K^y- ® OECOLOGY OF PLANTS AN INTRODUCTION TO THE STUDY OF PLANT-COMMUNITIES By BUG. WARMING, Ph.D. PROFESSOR OF BOTANY IN THE UNIVERSITY OF COPENHAGEN ASSISTED BY MARTIN VAHL, Ph.D. PRIVATDOCENT IN THE UNIVERSITY OF COPENHAGEN PREPARED FOR PUBLICATION IN ENGLISH BY PERCY GROOM, M.A., D.Sc, F.L.S. ASSISTANT PROFESSOR OF BOTANY IN THE IMPERIAL COLLEGE OF SCIENCE AND TECHNOLOGY, LONDON AND ISAAC BAYLEY BALFOUR, M.A, M.D., F.R.S. king's BOTANIST IN SCOTLAND, REGIUS KEEPER OF THE ROYAL BOTANIC GARDEN AND PROFESSOR OF BOTANY IN THE UNIVERSITY, EDINBURGH OXFORD ^ AT THE CLARENDON PRESS 1909 qoi Vi5 Co P-O- HENRY FROWDE, M.A. PUBLISHER TO THE UNIVERSITY OF OXFORD LONDON, EDINBURGH, NEW YORK TORONTO AND MELBOURNE NOTE It is now some years since expectation became prevalent of an English edition of Professor Warming's book — Plantesamfund. Nothing need now be said about the difficulties opposing its pro- duction, because Professor Warming has solved them happily by writing for the Delegates of the Oxford Press this present book, founded upon his original Danish work. To the manuscript, as it has been prepared by and received from Professor Warming at intervals. Professor Groom has applied with untiring patience his skill in interpretation and in apt expression, and the book as it now appears is therefore not an English edition of a foreign book — as are others of the botanical series issued by the Dele- gates of the Oxford Press, but is ' practically a new work ', as the author himself designates it. The book is a valuable addition to botanical literature, and will appeal to a wide audience. Its subject, Oecology, is the held in which the botanical morphologist, physiologist, and systema- tist happily meet, and to them this statement of the views of a pioneer and leader in oecological work will be welcome. Those whose interests lie in the practical application of a knowledge of plant life in the several domains of rural economy — Agriculture, Horticulture, Forestry — will find in the matters treated in the book the clue to many of the problems which they meet with. Teachers within whose sphere it lies to encourage a study of Nature will find its pages full of information and suggestion to guide them. Students of Botany will glean from it sound instruc- tion in a subject which now occupies a prominent place in botanical teaching. Every one, indeed, for whom the varying aspects of vegetation have interest will obtain by perusal of the book new lights by which that interest may be increased. Perhaps in no way will the value of the book be greater than as iv NOTE a stimulus to accurate observation and inquiry, through which Oecology will be advanced^from the stage of 'infancy', in which, as Professor Warming says, it now is. The character of the book and the method of its production have necessarily placed a restriction upon editorial functions, and have not allowed of modifications that might have brought more directly home some of its teachings to readers in Britain — for instance, by the introduction of a greater number of illustra- tive references to vegetation in the British Isles. If, as we beUeve will be the case, a new edition of the book is called for, attention may be given in it to matters of this kind. I. B. B. AUTHOR'S PREFACE In 1895 I published a Danish work entitled Plantesainfund, which was based upon lectures that I had delivered in the University of Copenhagen. I never imagined that the book would appeal to more than a few readers outside my audience, and was therefore greatly surprised shortly after its publication to receive from Dr. E. Knoblauch a request for permission to prepare a German version of my book — an act of courtesy, since Denmark had not subscribed to the Bern Convention, and my book was thus public property. Thanks to Dr. Knob- lauch's energy the German edition was published in 1896. In the short time available I found it impossible to introduce more than trifling changes into his edition, and was forced to post- pone the more important modifications that I contemplated. In 1902 the publishers, Gebriider Borntrager of Berlin, issued a second edition of this German translation. It was edited by Dr. P. Graebner. With this edition I had nothing whatever to do. The book was unchanged as regards plan and arrangement of the subject-matter. I had always entertained grave doubts as to the arrange- ment of the contents of Plantesamfund. When I wrote it I had no models to study ; mine was the first attempt to write a work on Oecological Plant-geography, of which the very name was then all but new. The present book is practically a new one ; for, not only have I myself introduced a number of new features, but I have also invoked the aid of the young Phyto-geographer, Dr. M. Vahl, in order that he might deal critically with purely geographical and climatic considerations. The following changes are amongst the most important that appear in the book as it is now presented in English : — Chapter II contains fresh subject-matter dealing with growth-forms, as well as an entirely new classification of these. The parts of the book referring to adaptations of water-plants and land-plants have been combined to form Section III ; and in the same section I have given my views on oecological classification in a more comprehensive and detailed manner. a 3 vi PREFACE For these alterations I am mainly responsible, but the new classification of the formations is largely due to Dr. Vahl, who has thus materially remodelled parts of the book. In place of the four Sections discussing hydrophytic, xerophytic, halo- phytic, and mesophytic communities respectively, thirteen Sections (IV-XVI) have been devoted to thirteen oecological classes established on the basis of edaphic and climatic distinc- tions. The arrangement of the subject-matter dealing with the several Formations is new in many respects, the changes involved being due partly to myself (for instance, in connexion with halophytes and lithophytes) and partly to Dr. Vahl (notably in Sections XI-XV). So far as my other varied work, including administrative duties, would permit, I have endeavoured, with the assistance of Dr. Vahl, to take into consideration the vast amount of pertinent literature issued since 1895. Since that year there have been published, not only the large general works by Schimper (1898, English Edition 1903), Solms-Lauhach (1905), and Clements (1904, 1905, 1907), which contain much that is original and suggestive, but also an immense number of original papers in various periodicals and countries. So far as possible, recognition has been made of all important contributions issued up to the present moment, and their titles will be found in the appended Bibliography. But there is considerable difficulty in selecting the most important from such a vast accumulation of literature. In many places I have felt the lack of definite, detailed, and truly oecological information concerning various questions, and, as in 1895, I must confess that my ideal is far from being realized. The oecology of plants is a subject still in its infancy ; numerous investigations must be made before the foundations can be truly and rightly laid, and before a consistent, clear, and natural classification of plant-communities is achieved. In conclusion, I must express my thanks to Dr. Martin Vahl for the great interest he has shown in efforts to improve the book, and to my colleagues in Britain for the exceeding care which they have bestowed upon the production of the book in English. EUG. WARMING. Copenhagen, March, 1909. CONTENTS INTRODUCTION CHAPTER I. Floristic and Oecological Plant-geography II. Growth-forms III. Plant-communities IV. Plan of this Book page I 2 12 14 SECTION I OECOLOGICAL FACTORS AND THEIR ACTION V. Light . VL Heat VII. Atmospheric Humidity and Precipitations Vni. Movements of the Air IX. Nature of the Nutrient Substratum X. Structure of the Soil XI. Air in the Soil XII. Water in the Soil XIIL Temperature of Soil XIV. Depth of the Soil. The Upper Layers of the Soil AND THE Subsoil XV. Nutriment in Soil XVL Kinds of Soil XVIL Are the Chemical or the Physical Characters of Soil the more important? .... XVIII. The Effect of a Non-living Covering over Vege tation XIX. Effect of a Living Vegetable Covering on Soil . XX. The Activity of Animals and Plants in Soil XXL Exposure. Orographic and other Factors 16 22 28 36 40 40 43 44 50 54 55 59 65 72 75 n 80 Vlll CONTENTS CHAPTER XXII. XXIII. XXIV. XXV. XXVI. SECTION II COMMUNAL LIFE OF ORGANISMS Reciprocal Relations among Organisms Interference by Man Symbiosis of Plants with Animals Symbiosis of Plants with one another. Mutual- ism Commensalism. Plant-communities page 82 82 83 84 91 SECTION III ADAPTATIONS OF AQUATIC AND TERRESTRIAL PLANTS. OECOLOGICAL CLASSIFICATION XXVII. Aquatic and Terrestrial Plants XXVIII. Adaptations of Water-plants (Hydrophytes) XXIX. Adaptations of Land-plants XXX. Regulation of Transpiration in Land-plants XXXI. Absorption of Water by Land-plants XXXII. Storage of Water by Land-plants. Water - reservoirs . ■ XXXIII. Other Structural Characters and Growth-forms of Land-plants, and especially of Xerophytes XXXIV. Oecological Classification XXXV. Physiognomy of Vegetation. Formations. Asso- ciations. Varieties of Associations 96 97 100 102 117 119 127 131 137 SECTION IV CLASS I. HYDROPHYTES. FORMATIONS OF AQUATIC PLANTS XXXVI. XXXVII. XXXVIII. XXXIX. XL. XLI. Oecological Factors Formations of Aquatic Plants . Plankton-formation Cryoplankton. Vegetation on Ice and Snow Hydrocharid-formation or Pleuston 149 154 155 163 164 LiTHOPHiLOUS Benthos 167 XLII. Benthos of Loose Soil 173 CONTENTS IX SECTION V HELOPHYTES. MARSH-PLANTS CLASS IL CHAPTER XLin. Adaptations. Formations XLIV. Reed-swamp or Reed-formation page 185 187 XLV. BusH-svvAMP and Forest-swamp of Fresh Water . 190 SECTION VI CLASS III. OXYLOPHYTES. FORMATIONS ON SOUR (ACID) SOIL XLVI. Xeromorphy. Formations . XLVII. Low-moor Formation XLVIII. Grass-heath. Tussock Formation XLIX. High-moor Formation L. Moss-tundra or Moss-heath LI. Lichen-tundra or Lichen-heath LII. Dwarf-shrub Heath LIII. Bush and Forest on Acid Humus Soil 193 196 199 200 205 208 210 214 SECTION VII CLASS V. HALOPHYTES. FORMATIONS ON SALINE SOIL 218 LIV. Introductory General Remarks on Halophytes LV. Adaptations of Halophytes LVI. LiTHOPHiLous Halophytes LVII. Psammophilous Halophytes . LVIII. Pelophilous Halophytes LIX. Salt-swamp and Salt-desert LX. Littoral Swamp-forest, Mangrove 219 224 225 230 233 234 SECTION VIII CLASS VI. LITHOPHYTES. FORMATIONS ON ROCKS LXI. Rocky Country 239 LXII. LiTHOPHYTES 24O LXIII. Chasmophytes 243 LXIV. Formations on Shingle and Rubble . . . 246 CONTENTS SECTION IX CLASS IV. PSYCHROPHYTES. FORMATIONS ON COLD SOIL CHAPTER PAGE LXV. Climatic Conditions in Subglacial Fell-fields . 248 LXVI. Adaptation of Species in Subglacial Fell-fields 251 LXVII. Subglacial Fell-field Formations . . . 256 SECTION X CLASS VII. PSAMMOPHYTES. FORMATIONS ON SAND AND GRAVEL LXVIII. Oecological Factors. Formations . . . 262 LXIX. Shifting, or White, Sand Dunes . . . 263 LXX. Stationary, or Grey, Sand-dune. Sand-fields. Dune-heath. Dune-bushland. Dune-forest. Other examples of Psammophilous Vegetation 265 SECTION XI CLASS IX. EREMOPHYTES. FORMATIONS ON DESERT AND STEPPE LXXI. Oecological Factors. Formations . . . 273 LXXII. Desert 274 LXXIII. Shrub-steppe 278 LXXIV. Grass-steppe 281 LXXV. Sibljak 288 SECTION XII CLASS VIII. CHERSOPHYTES. FORMATIONS ON WASTE LAND LXXVI. Waste Herbage LXXVII. BusHLAND ON Dry Soil .... 289 291 SECTION XIII CLASS X. PSILOPHYTES. SAVANNAH-FORMATIONS LXXVIII. Savannah-formations 293 LXXIX. Thorny Savannah 293 LXXX. True Savannah 295 LXXXI. Savannah-forest 299 CLASS XL CHAPTER LXXXIL LXXXIIL LXXXIV. LXXXV. CONTENTS SECTION XIV SCLEROPHYLLOUS FORMATIONS. AND FOREST XI BUSH SCLEROPHYLLOUS VEGETATION AND FORMATIONS Garigue. Tomillares MaQUI : SCLEROPHYLLOUS SCRUB SCLEROPHYLLOUS FOREST .... PAGE 304 305 308 SECTION XV CLASS XII. CONIFEROUS FORMATIONS. FOREST LXXXVI. Evergreen Coniferae 310 LXXXVII. LXXXVIIL LXXXIX. xc. XCI. XCII. XCIII. SECTION XVI CLASS XIII. MESOPHYTES Mesophytic Vegetation and Formations Arctic and Alpine Mat-grassland and herbage .... Meadow Pasture on Cultivated Soil Mesophytic Bushland . Deciduous Dicotylous Forest Evergreen Dicotylous Forest Mat- 317 318 322 326 328 329 337 SECTION XVII STRUGGLE BETWEEN PLANT COMMUNITIES XCIV. Conditions of the Struggle . . . . XCV. The Peopling of New Soil XCVI. Changes in Vegetation induced by Slow Changes IN Soil fully occupied by Plants XCVII. Change of Vegetation without Change of Climate or of Soil XCVIIL The Weapons of Species XCIX. Rare Species C. Origin of Species LITERATURE INDEX 348 349 358 364 366 368 369 374 407 INTRODUCTION CHAPTER I. FLORISTIC AND OECOLOGICAL PLANT- GEOGRAPHY Plant-geography deals with the distribution of plants upon the earth, and with the principles determining this. We may regard this distribution from two different standpoints, and accordingly may divide the subject into two branches, ftoristic plant-geography and oecological^ plant-geography ; but these are merely different aspects of the same science, touching at many points and occasionally merging into one another. Floristic plant-geography is concerned with — 1. The compilation of a ' Flora ', that is, a hst of species growing within a larger or smaller area. Such hsts form the essential basis of the subject. 2. The division of the earth's surface into natural floristic tracts (floristic kingdoms and so on-) according to their affinities, that is, according to the numbers of species, genera, and families common to them. 3. The sub-division of the larger natural floristic tracts — floristic kingdoms — into smaller natural tracts, regions, and districts, and the precise definition of these. 4. The discussion of the Hmits of distribution of species, genera, and famihes (their ' area ') ; of their distribution and frequency in different countries ; of endemism ; of the inter-relations between the floras of islands and of continents, and between those of mountains and of low- lands ; and so forth. The thoughtful investigator will not remain content with the mere recognition of facts ; he will seek after their causes. These are, in part, modern (geognostic, topographical, and climatic), and, in part, historical. The limits of distribution of a species may depend upon prevailing conditions, upon barriers now existing in the form of mountain, sea, soil, and climate, which oppose its spread ; but they may also depend upon geohistoric or geological and climatic conditions of ages long past, and upon the whole evolutionary history of the species, the site of this, and the facilities for and means of migration. In addition, problems must be dealt with concerning centres of development, the rise and age of species and genera ; and behind these lies the question of the origin of species. To deal with the yet undescribed floristic plant-geography of Denmark, * Haeckcl (1886) defined Oecology (fuKoj, a house, Xoyoy, theory) as the science treating of the reciprocal relations of organisms and the external world. Reitcr (1885) employed the term in the same sense; see MacMilIan, p. 950, 1897. ' Drude, 1884, 1886-7, 1890. WARMING B 2 INTRODUCTION chap, i for instance, it will be necessary to investigate the following items : the distribution of the species present, their arrangement in the country, the sub-division of Denmark into natural floristic sections, Denmark as a floristic portion of a larger natural district or its floristic affinity with Sweden and Norway, Germany, and other lands, the problems as to when and whence the species immigrated after the glacial epoch, the routes of their migrations and their means of migration, the problem of species left behind (vestigial plants), and the like.^ With these interesting and far-reaching results of floristic plant- geography we shall not deal in this work. This subject has been treated by Wahlenberg, Schouw, A. de Candolle, Grisebach, Engler, Drude, and others. Oecological plant-geography has entirely different objects in view : — It teaches us how plants or plant-communities adjust their forms and modes of behaviour to actually operating factors, such as the amounts of available water, heat, light, nutriment, and so forth. A casual glance shows that species by no means dispose their individuals uniformly over the whole area in which they occur, but group them into communities of very varied physiognomy. Oecology seeks — 1. To find out which species are commonly associated together upon similar habitats (stations). This easy task merely involves the determination or description of a series of facts. 2. To sketch the physiognomy of the vegetation and the landscape. This is not a difficult operation. 3. To answer the questions — Why each species has its own special habit and habitat, Why the species congregate to form definite communities, Why these have a characteristic physiognomy. This is a far more difficult matter and leads us — 4. To investigate the problems concerning the economy of plants, the demands that they make on their environment, and the means that they employ to utihze the surrounding conditions and to adapt their external and internal structure and general form for that purpose. We thus come to the consideration of the growth-forms of plants. CHAPTER II. GROWTH-FORMS Every species must be in harmony, as regards both its external and internal construction, with the natural conditions under which it lives ; and when these undergo a change to which it cannot adapt itself, it will be expelled by other species or exterminated. Consequently one of the most weighty matters of oecological plant-geography is to gain an understanding of the epharniony of species. ^ This may be termed its growth-form in contradistinction to its systematic form. It reveals ^ Warming, 1904. ^ Vesque (1882 a) defines ' L'epharmonie ' as ' I'etat de la plante adaptee '. Epharmosis, a term also invented by Vesque, on the other hand, denotes the act of adaptation (or the behaviour) of organisms exposed to new conditions. CHAP. II GROWTH-FORMS 3 itself especially in the habit, and in the form and duration, of the nutritive organs (in the structure of the foliage-leaf and of the whole vegetative shoot, in the duration of hfe of the individual, and so forth), but shows to a less extent in the reproductive organs. This subject leads us into deep morphological, anatomical,^ and physiological investigations ; it is very difficult, yet very alluring ; but only in few cases can its problems be satisfactorily solved at the present time. Thus we impinge upon the problem of the origin of different species. But difficulty is imparted to the question under discussion by the circumstance that, not only is a species changed in form by external factors and capable of adapting itself to these, but each species is also endowed with certain hereditary tendencies, which, for inherent but unknown causes, evoke morphological characters that cannot be corre- lated with the present environment and are consequently inexplicable. These inherent tendencies, differing as they do according to systematic affinity, render it possible for different species, in their evolution under the influence of identical factors, to achieve the same object by the most diverse methods. While one species may adapt itself to a dry habitat by means of a dense coating of hairs, another may in the same circum- stances produce not a single hair, but may elect to clothe itself with a sheet of wax, or to reduce its foliage and assume a succulent stem, or it may become ephemeral in its life-history. On the one hand, in very few families of flowering plants (e.g. Nym- phaeaceae) have the different species assumed approximately the same growth-form, or in other words acquired in harmony with the same environment the same external form, and similar adaptations and habits of life. As a rule, the members of a family differ widely from one another, both in form and in their demands upon the environment. On the other hand, species belonging to families systematically wide apart may be extraordinarily like one another in regard to the structural features of the vegetation-shoot. A striking example of this is afforded by Cactaceae, ractus-like species of Euphorbia and of Stapelia. These furnish an admirable example of a single, marked growth-form which is clearly adapted to definite external conditions, appearing in three families that are distant in affinity {epharmonic convergence). Another illustration is seen in the case of Hydrocharis, Limnanthemum, and others, which display so puzzling a likeness in form of their leaves to the Nymphaeaceae. The term growth-form used in this work nearly corresponds with the term vegetative form employed by some botanists, but involves more rigid scientific definition. The term vegetative form was introduced by Grisebach and has been employed in literature in various senses, so that it requires explanation. Within the same vegetative form were included all those species that are more or less closely similar in design and appear- ance, whether they be of close or very distant affinity. The design ' Anatomy, particularly stimulated by Habcrlandt, has recently been greatly enriched by numerous researches dealing with the question of the harmony between structure and function or environment. Duval-Jouve (1875) ^^.d already defined work of this kind in the following words : ' L'objet de la pr6sente etude est de constater les principales dispositions des tissus dans les fcuillcs des Graminces, et de determiner, autant que possible, le rapport de certaines dispositions avcc les ionctions imposees par le milieu.' B 2 4 INTRODUCTION chap, ii expresses itself not only in external features (form of the vegetative shoot, and of the leaves, position of the renewal buds, and so forth), but also in the anatomical structure and the behaviour of the plant in life (defoliation, duration of life, and the like). In this regard it is the vegetative organs, especially the vegetative shoots, that are of signifi- cance, whereas in systematic botany it is the floral structure that is of import. The vegetative shoot adapts itself to the conditions prevailing in regard to its nutrition ; but the flower follows other laws, other aims, and particularly adopts very diverse methods of pollination. In the morphology and anatomy of the vegetative shoot are reflected the climatic and assimilating conditions ; whereas floral structure is scarcely or not at all influenced by climate, but preserves the impress of phyletic origin under very different conditions of life. An examination of the catalogues or ' systems ' of vegetative forms compiled from time to time will further elucidate the matter. Humboldt (1805) was the first to lay stress upon the significance of plant-physiognomy in relation to the landscape : ' Above sixteen different forms of vegetation are principally concerned in determining the aspect or physiognomy of Nature.' ^ He treated the following nineteen forms in greater detail : those corresponding to the palm, banana, malvaceous and bombaceous plants, Mimosa, heath. Cactus, orchid, Casuarina, conifer, Pothos (aroid), liane, aloe, grass, fern, lily, willow, myrtle, melastomaceous plant, laurel. This is, of course, merely a superficial distinction among physiognomic and systematic types ; each of these ' forms ' in reality includes plants with very diverse modes of life. A purely physiognomic system is devoid of scientific significance, which is introduced only when physiognomy is founded upon physio- logical and oecological facts. Grisebach ^ made the next important attempt in this direction. He established fifty-four, and subsequently sixty, vegetative forms, arranged in a physiognomic ' system ', and he endeavoured to prove that this demonstrated a connexion between the external form and the environment, in particular the climatic conditions ; according to him a physiognomic type is for the main part also an oecological one. Whilst Grisebach clung in the main to physiognomy and entered into such minutiae as to distinguish, for instance, the laurel-form with stiff, ever- green, undivided, broad leaves, from the olive-form with stiff, evergreen, undivided, narrow leaves, and the liane-form with reticulate-veined leaves, from the rattan-form with parallel-veined leaves ; yet with his sixty forms he distinguished by no means all growth-forms, but rather, as he himself pointed out, only such as could serve to indicate country or chmate by reason of their growing in numbers together. Furthermore, he did not in the least take anatomical structure into consideration, nor did he adequately appreciate epharmony,^ In 1884, Warming, having in view the North-European Spermophyta, gave a general survey of growth-forms which he arranged in fourteen chief groups with many sub-groups, based upon morphological and biological characters ; among other characters the vegetative methods of migration occupied a prominent position. Drude, in 1895, justly ^ Humboldt, 1805, vol. ii, p. 18. * Grisebach, 1872. ' See Reiter, 1885; Warming, 1908. CHAP. II GROWTH-FORMS 5 remarked that these did not take sufficient note of geographical con- siderations. Reiter, in 1885, was the next to devote detailed consideration to the subject. His standpoint was sound and he laid stress upon internal structure, particular obser\'ation of truly adaptive features, and a due regard for all the types of a characteristic mode of life and of special design — as opposed to a regard for merely such of these types as occur in great numbers. Nevertheless Reiter's ' system ' is capable of im- provements. Subsequently Drude ^ dealt with the question. He adopted the ' biological-geographical ' standpoint as resting on the answers to the two questions : What functional role does any particular species of plant play in the vegetation of a definite country ? How does it complete the whole of its periodic life-cycle under the conditions prevailing in its habitat ? As features of the greater importance he denoted, ' the duration of organs and the protective measures against injuries during the resting period,' also ' the position of the renewal-shoot on the main axis in relation to hibernation '.^ In his later work he divided plants into thirty-five classes of vegetative forms. Krause,^ and later Pound and Clements,* gave the main outlines of systems. That of Pound and Clements approaches, as a whole, Drude's. It ranges plants in the following main groups : Woody plants, half- shrubs, pleiocyclic herbs, hapaxanthic herbs, water-plants, hystero- phytes, and thallophytes, and it contains thirty-four sub-groups. Raunkiar ^ devised a system, in which, like Drude, he laid greatest stress upon the adaptation of plants to enable them to five through the unfavourable seasons, as particularly evinced by the degree and kind of protection afforded to the dormant buds and shoot-apices. His five chief groups were phanerophytes, chamaephytes, hemicryptophytes, cryptophytes, and therophytes. The most recent treatment of this subject is due to Warming, who, since 1890, has published various papers dealing with the structure of growth-forms and the parts they play in formations. In 1908 ^ he attempted to map out the main lines of a system of w'hich the following is an outhne : — Just as species are the units in systematic botany, so are growth- forms the units in oecological botany. It is therefore of some practical importance to test the possibility of founding and naming a limited number of growth-forms upon true oecological principles. It cannot be sufficiently insisted that the greatest advance, not only in biology in its wider sense, but also in oecological phyto-geography, will be the oecological interpretation of the various growth-forms : from this ulti- mate goal we are yet far distant. It is an intricate task to arrange the growth-forms of plants in a genetic system, because they exhibit an overwhelming diversity of forms ' Drude, 1887, 1889, 1890, 1896. ^ Drude, 1890. p. 69 ; 1896, p. 46". •■' Krause, 1891. * Pound and Clements, 1898. ' Raunkiar, 1903, 1905, 1907. * Warming, 1908. 6 INTRODUCTION chap, ii and are connected by the most gradual intermediate stages, also because it is difficult to discover guiding principles that are really natural. Nor is it an easy task to find short and appropriate names for the different types. Genetic relationships, and purely morphological or anatomical characters, such as the venation and shape of leaves, the order of succes- sion of shoots, monopodial and sympodial branching, are of very slight oecological or of no physiognomic significance. Oecological and physio- logical features, particularly the adaptation of the nutritive organs in form, structure, and biology, to climate and substratum or medium, are of paramount importance. Cases, however, are not wanting in which oecological grouping runs parallel with systematic classification. Growth-forms may be arranged in the following six main classes, namely : — 1. Heterotrophic. 4. Lichenoid. 2. Aquatic. 5. Lianoid. -r 3. Muscoid. 6. All other autonomous land-plants. Heterotrophic growth-forms are shown by all holosaprophytes and holoparasites, which are undoubtedly derived from autophytes and are degenerate in form and structure. Hemi-saprophytes and hemi-parasites, on the contrary, are under the dominance of chlorophyll and exhibit the same rich diversity of form as other green plants. (See Chapter XXV.) Aquatic growth-forms differ from those shown by land-plants so widely as regards their morphology, anatomy, and physiology, that they must be regarded as constituting a separate class. (See Section IV.) The muscoid and liche^ioid growth-forms are seen, almost only, in mosses and lichens. Their powers of enduring extreme loss of water and of rapidly replacing this by means of absorption over the whole free surface, aie oecologically very important. Associated with these characters are a number of others. The distinction between the muscoid and lichenoid types lies in the method of nutrition, as autotrophic and symbiotic respectively. (See Chapter XXV.) The lianoid growth-form is mainly determined by social conditions, and shows pecuhar oecological and physiological characters. (See Chapter XXV.) Epiphytes, on the contrary, form an edaphic community of autotrophic land-plants including many different types. The sixth class includes the growth-forms adopted by all the remaining autotrophic land-plants that contain chloroph} 11 and, as regards nutrition, are independent of other plants, and are thus autonomous. The growth- forms of Pteridophyta are included here, although these differ so widely from those of Spermophyta as regards their reproductive organs. These main classes may, in turn, be divided into sub-classes. In particular, the growth-forms of the autotrophic plants of the sixth class admit of grouping in categories the foundation of which is the duration of the individual plant and of its parts. Upon this basis these plants are divided into monocarpic and polycarpic : 1 the former produce flower and fruit (or spores) once, and then die ; the latter may produce fruit repeatedly before death claims them. * In recent times these CandoUean terms have been suppressed often in favour of A. Braun's ' hapaxanthic ' {ana^, once ; avdos, flower) and Kjellman's ' pollakanthic ' {noXkaKis, several times ; avdos, flower). CHAP. II GROWTH-FORMS 7 We adopt the following sub-division : I. MONOCARPIC HERBS, which include the following groups : — (a) Aestival annual plants. The whole cycle of Ufe is completed within one vegetative period, varying from a few weeks, as in ephemeral desert-plants, to several months. Shoots foliaged with elongated inter- nodes (monocyclic). No vegetative organs of storage. The unfavourable season passed in the form of seed.^ Adaptation to dry climates and locahties, and to disturbed soil (littoral sand, cultivated soil, and the like). (6) Hibernal annual plants. These germinate in autumn, and con- clude their existence with the production of fruit in spring. Rosette- shoots are usually prevalent. Otherwise like those belonging to the preceding group (a). (c) Biennial-perennial (dicyclic, pleiocyclic -) herbs produce in their first vegetative period or in several successive ones rosettes of leaves, and in the following period the flowering shoot, which usually bears foliage. The foliage-leaves often live through winter. Buds open. Form of the shoot as in (b). Reserve-food often stored in tuberous axial organs (Beta vulgaris, Daucus Carota). Occurring in cold-temperate climates on open soil, also as cultivated plants. II. POLYCARPIC PLANTS In the case of the polycarpic plants it is necessary to consider, first, their adaptation to climate, and in particular the season unfavourable to plant-life ; secondly, the vegetative season ; and, finally, the conditions prevailing in regard to the soil, which Schimper terms edaphic conditions. Of greatest importance is — 1. Duration of the vegetative shoot: lignified axes of trees, shrubs, and undershrubs ; perennial herbaceous shoots ; herbaceous shoots deciduous after a short period. And closely associated with this is — 2. Length and direction of the inter nodes : whether the shoots have short internodes (rosette-shoots) or long internodes, and whether the latter are erect (orthotropous) or prostrate and creeping (plagiotropous). 3. Position of the renewal-buds during the unfavourable season [high up in the air, near the soil, under the surface of the soil, or buried in the soil (geophilous)].^ Of less importance is — 4. Structure of the renewal-buds or of buds in general. All stem-apices and very young leaves are protected by leaves ; this protection is accom- plished in some cases merely by older foliage-leaves (open buds), in other cases by specially differentiated protective organs, which are either parts of foliage-leaves or definite bud-scales. This depends less upon climate than upon the form of the assimilatory shoots ; short-jointed shoots with leaves in rosettes usually have open buds ; long-jointed shoots are more varied. These differences in the shoot are physiognomi- cally important, not only in themselves, but because the former shoots are branched httle or not at all, while the latter are usually richly so. ' See Ascherson, 1866. * Warming, 1884. * See the nomenclature in the paper by Raunkiar, 1905, 1907. 8 INTRODUCTION chap, ii 5. Size of the plant is of some moment, not only because in the struggle for existence the taller plants are enabled to establish a supremacy more easily, but also because they are more exposed to inclemency of climate ; shrubs reach greater altitudes and latitudes than trees, while dwarf- shrubs and herbs extend even farther than shrubs. 6. Duration of the leaves varies ; some live for only a few months, others for years. In all chmates deciduous (summer-green, rain-green) and evergreen plants are met with side by side. This distinction is associated with edaphic conditions, and can be utilized in the classifica- tion of sub-divisions. 7. The adaptation of the assimilatory shoot to the conditions of tran- spiration, is determined by the substratum, and by the climate. Some plants assimilate mainly by their leaves, which exhibit very great variety in shape and structure ; but others depute their assimilatory function to the stem, and reduce their leaves. The shapes of leaves (their vena- tion, division, and so forth) depend partly upon systematic affinity, and partly upon the surrounding medium and climate ; they are probably of but slight value as a basis of oecological classification. 8. The capacity for social life is of great importance in the struggle between species, and consequently in the composition and physiognomy of the plant-community. This capacity is due in some cases to the prolific production of seed, but usually to more vigorous vegetative multiplication by means of travelling shoots, or shoots given off from the root. And this latter is to some extent determined by the soil (moist or wet soil, loose sandy soil, and so forth). In accordance with these considerations polycarpic plants may be grouped under four sub-classes : — (a) Renascent (Redivivus) herbs. (6) Rosette-plants. (c) Creeping plants. {d) Plants with erect long-lived shoots.^ We will now proceed to discuss these sub-classes, and, as a final step in the process of sub-division, we shall be able to define types that can be named after definite species or genera, e.g. Primula-type, Bromeha- type, Cycas-type, and so forth. (a) Renascent herbs. Polycarpic herbs whose assimilating and flowering shoots develop at a definite climatic time. The plant there- fore passes through a resting period, during which its hypogeous or epigeous renewal-shoots are protected by scale-leaves. When the favourable season arrives, the plant once more reveals itself (and is thus renascent). The photophilous shoots are aestival-annual and usually have long internodes and mesophilous leaf -structure. The perennating hypogeous parts are necessarily provided with reserve food. A great variety of types is included amongst these herbs. Some are ' spot-bound ' [stationary, sedentary] ; others are travelling plants. Among them are the following : — ^ This classification approximates to that proposed by Krause (1891). For the further sub-division of these sub-classes, and for the distinction of the various types included among these fundamental forms, we must refer to the characters mentioned in the paragraphs numbered 4-8. CHAP. II GROWTH-FORMS 9 Multicipital rhizomes. Herbs with a multicipital rhizome ^ ordin- arily have axes with short internodes which are hypogeous, but often he well above the soil, and, at the bases of the connected stems, usually bear irregularly placed buds, which are protected with bud-scales, and give rise to erect flowering long-shoots and inflorescences. Growth is caespitose : — On grassland and savannah : Vincetoxicum ofiicinale, Silene inflata, Rheum, Dahlia variabilis (with food-storing roots). In savannah-plants the rhizome often becomes a thick, irregularly lignified ' xylopodium '.^ Mat-geophytes.^ Perennial spot-bound herbs, mostly monocotylous. The renewal-buds often deeply embedded in the soil on a short-jointed, feebly branched shoot, which usually contains a large amount of food- material. Occurring especially in hot dry lands. Rest during summer. Epigeous vegetative shoot with long internodes or, in some species, with a rosette of leaves — with stem-tubers : Crocus, Arum maculatum, Amorphophallus, Eranthis hyemalis, species of Corydalis ; with root-tubers : Ophrydeae ; with bulbs : many Liliaceae and Amaryllidaceae ; with perennial tuberous stem : Cyclamen. Travelling geophytes (Rhizome-geophytes). Geophytic perennial herbs with horizontal hypogeous branched scaly shoots, from which are emitted either foliage-leaves or erect epigeous shoots that bear foliage and flowers. The renewal-buds are inserted on the hypogeous shoots, and have bud- scales. Great variety is exhibited in the length of the internodes and other details : — On loose soil of dunes : Ammophila arundinacea, Carex arenaria. On loose humus soil in the forest : Polygonatum multiflorum, Anemone nemorosa, Asperula odorata. Zingiber officinale. On mud in water or swamp : Phragmites communis, Equisetum limosum, Hippuris vulgaris. In some cases the subterranean shoots become more or less tuberous food-reservoirs, and their elongated thin parts are then very short-hved : Solanum tuberosum, Stachys tuberifera. Special forms are where — Epigeous parts are exclusively foliage-leaves : Pteris aquilina. Hibernation is by roots : Cirsium arvense. {b) Rosette-plants. The erect foliaged shoot has short internodes and consequently closely set leaves ; it is usually epigeous and evergreen with naked buds, although often only the youngest leaves remain fresh and, in winter, are protected by the old faded foliage. The type occurs on — Open land (grassland, moors, arctic heaths, and so forth) : Arctic and alpine fell-fields : Papaver nudicaule and Draba. In the cold-temperate chmate there are many rosette-herbs, including those : — With leaves sessile elongated : Plantago major, Armeria vulgaris, Taraxacum vulgare ; ' 'Crown-formers' (Hitchcock) (1898); also see Pound and Clements, 1898, p. 106 ; Drude, 1896, p. 48. ' Lindman, 1900 ; also see Warming, 1892. ' See Raunkiiir, 1905, 1907. 10 INTRODUCTION chap, ii With leaves long-stalked, broad, more or less cordate : Soldanella alpina, Anemone Hepatica ; With leaves succulent, but thin in most cases and destroyed in winter, except the youngest : Crassulaceae. With runners, by which they can multiply freely : Ranunculus repens, Fragaria, Potentilla anserina, and Hieracium Pilosella. With flowers upon — Foliaged shoots with long internodes : Alchemilla, Geum urbanum ; Leafless scapes : Drosera vulgaris, Primula officinalis, Taraxacum, and others. Among rosette-plants must be reckoned a number of Graminaceae, Cyperaceae, Eriocaulonaceae, and other Monocotyledones, with grass- like leaves that are crowded together, close to the ground, in dense com- pound rosettes. These plants are particularly found in open country : grassland, steppe, savannah, and moor. On high mountains and in arctic lands there are numbers of rosette- plants, whose perennating leaves are more or less coriaceous and fleshy, as in species of Saxifraga. On hot, sunny, rocky sites the leaves become thicker and fleshy, so that there develop such forms as Sempervivum, Echeveria, and other Crassulaceae. These lead to such types as Aloe, Agave, Mesembryanthemum, Bromeliaceae, and others, with undivided, fleshy or leathery, often thorny, long-lived leaves. Musa-form. It is well to include among rosette-plants the types belonging to the Musa-form : Gigantic tropical herbs with a perennial, epigeous, evergreen, false stem, composed of the involute leaf-sheaths, and arising from a subterranean rhizome, and large leaf-blades of characteristic venation. Most of the species are stemless, but others have tall stems, and thus lead to tuft-trees. Tuft-trees.^ Shoots with short internodes ; leaves densely set on the end of the shoot, large, and few ; buds usually naked. Stem unbranched or with only a few thick branches, none of which are thrown off : — 1. Trunk unbranched and usually exhibiting no secondary growth in thickness ; leaves large and divided : tree-ferns, palms, cycads. 2. Trunk sometimes sparsely branched, undergoing secondary thicken- ing ; leaves undivided, linear : arborescent Liliaceae (Yucca, Dracaena, Cordyline, Xanthorrhoea, Vellozia). 3. Strelitzia-form. (c) Creeping plants. The assimilating shoots are prostrate, plagio- tropous, rooting, often long-jointed, and sometimes bear short erect branches. Buds naked, or encased in scales : — Some are herbs : Lycopodium clavatum, Lysimachia Nummularia, Hydrocotyle vulgaris, Menyanthes trifoliata, and Ipomoea Pes-caprae. Others are woody : Arctostaphylos Uva-ursi, A. alpina (deciduous), Empetrum, Vaccinium Oxycoccos, and Linnaea borealis. It is difficult to distinguish this group from the rosette-plants, such as Ranunculus repens, Potentilla anserina, and Fragaria, which possess epigeous means of travelling. Jungermannia.form. In the tropics there are many epiphytic creeping ^ Drude's (1896) Schopf bourne. CHAP. II GROWTH-FORMS ii plants : for instance, species of Philodendron, ferns, and others. The leaves are often short-stalked, more or less orbicular, and distichous (so that the species often bears such names as ' nummulariaefolia ' and ' serpyllifolia ') ; this constitutes the Jungermannia-jorm. (d) Land-plants with long erect long-lived shoots. To this sub- class belong very many species that possess more or less woody stems, or, less frequently, herbaceous stems. Among them arc the following : — Cushion-plants. Shoot-system richly ramified, often with the branches densely packed to form hemispherical cushions. FoHage-leaves usually small, more or less evergreen, remaining attached for a long time in a faded condition, and decaying slowly. Buds open. Adaptation to a cold, physiologically dry cHmate, or to cold dry air and a hot soil : dicoty- lous plants such as Azorella, Raoulia, Silene acauhs, species of Saxifraga, Draba, Dionysia, Aretia. Mosses assume similar cushion-shapes. Undershrubs (SufFrutices). There are a great many kinds of these which are transition-forms between herbs and shrubs, and have incom- pletely Hgnified stems, or lignified stems that soon perish. The yearling shoot often is branched. The buds are often naked. They include various types : — Labiate type. Considerable parts of the flowering shoots die after blossoming. Many are Mediterranean plants, adapted to a mild winter during which rain falls, and are particularly found in Continental steppes and maquis : species of Salvia, Lavandula, Thymus, Helianthemum, Artemisia, Ruta. Acanthaceous type. Erect, weakly lignified, tropical forest-plants, with thin leaves : species of Acanthaceae, Rubiaceae, Verbenaceae, Piperaceae. Rhizomatous undershrubs having subterranean runners : Vaccinium -Myrtillus and V. Vitis-Idaea. Cane-under shrubs} with lignified but commonly monocarpic shoots : Rubus Idaeus, other species of Rubus. Then there are also — Soft-stemmed plants. Stems thick, green, soft, scarcely lignified ; leaves usually very large. Plants essentially belonging to tropical forest and marsh, and epiphytes : Araceae specially. Succulent-stemmed plants : Cactus-form. Stem lignified, green, juicy, unbranched or feebly branched, often thorny. Fohage-leaves suppressed. Buds sunken, often protected by hairs. Adaptation to a hot chmate, with prolonged drought. On deserts and rocks, usually forming very open associations. Varying in size from trees to prostrate forms : Cacta- ceae, species of Euphorbia, Stapeha. Woody plants with long-lived, lignified stems. Buds naked or seal}'. Evergreen or deciduous. To these belong — Canopy-trees.^ Dicotylous and gymnospermous trees, with well- branched crown and many small leaves. The crown increases in size from year to year, and the stem necessarily exhibits corresponding secondary increase in thickness. Buds scaly, or at any rate not typically open. Leaves, deciduous or evergreen ; extremely varied in form, venation, and structure ; large, broad, and thin, simple or compound ; or small, broad and coriaceous (sclerophyllous trees) ; pinoid (Coniferae) ; ' Drudc's (1896) Schosslingsstrancher. ^ Drude's (1896) Wipfclbdione. 12 INTRODUCTION chap, ii cricoid (Ericaceae) ; cupressoid (Cupressus) ; scale -like in aphyllous trees (Casuarina, Halimodendron). Dwarf-forms occur, some of them having succulent leaves. In certain tropical forms the leaves are aggre- gated at the ends of long, feebly branched twigs : for instance, in Cecropia and Carica. Shrubs (frutices) and dwarf-shrubs {jruticuli). Low trunkless woody plants, with the variety in the construction of the leaf and shoot seen in canopy trees : — Switch-shrubs : erect, long assimilating stems and small caducous leaves. Succulent-leaved shrubs also belong here : species of Crassulaceae, Mesembryanthemum, Chenopodiaceae, and others. Gramineous shrubs represent a pecuhar type. In the bamboo-form there arise from the subterranean stem tufts of many richly and charac- teristically branched evergreen stems, which undergo no secondary thickening. Leaves grass-like. Particularly a tropical growth-form forming forest and bush. Aphyllous shrubs : which sometimes have equisetoid or salicornioid shoots. The sub-classes of the other classes of growth-forms will be referred to later, when the environmental conditions relating to them are discussed. CHAPTER III. PLANT-COMMUNITIES Oecological Botany has further to investigate the natural plant- communities, which usually include many species of extremely varied growth-form. Certain species group themselves into natural associations, that is to say, into communities which we meet with more or less frequently, and which exhibit the same combination of growth-forms and the same facies. As examples in Northern Europe may be cited a meadow with its grasses and perennial herbs, or a beech-forest with its beech-trees and all the species usually accompanying these. Species that form a community must either practise the same economy, making approximately the same demands on its environment (as regards nourishment, light, moisture, and so forth), or one species present must be dependent for its existence upon another species, sometimes to such an extent that the latter provides it with what is necessary or even best suited to it (Oxalis Acetosella and saprophytes which profit from the shade of the beech and from its humus soil) ; a kind of symbiosis seems to prevail between such species.^ In fact, one often finds, as in beech-forests, that the plants growing under the shade and protection of other species, and belonging to the most diverse families, assume growth-forms that are very similar to one another, but essentially different from those of the forest-trees, which, in their turn, often agree with one another .^ Oecological plant-geography has also to inquire into the kinds of natural communities in existence, their special methods of utihzing their resources, and the frequent intimate association together of species ^ See Chap. XXVI regarding Unlike Commensals. ^ Warming, 1901. CHAP. Ill PLANT-COMMUNITIES 13 differing in growth-form and economy. The physical and other charac- ters of the habitat play a fundamental part in these matters, and, for this reason, form the introductory subject-matter of Section I in this work. The oecological analysis of a plant-community leads to the recognition of the growth-forms composing it as its ultimate units. From what has just been said in regard to growth-forms it follows that species of very diverse physiognomy can very easily occur together in the same natural community. But beyond this, as already indicated, species differing widely, not only in physiognomy but also in their whole economy, may be associated. We may therefore expect to find both great variety of form and complexity of inter-relations among the species composing a natural community ; as an example we may cite the richest of all types of communities — the tropical rain-forest. It may also be noted that the physiognomy of a community is not necessarily the same at all times of the year, the distinction sometimes being caused by a rotation of species. In countries far apart there are to be found communities identical in type, but entirely different in floristic composition. Meadows in North America and in Europe, or the tropical forest in Africa and in the East Indies, may show the same general physiognomy, the same kinds of constituent growth-forms, and the same type of natural community, though of course their species are entirely different and thus introduce slight physiognomic differences. The different communities, it need hardly be stated, are scarcely ever sharply marked off from one another. Just as soil, moisture, and other external conditions are connected by the most gradual transitions, so likewise are the plant-communities, especially in cultivated lands. In addition, the same species often occur in several widely different com- munities ; for example, Linnaea borealis grows not only in coniferous forests, but also in birch-woods, and even high above the tree-limit on the mountains of Norway and on the fell-fields of Greenland. It appears that different combinations of external factors can replace one another and bring into existence approximately the same community, or at least can satisfy equally well one and the same species, and that, for instance, a moist climate often completely replaces the forest-shade of dry climates. It is evident that all these circumstances render very difficult the correct scientific interpretation, delimitation, diagnosis, and systematic classification of plant communities, especially when we consider the condition of our present knowledge — for we have only just commenced to investigate growth-forms and communities, and what we do not know seems infinite. Another difficulty, to which allusion has already been made, is to assign suitable names to the more or less comprehensive, principal or subordinate, plant-communities occurring on the Earth and imparting to the landscapes entirely different physiognomies. Nor is it easy to estimate the true significance of floristic distinctions. 14 INTRODUCTION chap, iv CHAPTER IV. PLAN OF THIS BOOK Whereas geography proper has to give information in regard to the species and the distribution of associations of plants in various parts of the earth, oecological plant-geography treats of the following : — 1. The external factors affecting the plant's economy ; the effects of these factors upon the external and internal structure of the plant, upon the duration of life and other biological relations, and upon the topographical distribution of species. These factors and their effects are discussed in Section I. 2. The grouping and diagnosis of the plant-communities occurring on the Earth. In connexion with each class the endeavour must be made to discover the determinant factors, the modes in which they are combined, and in which they possibly may replace one another. In Sections II and III the communities are treated generally, and their special treatment follows in Sections IV to XVI. 3. The struggle between plant-communities. This is dealt with in Section XVII. The various factors require to be dealt with separately, although this is a disadvantage, partly because they never work singly but work in complex combination, and partly because it is by no means clear in all cases what must be ascribed to one factor and what to another. Fol- lowing Schouw, one may divide the factors into directly and indirectly operating factors. Among the direct factors are — {a) Geographical factors. It is thus that Drude describes the factors that work over great areas, because they are dependent upon the Earth's course round the sun, and upon the latitude : composition of the air, light, temperature, atmospheric precipitations and humidity, movements of the air. (6) Topographical factors Those that operate within smaller, more localized, limits : the chemical and physical nature of the soil — the ' edaphic factors ' as Schimper ^ termed them. In the first section of this work the following arrangement is adopted : Atmospheric factors are treated in Chapters V-VIII, as follows : — Chapter V. Light. ,, VI. Temperature. ,, VII. Atmospheric humidity and precipitations, „ VIII. Movements of the air. To the atmospheric factors likewise also belongs the composition of the atmosphere. Excepting as regards the variable humidity of the air, this is very constant over the whole world. The two gases playing the greatest part in plant-life, oxygen and carbon dioxide, are present nearly everywhere in the same relative proportion. The more recent investiga- tions have proved that the relative amount of carbon dioxide at great alti- tudes and on lowlands is the same. And between the amount of carbon ' Schimper, 1898. CHAP. IV PLAN OF THIS BOOK 15 dioxide in the air of forests and of open land there is no difference.^ The composition of the atmosphere is therefore devoid of geographical signifi- cance. Edaphic factors are treated in Chapters IX-XVII as follows : — Chapter IX. The nutrient substratum — its constitution. „ X. Its structure. „ XL Its air. „ XII. Its water. „ XIII. Its temperature, „ XIV. Its dimensions. „ XV. Its nutriment. „ XVI. The kinds of soil. „ XVII. The problem as to the chemical or physical action of soil. Indirect factors are — The contour of the Earth's surface, configuration of land and sea, altitude, latitude, and other active and modifying factors. These are treated as follows in Chapters XVIII-XXI : — Chapter XVIII. The influence of a non-living covering over vegetation. „ XIX. The influence of a hving vegetable covering on soil. „ XX. The activity of animals and plants in the soil. „ XXI. Exposure. Orographic, and other factors. The atmospheric factors coincide approximately with Drude's 'geo- graphically operating ' factors, because they are usually almost constant over large parts of the Earth's surface. Consequently these are the ones that more than all others make their impress upon the vegetation of a country, for all plant communities are subject to their influence. Edaphic and indirect factors, on the contrary, determine the topographical differences in the vegetation.^ ' Hann, 1897, vol. i, p. 76. ^ Further particulars are to be found in the works by Sachssc, Deherain, Vallot, Ramann, Drude, Grabner, Schimper, and Clements. SECTION I OECOLOGICAL FACTORS AND THEIR ACTION CHAPTER V. LIGHT Radiant energy is a marked geographical factor the intensity of which varies according to the season, latitude, altitude, atmospheric humidity, and cloudiness. For the sake of brevity it will be designated by the term ' light '. Both the intensity and the duration of light vary, and therefore are of import. The intensity may be measured by the aid of the eye or, better, by the action on silver salts.^ The most powerful action is exerted on the eye by yellow rays, on silver salts by the violet and ultra-violet rays, and upon plants by the red and blue rays. Conclusions as to the action of light upon plants cannot be drawn directly from observations upon its luminosity and chemical intensity. Light plays a part — 1. By its chemical action on chlorophyll. Without light there would be no production of chlorophyll, no assimilation of carbon dioxide, and no life upon the globe. Commencing at a certain minimum intensity of light (which varies according to the species) assimilation increases as the intensity of light rises, until an optimum is attained. Light that is too strong is injurious in action.- 2. By its heating action. A plant exposed to insolation attains a con- siderably higher temperature than does the surrounding air ; while shaded organs, by reason of radiation, become colder than the air. 3. By promoting transpiration through rise of temperature. In this case also we must assume the existence of an optimum, which likewise varies with the species and generally does not coincide with the optimum for nutrition.^ Against excessive transpiration the plant makes various provisions. 4. By influencing growth movements, the lie of foliage-leaves, and nearly all vital phenomena. And in these cases, too, the composition of the light as regards the admixture of rays of short and long wave- length (especially a clear or a clouded sky)* is of great import.* 5. By influencing the distribution of plants. The earth, viewed as a whole, has scarcely a spot from which plant-life is excluded by insuffi- ciency of light ; for although the light may be too weak at certain seasons (e.g. during the polar night), yet it becomes at other times strong enough to call forth life. But when we descend to the depths either of the solid earth or of the water, life, dependent as it is upon ^' See Wiesner, 1876a, 1876&, 1893, 18956. ^ Wiesner, 1898, 1900, 1904, 1905, 1907; K. J. V. Steenstrup, 1901. * Sachs, 1865. * Kissling, 1895 ; Sachs, 1865 ; and. in regard to the physiological action of light especially, Wiesner's many papers. LIGHT 17 light, soon ceases, and only some of the most lowly organized plants reach any considerable depth. The intensity of light plays a leading part in determining the distri- bution of species and abundance of individuals of a community, because the optimum is very different for different species. With inadequate illumination plants do not thrive ; they become etiolated and undergo degeneration or die. The distinction between light-plants and shade- plants (for instance, in the forest) is well known. Stebler and Volkart,^ in Switzerland, have made comparative measurements of the intensity of light beneath trees, and investigations concerning the light required by meadow-plants ; these they classify into hght-demanding, light- loving, indifferent, light-avoiding, and light-dreading species. In accord- ance therewith the distribution of species is different in different localities. In arctic countries the nature of the sky (the number of sunny days, the frequency of clouds and mists) certainly causes the contrast mentioned by many travellers between the rich flora and vegetation in the seclusion of the fjords and those of exposed coasts as well as of the islands of the region. - 6. The development of plants depends not only upon the intensity but also upon the duration of the light to which they are exposed. For instance, in Finland or the north of Norway barley ripens its grain in eighty-nine days from the day of sowing, but in Schonen (in Sweden, 55-7° N. ) it requires one hundred days, despite the higher temperature and the more intense light ; and the explanation of this must in part be that in the former places prolonged illumination promotes anabolism. In the north the periodic vital phenomena of plants set in much more rapidly in summer than in spring, because of the longer duration of the dayhght. Arnell states that, going northw^ard from Schonen, for each degree of latitude anthesis is later by 4-3 days in April, 2-3 days in May, 1-5 days in June, and 0-5 of a day in July. 7. Direct light promotes the production of leaves and flowers. The side of a tree facing the source of light often acquires foliage before the reverse side : Brazilian Ficus-trees may be seen to be in leaf on their north side, whilst still leafless on their south side ; ^ tufts of Silene acauhs in arctic countries on their south side may be decked with flowers, which also point towards the south, while they are devoid of flowers on their north side.'* 8. The vegetative shapes of plants are greatly influenced by the intensity and direction of the light. The effects of insu-fficiently intense light are revealed not only in the phenomena of etiolation, which are essentially pathological in nature, but also in connexion with healthy normal individuals. Of this, forest-trees furnish admirable examples. Light, in the first place, determines the shape of the individual tree. The duration of life of the branches depends partially upon the intensity of light. The shade cast by younger branches retards the assimilatory activity of the leaves on older branches, and thus renders impossible the normal development of buds and the ripening of wood. The branches ' Stebler unci Volkart, 1904. ' In regard to Spitzbergen and East Greenland respectively, consult Nathorst, 18S3, and Hartz, 1895. ' Warming, 1892. * Rosenvingc, 1889-90 ; Stefansson, 1894. WARMING C i8 OECOLOGICAL FACTORS AND THEIR ACTION sect, i die off, become brittle, and break by reason of their weight or of storms ; it is because of their suppression that the central parts of trees and shrubs are leafless and have so few twigs. A spruce standing out in the open is conical and bears branches from its summit to its base ; whereas one standing in a dense forest has only a small green crown, and outside this no branches, or only leafless dead ones, because its illumination is different. Dicotylous trees, such as the oak or beech, standing in the open, have a full ovoid or conical crown, but when growing in dense woods have a small crown with upwardly directed branches.^ Light plays an important part in the struggles between trees that are growing in company. Forest trees may be divided into — (a) Light-demanding trees, which demand much light and endure but little shade ; (&) Shade-enduring trees, which are content with less light and can endure deeper shade. The reasons for these distinctions must be sought for in the specific distinctions in the chlorophyll, rather than in any difference in the archi- tecture (structure of the shoot, phyllotaxy, and form of the leaf) of the species. Arranging our commonest forest-trees in accordance with their demands for light when individuals of the same age are competing with one another, we arrive at the following series, the order of which approxi- mately denotes decreasing requirements as regards light : — 1. Larch, birch, aspen, alder. 2. Scots pine, Weymouth pme, ash, oak, elm, sycomore. 3. Pinus Montana, Norway spruce, lime, hornbeam, beech, silver fir. It is worthy of note and biologically important that nearly all trees can endure deeper shade in early youth than they can later in life. It may be added that the power to endure shade also depends upon the fertihty of the soil. Distinction between Sun-plants and Shade-plants. Between heliophilous or photophilous plants, which prefer sunlight, and heliophohous or sciophilous plants, which prefer the shade, there are great differences in external form and internal structure. I. Intense light retards the growth of the shoot ; consequently, helio- phytes are compact and have short internodes, but sciophytes have elongated internodes ; species clothing the forest soil are mainly tall and long-stemmed. The leaves of heliophytes are often small, narrow, of linear or some similar form ; but those of sciophytes under the same conditions are large and broad, longer in proportion to the width,^ and thinner. The leaves of Maianthemum Bifolium in sunlight attain scarcely one-third of the size that they reach in the shade.^ The leaves of many species, especially of cultivated plants, are larger in northern lands than in lower latitudes ; this is possibly due to the poverty of the light of high latitudes in rays of short wave-length. In gardens on the west coast of Norway, for instance, the flowers of Tro- paeolum ma jus lurk almost hidden beneath the mass of large leaves.* * See Vaupell, 1863. ^ Warming, 1901. * Kissling, 1895. * Bonnier et Flahault, 1878 ; Schiibeler, 1886-8, and others. CHAP. V LIGHT 19 2. Intense light decomposes chlorophyll. A whole series of structural peculiarities have been interpreted as affording protection against too intense light/ and among these are the following : — 3. The leaves of heliophytcs arc often folded (grasses, palms, screw- pines), or wrinkled and bent (Myrtus bullata), while those of sciophytes are flat and smooth ; this feature is well seen in many plants growing in hot, dry places in the West Indies.'^ 4. The lie of the leaves is differ oil in heliophytes and sciophytes. Leaves of heliophytes are often directed sharply or even almost vertically upwards (Lactuca Scariola in sunny spots, and other ' compass-plants '),^ or hang vertically downwards, particularly when young (mango and other tropical plants) ; whereas the leaves of sciophytes are extended horizontally, as we may see in the case of dicotyledons in our beech- woods. The sun's rays strike the leaves of heliophytes at acute angles, and therefore lose in efficiency, but the weaker light in the forest strikes the leaves of sciophytes at right angles. Young leaves are directed vertically or obliquely. In dicotylous sciophytes, a leaf-mosaic ^ is often formed by the juxtaposition of large and small leaves in such a manner that the interstices are reduced to a minimum (Fagus, Trapa, Trientalis, Mercurialis, and many other forest-herbs). In plants with acicular and linear leaves, such as Juniperus and Calluna, a great difference exists between heliophytes and sciophytes ; the former have erect ad- pressed leaves, and the latter have spreading leaves ; the former assume a permanent profile-lie, and the latter display their full surface ; these orientations are necessarily assumed by the plants during the young and growing stage. 5. The photometric movements exhibited by the leaves of many plants as a consequence of change in illumination may be mentioned here ; to light that is intense (or rich in rays of short wave-length) leaves oppose their edges, to light that is weaker (or composed mainly of rays of long wave-length) they oppose their faces.^ 6. The histology of leaves produced in the sunlight and shade respec- tively is not less different. Heliophylls are often isolateral, namely, when they are erect and their two surfaces are consequently equally illuminated ; sciophylls are universally dorsiventral.® Heliophylls have a thick pali- sade tissue, which owes its thickness either to the length of the palisade cells, or to the presence of additional layers of them, or to both of these characters (stems with little or no foliage likewise have a thick palisade tissue extending completely round them) ; sciophylls have a thinner palisade tissue or none at all. Palisade cells are often directed obliquely in reference to the surface ; this appears to be associated with the direc- tion of the rays of incident light.' Spongy parenchyma is relatively more developed in sciophylls than in heliophylls. Heliophylls are thicker than sciophylls. Heliophylls have small intercellular spaces, sciophylls have large ones. Heliophylls respire and assimilate more rapidly than do sciophylls of the same species. ' Wiesncr, 18766. ' Johow, 1884. * Stahl, 1S81, 1883. * Kerner, 1887 ; Warming, igoi. '- See Section III, Chapter XXX. * Heinrichcr, 1884. ' Pick, 1 881; Johow, 1884; Heinrichcr, 1884; Habcrlandt, 1886; Warming, 1897. C 2 20 OECOLOGICAL FACTORS AND THEIR ACTION sect, i The epidermis of the heliophyll is thick ; usually contains no chloro- phyll, at least on its upper face ; ^ sometimes it is converted by periclinal divisions into an aqueous tissue several layers in thickness (e. g. in Ficus elastica and some other tropical plants) ; its cuticle or cuticular layers are thick. The epidermis of the sciophyll is thin, one cell in thickness, sometimes contains chlorophyll, and its cuticle is thin. The heliophyll is often very glossy and a good reflector of light, as is demonstrated by many tropical examples ; ^ the sciophyll is dull in surface and, when subjected to dry air, fades much more easily than the heliophyll. The epidermal cells of heliophylls have less sinuous lateral walls than those of sciophylls. Stomata of the dorsi ventral helio- phyll are confined to the lower face, or are more numerous there than on the upper face (except in some alpine plants), and are often sunk below the level of the surface ; those of the sciophyll are on both faces, but perhaps on the whole more numerous on the lower face, and are inserted at or above the level of the surface. Many tropical sciophytes have velvety leaves that are beset with refractive papillae, which serve to collect the obliquely incident rays of light .^ 7. Lignified parts are more general in heliophytes than in sciophytes, for example, the production of thorns is more frequent. Heliophylls are often stiff and coriaceous (sclerophyllous plants), partly from lignification, partly because of their thickness, and partly because of the nature of the epidermis ; sciophylls are thin and, if large, flaccid (many herbs in European forests, such as species of Corydalis and Circaea, Lappa nemo- rosa, Lactuca muralis, Oxalis Acetosella, many ferns ; and in the tropics, Hymenophyllaceae, mosses, and others). 8. In the production of hairs variety is exhibited. Heliophylls often have a dense covering of hairs, a grey tomentum, a silvery coating, or are hairy in divers ways, especially on the lower face (e. g. many plants on rocks, heaths, and steppes) ; sciophylls are universally much less hairy, sometimes quite glabrous. g. In the sensitiveness of chlorophyll to light, great differences probably exist, for presumably the chlorophyll of sciophylls is more sensitive than is that of heliophylls, and is consequently better able to utilize weaker light. This suggestion harmonizes well with the fact that an alcoholic solution of the chlorophyll of sciophylls is very easily decolorized in the presence of light, 10. Light influences the coloration of plants by its action in regard to the production not only of chlorophyll but also of red cell-sap (antho- cyan or erythrophyll). This pigment occurs especially in young parts of plants (in young shoots and seedlings), in autumn leaves, in alpine * and arctic^ plants, in tropical sciophytes.*^ Engelmann has demonstrated that it absorbs the rays of light complementary to those absorbed by chlorophyll ; red leaves exposed to radiant heat acquire a higher tem- perature than green leaves do. The red pigment provides the means of storing up heat, which is available in connexion with metabolism when the temperature of the air is relatively low.' It may also be men- tioned that the colours of leaves, flowers, and fruits become deeper in ^ Stohr, 1879. "^ Volkens, 1890. ' Stahl, 1896; Haberlandt, 1905. * Kemer, 1887. ' Th. Wulff, 1902. " Stahl, 1896. ^ Stahl, 1896; Buscalioni et PoUacci, 1903; Jonsson, 1903; Overton, 1899. CHAP. V LIGHT 21 high latitudes ; ^ while a whole series of heHophytcs (Myrtus bullata, Perilla nankinensis, Prunus Pissardi and others) growing naturally possess the dark-red to blackish-red tints characteristic of the familiar copper- varieties of the beech, hazel, and other trees. The features described above will be treated in greater detail in subsequent chapters that deal with xerophytes. That light is of great significance in influencing the external and internal construction of plants is beyond doubt. This follows, not only from what has been already said, but also from the fact that many, perhaps most, plants can adjust their anatomical structure, especially of their leaves (' plastic leaves '), according to the intensity of the light. A beech-leaf exposed to sunlight is structurally different from a beech-leaf in the shade.'- The arrangement and movements of chloroplasts in cells, and therefore the tint of foliage, depend upon the light ; ^ stronger light causes the leaf to be paler in tint, weaker light causes it to become darker green. As to the exact method in which light acts physiologically our notions are very hazy. Some (Stahl, Pick, Mer, and Dufour) opine that it is light itself which determines, according to its intensity, the above- mentioned structural differences in the chlorenchyma ; but these investi- gators fail to explain how light acts. Others (Areschoug, Vesque and Viet, Kohl, and Lesage) suggest that the cause may be increased transpira- tion due to increased light. Still others (Wagner and Mer) are inclined to lay chief stress upon the strong assimilation following upon more intense light. The action of light of different composition upon the activity of proto- plasm and the arrangement of chloroplasts is treated in papers by Sachs and Kissling.* It is scarcely open to doubt that the structural differences between hcliophytes and sciophytes must be regarded as affording an example of self-regulation (direct adaptation ^) on the part of the plant. , We see this taking place before our eyes in plastic plants which adjust their structure to light ; opposed to this are other cases in which the structure probably has been modified during the course of phyletic development and become fixed by heredity in successive generations. Among the uses of the various structural features are the following : Protection of the chlorophyll from decomposition by intense light,® protection of the protoplasm itself (intense light can injure protoplasm, as is demon- strated by its destructive action on bacteria, its use as a means of disin- fection, and so forth), protection against excessive transpiration, and regulation of assimilation. When we consider that the volume of the palisade tissue is increased not only by more intense illumination, but also, as research has proved, by stronger transpiration, as well as by various factors (salts in the substratum, injury to the roots, and so forth) that influence the absorption of water from the soil and consequently affect transpiration ; and when we further consider that the palisade tissue increases in all stations where great atmospheric aridity prevails, then we shall be inclined to regard regulation of transpiration as the most ' Bonnier et Flahault, 1894; Schiibclcr, 1886. ' Stahl, 1880, 1883 ; Hesselman, 1904 ; Woodhcad, 1906. ^ Stahl, 1880, and others. * Kissling, 1895. * See Section XVII, Chapter C. • Wiesner, 1876. 22 OECOLOGICAL FACTORS AND THEIR ACTION sect, i essential reason for the structural differences in question. Transpiration increases with an increase of insolation. Thus hght is one of the most important factors influencing transpiration, and the plant regulates the latter according to the intensity of the former. But for further infor- mation in this matter we must look to the future.^ CHAPTER VI. HEAT Heat is to a far higher degree than is light an oecological and geo- graphical factor, not only in general, but also in detail. Each of the various vital phenomena of plant-life takes place only within definite (minimum and maximum) limits of temperature, and most actively at a certain (optimum) temperature ; these temperatures may even differ in respect to the different functions of one species. Heat is of import in the manufacture of chlorophyll, in the processes of assimilation, of respiration, and of transpiration, the functional activity of the root, germination, the production of foliage and blossom, growth, movement, and so forth. It is therefore clear that conditions as regards heat determine the boundaries of the distribution of species on the Earth. As the lower and upper critical temperatures vary greatly in different species, we can only say generally that the lower critical point (' specific zero' of the species) descends to o°C., or slightly lower in certain rare cases, which include many arctic and alpine species, mostly of low organ- ization. Algae in the Arctic Ocean, off the coast of Spitzbergen, at about 80° N., grow and fructify vigorously during winter, in darkness at a tem- perature of— 1-8° C. to o°C. ; of twenty-seven species Kjellman observed twenty-two with reproductive organs. Usually the functions do not com- mence activity until a temperature several degrees above 0° C. is reached, in some cases (especially in tropical plants) not before 10° C. or 15° C. The upper critical temperature does not attain 50° C, and generally not even 45° C. The different organs of a plant usually have different capabilities of enduring extreme temperatures. A species may therefore thrive in a country and produce blossom, but its seeds may not ripen, or if they do so may not be able to resist frost, or the seedlings may suffer from the cold. Such a species would be dependent upon vegetative propagation for its permanent existence in such a country. Heat has also an indirect significance, in that the relative humidity of the air and transpiration depend upon temperature. Temperatures outside those that are critical to species are not necessarily equally lethal ; in this respect there is a certain amount of latitude, which is greatest below the specific zero — that is to say, plants can, without ^ For further information reference should be made to the works of Areschoug, Stahl, Pick, Dufour, Haberlandt, Heinricher, Vesque, Viet, Mer, Lothelier, Johow, A. Nilsson, Eberdt, Schimper, Grabner, Wiesner, Hesselman, Woodhead, Stebler, and Volkart. In regard to photometry the works by Wiesner, K. J. V. Steenstrup (1901), and Hesselman (1904) should be consulted. CHAP. VI HEAT 23 suffering death, be exposed to low temperatures that are more degrees below the minimum than high, actually fatal, temperatures are above the maximum. (Possibly the sole exception is provided by many bacteria.) Moreover, temperatures below the minimum and above the maximum are not always devoid of significance to plant-life, even if they be not of direct utility. On the Earth there is scarcely a spot from which plant-life is absolutely excluded by reason of the thermal conditions ; for even in places where the temperature remains for months below the minima of species, or above the maxima (e. g. in parts of Africa), plants thrive at certain seasons of the year. Yet it may be necessary for plants to guard against extreme temperatures and against that which these involve, change of temperature. To the latter many plants (e. g. palms) are much more sensitive than to low temperatures. Sudden thawing is injurious to many plants because the tissue is ruptured ; forests often suffer from night-frosts on the east side, on eastern slopes, and on similar spots where the sun's rays strike them early in the day. The following means are adopted as affording protection against extreme temperatures, and particularly against such as are too low'^ : — 1. The cell-contents of some plants have certain (hitherto unexplained) characters in virtue of which they can withstand extreme temperatures for a long time : in phyto-geography, extremes of cold almost alone have to be considered. These resistant characters may be due to the proto- plasm itself, or to the admixture of sugar, oils, or resinous bodies, with the protoplasm or cell-sap. Protection of this kind is apparently exempU- fied by the snow-alga, Sphaerella nivalis, whose thin-walled isolated cells can endure the cold of arctic snow-fields and ice-fields.