roots buried in crevices of rock and no small number of herbaceous steppe-plants are quite thickly covered with glandular hairs of the kind. Centaurea Balsamita (see fig. 53^), a plant occurring on the elevated steppes of Persia, may be selected as an example of the latter group. The advantage of the structure of capitate hairs is not far to seek. In dry weather the thick cuticle (Pelargoniuvi) or the varnish coating {Centaurea Balsamita), as the case may be, prevents desiccation of the cells and groups of cells in question. But as soon as rain or dew falls, the cuticle and the coat of varnish take up water, and it is by their instrumentality that water reaches the interior of the cells. Thus, whilst the exhalation of water is hindered, its absorption is not. Other epidermal cells of foliage-leaves besides trichomes are capable of acting as absorption-cells, although this action, for reasons already given, is very restricted, and is only had recourse to when the turgidity of the cells of the foliage-leaves has diminished, and the water exhaled by those cells is not being restored by the ordinary apparatus of conduction from the roots. If branches are cut from plants which bear no glandular or other form of hair on their leaves or stems — as, for instance, the leafy stem of Thesiwm alpinum — and the cut ends are closed with sealing-wax, and the branches left to wither, and, when quite withered, are immersed in water, they freshen up speedily and the leaves become tense again, the cells having recovered their turgidity. Here, then, decidedly absorption has taken place through the ordinary cuticularized epidermal cells. Certainly these epidermal cells in Thesiiini are not protected against wetting. Wherever the epidermal cells are not susceptible of being wetted owing to a coating of wax or any other contrivance there could naturally be no question of water being absorbed. This very circumstance, however, leads to the supposition that an important part in water absorption is to be attributed to the alternation of wettable and non-wettable parts on one and the same leaf. In the case of many foliage-leaves one can see that only those cells of the epidermis which lie above the veins of the leaf retain the water which comes upon them, that is to say, are wetted by it, whilst the water rolls off the intervening areas of the lamina. Indeed, there are in many instances contrivances obviously designed for the purpose of conducting water from parts of the epidermis not liable to be wetted to parts that can be moistened. DEVELOPMENT OF ABSORPTION-CELLS IN SPECIAL CAVITIES AND GROOVES IN THE LEAVES. The contrivances last described are all only adapted to rather a casual appropri- ation of water from the atmosphere. But besides these we find a number of other contrivances, which render it possible for every rolling dewdrop and every passing shower to be made of use to the utmost extent. These contrivances consist of a variety of depressions and excavations, in which rain and dew are collected and protected against rapid evaporation. Some species have deep hollows or channels, others little pits, whilst others again have basins, vesicular or bowl- ABSORPTION-CELLS ON LEAVES. 231 shaped structures, to collect and absorb the water; and the construction of the protective apparatus, which prevents too rapid evaporation into the air of water that has once flowed into the depressions, is as various as the form of the depressions themselves. A short account of the most striking of these structures will now be given. Such water-collecting grooves as are closed, so as to form ducts, occur principally in petioles and in the rachises of compound leaves. For instance, in the Ash the leaf rachis, from which the leaflets arise, is furnished with a groove on its upper surface. Owing to the fact that the edges of this groove, which are strengthened by a so-called collenchymatous tissue, are bent up and curved over the groove, a duct or conduit pipe is produced, and this duct only gapes open at the places where the leaflets are inserted upon the rachis, and where, therefore, the drops of rain to which the leaflets are exposed flow off" into the groove (see fig. 54 ^ ). The simple hairs and peltate groups of cells developed in the grooves and ducts (fig. 54 ^ and 54^) are not merely transiently moistened, but inasmuch as the water is retained there for several days after a fall of rain, they are during that time immersed in a regular bath of water, and are able to absorb the moisture very gradually. In many Gentianese — most conspicuously in the large-flowered Dwarf Gentian (Gentiana acaulis) — the decussate pairs of radical leaves form a loose rosette (see fig. 52^). The larger dark-green blade of each leaf is flat and even, and only the pale-coloured base is fashioned into a groove. This groove is made deeper by the tissue of the leaf being pufled up round it, and as all the leaves of the rosette arise close together, the groove of each leaf is covered by the lamina above it. The rain or dew accumulated from the blade remains standing in this concealed nook for some time without evaporating, so that absorptive apparatus with the power of taking up water has plenty of time for the purpose. In this case the absorptive apparatus is in the hindmost extremity of the groove, and consists of long, club-shaped structures composed of extremely thin- walled cells (see fig. 54*), and these act so energetically that if leaves are cut off" and left to fade, and if the cut surfaces are stopped with sealing-wax, and the whole then bathed with rain- water, they take up in twenty-four hours about 40 per cent of their weight of water. A similar phenomenon occurs in the case of a number of Bromeliacese which adhere by a few roots to the bark of trees in the tropics, and have grooved rosetted leaves, the latter covering one another, and being arranged in such a manner as to form a regular system of cisterns. At the bottom of each cistern there are special groups of thin-walled cells which suck up any water that flows in when rain falls. On the under surface of the leaves of the Cow-berry (Vaccinium Vitis-Idcea) Httle depressions are formed, and in the middle of each depression there is a club- shaped structure composed of small thin- walled cells, which contain slimy, viscid substances and act as absorbent organs. The rain which falls upon the upper surface of the leaf gets drawn over the edges on to the under surface, fills the small depressions occurring there, and is taken up by the absorptive apparatus. A 232 ABSORPTION-CELLS ON LEAVES, similar contrivance is also exhibited by the leaves of alpine roses and those of the American Bacharis. For instance, on the under surface of the leaves of the Alpine Rose (Rhododendron hirsidum) there is a large number of discoid glands (fig. 54^), each of which is supported on a short stalk and sunk in a little hollow (fig. 54*") The cells composing the gland are arranged radially, and contain slimy, resinous matters capable of swelling up. These contents are also excreted, and then cover the entire glandular disc, and often even the whole surface of the leaf in the form Fig. 54.— Absorption of Water by Foliage-leaves. » Grooved rachis of the ash -leaf. 2 Section through the same ; x30. ' Peltate group of cells from the groove. « Section through the base of a leaf of the Dwarf Gentian ; x20. «Under side of a leaf of iiAododendroji /hwm^mhi; x30. sSection through a leaf of Rhododendron hirsutum. of a light-brown crumbly crust. When drops of rain fall upon Alpine Rose leaves, the whole of the upper surfaces, in each case, is in the first place moistened; but without delay, and partly through the action of the hairs fringing the margin, the water soaks on to the under side of the leaf. As soon as it reaches the glands it is taken up by the crumbly incrustation mentioned above, which swells up in con- sequence. The little cavities in which the glands are situated also fill with water, and each gland is then immersed, as it were, in a bath, and able to absorb as much moisture as is required. Owing to the glands being invariably developed above the vascular bundles of the leaf (see fig. 54^), the water that is absorbed can be conducted without delay by them to the places where it is required. As soon as the leaves of alpine roses become dry again, the mass of resinous mucilage again ABSORPTION-CELLS ON LEAVES. 233 forms a dry crust over the glands and protects their tender-walled cells from too great evaporation. Very remarkable also are the structures adapted to absorption on the leaves of saxifrages belonging to the group Aizoon, and on those of a large proportion of the Plumbagineae. The saxifrages in question have little depressions visible to the naked eye upon the upper surface of the leaves behind the apex, and along the margins. When the margin is dentate or crenate, as, for instance, in Saxifraga Fig. 55.— Absorptive Cavities and Cups on B'oliage-leaves. Leaf from a slioot of the Aspen, ^jlie base of this leaf; x3. s Section through an absorption-cup; x25. < Leaf of Acantholimon Senganense. 6 Section through part of this leaf; xllO. « Leaf of the Evergreen Saxifrage {Saxi/raga Aizoon). 1 Two teeth from the margin of this leaf. The absorptive cavity in the upper tooth incrusted with lime; the lower one with the incrustation removed. ' Section through a tooth from the leaf and its absorptive cavity ; x 110. Aizoon (see fig. 55^), one of these cavities occurs in the middle of each tooth. The cells forming the outer edge of the tooth or scallop are always much thickened, firm, and rigid; but the median portion of the leaf as a whole is fleshy, and composed of a bulky large-celled parenchyma. The vascular bundle, after entering the leaf at its base, divides into a number of lateral bundles which eitlier run towards the margin without further ramification (as in Saxicwsia), or else form a net-work by uniting one with another in their course (as in Saxifraga Aizoon). These lateral bundles terminate in the marginal teeth of the leaf and immediately beneath the little cavities which occur there, whilst the extremit}'- of each bundle swells into a knob or pear-shaped enlargement strongly resembling the roundish groups of spirally-thickened cells in the tentacles of the Sun-dew 234 ABSORPTION-CELLS ON LEAVES. (cf. fig. 26^). The bottom of each depression is made up of cells with very thin external walls, and the function of these cells is to suck up the water that flows into the cavity. It is obvious that the absorbed water passes thence into tho enlarged extremities of the branches of the vascular bundles, and may then be conducted to other parts of the leaf. Seeing that all these saxifrages have their habitat in crevices of rocks on sunny declivities, they are much exposed to desiccation in times of drought. The epidermal cells of the medial area and those of the extreme edge are no doubt protected by a very thick cuticle (see fig. 55^); but in the case of the thin-walled cells at the bottom of the depression there is the danger of as much or even more water escaping through them, in the form of vapour, than has been previously taken in during the prevalence of rain. In order to prevent this loss of moisture recourse is had to a very remarkable contrivance for closing the cavity, viz., an incrustation of carbonate of lime. In many saxifrages this crust covers the whole face of the leaf, in others only the margin, or the spot where the depression occurs. In the latter case it looks like a lid over the cavity. At that spot the crust is always thickened, and sometimes it forms a regular stopper which fills up the entire cavity. It rests upon the epidermis of the leaf, but is not adnata thereto, and may be removed with a needle. When a leaf is bent the crust is ruptured and breaks up into irregular plates and scales, and a strong gust of wind would then easily strip off the fragments and blow them away. In species subject to this danger, as, for instance, Saxifraga Aizoon, in which the resetted leaves curl strongly upwards and inwards in dry weather, the crust of lime is held fast by peculiar plugs which arise from individual epidermal cells projecting above the rest in the form of papillae (see fig. 55^). These plugs are found principally on the side walls of the cavities, but are also scattered every- where on the epidermis of the margin of the leaf. They are so incrusted with the lime that the latter cannot easily fall off, and a comparatively strong pressure must be applied with the needle to detach it from the substratum. The calcium carbonate of which these crusts consist is excreted in solution by the plant from pores occur- ring at the bottom of the depressions. The pores are constructed like ordinary stomata, but are, as a rule, somewhat bigger, and it is not improbable that, when once the lime crust has formed from the excreted solution, they take part in the function of transpiration. There is scarcely any need for further explanation of the manner in which the apparatus here described acts. When rain or dew falls on a saxifrage leaf the whole upper surface is moistened directly, whilst the water soaks under the crust of lime, and, diffusing itself there, fills in a moment the depressions, and is taken up by the absorption-cells situated at the bottom of the latter. The calcareous stopper imbedded in each cavity is only upheaved by this process to a trifling extent. In dry weather the crust is appressed closely to the epidermal cells, and the stopper descends again and impedes the evaporation of water from the thin- walled cells within the cavities. The absorptive organs on the leaves of Acantholimon, Goniolimon, and a few ABSORPTION-CELLS ON LEAVES. 235 other Plumbaginese, resemble in an extraordinary degree those pertaining to saxi- frages. The depressions are here found uniformly distributed over the entire sur- face of a leaf, and when they are closed by a crust or scale composed of calcium carbonate, the leaves are dotted with white spots, as may be seen in the drawino- of a leaf of Acantholimon Senganense given in fig. 55 ^ Upon the calcareous scale being removed, a little cavity is revealed beneath, and one observes that the floor of this cavity is composed of from four to eight cells, separated by radial partition- walls, and with exceedingly thin and delicate outer walls. The other epidermal cells adjoining the cavity are, on the contrary, always furnished with a thick cuticle (see fig. 55 ^). Whenever water is being copiously supplied to the roots, and the turgidity of the cells in the leaves is great, the cells forming the floor of the cavity excrete bicarbonate of lime in solution. Part of the carbonic acid escapes into the air, and the insoluble mono-carbonate of lime in the water then forms a crust, which fills and covers the cavity, and often even spreads over the whole leaf, constituting a coherent calcareous coat. All Plumbagineae which exhibit this contrivance — that is to say, the various species of Acantholimon, Goniolimon, and Statice — inhabit steppes and deserts, where in summer no rain falls for months together, and the soil becomes dry to a con- siderable depth, so that extremely little water is available for the roots. Although the rigid leaves are protected by a thick cuticle, and by crusts and scales of lime against excessive evaporation of their aqueous contents, still it is difficult to avoid some slight loss of water, especially when the noon-day sun beats down upon the steppe, and, owing to the extremely arid nature of the soil, it is scarcely possible to replace this loss, however small it may be, by absorption from the earth on the part of the suction-cells on the roots. All the more welcome to plants of the kind is the dew which sometimes falls copiously on steppes and in deserts in the course of the night; it wets the rigid leaves, and, soaking immediately underneath the crusts and scales of lime to the thin-walled cells at the bottom of the cavities, is absorbed with avidity by them. When drought returns with the day, the scales of lime close tightly down like lids on the epidermis beneath, and, so far as possible, prevent evaporation. In particular, they impede the exhalation of water from the thin- walled cells at the bottom of the cavities — a loss which would otherwise be quite inevitable, and would be followed by a rapid desiccation of the entire plant. To prevent the calcareous lids from dropping off, there are either, as in Saxifraga Aizoon, papilliform or conical projections from cells in the immediate vicinity of the cavities, which projections often have hooked ends and confine the crust of lime, or else each cavity is somewhat contracted at the top and enlarged below, so that the lime stopper, being shaped according to the contour of the cavity, cannot fall out. A significance similar to that attributed to calcium carbonate excretions belongs also to the saline crusts which are found covering the leaves of a few plants grow- ing on the arid ground of steppes and deserts in the neighbourhood of salt lakes and on the dry tracts of land near the seashore. Owing to the fact that in these 236 ABSORPTION-CELLS ON LEAVES. situations crystals of salt are sometimes to be seen separated out from the soil, and lying as a white efflorescence upon the ground, it used formerly to be believed that the salt incrusting leaves and stems was derived, not from the plants in question, but from the soil around, and had only spread from there over the various plant- members. But this is not the case. As a matter of fact, the salt observed on the leaves and stems of Frankenia, Reaumuria, Hypericopsis persica, and a few species of Taniarix and Statice, is produced from the substance of the leaves. It is excreted in just the same way as the crust of lime, above described, is from the leaves of saxifrages. To the naked eye the surfaces of the leaves in all the plants enumerated have a punctate appearance. On closer inspection, it is evident that, corresponding to each dot, there is a little cavity, the deepest part of which is constructed of cells with extremely delicate external walls. In quite young leaves only a single thin- walled cell of the kind is to be seen at the bottom of each shallow depression. But this divides, and, by the time the leaf is full-grown, from two to four cells are seen to have arisen by division of the one cell. Stomata are, in addition, intercalated in the membrane in the neighbourhood of these thin- walled cells, and, in the rainy season, when there is no lack of water in the habitats of the plants in question, a watery juice, containing a large amount of salts in solution, exudes from these stomata. The saline solution soaks over the whole surface of the leaf, and in a dry atmosphere crystals form from it and adhere to the leaf in the form of little gland- like patches or continuous crusts. If these tamarisks, frankenias, and reaumurias are observed during a rainless season, the crystals of salt are seen under the noon-day sun glittering on the leaves and stems, and may be detached in the form of a fine crystalline powder. But if the same place is visited after a clear night, no trace of crystals is to be seen; the little leaflets have a green appearance, but they are covered with a liquid with a bitter salt taste,^ and are damp and greasy to the touch. The crystals have attracted moisture from the air during the night, and have deliquesced, and the saline solution not only covers the whole of the leaf, but also fills the little cavities visible as dots to the naked eye. The thin-walled cells at the bottom of the cavi- ties differ from the rest of the epidermal cells and the guard-cells of the stomata, in that they are susceptible of being wetted, and they may act as absorption-cells, and allow the water, attracted by the salts from the air, to pass through their thin walls into the interior of the leaves. When the air dries under the rising sun, crystals are again formed from the solution of salts, and, covering the leaves once more in the form of crusts, fill up the depressions and protect the plants during the hot hours of the day from excessive evaporation. Whilst, therefore, in the dewy night these plants are indebted to their salt crusts for water, they are in the day-time preserved from desiccation by the action of the same contrivance. ^ The salt incrustations which were removed from plants of FranJcenia hispida, collected on a Persian salt-steppe, consisted principally of common salt (chloride of sodium). They contained in smaller quantities, gypsum, mag- nesium sulphate, calcium chloride, and magnesium chloride. ABSORPTION-CELLS ON LEAVES. 237 It is also worthy of mention that papillae are developed near the absorption- cells, with a view to the retention of the salt crystals, similar to those which hold the calcareous incrustations on the leaves of saxifrages and Acantholimon. The leaves of plants covered with crystals of salt are also for the most part furnished with little bristles, to which the salt adheres so firmly that it is not i-eadily detached, even by violent shaking. But however striking the analogy may be between the development and significance of lime crusts and salt crusts, there is the essential difference that the former have not, like the latter, the power of attracting moisture from the air. And on this particular stress must be laid. In the broken and hillocky tracts on the shores of salt -lakes or of the sea, where tamarisks and frankenias are especially wont to live, the sandy ground dries up to such an extent in the height of summer that it is scarcely conceivable how plants growing in it are able to preserve their vitality. The proximity of the sea has no immediate effect on the moisture of the ground in such situations. The sea- water does not penetrate into the ground far beyond the high- water line, and it is out of the question that the layers of soil serving as substratum to the frankenias and tamarisks should be irrigated by subterranean water. When in summer there is an absence of rain for months together, these plants — even though in close proximity to the sea — would necessarily perish of drought. Only the circumstance that they turn to account the moisture of the atmosphere by means of the excreted salts renders it possible for them to flourish in these most inhospitable of all inhospitable sites. Many plants which are periodically exposed to great dryness have the tips of the ! teeth on the leaf-margins thickened into little cones or warts. They also glitter i somewhat and at times are sticky. The glitter and viscidity are due to a resinous I slimy substance, which often contains sugar and tastes sweet. This substance I covers the teeth and sometimes spreads from the teeth inwards to a great dis- tance over the face of the leaf in the form of a delicate film-like varnish. The greatest resemblance exists between this varnish (sometimes known as "balsam") and the secretions of the glands on the leaves of the Alpine Rose and of the glandular hairs on those of Centaurea Balsamita. It is excreted by special cells, which are intercalated in the epidermis of the foliar teeth, and are at once marked out from the other cells of the epidermis by the facts that their protoplasm is of a brownish colour and that their external walls are easily permeable by water. The excretion of the varnish-like layer takes place at a time when the entire plant is dis- tended with sap, chiefly, therefore, in the spring. When summer is at its height the varnish dries and thenceforward affords an excellent preservative from the risk of too much evaporation from the cells it covers, and especially from those situated on the teeth of the leaves by which it was excreted. But if this dried film of varnish is wetted it saturates itself quickly with water and renders moisture accessible to the cells beneath it. Thus its value is similar to that of the crusts of lime and salt on the leaves of the plants above described. When moist it effects the absorption of water, when dry it guards against desiccation. 238 ABSORPTION-CELLS ON LEAVES. The reason for the contrivance just described being exhibited especially by the maro-inal teeth of the leaf, lies in the fact that dew is deposited particularly at those spots. If one looks at the leaves of the dwarf almond and plum trees in the steppe-districts, after clear summer nights, one finds a dewdrop suspended to every tooth on the margins; but by noon all the teeth are dry again and protected from loss of water by the coat of varnish. Moreover, not steppe-plants alone, but very many plants which grow in poor sandy soil on the banks of streams and rivers, exhibit this contrivance for the direct absorption of water from the atmosphere. Instances are afforded by the Sweet Willow, the Crack- willow, Poplars, the Guelder- rose, the Bird-cherry, and many others. It is at once evident that this contrivance is observed chiefly on the leaves of trees, shrubs, and tall herbs, whilst incrustations of lime occur only on shorter plants with rosulate leaves spread out on the ground, or with rigid acicular leaf-structures. The grounds of this distinction may well reside in the fact that the weight of a crust of lime is many times as great as that of the dry film of varnish. A load capable of being borne without hazard by the leaves of a Statice plant, they being spread out on the ground, or by the rosettes of Saxifraga Aizoon, would be unfit for the leaves of a Cherry or Apricot tree, or for those of the Sweet Willow, or the Crack- willow ; indeed the branches of these trees would break down under the burden if their leaves were incrusted with lime. In many cases only a few of the marginal teeth of the leaf are transformed into absorbent apparatus, and special contrivances then always exist to convey rain and dew to those teeth. The Aspen (Populus tremula) serves as a very good example of this. This tree has, as is generally known, two kinds of leaves. Those arising from the branches of the crown have long petioles and laminae of roundish outline and with somewhat sinuate margins; those which are borne by the radical shoots have shorter stalks and larger sub-triangular laminae sloping outwards; and the whole leaf is so placed and its margin so curved as to oblige the rain which strikes the upper surface in its descent to flow down towards the petiole (see fig. 55^). Now, situated exactly on the boundary of lamina and petiole are two cup-shaped structures (fig. 55^) originating from the lowest teeth of the leaf, and so arranged that every drop of rain descending from the lamina must encounter their shallow cavities and fill them with water. These cups are brown in colour and the size of a grain of millet; and the cells of their epidermis are furnished with a thick cuticle. Only the cells lining the shallow depression of each cup have thin walls, and they excrete a sweet-tasting, slimy, resinous substance which in dry . weather films over the cavity like a varnish, and protects, at all events, the cells lying beneath it against an injurious desiccation. When, however, this coat is itself in contact with water it swells up, and the moisture is then absorbed by the cells in the pit-like depression and is transmitted to the vessels running underneath the cups (see fig. 55^). A number of tall herbs, principally of the group of Compositse, have, like the Aspen, leaf -teeth which are developed at the part where petiole and lamina join and act as organs of absorption. In some, besides, the margin of the green lamina extends in the form of a narrow ridge down the pale canaliculate petiole; and, when ABSORPTION-CELLS ON LEAVES. 239 this is the case, teeth of the kind are found on this narrow green ridge which runs along the groove. In TeleJcia, a handsome herbaceous plant of wide distribution in the south-east of Europe, these teeth — conical or club-shaped — springing from the margin of the petiole-groove are incurved, and are in general so placed that their blunt apices project into the groove. But precisely on these obtuse tips of the teeth are situated cells with very thin outer walls easily permeable to water, and having contents with a strong attraction for it. Thus, as soon as the groove of the Fig. 56.— Water-receptacles. 1 In a Teasel, Dipsacus laciniatus. * In the American Silphium per/oUatum. petiole is filled with rain, collected from the surface of the leaf, the tips of the conical teeth are moistened, and they suck up the water. Lastly, we have to mention the curious receptacles appertaining to foliage- leaves in which water from the atmosphere accumulates and continues to stand for weeks without being protected from evaporation by the excretion of special substances. Any region or portion of the leaf may participate in their construction. In Saxifraga peltata the lamina is shaped like a shield and forms a shallow plate with the concave surface turned to the sky. In the Cloud-berry (Rtohus Chamce- morus) the formation of basins is brought about by the margins of the reniform lamina being superimposed over one another as if to make a spathe. In the various species of Winter-green, especially in Fyrola uniflora, the pale cauline leaves, 240 ABSORPTION-CELLS ON LEAVES. inserted above to the green leaves, are metamorphosed into little saucers. In one species of Teasel, Dipsacus laciniatus (see fig. 56^), and in the North American Silphium perfoliatiir)% (fig. 56 -) the two sheathing portions (vaginae) of every pair of opposite leaves are connate and form comparatively large and deep funnel-shaped basins, from the middle of which rises the next higher internode of the stem. In several Meadow-rues (Thalictrum galioides and T. simplex) the secondary leaflets, which are opposite one another and shut close, almost like the valves of a mussel, are moulded so as to form cavities for the retention of water, and in many Umbelliferae, such as Heracleum and Angelica, the vagina of each individual leaf is ventricose or inflated, thus forming a sac enveloping the segment of the stem which stands above it. These basins, saucers, and dishes are always so placed, relatively to their surroundings, that the water derived from rain and dew is directed into them from the surfaces of the leaves, or by the segment of the stem which rises fx'om their centres, and thus it is that the depressions are filled. Whether in all cases much of the water accumulated is absorbed is certainly open to doubt. In the case of the leaves of the Alchemilla (fig. 52 ^), which exhibit the phenomenon so conspicuously that the plant has received the popular name of Dew-cup; the absorption of water is, at anyrate, very inconsiderable, and here the retention of the dew secures advantages of a difierent kind to which we shall presently have occasion to return. On the other hand, it is established that in the case of basins belonging to tall herbaceous plants, particularly such as grow on steppes and prairies where often no rain falls for a long interval, the water collected is absorbed by the glandular hairs and thin-walled epidermal cells developed within them. The fact of this absorption may be proved by a very simple experiment. Let a stem of the Silphium, represented in fig. 56 ^ be cut ofl" beneath the pair of connate leaves, which form a basin by their union, and let the cut surface 1 e closed with sealing-wax, so that no water can be taken up by the stem from below. If the water accumulated in the basin is now emptied out, the leaves shortly become flaccid and droop; but if the basin is left full of water, the leaves preserve their freshness a long while and do not begin to wither until all the water has evaporated and disappeared from the basin. If oil is poured upon the collection of water in the basin, so that evapora- tion from the latter is impeded, a constant diminution of the water in the basin is observed notwithstanding; this leads to the conclusion that the water in question is really taken up by the absorption-cells at the bottom of the basin and conveyed to the tissue of the leaf. The first thing that strikes one on surveying once more all the plants possessing on their aerial organs special contrivances for water-absorption is that a large proportion of them have taken up their abode in swamps and on the banks of rivers and streams, or if not there, at all events in situations where no danger exists of the ground being thoroughly dried up. No doubt this appears to be inconsistent. How are we to explain the fact that Gentianese, ashes, willows, alpine roses, bog-mosses, &c., ax-e still in need of water from the atmosphere, when they all ABSORPTION-CELLS ON LEAVES. 