UNIVERSITY OF CALIFORNIA AT LOS ANGELES SOUTHERN BRANCH; UNIVERSITY OF CALIFORNIA. LIBRARY, IjOS ANGELES. CALIF. y RESEARCHES ON FUNGI VOLUME II BY THE SAME AUTHOR RESEARCHES ON FUNGI VOLUME I. An Account of the Pro- duction, Liberation, and Dispersion of the Spores of Hymenomycetes treated Botanically and Physically. Also some Observations upon the Discharge and Dispersion of the Spores of Ascomycetes and of Pilo- bolus. With 5 Plates and 83 Figures in the Text. Royal 8vo, 12*. Qd. net. LONGMANS, GREEN AND CO. LONDON NEW YORK TORONTO BOMBAY CALCUTTA AND MADRAS RESEARCHES ON FUNGI VOLUME II FURTHER INVESTIGATIONS UPON THE PRODUCTION AND LIBERATION OF SPORES IN HYMENOMYC ETES BY A. H. REGINALD BULLER B.Sc. (LOND.) ; D.Sc. (BiBM.) ; PH.D. (Lsip.) ; F.R.S.C. MEMBRE ASSOCIE DE LA SOCIETE ROYALE DE BOTANIQUE DE BELGIQUE PROFESSOR OF BOTANY IN THE UNIVERSITY OF MANITOBA WITH ONE HUNDEED AND FIFTY-SEVEN FIGURES IN THE TEXT LONGMANS, GREEN, AND CO. 39 PATERNOSTER ROW, LONDON, E.G. 4 NEW YORK, TORONTO BOMBAY, CALCUTTA AND MADRAS 1922 All rights reserved 67839 Made in Great Britaii Agric. ybept; QK I TO GULIELMA LISTER WHOSE BEILLIANT REVISION OP HER FATHER'S MONOGRAPH OF THE MYCETOZOA HAS ENRICHED BOTANICAL LITERATURE P2 If PREFACE IN 1909 there was published my Researches on Fungi, which treated i of the fruit-bodies of Hymenomycetes and certain Ascomycetes ^ considered as organs for the production and liberation of spores. x During the thirteen years which have elapsed since that time my ^investigations have been continued, and I now propose to embody the results which have accumulated therefrom in three new volumes. The 1909 volume and these new volumes are to be considered as parts of a larger work having the general title Researches on Fungi. . Of this work the 1909 volume now becomes Volume I and the pre- x, sent volume Volume II. The third and fourth volumes are already £> t in an advanced stage of preparation for the press. The morphological and physiological facts which have come to light in the course of my researches have taught me that the adapta- tion of structure to function in the higher fungi is just as remarkable as that found in the Phanerogamia. The form and arrangement of . parts exhibited in the sporophore of the Common Mushroom and its oS allies appear to be no less beautifully fitted for the efficient pro- . duction and liberation of spores than are the form and arrangement v of Orchid flowers for securing successful pollination by insects- :" Ample evidence supporting this conclusion will, I believe, be found in the pages of this book. The analysis of the hymenium of Panaeolus campanulatus, as described in Chapter X, will perhaps enable the reader to comprehend, in a manner not hitherto possible, how millions of hyphae combine their activities to produce the spore-stream which is emitted from beneath the pileus without a moment's interruption for the seven or more days of the spore- fall period. The work for this volume has been carried out chiefly in my own laboratory at the University of Manitoba, but in part also in the viii PREFACE Herbarium of the Royal Botanic Gardens at Kew and in the Botanical Department of the University of Birmingham. It is with much pleasure that I here acknowledge the courtesy of the authorities at Kew and Birmingham for the facilities accorded me. Of the one hundred and fifty-seven illustrations, all but twelve are here published for the first time. The source of each borrowed illustration is duly acknowledged in the text. Of the seventy-seven drawings, sixty-nine were executed by my own hand. Of the eighty photographs, thirty-one were made under my direction ; and the others, with the exception of four taken from previous publica- tions, were very kindly contributed by friends and correspondents : fourteen by A. E. Peck ; eleven by Somerville Hastings ; three each by J. E. Titley, J. H. Faull, and C. J. Humphrey ; two each by Miss E. M. Wakefield, C. W. Lowe, B. Leeper, and W. S. Odell ; and one each by G. G. Hedgcock, R. Hancock, and W. B. Grove, the last having been taken by the late J. Edmonds. I desire to thank my numerous correspondents in both Europe and North America for assisting me by sending specimens of fungi and reprints of papers, and by supplying information of various kinds in response to my enquiries. Dr. Guy Barlow has been good enough to help me with the sections relating to the revised formula for Stokes' Law. In the final revision of the proofs, my friend, W. B. Grove, M.A., has again given me the benefit of his wide mycological knowledge and experience. Finally, I wish to express my indebtedness to the Birmingham Natural History and Philosophical Society for a grant of £25 toward the cost of making the blocks for the illustrations of this volume, and to the Canadian National Council for Scientific and Industrial Research for a grant of $1,000 toward the expense of producing this volume and the two which are to follow it. A. H. REGINALD BULLER. WINNIPEG, October 22, 1922. TABLE OF CONTENTS PAOE PREFACE -.- . . vii CHAPTER I BASIDIA AND THE DISCHARGE OF SPORES The Elements of the Hymenium — The Discharge of Spores from the Basidia — The Water-drop— The Chemical Constituents of the Drop— Stokes' Law— The Mechanism of Spore-discharge — Does a Basidium Produce More than One Generation of Spores ? — The Sterigma and Spore-hilum in Hymeno- mycetes and Gastromycetes- — Balloons Falling from Rest or Fired from a Gun, Used to Illustrate the Movements of Spores — A Comparison of the Rates of Fall of Spores and Thistle-down — A Comparison of the Rates of Fall of Spores and Bacteria . . • . . . CHAPTER II THE RATE OF DEVELOPMENT OF INDIVIDUAL SPORES IN DIFFERENT SPECIES Introduction and Table of Results — Methods — Discussion of Results — Effect of Temperature . . . . CHAPTER III VARIOUS OBSERVATIONS The Suppression of the Gills of Lactarius piperatus as a Result of the Attack of Hypomyces lactifluorum — Sterile Fruit-bodies — The Monstrous Fruit- bodies of Polyporus rufescens — The Grafting of Fruit-bodies — The Dwarf Fruit-bodies of Coprinus lagopus — The Cultivation of Marasmius oreades for Food CHAPTER IV THE DISCHARGE OF SPORES FROM CERTAIN AGAKICINEAE AND POLYPOREAE The Brief Spore-fall Period of Coprinus curtus — Macroscopic Observations on the Fall of Spores of Armillaria mellea — Banker's Observations on Spore-discharge in Hydnum septentrionale — The Vernal Spore-fall Period of Fom.es fomentarius — Perennial Spore-production by One and the Same x TABLE OF CONTENTS PAGE Tube-layer in Fomes fomentarius — The Geotropism of Fomes fomentarius Fruit-bodies — The Attachment of Fomes fomentarius Fruit-bodies — -The Spore-fall Period of Fomes igniarius— Winter Break in the Spore-fall Period of Daeilalea confragosa — Solitary and Imbricated Fruit-bodies of Polyporeae, etc. . . . . . . . . . .95 CHAPTER V FOMES APPLANATUS AND ITS SPORE-DISCHARGE PERIOD Habitat and Hosts — A Wound Parasite — Spore Structure — The Longevity of the Fruit -body — The Longevity of Various Polyporeae — The Hymenial Tubes — The Spore-discharge Period — A Comparison of the Spore-fall Period of Certain Hymenomycetes — The Mechanical Consistence of Coprinus and Fomes Fruit-bodies Contrasted — The Number of Spores in Fomes applanatus and Other Basidiomycetes — The Cause of the Long Spore-discharge Period in Fomes applanatus — The Progressive Exhaus- tion of the Hymenial Tubes — The Significance of the Production of Vast Numbers of Spores 121 CHAPTER VI SPORE-DISCHARGE IN THE HYDNEAE, TREMELLINEAE, CLAVARIEAE, AND EXOBASIDIEAE Preliminary Remarks — Spore-discharge in the Hydneae — Spore-discharge in the Tremellineae — The Basidial and Oidial Fruit-bodies of Dacryomyces deliquescent — Spore-discharge in the Clavarieae — The Genus Calocera — Spore-discharge in the Exobasidieae ....... 149 CHAPTER VII THE RED SQUIRREL OF NORTH AMERICA AS A MYCOPHAGIST Introduction — Squirrels Observed Eating Fungi — Winter Stores of Fungi — Storage in Bulk — Storage in the Forked Branches of Trees — The Storage of Fungi in Relation to Climate— Two Chickens Hung in a Tree — Summary . 195 CHAPTER VIII SLUGS AS MYCOPHAGISTS Introduction — Slug-damaged Fungi in an English W7ood — Slugs and Poisonous, Acrid, or Cystidia-bearing Agaricineae — Absence of Slugs from a Wood in Central Canada — Some Conclusions — The Finding of Fungi by Slugs —Previous Chemotactic Experiments — New Chemotactic Experiments- Experiments I, II, III, IV, V, VI, and VII— Slugs and Mustard Gas- Conclusions ........ .212 TABLE OF CONTENTS xi CHAPTER IX THE TYPES OF FRUIT-BODY MECHANISM IN THE AGARICINEAE PAGE Introduction — The Characters of the Aequi-hymeniiferous or Non-Coprinus Type — The Characters of the Inaequi-hymeniiferous or Coprinus Type — The Characters of the Sub-types — Order of Description of the Sub-types — Order of Investigations . . . . . . . . . 236 CHAPTER X THE AEQUI-HYMENIIFERAE : THE PANAEOLUS SUB-TYPE ILLUSTRATED BY PANAEOLUS CAMPANULATUS The Panaeolus Sub-type — Panaeolus campanulatiis : General Remarks on the Sporophore — The Phenomenon of Mottling- — The Spore-fall Period — Apparatus and Method for Observing the Development of the Hymenium — Observations on the Developing Hymenium — Successive Generations of Basidia— Description of the Hymenium of Panaeolus campanulatus in Detail — Significance of the Development of the Basidia in Successive Generations on any One Hymenial Area, and of the Existence of Various Areas — The Spores which are Wasted — Significance of the Protuberancy of Mature Basidia — Significance of the Collapse of Exhausted Basidia- — The Relative Position of the Basidia and the Spores of One Generation — The Position of the Sterigmata in the Hymenomycetes Generally and in Panaeolus campanulafus — The Cheilocystidia . . . . 244 CHAPTER XI STROPHARIA SEMIGLOBATA General Remarks — Description of the Hymenium in Detail — Rate of Dis- charge of the Spores of a Basidium and Collapse of the Basidium-body— The Spore-fall Period— Wasted Spores 327 CHAPTER XII ANELLARIA SEPARATA General Remarks — -The Production and Liberation of Spores — A Photograph of the Hymenium 347 CHAPTER XIII PSALLIOTA CAMPESTRIS Introductory Remarks — Occurrence of the Fruit-bodies— Fairy Rings — External Appearance of a Fruit-body— Evolution of Ammonia from Dead Fruit-bodies — General Organisation of the Fruit-body for the Production and Liberation of Spores — The Radial Arrangement of the Gills— The Number of the Gills— The Depth of the Gills— The xii TABLE OF CONTENTS PAGE Thickness of the GUIs— The Ends of the Gills— The Gill-chamber and the Annulus — Conditions for the Origin of Fruit-bodies — The Fate of Rudimentary Fruit-bodies — Effect of Dry Weather on Development — The Spore-discharge Period and the Number of Spores — The Mushroom and the Panaeolus Sub-type — Text-book Illustrations of the Hymenium — The Mottling of the Gill Surface — Methods for Examining the Hymenium in Surface View — The Number and Size of the Spores on Individual Basidia — Rate and Mode of Development of the Spores on an Individual Basidium — An Analysis of the Hymenium of the Cultivated Mushroom — The Hymenium of the Wild Mushroom — Camera-lucida Studies of the Young Hymenium — Camera-lucida Studies of a Nearly Exhausted Hymenium — Camera - lucida Studies of a Completely Exhausted Hymenium — Secotium agaricoides ....... 360 GENERAL SUMMARY . ... v . . . .-- . . 457 GENERAL INDEX . 467 RESEARCHES ON FUNGI CHAPTER I BASIDIA AND THE DISCHARGE OF SPORES The Elements of the Hymenium — The Discharge of Spores from the Basidia — The Water-drop — The Chemical Constituents of the Drop — Stokes' Law — The Mechanism of Spore-discharge — Does a Basidium Produce more than One Generation of Spores ? — The Sterigma and Spore-hilum in Hymenomy- cetes and Gastromycetes — Balloons Falling from Rest or Fired from a Gun, Used to Illustrate the Movements of Spores — A Comparison of the Rates of Fall of Spores and Thistle-down — A Comparison of the Rates of Fall of Spores and Bacteria The Elements of the Hymenium. -- The investigations of numerous observers have led to the conclusion that the hymenium of Hymenomycetes, when most fully developed, consists of three kinds of elements : basidia, paraphyses, and cystidia. The essential element of the hymenium, always present under normal conditions, is the basidium. Typically, this is a club-shaped cell which, on attaining maturity, projects slightly from the general surface of the hymenium and bears at its free end four sterigmata and four spores. While the vast majority of species have basidia of this type, there are exceptions to the general rule. Thus mono- sporous basidia are characteristic of Pistillaria maculaecola, di- sporous of the cultivated Mushroom (Ptsalliotacampestris), trisporous of Coprinus narcoticus, hexasporous of Cantharellus cibarius, and octosporous of Corticium coronatum. These and other aberrant types of basidia will be treated of more fully in Chapter X in a discussion of the relative position of the sterigmata. Paraphyses are elements which more or less resemble the bodies 2 RESEARCHES ON FUNGI of basidia, but do not bear sterigmata and spores. They are destined to remain sterile from the first. The early investigators of the hymenium, Leveille,1 Berkeley,2 Corda,8 and others, believed that the majority of the cells associated with the basidia remain sterile. In 1842, Montagne,4 who held the same opinion, gave to these sterile cells the name of paraphyses and described them as " elongated, tubular, caeciform cells, placed parallel the one to the other, like the pile of velvet." Later on, certain observers expressed doubt whether true paraphyses exist. Thus, in 1862, Hoffmann 5 stated that in Coprinarius every hymenial cell becomes a fertile basidium. De Seynes, in 1864, admitted that, in general, para- physes in the Hymenomycetes are sterile cells analogous to the paraphyses of the Discomycetes,6 but in 1867 he expressed the view that in Fistulina hepatica the hymenium consists of basidia only.7 Levine 8 more recently has said : "It seems to me that in all cases the paraphyses are really immature basidia." That paraphyses, except in the genus Coprinus, are simply basidia destined to become fertile, is a view also held by Ruhland 9 who described them as " nothing more than young basidia which have not yet attained full development," byDemelius10 who has defined them as "derzeit nicht fertile Basidien," and by ReneMaire n who calls them " basides 1 J. H. Leveille, " Recherches sur I'Hymenium des Champignons," Ann. Sci. Nat., 2 ser., T. VIII, 1837, pp. 321-338. 1 M. J. Berkeley, " On the Fructification of the Pileate and Clavate Tribes of Hymenomycetous Fungi," Ann. Nat. Hist., London, vol. i, 1838, pp. 81-101. * A. C. I. Corda, Icones Fungorum, T. Ill, 1839, etc. 4 C. Montagne, " Organographic and Physiologic Sketch of the class Fungi," Translated by Berkeley, Ann. and Mag. of Nat. Hist., vol. ix, 1842, p. 288. 5 H. Hoffmann, /cones analyticae Fungorum, Giessen, Heft II, 1862, p. 46. * De Seynes, " Ape^us sur quelques points de 1'organisation des champignons superieure," Ann. Sci. Nat., 5 s6r., T. I, 1864, pp. 243-244. 7 De Seynes, " Recherches sur quelques points de I'anatomie du genre Fistulina " Compt. rend., T. LXIV, 1867, pp. 426-429. 8 M. Levine, " Studies in the Cytology of the Hymenomycetes, especially the Boleti," Butt. Torrey Bot. Club, vol. xl, 1913, p. 133. * W. Ruhland, " Zur Kenntnis der intracellularen Karyogamie bei den Basidiomyceten," Bot. Zeit., Jahrg. LIX, 1901, Abt. 1, pp. 199-200. 1 Frau Demelius, " Beitrag zur Kenntnis der Cystiden," Verh. der k. k. Zool- Bot. Qesetts. in Wien, Bd. LX, 1911, p. 279. " Rene Maire, " Recherches cytologiques et taxonomiques sur les Basidio- mycAtes," Butt. Soc. Myc. France, T. XVIII, 1902, p. 187. BASIDIA AND THE DISCHARGE OF SPORES 3 jeunes." De Bary,1 Fayod,2 and others, however, have regarded paraphyses as distinct from basidia. My own observations made during the last thirteen years have convinced me that this view is the correct one. I shall presently describe new methods for dis- tinguishing the hymenial elements from one another, and shall show that with their help it has been possible to determine that paraphyses, i.e. cells which in general resemble immature basidia but which are destined from the first to remain sterile, are present in the hymenium of various species of Coprinus, Psathyrello, dis- seminata, Panaeolus campanulatus, Anellaria separata, Stropharia semiglobata, Psalliota campestris, Galera tenera, Bolbitius flavidus, and Lepiota cepaestipes. During the gradual exhaustion of the hymenium by the pro- duction and liberation of spores, the paraphyses lose all their protoplasm except a very thin wall-layer and, at the same time, become greatly swollen. The manner in which they assist the basidia to perform their functions will be discussed subsequently in connection with the description of various hymenia. Cystidia, like paraphyses, are sterile elements, but they differ from these in their larger size, their peculiar form, their smaller number, and frequently in the nature of their cell-walls and of their contents. They are present in some species and not in others. In some species where they occur, but not in all, they produce character- istic excretions, while in certain Coprini they have a mechanical function. They will be treated of more fully in Volume III. It is now known that the two nuclei present in each basidium from the beginning of its differentiation fuse together and thus consummate a sexual act of which the first stages take place in the mycelium.3 According to Ruhland,4 the well-marked paraphyses of Coprinus porcellanus at first, like the basidia, contain two nuclei ; but these two nuclei never fuse and, finally, as the paraphyses swell up and become poorer and poorer in protoplasm, undergo degeneration. Ruhland also found pairs of nuclei in the large 1 A. de Bary, Vergleichende Morph. u. Biol. der Pilze, Leipzig, 1884, p. 326. 2 V. Fayod, " Prodrome d'une Histoire Naturelle des Agaricines," Ann. Sci. Nat., T. IX, 1889, p. 253. 3 Vide infra, vol. iv. 4 W. Ruhland, loc. tit. 4 RESEARCHES ON FUNGI cystidia of Coprinus atramentarius ; but, here again, no nuclear fusions could be observed. It thus appears that paraphyses and cystidia differ from basidia in their nuclear behaviour. The Discharge of Spores from the Basidia. — The discharge of spores from their basidia was first observed in 1843 by Schmitz who studied the hymenium of the living fruit-bodies of Thelephora sericea (probably Stereum hirsutum). He discovered that the four spores of each basidium are discharged in succession.1 Brefeld, in 1877, stated that the four spores of each basidium of Coprinus stercorarius are shot off simultaneously.2 Zalewski, in 1883, gave an account of some observations on species of Coprinus and Russula, and upon Cantharellus cibarius, which afforded evidence that the spores of each basidium in these species are often shot off in succes- sion.3 Fayod, in 1889, stated that successive discharge takes place in Galera tenera* Finally, in 1909, in the first volume of this work, I showed by means of detailed observations made upon a number of species of Hymenomycetes that successive discharge of the four spores of each basidium is the rule, and that simultaneous discharge is either very rare or never occurs at all.5 That the spores are violently projected from their basidia was first asserted by Brefeld 6 in the account of his investigations on Coprinus stercorarius and Amaniia muscaria. Zalewski 7 and Fayod,8 who used other species, confirmed Brefeld's observations. In 1909, by employing precise methods, I established the fact of violent projection for the Hymenomycetes in general.9 Brefeld,10 in his foot-note on the discharge of spores in Coprinus stercorarius, asserted that after discharge drops of water are left on 1 J. Schmitz, " Beitrage zur Anatomie und Physiologic der Schwamme," Linnaea, Bd. XVTI, 1843, p. 434. 1 0. Brefeld, Botanische Untersuchungen uber Schimmelpilze, Heft III 1877 pp. 65-66. 8 A. Zalewuki, " ttber Sporenabschniirung und Sporenabfallen bei den Pilzen " Flora, Bd. LXVI, 1883, p. 266. 4 V. Fayod, foe. dt., p. 272. ' A. H. R. Buller, Researches on Fungi, vol. i, 1909, pp. 144-146. * 0. Brefeld, foe. dt., pp. 65, 66, and 132. ' Zalewski, foe. & 8 y Fayod> ^ ^ 271-272. A. H. R. Buller, foe. dt., vol. i. Chap. X, pp. 120-147. 10 0. Brefeld, foe. dt,, pp. 65-66 BASIDIA AND THE DISCHARGE OF SPORES 5 the tops of the sterigmata and that the spores cany part of the fluid contents of the burst basidium with them. Zalewski,1 who worked with several species of Hymenomycetes, denied that drops are left on the sterigmata after discharge. Without knowing of Zalewski's observations, I, too, failed to find the drops. In the first volume of this work I said : "At the moment of discharge of the spores from the basidia of Coprinus comatus, Polyporus squamosus, etc., I have endeavoured to observe drops on the vacant sterigmata, but without success ; nor, by using my Method I, have I been able to detect drops on any spores as soon as they have settled on glass immediately after leaving the basidia." 2 Fayod, in his account of spore-projection, makes no mention of any drop left on a sterigma after spore-discharge or on a discharged spore, but he discovered that in Galera tenera a drop of water is formed at the hilum of the spore (i.e. near the place where the spore joins the ^terigma) before discharge takes place.3 When the first volume of this work was published, I had not seen any drop arise at the base of a spore. However, in 1910, when study- ing some fruit-bodies of Coprinus curtus with the high power of the microscope, I rediscovered the phenomenon of drop-excretion just before spore-discharge. Afterwards I found that my observations simply confirmed those of Fayod which had been made twenty-two years previously.4 Further and much more extensive studies of spore-discharge have convinced me that throughout the Hymenomy- cetes it is an invariable rule that, just before a spore is to be violently projected, a water-drop is excreted at its base. The results of my investigations will now be given in detail. Calocera cornea has tiny orange-yellow finger-shaped fruit-bodies which are not infrequently seen on decaying stumps of trees in the 1 Zalewski, foe. cit., p. 266. 2 A. H. R. Buller, foe. cit., pp. 148-149. 3 V. Fayod, foe. cit. 4 Fayod's observations on Qale.ro, tenera are recorded in a single paragraph (foe. cit., p. 272) and are not accompanied by any illustrations The first illustra- tions of basidia showing the exact position and size of the water-drops are contained in my paper entitled: "Die Erzeugung und Befreiung der Sporen bei Coprinus sterquilinus," Jahrb. f. wiss. Bot., Bd. LVI (Pfeffer-Festschrift), 1915, pp. 299-329, Taf. II and III. This paper, written for a volume celebrating PfefTer's seventieth birthday, was sent to Germany in July, 1914, a fortnight before the outbreak of the war. 6 RESEARCHES ON FUNGI autumn (Fig. 1). Its basidia are provided with two long slightly divergent sterigmata which project into the air above the gelatinous matrix in which the bodies of the hymenial cells are embedded. In order to watch the discharge of the spores in this species, I proceeded as follows. Some fresh fruit-bodies were gathered in a wood at Winnipeg and were allowed to dry in the laboratory. A few days afterwards some of them were revived by giving them access to water without their being submerged. New sterigmata and spores were quickly produced. A single small fruit-body was placed horizontally on a glass slide and covered with a cover-glass. The edge of the cover-glass which was raised above the glass slide by the thickness of the fruit-body, i.e. by about 1 mm., was then sealed round with vaseline. Thus the fruit-body came to lie in a small chamber contain- ing moist air. On examina- tion with the microscope, a FIG. l.-Fruit-bodies of Calocera cornea on number of sterigmata and a piece of wood. Natural size. After spores could be Seen pro- Brefeld (Vntersuchungen, Heft VII, Taf. 11, Fig. 14). jectmg near the top of the fruit-body in a horizontal direction. With a magnification of 440 diameters, I watched a spore come into existence as a tiny rudiment on the end of a sterigma, grow to full size, and be discharged (Fig. 2) . The total length of time which was thus occupied by one spore was only 1 hour and 20 minutes, and by each of two other spores approximately 1 hour and 30 minutes. The stages in the development and discharge of the spores were recorded by means of camera-lucida drawings made at intervals. For the spore shown in Fig. 2 the successive intervals were each 10 minutes. In 40 minutes, the spore developed from a minute rudiment to full size (B, C, D, E, F). Then there was a pause for 40 minutes. At the end of that time a tiny drop of watery fluid began to exude at the neck of the sterigma (G). The drop quickly grew in volume, and in about 10 seconds from its first appearance attained its maxi- mum size (H). Immediately thereafter the spore was shot violently from its sterigma, and both spore and drop disappeared from view. Subsequent observations, soon to be described, taught me that, BASIDIA AND THE DISCHARGE OF SPORES 7 at the moment of discharge, the drop is carried with the spore. Hence, in Fig. 2 at I, the drop has been represented diagrammatically as L i A B C D E F K FIG. 2. — Calocera cornea. The development and discharge of a spore. A, an immature sterigma pushing out beyond the gelatinous matrix of the hymenium. B, a full-grown sterigma. C, D, E, and F, stages in the development of a spore, drawn respectively 10, 20, 30, and 40 minutes after the stage B. F, the full- grown spore on the end of its sterigma. G, the same spore 40 minutes later than the stage F : a drop of water is being excreted at the hilum. H, the same spore, about 6 seconds later than the stage G : the drop has attained its maximum size. I, a second or less later than H : the spore is being shot away from its sterigma. J, appearance of a spore after it has settled and after the drop which it carried has evaporated. K, a few minutes later than I : the sterigma sinking back into the hymenium. L, remains of the collapsed sterigma, a few minutes after the stage K. Magnifica- tion, 1,740. clinging to the discharged spore. The sterigma, a few minutes after losing its spore, gradually contracted and was finally drawn down to the general level of the hymenium where it became lost to view (K, L) . The foregoing observations teach us : (1) that in Calocera 8 RESEARCHES ON FUNGI cornea the spores develop and are discharged with hitherto unsus- pected rapidity, (2) that the excretion of a water-drop precedes violent spore-discharge, (3) that there is no drop left on the sterigma after discharge, and (4) that a newly-developed sterigma collapses a few minutes after discharging a single spore and, there- fore, that a sterigma does not produce any more spores than one. A spore of Calocera cornea, when coming into existence, arises obliquely and not directly at the end of the sterigma, as may be seen in Fig. 2 at C. The result of this is that a hilum is formed which projects toward the axis of the basidium-body. The excretion of the drop of water always takes place from the hilum. It there- fore seems probable that the hilum is morphologically and physio- logically of considerable importance. It is evidently a structure which is specialised for the rapid excretion of water, and it doubtless plays a very important role in bringing about spore-discharge. In different species of Hymenomycetes, the interval of time between the moment when a spore begins to develop on the end of the sterigma and the moment of discharge is by no means constant, as will be shown by data given in the next Chapter. For the present it is sufficient to remark that this interval, whilst very short for Calocera cornea, is still shorter for certain other Hymenomycetes. Indeed, in some species, e.g. Collybia velutipes and Dacryomyces deliquescens, it is reduced to even less than one hour. The Water-drop. — The excretion of a water-drop just before spore-discharge occurs not only in Calocera cornea but in numerous other, and probably in all, species of Hymenomycetes. Among some fifty species in which I have observed it may be mentioned those given in the Table on the opposite page. The species in this list are fairly representative of the Hymenomycetes in general. They are sufficient to show that drop- excretion before spore-discharge is the rule in this great group. With the high power of the microscope (magnification usually 440) I have watched hundreds of spores leave their sterigmata in different species, and never once, since I became acquainted with the excretory process, have I watched for the preliminary water- drop in vain. In order to observe individual spores leaving their sterigmata BASIDIA AND THE DISCHARGE OF SPORES in one of the Agaricineae, my mode of procedure was as follows. A gill was removed from a living fruit-body, placed flat on a glass slide, and then covered with a cover-glass. No mounting fluid was used. The upper hymenial layer was then looked over with the low power of the microscope. Often a basidium was observed to have only three or two spores left upon it. The high power Hymenomycetes Excreting a Liquid Drop at the Spore-hilum before Spore-discharge. Family. Order. Species. Tremellineae | Auricularieae . { Dacryomyceteae j Hirneola auricula-judae (Exidiaalbida JDacryomyces deliquescens \ Calocera cornea Thelephoreae Stereum hirsutum Clavarieae . Clavaria formosa Hydneae . * . . ( Hydnum imbricatum \ Hydnum f errugineum Polyporeae . Leucosporae j Polyporus squamosus (Polystictus hirsutus I Collybia velutipes | Lepiota cepaestipes Porphyrosporae j PsaUiota campestris (Stropharia semiglobata Agaricineae . Ochrosporae Rhodosporae Galera tenera f Claudopus nidulans 1 Entoloma prunuloides /Coprinus atramentarius 1 Coprinus com at us Melanosporae - Anellaria separata j Panaeolus campanulatus Ipsathyrella disseminata was then applied, and such a basidium watched intently. The remaining spores, as a rule, were then seen to be discharged successively at intervals of a few seconds or minutes. Sometimes my attention was directed to a basidium with four apparently ripe spores upon it. Occasionally all four spores were then seen to be discharged one after the other. For Hirneola auricula-judae, Stereum hirsutum, and Polystictus hirsutus, sections through the hymenium were employed. In these sections the basidia were seen in side view. When the upper hymenial surface of a gill which has been mounted in the manner just described is observed with the io RESEARCHES ON FUNGI microscope, the spores are seen to be in groups of fours, each group corresponding to a subjacent basidium. The drop of water, which i i i 4(4 • • • • C M . 4 64 64 FIG. 3. — Psalliota campestris. Diagrammatic representa- tion showing the successive discharge of the four spores from a basidium, as seen from above. A-D, stages in the discharge of the first spore ; E-H, of the second ; I-L, of the third ; and M-P, of the fourth. A, a drop of water has just begun to be excreted where the right- hand top spore joins its sterigma. B, about two seconds later : the drop has grown. C, about three seconds after B : the drop has attained its maximum size. D, a fraction of a second after C : the spore has disappeared carrying the drop with it ; the old position of the spore is represented by a dotted circle. Similar explanations apply to the other three horizontal lines of sketches, each of which represents the discharge of a single spore. The pause between the stages D and E, H and I, and L and M, was in each case a few minutes. All the stages from A to P were passed through in about 10 minutes. Magnification, 1,320. is excreted a few seconds before a spore is to be discharged, always arises at the hilum of the spore, i.e. at the junction of the spore with its sterigma. Now the hilum always faces toward the longitudinal axis of the basidium-body. Hence the water-drop BASIDIA AND THE DISCHARGE OF SPORES n faces in this direction also. For Psalliota campestris, the sequence of events for a single basidium during spore-discharge is represented semi-diagrammatically in Fig. 3. The hilum of the first spore to be discharged (A) excretes a drop of water which quickly grows to its maximum size (B and C). At this stage the spore is shot away and, at the same moment, the water-drop disappears (D). After a short interval of time, a few seconds or minutes, another spore is shot away (E, F, G, and H) ; then, after another interval, a third (I, J, K, and L) ; and finally a fourth (M, N, 0, and P). For each spore, as is shown in the Figure, discharge is preceded by the excretion of a drop. A side view of a basidium, showing stages in the discharge of the last two spores, is given in Fig. 4. The interval between the moment when a water-drop first becomes noticeable and the moment of spore-discharge was observed to be about 5 seconds. Not infrequently, soon after a drop has begun to form at the base of one spore, another drop begins to form at the base of a second spore, and so on. Thus in some basidia one may see two, three, or even four drops in course of growth at one time. The almost simultaneous production of three such drops is shown in Fig. 5. The relative ages of the drops are directly proportional to their size. Thus, at C, the drop on spore No. 1 began to be excreted before that on spore No. 2, and that on spore No. 2 before that on spore No. 3. The sterigma which begins to produce its drop first shoots off its spore first, and so forth. When the drops on a basidium appear almost simultaneously, the four spores are shot off their sterigmata in close succession. A basidium therefore sheds its four spores, after spore-discharge has begun, much more quickly when the sequence of events is as represented in Fig. 5 than when it is as represented in Fig. 3. The size of the drop excreted before spore-discharge is very constant in the fruit-bodies of any one species, but in different species it varies considerably. However, it bears a certain relation to the size of the basidia and of the spores. The following Table shows the relation between the size of the spores and that of the drops in two species. The fourth column gives the number of seconds taken for the growth of a drop before spore-discharge. 12 RESEARCHES ON FUNGI Fia. 4. — Psalliota campestris. The discharge of the third and fourth spores from a basidium as seen in lateral view. A— D, stages in the discharge of the third spore ; E-H, of the fourth. A, a drop of water has just begun to be excreted at the hilum of the right-hand spore. B, about two seconds later than A : the drop has grown. C, about three seconds later than B : the drop has attained its maximum size. D, a fraction of a second later than C : the spore has been shot away and has carried the drop with it. Between D and E there is a pause of several minutes. E, F, and G show the growth of the water-drop at the hilum of the fourth spore, whilst H shows the appearance of the basidium after the fourth spore has been shot away. Magnifica- tion, 1,320. BASIDIA AND THE DISCHARGE OF SPORES Size and Rate of Growth of the Water-drop. Species. Maximum Diameter of the Water-drop in fj.. Size of Spores in /t. Time taken for the Growth of the Drop in Seconds. Coprinus sterquilinus Psalliota campestris 5 2-3 18-20 X 11-12 9 X 5-5 15-30 5 At the moment when a spore is suddenly shot away from its basidium, so far as I have been able to observe, there is no * A A A 4^^ ^^J ^^ ^^3 FIG. 5. — Psalliota campestris. Diagrammatic representation showing the discharge of four spores in rapid succession from a basidium as seen from above. The dotted circles show the positions of spores which have been shot away. The spores are discharged in order according to their numbers 1, 2, 3, and 4. The times of each stage, taking the stage A as the zero of the time scale, are approximately as follows : B, 2 seconds ; C, 7 seconds ; D, 7 seconds ; E, 10 seconds ; F, 10 seconds ; G, 13 seconds; H, 13 seconds ; I, 16 seconds ; and J, 16 seconds. Thus all the spores are shot away in a quarter of a minute ; but the dis- charge is rarely so rapid as this. Magnification, 1,320. contraction of the basidium as a whole. Even when all four spores have just been discharged, the basidium appears to have just as large and turgid a body as before discharge began (cf. Fig. 4). It is evident that the spores are not squirted away from the basidium by the basidium-body contents and, therefore, that the mechanism of discharge is not analogous to that employed for the discharge of spores from the explosive asci of the Ascomycetes or for the discharge of the sporangium from the sporangiophore of Pilobolus. For some time after beginning to observe the excretory pheno- menon in connection with spore-discharge, I was puzzled as to what becomes of the drop at the moment when the spore is shot RESEARCHES ON FUNGI away. There were three obvious possibilities. Firstly, the drop might run down the sterigma and spread itself over the basidium- body ; secondly, the drop might become divided, part of it remaining on the sterigma and part of it being carried away by the spore ; and, thirdly, the drop might be entirely carried away by the spore. Against the first two suppositions could be urged the fact that no trace of the drop is ever to be seen on the sterigma or basidium-body immediately after a spore has been dis- charged. On the other hand, it was observed that when spores are allowed to fall on to a glass slide placed beneath a fruit- body one cannot observe that, immediately they come to rest, they are enveloped in a liquid film. Close observation, how- ever, has convinced me that the spores do carry away the Fio. 6. — Entoloma prunuloides. Proof that the water-drop is carried with the spore. A, two adjacent basidia on a living gill seen from above. B, the left-hand basidium has dis- charged three of its spores. C, at the base of the fourth spore a drop of water has been excreted and has attained its maximum size. D, the spore has been shot upwards and has fallen upon one of the spores of the right-hand basidium. The drop of water which the discharged spore has carried away has run between the two spores. Magnification, 660. drops with them. Evidence of this fact was obtained for Entoloma prunuloides in the following manner. Two basidia were found near to one another in the relative positions shown in Fig. 6 at A. After one of the basidia had shed three of its spores, I watched the fourth intently (B). Quickly a drop of water exuded from the hilum of the spore and in .5-10 seconds attained its maximum size (C). Then the spore was shot away and the sterigma was left vacant (D). The flight of the spore, as is usual in such observations, could not be followed with the eye. However, at the moment of discharge, the spore BASIDIA AND THE DISCHARGE OF SPORES 15 fell upon one of the spores of the other basidium and stuck to it ; whereupon I saw that a drop of water instantly ran from the discharged spore toward the undischarged, and became held by adhesion between the two (D). In a few seconds the drop spread itself more widely over the spores, and then disappeared, doubtless by evaporation. This observation leaves in my mind the con- viction that the drop, which I had seen on the hilum before discharge, had been carried by the spore during its flight. Observations similar in their nature to that just recorded I have occasionally made when studying spore-discharge in other species, e.g. Coprinus sterquilinus, Psalliota campeslris, and Galera tenera. In each of these a spore, after being shot upwards from the horizontal hymenium, fell on to one of the basidia or paraphyses and, at the moment of settling, was seen to have a drop of water on its upper side. The drop in each case was approximately of the same size as that which normally is excreted at the hilum. In a few seconds it disappeared, partly by spreading slowly around the spore and, doubtless, partly by evaporation. After making the occasional observations on the carriage of the drop just described, I devised an experiment which, provided living fruit- bodies are available, can be set up at any time and the results of which are sufficient to prove that the hilum-drop is carried by the spore in any hymenomycetous species whatsoever. The arrangement of the apparatus and the mode of observation for Nolanea pascua were as follows. A compressor cell, like that shown in the first illustration of Chapter II, was used as a moist chamber to contain a piece of a gill with a developing hymenium.1 A very tiny drop of water was placed on the base of the cell, and then a piece of a living gill dis- sected from a just-gathered fruit-body of Nolanea pascua was laid flat over the drop. The drop spread out into a fine film between the under side of the piece of gill and the glass. The lid of the compressor cell was then pushed down over the box until it was nearer to the tops of the upward-projecting spores than 0-1 mm., i.e. so that the under side of its cover-glass would form a target 1 A similar compressor cell was used for various investigations recorded in 1909 in Volume I. Vide vol. i, Fig. 58, p. 167. i6 RESEARCHES ON FUNGI -a which would be struck by the spores when they were shot upwards. Then, using the high-power objective of the microscope with a magnification of 440 dia- meters, I concentrated my attention for some hours on the largest, most angular, and pinkest of the spores, i.e. the ones which seemed most mature. These, on being dis- charged, instantly stuck to the cover-glass just above the sterigmata upon which they had been developed. I ob- served that every spore which struck the cover-glass target was accompanied by a drop of water, which ran between the under surface of the cover- glass and the spore. Once, by carefully focussing a ripe spore seated upon its sterigma, I observed the excretion of a small drop at the hilum of the spore, saw the drop grow for about 4 seconds to nearly full size (Fig. 7, A), then instantly raised the tube of the micro- scope so as to focus the lower surface of the cover-glass just above the spore, and then had the satisfaction of seeing the spore and the drop strike the cover-glass together and remain adher- ing to one another and to the cover-glass (B). Here, therefore, the drop was seen attached to one and the same spore both before and after discharge. Immediately after the spore had struck the target, I focussed the microscope upon the sterigma from which the spore had been shot and found it to be free from any drop. These B FIG. 7. — Diagram to illustrate the method used for observing the discharge of a spore from a gill of Nolanea pascua. A : the basidium shown was in the upward-looking hymenium of a gill contained in a closed compressor cell. The microscopist looked through the cover-glass c in the direction shown by the arrow a and observed the excretion of the drop d at the hilum of the spore s which was destined to be shot in the direction shown by the arrow 6. He then raised the plane of focus to the under surface of the cover-glass c and, within two seconds, as shown in B, saw the spore s with its attached drop d strike and stick to the cover-glass. BASIDIA AND THE DISCHARGE OF SPORES 17 observations permit only of one inference and that is that the drop excreted at the spore-hilum is carried by the spore in its flight through the air. The carriage of a drop of water by every spore which has just been discharged helps to account for the extraordinary adhesiveness of hymenomycetous spores when they are newly liberated. They stick to any surface which they happen to touch. After they have settled, if the air is not saturated with moisture, owing to the very small amount of water which they contain and have adherent to them, they quickly dry up. Whilst drying they press themselves more and more firmly to the substratum on which they lie. If they have settled on a glass slide and have dried there, it often happens that, if one attempts to remove them with a needle, they do not come away without breaking. Probably, when a moist spore settles on a glass slide, owing to capillary attraction, a little water passes from the spore so as to lie between itself and the slide. Assuming the existence of this water, it seems likely that, on its evaporating, it drags down the smooth-coated spore to the slide in such a manner as to flatten it out and thus to increase the area of contact. Sometimes the excretion of water from the hilum takes place in an abnormal manner. Several times I have noticed that, when a gill of a Mushroom which has been kept under moist conditions is mounted in air under a cover-glass, certain basidia which at the ends of their sterigmata bear partly or fully grown yet immature spores excrete water in abundance. The water arises from the hila of the spores in the form of drops (Fig. 8, A). These at first resemble the drops which are normally produced just before spore- discharge, but they continue to increase in size until they meet and fuse (B, C). One large drop is then held between the two young spores (D). Finally, the drop becomes so large that it runs down between the two sterigmata and disappears by moving on to the surface of the hymenium. A* second instance of abnormal excre- tion of water by a basidium is afforded by the following observa- tion upon Lepiota cepaestipes. A transverse section was taken through some gills and was mounted in water under a cover-glass. The water, owing to the imprisonment of air, did not invade every i8 RESEARCHES ON FUNGI interlamellar space. Into one such air-filled space the basidia were projecting freely. One of the basidia bore four spores which were just beginning their development (Fig. 9, A). From the hilum of one of the spores a tiny drop of water was excreted (A), which quickly grew larger (B), until in a few minutes it had reached a relatively enormous size (C). The drop thus precociously excreted was larger in volume than the basidium itself. The basidium did not alter in size during the excretion. We must therefore suppose that as much water flowed into the basidium from neighbouring B Fio. 8. — Psalliota campestris (cultivated form). A, B, C, and D, stages in the abnormal excretion of water from the hila of two spores of a basidium. A, one drop is of normal size, the other larger than normal ; the latter has stepped up on to the side of the spore. B, the drops have grown larger and both are now adherent to the sides of the spores. C, the drops have touched, and have united to form one drop which has dragged the spores toward one another. D, the drop has continued to increase in size and has now become globular, thus enclosing the two spores. Magnification, 880. cells as was excreted at the top of the sterigma. Normal discharge of a spore of Lepiota cepaestipes is shown in Fig. 9 at D, E, and F. At E the drop has reached its maximum size. The next moment the spore is discharged and the sterigma is left vacant, as is represented at F. Certain other remarks upon the abnormal excretion of water by basidia just prior to spore-discharge will be made in Volume IV in the course of a comparison of the discharge of basidiospores in the Hymenomycetes and the Uredineae.- The Chemical Constituents of the Drop.- The largest drops which I have yet seen are those excreted from the hila of the large spores of Coprinus sterquilinus, but their diameter is only 0-005 mm. They are so excessively minute and have such a brief existence BASIDIA AND THE DISCHARGE OF SPORES that, at present, it seems impossible to subject them to chemical analysis. However, a drop doubtless consists chiefly of water : for it has a perfectly rounded form during its excretion, behaves like a drop of water when touched, on contact with another similar drop unites with its fellow to form a single drop, refracts light like water, is colourless and transparent, and undergoes rapid evaporation in unsatur- ated air. But it would be rash to assert that the drop consists only of water in view of the investigations by Knoll 1 upon the supposed drops of water excreted from the ends of the pileal hairs of Coprinus ephe- merus, Psathyrella disseminata, etc., various cystidia, and the sporangiophore of Pilobolus. Knoll discovered that all these drops contain a colloidal con- stituent which appears to be mucilage. This he proved by observing the irregular form of the drops when contracting during evaporation, the solu- bility of the drops in water FIG. 9. — Lepiota cepaestipes. normal excretion of water from the hilum of a very rudimentary spore. A, first appearance of a drop at the hilum of the right-hand spore. B, the drop has grown and has enclosed the spore. C, the drop has now attained its maximum size. D-G, normal dis- charge of a ripe spore as seen from above. D, one spore is left on the basidium. E, a drop of water has been' excreted at the hilum and has attained its maximum size. F, the spore has been shot upwards and has fallen at G. The drop of water which was carried with the spore has evaporated and is not represented. Magnification, 990. and their insolubility in alcohol, etc. ; and he came to the conclusion that the mucilage is formed by local mucilaginisation of the cell- wall. Knoll also showed that, as the drops dry up, crystals of calcium oxalate are often formed. This occurs for example in the drops on the ends of the cystidia of Inocybe trichospora. In view of these observations it may well be asked : do the drops excreted on the hila of the spores of Hymenomycetes contain mucilage ? By analogy I strongly suspect 1 F. Knoll, " Untersuchungen tiber den Bau und die Function der Cystiden und verwandter Organe," Jahrb. f. iviss. Bot., Bd. L, 1912, pp. 453-501. 20 RESEARCHES ON FUNGI that they do, but so far I can produce no evidence in support of this supposition. It is not unlikely that a hilum-drop is only excreted when a minute portion of the wall of the hilum changes into mucilage, and that the mechanism of excretion is therefore similar to that for the excretion of drops on the pileal hairs of Coprinus ephemerus, the cystidia of Inocybe trichospora, and the sporangio- phore of Pilobolus. It is also possible that the drops in addition to water and mucilage contain crystalloidal bodies such as calcium oxalate, etc., but here again there is at present no evidence to support such a supposition. We may conclude, therefore, that the drops, consist chiefly of water but that, judging by analogy, it is probable that they contain other substances in solution. Stokes' Law. — In the first volume I gave an account of some experiments upon falling spores which were made for the purpose of testing Stokes' Law.1 My calculations, which were based on measurements of the density, radius, and rate of fall, showed that the spores of Amanitopsis vaginata fall 46 per cent, more rapidly than they should do according to Stokes' Law. The Law was therefore not verified in detail. The observations on the rate of fall were made upon spores which were just passing out from between two gills in a small compressor cell, the air of which was saturated with water vapour. At the time when my experiments were made, I did not know, as I do now, that a tiny drop of water is carried away upon the exterior of each spore. Now, according to Stokes' Law, the rate of fall of a minute sphere varies directly as the square of the radius. The addition of a water-drop to a spore increases its effective radius and therefore also its rate of fall. It seems to me, therefore, that the excess rate of fall which I observed was, in part at least, due to the water-drop which I had not perceived and which my calculations therefore did not take into account. With a correction of the radius based on the size of the drop carried away by each spore, my experimental results could be brought into closer agreement with Stokes' Law. To continue my investigations in this direction, however, has become unnecessary on account of the fact that, since my observations were published in 1909, Zeleny and M'Keehan, using a method far 1 Vol. i, 1909, p. 173. BASIDIA AND THE DISCHARGE OF SPORES 21 more delicate than my own, have shown that Stokes' Law for the fall of small spheres of a certain range of size in air is approximately true.1 These authors measured the density, radius, and rate of fall of minute spheres of wax, mercury, and paraffin, which were obtained by spraying. The rate of fall, as actually observed for spheres about the size of spores, was found to be in almost exact accordance with the Law. MiUikan,2 who has contributed much to the recent upbuilding of the atomic theory of electricity, by employing a method even more delicate than that of M'Keehan and Zeleny has shown that Stokes' Law, while true within 2 per cent, for spheres with a diameter of 0-001 cm. or 10 fi, i.e. for spheres about the size of spores, becomes increasingly inexact for spheres of smaller and smaller diameter. Stokes' Law, in its original form, is represented by the following equation : The Law, as corrected by MiUikan, has the following form : where V = the terminal velocity, p — the density of the falling sphere, o- = the density of the medium, g — the acceleration due to gravity, a = the radius of the falling sphere, jj, = the viscosity of the medium, p = the pressure of the gaseous medium, 6 = a constant (according to Millikan = 6-25 x 10~4 when p is measured in centimetres of mercury). Zeleny and M'Keehan measured the rate of fall of the dry spores of Lycopodium, Lycoperdon, and Polytrichum, with the result that the spores were found to fall much more slowly than they should do according to Stokes' Law. Thus the spores of 1 Zeleny u. M'Keehan, "Die Endgeschwindigkeit des Falles kleiner Kugeln in Luft," Physikalische Zeitechrift, Bd. XI, 1910, pp. 78-93. 2 R. A. Millikan, The Electron, Chicago, 1917, pp. 88-122. 22 RESEARCHES ON FUNGI Lycopodium fell with a velocity which was only about one-half of that required by theory. The fraction for the spores of Lycoperdon was about four-sevenths and that for Polytrichum about two- thirds. Zeleny and M'Keehan were unable to account for the want of agreement between fact and theory in the fall of dry spores. There is therefore room for further experiments in this direction. It is of some biological advantage that spores should fall but very slowly in still air. Possibly dry spores have some peculiarity in their construction, yet to be found out, which serves in a special manner to reduce the rate of fall to a minimum. The Mechanism of Spore-discharge. — The exact nature of the mechanism by which the spores are shot violently from their sterigmata still remains a mystery. In Volume I, I showed that the mechanism is not like that of asci : the basidium does not explode and squirt the four spores away. I suggested that the spores are discharged in the same manner as those of certain Entomophthoraceae. In these fungi a double membrane is formed between the spore and its basidium. At the moment of discharge these two membranes separate from one another, and each, owing to the osmotic pressure in the spore and in the basidium, suddenly becomes convex towards the other. The spore is thus jerked away. However, I pointed out that, for the Hymenomycetes, no one has ever seen any double membrane separating a spore from its sterigma. The neck of a sterigma where it joins the spore has so small a diameter that an optical difficulty prevents one from directly seeing what it contains.1 I am now inclined to believe that no such double membrane 1 W. E. Hiley, in describing the hymenium of Forms annosus (The Fungal Diseases of the Common Larch, Oxford, 1919, pp. 103-104), says : " Each spore is attached to the basidium by a very thin extension of the latter called a sterigma (Fig. 41, B, «0>arid when ripe the spore which develops at its extremity is cut off from the sterigma by a transverse septum . . . the ejection is accomplished by the splitting of the septum between the spore and the sterigma " ; and in his Fig. 41 he shows a diagram of a sterigma with a septum across the sterigmatic neck. Hiley's account of the septum in Fomes annosus appears to be based on theory rather than on actual observation ; and his diagram is evidently imaginary, for the sterigma is represented as being curved convexly toward the basidium-axis, instead of con- cavely as it should be, with the result that the attachment of the sterigma to the hilum of the spore is incorrectly illustrated. BASIDIA AND THE DISCHARGE OF SPORES 23 exists as that to which reference has just been made. Using a fruit-body of Coprinus sterquilinus, I tried the effect of breaking off spores from their sterigmata at a stage when the spores had attained full size but were only half -ripened. The protoplasm was still flowing into the spores, for the basidia had become only about half-emptied. I found that a half-ripened spore, isolated from its sterigma in the manner indicated, remained quite turgid although the canal leading into it through its membrane at the hilum was still open so far as the cell- wall was concerned. Plasmolysis could be effected readily with a solution of 10 per cent, potassium nitrate. These observations seem to show that the canal has so small a diameter that the osmotic pressure of the cell-sap is insufficient to press the protoplasmic contents of the spore through it. It is also very probable that, when a spore is removed from a sterigma, the opened sterigmatic canal is too narrow to permit of the osmotic pressure of the basidium driving the basidial contents out through it. We may therefore draw the conclusion that a double wall is unnecessary to close the ends of a spore and its sterigma at the moment of spore-discharge ; for, even if they remain open, the cell-contents cannot be forced out through them. Assuming tentatively that no double membrane is formed between a ripe spore and its sterigma, and that, when a spore is discharged, the end of the sterigma is simply broken across, what can be the nature of the force which causes spore-projection ? As yet I can give no satisfactory answer to this question. The remarks which follow are intended to stimulate further investiga- tion which may lead to a solution of the problem. Before assuming the presence of a double membrane between spore and sterigma at the moment of discharge, we ought to try to solve the problem by taking into account all the following facts : (1) The existence of a structure called the hilum — a tiny blunt process projecting from the base of the spore. This is always developed and is the point where a spore separates from the sterigma. It is therefore probably the seat of the force which brings about discharge. (2) The extreme narrowness of the necks of all sterigmata. The 24 RESEARCHES ON FUNGI diameter varies from 1 ^ in the large basidia of Coprinus sterquilinus to about 0-25 /* in species with very minute basidia. In Psalliota campestris the diameter is about 0 • 5 /A. (3) The successive and not simultaneous discharge of the four spores. (4) The maintenance of the turgidity of the basidium-body during the successive discharge of the spores. (5) The non-collapse of the sterigma immediately after it has discharged its spore. (6) The excretion of a drop of water at the hilum as an invariable preliminary to spore-discharge. (7) The carriage of the water-drop with the spore when this is shot away. (8) The fact that a spore does not burst but remains turgid when it is violently broken off its sterigma, at a time when the canal through the spore-membrane leading down to the sterigma is still open. (9) The minute diameter of the canal through the spore- membrane and sterigma. It may be regarded as certain that the mechanism for the discharge of each spore is localised at the end of each sterigma, and that the four sterigmata, so far as spore-discharge is concerned, act independently of one another. It is at the hilum that the force applied in the discharge of a spore becomes developed and effective. The hilum, since it excretes a drop of water of relatively consider- able size in from 5 to 10 seconds just before spore-discharge, evidently is able to control the passage of fluid through its membrane. It is possible that a spore is shot from its sterigma by hydrostatic pressure. It may be that, at the hilum, water is pumped into a small place by the protoplasm and that, when the hydrostatic pressure is increasing in this small space, a certain amount of the water escapes in the form of the drop which has already been described. Possibly the drop weakens the cell-wall on which it rests by acting chemically or physically upon it. We may then suppose that, when the membrane has thus been BASIDIA AND THE DISCHARGE OF SPORES 25 weakened up to a certain point, the hydrostatic pressure in the hilum causes an explosion, so that the spore is shot from its sterigma. The explosion, owing to its localisation at- the top of the sterigma, does not involve the whole basidium. Further, we must suppose that the amount of water liberated by the hydro- static explosion at the tip of the sterigma is so minute that, even with a microscope, there is little hope of our ever perceiving it. I do not feel that this theory is altogether satisfactory, but have brought it forward as a basis for further discussion of the whole question. The absolute constancy of the preliminary ex- cretion of a drop seems to make it certain that the drop plays an essential part in preparation for spore-discharge. I have assumed that the drop is useful in that it acts upon the hilum-membrane from without and thus weakens it prior to discharge. However, in Pilobolus and the asci of Ascomycetes, the membrane which breaks is weakened from within and no drop analogous to that excreted by the hilum of the basidium is ever produced. No support for my supposition as to the function of the drop can be obtained from the Phycomycetes or the Ascomycetes. I have suggested that the drop acts either chemically or physi- cally on the wall on which it rests and thus weakens it in preparation for spore-discharge. Possibly, however, the growth of the drop only accompanies such a weakening of the wall and is not the cause of it. The excretion of drops from the pileal hairs of Coprinus ephemerus, Psathrella disseminata, the cystidia of Inocybe tricho- spora, and the sporangiophore of Pilobolus, etc., has, as already pointed out, been shown by Knoll to be accompanied by the mucilaginisation of those small portions of the outer part of the cell- wall from which the drops arise. Judging by analogy, there- fore, it seems not only possible but even probable that, whilst the hilum-drop is being excreted, the wall joining the sterigma and spore is undergoing change — perhaps mucilaginisation — and is thus becoming weakened in preparation for spore-discharge ; and that, as soon as this weakening has attained a sufficient degree, the explosive force in the sterigma causes the spore to be shot away. It may be that the supposed mucilage, as it is being formed, pulls water from the sterigma by its force of imbibition and thus 26 RESEARCHES ON FUNGI forms the drop. Possibly, therefore, the formation of a drop is merely the outward and visible sign of the weakening of the cell-wall which takes place in preparation for spore-discharge, and is nothing more than a necessary physical accompaniment of an alteration in the chemical or physical condition of the cell- wall. The drop of water is always carried away by the spore at the moment of its discharge. This is not surprising since the hilum which excretes the drop is part of the spore. From the physical point of view it is therefore clear that the drop serves to increase the mass of the projectile. This increase of mass doubtless permits of the spore being shot further from the basidium than it would be without the addition of water to it. After a spore has reached an interlamellar space and begins to fall vertically, the water-drop very quickly dries up, so that it does not hasten the fall of a spore to the ground. One must admit, therefore, that the drop of water is useful in that, by increasing the projectile-mass, it assists the sterigma in its work of shooting the spore to a safe distance from the hymenium before the spore begins to fall vertically. The fluid which in Pilobolus clings to the sporangium, and which in Empusa, Ascobolus, and Peziza clings to the spores, is certainly of considerable importance as a means of increasing the projectile- mass, and thereby of increasing the distance to which the mass can be shot.1 I am inclined, however, to think that, while the drop excreted at the hilum of the spore in Hymenomycetes does increase the mass of the projectile and thus permits of the pro- jectile being shot a little further than it otherwise would be, this is only an incidental advantage and does not give us a clue to the primary significance of drop-excretion. It may be that the force of surface tension is used in some way to effect spore-discharge ; but exactly how I cannot satisfactorily explain. Possibly we have an entirely new principle involved in the mechanism of spore-discharge in the Hymenomycetes which, when properly elucidated, will prove of the highest interest and once more show how organisms are able to avail themselves of the forces of nature in order to maintain their existence. 1 Cf. vol. i, 1909, Part II, Chap. II, pp. 25S-254. BASIDIA AND THE DISCHARGE OF SPORES 27 Does a Basidium Produce more than One Generation of Spores ? — Corda, in 1839, stated that in the Agaricini each sterigma pro- duces several spores in succession, one after the other, in the same manner as the terminal branches of the conidiophores of certain Moulds.1 Schmitz, in 1843, after studying the production of spores from the living hymenium of Stereum hirsutum (his Thelephora hirsute), expressed the view that one and the same basidium can produce sets of four spores not merely once but many times.2 Worthington Smith believed that the basidium of a Mushroom (Psalliota campestris) produces two generations of spores, first two spores on two sterigmata and then two more on two other sterigmata.3 Recently, Istvanffi and Rene Maire have revived the theory of Corda and Schmitz. In order to explain some of the results of their nuclear studies, they have assumed that in some species a single basidium can produce more than one genera- tion of spores. This assumption has been made by Istvanffi for Hydnangium carneum* Rene Maire, in synthesising the results of his researches, states that in the normal types of Basidiomycetes the basidium produces a single generation of four spores, but that in Psalliota, Godfrinia, Cantharellus, Clavaria, Mycena, Geaster, etc., it produces either two, three, five, or six spores in a single generation, or two generations of four spores, or two generations of spores of unequal number 5 ; and finally, in his chief conclusions as regards facts, he holds that in many species of Basidiomycetes the basidium can produce two generations of spores on the same sterigmata.6 Here I wish to challenge Maire's conclusions in so far as they assume that the same basidium may produce more than one generation of spores. Maire worked on dead material. He fixed it, stained it, and cut it up with the microtome. He counted 1 A. C. I. Corda, Icones Fungorum huciisque cognitorum, Prag, T. Ill, 1839, p. 43. 2 J. Schmitz, "Beitrage zur Anatomie und Physiologie der Schwamme," Linnaea, Bd. XVII, 1843, p. 437. 3 Worthington Smith, "Structure of the Common Mushroom," Gardeners' Chronicle, Oct. 1876, p. 456. The accompanying illustrations are certainly erroneous. 4 Gy. von Istvanffi, " Ueber die Rolle der Zellkerne bei der Entwicklung der Pilze," Ber. d. D. bot. Gesell, Bd. XIII, 1895, p. 464. 5 Rene Maire, " Recherches cytologiques et taxonomiques sur les Basidio- mycetes," Bull. Soc. Myc. France, T. 18, 1902, p. 187. R Loc. cit., p. 204. 28 RESEARCHES ON FUNGI the number of nuclei in the basidia and spores. Never once did he actually observe one and the same basidium in the living state produce more than one generation of spores. He deduced his conclusions from his nuclear studies. Now my experience, based upon extensive observations on the living hymenium, is that a basidium never produces more than one generation of spores and that, after producing a single crop of spores, its sterigmata and body quickly collapse. In writing of Calocera cornea, Maire states that the secondary (fusion) nucleus of the basidium divides to form either two or four daughter nuclei, and that in the latter alternative there are two generations of spores : each spore receives only one nucleus.1 Now my observations on the fruit-bodies of Calocera cornea, which are epitomised in Fig. 2 (p. 7), go to prove that a sterigma, soon after shooting away its spore, collapses. This was observed for living basidia not once but several times. In no case did I see a sterigma produce two spores in succession. Direct observations, therefore, do not support Maire's assumption. In the cultivated Mushroom (Psalliota campestris) and in Coprinus bisporus Lange, the basidia are disterigmatic and there- fore produce only two spores at one time (Fig. 8, p. 18). In both these fungi, as I have observed by examining the living hymenium, there is only one generation of spores. In the Coprinus, each basidium, soon after shedding its spores, becomes involved in the upward moving zone of autodigestion and is therefore destroyed before there is any possibility of its producing a second generation of spores. In the Mushroom, shortly after a basidium has dis- charged the second of its two spores, its sterigmata collapse and its body shortens and sinks down into the hymenium. It can then be seen with two sterigmatic stumps at its outer end. In this condition it is no longer living. There is therefore not the slightest direct evidence to support the supposition that a basidium of Psalliota campestris ever produces two generations of spores.2 There is a relation between the volume of a basidium and the 1 Rene Maire, loc. cit., p. 78. 