^ Likewise Coch- learia fenestrata is evidently protected ; for on the north coast of Siberia, in the winter of 1878-9, this plant endured unsheltered a temperature remaining lower than — 46° C, and in the following spring it continued its flowering which had been interrupted by winter.^ In a number of trees at autumn time the starch changes to fat ; * this is probably of use, in that fatty oil in the form of emulsion prevents sub-cooling and increases the power of resistance to frost. Fat-storing trees (birch, conifers) are precisely the ones that grow in the coldest lands. The change that takes place during winter of soHd reserve substances into dissolved substances, namely sugar, also prevents the under-cooling of plant-tissue and death of the plant. -^ 2. Amount of water. The water contained in plant-parts plays the leading role in regard to their power of enduring extreme temperatures ; the power of endurance is inversely proportional to the amount of con- tained water. Consequently the young shoots of North-European trees often suffer from late frosts, while the older shoots are not damaged ; also seeds, which are always poor in water — for instance those of the wheat — can hibernate uninjured for many years in arctic countries. The smallness of the amount of water possibly also explains the perennation ' For more recent work on the endurance of plants in winter, and on the eflfcct of freezing, see Mez (1904-5), and Lidforss (1907). - Wittrock, 1883 ; Lagerheim, 1892. ■" Kjellman, 1884. * A. Fischer, 1891 ; O. G. Petersen, 1896. ^ Mez, loc. cit. ; Lidforss, loc. cit 24 OECOLOGICAL FACTORS AND THEIR ACTION sect, i of many mosses, lichens, and other lowly organized plants. Lignified parts endure cold better than do herbaceous parts,^ consequently many arctic and alpine species are woody (dwarf-shrubs). Southern shrubs cultivated in North-European gardens, and trees and shrubs on the Faroes,^ often do not receive sufficient heat to enable them to ripen their wood ; the ends of their branches are killed by the cold in winter ; and they dwindle from shrubs to sub-shrubs. In places with a longer vege- tative season the same species (Broussonetia, Tamarix, and others in Hungary) endure, despite equally severe winter-cold. The forests in arctic Siberia withstand temperatures descending to -70° C. (in Ver- choyansk during January the mean temperature is — 5i-2°C., the mean minimum — 63-9°C.-, and the lowest recorded temperature — 69-8°C.) When a plant is frozen to death this is usually associated with a formation of ice, and thus with a drying up of the cell-sap. 3. Bad conductors of heat, for example bud-scales or hairs, often envelop the parts requiring protection ; their cells are mostly filled with air or have between them air-containing intercellular spaces, and contain as little water as possible. Very many protective devices are displayed by young shoots when in foliage.^ Many arctic and alpine plants have a grey cottony or white woolly coat of hairs (Leontopodium alpinum, which is ' edelweiss ') ; ' Frailejon,' which are Compositae occurring on the Paramos of South America and belong to the genera Culcitium and Espeletia,^ the shoots of these plants are encased by old faded leaves which hang on and envelop them,^ just as in Central Europe tender garden plants in autumn are artificially clothed with straw, hay, leaves, and the like. It must be pointed out that though these devices hardly exclude extreme cold (for this may reach the interior of the plant), yet they ward off three contingencies — rapid change of temperature, rapid thawing, and precarious transpiration. Experience and research have shown that, though cold itself is sometimes responsible for the death of plant-members (potatoes, petals, tropical plants in high stations of Brazil, and so forth) which have been fatally frozen, yet the act of thawing is the critical matter in the case of many plants which can be frozen solid without injury. Hence thawing ought to take place slowly, and in this direction assistance is rendered by the above-mentioned structural features, which are therefore particularly met with in sub-glacial com- munities. Repeated sudden freezing and thawing cannot be endured by the majority of the plants of Central Europe (e.g. beech, oak, and others). In contrast with this, Kihlman ® asserts that ' the extraordinary power of enduring great and rapid oscillations of temperature, and even of withstanding the recurrence of freezing-point several times within twenty- four hours, is an outstanding peculiarity ' of the tundra-vegetation of Lapland. Submerged aquatic plants are well protected by the surrounding water ; many of them sink to the bottom or possess buds that become detached and sink in like manner. The devices mentioned also serve as a protection against rapid transpi- ration— against the desiccating action of dry, cold winds, which are ' Mohl. 1848. '^ Borgesen, 1905. ^ See Griiss, 1892. '' Gobel, 1892. ^ See p. 75 and Chap. LXVI. ' Kihlman, 1890. CHAP. VI HEAT 25 dangerous to plant-life when the soil is cold and physiologically dry and the activity of the roots is arrested. To species both in respect of their conditions of life and their distribution it is by no means without import which of the efftcient temperatures (those between the maximum and minimum) prevail. The life of the individual is influenced not only by the height of the temperatures to which it is exposed, but also by the amount of efficient heat received or the duration of efficient temperatures. Annual mean temperatures are devoid of significance to plant-life. Only the season during which useful temperatures prevail is of import.^ Thus in Northern Siberia, where the mean annual temperature is below —15° C, forests occur, yet on Kerguelen Island, where the mean temperature €ven of the coldest month is above freezing-point, the vegetation is arctic. In most regions of the Earth change of season causes plant-life to undergo a period of rest. The cause in north-temperate climates is change of temperature, and particularly lowness of temperature ; in the tropics it is lack of water. The time during which efficient temperatures are available may be so short, sometimes only a few weeks in length, that many species are excluded because they cannot obtain sufficient heat. This certainly explains why annual species are so rare in arctic latitudes and at alpine altitudes ; they require for the completion of their life-cycle more time than is available. Perennial herbs in arctic countries and on high mountains display much variety in their adaptation to climate. For example, they may have perennating foliage-leaves, which sometimes contain reserve food, and with the aid of these they can utilize each passing moment that is favourable to assimilation, and 'lose no fraction of the vegetative season in producing new assimilating organs.- They display another adaptation in that they initiate their flowers in the year before these open, so that they can burst into blossom at the immediate commencement of the succeeding spring, and thus have as long as possible a period of blossoming and fertilization, and can utilize the warmest season for maturing the seed.^ The temperature and length of the vegetative season affect the physiognomy of the individual plant and of the whole vegetation. At one extreme are equatorial countries, where resting seasons are all but imperceptible, and where high temperature is linked with humidity ; here is developed the evergreen tropical plant-life whose luxuriantly growing species clothe the soil with the densest of vegetation. At the other extreme are arctic countries and regions on high mountains, where Nature doles out her gifts with niggardly hand perhaps for only three months in the year ; here the plants developed are, in places, not sufficient to cover the soil ; here, too, dwarf-forms present themselves, because, among other reasons, the vegetative season is too short and the efficient heat too feeble. With increasing heat the rate of growth accelerates until an optimum is attained. But in the two last-named sites, low vegetation, condensed shoots, rosette-plants, small leaves, and caespitose habit, inevitably result. In the tropics, dwarfed growth may result when a high temperature is combined with drought. * Koppen, 1884. ' Kcrner, 1896. * See Warming, 19080, 26 OECOLOGICAL FACTORS AND THEIR ACTION sect, i Phenologists have frequently endeavoured to estimate the accumu- lated temperature that species presumably require for their various func- tions. This reveals its existence most distinctly in spring, at which time the opening of flowers and leaves is clearly dependent upon the conditions prevailing in regard to heat and takes place in one year at one time, in another year at another time, and in one place earlier than in another. The number of days of vegetative activity commencing from a certain date having been estimated, and the temperatures prevailing at various places having been ascertained, endeavours have been made to explain upon this basis the differences in development and the facts of distribu- tion ; but in details there has been great diversity of treatment. Some investigators have sought to estimate the accumulated temperature by the addition of the daily mean temperatures ; others have multiplied the mean temperature of a certain period (particularly the period of growth) by the number of days ; others, again, have relied upon the square of the mean temperature or of the number of days ; still others opined that the daily maxima above o° C, registered by a thermometer exposed to the sun (insolation-maxima), should be added together. These investigations absolutely demand the support of strictly scientific experimental determinations of the temperatures important in relation to the vital phenomena of the different species. But the results of these estimations will not suffice to explain the extremely difficult and complex question of the relations between heat and the distribution of species, or between phenological phenomena, because other conditions, such as light, the temperature of the soil, the after-effects of the preceding vegetative season, are perhaps capable of replacing higher temperature. A general source of error is that shade-temperatures, and not temperatures resulting from insolation, are used in estimations; but even the sum of the insolation temperatures would hardly give a correct account of the temperatures prevailing during a definite period. In the following morphological features heat indubitably plays a part : I. Many sub-glacial plants, particularly woody plants (Salix, Betula, Juniperus, and others) assume the espalier-shape : that is to say, their stems lie on the ground, are pressed against it, and concealed more or less between other plants (mosses and lichens), stones, and such like ; only their tips are directed upwards, sometimes almost at right angles, and rise above the ground only a few centimetres. By this mode of growth the plants doubtless receive a greater amount of heat than they would were they erect ; but it is a question if it be not rather evaporation resulting from the dry cold winds that induces this change of shape. The same form of growth is exhibited by many littoral plants (Atriplex, Suaeda, Salicornia, Matricaria inodora, in Northern Europe ; Frankenia pulverulenta, on the Mediterranean shores) ; it is not the lateral shoots alone that lie prostrate and radiate in all directions ; but the main shoot itself bends down, sometimes almost at right angles, prone on the soil.^ Again this habit reveals itself on the desert and on sandy soil that is strongly heated by the sun (e.g. Aizoon canariense, Cotula cinerea, and Fagonia cretica, in Africa ; ^ Artemisia campestris and Herniaria glabra in Northern Europe). In the hot dry chmate of the lowland of Madeira ^ Warming, 1906. ^ Volkens, 1887. CHAP. VI HEAT 27 prostrate forms are rare, and such forms are protected by succulence, strong coatings of hair, and the hke> These famihar forms of growth have beyond doubt a common cause. With the popular explanation that the plant wishes to ' bend before the wind ', science cannot content itself. Probably the cause must be sought in the difference of temperature of the air and soil at the time when the shoots are developing. There are often to be met, growing side by side, erect and prostrate individuals, e.g. of Atriplex, Salicornia, Suaeda, and others on the northern coasts of Europe ; this fact denotes that the decisive cause is no general factor prevailing at all seasons in a definite place. Neither the wind nor its direction can be responsible, since indi- viduals of one species growing on the same shore may have their main shoots pointing in different directions, as may easily be observed on the North-European coasts. The explanation must, apart from individual pecuHarities, probably be sought in the different degree of heating of the plants during their development above ground, and the consequent exe- cution of thermotropic movements. Krasan ^ suggests that plants on a warm soil, particularly in a climate with a warm atmosphere, acquire vigorous erect shoots, but on a cold soil, and especially with an alpine climate, prostrate ones. That psychrocliny in reality is responsible for the espalier-like prostrate growth of plants in various cases, is confirmed by the admirable investigations of Vochting and Lidforss.^ Henslow ^ also expresses the view that thermotropism plays a part. On roads and soil frequently trodden down, prostrate forms, such as Polygonum aviculare, are frequently found. Here the cause is perhaps mostly strong vegetative heliotropism. 2. Rosette-plants. Many herbs have their basal leaves more or less liorizontally expanded in the form of a rosette ; even when they have elongated rhizomes or subterranean stolons their shoots, upon reaching the surface, become condensed. What factors are responsible for this is scarcely known ; but presumably heat plays some part. Bonnier ^ has experimentally demonstrated that great changes in temperature are among the most eflicient factors in determining the character of alpine plants ; those plants that were exposed to intense cooling action at night acquired condensed stems, smaller, thicker, and harder leaves, and they blossomed earher. Conditions of illumination exerted less influence. Rosette-herbs occur in great numbers in temperate countries and are particularly characteristic of sunny meadows which are covered by low vegetation ; they are found in great numbers in arctic countries and on high alpine situations on open grassy or rocky expanses, yet they occur in still greater quantity on the meadows of lowlands, but rarely in forests.'' In hot dry climates rosette-plants are less common, some occurring only on moist soil," some others being confined to hot dry spots and, in such cases, characterized by succulence (e.g. with succulent leaves, Echeveria, Aizoon, Agave, and Bromeliaceae) or by other potent means of protection against drought. 3. Tufted (caespitose) growth and scrub-plants are common in climates ' Vahl, 19046. ' Krasan, 1^82, 1884. "• Vochting, 1898 ; Lidforss, 1902, 1906 ; compare C. Schroter, 1904-8. * Henslow, 1894. " Bonnier, 1898. * Warming, 1901. ' Meigen, 1893, 1^94 5 Vahl, loc. oil. 28 OECOLOGICAL FACTORS AND THEIR ACTION sect, i with extreme temperatures, and in arctic and alpine situations are due to cold, but in deserts are called forth by great dryness, and evaporation arising from intense heat. The shoots become short and curved, in the former places because the warmth requisite for growth is wanting, and in the latter site because heat robs them of moisture.^ From what has been said it follows that various structural features apparently must be interpreted as due to the action of heat upon plants. Attention will be directed later in this work to the great importance of the temperature of the air in relation to atmospheric humidity and to transpiration ; these exercise great influence upon plant-form. The distribution and habitats of species in their main features (vege- tative zones of the earth, vegetative regions of mountains) are determined by heat. In terrestrial species of wide geographical distribution the difference between the maximum and minimum temperatures is as a rule especially wide (this is not true of aquatic species). But, above all, heat exerts an influence upon the habitat, economy, and struggles of plant- communities. It partly determines the distinction between the climates and vegetation of the coasts and interiors of countries ; this is most clearly shown in arctic countries, where the feeble vegetation of the cold coasts contrasts with that of the interior, which is relatively rich in species and individuals and displays more vigorous individuals.- Arctic countries also exhibit great contrasts between the feebler vegetation of lowlands and the richer, more luxuriant vegetation of sunny mountain-slopes ; for the angle of incidence of the sun's rays is far more acute in the plains than in the declivities. If steep mountains should occur near the Poles they certainly must have a relatively rich vegetation. The angle of inclination and the exposition of mountain-slopes obviously influence the result, since the soil and consequently the air will be heated to different degrees when these differ. But as these and other conditions depend intimately on the temperature of the soil, their treatment is rele- gated to Chapter XIII. That the contour of the earth's surface may influence details in the geographical distribution of plants may often be seen in situations where, on calm frosty nights, cold air remains suspended over depressions and valleys and causes the plants to be frost-bitten. CHAPTER VII. ATMOSPHERIC HUMIDITY AND PRECIPITATIONS The ©ecological importance of water to the plant is fundamental, and almost surpasses that of light or heat. Without water, vital activity is possible neither to plant nor animal. Its significance to the plant in a con- dition of full vital activity is as follows : — 1. As imbibition-water it is necessarily present in all protoplasm and all cell- walls. 2. As cell-sap it occurs in vacuoles and plays a part in turgidity and normal growth. 3. As a nutritive substance it is directly assimilated. ^ Further details are given in Chapter VIII, deaUng with the effects of wind. ^ In regard to the influence of light see Chap. V. CHAP. VII ATMOSPHERIC HUMIDITY 29 4. All absorption of nutriment from the soil, all osmosis, all transference of substance, take place solely through the aid of water. The germination of seeds and spores and the sprouting of sclerotia demand a supply of water for their initiation. The mineral nutriment of a plant must be present in a dissolved condition. 5. The assimilation of carbon dioxide depends upon water ; it is retarded in a plant that is not fully turgescent, because for one reason the stomata are closed, and it ceases in a fading plant. 6. Respiration ceases when the amount of water in a plant sinks below a certain limit. 7. The open or closed condition of the stomata, and consequently transpiration, or the evaporation of water from the plant, depends upon moisture. A moist condition of the leaves increases transpiration. 8. All movements, whether due to swelling or to irritability, take place only through the agency of water. g. The amount of water in a plant is the factor determining life or death when the temperature lies outside the critical ones. Dry parts of plants are the most resistant.^ It is therein not remarkable that death may ensue from lack of water or from desiccation ; yet many plants or their parts can withstand long and severe drought. The limits of desiccation vary greatly ; only very few, mostly lowly-organized plants such as lichens, mosses, Selaginella lepidophylla and allies, appear capable of withstanding almost complete desiccation. It is likewise not surprising that no other influence impresses its mark to such a degree upon the internal and external structures of the plant as does the amount of water present in the air and soil (or medium), and that no other influence calls forth such great and striking differences in the vegetation as do differences in the supply of water. It has been demonstrated by Hellriegel and others that a larger supply of water yields a richer crop (of leaves, straw, fruit, roots) ; if the plant be supphed with but little water, dwarf growth (nanism) ensues.^ But it may be noted that the vigour of an ordinary terrestrial plant is not pro- portional to the water supplied up to an indefinite amount, for there is an optimum which varies according to the nature, aeration, and other characters of the soil. For the purpose of ridding itself of any excess of water absorbed the plant exhibits certain devices (water-pores, gutta- tion, internal bleeding) ; but there is a limit to the amount of water that can be endured by a plant, for instance, plants that prefer drought mostly perish soon when supplied with large quantities of water (e.g. heath-plants). Water is conveyed to the plants by two channels — the air and the soil (or the water in case of aquatic plants). The power of absorbing and retaining water derived from the soil will be discussed in Chapter XII ; here moisture in the air and atmospheric precipitations alone will be considered. Atmospheric humidity. In the atmosphere there is always some water present in an in- visible gaseous state, but the amount of this varies greatly : it rises and ' See p. 23. * See Kraus, 1906a. 30 OECOLOGICAL FACTORS AND THEIR ACTION sect, i falls with the temperature of the air — the amount of the water that air is capable of retaining in a gaseous condition varying with the temperature. Cold air does not take up so much water as does hot air before becoming saturated ; consequently great fluctuations occur at different times of the day and year. It is not the absolute humidity of the atmosphere that is of greatest moment to plant-life, but the saturation-deficit. The satura- tion-deficit is the difference between the maximum and the observed vapour-pressure at a given temperature, and therefore indicates the additional amount of water which the atmosphere is capable of taking up, or the amount that it requires to become saturated. Evaporation from the surface of water at the same temperature as the air is nearly proportional to the saturation-deficit. Consequently, the saturation- deficit is one of the indices of the evaporating action of a climate ^ (if temperature be also taken into consideration). The saturation-deficit is, as a rule, smallest during the night and greatest during the day time. But among mountains this condition is often reversed owing to the daily alternation of winds blowing up and down the valleys. Transpiration is largely dependent upon the saturation-deficit, yet even in a very moist atmosphere the transpiration may be very considerable, because the stomata remain open and the plant is heated by rays of light. The amount of evaporation also depends upon several other con- ditions, including the temperature, the extent and other qualities of the evaporating surface, so that plants very naturally have produced many morphological and anatomical adaptations, enabling them to flourish under various conditions of humidity.^ The plant strives in some cases to depress transpiration to a certain low limit, in other cases to promote it ; certain plants, for example many sciophytes (mosses, ferns, particularly Hymenophyllaceae, and others) clothing forest-soil, can assimilate only in very moist air ; others are adapted to very dry air. The structural features guarding against dry air and depressing transpiration are, in part, identical with those guarding against intense light.^ It must be noted that it is very difficult to decide which features are to be correlated with atmospheric humidity, and which with other factors co-operating at the same time. The peculiarities of sciophytes mentioned on pages 19-20 can scarcely be attributed solely to the greater atmospheric humidity that usually prevails in the shade when compared with the open, but must be partially caused by weaker light, just as the peculiarities of heliophytes are caused by not only intense heat and intense transpiration but also by intense light. Sorauer, Mer, Vesque and Viet, Lotheher, and others have found that the effects of moist air are like those of weak illumination. Plants become more elongated, long-jointed, thinner, paler ; their leaves are smaller and thinner, more transparent, and have their dorsiventral anatomy obliterated because of the absence or feeble development of palisade tissue ; the vascular bundles become weaker, the intercellular spaces larger, and the mechanical tissue weaker or even suppressed. It is the difference in transpiration in the one case as in the other that is respon- sible for the difference in structure. Mosses and lichens presumably can absorb aqueous vapour from the atmosphere ; but it is uncertain to what extent spermophytous plants ^ See Hann, 1901. ^ See Section III. * See Chap. V. CHAP. VII ATMOSPHERIC PRECIPITATIONS 31 can accomplish this, for instance, by employing hairs or the velamen of aerial roots to condense the vapour. Possibly the cases in which such absorption has been assumed may be due to the deposit of liquid water taking place in or on plant-parts as a consequence of change in tempera- ture. The fact that plants drooping on a warm day become turgescent at night, does not imply that water-vapour has been condensed from the damper night air, but is certainly to be ascribed to the circumstances that transpiration is decreased owing to the smaller saturation-deficit, and that the roots, which may have been continuously conveying water to the plant, supply at night more water than is transpired. Certain desert-plants excrete hygroscopic salts which at night-time abstract water from the moister air ; but it is not certain that this water, which moistens the surface of the plant, is absorbed and utihzed by the cells.^ Atmospheric precipitations. If from any cause the air be cooled to dew-point, so that it cannot hold in a vaporous state the water which it contains, the water is deposited in one of the three known forms of atmospheric precipitation, mist (clouds), rain (snow), or dew (hoar-frost). Atmospheric precipitations in part are absorbed by the soil and thus become a source of profit to the plant,^ and in part are retained by epigeous portions of the plant,^ with which they come into direct contact and which in certain cases seem to be adapted to absorb them. Many plants (epiphytes and hthophytes) have no source of water other than direct atmospheric precipitations. Adaptation for the absorption of atmospheric precipitations} — In order that a plant may absorb atmospheric precipitations it is necessary that the cell-wall shall be permeable, the superficial cells shall contain osmo- tically active substances, and that the water shall not flow off the surface too rapidly. There are plants such as lichens, mosses, and certain algae, which can absorb liquid precipitations easily and rapidly over the whole surface, and thus become turgescent ; these plants also endure extreme desiccation. Other plants have on the surface definite parts which can be wetted and absorb water, but have other parts which do not permit this, or, at all events, can be wetted with difficulty (owing to thick cuticle, coating of wax, and the like). For the purpose of absorbing water from atmospheric precipitations a number of plants have special organs (aerial roots with peculiar absorbing tissue ; spongy old plant-remains that greedily suck in water ; hairs, like those in Bromehaceae, capable of taking up water ; characteristic leaf-cells with perforated walls, and so forth. ^ But it must be assumed that water is absorbed by epigeous organs, as a rule, only when these are in a half-faded condition, and when the root can provide no water and the plant contains no reserve-supply ; absorption by the root is a matter of necessity to the normal land-plant.^ ' Volkens, 1887; Marloth, 1887 ; J. Schmidt, 1904; Massart, 1898 a. ' See Chapter XII. ' Sec Burgcrstein, \(:)Oa. * See subsequent Sections, especially those dealing with xcroph>-tes (Chap. XXXI, ) ' See subsequent Sections, especially those dealing \\'ith xerophytcs. " Bohm, 1863; Detmer, 1877; Tschaplowitz, 1892; Kny, 1895; Willc, 18S7; see Burgcrstein, 1904. 32 OECOLOGICAL FACTORS AND THEIR ACTION sect, i The deposition of dew is of very great importance in tracts where there is but little rain ; many, especially subtropical tracts, would be almost devoid of vegetation were it not for the strong deposition of dew in the dry season. The deposition of dew is much greater in lower than in higher latitudes. It plays a remarkable part in plant-life, for instance, in the Egypto-Arabian desert ; ^ it must be the dew that in many places evokes in spring-time the phenomena of plant-life, which take place despite the fact that no rain has fallen for months.^ According to Mez,^ some epiphytic Bromeliaceae, Tillandsia usneoides for example, are adapted to take in dew especially by the aid of their loose, chaff-like, scaly hairs ; when the dew-absorbing leaves have an aqueous tissue of considerable extent, this is situated on the lower side of the leaf, but is on the upper side in species adapted for the absorption of rain. More- over, in European heath-moors dew is of the greatest importance to plant- hfe, and especially to bog-mosses, indeed it is the solitary source of water during the season of scanty rain. In temperate countries the deposition of dew may be very considerable, but it is of significance not so much as a source of water-supply as an influence depressing transpiration. It must be assumed that everywhere plants are adapted to the given mean supply of water. But as regards this, great specific differences exist. Wiesner's* researches have shown that many terrestrial plants are adapted to a definite average amount of rain, which in general varies with the species. He discovered the existence of two extreme kinds of plants, which he termed respectively omhrophilous (rain-loving) and onibrofhohous (rain-hating) plants, according to their power of enduring without injury the action of rain for a long period (often several months) or only for a short one. Xerophytes are mostly ombrophobous ; meso- phytes are ombrophilous or ombrophobous. Ombrophoby is usually associated with unwettability of the leaf-surface, ombrophily with wetta- bility. Many features have been regarded as adaptations for the removal of rain. Jungner and Stahl ^ have demonstrated in plants from rainy climates several characteristic structural features which serve to conduct rain rapidly from the leaves, so that transpiration may not be hindered by the blocking of the stomata, the plant may not be overloaded, fungal spores may be washed off, and so forth. Subserving this purpose are drip-tips, which are abnormally long, sudden, apical attenuations especially possessed by entire leaves of tropical plants, such as Ficus religiosa, Theobroma Cacao, and species of Dioscorea ; such tips facilitate the rapid removal of rain-water from the leaf. Whether certain other features to which Lundstrom ^ has directed attention subserve the same purpose is perhaps dubious ; lines of hairs, for example in Stellaria media and Veronica Chamaedrys, have been interpreted as affording means of carrying away water, so likewise have furrowed nerves and petioles of Lamium album, Humulus Lupulus, and Ai uncus Silvester ' ; and likewise velvety leaves in the tropical forest.^ It has been believed that falling rain, and particularly violent torrents ^ Volkens, 1887. ' Warming, 1892. ^ Mez, 1904. * Wiesner, 1894, 1897. ^ Jungner, 1891 ; Stahl, 1893, 1896. ° Lundstrom, 1884. ' Stahl, 1893. * See Section XVI, p. 346. CHAP. VII ATMOSPHERIC PRECIPITATIONS 33 of rain descending during storms, can mechanically damage parts of plants, and especially young delicate parts. The danger of injury arising from falling rain has certainly been greatly over-estimated. According to Wiesner ^ the weight of a drop of artificially produced rain is 62 gramme, but that of the largest rain-drops observed was only -16 gramme. The velocity of descent of rain is small and approximately constant. The maximum kinetic energy of a falling rain-drop is estimated by Wiesner as •0004 kilogramme-metre. As a means of protection in this regard, the following devices have been supposed to serve : — 1. The leaves of many, especially tropical, plants are directed upwards or downwards, so that the rain strikes them at acute angles and thus acts less violently ' ; in particular, young parts, either individual leaves or whole twigs, are pendulous and do not erect themselves until they have acquired a firmer texture (many tropical plants, Picea, and others). 2. Foldings, or corrugations of the leaf-blades, may operate in like manner. 3. Other plants having compound leaves execute paratonic movements when the sky darkens, before the rain itself descends ; consequently the rain impinges upon the leaflets at more acute angles. 4. Finely compound leaves of many tropical trees expose, as a whole, a less easily assailable blade than do broad and undivided leaves. 5. The possession by the leaves of most plants of a certain amount of free mobihty, due to their stalks or other causes, is probably the very best defence against the impact of falling rain-drops. Nothing beyond a shaking of the foliage or branches as a direct mechanical effect can be assumed to take place.' Hail can be very injurious to plants ; but there can scarcely be said to be adaptations protective against the damages threatened by hail- storms, though the contrary opinion has been expressed. Mist (clouds) absorbs light and thus can obstruct the assimilation of carbon dioxide.* It also retards the heating of the soil. Against it there can scarcely be said to exist any protection. Consequently the coasts of Spitzbergen, Greenland, and other northern lands are barren and poor in vegetation when compared with land distant from the coast. The vital and morphological significance of water to the plant in other respects can best be dealt with later on, partly in connexion with the different communities, but a few matters may be noted here : — A moist climate lengthens the life of individuals and of leaves. Aridity, on the contrary, shortens the vegetative period, hastens blooming, the inception of fruits, the maturation of seed ; also brings into existence a marked resting-season, and in steppes and deserts, very numerous annual species. The geographical significance of water is still greater than that of heat, because its distribution is still more uneven ; this is true not only in the main, but also and especially in details. Water is of all factors the most pregnant in relation to kind and distribution of plant-communities. The relation between rainfall and the amount of water needed by the plant is of great import in regard to differences in the vegetation. Upon this depends the development of equatorial forest-zones, where the rain- fall is very great, of desert-zones near the two tropics when the rainfall * Wiesner, 1895. - In this and other devices subsequently mentioned, illumination also plays a part, see p. 19. ' Wiesner, loc. cit. * See p. 16. WARMING D 34 OECOLOGICAL FACTORS AND THEIR ACTION sect, i is very scanty, and finally of the great temperate forest-zones. The rain- fall and its distribution during the seasons determine the great regional distribution of types of vegetation, while differences in the water- capacity of the various kinds of soil and the various conditions controlling the course of water above ground determine the finer topographical shades of distinction. On high mountains the regions are correlated with the distribution of the rainfall. There are often three regions : a lower one with scanty rainfall; a middle one, the cloudy region, with much mist and rain, and consequently clothed with forest ; an upper dry one above the clouds (e. g. Tian Shan, Madeira, Teneriffe). Mountains often show a dry lee-side and luxuriant rainy weather-side. The coast mountains of a country may arrest the rain so that in the interior, steppe, savannah, and the like, develop on the drier soil, whilst the coast-land yields a rich forest-vegetation (e. g. the coast of Brazil and the campos of the interior). The distribution of atmospheric precipitations. With the same rainfall there is a very great difference according as the rain falls uniformly throughout a long period (Central Europe), or falls for only a very short time in the form of heavy storms ; the number of rainy days is, in so far, therefore, of greater import than is the amount of rain. In the former case the rain is capable of being much more beneficial to the vegetation ; in the latter case the parched soil is not in a condition to absorb all the water, most of which, flooding and denuding the soil, flows away over its surface or percolates to its depths. Under the former conditions we find growth-forms and plant-communities quite different (mesophilous) from those under the latter, which are more extreme.^ /* It is remarkable that even in smaller districts relatively slight differ- ences in the amount of atmospheric precipitations are capable of evoking 'great distinctions in the vegetation. Thus, in the rainier parts of North Germany, especially in the north-west, heath dominates, and in its company grow a whole series of typical Atlantic plants that are wanting in the less rainy east. In this latter there is, consequently, a flora much richer in species which prefer dryness and which (also in cultivation) show themselves very sensitive to great humidity, especially in spring and autumn.^ Small quantities of rain are of little or no use to vegetation, because evaporation is rapid, and the water evaporates before it has time to sink into the soil. The time of the atmospheric precipitations (according to the season of the year) is of very great importance. Where in the tropics a heavy rainfall is distributed throughout the whole year, evergreen rain-forest prevails ; where the rainfall is likewise very heavy but is confined to a few months in the year, whilst the rest of the year is dry, high-forest may be present but it will consist of deciduous trees. Rain that is essentially winter-rain, as in Mediterranean countries or in South- West Australia ^, obviously favours a type of vegetation entirely different from that favoured by rain falling in summer. The Mediterranean district and South-West Australia are consequently poor in forest, rich in steppe and bushland ; whilst the vegetation of districts with summer-rain, for example East AustraHa, is characterized by rain-forest, and savannah rich in trees. In * Woikof, 1887 ; Koppen, 1900. '^ Grabner, 1895, 1901. * Diels, 1906. CHAP. VII ATMOSPHERIC PRECIPITATIONS 35 the former countries hot and dry seasons coincide, and the vegetation consequently has a xerophilous impress ; in districts with summer-rain, where the same quantity of rain falls, the vegetation has a more meso- philous impress. Against dry seasons plants may protect themselves by shedding the strongly transpiring surfaces {defoliation). Other plants that do not shed their foliage necessarily retain, also during the moist season, the features required to protect them from dryness during the dry season. In the tropics, where there is a prolonged rainless season, deciduous foliage is the rule. In sub-tropical (warm-temperate) zones, evergreen trees and shrubs predominate, while many herbs dry up during the dry season. In sub- tropical districts with summer-rain, we may perhaps regard the reduced evaporation during the cooler winter as being responsible for the evergreen nature of the trees. In districts with winter-rain, trees and shrubs that are leafless during winter subsist but badly, because the summer is too dry. Nor are shrubs that shed their leaves in summer so common as evergreen ones. In dry districts with a very short vegetative season (steppes and deserts), the herbs dry up during the dry season, and the shrubs, for the most part, shed their leaves. It should be noted that in many steppes (e.g. South Russia and Hungary) the summer months during which the vegetation is dried up, are the rainiest ; the rainfall in summer is not sufficient to supply the amount of water required by plants, as it is not great and is not sufficient to atone for the intense evaporation during the hot summer months when the air is dry ; the rain falling in spring, though still less in quantity is more efficient. In cold-temperate zones winter is to be regarded as a ' physiologically dry season \^ because, while low temperatures prevail the plants cannot absorb water from the soil. The trees and shrubs are either deciduous or have perennial pro- tection against drought. According to Grisebach,"^ deciduous trees have effective protection against evaporation during winter, but the means employed are not economical, because a not inconsiderable portion of the vegetative season is consumed in the issue of foliage. Consequently, evergreen Coniferae preponderate immediately that the length of the vegetative season sinks below a certain minimum. According to Koppen ^ the southern boundary of the predominant coniferous forests is parallel with the lines denoting equal duration of the warm season. Herbs in the cold temperate zone are mostly evergreen. During the frosty period they find protection under cover of the snow. This is also true of the herbs and dwarf-shrubs in the Arctic zone. It is clear that matters influencing the amount, distribution, and other distinctive features of atmospheric precipitations, are of indirect significance to oecological plant-geography. Such matters are especially topographical, and include : relief of the earth's surface, altitude above the sea-level, proximity to the sea, prevailing winds and their humidity. Heat and moisture may be the two weightiest factors determining the develop- ment of vegetation. According to the different quantitative i)roportions in which plants receive and are adapted to them, A. dc Candollc* has ranged plants into the following six groups : — I. Hydromegathermic : Plants making the greatest demands as regards ' Schimper, 1898. * Grisebach, 1872. * Koppen, 1900. ' A. de CandoUe, 1874. D 2 36 OECOLOGICAL FACTORS AND THEIR ACTION sect, i heat (a mean temperature of at least 20° C.) and water ; their present home hes particularly in moist tropical tracts, but at an earlier date they were certainly widely distributed. 2. Xerophilous : Plants caUing for much heat, but making the most modest demands for water. Here belong plants of the desert, steppe, and savannah. 3. Mesothermic : Plants calfing for an annual mean temperature of i5°-20° C, and, at least during certain periods, abundant moisture. In the Tertiary epocli these extended up to the North Polar lands. 4. Microthermic : Plants requiring an annual mean temperature of 0-15° C, little of the sun's heat, uniformly distributed atmospheric precipitations, and a period of rest caused by the cold. 5. Hekistothermic : Plants living beyond the Umits of tree-growth, where the annual mean temperature sinks below 0° C. ; they endure prolonged lack of Ught. 6. Megistothermic plants existed in earher ages of the world's history, and demanded high uniform temperatures (above 30° C). They were especially Crypto- gamia. A. de Candolle's groups suffer from the defect that no plants are dependent upon the mean annual temperature, but that plants depend upon the duration and temperature of the vegetative season, and upon certain minima of temperature and humidity which must not be transgressed. As an emendation more consonant with nature, the following arrangement is proposed.' 1. Hydromegathermic plants: mean temperature of the coldest month being more than 16° C. 2. Xerophilous plants : the rainiest month having less than twelve days with rain. 3. Mesothermic plants : mean temperature of the coldest month being below 16° C. yet not below 0° C. for long together. 4. Microthermic plants : winter having periods of prolonged frost (with snow remaining on the ground). 5. Hekistothermic plants : mean temperature of the warmest month being less than 10° C. CHAPTER VIII. MOVEMENTS OF THE AIR Wind exerts an influence upon both the configuration and the distri- bution of plants. This is to be seen most clearly when it blows over large stretches where its force is not broken by mountains, forest, or town, as is the case on sea-coasts and on extensive plains, such as the Asiatic steppes, the Sahara, and the like ; it is also seen where a definite wind, the trade- wind, prevails. The effects are revealed on tracts with a loose sandy soil and a scanty covering of plants, for instance, on many coasts and the Sahara, in the formation of dunes, with which is associated a highly characteristic vegetation. They are furthermore revealed in mountain-districts by the higher atmospheric humidity and greater rainfall, which are caused by the daily valley-winds.2 Likewise on lofty mountain-chains they are reflected in the distribution of the atmospheric precipitations, because the windward side catches the moisture brought by the wind (e. g. the east and south- east coasts of Australia, the east side of the Andes), while the lee-side remains dry. Dependent on these circumstances is the distribution of the different plant-communities according to the amount of moisture that they require, many species and whole formations being restricted as to their altitude above sea-level, and as to their environmental bounds. ' See Koppen, 1900. " Hann, 1897, 1901 ; Vahl, 19046; Scott-Elliot, 1900. CHAP. VIII MOVEMENTS OF THE AIR 37 Eminently worthy of note is the significance of the f ohn-vvind ^ to vegetation. The valleys where the fohn prevails are well known for the vegetation, which is that of a warmer chmate. In places sheltered from the wind the vegetation shows a development different from that in unsheltered situations. Wind, when strong and much inchned to prevail in one direction, exercises a remarkable influence on the form of tree-growth and on the whole character of the landscape. Distorted growth and nanism are the consequence. Trees display the following peculiarities in shape : — (i) They are low in stature. (ii) The trunk is often bent in the direction towards which the pre- vailing wind blows, and the boughs are curved and bent in the same sense. (iii) The shoots are short, often irregularly branched and interlaced. (iv) Many shoots are killed on the windward side, and sometimes one finds new shoots and fresh leaves only on the lee-side. (v) The crown assumes a peculiar shape by unilateral branching (Picea excelsa), or, because it inchnes from the windward side, appears as if clipped and rounded off, and exposes a very close-set surface.^ (vi) The whole forest or bush-wood inclines in like manner away from the windward side. (vii) Sometimes on the most exposed side the shoots springing from the root or from the base of the stems are the only ones to maintain a fair existence : so that on the windward side a forest may dwindle to scrub, and this in turn may be resolved into scattered or isolated cushion- like individuals (e. g. on the heaths of Jutland). (viii) The leaves become smaller than usual, and often are more or less brown in patches, or reddish (as if burnt), particularly at the margins. Like effects of the fohn-wind in East Greenland, on dwarf shrubs and perennial herbs, have been described and figured by Hartz ^ ; in this case the masses of sand and stone carried by gales have an erosive and destructive influence on the windward side.* Various explanations of the effects of wind have been given : — Borggreve assumed that all effects are essentially due to the mechanical action of wind, in that the shoots and leaves are beaten against one another, shaken, rubbed, and lashed. It is certain that wind can exert a direct mechanical action upon plants, and, for example, cause trunks of trees to slope and their branches to be eroded and barked. Other authorities ascribed to wind an indirect physiological action of some kind or other. Focke expresses the opinion that the injuries done to plants may be wrought by particles of salt conveyed by sea-breezes, but the same changes in form are to be observed far inland : for instance, in oak- scrub in central Jutland, or in the centre of Switzerland. Still others regarded cold as the cause, but on tropical coasts, for instance in the West Indies and at the mouth of the Amazon, we note, under the influence of the trade-wind, the re-appearance of shapes identical with those of our own latitudes, and the cessation of the effects of the wind when any sheltering object intervenes. ' In regard to the theory of the fohn-vvind consult Hann, 1897. ' Friih, 1901 ; Klein, 1899, 1905. * Hartz, 1895. ' See also Bematzky, 1901. 38 OECOLOGICAL FACTORS AND THEIR ACTION sect, i The truth is probably that the effects of the wind are largely due to the consequent increase in transpiration leading to desiccation, as was suggested by Wiesner in 1887, by Kihlman in 1890, and by Warming in his lectures in 1889.^ Wind has a desiccating action which increases with its force. It dries the soil : places very exposed to the wind acquire a relatively xerophilous vegetation. It dries plants, so that these must adapt themselves to their conditions in order to avoid desiccation. In a calm atmosphere the air adjacent to plants becomes humid, so that transpiration is obstructed. By even weak movements of the atmosphere air is constantly carried away, and fresh, less-humid portions of it come into contact with the plant. Even when the atmosphere is very humid its uninterrupted renewal will lead to strong transpiration. The drier the air and the stronger the wind, the greater will be the drying action. Evaporation is proportional to the square root of the wind-velocity. The shaking of the plant-organs also operates in the same direction. By this transpiration the growth in length of axes and leaves is decreased (nanism), many leaves and whole shoots are killed, so that irregular branching results : thus all the observed phenomena receive an explana- tion that is not forced. The deviation of the shoots in the direction of the prevalent wind is possibly almost without exception caused by a kind of sympodial growth. That the crown on the lee-side acquires a gradually ascending shape, is caused by the circumstances that the shoots, both living and dead, on the windward side screen the parts on the lee-side from too rapid a renewal of the air. In this matter once more we see the fundamental significance of water to plant- life. The force of the wind is far less on the ground than at some distance above it ; consequently short plants are much better protected from the wind than are taller ones.^ The danger arising from the wind is increased when at the same time the activity of the root is decreased by coldness of the soil, so that the loss of water is covered with difficulty or not at all ; hence in Central Europe when there is too little snow in winter cereals and other plants perish. This circumstance is of particular importance in arctic and alpine situations. The espalier-shape, mentioned in p. 26, as assumed by shrubs growing in these places, may be caused by wind, and we often see the shoots directed straight away from the windward side. The cushion-like growth of the herbs (Draba,^ Androsace helvetica,* and others) living under similar unfavourable conditions in windy cold places, may obviously arise in the same way. Even arctic mosses show similar construction. ^ Herbs of this kind acquire, for want of water, short shoots and small leaves, become as a whole very stunted pygmy- forms ; they are richly branched, consequently often of extraordinarily dense growth, and are very like miniature shrubs found in scrub. Often cushion-plants, for instance Silene acaulis, are dried up on the windward side. That dryness can really bring into existence such cushion-like forms is confirmed by plants growing in arid, hot, but tolerably calm, desert places. ' See Warming, 1884, p. 99. ' Wiesner, 1887. ^ See illustration in Kjellman, 1884, p. 474. * Kihlman, 1890 ; G. Andersson, 1902. "^ Ottli, 1903. CHAP. VIII MOVEMENTS OF THE AIR 39 The transverse section of tree-trunks is also influenced by the wind, since it becomes excentric — the diameter in the direction of the wind being longer than that at right angles to it. Plants vary according to their species in power of resistance to the wind.^ Of the trees common in Denmark the hardiest in this respect are the following : Pinus montana, P. austriaca, Picea alba, as well as some species of willow and poplar, and these are consequently also the species of greatest value in afforesting dunes and heaths. The importance of protection against wind has thus been made clear. Such protection is provided by elevations of the land, as well as by other natural or artificial protective barriers : careful study will often show that vegetation differing widely as regards density, stature, structure, development, and admixture of species, can arise respectively on the windward and lee sides of such a barrier, even when this is only an insignificant rock, stone, or shrub. The hills of Central Jutland appear when viewed from the east to be clothed with forest, but when viewed from the west to be clad with heath. In beech forests the vegetation clothing the soil of places where light and wind can penetrate is quite different from that where these are excluded. In this case the wind has inter alia an indirectly injurious action, in that it removes the carpet of dead leaves which protects and variously affects ^ the nature of the soil, and in that it leads to the conversion of mild humus into acid humus, or prevents the production of humus. Arctic and alpine vegetation, as was shown by Kihlman ^, receive very material protection from snow, and where this remains lying, particularly in calm and sheltered depressions, the vegetation is consequently of a stamp different from that on more elevated, windy spots.* The defences against wind that have hitherto been mentioned are topographical, but many plants have by adaptation acquired special structural features, both morphological and anatomical, by which they are protected. To this category belong bud-scales, covering hairs, remnants of leaves and stems that are long persistent, and other features which will be dealt with later in this work.^ Distribution of vegetation. It may be added that though the absence of trees from many places on the Earth is mainly due to wind, yet it is also due to cold and other conditions unfavourable to growth. Wind is partially responsible for the delimitation of the boundaries of forest in Polar lands, and of forest and bushland up high mountains. On mountains, forest ceases where the mountain commences to divide into separate peaks. Above this limit forest can still occur where there is local shelter from the wind, for instance, within the crater-valleys of Java.^ Also, it is in valleys sheltered from the wind that forest extends farthest north in arctic lands ; for instance, along the Lena and Mackenzie rivers. Middendorff' was the first to recognize the significance of the wind in assigning limits to the extension of forest.^ ' Illustrations are given in L. Klein, 1905. ' See Chapter X\III. * Kihlman, 1890. * For further particulars see C. Schroter, 1904-8, and Chapter XVI Tl. ■ Also see Chapter VI, p. 24; and Chapter XXX. " Schimpcr, 1893. '^ Middendorff, 1867. * The importance of wind has been treated in an attractive and detailed manner by Kihlman, 1890; and more recently by Bernatzky, 1901 ; Buchcnau, 1903; 40 OECOLOGICAL FACTORS AND THEIR ACTION sect, i The utility of the wind to vegetation is especially shown in the con- veyance of fresh supplies of carbon dioxide. The transport of pollen to anemophilous plants, such as coniferous and dicotylous trees of North and Central Europe, and in the dispersal of seeds ; many of our common trees have their seeds scattered by the wind.^ CHAPTER IX. NATURE OF THE NUTRIENT SUBSTRATUM The nature of the nutrient substratum, or edaphic ^ conditions, largely determines the habitats of plants and their topographical distribution ; and among all characteristics of the substratum the most important is the amount of water contained. There are two different forms of nutrient substratum available to autophytes : water and soil. Both these have to provide the plant with space and nutriment, as well as with external conditions suitable for the absorption and preparation of nutritive material : they make these provisions by entirely different methods and must, therefore, be treated separately.^ The air is not in itself a nutritive medium in which plants habitually live and feed, but is merely a temporary resort for organisms that are nearly all of them microscopic but are present in countless numbers, which vary according to time and space, being greatest in the vicinity of human dweUings, particularly in large towns, and being least over oceans, high mountains, and in forests. The weightiest geographical role of the air is that by its currents it affords ways and means for the transport of countless organisms from one place to another. The discussion on water and its characters that are of greatest import to oecological plant-geography will be deferred until hydrophilous com- munities are treated in Section IV. But the characters of soil will be dealt with at once : they depend upon the physical and chemical attributes of the soil-constituents. CHAPTER X. STRUCTURE OF SOIL The term soil is used here in a wide sense and includes — 1. Solid rock ; 2. Loose soil produced in situ by weathering ; 3. Secondary, loose, transported soil that is the product of weather- ing at some other spot. Solid rock. The characters of solid rock depend upon its miner alogical nature. It varies greatly in hardness, porosity, specific heat, and power of radiation, as is shown by contrasts such as granite, shale, and limestone. Loose soil. By the mechanical disintegration and chemical decom- Friih, 1901 ; Norton, 1897; Ganong, 1899; L. Klein, 1899, 1905; Kraus, 1905; Klinge, 1890; Schimper, 1898; see Schenck, 1905. ^ Cp. Warming, 1887 ; Sernander, 1901 ; P. Vogler, 1901. * TO «Sa0os, the soil. This term was introduced by Schimper (1898). =■ See Chapters XXVII-XXXIII. CHAP. X STRUCTURE OF SOIL 41 position of rock there arises loose soil : the active agencies are, particularly, changes of temperature, congelation of water as well as the chemical action of water and of the oxygen and carbon dioxide of the atmosphere. In certain cases lowly organized plants, such as lichens and bacteria, also play a part. Chemical decomposition and mechanical disintegration always go hand in hand. Secondary soil. This owes its origin to the transport, and partially to the separation of the different constituents, of soil produced by weather- ing ; the transporting agents are, particularly, currents of water, move- ments of glaciers, and wind. Rivers (Po, Nile, Ganges, and others) heap up at their mouths masses of loose substance that has been conveyed from mountains ; glaciers during the Glacial Epoch transported vast masses of soil to distant spots (for instance, from Norway and Sweden to Denmark and North Germany) and continue to do so in the present ; seas in their currents carry with them other masses of substance. Wind deposits sand from the seashore and from inland sandy soil in the form of dunes ; it also carries away fine particles from the surface of soil and deposits them in sheltered places (as loess). The characters of loose soil depend upon many and divers features, and particularly upon the fineness, chemical nature, arrangement, and cohesion of its constituents, as will be detailed in the sequel. From loose soil there often arise new kinds of rock, for instance sand- stone, shale, and conglomerate, which differ in character from the original rock and play a different role in plant-economy. Loose soil has the following structure. It is a mixture of — 1. Solid constituents; 2. Air (Chap. XI); 3. Water (Chap. XII). SOLID CONSTITUENTS OF THE SOIL The solid constituents of soil are : — {a) Larger miner alogical constituents, stones varying in quantity and size down to extremely small grains of sand ; if the soil be shaken up in water and allowed to stand, these constituents rapidly sink to the bottom. In chemical composition they vary greatly, but quartz is most common. (6) Very minute, powder-like particles, which remain suspended in water for a long time when the soil is shaken up in water and allowed to stand. By this process they can be easily separated from sand. They likewise vary greatly in chemical composition, but are mainly composed of aluminium sihcate, and compounds of iron and calcium ; they have an essential influence on the amount of nutriment in the soil, on its power of absorption, and on its physical characters. (c) Humus-substances arise from the corpses and by-products of plants or animals. They are destroyed by combustion. Many humus- substances clearly show their organic origin and mostly impart to the soil a black or dark brown colour. These three kinds of constituents occur in nearly all soils. All constituents that are too large to pass through a sieve with meshes 03 millimetre m width are termed by W. Knop the soil-skeleton (coarse 42 OECOLOGICAL FACTORS AND THEIR ACTION sect, i sand, grit, and stones, which can be further sub-divided into groups by the aid of the sieve) ; the remaining constituents are termed fine earth. The fine earth plays a special part on plant-life, directly as food material, and indirectly because of its power of absorbing important nutritive substances and because of its purely physical attributes. An admixture of stones and grit nevertheless considerably modifies physical relations in the soil. Pore-volume. The commixture, the relative amounts and the arrange- ment of the solid constituents enumerated is very different in different soils. Between the solid constituents there are small cavities termed pores. The sum of these spaces not occupied by solid constituents in a given volume of soil is termed its pore-volume. vSoil is very rich in con- tinuous spaces which become the more capillary the narrower they are. These pores are filled with air and water, the relative proportions of which depend upon the size of the pores and other circumstances. In the region of the ground-water, the pores are nearly completely filled with water ; at the surface of a sand-dune that has been exposed to prolonged drought we find the converse, a maximum of air and a minimum of water. Some kinds of soil are more or less crumbly, or capable of becoming so, that is, their individual particles do not remain separate, but combine to form larger particles, which may be termed compound particles or compound grains. Compound particles are especially found in humus ; according to Darwin,^ P. E. Miiller,^ and others, they are often the excre- menta and casts of subterranean animals, especially earthworms and insect-larvae.^ Soil having these compound particles acquires characters other than those of a soil consisting of simple particles : it is looser, more easily aerated, takes up water more readily, and allows roots to penetrate more freely. In the practise of horticulture and agriculture an endeavour is made to promote the formation of compound particles in the soil by turning over and ploughing the soil so that its bulk is easily changed by physical factors (especially frost), and by adding other kinds of soil or substance, such as sand, humus, marl, so that its tenacity is changed. Tenacity of the soil. The force with which particles of soil are held together varies greatly. As contrasts may be mentioned the dune, whose grains of sand are quite loose in a dry condition, and clay ; humusjike- wise, has little tenacity. Soils may be distinguished into such as are rigid, stiff {heavy), mellow (mild), lax, loose, shifting. Rigid soil on drying becomes hard, fissured, and crustaceous, so that the subterranean parts of plants may be ruptured ; the particles of shifting soil on drying become separated from one another, and are so light that they may be carried away by the wind. Tenacity depends, inter alia, upon the size and chemical constitution of the particles ; the smaller the particles are, the greater in general is the tenacity. Plant-form and vegetation as a whole are clearly influenced by the tenacity of the soil. In loose soil (sand, mud, humus in the forest, bog- moss, and so forth) the production of long, richly-branched roots and long, horizontal, subterranean stems (runners and rhizomes) with long internodes is favoured, doubtless because the resistance to be overcome ^ Darwin, 1881. ^ P. E. Miiller, 1887a. ' See Chapter XX. CHAP. XI AIR IN THE SOIL 43 during growth is small ^ ; in this way social growth is promoted, and the landscape may even acquire a special uniform physiognomy, for example, through Ammophila and Elymus on dunes, or Phragmites and Scirpus in swamps. Firm, very tenacious clay, on the other hand, on drying becomes hard and cracked, and therefore is not well suited to such plants ; in it are found plants with a vertical, short, thick root-stock (tuber or bulb), or with a multicipital rhizome and caespitose habit, for instance, on the compos of Brazil.- Rigid plastic clay is no favourable soil for plants, indeed (when it occurs beneath other layers) it may form an impenetrable obstacle to plants. Sohd rock (without any deposit of loose soil) does not in the least suit plants of the former habit, but may permit the enter- tainment of plants of the second kind in its splits and clefts (chasmo- phytes), and beyond this only such plants can settle on its surface (lithophytes) as have special organs of attachment.^ It remains to be said that the root-structure of the various species is very little known, and distinctions in this may perhaps often afford an explanation of the distribution of species. The capillary action of soil plays a very important part in the physical constitution. It depends especially upon the size and arrangement of the particles. The smaller are the particles and the more closely are they packed, the greater is the capillary action ; soil with compound particles has less capillary action than if consisting of simple particles ; stones and coarse grit in the earth likewise depress capillary action. CHAPTER XI. AIR IN THE SOIL Air in the soil is of most fundamental significance to plant life ; all living subterranean parts, like all other living parts, require air (oxygen) for respiration. In very wet soil, normal plants, adapted to soil rich in air, are suffocated ; alcoholic fermentation, the evolution of carbon dioxide, and consequently death and putrefaction * ensue ; in soil poor in oxygen decomposition takes place in a manner different from that in aerated soil ; humous acids are formed in great quantities, so that the soil becomes 'sour'. The aeration of soil depends essentially upon the structure; the more porous and loose the soil is, the more free is the aeration. Farmers and gardeners break up the soil with plough and spade, and drain and harrow it, so that, among other reasons, air may be freely admitted. In order to aerate the soil, the Dutch farmer causes the water-table of his meadows to sink to a depth of one metre during autumn and winter, but during the remaining months only to a depth of half a metre ; this is also the practice in the meadows of Soborg in Denmark.' A production of acid humus in the forest leads to an exclusion of the air, and consequently to an extinction of the forest. Air in the soil is somewhat different in composition from that in the atmosphere ; it contains more carbon dioxide and less oxygen, parti- ' Henslow, 1895. - Warming, 1892 ; Lindman (1900) terms certain woody subterranean tuberous structures ' xylopodia '. ' Ottli, IQ03. See Section VIII. * See Sorauer, 1886. ^ See Feilberg, 1890. 44 OECOLOGICAL FACTORS AND THEIR ACTION sect, i cularly in the deeper layers, because of the respiration of subterranean organs, plants (bacteria) and animals, and because of the decomposition of organic bodies. The amount of carbon dioxide varies with the quantity of organic matter in the soil, the vegetation, the contour and humidity of the land, the size of the soil-particles, the depth of stratum (the upper- most layers of soil have less carbonic acid than have the lower ones), and the temperature (season). The internal structure of the plant is correlated with the amount of air contained in the soil ; in very wet soil, as a rule (with the special excep- tion of bacteria), only such plants can thrive as have large internal air- spaces, which are in communication with one another throughout the whole plant, and can convey air from the atmosphere itself to the most distant root-tips and parts of the rhizome (aquatic and paludal plants ; horse- tails in firm clay contrasting with plants in heath-moors, which contain much more air).^ CHAPTER XII. WATER IN THE SOIL Water is the third component of soil. It is attracted by the solid particles of soil, and surrounds them as a thinner or thicker film.^ The amount of water varies greatly in different places and at different times in the same place. After Norlin we may distinguish the following grades which, as a rule, are only approximately estimated : i = very dry, 2 = moderately dry, 3 = moderately fresh, 4 = fresh, 5 = some- what moist, 6 = moist, 7 = very moist, 8 = moderately wet, 9 = wet, 10 = very wet.^ In more detailed scientific research, the amount of water must be expressed in percentages of the weight or volume of soil. The quantity of water in soil is practically indicated best of all by the plants growing on it ; for no factor has such an influence upon the disposi- tion of species as the amount of water in the soil. The amount of water in the soil is one of the most important direct factors operating on plant life : this follows from the statements in Chap. VII respecting the fundamental significance of water in plant-economy. Water must be present in certain proportions, which are definite for each species (in cultivated plants usually not more than sixty per cent, for any prolonged period) ; too much or too Uttle is injurious in this as in other cases. The significance to plant life of the quantity of water in soil is demonstrated, for example, by Fittbogen's investigations on oats : on soil, the humidity of which varied between forty and eighty per cent., there was no great difference to the resulting crops ; but with a humidity of twenty per cent, the crop was halved, and with one of ten per cent, the crop was reduced to an eighth. Lack of water in the soil causes the plants to be ill-nourished, because roots can obtain nutriment from such a soil only with difficulty. Water is also of indirect significance, as it affects animals and bacteria living in the soil ; a certain degree of humidity is essential to the production of humus. ' See Sections III, IV, V. ^ See Sachs, J. von, 1865, p. 171 ; Hedgcock, 1902. 3 See Hult, 1881. CHAP. XII WATER IN THE SOIL 45 Water in the soil is (i) chemically combined water, which for the most part plays no considerable part in plant-economy ; (2) water absorbed from aqueous vapour of the atmosphere ; (3) water received from atmospheric precipitations and retained by capillary action ; (4) ground- water, or water sucked up from it. Ground-water is that collected above the impermeable stratum of soil, and moving according to the laws of gravity or remaining in the soil in sheets, just as does water exposed above the surface of soil. A layer of clay mostly serves as the substratum of ground-water ; sand and gravel permit the passage of water. Ground-water may contain many soluble substances, especially calcic salts ; but when it lies deep it is, as a rule, poor in substances nutritive to plants (it is pure), because these have been retained by the over-lying layers ; it is also devoid of bacteria, because the upper layers of soil have acted as a filter. The level of the ground-water and fluctuations in this according to the seasons of the year depend upon the amounts of atmospheric precipita- tion and evaporation, and are of very considerable oecological signifi- cance, and play a most important part especially in the desert. In many cases ground-water lies too high for certain plants ; in other cases it is so far below that the roots cannot utilize it directly or indirectly ; in still other cases it is at such a depth as to be reached by the roots at certain seasons, but not at others. In these cases the height to which the water can be raised by capillary action is an important item. The level of ground-water obviously influences the temperature of soil.i The ability of plants to utilize water is very diverse, because the roots penetrate to different depths. Dry summers acquire great signifi- cance in relation to different species, some of which suffer or die sooner than others.- Trees with deep roots can thrive even in a dry climate when they are able to reach ground-water. It may be noted that, accord- ing to Ototzky, the level of ground-water invariably sinks in the vicinity of forest, and always hes higher in an adjoining steppe than in a forest ; forest consumes water. The significance of the level of ground-water is very clearly demon- strated in Denmark. Here chemical differences in the soil, which has been pulverized and deposited by glaciers, are scarcely so great as in mountainous countries where the rock lies near the surface and possibly reacts on the vegetation by reason of its chemical nature. A case in point, according to Feilberg,^ is provided by the sandy plains near Skagen in Jutland. When the ground- water in summer is at a depth of three inches, Juncus-vegetation and meadow - moor prevail ; at six inches mosses (Hypnaceae) and Cyperaceae still play a part, but grasses begin to occur ; at nine inches these latter become dominant ; at twelve inches, normal grass-growth occurs in ordinary summers ; at fifteen inches, cereals thrive in somewhat warm summers ; at from eighteen to twenty- four inches, cereals thrive in cold or moist summers ; at from thirty to forty inches, the soil is unsuitable to cereals, and xerophytes reign. Other examples are given by Feilberg, who lays greater stress, and rightly so, than perhaps the majority of other investigators do, on the importance of ' Sec Chapter XIII. ' See Deherain. i8(^j, and others. ' Feilberg, 1890. 