THE LIBRARY OF THE UNIVERSITY OF CALIFORNIA LOS ANGELES THE LIVING PLANT FROM SEED TO FRUIT BLOOD-LIPPED ONCID (Oncidiuu, The One-ids are a large group of Orchids, natives of tropical America and the West Indies. They all grow upon trees (epiphytes) and exhibit great variety ot form and colour. The species figured is a native of New Grenada. The leaves are thick and leathery, and the flowers home on erect spikes POPULAR BOTANY THE LIVING PLANT FROM SEED TO FRUIT BY A. E. KNIGHT AND EDWARD STEP, F.LS VOLUME II WITH 721 BEAUTIFUL ILLUSTRATIONS AND 18 COLORED PLATES NEW YORK HENRY HOLT AND COMPANY PRINTED IN GREAT BRITAIN VOL. II. CONTENTS. QK 50 v.o. CHAPTER X. THE LEAF IN KELATION TO ITS ENVIRONMENT (continued) . PAGE 289 XI. FLORAL FORMS AND THEIR EELATIONS TO INSECTS 312 XII. THE WELCOME OF THE FLOWERS 358 XIII. SOME PLANT MARRIAGES AND THE GUESTS THAT ASSIST AT THE FUNCTION . 422 XIV. THE PROMISE OF THE PLANT THAT is TO BE . . . . . . . 464 XV. HIDDEN MARRIAGES ............ 500 GLOSSARY OF TERMS 574 INDEX ... ........ 577 COLOURED PLATES. Blood-lipped Oncid (Oncidium hcematochilum) Frontispiece Facing page Peach (Prunus pcrsica) ............ 289 321 353 385 417 449 481 , 513 Garlic (Allium acuminatum) Close-headed Befaria (Befaria coarctata) Variegated Oncid (Oncidium variegatum) Blue Vanda (Vanda ccerulea) . Cervantes' Odontoglot (Odontoglossum cervantesii) Speckled Odontoglot (Odontoglossum ncevium) Toothed Ceanothus (Ceanothus dentatus) ILLUSTRATIONS IN THE TEXT. PAGE Alder . . . .514 p Barberry .... AGE 405 Alocasia spectabilis . 443 Bedeguar Gall, Flowering Antheridium, An 505 Branch of a Rose sur- Antherozoids 506 rounded by a 304 Anthers and their Dehiscence 354 Beech, Hairy Galls on. 305 Apple . . . 357, 473 Begonia, Abnormal Flower of 324 Aralia nudicaulis, Flower- — Section of 332 buds of ... 386 Bellflower, Nettle-leaved 425 Arbor-vita?, Cone of Chinese 489 Birth wort . 374 Archegonium, An 505 Bleeding Heart . 320 Arum, Giant 317 Brefeldia maxima 570 Asclepias cornuti, Pollinia of 408 Burweed . 486 Ash 368 Buttercup . 321 Aspen .... 448 — Apocarpous fruits . 334 Aspidistra . - . 376 442 — from below . — Rosaceous Corolla of 321 312 PAGE Buttercups, Section showing receptacle to which the carpels are attached . 322 Cabbage, Skunk . . 387 Cactus, A Prickly Pear . 300 Calla, ilarsh . .361, 441 Campion, White. . 340, 446 Candytuft, Evergreen . 486 Carrot . . . .332 Catasetum tridentatum 408, 409 Caterpillar-plant . . 481 Chestnut, Horse . . 479 — Section of (JIale Flower). 366 - Sweet . . 367, 462 Chrysanthemum, Annual . 352 Clavaria, Crested . . 538 I PAGE Clematis caorulea, Flower of 328- — Garden . . .335- Climbing Cobasa, Section of flower in first stage of development ; and fruit 373^ Sectionof flower in third stage of development . 373! Clubs, Fairy . 549 Cluster-cups 546 — Hawthorn 550 Coco-nut, Double 488 Columbine . 348 Comfrey . 407 Convolvulus, Seaside 345 Corallina officinalis 553 Corncockle 474, 496 2054333 Illustrations in the Text, Vol. II. PAGE mar PAGF PAGE Cotton-grass . . .453 Horn of Plenty . . .548 Oak, Bristle-gUls on . . 302 Sallow . . . .369 Cowslip . . . .315 Horse Chestnut . . 360, 479 — Cherry-galls on . . 301 Samphire . . . .493 Cranesbill, Meadow . . 374 Horsetail . . . . 512 — Spangle-galls on . . 301 Saxifrage . . . ,394 — Blood-Red . . .494 — Field . * . 514, 515 — Turkey . . . .495 — Mossy . . . .383 Great . . . 519 i Oar-weed . . . .291 Seaweed, A Red . . 553 Crocus . . . .334 Hydrangea . . .351 Oat 316 — Ash -leaved . . .559 — Indian . . . .380 Inflorescences, Definite . 327 Odontoglossum alexandwe , 412 — Forked . 563 Lip of flower showing — Indefinite . . . 331 i Oncid, Tiger-striped . . 384 Sensitive Plant . . . ->92 fringes and streaks of Iris 477 Orange, Myrtle-leaved . 490 Snapdragon, A portion of the colour— honey-guides . 382 Jania rubens . . . 557 j Orchis, Bird's-nest . . 437 stigma and style . 326 Hairs of a portion of Jericho, Rose of . . 485 ' — Butterfly . . .426 — Section through pistil o* 322 two of the fringes of the Job's Tears . . . 492 . — Dwarf . . . .432 Snowdrop . . . 359 tip highly magnified . 382 Kalmia latifolia, Flower of . 406 — Goodyer's . . .431 Sorrel, Common . . 458 Cuckoo-flower . . .393 Lady's Tresses . . .419 — Green-man . . . 433 Sparassis, The . . . 539 Cuckoo-pint . . .313 Lapageria, Section of. . 444 — Green-winged . . 436 Spear Plume-thistle . . 299 Cucumber, Squirting . . 469 Larkspur . . 390, 481, 496 — Fragrant . . .435 Spleenwort, Maidenhair . 500 Cyclamen . . . .360 — Garden . . . .339 — Musk . . . .440 Stamens of (a) Litsaea ; (6) Cyperus, False . . .459 Leocarpus fragilis . .573 — Purple .... 434 Pyrola; (c) Garcinia . 356 Dandelion . . 467, 492 Leptodermis lanceolata . 498 — Pyramidal . . .438 Star, Earth . . . 537 — Floret of . . .329 Lichen, A . . . .527 — Spotted . 427, 429, 430 Stinging Hair of Stinging Dead-nettle, White . . 346 — Sections through . . 528 Ovaries, Uniovular . . 465 Nettle . . .296 Dewberry . . . .471 Lily, African . . .395 Oxlip . . . .376 Stinging Nettle . . .303 Dog Lichen . . .533 Liverwort . . 522, 526 Pseony, Follicles of . . 481 Stock . . . .482 Dragon's-mouth, Mountain . 391 Loosestrife, Creeping . .312 Pansy Flower, Section of . 392 Stonewort . . . .552 Dry Rot .... 542 — Purple . . . 363, 377 Papilionaceous corolla . 348 Stork's-bill, Hemlock . 423, 424 Earth-Nut . . .484 — — Illustrating the tri- Pasque-flower, Section of . 396 Strawberry . . . 472 Ergot . . . .541 morphism of the Passion-flower, Blue . . 422 Sycamore .... 496 Fern, Cup -shaped Indusinm flowers . . .378 Pea, Campylotropous Ovule Tangle, Sugar . . . 554 of Filmy . . .506 Lousewort, Marsh . . 365 of . . .466 Teasel . . . .297 — Hard .... 501 Lucerne . 482 — Common . . . 330 Telegraph -plant . . . 293 — Male . . 502, 503 Lupin .... 482 — Flower of ... 402 Thorn-apple . . . 476 ^- Prickly Buckler . . 507 — Leaves of ... 289 Pea Flower, Section of . 402 Toadflax, Ivy-leaved . . 483 — Sea . . . 511 Magnolia . . . .379 — Gall, Spiked . . 308 — Yellow . . . .349 — Spores Germinating . 504 Mallow, Common . . 494 Peacock's-tail . . .560 Toadstool, The Solitary . 536 Fig . . . . .388 Marigold, Marsh. . . 362 Petty Spurge, Inflorescence Of 396 Touch-me-not Balsam Ex- Figwort . . . .350 Meadow-saffron . . .477 Pillwort . 509, 510, 512, 513 pelling its Seeds . . 470 Flamingo Plant . . .399 Meadowsweet, American . 32?, Pimpernel . . . .478 Tremella mesenterica . . 544 Flowers, ^Estivation of . 329 Medlar . . . 473, 475 Pine, Austrian . . 451, 499 Trichia varia . . .569 Forget-me-not . . . 385 Megarrhiza calif ornica, Spiny — Scots . . 449, 450 Trigonia villosa . . .498 Foxglove . . 343 Fruit of 466 Piiifc SIR Fuchsia .... 310 Milk-thistle . . . 489, — Mossy . . - .344 Urchin Crowfoot . . 490 Fungus, Bird's-nest _ . 546 Mitrula phalloides . . . 545 Pistils . . . .355 Valerian, Great . . 341 — Candle-snuff . . . 547 Monkshood . . .336 Plantain, Large . - .457 Vallisneria spiralis . .413 — Jew's-ear . . .540 Moonwort .... 508 — Lamb's-ear . . . 381 — Flowers of . . . 414 Ginger, Wild . . .403 Morel .... 543 Plants, Lime-secreting . 556 Violet, Sweet , . . 358 Goatsbeard . . 319, 364 Morning Glory . . . 373 i Plumbago . . . .360 Water-dock, Great . . 455 Gourd . . . .489 Moss, A Scale . . .524 Plume, The Crimson . . 557 Water-lily, White . . 464 Grape Vine . . 344, 463 Great Panicled Sedge, — Beard . . . . 529 Plum, Section through a . 488 Wheat .... 492 Pollen-grains of . .454 Gum-Arabic Tree . 485 — Hair . . . 520,523 American Plum-tree . 487 Willow-herb, Great . . 372 Gymnosporangium clavariio- — Lesser Alpine Club . 517 , Pollen-grain Magnified . 454 Winter Cherry . . .341 forme . . . .551 Hairs from the Leaves of — Marsh Club . . . 516 j — of Evening Primrose . 400 — Screw . . . . 521 1 — Passion Flower . . 400 Wistaria, Abnormal Flower of 324 " Witches Broom " on Scots Various Plants (ilagni- Mosses, Two . . . 525 — Pumpkin . . . 400 Pine . . . .309 fied) . . . . 295 Mould, Brown . . . 534 — Scot's Pine . . . 400 Wood Anemone , .417 Hard Heads . . . 371 Mucilago spongiosa . . 568 Polysiphonia . . .558 — Rush, Field . . . 418 Hazel . . . 456, 491 Mulberry .... 473 Poplar . 496 — Sage .... 460 — Flowers . . . 416 Mullein, Black . . .291 Potato-blight . . .532 — Sorrel . . . . 290 Heath, Hispid . . .350 Heather or Ling . . 447 Myosotis Flowers. . . 370 Primrose. Bird's-eye . . 389 Myxomycete, Development j Primula obconiea . . 307 — — Expelling its Seeds . 468 • Seed of , . 469 Helleborine, Broad-leaved . 421 of a . . . . 571 Prothallium, A . . . 504 — Spurge .... 397 Henbane . . . 338, 353 Nail-galls . . . . 306 Pternan'ira cordata . . 338 Wrack, Bladder 562, 564, 565 Henbit .... 370 Herb-paris. . . .314 Holly . . . .398 — Hedgehog Variety of . 296 Halymenia ligulata . . 561 Honesty . . . 340,485 Hop Hornbeam . . . 497 Nasturtium . . 337. 494 Puft-balls, A Cluster of . 535 — with spurred calyx . 337 j Purse, Shepherd's . . 37o Nectar Glands . . .393 Quillwort . . . .513 Nectaries . . . .390 Ramalina scopulorum . . 531 Needle Whin . . .428 Rampion . . . ,347 Nettle, Flower of . . 414 ! Rose, Dog- and Field- .' 333 Nut, Brazil . . .480 Sage, Flower of . .404 — Saw-Edged . . .555 Wych Elm . . . 493 Yarrow . . . .445 Yew . . . 497, 498 Yoke-thread . , .566 Yucca, Flower of . . 410 Moth . 410 Horehound, Black . .493 Hornbeam . . . 46i Oak • • • . . 452 St. John's Wort, Largc- — Apple . . . .3111 flowered . . .325 Gathering Pollen . 410 — Plant . . . 411, 439 The Peach is belies The PEACH fpr n Eii rope from very ancient days, "amily as the Plums and Sloe. CHAPTER X THE LEAF IX RELATION TO ITS ENVIRONMENT— (Continued} THOSE who have never studied under the microscope the singular forms of the covering hairs of leaves, have pleasures offered to them for many a winter evening. Possibly a glance at the illustrations which ac- company this part of our text will help to kindle interest in the subject. Pig. 3606 represents some simple hairs of a species of Brassica : i some fwked hairs of the Whitlow Grass (Draba verna) ; and h a stellate hair of the pretty Alpine Madwort (Alyssum spinosum). In these three specimens the hairs are unicellular, but multicellular hairs are met with in a large number of plants. When the cells grow together in a line, like the beads of a necklace, the hairs are said to be moniliform. Of this kind are the epidermal hairs of the Marvel of Peru (Mira- bilisjalapa, fig. 360j) and of the Virginian Spiderwort {Tradescantia virr/inica, k). When the cells spring from a common point, as in the Cretan Horehound (Marru- bium creticum, «), the hairs FlG- 352.— LEAVES OF LUPIN. are said to be tufted. The flrst shows thtejy°10 Branched hairs, which need no describing, are found in several plants. They give the downy (tomen- tose) appearance to the leaves of Nicandra anomala, a Peruvian plant of medicinal value (/), and to the decurrent leaves of the Great Mullein (IV- bascum thapsus). In a few plants of succulent habit, like the South African Eocheas, excessive evaporation is prevented by the development of special flinty cells on the epidermis of the thick fleshy leaves (g). They are many times larger than the ordinary epidermal cells, and their walls are largely composed of silica. A covering of this kind is simply the salvation of its possessor in time of drought. Not that fleshy leaves are ill-adapted to II— 1 289 of the leaflets clurinsr the day ; at night iown close to the stalk. 290 HUTCHINSON'S POPULAR BOTANY FIG. 353. — WOOD-SORREL (Oxalis). During the day the leaflets spread out from the leaf-stalk ; FIG. 354. — WOOD-SORREL. —at night, and dining rain, they fold down close to the leaf-stalk. dry climates ; the reverse is the case, the transpiring surface of such leaves being much smaller than if the}' were flattened out into thin and spreading forms. Yet even the smaller surface needs to be protected, and this it is which gives its value to the flinty armour. The vertical position which many leaves assume is likewise a means of checking excessive trans- piration. One has met with the expression, '• the shad&wless forests of Australia," and the phrase is not inappropriate, for the leaves of many Eucalyptus-trees and Acacias (the chief timber trees in Aus- tralian woods) do not assume a horizontal position like the leaves of most European forest-trees, but are placed vertically on edge ; and thus the shadows which they cast at midday are reduced to a mere line. This, as we need scarcely add, is due to the fact that the rays of light fall upon the up- turned edges of the leaves and not upon the broad surfaces of the blades.* The latter, indeed, escape altogether the meridian sun, though they get the full benefit of his less scorching rays at the beginning and close of the day. The interesting Compass-plant (Silphium laciniatum) should not be forgotten when speaking of the position assumed by leaves in refer- ence to transpiration. Longfellow's Evangdine contains a graceful de- scription of the plant (though he confuses flower and leaf), which * This is also a protection from the in- jurious effect on the chlorophyll corpuscles of the intense sunlight. FIG. 355. — BLACK MULLEIN (Verbascum nigrum). The leaves are downy with a dense growth of stellate hairs. The yellow flowers are closely packed in long racemes which are a foot and a half long. EUHOPR, SIBERIA. 