2 The collapse of the bisporous basidia of the cultivated Mushroom takes place in exactly the same manner as that of the quadrisporous basidia of the wild Mushroom. The phenomenon of basidial collapse will be dealt with in more detail in later chapters, e.g. Chapter X. BASIDIA AND THE DISCHARGE OF SPORES 29 total volume of the spores which it produces. In the cultivated form of Psalliota campestris the basidia are all of about the same size. Most of them produce two spores, but exceptions to this rule are not infrequent, for it is not difficult to find basidia which have but one sterigma and one spore only. The volume of the spore of a monosterigmatic basidium is twice that of one of the spores of a disterigmatic. A monosterigmatic basidium is equal in size to, and contains as much protoplasm as, a disterigmatic. There is therefore as much protoplasm available for the single spore of the first as for the two spores of the second. The sizes of the spores are evidently adjusted to the amount of protoplasm which they are destined to contain. In most species of Hymenomycetes, the spores take several hours to mature after coming into existence. After they have grown to full size, all the protoplasm of the basidium-body, with the exception of a thin lining layer, flows into them. There is therefore none left over which might be used for the production of a second generation of spores. This is true for all basidia, so far as my observations go,- whatever the number of spores which they produce. The whole mechanism of the basidium seems to me to be adapted for the production of a single generation of spores and no more. The Sterigma and Spore-hilum in Hymenomycetes and Gastro- mycetes. — Throughout the Hymenomycetes, the sterigma is well developed and consists of a distinct, longer or shorter, gently curved, tapering, conical body. Moreover, in all the species of this great group, a hilum, from which a drop of water is exuded just before spore-discharge, comes into existence in the first stage of each spore's development. The hymenomycetous spore, when full-grown, has a long axis which is inclined at an angle of about 45° to the long axis of the subjacent sterigma. Where the basidium is quadrisporous, the four sterigmata are symmetrically spaced about the free end of the basidium, so that the distances between successive sterigmata are all equal. The spores are symmetrically arranged in the same manner and all their hila are turned inwards toward the basidium-axis. This beautifully symmetrical quadri- sporous basidium is typical for the Hymenomycetes, and occurs 30 RESEARCHES ON FUNGI constantly in the vast majority of species, from Corticium to Coprinus, from Clavaria to Cortinarius, from Stereum to Schizo- phyllum, from Hydnum to Hygrophorus, and from Polyporus to Panaeolus. In the Tremellineae, which seem to form a link with the Uredineae, although the symmetry of the basidium is less striking than in the Thelephoreae, the Clavarieae, the Hydneae. the Polyporeae, and the Agaricaceae, the sterigmata and spore-hila are always well developed. In the non-tremelloid groups, a FIG. 10. — Scleroderma vulgare, a gastromycetous fungus, lacking sterigmata, commonly found under Oak trees, etc. Photographed by A. E. Peck at Scarborough, England. Natural size. few species have bisporous basidia and some others hexasporous ; but, in these and all other exceptional species, the sterigma and hilum have the normal development, and the different sterig- mata and spores are symmetrically arranged at the end of the basidium-body . In the Gastromycetes, on the other hand, the basidium has no typical or almost constant form. The number of spores on an individual basidium varies much. Thus in Scleroderma vulgare the number is from 2 to 5, while in Phallus impudicus it is usually about 9, these being packed tightly together, side by side in a bundle at the end of the basidium. The spore-hilum is usually not (perhaps never) developed at all. One may look for it in vain, BASIDIA AND THE DISCHARGE OF SPORES 31 for instance, in the two species already mentioned and in the various species of Lycoperdon. However, the most striking feature of the gastromycetous basidium is the variability of the sterigma. In some species, e.g. Phallus impudicus and Scleroderma vulgare (Figs. 10 and 11), the sterigmata are so exceedingly short that they only just suffice to attach the spores to the basidium-bodies from which they have been developed. In Lycoperdon, the sterigmata are usually long but extra- ordinarily thin. In L. gemmatum they are narrowly cylindrical in form and very variable in length and arrangement (Fig. 12). In L. nigrescens each basidium usually has four very slender sterigmata of about equal length, but the sterigmata break across and remain attached to the spores when these are set free. Further- more, in Gastromycetes, e.g. Phallus, Scleroderma, and Lycoperdon, each spore ,. . , . ,, is usually situated symmetrically upon the top of its sterigma instead of being laterally inclined. It will not be neces- n „ sary here to give any further details of the structure of gastromycetous basidia. Enough has been said by way of illustration, and we may now ask the question : why are gastromycetous basidia so different from hymenomycetous ? My own answer is as follows. The typical sterigma in association with the spore-hilum, such as we find generally in Hymenomycetes, is to be regarded as an organ for the violent discharge of the spore which the sterigma develops. This violent discharge of the spores, brought about by secret means by the sterigmatic guns, as I showed in Volume I, is of the very greatest importance to Hymenomycetes, in that it permits of the spores escaping from the hymenium into the air and thus becoming disseminated by the wind.1 In Gastromycetes, on the other hand, the mode of spore-liberation is entirely different from 1 Researches on Fungi, vol. i, 1909, pp. 133-134. n__Scleroderma vul. gare, one of the Gastro- mycetes. A piece of a mass Of tangled basidia lack sterigmata and the sessile spores are not violently discharged. After der Pike. Highly magni- RESEARCHES ON FUNGI that of Hymenomycetes. The spores of gastromycetous fungi are produced in closed chambers within the fruit-body instead of on a free external ^hymenium. These chambers only open after all the spores have ripened, so that the spores are unable to escape one by one as they come to maturity. When the spores become enveloped in a sweet fluid, as in the Phalloideae, the agent for spore-dissemination is insects ; but, if the spores dry up and become powdery, as in Lycoperdon, the agent is usually the wind. In no species of Gastromycetes do the basidia discharge their spores directly into the air. Now, in the closed chambers of gastro- mycetous fruit-bodies, violent spore-discharge would not only be useless but might be posi- tively harmful. Assuming that the Gastromycetes have been derived from the Hymenomy- cetes, it seems to me, therefore, that violent spore-discharge has become suppressed in the Gas- tromycetes. Assuming such suppression to have taken place, we may suppose that the well- developed sterigma, which, in association with the spore-hilum, had the function of violently propelling the spore in the ancestral Hymenomycetes, became useless during the development of internal hymenial chambers and, therefore, in the course of the evolution of the Gastromycetes, became either more or less suppressed or deformed. This degeneration of a once useful structure finds parallels in the reduction of the posterior pair of stamens in the flower of Sal via to functionless staminodes, in the suppression of the eyes but not the eye-stalks of certain American cave Crustacea, and in the almost total obliteration of the hind limbs of the Python. The degeneration of the typical basidium in cleistocarpous Basidiomycetes finds its parallel in the degeneration of the typical ascus in subterranean Ascomycetes. In Discomycetes, e.g. Peziza, Fia. 12. — Lycoperdon gemmatum, a common Puff-ball. Basidia from the hymenial chambers of a fruit-body. The sterigmata are cylindrical in form and variable in length, number, and arrangement. The spores lack hila and are not violently discharged. After Ed. Fischer, from the Pflanzen- familien. Highly magnified. BASIDIA AND THE DISCHARGE OF SPORES 33 Otidea, Ascobolus, and Morchella, which are subaerial, the ascus- gun is beautifully perfected : it is drawn out into a long tube with the eight spore-projectiles arranged in a row just behind the gun's apex, the ascus-wall is very elastic, the gun explodes at its apex by the breaking away of a little lid, and the eight spores are shot violently forward to a distance of some two or three centimetres. The smoothness of the spore- walls permits of the projectiles passing out of the end of the ascus-gun with the least possible friction.1 On the other hand, in some subterranean Tuberaceae, e.g. the well-known Truffles of the genus Tuber, where a violent discharge of the spores would no longer be of any biological advantage and does not take place, the ascus has lost its tubular form and is now oval or rounded, the spores are no longer arranged in a row one behind the other, the number of spores in each ascus is less than eight — usually two to four, the ascus-wall is not so elastic and does not open apically by a lid, and the spore- walls are no longer smooth. A comparison of gymnocarpous with cleistocarpous fruit-bodies in Basidiomycetes and Ascomycetes thus compels us to conclude that the typical forms of the basidium and of the ascus are intimately associated with the efficiency of these cell-organs in the performance of one of their chief functions, namely, the violent discharge of the spores ; and, at the same time, it teaches us that where the function of violent spore-discharge has become lost both the basidium and the ascus tend to undergo a series of degenera- tive modifications in structure. Balloons Falling from Rest or Fired from a Gun, Used to Illustrate the Movements of Spores. — It is impossible to observe the minute spores of the Hymenomycetes whilst they are being projected straight outwards from their sterigmata. In order to determine the trajectory of a spore I therefore proceeded indirectly. The maximum horizontal distance of propulsion and the terminal rate of fall of a spore in still air were first measured, and then the data so obtained were introduced into an equation based on Stokes' Law. By means of a series of calculations made with this equation the required trajectory was then plotted out.2 This trajectory, 1 For a discussion of the form and function of the typical ascus vide vol. i, pp. 240-247. 2 Vol. i, 1909, pp. 184-189, Figs. 64-66. 34 RESEARCHES ON FUNGI which I called a sporabola, is remarkable in that roughly it can be divided into two parts — a part due to the projection of the spore and a part due to the fall of the spore towards the ground. A spore, shot out horizontally from a gill, goes forward for about 0-1 mm. and then makes a sharp turn through a right angle, after which it falls under gravity vertically downwards with a constant terminal velocity. To demonstrate the nature of this peculiar trajectory on a large scale, I have invented the balloon-gun which has been so named because its projectiles are balloons. In the development of certain guns used in the Great War, the object has been to throw a projectile of large volume and high density a distance of several miles. In developing my balloon-gun, on the other hand, the object has been to throw a projectile of large volume but very small density (i.e. just exceeding that of air) a distance of a few feet only. Thus, while the size of my projectiles has not greatly differed from the size of the shells of large field guns, the density has been very much smaller. My projectiles, it must be admitted, although of considerable scientific interest, appear somewhat ridiculous to those who see them in action for the first time, but they have this ad- vantage over the shells of the battlefield : their motions provoke laughter rather than tears. The balloons which I have used as projectiles have been spherical and oval toy-balloons which, when inflated, resemble in form the two chief types of basidiospores. Their size was varied by choosing larger or smaller envelopes and by altering the amount of gas with which any one balloon was inflated. The spherical balloons were mostly about 6 inches in diameter and the oval ones 10-24 inches long and 6 inches wide. The gas pressed into a balloon was either air or a mixture of air and coal-gas. The mixture, of course, is lighter than air ; and, by filling a balloon with a mixture in which the components were suitably balanced, the rate of fall of the balloon could be reduced to whatever was desired. However, most of my experiments were carried out with balloons filled with air. A toy-balloon which falls very slowly in air resembles a spore in that it has an enormous volume and surface area relatively to its mass. Therefore, not only in having a spherical or oval BASIDIA AND THE DISCHARGE OF SPORES 35 shape and a smooth wall but also in its peculiar volume-to- mass and surface-area-to-mass relations, a toy-balloon shows itself particularly well fitted to be used as a large-size model of a basidiospore. The rate of fall of a spore varies not only with its density but also with its volume and with the amount of its surface exposed to the air, so that, with density constant, the smaller a spore, the slower is its terminal velocity. A variation of the rate of fall with volume can readily be demonstrated to an audience with a balloon. Thus, for instance, it was observed that the same balloon when falling 11 feet in still air took: (1) when uninflated, 1-5 seconds ; (2) when partly inflated so that it was 10 inches long and 6 inches wide, 4 seconds ; and (3) when fully inflated so that it was 23 inches long and 6 inches wide, 5 seconds. In its three states, in succession, the balloon had about the same weight in air but there was a progressive increase in volume and surface area. The larger the volume of the balloon, the slower was its rate of fall. Any homogeneous elongated body falling in still air tends to orientate itself in such a way as to present the maximum amount of surface in the direction of the line of fall and thus to fall as slowly as possible. In accordance with this law a piece of paper 4 cm. long and 0 • 5 cm. wide, when falling, does not approach the earth with an end- wise movement but rather keeps its long axis directed transversely to its line of fall. For certain reasons which will not be discussed here, the piece of paper usually rotates rapidly about its long axis during its descent. An elongated cylindrical object, such as a piece of cotton fibre, hair, or spider's web, when falling (as may be seen in a beam of sunlight), also keeps its long axis parallel with the earth's surface but, unlike the flat piece of paper, does not rotate about its axis. Still smaller cylindrical bodies, such as the spores of Claviceps pur pur ea, doubtless fall like bits of cotton fibre ; but even oval spores, which are only two or three times as long as they are wide, also tend to fall with their long axes held horizontally. This I directly observed with a horizontal microscope for the spores of Polyporus squamosus, which are three times as long as they are wide . 1 Now this mode of falling of elongated 1 Vol. i, 1909, pp. 185-189, Figs. 64-66. 36 RESEARCHES ON FUNGI spores can be beautifully illustrated with an elongated balloon as follows. A balloon which is 24 inches long and 6 inches wide and having a density just exceeding that of air, is taken in the hand and held with its long axis perpendicular ; and it is then allowed to fall from rest. Almost immediately after its liberation, it turns FIG. 13. — The balloon -gun for discharging balloons. A balloon is shown held in position on the platform just before the slamming-board is released by pulling the trigger. The base -board was 2 feet long. through a right angle and comes to place its long axis horizontally ; and so directed the long axis remains until the balloon has reached the ground. Such an elongated balloon, however liberated, always quickly turns, just as does an elongated spore, until its long axis comes into a horizontal position. When projected horizontally, toy-balloons filled with air or a mixture of air and coal-gas have a trajectory exactly resembling that which I have described for spores.1 The simplest way of 1 Vol. i, 1909. pp. 185-189. Figs. 64-66. BASIDIA AND THE DISCHARGE OF SPORES 37 projecting a balloon horizontally forward through the air is to use one's hands for the purpose. A spherical balloon 6 inches in diameter cast by hand as hard as possible in a horizontal direction travels rapidly forward for about 2 or 3 feet and then makes a sudden sharp turn downwards, after which it falls with a slow and even terminal velocity vertically downwards toward the earth. The resistance of the air reduces the horizontal projection- velocity of the balloon to zero in less than a second and, by the end of this time, the balloon has ceased to be carried forward and is simply falling under the influence of gravity. Thus the sporabolic trajectory, which for a spore has a horizontal component only 0 • 1-0 • 2 mm. long, can be demonstrated on a large scale to an audience of considerable size. The balloon-gun which I invented to save myself the trouble of projecting the balloons forward with my hands was of very simple construction (Fig. 13). It was made of wood and consisted of the following parts : (1) a base-board 20 inches long and 8 inches wide, (2) two straight upright arms, 8 inches high, (3) a platform for the projectile, attached to the two arms, (4) a slamming-board, 15 inches long and 8 inches wide at its base, hinged below to the base-board along a median transverse line, movable about its hinges through nearly a right angle, i.e. from the nearly horizontal position shown in Fig. 13 to a vertical position, in the latter position projecting 1 inches above the top of the platform, (5) a strong spring attached to the middle of the slamming-board and to the base-board, and tending to bring the slamming-board into a vertical position, (6) a triangular trigger-board attached to the base-board, and (7) a trigger attached to the trigger- board, passing upwards through a slot in the centre of the slamming-board, and provided above with a catch. In Fig. 13 the catch of the trigger is holding the slamming-board down. The balloon to be used as a projectile was set on the top of the platform with one side projecting toward the slamming-board ; and it was held in position very lightly by pressure above applied with a finger. The gun was then fired by pulling the trigger. The movement of the trigger released the catch, whereupon the slamming-board was caused by the spring to swing sharply through 38 RESEARCHES ON FUNGI nearly a right angle into a vertical position where further move- ment was stopped by the arms and platform. The slamming- board, which was 7 inches longer than the platform was high, was thus caused to strike the projecting part of the balloon with con- siderable violence. In response to the blow from the slamming- board, the balloon shot forward into the air through a horizontal distance of about 3 to 6 feet ; and it then made a sharp turn downwards, thus completing its sporabolic trajectory. The gun was used to discharge both spherical and oval balloons in a horizontal direction, and the results obtained with it did not differ in any essential from the results obtained by casting the balloons forward with the hand. The balloon shown in Fig. 13 was 1 0 inches long and slightly oval. When fired from the platform, it shot forward to a distance of 6 feet before beginning to fall verticaUy. When an elongated balloon, 13 inches long and 6 inches wide, was set on the discharging platform with its long axis in the direction of the axis of the gun and of the line of flight and was then dis- charged, it was observed that, during its flight, the balloon turned through an angle of 90° so that its long axis, while remaining horizontal, came to be turned in a direction which was transverse to the line of flight. The same result was obtained by casting elongated balloons forward by hand. This turning movement is exactly what one should expect on the principle that a moving body tends to present its greatest surface to the resistance of the air. Theory and observation both permit one, therefore, to draw the conclusion that an elongated spore, when shot from a sterigma or out of an ascus and whilst moving against the resistance of the air, must tend to turn its long axis from its original direction through a right angle. A Comparison of the Rates of Fall of Spores and Thistle-down. — If the spores of Hymenomycetes, when suspended in the air on an autumn day, were to have their diameters suddenly increased 100 times without this increase altering their normal rate of fall, we should see them with the naked eye, and they would often darken the air and astonish us with their vast numbers and extra- ordinary variety. We should be most struck, perhaps, by the slowness of their rate of fall. In particular, it would be necessary BASIDIA AND THE DISCHARGE OF SPORES 39 for us to watch the smallest spores very carefully indeed if we wished to make sure that they were actually falling. Every one has noted with delight the very slow rate of fall of thistle-down when being driven before the wind. Yet thistle-down falls very rapidly relatively to the spores of the Hymenomycetes. This fact I have proved by direct comparison of the rates of fall of thistle-down and spores in still air. Cnicus arvensis, which in Canada is called the Canada Thistle although it was introduced from Europe, is as common a weed in Manitoba as in England. I brought some capitula of this species, which bore ripe but not yet fully expanded fruits, into the laboratory at Winnipeg and allowed them to dry. Soon, each fruit opened its parachute-like pappus and some of the fruits became detached from their discs. I took four of these detached fruits and measured their rates of fall in still air through a distance of 6 • 9 feet. Many measurements were made and it was found that the fruits fell 6-9 feet in 11-12 seconds. The average rate of fall was 6-6 inches or 167-6 mm. per second. The rates of fall of the spores of various species of Hymeno- mycetes were measured, as described in Volume I, with a hori- zontal microscope and a drum recorder.1 The large spores of Amanitopsis vaginata fall in moderately dry air at the rate of 4-29 mm. per second, the medium-sized spores of the Wild Mush- room (Psalliota campestris) and of Polyporus squamosus at 1 -61 mm. per second and 1-03 mm. per second respectively, and the small spores of Collybia dryophila at 0-49 mm. per second. These rates of fall along with the rate of fall of the fruits of Cnicus arvensis may be set out in tabular form as follows : Rates of Fall in Still Air of Thistle Fruits and of the Spores of Hymenomycetes Cnicus arvensis . . . 167-6 mm. per second Amanitopsis vaginata . . 4-29 ,, ,, Psalliota campestris . . : .. 1-61 ,, ,, Polyporus squamosus . . 1-03 ,, ,, Collybia dryophila . . . 0-49 ,, ., 1 Vol. i, 1909, pp. 166-177. 67S39 40 RESEARCHES ON FUNGI From these data a simple calculation shows that the Thistle fruits fall with a steady terminal velocity which is : 39 times that of the spores of Amanitopsis vaginata, 104 „ ,, „ „ the Wild Mushroom, 162 „ „ „ „ Polyporus squamosus, and 342 ,, ,, „ „ Colly bia dryophila. These figures serve to indicate how very much more slowly the spores of Hymenomycetes fall than the parachute-like fruits of a typical Thistle. On a windy day, one often sees thistle-down blowing for very long distances over hills and plains ; but, on such a day, the invisible spores of the Hymenomycetes which, as we have seen, fall so very much more slowly than thistle-down, must be transported by the wind with even much greater ease than this delicate material and must, therefore, often be carried for many miles before they finally come to earth. A Comparison of the Rates of Fall of Spores and Bacteria. — In still air, the spores of the Hymenomycetes fall much more slowly than thistle-down or the winged seeds and fruits of Flowering Plants but, nevertheless, they fall very much faster than Bacteria. Among organic bodies adapted for passive dispersal by the wind, they therefore occupy only an intermediate position. Spherical species of Bacteria — Micrococci, Streptococci, etc. — are much smaller than the spores of the Hymenomycetes. Their diameters vary from about 2 to 0 • 2 /M. The rate of fall of individual bacteria may be calculated by using the second equation given on p. 21. Now, for our purposes, p may be taken as 1, a- as negligibly small, p, as 1 • 8 X 10"4, g as 1000, and p as 76 cm. of mercury. The following Table gives the diameter, radius, terminal velocity of fall in still air, and time required to fall 2 • 5 cm. (approx. 1 inch) for the individual cells of certain species of spherical bacteria. The cells of the largest species, Micrococcus Freudenreichii, which have a diameter of 2 p, and a radius of 1 p, fall only about one-eighth of a millimetre in a second and, to fall 2 • 5 cm. or 1 inch, require 3 minutes and 8 seconds. The cells of the smallest species, Streptococcus gracilis, which have a diameter of 0'2 ^ and a radius of 0 • 1 /*, fall only about BASIDIA AND THE DISCHARGE OF SPORES 41 one five-hundredth part of a millimetre in a second and, to fall 2 • 5 cm. or 1 inch, require 3 hours and 5 minutes. The ideas of most people concerning the speed of falling bodies are gained by observing the fall of comparatively large objects, and it is therefore a little difficult to realise that there are living organisms which, when falling in still air, require upwards of three Rates of Fall of Spherical Bacteria in Still Air. Species. Diameter inM. Radius in/u- Terminal Velocity in mm. per sec. Time taken to Pall 2- 5 cm. or 1 inch. Micrococcus Freudenreichii 2-0 1-0 0-133 3 mins. 8 sees. Micrococcus bicolor 1-2-1-6 0-7 0-0675 6 mins. 10 sees. Micrococcus luteus 1-2-0-4 0-4 , Micrococcus roseus 1-0-0-6 0-4 [ 0-0237 17 mins. 30 sees. Micrococcus pyogenes aureus . 0-8 0-4 J Micrococcus cerasinas 0-3 0-15 0-00428 1 hour 37 mins. Streptococcus gracilis x . 0-2 0-1 0-00224 3 hours 5 mins. hours to fall a single inch ; but such an organism is Streptococcus gracilis. The individual cells of this bacterium, under natural con- ditions, probably fall even more slowly than our calculation has shown owing to the fact that, when isolated in the air, they would be dried up and therefore have a smaller diameter than 0 • 2 ^ which was measured when they were immersed in a culture medium. The rates of fall per second and the length of time required to fall an inch for (1) thistle-down, (2) moist basidiospores,2 and (3) bacteria may be tabulated as shown in the Table on page 42. From a comparison of the data given in the Table it is evident that the rates of fall of basidiospores are intermediate between the rates of fall of Thistle fruits and of the cells of bacteria. Thus the rate of fall of a Thistle fruit is about one hundred times that of a Mushroom spore, but the rate of fall of a Mushroom spore is about twelve times that of the cells of Micrococcus Freudenreichii and about seven hundred times that of the cells of Streptococcus gracilis. The minute size of wind-disseminated cells, whether they be 1 The velocity of fall here given for Streptococcus gracilis is 82 per cent, greater than that given by the uncorrected equation for Stokes' Law (first equation on p. 21), while that for Micrococcus Freudenreichii is only 8-2 per cent, greater. 2 Vol. i, 1909, p. 175. RESEARCHES ON FUNGI basidiospores, mould spores, yeast cells, bacteria, or pollen grains, is advantageous to the species concerned in at least two ways : (1) it permits of these cells being produced in vast numbers so that the chances that some of them will settle in favourable positions is greatly increased and (2), being correlated with a very slow rate of fall, it enables the wind to carry the cells relatively very long distances before they settle, thus assisting the agent of dispersal. In concluding this section, it may be pointed out that a spore or Comparative Rates of Fall of Thistle-down, Basidiospores, and Bacteria in Still Air. Objects Falling in Still Air. Rate of Fall in mm. per second. Time taken to Fall 2 • 5 cm. or 1 inch. A fruit of Cnicus arvensis lfiT-6 0-15 second A spore of Amanitopsis vaginata A spore of a Wild Mushroom . 4-29 1-61 5 • 8 seconds 15-5 seconds A spore of Polyporus squamosus 1-03 24-3 seconds A spore of Collybia dryophila . 0-49 51-0 seconds A cell of Micrococcus Freudenreichii 0-133 3 minutes 8 seconds A cell of Streptococcus gracilis 0-00224 3 hours 5 minutes bacterium, on being cast into the air, must, unless the air is abso- lutely saturated with water vapour, dry up immediately. The rapidity with which basidiospores dry up when they fall from the hymenium of an agaric on to a slide placed beneath them in a laboratory, is well known, and I have remarked upon it in Volume I.1 In ordinary unsaturated air, either in the laboratory or in the open, the spores of Hymenomycetes dry up in a few seconds. Doubtless, under similar conditions, Micrococci and Streptococci dry up in less than one-tenth of a second. A spore or bacterium, therefore, when floating in the air, must be thought of as existing in a condition not of turgidity but of desiccation. 1 Vol. i, 1909, p. 162. CHAPTER II THE RATE OF DEVELOPMENT OF INDIVIDUAL SPORES IN DIFFERENT SPECIES Introduction and Table of Results — Methods — Discussion of Results — Effect of Temperature Introduction and Table of Results.— In order to increase the exact data concerning the production and liberation of spores in the Hymenomycetes, it seemed to me advisable to measure the interval of time which elapses between the first appearance of a spore on the end of the sterigma as a tiny rudiment and the moment of spore-discharge. The first observations of this nature were made upon Calocera cornea and have already been recorded in the previous Chapter. Subsequently, the investigation was extended to a number of other species with spores differing from one another in size, shape, colour, and thickness of cell-wall. The results obtained have been embodied in the adjoining Table. The time given in the fourth column is the average time taken for the de- velopment and discharge of the first of the four spores to be shot away from each basidium. The number of basidia upon which observations were made was usually from three to six. Methods. — The method generally employed for observing spore- development in the Agaricineae was as follows. A tiny drop of water was placed on the middle of the base of a compressor cell (Fig. 14) and spread out. The whole of a small gill or a large fraction of a large gill was then dissected from a freshly-gathered vigorously-growing fruit-body and laid over the drop, so that the water partly or wholly filled the space between the gill and the glass. Then another small drop was placed on another part of the compressor-cell base. Finally, the cell was closed by its glass cover which was pressed down a little way. The upper side of the 43 44 RESEARCHES ON FUNGI gill thus came to be exposed to moist air, without being wetted by free water or receiving mechanical interference from the compressor-cell cover. Usually the hymenium was observed with the low power of the microscope (Zeiss : objective AA, eye-piece 4, tube drawn out full, magnification about 130) ; but, sometimes, when the lid of the compressor cell had been pressed down so that it was in contact with part of the upper surface of the gill, it was Average Time taken for the Development and Ripening of an Individual Spore from the Moment the Spore appears on the Sterigma as a Tiny Rudiment until the Moment of Discharge. Species. Hrs. Mins. Spore-colour. Spore-wall. 11 Collybia velutipes . 0 47 colourless smooth '* S Dacryomyces deliquescens . 0 50 2 .2 Collybia fusipes . • 1 3 Q) 4) -° -3 Marasmius oreades 1 5 00 § S Exidia albida . 1 15 1 ept. of Agriculture, Butt. No. 149, p. 29. 2 G. P. Atkinson, Mushrooms, Edible and Poisonous, New York, 1911, p. 194. 3 Fomes officinalis is the original agaricum of the ancient Greeks and Romans and has been of importance in medicine as a purgative from the time of Dioscorides FOMES APPLANATUS 127 (Fig. 45). The number of its rings is 45. The average thickness of its rings therefore just exceeds one-half of an inch. In this fungus, owing to the distinctness of the rings on their exterior and owing to their great depth, I do not doubt that each ring represents the whole of the growth of the fruit- body for the year in which it was formed. When the 45-year-old fruit-body just described was only six years old, it must have had only six annual tube-rings and must have resembled in form the 6-year-old fruit-body shown in Fig. 46 ; and, when it was fifteen years old, it must have had fifteen annual tube-rings and must have resembled in general form the less evenly grown 15-year-old fruit-body shown in Fig. 47. The maximum age for certain species of Polyporeae, so far as onwards. Vide A. H. R. Buller, " The Fungus Lore of the Greeks and Romans," a presidential address, Trans. Brit. Myc. Soc., vol. v, 1914, pp. 58-60. J. H. Faull has given us an interesting account of the fungus (" Fames officinahs, a Timber-destroying Fungus," Trans. Roy. Canadian Institute, Toronto, vol. xi, 1916, pp. 185-209, Figs. 1-30) in which he makes the following statements. A fruit-body has the odour of fresh meal and an extremely bitter quinine-like taste. Long before the recognition of F. officinalis by botanists in America, the fruit - bodies were used by the early settlers in Ontario and Quebec for various purposes, including the preparation of yeast for bread-making, and the plant was known to them as the Pineapple Fungus (Pineapple tree, an obsolete English name for a pine or coniferous tree ; pineapple originally meant a pine cone). The first scientific record of the fungus in America appears to have been not earlier than 1886. There is some evidence that the Indians knew of its medicinal value. The active principle is a resinous substance agaricin ; and this, with other resins, constitutes up to 70 per cent, of the dry weight of the fruit-body. The resins are secreted in the form of amorphous granules to a very slight extent on the mycelium, but in great abundance on the hyphae of the sporophore. Quelet and certain other European systematists have assumed that F. officinalis is a variety of Polyporus sulphureus or specifically very close to it. However, these two species are very distinct, differing in : (1) size and branching of hyphae, (2) form, longevity, and content of sporophore, (3) structure of sporophore, (4) size of spores, and (5) cultural characters. F. officinalis causes a red heart-rot of conifers characterised by a removal of the cellulose, a fractur- ing of the wood into rectangular masses, and the formation of mycelial sheets in the crevices. Histologically the effects are similar to those caused by P. Schweinitzii. It occurs on living and dead timber and is a wound parasite. In Canada it has been found in Quebec, Ontario, and British Columbia ; and in the United States in Arizona, California, Oregon, Washington, Montana, Nevada, Idaho, Wisconsin, Michigan, and Wyoming. In Europe it occurs on Larix europaea and L. sibirica ; and in North America on Abies concolor, A. magnifica, A. grandis, Larix occidentalis, L. laricina, Picea Engelmanni, P. sitchensis, Pinus lanibertiana, P. murrayana, P. ponderosa, P. Jeffreyi, P. strobiis, P. monticola, Pseudotsuga taxifolia, Tsuga heterophylla, and T. mertensiana. 128 RESEARCHES ON FUNGI it has been ascertained by counting the successive tube-layers, is summarised in the following Table. Age of Oldest Known Fruit-bodies of Certain Polyporeae. Species. Age in Years. Observer. Polystictus versicolor Polyporus squamosus Fomes applanatus . Fomes officinalis 1 1 10 45 E. Fries, etc. . E. Fries, etc. J. H. White J. H. Faull Fomes igniarius 80 G. F. Atkinson The Hymenial Tubes. — The fruit-body of Fomes applanatus is bracket-shaped and broadly attached to its woody substratum. Its broad attachment and its remarkably hard upper crust serve to keep the hymenial tubes fixed permanently in one position. The hymenial tubes are very long and yet extremely narrow. In one large fruit-body, in one layer, I found them to be 12 mm. long, except toward the margin of the pileus where they became pro- gressively shorter, but they were only 0-17 mm. in internal diameter. In another very large three-year-old fruit-body (Fig. 48, p. 131), where the internal diameter of the tubes was also 0-17 mm., the length of the tubes (except near the margin of the pileus) varied in the first tube-layer from 9 to 12 mm., in the second from 19 to 28 mm., and in the third from 18 to 20 mm. In large well-grown fruit-bodies, therefore, the majority of the hymenial tubes vary in length from about 10 to 20 mm. and exceptionally attain a length of 28 mm., while their average internal diameter is only about 0*17 mm. From these data it can be calculated that the ratio of tube-length to tube-breadth in large fruit-bodies varies from about 60 to 120 and, exceptionally, is as great as 170. Now, as the hymenial tubes are relatively so very long and yet so very narrow, it is of high importance for the escape of the adhesive spores that the cylindrical lumen of each tube shall have its axis perfectly vertical ; for the least deviation from this direction, i.e. a deviation of 1° or even less, will cause the spores shot from the basidia in the upper and middle parts of each tube into the intratubular cavity to strike the sides of the tube during their fall and to adhere there FOMES APPLANATUS 129 (Fig. 49). The mode of attachment and the mechanical rigidity of the fruit-body as a whole are therefore important factors in bringing about the successful liberation of the spores.1 FIG. 46. — Fames officinalis. A fruit-body with 5-6 layers of hymenial tubes, and therefore 5-6 years old, growing upon the trunk of Paeudotsuga taxifolia in Montana. Photograph supplied by G. G. Hedgcock of the U.S.A. Department of Agriculture. About \ natural size. The width of the fruit-body was 7 inches. The hymenial tubes are not only often upwards of 1 cm. in length and less than 0-2 mm. in diameter, i.e.. very long and narrow, 1 Cf. vol. i, 1909, pp. 39-40. 130 RESEARCHES ON FUNGI but also closely crowded (Fig. 48). In one fruit-body which I investigated, there were 2,080 tubes in 1 square cm. of a tube-layer. Since each tube is lined with a hymenial layer from top to bottom, other things being equal it follows that the longer and narrower the tubes, and the more the tubes are crowded together, the more spores can be produced by the tubes above any given area of the pored surface. As recorded in Volume I, the inner surface of the tubes of a single layer of tubes in one fruit-body was found to be 148 times as great as the flat under surface of the flesh to which the tubes were attached, so that the specific increase of the hymenial surface of the fruit-body due to the presence of a single layer of tubes was 148. Taking three years together the specific increase was 49 3. * Owing to the high specific increase of hymenial surface it can be no matter for surprise that, as White has directly observed, Forties applanatus produces an almost incredible number of spores. White, using the methods I employed for Polyporus squamosus, found that a fruit-body of Forties appUnatus having a pored area of about 1 square foot liberated 30,000,000,000 spores in 24 hours.2 One square* foot equals 930 sq. cm. Reckoning 2,080 tubes to 1 sq. cm., the number of tubes on the under side of White's 1 Vol. i, pp. 32-33. Fomes vegetus is merely a synonym for F. applanatus. 2 J. H. White, loc. cit., p. 140. FIG. 47. — Fomes officinalis. A fruit-body with 15 annual layers of hymenial tubes, and therefore 15 years old ; length 1 foot, breadth and depth each 7'5 inches, weight 5 Ibs. Collected at Algonkin Park, Canada. Photographed by J. H. Faull. About | natural size. FOMES APPLANATUS 131 fruit-body must have been approximately 930 X 2,080 or about 2,000,000. Since from these 2,000,000 tubes 30,000,000,000 spores were liberated in 24 hours, the average rate of emission of spores from each tube must have been about 15,000 per day of 24 hours, or about ten per minute, or about one every six seconds. The Spore-discharge Period. — White has discovered that Fomes applanatus has a spore-discharge period which continues day and night uninterruptedly not for a few days only or even for a FIG. 48. — Fomes applanatus, a species with an annual spore-fall period of about six months' duration. Vertical section through a fruit-body three years old. The lowest and last-formed hymenial tube-layer is shedding clouds of spores. Specimen obtained at Munich, Germany. Reduced to \ natural size. few weeks but for six whole months ! No other fungus, so far as is yet known, sheds spores for anything like so long a time. White made continuous observations on spore-discharge from one and the same fruit-body for the full twelve months of each of the years 1915 and 1916. Spore-traps were set under the fruit-body from early spring until late autumn daily, and during the winter weekly. In addition to this unbroken series of observations during two years on one fruit-body, other observations upon spore-discharge were made upon other fruit-bodies during shorter periods each summer.1 White's observations show that, in the neighbourhood of Toronto, Canada, the spore-fall period of Fomes applanatus begins 1 J. H. White, loc. tit., p. 140. BllA RESEARCHES ON FUNGI early in May and continues without inter- ruption until the first heavy frost in late October or early November, when it suddenly ends. During this period of approximately six months, the liberation of the spores never ceases and is remarkably uniform (cf. Fig. 48). For the first few days after the commencement of spore- fall, the spore-deposit on the slides is rather scanty, but even then can be seen with the naked eye. Soon, however, the daily spore-deposit becomes almost constant ;• and it remains so for the rest of the spore-fall season. Any increase that may be observed is in the months of July and August.1 The daily slide collections for two seasons showed invariably their brown coating of spores. Changing slides at regular intervals throughout the day, and even throughout the night, gave no indication of any periodicity of discharge : the liberation of the spores continued steadily throughout the spore-fall period. On a still day White saw, just as I did for Polyporus squamosus? the spores FIG. 49. — Fomes applanatus. Diagram to show the mode of escape of the spores from the hymenial tubes. A, vertical section through four tubes of a fruit-body attached to a tree. The tubes have grown directly toward the centre of the earth so that the long axis of each is perpendicular. In each tube the spores are shot from the hymenium into the centre of the cavity and then fall down vertically in still air until they emerge from the pore. The trajectories of a few spores are shown by the arrows. B, a similar tube represented as tilted through an angle of 1°. As shown by the arrows at a and 6, all the spores discharged above c, i.e. in the upper five-sixths of the tube, hit and stick to the tube's side, and are lost. Only at and below c can the spores escape freely. C, a cross- section through a hymenial tube-layer. 1 J. H. White, loc. cit., p. 140. 2 Researches on Fungi, vol. i, 1909, p. 90. FOMES APPLANATUS 133 drifting away from a fruit-body in curling clouds like wreaths of very fine smoke as long as he cared to watch them ; and such clouds were observed as late as October 15. Night examinations of the spore-clouds were made with a lantern.1 The discharge of the spores did not appear to be affected by variations in light, humidity of the air, or change in temperature within very wide limits. However, frost caused an instant cessation of spore- discharge and, thereafter, no further fall of spores took place until a new set of hymenial tubes was organised in the spring.2 At the Spring Foray of the British Mycological Society held at Haslemere, May 14, 1921, on ascending a steep slope, a fallen log was observed from which projected a large fruit-body of Fomes applanatus. Upon looking at the fruit-body, the members saw spores emerging from its under side and floating away in the sun- light.3 At Kew Gardens, in the autumn of the same year, I found a large fruit-body of Fomes applanatus, about 1 foot from the ground, roofing over the space between two large buttress-roots of a Beech tree. On watching the exit of the hole under the fruit- body, with the aid of diffuse daylight only, I distinctly saw brown spore-clouds drifting outwards at intervals ; and I repeated this observation on August 29, September 9, and on several days in between those dates. It was then necessary for me to leave Kew. After my departure and at my request, Mr. S. Dickinson visited the tree on October 16 and a week later on October 23 ; and he kindly reported to me that on both occasions he clearly saw spore- clouds emerging from beneath the pileus. Thereafter in 1921 no further observations were made. In the following year, 1922, I observed spore-clouds coming away from the same fungus on June 17, August 1, and September 4. The evidence thus obtained goes to show that in England the spore-fall period of Fomes applanatus extends from early in May until late in October, i.e. for nearly six months. It is possible that, in England, owing to the milder climate, the spore-discharge period of this species may be even more prolonged than in Canada. Observations upon the Kew fruit-body just described convinced 1 J. H. White, loc. cit., p. 140. 2 Ibid., pp. 141, 167. 3 Trans. Brit. Myc. Soc., vol. vii, 1922, p. 221. 134 RESEARCHES ON FUNGI me that the numbers of spores coming away from the fruit-body each day was enormous. Near the fruit-body the bark of the tree and the leaves on the ground beneath had become coated with a brown layer of spores ; but direct inspection of the escaping spore - clouds indicated that this spore- deposit only repre- sented a very small fraction of the total number of spores actually produced. On the other side of the Beech a fruit-body was found roofing over a long narrow radial hole in the tree-trunk (cf. Fig. 50). From this hole the spores could not so easily escape owing to its form and the smallness of the opening. Hence FIG. 50. — Fomes applanatus. A fruit-body with about nine annual hymenial tube-layers, and there- fore about nine years old, roofing over a hole in the base of a tree. On the floor of such a hole the ares often accumulate as a thick brown layer. spor Photographed in Mulgrave Woods, Yorkshire," by theV had aCCUmU- A. E. Peck. About \ natural size. lated beneath the pileus. The brown spore-deposit there present was so thick that I had no difficulty in filling a wooden match-box with it. When full, the match-box doubtless contained many thousands of millions of spores. A Comparison of the Spore-fall Period cf Certain Hymeno- mycetes. — The length of the spore-fall period for a number of Hymenomycetes, together with indications as to the mechanical FOMES APPLANATUS 135 consistence of each kind of fruit-body, etc., is given in the following Table. Length of the Spore-fall Period in Hymenomycetes. Species. Spore-fall Period Mechanical Consistence. Attached to or Severed from Substratum. Observer. Coprinus curtus Small fruit-body 35 minutes Very watery Attached Buller V Large fruit -body 2 hrs. 25 min. Very watery ,, ,, » Coprinus sterquilinus 8-12 hours Watery ,, „ "1 Coprinus comatus . 2 days Less watery ,, ., "§ Psalliota campestris 7 days Firm ,, , © 0 © *n° ^0cP0^J n O r> C FIG. 94. — Panaeolus campanulatus. Analysis of the hymenium as seen in surface view on the second day of spore-discharge. A, an area equal approximately to 0-005 square mm. (0-06 x 0-09 mm.). This area contains: (1) past- generation basidia, a, which are recognisable by their four sterigmatic stumps and which are set out separately at B ; (2) present-generation basidia, b, each of which bears four black spores and which are set out separately at C ; (3) coming -generation basidia, c, heavily shaded with dots and set out separately at D ; (4) future-generation basidia, d, smaller than the coming-generation basidia and slightly shaded wfth dots, set out separately at E ; and (5) paraphyses, e, not shaded and partly hidden by the other elements, set out separately at F. Magnification, 580. in part be included in the count. This number is twice that of the present-generation basidia present on the same area. The inference PANAEOLUS CAMPANULATUS 277 is clear : the number of past generations of basidia, which the collapsed basidia really represent, is two. So far then, in classifying our basidia, we have accounted for two past generations and the present generation. We must now seek out the generations which are coming or are still to come. What we may term the coming generation of basidia, i.e. those basidia which will rapidly come to maturity and produce spores immediately after the present generation of basidia has become exhausted, can readily be made out. They are distinguished indi- vidually from other young basidia by the fact that, unlike these, they have attained their maximum diameter and have already become protuberant. Eight of them are present in the area under discussion and, in order to help the reader in finding them in Fig. 94, A, they have been set out by themselves in Fig. 94, D. The diameter of their bodies is equal to that of the bodies of the basidia of the present generation. The only protuberant elements in the hymenium are the basidia of the present and the coming generations. In our area, the basidia of the coming generation have already reached their maximum protuberancy, i.e. their bodies project above the general surface of the hymenium just as far as those of the basidia of the present generation. The basidia of the coming generation are also densely filled with protoplasm which, so to speak, is waiting to be put into the spores when these have been produced. On the other hand, the basidia of the present generation are poor in protoplasm owing to the fact that the transference of this sub- stance to the reproductive bodies has already been partially or completely accomplished. The basidia which are younger than those of the present generation, like the latter, are full of protoplasm, but their contents, owing to smaller volume, are much less con- spicuous to the eye. In some areas similar to the present one, it was found that, shortly before the spores of the present generation of basidia are to be discharged, the sterigmata of the basidia of the coming generation are in course of development. In the area taken for our analysis, however, the sterigmata were not observed. Probably the time had not yet come for their development. It may be said that, when an area of the hymenium has begun to shed spores, two generations of basidia are undergoing development 278 RESEARCHES ON FUNGI simultaneously, the present and the coming. The latter, however, is some eight or nine hours less advanced than the former. When a basidium of the present generation is developing its spores, a basidium of the coming generation is elongating so as to become protuberant, and filling with protoplasm. When the former is getting ready to shed its spores, the latter is developing its sterig- mata. Again, when the former is collapsing after shedding its spores, the latter is beginning to develop its spores. We must now turn our attention to what we may call the future generations, i.e. to generations of basidia which will be developed in succession after the coming generation has reached maturity. These later generations are represented by basidia which relatively are much smaller and less conspicuous than those of the coming generation. Eighteen of these elements are to be found in the area we are studying, and they are shown in Fig. 94, A, at d. They have also been set out by themselves at E. Their smaller size, non-protuberancy, and relatively less conspicuous contents enable one to distinguish them from the basidia of the coming generation ; but one cannot with certainty distinguish one future generation from another. The basidia of the future generations are distin- guished from those of the present generation by the absence of spores, and from those of past generations by the absence of the sterigmatic stumps, by their more symmetrical outline, and by their fine protoplasmic contents. The only elements with which the basidia of the future generations can be confused are the paraphyses. However, as a rule, these latter elements, which we shall treat of in greater detail directly, are relatively smaller and, owing to their large vacuoles, contain relatively little protoplasm. In Fig. 94 at A, the basidia of future generations (d) are all shaded with dots, while the paraphyses (e, shown by themselves at F) are left altogether unshaded. There can be little doubt that the eighteen basidia which are younger than those of the coming generation, represent several distinct generations. The number could not be less than three, if the same spacing as is seen for the basidia of the present generation is to be maintained. But, since we know from our observations with the horizontal microscope that the spacing of the basidia of future generations is not so compact as that of PANAEOLUS CAMPANULATUS 279 earlier generations, we have a right to conclude that the eighteen basidia represent more than three generations. The number is probably about five. The successive generations of basidia have been enumerated in their entirety. We must now turn our attention to those other elements of the hymenium which are called paraphyses. I have studied paraphyses in various Coprini, in Bolbitius, Psathyrella disseminata, Lepiota cepaestipes, Stropharia semiglobata, etc., and my conception of their nature may be stated as follows. Para- physes are hy menial cells sui generis, which (1) numerically con- stitute a large proportion — often 50 per cent, or more — of the total number of hymenial elements, (2) usually differ more or less in shape from the cystidia and basidia, (3) as they grow older and larger, become progressively poorer in protoplasmic contents, and (4) never produce either sterigmata or spores, so that they remain sterile from their first origin to their death. I regard paraphyses as destined from their first origin to remain sterile. I see no reason to believe that, after spore-discharge has begun, any paraphysis ever turns into a basidium or, vice versa, that any young basidium ever turns into a paraphysis. The question now arises whether or not paraphyses occur in the hymenium of Panaeolus campanulatus. If one cuts a transverse section through the hymenium of a gill that has been shedding spores only for a short time, say 24 hours, and if one studies the section with the microscope, it is not very easy to decide the question of the presence or absence of paraphyses. True it is that one can see here and there some small hymenial elements pushing up, as it were, between the bases of undoubted basidia and developing one or more vacuoles in their interior, but other cells which are similar in shape, not much larger, provided with but slightly more protoplasm, and which presumably are young basidia, occur along with them. One feels insecure when one pronounces that this small cell is a paraphysis but this other, differing from it so little in shape, size, and contents, is a very young basidium. The difficulties in distinguishing paraphyses from basidia are, however, not insurmountable, provided that they are approached in a suitable manner. In devising a new method for solving the problem of the presence 280 RESEARCHES ON FUNGI or absence of paraphyses, I made use of the discovery already recorded that exhausted basidia can always be distinguished from other elements by their concave ends and by the sterigmatic particles which remain on their tops after collapse has taken place. It was argued that, if one were to study the surface of a hymenium in face view, from the beginning to the end of spore-discharge, then not only would the basidia become more and more easily recognisable owing to more and more of them becoming exhausted, but the paraphyses with their plain rounded tops should become, by con- trast with the basidia, more and more conspicuous. It also seemed probable that the hymenium, after the spores had all been discharged, would present to the eye only two kinds of elements : (1) exhausted basidia to be distinguished by their sterigmatic particles, and (2) sterile paraphyses to be distinguished by their smooth rounded tops and impoverished protoplasmic contents. With a view to testing these suppositions, a fruit-body of Panaeolus campanulatus was grown on horse dung in the laboratory ; and from its pileus, day by day, from the beginning of spore-discharge to the end, there was dissected off a succession of gills for microscopic study. The hymenium on each" gill was examined in face view. The results of the investigation, which was continued for upwards of a week, fully justified the suppositions which had been made. A small piece (one two-hundredth part of a square mm.) of a completely exhausted hymenial layer is shown in Fig. 95 at A, In it there are two kinds of elements only — exhausted basidia and paraphyses. A waste spore, i.e. one which did not succeed in leaving the hymenium, is shown on the right. The paraphyses of Fig. 95, A, are set out by themselves in Fig. 95, B. A count (including elements sketched in part only) results in the finding that on the whole area the basidia are about equal in number to the paraphyses. Actually, the basidia are to the paraphyses in the proportion of 47 to 50. The tops of the basidia, which are by far their widest parts, pass down into an attenuated shaft, and they rest on the tops of the adjacent paraphyses. The latter are almost spherical elements and, since they are turgid, are doubtless still living. At the end of the spore-discharge period, therefore, the basidia are all dead and the paraphyses are all living. The paraphyses, as comparative PANAEOLUS CAMPANULATUS 281 examinations proved, gradually enlarge during the exhaustion of the hymenium and, by the end of the spore-discharge period, they have attained their maximum size. This is made very evident by a comparison of Fig. 94, F (p. 276), which shows the size of the paraphyses in a hymenium which has not long been shedding spores, and Fig. 95, B, which shows the size of the paraphyses in an exhausted hymenium. When the paraphyses have attained their A B FIG. 95. — Panaeolus campanulatus. Analysis of the ex- hausted hymenium from a fruit -body which has ceased to shed spores. A, the basidia (all belonging to past generations) can be recognised by their sterig- matic stumps : they slightly overlie the paraphyses. One waste spore lies on the hymenium. B, the para- physes of A sketched by themselves. Magnification, 580. full development, the layer of protoplasm upon their walls is very thin, and the major part of their contents consists of a large vacuole. In this condition, therefore, they are very transparent. Sometimes, in parts of an exhausted hymenium, aborted basidia are to be found. These can be distinguished from the paraphyses by their shape, which is more or less like that of a club, and by their proto- plasmic content. The latter, although containing a large vacuole in its centre, still forms a thick layer round the wall, and it has a peculiar dense, granular consistence totally different from that of the thin lining layer in the paraphyses. 282 RESEARCHES ON FUNGI What is the function of the paraphyses ? In some Hymeno- mycetes, like the Coprini, Lepiota cepaestipes, and Psaihyrella dis- seminata, in which the organisation of the hymenium (to be discussed later) differs from that of the Panaeolus Sub -type, the paraphyses are absolutely necessary to keep the spore-bearing basidia apart. In these species the paraphyses, which are of large size, are not merely in contact with one another but adhere together so as to form a continuous and firm tissue through the gaps of which the basidia protrude themselves at intervals. No doubt, under these conditions, the paraphyses, in addition to fulfilling their chief function of separating the basidia and therefore of preventing their mutual interference, also serve to support the basidia so as to preserve them in their upright positions. Further, they probably help to supply the basidia with water which is required to compen- sate for losses sustained in transpiration from the spores, etc. In Panaeolus campanulatus, however, the paraphyses have no very definite order and are not closely adherent (Fig. 94, F, p. 276 ; Fig. 95, B, p. 281), so that one often finds two basidia in contact with one another (Fig. 94, A ; Fig. 95, A). What, therefore, can be said for the function of the paraphyses in this species ? Owing to the fact that the basidia in Panaeolus campanulatus come to maturity in a succession of generations, paraphyses are not necessary as space-making elements : any two spore-bearing basidia can be separated by immature or exhausted basidia. I do not think, therefore, that the paraphyses here have the same chief function as they have in the Coprini. On the other hand, the paraphyses are probably useful in supporting the spore-bearing basidia in upright positions, especially toward the end of spore- discharge when most of the basidia are dead ; and, doubtless, they also conduct water into the basidia which, when protuberant and bearing four spores, must often transpire considerable quantities of it. If there were no paraphyses in the hymenium, many or most of the basidia would stand isolated on their narrow bases, and they would all expose a much larger superficial wall- area. Surely, under these conditions, the basidia would have greater difficulty in taking up positions perpendicular to the gill surface and also in obtain- ing the increased amount of water which would be required for PANAEOLUS CAMPANULATUS 283 transpiration. The paraphyses, in respect to their increase in cir- cumference in a direction parallel to the surface of the hymenium, bring an element of elasticity into the hymenium. For the basidia, the order of development and exact final thickness are fixed rather rigidly. The paraphyses, on the other hand, although at first all very small elements, come to have a relatively wide range of size toward the close of the hymenial activity. The difference in amount of variation in size between paraphyses and basidia will be realised by comparing in Fig. 95 (p. 281), the basidia in A with the paraphyses in B. Right up to the time of the coming to maturity of the first generation of basidia and often for some time after, the gills of Panaeolus campanulatus are expanding in superficial area. The paraphyses, during this expansion, themselves expand and thereby prevent the coming into existence of spaces between themselves, and between themselves and the basidia. Thus the hymenium is preserved as a continuous membrane and the basidia are kept in lateral contact with elements which live longer than they do and which, in respect to mechanical support and supply of moisture, can care for them during the whole of their existence. Our surface-view studies have taught us the essential facts concerning the arrangement of the elements of the hymenium in space and time. We are now in a position to interpret the basidia and paraphyses when seen from their sides. A cross-section through a gill that has been shedding spores for some time, say 24 hours, is represented in Fig. 96. This illustration may be considered to be an enlarged portion of the lower third of one of the long gills of Fig. 85 (p. 249). However, it does not show the exact position of the elements as they were actually observed in any one particular preparation, but rather the generalised result of the study of a great many preparations, i.e. it is semi-diagrammatic : various details drawn originally with the help of the camera lucida have been brought together and combined by the author so as to construct synthetically an ideal section which embodies as many facts of normal structure as possible. In Fig. 96, we can distinguish : the hymenium, hy, composed of a single layer of cells which are arranged perpendicularly to the plane of the gill-surface ; the subhymenium, sub, composed of two 284 RESEARCHES ON FUNGI or three layers of cells which are oval or rounded in shape and some- what irregularly disposed ; and the relatively broad trama, tr, the elements of which are more or less cylindrical in form, in general elongated in a direction perpendicular to the pileus-flesh and parallel to the hymenial layers, and often anastomosing with one another. On the left-hand side of the Figure, the section is represented as passing through : a white area of the hymenium, A ; a black area, B ; a brown area, C ; and another white area, D ; and on the right- hand side as passing through a black area, E ; a white area, F ; a brown area, G ; and another black area, H. These black, brown, and white areas correspond to similarly coloured areas which have already been described in surface view and which are shown in Figs. 89 and 90 (pp. 257 and 258). It was pointed out that a study of adjacent areas of the hymenium of a gill of Panaeolus campanulatus, seen in surface view, leads to the conclusion that the hymenium is so organised that its development takes place in a series of waves of small dimensions which move irregularly and simultaneously through it x ; and by means of camera- lucida drawings, reproduced in Figs. 89 and 90 (pp. 257 and 258), the direction of the waves on two small areas was clearly shown. Now in transverse sections through the hymenium the wave mode of hymenial development can also be made out. It has therefore been indicated iii Fig. 96 by means of an appropriate arrangement of the elements. A wave of development in the hymenium on the left-hand side of Fig. 96 is represented as passing from above downwards. The basidia opposite the Roman numerals, I, II, III . . . XII developed their spores in succession. The basidia I and II shed their spores an hour or two ago and have collapsed. The basidium III shed its spores about 15 minutes ago and is hi the act of collapsing : the sterigmata are being drawn into the concavity which is forming at the end of the basidium-body. The basidium IV is discharging its spores : two of them have already been shot away, and the lower one of the remaining two should be shot away at the end of about one second, for the drop of fluid excreted at its hilum has just attained its full normal size and is about 1 Vide supra, pp. 258-259. PANAEOLUS CAMPANULATUS 285 9 seconds old.1 The basidia V, VI, VII . . . XII will all discharge their spores in succession, V first and XII last. The basidia opposite the Arabic numerals 1, 2, 3 . . . 11 have developed spores in the area A and will develop spores in the remaining areas, in their numerical order. It will be noticed that the basidia 1, 2, and 3 bear spores of progressively smaller size. Basidium 4 possesses full-grown sterigmata but no spores. The sterigmata on basidia 4, 5, and 6 are progressively shorter. Basidium 7, although fully protuberant, has not yet developed even the bases of its sterigmata, and basidia 8, 9, 10, and 11 are in the same condition. They will all develop their sterigmata in succession following their numerical order. The series of basidia 1, 2, 3 ... 11 are developmentally about eight hours behind the adjacent series of basidia I, II, III . . . XII. Thus basidium 1 is about eight hours behind basidium I, basidium 5 about eight hours behind basidium V, basidium 10 about eight hours behind basidium XII, and so forth. Another wave of development in the hymenium is shown on the right-hand side of Fig. 96 as passing from below upwards. There is a series of basidia indicated by the Roman numerals I, II, III . . . XI, where I is the oldest, II slightly younger, III still younger, and so on up to XI. Similarly, in the series of basidia indicated by the Arabic numerals 1, 2, 3 . . . 12, basidium 1 is the oldest, basidium 2 slightly younger, basidium 3 still younger, and so on up to basidium 12. Here again, the series 1, 2, 3 . . . 12 is developmentally about eight hours behind the series I, II, III ... XI. The developmental waves shown in the hymenium on both sides of Fig. 96 correspond in their nature to the wave ac to 6 shown in the surface view represented in Fig. 90 (p. 258), for they are continuous in one direction. In the surface view represented in Fig. 89 (p. 257), a wave is shown which has originated at the periphery of the area and is spreading centripetally, i.e. toward the centre of the central black area. If one wanted to represent a cross-section through a wave area of this type, one could 1 The spores, normally, are shot away about 10 seconds after the water-drop begins to be excreted and immediately after it has attained full normal size. Vide Chap. I. 286 RESEARCHES ON FUNGI FIG. 96. — Panaeolus campanulatus. Semi-diagrammatic vertical section through part of a gill, 0-45 mm. long, showing the arrangements of the elements of the hymenium, subhymenium, and trama. On the left-hand side of the Figure, the section is represented as passing through : a white area of the hymenium, A ; a black area, B ; a brown area, C ; and another white area, D ; and, on the right-hand side, as passing through a black area, E ; a white area, F ; a brown area, G ; and another black area, H. In the white areas, basidia of the present generation are very young, their spores being between 0 and 2-25 hours old and still colourless. In the brown areas, the basidia of the present generation are older, being between 2 • 25 and 3 • 75 hours old ; their spores, which are brownish, are in the process of turning black. In the black areas, the basidia of the present generation are the most mature of all, their spores, which have already turned black, being between 3 • 75 and 7 • 25 hours old. The elements of the hymenium can everywhere be separated into the following five classes : ( 1 ) basidia of the past generations, a, which have discharged their spores and have collapsed ; their ends are concave but look flat in the side view here shown ; (2) basidia of the present generation, b, which are easily recognised because they bear spores ; (3) basidia of the coming generation, c, which are protuberant ; they may or may not have yet developed sterigmata (vide in B, C, and D, the basidia opposite the numbers 4, 5, 6, 7, 8, 9, 10, and 11) but have not yet developed spores ; (4) basidia of future generations, d ; these are non-protuberant and more or less pear-shaped ; they will bear spores later on ; and (5) the paraphyses, e, sterile elements which never bear sterigmata and spores ; at this stage of the development of the hymenium, they are distinguished from the basidia of future generations by their smaller size, their position between the basidia, their contents, and their tendency to become more or less spherical as they enlarge. A wave of development in the hymenium on the left of the Figure is repre- sented as passing from above downwards. The basidia opposite the numbers I, II, III . . . XII, developed their spores in succession. The basidia, I and II, have shed their spores an hour or two ago, and have collapsed. The basidium III has shed its spores about 15 minutes ago and is in the act of collapsing. The basidium IV is about to shed its last two spores ; one, which should be shot away the next second, has a drop of water of full size and about 9 seconds old attached to its hilum. The basidia V, VI, VII . . . XII will all shed their spores in succession. The basidia 1, 2, 3 . . . 1 1 will develop spores in succession in their numerical order. They are about 8 hours developmentally behind the adjacent basidia I, II, III ... XII. Thus 1 is eight hours behind I, 5 eight hours behind V, 10 eight hours behind XII, etc. Another wave of development is shown as passing upwards through the hymenium on the right-hand side of the Figure. Thus there is a set of basidia I, II, III ... XI where I is the oldest, II the next oldest, and so on down to XI. Similarly in the series 1, 2, 3 ... 12, 1 is the oldest, 2 the next oldest, and so on down to 12. Here again the series 1, 2, 3 ... 