I! I GRAY'S BOTANICAL TEXT-BOOK. VOLUME I. STRUCTURAL BOTANY. GRAY'S BOTANICAL TEXT-BOOK CONSISTS OF VOL. I. STRUCTURAL BOTANY. By ASA GRAY. II. PHYSIOLOGICAL BOTANY. By GEORGE L. GOODALE. III. INTRODUCTION TO CRYPTOGAMIC BOTANY, BOTH STRUCTURAL AND SYSTEMATIC. By WILLIAM G. FARLOW. (In preparation.) IV. SKETCH OF THE NATURAL ORDERS OF PH^ENOGAMOUS PLANTS ; their Special Morphology, Useful Pro- ducts, &c. (In preparation.) THE BOTANICAL TEXT-BOOK. (SIXTH EDITION.) PART I. STRUCTURAL BOTANY . ORGAXOGRAM-IY ON THE BASIS OF MORPHOLOGY. TO WHICH IS ADDED THE PRINCIPLES OP TAXONOMY AND PHYTOGRAPHY, AND 31 @I000arp of Botanical &erm0. BY ASA GRAY, LL.D., ETC., // FISHER PROFESSOR OF NATURAL HISTORY (BOTANY) IN HARVARD UNIVERSITY. LONDON: MACMILLAN AND COMPANY. Copyright. BY ASA GRAY, 1879. printcfc big mm. flvieon •ftcw Korfc, ia. S. H. PREFACE. THE first edition of this treatise was published in the year 1842, the fifth in 1857. Each edition has been in good part rewritten, — the present one entirely so, — and the compass of the work is now extended. More elementary works than this, such as the author's First Lessons in Botany (which contains all that is necessary to the prac- tical study of systematic PhcEnogamous Botany by means of Manuals and local Floras), are best adapted to the needs of the young beginner, and of those who do not intend to study Botany comprehensively and thoroughly. The present treatise is intended to serve as a text-book for the higher and completer instruction. To secure the requisite fulness of treatment of the whole -range of sub- jects, it has been decided to divide the work into distinct volumes, each a treatise by itself, which may be indepen- dently used, while the whole will compose a comprehensive botanical course. This volume, on the Structural and Morphological Botany of Phasnogamous Plants, properly comes first. It should thoroughly equip a botanist for the scientific prosecution of Systematic Botany, and furnish needful preparation to those who proceed to the study of Vegetable Physiology and Anatomy, and to the wide and varied department of Cryptogamic Botany. iv PREFACE. The preparation of the volume upon Physiological Botany (Vegetable Histology and Physiology) is assigned to the author's colleague, Professor GOODALE. The Introduction to Cryptogamous Botany, both structu- ral and systematic, is assigned to his colleague, Professor FARLOW. A fourth volume, a sketch of the Natural Orders of Phsenogamous Plants, and of their special Morphology, Classification, Distribution, Products, &c., will be needed to complete the series : this the present author may rather hope than expect himself to draw up. ASA GRAY. HERBARIUM OF HARVARD UNIVERSITY, CAMBRIDGE April 10, 1879. *#* The numerals in parentheses, which are here and there introduced into sentences or appended to them, are references to the numbered para- graphs in which the topic is treated or the term explained. CONTENTS. PAGE INTRODUCTION. THE DEPARTMENTS OF THE SCIENCE .... 1 CHAPTER I. OUTLINES OF THE GENERAL MORPHOLOGY OF PHvENOGAMOUS PLANTS . 5 CHAPTER II. MORPHOLOGY AND DEVELOPMENT OF THE EMBRYO AND SEEDLING 9 The Embryo, its Nature, Structure, and Parts ...... 9 Development of the Dicotyledonous Embryo in Maple ... 10 In Ipomoea, or Morning Glory, &c., with Albuminous Seeds . 13 In Embryos with thickened Cotyledons 16 As of Almond, Beech, Bean, &c 17 With Hypogaeous Germination and no Elongation of Caulicle 19 In Megarrhiza, &c., with concreted Petioles to the Cotyledons 21 In Ipomoea leptophylla with foliaceous and long-petioled Cotyledons and no elongation of Caulicle 22 In Pumpkin, &c., with no Primary Root 22 The Polycotyledonous Embryo 23 The Monocotyledonous Embryo of Iris, Onion, Cereal Grains 24 Pseudo-monocotyledonous and Acotyledonous Embryo ... 26 Dicotyledonous and Monocotyledonous Plants 27 CHAPTER in. MORPHOLOGY AND STRUCTURE OF THE ORGANS OF THE PLANT IN VEGETATION . . 27 SECTION I. OF THE ROOT 27 Nature, Growth, and Composition 28 Root-hairs 29 Kinds of Roots 29 Duration; Annuals 30 Biennials 31 Perennials 32 Aerial Roots 33 Epiphytes or Air-plants 35 Parasitic Plants, Green and Colored 36 CONTENTS. SECTION II. OF BUDS 40 Scaly Buds and Bud-scales 40 Naked, Subpetiolar, and Fleshy Buds 41 Bud-propagation 43 Normal, Accessory, and Adventitious Buds „ . 44 SECTION III. OF THE STEM 45 § 1. GENERAL CHARACTERISTICS AND GROWTH 45 Development and Structure 46 Ramification, Branches 47 Excurrent and Deliquescent Stems 48 Definite and Indefinite Annual Growth 49 § 2. FORMS OF STEM AND BRANCHES 50 Herbs, Shrubs, Trees, Culm, Caudex, Scape 60 Climbing Stems, Twining or otherwise 51 Leaf-climbers, Tendril-climbers, and Root-climbers .... 52 Suckers, Stolons, Offsets, Runners 53 Tendrils formed of Stems 54 Sympodial and Monopodial Stems 65 Spines or Thorns and Subterranean Steins 56 Rhizoma or Rootstock 67 Tuber, Tubercles 50 Conn or Solid Bulb 61 Bulb, Bulblets 62 Condensed Aerial Stems 64 Stems serving for Foliage, Phyllocladia, Cladophylla . . . 65 Frondose Stems 66 § 3. INTERNAL STRUCTURE 67 Anatomical Elements 68 Endogenous Structure 70 Exogenous Structure ; its Beginning 73 First Year's Growth .• 74 Pith, Layer of Wood, &c 75 Bark, its Parts and Structure 76 Annual Increase in Diameter 78 Demarcation of Annual Layers 70 Sap-wood and Heart-wood 80 Growtli and Duration of Bark 81 Living Parts of a Tree or Shrub, Longevity 83 The Plant composite , 84 SECTION IV. OF LEAVES 85 § 1. THEIR NATURE AND OFFICE 85 Parts of a Leaf 85 Duration, Defoliation, Normal Position 86 CONTENTS. yii § 2. THEIR STRUCTURE AND FORMS AS FOLIAGE 87 Internal Structure or Anatomy 87 Parenchyma-cells 88 Epidermis, Stomata or Breathing-pores 89 Framework, Venation 90 Parallel-veined or Nerved Leaves 91 Reticulated or Netted-veined Leaves 92 Pinnately or Feather-veined and Palmately or Radiately Veined 03 Forms as to Outline 94 Forms as to Extremity 96 Forms as to Margin or Special Outline and Dentation ... 97 Lobation or Segmentation 98 Number and Arrangement of Parts 99 Compound Leaves, Pinnate and Palmate or Digitate, &c. . . 100 Petiole or Leafstalk 104 Stipules, Ligule, Stipels 105 Leaves in unusual Modifications 106 Such as Inaequilateral, Connate, Perfoliate 107 Vertical and Equitant 108 Without distinction of Parts 109 Stipules serving for Blade 109 Phyllodia, or Petioles serving for Blade 110 § 3. LEAVES SERVING SPECIAL OFFICES llr Utilizing Animal Matter 110 Ascidia or Pitchers Ill Sensitive Fly-traps 113 Leaves for Storage 115 Bulb-scales and Bud-scales 116 CHAPTER IV. PHYLLOTAXY, OR LEAF-ARRANGEMENT . 119 SECTION I. DISTRIBUTION OF LEAVES ON THE STEM . Phyllotaxy either Verticillate or Alternate, Cyclical or Spiral 119 Verticillate or Cyclical Arrangement 120 Alternate or Spiral Arrangement 121 Its Modes and Laws 122 Relation of Whorls to Spirals 129 Hypothesis of the Origin of Both 130 Fascicled Leaves 131 SECTION II. DISPOSITION OF LEAVES IN THE BUD 132 Vernation and ^Estivation ; the Modes ...,..., 132 Direction. Dextrorse and Sinistrorse 140 Viii CONTENTS. CHAPTER V. ANTHOTAXY OR INFLORESCENCE .... 141 Bracts and Bractlets and their Modifications 141 Peduncles, Pedicels, Rhachis, Receptacle 143 Position of Flower-buds, Kinds of Inflorescence 144 Indeterminate, Indefinite, or Botryose 146 Raceme, Corymb, Umbel 146 Head or Capitulum 147 Syconium or Hypanthodium 148 Spike, Spadix, Ament or Catkin 149 Panicle and other Compound Forms 150 Determinate or Cymose 151 Cyme, Glomerule, &c 152 Botryoidal Forms of Cymose Type 153 Sympodial Forms 154 Scorpioid and Helicoid, the Pleiochasium, Dichasium, and Monochasium 155 Bostryx, Cincinnus, Rhipidium, Drepanium, &c 156 Mixed Inflorescence 158 Thyrsus, Verticillaster, &c 159 Relations of Bract, Bractlet, and Flower 160 Anterior and Posterior, or Inferior and Superior 160 Median and Transverse 160 Position of Bractlets 161 Tabular View of Inflorescence 162 CHAPTER VI. THE FLOWER 163 SECTIOK I. ITS NATURE, PARTS, AND METAMORPHY .... 163 Floral Envelopes, Perianth, or Perigone 164 The Parts, Calyx and Corolla 165 Androecium, Stamens 165 Gyncecium, Pistils 166 Torus or Receptacle of the Flower 167 Metamorphosis 167 Unity of Type illustrated by Position and Transitions ... 169 Teratological Transitions and Changes 170 SECTION II. FLORAL SYMMETRY 174 Symmetrical, Regular, and Complete Flower 175 Numerical Ground-plan 176 Pattern Flowers 176 Diplostemonous Type 177 SECTION III. VARIOUS MODIFICATIONS OP THE FLOWER . . . 179 § 1. ENUMERATION OF THE KINDS 179 § 2. REGULAR UNION OF SIMILAR PARTS 180 Coalescence or Cohesion . 180 CONTENTS. ix § 3. UNION OF DISSIMILAR OR SUCCESSIVE PARTS .... 181 Adnation or Connation 182 Hypogynous, Perigynous, Epigynous 183 § 4. IRREGULARITY OF SIMILAR PARTS 184 § 6. DISAPPEARANCE OR OBLITERATION OF PARTS .... 187 Abortion or Suppression of Parts of a Circle 187 Abortion or Suppression of whole Circles . ... v ... 190 1 Terms therewith connected 191 Suppressed Perianth 191 Suppressed Androecium or Gynoecium 193 Along with suppressed Perianth 194 Neutral Flowers 195 § 6. INTERRUPTION OF NORMAL ALTERNATION 196 Anteposition or Superposition 195 In Appearance only 196 Superposition by Spirals 196 Anteposition with Isostemony and Diplostemony 197 With Obdiplostemony 198 § 7. INCREASED NUMBER OF PARTS 200 Regular Multiplication 200 Parapetalous Multiplication 201 Chorisis or Deduplication 202 § 8. OUTGROWTHS 209 Their relation to Chorisis : Trichomes 209 Corona or Crown 210 Ligule 211 § 9. FORMS OF THE TORUS OR RECEPTACLE 211 Stipe, Thecaphore, Gynophore, Carpophore, &c 212 Disk 213 Hypanthium 214 SECTION IV. ADAPTATIONS OF THE FLOWER TO THE ACT OF FERTILIZATION 215 § 1. IN GENERAL 215 Close and Cross Fertilization, or Autogamy and Allogamy . 216 § 2. ADAPTATIONS FOR ALLOGAMY OR INTERCROSSING ... 216 Wind-fertilizable or Anemophilous Flowers 217 Insect-fertilizable or Entomophilous Flowers 218 Irregularity as related to Allogamy 219 CONTENTS. Dichogamy, either Proterandrous or Proterogynous .... 219 Proterogyny 219 Proterandry 220 Particular Adaptations in Papilionaceous Flowers .... 225 In Kalmia-blossoms, Iris, &c 229 Transportation of Pollinia 230 In Orchidaceae and Asclepiadaceae 231 Heterogonous Dimorphism and Trimorphism 234 § 3. ADAPTATIONS FOR CLOSE FERTILIZATION 240 Cleistogamy 241 SECTION V. THE PERIANTH, OR THE CALYX AND COROLLA IN PARTICULAR 243 Perianth as to Duration, Numerical Terms, Union, &c. . . . 243 Parts of Petals and of Gamophyllous Perianth 245 Forms of Corolla and Calyx 246 SECTION VI. THE ANDROSCIUM, OR STAMENS IN PARTICULAR . 249 The Stamen as a whole ; Numerical Terms 249 The Filament and the Anther; their Modifications .... 251 Pollen 256 Pollen-tubes 258 SECTION VII. THE PISTILS, OR GYNOSCIUM 259 § 1. IN ANGIOSPERMS 259 Carpel or Carpophyll 260 Ventral and Dorsal Sutures ; Placenta 261 Simple or Apocarpous Pistils 262 Compound or Syncarpous Pistil 263 With two or more Cells and Axile Placentae ; Partitions . . 264 With one Cell and Parietal Placentae 265 With one Cell and Free Central Placenta 266 Anomalous Placentation 267 § 2. IN GYMNOSPERMS 268 Structure in Gnetaceae 269 Structure in Coniferae 270 In the Yew Family 271 In the Pine Tribe, &c. 272 In the Cypress Tribe 273 Structure in Cycadaceae 274 SECTION VIII. THE OVULE 276 Its Structure and Position 277 Its Forms, Orthotropous, Campylotropous, Amphitropous, Anatropous 278 Origin and Morphological Nature of the Ovule 282 Origination of the Embryo 283 CONTENTS. Xi CHAPTER VII. THE FRUIT 285 SECTION I. ITS STRUCTURE, TRANSFORMATIONS, AND DEHIS- CENCE 285 Pericarp, its Alterations, Accessions, and Transformations . 287 Dehiscenee 288 SECTION II. THE KINDS OF FRUIT 291 Simple Fruits 291 Dehiscent Fruits, Follicle, Legume, Capsule, Pyxis, Silique . 292 Indehiscent Dry Fruits, Samara, Akene, Utricle, Caryopsis, Nut, &c 294 Fleshy Fruits, Drupe, Pome, Pepo, Berry, &c 297 Aggregate Fruits 299 Accessory or Anthocarpous Fruits 300 Multiple or Collective Fruits, Syconium, Strobile, &c. . . . 301 Table of Simple Fruits 304 CHAPTER VIII. THE SEED 305 Its Stalk, Coats, and Appendages 306 Aril or Arillus 308 Nucleus or Kernel, Albumen 309 The Embryo, its Parts and Positions 311 The Cotyledons as to Adjustment and Number 313 •CHAPTER IX. TAXONOMY 315 SECTION I. THE PRINCIPLES OF CLASSIFICATION IN NATURAL HISTORY 315 Individuals 315 Species 317 Varieties, Races, &c 318 Cross-breeds and Hybrids 321 Genera 323 Orders, Classes, Tribes, &c 325 Sequence of the Grades 327 Nature and Meaning of Affinity 327 Theory of Descent and Natural Selection ........ 328 SECTION II. BOTANICAL CLASSIFICATION 331 Ante-Linnjean Classifications 332 Linnamn Classification 333 Sexual Artificial System 334 Natural System 338 As presented by Jussieu 339 Some of its Modifications 340 Xii CONTENTS. CHAPTER X. PHYTOGRAPHY 345 SECTION I. NOMENCLATURE 345 Names of Plants, Binomial Nomenclature 346 Rules for naming Plants 347 Names of Genera 348 Names of Species, Varieties, &c 350 The Fixation, Precision, and Citation of Names 352 Subgeneric Names 356 Tribal and Ordinal Names 357 Names of Cohorts, Classes, &c 358 SECTION II. GLOSSOLOGY OR TERMINOLOGY 359 SECTION in. DESCRIPTION 861 Characters 361 Punctuation 364 Synonomy 365 Iconography 366 Habitat and Station, &c 366 Etymology of Names 366 Accentuation, Abbreviations 367 Signs 368 Floras, Monographs, &c 369 SECTION IV. SPECIMENS, DIRECTIONS FOR THEIR EXAMINA- TION, PRESERVATION, &c 370 Implements of Investigation 370 Diagrams 371 Herborizing 371 Drying Specimens 375 Poisoning Specimens 379 The Herbarium , 380 ABBREVIATIONS SIGNS GLOSSARY OF BOTANICAL TERMS, WITH INDEX STRUCTURAL BOTANY THE BASIS OF MORPHOLOGY. INTRODUCTION. 1. THE two Biological Sciences,1 considered as parts of Natural History, are Zoology and Botany. The latter is the natural history of the Vegetable Kingdom. It embraces every scientific inquiry that can be made respecting plants, their nature, their kinds, the laws which govern them, and the part they play in the general economy of the world. 2. We cannot distinguish the vegetable from the animal king- dom by an}' complete and precise definition. Although ordinary observation of their usual representatives maj- discern little that is common to the two, yet there are many simple forms of life which hardly rise high enough in the scale of being to rank dis- tinctively either as plant or animal ; there are undoubted plants possessing faculties which are generally deemed characteristic of animals ; and some plants of the highest grade share in these endowments. But in general there is a marked contrast between animal and vegetable life, and in the part which animals and plants respectively play in nature. 3. Plants only are nourished upon mineral matter, upon earth and air. It is their peculiar office to appropriate mineral mate- rials and to organize them into a structure in which life is mani- fested, — into a structure which is therefore called organic. So the material fitted for such structure, and of which the bodies 1 Biology, the science of life, or rather of living things, in its earlier use was equivalent to physiology : recently, it has come to denote the natural history of plants and animals, t. e. of the two organic kingdoms, including both their physiology and descriptive natural history. 1 2 INTRODUCTION. of plants and animals are composed, is called organic matter. Animals appropriate and live upon this, but have not the power of producing it. ISo the vegetable kingdom stands between the mineral and the animal ; and its function is to convert materials of the one into food for the other. Although plants alone are capable of building up living structure out of mineral mate- rials, and are the sole producers of the organic matter which is essential to animal life, and although animals consume that which plants produce, yet plants also consume organic matter, more or less, acting in this respect like animals in all their opera- tions, except in the grand and peculiar one by which they assimilate mineral matter. Most plants of the higher grades assimilate largely and consume little, except in special opera- tions. Some, on the contrary, are mainl}T consumers, and feed upon formed organic matter, living in this respect after the manner of animals. The living substance of plants and animals is essentially the same. 4. Botany deals with plants : 1. As individuals, and in respect to their structure and functions. 2. In their kinds, and as respects their classification, nomenclature, &c. Accordingly, the most comprehensive division of the science is into PHYSIO- LOGICAL or BIOLOGICAL BOTANY (using these terms in their widest sense) and SYSTEMATIC BOTANY. But as Plrysiology and Biology, in the restricted sense, relate only to functions or actions and their consequences, the first department naturally divides into two, viz. Structural Botany and Physiology. 5. STRUCTURAL BOTANY comprehends all inquiries into the structure, the parts, and the organic composition of vegetables. This is termed ORGANOGRAPHY, when it considers the organs or obvious parts of which plants are made up, and MORPHOLOGY, when the study proceeds on the idea of type. The term ORGANOGENY has been applied to the study of the nascent organs and their development; PHYTOTOMY, or VEGETABLE ANATOMY, to that of the minute structure of vegetables as re- vealed by the microscope, i. e. to the composition of the organs themselves. But, since anatomy in the animal kingdom includes the consideration of general as well as of minute structure, and indeed answers to organography, the minute anatomy of both kingdoms takes the special name of HISTOLOGY. The study of functions, or of the living being (animal or plant) in action, is the province of PHYSIOLOGY. 6. SYSTEMATIC BOTANY, or the study of plants in their kinds and in regard to their relationships, comprises TAXONOMY, or the principles of classification, as derived from the facts and ideas INTRODUCTION. 3 upon which species, genera, &c., rest; CLASSIFICATION or the SYSTEM OF PLANTS, the actual arrangement of known plants in 3}'stematic order according to their relationships ; PHYTOGRAPHY, the rules and methods of describing plants ; and NOMENCLATURE, the methods and rules adopted for the formation of botanical names. GLOSSOLOGY or TERMINOLOGY J is a necessaiy part of Phytography or Descriptive Botany, and hardly less so of Structural Botan}' : it relates to the application of distinctive terms or names to the several organs or parts of plants, and to their numberless modifications of form, &c. This requires a copious vocabulary of well-defined technical terms, by the use of which the botanist is able to describe the objects of his study with a precision and brevity not otherwise attainable. It will be convenient to exemplify the principal terms along with the modifications of conformation which they designate ; and also, for greater fulness and facility of reference, to append to this volume an alphabetical summary of them, or Vocabulary of Botanical Terms.'2 7. The present volume is mainly devoted to Morphological Botany ; that is, to Structural Botany on the basis of mor- phology. This department cannot be properly dealt with apart from considerable reference to intimate structure, development, and function, the subject-matter of vegetable histology and physiology. But these will here be treated only in the most general or incidental and elementary way, and only so far as is necessary to the understanding of the morphology of the stem, leaves, &c. The whole discussion of the histology and physiology of plants .is relegated to a following volume and to another hand. 8. The most comprehensive and important division of the vegetable kingdom is into plants of the higher and of the lower series or grade, i. e. into PH/ENOGAMOUS (or PHANEROGAMOUS) or FLOWERING, and CRYPTOGAMOUS or FLOWERLESS PLANTS. The first are all manifestly of one type, and therefore have a consist- ent and simple morphology. The second differ among them- selves almost as widely as they do from the higher series ; and 1 GLOSSOLOGY is the better word, but TERMINOLOGY, although a hybrid of Latin and Greek, is in common use. 2 What is called GEOGRAPHICAL BOTANY is the study of plants in respect to their natural distribution at the present time over the earth's surface, and the causes of it. FOSSIL BOTANY (Vegetable Palaeontology) relates to the plants of former ages, as more or less made known in their fossil remains. MEDICAL BOTANY, AGRICULTURAL BOTANY, and the like, are applications of Botany to medicine, agriculture, &c. 4 INTRODUCTION. their morphology is more special and difficult. Wherefore it is better to treat them separately and subsequently. This will be done in a third part, by an associate devoted to Ciyptogamic Botany. 9. Thus the field is here left clear for the Structural Botany of Phsenogamous or Flowering Plants, with which the stud}T of the science should naturally begin. In theory it may seem proper to commence with the simplest plants and the most ele- mentary structures ; but that is to put the difficult and recondite before the plain and obvious. The type or plan of the vegetable kingdom, upon which morphological botany is grounded, is fully exemplified only in the higher grade of plants, is manifest to simple observation, and should be clearly apprehended at the outset. CHAPTER I. OUTLINES OP THE GENERAL MORPHOLOGY OF PH^ENOGAMOUS PLANTS. 10. MORPHOLOGY, the doctrine of forms, as the name denotes, is used in natural history in nearly the same sense as the older term Comparative Anatomy. If it were concerned merely with the description and classification of shapes and modifications, it would amount to little more than glossology and organography. But it deals with these from a peculiar point of view, and under the idea of unity of plan or type.1 11. As all vertebrate animals are constructed upon one type (or ground plan) , which culminates or has its archetype in man, so all plants of the higher grade (8) are strictly of one type ; the different kinds being patterns or repetitions of it, with varia- tions. The vegetable kingdom, however, does not culminate in an archetype or highest representative. As respects the organs of vegetation, the higher classes of cryptogamous plants exhibit this same type ; but it is only in the most general or in a recondite sense that this can be said of their organs of repro- duction, and of the less differentiated structure of the lowest classes. Wherefore cryptogamous plants are left out of the present view, to be treated apart. 12. Viewed morphologically and as to its component organs, a plant is seen to consist of an axis or stem, which sends off roots into the soil, and bears lateral appendages, commonly as leaves, but which may be very unlike leaves in whole appearance 1 The term Morphology was introduced into science by Goethe, at least as early as the year 1817 (Zur Naturwissenschaft iiberhaupt, besonders zur Morphologic, Stuttgart und Tubingen, 1817-24). On page 9 of the first volume, he is understood to have suggested this word for the purpose and in the sense now adopted in botany and zoology. It essentially replaces an earlier and somewhat misleading word, Metamorphosis. (304.) Apparently the first botanist to adopt the term was Auguste de St. Hilaire, in his " Le9ons de Botanique, comprenant principalement la Mor- phologie Ve'getale, etc., Paris, 1841. The term seems not to have been taken up, in zoology, by Etienne Geoffroy Saint-Hilaire, the antagonist of Cuvier (who was of a wholly different family from that of the botanist), although the same idea was denoted by his phrase " unity of organic composition " 6 GENERAL MOliPHOLOGY and function. These appendages, whatever their form or use, accord with leaves in mode of origin, position, and arrangement on the axis or stem. Their most general and ordinary form is the familiar one of foliage ; hence the name of leaves has been by botanists extended in a generic way from the green expan- sions which constitute foliage to other forms under which such appendages occur. The proper morphological expression is, that the latter are homologous with -leaves, or are the homologues of leaves.1 13. Leaves are borne upon the stem at definite places, which are termed NODES. A node may bear a single leaf or a greater number. When it bears two, they occupy opposite sides of the stem. When three, four, or more, they divide the circum- ference of the stem equally, forming a circle, technically a WHORL, or in Latin form a VERTICIL. When only two, the pair evidently answers to the simplest kind of whorl. So that leaves are either single on the nodes, in which case they are alter- nate, that is, come one after another on the stem ; or in whorls (whorled, verticillate) , in the commoner case of a single pair being called opposite. The bare space between two successive nodes is an INTERNODE. This is longer or shorter, according to the amount of longitudinal growth, which thus spaces the leaves, or whorls of leaves, in most various degrees, either widely when the internodes are elongated, or slightly when they remain very short. The plant, therefore (roots excepted), is made up of a series of similar parts, i. e. of portions of stem, definitely bearing leaves, each portion developed from the apex of the preceding one. This constitutes a simple-stemmed plant. 14. Branching is the production of new stems from the older or parent stem. These normally appear in the AXILS of leaves, that is, in the upper angle which the leaf forms with the stem, — from which they grow much as the primary stem grew from the seed. The primary stem, connected with the ground, produces roots which develop downwardly into the soil, from which they draw sustenance. Branches, when developed above ground, 1 A common designation for all these appendages being desirable, a good one is furnished by the Greek name for leaf, <(>v\\ov, PHYLLUM, plural PHYLLA. This, used with prefixes, may be made to designate the kind of leaves in many cases, — as, prophylla, cataphi/lla, hypsophyl/a. Recent German botanists use the word Phyllome in this sense. It is a rather convenient and well-sounding word ; but phylloma is the exact Greek equivalent of our word foliage, and therefore not very well chosen as a common term for leaves which are not foliage as well as those which are. Nor will this word, like phyllum, readily take prefixes, as above, or the adjec- tive form, as it readily does in prophyllous, hypsophyllous, gamophyllous, &c. OF PHJENOGAMOUS PLANTS. 7 being in organic connection with their parent stem, do not usually produce roots ; but when placed in equally favorable conditions for it, i. e. on or in the soil, they may strike root as freely as does the original stem. 15. An incipient stem or branch, with its rudimentary leaves, is a BUD. The normal situation of a bud is in the axil of a leaf (axillary) , the development giving rise to branches ; or else at the apex of an axis (terminal), where there can be only one, the development of which continues that axis.1 16. As branches are repetitions and in one sense progeny of the stem which bears them, so the serial similar parts or leaf- bearing portions of a simple stem are repeti- tions, or in a like sense progeny, each of the preceding one from which it grew. The simple-stemmed plant is made up of a series of such growths, each from the summit of its predecessor ; the branched plant, of ad- ditional series, laterally developed, from ax- illary buds. These ultimate similar parts into which a plant may thus be analyzed, and which are endowed with or may produce all the fundamental organs of vegetation, were by Gaudichaud called PHYTONS. But phyton, being the common Greek name for plant, was not a happily chosen appellation for plant- elements, or homologous plant-units. A better (7)s>\ term for them is PHYTOMERA (yvrov, plant, v >\ I H^Qog, part) , equivalent to plant-parts, — the \\ \, structures which, produced in a series, make u up a plant of the higher grade. In English, « 11 the singular may be shortened to PHYTOMER. 17. This theoretical conception of the organic composition of the plant is practically impor- tant to the correct understanding of morpho- logical botany. The diagram, Fig. 1, serves to represent the organic elements, or pfn/fomera, in a simple case, such as that of a growing plant of Indian Corn, or other Grass. Here 1 Bifurcation by the division of a terminal hud into two, as in Acrogenous Cryptogams, is supposed by some to occur, even normally, in some Phaeno- gams, especially in certain forms of inflorescence ; but this has never been convincingly made out. FIG. 1. Diagram of a simple-stemmed plant, exhibiting the similar parts, or phytomera, a to h, of which it is composed. 8 GENERAL MORPHOLOGY the leaves are alternate ; in other words, each phytomer is single-leaved ; while in the subsequent illustrations of plants developed from the seed, at least the earliest phytomera are two-leaved. 18. The plan thus exhibited in the leafy stem begins in the embryo, or initial plant in the seed, and is carried on into the flower, in which the normal development of the axis finally ends. One plan prevails throughout. To illustrate it, the morphology and growth of the embryo, of the plant developed for vegetation and the general purposes of its individual existence, and lastly of the flower, through which sexual reproduction takes place, may be successively treated in this order. THE EMBRYO AND SEEDLING. CHAPTER II. MORPHOLOGY AND DEVELOPMENT OP THE EMBRYO AND SEEDLING. 19. The Embryo is the initial plant, originated in the seed.1 In some seeds it is so simple and rudimentary as to have no visible distinction of parts : in others, these parts may have assumed forms which disguise their proper character. But every well-developed embryo essentially consists of a nascent axis, or stem, bearing at one end a nascent leaf or leaves, or what an- swers to these, while from the other and naked end a root is normally to be produced. This stem is the primitive internode of the plant : its leaf or pair of leaves is that of the first node. The plant therefore begins as a single phjtomer. Some embryos are no more than this, even when they have completed their proper germination : others have taken a further development in the seed itself, and exhibit the rudiments of one or more fol- lowing phytomera. The embiyo of the Maple is an example of the first kind ; and, being large enough for handling and for the display of all its parts to the naked ej'e, and the character of these parts being manifest even in the seed, it is a good subject with which to commence this stud}'. And for this the Sugar- Maple is one of the best of the Maples. Its embryo (seen in Fig. 2 in the coiled condition which it occupies in the seed, and in Fig. 3 and Fig. 4 uncoiling and be- ginning to grow) is an initial stem, bearing a pair of leaves, and nothing more. These parts take the technical names of 1 Normally a seed contains a single embryo. Polyembry, the formation of two or more embryos, occurs occasionally as a kind of superfcetation in some seeds. In those of the cultivated Orange it is most common, and an evident monstrosity. In Coniferae and Loranthaceae, two or three embryos, of equal size and perfection, are not rarely produced. FIG. 2. Embryo of Sugar Maple, in vertical section, as coiled in the seed, merely somewhat loosened. 3. Embryo of same, just beginning to unfold in germination. 4. Same more advanced: a. its stem or caulicle; bb. its two leaves (y cotyledons. 10 MORPHOLOGY 20. Caulicle or Radicle, and Cotyledons. The name of radicle was early applied to the axis of the embryo below the cotyledons, on the supposition that it was the actual beginning of the root. But its structure and mode of growth show it is not root (24, 44, 78), but a body of the exact nature of stem, from the naked end of which the root is developed. Wherefore Gaulide (Lat. cauliculus, diminutive of caulis, stem) is the appropriate ' name ; and it would be gen- erally adopted, were it not that the older term is so incorporated into the language of sys- tematic botany (in which fixity and uniformity are of the utmost importance) that it is not easily displaced-. It may be continued in descriptive botany on this account, but in morphology it is apt to mislead ; and the name of caulicle, suggestive of the true nature of the organ, is preferable.1 The more fanciful name of Cotyledons was very early applied to what are now recognized as answering to the leaves of the embryo : it has the negative merit of suggesting no misleading analogy.2 21. Development of the Dicotyledonous Em- bryo, i. e. the two-leaved embryo. This, in the Red Maple (Figs. 5-8), usually germinates in summer, shortly after the fruits of the season have matured and fallen to the ground. It differs from that of Sugar Maple in the crump- ling instead of coiling of the cotyledons in the seed. Referring the whole physiology of ger- mination to that part of the work which treats of Vegetable Physiology, the development of the embryo into the seedling may here be described, taking that of a Maple for a convenient type or pattern, with which other forms 1 Linnaeus called it Rostellum, a name which, being etymologically mean- ingless in this connection, is not misleading. The French botanists named it Tigelle, diminutive of tige, stem: but some (like Mirbcl) applied the term to the developing axis above the cotyledons; others, to the early axis both above and below them. The name Radicula originated with Gaertner. 2 The name Cotyledon, which was adopted by Linnaeus, is a Greek word for a cup-shaped hollow or cavity, also for a plant with thickish and saucer- shaped leaves. It was primarily applied to the thickened " lobes " of the embryo, the foliaceous nature of which was not recognized. FIG. 5. One of the twin winged fruita of Red Maple (Acer rubrum), with body divided, to show the seed. 6. Seed extracted and divided, to show the embryo within. 7. Embryo parti" unfolded. 8. Embryo in early stage of R OF THE EMBRYO AND SEEDLING. 11 may afterward be compared. The first growth is seen in the elongation of the radicle or caulicle, and its assumption, as far as possible, of a vertical position, and the production of a root from the naked end. As it emerges from the seed in consequence of this elongation, the root-end of the caulicle points downward into the soil, the caulicle bending, if need be, to assume this position ; and the nascent root, partaking of this disposition, grows in a downward direction. Hence the root has been called the Descend- ing Axis of the plant. While this avoids, the opposite or budding end (as it may be termed) seeks the light, and when free takes an upward direction. The result of this, and of the elongation of the caulicle, is to carry the budding end out of the soil and into the air, where the growing cotyledons unfold or expand and become the first leaves, or Seed-leaves. This initial stem and its continuation therefore constitutes the Ascending Axis. If the budding end happen to lie pointing downward and the root-end upward in the ground when germination begins, both will curve quite round, as they grow, to assume their appropriate directions. If obstacles intervene, each will take as nearly as possible its wonted direction, through an instinctive tendency and action, which insures that each part of the plant shall be developed in its fit medium, — the root in the dark and moist earth, the stem and leaves in the light and air. 22. The plantlet, thus established, has now all the essential Organs of Vegetation, as they are called, i. e. root, stem, and leaves. Its subse- quent development, so far as vegetation (apart from proper reproduction) is con- cerned, consists in the addition of more of these, until the whole herb, shrub, or tree is built up. 23. In Maples (as in the Morning Glory, Fig. 16, and many others) the embryo in the seed, and until after the full develop- ment of its cotyledons or seed-leaves, shows no rudiments of the subsequent growth. The embrj-o grows into the plant- let wholly by the appropriation of prepared nourishing matter which was provided by the mother-plant and stored in the seed, — in the case of the Maple, wholly in the embiyo itself, mainly in its cot3*ledons. FIG. 9. Maple embryo developer! into plantlet of one phytomer, and producing rudiments of the second: the lower portion covered with root-'-airs is the root; tun naked portion above is t!ie canliele. 12 MORPHOLOGY After this is consumed and in good part converted into struc- ture, the plantlet must by the action of its root and leaves imbibe from the soil and air appropriate materials, and assimilate them into nourishing matter needful for further growth. Only then does the rudi- ment of new structure appear, in the form of a growing point, or bud, at the node or apex of the primitive stemlet, between the two seed- leaves. In this case it soon shows itself as a second pair of leaves, at first resting on the node (Fig. 9), next as somewhat upraised by the development of the second internode (Fig. 10, summit), and finally both this inter- node and the pair of leaves complete their growth (Fig. 11). Then the terminal bud which crowns the second node develops in the same way the third pair of leaves and their supporting internode or joint of stem (Fig. 12) ; and so on. 24. The root and the stem grow not only in opposite directions, but in a different mode. The primordial stem, pre-existing in the seed (though at first it ma}r be extremely short) grows throughout its whole length, but most in its upper part, so that it may become a stemlet two or three inches long. But, soon attaining its full growth as to length, the stem is carried upwards by the subsequent joints or portions, similarly developed and elongated, one after the other. Not that each portion necessarily waits until the growth of its prede- cessor is complete, — though this occurs at first in seedling Maples and other embiyos unprovided with much store of food, — yet the development follows this course and order of succession. The root, on the contrary, cannot be said to pre-exist in the seed, or at most it may be said to exist potentially in tissue of the caulicle from which a root or roots normally originate.1 It is formed 1 Yet from nothing which is special to this part of the embryo, nor to the embryo at all. The primary root is developed from subjacent tissue of the tip of the caulicle, just as it is sometimes developed from along the sides, and as secondary roots are from all or most stems under favoring conditions. This complete similarity, and the fact of what is called the " endogenous " origin of roots (i. e. their springing from subjacent rather than superficial tissue) appear fully to warrant the statement in the text above. FIG. 10. Maple plantlet with second internode developing. 11. Same with second internoile and nair of leaves complete, and bud of the third apparent. OF THE EMBRYO AND SEEDLING. 13 in the process of germination, and originates in tissue just back of that which covers the root-end of the caulicle, and which, being carried forward by the subjacent formation (to which it becomes a sort of cap or sheath) , is called the Root-cap. As the primary root thus began b}' a new and local growth at the extremity of pre-existing stem, so it goes on to grow in length wholly or mainly by a continuation of this formation, the new at the end of the old. That is, the root elongates by continual minute increment of its apex or near it, the formed parts very soon ceasing to lengthen. This is in marked distinc- tion from stem, which grows by suc- cessive individualized portions ; and these portions (internodes) , at first very short, attain or are capable of attaining a considerable and sometimes very great, but definitely terminable length, by interstitial growth through- out. Moreover, roots are naked, not producing as they grow either leaves or any organs homologous with leaves. They commonly branch or divide, but in a vague manner ; and their new parts bear what are called Root-hairs, which greatly increase the absorbing surface ; other- wise they are destitute of appendages or organs. 25. With the Maple embryo, here taken as a type, that of Morning Glory, Ipomoea purpurea, or any of its kin, may next be compared. The cotyledons are different in shape, being as broad as long, and notched both at base and apex. They lie in contact in Fig. 14, and are ver}T thin, leaf-like, and green ~P 14- while contained in the seed. Their thinness is shown in Fig. 13, T 14- where a section of the crumpled and folded embryo, as it lies in the seed, exactly divides them (passing through the terminal and basal notches) and also the caulicle, which here is thicker than both. The germination is similar to that of the Maple ; and like that (as Fig. 16 shows), and for the same reason, no bud or ~P /4 rudiment of the further growth pre-exists in the embryo or FIG. 12. Red Maple seedling, with three joints of stem and pairs of leaves developed, the first being the cotyledons. 14 MORPHOLOGY appears in the young plantlet, until that has established itself and had time to elaborate proper material therefor. This con- dition is correlated with thin foliaceous cotyledons, holding no store of nourish- ment. Here the}* do not contain sufficient material for the development of the initial stem and root. The maternal provision for this is .here stored up in the seed around but not within the embryo. This nourishing deposit, seen in the section (Fig. 13) filling the whole space between the seed-coats and the thin embrj-o, was named by the early botanists and vege- table anatomists the ALBUMEN of the seed.1 This substance, softened in germination and by chemical changes rendered soluble, is gradually absorbed by the cotyledons as material for their growth and that of the developing primary stem and root. 26. Seeds in this regard are accordingly distinguished into albuminous and exal- buminous, those supplied with and those destitute of albumen. The difference inheres neither in the character nor in the amount of the maternal provision for the development of the embiyo-plant, but merely in the storage. In exalbuminous seeds the nourishment supplied for this purpose is taken into the embryo itself, mostly into the cotyledons, during the growth and before the maturity of the seed. In albuminous seeds this same material is deposited around or at least external to the embryo. 27. The amount of this deposit is, in the main, inversely pro- 1 Grew appears to have first applied this name, and Gaertner to have introduced it into systematic botany, where it remains in use, although Jussieu replaced it by the term Per/sperm, and Richard by Endosperm, neither of them much better etymologically than the old word Albumen. But it must be kept in mind that it was intended to liken the " albumen " of the seed with the albumen or white of an egg as a body or mass, and not as a chemical substance ; the embryo being fancifully conceived to be analo- gous to the yolk of the egg, the surrounding substance of this kind not unnaturally took the name of the white, viz. albumen. FIG. 13. Section of seed of common Morning Glory, Ipomosa purpurea, dividing the contained embryo through the centre. 14. Embryo of same, detached and straight- ened. 15. Embryo in germination ; the cotyledons only partly detached from the coat of (he seed. 16. Same, Inter and more developed, the cotyledons unfolded and out- spread as the first pair of loaves. OF THE EMBKYO AND SEEDLING. 15 portional to the size and strength of the embryo, or the degree of its development in the seed. A comparison of the various illustrations sufficiently shows this. Figures 17 to 24 exhibit, in a few common seeds, somewhat of this relation, and also of the position and shape assumed in some instances. The upper rank of figures represents sections of seeds ; the embryo left in white ; the albumen as a dotted surface. The lower rank shows the embryos detached. That of Mirabilis has very broad and thin cotyledons, a caulicle of equal length, and the whole curved round the albumen which thus occupies the centre of the seed. That of Potato is coiled in the midst of the albumen, is slender ; the cotyledons narrowed down to semi-cylindrical bodies, not leaf-like in appearance, and the two together not thicker than the cauh'cle. In Barberry the embryo is straight, in the axis of the albumen, which it almost equals ir length ; the cotyledons considerably broader than the caulicle, but short and thickish. That of the Peon}r is similar, but very much smaller, occupA'ing a small space at one end of the albu- men, and seemingly without distinction of parts, but under the microscope and with some manipulation the broader end is found to be divided, that is, to consist of two minute cotyledons. The embryo of a Crowfoot is similar, but still more minute and the parts hardly to be distinguished ; and in some minute em- bryos there is no apparent distinction of parts until they develop in germination. 28. The stud}' of the formation of the emb^o in the seed teaches that all embryos begin with a still more simple, minute, and homogeneous structure ; and these comparisons suffice to show that all such differences are referable to different degrees and somewhat different modes of the development of the embryo while yet in the seed. It also appears that the size and shape , FIG. 17. Section of seed and contained embryo ft Mirabilis (Four-o-elock). 18. Embryo detached entire. FIG. 19. Section of a Potato-seed. 20. Embryo detached entire. FIG. 21. Section of Barberry-seed. 22. Embryo detached entire. FIG. 23. Section of Peony-seed. 24. Embryo detached entire. 16 MOKPHOLOGY of an organ do not indicate its nature, either in the embryo or in subsequent growth. But in all the cases yet mentioned the cotyledons actually demonstrate their nature by developing in germination in a foliaceous manner and becoming the first leaves of the seedling. Nor is this nature much disguised by the fact that the}' differ greatly in form in different species, and that the seed- leaves, or developed cotyledons, differ much in shape and often in texture from the succeeding leaves. (See Fig. 11, 12, 25, &c.) 29. To complete the comparison between the seedling Morning Glory and that of the Maple, it is to be noted that here, while the cotyledons or seed-leaves are two, the following internode bears only one leaf (Fig. 25), as also will the just de- veloping third internode ; and this continues throughout up to the blossom : that is, the leaves subsequent to the cotyledons are not opposite as in the Maple, but alternate. (13.) 30. All the preceding illustrations are from embryos which previous to germination have developed nothing beyond the cot3Tledons. In the following, a rudiment of further growth, or a primary terminal bud, is visible in the seed. It is most manifest in large and strong embryos with thick or fleshy cotyle- FIQ. 25. Further development of Morning Glory, Fig. 16, the root cut away, the internode above the cotyledons and its leaf completed, the next internode and its leaf appearing. FIG. 26. Emhryo (kernel) of the Almond. 27. Same, with one cotyledon removed, to show the plumule, a. FIG. 28. Section of an Apple-seed, magnified, cutting through the thickness of the cotyledons. 29. Embryo of the same, extracted entire, the cotyledons a little separated. FIG. 30. Germination of the Cherry, showing the thick cotyledons little altered, and the plumule developing the earliest real foliage. OF THE EMBRYO AND SEEDLING. 17 dons, i. e. cotyledons well charged with nourishing matter. The early vegetable physiologists gave to it the name of PLUMULE (Lat. plumula, a little plume). The name was suggested by its appearance in such an embiyo as that of the bean (Phaseolus) , in which it evidently con- sists of a rudimentaiy pair of leaves, while in the pea and the acorn it is a rudimentaiy stem, the leaves of which appear only later, when germination has considerably advanced. In any case, the plumule is the bud of the ascending axis already discernible in the seed. Fig. 27, a, shows it in the almond, one cotyledon being removed. Fig. 28 shows it in the section of a similar although much smaller embryo, that of an apple-seed, enlarged to nearly the size of the other. It is equally visible in the cherry, the bean, and the beechnut. The embryo in all these cases constitutes the whole kernel of the seed. For the nourishment, which in all the foregoing illustrations except the first (i.e. in Fig. 13, 17-23), is deposited around or exterior to the embryo, is in these stored within it. 31. The development of these em- bryos in germination proceeds in the normal manner, but with two cor- related peculiarities. First, by the lengthening of the radicle more or less, their thick cotyledons are usually raised to or above the surface of the soil; they expand, assume the green color needful to foliage ; but they imperfectly or in a small degree perform the function of green leaves. Their main office is to supply the other growing parts with the prepared nourishment which they abundantly contain. Then, being thus copiously nourished, the root below and the ready-formed plumule above grow rapidly and strongly, having accumulated capital to draw upon ; and the leaves of the FIG. 31. Beechnut cut across, filled by the fleshy embryo; the thick cotyledons partly enfolding each other. 32. Embryo of the same in early germination 33. Same more advanced; the plumule, which is just emerging in the preceding, here developed into a long internode and a pair of leaves. 18 MORPHOLOGY latter are practically the earliest efficient foliage of the plantlet. Thus, as in the germinating Cherry-seed (Fig. 30) , three or four intcrnodes of stem, with their leaves, may be produced before these leaves themselves are sufficiently developed to make any sensible contribution to this growth. And in the Beech and Bean, the leaves of the plumule come forward almost before the root has attached the plantlet to the soil. (Fig. 32, 35.) Between such cases and that of Maple and the like there are all degrees. There are also familiar cases in which the storage of nourishment in the cotyledons is carried to a maximum, with results which gravely affect the development. FIG. 34. The embryo (the whole kernel) of the Bean. 35. Same early in germi- nation; the thick cotyledons expanding and showing the plumule. 36. Same, more advanced in germination ; the plumule developed into an internode of stem bearing a pair of leaves. FIG. 37. Embryo of Pea, i. e. a pea minus the seed-coat. 38. Advanced germi- natior of the>ame. OF THE EMBRYO AND SEEDLING. 19 32. Thus, in the Pea, near relative of the Bean, the embryo (Fig. 37), which is the whole kernel of the seed, has the cotyledons so gorged with this nutritive store that they are hemispherical ; and the acorn of the Oak (Fig. 39), near relative of the Beech, is in similar case. These extremely obese cotyledons have not only lost all likeness to leaves, but all power of fulfilling the office of foliage, which is apparently no disadvantage ; for when two different duties are performed by the same organ, it rarely performs both equally well. Here they become mere receptacles of prepared food, the nature and office of which is the same as of the albumen, or nutritive deposit exterior to the embryo in what are called albuminous seeds. (25-27.) The difference is in the place rather than in the character of the deposit. The plumule in such cases is always apparent before germination ; and it develops even with more vigor than in the preceding cases. It usually rises as a stout stem of several internodes lengthen- ing almost simultaneously, or at least the upper strongly developing long before the lower have finished their growth ; and the latter are practically leafless, bearing only small and scale-like and useless ru- diments of leaves. This is correlated with the peculiarity that the caulicle does not lengthen in germination, or it lengthens very slightly ; the cotyledons remain within the coats of the seed ; and if this were buried beneath the surface of the ground, there it remains. The abortion of the earliest leaves of the plumule is in correlation with this hypogceous (i. e. underground) situation of the cotyle- dons throughout the germination. The slight elongation of the caulicle serves merely to protrude its root-end from the coats of the seed in a downward direction, and from this a strong root usually is formed. FIG. 39. Section of an acorn, filled by the embryo. 40. Advanced germination 01 the same. 20 MOiirHOLOGY 33. In some Oaks, notably in our Live Oak (Quercus vircns), and less so in the Horsechestnut, the two cotyledons coalesee or cohere by their contiguous faces. In some of these cases of hypo- gaeous germination, the short caulicle and plumule are extri- cated from the enclosing coats or husk, by the development of short stalks (petioles, 157) to the fleshy cotyledons ; as is seen in Fig. 42, and in most germinating acorns. These petioles are not visible in the seed, but are the first develop- ment in germination. 34. There are some curious cases in which, while the caulicle remains short and subterranean, the cotyledons are raised out of ground in germination by the formation of far longer stalks (petioles) than those of the 42 Horsechestnut. A singularly dis- guised instance of this kind is seen in Megarrhiza, a genus of Cucurbitaceous plants of California and Oregon, remarkable for their huge root. The large seed has very thick and fleshy cotyledons, and a very short and straight caulicle. In germi- nation, the whole seed is elevated, seemingly in the manner of the bean, upon a stout stem. One waits for a long time expect- ing to see the cotyledons throw off the bursting husk and expand, or else to put forth the plumule from between their bases. But at length the plumule makes its appearance from an unexpected place, coming separately out of the soil. Removing this, the state of things represented in Fig. 43 is presented, — that of the plumule seemingly originating from the base, instead of the apex, of an elongated caulicle ! But on examination of the cleft from which this proceeds, by making a section of the stem above (showing that it is hollow), and finally by separating the cotyle- dons and gentty tearing apart the two short stalks by which they are united to their stem-like support, it is found that the latter may be divided into two (as shown in Fig. 44) , even down to the cleft below. This explains the anomaly. The real caulicle has re- F1G. 41 Section of a Horsechestnut or Buckeye Feed, through the very thick cotyledons and the incurved caulicle. 42. Seed In germination, showing the petioles to the cotyledons, &c. OF THE EMBRYO AND SEEDLING. 21 mained short and subterranean, and is confluent with the upper part of the thickening root : the seeming caulicle, which raised the cotyledons above the soil, consists of the petioles of these combined into a tubular stem-like bod}-, no evident trace of which is visible in the seed, although in germination it attains the length of two or three inches : in age it is readily separable into the two leaf-stalks or petioles of which it is composed : the plumule is thus seen to be wholly normal, originating from between the cotj'ledons. All the ex- tensive growth so far, and until the proper foliage-leaves of the continu- ation of the plumule are developed and begin their action, is from nutri- tive material stored in the thickened cotyledons, a considerable part of which was transferred to the already enlarging root, before a remaining portion was used in building up the strong plumule. The economy of this elevation of cotyle- dons which never open, and of the lengthened distance through which the nutritive matter has to be carried, is not apparent. But it is the family habit in Cucurbitaceae to bring up the cotyle- dons that they may develop as leaves (as in the Pumpkin, Fig. 47) : here this elevation is brought about in a different way, but without securing the useful end.1 1 It may be inferred that Megarrhiza is a descendant of some Cucurbitacea with thinner cotyledons, which in germination developed into long-stalked leaves, in the manner described in the next following paragraphs. FIG. 43, 44. Peculiar germination of Megarrhiza Californica; explained above. 22 MORPHOLOGY 35. This same anomaly, as to the development of long stalks to the cotyledons and their union into a stem-like body, occurs in various species of Larkspur (notably in the Californian Delphin- ium nudicaule) ; but in these the cotyledons develop into a pair of efficient green leaves. 45 47 3G. A similar elongation of petioles of the cotyledons, but without any union, occurs in a species of Morning Glor}- of the plains be}Tond the Mississippi (Ipomoea leptophylla) ; the leaf- like cotyledons coming up on their long stalks separately from the ground (Fig. 45) ; the developed plumule rising somo time afterward between them. Compare this with the ordinar}' species (25, Fig. 15, 16, 25), and note that the difference is merely that the caulicle in the common Morning Glory elongates and the petioles of the cotyledons remain short. 37. In all instances thus far a single primary root so regularly develops from the lower end of the axis of the embryo (variously named radicle or caulicle) , and forms such a direct downward FIG. 45. Germination of Ipomoea leptophylla; the caulicle not developing, the plumule and the petioled cotyledons rise from underground. Dotted line marks the level of the soil. FIG. 46. Embryo of a Pumpkin, the cotyledons separated. 47. Same germinated ; a cluster of roots from the base of caulicle. OF THE EMBRYO AND SEEDLING. 23 prolongation of it, that it was called the descending axis ; and the body from which it originates was named the radicle, on the supposition that it was itself the nascent root. But, as already explained, the so-called radicle grows in the manner of stem (24), and is morphologically that initial internode the node of which Dears the first leaves or cotyledons. (20.) Let it now be noted that this descending axis or single primary root is far from universal. In Pumpkin, Squash, Echinocystis, and the like, the strong caulicle sends out directly from its root-end a cluster of roots or rootlets, of equal strength; i. e., it strikes root in nearly the manner that a cutting does. (Fig. 47.) 38. The Polycotyledonous Embryo is one having a whorl of more than two seed-leaves. The dicotyledonous embryo being a whorl of the very simplest kind, that is, with the members reduced to two, the pobycotyledonous may be regarded as a variation of it. In all but one group of plants it is simply a variation, of casual occurrence, or even a monstrosity, in which three or rarely four cotyledons appear instead of two. In Pines (Fig. 48, 49), however, and in most but not all Coniferse, a whorl of from 3 to 10 cot}'ledons is the normal structure, varying accord- ing to the species, but of almost uniform number in each. In germination these are brought out of \| the soil by the elongation of the caulicle, and when the husk of the seed is thrown off they expand into a circle of needle-shaped leaves. In the Pine tribe, all the subsequent leaves are alternate (spiral) in arrange- ment, with some disguises. In the Cypress tribe, the cotyledons are fewer (not more than four, and more commonly only two) , and the subsequent leaves also are in whorls of two to four ; i. e., are either opposite or verticillate. From the occasional union at base of the cotyledons of a polycot}Tledonous embryo in pairs or groups, and from a study of their early development, Duchartre 1 plausibly maintains that such cotyledons really consist of a single pair, parted into divisions or lobes. The ordinary- interpretation, however, is equally tenable. 39. The Monocotyledonous Embryo, although theoretically the simplest, is practically a more difficult study. It has a single cotyledon (as the name denotes) ; also a single leaf to each node 1 Ann. Soi. Nat. ser. 3, x. 207. This view, which originated with Jussieu, is adopted by Parlatore in DC. Prodr. xvi. FIG. 48. Section of a seed of a Pine, witli its embryo of several cotyledons. 49. Early seedling Pine, with its stem let, displaying its six seed-leaves. 24 MORPHOLOGY of the plumule ; that is, the leaves of the embryo are alternate. But the caulicle is usually very short, and there is no external mark by which its limits may be distin- guished from the cotyledon, until germi- nation has begun. For a type of it, the embryo of some aquatic or marsh plants \f \A |N^H| may be taken, where it forms the whole _ kernel of the seed (FiS- 50-53), and 'af si 52 68 the structure can be made out antecedent to germination. It is understood by supposing that the cotyle- don, which forms its principal bulk (the caulicle being only the very short thickish base) , is convolute around a short plumule, and the margins concreted, except a minute longitudinal chink at base, out of which the growing plumule protrudes in germination. The embryo of Iris may be similar in structure, but no distinction of parts is visible. It is very small in proportion to the size of the seed, the kernel being mostly albu- men, — a supply of food, from which the germinating embryo draws the materials of its growth. When this takes place, either the cotyledon or the whole embryo lengthens, its lower part is pushed out of the seed, a root forms at the free end of the excessively short caulicle, and the plumule develops from the other in a series of one-leaved nodes, the inte modes of which remain so short that the leaves continue in close contact, the bases of the older successively enclosing the inner and younger. (Fig. 55.) Here, therefore, the cotyledon mainly remains in the seed, and the seed remains underground (hypogseous) . 40. It is somewhat different in the Onion, which has a similar embryo, except that it is longer, and the cotyledon is curved in the albumen of the seed. The first steps are the same as in Iris ; but as soon as a root is formed and embedded in the soil, the cotyledon lengthens vastly more, into a long and filiform green leaf, which, taking an erect position, FIG. 50. Seed of Triglochin palustre ; the rhaphe, leading to the strong chalaza at the summit, turned towards the eye. 51. The embryo detached from the seed-coats, showing the longitudinal chink at the base of the cotyledon ; the short part below is the radicle. 52. Same, with the chink turned laterally, and half the cotyledon cut away, bringing to view the plumule concealed within. 53. A cross-section through the plumule, more magnified. FIG. 54. Section of seed of Iris, enlarged, showing the small and apparently simple embryo at the base of the albumen 55 Germinating seed and seedling of the same, of natural size. OF THE EM13JRYO AND SEEDLING. 25 carries up the light seed far above the surface of the ground, the tip only remaining in the albumen of the seed until that is ex- hausted, when the tip perishes and the emptied husk falls away. About this time the plumule shoots forth from one side of the subterranean base of this cotyledonar leaf, in the form of a second and similar filiform leaf, to be followed by a third, and so on. The sheathing bases of these succeeding leaves become the coats of the Onion-bulb. The internodes remain undeveloped until the plant is ready to blossom. Very similar is the germination of a date-seed, except that the protruding cotyledon does not lengthen so much, nor does it elevate the heavy seed. Instead of the seed being carried up, the lower end of the embryo, contain- ing the plumule, is pushed down more or less into the loose soil, from which in time the developing plumule emerges. 41. The embryo of Grasses, especially of those which yield the cereal grains, is more complex, owing mainly to the great de- velopment of the plumule and the manner in which its rudimentary leaves successively enclose each other. That of Maize or Indian Corn, one of the t ^ r ^c — /•" 59 60 61 The floury part of the seed, which makes most of its bulk, is the albumen, largely composed of starch. The embryo is exterior to this, applied to one of its flat sides, and reaching from the thinner edge to or above the middle in the common variety of corn here represented. The form of the embryo is best shown, detached entire, in Fig. 58 : its structure appears in the sections. The outer part is the cotyledon, which incompletely enwraps the plumule : it adheres closely to the albumen by the whole back, and remains un- changed in germination : its function is to absorb nutritive largest, is most convenient for study. (Fig. 56-59.) FIG. 56. Section, flatwise, of a grain of Indian Corn, dividing the albumen and the embryo. 57. Similar section at right angles to the first. 58. A detached embryo: corresponding parts of Fig. 57 and 58 indicated by dotted lines. FIG. 59. Vertical section of Indian Corn across the thickness of the grain, dividing the embryo through the centre and displaying its parts: c, cotyledon; pt plumule; r, the radicle or caulicle. FIG. 60. Similar section of grain of rice. 61. Same of an oat-grain; the parts as in Fig. 59. 26 MORPHOLOGY matter furnished by the albumen, and to transmit it to the growing plumule. The plumule consists of a succession of rudimentary leaves, sheathing and enclosing one another, on the summit of a very short axis, which is mainly the caulicle, otherwise called rad- icle. This is completely en- closed by a basal portion of the cotyledon and of the outermost leaf of the plu- mule, which form a peculiar sheath for it, named the Coleorhiza* i. e. root-sheath : consequently the first root or roots have to break through this covering. As in the Oak and Pea (3 2), the very first or outermost leaves of the plu- mule develop imperfectly and not into efficient foliage. The one in Fig. 62, which encloses the rest in the early growth, is left behind as a mere sheath to the base of the following and more perfect leaves : it is the same as the lowest in Fig. 63. The leaves are first developed : the internodes lengthen later, and the lowest lengthen very little. Not rarely the first root starts singly from the tip of the caulicle (Fig. 62, just as in Fig. 55) ; but others of equal strength follow from &ny part of the caulicle, and soon from the nodes above ; and no tap-root is ever formed. 42. A Pseudo-monocotykdonous emb^-o occasionally occurs ; that is, one of the dicotyledonous type, of which one cotyledon is wanting through abortion. This occurs in Abronia, a genus related to Mirabilis, and bearing an embryo very similar to that represented in Fig. 17, 18, except that one cotyledon is absent. The anomaly of an acotyledonous embryo occurs in Dodder, a plant of the dicotyledonous t}~pe, but with both cotyledons 1 This, the Coleorhize of Mirbel, should not be confounded (as by some it has been) with the "root-cap," or tissue which ordinary roots (whether primary or secondary) break through in their development or carry on their apex. FIG. 62. Early germination of Indian Corn. 63. More advanced germination of same : roots produced from portion of stem above the cotyledon as well as below. OF THE ORGANS OF VEGETATION. 27 actually wanting, — a correlation with its parasitic mode of life. (64, Fig. 78.) 43. The dicotyledonous and the monocotyledonous character of the embryo is correlated with profound differences in the whole ulterior development, as revealed in the structure of the stem, leaves, and flower ; which differences mark the two great divisions of Phaenogamous plants, viz. DICOTYLEDONES or DICOTYLEDONOUS PLANTS, and MONOCOTYLEDONES or MONOCOTYLEDONOUS PLANTS, — names introduced into classification by Ray, and adopted by A. L. Jussieu, in his Genera Plantarum. CHAPTER III. MORPHOLOGY AND STRUCTURE OF THE ORGANS OF THE PLANT IN VEGETATION. SECTION I. OF THE ROOT. 44. The Root, which has been called the descending axis, is that portion of the body of the plant which grows downward, ordinarily fixing the vegetable to the soil, and absorbing from it materials which the plant may elaborate into nourishment. As alread}* stated (24), the root grows in length by continuous additions of new fabric to its lower extremity, elongating from that part only or chiefly ; so that the tip of a growing root always consists of the most newly formed and active tissue. It normally begins, in germination, at the root-end of the caulicle, or so called radicle. But roots soon proceed, or may proceed, from other parts of the stem, when this is favorably situated for their production. The root does not grow from its naked apex, but from a stratum immediately behind it : consequently its blunt or obtusely conical advancing tip consists of older, firmer, and in part effete tissue. The tip of all secondary roots and rootlets FIG. 61. Magnified tip of root of a seedling Maple (such as in Fig. 9), sufficiently enlarged to indicate the cellular structure: a. the portion where growth is taking place; 6. the older ami firmer tip. 28 MOKPHOLOGY OF THE HOOT. is similarly capped or protected.1 But the so-called root-cap is seldom so distinct or separable as to deserve a particular name. 45. Nature of Growth, Cells. The development and growth of the root, as of other organs, results from the development, growth, and increase in number of certain minute parts, of which the plant is built up. These component parts are so much alike, at least in an early stage, and are so obviously formed all on one type, that they take one common name, that of CELLS. These are the histological elements of plants, i. e. the units of minute anatomical structure. While, in the morphology of the plant's obvious organs, analysis brings us to the phytomer (16) as the individual element which by a kind of propagation produces its like in a second phytomer, remaining however in connection with the" first, thus building up the general structure, so, in an analogous way, each of the obvious parts — each stalk or blade or rootlet — is microscopically determined to be com- posed of these ultimate organic units, generally called cells. The cell (cellula, by the French conveniently termed cellule) is the living vegetable unit, in the same sense that the brick is the unit of a brick edifice. To make this analogy fairly complete, the bricks should be imagined to have a firm exterior or shell, and a soft or at length hollow - interior, also to be living when incorporated into the structure, and finally to be produced in the forming structure by a kind of propagation. The production or increase in number of these cells by development from previous ones, and their successive increase in size up to maturity, are what constitutes vegetable growth.2 The inspection through a 1 The notion that the tip of the root consists of delicate forming or newly formed tissue, or bears some organ or structure of this nature (a " S/m»giole "), has hardly yet been eliminated from the text-books and popular writings. It had no proper foundation in fact. In Lemna, and in some other aquatics, and also in some aerial roots, this older tissue often separates into a real root-cap, free at base, like an inverted calyptra. 2 This, as to the structure, is the subject of Histology ; as to processes or actions, the subject of Physiology; both to be treated in a separate volume. FIG. 65, 66. Portions of surface of Fig. 64, more magnified, clearly displaying the Riiperftcfal cellular structure and the long processes from some of the cells, called root- h.iii-s, which abound on the upper part of Fig. 64. MORPHOLOGY OF THE BOOT. 29 simple microscope of a slender j'oung root, and of thin slices of it immersed in water, may serve to give a general though crude idea of the vegetable cellular structure, sufficient for the present purpose. Roots are naked ; that is, they bear no other organs. When they send off branches, these originate from the main root just as roots originate from the stem ; and in both cases without much predetermined order. The ultimate and veiy slender branches are sometimes called root-fibrils ; but these are only delicate ramifications of the root. Like any other part of the plant, however, roots may produce hairs or such like growths from the surface, which are wholly distinct from branches. (383.) ~7> * of 46. Root-hairs. Roots absorb water, &c., from the soil by imbibition through the surface ; that is, through the walls of the cells, which are in a certain sense permeable to fluids, more readily when young and tender, less so when older and firmer. Roots, therefore, absorb most by their fresh tips and adjacent parts ; and these are continuall}- renewed in growth and extended fur- ther into the soil. As the active surface of a plant above ground is enormously increased by the spread of foliage, so in a less degree is the absorbing surface of }'oung roots increased b}- the production of root-hairs. (Fig. 64, upper part, and more magni- fied in Fig. 65, 66.) These are attenuated outgrowths of some part of the superficial cells into capillary tubes (only one from each cell) , closed at the tip, but the calibre at base continuous with the cavity of the cell ; into which, therefore, whatever is imbibed through the thin wall may freely pass. These appear (as Fig. 64 shows) at a certain distance behind the root-tip. Further back the older or effete root-hairs die away as the cells which bear them thicken into a firmer epidermis. 47. To the general statement that roots give birth to no other organs, there is this abnormal, but by no means unusual excep- tion, that of producing buds, and therefore of sending up leafy branches. Although not naturally furnished with buds in the manner of the stem, yet many roots have the power of originat- ing them under certain circumstances, and some produce them habitually. Thus Apple-trees and Poplars send up shoots from the ground, especially when the superficial roots arc wounded. And the roots of Maclura or Osage Orange so readily originate buds that the tree is commonly propagated by root-cuttings. 48. Kinds of Roots. The root, commonl}- single, which origi- nates from the embryo itself, is called the PRIMARY ROOT. (37.) Roots which originate from other and later parts of the stem, or elsewhere, are distinguished as SECONDARY ROOTS. But the 30 MORPHOLOGY OF THE ROOT. latter are as normal as the primary root ; that is, to stems so situated that they can produce them. Most creeping plants emit them freely, usually from the nodes ; and so do most branches, not too old, when bent to the ground and covered with earth, thus securing the requisite moisture and darkness. Separate pieces of young stems (cuttings) can commonly be made to strike root. Upon this faculty of stems to originate roots depends all propagation by division, by laying or layering, b^' cuttings, &c. It is mainly annuals and common trees that naturally depend on the primary root ; and most of these can be made to produce secondary roots. Even leaves and leaf-stalks of some plants may be made to strike root and be used as cuttings. (77.) 49. Duration. By differences in respect to this, either the root or the plant, as the case may be, is distinguished into Annual, Biennial, or Perennial, according to whether life is contin- ued for a single year or season, for two, or for a greater number. The difference is not in all cases absolute or even well marked. 50. Annuals are plants which, springing from the seed, flower and seed the same }'ear or season, and die at or before its close. They produce fibrous roots, either directly from the embryo and succeeding joints of stem (as in Grasses, Fig. 63), or from a persistent primary or lap-root, more or less thickened into a trunk or divided into branches. The products of vegetation in all such herbs are not stored in subterranean or other reservoirs, but are expended directly in new vegetative growth, in the production of blossom, and finally in the maturation of fruit and seed. This completed, the exhausted and not at all replenished indi- vidual perishes. 51. But some annuals may have their existence prolonged b}T not allowing them to blossom or seed. Others, with prostrate stem or branches, may from these produce secondary roots, which, forming new connections with the soil, enable the newer growth to survive when the older parts with the original root have perished. And many herbs, naturally annuals, are continued from year to year through such propagation from the branches, used as la}Ters or cuttings. Moreover, certain plants (such as Ricinus or Castor-oil Plant) , which are perennial or even arbo- rescent in warm climates to which they belong, become annuals in temperate climates, early perishing lay autumnal cold. 52. The annuals of cool climates, where growth completely ceases in winter, germinate in spring, mature, and die in or before autumn. But, in climates with comparatively warm and rainy winter and rainless summer, many germinate in autumn, vegetate MORPHOLOGY OF THE BOOT. 31 through the winter, flower and seed in spring, and perish in early summer. These ma}- be termed WINTER ANNUALS. 53. Biennials are plants which, springing from the seed and vegetating in one season, live through the interruption of winter, and blossom, fructif}', and perish in the next growing season ; their life being thus divided into two stages, the first of vegeta- tion, the second of fructification. In typical biennials, nearly the whole work of vegetation is accomplished in the first stage, with the result of accumulation of a stock of nutritive matter, to be expended in the second stage in the production of blossom and seed. This accumulation is usually stored in the root or in the base of a very short stem in connection with the root. The root of a biennial accordingly enlarges and becomes flesh}", or obese, as this matter accumulates. At the close of the growing season, — the leaves perishing and the stem having remained very short (with undeveloped mternodes), — the root, crowned with the bud or buds, contains the main result of the summer's work, as provision for the next year's devel- opment and the completion of the C}'cle. This development, being thus amply provided for, is undertaken in spring with great vigor ; blossom, fruit, and seed are rapidly produced ; and the stock being consumed, but not at all replenished, the cells of the great root are now empty and effete, and the individual perishes. The Beet, Turnip, Parsnip, and Carrot are fa- miliar examples of biennials, with the store of nourishment in the root.1 The Kohl-rabi is a biennial with this deposit in the stem: the Cabbage, partly in the stem, partly in the head of leaves. 1 In tliese the caulicle enlarges with the root, so that the upper and bud-bearing end is stem. Tap-roots of this kind are said, in descriptive botany, to be . Fusiform or Spindle-shaped, when broader in the middle and tapering towards both ends, as in the common Radish (Fig. 67); Conical, when tapering regularly from base to tip, as in carrots, &c. ; Napifarm, i. e. Turnip-shaped, when the thickened part is wider than high, &c. Fascicled Roots are those which form in clusters ; these may be slender or thickened. When much thickened, either irregularly or not of the above shapes, they arc said to be tuberous. FIG. 67. Radish : a fusiform lap-root. 32 MORPHOLOGY OP THE BOOT. 54. But some plants, such as the Radish, which when they spring from seed in autumn are true biennials, will when raised in spring pass on directly to the flowering stage in summer, or when sown after tlae warm season begins will often run through their course as annuals. Then there are various biennials which thicken the root very little and hold their leaves through the winter. Between these and winter annuals no clear demarcation can be drawn. As respects annual and biennial duration, the terms may for the most part be applied indiscriminately to the plant or to the root. We may say either that the plant is a biennial, or that its root is biennial. 55. Perennials are plants which live and blossom or fructify year after year. They may or they may not have perennial roots. In trees and shrubs, also in herbs with growth from year to year from a strong tap-root, the root is naturally perennial. But in most perennials with only fibrous roots, these are produced anew from time to time or from year to year. Also, while some such roots remain fibrous and serve only for absorption, others may thicken in the manner of the ordinarj- biennial root and serve a similar use, *'. e. become reservoirs of elaborated nourishment. The Dahlia (Fig. G8) and the Peony afford good examples of this. Sweet potato is another instance.1 Most such roots have only a biennial duration : they are produced in one growing season ; they yield their store to form or aid the growth of the next. When perennials store up nutritive matter underground, the deposit is more commonly made in a subterranean portion of the stem, in tubers, corms, bulbs, &c. (See 115-122.) 56. The distinction between annuals and biennials is at times so difficult, and the particular in which they agree so manifest, — namety, that of blossoming only once, then dying, as it were by exhaustion, — that it was proposed by DeCandolle to unite 1 It is only by the readiness of this root to produce adventitious buds, especially from its upper part, that it has been mistaken for a tuber, such as the common potato. FTO. 68. Fnsdcled and tuberous or fusiform (secondary) roots of Dahlia : a, a. buds on base of the stem. MORPHOLOGY OF THE ROOT. 33 the two under the common appellation of MONOCARPIC plants, Planttz monocarpicce, taken in the sense of only once-fruiting plants ; and to designate perennials by the corresponding term of POLYCARPIC, Plants polycurpica, literally many-fruited, taken in the sense of many-times fruiting.1 57. But the distinction even here is no more absolute than that between annuals and biennials. For example, it is not quite clear whether the Cardinal Flower and related species of Lobelia should be ranked as annuals, biennials, or perennials. The plants may blossom and seed toward the end of the season in which they came from seed ; or, germinated in autumn, the small seedlings may survive the winter ; but whenever fructified the fibrous-rooted mother plant dies throughout ; yet usually not before it has established, and perhaps detached from the base, small offsets to blossom the next season ; and so on. Then Houseleeks (Sempervivum) and such-like fibrous-rooted succu- lent plants multiply freely by offsets which are truly perennial in the sense that they live and grow for a few or several years ; but when at length a flowering stem is sent up producing blos- som and seed, that plant dies as completely and in the same manner as any biennial, only the generation of offsets surviving. The same is true of the Century plant (Agave Americana, wrongly denominated American Aloe), which vegetates in the manner of the accumulating stage of a biennial, except that this continues for several or very many years, while the flower- ing stage, when it arrives, is precipitated and terminated in a single season. 58. Although the stem usually sends forth roots only when covered by or resting on the soil, which affords congenial dark- ness and moisture, yet these are in some cases produced in the open air. Roots may likewise subserve other and more special uses than the absorption of crude or the storing of elaborated nourishment. 59. Aerial Roots is a general name for those which are pro- duced in the open air. One class of these may serve the office of ordinary roots, by descending to the ground and becoming established in the soil. This occurs, on a small scale, in the stems of Indian Corn ; the lower nodes emitting roots which grow to the length of several inches before they reach the ground 1 These terms or some equivalents have a convenience in descriptive botany. But those employed by DeCandolle are not happily chosen, as lias often been said. MontStocous (bearing progeny once) and Polytocous (bearing many times) would be more appropriate. 34 MORPHOLOGY OP THE KOOT. into which they penetrate. More remarkable cases abound in those tropical regions where the sultry air, saturated with moist- ure for a large part of the }'ear, favors the utmost luxuriance of vegetation. In the Palm-like Pandanus or Screw-Pine l (Fig. 69), very strong roots, emitted in the open air from the trunk, and soon reaching the soil, give the appearance of a tree partially raised out of the ground. The famous Banyan-tree of India (Fig. 71) is a still more striking illustration ; for the aerial roots strike from the horizontal branches of the tree, often at a great height, at first swing- ing free in the air, but finally reach- ing and establishing themselves in the ground, where they increase in diameter and form accessory trunks, surrounding the original bole and supporting the wide-spread canopy of branches and foliage. Very similar is the economy of the Man- grove (Fig. 70) , which forms impenetrable thickets on low and muddy sea-shores in the tropics throughout most parts of the world, extending even to the coast of Florida and Louisiana. Here aerial roots spring not only- from the main trunk, as in the Pandanus, but also from the branchlets, as in the Banyan. Even the radicle of the embryo starts into growth, protrudes, and attains considerable length while the fruit is still attached to the branch. 59". Aerial Rootlets for climbing arc familiar in the Ivy of the Old World (Hedera) , Trumpet-Creeper (Tecoma radicans) , and our Poison Ivy (Rhus Toxicodendron) ; by the adhesion of 1 So named, not from any resemblance to a Pine-tree, but from a like- ness of the foliage to that of a Pine-Apple. FIG. 69. Pandanus, or Screw-Pine; and in the background, 70, a Mangrove-tree (Rhizophora Mangle). MORPHOLOGY OB" THE HOOT. o5 which the stems, as they grow, ascend walls and the trunks of trees with facility. In Khus a superabundance of these rootlets is produced, thickly covering all sides of the stem. 60. Epiphytes or Air-Plants also have roots which are through- out life unconnected with the ground. Epiphytes, or Epiphytic plants, as the name denotes, are such as grow upon other plants without taking nourishment from them. Deriving this from the air alone, they are called Air-plants. This name might be extended to the same or other kinds of plants attaching them- selves to bare walls, rocks, and the like, and unconnected with the soil, though such would not technically be epiphytes. Vciy many Lichens, Mosses, and other plants of the lower grade, and not a few phaenogamous plants, are in this case. The greater part of the phaenogamous Epiphytes pertain to two monocotyle- donous orders, the Orchis family and that to which the Pine- Apple belongs, viz. the Bromeliacese. Their thread-like or cord-like simple roots either adhere to the bark of the supporting tree, securing the plant in its position, or some hang loose in the air. Of these, Orchids, i. e. plants of the Orchis familv, are the most showy and numerous, and of the greatest variety of forms, especially of their blossoms, which are often bizarre and fantas- tic. They belong, naturally, to climates which are both warm and humid ; they are highly prized in hot-house cultivation ; and, along with the hardy and terrestrial portion of the order, they are peculiarly interesting to the botanist on account of the singular and exquisite adaptation of their flowers in relation to insects which visit them. In some the blossoms curiously FIG. 71. The Banyan-tree, or Indian Fig (Ficus Inclica). 6b MORPHOLOGY OF THE ROOT. resemble butterflies or other insects ; as, for example, Oncidium Papilio, Fig. 72. Epiphytic orchids are indigenous to the United States only from Georgia to Texas, and only in humble forms, in company with species of Tillandsia, representing Bromeliace- ous epiphytes. The commonest of the latter tribe, and of most northern range, is the T. usneoides, the so-called Long Moss, which, pendent in long and tangled gray clusters or festoons from the branches of the Live-Oak or Long-leaved Pine, gives such a peculiar and sombre aspect to the forests of the warmer portions of our Southern States. 61. Parasitic Plants have the peculiarity that their roots, or what answer to roots, not only fix themselves to other plants, but draw therefrom their nourishment, at least in part. Among ciyptogamous plants very man}1 Fungi are parasitic upon or within living plants or animals. But only phaenogamous para- sites are here under consideration. These may be divided into two classes ; those with and those without green foliage. 62. Green Parasites may be either wholly or partially parasitic ; that is, they may draw all their support from a foster plant, or FIG. 72. Oncidium Papilio, and, 73, Comparettia rosea; two showy epiphytes of tin Orchis family; showing the mode in which these Air-plants grow. MORPHOLOGY OF THE HOOT. 37 they may be likewise rooted in the soil, and receive from it materials of their food. Having green foliage, they are capable of elaborating such food, whether taken directly from the soil or from the crude sap of the foster plant. The Mistletoes (Viscum and its allies) are the principal examples of complete green parasitic plants. Seeds dropped by birds on the boughs of trees germinate there ; the root-end of the caulicle points thither instead of towards the earth ; the root, or what would be such, pene- trates the bark and in- corporates itself with the sap-wood so perfectly that the junction of par- asite with foster trunk is like that of branch with parent trunk. The parasite is probably fed by both elaborated and crude sap, that is, both by what the foster tree has assimilated and what it has merely taken from the soil and air : the former it can at once incorporate ; the latter it has first to assimilate in its own green leaves. Sometimes one Mistleto is parasitic upon an- other of the same or of a different species. 63. Partially parasitic plants (mostly green) may be either wood}' and arborescent or herbaceous. The species of Clusia in tropical FIG. 74. Native epiphytes of Georgia, &c.: the erect one at the right an Orchid, Epidendrum eonopseum ; the hanging one Tillandsia usneoides, called Long Moss. FIG. 75- Roots of Gerardia flava: some of the rootlets attaching themselves para- sitically to the root of a Blueberry. {From a drawing by Mr. J. Stauffer.) 38 MORPHOLOGY OF THE ROOT. America (called Cursed Fig) are examples of the former. They form trees, send down aerial roots in the manner of the Banyan ; but, while some roots seek the ground, some may attach them- selves to other trees parasitically, and draw from them a portion of their support. The parasitism of certain herbaceous plants with green foh'age is clandestine, the connection being under- ground and therefore long unsuspected. This occurs in species of Gerai'dia (at least of the section Dasystoma) and other plants of the same family, the uncultivability of which is thereby explained. Also in Comandra and in their relatives the Thesiums of the Old World, belonging to a natural order (the Santalaceae) which has much affinity with the entirely parasitic order (Loranthacere) to which the Mistleto belongs. 64. Pale or Colored Parasites, such as Beech-Drops, Pine-Sap, &c., are those which are destitute of green herbage, and are usually of a white, tawny, or reddish hue ; in fact, of an}" color except green. These strike their roots, or sucker-shaped discs, into the bark, mostly that of the root, of other plants, and thence draw their food from the sap alread}7 elaborated. They have accordingly no occasion for digestive organs of their own, i. e. for green foliage. The Dodder (Fig. 77) is a common plant of this kind which is parasitic above ground. Its seeds germinate in the earth, but form no proper root : when the slender twining stem reaches the surrounding herbage, it forms suckers, which attach surface of the supporting plant, penetrate its epidermis, and feed upon its juices ; while the original root and base of the stem perish, and the plant has no longer any connection with the soil. Thus stealing its nour- FIG. 76. Section of one of the attached rootlets of Gerardia, showing the union. FIG. 77. The common Dodder of the Northern States (Cuscuta Gronovii), of the natural size, parasitic upon the stem of an herb : the uncoiled portion at the lower end shows the mode of its attachment 78. The coiled embryo taken from the seed, con- sisting of naked caulicle and plumule; moderately magnified. 79. The same in germi- nation, elongating into a thread-like leafless stem. themselves MORPHOLOGY OF THE ROOT. 39 ishment ready prepared, it requires no proper digestive organs of its own, and, consequently, does not produce leaves. This economy is foreshadowed in the embryo of the Dodder, which is a naked thread spirally coiled in the seed (Fig. 78, 79), and presenting no vestige of cotyledons or seed-leaves. A species of Dodder infests and greatly injures flax in Europe, and some- times makes its appearance in our own flax-fields, having been introduced with the imported seed. Such parasites do not live upon all plants indiscriminately, but only upon those whose elaborate juices furnish a propitious nourishment.1 Some of them are restricted, or nearly so, to a particular species ; others show little preference, or are found indifferently upon several species of different families. Their seeds, in some cases, it is said, will germinate only when in contact with the stem or root of the species upon which they are destined to live. Having no need of herbage, such plants may be reduced to a stalk bearing a single flower or a cluster of flowers, or even to a single blossom developed from a bud directly parasitic on the bark of the foster plant. Of this kind are the several species of Pilostyles (para - sitic flowers on the shoots of Leguminous plants) in Tropical America, one species of which was discovered by Dr. Thurbcr near the southern borders of New Mexico. Its flowers are small, only about a quarter of an inch in diameter. The most wonderful plant of this kind is that vegetable Titan, the Raf- flesia Arnoldi of Sumatra (Fig. 80) which grows upon the stem of a kind of a Cissus or Vitis. It is a parasitic flower, measuring nine feet in circumference, and weighing fifteen pounds ! Its color is light orange, mottled with j'ello wish- white. 1 Monotropa or Indian Pipe (and perhaps some related plants), although probably parasitic on living roots in early growth, appears to live afterwards ' in the manner of the larger Fungi, upon leaf-mould and decaying herbage. Its mode of life should be investigated. FIG. 80. Rafflesia Arnoldi; an expanded flower, and a bud, directly parasitic on the stem of a vine: reduced to the scale of half an inch to a foot. 40 MORPHOLOGY OP BUDS. SECTION II. OF BUDS. 65. Buds are the germs of stems : they are axes with their appendages in an early state. LEAF-BUDS (GEMMAE) are those devoted to vegetation, and the parts, or some of them, develop as leaves. MIXED BUDS contain both foliage and flower or flowers. FLOWER-BUDS (ALABASTRA) are unexpended blossoms. These are considered in another chapter. 66. The conspicuous portion of an ordinary bud, or that which first develops, usually consists of leaves, or scales the homologues of leaves ; the axis itself being very short and undeveloped. If this remains comparatively short, the leaves as developed are crowded in a rosette, as in a Houseleek (Fig. 91*), a Barberry and the Larch : when the inte modes lengthen, the leaves are interspaced upon the axis. 67. The cotyledons and plumule of the embiyo are, morpho- logically, the first bud, on the summit of the initial stem, the caulicle. This in germination and subsequent growth develops into a leaf}- stem, in the manner already described. Normally this stem has the capacit}' of growing on in this way from the apex or growing point, which is always potentially a bud, the apical or terminal bud (15). Sometimes it is merely potential, and there is no external structure visible until the new growth begins, or the bud is said to be latent. 68. But commonly, in plants that live from year to year, growth is divided into seasons or stages, with intervals of repose. In such cases, especially in trees and shrubs, instead of a continuous succession of foliage, the period of interruption is apt to be marked by the production of scales (Bud-scales, Perulce, etc.) of dry teguments, which serve to protect the tender rudiments or growing point within during the season of rest. This being the winter-season in cold climates, Linnaeus gave to such bud-cover- ings the common name of HIBERNACULUM. From the usually s^uamose (scale-like) character of this covering, such buds take the name of 69. Scaly Buds. Large and strong ones of this kind, such as those of Horsechestnut, Magnolia, Hickory, Lilac, &c., may be taken as the type of bud. The scales serve to protect the ten- der parts within against injury from moisture and from sudden changes in temperature during the dormant or earliest growing state. To ward off moisture more effectually, they are sometimes coated with a waxy, resinous, or balsamic exudation, as is con- spicuous on the scales of the Horsechestnut, Balsam-Poplar or Balm of Gilead, and Balsam-Fir. To guard against sudden MORPHOLOGY OF BUDS. 41 changes of temperature, they are often lined, or the rudimentary leaves within invested with non-conducting down or wool. 70. Nature of Bud-scales. That they answer to leaves is made manifest by a consideration of their situation and arrangement, which are the same as of the proper leaves of the species ; and by the gradual transi- tions from the former to the latter in many plants. In the Turions, or sub- terranean budding shoots of numerous perennial herbs, and in the unfolding buds of the Lilac and Sweet Buckeye (-ZEsculus parviflora) , every gradation may be traced between bud-scales and foliage, showing that no line of distinc- tion can be drawn between them, but that the two are essentially of the same nature, are different modifications of the same organ. In the Lilac they may be regarded as the blade of the leaf, modified and depauperate ; in the Buckeye (Fig. 233), and therefore in Horsechestnut, as the base of leaf-stalks ; in Magnolia (Fig. 81, 82), in the Tulip-tree, and in the Beech, they are evidently stipules. They must therefore be referred to in the section on the morphology of leaves. (227.) 