46 OECOLOGICAL FACTORS AND THEIR ACTION sect, i the level of ground-water ; he gives, for instance/ one as showing how the vegetation of a district gradually changes with the descent of the water- table. Many trees assume a pecuhar shape, or cannot grow at all on soil with ground- water near the surface. Warming ^ gives additional exam- ples ; but in these cases further investigation is required to demonstrate what part is played by the level of ground-water, and what by other properties of the soil, including its power of raising water. In addition, periodic fluctuations in the level of ground- water embracing a number of years (Bruckner's thirty-five year periods) has been recog- nized. These observed fluctuations of climate cannot be associated with Blytt's theory of alternating moist and dry epochs, and corresponding changes in the vegetation. In the layer of soil lying above the water-table the amount of water is influenced by the following important characters : facility of percolation in soil, hygroscopic character of soil, its power of raising water, its water- capacity, as well as the amount of atmospheric precipitations^ and the influx of surface-water. Facility of percolation in the soil. Atmospheric precipitations do not penetrate all kinds of soil with equal facility, as may readily be seen if water be poured on sand, clay, and humus. The following factors play a part in this matter : the water-capacity of soil, also the kind and dryness of the particles of soil. The greater the water-capacity the more slowly does water sink in the soil. Very fine-grained soil, especially clay and certain humus- soils, are almost impermeable to atmospheric precipitations when the particles are densely packed ; whereas the more coarse-grained and loose is the soil the more freely is it penetrated by atmospheric precipitations. If the soil be rich in good-sized stones or in crevices and cavities, such as the burrows of earthworms, then the velocity of penetration is modified by these : it is decreased by stones, but increased by crevices and cavities. Water penetrates most readily into quartz-sand, less readily into humus, and least fully into clay. Clay soil permits the percolation of water with difficulty, not only because of the small size of its particles, but also because of their other characters. If the uppermost layers of soil be very dry, some time elapses before they are so wetted as to allow the infiltration of water to commence. The hygroscopic character of soil. All porous and dry soil can absorb aqueous vapour, though to a very varied extent. The absorbed water vapour is invariably available to the plant, because it is taken in only when the earth is dry ; it can never provide too much water. On the other hand it is not capable of alone supplying dry soil with water sufficient for the needs of plants ; these wither before the amount of water in the soil has decreased to such a degree that absorption of aqueous vapour takes place. Power of soil to raise water. The power that soil possesses of raising water from the deeper layers is obviously of importance to plant life. But we must distinguish between the heights to which, and the velocity with which, water is raised. These depend, inter alia, upon capillarity * Feilberg, 1891, p. 270. * Warming, 1887, 1890, 1891. =" See Chap. VII. CHAP. XII WATER IN THE SOIL 47 and upon the nature of the particles. Quartz-sand raises water rapidly ; clay and other very fine grained soils raise it slowly ; calcareous sand and humus fairly rapidly. But the height to which it is raised is least in sand (only about forty centimetres above the water-table in fine sand, according to Ramann ^), is greater in clay, and greatest in peat. (The widely accepted view that bog-mosses in heath-moors raise water out of the ground is nevertheless incorrect.) ^ If the grains of soil be more than two to three millimetres in size, the pores are too large to act as capillary tubes. The power of the soil to raise water is particularly of importance when evaporation from the surface of the soil is great. It may be added that it is more advantageous for a soil poor in water to have a small than to have a great power of raising water, because in the former case the soil is not so easily dried up. By the water-capacity of soil we mean its power to take up and retain liquid, so that none of it sinks into deeper layers of soil. This is measured by the quantity of water that a given weight, or better a given volume, of soil can retain. It depends upon the adhesion of water to the particles of soil, and varies with the capillary power of the soil and with the natuie of the particles. The water-capacity is greater the more numerous and narrow are the capillary spaces in soil and the more uniform their size, because the adhesion-surface is thereby increased. Quartz-sand with particles one to two millimetres in size can retain only one-tenth of the amount held when the size of the particles is •01—07 milUmetre.^ Research ^ has shown that the water-capacity is smallest in quartz- sand, greater in calcareous sand, still greater in clay or in fine, pure calcareous soil, and greatest in humus soil. In the last, the amount of water is reinforced by the presence of imbibition-water, which occurs in organic bodies ; of all soils peat has the greatest water-capacity. Some kinds of soil display so strong an adhesion to water that when ^his is added the interstices between their sohd components are widened and thus their volume is increased, that is to say, these soils swell ; on the contrary, when deprived of some water they shrink ; in this way a modification in the characters of these soils takes place ; when wet they are soft and partially plastic, when dry they are hard and brittle. These statements hold good in reference to clay and peat. In general, soil is not saturated with water (with the obvious exception of swamps and similar spots in the vicinity of ground-water) ; in soil clad with vegetation the maximum capacity is never attained because the plants are continuously expending water in transpiration. The drying of soil depends on various factors : the above-mentioned characters of the soil, the consumption of water by plants and animals, and evaporation. Evaporation obviously has a profound influence on the amount of water in soil and consequently on the economy and constitution of vegeta- tion. Soil retains a certain quantity of water when exposed to the most intense natural evaporation. The force with which water is held fast ' Ramann, 1893, J905- ^ '^^''^ ^- '-'riihner, igoi ; C. A. Weber, iqoj. ' See Livingston, ic/31, 1903, 1905. * Schiibeler, 1886-8. 48 OECOLOGICAL FACTORS AND THEIR ACTION sect, i is of high significance to vegetation. In this respect the various kinds of humus soils are very instructive. Heath-peat (from heath-moors, often composed of the remains of bog-mosses) dries uniformly, and for a long time remains moderately moist internally. The soil of meadow- moors may be as dry as powder at its surface yet greasily wet at a slight depth, for it does not readily permit the equable distribution of water within itself. This property renders it unfit for horticultural usage.^ The factors operating on evaporation are partly internal and partly external. Internal factors are those which depend upon soil itself, such as : the structure of soil, the form of the soil-surface (uneven or even), and so forth. From loose soil less water evaporates than from compact soil, because its power of raising water is less ; the formation of compound particles depresses evaporation. Soil with medium-sized particles permits the greatest evaporation ; large-grained soil permits less. The colour and kind of soil are of influence. From a darker soil more water is evaporated than from a paler one, because dark soil absorbs more radiant heat ; the order of gradation is : black, grey, brown, yellow, red, white. From quartz-sand and humus soil evaporation is most rapid, from calcareous sand and clay it is slowest ; Masure was able to render sand and humus completely dry in three days, clay and calcareous soil in seven days. But the amount of water evaporated in a given time is greater, the greater is the water-capacity of the soil ; in this respect humus stands at the top and quartz-sand at the bottom. In one experiment by Masure, humus retained 41 per cent., but sand only 2-1 per cent. Evaporation from a soil saturated with water is greater than from an equal water-surface. Among the external factors operating on evaporation from the soil must be reckoned : the saturation-deficit of the atmosphere,^ the slope and exposure of the surface, the strength of the wind,^ as well as the vegetation clothing the soil. Plants clothing the soil increase the surface exposed, and uninter- ruptedly extract from the soil water which is dissipated by evaporation from their leaves and other organs above ground. A field under cultiva- tion becomes more rapidly parched than does a fallow field (of course, the remaining conditions being the same). Plants clothing the soil rob it of moisture during the vegetative season, but to degrees that vary with the temperature and the kinds of plants present. The temperature of soil determines how much water is taken up by roots.* Herbs parch the soil more than do trees, and grass is particularly active in this respect. Colding's observations showed that at Copenhagen, from April to Sep- tember short grass consumed an amount of water greater than the rainfall. Feilberg^ estimated the daily amounts per 0-55 hectare of land during May, June, July, and August, at about 400, 500, 350, and 300 cubic feet respectively ; these estimations are, of course, only approximate, and vary with the conditions. The amount of water in soil therefore dimi- nishes from spring to autumn ; at this time of the year it is at its lowest, and may be from five to seven per cent, less than in spring ; subsequently it increases during winter until plant-hfe awakens. The ^ Grabner, 1901. ^ See Chapter VII. ' See Chapter VIII. * See Chapter XIII. ' Feilberg, 1890. CHAP. XII WATER IN THE SOIL 49 differences between species of plants depend upon the sum of the leaf- surfaces and on the leaf-structure, the nature of the root-system, and whether this last is shallow or deep ; thus, in the forest various species act as weeds because they consume water before this can reach the tree roots. We may thus explain how it is that some species are less protected than others in the same habitat. In forests the surface of the soil is protected by the tree-trunks, and consequently remains moist ; but the sub-soil, on the contrary, is robbed of its moisture to a greater extent than when under herbaceous vegetation, because of the activity of the roots of the trees. Roots can utihze water present in the soil only to a certain degree. The more the water in the soil decreases in amount the more firmly is the remaining water held fast, until a point is reached at which the plant can obtain no more, although a large quantity may still be left behind. Sachs was the first to demonstrate this by investigations on the tobacco plant.i A young plant began to wither when the soil (a dark humus) still contained water equivalent to 12-3 per cent, of its dry weight ; the water-capacity of the soil was determined by drying it at 100° C, as being 46 per cent, of its weight ; the plant was able to take up only 337 per cent., and the rest of the water was unavailable to it. Under the same conditions the plants withered on loam and on sand when the percentages of water remaining were 8 and 1-5 per cent, respectively. According to Heinrich's experiments, plants first began to wither in coarse sand when the amount of water had sunk to 1-5 per cent., but in peat when the amount was still 477 per cent. A soil from which a species is incapable of extracting water may be described as dry to that species, even though a large quantity of water may be present in it {physiologically dry)? Physiological dryness alone plays a part in distribution of plants.^ A dead vegetable covering also influences evaporation.^ A soil of considerable humidity may partially replace a moist cUmate. In tropical savannahs the banks of streams are clothed with forest. Furthermore, in steppes and deserts, trees occur where there is running water or where ground-water approaches the surface. Many perennial herbs which in Europe favour a dry sandy soil, occur in the hot dry lowlands of Madeira exclusively on wet soil in the vicinity of springs and water- courses.^ But it is worthy of note that a moist soil cannot always replace atmospheric humidity. Most species of Erica flourish on very dry soil, but are excluded from places with dry air. On the other hand, Tamarix gallica clings to the banks of rivers not only in the Sahara, but also in Central Europe. The significance of water in the soil to plant-form. In addition to what has been said in Chapter VII, concerning the significance of water, it may be mentioned that the production of adventitious roots on prostrate shoots is evidently promoted by moisture ; nowhere else is the production of adventitious roots so abundant and common as in moist places.*^ ' Sachs, J. von, 1865, p. 173. " Schimpcr, 1898; see p. 134. * Schimper, 1898 ; also see Kihlman, 1890 ; Hedgcock, 1902 ; Clements, 1904 ; Burgerstein, 1904. ' See Chapter XVIII. ° M. Vahl, 19046. ' Warming, 1884, 1892. WARMING E 50 OECOLOGICAL FACTORS AND THEIR ACTION sect, i This feature reacts on the duration of life of individuals ; in such spots annual species are rare.^ Moreover, roots branch more freely in moist than in dry soil. Upon the production of root-hairs water also exerts an influence.^ As regards the forms of roots, many ' water-roots ' are known to assume pecuHar forms, but we are ignorant of the actually operating causes.^ CHAPTER XIII. TEMPERATURE OF SOIL The temperature of soil is a geographical factor of paramount signifi- cance. In addition to what has already been recorded in Chapter VI regarding the general significance of heat, it may be mentioned that the junctional activity of the root depends upon the temperature of the soil, and that it increases as the temperature rises up to a certain optimum. A plant may wilt in a soil saturated with water if the temperature of the soil sinks below a certain degree, because in such circumstances the roots can absorb no water (the soil is physiologically dry) ; and a plant may be frozen to death by a soil-temperature that is too low, although it be capable of withstanding a far lower air-temperature ; beech, oak, and ash can withstand an air-temperature of —25° C, but their finer roots succumb to cold at from — I3°C. to — i6°C.* Many places on high mountains and in Polar countries would be certainly devoid of vegetation were it not for the temperature of the soil, for this may considerably exceed that of the air. The temperature of soil may rise exceedingly high in deserts. Bonnet observed a temperature of 59° C. in desert-sand between low plants, when that of the air was 33° C. Pechuel-Loesche observed a temperature of 75°-82° C. in the soil in Loango. The tempera- ture of soil and its fluctuations form the subject-matter of a considerable number of recent papers. The effect of the temperature of soil -upon plant-form is but little understood. Vesque ^ has experimentally shown that a high temperature of the soil gives rise to an abundance of sap (short, thick roots, stems, and leaves), possibly because the activity of the roots suffers from the heat. These features may also be regarded as affording protection against increased transpiration. Prillieux also concluded that a high soil- temperature directly induces the production of tubers. In this way it becomes easier to understand why succulent plants often grow on rocks between stones, or on soil that is easily heated. Nanism may result from a low soil-temperature, if it causes a diminu- tion in the amount of water absorbed and consequently of mineral nutriment taken in ; this factor probably co-operates in inducing the dwarfed growth generally prevailing in subglacial vegetation. It has already been mentioned on p. 26 that cold soil calls into existence prostrate shoots and rosette-like growths, whereas warm soil brings forth slender, tall plants, as Krasan ^ has proved in Pinus, Juniperus, ' Hildebrand, 1882. " F. Schwarz, 1888 ; also see Section III. ^ For further information on the general influence of moisture in the soil, Gain, 1893, 189s, should be consulted. ' Mohl, 1848. ^ Vesque, 1878. * Krasan, 1882-7. CHAP. XIII TEMPERATURE OF SOIL 51 Asperula longifiora, and others. Cold soil would appear to give rise to glaucous shoots, an abbreviation of the vegetative season, and other characters. The main sources of the soWs heat are two in number : (i) heat from the sun ; (2) chemical processes (especially decomposition) in the soil. These processes acquire particular importance in cold countries. The heating or coohng of soil, and consequently plant-life, is obviously greatly influenced not only by those factors (radiation, evaporation, heat-conduction, and so forth) that promote or retard coohng, but also by other factors which we may now consider briefly. Of these the first three to be considered concern the sun's heat, and the remainder relate to the soil itself. 1. Availability of the sun's heat. Particularly in Polar countries direct sunlight plays a leading part, as is clearly shown by the arrangement of the plant-communities over the landscape. In determining this a greater part is played by the heat of the soil than by the heat of the atmosphere.* 2. Angle of incidence of the sun's rays. The nearer this is to a right angle the greater is the heating power of the rays (their power being proportional to the cosine of the angle of incidence). Latitude, slope, and exposure of the land, all affect the result. In northern latitudes, south-west, south, and south-east slopes are warmest, while north-east, north, and north-west slopes are coldest. The relationships indicated in the two preceding paragraphs evoke great differences in the distribution of plant-communities in all latitudes. We see not only in Greenland, for example, that the southern slopes of a mountain-chain may have an open xerophytic vegetation appearing as if burnt up, while the northern slopes are at the same time covered with a dense fresh green, mossy carpet, in which flowering plants are scattered and which in summer are moistened by the slowly-melting snow ; ^ but, also in north-temperate latitudes we note, for instance, that the different faces of thatched roofs support different vegetation ; and again, in Mediterranean countries we observe on the southern mountain slopes, and ascending high up them, xerophilous Mediterranean vegetation with its characteristic forms and its early flowering season, whereas the northern and cooler slopes are stamped with the impress of Central European vegetation, with its more tardy development.^ Even close to the equator, for example in Venezuela (less than 10° N.), we observe most marked distinctions between northern and southern slopes ; near Caracas, stretch- ing from east to west, there are shallow erosion-valleys or folds in the land which, on their southern slopes, are so poor in vegetation that the red clay almost entirely determines the colour of the land, but on their northern slopes are clad with denser and taller vegetation. In lower latitudes (South Europe and the tropics) it must be remembered that prevailing north winds convey more moisture to northern than to southern slopes. This circumstance is perhaps of greater import than is the exposure to the sun's rays, because when the sun is overhead a steep southern declivity is exposed to less in.solation than is a more gently ascending northern slope : the lower the sun stands the more dependent is the intensity of insolation upon exposure. ' See Chapter VI. * See Warming, 1887. ' See FlahauU, 1893. E 2 52 OECOLOGICAL FACTORS AND THEIR ACTION sect, i It may be added that the snow-line may be at very different levels on the north and south sides of a mountain, and that the altitudes reached by many plants depend upon exposure. Selecting as an example the beech in the Alps, according to Sendtner,i the maximum altitude attained by it in South Bavaria is greatest on the south-eastern, and least on the north-eastern side. In the northern hemisphere, species ascend far higher up the southern side of mountains than up the northern (in the Pyrenees, for example, according to Bonnier). The statements above will suffice to show to what an extent heat — in this case the soil's heat (though atmo- spheric heat and radiation cannot be dissociated from this) — depends upon the relationships enumerated. 3. Duration of radiation. In this duration the tropics and Polar lands are very different, at least as regards the distribution of light according to seasons of the year. 4. The specific heat of soil varies with the mineralogical composition. The most easy to heat is quartz-sand, and the most difficult peat ; between these extremes stand calcareous sand, clay, and others. The specific heat of quartz-sand is 0-2, that of peat about 0-5 (water = i). The amount of humus in soil is of special importance in this relation. 5. Colour of soil. Darker soil is more readily and strongly heated than is that of lighter colour, other conditions remaining constant. Humboldt found that black basalt-sand on the island of Graciosa attained a temperature of 51-2° C, whilst white quartz-sand in the same circum- stances attained only 40° C. In radiation the conditions are reversed ; darker soil cools more rapidly at night-time than does lighter-coloured, but does not become colder than the latter. 6. Porosity of soil. A very porous, gravelly soil absorbs the sun's heat rapidly and becomes intensely heated at its surface, but the heat absorbed is equally readily lost by radiation. Soil rich in air conducts heat slowly, the more slowly the greater is the amount of air, because air is a bad conductor of heat. In a rock substratum the conductivity of heat is greater, and varies in velocity with the nature of the rock. Karst limestone, for example, is an excellent conductor of heat, because of its uniform density and its dryness. Granite, basalt, and other crystalline rocks are likewise good conductors. 7. The amount of water in soil is, of all the factors, the one that perhaps has the greatest influence on the temperature of soil, because heat is consumed in the heating and evaporation of the water. Water has a specific heat far greater than that of any kind of soil. The more abundant the water the colder is the soil ; dry soil is more easily heated than wet soil, but soil containing much water, on the other hand, retains heat longer than does dry soil, and for this reason is warmer than the latter in autumn. Sandy soils are ' warm ' because they rapidly lose water and become heated ; clay soils are ' cold '. Soil containing abundant water conducts heat to the subsoil better than dry soil. All these rela- tionships are of profound significance to the development of vegetation in spring for instance. A rock soil is the warmest of all kinds of soil, because no heat is expended in the evaporation of water. Heat penetrates rapidly and deeply into a rock soil, because this is a good conductor of * See Sendtner, 1854 ; see C. Schroter, 1904-8. CHAP. XIII TEMPERATURE OF SOIL 53 « heat. In the deeper layers the temperature extremes are great, whereas in loose soils only the superficial layers are heated.^ Frozen soil, which extends more or less deeply below the surface in Polar lands and on high mountains, naturally plays an important part in relation to vegetation, partly because roots bend away from it as from rock soil (also perhaps by reason of the thermotropism of the roots), and partly because the cold depresses the functional activity of the roots. 8. The texture of vegetation, in particular its density, influences the temperature of soil, because it more or less screens this from direct insola- tion and evaporation, and intervenes in radiation from the soil.^ 9. Internal heat of the Earth. According to Tabert's estimation the mean temperature of the earth's soil is raised by conduction from the internal heat of the earth by 01° C. — an indifferent quantity. Krasan's view of the great importance to vegetation of conduction of the earth's internal heat, is based upon an inadequate appreciation of the climatology of soil. In this connexion it may be mentioned that at Zwickau, owing to the heat liberated from the anthracite which undergoes slow subterranean combustion, it has been found possible to cultivate sub-tropical plants in the open. 10. The cooling of soil by wind is in many cases capable of playing an important part on vegetation. For instance, the vegetation on the coasts of the North Sea may suffer the greatest injury from the north- west wind, and this must be partly due to depression in the activity of the roots, caused by this cold wind cooling the soil. Concerning the relations between the heat of the soil and of the atmosphere, it may be stated that in winter the surface of the soil is warmer than the air only during a few hours at the middle of the day, but at other hours it is a httle colder than the air. Nevertheless the daily mean temperature of the soil is higher than that of the air. Only where there is a covering of snow is the temperature of this surface, also its daily mean temperature, lower than that of the air. In summer, the tempera- ture of the soil during the day is considerably higher than that of the air, but during the night a little lower or rarely higher. Consequently the annual mean temperature of the soil greatly exceeds that of the air. On mountains the maximal temperatures of the soil are nearly as high as in the lowland at their base, whilst the minima are not corre- spondingly lower, so that the excess of the temperature of the soil over that of the air increases with the altitude. The daily fluctuations of the temperature affect the soil to a depth of one metre, descending most deeply in compact kinds of soil that are good conductors of heat. Annual fluctuations penetrate much deeper : in Denmark, for example, to a depth of twenty-five metres, where approxi- mately the mean temperature of the soil remains constant. Thus it follows that the temperature of the soil undergoes greater fluctuations than does that of the air. The fluctuations are greater in warmer kinds of soil. But plants rapidly adjust their vital processes to variable temperatures. A variable temperature that often approxi- mates to the optimum is more beneficial tf) plants than is a constant temperature that remains far below the optimum. • See Homen, 1897. * See Chapters XII and XIX. 54 OECOLOGICAL FACTORS AND THEIR ACTION sect, i CHAPTER XIV. DEPTH OF THE SOIL. THE UPPER LAYERS OF THE SOIL AND THE SUBSOIL Depth of soil, that is, the thickness of the layers of incoherent soil above the solid rock is obviously of great import to plants. Great distinctions in the vegetation denote a shallow soil where rock lies at a very slight depth, whilst deep soil is indicated when this is not the case. Depth of soil affects the temperature, supply of water, amount of nutri- ment, growth of the roots, and so forth. On shallow soil vegetation is more adapted to dryness and is more dependent upon climatic changes than it is in deep soil ; shallow soil produces no such vigorous vegetation as does similar but deep soil, and the vegetation suffers more easily at seasons of drought. A transition from one formation to another may be caused by depth of the soil alone ; for instance, Rikli,^ dealing with Corsica, writes : ' When the soil, poor in humus, becomes still more shallow and consequently drier, the open maquis and garigues gradually give way to typical fell-heath.' In soil a distinction is made between the upper layers of soil (' the soil,' in the narrower sense), and the subsoil. In the former must be included the completely weathered uppermost portion of the soil, which, as a rule more or less intermixed with humus, is subject to the activity of plants and animals, is more influenced by light, heat, and air, and is richer in nutriment, partly because of the absorbent faculty of the soil. By absorbent faculty we mean the character of the soil, and particularly of fine soil, in virtue of which it retains, partly by chemical attraction and partly by surface tension (physical attraction) certain nutritive substances, which are soluble in water, and are filtered by it in such a way that they cannot be washed out by rain-water, or only with great difficulty. These nutritive substances are precisely the ones that are least abundant but most important : phosphoric acid, potash, ammonia ; on the contrary, nitric acid, and for the main part, lime are easily washed out by rain-water. Soil has a noteworthy power, that of regulating the nature of the aqueous solution in itself. This solution is usually very dilute, and its concentration varies according to circumstances. Different kinds of soil have different absorbent faculties. Certain soils, clay for instance, can even abstract nutriment from the atmosphere, in that they can absorb ammonia. The relationship between the upper layers of soil and the subsoil are very important. The depth of the upper layers of soil, the amount of water in them, and their other characters, all play a part ; broadly speaking, it seems that the relationship to plant-hfe is the more favourable the more opposed are the characters of subsoil and soil as regards power of raising water and as regards the amount of water contained. Deherain established the following series : — Light soil with a permeable subsoil is entirely dependent on climate. If this be dry the soil may be extremely sterile ; in a number of places in France there are on such soil coniferous forests, which transpire but little. If the atmospheric precipitations be abundant or the soil be irrigated, it can support tall vegetation. * Rikli, 1903. CHAP. XIV DEPTH OF SOIL 55 Light soil with an impermeaUc subsoil. In a moderately moist climate such soils are of very variable value, according as they are sloping so that water readily flows away, or are horizontal ; the former soils often sustain a rich vegetation, but the latter are very marshy and useless for cultivation. Heavy soil with a permeable subsoil is as a rule fertile, as the excess of water percolates into the subsoil. Heavy soil with an impermeable subsoil supports marsh-vegetation, and requires draining before it can be cultivated. As the constitution of the subsoil often changes from place to place with extreme suddenness, we see the character of the vegetation under- going entire change frequently at very short distances. The slope of the ground may essentially modify the significance of the subsoil, and in general it greatly affects the quality of the soil. CHAPTER XV. NUTRIMENT IN SOIL The plant obtains its nutritive substances partly from air, and partly from the substratum. It is therefore clear that differences in the substratum (edaphic differences) must play a leading part in plant- economy. Water will be discussed in Sections III and IV ; in this chapter we shall deal with soil. Soil in co-operation with the specific activity of the roots, which must be regarded as differing in different species, prepares nutriment, which contains three kinds of constituents : 1. Solid mineral particles ; 2. Salts dissolved in water ; 3. Humus substances. Soil, as was mentioned in Chapter XIV, collects nutriment in its upper layers by means of absorption. Essential nutritive substances in soil. Some substances are indispensable for the completion of the whole normal development of the plant. In the Higher plants hitherto investigated, the elements required are invariably only ten : oxygen, hydrogen, carbon, nitrogen, phosphorus, sulphur, iron, potassium, calcium, magnesium. If one of these in chemical form available to the plant be lacking, then the plant enters into a pathological condition, or entirely refuses to grow. Beyond this, all plants absorb various other substances that are of unknown utihty, and yet cannot be regarded as devoid of significance ; for instance, when present they may so act that certain essential substances are used in smaller quantities than would be the case if they were absent. Amount of nutritive substances in soil. Not only the nature of nutritive substances, but also their amount, is decisive. If a substance be present in a quantity less than a certain minimum the plant will not thrive ; but species vary greatly in their demands ; different species take in different amounts (one of the reasons for the farmer's adoption of a rotation of crops). The practical man distinguishes between poor and rich soil. The amount of soluble salt in soil depends upon — I. The minerals capable of being weathered present in the soil. 56 OECOLOGICAL FACTORS AND THEIR ACTION sect, i 2. The absorbent faculty of the soil, which has already been ex- plained. 3. The climate. Where but little rain falls, the soluble salts produced by weathering, being incompletely washed out, accumulate and may crystallize out, especially on clay surfaces.^ An inadequate supply of soluble salts is unfavourable to plant-growth, but too large an amount of them is also fatal to most species, because the osmotic absorption of water is thereby impeded. The same effect is produced by an abundance of humous acids in the soil. Such types of soil belong to those described by Schimper ^ as being physiologically dry. Plants on physiologically dry soil are often identical with those on arid {physically dry) soil, or are guarded from excessive transpiration by the same protective devices. The quantity and quality of nutriment influence plant-form. Defective nutriment (that is an inadequate supply of one or more substances) may be the cause of dwarf-growth (nanism) ; this has been demonstrated by many physiological investigations, and is shown in natural vegetation, for instance, on sand-fields and other poor soils. Dwarf-shrub is a growth- form characteristic of soil poor in nutriment, and particularly of heath. The amount of a single substance may determine the issue. It is a general rule that the size of a crop, in so far as it is dependent upon nutri- ment, is determined by that nutritive substance which is available to the species concerned in relatively the smallest quantity (Liebig's Law of the Minimum). When a nutritive substance occurs in so small a quantity that the crop is decreased for this reason, then, according to Atterberg's rule, the substance in question is present in the plant in relatively smaller amount than are those nutritive substances of which there is no deficiency ; it is then easy to surmise that other, morphological distinctions, may also thereby arise. The form of the root is adjusted to the characters of the soil. According to the investigations of Sachs ^ the more concentrated the nutrient solution the shorter are the roots. Roots are mostly long and feebly branched in poor soil, for example in plants on sand, especially on dunes, in Central Europe ; nevertheless the majority of heath-plants show the contrary. Roots branch very copiously and form dense clumps in rich soil. If roots encounter strata of soil with different quantities of nutriment, the contrasts in the ramification within the different strata are striking. " Roots search for food as if they possessed eyes.'* The chemical constitution of the nutritive substratum in certain cases evokes formal differences. This is particularly true of one substance, com- mon salt. It is recognized that all halophytes are distinguished by a special configuration ; they have fleshy leaves, transparent tissue, and so forth .^ The effects of calcium carbonate and other substances are less obvious. Distinctions in soil have probably led to the separation of new species : The calamine violet (Viola calaminaria) is presumably a form that has arisen from Viola lutea by the action of zinc in the soil.^ ' Hilgard, 1892. * Schimper, 1898. ' Sachs, J. von, 1865, p. 177. * Liebig. * See Section VII. ' Schimper, 1898 (1903, p. 93). CHAP. XV NUTRIMENT IN SOIL 57 On serpertine, a silicate of magnesium, there grow two species of Asplenium, A. Serpentini and A. adulterinum, which are closely alhed with A. Adiantum-nigrum and A. viride.^ These new forms of Asplenium have not yet become fixed, but in other cases fixation has probably taken place, so that only very prolonged action could reform them, if indeed it could do so at all. According to Kerner's - researches in the Alps, there exists a wide difference between parallel species occupying limeless slate and limestone mountains ; such parallel species, or, better perhaps, races, are the following : — Calcicolous Hutchinsia alpina Thlaspi rotundifolium Anemone alpina J uncus monanthos Primula Auricula Ranunculus alpestris Not Calcicolous Hutchinsia brevicaulis Thlaspi cepeaefolium Anemone sulphurea J uncus trifidus Primula villosa Ranunculus crenatus Dolomite Androsace Hausmanni Asplenium Seelosii Woodsia glabella Not Dolomite '^ Androsace glacialis Asplenium septentrionale Woodsia hyperborea.^ Since such species as replace each other on different soils are certainly derived from one parent species, it becomes of interest to ascertain wherein they differ from each other, because the effects of the soil will presumably be revealed. The first experimental investigations on the action of calcium were, according to Schimper, made by Bonnier.^ Observations in the open air were conducted by Fliche and Grandeau, and others.^ Kerner observed in parallel forms the following distinctions : Calcicolous Plants more strongly and densely clothed with hairs ; often clothed with a white or grey felt. Leaves often bluish green. Leaves more divided and more deeply so. If the leaves be entire Corolla larger Flowers mostly with duller surface but lighter hue. Not Calcicolous Hairs glandular. Leaves grass-green. Then the leaves not uncommonly glandular-serrate. ' See Schimper, 1898 (1903, p. 93) ; also Pfeffer, 1897-1904. ' Kerner, 1869. " Kerner, 18636. * Blytt doubts if the Norwegian Woodsia glabella be the dolomite form of W. hyperborea ; it occurs not only on dolomite, but also on slate. ' Bonnier, 1894. ' See Schimper, 1898 {1903, p. 95). 58 OECOLOGICAL FACTORS AND THEIR ACTION sect, i Although the characteristics of the Hme-flora are clear and distinct, yet in the past the influence of lime upon vegetation has been over- estimated. Indeed, a distinction has been made between calciphilous and calciphobus plants.^ Recently it has been definitely estabhshed that the amount of lime in itself, in so far as it does not operate physically, cannot be the cause of differences in the flora, for not only can calci- colous plants be cultivated in soil that is poor in lime, but sihcicolous plants, and even bog-mosses, which are regarded as pre-eminently calci- phobous, can grow vigorously in pure lime-water - if the aqueous solution be otherwise poor in dissolved salts. It has been overlooked that nearly all lime soils are rich in soluble mineral substances, and this wealth ex- cludes plants belonging to poorer soils ; beyond this the important physical characters of calcareous soil, compared with granite soil, come into play. Geographical significance of nutriment in soil. The nutritive substances indispensable to the higher plants occur in nearly all soils — certain ones, such as quartz-sand, being excepted — in quantities so considerable that in this respect there is no obstacle to prevent any plant growing almost anywhere on earth. It must be remembered that even when a nutritive substance is present in the substratum in very small quantity, the plants requiring it can yet absorb it in large quantities ; for instance, species of Fucus accumulate a great deal of iodine, though extremely little of this is contained in sea-water. The plant has a certain power of quantitative selection, in that it absorbs various substances in proportions other than those in which they occur within the substratum. There are, however, substances which exert a poisonous action on certain plants and exclude these from soils that contain a large amount of them. This is perfectly comprehensible when it is remembered that the plant can select its nutriment only to a certain extent. The larger the amount of a substance in the soil, the more of it, as a rule, is absorbed by the plant ; and in all cases substances that are useful or even essential in small quantities, may be absorbed to excess or act as poisons. Substances of this nature are common salt and ferrous salts. But a certain amount of latitude prevails in this matter, for one and the same species absorbs the various nutritive substances from various soils in different proportions. Indi- viduals of the same species on granite-soil contain much silica, and on calcareous soil much lime. Finally, it may be noted that certain sub- stances, for instance lime and magnesia, can replace each other to some extent. It is of profound significance to communities of plants that each species has its own peculiar economy, the nature of which is almost unknown to us ; for in virtue of its metabolic activity and the attributes of its root-system each absorbs substances in proportions different from those prevaihng in other species. For the communal life of species it is also of importance that substances are not absorbed at the same rate and time, or at the same ontogenetic stage. This renders it possible for many species to live side by side on the same soil without entering upon a struggle for food. Partially dependent upon this is also the system of ' rotation of crops '. ^ Sendtner, i860; Contejean, 1881. ^ See C. A. Weber, 1900 ; Griibner, 1901 ; and the critique of this by F. E. Clements, 1904. CHAP. XVI . ROCK AND SAND 59 CHAPTER XVI. KINDS OF SOIL In accordance with the different constitution of soil the following main kinds may be distinguished : — Rock soil Clay soil Sand soil ■ Humus soil Lime soil Saline soil These are connected with one another by gradual transitions and countless admixtures, so that in reality there exist innumerable varieties of soils of multifarious character. As, however, the kinds of soil above named have extremely different properties, and therefore necessarily support plant-communities very different oecologically, their charac- teristics are briefly recorded here : 1. Rock soiL In this case it is the nature of the rock that determines what vegetation can develop upon it. And in this there come into play distinctions in the hardness, porosity, specific heat, and thermal conducti- vity. Among the most important kinds of rock are : granite, gneiss, limestone, dolomite, sandstone, slate, basalt.^ 2. Sand soiL Sand consists of loose particles of various minerals, mainly of quartz, but also of felspar, hornblende, mica, even hme (for example, in coral-sand), volcanic products. The nutritive value of sand varies with the chemical nature of its particles ; pure quartz-sand is sterile, because particles of quartz are incapable of acting as nutriment ; sands containing hme, mica, felspar, and other minerals, have a greater nutritive value. Humus is formed only with difficulty in dry, loose, sandy soil, because in this organic bodies are easily decomposed and oxidized by the admission of air. In addition, sand, particularly quartz-sand, which is the most frequent kind, has only a slight absorbent faculty in regard to substances nutritious to the plant. Sand is loose soil, because its particles have but little cohesion, and this diminishes as the particles increase in size. Atmospheric precipita- tions easily percolate through sand, and with a facility that is greater the larger are the particles. Sand generally contains only a small amount of water ; the coarser are its grains the less water does it retain — approxi- mately from three to thirty per cent. ; dune-sand from Bordrup in Jutland, according to Tuxen, takes up twenty-seven per cent. The power of sand to raise water from the subsoil is as a rule very slight ; water is usually raised at most one-third of a metre. Sand dries as a rule very quickly, and consequently becomes rapidly and intensely heated in sunlight, but it also cools very rapidly and intensely at night. The surface of a shifting sand-dune becomes covered by a dry layer of sand, which becomes strongly heated in sunlight, yet though this layer is of but slight depth it hinders evaporation from the subjacent sand, which consequently always remains moist and cool ; this considera- tion is of very great importance for the proper understanding of dune- vegetation. In the deserts of Arizona, according to Livingston,^ a powdery superficial layer appears to act in like manner. The difference * See Section VIII. * Livingston, 1906. 6o OECOLOGICAL FACTORS AND THEIR ACTION sect, i between the temperature by day and by night may be very wide (from forty to forty-five centigrade degrees). As a result, at night-time dew is readily and richly deposited upon sand, whose water-content and vegetation are thus profoundly affected. On the other hand, plants suSer from frost more easily on sandy soil. The sand-flora develops early in spring — a feature that recalls the steppe. Plants habitually growing on sand are usually termed psammophytes or psammophilous plants-^ ^ 3. Lime soil. Calcareous sand, consisting of calcium carbonate, is less poor in nutriment than is quartz-sand, has a greater water-capacity, and dries less rapidly, but is nevertheless dry and warm. Mail is an intimate admixture of calcium carbonate (8 to 45 per cent., in calcareous marl 75 per cent.), clay (8 to 60 per cent), and quartz-sand : the lower diluvial marl from the Mark Brandenburg contains calcium carbonate 12 to 18 per cent., clay 25 to 47 per cent., quartz-sand 38 to 62 per cent. The characters of marl depend upon the relative proportions of the constituents, and generally stand between those of sand and clay. 4. Clay soil. This offers an almost complete contrast to sand soil. The particles that are invisible to the naked eye predominate over large grains. Clay consists mainly of kaolin (hydrated silicate of aluminium), and may contain more or less quartz, calcium carbonate, ferric oxide, and so forth. Kaolin is of no nutritive value to the plant, yet the presence of many additional substances may render clay very rich in nutritive bodies ; these are, however, available only with difficulty. With a favour- able admixture of sand, lime, and humus, a clay soil is a fertile soil. Clay soil has a large absorbent faculty, and is at the same time very hygroscopic ; it can absorb five or six per cent, of aqueous vapour from the atmosphere. Clay soil is tenacious or heavy, as its particles have great cohesive power ; aeration is mostly defective, a circumstance unfavourable to vegetation, and leading to the production of acids and swampiness. Clay soil is wet and cold, because its water-capacity is great (up to ninety per cent.), and because its capillary power is great ; it raises much water from the subsoil, and is almost impermeable to water. If we over- load it with water it swells, its volume increases, and its individual particles are forced asunder, so that a paste is formed. Clay soil containing an abundance of water is plastic. After prolonged drought, clay soil acquires a stony hardness, contracts, cracks ; and these occurrences react on vegetation.^ The unfavourable characters of clay may be ameliorated by com- mixture with substances of opposite character, such as sand and lime. Loam, which may be dealt with in connexion with clay, is weathered marl, the calcium carbonate of which has been more or less dissolved in water, and the ferrous salts of which have been converted into ferric oxide and hydroxide ; the soil consequently becomes brown and essentially contains clay and quartz-sand. 5. Humus soil.3 Humus is produced from the remains and products of plants and animals, often from animal excrement in all stages of decomposition, and is mixed in soil with various proportions of mineral ^ In Section X. " See page 42, Chapter X. ' Consult the important work by Friih and Schroter, 1904. CHAP. XVI HUMUS AND PEAT 6i constituents. Humus is black or brown, and rich in carbon, also some- times in nitrogen — the Russian ' Black Earth ' contains, according to Kostytscheff, as much as from four to six per cent. In the production of humus a prominent part is played by micro-organisms (bacteria, monerae, and the hke), and by larger animals, particularly earthworms. Humus substances combine with nutritive bodies that are soluble with difficulty to form easily soluble compounds, and thus they increase the nutritive value of soil. They also change the physical characters of soil ; when mixed with mineral soil they increase its absorbent faculty, its specific heat, its water-capacity, and so forth. There exist wide distinctions among humus soils, according to the degree of decomposition, and according to the species of plants and animals engaged in producing the humus. The first of these soils that we shall deal with is that which is richest in humus : (a) Peat soil. If water containing oxygen come into contact with organic bodies it is robbed of its oxygen by these. If then the admission of oxygen be prevented and the activity of micro-organisms excluded, or at least restricted, in many cases an incomplete decomposition and change of the organic substances ensues ; as a result, carbon will accumu- late in larger quantities the more the supply of air is restricted, and free humous acids make their appearance : there is a production of j)eat. Peat is humus rich in carbon, brown (from a light to a black-brown) in colour, and contains jnany free humous acids, and other acids which are contained in the remains of organisms buried in the peat. The organic portions (50 to 90 per cent.) of peat are chiefly plant-remains, which are readily recognizable ; animal-remains, on the other hand, are quite subordinate. By the removal of water and admission of air, peat can be changed into a humus that is well suited to plants. Peat contains 1-2 (-3) per cent, of nitrogen and 0-4 per cent, of lime (certain peats, for instance that of moss in Gothland, are stated to have up to 3.21 per cent, of nitrogen and much hme), but peat contains very little potash and still less phosphoric acid. The amount of these important nutritive substances is so small because the acids in peat combine with alkalis to form soluble salts, which are washed away. Peat soil has the following characters : — Of all soils it has the greatest water-capacity, so that it can take up much more water than its solid parts weigh ; air-dried peat contains only 15 to 20 per cent, of water ; peat swells on the addition of water to a far greater size, but contracts on drying and becomes cracked without however crumbling to pieces. When it is completely dried it becomes extremely loose, almost powdery (peat-dust ; driving peat-dust may be compared with driving sand). If one reckons the tenacity of clay as one hundred, that of peat is only nine. It is almost impermeable to water, and its power of raising water exceeds that of all other soils. It is powerfully hygroscopic (absorbing up to ten per cent, of aqueous vapour). In regard to their power of conducting water, different peats (for example those of heath-moors and meadow-moors) behave very differently. Sphagnum-peat conducts water rapidly and is therefore uniformly moist in all cases, but meadow-peat may be dry above and wet beneath. On account of its dark colour peat-soil is strongly heated by the sun, but is intensely cooled at night. Despite its dark colour peat 62 OFXOLOGICAL FACTORS AND THEIR ACTION sect, i forms a cold soil, because it is usually rich in water. Neither bacteria that produce nitrates, nor other bacteria, nor earthworms, can thrive in peat, because of its acid contents. Further details in regard to peat will be given in Section VI. (b) Raw humus is a ' production of peat in the dry ' ^ a black or black-brown peat-Hke mass, which is built up of densely interwoven, incompletely decomposed plant-remains, consisting of roots, rhizomes, leaves, mosses, fungal hyphae, and the like. Certain plants in particular give rise to raw humus, because they bear very thin, numerous, richly- branched roots (or rhizoids), which lie at the surface of the soil and weave the plant-remains into a dense felt work ; such species are, for example, Fagus, Calluna, Vaccinium Myrtillus, Picea excelsa. According to the constituent forming the main mass, we speak of heath (Calluna) raw humus, moss raw humus, beech raw humus, spruce raw humus, silver-fir raw humus, oak raw humus, pine raw humus — and so forth.^ In plants growing on raw humus mycorhiza is frequent. Raw humus may be so rich in plant-remains as to be employed as fuel (heath-peat) ; it may contain from fifty to sixty per cent, of organic matter. As it forms so dense and tough a felt above the mineral soil, on the one hand, it excludes air (oxygen) from the subjacent layers and, on the other, sucks up water as greedily as a sponge and holds it with great force ; in rainy European climates it is frequently wet for a large part of the year. Consequently in it, as in peat, free humous acids are produced in abundance. Like peat, it has an acid reaction. There occur in it only few animals, mostly Rhizopoda and Anguillulidae, but no earthworms. The part played by bacteria has not yet been ascertained. Raw humus appears in forest, especially in places exposed to wind, whilst ordinary humus, with its earthworms and other animals, reigns in fresh places sheltered from desiccation ; when ordinary humus in beech-forest has given way to raw humus, because of timber-falls and such like, then the beech, being no longer capable of regenerating, disappears, and is often replaced by Calluna-heath.^ The production of raw humus is linked with lowness of temperature, and is promoted by moisture.* The formation of a layer of raw humus also induces in the constitution of the subjacent layers of soil great changes, which have become best known through P. E. Miiller's ^ pioneer researches in Denmark, the main results of which are given in the succeeding paragraphs.^ From the raw humus, humous acids and their compounds descend with rain-water into the subjacent sand, which has been more thoroughly washed out and is poor in soluble salts ; here they are oxidized by contact with inorganic (particularly ferric) compounds rich in oxygen, and there arise, for example, freely soluble ferrous compounds, which are carried by water containing carbonic acid down from the upper layers of soil. These layers are consequently decolorized, lose their absorbent faculty ' P. E. Miiller, 1878, 1884, German edition (1887, p. 45). * P. E. Miiller, in the German edition of his worli, employs the terms heath-peat, moss-peat, beech-peat, etc., in this connexion. ' P. E. Miiller, loc. cit. * Ramann, 1895, p. 125. ' P. E. Miiller, loc, cit. ^ Also see Ramann, 1886, 1905 ; Warming, 1896 ; Friih and Schroter, 1904. CHAP. XVI HUMUS 63 almost completely, become poor in nutriment, and thus there is formed under the raw humus a light-grey or black ' bleached sand '.^ As raw humus dries, some of the humous substances that were originally freely soluble become scarcely soluble, and are precipitated as carbonized humus. Movements of the water also carry down particles of clay, ferric oxide, and humus, which are soluble only in water containing but httle salt, and convey these through the layer of sand that is poor in salt until, at the lower limit of this washed out sand, they meet with the particles of soil which are still undergoing disintegration, and which therefore still contain soluble salts. The water takes these salts into solution, and the humous acids are precipitated as a gelatinous mass, which at a definite degree of dryness becomes solid, possibly by chemical changes, and insoluble in water. The grains of sand are cemented together, and there is formed a reddish-brown or brow^n layer of earth known as hard pan or moor-pan, which may be half a metre in thickness, and when completed is impervious to plant-roots. The change from ordinary humus soil to raw humus is brought about by the following means : — (i) Plants with densely interlaced roots occur ; (2) Animals, particularly earthworms, vanish, so that the soil is not worked ; (3) Particles of soil, particularly grains of sand, sink down and leave the soil more compact and poorer in air. (c) Ordinary humus (leaf-mould, garden-mould, vegetable mould-) is an intimate mixture of sand and clay with completely disintegrated organic ingredients — rarely more than ten per cent. The mixture comes about mainly through the agency of animals and water.^ It is devoid of free soluble humous acids, and is neutral or alkaline in reaction. It contains many fungal mycelia, earthworms, insects, and so forth. That ordinary humus soil is so excellent a nutritive substratum for plants is due to — Its physical attributes (loose, having compound particles, aerated). Its chemical characters, as it contains many compounds of carbon and nitrogen. The humous substances, forming freely soluble compounds with nutritive substances that are otherwise scarcely soluble. The production of humus in forest partially replaces the manuring and amelioration of the soil in agricultural practice. Factors that strongly promote decomposition of organic rnatter hinder the production of humus ; according to Wollny, heat and moisture are the factors of greatest import. In connexion with both these, as in all other physiological processes, there exist a minimum, an optimum, and a maximum. Temperatures above the maximum are almost devoid of significance ; moisture beyond a moderate amount may exclude air from clay and humus soils, and thereby exert a restrictive action upon decomposition even before the soil is saturated with water. In lower latitudes decomposition proceeds very slowly during the dry season, but at the wet season it is greatly accelerated, and in most spots is so complete that a soil very poor in humus results.' In the • See Albert, 1907. * Darwin, 1887. ' See Chapter XX. * Hilgard, 189.-. 64 OECOLOGICAL FACTORS AND THEIR ACTION sect, i tropics and sub-tropics true humus soil occurs only in shady forest ; ^ peat soil is very rare, but still it occurs where the cHmate is sufficiently moist ; ^ typical moors are wanting.^ In steppe and desert likewise the soil is mostly poor in humus, because plant-remains are scanty, though the soil is sometimes sufficiently moist. Only in most richly clad grass-steppes is humus {' black earth ' occurring in South Russia) formed, especially upon closely deposited loess soil.* In cold-temperate lands vegetable mould is the most frequent kind of soil. Only in sunny open localities exposed to wind, such as dunes, do we find the soil with scanty humus. Raw humus is frequent wherever decomposition is restricted from any cause, which according to Ramann ^ may be lack of nutriment, exclusion of air, excess or lack of water, or lowness of temperature. The formation of raw humus is of specially wide distribution in the heaths of maritime Western Europe where the summers are cool, as well as in alpine and arctic situations.^ Different species of plants demand very different amounts of humus in the soil. Accordingly Kerner has ranged plants into three groups : (i) Plants which can settle upon bare rock, the most barren, sandy, or gravelly surfaces, and other spots where there is not a trace of humus ; their seeds or spores are mainly transported by wind : sub-glacial plants, many tundra-plants, desert-plants, and the like. (2) Plants requiring a moderate amount of humus : for instance, Gramineae and Cyperaceae. (3) Plants thriving only in rich humus, in the remains of a previous vegetation : many Orchidaceae, species of Pyrola and Lycopodium, Azalea procumbens, Vaccinium uliginosum, a number of other moorland plants, hemisaprophytes, and finally the highly modified holosaprophytes, Monotropa, Neottia, and others. We may regard it as certain that there is a correlation between the unusual forms of the last-named plants and their method of nutrition, and thus between their forms and the kind of soil upon which they live ; but beyond this we know nothing of the matter."^ 6. Saline soil is soil of varied constitution (sandy, clayey, and so forth) that is heavily charged with sodium chloride. It will be treated in detail in Section VII. Soils at the bottom of water. Deposits and varieties of soil are formed here. In the sea, fine particles of mud are accumulated by the action of animals and Blue-green Algae, in places where the water is calmest, and they form the foundations of the fertile marshes on the coasts of the North Sea.^ Mud of another kind is raised up in mangrove- swamps. On many coasts and at the mouths of many rivers there arise masses of mud that are rendered deep-black by sulphide of iron ; accord- ing to Beijerinck and Van Delden,^ anaerobic bacteria play a part in the production of iron sulphide. ^ See Warming, 1892 a; Vahl, 19046. ^ Ule, 190 1. ' See Friih and Schroter, 1904, p. 143. " Albert, 1907. * Ramann, 1893, 1895. ' Kerner, 1863 ; Warming, 1887. ' For further information on moors and peat, readers should consult the great work by Friih and Schroter (1904), which has already been cited. * Wesenberg-Lund and Warming, 1904, Chapter 58. ' Beijerinck, 1895 ; A. van Delden, 1903; Wesenberg-Lund and Warming, 1904. CHAP. XVI SOIL FORMED UNDER WATER 65 In fresh-water, owing to the advent of material brought by rivers, there arise deposits whose nature is determined by the geological character of the soils traversed by the river. ' Pollen-mud ' arises by the accumu- lation of pollen derived from anemophilous trees, such as conifers, beech, and the like. Other deposits are produced by chemical processes, more or less due to the activity of organisms. Calcic carbonate : In many spots in lakes a portion of the carbonic acid in the water is decomposed mainly, it appears, through consumption by green plants. The result is a precipitation of calcic carbonate, first, on plants (Potamogeton, Characeae, and certain algae), partly in their cell-walls ; and subsequently the precipitated chalk accumulates on the lake-bed. Characeae may contain as much as 80 to go per cent, of chalk. Compounds of iron are very often deposited, with or without the co-operation of bacteria. Blue-green Algae, and other plants. Deposits mainly constituted of the remains of plants and animals occur particularly in calm lakes, pools, and the like. They appear to be most frequent in cold-temperate regions. Some (termed ' Gytja ' in Scandinavia) are structureless, grey or brown masses, largely composed of the more or less disintegrated excrement of animals, also of the remains of small animals and plants. A large amount of fatty oil occurs in these deposits, according to Potonie. A very important constituent of this kind of soil comes from plankton. In lakes where diatoms abound these are deposited in great quantities. In other cases it is mainly Blue-green Algae, or the chitin of small freshwater crustaceans that are accumulated. By chemical processes these common deposits undergo change as the organic constituents are reduced. This ' Gytja ' is a kind of humus- production under water. Another kind of deposit that is amorphous, jelly-hke and brown when moist, but more black when dry, arises especially in shallow waters where the water is brown with humus and the vegetation is rich ; with this type of deposit nymphaeaceous vegetation is particularly asso- ciated.^ CHAPTER XVII. ARE THE CHEMICAL OR THE PHYSICAL CHARACTERS OF SOIL THE MORE IMPORTANT ? We have already learnt that there are numerous differences in the chemical and physical characters of soils, that is to say, in the amount and kind of the components, and in the water-capacity, tenacity, and so forth. Varied combinations of these bring into existence extremely diversified types of soil. ' Our knowledge of these various deposits, connected as they are with one another by very gradual intermediate types, is very defective. They were first studied by H. von Post, 1862 ; and more recently investigated by Ramann, 1895 ; Weber, 1903 ; Potonie, 1905 ; Fruh and Schroter, 1904; Wcsenbcrg-Lund, 1901 ; Ellis, 1907. WARMING F 66 OECOLOGICAL FACTORS AND THEIR ACTION sect, i Many species of plants are very indifferent as regards soil, inasmuch as they can grow on widely dissimilar kinds, for example : Phragmites communis grows both in very saline and in fresh water ; and Typha latifolia is said by Sickenberger to be capable of growing vigorously in the soda-lakes of Egypt, where a nearly saturated solution of salts is present. Many ubiquitous or cosmopolitan species display but little preference, yet most species are confined to soil that has quite definite physical and chemical relationships. Long ago it was noticed, particularly in mountainous countries with a substratum of varied geognostic nature, that the distribution of species and the whole appearance of the vegetation show a certain correlation with the soil. As an example we may select the case treated by Petry ^ of the Kyffhauser Hill, where there is a sharp contrast between the vegeta- tion on the rothliegende and that on the zechstein, not only in the forest and in its undergrowth, but also in the weed-flora, the vegetation of the sunny dry heights, and of the copses. The rothliegende, as a conse- quence of its poverty in nutriment, supports a meagre, uniform vegetation, partially agreeing with heath ; the zechstein tract, on the other hand, has beech-forests and a herbaceous flora of many species. The contrast between the two sub-formations is so sharp that we can detect from the vegetation, either in forest or field, whether we are standing upon the one or upon the other ; and the relations are such that this contrast must be attributed to conditions prevailing in the soil. Similarly in Montpellier,^ in Switzerland,^ and in many other moun- tainous lands we can observe the most emphatic contrast in the vegetation of two contiguous tracts, and even in a moraine-country like Denmark the same can be observed. In Jutland we can see sharply delimited patches with the Weingaertneria-association (Corynephorus-association), containing Weingaertneria canescens, Trifolium arvense, Scleranthus, Hieracium Pilosella, and others, dotted about a tract which hkewise has a poor arable soil, but which supports an entirely different vegetation composed of Leontodon autumnale, Jasione, Lotus corniculatus, Erigeron acris, Euphrasia officinalis, Trifolium pratense, T. repens, Achillea Mille- folium, Chrysanthemum Leucanthemum, Equisetum arvense, and others : in the former areas there are no mole-hills, whereas in the latter there are many. The reasons for these generally observed distinctions have been sought chiefly in two directions. Some authorities regarded the chemical constitution of soil as the decisive factor, while others laid greatest stress upon the physical characters, and particularly upon the relations prevailing in regard to heat and moisture. The main points in the discussion are the following : — The dominating influence of the chemical constitution of soil. One of the earliest advocates of the chemical theory was the Austrian, Unger. He directed special attention to the contrast between calcareous and siliceous or slate soils, and he ranged plants in three groups : Indifferent to soil, are those plants unaffected by the chemical nature of the substratum. ^ Petry, 1889. ' Flahault, 1893. ' Magnin, 1893: CHAP. XVII INFLUENCE OF CHEMICAL CONDITIONS OF SOIL 67 Partial to a certain soil, are those that show a preference for it without however being strictly confined to it. Restricted to a certain soil, are those Umited to it. In accordance with this we can distinguish : calcicolous plants, silicicolous plants, slate-plants, halophilous plants, and so forth.^ Of other botanists who likewise assume that the chemical constitu- tion of soil has a controlhng influence, we may mention, among Germans, Sendtner, Schnitzlein, Niigeli ; and among Frenchmen, Vallot, Fliche, Grandeau, Saint-Lager, Contejean (in later years), Magnin. Upon the whole, French investigators appear, in recent times, mainly to support this view. Various facts favour this interpretation. On p. 58 it was pointed out that certain substances in excess act on certain plants as poisons. This is seen most clearly in the case of common salt. Halophilous plants. Halophytes are not only of highly characteristic morphological and anatomical architecture, but have an absolutely defined topographical distinction on coasts, and in salt deserts and salt steppes. Common salt in excess has a highly exclusive action ; it acts as sterihzer, and only relatively few species, mainly belonging to definite families (Chenopodiaceae, and others), can endure much of it. Section VII should be consulted for further particulars in reference to halophytes. Calciphobous plants. In cases of other substances, lime for instance, the matter is more doubtful. Lime is essential to the plant. Certain plants are stated to avoid soil containing much calcium carbonate.^ Such reputedly calciphobous species are : Castaneasativa, Pinus maritima, Calluna vulgaris, species of Erica, Sarothamnus scoparius. Genista anglica, Ulex europaeus, Pteris aquilina, Rumex Acetosella, and other plants that we often find on heaths and on raw humus ; also Gramineae, Cyperaceae, many lichens and mosses, especially Sphagnum,^ and among algae the Desmidiaceae. The flowering plants named are reputed to be incapable of carrying on an existence in soil containing more than from 0-02 or 0-03 per cent, of calcium carbonate. But cultures made by C. A. Weber * and Grabner have clearly demonstrated that none of these plants suffer from lime when this is unaccompanied by a large amount of soluble salts. Calciphilous plants. Other plants that do not desert a soil rich in calcium carbonate are put forward as calciphilous plants, for example : Papihonaceae (Trifohum, Anthylhs, Vulneraria, Ononis Natrix, and others), Rosaceae, Labiatae, many Orchidaceae, Tussilago Farfara, and others. Unger gives a whole array of examples belonging to the hme-flora. Accord- ing to Blytt^, Ophrys muscifera and Libanotis montana are the sole vascular plants in Norway that occur exclusively on calcareous soil. Among algae the Mesocarpaceae are calciphilous. Silicicolous plants. These are brought forward in contrast to calci- colous plants. The calciphobous plants mentioned above are regarded as silicicolous. The truth may perhaps be that they are expelled from ' See Chapter XV, pp. 56-8. ' See p. 58. * Fliche et Grandeau, 1888 ; sec Contejean, 1893. * C. A. Weber, 1900. ' BIytt, 1893. F 2 68 OECOLOGICAL FACTORS AND THEIR ACTION sect, i calcareous soil by competition, and are compelled to select a soil containing less calcium, without having any real preference for silica, which is a very neutral substance ; thus Contejean interprets the matter. To the siUci- colous plants belong the majority of those growing on sand and moor in north-temperate Europe. Nitrophilous plants (nitrophytes, ruderal plants). These thrive best in soil where compounds of ammonium and nitric acid are abundant, and therefore especially in the vicinity of human dwellings (dung-heaps, highly-manured soil). They belong to certain special famihes (Chenopo- diaceae, Cruciferae, Solanaceae, and others), and nitrates occur in vheir cell-sap. Other species develop feebly on such soil, because they take into their tissues more nitrate than they can endure.^ Certain fungi and mosses (Splachnaceae) flourish only on dung. The solfataras of Java, according to Holtermann's assertions, have a pecuUar flora that differs from those of others.^ Other substances also can act as poisons if they be supplied in large quantities. If gypsum be scattered over a meadow certain ferns and grasses die off, while clover becomes more luxuriant. Similarly iron (iron sulphate, ferrous oxide) may act injuriously if present in quantity, though it is one of the absolutely indispensable nutritive elements. Investigations conducted at Rothamsted in England have demon- strated in a particularly clear manner the significance of the chemical constitution of nutriment ; they showed that with nitrogenous manure, especially with nitrates, grasses preponderated and expelled Leguminosae ; whereas, on the contrary, potassic salts favoured Leguminosae. Experi- mental manuring of high moors has, according to Weber, led to entirely similar results ; certain species of grasses were expelled by others. But one can hardly say that research has yielded any considerable sup- port to the chemical theory we are discussing ; calcicolous and silici- colous plants, the calamine- violet, and even halophytes, are perhaps always capable of flourishing in a soil not containing the respective substances they affect, or practically in any soil, in botanic gardens for instance. On the other hand, the amount of nutriment in soil plays a more prominent part. In course of seven years' wanderings, A. P. de Candolle found nearly all species upon soils of varied chemical nature ; and Blytt came to the conclusion that the very few species which in 1870 he had regarded as restricted to definite soils in Norway must be further reduced in number as a result of his later investigations. ' Every distributional relationship may be due to either a physical or a chemical cause, but the simultaneous presence of both prevents us from clearly distinguishing the part played by either singly'.' This is perfectly correct, and the history of botanical science shows that some botanists, in opposition to those previously mentioned, ascribe greater importance to physical than to chemical relations. ^ Schimper, 1 890-1. ' Holtermann, 1907. ' Vallot, 1 83 1. CHAP, xvir INFLUENCE OF PHYSICAL CONDITIONS OF SOIL 69 The dominating influence of physical characters of soil. The protagonist for the dominating importance of physical relation- ships was Jules Thurmann (1849). His doctrine can be summarized as follows : — It is the physical structure of soil that regulates the distribution of species ; Upon this structure depend the amount of water and the thermal conditions in soil. The same species can grow on very different kinds of soil, if it encounters the same conditions of moisture. Thurmann discusses the different weathering properties of kinds of rock under the action of air, water, and heat (both frost and warmth), As a result he di\'ides rocks into eugeogenous and dysgeogenous. Some kinds of rock are easily weathered and rapidly produce loose masses (grit, sand, and similar detritus) ; these soft types are the eugeo- genous, and in accordance with the fineness of their products of weather- ing they are pelogenous when the particles are very fine and powdery, especially clay and marl soils ; or psammogcnous, when the particles are coarser — sand. According as soil is more or less pelogenous or psammo- gcnous, Thurmann employs the prefixes ' per ', ' hemi ', and ' oligo ', to denote sub-divisions, or speaks of pelopsammitic soil. In opposition to types of rock that are easily weathered, are the hard, resistant types — dysgeogenous ; they give rise to scanty or no products of weathering. Finely comminuted soil absorbs more water than does slightly weathered rock.^ Eugeogenous rocks therefore bring into existence a moist, cold soil ; dysgeogenous, a dry, warm soil. To plants that exploit moist soil and eugeogenous land Thurmann applies the term hygrophilous ; plants that exploit drier soil and dysgeo- genous rock he terms xerophilous. His hygrophilous species correspond approximately to the silicicolous plants of Unger and others, and his xerophilous species to their calcicolous ones. Indifferent species of plants occurring on all kinds of soil Thurmann designates uhiquists. According to Thurmann, the obvious distinction between the floras on calcareous and siliceous soils is caused, not by the preference of species for lime or silica, but by the circumstance that calcareous rocks weather with difficulty, and permit water to flow away rapidly through clefts and fissures ; they produce a dry warm soil of slight depth, whereas quartz and felspar produce a loose, deep, moist, and cold soil. When species of rock, with identical chemical composition, in some cases are hard and resistant but in others become easily disintegrated, then calcicolous plants are found on the former soil even when it is sihceous, and silici- colous plants on the latter soil even when it is calcareous. Furthermore, a single species of plant in a definite climate may require a definite soil on account of the physical characters of the latter ; for instance, in a moist chmate it may choose a warm dry soil like lime ; but in a different climate it may prefer an entirely different soil ; for instance, in a warm dry chmate a moist, cold, sihceous soil. A favourable soil may facilit;ite the existence of a plant in an unfavourable climate. According to Blylt, * See Chapter XII, p. 47. yd OECOLOGICAL FACTORS AND THEIR ACTION sect, i many species in Norway reach their extreme stations in altitude and latitude on calcareous soil. Eugeogenous and dysgeogenous kinds of rock may bear the same flora. It is thus that the distribution of beech in Southern France must be explained. In Denmark it passes for a calciphilous plant ; yet in Mediterranean countries, according to Flahault/ only upon siliceous soil does it form extensive forests, while on dry, warm, calcareous soil it is sporadic, having been expelled by Quercus sessiliflora, excepting in cool valleys with north and east exposure. As supporting Thurmann's theory we may name Contejean, who however subsequently adopted the rival theory ; and approaching of nearly similar views are A. P. and Alph. de Candolle, Celakovsky, Krasan, Hoffmann, Kerner, H. von Post, Blytt, P. E. Miiller, Negri, and others.^ Yet Thurmann's theory certainly cannot explain all cases. In both theories there is some truth ; both chemical and physical relations operate ; the actual truth seems to be that in some (few) cases, where the soil is specially rich in a chemical substance, it is the chemical characters of the soil, but in other (far more frequent) cases it is the physical characters that are of greatest import. When we consider a country like Denmark or the North German Plain, with a soil produced by corrosion and commixture of multifarious kinds of rock, yet scarcely- possessed of any marked chemical characters, then the chemical signifi- cance of soil becomes evident in the halophytic vegetation on the coast, but probably there alone, whilst everywhere else conditions as regards moisture play the leading part.^ Temperature, illumination, air, atmo- spheric precipitations and humidity, chemical nature of the soil, may all be completely alike, and yet is the vegetation different. One solitary factor is different — the amount of water in the soil — and this it is that is decisive.