12 is developmentally about 8 hours behind the series I, II, III . . . XII. The exact size of every element may be read by using the scale at the base of the Figure. Magnification, 465. \ 12 XI 8 VIII VII H 288 RESEARCHES ON FUNGI accomplish it in Fig. 96 by removing the piece of hymenium on the left of the Figure from B VIII to the bottom of D, and by replacing it with the mirror-image of the remaining part of the hymenium which stretches from B 6 upwards. To observe such a mirror- image, it would only be necessary to place a mirror transversely across the Figure with its base along the line B 6 to G 5 and facing toward the top of the page. If a similar mirror-image were added to the upper half of the hymenium on the right-hand side of the Figure, we should have an arrangement showing a wave spreading centrifugally, i.e. outwards from the centre, like a wave started in the smooth water of a pond by a falling rain-drop. The elements of the hymenium in the section represented in Fig. 96, can be separated everywhere into the same five classes which we differentiated from one another in our surface-view studies : (1) basidia of past generations, a a, which have discharged their spores ; their ends are concave but look flat in the side view here shown ; they are no longer protuberant ; (2) basidia of the present generation, b b, which are easily recognised because they are protuberant and bear spores ; (3) basidia of the coming generation, c c, which are protuberant but do not bear spores ; they may or may not have developed sterigmata as yet (vide, in the areas B, C, and D the series of basidia opposite the Arabic numerals 4, 5, 6 ... 11 ; and on the opposite side of the Figure, in the areas, H. G, F, and E, the basidia of the series 1, 2, 3 . . . 11, as well as the basidium XI) ; (4) basidia of future generations, d d ; these are non-protuberant and more or less pear-shaped ; they will elongate and bear sterigmata and spores later on ; and (5) the paraphyses, e e, sterile elements which never bear sterigmata and spores ; at this stage of the development of the hymenium, they are distinguished from the basidia of future generations by their smaller size, their position between the lower parts of the basidia, their vacuolated contents, and their tendency to become more or less spherical as they enlarge. The most prominent elements of the hymenium in a cross- section are the basidia of the present generation, for they alone bear spores. It will be of interest here to state the relative ages of these basidia as they are represented in the various areas of PANAEOLUS CAMPANULATUS 289 Fig. 96. Such a statement is made possible through the appli- cation of the knowledge which was gained with the horizontal microscope in respect to spore-development. In the white areas A, D, and F, the basidia are relatively young, for their spores are only between 0-0 and 2-25 hours old. Some of these spores, A 2, A 3, F IX, and F X, which are only partially grown and there- fore not yet of full size, are even less than an hour old. The others which have attained full size, A 1, D X, D XI, D XII, F VI, F VII, and F VIII, are still colourless : they are only partially filled with the protoplasm and reserve foodstuffs which are slowly passing into them from the basidium-bodies through the narrow sterigmatic passages. In the brown areas, C and G, the basidia of the present generation, C IX, G IV, and G V, are relatively more advanced, for their spores are between 2-25 and 3-75 hours old. These spores have received most of their contents from the basidium-body, arid their walls have already become brown. The process of pigmentation, which will eventually turn the spores black, is in full swing, and the spore-walls are slowly deepening in tint. In the black areas, B, E, and H, the basidia of the present generation, B IV, B V, B VI, B VII, B VIII, E y, E z, HI, H II, and H III, are the most advanced of all ; for their spores are quite black, between 3 • 75 and 7 • 25 hours old, and preparing themselves for the moment of discharge. Two of the basidia, B IV and E y, are in the very act of discharging their spores and have therefore attained the maximum age. From E y the last spore, and from B IV the last spore but one, should be shot away within about one second, for drops of fluid of full normal size have just been excreted from their respective hila. The sporabolas of a number of spores, which have been shot from the hymenium and have fallen in still air, are shown magnified 20 times in Fig. 85 (p. 249). In Fig. 96 (p. 287), where the magnifi- cation is 465, it was found impracticable to represent any spore- trajectories owing to want of space on the page. A portion of Fig. 96 has therefore been reproduced in Fig. 97, and to it the sporabolic paths of several spores have been added. As in other Agaricineae, the four spores of each basidium of Panaeolus campanulatus are shot in succession outwards into an interlamellar 2QO RESEARCHES ON FUNGI space. Each spore is projected more or less horizontally from its sterigma with a considerable velocity ; but this horizontal velocity, in a small fraction of one second, is reduced to zero owing to the resistance offered to the passage of the spore by the air.1 Each spore, after having been shot more or less horizontally to a maximum distance of 0-1-0-12 mm., makes a sharp turn through a right angle and then, if the air is still, falls FIG. 97. — Panaeolus campanulatus. A vertical section taken transversely to the long axis of a gill showing the probable trajectories of spores discharged from the hymenium (a portion of the area B in Fig. 96, p. 287). The uppermost basidium has just discharged one of its spores which is being projected horizontally and is carrying a drop of water with it. From the hilum of another spore on the same basidium a water-drop has been excreted to about maximum size : this spore is just about to be discharged. Three trajectories of spores, discharged in still air, are indicated by the arrows ; their horizontal portions have been represented as being (from above down- wards) 0'12 mm., O'l mm., and O'll mm. in length. Magnification, 465. vertically downwards toward the earth (Fig. 97). The steady terminal velocity of fall for a spore of Panaeolus campanulatus is about 2-3 mm. per second.2 In Panaeolus campanulatus, just as in all other Hymeno- mycetes, the drop excreted at the spore-hilum just before spore- discharge is carried with the spore during its flight through the air.3 This is represented in Fig. 97 for the spore which has just left the uppermost present-generation basidium. We have seen that all the elements represented in the 1 Vol. i, 1909, p. 185. 2 Cf. p. 249, foot-note. Chap. I, pp. 14-16. PANAEOLUS CAMPANULATUS 291 hymenium of Fig. 96 belong to five classes. In order to demonstrate this point still more clearly, a small portion of the hymenium, namely, the black area B, has been represented in Fig. 98 on a larger scale and analysed in detail. This analysis of a piece of the hymenium in cross -section corresponds to the analysis given in Fig. 94 (p. 276) for a piece of the hymenium as seen in surface view. In Fig. 98, at A, the elements represented are exactly similar to those shown in Fig. 96 (p. 287) in the area B, except for the fact that the subhymenium, 5, has been toned down, so that by contrast the hymenium, h, stands out more clearly. At B, the four basidia of the past generations present in A are shown by themselves in their original relative positions. In a similar manner are represented : at C the basidia of the present generation, at I) the basidia of the coming generation, at E the basidia of future generations, and at F the sterile paraphyses. No single hymenial element shown at A has been omitted in this analysis, but every one has been placed in one or other of the five classes represented in B, C, D, E, and F. The letters A, B, C, D, E, and F in Fig. 98 correspond respectively to the letters A, B, C, D, E, and F in Fig. 94 (p. 276). Just as the area B in Fig. 96 has been analysed in Fig. 98, so might be analysed any of the other areas of Fig. 96. There is no hymenial element represented in Fig. 96 which does not belong to one of the five typical classes ; and an analysis of the hymenial elements of its areas B, C, F, and H, has been partially carried out by means of the letters a, b, c, d, and e which indicate past, present, coming, and future generations of basidia and paraphyses respectively. So far we have dealt with a cross-section of a hymenium taken when the process of spore-discharge has been going on for only 24 hours ; but the question may be asked : what is a cross-section of the hymenium like when several days have passed by and spore- discharge has ceased ? The answer is given in Fig. 99. Here the trama, t, and the subhymenium, s, have been toned down so that the hymenium, h, may be more easily distinguished. At w are two waste spores which were not successfully discharged and are therefore clinging to the hymenial surface. The hymenium is RESEARCHES ON FUNGI FIG. 98. — Panaeolus campanulattis. Analysis of the hymenium. A, cross-section through the hymenium h and the subhymenium « (same as area B in Fig. 96). The subhymenium has been softened so as to make the hymenium more distinct. The four past-generation basidia, a, are set out at B ; the five present-generation basidia, 6, are set out at C ; the three coming-generation basidia, c, are set out at D ; the twelve future-generation basidia, d, are set out at E ; and the twelve paraphyses, e, are set out at F. The present-generation basidium farthest to right in A and C, has just discharged two of its spores and is about to discharge a third spore. That the moment of discharge for this third spore has arrived, is indicated by the drop of water, now of full size, which has been excreted within the last ten seconds from the spore hilum. Magnification, 718. PANAEOLUS CAMPANULATUS 293 represented as being fully exhausted. Its elements are of two kinds only : exhausted basidia, 6, and more or less globular para- physes, p. The exhausted basidia can everywhere be distinguished by their shrunken bodies and flat ends. The paraphyses are every- where rounded or oval in shape and appear to be fully turgid. By comparison with Fig. 98, A (p. 292), which represents the hymenium in an early stage of spore-discharge, it will be seen that the paraphyses are now much larger than they were. It is evi- dent that, as more and more basidia shed their spores and collapse, FIG. 99. — Panaeolus campanulatus. Cross-section through the hymenium, h, after spore -discharge has been completed. There are only two kinds of elements present : collapsed basidia, 6, and swollen paraphyses, p. The tops of the basidia overlie the now rounded paraphyses. Two spores, w, which were not discharged, are shown lying on the hymenium. «, the subhymenium ; t, the trama. The subhymenium and trama have been softened so as to make the hymenium more distinct. Magnification, 718. the paraphyses gradually increase in size. This expansion on the part of the paraphyses makes up to a very considerable extent for the lateral shrinkage of the basidium-bodies. Owing to this compensation, the hymenium remains as a continuous membrane, even when it has become completely exhausted. Fig. 99, which shows the exhausted hymenium in cross-section, corresponds to Fig. 95, A (p. 281), which shows the exhausted hymenium in surface view. Up to the present we have studied the hymenium in two of its stages of development only : ( 1 ) during the time that it is actively discharging its spores, and (2) after it has ceased to shed spores and has become completely exhausted. In order to complete our analysis, it is necessary for us to consider the state of the 294 RESEARCHES ON FUNGI hymenium before it has produced any spores at all. A description of the very young hymenium has been deferred until now because the author's investigations began with the older states of the hymenium, and because one can only understand the nature and arrangement of the elements of the very young hymenium in the light of the knowledge gained by a study of the older hymenium which is producing and discharging spores. We shall consider the structure of the very young hymenium which has not yet produced any spores, in surface view first. A young gill was taken from a living fruit-body, placed flat on a slide, and covered with a cover -glass without the addition of any mounting fluid. The appearance of its hymenium is shown in Fig. 100 at A. One observes a large number of more or less prominent elements of various sizes projecting toward the eye. These elements are all basidia. The largest ones, indicated by the letter a, are the basidia of the first generation, i.e. those which will produce and liberate spores first. In the piece of hymenium under consideration, they had already attained their maximum diameter and their maximum protuberancy ; while, in other areas of the hymenium on the same gill, some of them had already developed sterigmata or were in the act of doing so. Several first-generation basidia can be seen scattered about over the area shown in Fig. 100, A. In this area can also be seen basidia of medium size, one of which is indicated by the letter 6. These are the basidia of the second generation, i.e. basidia which will produce and liberate spores, after the basidia of the first generation, already referred to, have shed their spores and collapsed. The other basidia on the area, indicated by the letter c, are much smaller. They are evidently the basidia of the third, fourth, and subsequent generations, which will develop and liberate spores in the middle and last stages of the hymenial activity. The paraphyses are deeply seated and more or less hidden between the bases of the basidia, and no attempt has been made to indicate their positions in Fig. 100, A. In order to observe the para- physes, a gill was mounted under a cover-glass in water, and the plane of focus of the microscope was lowered until the paraphyses were reached. In Fig. 100, B, the positions of some of the PANAEOLUS CAMPANULATUS 295 overlying basidia are represented by dotted circles, while the paraphyses which were in part seen through the basidia are shaded with dots. It will be observed that the paraphyses are considerably o. o c FIG. 100. — Panaeolus campanulatus. Analysis of a very young hymenium, seen in surface view, before any spores have been developed. A, (gill placed on slide under cover-glass, no water added) shows basidia only : those of the first generation, a, are the most prominent ; those of the second generation, b, less prominent ; and those of succeeding generations, c, least prominent. B, (gill mounted in water, plane of focus on the paraphyses) shows the paraphyses shaded with dots. The basidia, which are larger and overlie the paraphyses, have their positions indicated by broken circles. C, (gill mounted in water) a plan of all the elements in a small portion of the hymenium. The larger elements are the basidia which have been set out at D, and the smaller ones the paraphyses which have been set out at E. Magnification, 580. smaller than the basidia, and are ensconced between the basidia, particularly where three basidia adjoin one another. Another portion of the hymenium, which was observed when immersed in water, is shown in Fig. 100, C. Here, every element of the 296 RESEARCHES ON FUNGI hymenium has been shown in outline. The larger elements are the basidia and the smaller ones the paraphyses. An analysis of this area is given at D and E. The basidia of C have all been represented at D, and, similarly, the paraphyses of C have all been represented at E. If one compares the paraphyses in Fig. 100, E with those in Fig. 94, F (p. 276) and Fig. 95, B (p. 281), all of which have been drawn to the same scale, it is obvious that the paraphyses grow steadily in size during the gradual exhaustion of the hymenium. The paraphyses of an exhausted hymenium have diameters which are between two and three times as large as those of paraphyses in the hymenium just previously to the production of the first spores. A transverse section through the very young hymenium, which has not yet produced any spores but is about to produce them, is shown in Fig. 101 in the upper drawing at h, while the subjacent subhymenium is shown at s. The more prominent and larger elements of the hymenium, a, b, c, are all basidia ; while the smaller elements, p, ensconced between the lower parts of the basidia are all paraphyses. In order that the basidia and para- physes may be the more readily distinguished from one another, in Fig. 101 a second (lower) drawing has been added. This is a copy of the upper drawing except for the fact that the sub- hymenium has been toned down and the paraphyses have been left unshaded. Of the basidia : the four prominent ones, a a, one of which has already developed sterigmata, are basidia of the first generation ; the medium-sized ones, b b, are basidia of the second generation ; whilst the smaller basidia, c c, are the basidia of the third, fourth, and all subsequent generations. The basidia ab c in Fig. 101 should be compared with the corresponding basidia a b c in Fig. 100, A (p. 295). The paraphyses of Fig. 101 also correspond with the paraphyses of Fig. 100, B. We have now studied the Hymenium of Panaeolus campanulatus in all those stages of its development which are subsequent to the coming into existence of its elements. We have seen that all its elements, before, during, and subsequently to spore-discharge by the fruit-body as a whole, are either basidia or paraphyses ; PANAEOLUS CAMPANULATUS 297 and we have been able to divide the basidia into those of past, present, coming, and future generations. Surface-view and cross- sectional illustrations have been given which are in harmony with one another, and which clearly indicate the relations of the hymenial elements in space and time. For the first time, the hymenium of a Non-Coprinus Agaric has been analysed in detail. FIG. 101. — Panaeolus campanulatus. Above, a section through the very young hymenium, h, just prior to the development of spores by the first -generation basidia, and through the subhymenium, s. a a, basidia of the first genera- tion ; b 6, basidia of the1 second generation ; c c, basidia of subsequent generations ; p, paraphyses. Below, a similar section but here the sub- hymenium has been softened so as to make the hymenium more distinct, and the paraphyses have been left unshaded so that their size, number, and arrangement relatively to the basidia become obvious at a glance. Magnification, 718. The author feels that this marks a distinct stage in the advance of our knowledge of the wonderful way in which the fruit-bodies of the Hymenomycetes are organised for carrying out their one great function of producing and h' berating spores. Significance of the Development of the Basidia in Successive Generations on any One Hymenial Area, and of the Existence of Various Areas. — Now that we are acquainted with the arrange- ment of the elements in the hymenium in space and time, we are in a position to discuss the theoretical significance of the develop- ment of basidia in successive generations on any one hymenial 298 RESEARCHES ON FUNGI area and also the significance of the existence of the various areas to which the phenomenon of mottling is due. In regard to the question of successive generations, it is first to be remarked that on any one area, owing to the fact that the basidia come to maturity in succession, a great many more spores are safely ripened and discharged than would be the case if the basidia were all to develop simultaneously. The space above any B C FIG. 102. — Panaeolus campanulatus. A, surface view of a piece of exhausted hymenium. The basidia had produced about 159 spores as shown by the number of sterigmatic stumps ; a, exhausted basidia ; e, paraphyses ; s, a waste spore. B, the 159 spores of A set in the upright position close together ; the area is about 1 -5 times as large as A. C, a surface view of a piece of hymenium of the same size as A, seen in the middle of the spore- discharge period. The spores on A were distributed at any one time during the spore-discharge period as at C ; a, past-generation basidia ; b, present- generation basidia ; c, coming-generation basidia ; d, future-generation basidia ; e, paraphyses. Magnification, 387. given area for the protrusion of the basidia and the extension of their spreading sterigmata and spores is limited. If all the basidia on the area endeavoured to occupy it simultaneously, there would be nothing but confusion : the basidia would so jostle each other and interfere with each other's development that the spores would become matted together and their chance of successful ripening and discharge would be reduced almost to nothing. This can best be realised by reference to an actual area, one that was drawn with the camera lucida and is represented in Fig. 102, A. This area, as one may reckon by counting the number of sterigmatic stumps, produced 159 spores. Each spore in its upright position occupies a certain horizontal area. If all the 159 spores were to PANAEOLUS CAMPANULATUS 299 be placed side by side in contact, as shown in Fig. 102 at B, they would occupy an area which is roughly one and a half times the original area. The impossibility of all the basidia on any area successfully producing and liberating their spores at one and the same time thus becomes obvious. The actual distribution of the spores on the area A at any one time during the spore-discharge period is indicated by the camera-lucida drawing reproduced in Fig. 102 at C. If all the spores produced on A were to be developed simultaneously and were to have the spacing shown at C, the area A would need to be made about seven times as large as it actually is. It is clear from the above discussion that the production of successive generations of basidia enables a hymenium to produce and successfully discharge many more spores than would be possible if only one generation of spores were produced. Yet the further question remains : why should the succession on any small area take the form of successive generations of equal-aged or almost equal-aged basidia ? The answer may be obtained, perhaps, by comparing the basidia to soldiers. In order that a battalion shall pass across a narrow piece of ground, along a road, or through a pass, with the greatest speed and smoothness, it is organised into a column of fours. The soldiers in each four march shoulder to shoulder in step, and any one point on the line of march is passed by four after four at successive intervals. Efficiency in passing any given point in a given time by the whole force is only obtained by giving to each soldier suitable space-relations with his fellows in the same four and with those in the fours in front and behind. If a battalion becomes disorganised, so that each man loses his former space-relations with his fellows, it becomes reduced to a crowd of men who are apt to impede each other's progress by getting in each other's way. For the purpose of passing a given point on a road of limited width in a given time by a large number of men, the advantages of the column-of-fours arrangement over the go-as-you-please arrangement are sufficiently well known from experience. It seems to me probable, therefore, that the battalion method of bringing the basidia to maturity on any small area of the hymenium has come into existence in the course of evolution 300 RESEARCHES ON FUNGI because of its high efficiency — because it prevents jostling and saves time. The analogy, which has just been drawn between the basidia on a hymenial area coming to maturity in successive generations and the soldiers in a battalion passing a given point in the line of march, is only a rough one and fails in certain details. Thus, on any hymenial area, there is a wave of development (the details of which have already been discussed) of such a nature that the basidia of the present generation, proceeding from one side of the area to the other, unlike the leading four in a column-of-fours of a battalion, are progressively more and more advanced. Other defects in the analogy we need not discuss. We have seen that on each gill developmental waves pass over the hymenium in an irregular manner : they clash and meet one another in a way which can only be determined by observation ; and to these waves the phenomenon of mottling is due. What advantage, if any, is there in the existence of these waves ? With the wave-system of organisation we find that there are present on the gill a large number of hymenial areas in different stages of development : there are not only black, brown, and white areas, but even the areas of one kind, e.g. the black areas, do not keep developmental time, for some of them shed their spores sooner than others. The general want of synchrony between the areas is advantageous to this extent, that it enables the fruit-body as a whole to shed spores not intermittently, in discontinuous showers, but in a constant stream. Very many plants which produce a large number of organs of dissemination have arrangements which secure that these organs shall be gradually and not simultaneously set free. Thus in many Mycetozoa, Liverworts, and Puff-balls, elaters or capillitium threads are present from which the spore- dust only disentangles itself with difficulty with the passage of time ; and even in the capsules of certain Orchids elater-like hairs play a similar role in the dispersal of the seed-dust.1 In this connection one may also cite the capsules of Epilobium, Salix, Populus, etc. These, which contain many small seeds, open 1 E. Pfitzer, in Engler und Prantl, Pflanzenfamilien, Teil II, Abteilung 6, Orchidateae, 1889, p. 73. PANAEOLUS CAMPANULATUS 301 from the apex downwards and so allow their seeds to escape in succession. A gradual dispersion, as opposed to a simultaneous one, seems to result in a greater chance of wide distribution for the spores or seeds of any individual plant, owing to the fact that external conditions, such as the direction of the wind, rain-fall, presence of animals, etc., are constantly changing. Now let us suppose that the basidia of the first generation of a fruit-body of Panaeolus campanulatus were to begin the development of their sterigmata at exactly the same time. Then, for eight hours after- wards, no spores would be shed. Then, suddenly, millions of spores would be liberated simultaneously. Let us further suppose that the basidia of the second and of all the succeeding generations were to develop synchronously. Then spore-discharge would take place as follows : millions of spores of the first generation set free simultaneously ; then a pause of about eight hours ; then millions of spores of the second generation set free simultaneously ; then another long pause ; and so forth. The irregular wave-method of hymenial development leads to an avoidance of this inter- mittency in spore-discharge. In nature, a fruit-body of Panaeolus campanulatus liberates spores gradually from the first moment of spore-discharge until its close. So many spores are set free each minute for from seven to ten days. The never-ceasing stream of spores which pours out from beneath the pileus during the whole period of spore-discharge is directly traceable to the lack of synchrony in the development of the numerous hymenial areas. Upon a single gill of Panaeolus campanulatus there are between 500,000 and 1,000,000 basidia, the number varying according to the size of the pileus. An arrangement whereby all these basidia would begin and end their development exactly simultaneously for all of the series of generations into which they are divided would necessarily be complex. The protoplasm of the gill would require to act as a very perfect unit, the whole controlling each small part of the hymenium, if such perfect unison of development were to be attained. The physiological arrangements for the production of successive generations of basidia are perhaps more easy to carry out, i.e. require much less correlation of the protoplasmic mechanism involved, when numerous small areas of the hymenium can act 302 RESEARCHES ON FUNGI more or less independently of one another, as happens in Panaeolus campanulatus, than when all are required to act synchronously. An analogy may be drawn once more from the organisation of an army. It is no difficult matter for a battalion of soldiers to march in step and in proper order in ranks of four abreast, but it would be less easy, were the ranks composed of 16 men. Now imagine an army of 50,000 men all marching as massed formation across a piece of country together. The order of march might conceivably be 50 ranks of 1,000 men each, or it might be a parallel series of battalions say of 1,000 men each in which there are only 4 men in each rank in each battalion. Even if the country passed over were quite flat and presented no natural obstacles, with the first arrangement, i.e. 1,000 men to a rank, it would be extremely difficult for the ranks to be properly held. One part of a rank would tend to bulge forward in front of other parts, and the most exacting vigilance would be required by every soldier if the ranks were to be kept anything like straight. A much easier way of marching the army over the ground would be one commonly employed and already mentioned, namely, forming the army into battalions, arranging each battalion in a column-of-fours, and marching the battalions massed together more or less in parallel columns. Each battalion would have a certain amount of inde- pendence, and every rank could be properly held. This second method of organising the army is the one which is most nearly analogous to the arrangement for bringing the basidia to maturity in successive generations on a gill of Panaeolus campanulatus. Each area roughly corresponds to a single battalion acting more or less independently of its fellows. In concluding this section, I wish to call attention to the interesting parallel which exists between the undulatory develop- ment of the spore-producing hymenium of Panaeolus campanulatus and the undulatory development of the epithelium which produces spermatozoa in the testicular tubules of the Mammalia. When one of these tubules, e.g. of a cat, is examined in a series of microtome sections, waves of developmental activity can be traced along it.1 1 Cf. E. A. Schafer, The Essentials of Histology, ed. 10, London, 1916, pp. 393, 395, Figs. 549, 552. PANAEOLUS CAMPANULATUS 303 Here, in the tubule, all the spermatozoa appear to be quite ripe but are still held in place by the nurse cells (cells of Sertoli) in which they were nourished. A little farther along the tubule, the spermatozoa are being set free. Still farther along, it is evident that the spermatozoa have been recently liberated and that preparations are under way for the formation of their successors. Still farther along, a new crop of spermatozoa is partly developed. Still farther along, the spermatozoa are all ripe and thus resemble those which were first observed. Wave after wave may thus be traced in one and the same tubule ; and, doubtless, each part of every tubule produces a series of successive generations of spermatozoa. Thus the development of spermatozoa by the epithelium lining the tubules of the testis of a Mammal very much resembles in its time and space relations the development of basidiospores by the hymenium covering the flat and exposed gills of Panaeolus campanulatus. It seems, therefore, that the problem of efficiency in the production and liberation of vast numbers of minute organic particles has been solved in the same way by the higher animals and by a lowly toadstool. Just as a fruit-body of Panaeolus is organised so that, through- out its period of functional activity, it liberates a constant stream of millions of spores into the air, so the testis of a Mammal is organised so that, throughout its functional activity, it liberates a constant stream of millions of spermatozoa into its tubules. The spores are set free as soon as they are ripe, so that they may be scattered by the winds and thus find new substrata for the development of new individuals of the species. The spermatozoa, which are simply male gametes, are confined in a storage chamber and ever renewed, so that they may be always ready for employment in a sexual act. The Spores which are Wasted. — Whenever one examines a fruit-body of one of the Hymenomycetes at the end of the spore- discharge period, one finds that it has adherent to its gills a considerable number of spores which for some reason or other have failed to be properly liberated. These spores, since they contribute nothing to the dispersal of the species in nature, may be called wasted spores. 304 RESEARCHES ON FUNGI We are now acquainted with the fact that the production and liberation of spores is the outcome of a number of complicated developmental changes. Let us recapitulate the more obvious of these changes in so far as they concern the basidium. A basidium, on beginning to mature, grows in length and volume, and at the same time gradually fills with protoplasm. On attaining full size, its two nuclei unite. The fusion-nucleus then forms four daughter nuclei by two successive karyokinetic divisions. Then the four sterigmata are brought into existence, and afterwards the spores. The four nuclei then creep up, or are drawn through, the sterigmata into the spores, so that each spore comes to have one nucleus only. For some hours the basidium pours cytoplasm and food materials into the spores, so that in the end the spore-contents become very dense indeed. The metabolism taking place in a maturing spore involves, among other processes : the temporary production of glycogen, the thickening of the cell-wall, and the manufacture of a brown pigment ; and a mature spore excretes a tiny drop of water from its hilum just prior to its being forcibly shot away from its sterigma. Each spore/immediately after being discharged, provided that no hindrance arises from misplaced gills, creeping insects, violent winds, etc., escapes down an inter- lamellar space into the free air beneath the pileus and is then carried off by the wind. It would be astonishing, if, in view of all the complex changes which take place during the production and liberation of spores, every spore out of all the millions pro- duced were to succeed in escaping from a fruit-body. As a matter of fact, a considerable number of spores is always wasted. These victims of some accident, which has occurred either during develop- ment or discharge, adhere most tenaciously to the gills and only reach the earth when the fruit-body which has produced them collapses and undergoes putrefaction at the end of the spore-fall period. While the absolute number of spores which are wasted is frequently large, often amounting to several millions, the ratio of the wasted spores to the whole number produced may be small. It is well to bear this in mind : otherwise it may be thought that the wonderful arrangements for securing the successive production of mature basidia are to a large extent rendered nugatory by an PANAEOLUS CAMPANULATUS 305 inefficient mode of spore-discharge. We shall now study the wasted spores a little more closely. We shall first estimate their number, and then pass on to discuss a curious accident which may befall them at the moment when they should be liberated. In order to find out the number of wasted spores which may be present on a fruit-body of Panaeolus campanulatus, a culture of this species was made on horse dung in the laboratory under a large bell- jar. The first large fruit-body which came up was chosen for observation. During its spore-discharge period, which lasted about ten days, care was taken not to disturb it in any way. As soon as the spores had ceased to be liberated, the ex- hausted gills were examined under the microscope. A camera-lucida sketch of all the collapsed basidia and of all the wasted spores was made for an area of the hymenium which just filled the field of view (Fig. 103). The number of wasted spores was easily counted. The total number of spores which the whole area had produced was evidently equal to the number of sterigmatic stumps, and it was obtained by counting the exhausted basidia and multiplying the number found by 4. The total number of spores produced on the area was 1,832, of which 70 had been wasted. The wasted spores, therefore, formed only 3-8 per cent, of the whole number produced. In respect to the number of wasted spores, the area chosen for investigation was, so far as I could judge from inspection, of average character. Some areas, however, were richer in wasted spores and some poorer. In one of the poorer fields, which was equal in extent to the one already illustrated, only a single wasted spore could be observed, and in another only 15. From the results of these observations I think it would be safe to say that, for the fruit-body investigated, the wasted spores were approximately from 3 to 4 per cent, of the whole number produced. By tracing the outlines of the gills removed from one quarter of the pileus and measuring their superficial areas, both sides of each gill being taken into account, it was calculated that for the whole pileus : the 60 long gills had a superficial area of about 60 square cm., the 60 intermediate gills had an area of about 24 square cm., the 120 short gills had an area of about 4 square cm., VOL. II. X 306 RESEARCHES ON FUNGI and consequently the area of the hymenial layer of the whole fruit-body was approximately about 88 square cm. 0-0771. TO. 0~1 rum. 0-2.771.772. 03 FIG. 103. — Panaeolus campanulatus. An area of the hymenium equal to one- seventh of 1 sq. mm., drawn after spore-discharge had ceased, showing the exhausted basidia and the wasted spores. The paraphyses are omitted. Most of the spores left lying on the hymenium were black. Four sets of spores, a, b, c, and d — each set of four the product of a single basidium — were brown, and one set, e, colourless. The four black spores at / are abnormally small, and the single spores at g and h abnormally large. The number of sterig- matic stumps on the exhausted basidia indicate that the area produced 1,832 spores. The wasted spores number 70, i.e. 3-8 per cent, of those produced. Magnification, 293. On 0-01 square mm. of an exhausted hymenial area the number of basidia was found to be 61. It was therefore calculated that PAN AEOLUS CAMPANULATUS 307 the number of basidia on a square mm. was about 6,100 ; on a square cm. 610,000 ; and on 88 square cm., i.e. on all the gills taken together, 53,680,000. From this, reckoning four spores to each basidium, it was calculated that the total number of spores which the fruit-body had produced was approximately 215,000,000. If we take the number of wasted spores as forming 3-8 per cent, of this number, the absolute number of wasted spores on the whole fruit-body was about 8,000,000. Subtracting this from the whole number of spores produced, we find that 207,000,000 spores were liberated. We may conclude, therefore, that even when the wasted spores left on the gills amount to millions they may yet be very few compared with the vast number which have been set free. We may now enquire into the causes which may prevent full- grown spores from being properly liberated. If a gill is displaced so that its median plane is no longer vertical but inclined at an angle of several degrees, then a spore which has been shot out at one point on the upper side of the gill will soon strike the gill again at a lower point. After falling on a gill in this way, a spore never gets free again : it adheres tightly to the hymenium until the fruit-body decays. An illustration showing the effect of inclining gills shaped like those of Panaeolus campanulatus has been given in Volume I, Fig. 12, p. 40. When a large number of spores fall on to the upper side of the gill of a displaced fruit- body, the gill on its upper side becomes, in the words of the field-botanist, " powdered with the spores." In the fruit-body of Panaeolus campanulatus which was investigated the median planes of the gills were properly adjusted in space, so that this powdering could not occur. On almost all large fruit-bodies of Hymenomycetes growing in woods in the warm days of autumn one may find Fungus Gnats, Mites, or Springtails. These crawl over the gills. Doubtless they displace many of the spores, for they are bound to touch these in their peregrinations. However, the extent to which crawling insects are a cause of the accumulation of wasted spores on gills, etc., in nature, requires further investigation. So far I have made no special observations in connection with this matter. When the wind is blowing strongly in one direction, the spores, 308 RESEARCHES ON FUNGI when about to pass out from between the interlamellar spaces into the open air at the base of the pileus, may be blown on to the edges of the gills and stick there. In this way a certain number of spores is often wasted. I have noticed this phenomenon in connection with specimens not only of Panaeolus campanulatus, but also of Anellaria separata and a number of other species growing FIG. 104. — Panaeolus campanulatus. Excessive excretion of water-drops and the non-discharge of spores. A, four ripe spores of a basidium seen from above ; B, the same excreting drops of water from their hila ; C, the drops have attained full normal size (spores are usually shot away at this stage) ; D, the drops have become supernormal and have stepped up on to the sides of the spores ; E, the drops are still increasing in size and now almost touch ; F, the drops have fused and the spores have been dragged together ; G, excretion of water has continued and the spores in consequence have become separated to the normal extent (c/. A) ; H, owing to a further excretion of water the spores have been pushed more than the usual distance apart ; I, excretion has ceased, the water has evaporated, and the spores have been dragged together so that they adhere to one another ; J, the basidium has collapsed and has dragged its four spores to the surface of the hymenium to which they now adhere and from which they cannot escape. Magnifica- tion, 660. in the open. The edges of the gills of Panaeoli sometimes become quite black with the spores. Probably the chief cause of the accumulation of wasted spores on gills is a failure in the discharging mechanism. The nature of this mechanism has so far defied exact analysis.1 However, just before a spore is to be discharged, as was explained in Chapter I, a drop of water is exuded at the neck of the sterigma (Fig. 104, A, B, and C). When this drop has reached a certain definite size (C), the spore is shot away from the sterigma and bears the drop 1 For a discussion of this mechanism vide Chap. I, pp. 22-26. PANAEOLUS CAMPANULATUS 309 with it. Now, as I have observed when watching the discharge of spores with the horizontal microscope in the manner already described,1 it sometimes happens that the spores on a basidium are not shot off when the drops below them have attained their normal size, but the drops increase in size (D and E) until they finally meet and fuse together (F). The four spores, without being detached from their sterigmata, are drawn nearer together owing to the pull exercised upon them by surface tension (F). The drops may go on enlarging so that the spores are pushed further apart than they are normally (G and H). The excretion of water by the sterigmata then ceases and the drop gradually evaporates. As the drop diminishes in size, the four spores are drawn together so that they come in contact with one another (I). A few minutes after the spores have met, the basidium to which the spores belong contracts in length, its end becomes concave, and its sterigmata are drawn into the concavity. It thus comes about that the four spores are dragged down to the general level of the hymenium, where they remain until the decay of the pileus sets in (J). The loss of spores owing to over-enlargement of the water-drops in the manner described was observed a great many times on the gills of Panaeolus campanulatus when they were being viewed with the horizontal microscope, but I have also observed the phenomenon on the gills of several other Hymenomycetes. Sometimes, after two or three of the spores of a basidium have been successfully discharged, the remaining spores stay on their sterigmata until the basidium collapses. At the necks of the sterigmata belonging to the spores which are to be wasted no drops of water appear. In these cases the failure of the spores to leave their sterigmata may be connected in the first place with the failure in the excretory activity of the sterigmata. I have now watched the discharge of many hundreds of spores of different Hymenomycetes with a sufficiently high power of the microscope, but never once, since I began to pay attention to the excretory phenomenon, have I ever seen a spore shot away without a water- drop being formed, a few seconds before discharge, at the hilum of the spore where this is attached to the neck of the subjacent 1 Pp. 260-263, also Figs. 91 and 92. 310 RESEARCHES ON FUNGI sterigma. My observations therefore seem to warrant the conclu- sion that the excretion of a water-drop is in some way absolutely necessary for the discharge of the spores in Hymenomycetes. Non-excretion and excessive excretion of water at the sterigmatic neck appear to be equally fatal to the spores concerned : both lead to the turning of normal-looking spores into wasters. In Panaeolus campanulatus I believe that the formation of many of the wasters is connected with the non-excretion of the water-drop. Some of the waste spores, as judged by their size or colour or both, are evidently very immature (Fig. 103, a, b, c, d, e, p. 306). For some reason or other their development proceeded to a certain point only and was then inhibited. One could hardly expect such unripe spores to be shot away, for probably the development of the mechanism of discharge is connected with the final stages of ripening of the spores. Among the waste spores one often finds some which are unusually large or unusually small or misshapen (Fig. 103, /, g, h). Clearly these are abnormal spores. The forces which cause a spore to become monstrous may also interfere with the discharging mechanism. Sometimes one finds a little heap of wasted spores. In such a case it would seem that several adjacent basidia have failed in spore-discharge, and have each contributed spores to the heap. The spores of two basidia belonging to two successive generations may become stuck together so as to form a little group of eight. Smaller or larger groups of wasted spores may be formed by slight variations of this process. Significance of the Protuberancy of Mature Basidia.— It has been pointed out that a basidium, when beginning to get ready to produce spores, grows in length until it protrudes a certain distance above the general level of the hymenial surface. The amount of the protuberancy thus attained will be at once realised by reference to Fig. 96 (p. 287). After the basidium-body has reached its full length, then, and then only, are the sterigmata developed ; and only when the sterigmata have been pushed out to their full extent do the spores begin developing at their ends. The phenomenon of protuberancy, as will be shown in detail PANAEOLUS CAMPANULATUS 311 in Volume III, is most marked in the various species of Coprinus, but it is also very strikingly displayed in Lepiota cepaestipes and Psaihyrella disseminata. In the fruit-bodies of all these species, the basidia can be separated into different classes according to their length ; and, in a single species, they are dimorphic, trimorphic, or quadrimorphic. Subsequently it will be shown that the differ- ential protuberancy, where polymorphism of the basidia occurs, is correlated with the crowding of the basidia and is necessary to prevent the spores of adjacent basidia from touching one another and sticking together. For the present, therefore, let us take it for granted that basidial protuberancy, where most strongly developed, is evidently related to the mechanical requirements of the spores, i.e. to the provision of a free space for spore- development and spore-discharge. In the fruit-bodies of the Panaeolus Sub-type the basidia are monomorphic, i.e. when mature they are all about equally pro- tuberant beyond the general level of the hymenium. The question arises : why should these basidia be protuberant at all ? I shall attempt to answer this question in connection with Panaeolus campanulatus. First, it is to be remarked that the protuberancy of any basidium in the hymenium is strictly temporary : it is acquired shortly before the development of the spores, and lost almost immediately after the spores have been discharged. Evidently, basidial protuberancy is connected with the develop- ment and discharge of the spores. It seems to me that the lengthening of the maturing basidia, which results in their becoming protuberant, is of considerable advantage to the functional activity of the fruit-body in the following respect : it causes the spores, which develop on the basidia only after these have become pro- tuberant, to take their origin, grow to maturity, and be discharged above the general level of the hymenium in spaces which are mechanically free from neighbouring hymenial elements. We have seen that, owing to various imperfections, a certain number of spores, instead of escaping freely from a fruit-body, become wasted in that they stick tenaciously to the hymenium. Now there is always some danger that the waste spores of one genera- tion may become obstacles to the development, and particularly 3i2 RESEARCHES ON FUNGI to the discharge, of the spores of subsequent generations : but, owing to the fact that maturing basidia become protuberant and develop their sterigmata before they develop their spores, this danger is reduced to a minimum. With the horizontal micro- scope, when watching the coming into existence and disappearance of successive generations of basidia, I several times saw that a maturing basidium, during its development, pushed aside one or more of the waste spores which had overlain it or had been in contact with it. The result was that the sterigmata were reared above the hymenium in a free space, and that the new spores were developed free from all obstacles. The practical advantage of protuberancy thus became a matter of observation. Even when a hymenium is thickly strewn with wasted spores, a considerable number of basidia succeed in pushing up between the obstacles in their way and in raising the tips of their sterigmata into freedom in preparation for spore-development. Significance of the Collapse of Exhausted Basidia.— It has been pointed out that, some fifteen to thirty minutes after a basidium has shed its four spores, the basidium-body contracts in length so that it ceases to be protuberant (cf. Fig. 93, A, p. 272, and Fig. 98, A and B, p. 292), and that at the same time the sterigmata become drawn into the concavity which is formed at the basidium-body end. It might be argued that the rapid collapse of a basidium which has shed its spores is simply due to the fact that the basidium is exhausted. In support of this one might urge that the protoplasm in a basidium which has shed its spores is devoid of nuclei and extremely small in amount, and also that, since such a basidium has no further function to perform, its death might well be expected to take place just when it actually does. However, while all this may be true, yet it is conceivable that a basidium should remain living long after it has discharged its spores. Masses of cytoplasm separated from all nuclear influence, as we know from actual experiments with Amoebae and the cells of the Higher Plants, do not necessarily die immediately after their isolation. If it were for the advantage of the hymenium that basidia which have shed their spores should remain protuberant and turgid, it seems to me very likely from PANAEOLUS CAMPANULATUS 313 analogy that we should find this arrangement in existence. How- ever, the rapid collapse of basidia which have shed their spores, whatever its physiological cause, fits in very well with the general organisation of the hymenium. It would be distinctly disadvan- tageous for exhausted basidia to remain protuberant, owing to the fact that their sterigmata, and any undischarged spores which they might carry, would be liable to be in the way of the new spores in course of development or discharge. By their rapid collapse after shedding their spores, basidia get rid of their sterig- mata entirely, and they also drag down any spores, which they have not succeeded in discharging, to the general level of the hymenium. After having been brought to this low level, these wasted spores are mechanically harmless ; for, as we have seen, maturing basidia, when elongating, push them aside without difficulty. The rapid collapse of exhausted basidia is therefore significant from the point of view of the organisation of the hymenium as a whole, in that it leads to the rapid removal of useless sterigmata and wasted, spores from positions where they might form mechanical hindrances to the development and discharge of the spores of later basidial generations. The Relative Position of the Basidia and the Spores of One Generation. — The distribution of the basidia belonging to a single generation is shown for Panaeolus campanulatus in Figs. 89 (p. 257) and 90 (p. 258). It follows no precise mathematical rule. However, one can see from inspection that the basidia are separated from one another by such distances that their spores cannot come into contact with one another : mutual interference is rendered impossible. At the same time the basidia are sufficiently evenly spread over the hymenium, and sufficiently near to one another, to prevent there being any great waste of space. The distances separating the basidia are usually confined within fairly definite and circumscribed limits. If one considers the area shown in Fig. 89 (p. 257), one sees that there are not many places where one could insert additional spore-bearing basidia without reducing to a dangerous minimum the margin of safety for securing the isolation of the spores of each basidium. We may conclude that, for Panaeolus campanulatus, the distribution of the basidia of any 314 RESEARCHES ON FUNGI one generation over the hymenium is economical, in that the basidia are crowded as closely as is consistent with the conservation of that mechanical freedom which is required for the proper development and discharge of the spores. Exceptions to the rule just given certainly occur. One some- times finds that, here and there on the hymenium, a few of the basidia are more closely crowded than they are in the areas repre- sented in Figs. 89 and 90. Occasionally, two spores, one belonging to one basidium and the other to another, are so close to one another that they almost touch (Fig. 112, F, p. 323). In some rare instances even, actual contact may be observed (Fig. 112, D}. Sometimes also it happens that here and there the basidia are dis- tinctly more loosely scattered than is necessary for the safety of the spores. That such imperfections in distribution FIG. 105. — Psalliota cam- should sometimes make themselves mani- pestris (cultivated), a, f . . a normal disterigmatic fest 1S not to be wondered at, when one basidium; b , a young reflects that a single fruit-body of Panaeolus monosterigmatic basi- dium; c, a mature mono- campanulatus may in the course of its sterigmatic basidium; d -. -, -~ ,.,-,„ „„„ development produce some 50,000,000 is a seen from abovi e is c seen from above. The spore of c is about twice the volume of one of the spores of a. Magnification, 1,060. basidia. The Position of the Sterigmata in the Hymenomycetes generally and in Panaeolus campanulatus. — First let us consider the position of the sterigmata on a single basidium in the Hymeno- mycetes generally. In the vast majority of species the number of sterigmata on the end of each basidium is normally four. How- ever, there are some species in which the basidium has only one sterigma, others in which it has two sterigmata, others in which it has three, others in which it has six, and yet others in which it has eight. Further, in a species in which the number of sterig- mata is normally four, occasionally one may find basidia with three or five sterigmata ; and, where the number is normally two, monosterigmatic basidia are usually frequently present. Probably PANAEOLUS CAMPANULATUS 315 in all species occasional variations from the normal number of sterigmata occur, and in this respect basidia may be likened to certain compound leaves and flowers of the Phanerogamia, where analogous variations in the number of the elements are not infre- quent. We will now discuss the arrangement of the sterigmata in the different types of basidia, beginning with the simplest. Monosterigmatic basidia, according to Patouillard, occur normally in Pistillaria macu- laecola Fckl. and also in the Gastromycete Hydnangium monosporum Boud. et Pat., whilst in Pistillaria fulgida Fr., although the basidia are usually monosterigmatic, occasionally they are disterig- matic.1 I myself so far have not examined these species. However, I have observed monosterigmatic basidia in the cultivated Mushroom, Psalliota campestris2 (Fig. 105), and in Coprinus bisporus Lange (Figs. 106 and 107). In both these species the basidia are normally disterig- matic, but monosterigmatic basidia occur as by no means infrequent exceptions. Where a basidium is monosterigmatic, the sterigma always occupies a central position at the top of the basidium (Fig. 105, b and c). The spore, however, is not sym- metrically set on the end of the sterigma, but is attached unilaterally in the manner which is characteristic of the spores 1 N. Patouillard, Tabulae analyticae fungorum, eer. 1, 1883-1886, nos. 47, 50, 692. 2 Vide infra. Monosterigmatic basidia are shown for the Mushroom in Figs. 146 and 147, Chap. XIII. FIG. 106. — Coprinus bisporus. Two fruit- bodies growing on a stick sent from Kew. The disc remains prominent until exhaustion of the pileus. Photo- graphed at 10 A.M. at Winnipeg. Natural size. 3i6 RESEARCHES ON FUNGI of tetrasterigmatic basidia (Fig. 105, 6, c, e; vide also basidia nos. 48 and 57 in Fig. 147, Chap. XIII). Disterigmatic basidia have been observed by me as normal in •••'.X** •* .* .* ** > « v A # Fio. 107. — Coprinus bisporus. Photomicrograph of a portion of the hymenium in surface view. The spores of the long basidia only are in focus, but those of the short basidia can be faintly seen out of focus. All the basidia, except three, bore two spores. The three exceptions (one at the right top corner, one below the left top corner, and one in the middle of the lower edge) each bore one spore of large size. Magnification, 480. a form of Mycena galericulafa, in the cultivated forms of Psalliota campestris (Fig. 105), in Coprinus bisporus (Figs. 106 and 107), Galera hypnorum, Hygrophorus conicus, and Calocera cornea. They occur too, as the literature of mycology shows, in a number of other species, e.g. Amanita bisporigera Atkinson,1 Pistillaria 1 C. E. Lewis, "The Development of the Spores in Amanita bisporigera," Botanical Gazette, 1906, pp. 348-352. PANAEOLUS CAMPANULATUS 317 micans Fr.,1 P. sagittaeformis Pat.,2 P. Helenae Pat.,3 Clavaria cinerea Fr.,4 Hygrophorus agathosmus,5 Pholiota togularis,6 Tubaria conspersa,1 and Corticium commixtum v. H. et L.8 When the whole number of Hymenomycetes is considered, the species with disterigmatic basidia are relatively few. However, these exceptional species are scattered through a considerable number of genera from the highest to the lowest. Possibly, when a sufficient number of critical studies has been made, it will be found that almost every large genus of the Hymenomycetes has one or more species which normally have disterigmatic basidia. In Amanita bisporigera, Coprinus bisporus, and Psalliota campestris, there can be but little doubt that the two-spored condition has been obtained by reduction from the four-spored, and possibly this is true in all disterigmatic species. The suppression of two sterigmata out of four may be likened to the suppression of spines in the Prickly Opuntia and the suppression of colours in certain flowers. Evolution seems to have taken place by means of the loss of a character. In disterigmatic basidia the two sterigmata are always set opposite to one another in a symmetrical manner, one on each side of the top of the basidium-body. At their origin, they seem to repel one another and thus come to be situated as far apart as possible (Fig. 105; also Figs. 146 and 147, pp. 416, 429). Tristerigmatic basidia, so far as I know, are the rule for one species only. I have discovered that in Coprinus narcoticus (Fig. 108), the normal number of sterigmata for each basidium is three only. Mixed with the tristerigmatic basidia, however, are a certain number of tetrasterigmatic. The fruit-bodies used for my observations came up spontaneously in 1912 and again in 1922 in laboratory cultures of horse dung collected from the streets of Winnipeg. The species was readily recognisable from the first by the remarkably pungent odour of the pileus during autodigestion. A photograph of the hymenium seen from above i. 2, 3, 4 N Patouillard, Tabulae analyticae fungorum, s6r. 1, 1883-1886. 5, 6, 7 y Fay0(i, "Prodrome d'une histoire naturelle des Agaricines," Ann. sci. nat., 1 s6r., T. IX., 1889, p. 262. 8 F. von Hohnel und V. Litschauer, " Beitrage zur Kenntnis der Corticieen," Sitzungsb. der K. Akad. der Wiss. in Wien, Bd. CXVI, 1907, p. 83. RESEARCHES ON FUNGI is shown in Fig. 109. In many species in which the basidia are normally tetrasterigmatic one may sometimes find a few basidia scattered at rare intervals which are tristerigmatic. Such I have observed for example in several species of Coprinus, e.g. C. niveus (Fig. 110, b), in Panaeolus campanulatus, and in the field forms of Psalliota campestris. In tristerigmatic basidia the three sterigmata are always placed equidistantly at the top of the basidium. Their positions are like those of the three corners of an equilateral triangle. Tetrasterigmatic basidia are the rule in most species of Hymenomycetes. This arrangement is so frequent that there is no need to quote the names of species. Wherever it occurs, the four sterigmata (except in certain Tremellineae) are placed equidistantly on the top of the basidium. Their relative positions are like those of the four corners of a square. Pentasterigmatic basidia, according to Patouillard,1 occur in Cantharellus Friesii Q . The illustration given by this author shows a single basidium with five sterigmata and five spores. In Anettaria separata, Coprinus niveus, and some other species of Agaricineae, where the basidia are normally tetrasterigmatic, I have occasionally met with a basidium with five sterigmata (Fig. 110, c). In pentasterigmatic basidia the sterigmata have the same relative positions as the five corners of a regular pentagon. 1 N. Patouillard, loc. cit., p. 148. FIG. 108. — Coprinus narcoticus — a Coprinus with a strong and disagreeable odour. Fruit-bodies coming up in a pure culture on horse dung. Species obtained at Winnipeg. Photographed at 6.30 A.M. The opening of the pileus takes place during the night. Natural size. PANAEOLUS CAMPANULATUS 319 Abnormalities in the basidia of Coprinus niveus include not only tristerigmatic and pentasterigmatic basidia, but also spores which exhibit equal or unequal bifurcation to a greater or less degree. Some of these curious spores are illustrated in Fig. 110. ' Hexasterigmatic basidia occur in Sisto- trema confluens Fr.,1 Cantharellus cibarius, and some other species of Cantharellus.2 Basidia with six sterigmata are also not infrequent in Exobasidium vaccinii 3 and some species of Corticium.4 The bases of the sterigmata for each basidium have the same relative posi- tions as the corners of a regular hexagon. Heptasterigmatic basidia, so far as I FlG -109-— c°Prinus ™rcoticus. Photograph of the hymenium, to show that the basidia are tri- sporous. The basidia, as in most Coprini, are dimorphic. The spores of most of the long ones only are in focus, and the spores of the short ones out of focus. There are three spores only above each basidium. In the middle of the right-hand side the spores were removed from their sterigmata during the making of the preparation. Magnification, 440. know, are not the rule in any of the species of Hymenomycetes. However, they occur exceptionally in some Thelephoraceae where the normal number of sterigmata is eight. I have not seen any such basidia myself but, doubtless, where they occur, the relative positions of the sterigmata are like those of the corners of a regular heptagon. 1 N. PatouiUard, loc. cit., p. 111. 2 Cf. V. Fayod, loc. cit., p. 262. 3 N. Patouillard, Essai taxonomique sur les families et les genres des Hymeno- mycetes, Lons-le-Saunier, 1900, p. 35. 4 F. von Hohnel und V. Litschauer, loc. cit., p. 84. 320 RESEARCHES ON FUNGI Octosterigmatic basidia occur in some of the Thelephoraceae, e.g. Corticium coronatum (Schrot.) v. H. et L. Here, according to von Hohnel and Litschauer,1 almost always, eight sterigmata are present on each basidium. I myself have seen octosterigmatic basidia in one of the species of Corticium. ^n octosterigmatic basidia the sterigmata on the top of the basidium have the same relative positions as the corners of a regular octagon. So far as I am aware, there are no species of Hymenomycetes in which the normal number of sterigmata on each basidium is nine, ten, or any regular number greater than eight. From the foregoing, we have seen that, at their origin on the top of a single basidium, the sterigmata appear to repel one another with equal strength, so that in the end the spaces separating adjacent sterigmata are all equal. We may therefore state the following law with regard to the position of sterigmata on any one basidium : When only one sterigma is present, it occupies a median position on the top of the basidium-body ; when more than one sterigma are present, the sterigmata arise in such positions on the top of the basidium-body that they are as remote from one another as possible. From the observations of cytologists 2 it appears that the nuclei derived from the fusion-nucleus in the basidium attach themselves by means of their centrosomes to those parts of the basidial wall which are to bulge upwards and develop into sterigmata. We 1 F. von Hohnel und V. Litschauer, loc. cit., p. 94. 2 E.g. M. Levine, " Studies in the Cytology of the Hymenomycetes, especially the Boleti," Bull. Torrey Bot. Club, vol. xl, 1913, pp. 172-173. For reference to the papers of Rene Maire and R. E. Fries, vide infra, Chap. XIII. Fia. 110. — Coprinus niveus. Teratology of basidia and spores, a, a normal tetra- sterigmatic basidium ; 6, a tristerigmatic basidium ; c, a pentasterigmatic basi- dium ; d, a normal spore seen from in front and sideways ; e, f, and g, spores showing progres- sively deep bifurcation ; h, a small spore showing bifurcation ; i, j, and k, spores showing progres- sively asymmetrical bifur- cation ; I, an irregularly shaped spore ; m, two very small black spores, one showing a bifurcation. Magnification : a, b, and c, 293 ; d-m, 707. PANAEOLUS CAMPANULATUS 321 °0 OO 00 ©O oc C may therefore conclude that the positions of the sterigmata on the end of any basidium are decided by the positions of the nuclei in the basidium just before the sterigmata are formed. The number of sterigmata on a basidium is doubtless related to the number or mode of arrangement of the nuclei derived from the fusion-nucleus. Probably in tristerigmatic basidia there are only three nuclei present instead of the usual four, /€w _CL©^® while in pentasterigmatic a'nd octosterigmatic basidia there may be five nuclei and eight nuclei respectively. If the fusion-nucleus divided three times, eight nuclei would result, and this may happen in octosterigmatic basidia ; but how the fusion- nucleus divides so as to form only three nuclei or just five is rather puzzling. No doubt light might be thrown on this matter by cytological investigation. We have so far dealt with the position of the sterigmata on individual basidia. We shall now turn our attention to the position of sterigmata on closely adjacent basidia. In certain species of Coprinus, e.g. C. eckinosporus Buller, if one examines the hymenium in face-view, one can at once observe that the spores of adjacent basidia tend to have the domino-eight arrangement (Fig. 111). If one had to crowd as many four-spored basidia as possible into a given area of the hymenium so that the spores would not touch one another, this could be done more successfully on the domino- eight plan than on any other. Proof of this is best obtained by FIG. 1 1 1. — Coprinus echinosporus Buller. Surface view of the hymenium showing plan of spores which have a domino-like arrange- ment, as in the lines 0-6 and c-d-e ; /, a cystidium. The spores of the long basidia have been shaded, those of the short basidia left plain. The echinulations are only shown at g where two spores are represented in lateral view. Magnification, 440. 322 RESEARCHES ON FUNGI trial of different arrangements. In Coprinus echinosporus, there- fore, the basidia mutually influence one another in the arrange- ment of their sterigmata so that in the end the most economical arrangement tends to be established. In Panaeolus campanulatus I have not been able to observe any tendency toward the domino- eight arrangement. Whilst I have not been able to convince myself that in Panaeolus campanulatus adjacent basidia influence each other in respect to the places of origin of their sterigmata, yet, when two or more basidia are unusually crowded, it is evident that they accommodate themselves to one another by a suitable bending of the sterigmata. Evidence of this phenomenon is afforded by the eight camera- lucida drawings given in Fig. 112. At H are shown three basidia. The middle one, which is the youngest, is crowded in between the two others. Its two lateral spores are drawn together and its other spores are separated by more than the normal distance. The position of these four spores is determined by the bending of the sterigmata. One can at once perceive that the middle basidium has beautifully adapted itself to its crowded position. Had its sterigmata had their usual direction, so that the spores would have been at the corners of a square, actual contact of at least one of the spores with the spores of one of the neighbouring basidia would have taken place. The bending of the four sterigmata, so that one pair is drawn nearer together and the other pair becomes more divergent, results in a rhomboidal arrangement of the four spores when these are seen from above. From the various drawings in Fig. 112, it becomes evident that the rhomboidal arrangement is adopted only under conditions of crowding, and in such a way as to allow of the spores of any basidium affected being situated more favourably as regards space among the spores of the surrounding basidia. In Fig. 112 at A, D, and F, the rhomboidal arrangement comes out clearly in groups of mature basidia. At A, its adoption results in larger spaces being placed between the spores of several pairs of adjacent basidia. At D and F, where the necessity for the rhomboidal arrangement is very great, this arrangement is very marked indeed ; but, at D, it has not been quite sufficient to prevent the rare PANAEOLUS CAMPANULATUS 323 accident of actual contact of two of the spores. At B, 0, E, and H, the spores are all young, for none of them have yet turned •:• •;*• * © *• *v. . • . •:• ••• •:• v.« °o° •.