71. Naked Buds, &c., of shrubs and trees, even in climates with severe winter, are not unknown, that is, buds unprotected by special scales or other coverings. For example, the latest pair of leaves of the season in Viburnum nuclum, V. lantanoides (Hobblebush) , and the like, remain in a nascent state over winter without covering, and ex- pand into the first foliage in the spring. Yet V. Opulus (Snowball, &c.), another species of the same genus and inhabiting the same region, has well-formed scaly * FIG. 81. Branch of Magnolia Umbrella, of the natural size, crowned with the terminal bud ; and below exhibiting the large rounded leaf-scars, as well as the rings or annular scars left by the fall of the bud-scales of the previous season. 82. A detached scale from a similar bud ; its thickened axis is the base of a leaf-stalk ; the membranous sides consist of the pair of stipules united with it. 42 MORPHOLOGY OF BUDS. leaf-buds. In other hardy shrubs and trees, the buds, equally or almost destitute of scales, are minute, hidden in or un- der the bark, or otherwise inconspicuous until vernal growth commences. Phila- delphus and Taxodium are of this kind. 72. Subpetiolar Buds. Some leaf-buds are singularly cov- ered in their early state and through the summer, as in the Locust (Robinia) , Honey-Lo- cust Fig. 96 (where they re- main very undeveloped), in Yellow Wood (Cladrastis), and more conspicuously in the Plane-tree (Platanus, Fig. 87) : here they are all formed un- der the base of the protecting leaf-stalk, which in Plane-tree ^5f V-J forms a sheath or inverted cup, very like a candle-extinguisher, fitted to and concealing the conical bud until autumn, when by the fall of the leaves these buds are exposed. FIG. 83. Diagram of vertical section of a strong bud, such as of Horsechestmit. 84. The axis of the same developing, the elongation beginning with the lowest inter- node, soon followed by the others in succession. 85. A year's growth of Horsechestmit, crowned with a terminal bud: a, scars left by the bud-scales of the previous year: b, scars left by the fallen leaf-stalks: c, axillary buds. FIG. 86. Branch and buds (all axillary) of the Lilac. FIG. 87. Leaf-bud under the petiole of the Plane-tree. MORPHOLOGY OF BUDS. 43 73. Fleshy Buds. Bulbs are peculiar buds of certain herba- ceous plants, with fleshy scales, and often of a more permanent character. Their nature and economy may most conveniently be illustrated under subsequent sections. Usually bulbs are subterranean or partly so. But small bulbs (Bulblets, 123) regu- larly appear in the axil of nearly all the leaves of certain common Lilies, being obviously ordinary axillary buds, under certain modifications. They become detached at maturity, fall to the ground, produce roots, and grow as independent plants ; and their fleshy scales are storehouses of nourishment for the early support of this independent growth. 74. Bud-propagation is a normal mode of reproduction in cases like the above, the spontaneously detached bulblets or buds establishing themselves as progeny. In several species of Allium (Onions and Leeks), such bulblets usurp the place of flower-buds, making the analogy seem closer. Stems or branches which habitually root in the soil, or along its surface, equally propagate or divide into new individuals, becoming distinct by the perishing of the older connecting parts, or by breaking away from them. Propagation by cuttings is an acceleration or exten- sion of this same natural operation. The cutting is a portion of stem bearing one or more buds, which, through the faculty of the stem to strike root, is made to grow independently. In grafting, such a cutting, and in budding a bud only, with a small portion of wood and bark, is transferred to the stem of another plant of the same or of some related species, and made to grow there, uniting its wood and bark with those of the stock, and so becoming a limb or branch, in place of striking root into the soil and becoming a separate plant. The horticultural advantage of bud-propagation is, that the offsets or new individuals share in all the peculiarities of the parent as completely as if still branches of that tree. In propagation by seed, the special peculiarities or excellencies of individuals or varieties may not, and in some measure probably will not, be reproduced. 75. Normal or Regular Buds, as to position, are either terminal or axillary, as already stated. (15.) They are single, that is, one bud normally occupies the apex of a stem or branch, and appears, or usually may appear, in the axil of (or upper angle formed with the stem by) any well-developed leaf. In these, positions, buds are so usual, or so capable of appearing, that they are commonly regarded as potential when not actually present. The potentiality may be manifested by the actual development of these buds in shrubs or trees after the lapse of years. (84.) The terminal leaf-bud is to continue the axis it 44 MOBPHOLOGY OF BUDS. surmounts : axillary and any other lateral leaf-buds are to be- come branches. But even of buds which actually appear a large proportion do not grow. When a terminal bud is formed (as _ ,~'Fig. 81, 85, 91), this is commonly the strongest, or among the stronger. But in many cases it habitually or commonly fails to appear. In the Elm (with leaves and therefore buds alternate) , the bud axillary to the last leaf of the season takes its place. In the common Lilac, a pair of buds, > which were in the axils of the uppermost of the (opposite) leaves, seem to replace the terminal bud, which seldom develops. (Fig. 86.) When all the regular buds make their appearance, and the leaves are opposite, the stem will be crowned with the terminal bud, having an axil- lary bud on each side of it. (Fig. 88.) 76. Accessory Buds. These are, as it were, multiplications of the regular axillary bud, giving rise to two, three, or more, instead 0 of one ; in some cases situated one above an« other (superposed}, in others placed side b}' side (collateral). In the latter case, which occurs occasionally in the Hawthorn, in cer- tain Willows, in the Maples (Fig. 88), &c., the axillary bud seems to divide into three, or itself to give rise to a lateral bud on each side. On some shoots of the Tartarean IIone3-suckle (Fig. 90) from three to six buds appear in each axil, one above another, the lower being successively the stronger and earlier produced ; and the one immediately in the axil, therefore, grows in preference : occasionally two or more of them grow, and superposed accessory branches result. It is much the same in Aristolochia Sipho, except that the uppermost bud is there strongest. FIG. 88. Branch of Red Maple, at the middle bearing triple axillary buds, placed side by side. FIG. 89. Piece of a branch of the Butternut, with accessory buds placed one above another: a, the leaf scar: b, proper axillary bud: c, (I, accessory buds. FIG. 90. Part of a branch of Tartarean Honeysuckle, with crowded accessory buds superposed in the axil of each leaf. MOiiPHOLOGY OF STEMS. 45 So it is in the Butternut (Fig. 89), where the true axillary bud is minute and usually remains latent, while the accessory ones are considerably remote, and the uppermost, which is much the strongest, is far out of the axil : this usually develops, and gives rise to an extra-axillary branch. 77. Adventitious Buds are such as are abnormal and irregular, being produced without order and from any part. of the stem, or even from roots. The latter, like the internodes of a stem, although normally destitute of buds, do produce them notwith- standing in certain cases, especially when wounded, and in some plants (such as Blackberries) so freely that gardeners propagate them by root-cuttings. The stems share this tendency ; and buds are apt to break out on the sides of trunks, especially when wounded or pollarded, or to spring from new tissues produced on cut surfaces, especially where the bark and wood join. Even leaves may develop adventitious buds, and then be used for propagation. In Bryophyllum, such buds, followed by rootlets, are freely produced on the margins of the blade or of its leaflets. In Begonia, a leaf, used as a cutting, will root from the base of the petiole stuck in the soil, and produce buds on the blade, at the junction with the petiole, or elsewhere. SECTION III. OF THE STEM. § 1. GENERAL CHARACTERISTICS AND GROWTH. 78. The Stem is the ascending axis, or that portion of the trunk which in the embryo grows in an opposite direction from the root, seeking the light, and exposing itself as much as pos- sible to the air. All phaenogamous plants possess stems.1 In those which, in botanical descriptions, are said to be acaulescent, or stemless, it is either very short, or concealed beneath the ground. Although the stem always takes an ascending direction at the commencement of its growth, it does not uniformly retain it ; but sometimes trails along the surface of the ground, or burrows beneath it, sending up branches, flower-stalks, or leaves into the air. The common idea, that all the subterranean portion of a plant belongs to the root, is incorrect. Equally incorrect is the common expression that plants spring from the root. Roots^ spring from the stem, not the stem from the root. (21, 24, 37, 44.) ~P (6. 1 There are, however, reduced forms in which there is no distinction of axis and foliage ; but most of these are clearly leafless rather than stemless, and not even in Lemna and Wolffia can the stem be said to be wanting. 46 MORPHOLOGY OF STEMS. 79. While the root normally gives birth to no other organs, but itself performs those functions which pertain to the relations of the vegetable with the soil, — binding it to the earth and absorbing nourishing materials from it, — the aerial functions of vegetation arc chiefly carried on, not so much b}- the stem it- self as by a distinct set of organs which it bears, namel}-, the leaves. Hence, the production of leaves is one of the charac- teristics of the stem. These are produced only at certain definite and symmetrically arranged points, called nodes. (13, 23.) 80. Development and Structure. In a bud or undeveloped stem, the nodes are in contact or close proximity. In the develop- ment, growth in length takes place in such manner as to carry these apart more or less, according to the degree of elongation, that is, the internodes (13) elongate. The order of development is from below upward, the lowest internode first lengthening, the others in regular succession. Each completes its growth, with more or lees rapidity, although the length attained varies greatly in different stems, in different parts of the same stem, and under different conditions. Unlike the root, in which the elongation of formed parts is very soon finished and therefore only the tip is perceptibly growing, internodes go on growing throughout, and several formed internodes may be growing simultaneously, thus producing elongation throughout a consid- erable extent of stem and with considerable rapidity. But each internode grows independently. Some parts of an internode may lengthen faster or continue in growth longer than others ; this is usually the upper portion, at least in long internodes and when every part is equally exposed to light. 81. The development of a stem from a bud is wholty like that from the embryo, and has already been described in Chap. II. It exhibits similar variations as to rapidity and vigor, dependent upon the constitution of the bud, — which, like the plumule in the seed or seedling, may be either latent or much developed before growth begins, — also upon the amount of nourishment provided. Strong buds commonly have their parts, or some of them, ready formed in miniature, and a store of elaborated nour- ishment in the parent stem to draw upon. Those well-developed buds which in many of our shrubs and trees crown the apex or occupy the axils of stem and branches early in the preceding summer (as in Magnolia, Fig. 81, Horsechestnut, Fig. 85, and Hickory, Fig. 91) often exhibit the whole plan and amount of the next year's growth ; the nodes, the leaves they bear, and sometimes the blossoms being already formed, and only requiring the elongation of the internodes for their full expansion. As CHARACTERISTICS AND GROWTH. 47 the bud is well supplied with nourishment in spring by the stem on which it rests, its axis elongates rapidly ; and although the growth commences with the lowest internode, yet the second, third, and fourth internodes may begin to lengthen long before the first has attained its full growth. Such very strong buds are usually terminal ; but sometimes, as in Lilac (Fig. 86), they are the uppermost axillary, which take the place of a suppressed or abortive terminal bud. 82. Such woody stems, developed from a strong bud, and terminated at the close of the season's growth by a similar bud, may be continued from year to year in an unbroken series. A set of narrow rings on the bark (Fig. 85 a) commonly marks the limit of each year's growth. These are the scars left by the fall of the scales of the bud ; and these, in the Horsechestnut, and in other trees with large scaly buds, may be traced back on the stem for a series of years, growing fainter with age, until they are at length obliterated by the action of the weather and the distention caused by the increase of the stem in diameter. The same is the case with the more conspicuous Leaf-scars, or marks on the bark left by the separation of the leaf-stalk, which are for a long time conspicuous on the shoots of the Horsechestnut (Fig. 85 b) , the Magnolia (Fig. 81) , and Hickory, Fig. 91. 83. Ramification. BRANCHES (14-16) are secondary stems developed from a primary one, or tertiary ones from these, and so on. Ultimate or small ramifications of latest order are some- times called BRANCHLETS. The terminal bud continues the stem or axis which bears it. Lateral buds give rise to branches.1 As the normal lateral buds are axillary (75), so are normal branches. The symmetry or arrangement of branches, being that of the buds from which they are developed, is fixed by and follows that of the leaves. When the leaves are alternate, the 1 Dichotomy or forking, the division of an apex into two, although of com- mon occurrence in the lower cryptoganious plants, occurs so rarely and exceptionally, if at all, in phaenogamous plants that it may here be left out of view. In phasnogamous plants only the ramification of axes should take the name of branches. That is, roots and stems branch ; and the term may without confusion be extended to hairs and all THICHOMES (383) when com- pound, but not to leaves and their modifications. FIG. 91. End of a Hickory branch (Carya alba), with a strong terminal and smaller axillary buds. 48 MORPHOLOGY OF STEMS. branches will be alternate; when the leaves are opposite, and the buds develop regularly, the branches will be opposite, &c. This holds in fact sufficiently to determine and exemplify the plan of ramification ; but, if entirely carried out, there would be as many branches as leaves. This could rarely if ever be, even in primary ramification. 84. Non-development of Bads. Some of the buds arc latent or merely potential, that is, do not make, their appearance : of those which do appear only a part actuall}1 grow into branches ; and of these some are apt to perish at an early stage. In our trees, most of the lateral buds generally remain dormant for the first season : they appear in the axils of the leaves early in summer, but do not grow into branches until the following spring ; and even then only a part of them grow. Sometimes the failure occurs without appreciable order ; but it often is nearly uniform in each species. Thus, when the leaves are opposite, there are usually three buds at the apex of a branch ; namely, the terminal, and one in the axil of each leaf ; but it seldom happens that all three develop at the same time. Sometimes the terminal bud continues the branch, the two lateral generally remaining latent, as in the Horsechestnut (Fig. 85) ; sometimes the terminal one fails, and the lateral ones grow, when the stem annually becomes two-forked, as in the Lilac, Fig. 86. The undeveloped buds do not necessarily perish, but are ready to be called into action in case the others are checked. When the stronger buds are destroyed, some that would else remain dormant develop in their stead, incited by the abundance of nourishment, which the for- mer would have monopolized. In this manner our trees are soon reclothed with verdure, after their tender foliage and branches have been killed by a late vernal frost, or consumed by insects. And buds which have remained latent for several years occasion- ally shoot forth into branches from the sides of old stems, especially in certain trees. 85. Most branches springing from old trunks, however, as in Willows and Poplars, especially when wounded or pollarded, originate from adventitious buds (77), which occur without order. So also when accessory buds (76) develop into branches, normal symmetry is more or less disturbed, as by contiguous shoots standing directly over each other in Tartarean Honey- suckle, or by a branch far out of the axil in Walnuts (Fig. 89) and Honey-Locust, Fig. 96. 86. Excurrent and Deliquescent Stems. Sometimes the primary axis is prolonged without interruption, even through the whole life of a tree (unless accidentally destroyed) , by the continued CHARACTERISTICS AND GROWTH. 49 evolution of a terminal bud, or by some upper strong bud which equally becomes a leader, — forming an undivided main trunk, from which lateral branches proceed ; as in most Fir-trees. Such a trunk is said to be excurrent. In other cases, the main stem is arrested, sooner or later, either by flowering, by the failure of the terminal shoot, or by the more vigorous develop- ment of some of the lateral buds ; and thus the trunk is dissolved into branches, or is deliquescent, as in the White Elm and most of our deciduous-leaved trees. The first naturally gives rise to coni- cal or spire-shaped trees ; the second, to rounded or spreading forms. As stems extend upward and evolve new branches, those near the base, being overshadowed, are apt to perish, and thus the trunk becomes naked below. This strikingly occurs in the excurrent trunks of Firs and Pines, grown in forest, which seem to have been branchless to a great height. But the knots in the centre of the wood are the bases of branches, which have long since perished, and have been covered with a great number of annual layers of wood, forming the clear stuff of the trunk. 87. Definite and Indefinite Annual Growth of Branches. In many of our trees and shrubs, especially those with scaly buds, the whole year's growth (except on certain vigorous shoots) is either already' laid down rudimentally in the bud, or else is early formed, and the development is completed long before the end of summer ; when the shoot is crowned with a vigorous terminal bud, as in the Horsechestnut (Fig. 85) and Magnolia (Fig. 81), or with the uppermost axillary buds, as in the Lilac (Fig. 86) "'-'•< '^ and Elm. Such definite shoots do not die down at all the follow- ing winter, but grow on directly, the next spring, from these terminal or upper buds, which are generally more vigorous than those lower down. In other cases, on the contrary, the branches grow onward indefinitely, until arrested by the cold of autumn : the buds at or near their summit are consequently 3*oung and miniatured, or at least the lower and older axillary buds are more vigorous, and alone develop into branches the next spring ; the later-formed upper portion most commonly perishing from, the apex downward for a certain length in the winter. The Rose and Raspberry, and among trees the Sumac and Honey- Locust, are good illustrations of this sort; and so are most perennial herbs, their stems d}-ing down to or beneath the sur- face of the ground, where the persistent base is charged with * vigorous buds, well protected by the ground, for the next year's vegetation. 88. Many of the details and applications of ramification, of most importance in morpholog}- and descriptive botany, relate 50 MORPHOLOGY OF STEMS. to anthotaxy or inflorescence (Chap. V.), which has its own terminology. But some of its terms may be conveniently- employed in the description of ramification unconnected with flowering. § 2. FORMS OF STEMS AND BRANCHES. 89. On the size and duration of the stem the oldest and most obvious division of plants is founded, namely, into Herbs, Shrubs, and Trees. 90. Herbs are plants in which the stem does not become woody and persistent, but dies annually or after flowering, down to the ground at least. The difference between annual, biennial, and perennial herbs has already been pointed out in the chapter on the root (50—57), and the gradations between them indicated. Herbs pass into shrubs and shrubs into trees through ever}' gra- dation. The following definitions are therefore only general : — 91. Undershrubs, or Suffruticose plants, are wood}' plants of humble stature, their stems rising little above the surface. If less decidedly woody, they are termed Suffrutescent. 92. Shrubs are woody plants, with stems branched from or near the ground, and less than five times the height of a man. A shrub which approaches a tree in size, or imitates it in aspect, is said to be Arborescent. 93. Trees are wood}' plants with single trunks, which attain at least four or five times the human stature. Yet the name of tree is not to be denied to a woody plant having a single and stout trunk of less altitude ; and those which grow in a bushy manner, sending up a cluster of stems from the ground to the height of thirty feet or more, may still be called shrubs. 94. The erect position, elevation above the soil, and self-sup- port, are normal conditions of the stem, but are far from universal. And certain kinds of stem or branches are sufficiently peculiar to have received substantive names : other equally peculiar forms have no special names. There are, moreover, certain organs (such as spines and tendrils) which are commonly homologous (12) with stems, but not always. Two kinds of erect stems have special names in descriptive botany. 95. Culm is a name applied to the peculiar closed-jointed stem of Grasses and Sedges, whether herbaceous, as in most Grasses, or woody or arborescent, as in the Bamboo. 96. Caudex is the name technically applied to the trunk of Palms (Fig. 126), Tree-Ferns, and the like, consisting of a commonly simple column, the surface beset with scales, — the SPECIAL FORMS. 51 bases of former leaf-stalks, — or marked by scars, left by their fall. This name was used by botanists anterior to Linnaeus for an}- tree-trunk, but is now used for the peculiar stems above- mentioned ; also for the persistent base of a stem, otherwise annual, which throws up fresh herbaceous stems or stalks from year to }-ear. Such short and enduring stems, being usually near the ground or under it, were commonly mistaken for roots. The old English name of Stock is sometimes used in botanical description for all short and enduring, stems of this sort, whether rising somewhat above or concealed beneath the surface of the soil. 97. A Scape is a stem or branch which rises from beneath or near the surface of the ground and bears flowers, but no proper foliage. It therefore belongs to inflorescence. (265.) Scapes usually spring from some one of the subterranean forms of stern. 98. Of stems which do not stand upright in the air there are various modifications and gradations. 99. Scandent or Climbing Stems are those which rise by attaching themselves to some extraneous support. This is effected in various ways ; in some by the action of the stem itself, in others by that of organs which it bears.1 100. Voluble or Twining Steins, or Twiners, are those which ascend by coiling round a support, which must accordingly be comparatively slender, or at least not too large. Some ascend by coiling "with the sun" (that is, from right to left of the observer viewing the coil from the outside 2), as the Hop ; more, 1 See Darwin, The Movements and Habits of Climbing Plants, London and New York, 1875. Also the earlier paper on the subject in Journal of the Linnean Society, ix. 1865. Note that in North America climbing plants in general are in popular language called Vines (e. g. Hop- Vine, Grape- Vine, Squash- Vine, &c.), a name which properly belongs to Vitis only. 2 Dextrorse and Sinistrorse, i. e. to the right or to the left, are almost indis- pensable terms, but there is an ambiguity and discrepancy in their use. Darwin (in Climbing Plants, above referred to) seeks to avoid this by usually employing the terms " with the sun," and " against the sun," phrases which would be unmanageable in terminology. The writer (in Amer. Jour. Sci. ser. 3, xiii. 391) suggested Entropic for the former, Antitropic for the latter, to be used in case it is preferred to evade rather than to encounter the ambiguity. Probably the terms dextrorse and sinistrorse, or right and left, will continue in use, as most natural and convenient. Now, in the first place, it should be understood that a plant, or at least a plant's axis, having no front and back, can have no right and left of its own. These relations of direction must refer to the right and left of an observer. All depends, accordingly, upon the position which the viewing observer is supposed to occupy when he predicates the direction of the turns of a helix or of the over- lapping of the parts of a bud. Linnaeus supposed the observer to view the 52 MORPHOLOGY OF STEMS. by coiling in the opposite direction, as the Bean (Phaseolus), the woody Aristolochia Sipho, the Morning Glory (Fig. 91°) and other Convolvu- laceae. The Dod- der, a leafless par- asitic plant of the latter family, not only gains support by coiling on the stems of othe.r plants, but by attachment, through the development of sucker-like discs, along the whole contiguous surface. (Fig. 77.) The various actions through which plants climb, and the attendant phenomena, are physiological, and will be treated in the second part of this Text Book. The most complete and satisfac- tory discussion of the subject, of a readable sort, is that of Darwin's volume, referred to in a preceding note. 101. Leaf-Climbers are those in which support is gained by the action, not of the stem itself, but of the leaves it bears ; in most by the coiling or clasping of petioles, as in Clematis, Maurandia, Tropaeolum, and Solanum jasminoides (Fig. 235) ; in some by the incurvation of leaf-blades or portions of them, as in Adlu- mia ; or by an extension of the midrib into a hook or short ten- dril, as in Gloriosa ; or by the transformation of some of the blades of a compound leaf into hooks or tendrils, as in Cobtea and the Pea. 102. Tendril-Climbers (Fig. 92-95) are those in which the prehension is by a tendril, a slender filiform body, either simple or branched, specialty adapted to the purpose, and capable of coiling, either to secure a hold, or to draw the stem up to the coil or circle from the inside ; Mohl, Palm, Braun, and the DeCandolles adopt this, and the latter insist on it. Such authority should be decisive, if common usage and popular sense went along with it. But some of the botanists following Linnaeus adopted the reverse view ; and to the present writer, as to Bentham and Hooker, Darwin, Eichlcr, and in part G. Henslow, it was so natural to view the coil from the outside that we without concert adopted this position and mode of expression. A right-hand coil, or one turning to the right, with us, is one the turns of which pass from the left to right of a bystander who confronts the coil. It is in this sense that a com- mon screw is called a right-handed screw, and that the right bank of a river is that to the right of the person who follows the course of the stream. So natural is this, that even on a map or plate, which has face and back, and therefore a right and left of its own, the figures occupying its right or left portions are understood to be those which are toward the right or the left hand of the observer who stands before it. FIG. 91°. Dextrorsely twining stem of Morning Glory, Ipomoea purpurea SPECIAL FOBMS. 53 support. In certain tendrils the attachment to the support is by a sucker-like disc at the apex, as in the Virginia Creeper or Ainpelopsis, Fig. 94. 103. Root-Climbers are those in which the stems produce aerial rootlets (59°), which fix themselves to a supporting surface along which the stem creeps or ascends. In this way Trumpet Creeper (Tecoma radicans), Ivy, and Poison Ivy (Rhus Toxi- codendron) climb extensively. 104. Stems or branches which neither climb nor stand upright may have their direction or habit of growth expressed by certain adjective terms ; such as Ascending or Assurgent, when the}' rise obliquely upward ; Reclining, when from an ascending or erect base the upper part recurves and trails ; Decumbent, when trailing along the ground, but with apex assurgent ; Procumbent or Prostrate, when lying at length upon the ground ; Repent or Creeping^ when growing prostrate on the ground and rooting as they grow. Also applied to similar stems grow- ing under, as well as upon the surface of the soil, as in Couch- Grass and Mint, Fig. 99. 105. A Sucker (Surculus) is an ascending stem rising from a subterranean creeping base. The Rose and Raspberry multiply freely by suckers. Such plants are easiest to propagate " by division." 106. A Stolon is a prostrate or reclined branch which strikes root at the tip, and then develops an ascending growth, which becomes an independent plant. 107. An Offset is a short stolon or a short sucker. Houseleek (Fig. 91*) offers a familiar example. By offsets, some herbs, otherwise annuals, are continued from }-ear to year in a vegetative progeny (Lobelia cardinalis, &c.), and peren-| nials may thus establish colo- nies around a parent individual. 108. A Runner (Flagelluni) is a filiform or very slender stolon, naked and tendril-like except at tip, where it roots, develops a bud, and so a new plant. The Strawberry furnishes the most familiar example. FIG. 91&. Houseleek (Sempervivnm tectormn) with offsets. 54 MORPHOLOGY OF STEMS. 109. The two following are organs which may be of axial nature, or may not. This may ordinarily be determined by posi- tion. Any direct continuation of stem or branch must be of axial nature, that is, of the nature of stem ; and the same is true of whatever primarily develops in the axil of a leaf. Conversely, whatever subtends a lateral axis or branch ma}- be taken for a leaf or foliar production, being in the place of such. 110. A Tendril, a thread-shaped and leafless body, capable of coiling spirally, and used for climbing (102) , is homologous with stem in Grape- vines (Fig. 92) ; for the uppermost tendril is seen to be a direct continu- ation of the stem. The small bud which appears in the axil of the uppermost leaf will in its growth produce another internode and leaf, or some species more than one, but will terminate in a similar tendril: the present terminal tendril will have then become lateral and opposite the leaf, like the three in the lower part of FIG. 92. End of a shoot of the Grape-vine, with young tendrils: a sympodial stem. (See note.) FIG. 93. A portion of a stem of Ampelopsis quinquefolia, or Virginia Creeper, with a leaf and a tendril. FIG. 94. Ends of the latter, enlarged, showing the expanded tips or discs by which they cling. SPECIAL FOUMS. 55 the figure.1 The tendrils of Virginia Creeper (Fig. 93) are of the same nature and position. But, instead of laying hold \)y a coiling of the tip, when it has reached any solid surface, such as a wall or tree-trunk, the tip expands into an adhesive disc, which forms a secure attachment. (Fig. 94.) In a related plant, Vitis (Cissus) tricuspidata of Japan, these disks terminate the branches of very short tendrils : consequently the shoots as they grow are at once applied closely and secured firmly to the surface of the sup- port, — an admirable adaptation for climbing walls and trunks. 111. The simple tendril of a Passion- flower, being in the axil of a leaf (that is in the position of a branch) , is also of axial nature : it is a leafless and simple branch, composed of one long and slen- der internode, devoted to the purpose of climbing. Fig. 95 shows in all stages the admirably active tendrils of Passiflora sicyoides. This is a Mexican species, remarkable for the rapidity and freedom with which the tendrils move. The lowest tendril in the figure is attached and coiled : the next is free and coiled in one helix : the third is outstretched and seeking a support. For tendrils which are not homologous with stems, see Sect. IV. 228. 112. A Spine or Thorn (Fig. 96, 97) is usually a branch or the termination of a stem or branch, indurated, leafless, and attenuated to a point. The nature of spines is manifest in the Hawthorn (Fig. 97), not only by their position in the axil of a leaf, but often by producing imperfect leaves and buds. And in the Sloe, Pear, &c., many of the stinted branches become spinose or spntescent at the apex, tapering off gradually into a rigid and leafless point, thus exhibiting ever}' gradation between a spine and an ordinary branch. These spinose branches are less . 1 This forms what is called a Sympodinm or Sijmpodial stem, which is mor- phologically made up of a series of superposed brandies. (Sec Chapter V. __ 281,282.) In contradistinction, a stem formed hy the continued development /"* /S~ of a terminal bud is Monopodial or a Monopodium. Fig. 95 is an example. FIG. 95. Leafy slioot of Passiflora sicyoides, of Mexico, with fixed and coiled, free uiiil full grown, and forming tendrils. 56 MORPHOLOGY OF STEMS. liable to appear on the cultivated tree, when duty cared for, such branches being thrown mostly into more vigorous growth. In the Hawthorn, the spines spring from the main axillary bud, while accessor buds (76), one on each side, ap- pear, and grow the next sea- son into ordinary branches. In the Honey-Locust, it is the uppermost of several ac- cessor}- buds, placed far above the axil, that develops into the thorn (Fig. 96) . Here the spine itself usually branches, and sometimes becomes ex- tremely compound. 113. For spines which are homologous with leaves, or parts of a leaf, see Sect. IV. 227a. Prickles, such as those of Brambles and Roses, are superficial outgrowths from the bark, of a different nature (383) , and of small morpho- logical signification. 114. Subterranean Stems are hardly less diverse than the aerial. They are classed as RHIZOMES, TUBERS, CORMS, and BULBS, the forms passing one into another b}- gradations. 115. Rhizoma (Rhizome, or in English ROOTSTOCK) is a gen- eral name for any horizonal or oblique perennial stem, which lies on the ground or is buried beneath its surface. It sends off roots of a fibrous or slender sort wherever it rests on or is cov- ered bj* the soil, and usualty produces from its apex some kind of aerial stem, either leafy or as a flower-stalk (scape, 97), which rises into the air and light. Before morphology was understood, rootstocks were called creeping roots, scaly roots, &c. Some are slender, such as those of Mints (Fig. 99), of most Sedges (Fig. 98), and of Couch-Grass. Their cauline nature is evident from their structure and appearance ; their nodes and internodes are well marked, the former bearing leaves reduced to FIG. 96. Brandling thorn of the Honey-Locust fGleditschia). an indurated branch developed from an accessory bud produced above the axil. a. Three buds tinder the base of the leaf-stalk, brought to view in a section of the stem and leaf-stalk below. FIG. 97. Thorn of the Cockspur Thorn, developed from the central of three axillary buds; one of the lateral buds is seen at its base. SPECIAL FORMS. 57 scales , and the advancing apex rises at length into an ordinarj' stem, while the opposite and older end gradually dies away. A bud forms in the axil of each scale-like leaf, or » in some of them ; roots proceed from the nodes in preference ; the destruction of the ascending stem only brings these buds into activitj- ; and the cutting or tearing of the rootstock into pieces by the hoe or plough merely hastens the establishment of as many new plants, each with roots, bud, and a small store of nourishment ready provided. It is this which makes Couch- Grass or Quick-Grass (Triticum repens) very troublesome to the agriculturist ; and the Nut-Grass (Cj-perus rotundus, var. Hydra) of the Southern Atlantic States is even more so, portions of its rootstock being tuberiferous, i. e. en- larged into a tuber which contains a supply of concentrated nourishment to feed the growth. 116. Thickened rootstocks are common; nourishing matter, elaborated in the leaves above, being accumulated in them, just as it is in thickened roots, and for the same pur- 990 pose. (53-55.) Such are the so-called roots of Sweet-Flag, of ^ Ginger, of Iris or Flower-de-Luce (Fig. 216), of Bloodroot, of ' Solomon's Seal (Fig. 100), &c. These grow after the manner of ordinary stems, advancing from year to year by the annual development of a bud at the apex, and emitting roots from the under side or the whole surface. Thus established, the older . FIG. 08. Slender rhizoma of Carex arenaria, of Europe, which binds shifting sands of the sea-shore. FIG. 99. Rootstocks, or creeping subterranean branches, of the Peppermint. FIG. 99". A piece of the rootstock of the Peppermint, enlarged, with its node or joint, and two axillary buds ready to grow. 58 MORPHOLOGY OF STEMS. portions die and decay as corresponding additions are made to the opposite growing extremity. Each year's growth is often marked conspicuously, sometimes by a strong contraction where the interruption took place, as in certain species of Iris (Fig. 216) ; or by the circular im- pressed scar (likened to the impression of a seal) in Solomon's Seal ; this being the place where the annual aerial stem, bearing the vegetation, separated in autumn from the perennial rhi- zoma. The numerous slender lines encircling the rootstock are the scars left after the decay of the scale-like leaves or bud-scales, such as are seen at the young and growing end of the rootstock. The rootstock of Diphylleia, of the Alleghanj^ Mountains (Fig. 101), is similar; but the yearly growths are so exceedingly short that they become vertical, the bud of each j-ear is close to the stalk of the year preceding, and the scars marking previous growths are in contact.1 Trillium makes a short and mostly vertical rootstock, which, when it remains simple and dies away promptly below (as in Fig. 102), comes nearly within 102 V>"! the definition of a corm. But in several 1 The rootstock in Polygonatum and Diphylleia is a sympodium (110, note), the terminal bud developing yearly the growth above ground and perishing FIG. 100. Rootstock of Polygonatum or Solomon's Seal, with the terminal bud, the base of the stalk of the season, and three scars from which the latter has separated in as many former years. FIG. 101. Rhizoma of Diphylleia cymosa, showing six years' growth, and a bud for the seventh : a, the bud : ft, base of the stalk of the current year : c, scar left by the decay of the annual stalk of the year before; and beyond are the scars of previous years. FIG. 102. Shoot and young rootstock of Trillium erectum, with only terminal bud. SPECIAL FORMS. 59 species, and in older individuals, it is longer, often oblique, and branching, and bears the scars from which the annual aerial growths have separated.1 Nymphaea odorata, the sweet-scented white Water Lily, grows by very long, stout, and simple rootstocks. In N. tuberosa the sides of the rootstock produce short lateral branches or tubers. 117. A Tuber may be morphologically char- acterized as a short thickened rhizoma on a slender base, or a rootstock some portion of which — mostly a terminal portion and involv- ing several nodes — is thickened by the depo- sition of nourishing matter. A potato and a Jerusalem artichoke are typical examples (Fig. 104-107) ; and the difference between these subterranean branches and the roots which they may bear is very obvious. Their eyes are axillary buds ; the leaves which subtend them are plainly dis- cernible, in the form of short and closely appressed scales. In the attempt, occasionally seen, to form axillan- tubers above- ground by the Potato-plant, the leafy nature of the scales is evidenced. (Fig. 105.) 63- heaping the soil around the stems, the- number of tuber- ( iferous branches may be in- creased. The number of nodes and internodes involved in a tuber may be many or few. There is one instance of what may in autumn, to be renewed by an axillary bud, which makes its subterranean growth and the rudiments of the aerial in early summer. 1 This rhizoma is a monopodium, being continued year by year by the terminal bud, and the aerial stem or stems sent up in spring, bearing the whorl of leaves and blossom, are axillary branches. FIG. 103. An older and longer one of the same species, showing branches, scars left by former leaf- and flower-bearing stems : also at tip (stripped of the covering scales), the bases of two such stems of the season, and the terminal bud between them, for the con- tinuation of the growth of the rootstock, &c., the next season. FIG. 104. Base of stem of Helianthus tuberosus, or Jerusalem Artichoke, developed from a tuber, and producing a second generation of tubers. FIG. 105. Monstrosity of a Potato-plant, with an axillary bud developing into some- thing between a bulblet and a tuber ; the scales represented by obvious leaves. (From the Gardeners' Chronicle.) 60 MORPHOLOGY OF STEMS. be called a Monomerous tuber, namely in Nelumbium luteum (Fig. 108"), where it consists of a single thickened internode of an aquatic runner, which is accordingly quite destitute of scales or buds. The growth proceeding from this simple tuber is necessarily from a bud of the node at its apex, whence also a cluster of roots is produced. Of a somewhat similar nature are the con- catenate tubers of Apios tuberosa (several of which io8° are strung as it were upon a long filiform axis), the tubers not unfrequently being mo- nomerous, although the larger ones are not so. 117". Tubercles, as they may be termed, are of a mixed or ambiguous character between tubers and tuberous roots. A good example of the latter is afforded by Dahlia-roots. (Fig. 68.) They yield their nourishing substance to growing buds on the stem above, but do not themselves normally produce even FIG. 106. Forming potatoes in various stages. 107. One of the younger ones en- larged, log. Section of a small portion passing through an eye, or bud, more enlarged. FIG. 108a. A monomerous (i. e. one-membered) tuber of Nelumbium luteum, formed of a single internode. SPECIAL FORMS. 61 adventitious buds. Sweet potatoes (55), although equally roots, do produce adventitious buds, especially from near the upper end. The somewhat similar tubercles or tumefied roots of certain Orchises and other plants of the same tribe,1 definite in number and shape, and sometimes imitating a corm, are charged with a bud at the upper end, near their origin. Ap- parently, the origin is a bud from the base of the parent stem, which bud directly forms a tumefied short root from its very base.2 118. A Corm (Cormus) is to be compared on the one hand with a short rootstock or tuber, on the other with a bulb. It is a subterranean fleshy stem, of rounded or depressed figure and solid texture. Some of its buds grow into new corms, and these, upon the death of or separation from the parent, become new individuals : some develop above ground the vegetation and the blossoms of the season. A good type of corm is that of Cyclamen (Fig. 109), in which the very base of the seedling stem grows fleshy, and widens from year to year, but hardly at all lengthens, and so becomes far broader than high, or de- pressed. As the main bulk belongs to the first internode, or caulicle, the buds from which the yearly growths of leaves and flower-stalks spring are at the centre of the summit or upper surface, the roots from the lower, and the sides seldom pro- duce any buds. The corm of Indian Turnip (Arissema triphyllum, Fig. 110) is somewhat similar, but it sends up a single stout stem, and the roots sprir from around the base of this. These are completely naked corms. 119. But in Crocus (Fig. Ill, 112), Colchicum (Fig. 117). 7 Gladiolus, and the like, the sheathing bases of one or t\vo leaves enclose the corm with a membranous-scaly coat, giving it exactly the appearance externally of a coated bulb. Such have been not inappropriately named solid bulbs. In common parlance, they will doubtless continue to be called bulbs, and even in popular * 1 Not, however, such as those of Aplectrum, Tipularia, etc., which are genuine corms or tubers. - Irmisch, Beitr. Biol. & Morphol. Orchid. 1853, fide Duchartre, Elem Bot. 278. FIG. 109. Depressed corm of Cyclamen. FIG. 110. Corm of Indian Turnip, Arisaema triphyllum. 62 MORPHOLOGY OF STEMS. botanical descriptions. In fact, while they differ from naked corms in having some investment, they differ from true bulbs only in the greater size of the solid axis and the fewness of the investing scales ; the stem or solid body making the greater part of the corm, but a very small part of a proper bulb. There are, more- over, all gradations be- tween the two. 120. A Bulb, as compared with a corm, may be said to be an cxceedingl}' abbreviated stem, reduced to a flat plate, from the lower face of which roots are produced, from the upper face, leaves in the form of scales ; these scales being either reduced and thickened leaves or the thickened bases of ordinary leaves. Compared with buds (73), it is a very fleshy bud, usually large and subterranean, the axis of which never elongates. It is a provision for future growth, the stored nourishment of which Is deposited in the leaves, or the homologues of leaves, instead of in the stem. FIG. 111. Corm of Crocus, the few thin enveloping scales removed, showing their scars, which mark the nodes, the shrivelled vestige of the last year's corm at the base, and buds developing into new ones ou various parts of its surface. 112. Vertical section of a similar corm, with a terminal and one lateral bud. FIG. 113. Section of a tunicated bulb of the Onion. FIG. 114. Vertical section of the bulb of the Tulip, showing its stem or terminal bud (c) and two axillary buds (6, b). FIG. 1 15. Bulb of a Garlic, with a crop of young bulbs. FIG. 116. Vertical section of the corm of a Crocus: a. new buds. FIG. 117. Vertical section of the corm of Colchicum (6), with the withered corm of the preceding (a), and the forming one (c) for the ensuing year. SPECIAL FORMS. 63 121. A Tunicated or Coated Bulb (Fig. 113-115) is one in which the scales are broad and completely enwrapping, forming concentric coatings. These are tkickish when fresh, but thin when exhausted and dry, as in the Onion, Garlic, and Tulip. 122. A Scaly Bulb has the bulb-scales comparatively nar- row, thick, and small, imbricated, but not severally enwrapping each other. That of the Lily is the most familiar and char-# acteristic example.- (Fig. 118, 119.) "8 119 123. Bulblets are small aerial bulbs, or buds with flesh)- scales, which arise in the axils of the leaves of several plants, such as the common Lilium bulbiferum and L. tigrinum, the Tiger Lilies of the gardens (Fig. 120). Here they ap- pear during the summer as axillary buds: they are at length detached, and falling to the ground strike root, and grow as independent plants. In the common Onion, and in many other species of Allium, similar bulblets take the place of flower-buds in the umbel. Bulblets plainly show the identity of bulbs with buds. 124. All these extraordinary, no less than the ordinaiy, forms of the stem, grow and branch, or multiply, by the development of terminal and axillary buds. This is perfectly evident in the rhizoma and tuber, and is equally the case in the corm and bulb. The stem of the bulb is usually reduced to a mere plate (Fig. 114 a), which produces roots from its lower surface, and leaves or scales from the upper. Besides the terminal bud (c), which usually forms the flower-stem, lateral buds (5, b) are produced in the axils of the leaves or scales. One or more of these ma}- develop as flowering stems the next season, and thus the same bulb survive and blossom from year to year ; or these axillary buds may themselves become bulbs, feeding on the parent bulb, which in this way is often consumed by its own offspring, as in FIG. 118. Scaly bulb of Canada Lily, Lilium Canaclense, after flowering. 119. Ver- tical section of same, showing two new young bulbs within. FIG. 120 Bulblcts In the axil of the cauline leaves of Tiger Lilv. 64 MORPHOLOGY OF STEMS. the Garlic (Fig. 115) ; or, finally separating from the living parent, just as the bulblets of the Tiger Lily fall from the stem, they may form so many independent individuals. So the corm of the Crocus (Fig. Ill, 112) produces one or more new ones, which feed upon and exhaust it, and take its place ; and the next season the shrivelled remains of the old corm may be found underneath the new. The corm of Colchicum (Fig. 117) pro- duces a new bud on one side at the base, and is consumed by it in the course of the season ; the new one, after flowering by its terminal bud, is in turn consumed by its own offspring ; and so on. The figure represents at one view, a, the dead and shrivelled corm of the year preceding ; 6, that of the present season (in a vertical section) ; and, c, the nascent bud for the growth of the ensuing year. 125. Condensed Stems, homologous with corms, tubers, &c., and similar in mode of growth, but above ground, and multiply- ing in the same ways, are not uncommon. The Cactus famiby is mainly composed of such forms, of flat- or round-jointed Prickly Pears (Opuntia) , fluted or angled columns (Cereus) , and glob- ular Melon-Cactus, Mamillaria, and Echinocactus. The latter types, which completely imitate corms, are the most consolidated forms of vegetation. While ordinary plants are constructed on the plan of great expansion of surface, these present the least possible amount of surface in proportion to their bulk, their permanent spherical figure being that which exposes the smallest portion of their substance to the air. Such plants are evidently adapted to very dry regions ; and in such only are they naturally found. Similarly, bulbous and corm-bearing plants, and the like, are a form of vegetation which in the growing season may in the foliage expand a large surface to the air and light, while during the period of rest the living vegetable is reduced to a globular or other form of the least surface ; and this is protected by its outer coats of dead and dry scales, as well as by its subter- ranean situation ; — thus exhibiting another and very similar adaptation to a season of drought. And such plants mainly belong to countries (such as Southern Africa, and the interior of Oregon and California) which have a long hot season, during which little or no rain falls, when, their stalks and foliage above and their roots beneath being early cut off by drought, the plants rest securely in the corm-like forms to which they are reduced, and retain their moisture with great tenacity until the rain}' season returns. Then they shoot forth leaves and flowers with wonderful rapidity, and what was perhaps a desert of arid sand becomes green with foliage and gay with blossoms, almost in a day. SPECIAL FORMS. 65 126. Stems serving the purpose of foliage, Phyllocladia. Most of these condensed and permanent stems are illustrations of this, their green rind doing duty for leaves, which are either absent, or transient, or reduced to spines or other organs not effective as foliage. In the flat and broad-jointed species of Opuutia, and still more in Phyllocactus and Epiphyllum, the forms assumed give a considerable surface of green rind, which well answers the purpose of leaves. Flattened stems or branches of the same sort and economy not seldom occur in other than fleshy or succulent plants (such as the Cactuses) ; some- times accompanied by a certain number of real foliage-leaves, but these more or less transient, as in Bossiaea and Carmichaelia among Legu- minous shrubs, and MuhlenbecMa platyclada, now in common cultivation (Fig. 121) ; sometimes with all the leaves reduced to small and function- less scales, as in the Xylophylla (t. e. wooden- leaved) section of Phyllanthus, and in Phyllo- cladus (New Zealand and Tasmanian trees of the Yew family) . In all these, the cauline nature is manifest by the continuous or proliferous growth, by the marked nodes and internodes, and often by the bearing of flowers. 127. Cladophylla (literally, branch-leaves) are more ambigu- ous in character. The most familiar examples are found in the peculiar foliage of Ruscus, Myrsiphyllum, Asparagus, and in some other genera of the same family. In these the primary or proper leaves of the shoots are little scales, one to each node, and quite functionless. From the axil of each is immediately pro- duced a body answering in all respects to the blade of a leaf, both in appearance and in office. They also accord with leaves in being expanded horizontally, although they take a twist which brings them more or less into a vertical position, in the manner of phyllodia (that is, of leaf-stalks assuming the form and office of leaf-blades, 217); wherefore they may be regarded as the first and only leaf of an axillary branch with the internode under the leaf wholly undeveloped and no further growth ever taking place. But, on the other hand, their anatomical structure is said to be that of stems rather than of leaves. Moreover, FIG. 121. Foliiform branch of Muhlenbeckia platyclada, growing from the apex, bearing a small and transient leaf at some nodes, also a flower or two. 66 MORPHOLOGY OP STEMS. the cladophyll of Ruscus (called Butcher's Broom in England, Fig. 123) not only becomes firm, hard, and spiny-tipped, but it exhibits the character of a branch by bearing flowers on the middle of one face, in the axil of a little bract. Under this view such a dsdophyU would seem to be a- flattened branch of two in ternodes, or else of one internode with a flower-stalk adnate to it. In Myrsiphyllum (a South Afri- can climber, commonly cultivated under the erroneous name of Smilax, Fig. 122), the cladophyll is wholly leaf-like in appearance as well as in function, and it never bears either scale-leaf or blossom ; but the flowers are on slander stalks from buds out of the same axil. (See Dickson in Trans. Bot. Soc. Eclinb. xvi., and Van Tieghem, Bull. Bot. Soc. France, xxxi., for a discussion of the nature of cladophylla.) 128. To all such leaves or imitations of leaves, Bischoff has given the name PIIYLLOCLADIA, sing. PHYLLOCLADIUM. To those definitely restricted to one internode, and which so closely counterfeit leaves, Kunth gave the name of CLADOUIA, sing. CLADODIUM. The best common name for all productions which imitate leaves would have been that of phyllodiuin (meaning simply a leaf-like body) ; but that term was first applied and is restricted to the case of a petiole imitating the J>lade of a leaf. The name Phyllocladium (meaning a leaf-like branch) may properly be retained for the whole scries of leaf-like bodies here described. But for those of the preceding paragraph, which are so peculiarly leaf-like, Kunth's name of Cladodium (i. e. a branch-like body) is false in meaning, and may be replaced by that of CLADOPHVLLOI (i. e. leaf- branch), or in shorter English CLADOPHYLL. 129. Frondose Steins. Finally, in some few phaenogamous plants, the whole vegetation is re- duced to a simple leaf-like expansion, as in Duckweed (Lemna) , FIG. 122. Myrsiphyllum, with cladophylls serving for foliage; the true leaves con- sisting of minute and very inconspicuous scales subtending the former. FIG. 123. A single cladophyll of RUKCUR aculeatus in the axil of a scale-leaf, bearing another scale-leaf on the middle of its face, and flowers in the axil of this. FIG. 124. Letnna minor, a common Duckweed, whole plant in flower, magnified. INTERNAL STRUCTURE. 67 Fig. 124. Here is no differentiation whatever into stem and foliage ; but the expanded floating body which serves for both must be counted as stem developed horizontally into a flat plate, for it produces a root from the under surface and a flower from the edge. This simplification is common in some orders of Cryptogamous plants ; and such a body, which answers both for stem and foliage, is termed a FROND, from the Latin frons, which means either leaf or leafy bough. In some species of Lemna the frond is thickened or plano-convex : in Wolffla, the simplest and smallest of phaenogamous plants, it is a globular green mass, seldom much larger than the head of a pin, wholly destitute of root, propagated by proliferous budding from one side, and from within the top producing a flower or pair of flowers. § 3. INTERNAL STRUCTURE. 130. The investigation of the intimate structure of the stem, as of the other organs, belongs to vegetable anatomy or histology (treated in Part II.) ; but the general outlines of structure, so far as is requisite to the explanation of what is visible to the naked eye, should be here explained. 131. The stems of phaenogamous plants anatomically consist of two general elements, the cellular and the woody ; the former exemplified in the commoner stems by the pith and outer bark, the latter by the wood. Both are equally composed of cells, or origi- nate as such ; but those which form the woody system of the stem mainly undergo, at a very early period, transformation into tubes, some of which are of such small calibre that their common name of fibres is not inappropriate ; others, of larger size or ampler calibre, take the name of ducts or vessels. The latter are almost FIG. 124°. A magnified slice of a portion of the flower-stalk of Kichardia ^Ethiopica (the so-called Calla Lily), transverse witli some longitudinal view : mainly parenchyma, the cells built up so as to leave comparatively large vacancies (intercellular spaces or air-passages); near the centre a cross-section of a nbro-vaseular bundle, and next the margin or rind some finer ones. 68 MOKPHOLOGY OF STEMS. always associated with the wood-cells, so that they are in a general way taken together as constituting the wood, or woody tissue, and as forming what is more definitel}* termed fibro- vascular tissue or, when distinguishable into threads, fibro- vascular bundles. These run lengthwise through the stem, sometimes as such separate threads, sometimes confluent into a compact structure. The softer or at least the non-fibrous portions, formed of comparatively a b d e f g A short and commonly thin-walled cells, form cellular tissue. Its ordinary form (of roundish, cubical, or polyhedral and thin- walled cells) is called parenchyma. This abounds in herbaceous stems or herbaceous parts : in trees and shrubs, woody tissue largely pre- vails ; in most herbs, it forms a notable portion ; in some (especially FIG. 126. Fibro-vascular elements, a. Bast-cells (long wood-cells) of fibrous bark of Linden or Bass-wood, b. Some wood-cells and (below) a duct, and c. a detached wood-cell of the wood of same tree, equally magnified with a. <1. A detached wood- cell from a shaving of White Pine, showing the peculiar disk-like markings, e. Portion of same shaving. /. Portion of a dotted duct from the Vine, evidently made up of a series of short cells, g. Part of a smaller dotted duct, showing no appearance of such composition, h, i. Spiral ducts or vessels, of the ordinary kind. j. Spiral duct of Banana, k. Duct from Celery, the thread within spiral or annular below, reticulated above, and higher passing into the state of dotted duct. /. Duct from Impatiens, with the open spiral passing into rings at the middle. All magnified somewhat equally. INTERNAL STRUCTURE. 69 in certain aquatic herbs), it is reduced to a few threads or vessels, generally delicate, and sometimes obscure. The ac- companying anatomical illustrations (Fig. 124", 125, with their "P 6 explanations) will give a general idea of the nature of the ana- tomical elements of the stem. 132. In the forming state, the whole stem is parenchyma ; but an early differentiation takes place, converting certain portions into wood}- or fibro- vascular tissue. This is arranged in two wa}*s, giving rise to two kinds of stem in phsenogamous plants, which have been termed the Endogenous and the Exogenous,1 meaning inside and outside growers. 133. The two plans of stem are usually manifested in external conformation as well as internal structure, and are correlated with important differences in embryo, foliage, and flower.2 Palms, Lilies, Rushes, and Grasses are examples of the endogenous class ; the ordinary trees and shrubs, especially those of cool climates, and a large part of the herbs, are of the exogenous class. In an exogenous stem, the wood occupies annual concentric layers, one of each }-ear's growth ; the centre is occupied by a pith, composed of parenchyma only, the circumference by a separable bark ; so that a cross-section presents a series of rings or circles of wood, or in the first year one ring, surrounding the pith and surrounded by the bark. An endogenous stem has the wood in distinct threads or fibro- vascular bundles, travers- ing the cellular S3*stem or parenchyma with little or no obvious order, and presenting on the cross-section the divided ends of these bundles in the form of dots ; these usually (but not always) diffused over 1 Terms introduced by DeCandolle, following the ideas of Desfontaines, and which have played an important part in structural and systematic botany ever since DeCandolle adopted these names as those of the two primary divisions of phaenogamous plants, Exogence and Endorjence. But it has long been seen that the name of the second kind is not appropriate ; and the older and better (though longer) names of Jussieu, Monocotyledones and Dicotyledones, are reverted to. Yet the Candollean names are still much employed, with due explanation, to designate the two kinds of structure of the stem. 2 Yet with some more or less valid exceptions, as when the annual stem of Podophyllum and the rhizoma of Nymphaea, among dicotyledonous plants, imitate the endogenous structure ; or where the pith of an evidently exogenous stem, as in the Piperaceae, has scattered woody bundles in an endogenous fashion ; or where monocotyledonous plants have all their woody bundles in a definite circle, as in Luzula, Croomia, &c. FIG. 126. Section of a small Palm-stem, in two directions. 70 MORPHOLOGY OF STEMS. the whole section, or when few in number of somewhat definite position or arrangement. The ordinaiy appearance of such a stem, both on the longitudinal and the cross-section, is shown in Fig. 126 ; it may also be examined in the Cane or Rattan, the Bamboo, and in the annual stalk of Indian Corn or of Asparagus. The appearance of ordinary wood is very familiar. 135. The newer woody bundles of an endogenous stem are vari- ously intermingled with the old. When, DeCandolle gave the name, it was supposed, from Desfontaines's researches, that the older bundles occupied, or came at length to occupy, the circumference of the trunk, while only new ones were formed in the centre ; and that increase in diameter, when it took place at all, resulted from the gradual growth and distention of the whole. Hence the contrasting name of endogenous, or inside growing, and for such plants the name of ENDOGENOUS PLANTS, or ENDOGENS. Our actual knowledge of the structure and growth of these stems, as will be seen, cannot be harmonized with this view in any way which gives to the name endogenous an appropriate signifi- cation. The name continues as a counterpart to the more correct one of exogenous, and as a survival of former ideas. 136. The Endogenous Structure (so called) of the stem is cor- related with a monoeotyledonous embiyo (39), usually with a ternary arrangement in the flower (322), and commonly with parallel- veined leaves. (173.) Endogens, although they have many herbaceous and a few somewhat woody representatives in cool temperate climes, mostly attain their full variety of fea- tures and rise to noble arborescent forms under a tropical sun. Yet Palms — the arboreous type of the class — do extend as far north in this country as the coast of North Carolina (the natural limit of the Palmetto, Fig. 126°) ; while in Europe the Date and the Chamasrops thrive in the warmest parts of the Euro- pean shore of the Mediterranean. The* manner of their growth gives them a striking appearance ; their trunks being unbranched cylindrical columns, rising to the height of from thirty to one hundred and fifty feet, and crowned at the summit with a simple cluster of peculiar foliage. Palms generally grow from the terminal bud alone, and perish if this bud be destroyed ; the}' grow slowly, and bear their foliage in a cluster at the summit of the trunk, which consequently forms a simple cylindrical column. But in some instances two or more buds develop, and the stem branches, rarely and accidentally in ordinary species, regularly in the Doum Palm of Upper Egypt, and in the Pandanus, or Screw-Pine (Fig. 69), which belongs to a family allied to Palms : in such cases the branches are cylindrical. But when lateral INTERNAL STRUCTURE. 71 buds are freety developed (as in the Asparagus) , or the leaves are scattered along the stem or branches b}' the full development of internodes (as in the Bamboo, Maize, &c. ) , they gradually taper upward in the manner of most ex- ogenous stems. 187. This kind of stem comprises several subordinate types as to internal structure, which to be well understood must be studied his- tologically , under the microscope.1 To one of these, by no means , the simplest, belongs the ordinary palm- stem, the anatomy of which was made classical by Mohl, and has been supplemented by Na?geli. In this a large part of the bundles, or all of the more conspicuous kind, starting from the base of the leaf to which they respectively belong, curve inward more or less strongly toward the centre of the stem, and thence gradual!}' outward as they descend until they reach the rind, in which the attenuated lower extremity mostly terminates. Consequently, the bundles from different heights cross in their course, somewhat 1 For the best and most accessible memoir on the subject, of recent date, see Guillaud, Recherches sur 1'Anatomie comparee et le Devcloppement des Tissus de la Tige dans les Monocotyle'dones, published in Ann. Sci. Nat. ser. 6, v. 1-176, 1877. Six types of the stem of Monocotyledons are here recognized by anatomical characters and modes of growth, one of them having four modifications. FIG. 126- Sabal Palmetto in various stages; also the Yucca aloifolia or Spanish Bayonet. 72 MORPHOLOGY OF STEMS. as shown in Fig. 127. It is partly owing to this connection of these fibres with the rind that the latter is not separable from the stem. In some Palms, and in Grasses, there is no marked distinction between the wood and rind, or no proper rind at all. In others, such as the Palmetto (Fig. 126), there is a marked rind or false bark, .which receives independent fibro- vascular bundles from the leaf-stalks, and is traversed by them in parallel lines. In Grass-stems, and others with long inter- nodes and closed nodes, the fibro-vascular bundles all run approximately straight and parallel through the internodes, but are intricate and anastomosed in the nodes. The whole centre of the internodes, when not hollow or before it becomes so, is occupied by a true pith, like that of an Exogen, and in some»cases equally destitute of fibro-vascular bundles, but often with scattered ones, after the manner of certain Exogens anomalous in this respect, such as Nyctaginacese and some Araliaceae. Endogenous stems of simpler structure, as in herbaceous Liliaceae, Commelynaceae, &c., have a distinct cortical portion (at least in the root-stock or portion of stem properly comparable with palm-trunks and the like) ; but this is mostl}' destitute of fibro-vascular bundles. Most of them have two kinds of vascular bundles, one of which not rarely occupies an exact circle in the line of division between the cortical and medullary portion (between bark and pith), and the other is within this circle, either of very few and scattered bundles, as in Convallaria majalis, or numerous and scattered, as in Uvularia and the leafy stems of Tradescantia Virginica ; or these bundles are few and arranged nearly in an inner circle close around the centre. Finally, Luzula and Croomia have only one kind of bundles, answering to the outer ones of Convallaria ; in other words, the woody system forms a simple circle, dividing a purely cellular medullary from a similar cortical portion, thus closely imitating an herbaceous exogenous stem of the same age. 138. An annual endogenous stem increases in diameter by general growth until it attains its limit. Ligneous and enduring stems increase similarly up to a certain period. Then the rind PIG. 127. Diagrammatic view of the curved course of the fibro-vascular bundles ill a palm-trunk. INTERNAL STRUCTURE. 73 soouer or later ceases to distend or adapt itself to further in- crease in diameter, and there is no interior provision for indefinite increase in the greater number of woody endogenous trunks. But in Dracaena (Dragon-trees) , in the arborescent Yuccas, and the like, the zone intermediate between the cortical and interior re- gion, which is for a time active in many Endogens, here grows continuously and indefinitely. Such trunks increase in diameter throughout life ; they may attain a very great age (as some Dragon-trees have done) ; and they imitate exogenous trunks to a considerable extent in mode of growth. 139. The wood of an endogenous woody stem is hardest and most compact at the circumference ; in palm-stems commonly it is largely mixed with parenchyma or pith at the centre, even in old trunks. 140. The Exogenous Structure, that of ordinary wood, is char- acterized by the formation of a distinct zone of wood between a central cellular medullary portion (pith) and an outer chiefly cellular portion (bark) , traversed by plates from the pith (medul- lary rays) , and by increasing from the outer surface of this zone between wood and bark, the increase in enduring stems consist- ing of definite concentric annual layers. 141. Its Beginning, at the earliest growth of the embryo, is in the appearance of a few ducts (Fig. 125,/-£), at definite points in the common parenchyma of the initial stem (four equidistant ones in the Sugar Maple) ; each is soon surrounded by incipient 128 129 130 proper wood-cells (Fig. 1 25 , 5, c) , together forming a fibro- vascular bundle or thread. Additional ones are intercalated as the second and third internodes develop, and so a column (in cross- section a ring) of wood is produced, always so arranged as to . FIG. 128. Diagram of a cross-section of a forming seedling stem, showing the manner in which the young wood is arranged in the cellular system. FIG. 129. The same at a later period, the woody bundles increased so as nearly to fill the circle. FIG. 130. The same at the close of the season, where the wood has formed a com- plete circle, interrupted only by the medullary rays, which radiate from the pith to the bark. V 74 MORPHOLOGY OF STEMS. surround a purely cellular central part (the pith), while sur- rounded by a cellular external rind, the bark, or outer bark. The diagrams (Fig. 128-130) rudely show some stages in tne formation of the zone of wood. The fibro-vascular bundles originate in the bases of the leaves, and develop outward into the forming leaves as well as downward into the forming stem. 142. First Year's Growth. The wood, even in a herbaceous or annual stem, at the completion of the first year's growth, forms a zone or tube, enclosing the pith. But it is traversed by plates (in cross-section lines) of parenchyma, or eel lular tissue of the same nature as the pith, which radiate from that to the bark, and thus divide the wood into wedges. These lines, forming what is called the silver-grain in wood, are the MED- ULLARY RAYS. They represent the cellular system of the wood it- self, or untransformed parenchyma. Being pressed by the woody wedges, their cells are laterally flattened. In some stems, the med- ullary rays, or many of them, are comparatively broad and conspicuous ; in others, thin and inconspicuous or irregular. The growth of the woody wedges is soon complete, except at the outer portion, next the bark : here they usually continue to grow through the season ; that, is the wood grows externally. The general ana- FIG. 131. Longitudinal and transverse section of a stem of the Soft Maple (Acer dasycarpnm), at the close of the first year's growth ; of the natural size. FIG. 132. Portion of the same magnified, showing the cellular pith, surrounded by the wood, and that by the bark. FIG. 133. More magnified slire of the same, reaching from the bark to the pith: a. part of the pith ; b. vessels of the medullary sheath ; c. the wood ; d, d. dotted ducts in the wood; e,e. annular ducts;/, the liber, or inner fibrous hark; g. the cellular envelope, or green bark; h. the corky envelope; i. the skin or epidermis; k. one of the medullary rays, seen on the transverse section. INTERNAL STRUCTURE. 75 tomicai structure of a woody exogenous stem of a year old is displayed in the Fig. 131-133. Viewing the parts particularly, and in order from centre to circumference, there is, — 1st. The Pith or Medulla, consisting entirely of soft and rather large thin-walled cells,1 gorged with sap or other nourishing matter during the growing state, becoming light, dry, and empty when effete. 2nd. The Layer of Wood, traversed by the medullary rays. In Pines and other Coniferae, the wood is of uniform structure, being wholly composed of a woody tissue with peculiar markings (Fig. 125, d, e) : in other wood, ducts of one or more sorts occur ; the most conspicuous being what are termed dotted ducts. These are so large as to be evident to the naked eye in many ordi- nary kinds of wood, especially where they are accumulated in the inner portion of the layer, as in the Chestnut and Oak. In the Maple, Plane, &c., they arc rather equably scattered through the annual layer, and are too small to be seen by the naked eye. Next the pith, i. e. in the very earliest formed part of the wood, some spiral ducts are uniformly found, and this is the only part of the exogenous stem in which these ordinarily occur. They may be detected by breaking a woody twig in two, after dividing the bark and most of the wood by a circular incision, and then pulling the ends gently asunder, when their spirally coiled fibres are readily drawn out as gossamer threads. As these spiral ducts form a circle immediately surrounding the pith, they have collectively been termed the medullary sheath, but they hardly deserve a special name. The vertical section in Fig. 133 divides one of the woody wedges, and shows no medullary ray ; but there is one at the posterior edge of the transverse section. But, in the much more diagramatic Fig. 134, the section is made so as to show the surface of one of these plate's, or medullary rays, passing hori- zontally across it, connecting the pith (/>) with the bark (^). These medullary rays form the silver-grain (as it is termed) , which is so conspicuous in the Maple, Oak, &c., and which gives the glimmering lustre to many kinds of wood when cut in this direc- tion. A section made as a tangent to the circumference, and therefore perpendicular to the medullary raj-s, brings their ends to view, as in Fig. 135, much as they appear on the surface of a piece of wood from which the bark is stripped. They are here seen to be composed of parenchyma, and to represent the horizon- 1 In rare instances, a few fibro-vascular threads are found dispersed through the pith, presenting a somewhat remarkable anomaly. This occurs in Aralia racemosa, and more strikingly in Mirabilis and other Nyctaginacese, and in Piperaceae. (133. foot-note.) ~f £? 76 MORPHOLOGY OF STEMS. tal system of the wood, or the woof, into which the vertical woody fibre, &c., or warp, is interwoven. The inspection of a piece of oak or maple wood at once shows the pertinency of this illustration. 3rd. The Bark or rind. This at first consisted of simple parenchyma, like that of the pith, except for the green color developed in it,-the same as that which gives verdancy to foliage. This green matter is formed in the cells of all such parts when exposed to light, consists of green grains of somewhat complex chemical composition, has important functions to perform in assimilation (i. e. in the conversion of the plant's crude food into vegetable matter), and is named CHLOROPHYLL, i. e. leaf-green. The completed bark, when all its parts are apparent, as espe- cially in most trees and shrubs, is composed of three strata, of which the green bark, the most conspicuous in the young shoot, is the middle laj-er, therefore named the MESOPHLCEUM. This is soon covered, and the green color obscured, by a superficial stratum of cells, generally of some shade of ash-color or brown, occasional!}' of brighter tints, which gives to the twigs of trees and shrubs the hue characteristic of each species, the CORKY ENVELOPE or layer, or EPIPHLCEUM. The latter name denotes its external position ; the former, that it is the layer which, when much developed, forms the cork of Cork-Oak and those corky expansions which are so conspicuous on the twigs of the Sweet Gum (Liquidambar) , and on some of our Elms (Ulmus alata and racemosa) . It also forms the paper-like exfoliating layers of Birch-bark. It is composed of laterally flattened parenchy- matous cells, much like those of the EPIDERMIS (Fig. 133, z), which directly overlies it, and forms the skin or surface of the FIG. 134. Vertical section through the wood of a branch of the Maple, a year old, BO as to show one of the medullary rays, passing transversely from the pith (p) to the bark (6): magnified. But a section can seldom be made so as to show one unbroken plate stretching across the wood, as In this instance. FIG. 135. A vertical section across the ends of the medullary rays: magnified. INTERNAL STRUCTURE. 77 stem, and of the whole plant. Lastly, the inner bark, accord- ingly named ENDOPHLCEUM, takes the special name of LIBER, and is the most important portion of the bark in the stems of trees and shrubs. Complete and well-developed liber, like that of Linden or Basswood, contains two peculiar kinds of cells in addition to common parenchyma, both of the fibrous or vascular class: viz., 1. CRIBRIFORM or SIEVE-CELLS, a sort of ducts the walls of which have open slits, through which they communicate with each other; 2. BAST or BAST-CELLS, the fibre-like cells which give to the kinds of inner bark that largely contain them FIG. 135°. Portion of a transverse section (above), and a corresponding vertical sec- tion (below), magnified, reaching from the pith (p) to the epidermis (e) of a stem of Negundo, a year old : B. the bark ; W. the wood ; and C. the cambium-layer, as found in February. The parts referred to by small letters are : p. a portion of the pith : mr. small portion of a medullary ray where it runs into the pith; four complete med- ullary rays as seen on a transverse section, appear in the upper figure, running from pith to bark : ms. medullary sheath, a circle of spiral unreliable ducts, one seen length- wise with uncoiling extremity in the lower figure : to, w. woody tissue : dd. one of the dotted ducts interspersed in the wood : cl. cambium-layer or zone of new growth of •wood and inner bark : l-b. liber or inner bark, the inner portion of which is here cellular, the outer (6) composed of slender and thick-walled bast-cells or true liber-cella: ge. green envelope or inner cellular bark: ce. corky envelope or outer cellular bark: e. epidermis. 78 MORPHOLOGY OF STEMS. their strength and toughness. The}- are like wood-cells except in their greater length and flexibility, and in the thickness of their walls, which greatly exceeds the calibre. This is the material which gives to the bast or inner bark of Basswood, &c., the strength and pliability that adapts it for cordage and for making mats : it is the material of linen, and the like textile fibres. (For a view of the whole composition and structure of a woody stem at the close of the first year's growth, and immedi- ately before that of the second year begins, see Fig. 135°.) 143. Annual Increase in Diameter. An herbaceous stem does not essentially differ from a woody one of the same age, except that the wood forms a less compact or thinner zone ; and the whole perishes, at least down to the ground, at the close of the season. But a woody stem makes provision for continuing its growth from year to year. As the layer of wood continues to increase in thickness throughout the season, by the multiplication of cells on its outer surface, between it and the bark, and when growth ceases this process of cell-multiplication is merely sus- pended, so there is always a zone of delicate young cells in- terposed between the wood and the bark. This is called the CAMBIUM, or, better, the CAMBIUM-LAYER. It is charged with organizable matter, which is particularly abundant and mucila- ginous in spring when growth recommences. This mucilaginous matter was named Cambium b}" the older botanists : they sup- posed— as is still popularly thought — that the bark, then so readily separable, really separated from the wood in spring, that a quantity of rich mucilaginous sap was poured out between them and became organized into a tissue, the inner part becoming new wood, the outer, new bark. But delicate slices show that there is then no more interruption of the wood and inner bark than at any other season. The bark, indeed, is then very readily detached from the wood, because the cambium-layer is gorged with sap ; but such separation is effected b}- the rending of a delicate forming tissue. And if some of this apparent mucilage be scraped off from the surface of the wood, and examined under a good microscope, it will be seen to be a thin stratum of young wood-cells, with the ends of medullary rays here and there in- terspersed. The inner portion of the cambium-layer is therefore nascent wood, and the outer is nascent bark. As the cells of this layer multiply, the greater number lengthen vertically into woody tissue : some are transformed into ducts ; and others, remaining as parenchyma, continue the medullary rays or com- mence new ones. In this way, a second layer of wood is formed the second season over the whole surface of the former layer INTERNAL STRUCTURE. 79 between it and the bark ; and this is continuous with the wood}- layer of the new roots below and of the leafy shoots of the sea- son above. Each succeeding year another layer is added to the wood in the same manner, coincident with the growth in length by the development of the buds. A cross-section of an exoge- nous stem, therefore, exhibits the wood disposed in concentric rings between the bark and the pith ; the oldest lying next the latter, and the }-oungest occupying the circumference. Each layer being the product of a single year's growth, the age of an exogenous tree may, in general, be correctly ascertained by counting the rings in a cross-section of the trunk.1 144. Demarcation of the Annual Layers results from two or more causes, separate or combined. In oak and chestnut wood, and the like, the layers are strongly defined by reason of the accumu- lation of the large dotted ducts (here of extreme size and in great abundance) in the inner portion of each layer, where their open mouths on the cross-section are conspicuous to the naked eye, making a strong contrast between the inner porous and the exterior solid part of the successive layers. In maple and beech \vood, however, the ducts are smaller, and are dispersed through- out the whole breadth of the layer; and in coniferous wood, viz. that of Pine, Cypress, &c., there are no ducts at all, but only a uniform woody tissue of a peculiar sort. In all these, the de- marcation between two layers is owing to the greater fineness of the wood-cells formed at the close of the season, viz. those at the outer border of the layer, while the next layer begins, in its 1 The annual layers are most distinct in trees of temperate climates like ours, where there is a prolonged period of total repose, from the winter's cold, followed by a vigorous resumption of vegetation in spring. In tropical trees, they are rarely so well defined ; but even in these there is generally a more or less marked annual suspension of vegetation, occurring, however, in the dry and hotter, rather than in the cooler season. There are numerous cases, moreover, in which the wood forms a uniform stratum, whatever be the age of the trunk, as in the arborescent species of Cactus ; or where the layers are few and by no means corresponding with the age of the trunk, as in the Cycas. In many woody climbing or twining stems, such as those of Clematis, Aristolochia Sipho, and Menispermum Canadense, the annual layers are rather obscurely marked, while the medullary rays are unusually broad ; and the wood, therefore, forms a series of separable wedges disposed in a circle around the pith. In the stem of Bignonia capreolata, the annual rings, after the first four or five, are interrupted in four places, and here as many broad plates of cellular tissue, belonging properly to the bark, are inter- posed, passing at right angles to each other from the circumference towards the centre, so that the transverse section of the wood nearly resembles a Maltese cross. But these are exceptional cases, which scarcely require notice in a general view. 80 MORPHOLOGY OF STEMS. vigorous vernal growth, with much larger cells, thus marking an abrupt transition from one layer to the next. Besides being finer, the later wood-cells of the season are commonby flattened antero-posteriorly, probably b}~ growing under greater pressure. 145. Each layer of wood, once formed, remains essentially unchanged in position and dimensions. But, in trunks of con- siderable age, the older layers undergo more or less change in color, density, perviousness to moisture, &c. 146. Sap-wood (ALBURNUM). In the plantlet and in the developing bud, the sap ascends through the whole tissue, of whatever sort : at first through the parenchyma, for there is then no other tissue ; and the transmission is continued through it, especially through its central portion, or the pith, in the growing apex of the stem throughout. But, in the older parts below, the pith, soon drained of sap, becomes filled with air in its place, and thenceforth it bears no part in the plant's nourishment. As soon as wood-cells and ducts are formed, they take an active part in the conveyance of sap, for which their tubular and ca- pillary character is especially adapted. But, the ducts in older parts, except when gorged with sap, contain air alone ; and in woody trunks the sap continues to rise year after year to the places where growth is going on, mainly through the proper woody tissue of the wood. In this transmission, the new layers are most active ; and these are in direct communication with the new roots on the one hand and with the buds or shoots and leaves of the season on the other. So, by the formation of new annual layers outside of them, the older ones are each year removed a step farther from the region of growth ; or rather the growing stratum, which connects the fresh rootlets that imbibe with the foliage that elaborates the sap, is each year removed farther from them. The latter, therefore, after a few years, cease to convey sap, as they have long before ceased to take part in any vital operations. The cells of the older layers, also, usually come to have thicker walls and smaller calibre than those of the newer. Thus arises a distinction — sometimes obscurely marked, some- times abrupt and conspicuous — into sap-wood and heart-wood. The former is the popular name given to the outer and newer layers of softer, more open, and bibulous wood. The early physi- ologists named it alburnum from its white or pale color. Being more or less sappy, or containing soluble organic matter, and readily imbibing moisture, this part of the wood is h'able to decay, and it is therefore discarded from timber used for construction. 147. Heart-wood (or DURAMEN, so called from its greater hard- ness or durability) is the older and mature portion of the wood. INTEBNAL STRUCTURE. 81 In all trees which have the distinction between the sap-wood and heart- wood well marked, the latter acquires a deeper color, and that peculiar to the species, such as the dark brown of the Black Walnut, the blacker color of the Ebony, the purplish-red of Red Cedar, and the bright yellow of the Barberry. These colors are owing to special vegetable products, or sometimes to alterations resulting from age. In the Red Cedar, the deep color belongs chiefly to the medullary rays. In many of the softer woods, there is little change in color of the heart- wood, except from incipient decay, as in the White Pine, Poplar, Tulip-tree, &c. The heart-wood is no longer in any sense a living part : it may perish, as it frequently does, without affecting the life or health of the tree. 148. The Growth and Duration of the Bark, also the differences in structure, are much more various than of the wood. Moreover, the bark is necessarily subject to grave alterations with advanc- ing age, on account of its external position ; to distention from the constantly increasing diameter of the stem within, and to abrasion and decay from the influence of the elements without. It is never entire, therefore, on the trunks of large trees ; but the dead exterior parts, no longer able to enlarge with the en- larging wood, are gradually fissured and torn, and crack off in strips or pieces, or disappear by slow decay. So that the bark of old trunks bears only a • small proportion in thickness to the wood, even when it makes an equal amount of annual growth. 149. The three parts of the bark (142), for the most part readily distinguishable in the bark of j'oung shoots, grow inde- pendentty, each by the addition of new cells to its inner face, so long as it grows at all. The green layer commonly does not increase after the first }-ear ; the opaque corky layer soon excludes it from the light ; and it gradually perishes, never to be renewed. The corky layer usually increases for a few years only, by the formation of new tabular cells : occasionally it takes a remarkable development, forming the substance called Cork, as in the Cork vJak, and the thin and parchment- like layers of the White and Paper Birches. 150. The liber, or inner bark, continues its growth through- out the life of the exogenous tree, by an annual addition from the cambium-laver applied to its inner surface. Sometimes this growth is plainly distinguishable into layers, corresponding with or more numerous than the annual layers of the wood : often, there is scarcely any trace of such layers to be discerned. In composition and appearance, the liber varies greatly in different plants, especially in trees and shrubs. That of Basswood or 82 MORPHOLOGY OF STEMS. Linden, and of other plants with a similar fibrous bark, may be taken as best representing the liber. Here it consists of alter- nate strata of fibrous bast, and of the peculiar liber-cells called sieve-cells, in which nourishing matter is especially contained and elaborated. While the latter, or their equivalents, occur and play an important part in all inner bark, the bast-cells are altogether wanting in the bark of some plants, and are not pro- duced after the first year in many others. The latter is the case in Negundo, where abundant bast-cells, like those of Basswood, compose the exterior portion of the first year's liber, but none whatever are formed in the subsequent layers. In Beeches and Birches, also, a few bast-cells are produced the first year, but none afterwards. In Maples, a few are formed in succeeding }-ears. In the Pear, bast-cells are annually formed, but in very small quantity, compared with the parenchymatous part of the liber. In Pines, at least in White Pines, the bark is nearby as homogeneous as the wood, the whole h'ber, except what answers to the medullary rays, consisting of one kind of cells, resembling those of bast or of wood in form, but agreeing with the proper liber-cells in their structure and markings. 151. The bark on old stems is constant!}* decaying or falling away from the surface, without any injury to the tree ; just as the heart- wood within may equally decay without harm, except by mechanically impairing the strength of the trunk. There are great differences as to the time and manner in which the older bark of different shrubs and trees is thrown off. Some have their trunks invested with the liber of many years' growth, although only the innermost layers are alive ; in others, it scales off much earlier. On the stems of the common Honeysuckle, of the Nine-Bark (Spiraea opulifolia), and of Grape-vines (except Vitis vulpina) , the h'ber lives only one season, and is detached the following year, hanging loose in papery layers in the former species, and in fibrous shreds in the latter. 152. While the newer laj-ers of the wood abound in crude sap. which they convey to the leaves, those of the inner bark abound in elaborated sap, which they receive from the leaves and convey to the cambium-layer or zone of growth. The proper juices and peculiar products of plants are accordingly found in the foliage and the bark, especially in the latter. In the bark, therefore (either of the stem or of the root) , medicinal and other principles are usually to be sought, rather than in the wood. Nevertheless, as the wood is kept in connection with the bark by the medullary ra3*s, many products which probably originate in the former are deposited in the wood. INTERNAL STRUCTURE. 83 153. The Living Parts of a tree or shrub, of the exogenous kind, are obviously only these: 1st, The summit of the stem and branches, with the buds which continue them upwards and annually develop the foliage. 2d, The fresh roots and rootlets annually developed at the opposite extremity. 3d. The newest strata of wood and bark, and especially the interposed cambium- layer, which, annuall}- renewed, maintain a living communication between the rootlets on the one hand and the buds and foliage on the other, however distant they at length may be. These are all that is concerned in the life and growth of the tree ; and these are annualty renewed. The branches of each year's growth are, therefore, kept in fresh communication, by means of the newer layers of wood, with the fresh rootlets, which are alone active in absorbing the crude food of the plant from the soil. The fluid they absorb is thus combed directly to the branches of the sea- son, which develop leaves to digest it. And the sap they receive, having been elaborated and converted into organic nourishing matter, is partly expended in the upward growth of new branches, and partly in the formation of a new layer of wood, reaching from the highest leaves to the remotest rootlets. 154. Longevity of trees. As the exogenous tree, therefore, annually renews its buds and leaves, its wood, bark, and roots, — every thing, indeed, that is concerned in its life and growth, — there seems to be no necessary cause, inherent in the tree itself, why it may not live indefinitely. Some trees are known to have lived for one and two thousand 3'ears, and some are possibly older.1 Equally long may survive such endogenous trees as the Dragon tree (Dracaena) , which have provision for indefinite in- crease in diameter (138), and for the production of branches. The famous Dragon tree of Orotava, in Teneriffe, now destrc^cd by hurricanes and other accidents, had probably reached the age of more than two thousand j^ears. 155. On the other hand, increase in height, spread of branches and length of root, and extension of the surface over which the annual layer is spread, are attended with inevitable disadvantage, which must in time terminate the existence of the tree in a way quite analogous to the death of aged individual animals, which is not directly from old age, but from casualties or attacks to 1 The subject of the longevity of trees has been discussed by DeCandolle, in the "Bibliotheque Universelle " of Geneva, for May, 1831, and in the second volume of his " Physiologic Vegetale ; " more recently, by Alphonse DeCan- dolle in the " Bibliotheque Universelle ; " and in this country by myself in the " North American Review," for July, 1844. For an account of the huge Red- woods (Sequoias) of California, see Whitney's Yosemite Book. 84 MORPHOLOGY OF STEMS. which the aged are either increasingly incident or less able to resist. A tree like the Banyan (59, Fig. 71), which by aerial roots continues to form new trunks for the support and sustenance of the spreading branches, and thus ever advances into new soil, has a truly indefinite existence ; but, then, it becomes a forest, or is to be likened to a colony propagated and indefinite!