241 grow either in damp meadows, peat-bogs, on the borders of never-failing springs, or in ever-moist ravines, where their requirements in respect of nutrient water and imbibitions water can be supplied all around by means of the roots? A glance at the company in which these plants occur may perhaps lead to a solution of the problem. In the damp meadows and along the margins of springs where gentians, the Sweet- willow, and plants of that kind are found, the Butterwort (Pinguicula), which has been described in earlier pages amongst carnivorous plants, is never absent; whilst wherever the pale cushions of the Bog-moss spring, there also the Sun-dew is certain to spread out its tentacles for the capture of prey. With reference to community of site the assumption is warranted that all these plants which flourish under identical conditions of life endeavour to acquire the same material by means of their aerial parts. Now, this material cannot well be other than nitrogen, of which they do not find a sufficient store in the substratum. What then is more natural than that those plants, which are not adapted to the capture of animals, should use their aerial organs, when these are moistened with rain or dew, to take up direct nitric acid and ammonia, which are contained — though in small traces only — in the atmospheric deposits, instead of waiting till compounds of such great importance to them penetrate into the ground where they I may chance to be detained at spots whence the roots could only obtain them after long delay and by a highly complicated process? When one considers that plants, growing amid the sand and detritus of steppes, on ledges, and in crevices of steep rocks, or epiphytic on the bark of trees, are also able to acquire little or no nitrogenous food from the substratum by means of their roots, their especial equip- ment with apparatus for the absorption of atmospheric water becomes explicable on the ground of the latter being the medium of solution and transport of nitrogenous j compounds. In the case of epiphytes and of plants growing on steppes or rocks, there is the additional consideration that a supply of pure water, supplemental to that which can be withdrawn from the substratum, must be very welcome to them in dry weather, and that at such times it is a great advantage for the atmospheric water to be absorbed directly by the aerial organs instead of reaching them in a roundabout manner through the substratum. If this idea is justified, the atmospheric moisture taken up by the aerial organs with the help of the above-described contrivances, would be of value to the plant chiefly in being a carrier of nitrogenous compounds, and in this acceptation would have to be looked upon as water of imbibition. Whether it is also used, at least in part, as food-material can neither be asserted nor controverted. A separate absorp- tion of water which serves only for motive power, and of that which is in addition employed in the construction of organic compounds does not take place in a plant, it is not possible to make any a priori statement concerning the moisture taken up, as to which part it has to play in the plant. Most probably the allotment of functions is not at all uniform, but varies considerably according to conditions of time, place, and requirement. On a former occasion it has been mentioned that small animals are not 242 ABSORPTION-CELLS ON LEAVES. infrequently killed accidentally in the water filling the larger kinds of basins formed as parts of foliage-leaves, that pollen, spores, and particles of earth also are blown by the wind into these basins, and that, after the ensuing solution and decomposition of the organic and mineral bodies in question, the water exhibits a brownish colour and contains organic compounds as well as food-salts in solution. It is not necessary to repeat that these compounds are able to pass into the interior of the plant with the water through the action of the absorption-cells which are never absent from the bottom of the basins; but it seems proper to consider specially in this connection the most conspicuous cases of the phenomenon which have been observed. The greatest quantity of matter, dissolved and undissolved, is found in the flat, saucer-shaped laminss of Saxifraga peltata, which grows on the sites of springs in the Sierra Nevada of North America. The water in these saucers is sometimes coloured quite a dark brown by the presence of decayed beetles, wasps, centipedes, fallen leaves, and animal excreta; and when it evaporates a regular crust is left behind at the bottom of the reservoir. Three days after rain I still found in the inflated vagina of Heraclev/m palmatwm, a species of cow-parsnip, a pool of brown water 2 cm. deep, and at the bottom a deposit of blackish, oily mud in which the remains of decayed earwigs, beetles, and spiders, were still recognizable. The same thing is observed in the cisterns of Bromeliaceee and in the water-basins of Dipsacus laciniatus and Sil2:)hiwm perfoliatum (fig. 56), and it is interesting to find there are cells also at the bottom of the basins of the Dipsacus in question from which protoplasmic threads radiate forth, as in the case of the chambers of the Tooth wort, and that numberless putrefactive bacteria always make their appearance in the water in these basins. The quantity of organic residue is less considerable in the saucer-shaped leaves of pelargoniums, but, on the other hand, earthy particles are frequently met with in them to such an extent that, when the water has evaporated, the concave surface of the leaf is covered with an ashen-gray layer of earth. Observations of this nature establish the conviction that no sharp line of demarcation exists in respect of the absorption of water either between carnivorous plants and land plants, or between land plants and saprophytes, or between saprophytes and carnivorous plants; and they lead further to the conclusion that water, mineral food-salts, and organic compounds are susceptible of being taken up not only by subterranean but also by aerial absorptive apparatus. LICHENS. 243 6. SYMBIOSIS. Licheus. — Cases of symbiosis of Flowering Plants having green leaves with the mycelia of Fungi destitute of chlorophyll. — Monotropa. — Plants and Animals considered as a vast symbiotic community. LICHENS. In describing the vegetation of a limited area botanical writers are apt to desig- nate the various species of plants as "denizens" of the country in question. The conditions under which the plants live are likened to political institutions, and the relations existing amongst the plants themselves are compared to the life and strife of human society. By no means the least important factor in the suggestion of these analogies is the circumstance that often as a matter of fact one has opportunities of seeing how the species of plants which live together in a locality are dependent in various ways upon one another; how they exist in continual con- flict for the food, the ground, for light and air; how some are preyed upon and oppressed by others, whilst others are supported and protected by their neighbours ; and how, not infrequently, quite different species join together in order to attain some mutual advantage. As regards the preying of one upon another the subject has been treated in detail in a previous chapter, and it was also stated then that the term parasite can only be applied to those plants which withdraw materials from the living parts of other organisms without rendering a reciprocal service in return. The host attacked by a parasite supplies food and drink without being in any way compensated. One might suppose that nothing would be simpler and easier than to ascertain the existence of this relationship, and yet many difficulties are encountered in the determination of parasitism in individual cases. The main difficulty is due to the fact that one cannot always say with certainty whether the host does not perhaps get some advantage from the parasite which drains its juices. Should this be the case, however, the latter would be no longer a parasite, and the relationship between the two would rather be that of simple commerce and mutual assistance, an ami- cable association for the benefit of both. Whilst discussing the second series of parasites, the fact was mentioned that the plants upon which the various species of Eyebright fasten their suckers suffer no apparent injury as a consequence of this connection. The rootlet organically united to the suckers does, it is true, die away in the autumn; but the Eyebright also withers at that season, and it is not inconceivable that the useful substances existing in the green leaves of the Eyebright may be transferred, shortly before the latter withers, to the host-plant and deposited there at a convenient time in the permanent part of the root as reserve-material, and that in this way the host-plant ultimately derives benefit from the so-called parasite. The idea here suggested as a possibility for the case of Eyebright and the grasses connected with it is an ascer- tained fact in the case of some other plants. For plants are known which unite to 244 LICHENS. form a single organism and thenceforward so co-operate in their functions that ultimately both derive advantage from the arrangement. The one takes food-stuffs from the substratum and from the air and transmits them to the other; whilst, in the green cells of the other, the raw material is worked up, under the influence of sunlight, into organic compounds. The organic compounds thus created are used by both for the further production of organs, and therefore a connection such as this must be looked upon as a true case of symbiosis, i.e. associated existence for purposes of nutrition. The first place amongst social communities of the kind must be assigned to Lichens, a section of Cryptogams possessing an extraordinarily large number of species and differentiated into thousands of forms, representatives of which are Fig. 57.— Gelatinous Lichens. Ephebe Kerneri; x450. 2 Collema pulpostim ; natural size. 3 Section through Collema pulposum; x450. everywhere distributed, from the sea-shore to the highest mountain peaks yet scaled by man, and from the tropics to the arctic and antarctic zones. The partners in the Lichen communities appear to be, on the one hand, groups and filaments of round, ellipsoidal, or discoid green cells belonging to plant species included under the general name of Algse; and, on the other hand, pale, tubular cells or hyphge, which are destitute of chlorophyll, and pertain to species of plants comprised under the general n'ame of Fungi (see fig. 58). The form assumed by a large proportion of these lichens is that of incrustations on stones, earth, bark, or old wood-work ; the entire structure of the lichen is either ensconced and imbedded in the depressions of weathered surfaces of stone, or else between the cell-walls of dead fragments of wood and bark, so that it often happens that attention is only drawn to its presence by the altered colour of the substratum, or by the fructifications which lift their heads above the substratum. Lichens of the kind are termed Crustaceous Lichens, and the wide -spread Graphic Lichen (Lecidea geograpJdca) may serve as an example. A second great group nearly allied to the first is that of Foliaceous Lichens. The form of the 245 vegetative body in these is best compared to the foliage-leaves of the Curled ^lint, with their corrugated or sinuate margins, or to those of Malva rotundifolia. It may also be described as a number of lobes radiating irregularly and bifurcating repeatedly, and only lightly joined to the substratum by root-like fringes, and there- fore capable of being readily loosened and detached. The light-grey Farmelia saxatilis, which bear brown saucer-shaped fructifications, may be taken as a repre- sentative of these Foliaceous Lichens. The Fruticose Lichens are distinguished as a third group in which the thallus rises from the ground in the shape of a shrub, whilst the cylindrical, fistular, and ligulate stemlets, which ramify profusely, are only adherent to the substratum by a very small surface at the base. With these are associated the Beard Lichens, which hang down from the bark of old trees in the form of pale, copiously-branched filaments. Lastly, there is a fifth group, the Fig. 58.— Fruticose and Foliaceous Lichens. 1 Stereocaulon ramulosum in conjunction with Scytonema; x650. 2 Cladonia furcata with Protococcus; x950 ^ Coccocarpia molybdoea; section, x 650 (after Bornet). Gelatinous Lichens, which when moistened look like dark, olive-green, or almost black lumps of wrinkled and wavy jelly or as if composed of variously-divided bands and strips packed together into little cushions. In the gelatinous expansions last mentioned the algal cells are arranged in moniliform rows and are interwoven with the hyphal filaments of the fungus throughout the entire thickness of the thallus, as in Collema pulposum (see fig. 57 ^ and 57^), or else they form regular ribbon-shaped double rows, interwoven with few hyphse, as in Ujjhebe Kerneri (see fig. 57 ^). In crustaceous, foliaceous, and fruticose lichens, the algal cells constitute a disorderly heap and are crowded together in the middle stratum of the thallus, where they are imbedded between an upper and a lower layer of densely felted hyphal threads, as in Coccocarpia molybdcea (fig. 58^). Seeing the wide distribution of lichens it must be assumed that both partners occurring in the lichen-thallus are able to range about with extraordinary ease and latitude. When one observes how patches of the most various lichens are produced in a few years after a landslip on the freshly-broken surfaces of the stones which 246 LICHENS. have fallen down into the valley beneath, one can only explain the phenomenon by supposing that the algal and fungal cells concerned have been blown together, and that the opportunity has been afforded them on the blocks of stone of contracting a union. Now, so far as regards one of the two partners, viz.: the one devoid of chlorophyll, and known as a fungus — the idea that everywhere in the air spores of fungi are swarming about is so familiar to us that the supposition of an occasional stranding of individual spores, which are being blown about by the wind, upon the moist broken surfaces of stones can encounter no opposition. Respecting those spores in particular which are ejected from the aerial fructifications of lichens, the discussion of their life-history and distribution must of course be reserved for a later section; but it is necessary to make here the one statement that provision exists for the most profuse and distant dissemination of these spores. Thus, in the case of one of the partners, there is no difficulty in realizing its ubiquity. But when one comes to the Algae, the name at first calls up to mind the green filaments which occupy our pools and ponds, or the brown wracks and red Florideas of the sea-shore, and we ask ourselves how it can be possible for these plants to occur on fractured surfaces of stone, especially on the debris of mountain sides. Indeed, it is certainly not Algse of these kinds that take part in the construction of Lichens. The name Algae is properly only a general name for all Thallophytes containing chlorophyll, and it is applied to many small organisms besides those mentioned above, namely, to numbers of Nostocineae, Scytonemeae, Palmellaceae, Chroolepideae, and these are the kinds which fall in with the cells of fungi and form lichens in conjunction with them. Owing to their minute size, they are apt to escape observation, and, in general, only attract attention when myriads of them clothe the bark of trees, cliffs, stones, or earth. In these situations they need but little moisture, and it is not necessary for any of them to live under water like other algae^; they become desiccated without sustaining the slightest injury and make their appearance on the substratum occupied by them at the first stage of their development, as powdery coats, and, in this condition being extremely light, are liable to be blown away by a wind of moderate strength, and so distributed over mountain and valley. That this dissemination is not merely hypothetical but an actual fact has been susceptible of easy proof by the following experiment, made in a mountain-valley in the Tyrol. A plane surface covered with white filter-paper, which was kept moist, was exposed to a south wind; in the course of a few hours numerous particles, like dust, adhered to the paper, and amongst them cell-groups of Nostocineae and others of the above-mentioned algae occurred regularl}?-, in addition to organic fragments of the most various kinds, such as pollen-grains and spores of all sorts of mosses and fungi. All these bodies were deposited in the little depressions on the sheet of paper, and in the same way they rest in the grooves, cavities, and cracks in the surfaces of stone, bark, and old wood-work, where they succeed in reaching a further development as soon as the requisite quantity of water is provided. Now, if at these places the little algal cell-groups meet with hyphae belonging to the LICHENS. 247 other potential partner, the latter embrace and enmesh them, as is shown in the above figures, and thus is produced the confederacy called a Lichen. The member destitute of chlorophyll takes up nutriment from the external environment; it possesses, in particular, the property of condensing aqueous vapour, and has, besides, the power of bringing the solid substratum partially into solution by means of excreted substances; it effects adhesion to the substratum, and, in a majority of cases, determines the form and colour of the lichen-thallus as a whole. The second member, whose cells contain chlorophyll, undertakes the task of producing organic matter, under the influence of sunlight, from the materials conveyed to it; by this means it multiplies the number of its cells and increases in volume, whilst, at the same time, it yields to its mate so much as is necessary in order to enable the latter to keep pace with it in growth. The number of algse which enters into a partnership of this kind is, in any case, much less considerable than that of the fungi, and it must be assumed that one species of alga may unite with the hyphse of different lichen-fungi. The extreme variety, moreover, in the combinations of the two sorts of confederate occurring on a very small area is obvious from the circumstance that it is not rare for half a dozen different species of lichen to spring up side by side on a patch of rock no bigger than one's hand. Whether they all achieve an equally hardy development, or whether some perchance are not crowded out and overgrown by others depends on various external conditions — on the chemical composition of the substratum, and particularly on the conditions of moisture and illumination of the site in question. Lichens are very sensitive in this respect, and the different sides of a single rock often exhibit quite different growths of lichens. A very instructive example of this is afforded by a marble column near the famous castle of Ambras in Tyrol. This column is octagonal, and has been standing in its place for more than two hundred years, with all its sides exposed to wind and weather. Lichens have settled on all the eight faces, and, indeed, are present in such abund- ance that the stone is quite covered by patches the size of a man's hand. Many of these growths are but poorly developed, and not susceptible of being identified with certainty; but altogether on this column there must be over a dozen different species, the germs of which can only have been brought by winds. These species are, however, by no means uniformly disposed; some prevail on one side, some on another, and a few are confined exclusively to one of the eight faces. Of three species of Amphiloma, the one named A. elegans is restricted to the warmest side, i.e. the face exposed to the south-west; a second, Amphiloma murorum, is to be seen on the upper part of the southern face; whilst Amphiloma decipiens occurs on the same face, but only near the ground. On the side with a northern aspect Endocarpon miniatum predominates, and on the north-west face Calopisma citrinum and Lecidea are the prevailing forms. What thousands of spores and algal cells must have been blown on to this pillar to enable all these combinations to arise! What complex processes must have gone on before the selection of lichens best adapted to each different quarter 248 LICHENS. of the compass was effected on this little marble column! It is necessary to add, however, that lichens growing on stone, bark, or any situation of the kind do not in all cases owe their original appearance on the substratum to a fresh union of Algge and Fungi, but that there is a second mode of distribution of lichens. This method consists in the transportation by air-currents of already completed social colonies to places often situated at a great distance from the spots where the initial union between Al^a and Fungus was contracted. The process is as follows: — in the interior of an old, large, and fully developed lichen-thallu3 certain groups of cells separate from the rest, each group consisting of one or more green algal cells enmeshed in a dense weft of hyphse. When a sufficient number of these daughter-associations has been formed the thallus of the parent lichen is ruptured and the little miniature social-groups, which are termed "soredia", come to the surface. To the naked eye a single soredium is only visible as a bright dot, but all together they have the appearance of a mass of powder or meal lying loosely upon the old lichen-thallus. In dry weather this mealy efflorescence is easily blown away with other organic particles. If, then, a soredium thus removed comes to rest in the crack of a rock or on any suitable substratum, the alga and hyphse composing it continue to develop, and the organism grows into a larger lichen-thallus, which is able to repeat the process just described. In regions where lichens abound, soredia of the kind are found regularly amongst the elements of the organic dust, and occur, indeed, mixed with fungal spores and algal cells, so that it certainly happens not infrequently that two spots close together in the same cranny of stone exhibit both sorts of lichen-growth, the one newly produced by the concurrence and union of algal and fungal cells, the other a daughter- association which has arisen from an old lichen, as a soredium, and is continuing its development. Another case of symbiosis allied to that of lichens is manifested by certain Cryptogams which live socially together under water and have received the systematic names of Mastichonema, Dasyactis, Enactis, &c. In them also a plant containing chlorophyll, and belonging to the group of Nostocineoe, appears as one member of the partnership; whilst the second is some species of Lej^tothrix or Hypheothrix. The green moniliform rows of cells of Nostocinese are enmeshed and wrapped round by the delicate, filamentous cells devoid of chlorophyll of the Leptothrix or Hypheothrix; and later, by repeated processes of division, whole colonies of green cell-filaments ensheathed in this manner are produced, which to the naked eye appear as small soft tufts, usually clinging to porous limestone in the spray of waterfalls. In many cases the filaments destitute of chlorophyll rest upon the moderately thickened cell-membranes of the green algte, whilst in other cases they insinuate themselves into the thick cell-membranes, permeate them with their webs, and form in conjunction with them the sheathing envelope. SYMBIOSIS OF PHANEROGAMS AND FUNGI. 249 SYMBIOSIS OF GREEN-LEAVED PHANEROGAMS WITH FUNGAL MYCELTA DESTITUTE OF CHLOROPHYLL.— MONOTROPA. Another instance of symbiosis is observed to exist between certain flowering plants and mycelia of fungi. The division of labour consists in the fungus-mycelium providing the green-leaved Phanerogam with water and food-stufFs from the ground, whilst receiving in return from its partner such organic compounds as have been produced in the green leaves. The union of the two partners always takes place underground, the absorbent roots of the Phanerogams being woven over by the filaments of a mycelium. The first root that emerges from the germinating seed of the phanerogamic plant destined to take part in the association descends into the mould still free from hyphse; but the lateral roots and, to a still greater extent, the further ramifications, become entangled by the mycelial filaments already existing in the mould or proceeding from spore-germs buried there. Thenceforward the connection continues until death. As the root grows onward, the mycelium grows with it, accompanying it like a shadow whatever its course, whether the root descends vertically or obliquely, or runs horizontally, or re-ascends, as is sometimes necessary when it happens to be turned aside by a stone. The ultimate ramifications of roots of trees a hundred years old, and the suction-roots of year-old seedlings, are woven over by mycelial filaments in precisely the same manner. These mycelial filaments are always in sinuous curves and intertwined in various ways, so that they form a felt-like tissue, which looks, in transverse section, delusively like a parenchyma. As regards colour the cell-filaments are mostly brown, sometimes they are almost black, and it is rare for them to be colourless. The epidermis of many roots is covered as if by a spider's web, whilst the hyphse form a complex tangle of bundles and strands broken here and there by open meshes through which the root is visible. In other cases an evenly woven but very thin layer is wrapped round the root; and in others, again, the fungus-mantle forms a thick layer which envelops uniformly the entire root (see fig. 59). Here and there the hyphae insinuate themselves also inside the walls of the epidermal cells, and the latter are permeated by an extremely fine small-meshed mj^celial net (see fig. 59^). Externally the mantle is either fairly smooth and clearly marked off from the environment, or else single hyphse and bundles of hyphae proceed from it and thread their way through the earth. When these branching hyphae are pretty equal in length they look very much like ordinary root-hairs. And they not only resemble them, but assume the function of root-hairs. The epidermal cells of the root, which would in an ordinary way act as absorption-cells, being inclosed in the mycelial mantle cannot exercise this function, and have relegated the business of sucking in liquid from the ground to the mycelium. The latter undoubtedly acts as an absorptive apparatus for the partner on whose roots it has established itself; and the water in the soil, together with all the mineral salts and other compounds 250 SYMBIOSIS OF PHANEROGAMS AND FUNGI. dissolved in that water, are caused by the mycelial mantle to pass from the surrounding ground into the epidermal cells of the root in question, and thence onward, ascending into axis, branches, and foliage. Thus the fungus-mycelium not only inflicts no injury on the green-leaved plant by entering into connection with its roots, but confers a positive benefit, and it is even questionable whether a number of green-leaved plants could flourish at all without the assistance of mycelia. The experience gained in the cultivation of those trees, shrubs, and herbs, which exhibit mycelial mantles on their roots, does not, at any rate, lead to that conclusion. Every gardener knows that attempts to rear the various species of winter-green, the bog-whortleberry, broom, heath, bilberries, cranberries, rhododendrons, the spurge-laurel, and even the silver-fir and 2 // n\ Fig. 59. 1 Roots of the White Poplar with mycelial mantle. 2 Tip of a root of the Beech with closely adherent mycelial mantle; xlOO (after Frank). » Section through a piece of root of the White Poplar with the mycelium entering into tlie external cells; X480. the beech, in ordinary garden soil are not attended with uniform success. Therefore, as is well known, soil consisting of vegetable mould from the top layer of earth in woods or on heath is chosen for the cultivation of species of the genera Erica, Daj^hne, and Rhododendron. But it is not even every kind of forest- or heath- mould that can be made use of. When earth of that nature has been quite dry for a long time it is no longer fit for this purpose. On the other hand, it is known that the above-mentioned plants should be transplanted from their forest-home with the soil still clinging to the roots, and it is also laid down as an axiom that the roots of these plants should not be exposed and should be cut as little as possible. The following reasons account for all this. Firstly, fresh earth from a heath, or mould recently dug from the ground in a wood, contains the mycelia still alive, whereas in dry humus they are already dead; secondly, the mycelia woven round the roots are transferred together with the balls of earthy matter suspended to them into the garden; and, lastly, any considerable clipping of the roots would remove the ultimate ramifications which are furnished with the absorbent mycelial mantle. The failure of all attempts to propagate the oak, the beech, heath, rhododendron, winter-green, broom, or spurge-laurel, by slips or cuttings, if the shoot which is cut SYMBIOSIS OF PHANEROGAMS AND FUNGL 251 off and used for the purpose is put into pure sand, is explicable in the same way. Limes, roses, ivy, and pinks, the roots of which possess no mycelial mantle, are notoriously propagated very easily by putting branches cut from them into damp sand. Rootlets are at once produced on those parts of the branches which are buried in the sand, and their absorption-cells carry on the task of taking up nutriment from the ground. But though cuttings of oak, rhododendron, winter- green, bog- whortleberry, and broom strike root, no progress in their development is to be observed, because the superficial cells of the rootlets, in these cases, have not the power of absorbing food when they are not associated with a mycelium. It is only when the slips from these plants are put into sand with a rich admixture of humus, the latter having just been taken from a wood or heath and containing the germs of mycelia, that some few are successfully brought to further development. The result is even then often not assured, and the cuttings of several of the plants enumerated die even in sand mixed with humus before they have produced rootlets. Seeing also that the result of attempts to rear seedlings of the beech and the fir in so-called nutrient solutions, where there could be no question of any union with a mycelium, has been that the plantlets dragged on a miserable vegetative existence for a short time and ultimately died, we have good grounds for assuming that the envelope of mycelial filaments is indispensable for the Phanerogams in question, and that the prosperity of both is only assured when they are in social alliance. The facts ascertained in cases of analogous relationship lead one to expect that the fungus-mycelia also derive some advantage from the flowering-plants, the roots of which they clothe, and to which they render the service of acting as absorption- cells. The benefit in question is undoubtedly the same as that derived by the hyphae of a lichen- thallus from the en woven green cells. The mycelial mantles withdraw from the roots of the Phanerogams the organic compounds which have been elaborated by the green leaves in the sunshine above-ground, and which are conducted thence to all growing parts, that is to say, downwards as well as in other directions, to the tips of the swelling and elongating roots. According to this, therefore, the division of labour between the members of the alliance for joint nutrition consists in the mycelium supplying the green-leaved plant with materials from the ground, and the green-leaved plant supplying the mycelium with substances which have been worked up above-ground in the sunlight. The range of species which live in a social union such as is here described is certainly very large. All Pyrolacese, Vaccineae, and Arbutese, most, if not all, Ericaceae, Rhododendrons, Daphnoidese, and species of Umpetrum, Fpacris, and Genista, a great number of Conifers, and apparently all the Cupuliferae as well as several Willows and Poplars are dependent for nutrition on the assistance of mycelia. We find, too, that this condition recurs in every zone and in every region. The roots of the Arbutus on the shores of the Mediterranean are equipped with a mycelial mantle in precisely the same manner as those of the low -growing Whortleberry of the High Alps. 252 SYMBIOSIS OF PHANEROGAMS AND FUNGI. Special importance is given to the social life by the fact that the chief species of Phanerogams partic'pating in it are of gregarious growth and cover whole tracts of country, forming boundless heaths and measureless forests, as, for instance, the various heaths, the oak, the beech, the fir, and the poplar. The conception of this subterranean life affecting every moorland and vast timbered tract is one full of wonder and interest. We can now see why it is that the ground in woods is the abode of such a profusion of fungi. No doubt some of these fungi draw their nutriment exclusively from the store of dead plant-organs accumulated there; but others, as certainly, are in social connection with the living roots of green-leaved plants. It is true we cannot yet state precisely what are the species of fungi which contract this sort of union, or whether generally a definite elective affinity exists between certain fungi and certain green-leaved plants. There is much in favour of this supposition in a few cases: but, on the other hand, it is very unlikely that each of the various Phanerogams occupying a limited area of ground in a pine-forest, where a few square meters of earth contain so many tangled roots belonging to pines, spurge laurels, bilberries, cranberries, heath, and winter-green, that they can only be separated with difficulty, should select from the great host of fungi growing in the forest a different partner. In instances of this kind it seems just to suppose that the mycelium of one and the same species of fungus enters simultaneously into connection with all or several of the plants growing close together; it is similarly probable that the mycelia of different species of fungi render to one and the same flowering-plant the service of absorption according to the locality in which it occurs. This surmise is supported by the fact that when certain species, brought from distant parts and regularly exhibiting mycelial mantles on the ends of their roots, are reared in our gardens and greenhouses from seed, they unite in these abodes with fungus-mycelia, which certainly do not exist in the regions where the Phanerogams in question grow wild. Thus, for instance, the roots of the Japanese tree, Sophora Japonica, and those of the Epacrideae of Australia, are found in European gardens in social union with fungi, which with us are native, but which certainly do not occur in Japan or Australia; and it is therefore scarcely open to doubt that the Sophora Japonica, to take one example, associates itself with different fungi in different regions. Now that the symbiosis of fungi devoid of chlorophyll with green-leaved Phanerogams has been discussed, we are for the first time in a position to deal with that most remarkable of all cases of food-absorption wherein the subterranean roots of a flowering-plant are completely wrapped in a mycelial mantle, whilst the parts which shoot up above ground bear no green leaves, and, in general, possess no trace of chlorophyll. Such is the case of Monotropa, the various species of which are intimately allied in the structure of flowers and fruit with the Primrose and Winter- green, and are met with scattered everywhere in shady woods. Their stems, which are from 10 to 20 centimeters in height and emerge from the mould of the forest- ground in summer time, are thick, fleshy, succulent, and profusely beset with SYMBIOSIS OF PHANEROGAMS AND FUNGI. 