%•••* •:.:... . ::•': . ••*•£• o°0°°°<>0 y. °o° * • •• w*~ F G •• •• H FIG. 112. — Panaeolus campanulatus. Studies of the relative positions of spores on adjacent basidia. As if to avoid jostling with the spores of adjacent basidia, the spores of certain basidia have a rhomboidal arrangement. This is clear in A, D, F, and H, where the spores have attained full size, but the phenomenon is also to be seen in B, C, and E, where many of the spores are only partially grown. In D, actual contact of two spores occurs, and in F is only just avoided. At G, the spores have nearly a domino arrange- ment. Ripe or nearly ripe spores are black, those full-sized and turning black are dotted, spores which are still colourless are unshaded. Magnification, 293. black and some, indeed, have not yet attained full size. Yet, even here, the rhomboidal arrangement can be clearly seen. This shows that the sterigmata take up their final positions, so far as divergence is concerned, during their development, and further, that the sterigmata grow outwards in such directions that in the 324 RESEARCHES ON FUNGI end the spores will be placed in as favourable positions as possible. Evidently, adjacent basidia exercise a mutual influence of a complex kind upon one another during development : the two nearest sterigmata of two basidia which develop at the same time in close proximity tend to bend away from one another : when one sterigma of a basidium bends inwards toward the basidium- axis, the opposite sterigma bends inwards also, and at the same time the other pair of sterigmata become divergent as if repelled from the first pair. This results in the rhomboidal arrangement which can so readily be observed. The Cheilocystidia. — In addition to the basidia and the paraphyses, the hymenium of many of the Agaricineae contains elements of a third kind which are known as cystidia. The cystidia are usually very protuberant, and are often considerably larger than either the paraphyses or basidia. They were originally called cystidia because they are often swollen and appear as cysts. Cystidia occur in the genera Coprinus, Pluteus, Inocybe, Russula, etc. Now the cystidia on the gills of many Agaricineae can be divided into two groups according to their location. Those which occur on the sides of the gills we may call pleurocystidia on account of their lateral position, whilst those which are present on the free lower edges of the gills we may call cheilocystidia. Some lamellate fungi possess both these kinds of cystidia, others one kind only, and yet others are lacking in both. Panaeolus campanulatus is unprovided with pleurocystidia, but possesses a considerable number of cheilocystidia. The free gill-margin of this species is devoid of basidia and therefore of spores, and the cheilocystidia are present in such large numbers that the gill-margin appears to the naked eye as a white line. My investigations upon the cheilocystidia of Panaeolus cam- panulatus were made upon freshly-gathered fruit-bodies obtained from a grassy field in England close by a house which I was using as a laboratory. The fringe of cheilocystidia extends upwards above the extreme edge of the gills in some fruit-bodies to a maximum distance of 0-1 mm., but in others not so high. Each cheilocystidium has a somewhat swollen base embedded in the hymenium and is prolonged outwards beyond the general level of PANAEOLUS CAMPANULATUS 325 the hymenium into a fine cylindrical hair, so that the whole cell is often 0-05-0-06 mm. long (Fig. 113, A). The tip of the cell usually excretes a drop of mucilage (B), and often the drops of neighbouring hairs become confluent, as is shown in Fig. 113 at C. Some of the largest drops, formed by the confluence of drops from about twelve cystidia, may be as much as 0 • 1 mm. wide and can just be distinguished by the naked eye. It was FIG. 113. — Panaeolus campanulatus. The cheilocystidia and their excretions. A, two young isolated cystidia. B, two cystidia projecting from a gill-edge, one of which has excreted a drop of mucilaginous fluid. C, part of the free lower margin of a gill, turned upwards and seen from the side, showing the hymenium on the side of the gill and the cheilocystidia projecting outwards from the gill-edge ; a, a ball of mucilaginous fluid formed by the confluence of two drops excreted at the tips of two cystidia ; b, a hemispherical mass of mucilaginous fluid formed by the confluence of about ten drops excreted at the tips of as many cystidia. Magnification, 293. found that such drops tend to be plano-convex, apparently owing to the fact that the drops are unable to pass from the tips down the shafts of the cystidia (Fig. 113, C, &). When such a gill-edge as is shown in Fig.' 113 is placed in water, the mucilaginous drops can still be seen for some time. They are not therefore imme- diately soluble in water. When exposed to air, the drops do not dry up anything like so quickly as would equally large drops of pure water. Knoll investigated similar hairs to these for several species of Agaricineae and came to the conclusion that they are hydathodes. He therefore called them trichome-hydathodes.1 The 1 F. Knoll, " Untersuchungen iiber den Bau und die Function der Cystiden und verwandter Organe," Jahrb. f. wiss. Bot., Bd. 50, 1912, pp. 453-501. 326 RESEARCHES ON FUNGI extent, however, to which such organs reduce the amount of water in a pileus has not yet been measured. It is conceivable that the cheilocystidia are useful during the early develop- ment of the hymenium in removing from the gill-cells some of the superfluous water, and thus hastening the flow of nutrient substances which pour down the gill through the trama and subhymenium into the basidia. CHAPTER XI STROPHARIA SEMIGLOBATA General Remarks — Description of the Hymenium in Detail — -Rate of Discharge of the Spores of a Basidium and Collapse of the Basidium-body — The Spore- fall Period— Wasted Spores General Remarks.— In this and the two following Chapters, we shall endeavour to extend our knowledge of the Panaeolus Sub- type by studies made upon Stropharia semiglobata, Anellaria separata, and Psalliota campestris, the Common Mushroom. Stropharia semiglobata is one of the commonest of field fungi. Like Panaeolus campanulatus, it is coprophilous and grows on horse dung in pastures throughout Europe and North America. Whilst wandering over grazing grounds at Sutton Park in Warwickshire, England, in the autumn, I have often seen hundreds of fruit-bodies on a single day. The fungus also occurs at Winnipeg and has come up spontaneously on horse dung which has been brought into the laboratory. There can be but little doubt that the spores of this species, like those of the Coprini and coprophilous fungi in general, are able to pass uninjured through the alimentary canal of horses and cattle. A proof of this statement, so far as horses are con- cerned, seems to be afforded by the following consideration. The winter at Winnipeg is so severe and prolonged that, for a period of about five months, no fungus can develop its fruit-bodies out-of- doors. In mid-winter, when horses drop their faeces on the snow in the open, the dung-balls immediately become frozen. Now, at this season, infection by air-borne spores of Stropharia is impossible since, owing to winter conditions, no such spores are being liberated. Yet, on several occasions, dung-balls, after having been collected in January or February in the frozen condition and set in covered crystallising dishes in the laboratory, have given rise in the course 327 328 RESEARCHES ON FUNGI of a few weeks to fruit-bodies of Stropharia semiglobata ; and one such fruit-body is shown attached to its dung-ball in Fig. 114. It seems clear, therefore, that the spores of the Stropharia must have found their way into the dung-balls by passing down the alimentary canal with the hay. A fuller explanation of the spontaneous development of the fruit-bodies on the dung-balls is as follows. The fungus comes up in the summer on dung dropped by herbivorous animals upon their grazing grounds . Hence the spores liberated from the pilei, after being carried by the wind, settle upon, and become adherent to, the surrounding grass. The grass in due time is made into hay and in this state is fed to horses in Winnipeg during the winter. The horses swallow the spores with their fodder, but the spores resist the action of the digestive juices and are in no way injured during their passage through the alimentary canal. The spores therefore become embedded in the dung-balls FIG. 114. — Stropharia semiglobata. Photo- which, when dropped in the graph of a fruit-body grown on sterilised . • , . . , horse dung. Spores were being actively intensely Cold air upon the frozen snow of the streets, liberated under a bell-jar and some of them have settled on the annulus. Natural size. freeze the of STROPHARIA SEMIGLOBATA 329 few minutes into solid masses. As soon as such frozen dung-balls are brought into the laboratory and thawed, the spores within them rapidly germinate and produce an extensive mycelium which, at the end of a few weeks, gives rise to fruit-bodies. The mycelium retains its vitality in the dried condition for at least five months and probably much longer. I gathered some dung-balls in a field in England in August, 1911, allowed them to dry, and took them in the dried state to Winnipeg, where they were again moistened in January, 1912. A few days after the re-moisten- ing, I observed that several fruit-bodies of Stropharia semiglobata had begun to come up on the dung. These grew rapidly in size and, at length, after elongating their stipes and expanding their pilei, began to liberate a cloud of spores. Thus convincing evidence was afforded that the mycelium of the fungus, after five months of quiescence in the dry dung-balls, had been able to resume its development in a normal manner. The fruit-bodies of Stropharia semiglobata occur singly or in little groups, and occasionally two or three are joined at the base. In the wild state they are from 5 to 14 cm. in height, the cap 1 to 3 cm. broad, and the stipes, except for the somewhat swollen base, 2 to 4 mm. in thickness. Fig. 114 shows a fruit-body of large size grown in the laboratory, and Fig. 115 another fruit-body of about the maximum size which was gathered growing in a field in England. The outer surface of both the pileus and stipe are whitish yellow and very glutinous when moist. The fungus is regarded as poisonous. The pileus is rounded at first but soon becomes semi-globate, from which shape it has received its specific name. It is smooth, with white flesh fairly thick at the centre and thinner toward the margin, even, viscous or very viscid when moist, but like smooth leather when dry. The viscid pellicle which is more or less trans- parent can be torn off the white flesh. It consists of hyphae the outer walls of which are gelatinous and confluent. The gills (Fig. 115, B) are adnate and squarely set against the stipe, often with a tiny decurrent tooth, broad, smooth, thin, and mottled purplish- brown. The spores in the mass are brownish-purple. The stipe is viscid in moist weather. Springtails which try to ascend it, as well as bits of grass-stems, etc., sometimes become stuck in 330 RESEARCHES ON FUNGI the gluten (Fig. 115, A). The annulus is situated not far below the pileus, is glutinous when moist, and sometimes (Fig. 114) becomes discoloured with spores which have fallen and lodged upon it. There is a general resem- blance in the form of the fruit-bodies of a number of coprophilous fungi which grow on horse dung. Among these may be mentioned : Panaeolus campanulatus, Stropharia semiglobata, Galera tenera, and Coprinus ster- quilinus. In each of these the stipe is much elongated relatively to the breadth of the pileus, thin, and hollow. Moreover, the stipes are usually attached at a low point in the substratum (Fig. 114). We seem here to have presented to us an adapta- tion to a coprophilous mode of life. The fruit-bodies arise at the base of dung- balls, so that they become the better attached ; while a relatively long stipe is necessary in order that the FIG. 115. — Stropharia semiglobata. A fruit- pileus shall be brought Out body obtained from a field near Binning- ham, England. A, shedding spores ; from between the dung-balls of gmsT'^aturlTste11.' Sh°Wing mOttHng and raised Wel1 ab°Ve the surface of the dung-ball heap. The stipes are thin, elastic, and hollow, so that with the least expenditure of material the maximum of rigidity is attained. In B STROPHARIA SEMIGLOBATA 331 Fig. 114 the lashing of the stipe to the side of a dung-ball for a distance of about 2 cm. by means of mycelial hyphae can be clearly observed. The gills of Stropharia semiglobata possess all the characters of the Aequi-hymeniiferous Type, i.e. they are wedge-shaped in cross- section and positively geotropic, so that the hymenium everywhere looks more or less downwards toward the earth ; and every part of the hymenium of each gill actively produces and liberates spores during the whole period of spore-discharge. The gills also exhibit all the characters which have been enumerated for the Panaeolus Sub -type, as is indicated superficially by the mottling, which differs but little from that of Panaeolus campanulatus (Fig. 115, B). Whereas the spores of the Panaeolus are black, those of the Stropharia are purplish -brown. The dark areas of the latter fungus are therefore necessarily not so dark as those of the former. Description of the Hymenium in Detail.— In Fig. 116 is shown a portion of the mottled surface of a gill actively engaged in producing and liberating spores. In the centre is a white area which, as indicated by the dotted lines and arrows, is enlarging itself centrifugally at the expense of the two dark areas. The two waves of development are proceeding in opposite directions. Only the spores have been drawn : those shown black are pigmented purple- brown, whilst those shown white are very young and possess walls which are still colourless. The preparation from which the drawing was made was obtained by removing a gill from a living fruit-body growing on horse dung in the laboratory, laying the gill flat on a glass slide, and covering it with a cover-glass without the use of any mounting fluid. During the few minutes while the drawing was being made with the camera lucida, the basidium e collapsed and dragged down its spores on to the hymenium. The spores of d, f, h, and i were also dragged down in a similar manner but have been represented as they were first observed. Under normal conditions there is no doubt that all these spores would have been shot out into an interlamellar space and would have escaped from the fruit-body. At c, when the drawing was begun, two of the four spores had already been discharged and at d one of the spores. From other observations, shortly to be recorded, one may conclude : 332 RESEARCHES ON FUNGI °oV, 00 00 /oo o o o o • -' o o )O ^o (1) that the full-sized spores, shown at j in the white area a—b, are a little more than an hour old ; (2) that the partially-grown spores near 6 are only a few minutes old ; and (3) that the oldest spores in the two dark areas are those nearest to the dotted lines, their age being about five hours and thirty minutes. A camera-lucida drawing showing the spores from the middle of a large black area is given in Fig. 117. From this it may be observed that the basidia of the present generation are fairly com- pactly arranged, but that any one basidium is suffici- ently far from its nearest neighbours to prevent the risk of its being mechanically inter- fered with during the production and liberation of its spores. The rhomboidal arrangement of the four spores, the meaning of which was explained in connection with Panaeolus campanulatus,1 often occurs in Stropharia semiglobata and 1 Chap. X, pp. 322-324. Fio. 116. — Stropharia semiglobata. Surface view of hymenium of a young fruit-body showing part of a white area, a-b, separating two black areas. The white area is enlarging at the expense of the black areas : it is progressing by two waves in- dicated by the dotted lines and arrows. The basidia c, d, and g had already shed some of their spores when the sketch was begun. Whilst the sketch was being made, the basidium, et collapsed and dragged down the spores on to the hymenium. The spores of d, f, h, and i were also dragged down in a similar manner. Under normal conditions, all these spores would have been shot away into the interlamellar spaces, j, the oldest spores in the white area, possibly an hour old since their growth began. The spores in the white area which are still only partly grown are only a few minutes old. Magnification, 293. STROPHARIA SEMIGLOBATA 333 is well exhibited for the spores of a basidium just to the left of the middle of the Figure. By the same methods as those which were employed for Panaeolus campanulatus, the hymenium of Stropharia semiglobata has been analysed in detail. At A, B, and C in Fig. 118 are represented pieces of the hymenium of a fruit-body which had been shedding spores for one, two, and about six days re- spectively. We thus have presented to us a series of illustrations of the hymenium showing its gradual exhaustion, which is complete at C. A day was reckoned as 24 hours, and only such elements were drawn as could be clearly perceived with the microscope. The cells of the hymenium can be divided into the following six classes : (1) past - generation basidia, a a, every- where distinguished by their being collapsed, non-protuberant, and devoid of protoplasmic contents, and by containing sterigmatic stumps within their concave tops ; (2) present-generation basidia, b b, fully protuberant, with the maximum diameter, and each bearing four spores into which the protoplasm is passing or has passed ; (3) coming-generation basidia, c c, fully protuberant, with the maximum diameter, densely filled with protoplasm, lacking spores, but either with sterigmata, as in B, or without them, as in A ; (4) future-generation basidia, d d, non- protuberant, with less than the maximum diameter, filled with fine protoplasmic contents, and lacking both sterigmata and spores ; FIG. 117. — Stropharia semiglobata. A dark area of the hymenium with only the spores repre- sented, to show the density of distribution of the basidia of the present generation. The basidia bearing spores are only one-sixth to one-seventh of the whole number. Each side of the square = 0-15 mm. Drawn with the camera lucida. Magnification, 440. 334 RESEARCHES ON FUNGI (5) paraphyses, e e, numerous non-protuberant rounded or oval elements, more or less hidden by the enlarged tops of the basidia, never bearing sterigmata or spores, and becoming larger and more and more emptied of their protoplasmic contents with the gradual exhaustion of the hymenium ; and (6) pleurocystidia, f, elements ABC FIG. 118. — Stropharia semiglobata. Surface view of the hymenium : A, one day after the opening of the pileus and beginning of spore-discharge ; B, two days after ; and C, about eight days after, when spore-discharge has ceased, a, a past-generation basidium (with four sterigmatic stumps) ; b, a present-genera- tion basidium (bearing spores) ; c, a coming-generation basidium (heavily shaded with dots) ; d, a future -generation basidium (lightly shaded with dots) ; e, a paraphysis (unshaded) ; w, waste spores adhering to the hymenium ; /, a cystidium. In A and B respectively the number of elements on the equal areas is as follows : past-generation basidia, 10 and 20 ; present-generation basidia, 6 and 6 ; coming -generation basidia, 5 and 4 ; future -generation basidia, 23 and 10 ; paraphyses, about equal numbers. In C, all the basidia now belong to past generations, and the paraphyses have become larger and can therefore be more clearly distinguished. In A and B, only those para- physes are drawn which could be clearly distinguished, the rest are omitted. Magnification, 580. occurring singly (one shown in C), few in number, scattered irregu- larly throughout the hymenium, with dense persistent contents, differing from the basidia in never producing sterigmata and spores and from both basidia and paraphyses by ending in a protuberant peg-like point. Pleurocystidia, i.e. cystidia which are situated on the sides of the gills, do not occur in Panaeolus campanulatus, so that, in possessing these elements in this situation, Stropharia semiglobata differs from that fungus and comes to have six classes of hy menial elements instead of only five. STROPHARIA SEMIGLOBATA 335 The gradual exhaustion of the hymenium, as shown in Fig. 118, may be realised at a glance by noting the gradual increase of the past-generation basidia as one passes from A to C. Corresponding with this increase is a decrease in the number of future-generation basidia. The number of present-generation and coming -generation basidia is about the same for A and B. This accords with other observations which have taught me that, in Stropharia semiglobata, during the first few days of the spore-discharge period, on any given area of the hymenium, the number of basidia producing spores at any one time remains fairly constant, and that, as in Panaeolus campanulatus, during the last few days of the spore-discharge The Number of Elements in the Hymenium of Stropharia semiglobata. Age of Fruit-body after Commencement of Spore-discharge. "N"atnrp nf "Flompnt^ Area A : one day. Area B : two days. AreaC: sis days. §9 Past 10 20 32 i-a d'3 Present . . . 6 6 0 |1 Coming .... 5 4 0 5*0 Future .... 23 10 0 Paraphyses ..... 20 + x (= 24?) 20+*, (=24?) 44 period when the activity of the hymenium is approaching its close, there is a tendency for this number to decrease. In the area A there are present about 44 basidia. Let us assume that the average number of basidia which successively bring their spores to maturity on this area is 5. Then, by division, we can calculate that, alto- gether, such an area as A would give rise to 9 successive generations of basidia. If, however, we assume what may well be nearer the truth, namely, that the average number of basidia of each generation is 4 instead of 5, then the number of successive generations on the area A would be increased to 1 1 . The size of the paraphyses gradually increases with the progressive collapse of more and more basidia. This fact may be verified by comparing the few paraphyses clearly outlined in the area A with the paraphyses shown in the area C. In the above Table the nature and number of the elements in all three of the areas shown in Fig. 118 are set forth. Parts 336 RESEARCHES ON FUNGI of elements have been counted as wholes. The letters x and y in connection with the paraphyses indicate paraphyses of uncertain number which were not clearly perceived owing to optical difficulties. However, a comparison of the areas A and B with the area C suggests that x and y are each equal to about 24. The paraphyses, i.e. the permanently sterile elements, are just as distinct in the hymenium of Stropharia semiglobata as in that of Panaeolus campanulatus. They can be most easily distinguished in a piece of exhausted hymenium such as that shown in Fig. 119. Here the tops of the contracted basidia 6 somewhat overlie the paraphyses p. A pleurocystidium is shown at c, and waste spores at w. A more complete analysis of the exhausted hymenium is given in Fig. 120, where the basidia and paraphyses (from different areas) have been drawn separately. At A there are shown the basidia with- out the paraphyses, and at B the paraphyses without the basidia. The drawing C is similar to B except for the fact that it includes four pleuro- cystidia, whereas none of these ele- ments are present in B. By comparing A with B and C, which all represent equal areas of the hymenium, one may observe that the paraphyses are somewhat more numerous than the basidia and, further, that the paraphyses tend to be connected into chains so that they avoid isolation. On the other hand, individual basidia often stand apart from their neighbours, from which they are separated by paraphyses. Two transverse sections through the hymenium and sub- hymenium of Stropharia semiglobata are represented in Fig. 121. The upper one, A, shows the hymenium in the early part of the spore-discharge period, i.e. at a stage about equivalent to that indicated in the surface view represented in Fig. 118, A ; while the FIG. 119. — Stropharia semiglobata. Exhausted hymenium in surface view. 6, collapsed basidia, each with four sterigmatic stumps ; p, paraphyses ; c, a cystidium ; w, waste or undischarged spores. Magnification, 440. STROPHARIA SEMIGLOBATA 337 lower one, B, shows the hymenium in the exhausted condition at the end of the spore -discharge period, i.e. at a stage correspond- ing to that indicated in Fig. 118, C. In the upper drawing we can discern that the hymenium is made up of past-generations basidia (a), present-generation basidia (b), coming-generation basidia (c), future-generations basidia (d), small sterile paraphyses (e), and cystidia (/). The cystidium is club-shaped and has a slender stalk B C FIG. 120. — Stropliaria semiglobata. Sketches of portions of an exhausted hymenium after spore-discharge has ceased. A shows : the collapsed basidia, b, each with four sterigmatic stumps ; a cystidium, c ; and parts of two waste spores, w. The paraphyses are omitted. B shows paraphyses, p, and one waste spore, w, only. The basidia are omitted and no cystidia are present. C shows : paraphyses, p ; four cystidia, c ; and two waste spores, w. Magnification, 440. which springs from one of the lowest cells of the subhymenium. The present-generation basidium farthest to the left has just dis- charged its first spore and is about to discharge its second. As a preliminary to the second discharge a drop of water has been excreted at the spore-hilum. The drop has developed within the last five seconds and has almost attained its maximum size. In the lower drawing it will be seen that the exhausted hymenium is made up of collapsed past-generations basidia (a), swollen paraphyses (e), and cystidia (/). The large cell, g, is an aborted basidium. It contained thick protoplasm on its walls and a large central vacuole, and it seems to have become swollen instead of producing spores. That the basidia in any small area of the hymenium come to 338 RESEARCHES ON FUNGI maturity in a series of successive generations was proved not merely by inference from such studies as have been recorded in the drawings of Fig. 118, but also by direct observation. A fruit-body, which had not long expanded its pileus and was in the earliest phase of its B FIG. 121. — Stropharia semiglobata. Transverse sections through a gill showing the hymenium and subhymenium. A, about a day after the beginning of spore-discharge ; a a, past-generation basidia ; b b, present -generation basidia ; c c, coming-generation basidia ; e e, paraphyses ; /, a cystidium ; the present - generation basidium farthest to the left has just discharged one spore and is about to discharge another spore the hilum of which has just excreted a water- drop. B, from an exhausted gill at the end of the spore-discharge period ; a a, past-generation basidia ; e e, paraphyses ; /, a cystidium ; g, an aborted basidium which has become much swollen, but which has not produced any Magnification, 588. spore-discharge period, was obtained from a pasture. Straightway one of its gills was dissected off the pileus and placed flat in a com- pressor cell, and a little water was made to pass between the lower hy menial surface and the glass base of the apparatus. Another drop of water of small size was then placed on the glass base in a position where it could not touch the gill, and the preparation was completed by capping the compressor cell with its lid. On examin- ing the preparation with the low power of the microscope, it was found that the upper hymenium of the gill afforded an admirable STROPHARIA SEMIGLOBATA 339 object for study. There was a minute crack in the cover-glass, and it was found necessary to renew the water in the cell once during the investigation. A small portion of a dark area, from which the spores were being discharged, was chosen for special study ; and in the course of eight hours' continuous watching by myself and by my nephew, Mr. Bernard Workman, to whom I am much indebted for timely assistance, I succeeded in observing the following : (1) the last phase of one generation of basidia, including the discharge of the spores and the collapse of the basidium-bodies ; (2) the development and discharge of the spores of a second genera- tion of basidia, followed by the collapse of the basidium-bodies ; and (3) the development of spores on the sterigmata of a third generation of basidia. The horizontal-microscope method, which was employed in a similar investigation for Panaeolus campanu- latus,1 has various advantages over the just-described vertical- microscope method, for it permits one to observe a gill which is attached to a fruit-body in such a way that the spores discharged from it are shot clear of the hymenium, and it permits of the obser- vations being carried on for several days instead of for a portion of one. Nevertheless, the vertical-microscope method employed in the present investigation sufficed me to attain the end in view. Its disadvantages are that the gill has to be removed from the fruit- body and therefore will not continue to develop normally for many hours, and also that the spores which are shot upwards must neces- sarily fall back again on to the hymenium where their accumulation tends to hinder one from seeing the hy menial elements clearly. Its chief advantage lies in its simplicity, the only apparatus involved being an ordinary microscope and a compressor cell. The developmental waves, which have already been described as passing through the hymenium, were actually observed in their progress. In the centre of the field of the microscope was an area which was at first dark, and waves of development travelled out from it radially, somewhat in the way represented in Fig. 116 (p. 332). The passage of a single undulation across the field of the microscope occupied several hours. Close attention to the small area specially chosen for minute 1 Chap. X, pp. 260-264. 340 RESEARCHES ON FUNGI study revealed the following facts. The sterigmata of the coming- generation basidia begin to develop a short time (30 minutes ?) before the immediately adjacent present-generation basidia shoot away their spores (cf. Fig. 118, B, c, p. 334). Spores begin to develop on the sterigmata of a second-generation basidium about twenty minutes after the spores of an adjacent first-generation basidium have been shot away. The spores are rather slow in developing from their first rudiments to full size, for an hour and ten minutes are occupied in the process. Pigmentation of the spore-walls begins about one hour after the spores have attained full size, and continues for about two hours and a half. The spores remain on their sterigmata about one hour after they have become fully pigmented and, invariably, their violent discharge is preceded by the excretion of a drop of water at the hilum. Waste spores accumulate on the hymenium as a result of over-excretion of the water-drop, in the same manner as was described for Panaeolus campanulatus. The time occupied by the development and dis- charge of the spores of the second-generation basidia observed, from the moment when the spores of adjacent first-generation basidia were discharged, was about six hours. The chief facts observed in connection with the second-generation basidia under observation may be set out as follows. The times represent the average for six basidia. The zero of the time-scale is the moment of the first appearance of spore-rudiments on the tops of the sterigmata. Development of the Spores of a Single Generation of Basidia in Stropharia semiglobata. Observations. Hours. Minutes. First appearance of spore-rudiments on the sterigmata Spores have grown to full size The spore-walls begin to exhibit pigmentation The spore-walls are fully pigmented .... The first of the four spores is discharged The spores on basidia of the succeeding generation begin to develop . . . . ... 0 0 1 10 2 10 4 40 5 40 6 0 I suspect that the later generations of basidia, as in Panaeolus campanulatus, take increasingly longer periods to develop and STROPHARIA SEMIGLOBATA discharge their spores. Possibly the six hours actually observed for one of the earliest of the basidial generations may be lengthened out to seven or eight for the later generations. Rate of Discharge of the Spores of a Basidium and Collapse of the Basidium-body. — The following observations were made with the horizontal microscope in the manner fully described for Panaeolus campanulatus (Fig. 91, p. 261). The gill attached to the pileus was developing normally. For several basidia the times of discharge for the four spores were carefully observed. I called each observation and my laboratory attendant, Mr. S. G. Churchward, kindly recorded the time in seconds. The four spores of any one basidium were always discharged in succession in the course of about two to ten minutes. If we make the zero of the time-scale the moment at which the first of the four spores is shot away, then the times for the discharge of all four spores for three basidia which were observed may be set out as follows. The Discharge of the Four Spores of a Single Basidium in Stropharia semiglobata. Basidium I. Basidium n. Basidium m. Time of Discharge. Time of Discharge. Time of Discharge. Minutes. Seconds. Minutes. Seconds. Minutes. Seconds. First spore .... Second spore Third spore .... Fourth spore . ••' 0 0 4 48 5 33 5 53 0 0 0 53 5 48 7 15 0 0 1 32 1 37 1 55 One may conclude from these observations that the intervals between the discharge of the first and second, of the second and third, and of the third and fourth spores, are by no means constant but highly variable, ranging between nearly five minutes and only five seconds. It is not unlikely that very rarely two spores are shot off from the same basidium at practically the same in- stant ; but successive discharge is evidently the rule. The average length of time taken for the discharge of all the spores from each of three basidia was five minutes and one second. 342 RESEARCHES ON FUNGI During the time that a basidium is discharging its spores, the basidium-body and the sterigmata do not appear to undergo any alteration. The basidium-body, before, during, and immediately after spore-discharge, owing to the convexity of its top, appears light in the centre and dark at the periphery. I watched several basidia shed all their spores, and continued to watch them for some time after the last spore of each had disappeared, in order to become acquainted with the phenomenon of their collapse. It was found that, until about twenty minutes after the last spore has been shot away, a basidium remains prominent and apparently unaltered, but that at the end of this time it undergoes a rapid change. In the course of about one more minute, its sterigmata are drawn together and the end of the basidium-body becomes darkened and relatively difficult to observe. There can be no doubt that during this minute the basidium collapses : the end of the basidium- body becomes concave, the sterigmata are drawn down into the concavity so formed, and at the same time the basidium sinks so that its end is no longer protuberant beyond the general level of the hymenium marked by the exterior of the paraphyses. For four different basidia the actual moment at which the process of collapse began after the discharge of the last spore was observed to be 23, 16, 16, and 20 minutes respectively, or, upon the average, about 19 minutes. The actual moment of the collapse of the basidia of any one generation precedes only by a few minutes the beginning of the development of the spores on the sterigmata of the basidia of the succeeding generation. If, owing to any miscarriage of develop- ment, the spores of a basidium do not happen to be discharged when they should be under normal conditions, the basidium collapses apparently at the normal time, i.e. at the time at which it would have collapsed if it had discharged its spores. It is obvious, there- fore, that basidial collapse takes place just sufficiently soon to secure that any undischarged spores of one generation shall be dragged down to the general level of the hymenium before the already protuberant basidia of the succeeding generation begin to develop their spores. The undischarged spores of one generation are moved by basidial collapse from their place of origin to their STROPHARIA SEMIGLOBATA 343 place of rest in so timely a manner that they are unable even to touch, much less adhere to and interfere with, the spores of the basidia of the next generation. This protection of the newly- developing spores from the mechanical contact of old undischarged spores is one of those fine points in the organisation of the hymenium which only a very detailed investigation could possibly reveal. In concluding the discussion of this subject, it may be remarked that, since a basidium collapses very shortly after discharging its last spore, it is clear that it produces one set of spores and one only. The Spore-fall Period.— Spore-deposits were collected during successive days beneath the pileus of a fruit-body grown on horse dung under a bell-jar in the laboratory. The fruit-body had just opened when the observations were begun. A certain number of spores was shed during the first day, but the most copious deposits were collected during the second and third days. The spore- deposit made during the fourth day was relatively poor. During the fifth and sixth succeeding days, spores continued to be liberated but in ever diminishing numbers until, on the seventh day, spore- discharge ceased and the fruit-body collapsed. During the last stages of hymenial activity, the gills lost their mottled appearance and took on a uniform dirty-yellow colour. We may conclude, therefore, that spore-discharge from a fruit-body may continue for about six days but that it is most active during the first three days. Wasted Spores. — Spores which fail to be properly discharged become attached to the hymenium of Stropharia semiglobata in the manner already described for Panaeolus campanulatus, but, at least occasionally, in somewhat greater numbers than for the latter fungus. A fruit-body, which had been grown on horse dung under a bell- jar in the laboratory, was examined as soon as it had ceased to shed spores. All the wasted spores on an area of the hymenium 0 • 5 mm. long and 0 • 33 mm. wide were sketched with the camera lucida as shown in Fig. 122. The cystidia k k were drawn in a similar manner. The exhausted basidia and the paraphyses were put into the drawing semi-diagrammatically by knitting together various exact camera-lucida sketches similar to that shown in 344 RESEARCHES ON FUNGI Fig. 119 (p. 336). The resulting Fig. 122 forms a basis for calculating the percentage of wasted spores. It was found by counting that, on the area of the hymenium under discussion, 'there were 231 wasted spores. But the area contains 2,896 sterigmatic stumps. Since each sterigma must have given rise to a spore, the total number of spores produced by the area must have been about 2,896. From these figures it may be calculated that the wasted spores were just under 8 per cent, of the whole number produced. Some areas of the hymenium, as large as the one investigated, were observed to be much more free from wasted spores, and others less free. Probably, for the whole fruit-body, the percentage of wasted spores was between 7 and 8. Possibly, in nature, the number of spores which are wasted because they are not properly liberated is less than that obtained in my investigation in the laboratory. Under a bell- jar the air surrounding a fruit-body grown on moist horse dung tends, especially at night, to become saturated with moisture ; and it is not unlikely that inhibited transpiration is prejudicial to the discharge of spores. It may be that the water-drop defect,1 which was fully dealt with in connection with Panaeolus campanulatus, becomes augmented under these conditions, with a consequent decrease in the number of spores set free. Investigations upon the number of spores wasted by fruit-bodies of Stropharia semiglobata under natural conditions, however, have not been undertaken. The wasted spores are by no means uniform in appearance. Thus in Fig. 122, while most of the spores are pigmented, thirty- two of them, as indicated by the shading, have walls which are either colourless or only partially coloured. There are several heaps of four spores, each heap representing the product of a single basidium. The heaps shown at a, b, c, d, e, f, g, h, and i, consist of immature spores, and those shown at j and I of mature spores. Some of the spores, e.g. those at I, are under-sized, whilst others, e.g. three in the middle of the Figure at the top, are much larger than the average. Here, as in other species of Hymenomycetes, one often finds monstrous spores among those which are wasted. From data derived from Fig. 122, one may calculate that the number 1 Fig. 104, p. 308. STROPHARIA SEMIGLOBATA 345 FIG. 122. — Stropharia semiglobata. An area of the hymenium 0-5 mm. long and 0-33 mm. wide, drawn after spore- discharge had ceased, showing the exact number and arrangement of the spores which were left adhering to the hymenium. The spores shown in black were fully pigmented, those shaded heavily with dots partially pigmented, and those shaded lightly with dots colourless. The hymenium is made up of : exhausted basidia, m, each with four sterigmatic stumps ; paraphyses, n ; and seven cystidia, k, irregularly scattered. There are several heaps of four spores, each heap represent- ing the product of a single basidium : a, 6, c, d, e, f, g, h, i, groups of immature spores ; j and I, groups of fully pig- mented spores. The waste sporos form just under 8 per cent, of all the spores which were produced on the whole area. Magnification, 323. 346 RESEARCHES ON FUNGI of spores produced on one square millimetre of the hymenial surface of Stropharia semiglobata is about 9,000 and that of these, under laboratory conditions, some 720 may be wasted. When the figures are set forth in this way, one perceives that, notwithstanding the number of wasted spores, the efficiency of the fruit-body as a whole for the production and liberation of spores still remains very high. In concluding this Chapter, it may be stated that there is no essential difference in the structure of the hymenium of Stropharia semiglobata and of Panaeolus campanulatus. Both are built on the same plan and function in the same manner. The fruit-bodies of both species provide us with excellent examples of the Panaeolus Sub-type of organisation. CHAPTER XII ANELLARIA SEPARATA General Remarks — The Production and Liberation of Spores — A Photograph of the Hymenium General Remarks. — Anellaria separata, like Panaeolus campanu- latus and Stropharia semiglobata, is one of the best known of copro- philous fungi. It commonly occurs in England on horse dung or cow dung in pastures in the manner shown in Fig. 123, and I have also found it coming up spontaneously on horse dung brought into my laboratory from the streets of Winnipeg in winter. Some cow dung, obtained in South Wales, was dried by Miss E. M. Wakefield and sent to me at Winnipeg. Four months after being dried, it was moistened and placed in a damp chamber. After this, in the course of a few weeks, it gave rise to fruit-bodies of Anellaria separata (Fig. 125), from which fact we may draw the conclusion that the mycelium can retain its vitality in the desic- cated condition for a period of at least four months. In nature, during dry spells of weather in summer, it not infrequently happens that dung-masses are baked to dryness by the sun's heat. It is evident, from the observations just described, that such drying does not necessarily impair the vitality of the mycelium of Anellaria separata with which the dung may be infected ; and, doubtless, in dried dung in fields, the mycelium often remains at rest for many days or even weeks until rain comes again. Then the mycelium resumes its growth and may give rise to fruit-bodies. The exact state which the mycelium assumes when becoming quiescent, how- ever, requires further investigation. There is no difficulty in cultivating the fungus from spores in the laboratory. Some horse dung was sterilised and then infected with the spores of a fruit-body which had come up spontaneously 347 348 RESEARCHES ON FUNGI ANELLARIA SEPARATA 349 on cow dung. On the twenty-ninth day after the spores had been sown, the first fruit-body began to shed its spores. It is thus proved that the fungus can pass through the whole of its life-history, from spore to spore, in the course of a single month. The following is a brief description of the fruit-body, which differs but little from that given by Massee.1 Pileus 2 • 5 to 3 • 3 cm. high (varying up to 5' 5 cm., vide Fig. 124), 2' 5 to 4 -25 cm. broad (varying up to 6 cm., vide Fig. 124), ovate, then campanulate, not expanding, viscid, even, ochraceous, then whitish, shining, flesh rather thick ; gills adfixed, ascending, thin, crowded, broad, mottled greyish-black, margin paler ; stipe 8 to 16 cm. long, straight, base sub-clavate, attenuated upwards, whitish, shining, apex striatulate, ring persistent, distant ; spores black, opaque, 19-21 /i long, 10-11 /* broad, and 9 /* wide. Massee states that the fruit-bodies are from 3 to 5 inches long. However, I have gathered specimens 6 inches long, and according to Stevenson 2 the maximum length is about 8 inches. Some- times the pileus is very large and the stipe relatively short, as is shown in Fig. 124. Dr. Somerville Hastings has informed me that in the Alps of Switzerland, at a height of from 6,000 to 7,000 feet, the pilei of Anellaria separata are well-developed, but the length of the stipe is not more than twice the breadth of the pileus.3 It is evident that the length of the stipe of A. separata varies greatly in response to external conditions. The spores have three different dimensions : length, breadth, and thickness. The difference between the breadth and thickness may best be perceived by looking down on the spores, so that one obtains an endwise view of them like that shown in Fig. 128. Other authors have taken notice of the length and breadth only. Massee 4 states that the spores measure 10 by 7 p. My average measurements for length and breadth are 20 and 10 • 5 p, respectively, so that the spores of my specimens were twice as long as those of 1 G. Massee, British Fungus-Flora, London, 1892, vol. i, pp. 330-331. 2 J. Stevenson, Hymenomycetes Britannici, Edinburgh and London, vol. i, 1886, p. 338. 3 For a photograph of the Alpine form of Anellaria separata vide these Researches, vol. iii. 4 G. Massee, loc. cit. 350 RESEARCHES ON FUNGI Massee, and as broad as his were long. Keith's1 measurements Fio. 124. — Anellaria separata, on horse dung. A fruit-body with a very large pileus and a relatively short and thick stipe. The pileus (clay- whitish in colour) is smooth, and cracked in a characteristic manner. The annuius is unusually near the base of the stipe. Photographed at Scarborough, England, by A. E. Peck. Natural size. were 16-22 by 11-12 p. and Cotton's2 18-22 by 10-12 /*, i.e. 1 Keith, in Stevenson's Hymenomycetes. loc. cit. 2 A. D. Cotton, private communication. ANELLARIA SEPARATA substantially the same as my own. It Massee's measurements do not repre- sent the average for the species and that he either examined a fruit-body with unusually small spores or, by inadvertence, allowed some error to creep into his record of the figures obtained. The annulus upon the stipe is a structure which exhibits considerable variation. In nature, it is sometimes relatively much expanded and at other times relatively inconspicuous, although it is never absent. It is usually composed of a continuous membrane which takes the shape of an inverted cup (cf. Figs. 125 and 127). Under cultivation, the annulus is still more variable. In some of the fruit-bodies reared in the laboratory, the annulus was prefectly normal, while in others it became divided into long strings passing from the stipe to the periphery of the pileus and thus coming in a measure to resemble the cortina in the genus Cortinarius (Fig. 126). In yet other specimens no annulus was left attached to the stipe but instead there was a broad band of sterile tissue attached to the edge of the campanulate pileus. Some of the specimens, if considered by them- selves, would have been more fitly included in the ringless genus Pan- aeolus than in Anellaria. These obser- vations seem to me to show that, after all, the ring in Anellaria is not a very seems to me, therefore, that FIG. 125. — Anellaria separata. A fruit-body grown on horse dung in confinement under a bell-jar. The annulus, covered with spores at the top, is abnormally large. The black spore-deposit on the top of the pileus was formed owing to convection currents carrying spores up- wards after their discharge from the gills. Natural size. RESEARCHES ON FUNGI FIG. 126. — Anellaria sepa- rata. A fruit-body which came up spon- taneously on horse dung in a moist chamber at Winnipeg. The ab- normal veil is arachnoid like the cortina of Cortin - arius. Natural size. deep-set character. The species of Pan- aeolus and of Anellaria are so nearly alike, except for the ring, that Fries never separated them. The genus Anellaria was subsequently constructed by Karsten * to include the ringed Panaeoli. While this distinction, doubtless, is of some con- venience to field- workers, I cannot help feeling that, from the standpoint of rela- tionship, the view of Fries is the more correct one. The gap between Panaeolus and Anellaria appears to me to be very slight indeed, when compared with the gap between Panaeolus and other genera of Melanosporae, namely, Coprinus and Psathyrella. The Production and Liberation of Spores. — The organisation of the fruit-body for the production and liberation of spores was found to be the same in all essentials as in Panaeolus campanulatus. The fungus clearly belongs to the Panaeolus Sub-type of the Aequi-hymeniiferae. As a member of the Aequi-hymeniiferae, its gills are wedge-shaped in vertical section and posi- tively geotropic, so that every part of the hymenium in a normally oriented fruit-body comes to look more or less downwards. Moreover, every part of its hymenium (every square mm.) successfully produces and liberates spores during the whole period of spore-discharge. As a member of the Panaeolus Sub-type, its gills are mottled (Fig. 127, D and E). The mottling of its 1 P. A. Karsten, Rysslands, F inlands och den Skandinaviska Halfong Hattsvampar, Helsingfors, 1879, p. 517. ANELLARIA SEPARATA 353 gills is strongly marked and just as distinct as in Panaeolus cam- panulatus and Stropharia semiglobata. The organisation of the hymenium of Anellaria separata was found to be similar to that of Panaeolus campanulatus, so that it is unnecessary to describe it or to illustrate it in all its details. Among the facts yielded by a study of the living gills were the following. On each area of the hymenium the basidia come to maturity in a series of successive generations. Coming-generation basidia on any hymenial area do not develop their spores until the spores of the present generation have been shed. The time elapsing between the discharge of the spores of two successive generations of basidia on the same small hymenial area amounts to several hours. With the horizontal microscope, a single hymenial area of a gill which had been shedding spores for about 24 hours was watched in the manner already described,1 and it was found that from 10'5 to 11 hours elapsed between the spore-discharge of two successive genera- tions of basidia. Unfortunately, the fruit-body observed was not quite normal in its activities, as the proportion of its undischarged to its discharged spores was unusually large. However, notwith- standing the accumulation of wasted spores upon the hymenium, it became sufficiently obvious that the rate of development of each generation of basidia was of the same order as for Panaeolus campanulatus. Paraphyses are present in the hymenium, and they remain living until spore-discharge ceases and the fruit-body collapses. During the gradual exhaustion of the hymenium they grow in size. The shape and distribution of the paraphyses and basidia are similar to those of Panaeolus campanulatus. From the latter fungus Anellaria separata differs in possessing pleurocystidia. These, like those of Stropharia semiglobata, are few in number and scattered at relatively great intervals among the paraphyses and basidia. Each is more or less clavate and terminates in a nipple-like ending which projects slightly above the hymenium. The white margin of a gill, like that of Panaeolus campanulatus or Stropharia semiglobata, is clothed with hair-like cheilocystidia. The collapse of the basidia after they have discharged their spores was studied in more detail than in Panaeolus campanulatus. 1 Figs. 91 and 92, pp. 261 and 263. VOL. II. 2 A 354 RESEARCHES ON FUNGI FIG. 127. — Anellaria separata. Development of adnate gills. A, very young fruit-body of which B is a section. C, D, and E, other stages in succession. The long axes of the gills do not undergo a large movement just before spore - discharge begins. The mottling of the gills and the discharge of a stream of spores are also shown in D and E. Natural size. The observations were made with the horizontal microscope in the manner shown in Fig. 92 (p. 263). First, the length of time required for the discharge of all the four spores of a single basidium, counting from the discharge of the first spore, was noted. As a rule, it was found to be about five minutes, as in Stropharia semi- globata . Taking the time of discharge of the first spore as zero, the times for the discharge of the other three spores in one particular basidium were as given in the Table on the opposite page. At the moment when a basidium discharged its last spore, the time was noted. The basidium- body with its four sterig- mata could still be seen projecting from the hymenium. For some fifteen to twenty minutes the turgidity of the basidium-body appeared to be unimpaired, but at the end of this time collapse took place in a few seconds. The four sterigmata fell inwards ANELLARIA SEPARATA 355 toward the basidium-axis, the basidium-body shortened and its end changed from a convex to a concave shape. As the concavity was being produced, the basidium-body changed from a bright appearance to a uniform grey. Within a few seconds after the beginning of the collapse, the whole basidium came to resemble other collapsed basidia and could not be distinguished from them. The Discharge of the Four Spores of a Single Basidium in Anellaria separata. Spores. First. Second. 3 minutes, 30 seconds Third. Fourth. Moment of discharge 0 3 minutes, 5 minutes, 47 seconds 20 seconds I Altogether the time of collapse of four basidia was carefully noted, the zero of the time-scale being in each case the moment of discharge of the last of the four spores : The first basidium observed collapsed after 22 minutes, ,. second ,, ,, ., ,, 15 ,, ,, third „ „ „ „ 20 ,, fourth „ „ ,, ., 21 The average time of collapse was . . 19*5 ,, We may say, therefore, that in Anellaria separata the collapse of a basidium usually takes place about twenty minutes after the discharge of the last of its four spores. This interval of time is about the same as that which was observed for Stropharia Among the abnormalities of the basidia the following were noted. Occasionally, a basidium was observed to bear five spores instead of the usual four. Occasionally, a basidium was seen with one aborted spore and the other three of the usual size or possibly a little larger. The aborted spore was only about one-third of the usual diameter and therefore relatively very small in volume. A Photograph of the Hymenium. — When studying the hymenium of Panaeolus campanulatus, I endeavoured to photo- graph the living hymenium but, unfortunately, the results were 356 RESEARCHES ON FUNGI not clear enough for publication. However, with Anellaria separata, owing to improvement in my methods, I have had more success. In Fig. 128 is shown a photograph of the spores belonging to the present spore-bearing generation of basidia on one of the dark areas of the hymenium. One can see that the photograph re- sembles part of the drawing which was given for the hymenium of Panaeolus campanulatus in Fig. 89 (p. 257). The four black spores of each basidium can be clearly made out, and one can see that adjacent spore-bearing basidia are separated by such spaces that their spores cannot touch one another during development and discharge. Evidently the basidia of the present generation are all of about the same age, for their spores are highly pigmented and basidia with half -grown or colourless spores are entirely absent from the area. Another point demonstrated by the photograph is that the spore-bearing basidia are monomorphic, i.e. all protrude to about the same distance above the general level of the hymenium, for only on this supposition can one account for the fact that the spores are all in focus in a single plane. The reader may be reminded here that basidial monomorphism is one of the characteristics of the Panaeolus Sub -type. The tendency of the four spores of a basidium to be arranged in a rhomboidal manner, where a basidium is rather nearer than usual to its neighbours, is well shown just to the right of the middle of the Figure. The meaning of the arrange- ment is the same as that already explained in connection with Panaeolus campanulatus.1 A basidium at the bottom of the Figure, in the middle line, has already lost one of its spores and now only possesses three. All the spores are in view endwise on. Seen thus, they are not round but oval in outline. This confirms the state- ment, made on a previous page, that every spore of Anellaria separata has three differing dimensions : length, breadth, and thick- ness. In the middle of each group of four spores can be seen a light rounded spot. This is produced by the convex top of the basidium-body, which in concentrating the light-rays has a lens-like effect. Here and there between the groups of spores can be seen similar spots. These correspond to the convex ends of the bodies of basidia of the coming generation. 1 Pp. 322-324. ANELLARIA SEPARATA 357 The photograph reproduced in Fig. 128, so far as I know, is the first ever published which exhibits hymenial structure.1 I shall FIG. 128. — Anellaria separata. Photograph of part of a dark area of the hymenium of a living gill mounted in water. The plane of focus passes through the black spores of the present -gene ration basidia. Magnification, 640. not hesitate, therefore, to give the details of the method by which it was made. A fruit-body was grown in the laboratory and from it, on the first day of spore-discharge, a gill was dissected carefully 1 Except Fig. 107, p. 316, and Fig. 109, p. 319. 35» RESEARCHES ON FUNGI away. A drop of water was placed on a glass slide and on this the gill was floated. Another drop of water was then added at the side of the gill, whereupon the preparation was covered with a cover- glass. Under these conditions there was a layer of water between the under side of the gill and the cover-glass. On the upper side of the gill, as soon as the cover-glass had been applied, water began to make its way by capillary attraction over the hymenium, resist- ance to its passage being offered by the sterigmata and spores. When the water passed rapidly over the upper hy menial surface, as sometimes happened, the black spores were nearly all pulled off their sterigmata, whereby the whole preparation was spoiled ; but, with a very gradual passage of the water, the black spores often remained on. Thus, by gradually passing water between the upper hymenial surface and the cover-glass, the former became covered with water without being ruined. The photograph reproduced in Fig. 128 was made from such a water-immersed preparation. In actually taking the photograph, a Leitz microphotographic apparatus was employed. A microscope was fitted with a Leitz ocular, No. 4, and an objective, No. 7 ; and above the tube the camera was placed so that it had a vertical arrangement. The light was given by a Leitz " Lilliput " arc -lamp which was set 3 feet from the microscope mirror. The part of the hymenium to be photographed was first focussed in the usual way. The light employed for this purpose came from the arc -lamp ; but, in order to dim its rays, two sheets of glass, one blue and the other yellow, were set in front of it. After the focussing had been accomplished, the camera was swung into position at the top of the tube of the microscope. The piece of hymenium was then focussed on the ground-glass plate at the top of the camera. This plate was then removed and a photographic plate substituted for it. A black sheet of cardboard was next placed vertically against the base of the microscope so as to come between the mirror and the arc-lamp. Then the two sheets of coloured glass, already referred to, were removed from just in front of the arc-lamp. The carbons of the arc were then screwed close together so that the light coming from them was steadied. The cover of the photographic plate was then drawn out. Then, the black sheet of cardboard was lifted away so ANELLARIA SEPARATA 359 that the arc-light could shine on the mirror. The exposure was for about four seconds. At its end, the black cardboard sheet was replaced in its old position. Then the photographic plate was covered and taken to the dark-room for development. The magnification of the hymenium was 640 diameters. Before the spores develop pigment, or even after they have become slightly brown, they have almost the same tint as the background made up of the gill substance. It is therefore difficult to get good photographs of the white areas of the mottled gill- surface. The clearest results of my efforts in this direction do not seem to me worthy of publication. The best idea of a white area may be obtained from such a camera-lucida drawing as is shown in Fig. 89 for Panaeolus campanulatus (p. 257). We shall again return to the subject of photography of the hymenium in connection with the Coprini. With those fungi the gills were photographed without any immersion in wTater. The same method can be applied to Anellaria separata, but the results which I thus obtained were not so good as those obtained by the method just described. CHAPTER XIII PSALLIOTA CAMPESTRIS Introductory Remarks — Occurrence of the Fruit-bodies — Fairy Rings — External Appearance of a Fruit-body — Evolution of Ammonia from Dead Fruit-bodies — General Organisation of the Fruit-body for the Production and Liberation of Spores — The Radial Arrangement of the Gills — The Number of the GiUs — The Depth of the GiUs— The Thickness of the Gills— The Ends of the Gills— The Gill-chamber and the Annulus — Conditions for the Origin of Fruit-bodies — The Fate of Rudimentary Fruit-bodies — Effect of Dry Weather on Develop- ment— The Spore-discharge Period and the Number of Spores — The Mushroom and the Panaeolus Sub-type — Text-book Illustrations of the Hymenium — The Mottling of the Gill Surface — Methods for Examining the Hymenium in Surface View — The Number and Size of the Spores on Individual Basidia — Rate and Mode of Development of the Spores of an Individual Basidium — An Analysis of the Hymenium of the Cultivated Mushroom — The Hymenium of the Wild Mushroom — Camera-lucida Studies of the Young Hymenium — Camera-lucida Studies of a Nearly Exhausted Hymenium — Camera-lucida Studies of a Completely Exhausted Hymenium — Secotium agaricoides Introductory Remarks.— The present Chapter will treat of the production and liberation of spores in Psalliota campestris, the Common Mushroom (Fig. 129). The fruit-body of this fungus belongs to the Panaeolus Sub-type, and its hymenium is organised in all essentials exactly like that of Panaeolus campanulatus and Stropharia semiglobata. Since the Panaeolus Sub -type was described in great detail in connection with the two last-named species, it may seem superfluous that it should be dealt with any further. However, the Common Mushroom is so widely known all over the world as a familiar plant, is so celebrated as an article of food, and is so much employed in laboratory instruction, that it occupies .a unique place among the Hymenomycetes. Moreover its structure, particularly in regard to the hymenium, has been described hitherto but very imperfectly. For all these reasons I have thought it desirable that I should give as full an account of the organisation 360 PSALLIOTA CAMPESTRIS 361 of the Mushroom as possible, always bearing in mind that the one dominant function of the fruit-body is the production and liberation of spores. This account will necessarily involve a repetition of a number of statements which have been made in connection with other representatives of the Panaeolus Sub-type, as well as of observations scattered through the pages of Volume I. The reader who is chiefly interested in the systematic description of the Aequi-hymeniiferous and Inaequi-hymeniiferous Types and of their Sub-types is there- fore advised to pass over this Chapter and leave it to those who wish to increase their knowledge of a particular species. Among the new details of hymenial organisation which will be dealt with in the following pages may be mentioned : (1) the relation between the hymenium and the subhymenium, (2) the structure of monosporous and of bisporous basidia, (3) the arrangement of the spores on two adjacent bisporous basidia, and (4) the disappearance of exhausted basidia in late stages of hymenial activity. Also included in this Chapter will be found a new dis- cussion of the form of the pileus, the stipe, and the gill-system, which is of general application to the Agaricineae. In endeavouring to elucidate the organisation of the Aequi- hymeniiferous Type, my first attempts, made in 1911, were directed to the cultivated Mushroom. These partially failed owing to difficulties connected with the small size of the hymenial elements. I therefore turned my attention to Panaeolus campanulatus and Stropharia semiglobata, in which the basidia and paraphyses are relatively large ; and with these species my efforts, as we have seen in Chapters X and XI, were rewarded with a full measure of success. In 1916, I renewed my studies of the Mushroom ; and, in the light of a wide experience of hymenial organisation gained in the pre- vious five years, I at once perceived that the fruit-bodies of this fungus have an organisation similar to that of the Panaeolus Sub- type. The magnification of the microscope used in working out the finer details of structure in the gills of Panaeolus campanulatus and Stropharia semiglobata had been 440. Now the diameter of the basidia of Psalliota campestris is only about one-half that of the basidia of those two species. In investigating the hymenium of the Mushroom, therefore, it became necessary at least to double 362 RESEARCHES ON FUNGI the magnification. The actual magnification employed was 1,040 diameters. For all my investigations upon the Mushroom, I have used living material. Gills which have been fixed, stained, and cut with FIG. 129. — Psalliota campestris, the Common Mushroom. From an English meadow. Photographed by Somerville Hastings, £ natural size. the microtome, hinder rather than help one in perceiving the finer relations of the hymenial elements to one another. In such sections, the ripe and nearly ripe spores are always torn from their sterig- mata, thus leaving a number of basidia in a mutilated condition, and the basidia which still bear spores are relatively immature. Moreover, in a microtome section, it is often difficult or impossible to differentiate the hymenial elements from one another, owing to the fact that they were killed when being fixed and have therefore PSALLIOTA CAMPESTRIS 363 lost their turgidity. The microtome method of making preparations is, as everybody knows, of the highest value for certain kinds of research, especially for cytology ; but, for my purposes, it proved quite unsuitable. I am inclined to believe that, if the microtome had never been invented, the progress in our knowledge of the general organisation of the hymenium of the Hymenomycetes would have been much greater by now than it actually is. The first successful investigators of the hymenium, such as Leveille, Berkeley, and Schmitz, who flourished a century ago, studied their material in the living condition. My own work is in continuity with theirs. Occurrence of the Fruit-bodies. Fairy Rings.— The fruit- bodies of Psalliota campestris, the Common Mushroom, occur more especially in pastures and other grassy places, where the mycelium apparently feeds upon organic debris consisting of the subterranean organs of grasses and perhaps other herbs, which are present in the turf. They also sometimes come up on lawns, in well-manured gardens, and in beds rich in turf and rotted horse manure, such as are used for growing tomatoes and cucumbers. In addition, Mushrooms are grown on a very large scale on artificial beds made of stable manure. In pastures, they usually occur singly and with- out any regular order in respect to one another, but sometimes they occur in so-called fairy rings (Fig. 130). Such rings I saw near Banbury, England, about the middle of August, in a hay-field which had been mown only a few weeks before. One of the rings was 1 yard in diameter, two others 2 yards, and one 3 yards. There were also a number of incomplete rings. Each ring consisted of a circular dark-green zone of grass from 8 inches to 1 foot across. From this zone both centrifugally and centripetally the grass was of a normal lighter green colour. There were several fruit-bodies coming up in each ring (cf. Fig. 131). On making excavations, I found that the soil consisted of a stiff red loam and that, just beneath the dark-green zone but nowhere else, the mycelium, interlaced with roots and rhizomes, could be traced downwards for a distance of 3 inches (cf. Figs. 132 and 133). From our general knowledge of fairy rings, we may conclude that the mycelium had started its existence in the centre of the ring, that it had grown centrifugally outwards each year, 364 RESEARCHES ON FUNGI and that, in all probability, it had produced a series of successive annual crops of fruit-bodies above the dark-green zone. Bayliss observed that in one of the rings of Marasmius oreades, the well- known Fairy Ring Fungus, where the radius of the ring was already nearly four and a half feet, the radial increase of size of the ring through the growth of the mycelium was in four successive years FIG. 130. — Two fairy rings of Psalliota (= Agaricus) tabularis, containing large fruit-bodies about 15 cm. in diameter and of relatively uniform size. Photo- graphed at Akron, California, U.S.A., June 7, 1909, by H. L. Shantz and R. L. Piemeisel. Courtesy of the United States Department of Agriculture. 