tifp §. ^. pm pkarg ^9 EPH Rl/ZIHKA This book must not be taken from the Library bmlding. 2.05^?. ^ ^ ■ PREFACE BY THE AUTHOR. This book is intended cliiefly for those who, without desiring to become botanists by profession, wish nevertheless to be- come acquainted with the elements of scientific structural botany. It will likewise introduce the beginner to the various methods of microscopical manipulation. The study of vegetable structure is especially favourable as an initiation into the use of the microscope ; and any one whose future career will require command over this instru- ment should commence with the study under the microscope of vegetable anatomy. The manual is divided into thirty-two chapters, each of which is intended to provide materials for several hours' prac- tical work in the laboratory. The earlier chapters are easy, and the difficulties to be encountered increase almost con- tinuously up to the last chapter. The first chapter assumes on the part of the worker entire ignorance as to the use of his instruments, but nevertheless assumes the possession of some general botanical knowledge. With this elementary pre- paration the beginner ought to be able, by the diligent use of this book alone, to acquire a tolerably broad knowledge both of vegetable structure and of the methods of micro- scopical work. The objects for study have been so selected that most can be obtained with comparative ease. In many places I have recommended the use of plants preserved in alcohol, as the worker is thus rendered independent of the time of year. As, however, the objects may need to be collected even month:^ VI PEEFACE TO THE ENGLISH EDITION. have given at tlie head, of eacli chapter (task^ or lesson^ in the original) a list of the objects required for study in that chapter. I regret that I did not add to these lists any- special reagents which might be required for use ; possibly a future edition may give opportunity for this. I have considerably enlarged the scope of Appendices II. and III., and have added two new Appendices, I. and IV., which I hope may be useful to the student. Throughout the work I have likewise added the common English names (if any) of the plants referred to. The student will probably not be able to carry out all the investigations constituting a chapter at the same time. A careful note should be made of any which are thus postponed, so that they may be taken up in due season. It is not un- likely that some may not come at all within the range of the student's observation ; for these examples the' book must be looked upon in the light of a text-book. The student is earnestly urged to study from the beginning the Author's methods of work. These are especially note- worthy when he comes, perha,ps incidentally, to correlate structure with function. The interdependence of these two factors in the plants' life history is the great underlying principle of modern botanical teaching, and the student cannot too soon begin to exercise his thoughts in this direc- tion, resting assured that his methods are right even though his results may for the time being prove to be erroneous. As to translation, no one can feel so fully as myself its many and serious defects. I can only plead that the work was executed at a time of serious pressure, and, although cir- cumstances have delayed the issue of the book, the manuscript was out of my hands, and therefore only subject to such limited correction as proof-sheets would allow. W. H. Birmingham, Se^iemherj 1886. "t CONTENTS PA.OB Introduction. Instruments, Apparatus, Reagents, Materials . . xiii Fig. 1. Zinc frame for object-slides xxiii I. Use of the Microscope. Structure of Starch Fig. 2. Microscope Stand VIIa, of Zeiss 2 3. Starch-grains from Potato tuber 8 4. Starch-grains from Bean 10 5. Starch-grains from East-Indian Arrowroot ... 11 6. Starch-grains from Wheat-meal of Tn'ticuni duru)7i . . 11 7. Starch-grains from ^ueria safiva 12 8. Starch-grains from latex of EiipJiorbia Tiel ioscopia . . 12 9. Starch-grains from latex of Euphorbia splcndcus . . 13 II. Aleurone-grains, Protein Crystals, Fat Oil, mounting of permanent preparations, use of the simple Microscope 16 Fig. 10. Cells from cotyledons of Pea 18 ,, 11. Cross-section of outer part of grain of Wheat ... 19 ,, 12. Small dissecting microscope of Zeiss 21 „ 13. Large dissecting microscope of Zeiss 22 „ 14. Cell showing aleurone grains and albumen crystals, from the endosperm of Bicinus communis . . . . 25 III, Movements of the Protoplasm ; Nucleus. Drawing with the Camera, etc. ; calculation of Magnification 28 Fig. 15. Protoplasmic movement in hair of filament of Tradcscantia virginica 29 „ 16. Camera lucida of Abbe 30 IV. Chromatophores. Coloured cell-sap 38 Fig. 17. Chlorophyll-bodies of Funaria liygrometrica ... 39 „ 18. Colour-bodies from calyx of Troixjeolum majus ... 40 ,, 19. Epidermal cell from petal of Vinca minor .... 42 „ 20. Colour- bodies from root of Carrot 43 „ 21. Starch-builders, with starch grains, from rhizome of Iris germanica 44 vii vni CONTENTS. CHAPTT^H PAGE V. Tissues ; thickening of the walls ; reaction for Sugar ; Inuline, Nitrates, Tannin, Lignin 45 Fig. 22. Stone-cells (sclerenchyma) from fruit of Pear ... 47 „ 23. Striate cell from pitli of tuber of Dahlia variabilis . . 50 „ 24. Sphaero-crystals of Inuline in tuber of Dahlia variabilis . 51 „ 25. Thickened and pitted walls from endosperm of Ornitho- galum umbellatum, 54 ,, 26. Bordered pits of Pinus sylvestris 53 Vl. The Epidermis, Stomata, Water Stomata CI Fig. 27. Epidermis and stoma of leaf of Iris floreniina ... 03 ,, 28. Epidermis and stoma of leaf of Tradcscantia virginica . 68 ,, 29. Epidermis and stoma of leaf of Aloii nigricans ... 68 „ 30. Epidermic cell and stoma of leaf of Aneimia fraxinifolia, . 69 „ 31. Epidermis and water stoma of leaf of TropcEohim majus . 70 VII. The Epidermis {cont.) ; Hairs. Mucilage and Wax ... .72 Fig. 32. Hair of Cheiranthus Cheiri, and of Matthiola annua . . 73 „ 33. Hair from petal of Viola tricolor 74 „ 34. Scales from leaf of Shepherdia canadensis .... 75 ,, 35. Stinging hair of X'rtica dioica 78 ,, 36. Gland from ochrese of Rumex Tpatientia .... 79 ,, 37. Digestive gland (tentacle) of Drosera rotundifolia . . 80 [ „ 37*. Diagram of leaf of ditto 80J ,, 38. Glandular hair from bud-scale of ^sculus hipi^ocastanum . 81 ,, 39. Wax on node of stem of Saccharum officinarum ... 83 VIII. Closed collateral fibre- vasal (fibro -vascular) bundles. Mucus and Gum 83 Fig. 40. Cress-section of vascular bundle in stem of Zea Mais . 84 ,, 41. Radial section of same 90 „ 42. Cross-section of vascular bundle from leaf of Iris fiorentina 94 ,, 43. Crystals of oxalate of lime 96 ,, 44. Cross-section of stem of Draccena rubra .... 97 IX. Open collateral fibro-vasal bundles 100 Fig. 45. Cross-section of vascular bundle in stem of Eaniinculus repens 101 [ ,, 45*. Latex vessels in bast of Scorzonera hispanica . . . 103] [ „ 45**. Collenchyma in petiole of Pegonia 104] ,, 46. Cross-section of vascular bundle of Aristolochia Sipho . 106 X. Structure of the Coniferous Stem 114 [Fig. 46*. Diagrammatic section of junction of spring and autumn wood in Pinus sijlvestris 115] ,, 47. Development of wood and bordered pits in Pinus syl-uesfris 116 ,, 48. Resin-canal in wood of Pinus sylvestris 118 „ 49. Sieve-tubes of Pinus sylvestris 121 ,, 50. Section of walls of ditto, treated with chlorzinc iodine . 122 [ ,, 50*. Resin passages in young bast of Hedera ?ielia; . . . 123] XI. Structure of the stem of the Lime ; Bicollateral fibro-vasal bundles of the Cucurbitacea3 ; Sieve-tubes 125 [Fig. 50**. Diagram of cross-section of twig ot Tilia .... 127] „ 61. Isolated elements of wood and bast of Tilia . . . 129 „ 52. Sieve-plates of Cucurhita Pepo ...... 132 CONTENTS. IX XII. Axial fibro-vasal Cylinder, aud secondary increase in thickness of Eoots 136 Fig. 53. Cross-section of root of ^/h'ttm Ctjpa 1^7 „ 5k Cross-section of root of .4corus CaZawMS .... i:*-S „ f6. Crosfe-section Of root of .rri.s./foi-e7i(ina 1 1<> „ 56. Cross-section of young root of Taxus baccata . . . U-i XIII. The vascular bundle of the Ferns and Lycopodiacese (Club Mosses) 145 Fig. 57. Cross-section through vascular bundle of Pteris aquilina . 116 „ 63. Cross-section of stem of LycopoJLium complanatam . . 149 XIV. Cork, Lenticels ; the fall of leaves 152 Fig. 59. Cork-development in stem of Sambuctts 7ngra . . . l-"'{ „ 60. CrosS'Section through Lenticel of SambucKS nij/ra . . lil XV. Structure of foliage and of floral leaves. Terminations of the fibro- vasal bundles Fig. 61. Surface section of leaf of Rnta graveolens „ 62. Cross-section of leaf of Ruta graveolens [ „ 62*. Oil-gland of leaf of Bictamnus Fraxinella „ 63. Cross-section of leaf of Fagus sylvatica . 160 161 162 164] 16J 177 XVI. The growing Apex of the Stem. Differentiation of the Tissues. Course of the fibro-vasal bundles 170 Fig. 64. L-ongitudinal section of growing point of Hippuris vulgaris. 173 "■"„ 65. Apes of stem of Euonyirius jfiponicus 175 „ 66. Longitudinal section of growing point of Equisetuni o.rvense [ „ 66*. Schemes of division of apical cell of stem of Equisetum . 178] , 67. Longitudinal section of bud of E<2uiseiuw aruense . . 179 ,, 6S. Scheme of course of vascular bundles in stem of jK^uisetum. 181 XVII. Growing Apex (tip) of the root 183 Fig. 69. Longitudinal section of root-apex of HordeiimDiiIgare . 184 „ 70. Longitudinal section of root-apex of Thuja occidcntalis . 187 „ 71. Longitudinal section of root^apex of Pteris cretica . . 188 XVIII. Vegetative structure of the Mosses and Liverworts .... 190 [Fig. 71*. Germinating spores and protonema of Funnria liygro- metrica 191] „ 72. Thallus and air-pores of Marchantia. poly morpha . . . 196 ,, 73, Apex of thallus of Metzgeria f areata 198 XIX. Vegetative structure of Fungi, Lichens, and Algte. Staining the cell-contents 200 Fig. 74. Cross-section of byphal tissue of Agarious campestria . 201 „ 75. Cell of dadoplwra glomerata 203 „ 76. Cell of Spirogyra majuscula , 208 XX. Diatomacese, Protococcus, Yeast, Schizophycene (splitting Alga?). 210 Fig. 77. Pinnidaria viridis 212 ,» 78. Protococcns viridis -14 ,) 79. Sn ccharomyces cerevisia: 215 „ 80. Anahoiua Azollm 216 „ 81. Oscillaria princeps and 0. FrocUchii 217 „ 82. Gleocapsa polydermatica • 218 b CONTENTS. XXI. Schizomycetes (Bacteria). Use of immersion objectives . . 221 Fig. 83. Microscope Stand Va of Zeiss 225 [ „ 83*. Student's monocular Microscope Stand of Ross . . . 228] ,, 84. Spirochcete plicatilis 231 ,, 85. Bacillus suhtilis 237 XXII. The reproduction of Algae 246 Fig. 86. Swarm-spore of CladopTiora glomerata 250 „ 87. Sporangium and swarm-spore of Vaucheria sess/Hs . . 251 „ 88. Oogonium and antheridium of Vaucheria sessilis . . . 252 XXIII. The reproduction of Fungi 255 Fig. 89. Gonidiophores of Phytophthora infestans .... 258 ,, 90. Fruiting branches of Penicillium crustaceum . . . 260 XXIV. The reproduction of the higher Fungi and Lichens .... 262 [Fig. 90*. Puccinia graminis and ^cidium Berbendis .... 266] ,, 91. Conidia production of Hussula rubra 267 [ „ 91*. Coniciia production of Agaricus campestris .... 268] „ 92. Asci of MorcheUa esculenta 269 „ 93. Cross-section of spermogone of .4iioptj/c?ixa cilioris . . 271 XXV. The reproduction of Mosses and Liverworts 272 [Fig. 93*. Male receptacle and gemma-cup of Marchantia polymorpha 273] „ 94. Antheridium and spermatozoids of Marchanfi'a polymorpha 274 ,, 95. Archegonia of Marc/ia7ih'apolymorp?ia 276 95a. Antheridium and spermatozoid of Funaria hygrometrica . 279] 95b. Archegonia of Funaria hygrometrica 281] 95c. Development of sporogone of Funaria hygrometrica . . 281] 9£(J. Sporogonium of Funaria /ij/grometfica 285] 95e. Peristome of Fontinalis aniipyretica 285J XXVI. The reproduction of the vascular Cryptogams 287" Fig. 96. Sorus and sporangia of Scolopendrium vulgare . . . 28S „ 97. Antheridium and spermatozoids of Polypodium vulgare . 292 „ 93. ArchegonisL of Folypodium vulgare 295 XXVII. The reproduction of Gymnosperms 298 Fig. 99. Male cone, anther, and pollen of Pinus .... 299 ,, 100. Female flower of Taxus baccata 303 „ 101. Fruit-scale of Pinus sylvestris 305 „ 102. Longitudinal section of ovule of Piceo vulfiforis . . . 308 „ 103. Longitudinal section of ripe embryo of Picea . . . 310 XXVIII. The Androecium of Angiosperms 3X1 Fig. 104. Section of anther and pollen-development of HemerocalUs falva 313 ,, 105. Pollen-grains of Tradescantia virginica . . • . • 317 XXIX. The Gynaecium of Angiosperms 322 Fig. 106. Cross-section of ovary of Delphinium Ajacis . . . 323 „ 107. Longitudinal section of ovule of Aconitum Napellus . . 32S ,, 108. Ovule and embryo-sac of Monotropa Hypopitys . . . 331 „ 109. Ovule and embryo-sac of Orchis pallens .... 333 „ 110. Ovule and embryo-sac of Torenia asiatica .... 335 CONTENTS. x' CHAPTER XXX. Structure of the seed of Angiosperms ^333 Fig. 111. Longitudinal section of seed of CapseUa Biirsa-pastoris . 339 „ 112. Longitudinal section of achene of Alisma Flantago . ! 343 XXXI. The fruit of Angiosperms 3. y XXXII. Cell-division and Nuclear division 3rg Fig. 113. Nuclear division in stnminal hair of Tradescantia virginica . 358 „ 114. Nuclear division in pollen-grains of Fritillaria i)ersica . 362 „ 115. Nuclear division in pollen-grains of HeUeborus/ceeiclus *. 367 „ 116. Direct nuclear division in Tradescantia virginica . . . 369 Additions and Corrections ono Appendix I. English and Metric Weights and Measures .... 375 Appendix II. List of Plants used for Study 37g Appendix III. List of Eeagents used ; their Preparation and Use . . 388 Appendix IV. General Notes on Methods, and Select List of Eeagents. 400 I^^^^ 405 INTEODUCTION. Students at Universities, or properly equipped Colleges, or Schools of Science, will usually find in the Botanical Laboratory the instru- ments which are needed for their work. For those, however, who are not connected with such an institution, but may use this book independently as an introduction to the practical study of the minute structure of plants, as well as those who, under any circumstances, wish to become possessed of suitable instruments for microscopical work, the following lists, selected from the most recent catalogues of opticians, may be of service. The first list here following includes microscopes, with a price (affixed) ranging up to about £8. I. — English and American Makers. Some of these, such as the well-known firms of Ross & Co., Powell & Lealand, R. & J. Beck, and Zentmayer, are notably dearer than many other English and American, and than most foreign makers, and therefore probably for student purposes are less available. The microscopes built on the so-called " English model " are more massive and complicated in their construction than is really necessary for student purposes, and the object on the stage is usually moved about by means of a mechanical arrangement of screws, where, for ordinary purposes, the fingers had far better be used. Further, though the diameter of the body of the English microscope may be an advantage, its length is doubtfully so, and renders the erect position of the instrument in working, which is for most purposes far the best (though a joint permitting inclination is highly desirable), almost impossible. The distance of the stage from the eye renders delicate working with the fingers a matter of some difficulty ; for it is notorious that the nearer the fingers are to the eyes, within certain limits, the more delicately their movements can be con- trolled. These makers have, however, recognised the need of instruments of more compact form, simpler construction, and XIV INTRODUCTION. lower price, and, like the cheaper English makers hereafter noted, have brought out instruments suited for general use under the name either of "Student's," "Educational," or "Economic" microscopes. Of such kind we will specially indicate a few.* These will be exclusively monocular ; binocular microscopes are in no way needed. Ross & Co. (112, IS'ew Bond Street, London) produce an instrument of high quality and comparatively low price, called the " Student's Microscope," Avith rotating glass stage, coarse and fine adjustment by screws, and a " swinging tail-piece for oblique illumination," originally devised by Zentmayer, of Philadelphia, into which may be fixed various substage appliances, such as condenser, etc. Price, with one eye-piece (A), and in mahogany case, £10 IO5. Another cheaper stand is the " Student's Mono- cular Stand, No. 1," with coarse adjustment by sliding the tube, and fine adjustment by a screw, circular rotating glass stage, and draw-tube into which the continental eye-pieces will fit. (This is the case also with the above higher-priced instrument.) With one eye-piece the price of this stand is £4 10s.; a diaphragm to the stage costs 85., and a case for the microscope lis. extra. Ross's best objectives are too expensive for ordinary student use ; he offers some for the above microscope — e.g. 1 inch of 15° angular aperture at £1 5.S., and i of 75° for £2 2^. ; or, together with the stand, as above, £8 16s., to which ought to be added a second eye-piece, which costs £1. For 155. extra this microscope can be obtained with screw coarse adjustment. Instead of these objectives, the stand as above can be fitted with objectives by other makers. f R. & J. Beck (68, Cornhill, London, E.G.) offer a useful in- strument under the name of the " Monocular Economic Micro- scope " (No. 24, Catalogue 1885), having coarse adjustment by sliding tube, fine do. by screw, drawtube, 1 eyepiece, diaphragm, * For further information and particulars the reader is referred to any of the current works on the microscope, such as those of Dr. Lionel Beale,'Dr. Jabez Hogg, and, especially, that of the late Dr. Carpenter. t It is here perhaps desirable to note that all English objectives are made with the same size screw, the " Microscopical Society's Screw," known abroad as the " English Screw," and therefore will fit any English instrument. Foreign objectives, if instructions are given to that effect, are likewise made with this screw, usually with an extra charge of but a shilling or tAvo, sometimes none. Foreign microscopes can be made with the same gauge, or provided with adaptors. It is not improbable, and greatly desirable, that the English, or some other standard, screw may become universal, as the confusion amongst foreign makers is extreme. At the same time all of, at least, the smaller English microscopes would be better made for eye-pieces of continental size. INTRODUCTION. XV 1 inch and \ inch object glasses, in mahogany case, for £5 5s. Without objectives, but with 2 eye-pieces, £3 lOs. To this can be fitted an achromatic condenser for £1 2s., and other pieces of apparatus. Beck's best objectives are expensive; but he con- structs good student glasses at a lower rate. Amongst the cheaper optical firms, we may mention the following : — C. Baker (244 & 245, High Holborn, London, W.C.) publishes a " Medical Microscope " on the old continental model (of Nachefc & Hartnack), with draw-tube, coarse adjustment by sliding tube, fine by screw, and 2 eye-pieces, in mahogany case, for £3 3s. With 1 inch and |- inch object-glasses and condenser for opaque objects, £6 7s, Chas. Collins (157, Gt. Portland Street, Oxford Street, Lon- don, W.) offers a "Histological Microscope," with coarse adjustment by rackwork or by sliding tube, fine screw adjustment, one eye- piece, 1 inch and i or ^ objectives, in mahogany case, for £5 10s.; or with extra eye-piece, polariscope and stage condenser, for £7 10s. Also a " Student's Microscope " of rather larger size, with similar fittings, at an extra cost of £1 10s. H. Crouch (66, Barbican, London, E.C.) publishes "The Histologist's Mici-oscope," coarse adjustment by sliding in cloth- lined tube, fine by screw, glass stage with diaphragm, 1 inch and 4^ or i inch objectives and 2 eye-pieces, in mahogany case, for £5 5s. A Stand Condenser can be added for 8s. 6^., and an Achromatic Condenser for £1 Is. T. Darton & Co. (45, St. John's Street, West Smithfield, London, E.C.) have an " Improved Histological Microscope," on much the same model as that of Crouch, with draw- tube, screw fine adjustment, 2 eye-pieces, ^ inch and ^ inch objectives, glass stage, in mahogany cabinet, for £5 5s. Other apparatus can be fitted. Parkes & Son (St. Mary's Row, Birmingham) offer a " Portable Educational Microscope," a reliable and very steady instrument, with coarse adjustment by body sliding in cloth-lined tube, fine by screw, draw-tube, 2 eye-pieces, 1 inch objective, separating to 2 inches, J inch ditto, with magnifying power ranging, with use of draw-tube, from 140 to 470 diameters ; also with spot lens, con- denser on jointed arm attached to the stand, diaphragm, with disc for " white-cloud illumination," and glass stage, in mahogany case with leather handle, £6 10s. If with } inch objective instead xyi INTRODUCTION. of f , increasing magnifying power to 560 diameters, 5^. extra ; or, instead, with -f inch, magnifying up to 700 diameters, £7. The object-glasses are provided with a " patent sliding adapter," obviating the necessity for screwing in exchanging one glass for another while at work. A screw nozzle can also be had to adapt it for all other objectives with the English screw. An achromatic condenser can be supplied adapted for it. PiLLisCHER (New Bond Street, London), under the name of " International Microscope," offers a stand on the old continental model, but with rackwork coarse adjustment, 2 eye-pieces, -f and -i- inch objectives, giving, with draw-tube, a range of from 60 to 420 diameters, in case, for £7 10s. Swift & Son (81, Tottenham Court Road), under the name of the " College Microscope, No, 1," offer an instrument with coarse adjustment by sliding in cloth-lined tube, fine by screw, draw-tube (too large for continental eye-pieces), diaphragm, 1 eye- piece, 1 inch and J inch objectives, in case, for £h 5s. The same with screw coarse adjustment, glass stag'e, and jointed mirror, £1 10s. extra. Achromatic condenser, 12s. to £1 5s. Other instruments of equal excellence with the above are doubtless manufactured ; there is no pretence that this list is complete, nor is any comparison intended to be instituted. German Makers. Of these, the following two makers may be considered typical : — Carl Zeiss (Jena). Stand VIIa, with 3 eye-pieces, Nos. 2, 4, and 5, and objectives B and D, price £7 18s. This stand has an unjointed back, sliding tube for coarse and screw for fine adjust- ment, and swinging mirror. The instrument gives a magnification from 70 to 580 diameters. The glasses of Zeiss are unsurpassed. E. Leitz (Wetzlar). A stand of similar model to that of Zeiss, with eye-pieces 1 and 3, and objectives 3 and 7 (No. 17 in catalogue of 1882), and magnifying from 80 to 500 diameters, price £5 10s. The objectives of Leitz are low in price, but remark- ably good. Other makers are Seibert (Wetzlar), Ben^chb (Berlin), Hart- NACK (Potsdam), Winkel (Gottingen), Plosl & Co. (Vienna), E/EICHERT (Vienna) ; all good. French Makers. The two following are probably the best : — Bezu, Hausser & Co. (Paris, Hue Bonaparte, 1, successors to the old house of Prazmowski, formerly Hartnack & Prazmowski), Stand VIII., eye-pieces 2 and 4, objectives 4 and 8, magnifying INTKODUCTION. XVU 50 to 600 diameters, price about £8. Stand VIIIa, the same as above, but with jointed back, 125. extra. C. Verick (Paris, Rue de la Parcheminerie, 2). Stand V., with jointed back, with diaphragm disk and draw-tube, 2 eye-pieces, 1 and 3, 2 objectives, 2 and 7, magnifying 60 to 570 diameters, price 165 francs (about £6 12^.) ; or Stand IV., with which achromatic condenser and polariscope can be used, about £2 more. These two instruments are now very widely used in France. Most, or all, of the above makers, English and foreign, manufac- ture microscope stands of cheaper quality than the above ; it must, however, be borne in mind that accurate observation needs an instrument which is capable of it, and while there is, and ought to be, every desire to keep the cost within a reasonably small sum, true economy does not consist in purchasing an instrument which may be a constant source of dissatisfaction, and may have to be discarded when the student emerges from his swaddling clothes. The stand which is purchased ought to be adapted to the receipt of optical apparatus other than the simple eye-piece and objective. It should have a jointed back, and be thoroughly steady in any position ; the adjustment should be easy and true, and if the body is twisted, any object observed should not be thrown out of centre ; the mirror should be plane and concave, and should have a long jointed arm ; and the stage should be constructed for the reception of a condenser. Still more essential is it that the special optical parts, the eye-pieces and objectives, should be good. They should let through the largest possible amount of light (the diaphragm will easily control its quantity if needed), and there should be a complete absence of colour, both round the exterior of the field of view, and round any object, or particles of dust, in focus. The field should be flat, so that a small object moved from one part to another alters neither in distinctness, form, nor size. Lastly, the objective should have a fair working distance from the object, or the thickness of the cover-glass, to be hereafter noted, may become a matter of great importance. In all these points, except perhaps accurate centering, the stands of English makers either equal or excel, price for price, the foreign stands ; while, on the contrary, price for price, the eye-pieces and objectives of continental makers usually are far superior to those made by the English opticians, a superiority probably due solely to the more trained skill and more patient accuracy of the work- men. XVlll INTRODUCTION. All of the work in this book, perhaps, with the exception of Chapter XXI., can be performed with the aid of objectives up to i ; but the student who has gained some experience will probably wish to add to his microscope one or more stronger objectives, in order to increase the range of his work. Increased magnification can be obtained by increased power either of eye-piece or of object-glass. All the objectives we have heretofore noted are what are called " dry " systems, since they are used for work in a dry state, and a layer of air separates the objective from the object. "Dry" objectives of high power are subject to great disadvantage from the serious loss of light their use involves. The light, in passing from the mirror to the objective, passes in the first place through air ; then the object-slide, next the object and the medium in which the object is mounted, which may be glycerine, water, alcohol, etc., or even air ; then through the cover-glass, and finally through air again. In every one of these changes light is lost. Owing to this loss of light, as well as for other reasons, it is not wise to use high power eye-pieces with dry objectives, added to which, as the eye-piece does not magnify the object, but only the image of it as given by the objective, any errors of this latter are likewise magnified by the eye-piece. To obviate in part this loss of light, what are called " immersion" objectives have been for the last few years much in use. In these objectives the cover- glass and the front lens of the object-glass are connected by a drop of liquid. Such objectives are of two kinds : " water," in which that is the liquid used ; and " homogeneous," in which the liquid is in general oil, or a mixture of oils, but some- times is glycerine, etc. The homogeneous immersion objectives are dearer, less readily cleaned after use, and require a supply of the special fluid for which they are manufactured ; but on the other hand they transmit more light, bear a higher eye-piece, and are independent in their working of the thickness of the cover- glass. Dry objectives, and water-immersions, of high power are naturally dependent on the thickness of the cover-glass which the light-rays pass through after leaving the object. To provide for this, they are usually manufactured also with "correcting screw," for use according to the thickness of the cover-glass, and at a somewhat increased price. The correcting screw accommodates the objective to the thickness of the cover-glass which happens to be in use, but a right use of it requires considerable experience, nor is it usually needed with any of the weaker immersion systems. INTRODUCTION. XIX An immersion objective, without correcting screw, is made to suit a certain medium thickness of cover-glass, which is usually stated by the optician, and it is therefore preferable for the beginner, if he wants such an objective, to use with it cover-glasses of this definite thickness.* On the correcting screw, where the system has it, are usually divisions and figures, which allow the focussing for any given thickness of cover-glass, where this is known.f But whoever does not fear the expense would do well to provide himself at once with a system for " homogeneous immersion." They are all constructed without correcting screw, since, as already indicated, the thickness of the cover-glass, of course within the permissible limits, is of no importance. By selecting a single such objective, say yV? ^^^ purchasing a series of dye-pieces, one can obtain a range of possible magnification such as could only be given by several water-immersions, or dry objectives. A system for homogeneous immersion, provided it is perfectly constructed, can therefore replace several systems of another kind.;}: Even in the smallest stands mentioned above, objectives for homogeneous immersion can be used with great advantage without any special apparatus for increasing the illumination; but the highest capabilities of the homogeneous system are only brought out by the use of a sub-stage achromatic condenser. Several of the stands referred to above have sub-stage condensers constructed specially for them, and at a cost which, for these small micro- scopes, would rarely exceed £1 10s. Owing to the prevalent use of the standard screw of the Royal Microscopical Society of London, objectives of one maker can be attached to the instrument of another. Where this screw is not in use by the maker, the objective can have an adaptor attached. A point of some importance to English purchasers of continental objectives is this : — the customary length of the tube of the micro- scope on the Continent is 150 to 170 millimetres (6 to 7 inches), and the objectives are constructed to suit this length. If the tube exceeds this length, it should be stated in ordering the objectives, that they may be modified to suit. This is especially needed in ordering objectives for homogeneous immersion. All the micro- scope stands mentioned above have tubes of continental length, and most of continental size. * On this subject see a note on page xxii. t Further information on this point in Chap. XXT., p. 224. \ Leitz, of Wetzlar, produces a -f', of remarkable excellence, for £6 10s. XX INTRODUCTION. To give a theory of the formation of the microscopic image does not come within the range of our purpose, and for this we would refer to text-books on Physics and to special works on the micro- scope. Our task, on the other hand, will consist in making the beginner acquainted with the most important facts of microscopical botany, with the use of the microscope, and with microscopical manipulation. This instruction will be given by studies upon the objects themselves. Besides the compound microscope to which we have hitherto exclusively referred, a simple, or so-called dissecting, microscope is also more or less necessary. For all the purposes of this book, and, indeed, for most botanical purposes, whether in anatomy or morphology, a dissecting microscope of very simple construction is all that is needed. Most such instruments are unnecessarily complex and expensive. Some, for instance, are constructed to magnify up to 60, 80, or even 100 diameters ; if such magnifying power is needed, the low power of the compound microscope will do equally well, dissecting being done upon the stage, but the arms being carefully supported. The following are a few typical simple microscopes, any one of which would suffice : — Ross & Co.'s " Magnifier Stand," with two lenses of |-incli and 1-inch focus, magnifying 20 and 40 linear, in flat morocco case, £2 25.* C. Collins, *' Dissecting Microscope," with two lenses, to be used together or separately, 155. (No arm-rests.) Parkes & Son, " Simple Microscope " (No. 5030), on jointed arm, with universal movement, 15s. * Swift & Son, "Simple Dissecting Microscope," with three lenses, 185.* Zeiss, " Small Dissecting Microscope " (No. Ill), I85., to which double lenses, magnifying 10, 15, or 30 diameters, at 65. each, (Arm-rests.) The student can entirely dispense with a dissecting microscope, and dissect upon the stage of his larger instrument; but as the image of the object is inverted, and any movements he may make are likewise reversed, he wo aid probably be at first somewhat perplexed. Practice will overcome this difficulty; or it can be cleared away at once by purchasing an " erector " for insertion in * In these instruments the object is dissected on the table, or in any other convenient place. Those not marked have a special stage, with or without arm- rests, as indicated. See also p. 21 et seq. INTRODUCTIOJJ. XXI the draw-tube, costing usually 10s. or 10.$. 6d. It is desirable likewise to have a low power objective, e.g., 2-inch or IJ-inch, though dissection with the 1-inch is perfectly simple. The lowest power eye-piece should be used. Dissection under the compound microscope has, with very small objects, this further advantage, that there is no chance of losing them in removing from one in- strument to the other. To this may be added perhaps another advantage, in that the working-table is not cumbered with an extra instrument. For dissecting with the microscope the wrists must be supported on a level with the object, or slightly below it. Some dissecting microscopes have arm-rests for this purpose ; blocks of wood of proper height, or even stacks of books will answer admirably. A very necessary adjunct for microscopical work is a good magnifying lens, as it is often desirable first of all to study an object with it, afterwards using the microscope. The lenses of the dissecting microscope can be used as hand magnifiers, and low power objectives likewise make good hand lenses. It is worth while, however, to get a lens magnifying about six diameters; very convenient are the triplets, three lenses in a tortoise-shell case, usable separately or together, and sold at a price of about 3s. 6d. Remarkably beautiful are the Platyscopic Lenses of Browning (63, Strand, London, W.C), magnifying 15, 20, or 30 diameters, price I85. 6d. each, and the Aplanatic Lenses of Zeiss, magnifying 6, 10, or 20 diameters, price 125. or 155. each. As it is desirable that the student should from the first begin to draw the objects he examines (practical instruction in which will be found on p. 30 et seq.), it is desirable that he should have some form of drawing instrument to facilitate his work. Drawing can, it is true, be done without any such aid, but is more difficult. An apparatus for drawing (camera lucida) is constructed either for use with the body of the microscope placed horizontally, or placed vertically. Practically the latter is much preferable. Every microscope maker has appliances of his own make, but they vary very much in real utility. Probably the best in existence are two made by Zeiss, the new camera lucida of Abbe, price 3O5., or the camera lucida with two prisms, price 2l5. The former is specially constructed for the eye-piece No. 2 of Zeiss, and is mounted upon it ; it permits drawing upon a horizontal surface ; during observa- tion it is removed. The second is slipped by means of a ring upon the tube or the eye- piece (any eye-piece of continental size) ; it XXU INTRODUCTION. requires an inclined surface for drawing, but has, however, the advantage that it is always kept upon the microscope, and during the observation is only pushed on one side. Both apparatus re- quire drawing desks, Abbe's camera a horizontal one, the drawing prism one inclined about 25°. The height of the desk should in general correspond with the height of the stage of the microscope. In specially long or shortsighted observers, it should be arranged according to the distance of distinct vision. Most English opticians supply drawing prisms of one kind or another, capable of satisfactory work. None, however, in my experience, equal those of Zeiss. A stage micrometer is likewise necessary. This can be obtained from most opticians at a cost of from 55. to 10s., and ruled up to „_!__ of an inch. Zeiss has one at IO5. ruled to y^o of a milli- metre, i.e., about -o 5V0 ^^ ^^ inch. Any steady table can be used by the microscopist for working, but it should be looked to that it is not too small, and not polished or varnished on its surface. This surface is best painted a dull, dark colour. The table is so placed that the microscope is about, or somewhat less than, two yards from the window. Any position of the window is good which allows a free outlook. From direct sunlight we protect ourselves by a white roller blind, which is best made of tracing-linen. The dazzling white light which we obtain when the direct sunlight plays npon the blind gives the most favourable conditions for observation with high powers. The necessary object slides and cover-glasses can be obtained of most opticians. The former are procurable with either ground or unground edges at a cost of about 4^. 6d. or 3s. 6d. per gross, re- spectively. They are three inches long by one broad. The cover- glasses for ordinary observation should be about |-inch square : but the observer should also have larger ones for specially large objects, and also others somewhat smaller (|-inch square) which will usually suffice for permanent preparations. If we use power- ful objectives, it will be best to obtain these cover-glasses of definite thickness. For the beginner, this is not of special import- ance ; but the more advanced student will find it advisable to procure both object- slides and cover- glasses of a definite thick- ness.* * The latter, curiously enough, are difficult to obtain in England, where, never- theless, they are mostly made, and I get them from P. Steuder, in Leipzig, Konigstrasse 11. They are 18 mm. square, lettered " C," at a price of 2s. 6d. INTRODUCTION. XXIU Further necessary are some plane- and some liollow-ground razors ; a fine and a coarse pair of steel forceps ; a finely pointed pair of dissecting scissors, for whicli fine embroidery scissors will serve; a pair of needle-holders, somewhat after the fashion of crochet needle-holders, but so arranged that they will hold the finest needles firmly ; English needles from No. 8 upwards, for these holders ; some scalpels, some fine painting brushes, a small vice, such as used by watchmakers; some pipettes, glass tubes, and glass rods ; watch glasses of various sizes, and glass disks of suit- able sizes for covering them ; low glass bell- jars (receivers), in order to be able to fit up moist chambers ; zinc frames, somewhat as represented in half-size in Fig. 1, on which to place the object- Fio. 1. slides under the bell-jars ;* two bell-jars of suitable height, under which to be able to place respectively the compound and the simple microscope ; and lastly, elder-pith. For working, a tumbler of clean spring water is needed ; a saucer is useful for dirty slides. The list of the necessary reagents is to be found at the end of this book. "Where the word " alcohol," or " sj^irit " and not " absolute alcohol," is used, strong methylated spirits can always be understood and is far cheaper. For the preservation of permanent preparations, many kinds of cabinets and cases are advertised. It is very important to re- per 100. They are in square boxes, are very light, and come readily by post. Smaller sizes can be obtained, e.g. 15 mm. square, lettered " C," at 2.s. per 100. These are all 010 mm. thick. At 0-15 mm. (B) thick these sizes are 2s. 3d. and Is. 9rf. respectively. * Slides can be also left upon these frames to dry, after permanent mount- ing. If the frames cannot be kept perfectly steady, the slides may wriggle olT in time; to prevent this, sheets of paper ^ inch wider than the frame can bi bent over them on each stage, and the slides laid across these. By using blotting paper for these sheets, keeping wet, the bell-jar can be converted into a convenient moist chamber for a number of slide cultures at room temperature. XXIV INTRODUCTION. member that tlie objects should be kept in a horizontal position, and should be capable of ready supervision. NOTES TO THE INTRODUCTION. ^ From the special point of view of botanists : Naegeli und Schwendener, '•Das Mikroskop," 2 Edit., 1877 -, Dippel, "Das Mikroskop," 2 Edit., 1882; and ♦• Grundziige der allgemeinen Mikroskopie," 1885 ; Behrens, " Hilsbuch im Botanischen Laboratorium," 1883. [Carpenter, " The Microscope," 6 Edit., 1881.] In changing the objective in use from high to low power, or vice versa, much time and inconvenience is spared by the use of a "Nose-piece." This is screwed into the end of the microscope- tube, where the objective is usually placed, and is provided with apertures into which two or more objectives can be screwed j and, by rotating these on a centre, any one can be brought into a line with the tube of the microscope. "She best and cheapest are those of Zeiss, for 2 objectives, 205. ; for 8 objectives, 27^. Both of these are constructed with the " English screw," THE COMPOUND MICROSCOPE. ^J^ I. USE OF THE MICEOSCOPE. STRUCTURE OF STARCH. Material Wanted. Potato, fresh. Potato starch, air-dry. Bean meal, air-dry. East Indian Arrowroot {Curcuma leucorrliiza). West Indian Arrowroot {Maranta). Grains of Wheat. Grains of Oat. Stem of the Sun Spurge {Euplwrhia helioscopia) and of E. splendens, fresh. (Other species can replace these if necessary.) We will first obtain information about the separate parts of the compound microscope (Fig. 2, p. 2), and for this purpose we select Stand No. VII. A of the manufacture of Zeiss of Jena * Upon this stand we distinguish the horse-shoe foot (/s), the supporting pillar (si), the stage (ot), the body or guiding sheath (fh), the tube (t), the mirror (s), and the micrometer screw (m). The mirror-frame (s) combines two miiTors, that on the one side plane, on the other side concave. The former we use with weak the latter with strong enlargement, or magnification. [The mirror- arm is usually hinged, and sometimes jointed, so that it can be placed obliquely below the stage for oblique illumination. The beginner should always see that the mirror is directly below the aperture in the stage.] The stage is pierced in its centre by a circular aperture, which is intended to give passage to the light reflected from the mirror. Under this opening are found the cylinder diaphragms. They are fixed in a carrier, which can * I retain this, with modificatious, as it matters Httle on what instrument (provided it is of simple construction) the mode of use is described. As this particular instrument is not, and is not likely to be, largely used in England, I have added some supplementary paragraphs on the students' microscopes more commonly in use here. [Ed.] B THE COMPOUND MICROSCOPE. be withdi^wn laterally from the stage, and in which can be set dia- phragms of various widths, provided with the instrument. With the help of these diaphragms we regulate the illumination accord- ing to necessity, a diaphragm with a small aperture allowing little Fig. 2.— Stand No. VII. A of Zeiss, with prism for di-awing, cl, one-third actual size ; fs, foot; sV, lower, si", upper part.of the pillar; ot, stage; ch, cylinder-diaphragms; fd, clips; s, mirror; m, micrometer screw [for fine adjustment] ; /7i, guiding sheath [for] t, tube; ob, objective; oc, eye-piece. light to pass through, and so on in proportion to the size of the opening. Some of the stands of the same maker have, in place of the cylinder-diaphragms, an arched excentrically fixed diaphragm HOW TO USE THE MICROSCOPE. 3 disk, wliicli is rotated in order to bring different sized apertures into the optical axis of the microscope. [This is the kind of diaphragm-wheel which is possessed hj most of the students' microscopes of English makers. Though not perhaps quite so good, it is more convenient in use. Best of all is what is called an " Iris diaphragm," with which, by simply moving a lever, the size of the aperture can be regulated at pleasure, and with the utmost nicety.] Upon the stage are clips (/^), which serve to keep the object in position during examination [and are parti- cularly necessary where the instrument is used in the sloping position]. If it is possible to do so, we will first remove these. The tube (t) is movable in its guiding-sheath (fh) [which is often lined with cloth to make the movement easier] . In larger stands the sheath is wanting, and the tube is raised and lowered by rack and pinion movement. [Most of the better makes of English student microscopes have this rack and pinion coarse adjustment, and, for a small sum, most of those which are without it can be provided with it. It is, however, a doubtful advantage for the learner. The chances of accident with its use are perhaps numeri- cally fewer, but when they do occur they are more serious.] We withdraw the tube from the sheath and screw into its lower end the weak objective, about B of Zeiss, 3 of Leitz [or half-inch of English make. This will vary with the microscope. As seen in the Introduction, the English microscopes are usually supplied with 1 inch and j inch objectives. A much preferable com- bination would be a f and ^ inch. In purchasing it would be easy to arrange this. In microscopes provided with rack and pinion movement, the tube is not withdrawn, but raised suffi- ciently above the stage to allow the objective to be screwed in]. The relative power of the objectives can always be told by the comparative sizes of the front lenses ; the weakest power has the largest lens. We now replace the tube in the sheath, and ap- proach the objective so near the stage that it is only removed from it by somewhere over a quarter of an inch. In the upper end of tlie tube we place the eye-piece, No. 2 [or whatever our lowest power eye-piece may be. This likewise may be judged by the size of the glass. English eye-pieces are usually lettered]. It is on the whole desirable to use for general purposes the lower (weaker) eye- piece of the instrument of any maker. The drawing prism found over the eye-piece in the figure we pass over for the present.. We place oar instrument opposite to a window, and at a distance of 4 HOW TO USE THE MICROSCOPE. about, or somewhat under, a couple of yards from it. While we now look down through the eye-piece, we change with the fingers the inclination of the mirror until the field of view of the micro- scope appears to us bright and equally illuminated. In this we have to take care that the mirror is not (as, for example, it looks in the figure) pushed forwards or laterally out of the axis of the instrument, as we propose to observe by direct (not oblique) illu- mination. On the other hand, in this stand [and in most English stands], we can, according as required by the strength of the light, slide the mirror on its bearer upwards or downwards in the optical axis of the instrument, thereby approaching it more nearly to the stage, or removing it therefrom. [The majority of English microscopes, instead of being supported on a single pillar (sV) as in Fig. 2, have the body swung between two uprights, between which it is hinged, much as in Eig. 81, in Chap. XXL hereafter. This gives greater possibilities from the point of view of illumina- tion, has other advantages, and in the large " English stands," properly so called, is a necessity for observation. The learner is, howeyer, strongly urged to learn to work with the instrument erect. The clips, then, are unnecessary for ordinary work. With a sloping stage, some appliance for keeping the object-slide in position is a necessity.] An object-slide is now wiped clean, and upon it, by means of a glass rod, a drop of spring water is placed. We will now commence with the investigation of a potato tuber. We cut this through with a pocket knife, and transfer a little of the sap which exudes from the cut surface into the drop of water by means of the same knife. We then cover the drop with a cover-glass. This must also have been previously cleaned with special care. It is done best flat between the fingers with pieces of old linen.* [The cover-glass must be laid on as care- fully as possible, so as to exclude air from underneath it. For laying on, it can be held between the index- finger above, and a * This operation is not so simple as it seems. If the cover-glasses are thin they are very readily broken. The method I have found least destructive for learners is to hold the cover-glass by its edges between thumb and index-finger of one hand. Having slightly damped the same fingers of the other hand make a fold in a piece of silk, with the damp fingers flat above and below it, slip the glafes horizontally between, and gently rub the silk-covered fingers to and fro over its two surfaces. The silk will cling to the slightly damp fingers, and the process becomes easy. Some use little pstds between which the cover- glass is placed, and the pads then moved about over its surfaces. [Ed.] HOW TO USE THE MICROSCOPE. O needle underneath it. By gradually withdrawing the latter when the cover touches the drop of water, it is lowered into its place.] If the drop is of proper size, no water will flow out from the side of the cover-glass. [The size of the drop has usually to be calcu- lated from the point of view of (1) the size of the cover-glass, and (2) the thickness of the preparation to be covered. Here the latter does not come into the calculation.] If water does flow out it can be removed wdth blotting-paper, or it is better to make a second preparation, as in this case most of the grains which we wish to observe will be sucked out by the blotting-paper. We now place our preparation on the stage of the microscope, so that the object lies over the centre of the stage- aperture. In order to focus correctly, we first slide the tube, carefully con- trolling its motion, so far downwards that it almost touches the object. Then, while at the same time looking through the eye- piece, we move the tube as slowly as possible upwards. This movement is best combined with a twisting of the tube inside the body-sheath. Soon the moment arrives when the previously invisible object begins to show itself in the form of small grains. If, on the other hand, we find we have removed the objective (object-glass) more than about |-inch from the object-slide, with- out having caught sight of the grains, these either do not lie in the field of view of the microscope, or we have raised the tube too quickly, and so overlooked the rapidly appearing and equally rapidly disappearing object. We must not then attempt by sliding the tube downwards to find the object, as thereby we should run into the danger of breaking the cover-glass, injuring the object, and destroying the objective (object-glass) ; instead, we a second time slide the carefully controlled tube so far down- wards that it almost touches the object-slide, and begin anew to raise the tube, more slowly than before, and at the same time looking through the eye-piece. If this also should not realize our purpose, it is to be assumed that the object does not lie in the field of view, and must be looked for again after altering the position of the object-slide. After a short time it w^ill happen in all cases that the grains appear in the field of view, and we then discontinue sliding the tube, i.e., what we call the coarse adjustment, and attain the fine adjustment which now is wanted by the aid of the micrometer-screw (rn, Fig. 2). This we turn in one direction, or, in case the object thereby is made more indis- tinct, in the opposite direction. The adjustment (focussing) is b HOW TO USE THE MICROSCOPE. perfect when the figure appears as sharp as possible. In onr example of a microscope stand (Fig. 2), the micrometer- screw is at the upper end of the pillar (si") ; but it can be variously placed according to the make of the instrument. In instruments of larger size, as in many English students' instruments, the coarse adjustment is not effected by hand and sliding tube, but by rackwork and pinion. After we have determined by slight magnification the existence of small grains in the field of view of the microscope, and have noted, for subsequent use, the distance of this weak objective from the object, i.e., its focal or working distance, we leave the object-slide unmoved upon the stage, but withdraw the tube from its guiding sheath, unscrew the weak objective and screw in a stronger one, in no case as yet however an immersion objective, but rather about D of Zeiss, No. 7 of Leitz, [or a |: or -^ inch of the English makers]. We then replace the tube in its sheath, and push it down so far that once more the objective almost touches the cover glass. We again endeavour to catch sight of the object by raising the tube in its sheath. With a stronger magnification it must however be withdrawn far more slowly than with the weaker. As the preparation has lain unmoved upon the stage we know it to be certain that the object will be found in the field of view of the microscope. When the grains have become visible with the coarse adjustment, we complete the fine focussing (adjustment) with the micrometer screw. We shall find that the working or focal distance of the stronger objec- tive is considerably less than that of the weaker one [and always less in proportion as the objective is stronger]. We now begin the actual observation. The learner should accustom himself, so far as his two eyes are equally good, to observe with his left eye. The right eye is thus kept free and can be used in drawing while he continues to observe with the left eye. Many of the drawing prisms and appliances for the microscope are moreover constructed for the left eye (as shown in Fig. 2) ; and those who work with the right eye should inti- mate it on ordering such drawing prisms. The learner should also keep open the eye which is not in use. At first the surround- ing objects which are figured on the retina of the eye will disturb him ; but he will soon overcome the difficulty of concentrating all his attention on the eye in observation, and entirely suspending the activity of the other. POTATO STARCH. 7 We readily recognise that the colourless bodies which occupy the field of view of the microscope are solid and show lamination. They are starch grains. We slowly move the object-slide here and there, in order to find a place where the grains do not lie too closely, because it is easier here to fix attention on a single grain. We select for persevering observation a grain which shows the lamination with special clearness. As the movement of the object-slide under the microscope appears to be reversed, we shall at first find some diflBculty when we wish to place a selected grain in the centre of the field of view ; and we shall have as quickly as possible to accustom ourselves to sufficiently control the slight movements upon which it depends. If we have found a single specially favourable grain, we magnify it still more by now removing the weak eye-piece and replacing it by a stronger. [Hold the tube of the microscope firmly while you do this, or the focussing may be altered, and the objective perhaps run down on the cover- glass.] With perfect objectives the figure always remains good, though in all cases the light diminishes. We endeavour by im- proving the position of the mirror as far as possible to obviate this inconvenience. Now and then, after focussing the preparation, or after moving it, it will happen that the figure has lost in clearness. In all probability this is because fluid from the preparation has got upon the under lens of the objective. This will happen especially easily when too large a quantity of fluid has been used, and has run out from under the edge of the cover-glass. We must then withdraw the tube from its sheath, and after having proved the supposition, wipe the front lens of the objective with a clean and often-washed piece of linen rag, or, still better, rub it with a freshly broken surface of a piece of elder pith. The starch grains of the potato tuber ^ attain a comparatively considerable size. They are excentrically constructed, as their organic middle point (c, in A, Fig. 3) is not the geometrical centre, but lies considerably nearer to one end. The layers appear variously sharp (A) ; between those more strongly marked can be seen others more weakly marked. Towards the surface of the grain the layering becomes indistinct. For optical reasons, and on account of its smaller density, the organic centre, or nucleus, appears rosy coloured. It shows up most clearly when it is hollowed. It then shows as a rosy point, as a line, cross, or star with dark outlines. The layers immediately surround- 8 AIE-BUBBLES. ing the nucleus are developed concentrically, soon however the excentricity has influence, in that the layers diminish in thick- ness towards one end of the grain, so as partly in this direction to run out into a wedge. At this more weakly developed end of the grain, which we can distinguish as the anterior end, the layering, on account of the small distance from the surface, is indistinct. The individual grains vary considerably in size, and moreover they deviate from one another in outer form to a not unimportant extent, and show the layering with various sharpness. Between the starch grains in most preparations will be found rounded bodies, which with median focussing show a small, round, bright centre and a broad, dark margin ; this last is black at its inner edge, A c B dark grey outwardly, and interrupted by a clear ring. These structures are air-bubbles enclosed in the fluid under observa- tion. Their appearance under the microscope is so characteristic that, once known, they can scarcely ever be confused with other appearances. The rays of light which pass out of the denser medium into the air-bubble are, with the exception of the central ones, so strongly refracted, that they cannot get into the objective, and hence the broad dark edge and the comparatively small clear middle. If, by turning the micrometer screw, the tube is lowered, so that the under part of the air-bubble comes into view, the sharp- ness and brightness of the middle disk increases ; it diminishes at the same time in size, while the breadth of the surrounding dark ring increases. If the screw is moved in the opposite direction, in order to focus upon the upper part of the air-bubble, the middle disk enlarges, but losing somewhat in brightness ; a grey ring of differing brightness arises around it; the surrounding edge be- comes simultaneously narrower. Fig. 3. — Starch grains from a potato tuber, A simple grain, B half-compound grain, C and D entirely compound grains, c the nucleus ( x 540). STRUCTURE OF STARCH-GRAINS. 9 If tlie observer has selected a beautifully laminated starch grain, it should now be drawn. The greatest possible stress is decidedly laid upon drawing in microscopical observation. With the help of it we in general first learn to see quickly. Then the peculiarities of the figure first become present to the mind of the observer, when he concentrates his attention upon it for the pur- pose of reproduction. Drawing therefore protects from transient superficial observation, enforces a penetrating, thorough study of the figure, and sharpens more than any other means our power of observation. The learner should first endeavour to represent the object by free-hand drawing. So much drawing ability as is necessary for this he may indeed possess, or can however readily obtain by practice the necessary facility. The object should not be drawn too small, even if the observer believes he sees it very small. A correct opinion on the size of the object in the field of view of the microscope is only obtained after long practice, and it is better at first that the learner should draw the object too large, in order conveniently to include in his figure all the details of the object. No less important is it to provide the individual parts of the figure with suitable distinguishing names ("terms "), and to note the name of the plant, the object, and the most important results of the observation. The starch-grains of the potato are somewhat flattened, as can be easily demonstrated if, during the observation, you press care- fully with a needle against the edge of the cover-glass, and so set the grains rolling. Upon the smallest grains the layering is usually but little recognisable. Besides the simple grains (as in A^ Fig. 3) will be found also, after some search, semi-compound grains (as in J5). These grains enclose two, rarely more, organic nuclei (or centres). Each nucleus is surrounded by a number of its own layers, both to- gether by a smaller or larger number of common layers. Not infrequently the two inner complexes of layers are separated by a cleft, extending to the common layers (J5). The number of layers peculiar to the individual grains, as well as of those common, varies according to circumstances. The completely compound grain, which is found far more commonly than those half-compound, consists of two (C), less frequently of three (D), rarely of more than three component grains. As a distinction from the semi-compound grains, the common layers are wanting in those quite compound. In the 10 STARCH-GRAINS. line connecting the nuclei of the component grains the layers are most strongly developed. The component grains therefore turn their posterior ends towards one another, their anterior ends away from one another. The line of separation between two component grains often broadens internally into a cleft. For comparison we now put up a preparation of potato starch which has been preserved in an air-dry state. We proceed in this quite similarly to the preparation of the first object, and transfer a trace of the meal into a drop of water. As the object- slides may differ in thickness, it is advisable to raise the tube of the microscope prior to placing under it the new preparation. [This is not of course necessary in using the low powers.] The first preparation, as it will be again required later, we place in a large moist chamber. This moist chamber consists of a deep plate and a glass bell-shade with knob. On the plate stands the zinc frame, which we discussed and figured in the Introduction (Fig. 1) ; so much water is also poured into the plate till the bell-shade has its lower edge quite immersed in it. The preparation is laid upon the frame. But first we assure our- selves that the drop of water under the cover-glass of the prepara- tion is not already partially dry. If this should have happened, we place at the edge of the cover-glass, so that it shall be sucked in, a new drop of water. We also mark the object-slide, and best with a coloured crayon which writes directly on the glass. Upon examination of the new preparation we shall find that the lamination of the air-dry starch is at least as sharp as that of the fresh. This preparation also we place in the moist chamber. We further make a preparation of air-dry bean flour (Phaseolus vulgaris). The grains (Fig. 4), examined in water, appear circular or oval ; they are a little flattened ; a certain medium size predominates. The lamination is very clear and very uniform ; the lamellae show almost equal thickness. The structure is con- FiQ. 4— starch-grains from the cotyle- centric. The nUclcuS of grains dons ot Phaseolus vulgaris (x 540). . , . . -in ^ examined m water is nollowed, more isodiametric in the rounded, elongated in the oval forms. From the nuclear hollow extend radial clefts, which cut through the layers at right angles, and, thinning off, reach almost to the periphery of the grain. STARCH-GRAINS. 11 We lay a trace of this bean-meal, in similar manner, in a drop of glycerine instead of in water. In this fluid the starch-grains appear sectionally smaller ; of lamination not a trace can be recognised ; the inner hollow and the clefts are wanting. These are formed under the influence of water, in which the bean- starch swells somewhat. The starch of the East Indian arrowroot (Curcuma leucorrhiza) is otherwise constructed. We put up a preparation of the commercial starch, which is usually not difficult to ob- tain. Genuine East-Indian arrowroot shows in its grains a very excentric structure (Fig. 5J.), at the anterior end strongly tapering, beautifully and regularly layered, and very flat. Often a considerable number of grains cling together by their flat sides, and, viewed from the edge, appear like rolls of coins (B) . The size and form of the grains varies not inconsiderably. The West-Indian arrowroot, also called in short Arrowroot, from the rhizome of Maranta, especially of Maranta arundinacea, is easy to obtain in shops, but gives however, from the point of view of its structure, much less interest than the East-Indian arrowroot. Observed in water, the grains show great similarity to the starch-grains of the potato ; only they are usually less clearly, and, in exchange, more uniformly layered ; somewhat more rounded ; on the whole smaller ; also more uniform in their size. At the position of the nucleus is usually found a cleft in the form of a wide open V. Wheat meal shows the layering very badly ; as relatively the most favour- able, we choose the starch-grains of Triticum durum for observation. We halve the Triticum grain with the pocket-knife, and scrape off a little substance from the cut surface, and put it in the drop on the object-slide. The large starch-grains are circular, discoidly flattened, and re- gularly laminated (Fig. 6 A), but the layers are usually hard to Fig. 5.— Starch-grains from the commercial East-Indian arrowroot (from the rhizome of Curcuma leucorrhiza). A, seen from the sur- face ; B, several grains clinging to one another, seen from the edges (x 540). Fig. 6.— Wheat-meal from TH- ticum durum. A, a large, B, small grains. 12 STARCH-GRAIN! O o A B 7. — Starch from Avena A, a compound grain ; B, its component grains (x 540). Fig saliva. see. In many grains they will always be easily recognised, as well as the central nucleus . As a characteristic appearance will be found in the preparation, besides the large starch-grains, almost without transition sizes, small grains, with clear rosy nucleus, but without recognisable lamination. A number of such grains are recognised at B. In many preparations compound grains are not altogether rare ; in most they are sought for in vain, as they have fallen into their component grains. The starch-grains of the oat (Arena sativa) we take as the best, inasmuch as we halve an oat-grain and take a little for observation under water. The compound grains here are met with in great beauty, such as is represented in the adjoining figure. The size of these compound grains varies, and proportion- ally also the number of the component grains entering into its structure. The Fig. 7 A represents such a compound grain of medium size. The individual component grains appear polygonal, separated from one another by clearer looking boundary lines. Between the great grains are seen small ones, down to such as consist of but two component grains ; lastly also quite simple ones ; besides also numerous angular grains (B) which arise from the large compound grains broken down in making the preparation. A medium size, somewhere about our Fig. A, is met with by far most commonly amongst the compound grains. The lamination in this object is not visible, the nucleus is only exceptionally indicated. Of quite peculiar appearance are the starch-grains in the latex (milk) of the ^k \-"\ il Euphorbiaceas. A piece of the stem of a ^^fev Wl 11 spurge is cut off, and the cut surface is «v X ^ \\\ w plunged in the drop of water which is ready upon the object-slide. The latex which flows out from the cut surface mingles with the drop. We can select for example the universally distributed Euphorbia helio- scopia [sun-spurge] for our investigation. In the latex, which appears distributed in small drops, like an emulsion, in water, we shall see isolated, small, rod-like bodies (Fig. 8). These are the starch-grains in Fig. 8. — Starch -grains from the latex of Eupliorhia helioscopia (x 540). REACTIONS OF STARCH. 13 question, Thej appear pretty strongly refractive ; a lamination is visible only in the most favourable cases ; sometimes a longi- tudinal cleft is recognisable in the interior of the grain. The size of the rods is somewhat variable, many of them are a little swollen in the middle. Much more beautifully formed grains of this kind are possessed by the tropical Euphorbiacese. We choose for this examination Euphorbia splendens, so commonly grown in plant houses, and make the preparation in the same way as stated above. The starch grains which now put in an appearance (Fig. 9) have the form of bones. [In the same latex wdll be found others shaped like rods, and still others with greatly enlarged ends, like dumb-bells] ; they appear more or less swollen at both ends, are somewhat longer than those of oiir native forms, and in the swollen parts permit something of the lamination to be recognised. Very commonly we see a colourless vesicle adhering to the sides of the grain (A), the walls of which, however, are referable, not to the substance of the starch- grain, but the plasma mass adhering to it. It must strike the observer that the small latex globules distributed in the water are in tremulous motion. This is the so-called t^ « „ , Fig. 9.— Starch-grains Brown's molecular movement [the " Brownian from the latex of Eu- movement"], which we can therefore take p''^'-^'^ «pf^»f^"«- One -"' of the grams has a vesi- tliis ojDportunity of learning to know, and cie attached laterally which, not a phenomenon of life, is referable perhaps to fine streams in the fluid carrying with them the minute bodies. After getting this information on the form and structure of the starch-grains, we will produce some reactions upon them, and study directly, under the microscope, the result of the action. "We take first a preparation of potato-starch again out of the moist chamber. After focussing we place a drop of a solution of iodine (iodine-water, alcohol-iodine, or tincture of iodine, or potassium-iodide iodine) at the edge of the cover-glass. In using the reagent we must take special care that the drop does not run upon the cover-glass and thence upon the objective. If a drop comes upon the cover-glass, let it be immediately sucked off with blotting paper. If the reagent reaches the objective, plunge the- lower lens of this latter in pure water, and clean it afterwards with the pieces of linen rag already recommended. 14 REACTIONS OF STARCH. In order to see the action of the iodine solution directly, await its penetration to a spot previously selected, this spot, however, being chosen not too far from that part of the edge of the cover- glass at which the reagent is placed, and follow by movement of the object-slide the progress of the action. We see, immediately the influence of the iodine solution begins to make itself felt, the starch-grains stain bright blue, and rapidly ever darker till they are black-blue. At the 6rst moment of the action the lamination shows up clearly, only immediately to disappear in the grain when it becomes opaque. With potassium-iodide iodine solution, in case this is added in considerable quantity, the action produces quickly a dark-brown coloration of the grain. Similarly dry starch-grains, which are exposed to the action of iodine vapour, become deep dark-brown. If we add water to such a preparation the brown passes rapidly into blue. If the action of the reagent should not proceed rapidly enough under the cover-glass, it can be readily accelerated by fragments of blotting paper placed at the opposite end of the cover-glass. We should stain with iodine solution the rod, etc., shaped grains of the Euphorhia also, in order to demonstrate that, in spite of their variable form and of their scarcely noticeable lamination, these bodies are true starch- grains. Let us further study the phenomena of the swelling of starch- grains under the influence of potash (potassium hydrate). First we again take potato-starch, and await the entrance of the re- agent, j^laced at the edge of the cover- glass. The action of this must take place quite gradually, if it is to be instructive. We then notice, at the first moment of the action, that the lamination stands out more clearly, quickly, however, to disappear, while the grain increases in size. During this enlargement, which proceeds with more or less regularity, the nucleus of the starch-grain hollows considerably, upon which the wall of the weaker side, therefore towards the anterior end of the grain, sinks into the hollow. Later on the regularity of the phenomena disappears altogether, and the grain enlarges to a mass as clear as glass, of considerable volume, the limits of which are scarcely distinguish- able. Finally, we can endeavour by warming the preparation to cause the starch to swell, a treatment such as indeed is in use in the preparation of paste. The preparation is warmed over a spirit or gas flame, without allowing it to boil, and taking care to replace REACTIONS OF STARCH. 15 the evaporated water by fresh. If in warming a temperature of about 70° C [approx. 160 F] is reached, the grains will be found to be swollen just as in treatment with potash. [If it is wished to determine accurately the temperature at which swelling ensues, the warming of the preparation must be effected upon a special table which can be heated, and its temperature registered. Such a table by Ranvier,^ can be specially recommended.] With this we close our first Lesson.* Before we put the micro- scope on one side we carefully clean, in the manner before de- scribed, the objectives and eye-glasses [together with any other pieces of apparatus] that we have used. We withdraw the microscope tube from its sheath in order to rub it, and also the interior of the sheath, with a rough towel. Instead of again replacing the microscope in its cabinet, we prefer to place it under a glass bell-jar, which latter, in order to protect the instrument as much as possible from dust, can have its lower edge covered with felt. NOTES TO CHAPTER I. ^ Compare herewith Naegeli, Die Stdrkekorner, in Pfianzenphys. Untersuchun- gen, Heft 2 ; E. Strasburger, Bau und Wachsthum der ZellMute, p. 107, where the further literature will be found. 2 Eanvier, Traite d'Histologie, p. 41. 1875. * The chapters in the original are called " tasks," or " lessons. 16 HOW TO CUT SECTIONS. II. ALEURONE-GKAINS, PROTEIN CRYSTALS, FAT OIL, MOUNTING OF PERMANENT PREPARATIONS, USE OF THE SIMPLE MICROSCOPE. Material Wanted. Dried Peas. Grains of Wheat. Seeds of Lupine (Lupinus). Seeds of Castor-oil {Bicmus communis). Brazil INuts {Bertholletia excelsa). We examine, first of all, the Pea (Tisum sativum). A ripe seed is halved by a sharp pocket-knife, and in such a manner that the two cotyledons (seed-leaves) are cut across. Take then from the cut surface a thin cross section with a sharp, hollow-ground razor. On the subject of section-cutting with the razor the fol- lowing points can be noted : — 1. The cut surface is to be moistened before cutting the section, most commonly with water, though in this case with glycerine, since the preparation suffers from water, and we shall observe it in glycerine. 2. The first section is not to be used, as here the tissue would be too much injured by the pocket-knife. 3. In such resistant tissues as that of the pea only very small and exceedingly thin sections ought to be taken, as the edge of the razor would be very easily notched. If the razor has gone too deeply into the tissue, and it is seen that the resist- ance to its progress increases, it is better to withdraw the razor, in- stead of forcing it to the end of its cut. 4. Unless the investigation requii'es it, it is advisable not to commence the section with the outer surfacQ of the object, but rather to lay the razor on the cut surface, as thus a far firmer support is obtained in order to get a thin section. 5. In order to get a really good section, that is one in which the individual elements of the tissue are not torn, the razor must not merely be pressed with its edge against the object, but at the same time drawn across it. It is well, therefore, in order to cut as freely as possible, to accustom yourself not to rest the MANIPULATION OF SECTIONS. 17 thumb of the cutting hand upon the other hand. Instead of this, both hands can with advantage be rested against the breast, be- cause thereby lateral movement of the cutting hand is not hin- dered. The back of the razor should be supported on the index finger of the hand supporting the object. 6. As it is difficult to hold so small an object as a half pea, especially when it is also so hard, sufficiently firmly between the fingers, it is recommended to use for the purpose the small hand- vice described in the Introduction. The half pea is therefore fixed sufficiently deeply in this. 7. It is not advisable to be satisfied with a single section, but to take a considerable number, in order to make choice of the best. The section selected should be observed in glycerine, either concentrated or diluted with one-third distilled water. Pure water is not available for this, because it quickly sets up appear- ances of disorganization in the ground substance of the cells. The transfer of the section from the razor to the glass slide is best made with a fine camel-hair brush. The section is removed by pressing the brush upon it and sliding it off from the blade. If it adheres to a sufficiently broad surface of the brush, rolling up (" curling ") of the section will be prevented ; curling occurs very easily, on the other hand, if the section is taken directly by its edge with the tweezers and so transferred. The section adhering to the brush is immersed flat in the drop on the glass slide, and the brush withdrawn laterally with a simultaneous twisting move- ment. If it is desired to turn a section over when on the object- slide, the brush can be pressed down on the object- slide so that it is in contact with the edge of the section, and then begin to turn it over away from the section. In this way the section will be very easily drawn upon the upper surface of the brush, and can then be turned over with it. Other similar tricks will soon be acquired in practice. After every time of use the brush must be most carefully washed in water. Examine the section of pea with a strong magnifying power. It proves to be a tissue composed of rounded cells. At the places where three such cells adjoin one another a triangular intercellular space (i) filled with air, is present. The air appears black, like the edge of the air bubbles previously described ; here it naturally must show the form of the space, since it fills it. The wall of the cells (m) is pretty thick. In the adjoining figure the three middle cells are completely, the surrounding ones only partially, repre- sented. In each cell can be seen the large starch-grains (ain), and c 18 ALEURONE-GRAINS OF PEA. witli some care also the small grains (al) whicli lie between them. These grains are, for their part, imbedded in a very finely granular ground-substance (p). From thin parts of the section many a starch grain will have fallen out ; a hollow of similar form and size in the granular mass will indicate these places. The small grains are Aleurone or Protein-grains ^ ; they lie in the ground substance of the cell. If we run iodine solution into the prepara- tion, the coloration which ensues gives us immediate information as to the individual constituents of the cells. The drop of iodine solution is placed at the edge of the cover-glass ; as, however, the iodine solution diffuses very slowly in the glycerine, and it is not our present purpose to study the progress of the reaction, we accelerate it a little by slii the edg-e :htly raising .^^ ^^^. of the cover-glass with a needle, and so permit the mixture of the iodine vrith the glycerine. A second needle placed at the same time against the opposite edge of the cover- glass pre- vents it from slipping. The starch-grains colour blue to violet ; the aleurone-grains and the ground-substance yellow. By the use of potas- sium-iodide iodine the color- ation of the aleurone-grains and ground- substance be- comes very intense ; but the starch-grains are at the same time over-coloured, and appear then black-brown. If sections of pea are laid in a drop of borax-carmine solution, in an exceedingly short time the ground-substance, and also almost simultaneously the aleurone grains, colours dark-red ; the starch-grains remain colourless. The reaction becomes especially striking if, after the section is thoroughly soaked in the carmine solution, this is re- placed by dilute glycerine or by water. This is done by sucking out the carmine solution by a piece of blotting-paper placed at the edge of the cover-glass, while at the same time the water or dilute glycerine is run in under the opposite edge. If a section is placed in Millon'a reagent, the starch-grains swell very strongly, and Fig. 10.— From the cotyledons of the Pea. m, cell wall ; i, intercellular space ; am, starch ; al, aleurone grains ; p, ground substance ; n, nucleus. ALEURONE- GRAINS OF WHEAT. 19 become unrecognisable ; aleurone and ground-substance are im- mediately disorganized; the disorganized mass liowever, after some time, takes on a characteristic brick-red colour. If still another section is laid in acetic ized methyl-green, after a short time there appears in each cell, between the other constituents, a greenish-blue spot of rather indefinite outline. This spot is the Nucleus (n) . The other constituents of the cell have not stained ; the starch-grains are just a little swollen (they show radial clefts, which are wanting in glycerine) , and the aleurone-grains also have increased in size, and appear as if porous or even hollow. We recognise therefore in aceticized methyl-green a reagent which in the present case recommends itself as a special staining material for the nucleus. Simultaneously, it is true, the cell- walls also stain, but this does not injure the value of aceticized methyl-green as a reagent for nucleus staining. The cell-walls appear of a beautiful bright blue colour, and, as the result of this, are traced out in the glycerine preparations much more readily than before. The intercellular spaces also stand out more sharply. In the yellow-brown iodine reaction, the accumulation of colour materials, and the brick-red from Millon's reagent, we have learned to know the most important means whereby to recoo-nise under the microscope albu7)iinous bodies, for to these belong aleu- rone-grains as well as pr6toplasm andnucleus. Protoplasm, as will be seen again later, shows these reactions first when it is dead ; in this case death results from the action of the re- agents. The substance of the nucleus shows a specially strong affinity for the colour materials. A grain of wheat {Triticu'ni vulgare) can be recommended as a second object of investigation. The grain is first halved (across) with a pocket knife, then one half fixed in a small vice in order to Fig. 11.— Cross section through a grain of whca (Trilicum vulgare). p, pericarp of fruit ; t, testa of seed. In the endosperm cells succeeding to these : al, aleu- rone-grains ; am, starch-grains; n, nucleus (x 240). 20 HOW TO MOUNT A SECTION. have sections taken from it. This time it is desirable so to take the sections that a piece of the skin also is represented on them. In cutting, moisten the cutting surface with glycerine, and ob- serve the object in the same fluid (Fig. 11). Under the skin, formed of cells pressed closely together and dead (p), which represents the combined skin of the fruit (pericarp) and of the seed (testa), lies a layer of rectangular cells, which are thickly filled with small aleurone-grains (al). The aleurone-grains are embedded in a finely granular ground-substance. Then follow elongated, less regular cells, which contain large and small starch- grains. This is not difficult to determine with suitable reactions. We will now " mount " a successful section of the wheat-grain, and by this means learn how to put up a permanent preparation [or, to use the common phrase, how to permanently "mount" a preparation]. We will employ first the simplest method of pre- paration, which is here so much the more desirable, as it gives a very favourable result : we enclose the section in glycerine-jelly. Place upon the glass slide so much of this jelly-like substance as we believe will suffice to form a drop. Then warm the glass slide slowly over the flame of a spirit-lamp, till the jelly has become fluid. The section is then laid in the drop, and a cover-glass placed over it. It is advisable first to warm the coTcr-glass a little, as otherwise air-bubbles will easily remain in the prepara- tion, and for similar reasons it is desirable not to place the cover- glass on quite horizontally, but with a slight lateral movement. If, in spite of this, air-bubbles are enclosed, the glass slide can be warmed a little, and by careful raising of the cover-glass endea- vour to bring the air-bubbles to one side. If the air-bubbles are not troublesome, the task of removing them can be given up. If several sections are placed in the same drop thej should be uni- formly dispersed in it. Truly it often happens that, in laying the cover-glass upon them, the sections come into contact with one another, and even lie upon one another. If the cover-glass is raised on one side to secure order, the contrary to this is often produced. Another^comparatively simple method is therefore em- ployed. By warming the glass slide, make the drop as fluid as possible, and then, without lifting the cover-glass, pass in a hair from one side. With this hair seek out the object to be rectified, an operation which usually tends to succeed. Before covering with the cover-glass it is, above all, necessary to make sure that no particles whatever of dust have found access to the drop of THE SIMPLE MICROSCOPE 21 gljcerine-jelly ; any such should be removed with the needle. As these manipulations can only be carried on with a suitable magnification, this is at the same time the moment to learn the use of the simple microscope in connection with the methods of preparation under the compound microscope. I assume in the first place that the observer has at his disposal a small dissecting microscope (compare Introduction, p. viii.) , either as Fio'. 12, or some other of like construction. Over the stage (ot) Fio. 12.-Small dissecting microscope of Zeiss, on foot, two-thirds natural «i^e 0/ stage d. lens, sheathing toothed support for lens-arm ; sr. screw for fine adjustment ; s, mirror ]>' wooden supports for hands in dissecting, etc. of this small dissecting microscope (Fig. 12) is placed a lens (d) borne on a horizontal arm. The horizontal arm is fixed to a steel upright (st), which can be moved up and down inside a tube. By this movement is brought about the coarse adjustment. The fine adjustment is effected on the other hand by turning the screw (sr). The instrument is screwed into a dissecting foot, the high ends of which (p) serve as resting-places for the hands in the 22 THE SIMPLE MICROSCOPE. processes of preparation or dissection. The instrument is pro- vided with two, or with three lenses, magnifying 15, 30, and 60 diameters, and it is an advantage also to have lenses magnifymg live and ten fold. The larger dissecting microscope of Zeiss (comp. Introduction), Fio. 13.— Large dissecting microscope (Zeiss), half natural size, ot, stage ; p, wmgs as arm rests ; sr, screw- head for adjustment ; I, system of lenses, of which oh is the objective, oc the eye-piece. Upon the stage is an object-slide fixed with the clips. or other of similar construction, has also a system of lenses (I, Fig. 13), consisting of three achromatic lenses, which can be combined into an objective (oh), a tube, and an achromatic eye-piece. In order to work with slighter magnification, the objective can be USE OF THE SIMPLE MICROSCOPE. 23 used alone as a lens, the eye-piece, together with the tube, being unscrewed. The three lenses of the objective can also be un- screwed from one another, and the upper lens alone can be used, the two upper, or the three simultaneously. Magnification of 15, 20, and 30 diameters can be thus obtained. The adjustment is completed by turning the screw-head (sr). On both sides of the stage (at) "wings" (p) are fixed, to serve as hand supports in dissection.