}' in- creased by suckers, offsets, or other subterranean shoots. Ho the question of the secular continuation of the individual plant becomes merged in that of continuation of the race, — at least of a bud-propagated race, — the answer to which is wholly in the domain of conjecture.1 However this may be, it is evident that a vegetable of the higher grade is not justly to be compared with an animal of higher grade ; that individuality is incompletely realized in the vegetable kingdom ; 2 that rather 156. The Plant is a Composite Being, or Community, lasting, in the case of a tree, through an indefinite and often immense num- ber of generations. These are successive!}' produced, enjoy a term of existence, and perish in their turn. Life passes onward continually from the older to the newer parts, and death follows, with equal step, at a narrow interval. No portion of the tree is now living that was alive a few years ago ; the leaves die annu- ally and are cast off, while the internodes or joints of stem that bore them, as to their wood at least, buried deep in the trunk under the wood of succeeding generations, are converted into lifeless heart-wood, or perchance decayed, and the bark that belonged to them is thrown off from the surface. It is the aggre- gate, the blended mass alone, that long survives. Plants of single cells, and of a definite form, alone exhibit complete indi- viduality ; and their existence is extremely brief. The more complex vegetable of a higher grade is not to be compared with the animal of the highest organization, where the offspring always separates from the parent, and the individual is simple and indi- visible. But it is truly similar to the branching or arborescent coral, or to other compound animals of the lowest grade, where successive generations, though capable of living independently and sometimes separating spontaneously, yet are usually devel- oped in connection, blended in a general bod}', and nourished more or less in common. Thus, the coral structure is bnilt up by the combined labors of a vast number of individuals, — by the successive labors of many generations. The surface or the recent shoots only are alive ; beneath are only the dead remains 1 See Darwiniana, xii. 338-355. 2 As, perhaps, was first explicitly stated by Engelmann, in his inaugural essay, De Antholysi Prodromus, Introduction, § 4. MOliPHOLOGY OF LEAVES. 85 of ancestral generations. As in a genealogical tree, only the later ramifications are among the living. The tree differs from the coral structure in that, as it ordinarily imbibes its nourish- ment mainl}' from the soil through its roots, it makes a downward growth also, and. by constant renewal of fresh tissues, maintains the communication between the two growing extremities, the buds and the rootlets. Otherwise, the analogy of the two, as to individuality, is well-nigh complete. SECTION IV. OF LEAVES. § 1. THEIR NATURE AND OFFICE. 157. Leaf (Lat. Folium, in Greek form Phyllum), as a botani- cal term, has on the one hand a comprehensive, on the other a restricted sense. In its commonest sense, as used in descriptive botany, it denotes the green blade only. Yet it is perfectly understood that the footstalk is a part of the leaf, and therefore that the phrase " leaves cordate," or the like, is a short way of saying that the blade of the leaf is cordate or heart-shaped. Moreover, two appendages, one on each side of the base of the footstalk, when there is any, are of so common occurrence that they are ranked as a proper part of the organ. So that, to the botanist, a typical leaf consists of three parts: 1, BLADE or LAMINA ; 2, FOOT-STALK or LEAF-STALK, technically PETIOLE ; 3, A pair of STIPULES. (Fig. 142.) 158. The blade, being the most important part of an ordinary leaf, may naturally be spoken of as the whole. Petiole and stipules are indeed subsidiary when present, and are not rarely wanting. Yet sometimes they usurp the whole function of foli- age, and sometimes there is no such distinction of parts. 159. Physiologically, leaves are green expansions borne by tho stem, outspread in the air and light, in which assimilation (3) and the processes connected with it are carried on. Vegetable assimilation, — the most essential function of plants, being the conversion of inorganic into organic matter, — takes place in all ordinary vegetation onry in green parts, and in these when exposed to the light of the sun. And foliage is an adaptation for largely increasing the green surface. But stems, when green, take part in this office in proportion to the amount of surface, sometimes monopolize it, and in various cases increase their means of doing so by assuming leaf-like forms. (126-129.) Leaves, especially in such cases, ma}' lose this function, appear only as useless vestiges, or may be subservient to various wholly 86 MORPHOLOGY OF LEAVES. different uses. Form and function, therefore, are not sure indi- cations of the true nature of organs. 160. Morphologically, and in the most comprehensive sense, leaves are special lateral outgrowths from the stem, definitely and symmetrically arranged upon it ; in ordinary vegetation and in the most general form constituting the assimilating apparatus (or foliage), but also occurring in other forms and subserving various uses. Sometimes these uses are combined with or sub- sidiary to the general function of foliage ; sometimes the leaf is adapted to special uses only. So the botanist — recognizing the essential identity of organs, whatever their form, which appear in the position and conform to the arrangement of leaves — discerns the leaf in the cot}*ledons of a bean or acorn, the scale of a lily-bulb or the coat of an onion, the scale of a winter bud, and the petal of a blossom. Therefore, while expanded green leaves (which may be tautologically termed foliage-leaves) are taken as the proper type, the common name of leaves, in the lack of any available generic word, is in morphological language extended to these special forms, whenever it becomes needful to express their \)hylline or foliar nature. 161. In the morphological view, all the plant's organs except- ing roots (and excepting mere superficial productions, such as hairs, prickles, &c.), belong either to stem or to leaves, are either cauline or phylline in nature. To the latter belong all the primary outgrowths from nodes, all lateral productions which are not axillary.1 Whatever is produced in the axil of a leaf is cauline, and when developed is a branch. 162. The Duration of Leaves is transient, compared with that of the stem. They may be fugacious, when they fall off soon after their appearance ; deciduous, when they last only for a single season ; and persistent, when they remain through the cold season, or other interval during which vegetation is interrupted, and until after the appearance of new leaves, so that the stem is never leafless, as in Evergreens. In many evergreens, the leaves have only an annual duration ; the old leaves falling soon after those of the ensuing season arc expanded, or, if they remain longer, ceasing to bear any active part in the econom}' of the vegetable, and soon losing their vitality altogether. In Pines and Firs, however, although there is an annual fall of leaves either in autumn or spring, yet these were the produce of some 1 There are cases in which this rule is of difficult application, or is seem- ingly violated, sometimes by the suppression of the subtending leaf, as in the inflorescence of Cruciferae, rarely in other ways, to be explained in the proper places. THEIR STRUCTURE AND FORMS. 87 season earlier than the last ; and the branches are continually clothed with the foliage of from two to five, or even ten or more successive years. On the other hand, it is seldom that all the leaves of an herb endure through the whole growing season, the earlier foliage near the base of the stem perishing while fresh leaves are still appearing above. In our deciduous trees and shrubs, however, the leaves of the season are mostly developed within a short period, and they all perish in autumn nearly simultaneously. 163. Leaves soon complete their growth, and have no power of further increase. Being organs for transpiration, a very large part of the water imbibed by the roots is given out by the foliage, leaving dissolved earthy matters behind. Assimilation can take place only in fresh and vitalty active tissue. It is incident to all this that leaves should be of only transient duration, at least in their active condition. 164. Defoliation. The leaves of most Dicotyledons and some Monocotyledons separate from the stem and fall by means of an articulation at the junction with the stem, which begins to form early in the season and is completed at the close. There is a kin chlorophyll in their paren- chyma, exchange the ordinary green hue and herbaceous texture for the brighter colors and more delicate structure which are commonly seen in and thought to characterize flower-leaves. Such are said to be colored, meaning, as applied to foliage, of some other color than green. As petals are the type of such colored parts, the}' are said to be petaloid, i. e. petal-like. The}- are like petals, moreover, in one of the purposes which these sub- serve. (299.) Examples of these petaloid leaves are seen in the shrubby Mexican Euphorbia called Poinsettia, in Salvia splen- dens, most species of Castilleia or Painted Cup, also in the white hood of Calla and Richardia ^Ethiopica (called Calla Lily), and in the four white leaves which subtend the flower-head of Cornus florida, and of the low herbaceous Cornel, C. Canaden sis. (Fig. 294.) Such leaves, being in proximit}- to flowers, and all others which are within a flower-cluster or are borne by flower-stalks, receive the special name of BRACTS. More usually bracts are not petaloid, but different in size or shape from ordi- nary leaves, either by abrupt change or gradual transition. Not uncommonly they are reduced to scales or mere rudiments or vestiges of leaves, of no functional importance. 230. Flower- Leaves. The morphology of leaves extends not only to " the leaves of the blossom," more or less accounted as such in common parlance, but also to its peculiar and essential organs, the relation of which to leaves is more recondite. Their morphology needs to be treated separately, and to be preceded by a study of the arrangement of leaves and of blossoms. 1 The most satisfactory interpretation may be that of Braun and Wydler, adopted by Eichler (Bliithendiagramnie, i. 304) : that the flower of Cucur- bita and its peduncle represent the axillary branch, the tendril by its side answers to one of the bractlets (that of the other side being suppressed), and the supernumerary branch springs from the axil of the tendril. This makes of the tendril a simple leaf, of which the branches arc the ribs. But the tendril-divisions are evidently developed in spiral order, and in vigorous growths occupy different heights on the tendril-axis. This favors Naudin's view, that the main tendril is cauline, and its divisions leaves. PHYLLOTAXY, OR LEAF-ARRANGEMENT. 119 CHAPTER IV. PHYLLOTAXY, OR LEAF-ARRANGEMENT. SECTION I. THE DISTRIBUTION OF LEAVES ON THE STEM. 231. PHYLLOTAXY (or Phyllotaxis} is the study of the distri- bution of leaves upon the stem and of the laws which govern it. The general conclusion reached is, that leaves are distributed in a manner to economize space and have a good exposure to light, &c., and that this econom}- on the whole results from the formation of leaves in the bud over the widest intervals between the leaves next below.1 Leaves are arranged in a consider- able variety of waj'S, which all fall under two modes, the Verticillate and the Alter- ' nate (13), but which nuiy also be termed the Cyclical j and the Spiral. 232. Alternate leaves are those which stand singly, one after another ; that is, with one leaf to each node or borne on one height of stem . Verticillate leaves arc those with two or more at the same height of stem, circularly encompassing it, i. e. forming a Verticil or Whorl. Verticillate and whorled are synonymous terms to denote this arrange- ment. These two kinds of leaf-arrangement ^ are commonly ranked as three, viz. alternate, opposite, and whorled. But the opposite i* only the simplest case of the whorled, being 1 For the most comprehensive discussion of phyllotaxy in connection with development, and in view of these relations, see Hofmeister, Allgemeine Morphologie, § 11, and Chauncey Wright, Mem. Amer. Academy, ix. 389. FIG. 236, Alternate, 237, Opposite, 238, Verticillate or whorled leaves. 120 PHYLLOTAXY, OR LEAP-ARRANGEMENT. that in which the members are reduced to two. This case is so much commoner than whorls of three and of higher numbers that it took from the first its special name of opposite, so that in descriptions the phrase " leaves verticillate " implies more than two leaves in the whorl. But it should be kept in mind that "leaves opposite " is the same as " leaves in whorls of two." 233. The greater number of phaenogamous plants (all but the monocotyledonous class) begin with verticillate leaves, mostly of the simplest kind (t. e. cotyledons opposite) : sonic continue verticillate throughout ; some change in the first leaves of the plumule or after the first pair into alternate, and again into verticillate in or toward the blossom, in the interior of which the alternate arrangement may be again resumed. As Nature passes >-adily from the one mode to the other on the same axis, we ma}' expect that the two may be comprised under some common expression. But they have not yet been combined, except by gratuitous or somewhat forced hypotheses ; so that for the present the}' should be treated in morphology as primarily dis- tinct arrangements.1 234. Yerticillate or Cyclical Arrangement. Here the leaves occupy a succession of circles, or form whorls around the stem, two, three, four, five, &c., in each whorl. According to the number, the leaves are opposite, ternate, quaternate, quinate, and so on. The characteristic of the individual whorl is that the members stand as far apart from each other as their number renders possible, i. e. they divide the circle equally. Thus, when only two, or opposite, their midribs or axes of insertion have an angular divergence (as it is termed) of 180°; when three, of 120° ; when four, 90° ; when five, 72°. 235. The characteristic of the whorls in relation to each other is, that the members of successive whorls stand over or under the intervals of the adjacent ones. In other words, successive whorls alternate or decussate. This economizes space and light, or gives the best distribution which the cyclical S3'stem is capa- ble of. And it is in accordance with the general conclusion of Hofmeister's investigation of the origin of pbyllotaxic arrange- ments in the nascent bud, viz. that new members originate just over the widest intervals between their predecessors next below. Thus, in opposite leaves or whorls of two (Fig. 237), the suc- cessive pairs decussate or cross at right angles, and so four 1 It is readily seen that whorls may be produced by the non-development of the internodes between the leaves of a series of two, three, five, or more in alternate order. The difficulty is that the members of the next whorl do not follow the rrder that they should upon this supposition. DISTRIBUTION OF LEAVES ON THE STEM. 121 straight equidistant vertical ranks are produced. In ternate or trimerous whorls there are six vertical ranks ; in quaternate or tetramerous whorls, eight vertical ranks, and so on.1 236. The cases in which successive pairs of leaves do not decussate at right angles, or the members of whorls are not exactly superposed to intervals, but as it were wind spirally (as in Dipsacus, man}- Caryophyllacese, &c.) , may some of them be explained by torsion of the stem, such as is very manifest in numerous in- stances ; and others may be resolved into instances of alternate leaves simulating or passing into whorls by the non-development of internodes.2 237. Alternate or Spiral Arrangement. Here the leaves are distributed singly at different heights of the stem, and at equal intervals as respects angular divergence. (Fig. 236.) This angu- lar divergence (i. e. the angular distance of any two successive leaves) differs in the various kinds of this system of phyllotaxy, but is alwa}*s large enough to place the leaves which immediately 1 These vertical ranks have, by some German botanists, been named Orthostichies ; but this technical Greek is no clearer and no shorter than the equivalent English, which answers every purpose. 2 In Lilium Canadense, superbum, &c., with whorls of variable number of leaves and vague relation to each other (when of the same number some- times the members superposed), and above and below passing into the alter- nate arrangement normal to the family, these whorls are evidently formed of alternate leaves brought together by non-development of internodes. Here may also be mentioned the not uncommon anomaly in Fir-cones, notably those of Norway Spruce, the normal phyllotaxy of which is simply spiral, but in occasional instances the cone is composed of pairs of opposite scales, spirally arranged, i. e. the pairs not decussating at right angles, thus forming double spirals. In the abnormal spruce-cones, the fractions usually observed are f$ or ^, or, as expressed by Braun, (i)-j^ and (£)^. Braun's mode of notation for the ordinary succession (t. e. the decussation ) of opposite leaves is (|)i, the ^ meaning that the two leaves of the pair are half the circumference of the circle apart, the £ denoting that each leaf of the succeeding pair diverges one fourth of the circumference from the pre- ceding. Braun finds cases in which pairs (and equally whorls) are super- posed (e.g. certain species of Mesembryanthemum and Euphorbia), these are expressed in this notation by the formula (£)£, that is, the corresponding leaves of the succeeding pair diverge 180° from their predecessors. He recognizes also some cases of intermediate divergence; such as (£)$ in the upper leaves of Mercurialis perennis, (-J-)-^ on certain stems of Linaria vul- garis, (i)^ exceptionally in the leaves of Epilobium angustifolium and the scales of Norway Spruce, (i)-j8^ exceptionally in the scales of Norway Spruce. See Ordnung des Schuppen an der Tannenzapfen, 376, &c. FIG. 239. Ground-plan diagram of six trimerous whorls, showing their alternation. 122 PHYLLOTAXY, OB, LEAF-ARRANGEMENT. follow each other in the ascending order upon different sides of the axis : it also secures an advantageous spacing of the leaves over the whole length of the axis. Their vertical distance from each other of course depends on the length which the internodes attain, which is a matter of growth and is veiy variable ; but their angular distance is fixed in the kind or numerical plan of the particular phyllotax}-, and is uniform throughout. 238. The leaves are said to be alternate, because the}' come one after another, now on this side, then on that, as they ascend the stem. The arrangement is said to be spiral, because if a line be drawn or a thread extended from the base or insertion of one leaf to that of the next higher, and so on, taking in all the leaves, it forms a helix, more or less loose or close according to the development of the internodes. (See Fig. 242.) This imagined spiral line ascends continuously, without a break ; and on it the leaves are equably laid down.1 239. Almost all the ordinary instances of spiral phyllotaxy belong to one series, having very simple arithmetical relations. So that this may be taken as the type, and the few others re- garded as exceptions or sometimes as modifications of it. The kinds are simpry designated by the number of vertical ranks of leaves : they are technically named by prefixing the proper Greek numeral to the word meaning row or rank. The arrange- ment called Distichous, or Two-ranked, is the simplest and among the com- /nonest, occurring, as it does, in all Grasses and many other aionocotyledonous plants, in Lindens, Elms, and many dico- tyledonous genera. Here the leaves are disposed alternately on exactly opposite sides of the stem (as in Fig. 1) ; the second leaf being the farthest possible from the first, as is the third from the second ; the third therefore over the first, and the fourth over the second, and so on, thus forming two vertical ranks. The angular divergence is here half the circumference, or 180° ; and the phyllotaxy may be represented by the fraction \, which desig- nates the angular divergence, while its denominator expresses the number of vertical ranks formed. Tristichous, or Three-ranked, is the next in the series, and* is 1 But when we reach a leaf which stands directly over a lower and older one, we say that one set or s/>ire is completed, and that this leaf is the first of a succeeding set or spire. From analogy of such an open spire to the closed cycle of a whorl of leaves, it is not unusual to designate the former likewise as a cycle. Yet it is better (with Eichler) to restrict that term, and the adjective cyclical, to verticillate phyllotaxy, or to whorls, to which it properly and etymologic-ally belongs. DISTRIBUTION OF LEAVES ON THE STEM. 123 less common, though not rare in monocotyledonous plants. Fig. 240 illustrates it in a Sedge, and 241 is a diagram in horizontal section, as of a bud ; both extending to six leaves or two turns of the spiral. The fraction ^ designates this arrangement. The angular divergence, or distance of the axis of the first leaf from the second, and so on, is one third of the circum- ference (or 120°) : conse- quently the fourth leaf comes over the first, the fifth over the second, the sixth over the third, and so on ; that is, the leaves fall into three vertical ranks. The spiral character here begins to be manifest, or becomes so by drawing a line on either fig- ure from the axis or midrib of the first leaf to that of the second, and so on to the sixth, forming a helix of two turns.1 Pentastichous, or Five-ranked, sometimes termed the quincuncial arrangement. This is the most common in alternate-leaved dicotyledonous plants. It is shown in Fig. 236 (on a branch of Apple-tree), and by diagrams, displaying the spiral character, in Fig. 242, 243. The angular distance from the first to the second leaf (passing the shorter way) is f of the circumference, or 144°. But the spiral line makes two turns round the stem, on which six leaves are laid down, with angular divergence of f , 1 The line is supposed to follow the nearest way, and the divergence is counted as £, this being the simplest and most convenient. If for any reason the longer way is preferred, then the angular divergence would be expressed by the fraction £. FIG. 240. Piece of a stalk, with the sheathing bases of the leaves, of a Sedge-Grass (Carex crus-corvi), showing the three-ranked arrangement. 241. Diagram of the cross- section of the same. The leaves are numbered in succession. FIG. 242. Diagram of position of six leaves in the five-ranked arrangement : a spiral line is drawn ascending the stem and passing through the successive scars which mark the position of the leaves from 1 to 6. It is made a dotted line where it passes on the opposite side of the stem, and the scars 2 and 5, which fall on that side, are made fainter. 243. A plane horizontal projection of the same; the dotted line passing from the edge of the first leaf to the second, and so on to the fifth leaf, which completes the trim ; as the sixth would come directly before, or within, the first. 124 PHYLLOTAXY, OR LEAF-ARRANGEMENT. and the sixth is the first to come over any one below ; the seventh comes over the second, the eighth over the third, &c. The leaves are thus brought into five vertical ranks ; but these five leaves are laid down on two turns of the helix (the sixth beginning the second revolu- tion) ; the angular divergence of the leaves in order is f , or 144° ; the angular distance of the vertical ranks,- 72°. This is a very advan- tageous distribution for ordinary foliage on erect or ascending branches. Its formula is f , expressing the angular divergence, the denominator also indicating the number of vertical ranks, the numer- ator indicating the number of revolutions made in add- ing one leaf to each rank. Fig. 244 illustrates this ar- rangement on a cone of American Larch, the scales of which are homologous with leaves, the numbers in sight are affixed, and those of the whole cone displayed on a plane at the side. Octostichous, or Eight-ranked, a less common arrangement, occurs in the Holly, Aconite, the radical leaves of Plantago. It has the angular divergence of 135°, or f of the cir- cumference, and the leaves in eight ranks, the ninth over the first and at the completion of the third revolution : it is therefore repre- sented by the fraction |. 240. The obvious relations of the fractions b 4i £> I? representing the primary forms of spiral phyllotaxj', are that the sum of any two numerators is the numerator of the next suc- ceeding fraction, and the same is true of the denominators ; also the numerator is the same as the denominator of the next but one pre- ceeding fraction. Following these indications, the series may be extended to -fc, ^-, Jf , §£, &c. Now these FIG. 244. A cone of the small-fruited American Larch (Larix Americana), with the scales numbered, exhibiting the five-ranked arrangement. FIG. 245. An offset of the Houseleek, exhibiting the 5-13 arrangement; the leaves in sight numbered, the 14th over the first, the 19th over the 6th, &c. FIG. 246. Cone of White Pine (Pin us Strobus) with scales numbered from bottom, and some secondary spirals marked. DISTRIBUTION OF LEAVES ON THE STEM. 125 cases actually occur, and ordinarily only these.1 The -^ and -28T are not uncommon in foliage. The rosettes of the House- leek exhibit the ^ or thirteen-ranked arrangement, as also does the cone of Pinus Strobus, the 14th leaf falling over the first. (Fig. 246.) The ^ is perhaps little less common in foliage upon veiy short internodes, as likewise are higher ranked numbers ; and in many pine-cones and similar structures gf and f £ ph3'llotaxy may be readily made out. This actual series, £, £, f , |, &c., answers to and may be expressed by the con- tinued fraction, £ , , t + i + |, &c.2 1 When other instances are detected, they are found to belong to other series, following the same law, such as the rare one of \, %, f , ^\. 2 " The ultimate values of these continued fractions extended infinitely are complements of each other, as their successive approximations are, and are in effect the same fraction, namely, the irrational or incommeasurate inter- val which is supposed to be the perfect form of the spiral arrangement. This does, in fact, possess in a higher degree than any rational fraction the property common to those which have been observed in nature; though practically, or so far as observation can go, this higher degree is a mere refinement of theory. For, as we shall find, the typical irrational inter- val differs from that of the fraction f by almost exactly -rAtf, a quantity much less than can be observed in the actual angles of leaf -arrangements." "On this peculiar arithmetical property .... depends the geometrical one, of the spiral arrangement, which it represents ; namely, that such an arrange- ment would effect the most thorough and rapid distribution of the leaves around the stem, each new or higher leaf falling over the angular space be- tween the two older ones which are nearest in direction, so as to subdivide it in the same ratio in which the first two, or any two successive ones, divide the circumference. But, according to such an arrangement, no leaf would ever fall exactly over any other ; and, as I have said, we have no evidence, and could have none, that this arrangement actually exists in nature. To realize simply and purely the property of the most thorough distribution, the most complete exposure of light and air around the stem, and the most ample elbow-room, or space for expansion in the bud, is to realize a property that exists separately only in abstraction, like a line without breadth. Neverthe- less, practically, and so far as observation can go, we find that the fractions | and TV ^r, &c., which are all indistinguishable as measured values in the plant, do actually realize this property with all needful accuracy. Thus, | = 0.375, T§J = 0.385, and ?\ = 0.381, and differ from Jc [the ultimate value to which the fractions of this series approximate, or what is supposed to l>e the type-form of them] by —0.007, +0.003, and —0.001 respectively ; or they all differ by inappreciable values from the quantity which might therefore be made to stand for all of them. But, in putting k for all the values of the series after the first three, it should be with the understanding that it is not so employed in its capacity as the grand type, or source of the distributive character which they have, — in its capacity as an irrational fraction, — but simply as being indistinguishable practically from those rational ones." - Chaucey Wright, in Mem. Amer. Acad. ix. 387-390. 126 PTIYLLOTAXY OB LEAF-ARRANGEMENT, 241 The successive grades of angular divergence of alternate leaves, as expressed in degrees, are i = 180° | = 144° T*S = 138° 27' 41.54" I = 120° | = 135° 28i — 137° 8' 34.29" and so on ; and beyond, if not in the latter cases, the differences become quite too small for determination by inspection. They all fall within the J and ^ as to amount of divergence ; and they form a series converging to a deduced typical angle of 137° 30' 28", which, being irrational to the circumference, would place no leaf exactly over any preceding one, but alternately and more and more slightly on one and the other side of the vertical, and so on, in an endless spiral. That is, according to Bravais, the ranks in the higher grades tend to become curviserial, or actually become so; while in the lower grades they are obviously rectiserial. Unless, indeed, there is some torsion of the axis, by which the vertical ranks are rendered oblique, as is often the case in cones of the Norway Spruce. But, apart from this, the difference between rectiserial in a high order and curviserial soon becomes inappreciable. Any and all of the higher grades. and practically one as low as the |, secures the utility of the theoretical angle, viz., that " by which the leaves would be dis- tributed most thoroughly and rapidly around the stem, exposed most completely to light and air, and provided with the greatest freedom for symmetrical expansion, together with a compact arrangement in the bud." Even in the simpler grades of com- monest occurrence, each leaf (according to Wright) is so placed over the space between older leaves nearest in direction to it as always to fall near the middle of the space, until the circuit is completed, when the new leaf is placed over an old one.1 242. It is to be noted that the distichous or \ variety gives the maximum divergence, viz. 180°, and that the tristichous or ^ gives the least, or 120° ; that of the pcntastichous or f is nearly the mean between the first two ; that of the |, nearly the mean between the two preceding, &c. The disadvantage of the two- ranked arrangement is that the leaves are soon superposed and so overshadow each other. This is commonly obviated by the length of the internodes, which is apt to be much greater in this than in the more complex arrangements, therefore placing them vertically farther apart ; or else, as in Elms, Beeches, and the 1 This corresponds with Hofmeister's general rule, that "new lateral members have their origin above the widest gaps between the insertions of the nearest older members." Yet the fact that the character of the leaf- arrangement is laid down at the beginning in the bud does not go far ill the way of the mechanical explanation which he invokes. DISTRIBUTION OF LEAVES ON THE STEM. 127 like, the branchlets take a horizontal position and the peti- oles a quarter twist, which gives full exposure of the upper face of all the leaves to the light. The £ and f , with dimin- ished divergence, increase the number of ranks ; the | and all beyond, with mean divergence of successive leaves, effect a more thorough distribution, but with less and less angular distance between the vertical ranks. 242". The helix or primitive spiral upon which the leaves successively originate ascends, sometimes from left to right, sometimes from right to left,1 commonly without change on the same axis, and prevailingly uniform in the same species ; but occasionally both directions occur in the same individual. The earliest leaves of a stem or branch, or the last, are often on a different order from the rest ; or (as already stated) the spiral may change into the cyclical, or vice versa. 243. The relation of the phyllotaxy of a branch to the leaf from the axil of which the branch springs is somewhat various. But in Dicotyledons, the first leaf or the first pair of the branch is mostly transverse ; that is, the first leaves of the branch stand to die right or left of the subtending leaf. In Monocotjiedons, the first branch-leaf is usually parallel to and facing the subtending leaf, as shown in Fig. 304. 244. When the internodes are considerably lengthened, the normal superposition of leaves is not rarely obscured by torsion of the axis : indeed, this may equally occur in short internodes, sometimes irregularly or in opposite directions, sometimes uni- formly in one. Thus, in Pandanus utilis, or Screw-Pine, of tristichous arrangement, the three compact vertical ranks be- come strongly spiral by a continuous torsion of the axis. The later leaves of Baptisia perfoliata, which are normally distichous, become one-ranked by an alternate twist, right and left, of the successive internodes. 245. When the internodes are short, so that the leaves approx- imate or overlap, it is difficult or impossible to trace the suc- cession of the leaves on the primitive spiral, but it is easy to see which are superposed. The particular phyllotaxy may then be determined by counting the vertical ranks, which gives the denominator of the fraction. But in compact arrangements these vertical ranks are commonly less manifest than certain * oblique ranks, which are seen to wind round the axis in oppo- site directions. (See Fig. 245, 246.) These are termed second- ary spirals, also by some parastichies. These oblique spiral 1 That is, of the observer and as seen from without. See p. 51, foot-note. 128 PHYLLOTAXY, OR LEAP-ARRANGEMENT. ranks are a necessary consequence of the regular ascending arrangement of parts with equal intervals over the circumference of the axis ; and, if the leaves are numbered consecutively, their numbers will necessarily stand in arithmetical progression on the oblique ranks, and have certain obvious relations with the pri- maiy spiral which originates them, as will be seen by projecting them on a vertical plane. 245°. Take, for example, the ? arrangement, where, as in the diagram annexed to Fig. 244, the primitive spiral, written on a plane surface, appears in the numbers, 1, 2, 3, 4, 5, 6, and so on : the vertical ranks thus formed are necessarily the numbers 1_6-11; 4-9-14; 2-7-12; 5-10-15; and 3-8-13. But two parallel oblique ranks are equally apparent, viz. 1-3-5, which, if we coil the diagram, will be continued into 7-9-11-13-15 ; and also the 2-4-6-8-10 continues into 12-14, and so on, if the axis be prolonged. Here the circumference is occupied by two secon- dary left-hand series, and we notice that the common difference in the sequence of numbers is two ; that is, the number of the parallel secondary spirals is the same as the common difference of the numbers on the leaves that compose them. Again, there are other parallel secondary spiral ranks, three in number, which ascend to the right ; viz. 1-4-7, continued into 10-13 ; 3-6-9-12, continued into 15 ; and 5—8—11—14, &c. ; where again the common difference, 3, accords with the number of such ranks. This fixed relation enables us to lay down the proper numbers on the leaves, when they are too crowded for directly following their succes- sion, and thus to ascertain the order of the primary spiral series by noticing what numbers come to be superposed in the verti- cal ranks. Thus, in the small cone of the American Larch (Fig. 244) , which usually completes only three heights of leaves, the lowest, highest, and a middle one make a vertical row which faces the observer. Marking this first scale 1 , and count- ing the parallel secondary spirals that wind to the left, we find that two occup3T the whole circumference. From 1 , we number on the scales of that spiral 3-5-7, and so on, adding the com- mon difference 2, at each step. Again, counting from the base the right-hand secondary spirals, we find three of them, and therefore proceed to number the lowest one by adding this com- mon difference, viz. 1—4—7-10 ; then, passing to the next, on which the No. 3 has already been fixed, we carry on that se- quence, 6-9, &c. ; and on the third, where No. 5 is already fixed, we continue the numbering, 8-11, &c. This gives us in the vertical rank to which No. 1 belongs the sequence 1-6-1 1 , showing that the phyllotaxy is of the five-ranked, or f order. DISTRIBUTION OF LEAVES ON THE STEM. 129 It is further noticeable that the smaller number of parallel sec- ondary spirals, 2, agrees with the numerator of the fraction in this the f arrangement ; and that this number, added to that of the parallel secondary spirals which wind in the opposite direction, viz. 3, gives the denominator of the fraction. This holds good throughout ; so that we have only to count the number of par- allel secondary spirals in the two directions, and assume the smaller number as the numerator, and the sum of this and the larger number as the denominator, of the fraction which ex- presses the angular divergence sought. For this, we must, how- ever, take the order of secondary spirals nearest the vertical rank in each direction, when there are more than two, as in all the higher forms. But, in all, it is necessary to count only the most manifest secondary spiral of each direction in order to lay down the proper number on the leaves or scales, and so deter- mine the phyllotaxy.1 In a rosette of the leaves of Houseleek (Fig. 245) and a cone of Pinus Strobus (Fig. 246), the num- bers which can be seen at one view are appended, and in the latter the conspicuous secondary spirals are indicated : one to left with a common difference of 5 ; and two to the right, of which the most depressed and prominent has the common dif- ference of 3, the other, nearest the vertical, the common differ- ence 8. The 14th leaf is superposed to the first, indicating the ^ arrangement. The same conclusion is derived from the num- ber of the higher spirals, the smaller 5 for the numerator, and this added to 8 for the denominator. The mathematical discus- sion of these relations, and of the whole subject of phyllotaxy, leads into interesting fields. But this sketch may suffice for botanical uses. 246. Relations of Whorls to Spirals. Verticillate and alternate phyllotaxy, or whorls and spirals, in all complete exemplifica- tions, are to be considered morphologically as distinct modes, not to be practical!}' homologized into one. Nevertheless, transi- tions between the two, and abrupt changes from one to the other on the same axis are not uncommon, the former especially in the foliage, the latter in the blossom. If the spiral be assumed as the fundamental order, it is not difficult to form a clear con- ception as to how such changes come to pass. A single whorl 1 In applying this method to the determination of the phyllotaxy of a cone, or any such assemblage of leaves, the student should be warned that, although the cones of Pines and Firs are all normally on the alternate plan (while those of Cypresses are on the verticillate), yet in individual cases (common in Norway Spruce) the cone is plainly made up of pairs of oppo- site scales which are spirally arranged. See note under 236. 130 PHYLLOTAXY, OB LEAF-AKRAXGEMENT. may most naturally be produced by the non-development of the internodes between any two, three, or more alternate leaves. Two proximate distichously alternate leaves would thus form a pair ; the three leaves belonging to one turn of the spiral in the tristichous (£) arrangement would compose a trimerous whorl ; the five leaves of the two turns in the pentastichous ( J) arrange- ment, a 5-merous whorl, &c. Verifications of this conception, by whorls breaking up or reverting to 'spirals, are occasionally met with, and the successive overlapping in spiral order of the members of a trimerous or pentamerous whorl is very common. The few instances among phaenogamous plants in which the leaves are opposite and all in the same plane 1 (that is, the suc- cessive pairs superposed) may be deduced from the distichous alternate mode becoming opposite without further change, by the simple suppression of alternate internodes. The frequent disjunction of the members of the pair in similar and analogous cases goes to confirm this view. But the characteristic of whorls ordinarily is that proximate whorls alternate, that pairs de- •nissate. We cannot homologize this with spiral phyllotaxy ; tor in this lies the fundamental difference between the two plans. We can explain it only by a reference to Hbfmeister's law, which generally governs leaf-origination as to position, namely, that succeeding leaves appear directly above the intervals between the nearest preceding (241, note) : this gives decussation or alternation of successive pairs or whorls.2 247. Hypothesis of the origin of both. Instead of regarding the spiral path on the stem which connects successive alternate leaves as a purely formal representation, it may be conceived to be the line along which the members in some original form were physically connected, in the manner of a leaf-like expansion 1 As in Loranthus Europaeus, &c., according to Braun. See 236, note. 8 This renders the verticillate an advantageous arrangement, perhaps no less so than the distribution which spiral phyllotaxy effects. Both must be considered to have been determined by and for their respective utilities, and to have been independent determinations. For " there is no continuity or principle of connection between spiral arrangements and whorls " (Chauncey Wright) ; since, although individual whorls are easily reducible to spirals, each succession is an absolute break of that system. As whorls of four members often (as especially in calyx, bracts, &c.) may and sometimes should be viewed as two approximate pairs, so even the spiral of five members, as in a quincuncial calyx, has been conceived to consist of two whorls, one of two, the other of three leaves, the second alternating with the first as nearly as possible. But this appears far-fetched and of loose application. It is much clearer as well as simpler to regard the alternate as the fundamental phyllotaxy, and to deduce individual whorls from spirals, if need be, rather than to imagine spirals as somehow evolved from whorls. DISTRIBUTION OF LEAVES ON THE STEM. 131 resembling a spiral stairway. Upon this supposition, the leaves would be the relics, or rather the advantageous results, of the segmentation of such a frond-like expansion, the segments separated through the development of the stem in length and firmness, and modified in the various adaptations to the conditions of higher vege- table life ; even as leaves themselves are modified into tendrils, bud- scales, petals, or other usefully specialized structures. The type on this conception would be a frond, consisting of an elongating axis with a continuous leaf-blade on one side, and this taking a spirally twisted form. But the frond of Fucaceous Algae, Hepaticae, and the like, is two-bladed. While a one-bladed frond, or with one blade suppressed, might be the original of alternate-leaved spirals, the two-bladed frond, simi- larly broken up, would give rise to the opposite or other varieties of verticillate arrangement.1 248. Fascicled Leaves need to be mentioned here, in order that they ma}' be excluded from phyllotaxy. They are simply a cluster or tuft of leaves, belonging to more than one node, and left in a crowded con- dition because the internodes do not lengthen. They may belong either to the alternate or the verticillate series. In Barberry and in the Larch (Fig. 247), the}7 are evidently alternate ; and they may be inferred to be so in Pines (Fig. 248), or even may be seen to be so in the bud-scales which form the sheath sur- rounding the base of the 2, 3, or 5 foliage-leaves. In Junipers, the leaves of the fascicles are in the verticiUate order. 1 This is the conception of the late Chauncey Wright. See his elaborate and most suggestive essay in Mem. Amer. Acad. Arts and Sciences, ser. 2, ix. 379, mainly reprinted in Philosophical Discussions (posthumous), 296-328, in which the whole subject of phyllotaxy is acutely discussed, especially in its relation to questions of origin and developed utilities. His conception FIG. 247. Piece of a branch of the Larch, with two fascicles of leaves, i. e. two very short and stout branchlets, bearing scars of former leaves or bud-scales below, and a dense cluster of leaves of the season at summit. The main axis bears scars from which the alternate leaves of the developed axis of the preceding year separated. FIG. 248. Piece of a branch of Pitch Pine, with three leaves in a fascicle or bundle in the axil of a thin scale (a) which answers to a leaf of the main axis. The bundle is surrounded at the base by a short shealh, formed of the delicate scales of the axillary bud, of which the three leaves are the developed foliage. 132 PHYLLOTAXY, OR LEAF-AltRANGEMENT. SECTION II. DISPOSITION OF LEAVES IN THE BUD. 249. Vernation and ^Estivation are terms in general use, under which the disposition of leaves in the bud is treated. The first relates to ordinary leaves in this early condition ; the second, to the parts of a flower-bud ; not, however, as respects insertion, or position on the axis, which is phyllotaxy (231). but as to the ways in which they are coiled, folded, over- lapped, &c., either perse or inter se. Prcefoliation and Prcefloration are etymologically better terms, substituted by Richard.1 250. The descriptive terms which relate to individual leaves or ** parts, whether of foli- age or blossom, mostly range themselves under the heads of plications or of enrolling, and are such as the following, the sectional diagrams of which are copied from the original figures would make the two plans equally primordial. But the freedom with which these actually interchange on the same axis greatly favors the less hypothetical view that whorls may be condensed spirals. This assumes only the well-known fact that internodes may be completely non-developed. 1 Better formed and more expressive terms: but the Linnaean ones are most in use, and, though fanciful, are not misleading. In English description, it is as convenient and equally terse to say that the parts are imbricate, val- vate, &c., "in the bud." Linnaeus, in the Philosophia Botanica, described these dispositions of leaves in the bud under the term Foliatio, — not a happy name, — but did not treat of them in the flower-bud. Later, in Termini Botanici (Amcen. Acad. vi. 1762, reprinted by Giseke in 1781), he intro- duced the words Vernatio and sEstivatio in their now current botanical sense, to designate, not the time of leafing and of flowering (spring and summer condition), but the disposition of the parts in the leaf-bud and flower-bud (at least of the petals) as respects foldings, coiling, &c., of single parts, and modes of overlapping or otherwise of contiguous parts. The terminology as regards single leaves, Linnreus fixed nearly as it now remains. That of leaves or their homologues in connection, and as respects the flower-bud, was very imperfectly developed until its importance (and much of its termi- nology) was indicated by Robert Brown, in his memoir on Proteacea?, 1800, in the Prodromus a year later, and in other publications. Ptyxis (the Greek name) is coming into use as a general term for the folding, &c., of single parts. FIG. 249-254. I.innsean diagrams of sections of leaves in the bud 249. Conduph- rate. 250. Plicate or plaited 251. Convolute. 252. Revolute. 253. Involute. 254. Circinate or Cirdnal. VERNATION, OK PR^EFOLIATION. 133 in the Philosophia Botanica of Linnseus. They were applied only to foliage, but they are equally applicable to floral parts. Leaves, and all homologous or similar organs, if not simply plane, will be either bent or folded or else more or less rolled up in the bud. The first three of the following terms relate to the former, the remaining terms to the latter. They are as to the mode of packing Plicate or Plaited (Fig. 250), when folded on the several ribs, in the manner of a closed fan, as in Maple and Currant. This occurs only in certain palmately veined or nerved leaves. Conduplicate (Fig. 249), when folded lengthwise, or doubled up flat on the midrib, as in Magnolia ; a very common mode. The upper face of the leaf is always within. Reclinate or Inflexed, when the upper part is bent on the lower, or the blade on the petiole, as in the Tulip-tree (the blade of which is also conduplicate) . Convolute (Fig. 251), when rolled up from one margin, i. e. one margin within the coil, the other without, as in Apricot and Cherry. Involute (Fig. 253), both margins rolled toward the midrib on the upper face, as the leaves of Water Lily, Violet, &c. ; also the petals of Steironema and Tremandra. Revolute (Fig. 252), similarly rolled backward from both margins, as the leaves of Azalea and Rosemary. Cirdnal or Circinate (Fig. 254), when coiled from the apex downward, as the leaves of Drosera and the fronds of all the true Ferns. Corrugate or Crumpled, as the petals of a Poppy, applies to the irregular crumpling of the otherwise plane corolla-leaves. This is a consequence of rapid growth in length and breadth in a confined space. 251. The Ptyxis (or folding, &c.) of an individual leaf, of which the foregoing modifications arc the principal, should be distinguished from the arrangement in the bud of the leaves of a circle or spiral in respect to each other. The interest of the latter centres in the flower-bud, i.e. in {estivation. To this the following exposition is devoted, although sometimes applicable to leaf-buds also.1 252. The disposition of parts in {estivation, in respect to each other, is the result partly of their relative insertion, that is 1 In the succeeding paragraphs, it becomes necessary to presuppose so much knowledge of the flower as is implied in the free use of such terms as calyx and corolla, sepals or calyx-leaves, and petals or corolla-leaves. See, if need be, Chapter VI. Sect. I. 134 PHYLLOTAXY, OR LEAF-ARRANGEMENT. their phyllotaxy, and partly of the way in which they comport when their margins meet in growth. Those leaves which are within, or of higher insertion on the axis, will almost necessarily be enclosed or overlapped : those which are members strictly of the same whorl or cycle ma}- fail to come into contact, or may meet without overlapping at the contiguous margins or apex ; yet they may be overlapped,, since they may have grown unequally or some a little earlier than their fellows. Conse- quently, no perfectly clear line can be drawn in the flower between cycles and spirals except by their mode of succession. More- over, (estivation strictly so called should be concerned only with the disposition among themselves of the several members of one whorl, or of one complete spiral. So the alternation of contiguous whorls, as of the three inner with the three outer flower-leaves of a Lily or a Tulip (the alternative (estivation of DcCandolle) , is a matter of phyllotaxy, not of jestivation. The latter is properly concerned only with the relations of each three loaves to each other.1 253. The proper aestivations may be classified into those in which the parts do not overlap, and those in which they do. Of the first, there are two kinds, the open (cest. aperta) and the valvate, both characterized and named by Brown.2 Of the second, there is one leading kind, the imbricate (adopted b}' Brown from Linnaeus), with subordinate modi- fications.8 Accordingly, the aestivation is said to be 254. Open or Indeterminate (cest. aperta), when the parts do not come into contact in the bud, so as to 1 The same applies to the two sets of sepals and of petals in Barberry, in Menispermum, and of the petals in Poppy, &c. (359). 2 Linnaeus, indeed, has, " JEsticatio valvata, si petala se expansura instar glumae graminis ponuntur," — the name, but not the thing: the glumes of grasses are not valvate in the botanical sense. So the term as to its proper use may be said to originate with R. Brown. 8 For a brief discussion of " Estivation and its Terminology," see Amer. Jour. Sci. ser. 3, x. 339, 1875. As to names, it is perhaps more correct to say of the (estivation that it is imhricatire, convolntive, valvular, &c. (); also one ligulate and neutral ray-flower and part of another (c, c): in d, d, the bracts or leaves of the involucre are seen in section. FIG. 289. Catkin of White P-irch. 290. Young spike of Plantago major. 150 ANTHOTAXY, OB INFLORESCENCE. AMENT or CATKIN. This is merely that kind of spike with scaly bracts borne by the Birch (Fig. 289), Poplar, Willow, and, as to one sort of flowers, by the Oak, Walnut, and Hickory, which are accordingly called amentaceous trees. Catkins usually fall off in one piece, after flowering or fruiting. All true catkins are unisexual. 277. Any of these forms of simple inflorescence may be com- pounded. Racemes may themselves b'e disposed in racemes, spikes in spikes (as in Triticum) , heads be aggregated in heads, umbels in umbels, corymbs may be corvm- bosely compound, &c. ; forming compound racemes, spikes, utnbels, and the like, the terminology of which is eas}T. The most usual case of truly homomorphous com- pounding is that of umbels ; the inflorescence of much the larger part of Umbelliferae being in compound umbels, as in Fig. 290". There is then the general umbel, the rays of which become peduncles to the partial umbels, and the rays of the latter are pedicels. Umbella and Umbellnlee desig- nate in Latin terminology the general and its partial umbels. Umbellets (coined by the late Dr. Darlington) may well replace the latter as the English diminutive. But umbels are sometimes racemosely arranged, as in Aralia spinosa, heads may be arranged in spikes, and so on. 278. A Panicle, of the simple and normal sort (as illustrated in Fig. 291), is produced when a raceme becomes irregularl}T compound by some (usually the lower) of its pedicels developing into peduncles carrying several flowers, or more than one, or branching again and again in the same order. But in com- pound clusters generally the secondary and tertiary ramifications are apt to differ in type as well as in particular mode, giving rise to heteromorphous or mixed inflorescence. (288.) As Linnaeus defined the term, and as it has generally been em- ployed in botanical descriptions, the panicle is a general term FIG. 290« Compound umbel of Caraway. 291. A simple panicle. THE CYMOSE TYPE. 151 for an}' loose and diversely branched cluster, with pedicellate flowers. It is therefore difficult to restrict it in practice to the indeterminate type. 279. Varieties of Determinate or Cymose Inflorescence. The plan of this. type has been sufficiently explained. (270.) Its simplest condition is that of a solitary terminal flower, peduncu- late or pedicellate (as in Fig. 282), or sessile. The production of more flowers necessitates new axes from beneath, from the axils of adjacent leaves or bracts. These, being later, render the evolution centrifugal. The simplest flower-cluster (unless we call the solitary flower of Fig. 282 a one-flowered cluster) is that of Fig. 283, where a secondary floral axis or peduncle has developed from the axil of each leaf of the uppermost pair, or where with alternate leaves there is a single uppermost leaf, and then only one such peduncle, and thus is produced a three- (or two-) flowered cymose cluster. The flower of the primary axis is marked by its bractless peduncle (therefore a pedicel) ; the lateral and secondary peduncles are known (commonty or nor- mally) by their bracts or bract; the portion below the bracts is proper peduncle ; that above, of single internode, pedicel. Bracts, like other leaves, have potential buds in their axils ; these in an inflorescence give the third order of ramification, each branch tipped with its flower ; and so on. 280. The Cyme is the general name of this kind of flower-cluster in its various forms. One of these very simple cymes, by itself or as a part of a larger- cyme, may be called a Cymule. The regular usually accompanies opposite or other grades of verticillate leaves, but is not rare in the alternate arrangement. It is readiest understood in an opposite-leaved plant with regular opposite ramification, as in an Arenaria, Fig. 292. 63- its con- stitution, a cyme proceeds from simple to compound. It mat- FIG. 292. Dicliotomoiis or Iiiparous Eicbler) of Arenaria Michauxii. cyme (cyme bijKire of Bravuis. Dichasiuiu of 152 ANTHOTAXY, OR INFLORESCENCE. ters little whether its development is progressive, the flowers of the ultimate ramifications expanding after the earlier have matured fruit, and with subtending bracts conspicuous or foliaccous ; or whether, as in Elder and Hydrangea (Fig. 293, and in Fig. 273), the bracts are minute and caducous or abortive, and the ramification complete with all the flower-buds well formed before the oldest expand, so that the whole is in blossom almost at the same time. But a cyme may be properly said to be compound when the primary axis in it is a peduncle instead of a pedicel, 293 and supports a cluster (c3'me or c3Tinule) instead of a solitary central flower at the main divisions.1 One form of the regular cjTne, on account of its compactness, is named the GLOMERULE. This is merely a cymose inflorescence, of any sort, which is condensed into the form of a head, or approach- ing it. Of this kind is the so-called head of Cornus florida, and of the herbaceous C. Canadensis (Fig. 294), which shows the 1 The dichotomous or two-branched cyme is the commonest, but is some- times marked by suppression of internodes ; as, for example, where the branches are apparently in fours, in an umbelliform way ; but these are two sets of two, with the internode between the pairs extremely short ; or where, as in Elder, the brandies or rays are five, in this case consisting of the same two pairs and a central one, which is a many-flowered continuation of the primary axis. Or 5-rayed cymes, &c., may be founded upon alternate leaves with shortened internodes, the rays or peduncles axillary to them thus brought into an apparent whorl. Bravais distinguished cymes as multiparous, with three or more lateral axes ; biparotts, with two ; and uniparous, with only one (cyme multipare, bipare, unipare) To these Eichler gives the substantive names, severally, of Pleiochasium, Dichasium, and Monochasium. Only the latter needs illustra tion ; the others being as it were compounds of this. FIG. 293. Compound cyme of Hydrangea; with some neutral and enlarged mar- ginal flowers. THE CYMOSE TYPE. 153 composition best, on close examination. A condensed but less capitate cyme, or cluster of CN'mes, was called by Roeper and DeCandolle a FASCICLE ; and this terminology has been much adopted. It is properly enough said to be a fascicle, which, as used by Linnaeus and others, means a bundle, or close collection of parts, whether leaves, pedun- cles, or flowers ; but a fascicle is w .„,„,,.,_, -JBU^-X/X not necessarily a cyme (274) , nor is s*^/m\\tl*§$\ ^ there need of a special substantive name for a compact cyme, which may either be simply so called or it may pass into the glomerule. 281. Botryoidal forms of Cjrmose Type, or False Racemes, &c. The regular C3Tme seldom continues with all its ramifications. In Fig. 292, after the second forking, one of the two lateral peduncles mostly fails to appear, and in some parts one of the bracts also ; and ultimately the lateral peduncle present is bractless, like the central, therefore equally incapable of further ramification, being reduced to a pedicel of a single inter- node. This suppression some- times begins at the first fork- ing or at the very base ; and, when followed throughout, it reduces a biparous or dichoto- mous cyme to one half, and, converts this half (when the axis straightens) into the sem- blance of a raceme if the flowers are pedicelled, or of a spike when they are sessile. 2% 2.17 Fig. 296 is a diagram of such an inflorescence as that of Fig. 292. with one lateral branch uniformly suppressed at each division, the wanting members indicated by short dotted lines. Cases exem- plifying this occur in portions of the inflorescence of some of our FIG. 294. Plant of Cornus Canadensis: flowering stem bearing a cluster of leaves above, then continued into a peduncle, and terminated' by a glomerule of very smal; flowers; this subtended by a colored and corolla-like involucre of four bracts. 295 One of the flowers taken from the glomerule, enlarged. FIG. 296. Uniparous cyme or sympodial false raceme, with opposite leaves or bracts. FIG. 297. Form of the same, with alternate leaves or bracts. 154 ANTHOTAXY, OB, INFLORESCENCE. smaller H}Tpericums, and notably in H. Sarothra, in which the leaves are all reduced to bracts. It is not always easy to show why this is not a true raceme. But the other bract of the pair, upon that supposition, is unaccountably empty : the successive angular divergence of each joint of the axis of inflorescence in the }Tounger part, which commonly runs into a coil, finds explanation in the view that each portion is the lateral branch from the axil of the subtend- ing leaf: and occasionally the other axil produces a similar one, thus revealing the cymose character. When the bract from the axil of which the missing branch should come disappears also, as sometimes it does, and uniformly on the same side, a state of things like that of the upper part of Fig. 297 occurs. The same figure may serve for the arrangement corresponding to that of Fig. 296, only with alternate leaves. But then, close as the imi- tation of a raceme here is, the position of each flower in respect to the bract supplies a criterion. While in a true raceme the flower stands in the axil of its bract, here it stands on the oppo- site side of the axis, or at least is quite away from the axil. 282. Sympodial forms. The explanation is that the axis of inflorescence in such cases, continuous as it appears to be, is not a simple one, is not a monopode, but a sympode (110, 116, notes), i.e. consists of a series of seemingly superposed inter- nodes which belong to successive generations of axes : each axis bears a pair of leaves (Fig. 296) or a single leaf (Fig. 297), is continued beyond into a peduncle (or pedicel in these instances), and is terminated by a flower. From the axil promptly springs a new axis or branch, vigorous enough soon to throw the adjacent pedicel and flower to one side : this bears its leaf or pair of leaves, and is terminated like its prede- cessor with a flower ; and so on indefinitely. The fact that the alternate leaves or bracts arc thrown more or less strictly to one side and the flowers to the other, in Fig. 297, shows that these leaves do not belong to one and the same axis ; for alternate leaves are never one-ranked or disposed preponderatingly along one side of an axis, as in this diagram, and as is seen in the inflorescence of a Ilouseleek, &c. 283. A further difficulty in the morphology of clusters of this class comes from the early abortion or complete suppression of bracts. This is not unknown in botryose inflorescence, occurring in the racemes of almost all Crueiferae : it is very common in the cymose of all varieties, and especially in the uniparous ones in question, which characterize or abound in Borraginacese, Hydro- phyllaceae, and other natural orders. In some genera or species, the bracts are present, or at least the lower ones ; in others, THE CYMOSE TYPE. 155 absent ; in some, either occasionally present or wanting in the same species or individual. It is only by analogy, therefore, and b}- a comparison of allied plants, that the nature of some of these flower-clusters can be made out. With the botanists of a preceding generation, these one-sided clusters were all described as racemes or spikes. Botanists still find it convenient to con- tinue the use of these names for them in botanical descriptions, adding, however, as occasion requires, the qualification that they are false racemes or spikes, or cymose racemes, and the like ; or else, by reversing the phrase, with stricter correctness they call them racemiform or spiciform cymes, &c. 284. Commonly these false racemes or spikes (or bot^oidal cj'mes, if we so name them) are circinate or inrolled from the apex when young, in the manner of a crosier, straightening as they come into blossom or fruiting. Likening them to a scorpion when coiled, the earlier botanists designated this as scorpioid. As the coil is a helix, it has also been named helicoid.1 The flowers are then thrown, more or less strictl}*, to the outer side of the coiled rhachis, where there is room for them ; and so these false racemes or spikes are secund or unilateral. The particular anthotaxy and phyllotaxy of the various sympodial and botryoi- dal forms of cymose inflorescence become rather difficult ; and the sorts which have been elaborately classified into species (and have no little morphological interest) are connected by such transitions, and are based on such nice or sometimes theo- retical particulars, that the terminology based on them is seldom conveniently applicable to descriptive botany, at least as to sub- stantive names. 285. One of the la test and simplest classifications of cymes is that of Eichler in his Bluthendiagramme.2 1 Scorpioid and Helicoid have been carefully distinguished by later morphologists, on account of some difference in the mode of evolution and arrangement of the flowers along one side of the rhachis, by which they become two-ranked in scorpioid, one-ranked in helicoid. But practically the two kinds of clusters are not always readily discriminated ; and in gen- eral terminology a single name, with subordinate qualifying terms, is suffi- cient. Scorpioid is the older and commoner one, therefore the most proper to be used in the generic sense. 2 CYMOSE TYPE (classified without reference to bracts, which are so often wanting) ; divided into a. Lateral axes three or more: PLEIOCHASIUM, the multiparous cyme ot Bravais. 3. Lateral axes two : DICHASIUM, the liparous cyme of Bravais. y. Lateral axis one • MONOCHASIUM, the uniparous cyme of Bravais. The latter, or the corresponding divisions of the preceding sorts, may be divided as follows : 166 ANTHOTAXY, Oil INFLORESCENCE. 286. Sundry complications and obscurities are occasionally encountered in anthotaxy or phyllotaxy, which cannot here be * Lateral axes transverse to the relatively main axis. 1. Lateral axes in successive generations always falling on the same side of the relatively main axis: SCHRAUBEL [screw-like] or BOSTRYX [ringlet or curl], the uniparous helicoid cyme of Bravais. 2. Lateral axes falling alternately on opposite sides of the relatively main axis: WICKEL or CINCINNUS [a curl], the uniparous scorjjioid cyme of Bravais. * * Lateral axes medial [in the same plane] relative to the main axis. 3. Lateral axes in successive generations always on the back side of the axis from which it springs: FACHEL, RHIPIDIUM [fan]. 4. Lateral axes in successive generations always on the upper side of the axis from which it springs : SICIIEL, DRBPANIUM [sickle]. The subjoined simple diagrams from Eichler (Fig. 298-303) illustrate these forms. The ramification is given without the bracts, which theoreti- cally or actually subtend the axes of each generation. The student may add them, and so more readily apprehend the characters. Eichler recognizes the forms with median (antero-posterior) position of axes in Monocotyledons only. It is natural to distichous phyl- lotaxy, and it accords with the general rule that, in mono- cotyledonous plants, the first leaf of the branch, or the com- monly solitary bractlet of the )S peduncle, stands over against and facing the bract or leaf from the axil of which said branch or peduncle springs, i. e. is posterior and next the parent axis, as shown in the diagram, Fig. 304, 305. 301 300 304 FIG. 298. Diagram of the Cincinnus. 299. Diagram of the Bostryx. The flower- axes numbered in succession. FIG. 300. Diagram of the lihipidium. 302. Ground plan, indicating the order of evolution of the flowers. FIG. 301. Diagram of the Drepanium. 303. Ground plan, the flowers evolved in succession, from left to right. FIG. . with them. When the perianth-leaves are of only one set, it is not at all by color or texture that this perianth can be assigned to calyx or to corolla. Normal transitions from a stamen to a pistil could not, in the nature of the case, be expected. 308. Teratological Transitions and Changes. Teratology is the study of monstrosities. These in the vegetable kingdom FIG. 317. Cactus-flower (Mamillaria caespitosa), with bractlets, sepals, and petals passing into each other. FIG. 318. Series exhibiting transition from sepals to stamens in Nymphsea odorata ITS METAMOKPHY. 171 often elucidate the nature of organs.1 The commonest of these changes belong to what was termed by Goethe retrograde meta- morphosis ; that is, to reversion from a higher to a lower form, as of an organ proper to the summit or centre of the floral axis into one which belongs lower down.2 The most familiar of all such cases is that of the so-called double flower •, better named in Latin flos plenus. In this, the essential organs, or a part of them, are changed into colored flower-leaves or petals. Most flowers are subject to this change under long cultivation (witness " double " roses, camellias, and buttercups), at least those with numerous stamens. It occasionally occurs in a state of nature. The stamens diminish as the supernumerary petals increase in number ; and the various bodies that ma}' be often observed, inter- mediate between perfect ;- stamens (if any remain) and the outer row of petals, — from imperfect petals, with a small lamina tapering into a slender stalk, to those which bear a small distorted lamina on one side and a half-formed anther on the other, — plainly reveal the nature of the transformation that has taken place. Carried a step farther, the pistils likewise disappear, to be replaced by a rosette of petals, as in fully double 1 The leading treatises are Moquin-Tandon's Te'ratologie Vegetale, Paris, 1841, and Masters, Vegetable Teratology, London, published for the Kay Society, 1869. An earlier publication deserves particular mention, viz. the, thesis De Antholysi Prodromus, by Dr. George Engelmann, Frankfort on the Main, 1832. 2 To these abnormal changes, the term metamorphosis is obviously more applicable ; for here what evidently should be stamens, pistils, &c., on the testimony of position and the whole economy of the blossom, actually ap- pear in the form of some other organ : yet even here the change is only in the nisus formativus ; the organ was not first formed as a stamen, and then transformed into a petal or leaf. FIG. 319. A flower of the common White Clover reverting to a leafy branch ; after Turpin. Calyx with tube little changed, but lobes bearing leaflets. Pistil stalked; the ovary open clown the inner edge, and the margins of the pistil-leaf bearing leaves instead of ovules. 172 THE FLOWER. buttercups.1 In these the green hue of the centre of the rosette indicates a tendency to retrograde a step farther into sepals, or into a cluster of green leaves. This takes place in certain blos- soms of the Strawberry, the Rose, &c. Such production of " green roses," and the like, has been appropriately called chlorosis, or by Masters chloranthy, from the change to green. 309. A monstrosity of the blossom of White Clover, long ago figured by Turpin (Fig. 319), is such a case of foliaceous rever- sion, in which even the ovules are implicated. The imperfect leaves which take the place of the latter may be compared with the leaf}- tufts which form along the margins of a leaf of Bryophyllum, by which the plant is often propagated. (Fig. 322.) 310. The reversion of a simple pistil di- rectly to a leaf is seen in the Double-flowering Cherry of cultivation (Fig. 320, 321), usually passing moreover, by prolification of the re- ceptacle, into a leafy branch. 311. The reversion of pistils to stamens is rarer, but has been observed in a good number of instances, in Chives, in the Horseradish, in Gentians and Hyacinths, and in some Willows. In the latter, the opposite transformation, of stamens to carpels, is very common, and curious grades between the two are met with almost every spring. So also in the common Houseleek, and in perennial Larkspurs. Certain apple-trees are known, both in the United States and Europe, in which, while the petals are changed into the appearance of minute green sepals, the outer stamens are converted into carpels, these supernumerary and in the fruit superposed to the five normal carpels.2 In Poppies, many of the innermost stamens are occasionally transformed into as many small and stalked simple pistils, surrounding the base of the large compound one. 1 It must not be concluded that the supernumerary petals in all such cases are reverted stamens, or stamens and pistils. Some are instances of abnormal pleiotaxy, i. e. of the production of one or more additional ranks of petals (better deserving the name of double flower], with or without reversion of essential organs to flower-leaves. 2 These trees are popularly supposed to bear fruit without blossoming; the reverted green petals being so inconspicuous that the flower is un- noticed. FIG. 320, 321. Green leaves from the centre of a blossom of Double-flowering Cherry, one still showing, by its partial involution and its style-like apex, that it is a reverted carpel, the other a small but well-formed leaf. FIG. 322. Leaf or leaflet of Bryophyllum, developing plantlets along the margins. ITS METAMORPHY. 173 312. Another line of teratological evidence is furnished by prolification. The parts of the flower are, by the doctrine, homologous with leaves, and no leaf ever terminates an axis. Normally, in fact, the axis is never prolonged beyond the flower, but abnormally it ma}' be. It may resume vegeta- tive growth as a termi- nal growing bud, either from between the pistils after the whole flower is formed, or at an earlier period, usurping the central part of the flower. Thus, when a 324 rose is borne on a pe- duncle rising from the centre of a rose, which is not very unusual, or a leafy stem from the top of a pear (Fig. 323), the flower was probably complete before the monstrous growth set in. In Fig. 324, the reversion to foliaceous growth took effect after the stamens but before the pistils were formed. In rose-buds out of roses, the terminal proliferous shoot takes at once the form of a peduncle ; in the shoot from the pear, that of a leafy stem. 313. Again, axillary buds are normally formed in the axil of leaves. No such branching is known in a normal flower. But in rare monstrosities a bud (mostly a flower-bud) makes its appearance in the axil of a petal or of a stamen ; and it may be clearly inferred that the organ (not itself axillary) from the axil of which a bud develops is a leaf or its homologue. Fig. 325 exhibits a clear case of the kind, a flower in the axil of each petal of Celastrus scandens. Flowers, or pedunculate clusters of flowers, from the axil of petals of garden Pinks are sometimes seen. A long-pedunculate flower from the axil of a FIG. 323. A monstrous pear, prolonged into a leafy branch ; from Bonnet. FIG. 324. Retrograde metamorphosis of a flower of the Fraxinella of tho gardens, from Lindley's Theory of Horticulture ; an internode elongated just above the stamens, an^l hearing a whorl of green leaves. FIG. 325. A flower of False Bittersweet (Celastma scandens) producing other flowers in the axils of the petals ; from Turpin. 174 THE FLOWER. stamen of a species of Water-Lily (N}-mphaea Lotus) is figured and described by Dr. Masters.1 314. In the application of morphological ideas to the elucida- tion of the flower, nothing should be assumed in regard to it which has not its proper counterpart and exemplar in the leaves and axis of vegetation. SECTION II. FLORAL SYMMETRY. 315. The parts of a flower are symmetrically arranged around its axis.2 Even when this symmetry is incomplete or imperfect, it is still almost always discernible ; and the particular numerical plan of the blossom may be observed or ascertained in some of the organs. 316. Adopting the doctrine that the parts of the flower are homologous with leaves, the symmetry is a consequence of the phyllotaxy. It is symmetry around an axis, not the bilateral s}'mmetry which prevails in the animal kingdom. For parts of a flower disposed in a continuous spiral (which mostly occurs when they are numerous) , the arrangement is that of some order of this kind of phyllotaxy, which distributes the parts equably into superposed ranks. (237.) The much commoner case of 1 The fullest enumeration and discussion of the very various kinds of abnormal structures and deviations in plants is to be found in the Teratology of Dr. Masters, above referred to. Many technical terms are here brought into use, which need not be here mentioned, except the following, which relate directly to floral metamorphosis. Phyllody (called Pkyllomorphy by Morren, Frondescence by Engelmann) is the condition wherein true leaves are substituted for some other organs; i. e., where other organs are metamorphosed into green leaves. There is phyllody of pistils, ovules, filaments, anther, petals, sepals, &c. Sepalody, where other organs assume the appearance of green sepals. Petalody, where they assume the appearance of petals, as normally in Pinckneya and Calycophyllum, in which one calyx-lobe enlarges and becomes petal-like, and abnormally in Primroses where all the calyx-lobes imitate lobes of the corolla (this has been termed Calycanthemy) ; also of the stamens of common " double flowers." Staminody, where other organs develop into stamens. Cases of this as affecting pistils are referred to above : rarely sepals and petals are so affected. Pistillody, where other organs develop into pistils, which most rarely happens except with the stamens, as above mentioned. 2 It is stated that Correa de Serra (who published botanical and other papers in London, Paris, and Philadelphia during the first twenty years of the century, but who knew far more than he published) was the first botanist to insist on the symmetry of the flower. It was first made prominent by De Candolle, in the Theorie Elementaire, and elaborated in detail by A. St. Hilaire in his Morphologic Ve'ge'tale. FLORAL SYMMETRY. 175 equal number of parts in a cycle, and the cycles alternating with each other, is simply that of verticillate phyllotaxy. (234.) In either case, the members of the successive circles (or of closed spirals as the case may be) will be equal in number ; that is, the flower will be isomerous. 317. A Symmetrical Flower is one in which the members of all the cycles (whorls or seeming whorls) are of the same number.1 In nature, the symmetry is of all degrees : it is most commonly complete and perfect as to the floral envelopes when it is not so as respects the essential organs. The general rule is that the successive cj'cles alternate, as is the nature of true whorls. But the superposition of successive parts is not incompatible with symmetry of the blossom, although it is a departure from the ordinary condition, assumed by botanists as the type. An isomerous flower (meaning one with an equal number of mem- bers of all organs) is the same as symmetrical, if the reference be to the number in the circles, rather than to the total number of organs of each kind. 318. A Regular Flower is one which is symmetrical in respect to the form of the members of each circle, whatever be their number ; i. 3Ted by Jussieu, depend upon the degree of adnation, or the absence of it, viz. : — Hypogynous (literally be- neath pistil) , applied to parts which are inserted (i. e. are borne) on the receptacle of the flower, as in Fig. 336. This is the absence of adnation, or the condition which corresponds with the unmodified type. Periyynous (around the pistil) implies an adnation which carries up the inser- tion of parts (which always means apparent origin or place of attachment) to some distance above or away from the recep- FIG. 336°. Vertical section of a flower of the Common Flax, showing the normal or hypogynous insertion of parts upon the torus or receptacle. FIG. 337. Vertical section of a flower of the Cherry,' to show the perigynous insertion, or adnation to the calyx, of the petals and stamens. FIG. 338. Similar section of the flower of the Purslane, showing an adnation of all the parts with the lower half of the ovary. UNION OF DISSIMILAR PARTS. 183 tacle, so commonly placing this insertion around instead of beneath the pistil ; whence the name. The perig}*ny may be, as the figures show, merely the adnation of petals and stamens to calyx, the calyx remaining hypogynous, as in Fig. 337 ; or else the adna- tion of the calyx, involving the other organs, to the lower part of the ovary, as in Fig. 338, or up to the summit of the ovary, while the petals and stamens are adnate still further to the calyx, as in Fig. 339. The latter passes into what is called Epigynous (on the pistil) , where the adnation is complete to the very top of the ovary, and none beyond it, as in Fig. 340, 341. Yet here the parts so termed are not really on the ovary, except where an cpigynous disk (394) actually surmounts it. 333. Adnation brings some other terms into use in botanical descriptions, especially those of superior and inferior. In this connection, these words (in Latin taking the form of superus and inferus) denote the position in respect to each other of ovary and floral envelopes, — not the morphological, but the apparent posi- tion or place of origin. Thus, in Fig. 336 and in 337, the calyx is inferior, or in other words the ovary superior. Here real and apparent origin agree, this being the normal condition, which is otherwise expressed by saying that the parts are/ree, i. e. free from all adnation of one to the other. But, in Fig. 339-341, the FIG. 339. Similar section of a flower of Hawthorn, showing complete adnation to the summit of the ovary and of the other parts beyond. FIG. 340. Vertical section of a Cranberry-flower, and 341, of flower of Aralia nudicaulis. with so-called epigynous insertion of calyx, corolla, and stamens; the calyx of the latter completely consolidated with the surface of the ovary, or its limb obsolete, 184 THE FLOWER. ovary is said to be inferior and the calyx superior, the calyx and other parts, in consequence of the adnation of its lower part, seeming to rise from the summit of the ovary. 334. Adnation of floral envelopes to pistil rarely extends beyond the ovary ; yet, in species of Iris having a tube to the perianth, this tube is commonly adnate for most of its length to the style. But when the calyx has its tube or portion with united sepals prolonged, the petals -and the stamens are usually adnate more or less to it, i. e. are inserted on the calyx. And, when the petals are united and prolonged into a tube, the sta- mens, being within the corolla, are commonly adnate to or inserted upon this. 335. No one doubts that the view is a true one which repre- sents the perianth-tube as adnate to the style in Iris, petals and stamens as adnate to calyx in the Cherry (Fig. 337), stamens as adnate to base of corolla in Fig. 334, and a long way farther in Phlox, &c. That the calyx is similarly adnate to the ovar}r is nearby demonstrable in certain cases. 336. But, as the lower portion of a pear is undoubtedly recep- tacle, or rather the enlarged extremity of the flower-stalk, as in a rose at least a portion of the hip is receptacle, as the tube of the flower in a Cereus or other Cactacea has all the external char- acters and development of a branch, so it is most probable that in many cases the supposed cahyx-tube adnate to an inferior ovary is partly or wholly a hollowed receptacle (in the manner of a Fig- fruit) ; that is, a cup-shaped or goblet-shaped develop- ment of the base of the floral axis. This would bring the case under § 7. (326, 495.) § 4. IRREGULARITY OF SIMILAR PARTS. 337. Irregularity, or inequality in form or in union of mem- bers of a circle, is extremely common, either with or without numerical symmetr}*. One or two examples ma}" suffice. 338. Irregular flowers with S3rmmetrical perfection, except in the gyncecium, are well seen in the Pea Family, to which belongs the kind of corolla called Papilionaceous, from some imagined resemblance to a butterfly. (Fig. 342-344.) This flower is 5-merous throughout, has the full complement of stamens (10, or two sets) , but the g}Tno3cium reduced to a single simple pistil. The striking irregularity is in the corolla, the petals of which bear distinguishing names : the posterior and larger one, exter- nal in the bud, is the VEXILLUM or STAND AKD (Fig. 344, a) ; INEQUALITY OF SIMILAR PARTS. 185 the two lateral next and under the standard, ALJE or WINGS (b) ; the two anterior, covered by the wings and partly cohering to S !•*. form a prow-shaped body (c) , the CARINA or KEEL. The calyx is slightly irregular by unequal union, the two upper sepals united higher than the other three. The sta- mens are much more coalescent,but with an irregularity, nine com- bined by the lower part of their filaments, and one (the posterior) separate. (Fig. 345.) 339. The plan and floral symmetry in the Locust-blossom and its relatives are little obscured by the irregu- larities and the coalescence, hardly more so than in the plainer flower of its relative, m Baptisia (Fig. 347, 348), in which the petals are somewhat alike, and the ten stamens are distinct or unconnected. Only the calyx is more irregular, by the union of the two posterior sepals almost to the tip. (Fig. 348.) FIG. 342. Diagram of flower of the Locust, Kobinia Pseudacacia : a. axis of inflores- cence; 6. bract; first circle of 5, calyx; five remaining pieces, corolla; next anthers, 10 in number; in the centre a single simple pistil. 343 Front view of Locust-flower showing only the corolla. 344. This corolla displayed. FIG. 345. Androecium of the Locust, nine stamens coalescent, one distinct. 346 Same of a Lupine, all ten filaments coalescent below into a closed tube. FIG. 317. Calyx and corolla of Baptisia australis. 348. Same with petals fallen, showing ten distinct stamens and tip of the style. 186 THE FLOWER. 340. But in a Lupine-blossom, of equally near relationship, a casual observer might fail to recognize the very same type, although disguised only by cohesions. For while the two pos- terior sepals are united to the tip on one side of the blossom, the three others are similarly united into one body on the anterior side, giving the appearance of two sepals instead of five : in' the corolla, the two keel- petals are more strictly united into a slender scythe-shaped or sickle-shaped body; so that the petals might with the unwary pass for four : in the androecium, the coalescence includes all ten stamens (Fig. 346), which is an approach to regularity. 341. The 5-merous symmetry of the Violet- blossom is complete until the gynoecium is reached (but with only one circle of stamens) ; the main irregu- larity of the perianth is in the anterior petal, with its nectariferious sac at base (Fig. 349-351) ; the two stamens near- est this send into the sac curious appendages, which the other three do not possess ; the gynoecium is composed of three car- pels coalescent into one compound ovary in a manner hereafter explained. In Antirrhinum and Linaria (Fig. 480, 481), there is a similar <[//?(?r>,^\ f \irregularity accompanying coalescence of the \ I Vv^ ) I ) Pe^8' ^ne anterior one being extended at base V\\V\ di\l I into a nectariferous sac or hollow spur.1 The flower of a Lobelia (Fig. 488) has the same numerical plan and symmetry as that of Viola (except that the gj'noecium is dimerous) ; but thb members are adnate below and coalescent above, and the corolla is irregular through unequal coalescence of the five petals, and the absence of coalescence down one side. 1 PELORIA is a name given by Linnaeus to an occasional monostrosity of these flowers (imitated in sundry others), in which the base of every petal, or answering part of the corolla, is prolonged downward into a sac or spur. The sac is, morphologically considered, a departure from normal regularity : in the monster, symmetrical regularity is restored by the development of four more sacs. FIG. 349. Flower of Viola sagittata. 350. Its sepals and petals displayed. 351. Diagram of a Violet-blossom, from Eichler, with bract or subtending leaf (below), a pair of bractlets (lateral), and axis to which the subtending leaf belongs (above or posterior). DISAPPEARANCE OF PARTS. 187 § 5. DISAPPEARANCE OR OBLITERATION OF PARTS. 342. Abortion or Suppression are somewhat synonymous terms to denote the obliteration or rather non-appearance of organs which belong to the plan of the blossom. Abortion is applied particularly and more properly to partial obliteration, as where a stamen is reduced to a naked filament, or to a mere rudiment or vestige, answering to a stamen and occupying the place of one, but incapable of performing its office ; suppression, to abso- lute non-appearance. Such vestiges or abortive organs justify the use of these terms, the more so as all gradations are some- times met with between the perfect organ and the functionless rudiment which occupies its place. Such obliterations, whether partial or complete, may affect either a whole circle of organs or merely some of its members. The former interferes with the completeness of a flower, and may obscure the normal order of its parts. The latter directly interferes with the symmetry of the blossom, and is commonly associated with irregularh\y. 343. Of parts of a Circle. Among papilionaceous flowers (338), different species of Erythrina have all the petals but one (the vexillum, Fig. 344, a) much reduced in size, in some concealed in the calyx, and in every way to be ranked as abortive organs. In Amorpha, of the same family, these four petals are gone, leaving no trace, reducing the corolla to a single petal. (Fig. 352, 353.) This one is evidently the vexillum, both by position and shape ; and the 5-merous type, also the particular type of the family, are still discernible in the five notches of the calyx, the ten stamens, &c. In a related genus, Parryella, even this last petal is wanting, and the androecium is straight, all irregularity thus disappearing through suppression. 344. Delphinium or Larkspur and Aconite or Monkshood furnish good examples of flowers in which irregularity is accom- panied by more or less abortion. The calyx of the Larkspur (Fig. 354-356) is irregular by reason of the dissimilarity of the five sepals, one of which, the uppermost and largest, is pro- longed posteriorly into a long and hollow spur. Within these, and alternate with them as far as they go, are the petals, onry FIG. 352. Stamens and pistil of Amorpha fruticosa. 353. An entire flower of the same. 188 THE FLOWER. four in number, and these of two shapes, the two upper ones having long spurs which are received into the spur of the upper sepal ; the two lateral ones having a small but broad blade raised on a stalk-like claw ; and the place which the fifth and lower petal should occupy (marked in the ground-plan, Fig. 356, FIG. 354. Flower of a Larkspur. 355. The five sepals (outer circle) and the four petals (inner circle) displayed. 356. Ground-plan of the calyx and corolla. FIG. 357. Flower of an Aconite or Monkshood. 358. The five sepals and the two small and curiously shaped petals displayed ; also the stamens and pistils in the centre. 359. Ground-plan of the calyx and corolla; the dotted lines, as in Fig. 356, representing the suppressed parts. DISAPPEARANCE OF PARTS. 189 by a short clotted line) is vacant, this petal being suppressed, thereby rendering the blossom unsymmetrical. In Aconite (Fig. 357-359), the plan of the blossom is the same, but the uppermost and largest of the five dissimilar sepals forms a helmet-shaped or hood-like body ; three of the petals are wanting altogether (their places are shown b}7 the dotted lines in the ground- plan, Fig. 359) ; and the two upper ones, which ex- tend under the hood, are so reduced in size and so anomalous in shape that they would not be recog- nized as petals. One of these, enlarged, is exhibited in Fig. 360. Petals and other parts of this and of va- rious extraordinary forms were termed by Linnaeus NECTARIES, a somewhat misleading name, as they are no more devoted to the secretion of nectar than ordinary petals or other parts are. In these flowers, moreover, the stamens are much increased in number. m 345. Analogous abortion of some of the stamens, along with a particular irregularity of the perianth, especially of the corolla, characterizes a series of natural orders with coalescent petals.1 These flowers are all on the 5-merous plan (except that the gyncecium is 2-merous), but with corolla, and not rarely the calyx, irregular through unequal union in what is called the bila- bitite or two-lipped manner. The greater union is always median, or anterior and posterior, and two of the coalescent members form one lip, three the other. The two posterior petals form the upper lip, the anterior and two lateral form the lower lip of the corolla ; in the calyx, when that is bilabiate, this is of course reversed. In some, as in Sage and Snapdragon, the bilabiation of the corolla is striking (Fig. 479-481), and readily comparable to the two jaws of an animal ; in others, the parts are almost regu- lar. The suppression referred to is, in most of these cases, that of the posterior of the five stamens, as in Fig. 361, where it is complete. In Pentstemon (Fig. 362), a sterile filament regu- larly occupies the place of the missing stamen. The position sufficiently indicates its nature. This is also revealed by the rare occurrence of an imperfect or of a perfect anther on this 1 These natural orders in which this occurs, or tends to occur, are the Scrophulariacese (Snapdragon, Pentstemon, Mimulus, &c.), Orobanchaceas (Beech-drops), Lentibulacea? (Bladderwort), Gesneraceae (Gloxinia), Big- noniaceae (Trumpet Creeper, Catalpa), Pedaliaceae (Martynia), Acanthacese, Labiatae (Salvia, Stachys), &c. FIG. 360. A petal (nectary) of an Aconite, much enlarged. 190 THE FLOWER. filament, — a monstrosity, indeed, but the monstrosity is here a return to normal sj-mmetry. The two stamens nearest the suppressed or abortive one generally shai'e in the tendency to abortion, as is shown by their lesser length or smaller anthers : in the flower of Catalpa, these two also are either im- perfect or reduced to mere vestiges (as in Fig. 363) : in very many other plants of these families, even these vestiges are not seen, and so the five stamens are by abortion or complete suppression reduced to two. 346. Suppression in the gynoecium to a number less than the numerical plan of the flower (as shown in the perianth) is of more common occur- rence than the typical number, and the reduction is comparatively con- stant throughout the genus or order. A papilionaceous or other leguminous flower with more than one or with all five pistils is exceedingly rare, and except in one pentacarpellary genus is a monstrosity. Suppression of the interior is more common than of exterior organs. Want of room in the bud may partially explain this. 347. Suppression of whole Circles. Such suppression or rather non-production in the actual blossom of whole series of organs which belong to the type, and indeed are sometimes present in that blossom's nearest relatives, is very common. It gives occasion to several descriptive terms, which may be here defined together. First, and in general, flowers are Incomplete, in which any one or more of the four kinds of organs is wanting, whatever these may be ; Apetalous, when the corolla or inner perianth is wanting ; Monochlamydeous, where the perianth is simple instead of FIG. 361. Corolla of Gerardia purpurea laid open, with the four stamens, the place which the fifth should occupy indicated by a cross. FIG. 362. Corolla of Penlstemon grandiflorns laid open, with its four stamens, and a sterile filament in the place of the fifth stamen. FIG. 363. Corolla of Catalpa laid open, with two perfect stamens and the vestiges of three abortive ones. DISAPPEARANCE OF PAKTS. 