253 i membranous and transparent scales, and the extremity of each is bent back like a hook. The cylindrical flowers are developed at the top of the stem with their open ends turned to the ground, and are half-covered by the scales. Everything i about this plant (stem, leaf-scales, and flowers) is of a pale waxen-yellow colour, and the general impression it produces is much more that of a Toothwort, or one of the colourless forest orchids, than of a species of primula or winter-green. Towards autumn, when ripe fruits have been produced from the flowers, the hitherto drooping extremity of the stem lifts itself into an upright position, whilst the entire aerial portion of the plant turns brown and dries up. Every disturbance caused by the wind, however slight, shakes out of the spherical fruits many thousands of tiny seeds as fine as dust, which, like the winter-green seeds, consist of I only a few cells, and do not admit of the recognition of any diflerentiated embryo j within them. Moreover, underground, the rhizomes, from which the small group of ] pale stems have arisen in summer, continue to live through the winter, and a j number of new buds are developed on them. On digging down to the hibernating i plant and removing the mould which conceals it, one finds at a depth of from 10 to 40 centimeters bodies like coral-stems consisting of dense masses of roots crowded together and ramifying multifariously. All the root-branches are short, thick, fleshy, and brittle, and are matted together to form turf -like masses, which are not infrequently interwoven with the rootlets of pines, firs, and beeches, and have all their interstices filled with humus. Each rootlet is enveloped, right up to the growing apex, in a thick mycelial mantle. The hyphal filaments of this mycelium do not penetrate into the tissue of the root of Monotropa, nor do they send any haustoria into the superficial cells of these roots. The hyphae and the epidermal cells of the root are, however, in such close and continuous contact that sections exhibit a complete continuity of the tissues. Monotropa is therefore only able to withdraw nutriment from the hyphal weft of the mycelium so far as its subterranean parts are concerned, and, seeing that it is quite destitute of chlorophyll, and its aerial stem and leaves display no trace of stomata, the possibility of creating organic matter and of adding in general to its substance by means of its aerial parts is excluded. It therefore receives all the materials of which it is constructed from the mycelium of the fungus, whilst it is not in a position to render anything in return to this mycelium that it has not previously derived from the latter. If the mycelium subsequently withdraws any materials whatever from the still living or decaying Monotropa, the process is only one of restitution and not of exchange. Thus, in this case, there can be no talk of reciprocity in the processes of nutrition or division of labour such as occurs when there is symbiosis. The Monotropa grows in height and in circumference entirely at the expense of the mycelium in which it is imbedded, so that we have here the remarkable phenomenon of a Phanerogam parasitic in the mycelium of a Fungus. We so often come across the converse process in our experience that we cannot easily familiarize ourselves with the idea of a flowering-plant draining the mycelium of a fungus of nutriment: nevertheless there is scarcely any other inter- 254 ANIMALS AND PLANTS A SYMBIOTIC COMMUNITY pretation possible in this case, for all the other hypotheses, — such as that Monotropa enters into connection with the roots of trees, or that it is parasitic in the first stages of development, but subsequently detaches itself from its host and becomes a saprophyte, — rest on inaccurate observations, and have long been disproved. As a parasite Monotropa ought to have been discussed at the same time as others in earlier pages, but it was not without intention that the description of this plant was reserved for this place, for it would have been difiicult to state and explain the method of nutrition exhibited by it before some previous knowledge of the curious phenomena of union of the mycelia of fungi with the roots of green-leaved Phanerogams had been acquired. ANIMALS AND PLANTS CONSIDERED AS A GREAT SYMBIOTIC COMMUNITY. If we look back at the cases of symbiosis already discussed and inquire what is their value, we find it consists in an integration of the functions of plants possessing chlorophyll and plants not possessing it. The reciprocity here implied is, however, at bottom, but a copy of the complementary interaction of plants and animals which takes place on a grand scale in the organic world. The associated plant, destitute of chlorophyll, in which capacity fungi are always the organisms concerned, really plays the same part in the social life as is taken by animals in the great economy of nature, and this is in harmony with the fact that in other respects as well fungi exhibit so many similarities to animals that in many instances one looks in vain for a line of division to separate them from animal organisms. Hence there is no need for surprise when cases come under observation wherein a quite unmis- takably animal organism enters, instead of a fungus, as one of the partners in a symbiotic community. Certain Radiolarise have small yellowish spots upon them, which were formerly held to be pigment-cells, but have proved to be little algse, with cells furnished with true chlorophyll. Similar properties are exhibited by the fresh-water polyp, Hydra, and by the marine sea-anemones. Small algse occur in social union with these also in the shape of cells with membranes made of cellu- lose and containing chlorophyll and starch-grains in their protoplasmic bodies. These algse are in no wise injurious to the animals with which they are associated; on the contrary, their presence is beneficial, their partners reaping an advantage from the fact that the green constituents split up carbonic acid under the influence of the sun's rays, and in so doing liberate oxygen which may be again taken in by the animals direct, and serve a useful purpose in their respiration and all the pro- cesses connected therewith. Conversely, the alga, in association with the animal's body, will derive a further advantage from the latter, inasmuch as it receives at first hand the carbonic acid exhaled by the animal in breathing. The small algae living socially with animals cannot be reckoned as parasites in any case, nor can the animals be looked upon as parasites of the algae, but we have here the phenomenon of mutual assistance and of a bond serving for the benefit of both ANIMALS AND PLANTS A SYMBIOTIC COMMUNITY. 255 parties, precisely similar to that noticed in the case of lichens and in the others which have been described above. Several of the liverworts which live as epiphytes on the bark of trees exhibit on the under surface of their leaflets (which are inserted on the stem in two rows, and are pressed flat against the bark) little auricular structures, and in species of the genus Frullania, these take the form of definite hoods or pitchers. The rain that trickles down the trunks of the trees, washing the bark and wetting the liver- worts in its course, fills the hooded receptacles referred to with water, and is retained longer in these protected cavities than anywhere else, if a period of drought ensues and the liverwort becomes dry again. Now these cowls are the abode of tiny rotifers (Callidina symbiotica and G. Leitgehii), which live on the organic dust brought thither with the water. In return for the peaceful home thus afibrded them in the hooded chambers of the leaves, the rotifers supply the liverworts in question with nitrogenous food. For as such must serve the matter excreted by the rotifers in the interior of the cowls. Without the intervention of the rotifers, the living organisms (Infusoria, Nostocinefe, and spores) contained in the water could not be converted into food by the liverworts, whereas the liquid manure arising from the Infusoria, Nostocineae, and spores, digested in the bodies of the rotifers, contains highly nitrogenous compounds, which are of great value to the liverworts in question, as indeed they are to all epiphytes living on the bark of trees. It stands to reason that the symbiotic liverworts and rotifers derive also a mutual advantage from the fact that the oxygen set free by the former comes into the possession of the rotifers and the carbonic acid emitted by the rotifers into that of the liverworts by the most direct method. Moreover, these cases of partnerships further remind us of other analogous rela- tions existing between plants and animals, which it is necessary to refer to now, although they cannot be treated in detail till later on. A great number of flowering- plants excrete honey into their flowers, and so attract flying insects to them, which supply themselves plentifully, and in their turn render to the plants they visit the service of transferring the pollen from flower to flower, thus making possible the development of fruits and fertile seeds. Certain small moths which visit the flowers of Yucca bring the pollen to the stigmas, and force it into the stigmatic orifices in order that mature fruits and seeds may be produced from the rudimentary fruits, a result which is indeed a matter of vital importance to these moths. For the moths lay their eggs in the carpels of Yucca, and from the eggs larvae are developed which live exclusively on the seeds of this plant. If the Yucca were not fertilized, and did not develop any fruit, the larvae would die of hunger. A similar phenomenon occurs in many other cases of the kind, where both plant and animal reap some benefit. On the other hand, in the formation of galls, which are produced by animals laying their eggs in particular parts of plants, the advantage (with few exceptions) is all on the side of the animals, and these gall- structures might most justly be placed by the side of parasitic structures. It is obvious from all this that such of the mutual relations of plants and of 256 ANIMALS AND PLANTS A SYMBIOTIC COMMUNITY. their relations to animals as are occasioned by the endeavour to acquire nutriment are extremely various and often linked together and complicated or deranged by one another in the most curious manner. Cases occur of a particular plant being socially connected with another, and at the same time also beset by vegetable and animal parasites. The absorption-roots of the Black Poplar are covered with a dense mycelial mantle, so that this tree is associated for purposes of nutrition with the fungus to which the mycelium belongs. Such parts of the roots of the Black Poplar as are left free from the mycelium are fastened upon by suckers sent forth by Tooth wort plants, which withdraw from the roots the juices absorbed by the latter from the earth through the instrumentality of the mycelial mantles clothing them. Meantime, in the cavities in the leaves of the Toothwort various small animals are caught and made use of as nitrogenous food. Again, the poplar-tree bears Mistletoe on its boughs, and its presence there is due to the missel-thrush. The thrush takes the Mistletoe-berries for food, and, in return, renders the plant the service of dispersing the seeds and establishing them on other trees. The para- sitic Mistletoe takes its liquid nutriment from the wood of the poplar- tree; but, on the other hand, its own stems are covered with lichens, and these lichens are them- selves a symbiotic community of algse and fungi. Within the wood of the poplar- stems spread the mycelia of certain Basidiomycetes (Panus conchatus and Poly- porus populinus), whilst the foliage-leaves are covered with a little orange-coloured fungus, Melampsora populina. In addition, no less than three gall-creating species of Pemphigus live on the leaves and branches of the Poplar, and a number of beetles and butterflies are nourished by them. Certain lichens, mosses, and liver- worts regularly settle on the bark of old trunks, and included amongst these may be the species of liverwort which is inhabited by rotifers. If all the plants and animals which live upon the poplar-tree, within it or in association with it, are counted, the number turns out to be not much fewer than fifty. ACTION OF PLANTS ON THE SOIL. 257 7. CHANGES IN THE SOIL INCIDENT TO THE NUTRITION OF PLANTS. Solution, displacement, and accumulation of particular mineral constituents of the soil owing to the action of living plants. — Accumulation and decomposition of dead plants. — Mechanical changes effected in the soil by plants. SOLUTION, DISPLACEMENT, AND ACCUMULATION OF PARTICULAR MINERAL CONSTITUENTS OF THE SOIL RESULTING FROM THE ACTION OF PLANTS. Reference was made in the preceding section to a marble pillar on the faces of which a dozen different lichens have settled in the course of centuries. I again I introduce to the reader's notice this unobtrusive monument in order to demonstrate I in its case the changes to which stone is subjected by the plants clinging to it or j nestling in its crevices. It may be premised, as a matter of course, that when the marble column was erected two hundred years ago the eight sides were polished, I and presented perfectly even surfaces. But what is its appearance to-day? The whole is rough and uneven; in parts it is as though corroded, and there are little pits clustered together in places. The idea might arise that depressions have been formed in course of time by the impact of drops of rain, but nearer inspection shows that there can be no question that the inequalities have been produced in this way; on the contrary, it is by the influence of the lichens adherent to the stone. Especially on the two sides of the pillar facing south and south-west, one sees clearly how each pit corresponds exactly in size to a species of grey lichen there ensconced, and how this lichen, as it continues to grow and extends radially, corrodes and etches the marble it touches in ever-widening circles. The expression "to etch" may here be taken literally, for there is no doubt that the process, the result of which is manifested in the formation of little pits, is mainly caused by the excretion of carbonic acid from the lichen's hyphee, whereby the calcium carbonate is converted into bicarbonate. The latter, being soluble in water, is, in part, taken up by the lichen as nutriment, whilst part is washed away by the rain. In addition to this chemical action, the hyphal filaments exercise also a purely mechanical influence. A growing hypha penetrates wherever the merest particle of carbonate of lime has been dissolved and accomplishes regular mining operations at the spot. Projecting particles of the carbonate not yet dissolved are separated by mechanical pressure from the main mass; and at the places in question where a lichen is in a state of energetic growth, tiny loose rhombohedral fragments of the lime are to be seen, which are washed away by the next shower or else carried off as dust by the wind. The same process as that which may be so clearly traced on the marble pillar at Ambras takes place, of course, also on the limestone that has not been carved or polished, in every locality where lichens exist at all. We notice it in the case of other kinds of stone as well — in dolomite, felspar, and even in pure quartz rock — for even quartz is not able to withstand the lon^-continued action of Vol.1. ^ 17 258 ACTION OF PLANTS ON THE SOIL. carbonic acid and the mechanical operations above referred to in the performance of which the hyphge act like levers. Some of the powerful iron bands belonging to the great suspension bridge across the Danube at Budapest afford us the opportunity of observing the mining operations of lichens on a substratum of pure iron. Of course in these cases the decomposition and solution initiated by the carbonic acid varies according to the nature of the substratum; the result is, however, invariably the same; there is always a loss of substance on the part of the substratum, and a part of the dissolved matter is always taken up by the adherent plant, whilst another part is carried away either in solution or mechanically by wind or rain. Mosses act in precisely the same manner as lichens. If a tuft of Grimmia apocarpa is lifted away from the side of a block of limestone, it becomes evident that in the neighbourhood of the place where all the stemlets of the little moss- colony meet, the underlying stone is threaded through and through, and rendered friable. There lie the rhizoids imbedded between isolated particles of lime, which are as fine as dust, and have been disintegrated by the chemical and mechanical activity of the organs in question. At spots where plants of Grimmia have died, the limestone always exhibits an obvious loss of substance in the form of unevenly corroded depressions. The fact that the roots of Phanerogams also alter the subjacent stone in a similar manner may be proved by the following experiment. A polished slab of marble is covered with a layer of sand, and seeds of plants caused to germinate in this sand. The roots of the seedlings as they grow downwards come almost immediately upon the marble slab, and, turning round, creep onward in close contact with the stone. After a short time the parts of the slab against which the roots are pressed become rough as though they had been etched; a solution of individual particles of the carbonate of lime takes place under the influence of the acid juice saturating the cell- walls of the root's cells, and this circumstance reveals itself to the naked eye as a roughness which is readily perceptible. Whereas the loss of substance affecting the solid substratum of plants may thus be at once detected by sight, the removal of constituents of the air and of water eludes direct observation. The ingredients withdrawn by plants are instantly replaced in water and still more in the air by influx from the environment, and obviously no holes or pits are the outcome as in the case of a surface of limestone rock. In the discussions that follow it is important to retain the conception that in the process of vegetable nutrition certain substances may undergo local displace- ment, accumulation, and aggregation, and temporary consignment to a state of quiescence. Ingredients of the earth's crust are borne upwards into atmospheric regions, and constituent parts of the air are carried deep down into the ground. Lime, potash, silicic acid, iron, &c., pass from disintegrated rocks into the realms above ground — into stems and leaves, and to the tops of the highest trees, whilst carbon and nitrogen pass from the aerial shoots and from the foliage spread out in the sunshine into the deepest shafts which the roots have bored for themselves in ACTION OF PLANTS ON THE SOIL. 259 the ground. If one were to mark out the space of ground from which the lime, potash, and other nutrient salts used in the construction of a birch-tree were derived, its bulk would certainly be found to be much larger than that of the birch; and, if we were to try to estimate the volume of air through which the carbon, which has been converted into organic compounds in the tree, was previously dis- tributed in the form of carbon dioxide, it would turn out to exceed the volume of the birch a thousandfold. In this sense, every plant may justly be considered as an accumulator of those substances which serve for its nutriment. Every plant continues, so long as it lives, to store them up in ever-increasing quantities in its own body, and in the case of long-lived plants there is thus collected ultimately quite a considerable quantity. When the life of an accumulator of the kind is extinguished, those materials which were taken from the atmosphere are able to return into the atmosphere; but such mineral food as has been derived from the ground and lifted into the upper parts of the plant — particularly those above the ground — and has there been amassed in a confined space, does not return to its original place. A dead tree breaks down on the first provocation, and the trunk lies on the ground and rots. Such part of its substance as can pass into the atmos- phere in gaseous form escapes; but the salts accumulated within it, which it raised from deep under ground during its lifetime, are retained by the surface-layers of the soil. Even though some of them are washed out of the trunk by the lixiviating action of rain-water, the superficial layers of earth operate as a filter, and do not allow any part to return to the underlying strata. So, too, the nutrient salts which reach the foliage of plants are added to the top layers of the soil; for fallen leaves go through much the same process as the trunk which is broken by storms and undergoes decay as it lies prostrate upon the ground. Thus, wherever men do not interfere by clearing away the accumulative agents in question {i.e. plants), where there is no removal of the haulms of cereals from fields, or of mown grass and herbs from meadows to serve as hay, or of timber from the forest — wherever, in a word, the vegetable world is left to itself and the natural progress of evolution is not frustrated by any disturbing element — the food-salts which have been amassed will accumulate in the uppermost layers of the earth. Moreover, seeing that, as has been already pointed out, every plant has the power of possessing itself of substances of value to it, even when they are only present in the environment of the roots in scarcely appreciable quantities, it is possible for the top layers of soil to contain a considerable amount of a substance which only occurs in the subjacent rock in such small measure as to be detected with difficulty. The percentage of lime yielded by the subsoil on the Blockenstein, a granitic mountain 1383 meters high, on the borders of Bavaria and Upper Austria, was 2-7, whilst that of the top layer was 19-7; the percentage on Mount Lusen, situated to the north of the Blockenstein, was 1-9 for the subsoil and 8-6 for the superficial layer. When one considers that fresh plants strike root in the ground near the surface and these again act as accumulators, and remembers in addition that snails make their appearance in abundance wherever vegetable food containing 260 ACTION OF PLANTS ON THE SOIL. lime is to be found, that these snails again are to be reckoned as accumulators, and that their shells, which consist almost entirely of lime, remain after the animals' deaths in the top layer of soil, it is not surprising to find that the earth-mould on a granite plateau contains a proportion of lime not much less than that yielded by mould resting on argillaceous limestone. Still more striking than the influence of rock plants and land plants in trans- posing and accumulating lime is the agency of hydrophytes in causing the same results. In the trickling springs of mountainous regions as well as in the standing pools of level country and no less in the depths of the sea, plants occur which obtain part of the carbonic acid they require by the decomposition of the bicar- bonate of lime dissolved in the surrounding water. The monocarbonate of lime, which is insoluble in water, is then precipitated in the form of incrustations upon the leaves and stems of the plants in question. Many of these hydrophytes take up carbonate of lime into the substance of their cell-membranes; and in other cases both phenomena occur, that is to say, not only are they incrusted externally with calcium carbonate, but the cell-walls are also thoroughly impregnated by the same salt. In the streams arising from springs loaded with bicarbonate of lime in solution derived from the heart of a mountain, a number of mosses regularly occur — Gymnostomum eurvirostre, Trichostomum tophaceum, Hypnum falcatum, and others besides. These mosses and also several species of Nostocinese belonging to the genera Dasyactis and Eiiactis become completely incrusted with lime, in the manner referred to, but go on growing at the apical end as the older and lower parts imbedded in lime die off. In consequence, the bed of the stream itself becomes calcified and elevated, and, in course of time, banks of calcareous tufa are formed, which may attain to considerable dimensions. Banks raised in this manner are known which are no less than 16 meters in height; to construct them mosses must have worked for more than 2000 years. Numerous Stoneworts (species of Chara or Nitella), the Water-milfoil and Horn- wort (Myriophyllum and Ceratophyllum), Water-crowfoots (Banunculus divari- catus and R. aquatilis), and more especially many Pond-weeds {Potamogeton), which grow in continuous masses in still, inland waters, incrust their delicate stems and leaves with lime during the summer, but in autumn shrink away, that is to say, their stems and leaves fall and decay, leaving scarcely any trace of the mass of vegetation till the advent of the following spring. The calcareous deposits, how- ever, are preserved, and, sinking to the bottom of the water where the incrusted plants lived, form a layer which year by year increases in thickness. Anyone who undertakes the investigation of the sequestered wastes of water in the shallow lakes of lowland districts will be convinced of the magnitude of the scale on which this kind of accumulation must take place. As one's boat glides over places where there is a luxuriant growth of the lime-incrusted Chara rudis and G. ceratophylla, there is a crepitating sound in the water like the snapping of dry sticks of birch- wood. Great numbers of stoneworts are fractured by the boat as it strikes against them, and if one takes hold of the fragments they feel like a heap of brittle glass ACTION OF PLANTS ON THE SOIL. 261 fibres. What a quantity of carbonate of lime must be deposited yearly at the bottom of these lakes and ponds! Amongst pond-weeds, Potamogeton lucens, in particular, clothes its large shining leaves with a very stout, uniform crust, which drops off in scales as the plant dries, the weight of which can be exactly determined in the case of each separate leaf. The result of careful weighing showed that a single leaf equal in weight to 0*492 grm. was covered with a calcareous crust weighing 1-040 grm. Now, supposing one shoot of this pond- weed, having five leaves, and covering an area of 1 square decimeter, decays in the autumn, and lets its lime sink to the bottom of the pond, the approximate weight of lime deposited each year on a square decimeter of the ground at the bottom is 5 grms., and, if this process is repeated every year, a layer is deposited in ten years which weighs 50 grms., and consists of calcium carbonate and traces of iron, manganese, and j silicic acid.^ There is no doubt that it is possible for calcareous strata of great depth to be j produced in this way in fresh water. That also in times past lacustrine deposits of I lime have had a similar origin is inferred from the fact that the spore-fruits of I stoneworts (Characese) and the nutlets of pond-weeds have been found over and j over again inclosed in these formations of lime. Calcareous deposits originating ! in this manner are, at present at least, less frequent in the sea. Still, the Aceta- : bularioe undergo similar changes there, and may be the cause of an elevation of ! the sea bottom and of an accumulation of lime. On the other hand, in the sea, the Lithothamnia and Corallinas play a predominant part, and form — just like true corals, and often indeed in conjunction with these and other marine animals — lime reefs of great magnitude. The agency of plants may occasion accumulations of iron hydroxide, silicic acid, and salts of potassium and sodium at particular places besides lime. The formation of meadow iron-ore, spring iron-ore, and bog iron-ore, the construction of tripoli, agate, and flint, by the conglomeration of siliceous-coated Diatomaceae, and the accumulation of potassium and sodium salts in the superficial strata of salt steppes are processes which take place essentially in the same manner as the piling up of carbonate of lime, although upon a more modest scale. The question now arises, why it is that the substances which are stored in pre- ponderant quantities in the vegetable frame, which are the main constituents of the living part of plants, and represent the alpha and omega of plant life, are not pre- j served as well as the mineral food-salts in question. Why do not carbon and nitrogen, materials so eagerly appropriated by the living plant, compounded by it with the elements of water, secured in some measure in organic compounds, and constituting the fundamental mass of the vegetable structure, remain behind in the same condition after the death of the plant ? W^hen autumn comes and the lime- laden pond-weed dies, only the calcareous crust falls to the ground, and, at the bottom of the pond, enters upon a period of quiescence. The tissue of the plant ^In the case investigated 96 per cent calcium carbonate, 0 28 per cent iron oxide, 1'51 manganese oxide, and 1'51 per cent silicic acid ; the last, from the Diatomacese, settled on the calcareous crust. 262 ACTION OF PLANTS ON THE SOIL. itself — all its carbohydrates and albuminoid compounds — cannot remain dormant, but are split up without delay into those simpler compounds of which they were compounded in the summer; and, by the following spring there is nothing more to be seen of any of the pond-weed's stems and leaves. Certainly this is only to such a conspicuous extent true of plants living underwater; dead plants buried in earth or exposed to the atmosphere are resolved less rapidly, and under certain circum- stances deposits of organic remains on limited areas are preserved even almost unaltered through boundless ages. Let us try to obtain a somewhat closer knowledge of these various degrees of preservation. Thoroughly dried wood, leaves, and fruit, if protected from all but transient moisture, are capable of being preserved unaltered for long periods of time. When wood is exposed in a dry place to the sun, it turns brown, and in the course of years becomes quite black outside, the most superficial layers being regularly carbonized, as may be seen particularly well in the case of woodwork situated under the projecting roofs of old mountain chalets. This wood exhibits no sign of crumbling, mouldering, or rotting. In the dry chambers of old Egyptian graves fruits, foliage, and flowers have been found which were laid by the side of the dead 3000 years ago, and they had not undergone a greater change than if tliey had been dried but a few days. Even the colours of flowers of the Larkspur, the Safflower, and other plants of the kind, were still apparent, and the separate stamens in Poppy flowers were in a state of complete preservation. Dryness there- fore may be looked upon i:)ar excellence as one of the preventives of the decomposi- tion of organic matter. The same result as is secured by dryness in the cases cited is brought about in the ground of moors by humous acids. The dead plants saturated with these acids are not resolved into carbonic acid, water and ammonia, but preserve their form and weight almost unaltered, and are converted into peat. Above the mass of peat new generations of plants continue to spring up and produce ever fresh organic matter, which, in its turn, becomes peat, and is added to the mass beneath, so that gradually a very deep bed of organic matter may be accumulated in this manner. In the low country lying between East Friesland and the Hiimmling, from the river Hunte to the marshes on the Dollart, there is a stretch of nearly 3000 sq. kilometers covered with a layer of peat which has an average depth of 10 meters. Of minor importance is the preservation of dead plants and parts of plants in snow and ice. The leaves, twigs, and seeds, which are carried by the wind on to the snow-fields of the high mountains, remain there a long time almost without alteration in respect of form or size; they only turn brown under the influence of the intense sunlight, and at last become quite black as though they were carbonized, which, in fact, they are. So also such insects as meet their death on the snow-fields are converted there into a black, cindery mass. Indeed, even all the minutest organic fragments lying on a glacier become carbonized, and this explains the fact that the so-called cryokone, or snow-dust, which we have already had occasion to allude to, has a graphitic appearance. ACTION OF PLANTS ON THE SOIL. 263 Dead leaves, haulms, branches, and tree-trunks, when they rest upon damp ground, as also lifeless roots, rhizomes, bulbs, and tubers, buried in moist earth, pass into a state of putrefaction, provided that their temperature does not fall below freezing-point, that is to say, they are resolved into carbonic acid, water, and ammonia, the rapidity of the process varying directly as the supply of water and the degree of temperature to which the dead matter is exposed, and inversely as the quantity of compounds of humous acid present. If more dead fragments of plants accumulate within a particular interval of time on one spot than decay, a formation of vegetable mould takes place there; on the other hand, the ground remains destitute of humus when the entire accretion of organic matter is quickly decom- posed as soon as it is dead. The general fact turns out to be that the decomposition of organic bodies is prevented, or at least limited, by a dry condition, and is promoted by moisture, and that it can only be prevented in moist surroundings by the presence of large quantities of humous acids, or by the temperature being low enough to turn water into ice. This result directs attention to those inconceivably small animate beings, which, as has been proved by experience, are arrested in their activity by scarcity of water and are killed by the antiseptic substances referred to. That they are the cause of the resolution of dead plants is corroborated by the facts that they are always present where vegetable putrefaction is in progress, and that, on the other hand, decomposition can be prevented by rendering the access of these minute organisms impossible. First in importance in this respect of course are bacteria, the causal connection of which with processes of dissolution, and especially with those decom- positions, which are known by the name of putrefaction, is established. Of these bacteria. Bacterium Termo, and several micrococci, bacilli, vibriones, and spirilla, are the commonest. Their multiplication and the withdrawal for this purpose of substances from dead plants cause a splitting up of the organic compounds in the latter. The albuminoid compounds are first of all peptonized; next, tyrosin, leucin, volatile fatty acids, ammonia, carbon-dioxide, sulphuretted hydrogen, and water are formed, this stage of the process being accompanied by the evolution of an offensive odour of decomposition, and later, nitrous and nitric acids are produced by further oxidation. The carbohydrates, too, chiefly cellulose and starch, are split up, and the products of this analysis, in so far as they are not used up by the bacteria for their growth and reproduction, pass in a gaseous condition into the atmosphere, or into the water surrounding the dead plants. Moreover, the bacteria themselves do not remain at the spots where they have been battening on vegetable matter, but swarm away through the water, or else come to rest in a short time, in which case if the seat of their activity dries up they are blown away by currents of air, and so conveyed to other dead plants. Similar decompositions can be induced by moulds (Eurotium, Mucor, Botrytis cinerea, Penicillium glaucum) as well as by bacteria, and, in addition, the disintegration of wood occasioned by the mycelium of Dry-rot (Merulius lacrymans), the green-rot of trunks of oaks, and beeches, caused by Peziza ceruginosa, the mouldering of wood induced by the mycelium of Polyporus 264 ACTION OF PLANTS ON THE SOIL. sulfureus and various other fungi, the red-rot, &c., all depend on similar disruptions of the organic compounds in dead plants, and result in the ultimate dispersal of these in the air in the form of carbon-dioxide, ammonia, nitric acid, and water. Thus, ultimately, the exercise of this destructive activity only effects a return of the compounds just enumerated — the most important to plant-life — to the regions whence they had previously been withdrawn by the plants when living. Carbon and nitrogen, in particular, are set free from their bonds and given back to the atmosphere in the form and combination in which they are capable of being appropriated anew by living plants as food-material Considered from this point of view the phenomena of putrefaction and rotting appear as important and even necessary incidents in the history of the substances which are of the greatest importance to plants. Abhorrence of putrefaction is innate in us all, and everything connected with it — in particular, the entire race of bacteria — is looked upon with aversion. To estimate these processes according to their deserts requires a sort of self-abnegation. But when we overcome our repugnance and weigh the whole subject impartially, we come to the conclusion that the continued existence of vegetable life and of life in general depends upon the occurrence of putrefaction. If the untold numbers of plants which die in the course of a year did not rot sooner or later, but remained unchanged as lifeless forms, a certain quantity of carbon and nitrogen would be idle, being withdrawn from the sphere of activity and locked up, so to speak. Now, assuming this to be repeated year by year, a time must come when all the carbon and nitrogen would be imprisoned in dead plants. Thereupon, all life would cease, and the whole earth would be one great bed of corpses. Not only putrefaction, but also the minute organisms which excite putrefaction appear in a more favourable light when viewed from this standpoint. Let such bacteria as act in the capacity of foes to the human race, ravaging town and village in the form of infectious diseases, be exterminated if possible; but annihila- tion of putrefactive bacteria would mean a disastrous interference with the cycle of life upon the earth. These latter are not to be reckoned as enemies but friends to human beings. The effect of their invasion of dead plants and animals is certainly first made manifest, not in the most agreeable manner, for some of the substances mentioned as being evolved in the early stages of the onslaught, viz.: various ammoniacal compounds, sulphuretted hydrogen, and the volatile fatty acids, are disgusting to us; but as decomposition advances these phenomena, which are so unpleasant to our senses, abate, and the action of putrefactive bacteria becomes ultimately a beneficent process of purification of the last remnants of dead organisms. The final result of the decomposition of organic bodies by bacteria has been termed mineralization. It is a fact that nothing is ultimately left behind, in the ground or water, of bodies decomposed by the indefatigable exertions of bacteria excepting some nitric acid and the small quantity of mineral food-salts which has been taken up by the living organism in its time and are now in the form of dust and ash. MECHANICAL CHANGES EFFECTED IN THE GROUND BY PLANTS. 