291 292 HUTCHINSON'S POPULAR BOTANY FIG. 356. — SENSITIVE PLANT (Mimosa pudica). Day position of leaf and leaflets. specimens of this plant not only assume an almost vertical position, but by a singular twisting of their blades, bring themselves into a position which has earned it the name of Compass-plant. The lobes of each pinnate- parted leaf, extending like fingers on either side of the midrib, are said to point due north and south ; but some observers who have watched the plant in its native habitat have thrown considerable doubt upon the state- ment, as they have found the leaves point- ing in all directions. The leaves of the Marram (Psamma aren- aria) exhibit a special structure in view of this same purpose — the prevention of ex- cessive loss of water. The plant in ques- tion is the Common Matweed of our sand- hills, whose spreading fibrous roots are so useful in binding together the shifting sands on the coasts of Norfolk and Holland, and in many other places. It has been noted that " on wet days the leaves open longi- tudinally, so that their inner surface is freely exposed to the air, and the stomata which are situated there may not have their func- is widely distributed over the North Ameri- can prairies : Look at this delicate plant that lifts its head from the meadow, See how its leaves all point to the north, as true as the magnet ; It is the Compass-flower, that the finger of God has suspended Here on its fragile stalk, to direct the traveller's journey Over the sea-like, pathless, limitless waste of the desert. The leaves of young FIG. 357. — SENSITIVE PLANT. At night, or when touched, the leaf hangs down, and the leatlcts fold closely together. THE LEAF IN RELATION TO ITS ENVIRONMENT 293 tions restricted : in dry weather, on the con- trary, the leaves are rolled up so that the leaf almost forms a tube, the outer surface of which is hard and quite impervious to water." The mechanism by which this is effected is to be found in certain cells, which form longitudinal rows at the base of the furrows on the under surface of the leaves, and which are very sensi- tive to moisture. In damp weather the cells increase in turgid ity by absorption of water, and the leaf opens. In concluding these observations on the dangers to plants from excessive heat, and the means provided by Nature to counteract those dangers, one is naturally led to the op- posite side of the sub- ject, and the question arises, If too much heat be injurious to a plant, may not, under contrary circumstances, too great loss of heat be injurious too ? Moreover, if the all-wise Mother has devised means for protecting growing plants from the one evil, may she not also have devised means for protecting them from the other? To both questions an affirmative answer may be given. Loss of heat has no less to be provided against than excessive transpiration — the damp and chilly nights must be taken into account quite as carefully as the dry and sunny days ; for growth goes on in the plant more rapidly in some cases by night than by day, and if it were systematically deprived of heat during the hours of darkness, it would soon languish and die. Now the means which Nature has devised for protecting growing plants from loss of heat FIG. 358. — TELEGRAPH-PLANT (Desmodium gyrans). The small lateral leaflets move un and down, twisting at the same time, and describe a kini of ellipse. The movement i.s not continuous, and is much influenced by temperature. 294 HUTCHINSON'S POPULAR BOTANY are beautifully simple. What is known as the "sleep" of plants — in other words, the nocturnal drooping and folding of leaves and flowers — com- prehends the chief of those means, and we shall here confine our re- marks to this well-known phenome- non. If you have ever sauntered through a garden by night, and examined, lantern in hand, the dew-drenched vegetation, you will have stumbled upon some curious discoveries. It is easy to imagine the surprise evoked during such an excursion. As you pause before one of the well-ordered beds, and look down at the familiar plants, you involuntarily ask yourself, What has become of the flowers ? A few, indeed, are still plainly visi- ble : but there are others that you miss, nor do you realise what has become of them until, on closer ex- amination, you discover that some are closed, and others are hanging down their heads so that only their green collars (the calyces) meet the eye : while others, again, have skil- fully concealed themselves behind their own foliage leaves. The leaves, too, appear to be wonderfully changed. " We are all a-noddin', nid-nid-noddin'," seems to be their drowsy language. The Trapceolums no longer confront the vault of heaven with their green shields, which hang listless at their sides ; the Lupins have folded up their digitate leaves like umbrellas (fig. 352) ; and on every hand the foliage seems heavy with slumber. It may be cruel to unsettle so pleasing a fancy, but the phenomenon described, and which is popularly known as the sleep of plants, is simply a part of Nature's plan for guarding Ph/)to W [E. Slep. FIG. 359 — OAB-WEED (Laminaria digitata), One of the largest of our native seaweeds. The specimens photographed were fifteen feet long. FIG. 360. — HAIRS FROM THE LEAVES OF VARIOUS PLANTS (MAGNIFIED). (a) Glandular hairs of Snapdragon ; (6) simple unicellular hairs of a Brassica ; (c) glandular tentacle of ^"n|jp^' > (rf) sessile L-land of Hop ; (?) tufted hairs of Cretan Horehound ; (0 branched hairs of Nicaiulni ; (;/) ..M-/.\ coverm.,' of'Rochea : (h) stellate hair of Alpine Mr.dwort. ; (O forked hairs of Whitlow Grass ; (/) munform Hairs of Marvel of Peru ; (fc) muriform hairs of Virginian Spiderwort. ^- 295 296 HUTCHINSON'S POPULAR BOTANY FIG. 361. — STINGING HAIR OF STINGING NETTLE. her vegetable proteges against excessive loss of heat. The " sleep " position of leaves is, in fact, a protective arrangement. By folding themselves together and as- suming, as far as possible, a vertical position, radiation is materially checked, and thus the plants undergo no serious fall of temperature daring the night. We must not be misled s by this popular term "sleep" into sap- posing that the nutritive processes of the plant are suspended at this time. " The drooping position as- sumed by the leaflets of Oxalis is simply protective ; there is no correlation between the assumption of the drooping position and the temporary loss of the power of assimila- tion. Preparations made at night from the leaves of Oxalis when in the drooping nyctitropic position show a normally active power of assimilation, and the same is the case with leaflets of Mimosa. The movements per- formed in assuming the nocturnal nyctitropic position of certain ' sleep ' plants are not accompanied by any corresponding internal changes or alterations in the power of assimilation. In this respect the sleep of plants is more external and apparent than internal and real '' (A. J. Ewart, B.Sc., in Journ. Linn. Soc. [Botany], vol. xxxi., 1896). Plants of the great Leguminous order, to which the Acacias, Mimosas,. Peas, and Trefoils belong, exhibit the phenomenon of which we are treating in a very striking manner. The Wood-sorrels (Oxalis acetosella, corniculata, and strida), also, are extremely sensitive to changes of temperature, folding down their leaves even in the daytime if rain threatens (figs. 353, 354), while a blow from a stick will cause them to shrink together with affecting suddenness. Sensitiveness is carried to an extreme in a tropical species of this genus, Oxalis sensitiva, concerning which it is affirmed that even the disturbances of the air caused by the approach of man are sufficient to- induce the phenomenon, the petioles relaxing and the pinnate leaflets falling together like the leaves of a book. This is also said to be the case with several of the Mimosas ; but the two species which are most common in English stove-houses (M. pudica, figs. 356, 357, and M. sensitiva), though collapsing readily at the slightest touch, certainly do not exhibit such FIG. 362.-HEDGEHOG VARIETY OF HOLLY. extreme sensibility in this THE LEAF IN RELATION TO ITS ENVIRONMENT 297 country. Touch a leaf-point of Mimosa, and the small leaflets fold together, and the stalk to which they are attached drops suddenly. The leaflets on other branches of the compound leaf act in the same way ; and finally the main leaf-stalk drops suddenly. The Mimosas have received poetic treatment from more than one distinguished writer. Erasmus Darwin says : Weak with nice sense the chaste Mimosa stands, From each rude touch with- draws her timid hands ; Oft as light clouds o'erpass the summer glade, Alarmed she trembles at the moving shade ; And feels, alive through all her tender form, The whispered murmurs of the gathering storm ; Shuts her sweet eyelids to approaching night, And hails with freshened charms the rising light. The movements of so-called sensitive plants are probably in part due to a peculiar modifica- tion of certain of their leaf-cells, which — in the Mimosas, at least — are so constructed that delicate threads of pro- toplasm pass through their walls and maintain a connection with the living matter of adjoin- ing cells. Thus the effects of a touch on one photo W t£- S:eP- part Of a leaf may be FlG' 363.— TEASEL (Dipsacus sylvestris). ... -, ,-. . Showing the protective spines on the stem, and the spiny bracts inter- transmitted all over it ; ppersed with the f,0wers in the flower-head. and if, as is not un- likely, these perforated cells are distributed through the stem and branches as well as the leaves, the effects spoken of may be carried to every part of the plant. The drooping of the leaf-stalk (which, as in the leaf of Mimosa pudica, may alter its angle with the stein from ninety to thirty degrees) is caused by a beautiful piece of mechanism. At the base of the leaf-stalk there is 298 HUTCHINSON'S POPULAR BOTANY a little cushion-like swelling, called the pulviniis (the Latin word for "cushion"), which contains a woody centre surrounded by parenchymatous cells, rich in water. When one of the pinnate leaflets is touched, the effect is transmitted to the pulvinus by the threads of protoplasm, with the result that the water passes from the cells on the lower to those on the upper side, causing the former to pass from a distended into a flabby state. They thus become temporarily unfit to support the leaf-stalk, which in Photo by] FIG. 364. — HOLLY (Ilex aquifolium). Showing the spiny, dentate leaves and the clusters of red berri [E. Step. consequence falls of its own weight. By and by the water gains its original distribution, and then the leaf-stalk resumes its horizontal position. The celebrated Telegraph-plant (Desmodium gyrans) is even more interesting than the Mimosas. It is an East Indian plant with violet flowers and trifoliate leaves, and the latter are in motion night and day. In bright sunshine the two lateral leaflets jerk up and down, and from side to side, in a remarkable manner, while the large terminal leaflet goes through similar though less perceptible movements (fig. 358). Should these movements be artificially checked for a while, the leaf will start again with increased city directly the retarding influence is removed. During darkness the Photo by} \Jlcnry Ir FIG. 365. — SPEAK PLUME-THISTLE (Cnicus lanceolatus). All parts of tl.e plant above ground are armed with needle-like spines. The florets are specially adapted for the _^r visits of bees with long probosces, particularly humble-bees. 300 HUTCHINSON'S POPULAR BOTANY terminal leaflet assumes the perpendicular position which has been already shown to be characteristic of " sleeping " leaves. The trembling movement of the leaves of the Aspen (Populus tremula) has supplied many figurative allusions to prose-writers and poets, and the phenomenon deserves a passing notice. The quivering is due to the elasticity of the long flattened foot-stalks ; and Mr. Colbourn, of Hobart, suggests that the rapid movement in the air enables the leaf to throw off the excess of moisture which collects on it in the damp situations of the tree. Some force is given to this view if we look at the Aspen or the Black Poplar immediately after rain, when we shall find great numbers of the leaves held together bv moisture. Kerner, however, regards the motion as an arrangement for protecting the flat broad leaves against crushing ; but many other broad flat leaves are without this provision. He further remarks that the elasticity is due to the development of bast-strands in the leaf-stalks. We have now considered a few of the dan- gers to which the green leaves of plants are exposed, but the subject would be very im- perfectly treated were no mention made of a danger of another kind. This form of danger belongs to the animate rather than the inani- mate world — to " wild beasts and beasts of the field and creeping things'' rather than to heat and cold and other such phenomena and forces. Innumerable animals feed upon the green tissues of plants, and find in a vegetarian diet their only sustenance; indeed, if Nature had not provided special contrivances to keep off these devourers, it is next to certain that whole families of plants would long since have van- ished from the face of the earth. A few of these contrivances have been incidentally re- ferred to in former chapters. When speaking of the sap of plants, we showed that the milky juice of the Common Lettuce (Lactuca sativa) pro- tected the plant from the depredations of ants and other leaf-eating insects; and on a later occasion we saw that the thorns or spines in such plants as Blackthorn (/Yunua spinosa). Spiny Restharrow (Ononis spinosa), bpurges (Euphorbia), etc., render acceptable service by keeping off brows- ing cattle and herbivorous wild animals. But the subject was only lio-htlv ouched, and-from the nature of the connection in which it was intro- uced-many of these protective contrivances were not alluded to at all. ^or example no mention was made of pricttes. Prickles are another thorn. They are not, like spines, branches which have degenerated, FIG. 366.— A PRICKLY PEAR CACTUS (Opuntia multiflora), Protected by fine barbed bristles. THE LEAF IN RELATION TO ITS ENVIRONMENT 301 I P/fo/o &«/] FIG. 367. — SPANGLE-GALLS ON OAK. for they spring from the epider- mis or cortex of a plant member, and contain no fi br o -vascular bundles ; while spines, it will be remembered, are traversed by those bundles which connect them with the vascular system of the stem arid root. Good ex- amples of prickles are offered by the Dog-rose (Rosa canina) and the Holly (Ilex), to name no other plants. In the former they occur on the petioles and branches ; in the latter they spring from the margins of the leaves. Plants of a prickly nature seldom develop those structures on leaves and branches which are out of reach of grazing animals. The Common Holly (Ilex aquifolium), the bristly den- tate leaves of which form a characteri s t i c feature of shrubby speci- mens of the plant, produces only unarmed leaves, with en- tire margins, on its upper branches, when it attains to t h e dignity and di- mensions of a tree. Indeed, the gradations from FIG. 368. — CHERRY-GALL ON OAK. Photo by} [/. Ilolmns. 