191 double, in which case the wanting set is generally (but not quite always) the inner, or the corolla ; Dichlamydeous, when both circles of the perianth (calyx and corolla) are present ; Achlamydeous, when both are wanting, as in Fig. 365. (These three terms are seldom employed.) Unisexual (also Diclinous or Separated) , when the suppression is either of the stamens or the pistils. In contradistinction, a flower which possesses both is Bisexual or Hermaphrodite. Staminate, or Male, when the stamens are present and the pistils absent ; Pistillate, or Female, when the pistils are present and the stamens absent ; Monoecious (of one household), when stamens and pistils oc- cupy different flowers on the same plant ; Diceciov* (of two households), when they occupy different flowers on different plants ; Polygamous, when the same species bears both unisexual and bisexual or hermaphrodite flowers. This may occur in various ways, from the greater or less abortion of either sex, either on the same or on separate individual plants ; — as Monceciously or Diceciously Polygamous, according to the tendency to become either monoecious or dioecious. Recently Darwin has well distinguished the case of Gyno-dtcecious, where the flowers on separate individuals are some hermaphrodite and some female, but none male only ; and Andro-dioscious, of hermaphrodite flowers and male, but no separate female. The latter is a less common case. Neutral, as applied to a flower, denotes that both stamens and pistils are wanting, — a case neither rare nor inexplicable on grounds of utility. (356, 504.) Sterile and Fertile are more loosely used terms. A sterile flower may mean one which fails to produce seed, as a sterile stamen denotes one which produces no good pollen, and & sterile pistil one which is incapable of seeding. But commonly a sterile flower denotes a staminate one ; a fertile flower, one which is pistillate, if not also hermaphrodite. 348. Suppressed Perianth. Almost universally, when the peri- anth is reduced to a single circle, it is the inner, or corolla, which is not produced. Or, rather, when there is only one circle or sort of perianth-leaves, it is called calyx, whatever be the appearance, texture, or color, "unless it can somehow be shown that an outer circle is suppressed. For since the calyx is frequently delicate and petal-like (in botanical language, petaloid or colored, as in 192 THE FLOWER. Clematis and Anemone, Fig. 364), and the corolla is sometimes greenish or leaf-like, the only real difference between the two is that the calyx represents the outer and the corolla the inner series. Even this distinction becomes arbi- trary when the perianth consists of three or four circles, or of a less definite number of spirally arranged members. 349. Yet the only perianth obvi- ously present ma}* be corolla, as when the calyx has its tube wholly adnate to the ovary and its border or lobes obsolete or wanting.1 Aralia nudicaulis (Fig. 341) is an instance, likewise many Umbelliferse, some species of Fedia or Valerianella, the fertile flowers of Nyssa, and those Compositse which have no pappus. For PAPPUS, the name originally given to thistle-down and the like, answers to the border or lobes of a calyx attenuated and depauperated down to mere fibres, bristles, or hairs. The name is ex- tended to other and less obliterated forms. (644, Fig. 631-633.) When the obliteration is complete, as in Mayweed (Fig. 630) , in some species of Coreopsis, &c.,the corolla seems to be simply continuous with the apex of the ovary. A comparison with related forms reveals the real state of things.2 350. So also in Hippuris, in which (along with extreme numerical reduction of the other floral circles) the calyx as well as corolla seems to be wanting ; but the insertion of the stamen on the ovary (epigj'nous) suggests an adnate calyx, and near inspection detects its border. 351. Both cal}-x and corolla are really want- ing in the otherwise complete and perfect (symmetrical and 1 In the flowers of the two common species of Prickly Ash (Zanthoxy- lum) of the Atlantic United States, one has a double, the other a single perianth (as shown in Gray, Gen. Illustr. ii. 148, t. 156) : the position of the stamens gives a presumption that the missing circle of the latter is the calyx ; yet it may be otherwise explained. In Santalaceas there are some grounds for suspecting that the simple perianth, although opposite the stamens, is corolla ; and the foliaceous sepal-lobes of the female flowers of Buckleya would confirm this, if these are true sepals rather than adnate bracts. 2 In the pappus of Composite, every gradation is seen between undoubted calyx, recognizable as such by structure as well as position, and diaphanous scales, bristles, and mere hairs, wholly " trichomes " as to structure, although in the place of " phyllomes " and representing them. FIG. 364. Flower of Anemone Pennsylvania; apetalous, the calyx petaloi \. FIG. 365. Aelilamy.leons flower of Lizard's-tail (Saururus cernuus), magnlried. DISAPPEARANCE OF PARTS. 193 trimerous) flowers of Saururus, Fig. 365. But achlamydeous blossoms are usually still further reduced to a single sex. 352. Suppression of one circle of stamens is of very common occurrence. It is seen in different species of Flax ; which have mosth- 5-merous perfectly symmetrical and complete flowers with one set of stamens abortive. In some species (as in Fig. 367), vestiges of the missing circle of stamens are conspicuous in the form of abortive filaments, interposed between the perfect stamens ; in others, these rudiments are inconspicuous or even altogether wanting. 353. Suppressed Androecium or Gynoecinm. This occurs with- out or along with suppression in the perianth. In cases of the former, vestiges of the aborted organs often remain to sig- nify the exact nature of the loss. Sepa- ration of the sexes (monoecious, diceci- ous, &c.) is the re- sult of such suppres- sion. In Menisper- mum (Fig. 368, 369), this is accompanied by an actual doub- ling of both caryx and corolla. The dioecious flowers of Smilax are similarly complete, except by the abortion of one sex, but the calyx and corolla are single. FIG. 366. Flower of a Linum or Flax. 367. Androecium and gynoecium ; the former of 5 perfect stamens, alternating with 5 rudiments of a second set. FIG. 368, 369. Dioecious flowers of Moonseed, Menispermum Canadense : 368, Stami- nate or male blossom ; 369, Pistillate or female, but with six abortive stamens, before as many petals. FIG. 370. A catkin of staminate flowers of a Willow, Salix alba. 371. A single staminate flower detached and enlarged (the bract turned from the eye). 372. A pistil- late catkin of the same species. 373. A detached pistillate flower, magnified. 194 THE FLOWER. 354. Combined with suppression of Perianth. This, which is found in most amentaceous or catkin-bearing trees, in some with partial suppression of perianth, is well illustrated in Willows, the flowers of which are all achlamydeous and dioecious. (347.) The little scale (gland or nec- tary) at the inside of each blossom might be sup- posed to represent a perianth, reduced to a single piece ; but an extended comparison of forms refers it rather to the receptacle. Willow-blossoms (Fig. 370-372) are crowded in catkins, each one in the axil of a bract : the staminate flowers consist of a few stamens merely, in this species of only two, and the pistillate of a pistil merel}-. In Salix purpurea, the male flower seems to be a single stamen (Fig. 374) ; but it consists of two stamens, united into one body. Here extreme suppression is ac- companied with co- alescence of the existing members. 355. Still more simplified flowers, but more difficult A I \ / to comprehend, are l\ I , \ / those of Euphorbia, T^ ml/0 or Spurge. These are in fact monoeci- ous ; and the female flower is a pistil, the male is a stamen. The pistillate flower (of three carpels, their ovaries united into one three-lobed compound ovary) surmounts a slender peduncle which terminates each branch of the flowering plant. (Fig. 375.) From around the base of this pe- duncle rise other smaller and shorter peduncles, each from the axil of a slender bract, and surmounted by a single stamen, which represents a male flower. (Fig. 376, 377.) This umbel-like FIG. 374. A separate staminate flower of Salix purpurea, with the stamens coa- lescent (monadelphous and syngenesious), so as to appear like a single one. FIG. 375. Flowering branch of Euphorbia corollata. 376. Calyx-like involucre divided lengthwise, showingthe staminate flowers around a pistillate flower (a). 377. A more magnified stamiiiate flower detached with its bract, a; its peduncle or pedicel b. surmounted by the solitary stamen, c. 378. Pistil in fruit, cut across, showing the three one-seeded carpels of which it is composed. SUPERPOSITION OF SUCCESSIVE PARTS. 195 flower-cluster is surrounded and at first enclosed by an involucre in the form of a cup, which imitates a calyx ; and the lobes of this cup (the free tips of the calyx-leaves) in the present species are bright white, so that they exactly imitate petals. Here, then, is a whole cluster of extremely simplified flowers, taking on the guise of and practical!}7 behaving like a single flower, the invo- lucre serving as calyx and corolla ; the one-stamened male flowers collectively imitating the androecium of a polyandrous blossom, and surrounding a female flower which might pass for the pistil of it. A series of related forms, from various parts of the world, gives proof that this interpretation is the true one. 356. Suppression of both Andro3cium and Grynoecium. This occurs in what are termed Neutral Flowers (347) , such as are conspicuous at the margin of the cymes of Hydrangea (Fig. 293) and of Viburnum lantanoides and Opulus, also at the margin of the head of flowers of Sunflower, Coreopsis (Fig. 287, 288), and the like. In these and most other instances, the perianth of which only the flower consists is much larger and more showy than in the accompanying perfect flowers : in fact, their whole utility to the plant, so far as known, is in this conspicuousness. No plant normally bears neutral flowers only ; but in cultivation all sometimes become so by monstroshy, as in the form of Vibur num Opulus called Snowball or Guelder Rose, also in "full double" roses, pinks, &c. Occasionally flowers become sterile and neutral by mere depauperation and abortion of perianth as well as of essential organs, as in certain Grasses ; but such are mostly vestiges of flowers rather than neutral blossoms. § 6. INTERRUPTION OF NORMAL ALTERNATION. 357. Anteposition or Superposition is the opposition of succes- sive (or apparently successive) whorls which normally alternate. This result is brought about in different waj-s, some of which are obvious, while of some the explanation is hypothetical. 358. In the first place, there are cases of seeming anteposi- tion, which are explained away on inspection. In a tulip, lily, and the like, there is a perianth of six leaves and a stamen be- fore each. The simple explanation is that the flower is not 6-merous, but 3-merous : there is a calyx of three sepals, colored and mostly shaped like the three petals, which alternate with these and are clearly anterior in the bud ; next, three stamens alternate with the petals or inner circle of the perianth ^ then the three stamens of the inner circle, alternating with the preced- ing, necessarily are opposite the three petals, as the first three are 196 THE FLOWER. opposite the sepals. These organs altogether are in four whorls of three, not in two of six members ; and the pistil at the centre, of three combined members, is the fifth and final whorl. 359. The Barberry family exhibits a similar seeming ante- position, which is more striking on account of a multiplication of the members of the perianth. The calyx is of six sepals in two circles, the corolla of six petals in two circles, the stamens equally six ; and so each petal has a stamen before and a sepal behind it. But, when properly viewed as a trimerous flower with double circles of sepals and petals as well as of stamens, all is S3rmmetrical and normal. Menispermum in the related Moonseed famity is in the same case, but the flower is trimerous, as seen in Fig. 369 : in the male blossom this is obscured in the androe- cium (Fig. 368) by a multiplication of the stamens.1 The same thing occurs in the perianth and bracts of certain Clusiaceae, in which the members counted as in fours are superposed, and in some of which the double dimerous arrangement with apparent anteposition extends through the corolla ; while, in other closety related flowers, the corolla changes to simply tetramerous and to alternation with the preceding four sepals. This passes, in the same family and in the allied Ternstrosmiaceae, into 360. Superposition by Spirals, as where five petals are ante- posed to five sepals, b}* an evident continuation of pentastichous ph}*llotaxy ; and the stamen-clusters of Gordonia Lasianthus are probably in this way brought before the petals.2 The flower of Camellia is continuously on the spiral plan up to the gynoe- cium ; but upon one which, from the bracts onward, rises from the £ to the § and f order or higher, throwing the petals of the rosette in a full-double flower into numerous more or less conspicuous vertical ranks. 361. Anteposition in the Androecium. It is in the androecium that real anteposition is most common, and also most difficult to account for upon any one principle. Doubtless it comes to pass in more than one way. This condition is chiefly noticed when the stamens are definite in number, and mainly in isostemonous and diplosternonous flowers. (324.) 362. With Isostemony. Vitis (Fig. 379-381), also Rhamnus (Fig. 415, 416), and the whole Grape and Buckthorn families of 1 In Columbine (Aquilegia), multiplication of the stamens in successively alternating 5-merous whorls similarly brings the andrcecium into ten ranks ; so, when these stamens in double flowers are transformed into hollow-spurred petals, these are set one into another hi ten vertical ranks. 2 Gen. Illustr. ii. 1. 140. But the petals alternate with the sepals in the ordinary manner of the flower, though their strong quincuncial imbrication suggests the spiral arrangement. SUPERPOSITION OF SUCCESSIVE PARTS. 197 which they are the types, afford familiar cases of a single circle of stamens placed before the petals. In Vitis, there are green nectariferous lobes or processes from the receptacle, alternate with and inside the stamens : there is no good reason to suppose that they answer to a second row of stamens. All isostemonous Por- tulacacea? have the stamens before the petals ; and, when the stamens are fewer than the petals, those which exist occupy this position. Among the orders with gamopetalous corolla, such anteposition is universal in Plumbaginaceas, Primu- lacese, the related Myrsinaccae, and in most Sapotaceae, in the latter usually with some complications. 363. The earliest and the most obvious explanation of the anomaly is that of the suppression of an outer circle of stamens, and to this view recent morphologists are returning.1 Observa- tion supplies no vestige of proof of it in Rhamnacese and Vitaceas ; but, in the group of related orders to which the Primulaceae be- long, evidence is not wanting. For Samolus and Steironema both exhibit a series of rudimentary organs exactly in the place of the wanting circle of stamens, which may well be sterile fila- ments. In the allied order Sapotaceae, while Chrysophyllum has in these respects just the structure of Primulaceae, and Sideroxylon that of Samolus, Isouandra Gutta (the Gutta-percha plant) has a circle of well-formed stamens in place of the sterile rudiments of the preceding ; that is, alternate with the petals, 1 Eichler, Bliithendiagramme, passim, and in preface to Part II. xviii., relating chiefly to obdiplostemony. The principal opposing view is that of St. Hilaire, Duchartre, &c., maintaining that corolla and stamens here repre- sent one circle of organs doubled by median chorisis ; upon which see note under a following paragraph. According to that hypothesis, there is n6 andrcecial circle in such blossoms, or only vestiges of one, but the petals have supplied the deficiency by a supernumerary production of their own ! The more plausible hypothesis of Braun, that of a suppressed interior circle of extra petals, would restore the alternation, and make the extant sta- mens the fourth floral circle, as does the adopted explanation. Braun's hypothesis, if it insists that an extra row of petals is wanting, supposes the suppression of that which very rarely exists ; but, if of stamens, then the supposed suppression is of that which is so generally present, or with indi- cations of presence, as properly to be accounted a part of the floral type. FIG. 379. Flower of the Grape Vine, casting its petals before expansion. 380. The same, without the petals: both show the glands of the disk distinctly, within the stamens. 381. Diagram of the flower. 198 THE FLO WEB. completing the s}-mmetry of the blossom and the normal alterna- tion of its members. This explanation of the anteposition of a single circle of stamens is the more readily received, because it well accords with the idea here adopted, that the androecium of a typical flower should consist of two circles of stamens. (324. ) The only serious objections to this explanation rise out of the difficulty of applying it to analogous anteposition when both circles are present. 364. For Diplostemony^ the condition of two circles of sta- mens, each of the same number as the petals, is also itself very commonly attended by anteposition. In normal or Direct Diplostemony, — that which answers to the floral type com- pletely,— the antisepalous stamens (324, note) are the outer and the antipetalous the inner series, and the carpels when isomerous alternate with the latter and oppose the sepals ; the alternation of whorls is therefore complete, as in the diagram, Fig. 382. Such stamens, however, ma}r actually occupy a single line or coalesce into a tube, without derange- ment of the type. But it as commonly occurs that the antipetalous sta- mens are more or loss exterior in insertion, and then the carPels' whcn isomerous, are alternate with the inner and anti- sepalous stamens, and therefore opposite the petals, as in the diagram, Fig. 383. This arrangement takes the name of Obdiplostemony. In it the normal alternation of successive whorls is interrupted, so as to produce anteposition, 365. With Obdiplostemony. This condition prevails, more or less evident!}-, in Ericaceae, Geraniacese, Zygophyllaceae, Rutaceae, Saxifragacese, Crassulaceae, Onagraceae, &c. (but in some of these with exceptions of direct diplostemony) ; also, accompanied by a peculiar multiplication of members (380), in Malvaceae, Sterculiaceae, and Tiliaceae. The explanation is difficult. The h}-potheses may be reduced to three, neither of which is quite satis factor}*. There is, first, the hypothesis of St. Hilaire, ap- plied to this as to the preceding case (to Rhamnus, Vitis, &c.), that these exterior antipetalous stamens belong to the corolline whorl ; in other words, that the petal and the stamen before it FIG. 382. Diagram of pattern flower with direct diplostemony 383. Diagram of similar flower with Obdiplostemony. Both from Bidder's Bliithendiagramme. SUPERPOSITION OF SUCCESSIVE PARTS. 199 (whether adnate to or free from it) answer to one leaf which has developed into two organs by a deduplication (372) taking place transversely. This makes the inner and antisepalous stamens the third floral circle or the only truly androscial one, and sym- metrically alternate with the petals on the one hand and the carpels on the other. The second h}rpothesis conceives that there is a whorl suppressed between these antipetalous stamens and the corolla : this, ideally restored, gives symmetric succession and alternation to all the succeeding whorls. The five glands in a Geranium- flower, alternate with and next succeeding the petals (Fig. 384) , were plausibly supposed to represent this missing whorl, which according to Braun should be an inner corolla ; according to others rather a primar}- circle of stamens. The third is the recent hypothesis of Celakowskj-, which Eichler adopts : this regards the antipetalous stamens as really the inner or second circle, and conceives that in the course of development it has become external by displacement. The difficulties of this hypothesis are, first to account for this dis- placement, and then for the anteposition of the carpels to the assumed inner stamens in the great majority of these cases.1 1 In the first part of the Bluthendiagramme, Eichler inclined to the first hypothesis, that of St. Hilaire (now very much abandoned on account of the feeble evidence that there is any such thing as transverse or median chorisis); in the second, he discards this in favor of Celakowsky's view (published in Regensburg Flora, 1875). As to members which are morpho- logically interior becoming exterior by outward displacement, Eichler cites the staminodia or sterile stamen-clusters of Parnassia (Fig. 400, 401), and the corresponding antipetalous stamens of Limnanthes, as clearly interior in the early flower-bud, but exterior at a later period ; states that the vascu- lar bundles which enter these stamens generally are either inner as respects those of the episepalous stamens or in line with them ; that in some cases (as in many Caryophyllaceae) the real insertion of the stamens is that of direct diplostemony, while the upper part of their filaments and the anthers are external to the episepalous series ; that in most families with obdiplos- temony examples of direct diplostemony occur, and still more cases with both stamineal circles inserted in the same line ; and that, as a rule, the episep- alous stamens are either later or not earlier formed than the epipetalous. As to the position of the carpels before antipetalous stamens and petals, Celakowsky suggests that this may result from the outward recession of those stamens affording more room there, while in the normal case the greater space is over the episepalous stamens. And, indeed, exceptions to the prevalent position are not uncommon both in direct diplostemony FIG. 384. Diagram (cross- section) of the flower of Geranium maculutum, exhibiting the relative position of parts, and the symmetrical alternation of circles, ». e. sepals, petals, greenish bodies called glands, antipetalous stamens, antisepalous stamens, carpels. 200 THE FLOWER. 366. The case of stamens in a cluster before the petals is a complication of either of the foregoing with a peculiar kind of multiplication, termed deduplication or chorisis. (372.) § 7. INCREASED NUMBER OF PARTS. 367. Augmentation in the number of floral members is one of the commonest modifications of the type. It occurs in two ways : 1st, by an increased number of circles or turns of spirals in the flower, which is Regular Multiplication ; 2d, by the pro- duction of two or three or of many organs in the normal place of one, Chorisis or Deduplication. The first does not alter the normal symmetry of the blossom, although it may render it dif- ficult or impossible to trace or demonstrate it. The second apparently disturbs, or at least disguises, floral symmetry. Either may be definite, or of a constant and comparatively small number ; or indefinite, when too numerous for ready counting, or inconstant, as the higher numbers are apt to be. 368. Regular Multiplication, or Augmentation of floral circles or spirals, may affect any or all the four organs, but most com- monly the androecium. When the perianth is much increased in the number of its members, the distinction between calyx and corolla, or even between bracts and corolla, is apt to disappear, as in most Cactaceous flowers (Fig. 317), Nelumbium, Calycan- thus, &c. In these and similar cases, the members of the perianth are prone to take a spiral instead of cyclic arrangement ; and this and in obdiplostemony. Along with the lack of clear analogy to support St. Hilaire's hypothesis of transverse deduplication, the similar orientation of the vascular bundles in the petal and the stamen before it must, as Celakowsky insists, be good evidence that these represent independent leaves, and not superposed portions of one. The main objection to the second hypothesis (that of a suppressed circle outside of the antipetalous stamens) is that this missing circle, whether of petals or stamens, is not actually met with in any nearly re- lated forms (for in Monsonia the fifteen stamens are otherwise explained) ; also that there are transitions, as above mentioned, between obdiplostemony and direct diplostemony. To Braun's theory that the glands behind the antisepalous stamens in true Geraniaceae answer to suppressed phylla, Eichler objects that these are present behind all ten stamens in Oxalidese ; also that all are wanting when the office of nectar-secretion, which they sub- serve, is undertaken by some other part of the flower, as by the calyx-spur in Pelargonium and Tropaeolum. The first objection is forcible : the second mixes morphological considerations with functional, and is inconclusive. Abortive organs, preserved for their utility as nectaries, might totally dis- appear when rendered useless by a different provision for the same function. INCREASED NUMBER OF PARTS. 201 is even more true of greatly multiplied stamens and pistils, as in Magnolia and Liriodendron, most Anonaceae, Ranunculus, Anemone, and the like. But in Aquilegia, where the number five is fixed in the perianth, the cyclic arrangement with alternation of whorls prevails throughout 369. The definite augmentation of calyx and corolla by the production of one additional whorl of each, and the seeming anteposition which comes of it when the andrcecium remains simply diplostemonous (in the manner of the Berberidacese, Menispermaceae, &c., 359) has already been explained. 370. Similar increase to two whorls affecting the corolla only characterizes Anonaceae, Magnoliaceae, Papaveraceae, and Fuma- riaceae. In all but the last order, this is accompanied b}r indefi- nitely multiplied stamens, and mostly by an increased number of carpels. In Fumariaceae, which has dimerous flowers, there is a diminution by the suppression in most cases of half the normal androecium, and also an augmentation of the other half by chorisis. (372.) 371. Parapetalous Multiplication. Under this head may be described an anomalous arrangement of augmented stamens which prevails in the order Rosaceae, but is not peculiar to it.1 The simplest case, but a rare one, is seen in the 10-stamened vari- ety of some Hawthorns, as occasionally in Crataegus coccinea and Crus-galli. The ten are in one circle and in pairs, the pairs alternate with the petals. Some would say the pairs are before the petals ; but the space between two stamens before each petal is mostly rather wider than in the pair taken the other way. The next case in order, as in 15-stamened Hawthorns, and constantly in Nuttallia, adds to the above a simple interior circle of five stamens, one directly before the middle of each petal. Next, as in most Pomeae and many Potentilleae, there are twenty stamens, thus placed, but with an additional circle of five alter- nating with the preceding one. Next there are 25 in three circles, the second circle as well as the first having ten stamens ; and finally there are from 30 to 50, all probably in circles of ten each. There is little doubt that the circles develop in centri- petal order ; the inner successively the later.2 1 It was first clearly described by Dr. A. Dickson, in Trans. Bot. Soc. Edinb. viii. 468, and Seemann's Jour. Bot. iv. 473 (1866). He introduced the term, parapetalous, which is characteristic of it in its elementary form (254, note) : it is particularly illustrated by Eichler, in BlUthendiagramme, ii. 495-510. The former interprets it by chorisis, both median and collateral : the latter presents the facts and possible views, but declines to adopt either of them. 2 Accordingly, the whole is probably to be explained by some modifica- 202 THE FLOWER. 372. Chorisis or Deduplication. Both these terms, and the ideas which they denote, originated with Dunal, but were first expounded b}* Moquin-Tandon.1 The first word is Greek for a separating or separation. The second is a translation of Dunal's French word dedoublement (literally undoubling) , the ambiguity of which, and of the original presentation of the case, long retarded the right apprehension of the subject. Diremption has been suggested (by St. Hilaire) as a proper term. The mean- ing simply is, the division of that which is morphologically one organ into two or more (a division which is of course congenital) , so that two or more organs occup}* the position of one. As thus used, chorisis is restricted, or nearly so, to the homologues of leaves in the flower, and mainly to stamens and carpels ; the division or splitting up of a petal or a sepal, when it occurs, being expressed in the phrases which are applied to leaves. Yet a compound leaf, especially one of the palmate type, is a good type of chorisis, the several blades of a compound leaf answering to the single blade of a simple leaf. It has been ob- jected against the terms chorisis and deduplication that they assume the division of that which has never been united ; but so equally does the established terminology of foliage. A di- vided leaf has never been entire. 373. Chorisis is complete when the parts concerned are dis- tinct or separate to the very insertion, as in the stamen-clusters of Hypericum. The foliar form of this would be represented by tion of the augmentation of circles. Dickson's hypothesis, that the two, three, or five stamens which are more or less in face of each petal are all deduplications of that petal, would come to be noticed under the next head, but it may be dismissed at once. Yet that the pairs in the outer circle represent each an antisepalous stamen, divided by chorisis (sometimes incompletely) and much separated, is not improbable. The other tenable explanation (which may be harmonized with the last) is that the outer circle of stamens here rightly consists of ten members, respectively alternat- ing with the sepals and petals taken as a whole. This makes them para- petalous, and at the same time brings them under Hofmeister's general law that new organs originate over intervals of those preceding, in this case over the ten perianth-intervals directly. It also accords with Hartog's elucidation of the accessory parts in the flower of Sapotaceae (in Trimen's Jour. Bot. 1878). The inner circles are there sometimes 5-merous after the primitive type, sometimes 10-merous in regular alternation to the preceding circles. 1 Moquin-Tandon, Essai des Dedoublemens, &c., Montpellier, 1826; Con- siderations sur les Irre'gularite's de la Corolle, &c., in Ann. Sci. Nat. xxvii. 237, 1832; Teratologie Ve'ge'tale, 337. Dunal, Essai sur les Vaccinie'es, 1819, cited by Moquin (some pages printed, but never published) ; Conside- rations sur la Nature et les Rapports de quelques-uns des Organes de la Fleur, 1829. The next botanist to develop it was St. Hilaire, Morphologie Vegetale, 1841. CHORISIS OR DEDUPLICATION. 203 such sessile palmately compound leaves as those of some species of Aspalathus. It is incomplete when division does not extend to the base; as in Fig. 387, 393. Compare, as a proximate homologue of this, a petal of Mignonette, Fig. 385. But proper chorisis requires that the supernumerary organs should be developed like unto the original organ which is thus multiplied, or should complete their sj-mmetry, whatever it be. 374. St. Hilaire distinguished two kinds of deduplication ; viz., collateral when the members stand side by side, and parallel when an organ becomes double or multiple antero-posteriorly. The latter, sometimes called vertical, and sometimes transverse, is better named median chorisis. The collateral is the origi- nal and tj-pical chorisis. Most botanists incline to restrict the name to this, and to give some other explanation and name to the median form of augmentation. But some cases, such as those of Tilia and Sparmannia, are clearly of the same nature as the collateral, and may be a disguised form of it ; there are others which may be explained in accordance with it ; and there are such transitions between some of these and coronal out- growths that the term chorisis is most conveniently made, to comprise augmentation or doubling in either plane. Distinct anteposition, however, may be explained in other ways. (357.) 375. Typical or Collateral Chorisis, in which the members, together answering to one leaf, normal!}' stand side by side, occurs in many families of plants, and J^^^"5^ (DC in a variety of forms. A few are here //^^^^^ \. T presented. /^^A\ 8 376. Elodes Virginica (a common marsh plant of the Hypericum family) , like most of its near relatives, has its calyx and corolla on the plan of five, its stamens and carpels on the plan of three, as is shown in the diagram, Fig. 386. This makes a break in the symmetry between the corolla and the stamens ; but all within is in regular alternation when the three stamens of each cluster are counted as one as their union at base into a phalanx (Fig. 387) may suggest. These phalanges alternate with the three carpels, and therefore stand where single stamens belong. The three conspicuous green projections, which in a general way FIG. 385. A petal of Mignonette (Reseda odorata), with many parted blade, enlarged. FIG. 386. Diagram of flower of Elodes Virginica, with three phalanges of stamens forming the inner circle, and three glands answering to the outer circle. 387. A de- tached phalanx of three stamens. 204 THE FLOWER. are called glands, alternate with the phalanges, and so are taken to represent the outer circle of stamens. The morphologist accordingly sees in the glands the homologues or representatives of the outer series of stamens, reduced to three by abortion, and in the three stamen-clusters only the three alternating stamens of the inner series, trebled by chorisis, and this chorisis incom- plete, because it has not quite divided the filament into three. In Hypericum, the glands are completely suppressed, each pha- lanx is almost or quite divided into a cluster, either of about three stamens each, as in H. Sarothra, or of a few more (in H. mutilum and H. Canadense), or of an indefinite number, as in the common St. Johnsworts. Then in some other species (as in our H. pyramidatum) the carpels and the stamen-clusters rise to five, realizing complete pentamerous sj-mmetry, except that the almost numberless stamens all belong to the one inner circle. Morphologically, they are comparable to the leaflets of five (or in most species three) decompound and ses- sile or almost sessile leaves. The indefinitely numerous stamens of llicinus are similarly increased from five by compound ramification. 377. Fumariacese, the Fumitory famity, may furnish the next illus- tration. The flower is on the plan of two (dimerous) throughout. Taking Dicentra to show it, there is first a pair of small and scale-shaped sepals, not unlike the pair of bractlets on the FIG. 388. Dicentra Cucullaria (Dutchman's Breeches), a scape in flower and a leaf, severed from the singular bulb (formed of the enlarged bases of petioles). 389. Detached flower, of natural size, showing also the pair of bractlets on the pedicel. 390. Same with parts displayed, and 3!>1, inner petals placed above. 392. Diagram of flower of Dicentra or Adluinia, from a section across the summit. 393. One of the phalanges of stuiuuns of Adlumia; upper part only. CHOR1SIS OR DEDUPLICATION. 205 pedicel below (Fig. 389, 390) , and normally alternate with them : alternate with these is a pair of large petals, deeply saccate or spurred below ; alternate with these, a pair of smaller petals with spoon-shaped tips which cohere at the apex (the corolla therefore of two circles as in the related Poppy family) ; alternate with these, two phalanges or united stamen-clusters, of three stamens each ; alternate with these is nothing, for the second set of stamens is wanting ; alternate with this vacancy is a pair of carpels wholly combined into a compound 2-merous pistil. The statement itself explains the morphology. The three sta- mens of each phalanx stand in the place of a stamen, and are the divisions of one. In Dicentra the members of the phalanx are almost separate ; in Adlumia (Fig. 393) and Corydalis the ' undivided filament reaches almost up to the anthers. The middle anther of the phalanx is normal, or two-celled ; the lateral anthers are one-celled, as if halved.1 1 Eichler adopts this interpretation (proposed hi Gray, Gen. Illustr. i. 118), and applies it to the crucial instance of Hypecoum. In the flower of this Old World genus, there are four apparently simple and complete stamens, one before each petal : the simplest interpretation would be that which the facts appear to present, viz. that both dimerous circles of stamens are complete and normal. But Eichler — hi view of the early development and the double vascular bundles of the stamens before the inner petals, and some occasional slight disjunction of their anther-cells — considers that the ulterior stamen-circle is wanting here, no less than in the other genera of the order ; that what here takes its place before each inner petal is a stamen composed of the adjacent lateral member of the phalanx, congeni- tally severed from the group to which it belongs and soldered into one fila- ment, bearing the two one-celled anthers so brought together as to imitate a normal two-celled anther. The organogeny of the blossom is thought to favor this hypothesis ; and it certainly favors the view here adopted of the composition of the three-membered phalanx of the family generally. If this interpretation of Hypecoum seems far-fetched, it is no more so than its exact counterpart, through which DeCandolle, Lindley, an^l others explain the case of the rest of the family. Starting with that genus as the simple type, they conceive that the stamen opposed to each inner p* u'l is each severed into two, and that these half -stamens attached to the sides of the two intact stamens, thus producing the phalanges by coalescence. A good empirical conception of the formation, from a single leaf, of three stamens in Fumariaceae, or two in Cruciferae, is afforded by the petals of Hypecoum, as illustrated by Eichler. The outer petals are slightly three- lobed from the apex; the inner are deeply so and narrower. The mem- bers of the next circle in the family generally are just such three-lobe/1 bodies, the tip of each lobe transformed into an anther. There is an ap- parent congruity in the production by the symmetrical middle lobe of a symmetrical two-celled anther, and of a one-celled anther by each unsym- metrical lateral lobe or stipule-like portion. A fuller development of these sides of the leaf, and non-development of the middle portion (somewhat after the analogy of Lathyrus Aphaca, Fig. 219), with anther-formation, would convert the leaf into a pair of stamens. 206 THE FLOWER. 378. The obvious relationship of Cruciferae to Fumariaceae, their agreement in the rare peculiarity of having the two carpels :a5 side by side instead of fore and aft (median), and the characteristic anomaly which the andrrecium pre- sents (z. e. the teindynamy), would give reason to expect that its prob- lems might be solved by chorisis. Indeed, the doctrine was applied to this, long before its application to the other order. Beginning at the centre (Fig. 395, &c.), the pistil is of two carpels, right and left ; alter- nate with these is a pair of stamens on the side next the axis, matched by another pair on the opposite side of the pistil, the four longer and interior stamens ; alternate with these, and lower in insertion< 397 a single stamen on each side ; next> tbur petals, of somewhat various overlapping in aestivation, which essentially alternate with the two single stamens and the cwo pairs ; lastly, four sepals, alternating with the four petals as a whole, the anterior and posterior overlapping the lateral ones in the bud. Now the median (». e. the anterior and posterior) pairs of stamens occasionally have their contiguous filaments conjoined, as in Fig. 397. If this were at all constant, the inference would undoubtedly be that the case is one of chorisis, and that the flower as to its essential organs is dimerous. This is apparently the best explanation to be given. It assumes that the chorisis is normally complete in the androacium of Cruciferae, instead of incomplete, as in Fumariaceae.1 And this view is confirmed by the fact that the median stamens are simple and 1 The hypothesis here adopted, as to the andrcecium, is that of Steinheil (1839), and of Eichler (in Flora, 1865, 1872, and Bluthend. ii. 200), replacing that of Kuntli, 1833, &c., employed in former editions. The rejected view makes the flower 4-merous up to the pistil, and the stamens all of one circle, alternating with the four petals, the median stamens (as in our view) doubled by chorisis. Krause and Wretschko (cited as above by Eichler) would have the floral circles 2-merous and 4-merous by turns ; the calyx of two 2-merous circles (which it plainly is); the corolla of one 4-merous circle FIG. 394. A cruciferous flower. 395. Diagram of such a flower, with position of axis marked above it. 396. Tetradynaruous stamens and the pistil. 397. A common monstrosity of the same, two of the four inner stamens combined into a common 2-antheriferous body. CHOBJSIS OR DEDUPLICATION. 207 single in Senebiera and many species of Lepidium, in which the lateral or short stamens are at the same time abortive. 379. It is quite possible that chorisis may be extended to the corolla of the cruciferous flower, and reduce the whole to a symmetrical 2-merous plan, and to congruity in the perianth also with Fumariacese. The only obstacle is in the petals form- ing a whorl of four where all the rest is 2-merous, for the sepals are manifestly two decussating pairs. Now the median petals of Hypecoum are deeply 3-lobed. An abortion of their middle lobe would leave them almost two-parted : a little more would separate them ; then they would imitate the four cruciferous petals as in the diagram, Fig. 395. Applying this view to Cruciferae, the blossom in the two orders would accord in having a 2-merous three-whorled perianth, the first and third whorls median ; a as also in the dimerous androecium, the first whorl of which is lateral. The difference is that in Fumariacese the two members of the first whorl of stamens augment by chorisis into three, and the second is wanting, or is present only in Hypecoum ; while in Cruciferse the first whorl is simple (of the two short stamens) , and the second is doubled. In Fumariaceae only the first whorl of the perianth counts as calyx, and the corolla is of two whorls ; in Cruciferse, the first and second whorls are calyx, the inner sepals answering to the outer petals of Fumariacese. 380. Chorisis along with anteposition of stamens is well seen in Tilia or Linden, at least in the American species. In these the indefinitely numerous stamens are in five clusters, one before each petal (Fig. 398, 399), and there is a petal-like body alternating with the calyx-members as a whole ; the short stamens following as a 2-merous circle ; then the long stamens as a 4-merous circle ; lastly the 2-merous gynoecium. G. Henslow (in Trans. Linn. Soc. ser. 2, i. 195) would have the flower 4-merous by the suppression of the fifth members of a 5-merous type, and a further suppression of half of the remaining exterior stamen-circle, &c. Finally, there is the much better-maintained view that the cruciferous flower is 2-merous throughout, as explained in the following paragraph, 379. i This view was taken by Steinheil, :n Ann. Sci. Nat. ser. 2, 337 (1839), and is essentially reproduced by a Russian botanist, Meschajeff, in Bull. Soc. Imp. Nat. Mosc. 1872. FIG. 398. Diagram of the flower of Tilia Americana, the common American Lin- den or Basswood. FIG. 309. A detached stamen-cluster with its petal-like scale. 208 THE FLOWER. in each cluster with which the stamens cohere. The explanation by chorisis is that each cluster, petal-like body included, is a multiplication of one stamen. The diagram (Fig. 398) accu- rately shows that most of the stamens originate from the outer side of the base of the petal-like portion : this is most naturally explained by median chorisis. The superposition of the clusters to the petals will take the same explanation as that of Ilhamnus, Vitis, &c. (Fig. 363.) That the androocium is here composed of the inner circle merely is partly confirmed by the alternation of the carpels with the clusters. According to Duchartre,1 the development of the andrcecium in a Mallow indicates a similar structure ; for the whole united mass originates from five protu- berances, one before each forming petal and connected with it, this by collateral chorisis forming a cluster of stamens, and the five clusters coalescing as the}" develop into a tube of filaments, such as in Fig. 485. Now Hibiscus and its near relatives have a naked tip to tlie stamen-tube, ending usually in five teeth ; and Sidalcea, as is most striking!}' shown in the California!! S. diploscypha, has two series of stamens, the outer (answering to those of Malva and its relatives) in five membranaceous pha- langes, superposed to the petals ; the rather numerous inner series, more or less in phalanges, surmounts an interior filament- tube. Whence it is inferred that these, and the five teeth terminating the column in Hibiscus, represent the in- ner stamineal circle which is wanting in Malva, as it is in Tilia.2 381. The case of Parnas- sia would be explained as analogous to that of Tilia, but with the stamen-clusters before the petals wholly sterile, and of fewer divisions, while an inner circle of five stamens 1 Comptes Rendus, 1844, & Ann. Sci. Nat. ser. 3, iv. 123. Duchartre and others who draw freely upon median chorisis to explain anteposition, and consider that congenital union proves it, take the phalanges in these cases, like the single stamens in Vitis, to be an inner part of the petal itself. But this view appears to have had its day. 2 Gray, Gen. Illustr. ii. 44, 57, 75-82. The position of the carpels before the petals in Pavonia and Malvaviscus brings the former into symmetrical alternation with such an inner stamen-circle ; but it is not so in Hibiscus, which has the carpels before the sepals. FIG. 400. A petal of Parnassia Caroliniana, with a triple staminodium before it. FIG. 401 . Diagram of the flower of Parnassia Caroliniana. OUTGROWTHS. 209 alternate with the petals forms the effective androecium. For the scale-like body before each petal, and even slightly adnate to its base (in P. Caroliniana about 3-parted, as in Fig. 400, but in P. palustris a thin scale, fringed with more numerous gland- tipped filaments), is plainly outside the stamens in tho full-grown flower-bud. But Eichler and Drude have found that it is inside in the early bud.1 Wherefore, if these stamen-like bodies really represent a circle of the andrceciura, it must be the inner one ; and that is the more probable view. 382. Multiplication by chorisis in the g}TiO3cium is not common ; but there are well marked in- stances of it in all degrees. In Drosera, the styles and stigmas are doubled (Fig. 402) ; in Malvaceae, the same thing takes place in Pavonia and its allies ; while in Malope and two other genera of the same order the few normal carpels are multiplied, evidently by chorisis, into an indefinite number of wholly distinct § 8. OUTGROWTHS. 383. Proper chorisis is the congenital multiplication of one organ into two or more of the same nature and office ; or at least into two or more organs, even if dissimilar, as in the American Lindens, in which one member of the cluster is a kind of petal. Between this and the production by an organ of ap- pendages, or outgrowths of little or no morphological signifi- cation, there are many gradations ; as also between these and mere cellular outgrowths from the surface, even down to bristles and hairs. The latter, in all their variety and modifica- tions, are properly outgrowths of the epidermis only, and there- fore consist of extended cells, single or combined, unaccompanied by vascular or woody tissue. To them has been given the general name of Trichomes (Trichoma, pi. trichomata), that. is structures of which hairs are the t}-pe. They may occur upon the surface of any organ whatever. Their morphology is the morphology of cells rather than of organs. They will therefore be most conveniently illustrated under Vegetable Anatomy as 1 Eichler in Fl. Brasil., Sauvagesiacae, & Bliithend. ii. 424 ; Drude in Lin- naea, xxxix. 239. Eichler refers to this as a confirmation of Celakowsky's explanation of obdiplostemony by posterior displacement. (3G5.) FIG. 402. Pistil of Drosera filiformis with tricarpellary ovary (transversely divided), and six styles, i. e. three, and each two-parted. 210 THE FLOWER. respects their structure, and in the Glossary as respects ter- minology. 384. But into some bristles, such as those of Drosera, a sub- jacent stratum of tissue enters, including one or more ducts or even some woody tissue. Prickles are of this class ; and from the most slender, which pass into bristles, there are all grada- tions of stoutness and induration. Such outgrowths may even be formed in most regular order, as the prickles on the calyx-tube of Agrimonia and scales on the acorn-cup of Oaks, and yet have no morphological importance. On the other hand, true represen- tatives of leaf or stem may, by abortion and depauperation, be reduced to the structure as well as the appearance of trichomes. Examples of this are familiar in the pappus (answering to limb of the calyx) of many Composite, and in the bristles which answer to perianth in many Cyperacese. The scarious stipules of Paronychia and of Potamogeton, the ligule of Grasses, and even the corolla in Plantago, are equally reduced to mere cellular tissue. So that the structural difference between trichomes and outgrowths 1 is not at all absolute, and the morphological distinc- tion must rest upon other ground than anatomical structure. 385. Among the corolline outgrowths most akin to chorisis is the Crown ( Corona) of Silene and allied Caryophyllaceae, at the junction of the claw with the blade of the petals (Fig. 403), the analogy and probable homology of which to the ligule of Grasses (Fig. 150) is evident ; also the many-rayed fila- mentous crown of Passion-flowers (Fig. 404) , which consists of two or more series of such outgrowths. In Sapindus and some other Sapindaceae, these ligular outgrowths or internal appendages are more like a doubling of the petal ; as also in Erythroxylum, where they 1 This is the best English name for the Emergenzen of the Germans, the Epiblastema of Warming, &c. For the development and discussion of this subject, see Warming, in Kjobenhavn Vidensk. Meddel. 1872, and a larger treatise on Ramification in Phanerogams, Copenhagen, 1872. Also, Uhl- worm in Bot. Zeit. 1873 ; Celakowsky in Flora, 1874 ; and Eichler's note on Emergenzen in Bliithendiagramme, i. 48. FTG. 403. Petal of Silene Pennsylvanica, with its crown. FIG. 404. Flower of Passiflora caerulea, reduced in size. FOEMS OF THE RECEPTACLE. 211 are often more complicated in structure. They are always on the inner face, and are commonly two-lobed or parted. 386. Similar stamineal appendages are well known in Cuscuta (Dodder), in Larrea (Fig. 405) and other Zygophyl- laceae, and less conspicuously in Gaura. 387. To extend to them the name of LIGULE may not be amiss, whether they are regarded as mere outgrowths of floral leaves, without further morpho- logical relations, or whether they be, at least some- times, interpreted as the homologue of intrapetiolar stipules, as their ordinarily two-cleft form, and their m coincidence in Erythroxylum with an iutrapetiolar two-cleft stipule suggest. § 9. FORMS OF THE TORUS OR RECEPTACLE. 388. Torus is the more specific and proper name, RECEPTACLE is the more usual. (303.) A normal receptacle of the flower would be that of Fig. 316, the apex of the flower-stalk somewhat enlarged, roundish or depressed, and with surface mainly cov- ered by the insertion of the several organs ; the several inter- nodes which it potentially contains being undeveloped. As the members of the flower multiply and occupy numerous ranks, the receptacle enlarges or lengthens to give them insertion or standing-room. 389. Of elongated forms of receptacle, Magnolia and Liriodendron or Tulip-tree give familiar instances. The lengthening in the former is mainly for the support of both an- drcecium and gynoecium ; in the latter, as in M}-osurus, mainly for the gyncecium onl}-. The fall of the matured carpels reveals it as a very slender or bodkin-shaped pro- longed axis. Of broadened forms, the Straw- berry, even in blossom, affords a familiar example. (Fig. 406.) In the same order, Rubus odoratus shows a very broad and flat receptacle : in roses, it is so deeply concave as to become the reverse of the strawberry (Fig. 407) , being urn-shaped with a narrow mouth, upon which the petals and stamens are borne, FIG. 405. Stamen of Larrea Mexicana, with a conspicuous ligulate appendage at the base within. FIG. 406. Receptacle of a strawberry in longitudinal section. 407. Same of a rose, in diagram. 212 THE FLOWER. while the pistils line the walls of the cavity, the base or centre of this cavity answering to the apex of the strawberry. 390. Sometimes internodes are lengthened between certain members. In Schizandra, the receptacle, barely oblong in blos- som, lengthens greatly in fruiting, so as to scatter the carpels on a long filiform axis. 391. In many Gentians, in Stanleya and Warea among Cruciferae, and in most species of Cleome, the internode of the receptacle between stamens and pistil is developed into * ns stalk to the latter- \\\Uf/r \\\ni IS Gynandropsis (Fig. 409) is like its near relative, Cleome, except that this very long stalk lias the lower part of the stamens adnate to it : the in- ternode between the corolla and calyx is broad and slightly elevated (or in Cleome, &c., narrower and longer) ; and so the several floral circles are as it were spaced apart by this unusual development of receptacular internodes. In Silene (Fig. 408) and many other plants of the Pink family, an internode between the calyx and corolla is prolonged into a stalk or Stipe.1 1 STIPE is the general name of a stalk formed by the receptacle or some part of it, or by a carpel. To distinguish its particular nature in any case, the following terms are more or less employed : — THECAPHORE, for a stipe which belongs to a simple pistil itself (where it is homologous with a petiole), and is no part of the receptacle, as in Coptis or Goldthread. GTNOPHORE, where the stipe is an internode of receptacle next below the gyncecium, as the pod-stalk in some Cruciferae, Cleome, and Gynandropsis. GONOPHORE, when it elevates both stamens and pistil, as it seemingly does in the lower stipe of Gynandropsis, Fig. 409. ANTHOPHORE, when the stipe is a developed internode between the calyx and corolla, as in the Pink family, Fig. 408. GYNOBASE is a term properly applied to a short and comparatively broad portion of receptacle on which the gyncecium rests, as in Rue and Orange (Fig. 414), Houndstongue, Sage, &c. This may extend up between the car- pels and pass into, or the upper part become a CARPOPHORE, a name properly applied to a portion of receptacle which is prolonged between the carpels as a central axis, as in Geranium (Fig. 411) and many Umbelliferse, Fig. 412. FIG. 408. Section of a flower of Silene Pennsylvania, showing the stipe or anthophore FIG. 409. Flower of Gynandropsis, with floral circles separated on the elongated receptacle. FORMS OF THE RECEPTACLE. 213 392. Instead of forming a stalk, the elongation may be continued between the carpels in the form of a slender axis, as in Gera- nium (Fig. 410, 411), and in the carpophore of the fruit of Umbellifera, Fig. 412. In G cranium, this prolongation of receptacle extends far above the ovaries as a beak, to which the styles are adnate for most of their length. 393. In Nelumbium (Fig. 413), the gynophore, or portion of receptacle above the stamens, is enlarged into a singular broadly top-shaped body, with a flat summit, in which the pistils (a dozen or more isolated carpels) are separately immersed. 394. A Disk is a part of the receptacle, or a development of it, enlarged under or around the pistil. When under it or around its base and free from the calyx, the disk is hypogynous, as in Orange, Fig. 414. Here it is a kind of gynobase. When adher- ent to or lining the base of the calyx, it is perigynous, as in 415 Rhamuus (Fig. 415, 416) and Cherry (Fig. 337) : when carried by complete adnation up to the summit of the ovary, it is epigy- nous, as in Cornus, in Umbelliferae,- &c. Not rarely it divides into lobes, as in Vitis (Fig. 379, 380), in Periwinkle and most Apocynaceous plants, and in Cruciferse. These are termed glands of the disk, and indeed are commonly glandular or nectariferous. FIG. 410. Gyncecium of Geranium maculatum. 411. The same with fruit mature, the five ovaries or cells and the lower part of their styles separated and recurving away from the prolongation of the axis or receptacle, to which they were at flowering-time firmly attached. FIG. 412. Mature fruit of Osmorrhiza, the two carpels splitting away below from the filiform prolongation of the receptacle, or carpophore. FIG. 413. The top-shaped receptacle of Nelumbium, with the pistils, immersed in hollows of its upper face. FIG. 415. Flower of a Rhamnus or Buckthorn, and 416, section of the same, show- ing a thickened perigynous disk. 214 THE FLOWER. It is not possible by any direct demonstration to distinguish be- tween such productions of the receptacle, which are classed as belonging to the axis, and suppressed or undeveloped phyllous organs, such as stamens, which glands of the disk may some- times represent. 395. Hypanthium. Inspection of Fig. 415, 416, and 337, and comparison with Fig. 339, will suggest an explanation differ- ent frojn that which is generally adopted. Instead of regarding the calyx as b^tn lining on a level with the base of the ovary, and the cup as lined, more or less thickly, by an expansion of the receptacle (the perigynous disk) , the calyx may be understood to begin where this and the ovary become free from each other. Underthat view, tin receptacle, instead of convex or protu berant, is here con- cave, has grown up 419 around the ovar}', which, however, is free from the cup in the earlier cited figures, but immersed in it in Fig. 339 and the like. A comparison with a rose-hip, an apple, and a pear much strengthens this interpre- tation, which is rather largely adopted at this day, at least theoretically. It was perhaps first proposed by Link, who intro- duced the appropriate name of HYPANTHUM. A hypanthium or hypanthial receptacle is, as the name betokens, a flower-axis or receptacle developed mainly under the calyx. The name is a good one, in any case ; and such structures as those of Calycan- thus (Fig. 417-419), a rose, a pear (the lower part of which is evidently an enlargement of peduncle), and of Cactus-flowers (Fig. 317) , although quite compatible with the theory of adnatioii, are more simply explained by it.1 1 But, whether the cases are well distinguishable or not, it by no means follows that the receptacle plays such a part in all instances of perigyny and of inferior or partly inferior ovary. Such a view is attended by more diffi- culties than the other. Unless the mediation of an invisible receptacle must FIG. 41T. Flowering branch of Calycanthus. 418. Vertical section of the urn-shaped receptacle, the imbricated bracts or sepals on Its surface cut away. 419. Mature fructiferous receptacle entire, showing some scars from which the bracts have fallen. ADAPTATIONS TO FERTILIZATION. 215 SECTION IV. CERTAIN ADAPTATIONS OF THE FLOWER TO THE ACT OF FERTILIZATION. § 1. IN GENERAL. 396. The introduction into morphological botany of the con- siderations now to be mentioned should have dated from the year 1793, in which Christian Conrad Sprengel published his curious treatise on the structure of flowers in special reference to insect aid in their fertilization. For this book, which was wholl}" neglected and overlooked for more than sixty years, con- tains along with some fanciful ideas the germs of the present doctrine and many excellent illustrations of it.1 The interest in the doctrine now prevalent is witnessed by a copious special literature, beginning with the publication, in 1862, of Darwin's book on the fertilization of Orchids by the aid of insects.2 be invoked whenever there is a junction of two dissimilar organs, the petals and stamens of a Lythrum or a Cuphea are united with the calyx itself, instead of calyx beginning at the top of a long and simple tube. And if three or more of the floral whorls may be congenitally united, why not these also with the remaining one ? Van Tieghem, in his Anatomie Comparee de la Fleur, maintains wholly the old view, founding it upon anatomical struc- ture and his ability to trace down to the base of the ovary the distinct vascular bundles of the several involved organs. 1 C. C. Sprengel, Das entdeckte Geheimniss der Natur im Bau und der Befruchtung der Blumen, Berlin, 1793. Even earlier, Koelreuter ( Vorlaufigc Nachricht, etc., 1761-1766) recognized the necessity of insect-aid to various blossoms, and described some special contrivances for the purpose. 2 Charles Darwin, On the Various Contrivances by which British and Foreign Orchids are fertilized by Insects, and on the Good Effects of Inter- crossing, London, 1862. Ed. 2, 1877. This last contains a list of the papers and books which bear iipon the subject, published since 1862. Other leading works and papers on the subject are, exclusive of the other volumes and papers of Darwin, more or less referred to hereafter. Treviranus, Ueber Dichogamie, &c., in Bot. Zeitung, xxi. 1863. Hugo von Mohl, Einige Beobachtungen iiber dimorphe Bliithen, Bot. Zeitung, xxi. 1863. Delpino, Pensieri sulla Biologia Vegetale, &c., 1867. Relazione sull' Apparecchio della Fecondazione nelle Asclepiadie, &c., 1867. Ulteriore Osservazioni sulla Dichogamia, &c., 1868-69, 1870, and later papers. Axell, Om anordningarna for de Fanerogama Vaxternas Befruchtung, Stockholm, 1869. Hildebrand, Die Geschlechter-Vertheilung bei den Pflanzen, 1867, and other papers. Hermann Miiller, Die Befruchtung der Blumen durch Insekten, 1873, and papers in " Nature " and elsewhere. " Flowers and their Unbidden Guests," an English translation of a work by Professor Kerner, which describes arrangements in blossoms for exclud- ing unwelcome guests, has not yet reached us. It introduces the new terms -\ntotjumij and Allogamy, defined on the following page ; the latter compre- 216 THE FLOWER. 397. The subject, here considered as a part of morphology, must be fully treated, as regards acts and processes, under physi- ology. Every thing in the flower is in relation to fertilization and fructification, directly or indirectly. This section is con- cerned with those adaptations of structure by means of which agents external to the blossom are brought into service for its fertilization. 398. Linnaeus and his successors taught that the adjustments in hermaphrodite flowers were such, on the whole, as to secure the application of the pollen of its stamens to the stigma of its pistil or pistils. The present view is, that this is doubtless strictly secured in certain flowers of a moderate number of species, but never in all the flowers of any such species ; that in ordinary flowers, where it may commonly take place, it is not universal ; that in the larger number of species there is something or other in the floral structure which impedes or prevents it. Some flowers are adapted for close fertilization ; some for cross fertili- zation ; some for either. Here two terms need definition, viz. : Close fertilization or Self-fertilization, or Autogamy, the applica- tion and action of a flower's pollen upon its own pistil ; Cross fertilization, or Allogamy, the action of the pollen of one flower on the pistil of some other flower of the same species. This may be near, as when between flowers borne in the same cluster or on the same plant ; remote, when between flowers of distinct plants of the same immediate parentage ; most remote, when between different races of the same species. Any thing beyond this is hybridization, or crossing of species. § 2. ADAPTATIONS FOR ALLOGAMY OR INTERCROSSING. 399. The doctrine now maintained appears to have been first propounded by Sprengel in the statement that "Nature seems to have wished that no flower should be fertilized by its own pollen," — a proposition which is not wholly tenable, for there are blossoms specially adapted to self-fertilization. It was re- affirmed in our day by Darwin, in a similar adage, "Nature abhors perpetual self-fertilization," — a metaphorical expression to which no effective exception has been taken. And the infer- ence was drawn by him, that some important good to the species must result from propagation through the union of distinct individuals, and especially of individuals which have been dis- tinct for several or many generations. bending Geitonogamy, fertilization by pollen of other flowers of the same plant, and Xenogamy, by pollen from a flower on another plant. ADAPTATIONS FOR INTERCROSSING. 217 400. The actual proposition, simply stated, is that flowers are habitually intercrossed, and that there are manifold structural adaptations which secure or favor intercrossing, to such extent as to justify the proposition. The proof of the proposition is an induction from a very great number of particular observations. That intercrossing is beneficial is a rational inference from the arra}r of special adaptations for which no other sufficient reason appears, or (to resume the metaphor) from the vast pains which seem to have been taken to secure this end. This inference has been to some extent confirmed by direct experiment.1 401. Separation of the sexes is a direct adaptation to inter- crossing, rendering it necessary between individuals in dioecious, and largely favoring it in most monoecious and polygamous flowers. Strictly close fertilization can occur in hermaphrodite flowers only ; but it is in these that the most curious adaptations for intercrossing are revealed. 402. The agencies to the one or the other of which most flowers are structurally adapted in reference to intercrossing are mainly two ; viz., the winds and animals, of these chiefly insects. Delpiuo has accordingly classified flowers into Anemophilous and Entomophilous ; literally wind-lovers and insect-lovers, but de- noting wind-fertilized and insect- fertilized, according to the agent by which poUen is transported.2 There are hermaphrodite and unisexual flowers of both classes, but most wind-fertilized flowers are unisexual. 403. Wind-fertilizable or anemophilous flowers are mostly neu- tral or dull in color, destitute of odor, and not nectariferous. Their principal structural adaptations to this end, besides the separation of the sexes in most of them, are the superabundance, incoherencj', dryness, and lightness of the pollen, rendering it very transportable by wind and currents of air. The immense abundance of pollen, its lightness, and its free and far diffusion through the air in Pines, Firs, Taxodium, and other Coniferae, are familiar. Their pollen fills the air of a forest during anthe- sis ; and the " showers of sulphur," popularly so-called, the yellow powder which after a transient shower accumulates as a scum on the surface of water several or many miles from the 1 Darwin, The Effects of -Cross and Self-Fertilization in the Vegetable Kingdom, London, 1876. American Edition, New York, 1877. 2 Ornithophilous, i. e. bird-fertilized, flowers are to be ranked with entomo- philous. The large blossoms of Trumpet Creeper (Tecoma radicans) and of Trumpet Honeysuckle (Lonicera sempervirens), and others, are commonly visited and probably fertilized by humming-birds as well as by moths ; and other birds are known to play a similar part in equatorial regions. '218 , THE FLOWER. nearest source, testifies to these particulars. All amentaceous trees (Willows excepted) , Hemp, Hops, &c., are wind-fertilized ; and, among perfect flowers, those of most Grasses, Sedges, and Plantago. In the latter families especially, the anthers are pro- truded or hung out in the air only when just ready to discharge their pollen, and are at that moment suspended on suddenly lengthened capillary drooping filaments, fluttering in the gentlest breeze ; and the stigmas are either dissected into plumes, as in most Grasses, or beset with copious hairs on which pollen is caught. One physiological adaptation, very common in the fol- lowing class, is not unknown among hermaphrodite wind-fertiliz- able flowers, where it is important for securing intercrossing, viz. Dichogamy. It is best seen in the common species of Plantago or Plantain, and is described below. (408.) 404. Insect-fertilizable or entomophilous flowers are correlated with showy coloration (including white, which is most showy at dusk), odor, or secretion of nectar, often by all three modes of attraction to insects combined. Some insects, moreover, visit flowers for their pollen, a highly nutritious article, and ordina- rily produced in such abundance that much may be spared. The showiness of corolla or other floral envelopes is an attractive adaptation to fertilization, enabling blossoms to be discerned at a distance ; nor do we know that fragrance or other scent or that nectar subserves any other uses to the flower than that of alluring insects. Adaptations in the pollen of such blossoms for transportation by insects are various. Commonly the grains are slightly moist or glutinous, or roughish, or studded with projections, or strung with threads (as in (Enothera) , so as not to be readily dispersed in the air, but to have some slight coherence as well as capability of adhering to the head, limbs, or bodies of insects, especially to their rough surfaces ; and in two families (Orchidacese, Asclepiadacese) the pollen is com- bined in masses and with special adaptations for being trans- ported en masse. (421.) With this the stigma is usually correlated, by roughness, moisture, or glutinosity . * 405. Adaptations of the flower itself in reference to insect visitation are wonderfully various ; and most of these are found upon investigation to favor, or often to necessitate, intercross- ing. In dioecious flowers, this is necessitated b}r the separation ; in monoecious and polygamous flowers, of various kinds and 1 Thus nearly every Orchid genus but one has a persistently glutinous stigma ; in the exceptional one, Cypripedium, it is moist and minutely rough- ened, in correlation with the loosely granular or pultaceous pollen which it is to receive. ADAPTATIONS FOR INTERCROSSING. 219 degrees of separation, pollen is very commonly borne from plant to plant ; in hermaphrodite flowers only are more special arrangements needed to secure intercrossing or a certain measure of it, and in these such arrangements abound. 406. Irregularity is one of the commonest modifications of the flower (326, 337) : it is never conspicuous except in blossoms 7V7, visited by insects and generally fertilized by their aid ; and it finds rational explanation on the score of utility in this regard.1 407. Dichogamy, a term introduced b}- C. C. Sprengel, who first noticed and described it, is one of the most usual and effect- ual (rather physiological than morphological) adaptations for the promotion of intercrossing between hermaphrodite flowers. It means that such intercrossing is brought to pass b}" a difference » in the time of maturit}* of anthers and stigma ; this rendering Jiehogamous blossoms practically the same as dioecious or mon- oecious in respect to fertilization, while there is the economical gain that all the flowers are fertile. According to whether the anthers or the stigmas are precocious, dichogamous flowers are Proterandrous (or Protandrous) , when the anthers mature and discharge their pollen before the stigma of that blossom is recep- tive of pollen ; Proterogynous (or Protogynous) , when the stigmas are in receptive condition before the anthers have matured their pollen. Synanthesis* the maturing of the two sexes simultaneously or nearly so, is however made to secure the same result through special arrangements. 408. Proterogyny. The Plantains, such as Plantago major and P. lanceolata, are familiar instances of this in a wind-fertilized genus writh hermaphrodite flowers. The anthesis proceeds from base to apex of the spike in regular order, and rather slowly. AVhile the anthers are still in the unopened corolla and on short filaments, the long and slender hair}7 stigma projects from the tip and is receiving pollen blown to it from neighboring plants or 1 This did not escape the attention of Sprengel in the last century, and s.long with it the fact that strictly terminal and also vertical flowers, whether erect or suspended, are seldom irregular, while comparatively horizontal or obliquely set flpwers more commonly are so. The irregularity is in refer- ence to a landing place for the visiting insect, or also to storage of or access!- . bility to nectar, &c. Darwin (Forms of Flowers, 147) remarks that he does not know of a single instance of an irregular flower which is wind-fertilized. 2 Synacmy is the term proposed by A. W. Bennett, in Journal of Botany, viii. (1870), 316, with its opposite, Heteracmy, for proterandry and proterogyny. The latter names, in their shorter form (protandry and protoyyny), appear to have originated with Hildebrand, 1867. 220 THE FLOWER. spikes : a day or two afterwards, the corolla opens, the filaments greatly lengthen, and the four anthers now pendent from them give their light pollen to the wind ; but the stigmas of that flower and of all below it on that spike are withered or past receiving pollen. Among Grasses, Anthoxanthum is in the same case. The arrangement is somewhat similar to the Plantain in Amor- pha, which is fertilized by insects, the simple stigma projecting beyond the corolla in bud, while the anthers are still immature and enclosed. Scrophularia is a good instance of proterogyny in flowers fertilized by bees. The flower is irregular (Fig. 420-422) , and is approached from the front, the spreading lower lobe being the landing place. Fig. 420 represents a freshly opened blossom ; and Fig. 421, a section of it. Only the style tipped with the stigma is in view, leaning over the landing place ; the still closed anthers are ensconced below. The next day or a little later all is as in Fig. 422. The style, now flabby, has fallen upon the front lobe, its stigma dry and no longer receptive : the now-opening anthers are brought upward and forward to the position which the stigma occupied before. A honey-bee, taking nectar from the bottom of the corolla, will be dusted with pollen from the later flower, and on passing to one in the earlier state will deposit some of it on its fresh stigma. Self-fertilization here can hardly ever take place, and only through some disturb- ance of the natural course. 409. Proterandry. The process is the reverse, and is at- tended with much more extended movements in Clerodendron Thompson] ae, a Verbenaceous tropical African climber now com- mon in conservatories. The adaptations in this flower (which we indicated long ago) are exquisite. The crimson corolla and bright white calyx in combination are very conspicuous. The long filiform filaments and st}rle, upwardly enrolled in the FIG. 420, 421. Early opened flower of Scrophularia nodosa, and a longitudinal section. 422. Flower a day or two later. ADAPTATIONS FOR INTERCROSSING. 221 bud, straighten and project when the corolla opens : the stamens remain straight, but the st3'le proceeds to curve downward and backward, as in Fig. 423. The anthers are now discharging pollen : the stigmas are immature and closed. Fig. 424 repre- sents the flower on the second day, the anthers effete, and the filaments recurved and rolled up spirally ; while the style has taken the position of the filaments, and the two stigmas now separated and receptive are in the very position of the anthers the previous day. The entrance by which the proboscis of a butterfly may reach the nectar at bottom is at the upper side of the orifice. The flower cannot self-fertilize. A good-sized insect flying from blossom to blossom, and plant to plant, must transport pollen from the one to the stigma of the other. 410. Proterandry abounds among common flowers. It is conspicuous in Gentians and in nearly all that family. But. while in Gentians the short style is immovable and erect, in Sabbatia it is thrown strongly to one side, out of the way of and far below the stamens, the branches closed and often twisted, so that the stigma is quite inaccessible until the stamens have shed their pollen : then the style becomes erect, untwists, its two flat branches separate, and expose the stigmatic surface of their inner face in the place which the anthers occupied. In Sabbatia angularis, Lester F. Ward x observed that the anthers of freshly 1 In Median's Gardeners' Monthly, September, 1878, 278. FIG. 423. Flower of Clerodendron Thompsoniae, first day; 424, second day. 222 THE FLOWER,. opened blossoms are all thrown to one side almost as strongly as the style is thrown in the opposite direction. One of our common Fireweeds, Epilobium angustifolium or E. spicatum, as it is variously called, which is common all round the northern hemisphere, is similar to Sabbatia in behavior. In the freshly opened flower, while the anthers are in good condition and are giving their pollen to bees, the still immature style is strongly curved downward and backward, as in Fig. 425. Two or three days later, when the pollen is mostly shed, the style straightens, lengthens to its full dimensions, and spreads its four stigmas over the line of the axis of the blossom (Fig. 426), in the verj position to be pollinated by a bee coming from an earlier flower. 411. In the following instances of proterandry, the style is made the instrument of distributing the pollen which it is not itself to use. The anthers of a Cam- panula discharge all their pollen in the unopened bud, and it is nearly all deposited on the style which they surround, the upper part of which is clothed with a coat of hairs for holding 427 428 the pollen. (Fig. 427.) In the open flower, the stamens are found to be empty and withered, as in Fig. 428. These flowers are visited by bees and other insects for the pollen. While this is going on, and while the pollen is fresh and plentiful, no stigma is apparent. Later, the top of the st}*le opens into three (in some species five) short and spreading branches, the inner faces of which are the stigmas. Although FIG. 425, 426. Flowers of Epilobium angustifolium or spicatum; in the first, freshly expanded ; in the second, a few days older. FIG. 427. Vertical section of an unopened flower of Campanula rapunculoides : the broad white lines are sections of two anthers. 428. Same of an older flower. ADAPTATIONS FOR INTERCROSSING. 223 so close at hand, little if any of the pollen of that flower can reach the stigmas. These actually get fertilized by pollen brought by bees, which come loaded with it from other flowers and other plants. Symphyandra differs from a true Campanula chiefly in the continued cohesion of the five anthers into a tube around the style. (Fig. 429, 430.) The pollen is discharged on and held by the hairy upper portion of the style. 8oon after, the corolla expands, the lower part of the style lengthens, and carries the pollen- loaded part out of and above the anther-tube, as in Fig. 430 ; lastly, the three connivent tips of the style diverge and expose the stigmas to pollen mainly brought by bees from other flowers. By a slight further modification in Lobelia and in Composite, pollen is pushed out of the anther-tube by the tip of the style as it lengthens, or by the very back of the two stigmas, the faces of which, afterwards exposed, are not to receive this, but other m ^° pollen, though it may at times receive some of its own. The arrangement in Composites is here illustrated from Leptosyne maritima (Fig. 431-435), a showy plant of Southern California, now not very rare in cultivation. The large flowers around the FIG. 429. Stamens and pistil of a young, and 430, same from an old flower of Symphyandra pendula. FIG. 431. Head of flowers of Leptosyne marifimn, of Hie natural size. THE FLOWER. margin (ray-flowers, with ligulate corolla), one of which is sepa- rately shown in Fig. 432, are pistillate only : the enlarged and extended open part of the corolla (bright yellow in color) serves for attrac- tion, the circle of rays gives the appearance as of a single large flower. The flowers of the disk or whole central part are hermaphrodite, and with narrow tubular corollas, from the orifice of which projects the greater part of the tube of five co- alescent anthers. The pollen is early discharged into the interior of this tube. The style, with somewhat enlarged and brush-like tip, at first reaches only to the bottom of the anther-tube : it slowly lengthens, pushes the pollen before it out of the tube (Fig. 433) and into the way of insects of various kind, which, travelling over the surface, con- vey it to older flowers of the same head and of other plants. The style, elongating yet more, raises some of the pollen still higher (as in Fig. 434) ; and at length its two branches separate and diverge (Fig. 43ft), exposing to other pollen the stigmatic receptive surface which until now was un- approachable. 412. In Parnassia, which has sessile stigmas, their receptive surface is actually not formed until the anthers become effete ; FIG. 432 A ligulate female flower of the same, and a central hermaphrodite flower. 433. Upper part of the latter, more enlarged, the tube of anthers projecting from the corolla, and the pollen projecting from apex of the anther-tube, being pusheil up by the lengthening of the style beneath. 434. This style now projecting, and some pollen still resting on its tip 435. Tip of same style (more advanced and magnified); the two branches spreading, still carrying some pollen on the apex of each arm or branch, an*' by the divergence now exposing the stigmatic inner faces. ADAPTATIONS FOB INTERCROSSING. 225 and, as the plants or stems are single-flowered, they are function- ally dioecious while structurally hermaphrodite. 413. The adaptations for hermaphrodite intercrossing with synanthesis (407), i. e. where there is no essential difference of time in the maturing of anthers and stigma, are manifold. They ma}* be classed into those without and those with dimor- phism of stamens and pistils, or, in other words, those with Homogonous and those with Heterogonous flowers.1 414. The cases without dimorphism are the most various, certain families having special t}'pes ; and are of all degrees, from those that require intercrossing to those that merely favor or permit it. For the present purpose, having only morphology in view, it suffices to bring to view two or three cases or types of 415. Particular Adaptations in hermaphrodite blossoms, not involving either dichogamy or dimorphism. These are exceed- ingly various ; but they may be distinguished into two general kinds, namely: 1, where loose and powdery pollen is transported from blossom to blossom in separate grains, and 2, where pollen-masses or the whole contents of anthers are bodily so transported. 416. Papilionaceous flowers (such as pea-blossoms, 338) — having ten ^ stamens enclosed with a single pis- "* til in the keel. of the corolla, their anthers in close proximity to the stigma — were naturally supposed to be self-fertilizing; and so they sometimes are, yet with marked adaptations for intercrossing. None are less so than those of 1 Terms proposed in Amer. Jour. Sci. ser. 3, xiii. 82, and in Amer. Naturalist, January, 1877. Dimorphism in flowers may affect the perianth only, and not the yov-fi or essential organs ; or there may be two kinds of flowers as respects these also, but with no reciprocal, relations, as in cleisto- (jamous dimorphism (534) ; or of two kinds essentially alike except in stamens and pistil, and these reciprocally adapted to each other, which is heteroijorons dimorphism, or, when of three kinds, trimorphism. FIG. 436. Flower of Wistaria Sinensis natural size. 437. Same enlarged, with standard, wings, and half the keel removed. 438. Same with the keel depressed, as it is when a bee alights on this its usual landing place, the cluster of anthers and stigma, thus brought up against the bee's abdomen. 439. Style and stigma, with part of the ovary, more magnified, a fringe of fine bristles around the stigma. 15 226 THE FLOWEK. Wistaria (Fig. 436-439), in which the light fringe of stiff hairs around the stigma (shown in Fig. 439) would not prevent pollen of surrounding anthers from falling upon it. Yet when a bee alights upon the keel, with head toward the base of the flower, and proboscis is inserted for nectar between the foot of the standard and the keel, the latter is depressed by the weight, so that the ab- domen of the insect is brought against the ten anthers and the stig- ma, becoming thereby smeared with pollen, some of which when other blossoms are vis- ited cannot fail to be applied to their stigmas. The very similar flower of Locust (Robinia), like that of the Pea, adds an adapta- tion in favor of intercrossing. The style for some length below the stigma is covered with a short beard of hairs, as is seen in Fig. 442. The Anthers open early and dis- charge their pollen, which mainly lodges on this beard (Fig. 443), in a manner which may thus far be likened to the case of Campanula. (411.) The wings and the keel are yoked together, and are together depressed by the weight of an alighting bee. This does not bring out the anthers as in Wistaria, but these remain until effete within the sac, while the stigma and the pollen-laden part of the style (Fig. 441) are projected against the bee's abdomen, which, by the oblique movement, is first touched by the stigma and next brushed over with pollen by the style below. So that, in visiting a succession of blossoms, some pollen of one flower is transferred to the body of the bee, and thence to the stigma of the next flower, which flower immediately gives to the same spot some of its pollen, to be transferred to the next flower's stigma, and so on. 417. Two special modifications of the papilionaceous type FIG. 440. Flower of Robinia hispida, the standard and wings removed. 441. Same, as depressed by the weight of a bee, causing the stigma and pollen-laden tip of the style to protrude. 442. Enlarged section of same in the bud, leaving one keel-petal, half the stamens, and the pistil in view. 443. Style and stigma at a later period, the beard loaded with pollen ; more magnified. ADAPTATIONS FOR INTEflUltOSSlNG. 227 need particular mention. One of them, the Bean-blossom, is well known to botanists ; the other not so. The peculiarity in the common Bean, Phaseolus vulgaris, and its nearest relatives, is that the keel, enclosing the stamens and pistil, is prolonged into a narrow snout which is spirally coiled (as in Fig. 444-446) ; that the stigma is oblique on the tip of the style, and the beard on the style is mainly on the same side that the stigma is : the wing-petals stand forward and turn downward, forming a con- venient landing place for bees. As in the Locust-blossom, the anthers early discharge their pollen, much of which adheres lightly to the beard of the style. In the untouched flower, all from first to last is concealed in the coiled keel. Press down the wing-petals, and first the stigma and then the pollen-laden tip of the style projects from the orifice : remove the pressure, and they withdraw within. When this pressure is made by a bee, resting on the wing-petals while searching for nectar within the base of the blossom between the keel and the standard, the same movement occurs : the stigma first, and then the pollen on the style, strikes against a certain portion of the front or side of the bee's body, and the repetition of this operation causes the fertilization of each blossom by other than its own pollen. A slighter pressure or lighter movement of the wing-petals suffices FIG. 444. Flower of Garden Bean, Phaseolus vulgaris. 445 Same with wing- petals pressed down and tip of style projecting from the orifice of the keel. 446. Same as 444 enlarged, and standard and wings removed. 447. Upper part of keel, in the condition of 445, enlarged, showing plainly the projecting style. 448 Section of the keel, enlarged, showing the style within before the anthers open: stamens for sake of clearness not delineated. 449. Pistil detached from an older flower ; the brush loaded with pollen. 228 THE FLOWEE. to jostle some of the pollen down upon its own stigma, so that self-fertilization is not uncommon. 418. Apios tuberosa, a near relative of Phaseolus, exhibits a different and equally curious modification of the same parts. The wing-petals for landing place are similar : the standard is pro- portionally large, firm in texture, and shell-shaped or concave, with a sma^ boss at the tip as seen from behind, or a shallow sac as seen 451 from the front : the keel is narrow and sickle-shaped ; it arches across the front of the flower, and the blunt apex rests in the notch or shallow sac of the tip of the standard. (Fig. 450, 452, 453.) So it remains if untouched until the blossom withers : no self-fertilization has ever been observed, and none ordinarily occurs. The anthers are assembled close around the stigma, but a little short of it (Fig. 452) ; the pollen is not early nor copiously shed in the enclosure : the small terminal stigma is at first covered with a pulpy secretion, which at length collects into a soft ring- around its base over or through which no pollen passes. But when the keel is liber- ated by lifting from underneath, it curves promptly into the shape shown in Fig. FIG. 450. Flower of Apios tuberosa, unvisited. 451. Same after visitation, the keel dislodged from the retaining notch, and more incurved; the tip of the style pro- truded and thrust forward, followed by the anthers. FIG. 452. Enlarged vertical section of flower-bud of Apios tuberosa. 453 A flower with half the standard cut away, to show the blunt apex of the keel resting in t)!e notch. 454. Diagram of flower, with half of the standard cut away, to show what takes place when the apex of the keel is liberated. The figures (also those from 4L'3 to the present), and the first accoHnt of the adaptations of Apios, were published in the Amer- ican Agriculturist in 1876. ADAPTATIONS FOR INTERCROSSING. 229 451, or better in Fig. 454, where the dotted lines indicate its original position ; and first the end of the st3'le, tipped with its stigma, is pushed forward, and then the anthers come into view. The flowers are visited by humble-bees, and sometimes b}r hone}--bees. In searching for nectar at the base of the flower, they probably push forward into the space under the arching keel, and by slightly elevating dislodge its apex ; when first the stigma and then the anthers are brought against some portion of the insect's body, and against the same portion in succeeding blossoms, thus effecting cross-fertilization. This rationally ex- plains a remarkable adaptation, which seems to be not otherwise intelligible. 419. Special Adaptations. Two of these, each peculiar to the genus, may here be referred to. In Kahnia-blossoms (Fig. 455-458), the anthers discharge the pollen through a small orifice at the apex of each cell, in this respect agreeing with Rhododendrons and their other relatives ; but none of them utilize this family peculiarity in the manner of Kalmia. In the flower-bud, each of the ten anthers is lodged in a small cavity or pocket (externally a boss) of the corolla, in a vfay analo- gous to that in which the keel of Apios is lodged in the tip of the standard (418) : the expansion of the border of the corolla in an thesis curves the fila- ments outward and backward ; and when the bowed stamens are liberated by rough jostling they fly up elastically, and the pollen is projected from the two orifices. Some pollen may possibly be thrown upon the single small stigma at the tip of the st}ie, which rises much above the stamens. But the- anthers are not dislodged when undisturbed, at least until after the elasticity of the filaments is lost : they are dislodged by humble-bees, which circle on the wing over the blossom, the FIG. 455. Vertical section of a flower-bud of Kalmia latifolia, showing the anthers 1 xlged in the pockets of the corolla. 456. Expanded flower, with towed stamens 457. Vertical section of the same, 458. A stamen, enlarged. 230 THE FLO WEB. under side of the abdomen frequently touching the stigma, while the proboscis is searching round the bottom of the flower, liberat- ing the stamens in the process, which one by one project their pollen upon the under side of the insect's body. In the passage from flower to flower, pollen is thus conveyed from the anthers of one to the stigma of another. 420. Iris has three stamens, one before each sepal or outer lobe of the perianth, and behind each petal-like lobe of the style (Fig. 459) : the stigma, a shelf- like plate of each lobe, is just above the anther ; but, as the anther faces outward and the stigma is higher and faces inward, no pollen can find its way from the one to the other. But the adaptation of parts is admir- able for conveyance by bees, which, standing upon the only landing place, the re- curved sepal, thrust the head down below the anther, and in raising it carry off pollen, to be afterwards lodged upon the stigmas of other flowers which they visit. 421. Transportation of Pollinia, or of all the pollen in a mass, is effected in most of the species of two large orders, not otherwise allied, the Asclepiadacese and the Orchidacece. While in the Iris family the number of stamens is reduced from six to three, in all the Orchis family, except Cypripedium, the stamens are further reduced to a single one ; but the pollen is peculiarly economized. That of Arethusa is in four loose and soft pellets, in an inverted casque-shaped case, hinged at the back, resting on a shelf, the lower face of which is glutinous stigma, over the front edge of which the casque-shaped anther slightly projects ; and this anther is raised by the head of a bee when escaping out of the gorge of the flower. The loose pellets of pollen are caught upon the bee's head, to the rough sur- face of which they are liable to adhere lightly and so to be carried to the flower of another individual, there left upon its glutinous FIG. 459. Flower of Iris pnmila, with front portion and half of one petaloid style- lobe and stigma cut away. The section of the stigma is seen edgewise: the rough upper surface only is stigmatic. ^ ADAPTATIONS FOR INTEKCKOSSLSG. 231 stigma by the same upward movement which immediately after- ward raises the anther-lid and carries away its pollen, to be transferred to a third blossom, and so on. 422. But it is in Orchis and in the commoner re- presentatives of Orchis in North America (viz. Ha- benaria, &c.) that the most exquisite adapta- tions are found, and the greatest econonry se- cured ; paralleled, how- ever, by most of the very numerous and vaiious epi- phytic and by various ter- restrial Orchids of warmer regions. A single illus- tration may here suffice ; and Darwin's volume on the Fertilization of Or- chids (396, note), with its references to the copious literature of the subject, may be studied for full particulars and their bearings. The flower is trimerous, and the peri- anth adnate to the ovary, therefore apparently de- veloped upon its sum- mit. The three external parts of the perianth, which in Habenaria orbi- culata (Fig. 460) are much the broader, are the sepals: the three alternate and internal, the petals : the base of the long and narrow petal which is turned downward is hollowed out and extended below into a long tube, closed at bottom, open at top (the spur or nectary), in which nectar is FIG. 460. Flower of Habenaria or Platantbera orbiculata, enlarged. 461. Combined stamen and stigma, more enlarged. 462. One of the two pollen-masses (po/linia), with its stalk and glutinous disk or gland. 462". Lower part of this stalk and its disk.more magnified. 232 THE FLOWEIv. copiously secreted and contained. The central part of the blossom, beyond the orifice of the nectary (shown separately in Fig. 461), consists of one anther and a stigma, fused together (the dinandrium) : the marginal portions, opening by a long chink, are the two cells of the anther, approximate at their broader portion above, widely divergent below : most of the lower part of the space between is excessively glutinous, and is the stigma. The grains of pollen are united b}" means of short threads of very elastic tissue into small masses, and these into larger, and at length into pellets, having stalks of the same elastic tissue, by which they are all attached to a firmer central stalk, or caudick. (Fig. 463-465.) To the lower end of this caudicle (directly to the end of it in our Habenariae and Orchises gener- ally, in this instance to the inner side of the end, with a thick inter- mediate base intervening), is at- tached a button-shaped disk, the face of which is exposed, and is on a line with the surface of the anther ; so that these two disks look toward each other across the broad intervening stigma tic space, as seen in Fig. 461. The exposed face of the disk being covered with a durable layer of very viscid mat- ter, the body itself is sometimes termed a gland, and not improperly. The viscidity is nearly of the same nature as that of the interven- ing stigma, of which the glands are generally supposed to be detached portions. If so, then a portion of the stigma is cut off from the rest and specialized to the purpose of conveyance of the pollen. When a finger's end or any smaller body is touched to these disks, the}' adhere so firmly that the attached pollinia or pollen-masses are dragged out of the cell and carried away en- tire. Some of these pollen-masses have been found attached by the disk to the eyes of a large moth. "When a moth of the size of head and length of proboscis of Sphynx drupiferarum visits a spike of these flowers, and presses its head into the centre of the FIG. 463. A more magnified pollen-mass of Platanthera orbiculata. with its stalk and gland. 