265 By filling with water a glass which contains vegetable and animal remains in a state of putrefaction and swarming with bacteria, one is enabled to follow this process of mineralization from day to day. First, a decrease of the organic matter clouding the liquid, accompanied by simultaneous increase of ammonia and nitrous and nitric acids, is observed; then, after about two months, a complete clearing up of the liquid. The water is now colourless and odourless, but a precipitate has formed at the bottom, which contains, in addition to insoluble food-salts, bacteria in a state of temporary quiescence on the termination of their task and waiting till fresh prey becomes accessible. No doubt these processes occur in nature in just the same manner as in the glass of water, and the so-called self-purification of rivers, for example, has been rightly attributed to mineralization. It was long ago noticed that the water of such rivers as flow through great towns and consequently take up considerable quantities of animal and vegetable refuse contains no discoverable trace of all these impurities a few miles below the mouths of the drainage pipes and sewers. The water of the Elbe, which receives the refuse of the towns of Prague, Dresden, and Magdeburg, is so pure at Hamburg that it is there used for drinking purposes without protest^ The Seine, after taking up masses of rubbish in Paris, is already by the time it reaches Meulan, a distance of 70 kilometers, clear and pure again, and does not even exhibit there any traces of the organic matter received in the great city. Were it not for the activity of the putrefactive bacteria, this purification would never take place; and although the statement that putrefactive bacteria are the best of purifiers sounds at first like a paradox, it must be acknowledged to be consistent and based on experience. MECHANICAL CHANGES EFFECTED IN THE GROUND BY PLANTS. All the alterations hitherto spoken of as being brought about in earth and under the influence of vegetation subsisting therein are reducible to chemical transpositions. Added to these, there are always certain purely mechanical changes. The penetration of the rhizoids of a rock-moss or the hyphge of a crustaceous lichen into limestone is accompanied, as has been already stated, by a solution of part of the substratum and a mechanical separation of another part; the rhizoids or hyphae, as the case may be, becoming imbedded amongst tiny detached fragments of the underlying stone. When the hyphae and rhizoids die, the corresponding piece of the substratum is left porous, and admits air and water, whilst other plants are enabled to settle on it, although they may not perhaps possess the power of eating into stone and pulverizing it in the same degree as their predecessors. This is also true of the roots of Phanerogams. The food-seeking root-tips and their absorption- cells displace particles of earth as they insinuate themselves, and when they decay later on, the soil at those particular places is intersected by passages of varying size. No doubt these passages mostly collapse like the abandoned shafts and galleries of a mine, but some trace of root-action always remains behind in the shape of an * This was written before the last outbreak of cholera. 266 MECHANICAL CHANGES EFFECTED IN THE GROUND BY PLANTS. increased looseness of the soil in the locality, a result of the greatest importance, inasmuch as it enables air and water to permeate to a depth much more easily and quickly by the waj'S that the roots have previously opened up. Dead roots rotting underground constitute, moreover, the source of the carbonic and nitric acids which help to render available the mineral constituents, and so serve the turn of subsequent generations of plant-settlers on the same spots, whilst they accomplish fresh disin- tegration of the substance of the soil. If, however, the subterranean parts of plants are continually engaged in mining, and so change in various ways the position of the component particles of soil, the organs above ground exert an influence in some measure opposed thereto, in that they retain and bring to rest particles of earth which are set in motion by currents of air or water. In the section that treats of the absorption of nutrient salts by lithophytes, attention was directed to the fact that the dust pervading the atmos- phere, and blown from place to place by the wind, is arrested to a remarkable extent by mosses and lichens. One need only detach a small tuft of the common Barhula muralis, which everywhere occurs on walls by roadsides, to convince oneself of the extent to w^hich dust from the road is lodged amongst the leaves and stemlets, and of the tenacity of its adhesion. Moreover, not only such dust as rises from roads, but also that variety which, though not easily observed, yet fills the air of remote mountain-valleys, of arctic ice-fields, and of the most elevated parts of the earth's crust, is arrested in those localities by mosses and liverworts, and by many Phanero- gams besides, the growth of which is similar to that of mosses. There is not much less dust clinging amongst the stemlets of the dark Grimmias, Andreseas, and other rock-mosses, which grow in small cushion-like tufts on weather-beaten mountain crags, than is attached to the Barhula living by the dusty roadside. If one of the tufts in question is detached from its substratum, a fine powder composed of mica- scales, granules of quartz, chips of felspar, and a number of minute organic frag- ments pours out from between the moss-stems, whilst another portion of this finely powdered earth is left clinging to the leaves and stemlets, and is found to be regu- larly adnate to them. It is never, however, the still fresh and living upper parts of these leafy moss- stems that arrest and carry dust, but always the older dead parts below. The lower dead half of the moss, whether still in a state of preservation or already rotting, is alone capable (in consequence of characteristic alterations in the lifeless cell-tissue) of holding fast the atmospheric dust. The under part of moderate-sized cushions of moss constitutes a compact mass composed half of imprisoned dust and half of brown lifeless moss-stems. These little cushions, clothing rocky crags, become a favourable site for the germination of a whole host of seeds, which are conveyed thither by the wind and detained in the same manner as the dust. The seedlings arising from these seeds send their rootlets into the subjacent portion of the bed of moss, where the interstices are full of dust or finely-divided earth. Here they find all the conditions prevailing necessary for their nourishment, and they expand, and, little by little, crowd out the mosses which received them so hospitably. MECHANICAL CHANGES EFFECTED IN THE GROUND BY PLANTS. 267 forming ultimately a bed of flowering plants, including in especial abundance representatives of the orders of grasses, pinks, and composites. Many water-plants — in particular, aquatic mosses and algae — possess, in an almost greater degree than lithophytes or land plants, the power of laying hold of inorganic particles, and thus exercise a far-reaching influence as mud-collectors on the conformation of the ground. It is wonderful how plants are able to arrest large quantities of the fine sand hurried along by a flood, although they are exposed to the violent rush of the water. The tufts of the dark green alga Lemanea fiuviatilis and of the aquatic moss Cinclidotus riparius, which cling to rocks in the cascades of clear and rapid mountain torrents, are so conglomerated by mud and sand that they cannot be freed therefrom until the tissue has become dry and shrivelled. Limnohiwm molle, which grows in the turbid waters from glaciers, has such an abundance of earthy particles adhering to it that only the green tips of the leaf-bearing stems are visible above the grey-coloured cushions imbedded in the mud. The felted masses of Vaucheria clavata, filling the channels of apparently clear, gently-flowing streams, are so mixed with mud that if a lump of this alga is fished out, the weight of mud clinging to it exceeds that of the alga itself a hundredfold. In these cases of submerged plants, it is, again, not the living but the dead parts which serve to arrest the mud. On lifting up a lump one sees clearly that only the uppermost and youngest prolongations of the filaments — those situated at the periphery of the algal cushion as a whole — are living and filled with chloro- phyll; the fundamental mass has become colourless and lifeless. But these dead parts, which form a thick felt of interwoven filaments, alone retain in their meshes the finely-divided mud and sand in such surprising quantities; these particles slip off the green living parts without adhering to them. An important consideration in this connection is the fact that the dead cell-membranes swell up and become slightly mucilaginous, so that fine particles of mud lodge more easily in the soft swollen substratum thus formed. Wooden stakes stripped of bark and fixed in a strong current show this very clearly, as do also the trunks of trees that are thrown up by floods and lie stranded on the shore with their bared boughs projecting into the stream. However strong the current to which wood in that condition is ex- posed, it covers itself in a short time with a grey coat consisting of earthy particles brought down by the water. If a piece is cut ofif and exposed to the air, the earthy deposit does not become detached until the wood-cells have dried up and shrunk. As long as they are moist the particles of mud continue to adhere to them. This mechanical retention and storage of dust by rock-plants and of mud by aquatic plants is of the greatest importance in determining the development of the earth's covering of vegetation. The first settlers on the bare ground are crustaceous lichens, minute mosses and algae. On the substratum prepared by them, larger lichens, mosses, and algae are able to gain a footing. The dead filaments, stems, and leaves pertaining to this second generation arrest dust in the air and mud in the water, and thus prepare a soft bed for the germs of a third generation, which on rocks consists of grasses, composites, pinks, and other small herbs, and in water 268 MECHANICAL CHANGES EFFECTED IN THE GROUND BY PLANTS. of pond-weeds, water-crowfoots, hornwort, and various plants of the kind. The second generation is produced in greater abundance than the first, and the third develops more luxuriantly than the second. The third may be followed by a fourth, fifth, and sixth. Each successive generation crushes out and supplants the one preceding it. As on the rocky heights and in the roaring torrents of mountains, so also on the sandy plain and in the depths of the sea, a perpetual variation in the nature of the vegetation is taking place. At all times and in all places we see younger genera- tions displacing the older and building upon the foundations laid by their pre- decessors. The first settlers have a hard fight with uncompromising elements to seize possession of the lifeless ground. Years go by before a second generation is enabled to develop in greater luxuriance upon the earth prepared by the first occupiers; but there is no cessation in the productive and regulative effects of vegetable life, and its energy and aptitude in the work result in the erection of its green edifices over wider and wider areas. New germs are established upon the mouldered dust of dead races, and others on the plant forms adapted to the altered substratum, and so, for hundreds and thousands of years, the changes go on, until at length the tops of forest-trees wave above a black and deep soil, the battle-field of a number of bygone generations. Thus, the life of plants, like that of the human race, has its epochs and its history: as in the one so in the other a continual struggle prevails; processes of ousting and of renovation are always in progress, and there are ever new arrivals upon and departures from the scene. CONDUCTION OF FOOD. 1. MECHANICS OF THE MOVEMENT OF THE EAW FOOD-SAP. Capillarity and root-pressure. — Transpiration. CAPILLARITY AND ROOT-PRESSURE. Unicellular plants make use individually of the food material which they absorb from their surroundings, and work it up into the organic substances which they require for their structure and increase in bulk, and also for the production of future generations. In all plants composed, on the other hand, of aggregates of cells, there is a division of labour. Of the protoplasts occupying the cell-cavities of such larger plant-structures, one part provides for the absorption of the water and food-salts, another for the taking in of the gases which are used as food, and yet another part works up this food into organic substances for construc- tive purposes. The centres in which these various industries are carried on are frequently situated at some distance from one another, and it is obvious that there must not only be some communication between the various regions of activity, but that active forces must come into play which will effect the transport of the food from the cells whose function it is to receive it, to those in which it is to be elaborated into building material. It is evident that the greater the distance is between the various centres of the plant in question, the more difficult will be the performance of this task. In aquatic plants and lithophytes, all of whose superficial cells have the power of taking in nourishment from their environment, these distances are proportionately small, while they attain their greatest dimensions in land-plants whose roots are embedded in the earth, and whose leaves are surrounded by air. In trees the food materials which are taken up by the absorbing roots beneath the ground must frequently travel far more than 100 metres before reach- ing the topmost leaves. The path to the summit is very steep, and the fluid in rising must be able to overcome the force of gravitation, which has no inconsider- able significance at heights such as these. Naturally, desire for knowledge has at all times directed attention to this phenomenon, and the most diverse attempts have been made to explain by what means the food-sap taken in by the roots of trees is enabled to reach their summits. It was first considered to be in virtue of capillarity; that just as oil, alcohol, or water, is drawn up the wick of a lamp, the liquid food can rise in the delicate tubular cell-formations called vessels, which, united together in groups or 270 CAPILLARITY AND ROOT-PRESSURE. bundles, traverse the stems and leaves of plants. But the vessels are closed in above and below, and therefore it is impossible that capillarity should be sufficiently- developed in them. At best it could only raise the sap a trifling distance, and could never convey fluid to a height of many metres. It is a striking fact that in many plants the ascent of the sap is most vigorous after the evaporation from the superficial parts exposed to the air has been weakest. The so-called "weeping" of vines, i.e. the outflow of sap from the flat surface of a cut vine-branch, does not take place in summer and autumn, immediately after the branch has been fully adorned with foliage, and when its extensive leaf-surfaces have given up large quantities of moisture to the surrounding air; it occurs at the end of the winter sleep of the plants, when the brown branches rising above the ground are still in a bare and leafless condition. The (iause of the ascent, or at least of the ascent in the lower leafless branches, must therefore be sought for in the absorbent roots, and it may be assumed that here the same causes are at work which induce the fluid food materials of the surrounding earth to enter the superficial cells at the root-tips. It has already been shown that the contents of these cells suck up the water of the nutritive ground with great force in consequence of the chemical affinity they have for it, or in other words, that the fluid reaches the interior of plant-cells by endosmosis; it has also been mentioned that in consequence of the taking in of water the volume of the cell-contents increases, producing pressure from within outwards on the cell-wall, and the cell swells and becomes turgid. From this one of three cases might be deduced: — first, suppose that the cell-wall is so composed throughout that it allows the entrance of water into the cell, but not its exit, and that consequently the cell-contents absorb water, but that a filtration of the same towards the exterior cannot take place. Granted this hypothesis, the cell-wall by virtue of its elasticity would yield to the pressure of the cell-contents, but only within the limits of that elasticity; hence a condition of tension would be produced, in which the reciprocal pressures of the cell-wall and cell-contents would be in equilibrium. In the second case, suppose that the pressure of the cell-contents is greater than the force of cohesion between the molecules of the cell-wall, this consequently ruptures, and the cell-contents issue from the rent which is formed. This phenomenon is seen in certain pollen grains when placed in water. In half a second the cells absorb so much water that they double their volume; the cell- contents still absorb the fluid, and the cell- wall can at length no longer withstand the pressure; it bursts, and the contents, from which the pressure is now removed, pour through the opening, and are diftused in the surrounding water. There is a third case possible. Suppose that in a given cell the opposite walls are not of identical structure; that the wall which is in contact with the damp earth is so organized as to allow the entrance of water, but not its filtration to the exterior, while the opposite wall ofifers only a slight resistance to such filtration; then by the increasing pressure of the cell-contents fluid will be forced through that wall which offers least resistance, and the greater the affinity of the cell- CAPILLARITY AND ROOT-PRESSURE. 271 contents for the fluids in the nutritive earth, the more abundantly and energetically will this be carried on. The phenomenon can be well seen in some moulds, especially Mucor Mucedo, which makes its appearance in such quantity on : succulent fruits; and in the mycelium of the so-called Dry-rot, Merulius lacry- mans. Fluids are sucked up by the lower portions of the tubular cells which cover the nutritive substratum, and expelled again through the walls of upper parts ' of the same cells, which project freely into the air. These upper portions of the i mycelium cells appear as though ornamented with tiny dewdrops, which in the m\ lirnu .■ round the edge of the disc, on account of which the drops of water roll down from the middle of the leaf to the edge on the slightest rocking movement, and there coalesce with the water on which the leaves float. This puckering ,'- of the margin of the leaf is attended in the water-lilies by a phenomenon which, although not directly asso- ciated with the matter in hand, is so full of interest that it cannot be passed without notice. If we take a boat in the bright sunshine at mid- day, and float over the calm inlet of a lake, whose surface is overspread with the leaves of water- lilies, and if the water is clear to the bottom, we shall see the sha- dows of the leaves which float on the surface sketched out on the ground below. But we can scarcely believe our eyes — these do not look like the ! shadows of the I leaves of water- I lilies, but rather I of the fronds of huge fan-palms. From a dark central portion raaiate out long dark strips which are separated from each other by as many light bands. The cause of this peculiar form of shadow is to be found in the undulating margin of the floating leaves. The water of the lake adheres to the whole of the under surface of the arched portions 19 Fig. es.-o- disc as far as the edge, and is drawn up by capillarity to Vol. I. the 290 MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. of the undulating margin. The sun's rays are refracted as through a lens by this raised water, and so a light stripe corresponding to each convex division of the curved margin is formed on the bed of the lake, and a dark stripe corresponding to each concave part. These are arranged in a radiating manner round the dark central portion of the shadow. MAINTENANCE OF A FEEE PASSAGE FOR AQUEOUS VAPOUR. Special arrangements are met with in all plants which possess stomata, in order that the giving off of aqueous vapour may continue without hindrance. Water falling on the upper side of the leaf, in the form of rain and dew, threatens to cause the greatest obstacle to this free passage should it be able to collect directly in the stomata. The width of an open stomate does not render the entrance of water by capillarity impossible. As long as light and warmth exercise their power, as long as the temperature in the neighbourhood of the spongy parenchyma is higher than that of the surrounding air, and water- vapour in consequence is pro- duced in the spongy tissue and driven out with force from the stomata, such an entrance is indeed inconceivable. It is impossible for aqueous vapour to pass out and at the same time for fluid water to enter by the same passage and through the same gate. But should the leaf become cooled by radiation after sunset, and dew be deposited upon it, or should a cold rain trickle down over the leaves, and the stomata have been unable to close quickly enough, it is quite possible that water might enter, just as it enters a retort (whose narrow mouth dips into water, and whose contents have been vaporized by placing a lamp under them), when the lamp is removed, and the bulb of the retort with its contents becomes cooled. But putting aside the possibility of water thus pressing its way in, this much is certain, that the formation of a layer of water over the cells in the immediate neighbourhood of the stomata would cause great injury to the plants; and this, not only as affecting transpiration, but also the free entrance and exit of gases. Therefore, the im- mediate surroundings of the stomata must be kept open as a path for aqueous vapour, and no water must be allowed to collect and take up a position there. Stomata are much too small to be seen with the naked eye. However, it can be ascertained by a very simple experiment whereabouts, on a leaf or green branch, stomata are to be found. A twig or a leaf is dipped in water, and then withdrawn after a short time and lightly shaken; some spots will be found wet, while other places remain dry. Where water remains and spreads out to form an adhering film, no stomata will be found in the epidermis ; but where the twig or leaf is dry, one can be sure of finding them. In 80 per cent of cases experimented upon in this way, only the upper leaf-surface became wetted, while the under side kept dry; in 10 per cent both sides remained dry; and in the other 10 per cent the upper side kept dry, while it was the under side which was wetted. With this corresponds the actual fact that in far the greater number of instances the under side possesses most stomata, while the upper side is free from them. It seems as if MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. 291 this circumstance could be explained thus, that the upper side is usually turned towards the rain, and that the stomata are on this account collected too-ether on the under side, which is sheltered from it. This explanation, however, which at first sight seems so plausible, does not quite correspond to the true state of the case. The consideration of the reasons for believing that it is an advantage for the plant to have the upper side of the leaf free from stomata will indeed come later, but one thing must be noted here, — that the side of the leaf turned towards the ground, which in most cases contains all the stomata, remains anything but dry. Of course the rain-water only reaches the surface of the horizontal leaf-blade when the margin is so formed that the adherent layer of water which wets the surface is drawn over gradually from the upper to the under side, and that is very seldom the case; but the wetting of this surface by mist and dew is all the more important on this account. On taking a stroll through fields and meadows on a dewy morn- ing, as a rule only the upper surfaces of the leaves come into view, and one might easily be led to think that the dew is deposited only on this side. We constantly use the expression that the dew " falls ". Underlying this is the idea that the dew comes down like rain, and that only the upper leaf -surface becomes covered with dewdrops. But one has only to turn the leaf over to convince oneself that the lower surface is likewise bedewed, and on a closer examination it will even be seen that dew is of more importance in connection with the lower than the upper side, because it remains there so much longer. When the sun is already high in the heaven, and the dewdrops have long disappeared from the upper surface, and tran- spiration is in full force, the under side may still be found studded with dewdrops. If in the majority of cases the stomata lie on the under side, and this side is exposed to the danger of being covered with water as much as the upper one, it is evident why contrivances for hindering the access of water to the stomata are to be found much more abundantly on the under than on the upper side of the leaf. The most important of these arrangements are the following: — First the coating of wax. This is either in the form of a granular covering; or as a fine crust which fits closely to the epidermis; or, most commonly, as a continuous thin layer which is easily rubbed off, forming a delicate film popularly known as " bloom ". A group of primulas, belonging to mountainous districts and to the moors of low countries, of which Primula farinosa may be taken as the most widely distributed and best known representative, have a rosette of leaves spreading over the damp ground, and on the lower side of these leaves is a white coat, which under the microscope is seen to consist of a collection of short rods and knobs of a waxy nature. If such a leaf is plucked and placed in water for a short time, and then withdrawn, the upper side, which is quite free from stomata, will be moistened by a layer of water, while the under side, on which are the stomata protected by the granular coating of wax, remains quite dry . The lower surface of the leaves is covered with a fine closely adherent wax layer, in many of the willows growing in damp misty places near rivers (Salix amygdalina, jmrjmrea, pruinosa). as well as 292 MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. in a great number of rushes, bulrushes, and reed-like grasses. If when the dew falls heaviest one roams through a thicket of willows, or across a moor, one may see plenty of drops hanging from the under side of the leaves, but they do not actuall}'' wet this surface, and on the slightest movement of the leaves they roll off and fall down. It is, indeed, in consequence of this that one is more likely to get thoroughly wet by walking through meadows and dwarf willows than by an excursion through country overgrown with ordinary herbs. The two white stripes, so well known on the under side of fir leaves, are also formed by a waxy coat, which prevents the stomata below from being wetted. In species of juniper (e.g. J. communis, nana, Sahina) the two white stripes occur on the upper side of the leaf, and it is interest- ing to see how the distribution of the stomata again corresponds ; for junipers belong to that group of plants whose under leaf-surface is free from stomata, these being present only on the wax-coated region of the upper side of the leaves. Many grasses, to which we shall refer later for other reasons (e.g. Festuca punctoria), only possess stomata on the upper side of the leaf, and again only where the strips of wax are situated. Generally speaking, wax is a protection from moisture, and is most frequently formed when the stomata make their appearance on the upper side of the leaf. The leaves of peas, nasturtiums, woodbine, poppies, fumitory, many pinks, cabbages, woad, and many other cruciferous plants, which have stomata on the upper surface, also produce a covering of wax there. Water poured on the upper surface of a cabbage-leaf rolls off in the form of drops, exactly as it runs off a duck's back, without wetting the surface. In the fronds of ferns {e.g. Polypodium gtaucophyllum and sj)orodocarpum), on the upright leaves of Irises (Iris ger- manica, pumila, pallida), as well as on the vertical leaves and leaf -like branches of many Australian acacias and myrtles, and lastly in the erect whiplike, leafless or scantily-leaved papilionaceous plants (Retama, Spartium), the stomata are pro- tected from the wet by a coat of wax. The formation of hairs furnishes another barrier to the entrance of water into the stomata. We shall come back again to these structures, which serve so many different purposes in the plant economy, but here only those hairy and felted coverings whose task is to hinder the wetting of the stomata will be considered. Examples of these are furnished by many Malvaceae which grow in marshes and ditches (e.g. Althaea officinalis), and also by some mulleins (e.g. Verbascum Thapsus, phlomoides), whose leaves are provided with stomata on both surfaces, and with hairy coverings which it is impossible to moisten. In the damp meadows of the valleys of the Lower Alps grows Centaurea Pseudophrygea, whose large leaves, hairy on both sides, are very rough and much wrinkled. The stomata are only situated in the hollows between the ridges. When rain falls, or the leaf becomes bedewed, the water remains in the form of drops on the hairs of the elevated por- tions, and the cells in the hollows are not wetted. In many alpine plants, for example the Hairy Hawkweed (Hieracium villosum), after a fall of rain or dew the long projecting hairs of the leaves are thickly beset with drops of water, none of which can reach the stomata on the epidermis beneath. MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. 