^ When we further consider that the most important characters 6f soil (temperature, aeration, amount of water, evaporation) are mainly dependent on its structure, it then appears that the physical characters of soil are the weightiest, especially because they react upon the amount of water. Chemical differences are always accompanied by physical ones, and chemical characters seem to be capable of replacement by physical ones, but it would appear that physical attributes are in the last instance most frequently decisive. Yet it must be remarked that the amount of nutriment in soil is likewise of grave import — a principle of which Grabner and A. Nilsson ^ are advocates. But even the amount of nutriment depends upon the physical features of soil, that is, upon its water-capacity and absorbent faculty. Competition among species as a factor of distribution of plants. Darwin and Nageli^ directed attention to one factor affecting the distribu- tion of species and the production of plant-communities — the competition among species — which has not always been taken into consideration, but which must not be overlooked. How trivial a part may be played by chemical distinctions in soil is shown, for instance, by a botanic garden ^ Flahault, 1893. ° The most recent literature is cited by Woodhead, 1906. ' Consult pp. 45-8, dealing with ground- water. * Warming, 1894. ^ Grabner, 1898, 1901 ; A. Nilsson, 1902 b. * NageU, 1865, 1872. CHAP. XVII COMPETITION AMONGST SPECIES ON SOIL 71 in which there flourish on the same soil plants coming from the most diverse soils. But if we neglect the garden, only very few (mainly indi- genous) plants will emerge as victors from the ensuing struggle. Plants are evidently, in general, tolerably impartial as regards soil, if we except certain chemical and physical extremes (abundance of common salt, of lime, or of water), so long as they have no competitors ; only some few plants may perhaps be regarded as obligatory in the one or the other respect ; well-nigh all are facultative, and their occurrence depends upon competitors. If these be present, the one drives back the others, and the victorious species is the one that can best utilize the given combina- tion of soil, light, chmate, and so forth. For instance, according to Fliche, the Scots pine (Pinus sylvestris) over the whole Champagne is confined to calcareous soil and is wanting on non-calcareous soil ; the reason for this is that the Scots pine is an introduced plant, to which the climate, without being actually hostile, is yet not favourable ; on non-calcareous soil, upon which it thrives admirably elsewhere, it is here suppressed by other species, and only on calcareous soil does it become dominant, even then without developing really well. We should therefore err were we to describe it as being calciphilous ; hke many other forest-trees it will grow on the most diverse soils, and in Denmark it is most frequent on sandy soil. When in Denmark we find the oak growing sometimes on moist compact soil, and sometimes on dry poor soil, the reason for this is not that it prefers these soils, but that it is expelled from others by the beech. Similar competitors are hng (Calluna), and many other species, such as Anthemis Cotula and A. arvensis, Carlina vulgaris and C. acaulis. Prunella vulgaris and P. grandiflora, Veronica Teucrium and V. Chamaedrys.^ In the Alps, according to Nageh - Rhododendron ferrugineum and R. hirsutum, as well as Achillea moschata and A. atrata (silicicolous and calcicolous plants) struggle against one another. P. E. Miiller ^ has brought forward several examples of forest- trees in mountains driving each other back in the same manner ; lofty forests of silver-fir, for example, are sharply dehmited from lofty forests of another species without it being a question of inability to thrive at the boundaries. Moreover, Bonnier ^ and others came to the conclusion that species restricted to calcareous soil in one district may be calciphobous in another, and indifferent to soil in a third. In the middle of its distri- butional area a species often makes no selection as to soil, but outside this central position it is forced by other species to exercise a choice.'' As noteworthy examples of plants being able to flourish luxuriantly in countries other than their owti homes, we may cite : Erigeron canadensis, Galinsoga parviflora, from tropical Peru ; Oenothera biennis, and other American weeds that are now common in Central Europe ; Impatiens parviflora and Elodea canadensis may also be mentioned. On the other hand, Salsola Kali, a common httoral European plant, has become a most pestilent weed in the cornfields of North America ; in places it takes nearly complete possession of the soil.'' ' After Pietsch, according to Ludwig, 1895, p. 121. ' Nagcli, 1S72. " P, E. Miiller, 1871, 1887. * Bonnier, 1879. ' Sec Section XVII. * Among more recent literature on this subject readers should consult the works of Cowles, 1901 ; Saint-Lager, 1895; Schimper, 1898; Gillot, 1894; Gain, 1895; Ernst, 1907. The older hterature is to be found cited in Englcr, 1899, on pp. 164-6. 72 OECOLOGICAL FACTORS AND THEIR ACTION sect, i CHAPTER XVIII. THE EFFECT OF A NON-LIVING COVERING OVER VEGETATION A NON-LIVING covering exerts an action that depends upon, inter alia, its looseness or compactness ; the looser it is the greater is its action in the following respects : — 1. Water is sucked in, evaporation depressed, and moisture of the soil increased. 2. Radiation is lessened. 3. Fluctuations and extremes of temperature are diminished. In this relation there come into play two kinds of coverings : {a) snow ; (6) fallen foliage and withered grass. (a) Snow It has long been known that snow can protect vegetation very effi- ciently, and that it guards winter-crops from being frozen. In high alpine situations falls of snow in summer would appear sometimes to protect plants from exposure to the dry, cold weather, and consequent evapora- tion that often set in after such falls of snow. In arctic countries every patch of surface from which the snowy covering is blown away by winter storms has vegetation different from that on snow-clad depressions ; on the tundras of Lapland, for instance, Lecanora tartarea dominates in places exposed to wind, whereas in more sheltered places fruticose lichens can exhibit dense and tall growth.^ The distribution of the covering of snow determines the distribution of entire and definite commu- nities : some are protected at the expense of others ; the spots that in winter are covered with snow usually show in summer the greatest number of species and individuals. The snow-covering is thus of oecological impor- tance. Snow seizes upon the countless particles of dust in the air, purifies this, and collects other small organic and inorganic particles that the wind brings. When snow melts these masses of particles are deposited on the ground, and there is formed gradually a fine, fertile soil, which remains stationary in gentle depressions, and entertains special species of plants. This ' snow-patch flora' is subsequently alluded to in Chapter LXVII.^ The covering of snow influences plant-form. On the one hand, we may here include the influence exerted by heavy loads of snow at high alpine altitudes upon the shapes of trees and shrubs, whose stems are pressed down into a prostrate position on the soil and lie flat on slopes ; for instance, Pinus montana assumes the habit of elfin-tree or contorted shrub, while Juniperus, Alnus viridis, Fagus sylvatica, and other trees dwindle to form scrub, and spruce-birch-scrub develops in South Greenland.^ On the other hand in Lapland Juniperus and Picea excelsa become scrub ,^ in this case because all the twigs projecting above the snow regularly die off, and the individual plants acquire low, plate-hke, or umbrella-like crowns. * Kihlman, 1890. ^ See Schroter, 1904-8. * Kemer, 1863, p. 512 ; Rosenvinge, 1889 ; C. Schroter, 1904-8, p. 663 ; Szabo, 1907- * See the illustrations in Kihlman, 1890. CHAP. XVIII EFFECT OF NON-LIVING COVERING 73 The reasons for this significance on the part of a snow-covering arc the following : — (a) The thermal conditions in snow play a part, though scarcely the leading one. Snow is white because the spaces between its crystals are occupied by air, which may be very considerable in quantity. And it is this air that mainly renders snow a bad conductor of heat. Bv reason of its feeble thermal conductivity snow keeps the soil warmer, and the deeper one descends into snow the less cold is it, so that soil lying under deep snow is exposed to less cold than is bare soil. But this does not suffice to explain all the observed facts, for even under deej) snow plants may be exposed to very extreme cold.^ Neither can it be of very great importance that fluctuations in temperature are decreased so that plants are not exposed to the alternate heat of the day and cold of the night ; snow does serve to prevent too sudden thawing, which may constitute a danger.^ Snow acts as a protection against those changes of volume in frozen soil, occasioned by hoar-frost, which cause plants to be ruptured and uprooted. {h) Of far greater importance is the significance of snow in regard to the amount of water in the plant, as shown in succeeding para- graphs . Snow acts as a defence against transpiration. It is to this action that we must attribute the preservation of many species during winter, and, as described by Kihlman, the death of twigs which project above the snow. It is not low temperature that kills those twigs, but the great atmospheric aridity prevailing in arctic countries, and the violent storms which increase transpiration at a time when the roots are incapable of absorbing water. Twigs and whole plants wilt through desiccation •' ; the shapes of the shrubs serve to show how high the layer of snow stands in winter. The aberrant, sometimes bent and contorted, shapes are occasioned by the death of many twigs and the production of new ones in abnormal positions. The topographical distribution of species is also affected by relations in regard to water, namely, by the uneven distribution of water in the soil that results from the uneven distribution of snow on its surface. Depressions filled with snow remain moist for a longer time than do more elevated spots bare of snow, in fact, they may be moist throughout the vegetative season.^ In some places, for instance, in the steppes of Russia and North America, snow, by reason of its depth, acquires importance as a reservoir of water ; the greater or smaller the supply to the soil, the richer or scantier will be the vegetation in the succeeding vegetative season. When a covering of snow has an injurious effect, for example, on a dense vigorous winter-crop in depressed spots of fields, the cause may be that it suffocates them by restricting the supply of air. Snow affects adjoining slopes by wetting them when it melts. As mentioned on p. 39, in Greenland the northern slopes of a mountain chain may be fresh and vivid green (rich in mosses) in summer, while the southern slopes at the same time are dry and .scorched : this is becau.sc^ * Kjellman, 1884. ' See p. 23. '■ Kililman. 1890. 74 OECOLOGICAL FACTORS AND THEIR ACTION sect, i ct the northern slopes are kept moist by, inter alia, the slowly melting snow, which rapidly disappears from the southern slopes. A covering of snow shortens the vegetative season by preventing the soil-temperature from rising above freezing-point in spring-time, thus hindering plants from awakening into activity as early as on snowless spots. This has a profound effect upon the economy and distribution of plants ; certain species are excluded from places where the snow is wont to lie for a long time, because the vegetative season is too short \ or the soil too cold ; other species are actually favoured by these conditions. Blytt records that on Norwegian mountains, around accumulations of snow which melt to some extent during summer but scarcely ever entirely disappear, the flora is high-alpine in nature on account of the short vegetative season, and corresponds to an altitude above sea-level that is greater than those of the places in question. Even in places where the snow melts only in extremely warm summers we can find vegetation. This must have rested under the snow for several years before awakening. Obviously there are many spots where the snow lies so long that vegetation is absolutely excluded. It is easy to see that orographic and other relations — such as slope and exposure of the soil, nature of the wind, specific heat of the soil, and the like — that influence the melting of the covering of the snow, thereby acquire a phyto-geographical significance.^ {b) Fallen Foliage and Withered Grass The other kind of dead covering to the soil is made by fallen foliage or withered grass ; fallen foliage is met with especially in forest (both deciduous and evergreen), withered grass on meadow and savannah. These coverings have the same physical action as snow, diminishing the extremes of temperature, keeping the soil moister, and so forth. In the forest many plants can scarcely continue to exist without such protection against desiccation — that the protection is not merely against cold, is even more evident than when the covering is composed of snow.^ A covering of leaves over the soil in beech-forest and similar forests, where it is very thick, has a great influence upon vegetation on the ground inasmuch as it suppresses mosses and sundry other plants. A covering of leaves powerfully affects the production of humus in soil, which it thus improves, and is, further, of deep significance to animal life in forest soil, for it concerns moisture, and provides food for animals living on forest soil, among which earthworms seem to be the most important.^ Both circumstances prevent the humus soil of forest from changing into raw humus, and check all those modifications in the soil- covering that would accompany such a change, and would gravely interfere with the whole economy of forest.^ In this connexion may be mentioned the utility of their old dead ' For further information on the significance of snow see Woikof, 1887, 1889. ^ Concerning the characters of the various coverings on forest soil see Ramann, 1890, 1893, 1905. ' See Chapter XX. * See P. E. Miiller, 1878, 1894 ; Ramann, loc. cit. CHAP, xviii EFFECT OF NON-LIVING COVERING 75 parts to certain other plants, particularly to arctic, alpine, and desert plants. A fact that has long been known, and already mentioned on p. 24 of this work, is that the old, dead leaves remain attached in great numbers to the branches of sub-glacial plants, and thus envelop them with dense coverings, whose closeness is further increased b}' the production of con- densed short branches. This is evidently a result of the circumstance that the processes of disintegration and decay take place extremely slowly in the cold climate where bacteria and fungi do not thrive ; it is of utility to the plant in obstructing transpiration. Nature ensheaths plants just as a gardener in preparation for winter mulches sensitive forms. Certain species growing upon dry rock or similar arid spots are, in like manner, enveloped by remains of old twigs and leaves ; in this case it is lack of moisture, not of heat, that arrests the disintegrating action of fungi and bacteria. That the plants concerned derive any benefit therefrom cannot be generally asserted, though it is probable. It is possible to conceive that these old plant-parts serve partly as a protection against transpiration, and partly as organs for the absorption and reten- tion of water. In this connexion attention may be directed to the tunic grasses,^ also to the envelopes formed by the leaves and roots on the stems of the Velloziaceae, and by the roots in Dicksonia, and some other ferns.- CHAPTER XIX. EFFECT OF A LIVING VEGETABLE COVERING ON SOIL Every kind of covering formed by vegetation acts upon the physical relations in soil ; and the denser, taller, and longer-lived the vegetation is, the more powerful is its action. Forest therefore acts most powerfully ; and for this reason the vegetation clothing the ground in forest, on the one hand, and the plants forming the high-forest, on the other hand, are subject to entirely different physical relations. The effects partially agree with those wrought by inanimate coverings : I. The temperature of the soil is modified. A vegetable covering screens the soil, and therefore decreases the action of the sun's heat. But a vegetable covering is a very effective radiator of heat. Fluctuations of temperature, both diurnal and annual, are less considerable ; compared with soil clad with vegetation, bare soil is warmer by day and colder by night, warmer in summer and colder in winter. The maxima of temperature are much higher, but the minima only httle lower in bare soil than in shaded soil (clad with vegetation), so that the mean temperature of the latter soil is lower by, at any rate, one or two centigrade degrees in forest than in bare soil. The temperature at the surface of forest soil in Central Europe, according to Ebermayer, is rarely higher than 25° C. Inside forest the dead covering of course contributes to increase these effects. ' Hackcl, 1890; Warming, 1892 a. ' Warming, 1893; sec also Section III. 76 OECOLOGICAL FACTORS AND THEIR ACTION sect, i 2. The amount of water in soil is influenced. A portion of the atmospheric precipitations is lost to the soil of forest not bare of foliage, because it is deposited on the plants and thence evaporated ; this is specially true of small precipitations. In forest, about fifteen per cent, of the atmospheric precipitations is thus lost, and more in coniferous than in dicotylous forest. The power of soil of the forest to retain the moisture that reaches it is increased, as it is protected against evaporation. Snow melts more slowly, and water derived from melting snow is absorbed in larger quantities by the soil. On the other hand, a plant-covering tends to dry the layers of soil in which roots occur, and the more completely so the denser the vegetation is, because plants absorb water from the soil and dissipate it by transpira- tion.^ Soil contains less water if it be clothed with vegetation than if it be bare (other conditions remaining the same). A vegetation of weeds may have a very drying effect on soil. 3. Soil covered with vegetation is less compact than bare soil. This is so because descending rain can exercise no excessive mechanical action upon it ; moreover, animals (earthworms) play a more direct part in this connexion. 4. Light falling upon soil covered by plants is w^eakened. 5. The action of wind is decreased among dense, and especially among tall, vegetation. 6. Air underneath a vegetable covering is changed, especially in forest ; it is cooler and moister. The air above soil occupied by plants, particularly above forest, is also cooler ; and this may perhaps lead to an increase in the deposit of dew, in cloudiness, and in rainfall. It is certain that forest and dense vegeta- tion in general prevents atmospheric precipitations from flowing away rapidly, and thus being lost to plants and causing floods. 7. A covering of moss requires special mention, because it differs from any other vegetable covering in its effect on the amount of water in soil. The effect varies with the species of moss concerned. Some mosses (Hypnum and its allies) produce dense cushions, five to six centimetres in thickness, lying loose on the soil. The stems of other mosses (Polytrichum, Dicranum) are enveloped in a felt of rhizoids ; their protonemata and rhizoids permeate the soil in the form of a dense felt, and promote the formation of raw humus. Mosses therefore must act upon soil in divers ways. But, according to Oltmanns,^ the general facts of the case are as follows : — {a) A carpet of moss acts as a sponge. The dense, low carpet, with countless capillary spaces between leaves and rhizoids, absorbs capillary and superficial water, but obtains little or none by suction from the soil and internal conduction — the internal structure is an index of this.^ Consequently, living and dead carpets of moss imbibe and evaporate approximately the same amount of water. (6) A carpet of moss does not desiccate soil. Since mosses, particu- larly those forming loose-lying cushions, do not take much water from ' See Chapter XII, p. 48. ' Oltmanns, 1885. * Haberlandt, 1904, Absch. vii. CHAP. XIX EFFECT OF A LIVING COVERING • 77 soil, they dry it to a less degree than does other vegetation, and they protect dry easily-heated soil from desiccation. Evaporation takes place more rapidly from a covering of moss than from a dead covering, but the former keeps soil moist and cool upon the whole, and may easily occasion swampiness in wet, shaded soil. 8. The chemical relations of soil are influenced by a covering of plants, for different kinds of vegetation differently affect the nutriment in soil and its absorbent faculty, abstracting different inorganic substances and enriching it with organic bodies. A rotation of crops and the apphcation of manure are matters of necessity to the farmer, because with each crop he constantly removes from the soil certain quantities of nutritive sub- stances. The forester does this to a smaller extent, except when, as in Germany, he carries away forest-Utter, and the application of manure is as a rule not necessary in forest or, at any rate, has been but Httle practised. Nevertheless wind blows leaves out of many forests, and consequently brings about great changes in the soil and vegetation. A change in the forest-vegetation is known to have occurred in Denmark during past millenia,^ yet one would be only partially correct were one to seek to attribute this to a kind of rotation of crops practised by Nature, and due to each species of tree impoverishing the soil in such a manner as to render the soil less suitable for its own maintenance but more so for that of other species. Certain it is that the annual removal of wood from the soil withdraws some of the most essential plant-nutriment — potash for example. Where no forest-litter is taken away, one can trace no deterioration in the soil. The exodus of nutrient substances may be balanced by the advent of suitable salts derived from the weathering of deeper layers of soil. But where the removal of forest-litter is practised, easily satisfied species, e. g. Scots pine, replace species that are more exacting in their demands, e. g. beech and oak. CHAPTER XX. THE ACTIVITY OF ANIMALS AND PLANTS IN SOIL Between the plant-life and the animal-life of a place there exists an intimate and complex reciprocal relation, which expresses itself in various ways, and promises biological results of the deepest interest. Here we shall consider only two aspects of the matter : the effect on the soil of the tunnelling by animals and the effect of saprophytic plants. Tunnelling of the soil by animals. Soil is traversed by many species of animals : terrestrial soil, particu- larly by earthworms, larvae of insects, millipedes, wood-lice, ants, as well as by animals, such as moles, which search for these ; marine soil, by small Crustacea, Sedentaria or Tubicolae, and others. Terrestrial soil. The uppermost layer of soil in forest and field usually consists of an intimate mixture of mineral matter, animal remains, and '■ See Sect. XVII, Chapter XCVI. 78 OECOLOGICAL FACTORS AND THEIR ACTION sect, i the remains of previous vegetation in the form of leaves, fragments of twigs, fruits, seeds, and so forth, which occur in various stages of decom- position, and of demohtion wrought by animals. Terrestrial soil, if abounding in animal life, is favourable to vegetation because it is rich in humus bodies ; ^ but if wanting in animal life then the vegetation is usually low and stunted. Animals, especially earth- worms, work on the soil and thus on vegetation in four different ways : 1. They comminute vegetable remains by means of their jaws or, in the case of earthworms, by means of their alimentary canal with the aid of ingested stones. 2. In their intestines they mix their food with mineral particles of soil, thus promoting the formation of humus by producing a finely- mixed soil. 3. They bury vegetable fragments in the soil. 4. By the tunnels and passages due to their activity they render soil more porous and better aerated — the soil becomes ' mellow ' — thus promoting respiration in the roots and consequently growth in the plants. The excrement deposited likewise serves to render soil friable and porous. In this way animals also facilitate drainage. Earthworms play a special role in ordinary soil. In Denmark two large species, Lumbricus terrester and L. rubellus ; as well as L. pur- pur eus, Allolobophora turgida, and species of Euchytreus, are of signifi- cance. They make burrows which descend vertically into soil to a depth of two or more metres, and which reach down to deeply buried roots. The burrows are filled with substances, fragments of leaves and excrement, nutritious to the plant. Five other species live in arable soil. Some- times they are so numerous that some 400,000 individuals may occur in a hectare of land. At night, and in moist, dull weather they emerge from their burrows and deposit excrement in the form of friable castings on the surface of soil. They drag leaves into the ground, so that these decompose ; they comminute vegetable remains, acting on these mechanically, swallowing and intimately mixing them with mineral particles which they have likewise swallowed. In addition, their alkaline digestive liquids neutralize humous acids in soil. Shade, shelter from wind, and moist air contribute towards a wealth of animal life in the soil ; shade and shelter from wind are therefore of indirect importance to vegetation. When a forest-soil is exposed to desiccation and the fallen leaves are carried away by wind, the earthworms vanish, the soil becomes dry and hard, and the vegetation suffers. In acid soil (bog, heath,) and dune, earthworms are wanting. Upon their presence or absence depends the occurrence of a humus soil or a raw humus soil in north temperate forest and heath. Conversely they disappear upon the production of raw humus and humous acids. Even upon the growth of rhizomatous plants in the forest do they exert an action ; - their presence or absence causes a series of variations in the kinds of soil that corresponds to a series of variations in the plants clothing it.^ As an additional example, showing how animals may affect vegetation, * Seep. 42. ' P. E. Muller, 1894. ^ The natural history of earthworms has been investigated by C. Darwin (1881), P. E. Miiller (1878), and in the tropics by C. Keller (1887). CHAP. XX ANIMALS AND PLANTS IN SOIL 79 it may be mentioned that mole-heaps and ant-hills very often support a vegetation somewhat different from that on the surrounding soil.^ Marine soil. A role obviously of less general significance, though similar to that played by earthworms in relation to plant-life on terrestrial soil, is played by species of Arenicola in relation to Zostera-vegetation on the marine soil of European coasts.^ Saprophytic plants in the soil. A more important part is played by saprophytic plants in the soil, especially by fungi and bacteria, than by animals. Fungi in soil. In all kinds of soil with abundant humus, fungal mycelia live ; forest-soil in northern Europe in autumn, by its wealth of Basidiomycetes, reveals the extent to which it is permeated by fungal hyphae. But even when few or no fungi exhibit themselves above ground, microscopical examination will certainly demonstrate their presence in all humus-laden soil, even in acid heath-peat ; hyphae of Cladosporium humifaciens occur in this, while the ^oots of Calluna and other denizens have mycorrhizae, just like the majority of forest-trees and some perennial herbs living on humus.^ Saccharomycetes hibernate in soil.^ Bacteria in soil. These are of still greater importance than fungi. They occur in all soil and in all water, in terrestrial soil, in the various types of mud, in sahne and fresh water. In the uppermost layers of soil, especially near human dweUings, they occur in millions upon millions ; in soil occupied by vegetation their number increases with the depth as far down as about half or three-quarters of a metre ; it then rapidly decreases until, at a depth of live to six metres, there are as a rule no bacteria : the soil has filtered them out of the percolating water. The investigations of Adametz, according to Sacchse,^ gave the following results : — Number of Bacteria in one Gramme of Soil according to Adametz. Nature 0/ sou '^'P^ ^'X "/J."''""' ^'"""^^ °l "■<'"'«-'- Sandy soil At the surface 380,000 „ „ „ 20-25 460.000 Clay soil „ the surface 500,000 „ 20-25 464,000 Other investigators have found in one gramme of soil up to a million bacteria. The number of course depends upon various conditions. The number of species concerned is probably exceedingly great, and we know that some of them play a prominent part in the biology of soil. Some are aerobic, others anaerobic. There are present not only ordinary putrefactive bacteria, many of which are of the highest signifi- cance in regard to the composition of the air in soil, but also pathogenic species, for example the tetanus germ (Bacillus tetani), as well as others, including nitrifying and denitrifying bacteria, which cause the formation of important chemical compounds in soil. Schlosing and Miintz were ' Buchenau, 1876 ; Warming, 1804, 1906 ; P. E. Miillcr. iS()4 ; also sec p. 66. ' Rosenvinge, 1889-90, see Warming, 1906; concerning Corophium sec Warming, 1906. ' See Chapter XXV. * C. E. Hansen, 1881. ' Sacchsc. 1S88. 8o OECOLOGICAL FACTORS AND THEIR ACTION sect, i the first to prove that the formation of nitrates in soil is due to the activity of micro-organisms, since nitrogen-containing soil, in which this process can take place, loses the power of inducing it if heated up to iio° C, but regains that power when non-sterilized soil is added to it, and, further- more, since chloroform instantly arrests the process in question. Winogradsky first isolated these organisms. They flourish in a well- ventilated, moderately moist, alkaline soil that contains nitrogen, at temperatures between io° C. and 45° C. According to Miintz, the nitrate- bacteria play a prominent part in disintegrating rock, by penetrating the finest pores and exercising their chemical activity .^ It may be regarded as established that bacteria in the soil enrich this in nitrogen by utilizing free nitrogen from the air. Leguminosae with root-tubercles containing bacteria, Elaeagnaceae, and Alnus also have this power. According to P. E. Miiller's ^ investigations the moun- tain-pine (Pinus montana), which has both mycorhizae and peculiar, coralloid, branched root-tubercles, also belongs to the plants capable of fixing free nitrogen. Experience has shown that where Picea excelsa has been planted on heaths in Jutland it flourishes better in company with Pinus montana than without it. It is probable that in this case the mountain-pine provides the spruce with nitrogen. Bacteria do not flourish in a soil containing free acids (humous acids) ; consequently they are scanty, or lacking, in peat and similar soils. CHAPTER XXI. EXPOSURE. OROGRAPHIC AND OTHER FACTORS The different factors already considered are in nature so varied and connected bj^ such a number of transitions that the greatest diversity results in the nature of habitats and in the differentiation of vegetation, and it becomes extremely difficult to decide which factors are the most weighty in a given case. But this multiplicity and variety are further increased by modifications occasioned by certain geographical or orographic factors. Among these are included the direction of mountain- chains and valleys, the height of mountain-chains, the steepness and exposure of declivities, and so forth. The direction and height of mountain-chains. These are of paramount climatic significance. They steer wind into definite directions, occasion fohn-winds, capture moisture from the wind on certain sides, and con- dense aqueous vapour in higher regions in the form of clouds and rain ; consequently, on certain sides or at a certain altitude above sea-level luxuriant forests may prevail, whilst on other sides or at a lower level extreme aridity reigns. Thus the coast-mountains of Brazil are rainy and clad with forest ; but the interior is dry because the moisture of the trade-wind is condensed and deposited before the interior is reached. In like manner the coast of South Africa is moist, but the Karroos are dry. In the West Indies the more low-lying islands are dry and receive ^ See C. Schroter, 1904-8, p. 558. * P. E. Miiller, 1903. CHAP. XXI EXPOSURE 8i but little rain, whilst the more raised ones receive heavier atmospheric j precipitations and have more luxuriant vegetation. In miniature the conditions of the surface may produce an effect ; for instance Blytt ^ ! mentions that steep walls of rock facing south place the vegetation j concerned under unusual conditions as regards temperature ; beneath I lofty walls of rock at Christiania a vegetation occurs which is rich and ! varied, and includes a number of southern species ; intense heat prevails here on sunny days. Steepness of slope (angle or inclination to the horizon). This decides whether the products of weathering and the humus-substances can remain in situ or are carried away, the rapidity with which water t]ows away from the surface, the extent to which the surface is soaked with water, the density and height of the vegetation, and the intensity with which the sun's rays can heat the soil.^ Exposure of slope. This largely determines the kind of vegetation present. A slope exposed to sun and wind bears vegetation entirely different from that on one less exposed to either. In addition to what has been stated on p. 51, it may be noted that in the Russian east sea-provinces the south-western slopes bear a more mesophilous, and the north-eastern slopes a more xerophilous vegetation, because the south-west winds bring humidity, and the north-east winds aridity .^ Even in very small concerns, exposure may affect vegetation, for instance on dunes ; Giltay * has made some observations, showing the differences that can exist in temperature and atmospheric humidity only a few paces apart on the northern and southern slopes of sand-dunes in Holland.^ In like manner the vegetation on the opposite sides of a cutting or the embankments of a railway may be very different, as Stenstrom has pointed out.® On the southern side of slopes in the east of North Germany the flora of the sunny (Pontic) hills is especially characterized by the development of plants belonging to a Continental climate. Di^erences in geognostic structure, for instance in the inclination of the strata, evoke distinctions in vegetation. Inchnation of the strata acts on the course taken by water, on the emergence of springs, and therefore on vegetation. Moreover, the nature of the surface itself may be entirely different, according as to whether it forms an angle with the dip of the strata or runs approximately parallel with this ; in the former case the surface may be steep and gravelly as well as dry, so that only scanty and stunted vegetation can develop, while in the latter case it slopes gradually, is richer in water, and consequently bears dense and vigorous vegetation. Examples illustrating this are to be met with in many districts with slate mountains.' ' Blytt, 1893. ' See Chap. XIII. ' Klingc, 1890. * Giltay, 1886. ' Warming, 1907 (1909). * Stenstrom. 1905. ' The study of oecology will be much promoted by the preparation of maps in which the type vegetation is denoted by a special colour, and by a comparison with maps showing the geognostic surface. Excellent detailed studies of the kind have been made by Woodhead (1906) and W. G. Smith (1903-5). Flahault (1894, 1897, 1901), and Drude (1902, 1908), have pubhshcd vegetation-maps dealing with more extensive areas. Clements's (1905) work may also be consulted. WARMING /^ \4 SECTION II COMMUNAL LIFE OF ORGANISMS CHAPTER XXII. RECIPROCAL RELATIONS AMONG ORGANISMS The non-living (physical, chemical) and other factors dealt with in Section I do not suffice to impart a full comprehension of the production of communities in the vegetable kingdom. On p. 84 mention is made of another factor — the competition among species of plants — the importance of which is so great that many species are excluded from great areas on the Earth, not by direct interference on the part of non-living factors, but by the indirect interference involved in competition for food with other stronger species. Another factor, animal-life, also has a powerful influence upon the kind and the economy of vegetation. We have discussed the parts played by earthworms, insects, and other small animals in causing physical and chemical changes in soil, but animal-life affects the existence of plants in many other ways, and among all living beings man stands in the foreground as inducing the greatest modifications in plant-communi- ties and in their reciprocal struggles. The manifold, complex, mutual relations subsisting among organisms are matters of such profound import to plant -life and plant-communities that this Section of our book is set apart for their consideration. CHAPTER XXIII. INTERFERENCE BY MAN Very diversified are the reciprocal relations between the plant -world and mankind. Although the plant-world affects the human race, it is itself to a far greater extent influenced by mankind ; indeed, vegetation is the result of man's influence to such an extent that soon there will be but few places upon earth where he has not modified or destroyed the vegetation by directly turning it to his own use or by indirectly interfering with it. Here we merely draw attention to the extent to which man alters the condition and economic status of the original plant- communities by ameliorating the soil, also by tending cultivated plants and domestic animals, and further point out how, by introducing new cultivated plants (such as forest-trees) and new weeds, he voluntarily or involuntarily brings in fresh forms to compete with native plants. Old plant-communities are eradicated by man, and new ones inaugurated ; for instance, when we see in South America an abandoned plantation filled with weeds in the form of bushes, this is a new, secondary, commu- RECIPROCAL RELATIONS 83 nity which did not naturally occur before the soil had been drawTi into the service of man ; and the species which now occur in vast numbers, and form a community with its own special stamp and economy, must previously have been scattered singly at the edge of the forest or in other open places.^ Further information concerning interference by man will be given in Section XVII. CHAPTER XXIV. SYMBIOSIS 2 OF PLANTS WITH ANIMALS Modern biological investigations,^ to which Darwin's works gave the impulse, have elucidated the manifold and complex relations sub- sisting between the plants and animals that form one community, and have demonstrated the adaptations of plants to animals and the converse. From a floristic standpoint we may note between the distributional area of certain plants and animals a connexion which is due to exact reciprocal adaptation : as examples may be cited : Aconitum and Bombus * ; Vanilla, which was introduced into Mauritius at the com- mencement of the eighteenth century, but could be made to bear fruit only by artificial pollination, because the proper pollinating insects were lacking ; Angraecum sesquipedale, which is undoubtedly adapted to a moth with an immensely long proboscis ; Yucca filamentosa, dependent for pollination upon Pronuba yuccasella.^ Attention may also be drawn to the utterly different parts played by entomophilous and anemophilous flowers in the physiognomy of the whole plant-community and the landscape. Trees of the northern forests are anemophilous, those of the tropical ones are mainly entomo- philous, and there thus arise those differences in floral beauty that give to the forest an entirely distinct appearance. Many oceanic isles, the Galapagos Isles for instance, are poor in spermophytes with highly coloured blossoms, but abound in ferns and plants with small or inconspicuous flowers : apparently this is to be correlated with the scantiness of the insect fauna .^ But other matters have also to be taken into consideration. All structural features serving to protect plants against animals : poisons, bitter bodies, raphides, stinging hairs, sharp bristles, and so on ; '' the reciprocal adaptations between insects and flowers ; structural features enabling plants to utilize animals as agents dispersing their fruits (endozoic dispersal of seeds in juicy and coloured fruits, epizoic dispersal of fruits ;ind seeds provided with hooks and glandular hairs, myrmccochorous plants) 8 or even buds and parts of shoots ; symbiosis of ants and plants ' Warming, 1892. - [A somewhat extended significance is here given to the term Symbiosis.] ^ In this connexion may be mentioned the names of Axcll, Bcccari, Briquet, llurkill, Delpino, Scott-Elhot, Hildcbrand, Keller, Knuth, Lindman. Low, Ludwig, MacLeod, H. Miillcr, A. F. W. Schimper, Schumann, Warming, Wilhs, and many I 't hers. ' Kronfeld, 1890. ° Riley, 1873, 1891 ; see also Knuth, 1904, iii, p. 130. " Wallace, 1880 ; but see M. G. Thomson, 1880. ' Stahl, 1904, ' Scrnander, 1901, 1906. C 2 84 COMMUNAL LIFE OF ORGANISMS sect, ii to their mutual advantage (Myrmecodia, Cecropia, Acacia, and Triplaris, according to Belt,^ Delpino,^ Schimper,^ Schumann,^ and Warming^; the symbiosis of acari and plants in which the domatia (acaro-domatia) are constructed for occupation by the former 6; the symbiosis which, according to Cienkovsky, Entz, Brandt, and Geddes, prevails between green or yellow algae (Zoochlorella, Zooxanthella) and animals (Radi- olaria, Infusoria, Flagellata, Spongilla, and Hydra viridis), and which must be regarded as mutualistic, since the alga supplies carbonaceous food and oxygen, while the animal provides shelter and constantly fresh supplies of water containing carbon dioxide. Reference may be made here to the adaptations of insectivorous plants in accordance with their peculiar method of feeding ; also to the fact that an important oecological and geographical part is played by certain animals which search for and utilize certain plants as food, e.g. stags, hares, mice, and the Uke in forest, also large ruminants in savannah and desert. In this way certain species of plants are favoured at the expense of others, so that the whole stamp of the plant-community is changed. The manner in which plant-shap'j may be changed by animal bites has been illustrated and explained by L. Klein.' CHAPTER XXV. SYMBIOSIS OF PLANTS WITH ONE ANOTHER. MUTUALISM Various kinds of bonds of very various strengths can knit plants ■ together ; in some cases the symbiosis is very intimately bound up with i the existence of the species concerned, in others the connexion is far looser, even quite casual. In what follows we shall deal first with those types of symbiosis ini which species are most intimately and firmly linked, that is to say, organically united (symbiosis, in the strict sense, oecological guilds in " Schimper's sense),^ and shall gradually pass on to the looser types until we conclude with the great plant-communities which include many associated species, and which will be the special subject of our considera- tion. The various types of symbiosis are not sharply delimited from one another. PARASITISM. Parasitism is a form of symbiosis in which the two symbionts are i associated in the most intimate manner. One species provides the other ; with nutriment ; the parasite lives on or in its host, and at the expense of the living tissue of the host. ^ Belt, 1874. * Delpino, see Schimper, 1898 (1903, p. 155) ; Raciborski, 1898. ' Schimper, 1898 (1903, pp. 140-53). * Schumann, 1888, 1889, 1891 a, h. \ * Warming, 1893 ; see also Ule, 1900. 1 * Lundstrom, 1887 ; Penzig and Chiabrera, 1903. ' L. Klein, 1899. The interdependent and reciprocal relations between plants and animals have been dealt with by Ludwig, 1895. Reference should also be made to C. Schroter, 1904-8. * Schimper, 1898. CHAP. XXV SYMBIOSIS OF PLANTS WITH ONE ANOTHER 85 There are, however, stages in the degree to which parasite is dependent upon host and requires to abstract food from it : Most dependent of all are many rust-fungi, the species of Cuscuta and Orobanche, which are not only holoparasites, that is to say, incapable of utilizing inorganic food-material, but are able to live only upon one definite species of host. Less dependent are those species of parasites that can thrive equally upon several or many kinds of host belonging either to one or to several families. Cuscuta Epithymum (holoparasitic) is one such species, as it lives on Calluna, Labiatae, Papilionaceae, and even on Monocotyledones and Equisetum. While Viscum album (hemiparasitic) is another such species, of which one race can be parasitic upon about fifty species of dicotylous trees, and other races upon several kinds of coniferous trees ; these do not pass from dicotyledon to conifer or the reverse ; they are physiological races (the ' habitation-races ' of Magnus, the ' specialized forms ' of Eriksson, the ' biological races ' of Rostrup). While certain species are obligate parasites and therefore can exist only as parasites, there are others, less exacting, which on occasion flourish as saprophytes, for instance the honey-fungus (Armillaria mellea). Nectria cinnabarina and some other fungi are perhaps always saproph3rtes at first, but subsequently pass from dead stumps of branches into living tissue. Between parasite and host the relationship is hostile (a one-sided antagonism) : the parasite attacks the host and robs it of energy. The host may be so weakened as to perish — for instance, orange-trees are killed by Loranthaceae ; in such cases the parasite of course likewise perishes. The struggle between a species and its parasites has a highly important bearing upon the composition of the plant-community. Many forest- trees succumb to fungal attack — for example, Scots pine in Denmark attacked by Lophodermium pinastri — and the nature of the forest- vegetation of whole countries may be thus influenced. Pure forests are much more exposed to parasitic attack than are mixed forests, because parasites spread more easily through a homo- geneous assemblage of plants than through a heterogeneous one. Attack by parasites, as well as climatic conditions, is often the cause for one species giving way to another. HELOTISM The symbiosis between lichen-fungi and algae is obviously most correctly interpreted as helotism. A lichen is a dual organism consisting ol a fungus and an alga, which is enveloped by the fungal hyphac and is incorporated with the fungus. The relationship is usually described as mutualistic, that is to say, the two organisms arc said to be of mutual service to each other ; and this is to an extent true, since the alga by means of its chlorophyll provides the carbonaceous food and directs the metabolism to the best advantage of the community, while the fungus secures all else that is required. But the reciprocity is not equal, and the term consort seems inadequate, because though the fungus requires to combine with the alga before it can develop into its completest condi- tion, yet the alga has not the least need of the fungus, and indeed prefers 86 COMMUNAL LIFE OF ORGANISMS sect, ii to live apart from it. That the alga grows vigorously, multiplies rapidly, and may even acquire larger cells than when free, may be nothing more than an example of hypertrophy — a pathological condition. It has been suggested that the alga finds protection from desiccation within the fungal mass ; but this seems to be scarcely necessary, as the algae in question are certainly capable of enduring desiccation admirably ; moreover, it is not the case that they secure real protection against desiccation, for under given circumstances the lichen dries up so completely as to become brittle. Besides this, the alga is prevented from multiplying in its most efficient manner — for instance, by zoospores. The alga is in a condition of slavery in relation to the fungus, which is a kind of parasite differing from ordinary parasites in incorporating the host and in providing a portion of the food consumed in the host's maintenance. There is there- fore a certain likeness to hemiparasitism, but we must assume that green hemiparasites provide their own carbonaceous nutriment, whereas the lichen-fungus needs merely to secure for itself the non-carbonaceous food- material. In this case, too, the bond of union between the two organisms may be exceedingly close, in that the fungus selects definite species of algae. MYCORHIZA AND ENDOPHYTES A mutualism characterized by complete reciprocity, in which the symbiosis is equally advantageous to both partners, may or may not occur. Even in the most familiar forms of symbiosis the relations sub- sisting between the symbionts are not sufficiently understood to permit of our completely explaining the nature of their connexion. This is true of mycorhiza, in which the root of a highly organized plant enters into intimate connexion with fungal hyphae, which are either ectotrophic, and mainly form a sheath enveloping the distal surface, or are endotrophic, and live inside the cortical cells. Mycorhiza has been found in the majority of Amentaceae, Coniferae, Ericaceae, and many other plants, especially in perennial herbs growing on humous soils, such as acid humus, peat, and mould. Mycorhiza is present in the humicolous herbs, whether these do or do not contain chlorophyll ; plants belonging to the latter category (saprophytes) seem certainly to be dependent on the fungus for nutrition. The fungus possibly derives some benefit from the phanerogam, and there is scarcely a doubt that it is of use to the latter. As regards the endophytic mycorhizal fungus, Percy Groom ^ has proved that in Thismia it indulges in an interchange of nutritive material with the root, and promotes the production of protein bodies. The fungus abstracts from the root certain substances and provides others in exchange, and is itself perhaps digested to some extent. In the ecto- trophic form the relations are different. Stahl has expressed the opinion that the mycorhizal fungus in all cases undertakes the absorption of water and ash-constituents from the soil ; accordingly, it would be of special importance in soils poor in nutriment. This view harmonizes well with the distribution of mycorhiza. * Percy Groom, 1895 b. CHAP. XXV SYMBIOSIS OF PLANTS WITH ONE ANOTHER ^y Perhaps we have here a notable example of one plant being aided by another to settle in a habitat and to secure nutritive material on a soil from which it would otherwise be excluded ; Calluna-heath, spruce- forest, and the like, would then to a certain extent owe their existence to such symbiosis. But a very great deal concerning this form of symbiosis is completely unexplained.^ Apparently similar in some respects to endotrophic mycorhiza is the symbiosis of Leguminosae and bacteria which has already been mentioned on p. 80. The small root-tubercles of Leguminosae are entered and occupied by bacteria which manufacture nitrogenous food and finally perish, becoming changed into ' bacteroids ' and utilized as food by the Leguminosae. It is not definitely established that the bacteria profit by this symbiosis (they presumably acquire carbon- compounds for their host) ; but if they did not profit it would be remark- able that they, like endotrophic fungi, should enter roots. Going one step farther, we come to plants (algae) which inhabit others without, so far as we know, doing any service in return. They do not live at the expense of the host, in fact perhaps absorb nothing whatever from it, but have free quarters. In this category may be placed the cyanophyceous Anabaena living in the under-side of the leaves of AzoUa within special cavities ; these seem to exist only on the alga's account, as they occur in all four species of Azolla which are never free from Anabaena. The alga can flourish quite apart from Azolla. In like manner other algae live as endophytes, that is to say, inside other plants : in Sphagnum, whose leaves are occupied by Nostoc, which enters the colourless cells by way of the pores in their walls ; in certain liverworts, or in algae — for instance, Entoderma viride living in the cell- wall of Derbesia Lamourouxii. But perhaps the last is an example of parasitism. To some extent the same is presumably true in the case of those Cyanophyceae which enter the erect dichotomous roots of cycads, and stimulate a definite layer of parenchyma to grow in a special manner so as to provide space for themselves ; also, in the case of Nostoc punctiforme, which penetrates the stems of Gunnera but can live equally well apart from roots or stems.^ The present state of our knowledge does not permit us to define the exact nature of the symbiosis in every case. EPIPHYTES. From those endophytes that only seek for accommodation in other plants but do not absorb food from these, it is but a slight step to epiphytes, or plants living on others but abstracting no food from the living parts of the latter, and at most deriving sustenance from the dead tissue of these. Still it is not always permissible to say that epiphytes do not live at the expense of the supporting plants, for they may occur on these in such quantities as to necessitate the assumption that they do injury by this very quantity, by causing excessive humidity, or by diminishing respiration, as, for instance, in the case of lichens on trees. * Those who have worked on mycorhiza include Kamicnsky, 1881 ; Frank, 1887 ; Sorauer, 1893 ; Percy Groom, 1895 ; W. Magnus, 1900; Maz6. 1899; Stah], 1900 ; P. E. Miiller, 1886, 1902, 1903 ; and many others. * B. Jonsson, 1894. 88 COMMUNAL LIFE OF ORGANISMS sect, n The bond between epiphyte and the species upon which it rests is usually less close than in the preceding cases ; most epiphytes can grow upon various kinds of plants, some even upon rock or on the ground. Yet some are confined to definite species, because the nature of the cortex of the latter is of importance. There are epiphytes on aquatic as well as on terrestrial plants. Many kinds of algae live on other algae, or even on phanerogams, and some algae only upon quite definite species — for instance, Elachista fucicola on Fucus, E. scutulata on Himanthalia lorea.i Epiphytes upon terrestrial plants thrive best where atmospheric humidity and precipitations are rich. Yet a change of seasons, and movements of the air, seem to promote their welfare. Meyen (1836) devoted attention to this subject, and subsequently A. F. W. Schimper dealt with it in greater detail in his papers upon epiphytes.^ In cold and temperate climes epiphytes mostly belong to algae, lichens, and mosses ; but in warm countries they also include a number of ferns and phanerogams (Orchidaceae, Araceae, Bromeliaceae, Cactaceae, Piperaceae, and others) ; and in moist tropical forests there are many epiphyllous species of algae and lichens living upon leaves.^ Peculiarities in the habitat have resulted in a number of biological adaptations, which Schimper, Gobel, Raciborski, Mez, Treub, Karsten, and Beccari have explained in reference to flowering plants. The particular details in question are described in the succeeding paragraphs. The seeds (and spores) are calculated to gain double object — dispersal and fixation to the substratum. Either they are scattered by wind, in which case they are so small and light, or are so provided with long hairs (and the like), as to be easily conveyed by the wind on to trunks and branches where they encounter a fissure or some other depression in which they can become firmly lodged. Or the seeds are contained in fleshy fruits (e. g. Araceae, Bromeliaceae, and Cactaceae) which are eaten by birds, in whose excrement they are dispersed and fastened on to branches. An entirely exceptional method of multiplication characterizes the rootless Tillandsia usneoides, detached fragments of which easily become twisted round twigs of trees by means of their long slender shoots. Fixation of the epiphyte to parts of plants is accomplished either by rhizoids (as in the case of mosses and hchens) which penetrate the substratum (dead cortex) to some extent, or by attaching-roots which are irritable and sometimes adhere firmly to the substratum with the aid of fixing-hairs and secretions. There is often a division of labour between attaching-roots and absorbing-roots. Provision of water is a problem of difficulty to the epiphyte, because rain-water soon flows away. Some epiphytes obtain the necessary water rather from dew and mist than from rain ; others do the converse.* Many are able to seize the momentary opportunity, and in their dry condition can instantly absorb moisture over their whole surface, e. g. algae, mosses, lichens, Tillandsia usneoides and other Bromehaceae ^ Concerning the significance attached by Fritsch to the term ' consortium ', see Fritsch, 1906. * A. F. W. Schimper, 1884, 1888 ; compare also Mez, 1904 a. * Gobel, 1889-92; Raciborski, 1 898 ; Mez, 1904 a; G. Karsten, 1894; Treub, 1888. C. Jennings. * Mez, 1904 a. CHAP. XXV SYMBIOSIS OF PLANTS WITH ONE ANOTHER 89 with peculiar absorbing-hairs.^ Others (Orchidaceae, Araceae) have aerial roots with a special velamen adapted to absorb water ; yet others, for example Tillandsia bulbosa, have their leaves so constructed as to facilitate the retention of water among them ; others, again, possess two kinds of leaves, some of which, as ' pocket-leaves \^ are pressed closely against the substratum so that water is held by capillarity between them and the supporting-stem, or is actually taken up by them, as G. Karsten suggests in the case of the fern Teratophyllum aculeatum. Epiphytes are much exposed to desiccation. Against this, certain species (algae, lichens, and mosses) have no evident protection ; they can endure without injury existence in a dry condition for a long period, and awaken into life again at the first fall of rain or dew. But others have fashioned for themselves water-receptacles of various kinds : aqueous tissue in leaves and stems (in Orchidaceae, Peperomia, and others), water-storing cells in leaves (in Orchidaceae and others), urn-shaped water-bags or con- cavities of other shapes (as in liverworts,^ Dischidia, and others). Food-material is obtained by epiphytes as follows : Carbon is taken from the air, as all epiphytes are photophilous and evergreen ; some also accumulate humous and mineral bodies among their roots or with the aid of specialized leaves ('pocket-leaves', 'mantle-leaves'), as in the cases of ferns such as Asplenium Nidus, Polypodium quercifohum, and Platycerium alcicorne.^ The construction of the shoot and the whole architecture of the epiphyte varies widely. Some species, like Tillandsia usneoides, are rootless, while the vegetative organs of others, such as the orchid Poly- rrhiza (Aeranthus) funalis, consist almost entirely of green roots. With Schimper ^ we may divide epiphytes into four groups : 1. Those that find their nutriment on the cortex of their support. 2. Those that send aerial roots into the soil. 3. Those that collect moist humus within the large interwoven mass of roots, and, in some cases, behind or between their leaves. 4. Those whose leaves take over the functions of roots and absorb water and nutritive salts.^ Epiphytes have many structural features in common with terrestrial xerophytes, for, like these, they must be adapted to endure prolonged drought. They form in fact a group of xerophytes, and it is consequently easy to understand how it comes that certain species, such as Rhipsalis Cassytha and other Cactaceae, can live upon either trees or rocks. The characteristic structural features of epiphytes will therefore be considered in detail in Section III, which deals with xerophytic vegetation.' SAPROPHYTES In the case of many epiphytes it must be assumed that they derive nutriment from dead parts of plants (bark) upon which they grow ; they thus feed upon dead organic substance, and are saprophytes. Larger numbers and more pronounced forms of saprophytes are, ' Schimper, 1884, 1888a; Mez, 1904 a. ' Gobel, 1891. * Gobel, 1889-Q3 and 1898-1901. ' Gobcl, 1889-93. ' A. F. W. Schimper, 1884, 1888, 1898. * Sec G. Karsten, 1894. ' See Wittrock, 1894; Willis and Burkill, 1904; Ule, 1904; Cockayne, 1901. 90 COMMUNAL LIFE OF ORGANISMS sect, ii however, met with on the ground, especially in the forest, where all kinds of fallen fragments (withered leaves, twigs, flowers, and fruits) accumulate year after year and produce humus. Saprophytes are therefore bound up with other plants, but the bond is different from that associated with parasitism ; for it is the cast-off parts and redundant individual plants that they utilize. Some saprophytes select special kinds of vegetable remnants and are therefore inseparably associated with definite species of plants ; others are less narrow in their choice. Clavaria abietina, Lactarius deliciosus, and certain other fungi are only met with in coni- ferous forest, others select dicotylous forest, while others again grow only upon dung, as is the case with Poronia, Coprinus, Pilobolus, and Sordaria among Fungi, and Splachnum among mosses. Among cryptogamic and phanerogamic plants alike, saprophytes display a very varied degree of adaptation to the saprophytic mode of existence.^ Every kind of humus is permeated with fungal mycelia and bacteria, and the soil in forest at autumn reveals hosts of pileate fungi. Phanerogamic plants that are most completely adapted to a saprophytic life, that is to say, holosaprophytes, exhibit the following characters : — 1. They have little or no chlorophyll, being yellow, red, or brown in tint. 2. Their leaves are upwardly directed, and more or less reduced to adpressed scales. 3. Stomata are usually absent. 4. The root-system is more or less reduced, some forms like Corallo- rrhiza being quite rootless ; the roots are short, thick, and but feebly branched, and in the vast majority of cases form mycorhiza. 5. The vascular bundles are reduced. As examples may be cited : Neottia, Corallorrhiza, Epipogum, Pogo- nopsis, and other orchids ; some Burmanniaceae ; Triuridaceae ; Mono- tropa and Sarcodes among Pyrolaceae ; Voyria in the Gentianaceae.^ Hemisaprophytes have the external appearance and structure of normal plants that assimilate carbon dioxide. Their needs as regards organic nutriment probably differ extremely, for while some cannot exist away from a soil rich in humus, such as a forest soil, others, like many orchids and species of Pyrola, are tentatively to be regarded as facultative saprophytes.^ LIANES While the want of humus-containing food constitutes the bond uniting saprophytes with other plants, lianes are linked with fellow plants by reason of the want of support for their weak long- jointed stems by which they reach the light. The term liane is here employed in the widest sense, and includes twining plants as well as the various kinds of other chmbers. Lianes owe their origin to the grouping of plants into com- munities in the form of forest and bush ; the shade due to dense vegetation has caused them in the past to elongate, to produce long intemodes, and in the course of time to adapt themselves in various ways not only ^ See p. 85. " See Johow, 1885, 1889 ; Percy Groom, 1895 a and b. " See Heinricher, E., 1896, 1897, 190 1-3 ; Wettstein, R. v., 1902. CHAP. XXV SYMBIOSIS OF PLANTS WITH ONE ANOTHER 91 to hold firmly to their supports, but also by suitable internal structure to solve the problem of conduction of material, as well as other problems that they owe to the length and slenderness of their stems.^ Lianes display very different degrees of adaptation to the climbing habit. The lowest stage is represented by : — 1. ' Semi-lianes ' (Warming) and scramblers (Schenck), which occur particularly at the margins of forests, in hedges, and in bushlands. More elaborate and speciahzed in adaptation are : — 2. Root-climbers, which can clamber up thick tree-trunks and rocks. 3. Twining plants. 4. Lianes equipped with tendrils or other irritable organs, and capable of embracing slender parts of plants. These were termed by Darwin 'hook-climbers', 'tendril-bearers,' and ' leaf -climbers'. It is characteristic of the majority of lianes for the leaf to be broad, more or less cordate, and long-stalked.^ (But to this rule exceptions are provided by those Papihonaceae and other lianes that climb by means of tendrils on the ends of their leaves.) In structure of leaf and stem some hanes recall xerophytes ; it seems quite natural that lianes should be exposed to the possibility of losing more water by transpiration than can be balanced by supplies from the root, and hence should be structurally adapted to provide for this contingency.^ Certain species, for instance, of the genus Ficus occur both as hanes and as epiphytes.* The liane-form is a consequence of the congregation of plants to form communities, but hanes are partially independent of their fellows, since inanimate supports will sometimes serve them quite as well as living ones. Lianes belong especially to certain families, including Ampelidaceae, Asclepiadaceae, Apocynaceae, Bignoniaceae, Cucurbitaceae, Papihona- ceae, Sapindaceae, Dioscoraceae, and others. CHAPTER XXVI. COMMENSALISM. PLANT-COMMUNITIES In the preceding chapter we have dealt with the different bonds that may hnk plants together, and one individual with another ; parasite with host, master with slave (helotism of hchens) ; we then discussed mutualists, epiphytes, and finally species associated with the whole plant-community. We have yet to consider the great, highly complex plant-communities which form the essential foundations of oecological phytogeography. The term ' community ' implies a diversity but at the same time a certain organized uniformity in the units. The units are the many individual plants that occur in every community, whether this be a beech- forest, a meadow, or a heath. Uniformity is established when certain atmospheric, terrestrial, and any of the other factors discussed in Section 1 are co-operating, and appears either because a certain, defined economy makes its impress on the community as a whole, or because a number of ' For further details see C. Darwin, 1875 ; Schenck, 1892, 1893 ; Warming, 1892 ; A. F. W, Schimper, 1898. * Lindman, 1899; Warming, 1901. ' Warming, 1892. ' Schenck, 1892, 1893 5 VMiitford, 1906. 92 COMMUNAL LIFE OF ORGANISMS sect, ii different growth-forms are combined to form a single aggregate which has a definite and constant guise. The analysis of a plant-community usually reveals one or more of the kinds of symbiosis as illustrated by parasites, saprophytes, epiphytes, and the like. There is scarce a forest or a bushland where examples of these forms of symbiosis are lacking ; if, for instance, we investigate the tropical rain-forest we are certain to find in it all conceivable kinds of symbiosis. But the majority of individuals of a plant-community are linked by bonds other than those mentioned — bonds that are best described as commensal. The term commensalism is due to Van Beneden who wrote ' Le commensal est simplement un compagnon de table ' : but we employ it in a somewhat different sense to denote the relationship subsisting between species which share with one another the supply of food-material contained in soil and air, and thus feed at the same table. More detailed analysis of the plant-community reveals very consider- able distinctions among commensals. Some relationships are considered in the succeeding paragraphs. LIKE COMMENSALS When a plant-community consists solely of individuals belonging to one species — for example, solely of beech, ling, or Aira flexuosa — then we have the purest example of like commensals. These all make the same demands as regards nutriment, soil, light, and other like conditions ; as each species requires a certain amount of space and as there is scarcely ever sufficient nutriment for all the offspring, a struggle for food arises among the plants so soon as the space is occupied by the definite numbers of individuals which, according to the species, can develop thereon. The individuals lodged in unfavourable places and the weaklings are vanquished and exterminated. This competitive struggle takes place in all plant-communities, with perhaps the sole exceptions of sub-glacial communities and in deserts. In these open communities the soil is very often or always so open and so irregularly clothed, that there is space for many more individuals than are actually present ; the cause for this is obviously to be sought in the climatically unfavourable condi- tions of life, which either prevent plants from producing seed and other propagative bodies in sufficient numbers to clothe the ground, or prevent the development of seedlings. On such soil one can scarcely speak of a competitive struggle for existence ; in this case a struggle takes place between the plant and inanimate nature, but to little or no extent between plant and plant. That a congregation of individuals belonging to one species into one community may be profitable to the species, is evident ; it may obviously in several ways aid in maintaining the existence of the species, for instance, by facilitating abundant and certain fertihzation (especially in anemo- philous plants) and maturation of seeds ; in addition, the social mode of existence may confer other less-known advantages. But on the other hand it brings with it greater danger of serious damage and devastation wrought by parasites. The bonds that hold hke individuals to a Hke habitat are, as already indicated, identical demands as regards existence, and these demands CHAP. XXVI COMMENSALISM 93 are satisfied in their precise habitat to such an extent that the species can maintain itself here against rivals. Natural unmixed associations of forest-trees are the result of struggles with other species. But there are differences as regards the ease with which a community can arise and establish itself. Some species are more social than others, that is to say, better fitted to form communities. The causes for this are biological, in that some species, like Phragmites, Scirpus lacustris, Psamma (Ammo- phila) arenaria, Tussilago Farfara, and Asperula odorata, multiply very readily by means of stolons ; or others, such as Cirsium arvense, and Sonchus arvensis, produce buds from their roots ; or yet others produce numerous seeds which are easily dispersed and may remain for a long time capable of germinating, as is the case with Calluna, Picea excelsa, and Pinus ; or still other species, such as beech and spruce, have the power of enduring shade or even suppressing other species by the shade they cast.^ A number of species, such as Pteris aquilina, Acorus Calamus, Lemna minor, and Hypnum Schreberi, which are social, and likewise very widely distributed, multiply nearly exclusively by vegetative means, rarely or never producing fruit. On the contrary, certain species, for example many orchids and Umbelliferae, nearly always grow singly. In the case of many species certain geological conditions have favoured their grouping together into pure communities. The forests of northern Europe are composed of few species, and are not mixed in the same sense as are those in the tropics, or even those in Austria and other southern parts of Europe : the cause for this may be that the soil is geologically very recent, inasmuch as the time that has elapsed since the Glacial Epoch swept it clear has been too short to permit the immigration of many competitive species.- UNLIKE COMMENSALS The case of a community consisting of individuals belonging to one species is, strictly speaking, scarcely ever met with ; but the dominant individuals of a community may belong to a single species, as in the case of a beech-forest, spruce-forest, or hng-heath — and only thus far does the case proceed. In general, many species grow side by side, and many different growth-forms and types of symbiosis, in the extended sense, are found collected in a community. For even when one species occupies an area as completely as the nature of the soil will permit, other species can find room and can grow between its individuals ; in fact, if the soil is to be completely covered the vegetation must necessarily always be heterogeneous. The greatest aggregate of existence arises where the greatest diversity prevails.^ The kind of communal life resulting will depend upon the nature of the demands made by the species in regard to conditions of life. As in human communities so in this case, the struggle between the like is the most severe, that is, between the species making more or less the same demands and wanting the same dishes from the common table. In a tropical mixed forest there are hundreds of species of trees growing together in such profuse variety that the eye can scarce see at one time two individuals of the same species,^ yet all of them ' See Chap. XCVII. * See Warming, 18996. ' See Darwin, 1859. * Sec Warming, 1892, 18996. 94 COMMUNAL LIFE OF ORGANISMS sect, ii undoubtedly represent tolerable uniformity in the demands they make as regards conditions of life, and in so far they are ahke. And among them a severe competition for food must be taking place. In those cases in which certain species readily grow in each other's company — and cases of this kind are famihar to florists — when, for instance, Isoetes, Lobelia Dortmanna and Litorella lacustris occur together — the common demands made as regards external conditions obviously form the bond that unites them. Between such species a competitive struggle must take place. Which of the species shall be represented by the greatest number of individuals certainly often depends upon casual conditions, a slight change in one direction or the other doubtless often playing a decisive role ; but apart from this it appears that morphological and biological features, for example development at a different season, may change the nature of the competition. Yet there are in every plant-community numerous species which differ widely in the demands they make for light, heat, nutriment, and so on. Between such species there is less competition the greater the disparity in their wants ; the case is quite conceivable in which the one species should require exactly what the other would avoid ; the two species would then be complementary to one another in their occupation and utihzation of the same soil. There are also obvious cases in which different species are of service to each other. The carpet of moss in a pine-forest, for example, protects the soil from desiccation and is thus useful to the pine, yet, on the other hand, it profits from the shade cast by the latter. As a rule, a limited number of definite species are the most potent, and, like absolute monarchs, can hold sway over the whole area ; while other species, though possibly present in far greater numbers than these, are subordinate or even dependent on them. This is the case where subordinate species only flourish in the shade or among the fallen frag- ments of dominant species. Such is obviously the relationship between trees and many plants growing on the ground of high forest, such as mosses, fungi and other saprophytes, ferns, Oxalis Acetosella, and their associates.^ In this case, then, there is a commensalism in which individuals feed at the same table but on different fare. An additional factor steps in when species do not absorb their nutriment at the same season of the year. Many spring-plants — for instance, Galanthus nivalis, CorydaHs solida and C. cava — have withered before the summer-plants commence properly to develop. Certain species of animals are Mkewise confined to certain plant-communities. But one and the same tall plant may, in different places or soils, have different species of lowly plants as com- panions ; the companion-plants of high beech-forest depend, for instance, upon climate and upon the nature of the forest soiP ; Pinus nigra, accord- ing to von Beck, can maintain under it in the different parts of Europe a Pontic, a Central-European, or a Baltic vegetation. There are certain points of resemblance between communities of plants and those of human beings or animals ; one of these is the com- petition for food which takes place between similar individuals and causes the weaker to be more or less suppressed. But far greater are 1 See Hock, 1892, 1893. ' P. E. Miiller, 1887. CHAP. XXVI COMMENSALISxM. PLANT-COMMUNITIES 95 the distinctions. The plant-community is the lowest form ; it is merely a congregation of units among which there is no co-operation for the com- mon weal but rather a ceaseless struggle of all against all. Only in a loose sense can we speak of certain individuals protecting others, as for example, when the outermost and most exposed individuals of scrub serve to shelter from the wind others, which consequently become taller and finer ^ ; for they do not afford protection from any special motive, such as is met with in some animal communities, nor are they in any way specially adapted to act as guardians against a common foe. In the plant-community egoism reigns supreme. The plant-community has no higher units or personages in the sense employed in connexion with human communities, which have their own organizations and their members co-operating, as prescribed by law, for the common good. In plant-communities there is, it is true, often (or always) a certain natural dependence or reciprocal influence of many species upon one another ; they give rise to definite organized units of a higher order ^ ; but there is no thorough or organized division of labour such as is met with in human and animal communities, where certain individuals or groups of individuals work as organs, in the wide sense of the term, for the benefit of the whole community. Woodhead ^ has suggested the term complementary association to denote a community of species that live together in harmony, because their rhizomes occupy different depths in the soil ; for example, he described an ' association ' in which Holcus mollis is the ' surface plant ', Pteris aquilina has deeper-seated rhizomes, and Scilla festalis buries its bulbs at the greatest depth. The photophilous parts of these plants are ' seasonably complementary '. The opposite extreme is provided by competitive associations, composed of species that are battling with each other. The classification and nomenclature of plant-communities are fully discussed in Chapter XXXV. * See Chapter VIII, p. 2,7. ^ See, for instance, Grevillius, 1894. ^ Woodhead, 1906, p. 345. SECTION III ADAPTATIONS OF AQUATIC AND TERRESTRIAL PLANTS. OECOLOGICAL CLASSIFICATION CHAPTER XXVII. AQUATIC AND TERRESTRIAL PLANTS In the Introduction ^ brief mention was made of the fact that there are plant-communities which are characterized by a definite physiognomy, definite constituent growth-forms, and a definite economy : this is a consequence of the circumstance that those species which make approxi- mately the same demands in regard to the nature of the environment, or which are associated for other reasons, congregate naturally to consti- tute a kind of single entity. It will now be our task to inquire as to what communities there are, and on what principles they are to be most naturally defined and ranged into some sort of system, or, in other words, which of the factors men- tioned in Section II are of the greatest significance in this matter, and which of them play only a subordinate role, also what part in the estab- lishment of this classification is taken by the growth-forms already dis- cussed in Section I. Before we can deal with the individual communities we must consider oecological classification as a whole. The grouping of the classes of communities here adopted is based in the first place upon the -plants dependence upon and relation to water?' Pindar's aphorism, apirrrov ixev vboip, is wholly true of plant- life ; water is the condition of life that exercises the greatest influence in bringing into being external and internal differences among plants ; it is likewise water that plays the leading part in determining the creation of plant-communities and their distribution over the soil. It is quite true that the special attributes of a habitat result from the co-operation of the most diverse factors, edaphic and climatic, not one of which can be omitted without modifying those special attributes and consequently the vegetation. For instance, vegetation is greatly affected by fertility of the soil, and on a sterile soil there occur only communities of feeble productive power .