464. Five of the separate portions or pollen-packets, with some of the elastic threads of tissue connecting them. 465. A portion more highly magnified, with some of the pollen-grains in fours detached. ADAPTATIONS FOR INTERCROSSING. 233 flower so that its proboscis may reach and drain the bottom of the nectariferous tube, a pollen-mass will usually be affixed to each eye : on withdrawal, these will stand as in Fig. 466. Within a minute they will be turned downward (Fig. 466°), not by their weight, but by a contraction in drying of one side of the thick piece which connects the disk with the stalk. When a moth in this con- dition passes from the last open flower of one spike to that of another plant, and thrusts its proboscis down a nectary, the transported pol- len-masses will be brought in contact with the large glutinous stigma : on with- drawal, either some of the small pellets of pollen will be left adherent to the stigma, the connecting elastic threads giving way ; or else a whole pollen-mass will be so left, its adhesion to the glutinous ** ° stigma being greater than that of the disk to the moth's eye. The former is a common and a more economical proceeding, as then a succession of flowers are abundantly fertilized by one or two pollen-masses. In either case, new pollen-masses are carried off from fresh flowers and applied to the fertilization of other blossoms on the same and eventually on those of differ- ent individuals. Cases like this, and hundreds more, all equally remarkable, serve to show how sedulous, sure, and economical are the adaptations and processes of Nature for the intercrossing of hermaphrodite flowers. 422a. An arrangement analogous to that of Orchids, and similarly subservient to cross-fertilization, characterizes the otherwise widely unlike Asclepias family. In Asclepias (Milk- weed) there are five stamens surrounding a large stigmatic FIG. 466. Front part of Sphynx drupiferarum, bearing a pollen-mass of Platan- thera orbionlata affixed to each eye, in the early position. 466 a. Front view of the head. !ater, showing the pollen-masses deflexed. 234 THE FLOWER. body, and alternating with these five two-cleft glands, the ver- tical chink or groove of which is glutinous. To each gland is firmly attached, b}r a caudicle or stalk, a pollen-mass of an ad- jacent anther. (Fig. 522.) A slight force upraising the gland detaches it from the stigma and drags the pair of suspended pollen-masses out of their cells. Insects visiting the blossoms commonly dislodge them, the gland adhering to their legs or tongues when these happen to be drawn through the adhesive chink, and convey them from one flower to another. Without such aid the flowers of Asclepias rarely set seed.1 423. Dimorphism, i. e. the case of two kinds of blossoms, both hermaphrodite, on the same species, is another adaptation to intercrossing. Not all dimorphism, however, for in cleistogamous dimorphism (434) the intent to self- fertilize is evident. There may also be dimorphism as to the perianth, not particular^" affecting fertilization. One kind, however, and the commonest, is a special adaptation to intercrossing, viz. : 424. Heterogonous Dimorphism. (413, note.) This term U applied to the case in which a species produces two kinds oJ hermaphrodite flowers, occupying different individuals, the flowers, essentially similar except in the androecium and gynoecium, bui these reciprocally different in length or height, and the adapta tions such that, by the agency of insects, the pollen from the stamens of the one sort reciprocally fertilizes the stigma of the other.2 This dimorphism has been detected in about fort}- geneva belonging to fourteen or fifteen natural orders, widely scattered through the vegetable kingdom ; but there are far more examples among the Kubiacese than in any other order. Sometimes all the species of a genus are heterogonous, as in Houstonia, and 1 The reported sensitiveness of the gland, referred to in the first issue of this volume (1879), was founded upon misinterpreted observations. 2 This peculiar arrangement has been long known in a few plants, sucli as Primula vcris, P. grandiflora, and Houstonia. In Torrey and Gray's Flora of North America, ii. 38, 39 (1843), these flowers are said to be dicecio- dimorphous, not denoting that they are at all unisexual, but that the two forms occupy different individuals. Their meaning was detected by C. Darwin, and made known in his paper "On the Two Forms or Dimorphic Condition in the Species of Primula, and on their Remarkable Sexual Rela- tions," published in the Journal of the Linnean Society, vi. (1862), 77 : repub- lished, in 1877, as the leading chapter of his volume entitled "The Different Forms of Flowers on Plants of the Same Species." Mr. Darwin had termed these flowers simply Dimorphic. ; but in this volume he adopted Hilde- brand's name of Heterosti/led for this kind of blossom. The difference, however, affects the androecium, and even the pollen, as well as the style; wherefore we proposed for it the name of Heterogonous or Heteroyone dimor- phism, as mentioned in a former note, 413. ADAPTATIONS FOR INTERCROSSING. 235 Cinchona, sometimes only a part of them, as in Primula and Linum. In Hottonia, a Primulaceous genus of two species, the European one has heterogonous dimorphism * for cross-fertiliza- tion : the American one has homogenous show}* flowers with only the general chance for intercrossing, and earlier flowers which are cleistogamous for self-fertilization. 425. The nature of heterogone dimorphism ma}* be well under- stood from a single example. The most familiar one is that of Houstonia ; but, in larger blossoms, Gelsemium is a fine illus- tration in the Southern United States, and Mitchella (Fig. 467) mostly in the Northern. Raised from the seed, the individuals are about equally divided between the two forms : namely, one form with long style and short or low-inserted stamens ; the other with short style and long or high-inserted stamens. The stigmas in one rise to about the same height as the stamens in the other, both in the tall or exserted organs and in their low and included counterparts, as is shown in Fig. 468, answering to the left hand and Fig. 469 to the right hand flowers of Fig. 467. A bee or other insect with proboscis of about the length of the corolla-tube, visiting the blossoms of Mitchella, will brush the same part of its bodjr against the high anthers of the long- stamened and the high stigmas of the long- sty led forms ; and 1 C. C. Sprengel, as Darwin mentions, had noticed this, before 1793. He, " with his usual sagacity, adds that he does not believe the existence of the two forms to be accidental, though he cannot explain their purpose." Darwin, Forms of Flowers, 51. Some heterogonous Primulas are said to produce homogonous varieties in cultivation. In Primula, and in other genera, there are species which seem as if of one sort only, no reciprocal sort being known, as if one form had become self-fertile and the other had disappeared. FIG. 467. Partridge Berry, Mitchella repens, in the two forms, viz. long-stamened and short-styled, and short-stamened and long-styled. 236* THE FLOWER. the same part of the proboscis against the low anthers of the short-stamened and the low stigmas of the short-styled form. 426. Moreover, Dar- win has ascertained by microscopical examina- tion that the pollen of the two differs in size or shape, and by experi- ment that it is less active upon its own stigma than upon the other ; indeed, that in many cases (as in some species of Linum) it is quite inactive or impo- tent not only upon its own stigma but upon its own-form stigii'.a, while it is prepotent on the other, and this reciprocally of the two forms.1 Here, then, are flowers structurally hermaphrodite, but functionally as if dioecious, securing all the advantages of the latter, along with the economical advantage that both sorts of individual and every blossom may bear seed. With dioecism only about half the plants could be fruitful. 427. Heterogonous Trimorphism. A threefold heterogonism is known in certain species of a few genera ; and this complica- tion may have certain conceivable advantages over dimorphism. Where seedling dimorphous individuals are few and far between (those multiplying from root would all be alike), there would be an even chance that any two near each other were of the same form and therefore sterile or imperfectly fertile. But if the organization were of three forms, any two of which inter- crossed with perfect fertility, the chances (as Darwin remarks) are two to one that any two plants were of different forms, and therefore by fertilizing each other completely fruitful. 428. The earliest known instance of three forms as to recip- rocal relative length of stamens and pistil is that of Lythrum 1 Impotence of own pollen, either absolute or relative, occurs no less in certain flowers which are not dimorphous, as in Corydalis, some species of Passiflora, &c. On the contrary, many dimorphous flowers are in a certain degree self-fertile, especially in the long-stamened and short-styled form. These subjects are physiological, and belong to another volume. FIG. 468. Long-styled flower of Fig. 467. laid open, the same laid open Both equally enlarged. 469. Long-stamened flower of ADAPTATIONS FOR INTERCROSSING. 237 Salicaria. This was indicated by Vaucher in 1841, more par- ticularly described by Wirtgen in 1848, but was interpreted by Darwin, and the more recondite differences brought to notice, in 1864. 1 " The three forms may be conveniently called, from the unequal length of their pistils, the Iong-st3'led, mid-styled, and short-styled. The stamens also are of unequal lengths, and these may be called the longest, mid-length, and shortest." The pollen of the different classes of stamens is of two sorts as to color, and of three as to size, the largest grains from the largest stamens. " The pistil in each form differs from that in either of the other forms, and in each there are two sets of stamens, different in appearance and func- tion. But one set of stamens in each form corresponds with a set in one of the other two forms. Altogether, this one species includes three females or female organs, and three sets of male organs, all as distinct from one another as if they belonged to different species ; and. if smaller functional differences are considered, there are five distinct sets of males. Two of the three hermaphrodites must coexist, and pollen must be carried by insects reciprocally from one to the other, in order that either of the two should be fully fertile ; but, unless all three forms coexist, two sets of stamens will be wasted, and the organization of the species as a whole will be incomplete. On the other hand, when all three hermaphrodites coexist, and pollen is carried from one to the other, the scheme is perfect : there is no waste of pollen and no 1 In an article On the Sexual Relations of the Three Forms of Ly thrum Salicaria, in Jour. Linn. Soc. viii. 169. Also on the Character and Hybrid- like Nature of the Offspring of the Illegitimate Unions of Dimorphic and Trimorphic Plants. Ibid. x. 393, 1868. Reproduced and extended in his volume entitled " Forms of Flowers," 1877. FIG. 470. Diagram of the flowers of the three forms of Lythrum Salicaria, in their natural position, with the petals and calyx removed on the near side. The dotted lines with the arrows show the directions in which pollen must be carried to each stigma to ensure full fertility. (From Darwin.) 238 THE FLOWER. false coadaptation." The whole arrangement is displayed in the annexed diagram (Fig. 470), and in the following account of the operation.1 " In a state of nature, the flowers are inces- santly visited for their nectar by hive and other bees, various Diptera, and Lepidoptera. The nectar is secreted all round the base of the ovarium ; but a passage is formed along the tipper and inner side of the flower by the lateral deflection (not repre- sented in the diagram) of the basal portions of the filaments ; so that insects invariably alight on the projecting stamens and pistil and insert their proboscides along the upper and inner margin of the corolla. We can now see why the ends of the stamens with their anthers and the end of the pistil with the stigma are a little upturned, so that they may be brushed by the lower hair}' surfaces of the insects' bodies. The shortest stamens, which lie enclosed within the calyx of the long- and mid-styled forms can be touched only by the proboscis and narrow chin of a bee : hence they have their ends more upturned, and .they are graduated in length, so as to fall into a narrow file, sure to be raked by the thin intruding proboscis. The anthers of the longer stamens stand laterally farther apart and are more nearly on the same level, for they have to brush against the whole breadth of the insect's body. . . Now I have found no exception to the rule that, when the stamens and pistil are bent, they bend to that side of the flower which secretes nectar. . . . When nectar is secreted on all sides, they bend to that side where the structure of the flower allows the easiest access to it, as in Lythrum. ... In each of the three forms, two sets of sta- mens correspond in length with the pistil in the other two forms. When bees suck the flowers, the anthers of the longest stamens, bearing the green pollen, are rubbed against the abdomen and the inner sides of the hind legs, as is likewise the stigma of the long-styled form. The anthers of the mid-length stamens and the stigma of the mid-styled form are rubbed against the under side of the thorax and between the front pair of legs. And, lastly, the anthers of the shortest stamens and the stigma of the short-styled form are rubbed against the proboscis and chin ; for the bees in sucking the flowers insert only the front part of their heads into the flower. On catching bees, I observed much green pollen on the inner sides of the hind legs and on the abdomen, and much yellow pollen on the under side of the thorax. There was also pollen on the chin, and, it may be presumed, on the proboscis, but this was difficult to observe. I had, however, 1 All from Darwin, Forms of Flowers, 137-147, &c. ADAPTATIONS FOB INTERCROSSING. 239 independent proof that pollen is carried on the proboscis ; for a small branch of a protected short-styled plant (which produced spontaneously only two capsules) was accidentally left during several days pressing against the net, and bees were seen insert- ing their proboscides through the meshes, and in consequence numerous capsules were formed on this one small branch. . . . It must not, however, be supposed that the bees do not get more or less dusted all over with the several kinds of pollen ; for this could be seen to occur with the green pollen from the longest stamens. . . . Hence insects, and chiefly bees, act both as general carriers of pollen, and as special carriers of the right sort." 429. Finall}', a long series of experiments (requiring eighteen distinct kinds of union) proved that both kinds of pollen are nearl}- or quite impotent upon the stigma of the same flower, and that no ovary is fully fertilizable by other than a " legitimate union," i. e. by stamens of the corresponding length ; but that the mid-length pistil is more prolific than either of the others under illegitimate union of either kind ; which might perhaps be expected, as the pollen proper to it is intermediate in size of grains between that of the long and that of the shortest stamens. 430. Nesaea verticillata, a common Lythraceous plant of the Atlantic United States, is similarly trimorphous, but has not yet been particularly investigated. Several South African and American species of Oxalis are equally trimorphous, and have been investigated by Darwin and Hildebrand,1 with results quite as decisive as in Lythrum Salicaria. One genus of Monocotyledons has trimorphous blossoms, viz. Pontederia, of which the North American P. cordata is a good illustration.2 431. All known flowers exhibiting reciprocal dimorphism or trimorphism are entomophilous : no such wind-fertilized species is known. Few of them are irregular, and none very irregular : they do not occur, for instance, in Leguminosae, Labiatse, 1 Monatsber. Akad. Berlin, 1866; Bot. Zeit. 1871, &c. According to Darwin, Fritz Mueller " has seen in Brazil a large field, many acres in extent, covered with the red blossoms of one form [of an Oxalis] alone, and these did not produce a single seed. His own land is covered with the short-styled form of another species, and this is equally sterile; but, when the three forms were planted near together in his garden, they seeded freely." Forms of Flowers, 180. 2. Detected by W. H. Leggett. See Bulletin of Torrey Bot. Club, vi. 62, 170; and for the original discovery in Brazilian species, by Fritz Mueller, see Darwin's Forms of Flowers, 183, &c. Pontederia has three lengths of style and counterpart stamens, as in Lythrum Salicaria, each flower having two sets of stamens, three in each set. 240 THE FLOWER. Scrophulariacese, Orchidacese, &c. Nature is not prodigal, and does not endow with needless adaptations flowers which are otherwise provided for. § 3. ADAPTATIONS FOR CLOSE FERTILIZATION. 432. Even where cross-fertilization in bisexual flowers is obviously arranged for, it is apt to "be tempered with more or less of close- fertilization. The more exquisite the arrangements for the former are, the more completely is the plant dependent upon insect visitation. Failure to intercross is a remote and small evil compared with failure to set seed at all. In order therefore that the plan of cross-fertilization may not defeat even its own end, through too absolute dependence on precarious assistance, some opportunity for self-fertilization will usually be advantageous. Also there is a long array of insect-visited flowers, especially polyandrous ones, in which close fertilization must be much the commoner result, except where the pollen of another but wholly similar flower has greater potency. 433. Subsidiary self-fertilization is secured in a great variety of ways. In Gentiana Andrewsii, which is proterandrous, and usually cross-fertilized by humble-bees entering bodily into the corolla, an exposed surface of pollen long remains fresh upon the ring of anthers girding the base of the style : when the stigmas separate, they remain for some days simply divergent, but they at length become so revolute that the receptive surface is brought into contact with the ring of pollen below. The opening and closing of blossoms by day or night, the growth of style, fila- ments, or corolla after anthesis commences, or other changes of position, may secure a certain amount of self-fertilization in a subsidiary or even in a regular way. Then certain species, such as Chickweed, which blossom through a long season, close- fertilize even in the bud in early spring, when insects are scarce, but are habitually intercrossed by insects in summer. Somewhat similarly, according to Hermann Mueller,1 certain species, such as Euphrasia officinalis and Rhinanthus Crista-galli, habitually produce two kinds of blossoms, one larger and more show}-, usually affecting sunny localities, and with parts adapted to intercrossing bj- insects ; the other smaller or inconspicuous, and with anthers adjusted for giving pollen to the adjacent stigma without aid. There are gradations between these last arrange- ments, and the more special and remarkable one of dimorphism with 1 Befruchtung der Blumen durcli Insekten, 294 ; Nature, viii. 433. ADAPTATIONS FOR CLOSE FERTILIZATION. 241 434. Cleistogramy. Here the intention and the accomplishment of self-fertilization are unmistakable. This peculiar dimorphism consists in the production of very small or inconspicuous and closed flowers, necessarily self-fertilized and fully fertile, in addition to ordinary, conspicuous, and much less fertile, though perfect flowers. Two cases were known to Linnaeus,1 and one of them to Dillenius before him ; those of Viola have long been familiar in the acaulescent species ; Adrien Jussieu made out the structure of the cleistogamous flowers in certain Malpighiaceas in 1832, and recorded in 1843 that Adolphe Brongniart had well investigated those of Specularia, and that Weddel had discov- ered them in Impatiens Nolitangere. A full account of the then known cases was given by Mohl2 in 1863 ; but D. Mueller, of Upsala, who examined Viola canina, is said by Darwin to have given,8 in 1857, " the first full and satisfactory account of any cleistogamic flower." The appropriate name of cleistogamous was given by Kuhn,4 in 1867, and is now in common use. 435. Cleistogamous flowers are now known in about 60 genera, of between twenty and thirty natural orders, of very various relationship, though all but five are Dicotyledons. All but the Grasses5 and Juncus are entomophilous as to the ordinary flowers, and most of these such as have special arrangements for their intercrossing, either by dichogam}', heterogone dimorphism or trimorphism (in Oxalis), or such special contrivances as those of Orchids. 436. It has been said that the ordinary flowers in such plants are sterile, and perhaps they always are so except when cross- fertilized : in most cases they are habitually infertile or spar- ingly fertile. Probably they suffice to secure in every few generations such benefit as a cross may give, while the principal 1 Campanula (now Specularia) perfoliata and Ruellia clandestina, the latter a cleistogamous state of R. tuberosa. Linnaeus did not make out the structure of the flowers, but supposed them to want the stamens. 2 In Bot. Zeitung, xxi. 309. 8 In Bot. Zeitung, xvi. 730. 4 Ibid. xxv. 65. The name ( denoting " closed up " union or fertilization ) has been written cleistogmous, which is not so proper. We prefer cleistogamous to cleistogamic (and so of similar terms), as best harmonizing with the Latin adjective form, both in form of termination and in euphoniously taking the accent upon the antepenult. 5 Amphicarpum (Milium amphicarpon, Pursh) is the earliest recognized cleistogamous Grass, except perhaps Leersia oryzoides. Some species of Sporobolus are like the latter, and Mr. C. G. Pringle has recently detected such flowers concealed at the base of the sheaths in Danthonia. Amer. Jour. Sci. January, 1878, 71. 16 242 THE FLOWER. increase is by cleistogamous self-fertilization, which thus offsets the incidental disadvantage of the former mode. 437. In general, the cleistogamous are like unto the ordinary flowers arrested in development, some arrested in the almost fully formed bud, most at an earlier stage, and in the best marked cases with considerable adaptive modification. In these, " their petals are rudimentary or quite aborted ; their stamens are often reduced in number, with anthers of very small size, containing few pollen-grains, which have remarkably thin transparent coats, and generally emit their tubes while still enclosed within the anther-cells ; and, lastby, the pistil is much reduced in size, with the stigma in some cases hardly at all developed. These flowers do not secrete nectar or emit any odor : from their small size, as well as from the corolla being rudimentary, they are singularly inconspicuous. Consequently, insects do not visit them ; nor, if they did, could thejr find an entrance. Such flowers are therefore invariably self- fertilized ; yet they produce an abundance of seed. In several cases, the young capsules bury themselves beneath the ground, and the seeds are there matured. These flowers are developed before, or after, or simultaneously with the perfect ones." l In Grasses, however, as in some Dicotyledons, there is much less modifica- tion and more transition. For when Leersia half protrudes its panicle, in the usual way, the included half is fertile and the expanded portion sterile (or almost always so), although the flowers may open and exhibit well-developed anthers, ovaries, and stigmas. But when similar panicles remain enclosed in the leaf-sheaths, they are mostly fruitful throughout. 438. Fully to apprehend the economy of cleistogamy in pollen- saving alone, — and contrariwise to estimate the expense of intercrossing, — one should compare the small number of pollen- grains which so completely serve the purpose in a t}Tpical cleis- togamous flower (say 400 in Oxalis Acetosella, 250 in Irnpatiens, 100 in some Violets) with the several thousands of all entomo- philous cross-fertilized flowers, rising to over three and a half millions in the flower of a Peony, also their still greater number in many anemophilous blossoms. To this loss should be added the cost of a corolla and its action, also of the production of odorous material and of nectar. No species is altogether cleis- togamous. Thus cleistogamy, with all its special advantage, testifies to the value of intercrossing. 1 Darwin, Forms of Flowers, 310. PERIANTH, OR FLOWER-LEAVES. 243 SECTION V. THE PERIANTH,1 OR CALYX AND COROLLA IN PARTICULAR. 439. The distribution of the floral leaves around the axis, which belongs to phyllotaxy, and their particular disposition in the bud (aestivation), have already been considered in Chap. IV. Sect. I., II. And most of the morphology of calyx and corolla has been outlined in the preceding sections of the present chap- ter. What remains chiefly relates to particulars of form and to terminology. 440. Duration. The differences in this respect give rise to a few terms, such as the following. Caryx or corolla may be Persistent, not cast off after anthesis, but remaining unwithered until the fruit is formed or matured ; as the calyx in Labiatae, in Physalis, and most Roses. Marcescent, withering or drying without falling away ; as the corolla of Heaths, Drosera, &c. Deciduous, falling after anthesis and before fructification ; as the petals of Roses, the catyx and corolla of Columbine. Ephemeral or Fugacious, lasting for only a day ; as the petals of Poppy, Helianthemum, Purslane, and Spiderwort. In the two former, they are cast or early deciduous, the anthesis lasting but a day : in the two latter, the anthesis is equally or more brief, but the petals deliquesce or decay at once without falling, as does the whole flower of Cereus grandiflorus and other night- blooming Cactaceae. Caducous, falling when the blossom opens ; as the calyx of Poppy and Baneberry. 441. Numerical Terms, succinctly denoting the number of leaves, either of the perianth as a whole, or of an}' one of its circles, are common in descriptive botany. The most general are those which simply specify the number of component leaves, by prefixing Greek numerals to the Greek name of. leaves, ex- pressing them in Latin form, or transferring them to the Eng- lish. Thus Diphyllous, of two leaves (sepals or petals) ; Triphyllous, of three; Tetraphyttous,offour;PentaphyUous,offive; Hexaphyllous, of six, and so on. A tulip and a Tradescantia flower have a hexaphyllous perianth, but composed of two circles, answering to calyx and corolla ; each Triphyllous^ When the character 1 Perianthium, alias Perigone or Perigonium. (296.) 2 As elsewhere explained, when numerical composition is indicated without reference to nature of parts, the terms dimerous, trimerous, tetramerous, penta- tnerous, &c., may be used. 244 THE FLOWEK. of the organ, i. e. whether calyx or corolla is to be specified, the word sepal "or petal is employed in the combination ; as, Disepalous, of two sepals ; Trisepalous, of three ; Tetrasepalous, of four ; Pentasepalous, of five (also written 5-sepalous, and ac- cordingly 2-sepalous, 3-sepalous) , and so on : also, Dipetalous, Tripetalous, Tttrapetalous, JPentapetcdous (2—5- petalous), &c., when the corolla is concerned. 442. Monophyllous , Monosepalous, and Monopetalous are the proper terms for perianth (calyx, corolla, &c.) composed of a single leaf. Likewise Polyphyllous, Polysepalous, and Polypetal- ous for the case of a considerable but unspecified number of members. Unfortunately, in the Linnaean and long-prevalent use, monopetalous was the term employed to designate a corolla of one piece in the sense, or the fact, of a coalescence or grow- ing together of two, three, five, or more petals into a cup or tube ; and so of a calyx, of a whorl of bracts, &c. And poly- petalous, polysepalous, and polyphyllous were the counterparts of this, meaning of more than one distinct piece, whatever the number. The misleading use, consecrated by long prescription, is not 3'et abandoned, but will in time be obsolete. In present descriptive botany, a poH'phyllous cabyx, or a polypetalous corolla, or a 5-petalous corolla, would be taken to mean that the sepals or petals (as the case may be) were distinct or uncom- bined, and a monopetalous corolla to be one with petals combined by coalescence. (329.) 443. Terms of Union or Separation. The proper term for a corolla or a calyx the leaves of which are more or less coalescent into a cup or tube is Gamopetalous for such a corolla, Gamosepalous for the cabyx ; these terms meaning united petals or sepals. The older and mis- leading names Monopetalous and Monosepalous, although current up to a recent da}r, should be discontinued. Another term is not rarely used in Germany, that of Sympetalous, for the gamo- petalous (of formerl}- monopetalous) corolla, — therefore Syn- sepalous for a similar calyx. It is perhaps a more apt term than gamopetalous, and of the same et3'mological signification ; but the latter is already well in use. Choripetalous is, on the whole, the most fitting name for a corolla the petals of which are separate (as it literally expresses this) , that is, for what is still commonly called Polypetalous, as already explained. (442.) It is adopted by Eichler, &c. Chori- sepalous is the term applied to the cabyx. Dialypetalous (em- ployed by Endlicher) has the same meaning. Both this term and choripetalous carry the implication of separated, rather PERIANTH, OR FLOWER-LEAVES. 245 than of typically separate, parts. EleutJieropetalous (literally free-petalled) has also been used, but is inconveniently long. 444. Degree of coalescence is most correctly expressed by the phrases united (connate, or coherent, or coalescent) at the base, to the middle, or to the summit, as the case may be. But it is more usually and tersely expressed in botanical description by employing terms of division, identical with those used in describ- ing the lobing or toothing of leaves and all plane organs. (184-188.) That is, the calyx or corolla when gamophyllous is for description taken as a whole, and is said to be parted (3-parted, b-parted, &c.), when the sinuses extend almost to the base ; cleft, when about to the middle ; lobed, a general term for any considerable separation beyond toothing ; dentate or toothed (3-toothed, 5-toothed, &c.), when the union extends almost to the summit ; entire, when the union is complete to the summit or border. 445. Parts of Petals, &c. The expanded portion of a petal, like that of a leaf, is the LAMINA or BLADE : any much contracted base is the UNGUIS or CLAW. The latter is very short in a rose- 471 petal, but long and conspicuous in a pink and all flowers of that tribe (Fig. 471), in many Capparideae (Fig. 409) and Cruciferae. A sepal is very rarely distinguishable into lamina and claw. 446. Parts of Gamophyllous Perianth. The coalescent portion of a corolla, calyx, or of a perianth composed of both (such as a Lily or Crocus-blossom) , so far as the sides are parallel or not too spreading, is its TUBE : an expanded terminal portion, either divided or undivided, is the LIMB or BORDER. The limb may FIG. 471. Corolla of Soapwort, of five separate long-clawed or unguiculate petals, with a crown at the junction of claw and blade. FIG. 472. Flower of Gilia coronopifolia ; the parts answering to the claws of the petals of the last figure here all united into a tube. FIG. 473. Flower of the Cypress-Vine (Iponwea Quamoclit) ; the petals a little farther united into a five-lobed spreading border. FIG. 474. Flower of the Ipomnea coccinea ; the five component petals perfectly united into a trumpet-shaped tube, and beyond Into an almost entire spreading border. 246 THE FLOWER. be parted (that is, the component parts not united) quite or nearly down to the tube or base, as in Fig. 472, 475 ; or less so, as in Fig. 473, 476 (with limb 5-lobed) ; or with merely angles or points to represent the tips of the component members, as in Fig. 474 ; or with even and entire border, as in common Morning- Glory, Fig. 482. 447. The line, or sometimes a manifest or conspicuous portion, between the limb and tube (in the corolla always a portion above or at the insertion of the stamens, when these are borne by the corolla) is called the THROAT, in Latin FAUX, pi. fauces. This is mostly more open than the tube, yet less expanded than the limb ; but it often presents insensible gradations from the one to the other. 448. Such appendages as the CORONA or CROWN (385, shown in Fig. 403, 404, 471) usually belong to the throat of a gamo- petalous corolla or perianth, as in Oleander, Comfrey, Borrage, Narcissus, &c., or to a corresponding position when the parts are not coalescent. 449. Forms of Corolla, Calyx, &c. As to terminology, some of these are special and are applicable to corolla only, as the Papilionaceous, the peculiar irregular corolla of the tj-pical .portion of Leguminosse (388, Fig. 342-334), which has been already illustrated, and in which the petals, two pairs and an odd one, take particular names. Also the Caryophyllaceous, or Pink-flower (Fig. 471), a regular corolla, of five long-clawed (unguiculate) petals, the claws enclosed in a tubular calyx and the blades spreading ; and the Cruciferous, of four somewhat similar petals, the four abruptly spreading blades in the form of a cross (cruciate), as in Fig. 394. Rosaceous, with roundish and widely spreading petals on very short or hardly any claws, as in Rose and Apple-blossoms. Liliaceous, a 6-phyllous perianth of campanulate or funnelform shape ; the members either distinct, as in most common lilies and tulips, or gamophyllous, as in Lily of the Valley. All but the first and last of these sorts are examples of regular and chori- petalous perianth. Orchidaceous flowers are of a peculiar irregularity, combining both calyx and corolla : one member, the petal in front of the FIG. 475. Rotate or wheel-shaped and five-parted corolla of the Bittersweet (So- lanum Dulcamara). FIG. 476. Wheel-shaped and five-lobed corolla of the common Potato. PERIANTH, OR FLOWER-LEAVES. 247 stamen and stigma, differs from the rest in shape and in being nectariferous (as in Fig. 460) ; it is named the LABELLUM. Galeate is a term applied to a corolla the upper petal or part of which is arched into the shape of a casque or helmet, called the Galea; as in Aconite (Fig. 357) and Lamium, Fig. 479. ! In the former the galea is of a single petal ; in the latter, it consists of two, completely united. 450. Gamophyllous forms with special names are chiefly the following. Illustrations are usually taken from the corolla, but the forms and terms are not peculiar to it, excepting the first, viz. the Ligulate or Strap-shaped corolla (Fig. 288, &c.), which is / nearly confined to Composite. Here a corolla, formed of three or five petals, imitates a single petal, except at its very base, which is commonly tubular : the remainder is as though the tube had been split down on the upper side and flattened out. The corolla of Lobelia, type of a family most nearly related to Com- positse, illustrates this. (Fig. 488.) 480 481 451. The names of the general forms are mostly taken from some resemblance to common objects. All those in common use will be found in the Glossary : a few leading ones are here specified. They may be divided into the regular and the irregu- lar. The principal irregular form with a special name is the . Labiate, or lipped, also termed Bilabiate, as there are two lips, an upper and a lower (superior and inferior, or anterior and posterior, 290), although one of them is sometimes obscure or T/t abortive. This bilabiate character in the corolla, and often in the calyx also, pervades several orders with gamopetalous flowers, and gives name to one of them, the Labiates, to which FIG. 477. Campanulate corolla of the Harebell, Campanula rotundifolia. 478. Salverform (hypocraterimorphous) corolla of Plilox. 479 Laliiato diligent) corolla of Lunlnm; a side view 480. Personate corolla of Antirrhinum or Snapdragon. 481. Personate corolla of Linaria, spurred (calcarate) at the base. 248 THE FLOWER. the Sage and Mint belong. Such flowers are 5-merous, and have two members specially united to form one lip, and three in the other. The odd sepal being posterior (or next the axis of inflorescence) , and consequently the odd petal anterior, the calyx has its lower lip of two sepals and its upper of three ; while the corolla has its upper lip of two petals and its lower of three. But in Leguminosae, where the calyx is sometimes bilabiate, and where the odd sepal is anterior (or toward the bract), this is reversed, and two sepals or lobes of the calyx form the upper lip and three the lower. A bilabiate corolla is Ringent, that is gaping or open-mouthed, when the throat is freely open, as in Lamium, Fig. 479 ; Personate, or masked, when the throat is closed, more or less, by a projection of the lower lip called the PALATE, as in Antir- rhinum and Linaria, Fig. 480, 481. 452. Of regular forms, there are the following, beginning with that having least tube : Rotate, or Wheel-shaped (Fig. 475, 476), widel}' spreading from the very base, or from a short and inconspicuous tube. Crater if orm, or Saucer-shaped, like rotate except that the broad limb is cupped by some upturning toward the margin. Hypocrateriform, or rather (not to mix Latin and Greek) Hypocraterimorphous, in English Salverform, when a rotate or saucer-shaped limb is raised on a slender tube which does not much enlarge upward ; that is, where a long and narrow tube abruptly expands into a flat or flattish limb, as in Fig. 478. In Fig. 472-474 are seen salverform corollas with somewhat more upwardly dilated (trumpet- shaped) tube. The salver or hypo- craterium, which the name refers to, with a stem or handle beneath, is now to be met with only in old pictures. Tubular, when strictly used, denotes a gamophj-llous perianth with limb inconspicuous in proportion to the tube, as in Trumpet Honeysuckle, or as Fig. 472-474 would be if the limb were much diminished or wanting. But it is some- times used in the sense of having a conspicuous tube. FIG. 482. Calyx and funnelform (infundibuliform) corolla of a common Morning- Glory, Ipomosa purpurea. ANDKCECIUM, OR STAMENS. 249 Infundibuliform, or Funnelform, such as the corolla of common Morning-Glory (Fig. 482), denotes a tube gradually enlarging upward from a narrow base into an expanding border or limb. Campanulate, or Bell-shaped (Fig. 477), denotes a tube of length not more than twice the breadth, moderately expanded almost •from the base, the sides above little divergent. SECTION VI. THE ANDRCECIUM, OR STAMENS IN PARTICULAR. 453. The whole Stamen. For the general character and some of the modifications of the stamens, see the first (301) and por- tions of the succeeding sections of the present chapter. The terms peculiar to these organs, and of common use iii botanical description, were nearly all coined by Linnaeus, and employed as the names of classes in his sexual S3"stem. (672.) The sub- stantive names of those classes which are characterized by the number of stamens, and which were designated by Greek nume- rals prefixed to andria (the Greek word for man being used metaphorically for stamen), are put into adjective form, as follows : Mortandrous, for a flower with a solitaiy stamen ; Diandrons, for a flower with two stamens ; Triandrous, with three ; Tetran- drous, with four ; Pentandrous, with five ; Hexandrous, with six ; Ileptandrous, with seven ; Octandrous, with eight ; Enmandrous, with nine ; Decandrous, with ten ; Dodecandrous, with twelve ; Polyandrous, with a greater or indefinite number, or Icosandrous (meaning twenty-stamened) when a polyandrous flower has the ^ stamens inserted on the calyx, as in the Cherry (Fig. 337), Pear, &c. FIG. 483. Diadelphous stamens (9 and 1) of a Pea. 484. Monadelphous stamens of a Lupine. 485. Monadelphous stamens, &c., of Mallow. FIG. 486. Five syngenesious stamens of a Composita. 487. The same, laid open. 250 THE FLOWER. Didynamous is a term applied to an andrcecium of four sta- mens in two pairs, a longer and a shorter, as in Fig. 361. Tetradynamous is similarly applied to that of six stamens, two of them shorter, in the manner characteristic of Cruciferse, Fig. 396. 454. Terms which denote coalescence of stamens, whether by their filaments or their anthers, are Monadelphous , that is, in one brotherhood, by coalescence of the filaments into a tube, as in the Mallow (Fig. 485), Lupine (Fig. 484), Lobelia (Fig. 488), &c. Diadelphous, in two brotherhoods, by coalescence of the fila- ments into two sets ; sometimes an equal number in each, as in Fumariacese (Fig. 390), sometimes nine in one set and one separate, as in the Pea (Fig. 483) and most Papilionaceas. Triadelphous^ with filaments united in three sets or clusters, as in Hypericum. Pentadelphous, in five sets, as in Linden, Fig. 398, 399. But in general, when the sets are several, without regard to the number the stamens are said to be Polyadelphous. Syngenesious, when the stamens are united by their anthers into a tube or ring ; as in the whole vast order of Composite (Fig. 486, where they are five in number and the filaments dis- tinct), in Cucurbita (Fig. 489, 490, where they are three in number and the filaments partly monadelphous) , and in Lobelia (Fig. 488, where they are also five and the long filaments are mainly monadelphous) . FIG. 488. Flower of Lobelia cardinalis, with tube of corolla divided on one side ; filaments and anthers united into a tube : /. tube of filaments ; a. of anthers. FIG. 489. Male flower of Cucurbita (Squash), with limb of calyx and corolla cut away, to show the stamens, viz., three filaments, separate at base but monadelphous above, and three syngenesious anthers in a kind of head. 490. Stamens of the same, enlarged and the upper part cut away, to show the union. The anthers are sinuous. 491. A detached stamen of the Melon, with loosely sinuous anther. Fig. 492. Stamens and style of a Cypripedium. united into one body or column: a. anthers; st. enlarged sterile stamen; st'uj. the stigma. ANDHCECIUM, OR STAMENS. 251 455. Of terms relating to adnation of stamens, besides the general ones of hypogynous, perigynous, epigynous (332) , and epipetalous, or adnate with corolla, there is the special one of Gynandrous, having stamens borne upon the pistil, as in Orchidaceae. In Cypripedium, the filaments of two stamens, and an enlarged sterile stamen behind, are adnate to a style, while the two anthers are quite free (Fig. 492) ; in the proper Orchis tribe (as in Fig. 460, 461), anther and stigma are consolidated into one mass, and there is no evident style. 456. A complete stamen consists of FILAMENT and ANTHER. The latter is the functionally essential part of the organ, and therefore is wanting only in abortive or sterile stamens. (345, 352, &c.) The filament, being only a stalk or support, may be very short or wholly wanting : then the anther is sessile, just as the blade of a leaf is said to be sessile when there is no petiole. 457. The Filament, although usually slender and stalk-like, assumes a great variety of forms : it is sometimes dilated so as to resemble a petal, except by its bearing an anther ; as in the transition states between the true petals and stamens of Nj-m- phsea, shown in Fig. 318. 458. Such petaloid filaments would indicate that this part of the stamen answered to blade rather than to footstalk, while others would harmonize better with what seems at first sight to be the more natural view, that the filament is the homologue of the petiole, the anther of the blade of a leaf. Remembering that in large numbers of leaves there is no distinction into petiole and lamina or blade, such homologies should not be insisted on. The filament may be variously appendaged by outgrowths. Some of these appendages are very conspicuous, such as the scale of Larrea (Fig. 405), which is on the inside, and the nectariferous hood of Asclopias on the outside ; or there may be a tooth on each margin, as in species of Allium. 459. The Anther, the essential organ of the stamen, contain- ing the pollen, surmounts the filament, when that is present. It normally consists of two cells or lobes, the word cell being here used in the sense of sac. But, as each sac is not rarely divided into two cavities (locelli*), the best technical name for anther-sac is that of THECA. The two thecse, lobes, or cells are commonly connected by a more or less evident and sometimes conspicuous common base or junction, which is mostl}r a pro- longation of the filament, the CONNECTIVUM, or in English CONNECTIVE. 460. For the discharge of the pollen, the cells of a normal anther open at the proper time by a line or chink, usually 252 THE FLOWER. extending from top to bottom (Fig. 493), the suture or line of dehiscence. Commonly this line is lateral or marginal : not rarely it faces forward or backward. In the vast genus Solanum, to which the Potato belongs, in most Ericaceous plants (Fig. 458, 494), in Polygala, and in many other flowers, the anther- cells open only by a hole (foramen or pore} , or at most a short chink, at the tip, through which the pollen has in some way to be discharged. In Vaccinium (Cran- berry, Blueberry, &c.), the pore-bearing tip of the anther-cell is prolonged considerably, often into a slender tube, as in Fig. 340. In the Barberry (Fig. 495) and in most of that family, also in Lauraceas, the whole face of each anther-cell separates by a con- tinuous line, forming a kind of door, which is attached at the top, and turns back, as if on a hinge : in this case, the anthers are said to open by uplifted valves. In the Sassafras and many other plants of the Laurel family, each lobe of the anther opens b}- two smaller valves of the kind, like trap-doors. 461. The attachment of the anther to the filament presents three principal modes, which are connected b^' gradations. These are the Innate (Fig. 495, 496), in which the anther directly continues and corresponds to the apex of the filament, the cells usually dehiscent strictly marginal!}', the lobes or cells not looking or projecting either inward or outward. FIG. 493. A stamen, with its anther, 6, surmounting the filament, a, and opening in the normal manner down the whole length of the outer side of each cell. FIG. 494. Stamen of a Pyrola ; each cell of the anther opening by a terminal ori- fice or pore. FIG. 495. Stamen of a Barberry ; the cells of the anther opening each by an up- lifted valve. FIG. 496. A stamen of Isopyrum biternatum. with innate anther. 497. Stamen of Liriodendron, or Tulip-tree, with adnnte extrorse anther. 498. Stamen of OSnothera glanca, with the anther fixed by its middle and versatile FIG. 499. A stamen of Asarum Canadense, with adriate anther and prolonged tip to connective. ANDRCECIUM, OB STAMENS. 253 Adnate, in which the connective appears to be a direct con- tinuation of the filament, having the anther adherent to the anterior or posterior face of it, and the lines of dehiscence therefore looking inward or outward. Magnolia, Linodendron (Fig. 497), and Asarum (Fig. 499) furnish good examples ; the latter conspicuously so, on account of a prominent prolongation of the connective or tip of the filament. Versatile, when the anther is attached at some part only of its back or front to the tip of the filament, on which in anthesis it lightly swings ; as in Plantain, in all Grasses, the Lily, Evening Primrose (Fig. 498), &c. 462. The direction to which an anther faces, whether inward (toward the centre of the flower), or outward (toward the peri- anth) , has to be considered ; except in the case of an innate anther with strictly lateral or marginal dehiscence. An anther is Extrorse, i. e. turned outward, or Posticous, when it faces to- ward the perianth, as in Magnolia and Liriodendron (Fig. 497), Asaruin (Fig. 499), and Iris ; these all being cases of adnate and extrorse anthers, the cells attached for their whole length to the outside of the summit of the filament or the connective. Introrse, i. e. turned inward, or Anticous, when it faces toward the axis of the flower ; as in Nymphaeaceae (Fig. 318), in Violet and Lobelia (which are adnate and introrse), and in OEnothera. In the common Evening Primroses (as in Fig. 498) the anther is fixed near the middle, introrse, and versatile. 463. The direction in which the anther may be said to face, outward or inward, depends upon two characters, which do not always coincide, viz. the insertion or attachment of the cells, and the position of their line of dehiscence. In such a strongly characterized adnate anther as that of Liriodendron (Fig. 497), both the attachment and the dehiscence are plainly posticous or extrorse : in most species of Trillium, the cells are introrse as to attachment, but some are nearly marginal and some are even rather extrorse as to dehiscence : in the related Medeola, and in Lilium, where the anthers are extrorsely affixed toward the base or middle to a slender tip of the filament, the dehiscence is either introrse or nearly marginal. Parnassia is in similar case ; the anthers being clearly extrorse as to insertion and more or less introrse as to dehiscence. 464. Adnate anthers are perhaps as frequently extrorse as introrse. Others, whether basi fixed or medifixed, are more com- monly introrse. Those fixed by the middle, or at any other part of the back, and lying on the inner side of the filament, arc said to be Incumbent. 254 THE FLOWER. 4G5. The connective may be appendaged either by a prolon- gation or otherwise from the tip (as in Fig. 499), or from the back, as in Violets and in many Ericaceous plants. 4G6. The normal anther is two-celled, bilocidar, or (to use a less common term) dithecous, and its lobes or cells parallel, right and left ; but the cells at first, and sometimes at maturity, are bilocellate, that is each is divided into two by a partition which stretches from the connective to the suture or line of dehiscence. In an innate anther, and in many others, this line of dehiscence is marginal or lateral, either strictly or wo »i son' nearly so, as in Fig. 500. When introrse or extrorse (as in Fig. 501, 502), the sutures may still be considered to represent the margins turned inward or outward. The pollen is accordingly pro- duced in four cavities or separate portions of the interior. But -he two locelli on the same side of the midrib or connective (right and left) are usually confluent into one pollen-filled cavity or cell at maturity if not earlier, or at least the partition between them breaks up at dehiscence. Sometimes it remains, and, the groove at the sutures being deep, the anther is strongly four- lobed or quadrilocular at maturity, as in Menisi>ernunn (Fig. 504) ; but morphologically this is still only bilocular (dithecous) although quadrilocellate, and the anther opens at the sutures and through these partitions. 467. A stamen being the homologue of a leaf, the natural supposition is that the anther is homologous with the blade or an apical portion of the blade, therefore the two lobes or thecoj with the right and left halves of it, the intervening connective with the midrib, and the line of dehiscence with the leaf-mar- gins.1 This conception is exemplified by the accompanying 1 This is the view long ago taken by Cassini and Keeper, and it may still IK? maintained as the best morphological conception. Mohl interposed some objections to its universality ; but, as presented in Sachs's Text-Book, they arc not incompatible with the common morphology. Sachs takes the fila- ment with the connective to be the homologue of the whole leaf, and the anther-cells as appendages. Others, in likening the anthers to glands, adopt a similar view. FIO. 600. Innate anther, same as Fig. 496. In younger state, with transverse section, showing the four locelli. 601. Same of an adnate extrorse anther, such as Fig. 497. '•"•-' Same as the preceding but mature and dehiscent, the two locelli becoming one cell l>y the vanishing or breaking up of the partition. ANDKCEC'IUM, OB STAMENS. 255 diagram, Fig. 503, which should, however, show the median partitions in the cross-section, or traces of them. Pollen is a special development into peculiar cells of what would be parenchyma in a leaf. Its formation normally begins in four places, which may remain separate up to maturity, or the two on each side of the axis or connective may early be confluent into one cell. 4C8. Of the man}' deviations of the typical two- celled anther, with its cells parallel and united longitudinally by a connective, the simplest and commonest is that in which (as in Fig. 505) the two cells diverge below and remain united only at their apex. Next, the two cells may, in their early development, become confluent at the apex, as in the Mallow family (Fig. 506), so as to form a continuous pollin-' iferous cavity within, opening by a continuous suture round the margin : here the anther is unilocular or one-celled by confluence. In another way, the anthers of some species of Orthocarpus (gep eralty resembling Fig. 505, but the lobes or cells quite distinct or even separated at apex) lose one of the cells by partial o»- complete non-development and so become one-celled by abor- tion. The anther of Gomphrena (Fig. 507) is completely un» locular by abortion or suppression of the companion cell. Thu-., losing one half, it is said to be dimidiate, or halved. 469. The two anther-cells, such as those of Fig. 505, some- times diverge so much that they form a straight line transverse to the filament, as in Monarda (Fig. 508) , in which their contiguous ends so coalesce as to give the appearance of a one-celled anther fixed by the middle. Or, again, the two cells may be separated by the enlargement of the connective between them, as in Cal- amintha, Fig. 509. This enlargement is extreme in the great genus Salvia, in which a very long and narrow connec- tive gives the appearance of a filament astride the apex of the FIG. 503. Diagram to illustrate the morphology of the stamen, on the Idea that the anther answers to leaf-blade: the lower portion being filament and a part of the anther, in section, the upper a part of a leaf. FIG. 604. Stamen of Menispennum Canadense, the qnadrilorellate anther divided. FIG. 505. Stamen of Pentstemon pnbescens, with anther-cells divergent. FIG. 506. Stamen of Mallow (one of the cluster of Fig. 485), the two cells and suture* confluent into one. FIG. 507. Anther «f Gomphrena or Globe Amaranth, medifixed, of a single cell, dehiscent. 256 THi; FLOWER. proper filament, and bearing an anther-cell at each end. In a few species, the two anther-cells are nearly alike ; in more, the lower one is imperfect, as in Fig. 510° ; in more, it is abortive or wanting altogether, as in Fig. 510*. Then, in the related Californian genus Audibertia, the lower half of this connective is reduced to a short tail, as shown in Fig. 511°, or even in most of the species to so minute a vestige that, except for these transitions, the stamen might be supposed to consist of a simple filament, with an interruption like a splice in the middle, and surmounted by a one-celled anther, as shown in Fig. oil6. In Rosemary, the continuitj- is complete, although a minute reflexed tooth sometimes indicates the junction. 470. Pollen, the product of the anther, is usually a powdery substance, which when magnified is seen to consist of separate grains, of definite size and shape, uniform in the same plant, but often very different in different species or families. The grains are commonly single cells, globular or oval in shape, and of a yellow color. But in Spiderwort they are oblong ; in the FIG. 508-511. Anthers, with upper part of filament, of several Labiatse. 508. Of Monarda. 509. Of a Calamintha. 510 Of two species of Salvia. with long and slender connective, the upper fork of which tears one anther-cell ; the lower in a (from Salvia Texana), bearing the other cell in an imperfect condition ; in 6 (from S. coccinea), bear- ing none at all. 511. a. Same of Audibertia grandiflora, the lower fork of the connec- tive reduced to a naked spur ; 6, from A. stachyoides, in which this lower fork is nearly wanting, and the upper is in a straight line with the filament which it seems to continue. FIG. 512-515. Forms of pollen : 512, from Mimulus moschatus ; 513, Sicyos ; 514, Kchinocystis; 515, Hibiscus. POLLEN. 257 Cichory and Thistle tribes, many-sided ; in the Musk-plant, spirally grooved ; in the Mallow family and the Squash and Pumpkin, beset with briskly projections, &c. The pollen of Pine, as well as that of the Onagraceae, is not so simple, but appears to consist of three or four blended cells ; that of most Ericaceae evidently consists of four grains or cells united. (Fig. 512-521.) The most extraordinary shape is that of Zostera, or the Eel-grass of salt-water, in which the grains (destitute of the outer coat) consist of long and slender threads, which, as they lie side by side in the anther, resemble a skein of silk. 471. Pollen-grains are usually formed in fours, by the division of the living contents of mother cells first into two, and these again into two parts, which become specialized cells. As the pollen completes its growth, the walls of the mother cells are usually obliterated. But sometimes these cells persist, either as shreds, forming the cobweb-like threads mixed with the pollen of Evening Primrose, or as a kind of tissue combining the pollen into coherent masses, of various consistence. Of this kind are the elastically coherent pollen-masses (or POLLINIA, sing. POLLINIUM) of Orchises (Fig. 463), and the denser waxy ones of many other orchids and those of Asclepias or Milkweed, Fig. 522. 472. A pollen-grain has two coats. The outer coat is com- paratively thick, and often granular or fleshy. This is later formed than the inner, and by a kind of secretion from it : to it all the markings belong. The inner coat, which is the proper cell-wall, is a very thin, delicate, transparent and colorless mem- brane, of considerable strength for its thickness. The pollen of Zostera and of some other aquatic plants is destitute of the outer coat. 473. The cavity enclosed by the coats is filled with a viscid- substance, which often appears slightly turbid under the higher powers of ordinary microscopes, and, when submitted to a mag- FIG. 516-521. Forms of pollen : 516, Lily ; 517, Cichory ; 518, Pine ; 519, Circaea ; 520, Kalmia; 521, Evening Primrose. FIG. 522. A pair of pollinia of Asclepias, annexed by their caudicles to the gland. 17 258 THE FLOWER. nifying power of about three hundred diameters, is found to contain a multitude of minute particles ( fovillce) , the larger of which are from one four-thousandth to one five-thousandth of an inch in length, and the smaller only one fourth or one sixth of this size. When wetted, the grains of pollen promptly imbibe water by endosmosis, and are distended, changing their shape somewhat, and obliterating the longitudinal folds, one or more in number, which many grains exhibit in the dry state. Soon the more extensible and elastic inner coat inclines to force its wa}' through the weaker parts of the outer, especially at one or more thin points or pores ; sometimes forming projections, when the absorption is slow and the exterior coating tough. In many kinds of pollen, the grains, when immersed in water, soon distend to bursting, discharging the contents.1 474. Pollen-tubes. In others, and in most fresh pollen, when placed in ordinarily aerated water, at least when this is slightly thickened by syrup or the like, and submitted to a congenial tem- perature, a projection of the inner coat through the outer appears at some one point, and by a kind of germination grows into a slender tube, which may even attain two or three hundred times the diameter of the grain ; and the richer protoplasmic contents tend to accumulate at the farther and somewhat enlarging ex- tremity of this pollen-tube.2 475. In cleistogamous flowers (434), the pollen, while still in the anther, sends out its tubes, which may grow to a great length, in the mere moisture of the flower-bud, the growing tip always directing itself toward the stigma in a wonderful way. Similarly, in the open flower of Milkweeds, the pollen-tubes sometimes start from the pollen-mass even while yet in the anther, and in vast numbers, forming a tuft or skein of pollen-tubes, which may attain considerable length and direct itself toward the some- what distant stigma. Commonly, however, the pollen remains 1 In Conifers, the grains of pollen have a peculiar internal structure or rather a development (suggestive of a horaology with the microspores of some of the higher Cryptogamia), the contents at or before maturity undergoing division into two or three internal cells, only one of which acts in fertiliza- tion. When they act upon the ovule or are placed in water, and the inner coat swells by absorption, the bursting outer coat is commonly thrown off. In Pines and Firs (but not in Larch and Hemlock Spruce), the grain of pollen is singularly compound, consisting (as in Fig. 518) of a central arcuate body (the proper pollen-cell) bearing at each end an empty roundish cell. These are vesicular protrusions or appendages of the proper pollen-grain, of no known functional importance, except that they render such wind-dis- persed pollen more buoyant for transportation. 2 Van Tieghem, in Ann. Sci. Nat. ser. 5, xii. 312, &c., 1869. GYNCECIUM IN ANGIOSPEKMS. 259 unaltered until it is placed upon the stigma. The more or less viscid moisture of this incites a sim- ilar growth, and also doubtless nour- ishes it ; and the protruding tube at once penetrates the stigma, and by glid- ing between its loose cells buries itself in the tissue of the style, descending thence to the interior of the ovary and at length to the ovules. Fertilization is accomplished by the action of this pollen-tube upon the ovule, and upon a special formation within it. Consequent upon this an embryo is formed ; and the ovule now becomes a seed. SECTION VII. THE PISTILS, OR GYN 315> 316' 528' 531-533), and the stigma more or less oblique or even wholly lateral. The continuation of 531 532 533 the latter down the whole length of the ventral side of the style (as in Fig. 528, and also Fig. 549) is not uncommon. In Schizandra (Fig. 531) it is continued downward on the ventral edge of the ovary as far as to its middle.2 1 Abnormal specimens of Podophyllum peltatum are occasionally found having a gynceeium of from two to six separate carpels. 2 Pleurogyne, a Gentianaceous genus so named on this account, has no style nor apical stigma whatever, but has a long stigma extending down the outside of each ovuliferous suture of its dicarpellary ovary for most of its length. FIG. 530. Single simple pistil of Podophyllum, cut across to show the placenta, &c. FIG. 531. Vertical section of a pistil of Schizandra coccinea; a side view showing the stigma decurrent down to the middle of the ovary. 532. Pistil of Hydrastia ; ventral view. 533. Pistil of Actaea rubra, cut across, so as to show the interior of the ovary; ventral view. GYNCECIUM IN ANGIOSPEEMS. 263 487. As the placenta of a simple pistil belongs to the two united margins of the carpellary leaf, there is naturalty a double row of ovules, one to each margin. If the leaf- margins which are turned inward in the ovary be- low to bear the ovules are turned outward above to receive the pollen (see Fig. 531), then the typical stigma should also be double or bilamellar. So it is seen to be in such carpels as those of Fig. 528, 531—533, and indeed in very many stigmas of this class. Such division, or even a greater bifurcation of a monocarpellary stigma into two lobes or half- stigmas, is not anomalous. 488. The ovary of a simple pistil should be unilocular, that is, should have a single cavity or cell (loculus) , although, as will soon be seen, the converse does not hold true. Yet this cell in certain instances becomes bilocellate, being divided by a growth or intrusion from the back into two locelli. This occurs more or less in the larger number of species of the Leguminous genus m Astragalus, and the mode is shown in Fig. 534. 489. Compound or Syncarpous Pistil.1 This consists of two. three, or a greater number of carpels coalescent into one bod}-. A true compound pistil represents a whorl (in the simplest case a pair) of carpels united into one body, at least as to the ovary. 490. The coalescence of a capitate or spicate mass of carpels or simple pistils of the same flower, imbricately heaped on the torus, as in Magnolia (Fig. 648) and Liriodendron, cannot property be said to form a compound pistil. This heap of pistils may be called a SOREMA. 491. Morphologically, a compound pistil, as to the ovary, may be a pair or a circle of closed carpels or simple pistils brought into contact, and the contiguous parts united : this is illustrated in Fig. 535-538. Or it may be formed of a whorl of open car- pellary leaves, joined each to each by the contiguous margins, 1 The terms apocarpous and syncarpous for pistils, the first of separate, the second of combined carpels, were introduced by Lindley. They have little advantage over the terms simple and compound. Moreover, the word syncarp or syncarpium had been appropriated to a sort of fruit of the class now called multiple, formed by the coalescence of several flowers, and also to that of a heap, head, or spike of carpels more or less cohering at matu- rity, as in a blackberry, or confluent in the flower, as in Magnolia. FIG. 534. Ovary or forming legume of Astragalus Canadensis. transversely divided, to show the false partition which, intruded from the back, divides the simple cell into two half-cells or locelli. 264 THE FLOWER. in the manner of Fig. 542-545. Between these two there is every gradation. The first forms a compound ovary, 492. With two or more Cells and Axile Placenta;. For it is evident that, if the contiguous parts of a whorl of two or more closed carpels cohere, the resulting compound ovary should have as many cells as there are carpels in its composition, and that the placentae (one in the inner angle of each carpel) will all be brought together in the axis of the compound pistil. And the partitions, termed DISSEPIMENTS, which divide the compound ovary into cells, manifestly consist of the united contiguous por- tions of the walls of the carpels. These necessarily are composed of two layers, one belonging to each carpel ; and in fruit they often split into the two layers. True dissepiments and the true cells must accordingly be equal in number to the carpels of which the compound pistil is composed. That is, the ovary, or the resulting fruit, is bilocular or 2-celled, trilocular or 3-celled, quadrilocular or 4-celled, and so on, according to the number of dissepiments or cells. 493. There may also be false dissepiments, mostly of the same character as that which in Fig. 534 divides the cell of a single carpel. Such are found in Flax (Fig. 539-541) , in Amelanchier or Service-berry, in Huckleberry (Gaylussacia) , and in most of FIG. 535. Pistil of a Saxifrage composed of two carpels or simple pistils united below, but distinct above ; cut across both above and below. FIG. 536. Pistil of common St. Johnswort, of three united ovaries; their styles distinct. FIG. 537. The same of another species of St. Johnswort (Hypericum prolificum), the styles also united into one, which, however, may split apart in the fruit. FIG. 538. Pistil of Tradescantia or Spiderwort, even the three stigmas united into one. The ovary in all cut across to show the internal structure. GYNCECIUM IN ANGIOSPERMS. 265 the American species of Vaccinium. In all these, the false par- tition is a growth from the middle of the back of each carpel, which divides its cell more or less completely into two. 494. On the other hand, even the true dissepiments which belong to such a compound ovary may be abortive or evanescent, the placentae remaining in the axis combined into a column. (499.) The second modification of the compound pistil (491) normally has an ovary, 495. With one Cell and Parietal Placentae. That is, the placentae are borne (as the term denotes) on the wall or parietes of the ovary, as in the Poppy, Violet, Sundew, Cistus or Helianthemum (Fig. 543), Cleome, Gen- tian, and in all or most of the orders from which these examples are cited. The diagram Fig. 542 illustrates the morphological conception of a com- pound pistil of this kind. Not that it is ever sup- posed to be formed by the actual combination of once fid separate leaves, any more than a gamophyllous calyx or corolla is actually so produced. The conception in all such cases is that FIG. 539. Transverse diagrammatic section of a flower of the common Flax, show- ing the ovary with false partitions extending one from the back of each cell. 540. Sec- tion of a mature fruit and seeds of the same, the false partitions now complete, divid- ing the five cells into ten, each one-seeded. 541. Same of a wild Flax (Linum perenne), in which the false partitions remain incomplete. > FIG. 542. Plan of a one-celled ovary with three parietal placentae, cut across be- low ; the upper part showing the top of the three leaves it is theoretically composed of, approaching, but not united. FIG. 543. Ovary of Helianthemum Canadense, cut across, showing the ovules on three parietal placentae. FIG. 544. Transverse section of the ovary of Hypericum trraveolens ; the three large placentas meeting in the centre, but not cohering. 545. Similar section of a ripe capsule of the same ; the placentae now evidently parietal 266 THE FLOWER. of a congenital development of organs in union which, in the development of a vegetative shoot, would be leaves. This case is represented by the combination of open carpellary leaves, as the preceding one is by that of closed ones. As the edges of the leaves must needs be turned in, to bear the ovules, a compound ovary with parietal placentation may be likened to the unopened calyx of a Clematis, as shown in Fig. 256, 257. Eveiy gradation is found between axile and parietal placentation. Sometimes the placentae are strictly on the pari- etes or wall (Fig. 543, 547) ; sometimes borne inwards on incomplete dissepiments (Fig. 548) ; and sometimes they are brought firmly together in the axis, as in Fig. 544, though sepa- rable, and indeed separated in the fruiting stage. 496. A compound ovar}- with parietal placentae is necessarily one-celled (imilocular) ; except it be divided by an anomalous partition, such as is found in Cruciferae (Fig. 395) and in many Bignoniaceae. 497. Normal placentae are necessarily double : when parietal, the two halves belong to different leaves ; when axile, to the same leaf. These two halves may diverge or be widely separated. sometimes even at their origin, as in Aphyllon and some other Orobanchaceae, in which a dicarpellary ovary has four almost equidistant placentae ; or in such cases the placentae may be regarded as intra-marginal instead of marginal. 498. The placentae of a two-several-celled ovary, such as in Fig. 536, 537, &c., may be described in the plural number, being one in each carpel ; or when consolidated into a central column, and well covered with ovules, they may be said to form one (compound) placenta. Then when the dissepiments early disappear, or are abortive from the first, the result is a compound ovary of this class, 499. With one Cell and Free Central Placenta. In Caryo- phyllaceae (Fig. 549, 550) and Portulacaceae, this evidently results from the obliteration of the dissepiments (as many as there are styles or stigmas) , vestiges of which may be sometimes FIG. 546. Diagram (ground-plan) to illustrate free central placentation produced by abortion of dissepiments. 547. Same of strict parietal placentation. 548. Same with the placentae carried inward on imperfect dissepiments. GYNCEC1UM IN ANGIOSPERMS. 267 detected, while certain plants of the same families, of otherwise identical structure, retain the dissepiments even in the fruit. 500. But a similar condition may equally arise from a modification of parietal placentation, namely, with the margins of the leaves ovuliferous only at bottom, and the placentae there conspicuously devel- oped and completely united. The basal placenta- tion of Dionoea is unavoidably so explained, its nearest relative, Drosera (Fig. 553), having parietal placentas. And this leads to a probable explanation of the case in Primulacea?, where a large free central placenta fills the centre of the cell, and no trace of dissepiments can be detected.1 501. The idea maintained in former editions is still adhered to ; namely, that placentas belong to carpels and not to the cauline axis, in other words, that ovules are productions of and borne upon leaves, usually upon their margins, not very rarely upon other portions of their upper surface, rarely over the whole of it.2 502. Ovules cover the whole internal face of the carpels in Butomus and its relatives, also of the Water-Lilies (both Nym phaea and Nuphar, Fig. 551) excepting the inner angle, to whicU they are usually restricted in other plants. And in the allied Brasenia and Cabomba, where the ovules are reduced to two or three, one or more of them is on the midrib, but none on the 1 The placenta in this and like cases is rather to be regarded as an out- growth from the base of the carpellary leaves, combined over the floral axis. Upon this interpretation, a central portion of the column may be (and sometimes must be) of axile nature, yet the ovules be borne upon foliar parts. See Van Tieghem, in Ann. Sci. Nat. ser. 5, xii. 329 (1869) ; Celakowsky, Vergleichende Darstellung der Placenten, &c. (1876) ; Warming, in Ann. Sci. Nat. ser. 6, v. 192. 2 This view was first maintained as a general theory, and on critical grounds, by Brown, in Planta? Javanicse Kariores, 107-112. Schleiden, End- licher, and others took the opposite view, i. e., that ovules are productions of the axis, even in parietal placentation, — an exceedingly far-fetched suppo- sition. In later days, the commoner view has regarded ovules as of both origins, as productions of the carpels in parietal, of the axis in at least some free central or basilar placentation. But at present the theory of foliar origin without exception, revindicated by Van Tieghem, and espe- cially by Celakowsky and Warming, again prevails. For the bibliography and an abstract of the various views, see Eichler, Bliithendiagramme, espe- cially the note in the preface to the second part (where lie gives his entire adhesion to this conclusion) ; also Warming's memoir, De 1'Ovule, in Ann. Sci. Nat. ser. 6, v. 1877-78. FIG. 549. Vertical section through the compound tricarpellary ovary of Spergularia rubra, showing the free central placenta. 550. Transverse section of the same. 268 THE FLOWER. margins of the carpellary leaf. In many species of Gentian, as also in Obolaria and Bartonia, of the same family, the whole internal face of a dicar- pellary ovary is thickly ovuliferous. 503. Perhaps the parietal placentae in Parnassia (Fig. 552) are borne on the midribs of the carpels, for they are directly under the stigmas, instead of alternate with them, as they normally should be. The same thing occurs in Poppies and man}' other Papaveraceae, also in some Cruciferte ; and in some of the cases each stigma is more or less two-lobed. This sug- gests the explanation,1 here probably the true one, which supposes that the placentae are borne on the leaf-margins in the normal way, but that each stigma is two- parted (as if the carpel- lary leaf were deeply notched at the apex, and so its two stigmatic leaf- margins separate, as Drosera illus- trates, Fig. 553), and that the two half-stigmas of adjacent carpels have coalesced into one body, which would of course stand over the parietal placentae beneath. Each stigma in such a case, as well as each parietal placenta, would consist of the united margins of two adjacent carpels. § 2. IN GYMNOSPERMS. 504. GYMNOSPERMOTJS (that is, naked-seeded) plants are so named because the ovules, or bodies which are to become seeds, are fertilized by direct application of the pollen, which reaches and acts upon the nucleus of the ovule itself, not through the mediation of stigma and style. In the structure of their flowers, these plants are of a low or simplified type, in some respects not obviously homologous with the Angiosperms which now consti- tute the immense majority of phaenogamous plants. But, up to a comparatively late geological period, Gymnosperms appear to have been the only flower-bearing plants. They are represented 1 Given by Brown, in the Plantae Javanicae Rariores, above referred to. FIG. 551. Transverse section of an ovary of Nymphaea odorata, the carpels ovulifer- ous over the whole interior surface. FIG. 552. Pistil of Parnassia, with ovary transversely divided. FIG. 553. Pistil of Drosera filiformis, with ovary transversely divided. GYNOSCIUM OF GYMNOSPERMS. 269 in the extant vegetable kingdom b}- three (or four) groups or orders, two of them small, and one comparatively ample and of wide distribution ; and all are so strikingly different from each other that they cannot be illustrated b}r a common description. The largest order, Coniferae, is familiar, and contains a good share of the most important forest trees of temperate climates. The smallest, Gnetaceae, chiefly tropical or of warm regions, lies between Gymnosperms and common Dicotyledons. The third, Cycadaceae, is most remote from them, and as much so from Monocotyledons, except that it imitates Palms, as it also does the Tree-Ferns, in habit, both as to stem and foli- age. The particular morphology of Gymnosperms would re- quire for its illustration copious details and the history of various conflicting hypotheses. It must be relegated to the special morphology of the natural orders, premising, however, a brief sketch of the general floral structure.1 505. In Gnetacese, Gymnosperms and Angiosperms almost come together. The flowers have a perianth (diphyllous or tetr aphyllous) ; the stamens have a distinct filament and anther ; and the gynrecium is a sac (presumably of two carpophylls) open at the top and filled at bottom by a single ovule of the simplest kind, i. e. consisting of a nucleus destitute of coats. This pistillary body is attenuated and prolonged above the ovule into a style-shaped tube, with open and commonly two-cleft orifice. In the almost hermaphrodite sterile flower of Wei witschia, this takes the form of a much dilated stigma, which is even beset with seeming stigmatic papillae. If only the pollen were here to grow forth into pollen-tubes (with or without a closing of the tube) , angiospermy would be attained. But, in fact, the pollen- grains bodily reach the ovule itself through the tube, fertilizing it directly.'2 This interesting group of plants consists of the 1 References to the literature of gymnospermy and to the steps of the prolonged controversy over it, also the points of morphology still in part unsettled, need not here be given. The history and the idea of gymnospermy began with Robert Brown's paper on Kingia, " with Observations .... on the Female Flower of Cycadeae and Coniferae," read before the Linnean Society in the year 1825, and published in King's Voyage in 1827 ; and the bibliography down to a recent date is given by Eichler in Flora Brasiliensis, Gymnospermia, iv. 435, and in Bliithendiagramme, i. 55-69; also ii. preface x.. See also Alph. DeCandolle, Prodr. xvi.- 345, 524. In this volume, the late Prof. Parlatore adhered to the ancient ideas in his monograph of the Coniferae. 2 The view here implicitly adopted is that of Beccari, founded on the study of Gnetum, and published in Nuovo Giornale Botanico Italiano, ix. 1877. It was before nearly or quite reached in successive steps, by J. D. Hooker, in his classical memoir on Welwitschia, in Trans. Linn. Soc. xxiv.; Stras- burger, Die Coniferen und die Gnetaceen, 1872; and W. R. McNab, in Trans. Linn. Soc. xxviii. 1872. 270 THE FLOWER. genus Gnetum, shrubs or trees, with nearly the aspect of Angiosperms, having broad and pinnatery-veined leaves ; Wel- witschia of tropical W. Africa, remarkable for its persistent cotyledons which form the only foliage of a woody and long- enduring plant, and for its stem or trunk which broadens with- out lengthening, except in its flower-stalks ; also Ephedra, of much branched shrubs, mainly of warm-temperate regions, leafless or nearly so, one species of which inhabits Europe and two the southern borders of the United States. 506. The flowers in all Gymnosperms are diclinous, either dioecious or monoecious ; except that those of the strange Gneta- ceous genus Welwitschia are structurally polygamous, the male flowers having a well-formed but sterile gyncecium. 507. In Coniferae, the largest and most important type, are embraced all the familiar Gymnosperms of temperate regions, Pines, Firs, Cedars, Cypresses, which bear their flowers in catkin-like clusters and their fruit in cones, and also the Yews and allied trees which do not''' produce cones. Perianth being want- ing and the sexes wholly separate, the floral type is so degraded that it becomes doubtful whether each cluster of anthers, or of ovuliferous scales or ovules, constitutes a blossom or an inflores- cence. Certain botanists look upon a whole catkin, and others upon a male catkin only, of a Pine or Fir as forming one flower. It is here assumed that each stamen of the one and each ovu- f liferous scale of the other // answers to a flower of the v :''"::7 simplest sort.1 The anthers are extrorse, the cells or pollen-sacs belonging to the outer or lower side of a scale or a 1 It will be seen that, for the female flowers, this follows of course from generally accepted view ; and, where this is conceded, analogy may extend it to the male catkins also: yet in such cases, where all the phylla of an indefinite simple axis are stamens, spirally arranged on it, the difference between inflorescence and male flower completely vanishes. FIG. 554. Female flower of a Yew, an ovule surrounded by its bracts. 555. Longi- tudinal and more enlarged section of a female flower of Yew and of the upper part of the shoot it terminates : the thick coat of the ovule open at the top, the nucleus within, and the beginning of the disk outside of the coat, are seen in section. After Stras- burger. FIG. 556. Young fruit (berry-like cup surrounding the seed) of Yew. 557. Longi- tudinal section of a mature fruit of the same. After Decaisne. GYNGECIUM OF GYMNOSPEKMS. 271 connective : sometimes these sacs or cells are two, and the organ evidentl}' homologous with an ordinary stamen : often they are more numerous (from three to twenty) and variousty disposed. 508. The Yew Fam- ily (Taxinese) is next to Gnetaeese in structure. It is generally ranked as a suborder of Coniferae, but it may claim to be a distinct order. The gynoecium is a naked ovule, terminating a stem,1 and surrounded by several bracts. After fertilization, an outgrowth of the receptacle (or a kind of disk, 394) makes its appearance as a ring girding its base : this grows in height and thickness, and becomes a soft-fleshy cup, imitating a hollow berry, in the bottom of which the stony-coated seed nestles. (Fig. 554-557.) Very similar is the gyncecium of Torreya, except that the cup- shaped disk develops almost simultaneously with the ovules, and as it grows becomes adnate to the large seed in the form of a fleshy coating. In the Gingko, two or more similar ovules are nakedly developed on a naked peduncle, un- accompanied even by a bract (Fig. 558) , and one or more of these ripens into the berry-like seed, Fig. 559. In Podocarpus there are some sub- tending bracts, and the naked ovule 1 It does not therefore follow that the ovule is a part of the axis, or is terminal in the sense of being its direct continuation. In tin's regard it may be only what the pistil of a Cherry is, which to all appearance is equally a terminal production, but is really the representative of the last leaf of the axis. If so, that leaf is here suppressed to the utmost, and replaced by what is ordinarily its outgrowth, the ovular nucleus and its coat. The structure of Podocarpus favors this interpretation. FIG. 558. Female flowers of Gingko biloba or Salisburia aclian'ifolia. 558o. Portion of the same enlarged. After Strasburger. 559. A drupaceous seed of the same, in vertical section, exhibiting the mature disk which forms the flesh, the crustaceous seed- coat, within which is the kernel of the seed ; at the base on one side a sterile ovule is seen. After Decaisne. FIG. 560. Female flower of Podocarpus (an ovule inverted on a column or elevated support), subtended by bracts. After Eichler. FIG. 561. Magnified vertical section of a similar flower of Podocarpus. After Strasburger. 272 THE FLOWER. is inverted on a more or less lengthened and stout support, which is conceived to represent the carpel. (Fig. 560, 561.) 509. In the true Coniferae, to which Pines, Cypresses, and all such cone-fruiting trees belong, the ovules are borne on or in the axils of scales which are imbricated on a simple axis, in a spicate or capitate manner ; and the male flowers, each a single stamen, are also similarly spicate or capitate. Both are com- monly termed aments >or catkins ; and the female ones properly so, according to the present view ; but the only scales of the male catkins are parts of the anther, being a dilated tip of the connective in Pines, and a scale bearing anther-cells or pollen- 562 sacs on its back in Cypress. 510. In the Pine tribe the flowering female catkin consists of bracts, spirally imbricated on the cauline axis : in the axil of each bract or sterile scale is developed a scale which bears two ovules, and is therefore regarded as of carpellary nature. These ovules are pro- duced on the lower part of the upper face of this carpellary scale, and are wholly adherent to it quite to the orifice, which is directed downward. (Fig. 562, 563.) The ovuliferous scale in becoming fructiferous usually much and soon out- grows the bract, which is concealed in the Pine- cone (or sometimes obliterated) ; but it remains conspicuous in sundry Fir-cones. After fertil- ization, the scales, successively covering each other in close imbrication, protect the growing seeds as effectually as would a closed ovary. Sooner or later after ripening the scales diverge, and the seeds peel off the face of the scale with a wing attached, and fall or are dispersed by the wind.1 1 Among those who admit as well as those who reject gymnospermy, there has been much controversy over the morphology of the parts. With the former, the discussion turns on the character of the ovuliferous scale. As to this, the hypothesis originally proposed by Mohl, and adopted by Braun, is now said to be satisfactorily demonstrated by Stenzel, in Nov. Act. Nat. Cur. xxxviii. 1876. See note by Engelmann in Amer. Jour. Sci. Dec. 1876, and also the preface to the second part of Eichler's Bliithendiagramme, FIG. 562. View of the upper face of a carpellary scale of a Larch, showing the pair of adnate ovules. FIG. 563. Similar view of a carpellary scale of a Larch, and of a bract behind it. 564. Ground plan of the same in diagram, reversed ; the upper figure denoting the axis of the cone, the lower the bract, the middle one the carpellary scale and the two ovules borne on its face. After Eichler. GYNOECIUM OF GYMNOSPEKMS. 273 511. In the Araucaria tribe the ovuliferous or carpel-scale is throughout smaller than the bract, and is completely adnate to it, or with only the tip free ; that of Araucaria (Fig. 565) bears only one ovule, high on the carpel, the orifice downward as in the Pine tribe. In Taxodium, Sequoia, and the like, the cone- scale is equally inferred to be composed of bract and carpel-scale united ; and indications of this composition are to be observed. The ovules (from two to several) are at the base of the scale, erect and free. The cone-scales are alternate and spiral on the axis, but indistinctly so in Taxodium, the Bald Cypress or so-called Cypress of the Southern United States. 512. In the true Cypress tribe (Cupressineae) the cone-scales, which are never numerous, are opposite or verticillate, i. e. h'ke the foliage- leaves, in whorls of twos, threes, or sometimes fours ; and the ovules are from two to 1878, where it is fully adopted. It was suggested by certain rather common monstrosities, and by the two combined leaves of Sciadopitys. According to this view, the ovuliferous scale in the Pine tribe is com- posed of two leaves of an arrested and transformed branch from the axil of the bract, which are in the normal manner transverse to the subtending bract, are here carpellary, each bearing an ovule on the dorsal face ; the two are coalescent into one by the union of their posterior edges, and the scale thus formed is thus developed with dorsal face presented to the axis of the cone, the ventral to the bract. It is therefore a compound open carpel, composed of two carpophylls. This character of being fructiferous on the back or lower side of the leaf occurs in no other phsenogamous plants, but is the rule in Fefns, from something like which Coniferse may be supposed to have been derived ; the ovules of the one in this regard corresponding to the sporangia of the other. FIG. 565. Vertical section (in diagram) of a bract, adnate carpel-scale, and adnate ovule of Araucaria imbricata. After Eichler. FIG. 566. Brauchlet of the American Arbor-Vitse, considerably larger than in na- ture, with a forming fertile cone. 567. One of the scales removed and more enlarged, the inside exposed to view, showing a pair of naked erect ovules on its base. FIG. 568. Fertile flowers of true Cypress (Cupressus seaipervirens), after Baillon : a forming cone, with one scale cut away, to show the cluster of ovules under it. 18 274 THE FLOWER. several at or on the base of each cone-scale, always with orifice upward. Arbor- Vitae (Fig. 566, 567) has a single pair of ovules to the scale ; Junipers, sometimes only one ; true Cypresses (as in Fig. 568), often a dozen or more. At flowering time, the cone- scales mostly appear as if simple ; but in most genera they soon thicken greatly within ; and they are usually understood to be composed of bract and carpel-scale combined, the latter of the same constitution as that of PineS and Spruces, but perfectly consolidated and confluent with the bract-scale.1 513. In Cycadaceae, the type of the flower of Angiosperms is almost or quite lost ; yet the organs may be homologized with those of Coniferae, which these plants are wholly unlike in habit. 1 This internal and ovuliferous scale may seem to be wholly hypotheti- cal, and assumed to homologize the cupressineous with the abietineous cone. Without it, we should have to consider that, while in*5\.bietineae the ovules belong to leaves of a secondary axis, in Cupressineae they are borne on those of a primary axis, or else are axillary productions without carpels. But in the Araucaria tribe the internal scale is obvious ; and there are suffi- FIG. 569-575. Zamia, chiefly Z. media, after Richard. 569. A male plant. 570 Lower part of a male catkin. 571. A stamen removed, showing numerous small pollen- sacs under the peltate top. 572. A female catkin, with a quarter section cut away. 573. A female flower or carpel, with two enlarging ovules or young seeds. 574. Ripe seed, with the thick fleshy coat cut away at apex. 575. Longitudinal section of ripe seed, more enlarged. GYNGECIUM OF GYMNOSPERMS. 275 Their likeness to Palms and other Monocotyledons is confined to the port of their unbranched trunks and their pinnate leaves with parallel-veined or simple-veined leaflets ; nor have the}' any further resemblance to Ferns, except that in some the leaflets are circinate in vernation. Although a tropical type (of small present importance, compared with the part which it played in the Devonian and Cretaceous periods) , it has one small representa- tive (Zamia media, the Coontie) at the south-eastern extremity of the United States, and a more striking one (Cycas revoluta, well known in cultivation) in the southern parts of Japan. 514. Following the analog}' of Coniferae, each scale (whether of the pollen-bearing or the ovule- and seed-bearing ament) of Zamia (Fig. 569-575) is here regarded as a flower. Here the phylla, or scales with peltate top and stalk-like base, are exter- cient indications of similar composition in the cupressineous cone-scales tc induce the adoption of it by Parlatore, who rejected the idea of gymnospermy ; and, finally, this composition is nearly demonstrated by VanTieghem (1868) upon the anatomical structure, and by Strassburger (1872) on the development. FIG. 576-578. Carpophylla of Cycas revoluta, much reduced in size. 576. One hearing ovules below and leaflets or leaf-lobes towards the apex. 577. A similar carpo- pliyll with leaf-lobes reduced to mere teeth, and ovules in place of the lower teeth. 578. A similar carpophyll in mature fructification, bearing the large drupaceous naked seeds. The last two after Hichanl. 276 THE FLOWER. nally much alike in the two sexes, which throughout the family occupy separate plants. The male flower (Fig. 568) or stamen, if it may be so termed, bears indefinite pollen-sacs on the under side of the peltate portion, sometimes extending to che upper part of its stalk. The homologous female flower, or carpophyll, bears a suspended ovule on each side of the stalk (Fig. 573), which becomes a large fleshy-coated seed. In Cycas the male ament is not very dissimilar, although on a larger scale. But the carpophylls are evident leaves, not condensed into an ament, but loose or spreading, of a character and aspect intermediate between the lax bud-scales which precede and the pinnate foliage- leaves which follow them in development. Along the margin of what would be leaf-blade they bear ovules in place of leaflets, lobes, or teeth (Fig. 576-578) ; and these, when fertilized from the male flowers, mature into large and drupaceous naked seeds. Even without fertilization, such seeds grow to their full size on the female plant of the common Cycas (or falsely so-called Sago Palm) , but form no embryo. SECTION Vni. THE OvuLE.1 515. Omles (302) are peculiar outgrowths or productions of carpels which, upon the formation of an embryo within, become seeds. In the angiospennous gynoecium (476) they are nor- mally produced along the margins, or some part of the margins, of the carpellary leaf (478) , either immediately, or by the in- termediation of a placenta (485), which is a more or less evident development of the leaf-margins for the support of the ovules. Rarely, yet in a considerable number of cases (501, 502), ovules are developed from the whole internal surface of the ovary, or from various parts of it, in no definite order, directl}' from the walls, and without the intervention of any thing which can be regarded as placenta. In Gymnosperms (504-514) the ovules are borne on the face of the carpellary scale or at its base ; or on leaf-margins, as in Cycas ; or, when there is no representa- tive of the carpel, on the cauline axis, seemingly as a direct growth of it. (508, note.) 516. As to attachment, ovules are either sessile, i. e. stalk- less, or on a stalk of their own (Fig. 582, 584) , the FUNICULUS or PODOSPERM. As to number they are either solitary, few, or 1 Lat. Ovulum, pi. Ovula, diminutive of ovum (egg), perhaps first used by Adanson OVULES. 277 indefinitely numerous. They may^ also be indefinite or variable in number when not particularly numerous. 517. As to situation and direction within the ovary, the terms are somewhat special. Ovules are erect, when they rise from the very bottom of the cell, as in Fig. 580 ; ascending, when attached above its bottom and directed upward, as in Fig. 579 ; horizontal, when borne on one or more sides of the cell and not directed either upward or down- ward, as in Fig. 314, 315, 530 ; pendulous, when more or less hanging or declining from the side of the cell ; suspended, when hanging from the apex of the cell, as in Fig. 581. 518. The body and only essential part of an ovule is its NUCLEUS. This in most cases is invested by one or two proper coats. The coats are sacs with a narrow orifice, the FORAMEN. In the seed, the closed vestige of this orifice is termed the Micropyle ; wherefore this name is sometimes applied to it in the ovule likewise. When the ovule has two coats, the foramen of the outer one is called EXOSTOME, of the inner ENDOSTOME ; literally the outer and the inner orifice. The coats themselves have been named PRIMINE and SECUN- DINE, but with an ambiguity in the application which renders these names unadvisable : for in their formation the coats appear later than the nucleus, the inner coat earlier than the outer; and the name of primine has by some writers been applied to the earlier formed, by others to the external coat. The proper base of the ovule, from which the coats originate and where these and the nucleus are confluent, is the CHALAZA. The attachment of the ovule to its funiculus or support, which in the seed becomes the FI€r. 579. Ovary of a Buttercup, divided lengthwise, to display its ascending ovule. 580. Same of Buckwheat, with an erect ovule. 