6, 9-5, 10-5, and 13-5 inches respectively.1 In other instances she found very similar results which showed that the tendency of each ring was to extend more and more rapidly as it gets older and as its radius increases. Probably the rings of Psalliota, cam- pestris behave very similarly. The ring which I found to have a diameter of 3 yards may well have been enlarging for six or seven years with accelerated speed. Owing to the rarity of fairy 1 Jessie S. Bayliss, " Observations on Marasmius oreades and Clitocybe gigantea as Parasitic Fungi," Journal of Economic Biology, 1911, vol. vi, p. 118. PSALLIOTA CAMPESTRIS 365 rings in connection with Psalliota campestris, at least so far as my own experience is concerned, it may be well perhaps to insist that the rings which I observed did as a matter of fact belong to this species and not to Psalliota arvensis. The fruit-bodies were quite typically those of the Common Mushroom, and not those of the Horse Mushroom. Shantz and Piemeisel x found numerous fairy rings formed by Psalliota campestris in grass-land near Yuma, California ; and, in one instance, as is shown in Fig. 131, they observed several Mushroom rings enclosed by a much larger ring formed by the mycelium of Calvatia cyathiformis. Psalliota arvensis sometimes also produces rings, and a good illustration of such a ring occurring in the United States is given by Atkinson.2 I have seen similar rings at Malvern, England, many yards in diameter and dotted with the unmistakable, huge, white-topped fruit-bodies of the species in question. I have also observed some beautiful fairy rings of Psalliota sylvicola, indicated by numerous fruit-bodies, in a wood near Haslemere in England ; and Shantz and Piemeisel 3 found huge rings in eastern California formed by the mycelium of Psalliota tabularis (Fig. 130). It therefore appears that, under suitable conditions, there is a tendency for the formation of fairy rings in several species of the genus Psalliota. There can be no doubt that the mycelium of the fairy rings of Psalliota campestris is perennial and that it may live on for a number of years producing fruit-bodies annually. In this respect the vegetative part of the fungus is analogous to the creeping rhizomes of many Gramineae, Liliaceae, etc., which also have a perennial existence and serve as a basis for the annual production of a new crop of aerial shoots, flowers, and seeds. In general, in a fairy ring occurring in a field there is usually an outer stimulated zone of green plants, a central more or less bare zone, and an inner somewhat wider stimulated zone, the stimula- tion making itself apparent in the increased height and greater greenness of the vegetation. According to Shantz and Piemeisel, 1 H. L. Shantz and R. L. Piemeisel, " Fungus Fairy Rings in Eastern Colorado and their Effect on Vegetation," Journ. of Agri. Research, vol. xi, 1917. 2 G. F. Atkinson, Studies of American Fungi, Mushrooms, edible, poisonous, etc., Ithaca, edition 2, 1901, Fig. 18. p. 20. 3 Shantz and Piemeisel. foe. cit. 366 RESEARCHES ON FUNGI the bare zone is simply due to the fact that in dry weather the green plants in the zone are killed owing to the mycelium in the ground o 2 o° o o x * •o *;.< .*" O O o o 0*0^, So ° ° FIG. 131. — Fairy rings formed by two Basidiomycetes in the short grass north-east of Yuma, California, U.S.A. The signs represent the positions of fruit-bodies. The large outer almost complete ring, which has a diameter of 65 metres, was formed by the mycelium of a Puff-ball, Calvatia cyathiformis. The clear discs on the outside of the ring represent the positions of fresh fruit-bodies and the black discs on the inside of the ring dry fruit-bodies of the previous year. Interrupting the continuity of the Puff-ball ring, and also within the area bounded by the Puff-ball ring, are numerous rings formed by the mycelium of Psalliota campestris — the Common Mushroom. Mapped by H. L. Shantz and R. L. Piemeisel. Courtesy of the United States Department of Agriculture. beneath cutting off the supply of water from their roots, and the stimulated zones are due to the mycelium increasing the avail- ability of the soil's nitrogenous contents. These observers have also shown that the subterranean mycelium has essentially the same PSALLIOTA CAMPESTRIS 367 effect upon wheat plants in an arable field as upon wild grass plants in a natural sod (cf. Figs. 132 and 133). In many well-grazed pastures in England, Psalliota campestris does not form fairy rings which can be clearly discerned, and the mushrooms occur either singly or in ill-defined associations. Such pastures, however, often produce mushrooms year after year. Under these conditions is the mycelium perennial, living on from year to year in the turf ? Or must we suppose that the pastures are annually re-infected from spores in such a way that each mycelium, in a single season, develops to maturity, produces a single FIG. 132. — A vertical section through a fairy ring caused by the mycelium of Psalliota (= Agaricus) tabularis, in grass in eastern Colorado, U.S.A. The mycelium, which has produced a fruit-body in the outer stimulated zone, is progressing from right to left. Drawn by H. L. Shantz and B. L. Piemeisel. Courtesy of the United States Department of Agriculture. crop of fruit-bodies, and then dies from exhaustion ? Doubtless, new infections of the turf take place occasionally. However, there is reason to suppose that a mycelium, when once it has been pro- duced, often lives on from year to year just as it does in fairy rings but without giving rise to a ring. Ring- formation is only made possible by the turf being so even in its quality that the mycelium can develop centrifugally in an equal manner along all radii. In many fields the turf is organically uneven, owing to the occurrence of weeds, irregular grazing by animals, uneven drainage, uneven deposits of excrement or manure, heterogeneous character of the soil, or to the presence in it of various organisms other than herbage plants. In such fields the mycelium of Psalliota campestris must penetrate through the turf in an irregular manner ; but there is no reason why this irregular penetration should not often be continued perennially just as it is in fairy rings. I am therefore 368 RESEARCHES ON FUNGI inclined to believe that in fairy-ring-less pastures, where Mush- rooms make their appearance year after year, much of the mycelium is perennial and capable of producing fruit-bodies annually. It seems to me that, although much of the turf in certain fields may be capable of supporting the mycelium of Psalliota campestris, and although spores may be deposited upon it in great numbers, yet new infections from such spores only take place very rarely ; and the same statement holds for Marasmius oreades and for other -/oo -o -so -/OO FIG. 133. — A vertical section through a fairy ring caused by the mycelium of Psalliota (= Agaricus) tabularis in a wheat field in 1915, a year of ample moisture supply. Colorado, U.S.A. Sketched by H. L. Shantz and B. L. Piemeisel. Courtesy of the United States Department of Agriculture. ring-forming fungi. This view is based on the following reasoning. A fairy ring is produced from a mycelium which grows radially outwards from a single point and progressively exhausts its sub- stratum. The older parts of the mycelium die progressively, death taking place in a radial direction. Thus a fairy ring, so far as the fungus is concerned, is a living zone of mycelium which develops centrifugally. This centrifugal growth continues often for a con- siderable number of years. Now the turf outside the ring is evi- dently extremely well suited to support the mycelium of the fungus, for otherwise it would not be progressively invaded. Yet each year the surface of this turf must receive a vast deposit of spores from the fruit-bodies occurring in the same pasture. The spores, although thus showered by the million on the top of ground per- fectly suited for infection, rarely cause infection ; for, if infection frequently took place, the turf outside the ring would quickly PSALLIOTA CAMPESTRIS 369 become evenly occupied by the mycelium and all further ring- formation would be rendered impossible. The inevitable conclu-. sion from this argument seems to be, therefore, that the spores of Psalliota campestris, Marasmius oreades, Clitocybe gigantea, and of other ring-forming fungi occurring in pastures, even when scattered in prodigious numbers over fields where the turf is capable of supporting the mycelium, rarely succeed in causing infection. The reasons for the enormous waste of spores under these conditions require further elucidation. External Appearance of a Fruit-body.— The following descrip- tion of the fruit-body of Psalliota campestris includes the field characters only. It has been compiled, in the light of my own observations, from the descriptions given by Fries, Berkeley, Greville, Stevenson, Massee, and Atkinson, with the object of com- bining fulness with accuracy. For illustrations of the fruit-bodies of the wild Mushroom, as found in pastures, the reader is referred to Figs. 129 and 134 (pp. 362, 377) in this Chapter and to Figs. 9 and 17 in Volume I (pp. 36 and 51). The pileus is 5-13 cm. broad, at first rounded but, as expan- sion takes place, its top becomes first convex and then flat. The surface is at first smooth, presenting a soft, silky appearance from numerous loose fibrils, but later it is sometimes torn into triangular scales, especially as the fruit-body becomes old. The colour is usually white, but varies more or less to light brown, especially in the scaly forms where the scales may be quite prominent and dark brown. Sometimes the colour is brownish before the scales appear. The flesh is firm, thick, white, often more or less stained with reddish-brown, especially when bruised. The pellicle is extended slightly beyond the margin of the pileus and can be easily separated from the subjacent flesh. The gills are free but rather close to the stipe, 0-6-1-6 cm. broad in the middle, of various lengths, closely set together so that the interlamellar spaces are narrow ; the longest gills are equally attenuated at both ends and ventricose. The colour of the gills in the button stage of the pileus is white, but at the time of expansion it is a beautiful pink. As the pileus becomes older and more and more expanded, the pink changes successively 370 RESEARCHES ON FUNGI to purple-brown, chocolate-brown, and finally to almost black. As soon as the pileus has expanded, the surface of each gill is seen to be mottled, owing to the fact that pigmented spores are present on the darker areas and unpigmented spores on the lighter areas. The mottling is somewhat fine in texture, but can easily be detected with the naked eye, and it persists for several days until the pileus is nearly exhausted. The edge of the gill is whitish and minutely denticulate. The stipe is 5-10 cm. long, 1 -2-2 cm. thick, bulbous when young, at maturity cylindrical but often tapering somewhat to the lower end ; sometimes it is thicker below than above or even sub-bulbous at the base. It is white. The flesh is solid except for the central core which is more or less stuffed with looser fibrils. The stipe is easily separated from the substance of the pileus. The velum is thin, white, silky, and very fragile. It is stretched horizontally as the pileus expands and finally torn, so that an annulus is formed about the middle of the stipe while fragments cling to the pileus margin. The annulus is more or less sheathed to the stipe, margin spreading or reflexed, torn, often deciduous, shrivelling as the fruit-body ages and thus becoming more and more inconspicuous. Sometimes it is in the form of a cortina. The spores are purple-brown, elliptical, 7-9 ^ long, breadth and depth equal, each about 5-6 /JL. A fruit-body, when fresh, has a weak but pleasant odour and an excellent flavour when cooked for food. As soon as the pileus has ceased to shed spores, i.e. about the sixth day after expansion, the stipe and pileus both collapse and become converted rapidly into a very dark putrescent mass. A few comments on the descriptions given by other authors will now be made. The colour of the gills is chiefly due to the cell-sap in all the cells. This sap is at first colourless but, as the pileus expands, it becomes pink. As the pileus flattens, the pink colour of the cell-sap changes to brown ; and, as the pileus ages, the brown colour of the cell-sap becomes progressively darker until it is almost black. Atkinson x states that the gills "soon become pink in colour and after the cap is expanded they quickly 1 G. F. Atkinson, loc. cit., p. 19. PSALLIOTA CAMPESTRIS 371 become purple-brown, dark brown, and nearly black from the large number of spores on their surface." Although it is true that a certain number of waste spores do accumulate on the gills during the period of spore-discharge, yet it is incorrect to attribute the deepening of the colour of the gills chiefly to their presence. The change in colour of the gills is due in the main to the gradual darkening of the cell-sap of all the gill-elements. We have the same phenomenon in such Coprini as Coprinus comatus and C. ster- quilinus. In these species, a pink cell-sap comes into existence shortly before the spores become produced or become pigmented ; and, as the pileus grows older, just prior to the beginning of spore- discharge, the pink sap changes to a very deep brown. In the case of Coprinus sterquilinus, I have been able to prove experimentally that the sap-pigment arises through the action of an oxidising enzyme ; and, doubtless, a similar biochemical cause is responsible for the colour changes in the gills of Psalliota campestris. The mottling of the gills, which can be readily made out by anyone who looks for it, was observed by Berkeley,1 but is not mentioned by Fries, Greville, Stevenson, Massee, or Atkinson. Fries 2 described the gills as " saepe liquescentes," Stevenson 3 as " often deliquescent," and Massee 4 as " subdeliquescent." The gills, it is true, are very soft ; but, while the spores are being shed, they remain firm and do not undergo deliquescence (autodigestion). In this they differ very markedly from the gills of the Coprini. However, as soon as they have become exhausted by discharging all their spores or have been killed by bruising or other injury, they quickly collapse and become reduced to a pulp, partly perhaps through the action of their own enzymes but chiefly owing to the attack of putrefactive bacteria. On the surface of a dead gill these bacteria multiply with extraordinary rapidity. The only trace of autodigestion in the living gill of a Mushroom is to be found in the collapse of the individual basidia shortly after each has shed its spores, and in the entire disappearance of many of the 1 M. J. Berkeley, The English Flora, vol. v, Part II, Fungi, 1836, p. 107 ; also cited in Massee's British Fungus-Flora, London, vol. i, 1892, p. 411. 2 E. Fries, Monographia Hymenomycetum Sueciae, Upsaliae, vol. i, 1857, p. 406. 3 J. Stevenson, Hymenomycetes Britannici, London, vol. i, 1886, p. 306. 4 G. Massee, British Fungus-Flora, London, vol. i, 1892, p. 411. 372 RESEARCHES ON FUNGI collapsed bodies of such basidia before the death of the gill as a whole. Further particulars in regard to this matter, however, will be given later on. Evolution of Ammonia from Dead Fruit-bodies. — I placed three pilei of the Common Mushroom on the bottom of a crystallising dish which was about 8 inches wide, covered the dish with a glass plate, and kept the whole on a laboratory table at ordinary room temperatures. The pilei in the course of a few days died, collapsed, and became subjected to the action of putrefactive bacteria, which reduced them to a pulp and caused an exudation from them of a dark brown fluid. As soon as putrefaction was in full operation, a large amount of free ammonia was emitted. This gas, which rapidly turned red litmus paper blue, could easily be detected by its odour which was very characteristic and pungent, and it gave the usual reactions in chemical tests for ammonia. This experiment has been repeated in successive years, and I now use it to demonstrate to students that free ammonia may be liberated into the atmosphere through the action of putrefactive bacteria. General Organisation of the Fruit-body for the Production and Liberation of Spores. — The hymenium or spore-producing layer is situated upon the under side of the pileus. This position is fraught with various advantages. The dominating one is that it favours the escape of the spores, for, when these have been shot a little distance from the basidia which have produced them, they fall into a free space beneath the pileus whence they can be carried away by the wind without meeting with any obstacle. The sub- pilear position of the hymenium also secures the protection of the basidia from falling rain, from too rapid loss of water by trans- piration during dry weather, and from the direct heating effect of the sun's rays.1 The hymenium is situated on the surface of several hundreds of lamellae which are packed closely together. Each of these structures is shaped much like the blade of a pen-knife with the cutting edge directed downwards. The production of the lamellae throws the hymenium into a series of folds and therefore serves 1 Of. vol. i, 1909, pp. 21-24. PSALLIOTA CAMPESTRIS 373 to increase very greatly, and yet very compactly and very economically, the area on the under side of the pileus available for the development of spores. In one fruit-body, which was 9-8 cm. in diameter, the gills were found to have an area of a little more than 20 • 04 times the area of the under surface of the pileus-flesh to which they were attached. The specific increase of hy menial surface due to the production of gills in this particular fruit-body was therefore 20-04.1 The stipe is an organ which serves to raise the pileus, and therefore also the hy menial layer upon the gills, to such a height that a considerable space comes into existence between the under side of the pileus and the surface of the ground. The presence of this space enables the wind or air currents to pass freely below the pileus and thus to carry away the falling spores. Since the spores, after emerging from the interlamellar spaces, fall in still air at the rate of about 1 mm. per second,2 and since the free space beneath the pileus is often 50 to 70 mm. in depth, it is clear that even a very feeble lateral current of air is sufficient not only to prevent the spores from settling directly beneath the pileus but also to carry them for long distances before they come to earth. Everyone knows from observation what long distances thistledown or the hairy seeds of the Willow may travel on a windy day. Yet, in still air, the rate of fall of the spores of the Mushroom is much slower than that of any hairy fruits or seeds.3 It is clear, there- fore, that when the wind is at all strong the spores of the Mush- room must often be carried many miles before they settle. If in a particular fruit-body we suppose the free space between the base of the gills and the ground to be 6 cm., and if we take the approximate average velocity of fall of the spores to be 1*2 mm. per second, then a simple calculation shows that the approximate length of time required for the spores to fall from the base of the gills to the ground in still air would be 50 seconds, i.e. nearly a minute.4 Now, 1 Cf. vol. i, pp. 27-31. The total area of all the gills in this fruit-body was 195-2 square inches, i.e. upwards of a square foot. 2 Vol. i, p. 180. 3 In still air, it was found that the fruits of the Thistle, Cnicus arvensis, fall 104 times as fast as the spores of a wild Mushroom. Vide Chap. I, p. 39. 4 Vol. i, p. 216. 374 RESEARCHES ON FUNGI if we suppose, as a matter of theory, that the surface of the ground is perfectly flat and that the wind is moving over its surface quite uniformly at a rate of only 5 miles per hour, then the spores, before coming to earth after leaving the base of the gills, would be carried a distance of not less than 365 feet. But, in nature, on windy days, it must often happen that a gust of wind, after sweeping beneath a pileus, becomes deflected upwards and bears innumerable spores to a great height into the atmosphere, in which case it may be some hours before many of them settle, during which time they would probably be carried for some scores of miles. The adjustments by which the hymenial surfaces on the gills are placed in the optimum position for spore-liberation in the Mushroom are no less than four in number, and may be summar- ised as follows : (1) turning the pileus into an erect position by an upward curvature of the stipe ; (2) raising the pileus several centi- metres above the ground by growth in length of the stipe ; (3) placing the gills with their long axes horizontal by an expansion of the pileus ; and (4) setting the median planes of the gills in vertical positions by the turning of the gills themselves about their lines of attachment to the pileus-flesh. The expansion of the pileus is due to causes inherent in the fruit-body itself, but the other movements are controlled by the external stimulus of gravity : the upward curvature of the stipe and its continued growth in a direction away from the earth's centre are phenomena of negative geotropism, whilst the movement of the gills about their lines of attachment is a phenomenon of positive geotropism.1 The flesh of both stipe and pileus is firm and massive. The fruit-body is thus endowed with considerable rigidity, so that it remains immovable even when the wind is blowing. This rigidity is necessary in order to keep the lamellae fixed in their nicely adjusted positions. If the fruit-body were to sway about with every breeze like grass stems, a vast number of spores would be prevented from escaping from the sides of the gills. The spores are shot out into the inter-lamellar spaces a horizontal distance of not more than about 0 • 1 mm.2 In still air, after arriving at this distance from the gill, the spores make a sharp turn and then fall vertically 1 Vol. i, pp. 50-56. 2 Vol. i, p. 142. PSALLIOTA CAMPESTRIS 375 downwards. If, owing to insufficient rigidity of the whole fruit- body, the lamellae were permitted to oscillate about their normal positions, or if the pileus were suddenly tilted, a very large number of spores after being shot away from their sterigmata would, on falling, touch the sides of the lamellae, adhere there, and never escape into the free air beneath the pileus. In an investigation upon some fruit-bodies with gills having a depth of 5 mm. the following facts were elucidated. When the median planes of the gills are tilted about their lines of attachment to the pileus to an angle of 1° 30' from the vertical, the spores can all escape down the interlamellar spaces. If the tilt be increased to 2° 30', the critical angle is reached : all the spores can still make their way out between the gills, but with any increase in the tilt some of them fall upon the hymenium and adhere there. With a tilt of 5° half the spores are lost, and with a tilt of 9° 30' four-fifths of them.1 These particulars serve to emphasise the necessity for consider- able rigidity in the fruit-body as a whole. The rigidity of the stipe is due in part to the turgidity, density of arrangement, and mode of adherence of the long strings of hyphae making up its substance, but in part also to the fact that the stipe as a whole tends to have the form of a hollow cylinder.2 In one variety of Mushroom which I cultivated in England, the core of the stipe was always distinctly hollow.3 In the wild Mushroom the core is either hollow or stuffed with loose fibrils. The firm and massive pileus-flesh, constructed of evenly-woven plectenchyma, is important from the mechanical point of view in that it supports the gills and keeps them fixed in a constant position during the several days of the spore-discharge period ; but it also has other functions : it serves to conduct nutriment to the gills during their development and constitutes a reservoir of water upon which the gills may draw according to their needs when they have attained maturity. The form of the fruit-body of the Mushroom, or indeed of that of any other typical terrestrial agaric such as Lepiota procera, from the mechanical point of view, appears to be, within limits, the most 1 Vol. i, pp. 39-40. 2 Vol. i, pp. 3 Vol. i, Fig. 17. A and B, p. 51, 376 RESEARCHES ON FUNGI efficient possible, if one considers the conditions necessitated by the one great function to be performed, namely, the production and discharge of millions of basidiospores. A structure which must spread widely in a horizontal direction, so as to provide as great an area as possible on its lower surface for the hymenium, must be held firmly at a distance of several inches above the surface of the ground, so as to bring into existence a free space below the hymenium through which the wind may sweep ; and the support- ing structure must be slender so as to economise fungus-substance as much as possible, so as to offer the least possible obstacle to the stream of escaping spores, and so as not to occupy a greater area of the valuable under-surface of the structure to be supported than is absolutely necessary. The structure to be supported is, of course, the pileus, and the supporting structure the stipe. There are various forms which one can think of as possible for both stipe and pileus, and various ways in which these bodies might be attached to one another. By mathematical calculations, the nature of which is well known to mechanical engineers, it can be shown that the fruit-body as a whole is most efficiently constructed, i.e. brings into play the smallest stresses and strains in its various parts, when : (1) the stipe has the form of a hollow cylinder, (2) the pileus has the form of a cone, (3) the axes of the stipe and of the pileus are vertical, and (4) the stipe is attached immediately under the centre of the lower side of the pileus. Since these characteristics are all found' more or less perfectly embodied in the Mushroom, we must conclude that the general shape of this fruit-body approximates to the ideal. In Lepiota procera, as shown in Figs. 14 and 15 in Volume I (pp. 44 and 45), the very large pileus remains in the form of a low upright cone during the whole period of spore-discharge. In Psalliota campestris the pileus, during expansion, also assumes the form of a low upright cone or plano-convex structure (Fig. 134, A), but subsequently its edges turn up so that finally the top of the pileus becomes flattened ; and then the cone is, as it were, turned upside down (Fig. 134, B). The turning up of the edge of the pileus causes the gills to be raised in such a way that their long axes are inclined upwards through a considerable angle. The disadvantage accruing PSALLIOTA CAMPESTRIS 377 from the gills being thus upwardly inclined instead of remaining with their long axes directed horizontally lies in the fact that the vertical distances between the gills through which the spores (especially those liberated near the tops of the gills) have to fall are thereby increased, and that with these increases there is FIG. 134. — Psalliota campestris, wild form obtained in a pasture at Banbury, Eng- land. Sections through fruit-bodies to show the appearance of the gills at different ages. In C, where the gill-chamber is as yet unruptured, the surface of the gills has an even appearance. In A, where the pileus is expanding, the surface of the gills is faintly mottled. In B, where the pileus is fully expanded, the gills are very distinctly mottled. Natural size. increased risk that the spores in passing down the very narrow interlamellar passages will touch the gill-sides and adhere to them ; but there is a distinct advantage gained, namely, that the gills as a whole, and therefore also the basidia, are raised an appreciable part of an inch. This extra elevation, obtained by sacrificing to a slight extent what we considered to be the ideal form of the pileus, permits of the free space beneath the gills being increased in depth, with the result that the wind has so much the better chance of carrying off the spores. 378 RESEARCHES ON FUNGI A flattening of the pileus, with a consequent raising of the gills, occurs toward the end of pileus expansion not only in the Common Mushroom, but in many other large Agaricineae. For Amanita rubescens it is shown photographically in Figs. 135 and 136. The Radial Arrangement of the Gills. — Granted that the FIG. 135. — Amanita rubescens. A fruit-body with a convex, still ex- panding pileus. The annulus, which formed the base of the gill- chamber, is now hanging downwards and hence is not an obstacle to the falling spores. Photographed at Four Oaks, Warwickshire, by J. E. Titley. About £ natural size. The pileus was 4 inches in diameter. lamellae are useful to the fruit-body in that they serve to increase the amount of surface available for the hymenium, yet we may still ask why they should be arranged in a radial manner. Why should the gills stretch along radii from the stipe to the periphery of the pileus and never form concentric bands ? I believe that the answer is as follows : (1) the radial arrangement involves a less complicated adjustment in securing that the lamellae shall look directly toward the earth than would the concentric one, and (2) the radial arrangement, with its radial interlamellar spaces, PSALLIOTA CAMPESTRIS 379 subjects the gills to a lesser and more equal strain during the expansion of the pileus than the concentric one. (1) As a pileus of a Mushroom opens out, its periphery moves away from the stipe and upwards about the stipe apex (Fig. 134). During this movement, the median planes of the lamellae remain FIG. 136. — -Amanita rubescen Th same fruit-body as that shown in Fig. 135, but one day older. The top of the pileus is now flat. The expansion of the pileus since the previous day has raised the gills and inclined them upwards, thus giving to the wind a better chance of carrying away the spores. Photographed at Four Oaks, War- wickshire, by J. E. Titley. About £ natural size. The pileus was 4 inches in diameter. vertical, so that the hymenium on their sides continues to look slightly downwards. An upward movement of the whole pileus- periphery does not therefore seriously affect the escape of the spores which are being discharged from the basidia. But let us suppose that the lamellae in the closed pileus were to arise on the base of the pileus-flesh in the form of concentric bands. Then, as the pileus expanded, the median gill-planes would constantly alter their inclination to the vertical, and they could not take up their final vertical positions until the pileus-flesh had ceased to move. 380 RESEARCHES ON FUNGI The result would be that the beginning of spore-discharge would need to be considerably delayed or, if spore-discharge were to take place during the expansion of the pileus, there would be a great waste of spores, for a large proportion of the discharged spores would strike the gill-sides, adhere there, and never be carried off by the wind. One perceives that, with a concentric arrangement of the gills, the adjustments for securing that the hymenium shall look more or less downwards would be necessarily more complex, take longer to complete, and allow of less elasticity than with the radial arrangement. (2) The peripheral boundary of the pileus of a Mushroom increases very greatly and rapidly as the pileus expands. If the gills, instead of having the radial arrangement, had the concentric one, the amount of stretching to which the gills would be subjected during the expansion of the pileus would be by no means equal, but would increase progressively and considerably from the inner- most gill surrounding the stipe to the outermost gill at the periphery of the pileus. During the rapid expansion of the pileus, the outer- most gills in particular would be subjected to great strain, and only by extraordinarily rapid growth could they respond to the demands made upon them by the rapid increase of the pileus-periphery. During the expansion of the pileus, the innermost concentric gills would require to lengthen their circumferences only very slightly, but the outermost ones would require to lengthen theirs relatively enormously. I cannot help thinking that the concentric arrange- ment of the gills, which would involve such an inequality in the development of the gills, would result in inefficiency in the produc- tion of spores, and that the radial arrangement, where all the gills are equally stretched in the radial direction, is on this account much to be preferred. For the reasons given above we may conclude that the radial arrangement of the lamellae, as we actually have it in a Mushroom, is most beautifully adapted to secure the highest efficiency in the production and liberation of the spores. The radial arrangement of the gills comes into existence during the development of the pileus in the following manner. The first gills to make their appearance in the very rudimentary pileus are PSALLIOTA CAMPESTRIS 381 the long ones. As the pileus grows in size and the spaces at the pileus-periphery between the first-formed gills become wider, new and shorter gills are interpolated between the older ones ; and this process of gill-interpolation goes on continuously as spaces become provided until all the gills have been formed. The very shortest gills are always the last ones to appear. The determining factor in the origin of a new gill seems to be the divergence beyond a cer- tain degree of two older gills. As soon as an interlamellar space has attained a certain width, a new gill-rudiment arises midway between them. It is owing to this mode of development of the gills in general that the wonderful system of lamellae on the under side of a mature Mushroom comes to be established ; and, by comprehending it, we can understand why it is that two gills never touch one another and why the interlamellar spaces are never reduced beyond a certain minimum and never exceed a certain maximum. The law of gill-development secures that the under side of the Mushroom shall be as crowded with lamellae as is consistent with the efficient discharge of the spores. The Number of the Gills. — The number of gills in any gill- system varies from about 300 in small mushrooms to over 600 in large ones. It is dependent to a great extent upon the diameter of the pileus. This clearly follows from the consideration that as two long gills pass from the stipe to the periphery they diverge progressively more and more, and therefore leave room for the interpolation of more and more shorter gills. A certain variety of the Common Mushroom was cultivated artificially on a bed of horse manure. A small mushroom which came up had a diameter in the fully expanded condition of 4-5 cm. It possessed in all about 320 gills : 120 long ones, 120 intermediate ones, and 80 very short ones. The spaces between the long and intermediate gills numbered 240, but they were so narrow that only one-third of them were divided by very short gills. The gill-system in this particular mushroom, therefore, consisted for the most part of gills of two lengths only. Four other larger mushrooms, each of which was approximately 8 cm. in diameter, possessed the following respective numbers of gills at the pileus-periphery : 509, 586, 616, and 628, i.e. the gills of these 382 RESEARCHES ON FUNGI larger mushrooms were twice, or somewhat more than twice, as numerous as those in the very small mushroom. The larger mush- rooms had a gill-system which was everywhere made up of gills of at least three lengths : long, short, and shorter, while between the two last types of gills very short gills were frequently interpolated. Other factors, besides the diameter of the pileus, which affect the number of the gills, are : the diameter of the stipe and the thickness of the gills where they adjoin the pileus-flesh. The thickness of the gills depends 011 their depth, for the deeper a gill is, since it is wedge-shaped in cross-section, the thicker it must be where it is attached to the pileus. With a given diameter for a pileus, therefore, the deeper the gills are, the greater space will each occupy below the pileus and the fewer there can be of them. I carefully compared two sets of fruit-bodies : (1) wild mushrooms gathered from a field near Birmingham, England, and (2) culti- vated mushrooms of a particular variety grown artificially on a mushroom-bed. I found that, with fruit-bodies of the same diameter, the wild mushrooms had deeper and thicker but less numerous gills than the cultivated ones. The depth of the middle of the long gills in the wild mushrooms was 9 mm., whereas the corresponding depth in the cultivated mushrooms was only 6 mm. A wild mushroom with a diameter of about 8 cm. possessed about 413 gills and a cultivated one with an equal diameter about 600, the count in each case having included all the very smallest gills which could be clearly discerned at the pileus-periphery. It became obvious from this comparison and from others that, not merely theoretically but also actually, there is a relation between gill-depth and gill-number. The close packing of the gills in one of the cultivated mush- rooms is obvious from a glance at Fig. 137. This particular pileus was 6 cm. in diameter, but it was not yet quite fully ex- panded. It might perhaps have increased its diameter to 6 • 5 cm. if it had been left to continue its development upon the mush- room-bed. By throwing a photograph of this pileus on a screen from a lantern-slide, it became possible to count the gills at the stipe and at the pileus-periphery very exactly. The number of gills at the stipe was 132 and the number at the pileus-periphery PSALLIOTA CAMPESTRIS 383 639. A wild mushroom which measured 9-2 cm. in diameter in the fully expanded condition and which was therefore consider- ably larger than the cultivated mushroom just considered, had almost exactly the same number of gills, namely, 134 at the stipe and 644 at the pileus-periphery. This comparison points to the conclusion that on equal spaces beneath the pileus the culti- vated mushrooms tended to produce the larger number of gills. Let us suppose that a pileus is endowed with perfect radial symmetry and on this supposition investigate the nature of its gill- system. Let the num- ber of long gills, i.e. those which pass from the stipe to the pileus- FIG. 137. — Psalliota campestris, cultivated form. The pileus of a fruit-body photographed from below immediately after its removal from mushroom bed. Notwithstanding the close packing of the gill-system, there are inter- lamellar spaces, into which the spores can be shot, between every pair of adjacent lamellae. Natural size. periphery, be equal to x. Then, at a little distance from the stipe in the radial direction, owing to the gradual divergence of the long gills and the provision of the necessary interlamellar spaces, we should find a second set of gills, all of exactly the same length, regularly interpolated, one in each space or a total of x. The number of interlamellar spaces is now equal to 2x. At a farther distance from the stipe, we should find a third set of gills interpolated, all of exactly equal length, one in each space or a total of 2x. These by their presence would increase the number of spaces to 4x. Still farther from the stipe we should find a fourth series of gills interpolated, all of exactly equal length, one in each space or a total of 4x. If, still farther toward the pileus-periphery, a 384 RESEARCHES ON FUNGI fifth set of gills were present, they would number 8# and be inter- polated in as many spaces. The total number of gills on the whole pileus would be the sum of x + x + 2x + 4z + 8x or, for n sets of gills, x -f- x + 2x + 4# + Sx to n terms, the sum of which is equal to x X 2n~l. If a perfectly symmetrical mushroom had 132 long gills and three distinct sets of gills, the total number of its gills would be 132 + 132 + 264 = 528 and, if it had four sets of gills, the total number of gills would be 132 + 132 + 264 + 528 = 1,056, so that it is evident that every additional set of gills produced at the periphery of the pileus doubles the previous total. In nature, owing to the large number of factors which affect the growth of the cells making up the fruit-body, such perfect symmetry is never attained but only approximations thereto. The gills of each set vary considerably in length about a mean, and these variations become increasingly greater in each set as one proceeds from the stipe centrifugally. If one examines the under side of a pileus, one finds that the gills are distinctly more crowded at the periphery than in the region of the stipe and a little way from it. It so happens in the pileus shown in Fig. 137 that the circumference of the pileus, where the gills end, is just four times the circumference of the stipe. There are 639 gills at the margin of the pileus. If the crowding of the gills were the same in the region of the stipe as at the pileus-periphery, then there ought to be 159 gills at the stipe ; but, as a matter of fact, there are only 132. The extra crowding of the gills at the pileus- periphery is due to the fact that the gills in approaching their outer extremities become shallower and shallower (cf. Fig. 134, p. 377), and in accordance therewith thinner and thinner, so that each one occupies less and less space where it is attached to the pileus, much PSALLIOTA CAMPESTRIS 385 less indeed than do the longer gills at their mid-lengths where they are deepest and thickest. The inter lamellar spaces, as one can see from a glance at Fig. 137 (p. 383), are actually wider at the bottoms of the gills half-way between the stipe and pileus-periphery than at the periphery itself ; but at the tops of the gills where these adjoin the flesh they are all of approximately the same width. The Depth of the Gills. — Let us now enquire why it is that the gills of a mushroom are so organised that they have a certain depth which varies but little about a mean. Why should not the gills be twice or three times as deep, or only one-half or one-third as deep ? What factors have led to the evolution of a depth such as we are now able to observe ? I am inclined to believe that the present gill-depth of Psalliota campestris has been gradually attained as a result of variations which have tended toward the production of the greatest amount of gill-surface with a given expenditure of mushroom substance. The considerations on which this view is based will now be elaborated. Let us imagine a large wedge made of steel with the width and depth shown in the cross-section illustrated in Fig. 138 at ABC. Now let us suppose that we replace this wedge by two others which resemble the first in form but are only one-half as deep and wide (ADE + EFB). It is evident that the area of the lateral surface of the two new wedges taken together is the same as that of the first wedge but that their total volume is only one-half. Let us replace the two wedges by four others which resemble the two in form but are only one-half as deep and wide (AGH + HIE + EJK + KLB). The area of the lateral surface of the four wedges taken together is still equal to that of the original wedge, but the total volume has been reduced to a quarter. If we again replace the four wedges by eight, we shall find that, while once more there is no reduction of the original lateral surface, yet the volume is reduced to one-eighth. This process of substitution could be continued theoretically indefinitely, each substitution involving a reduction of the original volume to one-half but no alteration of the total lateral surface. Now let us suppose that it is desired to have the wedges of such a depth that with the least expenditure of material the surface of the lateral sides shall be collectively as VOL. n. 2 o 386 RESEARCHES ON FUNGI great as possible. Evidently, with the form of the wedge given, the area of the lateral sides cannot be increased, but the volume can be diminished by reducing the wedges in depth and increasing their H K B FIG. 138. — Diagrams to illustrate the relation of gill-depth to gill-surface and gill- volume. Left, a steel wedge in cross-section, with replacement by two, four, or eight wedges. Bight, a gill in cross-section, with replacement by two, four, or eight gills. The eight small gills have much less surface than the eight small wedges. number. One would therefore make the wedges as small as possible. The limit would be determined by human skill : it would be found that wedges could not be made less than a certain depth and yet retain their true symmetry. Instead of a wedge of steel let us now consider a gill (Fig. 138, A1TC'). A gill is wedge-shaped, but differs from the steel wedges already discussed in that its sides are slightly convex and in that it PSALLIOTA CAMPESTRIS 387 is rounded both at the free edge and on each side where it adjoins its two neighbouring gills. Let A'B'C' be the shape of a cross- section of a gill taken where the gill is deepest. Now let us replace this gill by two others which resemble the first in form and are of equal size (A'D'E' + ET'B'). The two gills wiU not meet at an acute angle, as did the two steel wedges, but the interlamellar space between them must be rounded as shown, so as to permit of the violent discharge of the spores for a distance of 0 • 1-0 • 2 mm. from the hymenium even at the narrowest part of the space, and so as to allow in addition for a margin of safety for the adjustment of the gill-planes in vertical directions even when the pileus happens to be slightly tilted (cf. Fig. 139, p. 390). Let us suppose that the radius of the rounded top of each interlamellar space, for the reasons given, is not to be reduced beyond 0 • 25 mm. Then we can speak of the narrowest portion of the interlamellar space as having a minimum width of 0 • 5 mm. Let us now replace the two gills by four. The new interlamellar spaces must be as wide as the old ones, i.e. 0-5 mm. wide at the top. Again, let us replace the four gills by eight. Here again, the seven interlamellar spaces all have a minimum width at the top of 0 • 5 mm. We thus see that, as the gills are progressively diminished in size, the minimum width of the interlamellar spaces remains constant. The gills, as they become more and more numerous and shallower, tend to be less and less like the original gill and more and more like rounded ridges, their exterior reminding one of the waves on the surface of water. If the process of sub- stitution were to be continued, the gills would eventually disappear and the hymenium become perfectly flat. It is evident, therefore, that, unlike what was found to happen with the steel wedges, reduction of gill-depth with increase in gill-number leads to a very considerable reduction of the absolute amount of gill-surface. The greatest amount of gill-surface is possessed by the original gill A'B'C', a slightly less amount by the two gills A'D'E' + ET'B', and progressively less amounts with every increase of gill-number until the gill-substance has been reduced to zero. We may regard two tendencies as having had effect in leading to the establishment of the particular gill-depth which we find in Psattiota campestris at the present day : (1) a tendency to increase the absolute amount 388 RESEARCHES ON FUNGI of hymenial surface, and (2) a tendency to economise gill-substance. Now these tendencies are antagonistic, and the gills, as they exist to-day, embody within them a compromise between the two. If the gills were four times as deep as they actually are, more than four times as much substance would need to be put into each of these deeper gills than was present in the shallower gills which each deeper one would replace, while the actual increase of hymenial surface would be very slight. The absolute amount of increase of fruit-body substance with this increased depth of the gills would be relatively considerable, for the volume and weight of the whole gill-system would be increased by about four times. If the gills were only one-quarter as deep as they actually are, somewhat less than one-quarter of the substance present in one of the original gills would be put into the shallower gills which would replace it, but at the same time the decrease in the area of the hymenial surface would be considerable. The absolute diminution of fruit-body substance would be relatively inconsiderable and it would be pur- chased at the cost of a marked decrease in hymenial area. Owing to the fact that in the course of evolution a certain percentage of fruit-body substance in Psalliota campestris has been converted into deep and narrow gills, the hymenial surface has been increased about twenty times.1 It seems to me, however, very probable that a further increase in the percentage of fruit-body substance put into gills would not be advantageous, owing to the fact, suffi- ciently indicated above, that the deeper the gills become the less the gain in surface relatively to the expenditure of gill substance.2 The Thickness of the Gills.— The amount of hymenial surface on the under side of a pileus is decided not merely by the depth of the gills but also by their thickness. With a constant depth, the thinner the gills, the more numerous can they be and the greater will be the hymenial area on their exterior collectively. As if with the object of producing as extended an hymenium as possible, the 1 Vol. i, 1909, pp. 30-31. 2 The geometrical principles relating to hymenial surface and gill-depth, which have been dealt with above, have been illustrated mathematically in an interesting manner by my colleague, Dr. C. D. Mailer ; but the mathematics involved would be considered by most mycologists as very abstruse. I believe that Dr. Miller intends to send his paper to one of the mathematical periodicals. PSALLIOTA CAMPESTRIS 389 gills of Psalliota campestris are extremely thin, and it is doubtful if they could be appreciably thinner and yet continue to facilitate the free escape of the spores. A Mushroom was gathered from a field near Birmingham, England, and some tangential sections were cut through the pileus so as to pass vertically downwards through the region of the long gills at their deepest part, which happened to be 7 mm. deep. In cutting the sections it was impossible to avoid bending the gills somewhat, so that their median planes came to be no longer straight. Two of these sections, which were of course still living, were very carefully sketched with a magnification of sixty with the help of a camera lucida. Fig. 139 shows these camera-lucida drawings after they had been adjusted to the extent of straightening the gills so as to give these the alignment which they had in nature. With the aid of this Figure let us now con- sider the thickness of a typical long gill. The thinnest part of the gill is at its free edge. Just above this edge the actual thickness is reduced to 0- 15 mm., and it could scarcely be thinner, owing to the fact that the gill-substance is bound to include two hymenial layers, two sub hymenial layers, and at least some tramal hyphae. A quarter of the way up the gill, the thickness becomes increased to O1 25 mm., half way up to 0' 3 mm., and just below the interlamellar sinuses at the top to 0 • 43 mm. The two sides of the gill converge from above downwards at a very acute angle which increases in size from above downwards. This angle in the upper part of the gill is only about 1°, but toward the middle of the gill it becomes increased to about 2°, whilst in the lower part of the gill it becomes 4°. The convergence of the two sides of every gill is of the greatest importance, for it is owing to its existence that every part of the hymenium on the under side of a normally adjusted pileus comes to look more or less downwards toward the earth, with the result that every small portion of the hymenium (every square mm.) can produce and liberate spores simultaneously during the whole period of spore-discharge. In Marasmius oreades, many Hygrophori, Russulae, etc., the gills are much thicker relatively to their depth than in Psalliota campestris, or, to state the same thing in another way : the angle of convergence of the sides of the gills is considerably greater in 390 RESEARCHES ON FUNGI B /7\ s 3 FIG. 139. — Psalliota campestris (wild form). Two cross-sections of gills. To show exactly the angle of convergence of the opposite sides of each gill and the shape of the interlamellar spaces as they are under field conditions. In cutting the sections of the freshly-gathered living mushroom with a hand-razor, each gill became unavoidably slightly curved to one side. The sections were mounted in water. Camera-lucida drawings were then made which were subsequently very carefully adjusted so as to make each gill look vertically downwards as shown above. h, the hymenium. The trajectories or sporabolas of a few spores, as they are in still air, are indicated by the arrows. The gills were originally 7 mm. deep. The scale enables one to measure their thickness at various depths. PSALLIOTA CAMPESTRIS 391 these fungi than in the Common Mushroom. On the other hand, there are some Agaricineae which have thinner gills than" Psalliota campestris, namely, the Coprini. In these, however, the gills are parallel-sided or subparallel-sided, and associated with this parallel- sidedness we find a special arrangement for the production and liberation of spores which is quite different from that in the Mush- room. If a Mushroom had parallel-sided gills with a thickness say of 0*2 mm., and its hymenium continued to be active everywhere at one and the same time, then, unless the gills could be kept practi- cally quite vertical during the whole period of spore-discharge, a great many spores would be wasted. Suppose that a gill of a Mushroom were parallel-sided and 7 mm. deep and that the spores were shot out from the gill-sides to an average distance of 0* 1 mm. Then, if the median planes of the gills were to diverge from the vertical by an angle of only 0*8°, which represents a slope just exceeding one in seventy, most of the spores on the upward-looking side of the gill would fail to escape from the pileus. It is evident that the margin of safety necessary for compensating for irregu- larities in development in such a case would be too small. In Mush- rooms, as we actually find them in the field, the lower part of the gills can be bent from the vertical by an angle exceeding 2° and still all the spores can escape. In a former investigation, of which an account was given in Volume I, the critical angle of tilt for the gills of a particular Mushroom was found to be 2'S0.1 The discussion just given may be taken as supporting the view that in Psalliota campestris the gills, provided they are to continue to liberate their spores successfully, have probably become reduced to their minimum thickness. The Ends of the Gills. — We have now to account for the fact that the gills, when observed in lateral view (Fig. 134, p. 377), are deepest toward the middle and become narrowed down to a vanishing point at each end. The narrowing of the gills toward the pileus-periphery, from the mechanical point of view, has the same significance as the corre- sponding thinning out of the pileus-flesh, i.e. it is in accordance with the mechanical requirements of the fruit-body as a whole. From 1 Vol. i, p 40. 392 RESEARCHES ON FUNGI the developmental and nutritional points of view, it also seems fitting that the gills should become shallower and shallower as the pileus-flesh immediately above them becomes progressively diminished in thickness beyond a certain point. The narrowing of the gills in the neighbourhood of the stipe has an entirely different significance from the narrowing at the pileus- periphery. The gills are cut away from the stipe in such a way as to permit of their being raised through an angle of about 90° during the rapid expansion of the pileus, without their being subjected to strains which would tear them. The question of the relation of the gills to the stipe in Agaricineae in general, however, will be dealt with more fully in Volume IV. The Gill-chamber and the Annulus. — The gills of a Mushroom are protected during their development and during the first stages of the expansion of the pileus by being enclosed in a gill-chamber. The wall of this chamber at the top and sides consists of the pileus- flesh and at the bottom of the velum partiale (Fig. 134, C, p. 377 ; also Fig. 140, D, F). As the pileus expands, the floor of the gill- chamber becomes drawn out horizontally into an extended but very thin and fragile membrane which, with further stretching, breaks, so that fragments of the velum are carried away upon the edge of the pileus (Fig. 137, p. 383) and an annulus is left surrounding the stipe (Fig. 134 ; also Fig. 140, A, B, H, I). The base of the thin membrane which closes the gill-chamber during the earlier stages of expansion, is, at its origin, a tubular structure ensheathing and apparently forming part of the outside of the stipe enclosed within the chamber. This sheath is pulled away from the stipe from below upwards during the expansion of the pileus (Fig. 140, B, C, E, F). The region of the stipe from which the sheath has been pulled can be distinguished in some field Mushrooms by its roughened surface, which indicates where cleavage took place : to its upper edge the annulus is attached, while at its lower edge there is not infrequently a trace of a second ring (Fig. 140, A). The origin of the roughened region and of the two annuli may best be realised by a reference to the semi-diagrammatic drawing given in Fig. 140, B, where the arrows indicate the movements which have taken place. Fig. 140, E, shows the exterior of a young Mushroom in which the PSALLIOTA CAMPESTRIS 393 floor of the gill-chamber is actually in course of formation from the sheath, and in which an inferior annulus can be clearly discerned. FIG. 140. — Psalliota campestris. From a field at Redditch, England. To show the occasional formation of an inferior annulus in addition to the superior annulus. A, a mature fruit-body, shedding spores, having a superior annulus s, a slight inferior annulus i, and a roughened area r. B, a diagrammatic vertical section to indicate the mode of origin of a double annulus : the velum has moved in the direction shown by the arrows from its position, o, and has left a slight inferior ring, i, behind. C, a very young fruit-body with the gills filling the gill-chamber ; the velum is about to be pulled outwards away from the stipe. D, the under surface of a young fruit-body with the velum, v, closing the gill- chamber ; st, the stipe. E, a young fruit-body which has just formed an inferior annulus, i, but in which the velum, v, still closes the gill-chamber ; r, a roughened area formed by the pulling-away of the velum. F, a vertical section through a young fruit-body showing the gill-chamber closed by the velum which is being pulled away from the stipe leaving a roughened area, r. G, an external view of a young fruit-body. H, a vertical section of G, showing the velum forming the annulus ; r, a roughened area like that of F. I, the under surface of the pileus of G and H, showing the gills, g, in the gill-chamber through a rent in the velum v ; st, the stipe, f natural size. 394 RESEARCHES ON FUNGI The existence of a gill-chamber causes the gills, during their de- velopment, to be subjected to more even conditions of temperature and moisture than would be possible in its absence. The chamber floor also bars out from access to the gills — the most nutritious and vulnerable parts of the whole fruit-body — such small marauding animals as Fungus Gnats (Mycetophilidae), Springtails (Collembola), and Mites (Arachnida) ; and it also keeps away Bacteria and the spores of parasitic Fungi. The development of a gill-chamber which persists until spore-discharge is about to begin, i.e. until the last possible moment when it can be of service, is a beautiful con- trivance which was probably added in a late stage of the evolution of the Mushroom. The annulus, owing to its delicacy, shrivels up soon after its definite formation and so comes to offer but a very slight obstacle to the removal of the falling spores by the wind. Conditions for the Origin of Fruit-bodies. — The conditions under which the fruit-bodies of Psalliota campestris come into existence in the first place are not as yet by any means fully understood. We know that the mycelium vegetates in the substratum of turf or rotted horse manure for some weeks or months and that it then rapidly produces mushrooms. The mushrooms never arise at any depth in the substratum, but always at its surface. Two questions in this connection await an answer : (1) what internal changes cause the mycelium to proceed from the vegetative to the repro- ductive stage, and (2) what external stimulus decides that the fruit-bodies, when formed, shall arise at the surface of the sub- stratum and nowhere else ? In the absence of exact knowledge our answers to these questions can only be speculative. The recent investigations of Mile Bensaude, Hans Kniep, and Miss Mounce have proved conclusively that in the Hymenomycetes, as in the Phycomycetes, some species are homothallic and others heterothallic.1 In a homothallic species, e.g. Coprinus sterquilinus, fruit-bodies are produced rapidly and perfectly upon a mycelium of monosporous origin. On the other hand, in a heterothallic species, e.g. Coprinus fimetarius, Schizophyllum commune, and Coprinus lagopu-s, this is usually not the case. Thus, according 1 For a detailed treatment of the problem of sex in Hymenomycetes vide infra, vol. iv, Chapters on Sex. PSALLIOTA CAMPESTRIS 395 to Mile Bensaude,1 monosporous cultures of Coprinus fimetarius are quite infertile ; according to Hans Kniep,2 monosporous cultures of SchizopJiyllum commune are usually, although not always, in- fertile ; and according to Miss Mounce,3 in Coprinus lagopus, while fruit-bodies arise on monosporous mycelia, such fruit-bodies often develop very few spores or even no spores at all. In all these heterothallic species, however, fruit-bodies are developed rapidly and perfectly upon secondary mycelia, i.e. mycelia derived from the union of two monosporous mycelia presumably of opposite sex. It thus appears that, in the heterothallic Hymenomycetes, the production of fruit-bodies is largely dependent upon the occurrence of a sexual act taking place between two mycelia presumably of opposite sex. We may therefore ask : Is Psalliota campestris homothallic or heterothallic ? If it is homothallic, then, doubt- less, its fruit-bodies can be produced on mycelia of monosporous origin. If, however, it is heterothallic, then probably one of the conditions for the normal production of its fruit-bodies is the union of two mycelia of opposite sex. No one has attempted to obtain fruit-bodies from Mushroom spawn derived from a single Mush- room spore ; so that, for the present, in respect to Psalliota cam- pestris, we lack knowledge which is vital for a full understanding of the conditions governing fruit-body formation. I took some mycelium from a Mushroom bed and examined its hyphae with the microscope. I did not find any clamp-connections either in the independent hyphae or in those making up the mycelial cords. Mile Bensaude 4 and Hans Kniep 5 have shown that the formation of clamp-connections in Coprinus, etc., stands in intimate relation with the conjugate division of a pair of nuclei, so that 1 Mathilde Bensaude, Recherches sur le cycle evolutif et la sexualite chez les Basidiomyc&tes, Nemours, 1918, pp. 1-156. 2 Hans Kniep, " Uber morphologische und physiologische Geschlechtsdifferen- zierung," Verhandl. der Physikal.-med. Gesellschaft zu Wurzburg, 1919, pp. 8-16. 3 Irene Mounce, ( 1 ) " Homothallism and the Production of Fruit-bodies by Mono- sporous Mycelia in the Genus Coprinus," Trans. Brit. Myc. Soc., vol. vii, 1921, pp. 198-217 ; (2) " Homothallism and Heterothallism in the Genus Coprinus," ibid., 1922, pp. 256-269. 4 M. Bensaude, loc. cit. 5 H. Kniep, " Beitrage zur Kenntnis der Hymenomyceten, III und IV," Zeitschr. /. Botanik, Bd. VII, 1915, and Bd. VIII, 1916. 396 RESEARCHES ON FUNGI clamp-connections are a sign that the nuclei in the hyphae are paired. Max Hirmer1 has observed that the cells of the stipe and pileus-flesh of Psalliota campestris contain not pairs of nuclei, but groups of individual nuclei numbering up to eleven in each group. The absence of clamp-connections in the mycelium of a Mushroom bed is therefore probably due to the fact that the nuclei are not paired but exist in little groups. In many Hymenomycetes, the occurrence of the nuclei in pairs in the mycelium and the formation of clamp-connections appear to be prerequisites for fruit-body formation, but this is not so for Psalliota campestris. The reduction of the nuclear groups in the cells of a fruit-body of the Mushroom takes place, according to Hirmer, in the gills, and is accomplished by the time the subhymenium is formed.2 This allows of a single pair of nuclei entering each basidium. Here nuclear fusion takes place in the usual manner. There are doubtless other conditions than that of the correct arrangement of the nuclei, which are necessary for the production of fruit-bodies by a mycelium. It appears that a mycelium never begins to form fruit-bodies until its hyphae have accumulated protoplasm laden with food materials, sufficient in amount to ensure that any reproductive effort shall not utterly fail through lack of constructive materials. Perhaps, therefore, the mycelium is subjected to an internal stimulus of such a kind that each hypha has a certain effect on the whole mycelium in that it stimulates it to begin the process of fruit-body formation ; and, perhaps, a reproductive effort on the part of the mycelium as a whole is only begun when the number of hyphae has become so large that the cumulative stimulus provided by all the hyphae exceeds a certain limit. Let us suppose that the older part of a mycelium which has been vegetating for some weeks in the turf of a pasture has received a stimulus of such strength that the reaction must be a reproductive effort. We may take it that it is impossible -for the mycelium to produce fruit-bodies at once owing to the fact that the external 1 Max Hirmer, " Zur Kenntnis der Vielkernigkeit der Autobasidiomyzeten, I," Zeitschriftfiir Botanik, Jahrg. XII, 1920, pp. 668-669. 