302 HUTCHINSON'S POPULAR BOTANY m the prickly to the unarmed forms of leaves in the plant named are so numerous and marked, that the species has been separated into varieties to distinguish them. Compare, for example, a leaf of Henderson's Holly (Ilex a. hendersoni) with one of the Donington variety (Ilex a. donington- ensis), and both with a leaf of the Hedgehog Holly (Ilex a. ferox, fig. 36*2) ; which latter has prickles not only at the margin, but also on the upper surface. Its popular name is, indeed, exceedingly appropriate ; so, too, is its Latin appellative — ferox, " savage." Several plants of the large genus Solarium, to which the Potato-plant and Woody Nightshade belong (e.g. S. fontanesianum, jac- quini, and maroniense), have spiny erections on both sides of the leaf. They are borne upon the midrib and veins, and make the plants ex- tremely awkward things to handle. Many other protestive arrangements more or less similar to those described occur readily to the mind ; as, for instance, the sharp, strong, needle-shaped (acicu- lar) leaves of many Grasses ; the formidable thorny ter- minations of the leaves of the Agaves, and the spine- bordered lobes of leaves like the Thistle (Carduus), Teasel (Dipsacus), and Acanthus. It has been asserted that in the- Southern Alps sheep will frequently return from pasture with their nostrils cut and bleeding, and the shepherds know at once that the cause of the mischief is a species f stiff-leaved Grass, Festuca alpestris, which they seek to destroy by burn- ing. In some instances, Grasses which cause discomfort to grazing animals will be dealt with by the animals themselves, who seize them low down with their teeth and tear them from the ground. Kerner saw thousands tufts of the Mat-grass (Kardus stricta), which had been rooted up by oxen lying dried and bleached by the sun on some meadows in the Tyrolese Stubaithal. FIG. 369. — BRISTLE-GALL ON OAK. These galls are caused by the gall-wasp. Andricus lucidus. Photo by] FIG. 370. — STINGING NETTLE (Urtica dioica). Theleaves and stem are covered with hollow stinging hairs through which a poison is introduced to the victim's flesh. The small green flowers are wind-pollinated. NORTH TEMPERATE REGIONS, N. AFRICA, THE ANDES. 303 304 HUTCHINSON'S POPULAR BOTANY More formidable than air^ of the protective weapons yet mentioned are the barbed bristles which surround the buds on many of the Opuntias. Each of these is a sort of compound fish-hook in miniature, and woe to the unwarv animal who pushes his nose against the smallest bunch of them! The little hooks enter the tender flesh, and cause intolerable itching. which is often succeeded by painful and. it may be, dangerous inflammation. The tormenting bristles are easily driven deeper, but the backward-pointing barbs tear the flesh unmercifullv when any attempt is made to withdraw them. One species, Opuntia ficus-indica, better known as the Prickly Pear, is abundantly naturalized in the Mediterranean area, where it forms inpenetrable fences. Another species. Opuntia feroXj is said by Schleiden to be especially remarkable because of the strength and size of its defen- sive thorns. " Among the hairs and smaller spines," he says, " arise very long and thick spines, in different form and number, which give the best characters for the determination of the species. In some, these are so hard and strong that they even lame the wild asses which in- cautiously wound themselves when kicking off the spines to reach the means to still their thirst. In Opuntia tuna, which is the kind most frequently used for hedges, FIG. 371. — FLOWERING BRANCH OF A ROSE SUR- ROUNDED BY A BEDEGUAR GALL. The gall-fly and grub of same are shown in the corner of drawing. they are so large that even the buffaloes are killed by the inflam- mation following from these spines running into their breasts. It was this species, also, which was planted in a triple row as a boundary line between the English and French in the island of St. Christopher." Fig. 366 represents a flowering branch of Opuntia multiflora. No account of the protective armature of green leaves would be complete without a reference to stinging hairs. The mention of these very remarkable structures brings to mind the Common Nettle (Urtica dioica}— a weed that is known to every child. Let us take a peep through the microscope at one of its stinging hairs, and try to realise what takes place when any rash THE LEAF IN EELATION TO ITS ENVIRONMENT 305 meddler with the plant gets stung. Our picture (fig. 361) represents a hair in section. It consists of a long, tapering single cell, rising from a cushion- like base, and widening at the apex into a little knob, which is bent some- what out of the perpendicular. At the point where this bend takes place the cell-wall is extremely thin — so thin that a very slight touch suffices to break off the knob. When, therefore, such a touch is given, the mischief is done, and the acrid irritating fluid contained in the cell escapes at the point of rupture and enters the tiny wound which the vitreous apex of the hair has made. The fluid consists of formic acid and a sort of unorganized ferment or enzyme, the latter being thought to be the more poisonous property of the two. It may be added that the break takes place obliquely (a conse- quence of the bend above described), so that the broken end resembles the poison-fang of a serpent — to which, indeed, it has sometimes been com- pared. A brush from the leaf of any of the British Nettles ( Urtica dioica, urens, and pilulifera) is doubtless a light matter, but to be stung by some Photo by} IE. Step. FIG. 372. — HAIRY GALLS ON BEECH, Caused by a two-winged fly (Hormomijia). of the Asiatic species is a very different thing. The great Shrubby Nettle (Urtica crenulata) of Northern India, for instance, is a nettle to beware of. " This plant," says Sir Joseph Hooker in his Himalayan Journals, ': called ' mealum-ma,' attains fifteen feet in height ; it has broad glossy leaves, and though apparently without stings, is held in so great dread that I had difficulty in getting help to cut it down. I gathered many specimens without allowing any part to touch my skin ; still, the scentless effluvium was so powerful that mucous matter poured from my eyes and nose all the rest of the afternoon in ii— 2 ' 306 HUTCHINSON'S POPULAR BOTANY such abundance that I had to hold my head over a basin for an hour. The sting is very virulent, producing inflammation ; and to punish a child with 'mealum-ma' is the severest Lepcha threat." The writer explains in a footnote that the hairs are microscopically small, and they only sting violently during the autumn. M. Leschenault, a French botanist who had the misfortune to be stung by the " mealum-ma '' at this particular season, while gathering one of the leaves for his herbarium, describes the symptoms that followed as anything but pleasant. At first he felt only a slight pricking which he wholly disregarded ; but the pain gradually increased, and at the end of an hour it had become excruciating. The parts affected — the first three fingers of his left hand— felfc as though they were being rubbed with a hot iron. Before long the pain had spread up the arm to the arm-pit : and within five hours of being stung the torture was increased tenfold by an ominous contraction of the muscles of the jaw, which made him fear an attack of lockjaw. However, the latter symp- toms passed away towards evening, and from, that time the pain continued to decrease, though upwards of nine days elapsed before it ^^H^^^^K^ entirely left him. ^W\ That the inferior animals are sensitive to "•^^ the stings of plants no less than man, and therefore that stinging hairs may be a real protection from grazing animals, is illustrated by the fact mentioned by Baillon, that the natives of Java rub buffaloes with a species of Nettle (Urtica stimulans) in order to excite them to fight with tigers. On the other hand, these vegetable fangs are innocuous to certain leaf-eating insects, which feed upon them with impunity ; indeed, it is well known that the leaves of our British Nettles, which are all furnished with stinging hairs, form the only food of the caterpillars of three of our most beautiful butterflies— namely, Vanessa atalanta, V. io: and V. urticce. But this fact affords us a very striking object-lesson on the way in which an offensive or merely defensive development in one organism may lead to the very con- siderable adaptation in some other organism that may ba seriously affected by it. The three caterpillars named have developed protecting spines which keep the stinging hairs of the Nettle from contact with their tender skins. Their relative, Pymmeis cardui, which feeds on Thistles, is similarly protected. It is remarkable that, so far as we have read, botanical writers have failed to note that hairs similar to those of the Nettle, but in a far FIG. 373. — NAIL-GALLS On leaf of Lime, produced by a r Photo by] 1E- FIG. 374. — Primula clccnica. This bai'atiful plint is coverei with easily detachable hairs. If these, by incautious handling, come in contact witli the hum in skin, they set up an inaimmitioa very similar to eczemi, for which it has been often mis -^ 307 308 HUTCHINSON'S POPULAR BOTANY less highly developed condi- tion, are found on the leaves and young twigs of the Com- mon Elm (Ulmus campestris}. They are abundant along the •j^t —pr • ^^^fr: midrib on the lower surface, " f^\ / - £' and though they have nothing M' I . "^^ 1|| like the malignity of the Nettle, ^rfjpy *'". ^i^fMi ^ej canse a considerable ^•8= LV^. 1 ^ amount of irritation to the hands and wrists of those who touch the leaves. The Elm sends up numerous suckers, and all down the bole it t}irows out new shoots, which would be probably browsed off but for the presence of these hairs. The Elm belongs to the same natural order (Urticacese) as the Nettle. When investigation has been carried farther it will be shown probably that the hair- structures of many other plants have a protective purpose. Many species of Mullein (Ver- bascum), for example, have branched radiating hairs which rub off easily when the plant is handled, and, though not stiff or prickly, "remain hanging to the smallest in- equalities on the surface of the disturbing body. If grazing animals bring the mucous membrane of their mouths into contact with the leaves of the Mullein," the Hock-like masses of hair adhere to the tongue and palate and produce sensations that can hardly be pleasant (Kerner). The hairs of several species of Primula, also (notably P. obconica), set up an inflammation very like eczema when handled incautiously. It may not be generally known that the singular growths called galls, so often to be seen on the leaves and branches of Flowering Plants, are due to insects. Formerly they were held to be entirely of a vegetable nature, and the insects found in them were thought to have been spon- taneously generated there. Many species of Cynips lay their eggs in the parts named, plunging their exceedingly delicate ovipositors into the soft tissues, and thereby set up local irritation, which induces a responsive action of the protoplasm, and galls are produced. In some cases, how- ever, the development of the gall does not commence until after the egg is Photo by} FIG. 375. — SPIKED PEA-GALL, On Dog-rose, caused by the gall-wasp, Rhodite, THE LEAF IN RELATION TO ITS ENVIRONMENT 309 hatched. No tree, probably, furnishes so great a variety of these growths as the Oak (Quercus), on which upwards of one hundred and fifty species have been observed, the well-known oak-apple being one of them. It is pro- duced by the punctures of Dryoteras tenninalis. Two species of Oak-gall often to be met with are produced by an insect named Spathegaster bac- carum. Fig. 373 represents a leaf of a species of Lime (TUia platyphyllos) with little conical excrescences or nail-galls, the work of a microscopic species of Phytoptus (P. tilce), whose portrait you will notice just above the leaf. Similar galls, but of a downy nature, occur on the leaves of Beech (Fagus sylvatica], in this case caused by a two-winged fly (Hormomyia piliger). The green (ultimately red), mossy-looking growths called bede- guars, or Robin Redbreast's Pincushion, so common on branches of the Rose, are also of insect origin (fig. 371). The gall-fly (Rhodites rosea) deposits its eggs in the shoot-bud, which presently swells and begins to put out what should be in the natural course three leaves : but embryo leaves have so little parenchyma between their veins that they fall into threads and thus give the mossy appearance to the galls. On cutting open one of the bedeguars the larvae of the insect will be found in the centre. In all these cases it appears that only the cells of the meristem can give rise to the galls. These growths do not appear to be actually injurious. The galls that are found on Hedge Bed- straw (Galium mollugo) are produced by a minute two-winged fly (Cecidomyia aparine) ; and the many-cham- bered gall on young shoots of Spruce (Picea excelsa), that look like half a pine cone, are the work of a Coccus (Kermes). The "Witches Broom " on the Scots Pine (Pimis sylvestris) is an excrescence consist- ing of a multitude of short shoots produced by a fungus (Perider- mium elatinum). What is known as fasciation, or the fusing together of parts of a plant which are norm- Photo by] \-E- FIG. 376.— "WITCHES BROOM " ON SCOTS PINE, Caused by the fungus Peridermium elatinum. 