^ But it is beyond doubt that water occupied the foremost position as a factor bringing about the greatest distinctions in vegetation and structure. We may say : The supply of water to the plant and the regulation of transpiration are the factors that evoke the greatest differences in plant-form and plant-life. In this connexion we at once meet with two extremes : There are many plants that pass the whole or the greater part of their ^ Page 12. * See Chaps. VII, VIII, XII. * Grabner (1898, 1901, 1908), indeed, would have it that nutriment in the soil is the paramount factor. AQUATIC AND TERRESTRIAL PLANTS 97 lives submerged in the water, which envelops them completely or, at most, leaves definite floating parts of them uncovered at its surface : these are water-plants (aquatic).^ On the other hand there is a still greater number of plants that expose at least their assimilating organs to the air and hence to transpiration : these are land-plants (terrestrial), and amongst them we include marsh-plants. To submerged water-plants transpiration is an impossibility ; in land-plants transpiration takes place, and it is incumbent upon them to maintain a balance between the intake and output of water, that is to say, they must regulate transpiration. If evaporation be greater than the supply of water the plant withers, and this has the gravest effect upon vital processes, even when it does not go so far as to kill the plant. Here reference may also be made to the part played by water in the general economy of nature, in its promotion of putrefaction and of the pro- duction of humus, as the micro-organisms responsible for these processes need water. The significance of water in relation to the distribution of plants is demonstrated most distinctly in flat countries such as the western parts of Denmark ; a marked zonal arrangement of the vegeta- tion reveals itself here, not only in water but also on land, round every lake or pool. Differences of a few centimetres in the level of the water- table sulftce to evoke wide distinctions in the vegetation.^ The story of man also indicates the importance of water to the plant. History has shown to what an extent the prosperity of countries (density and wealth of the population) is dependent upon water. In Asia, for example, civilization was confined to those lands where a well-watered soil ensured the existence of man. In Algiers the density of population runs almost parallel with the amount of rainfall.'^ Lack of water is that factor in plant-life in the face of which man is most helpless. The effect of their environment on water-plants is not only marked by the absence of transpiration, but is also impressed upon them by the other peculiar conditions belonging to water, such as its absorption and consequent weakening of light, its dissolved air, its movements, its buoy- ancy, and other characters. In the succeeding chapters the general structural relationships of the water-plant and land-plant respectively will be considered. CHAPTER XXVIII. ADAPTATIONS OF WATER-PLANTS (HYDROPHYTES) Various structural features and phenomena exhibited by submerged parts are to be regarded as adaptations to the peculiar physical qualities of water. Some of these are dealt with in the succeeding paragraphs. I. Roots and analogous organs. Since nutriment may be absorbed by the whole permeable surface of all submerged parts, there is in sub- merged plants a reduction in those organs which normally extract mineral food-material from the soil, that is to say, in the roots, or in analogous • Hydrophyta of Schouw, 1822. ' See Raunkiar, 1889; Warming, 1907; Massart, 1893, 1908, and others. * Deherain, 1892. WARMING H 98 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, iii organs among Cryptogamia. Some vascular plants, such as Salvinia, Wolffia, Ceratophyllum, Utricularia vulgaris, Aldrovanda, and Genlisea, are entirely rootless ; in others, such as Azolla, Lemna, Hydrocharis, Pontederia, and Pistia, the roots soon cease to grow, do not branch, and may shed their root-caps. Root-hairs are absent in Lemna minor and L. trisulca, Myriophyllum, Butomus umbellatus, Caltha palustris, and Hippuris vulgaris {' except at the collar '), Nymphaea alba, and others.^ Roots and root-hairs in many cases are merely anchoring-organs.^ 2. Water-carrying tubes are for the same reasons in less demand ; wood-vessels and the whole xylem are consequently reduced in vascular plants. Phloem, as the tissue conducting protein bodies, undergoes no reduction. The conducting tissues are always congregated more towards the centre of the organ, so that they finally constitute a central bundle. Van Tieghem establishes four types of degenerate roots. ^ The ramifica- tion and number of veins in the foliage leaves is less than in land-plants. 3. Mechanical tissue is either reduced or undeveloped because the buoyancy of water is greater than that of air. In particular, those struc- tural designs adapted to resist bending are not developed. In order to resist stretching, due to movements of the water, large fixed water-plants living in very troubled water have their mechanical tissue collected as closely as possible to the centre of the stem so as to form a design adapted to resist tensile stresses * ; while certain algae have strengthening rhizoids at the base of the thallus, as Wille ^ has demonstrated in detail. Lignifi- cation occurs to little (in the wood-vessels) or to no extent. Among submerged water-plants there are no woody plants. 4. Air-containing spaces are very abundant and large in submerged water-plants and marsh-plants, and serve partly to decrease their specific gravity (as a flotation-device), and partly to facilitate gaseous inter- change and especially respiration. In a number of the larger algae such as Fucus vesiculosus, Ascophyllum nodosum, Halidrys siliquosus, Sargassum, Macrocystis and other Laminariaceae, well-developed flota- tion-devices occur. Exceptions to these statements are provided by nearly all lithophilous hydrophytes, including the vast majority of algae, mosses, Podostemaceae, as well as by some small Spermophyta such as Bulliarda aquatica. 5. Secondary growth in thickness takes place only exceptionally in the axial organs of water-plants. This is correlated with the matters discussed in the three preceding paragraphs. On the contrary, sub- merged parts of stems and leaves of Spermophyta are far longer and thinner than are the corresponding members when developed in contact with air, and owing to weakness of illumination they almost acquire the appearance of etiolated parts. 6. The epidermis or the external cell-wall in contact with water is thin, and the cuticle very thin or wanting. Coatings of wax and cork are absent. In contact with air hydrophytes wither and dry up with extreme rapidity. Hairs are lacking from the assimilating organs of nearly all submerged Spermophyta, and when present may serve either ^ F. Schwarz, 1888. . ^ See Henslow, 1895 5 Warming, 1881-1901. 3 Van Tieghem, 1 870-1. * Schwendener, 1874. .^ Wille, 1885. CHAP. XXVIII ADAPTATIONS OF WATER-PLANTS 99 to produce mucilage, or to promote assimilation or respiration ; litho- philous species of algae and Podostemaceae provide examples of the last two cases. 7. The epidermis or external layer of cells often contains chlorophyll^ and in algae is actually the tissue richest in chlorophyll.^ This must be causally connected with the weakness of light, also with the lack of any necessity for the epidermis to function as aqueous tissue. 8. Excretion of water is not excluded from submerged water-plants, but when occurrent assumes the form of guttation (the excretion of liquid water) induced by internal activity. At the leaf-tips of many species water-pores occur, or the tips are detached and the ends of the vascular bundles are thus brought into direct contact with the water, as has been shown by the investigations of Sauvageau, Wieler, Weinrowsky, and Minden.2 Transpiration in the strict sense is of course excluded, and correlated with this is the usual absence of stomata. Where these do occur as exceptions, they may be regarded as functionless vestiges. (In floating plants stomata of abnormal structure occur. )^ 9. Mucilage is excreted by many water-plants, often in great quanti- ties, sometimes, as in many bacteria and algae, from the general surface of the body, sometimes from special organs, such as hairs in the higher plants, and at other times into internal passages. The function of mucilage has not been definitely ascertained, and possibly may vary widely. In many bacteria, algae, and buds, it perhaps serves as a protection against injury due to rapid physical or chemical change in the surrounding water, and according to the views of Gobel and others * accomplishes this by obstructing the passage of water ; according to Stahl ^ it acts as a defence against animal attack ; while Hunger ® suggests that it acts as a lubricant facilitating plant-movement. Littoral aJgae living on the shore and lying dry at the ebb, as well as other algae occa- sionally exposed to drying, are protected from desiccation by mucilage ; but in those algae that grow on a rocky shore and are also exposed to the violence of the water, mucilage may serve as a defence against the force of the waves.' Regarding the production of mucilage, reference should be made to the works of Hunger, Schilling,^ Gobel,^ B. Schroder,^" and other authors cited by these. 10. The chlorenchyma of water-plants is very slightly differentiated ; there is little or no indication of a distinction into palisade and spongy parenchyma in the submerged leaves of Phanerogamia. The leaves are therefore isolateral. This is possibly correlated with weakness of light. Dorsi-ventral structure reveals itself in floating-leaves. 11. The properties of water bring forth leaf-shapes entirely different from those of land-plants, as will be described in Section IV, which deals with water-plants. The submerged leaves of those plants that also possess aerial leaves are entirely different from the latter as regards both ' Wille, 1885. 'Sauvageau, 1889, 1890, 1891, 1894; Wieler, 1892; \\cinrowsky, 1898 j Minden, 1899 ; see Burgerstein, 1904, p. 246. ' Haberlandt, 1904, p. 413. * Gobel, 1898, 1901. '■ Stahl, 1904 a. " Hunger, 1899. ' ^^ "l^^^' '^^S- * Schilling, 1894 " Gobel, 1 898-1901. '• B. Schroder, 1903. H 2 100 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi shape and anatomy. This, according to MacCulland, is directly due to the arrest of transpiration and to the cells being overcharged with water. 12. Duration of life. The vast majority of water-plants, at least among spermophyta, are perennial ; this is in accordance with the favour- able environment, which is but slightly affected by seasonal change. Exceptions are, however, provided by many Cryptogamia and some vascular plants, such as Salvinia, Naias, and Subularia. Vegetative propagation far exceeds sexual reproduction in many water-plants : this may go to such a length that production of fruit is entirely elimin- ated. Certain species, such as Elodea canadensis (at least in Europe, where only the female plant occurs), many species of Lemna, and others, multiply exclusively in a vegetative way. It is a general biological phenomenon that humidity opposes the production of sexual organs, whereas aridity promotes it. The peculiarities of water-plants that have been mentioned here are to be interpreted in general as examples of degeneration, and of morphological and anatomical retrogression, if we compare water-plants with land-plants ; this retrogression we may, with Henslow, regard as adaptive.* CHAPTER XXIX. ADAPTATIONS OF LAND-PLANTS The land-plant contrasts most strongly with the water-plant as regards external and internal construction, in that parts in contact with air — the assimilatory organs in particular — are exposed to transpira- tion, and must therefore be adapted in an entirely distinct manner. Transpiration is a physiological process determined by factors of two kinds : external or environmental, and internal or those dependent on the precise structure or temporary condition of the plant. The external (climatic and edaphic) factors were mentioned in Section I : they are more particularly isolation, temperature, saturation-deficit, and movements of the atmosphere. The supply of water depends upon the nature of the soil, including quantity of water, temperature, acidity, amount and concentration of the salts in the soil. Correctly speaking, as Clements ^ insists, climatic factors to a great extent affect conditions in the soil. As regards internal factors, transpiration depends on the dimensions of the evaporating surface, and inasmuch as it is the foliage-leaves through which evaporation mainly takes place, it is likewise the size, disposition, and thickness of the foliage-leaves, as well as the whole development of the aerial shoot, that above all determine the amount of transpiration ; this is also influenced by the nature of the epidermis (cuticle, wax, cork, hairs, and stomata). The foliaged shoot gives the clearest indication of the conditions under which the plant has developed. An additional determinant factor is the nature of the root-system ; the larger the absorb- ing surface is, the more water can there be absorbed in the same time ; and the deeper the penetration of the root, the greater is the certainty that the supply of water will not be cut off by drought. The regulation of the amount of water within the plant is accomplished * Henslow, 1895. ^ Clements, 1904. I i CHAP. XXIX ADAPTATIONS OF LAND-PLANTS loi by means of water-excreting organs, to which Haberlandt ^ has given the name of hydathodcs. They are possessed by land-plants as well as by some water-plants, and cause water to be discharged by exudation- pressure in the form of drops. Not only in the tropical rain-forest, but also in the temperate countries, are there many plants, especially herbs, exhibiting the phenomenon of guttation. When transpiration is depressed by saturation of the atmosphere, there comes the danger that the plant, on account of continued and powerful root-pressure, may take up an excess of water from the wet soil, and thus attain a condition of maximal turgescence when the air is expelled from the intercellular spaces and completely replaced by water. The danger is averted by the hydathodes. These organs mainly belong to the following types : — 1. Epidermal cells, sometimes of remarkable structure, or peculiar hairs which are unicellular or multicellular, and in the latter case often assume the form of glandular hairs. As these organs occur on both leaf- faces, but particularly on the lower face, drops of water excreted over the leaves simulate dew-drops. 2. In some ferns hydathodes assume the form of peculiar glandular spots on the lamina. 3. The familiar water-pores, which are constructed like stomata, occur on the upper face of leaf-teeth above a small-celled, thin-walled, usually colourless tissue (epithema) in which the vascular bundles terminate. It must, however, be noted further that water may be excreted through the epidermis by means of pores opening outwards, without the co-opera- tion of hydathodes ; and also that water may be excreted without the aid of living cells, for example in grasses : this contrasts with the preceding cases in which living cells are the essential and functionally active organs. The adaptation of the land-plant to its existence in contact with air proceeds on the following lines : — 1. Control of the outgo of water, i.e. regulation of transpi- ration. 2. Increase of the intake of water, i. e. development of special mechanisms for absorption. 3. Arrangement for storage of water, i.e. development of water-reservoirs. In the following three chapters these adaptations will be discussed, and in a succeeding one we shall deal with some structural characters and growth-forms of land-plants, the utility of which to the plant is obscure, although their connexion with its existence in a dry environment is beyond question. At the outset it may be noted that the degree of adaptation of land- plants to their life in contact with air varies widely, according as the external conditions are more or less extreme. Those species that are adapted to meet the conditions of strongest transpiration and most precarious water-supply are termed xcrophytes :^ the remainder are termed mcsophytcs : ^ between these two classes there is of course no strict boundary. 1 Haberlandt, 1894-5, 1904. Xerophyta, Schouw, 1822 (ir]p6s, dry ; (jyvrt'tv, plant). Mesophyta, Warming, 1895 (ficVor, middle). 102 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi CHAPTER XXX. REGULATION OF TRANSPIRATION IN LAND-PLANTS The regulation of the transpiration, and the checking of it at critical times is, especially in xerophytes, effected by the following methods : — 1. Anatomical structure controlling transpiration. 2. Diminution of the evaporating surface, either by movements, or by reduction in the surface of leaves or of shoots which become irrever- sibly and characteristically adapted to the prevaihng conditions. 3. Regulation of illumination of the assimilating organs, either by their assumption of a temporary profile-He (accomplished by photometric movements dependent on intensity of illumination), or of a permanent profile-lie (as in compass-plants). 4. Investing organs, such as hairs, leaves, and the like, which weaken the light as well as directly decrease transpiration. 5. Ablation of rain-water from the leaves. I. ANATOMICAL STRUCTURE REGULATING TRANSPIRATION In this respect there is a fundamental distinction between land- plants and water-plants. It is clear that a great difference must exist between the surface of a plant that is permanently enveloped in water or moist air, and one that is surrounded by dry air and engaged in intense transpiration. But the difference concerns not only the construction of the integument and the aerating system, including stomata and inter- cellular spaces, but also the chlorenchyma. A. Cuticular Transpiration. Transpiration is either cuticular or stomatal. We shall first consider cuticular transpiration, which takes place through the external cell-walls of the plant, or in most of the higher plants through the epidermis. In connexion with the regulation of transpiration there occur the following devices : — Cuticle is the first important regulator of transpiration ; its thickness is adjusted in accordance with the need on the part of the plant to limit transpiration ; yet other conditions seem to play a part, for Bergen ^ found that the cuticle of young leaves is more impermeable to water than is that of old leaves. The cuticle of hydrophytes is as a rule very thin and permeable, but that of xerophytes is thicker and often com- pletely impermeable. The outer walls of the epidermis may be strongly thickened and cutinized, and in some cases may even include crystals of calcic oxalate or silica. The leaves, owing to the nature of the epidermis, are often leathery and glossy, and this is a frequent and striking feature in tropical (sclerophyllous) trees, but is also met with in temperate climes in leaves of evergreen plants, such as Ilex Aquifohum, some Coniferae, and Vinca. The polished surface reflects a portion of the incident hght from the leaves, and may thus be of use.^ Cuticle is often provided with fine processes, especially when the external wall is convex. Vesque ^ ' Bergen, 19046. ^ Wiesner, 18766. * Vesque, 1882. CHAP. XXX TRANSPIRATION IN LAND-PLANTS 103 and Henslow * suggest that the arrangement serves to scatter and weaken the incident rays of hght. Haberlandt,- on the contrary, regards these lenticular cells as organs for the perception of light. Wax may be excreted over the surface and depress transpiration, as has been experimentally established by Tschirch ^ and Haberlandt. Usually the coating produced is only a thin one ; but, to take an opposite case, Capparis spinosa, at the commencement of the dry season in the Egyptian desert, produces over the whole leaf-surface a very thick layer of wax that completely prevents transpiration.* The coating of wax may be very thick, more than one millimetre in Sarcocaulon in South Africa, and up to five milhmetres in Wax Palms. Incrustations of wax cause plant-members to have a dull, matt, bluish surface, which is then said to be covered with ' bloom '. Such ' bloom '-covered leaves usually have at their margins no sharp teeth, and possess, at most, rounded teeth provided with hydathodes. Wax prevents water from wetting leaves, so that it protects ombrophilous fohage from rain.^ Incrustatio7is of salt are produced on the surface of some desert-plants, w^hich thereby acquire a grey tint and are perhaps protected against excessive transpiration ; at night the incrustations dehquesce as they absorb moisture from the atmosphere.^ In the Plumbaginaceae and certain species of Saxifraga the hydathodes which excrete calcic carbonate may possibly ser\'e to check transpiration, but their main function would seem to be the excretion of injurious salts. Varnished leaves. Resins or similar bodies are excreted by hairs on the surface of many, particularly austral, xerophytes. The leaves are thus rendered viscid and appear as if lacquered, since they acquire a glossy, vitreous investment ; the epidermal walls are thin and feebly cutinized.' The creosote bush (Larrea tridentata) in the North-American deserts has leaves which, when unfolded, are thin, but which gradually become coated with shellac.^ Mucilage, or a mixture of gum, resin, and other bodies, is sometimes excreted by hairs (colleters^), in the buds of Polygonaceae and others: it may possibly aid in the absorption of water and perhaps check transpira- tion during the flushing of the foliage. The contents of epidermal cells may be designed to depress transpira- tion. The epidermis is perhaps a water-reservoir ^^ when, as is usually the case with land-plants, it is colourless. In virtue of various substances contained by it the epidermis may become less permeable to water- vapour. Tannin is often markedly present in the epidermis of ever- green leaves during winter,^^ and appears in connexion with the aqueous tissue of desert-plants and steppe-plants such as Alhagi, Monsonia, Astragalus, Tamarix ^^ ; but the functional significance of these facts is obscure. Anthocyan is a red pigment present in many plants, and particularly in the epidermis ; according to Engelmann and Stahl ^^ it acts as a heat- ' Henslow, 1894. - Haberlandt, 1905. * Tschirch, 1882. * Volkens, 1887. ' Burgerstcin, 1904, p. 207. " Compare pp. 31-2. ' Volkens, 1890. ' Coville, 1893, p. 51. " Hanstcin, 1868. '" Westermaier, 1882. " Warming. 1884. " Volkens, 1887 ; Henslow, 1894. " Engelmann, 1887 ; Stahl, 1896. 104 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi absorbing substance and thus promotes transpiration, but according to others it does precisely the reverse.^ An epidermis with mucilaginous inner walls, which are gelatinous, occurs in many land-plants, and specially in woody plants such as Empe- trum. Arbutus Unedo, and other Ericaceae. The mucilage possibly serves to depress transpiration ; ^ but it may perhaps function rather as a water-reservoir.^ A fact of perhaps supreme importance, and one that is the cause of some of the relations already mentioned in connexion with the epidermis, is that wettahle plant-parts wither much more rapidly than unwettahle parts. Wiesner regards the increase of transpiration in the former as due to a peculiar swelling of the cell-walls, which consequently oppose less resistance to evaporation. Many of the devices mentioned as decreasing transpira- tion also serve to prevent the plant-parts from being wetted, and in this way, too, prevent rapid transpiration. Cork, in virtue of its air-filled cavities and its other characters, depresses transpiration, as has been proved by experiment. Its thickness is some- times obviously and directly correlated with dryness of climate, as is illustrated by the difference between the trees of the Brazilian campos and of the adjoining forest. The desiccating action of fires occurring in the campos appears to stimulate the development of cork, and thus to provide an example of self -regulation.* Very thick investments of cork occur in a number of desert-plants, for instance, Dioscorea (Testudinaria) Elephantipes in South Africa, and Cocculus Leaeba in Egypt. In the aerial roots of some Orchidaceae and Araceae a mechanism designed for absorbing water assumes the form of a velamen, which clothes the root with an envelope of cells, usually of several layers in thickness : the cells resemble the water-absorbing cells of Sphagnum ; they are thin- walled, with annular, spiral, or reticulate thickenings. When these cells are filled with air the velamen is white ; but when they are occupied by water the chlorophyll-containing tissue of the root becomes more or less visible. Liquid water is rapidly sucked up by the velamen, and can be transported to the conducting tissue. It is possible that water in the form of vapour may also be absorbed by the velamen. Wehmer, how- ever, expresses another view, namely, that the velamen acts as a protection against transpiration. It is possible that both views are correct.^ Here it may be mentioned that the roots of many xerophytic land- plants produce a very strong endodermis, which probably protects against desiccation. B. Stomatal Transpiration and the Aerating System. Intercellular spaces are also seats of transpiration ; the transpiring surface of a plant is constituted not only by the surface exposed to the external atmosphere, but also by the cell-walls bounding all intercellular spaces ; it may therefore be anticipated that the air-containing inter- cellular spaces of land-plants, and especially of xerophytes, will sharply ^ See Chapter V, pp. 20-1. " Volkens, 1890. ^ Pfitzer, 1870-2; Radlkofer, 1875, p. 100; Vesque, 1884; Walliczeck, 1893; Westermaier, 1880 ; H. E. Petersen, 1908. * Warming, 1892. ° Burgerstein, 1904, p. 69. CHAP. XXX TRANSPIRATION IN LAND-PLANTS 105 (^ontrast with, and be narrower than, those of water-plants in which they are usually very large.^ At the same time there is an extreme difference in regard to the stomata present. (a) Stomata. Stomata, as Leitgeb and Schwendener have proved, are adapted by their mobihty and structure to regulate transpiration. They close when excessive transpiration is threatening, when leaves are withering because of lack of moisture in the soil, also while the leaves of many plants are resting during winter ; and they re-open when there is no further danger. The guard-cells of certain desert-plants are mobile only in young leaves ; but in the old leaves they become immobile owing to strong thickening of their walls, and the stomata may become blocked with wax or resin.''' Floating-leaves of water-plants have stomata on the surface exposed to the air, but these assume a pecuHar form and soon lose their power of movement.'^ The number of stomata depends upon the nature of the environment. As a general rule, the drier a habitat is the fewer are the stomata, as may be seen best when comparison is made between closely alhed species.^ The distribution of stomata is most intimately connected with transpira- tion and with the conditions of moisture. Meadow-grasses and other mesophilous land-plants, as a rule, have stomata on both faces of the leaf ; steppe-grasses only on the furrowed upper face ^ ; other xerophytes usually only on the lower face, where the stomata are often concealed in such a way as to render transpiration more difficult. Stomata of land-plants are frequently sunk beneath the general surface in pits, furrows, and the hke, which are often hned with hairs. There is thus formed a space containing air, which escapes only with difficulty, is screened from the wind, and becomes charged with aqueous vapour ; the result is that transpiration is retarded. Various arrange- ments ® of this kind are described in the succeeding paragraphs. The simplest device is the production outside a single stoma of a saucer-, urn-, or funnel-shaped cavity, which is formed either by cuticular processes giving rise to the outer stomatal cavity, or by the adjoining epidermal cells projecting above the stoma, which is sunk within an external respiratory chamber,^ as in Pinus sylvestris and some Proteaceae. In Euphorbia Paralias,'' also in various grasses and sedges,*^ the stoma is surrounded by low papillae. Groups of stomata in pits, whose narrow apertures are almost occluded by hairs, occur on the lower face of the leaves of Nerium, Banksia, and other xerophytes. In numerous plants stomata are lodged in longitudinal furrows, and then are usually confined to the furrows, whose margins are often more or less beset with hairs. Many stems, especially switch-shaped ones, have deep furrows to which the stomata are limited, as is the case in Casuarina, ' See p. 98. ' Wilhelm. 1S83; Volkcns, 1890: Gilg, 1891. ' Habcrlandt, 1904, p. 412. * Pfitzer, 1870-2; Zingeler, 1873; Czech, 1869; Tschirch, 1881; Volkcns. 1 881; Altenkirch, 1894. ' Pfitzer, 1870-2. • Tschirch, 1881, 1882. ' Giltay, 1886. * Volkens, 1890; Kihlman, 1890; Kaunkiiir, 1895-9. io6 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, iii Ephedra, Acanthosicyos horrida, and species of Genista. The furrows occur on the upper face of the leaf in many steppe-grasses, and others, such as Weingaertneria canescens, Festuca ovina, Psamma (Ammophila) arenaria, Aristida, Stipa, Sporobolus spicatus, Cynodon Dactylon ; in such cases the furrows may be narrowed above, and the stomata more completely enclosed by the rolling up of the leaf.^ Furrows or broader channels clothed with hairs occur on the lower face of the leaf in many plants, such as Empetrum, Phyllodoce caerulea, Calluna, species of Erica, Loiseleuria procumbens. Ledum palustre, Cassiope tetragona,^ and Dilleniaceae.^ To this category may be added leaves, such as those of Dryas octopetala, with margins revolute to a smaller extent and with stomata on their hairy under-surfaces. If leaves are permanently and steeply directed upwards as well as appressed, so that the lower face is the more strongly illuminated, then the lower side may be differentiated like the normal upper side and possess palisade tissue ; in such cases the stomatiferous furrow occurs on the upper face, as in Passerina filiformis, Ozothamnus, and Lepidophyllum.* In these plants, therefore, the easy egress of aqueous vapour is checked in more than one way. That these features stand in direct relation to dryness of climate is shown by species such as Ledum palustre and Andromeda pohfolia, whose leaves are smaller and more revolute the more they are exposed to wind and drought.^ The cases last mentioned form transitions to flat, broad leaves in which the sole screen over the stomata is formed by a dense investment of felted or peltate hairs, as in Olea, Rhododendron, and Elaeagnaceae, or some other kind of tomentum on the lower face of the leaf.® Sometimes leaves of this kind have veins strongly projecting on the lower face, and as the stomata lie in the meshes of the network of veins, they are to a certain extent sunk below the general surface, as for example in the West Indian Lantana involucrata. When stomata are in secluded cavities containing much aqueous vapour, or lie under a dense tomentum, they are usually raised above the adjoining surface, just as in the leaves of plants that live as a whole in contact with moist air. It may also be noted that stomata are enclosed in cavities, or sheltered under tomenta, in order that they may be protected from occlusion by water.' (h) Intercellular spaces. Respiratory cavities may be structurally fitted to regulate transpira- tion ; they may have cuticularized walls, or be surrounded by special cells as in the Restiaceae,^ or may be very small. In many instances cuticle extends from the outer face of the epidermis, through the stoma, and down over the walls of the respiratory cavity.^ The width of intercellular spaces varies with the external conditions ; ' See p. 267. ^ Warming, 1889; Gruber, 1882; Ljungstrom, 1883; H. E. Petersen, 1908. ^ Steppuhn, 1895. * Lazniewski, 1896; Gobel, 1891. Warming, 1887, p. no. " See p. 254. ' Kerner, 1887. * Pfitzer, 1870-2. ' A. de Bary, 1877, p. 79. CHAP. XXX TRANSPIRATION IN LAND-PLANTS 107 intercellular spaces are larger in shade-leaves (sciophylls) than in sun- leaves (heliophylls), better developed in moist than dry air. But for reasons already given it is generally true that in xerophytes growing on land the air-containing intercellular spaces are very narrow ; in this respect Altenkirch's ^ measurements of respiratory cavities may be mentioned. But exceptions occur ; for instance, in Rcstiaceae there are wide air-spaces, which possibly play a part in the assimilation of carbon dioxide, in addition to very narrow ' girdle-canals '. ' Girdle-canals ' also occur in the Australian desert-plant Hakea suaveolens, as well as in Olea europaea, Kingia,^ and in some arcnicolous grasses such as Festuca rubra and Triticum acutum.^ They are narrow intercellular spaces running round the palisade cells parallel to the leaf- surface ; by this tortuous course the escape of water-vapour is rendered more difficult. Certain desert-plants, such as Cynodon Dactylon, and Sporobolus spicatus, have a maze of extremely fine meandering inter- cellular canals,^ but it is not certain that these various forms of intercellular spaces are designed to depress transpiration. C. Chlorenchyma. It is characteristic of land-plants as opposed to submerged water- plants to possess dorsi-ventral leaves, and in particular pahsade tissue. The latter is greatly developed in xerophytes by an increase in either the number of layers or in the height of the cells, or by both means. It has already been mentioned ^ that there is a difference of opinion as to the significance and cause of this structural feature, and the suggestion has been made that it is most closely correlated with dryness of the atmosphere and with transpiration. Light doubtless also plays a part in the matter, for the oblique orientation of palisade cells must be due to illumination.^ In halophytes growing on land the height of the pahsade tissue is increased by the salts contained in the soil, as has been proved by Lesage. D. Other Means of Regulating Transpiration. Ethereal oils occur especially in xerophytes ; the garigues and maquis of Mediterranean countries,' and the canipos of Brazil are scented with Cistus, Labiatae, Verbenaceae, Compositae, and Myrtaceae, just as are European downs with wild thyme, and Asiatic steppes with Artemisia. Neither the origin nor the significance of the correlation between dryness of chmate or of soil and the occurrence of ethereal oil has yet been e.xplained. These oils evaporate more readily than water, and surround the plant with aromatic air. According to Tyndall, air rich in ethereal oil is less diathermanous, that is to say, permits the passage of radiant heat to a less extent than does pure air ; according to this view, ethereal oils diminish insolation and consequently transpiration.** It is possible that ethereal oils are of utility in other directions ; they ' Altenkirch, 1894. ' According to Tschirch. ' According to Giltay, 1886. * Volkens, 1887. ' Sec p. 21. " Sec Warming, 1897. ' See Beck von Mannagctta, 1901, and others. ^ Volkens, loc. cit., and others. io8 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi may, for instance, protect plants against herbivorous animals, as Stahl ^ suggests. Detto - doubts the correctness of the view that ethereal oils serve as a screen against transpiration, and takes the same view as Stahl. The significance of latex is not definitely established. According to Haberlandt, Schullerus, Pirotta, and others, laticiferous tubes are con- ducting channels, but according to Kerner they form a defence against animals at least in Cichoriaceae.^ By the various agencies described above, the transpiration of leaves is brought into harmony with the different environments. But it must not be concluded that xerophytic leaf-structure is inconsistent with capacity for vigorous transpiration : indeed Bergen * found that the absolute amount of transpiration (the amount of water lost in a unit of time) was scarcely less in the sun-leaves (heliophylls) of certain ever- greens, including Olea europaea and Quercus Ilex, than in Ulmus cam- pestris and Pisum sativum. II. DIMINUTION OF THE EVAPORATING SURFACE The extent of the transpiring surface plays an important part in determining the amount of transpiration : other relations being constant, the larger the surface the greater the transpiration. As foliage-leaves are essentially the organs of transpiration, it is their size and number which regulate this function and which therefore vary in the different species in accordance with climatic conditions. Divers means adopted to depress transpiration are treated in the succeeding paragraphs. A. Temporary Diminution of Surface. The most decisive method by which a plant can diminish its tran- spiring surface is the shedding of all strongly transpiring parts before the commencement of the dry season. This takes place, first, in all annuals which die after the seed has ripened : all seeds are very efficiently pro- tected from desiccation. In harmony with this is the very high per- centage of ephemeral species in deserts and similar places ; within the short rainy season, sometimes only from one to two months in length, these plants complete their whole life-cycle, germinating, flowering, setting seed, and dying, so that they pass through the dry season in the form of embryos enclosed in seeds ; Odontospermum (Asteriscus) pygmaeum, the 'Rose of Jericho ', is such a plant.^ Similar behaviour characterizes all bulbous and tuberous plants, as well as other ' renascent ' herbs whose subterranean shoots serve as reservoirs of food and water during the dry season : the epigeous shoots, with their extensive transpiring surfaces, are dispensed with while drought prevails, and the latent vitality is confined to the underground shoots. When moisture is once more supplied these species hasten to thrust new shoots and flowers into the light. ^ In fact, the rapid onset of spring after the first few showers of rain in deserts, steppes, and similar places, has often been mentioned with surprise by travellers. In like manner behave woody plants which shed their foliage before * But see Burgerstein, 1904, pp. 133, 214. ^ Detto, 1903, ^ See also p. 125. ' Bergen, 1Q04. ^ Volkens, 1878. ' See p. 8. CHAP. XXX TRANSPIRATION IN LAND-PLANTS 109 or during the dry or cold season, and remain leafless for a long time ; such species are described as deciduous, in opposition to evergreen. In these plants during the unfavourable season all parts above ground are usually protected from transpiration by means of cork or bud-scales, the latter being covered with cork or other bodies that check evaporation. In all these plants the structure of the foliage-leaf is usually not at all or only slightly xerophytic, but is mesophytic, if the vegetative season be sufficiently moist. In Egypt ^ and in the lowland of Madeira,^ where the atmospheric humidity is small even in winter, annual herbs growing on uncultivated land adopt protective measures against drought quite different from those employed by weeds in the irrigated fields. Protection against drought is more pronounced the more a species prolongs its vegetative season beyond the commencement of the dry season. Accord- ing to Kerner^ the fohage of deciduous trees on the Austrian coast is very hairy on the under-side, because the summer is exceedingly dry. The transpiring surface is reduced in quite a different manner in other plants, for example, in grasses whose leaves in dry weather become rolled up, so that they form tubes, and thus appear filiform or bristle- like. Such is the case with Psamma (Ammophila) arenaria, Weingaert- neria (Corynephorus) canescens, species of Festuca, and many other grasses inhabiting dune or heath ; and, in Mediterranean countries, Avith species of Stipa, Lygeum, Aristida* ; rolled-up leaves are particularly characteristic of steppe-grasses, and are also met with on sahne soil in Triticum junceum and other grasses. As the air becomes drier the leaf rolls up so that the transpiring upper surface, where the stomata mainly or solely occur, is less exposed to transpiration ; the stomata thus become enclosed in a space in which the air is more or less motion- less. In moist weather the leaf unfurls. Among Cyperaceae similar though less considerable movements are exhibited. In these movements a part is played by the hinge-cells lying in fun^ows on the upper face of the leaves of grasses ; these cells are deeper than the other epidermal cells, and their cellulose walls are easily folded as the leaf curls. The motive force would appear to reside in the bast-tissue, which is usually near the under-face of the leaf, and either absorbs or gives out water, thus swelling or contracting. But the turgor of the mesophyll seems to play an important part, at least in some cases.^ Similar movements are exhibited by a number of Dicotyledones, including Hieracium Pilosella, Antennaria dioica, Crepis tectorum.'' West-Indian species of Croton,'^ and Euphorbia Paralias,^ which grows on the dunes of western and southern Europe. The leaves of Erica Tetrahx,^ and Ledum palustre are less rolled on moist than on dry soil. Among Cryptogamia may be mentioned some ferns ^^ and mosses, including species of Rhacomitrium, Tortula, and Polytrichum. The leaves of Rhacomitrium canescens and Tortula ruralis in dry weather are folded together, and the shoots quite grey with densely set, long hairs ; but when the weather or soil is humid they are extended in a ' Volkens, 1887. ' Vahl, 1907 b. * Kcrncr, 1886. * See Duval-Jouve, 1875; Tschirch, 1882; Warming, 1891. ' Duval-Jouve, 1875; Tschirch, 1882. '■ Willc. 1887. ' Warming, 18996. ' Giltay, 1886. ' Grabncr, 1895. '• See Wittrock, 1891. no ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi stellate manner. Polytrichum can lay the marginal portion of the leaf over the thin-walled assimilatory cells clothing the more central portion.^ B. Permanent Reduction of Form of Leaf and Shoot. In very many xerophytes the transpiring organs, the foliage-leaves, are extremely and unalterably reduced in size and surface, and there result a number of speciahzed types of xerophytic shoots. The size of the foliage-leaf, and of the foliaged shoot as a whole, shows a certain dependence upon the amount of food-material and water available to the plant at the time of its development. Lack of water may induce nanism ; for instance, in dry sandy places many species are dwarfed ; again, one and the same species may be small-leaved on dry soil, and large-leaved on moist soil, as is the case with Urtica dioica, Viola canina, Erodium cicutarium and many others ; a number of desert - plants, including Zilla and Alhagi, produce at the commencement of the rainy season large leaves, but later on much smaller ones or none at all. The smallness of the leaf is a direct result of dryness.^ Lack of water has apparently also contributed to the evolution of a series of definite fixed and constant types, which are characterized by their relatively low assimilatory power and consequent slowness of growth.^ These types are described in the succeeding paragraphs. {a) Forms of Leaf. 1. The pinoid or acicular leaf is met with in Coniferae, Proteaceae, Ulex europaeus, and others. It is long, linear, pointed, and often has a more or less radial structure. The relations of this leaf to transpira- tion result from the fact that its surface in proportion to its volume is much less than in the case of a fiat leaf, hence its evaporating surface is relatively less. This is also true of the forms of leaves described in the succeeding paragraphs. 2. The ericoid leaf is a rolled leaf, in other words, its margins are curled downwards, or upwards (much more rarely, as in Passerina) ; there thus arises a furrow in which the stomata are secluded from movements of the air.* Ericoid leaves are short and linear ; they occur in Erica, Calluna, Cassiope tetragona and other Ericaceae, Epacridaceae, Myr- taceae, Berberis empetrifolia from Chile, South-African Thymelaea- ceae, Compositae, Rubiaceae, and among other families in species growing on maquis, heaths, or other places where transpiration is strong. 3. The cupressoid (lepidophyllous, lepidoid) leaf is broad and short, appressed, apically directed, and sometimes decurrent ; it is met with in many Cupressaceae, in some Scrophulariaceae (in Veronica thuyoides and V. cupressoides, which are alpine in New Zealand), Santalaceaej Tamaricaceae, Compositae, Umbelliferae (in Azorella at alpine altitudes in South America, and in Antarctic lands).^ 4. The setaceous ov filiform leaf occurs in very many grass-like Mono- cotyledones ; it is usually furrowed or channelled on its upper face, and ^ Kerner, 1887. '' Henslow, 1894; Scott-Elliot, 1905; Percy Groom, 1893. ' See Chap. XCIX. " See p. 105. * Gobel, 1891 ; Lazniewski, 1896. CHAP. XXX TRANSPIRATION IN LAND-PLANTS iii ronceals its stomata in hairy furrows. Movements associated with changes in humidity are met with — for instance, in Festuca ovina, Corynephorus ranescens, many grasses growing in deserts, steppes, or on high mountains.^ Dissected leaves, such as those of Artemisia campestris, often possess very similar small, terete segments. 5. The juncoid leaf is long, terete, devoid of furrows, and is seen in species of J uncus, a number of Cyperaceae, and some alpine Umbelliferae in South America. This form of leaf is mostly met with on wet, cold, acid soil that is exposed to wind.- 6. The succulent leaf may be mentioned here because, apart from its thickness, it is often more or less terete, hnear, oblong, or spathulate, devoid of teeth or other indentations ; examples are provided by Sedum acre, Sempervivum tectorum and other Crassulaceae, species of Mesembry- anthemum, Batis maritima, and other halophytes, and some Orchidaceae.^ This form of leaf is characterized by the relative smallness of the surface in comparison with the volume. Henslow's view that succulence is due to the direct action of environment is probably correct.* 7. The sclerophyllous or inyrtoid leaf. There are many other forms of leaves not belonging to any of the preceding types, yet adapted to resist ex- cessive transpiration ; among these may be specially noted the leaves of plants that Schimper describes as being ' stiff-leaved ' or sclerophyllous.^ The leaves may be small (as in Loiseleuria procumbens and Diapensia) ; or narrow and stiff, more or less revolute (as in Lavandula, Hyssopus, and other Mediterranean species) ; or broader (as in Myrtus communis Nerium, Olea, Rhododendron), obovate, oblong, elliptical, lanceolate, or of some other simple form, devoid of teeth or other indentations. They are flat, coriaceous, and stiff, largely owing to the thick-walled epidermis, and are evergreen. To protect themselves against excessive transpiration, such leaves usually have additional contrivances which will be described hereafter.^ Here we may include, too, the leaf-like cladodes of Ruscus aculeatus and other species, also of Semele androgyna. The shoots possessing leaves of the forms enumerated, and especially the pinoid, ericoid, and cupressoid, are usually extremely rich in leaves. By increasing the number of leaves the plant strives to compensate for the decreased assimilation that is caused by reduction in their size. Furthermore, possibly the close aggregation of the leaves on shoots with short segments may itself retard transpiration. (6) Forms of Shoots. Shoots with greatly reduced or caducous leaves occur in connexion with many xerophytes. The foHage-leaf has vanished, and its functions have been taken over by the stem, which produces palisade parenchyma. The epidermis of the stem in this case functions for a number of years. Such aphyllous shoots show several forms : — 1. The winged shoot is often aphyllous, and the light strikes its assimilatory tissue at acute angles.' 2. The switch, or Spartium form of shoot, is switch-shaped, erect, ' See p. 109. - See Gobel, 1889-03, Bd. II. ' Sec Warming, 1897. * See p. 371. * Schimper, 1898, * For further information see papers by Vesque, Volkens, Gobel, Warming. Hen- slow, and Schimper. ' Sec pp. 19, 113, 114. 112 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi slender, and often copiously branched ; the leaves of some species, such as Genista tinctoria and Spartium junceum, are of relatively considerable size, but fall off early when once they have performed their assimilatory functions ; but the leaves of other species are from the outset very reduced in form and function. The stem is terete, or deeply furrowed with stomata and palisade in the furrows and mechanical tissue in the ridges. This form of shoot is very common in Mediterranean Leguminosae, particularly in Genista, Retama, Cytisus, and Genisteae in general, also in Casuarina, Ephedra, a number of Chenopodiaceae, for example in Anabasis (which, however, is mainly halophytic), in Capparis aphylla, Periploca aphylla, and Polygonum equisetiforme.^ 3. The juncoid shoot, as represented by many species of Juncus and Cyperaceae, is tall, terete, aphyllous, and unbranched, being in form similar to the leaf of some of the same species. The relative proportions of surface and volume in such a shoot have already been explained. This form of shoot also occurs in numerous marsh-plants, such as Scirpus lacustris and S. palustris, J unci genuini, and others belonging to the same families.^ The Restiaceae also include shoots belonging to this type. 4. Acicular cladodes of Asparagus. 5. The flattened shoot is an aphyllous form which is upright, or exposes its profile : as examples may be cited Muehlenbeckia platyclada, Phyllo- cladus, and Carmichaelia australis. The stem in some cases (Ruscus, Semele) is so leaf -like that it is preferably included under the category of sclerophylls, a course that has been adopted above. 6. The spinose shoot of Colletia and others. 7. The salicornioid shoot, as represented by Salicornia, Arthrocnemum, and other Chenopodiaceae. 8. The cacti form shoot is met with under various forms in Cactaceae, Euphorbia, and Stapelia. It will be referred to subsequently in connexion with succulent-stemmed plants.'^ III. REGULATION OF ILLUMINATION As light has a heating effect upon the plant and thus promotes tran- spiration, and as intense light is injurious to chlorophyll, many land-plants possess devices by the aid of which the assimilating organs avoid too in- tense illumination. These devices are temporary or permanent in nature : — A. Movements by which Illumination is regulated. Many plants have an extremely deHcate power of appreciating the intensity of light, and can regulate the amount falling upon them by movements of leaflets, which place their blades at definite angles to the incident rays, the intensity of which determines the precise angle. When the light is moderate, as in the early morning, the blades are exposed as fully as possible to the light, whose rays strike them approximately at right angles. But when the light becomes stronger the leaves more and more assume a profile-lie, so that the angle of incidence becomes ^ See the cited works of Pick, 1881 ; Volkens, 1887 ; Schube, 1885 ; Ross, 1887; Nilsson, 1887; Kerner, 1887; Schimper, 1898. =* See Chaps. XLIII, XLV. * See p. 1 23. CHAP. XXX TRANSPIRATION IN LAND-PLANTS 113 increasingly acute ; they are thus relatively less illuminated and heated, and transpiration is inevitably decreased. These movements are executed by the compound leaves of numerous plants, and particularly of some growing in tropical, dry bushlands ; among such are many species of Acacia, and other Mimoseae, Papilionaceae, Oxalidaceae (including Oxalis Acetosella), Zygophyllaceae ; but simple leaves of some plants, including Hura crepitans, hkewise execute movements dependent on the intensity of hght.^ The leaves of the plants in question are wont to be more or less mesophytic in structure, for example, the leaflets of species of Acacia in the West Indies ; acacias endowed with the power of movement in response to intensity of hght are often or always thin, and have a smooth, thin epidermis.- Temporary vertical positions that must be of utility to the organs concerned are met with in connexion with many or most young developing leaves ; for as they shoot from the bud they are vertical, and sometimes remarkably so. B. Fixed Lie in Relation to Light. All foliage leaves as they unfold execute movements as a result of which they assume a favourable fixed he, in regard to which Wiesner has conducted investigations for many years.^ They place their blades perpendicular to the strongest diffuse light. In unusual circumstances, when more intense light prevails, they assume a profile-lie. A diminution in the action of the sunhght, and consequently in transpiration, will result from a permanent profile-pose or other similar arrangements on the part of assimilating surfaces, which thus receive the intense midday light at acute angles. This is the case with the so-called ' compass-plants ', represented in northern Europe by Lactuca Scariola, the leaves of which, in places exposed to strong sunlight, place themselves in the meridian with their faces vertical^ ; as a North American ' compass-plant ', Silphium laciniatum may be mentioned. Leaves exposing their edges to the light are met with in many other plants, for instance, in some species of Eucalyptus, phyllodinous species of Acacia, and Proteaceae, in Austraha ; species of Statice in South Africa ; Laguncularia racemosa in the West Indies ; and Bupleurum verticale in Spain. Leaf-blades that are erect, or directed sharply upwards, are common among xerophytes growing in intense sunlight : for instance, in the West Indian Coccoloba uvifera,^ in many grasses (including Brachypo- dium ramosum, Festuca ovina), in Calluna, Peucedanum Cervaria,^ and Hehchrysum arenarium ; among marsh- and moor-plants may be named Iris Pseudacorus, Narthecium ossifragum, and Totieldia ; among halo- phytes, Rhizophora and other mangrove plants.' More rarely, blades hanging vertically downwards are met with. The switch-plants may again be mentioned as being designed upon a very similar plan. Corrugations and folds in the lamina may play the same part, and become more frequent the drier the chmate is : as examples may be ' See C. Darwin. 1880. '' Warming, 18996. ' Wiesner, 1876. * Stahl, 1880-81. " Illustrations in Borgesen and O. Paulsen, 19CXD. " According to Altenkirch, 1894. ' Illustrations in Joh. Schmidt, 1903 WARMING I 114 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, iii cited Myrtus bullata in New Zealand, Lippia involucrata and Plumeria alba in the West Indies^, Salvia, Stachys aegyptiaca, Pulicaria and Urginea undulata in the Egyptian desert, ^ also Vicia Cracca in Europe. The lie in all the cases mentioned above is attained by means of torsions, folds, and curvatures that take place only during the develop- ment of the individual ; hence, in all kinds of plants which assume the shapes here described, the lie of the leaves varies with the nature of the habitat. Exposed to sunlight, drought, or wind, the leaves become more erect, or have their faces more vertical, or become more crumpled, than when either in the shade or in a humid habitat where the air is moist. This is the case with Calluna, Juniperus communis, Lycopodium Selago and L. alpinum.^ The leaves of Tiha argentea on which hot sunlight falls expose their edges to the light, but the remaining leaves present their fiat faces.* Hereditary profile-lie is met with in Austrahan phyllodinous acacias, the blade-hke petioles of which have their faces vertical, but bear no lamina ; also in many plants with flattened or winged stems, or with decurrent leaves, as in Baccharis triptera in Brazil, Genista sagittalis, Muehlen- beckia platyclada, Carmichaeha austrahs, and species of Colletia. These forms of shoots are usually aphyllous ; stem replaces the leaves. In this connexion must be mentioned the ensiform leaves of Iridaceae, Tofieldia and Narthecium. IV. INVESTING ORGANS IN THE CONTROL OF TRANSPIRATION It is evident that transpiration will be very materially reduced when the transpiring surface is clothed by air-containing bodies, in and between which the air is so firmly lodged that its circulation is obstructed.^ This method is adopted in various ways by many land-plants.^ A. Investing Hairs. In regard to hairiness the contrast between hydrophytes and xerophytes is especially marked : the former are glabrous, the latter often clothed with grey or white cottony and woolly hairs, or by ghstening silky hairs ; the optical effects are connected with the presence of air in and between the hairs. Only dead air-containing hairs are fitted to perform the function in question. In form, these hairs are extremely diversified.'' It has long been known that species which are elsewhere glabrous become hairy in dry places, and that hairy species become more hairy in dry than in moist sites, as is exemplified by Ranunculus bulbosus, Polygonum Persicaria, Mentha arvensis, and Stachys palustris ; moreover, etiolated potato-shoots are nearly glabrous in moist, but hairy in dry air.^ Woolly hairs coat many plants growing on rocks in Mediterranean countries (Corsica,^ for instance), in the dry bushlands of the West Indies, in the desert, steppe, or alpine situations. ^° The most thickly felted plant perhaps is Espeletia,!^ one of the Compositae, Hving on high mountains > Johow, 1884. ' Volkens, 1887, ' See illustrations in Warming, 1887. ' Kerner, 1887-91. * See Haberlandt, 1904, p. iii. ' See Burgerstein, 1904, p. 208. ^ See illustrations in Kerner, 1887-91. * Vesque et Viet, 1881. ' Rikli, 1903. '" See Lazniewski, 1896; Gobel, 1889-93, Bd. ii. " See illustrations in Gobel, loc. cit. CHAP. XXX TRANSPIRATION IN LAND-PLANTS 115 in South America. The woolly coat acts as a sunshade, and serves to moderate changes in temperature, also to reduce transpiration. As a particular form of hair may be mentioned the scaly hair, which when abundant lends a metallic lustre to plants such as the Elaeagnaceae, also some species of Croton and Styrax. The coating of hairs is almost invariably densest on the lower face of the leaf where the stomata occur. Young stems and leaves are frequently densely coated with hair, more densely than when older, as at the former stage they have greater need of protection against intense transpiration. Sometimes, in dry parts of tropical countries, the leaves produced just after the dry season and those developed later on differ widely in appearance, the former being more hairy, and the latter larger and greener. ^ But one group of xerophytes, namely, succulent plants such as Cacteae, species of Aloe and Agave, are usually smooth and quite devoid of any coating of hairs ; in this case other protective measures are adopted. The production of hairs, like all other self-regulatory devices of the plant, is possibly a direct adaptation to external conditions. According to Vesque, hairiness and dryness of the atmosphere increase side by side. Following Mer, Henslow - attributes the production of hairs to ocal supply of nutriment, which is correlated with suppression of paren- chyma ; according to him, the more the parenchyma is checked the greater is the compensatory production of hairs. But even if this hypothesis be correct it does not carry us much nearer to the compre- hension of the correlation between hairiness and dryness. B. Investing Leaves. The young parts of the shoot are usually protected against intense transpiration and intense light by older leaves. It is a quite general phenomenon for the youngest foliage-leaves to be protected by older ones in so-called ' open ' buds ; on the other hand, there are numerous buds that are provided with thick hud-scales, such as are met with in deciduous woody plants, not only in temperate and frigid countries, but also, though less frequently, in the tropics.^ By the production of cork, hairs, resin, and the hke, they are adapted not only to protect the young leaves resting within the bud from transpiration, but also during foliation to guard the buds against change of temperature."* In certain climates bud- scales are rare ; for instance, Coville ^ writes in reference to the Death Valley, ' scaly buds are almost unknown in the desert shrubs '. The same is true of Mediterranean countries, where the rainfall takes place in winter, and of the tropical rain-forest.*^ The young bud-parts of many xerophytic mosses are protected by white hairs that clothe the tips of the old leaves.' Stipules and leaf-sheaths (the latter, for instance, sheltering the young inflorescences of dune-grasses) may perform the same service, though tliey are not strictly included in the category of bud-scales ; ^ in this way the ' H. Schinz, 1893. ' Henslow, 1S94, 1895. ' Illustrations by Warming, 1892. ♦ Griiss, 1892 ; Feist; Cadura; Percy Groom, 1893. ^ Coville, 1893, p. 53. " Schimper, 1898 (1903. pp. 329-5 0- ■' See pp. 109-10. • See illu.strations in Warming, 1907-9- I 2 ii6 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, iii membranous stipules of species of Paronychia, Herniaria, and other plants, clothe young parts of the shoot with a dense silvery investment. Old leaves and remnants of leaves in many cases act in the same manner. ' Tunic-grasses ' is the term employed by Hackel ^ to designate those grasses in which the lower parts of the leaves remain attached long after their upper parts have died, persisting either as coherent, firmly closed sheaths, or in a macerated condition. These tunics depress transpiration, and store water ; they occur in grasses growing on dune, steppe, or desert, for example, in Nardus stricta, Andropogon villosus, Scirpus paradoxus, S. Warmingii, species of Aristida.^ A similar relationship exists in the Velloziaceae living on the mountain-tops and high plateaux of Brazil.^ In certain South African species of Oxalis the bulbs are invested by peculiarly constructed leaves ^ ; the dead bulb-scales of Tulipa praecox bear a dense felt of hairs. Here we may mention the compact clumps, such as those of the ' cushion-plants ' Raoulia and Azorella, consisting of closely-packed shoots and remains of shoots ; they are met with in subglacial vegetation, and especially in South America, and are often so hard that it is difficult to cut or break them ; in this case one shoot protects another — the old leaves protect the young. Many other methods are adopted to protect the youngest part of the stem and leaves, and are described in the papers cited .^ The roots of many epiphytes, including the asclepiadaceous Concho- phyUum imbricatum, are screened from excessive transpiration by leaves, which cover them closely, and keep them surrounded with moist air.^ The roots of some grasses living in the Egyptian desert, for instance, species of Aristida, Andropogon, Elionurus, Panicum, and Sporobolus, are surrounded throughout their length by a sheath of sand, the grains of which are glued together by an adhesive substance excreted by the root-hairs. To a less marked extent the. same is true of dune-grasses in northern Europe." Volkens ^ interprets this as a device for checking evaporation. V. THE ABLATION OF RAIN-WATER It is of importance to the land-plant that its leaves shall not remain too long wet with rain-water ; it is necessary for their surfaces to dry quickly, if transpiration is to be resumed.^ And there seem to be adapta- tions serving to carry rain-water rapidly away. It is in the tropical rain-forest that such devices are especially met with, though they are perhaps not entirely lacking in temperate countries. Jungner, working in the rainy Kamerun, and subsequently Stahl in Java, arrived at essen- tially the same conclusions. They regard as adaptations : — 1. A smooth cuticle that cannot be wet ; this device is very wide- spread. 2. Drip-tip, as Stahl terms the long leaf -tip terminating a blade that often becomes suddenly narrow ; it is typically represented in . ^ Hackel, 1890. ^ Hackel, 1890. See also illustrations in Warming, 1892. Henslow, 1894. ^ Warming, 1893. ^ Hildebrand, 1884. ^ Also see Lubbock, 1899. * Gobel, 1899. ' Warming, 1907-9. * Volkens, 1887. * Concerning ombrophobous and ombrophilous plants, see p. 3 2. CHAP. XXX TRANSPIRATION IN LAND-PLANTS 117 Ficus religiosa, but also in the most diverse plants (ferns, Monocotyle- dones, Dicotyledones), both in simple and compound leaves. The drip-tip serves rapidly to conduct the rain off leaves that are capable of being wet. Drip-tips are downwardly directed ; and the longer the tip, the more rapidly docs the leaf rid its surface of water. The sabre-like tip leads water away most rapidly, apparently at times in an almost continuous jet. Drip-tips are found neither on leaves that are incapable of being wet, nor among xerophytes. 3. Furrowed nerves that conduct superficial water to the leaf-tip are common. The arcuate course of the nerves in Melastomaceae and others is thus of additional use. 4. Velvety leaves are especially encountered among herbaceous species growing on the ground in forest, also among species forming the lower storey of the forest, where shade and moisture are at their greatest. The epidermal cells project in the form of countless short papillae, which give to the leaf a velvet-hke appearance, and produce a fine capillary system m which the water spreads over the whole blade as a thin film. The consequence is that water can evaporate much more rapidly than if it were not spread out in this manner. But it has been suggested that these papillae also serve to supply the leaf with an increased amount of light, or act as light-perceiving organs.^ CHAPTER XXXI. ABSORPTION OF WATER BY LAND- I PLANTS Submerged water-plants, or their overwhelming majority, exhibit j no organs specially adapted for the absorption of water, whereas the ! opposite is the case with land-plants. These possess adaptations that are described in the succeeding paragraphs. I i. Hypogeous Organs that Absorb Water. Subterranean organs in the form of roots, rhizoids, and myceha are designed for the absorption of water ; so likewise are some rhizomes that have absorbing hairs, like those of Corallorrhiza, Epipogum, Equi- setum, Psilotum, and Hymenophyllaceae. In xerophilous land-plants only few deviations from the normal type occur. Many xerophytes possess deeply-descending roots, which aid them in securing water at great depths during dry periods. This has been observed in the desert of Afghanistan in species of Astragalus 2 ; in the Egyptian desert in the colocynth, whose thin leaves would wither rapidly were it not for the deep roots, in Calligonum comosum, and in Monsonia nivea. Volkens^ observed roots in the Egyptian desert that were twenty times as long as the epigeous organs. The same features are to be seen in plants of the European dunes— for example, in Eryngium maritimum and Carex arenaria ; the latter has two kinds of roots— those that are very slender, branched, and lie near the surface, and others that are less branched, • See Haberlandt, 1905. ' Aitchison, 1887. ' Volkcns, 1887. ii8 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi and descend to a great depth. ^ Some plants in Hereroland, Acanthosicyos ^ for example, possess a specially large root-system which serves to raise the subterranean water that Hes very deep. Tall perennial herbs of the Hungarian steppe are extraordinarily deep-rooted. A peculiar device for the absorption of water is met with in the North African halfa-grass, Stipa tenacissima, the rhizome of which has peculiar epidermal cells, whose function is to absorb water.^ ii. Epigeous Organs that absorb Water. In general, the epigeous parts of land-plants are not fitted to supply the plant with water by absorption, since the more or less impermeable cuticle of the epidermis will permit them to take in water only to a slight extent.'* In ordinary phanerogams the amount of water that can be taken in thus is insufficient to compensate for the loss by transpiration. Exceptions to this are provided by lichens, mosses, and other thallo- phytes, which can endure prolonged desiccation and can rapidly absorb liquid water by their whole surface and may store it up ; ^ even aqueous vapour can be withdrawn from the air by many of them. To many other land-plants living exposed to periodically extreme drought it is of great importance that they should be able to seize the moment, often fleeting, when water is available ; and, as a matter of fact, there are devices enabling epigeous parts to absorb water with ease and rapidity. Hairs that absorb water were shown by Volkens ^ to occur on certain desert-plants, such as Diplotaxis Harra, Stachys aegyptiaca, and Convol- vulus lanatus ; and by Schimper ' in certain epiphytes, including Tillandsia and other Bromeliaceae. These hairs are not cuticularized at their base, and it is at this point that the water enters. The subject has been investi- gated by Mez,^ who regards some Bromeliaceae as adapted to absorb dew, and others to absorb rain. The numerous white hairs of cacti may subserve the same function.^ The same role has been ascribed to salt-glands, which Volkens '^^ dis- covered in the form of characteristic hairs on the leaves of various desert- plants, including Reaumuria hirtella, Tamarix, Cressa cretica, Frankenia pulverulenta, and Statice aphylla. These glands excrete solutions of hygroscopic salts (chlorides of sodium, calcium, and magnesium), which solidify during the day and impart to the plant-parts a white or grey appearance ; at night-time the salt deliquesces because of the increase in atmospheric humidity, and the parts concerned again become green and dotted with numerous drops of solution, even though there may have been no deposit of dew. Volkens expresses the opinion that the plants are thus enabled to absorb water. But Marloth " regards the ^ Warming, 1891, 1907-09 (see illustration). " Schinz, 1893. * Trabut, 1888. ■• Ganong, 1894 ; Wille, 1887 ; see Chapter VII. ^ The case of Sphagnum is discussed in connexion with bogs. See Chapter XLIX. ® Volkens, 1887. ' Schimper, 1884. * Mez, 1904 a. ^ As regards hairs which in temperate Europe are reputed to absorb water reference should be made to Lundstrom, 1884 ; Wille, 1887, and Henslow ; and, as regards the structure and function of hydathodes to Haberlandt, 1904; see p. loi. " Volkens, 1887. " Marloth, 1887 a. CHAP. XXXI ABSORPTION OF WATER BY LAND-PLANTS 119 incrustation of salts as a coating that decreases transpiration, and suggests that in this way the plants rid themselves of a portion of the salts absorbed; and this view is adopted by Haberlandt.^ The velamen of Orchidaceae and Araceae has already been discussed.- In sundry epiphytic ferns and Araceae the aerial roots remain short, grow more or less vertically, and collect among themselves humus, and consequently water .^ Felted iyivestments formed by roots or remnants of leaves, or both, occur in ferns such as Dicksonia antarctica, species of Alsophila, as well as in Velloziaceae, and palms. Some of the plants concerned are obvi- ously xerophytes, and the investment serves not only as a means of protection against transpiration, but also as a device for collecting and storing water.* According to Buchenau the same is true of the juncaceous Prionium serratum (P. Palmita), which grows in the periodically dry river-beds of South Africa. In this category must also be placed the grasses that Hackel '" terms tunic-grasses, which retain water between the macerated or scale-like persistent leaf -sheaths. To this group of devices for the obtaining of water may be added the felted mass of rhizoids of many mosses. Many arenicolous xerophytes, especially arenicolous grasses, form dense tussocks or cushions, which certainly benelit them by collecting and retaining water. Other organs, leaves for example, may likewise be designed to take up rain and dew. In such cases the leaves are usually more or less trough-like or, as in Umbelliferae, provided with large sheaths ; as marked examples may be cited the majority of Bromeliaceae, Pandanaceae, and the sugar-cane ; a specially remarkable form is Tillandsia bulbosa, whose narrow trough-like leaves easily obtain water and convey it to the cavities between the inflated leaf -bases. ^ Particular forms of leaves fitted to take up and hold water are possessed by many epiphytic liverworts. Of these GobeP distinguishes three types, according as the under-lobes of the leaves alone or with the co- operation of the upper-lobes form water-reservoirs, or as the leaves form peculiar bowl-like water-sacs. CHAPTER XXXII. STORAGE OF WATER BY LAND-PLANTS. WATER-RESERVOIRS A VERY important and common device, by which the land-plant is enabled to endure dryness of air and soil, is the construction of tissues or organs capable of conserving water for use when it shall be required for purposes of assimilation and other functions. This type of device is frequent among xerophytes, but completely lacking in hydro- phytes. » Haberlandt, 1904; see also Joh. Schmidt, 1903. ' Sec p. 104. » Gobel, 1891-2; Karsten, 1894. * Warming. i8o3- ' Hackel, 1890; also see figure in Warming, i8q2. • Schimper, 1884, and 1888 a. ' Gobel, 1898-1901. vol. u, p. 58. 120 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi CELL-CONTENTS There are plants and parts of plants, including various Thallophyta and spores of Cryptogamia, that can be killed by desiccation only with very great difficulty, although they are quite devoid of any particular morphological means of protection. This faculty is clearly correlated with the nature of their habitat.^ The more striking features in this connexion are treated in the succeeding paragraphs. Mucilage is common in mucilaginous cell- walls, or in the sap of cells which are frequently large ; it absorbs water readily, but parts with it very slowly, and is therefore manufactured by xerophytes in various organs, including hairs, foliage-leaves, stems, subterranean tubers and bulbs. There is a correlation between the production of mucilage-ceUs inside the plant, and the development of the tegumentary tissue. Cacta- ceae, such as Echinocactus, that have a well-developed hypoderma, possess no mucilage-cells. The mucilage-cells of the Cactaceae are often situated in the edges, the bosses, or other protuberant parts, which are most exposed to drying.^ Possibly acting in the same way as mucilage, there are other substances, such as — Acids — for instance, malic acid in Crassulaceae^ ; Tannin, which abounds in certain desert-plants * ; Salts, in halophytes ; Latex probably plays the same part.^ WATER-TISSUE Land-plants, particularly those exposed to strong transpiration, develop specialized water-storing tissues. True water-tissue is thin-walled, contains water but no chlorophyll, is devoid of intercellular spaces (as no gaseous interchange occurs in it), and its cells are usually very large. It is capable of collapsing when water is abstracted, and of expanding when the cells once more absorb water. Water-storing tissue may be peripheral {epidermal or hypodermal) or internal. Peripheral Water-tissue. The epidermis is the outermost layer acting (except in plants growing in water or shade) as a water-tissue, as was indicated first by Pfitzer ^ and subsequently by Vesque ' and Westermaier.® This view is supported by the facts that the epidermis usually contains no chlorophyll, that it forms a continuous layer which in certain cases is very deep and is directly connected with internal water-tissue, for instance in Velloziaceae.^ The epidermis is specially differentiated in the Graminaceae, Cyperaceae, and Velloziaceae, which have hinge-cells ^° along definite lines on the upper face of the leaf, and especially in the furrows of the upper face ; these cells are larger and much deeper than the other epidermal cells ; they ' See G. Schroeder, 1886; V. B. Wittrock, 1891. ' Lauterbach, 1889. ' G. Kraus, 1906 a. * Jonsson, 1902 ; Henslow, 1894. ^ See p. 125. " Pfitzer, 1872. ' Vesque et Viet, 1881. ° Westermaier, 1884. ' For illustrations see Warming, 1893. " See p. 109. CHAP. XXXII STORAGE OF WATER BY LAND-PLANTS 121 play some part in the rolling and unrolling of the leaf, and probably may function as water-reservoirs.^ Mucilage is present in the epidermis of not a few desert-plants, includ- ing Cassia obovata, Malva parviflora, Peganum Harmala, Zizyphus Spina- Christi, and other plants in the Egyptian desert^ ; in many plants mucilage arises in all the epidermal cells, but in others only in some of these. The mode of origin of the mucilage is not known in all cases ; often it arises from the inner epidermal walls. These swell to such an extent in some xerophytes that the lumen of the cell seems not to be more than about half the volume of the wall or, at least, not so large as the latter ; this is the case in Empetrum, a number of Ericaceae, Loiseleuria procumbens,^ Egyptian species of Acacia and Heseda, and species of Rosa.'