293 It should be particularly noticed here that plants with two-coloured leaves, such as those whose upper surfaces are green, smooth, free from stomata and easily- wetted, while their under surfaces, covered with gray or white hairs, and rich in stomata, which cannot be wetted, are generally to be found on the banks of rivers and streams. In the open woods which skirt the banks of rivers in the valleys of moun- tainous districts, i.e. in places where mist rises on summer evenings, and all the twigs, leaves, and stalks are covered with drops of water, the most characteristic plants are the Gray Alder (Alnus incana) and the Gray Willow (Salix incana), and as undergrowth everywhere the Raspberry — all plants adorned with the two- coloured leaves just described. Leaving the region of woods growing on river banks for the neighbouring meadows, through which ripples fresh water from a spring, and where everything drips with dew from evening until the middle of the following day, we come to the natural home of herbs and shrubs with leaves green on the upper and white on the under sides. There Fuller's Thistles (Cirsium heterophyllum and canum) grow luxuriantly, and the Meadow-sweet, with its large two-coloured leaves; whilst the whole course of the brook is bordered by the Colt's-foot {Tussilago Farfara) with leaves which may be considered typical of this group. What a contrast does this present to the lofty vaults of a dense forest, perhaps only a thousand paces away, where on the shady ground little or no dew is formed, and where the leaves which canopy the brown soil are never exposed to a thox-ough wetting! No parti-coloured leaf is to be found there, no leaves whose upper surface is green and smooth, while the under side is covered with white hairs; and plants which exhibit a thick coating of wax on their under surface, like the Primula farinosa of the moors, are also absent. On the other hand ferns are here, as for example the Hard Fern {Blechnum Spicant), whose leaves are furnished with stomata which open quite without protection on the tops of projecting undula- tions. This contrast between the leaves of plants in the open marshy country and in the interior of forests is found, not only in the colder territories of the north, but also in tropical districts. Moreover, plants whose leaves are covered with white hairs on the under surface are never to be found under the close leafy roof of huge trees which prevent nocturnal radiation and the formation of dew. Here occur, rather, plants having totally unprotected stomata opening on slightly raised areas of the surface, as for example in Pomaderis phylicifolia, and on the leaves of the Pepper family, e.g. Peperomia arifolia (see fig. 64^ and 64*), A very remarkable contrivance by which stomata are protected from moisture consists in providing the stomata of the upper surface with countless papillae and cone-shaped projections ; between them, of course, being innumerable hollows and depressions. Falling water-drops roll oft' such surfaces; the water cannot displace the atmospheric air in the depressions, and therefore the leaves and stems, in so far as their epidermis presents the aforesaid irregularities, appear covered with a thin layer of air. As the stomata are situated in the small hollows, they always remain 294 MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. dry, and even if that particular part af the plant is wholly immersed, they do not come into contact with the water. There are two causes for the unevenness of the leaves: first, the outer walls of a portion of the superficial cells may become strongly arched outwards; or secondly, solid peg-like projections may arise from the cuticle, and to these projections the air adheres so firmly that it cannot be displaced even by a considerable pressure of water. This protection of stomata against moisture by papilla-like outgrowths is to be found especially in marsh plants which are exposed to a changing water-level. On the banks of streams and rivers, and Fig. 64.— Stomata. 1 Surface view of a portion of the frond of the fern Nephrodium Filix-mas. * Vertical section through this portion, s Surface view of a portion of the leaf of Peperomia arifolia, * Vertical section through this portion ; x350. where water welling up from below forms pools and ponds, it may happen that plants are submerged for a week at a time, and then again remain dry for some months. Most of the plants growing in such situations, particularly the sedges (e.g. Carex stricta and paludosa), the rushes (e.g. Scirpus lacustris), most of the tall fistular grasses (Glyceria spectabilis, Fhalaris arundinacea, Eulalia japonica), the plants which grow with the sedges (e.g. Lysimachia thyrsijiora, Polygonum amphibium), and many other marsh plants, are all saved from the danger of having their stomata wetted during their submersion by the papilla-like out- growths of some of the epidermal cells, near the stomata, as shown in the figures on next page. Bamboos, and the grasses Arundinaria glaucescens and Phyllostachys hatn- MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. 295 busoides, which so much resemble the bamboos, besides some sedges {e.g. Carex 'pendula), exhibit on the other hand the above-named peg-like projections of the cuticle; these are shown in the section of a bamboo leaf in fig. 66^). On plunging such a bamboo leaf in water, a surprising sight presents itself. The upper side, covered by a dark green, smooth, flat epidermis, with no stomata, becomes wet all over, and retains its dark colour and dull appearance; but the under surface, blue- green in colour, and beset with stomata and thousands of cuticular pegs, does not allow the air to be displaced; and this layer of air, spread thin over the surface, glistens under water like polished silver! The leaf may be shaken under water to any extent, and may even be left submerged for a week, but the silvery glisten- ing air-stratum is not dislodged. If such a leaf is now taken out of the water, the upper surface is quite wet, but the under surface is dry, like a hand which has been dipped in mercury and then withdrawn, and not the smallest drop of water i^^i ^^1^-1^ Fig. 65.— Protection of Stomata from Moisture by Papilla-lilie outgrowths of tlie Surface. 1 Vertical section tlirough a portion of tiie leaf of Glyceria spectabilis. a Vertical section through a portion of the leaf ol Carex paludosa; x200. adheres to it. On placing a vessel of water, in which some bamboo leaves are half immersed, under the receiver of an air-pump, and then pumping out the air, numerous small air- bubbles are at once given off* from the submerged portions of the leaves. At length the silvery lustre disappears, and the air between the cuticular pegs is replaced by water. If now the leaves be completely submerged, the silver lustre is only shown on those parts which were not previously immersed, and where water could not replace the exhausted air; — the spaces round the pegs in this region having been again supplied with air on the opening of the stop-cock of the pump in order to submerge the leaves. It may be imagined from this experiment how much the stomata would be damaged by water if the plants mentioned were not protected from moisture by the pegs to which the air adheres so strongly. In many plants which grow in the sunshine, and particularly in those whose foliage is evergreen and only exposed to moisture at the time of the greatest activity of the sap (while later it is exposed for months to dry air), the stomata are to be found surrounded by an embankment, or sunk in special pits and furrows. Even in the leaves of many indigenous plants, which are green in the summer, e.g. those of the Carrot (Daucus Carota), the guard-cells of the stomata are so 296 MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. over-arched by the neighbouring epidermal cells that a sort of vestibule is formed in front of the true pore. It can easily be imagined that drops of water which come to such places are not able to press out the air from this vestibule, and there- fore cannot penetrate to the guard-cells of the stomata. In Hakea Jiorida and Protea Tnellifera, two Australian shrubs (see fig. 67), similar arrangements are met with, but here the stomata are still more over-arched, so that they are only visible to anyone looking at the surface of the leaf through small holes at the top of the dome. The stomata on the green branches of various species of Ephedra are surrounded by mound-like projections from the cuticle of neighbouring epidermal cells, and are at the same time somewhat sunken, so that an urn-shaped space is i,-1,r-v'-r^~r-l,^ "^ Fig. 66. -Protection of Stomata from Moisture by Cuticular Pegs. I Vertical section of a Bamboo leaf ; xl80. « Part of the lower portion of the section ; x460. « Part of the upper portion of the section ; x 460. formed above each stoma, from which water cannot dislodge the air. On the leaves of Dryandra Jioribunda, one of the Proteaceae which grows in the thick Australian bush, several stomata occur at the bottom of small pits on the under side of the leaf, and from the side walls of the depression spring hair- like structures which interlace and form a loose felt-work, easily penetrated by gases but not by fluids (fig. 68). The stomata on the leaves of the Oleander (NeHum Oleander) are similarly situated. These also are at the bottom of deep pits on the lower side of the leaf, and the entrance to them is beset with extremely delicate hair-like structures (see fig. 73^). The oleander fringes the banks of streams in the sunny open country of Southern Europe and the East, and in its natural position it is most exposed to wetting by rain, mist, and dew, just when transpiration is an absolute necessity for it. But even when the leaves are covered on both sides with a layer of moisture, none can force its way into the hair- lined depressions which conceal the stomata, and consequently transpiration is cot hindered even in the wettest season of the year. MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. 297 Stomata, which are spread over the green tissue of stems and flattened shoots, are frequently sunk in furrows, channels, and pits, in plants whose greatest activity occurs in the short rainy season, and they are saved from wetting in this position by the most varied contrivances. On the rocky shores of Lake Garda, and up over the mountain slopes to the heights of Monte Baldo, grows Cytisus radiatus, a shrub of unusual appearance (see fig. 69^). Its branches only possess rudimentary green leaves, and are themselves furnished with green tissue, which plays the same role as that assig-ned to the mesophyll of the leaf -lamina in normal foliaceous plants. Fig. 67.— Over-arched Stomata of Australian Proteacere. ' Vertical section through a leaf of Halcea jlorida. 2 Surface view of the same leaf; x320. of Protea rnellifera. * Surface view of the same leaf ; x 300. * Vertical section of a leaf These green branches bear very numerous secondary branches inserted in decus- sating pairs. On the secondary branches new shoots develop every spring exactly similar in form, and arranged in the same manner. At the period when this development is taking place, the humidity in that part of the Southern Alps, to which Monte Baldo belongs, is very great. In dull weather, rain and mist, or dew in fine weather, deposit large quantities of water on the soil, and on the plants covering it, particularly in the alpine region of the above-named mountains, on the westerly slopes leading down to the lake, which are thickly clothed with the shrubs in question. It is therefore important that the rod-like branches of this Cytisus should be able to breathe and transpire without hindrance, and that the short time during which the conditions for these vital transactions are favourable, should be fully and wholly taken advantage of. Here again the point above all others to be 298 MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. aimed at is to keep a free passage for the water-vapour which must escape from the stomata. To bring this about, the stomata are situated in grooves filled with air which are sunk in the green tissue, and which give a striped appearance to the branches. Water cannot force out the air from these narrow furrows which run along the green branches and twigs, eight of them to each branch. The branches may remain submerged in water for an hour without a trace of moisture entering the furrow. Moreover hairs are present in the furrows as a guard against moisture. These cannot be wetted, and the air adheres to them just as to the cuticular pegs of the bamboo leaf. A clear idea of this arrangement is given in the transverse section of the stem shown in fig. 69 ^ and 69 *. The adjacent section of the green branch of the Australian Casuarina quadrivalvis shows that these curious plants also have exactly the same arrangements, that the stomata He at the bottom of 2 Fig. 68.— Stomata iu Pit-like Depressions. ' Surface view of a leaf of Dryandra florihunda. A portion of the hairs which fill the pit is removed, in order to show the stomata; x360. 2 Vertical section through a leaf of i))i/a7idm/ori6i(jirfa; x300. narrow furrows which run along the green leafless branches, and that peculiar hair- structures are present in the furrows, to which the air adheres, forming a barrier against water, exactly as in those of the Cytisus. The Casuaringe, which must finish their work for the year during the very short rainy period of their native country, require during this time arrangements providing for unhindered transpira- tion no less than does the Cytisus in the Southern Alps. Altogether this con- trivance is found to be present in only a limited number of cases; in perhaps only twenty papilionaceous shrubs, most of which belong to the Spanish flora, of the genera Betama, Genista, Ulex, and Sarothamnus, in addition to the Australian Casuarinas, and in allied species of Cytisus (holopetalus, purgans, ephedroides, equisetiformis, candicans, albus, &c.). Most remarkably also this arrangement occurs in a small species of Broom (Genista pilosa), which is distributed over the mountains of Central Europe, over the heaths of the Baltic Lowlands, Denmark, Belgium, and England. And the presence of this contrivance here is the more strange, from the fact that the green branches with their furrows, in which lie stomata, are not leafless, but, on the contrary, are provided with a comparatively well-developed foliage. Among the most peculiar plants whose stomata are concealed in hidden nooks. MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. 299 impenetrable by water, are two very small orchids, of which one, Bolbophyllum minutissimum, grows in company with mosses on blocks of sandstone and on the bark of trees in the rocky ravines near Port Jackson, and on the Richmond River on the east coast of Australia; the other, Bolbopltyllum Odoardi, lives in similar Fig. 69.— Stomata in the Furrows of Green Stems. « Branch of Cytisus radiatus ; natural size. 2 Portion of a branch ; x 10. s Cross section of this branch ; x30. * Part of the same section; xl50. s Branch of Casuarina quadrivalvis ; natural size. « Portion of a branch; x8. ' Cross section of this branch; x30. 8 part of the cross section ; xl30. situations in Borneo. Both have a filamentous rhizome from which spring rootlets (from 2 to 5 mm. long and O'S mm. thick), arranged in pairs, by which they attach themselves to the stone and the bark of trees. Above the origin of each pair of rootlets is a little disc-shaped tuber, from 1^ to 3 mm. in diameter, and ^ mm. thick, with an aperture on the upper surface, scarcely xV mm. broad, leading into a hollow chamber within the disc -shaped tubers, about 0-5 mm. broad and O'l mm. high (see figure 70). The leaves of Bolbophyllum minutissimum are reduced 300 MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. to tiny pointed scales about ^ mm. in length; two of them are situated at the mouth of each cavity, and are inflected towards one another across it. In Bolbo- phyllum Odoardi, each of the small tubers bears only one small green leaf, which is about 1^ mm. long and 1 mm. broad, and is placed close to the opening of the chamber (see fig. 70*'^'^). Stomata are found exclusively in the interior of the hollow tubers. Water cannot enter through the narrow mouth into the air-containing chamber, and even when, in the rainy season, the whole of the mossy carpet, in which these smallest of all orchids are interwoven, is saturated with water, their transpiration continues unhindered, provided that the other conditions on which it depends are fulfilled. It is obvious that these structures which prevent moisture reaching the stomata during the wet season of the year can take on another function ^^ Fig. 70.— Orchids whose Stomata lie in Hollow Tubercles. ^ Bolbophylluin minutissimum. 2 j\^ tuber seen from above ; x8. * Vertical section through this tuber; xl6. * Bolbophyllti Oduardi. s A tuber; x6. « Longitudinal section through this tuber ; x6. in a succeeding dry period, which may follow immediately; but this must be spoken of again later. The occurrence of "rolled leaves", which are observed in so many plants of widely different affinity, is also connected with the keeping of water from the stomata. The rolled leaf is always undivided, of small area, generally linear, but sometimes ovate-linear, elliptical, or even circular in outline; always stiflT, and usually ever- green, and therefore living through two or three periods of vegetation. Its edges are bent down and more or less rolled back, even whilst still hidden in the bud. In consequence of this, the lower side which faces the soil is hollowed to a greater or less extent, while the upper side, turned skyward, is arched. Frequently the leaf is rolled so as to inclose an actual chamber, which only communicates with the outer world by a very narrow fissure, as is the case, for example, in the Crowberry (Evipetrum). The rolled-back margins of the leaves in this plant almost touch one another, and the epidermis of the lower side of the leaf forms the actual lining of the cavity which resulted from the rolling of the leaf (see fig. 71 ^). MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. 801 If the bent-back margins do not fit so closely together, a groove appears on the under side of the leaf, which is more or less sunken according to the extent of the rolling, as for example in the Heaths {Erica caffra, vestita, &;c., see fig. 71 ^). Occasionally a groove is developed which divides into two side furrows running s ^' ^ show these structures (which may be considered as counterparts of the cuticular pegs on the bamboo leaf) on the under side of Azalea procumbens, Erica caffra, and Andromeda tetragona, as well as on the edges of the fissure which leads into the hollow leaf of the Crow- berry (Empetrum nigrum). These structures are to be found almost without exception in the heathers of the northern moors as well as in the Mediterranean and Cape flora. The importance of this continuous delicate coat lies chiefly in the fact that air adheres to it as to the cuticular pegs of the bamboo leaf, and indeed so firmly that even water, under considerable pressure, is not able to displace it. On placing a leaf of Azalea procumbens under water, two elongated air-bubbles are seen along the two longitudinal furrows, which glisten like two strips of silver. Even shaking the leaf to and fro will not dislodge these air-vesicles, and even if the branch has been left submerged for a week, this air will still cling to the depressions in whose depths the stomata occur. If the branch be removed from the water it will be seen that the upper side of the leaves is wet, while water has been kept away from the stomata of the under side. And as with Azalea procumbens, so is MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. 303 it with all other rolled leaves, whether they belong to Cape plants or to heath plants of the Baltic lowlands. It cannot be doubted that the mechanism of rolled leaves, as just described, furnishes a protection for the storaata against moisture, and keeps open a passage for aqueous vapour and excreted gases. The question is now only how it comes about that this arrangement is to be met with in plants of such widely distant countries and under such differences of climate ? In order to understand this clearly, let us imagine ourselves in some of the regions which are specially characterized by the abundance of plants with rolled leaves. First, on one of the high ridges of the Central Alps, where the low-lying Azalea spreads a thick covering over the soil, where Erica carnea in great quantity Fig. 72. —Vertical Section through a Rolled Leaf of the Trailing Azalea {Azalea procumbens) ; x 140. covers broad slopes, where Di^yas octopetala, Salix reticulata, Homogyne discolor, Saxifraga ccesia, and many other plants which possess evergreen rolled leaves weave their carpet over the stony earth. The ground in which all these plants are rooted, and from which they draw their fluid nourishment, has many natural dikes and retains a large quantity of water, not only from the melting of the heavy winter mantle of snow, but also from the abundant rain of summer. For weeks together the heights are wrapped in a cold mist which saturates everything with moisture, and drops of water hang from the stems and leaves, unable to evaporate as long as the air remains so supercharged with vapour. At length the sky clears again, and the water on the plants begins to disappear. But even during the fine night following, all the plants become covered with a very heavy dew in consequence of their rapid cooling and radiation, and this not unfrequently remains until the middle of the next day. Transpiration at last occurs when the sun shines, and particularly if dry winds sweep over the heights. But who knows how long this state of things will continue ? Each moment is precious, and every hindrance to the evaporation, so important for the plants, would be a distinct disadvantage. The outlets for aqueous vapour on the under side of the leaves especially should not be obstructed, and the above described contrivance is arranged with this end in 304 MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. view. It can hardly be doubted that the earher mentioned plants of high moun- tainous regions cease to transpire for weeks at a time in the wet seasons, when a thick unbroken mist covers the slopes, and earth, stones, and vegetation are dripping with moisture; and of course the conduction of food-salts to the green leaves is interrupted to a corresponding extent. If one considers how short a period is afforded to plants of high mountain districts in which to perform their year's work, it will be understood how the most active means for promoting transpiration must be brought into play, and how everything which might interrupt or limit this process, so important to the welfare of the plant, must be avoided. A few months after the last snow has melted on the heights, fresh snow again falls, and entirely prevents growth and nourishment during the long winter. These climatic conditions account for the fact that so many Alpine plants, almost all those having rolled leaves, are evergreen. It is necessary that every sunbeam during the short vegetative period should be utilized, and that the leaves retained from the previous year should be able to transpire and to form organic materials on the first sunny day after the winter snow has melted, although the soil may have become only slightly warmed. It may perhaps be urged against this explanation that though, in the steppes the period of vegetation is restricted to the brief space of three months, nevertheless evergreen plants with rolled leaves are completely absent. But the conditions of moisture on the steppes during this three months' vegetative period are essentially different from those of the high mountain region. In the steppes, transpiration is never brought to a temporary standstill by too much moisture; evaporation can take place uninterruptedly from the leaves, and they have to be protected not from moisture, but from over-transpiration. With the exception of the halophytes and a few other growths w^hich are particularly well protected, no plants, on account of the extreme dryness of the air, can retain their green foliage in the height of summer on the steppes. Some of the plants which adorn the high mountains of southern regions make their appearance in the lower plains of the extreme north. The same carpet of Trailing Azalea, Dwarf Willows, and Dryas (Azalea procunihens, Salix reticulata, Dryas octopetala) is found on the soil underfoot. In addition are other small plants which remain green during the winter (e.g. Cassiope tetragona), which are similarly provided with rolled leaves. Even if we were not informed by Arctic explorers that the number of foggy days in the course of the short Arctic summer is much greater than on the mountain heights of the south, and that therefore a help instead of a hindrance to transpiration is required, the utmost use being made of the short time in which it is possible to draw food-salts from the soil, we might infer this to be the case from the frequent appearance of these small carpet-forming plants with their evergreen rolled leaves. Apart from other considerations, and disregarding the development of the various floral areas in point of time, the above signification of the evergreen rolled leaves explains the similarity and partial identity of the arctic flora with that of the heights mentioned. Let us turn now to the low-lying country along the North and Baltic Seas, and MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. 305 to the lowlands, which extend as far as the northern slopes of the Alps. Where man has not transformed the ground into arable soil, only moor and heath, heath and moor, are seen in wearisome monotony. On the moors especially are always the same plants — various Heaths (Calluna vulgaris, Erica Tetralix, Erica cinerea), Black Crowberry (Empetrum nigrum), Whortleberry (Oxycoccos palustris). Marsh Andromeda (Andromeda polifolia). Wild Rosemary (Ledum palustre) — all plants with evergreen rolled leaves, just as on the mountain heights. Some of these small evergreen bushes, viz. the Crowberry and the common Ling (Calluna vulgaris), may be traced in an unbroken range from the plains up to a height of 2450 metres on the slopes of the Alps. Strange to say, these plants do not blossom much earlier on the lowlands than on the high Alpine regions, and it has actually been shown that Calluna blooms rather sooner at a height of 2000 metres than in the northern portion of the Baltic lowlands. How is this ? The winter snow has long disappeared from the lowlands, while the hill-sides above are still concealed under their cold white covering. The winter snow has gone, to be sure, but not the winter! While every- thing around is already in blossom, while the ear is already visible on the stalks of rye, the neighbouring moor is still dismal, waste, and lifeless. A month or so later there is a stir on the dry soil of the cold moor, and the absorbent roots of the plants which have evergreen rolled leaves commence their activity. When the warm days of midsummer arrive and the sun sends down its powerful rays, the temperature of the soil quickly increases, and indeed rises far more than would be thought possible. The damp cushion of bog-moss at mid-day feels quite warm; and a thermometer placed 3 cms. below the surface in the uppermost mossy layer of a moor on a cloudless summer day (22nd June) showed a temperature of 31° C. while the tem- perature of the air in the shade was 13°! An unpleasant vapour rises from the damp earth, which settles on the surface, and makes a walk over the moor particularly disagreeable. Scarcely has the sun set in glowing red on the horizon when this vapour condenses into patches of mist which settle over the dark expanse; stems, branches, and leaves are covered with drops of water, and next morning everything is as thoroughly soaked as if it had rained throughout the night. This process, which is regularly repeated during the fine weather, is only interrupted when a damp wind from the sea blows, driving masses of cloud over the heath, or when copious rain saturates the soil. It needs no further showing that under such condi- tions an abundant and continuous transpiration from plants is impossible, and that in the short intervals which are allowed to the leaves for transpiration, the outlets from the wide-meshed spongy parenchyma must not be obstructed; and it does not need further proof that the evergreen rolled leaf is the form most suited and adapted to these conditions. The flora of the Cape of Good Hope may not unjustly be compared with that of the Baltic lowlands — countless low bushes which are very like Heaths, Wild Rose- mary, and Crowberry in appearance — all with small rigid evergreen leaves, and entire rolled-back margins; the upper side of the leaf usually dark green, the under side having the same arrangements as shown in the rolled leaves of plants growing on Vol. I. 20 306 MAINTENANCE OF A FREE PASSAGE FOR AQUEOUS VAPOUR. moors which border the Baltic Sea, and in the cold Arctic tundra. This shrubby evergreen vegetation of the Cape belongs indeed in part to the same families as these. Heaths especially are to be found in abundant variety; as many as 400 species can be counted — many more than are furnished by the whole of the rest of the world taken together. But a great number of species from other families, viz. Rharaneae, Proteaceae, Epacrideae, and Santalaceae, possess an exactly similar foliage, and without blossom and fruit are often indistinguishable from the heaths. This low evergreen bush vegetation is not distributed all over the Cape, but is restricted to the neighbourhood of the coast, to the country which slopes in terraces down to the south-west, and to the celebrated Table Mountain, rising abruptly above Cape Town. The aqueous vapour brought by the sea- winds condenses directly over these regions, and for five months, from May till the beginning of October, the soil is not only soaked by abundant rain, but what is perhaps of even greater moment, all the ever- green bushes are kept in a damp condition by the falling mist, and often are dripping with water just like the heaths on the moors of the Baltic lowlands. When the development of vegetation on the lower terraces of the south-west coast is at a standstill on account of the increasing dryness, the summit of the Table Mountain is still enveloped in the celebrated mass of cloud known as the "table- cloth ", and the plants growing on the ridges and plateaus are during this time as much saturated as the Trailing Azalea, which robs the south wind of its moisture on the mountain ridges of the Central Alps. It is, however, in this damp period that the growth of the plants in question takes place. Most of the plants on the heights of the Table Mountain blossom and put forth new shoots in February, March, and April; on the lower terraces from May to September. In the northern regions and on mountain heights the beginning and end of the year's work in plants is limited by the cold, but in the Cape the dryness of the soil is the cause which brings the upward current of the sap in vegetation to a standstill for so long a time. At the coast, however, this dryness is never so severe that the plants are exposed to the danger of withering up altogether, as on the steppes. As on the south-west coasts of the Cape, so is it round about the Mediterranean Sea and in the west of Europe, which is swept by sea-winds laden with vapour from the Atlantic; for example, Portugal and the south-west of France, which are in like case, characterized by an abundance of low bushes, with evergreen rolled leaves, and especially by some gregarious heaths. Here also the year's growth takes place in the wettest season, and arrangements must be made that during this period the formation of organic materials, the withdrawal of food-salts from the soil, and consequently unhindered transpiration may be carried on. Here, too, dry- ness interrupts the activity of the absorbent roots, and the evergreen vegetation of the coast-line extends inland as far as the damp sea-winds make themselves felt; while still further inland a steppe-like vegetation preponderates. The analogy pre- sented by the Mediterranean flora goes so far that, on the southern point of Istria, for example, which may be compared as to shape with the south point of Africa, quantities of the evergreen Erica arborea are only to be found on the south-west PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. 307 coast district on a comparatively narrow strip of land; while in the interior of Istria the waste dry terraces of the Tschitscherboden (which might be compared with the arid plains of the Cape) show no trace of a heath vegetation. Why the plants with evergreen leaves which grow in the far north, on the heights of the Alps, on the Baltic lowlands, on the shores of the Atlantic Ocean, on the borders of the Mediterranean basin, and at the Cape of Good Hope are not all of the same species, is a question which cannot be answered here; yet it seems proper to point out that all plants furnished with evergreen rolled leaves, whose year's work is stopped by dryness, would freeze in countries where the earth in winter is covered with snow, i.e. the molecular structure of their protoplasm would be entirely altered by the frost, which would kill it; while the protoplasm of the analogous northern forms would suffer no harm from the cold. It is well worthy of I remark in this connection that some of the last-mentioned plants have an extra- j ordinarily wide distribution; that they may actually be found, quite similar in appearance, in the bleak north, and in the southern districts, if only those conditions I of moisture which we have shown to account for the form of the leaves obtain in the places mentioned. Thus the Irish Heath (Daheocia polifolia) may be found along the Atlantic coast as far as Portugal, and the common Ling {Calluna vulgaris) grows just as well at a height of 2450 metres above the sea beside the glaciers of the (Etzthal in the Central Alps, as further south on the Abazzia, surrounded by laurel groves on the sea-coast of Istria. 