581. Same of Anemone, with a sus- pended ovule. FIG. 582. Diagrammatic section of a typical or orthotropous ovule (such as that of Fig. 582o), showing the outer coat, a, the inner, 6, the nucleus, c, the chalaza. or place of junction of these parts, d. (The coats are never so separated and the nucleus so re- duced in size as is represented in this mere diagram.) 583. An ovule similar to the preceding, but curved, or campylotropous. 584. An amphitropous ovule. 278 THE FLOWER. HILTJM, takes also this latter name in the ovule. In the simplest form of ovule (as in Fig. 582, 580), hiluin and chalaza are one. So also in cases where the body of the ovule incurves, as in Fig. 583. But very commonly the place of attachment, which becomes the hilum, is more or less distant from the chalaza ; as in Fig. 584 and 587, where the hilum is lateral, but the chalaza at the larger end, the two being connected by a short ridge ; and in Fig. 588 the two are separated by the whole length of the ovule. 519. The simplest and most rudimentary ovule is that with- out a coat, as in Mistletoe and the whole order Loranthacese, and in Santalacese and Gnetaceae. This has been called a naked ovule ; but long before ovules of such simplicity were known this term had been appropriated to. those of Gymnosperms, in the sense of destitute of ovarial or pericarpial covering, i. e. to uncovered ovule, not to uncovered nucleus. The ovule consist- ing only of nucleus may be termed (after Alph. DeCandolle) simple, or better achlamydeous.1 520. The tunicated or chlamydeous ovule is of three principal kinds, with one or two subordinate modifications. These are the orthotropous, campylotropous, and anatropous, and the modi- fication called half-anatropous, or amphitropous.'2 521. Orthotropous (Fig. 580, 582, 585), or straight ovule, is the simplest but least common species, being that in which the chalaza is at the evident base, and the orifice at the opposite extremity, the whole ovule straight (as the first part of the name denotes) and symmetrical. Atropous, meaning not turned at all, is a later and etymologically much better name, but it has 1 An epidermal stratum or tegument may not be wanting to such ovules, forming a sort of adherent covering ; but this in nature and origin is not similar to the ovular coats. 2 In Latin form, orthotropa, campylotropa, anatropa, amphitropa, — names given by Mirbel, and referring to the way in which the ovule is turned either on itself or on its support. Some English botanists incongruously write ortho- tropal, campylotropal, &c. FIG. 585. Orthotropous or Atropous ovule of Buckwheat. 586. Campylotropons ovule of Chickweed. 587. Amphitropous ovule of Mallow. 588. Anatropous ovule of a Violet. The letter h indicates the hilum ; c, the chalaza, which in 585 and 586 corre- sponds to the hilum; /, the foramen or orifice; r, the rhaphe. OVULES. 279 not come into general use. This ovule is characteristic of Polygonaceae, the proper Urticaceas, Cistacese, &c. 522. Campylotropous (Fig. 583, 586) is the name of the ovule 7 which in the course of its growth is curved on itself so as to bring the orifice or true apex down close to the base, here both chalaza and hilum. This and the orthotropous ovule begin their development on the placenta in the same way, but the Campylotropous develops unequal!}-, one side enlarging much more than the other, especially at the base, until the ovule becomes reniform, and chalaza and orifice are brought into close proximity. Campylotropous ovules are characteristic of Cruciferae, Capparidaceae, Resedacese, Caryophyllaceae, and Chenopodiaceae. 523. Amphitropous (Fig. 584, 587), also termed Heterotropous and sometimes Half-anatropous, is between the preceding and the following ; and it passes in various instances either into the one or into the other. The body of the ovule is straight o»- straightish, but it stands as it were transversely or at right angles to the funiculus and hilum ; and it is fixed by the middle, the chalaza at one end, the orifice at the other. An apparent, continuation of the funiculus, adherent to the outer coat, extends from the hilum to the chalaza. Compared with the preceding form, the explanation is, that the unequal development at its formation is confined to the basal half, and the axis remains straight, while the whole is half inverted by the very unequal growth. Compared with the next form, the inversion is less and the later growth or extension of the apical portion greater. The amphitropous ovule is characteristic of Primulacese, and is common in Leguminosae. 524. Anatropous (Fig. 588, also 579, 581, 597) is the name of far the commonest species of ovule, that in which the organ, under the course of its growth, is quite inverted on its base ; so that, instead of standing at right angles with the funiculus, it is parallel with it, or rather with the apparent continuation of it, which is adherent to its surface as a sort of ridge or cord extending along the whole length of the ovule, from hilum to chalaza. The latter occupies the seeming apex of the seed ; and the organic apex or orifice is at the other end, close beside , the hilum. At maturity, the ovule is straight, but not wholly symmetrical, the attachment being oblique or somewhat lateral, and the ridge or cord on that side not rarely prominent. 525. The cord or ridge, which extends along the whole length of the anatropous ovule, and for half its length in the amphi- tropous (Fig. 588, 587, r), is named the RHAPHE. This is not 280 THE FLOWER. at all a seam, as the Greek word denotes. Its origin, and the whole structure of such ovules will be apprehended by comparing various stages of its growth. 526. An ovule of any kind at the beginning is an excrescence or outgrowth of the placenta, or of some part of the leaf-surface if there is no developed placenta. This incipient ovule is the nucleus (518), or the nucleus surmounting a rudimentary funiculus. The nucleus is soft cellular tissue only, from first to /A I last. The achlamydeous ovule (519) undergoes no IJp / further development except in size or shape. Indeed i^' sometimes (as in Balanophoreae) this bare nucleus is reduced to a few cells of parenchyma. 527. In ordinary ovules a new growth early begins around the base of the nucleus, or is sometimes coetaneous with it, at first as a ring (or part of a ring) , soon as a cup, at length as an enclosing sac or covering, open at the top ; this is the inner coat of the ovule when there are two. The outer coat begins and goes on in the same wa}r, and at length grows over and encloses the inner coat as that did the nucleus. (Fig. 590-595.) When- ever there is a third and more exterior coat it is formed during the growth of the fertilized ovule into the seed, to which there- fore it belongs, and in which it takes the name of arillus. (597.) At the time of fertilization the apex of the nucleus, or a pro- longation of it, usually projects beyond the orifice and there receives the descending pollen-tube. Some fibro-vascular tissue, especially spiral ducts, may be found in the funiculus and cha- laza, sometimes extending into the coats. 528. The development of the orthotropous or atropous (un- turned) ovule proceeds symmetrically, without distortion, the parts keeping their primitive direction. In the campylotropous, the whole of one side of the ovule greatly outgrows the other. FIG. 589. Magnified view of a vertical section of a carpel of Magnolia Umbrella, about a month before anthesis, showing one of the two nascent ovules, at this time only nucleus. FIG. 590-597. Further development of the ovule of Magnolia Umbrella, showing the formation of the coats and the anatropy. 590. Ovule a week older than in 589. 591. Same a week or two later. 592. Same a few days later. 593. Same from a nearly full- grown flower-bud. 594. Same at time of anthesis. 595. Vertical section of the last through the middle of the rhaphe. 596. Cross-section of the same. (See Jour. Linn. Soc. ii. 108.) OVULES. 281 In the anatropous, the inequalit}- of growth is mainly confined to the base or chalazal region, which ends by becoming upper- most ; and the full-grown ovule has the ap- pearance of being inverted on and adherent to the upper portion of its funiculus, the rhaphe. Fig. 589-597 illustrate the course d of development from a comparatively early period. 529. The direction of anatropy or of other turning of the ovule in the course of growth is somewhat diverse. But in general, when- ever ovules are in pairs, the two turn from each other, in the manner of Fig. 315, and so present their rhaphes back to back. The rhaphe-bearing may therefore be called the dorsal side of the anatropous ovule. The same is true in the case of numerous ovules, viz., those of one half of the placenta (or one leaf-margin) turn their backs to those of the other. When such ovules are solitary or in single rows, and either ascending or hanging, the rhaphe is usually on the side next to the placenta or ventral suture, as in Fig. 579 : it is then said to be ventral (i. e., next the ventral suture), or adverse to the placenta. In certain cases, mostly in hanging ovules, as in Fig. 581, the rhaphe looks in the opposite direction, toward the dorsal suture or midrib of a simple ovary : it is then said to be dorsal or averse from the placenta.1 1 By comparison of Fig. 578 with 576 and the like, it may be perceived that the difference is explicable by a kind of resupination of the ovule of the former. That of Ranunculus, if inserted higher, would become hori- zontal ; and if the insertion were transferred to the very summit of the cell, it would be suspended and the rhaphe averse, as in Fig. 581. Upon this conception, Euphorbia and its allies has normally suspended ovules, the rhaphe being next the placental axis, and Buxus and its allies, resupinately suspended ovules, the rhaphe averse. The propriety of regarding the ad- verse rhaphe as the normal condition is confirmed by the fact that the only instance we know of solitary erect ovules from the base of the cell having the rhaphe averse is that of Rhamnus and its allies ; and here it was shown by Bennett (in PI. Javan. Rar. 131), and confirmed by the analyses of Sprague (Gray, Gen. 111. ii. 168, plates 163-169), that the rhaphe of the young ovules is ventral, so that the dorsal position, when it occurs, is the result of torsion. J. G. Agardh (in his Theor. Syst. PI. 178, &c.) maintains the contrary, but is not sustained by later observers. Accordingly, even if we adopted Agardh's estimate of the botanical value of the characters here considered, we should prefer to express these differences in the phraseology above indicated, and not to adopt his terms, FIG. 597. Same as 595 more magnified ; the outer coat, (a), the inner (6), nucleus (c), and the bundle of spiral ducts (d) in the rhaphe (running from placenta to chalaza) indicated. 282 THE FLOWER. 530. Origin and Nature of the Ovule. It has been already stated in general terms that ovules are peculiar outgrowths or productions, generally of the margins of carpellary leaves (515) ; that they are composed of parenchymatous cellular sub- stance, at least as to the nucleus, of which the simplest ovule wholly consists (526) ; that the coats originate subsequently to the nucleus ; and that the outer coat is of later origin than the inner one. (518.) The mamilifonri protuberance of which the forming ovule at first consists originates in one or more cells of a layer directly beneath the epidermis.1 531. The morphological nature of the ovule has been much discussed. The commonly prevalent view was that the ovule is homologous with a leaf-bud, and that its nature is in some degree illustrated by such buds as those which develop on the margins of the leaves of Bryophyllum, as shown in Fig. 322. But such buds, and the bulblets or fleshy buds which appear on the face of certain leaves, follow the universal order of budding growth, that is, are centripetal in development, the outermost parts being the earlier and the inmost the later formed. The ovule, on the contrary, is basipetal or centrifugal in develop- ment, the nucleus being first and the outer coat last formed ; therefore the coats are not homologous with sheathing leaves, nor the nucleus with a vegetative axis. The older theory has accordingly given way to the present one, in which the ovule answers to the lobe of a leaf peculiarly transformed, or to an outgrowth of a leaf, whether from its edges or surface. The apotropous, epitropous, and heterotropous (the first two new, the last employed in a new sense), the more so since the application is confused with hypo- thetical considerations and the necessity of bringing the ovules ideally back to ascending or horizontal positions. It may be stated, briefly, that fletero- tropous, in Agardh's terminology, applies to the normal position of collat- eral ovules, with rhaphes back to back, in opposite directions on the two halves of the placenta ; Apotropous, to an erect or ascending ovule with its rhaphe next the placental axis, and a hanging one has its rhaphe averse from it ; Epitropous, when an erect or ascending ovule has its rhaphe averse, and a hanging one has it adverse. 1 Hofmeister's statement that the simple ovule of Orchis originates in the division of a single epidermal cell (and is therefore a trichome) is con- troverted by Strasburger and by Warming. The latter adds the remark, that even if it were so in cases of extreme simplicity, this would not invali- date the proposition that the ovule is to be regarded as the homologue of the lobe of a leaf. Such a lobe is not rarely reduced to a single bristle. For the whole subject of the origin, development, morphology, and theory of the ovule, see Warming's very elaborate and perspicuous memoir, De 1'Ovule ; also the papers of Celakowsky, Van Tieghem, &c., referred to in notes to paragraphs 500, 501. OVULES. great advantage of this view is that it serves to homologize the fructification of Flowering Plants with that of the higher Flower- less Plants, or the Ferns, the sporangia or analogues of the ovule being outgrowths of the leaf.1 532. Origination of the Embryo. The whole process of fer- tilization and the resulting produc- tion of the embryo, also the history of the subject, belongs to the suc- ceeding volume, involving as they do questions of minute anatomy and of physiology. But a general idea may here be given of the wa}T in which the embryo originates. The tube which a grain of pollen sends forth into the stigma (574, 575) penetrates the style through loose conducting tissue charged with nourishing liquid, reaches the cavity of the ovary , enters the orifice of an ovule to reach the apex of the nu- cleus, although the latter sometimes projects to meet the pollen-tube. Meanwhile a cavity (the embryo- sac, which is formed by the great enlargement of a single cell of the tissue, or of two or more cells the product of a mother cell) forms in the nucleus, the upper part of it commonly reaching nearly or quite to the apex of the nucleus, which the pollen-tube impinges on or sometimes penetrates. A particular portion of the protoplasm contained in the embr3ro-sac forms a globule, and this at the time 1 The advocates of this view naturally maintain that ovules and placenta? always belong to leaves, and never truly to a cauline axis ; that in the pre- central placentation of Primulaceae, the actual ovuliferous surface is an out- growth of the bases of the carpellary leaves coalescent with each other and adnate to a prolongation of the torus ; also that in those Gymnosperms which have no carpophyll, such as Yew, the whole nascent carpellary leaf, or rather the papilla which would otherwise develop as such, is directly developed into ovule. This, being solitary and the last production of the axis, necessarily appears to terminate it. (500, 501, notes.) FIG. 598. Diagram representing a magnified pistil of Buckwheat, with longitudinal section through the axis of the ovary and orthotropous ovule ; some pollen on tlie stifrmas, one grain distinctly showing its tube, which has penetrated the style, reappeared in the cavity of the ovary, entered the mouth of the solitary ovule (o), and reached the embryo-sac (s) near the embryonal vesicle (v). 284 THE FLOWER. of fertilization is found at the apex of the sac, at or adjacent to the part reached b}~ the pollen-tube. Not rarely it adheres to the wall of the sac exactly opposite the termination of the pollen-tube. This is called the embryonal vesicle. To it the con- tents of the pollen-tube are in some manner trans- ferred. Upon which it takes a more definite shape, acquires a wall of cellulose, and so becomes a vegetable cell. This divides into two, the lower again into two, and so on, forming a chain (the suspensor or pro- embryo). The terminal cell of this divides again and again in three directions, producing a mass of cells which shapes itself 604 eos eo6 607 into the embryo, the initial afl f\ /^Sv/l plant of a new £enera- / I d IK IT " JlrBll tion- Ordinarily the sus- pensor soon disappears. It is attached to the ra- dicular end of the em- bryo, which consequently always points to the foramen or micropyle of the seed. The process in Gymnosperms is more complex, and has to be separately described. 533. Polyernbryony, the production of two or more emb^os in one seed, is not uncommon in Gymnosperms (there being a kind of provision for it), and is of occasional but abnormal occurrence in Angiosperms, in the seed of Mistletoe, Santalum, &c. In these it results from the production and fertilization of more than one embryonal vesicle. Strasburger has recently ascer- tained that the commoner polyembryony in the seeds of Onions, Oranges, Funkia, &c., results from the production of adventive embryos, which originate in the nucleus outside of the embryo- sac and wholly independent of fertilization.1 Two kinds of 1 Strasburger, Ueber Polyembryonie, in Zeitschr. Naturwis. Jena, xii. 1878 (see Amer. Jour. Sci. April, 1879). It was found that when, by exclu- sion of pollen, the formation of a normal embryo was prevented, no adventive FIG. 599. Diagram of the suspensor and incipient embryo at its extremity. 600. The same, with the embryo a little more developed. 601. The same, more developed still, the cotyledons faintly indicated at the lower end. 602. Same, with the incipient cotyledons more manifest. 603. The embryo nearly completed. FIG. 604-606. Form ing embryo from a half-grown seed of Buckwheat, in three stages. 607. Same, with the cotyledons fully developed. THE FRUIT. 285 anomalous reproduction are therefore now known, which are intermediate between sexual and non-sexual, between budding and fruiting propagation, viz., — Apogamy, which is budding growth or prolification in place of that which should subserve sexual reproduction. This was dis- covered in Ferns by Prof. Farlow, while a pupil of De Baiy, by whom our knowledge of the process has recently been extended, and this name imposed.1 The production of bulblets in place of seed or embryo answers to this in Flowering plants. Parthenogeny, the counterpart analogue of apogamy, is the non-sexual origination of an embryo extraneous to the embryonal vesicle or even the embryo-sac. However abnormal, its occur- rence is probably not so rare as has been supposed. CHAPTER VII. THE FRUIT. SECTION I. ITS STRUCTURE, TRANSFORMATIONS, AND DEHISCENCE. 534. The Fruit consists of the matured pistil or gj-nrecium (as the case may be) , including also whatsoever may be joined to it. It is a somewhat loose and multifarious term, applicable alike to a matured ovary, to a cluster of such ovaries, at least when somewhat coherent, to a ripened ovary with catyx and other floral parts adnate to it, and even to a ripened inflores- cence when the parts are consolidated or compacted. Fruits, accordingl}-, are of various degrees of simplicit}' or complexity, and should be first studied in the simpler forms, namely, those which have resulted from a single pistil. Such a fruit consists of Pericarp with whatever may be contained in it and incorpo- rated with it. embryo appeared in those seeds which habitually produce them. To this Caelebogyne offers an exception. The female of this dioecious plant habit- ually matures fertile seeds, with a well-formed embryo, in Europe when there are no male plants in the country. Strasburger ascertained that the embryo thus formed is adventive, the embryonal vesicle perishing. Parthenogenesis, of which Caelebogyne was the most unequivocal case, is thus confirmed, and is shown to occur in most polyembryony ; but it is at the same time explained to be a kind of prolification. i See Farlow, in Proc. Am. Acad. ix.68; De Bary, Bot. Zeit. xxxvi. 465-487. 286 THE FRUIT. 535. The Pericarp, or Seed-vessel, is the ripened ovary. It should, therefore, accord in structure with the ovary from which it is derived. Yet alterations sometimes take place during fruc- tification, either by the abortion or obliteration of parts, or In- accessory growth. 536. Internal Alterations. Thus, the ovary of the Oak con- rists of three cells, with a pair of ovules in each ; but the fru.t has a single cell, filled with a solitar}" seed, onl}' one ovule beinj/ matured, while two cells and five ovules are suppressed, the remains of which may be detected in the acorn. The ovary of the Chestnut has six or seven cells, and a pair of suspended ovules in each ; but only one of the dozen or fourteen ovules ever develops into a seed, except as a rare monstrosity. The three-celled ovary of the Horsechestnut and Buckeye is similar in structure (Fig. 608-611), and seldom ripens more than one or two seeds ; but the abortive seeds and cells are obvious in the ripe fruit. The ovary of the Birch and of the Elm is two-celled, with a single ovule in each cell : the fruit is one-celled, with a solitary seed ; one of the ovules being uniformly abortive, while the other in enlarging thrusts the dissepiment to one side, and obliterates the empty cell. Similar suppressions in the fruit of parts actually extant in the ovary are not uncommon. 537. On the other hand, there may be more cells in the fruit than there are primarily in the ovary. Thus the fruit of Datura is dicarpellary and normally two-celled, with a large placenta projecting from the axis far into the cells. But each cell be- comes bilocellate, that is, divided into two, by a false partition growing out from the back of each carpel and cohering with the middle of the adjacent placenta. So the 5-carpellary and nor- rnalry five-celled ovary of common Flax early becomes spuriously ten-celled (morphologically speaking, not 10-locular, but 10- locellate), by a false partition extending from the back of each FIG. 608. Longitudinal section of the ovary of a Buckeye (jEsculus Pavia), showing the pairs of ovules in two of the cells. 609. Transverse section of the same displaying all three cells and six ovules. 610. Same of half-grown fruit, with single fertile seed, abor- tive ovules and •! HU-rating cells. 611. Dehiscent one-seeded fruit, diminished in size ITS STEUCTUKE AND TRANSFORMATIONS. 287 carpel across its cell (Fig. 539-541) ; and the solitary carpel is similarly divided lengthwise in man}- species of Astragalus, as in Fig. 534. Transverse divisions or constrictions across a maturing ovary (such as is seen in Fig. 620) are not uncom- mon, especially in legumes and other pods, and are of little mor- phological significance. 538. External Accessions may here be referred to. The wing of the pericarp in Maple, Ash, and the like (Fig. 625-027) , are familiar instances of this ; and of the same nature are the im- bricated scales which cover some Palm-fruits ; the prickles on the pod of Datura, Ricinus, &c., and the hooked or barbed prickles of many small pericarps (as in various Borraginaceae) , which thus become burs and are disseminated by adhering to the hairy coat of cattle. All these are of the nature of superficial outgrowths, and these especially affect the pericarp or parts connected with it. 539. Persistence of Connected Organs. An adnate calyx (331) , being consolidated with the ovary, necessarily makes a constit- uent part of the fruit, in the pome (575) doubtless a very large part. The limb or lobes of such adnate organ may persist, as the tips of the sepals on an apple or quince, and may be turned to useful account, as is the pappus of Composite for dissemina- tion. Or, in small pericarps, the style may persist as part of the fruit, and subserve the same ends, either by becoming feathery for aerial dissemination, as in Clematis and in one section of Geum, or by becoming hooked at the tip for adhesion to fleece, &c., as in other species of the latter genus. Or adjacent parts which are not actually incorporated with the pericarp may play similar parts in the economy, as the hooks on the calyx-tube of the dry calyx of Agrimonia, which at maturity is detached with the included fruit, the flesh}- fructiferous calyx of Gaultheria (Fig. 651) and of Mulberry (Fig. 654) ; and the pulpy fructiferous re- ceptacle of the strawberry (Fig. 653) : the ultimate utilities in both classes of instances being similar, viz., wide dispersion of the seed by animals, whether by external carriage, or by being devoured and the voided seeds of fleshy fruits thus disseminated. 540. Transformations in Consistence. In the change from ovary to mature pericarp, various kinds of transformations may take place. In some the wall of the ovary remains thin and' becomes in fruit foliaceous or leaf-like, as in a pea-pod, the carpels of Columbine, and Marsh Marigold (Caltha) , or the pod of Colutea or Bladder Senna. In others it thickens and becomes at maturity either dry throughout, as in nuts and capsules ; or fleshy or pulpy throughout, as in berries ; or hard-rinded with- 288 THE FKUIT. out but soft within, as in a pepo ; or fleshy or berry- like without, but indurated within, as in all stone-fruits, such as the cherry and peach. 541. When the walls of a pericarp consist of two layers of dis- similar texture (as in a peach) the outer layer is called EXOCARP, the inner ENDOCARP, these terms meaning exterior and interior parts of a fruit. When the external layer is a comparatively thin stratum or film, it is sometimes termed the EPICARP. When it is fleshy or pulpy it is named S ARCOC ARP. When the endocarp within a sarcocarp is hard and bony or crustaceous, forming a shell or stone, this is termed a PUTAMEN. When three concentric layers are distinguishable in a pericarp, the middle one is called MESOCARP. 542. Fruits may be divided into two kinds, in reference to their discharging or retaining the contained seeds. They are dehiscent when they open regularly to this end ; indehiscent when they remain closed. There is a somewhat intermediate condi- tion, when they rupture or burst irregularly, as in Datura Metel, &c. Dry pericarps with single seeds are commonly indehiscent ; those with several or many seeds mostly dehiscent. Seeds pro- vided with a wing or coma or any analogous help to dispersion are always in indehiscent pericarps. Permanently fleshy peri- carps are indehiscent, stone-fruits as well as berries. But in some stone-fruits (i. e., with indurated endocarp and fleshy exocarp), such as those of Almond (Fig. 640) and Hickory, the barely fleshy exocarp or sarcocarp dries or hardens, instead of softening, as maturity is approached, and at length separates from the putamen by dehiscence. 543. Dehiscence, the opening of a pericarp for the discharge of the contained seeds, is r-gular or irregular; or, better, is normal and abnormal. For most of the abnormal or non-typical modes are as determinate and uniform in occurrence as the t3'pi- cal modes. A good English name for dehiscent pericarps in general is that of POD. 544. Regular or normal dehiscence is that in which a pericarp splits vertically, for its whole or a part of its length, on lines which answer to sutures or junctions, that is, along lines which correspond to the margins or midribs of carpellary leaves, or to the lines and surfaces (or commissures) of coalescence of con- tiguous carpels. The pieces into which a pericarp is thus sun- dered are termed VALVES. 545. The normal dehiscence of a carpel is by its inner, ven- tral, or ovuliferous suture, that is, by the disjunction of the leaf-margins, as in Fig. 618. Its only other line of normal dehiscence is by the opposite or dorsal suture, that is, down DEHISCENCE. 289 the midrib. Legumes usually dehisce by both sutures (as in Fig. 619), therefore into two valves. 546. A dehiscent pericarp formed of two or more carpels is called a CAPSULE. The two leading terms descriptive of capsular dehiscence were based upon the modes of opening of pericarps having as man}7 cells as carpels : they are the septicidal, that is, as the term denotes, cutting through the septa or dissepiments ; and the loculicidal, that is, cutting into the Loculi or cells. 547. Septicidal, the dehiscence through the dissepiments, is the disjunction of a pericarp into its constituent carpels, these then usually themselves dehiscing down their ventral suture, as in Fig. 612, illustrated by the diagram, Fig. 613. Good examples are furnished by the Hypericum Family (the pistil illustrated in Fig. 536, 537), where the placentae which compose the axis are carried away on the edges of the par- titions or introflexed valves ; also by Rhododendron, Kal- mia, and the like, in which the placentae remain combined into a column in the axis (the COLUMELLA or column) , from ^* *. which the edges of the valves f \ / >. break away. / \ / \ 548. The septicidal dis- * junction of the carpels does Onot of itself open the cells. Such separated carpels when 613 one-seeded not rarely remain closed, as in Mallow, Ver- bena, &c. Or when dehiscent they may open both by the ventral and dorsal sutures; i. e., the pericarp may first divide into its constituent carpels, and then each carpel break up into half carpels, as in Euphorbia. 549. Loculicidal, the dehiscence into the loculaments, loculi, or cells of the pericarp (shown in Fig. 614, and the diagram, 615), is that in which each component carpel splits down its FIG. 612. Septicidally dehiscent tricarpellary capsule of Elodes Virginica. 613. Dia- gram of septicidal dehiscence. FIG. 614 Loculicidally dehiscent tricarpellary capsule of an Iris, divided trans- versely at the middle. 615. Diagram of loculicidal dehiscence. 290 THE FRUIT. dorsal suture, as in Iris, Hibiscus, CEnothera, &c. In this, the dissepiments remain intact. If they break away from the centre then the}^ are borne on the middle of the valves, as in the figures above cited. If they remain coherent in the axis but break away from the valves, the result is one form of what is called — 550. Septifragal dehiscence, i. e., a breaking away of the valves from the septa or partitions, as shown in Fig. 616. This represents the loculicidal form of the septifragal mode, which is less common than that of the accompanying diagram, Fig. 617. Here the partitions alternate with the valves ; that is, the dehiscence of the pericarp is of the septicidal order, as near as may be, but the par- titions do not split, wherefore the valves break awa}- at the common junction. To this the term marginicidal has been applied. It occurs in the 2-3- carpellary capsule of Ipomaea (especially in the common Morning Glory) , in the 5-carpellary capsule of the North American species of Bergia ; likewise in the 2-carpellary pod of Cruciferae (Fig. 623) , with a difference that the placentae from which the valves break away are here parietal and the partition is abnormal. 551. The terms septicidal and loculicidal apply equally in plan, though not with etymological correctness, to one-celled capsules with either parietal (495) or free central (599) placentae. When the dehiscence is of the septicidal type and the placentation pari- etal, the (half) placentae are borne on the margin of the valves, as in the Gentian family and the species of Hypericum with one- celled capsule. When the placentae are borne on the middle of the valves, as in Violets, the dehiscence is of the loculicidal t}Tpe. In the case of free central placentae with no trace of partitions, the character of the dehiscence may usually be deter- mined by the position of the styles or stigmas relative to the valves. 552. Dehiscence may be quite normal although very partial, as when confined to the apex of the capsule, of Cerastium . and of Primula, and even to the pores under the radiate stigmas of Poppy. 553. Irregular or abnormal dehiscence is such as has no respect to the normal sutures ; as where the dehiscence is transverse ; FIG. 616. Diagram of loculicidally septifragal dehiscence. 617- Same of septicidally or rather marginicidally septifragal dehiscence. ITS KINDS. 291 either extending part way round, as in the pod of Jeffersonia, or completely round, so that the upper part falls off like an unhinged lid. This circumscissile dehiscence occurs in many plants of widely different orders ; such, for example, as Purslane (Fig. 621), genuine Amaranths, Plantain, Pimpernel, and Hen- bane. In other cases, as in Antirrhinum (Snap-dragon) and its allies, the cells burst by irregular laceration at a definite point, and discharge the seeds through the ragged perforation ; or one or more neat valvular orifices are formed on some parts of the wall, as in Campanula. SECTION II. THE KINDS OF FRUIT. 554. FRUITS have been minutely classified and named;1 but the terms in ordinary use are not very numerous. A rigorously exact and particular classification, discriminating between the fruits derived from simple and from compound pistils, or between those with and without an adnate calyx, is too recondite and technical, and sometimes too hypothetical, for practical pur- poses. It is neither convenient nor philosophical to give a substantive name to every modification of the same organ. For all ordinary purposes, both of morphological and s}Tstematic botany, it will suffice to characterize the principal kinds under the four classes of — Simple fruits, those which result from the ripening of a single pistil ; Aggregate, those of a cluster of carpels of one flower crowded into a mass ; Accessory or Anthocarpus, where the principal mass consists of the surroundings or support of either a simple or an aggregate fruit; Multiple or Collective, formed by the union or compact aggre- gation of the pistils of several flowers, or of more than one. 555. Simple Fruits may be distinguished, upon differences of texture, into Dry Fruits, Stone Fruits, and Baccate Frvits ; or, better, into Dry and Fleshy ; and the first may be divided into 1 The greater part of the forty-three substantive names of Desvaux's, and even of the thirty-six of Dumortier's and of Lindley's elaborate classi- fications of fruits have never found employment in systematic botany, and doubtless never will be used. Yet a detailed carpological classification has its uses for the student. Among the more recent attempts are the successive ones of Dickson, McNab, and Masters. See Nature, iv. 347 (also in Trimen's Jour. Bot. 1871, 310), iv. 475, and v. 6. 292 THE FRUIT. dehiscent and indehiscent kinds.1 Theoretically, each kind may be divided into those of a simple and those of a compound pistil, and some would make the primary division on this character. Some also would separate fruits with adnate or superior calyx from those free of all such combination. But in practice these differences can seldom be indicated by substantative names. The name of berry is equally applicable to the fruit resulting from the single carpel of Actsea, the sj'ncarpous ovary of the grape, and the similar ovary with adnate calyx of a gooseberry and cranberry. It should be understood that the kinds shade off one into another most freely. 556. Dehiscent Fruits (543), or Pods, are distinguishable into apocarpous, or of single carpels, and syncarpous, of more than one carpel, i. e. the first of a simple, the second of a compound pistil. The first kind is mainly represented by the Follicle and the Legume ; the second, by the Capsule and its modifications. 557. A Follicle is a pod formed of a simple pistil, and dehiscent by one suture (this almost always the ventral or inner suture) alone ; as in the Larkspur, Columbine, Peony, and Marsh-Marigold (Fig. 618) ; also in Milkweed and Dog- bane. There may be several follicles or only one to a flower, even in the same genus, as in Larkspurs, Cimicifuga, &c. In Magnolia (Fig. 648-650), fleshy carpels become follicles dehiscent by the dorsal suture. 558. A Legume is the pod formed of a simple pistil which is dehiscent by both sut- ures (as in the Pea, Fig. 619), so dividing into two pieces or valves. (544.) This is the fruit of the Pulse Family, accordingly named 61g Leguminosae (Leguminous plants) : indeed, the name of legume is restricted to the fruits of this family, and in descriptive botany is extended to all the modifications 1 Dr. Masters's modification of Dickson's and McNab's classification of simple fruits, as to primary kinds, is into 1. Nuts, or Achcenocarps, dry and indehiscent ; 2. Pods, or Reymacarps, dry, dehiscent ; 3. Stone-fruits, orPyrenocarps, fleshy without, indurated within, indehiscent ; 4. Berries, or Sarcocarps, fleshy throughout, indehiscent. PIG. 618. A dehiscent follicle of Marsh-Marigold, Caltha palustris. FIG. 619. Legumo of a Sweet Pea, already dehiscent. 620. Loment of a Desmodium. ITS KINDS. 293 which that order presents. Some of these, in fact, are in- dehiscent and reduced to akenes ; some break up at maturity into one-seeded indehiscent articulations or joints, which are dispersed as if they were so many seeds. A legume of the latter kind takes the special name of LOMENT, Lat. Lomentum. (Fig- 620.) In Mimosa (Sensitive-plant, &c.), such articulations de- hisce into two valves. They also fall away from the sutures, or from a persistent marginal border of them, or in some cases the valves thus fall away entire. The persistent frame which remains has been called a REPLTJM, an architectural word, here taken in the sense of door-case. 559. A Capsule is the pod, or dehiscent fruit, of any compound pistil. When regularly and com- pletely dehiscent, as already stated (544), the pod splits lengthwise into pieces or voices. The modes of regular dehiscence are illus- trated in Fig. 612- 617. Two modifica- tions of the capsule have received distinc- tive names which are in common use, viz. the Pyxis and the Silique. 560. A Pyxis or Pyxidium is a dry fruit which opens by a circular line, cutting off the upper part as a lid ; i. e., the dehiscence is circumscissile. (553, Fig. 621.) In the Purslane, Pimpernel, Henbane, and Plantain, the pyxis is a capsule; in Amaranths (Fig. 637) it is a * ™™ » ' utricle ; in Jeffersonia (Fig. 622) it is a modi- fication of the follicle, being of one carpel which dehisces transversely, and not all round, so that the lid remains attached. 561. A Silique is a narrow two-valved capsule, with two pari- etal placentae, from which the valves separate in dehiscence ; as in plants of the Cruciferous or Mustard family (Fig. 623), to the fruit of which this term is restricted. Usually, a false partition is stretched across between the two placentae, render- FIG. 621. Pyxis of Purslane, Portulaca oleracea, the top separating entirely and falling away. FIG. 622. Pyxis-like follicular fruit of Jeffersonia diphylla; the lid remaining attached dorsally. FIG. 623. Silique of Cardamine, in dehiscence. 624. Silicle of Capsella or Shepherd's- Purse, lateral view, and an oblique view of the same with one valve removed. 294 THE FRUIT. ing the pod two-celled in an anomalous manner. A SILICLE (Silicula, diminutive of siliqua) is merely a short silique, the length of which does not more than twice or thrice surpass the breadth ; such as that of Shepherd's-Purse (Fig. 624) . and of Lunaria, Candytuft, &c. 562. Indehiscent Dry Fruits are almost always one-seeded or very few-seeded. If numerous, the seeds thus placed would not be dispersed. The ordinary kinds are strictly one- seeded, and in common language are often con- founded with seeds. The ways in which such fruits are dispersed are various. In the following case, the adaptation of the pericarp to dispersion by wind distinguishes the species of fruit. 563. The Samara, sometimes called in English a Key, is an indehiscent one-seeded fruit provided with a wing. In the White Ash the wing is terminal (Fig. 625) ; in other species the whole fruit is wing-margined ; in Birch and Elm (Fig. 626) the wing sur- rounds the body of the pericarp ; and the Maple fruit is a double samara or pair of such fruits, con- 626 spicuously winged from the apex. 564. Akene (Lat. Achenium) is a general name for all the one-seeded, dry and hard, indehiscent and seed-like small fruits, such as are popularly taken for naked seeds. But that they are true pistils, or ovaries ripened, is evident from the style or stigma they bear, or from the scar left by its fall ; and a section brings to view the seed within, provi- ded with its own proper integuments. The name has been restricted to the seed-like fruits of simple pistils, such as those of the Buttercup (Fig. 628, 629) , Anemone, Clematis, and Geum. The st3*le in some species of the latter remains on the fruit as a long and feather}' tail, in others as a short and hooked one, both being agents of dissemination. The grains of the strawberry (Fig. 653) are also akenes. The name is extended to all one-celled seed-like fruits resulting from a FIG. 625. Samara or key of White Ash. Fraxinus Americana. 626. That of WTiite Elm. Ulmns Americana. 627. Double samara of Red Maple. Acer rubrum. FIG. 628. Achenium of a common Buttercup. 629. Vertical section, showing the seed within. ITS KINDS. 295 compound ovary, and even when invested with an adnate calyx- tube. Of the latter is the fruit of Compositae. (Fig. 630-635.) Here the tube of the calyx is incorpo- rated with the surface of the ovary ; and its limb or border, obsolete in some cases (Fig. 630), in others appears as a crown or cup (Fig. 631), or set of teeth or of scales (Fig. 632, 633), or as a tuft of bristles or hairs (Fig. 634, 635), &c., called the PAPPUS. In the Lettuce and Dandelion (Fig. 635), the achenium is rostrate, or beaked, i. e. its summit is extended into a slender beak. An akene with adnate ealyx has been termed a CYPSELA. 565. The Utricle is the same as the akene, only with a thin and bladdery loose pericarp, like that of Goosefoot. (Fig. 636.) This thin coat sometimes bursts irregu- larly, discharging the seed. In the true Amaranths, the utricle opens by a circular line, and the upper part falls as a rid, converting the fruit into a small pyxis (560), — a transition form. (Fig. 637.) 566. A Caryopsis or Grain differs from the utricle or akene in having the seed completely filling the cell, and its thin coat firml}- consolidated throughout with the very thin pericarp ; as in wheat, Indian corn, and all other cereal grains. Of all fruits this is the kind most likely to be mistaken for a seed. 567. A Nut is a hard, one-celled and one-seeded, indehiscent fruit, like an achenium, but larger, and usually produced from an ovary of two or more cells with one or more ovules in each, all but a single ovule and cell having disappeared during its growth (536) ; as in the Hazel, Beech, Oak (Fig. 638), Chest- FIG. 630. Achenium of Mayweed (no pappus). 631. That of Cichory (its pappus a shallow cup). 632. Of Sunflower (pappus of two deciduous scales'). 633. OfSneezeweed (Helenium), with its pappus of five scales. 634. Of Sow-Thistle, with its pappus of delicate downy hairs. 635. Of the Dandelion, tapering below the pappus into a long beak. FIG. 636. Utricle of Chenopodium album , or common Goosefoot. ft!?. Utricle of an Amaranth, by transverse dehiscence becoming a pyxis. 296 THE FRUIT. nut, and the like. The nut is often enclosed or surrounded by a kind of involucre, termed a Cupule ; such as the cup at the base of the acorn, the bur of the chestnut, and the leaf- like covering of the hazel-nut. The name Glans (sometimes Gland in English) is technically applied to such nuts, this being their classical Latin name. 568. The fruit of the Walnuts and Hickory is apparently a kind of drupaceous nut, or something intermediate between a stone fruit and a nut. But certain monstrosities give reason for supposing that the seeming exocarp (541), which in Hickory hardens and at maturity dehisces hi four valves, is of the nature of an adnate involucre. The cocoanut is a sort of fibre-drupaceous nut. 569. Nutlet, or in Latin form NUCULE (Nuculd), is sometimes superfluously employed in a literal sense, as a diminutive nut.1 Of late it has acquired a good and fairly legitimate use as the name of the seed-like, or rather akene-like, closed parts or lobes, of crustaceous or other hard texture, into which certain bilocular or plurilocular pericarps separate at maturity, i. e. for the seg- ments of a schizocarp, 571, which resemble akenes.2 These are sometimes carpels, sometimes half-carpels, as in Verbena, also in Borraginaceae and Labiatae (in which the segments are greatly separated in the ovary), and sometimes, as in Nolana, they are portions of compounded carpels which have been exceedingly multiplied by chorisis. 570. There are complete transitions between dry nutlets, with a thin and herbaceous epicarp, and the pyrence (574) or stony inner portion of such carpels when drupaceous or composing a drupe of two or more stones. It is therefore a hardly incongru- ous and very convenient use which extends the term nutlet to include these small seed-like stones also, as, for example, to those of Holly, Bearberry, Hawthorn, and the like. 571. The pair of achenium-like or often samara-like carpels, 1 Nut and akene, between which there is no fixed distinction, will cover this ground. The fruit of Cyperaceae, for instance, is truly an achenium, if this name is ever to be used (and it now commonly is) for any other than a monocarpellary fruit. It is often termed a nut, sometimes a nutlet, and by a late writer, Boeckler, a caryopsis. 2 Cocci (sing. Coccus, from a Greek word for kernel) is another name for fruit-carpels, or separating lobes of a dry pericarp, as well for dehiscent ones (of Euphorbia) as for indehiscent. Hence such lobed or partible fruits are said to be dicoccous, tricoccous, &c., according to the number of lobes or carpels. FIG. 638. Acorn (nut) of White Oak, with its cup, or cupnle. ITS KINDS. 297 united by their inner face but separating entire at maturity, which constitute the fruit of Umbelliferse, takes the name of CREMOCARP (Lat. Cremocarpium) ; and the halves are called MERICARPS. These names it may sometimes be convenient to use ; yet it is not advisable to have special names for the fruits of particular families ; and mericarp is here synonymous with carpel. For dry fruits in general (or such as become dry) which are composed of two or more carpels, and which at matu- rity split up or otherwise separate into two or more closed one- seeded portions, an appropriate recent name is that of SCIIIZOCARP. The component carpels of such a fruit were long ago named Car- cerules (carceruli, little prisons) by Mirbel. 572. Fleshy Fruits, which from their texture are natural!}' indehiscent, may be either fleslry throughout, or with a firm rind or shell, or fleshy externally and hard or stony internally. Of the latter, the t}-pe is 573. The Drupe or Stone Fruit proper (Fig. 639), that of the cherry, plum, and peach. True drupes are of a single carpel, one-celled and one-seeded (or at most two-seeded) , in the ripening of which the outer portion of the pericarp becomes fleshy or pulpy, and the inner stony or crustace- ous, i. e. divides into sarco- carp and putamen. (541.) But the name is extended to pericarps of similar texture resulting from a compound pistil, either of a single cell, as in Celtis, and (by abortion) in the olive, or of two or several cells, as in Cornus, Rhamnus, &c. The several pericarps of the aggregate blackberry and raspberry are diminu- tive drupes or DRUPELETS. 574. Small drupes are often confounded with berries, and the stone or stones taken for seeds. Especially is it so in drupes or drupaceous fruits of more than one cell, ripening into separate or separable hard endocarps or stones, each filled by a seed.1 Bearberries (Arctostaphylos) and Huckleberries (Gaylussacia) are good illustrations of this. The seed-like endocarps of this 1 The term Acinus, the original name of such a berry as a grape, has been used in descriptive botany for a small drupe or drupelet, and the ripened carpels of Rubus have been termed acini or acines, but without discriminating them from berries. FIG. 639. Vertical section of a peach. 640. An almond; in which the exocarp, the portion of the pericarp that represents the pulp of the peach, remains juiceless, and at length separates hy dehiscence from the endocarp. or shell 298 THE FKTJIT. sort are PYREN^E ; and the fruits are dipyrenous, tripyrenous, tetrapyrenous, &c., according as they contain two, three, or four pyrenae. When the sarcocarp is thin and dries up at maturity, these pyrenae pass by gradations into nuculae (569) or nutlets : hence pyrenae are not uncommonly in English descriptions called nutlets or nucules. 575. The Pome (Fig. 641, 642) is the name of the apple, pear, and quince. These are fleshy fruits, composed of two to several carpels (rarely by abortion only one) , of parchment-like or (in Hawthorns) bony texture, enclosed in flesh which morphologically belongs to adnate calyx and receptacle ; as may be ap- prehended by comparing a rose-hip (Fig. 407, in flower) with an apple or a pear. Of the quince, the whole flesh is calyx or hypanthium (395) ; in the apple and pear, the inner or core-portion of the flesh is of the nature of disk, investing the carpels. In the fruit of Hawthorns, the carpels become bony pyrenae (574), and so the fruit is drupaceous, is indeed nothing more than a syncarpous drupe. In Eriobotrya, or Cumquat, the carpels becoming very thin and membranaceous, the pomaceous fruit is in fact a kind of berry. 576. The Pepo, or Gourd-fruit (Fig. 643), of which the gourd and squash are the type, and the melon and cucumber equally familiar illustrations, is the char- acteristic fruit of Cucurbitaceae, fleshy internally and with a hard or firm rind, 11 or part of which is referable x> the adnate calyx complete!1 incorporate with the ovary. This is either one-celled with three broad and revolute parietal placentae, or these pla- centae, borne on thin dissepiments, meet in the axis, enlarge, and spread, unite with their fellows on each side, and are reflected to the walls of the pericarp, next which they bear their ovules. As the fruit enlarges, the seed-bearing placentae usually cohere with the walls, and the partitions are obliterated, giving the FIG. 641. Pome or apple in transverse section. 642. Quince in vertical section: the inner flesh answering to disk in the apple and pear is here wanting. FIG. 643. Section of the ovary of the Gourd. 644. Diagram of one of its constituent carpels. ITS KINDS. 299 appearance of a peculiar abnormal placentation, which the study of the ovary readily explains. In the watermelon the edible pulp all belongs to the greatly developed placentae. Fruits of this family in which the rind also is soft at maturity are true berries. 577. The Hesperidium (orange, lemon, and lime) is the fleshy fruit of a free mai^-celled ovary with a leathery rind, and is a mere variety of the berry. The name is applied only to fruits of the Orange tribe. 578. The Berry (Lat. Bacca) comprises all simple fruits in which the pericarp is fleshy throughout. The grape, gooseberry, currant, cranberry (Fig. 645), banana, and tomato are familiar examples. The first and last consist of an ovary free from the calyx ; in the others, calyx and ovary are combined by adnation. 579. Aggregate Fruits are those in which a cluster of carpels, all belonging to one flower, are crowded on the receptacle into one mass, as in the raspberry and black- berry taken as a whole. (Fig. 646.) The}' may be aggregates of an}- kind of simple fruits. But when dry and not coherent, the mass would simply and properly be described as a head or spike of carpels, more commonly of akenes, as in Ranunculus, Ane- mone, &c. Yet when numerous carpels thus compacted become fleshy, and sometimes more or less coherent, the aggregate may need to be taken into account. The best name for it is that of SYNCARPIUM, or in English form SYNCARP. But the term has been applied to multiple fruits as well.1 In II}-drastis, the numerous carpels imbricated on the upper part of the torus are baccate, that is, become berries ; in a raspberry, the seemingly baccate grains are drupaceous (being drupelets, 573) , , C If 1 The si/ncarp which is a gynoeciutn might be designated a simple syn- carpium ; that which is an inflorescence, a complex syncarpium, which may be biflorous, pauciflorous, or multiflorous. FIG. 645. The larger Cranberry, Vaccinium (Oxycoccus) macrocarpon ; the berry transversely divided. FIG. 646. Vertical section of half of a blackberry ( of Kubus villosus ), enlarged ; and, 647, of one of its drupelets more magnified. 300 THE FRUIT. and, slightly cohering together (though without organic union) , they fall as one boch' from the conical dry torus at maturity. It is the same in blackberries or bramble-berries (Fig. G46, 647), except that the drupelets persist on the torus, which partakes of the juiciness.1 In the aggregate fruit of Magnolia (Fig. 648-650), such carpels, imbricated over one another, cohere more or less at all contiguous parts, and become drupaceous ; never- theless, at maturity each opens dorsally, allowing the seeds to fall out : in age it dries and hardens, and also separates from its connec- tions, and so be- comes a follicle, but with the remark- able peculiarit}' of dorsal instead of ventral dehiscence. (Fig. 650.) In Li- riodendron, a tree of the same family, 648 649 ero such carpels are dry and indehiscent throughout ; and they largely consist of long and flat styles, imbricated in a cone, but separating from each other and from the slender torus at maturity, when each becomes a samara. 580. Accessory or Anthocarpons Fruits are those of which some conspicuous portion of the fructification neither belongs to the pistil nor is organically united with it, except by a common insertion. The part thus imitating a fruit, while it is really no part of the pericarp, is sometimes called a Pseudocarp, or an \nthocarp or Anthocarpiutn. This condition may occur either ;ii simple, in aggregate, or in multiple fruits. 1 The aggregate fruit like that of Rubus (named by some Conocarpium, by others an JEterio, Erytlvrostomum, &c.) was termed by Dumortier a Drupe- turn. A similar aggregation of baccate carpels he termed a Baccetum ; of follicles, a Follicetum, &c. All such names may look well in a system ; but they are both superfluous and unmanageable in phytography. FIG. 648. Aggregate fruit of Umbrella-tree, Magnolia Umbrella, reduced in size; a seed from a lower dehiscent carpel hangs on a thread, consisting of a tuft of extensile spiral ducts unravelled. 649. Same in longitudinal section. 650. One of the carpels detached, at full maturity, dried up, doreally dehiscent, exposing the pair of seeds of the nnturr.l size. ITS KINDS. 301 581. Gaultheria procumbens, the aromatic Wintergreen (Fig. 651, 652), affords a good example of the first. Its seeming berry (the checkerberry) , with summit crowned by the tips of the calyx-lobes, well imitates the true berry of a Vaccinium, such as that of Fig. 645. But it comes from a flower with thin calyx, underneath and free from the ovary. Its fruit is really a capsule : in the process of fructi- fication, the calyx enlarges, becomes succulent, completely encloses the capsule or true fruit, yet without adhering to it, and in ripening counterfeits a red berry. So in Shepherdia, or Buffalo Berry, the seeming sarco- carp of a drupe is realty* a free calyx, accrescent and succulent, enclosing an akene. So, also, the apparent achenium or nut of Mirabilis, or Four-o' clock, and of its allies, is the thickened and indurated base of the tube of a free calyx, which contracts at the apex and encloses the true pericarp (a utricle or thin akene) , but does not cohere with it. 582. Likewise the torus, although not con- spicuous, ma3*be said to be an accessory part of the aggregate fruit of the Blackberry or Bramble (579) : it becomes the solely con- spicuous and the sole edible part of a straw- berry (389, Fig. 406, 653), the akenes or true fruits dispersed over the surface being apparently insignificant. Equally in many multiple fruits the conspicuous flesh belongs to receptacle (either torus or rhachis) , to calyx, or even in part to bracts, or to all these parts combined, as in a pine-apple. 583. Multiple or Collective Fruits l are those which result from the aggregation of several flowers into one mass. The simplest of these are those of the Partridge-Berry (Mitchella,Fig.467), 1 Collective is the preferable name. The term multiple was applied by DeCandolle to what are here (following Lindley) called aggregate fruits ; and the aggregate fruits of DeCandolle are here called multiple or collective. - Moreover, the distinction between accessory or anthocarpous and collective or multiple fruits was not recognized by Lindley, who combined the two in his original "Introduction to Botany." In this work four classes are given : 1. Fruit simple, APOCARPI ; 2. Fruit aggregate, AGGREGATI ; FIG. 651. Forming capsule of Gaultheria procumbens, with enlarging calyx partly covering it. 662. Same, more advanced, and in longitudinal section. FIG. 653. Vertical section of half a strawberry. Compare with Fig. 406. 302 THE FRUIT. and of certain species of Hone}'suckle, formed of the ovaries of two blossoms united into one fleshy fruit. The more usual sorts are such as the pine-apple, mulberry, and the fig. These are, in fact, dense forms of inflorescence, with the fruits or floral envelopes matted together or coherent with each other ; and all or some of the parts succulent. The grains of the mulberry (Fig. 654-656) are not the ovaries of a single flower, like those of the blackberry, which it superficially resembles : they belong to as many separate flowers ; and the pulp pertains to the calyx, not to the pericarp, which is an akene. So that this, like most multiple fruits, is anthocarpous as well as multiple. Similarly, the mostly indefinite fructiferous masses of Strawberry Elite may resemble strawberries ; but the pulpy part is the calyx of many flowers, not the succulent receptacle of one. In the pine-apple, the flowers are spicate or capitate on a simple axis, which grows on beyond them into leafy stem ; this when rooted as a cutting 3. Fruit compound (ovaria compound), SYNCARPI ; 4. Collective fruits, Anihocarpi. Later, in his "Elements of Botany," Lindley reduced the classes to two: 1. Simple Fruits, those proceeding from a single flower; 2. Multiple fruits, those formed out of several flowers. FIG. 654. A mulberry, young. 655. One of the fleshy grains at flowering time, show- ing it to be a pistillate blossom with fleshy calyx. 656. The same later, with the succu- lent sepals in transverse section. FIG. 657. A young fig. 658. Longitudinal section of the same later, but in flowering time. 659. A small slice, magnified, showing some of the flowers. ITS KINDS. 303 bears another pine-apple, and so on : the constituent flowers have through immemorial propagation in this way become sterile and seedless, and all its parts, along with the bracts and the axis of the stem, blend in ripening into one fleshy and juicy mass. Few fruits of this class have ever been technically named, at least with names which have come into use. But the two following deserve special appellations, although only the latter is familiar either in ordinary language or in descriptive botany. 584. The Syconium or Hypanthodium, the Fig fruit. (Fig. 657- 659.) This results from a multitude of flowers concealed in a hollow flower-stalk, if it may be so called, which becomes pulpy and edible when ripe ; and thus the fruit seems to grow directly from the axil of a leaf, without being preceded by a blossom. The minute flowers within, or some of them, ripen their ovaries into very small akenes, which are commonly taken for seeds. The fig is to the mulberry what a rose-hip is to a strawberry. (389, Fig. 406, 407.) It is further explained by a comparison with a near relative of the Fig-tree, Dorstenia, in which similar flowers cover the upper surface of a flat peltate disk. This disk or plate sometimes becomes saucer-shaped by an elevation or incurvation of the margin. A greater degree of this would render it cup-shaped, or even pitcher-shaped ; from which it is a short step to the contraction of the mouth down to the small orifice which is found in the fig. 585. The Strobile or Cone (Fig. 660) is a scaly multiple fruit, resulting from the ripening of certain sorts of catkin. The name is applied to the fruit of the Hop, where the large and thin scales are bracts ; but it more especially belongs to the Pine or Fir cone, the peculiar fruit of Conifers (507), in which naked seeds are borne on the upper face of each fructiferous scale (Fig. 661), or some- times in their axils. Such a cone when spherical, and of thickened scales with narrow base, as that of Cypresses, has been termed a GALBULUS, an unnecessary name. The galbulus of Juniper is a FIG. 660. Strobile or Cone of a Pitch Pine, Pinus rigicla. 661. Inside view of on« of the scales, showing one of the winged seeds, and the place from which the other, 6fi2, has been detached. 304 THE FRUIT. remarkable transformation into a seeming berry ; the few scales cohering with each other as they grow and becoming fleshy at maturity, completely enclosing a few bony-coated seeds. 586. A Synopsis of the kinds of Fruit, as characterized in this chapter, is appended. The analysis extends only to simple fruits. For there are no commonly used special names of > kinds of Aggregate (579), Accessory (580), or Multiple (583) fruits, except that of Strobile. SIMPLE FRUITS are Dry and dehiscent, raonocarpellary, Opening by one (chiefly the ventral) suture, FOLLICLE. Opening by both sutures, LEGUME. Or transversely jointed, LOMENT. Dry and dehiscent, bi-pluri-carpellary, CAPSULE. When its dehiscence is circumscissile, PYXIS. When dehiscent by two valves from two parietal placentae, . . SILIQUE. A short and broad silique, SILICLE. Dry and bi-pluri-carpellary, splitting into one-seeded carpels, . SCHIZOCARP. The dimerous schizocarp of Umbelliferae, CREMOCARP. Each of its halves or carpels, HEMICARP or MERICARP. The akene-like or nut-like parts into which Schizocarps generally divide, NUCULES or NUTLETS. Dry and indehiscent, one-celled, one-two-seeded, Winged SAMARA. Wingless, and with the Thin pericarp consolidated with the seed, CARYOPSIS Thin pericarp loose and not filled by the seed, UTRICLE. Thick or hard pericarp free from the seed, Small, from a one-celled one-two-ovuled ovary, AKENE or ACHENIUM. Larger, mostly from a two-several-celled and ovuled ovary, . NUT. Nut borne in a cupule or involucre, GLANS. Fleshy and indehiscent, Heterogeneous in texture, having A stone (putamen) or nutlets within an exterior sarcocarp, . DRUPE. Papery or cartilaginous carpels in an inferior sarcocarp, . . POME. A harder or firm rind or exterior, and soft interior, From an inferior ovary (confined to Gourd Family), . . . PEPO. From a superior ovary (confined to Orange Family), HESPERIDIUM. Homogeneous, fleshy throughout, BERRY. THE SEED. 305 CHAPTER VIII. THE SEED. ~P £• oil THE SEED is the fertilized ovule (515), with embryo formed within it. It consists, like the ovule, of a nucleus or kernel, enclosed by integuments. The seed-coats are those of the ovule, viz. two, or sometimes only one, in certain plants none. Occasionally an accessory coat appears after fertiliza- tion ; and certain appendages may be produced, as outgrowths from some part of its surface or from its base. The nucleus or kernel is composed either of the embryo alone, or of a nutritive deposit in addition. (19-41.) All the parts of a seed are in- ' dicated in Fig. 663. 588. The SEED-STALK or PODOSPERM, when there is one, is the funiculus of the ovule (516), and retains this name. So 7 also do the CHALAZA, RHAPHE, and HILUM ; the latter being the scar left by the separation of the seed from its funiculus or directly from the placenta. The foramen of the ovule, now closed, is the MICROPYLE of the seed. 589. The terms which denote the char- acter of the ovule, such as orthotropous, campylotropous, amphitropous, and anatropous, apply equally to the resulting seed. 590. Seed-Coats. The integuments of the seed answer to the primine and secundine of the ovule. The main seed-coat is the exterior integument of the ovule when there is more than one. Being the most firm coat, and not rarely crustaceous in texture, it takes the name of TESTA, which is equivalent to seed-shell. It has also been named SPERMODERM (seed-skin) , and sometimes Episperm. The latter name (meaning upon the seed) is best applied to the pellicle or outer layer, sometimes a thick one, which the testa of certain seeds forms. The testa is extremely various in form and texture, is either close and conformed to FIG. 663. Vertical magnified section of the (anatropous) seed of the American Lin- den; with the parts indicated, viz. the hilum (a); testa (6); tegmen (c); albumen (d); embryo (e). 664. Vertical section of the orthotropous seed of Helianthemum Car.a- dense, with its funiculus, a. 306 THE SEED. the nucleus, or loose and cellular (as in Pyrola-seeds) , or vari- ously appendaged. 591. The inner coat, called TEGMEN and sometimes ENDO- PLEUEA, when present is always conformed to the nucleus, and is thin or soft and delicate. Sometimes it is inconspicuous through cohesion with the nucleus or with the inner surface of the testa. In ovules of one coat it is necessarily wanting. 592. Appendages or outgrowths 'of the testa generally have reference to dissemination. Two characteristic kinds of such appendages are the wing and the coma, both pertaining only to the seeds of dehis- cent fruits and calculated, by rendering seeds buoyant, to facilitate dispersion by the wind. The wing of a Pine-seed (Fig. 661, 662) is a part of the carpellary scale upon which the two ovules grew. In Trumpet Creeper (Fig. 665), an entire wing surrounds the body of the seed. In the related Catalpa (Fig. 666), it is mainly extended from the two ends, and almost dissolved into a coma, the name given to the tuft of soft hairs like that which forms the down at one end of the seed of Milkweed (Fig. 667), and of Epilobium, and at both ends in several Apocynacese. In the Cotton-plant, very long and soft hairs, admirably adapted for spinning, thickly cover the whole seed- ecr 666 coat. The wing and coma of seeds are functional!}' identical with the wing and the pappus of the pericarp in the samara and the akenes of Composite (563, 564), but morphologically quite unlike them. 593. There are other (mainly microscopic) structures on some seed-coats which come usefully into play in arresting farther dispersion at a propitious time or place. In many but not all Polemoniacese (notably in Collomia), in certain Acanthaceae, such as Ruellia tuberosa (and equally in certain Composite of the Senecio tribe and in Salvias, &c., among Labiatse, where this structure is transferred to akenes and nutlets) , the testa is coated with short hairs, which when wetted burst or otherwise open and discharge along with mucilage one or more very atten- FIG. 665. Winged seed of Trumpet Creeper, Tecoma radicans. 666. That of Catalpa, becoming comose: the body divided lengthwise through the embryo. FIG. 667. Comose seed of Milkweed, Asclepias Cornuti. ITS COATS AND APPENDAGES. 307 uated long threads (spiricles) which were coiled within. These, protruding in all directions and in immense numbers, form a limbus of considerable size around the seed, and evidently must serve a useful end in fixing these scja"! and light seeds to the soil in time of rain, or to moist ground, favorable to germination. In cress and flax-seed, the abundant mucilage developed when wetted comes from the gelatination of epidermal cell- walls, and subserves a similar use. 594. While the testa in many seeds is hard and crustaceous or bony, imitat- ing the pericarp of a nut, in others (such as Pseonia) it becomes berry-like (baccate), and in Magnolia, drupaceous.1 (Fig. 6G8- 671.) These may also be regarded as adaptations for dissemi- nation, here by the agency of birds, attracted by bright coloring and edible pulp. 595. The rhaphe of an anatropous seed (shown in Fig. 681, 685) is sometimes so salient as to form a conspicuous appen- T* dage, as in Sarracenia, Fig. 672. Again it may be wholly i See article On the Structure of the Ovule and Seed-coats of Magnolia, in Jour. Linn. Soc. ii. 106, from which the accompanying figures and Fig. 589-597 are reproduced. FIG. 668. Forming seed (one eighth of an inch long) of Magnolia Umbrella; the rhaphe toward the eye. 669. Magnified view of the same divided lengthwise through the rhaphe ; the outer coat, a, beginning to form a hard inner layer, a/. Within and distinct from this is the inner coat (6), immediately enclosing the nucleus, c. The oppo- site side of the testa is thicker on account of the rhaphe, in which d indicates the cord of spiral ducts. FIG. 670. A nearly full-grown seed, of the natural size. 671. Longitudinal section, emargeo, snowing the crustaceous or stony inner stratum of the testa well developed: the parts lettered as in Fig. 669. 672. A transverse section in the same position. 308 THE SEED. inconspicuous, as in the ripe seed of Magnolia, where it is at length completely merged and imbedded in the fleshy drupaceous testa, as shown in Fig. 670-672. . 596. Crest-like or other appendages are not uncommon either on the rhaphe or at the hilum. These are outgrowths produced during the deveiopment of the ovule into the seed. In Sangui- naria, such a crest develops from the whole length of the rhaphe (Fig. 673) : in Dicentra, Corydalis (Fig. 674), &c., from some part of it, mostly from its base next the hilum, or from the hilum itself, or even from just below it. Such an appendage, especially when attached to the base of the seed, is named a STROPHIOLE. A similar and commonly a wart-shaped appendage in Euphorbia, Ricinus (Fig. 675), &c., is produced by an out- growth of the external orifice of the ovule, the micropyle of the seed. This properly takes the name of CARUNCLE. But the two terms are not always discriminated. By further develop- ment, either of these may give rise, in certain seeds, to an acces- sory covering called 597. The Aril or Arillus. This term, rather vaguely employed by Linnaeus, was first well defined by Gaertner. The true arillus is an accessor}' seed-covering, more or less incom- plete, formed between the time of fertilization and the ripening of the seed, by a growth from the apex of the funiculus (when there is any) at or just be- low the hilum, in a manner similar to that in which the coat or coats of the ovule are formed. That of N3-mphaaa (Fig. 676) is a typical example ; only the arillus is developed from the funiculus at a point distinctly below its apex : here a ring forms, which grows into a cup, and this is soon extended into a sac, loosely enclosing the seed, and open at the top. This is membranaceous ; commonly it is fleshy. When there is absolutely no funiculus, the aril may originate from the placenta, as it does in Podophjllum, in which most of FIG. 672. Anatropous seed of Sarracenia purpurea, with very salient rhaphe. 673. Same of Sangiiinaria or Bloodroot, with rhaphe crested for its whole length. 674. Seed of Corydalis aurea, with crest or strophiole, attached at or near the hilum. 675. Seed (suspended) of Ricinus, with its caruncle. FIG. 676. Seed of White Water-Lily, Nymphsea odorata, in its loose and thin arillus. AKILLUS, ALBUMEN. 309 the pulp of the berry consists of these flesh}- arils, much com- pacted. (Fig. 677, 678.) 598. The laciniate aril of the nutmeg (mace) and, it is said, the bright red and pulpy aril of Euonymus and Celastrus begin in the manner of a ca- runcle, and are formed (mainly if not wholly) of an outgrowth at or around the micropyle. So that, if an orthotropous seed ever developed an aril of this sort, it would 6-7 be seen to begin at the apex of the seed and cover it from above downward. Planchon, who distinguished this from the true aril, gave to it the name of ARILLODE (Arillodium) or False Arillus. 599. The Nucleus, or kernel of the seed, consists of the Albu- men, when this substance is present, and the Embryo. 600. The Albumen, as described in the second chapter (25, &c.) , is the name generally employed by systematic botanists for a store of nutritive matter in the seed outside of the embryo, whatever its chemical composition. It is not here the name of a chemical substance (albumen or albumin) , but of a cellular structure, the cells of which are loaded commonly with starch- grains (as in the Cerealia), more or less mingled with other matters, or else filled with an encrusting deposit of some equiva- lent substance, as in the cocoanut, coffee-grain, &c. The cells in which this deposit is made belong either to the original tissue of the nucleus, or to a new formation within the embryo-sac, mostly to the latter. (503.) 601. Albumen may be said to belong to all seeds in the grow- ing stage. In what are called albuminous seeds it persists and forms either almost the whole kernel, the embryo remaining minute (as in Fig. 23, 54, 680), or forms a large portion of it (Fig. ' 13, 17, 19, 21, 48, 663, 664), or, by the growth of the embryo ' /f ' displacing it, it may in the ripe seed be reduced to a thin stratum or mere lining to the contiguous seed-coat ; or it may disappear altogether, as in the seeds of Maple, Almond, Squash, Pea, and the like, which are therefore said to be exalbuminous. The difference between albuminous and exalbuminous seeds is thai/- the maternal nutritive deposit is transferred to the embryo in FIG. 677. Section of pericarp and placenta of Podophyllum peltatum; the pulp of the latter mainly of the nature of arillus. investing the seeds. 678. The arillus of one seed detached and enlarged, divided lengthwise, showing the seed within. 310 THE SEED. the former during germination, in the latter during the growth of the seed. 602. The albumen was named Perisperm by Jussieu, and Endosperm by Richard (25, note) ; but neither name has in systematic botany displaced the earlier one of Grew and Gi»rt- ner. But both names have recently been brought into use to distinguish between two kinds of albumen, that formed within the embryo-sac, which is specific-all}- termed ENDOSPERM, and that formed without, which takes the name of PERISPERM. This use comports with the etymology of the two words, the former refer- ring to a comparatively internal and the latter to an external portion of the seed or kernel. 603. In most seeds the albumen is endosperm : in Canna it is all perisperin. In Nymphsea and its allies (except Nelum- bium, which has none) most of it is perisperm ; but a thin and condensed layer of endosperm surrounds the embryo, where with the per- sistent embryo-sac (or the apex of it) it forms the fleshy sac in which the embryo is enclosed. It is the same in the Pepper Family (Fig. 679) , except that there is a larger quan- tity of endosperm or inner albumen. 604. When the nucleus of a ripe seed is hollow, as in the cocoanut and nux vomica, the formation of endosperm, which usually begins next the wall of the embryo-sac, has not proceeded so as to fill the cavity. The embryo- sac in the cocoanut attains enormous size, and the cavity is filled by the milky fluid. 605. The texture or consistence of the albumen differs greatly. It is farinaceous or mealy when, consisting mainly of starch- grains, it may readily be broken down into a powder, as in wheat, buckwheat, &c. ; oily, when saturated with a fixed oil, as in poppy-seed ; fleshy, when more compact, but readily cut with a knife, as in the seed of Barberry ; mucilaginous, when soft and somewhat pulp}', as in Morning Glory and Mallow, but when dry it becomes fleshy or harder ; corneous, when of the texture of horn, as in coffee and the seed of Caulophyllum ; and even bony, as in the vegetable ivory, the seed of Phytelephas. It is mostly uniform ; but in the nutmeg, FIG. 679. Longitudinal magnified section of a seed of Black Pepper ; showing the large episperm, the small endosperm in the persistent embryo-sac, and in this the minute embryo. PIG. 680. Longitudinal section of a seed of the so-called Papaw, Asimina triloba, with ruminated albumen and minute embryo. THE EMBRYO. 311 in the seeds of Asimina (Fig. 680) and all the Custard- Apple family, it is marked by transverse lines or divisions (caused by inflexions or growths of the inner seed-coat) , giving a section of it either a marbled appearance, or as if it had been slit by incisions : it is then said to be ruminated. 606. The Embryo,1 being an initial plantlet or individual of a new generation, is of course the most important part of the seed. To its production, protection, and support, all the other parts of the fruit and flower are subservient. 607. In an embryo of full development, namely, one in which all the parts are manifest antecedent to germination, these parts are the Caulicle, otherwise called Radicle, the Cotyledons, and the Plumule. (20, 30.) The first is the initial axis or stem, a primary internode ; the second consists of the leaves of the primar}' node ; the third is a beginning of a farther growth which is to develop more stem and leaves. Such an embryo is usually unaccompanied by albumen, having in the course of its growth taken into itself (mostly into the cotyledons) the provision which in other seeds is mainly accumulated external to it until it is drawn upon in germination. 1 The word Embryo or Embryon was applied to this body in plants by Bonnet (Considerations sur les Corps organise'es),in 1762, and was introduced into systematic botany at about the same time (1763) by Adanson : it was taken up by Gaertner in 1788. Jussieu in the Genera Plantarum (1789) held to the term Corculum (the cor seminis) which came down from Caesalpinus. Being the germinal part of the seed, the embryo of the plant, like that of the animal, is in general language often called the Germ. FIG. 681. Seed of a Violet (anatropous), enlarged; with hilum or scar (a), rhaplie (6), and chalaza (c) indicated. 682. Vertical section of the same, showing the straight embryo in the axis of the mealy albumen. FIG. 683. Vertical section of the (orthotropous) seed of Buckwheat, showing the embryo folded round in the mealy albumen. FIG.. 684. Vertical section of the (anatropous) seed of Elodea Virginica, the embryo completely filling the coats. FIG. 685. Seed of Delphinium tricorne (anatropous). enlarged; the hilum, the rhaphe, and the chalaza lettered as in Fig. 681. 686. Vertical section of the same with c, the chalaza, rf, the testa, e, the tegmen,/, the albumen, g, the minute embryo near the hilum (a). FIG. 687. Embryo of the Pumpkin, with its short radicle and large and flat cotyle- dons, seen flatwise. 688. A vertical section of the same, viewed edgewise. 312 THK SEED. 608. The opposite extreme is an embryo (as in Fig. 686) which appears as a mere speck in the albumen, but in which close microscopical inspection ma}- commonly reveal some differ- entiation, such as a slight notch at one end (that farthest re- moved from the micropyle) of a dicotyledonous embryo, indicating the future cotyledons. Indeed, in Monotropese, Orobanchacese, and some other parasitic dicotyledonous plants, and in Orchids among the monocotyledonous, the embryo is a globular or oblong particle, with no adumbration of organs whatever antecedent to germination. There are all grades between the most rudimen- tary and the most developed embryos. 609. Under the circumstances of its formation (532), the radicular end of the embr}'o is alwaj's near to and points towards the micropyle of the seed, viz. to what was the orifice of the ovule ; and if the embryo be straight, or merely partakes of the curvature of the seed, the cotyledons point to the opposite extremity, that is, to the chalaza. 610. The position of the radicle as respects the hilum varies with ttie different kinds of seed. In the orthotropous form, as in Helianthemum (Fig. 664) and Pepper (Fig. 679), the radicle necessarily points directly away from the hilum.1 In the anatro- pous form, as in Fig. 663, 682, and 684-686, the extremity of the radicle is brought to the immediate vicinity of the hilum ; and so it is, although in a different way, in the campylotropous seed (Fig. 689, 690) ; while in the amphitropous the radicle points away from the hilum later- ally. As the nature of the ovule and seed may usually be ascertained by external inspection, so the situation 1 Two technical terms, early introduced by Richard to indicate the direc- tion of the radicle (caulicle), or rather its relation to the hilum, are Antitropous, when the embryo directs its radicle away from the hilum, as it must in all orthotropous seeds ; Orthotropous, also homotropous, when directed to the hilum (more strictly to the micropyle close to the hilum), as in anatropous seeds. These two terms are still employed by many botanists, although superfluous when the ovule or seed is stated to be anatropous or orthotropous, &c. And the term orthotropous, so used, is liable to be confused with orthotropous as applied to the ovule. Richard, moreover, termed the embryo amphitropous when curved or coiled, as in duckweed (Fig. 689) and all such campylotropous seeds ; and hetero- tropoiis when neither radicle nor cotyledons point to the hilum, as occurs in the semi-anatropous or amphitropous ovule. Many botanists describe the last by the expression " radicle vague," or, better, " embryo transverse." FIG. 689. Campylotropous seed of common duckweed, magnified. 690 Section ol the same, showing the embryo coiled into a ring around the albumen. THE EMBKYO. 313 of the embryo within, and of its parts, may often be inferred without dissection. But the dissection of seeds is not generally difficult. 611. The direction of the radicle with respect to the pericarp is also noticed by systematic writers ; who employ the terms radicle superior or ascending when this points to the apex of the fruit ; radicle inferior or descending when it points to its base ; centripetal, when turned toward the axis of the fruit ; centrifugal (or peritropous} , when turned toward the sides ; and vague, when it bears no evident or uniform relation of the kind to the pericarp. 612. The position of the embryo as respects the albumen, when that is present, is various. Although more commonly in the axis, it is often excentric, or even external to the albumen, .as in all Grasses and cereal Grains (Fig. 56-61), in Polygonum, &c. When external or nearl}- so, and curved circularly around the albumen, as in Chickweed (Fig. 690) and Mirabilis (Fig. 17), it is said to be peripheric. 613. The embryo ma}* be very variously folded or coiled in the seed. The two cotyledons, instead of plane and straight, may be crumpled ; or they may be simply convolute or rolled up from one edge, as in Calycanthus (Fig. 691) ; or circinately con- volute from the apex, as in Bunias ; or else doubled up and thus biplicately convolute, as in Sugar Maple, Fig. 2. Two modi- fications are more common, and are of such classificatory impor- tance in Cruciferae as to need special reference. Namety, when cotyledons are Incumbent (as in Fig. 692, 693), being so folded that the back of one is laid against the side of the radicle ; and Accumbent (Fig. 694, 695), when the edges of the pair of cotyledons are longitudinally applied to the radicle. These differences were first emplo3red in the classification Cruciferae by FIG. 691. Convolute embryo of Calycanthus, the upper half cut away. FIG. 692. Seed of a Cruciferous plant (Sisymbrium), with incumbent cotyledons, divided. 693. Embryo of the same detached entire. FIG. 694. Seed of a Cruciferous plant (Barbarea) with accumbent cotyledons 695. The embryo entire. 314 THE SEED. Robert Brown, and were adopted as primary and tiibal characters by DeCandolle. 614. As to number of cotyledons, the two types of embrj-o are the Monocotyledonous, with a single cotyledon, i. e. leaves at the first nodes alternate (39) ; and Dicotyledonous, with a pair of cotyledons, i. e. leaves of the first node in the most simple whorl, a pair, in other words, oppo site (21) ; with its modification of Polycotyledonous (38) , the leaves of the first node in whorls of three, four, or more. This occurs with constancy in a majority of Conifene (Fig. 48, 49), occasionally and abnormal!}- in sundry ordinary dicotyledonous species. 615. There are several embryos of the cotyledonous type in which one cotyledon is smaller than the other, viz. the inner one when the embryo is coiled or folded. And in all the species of Abronia (a genus allied to Mirabilis, Fig. 18) this cotyledon is wanting, so that the embryo becomes technically monocotyle- donous. In another genus, the Dodder (Fig. 78, 79), both cotyledons are constantly wanting ; and the plumule shows only minute scales, the homologues of succeeding leaves reduced almost to nothing. 616. Sometimes the two cotyledons are consolidated into one body by the coalescence of their contiguous faces ; when they are said to be conferruminate. This occurs more or less in the Horsechestnut and Buckeye (Fig. 41, 42), and is striking in the seed of the Live Oak, Quercus virens. 617. The general morphology of the embryo and its develop- ment in germination were described at the commencement of this volume. And so the completion of this account of plant, flower, fruit, seed, and embryo brings the history round to the starting point. (12—19, &c.) Having mastered the morphology and general structure of the higher grade of plants, the pupil may go on to the morphology and structure of cells (or Vegetable Anatomy or Histology), and to the study of Cryptogamous Plants in all their grades. TAXONOMY. 315 CHAPTER IX. TAXONOMY. SECTION I. THE PRINCIPLES OF CLASSIFICATION IN NATURAL HISTORY. 618. TAXONOMY, from two Greek words which signify arrange- ment and law, is the study of classification. This is of utmost importance in Natural Historj', on account of the vast number of kinds to be set in order, and of relations (of agreement and difference) to be noted. Botanical classification, when complete and correct, will be an epitome of our knowledge of plants. Arrangement according to kinds, and of special kinds under the more general, is common to all subjects of study. But the classification in Biological Natural History, that is in Botany and Zoology, has a foundation of its own. 619. The peculiarity of plants and animals is that they exist as individuals, propagating their like from generation to genera- tion in a series. Of such series of individuals there are very many kinds, and the kinds have extremely various and unequal degrees of resemblance. There are various gradations, but not all gradations of resemblance. Between some, the difference is so wide that it can be said only that they belong to the same kingdom ; between others, the resemblance is so close that it may be questioned whether or not the}- came from common parents or near ancestors. 620. The recognition of the perennial succession of similar individuals gives the idea of SPECIES. The recognition of un- equal degrees of likeness among the species is the foundation of GENERA, ORDERS, CLASSES, and other groups of species. 621. Individuals are the units of the series which constitute species. The idea of individualit}' which we recognize through- out the animal and vegetable kingdoms is derived from ourselves, conscious individuals, and from our corporeal structure and that of the higher brute animals. This structure is a whole, from which no part can be abstracted without mutilation. Each individual is an independent organism, of which the component parts are reciprocally means and ends. Individuality is a main 316 TAXONOMY. distinction between beings and things ; but, although the tend- ency to individuation begins with life itself, it is completely realized only in the higher animals. 622. In plants, as also in some of the lower animals, individ^ ality is merged in community. No plant (except one reduced to the simplicity of a single cell, of circumscribed growth, and without organs) is ail individual in the sense that a man or a dog is. (16, 156.) The herb, shrub, and tree are neither indivisible nor of definite limitation. Whether their successive growths are to remain parts of the previous plant, or to be inde- pendent plants, depends upon circumstances ; and there is no known limit to budding propagation. 623. There is, however, a kind of social or corporate indi- viduality in those animals, or communities (whichever we call them) of the lower grade which are multiplied by buds or off- shoots as well as by ova, and in which the offspring remains, or ma}' remain, organically connected with the stock. The poly- pidom or polyparium commonly has a certain limitation and a definite form ; and certain polyps may become organs with special functions subordinate to the common weal. This is more largely true in the vegetable kingdom. So that for de- scriptive purposes, and in a just although somewhat loose sense, the herb, shrub, or tree is taken as an individual. But only while it forms one connected body. Offshoots when separately established are equally individuals in this sense. 624. What it is in plants which philosophically answers to the individual in the higher animals is another question, to which various answers have been given.1 Some insist that the whole vegetative product of one seed makes one individual, whether connected or separated (as may happen) into a million of plants. But a common and less strained view restricts the individual to such product only while organically united. Others (of which Thouars at the beginning and Braun at the middle of the present century are leading examples) take each axis or shoot with its foliage to represent the individual, of which the leaves and their homologues are organs, the branches being usually implanted upon the parent axis as this is implanted in the soil, but also equally capable of producing roots by which they may make their own connection with the soil. Still others, on pre- 1 For the history of opinion upon and a full presentation of this topic, see Alexander Braun's Memoir (originally published in the Abhandl. Akad. Wissenschaften zu Berlin, 1853), Das Individuum der Pflanze, &c., and a translation by C. F. Stone in Amer. Jour. Sci. ser. 2, xix. xx. 1855. THE PRINCIPLES OF CLASSIFICATION. 317 cisely similar grounds, carry the analysis a step farther, and regard each phytomer (1C) as the individual. Finally, some, in view of their potentially independent life, take the cells, or units of anatomical structure, to be the true individuals ; and this with sufficient reason as regards the simplest cryptogamous plants. Upon the view here adopted, that plants do not rise high enough in the scale of being to reach true individuality, the question is not whether it is the cell, the phytomer, the shoot/ the tree, or the whole vegetative product of a seed which answers to the animal individual, but only which is most analogous to it. In our view, its analogue is the cell in the lowest grades of vege- table life, the phytomer in the higher.1 But, in botanical de- scription and classification, by the individual is meant the herb, shrub, or tree, unless otherwise specified. 625. Species in biological natural history is a chain or series of organisms of which the links or component individuals are parent and offspring. Objectively, a species is the totality o; beings which have come from one stock, in virtue of that mono general fact that likeness is transmitted from parent to progeny. Among the many definitions, that of A. L. Jussieu is one of the briefest and best, since it expresses the fundamental conception of a species, i. e. the perennial succession of similar individuals perpetuated by generation. 626. The two elements of species are : 1, communny of origin ; and, 2, similarity of the component individuals. But the degree of similarity is variable, and the fact of genetic relationship can seldom be established by observation or historical evidence. It is from the likeness that the naturalist ordinarily decides that such and such individuals belong to one species. Still the like- ness is a consequence of the genetic relationship ; so that the latter is the real foundation of species. 1 For just as successive branches are repetitions and progeny of the parent branch or stem, the phytomers of the branch are repetitions and progeny each of the preceding one, so forming a series of vegetative generations ; and the whole tree might almost as well represent the individ- ual as one of its branches. The phytomer, as well as the branch, is capable of completing itself by producing roots, but is itself indivisible except by mutilation. Least tenable of all is the conception that the whole product of a seed may be taken to represent the vegetable individual. For then individuals increased by buds and division are wholly unlimited both in ex- tent and in duration, so far as observation can show, and a multitudinous race, not only of the present and past, but perhaps in perpetuity, may con- sist of a single individual. There are, indeed, theoretical reasons for infer- ring that a bud-propagated race may not last so long as a seed -propagated species ; but there is no proof of it. See Darwiniana, Art. xii. 318 TAXONOMY. 627. No two individuals are exactly alike ; and offspring of the same stock may differ (or in their progeny may come to differ) strikingly in some particulars. So two or more forms which would have been regarded as wholly distinct are sometimes proved to be of one species by evidence of their common origin, or more commonly are inferred to be so from the observation of fa series of intermediate forms which bridge over the differences. Only observation can inform us how much difference is compat- ible with a common origin. The general result of observation is that plants and animals breed true from generation to genera- tion within certain somewhat indeterminate limits of variation ; that those individuals which resemble each other within such limits interbreed freely,, while those with wider differences do not. Hence, on the one hand, the naturalist recognizes Varieties or differences within the species, and on the other Genera and other superior associations, indicative of remoter relationship of the species themselves. 628. Varieties are forms of species marked by characters of less fixity or importance than are the species themselves. They may be of all grades of difference from the slightest to the most notable : they abound in free nature, but assume particular importance under domestication and cultivation ; under which variations are most prone to originate, and desirable ones are preserved, led on to further development, and relatively fixed. 629. If two seeds from the same pod are sown in different soils, and submitted to different conditions as respects heat, light, and moisture, the plants that spring from them will show marks of this different treatment in their appearance. Such differences are continually arising in the natural course of things, and to produce and increase them artificially is one of the objects of cultivation. Striking as they often are (especially in annuals and biennials) , they are of small scientific consequence. When spontaneous they are transient, the plant either outlasting the modifying cause or else succumbing to its continued and graver operation. But, in the more marked varieties which alone de- serve the name, the cause is occult and constitutional ; the deviation occurs we know not why, and continues throughout the existence and growth of the herb, shrub, or tree, and con- sequently through all that proceeds from it by propagation from buds, as by offsets, layers, cuttings, grafts, &c. 630. Some varieties of cultivation originate in comparatively slight deviations from the type, and are led on to greater differ- ences by strict selection of the most marked individuals to breed from. Most appear as it were full-fledged, except as to THE PRINCIPLES OP CLASSIFICATION. 319 luxuriance or development, more or less under the control of conditions, their origin being wholly unaccountable. They arise in the seed-bed, or sometimes from buds, which as the gardeners say " sport" 1 That is, some seedlings, or some shoots, are unlike the rest in certain particulars.2 631. Most varieties originate in the seed, and therefore the foundation for them, whatever it may be, is laid in sexual repro- duction. But Bud-variation, or the " sporting" of certain buds into characters in branch, flower, or fruit unlike those of the stock, is known in a good number of plants.3 It might also occur in corals, hydras, and other compound animals propagated by budding. Once originated, these varieties mostly persist, like seedling varieties, through all the generations of budding growth, but are not transmitted to the seed. 632. Upon the general principle that progeny inherits or tends to inherit the whole character of the parent, all varieties must have a tendency to be reproduced b}- seed. But the inheritance of the new features of the immediate parent will common!