* Hairs functioning as water-reservoirs form water-Madders. They occur in a number of African desert-plants, including Mesembryanthemum crystallinum, Malcolmia aegyptiaca, Hehotropium arbainense, Hyo- scyamus muticus, Aizoon, some Resedaceae,^ in many Chenopodiaceae, including Atriplex coriacea, A. Halimus,® A. (Halimus) pedunculata and A. portulacoides,' also as mealy hairs in other Chenopodiaceae, possibly also in Tetragonia expansa,^ and Rochea falcata,^ and others. In their typical form they are large, clear, watery vesicles, which project above the epidermis and glisten in the sunlight ; as their contents are gradually consumed they dry up ; in Atriplex Halimus and some other Chenopodiaceae, as well as in Oxalis carnosa,^^ the shrivelled hairs form an air-containing covering over the lamina. Whether or no all the hairs mentioned function to the same extent as water-reservoirs requires investigation. Hairs of a most remarkable form occur, according to Haberlandt,^ on the roots of an epiphytic fern, Drymoglossum nummulariaefolium. These hairs shrivel during the dry season ; the protoplasm withdraws to the base of the hair and shuts itself off from the dry part by a cell-wall ; when rain falls, the hairs grow out in a few hours and once more become filled with water. Voluminous peripheral water-tissue may arise either by tangential division of the epidermis or by the formation of a hypoderma. It is situated mainly on the upper face of the leaf, and if present on the lower face it is less developed. It checks the penetration of heat-rays rather than of luminous rays, and thereby retards transpiration in addition to acting as a water-reservoir. A multilamellar epidermis occurs frequently among xerophytes, but particularly among lithophytes and epiphytes ; there may thus arise a voluminous tissue, the thickness of which far exceeds that of the chloren- chyma, as in species of Peperomia, Begonia, Ficus, and Gesneraceae.^- Hypodermal water-tissue occurs in other xerophytes. It is composed of one layer of cells in certain Genisteae,^^ Velloziaceae,^^ and Orchidaceac'^; ' Duval-Jouve, 1875; Tschirch, 1882 b ; Volkens, 1887. ' Pfitzer, 1870, 1872; Volkens, 1887. ' Gruber, 1882; E. Petersen, 1908. * Vesque, 1882 a, b, 1889-92. ^ Volkens, 1887 ; Henslow, 1894 : Schinz, 1893. * Volkens, 1887. ' Warming. i8qi, sec illustrations. ic>o6. " W. Benecke, 1901. ' F. AreschouR, 1878. " Meigen, 1894. " Haberlandt, 1893. " Vesque, loc. cit. ; Pfitzcr. 1870, 1872. " Schube, 1885. '* Warming, 1893. " Kruger, 1883. 122 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi of two to three layers in Nerium ; and is very large in certain other plants belonging to the Commelinaceae, Scitamineae, Bromeliaceae, and Rhizo- phoraceae.^ A collenchymatous hypoderma functioning as a water- storing tissue is met with in a number of cacti ; it is traversed by narrow intercellular spaces leading from the chlorenchyma to the stomata. Mucilaginous cork may here be mentioned as a cork-tissue discovered by Jonsson ^ in a number of Asiatic desert -plants. Internal Water-tissue. Water-tissue may occur among xerophytes in various other forms, as is indicated in the succeeding paragraphs. [a) Longitudinal bands of water-tissue extending through the whole thickness of the leaf from the upper to lower faces occur in some desert- grasses,^ and in Phormium tenax. Strips of chlorenchyma, in which the veins are embedded, in this case alternate with the bands of aqueous tissue. In the Velloziaceae similar longitudinal bands connect the epidermis on the upper face of the leaf with the water-containing cells that form a sheath round the vascular bundles.* {h) Central water-tissue, occupying the centre of the leaf and surrounded by a thin layer of chlorenchyma, is met with in many xerophytes, including Aloe, Agave, Bulbine, Mesembryanthemum, Salsola,^ Atriplex, Halogeton, and Zygophyllum. In aphyllous stems the aqueous tissue may be distributed in this same manner, as is the case inSalicornia and Haloxylon.® Water-tissue and chlorenchyma may either be sharply delimited from each other, or may gradually merge, owing to the cells in the interior of the leaf containing but little chlorophyll, as in many Crassulaceae and Cactaceae. Water-storing idioblasts appear in the chlorenchyma of various desert-plants and halophytes.'^ SUCCULENT PLANTS Succulent plants are thick and fleshy forms which are provided with a water-tissue and parenchyma that contains abundant mucilage ; they are xerophytes which have especially pronounced water-storing tissue. They are commonly plump in form, and, like herbs, usually possess green stems which exhibit but feeble production of cork and of lignified tissue — ligniiication and succulence are, in a sense, opposed to one another. They are perennials, and often very long-lived. The cell-sap is rich in mucilage, the epidermis strongly cutinized as a rule, and the stomata are sunken. Succulent plants can store a large amount of water, which they give up extremely slowly, and they therefore dry only with great difficulty. The hottest and driest countries with a regular periodicity of climate are generally their homes.^ We can distinguish two main types of succulent plants, succulent- stemmed and succulent-leaved.^ ^ Warming, 1883; O. G. Petersen, 1893; Areschoug, 1902. ^ Jonsson, 1902 ; also see Haberlandt, 1904, p. 363. ^ Volkens, 1887. " Warming, 1893. " See figure in Areschoug, 1878. * Volkens, 1887; Warming, 18976. ' See figure in Volkens, 1887. * In regard to their adaptive features, consult Burgerstein, 1904, pp. 44, 205. ' Gobel, 1889-93. CHAP. XXXII STORAGE OF WATER BY LAND-PLANTS 123 i. Succulent-stemmed (Chylocaulous) Plants. In these plants the stem is fleshy and jiiic}'. The leaves are suppressed in the most marked types, or they are reduced to scales or thorns ; the stem has assumed the assimilatory functions of foliage, and the trans- piring surface is thereby greatly reduced. The most common and extreme types are Cactaceae in America, Stapelia in South Africa, and species of Euphorbia which occur mainly in Africa. To these may be added the geraniaceous Sarcocaulon in South Africa. In the various genera there occur a series of shapes whose efficiency has been demonstrated by Gobel,^ Noll,- and others. Frequent among such shapes are those like the sphere, prism, or cylinder, that combine smallness of surface with largeness of volume. This is advantageous in relation to storage of water, but disadvantageous in regard to assimila- tion. One stage towards increase of surface and therefore of assimilation is represented by the production of ridges, processes, bosses and the like, in Mammillaria, Echinopsis, and other Cactaceae. These protuberances are set vertically, that is, in a manner which does not render them so easily heated by the sun's rays — and this is of advantage as their internal temperature is often high.^ Here may be included pseudo-bulbs which occur mainly in epiphytic orchids ; they are tuberous green stems, consisting of one or more seg- ments (thus bearing one or more leaves) ; they persist long after the leaves have fallen, serving as water-reservoirs and often containing mucilaginous sap. ii. Succulent-leaved (Chylophyllous) Plants. In plants with succulent leaves the stem is normal in form, except that its internodes are often short and its leaves consequent!}' arranged in rosettes. The leaves are thick, stumpy, sessile, usually elongate and narrow, often cylindrical (if we except the sphere, a prism or cylinder has the smallest surface in relation to volume) ; they often are continued at the margins or apex into thorns, but apart from this are usually un- divided and entire. As examples of plants having their leaves in rosettes we may mention Agave, Aloe, Sempervivum, Echeveria, species of Mesembryanthemum, and some epiphytic orchids ; elongated internodes are developed by Sedum, BryophyUum, Portulaca, and Senecio (Kleinia). Succulent-stemmed and succulent-leaved plants are both represented among haloph}i:es. Succulent plants deviate from other chlorophyll-possessing plants in both respiration and assimilation. The divers structural features that obstruct transpiration at the same time constitute an obstacle to the assimilation of carbon dioxide ; at night-time, during respiration, there is produced only little carbon dioxide but much malic or other organic acid, which is utilized in the manufacture of carbohydrates on the following day.'* ' Gobel, 1889-93. ' ^'oll' '893- ^ Concerning the morphology of Cactaceae, see Vochting. 1874, 1894; Gobel, loc. cit. * Aubert, 1892; Jost, 1903, Lecture 15. 124 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi Succulent plants owe their origin, according to Vesque/ to two direct causes : — 1. Heating of the soil, which increases the osmotic power of the roots (succulent plants can endure without injury very high temperatures, and grow especially on warm rocks). 2. The supply of nutriment in alternately strong and weak solutions. Between succulent plants and xerophytes that are poor in water there are distinctions in appearance quite apart from those in thickness and the like. The former are as a rule of a fresher green (because they are glabrous) ; the latter, on the contrary, white-haired or grey-haired.^ Still there are some hairy succulent plants, Sedum villosum for example. In consequence of the production of wax, glaucous species occur in both groups of xeroph3rtes. BULBOUS AND TUBEROUS PLANTS These must be considered in connexion with succulent plants. They are adapted in a different fashion to endure prolonged periods of drought. In many cases it is not only reserve food, such as starch, but also mucilage- cells or mucilage-tissue which contribute to their fleshiness, and function partly as food-materials for the production of new shoots^ and partly as means of storage of water to provide against desiccation. Bulbous and tuberous plants belonging to the Liliaceae, Iridaceae, Amaryllidaceae, and other families, therefore occur especially in dry countries, and more particularly in South Africa ; also on the steppes of Asia, where they are among the species that develop rapidly after the commencement of spring or of the rainy season. Poa bulbosa is, according to Aitchison,"* the commonest grass on the great plains of Baluchistan, and is certainly enabled to exist there by the aid of the thick leaf -sheaths that form a kind of bulb. Marloth ^ suggests that South African bulbous plants are designed to resist the enormous pressure to which they are subjected by the drying soil ; for some of them, namely Cape species of Oxalis, are protected by a hard coat, others by numerous, superimposed, soft, finely- fibrous layers whose bundles of bast persist as rigid bristles. In South Africa there are many remarkable, partly epigeous tubers (certain stem- tubers) which in their leafless condition are distinguishable only with difficulty from the stones among which they grow ; as an example may be mentioned Dioscorea (Testudinaria) Elephantipes, which is protected from desiccation by huge cork structures. Likewise belonging to the category of epigeous tubers are the tuberous or at least swollen trunks of certain South American trees, occurring in the Caa-tinga forests, and including Chorisia crispiflora and Cavanillesia arborea (Bombaceae), Jaracatia dodecaphylla (Caricaceae),^ also Jatropha podagrica (Euphor- biaceae). Many tubers consist of root and stem combined, and thus lead the way to those consisting of root alone ; such is the nature of the lignified tuber (' xylopodium ' ') in many herbs and small shrubs in South American savannahs.^ In Crocus and other Iridaceae one sometimes sees clear, ^ Vesque, 1883. * See p. 114. ^ Tubers of this kind occur even in aquatic plants such as Sagittariasagittaefolia. * Aitchison, 1887. * Marloth, 1887 ; see also Hildebrand, 1884. * See figure in Warming, 1892. ' Lindman, 1900. CHAP. XXXII STORAGE OF WATER BY LAND-PLANTS 125 fusiform, juicy roots radiating from the tuber ^; these are also found on the bulbs of certain species of Oxalis,- and in the cactaceous Cereus tube- rosus, whose shoots cannot store much water but whose roots are tuberous, juicy, and enveloped by a sheath of cork. South African xerophytes have upon their long roots many fusiform or spherical tubers which are water-reservoirs encased by cork ; Elephantorrhiza has close beneath the surface of the soil a water-reservoir of this kind, which weighs up to ten kilogrammes, although the stem is scarcely a foot in height ; while a species of Bauhinia produces tubers weighing fifty kilogrammes.^ In Egypt there are species of Erodium with root-tubers which, according to Volkens,^ serve to store water. Spondias venulosa has gigantic sub- terranean tubers. In temperate Europe, Sedum maximum possesses thick fleshy roots. In some plants there have been discovered dwarf-roots that have been, correctly or incorrectly, regarded as water-reservoirs ; among such plants are Aesculus and some allies,^ some Australian conifers,^ and Sedum.' Dimensions of water-reservoirs vary greatly according to the part they have to play in the life-history of different species ; in some cases they necessarily function for months or even years without intermission, but in others — for instance, leaves of trees in tropical rain-forest — only for a few hours of the day ; some resign water rapidly, others slowly. The structural features must necessarily harmonize with these differences. Combinations of xerophilous characters — for instance, anatomical with morphological — are universal ; indeed, some characters demand the pre- exist ence of others before they can arise. Correlations. One character often entails another. For example, with peripheral water-tissue there appear accessory cells in connexion with the stomata, so that the latter may be protected when the plant-member shrivels as it dries .^ LATICIFEROUS PLANTS Up to the present we have dealt only with watery or slimy cell-sap in which various salts may be dissolved. But mention must be made of those plants that contain ' latex ', which is usually white and is con- tained in tubular organs (laticiferous cells and syncytes). The functional significance of latex is unknown, indeed it is probably multiple, and one function may be the protection of the plant against desiccation. In favour of this view^ is the fact that laticiferous organs are frequent in the tropics, particularly in hot, dry districts, and often precisely in plants which are thin-leaved and apparently lack any other means of replacing the water lost by transpiration.^ The occurrence of latex in subterranean bulbs, as in Crinum pratense,!^ harmonizes with this view when these bulbs grow in clay which becomes fissured during the dry season. ISOLATED WATER-STORING CELLS. NERVE-ENDS The succulent plants hitherto discussed possess coherent water-tissue, an arrangement that seems to be the most efficient ; laticiferous plants » Raunkiiir, 1895. Illustrations, 1905, 1907. ' HUdcbrand, 1884. » Schinz, 1893. * Volkens, 1887. ' J. Klein, 1880. " Berggren. 1887. ' Warming. 1891 : see illustrations, 1907-09. ' W. Benecke, 1892. * Warming. 1892. '" Lagerheim, 1892. 126 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, iii have long, tubular, branched receptacles. But there are still other kinds of water-reservoirs. Certain plants contain, scattered about in their general chlorenchyma, solitary or grouped cells which are larger than the other cells, have thin walls and clear contents. Such occur in Nitraria retusa, Salsola longifolia, Halogeton, Zygophyllum, and other plants belonging to the Egypto-Arabian desert,^ also in Barbacenia growing on mountains in Brazil,^ as well as in loranthaceous parasites.^ In certain cases it has been established that if a slice of a leaf be allowed to dry these cells collapse, but expand again when water is supplied. Lignified idioblasts (tracheids^) with spiral or reticulate thickenings occur in numerous other species, and are usually scattered in the same manner ; they resemble the cells of the velamen of orchids^ and the porous cells of Sphagnum,^ being short, tolerably thick-walled, but not perforated ; they fill with air when the contained water passes out. They show two types of distribution, being either grouped at the ends of vascular bundles or dissociated from these. The latter is the case in the leaves of many tropical orchids,' species of Crinum,^ Nepenthes,^ Sansevieria, Capparis and Reaumuria,^" Salicornia,^^ and Centaurea.^^ They occur in other xerophytes and halophj^es close to the ends of vascular bundles ; in this position, especially in desert-plants, they assume the form of huge, irregular tracheids with slit-like or elongated pits, and, as they stand above the delicate blind ends of vascular bundles in the leaf, they are often difficult to distinguish from tracheids apper- taining to the bundles : arrangements of this kind occur in species of Capparis and Caryophyllaceae.^^ These water-storing tracheids seem to play the same role as do wood-vessels in vascular bundles, since they fill with water and give it up again without collapsing. The -parenchyma-sheath surrounding vascular bundles may perhaps function as a water-reservoir in some Egyptian desert -plants ^* and in Restiaceae.^^ Translocation of water. Meschajeff ^® seems to have been the first to point out that the older leaves of succulent and semi-succulent plants often serve as water-reservoirs for the benefit of younger leaves ; for in times of prolonged drought water is transferred to the younger parts of the shoot from the older leaves, which shrivel and die. ^ Volkens, 1887. ^ Warming, 1893. ^ Marktanner-Turneretscher, 1885. * ' Reservoirs vasiformes ' of Vesque, 1882 ; ' spiral cells ', ' storage-tracheids ' of Heinricher, 1885. * See p. 104. « See Chap. XLIX. ' Trecul, 1855 ; Kriiger, 1883. ' Trecul, 1855 ; Lagerheim, 1892. ^ Kny u. Zimmerman, 1885. " Vesque, 1882 a, b. " Duval- Jouve, 1868. '' Heinricher, 1885. " See Vesque, 1882 a and b ; Heinricher, 1885 ; Kohl, 1886; Volkens, 1887 ; Schimper, 1898 ; Haberlandt, 1904. '* Volkens, 1887. ' Gilg, 1891. " Meschajeff, 1883 ; see Burgerstein, 1904, p. 228. 127 CHAPTER XXXIII OTHER STRUCTURAL CHARACTERS AND GROWTH-FORMS OF LAND-PLANTS, AND ESPECIALLY OF XEROPHYTES It has already been pointed out that some of the structural features of the growth-forms of land-plants are of such a nature that, while no one can doubt their connexion with a dry environment, their utility to the plant is not obvious. As a case in point we may cite the histology of the sun-leaf (heliophyll),^ where, as we know, there is a greater depth and number of layers of the palisade cells in the sunlight than in shade, and in dry air than in moist air, with the correlative greater thickness and smaller intercellular spaces of the spongy parenchyma, less-sinuous walls of the epidermal cells, and other characters. Among features of this problematic nature may be mentioned lignification, which is so common among land-plants and so restricted in aquatic plants. LIGNIFICATION Lignification is a mechanical utility because it increases the plant's power of resisting mechanical force. In many plants, including trees, it is of service in connexion with the storage of water. The birch and the Common spruce are splint-wood trees with shallow horizontal roots ; it therefore suggests itself that the supply of water is controlled by the splint-wood. It is worthy of note that lignification stands in direct relation to environment, for it becomes more extensive the drier the habitat (except in succulent plants). Families such as the Umbelliferae, Caryophyllaceae, Linaceae, Labiatae, Rubiaceae, and Dipsaceae, as well as genera which in temperate countries are rich in herbaceous species, become far richer in woody plants in tropical, warm-temperate, or even in Mediterranean countries, Lignification is particularly extensive in xerophytes that contain but little sap. In these the wood is dense and hard, but at the same time often brittle. It resembles the so-called ' autumn-wood ', because the lumina of the vessels and cells are narrow, and the reason for resemblance is presumably that the conditions of development are identical ; the narrowness of the constituents is correlated with weakness of transpira- tion, which is due to great reduction in the leaves and unfavourable conditions of growth.^ According to Cannon ^ ' the branches of irrigated plants in the desert about Tucson are poorer in conductive tissue than branches of the same diameter of non-irrigated plants '. The explanation of this must be sought in the difference ' in the length and character of the growing season of the two classes of plants '. What benefit desert- plants derive from this structural character of the wood is not clear. It must, however, be remembered that lignified parts withstand extreme temperatures better than watery thin- walled parts can, and that trees endure great fluctuations in humidity better than herbs do. Mechanical tissue is developed in the form of bundles of bast above and ' See p. 21. ' Henslow enumerates the peculiarities of desert-plants, 1894. ' Cannon, 1905. 128 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, in beneath the veins of leaves in land-plants, below or in the epidermis, and is more extensive the drier the climate, or the more transpiration is favoured ; by the environment. Parts of the fundamental tissue are developed as mechanical tissue in stems of some plants, including Restiaceae.^ Stone- cells and mechanical cells are differentiated in the chlorenchyma more or less as idioblasts of various forms, which Vesque ^ distinguishes by such terms as ' proteoid ', 'oleoid', and the like; they occur in leaves of Proteaceae,^ Rhizophora,^ Restiaceae, Olea europaea (as long, sinous, sclerenchyma-cells, which interweave and run both parallel and perpendi- cular to the surface), Thea, and others. In some cases the utility of these cells with thick, lignified walls is obvious ; it is to prevent the shrivelling, collapse, or distortion of the vitally important chlorophyl- containing tissue that would otherwise take place when the plant-member withered. The strong construction of the epidermis in sclerophyllous plants also performs the same service. In the production of thorns xerophytes show also their tendency towards lignification. It has been recognized that plants living in deserts and similar places are very thorny, and possess stiff, spiny or prickly leaves, thorny stems, and the like. Such plants are characteristic of the scrub in Australia, of stony steppes and high plateaux in Asia (the ' Phrygana-vegetation ' of Theophrastus), the Kalahari, the deserts |" of Egypt and North America, and others. Thorns vary widely in their ; morphological nature, and may represent complete leaves or portions of these, or emergences and prickles, or hgnified stems which are vegetative axes or flower-stalks ; in accordance with these features various growth-forms — for example Grisebach's ' thorn-shrub ' and Reiter's j ' thistle-form ' — have been defined by different authors. Thorns, according to Lothelier's ^ researches, are evoked by dryness , of atmosphere ; species such as Berberis and Crataegus, which are thorny ; in dry air, become thornless in moist air. It has long been known that spiny plants often lose their spines under cultivation on improved soil.*^ Nearly all those who have investigated the subject express the view that, while thorns play no direct part in assimilation, they can hardly be regarded as useless organs, since they presumably serve to protect the ; plant against animals."^ Wallace ^ points out that thorny shrubs are I especially abundant in those parts of Africa, Arabia, and Central Asia where large herbivora abound. It seems to be beyond doubt that this view is correct in certain cases, and that, for instance, the long spines of Acacia horrida, A. giraffae, and other species in the dry tracts of South Africa serve as defences against numerous wandering herds of ungulates ; Marloth ^ calls attention to specialized adaptation exhibited by certain species in that the longest and strongest spines occur on young individuals or on root-shoots, which are most accessible to animals, while the branches subsequently produced on tall trees are quite devoid of spines. A similar phenomenon is witnessed in connexion with Ilex Aquifolium, the upper leaves of which are usually not prickly when once the plant has grown , ' Gilg, 1 89 1. ' Vesque, 1882. * Jonsson, 1880. | ' Warming, 1883. ' Lothelier, 1890. , ^ See Henslow, 1894, p. 223 ; Vesque et Viet, 1881, ' Delbrouck, 1875 ; Marloth, 1887. « Wallace, 1891. * Marloth, 1887. CHAP. XXXIII SOME STRUCTURAL CHARACTERS 129 into a tall tree. It is evident that spiny plants by reason of their armed nature may defeat unarmed species and become more widely distributed ; but for all this we are not entitled to assume either that thorns are a direct adaptation to animals, or that they could arise by natural selection in a country rich in herbivorous animals. For example, against what animals did the Cactaceae and Agaves of Mexico and the West Indies require to defend themselves when they were evolved ? Would heredity have preserved these useless characters throughout the vast periods of time that may have elapsed since ungulates, which have recently been re-introduced, abounded in these lands ? (It is incontestible that spiny structures are now of use to Mexican succulent plants in protecting them from ungulates during the prolonged dry season). Kerner ^ assumes that the Mediterranean region is rich in thorny plants because animals also abound, and that on high mountains the absence of thorny vegetation is associated with the greater poverty of animal life. But in arctic countries there are many herbivorous animals, including large ones such as the reindeer and musk-ox, which roam about in great herds, yet no thorns occur, obviously because the conditions of humidity prevailing here and on high mountains do not conduce to the production of thorns.- In the north-temperate moist climate there occur many thorny growths the significance of which is at present obscure. This is likewise true of the strong spines of many palms, including Astrocaryum and Bactris, growing in Amazonian forests. There are other thorns whose definite utility can be demonstrated, and such is the case with those on stems of certain lianes. The physiological reason for the strong development of lignified constituents is still somewhat obscure. But intense light and vigorous transpiration seem to be the causes. Vesque and Viet,^ and subsequently Kohl,"^ and Lothelier ^ experimentally proved that mechanical tissue increases when transpiration is greater. Cockayne ® found that in the rhamnaceous Discaria Toumatou thorns are not developed in moist air. Stahl,' Dufour,^ and Lothelier^ found that mechanical tissue is more strongly developed in light than in darkness : etiolated plants are very weak-stemmed. On the other hand, experiment showed that with an increased supply of water there was a diminished production of wood in the oak and Robinia, and a reduction in the development of mechanical constituents among monocotyledons.^" STUNTED GROWTH. SCRUB. CUSHION-PLANTS It has already been mentioned " that lack of water, and strong transpiration, induce stunted growth. Wind, deficiency of water, and other conditions unfavourable to growth, bring into existence elfin-wood scrub, heath-scrub, ericaceous shrubs with bowed branches, and such growths as that of malformed and stunted Pinus sylvestris as it occurs in north-eastern Germany. Dry soil and strong transpiration impart to these ' Kerner, 1869. * Sec Warming, 1892; Hcnslow, i8g4 ; Cockayne, 1905. ' Vesque et Viet, 1881. * I^ohl. 1886. ■' Lothelier, 1890. ' Cockayne, 1905. ' Stahl. 1883. * Dufour, 1887. • Lotheiier, loc. cit. " Grabncr, 1895. " See pp. 29, i7. WARMING K sett 3se1 tit 130 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect. h.\p plants their characteristic habit, by inducing short, curved, and crool f^' shoots and stems, with short internodes, and with a feeble or irregi production of buds : abundant moisture causes shoots to be long i possessed of long internodes. In Mediterranean and other subtropi countries with winter-rain, many species assume the form of shrubs I ^S^ medium height, but in moist valleys they vary from this fo |iWO up to that of a tall tree. In scrub or in the desert, the branches a leaves are often closely packed, the ramification being extraordinar dense, and the plant as a whole being compact and rounded in fo] (hemispherical, or cushion-like) ; as examples may be cited Achil' fragrantissima, Artemisia Herba-alba, and Cleome arabica,^ all in t North African desert ; the globular bushes of Astragalus and Genist in Corsica ; Draba alpina,^ Silene acaulis,* species of Saxifraga, ai many mosses of cushion-growth in arctic countries ^ ; Androsace helveti( and others in the Alps.® The high mountains of South America and all other lands display many examples of cushion-like shrubs or herl which appear as if cleanly bitten or clipped ; for they are rounded 0 dense in growth or even solid, and have their numerous shoots, leave and remnants of these closely packed together : as examples may , cited the umbelliferous Azorella and Laretia, species of Oxalis, ar Cactaceae in South America. One of the most remarkable cushion-plan' is Raoulia mammillaria living in New Zealand.'^ Everywhere the cause is the same — dryness, occasioned by one c another factor. Dense ramification and tufted growth confer a benef upon the plants, in that their young shoots are thereby better shielde from transpiration ; they protect each other and are in turn protecte by older shoots from the desiccating action of wind in arctic countries. tk imii but pla tilt or ad art t ROSETTE-PLANTS Many xerophytes have their leaves arranged in rosettes on shoot which resemble the first year's growth of biennial dicotyledons : rosettt plants are encountered in arctic countries, at alpine altitudes, on steppe and deserts, among epiphytes and tropical lithophytes.^ The brevit of the internodes and the consequent arrangement of the leaves canno en perhaps always be explained in the same manner, nor is the utility alway identical. In many Bromeliaceae the rosette serves to collect and retai: water ; in other plants, such as Agave, it may be that the leaves formin, the rosette are better screened from the sun and from excessive transpira tion. In arctic and alpine plants the low rosette-shoot may benefi because the leaves spreading over the soil are not so much exposed t< desiccating winds, also because these leaves are situated in warmer ai and are better able to obtain heat from the soil. It is probable that ii the desert they can utilize to good advantage the dew deposited by night Meigen ^ also remarks that the leaves of many rosette-plants by over lapping one another produce niches screened from the wind, and thu reduce their transpiration. Rosette-plants thrive among open and lov ' Volkens, 1887. * Massart, 1898 ; Rikli, 1903. ' Figured in Kjellman, 1884, p. 474. * Figured in The Botany of the Faroes (Copenhagen, 1901-8), p. 993. * Andersson and Hesselnaan, 1900. * Schroter, 1904-8. ' See Section IX. * See p. 27. ' Meigen, 1894. d HAP. XXXIII SOME STRUCTURAL CHARACTERS 131 >getation ; the grasslands of northern and central Europe and of the ;ps, and similar types of vegetation, are very rich in low, perennial sette-herbs such as Plantago major, Taraxacum officinale, Achillea illefolium, and Pimpinella Saxifraga, species of Primula, Draba, Saxi- aga. Bonnier ^ showed that certain species which in the plains have ■loots with long internodes, when grown at alpine heights produce osettes. PROSTRATE SHOOTS Many species growing on dry, warm, sandy soil have prostrate shoots, t least so far as these are vegetative. As was shown on p. 26 this is 0 be attributed to the thermal relations prevailing in the soil. CHAPTER XXXIV. OECOLOGICAL CLASSIFICATION 2 eli The foregoing chapters have made it clear that the distinctions ^between water-plants and land-plants are deep-seated, and concern the external form as well as the internal structure. Plant-communities must therefore be grouped in the first place into aquatic and terrestrial ; but between these there is no sharp boundary, for there is a group of plants, marsh-plants {helophytes), which, like water-plants, develop their lower parts (roots, rhizomes, and, to some extent, leaves) in water lei or at least in soaking soil, but have their assimilatory organs mainly lei adapted to existence in air, as is the case with land-plants to which they are closely allied. Helophytes give rise to special forms of communities. Yet we must include among water-plants all those plants that, like Nymphaeaceae, approximate to land-plants in so far as they have floating- leaves, which are more or less adapted to existence in air, but are never- theless mainly designed for existence upon water. It has already been shown that land-plants exhibit many grades of adaptation to their mode of life in contact with air, and that those which encounter the greatest difficulties in regard to securing water are termed vxerophytes ; while others are described as mesophytes because in some HI respects they stand midway between the two extremes, hydrophytes njand xerophytes. The differentiation of the land-plant in one or the other direction is decided by the oecological factors, edaphic and climatic, that prevail in the station or habitat. But edaphic and climatic factors cannot be regarded separately : the plant-community is always the pro- 11 duct of both together. The nature of a soil is also influenced by climate, ii and it is incontestible that climate (rainfall) calls forth the wide differences i|between, say, desert and tropical rain-forest. But it is far from being rue that climate alone calls into existence the different communities f plants which will hereafter be defined as jormations. Characters of ' Bonnier, 1890, 1894, ■ In the classification of plant-communities these are grouped into successively mailer subdivisions that are, onlv to some extent, analogous with systematic iamiUes, genera, species, and varieties. The most comprehensive group is termed n German a ' Vereinsklasse ' or ' Formationsklasse ', wliicli we propose to translate is ' oecological class ', or when the context permits, as ' class '. K 2 132 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi the soil are of supreme importance in determining the production of formations, and they must therefore be the foundation of oecological classification. Clements,^ with reason, has objected to Schimper's scheme of distinguishing between climatic and edaphic formations, if indeed it was Schimper's meaning that a sharp distinction is throughout possible, and that both groups of factors are of equal potency. The importance of soil in determining the development of definite plant-communities is clearly revealed in the topographical distribution of these ; there is not a single community of land-plants that extends without interruption over great stretches of land ; all are discontinuous and, according to the nature of the soil, interrupted by other communities, however uniform the climate may be. On the other hand, it is often the case that one and the same formation, either of water-plants or of land-plants, is developed in very different climates. The differences existing in the climate of various parts of such a country as Denmark are quite inadequate to account for the great differ- ences in vegetation. A prominent part is played by chemical differences in soils (amount of common salt), also by their fertility or amount of nutrient salts ; for instance, soil poor in food-material favours the preponderance of heath. Grabner ^ regards the percentage of nutritive salts dissolved in the soil-water as the factor controlling the character of vegetation ; he therefore divides formations into three great groups according as the water is rich or poor in mineral matter, or contains common salt : 1. Formation, where the water is rich in mineral salts ; 2. Formation, where the water is poor in mineral salts ; 3. Formation, where the water is saline. The same opinion is expressed by A. Nilsson^, but this view seems to be based chiefly upon observations in the open air, and too little tested by soil-analyses. Grabner apparently exaggerates the importance of the factor in question, for his scheme would, on the one hand, unnaturally { separate formations, such as those of sour-meadow, heath-bog, and ling- ■ heath, which belong together, while, on the other hand, it would un- naturally group together formations, such as those of sand-field and peat vegetation, which are not allied. It would appear that the most potent and decisive factor is the amount of water in soil ; and this, in turn, depends upon the depth of the water-table and upon the physical characters of soil (its water-capacity, the amount of air-content, the plants and animals living in it, the production of humus, and the like). We must, however, admit that climate may favour a certain formation by causing this to become less exacting as regards its edaphic require- ments, and consequently enabling it to be distributed over a large area on very diverse soils ; whereas in another climate this formation will be vanquished by communities better adapted to localities where certain special soils occur, and will be more broken up and restricted in its dis- ^ Clements, 1899, 1904. Clements (p. 27) writes : 'From the above it follows that Schimper's so-called climatic formations, forest, grassland, and desert, are merely a somewhat incomplete expression of water-content association. As to the validity of his division of all formations into climatic and edaphic, there is also room for grave doubt ... all plants . . . are primarily influenced by soil, i.e. they are edaphic' * Grabner, 1898, 1909. * A. Nilsson, 19026. CHAP. XXXIV OECOLOGICAL CLASSES 133 tribution. In this sense we must interpret Schimper's division of forma- tions into climatic and edaphic. As an example, it may be mentioned ! that Whitford and Cowles ^ interpret coniferous forest in the eastern I United States as an edaphic, xerophytic formation occurring in the ' district where deciduous forest prevails ; but in the entirely different t Hmate near the Paciiic coast of the United States coniferous forest is I dominant, while the deciduous trees give rise to edaphic, mesophilous forest lining the watercourses. The occurrence of the same species in different formations renders explanation difficult ; Cowles, no doubt correctly, states that a species ' in general can grow in the largest number of formations at its centre of distribution, since there the climatic conditions favour it most highly. In other regions, especially near its areal limits, it can grow only in those formations which resemble most closely in an edaphic way the climatic features at the distribution centre.' Cowles also contends that in many rases a species can occupy very different soils (for instance, clay or dune- sand) only when the atmospheric conditions are the same, and that conversely on one and the same soil very different vegetation may prevail when the atmospheric conditions are changed. An exception is provided i by humus soil, for there are many species to which humus is absolutely essential. As an example showing that a species may inhabit entirely different kinds of soil, we may quote C. Schroter's ^ statement in reference to the mountain-pine. He writes : ' How fundamentally diverse are The habitats of the mountain-pine which can grow on the dry, loose, ilcareous talus of a hot southern slope, and on high-moors, which are iuainly composed of bog-mosses and are subalpine bogs dripping with vater but poor in mineral matter. The former soil is poor in humus, ich in mineral matter, and dry ; the latter is a substratum rich in humus, [loor in mineral matter, and always saturated with water. Common to ijoth is only one character — poverty in assimilable nitrogen.''^ Wlien endeavouring to arrange all land-plants, omitting marsh- plants, into comprehensive groups, we meet with, first, some communities that are evidently influenced in the main by the physical and chemical characters of soil which determine the amount of water therein ; secondly, other communities in which extreme climatic conditions and fluctuations, seasonal distribution of rain, and the like, decide the amount of water in soil and character of vegetation. In accordance with these facts, land-plants may be ranged into groups, though in a very uncertain manner. The prevailing vagueness in this grouping is due to the fact that occology is only in its infancy, and that very few detailed investigations of plant-communities have been conducted, the published descriptions of vegetation being nearly always one-sided and Horistic, as well as very incomplete and unsatisfactory from an oecological standpoint. It is to be hoped that Clements's remarks, in his Research Methods in Occology will stimulate detailed research and will banish the ignorance to which he alludes in the following terms : ' Our knowledge of soil-factors is in an ' Cowles, 1901 a; Whitford, 1905. - Schroter, 1904. See also P. E. Miiller, 1887, 1871. ^ Also may be added the character of dryness — physical in the first, physio- I logical in the second. See p. 134. 134 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi extremely elementary condition,' and 'We have no exact understand- ing whatsoever of the sum of physical factors which we term climatic' Communities of xerophytes must be set into different groups according as they are due to nature of soil or to nature of climate. The distinctions in this respect are recounted in the succeeding paragraphs. Soil may, to employ Schimper's ^ terms, be physically or physiologically Physical Drought. Soil is physically dry when it contains very little free water ; this is the case with — 1. The surface of rocks or stones occupied by plants which compose the lithophilous formations. 2. Sandy soil which lies so high above permanent subterranean water that this does not affect it, and which is very parched during dry seasons owing to rapid drainage and desiccation ; upon it are psammo- philous formations, which are allied to those on rubble, where soil is formed of gravel and stones. Here too must be placed epiphytes, which nearly all have definite adaptations for securing water (see p. 8y). Physiological Drought. Soil is physiologically dry when it contains a considerable amount of water which, nevertheless, is available to the plant only to a slight extent or can be absorbed only with difficulty, either because the soil holds firmly to a large quantity of water or because the osmotic force of the root is inadequate to overcome that of the concen- trated salt solution in the soil. This may be the case when — 1. Soil is rich in free humous acids, or in chemical bodies that by their peculiar action on the plant evoke xerophily ^ ; there result those forma- tions that grow on sour (acid) soil. 2. Soil rich in soluble salts, usually common salt, which brings into existence that form of xerophily whicli we see in halophilous formations. A halophyte is in fact a special form of xerophyte, as Clements repeatedly urges, and Wiesner ^ and Schimper * recognized. In addition to the xerophytic formations thus grouped according to char- acters of the soil, which is dry, or dries frequently or rapidly even in a moist climate, there is another series of formations to which the physical and chemical qualities of the soil are of subordinate import in comparison with the extreme climate. The soil is neither too acid, too saline, nor too poor in nutriment, and may be sufficiently moist to sustain luxuriant vegetation, yet the climate is so extreme that the soil is either too cold (as in the case of formations on subglacial tracts) or periodically so dry for a long season that only xerophilous formations can thrive on it, excepting in situations, such as marshes or river-banks, where the soil contains sufficient moisture throughout the year ; hence in this case, also, topographical features play a part. The vegetation of savannahs {campos) in the interior of Brazil is a formation evoked by a dry season ; yet it is everywhere confined to the higher ground in the hilly country, while forest always occurs along the watercourses and on the mountains, where greater humidity prevails in the soil ; there can be no doubt that were the chmate moist throughout the year the campos would be ^ Schimper, 1898. * See Livingston, 1904. ' Wiesner, 1889. * Schimper, 1891, 1898, I CHAP. XXXIV OECOLOGICAL CLASSES 135 clothed with forest.^ Among the formations belonging to this type must be reckoned those forming steppes and savannahs, also certain sclerophyllous formations. Mesophytes grow on soil which is of an intermediate character, and is neither specially acid, cold, nor saline, but is moderately moist, usually well-ventilated, also rich in nutriment and in alkahne humus or in other organic constituents. Mesophytic communities occur in very diverse climates, near the Poles or on the equator, yet they can never be exposed to the danger of prolonged drought. Adapted to such conditions are plants that show a relatively weak development of the above-mentioned arrangements for regulating transpiration ; in this respect these plants stand midway between hydrophytes and xerophytes. The leaves are large, and far more varied in form than in xerophytes ; teeth and other incisions of the margin are common, as are compound or richly divided leaves ; hydathodes seem to be frequent ; the vegetative organs are of 1 fresh green, and devoid of thick grey coatings of hair or bluish incrusta- tions of wax ; the leaves are usuafly dorsi-ventral in structure. Stomata ;ire numerous, often occurring also on the upper face of the leaf ; anatomi- cal peculiarities, such as aqueous tissues, are very rare, or at least not extremely developed. The greatest differences among mesophytes depend upon whether the leaves persist for only a few months in the favourable season, or for ;i year or more. Ilex Aquifolium is indubitably a mesophyte growing ;is underwood in forests of northern Europe ; but its leaves persist for up to two years, and, hke sclerophyllous leaves, are xerophytic in structure because they are exposed to the harsh conditions of winter — that is, to cold (physiologically dry) soil and possibly concurrent rapid transpiration caused by dry, cold winds ; the same is true of the spruce (Picea excelsa) ind most other evergreen woody plants in cold-temperate countries. In deciduous plants within the same countries the leaves are thinner, of L paler green, and more flexible ; the cuticle is thinner, and so on ; they are, in short, typically mesophytic in structure. In the tropical rain-forest, which may be regarded as a mesophytic community, there are many species possessing leaves that are xerophytic in structure because they persist for more than one year, and must consequently be adapted to endure all the changes during that period. It is also difficult to regard all conifers as xerophytes, even when their leaves are perennial. - Pound and Clements ' divide mesophytes into three groups : hylophytes (woody plants), po-ophytes (meadow plants), and aletophytes (ruderal plants). But it must be noted that there are forests and grasslands among xerophytic formations. Between the different groups there are very gradual transitional stages. Moreover, the peculiarities of a formation may be evoked by a com- bination of diverse factors of varied strengths ; for instance, there seem to be formations for the origin of which the co-operation of coldness and acidity of soil is responsible. Tropophyies. Schimper * has introduced the term ' tropophyte ', by which he designates land-plants which, in opposition to hygrophytes and xerophytes, have ' Warming, 1892, 1899. * See Section XV. • Pound and Clements, 1898; Clements, 1004, p. 22. * Schimper, 1898. 136 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi deciduous leaves, and ' whose conditions of life are, according to the season ot the year, alternately those of hygroph\-tes and of xeroph^-tes ' ; the structure of their perennial parts is xerophilous, and that of their parts that are present onlv in the wet season is hygrophilous. Tropoph\-tes as a whole are included in mesoph\'tes (a term also employed by Schimper), as are Schimper's hygrophytes. The cold-temperate flora is mainly tropophilous. The term tropophilous is a very ser\'iceable one as appUed to those plants that shed their assimilating organs during the unfavourable season, but there is no group of tropophytes contrasting with those of hydrophj-tes and xeroph^-tes ; for there are tropophilous hygroph^-tes and tropophilous xerophytes, as Schimper's own words in various passages indicate. In accordance wdth the considerations given in the present and previous chapters the following oecological classes ^ may be distinguished : — A. The soil (in the \\'idest sense) is very wet, and the abundant water is available to the plant (at least in Class i), the formations are therefore more or less hydrophilous : — Class I. Hydrophytes (of formations in water). Section IV. Class 2. Helophytes (of formations in marsh). Section V. B. The soil is physiologically dry, i. e. contains water which is available to the plant only to a slight extent ; the formations are therefore essen- tially composed of xerophilous species : — Class 3. Oxylophytes (of formations on sour (acid) soil). Section VI. Class 4. Psychrophytes (of formations on cold soil). Section IX. Class 5. Halophytes (of formations on saline soil). Section VII. C. The soil is physically dry, and its slight power of retaining water determines the vegetation, the chmate being of secondary import ; the formations are therefore likewise xerophilous : — Class 6. Lithophytes (of formations on rocks). Section VIII. Class 7. Psammophytes (of formations on sand and gravel). Sect. X. Class 8. Chersophytes (of formations on waste land). Sect. XII. D. The climate is very dry and decides the character of the vegetation ; the properties of the soil are dominated by chmate ; the formations are also xerophilous : — Class 9. Eremophytes (of formations on desert and steppe). Sect. XI. Class 10. Psilophytes (of formations on savannah). Section XIII. Class II. Sclerophyllous formations (bush and forest). Sect. XIV. E. The soil is physically or physically dry : Class 12. Coniferous formations (forest). Section XV. F. Soil and climate favour the development of mesophilous forma- tions : — Class 13. Mesophytes. Section XVI. ^ Concerning the classification of plant-communities, reference should be made to Clements, 1904 a ; and regarding nomenclature to Clements, 1902 a and b. 137 CHAPTER XXXV PHYSIOGNOMY OF VEGETATION. FORMATIONS ASSOCIATIONS. VARIETIES OF ASSOCIATIONS The large oecological classes indicated in the preceding chapter may be subdivided into vai"ious less comprehensive types of communities. Popular distinctions have for a long time been drawn among divers types of vegetation, to which have been allotted certain names (forest, bush, meadow, moor, heath, steppe, savannah, viaqui, and so forth) tluit have been adopted as scientific terms. The leading features upon which the pertinent distinctions depend are physiognomic, and thus dependent upon biological relationships. For the physiognomy of vegetation is not only of aesthetic, but also of scientific significance : vegetation often essentially determines the physiognomy of landscape, and in this respect plays a part very different from that played by animals.^ Physiognomy must therefore be scientifically considered. A. PHYSIOGNOMY OF VEGETATION The chief circumstances that determine the physiognomy of vegeta- tion are : — 1. Dominant growth- forms : trees, shrubs, and herbs, of varied appearance, size and shape of fohage ; mosses, lichens, and other types.- Thus arise the oecological types : forest, bush, heath, meadow, steppe, and other kinds of herbaceous vegetation, moss-tundra, lichen-tundra, and so forth, modifications being introduced by lianes and epiphytes. 2. Density of vegetation (number of individuals). This de- pends upon the struggles of plants with inanimate Nature, and upon the biological peculiarities of growth-forms. In some communities the soil is densely covered, as in the case of meadow, but in others the vegetable covering is so open that the colour of the soil imparts to the landscape its hue. A distinction must therefore be drawn between — {a) open formations evoked by shifting soil (seashore, dunes), extreme character of soil (rock), dryness of climate (in desert and steppe), or by extreme cold (in Polar countries), and — (b) closed formations composed of species that grow in company ^ for some reason or other, whether it be that, like some species of trees, they can suppress all competition by their shade, or, like Phragmites, can form dense associations by means of richly branched horizontal rhizomes. A number of different species can together form one closed formation.'* 3. Height of vegetation. Comparisons may be instituted between forest, bush, and heath, all composed of woody plants, or between the tall grass of a lowland meadow and the low sward of an alpine one, or between forest and tundra. Many formations exhibit strata or storeys of growth-forms : the greatest number of storeys will probably be found in well-lighted, therefore thinly wooded, forest. ' Darwin writes : ' A traveller should be a botanist, for in all views plants form the chief embellishment.' ' See Chapter NI. ' Plantes assocides, of Humboldt, 1807. * See Drude, 1905. iii. Tallest field-stratum iv. Middle field-stratum V. Lower field-stratum vi. Ground stratum It may, however, suffice to cons 138 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi Finnish botanists have adopted a kind of nomenclature to denote the different strata or storeys : i. Tree-stratum ii. Bush-stratum .... 90 centimetres to about 4-5 metres 45 ,, to 80-90 centimetres 10 „ to 45 5 „ to 10 immediately on the surface of the soil up to 5 centimetres (mosses, lichens, algae), der merely four strata : — i. Ground-stratum : immediately above the soil : mainly mosses, lichens, and algae, ii. Field-stratum : formed by grass and herbs, as well as dwarf shrubs of approximately the same stature. iii. Shrub-stratum : formed of taller shrubs, iv. Tree-stratum.^ 4. Colour of vegetation. We may compare brown heath with green meadow. Here, too, may be mentioned the colours of flowers, and the contrast between entomophily and anemophily. 5. Seasonal relationships. This involves the duration of the resting period, and other phases of vegetation (foliation, flowering, and defoliation). We may compare evergreen forests with those that shed their fohage for winter, or for the dry season ; steppe which is green for a few months, but yellow-brown and bare for much longer ; the vegetation of north-temperate Europe in winter, and in summer. 6. Duration of life of species : duration of epigeous parts, the role played by annual and renascent species, and by woody plants. ^ Only rarely do we find an assemblage of plants consisting solely of annual species, but they do occur, as in the case of Salicomia herbacea, and of certain weeds extending over limited areas. 7. The number of species present gives some indication of the struggles for space between species ; this struggle may be greatly inter- fered with, and is, in fact, interrupted by man. Sometimes — as in spruce-forest, beech-forest, and ling-heath — a single definite species dominates ; at other times there is an extraordinary admixture of species. Rich in species are communities in warm countries, such as tropical forests,^ the maquis of Cape Colony ; poor in species are the communities of northern Europe. It is evident that favourable conditions of fife call into existence a more complex flora ; but geological factors have often played some part. In the vicinity of Lagoa Santa in Brazil, on an area of about three geographical square miles, there grow about 3,000 species of Vascular plants (since more than 2,600 species have been determined, and there must be at least 400 species that have not been collected). Of these there are 1,600 species {^circa) in the forest and 800 species (circa) in the campos, of which 400 and 90 respectively are trees; such is the case, despite of the fact that the area of forest is much smaller than the area of the campos, and is essentially confined to the valleys where it bounds all the watercourses. The reason for the greater richness of this forest-flora is | certainly to be sought in the prevailing physical conditions (more abundant humidity j and food-material, especially humus). But possibly geological causes have played a part ; for probably the forest-flora is the older, and the flora of the campos \ gradually arose later, as South America raised itself more and more above the sea, and Brazil consequently acquired a more continental climate.* ' See A. Nilsson, 1902 a. ^ See p. 8. ' See Humboldt, 1807, * See Warming, 1892, 1899. .1 CHAP. XXXV PHYSIOGNOMY OF VEGETATION 139 As the number of associated species increases the number of growth- lorms as a rule does so hkewise ; in this respect the premier place is taken by the warm, moist tropical forest, which perhaps owes its boundless wealth to the circumstance that it has been able to develop for vast periods of time without interruption. The number of species depends,^ inter alia, upon the means of competi- tion possessed by the several species. Some species readily form dense nuisses of vegetation composed of many individuals ; others are univers- illy represented by isolated individuals. Many species can occur in several kinds of formations, because the demands they make are bounded by \Mde limits, and because the more habitats they can occupy the wider are these limits. The hardiest and most accommodating species can seize upon most kinds of habitat, nevertheless they are often found only in a few, because they have been crowded out of the better ones. The more pecuhar and extreme a habitat is the more uniform does its xegetation tend to be, because, as a rule, only few species are so specialized in their adaptation as to be capable of existing in such a place. In studying the vegetation of a certain area from a floristic and geographical standpoint, it is necessary to define the relative numbers of the various species. I ; very community consists of dominant and sub-dominant s-pecies, as well as of others ; iiat are more or less dependent upon these and occur only here and there. Drude ■inploys the following terms " : — social : dominant species whose individuals give the main character to the \ egetation ; gregarious : species whose individuals occur in small groups so as to form >inall unmixed collections in the main vegetation ; copious (with abbreviations cop.^ cop.", cop.\ to denote decreasing frequency), species represented by individuals scattered in smaller numbers among those jneviously mentioned ; sparse : species having only isolated individuals occurring here and there ; solitary : species of which individuals occur in extreme isolation. These terms may be used in combination ; for instance, solitary gregario7ts would mean a single clump composed of one species. The relative shares normally taken by the various plants constituting a community ought to be capable of numerical expression.' B. FORMATIONS It is not sufficient merely to distinguish among the broader, physiogno- mical types, for between these there occur differences that demand the establishment of subdivisions. In attempting to define these many difficulties are encountered.* The term 'formation', or 'vegetative formation', was introduced in 1838 by Grisebach in the form ' phytogeographical formation ', which subse- quently gave place to ' vegetative formation '. Grisebach wrotc,^ ' I give the name phytogeographical formation to a group of plants, such as a meadow or a forest, that has a fixed physiognomic character. It is characterized sometimes by a single social species, sometimes by a com- plex of dominant species belonging to one family, or it exhibits an aggre- gation of species which, though diversified in organization, yet have some feature in common, as is the case with alpine meadow-wastes consisting nearly exclusively of perennial herbs.' Other writers have, however, attached a narrower signification to the ' See p. 93 and Chap. XCVIII. ' Drude, 1888, 1889. 1895. ' Compare Clements, 1905. * See Chap. III. '' Gri.sebach, 1838; 1880, p. 2. 140 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi term 'formation'. Hult, in his excellent papers on the phytogeography of Finland ^ estabhshes about fifty ' formations ' as existent in North Finland — an Empetrum-formation, Phyllodoce-formation, Azalea-forma- tion, Betula nana-formation, Juncus trifidus-formation, Carex rupestris- formation, Nardus-formation, Scirpus caespitosus-formation, and others. Similarly Kjellman ^ divides the algal vegetation into numerous ' forma- tions ' named after predominant species ; and the same use of the term is made by Stebler and Schroter in connexion with the types of Swiss meadows that they recognized. In like fashion we must distinguish as different formations beech- forest, oak-forest, birch-forest, and other dicotylous forest ; ling-heath, Empetrum-heath, Erica-heath ; or in fresh water, Scirpus lacustris- formation, Phragmites-formation, Equisetum limosum-formation, and so forth. This means a subdivision of the vegetation based upon the local domination of certain species, and may consequently result in matters of wide and general import escaping notice, and communities naturally belonging together, as evidenced by their pursuance of identical economy, may not be recognized as such. The Empetrum-formation, Azalea- formation, and Phyllodoce-formation are oecologically essentially alike, and may be regarded as members of a more comprehensive natural community, the dwarf-shrub heath ; the Scirpus-formation and Phrag- mites-formation are likewise members of one community, and it often evidently depends upon mere accident whether the one or the other of these ' formations ' prevails at a given spot. Obviously, such special communities play a part, and must be dis- tinguished in any detailed description of the vegetation of a definite area ; but it is better to speak of them as associations, and to follow Grisebach in the use of the term 'formation ', since he was the first to introduce and define it, although he laid less stress on its oecological meaning.^ A formation may then be defined as a community of species, all belonging to definite growth-forms, which have become associated together by definite external (edaphic or climatic) characters of the habitat to which they are adapted. Consequently, so long as the external condi- tions remain the same, or nearly so, a formation appears with a certain determined uniformity and physiognomy, even in different parts of the world, and even when the constituent species are very different and possibly belong to different genera or families. Therefore — A formation is an expression of certain defined conditions of life, and is not concerned with floristic differences. The majority of growth-forms can by themselves compose formations or can occur as dominant members in a formation. Hence, in sub- dividing the groups of hydrophilous, xerophilous, and mesophilous plants, it will be natural to employ the chief tyPes of growth-forms as the prime basis of classification, or, in other words, to depend on the distinctions ' Hult, 1 88 1, 1887. ' Kjellman, 1878. ' The term ' formation ' is so employed also by Ascherson, 1883, p. 728 ; Kerner, 1891, p. 830; C. Schroter und Kirchner, 1902; Kearney, 1900; Ganong, 1902; Clements, 1902 a. Adamovicz, 1898, and Cowles, 1901, adopt at least approximately the same course. On the other hand, Flahault and W. G. Smith employ the word ' association ' in this sense. ir. I CHAP. XXXV FORMATIONS 141 between trees, shrubs, dwari-shrubs, undershrubs, herbs, mosses, and the hke.i Upon this basis the following types of formations must be dis- tinguished : — 1. Microphyte-formation or thallophyte-formations, in which the community is composed exclusively, or mainly, of lichens and algae. Here, except in the case of sea-weeds, there can scarcely be any question of more than one stratum (storey) of plants. 2. Moss-formation. Here algae may form a lower stratum (storey). 3. Herb-formations, such as meadow, prairie, grass-steppe, and others. Here may occur two or several strata (storeys) ; namely, a humbler vege- tation of thallophytes or mosses, and a taller vegetation of herbs ; the herbs may in turn be ranged into storeys of different heights. 4. Dwarf -shrub formations and undershrub- formations also include admixture of herbs, which sometimes overtop the dwarf-shrubs, and undershrubs. These longer-lived constituents preponderate, however, and among them there may occur several storeys of vegetation belonging to the types described under i, 2, 3. 5. Bush-wood or Shrub-wood is composed of taller, hgnified, many-stemmed plants. Compared with the previous ones the community gradually has become richer in growth-forms ; there may occur epiphytes and hanes, and below the highest story, all the forms of vegetation of I, 2, 3, 4 ; for instance, mesophilous herbs growing in the shade. Yet the oecological conditions prevaihng in bush are not the most favourable to plant-life, and the ground-vegetation is often very scanty, because bush-wood may be so dense as to permit the passage of even less light than does forest. 6. Forest is the tallest type, and exhibits the greatest multiplicity of growth-forms as well as the largest number of storeys : High forest, composed of trees — hght-demanding and shade trees ; - in tropical forest more than one storey. Underwood, composed of shrubs, dwarf-shrubs, and undershrubs. Forest-floor vegetation, composed of herbs, mosses, thallophytes, and many saprophytes, and depending upon the light, which is more or less enfeebled by the tree-crowns, upon moisture in the soil and air, upon the humus, and other factors. Beneath such species as beech, spruce, and silver fir, which grow close together and cast a deep shade, there is a very meagre vegetation, but below Hght-demanding trees, in harmony with their need for light, there is a richer vegetation. The flora at the edge of a forest may differ materially from that of the interior, because the conditions of illumination in the former position permit the develop- ment of many species that are excluded from the latter. Gri'\-illius ■' found that the tall herbs in thinly wooded and therefore well-lighted Scandinavian forests can be ranged into different types, which differ from one another in the arrangement of the inflorescences, the form and position of the assimilatory organs, innovation, time of flowering, and distribution in the different storeys of the general plant-community — all these relationships deserve closer attention and investigation. ' Sec Kerner, 1891. * Sec p. 18, ' Grcvillius. 1894. 142 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi It may be asked — why arrange the various types of vegetation in the classes named on p. 136 ? Why not use each growth-form as a foundation upon which to build a special class ? The following classes could then be distinguished : that of forest-formations, of bush-formations, of shrub-formations, of dwarf-shrub formations, of perennial-herb formations, of moss-formations, and of alga-forma- tions.' Within each such class one would further be able to distinguish hygrophilous, mesophilous, and xerophilous formations, and to define them. From a morpho- logical standpoint this would possess a certain interest, but from a phytogeo- graphical one it must be dismissed, because it would involve the separation of formations that are oecologically closely allied. It is nature of locality that must be represented by formations ; and naturally allied localities may include different collections of growth-forms, yet they must be grouped in the same class. According to the isolation or combination of the growth-forms in a formation we have to distinguish simple formations and compound formations. Simple formations : — Formations consisting solely of one type of growth-form are few. An example is the phyto-plankton formation. The species composing this microphytic free-floating flora belong to different families or even different systematic classes, but they may all be grouped together as belonging to one growth-form adapted to the free-floating mode of existence in water ; plankton is a purely edaphic oecological formation. Compound formations : — Usually many growth-forms, and often, to some extent, different formations, are combined to form a single whole. For instance, as will be explained hereafter, the reed-formation is one dominated by divers monocotylous herbs, which are social and perennial in habit and varied in stature ; but on the ground, also in the water between and beneath the reeds, there sometimes flourish other growth-forms comprising what may be termed subordinate commu- nities : thus there may be communities composed of Schizophyceae, plankton, pleuston, and limnaea,^ which are more or less influenced in their composition by the dominant community. Again, in forest the different lower storeys are constituted of growth-forms which, for the most part, are able in themselves to give rise to distinct formations (bush, grassland, moss-formation, and others), but in the two cases the species occurring would usually be different. For the shade of forest or of tall vegetation affects not only the conditions of illumination, but also the humidity and temperature of air and soil. As an example of a species capable by itself of giving rise to one independent formation, also of occurring as a subordinate member in another, we may cite Calluna vulgaris. This species as a dominant plant forms a widely distributed community belonging to the type of dwarf-shrub heath ; but it can occur as low vegetation in thin pine-forest ; to what extent . the latter differs from the open ling-heath has not been adequately | \ investigated. A formation consisting of several storeys may therefore be composed of both xerophytes and mesophytes : there are forests — for instance, sclerophyllous forest — in which markedly xerophytic species form the uppermost storey, but mesophytic species the lower storeys. ' This approximates to the method adopted by Kerner, 1891, p. 821, who distinguished nine kinds of societies. ' See Chap. XLIX. CHAP. XXXV FORMATIONS 143 Mixed vegetation : Quite different from the complex, several-storeyed formations just mentioned is mixed vegetation — which consists of small patches of different formations, occurring close together, but each retaining its own individuahty as a pure formation. This is specially the case where the terrain varies suddenly and greatly, and thus causes the oecological conditions to do likewise. In many mountainous districts, rocks, small bushlands, grasslands, perhaps pools of water, and the like, suc- ceed one another within a limited area, without one formation perceptibly affecting another, and without any principle, other than chance, revealing itself in the admixture. The more level and uniform does soil remain over a wide area, the larger and more uniform are the formations ; the more uneven and variable is it, the more mixed is the vegetation. It is impossible to draw a sharp distinction between a vegetation consisting of several pure formations mixed together, and one consisting of a single complex formation ; for the smaller the patch of a mixed vegetation is the more are the species influenced in their biotic conditions by other adjoining ones. One condition especially responsible for change of this kind is human intervention. By the mixing of formations, especially if these be extensive, there come into existence various physiognomical types of landscape which differ according to the constituents. Fixity of formations. ' Fixed physiognomic character ' is a part of Grisebach's definition of a plant-formation. Recent writers, like Beck \ and Drude -, emphasize the fact that ' fixity ' is a character essential to the concept of a formation. We may paraphrase Drude as follows : — A vegetative formation is an independent community of first rank, which consists of like growth-forms or of such as are necessarily asso- ciated, and is confined by its natural boundary to a site determined by the prevalence of identical conditions of existence ; it is thus assumed that without external interference no real change in the nature of any community occupying the site can set in — the community is ' fixed '. This fixity can be regarded as being only relative, as Drude distinctly agrees ; for, as the external conditions change, every formation will be able to undergo modification, and will always do so in the course of time ^ : there are formations that may have remained, and perhaps will remain, unchanged for thousands of years — tropical rain-forest, for example. There are others that will soon be exterminated in the places they now occupy, and driven to other sites. Secondary formations. By the term ' secondary formations ' we indicate formations which have arisen through human interference.'' There are various formations of this kind that are changed only in their flora : such may be termed semi-cultivated '" formations, and are exempli- fied by North European heaths, which have been modified by the browsing of animals, and by other agencies. But there are entirely new formations resulting from man's activity in destroying forest, in farming, and in otherwise utilizing the soil ; of such a nature are the ' sibljak ' in Servia '^, ' Beck von Mannagctta, 1902, ' Drude, 1896, p. 286. ^ Compare Section XVII. * Warming, 1892. They are the ' Substitute Associations ' of \V. G. Smith, 1905, p. 62. * Krause, 1892 a; see also Grabner, 1909. ' According to Adamovicz, 1902. 144 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi oak-scrub in Jutland, and the Tristegia glutinosa-grasslands in Brazil ^ : these formations arise and can only be preserved by cultivation, and must be described as secondary formations. Sub-formations. There are various formations which are so immense in extent and range, and display certain minor oecological distinctions, that it is advisable and convenient to subdivide them into sub-formations ; for example, plankton-formation, coniferous forest, and dicotylous forest, can thus be subdivided. In deahng with forest one character for consideration is its deciduous or evergreen nature, another concerns the question as to whether the ground-vegetation consists of the same or different growth-forms ; the differences in the ground-vegetation in forests may be so great that they give an entirely different appearance to these, which must therefore be accounted as sub-formations. The existence of sub-formations must, however, be justified on oecological grounds (by the depth, water-content, or kind of soil, or by other factors). Increase in our knowledge of oecology will shed light upon these questions. - C. ASSOCIATIONS* The same formation, in different districts and locahties, or even at different seasons of the year, may be composed of different species, and is perhaps mostly so. Plankton-formation in certain months may be composed of species different from those in other months ; reed- formation in various places has Phragmites communis as the dominant species, in others Scirpus lacustris or Typha, yet it remains the same formation. Beech-forest and oak-forest are specifically different forms of the same formation — the summer-green (deciduous) dicotylous forest in the temperate climate. Coniferous forest in one place may consist of Pinus sylvestris, at another of P. montana or P. halepensis, or in North America of entirely different species of Pinus ; or again, it may be formed by species of Picea or Abies ; furthermore, it may be an ad- mixture of coniferous species. A cornfield is a ' cultivated formation ' of annual or biennial species ; but the maize-field, rye-field, wheat- field, and buckwheat-field are different forms of it.* Such smaller, often more-localized subdivisions or kinds of the forma- tion may be distinguished most correctly by the general term association, which has been employed in this sense by various botanists.^ ^ Warming, 1892. ^ It is a matter of great difficulty to find suitable names for the various kinds of formations. Some of the many words in common use have been utiUzed as scientific terms, such as steppe, prairie, tundra, Caa Tinga, alang-alang, savannah, and others (see Warburg, 1900), but many popular words are unsuitable. Other scientific terms, such as plankton (Hensen), and garide (Chodat) are of recent introduction. ' Humboldt's ' plantes associees '. See p. 137. * In regard to Woodhead's ' complementary ' and ' competitive ' associations, see p. 95. * Including C. Schroter und Kirchner, Ganong, Kearnley, Cajander, Adamovicz, and partly W, G. Smith. Hult's, 1885, 'formations' (see p. 140) are associations, and so also apparentlv are Drude's ' unit formations '. Discussion of the different significations attached to the term ' association ' by authors is impossible here. I CHAP. XXXV ASSOCIATIONS 145 An association is a community of definite floristic composition within a formation ; it is, so to speak, a floristic species of a formation which is an oecological genus. Wliile the formations in different floras may be the same, associations are dependent upon the character of the flora of the country concerned. Terminology. Associations may conveniently be denoted according to the plan suggested by Schouw,' by the addition of the suffix -etum to the name of the characteristic species or genus ; for instance, in the reed-formation there are phragmiteta, scirpeta, typheta, and the like ; similarly there are saliceta, pineta, and so on ; in order to indicate the species concerned the specific name has to be added, thus, for example, scirpetum Scirpi lacustris, saUcetum Salicis albae, pinetum Pini sylvestris, pinetum Pini montanae ; or more briefly, scirpetum lacustris, salicetum albae, and the like.' This terminology corresponds somewhat to popular usage, in which two words are combined and one of them indicates the species, as is the case with the terms beech-forest, oak-forest, birch-forest, or Kerner's terms, gold-beard ' meadow, feather-grass * meadow. When an association is composed, not of a single prominent species, but of several in equal shares, it may be denoted by a compound term, for instance, scirpo- phragmitetum, fago-quercetum, or in some other way — for instance, by its popular name.* In many cases associations are brought into existence within the formation by minor distinctions in the soil, because different species react in a shghtly different manner. In other cases accident seems to decide the question : the species which first colonized the spot will subsequently be able to maintain itself against others, as in the instance of the reed-associations clothing the banks of rivers and lakes in north- temperate Europe. Associations may occur irregularly as patches in the formation ; or may exhibit a zonal arrangement. The latter is always the case when the formation grows in water (in which case depth decides the matter), or on the banks of rivers, lakes and pools, where subterranean water can play its part ; with increasing height of soil above the water-table vegetation y I displays a zonal change.