3. PKEVENTION OF EXCESSIVE TRANSPIRATION. Protective arrangements on the Epidermis. — Form and Position of Transpiring Leaves and Branches. PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. The relation of the form of the evergreen rolled leaf to transpiration is anything but exhausted in the foregoing account. The part played by this form of leaf, in particular during the dry season of the year, yet remains to be discussed. If it is necessary during the wet period that transpiration should be increased as much as possible, and that everything which might restrict the exhalation of aqueous vapour from the stomata should be kept away, it is also of importance that on the appear- ance of the dry season the equilibrium between the water taken from the soil and the water excreted by the leaves should not be destroyed, and that an excessive evaporation from the portions of the plant above ground should be hindered. New seasons bring new problems to be solved. At the time when the water-current begins to ascend from the soil saturated by the winter rains, we have an aid to transpiration; later on, in the dry period, we have a protection against the dangers which might attend excessive evaporation. It is certainly of great interest to see 308 PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. how a whole group of the arrangements discussed above, among which the rolled leaf is not the least noticeable, serve, at different seasons of the year, and often at different times of the same day, two distinct purposes, as indicated. First, the stomata themselves. While the green tissue has need of food-salts from the soil for the manufacture of organic materials, they cannot be too widely opened; everything is welcome, then, which promotes transpiration, and conse- quently assists in the elevation of fluid nourishment from the saturated soil. But if the temperature and dryness of the air increase after the green parenchyma has finished its yearly task, or if the soil from which the absorbent roots have hitherto derived their supply of fluid become so dry that the water exhaled from the aerial positions can no longer be replaced, it is of the greatest importance that the stomata should be closed. This is brought about by the two cells bounding the stoma, which have been termed the " guard " cells. In order to clearly understand the mechanism of the opening and closing of stomata, it is necessary to examine the structure of these guard-cells more in detail. Both are bean-shaped in outline, their concave surfaces being turned towards the stoma; they are only connected with one another at their extremities. By their convex sides they are in contact with ordinary epidermal cells; their outer walls are in contact with the atmospheric air, and their inner walls with the spongy parenchyma. Both the innermost and outermost walls of the guard -cell are strongly thickened, but the wall by which they are connected with neighbouring epidermal cells, as well as that portion which directly borders the stoma, is relatively thin, elastic, and extensible. If the figure of two such guard-cells be imitated in caoutchouc, and they be fitted together like an actual closed stomate — water being forced into them under considerable pressure — the curvature of the thin and elastic portions of the walls will be most altered. The side wall in contact with the neighbouring epidermal cell bulges out, and at the same time the whole cell becomes elongated in a direction perpendicular to the surface. By this means the two guard-cells are forced apart. When the water is allowed to flow out of the swollen caoutchouc cells, they again fall back into position, the two portions of the walls which border the stoma coming into contact with each other and closing the opening. The same thing occurs in the actual guard-cells of the living plant. As soon as they become distended, they separate from one another; when they relax and resume their original position, they come closely into contact again. This process bears a strong resemblance to the changes in the cells of the pulvini at the base of the sensitive leaves of Mimosa, which will be described later, and it is highly probable that it may be traced back to a similar stimulation. That the guard-cells actually separate from one another by swelling up, i.e. by absorbing fluid, and then close together again in consequence of the loss of water, can be shown by first supplying water and then withdrawing it by a solution of sugar. In the former case the stomata open, in the latter they close, and it may therefore be considered an established fact that a closing movement is brought about by the extra loss of water in dry air. But if these pores, through which water vapour escapes when the plants PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. 309 are full of sap, close as soon as there is a danger of too much evaporation, the mechanism must be considered as excellently regulating transpiration, and as pro- viding a true preventative against over-evaporation. This closure of the exhalent chambers in the interior of the leaf, important as it is, would alone be sufficient in but a very few cases to ward off this threatened danger. If the epidermis which stretches over the thin- walled transpiring cells of the spongy parenchyma is itself thin-walled and succulent, water will be exhaled from it also into the dry atmosphere; this loss of water from the epidermal cells is compensated for by water drawn from the neighbouring parenchymatous cells in the interior of the leaf, and ultimately the leaves would wither up if the supply of water from the roots were stopped or became insufficient. Therefore the epidermal cells must be adequately prevented from exhaling. When this is the case, and when the stomata are closed, the spongy parenchyma, and, generally speaking, all the succulent cells in the interior, are securely protected. The walls of the epidermal cells in the first stages of their development are composed mainly of cellulose, and are uniformly thin and delicate on all sides. The outer wall, which is in contact with the air, then becomes thickened and divided into an inner and an outer layer. The inner retains its original character, but the outer — the so-called "cuticle" — undergoes great modifications. The cellulose becomes changed, and is replaced by a mixture of stearin and the glyceride of a fatty acid (suberic acid), forming a tallow-like fat which is termed cutin or suberin. In consequence of this metamorphosis the cell-wall becomes less and less permeable to water, and when it has attained a considerable thickness it becomes at length almost entirely impervious to water and aqueous vapour. Frequently, between the inner cellulose and the outer corky layer, other so-called " cuticularized layers " are formed, whose chief constituent is again suberin, and which often attain to a con- siderable bulk. Aquatic plants, which are not exposed to the danger of excessive evaporation, of course do not require this protection. Plants whose leaves are surrounded by air, on the other hand, can never entirely dispense with it. The thickness of these corky layers is extremely variable according to the condition of humidity of the air. Where the air is very damp throughout the year, the outer wall of the epidermal leaf -cells appears to be only slightly thicker than the inner, and the cuticle only forms a thin continuous layer. On the other hand, plants which are temporarily exposed to dry air possess very highly developed cuticular strata. Especially when the leaves are evergreen and remain several years on the branches, as, for example, in the Holly (Ilex Aquifolium, see fig. 73 2), a^^j {^^ the Oleander (Nerium Oleander, fig. 73^), the cuticular layers are so strongly developed that the outer wall of the epidermal cells is many times thicker than the inner wall. Evergreen parasites, as, for example, the Mistletoe (fig. 73^), those tropical orchids and Bromeliaceoe which live epiphytically on the bark of trees and are often exposed to great dr;yTiess in the hot season of the year, cactiform plants, and generally the majority of succulent plants, possess epidermal cells with very strongly thickened outer walls. This is so 810 PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. also in the case of the pines with evergreen needle-shaped leaves, \\here, as a rule, the water compensating for that exhaled by the leaves cannot come quickly through open channels, but only slowly through the woody cells. Usually the cuticle and cuticular layers are of equal thickness over the whole leaf surface; this is so espe- cially in smooth, shiny, leathery evergreen leaves. Not infrequently, however, an irreo-ular thickening is seen, particularly in the neighbourhood of stomata, where circular ramparts are raised, as in Protea mellifera (see fig. 67 ^), or peg-shaped pro- jections are formed, as in the Bamboo (see fig. 66), or elongated hair-like filaments arise, as in the rolled leaves of Azalea and many Heaths (see figs. 71 and 72). It would, however, be erroneous to suppose that this formation of a thick cuticle on the epidermis is a peculiarity of evergreen leaves. Plants which are surrounded ^DtiCJQP r;^-^:V^»^r""'-wo'r,v^ -Thickened Stratified Cuticle. 1 Vertical section of a portion of the leaf of Mistletoe (Viscum album); x420. 2 Vertical section of a portion of the leaf of Holly {Ilex Aquifolium); x500. 3 Vertical section of leaf of Oleander (Nerium Oleander); x320. all the year by a damp atmosphere, and are never exposed in their natural condition to the danger of too much evaporation, very often have evergreen leaves, and yet the outer wall of the epidermal cells is not at all, or only very slightly, thicker than the inner; and conversely, plants with apparently thin delicate leaves, which are green only in the summer, have quite conspicuous thickening-layers. A knowledge of these conditions is of the utmost importance in plant culture, and gardeners know very well that many plants, although they appear to be capable of resistance, can never be removed from the damp air of the greenhouses, because the leaves then become desiccated like those of aquatic plants which have been taken out of water and exposed to the air. A species of palm, Caryota propinqua, which is repre- sented in its native habitat in fig. 74 opposite, was grown in the botanical gardens at Vienna, and it developed in the damp air a magnificent stem with fine large leaves. On a summer day, when the temperature of the open air coincided with that of the greenhouse, this Caryota, together with the tub in which it was rooted, was carried into the open and placed in a somewhat shady place, but partly exposed PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. 311 J If fir" I /fv Fig. 74. -Caryota propinqua. 312 PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. to the sun's heat. One day, after a warm dry east wind had swept for only a short time over the foKage, it became quite brown, and in the evening all the leaves were entirely dried up and dead. And yet leaf -segments of this palm appear to be firm, leathery, and dry, and one would have supposed them to be particularly well pro- tected against drying up. The section of part of a leaf which is represented in fig. 75, however, corrects this impression. This shows that the epidermal cells are certainly very compact, by which the firmness of the leaf is materially increased, but that their walls are not thickened, being only like those of a delicate fern in this respect. Under these small thin-walled epidermal cells lie large succulent cells which form the so-called aqueous tissue, the structure of whose walls likewise cannot limit evaporation; below these are the large succulent cells of the green tissue. A glance at this leaf section will make it clear that this palm is well , Fig. 75. — Vertical section of a portion of the leaf of Caryota propinqua; x2® adapted to its warm damp habitat, where it is never exposed to a strong evaporiza- tion, but not to the dry, even if warm, air of a Continental climate. To the wax-like excretions of the cell-wall which form a delicate bloom, easily rubbed ofi", on both sides of the leaf, frequently colouring it pale blue, grey, or white instead of dark green, it has already been stated that the role is assigned of protect- ing the stomata from moisture. From what has been said, one would expect that these waxy coverings, which are especially to be met with in the Cruciferse and Rutaceae of steppes, in many acacias and Myrtacese of Australia, and in the pinks and spurges of the Mediterranean flora, would also be able to limit transpiration in the epidermis — that is, in the structures over which the bloom-like covering extends. Experiments specially undertaken, have also shown that in the same space of time, and under otherwise similar conditions, leaves whose bloom had been carefully rubbed off lost almost a third more water than others whose waxy covering had been left intact. That the varnish-like covering of the epidermis, composed of a mixture of mucilage and resin (" balsam "), which is excreted from capitate hairs and other glandular structures, is able to restrict transpiration has also been pointed out. These coverings are especially developed in many plants of the Mediterranean flora, particularly in a whole group of Cistus (G. laurifolius, populifolius, Clusii, ladani- ferus, Tnonspeliensis, &c.); further in shrubby plants which develop late, in the height of summer — ^as, for example, in Inula viscosa, which is so abundant on the PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. 313 coast. Plants of steppes and prairies {e.g. Centaurea Balsamita of the Persian steppes and Grindelia squarrosa in the prairies of North America) are likewise protected throughout life from over- vaporization by varnish-like coverings of this kind; while the foliage of Cherry, Apricot, and Peach trees, as well as of Birches, Sweet Willows, Balsam and Pyramidal Poplars, and the Black Alder, is only covered with such a varnish while young, when it has just burst from the buds, and the outer walls of the epidermal cells have not yet become sufficiently thickened; later on, however, when the cuticularized layers have become fully formed, this covering which limits transpiration disappears. Only on those places of the epidermis, where the outer walls of the cells remain very thin and permeable by fluids and gases, is this coat of balsam retained until the leaf is to be thrown ofi"; but in this case it probably regulates the absorption of atmospheric w^ater. How far the incrustations of lime and salt excretions take part in the absorption of atmospheric water by organs situated above the ground has likewise already been considered in the section on water absorption. It is obvious that these concretions and coverings of the epidermis must be capable of restricting transpiration. Incrustations of lime are principally found in plants which grow^ in the clefts and crevices of rocks; excretions of salt are only observed in shore-plants and those of steppes and wastes, but then always on low bushes and shrubs with small narrow leaves, and herbs whose foliage rests on the soil. The reason for this is again easily found. High trees could not support the weight of leaves loaded with incrustations of lime and salt, even if their trunks and branches possessed the greatest strength imaginable. It has been observed that plants whose leaves are covered by incrustations of lime and salt, or whose epidermal cells are strongly thickened on their outer walls by corky layers, are almost always destitute of hairs; while plants, on the other hand, whose epidermal cells possess delicate outer walls, if they are not surrounded by a damp atmosphere throughout the year, nor submerged in water, are usually furnished with structures known as plant-hairs (trichomes); from which it may be inferred that the hairy covering of the leaf or stalk in question is able to protect it from drying up in just the same way as the corky layers. Of course only those hairs are meant whose protoplasmic contents have disappeared, and which have become sapless and filled with air; for those hair-structures, which consist of cells rich in sap and osmotic contents, would not help in preventing evaporation from the deeper tissue ; they are themselves in need of protection, and special protective arrangements exist for them, as already set forth in the discussion on the absorption of water by aerial portions of the plant. Such structures would, if unprotected, give off water to the surrounding air, and continually absorb fluid from adjacent cells below them. This action does not take place in air-containing cells, and if their dry membranes, and the air which they inclose, are interpolatetl between the dry atmosphere and the succulent tissue below, this latter wall be protected from evapo- ration, like damp earth covered with a layer of dry straw or reeds, or the fluid at the bottom of a bottle whose neck is closed with a plug of cotton-wool. 314 PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. The importance of air-containing cells as a covering for succulent tissue must also be considered in another relation. It is well known that evaporation from the surface of fluid or a damp body is much increased by the warmth of the sun's rays. On the other hand, if the heating is restricted, so also is the evaporation. If we use a dry cloth to shade from the sun, we lower not only the temperature, but also the amount of evaporation from the shaded body. The covering of air-containing hairs on leaves may be compared to such dry screens, and its action may be demonstrated by the following experiment: — Take two of the bi-coloured leaves of a Bramble bush, which are smooth on the upper side, but covered with a white felt-work of hairs on the lower, and which are exactly similar in size and position with regard to the sun, being situated very near each other on the stem. If these leaves are wrapped round thermometers, in such a way that the leaf which covers one thermo- meter bulb has its white felted side turned towards the sun, that covering the other, the green hairless side, it will be found that the temperature in the leaf whose smooth green side is directed towards the sun will in less than five minutes rise 2°-5° above that of the leaf whose white felted side is so directed. If such leaves are plucked and exposed to the sun, some with the white felted side, others with the smooth green side uppermost, the latter always shrivel and dry up much sooner than the former. There can be no doubt, after this, that a dry coat of hair over succulent plant tissue, which is exposed to the sun's rays, considerably restricts the heating of, and exhalation from this tissue. The significance of the coverings of hair on portions of plants turned away from the sun, particularly on the under sides of flat and rolled leaves, has already been discussed. These coverings are only of slight importance as a means of protection against over-transpiration. In rare cases, indeed, it happens that the hairy covering on the side of the leaf turned from the sun, the lining of the leaf, so to speak, must act as a protection, since the flat leaf-lamina is so twisted and turned that the sun's rays strike not on the upper but on the under surface. There are certain ferns of Southern Europe {Ceterach offLcinariim, Cheilanthes odor a, Notochlcena Marantoe), which, contrary to the habits of most of this shade-loving group, grow on blocks and walls which are exposed to the burning sun. In these ferns the upper surface of the leaf is smooth, but the under, on the other hand, is thickly covered with dry hair-scales. In wet weather the leaves are spread out flat, with the smooth surface uppermost; in dry weather they become rolled up, and the under cottony side is then exposed to the sun and to dry winds. Among the low herbaceous growths of the Mediterranean flora, a like behaviour is shown by the widely distri- buted Hawkweed, Hieracium Pilosella, whose radical leaves, forming a rosette on the soil, appear green on the upper and white oh the under side, by reason of a felt- work of star-shaped hairs. In places where the ground easily dries up, and when there have been no showers for a long while, it is usually seen that first the margins of the leaves turn up, and then by degrees the whole leaf becomes bent and rolled, so that the lower side is turned towards the sun's rays, and the white felt of hairs functions as a protective screen to the whole leaf. PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. 315 The relations between the hairy covering on the upper side of the leaf and trans- piration stand out, most strikingly, in those districts where plants during their veo-etation period are, as a rule, exposed to dry air only for a few hours each day, and where their activity is not interrupted by a long warm dry period, but by frost and cold — as is the case, for example, in the Alpine region of mountain heights. On the Alps, the drying up of flowering plants by the sun only occurs in a very few ^» J^ ti^' 70 — l,deU\eiss {Gnaphalium Leontopodnim). cases, viz., where the scanty soil on the narrow ledges of steep projecting rocks, and crags, and on rocky slopes, &c., is only watered by rain, mist, and dew. If no showers fall for several successive days, and the south wind blows over the heights with a clear sky day and night, these scanty layers of soil may dry up to such an extent that they are unable to supply the necessary fluid food to the plants rooted in them. Under these circumstances plants growing there have most pressing need of means of lessening transpiration in the leaves. In places such as these are to be found, almost without exception, plants whose leaves and stems are thickly covered on all sides with hairs, together with succulent plants and saxi- frages incrusted with lime. This is the habitat of the felted Whitlow-grass (Draha 316 PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. tomentosa, stellata), of the grey-leaved Ragwort {Senecio incanus and Carniolicus), of the magnificent silky Cinquefoil (Potentilla nitida), and of the white-leaved bitter Milfoil (Achillea Clavennce); especially is this the habitat of the most celebrated Alpine plants, of the scented Edelraut and the beautiful Edelweiss — the former (Artemisia Mutellina) with a grey shimmering silky coat, the latter {Gnaphalium Leontopodiutn) wrapped in dull white flannel. On looking at the vertical section of the Edelweiss leaf (see fig. 77^), one sees that the epidermal cells with their thin outer walls would be unable to regulate exhalation and drying in the sun, and that a powerful protection is afibrded against too rapid evaporation, in case of extraordinary dryness, by the possession of a layer of sapless, air-filled, interwoven hair-structures. The Edelraut, Ragwort, and the other plants named, which grow on the sunny rocks of the Alps, show these same characters of leaf structure, and what has just been said about the Edelweiss applies fully to them also. It should be mentioned that on the heights of the Pyrenees, Abruzzi, and Carpathians, as well as on the Caucasus and Himalayas, the plants growing on sunny ridges of rock, where they are exposed to the wind, are covered with silk and wool exactly after the model of the Edelraut and Edelweiss, and that there is on the Himalayas an Edelweiss which is wonderfully similar to that of the European Alps. In the far north, on the other hand, where the flora in other respects has so much in common with that of the Alps, these plants are absent, and generally a search over the rocky crags for herbs and shrubs, whose leaves are furnished with silky or felt- like coverings on the upper surface, is futile. The genera which grow on these places and form a characteristic feature of the vegetation in consequence of their great abundance — as, for example, Diapensia Lapponica, Andromeda hypnoides, Mertensia maritima, Draha alpina, and others, possess remarkably smooth green leaves. When hairy coverings are present, they are restricted to the under leaf- surface, especially to that of rolled leaves. They are never found on the plants of rocky slopes, but only on those of damp marshy ground, or by the side of water which is for a short time free from ice. Here, however, they certainly do not help to lessen transpiration, but function in the way described above in the discussion on rolled leaves. It is indeed not too much to connect these facts with the conditions of the climate, and especially to explain the absence of plants whose foliage is silky or felt-like on the upper surface, by saying that a drying up of the soil and a limiting of the water supply never occurs on the narrow terraces of steep rocky declivities in Arctic regions, and that therefore there is no danger of over-evapora- tion to plants growing in those regions. It is in keeping with this explanation that on Central and South European mountains, on whose heights an Alpine vegetation is to be found, the number of forms having silky and felted foliage increases as these mountains are situated further south, and the more they are exposed to temporary dryness. Plants of the Edelweiss type are still wholly foreign to the Riesen-Gebirge; in the Northern Alps their number is comparatively small, in the Southern Alps they increase in a surprising manner, and the summits of the Magellastock. the ridges of PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. 317 the Sierra Nevada, and the mountains of Greece are unusually rich in such forms. If plants growing in such situations are protected against the dangers of too rapid and too abundant evaporation, how much more must this be the case in those regions where, with the increasing warmth of summer, the number of showers steadily diminishes; and where the soil becomes dried more and more deeply, so that all the plants whose roots are near the surface are unable to derive a drop more water from it ? All plants which are to survive the dry period in such places must during this time entirely cease transpiring — they must, as it were, turn into a chrysalis and sleep during the summer. They actually do this in all sorts of different ways, and by tlie most diverse means. One of the commonest and most widely spread methods is, without doubt, by having the transpiring organs clothed with a thick covering of dry air-containing hairs. Plenty of examples of this are furnished by the flora of the Cape, Australia, Mexico, the savannahs and prairies of the New World, and the steppes and deserts of the Old. In the dry elevated plains of Brazil, Quito, and Mexico, there are large tracts covered with gregarious spurge- like growths and grey-haired species of Croton, and when the wind blows, moving these bushes to and fro, undulations are set up over wide extents of country, the whole appearing like a billowy sea of grey foliage. A similar picture is presented by the Painciras belonging to the Compositse, or by the Lychnophora, on the high plains of Minas Geraes in Brazil. Nowhere in the whole world, however, does the presence of hairs on foliage, as a protection against exhalation, appear in such an abundant and varied manner as in the floral region surrounding the Mediterranean, known as the Mediterranean district. The trees have foliage with grey hairs; the low undergrowth of sage and various other bushes and semi-shrubs (for which the name " Phrygian undergrowth ", used by Theophrastus, may be retained), as well as the perennial shrubs and herbs growing on sunny hills and mountain slopes, are grey or white, and the preponderance of plants coloured thus to restrict evapora- tion has a noticeable influence on the character of the landscape. He who has only heard from books of the evergreen plants of the Greek, Spanish, and Italian floras, feels at the first sight of this grey vegetation that he has been in some degree deceived, and is tempted to alter the expression "evergreen" into "ever grey". Every conceivable sort of hair structure is to be met with in these parts — coarse felt-work, thick velvet, and white wool mixed in endless variety. Here is a leaf looking as if covered with a cobweb; there another as if bestrewn with ashes or clay; here a leaf surface, covered with closely pressed hairs or scutiform scales, glistens like a piece of satin; and here again is a plant with such a long flock of hair that one might imagine that sheep in passing had left pieces of their fleece hanging on it. There is hardly a family in the flora of the Mediterranean district which does not possess members richly provided in this way. The Composites are the most remarkable in this respect, especially the genera Andryala, Artemisia, Evax, Filago, Inula, and Santolina; then come the Labiates of the genera Phlomis, Salvia, TeucHum, Marrubium, Stachys, Sideritis, and Lavandula; rock-roses, bindweeds, scabious. 318 PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. plantains, papilionaceous plants, and plants of the Spurge-laurel family — just those plants which constitute the main part of the vegetation on the shores of the Mediterranean Sea, and which possess a thickly -woven covering of hair. Indeed representatives of families such as the Grasses, whose members are usually bare, here appear to be quite shaggy with hair. It is also very interesting to see that so many species, which have a wide range of distribution, and which, from Scandinavia to the coasts of the Mediterranean, have bare foliage, can in the South protect them- selves from drying up, by developing hairs on their epidermis. For instance, from Northern and Central Europe as far as the Alps, the epidermis of the stems and foliage of Silene inflata, Cavipanula Speculum, Galium rotundifolium, and Mentha Pulegium is smooth and bare; in the South, — particularly in Calabria, — the leaves and stems of these species are covered with thick down. Next to the Mediterranean flora, the neighbouring Egyptian and Arabian desert regions, the elevated steppes of Persia and Kurdistan, as well as the lowlands of Southern Russia and the plains of Hungary, show a comparatively large number of species whose leaves are thickly coated with hairs on both surfaces. Their number is less than that of the flora of the Mediterranean district, because in the steppes and deserts the dryness of the summer is greater than in that region, and even thick hairy coverings are not always a sufiicient protection against this dryness, and also because in some of these districts the dry period passes directly into a severe winter, and the hairs would offer but a poor protection against the cold. Since on the coasts of the Mediterranean Sea the winter temperature never falls below freezing point, evergreen and grey leaves remain there unmolested, and recommence their activity at the beginning of the next season. The successive developments of certain plant forms are very instructive with regard to the relations existing between whole floral regions and transpiration. In the steppes, Mediterranean district, and at the Cape, bulbous plants and annuals first make their appearance; then follow the perennial grasses and woody plants; and finally succulent plants and thickly-haired immortelles. The numerous tulips, narcissi, crocuses, stars of Bethlehem, asphodels, amaryllises, and all the other bulbous growths, which begin to sprout immediately after the first winter or spring rain, always have bare foliage. Their transpiration is very active in consequence of the rapidly-increasing temperature of the air, but the saturated soil provides a sufficient substitute for the evaporating water, and also has ready in a free state the food-salts which a.re required for rapid growth. The shrubs which sprout at the same time, the peonies and hellebores, as well as the host of annuals which spring up, blossom, and fructify in an inconceivably short time, almost all possess bare foliage, especially in the steppes. Towards midsummer, when the drought com- mences, all these plants are already in fruit; their foliage, which until now has been actively at work, begins to turn yellow and to dry up; their succulent tubers and bulbs are imbedded below the surface in soil which is now as hard as a stone; and the seeds which have fallen from the annual plants are easily able to survive the aridity of the summer and the severity of the winter, since they are inclosed in PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. 