* In order to prepare or to dissect with the compound microscope, what is called an " erecting eye-piece " can be used in the place of the ordinary eye-piece of the microscope. This " erecting " eye- piece reverses the image of the object ; and as, in a compound microscope, the image is normally upside down, it is thus rectified. It is, however, quite possible, though to a beginner very difficult, to dissect, etc., with the ordinary compound microscope. With prac- tice one comes to realize that every movement is reversed, and to govern the movements accordingly. The low powers can then be freely used for dissection and preparation. In dissection, etc., with the compound microscope it is of advantage to have two blocks of wood of suitable size, which can be placed on either side of the stage, and will serve to support the hands.f Whichever of these instruments is used for preparation, we first lay the preparation on its stage, that w^e may free it from any foreign bodies which may happen to be present. For this purpose the lowest magnification that is at our disposal is used. This, in the larger microscope for preparation, of Zeiss (Fig. 13), is fifteen diameters. The distance of the object [from the lens] w^ould then be about Ij inch. With this instrument, even with the strongest magnification, viz., 100 diam., this distance is more than ^ inch After proper adjustment of the mirror (s) and of the image, take in each hand a needle fixed in a holder (see Introduction), steady the hands on the rests, bring the points of the needles into the axis of the instrument, and endeavour to see both simultaneously in the field of view of the microscope. This will soon be success- * I have retained the above descriptions intact, as they ilhistrate pretty fully the structure of dissecting microscopes in general. For an account of other instruments by English makers, see the Introduction. In choosing an instru- ment I would specially urge the importance of stable arm-rests, as in the above Fig. 12. An instrument satisfying all the requirements of even more than the beginner ought not to cost more than about 30s. [Ed.] t This remark equally applies to those forms of simple (or dissecting) micro- Bcopes which are unprovided with rests. [Ed.] 24 ALEURONE-GEAINS OF LUPINE. fully accomplished, and then by means of a few experimental attempts learn how to make the necessary slight movements with the needles. This easy problem of removing the foreign bodies out of the preparation with the points of needles will soon be com- pleted to our satisfaction, wherenpon we proceed to lay the cover- i^^lass upon the drop of fluid. If this in the meantime shall have become too viscid, it can be again warmed before being covered. The glycerine-jelly preparations need no further enclosing, are c^^ therefore prepared in the simplest possible way ; and as most vege- table objects, even stained ones, preserve very well in glycerine- jelly, we can recommend this method in preference to others. The preparation must then be labelled, preferably at both ends of the glass slide [with gummed circles or squares of paper], npon which must be written at least the name of the plant, the nature of the object, the direction of the section, if it be one, the medium in which it is preserved, any staining material used, and the date. If it is desired to keep the preparation slides stacked on the top of one another, then they must be protected from contact by card- board labels in the place of those of paper. The cardboard labels should be cut the breadth of the glass slide, by about -|-inch in the other direction. On these the information, as above, can be written. The card-labels are best fixed on with " Crystal Palace Cement," or other similar medium [or they can be fastened with Canada balsam dissolved in turpentine]. If it is necessary to fasten them with gum, it is best to cover each end of the slide first with a strip of gummed paper, the ends of which shall fold over and overlap under the slide, and fasten the card label on these ; otherwise the label would easily spring away from the slide. Take now the seed of the white Lupine {Lupinus albus), or other allied species. Once more halve the seed across, and take sections from the moistened cut surface. Sections observed in water show in the cells rounded aleurone-grains with vacuoles. In order to see the grains in their natural form they must be observed in gly- cerine. The grains then appear at first refractive, angular, gradually forming in their interior a fine network, granular. Lying closely together they fill up the cell ; a small quantity of ground-substance lies between them, more ground substance can be observed against the walls of the cells. The walls of the cells are very strongly thickened and pitted, a structure which we shall, however, study later on a more favourable object. In iodine-glycerine the grains take a beautiful golden-yellow colour. ALEURONE-GKAIXS AND PROTEIX-CRYSTALS OF RICINUS. 25 In the next place remove the shell-like testa from the seed of the castor oil plant (Bicinus communis) , cut it through across, and make preparations just as above from it. The tissue of the endo- sperm is capital material to cut ; it contains very much fat oil, and need not be moistened. The sections can be observed in water, the disturbing effects of which, by removal of oil, come but gradually into operation. The aleurone-grains, imbedded in a ground-sub- stance very rich in fat (Fig. 14, A), enclose in their interior usually one, but sometimes two or more. Protein- crystals [or so-called crystalloids], and usually a single rounded body, the Globoid, which is of inorganic composition, the combination of double phosphoric acid with lime and magnesia. With longer action of water the ground-substance in which the aleurone-grains lie is disorganized ; great masses of oil collect around and on the object. These cling partly to the object and the glass, and have an irregular form, partly lie free, and then are globular. They are mostly clouded with numerous vacuoles. If the microscope is adjusted so as to show an optical section of such an oil drop, it appears bright grey, and is surrounded by a narrow black limiting zone. If the tube of the microscope is lowered, the dark ring disappears; the disk appears somewhat more brightly surrounded. If the tube is raised, the dark zone, which in the mid-position of the tube is narrow, becomes broader. Oil-drops show, therefore, reverse appearances to those which have been previously observed in air- bubbles. Air refracts light less, oil more strongly, than water ; hence their opposite relations. These relations should be noted for future observation. Bodies which are less refractive than the medium in which they are observed, have an inner brighter part which, with deeper focussing, is so much the smaller, an outer darker part which is so much the broader ; with more strongly refractive bodies these relations are exactly reversed. If we run absolute alcohol under the cover-glass of the pi-epar- ation of Uicinus, which is at present in water, the preparation will y Fig. IJi. — From the endo^^perm of Riclnus com- munis. A, a cell of the endosperm with its contents, viewed iu water; B, single aleurone-grains seen in olive oil ; g, the globoid ; fc, the protein-crjstal (x540). 26 REACTIONS FOE OILS. " clear " somewliat ; and simultaneously the protein-crystals in the aleurone-grains come out very sharply. They are now so clearly defined that this method of manipulation is recommended in order to study their form, — hemihedral tetrahedra of the regular system. 2 After longer action of the alcohol, the oil-drops ,dis- appear more and more, as castor oil, in contradistinction to other fat oils, is miscible with absolute alcohol. Now make another preparation of Ricinus seed, lay it on the glass-slide in a drop of glacial acetic acid, and cover it with a cover-glass. The protein-crystals swell and disappear in the aleurone-grains. These latter increase considerably in volume, the globoids also enlarge, and show up very clearly in each aleurone-grain. Drops of fat are, however, not visible, because castor oil, again acting as an exception, mixes with glacial acetic acid. Otherwise absolute alcohol and glacial acetic acid, because normally they either not at all or but slightly dissolve fat oils, while on the other hand they are solvents of ethereal oils, are the best reagents for the purpose of distinguishing between these two classes of oil under the microscope. Of ethereal oils, the terpene dissolve somewhat less easily than the others in both the above reagents. Chloroform and ether dissolve fat and ethereal oils equally. To a preparation mounted in water run in alcanna (alkanet) tincture diluted with water. The fat masses soon accumulate colour and stain reddish brown, a reaction which ethereal oils and also resin alike show. Logwood (Hsematoxylin) added in small quantity to a prepara- tion in glycerine, stains the protein-crystals a beautiful violet. In olive oil the protein-crystals are not visible ; the whole grain on the other hand appears a strongly refractive, rounded body, at one of the ends of which the globoid simulates the appearance of a vacuole (Fig. 14, B). The protein-crystals come out very beauti- fully if the section is laid in a drop of 1% osmic acid ; they gradu- ally take on a brownish tint. By the same reagent the oil is slowly blackened, a peculiarity which fat oils have in common with ethereal oils ; this reaction is, however, not characteristic, as many other organic substances become black in osmic acid. Protein -crystals of extraordinary beauty, which show readily all the characteristic protein reactions, are to be found in the endosperm of the seeds of BerthoUetia excelsa, the well-known " Brazil nut." In this also the sections are exceedingly easy to obtain. If to a preparation laid in water is added absolute alcohol. PROTEIN-CRYSTALS OF BRAZIL-NUT. 27 the protein-crjstals come out very sharply. The fat oil is not touched to any extent by the alcohol. It remains unchanged also in glacial acetic acid, while the protein-crystals are immediately dissolved. In 1% osmic acid the crystals become very distinct. These crystals are so large, that their form can be made out even by comparatively smaller magnification. Near the crystal lies a globoid, this latter being here always in the form of an irregular aggregation of rounded bodies. The ground-substance is very rich in fat, and with 1% osmic acid becomes everywhere quite black. The granular contents of the aleurone-grain also take on quickly a dark coloration, while the crystals themselves colour slowly yel- low. The crystals are optically uniaxial. NOTES TO CHAPTEE IL ' Compare Pfeffer, Jahrb. fiir wiss. Botanik, viii. p. 429, ^\-liere the other literature will be. found. 2 Schimper, Unters. ii. d. ProteinJirystalle d. Pjianzen. Inaugural Dissertation, Strasburg, 1878. 28 MOVEMENTS OF PROTOPLASM. CHAPTER III. MOVEMENTS OF PROTOPLASM ; NUCLEUS. DEAWING WITH THE CAMERA, ETC.; CALCULATION OF MAGNIFICATION. Material Wanted. Flowers of Tradescantia (best T. virginica). Fresh. Or, very young shoot of a Gacarhita (gourd, pumpkin, cucumber, vegetable marrow, etc.). Young roots of the Frog-bit {Rydrocliaris morsus -ranee). Quite fresh. Strong, oldish, leaves of ValUsneria spiralis. Fresh. Young parts of Nitella. Fresh. We will first study now the phenomena of the movement of living protoplasm, and select as one of the most favourable objects for this purpose the hairs on the staminal filaments of Tradescantia (the Spider- wort). Tradescantia virginica, and other closely- allied species, are cultivated in every botanical garden, and flower from May or June till late into autumn. The long violet hairs in every flower will at once strike the eye. For observation, select hairs out of a flower which is either just opening or has just opened. The preparation is made by seizing a tuft of hairs at the base with the forceps ; remove them and lay them in water. Or the whole filament can be placed under a cover glass if the anther is previously removed. In this last case the masses of air clinging amongst the hairs will give trouble, and it- takes some pains to remove them. This is best effected by means of a 'fine camel-hair brush, with which the hairs are brushed over from below upwards, the tuft being at the same time held firmly at the base. After this the cover glass is laid on. Most of the hairs will not have suffered, provided the air has been removed with sufficient carefulness. The hairs in question are formed of numerous cells, swollen into a barrel form, and arranged into an unbranched row. At the points of constriction lie the partition walls which separate the CIRCULATION IN HAIR OF TRADESCANTIA. ;A individual cells from one another. Each cell (Fig. 15) shows a thin continuous lining layer [" peripheral layer "] of protoplasm, and is traversed in the interior by numerous thinner and thicker protoplasmic strands. Suspended within these strands is to be found the nucleus, surrounded by an enveloping layer of proto- plasm. (Shown somewhat below the middle in the figure.) The cell cavity in which the nucleus is suspended, and which is traversed by the protoplasmic strands, is filled by a violet-coloured cell sap. [It is the vacuole.] The protoplasm consists in a colourless, viscous, semi-fluid substance, which is distinguished by the name of Hyaloplasm [i.e., clear plasma], and which contains numerous minute granules, called by the name of Microsomata, or Microsomes. Besides these there can also be seen in the protoplasm, in greater or less number, somewhat larger, highly refractive bodies, which appear somewhat bluish in colour, and which will be designated by the terms Starch-formers, Starch- builders, or Leucoplasts. If we focus the object- glass from the peripheral jDrotoplasm inwards, it will be seen that this is not in movement as a whole, but that rather the fine, net-like, anasto- mosing, protoplasmic strands flow into and away from it. In the protoplasmic threads which sur- round the nucleus the movement is especially strong. These streams are of various thickness, they anastomose laterally with one another more or less frequently, and the nucleus manifestly furnishes a central point for them. Most of the threads end in the plasma layer surrounding the nucleus. The current in a single strand moves often only in one direction ; often, however, it can be seen that even in very thin strands or threads there are two currents in opposite directions. The movement is recognisable by the microsomes and leucoplasts borne in the clear basal hyaloplasm. With continued observa- tion it will be seen that the strands slowly change their thickness, arrangement, and other conformation. New branches of the system can be seen to arise, others can become constantly thinner in the middle, finally break through and withdraw into other strands. Thus by degrees the figure changes. The nucleus is almost globular, in many cases oval or somewhat flattened. AVith Fig. 15.— a cell from the hair on the filament of Tradescantia vir- ginica ( x 210). 30 THE CAMERA LUCIDA. the strongest magnification which is at our command it appears finely punctate, and in it can he readily distinguished some larger granules (Nucleoli). Often two nuclei lie close together in such a cell, because the original nucleus has divided. The nucleus is towed about hither and thither by the plasma strands, and thus slowly changes its position in the cell. In order to prove this, take rapidly a sketch of the cell, and compare this with the arrangement of the nucleus and the currents after the lapse of some time. Such a sketch can only be accurately taken by means of a drawing prism, and it alone has definite value for later com- parison. We will, therefore, endeavour here to become acquainted with the use of the drawing prism. The camera lucida of Abbe recommended first of all in the Introduction, which is represented in ideal longitudinal section Fig. 16.— Camera lucida of Abbe, nat. size. Ideal longitudinal section. The course of the rays of light indicated by the dotted lines ; o, the position of the eye ; s, the direction of the surface for drawing upon ; sr, clamping screw. in Fig. 16, is, as shown in the figure, placed upon the eye-piece and fastened with the clamping screw shown at its side (sr). It is best to remove the eye-piece before screwing on the camera, as in the performance of this manipulation upon the microscope there is the danger that the tube may be pressed down, and the preparation crushed. When the eye-piece with the camera is placed in the tube, then, in case we use the microscope with the left eye, the mirror of the camera should be placed in front ; but in case of use of the right eye, to the right hand, and inclined about 45°, in the manner shown in the figure. If now we look through the camera in the direction of the eye-piece, we see once more the figure of the object in the field of view of the microscope. I^ow place in front of, or to the right side of, the microscope a THE DRAWING PRISM. 31 horizontal drawing desk, this being thereabouts the height of the stage of the microscope. Lay a sheet of drawing-paper upon tliis desk, and rest the point of a lead pencil against it. If the point of the pencil is found under the mirror in the direction of s, this must now be visible in the field of view of the microscope at the same time with the figure of the object. The point of the pencil is, however, made visible by double reflection, the first time in the large mirror, the second time in the silvered surface of a small prism in the point of sight of the eye-piece (compare the figure) while the microscopic figure comes directly to the eye through a small opening in the silvering of this prism. If the surface of the drawing desk does not lie in the distinct visual distance of the observer, the point of the pencil will be seen indistinctly, and the drawing desk must be raised, or, though seldom, be made lower. We test the necessary height by means of books laid one upon the other. The microscopic figui^e is only well visible on the drawing surface when a definite relation of brightness exists between the two. Dimming of the drawing surface can be pro- duced by the aid of smoked glasses, which are made to turn on the camera. If the arrangement is perfect you can draw with the lead pencil the outline of the object as if drawing it in the field of view of the microscope. The second camera mentioned in the Introduction is seen in Fig, 2, set upon the microscope in the position for di-awing. This camera has the advantage that it can always be kept on the instrument, and with some practice will perform yeoman's service. It consists of two prisms, inclined to one another, in a common setting. The rays coming from the pencil take, after double reflection inside the prisms, a course parallel to the axis of the microscope, and thus coincide with the rays coming direct from the object. The camera is placed in the inclination represented in the figure, and so placed that its anterior edge, visible through the opening in the setting, approximately bisects the " pupil " of the emerging rays of the microscope, i.e., the bright circular disk which we notice when we look perpendicularly into the eye-piece from [a short distance, such as IJ inch] above it. If, then, on moving the head to one side we do not see the " pupil " notably displaced towards the edge of the prism, this latter stands also at the right height. We draw upon a sloping drawing desk, which is placed in front of the microscope. If, after bome attempts, we have found the point of the lead pencil upon the 32 HOW TO DRAW AN OBJECT. drawing paper, we can now follow with it the outlines of the object. If the object is not to be distorted in drawing, the draw- ing desk must have the correct inclination. In order to determine this, we use a method of procedure which quickly leads us to our end. We draw the circular outline of the field of view upon the paper with the aid of our camera, and obtain thus, if the inclina- tion of the drawing desk is correct, likewise a circle [i.e., the cross measurements of the figure from side to side and from top to bottom of the sloped surface will be like] ; if, on the other hand, we have an ellipse, the slope of the drawing desk is not correct, and must be varied until a circle is produced. Or, we set in position, and always with strong magnification, the stage micro- meter recommended in the Introduction, i.e., a millimeter divided into 100 parts, engraved upon an object slide. AYe now turn the stage micrometer around through 90°, so that the engraved lines shall run from side to side, and succeed one another fore and aft. In case the too small size of the stage does not permit such a position of the stage micrometer, we must change the position of the microscope 90°. The turning of the microscope naturally renders necessary a change of the direction of the mirror. If our instrument is provided with a " concentric rotating stage," or similar appliance, then it is only necessary to turn this ; such a stage is very useful for drawing, as it enables us to place the object in the desired position. If we have given the micrometer its proper position, w^e draw, with the help of the camera, its lines upon the paper on the drawing desk. The lines follow one another up the slope of the desk. We shall succeed, without much practice, in reproducing it exactly ; but, as the lines have a definite thickness, it is necessary that we should represent always a definite edge of the line. The inclination of the drawing desk is correct when the distance apart of the lines remains the same at all heights. If this distance increases upwards on the desk, the desk must be made steeper ; if it sinks, it must be placed in a less-inclined position. As, for the rest, small mistakes are not excluded from our measuring scale, it is necessary to represent several parts of it in the same w^ay. In this way we shall find that our desk should have a slope of about 25°. [Having once found the correct slope, it is well to have a desk made with its two supporting sides of the correct heights.] This figure, when we have obtained the correct inclination of the drawling desk, can be, at the same time, used in order to OTHER DRAWING APPARATUS. 33 calculate the magnification of the drawing [i.e., the magnifying power of the system, or combination of objective and eye-piece, in nse]. We know already that the lines which we have drawn are 0*01 millimeter (i.e., approximately -^-^^-^ inch) removed from one another. If we find that now they lie 2*4 mm. (i.e., nearly -^^ inch), we know that the drawing is enlarged 240 times. This method is also the simplest and best for measuring the size of the microscopical object. If we have, that is, attained the necessary accuracy in drawing, in order to reproduce even slight variations in size with fidelity, and if we know the definite enlargement of the object which we have drawn at exactly the same distance, it needs only to divide the size of the drawing by the known enlargement to get the actual size of the object. If, e.g., one cell of the hair of Tradescantia appears, with 240 times enlarge- ment of its figure, to be 9 mm. broad, this indicates an actual breadth of [^f o ^^-^ '^•^•' ^f] 0-0375 mm. This method gives in the simplest way such a close result, that in our investigations we can limit ourselves to it. [Various other contrivances have been introduced for the purpose of aids to drawing. Some of these, like the Wollaston Camera Lucida, require the body of the instrument to be placed horizon- tally, and the instrument as a whole to be raised on a pedestal. This can of course only be used with instruments which admit of this position ; and for working purposes it is, besides, objection- able in several ways. A very cheap form for use thus is Dr. Beale's neutral-tint reflector, which fixes on the eyepiece, making with its glass an angle of 45°. The student, when he chooses a camera or drawing-prism, should always select one for use with the instrument in the vertical position ; and, as he may not im- probably obtain one from a maker who is not the maker of his instrument, he should always send the eyepiece of the latter, so that the fittings of the camera may be adjusted to the size of this. Zeiss's camera is adjusted for eyepieces of the continental size, a size much used by English makers for their smaller instruments. Of whatever camera is chosen the method of adjustment upon the eyepiece must be learned from the maker (though usually very easy to find out for one's self) ; the rules laid down above for learn- ing how to draw are equally applicable to all of them. Lastly, the quality of the drawing depends on two factors : 1st, the ac- curacy of the observer, and 2nd, the skill of the draughtsman.] Now turn once more to the cell of the Tradescantia hair, and D 34 PLASMOLTSIS. endeavour with one or another drawing apparatus to make a figure of it. As in all drawing apparatus which are not strictly cameras, some manipulation for the regulation of the light is needed, so we mast endeavour, either by shading the drawing surface, or by changing the position of the mirror, to obtain thereabouts similar brightness for the surface of the drawing and the field of the microscope. For drawing, it is best to use stiff, smooth drawing- cards * and black-lead pencils. In order that they shall not be effaced, finished drawings can be washed over with very dilute gum-water. Take in this way a sketch of the entire outline of the cell, of the protoplasmic streams and the nucleus, and compare it after some hours, to see whether the form and circumstances now correspond. As already indicated, we shall most probably find that the dis- tribution of the streams has altered, and that the nucleus has changed its position in the cell. In order to determine that in their streaming the cells are independent of one another, and that the cell-wall does not in- fluence the movement, allow a neutral but water-containing fluid to act upon the cell. Under the cover-glass add to the drop of water a little concentrated sugar solution, or, better still, glycerine. Before long the reagent begins to withdraw the water of the cell- sap, and there results a decided contraction of the protoplasmic sac [i.e., the lining layer] into the cell. This withdraws from particular places of the cell-wall. This contraction of the proto- plasmic body of the cell under the influence of dehydrating (i.e., water- extracting) media is distinguished by the name Plasmolysis. It can be then observed, that so long as the contraction does not become too strong, the streaming of the protoplasm still goes on, even in those parts where it has withdrawn from the cell- wall. Soon, indeed, all movement in the cell is arrested. Yet in most cases to set it going again it suffices to wash out the water- extracting reagent by means of water. To this end water should be run under one edge of the cover-glass, while the fluid under the cover-glass is sucked out from the other edge by blotting-paper. The protoplasmic sac then again tends to expand and reach the cell- wall. It not infrequently happens that during the contraction single pieces of the protoplasm separate themselves from the cell-body, and remain lying against the cell-wall as rounded * Such, of excellent quahty and surface, are GoodaH's thin Bristol Boards. [Ed.] ROTATION OF PROTOPLASM. 35 balls. These balls can also be retaken into the expanding proto- plasmic sac. It is easy to determine that during the contraction of the contents, observed as above, the colour-material does not diffuse through the living protoplasmic sac, and that the coloration of the cell-sap becomes proportionally more intense. The appear- ances in dead cells are quite otherwise. For example, allow absolute alcohol to act upon the hairs. The protoplasm is imme- diately killed, and now the peculiar property of coagulated proto- plasmic masses, to accumulate colour materials, is set in action. The protoplasm withdraws from the cell-sap the violet colour, and this soon appears quite limpid, while the cell-plasm and the nucleus stain deep violet. The violet colour can now pass through the protoplasmic sac, and diffuse in the surrounding fluid. If Tradescantia should not be at the disposal of the observer, other hairs can be substituted for it. A very favourable object is pro- vided by the hairs which grow upon the youngest shoots of the genus CucurUta [gourd, pumpkin, vegetable marrow, cucumber, etc.]. The preparation is made by removing these hairs at their base by a razor, and bringing them into a drop of water on a slide. The stronger hairs are multicellular at the base, and pass into a tapering cell-row ; others bear multicellular heads. The proto- plasmic network in the cells is finely developed ; it contains micro- somes, and, though but sparely, large, green- coloured ChlorophyU- grains. The nucleus is large, suspended by the threads ; it has a brightly shining nucleolus, and is carried about hither and thitlier in the cell. A very peculiar object is provided by the root-hairs of llydro- charis morsus-rance [the Frogbit]. For the investigation are selected fresh young roots with stiff hairs. The hairs are visible to the naked eye. Cut off an entire root-point, and quickly place it on the slide in a sufficient quantity of water. The cover-glass is laid on in the usual way, and the largest cover-glass at our disposal should always be chosen. In this way the preparation is made, although it is true that, owing to the not inconsiderable thickness of the object, all parts will not be accessible with stronger magnification, because the object-glass will come into contact beforehand with the cover-glass. These hair-cells are very long and tubular, and, like all root-hairs, unicellular. The protoplasm, which it richly contains, is in active movement, but there are here, not numerously divided thin streams, formed into 36 EOTATION OF PROTOPLASM a network, but a single strong stream, moving round in the protoplasm lining the wall. This kind of movement is called Rotation, to distinguish it from the other kind, or Circulation. This stream, thus returning to the same place, presents the appearance of a broad, slightly spirally turned band, which, if projected upon a plane would form a very elongated figure 8. The movement must not, however, be represented as if the band, as a connected whole, were turned around inside the cell, for, in fact, the neighbouring parts during the movement are con- tinually changing their reciprocal position. The two streams going in opposite directions are, however, not in immediate juxta- position, but are separated by a narrow band of protoplasm which is at rest. This " neutral band " is reduced to a very thin layer of protoplasm. The leaves of Vallisneria spiralis furnish very instructive pre- parations for illustrating rotation of protoplasm. This plant is cultivated in all botanical gardens, and very commonly also in aquaria in rooms in houses. For investigation a strong leaf is selected, and a section taken from the lower part of it. For this purpose it answers best to lay the long, narrow leaf across the index finger, and to hold it down on both sides with the thumb and middle finger. The surface section is taken by moving the razor parallel to the long axis of the leaf. The aim should be to obtain a plate or " lamella " of tissue about half the thickness of the leaf [but if the section should at first sight appear too thick, parts of it which are sufficiently thin for the purpose will probably be found] . This lamella is laid on a slide, epidermis downwards, in a drop of water. Air clinging to it may make some parts of the section useless, but others will always be found which admit of undisturbed observation. The streaming always goes on for some time before it is discontinued ; it can be best followed in the wide elongated cells which form the interior of the leaf. At low room temperatures the movement is sluggish, but it can be hastened by slight warming of the microscope slide. The stream circles around the entire cell, without, in most cases, to any extent deviating from, its direction parallel to the long axis. The *' neutral band " is pretty broad. The stream carries with it green-coloured chlorophyll grains and the nucleus. The latter is flattened into the form of a disk. From time to time it comes into sight, but as a rule it is concealed by chlorophyll-grains. Not infrequently it sticks at a turning point, then the accompanying IN VALLISNERIA AND NITELLA. 37 cliloroplijll-grains also halt witli it, till, an instant later, all again are drawn into the stream. The direction of the streaming changes from cell to cell without any regularity. If glycerine or sugar solution is permitted to act upon the section, the proto- plasmic sac can be seen to withdraw from the cell- wall, and the continuance of the streaming at the first moment of contraction can be readily made out. The strongest protoplasmic currents known in vegetable cells are met with in the Characece (Stoneworts). We must, how- ever, take the genus Nitella, for the genus Chara has completely invested, and therefore opaque, internodes, while the internodes are specially suited for the investigation. For observation we select the younger members of the plant, and can state immediately that the rotating layer of protoplasm possesses a very distinct thickness. The outer layer of protoplasm [immediately lining the cell- wall], in which the chlorophyll-grains lie, is motionless. The motion- less layer is here, therefore, comparatively thick, while it is in general so thin as to escape observation. For in all earlier investi- gated objects also an outermost denser layer of protoplasm, the so-called primordial utricle (or Ectoplasm) takes no part in the movement. An obliquely mounting stripe or band on the wall of Nitella is free from chlorophyll grains ; it attracts the eye by its lighter coloration. This band, wanting in chlorophyll, marks the neutral band in the protoplasmic stream. It repeats here the like appearance with the root hairs of Hydrocharis, where we found the neutral band of the protoplasmic layer likewise extremely reduced. The intemodal cells of Characece are multinuclear, the protoplasmic current carries with it numerous elongated nuclei, which it is true show up as brighter spots only in the most favour- able cases. [If the piece of the plant is laid for 12 to 24 hours in 1% solution of chromic acid, they can often be very readily seen, and their peculiar rod-like, curved, and horse-shoe forms made out.] Not to be confused with these nuclei are the rounded balls which are seen carried around in the stream in larger or smaller number. These appear either smooth or with a spinous surface ; as to theii- significance there is uncertainty. NOTE TO CHAPTER III. Tradescnntia virjinica is a quite hardy perennial, and can be grown in any garden. It dies down in winter. Flowering period, June to August. [Ed.] 38 CHLOROPHYLL-BODIES. CHAPTER ly. CHKOMATOPHOKES. COLOURED CELL-SAP. Material Wanted. Funaria hygrometrica, or Prothallia of a Fern. The former (moss) very commonly grows on ground which has been charred, or limestone walls, etc. ; the latter, on pots and walls of fern-houses Flowers of the garden " Nasturtium " {Tropceolum majus). Flowers of the Snapdragon {Antirrhinum majus). Flowers of the Periwinkle {Vinca major or minor). Flowers of the Larkspur {Delphinium consolida). Flowers of Adonis fiammula. Root of Carrot {Daucus Carota). Autumnal leaves of "Virginian creeper {AmpelojJsis hederacea). Autumnal leaves of Ginkgo hiloha {Salishuria adiantifolia) ; or Maple. Flowers of the Mullein {Verbascum nigrum). Ehizome of Iris germanica. [All required fresh.] We have already had an opportunity in several objects of obtain- ing an insight into the structure and enclosures of the chlorophyll- grains [or bodies] ; nevertheless, we will give our attention some- what specially to these structures. We select for this purpose a very widely distributed moss, which is distinguished by very fine, large, lenticular Chlorophyll-bodies, and of which the leaves, unilamellar with the exception of the midrib, permit observation without further preparation. This moss is Funaria hygrometrica. Numerous chlorophyll-bodies of considerable size are to be seen in every cell ; in plants which are exposed to diffused daylight they are contiguous only to the free cell- walls ; that is, to those which form the upper and under surface of the leaf.* From this they present their broad side to the observer. That they are narrower in profile we see in the separate grains which underlie the side walls. All stages of division of the chlorophyll-bodies are easy to find, and often associated in the same cell (Fig. 17). The resting grains appear quite circular; they then become elliptic, afterwards constricted in the middle so as to be shaped like a figure of eight, * [This is commonly known as the position of Epistrophe.] CHLOROPHYLL-BODIES IN FUNARIA. 39 and finally completely diA-iVled across. The two yonng grains remain for some time still in contact. The starch-enclosures of the chlorophyll-bodies are, according to their varying sizes, in many leaves easy, in others dif- %'('®' ficult to see. They are, hoAvever, always clearly ^^ (^^ distinguishable when the chlorophyll-bodies get ^^ -^^ y^ out of an opened cell into the surrounding water, <^- ^ ^^ and are there disorganized. To this end we cut ©*^@) a leaf with a sharp pair of scissors into several ^^^ i7.-chioro- pieces. The starch-grains, liberated from the phyii-bodies from disorganized chlorophyll-bodies, augment in size, |fyg,^^„f,t^L!'''rr8T- and are identified as such with iodine. On the ing and in division. other hand an entire uninjured chlorophyll-body is coloured brown with iodine, always as a result of the combined blue coloration of the starch-enclosures, the yellowish brown coloration of the protoplasmic ground- substance, and the green of the chlorophyll. In order to obtain favourable iodine coloration of the uninjured chlorophyll-bodies, we take for investigation leaves which have lain some time in alcohol, and are thereby decolorized. The chlorophyll -bodies now appear colourless ; their starch-en- closures take on the coloration by gradual entrance of the lodme solution, earlier than the protoplasmic body. The iodine reaction is still more noticeable if the preparation is previously treated with potash, which causes the starch-grains to swell. ^ This last method also permits the smallest quantity of starch in the chlorophyll-bodies to be recognised. This succeeds so much the more sui-ely with fresh grains if they are treated with a solution of five parts of chloral hydrate in two parts of water 2 to which a little iodine solution has been added on the object-slide. The chlorophyll is dissolved, so that in a few minutes the leaf appears colourless; simultaneously the chlorophyll-body swells, and also the starch-grains which it contains, and these last come out clearly with their blue colour. Leaves decolorized with alcohol show also very beautifully, with the same treatment, the blue-stained starch- grains in the chlorophyll- bodies, while these last are not coloured. After the chlorophyll-bodies have been decolorized by alcohol they can be stained also very well with very dilute watery solution of Methyl violet or of Gentiana violet. The cell membranes also are always coloured hereby, but the chlorophyll-bodies are darker, and therefore stand out more sharply. With stronger magnification the living chlorophyll-bodies of the 40 COLOUR-BODIES OF TROP^OLUM. leaf of Funaria appear to be finely punctate, and thus betray a network structure. The same results as with the leaves of Funaria are obtained with . Fern prothallia, so that the two objects can mutually replace one another. Prothallia are always readily to be found in plant houses in which ferns are cultivated ; any species equally available for this investigation. In order to become acquainted with colour-bodies (Chromato- phores) of other coloration, let us turn next to Tropceolum majus [the so-called " Nasturtium" of gardens]. We choose for investi- gation flowers only just opened, because the colour-bodies begin to be disorganized in older flowers. Let us first take surface sections from the upper side of the sepals. The preparation can also be taken with a fine pair of forceps, if these are stuck pretty deeply into the tissue, and a strip torn therefrom. The preparation is laid in a drop of water, with the epidermis turned upwards. Proceed at once to the investigation, because the injurious action of water on the colour-body makes itself felt immediately. The margin of the section will have suffered from the beginning; therefore, cells that are still unchanged should be selected for more searching ex- amination. The colour-bodies are yellow with a shade of orange. They appear spindle-like, three or four angled (Fig. 18), in forms which border on the crystalline. The unchanged bodies are homogeneous. Under the influence of water they swell, become rounded off, and vacuolate ; that is, hollows filled with water appear in their interior. The bodies overlie in especial number the inner wall of the epidermal cells of the upper side of the calyx. The brown streaks on the upper side of the sepals proceed, as suitable sections show, from epidermal lines, the cells of which are filled with carmine-red cell-sap. These cells contain also yellow grains, which, however, the coloured cell-sap renders quite invisible. In the red cells the nu- cleus shows mostly as a clear spot. The petals show analogous relations; here the edges of the limb, as well as the cilia at the base of it, can be used for observation Fig. 18.— From the upper side of the calyx of Tropaeo. lum majus. The inner wall of an epidennal cell with the colour-bodies (chromato- phores) adjacent to it (x 540). TELLOW CELL-SAP OF VERBASCUM. 41 in their entire thickness. The air adhering to the limb hinders observation, but spots free from air will always be found, or can be made free by light pressure on the limb. The sepals, however^ always remain preferable for the observation of colour-bodies, since the papillae interrupt observation of the petals. It is evident that, with the exception of the brown stripes on the two lower petals, every epidermal cell of the upper and under side of it is prolonged in its centre into a blunt cone, the papillce already alluded to. These papillae are more strongly developed on the upper than on the under side. They give to the petals a velvety appearance. The air is entangled very strongly between the papillse. The fiery-red spots at the base of the petals arise from rosy cell-sap and yellow granules. During the investigation it will have been noticed that the surface of the epidermal cells of the upper side is longitudinally striate. The striations do not turn at the boundaries of the individual cells, and are folds of the cuticle which covers the epidermis. With watery solution of iodine the colour-bodies can be fixed pretty well, and take on at the same time a green coloration; they are very sharply defined. The nucleus is at the same time coloured yellowish-brown, its nucleolus becoming very visible. With Methyl violet or with Gentiana violet the colour-bodies are coloured violet. The yellow colouring matter is almost always combined with a protoplasmic basis ; but isolated cases are present w^here it is met with dissolved in the cell-sap. Let us fix our attention more closely on such a case in Verhascum nigrum. We can examine the petals in water without further preparation ; but here also we must remove the adhering air, even if only partially, either by pressure or under the air pump. The ej^iderraal cells of both upper and under side have undulating (sinuous) outlines ; the yellow colour of their cell-sap is at once noticeable. The brown spots at the base of the petals arise from a cell-sap coloured from purplish to brown. In the epidermis of the staminal filaments, from which lamellae can be easily cut with the razor, we see also a yellow sap ; but besides this there is in each cell also a cinnabar-red irregular lump of colour-material, and a number of colourless leucoplasts filled with starch-grains. Similarly it can be at once determined that the yellow-coloured parts of the lower lips of the corolla of Antirrhinum majns (the Snapdragon) contain a sulphur-yellow sap in their cells ; the parts coloured red have a rosy cell-sap and here and there one, seldom more, carmine-red balls of colour-material. 42 BLUE AND EED CELL-SAP. Fig. 19.— An epidermal cell from the under side of the petal of Vinca minor ( x 5iO). In tlie epidermis of the corolla of Vinca major or V. minor (the Periwinkle) we find a blue cell-sap. The epidermal cells, especi- ally of the Tipper side, are swollen out into papillae. The epidermis of either side can be readily torn off with the forceps. The side walls of the epidermal cells show ridges projecting into the cell cavity (Fig. 19), often swollen at their edges, so that they can even spread out into a T-form, and, on account of the stronger refraction of their outer surface and the weaker refraction in the interior, quite give the impression of folds. We see a red cell-sap in the petal of a rose. Here also the epidermis can be readily removed from either side. The upper side has pretty strongly developed papillee, and therefore appears so beauti- fully velvety. The cuticle shows strongly marked striation. In the blue sepals of Delphinium consolida (the Larkspur) we find the epidermis of both upper and under sides composed of cells with sinuous outlines. The epidermal cells of the upper side are elevated in their central part each to a papilla. The cuticular striations mount on all sides of this papilla, so that by focussing the microscope at the mid-height of the papillas, sun-like figures arise. The cells contain a blue cell-sap, somewhat shading into violet, besides also, in many cells, blue stars, which consist of short needles of crystallized colour- substance. The epidermis can be removed in small pieces ; moreover, the sepal is sufficiently transparent, after removal of the air, to permit examination at the edges through its entire thickness. Examples of blue and red cell-sap can be easily multiplied. Such are almost always met with in blue and red flowers ; so much the more remarkable, therefore, is the contents of the bright red flowers of Adonis fiammeus. In Adonis also the preparation can be removed with the forceps. In the epidermis we see beau- tiful red, from nearly round to elliptic, grains ; these are com- paratively large, and attain the size of chlorophyll-bodies. They appear finely granular, and in water separate quickly into very small granules, which show molecular movements [" Brownian movement"]. The epidermal cells are elongated; their cuticle longitudinally striate; the strise are clearly continued over the limits of the cells. COLOUR-CRYSTALS AXD AUTUMNAL TINTS. 43 \ ^ The root of Daucus carota (the Carrot) furnishes a very inter- esting object. The orange-red colour of this root arises from carmine and orange-red colour-bodies, which possess throughout a crystalline form. The most common shapes are found collected in Fig. 20. They are small rectangular plates or rhombs, the rhombs often acicularly elongated, and prisms of different lengths, often broadened out at one end to the shape of a fan. Such crystalline formations have often small uni- laterally projecting starch-grains attached to them. In origin, therefore, these crys- talline structures also are starch-builders, and must be placed in the same category with chlorophyll and other colour-bodies The colour-material which crystallizes out is here, however, what decides the shape. Only a small quantity of protoplasm ad- heres to the crystal, and from this, there- fore, the starch-grains also arise. If we examine also one of the variegated forms of our shrubs or trees, or else an herbaceous plant with leaves coloured reddish-brown, we see that the cells of the epidermis contain a rosy cell-sap, and that there- fore the joint action of the red of the surface and the green of the interior gives the reddish-brown compound colour. As to the autumnal coloration of the Virginian creeper, Ampe- lopsis hederacea, we can decide that the rose-coloured cell- sap arises in the cells of the [internal] tissue, and not of the epidermis. The distinctive yellow autumn coloration of leaves depends on the yellow coloration of the disorganized chlorophyll-bodies, as is shown in the most beautiful way in the leaves of Gingko hiloba [Salishuria adiantifolia], or, failing this, those of the various species of Maple. The autumnal brown coloration of leaves arises from a corresponding coloration of the cell- walls, chiefly, however, of the cell-contents, as is easy to determine in the case of the Oak. The starch-grains are found in specially individualized proto- plasmic structures. We have already learned to know the chloro- phyll-bodies as such, also the colour-bodies in which starch-grains are often present ; and lastly, we have already made reference to the colourless starch-builders. Upon these last devolves the Fig. 20. — Colour-bodies from the root of the Carrot. Partly with starch- grains (x540). 44 COLOURLESS STARCH-BUILDERS. formation of starch in the deeper layers of the body of the plant. We can comprise all three structures under the name of Chroma- tophores, and, further, distinguish the chlorophyll-bodies, colour- bodies, and colourless starch-builders as Chloroplasts, Chromo- plasts, and Leucoplasts respectively. These structures are nearly related, and can pass over into one another. They all belong to the protoplasm of the cell, and lie embedded therein. On the other hand the blue stars, which we found in the cell-sap of Delphinium consolida, do not belong to this ; they only represent colour-material crystallized out from the cell-sap, and are, like the lumps of coloui'-material which we found in the red cell- sap of Verbascum nigrum, not to be reckoned amongst the chromato- phores. The largest and most beautiful starch-grains are produced by leucoplasts ; but such leucoplasts are not exactly easy to see. A comparatively favourable object, and one not difficult to obtain, is furnished in the rhizome of Iris germanica. Surface sections of this are made parallel with the surface of the rhizome. The outer- most layer of tissue is removed, and to this succeed the starch layers. The observation is best made in water. In uninjured cells the leucoplasts appear as collections of pro- toplasm at the hinder end of the starch- grain (Fig. 21). These latter increase only at this end, and have a proportionally ec- FiG. 21.— Starch-build- x • i j_ mil i x ers with starch grains centric structure. The leucoplasts appear from the rhizome of Iris granular to the eye of the observer, and germanica (x 540). . j. i xi. • j. ii • i • i separate at length into smaller grains, which show molecular movement [Brownian movement]. Two starch- grains on one leucoplast is a not infrequent appearance. After further development such grains presently come into mutual con- tact, and receive thenceforth layers which are common to the two. These and similar phenomena lead, here and in other cases, to the formation of compound starch-grains. NOTES TO CHAPTEB IV. 1. Boehm'8 method. Sitzungsber. d. K. A. d. W. in Wien, Bd. XXII. p. 479. 2. According to A. Meyer, Das Chlorophyllkorn, p. 28. 3. A. F. W. Schimper. BoL Zeitung, 1880, col. 881 ; 1881, col. 185 ; 1883, col. 105, and 809. A. Meyer, Das Chlorophyllkorn, Bat. Zeituvg, 1883, col. 489. TISSUES OF THE BEETROOT. CHAPTER V. TISSUES ; THICKENING OF THE WALLS ; KEACTION FOE SUGAK ; INULINE, NITKATES, TANNIN, LIGNIN. Matehial Wanted. White Beetroot (Beta vulgaris). Fresh. A ripening Pear. Fresh. Tuber of DahUa (D. variahilis). Fresh. Tuber of Dahlia placed in meth. spirit, in or about October. Oak-apples or Oak-galls. Fresh and dried. Twig of Willow (e.g., 8nlix caprea). Fresh. Stems of Periwinkle {Virica major). Fresh, cut off close above the ground. Seeds of Ornithogalum sp. such as 0. umhellatum, the Star of Beth- lehem. Seeds (stones) of Date {Phoenix dadylifera) . Old Pine wood of any kind, preferably the Scotch Fir (Piniis sylves- tris). Dry, or, better, in alcohol. We commence with the white Beetroot {Beta vulgaris). A small piece of tissue is taken from the fleshy root, and from this is made a microscopical preparation. We choose as best for ex- amination a radial longitudinal section, i.e., therefore, a section which is taken parallel to the long axis, in the direction of the radius. This section cuts at right angles the concentric rings of the root, visible to the naked eye. Examined in water, this section shows us more or less rectangular cells, filled with a watery, colourless fluid. On the walls of these cells we notice also, here and there, larger and smaller, brighter, round or oval spots, which indicate shallow pits \_i.e., local thin places, or hollows, in the wall]. In individual cells the nucleus is visible. The intercellular spaces are usually filled with air, appearing black. In isolated parts of the preparation, the parenchymatous cells are narroTfer, elongated parallel to the long axis of the root ; between them are visible long tubes usually filled with air, 46 STRUCTURE OF THE BEETROOT. whicli are distinguished by a characteristic thickening of their walls. These tubes are vessels. The thickening of their walls is a network of pits [reticulated] ; that is, the wall shows thicken- ng bands combined into the form of a net, between which lie nnthickened places. These un thickened places or pits are elongated across the longitudinal direction of the vessels. Where the section has opened a vessel, there can be seen in it, from time to time, annular (ring-like) thickenings, which project into the interior of the cells. These are the diaphragm-like remains of originally complete partition walls, and from these remains it will be seen that the vessel has proceeded from a row of cells. The air present in the vessels often disturbs the examination ; it can be got out with the air-pump. When an air-pump is not at our disposal, we can endeavour to remove the air by laying the preparation in freshly boiled water. It is more quickly attained by a short immersion of the preparation in alcohol. It is true that by this the contents of the cells are killed; for the foreo-oing observation, however, this is not of consequence. Here and there also in the preparation we come across particu- lar cells, which are closely filled with small clinorhombic crystals, and appear almost black. These crystals consist of oxalate of lime. In order to prove this, we allow acetic acid to act upon them, and determine that they are insoluble in it. Into another preparation we run sulphuric acid, and the crystals are quickly dissolved. The quantity of sulphate of lime formed is so small that it remains dissolved in the surrounding fluid. The structural relations of the cells in the Beetroot show up still more beautifully and distinctly if the section is treated w^ith a watery solution of aniline green or of acetic aniline green. In both cases the cell-walls are beautifully stained green ; in the latter case the nucleus also is " fixed " and quickly stained. The walls of the parenchymatous cells, and of the vessels, are alike stained bluish-green. The surface of the pits in the walls of the parenchymatous cells, on the other hand, is not stained, and these therefore now show up more clearly ; they are places, in the otherwise not greatly thickened cell-walls, which have remained thin. Each parenchyma- cell contains nucleus, provided with a distinct nucleolus, and surrounded by minute leucoplasts, and a thin lining ["peripheral"] layer of protoplasm. The vessels con- tain neither nuclei nor plasmic contents. If chlorzinc iodine is added to a section lying in water, a characteristic violet cellulose- CELL-WALL AND PITS. CELLULOSE-REACTION. 47 reaction is soon set up. The coloration begins at the edges of the section, bnt is often not complete for hours. The walls of the vessels do not stain violet, but brownish-yellow ; they behave like lignified membranes. On the walls of the cells, the surfaces of the pits once more remain unstained, and stand out specially distinctly. These pit-surfaces are always rounded, of variable size, and irregularly distributed, singly or in groups. Large pit- surfaces are traversed by violet striae of various breadth ; they are formed into compartments by them, and give the impression of an irregular lattice. Bright granules, coloured yellow-brown by the chlorzinc iodine, adhere in larger or smaller quantity to the pit- surfaces. For the purpose of comparison we proceed now to the cellulose reaction with iodine and sulphuric acid. The section is first impregnated with iodine solution, best with potassium- iodide iodine solution, and afterwards transferred to diluted sulphuric acid (English), in the proportions of 2 volumes acid to 1 volume water. It commences at once, from the edges onwards, to indicate the action ; the section assumes a beautiful blue colour. The lesser pits here \ i also remain un- coloured ; the larger ones ap- pear latticed with blue. We further prepare a section from a ripening Pear. Constitut- ing the pulpy flesh of the fruit appears here also a regular thin- walled parenchy- ma of large cells, more or less rounded at the angles. These cells contain colourless cell-sap, a very reduced plasma-sac, and a nucleus. Scattered in the tissue are found nests of strongly-thickened cells (Fig. 22). The number of the stone cells so united is varied from part to part, and according to the kind of Fig. 22.— From the flesh of the fruit of the Pear. Stronply- thickeucd cells with branched pore-canals, surrounded by thin- walled parenchyma ( x 240). 48 EEACTIONS FOR SUGAR. pear. They form the so-called "grit" of the pear. The cells are distinguished by the considerable thickness of their walls, and by the numerous, fine, branched pore-canals [canaliculi] . The branch- ing arises from the diminution of the number of the pore-canals proportionally as the cavity of the cell becomes smaller [by the great increase in thickness of the walls], so that they open into the cell-cavity as common canals. Where two thickened cells are in contact, it can be determined that the pore-canals correspond in position with one another. In their perfected condition, in which they here appear to us, these cells no longer contain living cell-contents, but only a watery fluid. They represent, therefore, only dead cell-cases. After treatment with chlorzinc iodine, the thin parenchyma-cells take on gradually a violet coloration, the strongly thickened cells become yellow-brown. These latter are therefore lignified and belong, on account of their strong thicken- ing and lignification, to the sclerenchyma [or mechanical tissue]. The structural relations of the thickened cells become especially clear under treatment with chlorzinc iodine. We will use the flesh of the pear in order to learn to know the micro-chemical reactions for sugar .^ That most commonly used is with Fehling's solution. This is prepared with sulphate of copper and potassio-sodic tartrate in water. The proportions are 34' 64 gram, pure sulphate of copper with 200 gram, potassio-sodic tartrate dissolved in water. This solution can be preserved. In order to use it we add 600 ccm. soda ley of specific gravity 1"12, and dilute it to 1,000 ccm. This solution is heated to boiling. The section in which the reaction is to be produced should not be too thin, should contain at least two layers of uninjured cells, and naturally should not have previously been laid in water. Immerse the section, holding it with the forceps, in the boiling solution, and the section is coloured a beautiful vermilion-red. The reaction comes out in full beauty after two seconds. Under the microscope we can see in the cells the vermilion-red precipi- tate of reduced protoxide of copper. There is therefore present in the cells of the pear a substance which reduces the alkaline copper-oxide solution, a body from the grape-sugar group (Glucose), in this special case grape-sugar. For comparison we repeat the experiment with a section of Beetroot. This contains, as is known, a body from the cane-sugar group, viz., cane-sugar. Immersed for two seconds in the boiling fluid, it shows no precipitate in the cells ; the section, examined REACTIONS FOR NITRATES AND NITRITES. 49 microscopicallj, lias a blue coloration. If the section is kept for a longer time in the Feliling-'s solution, it begins to colour ver- milion-red on the surfaces also. The cane-sugar is inverted, and now gives the protoxide precipitate. Under the microscope the outer layers of cells show now vermilion- red grains, while, in case the action has not been too long continued, the inner cells still contain a blue fluid. Very much I'ecommended also for microscopical purposes is Barfoed's sugar reaction ^ with acidulated acetate of copper. This solution is prepared by dissolving 1 part of neutral crystallized ace- tate of copper in 15 parts of water. To 200 ccm. of this solution is added 5 ccm. of an acetic acid which contains 38 per cent, of glacial acetic acid. In a test-tube which holds from 5 to 8 ccm. of this solution we allow a section, not too thin, of the Pear, and in another similar teat-tube a section of the Beetroot to boil for a short time. The fluid in question, together with its section, is then poured out into a small evaporating dish, and allowed to stand. After some hours we find the section of the Pear covered with a fine precipitate of protoxide of copper, and likewise a little of the same precipitate in the evaporating dish, while the section of the Beetroot, as can readily be seen under the microscope, is free from the adhering precipitate, and this is wanting also in the evapo- rating dish. The result of the reaction should be observed after some hours, as after a longer time a very small precipitate re- oxidizes in the air, and can then dissolve. We will lastly again use the Beetroot, in order to learn to know the micro-chemical reactions for nitrates and nitrites by means of diphenylamine.^ This reagent, used by the chemist for the detection of very small quantities of nitrates and nitrites, performs, moreover, first-rate service for histological purposes. We prepare cross or longitudinal sections through the Beetroot, taking care, however, that the sections extend to the surface. These sections we allow carefully to previously become somewhat dry on the object-slide, and then first add the reagent. We use 0'05 gram diphenylamine in 10 ccm. pure sulphuric acid. Immediately after the addition of this, a deep blue coloration, formation of aniline- blue, shows in the outermost zone of the section. This zone contains the youngest tissue of the root, still in course of develop- ment ; it is this, therefore, which contains the nitrafe. From the parts coloured blue the colour quickly flows over the rest of the preparation ; but in the first moment of the reaction the coloured E 50 CROSS-STRIATION OF CELL-WALLS. zone is quite sliarply delimited. As, however, in plants, as analyses of sap show, the question is commonly of a nitrate, seldom of a nitrite, we can, therefore, from the resulting reaction, conclude with greater probability that it is a nitrate. If, instead of the somewhat dried section, a fresh one is used for the reaction, the colour-body which is formed is diffused far more rapidly in the surrounding tissue, and the coloured zone is less sharply delimited. As the next object of investigation we choose the tubers of the Dahlia (D. variahiUs). The tuber, halved longitudinally, allows one readily to recognise the central pith. A longitudinal section prepared from this shows under the microscope more or less rec- tangular cells, arranged in longitudinal rows (Fig. 23), with very reduced protoplasmic sac, with nacleus, and colour- less cell-sap. The inter- cellular spaces are filled with air ; the cell- walls finely striate. The striae are oblique, to the extent of from 35° to 40°. We be- lieve that we can see two diagonally opposed systems of striae in the same plane ; this is explained by the comparatively small thick- ness of the walls. In fact, the two opposing systems of strias belong to the walls of two contiguous cells re- spectively, as can be determined especially at the free edges of the section. With chlorzinc iodine the cell walls soon colour violet ; where, however, two striae come less closely together, a colourless line can be seen between them. The parts of the wall which remain unthickened are, just like pit-surfaces, not coloured by the chlorzinc iodine solution. Especially clearly show up in- dividual comparatively larger rhombic places as pits. Such pits lie always in the line of separation of two striae, and at the place of crossing of a line of separation of the system of striae running in the opposite direction. If the section is laid in absolute alcohol there arises in the cell- sap a fine precipitate of Inuline. Replace the alcohol by water Fig. 23.— From the pith of Dahlia variahiUs (X 2i0). REACTIONS FOR INULINE. 51 and warm the object- slide over a spirit flame, and the precipitate is again dissolved. In order to study the inuline in the shape of sphiero-crystals, which it forms,* we examine best pieces of tubers which have been placed in spirit at least eight days before. We examine the section best in water, and during the examination allow nitric acid very slowly to enter. The sphaero-crystals (Fig. 24) are found always on the cell-walls. They form more or less perfect balls. The ball can be traversed by one or by several cell-walls. Usually several variously sized balls form together a larger group. The balls allow more or less clearly a radial structure to be recognised ; this structure comes out more sharply when the nitric acid begins to work ; it arises from radially ar- ranged needle-shaped (acicular) crystals, which compose the ball. Besides the radial, a con- centric stratification is also usually visible, which is to be conceived as the expression of variations in the conditions of crystallization. Iodine solution produces no coloration. If the sphaero-crystals are warmed in a drop of water on the object- slide they quickly vanish. ^ In- order to demonstrate the tannin- reaction upon a typical object, we turn to the gall-apple or oak-gall, as it is to be found upon the leaves of our oaks. These gall-apples are due to the puncture of the oak-gall insect [Cynips querci], which lays an egg in the punctured tissue. We halve such a gall-apple while still young, and find on delicate radial sections taken from this that the interior hollow, occupied by the larva of the cynips, is sur- rounded by a shell, which consists of iso-diametric, rounded cells. These contain usually abundant starch-grains, becoming blue with iodine. The tissue following on to this inner portion is formed of radially elongated, polygonal cells, which diminish in length at the periplicry of the gall-apple, and finally end under the small- FiG. 24.— From the tuber oi Dahlia varta- hilis, after lying in spirits for many months. Sphaero-crystals on the walls ( x 240). 52 EEACTIONS FOR TANNIN. celled outermost layer, the epidermis, the cells of which are strongly thickened outwardly. This entire tissue, surrounding the inner shell, shows no enclosures of definite form. If, however, we lay a freshly-prepared section in a drop of watery chloride or sulphate of iron solution, we see that it colours throughout its entire mass of a dark-blue colour. This coloration is, moreover, communicated to the surrounding fluid, and produces for us, there- fore, the iron reaction for tannin, in its iron-blue form, while there is also an iron-green form. If the action is observed under the microscope, by allowing iron-solution to run into a dry section laid under a cover-glass, we see that first a fine dark-blue precipi- tate is formed, which, however, is soon again dissolved in the reagent, so that now a blue fluid fills the cells. The weakest tannin reaction is given by the starch-containing cells of the inner shell. For comparison, let us now lay a second section in a watery solution, about 10%, of bichroiiiate of potash, and we see a dense flocculent, I'ed-brown precipitate, which also persists, formed in the tannin-containing cells. Lastly, let us place a section in a concentrated solution of molybdate of ammonia, in concentrated ammonium chloride, and an abundant reddish- brown precipitate appears in the cells. This reaction will decide in doubtful cases, because those preceding can also proceed from other reducing bodies. The fihro-vasal bundles which traverse the oak-apple we will leave unnoticed, and also pass over other structural relations, because we have only taken this object in order to see a typical tannin reaction. Sections of dried gall-apples also give the above reactions, though less beautifully. In order to get the iron- green tannin reaction, we take a willow twig, say from Salix caprea, remove with the razor the outer green layer of bark, and then take a delicate tangential section from the green tissue of the cortex ; lay it in a drop of chloride of iron solu- tion. The section shows us mostly rectangular cells, somewhat elon- gated in cross direction, with walls pretty strongly thickened, and with simple pits. These cells contain chlorophyll-grains, and most of 'them, especially in winter, have each a white strongly refrac- tive rounded mass of cell contents, sharply defined, and filling the entire cell cavity. Other isolated cells contain a dark-looking stellate crystal of calcium oxalate, of which we shall, however, have an opportunity later of making a closer examination. The strongly . refractive masses of cell-contents contain tannin. As soon as the action of the iron chloride on the strons^ly refractive masses of SECONDARY THICKENING OF CELL-WALLS. 53 cell-contents has conimenoofl, these become griimons, and take on an olive-green to brown-green colour. In iron sulphate these masses become still browner ; in potassium bichromate they give a reddish-browm precipitate ; in ammonium mol jbdate dissolved in strong ammon. chloride, a yellow-brown grumous precipitate. With twn'gs of the alder (Alnus) the same results are produced. If a strong stem of Tinea major [the Periwinkle], cut off close above the ground, is broken, we see from the edge of the broken surface numerous small fibres project. We seize a number of such fibres with the forceps, draw them out, and place them in a drop of water on an object-slide. Under the microscope they appear to us as long, strongly-thickened sclerenchyma-fibres, tapering at both ends. The cavity is reduced to a narrow canal, which is en- tirely obliterated at both ends of the fibre. In slightly- thickened fibres the wall appears striate in one direction only. In more strongly-thickened fibres there are two oppositely oblique systems of striae, of Avhich one belongs to the outer, the other to the inner system of wall-layers [complex of lamellse]. Lastly, in still older sclerenchyma-fibres is often found still a third internal system of striae, directed almost perpendicularly to the long axis. This last arises from reticulated thickening bands, which leave between them elongated pits. This innermost system of thickening is usually sharply limited towards the exterior ones. With chlor- zinc iodine solution the fibres take on immediately a violet colora- tion, passing into brown. Specially instructive, however, is the relation with cuproxide ammonia, which reagent has the power of dissolving pure cellulose. The action must be observed directly. On the addition of the cuproxide ammonia solution the walls of the fibres swell strongly. At the first moment of the action the striation becomes more distinct, but quickly disappears. The outer complexes of layers are soon completely dissolved, while the inner reticulated one resists longer, and therefore the observer sees it completely isolated. At the beginning of the swelling a still finer stratification appears in the stratification which was previously visible. Each layer is therefore composed of numerous exceedingly thin lamellae. Such a fine stratification is stamped especially distinctly upon the inner more resistent layer. We now divide in halves, with the pocket-knife, the seed of Ornithogalum, say 0. iimhellatum [the Star of Bethlehem], clamp the half in the hand- vice, damp the cut surface with water, and make with the razor the thinnest possible preparation. This 54 PITS. preparation (Fig. 25) presents us cells with approximately rectan- gular contour. Tlie walls of these cells are strongly thickened, the thickening layer being, however, pierced by numerous simple pits. If the section has so grazed a cell -wall that it presents a surface view, the pits appear as round pores (7??), as can be seen in the upper cell of the adjoining figure. From the side the pits appear as canals, which pass out of the cell-cavity up to the primary cell-wall. The pits of ad- joining cells are directed towards one another ; they are separated by the primary wall (p), which we shall here designate the closing membrane. The inner surface of the thickening layer is distinguished by stronger refrac- tiveness ; it forms the limiting mem- brane. If sulphuric acid is allowed to act slowly on the preparation, from the edge of the cover-glass, the thickening layers of the cells are dis- solved, while a network of very delicate walls is at first left behind. The walls are the so-called middle lamellaB, which indicate the walls of the cells which were present before the thickening began, and which also traverse the closing membrane of the pits. By continuous action of the sulphuric acid, these middle lamellae also soon disappear. Chlorzinc iodine causes the thickening layers to swell and the middle lamellae become likewise visible. In con- sequence of the swelling, the coloration of the preparation is incomplete. The cells are closely filled with protojolasm and granular materials. These entire contents take on with iodine a greenish- brown coloration. In each cell the nucleus is readily distinguish- able with acetic aniline green ; this is, in general, wanting in no cell, living or capable of life. The thickening layers of the cells in the endosperm of the Date (Phoenix dadylifera) have a very similar appearance. The cells, however, are more elongated, their cavity narrower, the walls somewhat thicker. In the seed [" stone "] of the Date these cells are radially arranged. Cross and longitudinal sections of it, Fig. 2o.— From the endosperm of Omithogalum umhellatum. m, pits seen from above ; p, closing mem- brane [in pits seen in profile] ; n, nucleus (x 240). BORDERED PITS. 65 therefore, provided they correspond with the radii, show the cells in longitudinal view, while tangential sections, which cut the radii at right angles, show cross-sections of the cells. Chlorzinc iodine solution colours the thickening layers very beautifully violet. By slower swelling usually numerous lamellae are brought into sight. We turn now to the pine wood [Pinus, etc., any species, prefer- ably P.sijlvestris, the Scotch fir], in order to learn to know bordered pits. For this purpose we take a piece of wood, either dry, or, better still, preserved in alcohol, from a stem as old as possible. First we prepare with a sharp pocket-knife the suitable surfaces for cutting — one radial, parallel to the long axis of the stem, one tangential to the same, and one directed perpendicularly to this axis. The concentric yearly rings which are visible with the naked eye upon every piece of pine wood will provide us with the necessary bases from which to get information as to the direc- tions in question. The radial longitudinal section cuts the yearly rings perpendicularly. The tangential longtitudinal section is so much the more perfect, the more parallel it runs to the yearly rings. The cross-section is directed perpendicularly to both longi- tudinal sections. In the following preparation of microscopical sections, in order that the sections shall be good, and not to damage the razor, quite special precautionary rules must be adopted. If the razor is hollow-ground, rightly directed sections can be taken only from the edges of the piece of w^ood, i.e., so long as the back of the razor does not yet rest upon the cut surface. However, in general, only slightly hollowed razors should be used for cutting wood, as those greatly hollowed easily " give." It is recommended to use razors which are ground flat on one side, i.e., the side Avhich will rest upon the cut surface ; but these razors have the disad- vantage that they are not easily sharpened. The cut surface must always be moistened ; the sections must be as thin as possible. It is not necessary to have them of any particular size. A section which appears to be too thick should not be cut to the end ; it is better to withdraw the razor from the cut in order not to notch the edge. The razor must be sharp, otherwise it will tear the cell walls, and separate the inner thickening layers from the outer. The wood preserved in alcohol cuts more easily than when dry, especially when the former has been laid subsequently in a mixture of equal parts glycerine and alcohol. The surface of the cut surface prepared by the pocket-knife, as it contains the torn cell- 66 BORDERED PITS. walls, must be removed with the razor. The succeeding sections can be used.* A radial longitudinal section, correctly taken through the wood of the Pine, appears, with weak magnification, to be constructed of longitudinally elongated cells, which overlap one another with their tapering ends. Running across these cells we see the cell- rows of the medullary rays, with which we shall not at present concern ourselves. We focus now with stronger magnification upon a part in which we see only the walls of the longitu- dinally elongated wood-cells [fibres], and always the broader of them, and direct our whole attention to the bordered pits of these walls. The bordered pit appears to us in the form of two concentric circles >, C (Fig. 26, A). The j V ^{ { y inner small circle, or, it may be, the inner ellipse, indi- cates the opening of the pit into the cavity of the cell ; the larger outer circle, or outer el- lipse, the widest part of the pit, with which it joins on to the primary wall separating the two cells. In fact, this pit is only distinguished from the simple pit, as we have seen it in the Date and in Ornithogalum, in that it broadens at its base. The pits of the adjoining ceils, however, meet here in just the same fashion. If the mouth of the pit, as commonly, is an obliquely placed ellipse (as in A), by changing the focussing we shall find the corresponding mouth [of the other pit] oblique in the opposite direction. The two pit-chambers adjoining one .another are separated from one another by the primary wall, which, before the commencement of the secondary thickening, was already pre- FiG. 26.— Piftus sylvsstris. A, a bordered pit in surface view. JB, a bordered pit in tangential longitudinal section ; t, the torus. C, cross-section of an entire tracheide ; m, middle -lamella ; in*, a " seam " ; i, the limiting membrane (X 540). * It is of some advantage to keep an old razor (sharp, however,) for j^repar- ing surfaces, as it is keener than a pocjcet-knife, and will spare the actual section razor. Even then the first section cut with the latter should be rejected. [Ed.] BORDERED PITS. 57 sent, and subsequently is only slightly thickened. This delicate wall is the closing membrane. In the middle it is more strongly thickened, and forms the so-called torus. With most careful observation and suitable focussing we may even be able to see this torus. It forms a round, weakly-shining disk, which has about twice the diameter of the mouth (compare in A). In the most favourable cases, and here especially in preparations of dried wood, a radial striation is observable in this torus, and so that the delicate part of the closing membrane appears differen- tiated into radially dispersing lamellae.^ A complete insight into the structure of the bordered pit can only be obtained with the aid of tangential sections. As the bordered pits stand on the radial walls of the w^ood-cells,'' they are seen in cross-section (Fig. 26, B) in correctly taken tangential longitudinal sections. We search for these structures in the walls separating the wood-cells, stopping first at the dividing walls of the broader wood-cells, and not allowing ourselves to be led astray by the sectional view of the medullary rays, which are formed of a number of smaller cells, standing one over the other. The figure of the cut pit is, it is true, clear only in very delicate parts of the section. If this condition is fulfilled, the pit appears in the form of the two ends of a pair of tongs directed towards one another [or like a couple of extremely short screws placed with their heads flat together], after the type of the above figure (26, C). If once the structure of this large bordered pit is known, we can obtain information as to the structure of the smaller ones, which lie in the thicker walls of the narrower wood-cells. The difference, apart from the smaller size, is, that here on both sides a longer canal, corresponding to the thickness of the wall, runs out of the broadened pit-chamber. The largest bordered pits are connected w^ith the smallest by all intermediate stages. In the interior of the pit is seen, in the most favourable cases, the closing mem- brane, which in its centre is swollen into a torus (t). In the bordered pits of the air-dry wood it is usually pressed to one side of the pit-chamber (B). If, on the other hand, fresh wood, or alcohol material, is investigated, we shall find the closing membrane in the sap-wood [alburnum] stretched across the middle of the pit- chamber. In the heart-wood [duramen], on the contrary, t..o relations are just as we have given for the air-dry wood. The figure of the bordered pit is clearer after the action of chlorzinc iodine, which stains the cell-wall yellow-brown. This coloration 58 STRUCTUEE OF COXIFEROUS WOOD. is due to the strong lignification of the walls. Only in occasional places is a violet tinge still to be seen there, ^.e., where a not yet completely lignified inner thickening layer gives this colour re- action. The closing membrane is in general not stained by the chlorzinc iodine. After treatment with this reagent we can readily convince ourselves that the perfect wood-cells contain here neither protoplasmic sac nor nucleus ; they consist only of dead • cell-walls, and, as they functionally contain only water, and in this respect, as well also as in the nature of the thickening of their walls, they simulate the tracheae, i.e., vessels, they are known as tracheides, more recently as hydroides. N'ot infrequently the pine wood, which we examine, shows in longitudinal section a more or less distinct spiral striation mount- ing at an angle of about 45°. The mouths of the pits then appear elongated in the direction of the striation, and as do the striae of the two adjoining side-walls, so also the mouths them- selv-es of adjoining pits cross one another. We prepare a cross-section also of the pine ivood. This must be specially thin. The tracheides thus cut across appear as a rule rectangular. They form radial rows. We pause at one with the widest lumen (cavity). On its radial walls we see the sections of the pits (Fig. 26, G), the figure of which appears in no way different from the tangential longitudinal section. Between the cells the middle lamellae proceed as fine separating lines (m) . Where more than two cells are in contact, the middle lamella is broadened into a solid or hollow '"seam" {m*). The inner limit of the cell-wall is more strongly refractive and forms the limiting membrane {i), which is specially clear in the more strongly thickened tracheides with narrower cavities. It is ahvays clearer after the action of concentrated sulphuric acid. The thickening sheaths swell, and are finally dissolved ; the limiting membrane resists longer and stands out sharply. Between the swelling thickening layers are seen the primary avails of the cells, of which finally only the delicate netw^ork of middle lamellae is left behind, stained yellowish-brown. These middle lamellee, resisting con- centrated sulphuric acid, are cutinized [cuticularized]. With slower swelling in sulphuric acid it can be often determined, and especially on the strongly- thickened tracheides, that the thicken- ing layer consists of very numerous extremely delicate lamellae. With chlorzinc iodine the cross-section, as previously the longi- udinal section, is coloured yellow- brown ; in individual cells, how- REACTIONS FOR LIGNIN. 59 ever, part of the tliickening layer impinging directly upon the limiting membrane, takes on a violet tone. If we follow the treatment with chlorzinc iodine with dilute sulphuric acid (two parts acid, one part water), under the influence of this latter a blue coloration of the entire thickening layer is possible. If delicate sections are treated with concentrated chromic acid, an opposite action to that of sulphuric acid results. The middle lamellte are dissolved, and the individual cells, therefore, aie separated from one another. The thickening layer of the cells undergoes a not inconsiderable swelling ; the limiting membrane at the commencement of the action stands out sharply, but soon becomes unrecognisable. In order further to learn the characteristic reactions for Lignin, we will make use of phloroglucin and of sulphate of ani- line.^ We dissolve a trace of phloroglucin in alcohol, and lay some sections of wood in this solution. After this we place it in a drop of water on the object-slide, and allow, from under the edge of the cover-glass, hydrochloric acid to act upon it. The walls of the cells quickly take on a beautiful violet-red coloration. Other sections we place in a watery solution of aniline sulphate, Avhere they at once become bright yellow ; this colour is still more heightened by the addition of dilute sulphuric acid. In place of the phloroglucin we can use an extract, prepared with water or spirits of wine, from the wood of the Cherry, with almost the same result.^ If fresh sections of the stem of the Pine, passing from cortex to pith, are treated with concentrated hydrochloric acid, a yellow coloration of the w^ood is at once brought about, which, however, gradually shades off, inwardly and outwardly re- spectively, into a violet coloration.^" This also is the phloroglucin reaction, and indeed proceeds from the phloroglucin which comes from the contents of the cortical cells and pith cells respectively. Even the medullary rays of the young wood contain a little phloroglucin, so that the violet coloration also spreads from each of these. In the future, we shall make use of the different relations of lignified and unlignified cell-walls towards certain colour-bodies as an assistance in our investigations. 60 NOTES. NOTES TO CHAPTER V. 1 Compare Sachs, most recently in Jahrb.filr jciss. Bot. Bd. III. p, 187. - Barfoed de organiske Staffers qualitative analyse. Kjobenhavn. 1878, pp. 210, 217, 223. Notes. •'' Compare H. Molisch, Ber. der deutsch. hotan. Gesellsch. I. Jahrg. p. 150. * ?ya.chs, Bot. Zt(j. 1864, p. 77; Hansen, Arb. d. Bot. Inst, in ll'ilrzburg. Bd. III. p. 108 ; Meyer, Bot. Ztg. 1883, Col. 334 ; W. Gardiner, Proceedings of the Cambridge Philosophical Society. Vol. IV,, Pt. VI. p. 387. 5 Sanio, Jahrb. f. iciss. Bot. Bd. IX. p. 50. Strasburger, ZeUhriute, -p. S8. Russow, Bot. Centralbl., 1883. Bd. XIIL, Nos. 1-5. The other literature is there quoted. 6 Compare Russow, Bot. Centralbl, 1883. Bd. XIII., Nos. 1-5. 7 Bordered pits placed on the tangential walls occur rarely in the Pine, bat on the contrary are quite regularly met with in the autumn wood of the other Abietineffi. 8 Both introduced by Wiesner (compare Stzbr. der math. not. Klas. der Akad. der Wiss. zu Wien. Bd. LXXVII. 1, Abth., and before that in other places). 9 Von Hohnel, Stzber. der math. nat. Kl. der Wiener Akad. d. Wiss. Bd. LXXVI. p. 685. '0 The same, page 676. THE EPfDEKMlS. 