2 Ibid. PSALLIOTA CAMPESTRIS 397 conditions beneath the surface of the soil are not favourable thereto. The first reaction of the subterranean mycelium appears to be the formation of a network of cord-like strands. Probably these strands serve as a reservoir for the protoplasm and other materials which are to be put into the fruit-bodies ; and doubtless they are produced at the expense of substances supplied by the distant and actively-growing hyphae which are still attacking the substratum and extracting from it its nutriment. Certain of these strands, during their formation, grow upwards until they reach the surface of the soil and, as we know from observation upon artificial mushroom-beds made of soil-covered horse manure, can penetrate through an inch or more of non-nutritious compact soil or even through sand. This upward growth of mycelial strands absorbs both material and 'energy, but it is absolutely necessary in order that the reproductive bodies may be formed at the surface of the ground, where alone the spores may be liberated so as to secure their dispersal by the wind. One may ask whether or not the upward direction of growth of some of the strands is due to negative geotropism. Now, on artificial beds having various shapes with surfaces at the sides and bottom, mushrooms may arise anywhere on the surface (Vol. I, Fig. 17, p. 51). I am there- fore inclined to believe that the strands grow upwards to the surface of the soil in fields not as a response to the stimulus of gravity but for some other reason. It is not improbable that the upward growth is purely a matter of chance. The mycelium, owing to its wonderful power of branching and to the apical growth of its hyphae, doubtless tends to penetrate through the interstices of the soil in all directions of space. This being so, some of the hyphae must make their way upwards. The strands are then formed about these hyphae. After being formed the strands immediately proceed to the formation of rudimentary fruit-bodies. Why should the rudimentary fruit-bodies never be formed any- where else than at the surface of the soil ? We can only suppose that this position is decided by the reaction of the mycelium to one or more external stimuli or by the absence of mechanical pressure. The stimuli most likely to be involved seem to be light, moisture, or atmospheric gases. Now light is unnecessary for the 398 RESEARCHES ON FUNGI formation of fruit-bodies ; for, as every mushroom-grower knows, mushrooms can be reared in perfectly dark cellars and caves where not a ray of light enters. When the mycelial strands approach the surface of the soil, they are liable to lose considerable quantities of water by transpiration, for the free air above the soil is usually much less moist than the air in the soil. This increased transpira- tion may well be one of the conditions favouring fruit-body forma- tion. The production of fruit-bodies at the surface of the soil may be due in part to a stimulus provided by the gaseous environ- ment. The soil must be considerably richer in carbon dioxide and other gases produced by the decomposition of organic matter, and at the same time poorer in oxygen, than the free air above it. Another point of difference between the gases in the soil and those of the free air is that the latter are in constant motion owing to air currents and the wind, while the former are relatively still. If the change of gaseous environment should be shown by future experiment to be the cause of the formation of fruit-bodies at the surface of the soil, its exact mode of operation will require detailed analysis. At or very near to the surface of the soil the mycelial strands are freer from mechanical pressure than well below it ; and it is therefore possible that mechanical freedom is also a factor in deciding the place of origin of the fruit-bodies. Here again there is room for an experimental enquiry.1 The Fate of Rudimentary Fruit-bodies. — On artificial mushroom- beds, a great many rudimentary fruit-bodies are produced for every one that comes to maturity : " many are called, but few are chosen." Perhaps there are at least fifty rudimentary fruit-bodies brought into existence for every one which ultimately liberates spores. No doubt the same thing happens in nature, although in pastures it is usually impossible to observe any fruit-bodies at all in their very earliest stages of development, owing to their being hidden by herbage. It seems to me that there is a considerable advantage in the formation of so many rudiments, notwithstanding that the vast majority of them are always destined to suffer abortion. The surface of the ground is an external condition 1 For an experimental enquiry into the cause of the position of origin of the fruit-bodies of Coprinus sterquilinus, vide vol. iii, Chap. IV. PSALLIOTA CAMPESTRIS 399 which the fungus plant is unable to alter. So far as the fungus is concerned, it may be beset with irregularly disposed mechanical obstacles, it may be divided into moister and drier spots, it may harbour lurking enemies, or be crushed down locally by the feet of herbivorous animals. The mycelium, on arriving at the surface of the substratum, gives rise to a considerable number of rudi- mentary fruit-bodies which at first are all alike ; but very soon a few, which occupy the most favourable situations for further growth, draw to themselves the nutriment available in the mycelial strands and thus rapidly increase in size. Most of the other rudiments not so favoured cease to draw nutriment to themselves and soon wither. It is not unlikely that some of the protoplasm of the rudiments which become aborted is passed back again into the mycelium, so that it may not be wasted, but ultimately finds its way into the favoured fruit-bodies which are destined to develop to maturity and thus propagate the species. By forming a large number of rudimentary fruit-bodies at relatively little expense in material and energy, the mycelium, as it were, carries out a reconnaissance of the nature of the substratum-surface and finally produces its effective reproductive organs only at the most favourable points. Effect of Dry Weather on Development. — I think it probable that, in nature, the rudimentary fruit-bodies, like those of Coprinus niveus and certain other coprophilous fungi which I have actually observed, often have their development interrupted by lack of moisture. If during what may be called the mushroom season the weather continues dry, as everybody knows, no mushrooms appear. If, after such a dry spell, the weather breaks and heavy rain descends for a day or two so that the pastures are soaked, then large numbers of mushrooms make their appearance in the course of a very few days. I am inclined to believe that, when events happen in this way, the mushrooms are already present during the dry weather in the form of rudiments at the surface of the ground, but that their further development and final expansion have been delayed by drought which has temporarily checked the supply of constructive materials coming up from the mycelial strands. If the fruit-bodies which appear after a rainy day in 400 RESEARCHES ON FUNGI summer were to begin their development only after the soil had become watered, probably they would not appear in the pastures under the most favourable circumstances for at least ten days. The early stages of fruit-body development take place very slowly and cannot be hurried beyond a certain pace. The growth of a fruit-body in size undergoes acceleration from day to day, and the last stage, in which the stipe elongates and the pileus expands, is passed through so rapidly that many people are apt to think that a mushroom completes its full development in a single night. The rapidity of the expansion of a mushroom in its final phase of growth has been proverbial from the most ancient times and is, of course, based on the observations of the mushroom-gatherer bent on booty for the table ; but it must not be forgotten that this marvel of nature is only the climax of many days of slow and steady, although usually invisible, developmental progress. The Spore-discharge Period and the Number of Spores. — The length of time occupied by any fruit-body in the discharge of spores may be conveniently called the spore -discharge period.1 In order to determine this period for Psalliota campestris recourse was had to the following procedure. Two clods of turf each bearing a single mushroom, which was just bursting its gill-chamber and under- going rapid expansion, were dug up in a field with great care so as not to injure the fungi in any way, and were removed to a potting - shed. Here each clod, which was about 5 inches in diameter and 5 inches deep, was inserted in an 8 -inch pot and packed below and at the sides with wet soil, so that each mushroom took up a vertical position similar to that which it had had in nature. Each pot was then covered by a similar but inverted pot, and the hole at the top was partially closed with a plug of cotton-wool. The two preparations were then removed to a somewhat cool room in a house which was temporarily used as a laboratory. There- upon, at 10.15 A.M., some glass slides were placed beneath each mushroom. One of the mushrooms was a little further advanced in development than the other. Let us call the more advanced 1 A large Horse Mushroom (Psalliota arvensis) shedding spores in a field during the spore-discharge period is shown semi-diagrammatically in Fig. 76, vol. i, 1909, p. 218. PSALLIOTA CAMPESTRIS 401 one A and the less advanced one B. Already by 11 A.M. the mush- room A had produced a good spore-deposit on the slides, whilst B had produced a distinct but relatively poor one. It was clear that spore-discharge had already begun : possibly it had been in progress in A for two hours, but in B probably for only about half an hour. The glass slides were removed and replaced by new ones : and during the whole period of spore-discharge this process of periodical replacement was continued. Usually the slides were changed twice a day, at about 10 A.M. and 8 P.M. respectively. The continuance and vigour of spore-discharge were judged by the eye from the density of the spore-deposits as seen against a background of white paper. In order not to disturb the orientation of the gills, the pots containing the mushrooms were not moved during the whole period of observation. By making observations in the manner just described, it was found that the mushroom A yielded strong spore -deposits for four days and four nights. The spore-deposit made during the next 24 hours was only very thin, and at the end of this time the fruit-body toppled over and collapsed. The mushroom B yielded excellent spore-deposits for five days and five nights. During the next day and night the spore-deposit was very thin and, toward the end of this time, the mushroom fell over somewhat on its side. Spores then ceased to be liberated, whereupon the mushroom soon collapsed and became putrescent. From the above observations we may conclude that the spore- discharge period of a wild mushroom is five or six days in length, each day being reckoned as 24 hours. Doubtless, in nature, the period is affected by weather conditions such as temperature and moisture ; but, so far, I have not found it possible to investigate their influence. At the moment when spore -discharge began, the pilei were convex, but by the second day they had already become flattened. The spores were therefore being shed whilst the gills were being raised into their upward-inclined positions. This movement, as we have seen, does not affect the verticality of the median planes of the gills or prevent the hymenium from continuing to look more or less downwards toward the earth. The discharge of the spores VOL. II. 2 D 402 RESEARCHES ON FUNGI during the expansion of the pileus was therefore taking place without any spores being wasted by lodging on the gill-sides. The gills, when spore-discharge began, were pinkish, but by the second day were brown. The brown rapidly deepened so that the gills became almost black. The darkening of the ^^^^^^^^^^^^^^^^^^^^^^^^^^ gills, as has al- ready been pointed out, is simply due to the deepening of the pigmented cell- sap in the cells of the gills and is not due to decay or, in the main, to the accumulation of waste spores. Even when the gills had become dark choco- late - brown, the hymenium still con- tinued to liberate vast numbers of spores. The number of spores which a mushroom pro- duces can be deter- mined directly, by estimating the number of spores in the spore-deposit (Fig. 141) or, indirectly, by estimating the number of basidia which are formed in the hymenium. A spore-deposit, which had been produced by a wild mushroom within forty-eight hours from a pileus 8 cm. wide, was investigated with a counting apparatus. The result was given in Volume I.1 The number of spores was found to be 1,800,000,000, from which it was concluded that, on the average, about 40,000,000 had fallen during each hour of the spore-fall period. However, since it has now been shown 1 Vol. i, 1909, p. 82. FIG. 141. — A spore-deposit of a wild Mushroom, Psalliota campestris, gathered at King's Heath, England. The pileus was placed on white paper for about six hours. To the right, the black spore -lines were formed under interlamellar spaces, while the white spore- free lines were left under the free gill-edges. To the left, the clouding is due to the fact that the gills in this region did not at first look vertically down- wards. Natural size. PSALLIOTA CAMPESTRIS 403 that the spore-discharge period is more than double forty-eight hours, we must conclude that the number 1,800,000,000 was much less than the total number of spores which would have been liberated, had the pileus not been cut off from its stipe but allowed to continue its development to the point of exhaustion in nature. The number of spores which a mushroom can produce was found by counting the basidia in the following manner. A gill of a wild mushroom, which had been shedding spores for about two days and one night, was removed from the pileus and treated with chlor-zinc iodine. The reagent killed the hymenial cells, but all those basidia which were developing or were soon to develop spores were stained a decided yellow, owing to the presence in them of accumulated protoplasm containing proteins. The yellow basidia on an area of the hymenium measuring forty-nine ten- thousandths of a square mm. were all sketched on paper with the help of a camera lucida and were found to number 108. By filling up the spaces in the area and thus allowing for the exhausted basidia which had already shed spores during the period of about two days and one night already mentioned, the number of basidia for the area was increased to 164. The increase allowed for four generations of basidia, fourteen to a generation on the area under investigation, the number 14 being the same as that found by actual observation of similar areas of the living hymenium on other gills of the same fruit-body. The number of generations belonging to the past-generation basidia was taken as four, this being the number which could shed their spores successively in 36 hours if each generation took about 8 hours and 30 minutes for its full development.1 If we suppose, therefore, that the total number of basidia on forty-nine ten-thousandths of a square mm. is 164, then on one square mm. there would be 33,265 basidia, and these would produce 133,060 spores. The number of spores pro- duced by a square inch of hymenial surface would be 83,162,500 approximately. Now a mushroom about three and three-quarters of an inch in diameter was found to have a hymenial area of 195 square inches. Supposing the packing of the basidia per square 1 Vide Chapter II. 404 RESEARCHES ON FUNGI inch to be what we have calculated, this particular mushroom must have produced a total of 18,218,687,000 spores. As a result of the investigation just described, it may be stated in round figures that a wild mushroom, with a pileus about 4 inches in diameter, possesses a hymenial area of about 1 • 33 square feet, which contains about 4,000,000,000 basidia which discharge about 16,000,000,000 spores. There can be no doubt whatever that every large wild mushroom which develops in a normal manner liberates upwards of 10,000,000,000 spores. The Mushroom and the Panaeolus Sub-type.— The gills of Psalliota campestris possess all the general characters which have been described for the Aequi-hymeniiferous Type of organisation : the gills are wedge-shaped in cross-section and positively geotropic, the hymenium everywhere looks more or less downwards toward the earth, and every small part of the hymenium (every square mm.) produces and liberates spores during the whole period of spore-discharge. The gills also exhibit all the characters which have been enumerated for the Panaeolus Sub-type, as is indicated superficially by the mottling of the hymenium, which differs in no essential point from that of Panaeolus campanulatus or Stropharia Text-book Illustrations of the Hymenium. — For several decades Sachs' figure of a cross-section through the gill of a Mushroom has been considered sufficient by many text-book writers as an illus- tration for the hymenium of the Hymenomycetes in general. It has been copied from book to book and is familiar to every botanist. Sometimes it has been plagiarised, but slight alterations for the worse do not always prevent one from detecting its original source. Sachs' figure has done excellent service : it has enabled students to realise at a glance some of the main facts in gill-structure ; it teaches that the hymenium, subhymenium, and trama are differentiated from one another, and that in any section through the hymenium two kinds of elements may be distinguished, namely, those which bear spores and those which do not ; and, finally, it represents fairly correctly the shapes of the various cells, more especially those of the basidia with their club-shaped bodies, their sterigmata, and their oval spores. However, when one comes to PSALLIOTA CAMPESTRIS 405 examine the figure and its description in detail, a number of un- satisfactory features soon make themselves apparent. It is stated that " many of the tubes (hymenial elements) are sterile and are called paraphyses, others produce the spores and are the basidia." l In the illustration, four basidia are shown in various stages of development, all protruding beyond the much more numerous sterile cells and all provided with sterigmata. There is nothing in the description of the figure or in the figure itself to indicate whether or not any or all of the sterile elements eventually develop into basidia. So far as my experience goes, the student usually infers that the so-called paraphyses are destined to remain barren, and so comes to the incorrect conclusion that the paraphyses are much more numerous than the basidia. The basidium which has discharged its spores is represented as protuberant ; but, as a matter of fact, a basidium at such a stage in its history rapidly collapses and sinks down to the level of the paraphyses. Collapsed basidia are entirely unrepresented. There is nothing in the illus- tration to indicate that the whole number of basidia develop in an orderly series of successive generations. The hymenium is not nearly so sharply differentiated from the subhymenium as is shown, but the one passes gradually into the other. Finally, no surface view of the hymenium accompanies the cross-section, so that it is impossible to obtain more than a very limited idea of the organisation of the hymenium as a whole. These criticisms of Sachs' illustration of the hymenium of Psalliota campestris apply equally to Strasburger's illustration for Russula rubra.2 Van Tieghem 3 reproduces Sachs' figure of the hymenium and states that the shorter elements (those not bearing sterigmata) remain sterile and are the paraphyses, whilst the other longer elements which project above the general level of the hymenium (all of which in the illustration have sterigmata) are the basidia. This, as we shall see, is an erroneous division of the elements, for many of the shorter ones are certainly young basidia. 1 K. Goebel, Outlines of Classification, etc. A new edition of Sachs' Text-book of Botany, Book II, English translation, Oxford, 1887, p. 134. 2 Strasburger, Schenck, Noll, and Karsten, A Text-book of Botany, third English (based on eighth German) edition, London, 1908, p. 408. 3 Van Tieghem, Traite de Botanique, Paris, 1884, p. 1049. 406 RESEARCHES ON FUNGI I shall now endeavour to explain the organisation of the hymenium of the Mushroom in detail, and the reader will find that it is essentially the same as that of Panaeolus campanulatus and Stropharia semiglobata. However, the hymenial elements of the Mushroom are very much smaller than those of these two species, and their investigation has therefore been proportionally more difficult to carry out. The Mottling of the Gill Surface.— The hymenium covering the exterior of each gill which has become outstretched is finally mottled with small lighter and darker areas (Fig. 134, B, p. 377). The pattern is similar to that already described for Panaeolus campanu- latus and Stropharia semiglobata, but finer. This relative fineness is in part due to the fact that the diameter of the basidium of a Mushroom is only about one-half that of the basidium in the other species. If a dark area of a Wild Mushroom be examined with the microscope, it is found to possess a large number of basidia, each of which is bearing four spores which have already become pigmented. The darkness is simply due to the colour of the spores. The light areas, on the other hand, contain basidia which are just about to develop spores or which bear spores which may or may not have attained full size but are still unpigmented. The relative lightness is due to the absence of spore- wall pigment. With the help of the microscope, one finds that the light and the dark areas are irregularly shaped and very variously arranged in respect to one another. A semi-diagrammatic drawing of the hymenium of a Cultivated Mushroom, in which are shown only the spores and the basidia which have produced them, is repro- duced in Fig. 142. Here we see a light area surrounded by a dark area. The dotted contour-line indicates the limits of the former, while the arrows point to the directions in which the line is being pushed. Just outside the line, the spores on the dark area are being gradually discharged, while just inside it new basidia are coming to maturity and producing spores, so that as the dark area becomes reduced the light area becomes extended. There are such waves of development passing over the hymenium in a very irregular manner in all directions on every gill. In the course of a few hours, every dark area, owing to the discharge of its spores, PSALLIOTA CAMPESTRIS 407 becomes a light area, and every light area, owing to the develop- ment upon it of spores which have become pigmented, becomes a dark area. A single area becomes alternately dark and light a FIG. 142. — Psalliota campestris (cultivated ^form). Surface view of 'a piece of hymenium 0-15 mm. long and 0-1 mm. wide, semi-diagrammatically showing a white area surrounded by a dark area. The spores alone have been fully sketched, but the positions of the spore-bearing basidia are indicated by dotted circles. The white area is enlarging at the expense of the dark area : the dotted line indicates the limits of the white area and the arrows the directions of its extension. The basidia at a, b, and c, just outside the dotted ring, have each discharged one of their two spores ; and at a and c water-drops are being excreted at the bases of the remaining spores in preparation for spore-discharge. The basidia at d, e, /, and g in the white area are developing spores which as yet are only from 3 to 15 minutes old but which are rapidly attaining full size. At h the spores are older and of full size but colourless. At i and j the spores are still older and are becoming pigmented. At k, I, and m in the dark area the basidia have developed only one spore each, but the spore has twice the normal volume. At n and o two adjacent basidia have developed spores simultaneously and the four spores in each group have arranged themselves so that they occupy the corners of a square. At p, also, the spores are so arranged that those of adjacent basidia can not come into contact during development or discharge. Magnification, 706. number of times in succession during the several days of the spore- discharge period. The actual passage of slow waves of develop- ment over the hymenium of a Cultivated Mushroom was witnessed for a gill kept under continuous observation for eight hours in a compressor cell in the manner described in Chapter II. 408 RESEARCHES ON FUNGI Whilst the spores on an originally light area are in the early stages of pigmentation, the tint of the area is intermediate between that of a light and that of a dark area. If we wish, therefore, we can speak of three different kinds of areas as making up the mottled hymenial pattern : dark, light, and intermediate. With the microscope it is not difficult to detect intermediate areas here and there : they often mark the transition of a light area into an adjacent dark one. If the reader desires any further details in regard to the general phenomenon of mottling, he should turn to the fuller account given in the description of the organisation of Panaeolus campanulatus.1 Methods for Examining the Hymenium in Surface View. — The method described in Chapter II was made use of for the purpose of examining a large area of the hymenium with the low power of the microscope (magnification about 130), i.e. a living gill was mounted on a tiny drop of water in a compressor cell. However, owing to the small size of the basidia and spores, it was found necessary to magnify the hymenium with the high powers of the microscope. Now the gills are pigmented and absorb much of the light passing through them. Therefore, in order to increase the amount of light striking the hymenium when viewed with high powers, the following procedure was adopted. A gill was dissected away from a living fruit-body, wrapped around a finger, and from it a very thin surface-section was taken with the help of a hand- razor. A small drop of water was placed on a glass slide. The section was then mounted on the drop, so that its under sub- hymenial side alone was wetted, and then covered with a cover- glass. The result of these operations was that a thin and fairly transparent section was obtained which "could not dry up quickly and which possessed a hymenium which looked upwards under a cover-glass without being wetted. The cover-glass' only touched and injured certain parts of the section, so that a sufficient amount of the hymenium was left for examination in the normal condition. Strong daylight was directed through the preparation from the mirror. The Number and Size of the Spores on Individual Basidia. — The 1 Chapter X. PSALLIOTA CAMPESTRIS 409 number of spores on individual basidia was determined by the method just described, i.e. by the examination of the living hymenium in surface section. It was found that this number is not identical for Wild and Cultivated Mushrooms. In Wild Mushrooms gathered in several places in England (Birmingham, Redditch, etc.) the number of spores on each basidium .:. .. •• • • •• ®© . •*• Vn •• •* •• •• • O ••*.*.•• FIG. 143. — Psalliota campestris. Camera-lucida drawings of the spores of the present-generation basidia on the living hymenium. A and B, from a wild form of the fungus ; C, from a cultivated form. In A, the basidia are tetra- sporous with one exception which is trisporous. In B, in addition to the tetrasporous basidia, there are a number of trisporous basidia. The spores of the tetrasporous basidia are shown black, whilst those of the trisporous are shaded with lines. In C, most of the basidia are bisporous, but others are monosporous. The spores of the bisporous basidia are shown black, whilst those of the monosporous are shaded with lines. Magnification, 440. was usually four ; but here and there on the hymenium, mixed with these quadrisporous basidia, were frequently found a certain number of trisporous basidia which occurred either singly or in little groups. In Fig. 143, A, is shown a small area of the hymenium in which only one trisporous basidium is associated with the normal quadrisporous basidia, while at B is shown another similar area in which trisporous basidia are almost as numerous as the quadri- sporous. In these drawings, all the groups of four spores have been represented in uniform black, whilst the groups of three spores, for the sake of distinction, have been shaded with straight lines ; 410 RESEARCHES ON FUNGI but, in actual fruit-bodies, the colour of the spores is not in the least dependent on the number of spores produced per basidium, but only on the degree of spore-maturity. In a variety of Cultivated Mushroom, which I grew on a bed of horse manure at Birmingham, England, the basidia were found to be usually bisporous ; but, mixed with the bisporous basidia, were a larger or smaller number of monosporous basidia. An area of the hymenium showing spores in pairs (represented in uniform black) and single spores (shaded with lines) is illustrated in Fig. 143, C. Bisporous basidia, as well as a few monosporous basidia, are also shown in Fig. 142 (p. 407). In no fruit-body of the Wild Mushroom could I find a basidium with either one or two spores upon it ; and, corresponding with this, in no fruit-body of the Cultivated Mushroom could I find a basidium with either three or four spores on it. However, it would not be a matter of surprise to me if varieties of Mushrooms were discovered in which monosporous, bisporous, trisporous, and quadrisporous basidia occurred mixed together on the same hymenium. It was pointed out in the first volume of this work that both Atkinson and I have found two-spored varieties of Mushrooms coming up upon the campus of our respective Universities.1 It is therefore possible that, in North America at least, there is a Wild Mushroom with two-spored basidia ; but, to obtain certainty in this matter, further observations are necessary. It seems very likely that the two-spored varieties of Psattiota campestris have been derived by mutation from the four-spored.2 In regard to the size of the spores, it is to be noted that the spores on a trisporous basidium in a Wild Mushroom are slightly larger than those on a quadrisporous basidium (Fig. 143, A and B, p. 409), and also that the spore on a monosporous basidium in a Cultivated Mushroom is distinctly larger than the spores on a bisporous basidium (Fig. 143, C). In giving the average size of the spores, this fact ought to be taken into account. Usually, in a spore-deposit from a Cultivated Mushroom, one can distinguish by their size with a fair amount of certainty which spores have 1 Vol. i, 1909, p. 15. 2 Ibid. PSALLIOTA CAMPESTRIS 411 come from monosporous basidia and which from bisporous basidia. If one were to plot out a curve showing the variations in size of a large number of spores of a single Cultivated Mushroom, one would find that the curve would have two peaks. In a Cultivated Mushroom, the spores of the monosporous basidia have a volume of about twice that of the spores of the bisporous basidia. This is associated with the following facts. The basidium-bodies of the monosporous and bisporous basidia are equal in size (vide infra, Fig. 147) and, before the formation of the spores, are equally filled with protoplasm. This protoplasm is eventually all poured into the spores (with the probable exception of a very thin lining layer), and the spores become stuffed with it. Where there are two spores on the basidium, the protoplasm is equally divided and, where there is only one, it is undivided. The volumes of the spores on the monosporous and bisporous basidia are correlated with the lelative amounts of protoplasm which the spores are destined to contain. It seems very probable that the behaviour of the nuclei is different for monosporous and bisporous basidia on the same hymenium of a Cultivated Mushroom, and different also for the trisporous and quadrisporous basidia of a Wild Mushroom. This is a matter which invites cytological investigation. The complete emptying of the contents of a basidium-body (with the probable exception of a very thin lining layer necessary for maintaining turgidity) into the spores, whilst these are ripening, is a phenomenon which occurs not merely in Psalliota campestris but quite generally in Basidiomycetes. I have observed it in a large number of species. If, therefore, one finds a spore-bearing basidium only partially empty, one may be quite sure that the spores, although of full size and partially or completely pigmented, are not yet fully ripe, and that a stream of protoplasm is still pass- ing into them through the fine necks of the sterigmata. This is a fact that ought to be borne in mind by all cytologists when studying nuclear behaviour in connection with basidia.1 It has been shown that the relative size of the spores of mono- sporous and bisporous basidia in a Cultivated Mushroom is directly 1 CJ. Chap. I, pp. 27-29. 4i2 RESEARCHES ON FUNGI correlated with the amount of nutrient protoplasm which each kind of spore is to receive. This conclusion suggests another of a more general kind, namely, that the fluctuating variations in the size of the spores for any species belonging to the Hymenomycetes are correlated with the amounts of protoplasm in the basidia which produce the spores. Now the amounts of protoplasm in the basidia usually vary directly as the volume of the basidia. It therefore seems highly probable that, if of two individual fruit- bodies of the same species one has spores which on the average are larger than those of the other, the one with the larger spores will have the larger basidia. Spore- size is not improbably correlated with corresponding variations in size in all the elements of the hymenium, and perhaps of all the elements of the whole fruit- body. In studying a large number of species of the genus Coprinus, I have found that the species with the largest spores have the largest basidia, and those with the smallest spores the smallest basidia. So far as species of Coprinus are concerned, there is undoubtedly a strict correlation between spore-size and basidium- size. Rate and Mode of Development of the Spores of an Individual Basidium. — The rate of development of the four spores of a single basidium from their first rudiments up to full size was determined by direct observations made upon a Wild Mushroom gathered in England. A gill was dissected off the fruit-body and placed on a glass slide, and over it was laid a large cover-glass. No mounting fluid was employed ; but, on the autumn day when the investiga- tion was undertaken, the air was almost saturated with moisture, so that the gill under the cover-glass lost water but very slowly indeed. Two basidia, one with fully pigmented nearly ripe spores belonging to an older generation, and another which had developed sterigmata but not spores and belonged to a younger generation, were watched, and sketches of them were made at intervals. The first beginnings of the new spores on the sterigmata of the younger basidium are shown in Fig. 144 at A. After an interval of 10 minutes from the beginning of the observations, the spores had grown to half their full diameter (B), after 20 minutes to about three-quarters (C), and at the end of 30 minutes they had attained PSALLIOTA CAMPESTRIS 413 their full size (D). A similar set of sketches, made in succession with 10-minute intervals, is shown in Fig. 145. In this instance, the spores of both the younger basidia attained to their full size A B C n o°o o°o FIG. 144. — Psalliota campestris, wild form. Rate of growth of spores on the ends of the sterigmata. A, the spores of two basidia, one set full- grown and the other just coming into existence, as seen on a living gill. B, C, and D, sketches of the same made successively with intervals of 10 minutes. The spores attained full size in about 30 minutes. Magnification, 820. within 40 minutes. We may conclude, therefore, that the spores of a Wild Mushroom at room temperatures take only from 30 to 40 minutes to grow to full size. In order to observe the course of the development of the spores after they have attained full size, it was found necessary to resort A B C D E •A FIG. 145. — Psalliota campestris, wild form. Rate of growth of spores on the ends of the sterigmata. A, the spores of four basidia, two sets full-grown and two just coming into existence, as seen on a living gill. B, C, D, and E, sketches of the same made successively at 10, 20, 30, and 40 minutes respectively after the stage A. The young spores attained full size in a little less than 40 minutes. Magnification, 296. to the compressor-cell method. A piece of a gill was taken from a Cultivated Mushroom and placed on a very small water-drop in a compressor cell. The lid of the cell was then closed down, so that it just touched the gill at one spot. It thus became possible to observe the hymenium with a magnification of about 300 diameters. By using this method it was observed that the 4i4 RESEARCHES ON FUNGI spores, after attaining full size, remained colourless for about two hours. Then the process of pigmentation of the spore- walls set in, and the walls continued to deepen in colour for about two hours. After pigmentation of the spores had been completed, the spores remained on the ends of their sterigmata for about three hours and ten minutes. At the end of this time they were shot away. The time taken by a spore to develop from a tiny rudiment on the end of its sterigma up to maturity and ultimate discharge was approximately eight hours.1 An Analysis of the Hymenium of the Cultivated Mushroom. — The analysis of the hymenium about to be given is similar to that for Panaeolus campanulatus. Any small portion of the hymenium, such as that which is included within a dark or a light area of a gill and which has begun to shed spores, is made up of the following elements : 1. Past-generations Basidia, 2. Present-generation Basidia, 3. Coming-generation Basidia, 4. Future-generations Basidia, and 5. Paraphyses. On any small portion of the hymenium the basidia come to maturity in a series of successive generations, and the para- physes — the permanently sterile elements — undergo gradual enlarge- ment during the progressive exhaustion of the basidia in exactly the same manner as has been already described for Panaeolus campanulatus. The arrangement of the various elements of the hymenium in respect to one another in surface view is shown semi- diagrammatically in Fig. 146. On account of the small size of the basidia and paraphyses and the consequent increase in the optical difficulties of investigation, it was not found possible to make camera-lucida drawings showing the positions of all the elements in an actual surface view of the hymenium. Fig. 146, however, gives a synthesis of a number of detailed observations and is sufficiently correct. It will be found to agree in all 1 Cj. Chap. II. PSALLIOTA CAMPESTRIS 415 essentials with similar Figures for Panaeolus campanulatus and Stropharia semigtobata, which were actually made with the help of the camera lucida. The area of Fig. 146 is approximately equal to one-hundredth of a square mm. The present-generation basidia (those bearing spores) are identical with those shown in the right two- thirds of Fig. 142 (p. 407). The piece of hymenium is supposed to have been shedding spores for about 24 hours, so that already it contains a number of basidia belonging to past generations. However, it is far from exhaustion. Indeed, it should continue to shed spores for at least four days longer. New generations of spores are potentially resident in the coming-generation basidia and in the numerous basidia belonging to future generations. There are no cystidia on the surface of the gills. All the elements belong to one or other of the foUowing five groups : past-generations basidia, a a ; present-generation basidia, b b ; coming-generation basidia, c c ; future-generations basidia, d d ; permanently sterile paraphyses, e e. The past-generations basidia, a a, are collapsed and shrunken, devoid of protoplasmic contents, with tops drawn down to the general level of the hymenium. Their free ends are concave and contain the remains of the sterigmata which are reduced to mere stumps. When a mushroom has been shedding spores for five or six days and is in the last stages of hymenial exhaustion, most of the past-generations basidia are found to have entirely disap- peared. No such disappearance of these elements, however, is shown in the present drawing, as the piece of hymenium is supposed to have been shedding spores for not more than 24 hours. The present-generation basidia, b b, are distinguished from all others by the fact that they bear spores. The spores on the left- hand side of the Figure are colourless and many of them have not attained full size, while those on the right-hand side are of full size and completely pigmented. Most of the basidia of the present generation are bisporous, b b ; but here and there certain of them are monosporous, m m. The spores on the latter have decidedly larger diameters than those on the former. As soon as a present-generation basidium has shed the last of its spores, it becomes a past-generation basidium : after about twenty 416 RESEARCHES ON FUNGI "4 FIG. 146. — Psalliota campestris (cultivated form). Semi-diagrammatic surface view of the living hymenium of a Mushroom, about 24 hours after the beginning of the spore-discharge period. The [length of each side of the square is 0-1 mm., so that the area of the square is equal to 0-01 of a square mm. All the elements belong to one or other of the following five groups : past-generation basidia, a a : present-generation basidia, b b ; coming- generation basidia, c c ; future-generation basidia, d d ; permanently sterile paraphyses, A wave of de- velopment is proceeding from the light area on the left to the dark area on the right in the directions shown by the arrows. The series 1 to 7 are basidia in successively younger stages of development. No. 1, basidia with spores which are full-sized but young (only about 1 hour old) and unpigmented ; no. 2, basidium with spores not yet of full size (about 30 minutes old) ; no. 3, basidium with spores only one-third of their eventual diameter (about 10 minutes old) ; no. 4, basidium with spores which have just begun their existence (about 5 minutes old) ; no. 5, basidia with sterigmata fully developed but no spores ; no. 6, basidium with sterigmata just beginning to develop ; no. 7, basidium as yet without even a trace of sterigmata. Nos. 1 to 4 belong to the present-generation basidia of the light area, while nos. 5 to 7 belong to the coming-generation basidia of the dark area. Spore-discharge is taking place at the left-hand side of the dark area. The basidia * s have just discharged their spores and are collapsing ; their sterigmata are sinking into a concavity which is forming at the- end of each basidium-body. The basidium w is just discharging the second of its two spores and, as a pre- liminary thereto, a drop of water has been excreted within the last 10 seconds at the spore-hilum. Most of the basidia of the present generation are bisporous, b b ; but, here and there, certain basidia are monosporous, m m, and have larger spores. A past-generation basidium, if it has two sterigmatic stumps, has been derived from a bisporous basidium ; but, if it has only one such stump, as shown at cm, it has been derived from a monosporous basidium. An undischarged or waste spore lying on its side and adherent to the surface of the hymenium is shown at u. (The present-generation baridia are identical with those shown in the right two-thirds of Fig. 142.) Magnification, 1,060. PSALLIOTA CAMPESTRIS 417 minutes it collapses, its end becomes concave, and it is drawn down to the general level of the hymenium. A past-generation basidium which has two sterigmatic stumps in its concave end has been derived from a basidium that was bisporous; but one which, has only one such stump, as shown at cm, has been derived from a monosporous basidium, such as m m. The basidium-body of each present-generation basidium is shaped more or less like a club with the swollen end directed outwards. This swollen end pro- trudes slightly above the general level of the hymenium and there- fore hides to a greater or less extent the immediately adjacent past-generations basidia, future-generations basidia, and para- physes. Owing to this fact, the optical difficulties in making out the structure of the hymenium in surface view are considerably increased. The amount of massive protoplasm inside the body of a present-generation basidium, i.e. all the protoplasm except a very thin layer lining the walls which probably remains until the basidium collapses and dies, varies greatly in correspondence with the age of the spores. When the spores are just beginning their development on the ends of the sterigmata and are the merest rudiments, the basidium-body and sterigmata are filled or almost completely filled with protoplasm ; but, as the spores grow in size and for several hours after they have attained their full size, and even whilst they are undergoing pigmentation, the protoplasm flows through the sterigmata spore wards. This progressive evacuation of the basidium-body continues until the spores are ripe and ready to be discharged, and it is only then that the basidium-body and sterig- mata appear transparent and empty of solid contents. As the massive protoplasm moves out of the basidium, its place is taken by cell- sap. The coming-generation basidia, c c, are the basidia which are destined to produce a new crop of spores as soon as the crop upon the present-generation basidia has been discharged and the basidium-bodies of these elements have collapsed and sunk down to the general level of the hymenium. As soon as the present- generation basidia begin to develop spores, certain other basidia in their immediate vicinity become clearly marked out as their successors, i.e. as the coming-generation basidia, by their rapidly increasing size and prominence, and by the fact that they are 418 RESEARCHES ON FUNGI filled with dense protoplasmic contents. The development, ripen- ing, and discharge of the spores of the present-generation basidia, as we have seen, take about eight hours to accomplish. During this time, the coming-generation basidia interspersed between them are by no means quiescent. They first complete the growth in length and diameter of their basidium-bodies, so that these become equal to those of the present-generation basidia (cf. b and c, on the right-hand side of Fig. 146). Then the fusion-nucleus in each basidium divides and so produces nuclei which are subsequently to be translated to the spores. I have not attempted to study the nuclear divisions in the basidium, and therefore am unable to state from my own observations whether the fusion-nucleus divides into two or four nuclei. However, Rene Maire, who made a cytological investigation upon the bisporous basidia of a Cultivated Mushroom, states : that there are two nuclei in the young basidium which unite to form a fusion-nucleus (secondary nucleus) ; that this nucleus then behaves in the usual manner, i.e. it produces four nuclei by two successive divisions — a process probably accom- panied by the reduction of the chromosomes to one-half x ; and that, after the first of these nuclear divisions has taken place, the basidium begins to produce its two sterigmata.2 The development of the sterigmata is a slow process which requires at least thirty minutes. Usually a coming-generation basidium begins to develop its sterigmata about half an hour or so before the moment when the nearest present-generation basidium will discharge its last spore. Thus the sterigmata of the coming-generation basidia are almost ready (within 15 minutes or so of being ready) to develop spores at the moment when the hymenium in their immediate neighbourhood has become cleared of the spores of the previous generation of basidia. Thus one crop of spores is quickly followed by another. It is in consequence of this very orderly mode of development of one generation of basidia after another that by far the greater part of the hymenium at any moment is found to contain spore- bearing basidia. As soon as a basidium begins to develop the 1 Rene Maire, " Recherches cytologiques et taxonomiques sur les Basidio- mycetes," Bull Soc. Myc. France, T. 18, 1902, p. 151. 2 Ibid., p. 152. PSALLIOTA CAMPESTRIS 419 rudiments of spores on the tips of its sterigmata, it must be con- sidered to have passed from the coming to the present generation. The number and general distribution of the basidia of a coming generation on any given area of the hymenium are equal to those of the present generation, and it is a general rule for any generation of basidia that each basidium in it shall be sufficiently well separated from its neighbours to enable it to develop and discharge its spores without being jostled. We have seen that, according to Rene Mair.e, four nuclei are produced in the body of a coming-generation basidium as a result of two divisions of the fusion-nucleus. According to this author, two of the nuclei pass up one sterigma into one spore and two pass up the other sterigma into the other spore. Then each nucleus divides, so that each ripe spore comes to contain four nuclei. Maire noticed, however, that sometimes the fusion-nucleus only divides once, and then, he says, each spore only receives one nucleus, which afterwards may divide once or twice.1 He does not appear to have observed that monosporous basidia frequently occur in the hymenium of the Cultivated Mushroom. It seems to me that further investigation is required to clear up the irregularity to which Maire refers. It may be that, when the fusion-nucleus only divides once, a single sterigma and spore are produced and not two ; whereas, when there are two divisions, there may always be two sterigmata and two spores formed. In any case, cytologists, when studying the Cultivated Mushroom, must in future take into account the fact that there are two kinds of basidia present in the hymenium — bisporous and monosporous. One of the most remarkable events in the development of a basidium in the Hymenomycetes generally is the migration of the four nuclei from the basidium-body into the spores through the four extremely narrow sterigmatic necks. This has been studied cytologically by Ruhland,2 Maire,3 Fries,4 and more recently by 1 Rene Maire, loc. cit., p. 152. 2 W. Ruhland, " Zur Kenntnis der intracellularen Karyogamie bei den Basidio- myceten," Bot. Zeit., Jahrg. 59, 1901, Abt. I, pp. 187-206. 3 Rene Maire, loc. cit. 4 R. E. Fries, " Zur Kenntnis der Cytologie von Hygrophorus conicus" Svensk. Bot. Tidskrift, vol. 5, pp. 241-251. 420 RESEARCHES ON FUNGI Levine.1 The first of these observers showed that, when traversing a sterigmatic neck, a nucleus undergoes a great alteration in shape : it becomes drawn out into a filament. The other workers have observed fibrillar strands leading upwards from the nuclei into the sterigmata and sometimes into the spores. According to Levine, at the end of the second nuclear division in the basidium- body, the four centrosomes become attached to the free end of the wall of the basidium and the four daughter nuclei become reconstructed in close connection with them. The four nuclei then move downwards somewhat in the basidium but maintain their connections with the centrosomes by means of fibrillar strands. The positions which the centrosomes take up at the free end of the basidium-wall mark the points of origin of the sterigmata ; and, as these structures develop, the centrosomes are carried up in their tops. The fibrillar strand connecting each nucleus with its centro- some becomes elongated during this process. When the spores begin to be formed on the ends of the sterigmata, the centrosomes still remain attached to the cell-wall and come to occupy positions destined to be the spore-apices. Each fibrillar strand again elon- gates during the growth of the corresponding spore to full size, so that eventually the four nuclei which may be situated some distance down the basidium-body, are attached by relatively very long and thin fibrillar strands with the centrosomes at the apices of the spores.2 Maire, Fries, and Levine all agree that the fibrillar strands are associated in some way with the passage of the nuclei into the spores. The strands occupy the paths of the nuclei and possibly direct the nuclear migrations. Levine says that, since the strands pass not merely through the sterigmata but also into the spores, they may be of significance in the apparently difficult process of passing the nuclei through the extremely narrow sterigmatic necks. I am inclined to agree with this view. It remains, however, to be discovered whether the strands are kinoplasmic and contract, so that they drag the nuclei into the spores, or whether they merely represent lines of 1 M. Levine, " Studies in the Cytology of the Hymenomycetes, especially the Boleti," Bull. Torrey Botanical Club, vol. xl, 1913, p. 173. 2 M. Levine, loc. cit., pp. 172-173. PSALLIOTA CAMPESTRIS 421 least resistance of which each nucleus can take advantage during its migration. In the Wild Mushroom, where most of the basidia are quadri- sporous, the passage of the four nuclei into the four spores probably takes place in the typical manner described above. In the Culti- vated Mushroom, according to Maire, two nuclei travel up each of the two sterigmata into the two spores. Further investigation is required to determine whether the tip of each of the two sterig- mata contains two centrosomes instead of the normal one, and whether or not, just before the nuclear migrations, there are two fibrillar strands running through each sterigma, each strand connecting a nucleus with the apex of the same spore. The future-generations basidia, dd (Fig. 146, p. 416), are distinguished from the present-generation basidia by not bearing sterigmata or spores, and from the coming-generation basidia by their relatively small size and by the fact that they are not yet protuberant. They are usually filled or nearly filled with fine proto- plasm, and their basidium-bodies, which are always shorter than those of the present-generation and coming-generation basidia, are relatively transparent. It is possible that the two nuclei with which each of these basidia is provided at its origin have not as yet fused, but this point awaits decision by means of a direct investigation. The future-generations basidia vary somewhat in size, and one is led to suspect that eventually the larger ones produce spores before the smaller. In the piece of hymenium represented in Fig. 146, the future-generations basidia are more numerous than those of any other class. This is associated with the fact that the hymenium is still young (it is supposed to have been shedding spores for only about 24 hours). The future-genera- tions basidia will develop spores during the next three or four days of the spore-discharge period, at the end of which time they will all have become past-generations basidia. It is probable that the total number of generations of basidia produced on any one small area of the hymenium is at least ten. The paraphyses, ee (Fig. 146, p. 416), are numerous small elements which are non-protuberant and largely hidden by the basidia, especially by the swollen ends of the bodies of the 422 RESEARCHES ON FUNGI present-generation and coming-generation basidia. It is not easy, at the stage of development of the hymenium represented in Fig. 146, to distinguish the paraphyses from some of the future-generations basidia, but they undoubtedly are present in about the numbers shown. As the hymenium grows older and more and more basidia shed their spores and collapse, the paraphyses increase in diameter ; and then they can be seen more easily. When the hymenium is quite exhausted, the paraphyses alone have a turgid appearance, and are then seen to be very numerous. A mushroom had been shedding spores for six days and six nights. A few hours after the cessation of spore-discharge, one of its gills was carefully examined with the microscope and it was found that nearly all the basidia had completely dissolved, so that their outlines could only be traced here and there ; but the paraphyses could all be clearly distinguished. A surface view of this totally exhausted hymenium is shown in Fig. 152, C (vide infra, p. 448). It has frequently been denied that paraphyses, i.e. constantly occurring permanently sterile elements, are present in the hymenium of the Hymenomycetes 1 ; but the denials have been made by cytologists who have been content to examine fixed and stained microtome sections. By comparing the state of the living hymenium of a single fruit-body day after day throughout the whole period of spore-discharge for Panaeolus campanulatus and Stropharia semi- globata, I proved the existence of paraphyses in those species.2 A similar investigation upon the Mushroom has yielded a similar result. I cannot entertain any further doubt that the hymenium of Psalliota campestris contains numerous paraphyses destined from their first origin to remain sterile. There does not seem to me to be the slightest reason to suppose that these elements could be caused to develop into basidia by any normal variation of any of the natural conditions under which mushrooms grow. Since paraphyses are definite elements in the hymenium of the Mushroom, one may ask : what is their function ? I am inclined to think that they serve the purpose of assisting the basidia which are to shed spores. They are the elastic elements of the hymenium. 1 Cf. Chap. X, pp. 269-270. 2 Vide Chap. X, pp. 279-281, and Chap. XI, pp. 336-337. PSALLIOTA CAMPESTRIS 423 As the hymenium gets older and as more and more basidia shed their spores and collapse, the paraphyses become larger and larger, and so fill up the gaps between the still living basidia. To these they probably minister by giving mechanical support and moisture, and possibly in other as yet unsuspected ways.1 In Fig. 146 (p. 416), a wave of development is passing across the hymenium from the left to the right in the directions shown by the arrows. This wave has already been dealt with, so far as the present-generation basidia are concerned, in the section which treats of the mottling of the gills (cf. Fig. 142, p. 407, which has the same present-generation basidia in its right two-thirds as Fig. 146). But here all the elements of the hymenium are represented, so that it is now possible to explain the nature of the wave in greater detail. The series of elements nos. 1, 2, 3, 4, 5, 6, and 7 in Fig. 146 are basidia in successively younger stages of development. No. 1 indicates two basidia with full-sized but unpigmented spores. The spores are supposed to be only one hour old, and therefore to have attained their full size only about twenty minutes ago. No. 2 indicates a basidium which bears spores which have not yet attained full size. These spores are only about thirty minutes old. The basidium no. 3 bears spores which have attained only one-third of their eventual diameter and are only about ten minutes old. The basidium no. 4 bears spores which have just begun their existence and are only about five minutes old. No. 5 points to two basidia which possess fully developed sterigmata but no spores. The basidium no. 6 has sterigmata which are just beginning to develop. The basidium no. 7 is as yet without even a trace of sterigmata. Similar series of basidia can be seen anywhere in the Figure passing from left to right in the direction of the arrows. Nos. 1 to 4 belong to the present-generation basidia of the light area, while nos. 5 to 7 belong to the coming generation of the dark area. Spore-discharge is taking place on the left-hand side of the dark area and is proceeding to the right-hand side in the direction of the arrows. The basidia s s discharged their spores about twenty minutes ago and are in the act of collapsing : their sterigmata are sinking 1 Cf. Chap. X, pp. 282-283. 424 RESEARCHES ON FUNGI into a concavity which is forming at the end of each basidium- body. Soon the sterigmata will melt down to stumps and then these basidia will resemble the other past-generations basidia, a a, scattered throughout the Figure. The basidium w is just about to discharge the second of its two spores, the first having already been shot away leaving behind a stiff but naked sterigma. As a preliminary to the discharge of the second spore, a drop of water has been excreted at the spore-hilum within the last ten seconds. The drop is almost of maximum size, which fact indicates that the spore should be violently propelled from its sterigma within the next two seconds. The spores on all the other basidia at the extreme left edge of the black area should be shot away within about half an hour. The successive discharge of all the spores on the black area, proceeding from the left to the right of the Figure, might take an hour and a half or two hours. If one could watch the gradual disappearance of the spores in this way, one would perceive that the basidia indicated by the series which bear the numbers 1 to 7 would be continuing their development, so that the area which is now dark would gradually again come to be spore-bearing, proceeding from left to right. Thus, for instance, the basidium no. 6 would be observed to begin developing its spores within a few minutes of the moment of discharge of the last of the six spores on the three present-generation basidia with which it is now sur- rounded. If we could watch the basidium no. 6 developing its spores, we should find that, during the early stages of its develop- ment, the spores on basidia nos. 1 and 2 would undergo pigmenta- tion, those on nos. 3 and 4 would grow to full size, and those on no. 5 would grow to nearly full size. It is thus clear that, as the wave of development passes across the piece of hymenium, a great many basidia are undergoing progressive and synchronous changes, all leading to the climax of spore-discharge. Everywhere through- out the finely mottled surface of the hymenium, on every gill, waves similar to the one just described are moving ; but the movements are in various directions. Sometimes a series of waves following one another in regular succession moves over several light and dark areas in a straight line in one direction ; but more often the waves clash, move apart from one another, or interfere PSALLIOTA CAMPESTRIS 425 with one another in endlessly varied ways. One can find out only empirically the directions of the waves on any gill-surface, and it is impossible to predict their course theoretically. Even the waves which are opposite to one another on the two sides of the same gill go their ways quite independently of one another. The very numerous small irregular waves of development which we find in all representatives of the Panaeolus Sub-type, including the Mushroom, may be contrasted with the single large wave which, in all species of the genus Coprinus, passes from below upwards on each gill. A black-spored panaeolate fungus with a many- waved hymenium on each gill may well have been the ancestor of the genus Coprinus in which the hymenium on each gill is one-waved, the many small irregular waves having become reduced in the course of evolution to one grand wave always moving in an upward direction. We shall now proceed to a study of the hymenium of the Cul- tivated Mushroom in cross-section, so that we may gain an insight into the structure and mutual relations of the hymenial elements when these are seen in lateral view. Fig. 147 represents a cross- section of part of a living gill and shows semi-diagrammatically the structure of the hymenium h, the subhymenium s, and part of the trama t. The drawing is essentially synthetic and exhibits all the histological details which were made out by the study of numerous particular sections cut with a hand-razor from living gills and mounted in water. Parts of some of these sections were drawn with the camera lucida. Similar sections for the wild form of the Mushroom will be represented in subsequent illustrations. The gill in Fig. 147 is supposed to be taken from a mushroom which has been shedding spores for about 24 hours and which therefore will continue to shed spores for several more days, i.e. until the end of the spore-discharge period, which terminates only with the complete exhaustion of the hymenium. The stage of development of the hymenium is exactly the same as that for the hymenium represented in surface view in Fig. 146 (p. 416). In Fig. 147, a dark area of the hymenium is passed through pro- ceeding to the right from element no. 8 to element no. 34, a light area from element no. 34 to element no. 59, an intermediate area from 426 RESEARCHES ON FUNGI element no. 59 to element no. 63, and another intermediate area from element no. 1 to element no. 8. As indicated by the shading, the spores in the light area are unpigmented, in the intermediate areas partially pigmented, and in the dark area fully pigmented. The hymenial elements can all be classified as basidia or para- physes as follows : past-generations basidia, nos. 13, 21, 30, 38, and 51 ; present-generation basidia, nos. 3, 5, 11, 16, 24, 32, 41, 44, 48, 53, 57, and 61 ; coming-generation basidia, nos. 8, 18, 28, 36, 42, 46, 55, and 60 ; future-generations basidia, nos. 2, 6, 7, 9, 12, 14, 17, 19, 23, 27, 29, 31, 34, 37, 45, 49, 52, 54, 59, and 62 ; paraphyses (permanently sterile elements), nos. 1, 4, 10, 15, 20, 22, 25, 26, 33, 35, 39, 40, 43, 47, 50, 56, 58, and 63. The past-generations basidia have shed their spores and are all dead. They were once as protuberant as the present-generation basidia but their ends are now drawn down to the general level of the hymenium. The free end of each past-generation basidium appears flattened in side view, but is in reality concave, the concavity concealing the sterigmatic stumps. The collapsed shafts are often somewhat brownish, a discoloration which aids one in the not too easy task of identifying them. A basidium which fails for any reason to discharge its spores at the proper time, on collapsing, draws down its spores to the general level of the hymenium, so that they cannot interfere mechanically with the new spores developing on the sterigmata of neighbouring basidia. Two such undischarged or wasted spores which are now lying on, and adhering to, the hymenium, are shown in Fig. 147 at w w. Each of them was dragged down to its place of rest during the collapse of a past-generation basidium. The waste spore, w, on the right of the illustration, is supposed to have been produced by the element no. 51, but the basidium which produced the other waste spore is not shown. The present-generation basidia all bear sterigmata and spores. Their bodies protrude above the general level of the hymenium to a distance of about 0-005 mm. The spores. are raised up on sterig- mata, where, during their development and discharge, they cannot be interfered with mechanically either by the wasted spores or by the tops of neighbouring younger basidia. The age of the spores PSALLIOTA CAMPESTRIS 427 varies from about five minutes in no. 41 to about eight hours in nos. 24 and 32. Spores which are of full size but colourless are from forty minutes to two hours and forty minutes old (nos. 53 and 57). Spores which are undergoing pigmentation are from two hours forty minutes to four hours and forty minutes old (nos. 61, 3, and 5). Spores which are fully pigmented are from four hours and forty minutes to eight hours old (nos. 11, 16, 24, 32). These ages are approximations based on studies already recorded. The contents of the basidia are not shown but, if they were, we should see that the bodies of nos. 41, 44, and 48 were practically full of protoplasm and that the bodies of nos. 24 and 32 were empty. We should observe stages in the progressive emptying of the basidium-contents into the spores in nos. 53, 57, 61, 3, 5, 11, 16, and 24 : we should note the origin and increase in size of a basal vacuole in the basidium-body, and its extension upwards until it filled not only the whole basidium-body but also the sterigmata except perhaps their tips. Most of the basidia are bisporous, but two of them, nos. 48 and 57, are monosporous. The solitary sterigma of each of these is situated in line with the axis of the basidium-body. The spores of the monosporous basidia develop unilaterally on the ends of their sterigmata, and each possesses the usual hilum ; but they differ from the spores of the bisporous basidia in size. If one compares the spore on no. 57 with the spores on nos. 11, 53, or 61, one perceives that the former is distinctly longer, somewhat broader, and of about twice the volume of the latter. Nos. 3 and 5 show the typical arrangement of the spores on any two very closely adjacent basidia which develop their spores simultaneously : the four spores are situated at the corners of a square or rectangle (cf. Fig. 146, lower right-hand corner at b, p. 416). This prevents any jostling of the spores during their development and discharge. The positions of the spores are de- cided by the positions of the sterigmata at their first origin ; and, as we have seen, the positions of the sterigmata appear to be decided by the positions which the centrosomes of the daughter nuclei in the basidium-body take up after their formation.1 It seems probable, therefore, that the position of the spores in such 1 Cf. p. 420. FIG. 147. — Psalliota campestris (cultivated form). Cross-section of part of a living gill showing semi- diagrammatically the structure of the hymenium h, the subhymenium 8, and part of the trama t. The gill is supposed to have been taken from a mushroom which has been shedding spores for about 24 hours and the hymenium of which will therefore not be totally exhausted for several days. A dark area of the hymenial surface is passed through proceeding to the right from element no. 8 to element no. 34, a light area from element no. 34 to element no. 59, an intermediate area from element no. 59 to element no. 63, and another intermediate area from element no. 1 to element no. 8. As shown by the shading, in the light area the spores are unpigmented, in the intermediate areas partially pigmented, and in the dark area fully pigmented. The elements of the hymenium can be analysed as follows : past -generations basidia, nos. 13, 21, 30, 38, and 51 ; present-generation basidia, nos. 3, 5, 11, 16, 24, 32, 41, 44, 48, 53, 57, and 61 ; coming-generation basidia, nos. 8, 18, 28, 36, 42, 46, 55, and 60 ; future-generations basidia, nos. 2, 6, 7, 9, 12, 14, 17, 19, 23, 27, 29, 31, 34, 37, 45, 49, 52, 54, 59, and 62 ; paraphyses (permanently sterile elements which enlarge as the hymenium becomes more and more exhausted), nos. 1, 4, 10, 15, 20, 22, 25, 26, 33, 35, 39, 40, 43, 47, 50, 56, 58, and 63. Parts of three waves of development are shown proceeding across the section from right to left. The rear end (about three-ninths) of the foremost wave consists of the series of basidia, nos. 51, 38, 32, 24, 16, 11, 5, and 3. The middle part (about four-ninths) of the next succeeding wave consists of the series of basidia nos. 61, 57, 53, 48, 44, 41, 36, 28, 18, and 8. The forepart (about two- ninths) of the third wave which succeeds the second, consists of the series of basidia nos. 60, 55, 46, 42, and 34. There is a difference of about eight and one-half hours between each basidium of one wave and its nearest neighbour of the preceding wave. Thus after eight and one -half hours the basidium no. 8 will approximately resemble no. 5 as it is now, no. 18 will resemble no. 16, no. 28 will resemble no. 24, etc. The basidia of the rear part of the first wave will now be described. No. 51 has been in the collapsed condition for an hour or more ; its concave end, drawn down to the general level of the hymenium, conceals the sterigmatic stumps. No. 38 is in the act of sudden collapse about twenty minutes after the discharge of its last spore ; its end is rapidly becoming concave, and its sterigmata are being drawn into the concavity. No. 32 bears a spore which has a drop about ten seconds old and of the maximum size on its hilum and which therefore should be shot away at once. The other spore was shot away from its sterigma about a minute ago, in the direction indicated by the arrow, to a distance of about 0 • 1 mm. This spore is shown in its flight with the water-drop attached to it. No. 24 is about to shoot away its spores, for a water-drop is being rapidly excreted at each spore-hilum. The right-hand drop is about eight seconds old and of nearly maximum size which indicates that the spore to which it is attached should be shot away after the lapse of about two more seconds. The other drop is only about two seconds old and therefore will con- tinue to grow for about eight more seconds. The spores on both no. 32 and no. 24 are about eight hours old. No. 16 bears fully pigmented spores which we may suppose to be six and one- half hours old and therefore an hour and a half from the moment of spore-discharge. No. 11 also has fully pigmented spores. These we may suppose to be five and a half hours old and therefore two and a half hours from the moment of spore-discharge. Nos. 5 and 3 may be supposed to have begun their development at the same moment. Their spores are becoming strongly pigmented, and we may conclude that they are about four hours old and four hours from discharge. The basidia of the middle part of the second wave will now be described. No. 61 bears spores which are becoming pigmented. We may suppose them to be three and a half hours old. No. 57 bears a single sterigma and a spore of full size but unpigmented. We may suppose the spore to be two hours old. No. 53 bears two spores of full size but unpigmented. We may suppose them to be one hour old, so that they attained full size only twenty minutes ago. Nos. 48 and 44 bear spores which are far below full size and about ten minutes old. The spores of no. 41 are only five minutes old. No. 36 bears full-grown sterigmata but no spores as yet. It resembles in outward form a basidium which has just shot away its second spore, i.e. such a basidium as no. 32 will be like within two seconds when the left-hand spore has been discharged, but differs from it in being stuffed with contents instead of having its interior occupied by a large vacuole. No. 28 bears sterigmata about two-thirds developed and no. 18 sterigmata about one-third developed. No. 8 shows as yet not a trace of sterigmata. The basidia of the forepart of the third wave will now be described. No. 60 is almost, if not quite, of full length. No. 55 has certainly not yet attained its full length, for it is not yet fully protuberant. Nos. 46 and 42 are progressively younger and less protuberant. No. 34, which forms the foremost element of the wave, is simply a future -generation basidium which does not yet protrude beyond the general level of the hymenium. Most of the present -generation basidia are bisporous, but nos. 48 and 57 are monosporous. The spore on no. 57 has about twice the volume of any spore on a bisporous basidium. Nos. 3 and 5 show the typical arrangement of the spores on any two very closely adjacent basidia which happen to develop their spores simultaneously : the four spores are situated at the four corners of a square or rectangle. Two waste spores, w w, are shown lying on the hymenium ; the basidia to which they belonged failed to discharge them and they were therefore dragged down on to the hymenium when these basidia collapsed. We may suppose that the basidium no. 51 shot away one of its spores successfully but not the other ; hence the waste spore adjacent to its free end. The subhymenium s consists of several layers of rounded or oval cells. There is no sharply defined level plane separating the hymenium from the subhymenium. The older basidia (the first to develop spores) arise very deeply in the subhymenium, e.g. nos. 30 and 51 ; but nos. 5, 11, 13, and 38 are also noteworthy in this respect. The basidia which successively come to maturity during the several days of spore-discharge arise more and more toward the top of the hymenium and have shorter and shorter bodies. The trama t consists of elongated cells forming irregularly anastomosing chains which tend to run in a direction parallel to the median line of the cross-section of the gill. Magnification, 1,060. The scale is divided into hundredths of a mm. O 0-01 0-02 0-03 0-04 0-Oi 0-06 0-07 0-08 mi 430 RESEARCHES ON FUNGI a pair of basidia as 3 and 5 is decided by stimuli which affect the nuclei. Exactly how these stimuli operate is at present a matter for speculation. It may be that mutual stimuli are conducted from one basidium to the other by means of protoplasmic con- nections passing through one or more cells of the subhymenium ; or, possibly, stimuli are provided by the mutual pressure of the basidia, by a gaseous emanation, or by transpiration. Possibly some day the matter may be decided by performing minute operations on the hymenium with the aid of instruments for micro- dissection such as have been used by Kyte. The mutual adjust- ment of the positions of the spores on neighbouring basidia only takes place when the basidia producing the spores simultaneously are in contact