^ Plants react to infinitely small differences in the water-content of soil. In very many cases these belts of plants at the same time represent developmental series, inasmuch as each association is successively expelled by the next one. Associations may also be dependent upon shade and other oecological factors. An alpine meadow has a different floristic composition accord- ing as it clothes the northern or southern slope of a mountain, yet it remains a typical meadow, and thus unchanged as a formation ; grassy surfaces fining a railway differ floristically according to the aspect,' )ut they are merely different species, or, in other words, different associa- tions, of the same formation. D VARIETIES OF ASSOCIATION As in oecology a formation may be regarded as a genus, and an issociation as a species, so we may also recognize oecological varieties ' Schouw, 1822, who referred to ericeta, rhododendreta, arundineta, pineta, 'ageta, and the like. ' See Cajander, 1903. ' [Chrysopogon Gryllus]. * [Stipa]. ' See C. Schroter und Kirchner, 1902. •See Raunkiar, 1889; Warming, 1890, 1906; MacMillan, 1896; Magnin. [893, 1894; Pieters, 1894; Clements, 1905; Shantz, 1905. ^ Stebler und Volkart, 1904 ; Stcnstrom, 1905. WARMING 146 ADAPTATIONS. OECOLOGICAL CLASSIFICATION sect, hi dependent upon minor differences in an association.^ A beech-forest, fagetum Fagi sylvaticae, is not everywhere and on every terrain absolutely uniform and constant in composition, but owes its modification, inter alia, to differences in soil ; in one beech-forest the ground- vegetation may be dominated by Asperula odorata, which denotes alkaline, loose, well-aerated humus, while in others Aira flexuosa or Vaccinium Myrtillus may play a dominant part and thus indicate acid humus. There are also beech-forests in which still other species characterize the flora on the ground. Accordingly, beech-forests may be described as fageta asperulosa, fageta myrtillosa, and so on. Similarly there are betuleta hylocomiosa, betuleta cladinosa, and others, according as the ground is clothed with mosses of the genus Hylocomium, or lichens of the genus Cladina. The ground-flora in forests of Pinus sylvestris differs widely according to various conditions, including the character of the soil ; in Baltic forest there occur together many species that in the Hercynian hills and lower highlands are distributed in different formations .^ Grabner ^ distinguishes in the heath-formation various associations : namely, ling-heath (callunetum), with several varieties, according as Pulsatilla, Genistae, or SoHdago, or others predominate as perennial herbs ; a tetralix-heath (ericetum), which also has several varieties ; an Empetrum-heath, and others. Certain species are so accommodating that they can grow on very different soils, and can therefore give rise to different associations. Woodhead * has pointed out that this is the case with Pteris aquilina ; and he therefore distinguishes between two varieties of associations, which he terms meso-pteridetum (an association of Pteris with Holcus lanatus and Scilla festalis), and xero-pteriddum (an association of Pteris with Calluna, Vaccinium Myrtillus, Aira flexuosa, and others). Edaphic varieties. Varieties in an association may be distributed in patches, or zones, according to the conditions prevailing in the soil ; in a meadow we can see both these forms of arrangement, corresponding to the shapes and extent of the depressions in the ground. Here we are dealing with edaphic varieties. Geographical varieties. But varieties may have arisen through' histo- ric or climatic causes; for example, according to Beck von Mannagetta^ Pinus Laricio is distributed as a high-forest tree over so wide an area that its ground-vegetation belongs to three different floral districts : the Pontic, Baltic, and Mediterranean. Here, then, we have three geographical varieties. Hock's investigations show that the ' companion- plants ' of the respective forest-trees in different parts of Europe may be widely different, even if the limiting boundaries of the trees and of their companion-plants coincide to some extent. In the succeeding Sections an attempt will be made to give a survey * Cowles, 1899, p. III. These varieties are sometimes spoken of under the name ' facies '. The term ' facies ' seems to have been employed by Lorenz, 1863, first to denote small local differences in a formation, but subsequently to denote an association, and in a somewhat similar or slightly different sense it is employed by others. [The term ' facies ' has already the recognized meaning in English biological science of ' general aspect or appearance ' (N.E.D.), and its use in a restricted sense as that of a variety of an association is barred. P.G.] ^ Drude, 1902. ^ Grabner, 1901. * Woodhead, 1906, pp. 114, 363. ' Beck von Mannagetta, 1902. CHAP. XXXV VARIETIES OF ASSOCIATION 147 of the most important formations existing on the Earth. But to associa- tions and their varieties less attention can be devoted, partly because the subject has not been sufficiently investigated, and partly because it would involve too much detail ; for such detail reference should be made to special papers. It must be said that there still prevails great confusion in the use of the terms ' formation ', ' subformation ', and the hke, and this places difficulties in the way of a clear comprehension of the subject. In the following scheme an endeavour is made to arrange the for- mations within the respective classes according to definite principles in a definite order of succession. The scheme commences, so far as is practi- cally possible, with the simplest and shortest types, which form a low stratum of vegetation corresponding to the growth-forms described in Chapter II and on page 141 ; they can not only form separate communities, but also occur as subordinate constituents in formations composed of taller forms. For example, the formations on acid humus soil are, as a whole, arranged as follows : First, the low formations composed of thallophytes and mosses ; secondly, those composed of two storeys, dwarf-shrubs, and those forms belonging to the preceding type ; and, thirdly, still taller formations composed of several storeys, shrubs and trees, among which representa- tives of the first two types of formations are usually found and may, indeed, have developed as the real foundation of the whole formation. An attempt has been made to show between the different formations a logical and biological connexion, which in many cases is also develop- mental in significance. Formations commence often as open, not fixed, communities that are composed of short and lowly organized species ; with the passage of time their development is continued by the immigra- tion of species that are taller and more successful in the struggle for existence, until a final stage is reached.^ Cowles 2 strongly insists ' that the plant societies must be grouped according to origins and relationships, and the idea of constant change must be strongly emphasized '. A. Nilsson ^ followed the same train of thought in his arrangement of Swedish formations into four series — the heath-series, meadow-series, marsh-series, and moor-series — as he particularly took cognizance of the amount of water and nutriment in the soil. Many of the formations brought into his series are closely allied and can readily develop into one another, but there are others that are certainly not aUied, for instance, littoral meadow, alpine meadow, wood- meadow,"* meadow spruce-forest, and meadow oak-forest.^ It is true that changes in the physical relationships of soil are every- where and always taking place, and that in close correlation with this plant-communities also undergo modification.^ This will be discussed in the final Section of this book. In the future it will be an interesting and important problem to trace out in each country or district the development of the vegetation, as regards not only flora but also plant- ' Warming, 1890, 1906; Moss, 1906, 1907; A. Ernst, 1907. ' Cowles, 1 90 1 6. * A. Nilsson, 1902. * [Very open, thinly wooded forest, with an abundance of grass and other herbs representing meadow.] ' In regard to the genesis of plant-communities consult Engler, 1899, p. 179. ' Warming, 1899. L 2 148 ADAPTATIONS. OECOLOGICAL CLASSIFICATION communities as a whole.^ But it does not seem possible to use develop- ment as the fundamental basis of classification of plant-communities : for developmental changes are too dependent upon local conditions ; a formation does not develop merely in a single definite direction, but wiU modify in one direction at one place and in another at another place, according to the prevaiUng conditions .^ ' As regards Denmark see Warming, 1904, 1906, 1907-9. ' Concerning the oecological nomenclature and classification see : G. Beck von Mannagetta, 1902 ; Brockmann-Jerosch, 1907 ; Cajander, 1903 ; F. Clements, 1902, 1904, 1905 ; Cockayne, loc. cit., 1905 ; Cowles, 1899, 1901 ; Drude, 1896, 1905 ; Flahault, 1900, 1901 ; Ganong, 1902; Grabner, 1905, 1898 a, 1909; Harshberger, 1900, &c. ; Kearney, 1900; Kerner, 1891 ; A. Nilsson, 1902; Kirchner und Schroter, 1 896-1902; Shantz, 1905; R. and W. G. Smith, 1898, 1899, &c, ; Stebler and Volkart, 1 904 ; Woodhead, 1 906. SECTION IV CLASS I. HYDROPHYTES. FORMATIONS OF AQUATIC PLANTS CHAPTER XXXVI. OECOLOGICAL FACTORS Before considering the various communities of aquatic plants it is necessary to discuss the general characters of water and its oecological factors, in so far as these affect the distribution and existence of plants confined to water.^ Air in water. Air occurs dissolved in water in variable amounts. In the atmosphere - and in water the gases present are the same, but their relative proportions are different ; the gas absorbed by ordinary water contains more oxygen and much more carbon dioxide in proportion to nitrogen than does the atmosphere. Just as in the case of land-plants, these two gases are the only important ones, the former in respiration, and the latter in assimilation. Only certain bacteria can dispense with oxygen. Air can reach parts that are submerged in water with much greater difficulty than it can reach parts situated in the atmosphere or in ordinary soil ; indeed, stagnant water may become so poor in oxygen as to almost exclude the existence of higher plants and animals. Appa- rently in consequence of this, certain species occur particularly in places where the water is very troubled or has a rapid flow, and where there is a constantly fresh supply of water ; possibly for the same reason many submerged parts (leaves) or whole plants (algae) are divided into capillary segments — compare the construction of gills — by which means the surface in contact with water is greater than if the organ presented a single surface ; and perhaps for the same reason many algae and Podostemaceae bear long hairs that serve as probably respiratory organs or enlarge the assimilatory surface. The obstructed supply of air is also a reason, and perhaps the most important one, for the large air-containing spaces^ which occur in many aquatic plants (sometimes occupying more than seventy per cent, of the whole volume of a plant) ; these spaces serve, inter alia, to convey air, and especially oxygen, from parts in the air to those in the water or mud. Certain swamp-plants, and especially some living in mangrove swamps, possess special respiratory organs which will be described here- after.'* When access of air is prevented and water becomes poor in oxygen there are formed in soil humous acids, which are characteristic of moor and peat soils, and which render soil physiologically dry.^ As the temperature rises the power of water to dissolve gases decreases, and this is perhaps the main reason for the disappearance of certain ' See upon this subject Oltmanns, 1905, * See p. 14. " See p. 98, * See Chapter LX. ' See Chapter XVI and Section VI. I50 HYDROPHYTES sect, iv aquatic plants in summer, at which season the temperature is higher and the Hght more intense. Light in water. We must assume tliat every aquatic plant has its own minimum, optimum, and maximum intensity of hght. Illumination is of profound importance in relation to the distribution of algae, also to the abundance of their species at different seasons — ^but on this latter point very little is known. The farther apart lie the maximum and minimum the more extensive will be the area of distribution of the species concerned. Light plays in assimilation the same part as in the case of land-plants ; yet there are some peculiar relationships to be considered. Light is weakened, partly by reflection from water, partly by absorption in water, and partly by floating particles ; the weakening is therefore more con- siderable the dirtier the water. Submerged plants, for this reason and also because there is no transpiration, acquire as a whole the pattern of a shade-leaf ; they become lanky, like etiolated plants, thin, and their assimilatory tissue shows little differentiation. Light penetrates downwards only to a certain distance, and con- stantly weakens with increasing depth, so that the assimilatory energy varies greatly with the depth ^ ; consequently, except in the case of bacteria, plants cannot be active at great depths. A ' regional ' distribu- tion of the vegetation results from the different powers possessed by plants of living in different intensities of light. Distinctions have been drawn between : 1. Euphotic vegetation, which receives an abundance of light. 2. Dysphotic vegetation, which lives in weakened light. 3. Aphotic vegetation, which lives in very weak light or darkness. Spermophyta descend at most to thirty metres (Zostera, for instance, to twelve or fourteen metres in Denmark) ; algae to forty metres, but living algae have been found 120-150 metres below the surface (in clear alpine lakes in Switzerland Characeae descend to 25-30 metres, but in Baltic lakes only to 6-8 metres) ; in Lake Geneva, according to Forel, a moss, Thamnium alopecurum var. Lemani, has been found at a depth of 60 metres ; the extreme depth to which light apparently penetrates is 400-500 metres. The presence of the protococcaceous Halosphaera viridis at 2,200 metres below the surface of the sea is certainly to be explained as a result of sea-currents or as a periodic sinking. As the rays of different colours are unequally absorbed plants descend to different depths. Red rays are absorbed in the upper layers of water ; the green, blue, and ultra-violet not before the lower layers. Ultra- violet rays can be detected by means of photographic plates at a depth of 400 metres. Correlated with these facts is the ' regional ' distribution of Algae, according to depth? Green Algae assimilate best in red light, Brown Algae in yellow light, but Red Algae are most active in green and blue light ; consequently, the first named occur only in the upper layers of water, while the last are especially in the deeper layers. Against this theory maintained by Engelmann the objection is urged by Oltmanns that with algae it is only a question of the intensity of light, and that the colour of sea-water merely acts as a screen. Recently, Gaidukow^ has shown that when Oscillatorieae are cultivated in coloured light they ' Proved by B. Jonsson, 1903. ^ Borgesen, 1905. ' Gaidukow, 1903. •:i CHAP. XXXVI OECOLOGICAL FACTORS 151 change in colour and assume that which is complementary to the light acting on them, and by this means assimilate more vigorously. Temperature of water. Submerged aquatic plants are exposed to far smaller extremes and fluctuations of temperature, both diurnal and annual, than are land-plants, because water has a high specific heat and is a bad conductor of heat : annual changes of temperature descend only to relatively small depths except in shallow water. Many aquatic plants hibernate in their green state, because no considerable degree of cold reaches them, and most of them are perennials. Their optimum for growth is generally low ; certain species, including Hydrurus (an alga belonging to the Phaeoflagellata), thrive only in very cold water. The disappearance of many algae in summer may be due to their optimum temperature having been exceeded. Arctic fresh-water lakes are usually very poor in organisms — a circumstance that may possibly be ascribed to the temperature of the water. Algae are frequently very sensitive to sudden changes in temperature,^ in salinity or in other conditions. Each species has its own pecuharities. High temperatures are encountered only in hot springs, where the plants growing are almost exclusively Oscillarieae and other Cyano- phyceae, which may be representatives of the vegetation that was the first to appear on Earth. The temperature of water decreases as the distance below the surface increases, but at different rates in fresh and salt water. In standing fresh water, at the bottom of lakes which are so deep that the annual variations of temperature in the upper strata cannot affect the strata at the bottom, the temperature is about 4°C., because fresh water attains its maximum density at this point. Strata of water lying above this may thus be much colder in winter. The temperature at the bed through- out the year in Swiss lakes is about 5° C, but is always below 4° C. in Baltic lakes, which are frozen at the surface during winter. In the sea, on the contrary, strata of water are colder the deeper they lie ; moreover, warm, or cold and salt, currents may be intercalated between them. The influence of temperature upon the distribution of aquatic Spermophyta is proved by an observation made by Magnin,^ who discovered that they descend to a depth of 11 metres in the warmer Jura lakes, but only down to 6 metres in deep and cold lakes. Temperature affects the amount of gas dissolved in water ; the colder this is the richer is it in oxygen and carbonic acid, and the more favourable may be the conditions for nutrition and consequently for growth. This is possibly the cause of the luxuriant development of algal vegetation in arctic seas. Nutritive and other substances dissolved in water. Water con- tains in solution many substances which vary according to the kinds of rock or soils with which it has entered into contact. Calcic carbonate is commonly present, being dissolved by the contained carbonic acid ; and, as many aquatic plants (Characeae, species of Potamogeton, and mosses) seize upon the carbon dioxide contained in the calcium hydrogen carbonate, incrustations of chalk are excreted on their surfaces, and may lead to deposits of hme on lake-beds.'^ ' Oltmanns, 1892. ' Magnin, 1894. ' C. Wescnberg-Lund, 1901. 152 HYDROPHYTES sect, iv Brandt ^ asserts that the sea is rich in nitrogen, which is constantly being suppUed to it, and is reduced by denitrifying bacteria. This process is more active in tropical than in temperate seas, and partly for this reason the ocean in sub-tropical and tropical regions is relatively poor in organisms, while it is rich in them in cool and cold regions. Many waters contain organic compounds which, by consuming the oxygen, render water unsuitable for the existence of autophytes. The nutritive substances most important to plants, such as potassium, phosphoric acid, ammonia, and sulphur, occur in all water, but only in small quantities and in an extreme condition of dilution ; in fresh water compounds of potassium and nitrogen are present in far greater propor- tions than are the other compounds in question. But we have no know- ledge that these conditions have any distinct effect upon the distribution of aquatic plants. Certain desmids and diatoms are stated to prefer lime, others silica ; similar minor differences are attributed to other plants. Common salt, on the contrary, is of profound importance. In sea-water, among the numerous salts, including chlorides of sodium, magnesium, and potassium, also sulphates of magnesium and calcium, the first named is of far the greatest significance. The amount of common salt in the ocean fluctuates within only narrow limits, but that in smaller seas varies greatly, not only in different sites, but also in the same place at different times. The following are the approximate percentages : — Red Sea 4, Mediterranean Sea 3-5-3-9, Pacific Ocean 3-5, Skager-Rak 3, Kattegat 1-5-3, the Great Belt 1-27, the Oresund 0-92 (in the last two, very variable according to the currents), Gulf of Bothnia o-i-o-5, Gulf of Fin- land 0-3-0-7. These statistics relate to superficial water ; but in parts of the sea round Denmark at a greater depth there is a saline undercurrent from the North Sea. The great difference between the floras of salt and fresh water will be discussed later in this Section. Although not a few fresh-water algae, especially lowly organized forms, can adapt themselves to the presence of common salt, which causes an enlargement of their cells and other formal changes, yet there are no plants other than certain diatoms that are common to fresh and slightly saline water ; nevertheless, in the brackish water of the Baltic Sea there grow some Characeae, Enteromorpha intestinalis, and Potamogeton pectinatus, which also occur in fresh water. The cyanophyceous communities occurring in special places will be dealt with later. Specific gravity of water. Salt water and fresh water differ greatly in specific gravity, and consequently in buoyancy, which plays a great part in connexion with plankton-organisms ; for fresh water, as is well known, has a smaller power of keeping bodies from sinking. The regular seasonal changes in temperature of fresh water bring in their train corresponding ones in the specific gravity and viscosity of the water. Many plankton- organisms undergo periodic changes of shape, which all seem to be in the direction of increasing the resistance-surface, and which synchronize with the changes of temperature. It therefore seems highly probable that these seasonal variations in shape are to be regarded as responses on * Brandt, 1904. CHAP. XXXVI OECOLOGICAL FACTORS 153 the part of the organism to periodic changes in the buoyancy of fresh water.^ Colour of water. Water in a pure condition is blue. Any other colour may be caused by organisms,^ by suspended particles of clay and the like, or, especially in fresh water, by humous acids ; yellow or brown water often contains many humous acids and is acid in reaction, whereas alkahne (hard) water is clear (blue). Movements of water. Movements of water are of great importance to vegetation. They assume the form of waves (breakers) or currents, and lead to a fresh supply of oxygeti ; in streaming water assimilation is more active.^ Still water is very inimical to vegetation ; and for this reason many species are absent from stationary depths over large areas, or from enclosed calm inlets. In addition, moving water conveys addi- tional nutriment ; for instance, sea-water contains but little iodine and calcium, yet large quantities of these are stored by many Algae. Move- ments of water are all the more essential inasmuch as many aquatic plants, and particularly algae, as a rule have no far-reaching roots (in a physiological sense). Finally, movements of water act mechanically, in that they stretch and bend plants with a force that varies with their strength. In larger plants mechanical tissue is developed*; calcareous incrustations may also contribute to the stabihty of sea-weeds, but it is worthy of note that chalk-forming algae and many crustaceous algae grow, some in deep, and others in still water. The general shape of aquatic plants is adapted to the surroundings in divers ways ; thus in rapid currents we find very elongated plant-members (ribbon-like fohage, and long filiform algae). A distinction must be drawn between currents and wave-movements, as many species withstand the former but not the latter. Very many species flourish only in tranquil water. Movements of water also favour the dispersal of propagative organs, such as detached vegetative parts, spores, or seeds.^ Aquatic species of plants throughout have a very wide geographical distribution. This is partly because the conditions of life are uniform, or only slightly different, over wide areas, and the transport of marine plants over great distances is very easy, and partly because many species are carried far by water-haunting birds and insects, or are transported by air-currents, as is the case with the smallest, mostly microscopic, species. Differences in the aquatic flora associated with geographical situation show themselves in some respects more marked in the sea than in other waters ; this may be due to greater physical differences, and to the universally greater constancy in amount of salt, in temperature, and other characters of sea-water. The general morphological and anatomical adaptation of submerged organs and aquatic plants has already been considered in Chapter XXVII. ' Ostwald, 1903 a ; Wesenberg-Lund, 1900, 1908. ' See Chap. XXXVIII. ' F. Darwin and D. Pcrtz, 1896. p. 296. * Wille, 1885. * Compare Hemsley, 1885; Schimpcr, 1891 ; Sernander, 1901 ; Rosenvinge, 1905 ; Kjellman, 1906. 154 HYDROPHYTES sect, iv CHAPTER XXXVH. FORMATIONS OF AQUATIC PLANTS The different oecological factors just dealt with yield many distinc- tions in the environments of aquatic plants, according to the modes in which they are combined. But another factor of equal importance to the production of formations, and one that lies at the very base of this, is edaphic in kind : it is the nature of the substrattmi, which brings with it many differences, including those in the mode of fixation. The first formation to be founded is that of piaiikton, which is consti- tuted of those microphytes that float free in water and are adapted to this mode of life. Closely allied to plankton, but of a subsidiary and less important nature, is the glacial community forming the cryophyte-ionnsition, which is composed of microphytes that are periodically exposed to ice-cold water. A third type is the hydrocharid-iovmation or pleuston^ which is constituted of macrophytes floating on water, or, more rarely, floating in it, and adapted to this mode of existence. Forming a sharp contrast to the preceding is benthos,^ which includes aquatic plants that are fixed to the substratum, or, hke some diatoms, creep over it. In opposition to free-living species, these display many types of construction associated with mechanical rigidity, and various other structural features. The formations under this head may be referred to two great and widely different groups dependent upon the nature of the substratum, according as this consists of solid rock or loose material. Accordingly the fourth group of formations is constituted of litho- philous hydrophytes, which are water-plants hving attached to stones : this group may be subdivided into several formations. The remaining plant-communities associated with a loose substratum are to be separated into those composed solely or mainly of microphytes, and those in which vascular plants or larger algae play the chief part ; and after this a further subdivision may be adopted according as the plants live in saline or fresh water. In addition the kind of soil is of influence, according as it mainly consists of inorganic bodies or organic fragments.^ Thus there result the following formations : microphyte- formation ; enhalid-iovnidXion (in the sea) ; and limnaea-ioxnx^Xion (in fresh water) : the last two are closely allied. Hydrophytic formations and their associations are often distributed in a definite manner, usually m zones, according to depth of water and the conditions involved in this : there are deep-water associations and littoral associations. In the deepest parts of the sea and of fresh water there is often mud, in which only saprophilous communities of microphytes {abyssal vegetation) can exist. In less deep parts there is a miuch richer littoral vegetation, which consists of more highly organized algae and cormophytes, and is arranged zonally according to depth, that is, accord- ing to illumination. The arrangement of these associations is greatly influenced by differences of soil and movements of the water. Very ' (C. Schroter und) Ivirchner, i8q6, p. 14; ' macroplankton ' (Chodat). ^ Haeckel, 1890. * See Chapter XVI. CHAP. XXXVII FORMATIONS OF AQUATIC PLANTS 155 disturbed sand is utterly devoid of plants, as is the case with vast tracts of the North Sea bed ; Hehgoland hes hke an oasis in a desert whose sandy surface is ceaselessly set in motion by breakers or by tides, and is therefore absolutely unfitted for the germination of algal spores. On the shore, hydrophytic formations often very gradually pass over into the marsh-plant formations : this is especially true of communities living on loose soil. Marsh-plants are also zonally arranged.^ Subjoined is a general synopsis of the oecological class constituted of hydrophytes, and grouped into formations, which will be treated in the succeeding chapters. I. FREE-FLOATING OR FREE - SWIMMING (PLANKTON AND PLEUSTON)— 1. Plankton-formation. 2. Cryophyte-formation. 3. Hydrocharid-formation (Pleuston). II. FIXED (BENTHOS)— A. To rocks or stones. 1. Lithophilous spermophytic formation. 2, Lithophilous algal formation (Nereid). a. Halophilous nereid-fcrmation. b. Limnophilous nereid-fcrmation. B. To loose soil. 1. Microphyte-formation. 2. Enhalid-formation. 3. Limnaea-formation. CHAPTER XXXVUI. PLANKTON-FORMATION The term ' plankton ' was introduced in 1887 by Hensen,- to denote bodies, dead or living, plants or animals, that float passively in water and are conveyed about by wind or current. Here we are concerned only with phytopiankton, which always consists of minute plants [micro- phytes) ; some of these are autophytes, which can manufacture organic substances from inorganic material, while a far smaller number are bacteria and fungi living on the autophytes or on their products. FLORA Phytoplankton-organisms are all minute ; they are solitar\' or colonial, and unicellular or multicellular. They belong to widely different syste- matic groups of low organization, to wit : — I. Cyanophyceae'^ may occur in masses and colour the water bluish- green, sap-green, grey -green, or red. In the sea there occur species of ' Warming, 1906. * Henscn, 1887. ^ Sec Willc, 1904. 156 HYDROPHYTES sect, iv Trichodesmium, including T. erythraeum (which colours the water red, and occurs in the Red Sea, also in other seas, but especially near the coast); Nodularia spumigena (which causes a greenish-grey colour, and is common and even extremely abundant in the Baltic Sea) ; HeUotrichum (in tropical parts of the Atlantic Ocean) ; in fresh water there are species of Anabaena and Polycystis, Aphanizomenon fios aquae, Oscillatoria rubescens (especially in mountain-lakes), Coelosphaerium kuetzingianum, Gloeo- trichia echinulata, and others that give to water a sap-green or bluish- green tint, and a peculiar aroma. The majority of them are genuine plankton-organisms which are found floating below the surface, but when the water is quite undisturbed swim in numbers on its surface, just like cream on milk. Klebahn and Strodtmann^ have minutely investigated certain small irregular bodies in their cells, and express the opinion that these are air-containing spaces within the protoplasm and enable the Algae to ascend ; on the contrary, H. Molisch and R. Fischer 2 maintain that the corpuscles are not air-vacuoles, but are com- posed of a peculiar substance of undetermined nature. Fischer and Brand 3 come to the conclusion that they have nothing whatever to do with the power of flotation. Ripe spores contain no air-vacuoles, and they sink. 2, Schizomycetes may be mentioned after the Cyanophyceae. They are found in the ocean even far from land down to depths of 800-1 loo metres, and in considerable numbers at depths of 200-400 metres. One and the same species shows great variability in form and size. The Schizomycetes concerned are motile, and most of them are of the spiral type, while some are luminous. In lakes the number of bacteria present shows the widest differences, as it varies between a few and many thou- sands of individuals to each cubic centimetre. The pelagic region of most lakes has the fewest.* In Zurich Lake there were at a depth of 80 metres 28-30 per cubic centimetre, in Lake Constance at a depth of 60-65 metres 31-146. The number present is smallest at the surface, and larger in the somewhat deeper strata. In the opinion of some investigators the bacteria at the surface are killed by light, but according to others the larger number of bacteria in the deeper strata of water is due to the greater abundance of decaying organic matter (dead plankton). Among Schizomycetes nitrifying and denitrifying organisms are of special significance in regard to metabolism in water, as they oxidize ammonia to nitric acid, or reduce any excess of the latter to nitrogen. Brandt ^ has put forward the hypothesis ^ that the greater activity of denitrifying bacteria in warm seas is the cause of plankton here being poorer than in colder seas. 3. Diatomaceae occasion brownish or greenish ^ tints in water, especially in arctic seas, where they occur in huge masses composed of countless individuals, particularly belonging to the genera Thalassiosira, Chaetoceras, Rhizosolenia, Coscinodiscus, and Thalassiothrix.® In fresh water there occur Melosira (especially in lowland lakes), CycloteUa ^ Klebahn, 1895 ; Strodtmann, 1895. "" R. Fischer, 1904, 1905. " Brand, 1905. * Forel, 1878, 1901 ; Schroter, 1897. * K. Brandt, 1904. * See also H. H. Gran, 1905, and Reinke, 1904. ' In regard to greenish water in the North- Atlantic Ocean, see K. J. \\ Steen- strup, 1877. ' See Gran, 1905. CHAP. XXXVIII PLANKTON-FORMATION 157 (especially in Alpine lakes), Fragilaria, Asterionella, and others.^ The genera Rhizosolenia and Attheya, in fresh water are represented respec- tively by a few and by one species, which are widely distributed ; the remaining species of these genera are marine. Some are sohtary, but many live combined in chains of various kinds. They are all true plankton-organisms which are incapable of forming masses swimming at the surface of the water. Some are enveloped in mucilage. 4. Peridineae (Dinoflagellata) occur especially in salt water. They are found in greater quantities, but in smaller numbers of species in tem- perate seas ; while the individuals are fewer, but the species more numerous and diversified in warmer seas - ; the genera Ceratium and Peridinium are particularly common. The numerous forms (geographical races) of Ceratium tripos play the greatest part in the sea ; while C. hirundinella occurs in large quantities in fresh water. They are provided with two flagella and are motile. Some marine forms are luminiferous, and in the autumn, when they are most numerous in the North Sea, Skager-Rak, and western Baltic, cause the sea to be phosphorescent. Nearly all plankton-organisms are included in the four groups named above. But both in the sea and fresh water there are other species of Chrysomonadineae (Phaeocystis, Dinobryon, and others). Only in the sea are found Silicoflagellata, Coccolithophoridae, and other Flagellata. The Coccolithophoridae are present in vast numbers, but, on account of their minuteness, escape through the dredging-net, so that their true abundance has not been appreciated until recently.-' Phaeocystis Poucheti is a flagellate which seems to be common to the coasts of Norway, the Faroe Isles, and Iceland, and was discovered by Pouchet in huge quantities by the Lofoden Isles. The protococcaceous Halosphaera viridis, which has the form of a sphere one millimetre in diameter, commonly occurs in the temperate and warmer parts of the Atlantic Ocean at the surface, or down to a depth of 200 metres, and has been found at the depth of 2,200 metres. At considerable depths (100-300 metres below the surface) of the tropical Atlantic Ocean, Halo- sphaera and the diatomaceous Planktoniella represent a kind of shade-flora. In fresh water there live many Chlorophyceae, among which some can be described as true plankton-organisms ; among these may be mentioned Sphaerocystis Schroeteri, Dictyosphacrium, Oocystis, Botryo- coccus, and Golenkinia radiata.* Occasionally Desmidiaceae, or Scene- desmus, Pediastrum, and other forms are intermingled with plankton in fresh water. ADAPTATIONS AND DISTRIBUTION The power of flotation shown by plankton-organisms has recently been investigated by Wolfgang Ostwald.-^ Power of flotation constitutes the main difference between plankton and all other communities of organisms. This flotation merely denotes a tendency to sink exceedingly slowly. If a body is to sink in a liquid, its weight must exceed that of an equal volume of the hquid. The rate of sinking depends partly on ' (Schroter und) Kirchner, 1896. ' Schiitt, 1893. ' See Lohmann, 190J. * See Chodat, 1898 ; (Schroter und) Kirchner, loc. cit. ' See Wesenberg-Lund, 1908. 158 HYDROPHYTES sect, iv the extent of its surface and its precise shape, which constitute the factor known as ' form-resistance ', and partly on the buoyancy (specific gravity and viscosity) of the surrounding hquid. Ostwald conchides that — The rate of sinking Excess of weight (of the organism over that of an equal volume of water) Form-resistance x viscosity (of the liquid). If therefore a body is to float, or, in other words, if its rate of sinking is to be reduced to a minimum, the numerator (excess of weight) must be de- creased, and the denominator increased as much as possible ; so as to make the quotient approximate to zero. Hence, in order to decrease their rate of sinking, plankton-organisms in the first place endeavour to decrease the excess of weight. This is effected by the nature of the cell-contents (products of metabolism, such as fat and gases, play a part), also by the thickness of the cell-wall (which is always extremely thin), and it must necessarily be different in species belonging to salt water and fresh water respectively. Plankton-diatoms have thinner walls than have those living on the bed of the water. It should be specially noted that fatty oil supplies an effective means of flotation, and that diatoms manufacture oil ; this fact sheds light upon the universal and great part played by diatoms in plankton .^ Fat is also manufactured by other plankton-organisms, such as Flagellata. Flotation-devices. Again, the organism strives to increase the ' form- resistance ' as much as possible, by relative increase of surface due to decrease in volume, and an absolute increase in surface due to deviation from the spherical form. To obtain a shape presenting the maximum vertical projection (transverse section), the flotation-apparatus appears always to be placed horizontally, and therefore at right angles to the direction of sinking. Schiitt ^ has demonstrated several arrangements designed to enlarge the surface of microscopic plankton-organisms, which thus acquire an increased power of flotation and of evading too sudden ascents or descents. Such sudden movements tend to be caused by changes in the physical characters of sea-water, and may endanger the lives of the organisms. Plankton-organisms, and particularly diatoms, are, relatively speaking, extraordinarily large in surface : in some diatoms and Peridineae the surface is increased by a flotation-apparatus in the form of wing-like extensions, threads, bristles, and spines ; or the body itself is as a whole filiform, sometimes curved or spirally coiled, as in some diatoms ; others are helmet-shaped, or parachute-like, or possess sail-like or annular processes ; still others are combined into threads, or into gelatinous masses. All these features remain inexplic- able, except on the theory here given, though in some cases spines may serve as means of defence against foes. Temporary changes in the form of the flotation-apparatus are known to occur.^ The preceding argument is strengthened by a comparison between plankton-diatoms and ground-diatoms. The latter are fixed or creep about, and possess in their shells sutures through which the protoplasm projects, so that they can move about, seek the most favourably illumi- nated spots, and fix themselves there. The majority of plankton-diatoms * Beijerinck, 1895. " Schiitt, 1893, * Wesenberg-Lund, 1908, and others. 11 CHAP. XXXVIII PLANKTON-FORMATION 159 have no sutures. Ground-diatoms, on the contrary, do not show the various outgrowths described above. Quantity of plankton. Their power of rapid division seems primarily responsible for the frequently enormous multiplication and great abundance of plankton-organisms. Their quantity, however, varies with time and place. ' Pure blue is the desert-colour of the high sea. With the green of meadows we may compare the colour of the vegetation of arctic waters ; yet the colour of the most vigorous vegeta- tion, of the greatest abundance of plant-life, is the dirty greenish yellow of the shallow Baltic Sea.' ^ Hensen ^ invented and employed methods for estimating the quantity of plankton.^ His object was to determine tlie amount of organic matter produced in the sea at a definite time and place; and this object is of profound importance, since all marine animal- life, both lowly and highly organized, is dependent upon those plankton organisms that are plants, or at least assimilate carbon dioxide : plankton is the ultimate source of food, which it perhaps supplies largely in the form of fat, manufactured by diatoms and by Peridineae, from which originate the great quantities of oil in marine animals (sea-gulls, whales, and all zoo-plankton). It is of no slight interest to note that it is not, as in the land-flora, starch, with its greater specific gravity, but oil, which is the main product of assimilation in the floating plankton-community. The first to appreciate the significance of the microscopic plant-world of the sea as the ultimate source of food-supply for animals was the Danish botanist A. S. CErsted, who came to this conclusion as early as 1845-8, when on his journey to Central America.* Composition of plankton. A distinction may be made between homogeneous and heterogeneous plankton. Plankton is sometimes extremely rich in species, but at other times, particularly when the organisms are so abundant as to colour the water, it is dominated by one or a few species, as in the peridinia-plankton in western parts of the Baltic, and in the diatom-district of arctic seas. It is especially the Diatomaceae, Peridineae, and Cyanophyceae that lend a colour to water. The colour of Baltic lakes is mainly determined by those of the chromato- phores of the dominant plankton-organisms. In harmony with the periodicity of the fresh-water plants, the colour of Baltic lakes under- goes regular change. The season when plankton-organisms determine the colour of the lake only to a slight degree is usually comprised of the early days of June, when diatoms have vanished and Cyano- phyceae not yet appeared. The colour of alpine lakes in arctic regions is only slightly tinted by plankton, because this is present only in small amount. Seasonal changes. Here it may be mentioned that plankton-asso- ciations change with the season, just as does vegetation on land, because they are dependent on' the temperature, illumination, and chemical properties of the water. For instance, in Skager-Rak and Kattegat during February and March there occurs a rich diatomaceous plankton composed of species that later (in April and May) appear on the coasts ' Schiitt, 1893. ' Hensen, 1887. * Sec also Haeckel, 1890, 1891. * Sec VVille, 1904 b. i6o HYDROPHYTES sect, iv of Iceland and Greenland ; in April and May there is another rich diato- maceous plankton, whose species demand somewhat higher tempera- tures ; in June and July, a less rich and more uniform diatomaceous plankton (with Rhizosolenia alata) ; from August to November, the warmest season, there is a rich plankton of Peridineae, often with an admixture of a diatomaceous one, which is rich in species and hkewise occurs on the southern coasts of the North Sea ; finally, in December and January there is a poorer plankton, consisting of remnants of the preceding. Most investigators assume that in the case of coast-forms the spores, which are mainly produced at the conclusion of the vegetative periods of the respective species, sink to the ground, and rest there until the commencement of the favourable season. In fresh water lakes the seasonal changes are not less remarkable than in seas : this is shown, for instance, by the illustrations in Wesen- berg-Lund's great work (1905).^ Associations of the open sea are more stable, as the voluminous masses of water in the ocean do not experience such considerable changes. In general the amount of plankton is small ; yet, on the one hand, plank- ton sometimes occurs in such abundance as to cloud the water, while on the other hand, it is meagre where different kinds of currents impinge. False plankton (Tycholimnetic plankton) is formed in fresh water by various Chlorophyceae, including Zygnema, Mougeotia, Spirogyra, and others, which at first are fixed, but subsequently break loose, and, owing to the entanglement among their threads of gas that they have evolved, they rise to the surface of the water and float there. Other plants, such as Sargassum in the sea, that are normally fixed, but become detached by constant movement of the water, may also be placed in this category. Geographical distribution. Of the geographical distribution of plankton- organisms little is yet known. In this respect we are best acquainted with the temperate Atlantic Ocean, North and Baltic seas, and the sea surrounding Iceland. Diatomaceae play the greatest part in the colder seas, Cyanophyceae in the tropics, and Peridineae in temperate and tropical regions. The species throughout have a wide area of distribution, because the external conditions may be the same over such vast expanses, and in widely separated spots ; yet each region has a characteristic flora.2 SUB-FORMATIONS Plankton may be classified into the following three sub-formations : — Haloplankton, salt-water plankton. Limnoplankton, fresh-water plankton. Saproplankton, foul-water plankton. A. Haloplankton. Salt-water Plankton. The plankton of salt water may be subdivided into neritic and oceanic haloplankton. Neritic plankton is confined to the coast ; in the tropics it consists of Diatomaceae, Cyanophyceae, and Peridineae, but very little is known ^ See also Kofoid, 1903, 1908. ^ The recent literature is cited in papers by G. Karsten, 1898, 1905-6, 1907. I CHAP. XXXVIII PLANKTON-FORMATION i6i concerning the matter ; in temperate regions the associations during the cold season are identical with those occurring during summer in the arctic region, but in the warm season they are different. Diatomaceae dominate in the temperate region, except in autumn (the warmest season, with a temperature of about 20° C.) when Peridineae or, in brackish water, Cyanophyceae (Nodularia) may be the characteristic plants. As regards the arctic neritic plankton it is known that during spring- time Diatomaceae congregate in numbers on the under-surface of the \ce} but that when the ice disappears pelagic diatoms and yellow Flagel- lata dominate. Oceanic or pelagic plankton, i.e. plankton of the open sea, mainly consists of Peridineae and Coccohthophoridae, but also of Cyanophyceae, Diatomaceae (relatively few species), and Halosphaera. It includes a large number of species of Peridineae and Trichodesmium in the tropics, and of Peridineae and Diatomaceae in temperate and arctic regions. These terms ' neritic ' and ' pelagic ' or ' oceanic ' plankton approxi- mately correspond to Haeckel's ' neroplankton ' and ' holoplankton ' respectively ; as most species belonging to neritic plankton spend only a part of their life as pelagic organisms, and during the other part are associated with the soil as spores and the like ; whereas pelagic forms always occur as organisms floating free in the water. Among oceanic species resting spores are unknown, but some species, including Rhizosolenia styliformis, produce microspores that are apparently alhed to auxospores.^ The depth to which phytoplankton descends in the sea differs according to various conditions, such as transparence of the water and the like ; in general it may be asserted that phytoplankton occurs to a depth of 200 metres, but that even at 100 metres its amount is smalP: diurnal migrations are unknown among phytoplankton-organisms. In recent years investigators * have established a number of associa- tions in phytoplankton, which are brought into existence by difference in temperature, and salinity of the water ; but very little is known of their oecology. Cleve has given names to them in accordance with their dominant genera or species, thus : tricho-plankton (after Thalassi- othrix), styh-plankton (after Rhizosolenia styhformis), chaeto-plankton (after Chaetoceras), sira-plankton (after Thalassiosira), tripos-plankton (after Ceratium tripos). B. Limnoplankton. Fresh-water Plankton. This form of plankton in fresh water is constituted of autophytic species. In large lakes it may be differentiated into two zonal sub- divisions : the pelagic in the open water, and the neritic near the shore. Other subdivisions may be recognized, such as potamoplankton, heloplankton, and probably several more. Limnoplankton appears to be one of the most cosmopolitan of for- mations. Concerning that in arctic and tropical countries we are almost entirely ignorant, and it is only in regard to temperate Europe and ' Vanhoeffen, 1897. ' Gran, 1902; G. Karsten, 1Q05-7 ; P. Bergen, 1907. ^ Compare previous remarks concerning shade-flora. See G. Karsten 1905. ' P. Cleve, 1897, 1901 ; Gran, 1900, 1902 ; Ostcnfcld, 1898- 1900. \V.\RMING M i62 HYDROPHYTES sect, iv temperate North America that we are well informed. Apparently great similarity prevails everywhere. In the northern part of the temperate zone and in the more southern alpine lakes diatoms dominate during most of the year. In the flat countries of Central Europe it has been observed that nearly everywhere there is a regular alternation of diatom-plankton during the cold months, and Cyanophycea-plankton during the summer. In not a few lakes an almost monotonous plankton is formed during spring and summer by Flagellata, especially by Ceratium hirundinella, and Dinobryon.^ The maxima of the respective associations depend upon the temperature of the water, and other consequent changes in it ; moreover, light and the amount of certain nutritive substances in the water have their influence. ^ C. Saproplankton. Foul- water Plankton In this sub-formation is included vegetation consisting of Flagellata such as Euglena viridis and E, sanguinea, of species like the colourless Polytoma uvella, also of various Cyanophyceae and Schizomycetes. Such vegetation generally occurs in small pools of stagnant water that is rich in putrefying organic matter but very poor in oxygen, as is the case with water (manure-water, puddles in roads) near human dwellings which may be distinctively coloured. The water is usually very green and sometimes stinking. The green organisms, Chlamydomonadae and others, presumably assimilate carbonic dioxide, and obtain nitroge- nous compounds as well as other nutriment from organic constituents present in the water ; they are therefore probably hemisaprophytes. Euglena sanguinea and others cause a red colour, and are most fre- quently motile. Among the saprophilous organisms in such water are many Infusoria. The farther putrefaction has proceeded the better do chlorophyU-containing plants, such as Scenedesmus, Raphidium, and diatoms, flourish therein.^ The ' self -purification ' of rivers, after the water below large towns has been polluted, is due to the activity of bacteria and other microphytes ; the process can be traced right to the stage when the water is devoid of organic substance. Sometimes the final product assumes the form of sulphide of iron, which is a constituent of black mud. Schenck * investigated the Rhine between Bonn and Cologne and came to the conclusion that Green Algae play no great part i in this process, and that filamentous and rod-shaped Schizomycetes absorb the organic substances.^ * Wesenberg-Lund, 1904-8. ^ Whipple, 1894, 1896. ' See Kolkwitz und Marsson, 1902; Volk, 1903; Marsson, 1907-8. * Schenck, 1893. * An immense Uterature deaUng with plankton has sprung up within recent : years : among the investigators may be mentioned Gran and Wille in Norway, Cleve in Sweden, Ostenfeld, Ove Paulsen, and C. Wesenberg-Lund in Denmark, Apstein, Hensen, Brand, Zacharias, Chun, Haeckel, G. Karsten, Kirchner, Lohmann, and Schiitt in Germany, Kofoid in North America, Chodat, Bachmann, G. Huber and Schroter in Switzerland. See also literature quoted by Oltmanns, 1905. i63 CHAPTER XXXIX. CRYOPLANKTON.i VEGETATION ON ICE AND SNOW This glacial vegetation is most closely allied to plankton. It has long been known that animals and plants live on the extensive snow- fields and glaciers of arctic countries, and such high mountains as the Alps, Pyrenees, and Andes ; they are in the main microphytes, yet, like plankton, they can occur in such countless numbers as to colour the snow or ice. The animals specially include Poduridae (Desoria saltans, and the blue Achorutes viaticus), Tardigrada, Radiolaria, and threadworms. To Wittrock and Lagerheim^ we largely owe our know- ledge of the vegetation, which consists mainly of aquatic plants, and particularly of algae (Diatomaceae, Chlorophyceae, and Cyanophyceae), Schizomycetes, and mosses in their protonema-stage. Some Phycomy- cetes also occur. In 1892 Lagerheim assessed the number of species at seventy-two. According to colour we may distinguish red, brown, also green and yellow, snow. Red snow is the commonest, and has been known for the longest time ; the tint varies from blood-red to rosy red, and from brick-red to purple-brown. It is caused especially by Chlamydomonas (Sphaerella) nivalis, and C. nivahs var. lateritia. This unicellular, spherical or ovoid alga has red contents, and colours the uppermost layers of snow to a depth of a few centimetres ; in the water derived from melting snow it multiplies by zoospores. In addition there occur, among other species, Gloeocapsa sanguinea, Cerasterias nivalis, Pteromonas nivaUs, as well as diatoms, and, in Ecuador, species of Chlamydomonas. Brown snow owes its colour to various species, including the desmi- diaceous Ancylonema Nordenskioldii, which contains violet cell-sap and, together with other algae and ' cryoconite ' (very fine mineral particles), has an important action on the ice in the interior of Greenland, as it absorbs the sun's rays more strongly than does the ice and thus melts deep cavities in the latter. In company with it live Pleurococcus vulgaris, Scytonema gracile, diatoms, and other algae. Green Snow is a phenomenon due to the presence of Green Algae, for instance Desmidiaceae, also to Raphidium nivale,^ Cyanophyceae, protone- mata of mosses, and green individuals of Chlamydomonas (Sphaerella) nivalis. Bright yellow or greenish yellow snow is tinted by another alga, possibly Chlamydomonas flavivirens, which is known to occur on snow- fields of the Carpathians. These plant-communities provide very evident examples of extra- ordinary powers of resistance on the part of plant-cells : except for the peculiar properties of the protoplasm they seem to be devoid of any means of protection against cold, though perhaps red colouring-matter may enable them to absorb heat.'' Durnig the greater part of the year tliey lie frozen in ice or snow and in the lasting darkness of the arctic night ; when the sun's heat melts the ice and snow, they awaken into activity, carrying on their processes of nutrition and reproduction in ' Schroter, 1904-8, p. 623. * Wittrock, 1883, p. 88^ ; Lagerheim, 1892. ' Chodat, 1896. * VVum, 1902. M 2 i64 HYDROPHYTES sect, iv water whose temperature scarcely exceeds o° C. In many places the water melted during the day-time freezes every night ; and thus they pass their life in ice and icy water.^ But in yet another respect the snow-alga is remarkably resistant ; it can preserve its existence in a dry condition even though exposed for months to relatively high temperatures.^ The same is true of certain animals living in snow. CHAPTER XL. HYDROCHARID-FORMATION OR PLEUSTON On the banks of stretches of fresh water, in places protected from currents or from the violence of waves (for example among swamp-plants), and in small ditches and puddles, there occurs a type of vegetation which is not fixed but is free-swimming, or rarely even in part free-floating like true plankton, of whose organisms it may contain an admixture ; never- theless this type of vegetation differs so essentially from plankton that it must be regarded as a special formation {megaplankton). From plankton it is distinguished by two features : 1. The occurrence of quite other growth-forms, namely megaphytes, including Spermophyta, Hydropterideae, and mosses. 2. The occurrence of algae belonging to groups entirely different from those in plankton. The growth-forms in the two cases are however very different, and for this reason the two formations are at least theoretically to be placed apart. Kirchner in 1896 employed the term ' pleuston ' to denote the real typical representatives of the hydrocharid-formation, as he laid stress on their ' saihng ' character,^ with which is also associated their adaptation to existence in contact with air (transpiration and the like). But in 1902 Schroter adopted the course, which we follow here, of includ- ing in the hydrocharid-vegetation such rootless, free-floating, submerged Spermophyta as Ceratophyllum, Utricularia vulgaris, Lemna trisulca, and other species. Moreover, with perfect justice he distinguishes between the constant and the temporary floating-flora. In the temporary flora he includes the masses of algae swimming at spring-time. These are more especially composed of Conjugatae, including Zygnema, Spirogyra, Mougeotia, but also of Oedogonieae, which ascend in great quantities and remain at the surface of the water, whither they are raised by bubbles of gas, as we have already explained. FLORA The constant representatives of the hydrocharid-formation belong to the following groups : Bryophyta, namely Riccia (with both submerged and swimming species), Amblystegium giganteum, and others. In pools on heaths or in pockets on heath-bogs one often finds vegetation very poor in species, and consisting of floating Sphagnum which nearly fills the water. » See p. 22. ' ^ Wittrock, 1883. ttXuv, to sail; see Schroter und Kirchner, 1 896-1902. (HAP. XL HYDROCHARID-FORMATION OR PLEUSTON 165 Hydropterideae, with Azolla and Salvinia, both of which are swimming plants. Spermophyta, which can be ranged into the subjoined groups : — Submerged : Ceratophyllum, Utricularia, Aldrovanda, Lemna trisulca, Stratiotes aloides. Floating by leaves or shoots : Hydrocharis, Hydromystria stolonifera (Trianea bogotensis), Lemna minor, L. polyrhiza, L. gibba, Wolffia arrhiza, Pistia, and Eichhornia crassipes. Transitional to the rooted limnaea-types : Hottonia palustris, Jussieuea repens, and others. Many spermophytes, such as Lemna, Pistia, and Pontederia crassipes, can choke the water with their enormous numbers. ADAPTATIONS The submerged species, as in the case of plankton-organisms, must be approximately of the same specific gravity as water ; normally floating species are kept at the surface of water by air-containing cavities in their leaves and stems. This finds outward expression in thickness of the shoots and great convexity of the lower face of the floating organ of Lemna gibba and Hydromystria. Special floating-devices are shown by Eichhornia crassipes, Neptunia and Jussieuea repens.^ As in some cases the assimilatory organs projecting into the air are necessarily adapted to transpire, this formation shows a certain transition to land-plants, just as, on the other hand, through its submerged species it is allied to plankton. The morphology of the shoot varies. In the majority of submerged Spermophyta the shoots have very long internodes and very thin stems ; tire usually sessile or shortly stalked leaves are often divided into filiform segments, as in Utricularia, Ceratophyllum, and Hottonia. But in floating species the shoots mostly have short internodes and are condensed ; while the leaf-blades frequently assume the shape of typical floating-leaves, being very broad, peltate-cordate or ovate-cordate, as in Riccia natans, Salvinia, Hydrocharis, Hydromystria, Lemna polyrhiza and other species, and Azolla ; somewhat differently shaped are the shoots and leaves of Pistia. One of the offices of the floating leaf is to ensure equilibrium to the plant in water ; accordingly, floating leaves or analogous balancing- organs are developed early in the seedhngs of SaMnia, Lemna, and some others. 1 That this broad distinction between submerged and floating leaves represents a true adaptation to environment is clearly shown by Salvinia, and by aquatic plants such as Ranunculus (Batrachium), Trapa, and Cabomba, that are fixed by roots ; for all these have both submerged and floating leaves which differ in form. In free-floating submerged plants nutriment is absorbed over the whole surface, and in Vascular plants the root is accordingly either absent, as in Aldrovanda, Wolffia, Lemna trisulca, Ceratophyllum, and Utricularia vulgaris, or very reduced ; the most important part played by the root in such plants as Lemna and Hydrocharis is indubitably to secure the plant in a definite ' Gobcl, 1 89 1. i66 HYDROPHYTES position and to prevent its being overturned — and the same function is performed by the submerged leaves of Salvinia. Propagation. The division of the vegetative organs plays an important part in all cases. Not only algae, but also Pteridophyta Uke Azolla, and Spermophyta such as Lemna, Hydrocharis, Stratiotes, multiply with exceeding rapidity by division, and for this reason they are markedly social and occur in great abundance. Vegetative parts serve as convenient means of dispersal, for instance in Lemna ; the small shoots of Wolfiia brasiliensis are distributed by aquatic birds. Accordingly, the produc- tion of spores and seeds is in a number of cases almost unknown or, as in Lemna, very rare. Fertilization is necessarily connected with the water in Cryptogamia ; moreover, a few Spermophyta including Ceratophyllum open their flowers under water ; but the flowers of the others are developed in the air, and are mainly entomophilous, as in Utricularia, Hottonia, and Hydrocharis. The fruit is in most cases ripened under water. Hibernation and duration of life. Nearly all are perennial, as is true of aquatic plants in general. Salvinia and many algae are annual. Flowering plants often produce special bud-hke winter-shoots — hibernacula — which sink to the bottom in autumn ^ ; among such plants are Hydrocharis, Utricu- laria, Aldrovanda, and Ceratophyllum ; or after the death of the older shoots the younger ones, which are filled with reserve-food and do not yet contain much air, sink and hibernate without further modification — such is the case with Lemna. Certain algae, Cladophora fracta for instance, show similar behaviour, as they sink to the bottom in autumn, hibernate in the form of thick-walled cells which have rich contents, and develop in spring-time into new individuals.- DISTRIBUTION AND ASSOCIATIONS As the most prominent species in this formation float about on the surface of water, they are easily transported by wind and currents to quiet spots, where they may be collected together in vast numbers, as may be seen in the case of Lemna. Huber^ gives some information in regard to the floating islands in calm inlets of the Amazon ; these are often very extensive, and are formed partly of pleuston — for example, of Eichhomia azurea — but also of half-floating grasses which do not belong to this forma- tion, also of other marsh-plants that have broken loose. In hke manner immense multitudes of Eichhomia crassipes occur in North American rivers. This formation seems to be strictly confined to fresh water. Various associations may be distinguished according to the dominant species (pontederietum, lemnetum, pistietum, and the Hke). Schroter* separates the emersed hydrocharids (for example, lemnetum) from subtnerged types (including the associations, ceratophylletum, scenedesmetum, and zygnemetum) as separate formations. * See Schenck, 18866; Raunkiar, 1895-99. * Wille, communicated by letter. ' J. Huber, 1906. * Schroter, 1902. 167 CHAPTER XLI. LITHOPHILOUS BENTHOS This type of vegetation is confined to rocks, loose stones, mollusc shells, and similar solid substrata near shores and banks. Many of the species growing on these substrata also li\'e as epiphytes. FLORA The salt-water communities are solely composed of algae, which here reach their highest and richest development in all four colours (blue-green, pure green, brown, and red), and exhibit extraordinary variety of form. The fresh-water communities, though much poorer, consist of algae (nearly entirely Chlorophyceae, Cyanophyceae and Diatomaceae), of mosses (Fontinalis, Dichel3nTia, Cinchdotus, and others), and of Spermophyta, especially Podostemaceae. The chemical nature of the substratum plays a part that, so far as is known, is shght and solely concerns the presence of calcium. Some algae flourish only on Ume, which is perforated and corroded in furrows by their hypha-hke filaments.^ Most of the others grow equally well on stones, piles, shells, or on other algae. In addition, the inclination, illumination, and physical nature of the substratum (rock, stones, or calcareous shells) has its influence on the distribution of species.- According to Wille a sub- stratum of shells is distinguished by special algal associations, for instance, by Tilopteridaceae. ADAPTATIONS The general pecuharities of submerged hydrophytes, such as reduction or absence of stomata, of lignified constituents, and of wood-vessels, the production of assimilating chromatophores in the outermost layer of cells, and so forth, have already been dealt with.^ The assimilatory tissue extends to the surface ; but beyond this many algae possess an internal assimilatory tissue, which undoubtedly utilizes the carbon dioxide produced by respiration.^ Specialized adaptation is revealed in the undermentioned directions : Solidity of the substratum necessitates the possession of haptera,^ which m connexion with algae are sometimes described as roots though they arc widely different from true roots. They occur in two forms ; as circular disks, in Fucus vesiculosus and Laminaria soHdungula and others, or as branched digitate or coralloid structures, in Laminaria saccharina and Agarum Tumeri and others. Here, too, may be placed the tufted rhizoids of Fontinalis and of other aquatic mosses. The adaptive means of fixation ha\'e been discussed by Wille.^ Haptera in some cases have the structure of root-hairs, but in others they are solid, multicellular bodies. The firmest attachment is exhibited by such crustaceous algae as Lithothamnium, Lithophyllum, Hilden- brandtia, and Lithoderma, which form an incrustation on rock. Diato- maceae and Desmidiaceae that are fixed to other bodies by mucilage belong to a special type. ' Chodat, 1902; Lagerheim, 1892; Colin, 1893; Nadson, 1900 ; M. le Roux, 1907 ; P. Boysen-Jenscn, 1909. ' Sec Borgesen, 1905. ^ See Chap. XXVIII. * Wille, 1885. ' Warming, 1881-1901. " Wille, 1885, and others : see Oltmanns. 1905. i68 HYDROPHYTES sect, iv Creeping (migrating) lithophilous species are uncommon, yet are met with among Florideae, Caulerpa, and Podostemaceae. The last named have creeping roots, but agree with algae in their mode of attachment, since the roots only indirectly play a part in fixation by bearing haptera ; they have no organs specially set apart to absorb nutriment. Air-containing intercellular spaces are entirely lacking, or at most are very small and contain scarcely any air. Exceptions to this rule occur in the sub- aerial inflorescences of Podostemaceae, and in the floating-apparatus of cer- tain algae, such as Fucus vesiculosus, Halidrys siliquosus. Chorda filum, and AscophyUum nodosum, which live in the littoral belt or in shallow water. Through this character, lithophilous vegetation stands in sharp contrast with other types of aquatic vegetation. The reason for it is presumably that all the lithophilous plants in question live in troubled water, where they secure rich supplies of air : Podostemaceae mostly find their homes in cascades. The necessary power of resistance to rupture is structurally provided in various ways by mechanical, mainly collenchymatous, tissue.^ The excretion of calcium carbonate within ceU-walls takes place in a number of algae, and with this must be mentioned the silica occurring in Podostemaceae. These excretions in some cases play a mechanical role, while in others they seem to increase the longevity ; certain incrusted algae are perennial, whereas their non-incrusted allies are annual. Heavy production of mucilage takes place especially in species growing in the littoral belt, and perhaps serve to prevent desiccation while the tide is down. It is calculated also to diminish friction between water and plants, and thus to preserve the latter from the violence of breakers. ^ The Plant-shapes are extremely varied; and by no means aU seem capable of interpretation as adaptations to environment : — ^ Crustaceous type. There are among algae and Podostemaceae (Erythro- lichen, Lawia, Hydrobryum) crustaceous forms, which are weU-suited for existence in very perturbed water ; but many crustaceous Algae grow in deep, and therefore in but slightly disturbed, water. Myriophylloid type. There are species among algae and Podostemaceae that are structurally analogous to gills of animals, as they are cut up into many capillary segments, by which the surface and assimilatory activity are both increased. Muscoid type. There are moss-shaped forms among mosses (Fontinalis) and Podostemaceae (Tristicha hypnoides, species of Mniopsis and Podoste- mon). Filiform type. There occur species shaped like an unbranched thread, which experience passive undulatory movements in the water ; such are Chorda filum, many fresh-water algae, and the podostemaceous Dicraea elongata. Phylloid type. Leaf-Mke forms are presented by Porphyra, Laminaria, Ulva, and Monostroma, as well as by the podostemaceous Marathrum, Oenone, and Mourera. Special stress must be laid on the parallelism between the shapes of the marine algae and Podostemaceae, as demonstrating that these shapes are adaptive. In the sea at the southern point of South America there occur peculiar algae (Macrocystis and Durvillea) which have ' floating fronds '.^ ^ Wille, 1885. ' Wille, 1885 ; see p. 99. ' See Oltmanns, 1905. * Compare Hooker, 1847 a. CHAP. XLi LITHOPHILOUS BENTHOS 169 FORMATIONS As the lithophilous Spermophyta are biologically \'ery different from the algae, which are always submerged and are provided with quite other methods of reproduction, lithophilous \-cgetation must be divided into at least these two formations : 1. Lithophilous Spermophyta. 2. Algae (nereid formation). It may seem subsequently correct to establish several additional forma- tions, for instance, those of mosses and of diatoms. I Formation of lithophilous Spermophyta. Two remarkable families only, the Podostemaceae and Hydrostachydaceae (which were formerly combined into one) are represented in this formation. They are adapted for existence on submerged rock, in powerfully disturbed water (cascades) and have consequently evolved a number of structural features, affecting their external morphology as well as their internal struc- ture, which are unique in the plant-world.^ They are almost confined to the tropics ; they occur in America from Uruguay to the southern United States, in Africa (where many most interest- mg forms are found), in Madagascar, and extend from India to Java ; towards the east they become rarer, so that only a single species seems to hxe in Austraha. Mosses and algae may be intermingled with these Spermophyta. 2. Formation of Algae (nereid-formation). This must be divided into at least two sub-formations : (a) Fresh-water (Limno-Jiereid) ; (b) Marine (Halo-nereid). (a) Limno-nereid communities are poorer in species, individuals, and shapes than are the marine ones. In comparison with the latter they display far less luxuriance and variety of form. Nearly all belong to the Chloro- phyceae and Cyanophyceae ; but there also occur diatoms, a very few Phaeophyceae (including Pleurocladia lacustris) and Florideae (Lemanea, Batrachospermum, for example). In accordance with the variety displayed in environment, many associations (and possibly sub-formations) must be distinguished. For example : — Icy mountain streams have a quite peculiar flora, including Hydrurus, Prasiola fluviatilis, Tetraspora cylindrica, and others. ^ On stones in shallow water along lake-shores there is an entirely different flora, with species of Cladophora, Rivularia, and Diatomaceae. Sometimes associated with the algae are various mosses, among others, Fontinahs. Schroter and Kirchner^ recognize in Lake Constance an cncyoncmduni with different \-arieties, namely, spirogj^'retum, tolypotrichetum, and schizotrichetum. Stones near the shores of fresh-water lakes are often encrusted by lime- producing algae."* According to Wesenberg-Lund, incrustations of lime are produced mainly by Cyanophyceae (Schizothrix, Rivularia), but also by diatoms, Chlorophyceae (Cladophora), and the phaeophyceous Pleuro- cladia lacustris. These incrustations occur especially on stones where the ' See Warming, 1881-1901. * Lagerhcim, 1892. ' Schroter und Kirchner, 1896, 1902. * Chodat, 1902 : Forel, 1901 ; Schroter und Kirchner, 1896. 170 HYDROPHYTES sect, iv shore is flat ; only rarely are they met with at a depth of one metre below the water-surface. In summer, when the water sinks and many stones are uncovered, the incrustations crack and fall off; while in winter they are rubbed off by ice. Thus it is that algae can contribute to the deposit of lime in lakes. (6) Halc-nereid communities are those of salt-water. There are wide distinctions between the floras of different seas ; but even on a single coast there are many geographical features due to differences in the oecology of the \-arious species, as is briefly indicated in the succeeding paragraphs. The oecological distinctions depend largely upon differences in tempera- ture, salinity, movements, and illumination of the water, as well as upon fluctuations in these ; one important consideration is whether or no the algae are periodically laid high and dry owing to tides, and another is the height up the rocks to which breakers reach .^ A prominent part is played by differences in the soil (solid rock and its mineralogical nature, stones, rubble). The temperature of the sea is of importance. The most luxuriant ' forests ' of Brown sea-weeds are developed in the coldest seas (frigid seas, North Atlantic Ocean, coasts of Tierra del Fuego, southern point of Africa), possibly because cold water is richest in air. In the southern seas named, some individuals (Macrocystis, Lessonia) are hundreds of feet in length ; in the North, species of Laminaria attain considerable dimensions — for instance, near Greenland Laminaria longicruris attains a length of 25 metres, and Nereocystis 13^ metres in the Pacific Ocean. In tropical seas species are smaller throughout. At Spitzbergen at the depth where vegetation is richest the mean temperature of the water may not exceed at any season of the year 0° C.^ The seasonal phases of species, according to Rosen\ange and others,^ are strongly marked, and a number of species present utterly different appearances at different seasons. Some, including Chorda, Nereocystis, and a few other Laminariaceae, are annuals ; but in other species larger or smaller parts — for instance, haptera and the inferior portions of the thallus — perennate. Rhodomela subfusca in the Baltic Sea during April and May bears a richly branched shoot-system with reproductive organs which are subsequently shed. Desmarestia aculeata likewise varies greatly in appearance with the season. Some — Delesseria sanguinea, for example — fructify only in winter. The cold water is richer than warm water in oxygen and carbon dioxide and hence provides far better nutrition.* KjeUman's noteworthy account of algal life in extremely northern seas has already received attention in this work.^ Salinity of the water is another profoundly important factor affecting the composition and appearance of vegetation. The farther we proceed from the North Sea up the Baltic the less saline becomes the water,® and the poorer and more reduced the vegetation. The Siberian region of the Arctic Ocean is likewise poor in species, partly because the sea-bed is mainly sand or clay, and partly because of the volume of fresh water pouring out of Siberia. To fluctuations in salinity and temperature many species are exceedingly sensitive. Some species endure a slight decrease in salinity, others can adjust themselves to circumstances. Movements of the water (wave- violence, currents) and consequent increased ^ Borgesen, 1905. " Kjellman, 1875. ^ Rosenvinge, 1898 ; see Oltmanns, 1905. * See p. 151. '^ See p. 22. ® See p. 152. CHAP. XLi LITHOPHILOUS BENTHOS 171 supply of oxygen and food-material influence the distribution of associations. The algal vegetation on exposed coasts as a rule differs considerably from that found on sheltered coasts. In this connexion reference should be made to Hansteen's ^ investigations of the flora outside and inside the Norwegian rocky shoals and Borgesen's - on the Faroe Isles. Hedwig Loven^ investigated the respiration and the gas contained in the vesicles of algae, and came to the following conclusions : — The gas inside fucaceous vesicles is different in composition from that of the air in the water ; The amount of oxygen is at the maximum at midday, and at a minimum during night-time ; Algae can extract from water e\'ery trace of oxygen, but they live for a tolerably long time in water devoid of oxygen, and excrete into it a con- siderable quantity of carbon dioxide ; if oxygen be lacking in the water then they can completely exhaust the oxygen contained in the vesicles. Light. In the first place i7itensity of light is of significance ; the Green Algae are those most photophilous, and, according to Kjellman, it may be for this reason that they are enfeebled as well as scanty in the north Arctic Ocean (though they are luxuriantly developed along rocky coasts of Greenland.) The deeper one goes the more light is absorbed, and the fewer become the species until at last they are absent. According to Berthold, Florideae are generally shade-loving plants. Berthold * found at Naples a luxuriant algal vegetation at a depth of 120-130 metres, whereas in the Arctic and North Atlantic Oceans only a poor vegetation subsists at a depth of even 50-60 metres.