61 CHAPTER VI. THE EPIDERMIS, STOMATA, WATER STOMATA. Material Wanted. Leaves oi Iris jiorentina. Fresh. Leaves of Tradescantia virginica. Fresh. Leaves of Aloe (e.g. A. nigricans) or Agave. Fresh. Leaves of Anehnla (e.g. A. fraxinifolia). Fresh. Leaves of Nerium oleander. Fresh. Leaves of Tropceoliim majus (Indian cress, or so-called " ]S'asturtium"). Take a surface section of the outer side (morphologically the under side) of the " equitant " leaves of Iris florenthia. The section must be so thin that it only grazes the tissue underlying the epidermis, and should be observed in water with the outer side turned up- wards. It will be at once seen that the Epidermis is composed of elongated cells which run parallel to the long axis of the leaf. The cells are ended by cross partition walls ; they are connected together without any intercellular sjDaces (other than the stomata), contain colourless cell-sap, a nucleus, and a very reduced proto- plasmic sac. On its external side the epidermis is covered by an exceedingly fine-grained layer of wax. In a line with the cells of the epidermis lie the elliptic Stomata, which, however, are only indistinctly visible because the four cells of the epidermis which surround each spread over the Guard-cells of the stoma, par- tially covering them. Hence there remains only an elliptically elongated pit (/) which leads to the stoma (Fig. 27, ^4). This pit usually appears black, because filled with air. In order to see the guard-cells well, now turn the section over. It can then be easily proved that the stoma is formed by two half-moon-shaped guard-cells. In distinction from the neighbouring epidermal cells these cells contain chlorophyll bodies. The nuclei are wont to show as clear spots about the mid-length of the cells. Between the two guard-cells is a spindle-sliaped cleft {s), about half the length of these cells. Since the long axis of the stomata corre- sponds with the long axis of the leaf, it is easy to obtain correct 62 THE EPIDERMIS. cross sections of the stomata. The section is taken at right angles to the long axis of the leaf. For this purpose a narrow strip, about |-inch broad, should be cut out of the leaf in the direction of its length with a pair of scissors ; this strip can be supported between two pieces of the pith of the elder or of the sunflower.* The elder or sunflower pith necessary for this purpose is obtained from dried pieces of the stems of those plants by stripping off the cortex and woody bundles. A piece of pith about an inch long is cut in two lengthwise with a sharp razor. The flat strip of tissue Fig. 27. — Epidermis of the under side of the leaf of Iris florentina. A, surface view; B, in cross-section. /, stomatic pit; s, cleft, or stoma; c, cuticle; a, air-chamber (x 240). which has to be cut is now laid between the two halves of the pith, so that the end of the strip reaches to the end surface of the piece of pith. Thin cross sections are then taken through pith and object at the same time, and the sections are lifted with a camel hair brush from the blade of the razor on to the object-slide. While cutting, the two pieces of pith can either be held together simply with the fingers, or the two halves can be fastened together by tying round with a piece of thread. In cutting, the pith is so * Or, several such strips can be packed together without other support than they give to each other. [Ed.] THE MICROTOME. 63 held that the razor lights on the broad side and not on the thin end (edge) of the object; in this way many equal sections can be taken. For delicate objects the softer sunflower pith is preferable to the somewhat harder elder pith ; for more resistant objects, like that in question, elder pith is better used ; for still more resistant objects, not pith, but fine cork, as used for bottles.* The preparation of sufficiently thin sections need in this instance present no real difficulty ; under any circumstances such difficulty can be overcome by the use of a Microtome. A hand microtome of the simplest construction, such as Zeiss (of Jena) offers in his Catalogue for 1883, as No. 140, at I85., would suffice. This has a round cutting plate, ground smooth, of about three inches in diameter, which is fastened to a cylindrical tube, likewise serving as a handle. Inside this tube is placed a second, movable upwards and downwards by means of a screw. The extent of the move- ment can be read off upon a divided disk. The pieces of pith, between which the object is, are placed between two pieces of cork, hollowed out to receive them, and these are firmly fixed in the inner tube of the microtome. The pieces of pith project somewhat beyond the pieces of cork, and reach as high as the upper cutting plate. The sections can be taken either with an ordinary razor, or with one ground flat on one side ; [and while one hand holds the microtome] the razor is moved with the free hand over the cutting plate. [In the case of the razor with one side ground flat, this side should be applied to the cutting plate. Sec- tions are best cut by pushing the razor away from the operator.] * A few more practical hints on the subject of section-cutting by hand may be of use to beginners. The razor for most purposes should be what is called " hollow-ground," and of tolerably good quality, and should be kept sharp. The object to be cut should be held pretty firmly between the thumb and index- finger of the left hand, the index-finger being held as nearly as possible hori- zontally, and slightly bent, the thumb likewise very slightly bent, and wuth the joint depressed below the level of the finger, in order to secure its safety should the razor slip. In holding the object to be cut, the side of the tip of the index- finger should be rather higher than that of the tip of the thumb. The razor being then grasped firmly but not stiffly, the blade held quite flat and hori- zontal, the edge towards the body ; the index-finger of the left hand will serve as a table, on which the blade will lie and thus be greatly steadied. The section should be cut by a single forward and lateral movement of the blade. With all objects which will bear it, the razor-blade may float with alcohol on its upper side, and the object should be similarly wetted; otherwise the object, as here, may be kept moist with water. For this purpose two " wash-bottles " are a saving of time — one for distilled water, the other for alcohol. [Ei»-] 64 EPIDERMIS AND STOMATA. After cutting eacli section, the object should be somewhat raised by turning the screw. Microtomes of complex construction, such as are necessary to zoologists, are superfluous for botanists. In this way a considerable number of sections are prepared for further use, and they can be laid in the meantime [by means of a camel-hair pencil] in a watch-glass filled with water. Place some of the sections in water for observation under the microscope, and they will show, in favourable places, median cuts through the stomata, as shown in Fig. 27, B. As such a cross section will show, the epidermal cells of Iris florentina are more strongly thickened on their outer than on their inner side. The inner walls, however, are also pretty thick, while the radial walls are only slightly thickened. This structure is connected with the function of the Epidermis, which not only has to serve as an outer pro- tecting sheath, but also has to functionate as a water-reservoir.- The thin radial walls easily allow a change in the capacity of the Cells, which, by means of a bellows-like play, diminish in height through loss of water, and enlarge again with increase of water. The guard-cells lie recessed between the epidermal cells ; the manner in which the latter overlap the guard- cells can be at once seen. The pit leads down to the guard-cells. These latter show a cross-section quite peculiar to them. On the upper and under side they are strongly thickened. These thickened places are contiguous on that side on which is the stomatic cleft. Above each of these places is found a peculiar beak-like projection. On the opposite side, turned towards the interior of the epidermal cells, the guard-cells are comparatively thin-walled. This method of thickening of the wall is connected with the mechanism of the movement of the guard-cells, which would more strongly curve, and thus widen the cleft, when their turgidity increases, but which would straighten themselves, and thus diminish the cleft, when their turgidity decreases. It is clear, indeed, that with increasing turgidity the guard-cells must become more convex on the side of less resistance, more concave on the side of greater resistance ; just as an indiarubber ball, with a wall thicker on one side, must, by the forcing in of water or air under high pressure, become concave on the side of stronger re- sistance. The thin place on the side of the cleft, where the two thickened parts join together, facilitates the flattening of the cells on this side during curvature. In order that the movement of the guard-cells may not be prejudiced, we see the outer epidermal STOMATA. (55 wall join on to these guard-cells with suddenly diminishing rim ; the guard-cells are here fastened as with hinges,— the epidermal or stomatic joints, or articulations. Under the stoma is found the air-chamber (a), a large intercellular space, under natural conditions filled with air, surrounded by chlorophyll-containing cells, and 'connected with the intercellular spaces which are found between them. A cross-section laid in chlorzinc iodine shows us that the Avails of the epidermal cells stain in their entire extent, with the exception of a thin outer layer, somewhat corrugated, the so-called Cuticle (c) which becomes yellowish-brown. This cuticle swells out at the stoma into the beak-like projection which we have already mentioned, w^hich appears coloured yellow-brown by the chlorzinc iodine, and is therefore cuticularized. As an extremely delicate membrane, the cuticle is continued through the stomatic cleft, over the guard-cells, to the commencement of the chlorophyll-containing parenchyma. For the rest, the guard-cells are also violet over their whole extent [and are therefore of cellulose]. By the use of concentrated sulphuric acid the whole section is dissolved, the cuticle alone remaining behind, together with the cuticularized projections of the stoma. An exceedingly favourable object for the study of the stomatic apparatus is found in Tradescantia virginica. The epidermis on both sides of the leaf consists of polygonal cells, mostly elongated in the direction of the long axis of the leaf ; with these alternate narrow stripes of longer and narrower cells. These stripes are visible with the naked eye, especially on the under surface of the leaf, and appear green in colour, while the stripes of broader cells show grey. The lateral walls of the epidermal cells are pitted ; the outer surface is faintly striate. The number of stomata is markedly greater on the under side of the leaf ; therefore we choose this side for investigation. The stomata are always sur- rounded by four epidermal cells (Fig. 28, ^4). The guard-cells lie on the same level with the epidermal cells ; the cleft which they have between them is comparatively large; they contain chloro- phyll-grains, between which the nucleus is usually visible. In the epidermal cells also the nuclei are sharply conspicuous, and appear surrounded by colourless leucoplasts (Fig. 28, A, I) ; the cell-sap of the epidermal cells is here and there rose-coloured. The long axis of the stomata corresponds with the long axis of the leaf, so that here also it is easy to obtain correct cross-sections. The stomata present then the apjiearauce shown in Fig. 28, B. F 66 STOMATA. The stomatic side of the gnard-cells here also appears to be thickened, while the side turned towards the interior of the epidermal cells is thinner. Besides this, it happens that both of the epidermal cells bounding the guard-cells are flatter than the epidermal cells lying beyond, and are also less thickened on their outer sides. They appertain, therefore, to the* stomatic apparatus as " accessory cells " ; they form the hinge or joint which in Iris florentina is formed merely by the thin part of the membrane at the insertion of the guard-cells. The leucoplasts (Z), which surround the nucleus in the epidermal cells, offer here a very favourable object for observation. It is interesting that these leucoplasts, in spite of being in a position so strongly exposed to the light, remain small and colourless, and do not develop into chlorophyll-grains. The epidermis has here another purpose, and has not to functionate as an apparatus for assimi- lation. Fig. 28.— Epidermis of the under side of the leaf of Tradescantia virginica. A, seen from above ; B, in cross-section through the leaf ; I, leucoplasts ( x 24iO). The so commonly cultivated Tradescantia zehrina has a stomatic apparatus composed in the same way. Stomata are present only on the under side of the leaf. The cross-section is very instructive, though not easy to obtain thin ; thicker sections serve for general information. The epidermal cells on both sides of the leaf are alike distinguished, as cross-sections show, by their considerable size. Those on the upper side esj^ecially are so deep that they alone form half the thickness of the leaf. Many of these epidermal cells are seen to be divided by cross-walls. On both sides of the leaf the epidermal cells contain little besides watery cell-sap, that on the under side, moreover, mostly appears coloured red. The STOMATA. leaves of Tradescantla present, therefore, in their epidermis, a specially efficacious water-reservoir. The accessory cells of the stomata, almost always four in number, are, as tlie cross-sectioo shows, quite thin, so that a great air-chamber, of the depth of the surrounding epidermal cells, is formed under the stomatic appara- tus. In thicker parts also of a surface section, taken from the under side of the leaf, the form of the air-chamber can be traced out by deeper focussing, so long as the chamber is not opened by the razor, and remains filled with air. The leucoplasts around the nucleus of the epidermal cells are again clearly visible. The species of Aloii and Agave possess epidermal cells thickened very strongly on their outer sides, and stomata correspondingly deeply sunk in the epidermis. Because it is specially instructive, and not difficult to prepare, we select for observation Aloe nigricans, 8b greenhouse plant with ligulate leaves arranged in two series (ranks). Other species of Aloe can, if need be, serve as substi- tutes for this. In surface sections, the epidermis of upper as well as under side appears formed of regular polygonal cells, mostly hexagonal. The cavity (or lumen) of each of these cells is reduced to a relatively small, rounded space. This space appears dark, because the razor opened the cells from below, and the cavities filled with air. The stomata are found on both sides of the leaf ; deep pits lead up to them. These pits are always bounded by four cells, and have a rectangular contour ; a somewhat project- ing rim surrounds the pit. If you wish to see the guard-cells, it suffices to lay the section on the glass slide with the inner side upwards. The guard-cells are comparatively broad and short ; amongst their contents are noticeable strongly refractive spherical oil-globules. A.s the epidermis is very hard, the cross-section is best taken between two pieces of bottle cork. The whole thick- ness of the leaf need not be taken, but rather a piece of the tissue, about -^-^th. inch thick, is cut off from one surface of the leaf. As the stomata run parallel to the long axis of tlie leaf, we arrange the piece of leaf so that it shall be cut at right angles to this axis. We cut the sections from the inner towards the outer, i.e., from the soft towards the harder part of the tissue. The strong thickening of the epidermal cells is observable immediately in these sections (see Fig. 29) ; this thickening affects only the outer half of the cell ; con'esponding to it, the cavity of the cell tapers in an out- ward direction. The thickened parts of the cell-wall are white, strongly refractive, and are covered externally by a cuticle more 6S STOMATA. strongly refractive still, but not sharply delimited. The lateral boundaries of the cells are only indicated by delicate lines in the thickened mass, and outwardly by a slight ridge. The interior of the strongly refractive thickening sheath is clothed by a compara- tively slight, weakly refractive layer (i). This surrounds, there- fore, first the keel-shaped lessening part of the cell-cavity ; while gradually thinning off, it ends in the side walls simultaneously with the refractive thickening layer. This thickened part of the epidermis, viewed in the aggregate in the section, appears like a curtain cut into regular teeth. At the places where the hollows leading up to the stomata are found, is first to be noticed the projection which encloses the hollow as with a rim ; next, that the tooth, formed by the thickening layers, is here halved unilater- ally, and has also only half its usual depth. The guard-cells show, both above and below, on the stomatic side, projecting ridges, which in cross - section appear beak-like. Above the guard-cells are found the thin parts of the wall which serve as epidermal joints. The air-chamber is narrow and deep. Commonly a parallel, more or less oblique, striation w^ill be observed on the thickened walls of the epidermal cells ; it is caused by the razor in cutting, and recurs in the same way not infrequently on hard elastic objects. A section treated with chlorzinc iodine, shows the highly-refractive thickening layer coloured yellow- brown ; it is, therefore, cuticularized. The inner covering to this layer (i) is, on the other hand, coloured violet, as likewise is the rest of the tissue of the leaf. The yellow-brown coloration passes over the "hinge" on to the projections which are on the guard-cells above and below. Elsewhere the guard-cells are coloured violet. On treatment with concentrated sulphuric acid, the whole of the part which colours yellow-brown with chlorzinc Fig. 29.— Cross-section through the epidermis and stoma of Aloe nigricans, i, inner thickening layer (x 240). STOMATA. 69 iodine remains at first bcliind ; after some hours' action this also is dissolved, and then the delicate cuticle, and the fine middle lamellse found between the epidermal cells alone still persist. The cuticle is continued over the guard-cells to the junction with the chlorophyll-containing inner cells. The cuticular layers and the cuticle take a brown colour in the sulphuric acid. The oil present in the guard-cells " balls " together, immediately on the entrance of the acid, into a highly refractive spherule, which disappears after some time. Many modifications occur in the arrangement of the stomata in the epidermis. A very remarkable instance is that where the stomatic apparatus is surrounded by a single annular epidermal cell. This can be observed in Aneimia fraxinifolia, a fern which is to be found in every botanical garden. The cells of the epidermis have a strongly undulat- ing [" sinuous "] outline (Fig. 30), and, by this mutual dovetailing, so common a thing in epi- dermal cells, gain in firmness and solidity. Like all other ferns, Aneimia contains chloro- phyll-grains richly in its epidermal cells. Here, therefore, such a division of labour as exists in most Phanerogams is not carried out, and the ei3idermis forms part of the assimilating tissues. The stoma is set in the surrounding epidermal cell as in a frame. Cross sections (at right angles to the lateral veins) show us that they project somewhat above the surface of the epi- dermis. This extreme case is connected by intermediate forms, with other less remarkable ones, into which we shall not further enter. We need only to imagine the stomatic apparatus removed to the side wall of the surrounding epidermal cell, to do away with the unusual character of their insertion. Nerium oleander shows a peculiar condition. Neither on the upper nor the under surface of the leaf can stomata at first be seen. On both sides we find a comparatively small-celled epider- mis, which, especially on the underside, is covered with unicellular hairs, their walls thickened almost to the disappearance of the cavity. On the under side of the leaf, however, there a]>pear also larger or smaller depressions, filled with air, and garnished at their edges with short hairs, resembling those just mentioned, but with less thickened walls. These hairs, coming together. Fig. 30. — Aneimia fraxinifoUa. Stoma, surrounded by an epidermal cell ; n, nucleus of the epi- dermal cell ( X 240). 70 WATER-STOMATA. close up the aperture towards the exterior. A second surface section from the under surface of the leaf, taken from the same place, whence a previous one has already removed the epidermis, permits to us here and there a view of the bottom of the hollows. For this purpose it is, above all, necessary that the air should be previously removed from the hollows, either under the air-pump, or throuofh soaking: the section in alcohol. It is then shown that from the walls of the depression, project small conical elevations, whose apex is formed by a stoma. The side walls of the small cones consist of epidermal cells, which allow between them an air-chamber extending to the stoma. Between the cones bearing the stomata, the similar hairs to those which we have seen on the edges spring from the walls of the cavities. We will now turn our attention to a specially favourable object for observing Water-Pores or Water-Stomata, These show the same structure as the air- stomata, but are larger, the cleft, as well as the ad- joining intercellular space (air-chamber) is, at least partially, filled with water. The guard-cells of these stomata may be from the first immovable, quickly perish, and then at all events lose their mova- bility. The most favour- able object for the study of these water-pores is Trojpceolum majus [the In- dian cress or so-called "Nasturtium"]. The water-stomata are found in the upper side of the leaf, and always over the ends of the prin- cipal veins (or ribs). Here the edge of the leaf usually shows a small depression. A pretty clear view of the w^ater- stomata can be had if a suitable piece of a leaf throughout its whole thickness is brought into the field of the microscope, under water, and covered over with a cover-glass. The details are indeed only observable on surface sections taken from the proper part of the edge of the leaf. A w^ater-stoma then presents the appearance Fig. 31.— Water- stoma of the edge of the leaf of Troposolum majus, together with the surrounding epidermal cells ( x 240j. WATER-STOMATA. 71 in Fig. 31. The contents of the guard-cells were in this case already reduced tq a minimum. Several water-stomata are always found at a short distance from one another. NOTES TO CHAPTER VI. ^ Strasburger, Jahrh. fur iviss. Bot. V. p. 297 ; de Bary, Vergl. Anat. pp. 32 et seq. ; 70 et seq. (See trans, by Bower & Scott, pp. 29, et seq. ; 66 et seq.) Schwendener, Monatsher. d. kgl. Akad. d. Wiss. in Berlin, 1881, p. 833. For the remaining literature, see the two first-named authorities. - Westermaier, Jahrh. filr iviss. Bot. XIV. p. 43. 72 HAIRS. CHAPTER VII. THE EPIDERMIS (CONT.); HAIR?. MUCILAGE AND WAX. Material Wanted. Young branches of Wallflower {Cheiranthus Cheiri). Fresh. Leaves of Ten- week Stock {Mattliiola annua). Fresh. Flowers of Pansy {Viola tricolor). Fresh. Flowers of Mullein {Verhascum nigrum). Fresh. Leaves of Verhascum tliaiJsiforme. Fresh. Leaves of Shepherdia canadensis, or of Eleagnus angustifolius. Fresh. Young stems of Rosa semperflorens, or other rose. Fresh. Young stems of the stinging nettle {Urtica dioica). Fresh. Leaf-stalks of the Primula, P. sinensis. Fresh. Young stems of Bumex patientia. Fresh. Leaves of Sundew {Brosera rotundifolia) . Fresh. Winter buds of JEsculus Hippocastanum. Leaves of Echeveria secunda-glauca, or other like kind. Fresh. Piece of cortex of node of sugar-cane {Saccharum officinarum). Fresh. We are already acquainted with the root-hairs of Hydrocharis morsus-rance, and as with root-hairs it is always a case of similar rUnicellular sacs, we can abstain from further investigation of them. We have also seen the epidermal cells of numerous petals elongated into conical papillas (Tropceolum, Rosa), and also the staminal hairs of Tradescantia, threads formed of barrel-shaped, swollen cells (Fig. 15) ; lastly also the hairs of Cucurhita, passing over from a multicellular base into a simple pointed thread. Plant hairs are therefore known to us from many points of view ; it is, however, worth while to extend our special knowledge of them. On the leaves and stems of Cruciferae we find very many forms of much- branched unicellular hairs. On the stems and leaves of the Wall-flower, or GiWj-^o^ev {Cheiranthus Che^ri),^ye see spindle- formed structures (Fig. 32, A), with narrow cavities obliterated towards the two ends. These unicellular spindles are covered on their outer surface with protuberances, always fewer large ones BRANCHED UXI- CELLULAR HAIRS. 73 with numerous small ones between. As the spindles are all di- rected parallel to the long axis of the leaf, it is comparatively easy to obtain a good cross-section through them. It is indeed desir- able to hit upon the hair at its point of insertion in the centre of its length, and numerous sections must therefore be taken in order to increase the chance of success. Then we see (Fig. 32, J5) that the place of insertion of the hair lies somewhat depressed, and that the epidermal cell which broadens out outwardly into the body of the hair is smaller than its neighbours, that at the base it is somewhat swollen, rounded, and reaches more deeply into the surround- ing tissue. It forms the " foot " of the hair. Lon- gitudinal sections through the leaf show that the foot is not broader in the long direction of the hair than in the cross direction. We can readily satisfy ourselves that the cavity of the foot passes without interruption into the cav- ity of the body of the hair. We can obtain a still more complete figure of the form of the foot if we lay a thin surface-section with the under side up- wards. The foot is circular in cross-section. It can now be seen, also, that the chlorophyll-containing cells of the tissue of the leaf adjoin radially, and without interruption, the somewhat broadened part of the foot projecting below the epidermis. The hairs of the ten- week stock, Matthiola annua (Fig. 32, C), are repeatedly branched in one plane. These hairs, especially on the under surface of the leaf, are set so closely together tliat their branches interlace. The cavity of the hair, in consequence of the strong thickening of the walls, is well-nigh obliterated. Knobs Fig. 32, A and J?.— From the under side of the leaf of Chcimnthns Cheiri. A, the hair seen from above (x 90). B, in cross-section (x 240). C, from the under side of the leaf of Matthiola annua; hair seen from above ( x 90) . 74 BRANCHED MULTICELLULAR HAIRS. are scarcely at all developed on tlie surface. The view of tlie epidermis from the inner side (by means of surface sections placed upside down) is very instructive, for it shows a tolerably marked swelling of the globular foot of the hair, and around it an exceed- ingly beautiful radial arrangement of the chlorophjdl- containing cells. In the groove of the lower spur-like elongated petals of the pansy {Viola tricolor) are very peculiar long unicellular hairs (Fig. 33). They can be seen very well if a cross-section of the lower petal is taken near the place where the tubular spur opens out into the furrow or groove. Each of the epidermal cells concerned grows out, almost in its entire width, into a hair. This is covered with irregular knotted swellings. The membrane of the hair shows slight longitudinal ridges. The cell-sap is colourless, but yellow pigment-bodies (chromatophores) are often present in the protoplasmic sac. The staminal filaments in the flowers of the common Mullein {Verhascum nigrum) are covered with unicellular violet hairs. In order to examine them the anther should be removed from the filament, and this latter pulled to pieces with needles in a drop of water on an object-slide. The hairs are very long, swollen out at the end into the form of a club, and with violet cell-sap. The sui^face of the hair is covered with elongated protuberances which ascend in more or less regular spirals. Branched multicellular hairs are to be found in the same plant on the under side and edges of the corolla. Seen from above, these hairs have a certain likeness to those of Mattliiola, but all of the branches here arise from a common central point, and each branch is in itself a closed cell. Moreover, the branches do not spread out in the same plane, but arise at indefinite angles. Their walls are quite as strongly thickened as in Matthiola; outer protu- berances are wanting. The hairs oA the edges are seen in side view. The body of the hair is cut off by a partition wall from Fig. 33.— Hair from the fur- row of the lower petal of Viola tricolor ( x 240). FELT, SCALES. 75 the epidermal cell whicli bears it. It consists of a stalk or pedicel, almost always unicellular, and upon this the branches are mounted. Slight modifications of these conditions occur, which need no fur- ther explanation. Besides these branched hairs, the edge of the corolla also bears small glandular hairs. These have a two to three celled stalk, and a flattened head, which is covered here and there at the apex by a strongly refractive substance. These last we shall not, however, study here, but in another more favourable object. It is only necessary to imagine the multicellular branched hairs of the mullein placed one upon another several times in order to understand the hairs which form the felt on the leaves of Verhas- cum thapsifoi'ine. These hairs are sometimes as many as five stages high, each stage is separated from its predecessor by a uni- cellular joint, which continues the main axis of the hair. The cells of the hair are for the most part filled with air. They are best shown by cross-sections through the midrib of the leaf. Fig. 3i.— Scales from the under side of the leaf of Shcpherdia cana dciisis. il.from the surface; B, in cross-section ( x 210). To the same category as the branched hairs of the petals of Ver- bascum belong the scales of Shepherdia canadensis. On the under side of the leaf, even distinguishable with a hand-lens, we find more or less loosely-formed white, and more or less closely-formed brown (Fig. 34, A) stars. On the upper side of the leaf only white stars are to be found, and they always in small number. The cells of these looser white stars contain, as microscopical examination 76 PRICKLES. shows, only air ; they arise from a common central point, but are separated from one another laterally. On the npper side of the leaf they do not lie in one plane, but rather radiate stellately in all directions. The cells of the brown stars are connected together almost to their ends, and provided with living contents ; the nuclei in their interior can be seen without difficulty. A cross- section through the leaf, where it cuts a brown star centrally, shows that its stalk (Fig. 34, B) is multicellular, and that not only the epidermis but also the cell-layer next following passes over into it. The stalk bears aloft the stellate unilamellar biit multi- cellular expansion. Should Shepherdia canadensis not be at our disposal, Eleagnus angustifolius can to a certain extent replace it. Here, on the under side of the leaf, only the white air-containing scales are present. The disk consists of cells either laterally isolated or also grown together almost to the margin. Now take a horizontal section through the stem of a rose, say Rosa semperflorens of the gardens, at the place where one of the prickles arises. Try to halve the prickle as nearly as possible in the middle, and then to take a thin section. This last is, indeed, not so easy as it seems. In cutting, do not neglect to moisten the cut surface with water. In a successful section it can be seen that the epidermis of the stem is continued over the prickle. The cells of the epidermis are at the same time more strongly thickened and more elongated. Inside the epidermis there pass into the prickle pretty strongly thickened narrow-cavitied cells, and, fur- ther in, similar but broader ones. . These last fill up the whole central part of the prickle. All these cells are finely pitted. The epidermis of the stem is separated from the chlorophyll-containing inner tissue by a strong layer of considerably thickened elongated* cells joining on to one another with oblique end- walls, and con- taining no chlorophyll. These cells without chlorophyll are similar elements to those which form the inner tissue of the prickle. The elements of the tissue of the prickle, are, however, separated from the chlorophyll-containing tissue of the stem by a layer of flat-celled tissue. This strip of tissue arises by division from the undermost layer of the tissue of the prickle ; it follows only for a short space the chlorophyll- containing tissue of the stem, and then turns towards the epidermis, in order to bound the base of the prickle laterally also towards the chlorophyll-less tissue of the stem. This is a cork-layer, next to the outer surface STINGING HAIRS. 77 of which, by the interposition of a layer of separation (absciss- layer) the fall of the prickles will result in the older parts of the stem. Before this, it is possible to break off the prickle pretty smoothly from the stem, along the inner side of the cork-layer. If we select a prickle from the leaf -stalk for investigation, its structure is found to be in no way different from that on the stem, excepting that at its base the cork-layer is wanting. [Since the leaf as a whole will fall, separate provision for the fall of the prickles is unnecessary.] By careful examination of the cortical tissue adjoining the prickles of the rose, the presence of crystals in the cells can be made out. As they are not dissolved in acetic acid, nor in potash, but on the other hand are dissolved in hydrochloric acid without evolution of gas, they are crystals of oxalate of lime. They have here the form either of monoclinic prisms or of cluster-crystals. These last consist of a great number of crystals which are deposited on an original crystal. The cluster-crystals are specially distinguished by their size and stellate form. In order to get the stinging hairs of the common stinging nettle (Urtica dioica) uninjured, we must take them from the younger parts of the plant. They are found best on the veins, or ribs, of young actively growing leaves. The hair, which is visible with the naked eye, should be cut off below its point of insertion with the razor, and examined in water. If the hair is already dead, air will be found in its interior, and its apex is then no longer intact. An uninjured hair presents the appearance repre- sented in* Fig. 35. The hair is unicellular, sharply conical, swelling at its apex into a small knob. At the base the hair broadens out, and the bulb thus formed is sunk in a cup which is developed from the tissue of the leaf. As its developmental history shows, this hair springs from a single epidermal cell, lyino- at the same level with its neighbours ; afterwards the stronq-lv- swelling foot of the hair is lifted up on a column of tissue, which is covered by the epidermis, and is formed internally of hypo- dermal (sub-epidermal) tissue. In the hair itself is to be seen streaming of the protoplasm. The nucleus is usually to be seen inside the bulb, suspended by protoplasmic threads. The cuticle shows oblique striation, which ascends in the same direction in all the hairs. The wall of the hairs is siliceous, as can be readily proved by heating it red-hot on a mica plate. As already noted, hairs are often found with their points broken off. In case of 78 GLANDULAR HAIKS. careless contact, the hair, by means of this point, enters the skin, and as it is very brittle, breaks off, whereon the strongly acid sap enters the wound and causes slight inflammation. On the same piece of the epidermis, near to the stinging hairs, are also small unicellular bristles (cf. Fig. 35) ; these last are distinguished by the strong thickening of their walls, and their fine tapering points. We can find the same kind of bristle on the edge of the leaf. For this purpose it suffices to place a piece of the leaf in water under a cover-glass. In old leaves the bristles can be thickened almost to the obliteration of their cavity ; their sur- face is covered with small protuberances. We have already met wdth glandular hairs on the edge of the petals of Ver- bascum nigrum ; they can be studied under more favourable conditions in Primula sinensis. For this purpose cross-sections are taken through a leaf-stalk. The body of the hair is divided from the epidermoid foot-cell by a cross wall situated out be- yond the epidermis, and forms a cell row, which consists of usually two (sometimes more), longer, and at the same time broader, and one (rarely two), narrower and also shorter cells. This last cell bears the globular head. Upon^ this, how- ever, is formed a more or less strongly- developed cap of highly refractive resin- ous yellowish substance. The secretion takes place between the cuticle and the wall of the cell. The cuticle is raised, distended, and finally ruptured, where- upon the secretion overflows the upper part of the hair. The addition of alcohol removes the secretion, and then the raised cuticle can be clearly seen lying in folds. The cells of the hair show a beautiful network of protoplasm with suspended nucleus, in which lies a large nu- cleolus. Small chlorophyll-bodies are embedded in the peripheral protoplasm. Very beautiful for observation are the glandular hairs (CoUeters) upon the membranous sheathing stipules (ochreae) of. Fig. 35.— Stinging hair of JJrtica dioica, together with a fragment of the epidermis, on which is a small bristle ( x 60) . / GLANDULAR HAIRS. the leaf of Biimex patie^itia, one of the docks not found in Britain. The masses of secretion given off from the ghands are here so considerable, that in damp weather the apex of the stem and the young leaves are found entirely covered with slime. The membranous ochrese can be observed directly, and for that purpose they must be turned with their inner side upwards. A careful examination of the preparation will show the glands in the form of minute plates. These minute plates rise with a short unicellular foot from a small epidermal cell. To the one cell succeed two ; upon these usually four cells, which are elongated in the direction of the long axis of the plate, and are repeated in several stages. On the outwardly- turned walls of the cells of the gland are often to be seen bladder-like swellings, which sometimes occupy a part, and sometimes the whole wall of a cell. The secretion, therefore, is formed here also between the cuticle and the rest of the cell-wall, and lifts the cuticle up. At length the bladders open and let out the secretion. This secretion is not coloured by iodine, nor with chlorzinc iodine ; in water it swells to a perfectly clear solution, and behaves like a gummy body. The cells of the glands are rich in protoplasmic contents, and their nuclei are distinct. With Rosaniline violet the glands take an intense violet coloration, and the masses of slime are pale- red. Watery solution of nigrosine stains the slime steel-blue, without colouring the glands. Especially interesting in structure are the glandular hairs of the co;nmon sundew (Dwsera rotundifolia) , distinguished alike as digestive glands and tentacles. They arise as thread-like struc- tures from the edge and entire upper surface of the leaf. Tlie threads (Fig. 37) taper a little in the course of their length, and Fig. 36.— Gland from the ochreae of J? inner patientia ( x 240). 80 TENTACLES OF DROSEEA. swell into the form of an egg at their ends. The threads consist of delicate cells, elongated in the longitudinal direction ; the stronger threads are traversed by one or several tubes with screw- like thickenings, — the spiral vessels. The radial extension of the epidermis of the thread in forming the head, the fan-like arrange- ment of the elements of this epidermis, and their multiplication into two or three layers, are seen best in optical sections of the object (Fig. 37). The number of the spirally-thickened vessels is greater in the head; all the cells which lie inside the sheath formed by the division of the epidermal cells, take on this spiral thickening. The place of insertion of the thread, if correctly hit upon, shows that not only the epidermis but also the inner tissue of the leaf is continued into the tentacles. These digestive glands give out a slimy secre- tion, which clings to the head like a dew-drop Fig. 37. — Digestive ^land [tentacle] of Dro- sera rotundifolia (x 60). [Fig. 37* — Leaf of Droscra rotundifolia, dmgrarmnatized. A, expanded; d, tentacles on the edge of the leaf; m, the shoi'ter, stouter tentacles of the centre of the leaf. B, all the tentacle? have bent towards the middle at the touch of an insect, x (after Prantl).] [whence the common name of the plant], but does not arise under the cuticle, but rather flows out from its free surface. Small insects remain sticking in these slime drops, suffocated in the secreted slime, and are carried towards the centre of the leaf by a Qprres- ponding inflection of the [stalk of the] digestive gland. Then the other digestive glands all bend together over the body of the insect, and come in contact with it with their heads. Upon this GUM, RESIN, WAX. 81 the chemical nature of the secretion changes ; a free acid and a ferment, like pepsin, make their appearance, and these are enabled slowly to digest the albuminous bodies found in the body of the insect. The dissolved substances are absorbed into the plant. A cross-section through a winter bud of the horse-chestnut {JEsculus Mppocastanum) shows us button-shaped glandular hairs, situated on the scales covering the bud (Fig. 38). The inter- mediate scales of the bud bear glands on both sides ; on the external ones more are found on the inner side, on the inner scales the most on the outer surface. The structure of the glands is shown in the figure ; it shows an axial cell-row, which towards the top divides, and from which the secreting cells radiate. The figure gives the gland in longitudinal section. The cuticle is broken through by the secretion, and this is discharged between the scales, coating them and stick- ing them together. This secretion consists of a mixture of gum and resin. In water the gum-drops scattered in the resin can be seen to swell, while on the other hand, by the addition of Rosaline violet, the resin mass is coloured a beau- tiful blue. Here also the contents of the glands are red. Fig. 38.— Glandular hair on a scale of the winter bud of JSscttlus hippocasta- num, covered with secretion ( x 240). the epidermis ; we propose. On one object (Iris florentina), we have already drawn attention to the finely granular layer of wax which covers the outer surface of however, to investigate this point specially on some other plants Very suitable for this is Echeveria secunda-glauca, or other like plant, which is now so often used in gardens for " carpet- bed ding." The wax layer, which can easily be wiped off, gives to the plant a hoary or " glaucous " appearance. A surface view of the epider- mis shows us a net-like crust of aggregated grains. In an easily observed form, we see aggregated short rods forming a wax layer, in the surface view of the epidermis of Eucalyptus globulus. The most beautiful object is the sugar-cane (SaccJiarum officinarum), now so commonly cultivated in plant-houses. Here the wax covering appears in the form of long rods or filaments, often curved or curled at the end. We remove a surface section 82 WAX. from the nodes of the stem, which are noticeable from their glaucous appearance. As much air clings between the rods, it is best to immerse the section for a short time in cold alcohol. It can be then readily examined. On the other hand, it is difficult to obtain a good cross-section with the rods still adhering. 'Pig. 39 shows such an one. The rods stand closely crowded against . ■ MmIIm loQQOQ Fig. 39.— Cross-section through a node (knot) of the stem of Saccharum officinarum, with a rod-like, waxy layer ( x 540). • one another, many showing the bending already referred to. If a surface section is brought into proximity to a flame, the rods, under the microscope, show fused together. They dissolve in hot alcohol. NOTE TO CHAPTEK VII. ^ Compare this with de Bary, Com;p. Anat., §§ 10, 13, 16, et seq., and the literature also given there. CLOSED VASCULAR BUNDLES. 83 CHAPTER VIII. CLOSED COLLATEKAL FIBRO-VASAL (OR FIBRO- VASCULAR') BUNDLES. MUCUS AND GUM. Material AYanted. Stems of the Maize, or Indian Corn {Zea Mais), some time in alcohol. Or, stems of the Oat {Avena sativa), or other grass, likewise in alcohol. Full-grown leaf of Iris fioren Una, some time in alcohol. Stem of Dracaena {Cordyline) rubra- Fresh. A YERY favourable object for the study of the structure of the Closed collateral fibro-vasal bundles^ of the Monocotyledons is the stem of the Maize or Indian Corn (Zea Mais). We will investigate material which has lain for a considerable time in alcohol, in order the more readily to become acquainted at the same time ^yith. the cell- contents. First prepare a cross-section, taking care that it passes through an internode and not through a node. The comprehension of the structure will be much facilitated if the section is laid at once into a drop of chlorzinc iodine. Coloration of the section immediately begins, and the separate fibro-vasal bundles stand out quite clearly, even to the naked eye. If we lay the glass slide on a white object \_e.g. a sheet of paper], we can in the readiest possible way get information as to the isolated (or scattered) arrangement of the fibro-vasal bundles ; an arrangement, as a whole, peculiar to monocotyledons. It will also show that the fibro-vasal bundles are more closely crowded together towards the periphery of tlie stem. Every fibro-vasal bundle shows in cross-section as an oval spot : the tissue in which these bundles are embedded is the Ground, or Fundamental Tissue. A separation of the ground- tissue into pith and cortex is not present with the scattered or isolated arrangement of the bundles. Now find, under the micro- * In the course of the chapter it will be seen that the term " fibro-vascular," or "fibro-vasal," is open to objections. The author uses the term "vasal- bundles," or " vascular-bundles" (Gefassbiindel). The former ai'c, however, in common English use, and are retained here. [Ed.] 84 CLOSED VASCULAR BUNDLES. scope with a low power, a part of the section suitable for further investigation, choosing a fibro-vasal bundle which does not lie too near the periphery, because in this neighboui'hood the structure of many bundles is simplified, and fusion with one another occurs. In all cases it is necessary to settle definitely in which direction the periphery of the stem lies, in order that we may know which is the inner and which the outer side of the bundle. The bundle which we select may appear somewhat as in the adjoining Fig. 40. FiG_ 40.— Cross-Section through a fibro-vasal bundle from the inner part of the stem of Zea Mais, a, Segment of an annular vessel; sp, spiral vessel; m and m', pitted ducts ; v, sieve-tubes; s, companion cells; pr, crushed elements of the protophloem ; I, intercellular passage; ug, sheath (x 180). First to attract our attention is the Sheath (rg), which surrounds the fibro-vasal bundle, and has become coloured more or less red- dish-brown by the chlorzinc iodine. It consists of strongly- thickened and lignified sclerenchyma cells or fibres, and has for that reason stained as indicated above. It is more strongly de- CLOSED VASCULAR BUNDLES. 85 veloped on the inner than on the outer side of the fibro-vasal bundle, still weaker on its sides. Passing now from the inner side of the bundle towards the outer, we next see an Intercellular passage (or intercellular space) (Z), surrounded by narrow, only slightly-thickened cells, which are nevertheless coloured yellow by the chlorzinc iodine. Into this intercellular space projects a ring (a), belonging to an Annular vessel, which is usually torn by stretching. The intercellular passage, also, has usually arisen from the breaking down of cells. Such a method of development is indicated by the term lysigeyious, whereas when it arises only by the separation of the elements of a tissue, the process is schizogenous. This torn vessel, together with others Avhich may perhaps also be seen projecting into the intercellular passage, represents the first- formed elements in this part of the fibro-vasal bundle, elements which were developed at a time when the parts of the plant with which we are now concerned were still in process of rapid growth in length. Impinging on the intercellular passage on its outer side, are one or more other vessels. They are recognisable by their cavity, which is larger than that of the neighbouring cells. In the bundle represented in Fig. 40, only one such vessel {sp), and that a rather narrow one, is observable in the cross section. These vessels, present to the number of one or more, are, as can be demonstrated only in longitudinal section, thickened in a spiral manner [they are spiral vessels]. Next, in each half of the bundle, right and left, is a wide cell-cavity (m, w'). These are two vessels with netted (reticulate) or pitted, rarely spiral thick- ening [they are the so-called pitted ducts]. Often in the cavity of these great vessels a ring, or part of one (m'), can be seen pro- jecting as a thickening of the wall. This is the relic of a cross partition- wall which, diaphragm-like, was broken through. The cells which surround the two great vessels are reticulately thick- ened : on their sides turned towards the ground-tissue these vessels are usually, however, bounded directly by the elements of the sheath. The elements lying between the two great vessels show also, in general, the network thickening, and appear as a somewhat darker band joining these vessels. They are usually arranged in regular lines in the direction of the radial axis of the bundle. All the walls of the vessels, and especially those of the two great vessels, are coloured yellow by the chlorzinc iodine. In the two gi^eat vessels it happens that this coloration is more intense on that side where they are bounded by the sheath. The 86 CLOSED VASCULAR BUNDLES. elements between the two large vessels are coloured a somewliat deeper yellow than those surrounding the intercellular space. The part of the fibro-vasal bundle which we have thus far described is distinguished as the Wood, or Xylem, or as the " vascular part," and also by the name " Hadrome." On practical grounds we prefer the older name of Wood, or Xylem. These terms do not, therefore, involve, as we at once see in this first example, the presence of the strongly-thickened elements on which the common idea of wood is founded. The never-failing element of the wood portion of the fibro-vasal bundle is the vessel, and, therefore, the morphological term formed according to this is the most rational. The selection of the term "Wood," however, simpli- fies the terminology, and permits the primary part of the bundle, and the secondary growth which we shall hereafter describe, to have corresponding names given to them. For our earliest descriptions, therefore, we must give the preference to this older terminology, in accordance with which many terms still in use have been constructed. In the example studied above we have found, therefore, in the wood portion (the Xylem) of the fibro- vasal bundle, the primary wood, the Protoxylem, composed of primary wood-parenchyma and of vessels. In opposition to the wood portion, we must choose for the second part of the fibro- vasal bundle the term Bast, or Phloem, against which names we must raise the same objections as to the term wood. In the above example we have similarly a bast portion without the presence of what is usually spoken of as bast. As the sieve-tubes of the bast are never wanting, morphologically the most rational term for it is Sieve portion.^ In contradistinction to Hadi^ome, the bast por- tion is also, on physiological grounds, distinguished as Leptome. Wood and bast together constitute the Fibro-vasal bundle. As here the wood is in unilateral contact with the bast, these bundles are distinguished, in structure, as Collateral. If we wish to include the sheath, which mostly appertains to the ground-tissue, in one technical term with the fibro-vasal bundle, we speak of the whole as the Fibro-vasal string.* The physiological considerations which occasion a separation of the fibro-vasal bundle into Had- rome and Leptome, have led to the choice of the term Mestome for the entire bundle.^ The bast portion of the fibro-vasal bundle which we have under * This is in all respects a preferable term to the more common one, Fibro- vasal bundle. [Ed.] CLOSED VASCULAR BUNDLES. 87 observation takes on, with clilorzinc iodine a distinct violet colo- ration : it consists of unlignified elements. Cells with broader, and those with narrower openings appear in regular order. The first are, Sieve-tubes (v), the latter (s) are the companion-cells. Not infrequently the section cuts the cross-wall of a sieve-tube, and this cross-wall appears finely punctate, after the fashion of a sieve, (compare the figure). At the periphery of these elements are always to be seen a number of cells with strongly swollen walls, and cavities almost obliterated (pr) ; these are the sieve-tubes and companion-cells which were first formed, but now have their func- tion suspended ; they correspond with the first developed elements of the wood, and in contradistinction with it are distinguished as the primary bast, or Protophloem. With clilorzinc iodine they usually take on a brownish coloration. These cells are bounded by the cells of the sheath, and the innermost of these always are marked by the special width of their cavities. The sclerenchyma cells of the sheath pass over, by means of some intermediate members, into the large-celled parenchymatous ground tissue(/). The walls of these large cells of the ground tissue are also, in fully- developed stems, coloured yellow by the chlorzinc iodine, only here and there with a dash of violet, IPassing still^nearer to the periphery of the stem, we notice that the fibro- vasal bundles are here pressed more closely together, that the intercellular passage disappears from them first of all ; in some cases the elements, particularly those of the bast, are reduced, while in all the sheath augments in strength. We notice, even in the inner bundles, that the elements of the sheath are especially thickened and lignified on the inner and outer edges of the bundle. At the sides v^^e see the more strongly thickened and lignified elements only at the sides of the two great vessels. The weaker development of the sheath at the sides of the bast and of the inner portion of the wood facilitates the passage of nutrient fluids between the fibro-vasal bundle and the large- celled ground- tissue. In the fibro-vasal bundles lying more exter- nally, with more strongly developed sheaths, the communication is maintained on both sides of the bast. Lastly, on the most exter- nal bundles, with greatly reduced bast, almost sunk between the vessels of the wood, the sheath is proportionally weakened on the outer side of the bast. The communication between the bundles and their environs is provided for in this way in Zea (the Maize), and similarly in other cases. Lateral union (anastomosis) of small bundles with large ones is commonly to be seen in the periphery 88 CLOSED VASCULAR BUNDLES. of the stem, and the reciprocal meeting always takes place laterally at the places where the great vessels lie. Close in upon the epider- mis of the stem, is a more or less strongly developed ring of tissue, the elements constituting which appear just like those of the bundle-sheath, and also react clearly with chlorzinc iodine. Such a distinct sheath of tissue bounding the epidermis is distinguished under the name Hypoderma. This hypoderma is interrupted only under the spots where lie the stomata. The hypoderma and the sheath of the fibro-vasal bundle have alike to provide for the protection of the thin-walled tissue and for the stability of the entire part of the plant, and are included amongst the ele- ments of the mechanical system,* as Stereides, w^hile the tissue which they constitute forms the system of mechanical tissue, the Stereome. In proportion as the stem must be constructed secure against flexion, so must the mechanical appliance, the stereome, be removed as far as possible towards the periphery. The crowded peripheral fibro-vasal bundles, provided alike on the side of the bast and the wood with a strong cover of sclerenchyma, repre- sent here a system of complex upright girders. The sheaths of sclerenchyma are the ties, the fibro-vasal bundles themselves are the " filling." The hypodermal hollow cylinder of sclerenchyma strengthens this action, even when, as in this case, not strongly developed. This hollow cylinder is mechanically to be considered as a combination of numerous " ties," arranged in a circle.* * I have felt unable to give a satisfactory translation of the above passage in the text. I propose therefore briefly further to endeavour to illustrate it. Two sets of phenomena have to be mechanically provided for, the one affecting the stem as a whole, the other its separate fibro-vasal bundles. First, as to the stem as a whole. It has considerable weight to bear, its own and that of its leaves. It must therefore be mechanically constructed to resist crushing. It has to bear often considerable lateral strains, from winds. It therefore must also be con- structed to resist flexion. In both these respects it can be compared with a pier of a bridge, especially a cylindrical iron pier of a lofty railway bridge. To resist flexion this is made bollow, so as to throw all the strength to tbe outside ; and, to aid in resisting crushing, it may be filled with concrete. Secondly, as to the indi- vidual fibro-vasal bundles. The sclerenchyma layers will help in the above pur- pose ; but the bundles, being on the one hand water, and on the other hand food conduits, must be protected mainly from the lateral strains which would tend to crush their elements, make them " collapse," and so cease to functionate. This protection is the main duty of the sheath of sclerenchyma. Its being thickened most on the inner and outer side of the bundles, and taking thus the form largely of two arches concave to each other, makes its structure the most advantageous for its purpose, since the main strains in such a cylindrical stem are radial. To this we may add one more factor : the course of each of the fibro-vasal CLOSED VASCULAR BUNDLES. 89 Very mstructive is it to lay some cross-sections in a solution of soda-corallin. All the lignified elements of the fibro- vasal bundle, and of the ground-tissue, stain in a short time a brilliant coral-red, the non-lignified elements rose-coloured. In the section thus treated, the sclerenchyma cells of the sheath, especially at the inner and outer edges of the bundle, stand out conspicuously, and the walls of the vessels are coloured similarly to the sheath, but somewhat more brownish. The hypodermal ring colours the same as the bundle-sheath. It is worth while now to make a radial longitudinal section through the stem. [To obtain this, take a piece of the stem about ■i or I inch long, cut it in two longitudinally through the middle, and take the sections from the cut surface of either half.] Do not be satisfied with a single section, as otherwise the chance of obtain- ing in the preparation a fibro-vasal bundle cut actually median is too slight. Such a median-cut fibro-vasal bundle can be recognised on examination of the section, in that it shows at the same time the bast portion and the annular vessel projecting into the inter- cellular passage. If the longitudinal section is laid in chlorzmc iodine we can at once readily make out a violet coloration of the bast portion, and the thin-walled cells surrounding the intercellular passage have also a violet colour. The other elements, as we have seen in the cross-section, are distinctly coloured yellow to yellowish - brown. For the rest, we prefer to select, for further study, a section which we have previously stained with soda-corallin (Fig. 41) . Here also it is desirable first of all to determine in which direction lies the surface of the stem. As in the cross-section, we pass in our examina- tion from the inner towards the outer side of the bundle. We then see that to the broad cells of the ground-tissue, in outline well-nigh square, succeed narrower ground- tissue cells, and after these follow the narrow cells of the fibro-vasal bundle-sheath (vg). These last elements, deeply stained with corallin, show marked elongation, join one another with horizontal or more or less inclined end-walls, and are provided with small, cleft-like, obliquely ascending pits. In their bundles from its lower to its upper extremity is usually not straight, but in the form of an elongated arch, the concavity outwards, they thus become akin to " struts." If they anastomose, or join together, as they do most beautifully in some cases, the mechanical analogy is still more complete, since they then re- semble the network of connecting girders with which all observers of iron bridges are famihar. I select iron bridges for this illustration, for in them, as in nature, the smallest amount of material is made to go the greatest possible way. [Ed.] 90 CLOSED VASCULAR BUNDLES. interior is to be found a peripheral layer of protoplasm of very re- duced dimensions, and in each a small nucleus. We have here to do Tvith elongated sclerenchyma cells. To the cells of the sheath succeeds the intercellular passage, and we can determine that it follows without interruption the whole length of the bundle. [Thin sections are often entirely broken into two parts by this passage.] It is surrounded by thin-walled cells, which are far shorter than those of the sheath, have more cell-contents, end in hori- zontal walls, and can be described as primary wood-parenchyma. Fig. 41.— Longitudinal radial section through a fibro-vasal bundle in the stem of Zea Mais, a and a', segments of an annular vessel; sp, spiral vessel; v, sieve-tube; s, com- panion-cells ; pr, protophloem ; I, air passage (intercellular passage) ; vg, sheath ( x 180). Into the intercellular passage project usually isolated rings; they are attached to the outer side of the intercellular passage, i.e. to that side nearest the periphery of the stem. They arise from an annular vessel torn during the elongation of the internode. Other smaller isolated rings may also often be seen clinging to this or the other side of the intercellular passage (a). Collectively they represent the remnants of the eslements of the protoxylem. Im- CLOSED VASCULAR BUNDLES, 91 pinging on tlie larger rings outwardly are one or scYcral broader or narrower spiral vessels. In the case represented in Fig. 41, only one snch was present, and that a pretty narrow one (sp). Further succeed comparatively short primary wood-parenchyma cells, with pitted and partially reticulately thickened w^alls. These cells are somewhat more strongly thickened than are those by the intercellular passage. Thus we arrive at the bast portion, recog- nisable in the corallin preparation by some thick rose-red coloured cross-walls, the sieve-plates of the sieve-tubes (v). These sieve- plates are highly refractive; and stronger magnification shows that they are pierced by fine pores, after the fashion of a sieve, and that on one side, seldom on both, is collected a highly refractive plug of " slime." In the periphery of the bast (at pr), where in the cross-section were visible the swollen cell- walls of the protophloem elements, a specially beautifully-coloured cross-plate shows up. This is a sieve-plate covered with Callus, the structure of which we shall further study hereafter on other more favourable ob- jects. The callus-plate extracts the corallin wdth special avidity, and therefore stands out so sharply stained. ° By the side of the sieve-tubes can be distinguished the companion-cells. They are narrower and shorter than the sieve-tubes, and have besides other richer contents, and a readily visible nucleus, for which we look in vain in the sieve-tubes. Cells of the sheath again bound the fibro- vasal bundle. Their end walls are here in part so strongly inclined that we can speak of them as sclerenchyma-fibres. The innermost cells of the sheath have, as the cross-section has already shown, a comparatively broad cavity. Starch grains are not found in the fibro-vasal bundle ; here, however, they are wanting in the cells of the ground tissue also. All the cells of the fibro-vasal bundle and of the ground tissue, with the exception of the cells form- ing the vessels, and of the sieve-tubes, contain nuclei. — It is clear that such a median longitudinal section of the bundle as is de- scribed above can show neither of the two great vessels. [If the section be not exactly median, or not quite thin,] such may show by deeper focussing, but it is then however very indistinct. In order to study the longitudinal section of one of the great vessels, we must look for a section which cuts the fibro-vasal bundle later- ally. We shall then see that the great vessel is obliquely pitted, and more seldom spirally thickened. In these pitted ducts the thick- ened parts form a network. The pits broaden out at their bottom, but are however only unilaterally " bordered," in that the corre- 92 CLOSED VASCULAR BUNDLES. spending pit of tlie adjoining cell of the wood-parencliyma is want- ing in a " border." These cells, too, are far less thickened than the vessels. The diaphragms, or cross-walls, of the great ducts quickly attract attention in the longitudinal section. They show a doubly formed ring, which besides only projects a slight distance into the cavity of the duct. These rings originated in a thicken- ing of the outer edges of the cross-walls, the inner unthickened part of which was afterwards dissolved (resorbed). From the number of the diaphragms we can therefore draw a conclusion about the number and size of the cells of which the duct is composed. Corresponding with the diaphragms, the vessel shows slight constrictions on its outer side. It may be of interest to put up some well-chosen cross- and longitudinal sections of the fibro-vasal bundle as permanent preparations. The colours obtained by means of chlorzinc iodine and corallin are not permanent in such preparations ; but lasting colours can be given by means of saffranin or iodine-green. Very instructive double staining can be effected if the section is first placed for a short time in iodine-green, and then somewhat longer in Grenacher's [alum- carmine ^ or for a similar time in Hoyer's ammonio-acetic carmine; instantaneous double staining can be obtained by means of picric-nigrosine, or picric- aniline- blue. In this way the alum-carmine, the ammonio-acetic car- mine, the nigrosine and aniline-blue, respectively stain the un- lignified, the iodine and the picric acid the lignified tissues in the preparation. The cell-contents take the colour of the carmine, the nigrosine, or aniline-blue respectively. The preparation can be put up in glycerine jelly or in glycerine. In the latter case, the edge of the cover-glass must be hermetically closed. For this purpose we remove with blotting-paper any glycerine that may have flowed from under the edge of the cover-glass, and surround this edge with a solution of Canada-balsam * in turpentine, benzole, or chloroform, made as thick as syrup. The operation is best performed by means of a glass rod, about the thickness of a thick match, from which the superfluous Canada-balsam is first allowed to run. Gold-size is not suited for closing glycerine pre- parations, as it will not cling to a glass surface moistened with * Of these the solution in turpentine is best, as it does not become brittle when dry ; otherwise a jerk may make the cover-glass spring. The sokition can be kept in a bottle with a bell-shaped external ground cap, to keep out the air. [Ed.] CLOSED VASCULAR BUNDLES. 93 glycerine. It is however highly desirable to cover the Canada- balsam with gold-size after it has become quite firm. For this purpose it is best not to use the gold-size too thickly, but to put on a second layer. In this operation a fine camel-hair pencil should be used. Instead of the stem of Zea Mais, in case this plant is not at our disposal, can be taken, with very similar results, the stem of Avena sativa (the Oat), or that of some other grass. Now take cross and longitudinal sections of a fully-developed leaf of Iris Jlorentina preserved in alcohol. Here also the prefer- ence is given to alcohol material, because it is more easy to obtain good sections, it contains no air, and besides that, the cell-contents are fixed, so that we can more readily obtain information about them. The section-cutting will be facilitated if the material is previously laid in a mixture of alcohol and glycerine. Lay the sections for some hours in borax-carmine, and treat them after- wards for a short time Avith iodine-green. The cell-contents have taken up carmine, which in the form of borax-carmine does not stain the cell-walls ; on the other hand, the lignified walls are stained green with iodine-green. The vessels appear stained the least, and usually also the outer elements of the sheath, i.e. those impinging on the bast of the bundle. Besides this, a group of elements with swollen walls, the protophloem, stand out in the outer region of the bast by their blue coloration. We will there- fore commence with the study of such a preparation, from which the Fig. 42 is constructed. In this figure all the cells which are especially rich in contents, and therefore are conspicuous from their red coloration, have their interior shaded. The green-coloured walls of the vessels are, on the other hand, represented darker in the figure, "while the group of protophloem elements coloured blue are left clear. The thickened elements of the ground-tissue bounding the bast, when the section is taken through the base of the leaf, are yet unlignified, and therefore remain unstained. To rapidly stain a section, it can be treated with iodine-green alone ; the staining of the cell-contents red as here described is then absent. If the iodine-green should stain only the lignified cell-walls, the exact time for staining must be carefully watched. We proceed in the examination from the wood towards the bast, and therefore from the upper surface of the leaf, turned to the interior, towards the lower surface, turned outwardly. We first determine that the number of vessels in the wood is pretty considerable, and that 94 CLOSED VASCULAR BUNDLES. their mdtli diminislies towards the bast. The vessels directly impinge upon one another, or else are separated by the slightly thickened comparatively narrow cells, with abundant cell-contents, of the primary wood-parenchvma. Similar cells also surround the vessels on the ^ r .- flanks of the bun- dle, and separate them from the ground-tissue. At the inner margin of the wood are always to be seen some crushed elements, proto- xylem elements (ss), whose walls are stained like those of the ves- sels. The bast shows again an alternation of larger and smaller cells; the contrast is here however not so striking, nor is the regularity so great, as in Zea. The cells with broader cavities are the sieve- tubes, the smaller ones, marked out by their abundant cell-contents, the companion - cells. In the outer re- gion of the bast lie the crushed protophloem elements (pr), to which we have already referred, whose function is lost, and which are provided with swollen walls Fig. 42.— Uross.section of a fibro-vasal bundle from the leaf of Ins fiorentinn. With dark contour are the vessels; the cells of the bundle which are rich in contents are shaded, ss, crushed spiral vessels; sp, broader spiral vessels; sc, scalariform vessels; v, sieve-tubes between which are the narrow com- panion-cells; pr, crushed protophloem elements; vg, sheath with wavy radial walls ; fc, section through a crystal ( x 240). CLOSED VASCULAR BUNDLES. 95 more or less deeply stained blue. This outer beast portion is enclosed by the strongly thickened sclerenchyma of the sheath, which supports the fibro-vasal bundle with a more or less power- ful string. Around the remainder of the fibro-vasal bundle a clearly marked sheath is wanting ; yet it can be determined that the cells of the ground-tissue nearest to the fibro-vasal bundle are smaller, and that they join together without a break. At the flanks of the bundle these small cells are represented by but a single layer, at the inner boundary of the vascular part, on the contrary, by several layers ; here also the walls of some of these cells are well stained blue. The transition to the larger cells of the ground-tissue, which have intercellular spaces filled with air between them, takes place by intermediate forms. By a scrutiny of the tissue in the neighbourhood of the fibro- vasal bundle it will be seen that single small cells, between the large cells of the ground-tissue, contain a highly refractive crystal (Fig. 42 k). It offers itself to us either in cross-section, or in end view ; as to its form in longitudinal section we can readily inform ourselves. Good staining can also be rapidly obtained with corallin, by which the lignified sclerenchymatous elements are stained a fiery red, or if as yet only slightly thickened, and not lignified, a bright red, the walls of the vessels brownish red, and the remaining elements a pale yellowish red. In order to control the results hitherto obtained, take some cross-sections also from a fresh leaf. We then determine that the large cells of the ground- tissue in the outer parts of the leaf con- tain chlorophyll-grains, while the cells included in the fibro-vasal bundle sheath are however wanting in chlorophyll. In fresh preparations the vessels are filled with air, whence the structures are less clear than in alcohol material. A longitudinal section through the leaf, cutting through the middle of a bundle, shows us at the inner limits of this bundle the strongly extended, partly crushed spiral vessels, which we already saw in cross-section at ss, and distinguished as elements of the protoxylem, i.e. as the first-developed elements of the wood portion of the bundle. Following are broader more closely wound spiral vessels, and then again scalariform vessels with narrow cavity. In the bast the sieve-plates show clearly only in corallin preparations. Further outwards the sclerenchyma fibres are recognised by their strong thickening, their notable length, and pointed ends. 96 CLOSED VASCULAR BUNDLES. Jl Fig. 43.— Crystals. A, crystal of oxalate of lime enclosed in a cell from the leaf of Iris florentina (x24a). B-D, figures illus- trating occurring forms of crystals. Ba, Bh, and D, seen in optical longitudinal section. C, projection, showing its planes of symmetry. As tlie crystals are directed j^arallel to the long axis of the leaf, thej show in profile in» longitudinal sections (Fig. 43, A, D). They lie in elon- gated cells of the ground- tissue, which are only a little larger than the crystals them- selves. These cells contain no chlorophyll, while the neighbouring cells usually con- tain chlorophyll. The crystals in question dissolve readily in hydro- chloric acid with- out evolution of gas, whence we readily conclude that they consist of oxalate of lime. All the crystals occurring here have an elongated prismatic form, and belong to the monoclinic system ; most of them appear geminate (twin crystals), (D). The contents of the crystallogenous cells are not stained with corallin. The fibro-vasal bundles of the Monocotyledons, if we exclude immaterial modifications, reductions, and combinations, are con- structed upon the type of the two cases we have thus investigated, and we can therefore abstain from further study of these bundles. Closed fibro-vasal bundles are not capable of increase in thick- ness, and therefore where such occurs in the Monocotyledons, it cannot be brought about through the medium of the fibro-vasal bundles. This increase of thickness results from the action of of a Cambium-ring which is found outside the fibro-vasal bundles, and is confined to the families of DracEeneae and Aloineae [i.e., the so-called " Arborescent Liliaceae "] and the Dioscoreae. For this purpose we select as a favourable object of study the plant so commonly cultivated in gardens and nurseries as Dra- ccBna rubra (more properly, Cordyline rubra). The plant must be sacrificed for the purpose of the investigation. Examine CAMBIUM-RING IN DRACiENA, 97 first with the naked eye the stem cut across, we shall notice in- side the brown cork-layer the green soft cortex, somewhere about aV i^^^ thick, towards which the yellowish hard tissue of the stem is limited by less sharply-defined bounds. At this boundary lies the cambium ring. In the yellowish tissue cf the stem it is moreover distinguished from the cylindrical centre by its lighter coloration. We now submit a cross-section to micro- scopical examination, and first with weak magnification (Fig. 44). We then see, first, in the central portion of the stem, a ground- tissue composed of rounded cells (m), in which are irregularl;^ scattered isolated cir- cular or elliptic fibro- vasal bundles. Out- ward from a definite position (the inner /") the bundles are more numerous, elongated in radial direction, and press so closely to- gether that they appear separated only by com- paratively thin streaks of ground- tissue. In these latter the cells are more strongly thickened, coarsely pitted, more or less elon£yated in the direc- FiG. 44i.—Drac(Bna {Cordyline) rabra. Cross-section through the stem. /, fibro- vasal bundles, /' being primary, /"secondary, and /'".leaf-bundles ; 771, un- lignified elements of the ground tissue ; s, lignified elements of the ground tissue, sheathing the fibro-vasal bundles ; t, tracheides ; c, cambium ring ; cr, cortex ; I, cork; -pli, cork-cambium; r, bundles of raphides (x30). tion of the radius, and clearly arranged in radial rows with often wavy course. Further on, we come to the boundary between the yellowish inner tissue and the green cortex (c). We find 98 CAMBIUM-EING IN DRAC^NA. here a zone composed of flattened tliin- walled cells, strictly- arranged in radial rows. It is the Cambium-ring, which provides for the increase in thickness of the stem. It belongs apparently to the ground-tissue. Its most fl^attened cells lie in the middle of its cross-section. In it is found the real initial layer, it may be only a single layer of cells, the cells of which, in course of constant division, produce on their inner side new elements. These divisions result from the formation of tangential walls, and produce therefore radially arranged cell-rows, which from time to time are made double by the formation of radially directed walls. Embedded in the growing tissue resulting from this cambium-ring are numerous fibro-vasal bundles in all stages of development. The youngest consist of a group of thin- walled cells, the oldest are already perfect at their inner portion, while the thin-walled outer portion [i.e., the portion towards the perijDhery] are still immersed in the cambium-ring and in course of development. From the position where the fibro-vasal bundles appear crowded together, and the cells lying between them have acquired a radial arrangement, the tissue is Secondary, produced by the activity of the cambium-ring. The Cortex (cr), succeeding outwardly to the cambium-ring, consists of rounded cells. Between these, especially in the inner part of the cortex, occur single cells, in which lie fine needle-like crystals, in each cell closely packed together into a bundle (r) . These are the bundles of so-called Raphides, consisting of oxalate of lime. They are here seen in end view. Individual raphides-containing cells are sure to be opened by the razor in cutting, and the fine needles will therefore be found lying scattered over the section. The first of the cortical cells contain chlorophyll-grains. In the cortex are also visible single round cross-sections of bundles (f") which are passing outwards into the leaves. There succeeds a thick layer of thin-walled, colourless cells, arranged radially (Z), which on its outer side passes over into a brown, less regular, tissue. This is the Cork-layer, consisting of growing colourless cork-tissue in its more internal, but of old irregularly elongated and discoloured cork-tissue in its outer parts. Especially instructive are cross-sections treated with corallin. The fibro-vasal bundles are thereby sharply defined. The corallin also always colours deeply the lignified cells of the secondary ground-tissue, but of another shade. The unlignified cell-walls appear pale rose-coloured. In the cortex, the cells containing VEGETABLE MUCILAGE. 99 raphides appear now filled with clear coral-red to orange-coloured contents. We easily determine, by the aid of this coloration, that the raphides lie embedded in a homogeneous Slime [or Mucilage], which accumulates the corallin. Besides the power, which it shares with aniline-blue, of colouring the Callus of the sieve- plates, corallin has the specific property of staining Vegetable Slime [or Mucilage]. If we lay the longitudinal section of Draccena, stained with corallin, in alcohol, and, moreover, allow this last to boil, the slime remains none the less stained. From this we can conclude that, as far as knowledge goes, it consists of a Starch- slime, while the slime the result of the degradation of cellulose is decolorized sometimes even in cold, but at any rate in boiling alcohol. 7 Gums are not stained by corallin ; mixtures of slime and gum (Gum-slime), according to the proportions. On the other side, we can determine that watery solution of nigrosine does not stain the slime here present, even after long- continued action ; while it stains the slime of Bumex (see page 79). With this insight into the cross-section, which indeed suffices to give us information on the phenomena of increase in thickness, we will be content, and in this case abstain from the study of the further peculiarities, as also of the longitudinal section of the stem. EEMAEKS ON CHAPTER VIII. ^ On the fibro-vasal bundle compare, above all, De Bary, Comparative Anatomy (Eng. trans. 1884), especially in chapter viii., where will be found the entire older literature. Numerous more recent researches into the mori^hology of the fibro-vasal bundle have not since then received collective treatment. This has, on the other hand, been done, as far as the anatomo-physiological works are concerned, by G. Haberlandt, in Encyklopcidie der Naturivisseiischaften, — Handbuch der Botanik, Bd. II. p. 593. 2 The terms vascular-part (Gefasstheil), and sieve-part (Siebtheil), introduced by De Bary, Gomj). Anat. chap. viii. [In the English translation of this work the terms are replaced by " xylem" and " phloem " respectively.] ^ Compare Haberlandt, Bie Entwicklungsgeschichte des mech. Geioehesy stems der Pfianzen (The Development of the Mechanical System of Plants). ^ Schwendener, Das median. Princip. im anat. Ban der Monocotijlen (The Mechanical Principle in the Anatomical Structure of Monocotyledons). ^ This staining-fluid introduced by Szyszylowicz. Compare Bot. Centralb. Bd. XII. p. 138. ^ Compare Tangl, Jahrb. far wiss. Botanik. Bd. XII. p. 170. 7 Compare Szyszylowicz, as above. 100 OPEN VASCULAR BUNDLES. CHAPTER IX. OPEN COLLATERAL FIBRO-VASAL BUNDLES. Material Wanted. Eunners of the Creeping Buttercup {Banunculus repens). Fresh, or in alcohol. Stems of the Celandine {Chelidonium majus). In alcohol. Stems of Aristolochia Siplio, ^ to |, and f inch thick. In alcohol. As a first example for the study of open collateral fihro-vasal bundles, as they are peculiar to the Dicotyledons, we select the creeping stems (runners) of Ranunculus repens (the creeping Buttercup) . We stain the sections with corallin, in order to facili- tate the task. The cross-section shows that the fibro- vasal bundles are completely isolated from one another, but arranged in a single circle in the stem. The ground-tissue consists of rounded cells, which become smaller towards the periphery of the stem, contain chlorophyll-grains, and have between them large intercellular spaces. The epidermis forms the surface of the stems ; in the interior, through the stretching apart and destruction of the cells, the stem is hollow. The fibro-vasal bundles give throughout the same impression as those of the Monocotyledons ; the same parts are recognisable in the same order. The vessels nearest to the inner side of the bundle have not taken up much stain ; they are annular and spiral vessels (Fig. 45 s). The more remote larger vessels, in part, however, of the same or even smaller size, have stained brown- red. Their outline is somewhat angiflar ; even the cross-section betrays that their walls have bordered pits (m). Between these vessels lies the thin-walled primary wood-paren- chyma. In the bast is again very visible the alternation of larger sieve-tubes (v) and smaller companion-cells. The bast is, however, separated from the wood by a multilamellar layer of cells arranged radially. These cells have arisen from the activity of a cambium (c), and betray this by their radial arrangement. A cambium- OPEN VASCULAR BUNDLES. 101 layer, separating the wood from the bast, is met with here as a novelty, as a distinction from Monocotyledons. To be sure, the activity of this cambium is extremely limited ; still the layer serves to give to the bundles a place amongst the V " open " tan- ^^SS^Opl dies, i.e. those capable of fur- ther develop- ment. The cambium has here formed only a multi- lamellar layer of thin-walled cells, and with this has ceased activity. Out- wardly, the bast is protected by a string of s c 1 e r e nc hyma- tous elements, which have stained a beau- tiful coral -red. The inner boundary of the bundle is also surrounded by similar, but less strongly thickened elements of the sheath. At the flanks of the fibro-vasal bundle, the elements of the sheath do not close together ; a gap remains, which points out the limit between wood and bast. In the longi- tudinal section we can readily determine the presence of annu- lar, spiral, and pitted vessels [pitted ducts], and between them elongated cells of the primary wood-parenchyma ; then follow thin- walled cambium-cells, sieve-tubes, and companion-cells ; finally, elements of the sheath, which end on to one another with slightly oblique, porous, cross- walls. The fibro-vasal bundle of Chelidonium majiis is so similar to that of Banunculus repens, that the cross-section can be understood Fig. 45.— Cross-section through a fibro-vasal bundle from tte runner of Ranunculus repens. s, spiral vessels ; m, vessel with bordered pits ; c, cambium; v, sieve-tubes; vg, sheath (x 180). 102 LATEX-SYSTEM. without farther reference. Here, however, we prefer alcohol material. The wood shows large vessels, closely crowded together, which, in the older parts of the stem, have yellowish walls. The bast is strongly developed ; between the two lie the thin-walled radial rows of cells developed during the brief activity of the cambium. The sheath is only represented by a bundle of strongly- thickened sclerenchyma cells at the outer edge of the bast. In older parts of the stem these cells likewise assume a yellow colour. Sepai^ted from the epidermis by about two layers of cells is a strongly- formed ring, composed of the same kind of sclerenchyma- cells as protect and support the bundle, and serving as a common sheath around the inner tissue of the stem. In and near the bundle, however, we meet, for the first time, a new element — the milk-tubes [latex-tubes, laticiferous- cells]. In the bast portion of the fibre- vasal bundles, and also near the inner limits of the wood, but especially numerous at the flanks and near the outer edge of the string of sclerenchyma, and single examples also in the remoter ground-tissue between the fibro- vasal bundles, we notice cells with dark-brown contents. These contents consist of the orange- red latex of Ghelidonium, coagulated in alcohol. The cells in question are so striking that it is impossible to overlook them. They are all thin-walled, even those which are inserted within the outer edge of the string of sclerenchyma ; they are not even distinguished by any special form. The latex-tubes can be found very easily also in radial longitudinal sections.* They are recognised at once by their yellowish-brown contents. They present the appearance here of long tubes running thereabouts parallel to the long axis of the stem. Without difficulty the existence of cross-walls in these tubes can be determined. These cross- walls are more or less clearly pierced in their centres by one or more pores ; they are entirely wanting, also, here and there, where we should expect to find them. In not exactly rare cases, one or another of the vessels in the fibro-vasal bundle shows itself to be full of coagulated latex. Exceedingly instructive preparations of cross-sections can be obtained, alike for fibro-vasal bundle and latex-tubes, if the sections are stained with corallin, and then a drop of potash run * To obtain radial longitudinal sections, take a piece of the stem about J inch long, cut it in half down the middle, and take sections from the cut surface. If the half-stem is too tbin to hold readily, stick the point of a needle (in a holder) through it from side to side, and parallel with the cut surface ; then, laying the needle flat on the left-hand index-finger, cut the sections. [Ed.j LATEX-SYSTEM. 