310 HUTCHINSON'S POPULAR BOTANY ally distinct, as in the triple flower of Fuchsia shown in fig. 377, is likewise (at least, in a large number of instances) due to gall-mites. Many of the singular metamorphoses of plant organs, too — floral leaves which are changed into foliage leaves, petals which become stamens, etc., etc. — are probably attributable to the same exciting cause — a subject of which we shall have more to say when speaking of the Flower. In some instances galls have a positive economic value, though this, of course, is of no advantage to the plant. The galls of commerce are chiefly those which occur on Quercus infectoria, and the best of them come from Aleppo and Smyrna. They yield a fine black colour with any of the salts of iron, and are largely used in the manufacture of writing- ink. Perhaps the most dreaded of gall- producing insects is the grape-louse (Phyl- loxera vastatrix), which pierces with its proboscis the young leaves and roots of the European Vine ( Vitis vinifera), and thereby causes the growths referred to. The galls, by driving away nourishment from the roots (in which they are ably assisted by the insects themselves), starve and weaken those delicate organs, and at last destroy the plant. We have now touched upon all the more important facts connected with the forms, structure, and functions of foliage leaves, and with the means provided by Nature for their protection and preserva- tion. A few remarks on the decay and fall of the leaf may fitly conclude the subject. One of the first external signs of incipient decay in green leaves is the fading of their freshness. The green becomes dull, and gradually assumes a yellow, brown, or ruddy tinge, due to varying degrees of oxida- tion of the chlorophyll, contained in the cells. The fall of the leaf is not primarily nor necessarily due to external forces. Wind and frost may, and do, perform their part, but long before the leaf has attained its full growth and vigour the busy protoplasts have been weaving a layer of cells, which shall infallibly ensure the work of disarticulation. These cells, botanically known as the layer of separation, are formed in the base of the leaf-stalk, and run at right angles to the older and displaced cells, so that they divide the leaf from its branch or stem. FIG. 377.— FUCHSIA. Monster flower caused by fasciation. Photo by} IHenry Irving. FIG. 378. — " OAK-APPLE " (GALLS OF Dryoteras terminalis). About cwo-thirds of the natural size. They are found in May and June on the twigs of the Oak, ar numerous larvae are embedded in the soft tissue of the green- and red-skinned " apple." 311 CHAPTER XI FLORAL FORMS AND THEIR RELATIONS TO INSECTS To me the meanest flower that blows does bring Thoughts that do often lie too deep for tears. WORDSWORTH. AS there is a good deal of ground to be covered iu this chapter, we will not waste time on the threshold. Our subject is the Flower — in many respects the most important, as certainly it is the most interesting, of the subjects of which Botany treats. The Flower contains the organs of reproduction, a fact which accounts for its pre-eminent importance : while the manner in which those organs discharge their appointed functions — assisted often by the most unlikely agencies, as water, wind, and insects — gives to the study of the Flower an attractiveness all its own. Root, stem, and leaf are not with- out their fascinations, too, as we have sought to show in earlier chapters, but the Flower is the part of the plant which rightfully commands the lion's share of interest. As with the leaf, the beginning of the Flower is the bud. Flower-buds originate in much the same way as leaf-buds, and cannot be distinguished from the latter in their earlier stage. Like leaf-buds, too, they are formed either in the axils or at the ends of branches, and in accordance with those conditions are named respec- tively axillary and terminal. The reader will probably recognize the little flower shown in fig. 379. It is the Moneywort or Creeping Loosestrife (Lysi- machia nummulama), an English wild- flower partial to ruins and damp woods, and a favourite rockery plant under culti- vation. It, and its near relation, the Wood- 312 FIG. 379. — CREEPING LOOSESTRIFE. An example of solitary and axillary flowers. FLORAL FORMS AND THEIR RELATIONS TO INSECTS 313 land Loosestrife or Yellow Pimpernel, offer familiar examples of a solitary and axillary flower. The Herb-paris (P. quadrifolia, fig. 381) bears solitary flowers, too, but they are ter- minal, not axillary. The Herb-paris is one of the most singular of our wild-flowers, and, like the Loosestrifes, delights in moist and shady woods. But solitary flowers, whether axillary or ter- minal, are the exception rather than the rule. In by far the greater number of plants the bud unfoldsinto a branch system, consisting of several flowers, which are known collectively as the inflorescence. The Cowslip (Primula veris), Cherry (Prunus cerasus), and Forget-me-not (Myo- sotis palustris) may serve as examples. What is popularly known as the "flower" of the Daisy (Bellis perennis) and Dandelion (Taraxacum officinale) is likewise an inflorescence ; each of the so-called flowers being really a multitude of minute flowers (florets) crowded together on a single stem. We will consider the structure of one of these composite flowers later on. In describing a flower the presence or absence of a stalk should always be noted. Stalkless or sessile flowers are comparatively rare, but the flower- stems of stalked or pedicellate flowers may be so short as to be hardly perceptible. No more remarkable instance of a sessile flower could be named than that vegetable wonder, Rafilesia arnoldi, of which some account Photo by] [E. Step. FIG. 380. — CUCKOO-PINT (Arum maculatum). The front part of the spathe has been cut away to show the minute flowers around the base of the spadix. 314 HUTCHINSON'S POPULAR BOTANY was given in a former chapter ; and if we place side by side with Rafflesia the stalked inflorescence of the celebrated Lilium giganteum, we have a contrast indeed. A flower-stem of one of these Lilies, cut from a living plant in the Sunningdale Nursery in July, 1879, is preserved in No. 1 Museum at Kew. The circumference of the pedicel is eleven and a half inches, its height thirteen feet ! Truly a Brobdingriagian flower-stem. A stalk which supports a solitary flower, or the primary stalk of an inflorescence, is called a peduncle; while the branches or secondary stalks are known as pedicels. The stalk of Herb-paris, one of the flowers that we were looking at a moment ago, is an example of a peduncle ; so is the primary stalk of the Lily of the Valley (Convallaria majcdis), while its slender branches, curving with the weight of the dainty little bells, are pedicels. The portion of the floral stem (peduncle) which, in this plant, bears the stalked flowers, is the rachis. Rachis is the Greek word for " spine," and besides being the term used in anatomical science for the vertebral column of animals, is used for many things which suggest a resemblance to the spine, as the shaft of a feather, the stalk of the frond in Ferns, the axis of a compound leaf, and (as we have just seen) the axis of an inflorescence. The Lily of the Valley is an ex- tremely useful plant for an object- lesson. Who is not familiar with its pensile beauty? It is a favourite FlG 381 HERB-PARIS flower under cultivation, and one of Solitary and terminal flower. the HlOSt SOUgllt after of wild-flowers. The plant needs some seeking, too, for it loves to hide from sight in shady glens, covering its nodding white bells with its large and glossy leaves. You perceive that the leafless peduncle springs directly from a sub- terranean stem or root-stock — a rhizome, to use the botanical term. On this account the inflorescence is called a scape, and the plant itself a scapigerous herb. The Primrose, Cowslip, and Oxlip (Primula vulgaris, P. veris, and P. elatior) also offer familiar examples of scapigerous herbs. Notice, further, that the pedicels of the Lily of the Valley spring from the axils of what appear to be minute leaves — a fact of importance, as similar leaf-like forms are found on most branched inflorescences. They are called bracts, from the Latin bractea, a thin plate of metal. Bracts are usually green, but in certain plants — as the Flowering Dogwood (Gornus florida) and the celebrated Edelweiss (Leontopodium alpinum) — they are white or coloured like the petals of flowers, and are then called petaloid. In Poinsettia FIG. 382. — COWSLIP (Primula veris). A. familiar plant whose nodding yellow flowers are borne in an umbel on a tall scape, hence it is described a scapigerous herb. EUROPE, w. ASIA, N. AFRICA. 315 316 HUTCHINSON'S POPULAR BOTANY they form the most showy portion of the inflorescence, theflowers themselves being- small and inconspicuous. The names bracteate and ebracteate are applied to flowers according as they possess or do not possess these modified leaves. Here is a flower that will be readily recognized (fig. 380). No lover of green lanes and sunny meadows bordering banks can be a stranger to the Cuckoo-pint or Wake Robin (Arum maculatum), with its " ear-like spindling flowers,'' as Clare calls them, " betinged with yellowish, white, or purplish hue." But let us be clear as to what we are looking at. The figure does not represent a single flower, but an inflorescence, the numerous flowers of which are congregated round the narrow lower part of the club- like column or spa.*.)• (The keel, wings, and standard have been removed.) of what has been before us in this rapid sketch. The process that has been described is the same, in all essential particulars, as that which goes on in the great majority of flowering plants, and it is so simple as to be easily understood. The pollen may be conveyed to the stigma by other than insect agency, and the tubes may pursue a more winding course through the conducting tissue ; but these are details, and the above descrip- tion may be accepted as a fairly representative one. What actu- ally takes place in the ovules when the pollen-tubes have found an entrance will be explained farther on (Chap. XIV.) : it is sufficient here to have directed attention to the fact itself. The growth of the tubes may be conveniently demon- strated with the pollen of the cultivated varieties of Caladium, all that is needed being to leave a few grains on a damp microscope slide for five or six hours. We have already seen that the beginning of the flower is, like the beginning of the leaf, the bud ; and that a flower-bud and a foliage-bud are indistinguishable in the early stages of growth. The theory of the develop- ment of all parts of the flower from leaves was enunciated by the poet Goethe nearly a century ago : and though the announcement of his discovery was accompanied by a good deal of speculation which subsequent research has shown to be erroneous, his main position had much to recommend it. nor does it lack defenders even at the present day. Goethe taught that " the elementary floret expands into a leaf upon the stem, contracts to make the calyx, expands again to make the petal, to contract again FIG. 394. — ABNORMAL FLOWER OF Begonia, into sexual Organs, and expand for With sepals developed a? green foliage leaves. the last time illtO fruit." FLORAL FORMS AND THEIR RELATIONS TO INSECTS 325 On the other hand, there are physiologists of the first rank who hold the theory of the German poet in light esteem, and to whom the pronounce- ment that " every flower is simply a metamorphosed leaf-shoot " is a dangerous expression, implying that the flower has been developed in course of evolution from a leaf-shoot, for which, in their judgment, there is not sufficient evidence. We are disposed to concur in this view, and rather than yield allegiance too readily to G-oethe's theory would say Photo by} [E. Step. FIG. 395. — LAKGE-FLOWERED ST. JOHN'S WORT (Hypericum calycinum). Rose of Sharon. The lar re flov form :rs offer good examples of spiral activation. The profuse stamens number of little bundles. that there are flower-shoots and leaf-shoots, without attempting to derive one from the other. Yet the tendency of the floral organs to relapse into the foliar form in certain abnormally developed flowers, at least confirms the idea that floral leaves and foliage leaves are homologous structures. Most of us, doubtless, have met with flowers of the kind referred to — " monstrous " flowers, as they are called. In Science Gossip (1890) there is an interesting series of papers on the subject, by Dr. J. E. Taylor, with drawings of some of the more remarkable monstrosities. In one place we find an abnormal Knapweed (Gentaurea nigra}, of which some of the florets have become leaf- like; in another, a Daisy (Bell is perennis) has 326 HUTCHINSON'S POPULAR BOTANY developed a true foliage leaf instead of a bract in the involucre. A third specimen (a flower of the Enchanters Nightshade— Circcea lutetiana] appears with a portion of its stigma transformed into the anther of a stamen, and a stamen assuming the character of a pstal, while, in place of another of its petals, two sepals are developed! Not less remarkable is the figure of a Peach-flower, whose organs exhibit a steady gradation from petals to foliage leaves : and farther on in the volume we meet with an abnormal Rose, entirely devoid of petals, and with sepals which seem to have been trying ard to produce a serrated margin. The stamens of this flower are normally developed, but the pistil (if pistil it may be called) is a curiosity, the style being elongated into a green and healthy-looking shoot. bearing some two or three dozen ordin- ary leaves ! The " monster " flowers which have come under our own notice are not so singular as those figured in Dr. Taylor's remarkable articles, but they serve no less to illustrate Goethe's law. The first figure (fig. 393) repre- sents a flower of Wistaria, from which the coloured petals have been removed. In this example the abnormally de- veloped organs are the stamens, one of which has a small leaf-shaped purple petal growing out from the centre of the style ; while another has developed a similar petaloid organ in place of the anther-lobes. In the second figure (fig. 394) we have a "monstrous" flower of Begonia (B. octavia], whose outer petals — or rather sepals— have been metamor- phosed into green leaves, with midrib, veins, etc. In many species of the genus Clematis, the sepals are petaloid — that is to say, they are coloured like true petals, while the petals are absent. The former are spoken of collectively as the perianth, this being the name applied to the floral envelopes of a flower when calyx and corolla are not easily distinguished. It would be well, perhaps, to keep exclusively to this use of the term, rather than apply it in a loose way to the floral envelopes of any and every flower. In the figure of Clematis ccerulea (fig. 398), the six petal-like organs (really sepals) constitute the perianth. Probably no flower better illustrates