319 protective coverings of great variety. Any plants which are still to retain their activity during the summer on the steppes or in the Mediterranean floral district would succeed very badly if only furnished with the bare foliage of the spring vegetation. If such a plant is to be protected from drying up, its transpiration must be lessened. This is effected by various protective arrangements, but best of all by a thick coating of hair. The papilionaceous plants and species of Orache, above all the immortelles and wormwoods {Helichrysum, Xeranthemum, Arte- misia), which are still in bloom in the height of the summer and can bear the strongest heat of the sun, are, as a rule, thickly covered with hair, and regions, which perhaps only a month before were clothed in fresh green, are now shrouded in dismal gray. With the transition from the wet period of the spring and winter rains to the dryness of midsummer, there is a corresponding gradual transition from the green of the bare, succulent hyacinth leaf to the grey of the rigid felt-covered leaf of the immortelle. A peculiar appearance is shown in Mediterranean floral districts by many biennial and perennial plants which one spring give rise to a rosette of leaves close to the soil, and in the spring following to a stem bearing both leaves and blossom, which arises from the centre of the rosette. This rosette formed in the first spring has to live through the dry hot summer, and is therefore covered with felted grey hairs; the stem formed in the second year which gives rise to the blossom, since it is formed during the wet period, has no need of the protective hairs, and is there- fore furnished with green foliage. The Salvia lav andulw folia and Scabiosa pulsatilloides of Granada, the Hieraciwm gymnocephalum of Dalmatia, and in the Mediterranean flora the wide-spread Helianthemum Tuheraria may be mentioned as examples of such plants. Their appearance is so strange that one involuntarily asks whether this green leafy stem really belongs to the grey rosette of leaves, or whether some one has not been playing a joke by putting together the stem and rosette of two different kinds of plants. These hair- like structures, called "covering hairs", whose function is a pro- tection against excessive exhalation, exhibit a very great variety with regard to form. Notwithstanding this diversity, however, a certain degree of uniformity must not be overlooked, inasmuch as in individual species the same kind of hairs are always present. The coat of hair contributes not a little to the characteristic appearance of the species, and therefore has always been considered of especial value in description and discrimination. As a help to description the older botanists introduced a series of expressions into botanical terminology by which to denote shortly and tersely the most pronounced varieties, and this seems to be the most suitable place for explaining these terms — i.e., the forms of covering hairs which are signified by them. First, those covering hairs consisting of a single epidermal cell, which grows out beyond the other epidermal cells, must be distinguished and set apart from those which have become multicellular by the formation of separation walls. Unicellular clothing hairs in many cases only project slightly above the surface 320 PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. of the leaf to which they belong; they bend down nearly at a right angle almost immediately above their place of insertion, so that the long tapering part of the hair cell lies on the leaf -surf ace, as shown in fig. 77 ^ When such hair- forms, in great numbers and parallel to one another, entirely cover the surface of the leaf, light is strongly reflected from them, and the surface looks just like a piece of silk. Such a covering of hair, which is seen particularly well on the shining foliage of the South European bindweeds (Convolvulus Cneorutn, nitidus, olecefolius, tenuissimus, &c.), is termed " silky " (sericeus). Two varieties of this may be distinguished, viz. the more usual case in which all the hairs of the leaf lie parallel with the midrib, and the rarer case where the hairs assume a different position on the right and left of the midrib, the whole of those on either side being respectively parallel to the lateral ribs of their respective sides. The reflected light then only meets the eye of the observer, in any one position, from one half of the leaf, the other half therefore appearing dull. In such a case the whole leaf has that peculiar shimmer, changing on the slightest movement, which we admire on the wings of certain butterflies, and which is also shown by that variety of silken material known as satin. When the unicellular hairs do not lie on the surface, but rise up from it, the shimmer is altogether absent, or is only present to a small extent. If the hairs are short, very numerous, and closely pressed together, they are said to be "velvety" (holosericeus); if they are of greater length and situated further apart, the expression " shaggy " (villosiis) is used. Hairs which consist of single elongated air-containing cells, much twisted and bent, with thin and pliant walls, are called wool-hairs, and the covering formed by them is said to be " woolly " or " tomentose " (lanuginosus). Woolly hairs are always twisted spirally, sometimes loosely, sometimes tightly — frequently almost like a corkscrew. As a rule the spiral is in the opposite direction to the movement of the hand of a watch, whose direction is said to be to the left. It should also be noticed whether the elongated twisted cells of the wool-hairs are circular in cross section, as in the South European Centaurea Ragusina (see fig. 77 ^), or whether they are compressed like a ribbon, as in Gnaphalium tomentosum (fig. 77 ^). The latter case is by far the most common. Multicellular clothing hairs originate by the repeated division of certain epidermal cells caused by the formation of separation walls. These dividing walls are either all parallel to the surface of the leaf or stem, or some of them are perpen- dicular to the plane of the leaf. In the first case the cells are usually arranged like the links of a chain, and are termed jointed or articulated hairs. When such arti- culated hairs are short and not interwoven — as, for example, is the case in the beautiful gloxinias (see fig. 77 ^), the surfaces clothed with them appear like velvet; when they are elongated, curved, and twisted and entwined, the leaf appears to be covered with wool (see fig. 77 ^), and to the naked eye this form of covering is the same as that already stated to be shown by unicellular covering hairs. Silky coats are also produced by multicellular hairs, even by such a peculiar form as is repre- sented in fig. 78 ^ These hairs are developed in the following manner. A super- ficial cell by the formation of a septum parallel to the leaf-surface divides into two PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. 321 daughter-cells; the division is repeated and gives rise to a small chain of three, four, or five short cells which project slightly above the surface of the leaf. The top cell does not divide further, but enlarges in a striking manner, not, oddly enough, lengthening in an upward direction, but transversely, parallel to the leaf-surface, forming a lancet-shaped, rod-like structure, which shades the leaf, and is supported by its sister cells as if on a pedestal. Thousands of such curious hair-structures, ,lif\Mi ■v^ * r-;M| X Fig. 77.— Covering Halis. » Articulated woolly hairs of Gnaphalium Leontopodium. 2 Articulated velvety hairs of Gloxinia speciosa. » Silky hairs of Convolvulun Cneorum. < Ribbon-like flattened woolly hairs of Gnaphalium tomentosiim. ' Spiral woolly hairs of Cen- taurea Ragusina. « Stellate hairs of Alyssum Wierzbickii. ' Umbrella-shaped hairs of Koniga spinosa; surface view. 8 Vertical section of the same hairs. » Stellate hairs of Draba Thomasii. x about 50. which may best be compared to compass-needles, clothe the surface of the leaf in close proximity to each other, and when they are arranged in a regular manner, tliey reflect the light uniformly, and produce a distinctl}* silky lustre. If they are twisted, this lustre is lessened to a greater or less extent. This variety of hairs, called T-shaped, is distributed in a remarkable way. Numerous species of Astra- galus, the scabious of the Mediterranean flora (Scabiosa cretica, hymeltia, gramini- folia), several Crucifers (Syrenia, Erysimum), native on the steppes of Southern Russia, the magnificent Aster an Vol. I. gophyllus of Australia, and particularly numerous 322 PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. species of wormwoods; the South European Artemisia arborescens and argentea, the Artemisia sericea and laciniata belonging to the steppes and Siberian flora, the Common Wormwood, Artemisia Ahsynthium, and the frequently -mentioned Edel- raut, Artemisia Mutellina, growing on the rocky crags of mountain heights — all owe their silky appearance to these T-shaped hair-structures. It may also happen that the cell which is elongated ti-ansverselj?- (i.e. parallel to Floccose hairs of Verbascum thapsiforme. 2 Tufted hairs of Potentilla cinerea. « T-shaped hairs of Artemisia viutelli'ia. * Actinia-like hairs of Correa speciosa. « Scutiform scales of Elceagmis angiistifolia. « Stellate hairs of Aubretia deltoidea. x about 60. the leaf -surf ace), and which is the uppermost of the small group of cells projecting above the epidermis, is prolonged in three, four, or even more directions, so as to have a stellate appearance. Thus the covering of the leaf is seen to consist of three, four, or many-rayed stars, each supported on a short stalk (see figs. 78^ and 77^). The rays of the stellate cells are frequently forked, as in Draha Thomasii (see figs. 77^). In rare cases they have a comparatively large central portion, and are only divided at their circumference into short rays; they then look exactly like small sunshades spread out over the leaf-surface. This elegant form, which is represented in figs. 77^ and 77 ^ has a particularly beautiful appearance in PROTECTIVE ARRANGEMENTS ON THE EPIDERMIS. 323 Koniga spinosa, a member of the Mediterranean flora. All these clothing haii'S, with star-shaped indented upper cells, are grouped together under the name of "stellate hairs" (pili stellati). In Cruciferae and MalvaceoB they occur in endless variety. When the uppermost cell of the group forming the stellate hair is divided by separation walls, which in part are placed perpendicularly to the leaf -surf ace, branched hairs are the result. In branched hairs the branches, which are almost Fig. 79.— Flinty armour of Rochea falcata. I Section perpendicular to the leaf-surface. 2 Surface view ; on the right hand the vesicular distended portion of a few superficial cells is removed and the stomata are brought into view ; x350. always arranged in a stellate manner and are usually unicellular, can be dis- tinguished from the part which supports the branches. This portion usually looks like a pedestal, and is sometimes multicellular, sometimes formed from a single cell. When the pedestal is very short, and the cell supported by it is divided by several radiating divergent septa, which are either oblique or perpendicular to the leaf- surface, tufted hairs (pili fasciculati) are formed. These look like sea-urchins lying on the surface in close proximity to each other; they vary very much in the size, number, length, and direction of their branches, and they are particularly abundant on the cinquefoils (Potentilla cinerea and arenaria), cistus and rock- roses (Gistus and Helianthemum). A common form is represented in fig. 78 2. When the foot-stalk is very short, and the radiating branch -cells borne by it are 324 PROTECTIVE ARRAXGEMENTS ON THE EPIDERMIS. joined to one another, a star-shaped, ribbed, multicellular plate, indented at the margin, is produced (see fig. 78^). These plates are generally flat, lie level on the surface of the leaf or stem, overlap one another with their indented margins, and cover the green surface of the leaf so completely that it appears to be white instead of green, and invest it with a bright, almost metallic, lustre. Such leaves are said to be " scaly " (lepidotus). The best known examples of such leaves, covered with shining silvery hair-scales, are those of Mceagnus and of the Sea Buckthorns (Hippophae). If the plates are bent, irregularly fringed, and lustreless, the leaf covered with them looks just as if it were strewn with bits of clay, and is said to be " clayey " (furfuraceous). Examples of this are well shown by the leaf- coverings of many plants allied to the Pine-apple (Bromeliacese). When the top cell of the hair is supported on a moderately high pedestal, and is divided into numerous radiating daughter-cells which diverge from one another, a structure is produced which is somewhat like a knout, or, if the radiating cells are short, like a sea-anemone (Actinia). This form of hair is seen, for example, in the Southern and Eastern European Phlomis, in many mulleins (Verbascum Olynipicurti), and, with multicellular pedicels, on the leaves of Correa speciosa, an Australian shrub (see fig. 78*). Occasionally a branched hair produces several whorls of branches above one another, and then hair-structures are formed which resemble stoneworts (Characeae) or miniature fir-trees under the microscope. When many such tiny tree-like hairs are placed close together with interwoven branches, they look under a magnifying-glass like a small plantation, and the analogy is heightened if one- storied tree-shaped hairs, like the undergrowth in a high forest, occur under the higher many-storied ones. This is the case in the Torchweed, Verbascu7}i thapsi- fornie, whose hairs are represented in fig. 78^, Hair-structures like these appear to the naked eye like flock, and are described as " floccose " hairs (pili floccosi). Many of these have the peculiar habit of rolling themselves together into small balls, which make the leaf-surface look as if it were bestrewn with coarse white powder. This is the case, for example, in the mullein known as Verbascum veru- lentum. In the crowded condition of stellate and tufted hairs, of branched floccose and unbranched woolly hairs, it is unavoidable that the neighbouring hair-cells should cross one another, intertwine, and be more or less interwoven; and thus arises a felted mass which covers the surface of the organ in question. Such hair-masses are termed " felt " (tomentum), and the varieties are distinguished as " felted " (or " tomentose ") stellate or woolly hairs, &c. Often the felt only forms a thin loose layer, through which the green of the leaf -surface can b^ seen: but occasionally it is so thick that the leaf appears snow- white. While in all these cases the covering which protects the leaves and stem of the plant from over-transpiration is woven from air-containing cells, cylindrical and elongated — usually, indeed, very much elongated — in some thick-leaved plants, especially in species of the genus Rochea, a native of the Cape, these cells become vesicular and distended ; they are arranged in rank and file adjoining one FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. 325 another, so that taken together they form a layer which spreads over the other epidermal cells like a coat of mail. The ordinary epidermal cells are small and only slightly thickened on their outer walls, as shown in the illustration above. The cells which are placed together to form the armour, however, are enlarged in quite an unusual way; their stalk-like base, looking as if wedged in between the ordinary epidermal cells, is indeed comparatively large, but the bladder-like swollen portion exhibits dimensions which are about six hundred times greater than those of the ordinary epidermal cells. The vesicles are closely packed together, and become almost cubical by the mutual pressure they exert on each other. Where a space might occur, the bladders form protuberances and bulgings at the side which fit in together in such a way that a completely closed coat of mail is the result. The expression " coat of mail " is the more justified here since the swollen bladder-like cells of Rochea are as hard as pebbles. Large quantities of silica are present in the cell-walls, and by burning them a complete skeleton in silica can be obtained, as in the case of the silica-coated Diatomacese. It needs no further explanation that in the dry season such a coat of aimour affords to the succulent cells it covers an excellent protection against evaporation. There is, however, still another point to be considered. The vesicular swollen cells on fully-grown leaves are still occupied by protoplasm, which forms a thin layer round the walls, while in the centre is a large cavity filled with cell-sap; it is only in older leaves that the bladder-like cells become filled with air. As long as they contain watery cell-sap they serve as reservoirs of water from which the green chlorophyll-bearing cells below can obtain supplies at the periods of greatest drought, when all other sources are exhausted. This fact, that the water-reservoirs are situated on the exterior of the plants, where there exist so many aids to exhala- tion, shows how well the silicified walls of these bladders function. They may be compared to glass vessels whose mouths are directed towards the green tissue, and whose walls allow absolutely no water to pass through. FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. The enlargement of the green leaf-surface has been already explained as a means of increasing transpiration, which is of special importance when the plants con- sidered grow in damp air. Similarly a diminution of the green surface signifies a restriction of transpiration. This relation is illustrated by the fact that in all floral areas, in which the activity of the vegetation is restricted or entirely stopped by increasing dryness, the foliage of the plants is not so widely outspread, i.e. it under- goes a diminution. It is also a well-known fact that one and the same species, if grown in a dry sunny position, will exhibit smaller, and in particular, narrower leaves than when it has been grown in a damp situation. This is well seen in passing from the mountainous districts bordering the Hungarian lowlands to the plains of the lower regions. A number of shrubs and herbs, Anchusa officinalis, Linum hirsutum, Alyssum montanum, Thymus Marschallianus, &c., exhibit on 326 FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. the dry sands of the plains much narrower leaves than in the valleys of the moun- tainous regions. In conjunction with the narrowing of the foliage, the wrinkling of the leaves has to be considered, i.e. the formation of grooved depressions on the sur- face. Strictly speaking, there is no lessening of the whole surface of the leaf, but only of that portion of the surface which is exposed to sun and wind. This is the point with which we are concerned. With regard to the exhalation of water, only the extent of the surface directly influenced by the agents for increasing transpira- tion is to be considered; whilst the extent of the grooved depressions, which are not exposed to the sun's rays, nor to dry currents of air, may be in a certain measure neglected. On the whole, plants with wrinkled and grooved foliage are not very abundant. For the most part the crumpling is to be seen on quite young leaves when first they break through the bud-scales, and when their epidermal cells are not yet sufficiently thickened with cuticular material. Later, when the formation of the cuticle is advanced, the wrinkles gradually become smooth, and the leaf becomes flat. It has already been pointed out that those pit-like depressions, on the floor of which stomata are concealed (cf. figs. 68 and 73), may also serve to restrict trans- piration. There is no contradiction in the statement that the same structure at one time hinders the entrance of water and the wetting of the stomata at the bottom of the pit, and at another time prevents direct contact with dry winds and consequent over-transpiration. Each has its turn. When the foliage of the Australian Prote- acese, during the summer sleep, is exposed for months to the scorching rays of the sun and to the warm dry air, and when all supplies of water from the soil have ceased, evaporation from the leaves must be restricted as much as possible; it is then that the pit-like depressions perform their duty in this respect. When, later on, the plants are aroused from their long sleep, and have to provide themselves with food, to grow, blossom, and fructify in an extremely short space of time, while violent showers of rain are pouring down from the clouded sky, and all the leaves are dripping with wet; it is then very important that, in spite of these exceedingly unfavourable conditions for evaporation, an abundant transpiration should never- theless take place, and that the function of the stomata should be in no way impaired by the moisture. These pit-like depressions, which in the dry period pre- vented evaporation, now have to keep moisture away from the stomata. In many plants evaporation from the superficial tissue is restricted by the close contact of the leaves to their supports, like the scales on the back of a fish. The upper side of a leaf in contact with the stem, and frequently adhering to it, is thus deprived of the means of exhalation, and transpiration can only take place on the somewhat arched or keeled under side of the green scale-like leaf. This occurs, for example, in the Tree of Life (Arbor vitce), in several species of Juniper, in T/iujopsis, Libocedrus, and various other Conifers. It is not without interest to notice that in several of these Conifers the little green scale-like leaves only become close pressed to the stem when they are exposed to the sun, whilst they project from it if the branches in question are shaded. FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. 327 A further reduction of the evaporating surface is brought about by the development of thickened or fleshy leaves. In order to render the points under consideration as clear as possible, it is perhaps well to insert here the following observations. By altering the form of a sheet of lead 8 cms. square and 1 mm. thick into a solid cylinder, the diameter of this cylinder is seen to be only 1 cm., and the whole surface of the cylinder is only one-fifth of that of the previous flat sheet. The application of these figures to the tissue of a leaf demonstrates how much smaller is the transpiring surface of a thick cylindrical leaf than of a thin flattened one. Such thickened leaves, which approach more or less to the cylindrical shape, are to be found regularly where transpiration has to be reduced for a considerable time — as, for example, in the mountainous districts of Central and Southern Europe, in the genus Sedum, growing on sandy soil which easily dries up, and on stone walls and battlements {Sedum album, rejlexum, dasyphyllum, atratum, Boloniense, Hispanicum,, &c.). They also occur in a striking manner in many tropical orchids which grow on rocks, or epiphytically on the bark of trees in the East Indies, Mexico, and Brazil, exposed for more than six months to great aridity {Brasavola cordata and tuherculata, Dendrobium junceum, Leptotes bicolor, Oncidium Cavendishianum and longifoliuin, Sarcanthus rostratus, Vanda teres, and many others); but especially are they found in aloes and stapelias and species of Cotyledon, Crassula, and MesembryanthemuTn, whose habitat is in the dryest districts of the Cape. Several Umbelliferse, Compositae, and Portulaceae (Inula crithmoides, Crithmum maritimum, Talinum fruticosum) growing on stony places of the sea-shore in the burning sun, and many salsolas of the deserts and salt steppes, as well as finally some Proteaceae, which for two-thirds of the year are exposed to the droughts of Australia — all are characterized by their development of fleshy leaves. Just as thick-leaved plants have acquired their succulence by a modification of their foliage, similarly, in the so-called cactiform plants, it is the stems which become thick and fleshy, and take on the functions of leaves. Here the green tissue is situated in the cortex of the stem, the epidermis covering it contains stomata just like the epidermis of foliage-leaves, and the green cortex transpires, and functions on the whole exactly as the green leaves do. When the stems of the cactiform plants are richly branched and the branches are short, they sometimes much resemble thick -leaved plants. Frequently also the separate portions of the stem and branches take the form of fleshy leaf-like discs, as in the genus of the Prickly-pear (Opuntia), and such stem-structures are usually mis- taken by the uninitiated for thick leaves. Gardeners, as a rule, group the thick- leaved and cactiform plants together under the single term "succulent plants". To the cactiform plants belong the opuntias and cacti, species of Cereiis, Echino- cactus, Melocactus, and Mammillaria, which are distributed from Chili and South Brazil over Peru, Columbia, the Antilles, and Guatemala. These are, however, especially developed on the high plains of Mexico in astonishing variety of form. To the cactiform plants belong also the leafless candelabra-like tree-shaped spurges 828 FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. oi' Africa and the East Indies. These plants are exposed, far more than the thick-leaved plants, for the greater portion of the year to extraordinary dryness. Their usual habitats are dry sandy and stony plains, waste rocky plateaus, and crevices of rocks which are almost completely wanting in soil. They always inhabit regions where no rain falls for about three-fourths of the year, and which usually belong to the driest parts of the earth. The whole organization of these plants corresponds to these conditions of their habitat. Dry scales and hairs are produced instead of foliage-leaves, or these are often metamorphosed into thorns which project in great numbers from the thick stem -structures, and efficiently protect them from the attacks of thirsty animals. The epidermis of the pillar-like, disc- shaped, or spherical stem-portions is thickened on the outer wall, so as to almost resemble cartilage, and frequently it forms a coat of mail round the deeper-lying green tissue by the abundant deposition of oxalate of lime (as much as 85 per cent). Most of the succulent plants, whose cell- walls, which are in contact with the air, are fortified by oxalate of lime, silicic acid, or suberin, have in their tissue peculiar aggregates of cells which apparently serve for the storing-up of water for the dry season, and which have been termed " aqueous tissue " . The water in these reser- voirs is always so apportioned that it lasts from one rainy season to another; that is to say, the adjoining green tissue which exhales the stored-up water does not suffer from drought during the dry season. Also, it is contrived in these plants that, immediately after the fall of the first rain, the reservoirs are again filled with water, and that the emptying and filling of the cells and the decrease and increase of their volume exercise no harmful influence on the adjoining tissue. Succulent plants have been not inaptly compared to camels, the "ships of the desert", which provide themselves with a large quantity of water, and are then able to dispense with further supplies for a long time without injury. The cells of the aqueous tissue are comparatively large and their walls thin; the active protoplasm within forms a delicate layer round the walls — that is to say, a sac whose cavity is filled with watery, often somewhat mucilaginous, fluid. In the cactuses the aqueous tissue is hidden as much as possible in the interior of the thick rod-shaped or spherical stem; also in many thick-leaved plants, such as some of the European species of the genus Sedum (e.g. Sedum album, dasyjjhyllum, glaucum); in South African species of the genera Aloe and Mesemhryanthemum {e.g. Mesemhry- anthetnuTn blandum, foliosum, sublacerum), the aqueous tissue is concealed in the interior of the leaf, and is usually composed of cells surrounding vascular bundles there situated. In Sedum Telephium, known by the name of Orpine, as well as in species of House-leek (Semjjervivum), and many salsolas growing on steppes, the ramifications of the vascular bundles are enveloped in a mantle of green tissue, and the bundles, which are, as it were, overlaid with green cells, are so arranged with regard to the colourless aqueous tissue, that to the naked eye they look like green strands in a transparent matrix which is as clear as water. In the Mexican Echeverias the aqueous tissue is inserted as broad stripes in the green tissue, and in the thick-leaved orchids it appears as if sprinkled between the green FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. 329 cells. The epidermis in numerous other thick-leaved plants serves as a store-house for water in a marvellous way. Individual epidermal cells are then greatly enlarged and project beyond the others in the form of sacs, clubs, or bladders, as shown in the picture of Rochea (fig. 79). These bladders either fit together into a one-layered extended coat of armour, or they are frequently placed irregularly side by side or above one another. In some instances they form isolated groups or occur singly, and appear then to the naked eye like protuberances on the green stems and leaves, where they glitter and sparkle in the sunshine like an embroidery of dew- drops. Many leaves and branches — as, for example, those of the widely-distributed Ice-plant (MesembryanthemuTn cristallinum) — have the greatest resemblance to candied fruit covered with clear, colourless, sparkling sugar crystals. When the walls of the enormously-distended vesicular or bladder-shaped cells of the epidermis are silicified, as are those of the repeatedly-mentioned Rochea, it is easily understood that the watery cell-sap which they contain is not exhaled into the air; the fluid is, so to speak, inclosed in a glass bottle and can only be given off in the direction of the green tissue. But when the walls of the bladder-like giant cells are not silicified, and not even particularly thickened, what is the result ? From the aspect of the Ice-plant one would think that a single warm dry day would suffice to shrivel and dry up the watery vesicles. But this is certainly not the case. Leafy twigs cut from the Ice-plant may be left all day on the dry ground in dry air and sunshine, and the large bladder-like cells on the surface will not lose their aqueous contents. After a week they become collapsed, having given up their water, not to the atmosphere, but to the green tissue covered by this swollen coat. Without doubt this phenomenon is to be associated with a peculiar formation of the cell- wall; but it is as certain that the constituents of the cell-sap, which fills the vesicles are also important, and it must be assumed that substances are dissolved in this aqueous ffuid which restrict the evaporation of the water. These substances, which hold water with great energy, and thereby enable the plants in question to survive through periods of the greatest dryness, are partly viscous, gummy, and resinous fluids, partly salts. It is well known that the sticky, watery pulp of crushed mistletoe berries, used in the manufacture of " bird-lime ", may be exposed to the air for months without quite drying up, and the mucilaginous juices of many cactuses and thick-leaved plants behave in a similar manner, espe- cially those of the Cape aloes, which exhale no water, and enable the plants possessing them to withstand the drought for months. In the thick-leaved plants of the salt steppes and deserts, the fluids are rarely resinous or gummy, but they frequently contain a surprising quantity of salts dissolved in water, such as common salt, chloride of magnesium, and the like; and these also obstinately retain water in proportionately large quantities. It is one of the most surprising of phenomena to see the thick-leaved salsolas rising above the soil of salt steppes, green and succu- lent, when the ground is at its driest in the height of summer, when for months no clouds ha/e tempered the sun's rays and not a drop of rain has fallen, and when almost all other plants have long ago turned yellow and faded. The large amount 330 FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. of salts contained in the sap of these plants renders them capable of a resistance which is almost greater than that afforded by mucilaginous materials and gum-resins. It must, however, be remarked here that not all green leaf- or stem-cells contain- ing abundant water have the function of storing it up for a drj^ season, and that the 'aqueous cell-groups and strands adjoining the green tissue, especially the so-called outer aqueous tissue, in very many cases, has another important function, viz. the conducting of carbonic acid to places where it can be assimilated, but this will be described in the next chapter. An extreme reduction of the leaf-surface, combined with a formation of green transpiring tissue in the cortex of the stem, is also shown in another group of plants known by the name of " Switch " plants. They are characterized by thin rod- shaped stems and branches, while the cactiform plants, on the contrary, always have their axes but little branched, and massive, thickened, fleshy and rigid stem-struc- tures which are unaffected by the wind. The switch-plants may be subdivided into those which are flexible, hollow, and only slightly branched — as, for example, the horse-tails (Equisetum), reeds (Scirpus), rushes (Juncus), bog-rushes {Schvenus), and several cy peruses (Cyperus); and into broom-like shrubs with rigid woody boughs breaking up into innumerable branches and twigs. The former are distributed over the whole world ; the latter are principally to be found in Australia and in districts bordering on the Mediterranean Sea. In Australia it is chiefly Casuarinas and some genera of Papilionacese and Santalacese {SphcGrolobium, Vi'niinaria, Lepto- meria, Exocarpus) which take on this odd form, and some of them even attain to the size of trees. In the Mediterranean flora isolated species and groups from the families of Asparaginese, Polygalacese, and Resedaceae are seen with thin, stiff, rod- shaped, leafless branches, which project stifily into the air with green cortex; but again, most of these plants belong to the Papilionaceae and Santalacese. Several switch-plants of the papilionaceous genera Retaraa, Genista, Cytisus, and Spartiuvi, growing together, often cover wide tracts of country in densely-crowded masses, and thus contribute not a little to the scenic peculiarity of the district. Many small rocky islands off the coast of Istria are entirely overgrown by Spartium scoparium, which is represented in the illustration opposite. In May large golden flowers, scented like acacias, appear on the green rods of the Broom, and then for a short time the dark green of the switch-plant is changed into a brilliant yellow. On passing near the coast, just at this time, the remarkable phenomenon is seen of golden yellow islands rising above the dark blue sea. This floral adornment is, however, but transitory, and nothing more monotonous and desolate than such a dry unwatered islet, covered with these shrubs, can be imagined The SpartiuTn belongs to those switch-plants which are not entirely leafless, but which develop little green lancet-shaped leaves at intervals on their long twigs. But these are of such secondary importance that their green tissue can only form the smallest portion of the organic substances necessary to the further growth of the plants, and this duty chiefly falls to the share of the cortex of the switch-like branches. The cortex is also characteristically formed in accordance with this fact. FOllM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. 331 Under the epidermis, whose outer walls are much thickened and coated with wax, is the green transpiring tissue or "chlorenchyma", consisting of from five to seven rows of cells. This green tissue does not form a continuous mantle round the stem, but is divided into from ten to fifteen thick strands by strips of hard bast (see ficr. 81). Below this cortex of alternating green tissue and strips of bast are soft bast, cambium, wood, and a very large pith; but these have no further interest for us here. It is, however, worthy of remark that in the green strands of the cortex lij so — suitili I Imts Bushes of Spartium scoparium near Rovigno in Istvia. of the Spartium, the crowded green chlorophyll-containing cells of the chlorenchyma closely adjoin one another, and that only very narrow air-passages ramify between them, so that here there is no formation of a spongy parenchyma penetrated by wide canals and passages. On the other hand, large cavities occur where the green tissue touches the epidermis, and these act as substitutes for the wide ramifying canals. Over each of the cavities a stoma is to be seen in the epidermis tlirough which the water vapour, exhaled chiefly from the green cells, can escape (see fig. 81 2). The stomata are proportionately small, but their number is very great. Since the guard-cells are not so strongly thickened on their outer walls as are the other epidermal cells, the stomata appear to be somewhat sunken. By this arrange- ment, and also by the epidermal coating of wax, they are protected from moisture. 