}- be overborne b}' atavism, i. e. the tendency to inherit from grand- parents, great-grand-parents, &c. Atavism, acting through a long line of ancestry, is generally more powerful than the heredity of a single generation. But when the offspring does inherit the peculiarities of the immediate parent, or a part of them, its off- spring has a redoubled tendency to do the same, and the next generation still more ; for the tendency to be like parent, grand- parent, and great-grand-parent now all conspire to this result and overpower the influence of remoter ancestry. Close-breed- ing (398) is requisite to this result. In the natural wild state, varieties — many and conspicuous as they often are — must be much repressed \)y the prevalent cross-fertilization which takes place among the individuals of almost all species. Cultivators and breeders in fixing varieties are careful to secure close breeding as far as this is possible. This has fixed the particular sorts of Indian Corn, Rye, Cabbage, Lettuce, Radishes, Peas, &c., and 1 Both the technical English term, Sport, and its Latin equivalent, Lusus, are sometimes used for bud-variation only, yet as commonly for seedling variation also. 2 Darwin assumes that variation is of itself indefinite or vague, tending in no particular direction, but that direction is wholly given by the elimina- tion in the struggle for life of all but the fittest for the conditions. But what we observe in the seed-bed does not suggest this view. Xaegeli, Braun, and myself incline to the opinion that each plant has an inherent tendency to variation in certain general directions. 8 A list of known bud-varieties is given hi Darwin's Variation of Animals and Plants under Domestication, Chapter xL 320 TAXONOMY. indeed of nearly all our varieties of cultivated annual and biennial esculent plants, as well as of several perennials, many of which have been fixed through centuries of domestication, while others are of recent estabhVhment. What is now taking place with the Peach in this country may convince us that heritable varieties may be developed in trees as well as in herbs, and in the same manner ; and that the reason why most races are annuals or biennials is because these can be perpetuated in no other way, and because the desired result is obtainable in fewer years than in shrubs or trees. Varieties of this fixit}' of character are called G33. Races (Lat. Proles). A race, in this technical sense of the term, is a variety which is perpetuated with considerable certainty by sexual propagation. This distinction of varieties pertains chiefly to botany. In the animal kingdom all permanent varieties must be races. So are all indigenous varieties of plants.1 In most of these, the position of species and variety is more or less arbitrary or accidental, and capable of interchange. What is called the species may be only a commoner or better- known form, or the one first recognized and named by botanists ; whence the other forms as they come to be recognized are made to rank in the books as varieties. Instead of one varying from the other, all the forms have probably varied ages ago from a common type. 634. These varieties of the highest order and most marked characteristics, being perpetuable by seed, have the principal attributes of species. They are a kind of subordinate derivative species. Hence they are sometimes called Subspecies. We judge them not to be so many species, either because in the case of cultivated races we know something of their origin or history, and more of the grave changes which long domestication may bring to pass ; or because the forms, however stable, differ among themselves less than recognized species generally do ; or because very striking differences in the extremes are connected by intermediate forms. And our conclusions, it must be under- stood, "are not facts, but judgments, and largely fallible judg- ments." 2 For while some varieties appear strikingly different, some species are very much alike.8 1 The Horseradish and a few other plants of spontaneous growth, which through long dependence on bud-propagation seem to have lost the power of setting seed, can hardly be called varieties. 2 Darwiniana, 35. 3 Wherefore, since we hardly need the term race in the restricted sense of seed-propagated variety, it is sometimes convenient to use it in the man- ner proposed by Bentham (Anniversary Address to the Linnean Society, THE PRINCIPLES OF CLASSIFICATION. 321 635. One distinction between varieties and species is note- worthy and important, even if it ma}- not serve as a criterion. The individuals of different varieties in plants interbreed as freely as do those of the same variety and are equally prolific. Their union produces 636. Cross-breeds.1 In nature, cross-breeding doubtless re- presses variation or prevents the segregation of varieties into what would be ranked as species. In cultivation and domesti- cation, it is turned to important account in producing intermediate new varieties (cross-breeds) variously combining the different excellencies of two parent individuals or two varieties. Thus the great number of forms produced by variation (especially as to flowers and fruits) have been further diversified, and selected forms improved for special uses by judicious combination. 637. In general, the individuals of distinct species do not interbreed, although many are capable of it. There is great diversity in this regard among plants, some (such as Willows, Verbascums, and Verbenas) interbreeding freely and reciprocally ; some interbreeding in one direction, but not reciprocally ; others, even when ver}- similar, refusing to unite. But, on the whole, there seems to be few nearby related species in which the pollen of the one cannot be made to act upon the ovules of the other by persistent and proper management. Such crossing is an important resource in horticulture. Crossing of species, when successful, produces 638. Hybrids. In these, the characteristics of the two species are combined, sometimes in equal proportions, sometimes with great preponderance of one or the other parent ; and there is often a difference in the result in reciprocal fertilizations. Hy- brids do not play a very prominent part in nature, apart from cultivation, although the limits of some species ma}- be obscured by them, possibly of more than is generally supposed. In the animal kingdom, all the most familiar hybrids are sterile : in the vegetable kingdom, a majority may have a certain but very low degree of fertility ; but this is also the case in many unions May, 1869, 5) as the common designation of any group or collection of indi- viduals whose characters are continued through successive generations, whether it be permanent variety, subspecies, species, or group consisting of very-similar species, the term not implying any decision of this question. If this use of the term race prevails, Subspecies will probably take its place as the designation of the highest grade of variety. The objection to this is that the subspecific and specific names would be more liable to be confused. 1 Half-breed is a common equivalent term in the animal kingdom : Latin, Mistus or Mixtus ; French, Metis. 21 322 TAXONOMY. within the species, and especially in the application of the pollen to the stigma of the same blossom. Commonly the sterilit}- of hybrids is owing to the impotence of the stamens, which perfect no pollen ; and most such hybrids may be fertilized by the pollen of the one or the other parent. Then the offspring either in the first or second generation reverts to the fertilizing species. Moreover, certain hybrids, such as those of Datura, which are fully fertile per se, divide in the offspring, partly in the first gen- eration, and completely in two or three succeeding generations, into the two component species, even when close-fertilized.1 (In part this may come from adventive embiyo-formation, 533.) 639. There appears, therefore, to be a real ground in nature for species, notwithstanding the difficulty and even impossibility in many cases of defining and limiting them. 640. Species is taken as the unit in zoological and botanical classification. Important as varieties are in some respects, especially under domestication and cultivation, the}- figure in scientific arrangement only as fractions of species. Species are the true subjects of classification. The aim of S3'stematic natural history is to express their relationship to each other. 641. The whole ground in nature for the classification of spe- cies is the obvious fact that species resemble or differ from each other unequally and in extremely various degrees. If this were not so, if related species differed one from another by a constant quantity, so that, when arranged according to their resemblances, the first differed from the second about as much as the second from the third, and the third from the fourth, and so on, — or if the species blended as do the colors of the rainbow, — then, with all the diversity in the vegetable kingdom there actually is, there could be no natural foundation for their classification. The mul- titude of species would render it necessary to classify them, but the classification would be wholly artificial and arbitrary. The actual constitution of the vegetable kingdom, however, as ap- pears from observation, is that some species resemble each other very closely indeed, others differ as widely as possible, and be- tween these the most numerous and the most various grades of 1 According to Naudin in Comptes Rendus, xlix. 1859, & Iv. 1862. See also Naudin's memoir on hybridity in plants in Ann. Sci. Nat. ser. 4, xix. 1863, pp. 180-203, & in Mem. Aead. Sci. . . . For the literature on vegetable hybrids, see Koelreuter, Nachricht, &c., 1761, and Appendices, 1763-1766; Herbert, on Amaryllidaceae, 1837 ; C. F. Gsertner, Versuche und Bpobachtun- gen ueber die Bastarderzeugung in Pflanzenreich, 1849; Wichura, Die Bastardbefruchtung im Pflanzenreich, erlautertert an den Bastarden der Weiden ; and the memoir of Naudin referred to. THE PRINCIPLES OF CLASSIFICATION. 3^3 resemblance or difference are presented, but always with a mani- fest tendency to compose groups or associations of resembling species, — groups the more numerous and apparently the less definite in proportion to the number and the nearness of the points of resemblance. These various associations the naturalist endeavors to express, as far as is necessary or practicable, by a series of generalizations, the lower or particular included in the higher or more comprehensive. All kinds of differences are taken into account, but only the most constant and definite ones are relied on for characters, i. e. distinguishing marks. Linnaeus and the naturalists of his day used names for only three grades of association, or groups superior to species, viz. the Genus, the Order, and the Class ; and these are still the principal members of classification. 642. Genera (plural of Genus) are the more particular or special groups of related species. They are groups of species which are much alike in all or most respects, — which are con- structed, so to sa}', upon the same particular model, with only circumstantial differences in the details. They are not neces- sarily nor generally the lowest definable groups of species, but ire the lowest most clearly definable groups which the botanist recognizes and accounts worthy to bear the generic name ; for the name of the genus with that of the species added to it is the scientific appellation of the plant or animal. Constituted as the vegetable and animal kingdoms are, the recognition of genera, or groups of kindred species, is as natural an operation of the mind as is the conception of species from the association of like indi- viduals. This is because man}- genera are so strongly marked, at least so far as ordinary observation extends. Every one 1 knows the Rose genus, composed of the various species of Roses and Sweetbriers ; the Bramble genus, comprising Raspberries, Blackberries, &c., is popularly distinguished to a certain extent ; the Oak genus is distinguished from the Chestnut and the Beech genus ; each is a group of species whose mutual resemblance is greater than that of any one of them to an}- other plants. The number of species in such a group is immaterial, and in fact is very diverse. A genus may be represented by a single known species, when its peculiarities are equivalent in degree to those which characterize other genera. This case often occurs ; al- though, if this were universally so, genus and species would be equivalent terms. If only one species of Oak were known, the Oak genus would have been as explicitly discerned as it is now that the species amount to three hundred ; and better defined, for now there are forms quite intermediate between Oak and 324 TAXONOMY. Chestnut. Familiar illustrations of genera in the animal king- dom are furnished by the Cat kind, to which belong the domestic Cat, the Catamount, the Panther, the Lion, the Tiger, the Leop- ard, &c. ; and by the Dog kind, which includes with the Dog the different species of Foxes and Wolves, the Jackal, &c. The languages of the most barbarous as well as of civilized people everywhere show that they have recognized such groups. Natu- ralists merely give to them a greater degree of precision, and indicate what the points of agreement are. 643. If most genera were as conspicuous!}' marked as those from which these illustrations are taken, genus would be as defi- nitely grounded in nature as species. But popularly recognized genera, rightly based, are comparatively few. Popular nomen- clature, embodying the common ideas of people, merely shows that generic groups are recognizable in a considerable number of cases, but not that the whole vegetable or the whole animal kingdom is divisible into a definite number of such groups of equally or somewhat equally related species. The naturalist discerns the ground of genera in characters which the casual and ordinary observer overlooks ; and, taking the idea of genera from the numerous well-marked instances as the norm, applies it as well as possible to the less obvious or less natural cases, and groups all known species under genera. Resemblances among the species when rightly grouped into genera, though real, are often so unequal in degree, that certain species ma}r be about as nearly related to neighboring genera. So that the recognition of genera even more than of species is a matter of judgment, and even of conventional agreement as to how and where a certain genus shall be limited, and what particular association of species shall hold the position of genus. All the species of a genus must accord in every important structure ; but extended observation only can settle the question as to what are important and what are incidental characters. For example, the pinnatifid or sinu- ate leaf might have been thought as essential to the Oak genus as the acorn-cup ; but many Oaks are now known with entire leaves, resembling those of Willow or Laurel. An open acorn- cup beset with imbricated scales is a character common to all European and American Oaks ; but in numerous Asiatic species the cup bears concentric or spiral lamellae instead, and in others the cup takes the form of a naked and closed sac. Maples have palmately-veined and lobed leaves ; but one species has undi- vided and pinnately- veined leaves. The Apple and the Pear under one view are of the same genus, under another they rep- resent different genera. THE PRINCIPLES OF CLASSIFICATION. 325 644. The genus must be based on close relationship of species, but not necessarily on the closest. Raspberries differ from Blackberries, but must be ranked in the same genus ; and so of Plums and Cherries. For the groups which are to bear the generic name must be as distinct and definite as possible. 645. Orders are to genera what genera are to species. The}- are groups of a higher rank and wider comprehension, expressive of more general resemblances, or, in other language, of remoter relationship. As all species must be ranked in genera, so all genera must be ranked in orders. FAMILY in botany is synony- mous with order : at least natural orders and families (however distinguished in zoolog}-) have always in botany been inter- changeable terms, and will probably so continue.1 646. As examples of orders in the vegetable kingdom take the Oak family, composed of Oaks, Chestnuts, Beeches, &c. ; the Pine famil}', of Pines, Spruces, Larches, Cedars, Araucaria, Cypresses, and their allies ; the Rose family, in which Brambles, Strawberries, Plums and Cherries, Apples and Pears are asso- ciated with the Rose in one somewhat multifarious order. 647. Classes are to orders what these are to genera. The}* express still more comprehensive relations of species ; each class embracing all those species which are framed upon the same broad plan of structure, however differently that plan may be carried out in particulars. 648. Kingdom must be added, to represent the highest gener- alization. All subjects of biological classification belong either to the vegetable or animal kingdom. Mineralog}T, Chemistry, &c., may use the same terms (genus, species, &c.) in an analo- gous way ; but the classification of substances rests on other foundations than that of beings. 649. The sequence of groups, rising from particular to univer- sal, is Species, Genus, Order, Class, Kingdom ; or, in descending from the universal to the particular, KINGDOM, CLASS, ORDER, GENUS, SPECIES. 1 Order is the older term, and that which associates best with the technical Latin names. Family is a happy term, which associates itself well with English names. But its use is attended with this incongruity, that the tribe (653) in natural history classification is subordinate to the family. In zoology, order is distinguished from family as the next higher grade. 326 TAXONOMY. 650. This is the common framework of natural history classi- fication. All plants and all animals belong to some species ; every species to some genus ; every genus to some order or family ; ever}' order to some class ; every class to one or the other kingdom.1 But this framework, although all that is re- quisite in some parts of natural history, does not express all the observable gradations of relationship among species. And even gradations below species have sometimes to be classified. The series is capable of extension ; and extension is often requisite on account of the large number of objects to be arranged, and the various degrees of relationship which ma}' come into view. 651. This is effected by the intercalation of intermediate grades, to be introduced into the system only when there is occasion for them. And in botany one or more grades superior to the classes are needful ; for first and foremost is the great division of all plants into a higher and a lower SERIES 2 (or sub- kingdom), the Phaenogamous and the Cryptogamous. 652. The grades intercalated into the long-established sequence of Class, Order, Genus, and Species, with new names, are mainly two, Tribe and Cohort. 653. Tribe has been for a generation or two thoroughly estab- lished in both kingdoms, as a grade inferior to order and supe- rior to genus. In botanical classification, much use is made of this grade, genera being grouped into tribes. 654. Cohort (Lat. Cohors) is of more recent introduction, at least in Botany, but is becoming established for a grade next above that of order. Orders are grouped into cohorts. Lindley hit upon a good English name for this grade, that of Alliance. But this word has no available Latin equivalent ; while cohort takes equally well a Latin or an English form. 655. Finally, each grade is capable of being doubled by the recognition of one like it and immediately subordinate to it, and with designation directly expressive of the subordination. For 1 Not recognizing Haeckel's third kingdom of Protista, consisting of those lowest forms of being from which the animal and vegetable kingdoms emerge. 2 Answering to the French Embranchement in zoology. For this it is pro- posed to use the word Division (Divisio) : see Laws of Botanical Nomencla- ture adopted by the International Botanical Congress held at Paris in August, 1807 ; together with a Historical Introduction and a Commentary, by Alph. DeCandolle, — English translation, London, 1868 ; the original French edition, Paris, 1867. Perhaps no better name can be found ; but the elder DeCandolle brought Divisio into common use for a grade subordinate to tribe. Endlicher employed the term Regio. We have used Series, and much prefer it. THE PEINCIPLES OP CLASSIFICATION. 327 example, if Dicotyledones and Monocotyledones be the two classes of Phoenogamia, the former (and only the former) is divided upon very important characters into two branches, of far higher rank than the cohorts, viz. the Angiospermse and the Gymnospermse, which take the name of SUBCLASSES. Orders, especially the more comprehensive ones, often comprise two or more groups so distinct that it may fairly be a question whether they are not of ordinal rank : such take the name of SUBORDERS. Tribes in like manner may comprise groups of similar relative value : these are SUBTRIBES. Genera may comprise sections of species which might almost as well rank as genera themselves : to mark their importance and pretension (which may come to be allowed) , the}- are termed SUBGENERA. Finally, forms which are ranked as varieties, but which may establish a claim to be distinct species, are sometimes termed SUBSPECIES. Even what we regard as a variety may comprise more or less divergent forms, to be distinguished as SUBVARIETIES. 656. Some of the larger and most diversified orders, tribes, genera, or species may require all these analytical appliances, and even more, for their complete elucidation ; while others, comparatively homogeneous, offer no ground for them. But when these grades, or some of them, come into use, they are alwa3*s in the following sequence : — KINGDOM, SERIES or Division, or Sub-kingdom, CLASS, Subclass, Cohort, ORDER or Family, Suborder, TRIBE, Subtribe, GENUS, Subgenus, Section, Subsection, SPECIES, Subspecies or Race, Variety, Subvariety. 657. Nature and Meaning of Affinity. These grades, the higher including the lower, denote degrees of likeness or difference. Plants belonging respectively to the two great series or primary divisions may accord only in the most general respects, in that which makes them plants rather than animals. Plants of the same variety are generally as much alike as if they were of the 328 TAXONOMY. same immediate parentage. All plants of the same species are so much alike that the}- are inferred to have descended from a common stock, and their differences, however grave, are sup- posed to have arisen from subsequent variation, and the more marked differences to have become fixed through heredity. This is included in the idea of species. Descent from a common origin explains the likeness, and is the only explanation of it. 658. But what is the explanation of the likeness between the species themselves? As respects nearly related species, the answer is clear. Except for practical purposes and in an arbitrary way, no certain and unfailing distinction can be drawn between varieties of the highest grade and species of closest resemblance. It cannot reasonably be doubted that they are of similar origi- nation. Then there are all gradations between very closely and less closely related species of the same genus of plants. 659. The Theory of Descent, that is, of the diversification of the species of a genus through variation in the lapse of time, affords the only natural explanation of their likeness which has yet been conceived. The alternative supposition, that all the existing species and forms were originally created as the}' are, and have come down essentially unchanged from the beginning, offers no explanation of the likeness, and even assumes that there is no scientific explanation of it. The hypothesis that the species of a genus have become what they are by diversification through variation is a very old one in botan}T, and has from time to time been put forward. But until recently it has had little influence upon the science, because no clear idea had been formed of any natural process which might lead to such result. Doubtless, if variation, such as botanists have to recognize within the species, be assumed as equally or even more operative through long ante- rior periods, this would account for the diversification of an original species of a genus into several or many forms as differ- ent as those which we recognize as species. But this would not account for the limitation of species, which is the usual (but not universal) characteristic, and is an essential part of the idea of species. Just this is accounted for by 660. Natural Selection. This now familiar term, proposed by Charles Darwin, was suggested by the operations of breeders in the development and fixation of races for man's use or fane}' ; — in animals by breeding from selected parents, and selecting for breeding in each generation those individuals only in which the desired points are apparent and predominant ; in the seed-bed by rigidly destroying all plants which do not show some desirable variation, breeding in and in from these, with strict selection of THE PRINCIPLES OF CLASSIFICATION. 329 the most variant form in the particular line or lines, until it be- comes fixed by heredity and as different from the primal stock as the conditions of the case allow. In nature, the analogous selection, through innumerable generations, of the exceedingly small percentage of individuals (as ova or seeds) which ordi- narily are to survive and propagate, is made by competition for food or room, the attacks of animals, the vicissitudes of climate, and in fine by all the manifold conditions to which they are exposed. In the Struggle for Life to which they are thus inevi- tably exposed, only the individuals best adapted to the circum- stances can survive to maturity and propagate their like. This Survival of the Fittest, metaphorically expressed by the phrase natural selection, is in fact the destruction of all weaker com- petitors, or of all which, however they might be favored by other conditions, are not the most favored under the actual circum- stances. But seedlings varying, some in one direction, some in another, are thereby adapted to different conditions, some to one kind of soil or exposure, some to another, thus lessening the com- petition between the two most divergent forms, and favoring their preservation and farther separation, while the intermediate forms perish. Thus an ancestral type would become diversified into races and species. Earlier variation under terrestrial changes and vicissitudes, prolonged and various in geological times since the appearance of the main types of vegetation, and the attendant extinctions, are held to account for genera, tribes, orders, &c., and to explain their actual affinities. Affinh^ under this view is consanguinity ; and classification, so far as it is natural, ex- presses real relationship. Classes, Orders, Tribes, &c., are the earlier or main and successful branches of the genealogical tree, genera are later branches, species the latest definitely developed ramifications, varieties the developing buds.1 661. Except as to those changes in size, luxuriance, or depau- peration and the like, in which plants, especially seedlings, respond promptly to external influences, as to heat or cold, 1 For the inception of this theory of descent in the form which has within the last twenty years profoundly affected natural history, and developed a copious literature, see a short paper On the Variation of Organic Beings in a State of Nature ; On the Natural Means of Selection ; and On the Com- parison of Domestic Races and True Species, by Charles Darwin, also On the Tendency of Varieties to depart indefinitely from the Original Type, by Alfred Russell Wallace, both read to the Linnean Society, July 1, 1858, and published in its Journal of the Proceedings, iii. (Zoology) 45-62. For the development of the doctrine, see Darwin's " Origin of Species by Means of Natural Selection," " The Variation of Animals and Plants under Domes- tication," and various other works; Wallace's "Geographical Distribution of Animals," &c. For some expositions, see Gray's " Darwiniana." 330 TAXONOMY. moisture or diyness (which are transient and comparatively unimportant) , variation, or the unlikeness of progeny to parent, is occult and inexplicable. If sometimes called out by the external conditions, it is by way of internal response to them. In Darwin's conception, variation of itself does not tend in any one particular direction : he appears to attribute all adaptation to the sorting which results from the struggle for existence and the survival of the fittest. We have supposed, and Nsegeli takes a similar view,1 that each plant has an internal tendency or pre- disposition to vary in some directions rather than others ; from which, under natural selection, the actual differentiations and adaptations have proceeded. Under this assumption, and taken as a working hypothesis, the doctrine of the derivation of species serves well for the co-ordination of all the facts in botany, and affords a probable and reasonable answer to a long series of questions which without it are totally unanswerable. It is sup- ported by vegetable palaeontology, which assures us that the plants of the later geological periods are the ancestors of the actual flora of the world. In accordance with it we may explain, in a good degree, the present distribution of species and other groups over the world. It rationally connects the order of the appearance of vegetable types in time with the grades of differ- entiation and complexity, both proceeding from the simpler, or lower and more general, to the higher and more differentiated or special ; it explains by inheritance the existence of function- less parts ; throws light upon the anomalies of parasitic plants in their various gradations, upon the assumption of the most various functions by morphologically identical organs, and indeed illumi- nates the whole field of morphology with which this volume has been occupied. It follows that species are not " simple curiosities of nature," to be catalogued and described merely, but that they have a history, the records of which are impressed upon their structure as well as traceable in their geographical and palaeon- tologlcal distribution. This view, moreover, explains the re- markable fact that the characters in which the affinities of plants are mainly discerned (and which therefore serve best for orders, tribes, and other principal groups) are commonly such as are evidently of small if any importance to the plant's well- being, and that they run like threads through a series of species of the greatest diversity in habit, mode of life, and particular adaptations to conditions.2 1 Entstehung und Begriff der naturhistorischen Art. Zweite Auflage, 1865. 2 This is a corollary of natural selection, which can take effect only upon useful characters, i. e. upon structures which play some active part BOTANICAL CLASSIFICATION. 331 662. The fixity of species under this view is not absolute and universal, but relative. Not, however, that specific changes are necessitated in virtue of any fixed or all-controlling natural law. Some of the lowest forms have existed essentially unchanged through immense geological periods down to the present time ; some species even of trees are apparently unchanged in the lapse of time and change of conditions between the later tertiary period and our own da}-, during which most others have undergone specific modification. Such modifications are too slow to effect in any wise the stability and practical application of botanical classification. SECTION II. BOTANICAL CLASSIFICATION. 663. Natural and Artificial Classifications may be distinguished. A natural classification in botan}- aims to arrange all known plants into groups in a series of grades according to their resemblances, and their degrees of resemblance, in all respects, so that each species, genus, tribe, order, &c., shall stand next to those which it most resembles in all respects, or rather in the whole plan of structure. For two plants may be very much alike in external appearance, yet very different in their principal structure. Arti- ficial classifications single out one or more points of resemblance or difference and arrange by those, without reference to other considerations, convenience and facility being the controlling principles. The alphabetical arrangement of words in a dic- tionary, and the sexual system in botany by Linnaeus (or rather a part of it) , — in which plants are arranged in classes upon the number of their stamens, and in orders upon the number of pistils, — are examples of artificial classification. The arrange- ment of the words of a language under their roots, and with the derivative under the more primitive forms, would answer to a natural classification. in the life of the plant, and which therefore undergo modification under changing conditions. Unessential structures accordingly are left unaltered or are only incidentally modified. And so these biologically unessential points of structure, persisting through all adaptive changes, are the clews to relationship. Thus, Rubiaceae are known by insignificant stipules, Ano- naceae by ruminated albumen, Rhamnaceae by a valvate calyx and stamens before the petals, &c. Paradoxical as it may seem, it is not although, but because, they are of small biological importance that they are of high clas- sificatory (i. e. of genealogical) value. On considerations like these, characters are divided into adaptive or bio- loyicai on the one hand, and genealogical or genetic on the other. The saga- cious naturalist seizes upon the latter for orders and the like ; while the former are prominent in genera, &c. 332 TAXONOMY. 664. No artificial classification of plants could fail to be natural in some portions and some respects ; because plants which agree in any point of structure likely to be used for the purpose will commonly agree in other and perhaps more impor- tant characters. On the other hand, no natural classification can dispense with artificial helps ; nor can it express in lineal order, or in any other wray, all the various relationships of plants, even if these were fully determined and rightly subordinated. Natu- ralists now endeavor to make classification as natural as possible ; that is, to base it in every grade upon real relationships. What real relationships are, and how to express them in a general system and throughout its parts, has been the task of the leaders in botany from the beginning of the science until now ; and the work is by no means completed. 665. Linnaeus was perhaps the first botanist to distinguish clearly between a natural and an artificial classification. lie labored ineffectuall}' upon a natural classification of the genera of plants into orders ; and he devised an effective artificial classi- fication, which became so popular that it practically superseded all others for more than half a century, and has left a permanent impression upon the science. The last generation of botanists who were trained under it has not quite passed away. 666. Ante-Linnaean Classification. Linnaeus, in his Philosophia Botanica, divided systematists into heterodox and orthodox : the former, those who classify plants by their roots, herbage, time of flowering, place of growth, medical and economical uses, and the like ; the latter, by the organs of fructification. It is remark- able that all the orthodox or scientific classifications anterior to Linnaeus made a primary division of the vegetable kingdom into Trees and Herbs, referring the larger shrubs to the former and the under-shrubs to the latter, — an arrangement which began with Theophrastus and was continued by Ray and Tournefort. 667. The three most important names in botanical taxononi}- anterior to Linnaeus are those o" Cesalpini, Ray, and Tournefort. Scientific botany commenced with the former, in Italy, in the latter half of the sixteenth century. He first used the embryo and its cotyledons in classification, distinguished differences in the in- sertion of floral parts, and, indeed (excepting the primary division into trees and herbs) , founded all principal characters upon the organs of fructification, especially upon the fruit and seed. Conrad Gesner of Zurich had somewhat earlier recognized this principle, but Cesalpini first applied it. 668. A century later (1690-99) this principle was carried into practice by Rivinus (a name latinized from Bachmann), of BOTANICAL CLASSIFICATION. 333 Leipsic, in a wholly artificial classification founded on the corolla. His contemporary in England, Robert Morison, somewhat earlier began the publication of his great work, the Universal History v of Plants. In this was first attempted a grouping of plants into what are now called natural orders ; and these were defined, some- what loosely, some by their fruit, inflorescence, and flowers, others by their stems, the nature of their juice, &c. But the two great systematists of the time, who together laid the foundations of modern scientific botany, were John Ray in England and Joseph Pitton de Tournefort in France. 669. Ray's method of classification was sketched in 1682, and was anterior to Tournefort's, but was amended and completed in 1703. The leading fault of both was the primary division into trees and herbs. The great merit of Ray was his division of herbs into Flowerless and Flowering, and the latter into Dicotyle- donous and Monocotyledonous. These great classes he divided and subdivided, by characters taken from the organs of fructi- fication, into what we should call natural orders or families, but which he unfortunately called genera. He noted the coincidence of nerved leaves with the monocotyledonous embryo, although he did not notice that his first division of arborescent plants was monocotyledonous ; and he had a clear apprehension of genera. 670. Tournefort's method was published in French in the 3rear 1694, in Latin in 1700. It is more definite but more arti- ficial than that of Ray, being founded like that of Rivinus almost wholly upon modifications of the corolla, and it overlooked the dis- tinction between monocotyledonous and dicotyledonous embryos. Its great merit is that here genera, as we now understand them, are first established and defined, and all the species then known referred to them ; so that Tournefort was justly said ~by Linnaeus to be the founder of genera. Ray may be said to have indicated the primary classes, Jussieu (in the next century) to have estab- lished natural orders, and Tournefort to have given to botany the first Genera Plantarum. 671. Linnaean Classification. Linnaeus, the great reformer of botany in the eighteenth century, thoroughly revised the principles of classification, established genera and species upon a more scien- tific basis, and, in designating species by a word instead of a descriptive phrase, introduced binomial nomenclature. (704.) He likewise established for the stamens, and indeed for the pistils also, their supreme importance in classification (probably without knowledge of the clear suggestion to this effect made by Burckhard in a letter to Leibnitz, printed in 1702) ; their functions, so long overlooked, being now ascertained. He also 334 TAXONOMY. drew a clear and practical distinction between natural and arti- ficial classifications (663), and deferring all endeavors to make the "former available, except for genera, he devised a practical substitute for it, as a ke}- to the genera, viz. his celebrated 672. Sexual System, or arrangement of the genera under arti- ficial classes and orders, founded upon the stamens and pistils. Although now out of use, this artificial classification has been so popular and influential, and has left? so deep an impression upon the science and especially upon the language of botany, that it needs to be presented. The primary divisions are the classes, twenty-four in number. But the 24th class, Cryptogamia, con- sists of plants which have not stamens and pistils and conse- quently no proper flowers, and is therefore the counterpart of the remaining twenty-three classes, to which the corresponding name of Phanerogamia or, in shorter form, Phaenogamia (Phasnogamous plants) has since been applied. These twenty-three classes are characterized by certain modifications and associations of the stamens, and have substantive names, of Greek derivation, ex- pressive of their character. The first eleven comprise all plants with perfect (i. e. hermaphrodite) flowers, and with a definite number of equal and unconnected stamens. They are distin- guished by the absolute number of these organs, and are desig- nated by names compounded of Greek numerals and the word andria (from dv^(t) , which is used metaphorically for stamen, as follows : — Class 1. MOXANDRIA includes all such plants with one stamen to the flower : as in Hippuris. 2. DIANDRIA, those with two stamens, as in the Lilac. 3. TRIANDRIA, with three stamens, as in the Valerian and Iris. 4. TETRANDRIA, with four stamens, as in the Scabious. 5. PENTANDRIA, with five stamens, the most frequent case. 6. HEXANDRIA, with six stamens, as in the Lily Family, &c. 7. HEPTANDRIA, with seven stamens, as in Horsechestnut. 8. OCTANDRIA, with eight stamens, as in Evening Primrose and Fuchsia. 9. ENNEANDRIA, with nine stamens, as in the Rhubarb. 10. DECANDRIA, with ten stamens, as in Rhododendron and Kalmia. 11. DODECANDRIA, with twelve stamens, as in Asarum and the Migno nette; extended also to include those with from thirteen to nine- teen stamens. 673. The two succeeding classes include plants with perfect flowers having twenty or more unconnected stamens, which, in 12. ICOSANDRIA, are inserted on the calyx (perigynous), as in the Rose Family ; and in 13. POLYANDRIA, on the receptacle (hypogynous), as in the Buttercup, Anemone, &c. BOTANICAL CLASSIFICATION. 335 674. Their essential characters are not indicated by their names : the former merely denoting that the stamens are twenty in number ; the latter, that they are numerous. — The two fol- lowing classes depend upon the relative length of the stamens, namely : — 14. DIDYNAMIA, including those with two long and two short stamens, as in the majority of flowers with bilabiate corolla. 15. TETKADYNAMIA, those with four long and two short stamens, as in flowers with cruciferous corolla. 675. These names signify in the former that two stamens, and in the latter that four stamens, are most powerful. — The four succeeding are founded on the connection of the stamens, viz. : — 16. MONADELPHIA (meaning a single fraternity), with the filaments united in a single set, tube, or column, as in the Mallow. 17. DIADELPHIA (two fraternities), with the filaments united in two sets or parcels, as in Corydalis and in many Leguminosae. 18. POLYADELPHIA (many fraternities), with the filaments united in more than two sets or parcels, as in Hypericum. 19. SYNGENESIA (from Greek words signifying to grow together), with the anthers united in a ring or tube, as in the Sunflower and all Compositae. 676. The next class, as its name denotes, is founded on the union of the stamens to the style : — 20. GYNANDRIA, with the stamens and styles consolidated, as in Cypri- pedium and all the Orchis Family. 677. In the three following classes, the stamens and pistils occup}7 separate blossoms : — 21. MONCECIA (one household) includes all plants where the stamens and pistils are in separate flowers on the same individual ; as in the Oak and Chestnut. 22. DIOZCIA (two households), where they occupy separate flowers on different individuals ; as in the Willow, Poplar, Moonseed, &c. 23. POLYGAMIA, where the stamens and pistils are separate in some flowers and associated in others, either on the same or two or three different plants ; as in most Maples. 678. The remaining class is essentially flowerless ; or rather its organs of reproduction are more or less analogous to, but not homologous with, stamens and pistils. But, although Linnasus suspected a sexuality in Ferns, Mosses. Algjie, &c., there was no proof of it in his day. So he named the class, containing these, ' 24. CRYPTOGAMIA, meaning clandestine marriage, the sexes, if existent, hidden from view. 679. The characters of the classes may be presented at one view, as in the subjoined table : — TAXONOMY. isgg o so • -353 ' • • a ' m £ i! : • 1 .!§ ' s| s« • '*> S S g . • ' o_g • £•0 §3 tss ' "lo !£ 'ill ' ' j^S ' • 'IS gj : ns-s: g g -a « 2gS £| 1-1 N CO «* 10 (0 t- 00 OS 0 -H 0 -H gS I . f! eir filament eir filament eir filament eir anthers '• il!f-i . "5 C^3 § 1 11 SCtSTJ ££££ : 111- : 1- 1^ gl §1 ^1 -si each other . 2 .- 2 "8 •2 1 BOTANICAL CLASSIFICATION. 337 680. The orders, in the first thirteen classes of the Linnaean artificial system, depend on the number of styles, or of the stigmas when the styles are wanting ; and are named by Greek numerals prefixed to the word gynia, used metaphorically for pistil, as follows : — Order 1. MONOGYNIA, those with one style or sessile stigma to the flower. 2. DIGYNIA, those with two styles or sessile stigmas. 3. TRIGYNIA, those with three styles. 4. TETRAGYNIA, those with four styles. 5. PENTAGYNIA, those with five styles. 6. HEXAGYNIA, those with six styles. 7. HEPTAGYNIA, those with seven styles. 8. OCTOGYNIA, those with eight styles. 9. ENNEAGYNIA, those with nine styles. 10. DECAGYNIA, those with ten styles. 11. DODECAGYNIA, those with eleven or twelve styles. 12. POLYGYNIA, those with more than twelve styles. 681. The orders of class 14, Didynamia, are only two and are founded on the pericarp, namely : — 1. GYMNOSPERMIA, meaning seeds naked, the achenia-like fruits of a 4-parted pericarp having been taken for naked seeds. 2. ANGIOSPEKMIA, with the seeds evidently in a seed-vessel or peri- carp, i. e. the pericarp undivided. 682. The 15th class, Tetradj-namia, is also divided into two orders, which are distinguished merely by the form of the pod : — 1. SILICULOSA; the fruit a silicic (561), or short pod. 2. SILIQUOSA; fruit a silique (561), or more or less elongated pod. 683. The orders of the 16th, 17th, 18th, 20th, 21st, and 22d classes depend merely on the number of stamens ; that is, on the characters of the first thirteen classes, whose names they likewise bear : as MONANDRIA, with one stamen, DIANDRIA, with two stamens ; and so on. 684. The orders of the 19th class, Syngenesia, are six, namely : 1. POLYGAMIA ^QUALIS, where the flowers are hi heads (the so-called compound flower), and all hermaphrodite. 2. POLYGAMIA SUPERFLUA, the same as the last, except that the rays, or marginal flowers of the head, are pistillate only. 3. POLYGAMIA FRUSTRANEA, those with the marginal flowers neutral, the others perfect. 4 POLYGAMIA NECESSARIA, where the marginal flowers are pistillate and fertile, and the central staminate and sterile. 5. POLYGAMIA SEGREGATA, where each flower of the head [or glom- erule] has its own proper involucre. 6. MONOGAMIA, where solitary flowers (that is, not united into a head) have united anthers, as in Lobelia. 338 TAXONOMY. 685. The 23d class, Polygamia, has three orders, two of them founded on the characters of the two preceding classes and bearing their names, and the third named upon the same prin- ciple, namely : — 1. MON 728. Ordinal Names are formed in the same way, but with a preference for certain terminations which ma}- denote their rank, especially that of -acece, — as Rosacece, Myrtacece, Cucurbitacece, meaning Rosaceous, Myrtaceous, and Cucurbitaceous plants. 729. The names of what we now call natural orders, as sketched or adopted by Linnaeus, were mostly descriptive, such as Ensatce, Spathacece, Coronarice, Papilionacece , Coniftrce, Amen- tacece, Umbellatce ; but a few took their names from genera, as Orchidece, Liliacece. Jussieu, with whom the system of natural orders properly began, had no suborders, tribes, or any such gra- dation of groups to deal with. His one hundred ordinal names are some of them of the descriptive kind, as several of the above, also Leguminosce, Corymbiferce, &c. But the greater part are simply plurals of generic names, such as Asparagi, Junci, Liliu, Mu see, Ore/tides, Lauri, Convolvuli, Ericce, Accra, Cacti. To a few was given the lengthened termination in -ece, as Polygonecc, Solanece, JBerberidece, Caryophylkce ; to some, the termination in -acece, as in Oichoracece* Campanulacece, Rubiacece, Ranuncu- lacece. Malvaceae, Tiliarece, Cucurbitacece . Subsequent authors have necessarily changed all names which were plurals of gen- 358 PHYTOGBAPHY. era ; and the strongly prevalent tendency has been to give the termination in -acece to all such ordinal names, and to restrict this termination to orders. Lindley insisted upon making this an absolute rule even for names not formed from generic appel- lations ; but this will not be adopted. 730. In the first place, several large orders which have been known from the first by such characteristic names as Cruciferce, Leguminosce (and its suborder Papiiionacece) , Guttiferce, Umbelli- ferce, Composite, Labiatce, Cupulifcrce, and Coniferce, also Palmcce and Graminece, Filices, and even Afoidece and Ficoidcce, will retain these appellations ; but no new Ones of the kind will be made. 731. Also, names formed from genera which do not well take the termination in -acece may be allowed as orders to retain their natural form in -inece, -idece, -ariece, and the like. Ex. Tamaris- cinece, Salicinece, Scrophularinece, Berberidece, Lentibulariece. We ma}' prefer for the sake of uniformity to write Salicacece, Berberi- dacece, Lentibulariacece, and Scrophulariacece (as we should write Violaceai) , but this form cannot be insisted on. On the other hand, a termination in -acece has been allowed in the names of certain tribes to avoid excessive iteration of vowels. Thus, for the tribe of which Vernonia is the leading genus, authors write Vernoniacece, to avoid Vernoniece, which ends with four vowels. Spiraea and Staphylea are the types of tribes, for which the names, if they followed the rule, would be Spirceece and Staphyleece, ending one in five the other in four consecutive vowels. Some avoid this by writing Staphyleacece and Spirceacece. Others write Stapliylece, but this is only the plural of the generic name. 732. A few orders or other groups took their names long ago from superseded generic names. Ex. Caryophyllacece or Caryophyllece, Onagracece or Onayrariece, and Lentibulariece. 733. Names of Cohorts are distinguished by the termination in -ales. This was proposed by Lindley, and is adopted by Bentham and Hooker in the Genera Plantarum. Ex. Ranalcs, Parietales, Malvales, Rosales, Passiflorales, &c., most of them founded on the names of representative genera and orders. Euphony requires some to take other terminations. Ex. Poly- galinee, Caryophyllina:. 734. Names of Classes and other great divisions are plurals, either adjective or adjective nouns, expressive of the leading character. Ex. Polypetnlce, Gamopetalce, Apetalce ; Angiospermce and Gymnosperma ; Dicotyledones and Manor of yledones. The names of the two great series or sub-kingdoms, following the analogy of the Linnagan classes, end in -IOT, and are Phcenogamia, or Phanerogamin, and Cryptogamia. GLOSSOLOGY. 359 SECTION II. GLOSSOLOGY OR TERMINOLOGY.1 735. This is nomenclature as applied to organs or parts and their modifications. The actual botanical terminology owes its excellence in the first place to Linnaeus, and then to DeCandolle. The Theorie Elementaire of A. P. DeCandolle (the first edition of which was published in 1813) is still classical authority, and until recently has received few additions as regards terms need ful in phaenogamous botany. 736. The fundamental rule is that each organ or part shall have a substantive name, and that modifications of organs shall be designated by adjective terms. These names or terms should be as precise as possible : each object ought to be known by only one name, yet synonyms are unavoidable ; and no term ought to be used with two different meanings. The word flower, for instance, must not be used for a cluster of flowers, however it may imitate the appearance of one, nor for the corolla or other portions of a flower. Still, some terms have to be used in two or more senses, to be determined only by the connection, or else as having both a special and a more general meaning. Leaf (fo- lium) is a notable instance. A bract, to go no farther, is a sort of leaf; and the imperfect stamens of a Catalpa-flower and Pentstemon are stamens, although likewise called staminodia : these are liable to be called sometimes by one, sometimes by the other name. But, however frequent such ambiguities ma}r be in morphological treatment, they are usuall}- avoidable in descriptive botany, in which terms are held to their more special or partic- ular sense. Yet no rule can absolutely determine whether leaf or bract, bract or bractlet, is the proper term in many cases. Moreover, substantive names must also be applied to certain mere modifications of the same organ. In the same family, a simple carpel, differently modified in fruiting, is an akene in a Ranunculus, a follicle in Aquilegia, a berry in Hydrastis and Actaea ; while in another family an additional line of dehiscence makes it a legume. Moreover, in this latter famity it is called a legume when it is not dehiscent at all, and even when it becomes a drupe ! Arbitrary rules cannot absolutely fix technical any more than ordinary language. 737. Experience and judgment must determine what modifi- cations of organs should be regarded as a kind, and bear sub- 1 Although the former is the better name, the latter is well established in use as an English word, and perhaps it need not be objected to, inasmuch as the Latin terminus comes from the Greek reppa, of the same meaning. 360 PHYTOGRAPHY. stantive instead of merely adjective names. But the former should not be unnecessarily multiplied. 738. The classical language of scientific botany being Latin, all the organs of plants and their principal diversities are desig- nated by a Latin or Latinized name. Modern languages have also their own names and terms. Greatly to its advantage, English botanical terminology has adopted and incorporated terms from the Latin and Greek, with slight changes, not obscur- ing the identity, thus securing all their precision, and rendering the simple botanical Latin of descriptions of eas}" acquisition to the English student. 739. In a text- book like this, the principal names and terms applied to organs and their leading modifications, as also those which relate to their action (physiological tenns) , or to our study of them (didactic terms, such as phytograptry, phyllotaxy, glos- solog}'), are defined and illustrated in course. There remain the more numerous and varied characteristic terms, chiefly adjec- tives, applicable to more than one or to all organs, and which compose the greater part of glossology. These, which DeCandolle arranged systematically with much elaboration, ma}* best be reached by a glossary or dictionary, such as that at the end of this volume, which comprises the substantive terms likewise. 740. From characteristic adjective terms are derived the greater number of specific names of plants ; of which, therefore, the glossary may elucidate the meaning. 741. Capable as the existing system is, it cannot in single words define all observed forms and grades, nor well avoid various ambiguities of meaning. Some defects of the first kind are remedied by combining with a Iryphen two congruous terms to denote an intermediate state. Ex. ovato-lanceolutus, or ovate- lanceolate, for an outline between the two. Also a term may be qualified b}T the prefix sub, in the sense of somewhat, as in sub- rotundus, subcordatus (somewhat round or slightly heart-shaped) , or diminutives (such as integriusculus) , or superlatives (integer- rimus) or other strengthened forms (such as perangustus) may be emplo3-ed. Among terms of more than one form of meaning are such as ccdycinus, which may mean, according to the context, pertaining to the calyx, or of the appearance of calyx ; cymosus may mean in cymes, or bearing cymes, or in the manner of a C3'me ; and paleaceus may mean provided or beset with chaff, or resembling chaff in texture. Often the form of the word should distinguish the sense ; &sfoliatus, furnished with leaves, foliosus, with abundance of leaves, while foliacens may mean either bear- ing leaves, or properly of leaf-like texture or appearance. DESCRIPTION. 361 742. Absence of an organ or quality may be expressed by means of a prefix with privative signification, as indehiscent, not dehiscent, exannulote, destitute of a ring, apetalous, without petals. But the Greek privative « should not be prefixed to Latin words, nor the Latin sub to terms taken from the Greek. 743. When the Latin preposition ob is prefixed to an adjective term, it means obversely ; thus obcordatus is cordate inversed, that is, the broader end with its notch at the apex (instead of the base) of the leaf or other plane organ. SECTION III. DESCRIPTION. 744. Under this head may be convenient!}' comprised all that relates to the form of the exposition, in botanical terms, of the differences by which the species and groups of plants are distin- guished and recorded, the structure exemplified, and the history or bibliography indicated in systematic works or writings. Lin- naeus, in the Philosophia Botanica, treated these topics under the head of " Adumbrationes." 745. Descriptions may be full and general, comprising an account of all that is known of the structure and conformation of a plant or group, or rather all that is deemed worth recording, or the}- may be restricted to what is thought most important. In the former, the description is independent of all relative knowl- edge, or takes no notice of relationship to other plants or groups. The latter intends to portray the species or group in its relations to others, and to indicate the differences solely. Exhaustive descriptions of the former kind are seldom drawn up, but partial or supplementary ones are common. Descriptions of the latter kind, when reduced to what is essential or differential, are termed Characters, or the Character, of the group so described. There are all gradations in practice between characters and descrip- tions ; but the distinction should be maintained. 746. Characters are specific, generic, ordinal, &c. They are the differentiae, or marks which distinguish a group from any related group of the same rank with which it ma}- properly be compared. According to the occasion and purpose, they may specify only the fewest particulars which will serve as a diag- nosis, or they may be extended to all the known constant differ- ences between two or more related species, genera, orders, &C.1 1 The former would answer to what have been termed differential char- acters, the latter to essential characters. Linnaeus divided (generic) characters into factitious, essential, and natural; by the former denoting any difference which may effectively distinguish between any two groups brought arti- 362 PHYTOGRAPHY, What is now termed the specific character was the specific name with Linnaeus and his predecessors ; what we call the specific, Linnaeus called the trivial name. (703.) 747. Subordination of characters and the avoidance of vain repetitions require that as far as possible — regard being had to the form of the work — the ordinal character should contain only what is needful to circumscribe it, and to exhibit clearly its morphology ; that the characters of tribes or other divisions should not reassert any portion of the ordinal character, nor the generic character that of the superior groups ; and so of the sections and subdivisions of all grades down to the species. Equally from the specific character should be excluded every thing which belongs to the generic, or is common to its rela- tives generally, or has been already specified in the section or its subdivisions. So, likewise, of the varieties under the spe- cies. This can be done only by so arranging the species as best to exhibit their relationships, that is, by bringing together or into proximity those of greatest resemblance in all respects, or in the more important respects. What these are, and how a just subordination of characters is to be apprehended, cannot be taught by rules, but must be learned by experience and from the critical study of the classical botanical works. No one is competent to describe new plants without such study, and without a clear conception of the position which a supposed new species should occup}- in its genus, or a genus in its order. 748. Characters of orders, genera, and of all intermediate groups, are drawn almost without exception from the organs of fructification. In the description, these parts are mostly taken in order, beginning with the calyx and ending with the ovary, the fruit, seed, embr}-©. But, as to the orders, some writers pre- fer to preface these proper characters with a general sketch of those derived from the vegetation, which, albeit of less syste- matic value generally, are often very characteristic of particular families. Rubiaceae, for example, are known by their opposite entire and simple leaves and intervening stipules, along with a few floral characters ; Sarraceniaceae, by tubular or pitcher-like leaves, along with a certain combination of a few other charac- ficially together, as they might be in an artificial key, and as very anlike genera often were in his sexual system ; by the second meaning the distinc- tions, the fewer the better, which will separate a group from its nearest relatives ; by the third, all real marks of difference, i. e. all afforded by the organs of fructification, which only were taken into account for genera, &c. Upon the construction of this natural character Linnaeus prided himself, and justly. These are the characters in his Genera Plantarum. CHARACTERS. 368 ters, and so on. Where brevity is aimed at, such external and obvious characters, followed by a few diagnostic marks, may practically take the place of a full enumeration of particulars, man}' of which may be common to other orders, though not in the same combination. Generic characters alwa3*s commence with the calyx or most external of the floral organs and proceed to the ovary, thence to the fruit and seed, and end with subsi- diaiy (but often no less diagnostic) particulars furnished by the vegetation and mode of growth. 749. Detailed descriptions of species, as distinguished from technical characters, commence with the root, and proceed in order to the stem, leaves and their parts or appendages, inflor- escence, bracts, flowers, calyx, corolla, stamens, with filament, anther, and pollen, the disk, if any, gyncecium and its parts, ovules ; then the fruit, seed, albumen, if any, embrj-o and its parts. But descriptions of this sort in most works and in ordi- naiy cases are partial and subsidiary, comprising only certain details supplemental'}* to or in amplification of the character of the species or genus. In condensed works, such description is wholly omitted, or is reduced to a few specifications which do not readily find their way into the character. 750. Specific characters usually follow the same order of enumeration, from root to seed, so far as the several organs arc mentioned ; and in Latin the phrases are expressed in the abla- tive case. But these particulars are often very conveniently prefaced by statements applying to the whole plant rather than to any one organ ; and these are given in the nominative, and agree with the name in gender.1 751. Linnaeus required that neither the essential character of a genus, nor a specific character (his nomen specificum), should exceed twelve words. Latin characters take fewer words than English. But this arbitra^ rule is wholly out of date. Yet such characters should be brief and diagnostic : otherwise, their advantage is lost, and the distinction between them and descrip- tions disappears. In monographs and floras, the desirable brevity, or such as the case "admits, is secured by proper group- ing under a subordination of sections, subsections, and other subdivisions.2 1 Ex. " NEPETA CATARIA : erecta, elata, cano-pubescens ; foliis petio- Jatis," etc. In English, these adjectives without any substantive expressed will be seen to belong, as here, to " plant " or " herb " understood. 2 In the Synoptical Flora of North America, such a system of successive divisions is thoroughly carried out. And, if the specific characters are by no means short, it is mostly because nearly all separate descriptive matter is 364 PHYTOGEAPHY. 752. Punctuation. In proper descriptions, and in characters of genera and of higher groups, the account of each organ forms a separate sentence ; and in Latin the terms are in the nomina- tive case, except subsidiary portions, which are often thrown into the ablative. Excepting the latter part, the adjective terms are separated by commas. A specific character is always in one sentence. In Latin, its clauses are mainly in the ablative ; and much diversity prevails as to the punctuation.1 Subgeneric and other sectional characters are commonly framed like those of dispensed with : consequently various particulars are added to the char- acter which do not strictly belong to it. In Bentham's great Flora Austra- liensis, also in English, specific characters are replaced by a characteristic synopsis at the head of each genus ; and a terse description under each species completes the account. Moreover, Bentham, in recent works, such as his revision of the Genus Cassia, also that of the Mimosese, which have Latin characters, writes these in the nominative case and each member in a separate sentence, in the descriptive form, abandoning the long-used abla- tive form. 1 Linnaeus employed only the comma in the specific character, along with a subsidiary use of the colon in a manner very unlike its ordinary use in punctuation, making it a point of less value than the comma. Thus,^ "CHENOPODIUM ALBUM foliis rhomboideo-triangularibus erosis postice integris : summis oblongis, racemis erectis." Spec. PI. ed. 2, 319. Here, while the two main members of the sentence are separated by a comma, a subsidiary portion of the first member, relating to the uppermost leaves, is separated by a colon. Linnaeus employed the colon in the same way in generic characters. This anomalous usage is now abandoned. But most authors have followed the Linnasan pattern in distinguishing the prin- cipal members by commas only, so that these become the only points in the specific character, however complicated that may be. Thus, " RANUNCULUS ACKIS (Linn. Spec. 779) foliis pubescentibus subglabrisve palmato-partitis, lobis inciso-dentatis acutis, summis linearibus, caule erecto plurifloro subpubescente, pedunculis teretibus, calyce subvilloso, carpellis mucrone suberecto terminatis " DeCandolle, Prodromus, 5. 36. This is the punctuation throughout the Prodromus and in most contem- porary systematic works. Its imperfection is shown in the above-cited speci- men. The primary members of the sentence, which characterize the leaves, stem, peduncles, calyx, and carpels, are distinguished by the same grade of punctuation which serves for the parts of the first member, viz. the lobes of the leaves, and for a still subordinate portion, viz. the form of the upper- most lobes. This want of subordination is to be remedied by the use of semicolons between the principal members, and of the commas only for the secondary ones, — a punctuation now not uncommon, and which is adopted in the recent first volume of the Monographiae Phanerogamarum of the De- Candolles, which supplements the Prodromus. The portion of that volume contributed by Dr. Masters better exemplifies this than does the rest of the volume. For the latter sacrifices the advantage of the change by the inser- tion of commas between each adjective of a continuous ablative phrase (as, " Smilax laurifolia ; limbis foliorum oblongis vel ovato-oblongis, coriaceis, 3-5-nerviis, subtus pallidioribus," etc.), where they are generally deemed PUNCTUATION, SYNONYMY. 365 genera. Or the members may be united in one sentence, but in that case the principal ones are best separated by colons. 753. Should a point intervene between the specific name and that of the author cited? The practice varies. But, if the name is Latin, the comma is superfluous ; for the abbreviated name of the author is supposed to be in the genitive, and to read thus : Ranunculus repens Linncei. Still, since when the author's name is cited in full it is never written in the genitive, and since in English the comma is normally required, it seems on the whole proper to insert it. 754. In citations, the classical practice is to separate the refer- ences from each other and from the name by periods ; thus, " Anemone cylindrica, Gra}', Ann. N. Y. Lye. 3. 221. Torr. & Gray, Fl. 1. 113,"1 &c. It is becoming equally customary to separate the several citations by semicolons, thus bringing all the references under one name into one sentence. The bibliography of a species or group of species which a describer or other author has to refer to (with more or less fulness, according to the form of his work) is to be sought partly under the admitted name, and partly in the 755. Synonymy. This includes 'all other than the admitted names. Ex. Hydropeltis of Michaux is a synonym of Brasenia of Schreber, the latter being the earlier published name. Nectris of Schreber is a synonym of Cabomba of Aublet, the latter hav- ing priorit}'. Thalictrum anemonoides of Michaux has for syno- nyms Anemone thalictroides of Linnaeus and of many subsequent authors who followed him in referring this ambiguous plant to Anemone (721) ; and also Syndesmon thalictroides of Hoffmannsegg and AnemoneUa thalictroides of Spach, who proposed to consider superfluous. The preferable punctuation of the character above-quoted from the Prodromus would be RANUNCULUS ACRIS (Linn.) : foliis pubescentibus subglabrisve palmato- partitis, lobis inciso-dentatis acutis, summis linearibus ; caule erecto pluri- floro subpubescente ; pedunculis teretibus ; calyce subvilloso ; carpellis mucrone suberecto terminatis. The advantages of this style of punctuation will more and more appear, when applied to less simple cases. Commas between the ablative adjectives are superfluous and confusing. In English characters, commas are required between the adjectives which follow the noun. Rightly to express the subordination of characters, the plan adopted in the Synoptical Flora of North America is recommended ; that is, with colons separating the principal members, semicolons for subordinate and dependent ones, and commas between the adjectives of the same noun. 1 See Watson's Bibliographical Index to North American Botany (where this style is adopted) for a general model for the arrangement of synonymy and citations. 366 PHYTOGRAPHY. it an intermediate genus between Thalictrum and Anemone. In systematic works, the specific character immediately follows the name, and generally forms a pail of the same sentence ; and is followed first by citations of authors who have adopted the name, and then by the synonymy, or as much of it as the plan of the work calls for. The synonymous names and the references under them should be cited in the order of their publication. But, to economize space, all the authorities for the same name are brought together into one sentence, and arranged according to their date. Also, where the synonymy is not elaborately displayed, the various synonyms of the same generic name are usually placed in consecutive order. 756. Iconography. The leading and most essential citation is that of the author by whom and the work in which a plant is named and described, and also the work in which it is best char- acterized. Among the characterizations, published figures hold a prominent place. The citation of these is an important part of the synonymy. The best botanical plates are those which give detailed analyses of the parts of the flower, fruit, and seed, displaying their structure. 757. Habitat and Station are recorded in a sentence or para- graph following the name, character, and synonymy of a species. The habitation is the place, district, or region at or within which the plant is known to be indigenous, or to grow spontaneously. The complete habitat is the geographical range. The station is the situation it affects, whether in water, in marshes, on shores, on hills or mountains, in forests, on open plains, &c. 758. Discoyerer, &c. To the habitat and station of newly discovered, rare, or local plants should be appended the name of the discoverer or the collectors by whom the species has be- come known to science, at least when the plant is first published. Date of discover}- should also then be indicated. 759. Time of Blossoming should be recorded, either the month or the season, to which ma}' be added that of the maturity of the fruit. When the month or season is mentioned without farther explanation, flowering-time is intended. In a flora, this ma}' sometimes be indicated under the genus for all the species. In the flora of an extensive region, and in respect to species of considerable range in latitude or longitude, the time of flowering differs so widely at the extremes of the geographical range that it cannot well be specified except in general terms, as spring, summer, autumn, &c. 760. Etymology of Names. When a new generic name is pub- lished, its origin and meaning should always be given, if the ACCENTUATION, ABBREVIATIONS. 367 nature of the publication will allow it. So likewise of species, except where the source or signification of the name is mani- fest. This is commonly the case as respects most characteristic specific names, and also those drawn from station, habitat, and the like. 761. Accentuation of Names. The pronunciation of botanical names is settled by the rules of Latin prosod}*. All that is usually attempted in those botanical works which take this into account is to mark the syllable upon which the principal accent falls. This in words of two syllables is alwaj-s the first ; in words of three or more syllables, either the penult (the last sylla- ble but one) or the antepenult (next preceding syllable) . When the penult is a long syllable, it takes the accent ; when short, this recedes to the antepenult. The accentuation may accordingly be sufficiently indicated by marking the quantity of the penult, either long as in Erica, or short as in Arbutus and Gladiolus. Or else the accent may be marked by a proper sign, as Erica, Arbutus, Gladiolus. An endeavor has been made to represent the longer sound of the vowel by the grave accent-mark, as Erica, and the u*/kr> Benth. Bentham. Cam. Camerarius. Berg. Berk. Bergius. M. J. Berkeley. Camb. Cambes. Cambessedes. 386 ABBKEV1AT10NS. Campd. = Campdera. Eat. = Amos Eaton.. Cand. DeCandolle, usually DC. D. C. Eat D. C. Eaton, grandson. Casp. Caspary. Edgew. Edgeworth. Cass. Cassini. Edw. Edwards. C'atesb. Catesby. E/iren. Ehrenberg. Cav. Cavanilles Ehrh. Ehrhart. Cerv. Cervantes. Eichl. Eicliler. Cham. Chamisso. Eiseng. Eisengrein. Chapm. A. W. Chapman. EU. Elliott. Chao. Chavannes. Endl. Endlicher. Chois. Choisy. Engelm. Engelmann. Clayt. Clayton. Engl. Engler. Clus. Clusius. Eschs. Eschscholtz. Collad. Colladon. Eschw. Eschweiler. Coim. Colmeiro. Ettingsh. Ettingshausen. Comm. Commelin. Corn. Cornuti. Fendl. Fendler. Coss. Cosson. Feuil. Feuillee. Cunn. Cunningham, A. or J. Finger h. Fingerhuth. Curt. Wm. Curtis. Fisch. Fischer. M. A. Curt M. A. Curtis. Forsk. Forskal. Forst. Forster. Dalech. Dalechamps. Fourn. Fournier. Dalib. Dalibard. Fresen. Fresenius. Dad. Darlington. Freyc. Freycinet. DC. DeC. A. P. DeCandolle. Fred. Frcelich. A. DC. Alphonse DeCandolle,son. Gcertn. J. Gaertner. Cos. DC. Casimir DeCandolle, the Gartn.f. C. T. Gaertner. Deem. Decaisne. [grandson. Gardn. Gardner. Deless. Delessert. Garid. Garidel. Dennst. Dennstedt. Gasp. Gasparrini. Desc. Descourtilz. Gaud. Gaudin. Desf. Desfontaines. Gaudich. Gaudichaud. Desj. Desjardins. Germ. Germain. Desmar. Desmazieres. Gesn. Gesner. Deamoid. Desmoulins. Gillb. Gilibert. Desv. Desvaux. Ging. Gingins de Lassaraz. Dicks. Dickson. Gis. Giseke. Diesb. Diesbach. Gled. Gleditsch. Dieter. Dieterich. Gleich. Gleichen. Dietr. Dietrich. Glox. Gloxin. Dill. Dillenius. Gmel. J. G. Gmelin. Dittw. Dillwyn. C. Gmel. C. C. Gmelin of Baden. Dod. Dodonaeus (Dodoens). S. Gmel. S. G. Gmelin. D'Orb. D'Orbigny. Godr. Godron. Dorst. Dorstenius. Gaepp. Goeppert. Dougl. Douglas. Good. Goodenough. Drej. Drejer. Gren. Grenier. Dr,/and. Dryander. Grev. Greville. Dufr. Dufresne. Griseb. Grisebach. Duham. Duhamel du Monceau. Gram. Greenland. Dumort. Dumortier. Gron. ) Gronovius. Dun. Dunal. Gronov. ) ABBREVIATIONS. 387 Guett. = Guettard. Jacq. = N. J. Jacquin. Guib. Guibord. Jacq.f. J. F. Jacquin, son. Guillem. Guillemih. J. St. Hit. Jaume St. Hilaire. Guimp. Guimpel. Jord. Jordan. Gunn. Gunnerus. Jumjh. Junghuhn. Guss. Gussone. Juss. A. L. Jussieu. Adr. Juss. Adrien Jussieu, son. Hagenb. Hagenbach. Hall. Haller. Ham. Hamilton. Hanb. Hanbury. Hanst. Hanstein. Hartm. Hartmann. Uartw. Hartweg. Harv. Harvey. Hass. Hassall. Kcemp. Karat. Kaulf. Kindb. Kirscfd. Kit. Kcdr. Korth. Kaempfer. Karsten. Kaulf uss. Kindberg. Kirschleger. Kitaibel. Kcelreuter. Kort lulls. Hassk. Hasskarl. Hausm. Hausmann. Haw. Ha worth. Hebens. Hebenstreit. Hedw. Hedwig. Kostel Kremp. Kromb. Kuetz. Kosteletzky. Krempelhuber. Krombholz. Kuetzing. Ilcijelm. Hegelrnaier. Hcydsch. Hegetschweiler. L. Linnaeus. Heist. Heister. Labill. La Billardiere Hddr. Heldreich. Lcest. Lsestadius. lldw. Helwing. Lag. Lagasca. Hemsl. Hemsley. Lull. Lallement. [Marck) Henck. Henckel. Lam. Lamarck (Monnet de La Henfr. Henfrey. Lamb. Lambert. Hensl. Henslow. Lamour. Lamouroux. Herb. Herbert. Langsd. Langsdorf. Herm. Hermann. La Peyr. La Peyrouse. Hild. Hildebrand. La Pyl. La Pylaie. Hochst. Hochs tetter. Ledeb. Ledebour. Hoffm. G. F. Hoffmann. Lehm. Lehmann. H. Hoffm. Hermann Hoffmann Lem. Lemaire. /Tbjfmanns.Hoffmannsegg. Lesq. Lesquereux. Hofm. Hofmeister Less. Lessing. Hohen. Hohenacker. Lestib. Lestiboudois. Holmsk. Holmskiold. Lie. Leveille. Homb. Hombron. L'Her. L'Heritier. Hook. Wm. J. Hooker. L'He.rm. L'Herminier. Hook.f. J. D. Hooker, son. Liebnt. Liebmann. Hopk. Hopkirk. Light/. Lightfoot. Hornem. Hornemann. Lillj. Lilijeblad. Hornsch. Hornsehuch. Lindb. Lindberg. Horsf. Horsfield. Lindbl. Lindblom. Houst. Houston. Lindenb. Lindenberg. Hontt. Houttuyn. Lindh. Lindheimer. Huds. Hudson. Lindl. Lindley. Hueb. Huebener. Linn. Linnaeus. Also L. Humb. Humboldt. Linn. f. C. Linnaeus, son. HBK \ Humboldt, Bonpland, and Lodd. Loddiges. ( Kunth. Lufl. Loefling. ABBREVIATIONS. Lees. = Lceselius. Naud. = Naudin. Lois. Loiseleur-Delongschamps. Neck. Necker. Loud. Loudon. Lour. Loureiro. Nees or ] N.abE. \ C. F. Nees von- Esenbeck. l.'ulir. Ludwig. T.Nees T.F.L. Nees von Esenbeck. Lumn. Lumnitzer. Nestl. Nestler. Lyngb. Lyngbye. Newb. Newberry. Newm. Newman. Macf. Macfadyen. Ncegg. Nceggerath. Macgil. MacGillivray. Nols. Noisette. Magn. Magnol. [stein. Nord. Nordstedt. M. Bieb. Marschall von Bieber- Not. Notaris. Marsh. Humphrey Marshall. Nutt. Nuttall. Mars. Marsili. Nyl. Nylander. Mart. Martius. Nym. Nyman. Mass. Massalongo. Mast. Masters. (Ed. CEder. Maxim. Maximowicz. (Erst. CErsted. Med. Medikus or Medicus. Oliv. Olivier. Meisn. I Meigner or Mei8Sner. Meissn. ) D. Oliv. Orb. D. Oliver. A. or C. d'Orbigny. Meneg. Meneghini. Orph. Orphanides. Mem. Menzies. On. Ortega. Mert. Mertens. Oudem. Oudemans. Metten. Mettenius. Mich. Micheli. P.deBeauv.Palisot de Beauvois. Michx. ) A d , Michaux> Poll. Pallas. MX. ) Pom. Panzer. Michx. f. F. A. Michaux, son. Park. Parkinson. Midden. Middendorff. Parl. Parlatore. Mill. Philip Miller. Pasq. Pasquale. Mill. J. John S. Mueller or Miller. Pav. Pavon. Miq. Miquel. Perl. Perleb. Mirb. Mirbel. Pers. Persoon. Mitch. John Mitchell. Philib. Philibert. Mitt. Mitten. Planch. J. E. Planchon. Moc. Mo9ino. G. Planch. Gustave Planchon. Molk. Molkenboer. Pluk. Plukenet. Mont. C. Montague. Plum. Plumier, Lat. Plumerius. Moq. Moquin-Tandon. Pcepp. Pceppig. Marie. Moricand. Pair. Poiret. Moris. Morison. Pott. Poiteau. Morr. Morren. Poll. Pollich. Moug. Mougeot. Post. Postels. Muell.Arg.J. Mueller of Argau. Pourr. PourreJ. F. Muett. Ferdinand Mueller. Pringsh. Pringsheim. G. MueU. Otto Mueller of Denmark. Pritz. Pritzel. MuU. Muhlenberg. Putter. Putterlich. Munt Munting. Murr. J. A. Murray. Rabenh. Rabenhorst. A. Murr Andrew Murray. Radlk. Radlkofer. Raf. Kafinesque-Schmaltz. Nacc. Naccari. Rasp. Raspail. Nxg. Naegeli. Red. Redoute'. ABBREVIATIONS. Reich. = Reichard. Scop. = Scopoli. Reichenb. H. G. L. Reichenbach. Seem. Seemann. Reichenb.f. H. G. Reichenbach, son. Sendt. Sendtner. Reinw. Reinwardt. Seneb. Senebier. Reiss. Reisseck. Ser. Seringe. Retz. Retzius. Seub. Seubert. Rent. Reuter. Sibth. Sibtliorp. Rich. L. C. Richard. Sieb. Sieber. R.icll-f] \ AchiUe Richard. A. Rich. ) Sieb. Siebold. Sdand. Solander. Richards. John Richardson. Sow. Sowerby. Richt. Richter. Spenn. Spenner. Ridd. Riddell. Spreng. Sprengel. Riv. Rivinus. Sternb. Sternberg. Rcehl. Roehling. Steud. Steudel. Ram. J. J. Rcemer. Stev. Steven. M. J. RcemM. J. Rceiner. Sull. Sullivant. ^f JJ- & 1 Rcemer & Schultes. Sw. Swartz. Rasp. Rceper. Targ. Targioni-Tozetti. Rohrb. Rohrbach. Ten. Tenore. Rostk. Rostkovius. Thorns. Thomas Thomson. Rothr. Rothrock. Thuill. Thuillier. Rottb. Rottboell. Thunb. Thunberg. Rottl. Rottler. Thurb. Thurber. Roum. Roumegere. Thurm. Thurman. Roxb. Roxburgh. Tod. Todaro. Roy. Royen. Torr. Torrey. Rudb. Rudbeck. Torr.fr Gr.Torrey & A. Gray. Rupr. Ruprecht. Tourn. Tournefort. Tratt. Trattinick. Sacc. Saccardo. Traut. Trautvetter. Sadl. Sadler. Trev. Treviranus. St. Hil. A. SaintrHilaire. Trin. Trinius. Salisb. Salisbury. Tuck. Tuckerman. Salm-Dyck.Prmce Jos. Salm-Riffer- Turcz. Turczaninow. Sauss. Saussure. [schied-Dyck. Turn. Turner. Schimp. Schimper. Turp. Turpin. Schk. Schkuhr. Schltcht. Schlechtendal. Vaitt. Vaillant. Schleich. Schleicher. VeiU. Veillard or Vieillard. Schomb. Schomburgh. Vouch. Vaucher. Schrad. Schrader. Vent. Ventenat. Schreb. Schreber. Via. Villars, or Villar. Schueb. Schuebeler. Vis. Visiani. Schult. Schultes. Vittad. Vittadini. Schidtz \ C. H. Schultz, Bipontinus Viv. Viviani. Bip. J (Zweibrucken) Vog. T. Vogel. Schum. Schumacher. Schnitzl. Schnitzlein. Wahl. Wahlenberg. Schwtegr. Schwaegrichea Wahlst. Wahlstedt. Schwein. Schweinitz. Walds. Waldstein Schweinf. Schweinfurth. Wall. Wallich. Schwend. Schwendener. Wattm. Wallman. 389 390 ABBREVIATIONS. Wallr. = Wallroth. Wdp. Walt. Wang. Warm. Wats. Walpers. Walter. Wangenheim. Warming. P. W. Watson. I I.C. Wats.R. C. Watson. 5. Wats. Web. Wedd. Weinm. Welw. Wonder. Wendl. Wiks. Sereno Watson. Weber. Weddell. Weinmann. Welwitsch. Wenderoth. Wendland. Wikstrom. Wildb. = Wildbrand. Willd. Willdenow. WillL Willkomm. Wits. Wilson. Wlmm. Wimmer. Wisliz. Wislizenus. With. Withering. Woodv. Woodville. Wulf. Wulfen. Zanard. Zanardini. Zetllrst. Zetterstedt. Zucc. Zuccarini. Zuccay. Zuccagini. 2. ABBREVIATIONS OF NAMES OF ORGANS AND TERMS USED IN BOTANICAL WRITINGS. sEst. Estate, in summer. sEst. Estivation. Alb. Albumen. Anlh. Anther. Art. Artificial. Auct., Auctt. Auctorum, of authors. Aut. Autumnal. B. or Btat. Beatus, " the late," re- cently deceased. B>: Bract. Cat. Calyx. Cel. Celeberrimus, or Very cele- brated. Cent. Centimetre. Cl. Clarissimus. Char. Character. Coll. Collection. Car. Corolla. Cult. Cultivated. Decim. or Dec™. Decimetre. Descr. Description. Diff. Differentia?, the distinguishing marks. Ed. Edition. Embr. Embryo. Ess. Essential, as Char. Ess. Excl. Excluding, or being excluded. Excl. Syn. Excluding the synonym or synonyms. Fam. Family. Fil. Filament of the stamen. Fl. Flower (flos) ; Flora, or some- times Floret, it flowers. Faem. Female plant, flower, &c. Fol. Folium, leaf. Fr. Fruit. Fructif. Fructification. Gen. Genus or Generic. Germ. Germen, Linnaean name for ovary ; also Germination. H. Herbarium. Hub. Habitat, place of growth ; sometimes for Habeo, I have. Herb. Herbarium. Hort. Hortus, garden. Hortul. Hortulanorum, of the gar- deners. Ic. Icon, a plate or figure. 77/. Illustris, illustrious. Ined. Unpublished. Inf. Inferior. Inf.. Inflorescence. Inv. Involucre. Lat. Lateral, or relating to width. Lin. Linea, a h'ne (the 12th of an inch). Lit., Lift. In a letter or letters. /. c. Loco citato, in the place cited. ABBREVIATIONS. 391 Muse. Male plant, flower, &c. Mill, or mm. Millimetre. J/ss. Manuscripts. Mus. Museum. .V. or No. Number. Xat. Natural. ATom. Nomen, name. Obs. Observation . Ord. Order. Oo. Ovary. /'. Page, or sometimes Part. Fed. Peduncle or Pedicel, or for Pedalis, a foot long or high. Peric. Pericarp. Perig. Perigonium. Pet. Petal or Petiole. Pist. Pistil. Plac. Placenta. Poll. Pollicaris, an inch long. p. p. Pro parte, in part. Prodr. or Prod. Prodromus. Rud. Radix, root ; or Radical. Ram. Ramus, branch. s. Seu, or Sive, Latin for or. Sect. Section. Segm. Segment. Sem. Semen, seed. Sep. Sepal. Ser. Series. Sice. Siccatus or Siccus, dried or dry Spec, or Sp. Species, or specimen. Spont. Spontaneous. Stain. Stamen or Staminate. Sup. Superior. Syn. Synonym or Synopsis. T. or Tab. Tabula, plate. T. Tomus, volume. V. Volume : sometimes for Vel, or ; sometimes Vide, see. Far. Variety. Veg. Vegetation, characters of Fern. Vernal. v. s. Visa sicca, or Vidi siccam. v. v. Visa viva, or Vidi vivam ; the first indicating that a dried speci- men of the plant, the second that the living plant has been exam- ined. v. s. c. and v. s. s., indicates that the dried specimen was cultivated (c) or spontaneous (s). v. v. c. and v. v. s., that the living plant seen was cultivated (c) or spontaneous (s). SIGNS. 1. SIGNS USED BY LINN^US. O An annual plant. $ A biennial. ^ A perennial. b A tree or shrub. * Affixed to a reference, means that a good description will be found there. f Indicates an obscure or doubtful species. 2. SIGNS USED BY DECANDOLLE AND LATER WRITERS. O A monocarpic plant, i. e. which dies after once flowering and fruiting, either annual or biennial, or of longer duration. CD Annual. ® Biennial. @ Monocarpic perennial, such as Agave. 392 SIGNS. 11 Perennial herb. •^ Suffrutex, an nndershrub. 5 Fritter, a shrub. 5 Arbuscula, a tree-like shrub of ten to twenty-five feet in height. 5 Arbor, a tree. r\ A climbing plant. A An evergreen. $ Male plant or flower. J Female plant or flower. '$ Hermaphrodite plant or flower. co Indefinitely numerous, e. g. co-andrd, polyandrous. ? A sign of doubt. " Thalictrum ? Japonicum," doubts if the plant is really a Thalictrum. " Thalictrum Jnponicum, Thunb.?" doubts if the plant in hand is truly tlie species of Thunberg. Thalictrum Japoni- cum, Thunb., Willd.? doubts whether Willdenow's T. Japonicum is really that of Thunberg. ! A sign of certainty. As " Thulictrum anemonoides, Michx. ! Fl. Bor. Am. p. 322," as used by DeCandolle, affirms that he has seen an authentic original specimen of this author. Affixed to the name of a collector, as " Virginia, Clayton ! " it affirms that the writer has examined a specimen collected by the person to whose name it is appended. — Between two figures, as in " Stamens 6-10," indicates the extremes of difference, as that the stamens are from five to ten. 0 ' /; The signs for degrees, minutes, and seconds, as 1°, 2', 3", are used in Gray's Manual of Botany of the Northern United States, for feet (°), inches ('), and lines ("). With European authors, usually the sign for minutes is for leet ; that of seconds for inches : thus V, a foot high ; I", an inch long ; and 1'", a line long. O= Cotyledons accumbent to the radicle. OH Cotyledons incumbent on the radicle. GLOSSARY, OR DICTIONARY OF BOTANICAL TERMS, ENGLISH AND LATIN, COMBINED WITH AN INDEX. THIS Glossary is intended to contain all the principal technical terms (substan- tive as well as adjective) of structural and systematic Botany, as far at least as concerns Phfenogamous plants. Most of the special terms relating to the lower Cryptoijamia and to Vegetable Anatomy and Physiology are relegated to the vol- umes devoted to those departments. The annexed numbers refer to pages of this volume. Very many of the terms are seldom employed, or are wholly out of use. The principal Latin terms are given separately only when there is no English equiv- alent differing merely in the termination. When the word is essentially the same, the Latin termination (of adjectives in the nominative masculine only) is annexed to the English word in a parenthesis. The changed termination goes back mostly to the penultimate consonant. It is unnecessary in a work like this to accentuate all the technical words; but, in the case of words liable to mispronunciation, an accent-mark is placed over the syllable which takes the principal accent. The glossary, as here drawn up, may serve to indicate the meaning of the commoner descriptive specific names of plants. A, privative, as the initial in many words of Greek derivation, signifies the absence of the organ mentioned ; as, .4petalous, destitute of petals ; Aphyllous, leafless. In words be- ginning with a vowel, this prefix is changed to an; as, ^4nanthous, flow- erless ; u4nantherous, antherless. Abbreviations, 385. Aberrant. Wandering, applied to spe- cies, genera, &c., which differ in some respect from the usual or normal char- acter of the group they belong to. Abnormal (Abnormis). Differing from the normal or usual structure. Aboriginal. Strictly native ; indigenous. Abm-tion (Abortus). Imperfect develop- ment or non-development of an organ ; 179, 187. Abortive (-ivus). Defective or barren. Abrupt (Abruptus). Terminating sud- denly ; the opposite of tapering. Abruptly pinnate. Pinnate without a terminal leaflet or appendage ; 101. Acanthdcladous (-us). Having spin\' branches. Acanthopkorous (-us). Spine-bearing. Acaulescent (-ens). Stemless, or appar- ently so, with no proper caulis ; 45. Atiaulis. Stemless ; same as Acaulescent. Accessory. Something additional, or of the nature of appendage. Accessory Buds. 44. Accessory Fruits, 300. Accrescent (-ens). Increasing in size with age, as often occurs with the calyx after flowering. Accrete (-us). Grown together, or con- solidated with some contiguous body. Accumbent (-ens). Lying against an- other body. Accumbent Cotyledons. With edges against the radicle; 313. Acephalous (-us). Headless. 394 GLOSSARY. Acerose (-6sus). Needle-shaped, like the leaves of Pines. Acetdbuliform (-ormis). In the form of a shallow open cup or saucer. Ach(enium or Achenium. A small, dry and hard, one-celled, one-seeded, inde- hiscent fruit ; strictly one of a single and free carpel; but extended to simi- lar ones of more than one carpel, and also with adnate calyx; 294. (Achce- niiuii is etymologically the proper orthography ; but achenium is be- coming the commoner form.) Acheenocarp (-arpium). General name of a dry and indehiscent fruit ; 292. Achenodium. Such a double achenium as that of Umbelliferae ; a Cremo- carp. Achlamydeous (-eus). Destitute of peri- anth; 191. Acicula. A bristle. Acicular (-am). Bristle-shaped, or slen- der needle-shaped. Acindciform (-ormis). Scymitar-shaped ; curved with rounded point, thicker on the straighter edge than on the con- vex edge. Adnosus. Like grapes or grape-seed. Acinus. Classically a berry, particu- larly a grape, or its stony seed, or a bunch of berries ; now sometimes ap- plied to the separate carpels of an aggregate baccate fruit, or to the con- tained stone or seed ; 297. Acorn. Fruit of the Oak. Aci>tyledon,p\. Acotyledons, Acotyledones. A plant or plants destitute of coty- ledon, or Acotyledonous (-eus). Without cotyle- dons ; as the embryo of Cuscuta ; 26, 38. Mostly applied, as by Jussieu, to plants which have no proper seed nor embryo, and therefore no cotyle- don ; 339. Acrampkibrya. Plants producing side as well as terminal buds or growths ; 341 Acrdbrya. Plants growing from apex ^only; 341. Acroyen (Acrdgence). Name of class of plants which in growth are said to be Acroyenous. Growing from the apex or by terminal buds only. Acrosdrcum. Desvaux's name for a berry from an ovary with adnate calyx. Acroapira. An old name of the plu- mule of a grain in germination. Aculeate (-eatus). Prickly; beset with aculei. Aculeosus. Abounding with prickles. Aculeolate (-atus). Beset with diminu- tive prickles, or Aculeoli. Diminutive of aculei. • Aculeus. A prickle ; a pointed small excrescence of the bark. Acumen. A tapering point. Acuminate (-atus). Ending in a tapering point; 96. Acute (Acutus). Terminating in an acute angle; 97. Acropetal. Developing from below up- ' ward, or from base toward apex. Actinomorphous (-us). Capable of bi- section through two or more planes into similar halves, as is a regular symmetrical flower; 175. Acutiusculus. Somewhat acute ; acutish. Adelphous (-us, Adelphi, brothers). Sta- mens with coalescent or clustered fila- ments are monadelphous, diadelphous, &c., according to the number of Adel- phia or brotherhoods. Aden. Greek for gland, is compounded with Greek words with this meaning ; as, Adendphorus, gland-bearing; Ad- enophyllus, leaves bearing glands, &c. Adylutinate (-atus). Same as accrete Adherent (Adhcerens). Generally same as adnate ; may refer to adhesions not congenital. Adnate (-atus). Congenitally united to ; as the calyx-tube of the gooseberry to the ovary; 182. Adnate anther is one seemingly borne on the outer or inner face of the filament ; i. e. extrorsely or introrsely fixed by its whole length to the connective ; 253. Adnation. The state of being adnate; 179, 181. Adpressus. Latin of appressed. Adscendens. Latin of ascending. Adsurgens. Latin of assurgent. Adventitious, Adventive. That which has come from abroad or as a stranger ; as a plant lately or by chance introduced from another country. Adventitious Buds ; 45. jEquilaterdlis. Equilateral, equal-sided. jEqualijldrus. When all the flowers of the same head or cluster are alike in form as well as character. ^fjudlix, JEqunm. Equal ; equalling. Aerial roots, &c. ; 33. Aerophytes. Air-plants; 35. JEruyinosus. Verdigris-colored. jEstival (-dlis). Relating to summer. ^Estivation (-io). The disposition of the parts of a flower in the bud ; 132. GLOSSARY. 395 JZterio. A form of aggregate fruit ; 300. sEtheogamia, jEthogamvus, 340. Affinity. True and near relationship; 327, 330. Agamous or Agamic. Destitute of sexes . Agglomerate. j Heaped or crowded Aggregate(^tus). } 'nto a denf cluster' I. but not cohering. Aggregate Fruits. Those formed of aggregate carpels of the same flower ; 298, 301. Agrtstis. Growing in fields. Air-plants. Plants unconnected with the ground; 35. Akene, Akenium. See Achaenium. Ala (pi. aloe). A wing. Also the side petals of a papilionaceous corolla; 185. Has also been used in the sense of axilla. Alabdstrum. A flower-bud ; 40. Alar (Alaris). From ala in the sense of axilla, therefore axillary or in the forks. Alate (-atus). Winged. Albtscens, Albicans. Whitened, whitish, or hoary. Albumen of the seed. Any deposit of nutritive material within the seed- coats, and not in the embryo; 14, 309. Albuminous or Albuminose (Albuminosus). Said of seeds provided with albumen ; 13,309. • llburnum. Sapwood ; the newer wood of an exogenous stem ; 80. Albus. White. Allagostemonous. With stamens alterna- tively inserted on the torus and on the petals. Alliaceous (-ens). Having the smell of garlic. Alliance. Synonym of Cohort; 326. Allogamy. Fecundation of the ovules of a flower by other than its own pollen ; cross fertilization, 216. Alptstrine (Alpestris). Growing on mountains below an alpine region or one unwooded from cold. Alpine (-inns). Growing on the higher parts of the Alps, or (by extension of meaning) on other mountains above the limits of trees. Alternate (Altemus). One after an- other; as of leaves placed singly instead of in pairs (opposite) or in whorls. Also, standing before inter- vals ; as stamens alternate with petals instead of before them ; 6, 119. Alternative (-icus). In aestivation, with an inner whorl alternating with an outer one; 134, 136. Alv*olate(-atus). Honej-combed ; having deep angular cavities (Alveoli) sepa- rated by thin partitions, as the recep- tacle of cotton-thistle. Ambitus. The ray or circumference of a head, &c. Ament (Amentum). A catkin, or pe- culiar scaly spike ; 150. Amentaceous (-eus). Bearing catkins, or catkin-like. Amorphous (-us). Shapeless; of in- definite form. Amphanthium. One of the (needless) names coined for a dilated receptacle of inflorescence. Amphibrya. Equivalent to Monocotyle- dones; 341. Amphicarpous (-us). Producing two kinds of fruit. Amphigamous Cryptogams, 340. Amphigastria. Peculiar leaves (of He- paticse) imitating stipules. Amphisarca. A hard-rinded berry, or fruit succulent within and woody or crustaceous without, as a calibash. Amphispermium. Link's name for a one-seeded pericarp which is con- formed to the seed ; an akene. Amphitropous (-us), wrongly Amphi- tropal. Turned both ways; applied to an ovule with hiluin intermediate between micropyle and chalaza ; 279. Amphora. A pitcher; and the lower part of a pyxis. Amplectens, Amplexans, Amplexus. Em- bracing, clamping. Amplexicaul (-aulis). Clasping a stem, as does the base of certain leaves. Ampliate (-atus). Enlarged or dilated. Ampulla. A bladder or flask-shaped organ, as of Utricularia. Ampullaceous (-us), or Ampullteform. In the form of a bladder or short flask. Amylaceous (-eus). Resembling or com- posed of starch, or Amylum. Amyloid. Analogous to starch. Analogy (Analogia). Likeness in cer- tain respects. As distinguished from ' affinity, it means resemblance in cer- tain respects only, not in the plan of structure. Thus, a Ranunculus is analogous to a Potentilla, but there is no near affinity or relatio: ship be- tween the two. And the tendril of a Pea, that of a Smilax, and that of the Grape-vine are analogues ; i. e , are analogous organs, but are not homo- logues ; for the first answers to a leaf, the second to stipules, and the third 396 GLOSSARY. to a stem. The spur of a Larkspur is analogous to one of the five spurs of Columbine, but not homologous with it; for the first is a sepal, and the second a petal. Anandrous. Destitute of stamens. Anantherous (-us). Destitute of anthers. Ananthous (-us). Flowerless. Anastomdsis. The connection of veins, &c., by cross-veins, forming reticu- lation. Andtropous (-us), wrongly Anatropal. The reversed ovule, with micropyle close by the side of the hilum, and chalaza at the opposite end; 279. Anceps, Engl. Ancipital. Botanically always used in the sense of two- edged. Ander, andra, andrum. In Greek com- pounds, the male. Andro-dicecious. With flowers on one plant hermaphrodite, and on another staminate only ; 191. Andraecium. The stamens of a flower collectively ; 165, 249. Androgynous (-us). Said of an inflores- cence composed of both male and female flowers. Androphore (Andrdphorum). A sup- port or column on which stamens are raised. Androus. See Ander. Anemophilous. Literally wind-loving. Said of flowers which are fecundated by wind-borne pollen ; 217. Anfractuosus. Abruptly bent hither and thither, as the stamens of Cucur- bita. Angiocarpous (-us). When a fruit is covered by some envelope. Angiospermia. A Linnsean artificial order; 337. Angiospermous, Angiospermce, Angio- sperms. Plants with seeds borne in a pericarp; 259. Angular Divergence of leaves ; 123. Anisomerous (-us). Unequal in number in the different circles of the flower; unsymmetrical. Anisopetalous (-us). With unequal petals. Anisophyllous (-us). Unequal-leaved; i. e., the two leaves of a pair unequal. Anisostemonous (-M.I). When the sta- mens are not of the number of the petals. Annotinus. A year old, or in yearly growths. Annual (Annuus). Of only one year's duration; 30. Annular (-aris). In the form of a ring; or marked transversely by rings. The latter more properly. Annulate (-atus). Marked with rings. Annulus. A ring, such as that with which the sporangia of some Ferns and Mosses are furnished. Anophytes (Anophyta). Name of group comprising Mosses, &c. Anteposition Same as Superposition; 179, 195. Anterior, as to position, denotes the front side, or averse from the axis of inflo- rescence; 160. Anthela. A deliquescent and paniculate cyme, with median ramification, and the lateral axes overtopping the central, as in Juncus tenuis, &c. May be either a Drepanium or a Rhi- pidium. Anthemy, Anthemia. A flower-cluster of any kind ; 144. Anther (Anthera). The polliniferous part of a stamen; 165, 251. Antheridium. An analogue of the an- ther in Cryptogams. Antheri/'erous (-us). Anther-bearing. Anthesis. The time at which a flower is perfected and opens; or the act of expansion of a flower. Anthocarpous(-us),Anthocarpmm. Fruits in which some organ exterior to the pericarp is concerned ; 300. Anthoclinium. Name of a receptacle of inflorescence, such as that of Com- Anthodium. A name for the head of flowers (or so-called compound flowers) of Composite; 147. Antholysis. A retrograde metamor- phosis of a flower, in which normally ^ combined parts are separated. Anthophore (Anthdphorum). The stipe when developed between calyx and corolla; 212. Anthus or Anthos. A flower, in Greek compounds. Anticous (Anticus). Facing anteriorly ; 253. Antitropous (-us), less properly Antitro- pal. Said of an embryo with radicle pointing to the end of the seed oppo- site the hilum; 312. Antrorse. Directed upward or forward. Apetalous (-us). Having no petals ; 190. Apex. Besides its ordinary meaning, the top of a thing, it was once the technical name of an anther; 166. Aphyllous (-us). Leafless. GLOSSARY. 