the truth we are considering than FIG. 396. — SNAPDRAGON. A portion of the stigma and style, showing pollen- grains on the former putting forth tubes and pene- trating the style. FIG. 397. — DEFINITE INFLORESCENCES. (a) Panicled cyme of Chinese Privet (Ligustrum luddum). (b) Corymbose cyme of Di _ (c) Fascicle of Sweet-william (Dianthus barbatus). (d) Scorpioid cyme of Forget-me-not (My „„,-,,.,„,„-,) © Glomerulus of Box (Buxus sempermrem). (/) Verticillaster of Yellow Archangel (Laminm galeo 327 328 HUTCHINSON'S POPULAR BOTANY the "White Water-lily (Nymphona alba\ in which the gradual stages of transformation from the green sepals of the calyx to the yellow pollen- producing stamens may be seen to great advantage. We are not speaking now of '; monstrous " specimens of the flower (nor were the Begonia and Clematis last alluded to at all abnormal), but of the ordinary White Water- lily ; which thus shows, in a permanent fashion, the community which exists between the various members of the flower. For the sepals merge into petals, and the petals into stamens, by such imperceptible gradations that at certain points it is difficult to say to what set of organs particular parts belong. In drawings made to illustrate the fact, it is easy enough to see that one figure with its dark green colouring is a sepal, and that a second, though of a paler green, is probably a sepal too; but what of the next figure ? This is neither a decided green nor a pure white, but a cross between the two, and it might be called in- differently a sepal or a petal. In the next row we have petals beyond a doubt ; but we are again at a loss when we come to another row. Do these organs repre- sent petals or stamens ? They are broad and white like the former row, but the thickening at their apex is of a 3Tellow colour, and has all the appearance of rudimentary anther- lobes. The figure beside it is equally perplexing, and not till we get far in do we find a stamen pure and simple, with normally developed style and anther and abundance of pollen. The transition is far more gradual than the description might lead one to suppose. In the flower itself a large number of petaloid sepals and stameniferous petals have place between the organs named; and each differs in some slight degree from its neighbour. Here, then, you have an abiding witness to the facts of which we have been treating — a constantly accessible illustration of that homology of structure which seems to exist between the members of every flower. We are now in a position to carry our inquiry a step farther. A flower being only a modified shoot, it is not surprising that the Fio. 398. — FLOWER OF Clematis ccerulea, With petaloid sepals. FLORAL FORMS AND THEIR RELATIONS TO INSECTS 329 / FIG. 399. — ESTIVATION OF FLOWERS. (a) Valvate. (6) Imbricate, (r) Contorted or twisted. ( uguiate corona. FIG. 400.— FLORET OF DANDELION. (a) Cohering anthers, (ft) Extremity of style. 330 HUTCHINSON'S POPULAR BOTANY We approach more interesting ground when we begin to speak of the expanded flower. The question of form and arrangement again meets us on the threshold, but the study of the structural variations of flowers by no means confines one to the acquisition of name -lists or to tedious and meagre definitions. The question is one of scientific importance, involving as a preliminary step the arrangement of multitudes of appearances under primary points of view, and their classification according to rule and exception ; by which means alone a discovery of the actual laws of Nature is rendered possible. Now, manifold as are the structural arrangements of flowers, the variations really concern only a few simple subjects, and when these are grasped the investigation of the causes which produce the differences in the whole floral world is by no means a hopeless under- taking (Schleiden). The subjects of variation are, in fact, four; and the first to be mentioned is number. A Lily, for example, has three sepals, three petals, and six (i.e. twice three) stamens ; while its pistil, though looking like a single organ, is really made up of three carpels which have grown together. The Lily belongs to the great class of Monocotyledons, and three may be said to be the characteristic number of that class. In the floral whorls of Dicotyledons, on the other hand, threes are a rarity ; nor can any number be said to be charac- teristic, though fours and fives are very common. A Fuchsia has four coloured sepals, four petals, twice four (i.e. eight) stamens, and four carpels. A flower of Cherry, on the other hand, has its two outer whorls, the calyx and corolla, in fives, and its stamens form a multiple of five — namely, twenty. The ovary does not follow the rule — it is solitary ; nor is this case by any means an exceptional one, the parts of the pistil in the majority of Dicotyledons being fewer than in the other whorls. It might easily be shown that whorls of flowers correspond to cycles of leaves, but it would be exceeding our present limits to push our inquiries farther in this direction. The second of the four subjects of variation is cohesion. This term is applied by botanists to the union of like parts in a flower, as of sepal with sepal, petal with petal, stamen with stamen, and so forth. Both the calyx and corolla of a Primrose are good examples of cohesion. Here the five sepals unite in a tube and form what is called a gamosepcdoiis calyx ; the five petals are also fused together, and so the corolla is described as gamopetalous. A calyx whose sepals are not united is said to be polysepalous ; a corolla with separate petals is polypetalous. In the Laburnum (L. vidgare) the fila- FIG. 401. — COMMON PEA. A carpel with seeds removed. FIG. 402. — INDEFINITE INFLORESCENCES. (a) Cone of Larch (Larix europea). (6) Umbel of Cowslip (Primula i-eris). (<-) Strobile of Hop (Humulus lupulus). (rf) Hjrpanthodium of Fi* (Ficus carica). (*) Panicle of a Yuow. (/) Thyrsus of Horse chestnut (.isculus hippotaslanum). (g) Capitulum of Corn Blue-bottle (Centaurea cyanus). (h) Catkin of Hazel (Corylm avdlana). (0 Corymb of Cherry (Cmu) petals, (e) calyx, (r) receptacle. •*.< -», /'* V- FIG. 414. — GARDEN LARKSPUR (Delphinium exaltatum). A beautiful North American plant much grown in gardens, where crossing and selection have produced many varieties. The showiness of the flower is due to the sepals, the petals being reduced to two small nectaries, the dark centre of the flower in the photograph. 339 340 HUTCHINSON'S POPULAR BOTANY FIG. 415. — HONESTY (Lunaria). Section of flower showing saccatt calyx. In contrast to the Wallflower, take a Pansy (Viola"). A Pansy is a good example of an irre:/"f"r flower. By cutting through the centre perpen- dicularly we get two similar halves, but in no other way can the flower be divided symrne trie- ally. Flowers exhibiting this kind of bilateral symmetry are known as monosymmetrical or zyyo- morphic (Greek zeugnumi, I join, and morphe, a shape). Flowers the petals of which are placed like the spokes of a wheel, and which may there- fore be divided vertically into similar halves through two or more planes, are termed actinomor- phic (Greek aktin. a spurious form of aktis. a ray, and morphe]. All regular flowers are actinomor- phic : most irregular flowers are zygomorphic. Some flowers cannot be symmetrically divided in any plane — they are asymmetric. There is a curious fact about the Pans}'. What appears to be the lower- most petal is really the upper, for the flower has been reversed. It has been thought that the purpose of the curve in the flower-stalk is to strengthen the flower and thus to enable it to sustain the weight of insects. When a honey-bee visits a Pansy, it almost always turns round on alighting, and sucks with its head downwards, and the flower is bent down by the insect's weight — a circumstance which indirectly bears out this idea. Notice the streaks of colour on the petals, all leading towards the centre of the flower. They are honey-guides for the insect visitors — pathfinders, as some have named them. There will be more to say about these pathfinders on a future occasion, but allusion is made to them here by way of emphasizing the contrast which the petals of this flower present to those of another common flower, the White Cam- pion i Lychnis vespertina, fig. 416). The pure white corolla of this species of Campion has no guiding marks : in this case they would be of no service, for the flower expands in the evening, and insects are attracted to it by its whiteness. There are five petals, attached by claws to the base of a tubular calyx, and these are the characteristic fea- tures of a caryophyllaceous corolla. The Maiden Pink (Dianthus deltoides). Carnation (D. caryophyl- lus}: and Catchfly (Silene) exhibit the same features. The rosaceous corolla consists of five petals, too, but the}- are without claws : the Buttercup (Ra- nunculus, fig. 419) is a good example. FIG. 416. — WHITE CAMPIOX (Lychnis vespertina), With styles exserted. FLORAL FORMS AND THEIR RELATIONS TO INSECTS 311 FIG. 417. — WINTER CHERRY (Physalis), Showing the accrescent calyx. Notice how the corolla of the White Cam- pion is constricted near its upper end by the toothed or dentate calyx. This is another con- trivance for shutting out unwelcome visitors. Only a thin proboscis, at least three-fifths of an inch long (such, for instance, as that of the Small Elephant Hawk-moth (Chcerocampa porcellus], which often visits and cross-pollinates the flower), could reach to the fleshy part of the ovary where the honey is stored. In fact, both the calyx- tube and the elongated claws of the petals assist in preserving the nectar for those insects whose visits are really serviceable to the plant. In flowers with salver-shaped (kypocrateriform) corollas, like the Primrose and Common Lilac (Syringa vulgaris), the tubes formed by the cohering petals answer much the same purpose as the clawed bases of caryophyllaceous corollas, and also protect the pollen from drops of rain and dew — the latter an important consideration in the case of plants which dwell amid mountain-mists, like the delicate species of Primulacese belonging to the genus Androsace, and many species of Phlox (fig. 421). "If flowers of Aretia glacialis, a plant growing on the moraines of glaciers, are examined after a shower," says Kerner, " it is found that every one has a drop resting upon it which slightly compresses the air in the narrow tube of the corolla, but cannot reach the pollen upon the anthers lower down the tube. A subsequent shake or puff of wind causes the drops to roll off the limb of the corolla, or else they are got rid of by evapora- tion ; in either case, the flower becomes once more accessible to insects." Salver-shaped corollas must be distinguished from rotate or wheel- shaped corollas, such as we get, for instance, in purple-flowered Woody Nightshade (Solanum dulcamara). In the former the tubes are long and narrow, in the latter they are short ; but in both the limb is placed at right angles to the tube. We come now to the tubular corolla. From what has been said about the uses of caryophyl- laceous and salver-shaped corollas, the purpose of this third form may be readily divined. It, also, has reference to insect-pollination. The Com- mon Honeysuckle (Lonicera peryclymenum) may serve as an example. The sweet-scented flowers FIG. 418. — GREAT VALERIAN ( Valeriana officinalis). Showing the gibbous corolla. 342 HUTCHINSON'S POPULAR BOTANY are adapted for fertilization exclusively by long-tongued crepuscular and nocturnal moths ; and why ? Because the wax-like tubes are long and narrow — too long for short-tongued insects to reach the honey while standing at the mouth of the flower, and too narrow to enable them to descend bodily to the nectary. As a matter of fact, the flowers bloom at the very season when hawk-moths are most abundant — that is, during Ma}* and June — and they exhale their perfume most strongly in the evening, when these moths are on the wing. Another form of gamopetalous corolla which must not be passed over is that which, from its resemblance to a funnel, has received the name infundibuliform. The name is less elegant than the flower to which it is applied. The delicate white chalices of our beautiful Hedge Convolvulus (C. sepium) are of this form. Funnel-shaped flowers have not such a reputation for exclusiveness as have those of tubular and salver shape ; and it has been {^ rX observed of the Hedge Convolvulus that PP|jk\ all sorts of thrips and little flies frequent •L, ^ ^ the flowers by day, sheltering and feeding there, though they confer no benefit in return. Only on bright, moonlight nights, when the sphinx-moths are about, does the plant reap any advantage from its visitors. A friend of the naturalist Delpino, standing by a hedge overgrown with this Bindweed, was able to capture numbers of one species of sphinx-moth (S. convolvuli) simply by closing with finger and thumb the orifices FIG. 419.— ROSACEOUS COROLLA of the flowers as the moths inserted -their OF A BUTTERCUP. heads. A form of corolla closely related to the infundibuliform is the campanudate, or bell-shaped. We have examples of a regular campanulate flower in the Rampion (Campanula rapunculvs, fig. 426) and other Bell-flowers, and of an irreguJ.ar or oblique campanulate flower in the equally well-known Foxglove (Digitalis, fig. 420 1. The wide bell-shaped corollas of Campanula are specially adapted for humble-bees ; but the flowers number other kinds of bees among their visitors, and numberless beetles and small flies use them as shelters from the rain or make the comfortable bells their night quarters. The Foxglove is also specially adapted for humble-bees, for, says Hermann Miiller. " no other insects are large enough to touch the stigma and anthers with their backs when creeping into the tube." They are, in fact, the only pollinators of the Foxglove. Possibly the enlargement of the under side of the corolla, which gives the irregularity to the flower, is intended as a landing-stage for the insect (it is certainly so used), and may also be an arrangement \ \_E. Step. FIG. 420. — FOXGLOVE (Digitalis purpurea), With obliquely bell-shaped corollas which can be pollinated only by humble-bees, for whie specially adapted The lip on the lower side forms an alighting platform from wuicn me right into the bell. WESTERS EUROPE. 