103 nnder tlie edge of tlie cover-glass. The vessels now appear foxy- red coloured, the elements of the sclerenchyma rose-red, while the cross-sections of the latex- tubes stand out sharply with their dark-brown contents. If delicate longitudinal sections are laid in 45 per cent, acetic carmine, we may succeed in identifying nuclei in the latex- vessels ; this demonstration is not, however, the easiest possible task. Lateral communica- tions of the latex-tubes cannot be observed in Ghelidonium. [Anastomosing latex- vessels (laticiferous vessels) can, however, be found in the Poppies (Papaveracece), the Bellworts (Campanu- lacece), and in the ligu- lifloral section of the Composites (CicJioria- cece), as, e.g. in the Dandelion. Of these we can select as an example, the garden Scorzonera (S. hispan- ica) , not infrequently grown in kitchen-gar- dens for its parsnip-like roots. Tangential sec- tions, taken from the external part of the- root, a short distance below the exterior, if treated as described above, will show in the bast portion of the fibro-vasal bundles an extensive network of latex-vessels filled with their very granular contents, shown in Fig. 45*.] An extraordinarily favourable object for the study of the LFiG. to*. — Latex vessels in the bast of Scorzonera hhpanicn, tangential section. B, a portion of A, more highly magnified. (After Sachs.)] 104 OPEN BUNDLE OF ARISTOLOCHIA. growth in thickness of the dicotyledons is AristolocMa Sipho [a hardy decidnous climbing plant not infrequently cultivated in England] , and material for the investigation of which will there- fore be probably not difficult to obtain. Take first a cross-section through a twig about ^ or -^ inch in thickness. This section shows with a lens an internal spongy pith, around this a belt of isolated fibro-vasal bundles, round this further a continuous white ring, then green cortical tissue, and finally a yellowish -green peripheral rind. By weak magnification with a low power under the micro- scope we can determine that the pith consists of round, large cells, in part filled with air. In the fibro-vasal bundle the wood appears darker, pierced with holes which are the vessels ; then follows the cambium zone, com- posed of narrow, radially ar- ranged clear cells, and then the large-celled bast, which appears somewhat clear, and not with the regular arrangement shown by the cambium zone. Each bundle is bordered, especially in its outer part, by a paren- chymatous tissue, containing chlorophyll-grains to some ex- tent, perhaps also reserve food materials. The white ring, succeeding outwardly to these, is composed of strongly thick- ened sclerenchyma-cells ; be- tween the fibro-vasal bundles it projects inwards in a somewhat wedge-like fashion. Imping- ing on the ring towards the exterior is a tissue containing chlorophyll, the innermost layer of which, bounding upon the sclerenchyma ring, is marked by its richness in starch, and belongs to the category of so-called " starch-sheaths." After treatment with iodine this starch sheath stands out very clearly. There follows a tissue, likewise chlorophyll- containing and with narrow cavities, with glistening cell-walls, more strongly thickened in the corners, in which, by this last peculiarity, we recognise coUenchyma. [Compare Fig. 45**, show- ing collenchyma-cells from the leaf -stalk of a Begonia.'] Outside [Fig. 45**. — From a transverse section of a leaf-stalk of a Begonia, showing the col- lenchyma cells, cl, underlying the epider- mis, e ; the collenchyma cells contain a few chlorophyll-grains, cM ; p is an internal parenchyma cell ; c, the cuticle. These col- lenchyma thickenings are capable of greatly swelling. ("After Sachs, x 550.)] OPEN BUNDLE OF ARISTOLOCHIA. 105 all is found the epidermis. This general information Avill suffice, and we will now turn to the study of a single bundle. This can only be on very thin sections. Such we prepare with care from the alcohol material, which, in order to be able to cut it better, we have previously placed for some time in a mixture of half alcohol and half glycerine. These sections also are stained by a longer action of corallin. The figure of a fibro- vasal bundle in course of development from a twig of the current year, placed in alcohol about the beginning of June, appears as in Fig. 46. The fibro- vasal bundle begins at its inner margin with thin-walled primary wood-parenchyma (p), in which are enclosed narrow vessels (the protoxylem) gradually becoming broader as we pass towards the exterior. The primary wood-parenchyma becomes at the same time thicker walled. This applies still more to the vessels, while the space between is occupied by still more strongly thickened tracheides with bordered pits (m'). The fully deve- loped vessels and tracheides, as well as the thick-walled wood- parenchyma, stain an intense red in the corallin, the thin- walled wood-parenchyma only a light rosy colour, against which the innermost vessels therefore stand out very clearly. The two largest vessels (m") of the fibro- vasal bundle represented in the figure are in course of development. Between the two developing vessels lies a young thin-walled tissue, arranged in radial rows, and therefore arising through the activity of the cambium. The cambium-zone bounds the outermost limits of the two large vessels ; in this a specially flat, but otherwise not sharply- defined layer of cells, represents the initial layer. Succeeding towards the exterior are the thin- walled elements of the bast, the radial arrangement of the inner portion of which also indicates a se- condary derivative of the cambium. In this bast are the sieve- tubes (r), clearly distinguishable from the companion-cells accom- panying them by the abundant contents of these latter. Between sieve- tubes and companion-cells there are also scattered cells of the bast parenchyma, containing starch. The outer part of the bast, the protophloem, is composed of narrower sieve-tubes, which there- fore are not so sharply contrasted with their companion-cells. The bast is separated from the sclerenchyma ring (sk) by a large- celled cortical parenchyma, devoid of intercellular spaces. The sclerenchyma ring appears quite as deeply stained as the lignified part of the fibro-vasal bundle. Under the pressure of the elements newly developed from tlie cambium, the protophloem elements are 106 IXTEEFASCICULAK CAMBIUM. soon cruslied. — The formation of tlie interfascicular cambium is very instructive in such sections. With the commencement of the activity of the cambium in the fibro-vasal bundle, the ground tissue cells impinging laterally upon it (i.e. the cambium) have become stretched, and partition walls are formed in them (ic). Thus streaks of cambium are developed, passing through the c?>:^cro'='-'np, periphloem ; so, scalariform vessels ; sp, annular and spiral vessels; v, bast (x 26). 150 COMPOUXD YASCULAR BU^^DLE outer cortical cells have stained more clieiTy-red with the saffra- nin ; the inner, strongly-thickened ones, more rose-red. The thickened elements of the cortex cease suddenly, and there succeed two or three layers of polygonal cells, elongated somewhat tangen- tially, and united without gaps, which are coloured cherry-red. These cells have hei-e the position of the endodermis, but they are present in several layers, without the undulated band or the characteristic thickening. On the other hand, like the cells of the endodermis, they are cuticularized, and withstand sulphuric acid well. We will designate this sheath, therefore, as the inner sheath (vi). Further in follow several layers of equally wide cavitied cells, of like diameter with one another in the cross-section, often contain- ing starch, and with walls white and shining, as if swollen. With ' shorter action, these are not stained ; with longer, they are orange- red. These cells are here found in the position of the pericam- bium, and may therefore, as in the Ferns, be called periphloem (pp). We now notice the xylem bands stained beautifully cherry- red. They consist of broad scalariform yessels (^c) in immediate contact with one another, -i.e., without intermediate cells, and, at the narrow edges, of protoxylem elements, i.e., of narrow cavitied annular and spiral vessels (sp). The ligneous bands in Lycopo- dnim complaizaUim run aci^oss the cylinder, and more or less parallel to one another. They are somewhat concaye on one side, on the other correspondingly conrex ; and we can determine, if we take note of the natural position of the rising stem towards the earth, that the bands appear parallel to the surface of the earth, and always with the concave side turned upwards. The small vascular bundles of the leaves, after they have entered into the centi-al cylinder, join on to the spiral-vessel group of a ligneous band, just as in the Ferns. The ligneous bands not inft^equently anastomose, an example of which can be seen in the lower bands of the sketch (Fig. 58). In the erect stems oi Lycopodium Selago the whole of the ligneous bands are combined, and form a star. The ligneous bands are surrounded by a single layer of thin- walled, narrow-cavitied cells, which we, as in the Ferns, can desig- nate Avood-parenchyma cells. At the edges they pass, with their protoxylem elements and Avood-parenchyma cells, out to the tissue of the periphloem. Between the bands formed by the wood lie^ cells with white, sti-ongly refractive walls ; they have naiTow cavities, only a middle row is distinguished by somewhat broader cavities. These bands of tissue separating the portions of wood OF LYCOPODIUM. 151 form the bast ; the larger elements in this latter are the sieve-tubes (u). In specially favourable cases of staining, the walls of the sieve-tubes are rose-red, while the other elements of the bast re- main colourless. At the edges of these bands of sieve- tubes the protophloem elements are distinguished by the narrowness of their cavities. With these protophloem elements the sieve-tubes reach the periphloem, the considerably larger cells of which show up clearly against the wood and bast. At the inner limits of the periphloem, the inner part of the fibro-vasal cylinder, consisting of the wood and bast, can be easily broken away in cutting the sections. The longitudinal sections show us : most externally, the epidermis ; then, the broad cortical cells running obliquely towards it ; further, the sclerenchyma fibres of the outer sheath ; after this, the inner sheath of elongated parenchyma ; the periphloem with white, thicker walls, and cross-walls situated obliquely; the scalariform vessels, and the narrow, in part very greatly stretched, annular and spiral vessels ; finally, also, the elements of the bast. These last consist of very elongated cells, joining with one another with more or less oblique end- walls. With the aid of corallin and aniline-blue it is possible, but very difficult, to recognise the com- paratively inclined sieve-plates. Only the broader cells in the bast are sieve-tubes ; the much more numerous, narrow cells, filled with bright granular contents, are companion-cells. NOTES TO CHAPTEB XIII. [* The wood-parenchyma of the vascular bundle of Ferns is very generally designated "packing cells."] - Compare also de Bary, Comparative Anatomy (Engl, translation), p. 170. 152 SECONDARY DEVELOPMENTS IN CORTEX. CHAPTER XIV. COEK, LENTICELS ; THE FALL OF LEAVES. Material Wanted. Twigs of the Elder {Samhucus nigra) about i in. thick ; ditto about I in. thick. Fresh, or in alcohol. Pretty old twigs of the Laburnum {Cytisiis Laburnum). Fresh. Fine bottle cork. Pretty old twigs of the Ked Currant {Eibes rubrum). Fresh. Base of leaf -stalks, with piece of twig attached, of the Horse-Chestnut {JEsculus Hippo castanum). In autumn. Fresh, or in alcohol. Or the same of the Kentucky Coffee-tree {Gymnodadus canadensis), the Bastard Acacia {Eobinia Pseudo-Acacia), or one of the Poplars {Popuhts dilatata). Strong leaves of Gymnodadus canadensis, or Ailanthus glandulosa. Fresh. Or of the Ash {Fraxinus excelsior) or the Walnut {Juglans regia). We have already, upon various objects, had the opportunity of making ourselves acquainted with the position and structure of cork. None the less will we once again turn our attention to this object, in order to study on the one hand the Lenticels, and on the other hand the structure and reactions of the wall of cork cells. ^ Cross-sections through a twig, about | inch thick, of Samhucus nigra (the Elder) show us around the large-celled pith the sepa- rate fibro-vasal bundles already bound into a ring by the interfas- cicular cambium. The cambium ring has also already commenced its activity, and in the fibro-vasal bundles, as well as also between them, has formed in the usual fashion, inwardly secondary wood, outwardly secondary bast. The primary bast appears out- wardly supported by sclerenchyma fibres. The cortex is from ten to fifteen cells thick. The projecting ridges of the stem exhibit a strong hypodermal sheath of collenchyma, which in the grooves is reduced to a layer two or three cells thick. Under the stomata the collenchyma-sheath is interrupted by the green cortical parenchyma, which here extends to the epidermis. In parts of the stem about ^ inch thick the for- CORK. 153 mation of cork-layer commences, always by tangential division of the outermost collenchyma-cells immediately bordering on the epidermis. The inner of the sister-cells thus pro- duced again divides, and it is then the middle cell [of the three radially-disposed cells], which further acts as a cork- cambium cell. This is easy to recognise, even after the periderm has be- come multilamellar (Fig. 59, pK). Outermost in each [radial] row lies the outer, while innermost lies the inner portion of the original collenchyma cell (cZ) ; the flattened cell (jpli), bounding the inner portion externally, is the cork -cambium or phellogen-cell. In fortunate cross-sections we can, more- over, determine that the for- mation of a connected cork-layer is preceded by a peculiar process, which commences under the stomata. The primary cortical cells which surround the air-chamber commence to divide, and the divisions encroach laterally upon the surrounding collenchyma cells. Soon is formed under the stoma a layer of dividing cells in the form of a meniscus (Fig. 60, pi), which produces externally colourless cells, which become rounded (Z), and internally cork- cells, or Phellem {pd). The outer cells are distinguished as packing-cells.* They become brown, but not corky; and moreover, as they increase in number, they soon cause such a pressure on the epidermis that this is torn into fissures. In this way is produced the cortical pore, or lenticel. If a twig is examined with the naked eye, the lenticels appear as grooves, surrounded by two lip-like cushions. The brown colour of the packing-cells is specially noticeable. On younger paints of the stem the lenticels appear as oval, somewhat projecting spots. Still younger stages Fig. 59.— Cross-seetion through the.snrface of a young stem of Sanxhucus nigra. Epidermis ; ph. phellogen ; cl and cl, outer and inner parts of the original collenchyma cell (x 210). * Fiillzellen, translated by Bower & Scott (De Bary, Comp. Anai. cells." [Ed.J as " complementary 154 CORK. are marked ont by somewhat brighter colour. The section mnst be taken through such places in order to show the youngest stages of development. Not till after the splitting of the epidermis do divisions begin in the neighbouring collenchyma, which result in the formation of the periderm. The packing-cells of the lenticel are separated from one another ; proportionally as they outwardly undergo disorganization, they are replaced by the action of the cambium. The intercellular spaces of the packing-cells are filled with air ; between them is communication of the inner tissue of the stem w^ith the surrounding atmosphere. They compensate, therefore, for the stomata in older parts of plants, in which the *#f pd pi. Fig. 60.— Cross-section through a Lenticel of Sambucus nigro ; c, epidermis ; p7i, phellogen ; I, packing cells ; pZ, cambium of the lenticel ; pd, phellem (x 90). cork-formation has begun. For the winter, somewhat more com- pact and resistant packing-cells are formed. A specially formed closing layer of narrow cells close together is not present in Samhucus in winter, while they are met with in many other plants, as also are intermediate layers, which, formed just like the closing layer, are from time to time interposed between the packing cells during the period of vegetation. The cells of these closing and intermediate layers become corky, but allow radially running intercellular spaces between them; so that they do not effect complete closure.^ In older parts of the stems of Sambucus the periderm has longitudinal clefts. These pass through the lenticels, without, however, injuring them. The lenticels persist even on CORK. 155 quite old stems, while the outer layers of periderm between them scale off. It is recommended to study the structure of cork-cells in the first place upon Cytisus Laburnum [the Laburnum], because here they are remarkably thickened. Cross-sections through the cortex of older stems show the periderm formed of only one kind of cork-cells. These cork-cells are arranged in regular radial rows. The youngest cork-cells are colourless, the older coloured yellow, the oldest yellow-brown. Those lying at the periphery appear tangentially stretched, often to the disappearance of their cavity. All these cork-cells are greatly thickened, especially on their outer side. In them can be readily distinguished, even without the aid of reagents, the delicate middle-lamella, or primary membrane, separating the cells, a strong, distinctly laminate, secondary thickening-layer, and, on the inner side of this latter, a tertiary thickening-layer. Consequently each complete wall separating two cell-cavities consists of five distinct layers : — the middle- lamella, which here represents the primary cell-wall, and is lignified ; the two secondary thickening-layers, which alone are corky ; and the two tertiary thickening-layers, which often retain their cellulose character and are therefore distinguished as cellulose- layers, but here, however, are a little lignified. With chlorzinc- iodine the cork-cells colour yellow to yellow-brown, the younger darker than the older, their tertiary layers the darkest. The cha- racteristic reactions of the cork-material or Suberin are obtained by potash, maceration-mixture, and chromic acid.^ We first treat the sections with potash, and determine that the cork-cells become yellow. We warm the section carefully under a cover-glass upon the object- slide, and find at once that the intensity of the yellow coloration has increased. With the maceration-mixture (chlorate of potash and nitric acid) we obtain a reaction for Ceric acid. If unwarmed, the mixture first acts by colouring the cork-cells yellow-brown, besides which all their parts become clearer. If the preparation is now boiled upon the object-slide, if necessary more of the reagent being added, soon of the whole section only the corky layers of membrane remain behind ; these finally swell and fuse into a colourless, globular mass. It is the so-called Ceric acid, which is readily dissolved in alcohol, and still more so in ether. If pretty concentrated solution of chromic acid is permitted to work upon the section, of this there finally remains, as before, only the corky layers of the cork-cells. After a 156 CORK. longer time these themselves become so transparent that it is difficult to find them again, although they do not disappear. Notwithstanding that the middle-lamellae have been dissolved, the secondary thickening-layers adhere to one another. The bottle-cork (of Quercns suber [the cork-oak]) consists of almost cubical, thin-walled, comparatively large cells, which gradually pass over into somewhat more strongly thickened, flatter cells, marking the limits of the year's production, to which the cubical cells again succeed. Addition of potash solution colours the section yellow, and first of all the somewhat thicker walled cells marking the year's limits. Upon these it can now be determined that here also each [double] wall consists of five layers, just as we found it in Cytisus. Here also the tertiary thickening layer does not give at first the cellulose reaction, excepting after corresponding treatment. The reactions for suberin occur here more beautifully than in Cytisus, especially the Ceric acid reaction. Often from the phellogen are formed not only centrifugal cork-cells, but also centripetal cortical cells, the so-called Phelloderm. Rarely, however, does this phelloderm attain such a decided thickness as in the species of Bibes. If we prepare cross-sections through older stems of Bibes rubrum [the Red Currant], we find under the thin- walled brown cork- layer, first the phellogen, then a thick layer of chlorophyll- containing cortical cells. These last also are arranged in radial rows, which coincide with those of the neighbouring cork. In the inner part of the phelloderm the radial arrangement is lost, in consequence of subsequent extension. The innermost phelloderm- cells border on the collenchyma of the cortex. All the structures proceeding from the phellogen are collected under the term periderm; in Bibes, therefore, the periderm consists of cork (phellem) and cork-cortex (phelloderm). It is also of interest to take sections through this year's stem of Bibes rubrum, in which the cork formation has for a short time begun. We can here see the first commencement of the phelloderm formation, and at the same time determine that in the plant in question the phelloderm is situated pretty deeply in the cortex. The more external tissue, cut off by the cork-layer from access of sap, perishes, becomes brown, and forms the so-called Bark. The fall of foliage leaves in autumn * results from the inter- V * What follows is a translation of pp. 240-241 of the larger work, Das Botan- THE FALL OF LEAVES. 157 position of a separating layer [or what we may call an absciss layer] which is formed earlier or later during the period of vege- tation, and which cuts across the articulation of the leaf-stalk. This absciss layer is the only new formation the existence of which can be proved at the base of the leaflets of a compound leaf, and also at the base of the primary leaf-stalk of many leaves (as those of ferns, and numerous phanerogams); the scar is then somewhat later closed by a cork-layer, or, as in the ferns, by simple drying of the surface cells. In other cases, on the other hand, before the fall of the leaf, is formed at the base of the primary leaf-stalk a periderm, separated from the absciss-layer by a few layers of rounded cells, and Avhich, after the fall of the leaf, is only brought into a state of more active development. * We will examine the processes a little more closely in JEscidus Hippo- castanum (the Horse-Chestnut), during the fall of the leaf. The research is carried on upon alcohol material just as well as upon fresh, so that we can become independent of the time of the year. The absciss-layer, as well as the cork-layer, lie in the position which is clearly visible externally as the boundary between the brown tissue of the cortex and the green tissue of the leaf-stalk ; upwards this boundary strikes the angle which the leaf-stalk forms with the bud in its axis. We cut off the leaf-stalk, with the surrounding parts of the cortex, from the twig, and halve it in a median line. We take now a number of delicate longitudinal sections with the razor, in which we take care that some of them also cut through a fibro- vasal bundle. In such longitudinal sections, prepared from fresh material, and examined in water, the cork-layer is at once observable, even with low magnification, as a clear brownish streak, between the deeper brown cells of the cortex, and those of the leaf-stalk. In alcohol-material the cell- walls of the cortex and of the leaf-stalk remain colourless. The cork layer is clearly reddish-brown, especially on the cortical side. It consists of six or eight layers of cells, and joins on to the peri- derm of the twig with its margins. Its phellogen lies on the side of the stem. This cork-layer is penetrated by the fibro-vasal bundles of the leaf. Separated by some layers of cells from this periderm [and on the leaf-stalk side of it] the absciss-layer, only ische Practiciim, inserted here by request of the Author. I commenced an investigation into this subject, still in progress, in the autumn of 1882, in the Botanical Laboratory at Bonn, under the guidance of the Author. The leaves referred to above were included in that research, and the results, as far as they go, substantially coincide with what follows. [Ed.] 158 MECHANISM OF THE a few cell-rows thick, runs within the roundish cells of the leaf- stalk, recognisable by its yellow colour, the newly intercalated dividing walls, and the more copious contents of its cells, which likewise contain small starch-grains. It is first formed shortly before the fall of the leaf, while the periderm was already present much earlier, and is continued through the living elements of the fibro-vasal bundle. For the rest, the cells of the leaf-stalk are almost completely emptied of reserve food materials ; they con- tain, as treatment with iodine shows, only a trace of starch. In the same way starch is w^anting, alike in the leaf and in the cortex, within the fibro-vasal bundle, although in the cortex it is very abundantly represented in the vicinity of the fibro-vasal bundle. The thin-walled elements of the fibro-vasal bundle are, on the other hand, filled with highly refractive masses, which give a tannin reaction. If fresh sections are examined in water, this latter commences very quickly to fluoresce wath a bluish tone, from the sesculin which comes out of the stem. Numerous cells of the leafstalk contain clusters of crystals, or a single crystal, of oxalate of lime. Preparations treated with aceticized methyl- green show in the cells of the leaf- stalk a remnant of the proto- plasmic sac, the nucleus, and chlorophyll-grains. The yellow grains, into which the chlorophyll-grains break up, give to the leaves their autumn tint. The fall of the leaf takes place inside the absciss-layer, the cells of which become rounded, and so dis- united ; the fibro-vasal bundle is torn through in the correspond- ing part. The leaf-scar is covered by the roundish parenchyma- tous cells, which lie between the absciss-layer and the cork-layer, and therefore at first appears greenish. These cells become brown, and dry up quickly in air. The exposed and broken elements of the fibro-vasal bundle wither, and their walls, as well as their contents, become dark-brown. Under these decayed cells a phello- gen is now formed also in the fibro-vasal bundle. It arises through division of all the elements provided w^th living contents. In the vessels which are devoid of a protoplasmic cell-body, the process naturally is interrupted. These, on the other hand, are quickly crushed by the dividing cells. Thus is developed on the leaf-scar, a completely closed cork-layer, which further increases somewhat in thickness. Between the cell-rows of this, the flattened and drawn- out ends of the vessels can later on be still recognised. The dead ends of the fibro-vasal bundles, hoAvever, continuously project, to the number usually of 5 or 7, out of the shield-like leaf-scar. As AUTUMNAL FALL OF LEAVES. 159 a specially favourable object for the study of the processes here de- scribed, may be mentioned Gymnocladus canadensis [the Kentucky Coffee-tree], where it is at our disposal, and also llobinia Fseud- Acacia [the common bastard Acacia], or Populus dilatata. The results of the investigation will agree in the main with the pro- cesses above described. If strong leaves of Gymnocladus canadensis, or of Ailanthus glandulosa are laid in a damp, dark chamber, the former in about 48 hours, the latter in four days, lose their leaflets on the slightest touch. ^ Longitudinal sections through the place of insertion of their leaves shows that an absciss-layer has been formed at their base. Such an absciss-layer commences its forma- tion also at the base of the common leaf-stalk at about the sixth or seventh day. Under these conditions, however, a periderm is not formed under the absciss-layer. Fraxinus excelsior (the Ash) and Juglans regia (the Walnut) can also be used in this experi- ment. NOTES TO CHAPTER XIV. ' For literature see De Bary, Comparative Anat. (Engl, trans.), pp. 544 et seq. V. Hohnel, Stzber. d. math, naturw. CI. d. k. Akad. d. Wiss. in Wien, Bd. LXXVI. 1877. 2 Klebahn, Jen. Zeitschr. f. Naturw. Bd. XVII. 3 Introduced by v. Hobnel, see above, p. 522. 4 Von Mohl, Bot. Zeitang, 1860, pp. 1, 132, 273 ; Bretfeld, Jahrh.f. iciss. Bat. Bd. XII. p. 133 ; Van Tieghem et Guignard, Bull, de la Soc. Bot. de Fiance, 28 July, 1882. 5 Von Mohl, as above, p. 271. 160 ANATOMY OF LEAF-STRUCTURES. CHAPTER XV. STBUCTUEE OF FOLIAGE AND OF FLOKAL LEAVES. TERMINA- TIONS OF THE FIBEO- VASAL BUNDLES. ■ Material Wanted. Leaves of the garden Eue {Buta graveolens). Fresh. Leaves of the Beech {Fagus sylvatica). Fresh. Flowers of the Mullein {VerhasciLm nigrum). Fresh. Petals of the Poppy {Papaver Bhoeas). Fresh. We will now endeavour, by means of a series of examples, to make ourselves acquainted with the structure of leaves. We turn first to foliage leaves, and to kinds which exhibit the smallest amount of differentiation of their inner structure. Our first example shall be Ruta graveolens [the garden Eue], the leaves of which also usually remain fresh during the winter. The leaves of this plant are bipinnate, the leaflets ovate. Held towards the light, these leaflets show clear spots ; these are the glands, filled with etherial oil, internal glands in the tissue of the leaf. [To the oil contained in them the leaf owes its strong smell when bruised.] We take first surface views [by means of surface-sections] of the epidermis, and determine that the upper side (Fig. 61, J.) universally has no, or but few, stomata ; these, on the other hand, are numerous on the under side (B). Elongated pits, filled with air, lead up to the stomata. Above the glands, as can be determined upon either upper or under epidermis, lie usually four cells (A, sc). These cells form in the centre a shallow depression. In thicker parts of the section, where the glands are not opened by the razor, can be seen in these a highly refractive yellow drop. With still deeper focussing we can determine that under the epidermis of the upper side lies a green tissue of cells, which appear round in optical section (A, p). These cells are almost completely separated from one another, and the intercellular spaces filled with air. Below the under epidermis are situated cells, likewise green and rounded in optical section, but in much smaller number (B, s). These cells, STRUCTURE OF THE LEAF. 161 also, are separated bj air, and leave, especially under the stomata, wide air-chambers. After obtaining this information, we proceed to cut cross-sections ; these we prepare, perpendicularly to the long axis of the leaflet, in the manner already known to us ; viz., by placing the leaflet, for the purpose of cutting, between two pieces of elder-pith. The cross- section shows us the leaf -tissue or mesophyll, between the upper and under epidermis. Proceeding from above downwards, we see first the epidermis of the upper side (Fig. 62, ep'), then a double parallel layer of chlorophyll- containing cells, elongated perpendicularly to the surface of the leaf, which we call the palisade layers. We already proved by the surface-section that these cells are laterally more or less com- FiG. 61. — Epidermis and underlying tissue of the leaf of Ruia gravcolcns. ^.epidermis of the upper side; sc, epidermal cells over a gland ; p, palisade cells ; B, epidermis of the under side ; s, spongy parenchyma. In A, the intercellular spaces, filled v?ith air, are shaded ; in B, are left clear ( x 210). pletely separated from one another ; on the other hand, the two successive layers are closely joined together by their ends. The elements of the second palisade layer {pi") are somewhat less numerous than those of the first ; and two of the outer palisade cells often stand upon one of the inner. To these two palisade layers follows a loose tissue, that extends to the epidermis of the under side, and forms a net with wide meshes ; this tissue we call spongy parenchyma ; it contains fewer chlorophyll-grains than the palisade tissue. The cells of the upper layer of spongy parenchyma {sp) are fast joined to the inner palisade cells, each one usually being attached to several of the latter cells. None of the palisade cells remain with their under ends free ; where this appears to be 162 STRUCTURE Oi THE LEAF. STRUCTURE OF THE LEAF. 163 the case (asm some of the palisade cells in the figure), the junction does not lie in the plane of the figure. So also in the spongy parenchyma, the cells have no free ends ; the ends of all cells join together. The lowermost layer of spongy parenchyma (sp'") is elongated in the direction of the lower epidermis, and directed more or less perpendicularly to it ; here, therefore, we have an intermediate formation between spongy parenchyma and palisade parenchyma. The air-chamhers (a) under the stomata (st) are left free. Single cells in the spongy parench;y-ma contain com- pound crystals of oxalate of lime (/c). These cells are devoid of chlorophyll, swollen into a barrel-shape, and appear as if suspended between the green cells. At the edges of the leaflets the outer sides of the epidermal cells are strongly thickened. The palisade layer is single at the edges, and passes over on the under side of the leaf into the elongated layer of spongy parenchyma (sp'"). The fibro-vasal bundles lie in the spongy parenchyma; the largest, the mid-rib of the leaflet, extends on the one side almost to the inner palisade layer, on the other side to the under- most elongated layer of spongy parenchyma. In the fibro-vasal bundles themselves, we recognise readily the darker-looking vessels and the lighter bast. The radial arrangement of the elements arises from the bilateral [but temporary] activity of the cambium. Around the bundle is a sheath of parenchyma, the elements of which contain chlorophyll-grains, and which join on to the sur- rounding spongy parenchyma. The relations of the smaller bundles are similar, as is shown, for example, in the figure. Still smaller fibro-vasal bundles (vs), wdiich are reduced to a few vessels and bast-elements, are also met with in the cross-section. These remain to the last surrounded by a sheath of elongated parenchyma. The glands (sc) impinge upon the epidermis of the upper or under side. They are circular in outline, clothed by a layer of thin-walled, more or less disorganized cells, to which follows a layer of flat cells with granular contents, and pretty thick white walls. The surrounding chlorophyll-containing meso- phyll joins on to these cells. The epidermal cells which lie over the gland are flatter than their neighbours. The volatile oil can be readily removed by alcohol. [The mode of origin of these glands is interesting, especially for the purpose of comparison with resin canals. They can be seen well in sections of young growing leaves of either this plant, Biita graveolens, or of another nearly-related plant, the Dittany (Die- 164 STEUCTURE OF THE PETIOLE. tamnus Fraxinella). It will be readily seen that the gland is lysigenous in origin ; that is, arises from the breaking down of cells, instead of from their separation. This breaking down com- mences at the centre of the mass of gland-cells. See Fig. 62*.] Surface-sections at the base of the common leaf-stalk [petiole] show the epidermis elongated, and interrupted alike on the upper and under surface by stomata. Oil glands are not wanting. Under the epidermis lies a layer of elongated, collenchymatous cells, and then follows the chlorophyll-containing tissue. In cross- section, the epidermis is seen to be thickened on its outer side, then follows the single layer of thickened collenchymatous cells, this layer being wanting only under the stomata. The two or three layers of palisade-like, elongated, green cells are toler- ably similarly developed all round the stalk, but are looser on the under side. To these follow rounded cells, the outer green, the inner colourless, and which get larger more intern- ally. In this inner cylinder of colourless cells run the fibro- yasal bundles, the strongest in the vertical median plane, and nearer to the under side ; the others on either side of the large one, and each with its wood portion turned towards the centre of the leaf-stalk. The larger of these fibro-vasal bundles are provided on their external side with a string of sclerenchyma fibres. The activity of the cambium has also apparently lasted longer in these bundles, and it has cut off inwardly secondary wood, and outwardly se- condary, thin-walled bast. Only in the inner part of the bundle do we see larger vessels ; in the outer portion [of the wood] are only tracheides with bordered pits. As a second object for investigation, we choose the leaves of Fagus sylvatica [the Beech]. On account of the small thickness of the leaf, a thin section is here less easy to obtain. It will be well [Fig. 62*.— Oil-cavity below the upper surface of the leaf of Dictamnus Fraxinella. B, early stage, showing the breaking down of the central cells only ; c (shaded), cells not yet broken down ; C, mature state ; o, a large drop of oil. (x 320, after Sachs.)] STRUCTURE OF THE LEAF. 16^ to place straight narrow strips of the leaf between the two pieces of elder-pith [or, to pack together several of such strips, and then place them between the pieces of pith]. Only the epidermis of the under side has stomata. Adjoining the epidermis of the upper side (ep. Fig. 63), in somewhat radiating groups of cells, is a layer of elongated palisade cells (pi). These palisade cells are more or less completely separated from one another by intercellular spaces. At their lower ends they bend together into bunches, and to each bunch is joined one or several funnel-shaped, broadened cells of the spongy parenchyma (sp'). These latter are bound together with the elongated cells of the spongy parenchyma into a loose network, which extends to the epidermis of the under side (ep"). Fig 63.— Cross-section through the leaf of Fagtis iylvaticn. ep, epideraiis ; pi, palisade parenchyma ; sp, spongy parenchyma ; fc, crystallogenous cells, in k', a cluster-crystal ; st, stoma (X 360). Single cells, devoid of chlorophyll, but with a cluster-crystal (A-'), are interposed in the spongy parenchyma. The chief veins, and the lateral veins of the first order, project strongly from the under surface of the leaf in the form of ribs. The projecting part is about as thick again as the other parts of the leaf. The fibro- vasal bundle has its course in the projecting rib. This latter is covered with elongated epidermal cells, to which follow elongated collenchymatous cells. To these adjoin cells, each of which contains a simple crystal ; and then follow the multilamellar sclerenchyma-fibres, which ensheath the whole bundle. On the upper side, the palisade layer is interrupted at a narrow part over the fibro-vasal bundle, and is replaced by collenchj^raa, to which a narrow strip of elongated epidermal cells follows (cf. also at ep"). 166 CORRELATION OF STRUCTURE A layer of chlorophyll-containing cells surrounds the sclerenchyma sheath, and to these the cells of the spongy parenchyma join on. The ribs represent the mechanical system of the leaves, which must be constructed firm against flexure. [The ribs may again be, as far as the mechanical development is concerned, likened to girders.] The girders are arranged symmetrically wdth regard to the surface of the leaf, the plane of the girder being perpendicular to this surface. The upper side of the leaf is " stayed " especially against traction, the under side against compression. The girders in this case are arranged, in each rib, in the form of an I, the fibro-vasal bundle forming the " filling." The mechanical capacity of the under part of the girder, constructed against compression, is heightened by its rejnoval as deeply as possible out beyond the under surface of the leaf into the projecting rib of the leaf. By means of the veins the leaf-blade is tightly expanded, and attains thereby the necessary firmness to protect it from tearing.^ Smaller fibro-vasal bundles, as those of the figure (63), are pro- tected on the upper and under side only by some sclerenchyma fibres. The ultimate branchlets of the veins are devoid of sclerenchymatous cover, and directly surrounded in their whole circuit by the sheath of parenchyma. The smaller fibro-vasal bundles are accompanied on Avood and bast sides by the crystal- logenous cells (k). Above and under them the epidermal cells are somewhat elongated, and form shallow, depressed streaks. From the epidermal cells upon the veins arise long hairs, like sclerenchyma fibres, which, however, in the fully-developed leaves are mostly thrown off. It can, without difficulty, be determined that the leaves of the beech have grown especially thick in sunny places, and are so much the thinner in deeper shade." This increase in thickness, as microscopical investigation shows, affects the palisade parenchyma, which can become very considerably elongated and multilamellar. The palisade parenchyma is indeed a tissue specially adapted for strong light-intensity, while the spongy parenchyma is suited for slight intensity. In the palisade-cells we see the chlorophyll- grains only in profile, i.e., distributed over the elongated side-walls, and therefore, according to the intensity of the illumination, only projecting somewhat more or less into the cavity of the cell. In the spongy parenchyma, on the other hand, the chlorophyll-grains, according to the intensity of the illumination, show surface or AND FUNCTION IN THE LEAF. 167 profile arrangement, i.e., lie parallel or perpendicular to the upper surface of the leaf. The chlorophyll-grains in the palisade layer are first affected by the sun's rays ; while the spongy parench^Tna only receives the light weakened by absorption in the palisade- cells. This disadvantage is partially equalized by the surface arrangement possible in the spongy parenchyma. If, however, the intensity of the illumination is too great for the spongy parenchyma, its chlorophyll-grains assume the arrangement in pro- file. In the Beech leaves which are developed in the most intense sun-light, almost the whole green tissue is formed of palisade parenchyma, while the leaves, somewhere about a third their thickness, which have grown in deep shade, have well-nigh only spongy parenchyma. In connection with these morphological studies we will enter into a few more physiological conceptions,'^ and test their accuracy by means of the above microscopical structure. In certain coloured chromatophores, and, indeed, in the more highly organized plants, always in the green-coloured chlorophyll- bodies, the assimilation of carbonic acid takes place. Therefore these coloured plasma-bodies only are capable, in light of sufficient intensity, of decomposing carbonic acid gas and water, and out of them constructing combinations rich in carbon. This process takes place to by far the largest extent in the palisade-cells, and these can, therefore, be ph3\siologically designated, as in the highest degree, the assimilating cells. The palisade-cells are further, as we have already seen, laterally, more or less completely separated from one another, and come together internally into bundles. The assimilated materials, therefore, are not passed laterally from cell to cell, but rather make their way into the interior of the leaf. Here the bundles of palisade-cells join on to cells of the spongy parenchyma, which often, at the point of junction, are broadened into a funnel form (sp', Figs. 62 and 6S), and their function can therefore be that of receiving or collecting cells. The spongy parenchyma-cells which follow these (sp", I'igs. 62 and 63) may, from the same point of view, be designated conducting cells. The spongy parenchyma further contains air-cavities, Avhich are in communication with the air-chambers of the stomata ; it is, therefore, also a " ventilating tissue " [perhaps preferably an aerating tissue]. It is also a transpiration- tissue, since from the surface of its cells especially copious evaporation takes place into the intercellular spaces. Lastly, the collecting and conducting 168 STRUCTURE OF PETALS. tissue is, by reason of its chloropliyll-contents, also an assimilating tissue. The spongy parenchyma joins on to the parenchyma- sheath of the fibro- vasal bundle. To these they ultimately lead the assimilated materials, which are partially conducted in the parenchyma-sheath itself, partly in the bast portion of the fibro- vasal bundle ; hence these last together represent the conducting strings or conducting bundles. The tibro-vasal bundles are, how- ever, at the same time conducting strings for water, which flows in the wood, from this is given off to the surrounding tissue, and is collected in the epidermis, which, in part, functionates as a water- reservoir. It is this conducting tissue of the parenchyma- sheath around the fibro-vasal bundle which, as nerve, or vein-paren- chyma,* together with the strongly thickened "mechanical" cells, promoting firmness, forms the tissue of the projecting ribs of the leaf. This vein-parenchyma is continued into the ground-tissue of the leaf-stalk, which, as we have seen in Buta, is mainly composed of conducting (to or fro) and mechanical elements. Assimilating cells play in this only a subordinate part. We will now make ourselves acquainted with the inner structure of a petal, and also avail ourselves of this favourable opportunity to learn in it the course and termination of the fibro-vasal bundles. The petals of Verhascum nigrum [the Mullein] readily permit us to follow the branching of the bundles, and their ends, and to obtain also an insight into the structure of a petal. The air which clings to the bright yellow petal can be easily removed by tapping on the cover-glass. Alcohol cannot here be used, as it makes the structures indistinct. The petal [viewed in water] shows a delicate epidermis on the upper and under side, and from two to four layers of spongy parenchyma. Only two layers are found at the edge, from which the thickness of the leaf increases till it reaches four layers. The strongest fibro-vasal bundles, as well as the finest branches, reduced to spiral vessels only, are covered by a layer of elongated thin-walled paren- chymatous cells. This parenchyma-sheath closes together over *_It will be seen that three terms, viz., rib, vein, and nerve, are used almost indiscriminately by botanical terminologists to represent the same thing. The term rih is correct so far as it refers to the mechanical nature of the parts in question, acting just as do the ribs of an umbrella. The term vein is correct so far as it refers to the conducting (water and food) function of the fibro-vasal bundles contained in them. How far the term nerve may be looked upon as correct in its implication must be left to the future to solve ; but under any circumstances, it must be considered far inferior in appropriateness to either of the other two. [Ed.] STRUC'J'UKE OF PETALS. 1G9 the ends of the bundles. In the cells composing it protoplasmic movement can be seen. The strongly-branched cells of the spongy parenchyma join on to the elements of this parenchyma- sheath. Very instructive is the view of the ends of the fibro-vasal bundles, Avhich show a radiating junction of the cells of the spongy parenchyma with the sheath. The petals of Papaver Bhoeas [the common Poppy] can be like- wise studied without further preparation, after the air has been removed by tapping on the cover-glass. Besides the upper and under epidermis, there is here present only one layer of spongy parenchyma. The ends of the fibro-vasal bundles are never free ; they join, on the contrary, in connected arches at the edges of the leaf. In their entire course they are surrounded by a unilamellar parenchyma-sheath. To this the cells of the spongy-parenchyma join on from both sides. NOTES TO CHAPTEE XV. 1. Compare Haberlandt, in Encykl. d. Natunviss., Uandb. der Botanik., Bd. II., p. 614 ; J. von Sachs, Vorlesimgen fiber Pfianzen-Physiologie, pp. 59 et seq. 2. Compare herewith Stahl, Jen. Zeitschr.f. Natunv. Bd. XVI., 1883 ; Ueber den Elnfi. des sonnigen oder schattigen Standortes auf die Ausbildung der Laiib- bldtter. (Ou the influence of a sunny or shady position on the perfectiou of the leaf.) 3. Compare here^Yith Haberlandt, as above (1), p. GiO. 170 APICAL GROWTH. CHAPTER XVI. THE GROWING APEX* OF THE STEM. DIFFERENTIATION OF THE TISSUES. COURSE OF THE FIBRO-VASAL BUNDLES. Material Wanted. Shoots of the Mare's-tail {Hiio;puris vulgaris). Fresh, or in alcohol. Buds of a Spindle-tree (e. g., Euonymus japo7iicus). Buds of the field "Horse-tail" {Equisetum arvense). Fresh, or in alcohol. It will now be our task, by means of carefully-chosen examjDles, to become acquainted with the structure of the growing apex. As the first example, we choose a phanerogamous plant, with a very strongly-developed and easily-prepared "growing j)oint," viz., Hippuris vulgaris [the Mare's-tail] } We take strong shoots for the investigation. From these we cut off the end bud about i inch under the apex of the stem, and first remove from it all the larger leaves. We then hold the bud with the point downwards, flat between the thumb and index-finger, and endeavour to obtain a median longitudinal section of it. For this purpose the razor is passed as perj)endicularly as possible between the two fingers in question. First the bud is halved. Each half is cut up subse- quently in the same way. Then the section which appears most nearly median is chosen, and in case it does not yet appear thin enough, it is again halved, and so on until a sufficiently thin section is obtained. The operation will at first, perhaps, not be successful, yet in general it j^resents no insuperable difficulty, and can, at any rate, be attempted. If, however, the difficulty which presents itself cannot be overcome by the beginner, our object can be attained in another way. Instead of between the fingers, place the bud between two flat pieces of elder-pith, and draw the razor between * Variously known by the terms "growing point," '' imnctum vegetationis,'' and '• vegetative cone." I adopt the term " growing apex " as at once correct, and a suitable complement to the expression " apical growth," in so general THE GROWING POINT OF HIPPURIS. 171 these. The connect cutting of the object is then, it is time, left more to accident.* Objects which, like the foregoing, have a certain thickness and firmness, can be also clamped between the ends of two pieces of elder-2:>ith, and cut horizontally, together with these, as has been done in earlier cases. From the sections thus prepared we select one sufficiently median for examination ; we recognise it by the slender, regularly- formed growing apex. This forms the leaves in whorls or circles of manj^ members, and so we see them at a little distance from the apex as isolated protuberances I'ising symmetrically from the peri- phery of the vegetative- cone. Under the second youngest whorl the noies of the stem begin to be marked as horizontal denser plates [diaphragms] of tissue, above and below which, in the cortex of the stem, proceed the air-passages. These air-passages, ■which reach from one nodal diapln^agm to the other, are larger in size as the stem increases in volume. The internodes elongate very rapidly and symmetrically, and in the same proportion their thickness also increases. Under somewhere about the fourth youngest leaf-w^horl begins the formation of vessels in the stem. These are very beautifully seen after the addition of a little potash. These vessels run along the long axis of the stem. They appertain to a fibro-vasal bundle, which grows acropetally, and ends above with some annular vessels. In the tenth to the twelfth nodes are the vessels first visible which appertain to the leaves. These join the vessels of the stem-bundle. In Hippnrisy therefore, we have to do with a single fibro-vasal bundle, belong- ing to the stem, and therefore a " cauline " bundle, with which ai-e * Another method is as follows :— Cut the bud by as near as possible a me- dian cut into two halves ; place the halves in water or alcohol, as the case may be. Examine the cut surfaces, and judge by the regularity of the shape which one includes the actual growing apex, or, if it be a large apex, the most central pai'ts of it. Select this lialf ; stick a needle in a holder through it, well outside the median line, at right-angles to the length of the half, so that the cut surface of the half shall be in a plane parallel with the plane of the needle, and, when upwards, sliall have the actual apex to the right hand. With the left hand grasp the needle-holder between thumb and other fingers, but extending the index finger straight out, and fiat, so that the curved side of the half-bud lies on and across the finger, and about a third or half an inch from its end. If the needle be lightly pressed upon the finger, the flesh will yield a little, and the object will sink in and be held somewhat firml}', while at the same time the raised part of the finger beyond the object will serve as a good support for the blade of the razor. Holding the razor-blade as flat as possible, take section after section, quite cleanly, until you consider that you have fully passed the central portion of the bud. If in doubt as to which half-bud contains the actual apex, both halves can be treated in the same way. The projier central section must then be selected under a low power. Probably more than one section will be suitable. [Ed.] 172 THE GROWING POINT articulated the fibro-vasal bundles appertaining to the leaves — the " foliar " bundles. In the axils of the leaves, at a short dis- tance from the apex, small flat protuberances begin to be raised, which are the commencement of fan-like scales each borne upon a short stalk-cell. Only in specimens in course of flow^er-production do we here meet with the commencing formation of axillary buds. — In order to become more closely acquainted with the struc- ture of the growing apex, we select a good median longitudinal section, and treat it with Eau de Javelle (Potassium hyposulphite). Gas-bubbles soon begin to escape from the preparation. The action must last shorter or longer according to circumstances. The most beautiful results are obtained with alcohol material. The Uau de Javelle dissolves out the cell-contents, while the cell- walls stand out sharply. The series of cells are then easy to follow. As soon as the necessary degree of transparency is attained, the prepara- tion must be ^vashed with water. [If the section has become too transparent, it can be partially restored by treatment with a solu- tion of alum, or with alcohol.] If grains of calcium, which are separated out, should cling to the preparation, dilute acetic acid should be allowed to run in, in order to remove them. The washed preparation can be preserved in glycerine, but must be first laid in very dilute glycerine, and this concentrated slowly in air. In other cases, just as in this, Uau de Javelle can be used when it is desired to dissolve the cell-contents, and thus make the cell- walls prominent. Cuticularized cell- walls, after some time, are attacked by JSau de Javelle. If the cells are very rich in reserve food materials, the Eau de Javelle offers few advantages. If Eau de Javelle is not at our disposal, then treat the section with concen- trated potash solution, wash it out, and lay it in concentrated acetic acid. After some time Ave examine it either in acetic acid or in acetate of potash. It is an advantage not to place the section directly upon the object-slide, but to lay it upon a cover- glass placed upon this, and to cover it with a second cover- glass. We can then, if needed, turn over the section together with the cover-glasses, and so examine it on both sides ; we must, therefore, take care that no fluid gets between the under cover- glass and the object-slide. With pretty strong magnification, we settle, in the first place (compare Fig. 64), that there is a thoroughly definite arrangement of the cells in the "meristem" of the growing apex. There are cap-like layers of cells, the separating walls of which form a series OF HirruRis. 173 of confocal parabolas. The outermost layer of cells, passing also over the rudimentary leaves, is the initial layer of the epidermis — the Dermatogen (d). Under this lie four or more undifferen- tiated layers of cells (meristem layers), which appertain to the Periblem (pr), and from which the cortex of the stem proceeds. Lastly, Ave find a central cylinder, which tapers conical ly upwards, ending with usually one cell, and out of which, as can be demon- strated lower down in the section, the axial fibro-vasal bundle of the stem is formed. This tissue we designate the Plerome (pi). Epidermis, cortex, and axial fibro-vasal bundle, have therefore in Hippuris their own special "histogens" or histogenic layers. There is no single apical cell, though the individual "histogens " of the apex of the growing point may terminate in one or several initial cells. Not, however, in all the growing apices of Phanerogams is the sepa- ration of the "histogens " so sharply marked as in this case. In many Gymnosperms (Abietineee, Cycads) a clear separation between dermato- gen and periblem does not exist, and often the periblem is not clearly defined from the plerome. In the Angio- sperms the dermatogen is always sharply limited, but often there is no limit be- tween periblem and plerome. It is not in any way a ques- tion of a differentiation of the tissues which is continued into the meristem of the growing apex, but rather of the mechanical arrangement of the cell-walls, which give to the young tissues the necessary firmness. Very clearly marked in this arrangement is the rectangular junction of the anticlinal walls, i.e., those running perpendicularly to the surface of tlic apex, and of the periclinal, i.e., those parallel to that surface. - For all that, we can retain the terms dermatogen, periblem, and ple- rome, because the arrangement of the layers of cells, as we have observed it in Hippuris, frequently recurs in the growing apex Fig. 64.— Longitudinal section through the growing apex of Rlit-puris vxdgaris. d, dermato- gen ; pr, periblem ; pi, plerome ; /, commence- ment of the leaves ( x 2-iO). 174 THE GROWING POINT of Phanerogams. These terms can be therefore conveniently used for denoting definite regions of the growing apex. From the dermatoofen, in fact, amongst Angiosperms, if we exclude the very few exceptions, proceeds only the epidermis. The fibro- vasal system is, however, not always traceable in its origin to the ple- rome, but often to the periblem also. In the earliest rudiments of leaves we see first in the outermost layer of the periblem pericli- nal divisions set up (at/), then follow anticlinal. The dermato- gen in the places of protuberance remains unilamellar ; it divides only anticlinally. In the same way in the development of buds periclinal and anticlinal divisions take place in the outer periblem layer, and anticlinal only in the dermatogen. We will now investigate a flat growing apex, as it occurs in most Phanerogams. As an example, we may take Buonymus japonicus^ [the Japanese Spindle-tree], cultivated as an orna- mental shrub in many gardens, which can be examined at any time of the year, and the buds of which cut very well. We first prepare cross-sections, in order to obtain a surface view of the growing apex. Treat the sections in this case as we have done with Hijypuris. With weak magnification we recognise the growing apex as a flat hump, surrounded by the youngest leaf protuberances. These stand in two-membered, alternating whorls, and therefore decussate [" opposite decussate "], as we are wont to say. Every new pair of leaves starts, after con- siderable enlargement of the gi^owing apex, in the gaps present between the preceding pair of leaves (Fig. 65 A). With suitable magnification, it is here exceedingly easy to follow the arrange- ment of cells at the apex. Fig. 65 B presents such a figure ; an apical cell is therefore not present. Cross-sections taken close under the apex show us a rapidly initiated differentiation of the tissue into primary pith, into procamhium, which will form the fibro-vasal bundles, and into primary cortex. The zone of pro- cambium shows in the cross-section a rhombic figure, with some- what projecting and rounded angles. The procambium consists of thin-walled, narrow, radially-arranged cells. At the angles begins the formation of the elements of the fibro-vasal bundles : protophloem elements at the outer, spiral vessels at the inner, side of the zone. This region of commencing differentiation of the elements of the fibro-vasal bundles is not defined towards the rest of the procambium tissue. The procambium zone opens at the places where the foliar bundles enter, in order to admit OF EUONYMUS. 175 them. In the axil of each of the young leaves we can see the rudiments of an axillary bud. The form of Fig. 65 C is shown by median longitudinal sections, with weak magnification. The flat growing apex, the rudimentary leaves, increasing in size, the axillary buds (g), the differentiation of the primary pith (m), the procambium zone (pc), the fibro-vasal bundles, com- Xi"! ^- ''''■" "■••" """ m Fio. 65.— Apex of the stem of Euonymus japonicus. A, apical view of the same (x 12^ B apical vie«- of the growing point (x 2iO). C, median longitudinal section through / %^j;?''/ "'"^ ^^^"^ ^"^ ^^^' ^' ™'''^^^'' longitudinal section through the growing apex (X 240) ; d, dermatogen ; pr, periblem ; pi, pleromc; /, leaf protuberance ; g, bud protuber- ances ; p/, leaf -traces ; pc, procambium ring ; hi, pith ; c, cortex. mon to both leaves and stem [the so-called leaf- traces] (pf), and the primary oortex (c), are to be seen at a glance. Pith and cortex contain great quantities of clustcr-crystals of oxalate 176 THE GROWING POINT of lime. In fresli sections examined in water, tlie pitli and cortex appear greenish, wliile the procambium-zone appears clear. In order to follow the arrangement of the cells at the growing apex, we again use potash and acetic acid. Outermost on the growing apex we find the unilamellar dermatogen (Fig. 65, D, d) ; under that three casing layers, w^hich we have to designate as Periblem (p-) ; and then the central solid cylinder of tissue, which is not everyw^here sharply defined towards the periblem, the plerome (pZ). The growing apex appears very narrow between the two last progressing leaf-protuberances ; it is as a rule thus to be seen. On the other hand you have often to cut for a long time before the section passes through the first leaf- protuberances. If you are successful in this, the form presented is that of Fig. 65 D. The growing apex appears then much broader, the histogens [or histogenic layers] can be better followed in it. The formation of the leaves is initiated by periclinal divisions in the two outermost layers of periblem (at /) ; the dermatogen remains unilamellar. Just the same divisions as for the commencement of the leaves take place in the axils of the third youngest pair of leaves, for the formation of axillary buds ; the process is likewise initiated by periclinal divisions in the hypodermal layers of cells. It can be determined with certainty that the dermatogen produces only the epidermis, the periblem the cortex, and the plerome the pith of the stem. Less certain is the proof that the procambium ring also proceeds from the plerome. That the formation of fibro-vasal bundles is not ex- clusively confined to the plerome, follow^s of necessity from the fact that the part of the fibro-vasal bundle passing into the leaf arises inside the cortex, and therefore out of periblem, and that the entire inner tissue of the leaf, with all its fibro-vasal bundles, is a product of the periblem. We wdll investigate lastly a Yascular Cryptogam, growing by means of an apical cell [as distinguished from the preceding cases in which growth takes place by an apical meristem], and choose as the most favourable object Equisetiim arvense [the common field " Horse-tail " ] *. It is here comparatively easy to bring into view the apical cell. Shoots in course of development are studied either fresh or by alcohol material. We remove from the apex of the stem a piece about ^ inch long, or rather more, and cut it, as in the above cases, between the fingers, with the apex down- wards [or by the needle method described above]. OF EQUISETUM. 177 Amongst the longitudinal sections produced we look for one which shows the conical growing apex intact. In order to obtain an insight into the arrangement of the cells of this apex, it is usually necessary to make them more transparent. This may be effected, in this case also, with Eau do Javelle, or else by the addition of a little potash. Should this latter act too strongly, and have " cleared " the growing apex, until its cell- walls become unrecognisable, we can remedy the evil by a suitable addition of water. In fresh sections we must avoid the use of any w^ater- withdrawing medium, as otherwise the growing apex Fig. 66.— Longitudinal section through the growing apex of a main vegetative shoot of Equisctum arvcnse. t, apical cell ; s', youngest, s", next older segment ; p, primary wall ; m, segmenting wall ; pr, later periclinal ; a, anticlinal walls; /.first; /', second, /", third whorls of leaves; g, initial cell of an axillary bud (x 240). will collapse. Sections of alcohol-material can, on the other hand, be laid in glycerine direct, but not after previous sojourn in water. Sections treated with Eau de Javelle cannot be placed at once into concentrated glycerine, but must be placed in very dilute glycerine, which is allowed to concentrate by standing in the air. Sections made transparent with potash can be neu- tralized with acetic acid, and preserved in acetate of potash. As it is here of special importance to be able alternately to view the 178 THE GROWING POINT section from its two sides, we lay it, as we have already done in the case of tlie growing apex of Eippuris, between two cover- glasses. If the growing apex is cut in the proper direction, it presents its apical cell, a three-sided pyramid upon a convex base, in the form of a wedge, the apex of which is sunk in the tissue of the growing point, and its base is arched free towards the exterior. This apical-cell divides by means of partition walls, which are parallel to the existing side-walls, follow one another in spiral sequence, and form segments arranged in three straight rows. These segments (*S) are shown in profile in our figure 66. They further divide up in definite fashion, and so gradually construct the body of the plant. [The adjoining Fig. 66*, taken from Nageli and Schwendener's Das MihrosJcoj}, will help to fui-ther elucidate this process. A shows a median longitudinal section of the apex of an Equisetum stem, which corresponds pretty closely [Fio. 66*.— Schemes showing the division of the apical cell in the stem of JEguiscfiim. A, in sectional, B, in apical surface view.] with that in the text, the figures 1, 2, etc., showing the order of priority of the original dividing walls, cutting off segments from the apical cell. As this cell is three-sided (besides its base), every third wall is not shown in the longitudinal section ; in this case walls 3 and 6 are omitted from the figure. B gives a dia- grammatic view from above of the three-sided apical cell, showing the order of formation of the primary dividing walls, 1, 2, etc. In both figures the lighter lines are intended to show the subse- quent divisions of the primary segments.] At some distance from the apical cell, a "bank" is raised upon the growing apex, which grows at its edge with w^edge-shaped initial cells. Certain parts of this edge, later on, get in advance in their development, and form the free leaf-apices [" teeth "] of the, lower down, connate whorl of leaves [" leaf -sheath "]. The further removed OF EQUISETUM. 179 from the apical cell, so much the greater are the 3'oung leaf- whorls ; simultaneously progresses the differentiation of the inner tissue of the stem, especially the separation into denser, small-celled, thin nodes, and less dense, elongated-celled, long internodes (Fig. 67). First the larger-celled pith begins to separate out in the middle of the stem. In the fifth internode, counted from above (in the figure), the first annular vessels become visible in the procambium strings, at the outer limits of the pith, and can be traced from here into the next higher com- mencement of the leaf -whorl. Each in- dividual fibro - vasal bundle is here com- mon to stem and leaf, and is therefore de- signated as leaf-trace. In each internode just so many fibro-vasal bundles run outwards, as leaves are repre- sented in the leaf- whorl. The leaf- traces, lying at first separated from one another, are, in about the node underlying the seventh inter- node from the apex, connected by side branches, whereby a complete fibro-vasal system is formed. Approximately in the tenth internode the pith begins to become hollow, through the breaking apart of its cells. In the node, on the other hand, the pith-cells Fig. 67.— Median longitudinal section through a main vegetative shoot of Enuisetum arvense. pv, growing apes of the main axis; g, initial for a bud; g', g", g'", g"", stages in the development of such a bud ; r, r', the origin of a root on the bud; m, differentiation of the primary pith ; vs, spiral vessels making their appearance ; n, differentiation of the nodal diaphragms (x 2G). 180 COURSE OF THE VASCULAR BUNDLES undergo a corresponding augmentation, and remain in union. — The lateral buds are initiated bj single cells in the axils of the leaf- whorl [leaf- sheath]. They stand in whorls, and, as examination of the mature condition shows, alternate [in position] with the free leaf-teeth of its leaf-sheath, the tissue of which they finally break through at the base, in order to come outside. The longi- tudinal section, therefore, shows the somewhat larger buds, grown into the tissue of the leaf-whorl, lying close over the surface of the stem. At about the seventh node, the buds are so far developed that they already possess several embryonic leaf -whorls. Their growing apices can, with care, be used for the study of the apical-cell. Amongst the Vascular Cryptogams, the. Equisetacese and Ophio- glosseae possess only collateral fibro-vasal bundles, as we can prove readily enough by means of cross-sections through an older internode of Equisetum arvense. The fibro-vasal bundles are arranged in a single ring around the hollow pith. In the w^ood portion, placed internally, of each fibro-vasal bundle, will be noticed an intercellular passage, the carinal canal ; the thin-walled bast, placed externally, is environed on its sides by the annular and reticulated vessels of the wood. An endodermis surrounds the entire ring of fibro-vasal bundles. In the thick cortex, alternating with the fibro-vasal bundles, will be specially noticed broad inter- cellular passages, the Vallecular canals. If w^e count the free leaf- teeth in the next higher leaf- whorl, we find that the number of the fibro-vasal bundles corresponds with this number. — In order to obtain information upon the course of the fibro-vasal bundles, "we now prepare successive cross-sections, so long as to have passed out of one internode through the node into the next internode. We can use for this purpose either fresh or alcohol material ; only it is advisable that we use the youngest possible part of the stem, as older parts are strongly silicified, and will quickly blunt the razor. In order to prepare the sections of equal thickness, we can make use of the microtome referred to on page 63. The sections are arranged in proper order upon an object-slide, and can be made more transparent by means of potash. A close comparison of these successive sections enables us to prepare a scheme of the whole course of the fibro-vasal bundles, as in the adjoining Fig. 68, in which w^e have cut the stem open along one side, unrolled it, and projected the course of the fibro-vasal bundles on the surface of the cylinder thus laid open. AVe find IN EQUISETUM. 181 n S S/ that each of the fibro-vasal bundles (a, 6, or c), coming down from a higher internode, divides in the node into two forks ; and that one of the forks from each of two adjoining bundles combines with a new fibro-vasal bundle, which here enters out of the leaf- whorl (thus the forks from a wdth h and 6, a fork from each of h and h with c). If the fibro-vasal bundles of the lateral buds are already formed, the figure is somewhat complicated by them. Each lateral bud joins (jj) the vascular system of the parent axis, with two fibro-vasal bundles {cj), and ahvays with one fibro-vasal bundle on to each of the two forks of the next higher stem bundle, immediately after this divides into its two forked branches. The lateral buds alternate with the fibro-vasal bundles of the leaf- whorl which conceals them, and correspond in their position with the fibro-vasal bundles of the next higher and next lower leaf- whorl. — It follows from our observations that the entire system of fibro-vasal bundles of our stem of Equisetum is common ; it is formed of leaf-traces, which fork at their base inside the node, in order to join there, through the medium of their forks, with newly entering fibro- vasal bundles. That the leaf-traces combined with one another form the entire fibro-vasal system, is every- where in vascular plants the most common case ; we will therefore limit our studies upon the course of the fibro-vasal bundle to this one simplest possible example. — In the investiga- tion of a more complicated case, it is necessary to arrange the successive sections in the same [sequence and] direction on the object-slide, in order to be able to compare them more readily. This last task is facilitated if one side of the stem is marked by a shallow longitudinal cut. It is often necessary to draw the successive sections, in order to be able to prove the displacement of the Fig. G8. — Scheme showing the longitudinal course of the fibro- vasal bundles in the stem of Eiiui- sctum, supposed to be opened out into a plane, g, the junction of the bundles of the bud. 182 KOTES. individual bundles. Longitudinal tangential sections, made trans- parent with potash, will in many cases lay bare at once the entire course of the fibro- vasal bundle. NOTES TO CHAPTER XYI. 1 Sanio, Bot. Zeitnnri, 1864, p. 223. Note **, 1865, p. 184. De Bary, Com- parative Anatomy (Engl, trans.), p. 8 ; L. Kny, Wandtafcln, III. Abth., p. 99. 1* Noll, Botan. Centralblatt, Bd. XXI., 1885, p. 377. - Sachs, Arheiten des bot. Inst, in Wiirzhurg, Bd. II., pp. 46 and 185. 3 Hanstein, Die Scheitelzellgruppe im Vegetationspunkt der Phanerogamen, p. 9. Warming, Eecherches s. I. ramif. d. Phanerogames. •* Compare Cramer, Pfianzenpliysiol. Unters. v. Naegeli, Heft 3, p. 21. Eeess, Jahrb. /. iciss. Bot., Bd. VI., p. 209. Sachs, Text-Book of Botany (English translation of Ed. IV. by Vines), pp. 398-402; and Goebel, Grundzuge, p. 291. De Bary, Comparative Anatomy (Engl, translation), pp. 18, 19. THE EOOT-APEX. 183 CHAPTER XVII. GrtOWING APEX (TIP) OF THE EOOT. Material Wanted. Eoots of barley {Tlordeum vulgare), grown in a flower-pot. Fresli, or in alcohol. Eoots of Thuja {e.g., T. occidentalls), grown in a flower-pot. Fresh, or in alcohol. Boot of fern {e.g., Pteris cretica), grown in a flower-pot. Fresh, or in alcohol. It is des^L^able now to become acquainted with the growing apex (tip) of roots. We commence with Angiosperms. The structure of their root-apex^ can be studied with comparative ease in the Graminea3 [grasses]. They provide us, it is true, with only one of the possible types of root-growth amongst Angiosperms, but still one widely spread and instructive, and therefore very suited to give us an insight into the processes in question. In order to obtain favourable material, w^e choose plants removed with care from flower-pots. If we turn the flower-pot upside down [so that the whole contents come out bodily, a result often assisted by tapping the flower-pot lightly on its rim], the root-apices will be usually found free in the exterior of the mass of earth. For careful study we choose the common barley, Hordeum vulgare. In order to get general information, we first prepare a cross-section through an older part of the root. In the middle of the axial fibro-vasal cylinder we find a large duct or vessel, then in the periphery of it about eight vascular rays alternating with the same number of portions of bast. As, however, customary in grasses, the vascular rays extend to the endodermis, and therefore interrupt the pericambium. The endodermis shows more or less clearly the dark radial shadings ; to it follows the pretty thick cortex. The longitudinal section of the root-apex w^e prepare between the thumb and forefinger. It must be sufliciently 184 THE ROOT -APEX median; then the structure is plain, even without the use of reagents, but here also Eau de Javelle can be used with advantage (cf . p. 172). Above all, it is observable that the body of the root is sharply defined from the root-cap. We can, in fact, follow a line, which is prolonged from the surface of the epidermis, continuously Fig. 69.-Median longitudinal section through the root-apox of Hordeum vulgare li Calyptrogen; c, thickened outer wall of the epidermis ; d, dermatogen; pr, penblem ; i^l, pierome ; en, endodermis; i, intercellular passage filled with air ; a, row of cells which will form the'central duct ; r, disorganized cells of the root-cap ( x 180). over the apex, between the body of the root and the. root-cap (cf. Eig. 69). Still the dermatogen does not pass, as such, over the apex, but it must rather be said that the dermatogen (d) and the OF GRASSES. 185 periblem (pr) of the apex come together in common initial-cells. In Fig. 68 only one such common initial layer is present, but there may be several. The dermatogen, as such, can be traced up to this initial layer ; the periblem also, but one cell thick, merges with it. The plerome, under this common dermatogen-periblem cap, is crowned by a few initial cells. Externally bounding the line which separates the root-body from the root-cap are the initial cells for the root-cap, forming a layer of flattened cells which may be designated the calyptrogen (k). The cells from the calyptrogen are given off outwardly, and, as the result of their origin, arranged in straight rows ; at first flat, they soon increase in height. At the apex of the root-cap they become rounded ; finally separate from one another and become disorganized (r). It is a peculiarity of the Graminese that their dermatogen is strongly thickened on the outer side (c). This thickened outer wall is bright white, swells strongly, and appears so much the thicker the longer the section lies in water. At the lateral boundaries of the cells we see highly refractive strise pass more or less deeply into the thickened outer wall. These are the primary walls of the cells ; and the older they are, the more deeply they always penetrate into the thickened Avail. These walls clearly show lamination. The periblem has rapidly increased the number of its layers by periclinal divisions. Between its inner cell-layers intercellular passages filled with air quickly put in their appearance, as is represented in our figure by dark lines (e.g., at i). The periblem forms the cortex, the innermost layer of which becomes the endodermis. — The plerome ends conically in a group of initial-cells; two such initial cells can be seen in the longitudinal section which we have represented. The plerome forms the axial fibro-vasal cylinder. The diffei-entiation of the large central duct or vessel in this last can be traced from the initial group. The cells from which this duct will be developed are distinguished by their greater breadth (a). The elements set apart for the smaller vessels are first distinguishable much later. The roots of Gymnosperms^ show, in many respects, a peculiar organization in the meristem of their growing apex. We will study more closely Thuja Occident alls. The cross-section through a fully-developed root resembles the already-known cross-section of the root of Taxus baccata [the Yew], excepting that the roots of Thuja are usually tetrarch [i.e., have four primary ligneous rays, or bundles]. The median longitudinal section through the apex 186 THE ROOT-APEX of tlie root shows a sliarplj^-liraited plerome cylinder, which, terminates in a few initial cells, and is surrounded by a covering of periblem, some twelve to fourteen cell-layers thick. This last passes over the apex, and forms there its terminal initial layers of eight to ten inner rows of cells, while the outer rows pass over into irregularly arranged, comparatively large cells. These large cells extend to the apex of the root-cap, where they ultimately lose their union, and become disorganized. The root-cap of TJucja, and of Gymnosperms generally, consists of the outer parts of the periblem ; dermatogen and calyptrogen are wanting. The initial layers of the plerome, passing over the apex of the plerome, divide by periclinal and anticlinal walls. The periclinal divisions increase the number of layers of the periblem, and replace from the interior the elements which are exfoliated from the periphery. The anti- clinal walls increase the number of cells in the individual layers, and provide chiefly for the formation of the cortex. As the anti- clinal Avails in successive layers correspond pretty regularly with one another, they form anticlinal rows of cells, which, straight in the interior, separate from one another externally, like the com- ponent rays of spray which collectively constitute the jet issuing from a fountain ; and forming therefore a constantly extending series of co-axial parabolae. The periclinal divisions in the initial layers of the apex have as result that the cell-rows of the cortex, when these are followed towards the point, appear constantly doubled. The most median, straight, anticlinal rows of cells in the periblem of the root-apex are distinguishable from their neighbours. They form a "periblemic column " (Periblemsaule), which is lost in the outermost, brown elements of the root-cap. This column appears clearer, its cells immediately adjoin one another, while those bordering laterally form air- containing intercellular spaces. Moreover, the cells of the column are distinguished by especial richness in starch. As results from the foregoing relations, the root of Tlmja possesses no epidermis, the outer surface of the root is composed of the, for the time being, outermost layer of the periblem. If such a layer is followed in the direction of the apex, we shall soon see it pass under another, which for a time pro- vides a surface. The outermost living layers of cells are protected at their surface by the collapsed walls (become brown) of the disorganized layers of cells. The roots of the Gymnosperms have, in general, no root-haii'S ; we search for such in vain in TJnija occideyitalis. The adjoining figure, 70, gives, with low magnification, OF CONIFERS. 187 the structure of a longitudinal section, and can facilitate our obtaining information about it. Naturally, the arrangement of the cells can only be indicated with such a low magnification. Pass- ing from the exterior towards the interior, we see, therefore, the brown, collapsed covering of cells (x) ; then the periblem (_pr), which can be traced over the apex of the root, and whose outer- most layers, therefore, form the root-cap ; lastly, the plerome (pi), the termination of which is not quite clear with low magnification. "We are inclined to imagine that the upper part of the plerome is larger than it really is, because the innermost layers of the periblem, bordering on the plerome, are devoid of intercellular spaces, and there- fore (as is shown in the figure) appear just as clear as the plerome cylinder. In the oldest parts of the section the plerome cylinder shows itself surrounded by a red layer of cells, which, as a comparison with the cross-section shows, indicates the en- dodermis filled with red cell-sap. As we approach the apex these endodermic cells become unrecognisable. Vessels (s) also appear in the older parts of the plerome cylinder. The more clearly showing column (c) penetrates the apex of the periblem. Upon this impinge laterally the air-containing layers of the periblem. These last, however, reach entirely neither to the plerome nor to the surface of the root. The last is composed of large brown cells. The roots of Coniferas will serve to make us acquainted with the methods of branching of roots in general. In the ex- amination of the roots of Thuja occiden- talis, it will strike us that they bear their lateral roots in four or in three straight rows. We readily prove by cross-sections that three rows of lateral roots indicate triarch [i.e., with three lig- neous rays], four rows tetrarch, fibre- vasal cylinders. We prepare now a cross-section through a root at the place of insertion of a lateral root, and determine that the lateral root projects from one Fig. 70.— Longitudinal sec- tion through the root-apex of Thuja occidenfalis ; x, outer brown hijer of disorganized cells ; jir, pericambium ; pi, plerome ; e, endodemiis ; s, spiral vessels; c, periblemic column ; k, root-cap ( x 26 ). 188 THE KOOT-APEX of tlie ligneous rays. As now the ligneous rays run in straight lines in the axial, fibro- vasal cylinder, the arrangement of the lateral roots in straight rows is hereby explained. We will now endeavour to become acquainted also with the growing ajDex of a root which grows by means of an apical cell.^ In such roots the same variability as with stems growing by means of apical cells does not exist. The apical cell is always a Pro. 71.— Medium longitudinal section through the root of Pteris cretica ; t, apical-cell ; 7c, initial cell of root-cap ; Icn, outermost layer of roQt-cap ; {e, wall cutting off epidermis ; r, ditto cortex; c, ditto cambium ; 2', ditto pericambium) (x 2iO). trilateral pyramid, and the co-ordination of the segments formed from it remains constant. We investigate the root of Pteris cretica (Fig. 71), but can equally well choose any other species of Fern. By turning a flower-pot upside down, we easily obtain uninjured root-apices. The roots of Fteris cretica, as of ferns generally, are diarch ; with the woody portions alter- OF FERNS. 