or nearly in contact ; but basidia which are relatively distant from one another, e.g. nos. 11 and 16, do not seem to in- fluence one another in any way, and a prediction of the probable relative positions of their spores is impossible. The corners- of-a-square arrangement of the spores of two adjoining bisporous basidia of a Cultivated Mushroom exactly resembles the corners- of -a -square arrangement of the four spores of a quadrisporous basidium of the Wild Mushroom. Two adjoining bisporous basidia, therefore, appear to respond to the stimulus which decides the posi- tions of the spores like the two halves of a quadrisporous basidium. The coming -generation basidia (nos. 8, 18, 28, 36, 42, 46, 55, and 60) are destined to produce the next crop of spores after the spores on the sterigmata of the present -generation basidia have been discharged. Like the present -generation basidia they are distinctly protuberant, but they differ from these in that they do not bear spores. They develop sterigmata in the later stages of their growth. Nos. 42, 46, and 55 are not yet fully protuberant, as they are relatively young : they are associated with present- generation basidia which bear very rudimentary spores. On the other hand, nos. 60, 8, 18, 28, and 36 are fully protuberant, for their bodies have attained maximum length : they are associated with present-generation basidia which bear relatively mature spores. The cell-contents are not shown but, if they were, we should see that the body of each coming-generation basidium was filled, or almost completely filled, with fine protoplasm. PSALLIOTA CAMPESTRIS 431 The future-generations basidia (nos. 2, 6, 7, 9, 12, 14, 17, 19, 23, 27, 29, 31, 34, 37, 45, 49, 52, 54, 59, and 62) are smaller than those of the coming-generation basidia and never bear sterigmata. They are not yet protuberant ; indeed, it is their free ends, together with those of the past-generation basidia and of many of the para- physes, which make up what I have often referred to as the general level of the hymenium, above which the bodies of the basidia of the present-generation and coming-generation basidia project by about 0-005 mm. The basidium-bodies are filled with fine protoplasmic contents. Since the hymenium represented in Fig. 147 is sup- posed to have been discharging spores for only 24 hours and will discharge further spores for some four or five days, the future- generations basidia still outnumber those of the past generations. The paraphyses (nos. 1, 4, 10, 15, 20, 22, 25, 26, 33, 35, 39, 40, 43, 47, 50, 56, 58, and 63) are permanently sterile : they never produce sterigmata or spores. They are usually attached to the uppermost cells of the subhymenium and are the shortest of the hy menial elements. At the stage of the development of the hymenium shown in Fig. 147 (p. 429), they have not yet grown to their maximum size. As more and more generations of basidia shed their spores and collapse, the paraphyses grow larger and larger in diameter, so that the lateral space which they occupy in the hymenium is progressively increased. Drawings of older swollen paraphyses will be given in a subsequent illustration. The paraphyses in the hymenium represented in Fig. 147 are to a large extent covered laterally by the bodies of the basidia. Owing to their small size in the young hymenium and to the fact that they are largely hidden by the basidia, they are often difficult to make out clearly. At first they contain a large amount of protoplasm in which one can usually detect one or more vacuoles ; but, as they grow older and enlarge, their vacuoles increase in size and their massive protoplasm gradually disappears, so that, finally, in late stages of the development of the hymenium, the lumen of each paraphysis comes to be occupied by one large vacuole. Parts of three waves of development are shown in Fig. 147 (p. 429) proceeding across the section from right to left. The rear end (about three-ninths) of the earliest wave consists of the series of 432 RESEARCHES ON FUNGI basidia nos. 51, 38, 32, 24, 16, 11, 5, and 3. The middle part (about four-ninths) of the next succeeding wave consists of the series of basidia nos. 61, 57, 53, 48, 44, 41, 36, 28, 18, and 8. The fore part (about two-ninths) of the third wave, which succeeds the second, consists of the series of basidia nos. 60, 55, 46, 42, 34, and 31. The stretch of hymenium shown in the illustration is unfortunately too short to show the whole length of any one wave ; but a whole wave can easily be imagined if we suppose that the illustration is enlarged by adding to its left side in succession two copies of it. With this arrangement we should have the following set of basidia making a complete wave : nos. 51 (the hindmost element), 38 (the wave declining), 32 (crest of the wave : spore-discharge), 24 (beginning of the fore part of the wave), 16, 11, 5, 3, 61, 57, 53, 48, 44, 41, 36, 28, 18, 8, 60, 55, 46, 42, 34, and 31 (the foremost element of the wave). There is a difference in development of about eight and one-half hours between each basidium of the second of the three waves and its nearest neighbour of the first wave. Thus, for instance, after about eight and one-half hours, the basidium no. 28 will resemble no. 24 as it is now, no. 18 will resemble no. 16, and no. 8 will resemble no. 5. There is a similar difference between the third wave and the second. Thus, after about eight and one-half hours, the basidium no. 55 will resemble either no. 53 or no. 57, and no. 46 will resemble either no. 44 or 48. The basidia of the rear part of the first wave will now be described. No. 51 has been in the collapsed condition for an hour or more ; its concave end, drawn down to the general level of the hymenium, conceals the sterigmatic stumps. No. 38 is in the act of sudden collapse about 20 minutes after the discharge of its last spore ; its end is rapidly becoming concave and the sterigmata are being drawn into the concavity. No. 32 bears a spore which has a drop about 10 seconds old and of the maximum size on its hilum and which, therefore, should be shot away at once. The other spore was shot away from its sterigma about a minute ago, in the direction indicated by the arrow, to a distance of about 0-1 mm. This spore is shown in its flight with the water-drop attached to it. No. 24 is about to shoot away its spores, for a water-drop is being rapidly excreted at each spore-hilum. The PSALLIOTA CAMPESTRIS 433 right-hand drop is about eight seconds old and of nearly maximum size, which indicates that the spore to which it is attached should be shot away after the lapse of about two more seconds. The other drop is only about two seconds old and therefore will continue to grow for about eight more seconds. The spores on both no. 32 and no. 24 are about eight hours old. No. 16 bears fully pig- mented spores, which we may suppose to be six and one-half hours old and therefore an hour and a half from the moment of spore- discharge. No. 11 also has fully pigmented spores. These we may suppose to be five and a half hours old and therefore two and a half hours from the moment of spore-discharge. Nos. 5 and 3 may be supposed to have begun their development at the same moment. Their spores are becoming strongly pigmented, and we may there- fore suppose that they are about four hours old and four hours from the moment of spore-discharge. The basidia of the middle part of the second wave will now be described. No. 61 bears spores which are becoming pigmented. We may suppose them to be three and a half hours old. No. 57 bears a single sterigma and a spore of full size but unpigmented. We may suppose the spore to be two hours old. No. 53 bears two spores of full size but unpigmented. We may suppose them to be one hour old, so that they attained full size only twenty minutes ago. Nos. 48 and 44 bear spores which are far below full size and about ten minutes old. The spores on no. 41 are only five minutes old. No. 36 bears full-grown sterigmata but no spores as yet. It resembles hi outward form a basidium which has just shot away its second spore, i.e. such a basidium as no. 32 will be within two seconds when the left-hand spore has been discharged. But one can easily distinguish between a basidium which is just about to develop its spores and one which has just discharged them, for there is a marked difference in their cell- contents. The former has a basidium-body filled or almost filled with protoplasm, while the latter has a basidium-body occupied by one large vacuole. No. 28 bears sterigmata about two-thirds developed, and no. 18 sterigmata about one-third developed. Finally, no. 8 shows as yet not even a trace of sterigmata. We shall complete the account of the basidia entering into the VOL. n. 2 F 434 RESEARCHES ON FUNGI three hymenial waves by describing the basidia of the fore part of the third wave. No. 60 is almost, if not quite, of full length. No. 55 has "certainly not attained its full length and is therefore not yet fully protuberant. Nos. 46, 42, and 34 are progressively younger and less protuberant. No. 31 which forms the foremost element hi the wave is simply a future-generation basidium, which does not yet protrude beyond the general level of the hymenium. The subhymenium $ consists of some three or four layers of rounded or oval cells. There is no sharp and level plane separating the hymenium from the subhymenium, such as many writers have imagined. Some of the basidia, namely, the oldest (the earliest to develop spores), arise very deeply in the subhymenium, either toward its middle or sometimes even lower. In Fig. 147 (p. 429), the most striking basidia, so far as depth of origin is concerned, are nos. 30 and 51, but nos. 5, 11, 13, and 38 are also noteworthy in this respect. The basidia which successively come to maturity during the several days of the spore-discharge period arise more and more toward the top of the subhymenium. The paraphyses are situated quite superficially : the small subhymenial cells from which so many of them arise appear to be almost pedicellar in their nature (cf. the cells beneath the paraphyses nos. 1, 4, 10, 25, 26, 33, etc.). The trama t consists of large elongated cells attached together in chains which are irregularly anastomosing and which tend to run in a direction parallel to the median line of a cross-section of the gill. Sometimes there are a few narrow elongated hyphae just beneath the subhymenium where the trama begins. Some have been represented in Fig. 147. At the stage of development shown in Fig. 147 (p. 429), the hymenium contains a great deal of protoplasm and accumulated food-stuffs. Cell-contents are particularly heavy in the younger present-generation basidia, in the coming-generation and in the future-generations basidia. There is a less amount of protoplasm in the paraphyses. On account of the presence of heavy cell- contents, the hymenium appears very dense even in thin sections, which fact constitutes one of the difficulties in making out the cell-outlines. The subhymenium also contains a large quantity of PSALLIOTA CAMPESTRIS 435 protoplasm, and a less amount is present in the cells of the trama. Part of this protoplasm and its inclusions, such as glycogen, etc., is doubtless used up in the growth of the basidia in length and of the paraphyses in breadth, and part of it is necessary for the formation of the sterigmata and the spore-walls. Part of it must also be used up in processes of destructive metabolism, including respiration. But there can be little doubt that most of it is destined to pass through the narrow sterigmatic necks and to be crowded into the spores, there to await the important process of germination and the construction of germ-tubes. As more and more generations of basidia shed their spores, the hymenium, sub- hymenium, and trama become more and more emptied of proto=- plasm, until, when the process of spore-discharge ceases, the cells are all occupied by large vacuoles. The contrast in density between a section of a gill which has just begun to shed spores, and of a gill which has just ceased to shed spores, is very striking. No doubt, chemical analysis of two such sections would show a very marked difference in proteins and other cell-contents. There is every reason to suppose that a mushroom must be much more nutritious when it is young than when it is aged, for the most desirable food substances which it contains are gradually carried away from the pileus by the spores during the spore-discharge period. The Hymenium of the Wild Mushroom. — Up to the present we have confined our analysis of the hymenium to the cultivated form of Psalliota campestris. We shall now turn to the wild form, the hymenium of which is similar in all essentials of its organisa- tion to that of the cultivated form, although it exhibits some interesting differences in detail. The main points of difference are : (1) the "Wild Mushroom, typically, has quadrisporous basidia and the Cultivated Mushroom bisporous ones, (2) the basidia of the Wild Mushroom are distinctly larger than those of the Cultivated, and they produce longer spores (cf. Figs. 148 and 149 with Fig. 147, p. 429). The cross-section of the Cultivated Mushroom represented in Fig. 147 (p. 429), which has just been described, is a synthetic one, embodying as many morphological details as possible ; but it is unaccompanied with drawings of any of the special sections 436 RESEARCHES ON FUNGI upon which it was based. In the case of the Wild Mushroom, on the other hand, no synthetic drawings will be given, and we shall confine our attention to a set of special sections from which a synthetic illustration might be constructed, were such required. Our plan will be to study a series of such sections which show the hymenium in : (1) an early stage of its activity, (2) an almost exhausted stage, and (3) a completely exhausted stage. These sections were all cut with a hand-razor from living gills and mounted in water, after which the cells were drawn with the camera lucida. The illustrations representing these sections on subsequent pages of this Chapter are the first of their kind to be published, and I shall not hesitate, therefore, to tell the secret of my success in making them. I used perfectly fresh mushrooms, cut sections through the thickest part of a gill with a sharp razor held in the hand, examined the sections in the living state immedi- ately after they had been mounted in water, and interpreted the sections in the light of the knowledge and experience which I had gained during my study of Panaeolus campanulatus and Stropharia semiglobata. The microscope employed was a Zeiss, eyepiece no. 4 and objective FF. The original camera-lucida sketches were made with a magnification of 1,040 diameters and then photo- graphed with a lantern-slide-making apparatus. The negatives so obtained were then put into a lantern and projected on a sheet of white paper, so that the magnification- of the original drawings was increased from 1,040 diameters to 1,560. The projections were then traced out on the paper with a pencil. The tracings so obtained were then made into black-and-white drawings with ink, and these drawings, after reduction to two-thirds by the block- maker, are reproduced as Figs. 148-152. The illustrations of the special sections, about to be described, not only show the general succession of basidia but exhibit for the first time, so far as the Mushroom is concerned : (1) collapsed past-generations basidia which hitherto have been overlooked, (2) paraphyses which never produce spores and are destined from the first to remain sterile, and (3) the true relations of the hymenial elements with the subhymenium. Camera-lucida Studies of the Young Hymenium. —In Fig. 148 PSALLIOTA CAMPESTRIS 437 is represented part of a cross-section of a living gill of a fruit- body which had just expanded and had only very lately commenced to shed spores. The past-generation basidia were so few in number that none happened to be included in the section. In cutting the section, some of the spores on the basidia nos. 1, 2, and 3 were either displaced or knocked off their sterigmata, while the riper spores on nos. 4 and 5 were all knocked off. The details of the spores and sterigmata of these five basidia have been adjusted semi- diagrammatically ; but all the other cells are shown as observed. It may be remarked that nearly ripe spores are very easily detached from their sterigmata by the pressure of surface tension when the section to which they belong is irrigated with water ; and the premature detachment of the riper spores is probably a mutilation characteristic of all preparations mounted in fluids. Histologists should beware of being deceived by the supposition that a hymenium, after being fixed and stained and cut into sections with the microtome, necessarily exhibits all the delicate morphological relations which the living hymenium possesses in nature. The section shown in Fig. 148, although not including any past- generation basidia, includes five present-generation basidia, nos. 1, 2, 3, 4, and 5. These, as indicated by the degrees of pigmentation of the spores, form part of a wave of development progressing from the right-hand to the left-hand side of the section. Two coming- generation basidia, nos. 6 and 7, have both attained about their maximum protuberancy and, doubtless, are preparing for the pro- duction of sterigmata. The future-generations basidia are nos. 8, 9, 10, 11, 12, 13, 14, 15, 16, and 17. Their free ends, which are not yet protuberant, constitute the general level of the hymenium. The permanently sterile elements or paraphyses are nos. 18, 19, 20, 21, and probably no. 22. But how, it may be asked, does one really know that these elements are paraphyses ? Why should they not be very young basidia which subsequently will produce spores ? These questions are certainly pertinent and demand an answer. Now a study of the progressive exhaustion of the hymenium and of the completely exhausted hymenium proves (vide, infra) that a large number of elements in the hymenium never produce spores but gradually enlarge in diameter, as the hymenium gets 438 RESEARCHES ON FUNGI older, without growing appreciably in length. In the exhausted hymenium these cells are found to be short elements which have a superficial origin from the subhymenium. But, in the section we FIG. 148. — Psalliota campestris (wild form). Cross-section of a living gill of a fruit -body just expanded, drawn with a camera lucida. In cutting the section some of the spores on the basidia nos. 1, 2, and 3, were either displaced or knocked off their sterigmata, while the riper spores on nos. 4 and 5 were all knocked off. The details of the spores and sterigmata of these five basidia have been adjusted semi-diagrammatically. All the other cells are shown as observed. Special objects of jdrawing : to show (1) presence of paraphyses and (2) relations of hymenium with subhymenium. h, the hymenium ; s, the subhymenium ; and t, part of the trama. Nos. 1, 2, 3, 4, and 5, present -generation basidia. Nos. 6 and 7, coming-generation basidia. Nos. 18, 19, 20, 21, and probably 22, are paraphyses. The elements nos. 4,' 7, 21, etc., arise at diverse levels of the subhymenium. Magnification, 1,040. are discussing, the elements nos. 18, 19, 20, 21, and 22 are the shortest elements ; of all the elements present they resemble basidia least ; moreover, they are seated on the outermost sub- hy menial cells. Their contents are already provided with large vacuoles (although this is not indicated in the illustration), as if they were already beginning to empty themselves, and were thus PSALLIOTA CAMPESTRIS 439 coming to resemble the paraphyses with clear lumina which cer- tainly occur in a fully exhausted hymenium. All the other cells in the section are too long to be paraphyses, as will be seen by comparing Figs. 148 and 152, A (p. 448). The process of elimina- tion, by itself, therefore, leads to the conclusion that the elements nos. 18, 19, 20, 21, and 22 are paraphyses. This conclusion is also supported by comparative studies of other hymenia such as those of Panaeolus campanulatus and Stropharia semiglobata, where the elements are much larger and more easily differentiated from one another. A study of Fig. 148 shows that there is no sharp and level plane dividing the hymenium from the subhymenium, for the bases of the hymenial elements are situated at very various levels in the subhymenium. In this regard the basidium no. 4 is noteworthy, for it arises at about the middle depth of the subhymenium. On the other hand, basidia nos. 13 and 7 arise at a higher level, and the paraphyses nos. 20 and 21 at a still higher level. The large subhy menial cell which has given rise to the basidium no. 6 has also produced a small subhymenial cell which acts as a pedicel for the paraphysis no. 20. A similar relation appertains for the basidium no. 7 and the paraphysis no. 21. The same subhymenial cell no doubt often produces at least two and possibly more basidia, but no good instance of this is shown in this particular illustration. The section shown in Fig. 149 was cut from the same mush- room and at the same time as the one which has just been described. The spores on basidium no. 1 were bent out of their normal positions in cutting the section and have been readjusted semi-diagrammatically. The other cells are all drawn as observed. Here again, there are no past-generations basidia to be seen : one might have been included in the section but was not. Nos. 1, 2, and 3 are present-generation basidia, evidently bearing very young spores, for the spores on nos. 2 and 3 are only partly grown, while those on no. 1, although of full size, are as yet unpigmented. Nos. 4, 5, and 6 are coming-generation basidia, a fact indicated by their protuberancy which is greatest in no. 6. Basidia nos. 1, 2, 3, and 6 appear to form part of a wave of development which is passing across the section from left to right. Nos. 7, 8, 9, 10, 11, and 12 440 RESEARCHES ON FUNGI are future-generations basidia, while nos. 13, 14, 15, 16, and 17 from their size, position, and contents appear to be paraphyses. fcn c... FIG. 149. — Psalliota campestris (wild form). Cross- section of a living gill of a just expanded fruit-body, drawn with the camera lucida. The spores on basidium No. 1 were bent out of their normal positions and have been adjusted serni-diagrammatically. To show (1) paraphyses, and (2) relations of hymenium and subhymenium. h, the hymenium ; .9, the subhymenium ; t, part of the trama. Nos. 1, 2, and 3, present-generation basidia. Nos. 4, 5, and 6, coming-generation basidia. Nos. 7, 8, 9, 10, 11, and 12, basidia of future generations. Nos. 13, 14, 15, 16, and 17, paraphyses. From the uppermost layer of the subhymenium, cells I, II, III, IV, and V, arise paraphyses or occasionally, as from cell 1, a basidium ; from the second layer, cells VI, VII, VIII, IX, X, XI, XII, and XIII, arise either basidia or cells supporting paraphyses, or occasionally, as from cell XI, a paraphysis ; from the lowest layer, cells XIV, XV, XVI, XVII, XVIII, and XIX, arise sometimes a basidium as at XVIII or, more fre- quently, cells of the second layer of the subhymenium. Magnification, 1,040. The subhymenium can be divided roughly into three layers. The highest layer, cells nos. I, II, III, IV, and V, chiefly give rise to paraphyses ; from the second layer, ceUs VI, VII, VIII, IX, X, XI, XII, and XIII, arise occasionally a paraphysis as from no. XI, PSALLIOTA CAMPESTRIS 441 but more usually either basidia or cells supporting paraphyses ; from the lowest layer, cells nos. XIV, XV, XVI, XVII, XVIII, and XIX, arise sometimes a basidium as at XVIII, but more frequently cells of the second layer of the subhymenium. Here again, it is clear that there is no level plane separating the hymenium from the subhymenium. The basidium no. 3 arises from one of the lowest cells of the subhymenium, whereas the basidia nos. 11 and 5, etc., arise at a higher level, and the para- physes nos. 14 and 16, etc., at a still higher level. The oldest basidia, i.e. those which produce spores first, are as a rule the longest, owing to their depth of origin in the subhymenium. Basidia of later generations tend to arise more and more super- ficially, so that their shafts become shorter and shorter. In this particular section, the basidia nos. 1, 2, and 3, which alone are producing spores, are the most deep-seated so far as their origin is concerned. Between the subhymenium and the swollen cylindrical cells of the trama, there are often placed a certain number of thin elongated hyphae. One of these happens to be present in Fig. 149. The last of the sections of the very young hymenium, cut at the same time as the two just described, is represented in Fig. 150. In cutting and mounting the section, all the spores on basidia nos. 1, 2, 3, 4, and 5 were knocked off their sterigmata except the one shown on no. 1. Probably the spores on all these basidia were nearly ripe and, on this account, easily detachable. None of the spores accidentally removed are restored in this Figure, so that the present-generation basidia are all represented in their mutilated condition. It may be asked : is it not possible that the basidia nos. 2, 3, 4, and 5 are really basidia which have just shot away their spores ? The answer is no. These basidia all contained a certain amount of massive protoplasm in their upper ends (not shown in the illustration) which was doubtless still being passed into the spores ; but such protoplasm is not contained in basidia which have just shot away their spores. Moreover, basidia which have shot away their spores usually collapse about twenty minutes after the discharge of the last spore. It is most unlikely that four basidia, which have discharged all their spores and are all about 442 RESEARCHES ON FUNGI to collapse together, should be contained in one small section such as is here represented. There is only one coming-generation basidium, namely, no. 6. Others, doubtless, were only just missed when the section was cut. The future-generations basidia are nos. 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, and 18. The paraphyses are nos. 19, 20, 21, and 22, and possibly nos. 7 and 9 ; but I have assumed that nos. 7 and 9 are very short basidia on account of their being already so broad in an early stage of the development of the hymenium. The contents of nos. 7 and 9, unfortunately, were not observed. It is likely that there were paraphyses attached to the subhymenial cells nos. V and VI, but that they were torn away in cutting the section. Other paraphyses, resembling no. 19 with its subhymenial cell no. Ill and no. 21 with its subhymenial cell no. IV, may have been just excluded from the section. One such set of cells would fit in nicely in front of basidium no. 13, from which position it may well have been removed. Once more we can notice the absence of any level plane separating hymenium and sub hymenium. The oldest basidia, or at least some of the oldest, are the longest of all. Nos. 2 and 4 are remarkable for the depth to which their shafts descend to attach themselves to subhymenial cells of the lowest layer. The basidium no. 2 arises from the cell no. XXIII, and no. 4 from no. XXIX ; and the lower part of the basidial shaft of each is completely enveloped by subhymenial cells. With the position of origin of basidium no. 2 should be com- pared the positions of origin for elements nos. 10, 9, and 19 ; and with the position of origin of basidium no. 4 should be compared the positions of origin for elements nos. 15, 16, and 21. The sub- hymenium consists of rounded or oval cells, and can be divided very roughly into four rows of cells : I to VI inclusive, VII to XXI inclusive, XXII to XXXIII inclusive, and a fourth layer of a few cells only, XXXIV to XXXVI inclusive. The paraphyses are always superficial in their origin, but basidia arise at various levels of the hymenium, the older ones tending to be most deeply seated and the younger ones to be situated more and more toward the exterior. A little more of the trama is shown in Fig. 150 than in the two preceding Figures. It will be noticed that the larger tramal cells, PSALLIOTA CAMPESTRIS 443 which tend to form cell-chains, greatly exceed in size even the largest elements of the hymenium and subhymenium. FIG. 150. — Psalliota campestris (wild form). Cross-section of a living gill of a fruit-body just expanded, drawn with the camera lucida. In cutting the section all the spores on the basidia nos. 1, 2, 3, 4, and 5, were knocked off their sterigmata except the one shown on basidium no. 1. The extreme top and bottom of basidium no. 6 were not drawn owing to an oversight, and have been completed semi-diagrammatically ; otherwise all elements are repre- sented as seen. Chief object of drawing : to show the relations of the hymenium with the subhymenium. h, the hymenium ; s, the subhymenium ; t, part of the trama. Nos. 1, 2, 3, 4, and 5, present-generation basidia. No. 6, a coming- generation basidium. Nos. 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, and 18, future- generations basidia, possibly with the exception of nos. 7 and 9 which may be paraphyses. Nos. 19, 20, 21, and 22, and possibly also nos. 7 and 9, are para- physes. The subhymenium consists roughly of four layers of cells (reading from left to right) : the highest layer, usually supporting paraphyses, I, II, III, IV, V, and VI ; a second layer, VII, VIII, IX, X, XI, XII, XIII, XIV, XV, XVI, XVII, XVIII, XIX, XX, and XXI ; a third layer, XXII, XXIII, XXIV, XXV, XXVI, XXVII, XXVIII, XXIX, XXX, XXXI, XXXII, and XXXIII ; a fourth layer of a few cells only, XXXIV, XXXV, and XXXVI. Some of the basidia arise deep down in the subhymenium, e.g. nos. 2 and 4 which spring from the cells XXIII and XXIX respectively : the bases of these basidia are enveloped by upper subhymenial cells. It is evident from a study of this Figure that there is no sharp plane of demarcation between the hymenium and subhymenium. Magnification, 1,040. 444 RESEARCHES ON FUNGI Camera-lucida Studies of a Nearly Exhausted Hymenium. — The studies of the hymenium represented in Fig. 151 were made upon an old and nearly exhausted mushroom obtained from a field in. England. The top of the mushroom was flat, and the gills were of a dark reddish-brown colour, indeed almost black ; but the mottling of the hymenial surface could still be discerned. Evidently the fruit-body had been shedding spores for some days and was nearing the point of exhaustion. Sections of the living gills were made with a hand-razor as before, and two of them are shown at A and B in Fig. 151. In Fig. 151, A, the elements nos. 1, 2, and 3 are present-genera- tion basidia from all of which the spores were knocked off in cutting the section as at no. 3 ; but in nos. 1 and 2 the spores have been restored semi-diagrammatically. No. 4 is an old past-generation basidium arising deep down in the subhymenium from the cell no. VIII. The outlines of the other past-generations basidia cannot be clearly discerned, but it is evident that their gelatinised tops make up the general level of the hymenium which can be seen at I. There are no coming-generation basidia included in the section. Elements nos. 5, 6, and 7 are paraphyses. Two waste spores, w, are lying on the hymenium. The subhymenium consists roughly of three layers of cells : I to IV inclusive, V to VII inclusive, and VIII to X inclusive. It is evident that the past- generations basidia have all disappeared from the section except no. 4 ; otherwise it would be possible to observe their outlines and their attachments to the cells of the two upper layers of the subhymenium. In Fig. 151, B, the elements nos. 1 and 2 are present-generation basidia from which the spores were knocked off in cutting the section ; but for no. 2 the spores have been restored semi- diagrammatically. Nos. 3 and 4 are coming-generation basidia, the presence of which indicates that the mushroom, although old, would continue to discharge spores for at least half a day and possibly for twenty-four hours longer. The past-generations basidia which can be clearly distinguished are elements nos. 5, 6, 7, 8, and 9. The tops of other past-generations basidia, the walls of which have apparently been gelatinised, make up the PSALLIOTA CAMPESTRIS 445 general level of the hymenium /. In the unclear spot, u, no FIG. 151. — Psalliota campestris (wild form). Camera-lucida drawings of an old, nearly exhausted hymenium showing past-generations basidia arising at various levels of the subhymenium. A and B are cross-sections. In A, the elements nos. 1, 2, and 3 are present-generation basidia from which the spores were knocked off in cutting the section as at 3 ; but in 1 and 2 the spores have been restored semi-diagrammatically ; no. 4, an old past-generation basidium arising from the subhymenial cell no. VIII ; nos. 5, 6, and 7 are paraphyses ; I, the general level of the hymenium including tops of no longer distinct past- generations basidia ; w, waste spores lying on the hymenium ; the subhymenium consists roughly of three layers of cells : I, II, III, and IV ; V, VI, and VII ; VIII, IX, and X. In B, nos. 1 and 2 are present-generation basidia from which the spores were knocked off, but for no. 2 the spores have been restored semi-diagrammatically ; nos. 3 and 4, coming-generation basidia ; nos. 5, 6, 7, 8, and 9, past-generations basidia ; nos. 5 and 7, the longest and oldest, arise from near the base of the subhymenium ; nos. 9, 6, and 8 arise successively at higher levels, while the basidium no. 4 arises at a still higher level ; u, unclear spot where outlines of the cells could not be distinguished ; I, general level of the hymenium made of indistinct tops of past-generations basidia ; there are roughly four layers of subhymenial cells : I, II ; III, IV, V, VI, VII ; VIII, IX, X ; XI, XII, XIII ; basidia arise at all these levels. C, a surface view of the hymenium of the same material, showing such elements as could be clearly distinguished ; a, past-generation basidia (as seen in water, brownish and with no distinguishable sterigmatic stumps) ; b, a coming- generation basidium ; c c, small paraphyses. Magnification, 1,040. individual element can be made out. The absence of any level plane sharply separating the hymenium from the subhymenium is 446 RESEARCHES ON FUNGI especially clear in this section. The subhymenium consists roughly of the following four layers of cells : I and II, III to VII inclusive, VIII to X inclusive, and XI to XIII inclusive. Now it will be noticed that the basidia are attached at very various depths to the cells of the subhymenium. The basidia nos. 5 and 7 are attached to the cells XI and XII, respectively, which belong to the deepest layer of the subhymenium. The basidia nos. 9, 6, 8, and 4 arise successively at higher and higher levels, and in consequence their shafts are shorter and shorter. If one compares the basidia nos. 1, 2, and 3 in Fig. 151, A, and no. 4 in B, i.e. some of the last basidia to produce spores in an old hymenium, with basidia nos. 1, 2, and 3 in Fig. 149 (p. 440), nos. 1, 2, 3, 4, and 5 in Fig. 150 (p. 443), and nos. 4 and 5 in Fig. 148 (p. 438), i.e. with some of the earliest basidia to produce spores in a young hymenium, one can at once perceive that the former set of basidia is much shorter than the latter set. The comparison of hymenial sections from a mushroom at different ages seems to show that, in general, the first basidia to come to maturity, i.e. to produce spores, are the longest, the last basidia to produce spores are the shortest, and that there is a tendency to gradual shortening of the basidia coming to maturity as the hymenium becomes more and more exhausted. In Fig. 151, C, is shown a surface view of the hymenium taken from the same material as the sections A and B. The hymenium was submerged in water under a cover-glass. Only such elements are shown as could be clearly distinguished. The past-generations basidia, a, were brownish. Their sterigmatic stumps could not be seen ; possibly they were disappearing or had disappeared. The large element & is a coming-generation basidium, and the elements c are paraphyses. As already explained in an earlier section, in the Mushroom, on account of the small size of the hymenial elements, the optical difficulties are too great to permit of one clearly distinguishing all the elements in a face view of the hymenium. The clear spaces left in the drawing C indicate that the cells in those regions seemed too vague and uncertain in their outlines to be worth sketching. The elements which have actually been drawn, with the exception of b, appear to be either PSALLIOTA CAMPESTRIS 447 past-generations basidia or paraphyses. No present-generation basidium happened to be included in the area. Camera-lucida Studies of a Completely Exhausted Hymenium. — The studies of the hymenium represented in Fig. 152 were made upon a completely exhausted mushroom which had discharged spores for six days and six nights. The sections were cut with a hand-razor a few hours after the last spores had been shed. In many of the cross-sections, one of which is represented at A, only the paraphyses could be made out, and the basidia, all of which belonged to past generations, were only represented by their gelatinised remains. In Fig. 152, A, the hymenium h is composed of paraphyses which still have clear outlines. All these cells were devoid of protoplasm and in this respect resembled the cells of the subhymenium s and trama t. The paraphyses, p, evidently correspond to the elements nos. 18, 19, 20, 21, and 22 in Fig. 148 (p. 438) and to nos. 13, 14, 15, 16, and 17 in Fig. 149 (p. 440). They were smaller when the hymenium began to discharge spores and have become somewhat swollen during the gradual exhaustion of the hymenium. The paraphyses, g g, appear to have arisen as outgrowths from the uppermost subhymenial cells, i.e. from cells like nos. I, II, III, IV, and V in Fig. 149 (p. 440) and like the cells upon which the paraphyses nos. 18, 19, 20, and 21 are situated in Fig. 148 (p. 438). The past-generations basidia have disappeared except for their gelatinised remains, which are indicated at r r. There are a number of waste spores, w, lying on, and adherent to, the hymenium. In the drawing B of Fig. 152 is represented a piece of a section of the totally exhausted hymenium in which, in addition to the paraphyses p, two past-generations basidia can still be recognised at 6. Their bodies were very faintly outlined and are so short that there can be little doubt that their sterigmata were among the last to produce spores. The subhymenium is shown at s, and a waste spore at w. The surface view of the exhausted hymenium shown at C corresponds to the section at A. The only elements we can clearly distinguish in it are the paraphyses p. The gelatinous remains of the basidia fill the spaces between the paraphyses and are FIG. 152. — Psalliota campestris (wild form). To show the existence of paraphyses in a completely exhausted hymenium. The hymenium was examined a few hours after total exhaustion, before which the mushroom had shed spores for six days and six nights. A and B, transverse sections through the living hymenium ; C, a surface view of the same. Preparations mounted in water and drawn with aid of the camera lucida. In A, h is the hymenium composed of paraphyses, p, g, and of the gelatinised remains of basidia, r. The paraphyses g g appear to have been formed by mere extensions of two subhymenial cells. The waste spores w are lying on the hymenium. The subhymenium, composed of about three layers of swollen cells, is shown at s, and part of the trama at t. In B, the structure of the hymenium is similar to that in A except for the fact that the outlines of two short lately collapsed basidia could be definitely made out at 6. The remains of other basidia are shown at r, paraphyses at p, a waste spore at w, and subhymenial cells at 8. In the surface view C, the paraphyses are shown at p, and the gelatinised remains of the exhausted basidia at r. Lying on the hymenium are two fully pigmented and full-sized spores, w, and four unripe, unpigmented spores at u. Magnification 1,040. PSALLIOTA CAMPESTRIS 449 indicated at r. A mature waste spore is present at w, and four very immature waste spores which failed to attain even full size are present at u. This surface view serves not only to demonstrate the presence of paraphyses, but also to show how numerous they are and how evenly distributed throughout the hymenium. Secotium agaricoides. — In concluding this Chapter, I shall make mention of Secotium agaricoides (Czern.) Holl. (=S. acuminatum Mont.), a peculiar fleshy fungus which in some ways resembles a Puff-ball but which, from its mode of development as described by Conard, may be nothing more than a much modified Psalliota, pos- sibly a near relative or direct descendant of the Common Mushroom. Secotium agaricoides (Figs. 153, 154, 155) was first found hi the Ukraine by Czerniaiev 1 in 1845. In 1882 it was found by Peck 2 in the United States of America and was described by him as S. Warnei. In 1903, Hollo's,3 in his Gasteromycetes Hungariae, treated of it at some length with the help of an excellent series of illustrations which showed its variability in form, size, and general appearance. This author examined some of Peck's specimens of S. Warnei, pronounced them to be identical with the European plants, and recorded $. agaricoides not only from Europe but also from Asia, Africa (Algeria), North America (Ohio to Wisconsin and Kansas), New Zealand, and Australia (Banks Peninsula). I can now add that the fungus has recently been found in Canada, by myself near Winnipeg (Fig. 155) and by Mr. Odell near Ottawa (Figs. 153 and 154). The fruit-bodies of Secotium agaricoides, when found upon a grassy sward, look very much like small Puff-balls ; for they are more or less spherical or oval, whitish, .and closely adherent to the ground. However, they each possess what all true Puff-balls lack, namely, a stipe which continues upwards through the body of the sporophore as a columella. The body-wall (peridium) encloses a compact mass of branched and much folded lamellae instead of a capillitium (Fig. 154). 1 B. M. Czerniaiev, " Nouveaux Cryptogames de 1'Ukraine et quelques mots sur la flore de ce pays," Bull. Soc. Imp. Nat. Moscou, T. XVIII, 1845, pp. 132-157. 2 C. H. Peck, " New Species of Fungi," Bull. Torrey Bot. Club, vol. ix, 1882, pp. 61-62. 3 L. Hollos, Gasteromycetes Hungariae, Budapest, 1903 ; Leipzig, 1904, pp. 33-37, Plates III-VI bis. IT^T TT * G 450 RESEARCHES ON FUNGI At maturity the fruit-body, under dry climatic conditions (as at Winnipeg), may sometimes remain closed ; but usually it dehisces FIG. 153. — Secotium agaricoides — a fungus resembling a Puff-ball, but closely allied to Psalliota. A, a fruit-body developing its spores. B, a normal fully- developed ripe fruit-body showing the peridium slightly separated from the stipe below and longitudinally cracked ; branched gills and spores, concealed under the peridium, now pulverulent. C, a fruit-body expanded to a much greater extent than is normal, its body thus resembling the expanded pileus of a Psalliota. Collected by W. S. Odell at Ottawa. Photographed by the Photographic Division of the Canadian Geological Survey. Natural size. irregularly by means of one or more longitudinal rifts (Figs. 153, B, and 155) or by part, or occasionally the whole, of the base of the PSALLIOTA CAMPESTR1S 45l peridium separating slightly from the stipe and columella after the manner of an agaric such as a Mushroom. The photographs reproduced in Figs. 153, B, and 155 show common types of mature fruit-bodies, while those reproduced in Figs. 153, C, and 154 show an amount of expansion beyond the ordinary. After a median vertical section had been made through a fruit- FIG. 154. — Secotium agaricoides. A vertical section through a ripe fruit-body which has expanded on the left side only. Exposed to view are: the peg-like stipe and its extension, the colu- mella ; the peridial flesh ; and the pulverulent, spore -enclosing gills. Collected by W. S. Odell at Ottawa. Photographed by the Photographic Division of the Canadian Geological Survey. Natural size. body freshly gathered at Winnipeg, the upper part of the columella turned brown on bruising and the anastomosing gill-plates greenish. Conard l describes the general development of the fruit-bodies of Secotium agaricoides in the following words : ' ' The youngest specimens are nearly globular, white, and smooth. 'Then comes a ' button stage,' with the upper portion slightly larger than the stalk and the surface still smooth. Later, the fertile portion increases greatly in height and diameter, leaving the free stalk as a mere basal projection. The superficial brown scales of the peridium appear only when near full maturity. The spores are fully mature 1 H. S. Conard, "The Structure and Development of Secotium agaricoides," Mycologia, vol. vii, 1915, p. 95. 452 RESEARCHES ON FUNGI while the trama, peridium, and stalk are still fleshy. The old speci- mens, at least in our climate (Iowa), dry up and do not putrefy." With the help of the microtome, Conard found that the develop- ment of a fruit-body of Secotium agaricoides, in its earliest stages, is precisely like that of Psalliota campestris and P. arvensis as described by Atkinson,1 and not at all like that of Phallus and Mutinus. The hymenophore was distinctly seen as a horizontal ring of deeply-staining tissue within the globular fungus body of a sporophore only 3-5 mm. in diameter (Fig. 156, No. 1) ; and in another fungus-body, 9 mm. in diameter, the branched and anastomosing gills were found to run primarily in a radial direc- tion and to be extending from the cap to the columella (Fig. 156, No. 22). The spores are olive-green in the mass, ovoid, 7 p long and 5 p wide, smooth, thick-walled, and possessed of an apical germ-pore.3 This pore is evidently homologous with the germ-pore that one finds in the thick- walled spores of Goprini, Panaeoli, Psalliotae, and other chromosporous Agaricineae. The basidia are tetra- sterigmatic and quadrisporous, and a spore often carries with it a part of the sterigma as an appendage.4 "The mature gills," says Conard,5 "are much branched and folded. The tramal tissue consists of long branching and nearly parallel hyphae, whose ultimate branchlets form the densely crowded basidia. There are no cystidia or other aberrant cells. From a fairly dense web in growing stages, the trama becomes looser toward maturity and finally becomes dry and fragile. From the above account it is clear that the carpophore of Secotium agari- coides presents in its origin and development an exact counterpart of that of Agaricus 6 (Atkinson, 1906, 1914). There is at first a universal veil like that of evolvate agarics (Agaricus, Armillaria, and Stropharia). There is an ill-differentiated partial veil. The origin of the hymenophore agrees precisely with that of all recently reported agarics except Hypholoma (Allen, 1906) and Coprinus 1 G. F. Atkinson, " The Development of Agaricus campestris,''' Bot. Gaz., vol. xlii, 1906, pp. 215-221 ; also, " The Development of Agaricus arvensis and A. comtulus" Am. Journ. Bot., vol. i, 1914, pp. 3-22. 2 H. S. Conard, Joe. cit., pp. 95-98. 3- 4 Ibid., p. 94. 5 Ibid., pp. 97-98. 6 Agaricus = Psalliota. PSALLIOTA CAMPESTRIS 453 (Levine, 1914). The marginal growth of the gill system is also familiar in the mushrooms. The columella comes into being exactly as does the stipe of mushrooms. Indeed, we might well speak of stipe and pileus of Secotium, rather than of peridium and columella. The basidia have the shape, size, and arrangement found in agarics. The copious branching of the gills exceeds anything seen in agarics. The failure of the cap and gills to expand, the drying up of the trama into a friable mass of tissues and spores, the olive-brown color of the spores, and the freedom of the spores to ' puff ' when the exposed mass is touched, are all lycoperdinean characters." A simple forking of the gills occurs in certain Agaricineae, e.g. Russula furcata and Can- tharettus cibarius ; but it is a perfectly normal phenomenon and in no way interferes with the liberation of the spores. On the other hand, an irregular branching and anastomosis of the gills sometimes occurs in Pholiota erebia (Fig. 157), etc., and I myself have observed it in a Psalliota campestris fruit-body which I gathered in a field near Banbury, England, and in some large Coprinus niveus fruit-bodies which I grew in pure cultures on horse dung at Winnipeg. Here the lamellar irregularities were certainly abnormal and seriously inter- fered with spore-liberation. I am inclined to believe that the copious branching and anastomosis of the gills in Secotium agaricoides arose in the first place as an abnormality affecting simple radial gills and that it is therefore comparable with the FIG. 155. — Secotium agaricoides. A full- grown conico-cylindrical fruit-body showing its characteristically peg- like stipe and a longitudinal crack in the upper part of the peridium. Collected at Winnipeg by the author. Natural size. 454 RESEARCHES ON FUNGI abnormalities in Pholiota erebia, Psalliota campestris, and Coprimis niveus just described. "On the whole," concludes Conard,1 " Secotium agaricoides would best be placed near to Agaricus (Psalliota), either in the Agariceae or Marasmieae of Hennings (1897). It clearly falls within the Agaricaceae of Maire (1902). It is to be regarded as a primitive FIG. 15G. — Secotium agaricoides. Stages in the development of two fruit- bodies. No. 1, a median longitudinal section of a fruit-body 3'5 mm. in diameter, showing the fundament of the hymenophore and the gill- chamber. No. 2, the left half of a median longitudinal section of a fruit-body 9 mm. in diameter, showing a ring of well-formed gills extend- ing from the cap toward the columella. The gills are much branched . ' and anastomosing, and they run primarily in a radial direction. At a they are attached to the columella and are therefore decurrent ; and at b they are extending centrifugally. Photographed by S. Conard. or arrested agaric — perhaps a paedogenic form, reaching its re- productive maturity in the * button ' stage." It seems, therefore, as a result of Conard's investigation, that we are justified in thinking of Secotium agaricoides as a fungus which superficially resembles a Puff-ball (Lycoperdon), but which, nevertheless, has been derived from an ancestor which must have had a structure practically identical with that of the Common Mushroom and its allies. This ancestor, doubtless, possessed a simple radial gill-system, expanded its pileus in the usual way, and 1 H. S. Conard, loc. cit., p. 101. PSALLIOTA CAMPESTRIS 455 shot its spores from its basidia into the inter-lamellar spaces, the production and liberation of the spores thus continuing for a spore - fall period occupying several days. In this ancestor, in the course of evolution, we must suppose that a series of changes took place which profoundly altered the mode of spore-liberation and rendered the beautiful gunnery of the basidia super- fluous and useless. Owing, possibly, to one or more mutations : (1) the gills became branched and anastomosing ; (2) the pileus ceased to expand in the usual way at maturity ; (3) the spores ceased to be shot away from their basidia and, on ripening, merely accumulated in the gill- folds and cavities ; (4) the pileus-flesh became tougher and more resistant to decay ; and (5) the liberation of the spores imprisoned in the gill-folds and cavities came to be effected by the gills drying and becoming friable and by the dry spore-powder falling through cracks in the unexpanded pileus. Thus, to a certain extent, the fungus came to resemble a Puff- ball in both its outward form and its mode of evacuating its spores. However, whilst a Puff-ball, e.g. Lycoperdon gemmatum or L. giganteum, gives one the impression of having a structure which is beautifully adapted to secure the liberation of the spores,1 one does not gain a similar impression on beholding the irregularly cracking, often misshapen fruit-bodies of Secotium agaricoides. These fruit-bodies rather look like monstrosities, indeed, as if they had recently changed their mode of spore -liberation but were still structurally imperfectly adapted to the new functional demands made upon them. 1 Cf. these Researches, vol. i, 1909, p. 258. FIG. 157. — Pholiota erebia. An abnormal fruit-body : the gills, instead of being simple, exhibit irregular branching and anasto- mosis, n Photographed at Scarborough, Eng- land, by A. E. Peck. Natural size. GENERAL SUMMARY THE FOLLOWING IS A SUMMARY OF THE MORE IMPORTANT RESULTS OBTAINED DURING THE INVESTIGATIONS Chapter I. — Typically, the hymenium of Hymenomycetes consists of basidia, paraphyses, and cystidia. Two nuclei fuse together in each young basidium, and every basidium eventually produces spores. There is no nuclear fusion in paraphyses and cystidia, and these elements are sterile from the first. In all Hymenomycetes, the spore has a little projection or hilum at its base, which projects toward the axis of the basidium. From this hilum, a few seconds before spore-discharge, a tiny fluid drop is excreted. The drop grows to a maximum size which is usually equal to about one-half the diameter of the spore. Then the spore is shot violently away from its sterigma. The drop clings to the spore and is always carried away by it. Sometimes the drop is not excreted at all and sometimes the excretion is excessive. Under these conditions the spores are never discharged from their sterigmata. The author discusses, but is not able to explain satisfactorily, the mechanism of spore-discharge. The fluid drop excreted by the hilum of the spore is not improbably an indication of some chemical or physical change taking place in the cell-wall. A basidium never produces more than one generation of spores upon its sterigmata. Direct observations made upon living disterigmatic and tetrasterigmatic basidia (Psalliota, Calocera, Panaeolus, etc.) show that a basidium, 15-30 minutes after discharging the last spore it bears, collapses and dies. There is a relation between the volume of a basidium and the total volume of the spores which it bears. In Psalliota, campestris (cultivated form) the volume of a spore of a monosterigmatic basidium is twice that of one of the spores of a disterigmatic basidium. In the Hymenomycetes, the sterigma, in association with the hilum of the spore, is to be regarded as an organ specialised for bringing about violent spore-discharge. In the Gastromycetes, in which the spores are not violently discharged, the sterigmata are either suppressed or imper- fectly developed, and the spore-hilum is altogether lacking. Typically, in the Hymenomycetes, each basidium bears four spores separated from 457 458 RESEARCHES ON FUNGI one another by well-marked spaces ; whereas, in the Gastromycetes, each basidium often bears more than four spores, and the spores are often crowded together on the end of the basidium-body so that at maturity they touch one another. This difference, again, is correlated with the fact that in the Hymenomycetes the spores are violently discharged, whereas in the Gastromycetes they are not. A toy balloon resembles a spore in having an enormous surface area relatively to its mass. The author has invented a balloon gun with to3T balloons for projectiles. The trajectory of a balloon shot horizontally resembles the trajectory (sporabola) of a spore shot horizontally from its sterigma. The rates of fall of thistle-down, basidiospores, and bacteria have been compared. The spores of Hymenomycetes fall much more slowly than thistle-down, but much faster than bacteria. To fall a single inch in still air certain very small bacteria require somewhat more than three hours. To calculate the rate of fall of bacteria the author has employed the equation for Stokes' Law as corrected by Millikan. Chapter II.- — The interval of time which elapses between the first appearance of a spore on the end of its sterigma as a tiny rudiment and the moment of spore-discharge has been measured for a number of Agaricineae. In Collybia velutipes it was found to be only about 47 minutes and in Panaeolus campanulatus about 7 hours and 30 minutes. The interval is usually very short (50-90 minutes) for spores having thin, smooth, colourless walls and much longer (3-9 hours) for spores having thick, pigmented walls or walls which are polyhedral or warted. In general, the interval is short for the spores of the Leucosporae and long for the Chromosporae (chromosporous Agaricineae). The longest interval observed was about 32 hours for Coprinus sterquilinus. In this fungus, on any small area of the hymenium, the basidia all develop their spores simultaneously. The ripening of a spore, after the spore has attained full size, takes a longer time than the growth of the spore to full size. Thus in Collybia velutipes the spores take about 15 minutes to grow to full size and a further 32 minutes to ripen and become discharged ; and in Psalliota campestris the spores take 40 minutes to grow to full size and a further period of about 7 hours and 20 minutes to ripen and become discharged. Chapter III. — When the parasite Hypomyces lactifluorum attacks Lactarius piperatus, the host fruit-bodies grow to normal size but fail to develop gills and remain completely sterile. On the other hand, the Hypomyces gives rise to great numbers of perithecia on the under side of the pileus. The author discusses the relations between the parasite and its host, and describes his observations upon the discharge of spores from the perithecia. Occasionally the fruit-bodies of certain Hymenomycetes grow to full size but remain partially or completely sterile. Some fruit-bodies of GENERAL SUMMARY 459 Coprinus lagopus which developed no spores at all were found to possess basidium-bodies but neither steriginata nor spore-rudiments. The author discusses the possible cause of sterility. It is suggested that the monstrous fruit-bodies of Polyporus mfescens owe their abnormal form to a loss of power to respond to geo tropic stimuli. The pileus of a Coprinus sterquilinus fruit-body was successfully grafted upon the decapitated stipe of another fruit-body of the same species ; but attempts to graft the pileus of one species of Coprinus upon the stipe of another species failed. Coprinus lagopus, which under favourable conditions produces large fruit-bodies, 100-200 mm. high, under less favourable conditions produces dwarf fruit-bodies which may have stipes only 1-10 mm. high and ex- panded pilei only 0*75-3 mm. in diameter. These minute fruit-bodies, Avhich may lack cystidia, develop and liberate a small number of perfect spores which, if sown on favourable nutrient media, give rise to large normal fruit-bodies one hundred or more times larger than the dwarfs. The dwarf fruit-bodies of C. lagopus have been regarded by certain systematists as belonging to an independent species. At Winnipeg Marasmius oreades has been successfully cultivated for food purposes from spawn on a soil-covered bed of manure. The author describes how the bed was made and the large abnormal fruit-bodies which appeared upon it. Chapter IV. — The spore-fall period for Coprinus curtus, which has small fruit-bodies (expanded pilei 2-5-15 mm. wide) occurring on horse dung, is extremely brief, being only from 30 minutes to 2 hours and 30 minutes in length. The fruit-bodies are produced in diurnal crops, and the spores are usually liberated about the middle of the morning. The escape of the white spore-clouds from the pilei of Armillaria mellea was successfully observed by the author with the naked eye, under natural conditions in a wood. From the fruit-bodies of Hydnum septentrionale there is probably not " a simultaneous liberation of spores followed by a period of rest," as has been supposed ; but, on the contrary, the spores are probably liberated in a constant stream, as in the Agaricineae, the Polyporeae, and Hymenomycetes in general. The fruit-body of Fomes fomentarius develops a new annual tube- layer in the autumn, but the production and liberation of spores from it is delayed until spring. This discovery, originally made by J. H. Faull, has been confirmed by the author. Faull has also discovered that, in F. fomentarius, each annual tube-layer may produce a crop of spores for four years in succession. The form and mode of growth of a F. fomentarius fruit-body is profoundly modified by various reactions to the stimulus of gravity. The attachment of a F. fomentarius fruit- body to the mycelium in the trunk of a tree is localised at the back 460 RESEARCHES ON FUNGI •of the pileus-flesh at the upper end of the fruit-body, and the attachment of the rest of the fruit-body to the tree-trunk is merely by adhesion to the bark. Fomes igniarius develops a new layer of hymenial tubes each summer and has a summer spore-discharge period of somewhat more than two months' duration. In Manitoba, owing to winter frost, the spore-discharge period of Daedalea confragosa and certain other lignicolous Hymenomycetes is often divided into two parts, one autumnal and the other vernal. In the Polyporeae, the pilei : (1) may be solitary as in Fomes offtcinalis, (2) may be imbricated as in Polyporus sulphureus, or (3) may form part of a compound fruit-body as in P. umbellatus. The author discusses the significance of these arrangements. Chapter V. — Fomes applanatus is an important wood-destroying fungus which attacks upwards of fifty species of trees. The author treats of its biology, more especially of its production and liberation of spores, in the light of the investigations made by J. H. White and by himself. The basidiospore should not be regarded as a chlamydo- spore. Fomes applanatus may produce 10 successive annual tube- layers ; F. officinalis, 45 ; and F, igniarius, 80. Owing to the narrowness, length, and crowding of the hymenial tubes a fruit-body of Fomes applanatus has a very large hymenial surface. White found that a large fruit-body liberated about 30,000,000,000 spores in 24 hours and that the spore-fall period had a duration of six months (the longest known). The visible emission of spore-clouds from the pileus has been observed by White at Ottawa and by the author at Kew. The length of the spore-fall period in the Hymenomycetes is roughly correlated with the mechanical consistence of the fruit-bodies, the most watery fruit-bodies having the shortest spore-fall period, and the most woody the longest. The number of spores liberated per annum by a very large fruit-body of Fomes applanatus is of the same order of magnitude as the number liberated from a fruit-body of the Giant Puff-ball. The great length of the spore-fall period appears to be correlated with the long-continued growth in length of the hymenial tubes ; but a full explanation of the duration of the spore-fall period can be given only after the organisation of the hymenium has been elucidated in detail. The production of vast numbers of spores by the fruit-bodies of wood- destroying fungi appears to be necessary to permit of these fungi overcoming the difficulties of invading new woody substrata with sufficient frequency to maintain themselves under natural conditions. Chapter VI. — In the Hydneae, the Tremellineae, and the Clavarieae, the form of the fruit-body is intimately correlated with the size and strength of the basidial guns which discharge the spores. In the Exobasidieae, the absence of a fruit-body is correlated with the GENERAL SUMMARY 461 fact that the fungi in this group are parasitic on shrubs and produce their hymenial layers high above the ground on the surface of living leaves, etc. In the Hydneae, the increase in the surface-area of the hymenium beneath the pileus has been effected by the development of positively geotropic spinous processes (teeth), instead of by gills as in the Agaricineae, or by hymenial tubes as in the Polyporeae. Each spore is discharged from its sterigma only after the emission of a drop of fluid from the hilum, as in other Hymenomycetes. The gelatinous consistence of the fruit-bodies of the Tremellineae is correlated with the lignicolous habit of the group. When sticks bearing Tremellineae dry up, the fruit-bodies dry up too but retain their vitality. When rain comes again, the water is rapidly absorbed, not by capillarity as in Lenzites and other non-gelatinous lignicolous fruit-bodies, but by imbibition. After absorbing water and expanding, a gelatinous fruit-body immediately resumes its work of producing and liberating spores. The period between the first appearance of a spore on its sterigma as a tiny rudiment and its discharge with drop-excretion is very short in the Tremellineae : 1 hour 20 minutes in Calocera cornea ; 1 hour 15 minutes in Exidia albida and 50 minutes in Dacryomyces deliquescens. This rapidity of spore- development is correlated with the fact that the spore-walls are thin and smooth, and with the fact that the fungi are lignicolous. The author discusses the problem of septation in the basidia of the Tremellineae and attempts to solve it from the physiological point of view. The spores of the Tremellineae are more violently discharged than those of the non-tremelloid Hymenomycetes, but somewhat less violently than those of the Uredineae. In violence of spore-discharge the Tremellineae thus take up an intermediate position between the Hymenomycetes in general and the Uredineae. In those Tremellineae in which the hymenial surface often looks more or less upwards, great violence of spore-discharge is distinctly advantageous from the point of view of the dissemination of the spores by the wind. The author redescribes the basidial and oidial fruit-bodies of Dacryomyces deliquescens and points out that, hitherto, they have been regarded by certain systematists as belonging to different species. In the Clavarieae, the hymenium is exposed on the surface of the fruit-bodies. Some simple fruit-bodies, e.g. those of Clavaria pistillaris, are obconic in form, so that the hymenium looks more or less down- wards. In compound Clavariae the successive branches are usually more or less obconic in form , so that here again the hymenium tends to be so placed that it looks horizontally more or less downwards, thus favouring the dispersal of the spores. In most Clavariae the basal 462 RESEARCHES ON FUNGI parts of the fruit-bodies, the tops or tips of the ultimate branches, and the upper surfaces in the angles of each fork are usually sterile. In the Clavariae the discharge of the spores from the basidia is accompanied by a drop excretion from the spore-hilum, as in other Hymenomycetes . Sparassis has its hymenium restricted to the lower surfaces of its laminae. In this it resembles Stereum, but it may have been evolved from a Clavaria. Calocera resembles Clavaria in its form and in the small distance to which its spores are shot from their sterigmata, but it differs from Clavaria in being subgelatinous, in becoming horny on drying, in reviving after desiccation, and in possessing long cylindrical bifurcated basidia. The galls formed by Exobasidium species upon their hosts are advantageous to the parasites in extending the surfaces over which they may spread their hymenium. Chapter VII. — The Red Squirrel of North America not only feeds on the seeds of fir-cones, hazel-nuts, etc., but is also an habitual mycophagist. In the late autumn, it often collects fleshy fungi in large numbers for its winter supply of food, and it stores these fungi some- times en masse in holes in tree trunks, old birds' nests, etc., and some- times on the branches of certain trees. The fruit-bodies hung out in the branches of trees dry up and are thus preserved ; and squirrels have been watched eating them during the winter. Chapter VIII. — Slugs attack and feed upon most species of fleshy Hymenomycetes occurring in woods, including Amanita muscaria and A. phalloides which are poisonous to human beings, and Pluteus cervinus the gills of which are provided with numerous cuspidate cystidia. The attacks of slugs often interfere seriously with, and sometimes entirely prevent, the production and liberation of spores by individual fruit- bodies. Most species of fleshy fungi are in no way protected against slugs. Very many species of fleshy fungi do not depend on slugs for the dissemination or germination of their spores. Slugs find the fungi upon which they feed by their sense of smell. Chemotactic experiments made upon Limax maximus with Phallus impudicus, Boletus scaber, Russula nigricans, etc., in which the fruit- bodies were set out on bare gravel, showed that this slug can find various fruit-bodies at night by its sense of smell even when the fruit-bodies are 20 feet distant. Chapter IX. — In the Agaricineae there are two mam types of organisation for the production and liberation of spores : (1) the Aequi- hymeniiferae or Non-Coprinus Type, and (2) the Inaequi-hymeniiferae or Coprinus Type ; and each Type includes several distinct Sub-types. In the Aequi-hymeniiferae : (1) usually the gills are relatively thick ; (2) the gills are wedge-shaped in cross-section ; (3) the gills are positively geotropic ; (4) the hymenium looks downwards to the earth ; (5) every GENERAL SUMMARY 463 part of the hymenium produces and liberates spores during the whole period of spore-discharge; and (6) the gills are not destroyed from below upwards by autodigestion. The name Aequi-hymeniiferae refers to the equal development of the different parts of the hymenium. In the Inaequi-hymeniiferae : (1) the gills are very thin ; (2) the gills are not wedge-shaped but, on the whole, parallel-sided or sub- parallel-sided ; (3) the gills are not positively geotropic ; (4) usually the hymenium on one side of a gill at maturity looks slightly downwards and that on the other side slightly upwards ; (5) the spores ripen in succession from below upwards on each gill ; (6) the spores are dis- charged in succession from below upwards on each gill ; and (7) auto- digestion proceeds from below upwards on each gill and removes those parts of the gills which have become spore-free and which, if they continued in existence, would become mechanical hindrances to the fall of the remaining spores. The name Inaequi-hymeniiferae refers to the unequal development of the different parts of the hymenium, the pro- duction and liberation of the spores proceeding in a zone-wise manner from below upwards on each gill. Chapter X. — In the Panaeolus Sub- type the gills are mottled. Mottling is an expression of the fact that the hymenium is made up of a mosaic-work of small irregular local areas, the basidia in the darker areas bearing older pigmented spores and those in the lighter areas either no spores at all or spores which are younger and as yet unpig- mented. With the discharge of the spores, a dark area turns light ; and, with the ripening of the spores, a light area turns dark. Mottling of the gills occurs not only in Panaeolus but also in Anellaria, Psilocybe, Hypholoma, Stropharia, Psalliota, Cortinarius, Crepidotus, Flammula, and Pholiota. The changes taking place in one and the same piece of hymenium of a living gill of Panaeolus campanulatus were observed during three days and three nights with a horizontal microscope. It was discovered that each area of the hymenium gives rise to a series of successive generations of spore-bearing basidia, each area changing from black to white alternately many times in succession. The fall of spores from a fruit-body of Panaeolus campanulatus continues for about 8-11 days ; and it is the development of successive generations of basidia on the local