343 344 HUTCHINSON'S POPULAR BOTANY to relieve the strain upon that part of the bell where the weight of the humble- bee presses most. We find ourselves among flowers of more intricate construction when we come to speak of labiate or lipped corollas. The White Dead-nettle (Lamium album}, one of the commonest of our weeds, is the flower chosen for illustration (fig. 425). The corolla is decidedly irregular, and on a cursory examination it is not easy to distin- guish from one another the five coher- ing petals which compose it; though by bearing in mind the simple rule that petals should alternate with sepals, the difficulty will vanish. Guided by this rule it will be found that the lower lip or cleft piece in front is one petal, that the pointed and inconspicuous FIG. 421,-MossY PINK (Phlox subulata), appendages on either side the corolla- With salver-shaped corollas. ° ,, ,. „ , tube are the rudiments ot two more, and that the remaining pair form the overshadowing hood or upper lip of the flower. "We can estimate by direct observation," says Hermann Miiller, "how perfect the adaptation of this flower is to bees' [humble- or other large bees'] visits. The bee alights on the under lip, and in doing so thrusts its head between the broad lateral lobes of the mouth, clings with its fore feet to the base of the under lip, and with its inid- and hind-feet to the two lobes of the under lip; then if its proboscis is not less than ten millimetres [about two-fifths of an inch] long, it can at once reach the base of the flower. While sucking, the thorax, and in the case of small workers the base of the abdomen also, fills up the space between the upper and lower lips, and the vaulted upper lip fits the bee's back, which is pressed against the stigma and the open face of the anther." Personate or mask-like corollas, of which the Toadflax (Linaria) and Snapdragon (Antirrhinum^) offer convenient examples, are even more elaborate in their construction than the labiate form. They re- FIG. 422. — WHORTLE- BERRY (Vaccinium). TJrceolate corolla. FIG. 423.— GRAPE- VINE (Vitis). ilitMBform corolla. FLORAL FORMS AND THEIR RELATIONS TO INSECTS 345 semble the latter in possessing a double lip, the essential point of difference being that their lower lips approximate to the upper, so as to close the orifice of the tube or throat. In the Toadflax (fig. 429) this arrangement shuts out flies and beetles, which lack the requisite strength to force an entrance, while the length of the nectar-storing spur excludes short-lipped bees, which are not so incapable of breaking in. Thus the flowers become exclusively adapted for the long-tongued species, by which, indeed, they are diligently visited. The Toadflax, it may be noted in passing, is one of those flowers which, though normally irregular, will sometimes become regular by producing in all their petals or sepals the very fea- ture which is the cause of their ir- regularity. In the flower in question, the peculiarity is that each of the five petals, by vir- tue of this excess of irregularity, pro- duces a spur, while the upper part of the flower loses its personate character §p^ \\ ^/8S2K and becomes regu- ki^3itlir-iry or in a few-flowerei raceme. EUROPE, NORTH AFRICA, NORTH ASIA, NORTH AMERICA. 347 348 HUTCHIXSON'S POPULAR BOTANY FIG. 427. — COLUMBINE (Aquilegia vulgaris), With each of the five petals ending in a spur. honey. This nectary is presented to the humble-bee just like a spoon as it sits on the lower lip. Directly the bee goes, the lower lip snaps to, and the nectary disappears from view." The Calceolarias, indeed, bring before us a new form of corolla, the calceolate, or slipper- shaped. They offer good illustra- tions of special thickenings in parts of flowers where the strain caused by insects is most felt. Calceolaria pavonii is a striking example of this. Along each of the upper edges of the curved basal part of this flower — that part which carries the inflated end upon which the bee stands — there is a thickened ridge, and this gives wonderful strength to the support and prevents injury when the bee alights. The very cohesion of the petals, in this as in all flowers where cohesion takes place, is also a source of strength : and swellings and hollow projections in particular places may subserve a similar end. The bulge in the tubular part of the Foxglove flower is evidently for this purpose : as in all probability are the curious projections in many kinds of gibbous or pouched corollas. The Great Wild Valerian ( Valeriana ojficinalis. fig. 418) offers a good example of this form. The small pouch or hump in this particular flower is the nectary, in the green fleshy floor of which the honey is secreted. As the pouch is short and easily accessible, the flower is largely patronized by insects, and cross- pollination regularly takes place. Returning for a moment to the simpler kinds of gamopetalous flowers, notice a familiar example FIG. 428.— PAPILIONACEOUS COROLLA, of faQ globose corolla (fig. 430). It is With the parts separated, (r) Yexillum or standard j.1 T\'' d LI' 1 (a) Ate or wings, (c) carina or keei. the Figwort (Scrophiilaria noctoso), FLORAL FORMS AND THEIR RELATIONS TO INSECTS 349 which blooms in moist places from July to September. Perhaps it is not the best instance that might be given of the globular form some of the Heaths (Erica) would better fulfil the requirements of the name — but we have adopted it because the flower is one of those which, unlike the flowers which we have been hitherto considering, lay them- selves out for pollination by insects with short probosces. Wasps are its chief visitors, and easily reach down to the honey at the base of the corolla : while long-tongued insects, such as humble-bees and butterflies, avoid the flower altogether. The globular form often merges impercep- tibly into the urceolate, or urn-shaped — indeed, a single spray of Heath will sometimes show both forms (fig. 431) — but a better example of an urceolate flower than the Heath is the Whortleberry ( Vacci- nium myrtiLlus), the corolla of which might have furnished designs to the sepulchral-urn- makers of ancient Etruria (fig. 422). As a rule, only insects with a proboscis long enough to reach from the ex- terior to the base of the corolla, where the honey is lodged, can reach that coveted treasure, as the opening is too narrow to admit the bodies or even the heads of the majority of nectar-sipping insects. A curious and exceptional form of corolla is the mitrseform or mitre- shaped (fig. 423). It occurs in the Grape Vine (Vitis vinifera). The five petals, which are coherent at their tips, form a dome-like covering to the stamens and ovary ; but they are green and insignificant, and hardly dis- tinguishable from the foliage ; on which account they are valueless as insect lures. The plant, therefore, does well in getting rid of them as soon as possible, Photo by} [E. Step. FIG. 429. — YELLOW TOADFLAX (Linaria vulgaris), With personate or mask-like corolla. The upper and lower lips close tightly to keep out all insects but bees. 350 HUTCHINSON'S POPULAR BOTANY FIG. 430. — FIGWORT, With sub-globose corolla. Adapted for pollination by wasps. and this is effected in a beautiful manner. The petals become detached at their base, and curl up spirally — that is to say, towards their common point of attachment at the apex of the flower ; and in this condition they remain till, by the expanding of the stamens, they are thrown off completely. A more interesting, because more intricate, form of irregular flower than any which we have yet considered is that which constitutes the characteristic feature of the well-known and extensive sub-order of Legumiiiosse, to which the Pea and Bean belong — namely, Papilionacese. This name at once suggests the form of the corolla, which Tournefort and Ray, and the early botanists generally, conceived to bear a resemblance to a butterfly with expanded wings. Papilio is the Latin word for '• butterfly," and hence the corolla came to be called papilionaceous and the sub-order Papilionaceae. But it was reserved for a modern philosopher to advance the startling suggestion that possibly the first butterflies were flowers "which got loose from their stalks and flew away " ! Papilionaceous flowers are pollinated almost exclusively by bees, to which the several parts of the corolla bear the most evident relation. The reproductive organs (stamens and pistil) are contained in the two inferior petals (fig. 428), which cohere to form the keel or car-ilia (cc), the latter an excellent contrivance for protecting the anthers from pollen-feeding insects and from rain. The lateral petals (a a) — known as the wings or aloe, — besides affording a platform for bees, serve as a lever to depress the keel, as well as to bring it back to its place after depression, if repeated insect-visits are necessary for pollination. Lastly, the large posterior petal, which is known as the standard or vexillum (v), gives conspicuousness to the flower, as well as closes the entrance to the nectary from behind, so that insects seeking honey must sit either on the keel or wings. We must not conclude these remarks upon the forms of calyxes and corollas without some reference to composite Row GIB A composite flower is really a number of florets crowded together on a single base or receptacle, such as we find, for instance, in the flower-head (capitulum) of a Daisy (BMis perennis) or Dandelion (Taraxacumofficinale). We have chosen for illustration the flower-head of a Chrysanthemum (C. carinatuTri), the florets of which are more easily distinguished than are those of the commoner flowers above named. What you see (fig. 433) is a section, the flower- FIG. 431.— HISPID HEATH. Complete flower (upper), and the same after removal of calyx and corolla, to show position of stamen* and pistil. 351 352 HUTCHINSON'S POPULAR BOTANY FIG. 433. — ANNUAL CHRYSANTHEMUM (Chrysanthemum carinatum). Section througn a flower-head (capitulum) to show how tlie numerous florets are arranged upon the receptacle. head having been cut through longitudinally in order to exhibit a perfect row of the disc florets. Each of these florets, which you will observe are tubular in form, is a perfect flower, capable of setting seed. The corolla- tube covers at its base the ovary, and hides from view the greater part of the styles, the only portion visible being the end which bears the branched stigmas. Surrounding the flower-head are some florets of another kind, each of the so-called petals being a female flower, or, in scientific parlance, a rayed floret with ligulate (i.e. strap-shaped) corolla. One of the great purposes effected by the massing together of the florets in composite flowers is admirably suggested by Kerner in his remarks upon the taking up of pollen by insects. " Great quantities of pollen," he tells us, " adhere to the under parts of insects in the case of composite inflorescences. Shortly after the opening of the corollas the style bearing an external load of pollen is exserted from each of the little tubular and ligulate florets composing the capitulum in this group, and, owing to the fact that large numbers of these florets invariably open simultaneously, numbers of styles laden with pollen project close together from the discoid head. A largish insect settling on a capitulum may therefore be dusted with the pollen of numerous florets at once." Darwin observes that the ray-florets of composite flowers protect the florets of the disc by folding inwards at night and during rainy ? r F CLOSE-HEADED BEKARIA (Befario ctata). The Befarias are a semis of plants districts of Peru. The apparently undersides. The purple flowers for elated to the Rhododendrons. The species illustrated is a native of the Alpine owiU-il. but really alternate leaves are oblong-, smooth and leathery, with grey a terminal corymb, and their footstalks and sepals are covered with rusty cotton. FLORAL FORMS AND THEIR RELATIONS TO INSECTS 353 weather. Composite flowers constitute the largest order in the Vegetable Kingdom (Composite), about one-tenth of the Flowering Plants belonging to it. A characteristic of this order not yet touched upon, yet intimately con- nected with our present subject, is the production of a pappus from the limb of the calyx. Two forms of such hairy crowns are the sessile and the stipitate. The formation of a pappus is looked upon as a modification of the calyx. It is made subservient, as every child knows, to the scattering of the fruit. In not a few flowers the corolla is provided with a supplementary organ known as the corona or crown, sometimes called the paracorolla, which in some cases is small and inconspicuous, and in others large enough to add materially to the beauty or singularity of the plant. A corona is one of the dis- tinguishing marks of the large genus Narcissus , to which our own Wild Daffodil (Narcissus pseudo-narcissus) belongs. In a less exaggerated form this " corona " will be found also in the Forget - me -not (Myosotis) and Primula. Henslow (Journal of the Linnean Society, vol. xvi., 1877) re- gards it as a de- velopment of a fold in the inner epider- mis of the corolla. In the well-known Poet's Narcissus (JV. poeticus) the white crown is surrounded by a cinnabar-red border, which is probably a means of attracting in- sects ; and in most species of the family this organ is either delicately marke d or the whole of the corona is of a ii— 5 Photo by] \.E- SleP- FIG. 434. — HENBANE (Hyoscyamus niger). The yellow corolla is veined with purple, and the urn-shaped calyx is persistent. The seed-vessel is a box or pyxis with a well-defined lid. 354 HUTCHINSON'S POPULAR BOTANY deeper colour than the rest of the perianth. The crown of the Passion- flower (Passiftora), which in some species is single, in others double, is split up into narrow threads which Fritz Miiller, a naturalist who has given much attention to the mechanism of floral organs, believes to be of service in detaining small insects in the lowest chamber of the flower, and keeping them caged for humming-birds, the chief pollinators of many species of Passiflora. In the White Dead-nettle (Lamium album} there is a circle of hairs in the narrow part of the corolla-tube, near the base, which serves the purpose of excluding flies and small bees from the nectar, these being useless to the plant. The corona of our beautiful bog-flower, Grass of Parnassus (Parnassia palustris), is very interesting. It consists of five scales (the nectaries) terminating in hairs, each of which is surmounted by a yellow FIG. 435. — ANTHERS AND THEIR DEHISCENCE. (a) Longitudinal, (6) transverse, (c) valvular, (d) porous dehiscence ; (e) cross-section of stamen of a Lily (/) stamens of a Milkwort (Polygala erioptera). glandular body, which has all the appearance of a drop of fluid. Even flies are deceived by these shining knobs, mistaking them for drops of honey. We come now to the stamens or male organs, which are known col- lectively as the andrwcium. The forms of these, like the forms of the calyx and corolla, vary considerably. Bearing in mind what has been said already about the stamens in connection with cohesion and adhesion, there is really little else to be acquired. It is instructive, however, to notice the diversities of form and mechanism which the andrcecium presents, and which are usually connected with that most important function, the scattering of the fertilizing meal or pollen. Thus, in the Grasses (Graminese) and Plantains (Plantago) whose pollen is carried by the wind, the anthers swing loosely on their connectives, and so assist the wind in its useful labour — they are versatile ; while in cer- tain species of Sage (Salvia) the connective forms a curved bar or lever, and FIG. 436.