189 nate flattened bast portions ; the pericambium is unilamellar, the endodermis flattened, the cortex become brown, and in its inner part strongly thickened. We now endeavour to obtain, between thumb and forefinger, a thin, median, longitudinal section of the root-apex. It is not very difliciilt to bring to view the apical cell ; it does not here, however, occupy the apex of the root, but is covered by the tissue of the root-cap. This apical-cell {t, Fig. 71) has, as in the stem of Equisetum, the form of a three- sided pyramid, whose convex base is turned towards the cap, while the apex formed by the junction of the three side- walls is sunk in the body of the root. The divisions, as in the stem of Equisetum, take place parallel to the side- walls ; besides these however, from time to time (usually after three or four of the above consecutive divisions), a wall parallel to the convex base is formed (compare the figure). The apical-cell retains its form throughout its divisions ; the cell cut off from the base has, however, the form of a segment of a sphere. This cell (k) is an initial cell of the root-cap, giving to this latter its origin. It divides first by a wall perpendicular to its base into two halves ; each half repeats this division, so that four cells, quadrilateral in outline, are formed. In these the division is reiDcated, and always by walls at right-angles to the original base, so that an older layer of the cap (/b") consists of a large number of cells. The cells of the older cap-layers are full of starch-grains. They are gradually disorganized while the apical cell cuts off continually new initial cells. The outer walls of the, for the time being, outermost cells of the cap, are strongly thickened. The dividing walls, formed parallel to the side-walls of the apical cell, follow, as in Equisetum, the direction of a spiral. NOTES ON CHAPTEE XVII. ^ Sachs' Text-hook of Botany (Engl, trans, of 4th edit.), p 147 ; Janczewski, Annales des sc. nat., Botanique. V. Ser., torn. XX., 1873, pp. 162 ctseq. ; Treub, Musee hot. de Leide, torn. II., 1876 ; De Bary, Comparative Anatomy (Engl, trans.), pp. 9 et seq. 2 Strasburger, Coniferen und Gnetaccen, p. 340 ; De Bary, Comp. Anat. (Engl, trans.), pp. 13 et seq., where see the further literature. ^ NiigeU and Leitgeb in Bcitr. zur ichs. Bot., 4 Heft, 1868, pp. 74 et seq. 190 VEGETATIVE STRUCTURE OF MUSCIJS'E^. CHAPTER XVIII. VEGETATIVE STEUCTUKE OF THE MOSSES AKD LIVERWORTS. Material Wanted. Plants of a strong moss, such as 3Imum tmdulatum, M. hornum, or Polytriclinm. Fresh, or in alcohol. Plants of a Bog-moss {e.g., Sphagnum acutifolium) . Fresh. Thallus of a Liverwort {e.g., Marcliantia 'polymorflia). Fresh. Thallus oi Metzgeria furcata. Fresh. Hitherto we have studied only the structure of the stem and leaves in Vascular Plants ; we turn now to the stem and leaves of Mosses, from which vessels are absent. ^ We commence with a comparatively complicated case, where the differentiation of tissue is already somewhat advanced ; with Mnium undulatum. We first of all prepare delicate cross-sections through the stem. In the midst of the stem appears an axial cylinder, composed of narrow thin- walled cells. We can conceive this cylinder as the simplest of all conducting bundles. Its cells contain no living contents, only water ; they are distinguished from their surroundings by their yellow-brown coloration. To this conducting bundle, which there- fore consists of only water-bearing tissue, adjoin the wider cells of the cortex, with greenish-yellow walls, and living chlorophyll- holding contents. At first they increase somewhat in diameter in passing from the interior outwards ; at the perijDhery they be- come rapidly narrower and thicker walled, and pass over at length, without special limits, into a uni- or bi-lamellar, narrow, strongly- thickened epidermis. At two or three places the outer cell-layer of the stem is prolonged externally into a unilamellar plate of cells, which represents the leaf -wings running outwards from the stem. Cross-sections which are taken from the lower, leafless, strongly-browned part of the stem, show the walls of the peri- pheral layers of cells coloured dark-brown. From single cells of the surface have grown long, broAvn- walled, repeatedly branched threads of cells, which here take the functions of roots, and are KOOT-HAIRS AND PROTONEMA OF MOSSES. 191 distinguished as root-hairs or rhizoids. These rhizoids, as can be readily seen, are distinguished by obliquely-placed partition walls, which therefore form an exception to the rule, so generally obtain- ing, of rectangular division. Under numbers of such partition walls, and always under its elevated side, arise the subsequently still inoro branching lateral branches. Only the apices of the rhizoids which are still growing are provided with colourless walls. The closest similarity with such root-hairs, from the point of view of the branching and the oblique position of the dividing walls, is shown by the proembryo of the typical leaf-bearing mosses, the so-called protonema, which is developed from the [Fig. 71.*— A moss, Funaria Tiy grometriea. J, germinating spores ; w, root-hair, or rhizoid; s, exospore (xo50). B, part of a developed protonema, about three weeks after geinniiia- ting ; 7i, a procumbent primary shoot, with brown wall and oblique septa ; from this arise the ascending branches, having limited growth. K, rudiment of a leaf-bearing axis with u', rhizoid or root-hair (x about 90). (After Prantl.j] germinating spore. Its branches, however, so far as they do not penetrate into the soil, are colourless, and contain numerous chloro- phyll-grains. The buds, which develop into the moss-stems, are side-branches of this protonema [compare Fig. 71*]. The near relationship of rhizoids and protonema is shown also in the circumstance, that the rhizoids, if kept damp and exposed to light, can develop a protonema which can give rise to numerous new plantlets. It needs only to lay a turf of Mnium with the underside upwards, and keep it damp, in order to produce 192 THE LEAF OF MOSSES. numerous green protonema-threads from the rhizoids. This lat- ter in its macroscopic aspect reminds us of terrestrial tufts of Vaucheria. If the cross-section has passed through an injured part of the stem of Milium, the place is seen not to be covered with cork, since this cannot be formed by the Crj^ptogamia, with the ex- ception of Botrychium [the Moonwortfern] ; but, on the other hand, the walls of the limiting cells are thickened and browned, and so, apart from their broader cavities, resemble the other surface-cells. Near the surface can be seen, in the cross-section, isolated small strings of thin-walled cells, which moreover in their coloration resemble the elements of the central cylinder, and which, like them, are without living contents, but on the other hand contain water. These are the conducting bundles belonging to the leaves, which have " blind " ends in the cortex of the stem, while they occasionally, in Folytrichum* join on to the axial conducting bundle of the stem. A leaf, which we examine Avithout further preparation in a drop of water on the object- slide, exhibits a unilamellar lamina, and a multilamellar midrib. This last ends in a terminal tooth, which consists of a number of rhombic cells. The cells of the midrib are elongated, the peripheral cells contain chlorophyll-grains. The lamina of the leaf is unilamellar ; it consists of polygonal chlorophyll- containing cells. The broad, seam-like edge of the leaf is formed of elongated, strongly thick- ened cells. The outermost bear on their edge, at nearly equal distances, one- to two-celled, sharply tapering teeth. Cross- sections through the leaves are obtained at the same time with the cross-sections of the stem. If it is desired to cut cross- sections of separated leaves, which, from their small thickness, is no slight task, it can be considerably facilitated if a considerable number of leaves are stuck together with glycerine-gum, and, without waiting for the gum to dry, the object, thus made thicker, is cut between elder-pith. f The cross-sections are then laid in water, which at once dissolves out the gum. This method can be used at all times, when it is desired to obtain cross-sections of very thin surfaces. Upon these cross-sections of our moss-leaf, we can determine that the lamina is unilamellar, the cells at the * In Pohjtrichum, the inner part of this conducting bundle is commonly collencbymatous. [Ed.] t Good cross-sections of leaves can be obtained without preparation by cutting through the crowded leaves at the apex (" bud ") of the stem of an actively growing moss, such as Mnium undidatum, M. Iwrnum, or a Folytrichum. [Ed.] STEM OF SPHAGNUM. 193 margin of the leaf are strongly thickened. The midrib projects more strongly from the under than the upper surface of the leaf. In its centre, somewhat nearer the under side, lies a strino- of thin- walled cells, in which we again recognise the conducting bundle which we previously saw^ in the cortex. These thin-w^alled strings are protected on their under side by some strongly-thickened narrow cells. The structure reminds us not a little of certain greatly-reduced monocotyledonous fibro- vasal bundles, limited to a few bast elements and a weak layer of sclerenchyma. A withered plant, with the lower cut surface of its stem placed in water, remains withered, but becomes, on the other hand, rapidly turgid if it is immersed with its leaves in water. The admission of water through the leaves is here, therefore, very active. The structure of the stem of the Bog-mosses oifers special pecu- liarities, and shall therefore be brought here within the range of our observation. We prepare cross-sections of the stem of Sphag- num acutifolium. These cross-sections show us a broad central cylinder, which in its interior is constructed of broad, somewhat collenchymatously thickened cells ; towards the periphery its cells become gradually narrower, and, in the outermost layers, are coloured yellow-brown. A special conducting bundle is not present in the interior of this cylinder. It is surrounded by a large-celled outer cortex of three layers of cells. The elements of this impinge immediately upon the narrow, yellow-brown cells of the inner cylinder. They are distinguished by large round or oval holes [pores] and delicate spiral bands. These pores are easy to see, and that they directly join together the cavities of these cells can be readily proved in places where the sections have cut through such pores. Not infrequently, moreover, fungal threads [hyphae] are seen in these cells, which pass without hindrance from one cell to the other through the pores. These porous elements of the outer cortex of Sphagnum contain, moreover, only water or air, and are wdthout living cell-contents. To the plant they serve as a capillary apparatus, by which the water may be carried to a place of need. The plants are devoid of cuticularized parts ; con- centrated sulphuric acid immediately dissolves the entire tissue ; comparatively the most resistant are the middle-lamellae and their junction " seams " of the yellow-brown outer cells of the central cylinder. The leaf expansion is ovate, entire, unilamellar, and consists, as either surface- view shows, of two kinds of elements. The one are 194 LEAF OF SPHAGNUM. small, cliloropliyll-coTitaining (and therefore also containing proto- plasm and nucleus), living cells ; and others are dead, filled with air or water, provided with rings or with spiral bands and intermediate open pores. The fact, which must have repeatedly struck us, that dead air- or water-containing cells, so far as they are not strongly thickened, so often need spiral band, rings, or network as thick- ening of their walls, derives explanation from the circumstance that the said cells are deprived of their turgidity, and need this me- chanical apparatus in order not to collapse nor be crushed. The green cells of the leaf-expansion are all joined together, and form a network with elegantly winding walls, each mesh being occupied by one of the empty cells. The green cells serve for the assimila- tion of carbonic acid gas ; the empty cells serve, just as do the corresponding cells of the outer cortex of the stem, as a capillary apparatus for the supply of water. Careful observation shows that the number of pores diminishes towards the edge of the leaf, that they are more prevalent on the under side of the leaf, and stand laterally on projecting cell-walls. The edge itself of the leaf is composed of the narrow green cells, and adjoining these of a sino-le-rowed " seam " of narrower collapsed elements containing watery contents, and slightly thickened on the outer surface. Only the end surfaces of these elements appear to be thickened more strongly, and project outwards proportionally. A midrib is wantino' in the leaves, just as is a conducting bundle in the stem ; the plants are in this respect, therefore, much more simply con- structed than Mnium ; more complex on the other hand as to the formation of a special capillary apparatus. The thallus of MarcJiantia polijmorpha [the common Liver-wort], so widely spread upon damp ground [and especially in damp green-houses on the surfaces of the flower-pots], and so readily recognisable by its bulbil or gemmse-cups, and ultimately also by its [more rarely produced] disk-like or umbrella-like receptacles, shows a tolerably complicated structure. The absence of cormo- phytic development does not therefore necessarily entail simple anatomical structure. The thallus is tough, like leather ; it branches by forking (bifurcation) of its apex, which lies at the base of the apical sinus (or depression). If the shoot has forked shortly before, the centre of the previous depression is occupied by a lobe of the thallus, on both sides of which the apical de- pressions lie. Along the centre of each shoot projects, on its ventral [under] side, an indistinctly bounded midi^ib. From this THALLUS OF MARCIIANTIA. 195 proceed outwards and forwards obliquely directed strice arching towards the margin of the thallus. At some distance from the apex the thallus is fixed to the substratum by delicate rhizoids springing from its centre. If we bring the thallus, Avith its ventral side turned upwards, under a simple microscope, we can determine, by the aid of needles, the existence of scales which arise from the surface of the thallus. There are present here three diiferent forms of ventral scales : marginal scales, which usually extend somewhat over the edge of the thallus, and have become brown ; median scales, which lie in the middle line, and laminar scales, which are inserted upon the thallus on both sides of the middle line, can also be wanting. The median scales, often purple-coloured, alternate with one 'another, their edges overla]) in the middle line. Together with median or laminar scales, or with the former only, arise out of the frond fine rhizoids, which, covered by the scales, and following their insertion, attain to the mid-rib, and here run further forwards in bundles. It is the median and laminar scales which produce the striation on the under side of the thallus, which we have already observed with the naked eye. If we examine the dorsal [upper] side of the thallus Avith the lens, this appears to be divided into small diamond-shaped areas. The limits of the areas are dark-green, the areas themselves appear more grey. In the middle of each area a dot-like opening is visible. We now examine, with stronger magnification, a sec- tion which is taken parallel to the dorsal side of the thallus. We see that the outer cells of the dorsal surface are polygonal, fii-mly united together, and contain numerous large chlorophyll granules. The boundaries of the areas show clearly ; each area has its centre occupied by a round opening, which is surrounded usually by four narrow cells containing no chlorophyll, and curved into the form of a crescent (Fig. 72, A). Where the section is somewhat thicker, air is seen to be collected under the free outer surface of the area. Into this air space, the air-chamber, project chlorophyll-containing threads of cells. The walls bounding the air-chamber laterally are constructed of closely-combined cells. These walls are uni- to multi-lamellar; their cells contain chlorophyll. Single cells of the surface, and also of the interior, are distinguished by a highly-refractive, irregularly-outlined, grape-like body. These bodies in the younger shoots are slightly brownish, in older arc coloured brown, contain mostly fat oil, and form the so-called 196 STRUCTURE OF THE oil-bodies of the Liverworts.^ The cells which contain such a body show no other formative contents. Surface-sections taken from the ventral side of the thallus show no division into areas. The cells are here more elongated and poorer in chloroiDhyll than on the upper side. The rhizoids which spring from the ventral surface, show a double structure. They are more slender, and provided with peg-like projections into the interior, or thicker and without such thickenings. Those with the peg-like projec- tions arise out of the frond as far as the median or laminar scales, or only the former, extend. They lie close to the frond, and follow the mid-rib in bundles, covered by the scales. They serve, perhaps, the purpose of stiffening the thallus. Fig. 72.— Mar cliantia poIymorpTia. A, an air-opening from above. B, in cross-section (x 240). The slender rhizoids proceed chiefly from the midrib, and turn at an acute angle towards the substratum, to which they fix the thallus. At their apex they often appear sinuately lobed, at the base commonly purple-coloured. All ventral scales are uni- lamellar, the median consist of still living cells, the laminar and marginal scales of cells which quickly die. — A cross-section through the thallus shows us on the dorsal surface first a zone of chloro- phyll-containing tissue. The interior of the thallus is composed of broader cells, almost free from chlorophyll. In the walls of these cells stellate broad elliptic pits are to be seen. At the ven tral surface the two last layers of cells are again narrower, flatter, rich in chlorophyll, and form the so-called ventral cortical layer. Oil-bodies are scattered through the entire tissue. Other individual THALLUS OF MARCHANTIA. 197 cells are noticeable from their size and highly refractive contents; these are the mucilage-cells, which in Marchantia are feebly, but in other Marchaiitiacea3 are more strongly, represented. — A closer study of the outer layers, rich in chlorophyll, of the doi-sal surface, completes the conception which Ave had obtained from the surface- view. Outermost we see a single layer of fiat cells, which pro- ceed from the walls bounding the air-chamber laterally, free over the chamber. In the midst of the free outer wall is found the air-opening [the so-called stoma], which is surrounded by several, from four to eight, tiers of cells ^ (Fig. 72, B). The opening is narrowed at its upper and under apertures, especially at this latter, and therefoi-e shows with a barrel-shaped form. The cells of the upj)ermost stage are prolonged into a membranous border. As the air is very strongly retained in the air-opening, and the structure is thereby made indistinct, it is desirable previously to pump the air out of the preparation. Into the air-chamber project from below threads of cells, two or three cells long, and now and again branched. These threads are especially rich in chloro- phyll ; they arise from the flat cell-layer next below, which is poor in chlorophyll. On the ventral side of the thallus we see on the midrib the lateral, alternate overlapping of the median scales. Between the scales lie the cross-sections of the bundle of rhizoids. Median longitudinal sections show the insertion alike of the stronger ordinary rhizoids, turning ofE at once from the thallus, and the " pegged " rhizoids overlying the midrib. A very simply constructed, and in many respects very instruc- tive, thallus is that of Metzgeria furcata.'^ This inconspicuous plant is met with on the cortex of leafy trees [is cosmopolitan, and very variable]. The thallus is riband-like, bright-green, bifurcate, and traversed by a midrib visible even with the naked eye. Apart from this midrib, the thallus is unilamellar, as can be readily proved under the microscope. It consists of polygonal cells, richly provided with elongated chlorophyll grains. The narrow midrib projects from the ventral [under] surface much more strongly than from the dorsal [upper] surface. It consists, passing from above downwards (as can be readily proved by focus- sing to different depths), of broad, slightly elongated, then of narrow, elongated, and finally again of broader cells. The two outer layers of cells contain chlorophyll, the inner, on the other hand, do not. At the growing-point arise, from the ventral surface of the midrib, some rather short club-shaped hairs, at their for- 198 STRUCTUEE OF THE ward end filled with higlilj refractive contents. From older parts of the midrib, and also from the marginal cells of the thallus, arise bristles, which, under favourable circumstances, can at their extremity give rise to a lobed suctorial disk, and then functionate as rhizoids. These bHstles always stand at the hinder end of the cell (i.e., most removed from the growing point), from which they are cut off by a curved partition wall, which does not pass through the entire height of the cell in question, but only cuts off a corner or angle from it. As the cross-section shows, the inner cells of the midrib are distinguished by somewhat more strongly Fig. 73.— Apes of ashootof Metzgeriafurcaia. t, Apical cell ; si -sVil. successive segments ; mi. marginal cell of the first, inii. of the second grade ; pi surface-cell of the first grade; ii, inner cells of the midrib; c, club-shaped hairs. inner cells of the midrib ( x 54iO) . The figure drawn by focussing into the thickened glancing white walls, appearing almost collenchymatous \_cf. footnote to Pohjtrichum, p. 192]. In the most instructive and easy manner the processes of division in the growing point of Metzgeria can be followed.^ The growing apex of Metzgeria shows a comparatively slight depression. The base of this apical sinus, close up to the place where the midrib commences, is occupied by the apical cell. We examine it from the dorsal sur- face of the thallus, in order not to be disturbed by the club- shaped hairs. The apical cell is wedge-shaped (Fig. 73, t). It TUALLUS OF METZGERIA. 199 has the form of an isosceles triangle, with the base directed forwards (outwards), and usually somewhat convex, and slightly arched side-walls. It divides by walls which are parallel with its side- walls, and, thus developing, gives off segments right and left (5), and therefore all lying in one plane. NOTES TO CHAPTER XVIII. 1 Compare P. G. Lorentz, Jahrb. f. zciss. Bot. Bd. VI., 1SG7-8, p. 3G3 ; Goebel, Grundriss der systematisclien und specicllen Pjlajizenmorpholorjie, 1832, p. 18-4 (for the literature see also on p. 179); recently also G. Fritsche, Ber. d. deutsch. hot. Gesell , I. Jahrg. p. 83 ; Haberlandt, ditto, p. 203 ; and Oltmanns, in Colm's Beitr. zur Biol. Bd. IV. p. 1. 2 Compare Leitgeb, Untersuch. ilb. die Lebermoose, VI. Heft, 1881, where the other literature is to be found. ^ Pfeffer, Die Oelkorper der Lebermoose, Flora, 1874, No. 2. 4 Voigt, Beitrag zur vergl. Anat. der Blarchantien, Bot. Zeit. 1879. Col. 729. 5 Compare Leitgeb, Unters. lib. die Lebermoose, Heft. III., p. 34, where also is the other literature. fi Compare Knv, Jahrb. f. loiss. Bot. Bd. IV., p. 85. 200 VEGETATIVE STRUCTURE OF THALLOPHYTES. CHAPTER XIX. VEGETATIVE STKUCTUEE OF FUNGI, LICHENS, AND ALG.E. STAINING THE CELL-CONTENTS. Material Wanted. Mushroom {Agaricus caiiipestris). Fresh. A Lichen, such as Parmelia {AnaiJtycMa) cUiaris, common on trees. Fresh. Cladophora glomerata* A common fresh-water Alga, Fresh. Sinrogyra majuscula,* or other similar species. Fresh. The yegetative organs of the Fungi consist, apart from a number of the simplest forms, of elongated, thread-like, mor(3 or less copiously branched elements, the Hypliae. These are either with- out partition walls [unseptate], unicellular throughout their entire body; or, by means of partition walls [septa], seg- mented into a number of consecutive cells. Moreover, the most massive fungal structure is composed of such hyph^B, then very much interwoven with one another. The hyphce can indeed, in many cases, become so firmly united, side by side, that a tissue is produced, which, as pseudo-parenchyma, delusively imitates the appearance of the parenchymatous tissue of the higher plants. The pseudo-parenchyma, however, is a product of the union of cell-threads, and not the result of cell- division taking place in three planes. In order to inform ourselves about this kind of structure, we take the fruiting body of a hymenomycetous fungus as subject for investigation. We choose the spore-producing body of the Mushroom Agaricus campestrisy because the fungus can now be obtained at any season of the year, and shows, besides, a comparatively simple structm'e. We prepare first a delicate longitudinal section, from the stalk of a fully-developed speci- men. We recognise clearly a structure of longitudinally disposed hyplise, and can readily tear the section in longitudinal direction * This, and many other Algfe, etc., can usually be obtained from T. Bolton, Newhall Street, Bhmingbam. [Ed.] STRUCTURE OF THE MUSHROOM. 201 with the needles. The hyphag are directed more or less parallel to one another; single ones run obliquely between the others. Each hypha forms a cell-thread, which is branched here and there by the formation of side-branches. These arise either close under a partition-wall, or else lower down on the side surface. Here and there we come across a " blind " end of a branch. The cells of neighbouring hyphee not infrequently appear connected by a horizontal branch, and openly communicate with one another. In the periphery of the stalk the hypha? are narrower, and at the same time more closely pressed together; just under the surface the walls become brown, and their cavities more or less completely collapse. Towards the middle of the stalk likewise the hyphce become smaller, but their texture much looser, and, at the same time also, their course quite irregular ; great masses of air here fill up the interspaces between the hyph^e. So long as the destructive influence of water has not made itself felt upon the contents of the hyphoe, very little of these contents is to be noticed ; only at the cross-walls does it show, here and there, more markedly collected. Later on large vacuoles begin to form in the cells. Here and there small crystals are met with in the cells. The cross-section of the stalk has a parenchymatous appearance, which is only lost in the middle parts of the section, where the hyphse also offer their side views. This pseudo-parenchymatous tissue appears as if composed of unequal, irregular polygonal cells, which leave between them more or less numerous inter-cellular spaces and gaps (Fig. 74). On careful examination of the section, we notice close in the middle of many cells a refractive point (c/. the figure). The section has here grazed a cross-wall, and the middle point shows the position of a pit, which is clothed, on either side of the partition wall, with a small collection of a highly refractive Fig. 7i.—A,jaricns cam- substance. Such pits in the centre of the J""'"'- ^*^' "^ * "°'^- ^ . . section through the stalk. cross- wall arc universally distributed amongst in two hypha> the section Basidiomycetes and Ascomycetes2. The cells ^^^ sv^^<^^i the cross- •^ ... walls ; a central point can of the hyphco contain in the periplieral proto- be seen upon them (x plasm, numerous very small nuclei, which, ^^^' however, are not easy to see, and the method of identification of which we shall postpone. 202 STRUCTURE OF LICHENS. UiDon the structure of the stratum (thallus) of Lichens, we shall best obtain information by AnaptycMa {Parinelia) ciliaris, universally distributed on tree stems. The thallus itself is erect, leaf-like, and shrubby [foliaceous-fruticose] ; on the dorsal [upper] surface, grey-green to bright green; on the ventral [under] surface grey. From the edges of the thallus arise stiff cilia, which often become lobed at their ends, and, where they extend to the sub- stratum, adhere to it. We hold a piece of the thallus between two pieces of elder-pith, and cut cross-sections through it. By a sufficiently strong magnification, we see that the thallus consists, on its dorsal surface, of closely interwoven, thick-walled hyphas. These form the so-called riud, or cortical layer. Passing farther inwards, the curves of the hyphse separate from one another, in order to form the looser central layer. We can readily decide that the hyphse are long sacs, branched from time to time, and divided by cross- walls. At the limits of rind and central tissue lie scattered comparatively large, green, globular cells, the Gonidia. They correspond with the alga Gystococcus humicola [ = Ghloro- caecum humicolum']. The hyphee fit closely to the gonidia, and carry to them the crude sap, for which they receive in retui-n a portion of the substances assimilated in the gonidia. There exists here a symbiosis, a conjoint existence of fungus and alga, which is based upon reciprocal service. At the under surface of the thallus of AnaptycMa the fungal hyphas again interlace more closely, so as to form a kind of under rind; or this closer combina- tion does not exist, and the looser central tissue extends to the ventral surface. This latter is in general the case. At the edges of the thallus, however, the rind of the dorsal surface, in all cases, extends underneath to the ventral side. From these edges arise, as we have already determined macroscopically, the fixing cilia (rhizines), which now can be made out to consist of parallel closely-combined hyphss. The walls of these hyphge have a brownish colour. At their base the threads often fork. In other Lichens the rhizines are apt to spring mostly from the ventral surface of the thallus. Chlorzinc iodine stains the walls of the gonidia immediately a beautiful blue, while the hyphse only appear yellow to yellow-brown, showing the reactions of the so-called fungal cellulose. In AnaptycMa ciliaris we have a Lichen with what is called a layered or heteromerous thallus, so called because the gonidia form a special layer in the thallus. In more lowly organized STRUCTURE OF CLADOPHORA. 203 tlie liclien-tb alius, differ green or blue-greeu, but lichens the tliallus is homoiomerous, i.e., the gonidia are distributed through the whole tissue. To the last belong also the Gelatinous lichens, in which the gonidia lie in a translucent jelly, which is penetrated by the hyph89 of the fungus. Moreover, the Alg83, which take part in the formation of according to their species, are coloui c I belong almost exclusively to the lowest divisions of the Algce. The Cladophorea3'^ present themselves to us as abundantly branched green threads, whose segments decrease in thickness with the grade of the branch- ing. They are the most widely dis- tributed of all fresh-water alga), and any species is suited for examination. The determination of species is, however, very uncertain in this genus. We select a dark-green, undulating, tuft-forming Cladophora glomerata for further examin- ation. This is corymbosely branched, the side-shoots arising, as in all other Cladophorea3, from the upper end of the constituent cells. The branching pro- ceeds acropetally, so that the end cells of the branches act as apical cells. Sub- sequently branches arise also from the older segments, producing what we may call adventitious shoots. With sufficiently strong magnification, the green peri- pheral protoplasmic la3'er of the cells shows to be composed of small polygonal plates, the chromatophores (Fig. 75, ch), separated by delicate colourless lines. In each plate more or less numerous pale grains (of starch) can be seen ; be- sides these, in some plates, lie compara- tively large, more or less regularly globular bodies, more strongly refractive, which arc generally known as amylum-hodies, and more recently as pyi'enoids * (p), and in which an inner grain is more or less clearly distinguishable from an outer layer. The cells are seen to be filled internally with %?p lo QX^r^ "a ' o 5"^ y ^^CL" "i o @^ o / Fig. 75. — Cladopliora glome- rata. A cell of a thread from a chromic-carmine preparation. «, a nucleus; ch, chromato- phores (colour-bodies) ; )'. nmy- lum-bodics(pyrenoid!!;); a, starch grains ( x 610). 204 FIXING AND STAINING cell-sap, wliich is traversed by colourless, exceedingly thin proto- plasmic plates, which, proceeding from the peripheral layer, divide the cell cavity into irregular, unequal, polygonal chambers. Here and there chromatophores [colour-bodies] can be seen in the inner protoplasmic plates. By focussing so as to get an optical section, it will be seen that colourless protoplasmic balls here and there project from the peripheral protoplasm into the cavity of the cell. These are the nuclei, in which, in especially favour- able situations, a nucleolus can also be distinguished. In Clado- phora, as is clearly proved by this examination, we have to do with multinuclear cells. If now the preparation is somewhat firmly crushed, we see in the flattened cells, the contents of which are somewhat withdrawn from the walls, the individual chloro- phyll-plates, separated from one another, and rounded. At the same time the small grains and amylum-bodies show up clearly in the chromatophores, which now a^Dpear, like the chlorophyll- grains of higher plants, acted upon by the water. If we now add a little potassium-iodide-iodine solution to the preparation, the small grains, and also the outer layer of the amylum-bodies, colour violet ; in the green chromatophores, however, they appear brown, and the partially visible nuclei also take a brown colour. We must not omit in this preparation to look out uninjured cells, in which starch-grains and amylum-bodies are stained in their natural position, and are very well defined, and we can also, by deeper focussing, distinguish the nuclei. We now examine another thread, which we lay directly into a drop of picric-alcohol, when the nuclei of the amylum-bodies are sharply defined in the yellowish-brown stained protoplasm. Sufficiently strong magni- fication presupposed, these bodies appear angular ; they are protein crystals,^ of which, moreover, two not infrequently lie in an amy- lum-body. After a short time irregular brown bodies appear in the chlorophyll-plates, which proceed from the disorganized chloro- phyll, and give us the hypochlorin or chlorophyllan reaction^. The same reaction will be obtained under the influence of other acids. However, in order to be able to study the nucleus more closely, and to obtain a complete insight into their distribution, we will bring other methods into use. This will besides give us oppor- tunity of learning some approved methods of " fixing " and staining, which histological studies have, in recent times, to thank for not unimportant advances. We place some branches of the CladopJiora in 1 per cent, chromic acid, other small portions in con- THE CELL-CONTENTS. 205 centrated picric acid, still others in 1 per cent, clirom-acetic acid (chromic acid 07 per cent., acetic acid OS per cent.) 7. In doing this, we must take care that the reagent is at least 100 times the bulk of the object to be fixed. The 1 per cent, chromic acid, and the chrom-acetic acid, we allow to act for some hours, even without disadvantage for 24 hours, and the picric acid for about 24 hours. All these objects must afterwards be washed most carefully in distilled water; they can with advantage remain for up to 24 hours in water which is frequently changed. Specially careful treatment is required by the i^icric-acid preparations when the}' have to be stained with hcematin-amnionia. — The variously "fixed" and well-washed preparations are now laid in watch glasses with Beale's carmines,^ with Thiersch's or Grenacher's borax-carmine, and also with Hoyer's neutral carminic-ammonia. In Beale's carmine the sections must remain for up to 24 hours, about half the time in Hoyer's carmine, several hours in borax- carmine. Another portion of the threads we stain with Grenacher's or Boehmer's hasmatoxylin [logwood], which, if it is to stain well, must be as old as possible. This solution is used very greatly diluted. It is best, from time to time, to control the extent of the staining of the object by slight examination under the microscope, and to take it out when it has taken up sufficient colour material.* If, in spite of this care, the object should be ovei'stained, that is should be stained too darkly, it is laid in pure water, or in watery alum solution, or in water containing a trace of hydrochloric acid, and left in the fluid in question until the intensity of the coloration is diminished to the required degree. If the preparation has been treated with acidulated water, it is necessary afterwards to wash the preparation for some minutes in very weak ammonia water. In order to be able to stain the preparation according to the hoematin-ammonia method,^ we must have previously removed from it every trace of picric acid. For this purpose we transfer it to a comparatively large quantity of boiled water, which we repeatedly change. In this water, freed from carbonic acid gas by its previous boiling, the object remains for from 24 to 48 hours, after which it can be stained. For this purpose we throw some crystals of hasmatoxylin in a small quantity of distilled water, * This is perhaps best effected if, instead of staining in a watch-glass, the process is carried ou upon an object-slide, in which a round or oval hullow h is been ground. The slide can be placed bodily on the stage of the microscope, and is handier in use for this than a watch-glass. [Ed.] 206 FtXING AND STAINING and aerate it with ammonia gas. This latter we effect with the aid of a wash-bottle containing some ammonia solution, in which the two glass tubes do not reach the fluid. The hoematoxylin now dissolves with a beautiful violet colour. The solution is greatly diluted with distilled water, and the preparation allowed to lie in it for about two hours. The exact time for coloration can here also be directly controlled. The preparation is, with advantage, somewhat overstained, and afterwards steeped for several hours in distilled water. This method of staining is somewhat trouble- some, but often gives, however, the most exquisite results. Pre- parations hardened otherwise than with picric acid, are little suited for staining with hoematin-ammonia. The preparations treated with Beale's carmine, with borax-carmine, or with Hoyer's carmine, are likewise most beautiful when they are overstained, and afterwards laid for some time in a watch-glass, in 50 per cent. to 70 per cent, alcohol, to which is added a drop of hydrochloric acid. (For this purpose we can keep ready prepared a solution of about f per cent, hydrochloric acid in 70 per cent, alcohol.) Previously these preparations show a more or less diffused color- ation; they first acquire a definite staining in the hydrochloric- alcohol. The preparations laid in acidulated alcohol are in all cases washed afterwards with alcohol containing no acid. If after completed examination we wish to make permanent preparations of the stained objects, we choose for our preserving medium either glycerine or glycerine- jelly, or, for carmine prepa- rations, Hoyer's mounting fluid. If the heematoxylin stain is to be preserved in glycerine or glycerine-jelly, this must be completely free from acids. The foregoing preparations must not be trans- ferred immediately to the enclosing medium in question, as other- wise the cells, as the result of sudden withdrawal of water, would collapse. These preparations are, therefore, first laid in very dilute glycerine, which, by standing exposed to air, very slowly concentrates. The threads can then, without prejudicial results, be transferred to glycerine, or to glycerine-jelly. The glycerine preparations are closed with Canada balsam. The glycerine- jelly, or Hoyer's mounting fluid, needs, as we have already seen, no further enclosing. We will now submit the various preparations to close study^ and find that the chromic acid, or chromic S^cid mixture, prejjara- tions, stained with forms of carmine on the one hand, and pre- parations which are suitably fixed and stained with haBmatoxylin THE CELL-COXTENTS. 207 and htematin-ammonia on tlie other hand, show themselves in the foregoing cases to be the best. It must, liowever, be explicitly- stated that this result is limited only to the objects in question, and for other objects another method, which here is less advan- tageous, might have the preference. It also happens only too fre- quently that a stain formerly approved fails for unknown reasons, and, therefore, a conclusion should never be based upon an isolated case. In general, the fixing and staining of the cell- contents has become a special art, which must be learnt, and requires practice, so that in our first attempts we must be prepared for failures. We have chosen Cladopliora as a suitable object for introduction to the various methods of hardening and staining ; \vhocver wishes to limit himself to the most certain, rarely failing, method, will harden in the above way in 1 per cent, chromic acid, and after- wards stain, one part with borax-carmine, another with hema- toxylin. The borax-carmine stain almost always succeeds. In the borax-carmine preparation (Fig. 74), the nuclei stand out quite sharply. The amylum-bodies (pyrenoids), together vfith the rest of the protojDlasm, remain as good as unstained, and the starch-grains also take no coloration. The amylum-bodies now show clearly in their interior the more strongly refractive protein crystal, which is surrounded by a hollow ball, which gave us earlier the starch reaction with iodine. The nuclei, to which we specially turn our attention, are distributed pretty uniformly in the cell ; they lie on the inner side of the chlorophyll-layer, and project into the cavity of the cell. Each nucleus shows a more darkly-stained nucleolus, and appears, besides, as if finely granu- lar or finely porous. The h^ematoxylin or hfematin preparations show the nuclei stained dark, and besides, though more faintly, the crystal in the chlorophyll-vesicle. The starch-grains are not stained, but on the other hand the microsomata (microsomes) of the cell protoplasm are, and almost as darkly as the crystals of the chlorophyll- vesicles (amylum-bodies). The genus Spirogijra furnishes us with a simple filament or thread of cells. We choose for examination a species wliieh has a central, readily-visible nucleus. So constituted, for example, is Spirogyra maj iiscula ^^ \_S. orthospira'], which is met with now ,and then, not exactly rarely, but sporadically, in pools. For this purpose other species with central nucleus will serve equally well for examination, and will differ but slightly in the essential relations of their structure. If once in possession of good 208 STRUCTURE OF SPIROGYRA. Spirogyra material, you should endeavour to preserve it in culti- vation. This is effected best in comparatively shallow vessels, whose vt^alls are either opaque, or are made opaque by means of black paper, as light falling unilaterally acts disadvantageously. The vessels must stand in a light place, but protected from the direct action of the sun. Into either river or spring water, which is not too rich in chalk [too "hard"], are thrown fi-om time to time boiled pieces of turf soaked in a nutrient fluid. This nutrient fluid will be prepared suitably if we add to 100 ccm. water, 1 gi'am nitrate of potash, |^-gram sodium chloride, |-gram sulphate of lime, "l-gram sulphate of magnesia, |-gram finely pulverized phosphate of lime (of this latter salt, only a trace is soluble) .^^ Under such circumstances the Spirogyra, and fresh-water algae in general, thrive well. The cells of Spirogyra majuscula, when fully developed, are about IJ to twice as long as thick (Fig. 76). The cell- wall is lined by a de- licate, colourless, peripheral layer of protoplasm, which becomes clearly visible if the cells are plasmolysed, i.e., if the protojDlasmic body of the cell is made to contract by some water- withdrawing me- dium, such as sugar-solution, glycerine, solution of common salt, or of salt-petre. To the colourless lining-layer follow 8 to 10 chlorophyll -hands, which usually appear pretty steep and closely wound. The bands have a finely undulating outline, and are transparent enough to admit of a view into the interior of the cell. At irregular distances in the bands are imbedded denser, globular, colourless, bodies — the amylum-bodies with which we are already acquainted. The amylum-bodies show a protein-crystal, and a hollow globe of small starch-grains as an enclosing sheath. We know the angular outline of the crystals even without reagents ; they stand out more sharj^ly if some picric alcohol is run under the cover-glass. By treatment with potassium-iodide-iodine, from the conjoint staining of the starch-sheath and the protein-crystal, the whole body appears dark-brown. The central nucleus in this species is spindle-shaped; it becomes, nevertheless, by pressure Fig. 76. — Spirogyra majuscula. A cell of a thread gradually focussed into, showing there- fore, besides the chlorophyll-bands, the nucleus with its suspending threads (x 240). STRUCTURE OF SPIROGYRA. 209 ujDon the cell, brought out of its position and visible from its side, and then presents the form of a disk ; it has, therefore, in reality, the form of a bi-convex lens. In its centre lies a large distinct nucleolus; seldom two or three such bodies are distributed sym- metrically in the interior of the nucleus. In other more nearly- allied species the nucleus is thicker, and appears in its natural position in the cell as rectangular with rounded corners. The nucleus is surrounded by a very thin layer of protoplasm, from which delicate protoplasmic threads run out towards the peripheral protoplasm of the cell. By these threads the nucleus is suspended in the cell-sap filling the cavity of the cell. The threads all arise from the thin margin of the nucleus, usually fork repeatedly in their course, and join on to the inner side of the chlorophyll bands, and in all cases at the projecting parts which cover the chlorophyll- vesicles (i^yrenoids) . We can convince ourselves of this in most cases easily by slowly changing the focus. NOTES TO CHAPTEE XIX. ^ H. Hoffmann, Icones anal, fung., I.-III. ; De Bary, Morph. d. Pllze, etc., pp. 49 et seq. " On the pits in the partition walls of the Florideas, compare Bornet, Etudeit Xihycologiques, p. 100; and Schmitz, Stzber. d. kgl. Akad. d. IViss. z. Berl, 1883, p. 218. 3 Schmitz, Siphonocladiaceen, p. 17 ; Strasburger, Zellblld. u. Zcllth., III. Aufl., p. 204. •* Schmitz, Chromatophoren d. Algen, p. 37; compare also pp. 16 and 35. ^ According to a communication from A. W. Schimper. ^ Pringsheim, especially in the Jahrb. fiir %ciss. Bot., Bd. XII., p. 294; A. Tschirsch, Ber. der dent. hot. GeselL, Bd. I., p. 140, where the literature is also given. ' Flemming, in Kernsuhstanz, Kern- und Zelltheihuig, 1882, p. 379, where the literature also is given. 8 The property of the nucleus to take up and accumulate colouring matters was discovered by Th. Hartig: " Ueber das Verfahreu bei Behandlung des Zellkerns mit Farbstoffen," Bot. Ztit., 1854, col. 877. Enlwickhingsgesch. d. Pjlkebns, 1858, p. 154. In animal histology the treatment was introduced by Gerlach. Mikr. Stud. a. d. Geb. d. vienschl. Morpholg., 1858. 9 Compare Schmitz, Stzhr. d. niederrh. Geselbch., 13th July, 1880 ; separate reprint, p. 2. 1" Strasburger, Zellbildung und Zelltheilung, III. Aufl., p. 173. 11 Nutrient fluid, according to Sachs, Vorlcsnngcn iiber Pjlanzcn-Phyiiologle, p. 342. 21U - UNICELLULAR PLANTS. CHAPTER XX. DIATOMACE.E, PEOTOCOCCUS, YEAST, SCHIZOPHYCEiE, (SPLITTING ALG.E). Material Wanted. Some large diatom, e.g., Pinmdaria {Navicula) vlridis. Living. Protococcus vlridis, from a tree trunk. Living. Yeast [Saccharomijces cerevisioe), from a brewery. Living. Anahcena AzolhE, or Azolla caroliniana. Living. Oscillaria, sp., from standing vrater or wet soil. Living. Gleoca-psa, sp., from damp walls, or the glass of a fernery. Living. The DiatomaceaB or BacillariaccEe are unicellular organisms, occupy- ing an intermediate position between animals and plants, and form an isolated group. The most favourable object upon which to get information as to the structure of the Diatomaceae is, perhaps, Pinmdaria \_Navicula'] viridis,^ a species very common in standing and flowing [fresh] water. It is distinguished amongst fresh- water forms by its comparatively large size, and allows in general an easy insight into the structural relations of its body. Under the microscope, in w^hich w^e must study them with the strongest objective at our command, they appear either in the form of an elongated ellipse or as a rectangle with somewhat rounded ends. In the former case, we see them from the side of the valve [frustule] (Fig. 77, A), in the latter, of the girdle [or joint of the valves] (Fig. 77, B). We will call these the valve-side and girdle- side respectively. On the valve-side, the cell- wall appears marked with narrow furrows, running from the edges towards, but with- out reaching, the middle (compare the figure). They are usually considered to be depressions in the outer surface of the valve, i.e., thin places therein. The central, smooth space, free from the furrows, shows at its middle and each end, a strongly refractive thickening, which we distinguish as a nodule. The tw^o end nodules are joined to the median nodule by a line, which bends out symmetrically close on either side of the median nodule, and DIATOMACEyE. 211 ends in a slight enlargement. The end nodules are surrounded bj the ends of the line in the form of a crescent, to permit which the lines bend out at both ends laterally in the same direction as at the median nodule. In its course between the nodules, the line broadens a little. We assume that it is a cleft leading into the interior of the cell ; it is the raphe. The furrows do not pass on to the girdle side (B) ; we see them only at the edges of the figure. By focussing for the optical section, and careful examination of the ends of the cell, we can demonstrate the remarkable fact that the middle line of the wall is double. From exhaustive investi- gation, it is settled that there is here an overlapping, box-wise, of the separate parts of the wall. At the edges of the two elliptic wall-segments, which we saw in the view of the valve-side, portions of a membrane adjoin, which end with free margins. The wall of this cell, therefore, consists of two halves, of which the one is inserted inside the other. The structure of this wall indicates throughout that of an elliptic box with a cover placed upon it. The side walls of the cover are just as high (deep) as those of the box, but they are not completely slipped the one into the other. If we return, in our cell, from the optical section to the surface view, we can follow the thin edges of the two halves of the cell as delicate lines. The grooved surfaces of the cell-wall we distin- guished as valves, the smooth free-ending side walls as girdles, whence the use of the terms in question to indicate the two views. In Finnularia it is easy to free the one half of the cell- wall from the other by pressure or by chemical reagents, and, moreover, here, and there dead specimens are found in which this process has more or less completely taken place. With pressure the girdles easily break at some little distance from their edge, and along a line parallel with it. These lines, one near each edge, and there- fore two in girdle-view, are often recognisable, and may be thin parts of the girdle. They do not extend to the ends of the cell. The contents of the cell present a somewhat different appearance according to whether we have a valve- or girdle-view. In the former (Fig. 77,^1), a median clear strip traverses the cell from end to end ; the colourless cytoplasm of the cell is therefore visible. In the mid-length of the cell it a])pears collected into a bi-concave protoplasmic bridge. In this " bridge " lies the nucleus, not always readily visible without the use of reagents, and with a comparatively large nucleolus. Bounding both sides of this clear band, with a tolerably smooth or undulating outline, are the brown- 212 DIAT0MACE-S3. coloured cliromatopliores (colour-bodies), the endochrome plates. These lie, therefore, on the sides of the girdle. In the proto- plasmic "bridge " can be seen narrow rodlets, connected in pairs, the meaning of which is unknown. Lastly, in the cell-sap lie usually, but not always, larger and smaller oil-drops. In the girdle-view the cell-body appears uniformly brown, because here the chromatophore covers the whole colourless peripheral proto- plasmic layer. Only at the tAvo extreme ends of the cell does the colourless protoplasm come to view. The chromatophore is uniformly dense and uniformly coloured, without visible differentiation. In girdle- view also the central collection of proto- plasm appears to have the form of a bi-concave bridge. If we examine now our former pre- paration of Gladophora, we are pretty certain to find diatoms clinging to this alga. They were fixed and stained at the same time with the alga, and we shall see the stained nucleus beautifully in each cell. Amongst a large number of examples of Pinnularia we may here and there find one double. These are sister-cells, which have recently resulted from the division of a mother-cell. They cling to one another with their valve sides, and, if the wall is fully developed, we can determine that the girdles of the two inner valves are inserted in the two outer valves. After division of the con- tents of the mother-cell, these inner halves of the wall are developed for each individual. Each cell, therefore, has an older and a younger half of its wall [i.e., one valve, the outer one, belonged to the mother-cell, and the other, inner valve, is peculiar to the present individual], and this consideration shows that the difference of age between the two valves may be very considerable. The Pinnularia cells are motile. They commonly progress in the direction of their long axis, either uniformly or by jerks, also turning off now and then laterally from their path. They do not Fig. 77. — Pinnularia viridis. A, View of the valve-side. B, View of the girdle-side ( x 640 ). DIATOMACE^. 213 swim free, but rather creep on some substratum, and it is there- fore probable that from the line indicated as a cleft, which we saw in the middle of the valves, a delicate protoplasmic edge is protruded, and forms the organ of movement as a kind oE pseudo- podium. We now place a preparation of Finnularia on a plate of mica, and heat it over a gas or spirit flame. We then lay the plate of mica [when cold] upon our object-slide, and observe the prepara- tion dry, but under a cover-glass, with strong magnifying power. We can see that the Finnularia remain as perfect skeletons. With short heating they become somewhat brown, from the carbonized organic substances; with longer continued heating they are colour- less. Hydrochloric acid does not touch them; they consist of silicic acid [like flint], and retain and show the finest peculiarities of the cell-wall, which must therefore have been silicified in a high degree. The furrows show in this preparation very clearly as dark s trice, and are besides extremely good for studying the structural relations of the wall. Especially beautifully visible in valve-view are the clefts, which run on both sides from the median nodules to the terminal nodules. Their enlargement at mid-length i? manifest. In the girdle- view the edges of the two halves of the cell- wall show clearly ; moreover, on the overlapping parts are seen two lines, parallel with one another and with the edges of the valves, which do not extend to the ends of the cell. Flint-skeletons quite as beautiful as these are also obtained if we first allow a di-op of concentrated sulphuric acid to act upon our diatoms, and after some time add 20 per cent., and then gradually concentrated chromic acid, and finally remove these reagents with water.^ Diatom valves which are poor in silex (flint) will neither bear heating red- hot, nor this last method of procedure ; they must instead be laid for from four to seven days in hydrochloric acid, to Avhich a little chlorate of potash has been added. In case the valves are still not quite clear and separated, it is advisable after this to lay them for two days in ammonia, and afterwards transfer them to nitric acid. The remarkable phenomenon of the composition of the cell -wall out of two pieces is, moreover, present in the other Diatomacea?. Similar motility is likewise observable universally in the free living forms. Even many which grow upon and enclosed in a gelatinous tube, are, if freed, capable of movement, while this appears to be usually wanting in thread-forming species. On 214 PROTOCOCCUS. account of the often exceedingly delicate [and regular] structural relations of their cell-walls, diatoms are much used as test objects, in testing the quality of the more powerful microscopic objectives. Especially used are the valves of Pleurosigma angulatum, which, with sufficiently strong magnification, shows regularly arranged hexagons. In order to come to know one of the simplest possible forms of the unicellular green algoe we will examine a Frotococcus. To this belong in the main all the green incrustations which are found on the stems of trees, damp boards [e.g., wood palings, etc.], walls, and other similar places. In this let us note that it is quite un- certain whether our' Frotococcus is an independent species, or is not rather to be considered a stage in the development of another alga.^ The form (Fig. 78), which we have removed from an old tree trunk comes under the name Frotoccocuff viridis. We examine this with a strong magnifying power, and find it composed of globular cells, iso- lated or united into small families (Fig. 78, A to F). The contents of the cells are bright green, but the whole protoplasm is not uniformly col- oured, but rather, as sufficiently strong magnification shows, a number of chromato- phores are present, which, in lateral contact, occupy the surface of the cell-contents. Where their contact is not complete, the colour- less protoplasm comes into view. More or less in the middle of the cell lies the nucleus, with its nucleolus, which, however, is not usually visible without the help of reagents. The cells have a thin wall, which stains violet with chlorzinc iodine. Numerous cells are usually in course of bipartition by means of a partition wall, which cuts the globular cell in halves (Fig. 78, D). The divisions of adjoining cells take place in planes either parallel or cutting one another at right angles. The daughter-cells, becoming rounded off, soon go out of union with one another (G, F) ; they Fig. 78. — Protococcus viridis, after treatment with potassium-iodide -iodine. In D, the cells on the left have just divided (x 540). YEAST-FUNGUS. 215 remain, however, for some time clinging to one another, or else become completely separated. If the cells are treated with potassium-iodide-iodine, the nuclei show up clearly (our fioures were sketched from iodine preparations). In each nucleus the nucleolus is clearly visible. In the cells which have just ai'isen by division, the nuclei lie against the young partition wall (D). The iodine solution shows small starch-grains in the chromato- phores, but no amylum-bodies. Very simply constructed organisms are met with in the colour- less fungal cells hitherto collected together under the name of Saccharomycetes. We provide ourselves with some yeast, the ferment used in brewing beer, and examine a trace of it, diffused in water, under a high power. We find the field of view filled with small cells, individuals of the so-called yeast fungus, Sac- charomyces cerevisicB.'^ The cells appear globular or ellipsoid; they have a delicate membrane, and in the interior can be recog- nised a large or several small vacuoles, and some strongly refrac- tive granules (Fig. 79, 1). A nucleus cannot be distinguished; such is, however, present, and can, though not at all easily, be recognised.^ For this purpose g^-~. ^ it is necessary to fix the object with picric acid, >-x ^^ /t^ in the way given for Cladophora, and then to \|^ (g) ^ stain with Heematin-ammonia. We then find in ^ , 11 ,-. , TT 111 Fig. 79. Saccharo- each cell near the centre a small, round, darker- myces cereuisup. i, not stained nucleus. — The living object, which we budding; 2 and 3, , 1 1 X- 1 n budding cells ( X 5iO). nave under observation, shows us numerous cells in course of multiplication. This takes place here in a quite cha- racteristic and peculiar fashion, by the cells forming one, seldom several, small, knob-like swellings, which gradually attain the size and form of the mother-cell, and then can be separated from it (2, 3). In very energetic development we find the daughter- cells united into small occasionally branched chains ; in slower development, separation of the cells takes place before any new one begins to form. This is multiplication by budding, peculiar to the Sa^ccharomycetes. — In sugar-containing fluids it induces alco- holic fermentation. Recently,^ the individuality of the Saccharo- mycetes has been questioned, and they have been declared to be conidia (spores of a kind) of different fungi, conidia which hav-e the power, in a suitable nutrient fluid, of multiplying by budding in endless sequence. We will now turn our attention to one of the Xostocacea', of 216 NOSTOO. interest to us on account of its symbiotic * relations with another plant. This last plant, widely cultivated in botanical gardens, is Azolla caroliniana. As the Azolla winters in plant-houses, we are therefore in a position to obtain material at any time for the investigation of the Nostocacese. The I^ostocacefB are, in general, specially disposed towards symbiosis, and we find them in very various plants, especially, however, as constituents of the body of lichens. — The Anahmna Azollce, living in the Azolla, is found in definite parts of the plant in question. The leaves of Azolla are each two-lobed. The upper lobe is fleshy, and floats on the water; the under is membranous, and im- mersed. The upper lobe shows in the interior a broad hollow, into which a narrow opening, found on the inner surface of the leaf, leads. This cavity is filled with Anahcena, and from the walls of the hollow also grow branched hairs between the coils of this Anabcena. In order now to obtain the Anahcena for our ex- amination, we pull the upper lobes of some leaves to pieces with the needles, lay on a cover-glass, press upon this a little, and are now pretty sure to find the Anabcena strings. This much is certain, that no specimen of Azolla is devoid of them. We examine the strings with our highest possible power (Fig. 80). These consist of a row of barrel-shaped cells, which from time to time are interrupted by a larger, ellipsoid or globular cell, the limiting cell, or heterocyst. The threads are serpentine, coiled here and there, without any visible gelatine coat. The entire content of the vegetative cells is coloured verdigris-green, of the limiting cells is olive green ; small, darker-looking granules are distinguish- able in these contents ; nucleus is wanting. Individual cells are usually found in division (Fig. 80, ato d). If a twig of Azolla is taken between the fingers, and surface-sections taken from it, not infrequently the Anahcena can be seen under the microscope in its natural position inside a leaf -cavity. It must, however, have happened by chance, that a leaf-cavity has been cut in the proper Fig. 80. — Anabcena Azollce, a to d, successive stages in the division of vegetative cells; 7i, a limiting cell, or hetero- cyst ( X 510 ). * Symbiotic, from the substantive symbioai!^, implying the co'-existence in more or less mutual iuterdependence, of two different organisms. [Ed.] OSCILLARTA. 217 direction. This however frequently occurs ; then we see also the segmented hairs which permeate the Anahcvna. Quite similar is the structure of the threads in the olive-green lobed gelatinous masses, sometimes found in great masses on paths, and which belong to Nostoc ciniflonum, Tournefort (commime^ Vauch).7 In examining any terrestrial form of Vaucheria, particularly that collected from flower-pots, we meet with Oscillaria, likewise be- longing to the Schizophyta (splitting plants), in closest affinity to the Nostocace£8. They are found, moreover, almost everywhere in standing water, on muddy ground, and under similar conditions. Their presence is often betrayed by an unpleasant muddy smell. Caltivated in vessels, they creep in part to the walls of this, over the surface of the water. They are nearly straight, ^ yi or even coiled threads, coloured from blue-green, verdigris - green, olive- green to brown ; can, however, be colourless, and in many forms dis- tinguished by active mo- tility. The threads are free, or enclosed in a gelatinous sheath. They can be inserted indivi- dually, or in numbers, in such a sheath. The sheaths arise from the outer layers of the mem- brane of the threads ; sheaths are wanting. I Fig. 81. — A, Oscillaria princeps ; B, Oscillaria Froc lichii; a, ends of the threads ; b, piece from the middle of the thread ; in B, h, the grsnules collected against the partition walls; in ^, c is a dead cell between two livinpr ones. ; where these layers become fluid, the The threads are divided by cross partition walls into a multitude of similar short cells. The partition walls in some species can be seen very easily, in others with great difficulty. With the exception of this difference, there is great uniformity in the structure of these organisms. The cell-contents are, in general, coloured throughout the entire mass ; no nucleus can be recognised in the interior, but numerous small granules. The granules are either distributed through the entire cell -con- tents, or are specially collected at the partition walls. It matters not what species is chosen for examination, but preference should 218 MOVEMENT OF OSCILLARIA. be given to the thicker forms, -with clearer partition walls, as represented in Fig. 81. The phenomena of movement, as we must have noticed from the very beginning of our observation of the Oscillarise, are very- interesting. Especially in the thicker forms, wdth somewhat bent end and distinct granules, and with a sufficiently strong power, we shall be able accurately to study the phenomenon. We then determine that wdth the movement of the thread is combined a slow rotation on its axis. Simultaneously the thread shows irregular flexions, or nutations, which are the expression of ex- isting differences in the intensity of growth on its different sides. These flexions usually take place slowly ; can, however, induce violent movements when the flexion is stopped by some obstacle, and then by overcoming this, the tension is suddenly equalized. The Oscillaria-threads move now forwards, now backwards. The movements can only take place when the threads have a point of support on some other object. The straight threads move like those which are bent ; in these latter the phenomenon is, however, especially striking, and at once visible, while in the straight threads it is necessary to fix the attention upon the individual gran- ules of the surface, in order to de- monstrate a rotation of the thread on its axis. The origin of the movement is not yet known with certainty; it has recently been maintained that it is occasioned by protoplasmic processes [whether pseudopodia or cilia], w^hich pass through the membrane to the exterior.^ To the same class of organisms as the Nostocacese and the Oscillariae, belong the still simpler-constructed Chroococcaceee, which we will study upon one of the wddely- distributed species of Gleocapsa. We choose G. polyde7-'matica* (Fig. 82), growing upon damp walls or rocks, recognisable from their dirty green to olive colour, and their firm, clearly and repeatedly layered, gelatinous envelopes. Other species, with less beautifully laminate Fig. 82. — Gleocapsa polyderniatica. In A, at the commencement of divi- sion ; in B, to the left, shortly alter division (x 540). * More readily obtainable, and very like, is G. caldariorum, a species grow- ing commonly on the walls, flower-pots, and glass, etc., in conservatories and greenhouses. [Ed.] GLEOCAPSA. 219 gelatinous envelope, will serve the same end. In all of tliem we find in the gelatinous envelope uniformly coloured cells, more or less clearly granular, and devoid of nucleus. By these peculiari- ties of their cell-body the Chroococcacese are distinguished from Protoeoccacece and especially Palmellacco3, which in many forms very strongly resemble them, but which have a nucleus, and chro- matophores, separated from the rest of the protoplasmic body. In Gleocapsa polydermatica the cell-bodies arisen from just previous division are quite globular (Fig. 82, C). They then begin to grow in length, and become ellipsoidal. They then show a weak hour-glass-like constriction (A) in mid-length, after which a delicate partition wall becomes visible at this place. The daughter-cells now round off towards one another, and, by swell- ing of the separating wall, and the thickening layers afterwards developed, become thrust back from one another. Owing to the ever-new development of gelatinous layers in the interior, the older ones become stretched, finally ruptured and cast off ^ [often parts of such layers are found, while the rest have disappeared]. A considerable number of generations is therefore combined by the gelatinous envelopes into a single cell-family [or colony, whence often called colonial algae]. By rupture of the outer envelopes the families fall apart. An isolated cell is rarely found, and then is usually surrounded by a considerable number of gelatinous envelopes (Fig. 82, A). In such cases the cell- division is discontinued, not the thickening of the wall. We have therefore found that in Nostocacete, Oscillarice, and Chroococcacese, the cell-contents differ from those of the plants hitherto considered by us : while in these latter we find the sepa- ration of the protoplasm into cell-plasma, nucleus, and chroma- tophores, we find here all these elements of the cell -body still united into a single substance. ^^ Distinguished by their coloration from the pure green of other plants, these plants have been collected together under the name of Phycochromaceas, or Cyano- phycea). The simplicity of organization of these organisms is betrayed also by the absence of sexual multiplication. One kind of asexual multiplication is, however (often by the side of other kinds of asexual maltiplication), quite peculiar to them, viz., that by vegetative bipartition ; and therefore these organisms have been called segmenting or splitting alga?, or Sehizophyccje.^^ Recent researches 12 suggest that the thread-like Schizophyceoe are capable of separating into globular cells surrounded by 220 GLEOCAPSA. gelatinous layers ; i.e., of entering into a chroococcaceous condition, like to Gleocapsa. An analogous phenomenon, found amongst the green Algse in the case of Protococcaceee,- gave rise to the question whether Protococcus viridis was an independent organism. This question, therefore, repeats itself with the Chroococcaceae, which are perhaps merely developmental stages of the thread-like Schizophyta. NOTES TO CHAPTER XX. ^ Compare Pfitzner, in Hanstein's Bot. Ahhand. Bel. I., Heft, II., p. 40, and Scheuk's Handbuch der Botanik, Bd. II., p. 410. In the former the litera- ture is given. 2 Miliarakis, Die Verkieselung, Wiirzburg, 1884. ' Compare especially Cienkowski, Botan. Zeitung, 1876, Col. 17, and Melang. biol. de St. Petershourg., torn. IX., p. 531. "* Eeess, Alcoholgahningspihe, 1870. 5 Schmitz, Stzber. d. niederrh. GeselL, 4th Aug., 1879. ^ Breield, Botan. Unters. ilber Hefepilze ; der Schimmelpilze, V. Heft, 1883, p. 178. ' Compare Thiiret et Bornet, Notes algologiques, II., p. 102. ^ Engelmann, Botan. Zeitung., 1879, Col. 49. 9 Schmitz, Stzber. d.niederr. GeselL, 6th Dec, 1880. 1" Schmitz, Die Chromatophoren der Algen, p. 9. 'I Compare, for example, Falkenberg in Schenk's Handbuch der Botanik, Bd. II., p. 304. 12 Zopf, Bot. CentralbLfBd. X., p. 32; Zur Morplwlogie der Spaltpfianzen, 1882. BACTERIA. 22 1 CHAPTER XXI. SCHIZOMYCETES * (BACTEKIA). USE OF IMMEKSION OBJECTIVES. Material Wanted. Some green leaves, e.g. those of the lettuce. Fresh. A carrot, turnip, or potato. Some peas, dry or green. Vaccine lymph, best in capillary glass tube. Hay. Beggiatoa alba, p. 232, can usually be obtained freely upon pieces of indiarubber tubing kept in water. [Ed.] (Other materials may be available in addition to these.) Let us now turn our attention to some examples from the group of the smallest known organisms, the Bacteria,^ in order to obtain some information as to the general form which they assume. We shall not endeavour, in the first place, to study any particular species ; we will rather leave it to the accident of what form hap- pens to be at our disposal. We boil some green leaves, say lettuce leaves, in a Florence flask, and leave it standing open at a com- paratively high temperature. Into another flask we place some peas — killed by steeping in boiling water — with a little water. At the same time we distribute disks of boiled carrot, turnip, and potato, on watch-glasses or object-slides, and place them about in warm, moderately moist places; some free, others covered with glass bell- jars. Upon the decoction [or infusion] of leaves after two days a skin may have been formed, which we wall call the pellicle. On the different vegetable disks we see small whitish, rarely coloured, masses of gelatinous substance appear. If we bring a trace of such a mass of jelly into a drop of water on an object- slide, and examine it with the strongest possible magnifica- tion, we find an enormous number of exceedingly minute bodies, appearing almost dot-like, imbedded in the jelly. These bodies show a necklace-like arrangement ; we find them singly, or in pairs, or united in large number into a thread. Embedded in the * Segmenting, splittiug, or fissing fungi. As a convenient term, not imply ing any individual kind, 1 shall use the ^Yord bacteriad. [Ed.] 222 BACTERIA. jelly, therefore, we have the Coccus-form of a bacteriad. If we wish to define the outer limits of the jelly, which, in its refractive relations, differs only little from water, we can readily accomplish it with the aid of Indian ink.- The ink must be of good quality, and should be ground down very carefully in water. A drop of this ink can then be placed npon the object- slide, the gelatinous mass which is to be invBstigated placed npon a cover glass, and the cover-glass then laid upon the drop. In this way the particles of ink are prevented from passing between the jelly and the cover- glass. The limits of the jelly are now sharply defined, on account of the surrounding fluid being filled with fine particles of ink, which exert no injurious influence on the object. Such masses of bacteria embedded in jelly are distinguished as ZOOglcea [or the zooglcea stage of the bacteriad]. The jelly arises from the swollen membranes of the bacteria. In the bacteria of putrefaction these membranes are composed of a peculiar albuminous substance, mycoprotein ; with bacteria not provoking putrefaction they con- sist of cellulose. We make use of the property of bacteria of eagerly taking up certain aniline and azotic colours, in order to stain them. For this purpose we only need to add a little methyl violet, gentiana violet, methyl blue, fuchsin, Bismark brown, or Vesuvin, to the preparation. Hasmatoxylin (logwood) at the same time colours the jelly ; and we therefore use this in order to make the jelly distinct.* We will confine ourselves at first to gentiana violet, which stains bacteria with extraordinary rapidity and intensity. We then see the bacteria very clearly, and can form an opinion as to their mode of multiplication, which takes place by successive bipartition [or fission]. This multiplication, in con- tradistinction to the " budding " of the yeast fungi, has given to the bacteria the name of " splitting or fissing fungi," or Schizo- mycetes. — It is quite possible that the jelly taken under our observation does not contain round " Cocci," but rodlets (compare Fig. 85 A, on p. 237). The rodlets can be identified as composed of shorter segments, which stand out very clearly if we add iodine- solution to the preparation. The segments now appear much shorter than they appeared in the fresh state ; partition walls are now shown which formerly were invisible. Some bacteria are distinguished by the fact that in the stages preceding spore-formation they form a starch-like substance in their body, and then, on the addition of iodine solution, colour blue or violet. BACTEraA. 223 In the pellicle which has formed upon the leaf-infusion (cf. Fig. 85, A, page 237) we have also a form of zoogloea, in which the cell-rows are held together by a jellj into a superficially de- veloped skin. This is seen to be traversed by fine, undulating partly parallel threads, formed of cocci or, more commonly, of rod- lets. The segmentation of the rodlets is again made specially clear by the addition of iodine-solution. In such material obtained by cultivation, the swarming stages of development will often be found. We are almost certain to find such in the water in which the peas have been soaking for one or two days. We then see the bacteria in dancing movement, now forwards now backwards, hurrying about in different directions. Exceedingly fine cilia have been repeatedly demonstrated as the cause of this movement (Fig. 85, B) ; in other cases these have not been found. If we examine the pellicle of a leaf -infusion which has already stood for some time, we shall find the rods or threads ultimately in course of formation of spores (Fig. 85, C). The contents of the rods have aggregated at one or several points, and produced rounded or elliptic highly-refractive structures, which appear as darker bodies, and are resting -spores. These persist, while the emptied membrane of the rodlets is finally disintegrated. In material from other cultures we shall quite as commonly find rod- lets which have formed but one resting-spore, at one end, and have thus taken the form of a pin or tadpole. Such forms, for example, are very frequent in the very widely-spread bacteria of butyric fermentartion {Clostridium hutyricum). As the Bacteria are the smallest of known organisms, for a more complete study of them the most powerful and best objectives, and the most favourable possible illumination, are alike necessary. As objectives, those for homogeneous immersion are especially to be re- commended ; while the most advantageous conditions of illumination can be attained by the aid of an achromatic condcnsor, or ap- paratus such as Abbe's " Beleuchtungs-Apparat." In most cases, however, ivater-immersion objectives will suffice. Objectives for water-immersion, as well as those for homogeneous immersion, can be used with any of the microscopes we have heretofore specified ; but a condensor, or other similar apparatus, can only be added to a microscope specially arranged for its use. They require, as was indicated in the Introduction, one of the larger microscope stands. If a water-immersion lens is employed, cover-glasses of a definite thickness, indicated by the optician (compare Introduction), must 224 IMMERSION OBJECTIVES. be used. If the objective is provided with a "correcting screw," the objective is regulated for any thickness of the cover-glass, so far as this lies within the possible range, by turning the correcting screw placed in the upper part of the objective. Every objective provided by an optician has usually a definite value to its screw for each hundredth of a millimetre thickness of the cover-glass. In order to use the objective, a small drop of distilled water is placed on its front lens. Care must be taken that this drop of water is not allowed to dry ; between the cover-glass and objec- tive, however, it is so protected against evaporation that it will usually last for several hours. In moving the object-slide, care must also be taken that the drop of immersion-fluid does not go to the edge of the cover-glass, and so mix with the fluid in which the object is immersed. If this should happen, the objective should be at once cleaned, and the fluid on the cover-glass removed. In case an object, already covered with a cover-glass, is examined with the water-immersion objective, and the thickness of the cover-glass is not known, the correction, if necessary, must be arranged during the observation. While observing, we turn the ring towards the one and the other side, and compare the effects produced. As the correcting screw in almost all objectives is so arranged that the front lens remains immovable, and only the upper lenses of the objective are moved, the object remains, during the correction, approximately focussed. The correction is complete when the figure appears sharpest. Objectives for " homogeneous " immersion have no correcting screw, and the thickness of the cover-glass, within the permissible limits, matters nothing. For these a drop of the immersion fluid provided by the optician (oil of cedar- wood, or a mixture of oil of fennel and castor oil, or zinc-iodide glycerine) is placed upon the front lens of the objective. We take the smallest possible quan- tity of the immersion fluid ; as this does not evaporate, it need not be replenished during the observation. As with the water-immer- sion, we must take care that in moving the object-slide the immersion fluid does not run to the edge of the cover-glass. For cleansing the objective after use, a very clean and of ten- washed piece of linen is best. In order to clean the cover-glass, we use a piece of linen moistened with chloroform. As objectives for homogeneous immersion bear a change of eye-pieces very well, we can procure a complete series of these. In case the observer has a larger microscope stand, to which [as LARGER MICROSCOPE STAND. Fig. 83.— St! 225 , ^ ^^ ~-' -. ^ttuiUi/o, with Abbe's "BclLULhLunt deusor; d, DiAphra^^-m-boaier , t, Screw tothis; s, Doable mirror. Con- 226 LARGER MICROSCOPE STAND. e.g. in that illustrated in Fig. 83*, a "condensor" can be, or], as in Fig. 83, an Abbe's " Beleuclitungs-Apparat," is fitted, this acces- sory apparatus can now be brought into use. In order to fix Abbe's apparatus, the body of the microscope must be sloped (even more than in Fig. 83), remove the ordinary mirror, and slip the apparatus in its place in the same groove. The apparatus is con- structed in one piece, and consists of condensor (c), diaphragm- bearer (d)^ and double mirror (s). The apparatus is pushed so far up that the upper surface of the condensor lies only a little below the upper surface of the stage (as is shown in the figure). The ap- paratus is then fixed in the groove with a screw found just above the mirror. Of the two mirrors the plane one is, as a rule, used with the apparatus. The concave mirror should be used only with very weak objectives, if the plane mirror does not illuminate the whole field of view quite equally. With the exception of a special case of bacteria investigation, which we shall speak of directly, we ought not to use Abbe's apparatus without a diaphragm. The narrowest diaphragm which gives sufficient brightness, is in all cases the best. In order to bring into use the diaphragm disks provided with the instrument, we turn the diaphragm-carrier (cZ), which is found under the condensor, away to the right hand from under the stage, introduce a diaphragm-disk, and then return it into its place. The screw (i) on the diaphragm- carrier serves to remove the diaphragms out of the central position, and then, as the diaphragm-carrier also turns in its sheath, we can rotate them about the axis of the microscope. We can thus obtain oblique illumination; to which, however, we shall seldom have recourse. Abbe's " Beleuchtungs-Apparat " is so convenient in use, and offers such great advantages, especially for difficult investigations, that it cannot be too highly recommended. If we are in posses- sion of a suitable stand with such an apparatus, it should be used for all investigations. Abbe's apparatus can be used also with advantage with w^eak objectives, and, by changing and moving the diaphragms, permits all gradations and modifications of the illu- mination. [The apparatus as above is constructed by Zeiss.] [This microscope of Zeiss (Fig. 83), is not, however, much in use outside Germany ; nor is Abbe's apparatus suited, on its present lines of construction, for use with English microscopes. — As the English student will probably purchase a microscope of home manufacture it is desirable to state here that the larger, and LAEGER MICROSCOPE STAND. 227 typically English, stands are not to be recommended for student use. Their length of body makes it exceedingly difficult to use them upright without a special table ; and the upright position is, all round, the more convenient for student work. Nor are me- chanical appliances for moving the object-slide about on the stage of utility commensurate with their cost, and the want of inde- pendence which they induce. Most of the English makers manu- facture microscopes with tubes of about the Continental length, but of better workmanship than the ordinary Student stands, and suited for the addition of accessory illuminating and other appliances. Without desiring to assert or imply its absolute superiority, where at least three or four excellent instruments exist, the " Student's Stand " of Ross (monocular) may be looked upon as typical of instruments of this class, in which workmanship of the highest excellence is combined with simplicity and moderate jDrice. It is illustrated in the accompanying Fig. 83*. The model shows peculiar lightness and stability. The body is swung between two pillars, so that it can be removed from the vertical to any extent in the backward direction, and can even be thrown a little beyond the vertical in the forward direction. It has coarse rack- work adjustment of considerable delicacy, and a fine adjustment. The eyepieces are of Continental size. The stage rotates upon a centre in the optic axis of the instrument. The mirror slides up and down in a groove upon a "swinging tail-piece " (Zentmayer's) which replaces the jointed arm of most mirrors, and enables rays of any degree of obliquity to be thrown upon the object ; and the mirror can even be swung round above the stage so as to illu- minate an opaque object from above. The same tail-piece carries a cylindrical bearer (seen in the figure between mirror and stage), likewise sliding up and down in the same groove with the mirror, in which can be placed a condenser or other accessory apparatus. If the student should not wish to purchase a special condenser, a small adaptor can be obtained, by means of which his one-inch, or other low-power, objective can be screwed into the carrier, front lens uppermost, and then, after sliding in the groove till its proper focal distance from the object, as ascertained by experiment, is attained, this objective will serve as a thoroughly good condenser. As the condenser is never required excepting with high powers, the low-power objective will always be at liberty for this purpose. In place of the diaphragm wheel, as usually supplied, an " Iris- diaphragm," which, by simple movement of a lever arm, gives a 228 LARGER MICROSCOPE STAND. Fio. 83 ♦.—Student's monocular stand of Ross, with concentric rotating glass stage, one eye-piece, etc., as in above figure, but no objective. Price, £8 18s. 6d. About | actual size- GENERAL METHODS OF WORKING. 229 nearly circular aperture of any chosen size, can be added at an extra cost of about £1.] In order to be able to work with the microscope in dull weather, or in the evening in general, it is an advantage to have a lamp with a wide burner, and to shade between this and the mirror of the microscope by means of a plain glass water-bottle, or the largest size possible of Florence-flask, filled with a very dilute solution of ammonio-cupric oxide. Microscoping in the evening exhausts the eyes very little, provided care be taken that the surroundings are illuminated just as brightly as the field of view of the microscope. Small incandescent electric lamps have been recently recom- mended as sources of light ; a current from about three Bunsen elements, each about eight inches high, suffices. It is best and simplest to place the incandescent lamp in front of the micro- scope, and between this and the mirror of the instrument to place a globe (as above) filled with very dilute ammonio-cupric oxide. The comparative richness of the incandescent light in rays of short wave length, although far less marked than that of the arc light, makes it very suitable for the study of delicate structural relations! As already noted, methyl- violet, gentiana-violet, methyl-blue, fuchsin, Bismark-brown, and Vesuvin, are especially serviceable for staining bacteriads. These stains are best used in watery solution, which must be either fresh, or at least freshly filtered. For this purpose we keep a saturated alcoholic solution of these colours ready, and add one of them drop by drop to a large quantity of distilled water. Bismark-brown and Vesuvin, how- ever, as they are altered in alcohol, must be kept in watery solu- tion, and this must be filtered before each use. The bacteria found in a fluid medium should be spread in the thinnest possible layer on the cover-glass, and allowed to dry at the temperature of the room. If the fluid contains albuminous bodies, or mucilage, these must, after the preparation is completely dried, be fixed either by laying the cover-glass for several days in absolute alcohol, or, still simpler, by a higher temperature. For this purpose w^e pass the cover-glass pretty quickly several times through the gas or spirit flame, during which the surface covered by the bacteria should be turned upwards. We stain it by spreading over the cover-glass, prepared in this or any way, but which in all cases must be dry, a drop of colouring fluid, and allow it to act for from five to ten minutes. Or we stain it in a saucer, which contains a larger 230 GENERAL METHODS OF WORKING. quantity of the colouring fluid, on which the cover-glass is allowed to float for from ten to thirty minutes. Warming the fluid from 30° to 60° C. (86° to 140° F.), hastens the operation. After com- plete staining, the cover- glass is w^ashed in distilled water, dried at the temperature of the room, a drop of oil of turpentine, xylol, or cedar oil placed upon it, and the examination carried on in this. If the preparation is to be preserved, remove the oil with blotting paper and imbed in dammar or in Canada balsam, which, however, must be dissolved in turpentine and not in chloroform. Care should be taken, in case the preparation should be later on ex- amined with a homogeneous immersion objective, that the dammar or Canada balsam does not come out from the edge of the cover- glass, since both of these are soluble in the immersion oil, and the whole cover-glass would thus be soiled. This inconvenience can also be obviated by making at the edge of the cover-glass, after the dammar or Canada balsam has become dry, a border of black varnish, or of gold-size. To do this, a small camel-hair brush is used, and care is taken that the varnish does not run over the cover-glass more than is necessary. A concentrated solution of pure dry iodide of zinc in pure glycerme can also be used as a fluid for immersion objectives. This, after filtering, if necessary, is evaporated in a water-bath to the refractive index 1-518 (for line D of spectrum). This fluid does not attack balsam in settings of the cover-glasses, and has the further advantage that it can easily be washed off from the cover-glass with water.