hymenial areas which enables spores to be discharged as an uninterrupted spore-stream during the whole of the spore- discharge period. In any small area of the active hymenium of Panaeolus campanulatus (0-2 square mm.), the hymenial elements can all be placed in one or other of the following five classes : (1) past-generations basidia, (2) present-generation basidia, (3) coming-generation basidia, (4) future- generations basidia, and (5) paraphyses. This analysis greatly simplifies our conception of the organisation of the hymenium of Panaeolus 464 RESEARCHES ON FUNGI campanulatus and applies equally to all other members of the Panaeolus Sub- type. The author has studied the hymenium of Panaeolus campanulatus : (1) before the development of the spores, (2) in the middle of the spore- fall period, and (3) at the end of the spore-fall period when the hymenium has become exhausted ; and he has illustrated his analyses with a complete series of drawings, showing the hymenium both in surface view and in cross-section and indicating the time and space relations of all the elements. The paraphyses of Panaeolus campanulatus, unlike the young basidia, are destined from the first to remain sterile. They are about equal in number to the basidia ; but they differ from the basidia in : (1) size, (2) position, (3) form (as they grow older), (4) early vacuolisation and ultimate disappearance of their cytoplasm except for a very thin wall- layer, (5) never developing either sterigmata or spores, and, doubtless, also, (6) in the non-fusion of the two nuclei with which they are at first provided. As, during the spore-fall period, more and more basidia discharge their spores and collapse, the paraphyses swell up and to a large extent fill up the gaps made in the hymenial layer by the shrink- age of the exhausted basidia. In general, the paraphyses appear to act as mechanical supports and nurse-cells for the spore-bearing elements. A basidium, about 20 minutes after discharging the last of its four spores, collapses : its convex end bearing the sterigmata sinks down and becomes concave, thus drawing the sterigmata into a concavity. The shaft of the basidium then shrinks laterally and the four sterigmata become melted down to sterigmatic stumps. Exhausted basidia can be recognised in cross-sections by the flat rims of their concave ends and by their shrunken shafts, and in surface view by their lack of contents and by their sterigmatic stumps. Irregular waves of hymenial development pass across the mottled surface of each gill. These waves are comparable with the undulations which may be observed in the epithelium which produces spermatozoa in the testicular tubules of the Mammalia. The wasted spores, i.e. those which are not properly discharged from the sterigmata and which cling to the hymenium, form but a very small percentage of the total number of spores produced. The author shows that the protuberancy of the mature basidia, the collapse of the exhausted basidia, the relative positions of spores on individual basidia with 2-8 sterigmata, and the relative positions of the spore-bearing basidia of a single generation are all factors making for the success of the fruit-body as an organ for the production and liberation of the spores. Many Hymenomycetes are known in which the basidia are disterigmatic and bisporous ; but Coprinus narcoticus is at present the GENERAL SUMMARY 465 only known hymenomycetous species in which the basidia are regularly tristerigmatic and trisporous. Panaeolus campanulatus lacks pleurocystidia, but possesses a fringe of cheilocystidia along the free edge of each gill. The cheilocystidia excrete mucilaginous globules. Chapter XI. — Stropharia semiglobata belongs to the Panaeolus Sub-type, and its organisation for the production and liberation of spores very closely resembles that of Panaeolus campanulatus. The author describes the hymenium in detail and illustrates it with drawings showing surface-views and cross-sections. The hymenial elements all belong to one of six classes : (1) past-generations basidia, (2) present- generation basidia, (3) coming-generation basidia, (4) future-generations basidia, (5) paraphyses, and (6) cystidia. The efficiency of a given area of the hymenium was determined by observing the total number of spores which the area produced and the percentage of the spores which failed to be properly discharged. Chapter XII. — Anellaria separata belongs to the Panaeolus Sub- type, and its organisation for the production and liberation of spores resembles that of Panaeolus campanulatus and Stropharia semiglobata. The author describes the fungus and, among other illustrations, shows a photomicrograph of the spores belonging to the present spore-bearing generation of basidia situated in one of the dark areas of the mottled hymenium. This is the first photograph of the kind ever published. Chapter XIII. — The author gives a full description of the general characters of the Psalliota campestris fruit-body, including the mottling of the gills, and points out that the gills are not truly deliquescent during the spore-fall period. Dead fruit-bodies, enclosed in crystallising dishes, give off ammonia gas during putrefaction. The occurrence of Psalliota campestris in fairy rings affords evidence that new substrata below the turf are not easily invaded by mycelium produced from spores. The author discusses the principles involved in the construction of the stipe and of the pileus-flesh in relation to their function, the radial arrangement of the gills, the depth and thickness of the gills, the presence of the gill-chamber, the conditions of origin of the fruit-bodies, the fate of fruit-body rudiments, and the effect of dry weather on fruit- body development. Psalliota campestris possesses finely mottled gills and belongs to the Panaeolus Sub-type. The organisation of its hymenium for the production and liberation of spores closely resembles that of Panaeolus campanulatus and Stropharia semiglobata. For the first time, the author describes the structure of the hymenium in detail and illustrates it with drawings showing both cross-sections and surface- views. For the study of the hymenium, Psalliota campestris affords much less favour- able material than Panaeolus campanulatus or Stropharia semiglobata, 466 RESEARCHES ON FUNGI owing to the relative smallness of its basidia and paraphyses and the consequent denser packing of these elements. The spore-discharge period of a wild Psalliota campestris was observed to be of 4-6 days' duration. Spores begin to be discharged from a fruit-body when the gills are pink and the pileus is still expanding, and they continue to be discharged even when the top of the pileus has become quite flat and the gills have turned dark chocolate-brown. The number of spores produced by a large Mushroom was calculated to be upwards of 10,000,000,000. In Psalliota campestris, just as in Panaeolus campanulatus, irregular waves of hymenial development constantly pass over the surface of each gill, thus changing the pattern of the mottling ; and the elements of the hymenium can all be referred to one or other of the following five classes : (1) past-generations basidia, (2) present-generation basidia, (3) coming-generation basidia, (4) future-generations basidia, and (5) paraphyses. In Psalliota campestris, as in Panaeolus campanulatus, the paraphyses are destined to be sterile from the first. They become vacuolated early and swell up considerably during the gradual exhaustion and collapse of the basidia. At the end of the spore-discharge period, all the basidia are collapsed and dead, but all the paraphyses are living. The basidia of wild forms of Psalliota campestris are usually tetra- sterigmatic and quadrisporous, whilst the basidia of cultivated forms are usually disterigmatic and bisporous. In the hymenium of cultivated forms of Psalliota campestris a certain proportion of the basidia are usually monosterigmatic and unisporous. The unisporous basidia are equal in size to the bisporous ; but the spore of a unisporous basidium has about twice the volume of one of the spores of a bisporous basidium. A cultivated Mushroom therefore has spores of two sizes, large and small. The development of the four spores on a basidium of a wild Mush- room, from their first appearance as tiny rudiments to their discharge, was observed to occupy about 7-5 hours. A single spore (1) takes 30-40 minutes to grow from a tiny rudiment to full size, (2) remains colourless for the next two hours, (3) gradually becomes pigmented during the next two hours, (4) remains in the fully pigmented condition on the end of its sterigma for a further 3 hours and about 10 minutes, and (5) is then discharged. The discharge of a Mushroom spore takes place with drop-excretion as in other Hymenomycetes. In Psalliota campestris there is no plane sharply dividing the sub- hymenium from the hymenium ; but the subhymenium passes gradually into the hymenium, owing to the fact that the basidia are attached to subhymenial cells at different depths. Secotium agaricoides resembles a Puff-ball in general appearance, but is nevertheless closely allied to Psalliota. GENERAL INDEX ABNORMALITIES, 60-61, 81-83, 319-320, 453-454 Acrid fungi, and slugs, 217-219 Adams, L. E., on Limax maximus, 223 Aequi-hymeniiferae, 244, 462^163 Aequi-hymeniiferous Type, 237-240, 244 Agaricaceae, spore-fall period, 62 ,, sterigmaandspore-hilum,30 Agaricin, in Fomes officinalis, 127 Agaricineae, and Clavariae, 184 ,, and Exobasidium, 193 ,, and Fomes applanaius, 141 basidia, 164 beam-of-light studies, 105 „ chromosporous, 124, 125 „ confluent pilei, 81-82 ,, drop-excretion, 9, 154 ,, gills and hy menial tubes, 143-144 ,, investigations on, 149 ,, non-septation of basidia, 166-167 ,, position of hymenium, 181, 182 response to gravity, 163 ,, spore-fall periods, 135 „ spore-hilurn, 168 ,, types of fruit-body mechan- ism in, 236-243 ,, variations in size, 83 ,, violence of spore-discharge, 168 Agaricus (= Psalliota), 364, 367, 368, 452, 454 Agriolimax agrestis, as a mycophagist, 217, 218, 219, 220, 221 Agriolimax campestris, in Manitoba, 220 Agriolimax hyperboreus, in Manitoba, 221 Alpine form of Anellaria separata, 349 Amanita, and Hydnum, 151 and slugs, 212, 216, 221 Amanita bisporigera, basidia, 316 Amanita muscaria, and cows, 218 and slugs, 217-218 ,, ,, avoided by squirrels, 202 „ illustration of, 219 ,, ,, violent spore - dis- charge, 4 Amanita pantherina, and slugs, 217 Amanita phalloides, and slugs, 217 „ „ illustration of, 217 Amanita rubescens, and slugs, 214 „ ,, flattening of pileus, 378-379 illustrations of, 378, 379 Amanitae, effects on animals, 218 Amanitopsis, and slugs, 216 Amanitopsis vaginata, and slugs, 214 ,, ,, and Stokes' Law, 20 rate of fall of spores, 39-40 „ ,, sporabola, 67 ,, ,, variation in size, 83 ,, ,, violence of spore- discharge, 169 Amauroderma salebrosum, confluence of fruit-bodies, 81 Ammonia, from decaying mushrooms, 372 Amoebae, and basidium-bodies, 312 Analysis of hymenium, remarks on, 297 Anellaria, and Panaeolus, 253, 351-352 Anellaria separata, account of, 347-359 Alpine form, 349 annulus, 351-352 cultivation, 347-349 drop-excretion, 9 ,, ,, hymenium, 353 ,, ,, illustrations of, 51, 348,350,351,352, 354, 357 467 468 GENERAL INDEX Anellaria separata, mottling, 254 ,, ,, paraphyses, 3, 353 ,, ,, pentasterigmatic basidia, 318 ,, ,, photograph of hy- menium, 355-359 ,, „ rate of spore-develop- ment, 51 ,, ,, summary, 465 Annulus, 351-352, 392-394, 378 Apparatus, for observing hymenial development, 260-264 Arachnida, and gill-chamber, 394 Arion, attracted by a bean, 222 Arion ater, as a mycophagist, 213, 217, 219 ,, ,, senses of sight and smell, 22 1 Arion empiricorum, as a mycophagist. 217, 219 Arion hortensis, and Russula emetica, 219 Arion intermedius, and Russula emetica, 219 Arion subfuscus, as a mycophagist, 213, 216, 217 Armillaria, and Panaeolus Sub-type, 255 ,, rapid spore-development, 50, 51-52, 53 ,, spore-wall, 52 ,, universal veil, 452 Armillaria mellea, and Armillaria Sub- type, 237 ,, ,, clouds of spores seen macroscopically, 100-102, 459 ,, ,, eaten by squirrels, 197-199 ,, ,, illustrations of, 102, 103, 203 „ ,, rate of spore-develop- ment, 44, 49, 54 Armillaria mucida, heterothallism, 148 Armillaria Sub-type, 237 Ascobolus, ascus-gun of, 33 ,, ascus-sap and spore-dis- charge, 26 Ascobolus immersus, distance of spore- discharge, 67 Ascomycetea, ascus degeneration, 32-33 „ explosive asci of, 13 „ spore-discharge, 25 Ascomycetes and Basidiomycetes, com- parison of guns, 33 Ascospores of Hypomyces lactifluorum, dispersal of, 61-67 Ascus, contrasted with a basidium, 170 ASCIIS, discharge of spores from, in Hypomyces lactifluorum, 63-67 „ extrusion of, from a perithecium, 63,65 „ its degeneration in Tuber, 33 Ascus-gun, and basidial gun compared, 67 ,, and sporophore structure, 67 Atkinson, G. F., on bisporous Mushroom basidia, 410 on development of Psal- liotae, 452 ,, ,, on fairy rings, 365 ,, ,, on finding spores of Fomes fomentarius, 106-107 ,, ,, on longevity of Fomes igniarius, 126 ,, ,, on Psalliota campestris, 369-371 ,, ' ,, on spores of Fomes applanatus, 123, 125 Atramentarius Sub-type, 238 Auricularia, absorption of water, 157 Auricularia mesenteries, and slugs, 212 ,, ,, position of hy- menium, 163, 171 resembles Ster- eum, 163-164 „ • revival after desiccation, 160 ., ,, spore-discharge. 156 ,, ,, violence of spore - dis- charge, 169, 192 Auriculariese, and Tremellineae, 156 „ basidial structure, 164, 165-166 ,, drop-excretion, 9 ,, violence of spore-discharge, 169 Autodigestion, in Coprinus Type, 241 of sterile Coprim, 70 ,, supposed, of Mushroom gills, 371 BACTERIA, and ammonia gas, 372 „ and gill-chamber, 394 „ and Mushroom gills, 371 ,, desiccation of, 42 rate of fall in still air, 40-42 GENERAL INDEX 469 Baker, F. C., on Agriolimax agrestis, 221 „ „ on Manitoban slugs, 220- 221 Balloon-gun, and movements of spores, 33-38 „ ,, illustration of, 36 Banker, H. J., on spore-discharge in Hydnum septentrionale, 101, 102-105 Bartsch, P., on slugs and mustard gas, 234-235 Basidia, bisporous, in Mushroom, 1, 30, 315, 407-412, 416, 430 „ collapsed, and Sachs' illustra- tion, 405 „ collapse of, 192, 312-313, 342, 353-355, 417, 457 „ death of, 273 degeneration of, 32, 33 ,, density in successive genera- tions, 269 disappearance of, 371, 415 ,, domino-eight arrangement, 321 evacuation of, 29, 55, 417 ,, exhausted, recognition of, 271- 275 „ failure to develop spores, 70, 72, 73 „ forms in Trernellineae, 164-168 „ general remarks on, 1 „ hexasporous, 1, 30, 319 ,, monomorphic, 245 „ monosporous, 1, 314-315, 407, 408, 410^11, 416, 430 . inonosterigmatic, 315 ,, octosporous, 1, 320 „ of Gastromycetes, 30-33 ,, projection of the four spores, 355 „ quadrisporous, 1, 29-30, 318 ,, relative position of spores, 313- 314 ,, relative sizes of, 361 „ significance of protuberancy, 310-312 , , significance of s uccessive genera - tions, 297-303 ,, successive generations of, 260, ! 333, 414-421 „ trisporous, 1, 317-319, 409 Basidial gun, and ascus-gun compared, 67 ,, ,, and sporophore-structure, 67 Basidiomycetes, R. Maire on spore- production in, 27 Basidiomycetes and Ascomycetes, com- parison of guns, 33 Basidium, aborted, 337, 338 ,, discharge of four spores, 341- 342 ,, number of spore-crops, 27-29, 270, 457 ,, series of changes in a, 304 volume, 29, 411, 457 | Bayliss, J. S., on fairy rings, 94, 364 Beam-of-light method, and continuity of spore-streams, 105 Bsll, C. N., on squirrels as mycopha- gists, 206, 208 Bell, G., on squirrels as mycophagists, 209-210 Benecke, W., on relations of slugs with fungi, 217, 219, 224 Bensaude, M., on fertility and sex in Hymenomycetes, 394-395 Berkeley, M. J., on hymenial structure, 263, 270 „ ,, on paraphyses, 2 ,, ,, on Psalliota campestris, 369, 371 Bisby, G. R., on the spore-fall period in Fomes igniarius, 113-116 Bolbitius flavidus, and Bolbitius Sub- type, 237 „ „ paraphyses, 3 ,, ,, rate of spore-develop- ment, 44, 48, 51 Bolbitius Sub-type, 237 Bolbitius titubans, and Bolbitius Sub- type, 237 Boleti, development of tubes, 150, 151 ,, stored by squirrels, 204 Boletus, and Clavaria, 179-180 and slugs, 212, 216, 235 „ variation in size, 83 Boletus bad.ius, eaten by rabbits, 195-196 „ „ illustration of, 196 Boletus chrysenteron, and slugs, 214 Boletus edulis, size variations of, 83 Boletus elegans, eaten by slugs, 213 „ illustration of, 213 Boletus flavus, eaten by slugs, 214 Boletus luridus, and slugs, 217 Boletus luteus, eaten by slugs, 213 Boletus scaber, and squirrels, 202, 204 „ attracts slugs, 214, 229- 233 „ illustration of, 205 Boletus subtomentosus, confluent fruit- bodies, 81 470 GENERAL INDEX Boletus versipellis, stored by squirrels, 204 Boyce, J. S., on squirrels eating Poly- porus amarus, 202 Brefeld, 0., illustration of Calocera cor- nea, 6 „ onDacryomycesdeliquescens, 172 „ on ejaculation of spores, 4-5 ,, ,, on hymenial structure, 270 „ on spore- hila in Tremel- lineae, 168 Brierley, W. B., and Sparassis, 189 Brooks, F. T., on Polyporus squamosus, 100, 138 CALCIUM oxalate, and drop-excretions, 19-20 Calocera, absorption of water, 157 and Clavaria, 193 basidia of, 191 ,, forms of, 191 „ spore-discharge, 168, 190-193 „ summary, 457 Calocera cornea, and Tremellineae, 156 ,, ,, development and dis- charge of spores, 55 . ,, ,, drop-excretion, 6-7, 9, 171, 192 ,, , form and spore-dis- charge, 191-192 ,, „ illustrations of, 5, 7, 191 ,, ,, nuclei and spore-pro- duction, 28 ,, ,, rate of spore-develop- ment, 44, 49, 54, 55, 161 „ „ resembles simple Cla- variae, 163-164 ,, ,, revival after desiccation, 160 „ „ spore-hilum of, 168 „ „ violence of spore-dis- charge, 169, 171, 192 Calocera viscosa, and Tremellineae, 156 ,, ,, form and spore dis- charge, 192 , , , , mistaken for a Clavaria , 191 ., ,, resembles branched Cla- variae, 163-164 ,, ,, spore-deposits of, 192 Calvatia cyathiformis, fairy rings, 365- 366 illustration of, 366 Calvatia gigantea, number of spores, 137, 139 Cameron, R., and Forms officinalis, 126 Cantharcllus, number of spores on a basidium, 319 ,, parasitised, eaten by squirrels, 201 ,, spore-crops on a basi- dium, 27 Cantharellus cibarius, eaten by squirrels, 202 ,, „ forking of gills, 453 „ ,, hexasporous basi- dia, 1, 319 ., ,, successive spore- discharge, 4 Cantharellus Friesii, pentasterigmatic basidia, 318 Capillarity, absorption of water by, 157 Cheilocystidia, 248, 324-326 Chemotaxis, slugs and fungi, 221-235, 462 Chickaree, as a mycophagist, 197 Chickens, and a squirrel, 211 Churchward, S. G., assistance given by, 267 Clamp-connections, 395 Claudopus nidulans, drop-excretion, 9 Clavaria, and Calocera, 193 „ branches of, and tubes of Polyporeae, 179-180 ,, branching of, 155-156 ,, form and hymenial position, 179-185 quadrisporous basidium, 30 ,, spore-crops on a basidium, 27 Clavariae, basidia of, 185 ,, escape of spores, 181-182 form, 163, 164 Clavarieae, basidia, 164 ,, drop-excretion, 9 ,, small size of group, 180 spore-discharge, 149,179-190 ,, sterigma and spore-hilum, 30 ,, summary, 460-461 ,, violence of spore-discharge, 169 Clavaria abietina, illustration of, 184 Clavaria cinerea, and slugs, 214 barren parts, 183-184 ,, ,, disterigmatic basidia, 317 Clavaria fistulosa, flesh of, 181 Clavaria jormosa, barren parts, 183-184 GENERAL INDEX 471 Clavaria formosa, drop-excretion, 9. 185 ,, ,, violence of spore-dis- charge, 169, 171, 186, 192 r, ,, violent spore-discharge, 185-186 Clavaria ligula, form of, 181 Clavaria pistillar is, flesh, 181 ,, ,, form and hy menial position, 181 illustration of, 180 Clavaria pyxidata, barren parts, 183-184 ,, ., confluence of branches, 185 ,, ,, eaten by squirrels, 201 ,, ,, form of fruit-body, 182-184 illustration of, 183 Clavaria rosea, form and hy menial posi- tion, 181 Clavaria vermicular is, form and hymenial position, 181 Claviceps, form of, 186 Claviceps purpurea, mode of fall of j spores, 35 Clitocybe clavipes, and slugs, 214 Clitocybe gigantea, spores and fairy rings, 369 Clitocybe maxima, eaten by squirrels, 202 Clitocybe monadelpha, eaten by squirrels, 202 Clitocybe nebularis, abnormality in, 82 Clitocybe Sadleri, sterility of, 69 Cnicus arvensis, rate of fall of fruits, 39-40 Coleosporiaceae, violence of spore-dis- charge in, 169 Coleosporium Campanulae, violence of spore-discharge, 169 Coleosporium Petasitidis, violence of spore-discharge, 169 Collapse of basidia, 271-272, 312-313, 342, 417 Collema granulosum, jelly of, 160 Collembola, and gill-chamber, 394 Collybia, and Panaeolus Sub-type, 255 ,, rapid spore-development, 50, 51-52, 53 ,, spore-wall, 52 Collybia butyracea, and slugs, 214, 216 Collybia dryophila, rate of fall of spores 39-40 ., ,, rate of spore-de- velopment, 49, 54, 161 Collybia Jusipes, rate of spore-develop- ment, 44, 49, 54 Collybia radicata, rate of spore-develop- ment, 44, 46, 49, 54 ,, ,, temperature andspore- development, 56-57 Collybia velulipes, and Armillaria Sub- type, 237 „ ,. drop-excretion, 9 ,, „ illustration of, 47 rateof spore-develop- ment, 8, 44, 45, 46, 48, 49, 54, 55, 458 Comatus Sub-type, 238 Compressor cell, illustration of, 45 „ use of, 15-16, 43-45, 63,154,168-169,185, 408 Conard, H. S., on Secotium agaricoides, 451-454 ,, ,, photograph by, 454 Cooke, A. H., on slugs finding fungi, 222 Cooke, M. C., on Clitocybe Sadleri, 69 Coprini, autodigestion of gills, 371 „ effect of water on hymenium, 174 ,, germ-pore, 452 ,, paraphyses of, 282 shape of gills, 391 Coprinus, and Inaequi-hvmeniiferouf* Type, 238 and slugs, 212 „ cystidia, 324 ,, development of, 452 „ diurnal rhythm in, 97 mechanical consistence, 135- 137 ,, mottling absent in, 253 order of investigations, 243 ,, originfrompanaeolateagarics, 425 paraphyses, 3, 243 ,, slow spore-development, £2, 53 ,, spore-discharge periods, 95, 135 spores of, 124, 125 „ sterile fruit-bodies, 69 ,, successive spore-diseharge, 4 „ symmetrical basidia, 30 ,, wave development in, 425 Coprinus atramentarius, and Atramen- tarius Sub- type, 238 „ ,, drop - excre- tion, 9 472 GENERAL INDEX Coprinus atramentarius, mode of shed- ding spores, 98 ,, ,, nuclei in cys- tidia, 4 Coprinus bisporus, illustrations of, 315, 316 „ ,, monosterigmatic and disterigmatic basi- dia, 315-316 ., „ spore-crops on a basidium, 28 Coprinus comatus, and Comatus Sub- type, 236-238 ,, „ drop-excretion, 5, 9 gill pigment, 370-371 „ ., mode of shedding spores, 98 „ ,, number of spores, 137, 139 ,, ,, size variations, 83 „ „ spore-fall period, 135 Coprinus cordisporus, method for ob- taining, 83 Coprinus curlus, and Curtus Sub-type, 238 ,, ,, brief spore-fall period of, 95-99, 135, 459 „ „ drop-excretion, 5 „ ,, dwarf fruit-bodies, 88 „ „ method of obtaining, 83, 97 ,, ,, rhythmic fruit-body production, 97 „ ., violence of spore-dis- charge, 169 Coprinus domesticus, and Lagopus Sub- type, 238 Coprinus echinosporus, domino-eight ar- rangement of basidia, 321- 322 „ ,, grafting of, 80 „ „ illustration of, 321 Coprinus ephemerus, collodial drops on pileal hairs, 19, 25 „ „ dwarf fruit-bodies, 88 ,, „ method of obtain- ing, 83 ,, ,, rhythmic fruit- body production, 97 Coprinus fimetarius, = C. lagopus, 238 ,, ,. heterothallism of, 394-395 Coprinus lagopus, and Lagopus Sub-type, 238 ,, ,. dwarf fruit-bodies,82- 88 ,, „ grafting of, 77-80 heterothallism, 394- 395 „ ,. method of obtaining, 83-84 „ ,, spore-fall period in dwarfs, 99 sterility of, 70-74, 459 Coprinus macrorhizus, and Atramenta- rius Sub-type, 238 „ „ grafting of , 78-79 Coprinus micaceus, and Micaceus Sub- type, 238 and slugs, 219 Coprinus narcoticus, and Atramentarius Sub-type, 238 ,, ,, grafting of, 80 ,, ,, hy menial photo- graph, 317, 319 ,, „ illustrations of, 318, 319 „ „ method of obtain- ing, 83-84 ,, ,, trisporous basidia, 1, 317, 319, 464- 465 Coprinus niveus, and drought, 399 ,, ,, bifurcate spores, 319, 320 „ „ confused with C. lago- pus, 78 „ illustration of, 320 „ „ number of basidial sterigmata, 318, 320 ,, ,, violence of spore-dis- charge, 169 Coprinus plicatilis, and Plicatilis Sub- type, 238 ,, ,, spore-fall period, 95 Coprinus plicatiloides, = C. curtus, 83, 96 Coprinus porcellanus, nuclei in para- physes, 3 Coprinus radiatus, and C. lagopus, 86-87 Coprinus stercorarius, double spore-wall, 52, 124 „ method of obtain- ing, 83-84 GENERAL INDEX 473 Coprinus stercorarius, violent spore-dis- charge, 4 Coprinus sterquilinus, and Comatus Sub type, 238 ,, and other copro- philous fungi, 330 ,, double spore- wall, 52, 124 ,, drop carried with spore, 15 ,, drop too small to analyse, 18-19 gill pigment, 370- 371 grafting experi- ments, 78-80 ,, ,, homothallism, 394 ,, method of obtain- ing, 83-84 ,, number of spores, 139 pigmentation of spore-walls, 56 ,, ,, rate of spore-de- velopment, 44, 45-46, 48, 51, 53, 54, 55 ,, ,, removal of half- ripened spores, 23 >, size and rate of growth of drop, 11, 13 ,, ,, spore-fall period, 135 ,, „ violence of spore- discharge, 169 Coprinus Type, 236 Coprophilous fungi, and drought, 379 „ ,, common species of, 347 „ similarities in, 330 Corda, A. C. I., on hymenial structure, 270 ,, ,, on paraphyses, 2 „ „ on spore-crops of a single basidium, 27 Cordyceps, form of, 186 Corticium, hexasterigmatic basidia, 319 ,, symmetrical basidium, 30 Corticium commixtum, disterigmatic basidia, 317 Corticium coronatum, octosporous basidia, 1, 320 Corticium Solani, rate of spore-develop- ment, 48 Cortinarii, and slugs, 221 Cortinarius, mottling, 253 ,, stored by squirrels, 204 ,, symmetrical basidia, 30 Cortinarius anomalus, and slugs, 214 Cortinarius arvinaceus, mottling, 254 Cortinarius caninus, and slugs, 214, 229- 233, 235 Cortinarius collinitus, mottling, 254 Cortinarius paleaceus, and slugs, 214 Cortinarius rigidus, and slugs, 214 Cortinarius sanguineus, and slugs, 214, 215 Cotton, A. D., observes spore-clouds of Armillaria mellea, 102 . ,, ,, on hymenium of Spar- assis, 188 „ ,, on spores of Anellaria separata, 350 Cows, and Amanita muscaria, 218 Craterellus cornucopioides, form, 181 Crepidotus, mottling, 253 Crepidotus mollis, mottling, 254 Griddle, N., on winter stores of squirrels, 206 Griddle, S., on fungi stored by squirrels, 204, 208 Cronartium ribicola, eaten by squirrels, 202-203 Curtus Sub-type, 238 Cystidia, and slugs, 219 ,, as hymenial elements, 3-4, 457 , , colloidal drops excreted by , 1 9- 20,25 illustrations of, 73, 325, 336, 337, 338 ,, in sterile fruit-bodies, 72-73 nuclei of, 4, 457 ,, origin of name, 324 reduction or absence in dwarf fruit-bodies, 86 Cystidia -bearing fungi, eaten by slugs, 217 Czerniaiev, B. M., on Secotium agari- coides, 449 DACRYOMYCES, absorption of water, 157 form of, 161 „ position of hymenium, 170 „ spore-hilum, 168 474 GENERAL INDEX Dacryomyces deliquescens, absorption of water, 157- 158 ,, ,, basidial and oidial fruit- j bodies, 171- 179 „ ,, drop - excre- tion, 9, 171 ,, „ illustrations of, 173, 175 ,, ,, position of hymenium and violence of spore- ! discharge, 170, 176 ,, ,, rate of spore - develop- ment, 8, 44, 46, 49, 54, 161 ., „ revival after desiccation, 160, 176 ,, ,, spore - dis- charge, 156 „ ,, systematic description, 178-179 „ „ violence of spore - dis- charge, 169, 176, 192 Dacryomyces stillatus, 172, 173, 177, 179 Dacryomyceteae, and Tremellineae, 156 ,, basidial structure, 164, 167-168 ,, drop-excretion, 9 ,, violence of spore-dis- charge, 169 Daedalea confragosa, interrupted spore- fall period, 116- 117, 460 ,, „ number of spore?, 138, 139 Daldinia concentrica, spore-wall, 52 Daldinia vernicosa, spore-wall, 52 Dangeard, P. A., on Dacryomyces deli- quescens, 174 Deakin, P. T., slugs named by, 216, 228 Death of basidia, 273 De Bary, on paraphyses, 3 Deer, and fungi, 207 Deliquescence (autodigestion), 241, 371- 372 Demelius, Fran, on paraphyses, 2 De Seynes, on paraphyses, 2 Desiccation, and fairy rings, 366 and fruit-bodies, 157, 160, 161, 399^00 and mycelium, 246, 347 Dickinson, S., on Fames applanatus, 133 Dietel, P., on spore-discharge in Ure- dineae, 168 Discomycetes, ascus-gun, 32-33 ,, violence of spore-dis- charge and position of hymenium, 170 Disterigmatic basidia, 316-317, 457 Doern, A. H., on squirrels as mycopha- gists, 209 Drop, carried with spore, 7, 13-17 Drop-excretion, abnormal or absent, 17- 19,48 ,. and rate of fall of spores, 26 ., and Stokes' Law, 20 chemistry of, 18-20 ,, function of, 24-26 ., general account, 8-20 „ historical remarks on, 4-5 illustrations of, 7, 10, 12, 13, 14, 16, 18, 19, 272, 287, 290, 292, 308,338,407,416,429 ,, in Calocera cornea, 4-8 ,, size of drops, 11-13 Dwarf fruit-bodies, 82-88, 459 EDIBLE fungi, Marasmius oreades, 93- 94 ,, ,, parasitised Lactarius piperatus, 68-69 Edmonds, J., photograph by, 81 Elaphomyces granulatus, eaten by squir- rels, 195, 198 „ „ illustration of, 198 Elliott, W. T., on slugs as mycophagists, 219, 221 Empusa, spore-discharge, 26 Endophyllaceae, 169 Endophyllum Euphorbiae-sylvaticae, vio- lence of spore-discharge, 169 Entoloma prunuloides, drop carried with spore, 14-15 „ ,, drop-excretion, 9 GENERAL INDEX 475 Entoloma prunuloides, illustration of, 14 | „ „ not mottled, 253 Entomophthoraceae, mechanics of spore - discharge, 22 Epilobium, escape of seeds, 300 Excretions, from cystidia, 325-326 Exhausted basidia, finding of, 271-275 ! Exidia, absorption of water, 157 „ jelly of, 159 Exidia albida, absorption of water by, 157-158 „ and Tremellineae, 156 ., ,, drop-excretion, 9, 171 ,, ,, hy menial position and violence of spore-dis- charge, 170 ., „ rate of spore-develop- ment, 44, 46, 161 ,, ,, revival after desiccation, 160 „ ,, violence of spore-dis- charge, 169, 192 Exidia glandulosa, position of hymenium, 163 Exobasidieae, lack of fruit-bodies in, and parasitism, 193 ,, spore-discharge, 149, 193- 194 Exdbasidium Leucothoes, galls of, 194 Exobasidium Vaccinii, galls of, 193 „ „ hexasterigmatic basidia, 319 FAIRY rings, 88-94, 363-369 Faull, J. H., on Fames fomentarius, 100, 106-110, 112 ,, ,, on Fomes officinalis, 127 photographs by, 109, 126, 130 Fayod, V., on ejaculation of spores, 4-5 ,, ,, on number of sterigmata, 317 ,, ,, on paraphyses, 3 ,, ,, on spore- walls, 52 Femsjonia luteo-alba, hymenial position, 163 „ ,, illustration of, 165 Field-mouse, and fungi, 207 Fistulina hepatica, hymenium of, 2 Flammula, mottling of gills, 253 Flammula carbonaria, mottling, 254 Flammula inopus, and slugs, 214, 215 Flammula sapinea, and slugs, 214 Flies, and Phallus impudicus, 225-226 Fly Agaric, avoided by squirrels, 202 Fomes, heterothallism of, 148 Fomes, high fruit-body efficiency, 109 „ large fruit-bodies of, 118 ,, mechanical consistence, 135-137 „ spore-fall period, 105-106, 108, 135 Fomes annosus, ejaculation of spores, 22 Fomes applanatus, at Kew Gardens, 133 hosts of, 121 ,, ,, hymenial tubes, 128- 132 illustrations of, 122, 124, 131, 132, 134 „ „ longevity, 125, 128 ,, ,, long spore-fall period, 108, 121-148, 460 „ ,, mechanical consist- ence, 136 „ ,, number of spores, 130-131, 137-141 „ ,, possibly hetero- thallic, 148 „ ,, progressive exhaus- tion of hymenial tubes, 143-144 ,, ,, significance of vast numbers of spores, 144-148 „ „ solitariness of fruit- bodies, 118, 120 „ „ visible spore-clouds of, 101, 132-133 Fomes fomentarius, and squirrels, 201 „ attachment of, 112 „ form of, 163 illustrations of, 106, 109, 110, 113 ,, mechanical consist- ence, 136 „ perennial spore- production by same tube-layer, 108-110 ,, ,, rate of growth of tubes, 142 relations with gra- vity, 110-112 solitariness of fruit - bodies, 118 „ spore-fall period, length of, 135 „ vernal spore-fall period, 105-108, 116, 459 ,, visible spore-clouds of, 100, 107 476 GENERAL INDEX Forties igniarius, attachment of, 112 development of, 114 longevity, 113, 125-126, 128 ,, ,, mechanical consistence, 136 ,, ,, number of spores, 116 „ ,, rate of growth of hymenial tubes, 142 „ ., spore-fall period, 108, 113-116, 460 Fomes officinalis, account of, 127 ,, „ illustrations of, 126, 129, 130 „ longevity, 126-128, 460 ,, . „ mechanical consist- ence, 136 ,, ,, solitariness of fruit- bodies, 118 Fomes pinicola, visible spore-clouds of, 100 Fomes pomaceus, continuity of tubes, 1 14 Fraser, W. P., on squirrels as mycopha- gists, 200 Fries, E., and Coprinus curtus, 95 ,, ,, on Hydnum septentrionale, 103 „ „ on Panaeolus, 252-253, 352 ,, ,, on Psalliota campestris, 369, 371 ,, ,, on sterility of Russula integra, 69-70 Fries, R. E., on basidial nuclei, 419- 420 Frost, effects of, 117, 132, 133, 145, 210, 245, 327-329 Fruit-bodies, abnormal, 81-83, 75-77, 453-455 attachment of, in Fomes fomentarius, 112 ,, conditions of origin, 394- 398 „ confluence of, 81-82 ,, development of, in Pan- aeolus campanulatus, 250 dwarf, 82-88 ,, form of, in Tremellineae, 161-162 ,, geotropism of, in Fomes fomentarius, 110-112 ,, hung in trees by squirrels, 207-210 ; , longevity of, in Polyporeae, 125-128 ,, mechanical consistence, 135-137, 375-378 Fruit-bodies, monstrous, of Polyporus rufescens, 75-77 „ proliferation in, 82 ,, rhythmic production of, 97 rigidity, 128, 129, 135 ,, rudimentary, fate of, 398- 399 solitary and imbricated, 118-120 sterile, 69-74 ,, subterranean, and squir- rels, 195 ,, types of mechanism in, 236-243 „ vigour of, 99 Fucaceae, gelatinous nature of, 158 Fungus gnats, and gill-chamber, 394 and poisonous fungi, 218 ,, ,, and spores on gills, 307 Galera hypnorum, disterigmatic basidia, 316 Galera tenera, drop carried with spore, 15 „• ,, drop-excretion, 5, 9 ,, ,, paraphyses of, 3 ,, ., structure and habitat, 330 „ „ successive spore-discharge, 4 Gastromycetes, basidial degeneracy, 29- 33, 457 ,, origin from Hymeno- mycetes, 32 Geaster, spore-crops of a single basidium, 27 Gelatinous hyphae, of Panaeolus cam- panulatus, 250 ,, ,, of Stropharia semi- globata, 329-330 ,, ,, significance of, 156- 158, 161 Generations of basidia, 260, 268, 275- 279, 297-303 Geoglossum, form of, 186 Geotropism, of Clavaria pyxidata, 182 ,, of Fomes fomentarius, 110- 112 of gills, 239, 240 of hymenial spines, 151 of Mushroom, 374 of Panaeolus campanulatus, 250 of Polyporus rufescens, 75 of Typhula, 186-187 Gilbert, E. M., on squirrels as myco- phagists, 200-201 GENERAL INDEX 477 Gills, and Bacteria, 371 ,, and tubes, development of, 142 ,, branching and anastomosis, 453 „ cheilocystidia of, 248 „ coarse and fine adjustments, 251 crowding of, 381, 384 depth of, discussed, 385-388 effect of tilting, 375 ends of, 391-392 function of, 372-373 gelatinous, 164 geotropism of, 239, 240 mottling of, 244-245 mottling of, in Mushroom, 406-408 movement during spore- discharge, 401^02 „ number of, discussed, 381-385 ,, number of, in Panaeolus campanu- latus, 250 ,, packing of, and the basidial guns, 67 ,, radial arrangement discussed, 378- 381 ,, reduced number in dwarfs, 86, 88 „ shape of, 391 ,, suppression of, in Lactarius pipe- ratus, 58-69 „ thickness of, 388-391 Gill-chamber, of Psalliota campestris, 392-394 Gill-system, 361 Gluten, on stipe of Stropharia semi- globata, 329-330 Glycogen, and slugs, 217 in gills, 434-435 ,, in spores, 55, 304 Godfrinia, spore-crops of a single basi- dium, 27 Gomphidius, mottling absent, 253 Gorman, M. W., on squirrels as myco- phagists, 208 Grafting, of fruit-bodies, 77-80, 459 ,, illustration of, 79 Gravity, and hymenium of Clavariae, 184-185 ,, and hymenium of Sparassis, 188-190 „ and Mushroom, 374 „ and mycelial cords, 397 and Tremellineae, 163 Greville, R. K., on Psalliota campestris, 369, 371 Griggs, R. F., assistance of, 235 Grove, W. B., and slug-damaged fungi, 213 Grove, W. B., on sterile fruit-bodies, 69- 70 Guepinia spathularia, position of hy- menium, 163 HAMMER, on spore-clouds, 101 Hancock, R., photograph by, 348 Hansen, E. C., on spore-wall structure, 52 Hartig, R., photograph by, 124 Hastings, S., observed spore-clouds of Armillaria mellea, 102 ,. „ on Anellaria separata, 349 ,, ,, photographs by, 100, 102, 103, 150, 155, 159, 180, 191, 196, 198, 215, 219, 362 Hedgcock, G. G., on hosts of Fomes applanatus, 121 „ „ photograph by, 129 Heliotropism, in Panaeolus campanu- latus, 250 Heptasterigmatic basidia, 319 Heterothallism, and number of spores, 147-148 ,, and sterile fruit-bodies, 73-74 Hexasterigmatic basidia, 319 Hiebert, E., on squirrels as myco- phagists, 208-209 Hiebert, R., and Clavaria pyxidata, 183 Hiley, W. E., on ejaculation of spores in Fomes annosus, 22 Hirmer, Max, on nuclei in Psalliota campestris, 396 Hirneola, absorption of water, 157 Hirneola auricula-judae, and slugs, 212 ,, ,, drop-excretion, 9,171 „ ., illustration of, 162 ,, ,, position of hy- menium, 163 171 ., „ revival after desiccation, 160 ,, spore-discharge, 156 „ „ violence of spore - dis - charge, 169, 192 478 GENERAL INDEX Hoffmann, H., on hymenium of Cop- rinarius, 2 ,, on spore-clouds of Poly- porus destructor, 100 Hollos, L., on Secotium agaricoides, 449 Horses, and coprophilous fungi, 327-329 Horse Mushroom, 365, 400 House-flies, and the Fly Agaric, 218 Howitt, J. E., on squirrels as myco- phagists, 202 Hubert, E. E., on squirrels as myco- phagists, 201-202 Humphrey, C. J., photographs by, 106, 122 Hydathodes, 325 Hydnangium monosporum, monosterig- matic basidia of, 315 Hydneae, a giant species of, 102-103 and Clavariae, 184 and Tremellineae, 163-164 basidia of, 30, 164 ,, drop-excretion, 9, 151 ,, organisation of hymenium, 105 ,, position of hymenium, 170, 181 response to gravity, 163 ,, spore-discharge in, 149-156 „ sterigma and spore-hilum, 30 Hydnum caput-ursi, eaten by squirrels, 201 Hydnum coralloides, spines of, 151 Hydnum erinaceus, form of, 154 ,, ,, illustration of, 155 spines of, 151, 155- 156 Hydnum ferrugineum, drop-excretion, 9, 153-154 Hydnum imbricatum, drop-excretion, 9, 153-154 Hydnum repandum, eaten by squirrels, 201 form of, 154 ., „ illustration of, 150 ,, ,, spore-discharge, 150-151, 154 Hydnum septentrionale, form of, 154 ,. ,, illustrations of, 152, 153, 154 ,. imbrication in, 120 ,, ,, spines of, 151 , ., visible spore - clouds of, 101, 103-105, 459 Hygrophorus, rate of spore-develop- ment, 50, 53 ,, symmetrical basidia, 30 thickness of gills, 389 Hygrophorus agathosmus, disterigmatic basidia, 317 Hygrophorus ceraceus, rate of spore - development, 44, 46, 48, 50, 54 Hygrophorus chrysodon, eaten by squir- rels, 198 Hygrophorus conicus, disterigmatic basi- dia, 316 ,, ,, rate 'of spore-de- velopment, 48, 50 Hymenial spines, 150-151, 155 Hymenial tubes, and Clavaria branches, 179-180 ,, ,, development and acti- vity of, in Fomes Jomentarius, 106 „ illustrations of, 131, 132 ,, increase hy menial area, 130 ,, of Fomes applanatus, 128-132 ,, ,, perennial spore-produc- tion by, in Fomes fomentarius, 108-110 ,, ,, progressive exhaustion of, 143-144 ,, ,, slow elongation of, in Fomes applanatus, 142 „ verticality of, 112, 128, 132 Hymenium, addition of new areas, 142 , , adjustments of position, 374 analysis of, 274-297 area of, 140, 250 ,, criticism of illustrations of, 270-271 ,, early workers on, 270 elements of, 1-4, 457 exhausted, 291, 293, 298, 336, 345, 447-449 ,, exposed in Clavaria, 180 ,, first complete description of, for a Non-Coprinus agaric, 297 folds in, 372-373 illustrations of, 251-325, 330-345, 357, 407-448 „ incompletely described, 270 GENERAL INDEX 479 Hymenium, in Types of Agaricineae, 239-241 „ method of observing de- velopment, 260-264, 408 ,, modes of increasing areas of, 151 ,, observations on develop- ment of, 264-269 ,, of Fames applanatus, dis- cussion of, 141-144 of Mushroom, 406-449 ,, of Panaeolus campanulatus, 269-297 ., of Stropharia semiglobata, 331-341 „ photographs of, 316, 319, 355-359 ,, position of, in Sparassis, 188-190 ,, position of, in Tremellineae, 161-163 ,, subpilear position of, 372 ,, unilateral in some Pterulae, 187 ,, upward-looking in some Tremellineae, 170 ,, wave-development of, 256- 259, 284-288 Hymenomycetes, and spore-discharge, 26 „ basidia of, 164 „ chief groups of, 149 „ comparative spore-fall periods, 95, 134-135 „ confluent fruit-bodies, 81-82 ,, drop-excretion in, 5 ,, eaten by squirrels, 200 „ heterothallism in, 147, 148 ,, organisation of hymen- ium, 105 ,, phylogenetic conver- gence, 163-164 ,, separation of spore from sterigma, 22 „ significance of vast spore numbers, 148 spore-hila, 168 „ sterile fruit-bodies, 69- 70 Hymenomycetes and Gastromycetes, basidia of, contrasted, 29-33 Hypholoma, development of, 452 „ mottling in, 253 Hypholomafasciculare, and slugs, 214, 215 mottling of, 254 » ,, sterile form, 69 » „ stored by squir- rels, 204 Hypholoma pyroirichum, mottling, 254 Hypholoma sublateritium, and slugs, 214 Hypholoma velutinum, mottling, 254 Hypomyces lactifluorum, and squirrels, 204 », „ as a parasite, 58-69, 458 „ ,, illustrations of, 60, 61, 65 Hypomyces htteo-virens, action on hosts, 59 Hypomyces transformans, and squirrels, 201 IMBIBITION, absorption of water by, 157 Inaequi-hymeniiferae, 462^463 Inaequi-hymeniiferous Type, characters of, 240-241 Inocybe, cystidia of, 324 'Inocybe asterospora, and Inocybe Sub- type, 237 „ „ and slugs, 214 Inocybe geophylla, and slugs, 214, 216 Inocybe rimosa, and slugs, 219 Inocybe Sub-type, 237 Inocybe trichospora, calcium oxalate in cystidial drops, 19-20 Interlamellar spaces, 249 Istvanffi, Gy. von, on spore-crops from one basidium, 27 JELLY, significance of, 156-158, 161 KAKSTEN, P. A., on Anellaria, 253, 352 „ ,, on Hypomyces luteo- virens, 59 Keith, on spores of Anellaria separata, 350 Kew, 69, 81, 87, 101, 133, 143, 189 Kniep, Hans, on sex in Hymenomycetes, 147-148, 394-395 Knoll, F., on chemical constituents of excreted drops, 19, 25 „ ,, on trichome-hydathodes, 325 Kyte, and microdissection, 430 Laccaria laccata, and slugs, 214, 216 Lactarii, and slugs, 221 and squirrels, 202 480 GENERAL INDEX Lactarii, parasitised in Finland, 59 Lactarius glyciosmus, and slugs, 214, 215, 216 Lactarius piperatus, and slugs, 213, 214 ,, ,, illustrations of, 60, ! 61,65 ,, ,, parasitised fruit- i bodies edible, 68- 69 „ „ stored by squirrels, 204, 209 ,, ,, suppression of gills, 58-69 Lactarius quietus, and slugs, 214, 216 Lactarius rufus, and slugs, 214, 216 Lactarius subdulcis, and slugs, 214 Lactarius turpis, and slugs, 214 • Lactarius vellereus, sterility of, 69 Lagopus Sub-type, 238 Lange, J. E., on Coprinus curtus, 95 „ „ on genus Coprinus, 84 Leeper, B., onPolyporusrufescens,75-ll „ „ photographs by, 76, 77 Lentinus lepideus, and squirrels, 202 Lenzites, absorption of water by, 157 ,, lignicolous habit of, 157 ,, retention of vitality, 160 Lenzites betulina, spore-fall period, 95 Lepiota cepaestipes, abnormal drop-ex- cretion, 17-19 basidial protuber- ancy, 311 drop-excretion, 9 , ,, ,, illustration of, 19 „ „ paraphyses, 3, 282 Lepiota procera, and Panaeolus Sub-type, 255 ,, „ mechanical structure, 375, 376 „ „ rate of spore-develop- ment, 50, 53 „ spore-walls thickened, 50 „ „ variations in size, 83 Leucosporae, and mottling, 254-255 „ and Panaeolus Sub-type, 254 „ drop-excretion, 9 ,, rate of spore-development, 53,55 „ simple spore-walls of, 52 LeVeill6, J. M. H., on hymenial structure, 2, 270, 363 „ ,, on sterility in"" Lac- tarius vellereus, 69 Levine, M., on basidial nuclei, 420 ,, ,, on Coprinus, 453 „ „ on paraphyses, 2 Light, and hymenial development, 267 „ and mushrooms, 397-398 Lignicolous fungi, 49, 156-161 Limax arborum, and lichens, 217 Limax cinereo-niger, and Russula emetica, 219 Limax maximus, and mustard gas, 234- 235 „ ,, as a mycophagist, 217, 218, 219 ,, „ homing of, 225 „ „ illustration of, 230 ,, „ new chemotactic ex- periments upon, 224- 234 senses of sight and smell, 221 Limax tenellus, as a mycophagist, 217 Liverworts, elaters of, 300 Lloyd, C. G., on Phyllotremella ajricana, 163-164 Lowe, C. W., on Fomes fomentarius, 110 „ „ photographs by, 110, 152 Lycoperdon, aberrant sterigmata of, 31 ,, and Secotium agaricoides, 454 ,, dispersal of spores, 32 „ rate of fall of spores and Stokes' Law, 21-22 Lycoperdon gemmatum, and Secotium agaricoides, 455 ,, ,, illustration of, 32 „ „ sterigmata of, 31-32 Lycoperdon giganteum, and Secotium agaricoides, 455 „ „ (vide also Cal- vatia gigantea) Lycoperdon nigrescens, sterigmata of, 31 Lycoperdon pyriforme, and slugs, 214, 216 Lycopodium powder, and Stokes' Law, 21-22 MAIEE, Rene, on Agaricaceae, 454 „ ,. on basidial spore-crops, 27- 28 „ on nuclei in basidia of Mushroom, 418, 420 „ ,, on paraphyses, 2 GENERAL INDEX 481 Mammals, and Amanitae, 218 Marasmiae, of Hennings, 454 Marasmius, and Panaeolus Sub-type, 255 ,, rapid spore-development, 50, 51-52, 53 spore- wall of, 52 Marasmius oreades, cultivation for food, 88-94, 459 ,, ,, fairy rings of, 364, 368-369 gill thickness of, 389 illustrations of, 89, 91,93 rate of spore-devel- opment, 44, 45, 46, 48-49, 54, 161 ,, ,. supposed parasitism of, 94 Maskell, W., on Marasmius oreades, 89- 90 Massee, G., on Anellaria separata, 349, 351 „ ,, on Coprinus radiatus, 86-87 ,, ,, on Dacryomyces deliquescent*, 171, 176-178 ., ., on Psalliota campestris, 369, 371 Mcllvaine, C., on Hypomyces lacti- fluorum, 58-59 Melanosporae, drop-excretion, 9 mottling, 253-254 Merulius, and Tremellineae, 163, 164 Micaceus Sub-type, 238 Micrococcus bicolor, rate of fall, 40-42 Micrococcus cerasinus, rate of fall, 40-42 Micrococcus Freudenreichii, rate of fall, 40-42 Micrococcus luteus, rate of fall, 40-42 Micrococcus pyogenes aureus, rate of fall, 40^2 Micrococcus roseus, rate of fall, 40-42 Microscope, horizontal, use of, 262-264, 354 Microtome, avoidance of, 274, 362-363 Miller, C. D., mathematical treatment of gills, 388 Millikan, R. A., Stokes' Law corrected by, 21 Mites, and gills, 307, 394 Monosterigmatic basidia, 315, 457 Montagne, C., on paraphyses, 2 Moquin-Tandon, on Limax maximus, 222 Morchella, ascus-gun of, 33 Morchellae, eaten by squirrels, 201 Mottling, of Anellaria separata, 352, 354 Mottling, of Psalliota campestris, 370, 377, 406-408 of Stropharia semiglobata, 330, 331, 332 phenomenon described, 251- 259, 463 significance of, 297-303 Moulds, and spore-production, 27 Mounce, Irene, on heterothallism, 147- 148, 394-395 ,, ,, on sex and sterility in Coprinus lagopus, 73- 74 Mucilage, in drop-excretions, 19-20, 25- 26 - „ on cheilocystidia, 325 Mucorineae, heterothallism in, 147 Murrill, W. A., on squirrels as myco- phagists, 208 Mushroom, abnormal drop-excretion, 17- 18 cultivated, 406-435 good value at different ages, 435 illustrations, 18, 362-448 wild, 409, 435-449 ,, (vide also Psalliota campes- tris) Mushroom Type, 236 Mustard gas, and slugs, 234-235 Mutations, 410 Mutinus, and Secotium agaricoides, 452 Mycelium, desiccation, 329, 347 ot fairy rings, 94, 363-369 of Psalliota campestris, 395- 397 perennial, 367-368 Mycena, spore-crops of a single basidium, 27 Mycena galericulata, disterigmatic basi- dia, 316 Mycetophilidae, and fungi, 218, 394 Mycetozoa, capillitium of, 300 Mycophagist?, fungus gnats, mammals, cows, herbivora, 218 human, 88-94 rabbits, 195-196 rodents, 180 slugs, 212, 235 squirrels, 195-211 Nolanea pascua, and Armillaria Sub- type, 237 ,, ,, drop carried with spore, 15-17 2 i 482 GENERAL INDEX NoUnea pascua, illustration of, 16 not mottled, 253 ,, ,, rate of spore-develop- ment, 44, 50, 54 Non-Coprinus Type, 237, 238-240 Nostoc commune, jelly of, 160 Nuclei, and sterigmata, 320-321 ,, of Dacryomyces deliquescens, 174 ,, of Polyporus rujescens, 77 ,, in a basidium, 304 ,, in basidia of Calocera cornea, 28 ,, in basidia of Mushroom, 411, 418, 419, 421, 430 ,, in gills of Mushroom, 396 ,, in spores, 55 OCHROSPORAE, drop-excretion, 9 mottling in, 253, 254 Octosterigmatic basidia, 320, 321 Odell, W. S., and Secotium agaricoides, 449 ,, ,, photographs by, 450, 451 Odour of Phallus, attracts flies and slugs, 225-226 Oidia, of Dacryomyces deliquescens, 174, 175 Orchids, elater-like hairs of, 300 Order of investigation of Sub-types, 242-243 Otidea, ascus-gun of, 33 PANAEOLTJS, and Anellaria, 351-352 ,, and Panaeolus Sub-type, 238 ,, germ-pore, 452 ,, origin of name, 252 ,; rigidity and spore-fall period, 135 slow spore-development, 52, 53 ,, spore-walls, 124 „ symmetrical basidium of, 30 Panaeolus campanulatus, abnormal drop- excretion, 308-310 ,, „ and Mushroom, 361,404,406, 408,415, 422, 439 „ „ and other cop- rophilous fungi, 330 ,, „ appearance of fruit-bodies, 247-248 Panaeolus campanulatus, cheilocystidia, 324-326 „ ,, collapse of basidia, 312 „ ., cultivation of, 246-247 „ ,, description of, 245-326 „ ,, developing hy- menium of, 264-266 ., drop-excretion, 9, 272, 308- 309 ,. ,, exhaustion of hymenium, 143 „ ,, geotropism and heliotropism, 250-251 ,. ,, hymenium de- scribed in detail, 269- 297 „ ,, illustrations of, 246-325 „ „ mottling, 251- 259 ,, „ mutual influ- ence of adja- cent basidia, 322-324 ,, ,, number of basi- dia on a gill, 301 ,, ,, number of spores, 259 „ „ order of in- vestigations, 242-243 „ ., paraphyses, 3, 278, 279- 281 „ ,, pigmentation of spore- walls, 56 „ ,5 rate of spore- development 44, 45, 48, 50, 51, 438 ,. „ relative posi- tion of spores on a basi- dium, 313- 314 GENERAL INDEX 4^3 Panaeolus campanulatus, spore - fall period, 135, 259-260 .., ,, spore - projec- tion, 289- 290 ,. ,, spores and spermatozoa, 302-303 ,. ,, sterile form, 70 „ „ successive gen- erations of basidia, 267- 269 ,. ,, summary, 463- 465 .. " ,, tristerigmatic basidia, 318 „ wasted spores, 303-310 ,, „ waves of de- velopment, 256-259 ,, ,, why studied, 245-246 Panaeolus papilionaceus, mottling, 254 Panaeolus Sub-type, 238, 242, 243-245 ,, ,, and Coprinus, 425 and Mushroom, 404 Panus stypticus, frost and the spore-fall period, 117 Paraphyses, as sterile elements, 1-3 existence of, 269-270 functions of, 282-283, 422- 423, 431 ,, in exhausted hymenium, 293 „ in young hymenium, 294- 297 ., nuclei of, 3 of the Mushroom, 421-423, 447 ., of Panaeolus campanulatus, 278, 279-281 ,, of Stropharia semiglobata, 334, 336-337 ,, Sachs' illustration of, 405 ,, summary, 457 „ swelling of, 283 Parasites, and gill-chamber, 394 ,, Cronartium ribicola, 202 ,, Fomes applanatus as a wound- parasite, 121-123 ,, Hypomyces lactifluorum, 58- 69, 204 Parasites, Hypomyces luteo-virens, 59 „ Polyporus amarus, 202 „ Ustilago tritici, 68 Parasitism, of Exobasidium, 193-194 „ of Fomes applanatus, 145- 147 „ of Hypomyces transformans, 201 „ of Marasmius oreades, 94 Parenteau, on Arion finding a bean, 222 Patouillard, N., on number of sterig- mata, 315, 318, 319 „ on Pterula, 187 ,, ,, on Sparassis, 188 Paxillus involutus, and slugs, 214 Peck, A. E., and confluent fruit-bodies of Boletus subtomentosus. 81 „ „ photographs by, 30, 47, 51, 82,119,134,162,184,203, 205, 217, 228, 350, 455 Peck, C. H., on Polyporus abortivus, 75 „ ,, on Secolium agaricoides, 449 Pentasterigmatic basidia, 318-319, 321 Perithecia, of Hypomyces lactifluorum, 62-69 Peziza, ascus-gun of, 33 ,, ascus-jet of, 62 „ ascus-sap and spore-discharge, 26 Pezizae, distance of spore-discharge, 67 Peziza vesiculosa, odour of, and slugs, 223 Pfeffer-Festschrift, 5 Pfitzer, E., on fruits of Orchids, 300 Phalloideae, dispersion of spores, 32 Phallus, and Secotium agaricoides, 452 „ and slugs, 235 Phallus impudicus, basidial spore-num- ber, 30 ,, „ description of, 225- 227 illustrations of, 226, 228 its attraction for slugs, 224, 227- 229, 235 ,, ,, sterigmata of, 31 Philomycus carolinensis, in Manitoba, 221 Pholiota, mottling in, 253 Pholiota erebia, anastomosis of gills, 453 „ „ illustration of, 455 Pholiota mutabilis, mottling, 254 Pholiota praecox, rate of spore-develop- ment, 51, 54 2i2 484 GENERAL INDEX Pholiota togularis, disterigmatic basidia, 317 Photographs of hymenium, method for making, 355-359 Phycomycetes, spore-discharge in, 25 Phylogenetic convergence, 163-164 Phytopathology, vide parasites Pigment, in Mushroom gills, 370-371, 402 Pigmentation of spores, 52, 56, 266 Pileus, form and function, 361, 376 „ flattening of, 376-378 „ scales on, 86 Pilous-flesh, abnormal thickening of, in Marasmius oreades, 91- 93 „ and fruit-body rigidity, 374, 375 ,, of Clavaria pistillaris and C. fistulosa, 181 Pilobolus, colloidal drops on, 19-20, 25 „ discharge of sporangium, 13, 25 ,, sporangiophore-sap, 26 Pineapple Fungus, 127 Pistillaria, form of, 181, 186 Pistillaria fulgida, monosterigmatic basidia, 315 Pistillaria Helenae, disterigmatic basidia, 317 Pistillaria maculaecola, monosterigmatic basidia, 1, 315 Pistillaria micans, disterigmatic basidia, 317 Pistillaria sagittaeformis, disterigmatic basidia, 317 Pleurocystidia, 324, 334, 337, 338, 353 Pleurotus, and slugs, 212 Pleurotus ostreatus, grafting of, 78 ,, „ visible spore-clouds of, 101 Pleurotus ulmarius, and squirrels, 200 ,, ,, spore-fall period, 95, 135, 143 Plicatilis Sub-type, 238 Plo wright, C. B., on Hypomyces luteo- virens, 59 Pluteus, cystidia of, 324 Pluteus cervinus, and Armillaiia Sub- type, 237, 253 „ „ and slugs, 219 „ rate of spore-develop- ment, 44, 50 Poisonous fungi, and slugs, 217 Polyporeae, and Clavarieae, 184 Polyporeae, and Exobasidiurn, 193 and Tremellineae, 163, 164 beam- of -light studies of, 105 confluent fruit-bodies, 81 drop-excretion, 9 investigations on, 149 large, 102 longevity of, 125-128 new conception of hymenial activity in, 109 „ position of hymenium, 181, 182 „ solitary and imbricated, 118-120 ,, sterigma and spore-hilum, 30 „ summary, 460 „ violence of spore -discharge, 168, 169 „ visible spore -discharge in, 100, 101 Polyporus, and slugs, 212 branched fruit-bodies, 120 hymenial tubes, 179-180 large fruit-bodies, 118 position of hymenium, 170 symmetrical basidia of, 30 Polyporus abortivus, monstrous fruit- bodies, 75-77 Polyporus adustus, imbrication of, 120 Polyporus betulinus, and squirrels, 201 ,, „ solitariness of fruit- bodies, 118 Polyporus cuticularis, exhaustion of tubes, 144 Polyporus destructor, visible spore-clouds, 100 Polyporus Jrondosus, compound fruit - bodies, 120 Polyporus giganteus, compound fruit - bodies, 120 Polyporus hispidus, exhaustion of tubes, 144 ,, ,, solitariness of fruit- bodies, 118, 120 Polyporus radiatus, illustration of, 119 ,, ,, imbrication of, 119- 120 Polyporus rufescens, illustrations of, 76, 77 ,, „ monstrous fruit- bodies of, 75-77, 459 Polunorus Schweinitzii, effect on wood, 127 GENERAL INDEX 485 Polyporus Schu'einitzii, illustration of, 100 „ ,, visible spore- clouds, 100, 104 Polyporus squamosus, annual fruit- bodies, 128 „ „ drop-excretion, 5,9 ,, ,, elongation of tubes, 142 exhaustion of tubes, 144 ,, „ mode of fall of spores, 35 „ ,, number of spores, 138, 139 rate of fall of spores, 39-40 ,, ,, significance of vast spore numbers, 144 spore - fall period, 95, 135, 138 violence of spore-dis- charge, 169 „ „ visible spore- clouds, 100, 132 Polyporus sulphureus, and Fomes offici- nalis, 127 „ ,, illustration of, 117 ,, ,, imbrication of, 118-120, 460 Polyporus utnbellatus, compound fruit- bodies of, 120, 460 Polystictus, absorption of water, 157 „ lignicolous habit of, 157 ,, retention of vitality, 160 Polystictus hirsutus, drop-excretion, 9 Polystictus perennis, confluent fruit- bodies of, 81 Polystictus versicolor, andelugs, 212, 214, 216 „ „ annual fruit- bodies, 128 ,, ,, imbrication of, 120 „ „ spore-fall period, 95, 135 Polytrichum spores, and Stokes' Law, 21-22 Popuhis, escape of seeds, 300 Porphyrosporae, drop-excretion, 9 „ mottling in, 253, 254 Position of sterigmata, 314-324 Powdering of gills with spores, 307, 308 Protohydnum, and Hydnum, 163-164 Protomerulius, and Merulius, 163-164 Psalliota, and Panaeolus Sub-type, 238 and slugs, 212 „ germ-pore, 452 mottling in, 253 ,, position of hymenium, 170 ,, rigidity and spore-fall period. 135 ,, slow spore-development, 53 ,, spore-crops of a single basi- dium, 27 ,, summary, 457 Psalliota ari'ensis, fairy rings, 365 „ „ liberating spores, 400 „ „ mottling, 254 Psalliota campestris, account of, 360-449 „ ,, ammonia given off from, 372 „ ,, analysis of hymen- ium, 414-449 and Hydna, 151 „ „ and Panaeolus Sub- type, 361, 404 „ „ and Secotium agari- coides, 451 andtext-books,404- 405 „ „ and the Aequi-hy- meniiferae, 236- 238 ,, ,, bisporous basidia, 1, 407, 408, 410- '411,416 ,, ,, clamp - connections absent, 395-396 „ „ drop carried with spore, 15 „ „ drop-excretion, 9- 13 ,, ,, each basidium pro- duces only one crop of spores, 28 „ eaten by squirrels, 201 „ „ effects of tilting,375 „ „ external appear- ance of fruit - bodies, 369-372 „ „ fairy rings of, 363- 486 GENERAL INDEX Psalliota campestris, flattening of pileus, 376-377 form of fruit-body, 375-378 „ gill depth, discus- sion of, 385-388 ,, gill ends, discussion of, 391-392 gill number, dis- cussion of, 381- 385 ,, ,, gill thickness, dis- cussion of, 388- 391 „ ,, gill-chamber and annulus, 392-394 ., ,, gills and tubes com- pared, 141-143 ,, ,, illustrations of, 10, 12, 13, 18, 362- 448 „ ., in dry weather,399- 400 „ „ mechanical struc- ture, 149 ,„ „ mode of investigat- ing, 361-362 ;, „ monosporous and bisporous basi- dia, 315, 407, 408, 410-411, 416 mottling, 254, 255, 406^08 nuclei of, 396, 411, 418, 419-421 „ „ number of spores, 137, 139-141, 402-404 ,, ,, occurrence in fields, 367-369 ,, ,, one spore-crop only from each basi- dium, 27-28 ,. „ order of investiga- tions, 242-243 ,. „ organisation of fruit-body, 372- 378 ,, ,, origin of fruit- bodies, 394-398 ,, ,, paraphyses of, 3, 414, 421-422, 429, 438,440,443,448 ,, „ perennial mycelium of, 367-368 Psalliota campestris, position of gills, 374, 379 ,, ,, radial arrangement of gills discussed, 378-381 rate of fall of spores, 39-40 rate of spore-de- velopment, 44, 48, 54, 55, 412-414 rigidity of, 375 ,, ,, rudimentary fruit - bodies, fate of, 398-399 sex in, 395-396 „ „ significance of vast spore numbers, 144 ,, ,, sporabola of, 67 ,. ,, spore -deposits, 401, 402 spore-fall period, 95, 135, 400-402 ,, „ successive genera- tions of basidia, 269 „ „ summary, 465-466 taste of, 69 ,, ,, tristerigmatic basi- dia, 318, 409 ,, ,, violence of spore- discharge, 169 ,, ,, volume of spores, 29,411,457 Psalliota sylvicola, fairy rings, 365 mottling, 254 Psalliota tabularis, fairy rings, 364-367 illustrations of, 364, 366 Psathyrella, mottling absent, 253 Psathyrella disseminate, and Psathyrella Sub-type, 238 ,, ,, basidial protu- berancy, 311 ,, ,, colloidal drops on pileal hairs, 19,25 ,, ,, confluent pilei, 81 ,, ,, drop - excretion, 9 ,, „ illustration of, 81 ,, ,, paraphyses, 3, 282 GENERAL INDEX 487 Psathyrella Sub-type, 238 Psilocybe, and Panaeolus Sub-type, 238 „ mottling in, 253 Psilocybe ericaea, mottling, 254 Psilocybe foenisecii, mottling, 254, 255 Psilocybe semilanceata, mottling, 254 Pterula, form of, 187 ,, unilateral hymenium, 187, 190 Pterula subulata, confluence of branches, 187 Pnccinia annularis, violence of spore- discharge, 169 Puccinia Glechomatis, violence of spore - discharge, 169 Pucciniaceae, 169 Puff-ball, fairy rings, 365-366 „ Giant, number of spores, 137, 139 Puff-balls, and Secotium agaricoides, 449, 454-455 „ capillitium of, 300 Pyrenomycetes, discharge of spores in, ' 63-64 RABBITS, as mycophagists, 195-196, 197 Rate of spore-development, method of observing, 43-48 Red Squirrel, as a mycophagist, 195-211 illustrations of, 199, 200 Rhizoctonia Solani, rate of spore-develop- ment, 48 Rhoads, A. S., on spore-walls in Daldinia, 52 Rhodosporae, drop-excretion, 9 ,, mottling absent in, 253 Rigidity of pilei, 374-375, 376 Rodents, as mycophagists, 180, 195, 196 Ruhland, W., on nuclei in paraphyses and cystidia, 3-4 „ ,, on paraphyses, 2 Russula, and Panaeolus Sub-type, 255 „ and slugs, 212, 216, 221, 235 and squirrels, 200, 201, 202, 204, 207, 209 cystidia of, 324 ,, rate of spore-development, 53 ,, successive spore-discharge, 4 Russulae, damaged by slugs, 215 ,, milder species preferred by slugs, 219 thickness of gills, 389 Russula adusta, and slugs, 214 Russula alutacea, and slugs, 219 Russula citrina, and slugs, 219 Russula cyanoxantha, and Armillaria Sub-type, 237 »» „ and slugs, 219 » „ rate of spore-de- velopment, 44, 50, 54, 55 Russula emetica, and slugs, 214, 218-219 Russula fragilis, and slugs, 219 Russula keterophylla, and slugs, 213, 214 „ attracts slugs, 229, 235 „ „ illustration of, 213 Russula Integra, and slugs, 219 „ „ sterility of, 69-70 Russula lutea, and slugs, 219 Russula nigricans, and slugs, 214 attracts slugs, 229- 234, 235 Russula ochroleuca, and slugs, 214 Russula pectinata, and slugs, 219 Russula rubra, and squirrels, 202 ,, ,, hymenium of, 270 „ „ text-book illustration of hymenium, 405 Russula sardonia, and slugs, 214, 219 SACHS, J., on Psalliota campestris, 271, 404-405 Salix, escape of seeds, 300 Schizophyllum, absorption of water, 157 „ lignicolous habit, 157 ,, retention of vitality, 160 ,, symmetrical basidium,30 ScJiizophyllum commune, and slugs, 212 frost and the spore - fall period, 117 „ heterothallism of, 148, 394- 395 „ ,, spore - fall period, 95, 135 Schmitz, J., on spore-crops of a single basidium, 27 ,, ,, on spore-discharge, 4 ,, ,, on structure of hymenium, 363 Schrenk. H. von, on longevity of Fomes igniarius, 126 ,, ., „ on spore-clouds of Poly- porus Schweinitzii, 100, 104 Sciurus hudsonicus, as a mycophagist, 197, 199 488 GENERAL INDEX Scleroderma vulgare, basidial spore- number, 30 „ „ illustrations of, 30, 31 „ „ sterigmata of, 31 Secotium acuminatum, 449 Secotium agaricoides, account of, 449- 455, 466 ,, ,, illustrations of, 450, 451, 453, 454 Seton, E. T., drawings by, 199, 200 ,, ,, on storage of fungi by squirrels, 207-208 „ ,, on the Red Squirrel, 197 Sex and sterile fruit-bodies, 73-74 Shantz and Piemeisel, on fairy rings, 94, 365-368 Sistotrema confluens, hexasterigmatic basidia, 319 Slugs, and stimulatory substances in food, 217 „ as mycophagists, 212-235, 462 ,, chemotactic experiments with, 222-234 ,, general absence of, in central Canada, 220 ., homing of, 225 ,, plan of ground used for experi- ments on, 230 ,, rate of progress, 225 „ senses of smell and sight, 221, 235 Smith, W., on spore-production in a Mushroom, 27 Sparassis, hymenium of, 188 illustrations of, 188, 189, 190 „ placed in Thelephoreae, 188 „ possible origin from a Cla- varia, 190 Sparassis crispa, experiment upon, 188- 190 ., „ gravity and position of hymenium, 184 Spaulding, P., on longevity of Fomes igniarius, 126 ,, ,, on squirrels and Cronar- tium ribicola, 202-203 Spermatozoa and spores, 303 Spines, packing of, and basidial guns, 67 Sporabola, nature of, 34 ,, of Panaeolus campanulatus, 249, 287, 289-290 „ of Psalliota campestris, 390 Spores, abnormalities, 320 „ adhesiveness of, 17 „ and fairy rings, 368-369 Spores, and microtome sections, 362 and spermatozoa, 303 ,, arrangement on adjacent basidia of a Mushroom, 427^28 „ bifurcate, 319, 320 ,, carry away proteins, 435 ,, chromosporous, and rate of de- velopment, 51-52 ,, desiccation of, 42 ,, development and temperature, 56-57 ,, development in Panaeolus cam,- panulatus, 266 ., development in Stropharia semi- globata, 340 ,, falling down narrow hymenial tubes, 112, 128, 132 ,, germ-pore, 452 „ individual, rate of development, 43-47 „ misshapen, 310, 320 ,, monstrous, 344 ,, normal and abnormal develop- ment, 46-48 ., number emitted from hymenial tubes, 130-131 ,, number of, in Fomes igniarius, 116 ,, number of, in Hypomyces lacti- fiuorum, 62 ,, number of, in Panaeolus cam- panulatus, 259 ,, number of, in Psalliota campes- tris, 402-404 ,, number of, in various Basidio- mycetes, 137-141 ,, number produced by a single basidium, 27-28 ' »* „ number produced by a small hymenial area, 298 . , of Armillaria mellea, seen falling, 100-102 ,, of Coprinus curtus, seen falling, 98 ,, of Mushroom, dispersal of, 373- 374, 377 „ pink, 50 ,, position while falling, 35 ,, projection of, in Panaeolus campanulatus, 289-290 ., rate of development, 5, 41-57, 161, 412-414 „ rate of fall, 22, 39-40, 249, 290, 373 ,, ripening processes, 55 GENERAL INDEX 489 Spores, size and rate of development, 48- 49 successive discharge, 289 supposed septation of, 177 trajectories of, 249 variations in size, 410-411, 412 vast numbers, significance of, 144 ,, volume and number, 29, 411- 412 „ wasted, 303-310, 447-448 Spores and Bacteria, rates of fall, 40-42 Spores and Thistle-down, rates of fall, 38-40 Spore-deposits, otArmillaria melka, 101 of Clavarieae, 183, 184 „ ,, of Daedalea confragosa, 117 ,, ,, oiFomesapplanatus,IQ8, 134 „ ,, of Hypomyces lactiflu- orum, 62 ., ,, of Psalliota campestris, 401, 402 Spore-development, rate of, 41-57, 458 Spore- discharge, and radial arrangement of gills, 379-380 ,, ,, and surface tension, 26 ,, „ gradual, advantage of, 300-301 ,, ,, history of observations upon, 4-5 in Calocera, 190-193 ,, ,, in Calocera cornea, 4-8 in Clavarieae, 179-190 in Exobasidieae, 193- 194 in Hydneae, 150-156 ,, ,, in Hypomyces lacti- fluorum, 63-68 „ ,, in Panaeolus campanu- latus, 249 ., ,, in Tremellineae, 156- 179 „ ,, mechanism of, 22-26 ,, ., summary, 457 Spore-fall period, longest known, 131 ,, ,, of certain Hymeno- mycetes, 134-135 ,, ., of Coprinus curtus, 95- 99 „ ,, of Fomes applunaius, 131-135 ., ,, of Fomes igniarius, 113-116 Spore-fall period, of Lactarius piperatus, 62 ,, ,, of Panaeolus campanu- latus, 259-260 of Psalliota campestris, 400-402 ., ,, of Stropharia semi- globata, 343 » . • vernal, of Fomes fomentarius, 105-108 Spore-hilum, and spore discharge, 55, 151-153, 171, 457 „ illustrations of, 7, 12, 426 importance of, 8, 23 „ in Hymenomycetes and Gastromycetes, 29-32 ,, not developed in certain Gastromycetes, 30-31 „ strongly developed in Tre- mellineae, 168 ,, summary, 457 Spore-wall, exospore and endospore, 52, 55 „ of Fomes applanatus, 123- 125 ,, of Ganoderma colossus, 123- 124 ,, pigmentation of, 56, 266, 340, 414 polyhedral, 50 ,, structure and rate of spore- development, 49, 161 Springtails, and gill-chamber, 394 „ on gills, 307 on stipe, 329-330 Squirrels, as mycophagists, 195-211, 462 Stahl, chemotactic experiment of, on a slug, 223-224 ,, on food of slugs, 217 ,, on gelatinous frogs' eggs, lichens, and algae, 159-160 Steccherinum septentrionale, 103 Steraum, absorption of water by, 157 ,, and Sparassis, 188 form of, 163, 164 ,, lignicolous habit of, 157 position of hymenium, 170, 171 „ retention of vitality, 160 „ symmetrical basidium of, 30 Stereum Ttirsutum, and slugs, 212, 216 „ attachment of, 163 ,, ,, drop-excretion, 9 „ grafting of, 78 imbrication of, 120 4QO GENERAL INDEX Stereum hirsutum, spore-discharge, 95 „ „ spore-fall period, 95 Stereum purpureum, grafting of, 78 „ „ heterothallism of, 148 Sterigma, collapse of, 7, 271-273, 342, 354-355 function of, 31, 457 ,, in Gastromycetes, 31 Sterigmata, number and relative position on a basidium, 314-324, 332 ,, rhomboidal arrangement of, 322, 332, 356-357 „ rule for position of, 320 Sterile fruit-bodies, illustrations of, 72, 73 ,, ,, observations on, 69- 74, 458 Sterility, degrees of, in Coprinus lagopus, 70 ,, of Lactarius piperatus caused by a parasite, 59 Stevenson, J., on Anellaria separata, 349 ,, ,, on Psalliota campestris, 369, 371 , , , , on sterility in Stropharia obturata, 69 Stipe, form of, 361 „ function of, 373, 375 „ of coprophilous fungi, 330 ,, of Marasmius, abnormal thickness of, 91-94 „ variation in length of, in Anellaria separata, 349 Stokes' Law, and fall of spores, 20-22, 33-34, 41, 458 Stone, R. E., on spore-clouds of Fomes pinicola, 100 Strasburger, on hymenial structure, 270- 271, 405 Streptococcus gracilis, rate of fall of, 40- 42 Stropharia, and Panaeolus Sub-type,238 ,, mottling in, 253 ,, rigidity and the spore-fall period, 135 ,, slow spore-development, 52, 53 ,, universal veil, 452 Stropharia aeruginosa, mottling, 254 Stropharia obturata, sterility of, 69 Stropharia semiglobata, account of, 327- 346 Stropharia semiglobata, and horses and cattle, 327-329 „ ,, and Mushroom, 361, 404, 406, 415, 422, 439 ,, ,, and other copro- philous fungi, 330 „ ,, basidial genera- tions, 333 description of fruit - bodies, 329-331 ,, ,, development of hy me nium, 338-341 ,, ,, drop-excretion, 9, 340 ,, ,, exhaustion of hv- menium, 143, 333, 334, 335 ,, hymenium of, 331-341 ,, ,, illustrations of, 328-345 mottling, 254,255, 330-332 ,, ,, order of investiga- tions, 243 ., paraphyses, 3, 335-336 ,, „ pigmentation of spores, 341-342 ., „ rate of spore-de- velopment, 44, 45, 46, 50, 54, 340 ,, ,, rhomboidal ar- rangement of spores, 332 ,, spore-fall period, 135, 343 ,, ,, stages in spore- development, 55-56, 340 ,, ,, sterile form of, 70 ,, ,, summary, 465 ,, ,, wasted spores, 343-346 Subhymenium, in Mushroom, relations with hymenium, 361, 405, 434, 439- 441 Sub-types of fruit-body mechanism, 237-238, 241-242 GENERAL INDEX 491 Successive generations of basidia, 260, 267-269, 333, 414 Summary of results, 457-466 TAYLOR, J. W., monograph on Mollusca, 222 Temperature, and spore development, 56-57 ,, and spore-discharge, 114, 116-117, 132, 133 Teratology, 58-59, 75-76, 81-83, 91-93, 320 Text-book illustrations of hymenium, 270-271, 404-405 Thelephora sericea, spore-discharge, 4 Thelephoreae, and Tremellineae, 163,164 . ,, drop-excretion, 9 „ octosterigmatic basidia, 319-320 ,, previous investigations upon, 149, 150 ,, sterigma and spore-hilum, 30 Thistle-down, rate of fall of, 38-40, 42 Titley, J. E., photographs, 117, 378,r,79 Transpiration, of basidia, 282 ,, of mycelium, 398 Tremella, form of, 161 jelly of, 159 position of hymenium, 170 ,, spore-hilum of, 168 Tremelleae, and Tremellineae, 156 basidial structure, 164, 166- 167 „ violence of spore-discharge, 169 Tremellineae, drop-excretion, 9, 171 fruit-body form, 161-162 ,, phylogenetic convergence toward other Hymeno- mycetes, 163-164 ,, position of hymenium, 170 ,, position of sterigmata, 318 ,, related to Uredineae, 170 ,, spore-discharge in, 156- 179 ., spore-walls of, 161 ,, sterigma and spore-hilum well developed, 30 violence of spore-dis- charge, 168-171 with gills, 164 Tremellodon, and Hydnum, 163-164 Tremellodon gelatinosum, position of hvmenium, 163 Tremelloid and non-tremelloid fungi, 157 Tricholoma nudum, confluent fruit - bodies in, 82 Tricholoma personatum, eaten by squir- rels, 202 Trichome-hydathodes, 325 Tristerigmatic basidia, 317-318, 321 Trogia crispa, eaten by slugs, 220 Tubaria conspersa, disterigmatic basidia, 317 Tubaria furfuracea, and slugs, 214, 216 Tuberaceae, degeneration of ascus-gun in, 33 Tubes of Polyporeae, and the basidial guns, 67 Tulasne, L. R., on Dacryomyces deli- quescens, 172, 173- 174 ,, ,, on Scleroderma vul- gare, 31 Types of fruit-body mechanism, 236-243 Typhula, form of, 186 ULOCOLLA, form of, 161 jelly of, 159 ,, position of hymenium, 170 Uredineae, and Auricularieae, 165-166 ,, and Tremellineae, 30, 170 ,, violence of spore-discharge, 168, 169, 170, 461 Ustilago tritici, and Hypomyces lacli- fluorum, 68 VAN TIEGHEM, on hymenium of Mush- room, 405 Velum partiale, of Mushroom, 392 Voglino, on slugs and fungi, 216 Von Hohnel and Litschauer, on Cor- ticium, 317, 319, 320 WAKEFIELD, E. M., and Anellaria sepa- rata, 347 ,, ,, on Hypholoma Jas- ciculare, 69 ,, ,, photographs by, 182, 188 Wallis, J. B., on chickens killed by a squirrel, 211 Wasted spores, of Panaeolus campanu- latus, 303-310, 311- 312 ,, ,, of Psalliota campestris, 447-448 ,, ,, of Stropharia semigldbata, 343-346 492 GENERAL INDEX Wasted spores, percentage, 305, 344 Water, absorption of, by capillarity and imbibition, 157 Water-drop, account of, 6-20 ,, carried with spore, 13-17, 290 ,, excessive excretion of, 17- 19, 308-310 ,, in Clavaria Jormosa, 185 in Dacryomyces deliquescen s, 176 in Hydneae, 151-152 ,, in Panaeolus campanulatus, 284, 292 ,, in Psalliota campestris, 424 ,, in Stropharia semiglobata, 337, 338 „ in Tremellineae, 171 Wave development, and testicular tubules, 303 ?, „ discussion of, 301-303 Wave development, in Panaeolus campanulatus, 284-288 ,, .. in Psalliota cam- pestris, 406- 407, 416, 423- 425,428,431- 434, 439 Weir, J. A., on grafting fruit-bodies, 77- 79 Weismann, A., on telescope-eyes, 164 White, J. H., on Fomes applanatus, 121- 125, 128-133, 137, 142, 143 Winter break in spore-fall period, 116- 117 Workman, B. A., assistance given by, 57, 339 Worsdell, W. C., on proliferation, 82 ZALEWSKI, A., on ejaculation of spores, 4-5 Zelenyand M'Keehan, on Stokes' Law, 21 Printed in England at THE BALLAXTYNE PRESS SPOTTISWOODE, RALLANTYNE & Co. LTD Colchester. London & Eton UNIVERSITY OF CALIFORNIA AT LOS ANGELES THE UNIVERSITY LIBRARY This book is DUE on the last date stamped below A 000425398 5 J1 ERT* IS ' OF CALIFORNIA <3RARY, >S SELES. CALIF.