-^ The differences in their demands for light cause algae to be distributed in zones according to depth. ^ Then the colour of light changes with the depth,' and correlated with this change are the colour of the algae and their distribution in ' regions '. Lyngbye in 1836 established the ' regions ' of the Ulvaceae, Florideae, and Laminariaceae ; Agardh in 1836 and Orsted in 1844, established those of the Green, Brown, and Red Algae. But Orsted in 1844 was the first to assume the connexion between colour of light and depth of the strata in which algae occur ; in the Ore-Sund he recognized the following ' regions ' commencing at the surface and descending : — 1. Regio algarum viridium S. Chlorospermearum. Sub-regio Oscillatorinearum. Sub-regio Ulvacearum. 2. Regio algarum olivacearum S. Melanospermearum. Sub-regio Fucoidearum et Zosterae marinae. Sub-regio Laminariarum. 3. Regio algarum purpurearum S. Rhodospermearum. Kjellman has divided the algal vegetation off the Swedish part of the Murman coast, and in other seas, into three ' regions ', which run parallel with the coast, and are in turn composed of a great number of small ' formations ' (i.e. associations), according as one or another species predominates. The three regions are the following : — ' Hansteen, 1892. ■ Borgesen, 1905. ' Loven. i8qi. * Berthold, 1882; see Oltmanns, 1905. ' Rosenvingc, 1898; Borgesen, 1905. ' Concerning algal vegetation in caves of the Faroe Isles, sec Borgesen, loc. cit. ' See p. 150. According to Gaidukow, 1904, the colours of algae arc to be regarded as adaptations to the quality of the light present. 172 HYDROPHYTES sect, iv 1. Littoral ' region ' ; — Stretches between the high-tide and the low-tide marks, and includes many Green Algae, Brown Algae, and some Red Algae. At low tide these lie uncovered; many may be described as nearly amphibious, for they become dry on bright sunny days. In extreme arctic seas the rubbing action of the masses of ice prevents any strong development of this ' region '. 2. Sub-littoral ' region \' — Ranges from below low-tide mark down to a depth of twenty fathoms (40 metres) ; here algae of all colours are represented, but Green Algae cease and Red Algae become more numerous with increased depth. 3. E littoral ' region ' ; — Is below the preceding and descends as deep as light ; it is poorer in species and individuals ; moreover, the latter are smaller and distorted, as Lyngbye has already noted. Hansteen and Gran on the whole approve of the preceding scheme ; but Reinke, basing his conclusions on investigation of the Baltic Sea, from which ' region 3 ' is absent, suggests a further partition of regions i and 2 into two sub-divisions each. At a depth of 4 metres many species find their lowest limit. Rosenvinge and Borgesen ^ indicate that the littoral ' region ' ' extends far beyond the highest tide-mark, in the Faroes even in some places about 25-30 metres beyond ', and they set the ' lowest limit of the sub-littoral region above the lowest ebb-marks '. Rosenvinge and Borgesen do not recognize any elittoral ' region '. Some algal hairs are assimilatory, as in Desmarestia aculeata and Chorda tomentosa, but others are colourless, especially among Red Algae. Those of the latter kind are strongly developed when the light is more intense, and Berthold has made the hardly probable suggestion that their function is to regulate the illumination ; they would seem rather to be respiratory or absorbing organs. ^ The factors already enumerated affect the vegetation notably as a whole but also in detail, and, possibly with the co-operation of other factors (kind of rock, and other topographical features), they aid in evoking a number of miniature topographical distinctions and a number of associa- tions, which may be profusely intermingled and may owe their appearance mainly to one or a few species that compose their main mass.^ (Phycolo- gists often apply the term ' formations ' to these miniature purely floristic groups ; but it must be noted that ' formations ' should be oecologically founded upon the forms of the algae, and that true ' formations ' may here be quite out of the question. These groups then are, at least tentatively, to be regarded as associations.) Within extensive communities of large algae — for example, among Laminaria stems — many epiphytes and many humbler forms find suitable homes ; thus an underlying vegetation of plants requiring less light arises, just as in a forest. As the various factors enumerated act with unequal intensity at different seasons of the year, there arises a differentiation in time of the development of the nutritive and reproductive organs. Each species of marine alga seems to have its definite season of development, which may differ with the latitude : species that in Denmark disappear with ^ Rosenvinge, 1898 ; Borgesen, 1905. ' See p. 168, and Wille, 1885 ; Rosenvinge, 1903. ^ Kjellman, 1878 ; Hansteen, 1892 ; Borgesen, 1905 CHAP. XLi LITHOPHILOUS BENTHOS 173 the commencement of summer, may in arctic seas persist throughout summer.^ In Danish waters the marine algal vegetation in summer differs widely from that in winter,- and even in the more southern latitudes of Naples the same has been noticed.^ In the latter place illumination and breakers are the determinants, but in higher latitudes temperature certainly plays a more important part. A pecuhar group, that of the Diatomaceae, merits special notice, as it includes forms deviating from all other plants. Among them are the ground-diatoms showing various growth-forms, including motile forms which creep over the substratum (stones or other algae), and stalked non-motile forms, which especially inhabit the marginal zones of salt-water, are easily detached and then can intermingle with plankton."* These communities of diatoms may perhaps be regarded as constituting a separate formation which includes many associations. Moist rocks on sea and land may bear a vegetation which is transitional between submerged rock-vegetation and land- vegetation, and may give rise to a special formation. Inland nereid-communities are dependent upon great atmospheric humidity and trickhng water, and they therefore develop luxuriantly only near waterfalls, whose spray habitually wets the rocks, and in countries where atmospheric precipitations are heavy and fall throughout the year (as in Java), and in the cloud-belt of moun- tains. On rocks that are wet by fresh water there may be formed a spongy felted carpet of algae (including Trentepohha, Rhodochorton islandicum, R. purpureum and others), mosses, ferns, and other herbs ; and there may even be found small shrubs that are always wet or dripping with water. On rocky coasts foam of the breakers may be carried especially high, and in these same places marine algae, such as species of Prasiola, Ulothrix and Enteromorpha, Calothrix scopulorum, also Bangia fusco-purpurea and Hildenbrandia rosea, may occur far above the high- tide mark.^ Thus arises a kind of supra-littoral ' region '. Various crus- taceous lichens, including Verrucaria maura, are intermingled therein. The oecology of this community differs from that in water in so far as the constituent species must be fitted to endure greater dryness than in the case of submerged types. CHAPTER XLII. BENTHOS OF LOOSE SOIL The structure of soil has already been described,^ but in the soil now under discussion the interstices arc filled with water, and air occurs in extremely small quantities or not at all. The texture of the soil may vary, being mud, clay, or pure sand, which is mostly quartz-sand or, in the tropics, coral-sand, and may be mixed with marine shells and stones, more or less small according to the violence of the waves. These differences in texture cause floristic distinctions, though hardly anatomical or morphological ones. But nothing furtlier is known in regard to this matter. An exceptional kind of soil is mud that consists of dead organic matter. ' Rosenvinge, 1898. ' Kjellman, Rosenvinge, and others; as cited by Oltmanns. 1905. ' Berthold, see Oltmanns, IQ05. * Schiitt, sec Oltmanns, 1905 ; compare p. i 56. ' Rosenvinge, 1903, see Oltmanns, 1905; Borgcsen, 1905. ' See Chap. X. 174 HYDROPHYTES sect, iv Movements of the water, on the contrary, are of great morphological and floristic significance. Salinity of the water is of even greater import. Vegetation in the sea is morphologically and oecologically very different from that in the majority of fresh waters (rivers, lakes, and pools). Loose soil, in contrast to a stony substratum, entertains very few algae, but mainly Spermophyta. ADAPTATIONS Of the modifications of vegetation in such environment we may note — Roots, or root-hke organs branching in the soil, serve to attach the plant and absorb nutriment ; apart from these, special organs of fixation are lacking. Roots do not attain the dimensions or degree of branching displayed by land-plants, and some are devoid of root-hairs,^ for instance, Hippuris (excepting at the 'collar'), Elodea, Hottonia and some others. Some tropical algae (species of Udotea, Halimeda, Penicillus ^) on sandy soil and mud are fixed to the loose soil, and obtain nutriment therefrom by means of hypha-like hairs attached to the lower parts of the thallus that penetrate the mud ^'^ ; the same is true of Characeae. Horizontal rhizomes, or their analogues (in Caulerpa, for instance), creeping on or, more usually, in the soil are very common, and bring into existence a dense vegetation composed of numerous social individuals, such as is exemplified by the submarine ' meadows ' of Zostera and other ' grass-wracks '. This mode of growth clearly harmonizes with the loose texture of the soil.^ Gaseous interchange on the part of submerged plants is aided by the large air-containing intercellular spaces which are peculiar to aquatic plants.* FORMATIONS AND ASSOCIATIONS Three formations may be distinguished : i. microphyte-formation ; ii. enhalid-formation; iii. limnaea-formation. i. Microphyte-formations. Pure associations of microphytes, particularly of Cyanophyceae appear in extreme circumstances, chiefly in hot springs, and in the shallow beds of seas and fresh waters where organic constituents abound ; also not infrequently in richly humous shallow waters of heaths. These microphytic communities differ so widely from the limnaea-formation that they must be regarded as constituting a distinct formation, or perhaps two — the autophytic and the saprophytic — in which several sub-formations may be distinguished. A. Autophytic Microphyte-formation. I. Sub-formation living in hot springs. This occurs in various parts of the Earth. The temperature varies widely in these springs ; at relatively low temperatures Phanerogamia still occur in them, but high temperatures exclude all plants save Cyanophyceae (Beggiatoa, Lyngbya, Oscillaria, Hypheotrix, and others). Many of these species are ubiquitous. They form green, yellow, white, red or brown mucila- * See p. 97. ' Borgesen, 1900. ^ See p. 42. * See p. 98. CHAP. XLii BENTHOS OF LOOSE SOIL 175 ginous or filiform masses, which are often several centimetres thick and sometimes present the appearance of almost structureless jelly. In European hot springs we know of Anabaena thermalis (in water at a temperature up to 57° C), species of Leptothrix (in Karlsbad at 557° C.), Beggiatoa, Oscillatoria (44°-5i° C), Hypheothrix (in Iceland), Tolypothrix lanata (in Greenland at Unartok, 40° C), as well as others. Lyngbya thermalis is known to occur in Iceland, in Italian mud-volcanoes, and in hot springs in Greenland. Many Schizomycetes occur, including true thermophilous species (of which some dozen have been described), sulphur-bacteria, iron- bacteria, and others. Moreover, many diatoms and other more highly organized algae occur.^ The highest temperatures so far noted in this connexion are 8i°-85° C. in Ischia, 90° C. in the Azores,'- and even 93° C. (200° F.) in California.^ At Las Trincheras in Venezuela there is a warm spring whose temperature at its source is 85°-93° C. ; the algae in this grow in water whose tempera- ture exceeds 80° C. The water of many hot springs, which mainly occur in volcanic regions, contains sulphur, calcium, or other mineral bodies, without changing the composition of the vegetation. A special form of activity is displayed by certain of these algae, which excrete crystalhne concretions of calcium carbonate or of siliceous sinter ; in the Amo travertine is deposited by Cyanophyceae ; and in the hot springs at Karlsbad large masses of calcareous sinter are excreted. In North America numerous hot springs and geysers occur in Yellow- stone Park, and Weed* describes the remarkable stone-forming activities here exhibited by algae ; these grow in waters at temperatures of 30°- 85° C. and lend to the water their various rainbow tints, ranging between red, orange, white, and green, according to the temperature. Cohn '" expresses the opinion that these algae have a peculiar faculty of storing up calcium carbonate. Is it not possible that these thermophilous communities of the most lowly organized algae living in hot springs present us with a picture of the oldest vegetation on Earth? Perhaps the Cyanophyceae can even assimilate free nitrogen. 2. Sub-formation of Sand-algae — the aestuarium of tidal shores. On the sandy shores of seas of Northern Europe and of lakes (for example, Michigan, Fureso in Denmark^), there lie, within reach of the tide-water, communities of algae, which form a thin film on or under the surface of the sand, and give to the shore a distinctive colour if they be present in large quantities. On the coasts of Denmark there are various associations of such : chlaniydomonadeia, composed of species of Chlamy- domonas and Diatomaceae, which he loose on the sand which is not cemented together : phycochromaceta formed by Blue-green Algae and diatoms, which, by means of their mucilage, glue together the grains of sand so as to form a thin, firm, more or less crustaceous layer, which is * Respecting North America, see Harshberger, 1S97 ; Josephine Tildcn, 189S ; respecting Japan see Miyoshi, 1897 ; and respecting Hungary, see Istvanffi, 1905. * Moseley, 1875. ' Brewer, 1864. ' See the hterature quoted by Weed, 1887-9. ' Cohn, 1892. • Cowles. 1899. 176 HYDROPHYTES sect, iv usually visible immediately beneath the actual surface of the sand. With this plant-community is combined a remarkably peculiar community of animals. This association has an enormously wide distribution along shores of the North Sea, on places where high-sands and flats (aestuaria) are miles in width. Closely allied to this subformation is a community of Cyanophyceae and diatoms which is hkewise placed within reach of the tide-water, as it occupies mud-flats and thrives on marshy soil that is flooded by sea- water during the spring tides (aestuarium).^ We treat of these sand-algae here as essentially belonging to the benthos of loose soil, but in our classification they have no less claim to be considered amongst both the halophytic and psammophilous com- munities. B. Saprophytic Microphyte-formation. The vegetation on dead organic accumulations at the bottom of calm water consists of Oscillatorieae, Beggiatoeae, and Schizomycetes, but sometimes also of additional algae whose true home hardly lies here. The accumulations usually lie loose on decaying soil ; Beggiatoa lives in chalky white flocculent masses (beggiatoeta) ; while Clathrocystis roseo-persicina, as well as Bacterium sulphuratum, B. Okeni, and other sulphur-bacteria coloured with bacterio-purpurin, are in red masses. It is in calm inlets with shallow brackish water and accumulations of Fucaceae and other algae, that they specially occur in large quantities and form these associations which are rich in individuals and species.^ The sulphur-bacteria here, as in hot springs, deposit sulphur within their cells ^ ; for by the reciprocal action of the dead organic matter and water there is produced sulphuretted hydrogen, which is absorbed and oxidized by the bacteria, thus giving rise to water and sulphur. According to Sickenberger red sulphur-bacteria play an essential part in the production of soda in the Egyptian saHne lakes. Abyssal saprophytic associations. At great depths in seas and lakes where the water is untroubled, the light is often feeble, and the tempera- ture often low, there is frequently a collection of black mud which is filled with decaying bodies, alive with animals (thread-worms), but allows the growth of no highly organized autophytes. Probably there occurs here a rich saprophytic vegetation composed of species of Beggiatoa and other, perhaps mainly, anaerobic bacteria. But we know practically nothing of this vegetation. As an example of a place where there is probably a rich bacterial vegetation we may mention the Black Sea. According to Andrussow,* in this sea, at a depth of 100-600 fathoms and more, one encounters vast masses of mud containing sub-fossilized remains of mollusca, which belong to brackish water and arose in the not distant past when the Black Sea was a brackish lake, but which were exterminated when the Mediterranean Sea forced its way in. The currents are so constituted as to cause defective aeration of the deep water, which is therefore poor in oxygen though rich in sulphuretted hydrogen. It ' See Warming und Wesenberg-Lund, 1904; Warming, 1906. - Warming, 1875. ' As was first shown by Cohn. '' Andrussow, 1893. M 1 CHAP. XLii BENTHOS OF LOOSE SOIL 177 is assumed that not an animal lives in this, and that the organic consti- tuents of the mud are not consumed by animals ; but it is highly probable that a rich anaerobic vegetation of bacteria occurs here. In like manner in north European seas and fiords there are muddy spots where saprophytic vegetation may lurk. The black mud so extremely common in lakes, on sea-coasts, as well as in the depths of the sea, is usually very rich in iron sulphide. According to Beijerinck,^ and van Delden,- the reduction of sulphates in water on ferruginous soil is accom- pUshed by definite anaerobic bacteria, Microspira desulfuricans and M. Aestuarii.^ Also in north European fresh-water lakes in deep situations one encounters but scanty vegetation of an elaborate type ; here many lowly organized animals, including worms and larvae, thrive, and the eel waxes fat by preying on them. Here, we anticipate (but do not know) there will be found vegetation consisting of saprophytic Schizomycetes. Accord- ing to Forel *, in Lake Geneva at a depth of a hundred metres there is a brownish layer of lowly organized algae, mainly Schizophyceae and Diatomaceae, which thus form an organized carpet. ii. Enhalid-formations. To this formation belong all communities of Spermophyta, and larger algae growing on loose soil in salt water. FLORA Of Algae there are very few : in tropical seas species of Caulerpa^ and Penicillus, and in European waters (especially if these be brackish) Characeae, all of which send capillary root-hke organs into the soil, (The algae which casually occur here are attached to stones, and belong to the lithophilous formation.) Spermophyta, all of which are herbaceous, preponderate in number and dimensions, though the number of species present is small (twenty- seven). They belong to two families : Potamogetonaceae, with Zostera, Phyllospadix, Posidonia, Cymodocea, Halodule, Althenia, also in brackish water Ruppia and Zannichelha ; and Hydrocharidaceae, with Halophila, Enhalus, and Thalassia. Divers epiphytic Algae occur. ADAPTATIONS The grass-wracks, though belonging to two different famihes, are externally so alike that mistakes have often been made in the identifica- tion of flowerless specimens. The typical form is well illustrated by Zostera ; like this, all the represen- tatives are submerged ; true floating-leaves are wanting, probably because of the violence of the waves ; the leaves, except in Halophila, are ribbon- hke, rounded or blunt at the tip, and entire. The ribbon-like or zostcroid form of leaf is well fitted for existence, not only in currents, but also in water that is deep and therefore ill-lighted ; accordingly, it reappears in similar > Beijerinck, 1895. * van Delden, 1903. * See also Warming, 1904. * Forel, 1891. ' Svedehus, 1906; Borgcsen, 1907. WARMING N 178 HYDROPHYTES sect, iv circumstances in species belonging to the limnaea-formation. The breadth of the ribbon-hke leaf of Zostera marina is obviously adjusted to the depth of water ; the shallower this is the narrower the leaf (variety, angusti- folia) ; whereas in deeper water the plant becomes more vigorous and broader-leaved. The narrow-leaved species, Zostera nana, also species of Ruppia and Zannichellia, occur in shallow water not far from the shore. In consequence of the far-stretching rhizomes a social mode of growth results, so that dense, grass-green, submarine ' meadows ', often extending for miles, are formed. The ftowers are very reduced, and inconspicuous ; ^ flowering takes place on or under the water, and with its aid ; the pollen grains in species with submerged flowers are filiform, as in Zostera and Cymodocea, or linked together in long threads, as in Halophila,^ obviously in order that they may be more easily captured by the long stigma, when conveyed thither by the water, with which they agree in specific gravity. The stalks of the female flowers of some species, including Enhalus and Ruppia spirahs, are long and spirally coiled, and they contract after pollination. The pollination of Enhalus is dependent upon the ebb-tide.^ GEOGRAPHICAL DISTRIBUTION In arctic seas this type of vegetation seems to be almost wanting, possibly because the ice will not permit of its development. Several associations of grass-wracks and the like may be distinguished.* In north-temperate seas of Europe, Zostera marina and Z. nana occur, while in the Mediterranean added to these are Cymodocea nodosa and Posidonia oceanica. Zostera marina forms a shallow zone along coasts ; in Danish seas its lower limit is eleven metres ^ ; the depth attained necessarily depends upon intensity of the light, and thus upon clearness of the water. It demands a soil that is to some extent protected. Other plants also are partly confined to submarine meadows of this kind ; among such are certain algae, some of which are epiphyllous, for example, diatoms, Desmotrichum undulatum, species of Ectocarpus, Ceramium, Polysi- phonia, often in great masses, while others, including Fucus, Laminaria, Cladophora gracilis, and Fastigiaria furcellata,^ grow among the rhizomes. This is the case particularly where the soil includes stones. In addition there occur a series of more or less modified forms of various species of algae, which have been brought hither by currents, and are detained among the Zostera plants, where, remaining in a sterile condition, they undergo more or less modification in their subsequent growth ; such species are Ascophyllum nodosum var. scorpioides, Phyllophora Bangii which is a metamorphosed form of P. rubens, also forms of Phyllophora Brodiaei, Anfeltia phcata, Cladostephus verticillata, and others. In the shallows of the sea off Schleswig, Cyanophyceae, as a microphyte-forma- tion, also mingle with Zostera.' In Danish waters and the western Baltic, close to land there is usually a shore-zone of Ruppia and Zostera nana, while in deeper water there is one of Zostera marina. ^ Schenck, 18866. ' Balfour, 1878; Theo. Holm, 1885. ' Svedelius, 1904. * Ascherson, 1871. ' Ostenfeld, 1905, 1908 a. ' Communicated by Rosen vinge. ' Warming, 1904, 1906. CHAP. XLii BENTHOS OF LOOSE SOIL 179 In lagoons on the coasts of the Danish West Indies, according to Borgesen,^ there are species of Thalassia, Cymodocea, Halophila, and Halodule, creeping algae, such as Caulerpa, and non-rhizomatous Algae, such as Penicillus, Udotea, and Halimeda. The vegetation of brackish water on many coasts is closely allied to that just described, but it also includes other more slender species and other genera, such as Chara, Zannichellia, Batrachium, Naias, Potamogeton pectinatus, and Myriophyllum, which reappear more bountifully in fresh water. Several of the species mentioned grow only in shallow water, down to a depth of two metres at most. Grass-wracks play an important part in the biology of the sea as homes for marine animals, in connexion with oviposition by fish, and as food- material in the case of Thalassia testudinum, which is eaten by turtles. Many sahne waters near salt-works have a peculiar algal vegetation often intermingled with Ruppia and Zannichellia. iii. Limnaea-formations. To these formations belong aU those fresh-water communities of auto- phytic spermophyta, and other plants of considerable size, the individuals of which live on loose soil, whether sand, clay, or mud, and are completely submerged, or at most possess floating leaves. (Their flowers, however, in nearly all cases rise out of the water). They are therein distinguished from marsh-plants, whose assimilatory organs for the most part project above water. But there is no sharp hmit between these formations. I ' FLORA The flora is composed of — ' Green Algae, namely Characeae, which occur particularly on marl ! soil, and clothe this with a dense, peculiar-smelling carpet (characetum). Musci, including Fontinahs, Hypnum, and Sphagnum. Pteridophyta, including Marsiha and Pilularia among Hydropterideae, IS well as Isoetes. j Spermophyta, including more numerous species of Potamogetonaceae ' than occur in the sea, of Potamogeton for instance, also Elodea, Vallis- neria and Hydrilla representing the Hydrocharidaceae, Sparganium minimum, S. affine, and other species, in addition to many Dicotyledones such as Nymphaeceae, Cabombaceae, most species of Batrachium, Myriophyllum, Helosciadum, Callitriche, Subularia, Elatine, Montia, Limosella, and others. Epiphytes, among which are many Diatomaceae and Cyanophyceae which are often enveloped in mucilage. ADAPTATIONS The diversity of form among Spermophyta, in contrast to that in I the enhalid-formation, is extreme. This is to be attributed to the great variety of environment, which includes not only powerfully streaming water but very often very calm water, sucli as is never met with in the ' Borgesen, 1900. N 2 i8o HYDROPHYTES sect, iv > sea. The chief distinction in shape concerns the occurrence not only of completely submerged types, but also of species with floating leaves, or with shoots floating on the surface. All the species are herbaceous, and nearly all are perennial. The Shoot. The construction of the shoot varies widely. In agree- ment with the loose nature of the soil, the vast majority possess creeping axes, and therefore display social growth : such is the case with Pota- mogeton, Hippuris, Nymphaea, Nuphar, all of which have subterranean horizontal stems ; and Myriophyllum, Ranunculus, Callitriche, which have epigeous creeping stems ; the Characeae also belong here mutatis mutandis. Others, such as Littorella and Vallisneria, emit long runners, and on these, at certain distances from the mother-plant, they produce rosette-shoots which become firmly rooted. All such species can give rise to extensive and dense associations which clothe the lake-bed, being rich in individuals though poor in species. A minority of species, including Isoetes and Lobeha, have also vertical rhizomes with the leaves separated by short internodes and ranged in a rosette : but these are devoid of the above-mentioned methods of migration, and are rather represented by isolated individuals. Finally, there is a small number of annual species, such as Subularia, Naias, and Trapa, which grow socially only when numerous seeds are strewn over the same ground. There are three essentially different forms of assimilatory shoots,, namely : — A. Rosette-type. The shoots are vertical, short, unbranched, with short internodes ; the leaves are in rosettes, sessile, and mostly submerged. Such is the case in Vallisneria with ribbon-like leaves, Isoetes, Lobeha Dortmanna, Subularia, and Littorella unifiora with more terete leaves. B. Nymphaea-type. The shoots assume the form described for the rosette-type, or are horizontal rhizomes creeping in the soil ; there are, however, long-stalked floating leaves. This type is represented by Nymphaeaceae. C. Long-stemmed type. From a rhizome or a stem creeping over the ground there arise under water erect stems that have long internodes, and are slender and branched ; the main and lateral axes, as a rule, are of equal thickness, showing no secondary thickening, as is precisely the case with certain plants described in Chapter XL. The stems, which are often very long and slender, are extremely flexible and can yield to movements of the water. Their length depends upon the depth and flow of the water. Terrestrial forms of the same species have shorter inter- nodes. These shoots are of two kinds : a. Completely submerged, as in Potamogeton pectinatus, P. perfoliatus, some species of Batrachium, Myriophyllum, Zannichelha, Callitriche autum- naUs, Elodea and Naias (annual). The leaves are linear or oblong (only rarely broad), and in some are very finely segmented. b. Possessed of floating leaves in addition to submerged leaves, the former leaves sometimes being arranged more or less in a rosette at the end of a condensed shoot, and then sometimes having tolerably short petioles. As examples may be cited Callitriche verna, Trapa (annual), the majority of species of Batrachium, Potamogeton natans, and EUsma natans. CHAP..XLII BENTHOS OF LOOSE SOIL i8i The Leaf. The dependence of the leaf-shape (and partly shoot-form) on the medium is particularly striking. The following are the five types of leaf-shape : — 1. The floating. 2. The submerged. a. The zosteroid. h. The elodioid. c. The isoetoid. d. The myriophylloid. The submerged types fall into two groups : that including very finely segmented leaves occurring mainly in Dicotyledones (type 2 d), and that including essentially long and linear leaves (types 2a, 2b, 2c)} I. The floating leaf has already been mentioned as occurring in the hydrocharid-formation."^ It is found especially in associations thriving in calm inlets or under shelter of reed-swamps. This type is possessed by Nymphaea, Nuphar, Cabomba, Brasenia, Limnanthemum, Hydrocleys, Elisma, Batrachium, Trapa, Calhtriche, species of Potamogeton (P. natans), Polygonum amphibium, and other genera with the same general form. The leaf is broad (orbicular, ovate, cordate, reniform, rhombic or elliptic ; rarely lanceolate), undivided and entire, rarely crenate or incised (as in Trapa and Batrachium), comparatively thick and tough (coriaceous), sometimes possessed of a mechanically strengthened or an upwardly bent margin, and excellently adapted to rest on water and resist movements of the latter ; the gigantic floating leaves of Victoria regia, Euryale ferox, and others are, in addition, strengthened by stout ribs on the under-surface. The floating leaf is necessarily adapted to transpire, and thus provides a transition to the land-plant. Stomata occur solely or mainly on the upper face, whose epidermis contains no chlorophyll ; they are protected from being plugged with water by the deposition of wax in or on the cuticle, which is thus rendered unwettable. This it is which gives often a glossy or whitish appearance to the upper face.^ The lamina of the floating leaf is dorsi-ventral, showing paHsade tissue towards the upper face, and very lacunar spongy parenchyma towards the lower face. The lower face is often coloured dark red by erythophyll, the significance of which is not yet known. Prickles on the lower face of the lamina and on the stalk are shown by Victoria and Euryale. The petiole of the floating leaf has the power of adjusting itself according to the depth of water, in such a way that its growth ceases when the lamina comes in contact with the atmosphere. In the case of shoots with long internodes, the latter are hkewise similarly arrested in growth, as in Trapa and Callitriche ; in such cases, the proportions as regards length between the petioles and insertion of the floating leaves is such that all the blades are accommodated with space on the water. Frank suggested that growth of the petiole is promoted by the pressure of the overlying column of water ; other investigations have shown that contact with the atmosphere and more intense illumination are respon- sible for the shaping of the floating leaf."* ' For the literature, see Schcnck, i886fc. ' See Chapter XL. ' See Jahn, 1886. * Frank and others; see Hterature in Schenck, 18866. i82 HYDROPHYTES sect, iv Various species are heterophyllous, possessing not only floating leaves but also submerged leaves. According to Askenasy ^ and others, the floating leaves of Batrachium and Cabomba do not appear until the plant is about to flower, so that they may serve specially to raise the blossom above water. 2. The submerged leaf differs both morphologically and anatomically (particularly as regards epidermis and chlorenchyma) from the floating leaf : ^ a. The zoster oid, or ribbon-like leaf, which generally occurs among grass-wracks, is less frequent in this formation, though it occurs in ValHs- neria, Sparganium, species of Potamogeton, and others. Convincing evidence is forthcoming that this form of leaf is adapted to and evoked by deep or running water (both these conditions appear to act in the same direction), when we observe that it appears on certain marsh-plants, including Alisma Plantago, Sagittaria sagittifolia, Echinodorus ranuncu- loides, and species of Sparaganium, if these be compelled to grow in such water. Similar forms of leaf are met with under the same conditions in Scirpus lacustris and Potamogeton natans, which produces ' current- leaves ' half a metre long.^ b. The elodioid leaf, which is narrow, linear, undivided, flat, sessile, and short, is frequent, as shown by Elodea, Potamogeton densus, P. obtusif olius, P. pusillus and other species, Hippuris, Zannichellia, Callitriche autumnalis and other species, and Naias. In this category may be placed the leaves of the aquatic mosses. Broader forms of leaf are displayed by other species of Potamogeton. c. The isoetoid leaf is linear, undivided, terete, often tubular, and sessile, and occurs in Pilularia, Isoetes, Lobelia Dortmanna, Littorella lacustris, and others, most of which are rosette-plants. Subularia and the Characeae may most fittingly be appended to these. It becomes clear that the two, tolerably similar, hnear types of leaves just described result, at least partially, from the action of the water, when we observe the behaviour of Juncus supinus, Hippuris vulgaris, Elatine Alsinastrum, Isoetes lacustris, Pilularia, and other plants that assume terrestrial and aquatic forms ; the submerged leaves are much longer, more flaccid than the subaerial leaves. d. The myriophylloid leaf, or leaf dissected into filiform or linear segments (like the gills of a fish) is very widespread, occurring in Myriophyllum, Heliosciadium inundatum, Batrachium, and Cabomba, as well as in certain marsh-plants, including Oenanthe Phellandrium, O. fistulosa, and Slum latifolium, when these grow in deeper water. Allied to this type of leaf is the uncommon fenestrated leaf of Ouvirandra fenestrahs. Many observations show that the depth of the incisions, and the fineness of the segments are due to the influence of the medium (depth of water, strength of its flow, and so forth) : when the shoots reach the surface there appear floating leaves, as in Batrachium, or leaves with shorter, broader, thicker segments, especially when the shoots project out of the water as in the case of Myriophyllum. The physiological cause of this difference must presumably lie in the elongation due to decreased illumination, and in J Askenasy, 1870. " In reference to the different forms of Polygonum amphibium, see Massart, 1902. ' See Costantin, 1884, 1885, 1886; Gobel, 1889; Raunkiar, 1895-9; Gluck, 1905, 1906. CHAP. XLii BENTHOS OF LOOSE SOIL 183 the exclusion of transpiration. Finely divided leaves are well suited to the medium, because their surface is thus relatively increased, and consequently the absorption of food-material and of oxygen is facihtated. The movements of the water would hardly permit of larger surface. Reproduction. The reproduction of the Cryptogamia takes place under water. But nearly all the Spermophyta thrust their flowers above water ; some, including Hottonia and Nymphaeaceae, are entomo- philous, but others, represented by Hippuris, Myriophyllum and Pota- mogeton, are polhnated by the aid of wind or water, or adopt self-pollina- tion. The pollen is conveyed by the water in Zannichelha, Callitriche, and Naias ; while semi-cleistogamy is enacted under water by Subularia aquatica, Limosella aquatica, Euryale ferox, Ehsma natans, and rarely by Batrachium. Pecuhar behaviour, paralleled by that of Ruppia spiralis, is displayed by ValHsneria, whose small male flowers break loose, swim on the surface and polhnate the stigmas of the female flowers which rest on the water ; nearest in this respect to Valhsneria stands Elodea.^ After polHnation, the developing fruits of many species, including Trapa and Ranunculus, are dragged or curled down under water where they ripen. Dispersal of the seed is often accomplished by special means which are suited to the medium : the seeds or fruits of many species, owing to their peculiar structure, are lighter than the water, and are thus conveyed to other habitats. - Vegetative Propagation. Propagation by vegetative means is very widespread among all aquatic plants, and very easily takes place by mere separation of fragments of the shoot ; it is of profound biological signifi- cance, and some species have almost ceased to reproduce sexually.^ Calla palustris has special buds that easily become detached.^ The rapid spread of Elodea and the inconceivable number of its individuals in Europe result from vegetative multiphcation, as it produces no seed because the female plant rarely occurs in Europe. The great power of vegetative multiplication is to be ascribed to the production of propagative buds, to branching, and to the facile manufacture of adventitious roots. Hibernation.! Most of the species hibernate as green plants at the bottom of the water, where thermal conditions are not so extreme as in air ; such is the case with Callitriche, Zannichelha, Nymphaeaceae, Valhsneria, and others. Hibernating organs of a special kind, which become detached from the decaying shoot, are represented by the cartila- ginous winter-shoots (' hibernacula ') of Potamogeton crispus and other species,'* the spherical buds of Myriophyllum and species of Utricularia which include densely packed leaves, also by the gemmae (buds) of Hydrilla and Elodea. FORMATIONS, AND DISTRIBUTION A great number of associations occur in the limnaea-formation, and some of these, hke those composed of Cryptogamia, might perhaps be regarded as being themselves complete formations or sub-formations. The common zonally arranged associations are evidently identical over wide areas in temperate countries. In the deepest parts of some lakes in Europe there occurs a zone of ' Schenck, 18866. = Ravn, 1894. ' J. Erikson, 1895. ' Sauvageau. 1888-94; Raunkiar, 1895. i84 HYDROPHYTES ground-algae, which may include vast numbers of Cladophoraceae, with which Diatomaceae and Fontinahs antipyretica may be mingled.^ Brand ^ found a zone of this kind at a depth of 20 metres. Generally, characeta form the deepest zone of macrophytes ; they are ranged in dense sub- lacustral ' meadows ', in which no plants, other than possibly mosses, may find place. In Lake Geneva they descend to 20-25 metres, according to Forel,^ while in Lake Constance they go down to 30 metres according to Schroter and Kirchner,* but the most frequent depth is 8-12 metres. The association next in depth is that of Elodea, which descends to 6 metres ; then succeed submerged species of Potamogeton, including P. lucens, at depths as great as 4-6 metres, and with them may be Myriophyllum. In the next higher zone appear plants with floating leaves, 7iymphaeeta and nuphareta down to 3-5 metres, and hatrachieta to 2-3 metres. As a rule, Spermophyta seem to cease at a depth of 10 metres.^ In North America ^ there are zones completely corresponding to these European ones. By the shores of shallow waters in Europe, and especially where the soil is sandy, there occurs a peculiar association composed of the limnaea rosette-forms, Lobelia, Littorella, Isoetes, and Subularia, which are charac- terized by shortness of stems and by rosulate leaves that belong to type 26 described on p. 182. The distribution of the associations is determined by — a. The conditions of depth, light, and clearness of water. Some species can descend to much greater depths than others. The limnaea-com- munities in larger lakes are confined to a zone of slight depth which borders the shore and entertains an abundant fauna. h. Distinctions in soil. Some species prefer sand, and others mud. The Characeae are calciphilous, yet they can be found in pools on heaths and moors where the water contains but little lime. c. Movements of the water. Some species, and particularly those with floating leaves, live only in placid water. Limnaea- vegetation is allied to hydrocharid-vegetation. The boundary between the two is not sharp ; they are often intermingled, and in both there occur genera that are the same, although represented by different species. Certain plants, including Eichhornia crassipes, Stratiotes aloides, Hydrocharis, and Pistia, that usually are free-floating, may on occasion become attached by roots ; and conversely other species, of Ceratopteris ' for example, that are normally fixed and rooted, may become free-floating. Plankton necessarily must occur in limnaea- formation : the two give rise to a combined formation. There is, of course, no sharp distinction between aquatic plants that are fixed by their roots and marsh-plants ; among them are many ' amphibious ' species, including Polygonum amphibium, that can assume an aquatic or a terrestrial form. Plants such as Montia rivularis living in springs are in a sense transitional between land-plants and water-plants ; they seek water that flows rapidly and is rich in oxygen and carbonic acid.^ * (Schroter und) Kirchner, 1896; G. Huber, 1905. ^ Brand, 1896. ^ Forel, 1 891. * Schroter und Kirchner, 1896- 1902. * Magnin, 1893, 1894. * According to Coulter, Cowles, Transeau, and Pieters. ' Gobel, 1889-92,11. * In addition to the hterature already cited, attention should be directed to papers by Chatin, 1856; Friih und Schroter, 1904; Pond, 1905; Gliick, 1905-6; Gobel, 1908. SECTION V CLASS II. HELOPHYTES. MARSH-PLANTS CHAPTER XLIII. ADAPTATIONS. FORMATIONS x'Vmong aquatic plants are included all plants whose assimilatory organs are submerged, or, at most, swim on the surface of water ^ ; marsh- plants or helophytes include all those which normally have their roots under water or in soaking soil, but, like land-plants, raise their foliage branches above the water-surface. It has already been pointed out ^ that there is no sharp hmit between marsh-plants and land-plants. Moreover, many marsh-plants are more or less ' amphibious ' and plastic, so that they can change their structure according as they are submerged or not.^ Marsh-plants and bog-plants are confined to shallow, still, or gently flowing water, and to soil that contains a large amount of water (appa- rently more than eighty per cent.) at least for a prolonged period. The soil is loose, often very loose and soft, also usually rich in humus in the form of peat or mud."* ADAPTATIONS 1. Marsh-plants, Hke aquatic plants, are largely perennials.^ But in marshes that are completely dried up during the dry season annual species may prevail. 2. Many marsh-plants readily produce adventitious roots and possess horizontal rhizomes or runners ; in Europe such is the case with Equise- tum limosum, Iris Pseudacorus, Phragmites, Typha, Acorus, Butomus, Scirpus lacustris and S. (Heleocharis) palustris, Eriophorum angustifohum and E. alpinum, Sparganium, Carex acutiformis, C. rostrata, and other >pecies, Cladium Mariscus, and other Monocotyledones, Lysimachia \ulgaris and L. thyrsiflora, Ranunculus Lingua, Sium latifolium (with roots that produce buds) and S. angustifohum. Caespitose plants with a small power of vegetative migration or devoid of such a power are exemplified by Ly thrum Salicaria, Cicuta virosa, Alisma Plantago, and Rumex Hydrolapathum. Plants of this form often grow partially on their own dead fragments by which they are jirradually lifted upwards ; one obvious reason for this is that water is raised by capillarity in the sponge-like tufts formed by the interwoven remnants of stems, leaves, and roots ; such is the case with Eriophorum vaginatum, Carex stricta, C. paniculata, and many other species. In addition there occur plants showing other modes of growth ; for ' See Chap. XXVII. ' See p. 131. ^ Costantin, 1897; Schenck, 1884, 1886; Massart, 1902. ' See p. 61. ' See p. 100. i86 HELOPHYTES sect, v example, those living on Sphagnum must necessarily have the power of raising themselves as their substratum grows. ^ 3. As in aquatic plants, internal air-containing spaces develop in stems, leaves, and roots, and are adapted to meet the scarcity of air in wet soil, whose air-content is sometimes further decreased by special conditions, such as the accumulation of organic remains, production of peat,^ the interweaving of roots, and other means by which there is formed a covering that shuts off the supply of air. In order to secure adequate aeration for the plant the following special devices exist : — (a) Aerenchyrna^ : a tissue which, like cork, has its own phellogen, but consists of thin-walled, non-suberized cells, and includes large, air- containing, intercellular spaces. The tissue shows itself as a white, spongy envelope. It occurs in Epilobium hirsutum and other species, Lythrum__Salicariaj__Lycopus europaeus, the mimosaceous Neptunia oleracea, andTotliers. {h) Respiratory roots (pneumatophores). In some trees and shrubs there are developed erect roots which thrust their tips above water and convey air to the intercellular system of parts in the mud by means of their pneumathodes, that is to say, by means of lenticels or other com- munications with the atmosphere.^ They occur particularly in mangrove- swamps ; also on certain palms, Taxodium distichum,^ and possibly on Jussieuea.^ 4. Marsh-plants commonly have mesophytic leaves ; but in a remark- able number of them xerophytic structure is encountered.' 5. The seeds and fruit of many marsh-plants are provided with air- containing spaces and other devices that facilitate their dispersal by water.^ FORMATIONS AND ASSOCIATIONS Qualities of the soil play a great part in evoking floristic, and, more or less, oecological distinctions in the communities : Saline swamps are not only floristically, but also anatomically and morphologically so pecuHar, that they differ widely from fresh-water | swamps, and they will be considered in Section VII, deahng with halo- philous vegetation. Fresh-water swamps, which alone will be dealt with here, display many differences according as the water is troubled or not, according as the soil \ is muddy, sandy, or gravelly, and so forth. The constituent formations have been but Uttle investigated. These are probably several, but certainly there are two : Reed-swamp and Bush-swamp. There are also communities on the ' boundary ' zone of wet land which hve an amphibious life, and are hydrophytic in adaptation, and for these Schroter and Kirchner set up a third formation consisting of amphiphytes. They write ^ : ' Every point on the boundary zone is annually inundated for a shorter or longer period, which is longer the nearer it stands to the lake. . . . Thus this zone represents a very gradual transition from terrestrial ' See P. E. Miiller, 1894. ^ See Chap. XVI. ' Schenck, 18896; Gobel, 1889-92, ii. ' Gobel, 1886; Wilson, loc. cit. ; Schenck, 1889a; Schimper, 1891 ; G. Karsten, 1 80 1. ^ Kearney, 1901. * Gobel, 1889. ' See Chap. XLVI. * Guppy, 1891-3 ; Ravn, 1894. ' Schroter (und Kirchner), 1902, p. 42. k CHAP. XLiii ADAPTATIONS. FORMATIONS 187 to lacustral conditions. It therefore exhibits a zonal arrangement of its occupants according to the degree of adaptation of these to the lacus- tral conditions. . . . The geographically defined " boundary zone " must be classified biologically into three main subdivisions : {a) Meadow- swamp, lying nearest the dry land and inundated only for a short time; {b) zone that is being converted into land . . . : (c) strips of gravel or sand poor in vegetation.' Only the last two concern us here. In these there grow land-forms of plants derived from the lake-flora, also typical occu- pants of the ' boundary zone ', as well as plants that have advanced to this point though belonging to swamp meadows and ditches. The above- named botanists distinguish, in connexion with Lake Constance, two associations : a heleocharetum (with Heleocharis acicularis, Littorella, Ranunculus reptans, Myosotis palustris var. caespititia, Agrostis alba, and others) and a polygonetum. They furthermore distinguish a fourth formation, namely that of alluvial plants, including a tamaricetum- association (with Tamarix germanica and Hippophae and others), which occupies the ' boundary zone ' where this takes the form of a gravelly or sandy shore, and including also plants from the riparian alluvia as well as alpine plants. Swamp vegetation is also allied to that which is described by Drude under the heading of Formations near springs and brooks, and includes tall herbs hke Ulmaria, Geranium palustre, Impatiens noU-me-tangere, Equisetum Telmateia, and small herbs, mosses, and algae. As these last formations have been oecologically investigated only to a slight extent, we shall here consider in detail only — 1. Reed-swamp Formation. 2. Bush-swamp Formation. CHAPTER XLIV. REED-SWAMP OR REED-FORMATION Reed-swamps occur either in fresh flowing water, or in stagnant, more or less acid, water. Many species, such as Phragmites communis, seem to be not in the least exacting as regards choice of soil. The swamps of Arundinaria macrosperma are on acid peat soil, and the vegetation thus provides a transition to oxylophytes. The vegetation is mainly composed of tall monocotylous perennial herbs, grows in more or less deep, usually standing, water, and seems to be most nearly alHed to the communities of fresh-water plants ; between the individual shoots and leaves one sees everywhere clear water, which entertains representatives of the plankton-, hydrocharid-, and e\-en limnaea-formations. ASSOCIATIONS IN TEMPERATE ZONES Among the various genera and species to be found are Phragmites communis, Scirpus lacustris, Typha, Butomus umbcllatus, Glyceria spectabilis, and other species, Phalaris arundinacea, Iris Pseudacorus, Cladium Mariscus, Carex paniculata, C. rostrata, C. gracihs, C. fihformis, C. acutiformis, C. stricta, C. riparia, C. vesicaria, and other species, i88 HELOPHYTES sect, v Alisma Plantago, Sagittaria, Sparganium ramosum, S. simplex, Acorus Calamus, and Calla palustris, which together represent the most important monocotylous representatives of this oecological class in temperate Europe ; to these may be added Equisetum hmosum, and, among Dicoty- ledones, Senecio paludosus, Sonchus palustris, Menyanthes trifoliata, Lythrum Sahcaria, Epilobium hirsutum, Rumex Hydrolapathum, Lysima- chia vulgaris, L. thyrsiflora, Ranunculus Lingua, Oenanthe fistulosa, O. aquatica. Slum latifohum, S. angustifohum, Cicuta virosa, and many others. Many diatoms and Green Algae occur as epiphytes. Especially on the banks of water-stretches is found this vegetation, which advances as a pioneer of land-vegetation, acts as a check to wave violence, and adds to the land.^ According to depth of water and other conditions dependent thereon (light, temperature, and water-movement) this formation is hkewise arranged in zones, which in Denmark and over the greater part of Europe are identical, and may be pure associations, such as phragmiteta, scirpeta, and so forth. Magnin '^ and Schroter ^ have observed, in Lakes Jura and Constance respectively, the following zones : scirpeta (Scirpus lacustris down to 3-5 metres in Lake Constance), phragmiteta (Phrag- mites communis down to 2 metres), after which succeed heleochareta (Heleocharis palustris), and cariceta (in Denmark, C. aquatilis, C. rostrata, and others). All these plant-communities work in the direction of filhng up collections of water and draining them dry. There also occur typheta, equiseteta, and others. Precisely the same zones present themselves in North Europe and North America. According to Transeau,* in the Lakes of Michigan, after the aquatic societies containing Potamogeton and Nymphaea, there succeeds the ' cat-tail-DuHchium association ' with Typha, Phragmites, and Dulichium. Further inland follow the ' Cassandra society ', ' shrub- and-young-tree ' association, and forest. Cowles ^ finds in the vicinity of Chicago the following zones : (i) Chara ; (2) Nymphaea ; (3) Carex and Scirpus ; (4) Cassandra calyculata and other shrubs ; (5) Forest. In other places the cariceta are succeeded by grass meadows.^ Adaptations. Vigorous creeping stems fasten the plants to the loose soil, causing social growth and dense pure associations of Phragmites, Scirpus lacustris, Equisetum limosum, Typha, and many others; also, on the Nile, Cyperus Papyrus. The production of shoots by roots, which is so frequent in dry spots, rarely occurs in the vegetation of reed- swamps, but is exhibited by Sium latifolium. Caespitose species are likewise rare. The vegetative shoots vary in construction, but mainly belong to one of three types : — (a) Leafless, bare stem consisting of one internode, which may be either one or two metres in length, and is capped by the inflorescence ; as examples, we may mention Scirpus lacustris, S. Tabernaemontani, which are as much as 1-2 metres in height, also Heleocharis palustris and species of J uncus which are shorter. ' See Section XVI. " Magnin, 1893, 1894. ^ Schroter und Kirchner, 1896-1902. * Transeau, 1903, 1905. * Cowles, 1901. * See also Pieters, 1894, 1901 ; Hitchcock, 1898 ; V. Borbas (Bernatsky), 1907 ; Friih und Schroter, 1904. II I CHAP. XLiv REED-SWAMP OR REED-FORMATION i8q I {b) In addition to long linear leaves rising from a rhizome or radiating from the base of the flowering axis, there are tall culms bearing inflo- I rescences, as may be seen in Typha, Acorus, Butomus, and others. (c) Tall haulms with long distichous, spreading leaves, as in Phragmites and other grasses. I The character coinmon to all is that the dominant, mainly monocotylous, plants which give the stamp to the vegetation, are tall, slender, upright, and unhranched. ^ven in Ranunculus Lingua and in Rumex Hydrola- pathum there is a recurrence of the same habit which thus suggests an adaptation of obscure significance. It may, however, be pointed out that these tall slender shoots easily bend to breeze or current, and elasti- cally recover ; this is specially true of the unbranched stems of plants such as Scirpus palustris, or the tall, long leaves projecting above the water from tlie stems of Typha and Sparganium. Nearly all the species are perennial herbs, or, like Ranunculus scelera- tus, biennials. Special hibernating and propagative organs are produced by Sagittaria, in the form of stem-tubers on runners. An occasional woody plant, such as Sahx cinerea or Alnus glutinosa, may also occur. x\mong the plants of reed-swamps many are almost devoid of protection against desiccation — they are mesophytes ; but others are xeromorphic, for one finds among them leaves with a profile-he in Iris, rolled leaves in Cladium, juncoid shoots in J uncus and Scirpus, dense hairiness in Epilobium hirsutum, and so forth. Aerenchyma is very widely represented in plants of the reed-swamp. Geographical Distribution. Associations of this same stamp are ubiquitous on the Earth. Phragmites, the reed, is extraordinarily wide- spread ; over an area of many square miles it forms impenetrable associa- tions (phragmiteta) in the Danube delta, in deltas of the Caspian Sea, and Lake Aral, and even in Austraha ; on the Syr-daria it attains a height of 6 metres and endures salt water quite well, while in Lusatia in Germany its variety pseudo-donax attains nearly lo metres.^ It can grow in water 3 metres in depth. In Mediterranean countries this reed forms communities, and is sometimes accompanied by Arundo Donax and Erianthus Ravennae, two grasses which are often even 5 and 6 metres in height. As an illustration of its capacity of adjusting itself to external conditions, we may mention that on the shores of the North Sea and of \\ many inland lakes it produces long epigeous runners.^ Extensive reed-swamps of Glyceria spectabihs on the saline soil of Neusiedler Lake assume the form of veritable ' grass-forests ' 2 metres in height ; the same is true of Cyperus syriacus in Sicily, and in an exaggerated degree of C. Papyrus on the Upper Nile, where this, together with Panicum pyramidale, Phragmites communis, Typha australis, and others, give rise to a ' sudd-formation ' ^ ; the shores of Valencia Lake in ■ Venezuela are fringed with dense reed-swamps of Typha domingensis, I which exceeds man's height, and the same is true of the cyperaceous Malacochaete Tatora on the shores of Lake Titicaca. j In Virginia, according to Kearney,^ there occur similar reed-swamps, ' According to Ascherson und Grabner 1898-9. * Illustrations in Warming, 1906 ; Schroter und Kirchner, 1902. * Broun, 1905. * Kearney, 1901. igo HELOPHYTES sect, v in which he distinguishes associations of Typha-Sagittaria along the rivers, and of Scirpus-Erianthus at the edge of the swamp-forest. Here also occurs Arundinaria macrosperma - association, which clothes large expanses of the Dismal Swamps. Along the rivers of Pennsylvania, according to Harshberger,^ there occur extensive reed-swamps, in which he distinguishes various associations, including those of Zizania, of Typha, of Sagittaria latifoHa, and of Ambrosia trifida. In swamps with slowly flowing water one finds other associations, including those of Symplocarpus (with Spathyema foetida, and species of Osmunda), of Iris-Typha-Acorus, and of Heracleum-Veratrum-Eupatorium. Every- where the physiognomy, and to some extent the genera, are the same as in Europe. ASSOCIATIONS IN TROPICAL ZONES Other species and families occupying similar habitats play the same part in nature within the tropics, but, as their associations include entirely different forms, they present a different physiognomy. They have been studied only to a slight extent. Many species of Araceae are swamp- plants — just as are Calla and Acorus in Europe — and usually have sagittate or cordate leaves ; dense associations, often several metres in height, are formed by them, for instance by Montrichardia arborescens in Trinidad and the adjacent parts of South America, also by Caladium.^ Among Scitamineae, species of Heliconia likewise occur in tropical America, and gigantic Amaryllidaceae, represented by Crinum, fringe the rivers of Guiana. These types of vegetation are never absolutely pure ; other, possibly many other, species mingle with those that have been mentioned as giving the tone to the vegetation. In the tropics larger numbers of woody plants occur and affect the appearance of reed-swamps. But there are also swamps with grasses and perennial herbs very like those of temperate countries, for example in Brazil.^ CHAPTER XLV. BUSH-SWAMP AND FOREST-SWAMP OF FRESH WATER In reed-swamp and low-moor there often occur some woody plants, but in other places these become so numerous as to create hush- land and forest (woodland-swamp, paludal forest, foHage-moor). In North Europe the first steps towards these are shown in collections of alders, birches, and willows growing on the banks bounding fresh water. Alneta may occur on mud on which only few plants can thrive as undergrowth ; these latter include ferns, mosses, Calla, Lythrum, Spiraea Ulmaria, Menyanthes, Carex, and others, which particularly grow on the drier spots at the foot of the alders. Intermingled with the alders there may be species of SaUx, Viburnum Opulus, and Rhamnus Frangula. In other spots Urtica dioica forms impenetrable thickets.^ Saliceta at other places in North Europe form the riparian bush- » Harshberger, 1904. ' Martius, 1840-7. ' Warming, 1892. * See Domin, 1904. CHAP. XLV FRESH-WATER BUSH- AND FOREST-SWAMP 191 lands, mainly composed of species of Salix, S. alba, S. fragilis, S. cinerea, and S. pentandra, between which there grow dicotylous perennial herbs, including Lysimachia vulgaris, Epilobium hirsutum, species of Valeriana, and Spiraea Ulmaria, as well as grasses such as Calamagrostis lanceolata cind Phragmites. The lianes in these bush-swamps are Solanum Dul- camara, Convolvulus sepium, and Humulus Lupulus. Betuleta and Pineta, according to Fleroff,^ occur on bog-lands in Russia. More extensive bush-swamp and forest-swamp occur in the southern part of the United States, where they give rise to large forests on wet, peaty soil. In Virginia there are two main associations — juniper-swamp and black-gum swamp — with several (subordinate) associations.'^ Juniper-swamp is formed of Chamaecyparis thyoides, and may be pure. The soil is a very acid peat, which even in summer is covered by water 30-60 centimetres deep. Black-gum swamp is composed of Nyssa biflora and Taxodium distichum. On the horizontal roots of the latter there arise conical roots which attain a height of a metre, are similar to those of Bruguiera in mangrove-swamps, and in like manner serve as respiratory organs. On the mud they form the solitary firm spots over which man can walk. Many pseud-epiphytes occur on the trees.^ In the water between the trees grow Azolla, Wolfhella, and others. The soil is acid, but not so peaty and dry as in juniper-swamp. The water as a rule forms a sheet jo-ioo centimetres deep lying above the soil. Nyssa and Taxodium are deciduous. And the same is mainly true of the subordinate species in Virginia. Farther south there appear a number of evergreen shrubs, and several short palms, including Sabal and Chamaerops; near the tropics Tillandsia usneoides and other epiphytes show themselves on the tree- crowns. Nearly all woody plants growing in the American forest- swamp are protected against rapid transpiration. Stomata in nearly all species occur exclusively on the lower face of the leaf, and are sunken m some species. In addition, the following features are present : coating of hairs or of wax, thick cuticle, and thick outer wall to epidermis, con- version of epidermal cells into mucilage, hypoderma, multiplication of the pahsade-layers. This strong development of measures guarding against desiccation is a consequence of acidity of soil, which abounds in organic remains. Bamboo-forest (bambusetum). Tropical bamboo-forest must apparently be regarded as one type of association belonging to swamp- forest. Tropical rivers are often fringed with bamboo-brake, which forms most impenetrable vegetation. Humboldt mentions that, along the river Magdalena, there are uninterrupted forests of bamboo and banana-leaved species of Heliconia. Nipetum. Under this heading is included the eastern Asiatic and .\ustralian vegetation composed of Nipa fruticans. This palm is all but stemless, yet it possesses immense pinnate leaves, which may be six metres in length, and its growth may be so dense that one requires an axe to cleave a way through the vegetation, in which other species, mcluding Chrysodium aureum, also occur. Nipetum lies on the land- ' Fleroff, 1907. * Kearney, 1901. ^ Thco. Holm, in letter. 192 HELOPHYTES ward side of mangrove-swamp, from whose species it derives subordinate constituents ; it also appears in connexion with lagoons and swamps, but mostly on less saline soil. According to J. Schmidt ^ it does not truly belong to the saline mangrove-swamp, next to which Schimper places it, but rather belongs to fresh-water swamps along rivers. In the tropics several other, but httle known, forms of forest-swamp and bush-swamp occur. A small fan-palm, a species of Bactris, clothes large swampy tracts along the river Caroni on the plains of the island of Trinidad. Another palm. Phoenix paludosa, is encountered in eastern Asiatic swamps. While, according to Kurz, in Burma there are swamp-forests that are leafless during the rainy season. Reorders ^ gives an interesting account of a forest-swamp in the interior of Sumatra, in which respiratory roots (in Calophyllum, Eugenia, and others), prop- roots, plank-roots, and remarkable besom-like aerial roots (i-i|- metres in length) occur. The physiological dryness and these peculiarities of construction are to be attributed to lack of oxygen in the soil. It is impossible to establish any sharp distinction between swamp- forests and forests on dry land ; this is shown by the semi-aquatic varieties ('facies') of the primaeval forests that line the Amazon, which are annually inundated, and are locally known as ' igapo '.^ Bush-swamp and forest-swamp occur not only in tropical and temperate, but also in arctic lands ; for instance, in Kanin peninsula, on the White Sea,^ Picea excelsa, Betula pubescens, and Pinus sylvestris, occur as dwarfed trees on insufficiently drained soil, and are accompanied by genuine marsh-herbs such as Eriophorum vaginatum. Again, in the rainy zone of the territory near the Magellan Straits the forests composed of evergreen species of Nothofagus (N. betuloides and others) are regarded by Dusen ^ as hydrophytic communities ; here, extraordinary humidity brings forth a swampy soil nearly wholly carpeted with mosses. In this, as in all other cases, sharp distinctions are lacking. But it must be insisted that in these bush-swamps the soil is always more or less sour (rich in humous acids), and that consequently this formation is allied to the one about to be described.^ ^ J. Schmidt, 1903. ^ Koorders, 1907. ' J. Huber, 1906. * Pohle, 1903. * Dusen, 1905. " W. G. Smith, 1903 ; N. Walker, 1905 ; see also, on the subject of this chapter. Coulter, 1903 ; Hitchcock, 1898 ; Kearney, 1901. I SECTION VI CLASS III. OXYLOPHYTES. FORMATIONS ON SOUR (ACID) SOIL CHAPTER XLVL XEROMORPHY. FORMATIONS On a soil that contains an abundance of free humous acids, and is more or less peat-Hke, there occurs a group of closely-related formations ; these rise in stature from humble communities of mosses and lichens, to dwarf-scrub and bushland, and finally to forest. All these communities share the character of choosing soil that is poor in nutriment and particularly in easily assimilable nitrogen — they are oligotrophic ; furthermore, they are not calciphilous. They often clothe sterile sand, on which they themselves soon produce raw humus, and usually give rise to dense, exclusive, and extensive communities. Sour soil is intimately associated with a moist, cold or temperate, chmate. These communities all exhibit xeromorphy, that is to say, they are protected from desiccation by certain devices, of which the most important are : — 1. Well-developed coating of hairs : Hairs on the lower face of the leaf form a felt in Ledum, Salix repens, S. lanata, and S. glauca, but are scale-like in Cassandra calyculata, and in the North American Nyssa uniflora, Persea pubescens, and Magnolia virginiana, which grow in swamps. 1 The essential function of the hairs may be to prevent water from occluding the stomata, but the hairs also depress transpiration. In this connexion it may be noted that Salix Myrsinites, growing in the swamps of Lapland, retains its faded leaves which serve to protect the year's shoot .- 2. Papillae project over the stomata of various Gramineae and Cyperaceae, such as Carex limosa, C. panicea, and C. rarifiora ; also of Lysimachia thyrsiflora, Polygonum amphibium.^ They also may prevent the stomata from being blocked by water. 3. Wax forms incrustations over the whole leaf, as in Vaccinium uliginosum, or only over the stomatiferous lower face, as in Andromeda polifoUa, Vaccinium Oxycoccos, Primula farinosa, Sahx groenlandica, Carex panicea, and in the North American swamp-plants,"* Acer rubrum, Persea pubescens, and others. 4. Thick cuticle is shown by various leaves, and by the stems of Scirpus caespitosus and others. ' Kearney, 1901 ; see p. 191. ' Kihlman, 1890. ' Volkens, 1890; Kihlman, 1890; Raunkiar, 1895-9, 1901. * Kearney, 1901. WARMING O 194 OXYLOPHYTES sect, vi 5. Sclerophylly is frequent and is due to thickness of the epidermal, wall, as in Andromeda polifoUa, Vaccinium Oxycoccos, V. Vitis-Idaea, and Ledum palustre, and is perhaps correlated with the perennial | character of the leaves concerned.^ 6. Mucilage occurs, for instance, in the epidermal cells of Berchemial scandens ^ ; and a continuous hypoderma is present beneath the epidermis] on the upper leaf-face of Pieris nitida. 7. Ericoid leaves. Many moor-plants have flat broad leaves ; but ' some species possess ericoid or filiform leaves whose stomata are enclosed in secluded spaces, so that water-vapour escapes with difficulty — such is the case in Erica Tetralix, Empetrum, Calluna vulgaris, and other species to be mentioned in the next group. 8. Terete leaves, aphyllous stems. The assimilatory organs of many moor-plants and marsh-plants assume the form of erect, terete leaves or aphyllous stems, as in Equisetum limosum, species of Junci genuini, and other species of Juncus to a less extent, Scirpus palustris, S. caespi- tosus, S. (Heleocharis) lacustris, and other species, Eriophorum vaginatum, Carex microglochin, C. dioica, C. chordorrhiza, and C. pauciflora. 9. Bilateral leaves. Leaves exposing their edges (profile-lie) are met with in Iris, Narthecium, Acorus, and Xyris. The leaves are fiat, broad, but likewise erect or upwardly directed, long, and undivided, in AUsma Plantago, Sagittaria, and other Ahsmaceae, Butomus, Typha, Sparganium, Ranunculus Lingua, and Lathyrus Nissolia. ID. Closure of leaves. Broad-leaved Cyperaceae can close their leaves together (always ?), and distinctly so in Carex Goodenowii ; yet the stomata are not confined to the upper face. This xeromorphy of plants growing on wet moor-soil occurs all the world over ; it is known not only in Europe,^ but also in America * and New Zealand.^ It is evident that there must be a causal connexion between the soil and the xeromorphic structure which has been described, but it would never be anticipated. The soil must be physiologically dry. In genera which include not only paludal species, but also mesophytic species not growing in dry places, we frequently find that the latter have the broadest leaves, although the contrary might be expected. The paludal Epilobium palustre and Lysimachia thyrsiflora are the narrowest-leaved species of their genera in our country ; Galium palustre ' and G. elongatum likewise have narrower leaves than the mesophilouS' species possess ; and other cases might be cited. Here attention may be directed to the remarkable fact that many species of heather-plants can grow both on extremely dry, warm soil, and on extremely cold, wet soil ; such is the case with Calluna, Empetrum, several species of Pinus, Juniperus communis, Betula nana, Saxifraga Hirculus, Ledum palustre, and Vaccinium Myrtillus in Europe,® Pinus Taeda in the Dismal Swamps of the United States,^ and Phormium tenaxj in New Zealand.' One would, therefore, assume that between the two, * For anatomy see H. E. Petersen, 1908. * Kearney, 1901. ' Volkens, 1890; Kihlman, 1890; Raunkiar, 1895-9,1901 ; concerning the British Isles, see Miall, 1898 ; Rob. Smith, 1899. ' Kearney, 1901 ; Pound and Clements, 1900. ° Cockayne, 1901, 1904, 19050. * See Grabner, 1895, 1901. ' Cockayne, 1904. J CHAP. XLVi XEROMORPHY. FORMATIONS 195 kinds of soil there is some essential agreement, and that some of the life-conditions under which marsh-plants exist compel them to deal economically with water. Several different factors may be of moment and co-operate : for instance — Johow ^ and Kihlman - directed attention to the observation made by Tschaplowitz ^ indicating the existence of a transpiration optimum, ;md that therefore even marsh-plants may be forced to depress their transpiration. Wet soil is cold, and therefore physiologically dry.^ Consequently, on moors and swamps vegetation develops late, and flowering is late, except in certain species. Kihlman^ and Gobel ^ point out that many plants, though growing in soaking spots, are clothed with woolly hairs (as is the case with species of Espeletia in Venezuela) or are protected from rapid transpiration in some other way, because strong winds dry the vegetation at a time when the activity of the root is checked by coldness of soil. This well accounts for the xerophytic structure of plants in the extreme north and high up on mountains, and it plays an important part in the places under discussion. Another circumstance of potential significance is that in every wet, badly-aerated soil, which is poor in oxygen, respiration is obstructed, and consequently the root's functional activity is depressed. '^ According to Freyberg, roots of marsh-plants consume less oxygen than do land- plants in a given time, and in order to maintain the balance between their working and that of epigeous parts the functional activity of the latter must be depressed. The fact that many plants, such as Calluna and species of Pinus, growing on heath and on other dry warm soils can also grow on moors, is no longer incomprehensible when we remember that heath often has an extremely ill aerated, periodically soaking, raw- humus soil, which exemphfies the ' dry production of peat '. It must also be noted that peat has a great power of retaining water.^ Many moors and heaths may become very dry in their upper layers in summer. We can often walk dry-footed not only over moors clothed with such marsh-plants as Scheuchzeria, Rhynchospora alba, and Carex limosa, but can note that the bog-mosses are so dry as to crackle as we tread on them. Many arctic swamps and moors often become completely dried up. Livingston ^ arrives at the conclusion that at least in some bog- waters there occur chemical substances ' which are not in direct relation to the acidity of the water ', but which ' act on the vegetation ' : and * it is suggested that these substances may play an important role in the inhibition, from bogs, of plants other than those exhibiting xerophytic adaptations '. But the weightiest cause of the physiological dryness of the soil probably lies in the presence of free humous acids and other dissolved substances which chemically affect the roots.^" Moor-soil is probably • Johow, 1884. * Kihlman, iSqo- * Tschaplowitz, 1883. * Seep. 50. ' Kihlman, 1890. * Gobcl, 1889-91. ' Transeau, 1903, 1905. * Sec pp. 47.61. * Livingston, 1904. " See Weber, 1902, 1903 ; Schimpcr, 1898 ; Cowlcs, 1901 ; Brunckcn, 1902 a ; Friih und Schroter, 1904. O 2 196 OXYLOPHYTES SECT. VI always acid ; humous acids depress the root's activity and render it more difficult for the plant to replace the water lost by transpiration. In fact, various factors enter into the question, and possibly all of them play a part in evoking xeromorphy. Finally it may be noted that there are not only moors inclining to xerophily, but also others leaning rather to hydrophily, and that, in addition to the structural types ^ and forms of leaves mentioned, there are others which apparently show no signs of xerophily and cannot be shown to be in harmony with this habitat : for example, broad, hastate, sagittate, or cordate leaves occur in many Araceae concerned, while broad, orbicular or reniform leaves are shown by Rubus Chamaemorus, Caltha palustris, and Viola palustris. The formations growing on acid soil may be arranged according to the following scheme : — 1. Low-moor formation : Often represents the first stage and the one most nearly related to hydrophytic formations. 2. Grass-heath. Tussock-formation (Pecuhar to the Southern Hemisphere). 3. High-moor formation : Likewise very wet. 4. Moss-tundra formation. 5. Lichen-tundra formation. The others are drier formations in which woody plants are the domi- nant constituents : 6. Dwarf-shrub heath formation. 7. Bushland and forest formation : The formations 4, 5, 6 may occur as undergrowth here.^ CHAPTER XLVIL LOW-MOOR FORMATION The vegetation of low-moor requires less water than does that of reed-swamp, with which it is often continuous on the landward side and at the expense of which it often develops. It shows less open water, , which, moreover, is visible to a less extent than in reed-swamp and I frequently is only to be seen at certain times. The water-table is always] at a high level. The vegetation is closer, and the vegetative shootsj project nearly entirely into the air. The water is still or flows but slowly ; the land fiat and horizontal, though in arctic countriesl it may slope slightly. Humous acids arise in the soil, which becomes moor-like because of the accumulation of vegetable fragments ^ ; thick layers of peat may be produced, especially from certain species, including ■- at times some, such as Phragmites, also present in reed-swamp. The peat contains much nitrogen, which, however, is not always in a form easily available to plants. * In regard to the great variation in the anatomical characters of monocotylous marsh-plants, see Grabner, 1895. * In reference to these formations, special reference should be made to the great work by Friih and Schroter, 1904. See also Pound and Clements, 1900; Clements, 1904; Weber, 1902, 1903; MacMillan, 1893, 1896, 1897; Livingston, 1904; Ramann, 1895 and 1906; Yapp, 1908. ' See Chap. XVI. CHAP. XLVii LOW-MOOR FORMATION 197 The water coming from low-moor is rich in calcium and potassium. Low-moors arise not only around reed-swamps, but also at the margin of standing or flowing water whose boundaries they constantly narrow, as the reed-vegetation gradually advances into the water. The moor nearly always commences to form on the side towards the prevailing wind ; waves caused by the wind prevent or obstruct its production