^ Preparations stained with Bismark- brown or with Vesuvin retain their colour also in glycerine, and can therefore be preserved in it. The closure of the edges of the cover-glass can be effected with Canada-balsam in chloroform. After some days or weeks, just as is convenient, the Canada balsam can be covered with a layer of gold-size or of varnish, not in order to prevent the immersion-oil from attacking the closing cement, but because it is recommended especially as very durable. [Canada balsam in chloroform is apt, when dry, to spring. Hence the advantage of turpentine, since the fat is left behind in drying, and the balsam never becomes brittle. There is a concurrent dis- advantage, however, that the slides must not be pressed one no another for any length of time, or they may stick together. Nor must the proportion of fat in the balsam be allowed, by adding repeatedly more turpentine to dilute it in its bottle, to become considerable. Covering with gold-size is advisable.] MICROCOCCUS AND SPIROCHETE. 231 If one of the larger forms of bacteria is under examination, we may also, with the aid of our strongest objective and the most successful staining, learn something about the content of the cells. This appears to be a homogeneous plasma, in which can be embedded finer or coarser granules, which probably consist of fat. Nuclei cannot be identified even in the largest forms. Only in rare cases are the bodies of the bacteria coloured while in living condition. We will now make use of the experience we have accumulated in order to study an exceedingly small Coccus species, namely Micro- coccus Vaccinae (Cohn), the globular bacteria of vaccine lymph.^ If we place some fresh vaccine lymph upon a cover-glass, allow it to dry, and then stain it with gentiana violet, it will be possible for us to make out the small, round, darkly stained cocci, singly or joined in j^airs; though even with strong magnification they are only dot-like. Fresh lymph under a cover-glass, and protected from drying, kept for some hours at a high room-temperature, or, better still, at 36° C. (about 96° F.), in a warm chamber, shows sooner or later bead-like threads, or after a longer time, complete masses of cocci. Such masses can be seen at once in a lymph which has been preserved in the capillary glass tubes, in which these masses can be seen even with the naked eye as small flecks. It is these cocci which, by inoculation [vaccination] , are introduced into the human body, then multiply, produce the so-called cow- pox, and, for unknown reasons, give the body immunity against small-pox. If we have at command water in which j^ y^- algae, especially Spirogyra and Vauclieria, are J , k decaying, and examine a little of this fluid, V > f we shall find in it, almost to a certainty, I < ^ motile, exceedingly fine spiral threads (Fig. J J 84). These flexible, corkscrew-like threads J^ ) move rapidly in the water. They turn on their axis, and at the same time bend to and fro. Individuals suddenly stand still, then Fig. si.—SpirociKstc pU- hasten on again. The spirals found under ^attii.. after aniline stam- ° _ •■■ i2g, partially showing the such circumstances in all probability belong segmentation into rootlets to Spirochcete plicatilis, the Spiroclicvte of (^^^^)- marshes. If these spirals are allowed to dry, and stained, we shall see that they are not unicellular, but consist of successive segments, which may vary in length according to circumstances. 232 BEGGIATOA AND LEPTOTHRIX. On the same decomposing algfe, or on pieces of other decompos- ing aquatic plants, or other similar substratum, we commonly see growing fine threads, which are Beggiatoa alba (Vauch).^ These bacteriads are especially widely diffused in water which receives the refuse from factories, and in sulphur springs. They then often cover the masses of mud on the bottom and sides with a dirty- white layer. They are amongst the largest known bacteria, and can be distinguished with even comparatively low magnifi- cation. The threads are of variable thickness (from O'OOl to 0'005 mm.), are attached or free, the free however only being part of those attached. A segmentation of the threads into shorter or longer rodlets is more or less distinct ; the cell-contents are usually distinguished by a greater or less number of strongly refractive grains. If we allow the preparation to dry, and run in bisulphide of carbon, the grains are dissolved ; they consist of sulphur. In threads very rich in sulphur the segmentation is indistinct, and only appears after aniline staining, or after heating in glycerine or in sulphite of soda. By the glycerine the grains are partly, by the sulphite of soda completely, dissolved. By cross segmentation the threads can separate into cocci, and it has been observed that, in thicker threads, even a quadripartition of the cells can result from this cross division. Moreover, "swarming" cocci, rodlets, and spirals have been observed as developmental stages of Beggiatoa. The attached threads can, in their upper parts, be spirally bent. The straight as well as the spiral frag- ments of threads are flexible, and show creeping movements. These characters serve to connect Beggiatoa with the Oscillatorians. Beggiatoa separates out the sulphur compound from the water in which it dwells, and thus sets free a more or less considerable quantity of sulphuretted hydrogen. We will now turn our attention to another object, which shows the coccus, rodlet, and spiral, and at the same time the thread form, combined. For this purpose we will use the white "fur" of teeth. If a little is diffused in a drop of water, and examined with the strongest possible magnification, we shall see long, ap- parently unsegmented threads, rodlets of various lengths, spii^al Spirochcete, and also minute crowded cocci. It has recently been determined 7 that all these forms are developmental stages of the same segmenting fungus, Leptothrix buccalis, Robin. This lives as a saprophyte upon the mucous membrane and in the "fur" of teeth J it can, however, under certain conditions, become parasitic, CLASSIFICATION OF BACTERIA. 233 penetrate into the tissue of the tooth, and induce the " decay." If the preparation is treated with iodine solution, the long threads also prove to be composed of longer or shorter rodlets. The aggre- gated cocci are clearly differentiated into their individual elements. These cocci are never wanting, though it is doubtful whether they always belong to Leptothrix itself. Recent researches have, as a whole, tended to show that the genera and species hitherto distinguished by their outer form, as Micrococcus, Bacterium, Bacillus, Vibrio, Spirillum, Spirochete, etc.,^ may enter into the developmental cycle of one and the same species.9 Hence we now employ these terms only to distinguish a given form of growth, and speak oi:— cocci, globular or ellipsoid structures; rodlets, threads, and spirals. The shorter rodlets are distinguished as Bacteria from the longer Bacillus; the simple threads as Leptothrix, the branched Cladothrix ; the spirals with comparatively wide turns and greater thickness of the threads are called Spirillum; or, if they contain sulphur, Ophidomonas; spirals with elongated turns, Vibrio ; very thin spirals, with small diame- ter and also smaller distance of the turns, Spirochcete ; riband- like tapering spirals, Spiromonas ; flexible spirals, whose ends coil back towards one another, Spirulina}^ We have seen in the study of the segmenting algje that they also are distinguished by like variability of form in the different stages of development; and the comparison of the bacteriads with these segmenting algjs leads, in fact, to the presumption of some close relationship between these organisms. In the segmenting algfB also we have made the acquaintance of cocci, rodlets, threads', and spiral forms. Moreover, we have met with the phenomena of movement amongst them, and even in their ability to resist high temperatures the segmenting algce approach the segmenting fungi. The first plants which show themselves in hot springs are seg- menting algas, though it is true they do not resist so high tempera- tures as, e.g., the spores of the bacteria of hay, whose capabilities for germination temporary boiling seems only to heighten. More- over, in the structure of their cell-body, segmenting algjB resemble segmenting fungi, since both groups are devoid of nucleus and of separated chromatophores. To this we may add the mode of vegetative multiplication, which gives to the two sections their respective names. For these reasons we can consider the segment- ing fungi to be a colourless section of the segmenting algjp^ or, at any rate, one devoid of a colour which enables carbon-assimilation, 234 BACILLUS TUBERCULOSIS. and wliicli, together with the segmenting algas, form the class of segmenting plants, the Schizophyta. Bacillus tuberculosis, of recent times considered to be the cause of tuberculosis in the sputum of consumptives,^^ is always motion- less, very minute, somewhat tapering at its ends, and now and then with four to six grains, which are considered to be spores, in the interior. This Bacillus is distinguished by special relations towards staining reagents, which render it possible to distinguish it from other Bacilli. The substance to be tested is spread as flat as possible upon a cover-glass, and allowed to dry at the tem- perature of the room. The albumen present is then fixed by passing the slide bearing the cover-glass three or four times through a spirit or gas flame, the preparation side turned up- wards. We then saturate a quantity of water with aniline by shaking up the water with an overplus of this body. We filter it through paper previously damped with distilled water, and add to 100 parts (by measure) of the fluid, drop by drop, 11 parts of a saturated alcoholic solution of fuchsine or of methyl violet, and then 10 parts of absolute alcohol. This staining fluid can be pre- served for at least ten days in a well-closed glass, without its being necessary to filter each time it is used. The cover-glass is now allowed to float for half a day in this fluid. The staining proceeds more quickly if the solution is warmed till evolution of bubbles commences. The action need then only last ten minutes. After this the coverrglass is laid for, at the most, a half -minute in a solution of 1 part nitric acid to 3 or 4 parts distilled water, and then for some minutes in 60 per cent, alcohol. The entire pre- paration is thus coloured, with the exception of the tubercle bacilli, if any such are present. The preparation is then observed in water ; or it can be washed in water, allowed to dry, and after- wards mounted in Canada-balsam dissolved in turpentine. — The material for preparation of sections must be well hardened in absolute alcohol, or, if hardened in other ways, must lie a long time in alcohol. The sections are then stained in the same way as above described. They must remain in the staining fluid at least twelve hours. After being passed through 60 per cent, alcohol, they can be placed for some minutes in dilute watery solution of Vesuvin or methyl-blue. They are then once more washed in 60 per cent, alcohol, passed from thence into absolute alcohol, in order that, when completely deprived of water, they may be placed in oil of cedar (which does not extract the aniline STAINING AND DOUBLE-STAINING. 235 colours from the preparation), in which they can be examined. In order to preserve them, the preparations are mounted in Canada- balsam dissolved in turpentine.^^ Tubercle bacilli are beautifully visible with magnification of 300 diameters. — These bacilli are likewise stained with great intensity by fuchsin prepared in the following manner : — In 100 gram of a 5 per cent, watery solution of carbolic acid is dissolved 1 gram fuchsin, and then 10 grams alcohol added. Filter. The solution keeps well. It is advisable to warm the fluid in using it.^^* — For bacteria found in fluids double-staining has also been employed. According to one of these methods,^^ the fluid diffused on the cover-glass is dried, and fixed with osmic acid vapour, or with a 0'5 per cent, solution of chromic acid. It is then washed with distilled water, and stained, usually for from half an hour to an hour, with 0*001 per cent, aniline green. It is again w^ashed for from twenty-four to forty minutes with distilled, weakly acidulated, water, in order to de- colorize the elements of the tissue. After once more washino- in distilled water, the preparation is placed for some minutes in a weak solution of picro-carmine. After being once more washed, the preparation is dehydrated by absolute alcohol, or simply by drying, and finally, if necessary, is cleared with oil of cloves and put up in Canada-balsam. In order to study bacteriads in the interior of the tissues, it is best to harden the tissues by placing them for at least from one to two days in absolute, or at least 90^ to 95° alcohol. For staining the bacteriads the colours already known to us come into use. In preparations stained Avith gentiana- violet or with methyl- violet, the tissues are comi^letely decolorized with strong alcohol in which is a trace of potash, while the bacteriads retain the colour. A like effect can be attained by laying the preparation for at most a half-minute in picric acid, whereby the tissue takes at once a yellow coloration. After decolorizing the tissue in alcohol this can be again stained with iodine-green, methyl-green, eosin, magdala, acid-fuchsine, and other stains, which are not taken up by the bacteriads. ^^ Good double-staining is also attained by gentiana- violet and picro-carmine.^^ The best means for stain- ing bacteriads in the interior of the tissues is, however, usually a solution of gentiana-violet in aniline water, and a solution of potassium-iodide iodine. ^^ The aniline water is prepared in the way given upon page 234, and dry gentiana-violet is dissolved in it to saturation, or 5 parts of a saturated alcoholic solution of 236 CULTURE AND LIFE-HISTOEY gentian a- violet are added to 100 parts of this water. This is filtered before each time of use. The solution can be kept for months. The sections are transferred from absolute alcohol, for some minutes, into the staining fluid, then for from one to three minutes into dilute jDotassium-iodide iodine solution (1 part iodine, 2 parts potassium- iodide, and 300 parts distilled water), then into absolute alcohol. In this the sections must be decolorized. They are then cleared in oil of cloves, and embedded in Canada-balsam dissolved in xylol. The tissue now appears decolorized, the bac- teriads stained dark blue. The hacilli 'of typhus, also the cocci in many cases of pneumonia, are decolorized by this process, and are distinguished thus from most other hacilli. Treatment for a very short time with a weak solution of Yesuvin before placing in oil of cloves gives beautiful double coloration, in that the tissue now appears stained brown. Instructive coloration is also obtained with safranin upon sections which were hardened in alcohol or in chromic acid.^'' Equal parts of a concentrated watery and a con- centrated alcoholic solution of safranin are mixed together, and the sections allowed to lie in this for half an hour, then washed a little in water, and for some minutes in absolute alcohol, trans- ferred to oil of turpentine, and put up in Canada-balsam. In order to find bacteriads in tissues, after they have been com- pletely stained. Abbe's apparatus [or other condensor] can be used with great advantage, and in a special fashion.^^ After focussing the preparation the diaphragm is completely removed, so that the cone of illumination filling the entire objective comes into use. Thereby the figures of all uncoloured parts, which are only distinguishable by differences in their refractive indices, more or less completely disappear, while the coloured light- absorbing bodies remain visible. We distinguish this method as " isolation of the coloured images." After we have thus made ourselves acquainted with the different developmental forms and methods of research, we will now point out the methods of culture, which come into use for breeding bacteriads ; we will breed a definite bacteriad, and follow out more- over its entire development. For this purpose we soak dry hay^^ in the smallest possible quantity of spring water, and let the infusion stand for four hours in a warm chamber at a constant temperature of 36° C. ( = about 96° F.). Then pour off the extract without filtering, and if it is too concentrated dilute it, so as to be more safe, to a specific gravity of 1*004. ^NTow place the fluid OF BACILLUS SUBTILIS. 237 in a flask holding at least 500 ccm. The flask is stopped with cotton-wool, and the fluid then boiled very gently for an hour. Then let the temperature sink to, and remain at 36° C. In the course of a day or a day and a half, a delicate grey skin will have formed on the surface of the fluid; this consists of the zooglcta stage of Bacillus suhtilis, the bacteriad of hay. We have made use of the power possessed by the spores of this bacteriad, of resisting boiling heat for a long time, in order to obtain a pure culture of them. Bacteriads in general are distinguished by their power of resisting high temperatures, the bacteriad of hay stands, however, foremost in this respect. Of the pellicle obtained as above we now transfer a little, with a suitable quantity of the fluid to an object-slide, and examine the object with the strongest magnification which we have at command. We find the pellicle Fig. S5.— Bacillus sxiltilis. A, the pellicle (x 500); Z?, " swarming" rodlets (x 1000); C, spore formation ( X 800). formed of long, segmented, wavy threads, running parallel to one another. The threads remain for the most part in their position, because they are held together by an invisible jelly (Fig. 85 A). The threads consist of cylindrical rodlets of various length, in general however, twice or thrice as long as broad. The substance of the threads appears homogeneous, colourless, pretty strongly refractive. Even with the strongest magnification no other structure is recognisable. With chlorzinc iodine the rodlets are stained throughout, and very clearly, a brownish yellow. The figures are thus obtained better than with the other solutions of iodine. The segments of the threads appear thereby in general shorter than in the fresh state, because now all the limits are 238 CULTURE AND LIFE-HISTORY visible. In order sharply to differentiate the rodlets, we can stain them, according to the methods already known to us, with fuchsine, methyl- violet, gentiana-violet, or Vesuvin, and then keep them as permanent preparations in Canada-balsam or in dammar. Picro-nigrosine can also be used with advantage for fixing and staining the preparation. If we focus upon a particular spot in the pellicle with a magni- fication of about 1000, we can observe the division (segmentation) of the rodlets direct.^o It is best to draw the piece of the thread in question at short intervals with the camera, and compare the drawings, so as to show the changes which have taken place. If abundant food-stuff is still in the fluid, the individual rodlets divide every half-an-hour to an hour and a half. The higher the temperature of the room, the more rapid the division. The rodlets increase in length without becoming thinner ; when they have attained, however, a definite size, a dark partition- wall appears across their middle. This process of division explains the arrangement of the rodlets and threads; it explains also the wavy course of the threads, which grow at all points by inter- calary growth, and if the ends cannot become further removed, the thread must become laterally contorted. For this reason, the whole pellicle shows a .wrinkling visible to the naked eye. We next transfer a fragment of the pellicle into a moist chamber, in order to examine it in a suspended drop. For this purpose we will use the simplest possible moist chamber, to wit a small frame of pasteboard. Such a frame is cut out of tolerably thick paste- board, its inner aperture being somewhat smaller than the size of the cover-glass we propose to use, while its outer diameter does not exceed the width of the object-slide. This frame is soaked in water till it is completely saturated, and then laid upon an object- slide.* On the middle of a cover-glass is placed a drop, spread flat, of the culture fluid, into which the object for investigation is transferred. The cover-glass is turned rapidly upside down, with the drop hanging below, and laid upon the pasteboard frame. If the observation is to be long and continuous, a drop of water is from time to time placed upon the frame, so that it shall not be- come dry. If the observation is interrupted, the preparation can * After and before using the pasteboard culture-cell it is desirable to place it for a few minutes in alcohol, so as to kill any organisms which may adhere to it, or the culture may be vitiated. This applies equally strongly to other cultures. [Ed.] OF BACILLUS SUBTILIS. 239 be protected from evaporation in a large moist cliamber. In order to again find a definite spot in the preparation, the object-slide must be brought back again into its original position, for which its outline can be drawn with a sharply pointed pencil upon the stage. It is still better in this and similar cases, to cut a cross on the stage by means of a sharp instrument, right and left of the central aperture. Then, when the object-slide is in the required position, similar crosses can be made upon it with one of the sharply pointed colour pencils mentioned in the Introduction [or with a writing diamond]. It is then easy later on to replace the object-slide thus marked in exactly the same position.* If the food materials of the drop are exhausted, the vegetative segmenta- tion or bipartition is arrested and the spore formation at once begins. After the lapse of from six to eight hours there can be seen in the threads, at thereabouts equal distances, ellipsoid, strongly refractive spores (Fig. 85 C). Elsewhere the threads appear empty ; only the colourless sheaths unite the spores. At some places in the preparation, one is certain to find spores still in course of formation. They appear in the form of collections of more refractive material situated most usually towards the middle of the rodlet. The aggregation becomes continually stronger, while the rodlet becomes emptied, and at last the for- mation of the spore is complete. If the culture is allowed to continue some hours longer, the sheaths of the rodlets will have become indistinct, and after the lapse of about a day the spores appear free, and sunk to the bottom of the drop. In contra- distinction to the rodlets, the spores hardly, or not at all, stain with gentiana violet and the other stains we have recommended, with the exception of the carbolized fuchsine and alcohol solu- tion given on page 235, which, especially when warmed, stains the spores very deeply. The spores germinate very easily if they are transferred to fresh nutrient fluid ; more slowly at the temperature of the room, quicker at 30° C. [ = 86° F.]. It is best to boil them for five minutes, and cool them slowly. Then in about two to three hours wc shall see the commencement of germination.2i The spore-membrane is opened on one side, the minute germ begins to protrude here, and elongates gradually * For the purpose of keeping a particular spot under observation for several successive days, ue., without removal from the stage of the microscope, one end of a few strands of loose wick can be inserted between the layers of the pasteboard, while the other end can dip into a vessel of distilled water. The water sucked up by the wick will keep the pasteboard moist. [Ed.] 240 CULTURE-METHODS. into a rodlet. Its hinder end remains inserted in tlie spore-case. About twelve hours elapse before the rodlet divides for the first time. In the meantime, the preparation will show all stages of germination. As a rule, the germinated rodlets at once set up movement, they enter into the " swarming" [or "roving"] stage. Such a swarming rodlet still carries about at its hinder end its spore-case. The number of the " swarmers " becomes by succes- sive divisions continually greater, and they fill the entire fluid before the beginning of the formation of the pellicle. Then the swarmers collect on the surface of the fluid, come to rest there, and produce the pellicle. The rodlets are of unequal length, and consist of a varying number of segments (Fig. 85 JB). Their movement is serpentine. We allow the fluid containing the swarmers to dry upon the cover-glass, and stain them then by one of the methods given upon pp. 234-5.22 The swarmers have a cilium at each end, the identification of which is not easy.-^ Breeding experiments with bacteriads are chiefly carried on in Florence or conical flasks, or in test-tubes. ^^ Many cultures are carried on direct upon the object-slide. Object-slides, vessels, and all the utensils to be used must be sterilized. This is effected by passing them quickly through a gas or spirit flame, or laying them before the beginning of the experiment in absolute alcohol, which quickly eva^Dorates after removal. The particular nutrient fluids for the cultures are boiled in the vessels which are to be used, and which must be closed with a cotton-wool stopper. In general it is desirable to boil the nutrient fluids for a short time on each of several successive days. In this way all the bacteriads which have in the meantime germinated, and which then bear, a high temperature far less than do the spores, are killed. It is assumed that after five days all the spores are killed. For greater certainty the nutrient fluid can be allowed to stand another day, before it is used for the inoculation; if it remain clear, it is assumed that it is sterilized. That boiling for an hour does not always serve for killing the spores, we can see in the culture of Bacillus subtilis. The infection of the cultures arises usually, not from the air, but from incompletely sterilized vessels. The danger of infection from, the air in the temporary opening of the vessels for the purpose of sowing (" inoculation ") is far less great than that wliich arises from incompletely sterilized vessel s.^^ To obtain pure material for inoculation in cultures on a large scale, various methods can be followed : — CULTURE-METHODS. 241 1. The Mrthod of Fractional Cttlturer^ This is based upon the experimental fact that of several kinds of bacteriads growing in the nutrient fluid, one ultimately gets the upper hand. If now from a culture which has progressed thus far a little is transferred into a second solution, free of fungi, and after a similar length of time from this into a third, and so on, there is a chance of ultimately obtaining a perfectly pure culture, that form always remaining last, which, under the given conditions, multiplies most quickly. 2. The Method of Dilution.-'^ When the bacteriad which is to be bred preponderates in the fluid, this method gives usually very good results. The fluid containing the bacteriad is diluted with water free of fungi, until by casual estimation only one bacteriad comes into a drop of the fluid. If the form to be bred distinctly preponderates, and a series of vessels containing the nutrient fluid are inoculated each with a drop of the fungus- containing solution, all the probabilities are, that in the majority of the vessels pure cultures will be obtained. [See note -'*.] Whether a culture in a nutrient fluid is pure, can in general be determined even macroscopically, by the fluid being uniformly turbid, or showing uniform formation of skin on the surface, or uniform formation of clouds at the bottom, or uniform coloration, or uniform formation of jelly. The purity of a culture is likewise assumed in which strong fermentation or intense putrescence results.-^ 3. Gelatine Culture.-^ This method gives the best results, and has led to the greatest progress in our knowledge of bacteria. In it the nutrient fluid is mixed with gelatine, with agar-agar,* or witli blood serum. Most commonly used is a mixture of infusion of pep- tone and gelatine in which the gelatine forms 5 per cent.. 50 grm. gelatine is soaked and boiled in 500 ccm. water. Half a kilo, of chopped meat is allowed to stand cold in 500 ccm. water for 24 hours, then the meat-infusion, obtained by pressing the meat, is boiled, filtered through fine gauze, and mixed with the gelatine ; add 10 grm. peptone and 1 grm. common salt, neutralize it with carbonate of potash, or with carbonate of soda or phosphate of soda, and filter through filter paper, f Put into a test-tube * Agar-agar, or Bengal isinglass, is a species of dried seaweed from Singajiore, consisting of small, colourless, transparent strips; is almost completely soluble in water, and forms a thick, tasteless, and odourless jelly. [Ed.J t One-tenth of these proportions would fully suffice. [Ed.] R 242 CULTURE-METHODS. 10 to 15 ccm. of this nutrient gelatine, close it with a plug of cotton wool, and sterilize it by continuous boiling for many hours, or better by boiling for half an hour on each of several successive days. In many cases it is recommended during the final cooling of the nutrient gelatine to bring the test-tube into a somewhat inclined position, whereby the free upper surface of the fluid is enlarged. According to need, the quantity of nutrient gelatine can be reduced to 25 per cent, or raised to 10 per cent. In similar manner to this meat-infusion peptone gelatine, can also be pre- pared hay-infusion gelatine, wheat-infusion gelatine, aqueous humour gelatine, meat-extract peptone gelatine, meat-infusion peptone gelatine, with 1 per cent, cane or grape sugar, etc. In case the culture should be maintained at incubation temperature, it is preferable to add agar-agar or blood serum to the nutrient fluids instead of gelatine. Such a nutrient basis remains solid at incubation temperature, while gelatine nutrient basis becomes fluid. It is prepared by adding 1 per cent, agar-agar to the nutrient fluid. The method of preparation of cooled blood-serum is more complicated. The blood of the killed animal is drawn im- mediately out of the wound into a pretty tall vessel provided with a glass stopper and previously sterilized. This vessel is filled light up to the brim, and placed for 24-30 hours in a refrigerator or ice- bath, until a copious layer of entirely transparent amber-yellow coloured serum is formed over the cake of blood. By means of a pipette a test-tube is filled Avith the serum, and stopped with a cotton-wool plug. The plug has been previously heated for an hour in a warm bath of 150° to 160° C, and so sterilized. The blood-serum should now be warmed in the open w^ater-bath, on five successive days, for one hour each day, to a temperature of 58° C. On the last of these days the temperature should be allowed to rise, from half to one hour, to 65° C, by which the blood-serum "sets." Sheep's serum sets the most quickly, calf serum the most slowly. The coagulated serum must be completely clear and pel- lucid; if it is not perfectly sterilized it becomes cloudy imme- diately.* It can be used by itself or be added as a "setting" constituent of the nutrient fluid. The solid nutrient basis can also be used with good results for microscope slide culture. While the sterilized nutrient gelatine, agar-agar, or blood-serum respect- ively are still fluid, a small quantity of either can be poured on a * All these culture fluids are to be had, ready prepared and sterilized, from Dr. Hermann Rohrbeck, in Berlin, Friedrichstrasse, 100. CJLTURE-METHODS. 243 sterilized object-slide, so that the layer of it after setting is about 2 mm. (y\ inch) thick. After the subsequent inoculation this object-slide is placed under a bell- jar closed with water, or in a case made out of plaster of Paris. A case made entirely of plaster of Paris, with a plaster of Paris cover, is very suitable as a moist chamber for the cultivation of fungi and bacteria, which do not need the light, because the moisture is very uniformly distributed in it, and no drops of water fall down from above on to the preparation.^^ Instead of inoculating on the object-slide this can be done with the nutrient gelatine while still in the test-tube, warmed up to about 25° C. ( = 77° F.), and so made fluid, with which the inoculating material is uniformly mixed, and which is then poured on the object-slide. If different organisms are repre- sented in the inoculating material, they now form on the object- slide separate colonies, each of which usually represents for itself a pure culture. The purity of individual colonies can be proved directly under the microscope ; and in that way pui'e material can be selected from them for future inoculations. The macroscopical appearance of the colonies is moreover often characteristic, and can lead to the recognition of forms which are otherwise difficult to distinguish microscopically. Thus the form of the colonies, their coloration, the circumstance whether the nutrient body be- comes fluid or not, whether the colonies themselves ultimately stain, serve as good indications of the nature of the bacteriads. The inoculation of a nutrient fluid, or of a solid nutrient body, is effected by means of a needle w^hich has just been heated red-hot, but is already quite cool, or with a platinum wire just heated. For this purpose the solid nutrient body on the object-slide is scratched [with the wire or needle after it has been dipped in the bacterial fluid] . If the solid nutrient body is in the interior of a test-tube, the needle or w^ire is stuck in to the depth of from |- to f inch. The mode of development in nutrient bodies inside the test-tube is also characteristic, and often, like to the characteristic signs in cultures on the object-slide, permits macroscopic dis- tinctions of separate forms. If it is desired to follow the processes of development of a form directly under the microscope, it is done with the aid of small moist chambers. For pure cultures lasting a longer time the pasteboard chamber previously referred to is not sufficient. For such purposes a chamber made out of a glass ring is recommended.^^ Such a glass ring, about \ inch thick (high), is broken off from a 244 CULTURE-METHODS. glass tube of suitable diameter. The glass ring is flattened on both of its ends (sides) on a whetstone, and fixed upon the object- slide with Canada-balsam. A round cover-glass of suitable size serves as a cover. The thinnest possible layer of the gelatine, agar-agar, or blood-sei-um nutrient body is placed in the middle of the cover-glass, and this layer is afterwards inoculated. The cover-glass is fixed on the glass chamber by means of oil drops run round its edge. A thin layer of water at the bottom of the glass chamber secures the necessary moisture. Such a moist chamber can be converted into a gas chamber when the glass ring has two lateral openings into which are melted or stuck glass tubes which serve for the introduction and removal of gases. Another moist chamber, likewise to be recommended,^- consists in an object-slide with a flat central round or quadrangular hollow, which is sur- rounded by another narrow groove (or channel) cut still more deeply. This groove is filled wdth water. The cover-glass used must be of suflicient size to completely cover up this external groove, and project all round on to the unhollowed portion of the object-slide. For cultures at constant and more elevated tempera- tures double-walled propagating chambers, with suitable warming appliances, will be needed.* NOTES TO CHAPTEE XXI. ^ For the statements here following compare Zopf, die Spaltpilze ; and de Bary, Vergl. Morph. il. Biol, der Pilze, Mycetoz. und Bacterien, p. 490 ; in both these works the general literature is given. For the staining methods I adhere chiefly to Hoyer, Gazeta lekarska, 1884. ^ According to Errera, Bull, de la Soc. Beige de Blicr., torn. X., No. 11. ^ Zeitschr. filr wiss. Mikros., Bd. I., p. 411. "^ According to Brun, taken from Fol, in Lehrt. d. vergl. mikr. Anat., p. 37. 5 Cohn, Beitr. d. Biol., Bd. I., pp. 16-17 ; Zopf, I.e., p. 92. ^ Engler, Bericht der Co7nmission zur Erf. d. deut. Meere, 1881 ; Zopf, die Spaltpilze, pp. 13, 75, et seq., where the other literature is quoted. 7 Here also compare Zopf, I.e., p. 80. ** Compare Cohn, Beitrdge zur Biologie, Bd. I., p. 125. 9 Compare the literature on this subject in Zopf, die Spaltpilze, 1883. ^° Zopf, I.e., p. 5. [See also W. B. Grove, Bacteria and Yeast Fungi, 1884] . '^ From E. Koch, Berliner klinische Woclienschrift, 1882, p. 221. * Such an apparatus can be obtained from Dr. Eobert Muencke, in Berlin, Louisenstrasse, 58, or from Dr. Hermann Eohrbeck, Berlin, Friedrichstrasse, 100, at an expense of from 25 to 50 M. The propagating chambers of d'Arsonval, which cost however of Dr. Muencke from 72 to 108 M., or of Dr. Eohrbeck 28 to 130 M., are to be specially recommended. NOTES. 215 ^- Compare on this point C. Friedljinder, Mikr. Tcchnih, II., edit., p. 53. '-* Neelson's method, according to Hoyer. 15 According to Soubbotine, Arch, de Phys. norm, et path., Tom. XIII., 1881, p. 477. 1* According to Hoj'er, I.e. J5 Weigert, Virchoiv's Archiv, Bd. LXXXIV., p. 201 ; Firket in Bizzozcro's French translation of the Manuel de micr. din., p. 314. 16 Gram, Fortschr. d. Med. 1884, p. 185. 17 Victor Babes, Archiv. fur jnikr. AnaL, Bd. XXII., pp. 359 and 361. 18 Introduced by R. Koch; Unters. uber Aet. d. Wundinjectionskrankheitcn, Leipzig, 1878. ^3 Accordhig to a method recommended by Roberts and Buchner ; compare Zopf, die S2)alt2)ilze, p. 57, to which work I have in general referred as a source for the other literature. 20 Compare Brefeld, SchmmelpUze, Heft IV., p. 38. ■■^1 Brefeld, I.e., p. 43. "2 Cf. Koch, in Cohu's Beitrage z. Biolog., Bd. II., p. 402. '3 Brefeld, I.e., p. 40. 2* Buchner, in Naegeli's Unters. iib. niedr. Pilzc, p. 192, where are representa- tions of the special forms of glasses used for culture experiments. 25 Buchner, Stzher. d. kongl. hair. Akad. der Wissensch., 1880, p. 381, and in Naegeli's Unters., as above, p. 159. 26 Employed by Klebs ; Archiv. f. exper. Path., Bd. I., p. 46; for the rest I have again had recourse to Zopf, Spaltpilze, pp. 43 ff. 27 From Naegeli, Stzher. d. kgl. hair. Ak. d. U'iss., 1880, p. 410, and Unters. ilber niedr. Pilze, p. 13 ; Buchner, Stzher. d. kgl. hair. Akad. d. Wissensch., 1880, p. 374, and in Naegeli's Unters. ilber niedr. Pilze, p. 146. [27* According to H. Fol and P. L. Duuant [Arch, des sc. p/i!/s. et nat. de Geneve, tom. XIII., 1885, p. 116), this same method can be employed to determine the number of Bacteria contained in a given quantity of a fluid. This fluid is diluted up to a certain point with sterilized water ; then a portion is taken and added to a known quantity of nutrient fluid, and this mixture is divided equally amongst a number of glasses. The number of glasses which remain sterile, compared with the total number, and with the quantity of fluid taken, enables us to calculate the number of Bacteria in the original fluid. If all the glasses contain Bacteria, the fluid employ td has not been sufficiently diluted.] 23 According to Zopf, Ic, p. 44. 29 Introduced by Brefeld ; compare Schimmclpiize, Heft I., p. 15. Completed by R. Koch. Compare R. Koch, Zur Untersuch. pathol. Organismen, extracted from Kaiserl. Gesundheitsamte, 1831, p. 18, and numerous other researches in the same. 30 Rainier, Annal. des Sc. Nat. Bot., Scries VI., tom. XV., p. 346. 31 According to Van Tieghera and Le Mouuier, Annates des Sciences Naturalcs, notanique, V. ser., tom. XVII., p. 263. 32 Dippel, Das Mikroskop, 2nd edit., p. 062 ; Grundziige dcr allg. Mikr. p. 295. [For the moi-phology of Bacteria see also W. B. Grove, A Synopsis of Bacteria and Yeast Fungi, 1884.] 246 REPRODUCTION OF ALG^. CHAPTER XXII. THE EEPRODUCTION OF A.LGM. Material Wanted. . Spirogyra in conjugation. Fresh. Cladophora glomerata, taken from quickly-flowing water. Fresh. Vauchena sessilis. Fresh. Also the terrestrial form of the same, with the sexual organs.* Now that we have obtained information in the general sphere of morphological investigation amongst higher as well as lower vegetable organisms, it will be our task to make ourselves ac- quainted with the most important of those problems "which the special morphology of microscopical investigation provides. In this we shall take just the opjDOsite way to that we have hitherto followed, and slowly mount from the simplest groups of organisms to those most highly organized. We have already made a commencement in our last chapter upon the Bacteria, to the entire cycle of development of which we have directed our attention. [In the whole of this life-cycle there was no indication of separate sexuality. The process of multiplication was vegetative, or asexual. Organisms of somewhat higher grade, however, show both of these processes, vegetative or asexual multij)lication, on the one hand of entire organisms, or on the other hand of the constituent cells of those organisms, and the commencement of the life-cycle of new individuals through pro- cesses of more or less complicated sexuality.] We will now continue with the examination of the asexual and sexual processes amongst the Algoe. Opportunity is not rare for examining the various species of Spirogyra in process of Conjugation.^ This is recognisable out of doors by the crinkled look [rather yellowish] and hangmg together * All these Algae can probably be obtained from T. Bolton, Newhall Street, Birmingham. [Ed.] REPRODUCTION IN SPIROGYRA. 247 of tlieir masses of threads. The process can be easily followed, but the threads must not be directly covered with a cover-glass upon the object-slide ; but on the other hand the small pasteboard moist chamber, described on page 174, will serve with advantage, and then the Spirogyra is placed in the drop suspended from the cover-glass. Conjugation in most species takes place in ladder- like fashion, i.e., two threads lying alongside one another are united by cross-bridges. The cells put out short blunt projections, which come into contact and fuse with one another. In many cases it can be distinguished, before conjugation, which thread is male and which female, since the cells of this latter sw^ell out into barrel-shape. After the union of the conjugating processes, the contents of the male cell tend first to become rounded off, and finally withdraw on all sides from the cell-wall. They then pass into the conjugating canal, and through the partition walls [of the two conjugating processes], which in the meantime had become softened. The female cell had simultaneously rounded, or rounds off on entrance of the male cell. Both cells come into contact, and after a few minutes coalesce. Their contents blend ; the chloro- phyll bands join together ; the two nuclei unite into a single one,^ this, however, not being visible without the use of staining re- agents. The zygote [zygospore] thus formed begins at once to contract; after the course of an hour its cavity [hitherto filled with cell-sap] has completely disappeared [the cell-sap being expelled]. In this contraction the chlorophyll bands are drawn more into the interior, while the periphery appears composed of colourless frothy protoplasm. The zygote [zygospore] is more or less globular. After the lapse of twenty-four hours, it has again enlarged, acquired a cavity, and taken an ellipsoidal form ; the chlorophyll bands are pressed to the periphery, and a clear membrane, Avith double contour, now covers the zygote. This process of conjugation we have just studied is characteristic of the entire section of Alga3, collected together as Conjugatae. To this, besides the species of Sjjirogyra, so widely dilTused in fresh water, belong also the almost equally widely-spread species of Zygnema, recognisable by two stellate chromatophores in each cell, and the elegantly formed Desmidca\ Into close connection with these latter we can bring the Uiatomacca?, in whicli typu-al conjugation is likewise present. The genus Cladophora, belonging to the Chlorophyceae, the structui-e of which is already known to us, provides us with a 248 REPRODUCTION IN CLADOPHORA. riglit favourable object for the study of swarm-spores;'^ it is only to be regretted that they are not always inclined to the production of swarm-spores. It is comparatively easy to obtain swarm-spores of the marine forms, which we lay in a large vessel with sea- water; still amongst fresh- water forms,Cladophoraglomerata,if taken from rapidly-flowing water, and laid about evening in a shallow- vessel with a layer of water about |-inch deep, will be found with swarm-spores usually next day. The formation of these com- mences at the apex of the branches, and proceeds towards their bases. Hence all stages of development are easily found close together. We look at these in the direction from the base towards the apex of the branch, and commence with a still un- changed cell. The structure of this is already known to us. What is visible without reagents we soon recognise again : the polygonal, closely-crowded chromatophores, which contain small, pale starch-granules, in part also having larger amylum-groups ; the plasmic plates which traverse the cavity of the cell, and in jDart also contain chromatophores. If we pass now gradually from such a cell to such as are being transformed into sp Drangia, first of all a change of the colour of the contents strikes us. With a sufficiently high power we can determine at the same time the absence of the amylum-groups ; these have fallen into individual starch-granules, and simultaneously has come about a subdivision of the chromatophores into smaller ones. In the next stage the chromatophores begin to arrange themselves into a net, so that the entire contents of the cell, surrounding a narrower or wider cavity, appear divided into approximately equal polygonal sections. The middle of each of such sections is free of granules, and fixed and stained objects teach us that a nucleus lies there. At the same time the peripheral layer around the whole contents of the cell increases in thickness, and becomes readily visible. It is especially thick at the angles of the cell. At one place, usually in the neighbourhood of the front end of the cell, and in terminal cells occupying the anterior end, is still noticeable a lenticular aggre- gation of colourless protojDlasm. Corresponding with the centre of this aggregation the membrane of the cell swells and bulges out, likewise as the result of the increase in volume due to the swelling, into a papilla-like projection. The next change consists in that the chromatophores withdraw towards the interior of the polygonal sections, and these latter appear bounded towards one another by clear lines. The sections then begin to round off, and REPRODUCTION IN CLADOPHORA. 249 SO partially to separate from one another. The i)eriphoral sections now project outwardly as roundish knobs. The colourless peri- pheral protoplasm takes no part in the differentiation of the chlorophyll-containing contents into individual sections, but is transformed into a colourless mucus, which plays a part in the evacuation of the swarm-spores [zoospores]. Corresponding with the strong aggregation of colourless jirotoplasm at what later on is the place of exit, the mass of slime formed is here the largest, and the still connected mass of swarm-spores remains, therefore, at this place, somewhat removed from tlie swelling cell- wall. In the mulberry-like mass of the swarmers the cylindrical, more strongly or feebly developed cavity is now easy to see. In a sporangium very rich in contents it may be wanting. In general it is, however, present, so that the swarm-spores form a double or triple layer around this inner hollow. The swarmers soon take on a pear-like form. The anterior, colourless, tapering end is easily distinguish- able from the rounded chlorophyll-containing posterior end; on the surface of each swarm-spore appears a narrow, reddish-brown fleck, the so-called eye-spot. The cell-membrane, at the part corresponding with the papilla, is already so strongly swollen that its outline is difficult to recognise. With continuous observation we shall now soon see the moment arrive when the sallying of the swarm-spores begins. Under the pressure of the contents the swollen substance of the papilla is broken through, and the mass of the swarm-spores strongly pressed forwards. At the same time with the swarm-spores, finely granular masses of contents of the cell-cavity move outwardly. After a while the forward-pressed swarm-spores set up a motion. The contents of the sporangium, decreasing in bulk, withdraw from the cell- wall; apparently the mass of jelly which lies there presses on the cell-contents. If only a few swarm-spores are present in the sporangium, they now begin to move about confusedly, and one after another pass outwards through the papilla. A small number can also remain behind in the sporangium. If the object is examined in a suspended drop, under the influence of light, the swarmers ultimately collect either at the side of the drop turned towards, or at that turned from, the window. These swarmers are not, however, amongst those most sensitive to light ; they remain for a longer time scattered in the drop, move about in indefinite directions, and only gradually, while their motile energy diminishes, arrive at the edge of the drop, where they come to rest. They tlien round oft', and surround 250 REPRODUCTION IN VAUCHERIA. themselves with a cell-wall. Witli a little potassium-iodide-iodine, these swarm-spores can be very well fixed (Fig. 86). We recos- nise now two cilia upon each (or, with the species of Cladophora, even four), which arise from a small projec- ftion from the anterior end of the swarm-spore. [The swarm-spores move with the ciliate end forwards.] In swarm-spores lying in a favour- able position, the small nucleus is thoroughly recognisable, after treatment with iodine, Fig. 86.-aadopiwra ^J^'^S in the anterior colourless end (compare giomerata. A swarm- the figure) ; the nuclcolus usually stains very spore fixed with iDotas- -i ■. sium-iodide-iodine. On Sliarply. the right hand is seen These swarm-spores observed by us are anterior colourless por- asexual, but in Gladophora still other, smaller, tion the nucleus is to be sexually differentiated swarmers, or Gametes, seen(x 510), t i mi . . , are produced. These conjugate with one another, but have hitherto been observed only in the marine forms.'*' From amongst the SiphoneSB we select for examination the widely- spread Vaucheria sessilis, in order to study the formation of its swarm-spores (zoospores) and sexual organs. If we have collected strong specimens of this alga in standing, or better still in flowing, water, and afterwards placed then in a shallow vessel with fresh water poured over them, we can pretty certainly reckon on numerous swarm-spores by next morning. These are being turned out the whole forenoon, so that we can easily find all the desired stages.'' If we examine the culture with a lens of considerable focal distance, we can easily recognise the first for- mation of the sporangia by the dark colour of the ends of the threads. If now we seize with the forceps, at their point of junction, a group of threads which appear to be in the desired condition, and transfer them, without allowing them to be bent, to an object-slide, we may study upon them directly the further processes of development. Moreover, these often go on undis- turbed under the cover-glass, if only the object is protected from the pressure of the cover-glass by minute fragments of elder-pith, or horse-hairs, placed under its edges. If a sporangium has been formed out of the end of a branch, contents rich in chlorophyll collect in this, and at the same time the end of the twig begins to swell into a club. The hollow in the club is narrowed (Fig. 85, A), and is soon separated off in the upper part of it as a REPRODUCTION IN VAUCHERIA. 251 spherical vacuole. The sporangium is then cut off by a partition wall, in the formation of which the chlorophyll-containing con- tents of the young sporangium and of the rest of the sac temporarily separate from one another, so that we can see them separated by a clear interspace (Fig. 87, B). Around the contents of the sporangium is now formed a clear border (E), which soon shows radial structure. This border consists of colourless proto- plasm, the radial structure arises from the elongated, radially- arranged nuclei, which are here collected (F, G). These nuclei show up clearly only after treatment with suitable reagents, and are only visible with strong magnifi- cation.^ The swarm- spore of Vaucheria is therefore multi- nuclear. When the swarm -spore is fully formed, it is at once evacuated. The apex of the sporangium rup- tures with a jerk, and at the same moment the anterior part of the swarm- spore flows out of the opening, and simultaneously be- gins to rotate upon its axis. The swarm- spore has to squeeze through the opening. Its birth lasts usually somewhat over a minute. A swelling substance formed in the sporangium helps to expel the swarm-spore. In many cases, though comparatively seldom, the anterior end of the swarm-spore twists off from the hinder part, still in tlie sporangium; the anterior part then hastens to form a complete, but correspondingly small, swarm- spore, and the posterior part forms a second swarm-spore. This is only possible by virtue of the multinuclear character of these Fig. 87.—Vauchcvia scusiUs. A and B, formation of the sporangium; C-E, formation of the swarm-spore out of the contents of the sporanrrium ; F, a swarm-spore set free ; G, a portion of the outer, colourless plasmic layer ["ectoplasma"], taken from the anterior end of the swarai- spore. <4-E(x 95 ); F (X 205); G (X 450). 252 EEPRODUCTION IN VAUCHEKIA. swarm-spores, in that each half contains the nuclei necessary to its existence. The movement of the swarm-spore lasts about a quarter of an hour ; the direction of the movement is not in- fluenced by the direction of the rays of light falling upon it. The swarm-spore is egg-like in form ; in the anterior part it is broader ; in this anterior end lies the cell-cavity [vacuole]. Only in the moment when the swarm-spore comes to rest are the cilia visible ; they cover the whole body as a short down. In the next moment they are withdrawn into the body of the swarm-spore, which, during this process, shows a wrinkled surface. Afterwards the body is again smooth. During the withdrawal of the cilia it is noticeable that the swarm-spore has already surrounded itself wdth a very delicate membrane. The spore now rounds off slowly ; its colourless border disappears, while the chlorophyll- grains come to the surface ; the cell- wall rapidly becomes thicker. In the terrestrial form of Vauclieria sessilis, Vauch., the sexual organs are found very easily. The species is re- cognisable in that the female organs, the oogonia, are ses- sile (i.e., unstalked) upon the filamentous thallus, and that the male organs, the antheridia, on the other hand, arise as a direct con- tinuation upon a short, horn- like curved branch, arising immediately from the fila- mentous thallus. An antheridium and an oogonium arise usually in a pair quite close together ; not infrequently, however, an antheridium can be seen between two oogonia. This species of Vauclieria is chosen for examination, and not that which is met with quite as commonly upon damp earth, in which oogonia and antheridia are seated upon a common lateral branch, which is ended by the oogonium. This last sjDCcies, Vauclieria terrestris Lyngb., is little suited for examination. The aquatic Vaucheria sessilis, when in culture, forms at first the swarm-spores already described, and tends only after some weeks to produce sexual organs. The oogonium is obliquely ovate, thickly filled with plasma containing chlorophyll and oil, separated from the thallus- i^^^M^^^:^^ Fig. 89. — Vauclieria sessilis. Portion of the thallus with sexual organs, o, oogonium ; a, an- theridium ; ch, chromatophores ; ol, oil-drops. The nuclei 71 are inserted, although they are only visible after suitable staining ( x 210 ). REPRODUCTION IN VAUCIIEniA. 253 thread by a partition wall placed somewhat above its point of insertion. The oogonium is provided with a unilateral, beak-like outgrowth, in which colourless protoplasm is collected. In ad- vanced stages of development of the oogonia this latter occupies the entire upper third of the egg [-cell, or oosphere].* If now we observe such an oogonium continuously, we shall see the colourless substance at the beak end put out a papilla-like pro- jection, which rounds off more and more into an independent ball ; this separates finally from the contents of the oogonium, and is thrust out into the surrounding water, where it slowly goes to the bottom. Direct observation shows us that in this the membrane of the oogonium at the end of the beak is not per- forated, but rather swells into a jelly, and that the issuing plasmic drop is pressed out through the jelly. The remaining contents of the oogonium round off, its colourless apex is the place of fertilization [receptive spot of the oosphere, or egg-cell].— The antheridial branch is more or less strongly curved. Its upper third is formed into an antheridium, and is cut off by a partition wall (Fig. 88, a). In the ripe condition it is distinguished by its colourless contents, while the branch Avhich bears it is rich in chlorophyll-grains. The antheridium usually turns its apex away from the oogonium. In the colourless contents of the antheridium short rodlets, arranged longitudinally, are more or less clearly distinguishable. At the moment when the oogoni-j.m presses out a portion of its colourless protoplasmic substance, the antheridium opens at its apex, and evacuates its slimy content. The greater part of this remains in the surrounding water in the form of colourless bubbles, where it slowly disorganizes ; a smaller part hastens away in the form of minute glancing spermatozoids [antherozoids]. These actively swarming sper- raatozoids soon collect in the mass of jelly at the apex of the oogonium. Individuals press forwards to the colourless receptive- spot of the egg [-cell], and as it were grope around it. In specially favourable cases a fusion of such a spermatozoid witli * Some explanation of the terminolofry is needed here. The female cell, prior to fertilization, the Author calls the "o^'g." By otliers it is known as the "egg-cell" or "oosphere." After fertilization, keeping up the analogy with conjugation, the Author calls it "zygote"; by others this fertilized cell is called "egg" and "oospore." The use of the terms "zygospore" and " oospore " implies the morphological difference between the fertilizing cells in the two cases ; the use of the one term " zygote," as comnu>n to both, emphasizes the physiological unity of the sexual process, even when more highly evolved than here. [Ed.] 254 NOTES. the receptive-spot can be determmed. After a sliort time the fertilized egg [or oospore], the zygote, has surrounded itself with a delicate membrane, which is especially clearly visible at the receptive-spot. In the course of some hours the colourless proto- plasm of the receptive-spot is diffused equally in the zygote. Older zygotes are thickly filled with oil- drops, show some brown spots in the interior, and have a firm membrane. If a spermatozoid in course of movement is fixed with potas- sium-iodide-iodine, two cilia, unequally long, unilaterally inserted, and extended in opposite directions, can be seen attached to it. NOTES TO CHAPTEE XXII. ' De Bary, Conjugaten, p. 3 ; Strasburger, Befruchtung und Zelltheihmg, p. 5 ; Kny, Waudtafeln, Text, p. 11. - Schmitz, Stzher. der niederrh. GeselL, 4th Aug., 1879, p. 23. 2 Compare Thuret, Ann. des sc. nat. Bot., III. Ser., torn XIV., p. 219 and plate 16 ; Schmitz, Siphonocladiaceen, p. 31, and Chromatophoren, p. 119, Note ; Strasburger, Zellbildiing und Zelltheilung, 3rd Edit., p. 72. •* Compare Areschoug, Observ. phycoL, II., Acta soc. sclent. Upsala., vol. IX., 1874. 5 Thuret, Annales des Sciences Naturales, Botanique, II. Series, torn. XIX., p. 270 ; Strasburger, Zellb. u. Zellth., 3rd Edit., pp. 213 and 84. 6 Schmitz, Stzber. der niederrh. Gesellschaft, 4th Aug., 1879, separate reprint, p. 4 ; Strasburger, Zellb. u. Zellth., 3rd Edit., p. 88. '' Compare Pringsheim, Monatsber. der k'ongl. Akad. d. Wiss. zu Berlin, of the year 1855 ; De Bary, Ber. der Freib. Naturf. GeselL, 1856 ; Strasburger, Zellb. II. Zellth., 3rd Edit., p. 90. REPRODUCTION OF FUNGI. 255 CHAPTER XXIII. THE KEPEODUCTION OF FUNGI. Material Wanted. Fresh horse-dung to grow moulds upon. Piece of diseased potato plant. Fresh. A piece of bread to grow blue mould upon. If a piece of damp bread is placed under a glass bell-jar, it is covered, even in a few days, with a thick felt of fungus threads, [mycelium], which almost always belongs to Mucor Mucedo} one of the Phycomycetes. This fungus soon shoAvs itself very luxuriantly upon fresh dung, kept in a closed moist chamber. From the substratum arise erect fruiting brancl>es [conidiophores, or gonidiophores] , an inch or more in length, which turn towards the source of light, and end each one wdth a globular, yellow or brown head, readily visible with the lens [and even with the naked eye]. If we lift some of this material carefully from the substratum, and place it in a drop of water, we can determine, by means of sufficiently strong magnification, that the mycelium consists of thick, copiously branched, irregularly septate sacs [hyphae], and that from these arise the straight, unseptate and unbranched fruiting branches, which bear each one of the globular heads, the sporangium. If still unripe, this remains unchanged in the water; its contents consist of yellowish-brown protoplasm. In the youngest stages the fruit-stalk is not cut off from the sporangium ; later on there arises a partition wall, strongly arch- ing into the interior of the sporangium, so that the stalk ends inside the sporangium with a swelling like a ninepin, the so-called columella. The ripe sporangium deliquesces in water, and of its wall only fragments formed of fine needles remain behind, of which it has been determined that they consist of oxalate of lime.^ The expelled spores lie at pretty regular distances from one another ; and by pressure on the cover-glass we can determine 256 siuccR muced:. that tliej lie embedded in a colourless slime. On tlie stalk, under the columella, is usually to be seen a small collar, as a relic of the lime-crust which was attached there. In the peripheral proto- plasmic layer of stalks which are not too old, we can follow fine, in the main longitudinal, streaming of the protoplasm. The sacs of Mucor are multinuclear, the nuclei very small, only distinguish- able by suitable staining. In dung-cultures the fungus occasion- ally forms zygotes [zygospores], which present themselves as dark points. They can usually be forced into the formation of zygotes [zygospores] in the months of March and April, if the spores are sown in fresh, flattened-out horse-dung. The zygotes are ready in from eight to fourteen days. At other times, in order to obtain the zygotes, it succeeds well if the sowing is made in some drops of concentrated plum-juice, sterilized by long boiling, and then mixed with ten to twenty per cent, of alcohol [not methylated]. The sowing is made on a cover-glass in a damp chamber constructed of a glass ring (see p. 244), and the object-slide placed in the large plaster-of -Paris moist chamber (p. 243). The zygotes [zygo- spores] arise by conjugation of the ends of mycelial threads swollen into club shape. On the ripe black zygotes, covered with warts, the positions of these two mycelial threads can be seen opposite to one another, as clearer places with circular outline. The cause of the potato-disease is likewise a Phycomycete, the Phytophthora infestans, de Bary,^ germinating hyphje of which penetrate through the membranes of the epidermal cells of the leaf into its intercellular spaces, and spreading about in these destroy the tissue of the host,* forming brown spots of constantly increasing diameter. In order to obtain the fungus fructifying in large quantity, we place a piece of a diseased potato-plant in an atmosphere saturated with moisture under a bell-jar, and let it lie there for about two days. The diseased leaves are now covered over on both sides, but especially on the under, with white "mould," formed by the filamentous fruiting branches [gonidio- phores] of the Phytophthora. These tufts of mould are especially developed at the edges of the brown spots. On surface sections of the parts covered with mould we see the gonidiophores f pro- * I have thus rendered the Author's term " Nahrpflanze," as the word " host " is fully incorporated into English scientific phraseology as signifying the Hving organism upon which another organism, animal or vegetal, lives, and more or less completely preys. [Ed.] t lu the Author's corrections for the EngHsh edition, he has throughout, following the terminology of de Bary, erased the term " conidia." and inserted PIIYTOPHTIIORA INFESTAXS. 257 jecting through the widely opened stomata. We can demonstrate this, though less completely, in fragments of leaves, which we place in their entire thickness under the microscope. The gonidiophores appear as delicate, unseptate threads, branched above, and filled with finely granular protoplasm (Fig. 88, A). The branching is monopodial [or racemose] ; the number of branches usually but two or three. These branches are irregularly swollen at intervals. In dry air the gonidiophores, collapsing, are twisted upon their axis. Here and there we see at the end of a branch a gonidium in course of development ; the ripe lemon-shaped gonidia, however, have fallen off in laying the preparation in water. In order to find the gonidia on the gonidiojDhore, we must examine the pre- paration dry. The preparation is, however, to be covered with a cover-glass, and a trace of water placed under it from the edge, because otherwise the gonidiophores, as already indicated, rapidly drying, shrivel up. In jDlants collected from the open air the gonidiophores are found only on the under side of the leaves, and do not grow so tall as in the moist chamber ; are much less notice- able, therefore, with the naked eye. Delicate cross-sections through diseased leaves, made by means of elder-pith, and always at the margins of the spots, permit us to clearly follow the exit of the gonidiophores from the stomata. Often several such hypha3 come side by side out of the same stoma ; or, more commonly, the hypha branches at its exit, and gives correspondingly more gonidiophores. From these places we can, though with great difficulty, follow the liyphas inwards, into the tissue of the leaf, and determine that they pass into the intercellular spaces. As a distinction from the most nearly-allied species of Peronospora, Phytophthora forms but spar- ingly, and then only short, suctorial organs (haustoria), penetrating into the cells of the host, so that usually they may be looked for in vain. The delicate mycelial threads, on the other hand, cling closely to the cells of the host. Such cells show first a brown- ing of their chlorophyll-grains; these fuse finally together, and with 'he other constituents of the cell, into a dark-brown, coagulated mass ; at the same time the whole cell collapses. The gonidia are lemon-shaped (Fig. 88, B), with short stalks, somewhat tapering apex, and finely granular contents. The membrane of the gonidium is very delicate, a little swollen at the apex. Tliey are, as we "gonidia." I have retained this altei-ation. In most works, however, the student will find " couidia," with its derivatives, applied to Fuuf,'i, "gonidia," and its derivatives, applied to AlgJE— a separation not without advantage, l^d. j 258 PHYTOPHTHORA INFESTANS. have already seen, situated at the ends of the branches of the gonidiophore ; if they have attained their full dimensions, the apex of the branch under the point of origin of the gonidium further grows unilaterally, presses the gonidium over to one side, so that this comes to lie in a position at right angles with the branch. At the apex of the branch soon arises the foundation of a new gonidium (compare Fig. 89, A) . We sow the gonidia in a drop of water upon a cover- glass, and take care by stirring the drop that the greater part of the gonidia are immersed. The cover- glass is laid upon a small moist- chamber, and the drop suspended from it. The culture must not be carried on in too intense light. After the lapse of about an hour, joerhaps later, the formation of swarm-spores from the contents of the gonidia begins. The gonidia are ^converted, therefore, into sporangia; they can, how- ever, germinate directly, when we see some of those lying at the surface or at the edge of the drop put out a hyphal sac from the anterior papilla. In those that are immersed and form swarm-spores, the contents divide into an indefinite number of cells (C), in each of which we can see a small central vacuole. The apex of the gonidium swells out into a papilla, is finally dissolved, and the separated masses of its contents are successively pressed out through the small round aperture. They hasten away as swarm-spores. If we fix these Fig. 89,—^, surface-view of the epidermis of the leaf of Solanum tuberosum [the potato], with the gonidiophores of Phytophfhora infcstans projecting out of the stomata (x 90); B, a ripe gonidium; C, another, with divided contents. D, a swarm- spore (B-D X 540). PENICILLIUM. 259 swarm-spores with iodine solution, we can determine the presence npon them of two cilia. These are inserted laterally in the proximity of the now peripheral vacuole (B). The movement of the swarm-spores lasts up to half an hour. They then come to rest, surround themselves with a cellulose membrane, and ger- minate soon into a hyphal sac. It is this sac, developed directly from the gonidium, or from a swarm-spore, which penetrates through the epidermis into the stem and leaves of the potato-plant, and can, as may be proved, in this way infect a completely healthy plant. The rapid multiplication of the parasite is provided for by the formation of gonidia. Sexual organs have not yet been found upon Phytojjlithora infestans, although known for the nearly allied Peronosporea3. In these, mycelial branches in the interior of the host swell, usually at their end, globularly, and form the OOgOnia by catting off these swellings by partition walls. By each oogonium is found a mycelial branch, with its end cut off as an anther idium. The greater part of the protoplasm present in the oogonium forms a central globular egg [-cell or oosphere]. The antheridium puts out a fertilizing sac to the egg [-cell], and this surrounds itself afterwards with a firm membrane.* Upon the most variable objects in damp positions, even if only traces of nourishment can be obtained from them, soon is wont to be found the blue-green mould, Fenicillium crustaceum, Fries. ^ It is the most widely distributed of all moulds ; we meet with it everywhere. We shall not, therefore, need to seek long for material for examination. It will be, however, most convenient to moisten a piece of bread, and place it under a bell-jar. Not improbably Mucorineoe will first show themselves on the bread ; but soon the, at first, more slowly developed Penicillium will have supplanted it, and after about eight days covers the substratum with a dense, blue-green covering. The blue-green coloration arises from the spores of rcnicillium, which, however, only show this coloration when in great quantity. We now lift a little mate- rial from the substratum, and examine it in water. The mycelium consists of branched hyphre, which are divided by partition walls. The contents directly visible are finely granular proto- plasm and small vacuoles. Individual threads, not distinguish- able from other mycelial threads, have developed into fruiting * In our customary terminology, by fertilizing, the egg-cell (oosphere) be- comes an egg (oospore). [Ed.] 260 PENICILLIUM. branches [gonidiophores]. At their apex they bear a whorl of short branches, which branches (Fig. 90, s') on their part either bear directly whorls of basidia, or previously each one again bears a whorl of shorter lateral branches, and then these bear the Avhorls of basidia. This branching gives to the fruiting branch the appear- ance of a brush. Commonly other lateral branches, which arise just under a partition wall of the primary fruiting branch (as in the right hand of the figure), come up laterally to this terminal brush, and form secondary fruiting branches. The basidia, as suffici- ently strong magnification shows, are cylindrical, prolonged at their end to a finer projection, the sterigma [pi. sterigmata]. This sterigma swells to a globular point, and forms a quickly-growing spore. Under the first spore soon shows a second swelling, which becomes a spore, and so on, so that chains of spores arise [the terminal spore being the oldest]. The uppermost spores of the chain are thrown off, while new ones press outwards from below. Tufts of Penicillium, fixed with alcohol, stain very well with haematoxylin [logwood], by which it can be determined that in the cells of the mycelium and the fruiting branch numerous nu- clei are present.^ The nuclei are very small, so that they require strong magnification. They are elongated in the longitudinal direc- tion of the cell, and joined by fine plasma-strings. In long cells very many can be counted ; in the shorter branches of the whorls on the fruiting branches, only one or two ; in the basidia, pro- FiG. 90.— Penicillium crustaceum, fruit- ing branches with verticils of branches (s' and s"), basidia (b), sterigmata (st), and spores ; nuclei visible. From an alcohol-hEematoxylin preparation (x 540). ASCOSPORES IN PEiNICILLlUM. 261 bably only one at the upper end. The basidia, liowever, are usually filled so thickly with contents at the apex that the identification of the nucleus in them is impossiWe. In the spores also, with the strongest magnification, a nucleus can be distinguished with certainty for each. To complete what we have already said, it may be added that besides the above-described fruiting branches, it is possible to rear upon Penicillium a second kind of fruiting body.^ These arise in suitably managed culture en masse, have the size of small pin- heads, and a yellowish colour. In their interior, after longer period of rest, asci are formed, each of which produces 8 spores [ascospores]. Therefore Penicillium must be set down as an Ascomycete, one representing the section of cleistocarpous Asco- mycetes, with closed fruit-body. Out of the spores developed in the asci the brush-like gonidiophores have been again developed upon the object-slide. NOTES TO CHAPTER XXIII. 1 Brefeld, Schimmelpilzc, Heft I., p. 10 ; the other literature is given here. 2 Brefeld, I.e., p. 18. "^ Compare de Bary., Ann. dea Sc. nat. Lotanique, IV. Series, torn. XX., p. 32, and Beitrage zrir Morph. und P]iys. der Tilze, Heft II., p. 35. 4 Brefeld, I.e., Heft II. 5 Strasburger, Zcllh. m. Zellth, 3rd edit , p. 221. 6 Brefeld, I.e., p. 39. Compare Bainier, Annales d''S Seiences naturelles, Botanique, VI. Series, tom. XI., p. 315, for further particulars on the culture of the Mucorineae. 262 KEPRODUCTION OF FUNGI AND LICHENS. CHAPTER XXIV. THE EEPEODUCTION OF THE HIGHEE FUNGI AND LICHENS. Material Wanted. Leaves of Barberry with cluster-cups. Fresh (gathered in May or June), dry, or in alcohol. Plants of grass (wheat or oat, etc.) affected with rust. Fresh (gathered mid-June to August), dry, or in alcohol. Bussula sp. Fresh, or in alcohol. Failing this, the common Mush- room {Agaricus carnpesiris). Fresh, or in alcohol. The Morell {Morchella esculenta). Fresh, dry, or in alcohol, Anaytijchia ciliaris, in fructification. Fresh, dried, or in alcohol. In the months of May and June are found not infrequently upon the under side of the leaves of the Barberry (Berheris vulgaris) orange-coloured warts, which, to the naked eye, appear finely pitted. Examination with a lens shows them as cushion-like yellow swellings, upon which are placed small orange-red cuplets. The corresponding positions on the upper side of the leaf present themselves as reddish spots edged with yellow. Examined with a lens, usually numerous brown points, surrounded with orange-red, show out in the inner parts of them. Individual similar points are often to be found on the edges of the cushion on the under side of the leaf. The cuplets on the cushion of the under side of the leaf are the aecidium-fruits of ^ciditim Berheridis [the " cluster- cup " of the Barberry] ; the corresponding points on the spots on the upper side of the leaf, and also upon the edges of the cushion on the under side of the leaf, are the spermogones appertaining to them. Together they form the first generation of the common fungus, rust of wheat, etc. {Puccinia graminis), belonging to the JEcidiomycetss or Uredineae, of which the second generation is gone through upon our corn and other grasses, giving rise to the appearance of the disease called "rust."^ By means of elder-pith we prepare delicate cross-sections through an infected leaf, and examine them with weak, and afterwards with strong, magnifica- iEClDIDM. 263 tioD. We assume that fresh material stands at our disposal ; the investigation can, however, be carried on satisfactorily upon dried and soaked, and well upon alcohol material. The sections pre- pared from the fresh leaf are especially clear if we run in a little potash solution. In the uninfected parts the barberry leaf shows, proceoding from above downwards : an upper epidermis ; a single layer of elongated palissade- cells ; a loose spongy parenchyma, about five cells deep ; the under epidermis. The cushions of tissue of the infected parts have attained more than double the thick- ness of the leaf. Upon the palissade layer of the upper side, which is higher, but otherwise appears little changed, impinges a closed tissue, which also shows to be more or less elongated in a direction at right-angles to the surface of the leaf, and from the small development of its intercellular spaces is essentially distinguished from the spongy parenchyma of the surrounding parts of the leaf. The epidermis of both surfaces of the leaf has not been affected in the form of its cells. The contents of all these cells are disorganized, and consist partly of colourless oil- drops, partly of greenish-yellow and reddish drops and granular masses, proceeding from the chlorophyll-grains and the cell pro- toplasm. The entire tissue of the cushion shows its intercellular spaces traversed by delicate, partially-branched fungal hyphae, septate by cross- walls, and containing oil-drops. These extend on both sides to the epidermis. With chlorzinc iodine, as also with iodine and sulphuric acid, blue coloration is not induced, since fungal-cellulose rarely shows .this reaction. The aecidium-cups, as we have them before us in longitudinal section, are sunk above the middle in the tissue of the cushion. We easily determine that the mycelial hypha3 under the cups form a dense, almost pseudo- parenchymatous, layer, from which, perpendicularly outwards, and parallel to one another, rise numerous thicker club-shaped hyphae, m gap-less union, forming the so-called hymenium. These hj-phcc, the basidia, pass over at their ends into straight rows of spores, which at the basidia are colourless and, from mutual pressure, polygonal, but gradually become orange-red and rounded. Higlier up the spores separate from one another, and are evacuated from the opened fruit. The observation of the youngest spores upon the basidia convinces us, however, that they are cut off one after another by cross-walls, from the apex of the growing basidia. The unilamellar wall of the fruit (the peridium) consists of cells which look very like the spores, but remain polygonal, and do not sepa- 264 LIFE-HISTORY OF rate laterally from one another. Their fine delicately porous walls are especially strongly thickened on the outer side. The develop- ing peridium pushes back and destroys the sun^ounding tissue of the cushion, and tears open the epidermis in order to open out to the exterior.— The pear-shaped spermogones, especially found upon the upper side of the leaf, are, like the secidium-fruits, sur- rounded by a weft of hyphse, though less strong, from which arise densely-crowded, parallel threads, running towards the middle line of the structure. These threads are very delicate ; those found in the upper part of the organ project as delicate bundles towards the exterior [compare Fig. 90* later]. These delicate threads, the sterigmata, abstrict at their points exceedingly small, globular cells, the spermatia, which, as a slimy mass, are evacuated out- wards from the organ. The sterigmata themselves contain orange-red oil-drops, which gives to the entire organ, especially in its outer parts, its special coloration. The spermatia do not germinate ; their significance is still unknown. There is a dispo- sition to take them for male sexual products, and to consider that the sexual act leads to the formation of the secidium fruits. — As already mentioned, the fungus lives as a second generation upon Grramineae. It belongs to the hetercecious parasites, which, in contradistinction to the autoecious, go through their alternation of generation upon different hosts. This can be demonstrated by direct sowing of the ascidium-spores upon seedlings of the cereals.'^ The uredo patches of Puccinia graminis we encounter not in- frequently in the open field, from mid-June to autumn, upon rye, wheat, barley, oats, and especially also on the couch-grass or twitch (Triticum repens). They especially take possession of the haulm and the leaf -sheaths of the infected plants. They are easily recognised as narrow, rust-coloured to dark-brown streaks, run- ning parallel with the veins. Upon the leaf-sheaths and haulm (stem) they attain to even two or more inches in length. The epidermis of the host is torn open and raised by the protruding spore-patches. First appear the rust-coloured patches of uredo- sporeSjwith which gradually are associated brown teleutospores. They take possession of the patches of uredospores, and at length completely supplant these latter, whereon the patch becomes dark- brown, almost black. Towards the end of summer only teleuto- spores are to be found. — If fresh material is not obtainable, those preserved in alcohol, and even dry plants, will answer for the investigation. We first prepare a cross-section through the haulm ^GIDIUM-PUCCINIA. 265 of an oat which is infected with rusty ureclo-patches. We can easily demonstrate upon the cross-section that the fungal hyphoe only traverse definite tissues of the host ; it is the chlorophyll- containing, looser strips of tissue, wliicli alternate with sclerencliy- matously-thickened strips, in the periphery of the stem, and are covered with an epidermis provided with stomata. Here the cells are densely enveloped in segmented hyphse, and their contents disorganized. In the places where the section has passed through a patch, we can see numerous short and delicate twigs, directed outwardly, arise from the mycelium, which at their swollen ends abstrict a unicellular spore, the uredospore. The surface is rup- tured, its edges thrown up laterally. The spores are in different stages of development. Those that are ripe appear an elonGj-ated oval, and with sufficiently strong magnification permit us to dis- tinguish two layers in their wall. The outer, dark-brown, is covered with numerous small warts ; the inner, less dark, shows several, usually four, regularly distributed pits in the equator. The contents of the spore are granular, in the interior parts a lively orange-red. Cross-sections through the haulm of oat, bearing the dark- brown patches of teleutospores, show the same structure, as far as the hypha3 are concerned, as we have already seen. The teleuto- spores are mounted upon just the same, but thicker- walled, stalks as the uredospores. The teleutospores are two-celled. The two cells together form an obovate body, somewhat tapering at its two ends. The spore-wall is dark-brown. Plants examined in the course of the summer may have at the same time urodo- and teleutospores in the patch. "We may here supplement this by adding that these teleuto- spores hibernate, and are first capable of further development in the next spring. Each of the two cells puts out a delicate tube, the so-called promyceliiiui, which is segmented into several cells, and from these are put forth short awl-shaped outgrowths, which cut off at their apex a kidney-shaped sporidiiim. These sporidia can only infect [i.e., germinate upon] the Barberry leaf ; if they happen upon a sufficiently young leaf, their germinating tube pierces straight through the outer wall of the epidermal cell directly into the interior of the host. As we therefore see, the way through the stomata, by which the germinating tubes of the secidio- and uredospores enter, is not the only one by which infec- tion is possible. 266 iEClDlUM-PUCCINIA. [In order further to elucidate tlie various structures referred to in detailing tlie life-lii story of this ^cidium-Piiccmia, I introduce the adjoining Figure 90*. The description given at the foot of the figure should be compared with that given in the text.] [Fig. 90*.— Puccim'o graminis and ^Ecidium Berheridis. I., transverse section of the leaf of Bcrheris, with secidia (a) ; p, the peridium, or wall of the secidia ; u, lower, o, upper epidermis of the leaf ; from ii to y" the leaf has become thickened by the action of the parasite, thus forming the cushion; on the upper surface are spermogonia (sp). A, a young secidium which has not yet burst. II., layer of teleutospores (f) on the leaf of Tviticum repens; e, its epidermis. III., part of a layer of uredospores on the same plant; tir, the uredospores ; t, a teleutospore. (From Prantl, after Sachs.)] In order to make ourselves acquainted with the structure of the hymenium of the Hymenomycetes,^ we select as best one of the numerous species of Toadstools (Amanita), Mushrooms (Agaricus- Psalliota), or Bussula. We select here for examination a Bussula. because this possesses, moreover, the cystidia, which we have at RUSSULA RUBRA. 