— PISTILS. (a) Canterbury Bell ; (6) Pansy : transverse section of young ovary ; (c) Pansy ; (d) Begonia ; («) White Campion (/) Calceolaria ; (?) Clarkia ; (*) Periwuikle. 355 356 HUTCHINSON'S POPULAR BOTANY runs transversely to the filament, to which it is attached by a movable joint, the purpose of which will be explained hereafter. From some stamens the filament is entirely absent, and the anthers, which are then described as sessile, may be attached either to the petal of the flower, as in Vervain (Verbena), or to some part of the pistil, as in the Birthwort (Aristolochia). In a large number of cases the connective is perfectly continuous with — or, in other words, is a direct prolongation of — the filament, the point of attachment of the anthers being immediately upon the top of the filament ; but far more frequently the latter is prolonged up behind the anther-lobes. The terms basifixed or innate and dwsifixed or adnate are applied respectively to these two kinds of attachment. When the face of the anther — or, in other words, the surface opposed to that to which the connective is attached — is turned towards the pistil of the flower, the stamens are said to be introrse : when, on the other hand, the face is turned towards the petals, the stamens are extrorse. The direction in which the anthers dehisce — that is, discharge their pollen — is thus largely- dependent upon the position of the stamens — a fact to be carefully noted. The mode of de- hiscence is chiefly determined by the nature of the aper- tures through which the pollen escapes. When, as is most frequently the case, FIG. 437.-STAMENS OF (a) LITS^A ; (6) PYKOLA ; (c) GARCIXIA. the anther opens by narrow upright slits running from top to bottom of the lobes, the dehiscence is said to be longitudinal ; when the lobes are slit crosswise, the dehiscence is transverse. In not a few flowers the lobal orifices are wide and concealed by delicate valves or flaps, which lift up like a trap-door at the moment of shedding the pollen. This is valvular dehiscence. When, as in the case of the Common Whortleberry (Vaccinium myrtillus), the dehiscence is effected by minute fissures or pores at the tips of the lobes, we speak of it as porous. The anthers assume the most various shapes — globular and oval, pear-shaped and worm-shaped, curved and undulating (the latter a form which reminds one of the convolutions of the human brain), pitcher-like and box-shaped, etc., etc. ; while many of these forms are rendered still more striking by the presence of appendages in the shape of hairs, tubes, fleshy hooks, leaf-like expansions, coloured bladders for attracting insects, feathery growths, and other singular contrivances whose name is Legion. Such appendages are usually prolongations of the connective. The almost naked stamens of Leitneria floridana, a shrub with willow-like leaves FLORAL FORMS AND THEIR RELATIONS TO INSECTS 35? inhabiting the marshes of the southern United States, are seated upon bracts, furnished with hairy appendages, which give a curious flocculent appearance to the whole inflorescence. The morphology of the gyncecium, or pistil, need not detain us long. A pistil also consists of three parts — the ovary, the style, and the stigma. Just as the stamen assumes different forms according as its filament is long or short, its connective with or without appendage, its anther-lobes globular or oval, pear-shaped or linear, etc., etc., so the form of the pistil varies in accordance with analogous diversities in the ovary, style, and stigma. The mode of insertion of the style is also a cause of variation. It may spring either from the top of the ovary, when it is terminal ; or from the side, when it is lateral ; or from the base, when it is basilar. Sometimes it is entirely absent, and then the stigma rests upon the ovary, and is said to be The style, indeed, is not an essential part of the pistil ; the stigma is. It will be remembered that the latter is the organ which serves for the detention of the pollen-grains. Nature, as might be guessed, has all sorts of devices for facilitating this object, and hence the multitudinous forms of stigma— peltate, plumose, penicillate, petaloid, etc. — which the pistils of flowers present. It should be added that the relative position of the calyx, Corolla, and Stamens to the pistil Section through perigynous flowc is a matter of considerable importance. Observe the flower of a Geranium. Here the ovary is superior to the other floral organs, which are attached to or below its base. Such a flower is termed hypogynous, from the Greek words hupo, under, and gune, a woman. Compare these with the sections of the flowers of Rose and Apple, in which, by the more vigorous growth of the axes, a tube is formed around the carpels, and the stamens and perianth are raised so that they stand on the apex of the rim of the tube. The Rose and Apple are perigynous flowers (Greek peri, around, and gune).* In the Begonia and Carrot we have a third mode of insertion. In both cases the ovaries are situated below the perianth— they are inferior; whilst the flowers as a whole are described as epigynous (Greek epi, upon, and gune). * In the case of the Rose, each of the carpels has a distinct ovary ; in the Apple the ovary fills the \vhole cavity of the tube, with the inner wall of which it is fused. The immature " pips " are ovules, not ovaries. Photo by] FIG. 439. — SWEET VIOLET (Viola odorata). Showing the reversal of the flower by the curving of the flower-stalk. The plant to the left is the white-flowered form. CHAPTER XII THE WELCOME OF THE FLOWERS These have their sexes! and when summer sliines. The bee transports the fertilizing meal From flower to flower, and e'en the breathing air Wafts the rich rrize to its appointed use. COWPEB. THE important part which insects and other external agents play in the fertilization of flowers was indirectly alluded to in the last chapter, and we propose now to follow out this interesting subject in some detail. The insects which visit flowers may, for popular purposes, be divided into two classes — bidden and unbidden guests : the former being useful to the plant, the latter useless. It may be added that an insect which is a welcome visitor to one kind of flower may be an unwelcome visitor to another; so that the terms " bidden" and " unbidden'7 have only a relative significance. Incidental allusion has been already made to some of the contrivances in plants for the exclusion of certain visitors : we may now consider a few more of these contrivances, with special reference to the flower. Of the necessity of excluding various kinds of insects from certain flowers there can be no question. ;' Guests might come,'' says Dr. Ogle, 358 [Henry Irving. FIG. 440. — SNOWDROP (Galanthus nivalis). The Snowdrop produces but a single flower throughout the year, but thuMower, if unfertilized, •£«»*£ fresh and open for a lona period ; and hence, if deprived of the visits of hive-bees (its chief pollinatore) for many days .at a time which may often happen in consequence of unfavourable climatic conditions), the ability to hold out is of greal advantage to the plant. EUROPE and w. ASIA. 359 360 HUTCHINSON'S POPULAR BOTANY FIG. 441. — CYCLAMEN (Cyclamen europceum). in his amusing preface to Kerner's famous little book, Flowers and their Unbidden Guests, " who were not of sufficient importance, and the ban- quet [whether of nectar or pollen] be wasted on them ; for it is only when insects have a certain shape, size, or weight that she requires their visits, and can use them profitably for her purposes. . . . All insignificant and unremunera- tive visitors, all such, moreover, as would creep in by the back entrance, must be kept out." Thus the opposite leaves of a plant may form a kind of collar, or series of collars, to the inflor- escence, insurmountable to wingless insects from below, as is the case in many Gentians ; or even the stipules and alternate leaves may act in the same way, as in the Common Pear and Thorow-wax respectively. How excellently, again, is a pendulous flower adapted for the exclusion of small creeping insects ! Take the Snowdrop (Galanthus nivalis, fig. 440). Where is the ant that could get inside the hanging flowers of this February maid ? — if the ant were in the habit of climbing up plants at that season. The slippery curved walls would defy all its efforts ; and, as a matter of fact, only winged insects pollinate the flower. Hive-bees, which are the most useful to the plant, enter the drooping bells without difficulty. When the object of a hive-bee's search is the pollen, " it thrusts its head, fore-legs, and mid- legs into the flower, clinging by means of its hind-legs to the outer surface of an inner perianth- segment. With the tarsal brushes of its fore- and mid-legs it sweeps pollen from the anthers and places it in the baskets on its hind-legs. If it wishes to suck honey, it usually finds it more convenient to use its own fore- and mid-legs for clinging to the perianth." * In either case the bee's head gets well covered with pollen, some of which is sure to be deposited on the stigma of the next flower which it visits, for the style of the Snowdrop projects beyond the anthers, and .the bee's head must come in contact with the stigmatic surface on entering. The curvature of many perianth leaves also subserves the purpose of excluding wingless insects from the nectar and pollen. "I placed," FIG. 442.— PLUMBAGO sa^s Kerner, " some small and by no means timid (Plumbago capensis). * Miiller : Fertilization of Flowers. THE WELCOME OF THE FLOWERS 361 ants, of a kind (Lasius nigra) which under ordinary circumstances show themselves to be capital climbers, on the flowers of Cyclamen europceum (fig. 441). At first they tried to make their escape downwards by the peduncle ; but as I had put the flower-stalk in water, they turned back and managed to recross the calyx and get back to the corolla. After some useless clambering about the reflexed tips of the petals, they at last reached their curved margins, and here all their skill was baffled, and they fell either into the water or to the ground." The calyx, epi-calyx, and bracts may be further protections to the flower, by prevent- ing insects from eating their way through the corolla to the nectary, a burglarious proceeding of which even bees are sometimes guilty. We have seen how admir- ably an inflated calyx effects the same results, holding the would-be intruder at a distance from the honey, even when the tissue has been gnawed through ; and the fact might be en- forced by other exam- ples. Hermann Miiller remarks of one of the Louseworts (Pedicularis Pfotow ^. FIG. 443. — MARSH CALLA (Calla palustris). The white spathe serves as an alighting platform for the flies that are attracted by the unpleasant odour to pollinate the flowers. verticillata) that "the calyx is swollen, and the lower part of the corolla-tube is bent at right angles within the calyx; the honey is^thus guarded from Bombus mastrucatus, which tries in vain to reach it." In the Canterbury Bell (Campanula medium) the tough inflated hairy calyx, with its valvate divisions, stands above the ovary, the nectar being sufficiently guarded by the expanded bases of the five stamens which surround it (fig. 436 a); while in Clarkia (fig. 436 g) the curious boat- shaped gamosepalous calyx, like the curved petals of the Cyclamen referred 362 HUTCHINSON'S POPULAR BOTANY to by Kerner, is an admirable contrivance for keeping out small wingless insects. We have seen how perfectly the flowers of Antirrhinum and Toadflax are adapted for the exclusion of certain intruders, and many equally perfect adaptations of corolla and calyx might be enumer- ated, where the same end is to be gained. In a number of in- stances, the corolla " forms a narrow tube, still further protected by the presence of hairs, sometimes scattered, sometimes, as in the White Dead-nettle, forming a row. In others the tube itself is so narrow that even an ant could not force its way down : while in some of the Gentians the opening of the tube is protected by the swollen head of the pistil. ... In Clover 1 Trifolium), Birds- foot Trefoil (Lotus), and many other LeguminossB, the ovary and the stamens, which cling round the ovary in a closely fitting tube, fill up almost the whole space between the petals, leaving only a very narrow tube. In still more numerous species the access of ants and other creeping insects is prevented by the presence of spines or hairs, which constitute a veritable cheval de frise. Often these hairs are placed on the flowers themselves, as in some Verbenas and Gentians. Occasionally the whole plant is more or less hairy : and it Photo fry] [B. Step. FIG. 444. — MARSH MARIGOLD (Caltha palustris). A splendid Buttercup whose brilliance is due to the golden sepals, t!ie petals being absent. Photo by} [E. Step. FIG. 445. — PURPLE LOOSESTRIFE (Lythrum salicaria). The flowers, though outwardly all alike, are of three forms, differing in the length of the pistils and the filament of tfie stamens : a plan that secures cross-pollination. NORTHERN TEMPERATE REGIONS and AUSTRALIA. 