397 Apical (-alls). Relating to the apex or tip. The name for anthers anterior to Ludwjg and Linnaeus ; 166. Apiculate (-us). Ending in a short pointed tip or apicula. Apocarpous (-us). When carpels of a gynceeium are separate ; 261, 262. Apophysis. An enlargement or swelling of the surface of an organ at some par- ticular part. Apothecia. The "shields " or fructify- ing disks of Lichenes. Apotropous (-us). Said of an anatropous ovule which when pendulous has rhaphe averse ; 282. Appendage, Appendix. Any superadded or subsidiary part. Appendiculate (-atus). Furnished with a small appendage (Apptndiculum), or with any appendage. Appositus. Placed side by side. Appressed (Lat. Adpressus). Lying flat against or together for the whole ^length. Jlpricus. Growing in dry sunny places. Apterous (-us). Wingless; not alate. Aquatic (-icus). Living in water. Aqudtilis. Living under water. Arachnoid (-oideus). Cobwebby ; com- posed of slender entangled hairs. Araneose (-osus), Araneus. Like spider- web; same as Arachnoid. Arbor. A tree ; 50. Arboreous (Arboreus). Tree-like, or re- lating to a tree. Arborescent (-ens). Tree-like; approach- ing the size of a tree. Arboretum, also Arbustum. A place where trees are -grown ; an arranged collection of trees. Arbuscula. A small shrub of tree-like growth or form. Arbuscularis. Ramified like a little tree. Archegonium. The spore-case of mosses, &c., in an early state. Arcuate (-atus). Moderately curved, as if bent like a bow. Areola, pi. Areolential components of bast ; long and flexible but thick-walled attenuated cells; 77. Beak. A narrowed or prolonged tip. Beaked. Ending in a beak. Bell-shaped. Same as Campanulate ; 249. Berried. Baccate. Berry. A fruit, the whole pericarp of which is fleshy or pulpy; 299. Bi- or Bis. As a prelix to Latin words (Greek words have Di-)> two, twice, or doubly. Biacuminate (-atus). Two-pointed, as malpighiaceous hairs, fixed by the middle and tapering to each end. Blarticulate (-atus). Two-jointed. Biduriculate (-atus). Two-auricled. Bibracteate <-atus). With two bracts. Bibracteolate (-atus). With two bract- lets. Bicallose (-osus). With two callosities. Bicdrinate (-atus). Two-keeled. Biceps. | With two supports or stalks, Bicipital. i or two-headed. Bicolor. Two colored. Biconjuijate (-atus). Twice paired. Bicmtiis. Two-horned. Bicornute. Same as preceding. GLOSSARY. 399 Bicruris. Two-legged, or with two sup- ports. Bidentate (-atus). Having two teeth. (Not doubly dentate.) Biduus. Lasting two days only. Biennial (Biennis). Of two years' dura- tion ; 31. Bifarious (-ius). Two-ranked; in two vertical rows. Biferus. Double-bearing ; fruiting twice a year. Bifid (-idus). Two-cleft, to the middle or thereabout. Biflorous (-us). Two-flowered. Bifoliate. Two-leaved. Bijoliolate. Of two leaflets. Biforate (-atus). Having two open- ings. Bifoi-mis. Two-formed ; in two shapes. Bifrons. With two faces or aspects. Bifurcate (-atus). Two-forked ; i. e. of two prongs or forks. But it may mean bisfurcatus; i.e., forked and again forked. Bigeminate (-atus). Twice twin ; same as Biconjugate. Bigener. The offspring of a cross be- tween two generically different plants. Bijugate (Bijuyus). Two-paired, as a pinnate leaf of two juga or pairs of leaflets. Bilabiate (-atus). Two-lipped; 247. Bildmellate (-atus), or Bilamellar. Of two plates or lamellae. Bilobed (Bilobus), or Bilobate. Of two lobes, or cleft into two segments. Bilocellate. Divided into two locelli; 263. Bilocular (-am). Two-celled. Bimestris. Lasting two months. Bimus. Lasting two years ; two years old. Binary (-arius). Consisting of two members; 176. Binate (-atus). -In pairs or twos. Bini. .Twin, or two together. Binodal (Binodis). Having two nodes. Binomial Nomenclature, 346. Biology. The natural history of plants and animals, i. e. of living things ; 1. Bipalmate (-atus). Twice palmately com- pound. Biparous. Bearing two; as a cyme of two rays or axes ; 152, 155. Bipdrtible (-ibilis). Capable of division into two similar parts. Bipartite (-itus). Divided almost into two pieces ; two-parted. Bipes. Same as Bicrurit. Bipinnate (-atus). Doubly or twice pinnate; 103. Bipinndtifid (-idus). Twice or doubly pinnatitid; 100. Bipinndtisect (-us). Twice pinnately di- vided. Biplicate (-atus). Twice folded or plaited. Biporose (-osus). Opening by two pores. Biradiate (-atus). Of two rays. Birimose (-osus). Opening by two slits. Bisected (-us). Completely divided into two parts ; 99. Biseptate (-atus). With two partitions. Biserial (-ialis), or Biseriate (-iatus). In two series, one above the other. Biserrate (-atus). When serratures are again serrate ; doubly serrate. Biseomal. Having both stamens and pistil; hermaphrodite; 191. Bisulcate (-atus, Bisulcus). Two-grooved; having two furrows. Biternate (-atus). Twice ternate. Bladdery. Thin and inflated. Blade. The lamina, limb, or expanded portion of a leaf, &c. ; 85, 245. Blastema. The budding or sprouting part or point. First used for the axia of an embryo ; now used for the ini- tial growth out of which any organ or part of an organ is developed. Bloom. Besides its use as equivalent to blossom, it denotes the white powdery and glaucous covering of the surface of many fruits and leaves, of a waxy nature. Boat-shaped. Of the shape of a boat, of the deeper sort, with or without a keel. Bostrychoidal. Having the form or char- acter of a ringlet, or Bostryx ; 157. Bosti-yx. An uniparous helicoid cyme ; 156. Bothrenchyma. Tissue of plants com- posed of dotted or pitted ducts. Botry-cymose. Racemes or any botryose clusters cymosely aggregated ; 159. Botryose (-osus), Botryoidal. Of the ra- cemose type ; 144, 145, 146, 153. Botrys. The equivalent of Raceme ; 146. Botulifoi'mis. Sausage-shaped. Brdchiate (-idtus). With spreading arms, as branches (especially opposite and decussate) widely diverging. Brachys. Greek for short, and used in compounds; as, Brachypodus, short- stalked. Bract, Bractea. The leaves (more or less modified) of a flower-cluster; 118, 141. Bracteate (-eatus). Having bracts. 400 GLOSSARY. Bracteola, Bracteole. See Bractlet. Bracteolate(-atus). Having bractlets. Bractlet. A bract of the ultimate grade, j as one inserted on a pedicel or ultimate i flower-stalk, instead of subtending it; 141, 142, 160. Bracteose (-<>sus). Full of, or with con- spicuous bracts. Branches. Secondary axes, or divi- sions of an axis; 47. Branchlets. Ultimate branches or divi- sions of an axis ; 47. Breathing-pores. See Stomata, 89. Bristle. A stiff hair, or any slender body or outgrowth which may be likened to a hog's bristle. Bristly. Beset with bristles. Briinneus. Deep brown. Brush-shaped. See Aspergilliform. Bryology. The botany of Mosses. Bud. The undeveloped state of a stem or branch, with or without leaves; 6, 40. Bud-scales. The teguments of a bud ; 40. Bulb (Bulbus). A leaf-bud (commonly subterranean) with fleshy scales or coats ; 43, 62. Buibiceps. A stem with bulbous base. Bulbiferovs (-us). Bulb-bearing. Bulbillus, Bulbulus. Diminutive bulb. Same as tfulblet. A small bulb, especially such as is produced in the air, in the axil of or- dinary leaves, or upon them ; 63. Bulbodium. A synonym of Corm, the "solid bulb." Bulbo-tuber. Synonym of Corm. Bulbous, Bulbosus. Having bulbs or the structure of a bulb. Bullate (-atus). Said of a puckered sur- face (as if blistered), thrown into por- tions which are convex and projecting j on one side and concave on the other. Also used in specific names, in its more literal sense for inflated. Bursicula, A small pouch (bursa); such as that which encloses the disk or gland of the caudicle of the pollin- ium of an Orchis. Bnrsiculatus. Furnished with a bursi- cula or pouch. Bijssaceous (-eus). Composed of fine threads, like byssus or fine flax. Caducous (-us). Dropping off very early, as the calyx of a Poppy at the time of expansion ; 243. Cceruleut. Sky blue, or pure blue. Ccesius. La vender- color ; pale green with whitish or gray. Calathidium, Cdlathis. Literally a bas- ket; a name for the head of flowers (or better for the involucre only) of Com- Calathiform (-ormis). Cup-shaped ; of somewhat hemispherical outline. Calcar. A spur; mostly used for the nec- tariferous one of a calyx or corolla. Cdlcarate (-atus). Furnished or pro- duced into a spur. ^Cdlceolate (-atus), or Calceiformis. Shaped like a slipper or shoe. Callose (-osus). Bearing callosities (calli), or hard protuberances. Calvus. Bald, as an akene without pappus. Calycdnthemy. Name of the monstros- ity in which the calyx imitates an exterior corolla ; 174. Calyciflorous (Calyciflorce), 340. Calycine ( Calycinus). Relating to calyx. Calyculate (-atus). Bearing bracts next the calyx which imitate an external or accessory calyx. Calyculus. An involucre or involucel imitating an additional calyx. Culyptra. The hood or veil of the spore-case of a Moss ; or some cover- ing body like it. Calyptrate (-atus). Furnished with a calyptra, or something like it. Calyptriform (-ormis). Calyptra-shaped ; as the calyx of Eschscholtzia. Calyx. The flower-cup, the exterior perianth; 164. Cdmara and its diminutive Camerula (chamber) are sometimes used for the cells of a fruit. Cambium. Old name' of the viscid mat- ter between bark and wood in com- mon trees or shrubs in spring; now used for the nascent structure there forming, or Cambium layer; 78. Campdnulate (-atus). Bell-shaped; elon- gated cup-shaped or shorter, and broad from the base ; 249. Campanif ormis. Same as Campanulate. Campylospermous (-us). Curved-seeded. Said of seed-like fruits or carpels, as those of some Umbellifere, in which the contained seed is involute by the lateral edges, so as to produce a longi- tudinal furrow on the ventral face. Campylotropous (-us), or less correctly Campylotropal, or Campulitropous. An ovule or seed which is curved in its formation so as to bring the GLOSSARY. 401 micropyle or true apex down near to the hifum; 279. Canalicaulate (-atus). Channelled, or with a longitudinal groove. Cdncellate (-dtus). Latticed; resembling lattice-work. Cdndidus. Pure white. Canescens. Hoary, usually with gray pubescence. Canus. Graj'-white; whiter than the preceding. \ So slender that it Capillaceous(-eus). I may be compared Capillary (-aris}. < with the hairs of ) animals. Capitate (-atus). Head-shaped, or col- lected in a head; 147. Capitellate (-atus). Diminutive of Cap- itate. Cupitulum. A head of or simple globu- lar cluster of sessile flowers ; 147. Capreolate (-atus). Bearing a tendril (capreolus). Capsule (-ula). A dry and dehiscent pericarp composed of more than one carpel ; 289, 293. Capsular. Of the nature of, or relating to, a capsule. Capsuliferous. Capsule-bearing. Carcerulus. An unused name for an indehiscent and several-celled dry fruit; 297. Carina. A keel; used either for the two combined lower petals of a papil- ionaceous corolla (185); or for a sa- lient longitudinal projection on the centre of the lower face of an organ, as on the glumes of many Grasses. Cdrinate (-aim). Keeled. Cariopsis or Caryopsis. A grain ; a seed- like fruit with thin pericarp adnate to the contained seed ; 295. Carneus. Flesh-colored, very pale red. Caro. Flesh, as the pulp of a melon, or the fleshy part of a drupe. Carpadilium. Synonym of Cremocarp. Carpel, Carpellum. A simple pistil, or an element of a compound pistil, an- swering to one leaf ; 167, 260. Carpid, Carpidium. Synonym of carpel. Carpolngy. The botany of fruits. Carpophore (Carp6pho>-um). A portion of receptacle prolonged between the carpels; 212. Cdrpophyll ( Carpophyllum). Literally fruit-leaf ; synonym of Carpel ; 260. Cartilaginous or Cartilagineous (-eus). Of the texture of cartilage or gristle ; firm and tough. Caruncle (Caruncula). An excrescence at or about the hilum of certain seeds ; 308. Caryophyllaceous (-eus). Resembling or relating to the corolla of Dianthus Caryophyllus (246), or to the Pink family. Caryopsis. See Cariopsis. Cassideus. Helmet-shaped. Cansus. Empty, as an anther contain- ing no pollen. Castrate (-atus). Said of a stamen which wants the anther. Catapetalous (-us). Where petals are united only by cohesion with united stamens, as in Mallow. Cataphylla. Answers to the German " Niederblatter," or under-leaves, those at the beginning of a growth, cotyledons, bud-scales, scales on rhi- zomes, &c. ; 6. Catenulate (-dtus). Formed of parts united end to end, like the links of a chain. Catkin. A scaly spike (see Ament); 150. Caudate (-atus). Furnished with a tail (cauda), or with a slender tip or ap- pendage resembling a tail. Caudex. A trunk or stock of a plant ; 50. Caudicle (Caudicula). The stalk of a pollinium, &c. Caulescent (-ens). Having an obvious stem. Caulicle (Cauliculus). The initial stem in an embryo, generally named the Radicle; 10. Cauline (-inus). Belonging to the stem. Caulis. Greek form Caulon. The stem of a plant. Caulocdrpic or Caulocarpous. Applied to plants which live to flower and fructify more than once or indefi- nitely. Cauldme, Cauloma. The stem-part of a plant. Cephalanthium. One of the names of the head or capitulum in Composite ; 148. Cell ( Cellula). The anatomical element of plants; 28. The cavity of an anther which contains the pollen, or an anther-lobe, thus taken in the sense of the circumscribing wall as well as the cavity; 251, 254. The cavity, or any one cavity of an ovary; or pericarp, containing the ovules or seeds; 262. Cellular Plants, Cellulares, 340. Cellule (-ula). Diminutive of cell; of 402 GLOSSARY. the same meaning as Cell in vegetable anatomy ; 28. Cellulose. The material, chemically con- sidered, of which the wall of the cell consists. Cenobium. A name of the peculiar four- parted fruit (or the four nutlets around a common style) which distinguishes Labiate and Borraginaceae. Centrifugal. Tending or developing from the centre outward. Centripetal. Tending or developing from without toward the centre. Cephalanthium. Synonym of Antho- dium. Ceratium. A siliquiform capsule, such as that of Corydalis, Cleome, &c. Cereal. Belonging to corn and the allied grains. Cerinus. Of the color of wax. Cernuous (-uus). Nodding. Chceta. Greek for a bristle, Latin Seta. Chaff. Small scales; dry and depau- perate bracts ; such as those on the re- ceptacle of a sunflower and many other Composite ; also glumes of Grasses. Chaffy. Provided with or having the texture of chaff. Chaldza. The part of an ovule where coats and nucleus are confluent; 277. Channelled. Hollowed out longitudi- nally like a gutter. See Canaliculate. Character. A diagnostic description, or the enumeration of essential differ- ences; 361. Chasmoyamy. The opening of the peri- anth at flowering time ; the opposite of Cleistogamy. Chartaceous (-eus). Having the texture of writing-paper. Chlorophyll. The green matter of leaves and other vegetation ; 76, 88. Chloros. Greek for green. Enters into compounds, such as Chloranthus, green-flowered, C/iloranthy, same as Chlorosis, as when petals turn green ; 172. Chlorosis. Literally becoming green, as some flowers in retrograde metamor- phosis. Also used contrariwise for the loss of a normal green color; 172. Chorda Pistillaris. A line of tissue reach- ing from stigma to ovary. Choripetalous (-us). Same as Polypeta- lous, i. e. petals unconnected : 244. Chonsepalous. Same as Polysepalous.&c. Ckorisis. The separation of a leaf or phyllum into more than one ; 202. Ch oristophyllus. Separate-leaved. Chromule (-ula). Coloring matter of plants other than chlorophyll, espe- cially that of petals. Chrysos. Greek for golden, or golden- yellow ; as Chrysanthus. Yellow-flowered, &c. Cicatrix, Cicatricula. A scar left by the fall of a leaf or other organ. Ciliate (-fitus), Ciliaris. Marginally fringed with hairs. Ciiium, pi. cilia. Marginal hairs, form- ing a fringe, like the eyelash. (The , name has been extended in scientific books to undividual hairs, and of a surface as well as edge.) Cincinnus. A curl : name of a uniparous scorpioid cyme, which is Cincinnal ; 156, 157. Cinenchyma. Laticiferous tissue. Cinerascens, Cineraceus. Ash-grayish. Cinereous (-eus). As^h-gray. Clnnabarinus. Cinnabar-color ; scarlet touched with orange. Circinal (-alls). Involute from the tip into a coil; 133. Circinate, or Circinnate (-atus). Same as preceding; or sometimes meaning coiled into a ring only. Circumscissile, or Circumcissile ( Circum- scissus). Cut circulary and transverse- ly; divided transversely ; 25)1, 293. Circumscription (-io). The general out- line of the margin of a flat body. Cin-hiferwu (-us) and Cirrhose. Tcn- "dril-bearing. Cirrhus. A tendril ; 54. Citreus, Citi-inus. Lemon-colored. dados. Greek for branch ; whence such terms as Cladodium. Same as Cladophyllnm. Clddophyll, Cladophylla. Branches as- suming the form and function of foli- age; 65,66. Clavate (-atus\ Claviformis. Club- shaped. Clavellate. Diminutive of Clavate. Clavicuhte (-alus). Furnished with clnv- iculce ; viz., tendrils, hooks, or other appliances for climbing. Class, 325. Classification, 315. Clathrate. (-atus). Latticed. Claw. The narrowed base or stalk which some petals, &c., possess; 245. Cleist.dgamy, Cltistoyamous, Cleistor/amic. Close-fertilization in unopened blos- soms; 241. Cleistofifny, Cleistogenous, 241. Same as Cleistogamy. GLOSSAKY. 403 Cleft Cut taff-way down or there- about; y8. Climbing. Kising by laying hold of surrounding objects for support; 51. Clinandrium. The anther-bed in Orchi- daceae. Clinanthium. A name for the receptacle of inflorescence in Compositae ; 148. Clinium. Used in Greek compounds for receptacle, e.g. Periclinium, for an involucre around the receptacle of inflorescence. Close-fertilization. Fecundation by own pollen ; 216, 280. Cloves. A gardener's name for young bulbs developed around a mother bulb. Club-shaped. Gradually thickened up- ward from a slender base. Clustered. Collected in a bunch of any sort. Cluster is a good indefinite name for any assemblage of flowers on a plant. Clypeate (-atus), Clypeiformis. Buck- ler-shaped. Cuuctrrate (-atus). Heaped together. Coddnate (-atus), Coadunatus. Same as Adnate. Coalescence. Union of similar parts ; 179, 180. Coakscent (-ens), Coalitus. Cohering; properly applied to the organic cohe- sion of similar parts. Coarctate (-atus). Crowded together. Coated. Composed of layers as an onion, or furnished with a covering or rind, Cobwebby. Bearing long and soft entan- gled hairs. Coccineus. Bright red or scarlet (red with a little yellow). Coccus. Greek for a kernel or nutlet, from which the Latin Coccum, the kermes or scarlet grain (supposed berry) of the Quercus coccifera; used botanically, mostly in the form of "coccus," for the portions into which a schizocarp, or lobed fruit with one- seeded cells, splits up; these portions are Cocci or Coccules ; 296. Cochlear (Cochlearis). Spoon-shaped. Unmeaningly applied also to a form of imbricative aestivation with one piece exterior; 137. Cochleate (-atus). Shell-shaped, i. e. spiral in the manner of a snail-shell. C&lospermous (-us). Hollow-seeded ; ap- plied to seeil-like carpels of Umbelli- ferce with ventral face incurved at top and bottom, as in Coriander. Ccenanthium. Synonym of Clinanthium. Ctenobio. Synonym of Carcerulus. Cohesion. The congenital union of one organ with another ; either similar parts (coalescence), or dissimilar parto (adnation). Cohort. In classification a group next superior to order, 326. Coleorhiza. Root-sheath; the invest- ment (belonging to the cotyledon or plumule) through which the primary root in many Monocotyledons bursts in germination; 26. Collar ( Collum). Name of an imaginary something intermediate between pri- mary stem and root. Collateral. Standing side by side. Collective Fruits. The aggregation of the fruits of several flowers into one mass; 301. Colored. Of other color than the green of herbage; 118. Columella. The persistent axis of cer- tain capsules, spore-cases, &c. ; 289. Column (Columna). Body formed l»\ the union of the filaments among thern- selves (as in a malvaceous flower), or with the style or stigma, as in Or- chids; 250. Columnar. Column-shaped ; pillai shaped. Coma. Literally a head of hair ; a tuft of hairs of any sort ; specially a tuft of hairs on a seed; 306 Also the name of the whole head of a tree. Commissure (-ura). The face by which two carpels cohere, as in Umbellifenp. Common ( Communis). General or prin- cipal, as opposed to partial. Comose (-osus), sometimes Comatus. Furnished with a coma. Complanate (-atus). Flattened. Complete (Completus). Having all the parts belong to it or to the type; 175. C6mplicate (-atus). Folded upon itself. Compound. Said of similar parts aggre- gated into a common whole. Com- pound Flower, 147. Compound Pistil, 263. Compound Inflorescence, 159. ' Compound Leaf. One divided into sep- arate blades ; 100. Compressed (-us). Flattened lengthwise. Concaulescence. A name for the coales- cence of axes : 158. Conceptacle (-aculum). Originally used by Linnaeus for what is now called Follicle ; and later for the pair of fol- licles of Asclepiadaceae and Apocy- 404 GLOSSARY. Conckiformis. Shaped like one valve of a bivalve shell. Concinnus. Neat or elegant. Concolor. Of the same or of uniform color. Conduplicate (-atus), Conduplicativus. Folded together lengthwise; 133. Cone. See Strobile. Confertus. Closely packed or crowded. Conferruminate (-atus). Stuck together by adjacent faces, as the cotyledons of Horsechestiiut ; 314. Confluent (-ens). Blended into one; pass- ing by degrees the one into the other. Conformed (-ormis). Similar to in form ; or closely fitted to, as a seed-coat to the nucleus. , Congested (-us). Crowded together. Conglobate (-atus). Collected into a ball. Conglomerate (-atus). Densely clus- tered or heaped together. Coniferous (-us). Cone-bearing. Conjugate (-atus). Coupled; in single pairs. Conjugate-pinnnte, 104. Connate (-atus). United congenitallv; 107, 182. Connate-perfoliate. United at base in pairs around the supporting axis ; 108. Connective (-ivum). A portion of a sta- men which connects the two cells or lobes of an anther; 251. Connivent (-ens). Coming into contact or converging. Conocarpium. An unused name for an aggregate fruit, such as a strawberry, consisting of many carpels on a coni- cal receptacle; 298 Consolidated. When unlike parts are coherent. Continuous. The reverse of articulated or interrupted. Contorted (-us). Twisted; or bent or twisted on itself. In ^Estivation, the same as Convolute ; 138. Confer tuplicate (-atus). Twisted and plaited or folded. Contracted. Either narrowed or short- ened. Contrary (-a.rius). Opposite in direction to the part compared with ; as a silicle compressed contrary to the dissepi- ment. Convolute (-utus) or Convolutive (-ivus). Rolled up from the sides or longitudi- nally. In ^Estivation, 138. In Ver- nation, 133. CSralloid (-eus). Coral-like. C6rculum. Old name for the embryo, or Cor seminis ; 311. Cordate (-atus), sometimes ConKform (-ormis). Heart-shaped ; like the fig- ure of a heart on cards ; the stalk at the broader and notched end ; 96. Coriaceous (-eus). Leathery in consist- ence. Cork, 81. Corky. Of the texture of cork. Corky Envelope, 76. Corm ( Cormus). A bulb-like fleshy stem, or base of a stem ; a " solid bulb ; " 61. GormopJiytes ( Cormopl/yta), 341. Corneous (-eus). Of the texture of horn. Corniculate (-atus). Furnished with a little horn. Cornu. A horn ; i. e. a horn-like process ; sometimes used for Calcar, a spur. Cornute (-utus) . Furnished with a horn- like process or spur. Corolla. The interior perianth, com- posed of petals; 165, 243. Corollaceous (-eus), Corollinus. Pertain- ing to, or resembling corolla. Corolliferous (-us). Bearing a corolla. Coroliijlorous, Corollijtorce, 340. Corollula. Diminutive of corolla. Corona. A crown : an inner appendage to a petal, or to the throat of a corolla ; 210, 246. Or any coronet-like append- age at the summit of (crowning) an organ. C6ronate (-atus). Crowned, having a corona, &c. Coroniform (-ormis). Shaped like a crown or coronet. Corrugate (-atus or -ativus). Wrinkled or in folds ; 133. Cortex. Rind or bark. Cortical (-alis). Relating to bark. Corticate (-atus). Coated with a bark or with an accessory bark-like cover- ing. Cdrymb ( Corymbus). A flat-topped or merely convex and open flower-cluster of the indeterminate or centripetal order; 146. Corymbiferous (-us). Bearing corymbs. Corymbose. In corymbs, or in the man- ner of a corymb. The corymb of Linnaeus and of other writers down to Roeper included most cymes. So that much cymose inflorescence is in descriptions loosely said to be corymbose, or a stem is said to be corymbosely branched, even when the evolution is centrifugal ; 140. Costa. A rib; when single, a midrib or mid-nerve. GLOSSARY. 405 Costal-nerved. With nerves springing from a midrib ; 92. Costate (-atus). Ribbed; furnished with one or more longitudinal primary veins or ribs. Cotyledons ( Cotyledon, pi. Cotyledones). The " seed-lobes," being the leaves or first leaves of the embryo; viz., the one, or the pair, or rarely the whorl of leaves borne by the radicle or caulicle ; 10, 311, 313. Cotyliformis. Dish-shaped, or wheel- shaped with an erect or ascending border. Cratiiriform (-ormis). In the shape of a goblet or cup, of hemispherical con- tour or more shallow ; 248. Cremocarp ( Cremocdrpium). A dry and seed-like fruit, composed of two one- seeded carpels, invested by an epigy- nous calyx, and separating at matu- rity; 297. Creeping. Running along or under ground and rooting; 53. Crena, Crenatura. A rounded tooth or notch. Crenate (-atus). Toothed by crenatures ; scalloped; 98. Crenel, Crenelled. Same as Crenature and Crenate. Crenulate (-atus). Diminutive of Cre- nate, t. e. with small crenatures. Crested. Furnished with any elevated line, ridge, or conspicuous elevation on the surface, especially suc.h as may be likened to the crest of a helmet. Cretaceus. Chalk-white; chalky. Cribrose (-osus) and Cribriform (-ormis), Pierced like a sieve. Cribriform Cells', 77. Crinttus. Bearded with long and weak hairs. Crispatus. Curled or crispy. Cristate (-atus). Crested. Croceus, Crocatus. Saffron-colored, t. e. deep reddish-yellow. Cross-breeds. The progeny of interbred varieties; 321. Cross-fertilization. Fecundation by pol- len of another flower and of another individual; 216. Crown. See Corona, 210; 246. Crowned. See Coronate. Crowning ( Coronans). Borne on the summit of an organ. Cruciate (-atus), Cruciform (-ormis). Cross-shaped. Cruciferous (-MS). Cross-bearing; used in the sense of Cruciform ; as the "cruciferous" corolla of the order Cruciferae; 246. Crumpled. See Corrugate. Crustaceans (-us). Of hard and brittle texture. Cryptos. Greek for concealed ; whence Cryptogamia. Cryptogamous or Cryp- togamic plants ; 3, 335, 344. Cryptogamous. Pertaining to the above. Cucullate (-atus), Cuculld>-is, Cuculli- formis. Hooded, or hood-shaped, cowled. Culm ( Culmus). The peculiar stem or straw of Grain-plants and Grasses ; 50. Cultrate (-atus), Cultriformis. Shaped like a broad knife-blade. Cuneate (Cuneatus), Cuneiform (-ormis). Wedge-shaped; triangular with an acute angle downward ; 95. Cup-shaped. In the form of a drinkiug- cup. Cupule (Cupula). The acorn-cup and the like; 296. Cupularis, Cupulatus. Furnished with or subtended by a cupule or any re- sembling body. Cupuliferous (-us). Cupule-bearing. Curvinerved (-ius). When the ribs of a leaf are curved in their course ; 92. Cuniserial. In curved or oblique ranks ; 124. Cushion. The enlargement at or be- neath the insertion of many leaves. Cuspidate (-atus). Tipped with a Cusp, or sharp and rigid point; 97. Cut. Same as incised, or in a general sense as cleft. Cuticle ( Cuticula). The outermost skin or pellicle. Cutting. A severed portion of a plant used for bud-propagation ; 43. Cydneus. A clear bright blue. Cydthiform (-ormis). Cup-shaped ; in the form of a Cydthus. A drinking-cup, such as a goblet or wine-glass. Cycle. A circle. Sometimes used for one turn of a helix or spire; 122. Cyclical. Relating to a cycle ; or coiled into a circle ; 119, 120. Cylindraceous. Somewhat or nearly cylindrical. Cylindrical (-us). Elongated and with circular cross-section ; in the form of a cylinder. Cymbceform or Cymbiform (-ormis). Boat-shaped. Cyme ( Cyma). A flower-cluster of the determinate or centrifugal type, 406 GLOSSAEY. especially a broad and flattish one ; 151. Cymo-botryose. When cymes are ar- ranged in botryose manner; 159. Cymose (-osus). Bearing cymes, or re- lating to a cyme ; 151. Cymule ( Cymula). Diminutive cyme, or a portion of a cyme ; 151. Cynarrhodium. Name of such a fruit as that of the Rose; fleshy, hollow, and enclosing achenia. Cypsela. Name of an ache ium in- vested by an adnate calyx, as the fruit of Composite ; 295. Cystdith. One of the mineral and usually partly crystalline concretions of the cells of the epidermis of or subjacent tissue of the leaf in various plants, especially in Urticaceae. Cytoblast. An obsolete name for the nucleus of a cell of cellular tissue. Ddctylose (-osus). Fingered, or finger- shaped. Dasypliyllous (-us). Woolly-leaved. DeaUi.ite (-atus). Whitened over (as if whitewashed) with a white powder or minute pubescence. Deca. Greek for ten, compounded with various words, such as DtC'igynia. One of the Linnaean artificial orders; 337. Decdyynous (-us). With ten styles or carpels. Dtcdmerous (-us). Of ten members ; 176. Dccandria. A Linnsean class with ten stamens; 334. Decandrous (-MS). With ten stamens; 249. Decapetalous (-us), Decasepalous, &c. With ten petals or sepals, &c. Deciduous (-us). Falling, or subject to fall in season, as petals after anthesis, and leaves (except of evergreens) in autumn; 243. Decimate (-atus), or Declined. Bent or curved downward or forward. Decompound. Several times compound- ed or divided; 102, 104. Decompositus. Decompound. Decumbent (-ens). Reclining, but with summit ascending; 53. Decurrent (-ens), Decursive. Running down into ; as where leaves are seem- ingly prolonged below their insertion, and so run down the stem. Decussate (-atus). In pairs alternately crossing at right angles. Deduplication, Fr. Dedoublement. Same as Chorisis; 202. Definite (-itus). Of a fixed number, not exceeding twenty ; or of a fixed or..er. Definite Inflorescence. Where axes of inflorescence end in a flower; 144, 151. Deflexed (-us). Bent or turned abruptly downward. Deflorate (-atus). Past the flowering state. DejUl ite (-atus). Having cast its leaves. Defoli 'lion, 87. Dehiscence (-entia). The mode of open- ing of a capsule or anther by valves, slits, or regular lines ; 288. Dehiscent (-ens). Opening by regular dehiscence; 292. Deliquescent (-ens). Dissolving or melt- ing away, as a stem divided into brandies; 48. Deltoid (-oides). Having the shape of the Greek letter A. Demersed (-us). Under water ; same as submersed. Dendritic (-icus), Dendroid (-oideus). Tree-like. Dendron. Greek for tree. Deni. Ten together. Dens. A tooth. Dentate (-atus). Toothed; specially with salient teeth not turned forward ; 98. Denticulate (-atus). Minutely toothed; having denticulations, or diminutive teeth. Deniulate (-atus). Made naked ; stripped. Deorsum. Downwards. Depauperate (-atus). Impoverished ; as if starved; or diminutive for want of favorable surroundings. Depressed (-us). Having the appear- ance or shape as if flattened from above. Derma. Greek for skin or surface of a plant or organ. Descending (-ens). Tending or turning gradually downward. Descending Axis. Primary root; 11. Determinate. Limited in number or ex- tent; as are the axes of determinate inflorescence; 144, 151. Desinens. Terminating in. Desmos. Greek for things bound, or as if chained together. Dextrorse (Dextrorsus: adv. Dextror- surn). Toward the right hand, or re- lating to it; 51, 140. Di, Dis. In Greek compounds, two, or double Diachenium. Synonym of Crsmocarp. GLOSSARY. 407 Diadelphia. A Linnaean (335) class having the stamens. Diaddphovt (-us). Combined by their filaments into two sets ; 250. Diagnosis. A brief distinguishing char- acter. Dialypetake, 341. Dialypetalous (-us). Same as polypeta- lous, i. e. of separate petals ; 244. Dialyphyllous (-IM). Bearing separate leaves. Diandria. A Linnaean class with per- fect flowers having only two stamens; 334. Dlundrous (Diander, &c.). Having two stamens; 249. Diaphanous (-us). Letting the light shine through. DicarptlUiry. Composed of two carpels or pistil-leaves ; 261 Dichdsium. A two-parted or two-rayed cyme; 152, 155. Dichlamydeous (-eus). Having a double perianth; 191. Dichotomous (-us). Forked in pairs; two-forked. Dichoyamous (-us), Dichogamy. Her- maphrodite with one sex earlier de- veloped than the other in the blossom ; 219. Diclesium. Name of a fruit consisting of an achenium within a separate and free covering made of perianth, as that of Min.bilis. Diclinous (Diclinis). When flowers are of separate sexes ; 191. Dicocctrus (-us). Fruits of two cocci. Dicotyledons, Dicotyledones. Plants of the class marked by having two coty- ledons ; 27, 339, 340, 344. Dicotyledonous (-eus). Having a pair of cotyledons; 10, 314. Diilymous (-us). Twin, found in pairs. Didyndmia. The Linnaean class marked bv didynamy (335), i. e. Dithinamous (-us). When a 4-androus flower has the stamens in two pairs, and one pair shorter than the other; 250. Dieresilis. Mirbel's name for a dry fruit composed of several cells or car- pels connate around a central axis, and separating at maturity, as that of Mallow. Difformis. Of unusual formation. Diffuse (-usus). Widely or loosely spreading. Diyamous (-us). Of two sexes in the same cluster. Digitate (-atus). Fingered ; a compound leaf in which all the leaflets are borne on the apex of the petiole ; 101. Diyitattly. In a digitate mode; same as Palmately. Diffitate-Pinnate, 104. Diyynia. A Linnaean order character- ized by having the gynoecium Diyynous. With two separate styles or carpels; 261. Dimerous (-us) Of two members in each circle; 176. Dimidiate (-atus). Halved, or as if one- half was wanting. Dimorphous (-us), Dimorphic, Dimor- phism Occurring under two forms; 225, 234 Diaecia. Linnaean class (355) of plants with the flowers Dioecious (Dicecius, Dioicous). Unisex- ual, and the two sexes borne by dis- tinct individuals ; 191. Dicecio-poly yamous. When some indi- viduals bear unisexual and others bi- sexual flowers. Dipetalous (-us). Two-petaled; 244. Diphyllous (-us). Two-leaved ; 243. Diplo. See Duplo. Diplostemonous, Diplostemony. Having twice as many stamens as petals or sepals ; 177, 198. Diplote yium. A capsule or other dry fruit, invested with adnate calyx ; an inferior capsule. Dipterous -(us). Two-winged. Diremption (-io). Syn. of Chorisis; 202. Disciferous (-us). Disk-bearing. Disciform (-ormis). Depressed and cir- cular, like a disk or quoit. Discoidal or Discoid (Discoid eus). Ap- pertaining to a disk. A discoid head is one destitute of ray-flowers. Disc or Disk (Discus). A word used in several senses. The disk or disc of a flower is a development of the torus within the calyx, or within the corolla and stamens ; 213. In a capitulum or head of flowers it is the central part of the cluster, or the whole of it -as opposed to a border or ray. It is the face or surface of any organ, such as a leaf-blade, as opposed to the mar- gin. In vegetable anatomy, certain round spots or markings on cell-walls are termed discs. Discolor. When the two faces of a leaf, &c., are unlike in color. Discrete (-etus). Separate ; not coales- cent. 408 GLOSSARY. nitepilous (-us) Two-sepaled; 244. Disk-flowers. Those belonging to the disk, or body, and not to the margin or ray of a capitulum. Dissected (-MS). Deeply cut or divided into numerous segments. Dissepiment (-entum). A partition in an ovary or pericarp ; 264. Dissident (-ens). Bursting asunder or in pieces. Distichous (-us). Disposed in two ver- tical ranks ; 122. Distinct (Distinctus). Separate from; not united. Distractile (-ilis). Carried widely apart. Dilhecous (-us). Of two thecae, or cells, as are most anthers ; 254. Diurnal. Daily ; occurring in the day; sometimes used for ephemeral. Divaricate (-atus). Extremely divergent. Divergent, Diverging (-ens). Inclining away from each other. Divided (Dicisus). Where lobing or segmentation extends to the base ; 98. Dodeca. Greek for twelve. Used in Dodecayynia. Linnaean order with flow- ers. Dodecdffynous. Having twelve styles or distinct carpels. Dodecdmerous (-«*). Of twelve parts in the circle. Dodecandria. A Linnaean class (334) with the flowers. Dodecdndrous. Having twelve (or from 12 to 19) stamens; 249. Dodrantdlis. A span (about nine inches) long. Dolabriform (-ormis). Axe-shaped or hatchet>shaped. Dorsal (-alls). Relating to the dorsum or back. Dorsal Suture. That which answers to the midrib of a carpel ; 261. Dorsiferous. Borne on the back. Double. Has a technical use when a flower is said to be "double;" this denoting one in which the leaves of the flower are monstrously increased mostly at the expense of the essential organs. Downy. Pubescent with fine nnd soft hairs. Loosely synonymous with soft- pubescent, tomentose, &c. Drepdnium. A sickle-shaped cyme; 156. Drupaceous (-eus). Resembling or relat- ing to a drupe. Drupe (Drupa). A stone fruit; 297. Drupelet, Drupel (Drupeola). A dimin- utive drupe; 297. Drupetum. An aggregation of drupes, 300. Duct. In vegetable anatomy, an elon- gated cell or tubular vessel, lound espe- cially iu the woody (fibre-vascular; parts of plants. Dumetuse (-osus). Pertaining to Dume- tum, a thicket, or Dumus, a bush. Dumose (-vsus). Bushy, or relating to bushes. Duplo. Twice as many. In Greek com- pounds, Diplo. Duramen. The heart-wood of an exo- genous stem ; 80. Dwarf. Of small size or height com- pared with its relatives. Dyclesium. See Diclesium. E- or Ex- As a prefix to Latin words, carries a privative meaning, as Ecos- tate, without ribs, Exalbuminous, with- out albumen. Eared. Same as Auriculate. Ebracteate, Ebracteolate (-atus). Desti- tute of bracts or bractlets. Eburneus. Ivory-white. Ecdlcarate (-pous (-us, Heterocarpictts). Producing more than one kind of fruit. GLOSSARY. 415 Heterocephalous (-us). Bearing two kinds of head or capitulum. Heteroclite (-itus). Anomalous in forma- tion. Heterocline (-inus). Nearly same as Heterocephalous ; on separate recepta- cles. Htterodromous (-us). Spirals of changing direction. Heterdyamous (-us). Bearing two kinds of flowers. Heterogeneous. Not uniform in kind. Heteroyone or Heteroyoneus. When the flowers are dimorphous or trimor- phous as respects relative length, &c., of stamens and pistil ; 225, 234. Heterostyled. Same as Heterogone ; 234. lleteromerous. Of members not corre- sponding in number. fltttr» divided. Palminerved. Palmately nerved ; 93. Pdludose (-osus), Palustrine (Palustr** or Paluster). Inhabiting marshes. Pdndurate (-atus),Panduriform (-oi-mi*^ See Fiddle-shaped. Panicle (Panicula). A loose compound flower-cluster, such as is produced by the branching of a raceme, or the ir- regular branching of a corymb ; 150. Panicled, Paniculate (-atus). In a pani- cled manner or borne in a panicle. Pannosus, Panniformis. Having the ap- pearance or texture of felt or woollen cloth. Papery, Papyraceous. Having the text- ure of paper. Papilionaceous (-eus). Butterfly-like; applied to a peculiar polypetalous corolla ; 184, 246. Papillar (-aris), Papillose (-osus), Papil- late (-atus). Bearing or resembling papillce, minute nipple-shaped projec- tions. Pappiferous (-«5), Pappose (-osus). Bearing a pappus. Pappus. Thistle-down; thence applied to various hairy tufts on akenes or fruits ; and thence to any production or structure which takes the place of the limb of the calyx on the akenes of Composite; 192, 295 Papuliftrous (-us), Papulose (-osuf\ Covered with Papula, or small pim- ples 424 GLOSSARY. Paracarpium. Unused name given to an abortive pistil or carpel. Puracorulla. A crown or internal ap- pendage or deduplication of a corolla. Parallel-nerved, veined, &c. Same as Nerved; 91. Parapetalous (-us). Said of stamens, &c., which stand at each side of a petal; 178, 201. Pardphysis, pi. Parapkyses. Jointed thread-like bodies, of no known func- tion, accompanying the archegonia of Mosses. '< Parasitic (-icus). Growing on or in and living upon another plant or even ani- mal; 36. Parastemon. Name rarely applied to an abortive stamen or body in place of or accessory to a stamen; same as Sta- minodium. Parastichies. Secondary spirals in phyl- lotaxy; 127. Parenchyma. Common or soft cellular tissue. Parenchymatous. Of the nature of or composed of parenchyma. Paries, pi. parietes. The wall of any organ. Parietal (Parietalis). Borne on or re- lated to the wall; 265. Paripinnate (-atus). Even-pinnate ; same as abruptly pinnate; 101. Parted, Partite (-itus). Cleft nearly but not quite to base; 98. Parthenogenesis, Parthenoyeny. Pro- duction of seed without the interven- tion of pollen; 285. Partial (Partialis). Secondary, as Par- tial involucre (142), peduncle (143), petiole (105), umbel (150). &c. Partible (Partibilis). At length sepa- rating or easily to be separated. Partition. In one sense a separated por- tion or segment ; in another and the more usual, a wall or dissepiment. Patelliform (-ormis). Disk-shaped, cir- cular with a rim, of the form of the patella or kneepan. Patent (Patens). Spreading ; either widely open or diverging widely from an axis. Patentissimus. Superlative of Patens; extreme!}' spreading. Pdtulous (-us) Slightly or moderately spreading. Pauciflorous (-us). Few-flowered. Paucifolius. Few-leaved. Pear-shaped. Obovoid or obconical with more tapering base. Pectinate (-atus). Pinnatifid with nar- row and closely set segments, like comb-teeth. Peddlis. A loot lung or high. Pedate (-atus). Palmately divided or parted with the lateral divisions two- cleft ; resembling a bird's foot. Pedatipartitus, -looutus, -sectus, &c. Pe- dately parted, lobed, divided, &c. Pedicel (-tllus). An ultimate flower- stalk or its division ; the support of a single flower ; 143. Pedicellate (-atus). Pedicelled, borne on a pedicel. Pediculus. Name sometimes used for Pedicel. Peduncle (Pedunculus). A general flower-stalk, supporting either a clus- ter or a solitary flower; in the latter case,the cluster may be regarded as reduced to a single blossom ; 143. Pedunculate (-atus). Peduncled, borne on a foot-stalk. Ptloi'ia. An irregular flower become regular by a monstrous development of complementary irregularities ; 186. Peltate (-atus), Ptltifonn (-ormis). Shield-form; target-shaped; a plane body attached by its lower surface (instead of margin or base) to a stalk ; 96, 107. Peltinerved (-ius). Radiately-nerved or ribbed all round the circle. Peliiform (-ormis). Basin-shaped ; shal- low cup-shaped. Pendent (-ens). Hanging on its stalk or support. Pendulus (-us), Pendulinus. Hanging more or less, as if from weakness of the support. Penicittate (-atus), Penicittiform (-ormis). Pencil-shaped, the pencil (penicillum) being a brush or tuft of hairs. Pennate (-atus). Same as Pinnate. Penniform (-ormis). In the form of a feather or its plume. Penninerved (-ervius). Same as pin- nately nerved or veined ; 93. Penta. Greek for five ; gives compounds such as Penta cdrpellary. Composed of five car- pels; 261. Pentachcenium. Name of a pentacarpel- lary fruit otherwise like a cremocarp. Pentadelphous (-us). With stamens in five clusters; 250. Pentnrjynin. Linnaean artificial order characterized by Pentagynous, i. e~ five-styled flowers; 337. GLOSSAKY. 425 Pentdmermis (-us). Composed of five members in a circle; 176. Pentandi~ia. The Linnaean class with Pentandrous, i. e. five-stamened flow- ers; 249, 334. Pentapftalous (-us). Five-petalled ; 244. Ptntaphyllous (-us). Five-leaved: 243. Pentdpterous (-us)- Five-winged. Pentasepalous (-us). Of five sepals ; 244. Ptntdstichous (-us). In five vertical ranks ; 123. Pepo, Peponida, Peponium. A gourd- fruit; 298. Perennial (Perennis, Perennans). Last- ing year after year ; 32. Perfect (Perfectus). Said of a flower which is hermaphrodite. Perfoliate (-atus). Where a stem seem- ingly passes through a leaf; 167. Perforate (-atus). Pierced, or having translucent dots which look like holes. Pergameneus, Pergamentaceus. Parch- ment-like in texture. Peri. Greek for around; hence such compounds as Perianth (Perianthium). The floral en- velopes or leaves of the flower, consist- ing of calyx, corolla, or both ; 164, 243. Pericarp (-arpium). The fructified ovary; 286. Pericdrpic (-icus). Relating to the peri- carp. Perichcetial (-ialis). Relating to the Perichcetium, a set of bracts around the fruit-stalk in Mosses. Peridddium. The sheathing base of a leaf when it expands and surrounds the supporting branch. Periclinium. Involucre of the capitu- lumof Compositae; 148. Periderm (-erma or -ermis) Outer bark or Epiphloeum. Perigone, Perigonium. Synonym of Perianth; 164. Perigynium. Name of hypogynous bristles, scales, or a sac, which sur- rounds the pistil (also the stamens when present) of many Cyperaceae. Perir/ynous (-us). Literally around the ovary; said of organs which are ad- nate to the perianth, or to this as con- nate with the low T part of the pistil ; 182. Peripetalous (-us). Around the petals. Peripherie (-icus). Of or belonging to the circumference ; as of an embryo coiled round the outside of the albumen. Periphoranthium. Synonym of the involucre of Composite : 148. Peripterus. Surrounded by a wing or thin border. Perisperm (-ermium). The albumen of the seed, at least the exterior and or- dinary albumen; 14, 310. Peristome (Peristoma or Peristomium). The fringe or other structure surround- ing the orifice (stoma) of a Moss. Peritropous (-us) or Peritropal. Said of a seed which is horizontal in the pericarp; or of a radicle pointed to- ward the sides of the pericarp. Persistent (-ens). Remaining even on the fruit, or over winter ; 243. Personate (-atus). Masked, as when a bilabiate corolla has a prominent pal- ate; 248. Pervious (-ius). With an open passage- way. Perfuse (-usus). Having slits or holes. Perula, pi. Peruke. Scales of leaf-buds and the like; 40. Perulate (-atus). Furnished with peru- Ice or scales. Pes, gen. pedis. A foot. Hence in Latin compounds Longipes, long- stalked, Brevipes, short-stalked, &c. Petal (Petalum). A corolla-leaf; 165. Petaline (-inus), Petaloid (-oideus). Petal-like, or relating to petals ; 118. Petalody. Name for the metamorphosis of other organs (such as stamens) into petals; 174. Petiolar (-arts). Borne on or relating to a petiole. Petiolate (-atus), Petioled. Having a petiole. Petiole (Petiolus). The footstalk of a leaf; 85, 104. Petiolulate (-atus), Petiolular (-aris). Having a Petiolule (Petiolulus). A footstalk of a leaflet; 105. Petraeus. Growing among rocks. Petrosus. Growing in stony places. Phcenogams, Phanogamia, Phasnoya- mous plants. Plants sexually propa- gating by flowers, of which the essen- tial organs are stamens and pistil ; 3, 334, 340, 344. Phalanges, sing. Phalanx. The bundles of stamens in diadelphous or polyadel- phous flowers. Phanerogams, Phanerogamia, &c. See Phaenogams, &c. Phlceum. Greek name for bark. Phceniceus. Deep red with some scar- let. Phoranthium. A name for the recep- 426 GLOSSARY. tacle of the capitulum in Composite ; 148. Phycology. The botany of Algae. Phylla. Leaves in Greek; combined with Greek numerals, forming such terms as Diphyllous, Triphyllous, &c., to Polyphyllous. Phyllocladium. A branch assuming the function of foliage ; 65. Phyllodineous (-eus). Relating to a Phyllodium. A petiole usurping the form and function of a leaf-blade; 110. Phyllody, Phyllomorphy. Names for the transformation or metamorphosis of floral organs into leaves ; 174. Phyllotdxis, Phyllotaxy. Leaf-arrange- ment; 119. Phyllomania. The unusual or abnormal production of leaves. Phyllophore (-orum). The budding sum- mit of a stem on which leaves are de- veloping. Phyllam. Greek for leaf; 6, 85. See Phylla. Phyliome, Phylloma. An assemblage of leaves, or of incipient leaves in a bud. Also recently used by German botanists for leaf genetically or poten- tially, that which answers to a leaf; 6. Phytography. Botany as relates to the de- scription and illustration of plants ; 345. Phy'dlogy. Synonym of Botany. Phytomer, pi. Phytomera. Plant-ele- ments in morphology ; same as Phyton. Greek name for plant; has been used in the sense of plant-ele- ment, or plant-unit ; 7. Phytotamy. Same as Vegetable Anatomy or Histology; 2. Ficeus. Pitch-black or brownish-black. Pictus. Painted, or rather as if painted. Pileate (-atus), Pikiformis. Having the form of a cap or Pileus. Plltui'hiza. The root-cap. Piliferous (-us). Bearing or tipped with hairs (pill). Pilusciusculus. Slightly hairy. Pilose (-vsus). Hairy, in general with any sort of pilosity ; in particular with soft and distinct hairs. Pinna. One of the primary divisions of j a pinnate leaf, either simply pinnate, 1 when it is a leaflet, or a partial petiole j or rhachis with the leaflets when the ' leaf is bipinnate ; 104. Pinnate (-atus). When leaflets are ' arranged along each side of a com- | inon petiole; 100. Pinnately cleft, lobed, parted, &c. ; 99. Pinnately veined. Feather-veined; 93. Pinndtifid (-idus). Phmately cleft. Pinnatilobatus, Pinnatilobus. Pinnately lobed. Plnnatipartitus. Pinnately parted. Pinnatisectus. Pinnately divided quite down to the rhachis. Pinnule (Pinnula ). One of the pinnately disposed divisions of a pinna; a sec- ondary pinna; 104. Pisiform (-ormis). Pea-shaped; resem- bling a pea. Pistil (Pistillum). The female organ of a flower, consisting of ovary, style, and stigma, or at least of ovary and stigma ; 302, 259. Pistillate (-atus), Pistilliferous. Said of a plant or a blossom provided with pistil, most properly for one having pistil only; 191. Pistillidium. One of the names of the analogue of pistil in Mosses, &c. Pistillody. Name for the metamorphosis of other organs into carpels ; 174. Pitcher. See Ascidium. A tubular or cup-shaped leaf, which usually holds some liquid; 111. Pith. A central cellular part of a stem, especially of an exogenous stem ; 75. Pitted. Marked with small depressions or pits. Placenta. That in the ovary which bears the ovules, sometimes the mere united margins of the carpel-leaves, sometimes a thickening or enlarge- ment of them, or even of some other part of the ovary ; 261. Placentation (-io). The disposition of the placentae. Placentiform (-ormis). Quoit-shaped, or in form like a flat cake. Plaited. See Plicate. Plane (Planus). With flat surface or surfaces. Platys. Greek for wide, in such com- pounds as Platyphyllus, broad-leaved, &c. Pleios. Greek for full, used in com- pounds for several or many ; as Pleio- phyllous, several-leaved, &c. Simi- larly Pleistos for a great manv. Pleiochdsium. A several-rayed cyme; 152, 155. Plenus. Full. Flos pltnus is what gar- deners call a "double flower," that is one in which the petals or other flower- leaves are abnormally multiplied. Pleurcnchyma. Same as woody tissue. GLOSSARY. 427 Pleurorhizal (-us). Embryo with radicle against one edge of the cotyledons; i. e. the latter accumbent. Plicate (-atus), Pticaticus. Folded into plaits (plicce), usually lengthwise ; 133, 139. I'lumbeus. Lead-colored; dull gray with some metallic lustre. Plumose (-osus). Feathered; when bris- tles, &c., have fine hairs on each side like the plume of a feather, as the pap- pus of Thistles. Plumule (Plumula). The bud or grow- ing point of the embryo above the cotyledons; 17. Plures. Many or several; used as a prefix in Latin words, such as Pluri- florous (-us), several-flowered; Pluri- locular (-am), several-celled ; Pluri- fuliolate, with several leaflets, &c. , Pluri/ugate, in several pairs, &c. Poculiform (-ormis). In the shape of a drinking-cup or goblet. Pad. A dry*and several-seeded dehis- cent fruit; strictly a Legume or a Silique ; 288, 292. Podium, Podus. A footstalk, stipe, or other such support ; used only in Greek compounds, as Podocephalus, head pedunculate; Podocarpus, fruit stipi- tate ; or as a suffix, in such words as Leptopodus, slender-stalked; Brachy- podus, short-stalked, &c. Podetium. Any stalk-like elevation. Podogynium. Same as Gynophore. Podosperm (-ermium). The stalk of a seed; 276, 305. Pogon. Greek for a beard ; enters into various compound w.ords. Polembryony. See Polyembryony. Politus. Polished ; applied to a smooth and shining surface. Pointless. Same as Muticous. Puintletted. Minutely pointed; same as apiculate or as minutely acuminate. Pollen, Pollen-grains. The fecundating grains or cells contained in the anther; 165, 256. Pollen-tube. The slender tube which begins as a protrusion of the inner coat of a pollen-grain, and elongates by growth, at least when in contact with the stigma ; 258. Pollicdris. An inch long; the length of the terminal joint of the thumb, pollex. Poliniferous (-us). Pollen-bearing. Pdllinium. A mass of pollen-grains more or less coherent ; 257, 230. Pollinated (-atvs). Said of a stigma when supplied with pollen. Poly. In Greek compounds, denotes numerous ; as in Polyadelphia. Name of a Linnaean ar- tificial order with stamens Polyadel- phous, or in several phalanges or brotherhoods; 250, 335. Polyandria. Name of a Linnaean class with flowers Pulyandrous, or having an indefinite number of stamens; 249. 334. Polydnthous (-us). Many-flowered; in the Latin form same as multiflorous. Polycdrptllary. Of many carpels ; 261. Polycdrpic (-icus). Fruiting many times or indefinitely ; DeCaudolle's name for a perennial herb ; 33. Polycephalous (-alus). Consisting of or bearing many heads, capitula. Polycoccus. Of several cocci. Polycotyledonous (-eus or es). Having several cotyledons ; 22, 314. Polyydmia. Name of a Linnsean class having Polygamous flowers, i. e. some hermaphrodite, some unisexual ; 191, 335. Also of Linnaean orders of Syn- genesia; 337. Polygynia. Name of a Linnaean artifi- cial order with flowers Polygynous, i. e. containing numerous carpels; 261, 337. Polymerous (-us). Of numerous mem- bers to each series or circle. Polymorphous (-us). Of several or vari- ous forms. Polypetnlous (-us). Having separate petals; 244. Polyphore (-6rium). A torus which bears many pistils, as that of a straw- berry or raspberry. Polyphyllous (-us). 'Many-leaved ; 244. Polysepalous (-us). Of separate sepals; 244. Polyspermous (-us). Many-seeded. Polystemonous (-us). With many sta- mens. Polystdchyus. Bearing many spikes. Polystylous (-us). Bearing many styles. Polysymmetrical. That which can be divided into similar halves in several or more than one plane ; 175. Polytocous (-us). Bearing progeny (fruit- ing) many times, i. e. year after year; 33. Pome (Pomum). Kind of fruit of which the apple is the type; 298. Pomeridianus. In the afternoon. Pomiferous (-us). Pome-bearing 428 GLOSSARY. Pomology. A treatise on or the subject of fruits considered as esculent. Porose (-osiis), Porous. Pierced with small holes or pores. Posterior. In an axillary flower is the side next the axis of inflorescence; 160. Posticous (-us). On the posterior side, which in a flower is that next the axis of inflorescence : an adnate anther is posticous when on the outer side of the filament, i. e. when it faces the petals; 253. Pouch. See Silicic. Prcecox. Appearing or developing early. Prceftoration. Same as ^Estivation; 132. Prcefoliation. Same as Vernation ; 132. Prcemorse (-orsus). With end as it were bitten off. Prasinus. Grass green. Pratensis. Growing in meadows. Prickly. Armed with Prickles (56), which are outgrowths of the bark or rind. Primine. Outer coat of the ovule ; 277. Primordial (-ialis). The first in order of appearance. Primordial leaves are those of the plumule. Prismatic (-icus). Prism-shaped, with flat faces separated by angles. Procerus. Very tall. Process (Processus). Any projecting appendage. Procumbent (-ens). Lying along the ground; 53. Productus. Produced, i. e. extended or prolonged into. Pro-embryo, 284. Proles. Progeny; sometimes used for race; 320. Proliferous (Prdlifer, Proliferus). Bear- ing progeny, in the way of offshoots. Proliferation or Prolification is usually taken as the production by one organ of something different, such as the development of buds and plantlets on leaves, of leafy shoots in place of flowers, &c. ; 73. Proligerous (-us). Same as Proliferous. Prone (Pronus). Lying flat, especially face downward. Propdculum, Propdgulum. Name of a shoot, such as a runner or sucker which may serve for propagation. Propagines. Same as Bulblets. Prophylla. Primary leaves, as the first leaves of a branch or axis. Prosenchyma. Plant-tissue consisting of lengthened, tubular, or fusiform cells. Prostrate (-atus). Lying quite flat on the ground ; 53. Protos. Greek for first; used in various compounds, such as Protandrous, Protandry. See Proter- androus. Proterdndrous, also Protandrous, Pro- terandry. When the anthers of a flower are in anthesis earlier than the stigma; 219, 220. Proteranthous (-us). Where flowering precedes leafing. Proteroyynous, Proterogyny, or Pro- togynous, Protoyyny. When the stigma is ready for its functions ear- lier than the anthers of the same blossom; 219. Protophytes, Protophyta. Alga, &c., the supposed first plants. Protoplasm, Protopldsma. The forma- tive organic material of plants and animals, in its living state. Pruinate (-atus), Pruinose (-osus). As if frosted over with a bloom or powder. Pseudos. Greek for false, a prefix in various compounds, as Pseudo-mono- cotyledonous; 26. Pseudo-bulb. A thickened and bulb- like internode in epiphytal orchids; a corm. Pseudocarp (-arpium). The principal or accessory part of an anthocarpous fruit; 300. Pseudo-costate. False-ribbed, as where a marginal or intramarginal vein or rib is formed by the confluence of the true veins. Pseudospermium. Name given to any kind of one-seeded fruit which is inde- hiscent and resembles a seed, such as an akene, &c. Psilos. Greek for naked or bare; as in Psilostachyus, with naked spike. Pteridium, Pterodium. Names for the Key-fruit or Samara. Pteris. Used for wing in Greek com- pounds, also for a Fern. Pteridographia. The botany of Ferns. Pterocarpous (-us). Wing-fruited. Pteropodus. Wing-footed, i. e. petiole wing-margined, &c. Ptyxls. Greek name for folding, as of leaves in a bud ; 132, 133. Pubens, Pubes. Used for Pubescent. Puberulus. Minutely pubescent. Pubes. Pubescence, hairiness. Pubescent (-ens). Clothed or furnished with hairs or down, especially with soft or downy and short hairs. GLOSSARY. 429 Putjioniform (-ormis). Dagger-shaped. Pullus. Dark-colored ; dusky-brown or blackish. Pulvereus, Pulverulentus. Powdered ; as if dusted with powdery matter or minute grains. Pulvinate (-atus), Pulviniform (-ormis). C ushion-shaped. Pulmnus. A cushion ; name given to an enlargement or swelling close under the insertion of a leaf, or sometimes to the swollen base of a petiole. Pumilus. Low or little. Punctate (-atus). Dotted, either with depressions like punctures, or trans- lucent internal glands, or with colored dots. Puncticulate (-atus). Minutely punctate. Pungent (-ens). Terminating in a rigid and sharp point or acumination, like a prickle. Puniceous (-eus). Bright carmine-red. Purpureus. Originally the red of arte- rial blood ; but our purple is some- what dull red with a dash of blue or violet. Purpurdscens. Purplish. Pusillus. Very small, or weak and slen- der. Pustular, Pustulate (-atus), Pustulose (-osus). Having low elevations, like blisters. Putdmen. The shell of a nut; the endo- carp of a stone-fruit; 288. Pycnos, Greek for thick ; whence Pycno- cephalus, thick-headed, &c. Pygmceus. Dwarf, pygmy. Pyramidal (-alis). Pyramid-shaped. Pyrene (Pyrena). Same as Nucule or Nutlet; one of the small stones of a drupaceous fruit ; 298. Pyrenarium, Pyridium. A pear or pear- like fruit, same as Pomum. Pyrenarius. Name of a drupaceous pome, as of Medlar and Crataegus. Pyridion. Synonym of Pome. Pyrenocarp (-arpium). A general name for any drupaceous fruit ; 292. Pyriform (-ormis). See Pear-shaped. Pyxidate (-atus). Furnished with a lid. Pyxidium, Pyxis. A capsule with trans- verse dehiscence, making a lid of the upper portion ; 293. Quadri-. In Latin compounds, denotes four; as Quadrangular, Quadrifari- ous (in four vertical ranks), Quadriju- gate (in four pairs), &c. Quaternary, Quaternate. In fours or composed of four; 176. Quini, Quinary (-ius), Quinate (-atus). In fives ; 176. Quinque. Five. In Latin compounds, giving rise to such terms as Quincuncial, in a Quincunx; also five- ranked; 123, 136. Quinquefarious (-ius). In five vertical ranks. Quinquefoliate (-atus). Five-leaved. Quinquefoliolate, with five leaflets. Quintuple. Dividing into five parts, or five-fold. Quintuplinerved or -veined. With mid- rib of leaf dividing into five (i. e. two lateral pairs) above the base; 93. Race. A variety of such fixity that it is reproduced by seed ; also used in a looser and more extended sense for a series of related individuals without particular regard to rank ; 320. Raceme (Racemus). An indeterminate or centripetal form of inflorescence with lengthened axis and equal-pedi- celled flowers; 146. Racemiferous. Bearing racemes. Racemiform (-ormis). In the form of a raceme. Racemose (-osus). Having the character or appearance of a raceme, or in ra- cemes. Rachis. See Rhachis. Radial. Belonging to the ray. Radiate (-atus). Spreading from or arranged around a common centre, or around the circumference of a cir- cle ; bearing rays or ray-flowers. Radiately veined. Same as Palmately veined; 93. Radiatiform (-ormis). Said of a capitu- lum of flowers which is radiate by en- largement of some of the outer flowers, which however are not truly ligulate, as in species of Centaurea. Radical (-alls). Belonging to or pro- ceeding from the root, or from a root- like portion of stem at or below the surface of the soil. Radicant (Radicans). Rooting. Radicel. A minute root or a rootlet. Radiciflorous (-us). Flowering (appar- ently) from the root. Radiciform (-ormis), Radicinus. Of the nature or appearance of a root. Radicle (-icula). Literally a diminutive root; but the " radicle " of the embryo, 430 GLOSSARY. so called in descriptive botany, is the hypocotyledonary aiid primal inter- node. (See Caulicle); 10. Radiculose (-osus). Bearing rootlets. Radix. The root. Rameal (-alts), Rameus. Belonging to (Ramus) a branch. Ramenta. Thin chaffy scales belonging to the surface or epidermis, such as the chaff on the stalks of many Ferns. Ramification. Branching; 47. Ramiflvrous (-us). Flowering on the branches. Ramose (-osus). Branching or branchy. Ramulose (-osus). Bearing many branch- lets, i. e. Ramuli or Raphe. See Rhaphe. Raphides or Rhaphides. Crystals in the cells of plants, especially needle- shaped crystals. Ray (Radius). One of the radiating branches of an umbel (147); also the marginal as opposed to the central part (or disk) of a head, umbel, or other flower-cluster, when there is a differ- ence of structure. Also used as an abbreviated expression for Rdy-jlowers. Those which belong to the margin of a circular flower-clus- ter, and differ from (being usually larger than) those of the disk. Recaulescence. The adhesion of leaves or their stalks to a stem ; 158. Receptacle (Receptdculum). A portion of axis forming a common support or bed on which a cluster of organs is borne. The receptacle of the flower, or the torus, is the axile portion of a blossom, that which bears sepals, petals, stamens, and pistils; 167, 211. The receptacle of inflorescence is the axis or rhachis of the head, spike, or other dense cluster ; 143. Redinate (-atus), Reclined, Reclining. Falling or turned toward downward, so that its upper part rests on the ground or other object; 53, 133. Itectineiirius. Straight-veined or straight- nerved; 92. Rfctiserial ( -ialis). In rectilinear ranks ; 124. Recurved (-us), Recurvatus. Curved backward or downward. Reduplicate (-atus) or Reduplicativus. Folded and projecting outward. Re flexed (-us). Abruptly bent or turned downward or backward. Refracted (-m). Same as reflexed, but abruptly bent from the base. Regma. A two-several-lobed two- several-celled fruit (2-pluricoccous), which separates at maturity into as many 2-valved carpels, as in Euphor- bia ; one form of Schizocarp. Regmacarp (-arpium). A general name of a dry and dehiscent fruit , 292. Regular (-am). Uniform in shape or structure; symmetrical as respects shape; 175. Reniform (-ormis). Kidney-shaped; having the outline of the longitudinal > section of a kidney ; 96. Repand (Repandus). With slightly un- even margin, which, if more pro- nounced, would be sinuate ; 98. Repent (Repens). Creeping, f. e. pros- trate or horizonal and rooting; 53. Replicate (-atus), Replicativus. Folded backward. Replum. A frame-like placenta (like a door-case), from which the valves of a capsule or other dehiscent fruit fall away in dehiscence, as in Cruci- ferae, certain Papaveraceae, Mimosa, &c.; 293. Reptant (Reptans). Same as Repent. Resupinate. Upside down, or having that appearance. Rete. Network. Reticulated (-atus), Retiformis. In the form of network ; netted. Reticulate-veined, 92. Retindculum. Name sometimes applied to the gland to which one or more pollinia are attached in Orchids, &c. The persistent and indurated hook- like funiculus of the seeds in most Acanthaceae. Retinerved (-ius). Same as Reticulate- veined. Retrocurved (-us). Same as Recurved. Retroftexed (-us). Same as Reflexed. Retrorse (-orsus). Directed backward or downward. Retroverted (Retroversus). Inverted. Refuse (Returns). With a shallow or obscure notch at a rounded apex ; 97. Reversion. A changing back, or in the reverse direction ; 171. Rerolute (^utus). Rolled backward from the margins or apex; 133. Rhachis. The axis (backbone) of a spike or of a compound leaf; 101, 143. Rhaphe. The adnate cord or ridge which in an anatropous ovule con- nects the hilum with the chalaza; 279, 307. Rhipidium. A fan-shaped cyme; 156. GLOSSARY. 431 Rhizanthous (-us). Root-flowered ; flower- ing from the root or seeming root. Rhizina. The peculiar roots or root- hairs of Mosses, Lichenes, &c. Rhizocarpous (-us). Rhizocarpic (-icus). Literally root-fruited; used by De- Candolle for a perennial herb. Rhizome, Rhizoma. A rootstock; a stem of root-like appearance pros- trate on or underground, from which rootlets are sent off; the apex pro- gressively sending up herbaceous stems or flowering stalks and often leaves; 56. Rhizomorphous (-us). Root-like in ap- pearance. Rhombic (-icus). Rhomb-shaped. Rhomboidal (-alls). Approaching a rhombic outline ; quadrangular, with the lateral angles obtuse. Rib. A primary and strong vein or conspicuous portion of the framework of a leaf; 92. Ribbed. Furnished with prominent ribs. Rictus. The mouth or gorge of a bila- biate corolla. Mima. A chink or cleft. Rimose (-osus). With chinks or cracks, like those of old bark. Ring. In Ferns, &c. See Annulus. Ringent (-ens). Grinning or gaping ; as is the mouth of an open bilabiate corolla; 248. Riparlus. Growing along the banks of rivers, &c. Rivalis Growing along brooks. Rivularis. Growing in watercourses or rivulets. Root. The descending axis. Roots are axes which grow in the opposite di- rection from the stem, are not com- posed of nodes and internodes, are mostly developed underground, and absorb moisture, &c., from the soil; 27- Root-cap, 13, 28. Hoot-hairs. Attenuated unicellular outgrowths or hairs from the newly formed parts of a root, for absorp- tion ; 13, 29. Rootlet. A very slender root or branch of a root. Rootstock. See Rhizoma ; 56. Roridus. Dewy ; covered with particles resembling drops of dew. Rosaceous (-eus). Arranged like the five petals of a normal rose; 246. Sometimes used for rose-color. Rosens. Rose-colored; pale red. Rostellate (-atus). Diminutive of Ros- trate. Rostellum. A diminutive beak. Also the name applied by Linnaeus to the Caulicle or Radicle. Rostrate (-atus). With a Rostrum, a beak or spur; narrowed into a slender tip or process. Rosular, Rosulate *(-atus). Collected in a rosette. Rotate (-atus). Wheel-shaped; circular and horizontally spreading very flat ; 248. Rotund (Rotundus, Rotundatus). Round- ed in outline; 95. Rough, Roughish. See Scabrous. Rubellus, Rubescent (-ens), Rubens. Red- dish. Rubescent also is turning red. Ruber. Red in general. Rubicundus Blushing, turning rosy- red. Rubiginose (osus). Brownish rusty-red. Ruderal ( -alls). Growing in waste places or among rubbish. Rudiment. An imperfectly developed and functionally useless organ; a Vestige. Rufous (-us), Rufescent (-ens). Pale red mixed with brown. Rugose (-osus). Covered or thrown into wrinkles, Rugce. Ruminated (-atus). As if chewed; said of the albumen of a nutmeg, &c. ;^511. Runcinate (-atus). Saw-toothed, or sharply incised, the teeth or incisions retrorse. Runner. A prostrate filiform branch which is disposed to root at the end or elsewhere ; 53. Running. Same as Repent. Rupestris, Rupicola. Growing on rocks or in rocky places. Ruptilis. Bursting irregularly. Rusty. Same as Rubiginose, Rufescent, and Ferruginous. Rutilans. Deep red with a metallic lustre. Sabulosus. Growing in sandy places. Saccate (-atus), Sacciform. Sac-shaped; baggy. Sagittate (-atus), Sagittiform (-ormis). Arrow-head-shaped. Salsuginosus. Growing within reach of salt water. Salver-fhaped. See Hypocraterimor- phous; 248. 432 GLOSSARY. Samara. An indehiscent winged fruit ; 294. Samaroid. Resembling a samara. Sap-wood. New wood of an exogenous stem; 80. Sarcocarp (-arpium). The succulent or fleshy portion of a drupe ; 285. Has been proposed also as a general name for a baccate fruit ; 292. Sarmentose (-osus). Producing long and lithe branches or runners, viz. Sarments (Sarmenta). Sativus. That which is sown or planted. Saw-toothed. See Serrate. Saxdtilis, Saxosus, Saxicolus. Living on or among rocks. Scabridus, Scabriusculus. Roughish ; diminutive of Scabrous (Scaber). Rough to the touch. Scalariform (-ormis). Ladder-shaped; with transverse markings like the rounds of a ladder. Scales. Any thin scarious bodies, usu- ally degenerate leaves, sometimes of epidermal origin. Scalloped. Same as Crenate ; 98. Scaly. See Scarious, Squamose. Scaly Suds, 40. Scandent (-ens). Climbing, in what- ever mode; 51. Scape (Scapus). A peduncle rising from the ground ; 51, 143. Scapiform (-ormis), Scapose (-osus). Re- serrfbling a scape. Scapigerous (-us). Scape-bearing. Scar. The mark left on the stem by the separation of a leaf, or on a seed, &c., by its detachment. Scarious or Scariose (-osus). Thin, dry, membranaceous, and not green. Schizocarp (-arpium). A pericarp which splits into one-seeded pieces ; 296. Scion. A young shoot ; a twig used for grafting. Sciuroideus. Like a squirrel's tail. SclerantJtium. Name of the fruit of Mirabilis, and the like; an akene enclosed in an indurated portion of calyx-tube. Scleroideus. Having a hard texture ; from Scleros, hard. Scobiform (-ormis). Having the appear- ance of sawdust. Scorpioid. A form of unilateral inflo- rescence which is circinately coiled in the bud : in the stricter sense, a form with the flowers two-ranked, these being thrown alternately to the right and left; 155,157- Scrobiculate (• atus). Marked by minute or shallow depressions. ScrotlJ'orm (-ormis). Pouch-shaped. Scurf. Small and bran-like scales on the epidermis. Scutate (-atus), Scutiform (-ormis). Buckler-shaped. ScutelKform (-ormis). Platter-shaped. Scymetar-shaped. See Acinaciform. Sectile (-His). As if cut up into portions. Section (Sectio). In classification, is applied in a general way to a divi- .jsion in the arrangement of genera, species, or other groups ; 327. Sectus. Completely divided ; 99. Secund (Secundus). When parts or organs are all directed to one side. Secundiflorus. With flowers of a cluster all secund. Secundine. The second (inner) coat of an ovule ; 277. Seed. The fertilized and matured ovule ; the result of sexual reproduction in a phaenogamous plant; 305. Seed-leaves. Cotyledons, 11. Seed-stalk. See Funiculus and Podo- sperm. Seed-vessel. See Pericarp. Segetdlis. Growing in grain-fields. Segment (Segmentum). One of the divisions into which a plane organ, such as a leaf, may be cleft. Segregate (-atus). Separated; kept apart. Semen. Seed. Semi. Half, in Latin compounds ; such as Semi-adherent. The lower half adhe- rent, &c. ; Semi-amplexicaul (-aulis), half clasping the stem ; Semiovate, ovate halved lengthwise, &c. Semiandtropous. Same as Amphitro- pous; 279. Semilunar, Semilunate (-atus). A syn- onym of Lunate, being like a half- moon. Seminal (-alis). Relating to the seed. Seminiferous (-us). Seed-bearing. Sempervirent (Sempervirens). Ever- green. Senary (-arhis). In sixes; 176. Sepal (Sepalum). A calyx-leaf ; 165. Seprtline (-inus), Sepalous. Relating to sepals. Sepaloid (-oideus). Resembling a sepal. Sepalody. Name for the metamorphosis of petals, &c., into sepals or sepaloid organs; 174. Separated Jlowers. Those of distinct sexes; same as Diclinous; 191. GLOSSARY. 433 Septate (-atus). Separated by a parti- tion or septum. Septictde, Septiculal (-cidus). When a capsule dehisces through the dissepi- ments or lines of junction ; 289. Septiferous (-us). Bearing the partition or dissepiment. Septifraijal (-us). Where the valves in dehiscence break away from the dis- sepiments; 290. Septum. Any kind of partition, whether a proper dissepiment or not. Septulate (-atus). Divided by spurious or transverse septa. Serial (Serialis) or Seriate (Seriatus). Disposed in series or rows, whether transverse or longitudinal. Sericeous (-eus). Silky; clothed with close-pressed soft and straight pubes- cence. Serotinous (-us). Produced compara- tively late in the season. Serrate (-atus). Beset with antrorse teeth; 97. Serrulate (-atus). Serrate with very small or fine teeth; 97. Sesqui. A Latin prefix denoting one and a half; as, Sesquipedalis, a foot and a half. Sessile (-ilis). Sitting close, without a stalk ; destitute of peduncle, pedicel, or petiole, as the case may be. Set'i. A bristle, or bristle-shaped body. Setaceous (-eus). Bristle-like. Setiform (-ormis). In the form of a bristle. Setigerous (-us). Bristle-bearing. Setose (-osus). Beset with or abounding in bristles ; bristly. Setula. Diminutive of Seta. Setulose (-osus). Bearing or consisting of minute bristles. Sex. Latin for six; as in Sexangular, Sexfarious, Sexpartite, &c. Shaggy. Pubescent with long and soft hairs ; same as Villous. Sheath. A tubular or enrolled part or organ, such as the lower portion of the leaf in Grasses. See Vagina. Sheathing. Enclosing as by a sheath. Shield-shaped. In the form of a buckler ; plane and round or oval, with stalk attached to some part of the under surface; 96. See Clypeate, Scutate, Peltate. Shrub. A woody perennial of less size than a tree : 50. Shrubby. Having the character of a shrub. Sieve-cells^ 77. Sigillate (-atus). As if marked with the impression of a seal, as the rootstock of Polygonatum. Sigmoid (-oideus). Doubly curved like the Greek s or the capital S. Silicle (Silicula). A short silique, not very much longer than wide ; 294. Siliculosa. Name of the Linmvan arti- ficial order of the class Tetradynamia, having Siliculose pods ; 337. Silique (Siliqua). The peculiar pod of Cruciferae, especially when much longer than wide; 293. Siliquosa. Name of the other order of Tetradynamia, with Siliquose fruit, Le. a Silique; 337. Silky. See Sericeous. Silver-grain. The glittering plates in exogenous wood belonging to the medullary rays; 74. Simple (Simplex). Of one piece, series, &c. A simple pistil is of one carpel ; a simple leaf, of one blade, &c. Simple Fruits, 291. Simplicissimus. Most simple ; complete- ly simple. Sinistrdrse. Turned or directed to the left; 51, 140. Sinuate (-atus). With a strongly wavy or recessed margin ; 98. Simts. A recess or re-entering angle. Slashed. Same as Laciniate. Smooth. Either opposed to scabrous, i. e. not rough, or to glabrous, »'. e. not pubescent ; the former is the more correct application. Soboles. Shoots, especially those from the ground. Soboliferous (-us). Bearing vigorous lithe shoots. Solid Bulb. A corm ; 61. Solitary (-arius). Single, only one from the same place. Solubilis. Separating into portions or pieces. Solutus. Loosed; becoming separate. Sordidus. Of a dull or dirty hue. Sorediate (-atus). Bearing small patches on the surface. Sorema. A heap of carpels belonging to one flower; 263. Sori, sing, sorus. Heaps, such as the clustered fruit-dots of Ferns. Sorose. Heaped or bearing Sori. Sorosis. A flesh}- multiple fruit, such as a mulberry, bread-fruit, and pine- apple. Spadiceus. A bright and clear brown, or chestnut color 434 GLOSSARY. Spadiceous. Having the nature of or bearing a SpcuJix. A spike with a fleshy axis ; 1 49. Span. The length of the space between the tip of the thumb and that of the little linger, when outstretched; about nine inches. Sparsus. Sparse or scattered; whence Sparsifloi-us, with scattered flowers; Spa rsifolius, with scattered leaves, &c. Spathnceous (-us). Spathe-bearing, or of the nature of a Spatke (Spatha). A large bract, or a pair of bracts, enclosing a flower-clus- ter; 142. Spathella. An unused name for the glumes of Grasses. Spathilla. A secondary or diminutive Spalulate (Spathulatus). Oblong with the lower end attenuated, shaped like a druggist's spatula; 95. Species. The particular kind, the unit in natural history classification; 317. Specific Character, Name, &c., 349, 363. Spermaphore or Spermophore (-orum). A name for the Placenta. Spermoderm (-ermis). The outer seed- coat; 305. Spermodophorum or Spermophorum. An unused name for the gynophore in Umbelliferae. The latter also an un- used name for the Placenta ; 261 . Spermotheca. An unused name for peri- carp. Spermum. Latin form of the Greek word for seed. Lat. Semen. Sphalerocarpium. Name proposed for an accessory fruit, such as that of Shepherdia, in which an akene is enclosed in a baccate calyx-tube. Spica. See Spike. Spicate (-atus). In the form of or resem- bling a spike, or disposed in spikes. Spiciform (-ormis). Spike-like. Spicula. A diminutive or secondary spike; a Spikelet. Spike (S/rica). A form of indeterminate inflorescence, with flowers sessile on an elongated common axis ; 149. Spikelet ( Spicul/t ). A secondary spike ; the name given to the Locusta or clus- ter of one or more flowers of Grasses subtended by a common pair of glumes. Spindle-shaped. See Fusiform. Spine (Spina). A sharp-pointed woody or i ndurated body, commonly a branch, sometimes a petiole, stipule, or other part of a leaf; 55, 117. Spinescent (-ens). Ending in a spine or sharp point ; 55. Spinose (-osus). Furnished with spines, or of a spiny character ; 55. Spinuliferous or Spinulose (-osus). Fur- nished with diminutive spines or Spin- ulce. Spiral (Spiralis). As if wound round an axis. Spiral Ducts, 68. Spiral Fhyllotaxy, 119, 121. Spirides. The delicate coiled threads in the hairs on the surface of certain seeds and akenes, which uncoil when wet; 307. Spitkamceus. A span long; the length spanned between the tip of thumb and forefinger when extended. Splendens. Resplendent or glittering. Spongelet, Spongiole (-iola). Name given to young root-tips ; once sup- posed to be a peculiar organ; 28. Sporadic (-icus). Widely dispersed or scattered. Sporangium. A spore-case or theca con- taining the analogues of seeds (spores) in the higher Cryptogams. Spore (Spora, Greek for seed). The analogue of seed in Cryptogams. Spore-case. See Sporangium. Sporidium. Synonym or diminutive of Spore. Sporiferous. Spore-bearing. Sporocarp (-arpium). Name given to certain spore-cases, as of Lycopodi- Sporophore (-orum). One of the syno- nyms of Placenta. Sporule (Sporula). Diminutive spore or a sort of spore. Sporullferous (-us). Bearing or con- taining spores. Sport. A bud-variation or seed-varia- tion; 319. Spumescenl (-ens), Spumose. Froth-like in appearance. Spur. A hollow and slender extension of some portion of the blossom, usu- ally nectariferous, as of the calyx of Larkspur and the corolla of Violet: rarely applied also to a solid spur-like process. Spurred. Producing a spur. See Cal- carate. Squama. A scale of any sort, usually the homologue of a leaf. Squamate (-atus), Squamiferous, Squo- mosus. Furnished with scales. GLOSSARY. 435 Squamella, Squdmula. Diminutive squama; scales of secondary order or reduced size. Squamiforrn (-ormis). Scale-like. Squamulose (-osus). Covered or beset with minute scales. Squarrose (-osus). Literally rough- scurfy; applied to bodies rough with spreading and projecting processes, such as tips of bracts, &c. Squarrulose (-osus). Diminutively squar- rose. Stachys. Greek for spike. Stalk. Any kind of lengthened support on which an organ is elevated. Stamen. One of the elements or phylla of the androecium ; 105. Stamineal, Stamineous (-em). Relating to the stamens ; 191. Staminiferous (-us). Stamen-bearing. Staminodium. A sterile stamen, or what answers to a stamen, whatever its form, without anther. Staminody. Name for the metamor- phosis of other floral organs into stamens ; 174. Standard. The posterior petal of a papilionaceous corolla; 184. Stans. Supporting itself in an erect position. Station. Particular place as to soil, ex- posure, &c., which a plant affects; 366. Stellate (-atus). Star-shaped, arranged like the rays or points of a star. SteUulale (-atus) or Stellular. Dimin- utive of Stellate. Stem. The main ascending axis of a plant; 45. Stemless. See Acaulescent ; with no leaf- bearing stem above ground ; 45. Stemlet. Diminutive stem ; as that of the plumule. Stenos. Greek for narrow ; hence Stinophyllus. Narrow-leaved, &c. Steriytna. Any foliaceous prolongation of the blade of a leaf down on the stem by decurrence. St-rigmum. Name of Desvaux for the Dieresilis of Mirbel. Sttrile (-ills). Barren, as a blossom destitute of pistil, 191 ; a stamen without anther, or an anther without pollen; an ovary, without good ovules, seeds without embryo, &c. In com- mon English use, a male or staminate flower is said to be a sterile flower. Stichi.s. Greek for row or rank, usually meaning vertical rank ; hence such compounds as Distichous, two-ranked ; Tristichous, three-ranked, &c. Stiyma, pi stigmata. That part or sur- face of a pistil (usually on or a part of the style, or in place of it) which re- ceives the pollen for the fecundation of the ovules; 166. Stigmatic (-icus), Stigmatose (-osus). Relating to stigma. Stigmatiferous. Stigma-bearing. Stings. Stinging hairs, seated on a gland which secretes an acrid liquid, as in Nettles. Stipe ( Stipes). A stalk of various sorts ; the support of the cap of a mush- room ; the leafstalk of a Fern ; any stalk-like support of a gynoecium or a carpel; 212. Stipel (Stipellum). An appendage to a leaflet analogous to the stipule of a leaf; 106. Stipellate (-atus). Provided with stipels ; 106. Stipitate (-atus). Having a stipe or special stalk. Stipitiform (-ormis). Shaped like a stipe; stalk-like. Stipulaceous (-eus), Stipular (-arts). Belonging to stipules. Stipulate. Possessing stipules. Stipules. Appendages or adjuncts of a leaf one on each side of the insertion ; 85, 105. Stirps, pi. stirpes. A race. Stock. Synonym of Race ; also the portion of a stem to which a graft is applied; a caudex, rhizoma, or root- like base of a stem from which roots proceed; 51. Stole, Stolon (Stolo). A sucker, runner, or any basal branch which is disposed to root ; 53. Stoloniferous (-us). Sending of or propa- gating by stolons, runners, &c. Stoma, pi. stomata, Stomate. One of the apertures in the epidermis of folia- ceous parts, through which cavities within communicate with the external air ; 89. Stomatiferous (-us). Bearing stomata or "breathing pores." Stone. The hard endocarp of a drupe. Stone-fruit. A Drupe, such as a peach or plum : 297. Stool. The plant from which layers are propagated, by bending down to the ground to be rooted. Stramineous (-eus). Straw-like or straw- colored. 436 GLOSSARY. Strap-shaped. See Ligulate (247) and Lorate. Striate (-atus). Marked with fine longi- tudinal lines, streaks, or diminutive grooves or ridges (Strife). Strict (Strictus). Close or narrow and upright; very straight. Striyillose (-osus). Minutely strigose. Strigose (-osus). Beset with striyce, or sharp-pointed and appressed straight and stiff hairs or bristles. Strobilaceous (-eus), Strobiliform (-or- mis). Relating to or resembling a Strobile. Strobile (Strobilus). An inflorescence formed largely of imbricated scales, as that of Hop and a Fir-cone; 303. Strombuliformis, Strombuliferus. Twist- ed spirally into a screw shape, as the legumes of the Screw-bean (Proso- pis, sect. Strombocarpa) and of some species of Medicago. Strophiole (-iola). An appendage at the hilum of certain seeds; 308. Structural Botany, 2. Struma. A wen or any cushion-like swelling on an organ. Strumose (-osus), Strumiferous (-us). Furnished with a struma or goitre-like swelling. Stupose (-osus). Tow-like; with tufts or mats of long hairs. Style (Stylus). The usually attenuated portion of a pistil or carpel between the ovary and stigma; 166. Styliform (-ormis). Style-shaped. Styliferous. Style-bearing. Stylinus. Belonging to the style. Stylosus. With styles of remarkable length or number, &c. Stylopodium. An enlargement or a disk- like expansion at the base of a style, as in Umbelliferse. Sub. In composition of Latin words in terminology, denotes somewhat or slightly; as, Subacute, Subcordate, that is acutish, somewhat cordate, &c. Subclass, 327. Subconvolute and Subimbricaie in aestiva- tion, 137. Suberose (-osus). Of a corky texture. Subyenus, 327. Submerged, Submersed (-us). Growing under water. Suborder (Subordo), 327. Subpetiolar (-aris). Under the petiole, as the leaf-buds of Platanus ,• 42. Subsection, 327. Subspecies. A group which is ambigu- ous in rank between variety and spe- cies; 320. Subtribe (Subtribus), 327. Subulate (-atus), Subuliform (-ormis). Awl-shaped. Subvnriety, 327. Succise (-isus). As if cut or broken off at the lower end. Succubous (-us). When in leaves crowded on a stem the apex of eac1} leaf is covered by the base of the ed. See Bilabiate. Type. The ideal plan or pattern. Typical. Representing the plan or type. Uliginose (-osus). Growing in swamps. Ulndris. Of the length of the ulna or fore-arm. Umbel ( Umbcllit ). An inflorescence (properly of the indeterminate type) in which a cluster of pedicels spring all from the same point, like rays of an umbrella; 146. Umbellate (-atus). UmbeUiform (-ormis). In or like umbels. Umbellet. A partial or secondary um- bel; 150. Umbelliferous (-us). Bearing umbels. 440 GLOSSARY. Umbellula. A partial or secondary um- bel, or umbellet; 150. Umbilicate (-atus). Depressed in the centre, navel like. Umbilicus. The hilum of a seed. Umbonate (-atus). Bearing an Umbo or loss in the centre Umbrdculiform (-ormis). Having the general form of an umbrella. Umbrosus. Growing in shady places. Unarmed. Destitute of prickles, spines, or other armature. Uncate (-atus), Uncinate, (-atus), Unci- J'orm (-ormis). Hooked; bent or curved at tip in the form of a hook. Untidlis. An inch (uncia) in length. Undate (-atus) or Undulate (-atus). Wavy; 98. Undershrub. A very low shrub ; 50. Unequally pinnate. See Impari-pin- nate. Unguiculate (-atus). Contracted at base into an Unguis. A claw, or stalk-like base of a petal, &c. ; 245. Uni-. In Latin compound, one; as Unicellular. Of one cell; Unicolor, of t one color, &c. Unicus. Singly or single, solitary. Uniflorous (-us). One-flowered. Unifoliate (-atus). One-leaved. Unifoliolate, of one leaflet ; 102. Unijuffate (Unijugus). Of one pair; 102.' Unilabiate (-atus). One-lipped, like the corolla of Acanthus, in which the upper lip is obsolete. Unilateral (-alls). One-sided; either originating on or more commonly turned all to one side of an axis. Unilocular (-am). One-celled. Uninervate ( Uninervis, Uninervius). One-nerved. Uniovulate (-atus). Having only a soli- tary ovule. Unij)arous. Bearing one ; as a cyme of one axis or branch ; 152, 155. Uniserial (-ialis), Uniseriate (-atus). In one horizontal row or series. Unisexual (-alis, Unisexus). Of one sex; having stamens only or pistils /0nly; 191. Univalved ( Univalvis). Of one piece or valve. Urceolate (-aim). Hollow and con- tracted at or below the mouth, like an urn or pitcher ( Urceolus). Urens. Stinging, in the manner of net- tles. Utricle ( Utriculus). A small bladdery pericarp; 295. Or any small bladder- shaped body or appendage ; also a synonym of a cell of parenchyma. Ucricular (-aris), Utriculate (-atus), Utriculiform (-ormis), Utriculose (-osus). Having or consisting of utricles, or bladder-like in appear- ance. Vacillans. Swinging free, as the anth- ers of Grasses on their filaments. Vdcuus. Void or empty of the proper contents. Vayina. A sheath, as of a leaf, &c. Vayinate. Sheathed. Vail culce. The intervals or grooves between the ridges or ribs of the fruit Umbellifer*. Valcate (-atus), Valvular (-aris). Opening as if by doors or valves, as do most dehiscent fruits (capsules), and some anthers; also the parts of a flower-bud when they exactly meet without overlapping ; 135. Valce ( Valva). One of the pieces into which a capsule splits, 288. Valved. Same as valvate: hence 3 -valved, 5-valved, many valved, &c. Vdlvula. A diminutive valve. Also used (after Linna;us) for the inner or flower-glumes of Grasses. Variegated (-atus). Irregularly colored ; in patches of color. Varitty ( Varietas). A sort or modifi- cation subordinate to species; 318. Variolate, Varioldris. Marked as if by the pustules or pittings of small- pox. Vascular (-aris). Relating to or fur- nished with vessels ( Vasa) or ducts. Vascular Plants ( Vasculares), 340. Vdsculum. Same as Ascidium. Also the botanists' collecting box ; 372. Vasifvrm (-ormis). In the form of a vessel, duct, &c. Veined. Furnished or traversed with fibro-vascular bundles or threads, es- pecially with those which divide and are reticulated. Veins ( Vena). In general any ramifi- cations or threads of fibro-vascular tissue in a leaf or any flat organ; especially (as distinguished from nerves) those which divide or branch; 92. Veinless. Destitute of veins. Veinlet ( Venula). One of the ultimate GLOSSARY. 441 or smaller ramifications of a vein or rib ; 93. Velate (-atus). Veiled. Velutinous (Vdutinus). Velvety: the surface covered with a soft coating of fine and close silky pubescence, or velumen. Venation ( Venatio). The mode of vein- ing; 90. Venenatus, Venenosus. Poisonous. Venose (-osus) Veiny; abounding in veins or network. Ventral (-alls). Belonging to the an- terior or inner face of a carpel, &c.; the opposite of dorsal. Vtntricose (-osus). Swelling unequally or inflated on one side. Ventriculose (-osus). Minutely ventri- cose. Venulose (-osus). Abounding with vein- lets or venulae. Vermicular (-am). Worm-shaped. Vernal ( Verndlis). Appearing in spring. Vernation (-atio). The disposition of parts in a leaf -bud ; 132. Vemicose (-osus). As if varnished. Verrucose (-osus). Covered with warts (verrucas) or wart-like elevations. Versatile ( Versdtilis). Swinging to and fro ; turning freely on its support ; 253. Versicolor. Changing color, or of more than one tint or color. Vertex. The apex of an organ. Vertical (-alis). Perpendicular to the horizon ; longitudinal. Verticil (-illus). A whorl ; 6. Verticillaster. A false whorl, composed of a pair of opposite cymes ; 159. Verticillastrate. Bearing or arranged in Verticillasters. Verticillate (-atus, -arts). Disposed in a whorl; 6, 119, 120. Vescicle (-icula). A small bladder or air-cavity. Vesicular (-am), Vesiculose (-osus). As if composed of little bladders. Vespertine ( Vespertinus). Appearing or expanding in early evening. Vessels ( Vasce). See Ducts. Vexillary (-aris), Vexillar, 137. Per- taining to the Vexillum. The standard or large pos- terior petal of a papilionaceous corolla; 184. Villose (osus) or Villous. Bearing shaggy or long and soft (not interwoven) hairs or VOK. Vitnineous (-eus). Bearing long and flex- ible twigs, like those used for wicker work. Vine. Any trailing or climbing stem : originally that of the Grape from which wine is made. Vinedlis. Growing in vineyards. Violaceous (-eus). Violet-colored. Virens. Green, or evergreen. Virescens. Greenish or turning green. Virgate (-atus). Wand-shaped, or like a rod ; slender, straight, and erect. Viryultum. A vigorous twig or shoot. Viridescent (-ens). Same as Virescens. Viridis. Green. Viridulus. Greenish. Virosus. Venomous. Viscid (-idus), Viscous (-osus). Sticky from a tenacious coating or secretion. Vitellinus. The yellow hue of the yolk of egg. Vitellus. Name formerly given to the peculiar albumen which is in some cases deposited within the embryo-sac. Viticulose (-osus). Sarmeiitaceous ; pro- ducing vine-like twigs or suckers, viticulce. Vittce. The fillets or stripes (oil-tubes) of the pericarp of most Umbelliferre, which contain an aromatic or peculiar secretion. Vitiate (-atus). Bearing vittoe ; or with any longitudinal stripes. Viviparous ( -us). Germinating or sprout- ing from seed or bud while on the parent plant. Volulile (Volubilis). Twining round a support; 51. Volutus. Rolled up in any way. Volva. A wrapper or external covering, especially that of many Fungi. Wavy. See Undulate. Waxy. Resembling beeswax in consist- ence or appearance. Wedye-shaped or Wedge-form. See Cuneate; 95. Wheel-shaped. See Rotate. Whorl. Arranged in a circle round an axis; a Verticil; 6. Whorled. Disposed in whorls. Wild. Growing without cultivation; spontaneous. • Wing. See Ala. Any Membraneous or thin expansion by which an organ is bordered, surrounded, or otherwise augmented. Also the two lateral petals of a papilionaceous corolla are termed wings; 185. 442 GLOSSARY. Winged. See Alate : bearing a wing or wings. Witlierln,]. See Marcescent. Wood. The hard part of a stem, &c., mainly composed of Wood-cells, Woody Jibre or tissue, 68. Woolly. See Lanate and Tomentose : clothed with long and tortuous or matted hairs. Xanthos. Greek for yellow in com- pounds, such as Xanthophyll, the yel- low coloring matter in leaves. Xenogamy. Fecundation of the ovules of a flower by pollen from some other plant of the same species; cross-fer- tilization; 216. Xylinus. Woody, pertaining to wood. Zoospore. One of the free-moving spores of the lower Cryptogams. Zy (/amorphous {-us}. That which can be bisected in only one plane into similar halves; 175. ADDENDA. Antidromous, Antidromy. When the course of a spiral is reversed, 157. Infertile (-ilia). Said of a pistil or flower which fails to set fruit. P<'lyembryony. The production of two or more embryos in a seed, 284. Saprophytes (-yta). Plants feeding upon decaying vegetable or animal matter . THE LIBRARY UNIVERSITY OF CALIFORNIA Santa Barbara , THIS BOOK IS DUE ON THE LAST DATE STAMPED BELOW. APR 3 1330 RETDAPR4 19902 I