267 the same time to make mention of. The cap [or pileiis] shows on the underside radially-arranged lamella;. These bear tlie liymenium. We cut, parallel to the course of the lamella?, a small piece out of the cap [pileus], and make through this cross-sections perpendicular to the course of the lamellae ; these must be as thin as it is possible to make them. The entire cross-section appears like a comb, on which the sections through the lamellce form the teeth [see Fig. 91*, A, later on]. With a low power we see that the hyphjB pass out of the cap into the lamella?, run rectilinearlv in the middle of these, and gradually give off ramifying branches, which are directed obliquely towards the flanks of the lamellae, and again branch [see Fig. 91*, B and 6']. Some of these branches swell into club form, and end blind. A large proportion re- main slender, . and form, outside the club-shaped swollen branches, a dense layer of short, rounded segments, which we will dis- tinguish as the sub- hynienial layer. This is limited more or less sharply to- wards the inner tissue of the lamella — the trama. The club-shaped swollen branches of the trama serve to give to the lamella the necessary stiffness. From the sub-hymenial tissue spring the basidia, and paraphyses (Fig. 90). These have an approximately parallel course, stand out perpendicularly from the flanks of the lamella", and form the hymenium. The basidia (h) are club-shaped. On their flattened ends they form four, symmetrically-placed, thin branches, the sterigmata (c). These swell gradually at their apices, each into an ellipsoid spore, basidiospore. The basidio- spores, when they have attained their full dimensions, remain in most cases smooth, or, in many species of Russida (cf. Fig. 90), they form short spines on their surface. They are cut off from Fig. 91.— Kussula rubra. Part of the hymenium. s?i, sub-hymenial layer; 6, basidia; s, sterigmata ; si>, spores; p, paraphyses; c, a cystidium ( x 510). 268 AGARICUS CAMPESTRIS. the sterigmata by a partition wall, and nltimatelj fall off. The septation and separation takes place a little below the spore- swelling, in the position where the sterigma shows a slight nick. The fallen spore has therefore a short stalk. Smaller basidia, which remain sterile, form the paraphyses {p). — So far the Toad- stools and Mushrooms I agree with the above- described Bussula. In Bussula, however, be- tween basidia and paiaphyses occnr also individual cystidia(c), structures of the size of the basidia, which project a little with their pointed ends above the hymenial surface, with their narrowing base pene- trate to the sub-hy- nienial tissue, and show themselves to be direct branches of the median elements of the trama. All the elements in ques- tion are bounded at their base by jDarti- tion-walls ; they con- tain finely granular protoplasm, and not infrequently individ- ual oil-drops. [The accompanying Fig. 91* will further elucidate the struc- ture of the Basidio- mycetes. It is taken from the common Mushroom, Agaricus campestris. There are no cystids, and each basidium produces only two, instead of four, basidiospores. The [Fig. 91*. — Agaricus campestris. A, tangential section of the pileus (Ji), showing the lamellse, I. B, a similar section of a lamella more highly magnified; hy, the hymenium ; t, the central tissue, or trama ; s7i, the sub- hymenial layer. C, a portion of the same section still more highly magnified (x 550); q, young basidia and paraphyses. At this stage these are practically alike ; s', first formation of conidia on a basidium ; s", s'", more advanced stages ; at s"", the conidia have fallen off. (From Prantl, after Sachs.)] MORCHELLA ESCULENTA. '269 Mushroom offers, moreover, the advantage of being obtainable fresh all the year round..] In order to investigate the structure of the hymeiiinm of a highly-developed form of the Ascomycetes, we will take the Morell, Morchella esculenta. Even dried specimens can here, after softening, be employed for the investigation. Fresh are naturally to be preferred. The well-known [edible] Morell has an irregularly ovate, stalked fructification, which in its interior conceals a simple hollow, and whose upper swollen part has deep folds. The areas or chambers [betw^een the ridges or folds] are clothed with hymenial tissue, while these are not developed upon the projecting exposed ribs or folds. Suitable sec- tions are very easy to obtain, and they must be taken perpendicularly to the surface of a chamber. The hymenium consists of approximately parallel spore- sacs (asci) and para- physes (Fig. 92). The sacs, or asci (a) are almost cylindrical, and con- tain in their ujjper part eight ellipsoid, unicellular spores [asco- spores], crow^ded together. Besides the spores, there is also present in the asci the, in part, strongly refrac- tive epiplasm. The paraphyses (p) are brownish, septate threads, some- what swollen above. The upper- most cell is especially long. They do not attain the length of the asci. Asci and paraphyses arise as ends of hyi^hoe of the densely-interwoven, flatly-extended, sub-hymenial tissue loosely-constructed inner hyphal texture of the fructification. Addition of potassium-iodide-iodine colours the masses of epi- plasma in the asci reddish-brown. This reaction is characteristic of epiplasma, and has been recently pointed out as a reaction for glycogen.* The characteristic peculiarity of this i-eaction is shown upon warming. To the preparation lying in water, and stained b}- the addition of potassium-iodide-iodine, some water is added, yet not so much as to decolorize it ; then it is carefully warmed, with- point, and held over white paper, in Fig. 92.— Portion of the hymenium of Morchella esculenta. a, asci ; y, paraphyses ; sh, sub-hymenial tissue (x 240). This arises fi-om the more out the boiling 270 REPRODUCTION OF LICHENS. order to see if the colour has become paler. If this happens, the preparation is then rapidly cooled, and the darker coloration, in large preparations even visible with the naked eye, again appears.^ With the aid of potassium-iodide-iodine, the base of many asci can be traced pretty deejDly into the sub-hymenial tissue. The contents of the spores, paraphyses, sub-hymenial tissue, and of the tissue in the interior of the fructification colours simultaneously yellow or yellowish-brown. The fungus appertaining to the thallus of Lichens belongs, with few exceptions, to the Ascomycetes. The Anaptychia ciliaris, already known to us, fructifies very freely. The apothecia are bowl-shaped, with a frame developed from the thallus. This con- tracts under the apothecium like a stalk. A cross-section through this stalk shows radial structure, with a uniformly thick cortical layer, and following this a homogeneous gonidial layer in its entire circuit. The interior of the stalk is occupied by a medulla, or " pith," formed of a looser hyphal texture. We further prepare median longitudinal sections through the apothecium. This shows us the frame formed of the tissue of the thallus. The gonidial layer extends to its rim, which grows out in places into ciliary outoTowths. The apothecial stalk has expanded into a bowl-like form, in order to admit the hymenium, which arises from the medullary tissue. The hymenium is recognisable by its somewhat brownish colour. It consists of very numerous, long, exceedingly narrow, septate threads, the paraphyses ; between these, far less numerous, stand the club-shaped sacs, the asci. These latter are always of different ages ; the ripe ones contain eight brown- walled spores [ascospores]. These spores are ellipsoid, two-celled, a little constricted at the boundary of the two cells. Paraphyses and asci arise from a like-coloured, felted, horizontally expanded layer of little thickness, which is distinguished as the sub-hymenial layer. This originates from the medullary tissue of the stalk, from which it is cut off by its brown colour and the want of air- containing spaces. While, as we have seen, the hyphae of the thallus itself are not coloured blue even by chlorzinc iodine, the hymenial layer take a dark-blue coloration, even with the addition of a little potassium-iodide-iodine. The walls of the hymenial elements are formed of a special modification of cellulose, which has been distinguished as starch- cellulose. — If we examine the thallus of Anaptychia ciliaris with the lens, we shall notice in individual spots upon it wart-shaped prominences, standing singly REPRODUCTION OF LICHENS. 271 or in groups. If in such places delicate cross-sections are taken in considerable numbers, we shall probably cut through such a swelling (Fig. 93). It appears then as an ovate structure, sunk in the thallus, and opening outwardly with a pore, and is now known as a sperm ogonium.* It occupies almost the entire depth of the thallus, is laterally surrounded by the gonidial layer, and in the interior shows itself to be constructed of delicate, shortly segmented, approximately radially-arranged threads, singly or in bundles, — the sterigmata (compare the Figure) . The long axis of the organ is traversed by a cylindrical cavity, which receives the rod -like spermatia, which are segmented off from the ends of the sterigmata. Through the upper open- ing of the spermogone, the spermatia can pass into the exterior. In the CollemaceoB [gelatinous lichens] the function of the spermatia as male sexual organs has been determined^; in other lichens their significance is as yet unknown. Fig. 93.— Cross-section through the thallus of Anap' tychia ciliaris. NOTES TO CHAPTER XXIV. ^ Compare de Bary, MonaUhcr. d. k. Akad. d. WUs. in Berlin, for the year 1865, p. 15. Kny, Bot. Wandtafeln, text p. 68. Frank, Die Krankheiten der Pjianzen, p. 454. 2 De Bary, as above, for the year 1866, p. 206. 3 Compare de Bary, Morph. nnd Pliys. der Filze, p. 112 ; Goebel, Grundziige der Pfianzen-morphologie, p. 143. In both the other literature is quoted. * Leo Errera, L'epiplasme des Ascomrjcetes^ 1882. The literature of epiplasm is here given. 6 Leo Errera, I.e., p. 45. 6 E. Stahl, Beitrdge zur Entwicklungsgescliichte der Flechten, Heft. I., 1877. * As AiKipttjchia ciliaris may not bo at the disposal of the student, any one of the following lichens will serve to show spcrmogones : Parmelia (Phyxcia) parielina, Verrucaria nitida, Collema nielcemim, or Cladonia rangiferina. [Ed.] 272 EEPRODUCTION OF MOSSES AND LIVERWORTS. CHAPTER XXV. THE EEPEODUCTION OF MOSSES AND LIVERWORTS. Material Wanted. Marchantiapolymorpha (Liverwort), with male and female receptacles. Preferably fresh. May be kept in alcohol. Male " flowers" of a Moss, e.g., Mnium liornum, or Polytriclnim. Fresh, or in alcohol. {Mnhim hovnum is very common in woods and on shady banks.) Female " flowers " of the same. Fresh, or in alcohol. (Both of these gathered in April, May, or June Spore-capsnles of the same. Fresh, or in alcohol. Amongst Liverworts, Marchantia polymorjpJia, already known to us, rapidly multiplies vegetatively by its gemmgB. These are common amongst liverworts in general, and are here met with in especially beautiful form. The gemmee of Marchantia arise •upon the dorsal [upper] surface of the thallus in cup-shaped receptacles or cupules. [See Fig. 93*, B.] The cups have a beautifully toothed rim, and at their bottom the bright green gemmoe are visible. A median longitudinal section through the cupule, parallel to the long axis of the shoot which bears it, shows that the cup is at first slightly narrowed upwards, and then some- what suddenly expands into the broad rim. The tissue which forms the air-chambers jDasses into the exterior of the cup, to above the point w^here its outward broadening begins. The bottom of the cup is occujDied by unicellular club-shaped papillae, the membrane of which swells into mucus. Between these papillae are also found individuals which are two-celled ^ ; and some also, the upper cell of which has been further cross-septate. The lower cell remains simple, and forms the pedicel (or stalk) ; the off- spring of the ujDper cell soon divide longitudinally, and the struc- ture becomes constantly more multicellular, enlarges considerably by surface expansion, and becomes several cells thick in the middle. Others wdll be found w^hich have attained their ultimate biscuit- REPRODUCTION OF MARCHANTIA. 273 like [bi-convex] condition as fully developed gemrasB. The uni- cellular pedicel can easily be broken through. The separation of the gemmce, and their ejection from the cups, results from the strongly-swelling mucus, which is developed from the unicellular club-like papillae at the bottom of the cup. Each of the two lateral indentations of the gemma conceals a growing point, pro- tected by short papilla?. The cells of the gemma are rich in chlorophyll, but on both sides large cells, devoid of chlorophyll, occur, which keep near the middle, but otherwise are irregularly scattered. At the edge, individual cells contain oil-bodies. After the dissemination of the gemmae, the large cells, devoid of chloro- phyll, develop in one or two days into hair-roots [rhizoids], in all cases on the shaded side of the gemma only [hence becoming the ventral side], while the side exposed to the light forms morpho- logically the upper [or dorsal] side.- [Fig. 93*.— ^1, portion of a thallus of Marchantia pohjmorpha (i], with the upright male receptacle (hu), bearing antheridia. B, portion of a thallus with a receptacle containing gemmse ; v v, growing points of the two branches of the thallus. (From Prantl, after Sachs.)] The sexual organs of the Marchantiaceoe are situated upon special receptacles, which avc will examine in Marchantia poly- morpha.^ Male and female receptacles are readily distinguishable; the former shield-shaped, with scolloped outline (Fig. 93*, ^1), the latter radiating like bare umbrella ribs. The two sexes are situated upon different plants [the plants are dioecious] ; the receptacles and their stalks are metamorphosed branches of the thallus. We prepare between elder-pith delicate longitudinal sections through the male receptacle, and can demonstrate that .its upper side has exactly the same structure as tlie dorsal surface of the thallus, T 274 REPRODUCTION OF and that in the same way the under side resembles the ventral surface of the thallus, and bears rhizoids and scales. On the uppei' side, however, sunk in special cavities, are the antheridia (Fig. 94, A). On satisfactory sections we can determine that in each cavity is found only one antheridium, besides some short, unicellular paraphyses (p) ; the cavity closes together above the antheridium into a narrow canal. The antheridium is a shortly stalked, oval body, Avith a unilamellar chlorophyll-containing wall. The special mother-cells of the spermatozoids [antherozoids] have been produced by successive divisions at right angles, and even in the almost ripe antheridium still form rectilineally-arranged cross and longitudinal rows (compare the Figure). Shortly before the ripening of the antheridium, the special mother-cells, rounding off, pass out of union, the wall of the antheridium tears at its apex, and the small, round cells are evacuated. If a drop of water is placed upon a fully-developed re- ceptacle, the water is seen rapidly to, spread over its whole surface, and soonbecomes milky. If. this water is now examined with a high power, we can see in it innumerable evacuated spermato- zoidal cells. They remain at rest only a short time, when the cell-membrane swells. Finally it is torn through, and the sper- matozoid escapes into the surrounding water. The spermatozoids are comparatively very small, have a thread-like body and two long cilia; to the posterior end clings a bladder, which is lost during the swarming. In order to see them clearly, we run into the preparation a drop of 1 per cent, osmic acid, and as they are fixed beautifully by the reagent, we can now study them con- veniently. (Fig. 94, B.) We can effect the same purpose by the addition of a trace of potassium-iodide-iodine solution. Fig. 91. — Marcliantia poTymorpJm. A, an almost ripe anther- idinm in optical cross-section; p, paraphyses. B, spermato- zoids [antherozoids], fixed with 1 per cent, superosmic acid {Ax 90, Bx600). MAECHANTIA POLYMORPHA. 275 The female receptacle forms, like the male, a radially-spreading inflorescence, and in general there are nine rays, and hetween these are eight rows of archegonia on the under side of the receptacle. The distinction from the male receptacles is striking, in that here the sexual organs stand upon the under side ; but this phenomenon is connected with an early displacement of the growing point to- wards the under side of the receptacle. Under the simple micro- scope we can demonstrate that each row of archegonia, lying betAveen two rays, is enclosed in a common, veil-like covering, fringed at its edges. We prepare, between thumb and forefinger, delicate longitudinal sections through a comparatively young receptacle, and upon some of these sections find, without difficulty, the female sexual organs, the archegonia. The oldest lie nearest the edge, the ^^ounger progressively nearer the stalk. The first ripening archegonia show alongside the edge of the disk with their neck curved outwards, the succeeding ones hang straight down- wards. In an approximately ripe archegonium (Fig. 95, A) we can distinguish a short stalk, a ventral portion [the body], and a neck. The wall of the body, as of the stalk, is unilamellar. The central-cell of the body is filled by the egg [-cell, or oosphere], and the ventral canal-cell {k") cut off from it shortly before ripening. In the egg [-cell] the nucleus is readily visible. Tlie neck is traversed by the neck-canal, which is composed of a series of neck canal-cells, the walls between which are dissolved, and the disorganized contents of the four neck canal-cells are thus fused into a connected string. Between the archegonia, numerous small, leaf-like scales of the receptacle can be seen to arise. In many preparations we have in view the unilamellar veil-like covering, fringed at its edges, which protects the entire row of archegonia. jS'umerous cells of this contain oil-bodies. It is comparatively easy to see the opening of the archegonium directly under the microscope. We take quickly longitudinal sections through a female inflorescence, which has not yet raised itself, or only a little, upon its stalk, lay it dry under a cover- glass, and examine it under the microscope. When we appear to have found a ripe archegonium, and while still observing, we place a drop of water at the edge of the cover-glass. After the entrance of this, the archegonium opens almost immediately. The cause of the opening lies in the strong swelling of the contents of the neck- canal. The neck-cells separate from one another at the apex of the neck. The contents of the neck canal-cells pass out, then 276 REPRODUCTION OF tlie contents of the ventral canal-cell follow. The homogeneous portion of these contents is formed of a strongly- swelling slime, Avhich diffuses in the surrounding water; the granular contents remain in the surrounding water, where they are slowly disorgan- ized. Immediately after the ejection of the ventral canal-cell, the egg-cell in the central part of the body rounds off (Fig. 95, B). At its anterior margin [^,e.^ that in apposition to the canal], a clearer spot, the receptive-spot, is often, though not always, to be distinguished. Moreover, the penetration of the spermatozoids in- to the neck-canal can in this plant be easily ob- served. For this purpose, instead of pure water, we add to the pre- paration a drop which has pre- viously lain on a male receptacle. The spermato- zoids quickly col- lect in the slime expelled from an archegonium ; we see them enter the neck, where they become invisible. A substance is given off from the archegonium, which acts as a chemical stimulus, and determines their direction of movement. Thus they get into the slime given off from the arche- gonium, in which they slowly move in the direction of the opening of the neck. It is interesting to prove that in an unfertilized arche- gonium the neck does not close, and under such conditions the archegonium slowly decomposes. If, on the other hand, water, containing spermatozoids fs added to the preparation, and the Fig. QS.—Marchantia polymorplia. A, young; B, opened archegonium ; C, fertilized archegonium after the commence- ment of the formation of the embryo, k', neck canal-cells; fc", ventral canal-cell ; o, egg [-cell] ; pr, perianthium ( x 540), MARCIIANTIA POLYMORPIIA. 277 egg-coll becomes fertilized, the neck closes, even after a few hours, by means of a contraction proceeding from above down- wards. Keep the preparation, and after twenty-four hours the presence of a cellulose membrane around the fertilized egg [oospore] is easy to recognise. In the course of the next few days the thickness of this cellulose wall still increases. The fertilized archegonia, which we may meet with upon the longitudinal section, show a shrivelled and brown neck, while the egg [oospore] has divided (Fig. 95, G). Around the base of the archegonium, from its foot, a cup-shaped sheath, the so-called perianthium (pr) begins to develop. This soon encloses the entire swollen archegonium. Upon longitudinal sections of receptacles, which have already spread out their radiating ribs, we see the bright green, swollen archegonia, with base broadened to corre- spond, situated upon the surface of the receptacle, and decorated at the apex by the remnant of the neck. — From the fertilized egg gradually proceeds the sporogonium, which we ultimately see in longitudinal sections, prepared from still older recejitacles. The sporogone consists in a shortly-stalked, oval, yellowish-green capsule. The wall of this capsule is unilamellar ; if we spread it out with needles, and examine it with stronger magnification, the characteristic thickening rings in the otherwise thin-walled cells will appear. The yellow-walled spores are finely pitted. Between them lie narrow, long cells, tapering at both ends, and distinguished each by two brown spiral bands on its wall ; these are the elaters. The interior of the capsule is filled exclusively with spores and elaters. In capsules already opened (dehisced), we can see that this opening takes place by means of a number of recurved teeth. The elaters are strongly hygroscopic, bend to and fro with changes in moisture of the atmosphere, and so assist the dissemination of the spores. — The sexual organs are not raised upon special receptacles in all the Marchantiaceoe, and in other Liverworts this appearance is altogether wanting. On the other hand, the stalk of the sporogonium in many cases elongates con- siderably, and carries up the capsule with the spores, which assists the dissemination of the spores. The antheridia of the leaf-bearing Mosses are best examined in a ^enus which has striking male " flowers." We choose a repre- sentative of the genus Mnium, to wit the widely-distributed Mnium hormim, which in May [and June] "flowers" very freely, and bears female " flowers " and sporogonia at the same time. The 278 REPRODUCTION OF male flowers are^ it is true, mucli more striking than the female, and it is often necessary to search longer for these latter. The male flowers are dark-green, disk-shaped, surrounded bj a rosette of leaves, the so-called pericbaetium or perigonium. Towards the interior of the flower these leaves decrease rapidly in size. In the axils of the outer, but chiefly, however, of the inner, perichastial leaves, stand numerous antheridia, and paraphyses, which, more- over, spread over the entire apex of the axis. This is easily shown by median longitudinal sections of the flower, which are best pre- pared between the fingers, turning the apex downwards in cutting. On these longitudinal sections w^e see that the flower-axis broadens, after the fashion of a floral receptacle, at the place of insertion of the sexual organs, and in the middle is even a little hollowed. The central conducting bundle, peculiar to species of Mnium, has undergone a corresponding broadening, and ends in a chlorophyll- containing tissue, which spreads out under the receptacle. The antheridia and the paraphyses are at once recognised as such, and their structure easy to understand [see Fig. 95a]. The antheridia are club-shaped, shortly stalked bodies, somewhat tapering at both ends. The cells of their wall contain numerous chlorophyll-grains. Where the longitudinal section has opened an antheridium, we see that its wall is composed- of a single layer of cells. The contents [of the antheridium] consist of small, colourless cells, the partition w^alls of which in young stages of development clearly show rectangular arrangement. The ex- truded contents of older antheridia opened by the section prove to be composed of rounded cells, still " glued " together, the mother-cells of the spermatozoids, in which the thread-like body of the spermatozoid is already often recognisable. The chlorophyll- grains at the apex of ripening antheridia assume a somewhat brownish tone. Emptied antheridia are open at their apex. The paraphyses are simple cell-rows, the cells of which gradually enlarge upwards, when they are, however, at least the uppermost, again tapering ; hence the uppermost cell is always pointed. The walls of the cells are often browned in the lower part of, and not infrequently even higher up upon, the paraphyses ; they contain chlorophyll. Cross-sections through the lower part of the flower show in an instructive manner the distribution of the antheridia, their relations with the perichgetial leaves and the paraphyses, and also provide us with numerous cross-sections through the antheridia. THE TRUE MOSSES. 279 Still more striking than tlie male flowers of Mnium are the red-coloured ones of species of Polytrichum, likewise found in May [and June]. For examination we choose Pol/jtrichum juniperinum. The outer leaves forming the perichoetium, beyond their colour, differ from the ordinary leaves also, in that their unilamellar sheathing portion is continued up to the apex of the leaf. The green lameUce* characteristic of the genus Polytrichum, are found only towards the end or apical portion of the leaf, and almost always confined only to the midrib. On the rapidly-decreasing reddish - brown perichcetial leaves, near the interior of the flower, the green lamellae are de- veloped only on the outermost, sharply out- ward-bent points. The leaf thus appears ultimately reduced almost to its sheath- portion alone. The antheridia and para- physes stand in the axils of the pericheetial leaves. The middle of the flower is, how- ever, occupied by a vegetative bud, into which the central string of the stem is continued. Thence comes the later growth through the male flowers [proliferation], which is normal for Polytrichum. The antheridia have the same structure as in Mnium. The paraphyses, forming in their lower part a long cell-row, usually broaden at their tip into a sjDathulate unilamellar cell-surface. If a male flower of Polytri- chum is squeezed somewhat between the fingers, the contents of the antheridia come out as a milky slime, clearly visible against the reddish ground. [The form of the antheridium of Mosses varies veiy little, and the accompanying Fig. 95a of that of a moss especially com- mon upon shaded cinder-paths and other places where the substratum has been burnt, [Fig. 95a. — Funana hyjro- metrica. A, an anlheridiuin burstinfj; the body of the antheridium shows its wall of cells containing chloro- phyll-grains ; a, the anther- ozoids (spermatozoids) (x 350). B, the spermatozoids more strongly magnified ; h, in the mother-cell ; c, free antherozoid of Polytvichum ( X 800 ). From Trantl.] • The leaves of Polytrichum, though really unilamellar, like those of other mosses, are rendered ()pa(iue by being more or less covered by vertical green scales, or lamelht, produced upon their upper side. In P.Juntperinum, each foliage leaf shows about forty-eight such lamcllaa, running, as usual, longitudinally, and from 4 to G cells long. [Ed.] 280 AECHEGONIA OF MOSSES. as well as on old walls, viz. Funaria Jiygromp.trica, will serve to illustrate it.] The female jQowers of Mnitim Jiornum, however, are throughout not so visible. as the male, and it is often necessary to seek for them longer. The plants bearing them are far shorter than the male, and somewhat darker in foliage. The upper leaves close to- gether, after the fashion of a bud, in order to protect the female sexual organs, the archegonia. As is shown by median longi- tudinal sections, the apex of the flowering axis is not broadened to any extent, but greatly blunted, and from this we can at once assume that we have to do with a female flower, even if we do not happen at once to find the archegonia. The central con- ducting bundle of the stem is somewhat swollen under the recep- tacle, and ends, just as under the male flower, in a chlorophyll- containing tissue. The modified leaves which form the female perigynium (equivalent to male perichtetium, or, if surrounding hermaphrodite flowers, the perigamium), while remaining leaf-like, decrease in size towards the middle of the flower ; the apex of the flower is occupied by only a few archegonia, so that it is necessary to take strictly median sections in order to disclose the archegonia. The archegonia are constructed essentially like those of the Liver- worts [see Fig. 95b], but the foot-portion is far more strongly developed, only tapering a little downwards, and forms the greater part of the lower half of the archegonium. On these grounds the egg [-cell, or oosphere] apjDcars comparatively small. We must look for it close under the commencement of the neck, which here appears only a little narrower than the ventral portion. The chlorophyll contents of the cells make the archegonium anything but transparent, and hence the oosphere and the canal-cells of the neck usually need addition of potash to make them visible. In the axils of the perigynial leaves stand numerous paraphyses. Each consists of a row of short cells, swelling somewhat upwards. The lowermost cells of these paraphyses have often become brown. [To illustrate the general structure of the archegonium of the mosses, I here introduce Fig. 95b, showing its form and relations with the perichjetium in Funaria liygrornetTica.~\ [Fertilization in the Mosses takes place in all essentials aa in the Liverworts, already described.] The sporogonium, the so-called " moss-fruit," the study of which we will carry on upon the same Mnium hornum, consists of stalk or seta, and capsule. The base of the seta is sunk in the tissue of the mother-plant. SPOROGONE OF MOSSES. 281 [The result of fertilization is here, therefore, somewhat different to that in the Liverworts, and needs a few words of exphmation, further illustrated by Fig. 95c. After fertilization, the oospore develops into an embryo, an early stage of which is shown in Fig. 95c, A. This embryo develops in length both uji wards and down- H^ [Fig. 95b. — Fanaria hygrometnca. A, 1 )ngitudinal section of the summit of a weak female plant (x 100) ; a, archegonia ; b, leaves. B, an archegoninm (x 550), ventral portion with the oosphere ; h, neck ; TO, mouth still closed ; the cells of the axial row are beginning to be con- verted into mucilage. C, the part near the mouth of the neck of a fertilized archegonium, with dark-red cell-wall=:. (From Prantl, after Sachs.)] [Fig. 95c.— -1, origin of the sporogonium (//') in the ventral portion, (b h) of the arche- gonium, seen in longitudinal section (x 500). B, C, different further stages of development of the sporogonium (/), and of the calyptm (r); h, neck of the archegonium ( x about 4'>). iFrom Prantl.)] wards ; downwards it grows into a foot which, as the base of the seta, passes through the tissue of the foot or stalk of the archego- nium, and plunges into that of the apex of the moss-stem. (See Fig. 95c, B and C). Upwardly, the embryo develops into the 282 STRUCTURE OF TUE capsule, to be hereafter described. The seta remains for a long time short. Accompanying the increase in length, and likewise in thickness, of the young sporogonium, the body of the arche- gonium, which had enclosed the oosphere, also undergoes further development, keeping pace with the sporogonium in its growth, so as continuously to cover it. The upper part of the neck shrivels, as shown in Fig. 95c, B and (7, at the top. When, later in the development of the sporogonium, the seta rapidly elongates, the body of the archegonium, etc., is ruptured round its base, and is carried upwards, covering the capsule as with a cap, — the calyptra.] This calyptra, proceeding from the enlarged archegonium, which covers the growing capsule, is in Mniuin early cast off, so that it is usually difficult to find. It is split uj^ one side to its narrowed apex, and is composed of one, and in part also two layers of elongated cells. The narrowed apex ends in a brown point, which indicates the neck of the archegonium. At the base, where it was ruptured by the growing S23orogone, it appears as if cut off. The aj^ex of the capsule, denuded of its calyptra, has a cover or lid [operculum] provided with a short beak. With a needle it can be easily loosed, when the edge of the capsular urn, fringed with its teeth, comes to view. The teeth form the peristome [the form of which is an important feature in the delimitation of genera, as it is characteristic in each group]. The upper part of the seta, passing into the capsule, is called the apophysis. In the present case this last is separated from the capsule by a very slight con- striction, and is distinguished from it by its brown colour. In some mosses, the apophysis is far wider than the capsule. In order next to learn the structure of the peristome, we take a cross-section through the capsule, close under the brim of the urn, lift it up, and place it, with the teeth turned upwards, upon an object- slide. We remove the mirror from the microscope [or turn the diaphragm so that no aperture lies under] and observe the object with direct light. In this we need use only a low power. We can decide that the teeth are inserted in the inner brim, that they are wedge-shaped, and cross striate. If we breathe lightly on the object while still looking at it, we shall see the teeth curve together inwards. They are hygroscopic ; in damp weather they bend inwards, and so close the open capsule, while in dry weather they bend outwards, and again open the capsule. We count sixteen teeth on the urn. We now lay the same section in a drop of water, and, tearing it through on one side with the SPOROGONE OF MOSSES. 283 needles, spread it out flat, cover it with a cover-glass, and observe it by transmitted light, and first from its outer side. We then notice, quite at the edge of the urn, a double layer of obliquely- arranged cells, papillately prolonged, pretty strongly thickened, and containing abundant chlorophyll-grains. These cells have colourless walls, browned only at their very base, and there thev are very easily disconnected from the edge of the urn, remaining, however, connected together. By means of these cells, the separa- tion of the operculum (lid) is effected ; they form the so-called amiulus at the rim of the capsule. Now laying it with the inner side upwards, the preparation shows us that the cross-strice already noticed on the teeth are ridges projecting from their inner surface. Besides the outer peristome formed by the teeth, an inner one is also present ; it consists of the so-called cilia. Mnium kornumi has, therefore, a double peristome, w^hile there are Mosses with only one, and also without any such peristome. The cilia, like the teeth, are here flat . lamellae, which in their lower part appear divided into chambers, and in their upper part cross-striate, by slight projecting ridges on their inner surface. In the lower part they are fused together into a continuous membrane, which, between each pair of teeth of the outer peristome, is a little bulged. Two cilia stand between each pair of teeth, and present themselves obliquely from the corner. Their edges, the outer in their entii'e height, the inner only in the upper part, are fringed with small serrate projections. In these the cross-ridges of the sui-face of the cilia end. Through these serrations the pair of cilia in their upper part are combined by the outer edge, and finally the two fuse into a single narrow, elongated apex. With these pairs of cilia alternate very small ones, which, from three to five in number, stand in front of the teeth of the outer peristome. A delicate cross-section, taken somewhat deeper through the capsule, shows in the interior of this the column formed of large-celled tissue, the columella. Around this columella lies the cavity filled with spores. The inner wall of this is formed by the columella itself, the outer by a layer of tissue, usually two cells thick, and contain- ing chlorophyll, which appears separated from the wall of the capsule by a very loose chlorophyll-containing tissue. The wall of the capsule consists of two or three layers, and is covered by a sharply- defined epidermis. The cells of this latter are more strongly thickened on their outer walls. The spores contain chlorophyll-grains, their wall is bi'ownish, and studded with fine 284 STRUCTURE OF THE warts ; in favourable cases a tliree-sided pyramidal tapering of one side of the spore can be noticed. This tapering arises from the tetrahedral position of the spores inside their mother-cell ; it in- dicates the contact surfaces of its three sister-spores.* — A per- fectly median longitudinal section, which we prepare from a capsule which is still green, and provided with its lid [operculum] but is already fully formed, shows us uppermost the lid, con- sisting externally of one sheath of browner, strongly-thickened cells, and internally of many layers of thin- walled cells. At the limits between lid and urn lies the double layer of the obliquely- arranged, chlorophyll-containing cells, already known to us, by which the separation of the lid is effected. The brown cells, which adjoin the urn below, are distinguished by their small height. Similar cells adjoin these small ones towards the interior, and form thus an inward projecting ledge of thickened, brown cells, on which are set the teeth of the outer peristome. About the thickness of a cell removed arise the cilia. As the history of their development teaches us, these teeth and cilia arise by local thickening of the opposite walls of one and the same layer of cells adjoining the inside of the lid. The teeth proceed from definite portions of the outer walls, connected in the ascending direction ; their cross-ridges indicate inner adjoining cross- walls, upon which the thickening has continued for some little distance. The cilia proceed from the thickened parts of the inner walls of this same layer of cells, and bear slight ridges at the places of junction of the next inner partition walls. In our median longitudinal section the lid is hollow ; the inner tissue, after the formation of the teeth and cilia, has shrivelled up, separating from the inner surface of the cilia, which extend to the top of the lid. This tissue now forms on the columella only a projecting conical knob. The columella is visible in its entire length ; similarly we can survey the spore-sac, its outer wall, the looser tissue lying between this and the wall of the capsule, and lastly this wall. The spore-sac, so long as the lid has not been cast off, is closed above by a thin layer of tissue. Later on, it opens by the tearing of this layer. At the base of the capsule, under the spore-sac, an annular cavity has been formed. The apophysis, as is now seen, is provided with stomata, for on well- nigh every median longitudinal section, such will be cut. They lie * The contents of the spore mother-cell divide into four, which are situated as if at the four corners of a tetrahedron. [Ed.J SPOROGONE OF MOSSES. 285 below the level of the epidermis ; a pit leads down to each ; an air-chamber adjoins it intei-nally. It is surrounded by chlorophyll, containing tissue, the intercellular spaces of which communicate with the annular cavity under the spore-sac, and with the inter- cellular spaces of the entire chlorophyll-containing tissue separ- ating the wall of the capsule from the spore-sac. All the stomata are cut in the direction of their length, and give figures which, so far as can here be determined, agree with those of the Vascular Cryptogams, and of Phanerogams. This latter is so much the more striking since the apophysis (or, in other cases, the wall of the capsule as well) is the only place in mosses where true stomata, constructed after the type of the higher plants, are borne.— In [Fig. 95b. Mouth of the capsule of Fontinalis antip'jreiica, the oper- culum having fallen off. ap, outer peristome of teeth ; ip, inner peri- stome of cilia (x 54). (From Prantl.)] [Fig. 95D.—Funavia hygrometvica. A, a young leaf-bearing plant (g), with the calyptra (o). B, a plant (g) with nearly ripe sporogonium ; s, its seta; J, the capsule; c, the calyptt-a. C, longitudinal section of the capsule bisecting it symmetrically; d, the operculum ; a, the annulus ; p, the peristome ; c, c, the columella ; ?i, the air-cavity (which here extends around as well as below the spore-sac); s, the spore-layer, consisting of the primary mother-cells of the spores ( x about 20). (From Prantl.)] order to complete the impression we have obtained, let us now examine sections of the surface of the capsule and of the apophysis. We can decide that on the surface of the capsule, stomata are wanting ; between the brown-walled cells of the apophysis we see, however, pits which lead up to the stomata. If we turn the sec- tion over, and examine it from the inner side, we can in favoui-able cases distinguish the two guard- cells of the stomata, formed as in higher plants. Upon such sections we can at the same time deter- mine that the green cells between the wall of the capsule and the 286 SPOEOGONE OF MOSSES. spore-sac are joined together in longitudinal direction, that they are branched, and have all the aspect of algal threads. Moreover, on cross-sections through the apophysis stomata have usually been cut, the two guard-cells of which are not difficult to see. At the seta, the differentiation of the epidermis ceases ; its surface is oc- cupied by two or three layers of yellow to reddish-brown strongly, thickened cells, the cavities (lumina) of which, passing inwardly, become gradually larger. In the interior of the seta a central conducting bundle is differentiated. Median longitudinal sections taken near the apophysis show that these relations, beginning close to this region, are stamped upon the seta quite gradually. [The accompanying Figures 95d and 95e will serve to render more easy the comprehension of the foregoing description of the structure of the capsule in the mosses. It should be noted that the author's descriptions are confined to the Bryaceee, just as in the Liverworts they were confined to the Marchantiacea3.] NOTES TO CHAPTEE XXV. ^ Goebel, Die Miiscineen in Schenk's Handbuch der Botanik, Bd. II., p. 338. 2 Compare, A. Zimmermann, Ueber die Einwirkung des Lichtes auf den Mar- chantienthallus. Arb. axis d. Bot. Inst, in Wiirzbnrg, Bd. II., p.. 665. 3 Leitgeb, Untersuchungen ilber die Lebermoose, VI. Heft, 1881, pp. 20, 117 ; Goebel, I.e. ; Strasburger, Jahrb.f. wiss. Botanik, VII., p. 409, and Befruchtung und Zelltheiltuig, 1877, p. 12. liEPRODUCTION OF VASCULAR CRYPTOGAMS. 287 CHAPTER XXVI. THE KEPKODUCTION OF THE VASCULAE CRYPTOGAMS. Material Wanted. Fertile fronds of Scolopendrium vulcjare, the Hart's-tongue fern. Fresh. (Alcohol material in part answers.) The same of the Male Fern (Aspidlum Filix-mas). The same of the common Polypody {Foly podium vulr/are). Fresh spores of Cerato'pteris thalidroides. Prothallia of Fohjpodium vulgare. Fresh. Fructifying plant of Selaginella Martensil. Fresh, or dried. The sporangia of Ferns stand, with few exceptions, on the nnder side of the leaves. They usually form groups, which are called sori. The whole sorus is commonly covered hy a strong outgrowth of the leaf, the indusium. The indusium can be very variously developed. If the edge of the leaf turns over the sorus, we speak of it as a false indusium.-^ As an example for investigatipn, we select [the common Hart's-tongue fern] Scolopendrium vulgare. The leaf is traversed by a strong midrib, from which arise weak lateral veins, only slightly inclined forwards. In the upper half [or along the greater part of the length] of the fertile leaf the sori are formed. They retain the same direction with the lateral veins. Externally they appear more or less completely covered by two [at first] overlapping lip-like indusia, which later [are more widely separated and] spread open. It is only necessary to prepare a delicate cross-section of a piece of a fertile leaf. For this purpose we select a leaf on which the sori are already brown, but the edges of the indusium have not yet spread open. We cut with the scissors a narrow strip out of the leaf, parallel with the sorus, clamp this strip between pieces of elder-pith [or pack several such strips together, one behind the other, in which case no elder-pith is needed], and take delicate cross-sections through them. The cross-section (Fig. 96, A) through the tissue of the leaf shows us an epidermis on the upper and under side, and a 288 SPORE PRODUCTIOX spongy-parenchyma, tlie cells of which lie more densely together under the upper epidermis. [There is no palisade layer.] The apparently simple linear sori now appear divided into two. These stand right and left, inclined to one another [each in the angle between the leaf -surf ace and an indusium], and each close over a fibro-vasal bundle. The surface of the leaf at the places in question is hollowed into a furrow, and between the two sori rises Fis. 96.—Scolopendr{um vulgare. A, cross-section through the fertile part of the leaf ; i, indusium; sg, sporangium. B-E, sporangia; B and E, seen from the flanks; D, from the dorsal side ; C, from the ventral side j -F, a spore. (A, x 50; B-E, x 145 ; F, x 540.) into a ridge. The ej^idermis at the base of the furrow, studded with sporangia, impinges immediately upon the bundle -sheath. The epidermis of the under side of the leaf, and of the furrows, unite in order to pass over into the indusium (i). This begins, therefore, with a double layer of cells, which quickly passes over IN FERNS. 289 into a single one. This layer of cells has the structure of the neighbouring epidermis, except that it is wanting in stoniata and chlorophyll-grains. Yet it contains smaller colourless chromato- jjhores. From the base of the furrow arise the sporangia (sg) ; they can be seen in different stages of development ; each derives its origin from a single epidermal cell. Even with weak magnifi- cation (Fig. 96, A) we can distinguish in each sporangium a stalk and a capsule, and on older a yellow-brown ring [the annulus] can be noticed on the capsule. — For further study we make use of somewhat stronger magnification (Fig. 96, B). The stalk passes over from a single to a double row of cells. The capsule has a unilamellar wall of cells. As is shown by different views of the wall of the capsule (B-E) the annulus is composed of a row of cells, of this capsule-wall, which project outwards. These cells form a row, which, commencing at the stalk, passes over the apex, and down the opposite side, and, flattening and becoming broader, dies away without again reaching the stalk. The inner and tranverse walls of the cells of the ring are strongly thickened and browned ; the thickening decreases in the transverse walls in the direction of the outer surface. The sporangium opens between the broad cells in which the ring ends (Fig. 96, 6', -E") ; the one half of these broad cells then lies on the one, the other half on the opposite side of the fissure. The cause of the rupture lies in the ring, which in drying tends to diminish its curvature. The brown wall of the ripe spore shows a beautiful structure (Fig. F). It is covered on its outer surface with a network of cockscomb- like projections. — In Aspidiuvi Filix-vias [the " male Fern "] we find indusia, in shape between a heart and a kidney, which with age become leaden-coloured, and finally brownish, shrivel some- what, and no longer completely cover the dark-brown sori. The sporangia have almost the same structure as those of Scolopeudrmm. Upon some of them we see a short glandular hair, ending in a unicellular head, arise from the stalk. The sporangia are attached to a cushion-like prominence, a placenta, which lies over a fibro-vasal bundle. To these latter adjoin reticulately thickened tracheides, which are distributed in the placenta. At its apex the placenta bears the indusium, inserted by being curved down into the form of a stalk. — If we take a preparation in water which includes sporangia that are ripe, but still closed, and run in from the edge of the cover-glass a water-withdrawing medium, best glycerine, the sporangia slowly open before our eyes. The u 290 REPRODUCTION OF FERNS. annulus nltiniately becomes strongly concave. Then follows, with a jerk, an opposite movement, which more or less completely closes the sporangium. The entire phenomenon can in lessened degree be repeated once or several times. Careful observation shows that during the dehiscence the outer walls of the annulus project strongly into their cells. The closing movement tallies with the peculiar phenomenon that in the cells of the annulus, at the maximum of loss of sap, gas is separated out from the cell-sap. If gas has not come out in every cell, the outward curvature still continues in those in which this has not happened, which occasions the secondary movements of dehiscence. If now the glycerine is replaced by water, the gas filling the cell-cavities runs together into one gas-bubble in each, which decreases in size and at length disappears, while the sporangium almost completely closes. With renewed addition of glycerine^ the reverse phenomenon can again be produced. — It may be of interest to us also to turn our attention to the naked sori of Polypodium vulgare [the common Polypody fern] . The sori are entirely without indusia, and each one lies over the end of a fibro- vasal bundle. The placenta scarcely projects above the surface of the leaf. The sporangia are con- structed upon the same type as in the foregoing species. We select the Ferns likewise for the purpose of studying the structure of the sexual organs, and of following the processes of fertilization, in the group of Vascular Cryptogams. The pro- thallus, which is the first and sexual generation of Ferns, is always easy to produce. We obtain them by sowing the spores, or else collect fertile prothallia. In this we will confine oui'selves to the family of Polypodiaceae, the most widely-spread, and by far the richest in species. For sowing, we take the spores of Ceratopteris thalictroides, cultivated in all botanical gardens, and therefore easy to procure. [If this should not be procurable, the spores of almost any fern will do equally well.] If on the othei' hand we would collect fertile prothallia, those of any species of the Polypodiace88 will serve for examination. To find pro- thallia in the open air is attended with considerable difficulty, and wfi shall therefore do well to look for them in plant-houses. On damp shaded walls, on the stems of tree-ferns, on flower-pots we can almost always find prothallia. On the fibrous peat,^ much used now in the culture of Orchids, Sarracenia, etc., and which is often permeated by Polypodium vulgare, are usually found numer- ous prothallia of this fern, which we will here select for closer PROTRALLUS. 29] examination. As in most other Polyp odiaceoc, the prothallia of the common Polypody fern have the form of small, heart-shaped, brio^ht-green leaves, lying on the substratum. We seize a pro- thai lus of medium size, with the forceps, always taking hold of the place Avhere it is attached to the substratum, and lift it away. We immerse it in water, in which we move it for some time here and there, in order vo wash off the fragments of adhering soil, and then lay it, with the ventral side upwards, in a drop of water on the object-slide, and examine it under a cover-glass. The prothallium, as already noted, is heart-shaped. It consists of polygonal cells, containing numerous chlorophyll-bodies. In the anterior indentation lies the small- eel led meristem of the growing point. Only in its central portion is the protliallus multilamellar, as can readily be proved by changing the focus. This median portion is the so-called cushion. It passes over at the sides into the unilamellar thallus, and slopes gradually also towards the base of the prothallus. From the after-parts of the protliallus [i.e., those furthest from the growing apex] arise the root-hairs or rhizoids ; they are especially produced in the median portion of the prothallus. They are long, unicellular sacs, which soon become brown. At the edge and under side of the prothallus, individual cells, moreover grow out into short, almost without exception, unicellular papilla?, which, like the rhizoids, are cut off by a partition- wall at their base. If we have chosen for investigation comparatively young prothallia, they are male; if we have taken too old ones, they bear exclusively female sexual organs. Between these two are such as unite both sexual organs. The sexual organs, like the root-hairs, stand only on the ventral side of the prothallus. The male sexual organs, Antheridia, are found on the hinder parts of the prothallus ; they arise between the root-hairs ; but also further beyond these laterally. Their formation proceeds in the direction of the apex [acropetally] . They appear as globular arched structures (Fig. 97, A), which, in a ripe condition, contam smaller globular cells in greater number inside a unilamellar wall. On the other side [more behind] the ripe antheridia stand those which are already emptied, recognisable by the browning of their inner walls, and showing a stellate gap in their lid-cell. A full insight into the structure of the antheridia is only obtained when we examine tliem in profile. Such profile vie^s are not seldom ob- tained in many accidentally-bent parts of the prothallus; we obtain them easily, also, if we suitably bend around with needles 292 KEPRODUCTION OF FERNS. prothallia which are rich in antheridia. In correct side-views (Fig. 97, A) v,'e now readily determine that the antheridium is seated upon the middle of a weakl;^- arched prothallium cell (p), and cut off from it by a partition membrane. The wall [of the antheridium] consists, almost without exception, of two stages of lateral cells (1 and 2) and a lid cell (3). The cells of the lower stage have a broader cavity than in the upper stage or the lid. The side-view of an emptied antheridium (Fig. 97, JB) shows the lateral cells very strongly swollen ; they therefore stand out very clearly. The cavity of the antheridium is then correspondingly narrowed, the lid-cell pressed flat, and ruptured. If we now turn again to the surface-view of the prothallium, and observe an emptied antheridium from above, we can determine upon it that the lateral cells are without inner seg- mentation. Inner partition walls are in no way visible, and we come, therefore, to the conclusion that the wall of the antheridium consists of annular cells. Each stage is there- fore formed of but one ring- like cell. The entire wall of the an- theridium consists, therefore, of two such superposed ring-like cells, and a lid-cell. Annular cells of this kind are a rare phenomenon,* but constantly recur in the antheridium of the Polypodiaceae. — In general we should find similarly-constructed antheridia on the prothallia of other PolypodiaceiB. The only common departure from the form here represented, is that in which the antheridium has a lower, flat stalk-cell, and the side wall consists of only one annular cell. — If we have for examina- tion prothallia which have not been wetted for a long time, we shall not have long to wait for the emptying of individual ripe antheridia. The mechanism of the evacuation consists in the pressure which the annular lateral cells bring to bear upon the contents, besides which a swelling substance is developed between the separated internal cells of the antheridium. The lid-cell is * Annular cells are likewise met with in some Ferns in connection with the development of stomata. [Ed.] Fig. 97. — Polyvodium vulgare. A, ripe ; B, emptied antheridium ; p, cell of the prothallus; 1 and 2, lateral cells ; 3, lid or cover cell. ^ and B (x2-40). C, a spermalozoid in move- ment; D, one fixed with iodine solu- tion. CandD(x 540). A.NTHER1DIA. 293 ultimately ruptured, and the contents of the antheridium squeezed out, whereon the annular cells increase in size. The contents of the antheridium come out in the form of isolated, globular cells, tiie spermatozoidal cells, which first lie resting for a short time in the surrounding water. In each cell, even with weak magnifi- cation, is to be recognised a coiled thread, the spermatozoid [antherozoid] , and a central collection of fine granules. The walls of these cells dissolve in the surrounding water, and even in a few seconds individual spermatozoids begin to free themselves. This occurs with a jerk, whereby the coils of the body of the spermatozoid separate. One spermatozoid after another thus escapes. We follow individuals in the surrounding water, and notice that they progress comparatively rapidly, and at the same time rotate upon their axis. If a condenser is at our disposal, we proceed, by cutting off the direct rays of light by means of the diaphragm, to obtain a dark field of view. In this dark field the spermatozoids swarm about as illuminated objects.* After about 20 to 30 minutes, the movement slackens, and finally ceases. During these last stages of the movement, the form of the spermatozoids is not difficult to recognise. This is more easily attained if to the drop of water containing the spermatozoids is run in a 10 per cent, clear filtered solution of gum arabic, and so the rapidity of their movement is diminished.'^ The spermatozoid (Fig. 97, C) is composed of a band, rolled after the fashion of a corkscrew. The turns at the anterior end are narrow, but towards the posterior become broader. The anterior narrow turns bear long fine cilia. Between the posterior turns lie fine granules, and we often recognise a " vesicle " or " float " containing them. By the addition of a little 23otassium-iodide-iodine the spermatozoids are very beautifully fixed. At the anterior indentation of the prothallus, we see the female sexual organs, the arcliegonia. Nearest the indentation, they are still imperfect ; further in, are I'ipe but unopened ; finally, dead and opened, brown inside. The female sexual organs are very easy to distinguish from the male. They project above the surface of the prothallium in the form of short, cylindrical structures, curved away from the anterior indentation. This free jjortion of the archegonium is only its neck, whilst the ventral portion is found sunk in the tissue of the prothallium. At the neck we distinguish * For further information on this head, see " Dark field illumination " in any handbook on the microscope. [Ed.] 294 REPRODUCTION OF FERNS. a unilamellar wall, formed of four cell-rows, and a central canal, the contents of which, in ripe archegonia, appear granular in the central portion, and strongly refractive peripherally. This inner canal, the neck-canal, broadens upwards like a club. Below it passes into the central* cell of the archegonium, in which is found the oosphere. This last, it is true, is scarcely distinguishable. — If the protliallia had been allowed to remain dry for several days before the commencement of the investigation, we shall probably be successful in seeing the opening of an archegonium. We choose for continuous observation, an archegonium the contents of the canal of which appear strongly refractive. Often the opening results almost instantaneously ; often it is necessary to wait some time. The opening of the neck is the result of the pressure which the strongly-refractive swelling substance of the neck- canal exerts upon the wall of the neck. The four cells at the apex of the neck suddenly separate from one another, and the contents of the neck- canal pour out. The strongly-refractive substance of this diffuses as a colourless mucilage in the surrounding water, while the granular contents are gradually disorganized. The evacuation of the contents takes place interruptedly ; first come out the contents of the neck-canal, then those of the ventral canal-cell last cut off from the oosphere. Under specially favourable conditions we may now see the entrance of the spermatozoids into the archegonium. The chances of this are increased if we have placed with the older prothallium, selected for the examination of the archegonia, some quite young ones, rich in antheridia. If spermatozoids are diffused in the preparation, we see them, so long as the archegonia are closed, quietly swimming by them. If on the other hand an archegonium has opened, the spermatozoids, from a measurable dis- tance round, take the direction of the mouth of the canal, and are intercepted by the mucilage. Inside this mucilage their movement is slackened, while they retain their original direction ; they enter into the neck-canal, and reach the oosphere, into which they are taken up. As has been recently determined, here also the secre- tion of a substance from the neck of the archegonium takes place, which acts as a chemical stimulus on the spermatozoids, and determines the direction of their movement.* The specific stimu- lant in this case is malic acid, which to the extent of about 0"3 per cent, is represented in the mass evacuated from the neck of the archegonium. Thus these spermatozoids can be successfully enticed into capillary tubes, which are fused at one end, and under ARCHEGONIA. 295 tlie air-pump are injected with a fluid wliicli contains 0"01 to 0"1 per cent, malic acid, combined with any base, just the same as into the neck of the archegonium. The spermatozoids of Ferns swarm into such capillaries, likewise into large hairs, best of all those of the leaves of Heracleum Sphondylhmi [the Hog- weed, or Cow-parsnip], if these are laid, 'with their ends cut off, in water containing spermatozoids.^ For the spermatozoids of the mosses, cane-sugar is the specific stimulant, while with Marchantia, an- other, not yet determined, substance proceeds from the arche- gonium. — It has been experimentally determined^ that a single spermatozoid suffices for fertilization ; but usually several penetrate into the archegonium, of w^hich, however, only one finds admittance. These processes cannot be followed in detail, as the prothallium of Folypodium is too opaque ; they can be seen much better in Cera- topteris. We can, however, X-^KTZX A state here, that the spermato- zoids do not take their pos- terior vesicle with them into the archego- nia, but, so far as they arrive there with it still clinging to them, it is left in the mucilage in front of the opening. Now and again the number of the spermatozoids which arrive is so large that they ultimately bore in between one another, and elongating thread- like, fill up the entire canal of the archegonium, and still form a tuft before its opening. — There still remains one thing, to see the archegonia in sections. These must only be cut median, as the archegonia are found only on the median line of the prothallus. In order to facilitate the cutting, we lay several prothallia, which are carefully arranged, one upon another, after we have previously removed all grains of sand. We now find the desired structures very easily on the sections. The archegonium, as we sec (Fig. 08, A and I?), with its ventral portion sunk in the prothallium, the neck being bent. Neck canal-cell {K') and ventral canal-cell Fig. 98.— PoIy)iodii(mi;i(l3are. A, unripe arclie-oniuui ; X', neck canal-cell; K", ventral canal-cell; o, oospbere; B, ripe oi)ened archegonium (x2iO). 296 KEPRODUCTION OF SELAGINELLA. (K") are now distinguisliable ; as also the oospliere (o), together with its nucleus. The ventral portion of the archegonium has become covered by a layer of flat cells. In the ripe opened arche- gonium (B) a colourless spot, the receptive spot, can often be noticed at the apex of the oosphere, at which takes place the reception of the spermatozoids. Individual less median sections may show us the antheridia also in profile. [In order to grow prothallia from spores, as of Ceratopteris recommended above, we can sow the spores on a piece of moder- ately soft tile, laid in water in a saucer, or upon a flower-pot or flower-pot saucer similarly kept constantly moist. In a room it may be covered over with a bell-globe. In this way all the early stages of development can be well obtained, and it needs only to scrape off some of the germinating spores day by day with the blade of a pocket-knife, and lay them in water on an object-slide, to be able to follow the development. For full-grown prothallia for section-cutting, the spores can be well sown on a bed of cocoa-nut fibre refuse, flattened down in a large flower-pot saucer, with a hole in the bottom, or a seed-pan, and well drained, kept moist until towards the time they are needed for examination. Over- head watering, if needed, can be given with a spray such as is used for diffusing scent.] The Selaginelleae are heterosporous Lycopodine^ ; they possess two kinds of sporangia and spores, and we will therefore turn our attention to them, in order to complete the view we have taken of the other Vascular Cryptogams. The Selaginellesa are also known as the Ligulatae, because their leaves are provided with a small ligule at the base. .We will examine more closely the Selaginella Martensii (Sprg.), universally distributed in plant-houses. Fertile specimens are easy to recognise by the spikes which they develop on the last branches of usually numerous shoots. The vegetative body of the plant is spread in one plane ; it bears four rows of leaves in pairs, which cross one another obliquely. In each pair the upper leaf remains small, the under is considei'ably larger. The two rows of upper leaves on the dorsal surface press against the stem with their upper surface. The two rows of under leaves on the ventral surface are placed laterally, flatly spread out, with their upper surface above. The vegetative body of the plant is therefore "bilateral and dorsi-ventral ; that is, it admits only one plane of symmetry, which divides the body into a right and left half, and exhibits a ventral and dorsal surface. The fertile terminal MACROSPORES AND MICROSPORES. 297 spikes, on the other hand, are quadrangular, provided with four rows of symmetrically-arranged leaves, directed outwards. — "We next inform ourselves as to the structure of the spike, by pulling off one leaf after the other with needles under the simple micro- scope, beginning at the base. We see an ovate, somewhat flattened sporangium stand in the axil of each leaf. Even in this operation we shall liave noticed that many sporangia are larger, and show projecting bosses. If we open the large, bossed sporangia with the needles, four large spores will come into view, which completely filled the sporangium, and arched its wall out locally ; if we open a small sporangium, this proves to be filled with numerous small spores. The large sporangia are female sporangia (macrosporangia), the large spores female spores (macrospores) ; the small sporangia and spores are male, and are distinguished as microsporangia and microspores. The small spores are triangularly pointed on one side, with reticulate markings, and usually hang together in tetrads. The same relations, increasing in accordance with size, are met with on the four macrospores. We see clearly upon them the triangular tapering of one side ; in order, on the other hand, to be able to distinguish well the reticulately connected ridges on the cell-^vall, it is desirable to crush the spores. The walls of the microspores soon become dark brown, while the macrospores remain far clearer. If we examine the leaves, from which we have removed the sporangia, we see the ligule arise close under the place of insertion of the removed s^Dorangium, as a tongue- shaped membrane. A further removal of leaves from the spike shows us that the macrosporangia are far scarcer upon it than the microsporangia, and always appear preponderatingly on the low^er parts of the spike. The ripe sporangia dehisce transversely into two valves. In conclusion, it may be mentioned that the Selagiuellea^, in drying, preserve so excellently, that we can use softened herbarium specimens in order to study the growing point and the origin of the sporangia. Sections through fresh material, as well as material thus softened, can be made very transparent with potash solution. NOTES TO CHAPTER XXVI. ^ Compare Lcclcrc du Sablon, Ann. des Sci. Nat. Bot., VII. Scr., vol. ii. , p. 10, 1885. - Terre fihrexise of the Belgian nurserymen. 3 Compare Pfoffer, Vntcrs. a. d. Bot. Inst, zu Tnhiufjen, T>d. I., p. .370. 4 The same, p. 8(10. ^ The same, p. 410. 6 Strasburger, Jahrh.f. xviss. Botanik., Bd. VII., p. 105. 298 EEPKODUCTIOX OF GYMXOSPEEMS. CHAPTER XXVn. THE KEPKODUCTION OF GYMNOSPEKMS. Material Wanted. Male floTvers of Finns {e.g., the Scotch Fir, P. sylvestris). Best in alcohol. March or April. Male flowers of the Yew {Taxus haccata). Fresh, or in alcohol. March. Fertilized ovules of Yew. Fresh, or in alcohol. End of April. Young female cones of the Scotch Fir (or other Pinus). Fresh, or in alcohol. End of May. Cones of the Red Fir {Plcea vulgaris, Lk.). Fresh, or best in alco- hol. Mid- June. Seeds of the same. October. Phanerogamic plants fall into the two great divisions of naked seeded, or Grymnosperms, and enclosed seeded, or Angiosperms. These divisions are especially distinguished by the structure of the flower and the processes of fertilization and embryology, which we will first examine in the Grymnosperms. We first make ourselves acquainted with the structure of the male flower ^ of the Scotch Fir (Pinus sylvestris) . This plant flowers in May [or June, according to the district] ; but it can be investigated very well in alcohol material, which, because too brittle, should be laid, at least one day before the commencement of the investigation, in a mixture of equal parts of alcohol and glycerine. Material thus prepared can be cut much better than if fresh. — We first make out that the male flowers here stand in large numbers on the lower parts of a shoot of the same year. They are arranged according to a y^- phyllotaxy, and correspond in their arrangement exactly to the condensed shoots, each bearing two needle-leaves, which succeed the flowers in interrupted series. The flowers also, like the con- densed leafy-shoots, stand in the axils of scale-leaves. Upon the stalk of the male-flowers, we find first three decussating pairs of bracts. The low^ermost pair is placed laterally with regard to the MALE FLOWER OF I'lNUS. 299 mother axis, an arrangement which is due to the necessities of space, and which recurs almost without exception with the first pair of leaves of the vegetative buds of Gjmnospcrms. To the bracts of the short flower-stalk succeed the stamens, closely crowded, usually arranged in ten vertical rows. The floral axis is elongated, fusiform. A single stamen separated and examined under the simple microscope appears circular ; its under-side is occupied by two longitudinally-inserted pollen-sacs, touching one another in the middle line ; at its apex running out into a short, outwardly-directed border, Median longitudinal sections througli Fig. 93.— Piniis Pu inil is, rosemblinf? P. sylvestris. D, from P. sylvcstris. A, longitu- dinal section through a nearly ripe male flower (x 10). B, longitudinal section through a single staniinal leaf (x 20). C, cross-section through a staminal leaf (x 27). D, a ripe Pollen-grain ( x 400 ). the flower, shortly before the dehiscence of the anthers (Fig. 09, A), show, especially after treatment with potash, the course of the flbro-vasal bundles in the floral axis, the series of staminal leaves, each with a single flbro-vasal bundle, the insertion of the poUen- sacs on the staminal leaves. Upon less complete longitudinal sections, thinner spots can be readily found, in wliich the struc- ture of the individual staminal leaf {B) is followed still better. 300 REPRODLTTION OF GYMNOSPERMS. We then, prepare tangential longitudinal sections througli the flower, in order to obtain cross-sections of single staminal leaves, and pick out such an one for closer study (C). We see that the two pollen-sacs adjoin in the middle line, and, when perfect, are nsuallj separated only by a flat wall oi collapsed cells, in the middle of wliich may be interposed one or more layers of flat starch- containing cells. Upon their free outer surface the pollen-sacs are covered by the epidermis, to which, towards the interior, usually only collapsed cells adjoin ; towards the dorsal surface of the leaf likewise the anther cavities are closed in the same way. In the median line of the staminal leaf, above and below the partition wall separating the two pollen-sacs, runs a strip of mesophyll. The npper is thicker, and is traversed by the very delicate fibro-vasal bundle. At the two side-edges of the sta- minal leaf, the epidermis projects into a weak or more strongly- developed wing ; in the latter case, a little mesophyll can be found between the two layers of epidermis (G). On the under side of the pollen-sacs the epidermal cells diminish in size from both sides ; at the places of weakest development, the pollen sacs open. These pollen-sacs closely resemble the sporangia of Lycopodiaceae ; researches in comparative development have, in fact, led to the conception that the poUen-sacs of Phanerogams, and the micro- sporangia of Cryptogams are homologous structures. — If we look now to the pollen-grains developed in the pollen-sacs, where possible in the fresh state, we shall note that each of these consists of a central body, upon which are placed laterally two vesicles (D). If the flower is ripe, the two vesicles appear dark, because filled w^ith air. They show delicate markings npon their sur- face. The interior of the central, true pollen-grain, contains finely granular protoplasm, and a large nucleus. Shortly before dehiscence — i.e., before the opening of the pollen-sacs — a division takes place in the pollen-grain, by means of a convex partition wall (-D), which limits a lenticular cell on that side of the pollen- grain which is turned away from the place of insertion of the vesicles. This cell is best seen when the pollen-grain, as in our figure, lies on its side. An exactly similar cell is also cut ojff from the microspores of the heterosporous Lycopodiaceae, before the commencement of the stages of development w^iich lead to the formation of the antherozoidal cells. In both cases alike we can distinguish these cells as vegetative cells. The wings (vesicles) of the pollen-grain arise, as the story of their development shows, MALE FLOWER OF TAXUS. 301 rather late, and by the upheaval of the cuticle, between wliicli on the one hand, and the inner thickening layers of the wall on the other hand, a watery fluid collects. From the structure of the male flower of Finns sylvestris exam- ined above, the male flower of Taxus baccata (the Yew) differs most. This flowers somewhere in March, but by means of alcohol material we can be independent of time. The male flowers of Taxus stand in the axils of leaves on the previous year's twigs. They commence with some decussating pairs of scales, and pass over into scales arranged on a f phyllotaxy. The scales become saccossively larger, and at length follow, in quite indefinite arrange- ment, the shield-shaped staminal leaves upon the elongated axis. These, as examination with the lens will at once show, have a by no means slight resemblance to the fertile sporangiferous leaves of the spikes or cones of Equisekim. If we remove a staminal leaf with the scalpel, and examine it under the simjDle microscope, we shall find from five to seven pollen-sacs inserted on tlie inner side of the shield and its stalk. These are mounted on the shield with their base, on the stalk with their inner side. Laterally, to- wards one another, they are mostly free, and quite free on their outer surface and at their apex. We can fully inform ourselves on this point, if we further bring median and tangential longitu- dinal sections to our aid. The former show the staminal leaves and pollen-sacs in longitudinal section, the latter in cross-section. In longitudinal section the whole staminal leaf has a wedge-like outline, because the pollen-sacs broaden outwardly. In cross- section, as in longitudinal section, we see that the wall of the ripe pollen-sac is reduced to the epidermis and a layer of collapsed cells. The walls of these epidermal cells are provided with thick- ening ridges. So far as the walls of the pollen-sacs will separate from the stalk of the staminal leaf, their epidermal cells, as cross- sections teach us, show a considerable reduction in size. In order to become quite clear as to the kind of thickening of the wall of the pollen-sacs, we lift a wall from the staminal leaf with needles, and determine that they are U-shaped ridges, with which the inner and side-walls of their epidermal cells are thickened. The same thickening is present also upon the epidermal cells of the outer surface of the shields. The opening of the pollen-sacs is brought about by the wall separating from the stalk and stretching straight. The pollen-grains are ellipsoidal, studded with small knobs. Short- ly before dehiscence, a small cell is cut off from the end of the 302 REPRODUCTION OF GYMNOSPERMS. grain. In alcohol-material the contents of the pollen-grain are contracted, and unfit for examination. The pollen-grains of Taxus are without vesicular appendages to the wall ; these latter are not present in all Abietineae * ; and on the other hand are found, in the Taxineae, in Podocarpus. In many genera more than one vegetative cell is cut off from the contents of the pollen-grain, whence arise cell-masses projecting into the interior of the pollen-grain. Amongst the Abietineee the genus Tinus alone shows a single vegetative cell. The female flowers of Taxus haccata - are found, like the male, in the axils of leaves of the previous year's twigs (Fig. 100, A) ; but upon other individuals, as the plant is dioecious. The time of flower- ing, as we already know, is in March ; in alcohol the flowers pre- serve very well and can be very conveniently studied after they have been laid for at least twenty-four hours in a mixture of equal parts alcohol and glycerine. The flowers apparently terminate a small shoot, but are in reality not terminal. Not infrequently two flowers are found on the same shoot (Fig. 100, at*); in rare cases we even come across monstrosities, which show a leaf -bearing shoot developing laterally from the flower (Fig. 100, B). First we examine the flower-axis with the lens, and determine that this begins with a lateral pair of scales, to which succeed spirally- arranged scales, gradually becoming larger. The flower itself is enclosed by three decussating pairs of scales, and only its apex shows between them. This apex shows a point-like opening, the micropyle. We arrange the shoot in a definite way, in order to obtain a median longitudinal section. This must pass through the middle of the pair of scales last but one under the flower. We select for the examination a somewhat older flower, already pollinized, at about the end of April, because they are more suitable for cutting, and in many respects- also are more instruc- tive. If the direction of the section has been properly observed, the structure appears as in the adjoining Fig. 100, C. The flower does not appear to be terminal upon the primary shoot ; this on the other hand closes its development, after it has formed a secondary shoot in the axil of the uppermost scale. It is this latter which ends in the flower, after it has previously given rise to three decussating pairs of scales. Pressed on one side of the point of * AbietineaB, a sub-order of Coniferae, which includes the well-known genera or sub-genera Finns (the Piues), Abies (the Firs), Picea (the Spruces), Larix (the Larches), Cedrus (the Cedars) [Ed.J FEMALE FLOWER OF TAXUS. 303 insertion of the secondary shoot is the growing point (v) of the primary shoot (to the right in the figure). Now and then, the last scale but one of the primary shoot also gives rise to a second- ary shoot ending in a flower. Rarely, as we have seen {B), the primary shoot further develops into a leaf-bearing axis. The pairs of scales which precede the flower are to be considered as its Fig. 100.— Tcuus haccata. A, figure ot a twig with female flowers aL Lbe time of pollina- tion, at • two ovules upoa the same primary shoot. Nat. size. B, a leaf with an ovule standing in its axil; the primary shoot has further ilevelopol laterally [is proliferou8\ (X 2). C, longitudinal section through the common median plane of the primary and irccnudary shoot; v, growing point of the primary shoot ; a, commencement of the aril; e, rudiment of the embryo-sac; n, nucellus; i, integument ; m, micropyle (xl8). bracteoles; the flower itself is reduced to an ovule. Such is, for example, the terminal structure which we see at the a])cx of the secondary shoot. In the longitudinal section of this we distinguish a simple case, the ovular integument (Q, which leaves above a 304i REPEODUCTION OF GYMNOSPERMS. narrow opening, the micropyle (m) free, and in the interior the so- called nucleus of the ovule, the nucellus (n). At the base of this, only, however, in specially favourable cases, or after treatment with potash, a large cell (e) is to be recognised as the rudimentary emhryo-sac. ^ As the pollen-sac resembles a microsporangium, in the same way the ovule corresponds with a macrosporangium; as the pollen-gi-ains resemble microspores, so the embryo-sac a macrospore, Developmental researches ^ have disclosed consider- able agreements between the initiation of these structures, but have at the same time shown that a progressive reduction affects the processes which amongst Phanerogamia lead to the development of the macrospore. To compare the integument with the in- dusium of the Vascular Cryptogams offers, however, no sufficient grounds. The integument is a newly-evolved structure on the macrosporangium of Phanerogams. Upon the stalk of the ovule of Taxus can be seen a small wall of tissue (a), which for a long time, even into June, remains stationary ; later, however, begins to grow, and forms the bright-red aril, which in autumn surrounds the ripe seed. Upon the already pollinized flowers which we have taken for investigation, we can see the pollen-grains lying on the apex of the nucellus. Each of them has put out a short sac into the tissue of the apex of the nucellus. It is the large cell of the pollen-grain which grows out into the sac, while the small vegeta- tive cell shrivels. The inner wall of the pollen-grain, the intine, forms the pollen-tube, while the extine, studded with small pro- tuberances, which we have already seen upon the ripe pollen- grains, is stripped off. The pollen-grains lie in this case upon the surface of the papillose nucellar apex ; while with various other Taxinege, and their near allies, the nucellar apex is hollowed out ^ in order to receive the pollen-grains, giving rise to the so-called pollen-chamber. If we wish to know the mechanism which brings the pollen-grains to the ovule, we must make the observa- tion in the open air, during the. time of pollination.^ If we examine the female flowers at the time when the pollen-grains are being emptied from the pollen-sacs, we shall see that each flower exudes a small drop of fluid from its micropyle. In this drop the pollen-grains, carried by the wind, are caught, and in the evening are absorbed at the same time with the drop. The Scotch Fir (Pinus sylvestris) will serve as a second, and at the same time extreme, example of the structure of the female flower of the Coniferse. The Scotch Fir is monoecious, so that we FFMALE CONE OF PINUS. 305 find male and female flowers upon the same i)lant. I'lie ovules in the Scotch Fir do not stand alone, as in the Yew, but are developed in cones, in which numerous ovules, inserted upon scale-like structures, are found combined. The small cones, either singly or several together, occupy the apex of twigs of the same age. They stand in the axils of bracts like those of the condensed branches, each bearing two needle-leaves, inserted lower down on the axis ; their position at the end of the shoot corresponds, how- ever, with that of the normal twig-bearing branch. The small cones are usually in the receptive state at the end of May, and, though small, are recognisable by their brown-i-ed coloui-. They are stalked, and stand erect ; the stalk is covered with brown scales. — Hei-e also alcohol-material treated with glycerine can serve for the investigation. If we bring a portion removed from the axis of the cone with the scalpel under the simple microscope, and isolate it with needles, we can see (Fig. 101), that in the axils of delicate obovate bracts (Z>), somewhat fringed at their margin, arise scales (fr) of similar form, but fleshy, smooth-edged, provided on the in- ner side with a central projecting rib (c). These are distinguished as fruit-scales. At the base of the fruit-scale, right and left, is found on each side of the rib an ovule (if), with its micropyle turned below and towards the outer side. The edge of the integument is pro- longed at the micropyle into two lobes (m), placed right and left. Bract and fruit-scale have grown together at the base, and are therefore removed together from the axis of the cone. The cones of the Abietinece and other true cone-bearing Coniferse are conceived to be either single flowers or inflorescences, according to the significance which is given to the fruit-scale. That is, this is either considered to be a flattened, metamorphosed axial shoot [growing in the axis of a modified leaf which we have here called the bract], partially adnate to the bract ; or as a development of the placenta of a V Oi^' Fig. 101. — Pinus sylvestris. Fruit- ecalo / with its two ovules s, and tbe central rib c. Behind is the bract b. Upon the ovule the integument has grown out into two prolongations, m (x7). 306 REPRODUCTION OF GYMNOSPERMS. carpellary leaf, Avhicli Ave have hitherto called the bract. In the former case, therefore, we should have a bi- ovular branch in the axil of each bract, in the second a bi-ovular placenta on the upper side of a carpellary leaf. In the former case, the cone would therefore be an inflorescence composed of many fertile axillary branches ; in the second, the cone would be a single flower composed of numerous [open] carpellary leaves. — The remarkable structure of the fruit-scale is explained by the machinery for pollination,^ which can only be followed upon fresh material at the time of j^oUination. As soon as the male flowers begin to free their pollen, we can demonstrate an elongation of the axis of the cone, whereby the fruit-scales, together with the bracts appertaining to them, are separated. The pollen can now fall upon the lifted fruit-scales, slip down them, and, guided by the projecting rib, come between the two prolongations of the integu- ment. Later on, these prolongations curve inwards, and in this way carry the pollen into the micropyle, and to the apex of the nucellus. After full pollination the fruit-scales soon close together again by their edges, and are glued together by resin. The bracts do not further develop, nor does the central rib of the fruit-scale, which is of no further use. The red colour of the cone passes over into brown, and then into green [when ripe, again becom- ing brown], and the cone slowly sinks, and finally takes a pen- dant position. We will now turn our attention to the further changes which take place in the pollinated ovules of the Coniferee.^ With the structure of the ovule we have already become acquainted in Taxus, and have proved that at the time of pollination only the first rudiments of the embryo-sac were present. After this a further development of the ovule takes place, always variously quickly, according to the greater or less time which has to separate the periods of pollination and of fertilization. In Taxus, fertilization takes place about the middle of June in the same year ; in the Scotch Fir not until the next year, about thirteen months after pollination. In the Spruce Fir (Abies, Ficea), pollination and fertilization are separated by about six weeks only. We will consequently, in what follows, keep to the Spruce Fir, because this offers many advantages for the investigation. It would lead us too far to follow step by step the enlargement of the embryo-sac, the origin of the tissue of the prothallus (endo- sperm) and of the sexual organs in its interior, the increase in THE OVULE OF PINUS. 