364 HUTCHINSON'S POPULAR BOTANY will be observed that the hairs of plants have a great tendency to point downwards, which, of course, constitutes them a more efficacious barrier " (Lord Avebury). Lastly, in not a few cases creeping insects are kept away from the interior of the flower by viscid secretions on the stem or calyx, to which the unfortunate visitors get glued, and from which there is usually no escape. Thus the calyx of Plumbago (fig. 442) is furnished with glandular hairs, which stand out horizontally from the epidermis, and are fatal to many a wandering aphis and small fly. The Honeysuckle (Lonicera) is another familiar instance of a flower which produces these viscid protective hairs on its calyx, but in the Catchflies (Silene nutans and S. noctiflora) they are much more effective, as may be judged by the number of small insects usually found glued to the calyx and flower-stalk. The Tutsan-leaved Dogsbane (Apocynum androsemifolium) is a plant which re- sorts to ex- treme mea- sures in dealing with un invited guests. The French call it G o b e - mouche, or Fly-gulper, and the name is well bestowed. In tact, the P. stamens ot FIG. 446. — GOATSBEAKD (Tragopogon pratensis). The involucral bracts are longer than the ray-florets. The flower is supposed to close at noo (" John-ffo-to-betl-at-noon "). but it is seldom open so late. the flower have an ugly trick of nipping intruding flies by their probosces and detaining them as captives till death puts an end to their miseries. Then the filaments open and the dead insects are released. " Allured by the honey on the nectary of the expanded blossom," says Knapp, in his Journal of a Naturalist, " the instant the trunk [of the fly] is protruded to feed on it, the filaments close, and, catching the fly by the extremity of its proboscis, detain the poor victim writhing in protracted struggles till released by death, a death apparently occasioned by exhaustion alone : the filaments then relax, and the body falls to the ground. The plant will at times be dusky from the numbers of imprisoned wretches." More than one plant is known in which the expulsion of unremunerative visitors is effected, not by the plant itself, but by other and remunerative insects. Kerner has enumerated four of such plants — Centaurea alpina, G. ruthenica, Juritiea mollis, and Serratula lycopifolia, all of them belonging THE WELCOME OF THE FLOWERS 365 to the great Composite order. " The young capitula of these Composites," says the eminent naturalist, " are particularly liable to the attacks of devouring beetles, especially of Oxythyrea fanesta, which bites big holes in the heads, destroying crowded flower-buds and involucral scales without the least difficulty. To meet this danger a garrison of war- like 'ants is em- ployed. Honey is secreted from big stomata on the im- bricating scales of the st ijl 1-clo se d capitula in such quantities that one can see a drop of it on every scale in the early morning, while later in the day, as the water evapor- ates, little masses or even crystals of sugar are to be found. This sugar, either in its liquid or solid form, is very palatable to the ants, which habitually re- sort to these capitu- la during the period of its secretion, and to preserve it for themselves they re- sent any invasion from outside. If one of the afore-men- tioned beetles ap- pears, they assume a menacing attitude. They hold on to the involucral scales with their last pair of legs and present their forelegs, abdomen, and powerful jaws to the enemy. Thus they remain till the beetle withdraws, if necessary hastening its retreat by squirting formic acid in its direction." It has been remarked that wingless insects are most active when the dew , FIG. 447. — MAKSH LOUSEWORT (Pedicularis paluatris). Parasitic upon the roots of other plants. It has a dull pink corolla and reddish-green calyx. 366 HUTCHINSON'S POPULAR BOTANY FIG. 448. — FLOWER OF HORSE CHESTNUT. Hermaphrodite or complete flower with stamens and pistil. has evaporated, and that this is the signal for the closing of many flowers ; but it may be doubted whether these circumstances stand in the relation of cause and effect. Flowers open and close at almost all times of the day and night ; and though in some cases the periods of their opening have distinct reference to the visits of bidden guests, it seems extremely doubtful that the periods of their closing have any reference to unbidden ones. Be that as it may, the phenomena of the open- ing and closing of flowers are full of interest, and as the subject has been named, we will offer a few remarks upon it before passing on. It is well known that Linnaeus devised a floral dock at Upsala, by grouping together plants according to the hours at which they open and close, and that for a time the growing of these flower-clocks in public and private gardens became quite a rage. In recent years Kerner has repeated the experiment at Innsbruck (47° N. lat.), and a comparison of his tables with those of the older naturalist has shown that flowers both open and close earlier in the day at Upsala than at the more southerly situated Innsbruck. " The result," he says, " especially the earlier opening, is probably connected with the fact that the sun during the flowering season of the plants in question rises about an hour and a half earlier at Upsala than at Innsbruck." Of the flowers in Linnaeus' clock the earliest to open was the Goatsbeard (Tragopogon pratensis^ fig. 446), a Composite flower like a large Dandelion, which has received in this country the name of John-go-to-bed-at-noon, from its habit of closing about midday. Broad o'er its imbricated cup The Goatsbeard spreads its saffron rays, But shuts its cautious florets up, Retiring from the noontide blaze. Owing to its very early opening it has been pollinated usually long before noon ; and as soon as a flower has been pollinated it either begins to close, changes colour, or casts off its now useless non-essential organs. The latest flower to open was the magnificent Queen of the Night, which unfolded its scentless white petals two hours before midnight. We approach now a new and more important division of the subject — namely, the means by which the pollen of flowers is conveyed to the FIG. 449. — SECTION OF HORSE CHESTNUT, MALE FLOWER. (ov) Rudimentary ovary. There is no style or stigma. 367 368 HUTCHINSON'S POPULAR BOTANY stigma as a preliminary to fertilization. The consideration of these means brings us in touch with those denizens of the insect world which may be reckoned among the bidden guests of flowers. The subject has received an extraordinary amount of attention during recent years, and is practically inex- haustible. The phenomena of which it treats are them- selves often spoken of as fertilization, but this use of the term — albeit we con- fess to being fre- quent offenders in this respect—is hardly correct. Pol- lination is a better word. Let us begin with flowers which pol- linate themselves. Now it will be evi- dent at a glance that such flowers must bear the male and female organs on the same indi- vidual ; hence we call them bisexual or hermaphrodite. The presence or absence of one or both of the essential organs of a flower is a matter of great importance. One of ^.steP. the commonest FIG. 451.-ASH (Fraxinus excelsior). hermaphrodite The flowers are without petals or sepals, and yet are perfect because each one consists of two stamens and a pistil. flowers IS the Buttercup (Ranun- culus). Here the male and female organs are both present in the shining 3^ellow cup, and hence it is not only bisexual, but perfect. Even though the Buttercup had no calyx or corolla, it would still be perfect. As a matter of fact it possesses both those organs ; and since, in addition, it THE WELCOME OF THE FLOWERS 369 possesses both pistil and androecium, it is said to be complete. The absence of any of the four organs— calyx, corolla, androecium, or pistil — renders a flower incomplete ; but only the absence of one of its essential organs (i.e. the andrcecium or the pistil) j renders it imperfect. We are speaking, of course, not of that which is accidental and abnormal, but of that which is characteristic of the flower. Thus the Common A.sh(Fraxi- nus excelsior), which bears its male arid female organs on the same flower (fig. 451), but has no floral envelopes whatever, is a perfect flower ; while, on the other hand, the Arrow- head (Sagittaria), in the different species of which both calyx and corolla are al- ways present, but which bears the sexes on different flowers, is imperfect. We need hardly add that, in both cases, the flowers are in- complete. It is, then, evident that all im- perfect flowers are either male or fe- male ; if the former, ,, ,, , Photo by] \E. Step. they are called 8ta- FIG. 452. — SALLOW (Salix caprea). minate ; if the latter, The spray pistillate — na m e s which explain themselves ; while flowers which have neither male nor female organs are described as neuter. Of this latter kind are the outer florets of some of the Composites. Now, though the sexes are often separated in the manner described, staminate and pistillate flowers are not always or necessarily borne on ii— 6 the left consists of the male catkins, the so-called "palm"; that on the right is made up of female catkins. 370 HUTCHINSON'S POPULAR BOTANY FIG. 453. — HENBIT (Lamium amplexicaule). (a) Perfect flower ; (6) cleistogamic flower ; (f) section of (&). different plants. We might, indeed, liken the androecium and pistil to tenants in a house where the rooms are flowers and the house itself is the plant. When, as in the Oak (Quercus robur), Walnut (Juglans regifi), and Sweet Chestnut (Castanea saliva, fig. 450), the male and female flowers occupy the same house, they are saicl to be monoecious (Greek monos, one, and oikos, house); when, as in the Juniper (Juniper us), Poplar (Populus), and Willow (Salix. fig. 452), the sexes not only occupy different rooms, but different houses, being borne on distinct plants, the flowers aie appropriately termed dioecious (Greek di, two. and oikos}. Hence, too, the plants themselves are capable of sexual classification, some being male, others female, others bisexual, and a fourth class, in which male, female, and hermaphrodite flowers are found on one and the same individual, polygamous. With this digression, let us revert to the subject which led to it — the pollinating of a flower by its own anthers. The process of self-pollination is the first stage of autogamy, which, as already explained, is the fecundation of a flower by its own pollen ; and the whole process may, and sometimes does, take place without the flower opening at all. It has, indeed, been discovered by comparison of several closely allied flowers that the smaller ones are frequently self-pollinated — nay, that in some cases the same plants which produce ordinary cross-pollinated flowers produce minute self- pollinated ones which never open. Thus we may say that there are two kinds of autogamous flowers — those which open like ordinary flowers, and those which remain closed. A glance at an example or two will make the fact additionally clear. To take the last-named first. These, as the whole process is effected in the closed flowers, are called cleistogamic, a name derived from the Greek kleislos, closed, and gamos, marriage. The Henbit Dead-nettle (Lamium am- plexicaule) offers excellent examples of cleistogamic d ^^^ flowers. The richly tinted reddish-purple corollas of FIG. 454. — FLOWERS OF Myosotis IN SECTION. (a) Wood Scorpion-grass; (6, c) Yellow and Blue Scorpion-gra; (d) Forget-me-not. the. expanded blossoms of this not uncommon weed are familiar to most persons Photo fty] ** | ,7 £ |[ FIG. 455. — HARD HEADS (Centaurea scabiosa). utiful composite flower, with all the florets rayed, and of a bright purple colour. The overlapping bracts ol the involucre have toothed brown margins. EVROPE, SIBERIA, W. ASIA. 371 372 HUTCHINSOX'S POPULAR BOTANY (fig. 453, «). They are met with in dry and sandy fields during the sultry months of July and August : but few probably have noticed, earlier in the year, the small bud-like unexpanded flowers of this plant ifig. 453, 6). These are cleistogamic flowers, which never open — undeveloped flower- buds, with anthers and stigmas that mature so that perfect fruits are pro- duced. "Were you to open one of these buds at the moment of pollination (fig. 453, c). you would find that the long and flexible style (st) had curled round so as to bring the inner side of its forked stigma in con- tact with one of the anthers. Perhaps you would even find that the anthers had not opened, but that the pollen- tubes had perforated its delicate walls and were growing in the direction of the stigma. The Dog-violet (Viola canina) is another plant which produces these un- developed flowers. Pro- fessor Ainsworth Davis re- marks that " in summer the ripe fruit of the cross- pollinated flowers will be found, and, close to them, minute bud-like structures. These are the cleistogamic flowers : their anthers are so placed that the pollen-grains can send their tubes straight to the stigma. Such a flower produces, perhaps, only two hundred pollen- grains, as opposed to some thousands in an ordinary blossom." As a rule, indeed, cleistogamic flowers are pollen saving. Thus, a single self-pollinating flower of Wood-sorrel (Oxalis acetosella) contains about four hundred grains : a flower of Touch- me-not Balsam (I'mpatieiis noli-me-tangei'e) about two hundred and fifty : and of Cut-grass (Leersia) not above fifty. Contrast these figures with the number of pollen-grains in the Peony, 3,500,000, or in a single flower- head of Dandelion. 365.000 ! FIG. 456. — GREAT WILLOW-HERB (Epilobium hirsutum). (a) Flowering branch ; (6, c) stamens and pistil arranged for cross- pollination. In (