332 FORM AND POSITIOX OF THE TRANSPIRING LEAVES AND BRANCHES. In the Casuarineos and in Cytisus radiatus (see fig. 69), tlie green tissue is distri- buted in the cortex of the branches exactly as in the case just described; but the strips of green tissue traversing the stem are deeply cut into by longitudinal furrows. In some other leafless switch-shrubs, such as species of the genus Ephedra, the chlorenchyma forms a continuous and uniform mantle round the stem, uninterrupted by strips of bast. But in this case the stomata are distributed uniformly over the whole surface of the rod-shaped branches, while in the brooms, Casuarinese, and in Cytisus radiatus they are absent from those portions of the epidermis which cover the strips of hard bast. Plants with leaf-like branches or cladodes are distinguished from switch-plants .^^ T 11 111 1 J ' ,','A■c,_-,?~ '• "IJ,' ;j''i'i,LM> ''t^' :i.«MiJiiw,iiiiiw , i U'- '- JU ^»j^ ^^^Tc -^,;^^-^, ~ ^^ Fig. 81.— Switch-shrubs. 1 Part of stem of Spariiumsco^armrn cut transversely; x30. 2 Part of the transverse section ; x240. by the fact that all their shoots are not circular in section, but some are flattened, looking as though they had been pressed out. When this flattening is restricted to the so-called "short branches ", i.e. when on a stem only the ultimate, comparatively short branches are flattened, the main axes remaining cylindrical, like ordinary stalks, these structures have quite the appearance of leaves which are sessile on the rounded stems. This explanation of them, however, given by botanists, is not at first sight satisfactory to the uninitiated. Why should these flat green structures be branches, and not leaves? The illustration opposite at once makes the matter clear. It represents two cladode-bearing plants, viz. two species of Butcher's-broom {Ruscus Hypoglossum and aculeatus), each at an early stage of development and also when fully grown. On the young shoots, which have just made their way out of the soil (see figs. 82 ^ and 82^), the true leaves can be seen in the shape of small sessile pale scales on the long, rounded, finely-ridged axis; and from the angles which these scales make with the long axis arise darker, much thicker organs which rapidly increase in size, while the supporting covering-scales become dry, shrivel up, and finally FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. 333 disappear, leaving no traces. Since the members which arise from the axils of leaves (whether these are small clothing-scales, or large green laminoe does not matter) are not considered to be leaves, but shoots, the flat leaf-like structures of the Butcher's-broom are also regarded as shoots, and are named "flattened shoots" (cladodes) — or, considering their similarity to leaves, " leaf -branches " (phylloclades). Fig. 82.— Plants with leaf-like Branches (Cladodes). » Ynuiifi shoot of 7?w«!W Bypoglossum. 2 The same branch fully grown, with flowers on the nladod«B « Youne pboot of Rmmi. aeuleatvj, * The pam6 braach with flowers ou tho cladodes. This view is strengthened materially by the fact that these leaf-like structures, in their further development, and in the production of shoots, behave exactly like ordinary cylindrical axes. That is to say, small scale-like leaves spring from them, and from the axils of these scales arise stalked flowers (see figs. 82 - and 82 *) which ultimately fructify. Plants possessing such phylloclades are not very numerous on 334 FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. the whole. The Butcher's-brooms, chosen above as examples, belong to Southern Europe, and occur in large quantities on the soil of dry woods, where everything is wrapped in deep sleep during the height of summer. In the Antilles, and in the prairies of the East Indies, are about twenty shrub -like species, belonging to the genus Phyllanthus of the Spurge-family. New Zealand also possesses one of these peculiar phyllocladous plants, belonging to the papilionaceous genus Car- michelia. In the species of both these genera (see fig. 83) the flattened shoots are exceedingly like lancet-shaped foliage-leaves, and the true leaves are transformed into small pale scales. These tiny scales are situated on the margins of the phylloclades, and from their axils arise stalks bearing the flowers and fruit. On the Andes of South America occur the remarkable colletias, of which a species, Colletia cruciata, is represented in fig. 83. The leaflets on these extraordinary shrubs are diminutive, but not pale and scale-like; whilst the green phylloclades, which play the part of the foliage-leaves, form very strong flattened organs, tapering to a point, and placed opposite one another in pairs, so that each pair is always at right angles to the couple next above or below. Yet another arrangement is seen in Coccoloba platyclada (Polygonaceae), a native of the Salomon Islands, and in Coceulus Balfou7'ii, growing in the island of Socotra. But it is impossible here to enter into all these variations in detail ; it is enough to have brought forward the most striking forms of phyllocladous plants which are represented in figs. 82 and 83. If in all these peculiar plants the branches are flattened and spread out, it cannot indeed be asserted that the surface of their transpiring tissue has undergone diminution, and thus far of course this strange development has nothing to do with the restriction of transpiration. The arrangement by which this is brought about must be sought for elsewhere. It consists in this: the leaf-like shoots are so directed that their surfaces are vertical and not horizontal. Contrary to most flat leaves, which turn their broad surfaces fully to the incident light, the flattened shoots are placed vertically so that at mid-day they only cast a very narrow shadow, and do not stop the sunbeams on their way to the soil. It is obvious, however, that such a leaf-like structure placed vertically, as it were on edge, will exhale much less than a foliage-leaf whose surface is opposed to the mid-day sunbeams. The work carried on in the green cells, under the influence of light, is not hindered by this position of the leaf-like organs. If the vertical green surfaces are not so well illuminated by the sun's rays during the warmest part of the day, this is abundantly compensated for by the fact that their broad surfaces are exposed to the light both of the morning and evening sun. On the other hand, when the sun rises and sets, the heat is not so powerful, and consequently there is no such rapid exhalation to be feared as when the sun is in the zenith. To put the matter shortly, transpiration alone — not illumination — is restricted by the vertical position of the green laminae, and therefore this metamorphosis has rightly been considered a protective measure against excessive transpiration. This arrangement is only found in plants of dry regions, where transpiration FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. 335 requires no assistance, but where, on the contrary, the danger is often imminent that water cannot be drawn from the soil in sufficient quantity to replace that lost by exhalation. The phylloclades, moreover, are only a type of a large number of organs which, in a word, all agree in this ; the edge or narrow side of the flattened exhaling structure, not the broad surface, is turned towards the zenith. In many of the Fig. 83.— Plants with Leaf-like Branches (Cladodes). 1 Colletia crueiata. * Carmichelia atcstralis. » Phyllanthus speciosui. vetches of the Southern European flora (Lathyrus Nissolia, Ochrus), but especially in a large number of Australian shrubs and trees, principally acacias (Acacia longi- folia, falcata, myrtifolia, armata, cultrata, Melanoxylon, decipiens, &c.), it is the leaf-stalks which are extended like leaves placed vertically, and then the develop- ment of the leaf-lamina is either entirely arrested, or has the appearance of an appen- dage at the apex of the flat green leaf-stalk, or " phyllode ", as it is called. In many Myrtacese and Proteaceae, especially in species of the genera Eucalyptus, Leucaden- 336 FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. dron, Melaleuca, Protea, Banksia, and Grevillea, the leaf -blades themselves are not placed horizontally like those of our maples, elms, beeches, and oaks, but vertically on edge, like the phylloclades and phyllodes. Imagine now an entire wood of such eucalypti and acacias, on which the mid-day sun is pouring down its rays. If it is not exactly literally true to say that each vertical leaf only casts a linear shadow at noon, it is at least certain that there is not much shade on the ground of such a wood. The sunbeams find their way everywhere between the erect leaf-blades, penetrating the depths below, and it is impossible to speak of the dim forest-light under such circumstances. The Casuarinese, which grow with eucalyptus, acacias, and Proteaceae do not help to make such woods shady, and thus one is quite justified in speaking of the shadowless forests of Australia. Although Australia stands alone in the variety and abundance of its plants possessing vertical leaf -blades, other floral areas furnish numerous and remarkable examples of this arrangement. One has only to think of the curious shape of the so-called "equitant" leaves belonging to many plants of the Lily family (Tojieldia, Nartheciuni), numerous Iride83, and the closely-related genera. Gladiolus, Ferraria, Witsenia, Montbretia, &c., chiefly natives of the Cape. The leaves exhibit the peculiarity of being folded together lengthwise, and the sides thus brought into contact become fused with one another. Only at the point where they join the stem do the two halves remain distinct, forming a groove in which is inserted the base of an upper leaf. The formation of such equitant leaves from ordinary leaf -blades may perhaps be illustrated by taking a strip of paper smeared on one side with paste and folding it longitudinally so that the pasted sides are in contact and become joined together. Such equitant leaves are so directed that their broad surfaces are much less exposed to the perpendicular rays of the mid-day than to those of the rising and setting sun. In the Mediterranean flora, and on many steppes, plants are not seldom to be met with whose leaves look as if they had not been able to free themselves from the stem. In such plants the projecting portion of the foliage-leaf is very small, but the margins are continued for some way down the stem as projecting strips and wings. Leaves of this kind are termed "decurrent". They are particularly abundant amongst Composites, viz. in the genera Centaurea, Inula, Helichrysum; but they also occur in many Papilionaceous plants and Labiates. The position of these vertical wings, which traverse the stem, is exactly the same, with regard to the sun, as that of the phyllodes, phylloclades, and equitant leaves, and they behave in respect to transpiration in exactly the same way. In many plants the blades of the foliage-leaves when young have not a vertical position^ but gradually assume it during development, i.e the blades at first arc turned so that the flattened surfaces are horizontal and face upwards and downwards. Later they twist round at the point where they are inserted on the stem, so that their margins become directed upwards and downwards. As already stated, this peculiarity is observed in many eucalypti and various other Australian trees and shrubs. But plants in sunny situations in other regions also exhibit this FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. 337 peculiarity. In the Spanish flora, for example, is an Umbellifer (Bupleurum verticale) whose leaves are so twisted with regard to the sun that they remind one forcibly of the Australian acacias. Many Composites, especially the widely-distri- buted Wild Lettuce (Lactuca Scariola), growing on dry soil in Central Europe, Fig. 84.— Compass Plants. ^Silphiwn laciniatum, seen from the east. « The same plant seen from the south. « Lactuca Scariola, seen from the east. * The same plant from the south. Both species are considerably reduced. exhibit this contrivance in a striking manner. A Composite shrub, Silj^hiuin laciniatum, to be found in the prairies of North America, from Michigan and Wisconsin as far south as Alabama and Texas, has obtained a certain renown by reason of the remarkable twisting of its leaf-blades. It long astonished hunters in the prairies that in these plants (represented in fig. 84) the leaf-laminae, especially those springing from the lowest portions of the stem, not only assumed a vertical Vol. :. 22 338 FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHEIS. position, but that the broad surfaces of each leaf always faced the rising and setting sun. Healthy living plants as they grow in the sunny meadows look as though they had been laid between two gigantic sheets of paper, somewhat pressed, and dried for some time in the way plants are prepared for herbariums, and had then been removed from the press and set up so that the apex and profile of the vertical leaf-blades point north and south, i.e. in the meridian; while their surfaces face the east and west. This inclination is so well and regularly observed by the living plants on the prairies, that hunters are enabled to guide themselves over such regions, even under a clouded sky, by means of these plants; for this reason Sil- phium laciniatum has been called a " compass " plant. The life of the compass plant is assisted by this placing of the vertical leaves in the meridian, in that the broad surfaces are placed almost at right angles to the incident sunbeams which illuminate them in the cool and relatively damp morning and evening, while at the same time they are not too strongly heated nor stimulated to excessive transpiration. At mid-day, on the other hand, when the sun's rays only fall on the profile of the leaves, the heating and transpiration are proportionately slight. It is of interest that the leaves of these compass plants, as well as those of the above-mentioned Lettuce represented with the compass plant in fig. 84, show this inclination and position when they grow on level, moderately dry, unshaded ground, and that in damp shady places, where there is no danger of over-transpiration from the powerful rays of the noon-tide sun, the twisting of the leaves does not take place, and they are not brought into the meridian. The placing of their leaf-blades parallel to the ground when in the shade, but vertically when in dry sunny places, is, generally speaking, a phenomenon which may be seen in very many plants, including shrubs and trees. A species of lime, a native of Southern Europe, viz. the Silver Lime (Tilia argentea), is particularly noticeable in this respect. On dry hot summer days the leaves assume an almost vertical position, but only on those boughs and twigs which are exposed to the sun. If the tree stands at the foot of a wall of rock, or on the edge of a thick wood, so that a portion of it is shaded, the leaves on this shaded part remain extended horizontally. Such a tree then presents a strange aspect, as the leaves are of two colours — dark green on the upper side, and white on the under surface by reason of a fine felt-work of white stellate hairs — and it is scarcely credible at first sight that the shaded and sunny portions of the tree belong to one another. In the compass plants and also in the Silver Lime the alterations in the direction of the leaves are brought about by alterations in the turgidity of certain groups of cells in the leaf -stalk. It is exactly the same cause which produces the periodic movements of numberless plants with pinnate or palmate leaves, and the leaf -folding of many grasses; and it is natural to conjecture that these phenomena of movement are also connected with transpiration. This is in part actually the case. In consequence of alterations in turgidity of the pulvini, the pinnate leaflets of the Gleditschias and some Mimosas rise up after sunset, while those of the Amor- phas fall down, and assume a vertical position during the night; but this is con- FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. 339 nected with the nocturnal radiation of heat (as will be explained later) and not with exhalation. It is, however, equally certain that the placing together and folding up of leaves and leaflets in many other plants is brought about in order to prevent over-transpiration and consequent withering up. Many shrubby, thorny mimosas of Brazil and Mexico, when in their native habitat and position, extend their leaflets horizontally when evening approaches, contrary to the behaviour of the well-known Sensitive Plant (Mimosa pudica), and they remain in this position throughout the night. Next morning they are still widely outspread. As soon as the sun has risen, and its beams fall on the foliage, the leaflets shut together; the menacing thorns, which until now have been hidden by the extended leaves, become apparent; and the leaflets remain in the vertical position during the hottest and driest hours of the day. Towards sunset they again rise and are extended horizontally. There is but one exception to this cycle of changes — if the opened leaf is shaken by the wind, and if the sky has been gray and clouded all day. In the former case, under the influence of the wind, a rapid closure occurs; in the latter case, when the weather is bad, they remain open all day. One of the Rutacese, Porliera hygrometrica, behaves like these mimosas. In Peru, the native country of these plants, where they abound, the opening and closing of the leaves has even been made use of for weather predictions, for when the vertical leaves are closed, dry hot weather can be reckoned upon; when they are open, damp cool weather. In the cultivated Bean (Phaseolus), moreover, alterations of position in parts of the leaflets may be observed to take place during the day. When the sun is powerful, the leaflets assume a vertical position, so that at noon the sun's rays only reach a small portion of the blade. In several species of Wood-sorrel belonging to the South African flora, and also in the widely-distributed Common Wood-sorrel (Oxalis Acetosella), it may be noticed that the leaflets, as soon as they are directly struck by the sun's rays, sink down, so that their under surfaces — on which the stomata are situated — face one another, the three leaflets together forming a pyramid; while these same leaflets in damp shady places remain extended. The leaflets of the water fern, Marsilea quadrifolia, which grows in marshes and is distributed through Central and Southern Europe, temperate Asia, and North America, are very similar to those of the Wood-sorrel, but carry their stomata on the upper surface. As long as they remain floating on the surface of water, these leaflets are extended, but as soon as the water-level sinks and the leaflets become surrounded by air, they fold together above in the sunshine, and their position becomes vertical, precisely as in the compass plants. As another phenomenon of this kind the periodic folding or closing of the leaves of grasses must be specially mentioned. It has long been noticed that certain grasses exhibit a very different aspect according as they are observed on a dewy morning or in the noon-day sunshine. In the morning their long linear leaves are fluted on the upper surface, or spread out quite flat. As soon as the humidity of the air diminishes, in consequence of the higher position of the sun, they fold 340 FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. together lengthwise; again after sunset they widen and become flat or fluted. This process may be repeated twice on a summer's day within twenty-four hours, if a storm intervenes at mid-day and is followed by a sunny afternoon. How much this depends upon the conditions of humidity of the air, is demonstrated by the fact that such grasses, when gi-own in pots, can be easily made to open and close their leaves by alternately sprinkling them with water and placing in damp air, and then for a short time exposing them to dry air. The leaf -folding in various species of Sesleria is exceedingly quick and also very interesting. The species of this genus grow principally on the Alps, Carpathians, and Balkans. They always grow together and often cover wide stretches of hilly and elevated districts with thick grassy turf. One species (Sesleria coerulea) is distributed over Northern Europe in Finland, Sweden, and England. The closing of the leaves of these moor- grasses reminds one strongly of the Venus Fly-trap (Dioncea muscipula), which has already been fully described. It is indeed an actual shutting together of the two halves of the leaf. As in the leaf of the "Fly-trap", the midrib of the leaf of the Sesleria remains in its original position unaltered; also the two halves of the leaf do not come flatly in contact, but rise up obliquely so as to leave between them a deep, narrow, groove-like cavity, widest at its lowest part (see fig. 85 ^). While the open leaf turns its upper surface, rich in stomata, towards the sky, the two raised halves of the folded leaf are parallel with the incident sunbeams, and the folded leaf of the moor-grass may then be compared to the equitant leaf of an iris. In the cavity produced by the closing up of the leaf are the stomata, however, and thus the green tissue next them is excellently protected from the sun's rays as well as from the direct action of the wind. The epidermis of the lower surface, which is exposed on the folded leaf to all the agencies which excite transpiration, possesses no stomata, but is provided with a thick cuticle. A leaf -folding similar to that of Sesleria, along the midrib, has been observed in the leaves of Avena planiculmis, which grows in sunny fields on the Sudetics and Carpathians. It also occurs in Avena compressa, and many others related to these species. The folding or closing of the leaves in the large section of fescue-grasses (Festuca) is carried on somewhat diflferently. In Sesleria, the opened upper sur- face of the leaf forms only a single shallow groove, and the folding only occurs at the midrib; but on the upper side of the fescue-grass leaf several parallel grooves are to be seen, and the green tissue is divided up by these grooves into several pro- jecting ridges, exhibiting a very remarkable structure. In each ridge can be dis- tinguished the base which forms a part of the under side of the whole leaf; then the apex opposite the base, belonging to the upper surface of the entire leaf; and finally, the two side portions forming the sloping sides of the grooves which run between the ridges (see figs. 87 and 88). The greater part of each ridge consists of green tissue. The stomata on the ridge only open on the sloping sides facing the grooves. Neither the crests of the ridges nor the lower surface of the leaf exhibit a single stomate. The apex is without chlorophyll, and almost always has a border of elongated cells with strong elastic FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. 341 walls under the epidermis; the same thing occurs on the under side of the leaf (^.e. at the base of the ridges), which is formed of one or several layers of cells without chlorophyll, but furnished with thickened walls. The closing of the leaf is not so simple here as in the Seslerias. There the leaf-folding only produced a single deep channel, widened at its base; in the fescue-grasses all the small grooves between the ridges become narrowed by the closing, i.e. by the upward inclination of the right and left halves of the leaf, those adjoining the central ridge to the greatest Fig. 86.— Folding of Grass-leaves. 1 Vertical section through an open leaf of the thin-leaved Moor-grass (Sesleria tenuifolia). « Vertical section through a closed leaf ; x40. » Portion from the centre of an open leaf; x300. extent, those in the neighbourhood of the approximated margins in a lesser degree (see fig. 88^). Since the stomata lie on the sides of the ridges, it is obvious that transpiration is checked to the utmost by the closing and consequent approximation of the opposite sides of each groove. In individual cases among various fescue-grasses are to be found manifold differences in the number and shape of the ridges, also with respect to the formation of the under surface of the leaf, and most of all in the form assumed by the leaf in its expanded condition. There is a large group of festucas which are said to be poisonous by the shepherds in the mountain regions of Spain, and in the Alps, the Taurus, and the Elbruz, These will be spoken of again later. When open in damp weather they form only a moderately narrow main furrow, with several 842 FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. narrow secondary grooves leading from it, as can be seen in a vertical section of an open leaf of Festuca alpestris, a plant very abundant in the Southern Alps (see fig. 86^). In Festuca alpestris, the blunt apex of each ridge has a border, three layers deep, of cells destitute of chlorophyll, and the lower side of the leaf is pro- vided with an actual armour of thick-walled bast cells, covered by an epidermis. ^sio-S?=i?^;^:o'^ \ / \u.. \^uy . i\v:r- ^ Fig. 86.— Folding of Grass-leaves. 1 Vertical section through part of the open leaf of Stipa capillata; x240. 2 Vertical section through an entire open leaf. 8 Vertical section through a closed leaf; x30. * Vertical section through a portion of the leaf of Festuca alpestris; x210. 5 Vertical section through an entire open leaf. * Vertical section through a closed leaf; x30. whose outer walls are much thickened. A vertical section through the leaf of Festuca punctoria, a native of the Taurus, is represented in fig. 88. In this plant, the leaves, when open, present a fairly shallow depression; the under surface is clothed with a protective mantle of five layers of strong cells devoid of chloro- phyll; the ridges are rounded off and possess only a single layer of covering cells, provided with an extremely strong wax-like coat. The open leaves of Festuca Porcii, FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. 343 a native of the Carpathians, are relatively thin (see figs. 87 * and 87 ^). Below the epidermis of the under side is no mantle of bast cells as in the species already described, but only isolated strands of bast; however, the crest of each ridge is furnished with a strand of bast cells; the ridges themselves project very much, and the whole leaf is traversed by six deep narrow grooves. In the three fescue-grasses cited here as examples, and in all species of the genus Festuca, forming the main part of the turf of our fields, a vascular bundle Fig. 87.— Folding of Grass-leaves. Vertical section through a closed leaf of Lasiagrostis Calamagrostis. ^ Vertical section through an open leaf ; x 24. * Vertical section through a portion of the open leaf; x210. * Vertical section through a closed leaf of Festuca Porcii. 5 Vertical section through an open leaf; x24. « Vertical section through a portion of the open leaf; x210. surrounded by green tissue traverses each ridge. In the hinged leaves of many other grasses, the green tissue of each ridge is divided into two portions. The vascular bundle is bordered above and below by strands of thick-walled cells devoid of chlorophyll, and thus arises a strong septum in the green parenchyma, beautifully shown in the transverse section of a leaf of Lasiagrostis Calamagrostis, illustrated in fig. 87. In the leaves of the Feather-grass (Stipa capillata) are alternating higher and lower ridges; a vertical section is shown in fig. 86 ^-2'^. In the higher ridges occur septa similar to those in Lasiagrostis, but in the lower there is only a vas- cular bundle surrounded by green tissue as in the fescue-grasses. No less than 344 FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. twenty-nine ridges can be counted on the leaf of the above-mentioned Lasiagrostis, a plant widely distributed in the valleys of the Western and Southern Alps, where it clothes the sunny slopes in thick masses. When the leaf folds up, the twenty- eight grooves between the ridges, on whose sides are the stomata, become narrowed, and the entire leaf assumes a tubular form, so that transpiration is almost com- pletely suspended. In Stipa capillata, which is very abundant on clay steppes, the same thing occurs (see fig. 86 ^ ). In both grasses the closure of the grooves on whose sides are the stomata, is completed by short stiff hairs on the summit of the ridges, which interlock when the ridges approach one another, and so block up access to the grooves (fig. 86 ^). It would take us much too far to describe the numerous other modifications which are to be met with in the structure of hinged grass -leaves. The examples given suffice to make it evident that the danger of over-transpiration is avoided by the folding of the leaf, and that amongst the grasses very many arrangements obtain in order, sometimes, to expose those green parts of the leaf whose epidermis is supplied with stomata to the rays of the sun, and at other times to withdraw them, according to the humidity of the soil and of the surrounding air, thus suitably regulating transpiration to the existing circumstances. The mechanism by which grass-leaves open and close may be explained in two ways — either the process is due to hygroscopic changes, as in the opening and clos- ing of the " Rose of Jericho ", or to alterations in the turgidity of certain groups of cells, as in the mimosas. If the former alone were the case, a dry, dead grass-leaf should be still capable of opening and closing in accordance with its damp or dry condition; but a leaf of any of these when cut off and dried no longer opens, even after being moistened for a considerable time, and therefore the first explanation cannot be accepted, at any rate for most of the grasses. Apparently, the mechanism consists of alterations in the turgescence of those groups of cells situated in the angle of the grooves. Since the floor of the grooves was frequently found to con- sist of peculiar thin-walled cells destitute of chlorophyll, and filled with colourless watery sap, it was concluded that the opening and closing of the grass-leaves was due to the change in turgidity of these cells. However, this was going too far. These cells in most instances, for example, in Festuca punctoria (see fig. 88 -), would be much too delicate to effect, unaided, the closure of the leaf by their loss of turgidity, or to open it by their increasing turgescence. In many grasses these cells are completely wanting {e.g. in Festuca alpestris and Stipa capillata, fig. 86). Moreover, it is observed that the opening and closing of the leaf is still carried on when the thin-walled cells at the bottom of the grooves are destroyed, artificially, by puncturing with fine needles. The cause of the movement must therefore be looked for in the alteration of turgescence of other cells below the grooves. When a mantle of several layers of thick-walled cells is present on the under side of the leaf, their walls are seen to swell up simultaneously with the alterations of tur- gescence of the parenchymatous cells. Of course the inner cell-layers of the mantle must be capable of swelling up to a greater extent than the outer, and this has FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. 345 actually been shown to be the case in some species. Moreover, although the thin- walled cells at the bottom of the furrows are not consiaered strong enough to bring about the opening and closing by changes in their turgidity alone, it is by no means asserted that they have no other part to play. When they are constructed as in the leaves of the moor-grasses and in the fescue-grass of the Taurus (Festuca punctoria, figs. 85 and 88), they certainly are not without a purpose. Their advan- tage to the plant lies in the fact that they can be much compressed without harm by the closure of the leaf, whereby the neighbouring parenchymatous cells are pro- ■^^^ Fig. 88.— Folding of Grass-leaves. ^ Vertical section through an open leaf of Festuca punctoria, of the Taurus. * Vertical section through a closed leaf ; x 40. 8 Vertical section through a portion of the open leaf ; x280. tected from injury; also that by means of these cells, which are filled with watery «ap, carbonic acid from the atmosphere is conducted to the underlying green tissue; and lastly, that in case of necessity, water can be absorbed from the air. They re- mind one strongly of the thin-walled groups of cells of foliage-leaves used for the direct absorption of moisture, and possibly they can function in this way. If, in places where these grasses grow naturally, a slight shower of rain falls after a long period of drought, or if dew falls during clear nights, little or none of the water reaches the roots, since it is retained by leaves overspreading the soil. But the water easily runs into the furrows of the folded leaves of grass, and since the large thin-walled cells at the bottom of the grooves can be wetted, they offer to the water which can pass through them the shortest path to the green cells in the interior of the leaf. 346 FORM AND POSITION OF THE TRANSPIRING LEAVES AND BRANCHES. A process, very similar to the opening and closing of grass-leaves, is also to be observed in the true mosses, in all species of the genus PolytrichuTn, and in some of the Barbulas. The peculiar structure of the leaves of these mosses has been already i treated of. In addition to the description there given, it may be mentioned that the ridges of thin-walled green cells, which are present on the upper surface of such a i i leaf (see fig. 89), only remain exposed to currents of air as long as this air possesses the requisite degree of humidity; that is to say, the blade of the leaf from whose upper surface the bands project only remains expanded while that is the eg (fig. 892). As soon as the air becomes dry, the lateral portions of the leaf-blade bend upwards, and envelop the green ridges like a mantle (fig. 89^). These are then & i^fs) -.^ ^^r^(^]'c^