307 size and corresiDonding differentiation of the entire rudimentary secd. We will therefore turn at once to the stage in which the oosphcrcs are fully formed and in a receptive condition. Tliis condition, in the common Red Fir (Picea vuhjarls, Lk.), is reached about the middle of June, the fertilization is then completed in the course of a few days. Either fresh or alcohol material must be at command. For this investigation, alcohol material is better suited than fresh, as it shows the oosphere fixed. It is, above all, recommended not to lay entire cones, but separate fruit-scales, in the alcohol. Before cutting the alcohol material, it should be transferred, as we have already repeatedly done, to a mixture of equal parts alcohol and glycerine for at least twenty-four hours. — In beginning the investigation, we first inform ourselves as to the appearance of the entire scale. This is obovate, shows below, on its inner surface, the two rudimentary seeds, also already the outlines of the wings, which later on will be separated, with the ripe seed, from the inner surface of the fruit-scale. On the outer surface of the fruit-scale, and below, can still be found the bract now appearing comparatively very small. The ovule to be cut can be easily separated uninjured from the fruit - scale with the points of the needles. We prepare longitudinal sections of it between thumb and forefinger. Cuttino- is made more difficult by the integument having become comparatively hard, therefore we must somewhat modify our method of pre- paration. We cut the ovule in two with the scissors at about half its height ; we then take the upper half of the ovule, z.e., that which contains the apex of the ovule, between the fingers, and with the forceps withdraw out of the cut surface the upper part of the embryo-sac, together with the nucellus. Through these soft parts longitudinal sections can now be readily made. Staining reagents arc only to be • used with great precaution, as they stain the entire protoplasm of the oospheres, and can easily make them opaque. We first examine the longitudinal section of a receptive ovule with a low power. The entire ovule, with integument, is cut perpendicularly to its surface of insertion ; it is displayed, therefore, in median longitudinal view (Fig. 102). We see in it the integument (i), which develops into the skin of the seed, and from half its height is separated from the nucellus ; the nucellus, bearing upon its apex pollen-graius, which partly are external, and partly lie sunk in its tissue ; or may even show pollen-tubes {t), developed from these pollen-grains, which 308 REPRODUCTION OF GYMNOSPERMS. pierce the upper part of the nucellus, in order to reach the external layer of the embryo-sac ; the embryo-sac (e), of elliptic outline, filled with endosperm (or, more correctly, prothalloid tissue) ; the archegonia, here known earlier as corpuscula, whose ventral portion (a) is easy, but neck more difficult to recognise ; in the interior of each archegonium is an oosphere (o), which in alcohol material is noticeable from its yellow-brown colour, and shows a central large nucleus (n) ; and lastly, under the ovule, the commencement of the wing (s). If Ave prepare a similarly- di- rected section through a fresh ovule of the same age, we shall again find the same re- lations ; but very com- monly the contents of the archegonium will have run out. If the section has laid bare individual archegonia, without opening them, the oospheres will appear as yellowish frothy masses of proto- plasm, in which the cen- tral nucleus is scarcely distinguishable, or else, in the best cases, has only the appearance of a large central vacuole. The oospheres quickly suffer under the influ- ence of the water taken from the neighbour- hood ; if the section is to be kept for a longer time, it is recommended to use as fluid for observation white of egg diluted with water, to which, for greater durability, a little camphor has been added. ^ In such preparations the neck of the archegonium is not difficult to see. It consists of from two to four stasres of cells. Under the neck is to be found Fig. 102. — Median longitudinal section through a receptive ovule of Picea vulgaris, Lk. e. Embryo-sac filled with endosperm ; a, ventral portion, and c, neck of an archegonium ; n, the nucleus of the oosphere ; nc, the nucellus of the ovule ; p, pollen-grains upon and in the nucellar apex ; t, pollen-tubes, traversing the nucellus ; i, integument ; s, the wing of the seed ( x 9 ). THE EMBRYO OF FINDS. . 309 a small cell, which corresponds with the ventral canal-cell of the Vascular Cryptogams ; the oosphere divides, in order to form it, shortly before it is ripe. The ventral part, or body, of the arche- gonium is surrounded by a layer of flattened cells, richer in cell- contents, like to the layer which we saw around the body of the archegonium in Ferns. — In order to inform ourselves as to the number and position of the archegonia, we prepare a number of successive cross-sections through the upper part of the ovule. In this way we show that from three to five archegonia, arranged in a circle, stand in the apex of the embryo-sac. Sections which have laid bare the ajDex of the embryo-sac show us the neck of the archegonia in apical view as rosettes of six or eight cells. If our material has been gathered at the time of fertilization, we may be able to follow individual pollen-tubes to an oosphere, and find in the lower end of individual oospheres [or, as they have now been fertilized, oospores] a four-celled rosette, from Avhich four connected sacs, or tubes, can be followed into the tissue of the prothallus. The four end cells of such sacs produce the embryo [the long sacs themselves are the suspensors]. The seed ripens in October. It then easily separates, together with the wing, from the fruit-scale. The wing is developed on the inner side of the seed, between it and the fruit-scale, and the seed later falls easily from the wing, leaving behind upon this a corresponding hollow. The cells of the skin of the seed are, as cross and longitudinal sections readily show, thickened almost to the obliteration of thek* cavity. A portion of the tissue of the prothallus remains in the seed, as albumen or endosperm, densely filled with reserve food materials. It forms a sac, enclosing the embryo. This sac is open at its micropylar end, and here the radicle of the embryo is placed against the displaced remnant of the nucellus. The embryo can be easily made out in seeds cut in the direction of their length. It looks like a cylinder, gradually getting thicker towards the cotyledonary end. In consequence of being filled with reserve food-materials it is white, and as opaque as the albumen or endosperm of the seed. We prepare a median longitudinal section through the seed between the fingers, and lay it»in carbolic acid diluted with alcohol. The figure becomes very beautifully clear, far better than in potash, and better even than in chloral hydrate, so that we can follow every row of cells. We see (Fig. 103) that the cotyledons (c) do not reach quite a third of the whole length of the embryo. ' At 310 :he embryo of pinus. the base between them is to be seen the growing point \_punchtvi vegetationis] of the embrjonic stem [the plumule]. The stem (caulicle) itself, which is distinguished as the hypocotyledonary- axis, or hypocotyl (/i), passes without clear limitation into the root (the radicle). This is for the most part re- presented only by a growing apex, which shows clearly in the interior of the body of the embryo, but is in reality only the apex of the plerome {pi) of the root, while the cell-rows of the cortex [psriblem] of the hypocotyl pass directly into the parabolic layers of the rootcap (cp), a re- lation which recurs in all roots of the Gymno- sperms, inasmuch as we can see the cell-rows of the cortex of the body of the root pass over direct into the cell-layers of the root-cap (cf. Thuja, p. 186). The root-cap is traversed in the direction of its long axis by a distinctly-marked column of tabular cells, arranged in straight rows. In the hypocotyl the tissue of the pith (in) already begins to show, and around this the elongated cells of the pro cambium ring (op), in which the fibro-vasal bundles will make their appearance. These cells can be traced, moreover, for a short distance along the median section of the cotyle- dons (compare the Fig.). Thus in the embryo the essential parts of the future plant are already established. Fig. 103.— Longi- tudinal section through the ripe embryo, c, cotyle- dons ; 7i, hypocotyl ; jjI, growing ajDex of the plerome ; cp, root-cap ; cl, its central column ; m, pith ; op, procam- bium ring in the hypocotyl (x 10). NOTES TO CHAPTER XXVII. 1 Upon this compare : Strasburger, Coniferen und Gnetaceen, p. 120. Eichler, Blilthendiagr amine, Bd. I., p. 58. Goebel, Grundzuge, p. 363. 2 Strasburger, I.e., p. 2. 3 Strasburger, Anniospnmen und Gymnospennen, p. 109. * Strasburger, I.e., p. 109. Goebel, Botanische Zeituiuj, 18^1, Sp. G81. ^ Strasburger, Jfiuiische Zeltschr. f. Natuna., Bd. VI., 1871. p. 250. " 6 The same, p. 250 ; Conif. u. Gnet., p. 2G5. 7 Strasburger, Jen. Zeitschr., Bd. VI., p. 251 ; Conif. u. Gnet., p. 267. 3 Compare Strasburger, Befr. b. d. Conif. ; Coniferen und Gnetaceen, p. 274. Befrnchtunrf luid Zelltheilnng. Angiospermen und Gyni'Ko^pcrmcn, p. 140. Goroschankin: On the Corpuscula and fertilization in Gyvino^permSyin Bussiun, 1880. ^ Strasburger, Befr. b. d. Coniferen, p. 8. I'HE ANDilCECIUiM OF ANG-IOSPEKMS. 31] CHAPTER XXVIII. THE ANDRCECIUM OF ANGIOSrERMS. Materfal Wanted. Flower-buds of varions ages of the Day Lily (Hcnierocallis faloa). Fresh, or in alcohol. July. Any other large liliaceous flower will do— e.^., any Lily, Tulip, Hyacinth— thus providing fresh material for various times in the year. The same of Trad^scantia virginica (July), or of a Leucojum. Flower-buds, ready to open, of the Evening Primrose {GEnothera hiennis), Epilohium, or Fuchsia. Fresh. Flowers of Hollyhock or Mallow (July), any Curcurbit, Calhuia vulgaris (the Ling), or other heath, Azalea, or Rhododendron. Fresh, or in alcohol. Quite freshly-opened flowers of Sweet Pea, Pa3ony, or Everlasting Pea. Fresh. The male sexual organs of an Angiospsrmous flower form col- lectively the andrcecium. The individual stamen^ consists of a usually thread-like stalk, the filament, and the anther. This last is formed of two longitudinal halves [or anther-lobes], which are separated by the upper part of the filament, the so-called con- nective. It is desirable, however, to include the connective with the anther. In the tissue of each anther-lobe are usually im- mersed two compartments, or poUen-sacs. Each compartment corresponds with a microsporangium. — We first inform ourselves about the stamen of some one of the large-flowered Liliacere ; foi* example, HemerocalUs fulva [a hardy herbaceous perennial], very widely cultivated in gardens [or any of the still more universally cultivated white or tiger Lilies, Tulip, Crown Imperial, etc., will do equally well]. The yellow filament is hei-e very long, becomes thinner towards its upper end, and tapers very sharply at the place of insertion of the anther. This latter is brown [i.e., in HemerocalUs'], and movable (versatile) upon the filament. Tho connective can be followed along the outer side of the anther as a 312 REPEODUCTION OF ANGIOSPERMS. thin stripe between tlie two anther-lobes. The ripe pollen, ob- served dry upon the object-slide, shows the form of coffee-berries [the general form in Liliaceae]. It appears yellow, ornamented with a network of ridges on its surface. If, while examining, we allow water to enter from the edge of the cover-glass, we see that each pollen-grain, as soon as wetted, levels up its furrow, strongly bulges out on the corresponding side, and takes the form of a unilaterally flattened ellipsoid. The membrane of the previously furrowed part shows a relatively considerable thickness, is colour- less, has no markings, and is limited sharply against the sculptured, brownish membrane. Careful focussing of a pollen-grain in a favourable position shows us that only a single skin surrounds the pollen-grain, that the colourless part thins off at its edges and passes direct into the coloured. Between the grains in the pre- paration orange-red oil is everywhere present, and clings also to the surface of the grains, giving to them in the dry state their yellow coloration. The contents of the pollen-grain appear grey and finely granular. After a short time, during which the pollen- grain slowly and continuously enlarges, it bursts and empties its contents, in the form of a worm, into the surrounding water. In sugar solution of suitable concentration the grains round off without bursting, and can be examined uninjured. If we allow concentrated sulphuric acid to act upon the pollen- grains, the colour- less smooth part of their wall is at once dissolved, wdiile the sculptured, brownish part, on the other hand, resists : it is cuti- cularized. The cuticularized portion has therefore, in the open anther, where the pollen-grain is furrowed, to serve for the pro- tection of the entire grain. As can be seen upon the dry grains, the edges of the cuticularized parts are in contact along the fold, or furrow, so that the non- cuticularized portion lies completely concealed in the fold. It first comes into view when the grain swells, and grows out into the pollen- tube. An extine and an intine — i.e., a special outer and inner coat — is, however, as we see, not to be distinguished upon the pollen-grains of Hemerocallis, because the wall nowhere shows a double composition. Its cuti- cularized portion functionates as an extine, while the non-cuticu- larized part behaves just as does the intine in other cases. — Under the influence of sulphuric acid the structure of the cuticularized membrane is very clear. Examined from above with strong magnification, it shows a meandering network with elegantly wavy walls. In many meshes we can see lying a blue body, with STRUCTURE OF THE ANTHER. 313 irregiilai' outline, which represents the oil, previously yellow, but become blue with sulphuric acid. The cuticularized membrane itself has become yellow. If we now focus for the optical section, we recognise readily a connected basal wall, upon which are the projecting ridges^ The ridges are swollen at their outer angles, so that in optical section they appear club-shaped. In surface- view the areas at the bottom of the meshes ajipear finely dotted, and the optical section shows that these dots are in reality minute knobs, which are upon the basal wall. After some hours' action of the sulphuric acid, the membrane assumes a red-brown colora- tion, while the contents of the pollen- grain, which have come out, are at the same time stained rose-red, a re- action which proto plasm often shows with sulphuric acid.2 We now pre- pare cross-sec- tions through the anthers ; first it would be well to turn to a flower-bud only about t w o - 1 h i r d s grown, and cut cross-sections through this. The sections of the perianth are then removed from the preparation with the needles. Although we have chosen so young a flower for investigation, we nevertheless find all the pollen-sacs open. Their opening is effected very easily, and is brought about by the pressure of the razor in cutting. The adjoining figure (Fig. 104, A) will assist our con- ception. The walls of the pollen-sacs separate away (at p) from the partition walls separating the two sacs of each anther-lobe. Fig. 10 i.—IIcmcrocallis fnlva. A, cross-section through an almost ripe anther, with pollen-sacs opened by cutting; p, the partition wall between the sacs ; /, fibro-vasal bundle of the connective ; a, groove along the connective ( x 1-1 ). B, cross-section through a young anther (x 28). C, Part of the previous cross-section of a sac. c, epidermis ; /, the fibrous layer [mesotheciiun] formed later; c, the layer to be displaced; t, the tapetal layer, to be re- sorbed later on ; pm, pollen mother-cells ( x 210 ). D and E, division of the pollen mother-cells ( x 210 ). 314 REPRODUCTION OF ANGIOSPERMS. [Hence sucli an anther when ripe is commonly, by systematists, said to be " 2-celled,"] They thus reduce their curvature. The two anther-lobes are connected together by the narrow connective, traversed by a fibro-vasal bundle (/). If Ave now examine the cross-section with stronger magnification, we see most outwardly the epidermis of flat cells filled with violet cell-sap. These epidermal cells are bulged outwards. At the edges of the walls of the sacs they are rapidly reduced to a small height. Here the separation from the middle partition Avail takes place. Stomata are scattered over the whole surface of the anthers. A small air- chamber lies under each of these. To the epidermis folloAvs, on the wall of the sac, a single layer of comparatively high cells, with annular thickenings, the so-called fi.brous-layer [or mesothecium]. The rings on these cells are arranged perpendicularly to the surface; they pass over partially into spiral thickenings, and branch frequently into a network. Towards the dorsal side of the anther the Avails of the sac become gradually thicker, the fibrous layer being doubled. The remainder of the body of the anther is likewise constructed of fibrous cells. Only the cells Avhich sur- round the fibro-A^asal bundle of the connective, and those (p) Avhich form the partition Avail between the pollen-sacs, are without thickening ridges. In order to prepare surface-sections of the anther, we again select a flower-bud about tAvo-thirds dcA^eloped. The surface-sections show that over the sacs the epidermal cells are longitudinally, the cells of the fibrous layer, on the other hand, are transversely, elongated. Not so on the dorsal surface of the anther, Avhere the fibrous cells appear more isodiametric. Over the sacs the thickening ridges on the outer side of the fibrous cells are weaker, often scarcely recognisable. In drying, the cells of the fibrous layer contract in tangential direction transverse to the long axis of the loculus ; they are hindered in radial contraction by the thickening ridges. On the outer surfaces, Avhere the thick- ening ridges are weak, the contraction of the fibrous cells has more effect, whence the outward curvature which results in the rupture of the loculi.'^ — Often in Angiosperms, as in Taxus, the thickening of the outer surface of the fibrous cells on the Avail of the sac ceases entirely, so that the thickening ridges show U-shaped or basket-shaped figures open toAvards the exterior; it is clear that such a disposition assists the Avail of the sac in becoming concave on its outer side, — In order to study closely the relations of the fila- ment Avith the anther, Ave j^repare a median longitudinal section DEVELOPMENT OF POLLEN-GRAINS. 315 which falls between the two anther-lobes, through the upper part of the stamen. We see the filament thin off very strongly at the point of insertion of the anther. Its bundle enters into the con- nective, and passes through it, gradually becoming attenuated, almost to the apex of the anther. The non-fibrous cells, sur- rounding the fibro- vasal bundle, which we saw in the cross-section, can likewise be followed out of the filament into the connective. In order to obtain closed pollen-sacs in cross-sections, we must go back successively to younger and younger flower-buds so long as it proves necessary (Fig. 104, i>). Now prepare cross-sections through a flower-bud about J inch high, and we shall find the walls of the sacs consisting, besides the epidermis (Fig. 104, C, e), of two or three layers of flat (/, c), and one layer of radially elongated cells (t).^ These last surround the entire sac. The interior is filled with iDolygonal pollen mother-cells. • If we next prepare cross-sections through a flower-bud about f inch in height, we shall see the pollen mother-cells already isolated and in course of division. These pollen mother-cells are recognisable by their white, thick, strongly refractive Avail ; their contents are divided into two, or already into four cells, which lie in one (Fig. 104, D), or in two planes at right angles (Fig. 104, E). These pollen- grains, therefore, like spores, are produced by quadri- partition inside their mother-cells. The wall of the anther is lined by tapetal cells, which are filled with yellow-brown contents. These proceed from the innermost lnjer (t) clothing the sac. In the next older flower-buds the walls of the pollen mother-cells are dissolved; the young pollen-grains lie free ; the tapetal cells have for the most part lost their independence, their contents have penetrated between the young pollen-grains. The layer of flat- tened cells (/) underlying the epidermis has strongly developed, and forms the fibrous layer, while the next inner layer is crushed and disorganized. Ultimately, as still older buds show, the un- consumed portion of the tapetal cells, especially in the periphery of the sac, takes on an intense yellow-brown coloration, a glis- tening oily appearance, and so forms the oily substance which clings around and upon t'le pollen-grains. The species of the genus Lilium agree with IlemerocaUis. The processes of differentiation in the anther commence here, however, later. In flower-buds of the white Lily, Lilium caiididiim, of L. croceum and others, four-fifths of an inch high, the pollen mother- 316 REPRODUCTION OF ANGIOSPERMS. cells first begin to divide. In cross-sections tlirougli fresli flower- buds the large tapetal cells are very striking from the yellow - brown coloration of their contents. The hypodernial cells, as well as all the others which are later on provided with thickening ridges, are densely filled with starch-grains. Funhia ovata [May] provides likewise a very favourable object for investigation, and agrees with Heinerocallis and Liliuin, as also do Agapanthus umbellatus [Grreenhouse, April] and many others. Tulipa [the Tulips, April, May], and Hyacinthus orientalis [the Hyacinth, January — ^lay] are likewise good to use. In Tulipa the filament under the anther tapers so sharply that this latter will draw oif ; in Hyacinthus the anthers are almost sessile on the perianth segments. Tradescantia virghiica does not cut so well, but we examine them with respect to their pollen-grains. Cross-sections through flower- buds which have attained about two-thirds of their definite length, show us the two halves of the anther separated by a connective elongated somewhat considerably in cross-direction. The walls of the sacs are already reduced to two layers, and the thickening ridges already formed in the inner layer. The young pollen-grains lie embedded in a yellow-brown substance, the origin of which from the tapetal cells is already known to us. The partition wall between the two sacs of each anther-lobe is here strongly devel- oped, and projects so far, that externally scarcely any depression between the two sacs is to be seen. At the place of insertion of the walls of the sac upon the partition- wall, the fibrous layer ends suddenly, and here also the dehiscence takes place later on. Sur- face examination of the walls of the sacs shows in this case also a longitudinal direction of the epidermal cells, a radial direction of the fibrous cells, and an almost complete absence of thickening ridges on the outer wall of the cells. If we examine with a lens the stamens of a bud which is ready to expand, we shall see the beautiful sulphur-yellow anthers fixed upon violet filaments covered with violet hairs. The dry pollen- grains are folded (or grooved) on one side (Fig. 105, A). In water the fold is levelled out, and the grains become almost ellipsoid ; but the side which was furrowed is more strongly bulged. Its membrane is decorated with fine sinuous lines ; the furrowed side also shows this structure, and is distinguished Only by its somewhat brighter coloration, and somewhat weaker cuticulariza- tion. In the finely granular contents can be distinguished two STRUCTURE OF POLLEN-GRAINS. 317 brighter homogeneous-looking spots (13). These are the two nuclei, of which the one appears worm-shaped, and the other elliptic. The other contents of the pollen-grains are pretty uni- formly finely granular. The pollen-grains after some time begin to flatten, whereby the nuclei, together with the contents, are pressed out. The two nuclei can be seen very beautifully if the pollen-grains are crushed in a drop of acetic methyl-green, or acetic iodine-green. The worm-shaped nucleus stains more deeply, and in coming out often elongates considerably. If the pollen- grains are placed in the reagents in question, but without crush- ing, the nuclei show in A their natural position in- side the grain, the worm- shaped nucleus always staining very strongly, the elliptic, on the other hand, somewhat more weakly. The rest of the pollen-grain remains at the same time unstained. If the pollen-grains in water have a drop of po- tassium-iodide-iodine solu- tion added, we see, after crushing the grain, numerous small blue starch-granules in the extruded yellow-brown contents. — If we go back to the younger flowers, and remove the anthers from a bud about J inch long, and crush them in water, we shall see part of the pollen-grains with one nucleus, part, as in Fig. 105 C, where two nuclei lie close together. These two nuclei are, however, separated by a curved partition wall, which encloses one nucleus together with a little protoplasm. This cell, in basal outline almost circular, lies always upon the flatter side of the grain, which later on is opposed to the fold, or furrow. In somewhat older flower- buds we can see that this cell has separated from the wall of the pollen-grain, and lies free in the contents of the grain. It has elongated, and correspondingly thinned, and at the same time tapered at both ends ; with the exception of the two ends, it is filled by its nucleus.^ In pollen-grains which are almost ripe, the special boundary around this nucleus has disappeared ; it lies therefore completely free, and has elongated still more vermi- formly. Comparison with the Gymnosperms induces us at first to Fig. 105. — Tradcscaniia Virginica. J, pollen grain dry ; B, in water ; C, young pollen-grain in water, showing the reproductive and vegetative cells (x 510), 318 KEPRODUCTION OF ANGIOSPEEMS. consider the small cell as the vegetative one ; in point of fact, however, it is the generative cell, and it is its more deeply stain- ing nucleus which effects fertilization.^ — These observations can, as far as the youngest stages are concerned, be carried on in pure water ; for the oldest stages we must bring acetic methyl-green or acetic iodine-green to our aid. — The species of Leucojum [Snow- flake] agree Avith Tradescantia. [The Crown Imperial, Fritillaria imperialis, presents another favourable object for investigation. In sections of alcohol material, first placed in water, and then stained with logwood, the two nuclei in each pollen-grain can be seen very beautifully.] If we open a bud of (Enothera biennis [the Evening Primrose] which is ready to expand, we shall find that the anthers have already dehisced and set free their pollen. These latter are suspended between the anthers by cobweb-like threads. If such threads are stretched out upon an object-slide, they appear under the micro- scope as exceedingly delicate threads, partly stretched straight, partly tangled. The pollen-grains in a dry state are opaque, but their three-cornered form is at once noticeable. In water, with stronger magnification, they show as flattened, symmetrically triangular bodies, with rounded projecting angles. At the base of each of these rounded corners an annular thickening of the mem- brane of the pollen-grain is to be seen. The contents of the pollen- grain appear finely granular; the two nuclei are only recognisable in the contents of the ripe grain with extreme difficulty. In sulphuric acid the pollen membrane assumes a red-brown colour. By it an outer, thin, yellow-coloured layer is raised from the body of the pollen-grain, forming folds upon an inner, thicker, red-brown layer. Both layers coalesce in the wall of the corners. From the side walls of the corners fine teeth project inwards, so that these walls appear as if porous. The apex of the corners is dissolved by the sulphuric acid. The fine threads, binding the pollen-grains together, resist water, sulphuric acid, and potash, and are insoluble also in alcohol. If the grains are treated with 25 per cent, chromic acid, their membrane is soon dissolved, and in all cases the strongly cuticularized portion somewhat earlier than that which is either not, or but slightly cuticularized, which remains for a time as colourless swollen caps on the projecting corners of the plasmic contents. Later on, these also are dissolved ; and ultimately even the cobweb threads between the grains do not resist the chromic acil. — From the stigma of an older flower pollen-grains can be STRUCTURE OF POLLKX-GKAIXS. 819 lifted off which have already developed pollen-tubes. The forii'- ation of tubes takes place commonly only at one corner, or else, of the tubes formed, only one further develops. The membrane of the tube is continuous with the side-walls of the corner ; a specially limited intine is not present.^ [Longitudinal sections of the arms of the style of flowers which are past their prime, will show the pollen-tubes developed from the corners of the pollen-grains, traversing the tissue of the style. With magenta the contents of the pollen-grains and tubes will com- monly stain deeply, the tissue of the style only very slightly ; and hence the pollen-tubes can be traced with great ease, and often in large numbers.] Instead of CEnothera, an Epilohium [Willowherb] or a Fuchsia can be used for the investigation. We will consider still some other pollen-grains of specially characteristic form. The Malvacea? are distinguished by remark- ably large pollen-grains ; we examine those of [the Hollyhock] Althcea rosea. In water these appear globular, opaque, studded with colourless spines. They become very beautifully transparent in carbolic acid and in chloral hydrate, much less so in oil of cloves, still less in oil of lemon. The preparations are best in carbolic acid, so that we will keep to this. Their surface view shows us that the colourless membrane is studded, at approxi- mately equal distances, with large pointed spines. Between these are scattered others, short, blunt, of variable thickness. Regularly distributed circular openings, appearing rose-coloured, traverse the membrane. The basal surface of the membrane is finely punctate. The contents of the pollen-grain appear uniformly finely granular, the nuclei are very difficult to distinguish. The optical section of the grain shows us clearly the foi-m of the large and small spines, and of the canals penetrating the membrane. An exceedingly delicate, but nevertheless existing, intine can be traced only as a boundary to the contents ; it bulges a little, papilla-like, into the canals of the extine. In concentrated sul- phuric acid the extine is quickly stained red-brown, and its struc- ture is then very cleai-ly shown. The pollen-grains of most other Malvacea) resemble those of Althcea. In Malva crispa, a [hardy annual] sjjecies not infre- quently cultivated, for example, the pollen-grains are shaped just as in Althcea, excepting that the spines on the membrane are all alike ; between the spines lay scattered the pores or canals ; the membrane appears besides finely punctate. 320 EEPrvODUOTION OF ANGIOSPERMS. The lai'ge pollen-grains of the species of Cucurhita have always been specially noticed on account of the valves which close the places of egress in the extine. In water, yellow oil-drops come off from the surface of the extine, the grains soon evacuate their contents, and the structure of the membrane then becomes clear. The extine is studded with regularly distributed large spines, and between them very numerous small ones. The places of egress are round, the valve is lifted up either on one side or altogether, by the j^apilla-like bulging of the intine. The valve has the struc- ture of the surrounding extine, and bears one or more spines. Very good figures are obtained in oil of lemon, less useful in oil of cloves. On the other hand the figures in chloral hydrate are to be preferred to those in carbolic acid. In a word, the most favourable clearing medium for each object must be found by experiment. U23on the preparations in oil of lemon and chloral hydrate we can determine, by optical sections, the position of the valve inside the extine, in which it is found wedged with its base somewhat broadening inwards. Under the valve can be seen the bulging of the intine. In suliDhuric acid the oil-drops on the extine become blue. The extine slowly becomes brown. The valve is thrust off by the swelling contents. In 25 per cent, chromic acid the entire pollen-membrane is soon dissolved; the intine resists somewhat longer, and, at the moment when the extine disappears, can be followed as a strongly swollen, homogeneous membrane. The pollen-grain has previously emjDtied itself, whereby the obser- vation of the intine is considerably facilitated. In sulphuric acid, on the other hand, the intine is immediately dissolved, the extine remains, the extruded contents of the pollen-grain are gradually, as in other cases, coloured rose-red. Of compound pollen-grains, which occur alike in Monocotyledons and Dicotyledons, we will first take those of [the Ling] Calluna vulgaris. The grains here are joined in fours, and usually grouped tetrahedrally. The pollen-membrane shows only slight protuber- ances, and usually three places of egress for each grain. The various species of Erica, Azalea, and Bhododendron agree in all essentials with Calluna. In species of Acacia, as in the Mimoseas generally,^ the pollen-grains form groups of 4, 8, 12, and 16, and even more cells, but can occur also separately. In a from 3 to 30 per cent, sugar solution, which contains 1"5 per cent, gelatine, most pollen-grains easily put out tubes, in which protoplasmic movement is beautifully seen. The formation of tubes NOTES. 321 takes place quite certainly and rapidly in 5 per cent, solution of sugar and 1*5 per cent, gelatine from the pollen-grains of Pceonia, Staphylea, and even from Tradescantia when the pollen-grains are taken from freshly oiDoned flowers. The most favourable objects are perhaps species of Lathyrus \_e.g. Sweet Pea, Everlasting Pea, etc.], in 15 per cent, solution of sugar and 15 per cent, gelatine. This solution must be freshly prepared, the sowing is best per- formed in a suspended drop in a moist chamber (see p. 238). NOTES TO CHAPTER XXVIII. ^ On Stamen and Pollen compare v. Mohl, Ueber den Bau und die Formen der Pollenkurner, 183J:. Fritsche, Ueber den Pollen, Mem. de sav. etrang. 1836. Naegeli, ZurEntwickliings. d. Poll, bei den Phanerogamen, 1842. Schacht, Jahrb.f. wiss. Bot., Bd. II., p. 109. Warming in HansichVs bot. Abh., Bd. II., Heft. II. Strasburger, Befr. und Zellth., p. 15, and Bau der Zellhdute, p. 8G. Elfviug, Jen. Zeitsclir. f. Natunv., Bd. XIII., p. 1 [trans, in Quart. J. of Mic. Science, 1879]. Goebel, Grundz, der. Sijst. Bot., p. 398. Luerssen, Grundz. d. Bot., III. Aufl., p. 359 ; Med. Pliarm. Bot.. Bd. II., p. 198. Prantl, Lehrh. der Bot., HI. Aufl., p. 192 [English trans., by Vines] . - Sachs, Bot. Zeitung, 1862, p. 242. ^ Compare Leclerc du Sablon, Annales des sc. nat., Bo'aniqne, VII. Ser., Vol. I. p. 97, 1885. * Warming, in Hanstcin's bot. Abh., Bd. II., Heft. II. Goebel, Gnnuhiige, p. 409. 5 Compare herewith, Elfving, Jenaische Zeitsch., Bd. XIII., p. 12 [EngHsh translation as above] . ® Strasburger, Neue Untersuchiingen ilber den Befruchtungsvorgang bei den Phanerogamen, 1884, p. 5. The differential coloration of vegetative and generative nucleus is in general more strongly marked than in Tradescantia. ' Strasburger, Baa der Zellhdute, p. 95, where the development is given. s llosanoff, Jahrb.f. wiss. Bot., Bd. IV., p. 441. Engler in the same, Bd. X., p. 277. The other literature is there given. 322 THE GYNJICIUM OF ANGIOSPERMS. CHAPTER XXIX. THE GYN^CIUM OF ANGIOSPERMS. Material Wanted. Fading flowers of Larkspur, Delphinium Ajacis. Fresh ; or in alcohol. Or, the same of the Hellebore, Hellehorus sp. The same of the Flowering Eush, Butomus umhellatus. The same of some liliaceous plant, e.g., Tulip, Hyacinth, Lilium, Hemerocallis, Yucca. The same of a Primula, e.g., Primrose, Cowslip, Auricula, Polj'anthus, or of Lysimachia, or Anagallis. Withering flowers of an orchid. Full-blown flowers of Monkshood {Aconitum Napellus, etc.). The same of the Bird's-nest Rape, Monolropa hypopitys. Fresh only. Or, the same of some sp. of Pyrola (Winter-green). Or some orchid. Or Gloxinia. Fresh. Alcohol -material will do. Full-blown flowers of Torenia Asiaiica. Fresh. Alcohol-material will do. The same which we have ourselves pollinated from 36 to 48 hours previously. Let us first obtain a general idea of the structure of the Ovary. ^ For this purpose one of the Ranunculacese is very well suited, e.g. Delphinium Ajacis, the Larkspur of the gardens. We choose an old flower, from which .the petals and stamens are easily removed, and observe three pistils * left standing in the central position. Even by superficial observation we can distinguish upon the pistil the lower, green, swollen portion — the ovary, and the thin part, here rose-coloured, into which the ovary narrows above — the style. This last ends with the stigma, which in this case is not specially * It is high time an end was put to the confusion existing in systematic works as to the use of this term. A pistil, I cletine as a distinct ovary, of one or more cells, and composed of one or more conjoined carpels, with its stigma or stigmas, and, if present, style or styles. A gyncBcium is the whole female part of the flower, consisting of one or more such pistils. Pistil and gynsecium may thus be synonymous, as when there is but one pistil, but are not necessarily so. This may not be throughout consistent, but it is at least clear. [Ed.] STRUCTURE OF A CARPEL. 323 delimited, and merely ends the style. — We now prepare cross- sections through all three ovaries together, and examine them with a low power, or with the addition of a little potash. The cross- section (Fig. 106) shows us a single cavity [cell or loculus] in each ovary. Apparently it is a single fertile leaf, or carpellary leaf, which forms each such ovary. We can conceive the car- pellary leaf folded iuAvards and its ^dges here grown together. To such an origin points, moreover, the ventral suture,* which we find, in fact, in the median plane of the ovary on its side turned towards the middle of the flower. Such an ovary, composed of one fertile leaf, is monocarpellary ; when a number of such mono- carpellary ovaries are combined in a single flower, as is the case in our example, the flower is said to be [apocarpous or] poly- carpous. The ovaries are here free to their base, and inserted upon the floral axis, i.e., they are superior [or free]. The entire female sexual apparatus of the flower may consist of one or of numerous pistils, and is designated the gynaBcium. Our cross- section shows clearly the groove on tlie ventral side ; and with stronger magnifica- tion we can, at this place, follow the ex- ternal epidermis through the entire thick- ness of the wall, and see it continue into the epidermis of the ovarian cavity. It f.g. \q6.— Delphinium A}a. is interesting that this inner epidermis '''• Cro.s-secticn through an '^ ^ ovary, o, wall of the ovary ; also possesses Stomata. The wall of the v, fibro-va?al bundles in the ovary is traversed by a number of fibre Te^U .tuoTuLl'.x™^: vasal bundles, of which most appear on the dorsal side, and some near the edges of the carpel on the ventral side. The edges of the carpellary leaf are a little swollen, and form, on the cavity side, the placentae {p). From these the ovules (s), corresponding with the number of placenta}, arise in two rows. With the ovules we shall concern ourselves later on, and for this purpose we put our preparation on one side. [Instead of Delphmium, and available so early in the year as * While there can be no sound objection to the use of the term vvutral suture to imply this miion of the margins of the leaf, there is uo reason what- ever for retaining the absurd term dor.tal suture for the midrib of the car- pellary leaf. The tei-m " dorsal " is however of advantage in connection with dehiscence, and it would greatly facilitate a much-needed simplification of the methods of classification to sjjeak, for instance, of the " ventral " dehiscence of a follicle, the " dorsi-ventral " dehiscence of a legume, and so on for the various kinds of capsule. [Ed.] 324 EEPRODUCTION OF ANGIOSPEEMS. February, we can use the ovaries of tlie stinking Hellebore, Helleborus foetidus. They agree in all essentials with the above. Sections taken through flower-buds at various stages will show moreover the method of union of the margins of the carpellary leaf. Some of the sections will show these margins unjoined but in contact ; in others, the epidermal cells of one thickened margin will be seen to grow out, papillately, between the similar cells of the other margin, so as to "dovetail " the edges together. Sections of the same young flower-buds may show the division of the pollen mother-cells into tetrads. For earlier examination still the Christmas Rose, S. niger, may be used.] In the flower of [the Flowering Rush,] Butomus umhellatus, as in Delphinium, we find a number of ovaries — always six ; but these ovaries are free only in their upper half ; in the under half they have grown together laterally, and cannot be isolated uninjured. The style is very short, and bears the stigma on its upper edge. We prepare cross-sections through the free and the combined portions of the ovaries. The figure of the free upper part is, from the point of view of the carpellary leaf, the same as in Delphinium ; the individual carpellary leaves remain distinguish- able from one another to their base, but in the lower part it is no longer possible upon the cross-sections to separate the individual carpellary leaves intact. In Butomus we have therefore an inter- mediate stage between apocarpous and syncarpous flowers ; and this example is suitable to introduce us to the multilocular ovary, composed of more than one carpellary leaf. Besides this, we find another novelty in Butomus. The ovules arise not only from the edges, but rather, the median line excepted, from the whole inner surface of the carpellary leaf ; they have superficial placentation. The entire walls are covered with ovules, and functionate as placenta. At the place of inserbion of each ovule a fine fibro- vasal bundle is to be seen, which provides for the ovule. They are branches of the stronger, larger, fibro-vasal bundles, lying deeper in the tissue. The ovule of Liliaceae is on an axile placenta ; we select, with like results, the Tulip, Hyacinth, a Lily, or Hemerocallis for examina- tion. In the Tulip the three stigmatic lobes are sessile upon the ovary, without style. In the Hyacinth the style is short, the stigma small, slightly trifid. In Lilium the style is long, the stigma tripartite. In Hemerocallis the style is very long, the stigma likewise tripartite, but very small. Cross-sections show STRUCTURE OF SYNCARPOUS OVARIES. 325 us a trilocuLir ovaiy, composed of three closed carpellary leaves which have grown together. Here neither laterally nor in the middle is a limit between the tissues of the individual carpellary leaves to be recognised ; and a single continuous epidermis covers the exterior of the whole structure. [In Yucca, another genus of the Liliace^e, the limits of the three carpellary leaves are marked externally by grooves, and internally by narrow radial pear- shaped cavities in the tissue, each cavity being lined by a distinct epidermis.] Three carpellary leaves therefore form here a syn- carpous, trilocular ovary. Each of the three carpellary leaves combined into this trilocular ovary bears, corresponding to its two edges, tw^o rows of ovules ; i.e. the placentse lie here in the inner angles of the loculi or cells of the ovary. The placentation is therefore marginal, as in Belpliinium ; but as they arise from the angles of the cells, and therefore in the centre, it is specially designated central [or axile] placentation. — Cross-sections through the style of Hemerocallis show us in it a central triangular passage, the pollen-canal. Three fibro-vasal bundles are distributed at the three angles of the pollen-canal. A longitudinal section through the apex of the style, and therefore through the stigma also, shows us the surface of this latter grown out into long papilla. This phenomenon is very general upon stigmatic surfaces ; Hemerocalhs however offers still another interesting condition, in that the cuticle of the papillae is raised up by the formation of slime or mucus. This cuticle is spirally striate, and in accordance with this its upheaval follows a spiral line. At length the cuticle is entirely loosened from the inner layer of the membrane, and disappears from the papillae. — The other Liliacea3 likewise show a hollow style ; in most cases, on the other hand, the style is solid, but filled either with cells easily passing out of lateral union, or else provided with swollen side- walls, between which the pollen-tubes can easily grow downwards. Another free or superior ovary exists in the flowers pi the species of Primula [Primrose, Auricula, Cowslip, etc.]. These are dimorphous, i.e., have short-styled and long-styled ovaries, and stamens inserted high, up or low down upon the tube of the corolla. A median section taken through the ovary shows us that the floral axis is prolonged into the cavity of the ovary, and hero enlarges into a mushroom-like swelling. In the middle this swelling projects, papilla-like, into the pollen-canal of the style. The entire surface of this swelling is covered with ovules. \^ e 326 REPRODUCTION OF ANGIOSPERMS. liave here a free-central placenta. The wall of the ovary is in no way connected with this placenta. We can be quite convinced of this by cross-sections in which the wall of the ovary appears as a free ring around the central placenta [or, if an equatorial incision is made round the ovary, the style and upper part of the ovary can be lifted off like a cap from the mass of ovules, and the prolongation of this central swelling will be withdrawn from the pollen-canal.] Wanting also in the ring are the points of se- paration which enable us to determine the number of carpellary leaves concerned in forming the ovary ; these however are assumed to be five, from the point of view of the numerical symmetry of the other floral parts, and from the circumstance that in many Primu- laceae the fruit-capsule dehisces at its apex with five teeth. In Primula itself the number of the teeth with which the capsule opens is undetermined. — Instead of Primula, species of Lysimachia [Loose-strife, Creeping Jenny, Money-wort, etc.,] or of Anagallis [Scarlet Pimpernel, Bog Pimpernel, etc.,] can be used with the same results ; they all bear their ovules on a free central placenta. Let us examine now an inferior [adherent or adnate] ovary, selecting first that of Epipactis palustris, or of some other orchid. The brown ovary lies below the point of insertion of the other floral parts. We select for cutting a young rudimentary fruit, upon which the petals have already begun to go brown. Cross- sections are very instructive, and show us a unilocular ovary, which bears equidistantly upon its wall three pairs of placentas. The placentas repeatedly divide at their inner edges, and bear a great number of ovules. The wall of the ovary has on its outer side six projecting ribs, of which three correspond to the places of insertion of the placentae ; three specially strong ones alternate with these places of insertion. Each rib is traversed by a fibro- vasal bundle, or a complex of fibro-vasal bundles, besides which a small bundle lies at each place of separation of two placentas. In a superior ovary, the cross-section of which should agree completely with that here described, we should in no way scruple to consider the ovary as composed of three carpellary leaves, and to look upon the pairs of placentce as arising from the conjoined edges of two adjoining carpellary leaves. The three ribs which alternate with the lines of insertion of the placentae, we should take for the midribs of the three carpellary leaves. As we have here, however, an inferior fruit, the matter is not so simple. We can -either conceive that the inferior ovary consists 'STRUCTUKE OF AN INFERIOR OVARY. 327 of the liollowed floral axis, and is only closed above by the carpelJarj leaves, that from these latter, however, the placentae grow downwards into the hollowed floral axis ; or we can consider that the carpellary leaves and hollowed floral axis have grown together, and that in the wall of the inferior ovary, therefore, the outer portion appertains to the stem, the inner to the carpel- lary leaves. This latter theory is decidedly to be preferred ; it has, however, no other than a ph3dogenetic [or evolntionary] value ; i.e., we conceive that in course of time the inferior ovary has so arisen.* In point of fact, however, the anatomical and physiological data for such a conception are wanting, and we must therefore be contented with stating that the structure of this inferior ovary is not different from that of a polycarpellary, unilocular, superior ovary. — If ripe fruit-capsalcs of Epipadis are at our disposal, we shall find in these, as in most other Orchidece, that the wall of the capsules dehisces by six longitudinal clefts. The six bands separating the clefts remain joined at the base and at the apex of the capsule. Three of them are broader and fertile, three are narrower and sterile. The three sterile correspond with the three mid-ribs, which we saw in the cross-section of tlie ovary ; the three fertile bands bear in their middle the placentas. We will now endeavour to become acquainted with the struc- ture of the ovule, and at the same time turn our attention to the processes of fertilization in Angiosperms. In order to become acquainted with the individual parts of the ovule, we first prepare cross-sections through the ovary of [the Monkshood] Aconihim Napellus., or of some other species of Aconitum. We select a flower in full bloom, strip off the other parts of the flower, and then cut through the three ovaries together. Care should be taken that the sections are taken correctly at right-angles with the long axis of the individual ovaries. The number of the sections must be con- siderable, so as more probably to cut an ovule correctl}'. We glance over the sections, and select those which appear likely. In case the section is not delicate enough, we can help matters with a little potash. The figures are almost identical with those which we have just examined in Delphinium ; but in the structure of tlie integument of the ovule there is a slight difference, whicli in- duces lis to give Aconitum the preference now. If an ovule is cut centrally, it appears as in the adjoining Fig. 107. The ovary * A plant of great interest in this connection is the Californiau poppy, Esclischoltzia Califoniica, now so commouly grown in gardens. [Ed.J 328 REPRODUCTION OF ANGIOSPERMS. is monocarpellary, the ovule arises from a marginal placenta. It is inserted thereon with a stalk, the funiculus or funicle (/) ; the free part of this is very short, the rest of it is grown to the ovule, forming upon it the raphe (r) . In the body of the ovule we distinguish first of all the inner conical mass of tissue as the so-called nucleus of the ovule — the nucellus (n). This corresponds with the macrosporangium of the Vascular Cryptogams. The nucellus is encased in two integuments, an inner (ii) [originally called the " secundine " by Mirbel ; but, as it is developed before the other, more recently known as the "primine"], and an outer (ie) [originally called the " primine " ; but, as it is developed after the other, more recently called the "secundine"]. The inner is developed on all sides to the base of the nucellus, the outer is want- ing on the side of the raphe, in that it joins on both sides to the funiculus. The inner integument leaves a narrow canal free between its upper edges, which extends to the nucellus ; this canal is known as the micropyle. The funiculus is traversed by a fibro-vasal bundle, coming from the placenta, which in many, but not, however, in all, cases can be traced to the base of the nucellus. The tissue adjoining the base of the nucellus, here dis- tinguished by its brighter color- ation, is known as the chalaza (ch). In the long axis of the nucellus is noticeable a larger cell, forming quite a cavity ; this is the embryo-sac (e). At its base can be seen some globular cells, which in Aconitum (and Ranunculaceaa gener- ally) are very strongly developed — the antipodal cells (a). In specially favourable cases we can determine that they are three in number. In the apex of the embryo-sac we can also see a small cell, which, however, is only recognisable in perfectly median sections ; it is the egg [-cell or oosphere, sometimes called the embryonic vesicle or germinal vesicle] (o) . The ovule as a whole is distinguished as anatropous, i.e. turned back, because the body of the ovule does not lie in direct continuation of the funiculus, Fig. 107.— Aconitum Napellus; median longitudinal section of an ovule. /, funi- culus ; r, raphe ; v, fibro-vasal bundle of the funiculus ; i.e., outer integument ; ii, inner integument; n, nucellus; ch, chalaza; e, embrjo-sac ; a, antipodal cells; 0, the oosphere; ne, nucleus of the embryo-sac ; m, micropyle ; or, wall of the ovary (x 53). STRUCTURE OF THE OVULr. 329 but appears laid b}^ the side of it, with one side grown to it, and the micropyle turned to the base of the funiculus, Tliis form of ovule is by far the most common in Angiosperms. If we noAv compare our preparation of Delphinium (Fig. 106) with that of Aconitum (Fig. 107) we shall see that the structure of the ovary and ovule in the two cases is quite identical ; the distinction only is that in Delphinium [as very commonly in Ranuneulacea)] the two integuments of the ovule are blended together. In order to prepare sections of the ovule of Aconitum, we can separate one from the ovary, and cut it singly between the thumb and fore-finger, in the method already known to us. If the ovule is correctly arranged between the fingers, we shall in this way more rapidly obtain true median sections. In this and in other like cases the ovule may, with advantage, be first em- bedded in glycerine- jelly or in celloidin (celluloidin), and then cut. The glycerine-] elly must be tolerably firm, i.e., must contain a comparatively large proportion of gelatine. In celloidin only alcohol material can be embedded. We pour the solution of celloidin * into a small box made of writing paper, and immerse the ovule in it. The celloidin is then allowed to stand in the air till it has acquired a firm skin, when it is laid in 82 p.c. alcohol. Here, after some hours, it acquires the consistence of cartilage, remaining transparent. Celloidin and object are cut together, and the sections laid in glycerine or glycerine- jelly, without its being necessary to remove the celloidin. The sections can be stained with carmine or with logwood (hsematoxylin), but not with aniline colours, as these latter colour the celloidin as well. If the celloidin has been procured in cakes, it must be dissolved before use in equal parts of ether and absolute alcohol. In order to make ovules which are to be embedded in glycerine- jelly or in celloidin still more visible, they can be previously stained with watery logwood ; the ovules must then, however [after previous washing in water], be again dehydrated in absolute alcohol, before being placed in the celloidin. Objects which, in order to make them available for section-cutting, must be permeated with celloidin, are first treated with dilute solution of celloidin, in which the object must lie often for days before it is transferred to, and embedded in, the thick celloidin solution. * To be obtained of Dr. Griibler, iu Leipzig, Dufour-strasse, 17. In cakes at about 3s. each, or iu solution at lis. the kilogramme. Also of Messrs. Southall Bros. & Barclay, manufacturing chemists. Dalton St., Birmingham. [Ed.] 330 REPRODUCTION OF ANGIOSPERMS. We will now iake in liand the study of the interior of the embryo-sac. The most favourable object for this is Monotropa Hypopitys [the Bird's-nest Rape, a pale yellowish parasite, found chiefly under beech and fir trees] .^ This plant is so favourable for the otherwise difficult investigation of the embryo-sac that no pains should be spared, if possible, to obtain it.* It Howers in July and August, and must be examined fresh, as in alcohol it becomes brown and oj^aque. The plant bears carriage very well, and can be preserved healthy for a very long time in a glass of water. [Supplies may in this way be obtained from a distance.] With Monotropa agree the various species of Pyrola, or "wdnter- green," [likewise Ericaceous plants,] excepting that their ovules are smaller. [About a dozen species of Pyrola, all hardy herbaceous perennials, can be readily enough cultivated in gardens, either from seeds or division of the roots, selecting for the purpose a shady border, w4th a sandy peat soil.] The cross- section through the under part of the ovary shows this to be four-celled [5-celled in Pyrola']. The placentae are strongly swollen, and bear on their surface very numerous, slender, closely serried ovules. The two halves of the placenta in each cell are removed to some little distance by a radial line of separation. In the upper part of the ovary these lines of separation extend to the centre, and there adjoin one another. We see then four strong pairs of placentae, each placed on the centre of a partition w^all, which appertain to the two neighbouring loculi ; the pairs are easily separated from one another w^ith the needles. We get the ovules for our investigation by removing a portion of the wall of the ovary with the forceps, and stripping off the ovules with the needles from the placenta thus exposed. We place them in pure water, or in 3 p.c. solution of sugar, in which the ovules remain longer unchanged. We take this material from an oldish flower? in -which the stamens have already dehisced, so that we shall find the ovules in part ripe and not yet fertilized, in part already fertilized. Between the ovules we come often upon pieces of pollen tubes. The receptive ovule has the appearance of the adjoining fig., 107. It is transjDarent, and can be focussed for optical section. We recognise in it an anatropoiis ovule, with but one integument (i). The whole interior of the ovule is filled by the embryo-sac ; we miss the nucellus, which, during the develop- * It is hardly likely to be regularly obtainable in England, so that the alter- native plants referred to later on must be relied on for material. [Ed.] THE EMBRYO-SAC. 331 ment, is pressed back bj the embryo-sac. The apex of the embryo-sac is occupied, as we can clearly see, by three cells. These three cells form the egg-apparatus [or germinal apparatus]. They are not of eqnal value. The two upper are the assisting- cells, or synergidae (Fig. 108, B) ; that more deeply inserted, is the oosphere (o) [germinal vesicle, embryonic vesicle]. The synergidcT, as can be easily seen, have in their lower part a vacuole, are filled above with protoplasm, and here contain the nucleus. The oosphere inversely has its cavity above, and below the main mass of its cell-protoplasm, and the nucleus. Both 332 REPRODUCTION OF ANGIOSPERMS. synerq-idae are not always seen, as one can cover tlie other (Fig. 108, C). At the base of the embryo-sac the antipodal cells can usually be recognised without difficulty, and we can count that three also of these are present. In the interior of the embryo-sac is usually found a nucleus, with a nucleolus (Fig. 108, A) ; but in other cases there are two nuclei (B) or a nucleus with two nucleoli (C) ; and we judge from this, that the one nucleus which we always ultimately find arises from the union of two. Ovules, the fertil- ization of which has already commenced, can be recognised by the changes which the synergidae have undergone. These appear strongly refractive, both or only one being thus modified. It is then certain that a pollen-tube has penetrated to the embryo-sac, and if it is not easy to see it in the interior of the micropyle, it is still not difficult to recognise the piece torn off from it in the preparation and projecting beyond the micropyle. The apex of the pollen-tube, however, has penetrated to the synergidae, and the protoplasm of the pollen-tube between the synergidae to the oosphere. With careful examination we may happen, in oospheres which border on synergidas that are thus changed, to find two nuclei (D), one larger, the original nucleus of the oosphere, and close by it also a smaller, the spermatic-nucleus, which has pene- trated from the pollen-tube. This latter increases quickly in size. We can find stages of conjugation between the 00-nucleus and this spermo-nucleus, and afterwards see only one embryo-nucleus, with two unequal nucleoli, of which the smaller arose from the spermatic-nucleus (E), and ultimately an embryo-nucleus with only one nucleolus. While the oosphere is being fertilized, the highly refractive masses of substance in one or both synergidae diminish ; they are apparently used for the nourishment of the oospore. At the same time with these changes in the egg- apparatus the formation of endosperm has commenced in the cavity of the embryo-sac ; i.e., we see the embryo-sac divided by walls. The endosperm-formation here, therefore, takes place by cell division. In other equally frequent, even more frequent, cases, the nucleus of the embryo-sac and its descendants at first divide free ; and only at a later stage of the development the formation of partition-walls between these nuclei commences. The process, as we have it here, takes place in general in such embryo-sacs as show slow, and on the whole inconsiderable, increase in size. Where, on the other hand, the embryo-sac increases very rapidly in size after the fertilization of the egg-cell, there nuclear division THE EMBRYO-SAC. 333 without cell-division first takes place, and cell-formation [i.e., the formation of the partition- walls] first begins when the embryo-sac is approximately fully developed. — In consequence of fertilization the oospore has acquired a delicate cellulose membrane, and soon begins to elongate into a sac, and after some time penetrates with its apex into the body of the endosperm, where the apex of the sac produces a few-celled embryo, [the rest of the sac forming the suspensor] . — We have thus far examined these ovules only in pure water or in sugar-solution ; if we wish to see the nuclei come out clearly, we must treat the ovules with two per cent, acetic acid. In this way we obtain very sharply- defined structures in most ovules, and at the same time fix the stages of division of the nuclei, although into these processes we do not pro- pose at present to go more deeply. Staining media cannot be recom- mended, since they stain also the nuclei in the integument, and in this way injure the view into the interior. Instead of Monotrojpa various Or- chids {Orchis and other genera) can serve for this investigation. Fer- tilization takes place in these a good while after pollination, and in ovaries which are already greatly en- larged. These are cut open, ovules removed with the needles from a placenta, and transferred to water or three per cent, solution of sugar. We can, without further steps, in- form ourselves as to the structui^e of the fully-formed ovule (Fig. 109) ; this is very like to that of Moiiotropa, but there are two integu- ments [as commonly in monocotyledons], and an air-cavity in the neighbourhood of the chalaza. This air-cavity makes observation more difficult if it is filled with air, and this latter also penetrates between the integuments. The ovule in water or in three per cent, sugar solution, must therefore be freed from aii' under the air pump. Often even a slight pressure upon O.V Fig. lOd.— Orchis pallcns. Receptive ovule. OS, egg-apparatus; ii, inner, ic, outer integument ; I, air-cavity The other letters are as in the pre- vious figures (X 240). the cover-glass 334 REPRODUCTION OF ANGIOSPERMS. serves to remove tlie most disturbing air, found bet^veen tlie integuments. The nucellus in the Orchideoe also is quite displaced hj the embryo- sac ; as a relic of the nucellus a strongly refractive cap of substance is still to be seen at the apex of the embrjo-sac. The egg-apparatus {os) is constructed as in Monotropa, only that the oosphere is less deeply inserted. The antipodal cells are not to be seen; in their place is a strongly refractive substance, in which lie, in fact, three nuclei, recognisable, however, with great difficulty. The pollen-tube can, more easily than in Monotropa, be traced to the synergidee ; the changes which the synergidae undergo are the same. The two nuclei, moreover, are again found in the fertilized oosphere. Endosperm is in general not formed in the Orchidere. In default of Monotropa and of Orchidace83 transparent ovules for investigation are provided by various Gesneraceae,"* and, above all, the large-flowered Gloxinia hyhrida of the gardens. The ovule, having only one integument, is so far translucent that the egg- apparatus is clearly visible. It shows the two synergido3, and in this case flask-shaped oosphere. Under some circumstances two oospheres can be present. The embryo-sac in its npper part is swollen, bnt narrows suddenly below ; the antipodal cells in the lower end are not distinguishable with certainty. One of the most favourable plants for the study of fertilization is however the Scrophularineous plant Torenia Asiatica.^ This [stove-evergreen from the East Indies] is now cultivated very generally in gardens, and bears flowers the whole year through. It is distinguished in that its embryo-sac grows upwards into the micropyle, and heuce the whole egg- apparatus appears without any other covering than the wall of the embryo-sac. Cross- sections through the superior, elongated ovary show this to be two-celled ; the two axile placentae project as pads into the loculi. They are covered with numerous ovules. For the purpose of observation we remove a wall of the ovary, and strip off the ovules from the placenta, best under the simple microscope. We observe them with advantage in a 3 p.c. solution of sugar. The ovules are anatropous, or, more correctly, somewhat campylo- tropous, for the embryo-sac and the integument are bent in their upper part (Fig. 110, A). The free part of the funiculus (/) of the ovule is pretty long. There is only one strong integument. The embryo-sac (e) projects with its upper end out of the micro- pyle. This protruding part is convexly swollen and pointed at its THE EMBRYO-SAC. 335 apex ; it lies against the funiculus. It is difficult to follow the embiyo-sac in the interior of the ovule, but by running in a little potash we can, during its commencing action, convince ourselves that it immediately adjoins the integument, is first very narrow, then swells somewhat spindle-shaped, and (e*) again narrows at the base. Our preparations in sugar solution show, in the free apex of the embryo-sac, the two synergida3 and the oosphcre ; once more therefore, as always, three cells form the egg-apparatus. 336 REPRODUCTION OF ANGIOSPERMS. According to tlie position of the preparation two synergidae are to be seen (Fig. 110, B), or one covers the other (G). At the apex of each synergida we notice here a homogeneous, strongly refractive cap, sharply defined against the finely granular portion behind; this is the so-called filiform apparatus. If such a pre- paration is treated with chlorzinc iodine, the caps of the synergidce are seen to colour blue. They consist therefore of cellulose. The other, substance of the synergidos and of the oosphere colours yellow-brown. Careful examination shows that the membrane of the embryo-sac is open over the caps of the synergidae (jB, C) . The filiform apparatus therefore now forms the stopper to the opening of the embryo-sac. This apparatus is very widely distributed, especially amongst monocotyledonous ^Dlants, and projects often very far out of the embryo-sac. The longitudinal striation, w^hich they very commonly show, arises from fine pores filled with protoplasmic contents. — We turn- again to our preparation lying in water or in sugar solution, and further determine that here also the distribution of the contents in the synergidae and the oosphere is entirely the same as in Monotropa and Orchis (B, G). In the synergidae the nuclei lie in the upper, the vacuole in the lower part ; in the oosphere this is reversed. — If we wish to study the process of fertilization in Torenia we must pollinate the flowers for this purpose. From pollination to fertilization thirty-six hours elapse, so that we must renew our observations after from a day and a half to two days. As before, we free the ovules from the placenta, but as carefully as possible, under the simple microscope, in order to remove the largest possible pro- portion of the pollen-tubes. These are followed here, with the greatest possible ease, to the apex of the embryo-sac, between the caps of the synergidae and right up to the oosphere (D, IS). We see that the pollen-tubes conducted by the placentae are still further led by the funiculi till they attain the apex of the embryo-sac. A direct influence makes itself felt at the same time from this latter, which effects the direction of growth of the apex of the pollen- tube. It can be assumed that the synergidae secrete a definite substance which acts as a stimulus upon the pollen-tube. The caps of the synergidae, on account of their soft consistence, oppose little resistance to the escape of the secretion. Where the caps of the synergidae are specially strongly developed, they appear besides traversed by very fine canals, which conduct the secreted substance outwards. The synergida in Torenia, as in others, become dis- NOTES. 337 organized after the entrance of the jjollen-tnbc, and take on the strongly refractive appearance already known to us. For the study of the further stages of development of the embryo, etc., tl:is object is not favourable. NOTES TO CHAPTER XXIX. ^ Goebel, Gnindziige d. Sijst., etc., p. 417. Liirssen, Grundz. der Botanik, p. 356. 2Ied. Pharm. Bot., Bd. II., p. 244. Prantl, Lehrbuch der Boianik, 4th edit., English edition by Vines, p. 204. - Strasburger, Befruchtuug iiiid Zelltheilung, iDp. 34, 35. •^ The same, p. 55. •* The same, p. 54. ° The same, p. 52. 338 THE SEED OF ANGIOS PERMS. CHAPTER XXX. STRUCTUEE OF THE SEED OF ANGIOSPERMS. Material Wanted. Flowers and seed-pods of various ages of the Shepherd's Purse {Ga'p- sella Biirsa-pastor Is). March to November. Fresh, also in alcohol. The same of the Water-Plantain {Allsma Plantago). June to August. Pipe seeds of the Castor Oil plant {Rlcinus communis). We will now endeavour to make ourselves acquainted with the structure of a ripe seed, and to give especial attention to the einbyro which it contains. As a comparatively favourable plant, we select one of the Cruciferas, Capsella Bursa-jpastoris [the shepherd's purse], a plant which has been very commonly made use of for embryological studies. ^ Its seed is comparatively small, but provides special advantages for developmental investigation. For this reason therefore we will endeavour to overcome the difficulty that the cutting of the ripe seed presents. It is advisable first of all to prepare a median longitudinal section through the seed, as we want to know what the object looks like, the development of which we are about to study. This section presents no insuperable dif- ficulty, if we have fresh seeds at our disposal, in preparing between the fingers. It is still easier if we place the seed between two flat pieces of cork, and draw the razor between them. Or the seed can be fastened with gum in the desired position between two pieces of soft lime or poplar wood, and after it is dry the sections made through wood and seed at the same time. Or the seed can be embedded in a drop of gum, to which a little glycerine has been added, at the end of a piece of elder pith, and after drying, gum and seed can be cut at the same time. The sections, in whichever way prepared, should be examined in glycerine, as water makes the embyro swell, and come out of the testa. The embyro (Fig. Ill, A), fills up the entire seed; it is bent at its mid-length, so that the cotyledons (c) lie alongside the hypo- SEED OF CAPSELLA. 339 cotyledonarj axis, or hypocotyl (h) [i.e. are incumbent]. (Cora- pare the figure.) Tliis disposition is characteristic of [several tribes of the Cruciferoe, by some systcmatists collected into] the section Notorhizeos, and may be represented by the sign | | Q [Another characteristic method of folding of the embryo in Cru- ciferae, is where the hypocotyl is folded over and applied to the edges of the cotyledons. This is called accumbent, and may be expressed by the sign ZZ O- They may likewise be represented ^^y C C H ^^^ c H respectively.] If the section is delicate and has cut the seed perfectly in the centre (as in the adjoining Fig. A), we see at the base of and between the cotyledons the small o-row- ing apex of the stem [the plumule], and can also see, at the lower end of the hypocotyl, the axis closed by the radicle covered by a root-cap only a few cells thick. There is no endosperm to be seen ; the embryo is immediately surrounded hj the skin of the seed — the testa, [and is therefore said to be exal- buminous]. If we use a somewhat higher magnifying powder, we can determine that this testa (Fig. Ill, B) consists of three layers of cells. The innermost layer (a) is composed of cells with comparatively little thickened and almost colourless walls, and with granular contents. Addition of iodine shows us that these grains are coloured yellow-brown, and are there- fore aleurone. Outwardly follows a second layer (c), the cell- w^alls of which are coloured deep brown, and on the inner side are very strongly thickened. The outermost layer of cells ap- pears in concentrated glycerine as a colourless, apparently homo- geneous membrane ; its cells are strongly flattened, and thickened almost to the obliteration of the lumen. Between the innermost and the second outer layer is often to be distinguished a crushed layer of cells, appearing like a single membrane. — If we examine the testa from the outside, we easily recognise the contour of the polygonal cells of the external tabular layer. These cells have /t Fig. 111. — Capsolla Bursa-pasloris. A, longi- tudinal section through a ripe seed ; 7i, hypo- cot.yl : c, cotyledons ; v, fibro-va?al bundlo of the funiculus (x 2(i). B, part of a longtitudinal section through tlie testa, after the action of water ; c, the swollen epidermis ; c, the brown strongly thickened layer ; * the crushed laj-er of cells ; a, the aleurone layer ( x 240 ). 340 THE SEED OF AXGIOSPERMS. their more internal portions partly separated by intercellular spaces full of air. In the middle of each cell is to be distinguished a weakly defined more strongly refractive portion. The walls of the next inner layer of cells are brown, strongly thickened, the cells themselves only a little smaller than in the outer layer. Considerably smaller, on the other hand, and weakly thickened, are the cells of the third, aleurone-containing, layer. — If we now allow water to run into the section from the edge of the cover- glass, we see, in the cross-section, the cells of the outer layer rapidly swell ; each bulges strongly outwards ; at their centre a highly refractive column is noticeable. The lumen is now hardly dis- tinguishable ; the entire cell is filled by the thickening layers of the wall, and in all cases the outer thickening layers are weakly, the internal strongly refractive. These internal thickening layers form the striking central columella, which now shows up very strongly on the surface of the seed, while at the same time the intercellular spaces between the cells disappear. The swelling walls usually show clear lamination. With further addition of water, the cuticle of the cells is ruptured, and the outer thickening layers come out, diffusing in the surrounding water as invisible mucilage. The refractive columella remains behind, marking the centre of each cell (Fig. Ill, B, at e). It has increased not incon- siderably in size ; at its apex can be seen the relics of the dissolved thickening layers. In the same way the lateral middle-lamello3 of the cells remain, and, as they do not swell, show now a much less height than the columellse. All this can be seen in our Fig. Ill, B, which represents the testa after the action of water. These phenomena of swelling can be observed more quickly if the section is first examined in alcohol, and then water run in. This muci- lagination of the thickening layers of the outer cells of seeds and mericarjDs* is a comparatively common phenomenon, which facili- tates the sticking of the seed to foreign objects, and therefore serves in their transport and dissemination, and on the other hand has as a result, the firm retention of water on the surface of the seed. As the section-cutting of ripe seeds presents some difficulty, we can, so far as informing ourselves about the position and structure of the embryo is concerned, prepare the sections from seeds which * Mericarps, the segments of fruits which, like the Pelargonium (Geranium), Cranes' Bills, and Parsley Worts, do not dehisce to let out their seeds, but split up bodily into seed-containing segments. Hence these fruits are called scluzo- ca77)S—sphtting fruits. [Ed.] SEED OF CAPSELLA. 341 are not quite ripe, and far softer, and only study the testa upon fully ripe seeds. In this we can go back to younger stages, and at first lay the entire ovule in potash. These ovules can be best obtained by halving the seed-vessel in its entire length, and then removing the ovules from each half with the scalpel. The ovules, almost to the fully ripe stage, can be made so far transparent that we can inform ourselves sufficiently as to the position of the embryo. The embryo goes a beautiful green in potash, which arises from the swelling of the starch-grains, so that the chloro- phyll-grains become visible. Proceeding to progressively younger ovules, we see that the embryo (and at first especially its coty- ledons), becomes continually shorter. It continually withdraws further and further from the lower, outward-bent, half of the embryo sac. Ovules from fruits which, without stalk, measure ^ inch in height, show the embryo as a small body of cordate form. The two divaricating anterior protuberances are the rudiments of the cotyledons. If we trace back the various stages of develop- ment of the embryo, we can at the same time determine that endosperm is formed only at the two ends of the embryo-sac, and appears especially at the chalazal end as a small green-coloured mass of tissue. The latter is not reached and absorbed till the seeds are well-nigh ripe. We can also see that the testa proceeds from the two layers of cells of the outer integument, and the inner cell-layer of the inner integument. This latter layer is easily distinguished by its richness in contents. The one or two layers of cells lying between this innermost layer and the outer integument are gradually pressed back and crushed, so that they ultimately form only the compound membrane lying between the second and third layer of the testa. [If unripe seeds of various ages ai-e laid upon an object-slide in a drop of glycerine, and covered with a cover-glass, and then subjected to carefully con- trolled pressure with the handle of the needle-holder, the ovules will burst, and the often uninjured embryo will come out. This process is effective with embryos from a very young state up to those that are well-nigh fully formed.] — In order to inform our- selves as to the structure of the egg-apparatus in the ovule at the receptive period, we must have recourse to alcohol-material, which we make transparent to the desired degree by careful addition of potash. We can thus state the jn'esence of two syncrgida?, and an oosphere, in the egg-apparatus, while the antipodal cells are very difficult to see. The structure of the ovule is easy to follow fresh, 342 THE SEED OF ANGIOSPERMS. in pure water, or with a trace of potash added to make it still more transparent. The ovule is campylotropous, i.e., its nucellus and embrjo-sac, as w^e have already seen in later stages, are curved. The outer integument consists of two layers, the inner in its upper part has two, further in has three layers. The nucellus at this stage is already absorbed, so that the embryo-sac is directly in contact with the inner integument. The funiculus is pretty long ; it is traversed by a fibi'O- vasal bundle, which ends in the chalaza, and is still visible even in the ripe seed (Fig. 111,^, i') Very beautiful, in the next older stages, especially after the addi- tion of potash, are the views one gets of the rudimentary embryo. We can see that the fertilized oosphere [oospore] grows into a thread-like pro-embryo, about six cells long ; the uppermost cell, i.e., that most removed from the micropyle, rounds off into a head- cell, the embryo-cell, [the other cells collective form the suspensor,] while the lowermost cell of this embryo-bearer or suspensor, the attaching-cell, swells at the same time bladder-like, absorbs the entire nucellar tissue up to the integument, and forms the bladder which we find at this place even in the ripe state (comp. Fig. Ill, A). This swollen cell may serve to help on the absorption of nutri- ment for the embryo. The tissue of the chalaza swells consider- ably at the same time, and the cell-contents become dark-coloured. Soon we see there the green endosperm-cells, which in smaller number surround the rudimentary embryo at the microi:)ylar end also. In such preparations we can determine that the swollen globular embryo-cell is already separated from the suspensor by a partition wall, and soon after is divided by a longtitudinal wall, at right angles to which follows a second longtitudinal wall, and then at its mid height a cross-wall. The globular embryo thus appears divided into octant-cells, in which are subsequently formed periclinal and anticlinal walls. The globular embryo increases in size and in number of constituent cells, flattens somewhat, and then from its anterior end the cotyledons grow out. These at first are in contact at their base, and then subsequently the growing point or plumule of the stem bulges out between them. [All of these stages can likcAvise be followed in fresh material, and in situ, by laying the ovules in glycerine, covering with a cover-glass, and then carefully heating over a spirit or gas-flame. The ovule is thus made transparent, and the embryo clearly visible.] For the study of the monocotyledonous embryo we select the common Water Plantain, Alistna Flantago.^ This object is, in fact, SEED OF ALISMA. 343 highly suited to this kind of investigation, and is therefore very commonly used for it. First of all we will make ourselves familiar with the fully-developed state. The flower of Alisma Plantago contains numerous monocarpellary pistils ; it is apocarpous [much resembling in this respect the Ranunculaceas] . From each flower therefore proceed numerous fruits, which are closely pressed together [in a single whorl] into a collective fruit [a dry etoerio] of triangular outline. Each individual ripe carpel [achene] is strongly flattened [laterally], somewhat thicker above, obovate in profile, with a median dorsal groove. At about mid-height of each achene, on its internal edge, is a short thread-like outgrow tli, repre- senting the withered style [which, therefore, / is ventral]. For fur- ther investigation we select an almost ripe etoerio, place a single achene between the two halves of a split cork, and draw the razor be- tween these two halves, we shall thus, without trouble, obtain tolera- bly median longitudinal sections; while cutting between the fingers presents difficulties, as the testa is so hard. At the same time we will prepare some cross-sections in the ordinary way, between two pieces of cork. The longitudinal sec- FiG. 112. — Jlisina Plantago. Mcclian longitudinal section through a ripe acheue. cp, epicarp (epidermis); in, mesocarp ; en, endocarp of the wall (pericarp) of the fruit; v, a fibrovasal bundle in this; v*, the end of the fibro-vasal bundle; sf, the withered style; <, the pollen-passage ; /, funiculus of the seed, with its fibro- vasal bundle /d; mp, micropyle; ch, chalazal end; ). Calyptrogen, 185 et seq. (Fig. 69). Cambium, 97 (Fig. 44), 101; interfasci- cular, 106. See also Thickness, Increase of; Vascular bundles, etc. (Figs. 44, 45, 46,47, 50**). Camera lucida of Abbe, 30 (Fig. 16) ; Zeiss, 31 (Fig. 2); Wollaston, 33; Beale, 33. See also Drawing. Camphor, Use of, 308. Campylotropous ovule, 342. 408 INDEX. Canada Balsam, Use of, 92, 230, 366. „ in benzole. Use of, 92. „ in chloroform, Use of, 92, 230, 365. ,, in turpentine, Use of, 92, 230, 234, 365. ,, in xylol, Use of, 236. Canal, cariual, 180 ; vallecular, 180. Canal-cell, of Marchantia, 275 (Fig. 95); of Polypodium, 294 (Fig. 98). ,, Ventral, of Marchantia, 275. Capillary apparatus of Sphagnum, 193. Capsella Bursa-pastoris. Structure and development of embryo and seed, 338 (Fig. Ill) ; structure of testa, 339 (Fig. 111). Capsule of Moss, 280 (Fig. 95 c, d, e). Carbolic acid, Use of, 309, 319, 320, 343. Carbon bisulphide, Use of, 232. Carmine, Alum, Use of, 92. ,, Ammonio-acetic, Use of, 92, 103. „ Beale's, Use of, 205. ,, Borax, see Borax-carmine. ,, Picric, see Picric -carmine. Carpellary leaf, 306, 323. Carrot, see Daucus Carota. Caulicle, 310. Cauline bundles, 171. Cedar, Oil of, Use of, 230. Celandine, see Chelidonium majus. Cell- division, 356; in Cladophora glomerata, 368; anthers of Fritillaria persica, 360 ; of Helleborus fcetidus, 367 ; of Tradescantia virginica, 356. Cell-division, anticlinal walls, 173 ; periclinal, 173 ; oblique, 191 ; rectangular, 173. Cclloidin (Collodion), Use of, 329. Cell-plate, 359 (Fig. 114 ry). Cells, Multinuclear, see Nucleus. Cell-sap, 29 ; blue, 42 ; purple, 41 ; red, 40, 41, 42, 43, 65 ; violet, 74 ; yellow, 41. Cellulose, Eeactions of, 47. „ Fungal-cellulose, Eeactions of, 202, 263. ,, Starch-cellulose, Eeactions of, 270. Cell-wall, Structure of in endosperm of Date, 54 ; seed of Oniitho- galum unibeUatum, 53; in Pin- nularia viridis, 213 ; Pimis sylvestris, 57. „ Middle-lamella, 54, 58, 140, 155 ; lamination, 53; striation, 50, 53. ,, Thickening, 47 ; pits, 54. , , Cuticularized and suberized, struc- ture, 153 ; reactions, 58, 65, 68. ,, Lignified, reactions of, 59, 119. Cement (Glass), 24. Central cell, of Marchantia, 275 ; Mnium, 280 ; Polypodium, 294. Ceratopteris thalictroides, germination of spores, 290. Ceric acid, Eeactions for, 155. Chalaza, 328 (Fig. 107). Chamber, Moist, see Moist chamber. Chara, Protoplasmic movement in, 37. Cheiranthus Cheiri, hairs, 72 (Fig. 32). Chelidoniiivi majus, Vascular bundle of, 101 ; latex vessels of, 102. Cherry, Structure of fruit, 348. Cherry-wood, Extract of, Use of, 59. Chloral hydrate, Use of, 39, 319, 320. Chlorococcus humicolum = Cystococcus humicola. Chloroform, Use of, 26. Chlorophyllan reaction, 204. Chlorophyll-bands, 208. Chlorophyll-bodies, 35 ; in Funaria hygrometrica, 38 (Fig. 17) ; division of, 38 ; function of, 167 ; starch in, 39 (Fig. 17). Chlorophyll-corpuscles, see Chloro- phyll-bodies. Chlorophyll-grains, see Chlorophyll- bodies. Chlorophyll-vesicles (Amylum bodies) , see Pyrenoids. Cliloroplasts, see Chlorophyll-bodies. Chlorzinc Iodine, Use of, 46, 48, 50, 53, 54, 55, 57, 58, 68, 84, 87, 110, 119, 122, 124, 155, 202, 237. Chrom-acetic acid, 1 per cent., Use of, 205. Chromatophores, see Colour-oodies, Chloroi)hyll-bodies. INDEX. 409 Chromic acid, Use of, 70, 213. ,, 0-5 per cent., Use of, 235. „ 1-0 per cent.. Use of, 37, 204. „ 20 per cent., Use of, 213. ,, 25 per cent.. Use of, 318. „ Concentrated, Use of, 59, 155, 213. Chromoplasts = Cbromatophores. Chroococcaceae, 218. Cilia, 218, 223, 240, 250, 251, 254, 293 (Figs. 86, 87, 89, 94, 95 a, 97). ,, of Lichens, 202. „ of Mosses, 283 (Fig. 95 e). Citrus vulgaris [C. Aurantium), ad- ventitious embryos, 354 ; de- velopment of fruit, 352 ; struc- ture of fruit, 350. Cladophora glomerata, 203 (Fig. 75) ; cell-division in, 368 ; chroraato- phores, 203; nucleus, 204; pyrenoids, 203 ; swarm-spores, 248 (Fig. 86). Cladophorere, 203. Clearing sections, 172. Cleistocarpous, 261. Clips on microscope stage, 3. Closing membrane, see Pits. Clostridium butyricwn, 223. Cloves, Oil of. Use of, 235, 365. Club- mosses, see Lycopodium. Cluster-cup, see Mcidium. Collateral vascular bundles, 86, 180. Collecting cells, 167. Collemaceae, 271. Collenchyma, 104 (Fig, 45**), 152, 164, 165. CoUeters (glandular hairs), 78. Collins' microscopes, xv. Colonial algOB, 219. Colour-bodies, in flower of Adonis fiammeus, 42; of Delphinium, 42 ; of Pansy, 74 ; of Tropcsolum mojus, 40 (Fig. 18). ,, in root of Daucus Carota, 43 (Fig. 20). Columella, of Blucor, 255 ; Mnium, 283 (Fig. 95 d) ; in the cells of the testa of seeds, 340 (Fig. 111). Companion-cells, 146. See also Sieve- tubes. Condenser, se